Selective alteration of polymer surfaces by thermal cleavage of fluorinated side chains

Article abstract of DOI:10.1021/ma011938v

We have synthesized a series of fluorinated side chain block copolymers based on either a hydroborated polystyrene-b-polyisoprene or a poly(methyl methacrylate)-b-poly(2-hydroxyethyl methacrylate) backbone. Through polymer-analogous reactions, perfluorinated ester, fluorinated urethane, and fluorinated carbonate side chains were attached to the pendant hydroxyl groups. Their selective thermal degradation behavior was investigated using thermogravimetric analysis (TGA) and thermal desorption mass spectrometry (TDMS). In addition, we investigated the heat- and laser-induced selective alteration of polymer film surfaces by chemically highly selective cleavage of the fluorinated side chains using X-ray photoelectron spectroscopy (XPS), FT-infrared spectroscopy (FT-IR), and contact angle measurements. Finally, we demonstrate the potential to create a laterally patterned surface with different regions of chemically well-defined surface moieties from our fluorinated block copolymers by a dry laser process.

Full text of DOI:10.1021/ma011938v

Macromolecules 2002, 35, 4929-4937
4929
Selective Alteration of Polymer Surfaces by Thermal Cleavage of
Fluorinated Side Chains
Alexa n d er Bo1k er ,† Th om a s Her w eg,‡ a n d Ka r sten Reih s*,‡,§
Lehrstuhl fu¨r Makromolekulare Chemie II, Universita¨t Bayreuth, D-95440 Bayreuth, Germany, and
Central Research, Bayer AG, D-51368 Leverkusen, Germany
Received November 6, 2001; Revised Manuscript Received March 11, 2002
ABSTRACT: We have synthesized a series of fluorinated side chain block copolymers based on either a
hydroborated polystyrene-b-polyisoprene or a poly(methyl methacrylate)-b-poly(2-hydroxyethyl meth-
acrylate) backbone. Through polymer-analogous reactions, perfluorinated ester, fluorinated urethane,
and fluorinated carbonate side chains were attached to the pendant hydroxyl groups. Their selective
thermal degradation behavior was investigated using thermogravimetric analysis (TGA) and thermal
desorption mass spectrometry (TDMS). In addition, we investigated the heat- and laser-induced selective
alteration of polymer film surfaces by chemically highly selective cleavage of the fluorinated side chains
using X-ray photoelectron spectroscopy (XPS), FT-infrared spectroscopy (FT-IR), and contact angle
measurements. Finally, we demonstrate the potential to create a laterally patterned surface with different
regions of chemically well-defined surface moieties from our fluorinated block copolymers by a “dry” laser
process.
In tr od u ction
shown previously, there is a significant difference in the
selective side chain cleavage if one compares hydrocar-
bon ester side groups to perfluorocarbon ester side
chains.8 Through incorporation of urethane and carbon-
ate functionalities, which are supposed to decompose
earlier than an ester linkage,13-15 we are searching for
ways to lower the selective decomposition temperature.
Furthermore, we shall describe the possibility to
create a laterally patterned surface with different
chemically well-defined regions from our fluorinated
block copolymers. These patterns can be produced from
a “dry” laser treatment only, thus avoiding any wet or
vapor chemical process to produce lateral patterns with
well-defined surface moieties. However, a local post-
treatment by a wet chemical surface reaction is greatly
assisted by the fact that regions with different chemical
moieties are also very different in hydrophobicity. Thus,
our laser patterned surfaces can be used as templates
to produce a variety of controlled chemically heteroge-
neous surfaces.
The low surface energy of fluorinated polymers1-4
allows for controlled alteration of surface properties of
coatings or polymer blends. Upon blending a fluorinated
polymer with another miscible component, the surface
composition of the mixture is dominated by fluorinated
groups due to their low surface energy leading to highly
hydro- and oleophobic surfaces.5-7
Recently,8 we have developed polymers with fluori-
nated side chains, i.e., perfluorinated groups, in order
to produce extremely low-energy surfaces. Furthermore,
we have demonstrated that upon thermal treatment the
fluorinated side chains can be cleaved selectively using
a well-known ester pyrolysis reaction.9 The absence of
fluorinated side chains at a polymer surface leads to a
substantial increase of the surface energy, resulting in
significantly different wetting properties compared to
an untreated fluorine-rich surface. Since the thermal
cleavage can be carried out locally, such fluorinated side
chain polymers can be used to create chemically defined
hydrophobic/hydrophilic patterns on surfaces. These
properties may favorably support applications of surface
patterning, e.g., printing or the preparation of micro-
sample arrays. In addition, defined chemically struc-
tured surfaces can be formed which allow the study of
wetting phenomena on chemically heterogeneous sur-
faces.10,11 Furthermore, well-defined laterally patterned
surfaces are important for a variety of technological
applications, e.g., as compatibilizers for polymer blends
or as templates for growing biological cells with con-
trolled shapes and sizes.12
Exp er im en ta l Section
The acronyms of the polymers are chosen according to a
previous publication;8 i.e., SI-46/11 represents a block copoly-
mer with a molecular weight of 46 kg/mol of the polystyrene
(S) block and 11 kg/mol of the polyisoprene (I) block. Similarly,
MH-26/5 denotes a poly(methyl methacrylate) block (M) of 26
kg/mol and a poly(2-hydroxyethyl methacrylate) (H) block of
5 kg/mol. Unless noted, all chemicals used in this work were
purchased from Aldrich and used without further purification.
Syn t h esis of H yd r oxyla t ed P olyst yr en e-b-p olyiso-
p r en e Block Cop olym er s. The polystyrene-b-polyisoprene
block copolymers used in this study were synthesized as
reported previously.8 Their composition is shown in Table 1.
After hydroboration of the polyisoprene block the pendant
hydroxyl groups were reacted with fluorinated isocyanates, a
fluorinated chlorocarbonate, and perfluorinated carboxylic acid
chlorides, yielding urethane, carbonate, and ester side groups,
respectively.
The aim of our present study is to investigate the
temperature dependence of the cleavage of different side
chains in our polymers on their chemical nature. As
† Universita¨t Bayreuth.
‡ Bayer AG.
Ur eth a n e Sid e Ch a in s (URF 1). The urethane side chains
were introduced by reaction of the alcohol groups of the
hydroxylated isoprene block with a fluorinated isocyanate as
shown in Scheme 1. Care has to be taken in order to eliminate
§ New address: SuNyx Surface Nanotechnologies GmbH, Stol-
berger Str. 370, D-50933 Ko¨ln, Germany.
* Corresponding author: phone +49 221 485 2453; Fax +49 221
485 2479; e-mail karsten.reihs@sunyx.de.
10.1021/ma011938v CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/18/2002

4930 Bo¨ker et al.
Macromolecules, Vol. 35, No. 13, 2002
Ta ble 1. Syn th esized P olym er s w ith P olystyr en e-b-p olyisop r en e (SI) Ba ck bon e
deg of polymerization
(PS:PI)
block ratio
(PS:F-PI) [kg/mol]
vol fraction of
fluorinated block
polymer
Mw(GPC) [kg/mol]
polydispersity
1.02
1.03
1.03
SI-26/7
250:100
25.9:6.9
21.2
30.0
52.1
65.0
67.3
51.3
19.2
30.7
49.6
62.7
65.0
48.8
49.4
22.1
34.8
53.5
54.3
67.1
69.2
35.2
39.4
41.7
SI-26/7-H1c
SI-26/7-F3
25.9:11.1
25.9:28.2
25.9:48.2
25.9:53.2
25.9:27.3
45.6:10.8
45.6:20.2
45.6:44.8
45.6:76.8
45.6:84.8
45.6:43.4
45.6:44.6
65.9:18.7
65.9:35.2
65.9:75.8
65.9:78.3
65.9:134.3
65.9:148.3
SI-26/7-F7
SI-26/7-F8
SI-26/7-URF1
SI-46/11
micelle: 36 and 560
each monodisperse
micelle: 33 and 250
each monodisperse
97.9d
63.6
65.3
63.7
a
1.27b
1.03
1.03
1.05
440:160
SI-46/11-H1c
SI-46/11-F3
SI-46/11-F7
SI-46/11-F8
SI-46/11-URF1
SI-46/11-OCOH1F2
SI-66/19
a
226.6d
67.8
92.5
95.2
503.7d
a
1.48b
1.05
1.03
1.02
1.65b
640:280
SI-66/19-H1c
SI-66/19-URF1
SI-66/19-F3
SI-66/19-F7
SI-66/19-F8
a
a
a
b
No signal could be detected by GPC. Tailing due to extended interaction with column. c Acetylated after hydroboration to improve
d
solubility. PSS GR_AM columns, high molecular weight due to aggregation. Side chains: F3 ) perfluorobutyryl, F7 ) perfluorooctanoyl,
F8 ) perfluorononanoyl, H1 ) acetyl, URF1 ) R,R,R-trifluoro-p-toluyl, OCOH1F2 ) fluorinated carbonate.
Sch em e 1. Syn th esis of Ur eth a n e Sid e Ch a in Block
Cop olym er s
tion and a reduced yield due to the higher nucleophilicity of
the amine compared to that of the alcohol.
0.5 g (1.42 mmol of -OH) of SI-46/11-OH are dissolved in
15 mL of freshly distilled THF (refluxed over potassium for at
least 3 days) in a nitrogen atmosphere and 0.22 mL of R,R,R-
trifluoro-p-toluyl isocyanate (F 1.31 g/mL, Mw 187.12 g/mol,
5% excess) were added at room temperature, and the reaction
mixture was slowly heated to 40 °C and stirred overnight.
Finally, the solution was heated to 85 °C to complete the
reaction. After cooling to room temperature, the crude product
was precipitated in water. The polymer was dissolved in 10
mL of THF and reprecipitated in n-hexane. This procedure
was repeated twice before the product was dried under vacuum
at room temperature.
Yield: 0.7 g (90%) of white and brittle powder (SI-46/11-
URF1).
1H NMR (CDCl3, 250 MHz, TMS) δ [ppm]: 7.2-7.6 (-NH-
C6H4-CF3), 6.3-7.15 (aromatic polystyrene units), 3.6-4.5
(OOC-NH-)/(R-CH2-O-), 0.5-2.3 (hydrocarbon backbone
and aliphatic isoprene units).
IR (film on NaCl plate) ν˜ [cm-1]: 3323, 3025, 1709, 1602,
1546, 1328, 1164, 1119.
F lu or in a ted Ca r bon a te Sid e Ch a in s (SI-46/11-
OCOH1F 2). The carbonate side chains have been introduced
via reaction of a fluorinated chlorocarbonate that has been
synthesized previously using excess phosgene and the corre-
sponding fluorinated alcohol as shown in Scheme 2. Care has
to be taken to remove phosgene. As a consequence, cross-
linking reactions via the phosgene could be eliminated, and
the narrow molecular weight distribution of the starting block
copolymer could be retained throughout the reaction.
3 mL (4 g, 0.04 mol) of phosgene was added at 0 °C to a 200
mL flask with 50 mL of freshly distilled THF. After that, a
solution of 3.36 g (0.02 mol) of 1,1,1,3,3,3-hexafluoro-2-propanol
and 2.02 mL (0.02 mol) of triethylamine in 10 mL of THF was
added dropwise to the reaction mixture. A fine white precipi-
tate appeared immediately. Then nitrogen is bubbled through
the solution for 45 min at 10 °C in order to remove excess
phosgene. After that, a solution of 1 g of SI-46/11-OH (2.8 mmol
of -OH) and 2.02 mL (0.02 mol) of triethylamine in 70 mL of
THF is added, and the reaction is allowed to proceed for 1 h
at 10 °C before it is slowly heated to room temperature.
Subsequently, distilled water is added to destroy excess
1,1,1,3,3,3-hexafluoro-2-propyl chlorocarbonate, until a clear
solution appears. Finally, the crude product is precipitated in
600 mL of water, redissolved in methylene chloride, and
washed with diluted hydrochloric acid and distilled water. The
organic phase is concentrated to 20-30 mL, precipitated
side reactions like allophanate formation by reaction of iso-
cyanate with the urethane group. Therefore, only a slight (5%)
excess of isocyanate is used for this reaction. In addition,
considerable care has been taken to keep the reaction media
moisture-free to avoid Hofmann degradation of the isocyanate
to the corresponding amine, which would lead to urea forma-

Macromolecules, Vol. 35, No. 13, 2002
Selective Alteration of Polymer Surfaces 4931
Sch em e 2. Syn th esis of F lu or in a ted Ca r bon a te Sid e
Ch a in Block Cop olym er
Sch em e 3. An ion ic Block Cop olym er iza tion of Meth yl
Meth a cr yla te a n d 2-[(Tr im eth ylsilyl)oxy]eth yl
Meth a cr yla te
in 300 mL of methanol, and filtrated. The polymer is redis-
solved in methylene chloride and reprecipitated in methanol
two more times and finally dried under vacuum at room
temperature.
Yield: 1 g (63%) of white powder (SI-46/11-OCOH1F2).
IR (film on NaCl plate) ν˜ [cm-1]: 3025, 2925, 1697, 1600,
1492, 1452, 1271, 1172, 757, 698.
Syn th esis of P er flu or in a ted Sid e Ch a in Block Cop oly-
m er s on th e Ba sis of P oly(m eth yl m eth a cr yla te)-b-p oly-
(2-h yd r oxyet h yl m et h a cr yla t e) (P MMA-b-P HE MA). (a )
Syn th esis of 2-[(Tr im eth ylsilyl)oxy]eth yl Meth a cr yla te
(TMS-HEMA). The trimethylsilyl protecting group was in-
troduced as previously reported by Hirao et al.16 using hexa-
methyldisilazane and trimethyl chloride.
over calcium hydride overnight and distilled under reduced
pressure. After that it was treated with triethylaluminum and
finally condensed into a glass ampule.
The polymerization is carried out in THF that is freshly
distilled from potassium after refluxing for at least 3 days.
On the day before the reaction, the solvent is titrated with a
few drops of sec-BuLi (1.3 M solution in n-hexane) at -20 °C
until a slightly yellowish color appears. The polymerization is
initiated with sec-BuLi, capped with diphenylethylene (DPE),
which gives diphenylhexyllithium as active initiator species
as shown in Scheme 3. The reaction time of the polymerization
of TMS-HEMA should not exceed 20-25 min in order to
prevent proton abstraction by the living chain end from the
protecting group.16 In this way, a very narrow molecular
weight distribution can be achieved as shown in Table 2. GPC
elugrams of both PMMA precursor and PMMA-b-PHEMA
show single equally narrow peaks.
600 mL of distilled and titrated THF was cooled to -78 °C.
0.513 mL of sec-BuLi was added and capped with 0.26 mL of
DPE, which results in a deep red color of the reaction mixture.
After 5 min, the polymerization was started by adding 20 g of
MMA, which causes an immediate disappearance of the red
color and a temporary rise in temperature to -50 °C. After
an additional 45 min, 7 g of TMS-HEMA was added quickly
and polymerized for another 25 min. Finally, the reaction is
stopped with 1 mL of degassed methanol. The polymer is
precipitated in 3 L of methanol, which also leads to deprotec-
tion of the PHEMA block. Then the product is redissolved in
THF and reprecipitated two more times in n-hexane and dried
under vacuum at room temperature. The block ratio has been
determined by 1H NMR spectra using the integrated aromatic
signals of the benzoylated block copolymers as well as the
protons of the ethylene spacer as a reference.
30 g (0.23 mol) of 2-hydroxyethyl methacrylate was added
under a nitrogen atmosphere to a 50 mL round-bottom flask
closed with a rubber septum. After that the mixture was cooled
to 10 °C, and 37.5 g (0.23 mol) of hexamethyldisilazane was
added dropwise via a syringe. Finally, a few drops of trimeth-
ylsilyl chloride were added as catalyst, and the reaction
mixture was stirred overnight at room temperature. The next
day ammonia was removed under reduced pressure, and the
crude product was distilled from the reaction mixture and
finally purified by fractionated distillation.
Yield: 30.4 g (65%) of colorless liquid. Kp(3) 60-62 °C. Lit.:16
Kp(2) 71-72 °C.
1H NMR (CDCl3, 250 MHz, TMS) δ [ppm]: 6.20 (1H, m,
HCdCCO), 5.60 (1H, m, HCdCCH3), 4.40 (2H, t, OCOCH2),
3.60 (2H, t, CH2CH2OSi), 2.20 (3H, s, CH3), 0.18 (9H, s,
Si(CH3))
(b) Block Cop olym er iza tion of MMA a n d TMS-HEMA.
The two methacrylate monomers were purified according to
standard methods. The MMA was stirred with triethylalumi-
num for 2 h and finally condensed into a glass ampule with a
rotaflo Teflon stopcock. The TMS-HEMA was twice stirred
Yield: 24 g (76%) of white powder (MH-26/5).

4932 Bo¨ker et al.
Macromolecules, Vol. 35, No. 13, 2002
Ta ble 2. Syn th esized P olym er s w ith P oly(m eth yl m eth a cr yla te)-b-p oly(2-h yd r oxyeth yl m eth a cr yla te) (MH) Ba ck bon e
deg of polymerization
(PMMA:PHEMA)
block ratio
(PMMA:F-PHEMA) [kg/mol]
vol fraction of
fluorinated block
Mw(GPC)
polymer
[kg/mol]a
polydispersity
MH-26/5
260:40
26:5
16.1
25.3
43.4
45.6
51.3
24.5
28.2
29.2
24.9
29.0
1.05
1.04
1.05
1.05
1.05
MH-26/5-Bz
MH-26/5-F7
MH-26/5-F8
MH-26/5-F11
26:8.8
26:19.9
26:21.8
26:27.4
a
According to PS calibration, MMA precursor calculated using Mark-Houwink equation. Side chains: Bz ) benzoyl, F7 )
perfluorooctanoyl, F8 ) perfluorononanoyl, F11 ) perfluorododecyl.
Ta ble 3. TGA Da ta of th e SI Block Cop olym er s w ith
1H NMR (CDCl3, 250 MHz, TMS) δ [ppm]: 8.05 and 7.5
Ur eth a n e a n d Ca r bon a te Sid e Ch a in s a n d MH Block
aromatic hydrogen atoms, 4.4 (OCH2CH2OBz), 4.2 (OCH2CH2-
Cop olym er s
OBz), 3.6 (OCH3), 1.8 (CH2), 0.8 (CH3), benzoylated to improve
solubility.
calcd loss exptl loss
1st
2nd
IR (film on NaCl plate) ν˜ [cm-1]: 3443, 2993, 2951, 1731,
1484, 1448, 1272, 1241, 1191, 1149, 1067, 987, 966, 749.
(c) Atta ch m en t of P er flu or in a ted Ester Sid e Ch a in s.
The introduction of the perfluorinated ester side chains is
performed as published previously for hydroxylated PS-b-PI
block copolymers.17 As shown in Table 1, the narrow molecular
weight distribution of the starting block copolymer could be
retained throughout the polymer-analogous reactions.
In a 10 mL round-bottom flask equipped with a rubber
septum, 0.2 g of MH-26/5 (0.46 mmol of -OH) were dissolved
in 5 mL of freshly distilled THF under a nitrogen atmosphere.
Then 2 mL of absolute pyridine was added via a syringe, and
finally 0.5 mL of perfluoroctanoic acid chloride (30% molar
excess) was added dropwise to the reaction mixture. A fine
white precipitate occurred immediately. The reaction was
heated to 45 °C and allowed to proceed overnight, before it
was stopped with 1.5 mL of methanol. Subsequently, the
polymer was precipitated twice in a 50 mL methanol/water
mixture (1:1, 2:1, v:v), redissolved in 5 mL of THF, and
precipitated in 50 mL of n-hexane. Finally, it was dried under
vacuum at room temperature.
of mass
of mass degradation degradation
[%]
polymer
[%]c
step [°C]
step [°C]
SI-26/7-URF1
SI-46/11-URF1
SI-66/19-URF1
SI-46/11-
OCOH1F2
MH-26/5
36
36
36
37
39
35
37
30
232
326
318
354
389
397
397
419
100
33
37
100
a
a
364
MH-26/5-F7
MH-26/5-F8
316b
339b
354b
370b
a
b
No precise evaluation possible. Evaluated by peak-fitting
procedure. c Assuming first degradation step of thermal side chain
cleavage.
FT-infrared spectrometry (FT-IR), 1H NMR, X-ray photo-
electron spectroscopy (XPS), contact angle measurements,
thermogravimetric analysis (TGA), and thermodesorption
mass spectrometry (TDMS) were performed as described in a
previous paper.8
Resu lts a n d Discu ssion
Yield: 0.4 g (90%) of slightly yellowish powder (MH-26/
5-F7).
Degr a d a t ion St u d y Usin g Th er m ogr a vim et r ic
An a lysis. In a previous paper,8 we presented polysty-
rene-b-polyisoprene block copolymers that allowed for
selective thermal cleavage of fluorinated side chains
attached by ester groups. In addition, we observed a
distinct difference between perfluorocarbon and hydro-
carbon ester side chains. Perfluorocarbon ester side
chains showed distinctly lower decomposition temper-
atures than hydrocarbon ester side chains, which de-
composed at temperatures similar to the degradation
temperature of the polymer backbone itself. To increase
the selectivity of the fluorinated side chain decomposi-
tion, our aim in the present study is to investigate the
dependence of the cleavage temperature of different
fluorinated side chains on their chemical structure.
Urethane and carbonate functionalities are supposed to
thermally decompose at lower temperatures compared
to ester groups, thus allowing to thermally detach
fluorinated side chains at lower temperatures and to
facilitate processes to incorporate hydrophobic/hydro-
philic patterns into polymer surfaces.
(a ) Ur et h a n e Sid e Ch a in s (-UR F 1). A series of
polymers with the same urethane side chain (URF1) and
different overall molecular weights of the SI blocks have
been investigated by TGA; the results are shown in
Table 3. As can bee seen, all urethane side chain
polymers show a two-step degradation process. The
second high-temperature process around 390-400 °C
can be attributed to the decomposition of the polymer
backbone as derived from earlier studies.8 N-Monosub-
stituted urethanes are expected to decompose via a
N-H hydrogen transfer initiated depolymerization mech-
anism which leads to the formation of isocyanate
IR (film on NaCl plate) ν˜ [cm-1]: 2951, 1788, 1731, 1484,
1448, 1241, 1206, 1149, 988, 748.
Exp er im en ta l Setu p
Sa m p le P r ep a r a tion . The polymer films were prepared
from 5 wt % solutions of the fluorinated block copolymers in
R,R,R-trifluorotoluene (SI blocks) and THF (MH blocks). Thin
films (approximately 200 nm as obtained by scanning force
microscopy (SFM) at an edge of a partially delaminated film)
were prepared by spin-coating 0.1 mL of the solutions onto a
1 × 1 cm2 piece of a polished silicon wafer (2000 min-1, 20 s).
Prior to use, all films were dried overnight under vacuum at
room temperature and annealed at 140 °C for 3 h in a vacuum
oven. The polymer films used for the laser writing experiments
were prepared by casting from 5 wt % solutions in THF onto
aluminum foil.
Gel P er m ea tion Ch r om a togr a p h y. GPC measurements
were performed using a WATERS GPC equipped with dual
detectors (RI and UV [λ ) 254 nm]) and SDV-gel columns (PSS
Mainz, Germany) of 5 µm particle size having 105, 104, 103,
and 102 Å pore size. In addition, polyester-gel columns (GRAM,
PSS Mainz, Germany) with 10 µm particle size and 104, 3 ×
103, and 102 Å pore size were used for polymers with urethane
side chains. In each case, the elution solvent was THF at room
temperature with an elution rate of 1 mL/min. Narrow
polydispersity polystyrene samples were used as calibration
standards.
Th er m a l Tr ea tm en t. Thin polymer films spin-coated on
silicon wafers were heated to 330 °C for 15 min in a vacuum
heating chamber. Finally, the samples were cooled to room
temperature, and the chamber was flushed with nitrogen.
LASER Tr ea tm en t. The laser-induced decomposition stud-
ies were performed using a neodymium glass laser operated
at 1064 nm and 295 mW. The velocity of the writing beam
was set to 3.2 m/s.

Macromolecules, Vol. 35, No. 13, 2002
Selective Alteration of Polymer Surfaces 4933
F igu r e 3. TGA curve of SI-46/11-URF1.
F igu r e 1. TGA curve of SI-26/7-URF1 (allophanate structure).
F igu r e 2. FT-IR of SI-26/7-URF1 before (s) and after heating
(‚ ‚ ‚) at 340 °C for 1 h.
F igu r e 4. TGA curve of SI-46/11-OCOH1F2.
Ta ble 4. TGA Da ta of th e P er flu or in a ted Ester Sid e
Ch a in Block Cop olym er s
monomers at temperatures between 285 and 325 °C.14
We therefore assume that the first degradation step can
be attributed to such thermal cleavage of the fluorinated
urethane group in SI-46/11-URF1 and SI-66/19-URF1
at temperatures of 326 and 318 °C, respectively. SI-26/
7-URF1 only shows a weak peak at this temperature
(see Figure 1). However, a strong decomposition occurs
at a much lower temperature of 232 °C. This decomposi-
tion step can be attributed to the cleavage of an
allophanate structure, formed during synthesis of SI-
26/7-URF1. According to literature, the allophanate
structure is much more instable and decomposes at
relatively low temperatures (200 °C18). Further evidence
of an allophanate decomposition is found since the
calculated loss of mass for this step matches well with
the experimentally found loss. In addition, FT-IR specif-
ically detects the disappearance of the allophanate
vibration at 1654 cm-1 after a thermal treatment (1 h
at 340 °C) as shown in Figure 2, whereas all other
signals attributed to R,R,R-trifluoro-p-toluyl side groups
only decrease. All other degradation steps that appear
at similar temperatures in every polymer investigated
can be attributed to the side chain cleavage of the
urethane and the degradation of the polymer backbone
(as an example see Figure 3). The fine structure
observed in the first derivative can be explained by an
additional two-step process that involves partial cy-
clization of monomer generated during the depolymer-
ization process and subsequent degradation.14,15,19
(b) F lu or in a ted Ca r bon a te Sid e Ch a in s (-OCO-
H1F 2). Only the SI-46/11-OH backbone was modified
with a fluorinated chlorocarbonate. The resulting poly-
mer shows two distinct degradation steps at 354 and
calcd loss exptl loss
1st
2nd
of mass
[%]
of mass
[%]
degradation degradation
polymer
step [°C]
step [°C]
SI-46/11-F3
SI-46/11-F7
SI-46/11-F8
SI-66/19-F3
SI-66/19-F7
SI-66/19-F8
40
56
59
40
56
59
54
63
68
53
65
63
340
345
346
342
344
347
405
395
401
401
394
407
419 °C as shown in Table 3 and Figure 4, which are
close to the cleavage temperature observed for perflu-
orinated ester side chains (see Table 4). We assume a
decomposition of the carbonate side chain in the first
degradation step as can be seen from the comparison
of the calculated and experimentally found loss of mass,
which agree quite well as shown in Table 3. Surpris-
ingly, the block copolymer did not show any expected
advantage with respect to a decrease of the decomposi-
tion temperature compared to the fluorinated ester side
chains. However, the fluorinated carbonate side chain
also allows for selective cleavage of fluorinated side
chains without decomposition of the polymer backbone.
(c) F lu or in a ted Ester Sid e Ch a in Meth a cr yla tes.
As it is impossible to perform any hydroboration reac-
tion in the presence of a methacrylic acid ester, a
monomer had to be found that already carries a hy-
droxyl function. Having in mind the previously dis-
cussed mechanism of thermal ester cleavage8 via a
retro-en reaction in a six-membered-ring transition
state, it is important to incorporate an ethylene spacer
between the hydrocarbon backbone and the ester func-

4934 Bo¨ker et al.
Macromolecules, Vol. 35, No. 13, 2002
tionality. Therefore, the 2-hydroxyethyl methacrylate
was used as monomer for the second block, as it fitted
well into these presumptions and allowed well-known
protection group chemistry as described earlier in this
paper.
The fluorinated side chain block copolymers based on
PMMA-b-PHEMA do not show any clearly separated
degradation steps. Therefore, an evaluation of the
experimentally found loss of mass was not possible, but
the temperature, at which the second degradation step
presumably occurs, could be estimated using a peak-
fitting procedure for the plot of the first derivative. If
one compares the TGA curve of MH-26/5 and MH-26/
5-F7 (not shown here), the main difference is a slight
shoulder in the first derivative of the curve of MH 26/
5-F7 with an assumed maximum at 316 °C (Table 3),
which can be attributed to the decomposition of the
fluorinated ester side chain. The same result is found
for MH-26/5-F8, which is slightly shifted to higher
temperatures. Obviously, the degradation temperature
of the polymer backbone is much lower for the meth-
acrylate (MH)-based copolymers compared to the poly-
styrene-b-polyisoprene (SI) copolymers, thus preventing
selective cleavage of perfluorinated ester side chains for
MH backbones. This behavior of the methacrylate-based
fluorinated side chain block copolymers can be explained
by the tendency of methacrylate polymers to depoly-
merize. Heating homopolymers of deliberately chosen
molecular weight to 500 °C showed that depending on
the nature of the neighboring group to the hydroxyl
function the monomer yield was 20% (free -OH), 50%
(acetyl group), or 70% (methyl group).20 Fundamental
studies21,22 on the degradation of PMMA have shown
that the thermal degradation takes place by a depolym-
erization reaction with a maximum rate of decomposi-
tion at ∼390 °C (Ea ) 130-230 kJ /mol22). Additionally,
recent molecular orbital studies have underlined that
in the case of poly(alkyl methacrylates) there is a
significant influence of the side chain on the degradation
process.23 A long and bulky side group finally leads to
a transition to a typical ester pyrolysis mechanism, e.g.,
poly(tert-butyl methacrylate), which decomposes by an
ester decomposition mechanism yielding isobutene and
poly(methacrylic acid) almost exclusively.24 Further-
more, Lee et al.25 demonstrated that free hydroxyl
groups lead to a significant decrease in the decomposi-
tion temperature and energy as they attack the ester
linkage and therefore facilitate the overall degrada-
tion.
F igu r e 5. TGA curve of SI-66/19-F8.
F igu r e 6. TDMS spectrum of SI-66/19-F8; heating rate 5
K/min.
will be investigated later in this work, after the bulk
behavior has been determined by TGA measurements.
All perfluorinated ester side chain block copolymers
in this study show a two-step degradation process that
can be explained by selective cleavage of the perfluori-
nated ester side chain and the decomposition of the
hydrocarbon backbone (see Table 4). The maxima of the
first derivative reveal a temperature gap of approxi-
mately 60 °C, which is around 20 °C lower than the gap
observed in the previous study due to a lower backbone
decomposition temperature as shown in Table 4 and
Figure 5. A clear explanation for this observation cannot
be given at this point. Additionally, the calculated and
experimentally obtained losses of mass do not fit as well
as found for SI-26/7-F3 and -F7 in a previous study.8
Nevertheless, within the experimental error, these
results again show the possibility to selectively cleave
the fluorinated side chains by thermal treatment. The
results will be further supported by TDMS measure-
ments in the next section. The TDMS spectrum of SI-
66/19-F8 is shown as an example in Figure 6. Here, the
desorption maxima of the fluorocarbon and the hydro-
carbon fragments can be found at 321 and 392 °C,
respectively. Slight deviations in temperature between
TGA and TDMS can be attributed to experimental
errors for the different setups.
The fact that our pure unmodified MH-26/5 degrades
in a single step at 364 °C, which is nearly 30 °C lower
than reported earlier25 for pure PMMA, may be due to
the strong influence of the hydroxyl groups. In the case
of the fluorinated ester side chains already a few
residual hydroxyl groups may be sufficient to induce an
OH-assisted degradation.
(d ) F lu or in a ted Ester Sid e Ch a in P S-b-P I Block
Cop olym er s. To complete the experimental results
concerning fluorinated side chain block copolymers
based on a SI backbone, the block copolymers based on
SI-46/11 and SI-66/19 backbones from previous studies8
were similarly investigated by TGA. These polymers
exhibit the highest fluorine enrichment in the surface
region as previously investigated by XPS. Therefore,
they are most favorable as materials for thermally
processed hydrophobic/hydrophilic patterned surfaces.
The selectivity of thermal cleavage in thin polymer films
Degr a d a tion Stu d y Usin g Th er m od esor p tion
Ma ss Sp ectr om etr y. Deter m in a tion of Degr a d a -
tion En er gies for Ester Bon d Clea va ge. Degradation
energies for the ester bond cleavage of the SI-block
copolymers were determined from heating rate varia-
tions in TDMS by several orders of magnitude. However,
experiments were practically limited to heating rates

Macromolecules, Vol. 35, No. 13, 2002
Selective Alteration of Polymer Surfaces 4935
F igu r e 7. TDMS spectrum of SI-26/7-F7; heating rate 1
K/min.
F igu r e 9. Evaluation of degradation energies for peak I (b)
and peak II (9).
the decomposition process in our example. The temper-
atures at peak maximum and resulting decomposition
energies are shown in Table 5.
The calculation of the frequency factor νm which can
be interpreted on a molecular level similarly as the
Arrhenius frequency factor yields 1.1 × 1013 s-1 for peak
I, which scales well compared to common Arrhenius
factors for monomeric ester pyrolysis.9 The degradation
energy of 167 kJ /mol for the first peak which is at-
tributed to the cleavage of the fluorinated ester side
chain is of the same order of magnitude as for example
previously reported for the thermal decomposition of
acetic acid esters that revealed activation energies for
the pyrolysis of around 190 kJ /mol.9 The value calcu-
lated for our fluorinated ester is smaller than for the
hydrocarbon ester cited in the literature. As previously
reported,8 we assume that the strong electron with-
drawing effect of the fluorocarbon chain leads to activa-
tion of the ester linkage resulting in the smaller
decomposition energy.
XP S In vestiga tion s on Th er m a lly Mod ified P oly-
m er F ilm s. Since the characteristics of the XPS spectra
of fluorinated side chain block copolymers based on
polystyrene-b-polyisoprene backbones were discussed in
a previous paper,8 only the XPS spectra of the meth-
acrylate-based block copolymers (poly(methyl methacry-
late)-b-poly(2-hydroxyethyl-[g-perfluoroctanoyl] meth-
acrylate) and their blends will be discussed in detail. A
meaningful interpretation of the XPS data of the start-
ing polymer films is necessary before dealing with
thermally degraded polymer films.
The C 1s region of the high-resolution XPS spectrum
of the pure block copolymer MH-26/5-F7 at a takeoff
angle of 90° (Figure 10a) reveals complex signals
extending over an energy interval of approximately 12
eV. Although there are six chemically differently bonded
carbon atoms present in the polymer, only five could be
resolved clearly and were used for curve-fitting proce-
dures. According to published literature,30 the aliphatic
carbon signals were calibrated to 284.6 eV. The result-
ing peak assignments are also given in Figure 10a and
are based on the characteristic binding energies and
agree well with previously established references.30 The
ratios of the different carbon moieties as obtained by
curve fitting of the C 1s lines do not agree at all with
the expected ratios of the fluorinated side chain of the
fluorinated PHEMA block as expected for a pronounced
surface segregation of the fluorinated PHEMA block.
For the polymer MH-26/5-F7 one obtains the following
ratios of CF3:CF2:O-CdO:CH2-O groups for the per-
F igu r e 8. TDMS spectrum of SI-26/7-F7; heating rate 100
K/min.
ranging from 1 to 100 K/min. Previous studies showed
a very good agreement between the TDMS and TGA
data using SI copolymers.8
As an example, SI-26/7-F7 was selected for the TDMS
investigation and was heated starting from room tem-
perature to 600 °C as shown in Figure 7 and Figure 8
for heating rates of 1 and 100 K/min, respectively. Here,
fragments originating from the fluorocarbon side chain
can be well distinguished from those of the hydrocarbon
backbone. Fluorocarbon fragments are observed by a
+
homologous series following the formula CnF2n+1 and
+
CnF2n (CF3+/69 amu, C2F5+/119 amu, C3F7+/169 amu,
CF2+/50 amu, C2F4+/100 amu, and C5F11+/269 amu for
the perfluoroctanoyl side chain). For the hydrocarbon
backbone, the most intense fragments observed were the
highly stable aromatic ions: C6H5+/77amu, C6H6+/78
amu, C7H7+/91 amu, and C8H8+/104 amu. The relative
yield of these ions was monitored during heating.
Degradation energies ED were determined from a
kinetic model similar to the Polanyi-Wigner model26,27
of thermal desorption. Here, the peak temperature of
desorption (in our case decomposition) Tp at a heating
rate â is determined, yielding28,29
2
ln(Tp /â) ) ED/RTp + ln(ED/Rνm)
(1)
2
Plotting ln(Tp /â) vs 1/Tp yields a slope ED/R as shown
in Figure 9. The intersection with the ordinate gives the
frequency factor νm that can be calculated by the term
ln(ED/Rνm). A first-order desorption process yields a
constant desorption energy and a desorption tempera-
ture (i.e., temperature at peak maximum) that only
depends on the heating rate. The same is assumed for

4936 Bo¨ker et al.
Macromolecules, Vol. 35, No. 13, 2002
Ta ble 5. Degr a d a tion Tem p er a tu r es a n d En er gies for SI-26/7-F 7
heating rate
[K/min]
temp at peak
max I [°C/K]
temp at peak
max II [°C/K]
ED for peak I
[kJ /mol]
ED for peak II
νm for peak I
νm for peak II
[kJ /mol]
[1013 s-1
1.11
]
[1013 s-1
24.5
]
1
10
100
303/576
340/613
385/658
376/649
410/683
460/733
167a
205a
a
With an estimated error of (10% derived from the error in temperature detection and the linear fit.
F igu r e 11. Fluorine content in the upper surface region of
MH-26/5-F7 (9) and its 10 wt % blend (2) with APEC 35.
Ta ble 6. XP S Resu lts of Hea ted a n d La ser -Tr ea ted F ilm s
amount of fluorine [at. %]
F igu r e 10. (a) C 1s regional scan of MH-26/5-F7. (b) C 1s
regional scan of SI-46/11-F8 before (s) and after heating
(‚ ‚ ‚) at 330 °C for 15 min. (c) 10% MH-26/5-F7 in APEC 35
before (s) and after (‚ ‚ ‚) laser-induced cleavage.
before
after
heating
before
laser
after
laser
polymer
heating
SI-46/11-F3
SI-46/11-F8
10% MH-26/5-F7
in APEC 35
32.5
45.8
18.7
<1
<2
<1
fluoroctanoyl side chain (F7, expected 1:6:1:1, observed
1:10:14:24). The high presence of CH2-O groups sug-
gests that a remarkable amount of PMMA block resides
in the uppermost 50 Å of the polymer film as detected
by XPS. This assumption is supported by a quantitative
evaluation of the C 1s , O 1s, and F 1s peak integrals of
the spectra measured at a takeoff angle perpendicular
to the sample surface (90°). They show a fluorine
amount in the uppermost 50 Å of the polymer film that
is as low as calculated for the bulk composition (17 at.
%).
Angle-dependent XPS measurements of films of the
pure block copolymer reveal that the fluorine enrich-
ment increases remarkably at smaller takeoff angles,
reachingswithin the experimental errorsthe value of
the isolated fluorinated PHEMA block (45 at. %) at a
relative depth of 0.2 (takeoff angle: 10°).
The result that the fluorine surface enrichment is
detected in a much shallower surface layer than for the
SI-block copolymers discussed in our previous study can
be explained by the fact that the degree of polymeriza-
tion (DP) of our fluorinated PHEMA blocks is much
lower than the DP of the fluorinated isoprene blocks.
Therefore, the domain spacing is smaller, which results
in a much thinner fluorinated surface layer. In addition,
a comparison between angle-dependent measurements
of the pure block copolymer and its 10 wt % blend with
polycarbonate APEC 35 (polycarbonate copolymer based
on BPA and TMC bisphenols containing 35 wt %
trimethylcyclohexylbisphenol (TMC) and 65 wt % bis-
phenol A (BPA, Bayer) is shown in Figure 11. A very
similar fluorine enrichment in the upper surface area
is observed, which implies thatssimilar to the fluori-
nated SI block copolymer blends8sonly a small amount
of fluorinated component is sufficient to change the
surface characteristics remarkably. We used these
blends for surface patterning experiments by laser-
19.3
3.3
induced thermal cleavage as presented in the next
section.
(a) Th er m al Cleavage of Flu or in ated Side Ch ain s.
Film surfaces were investigated by XPS before and after
heating. The results are summarized in Figure 10b and
Table 6. For SI-46/11-F3 as an example, the fluorine
content in the surface region of the film decreased from
32.5 at. % to less than 1 at. %. The fluorine content of
SI-46/11-F8 decreases from nearly 46 at. % to lower than
2 at. % after the thermal treatment. The resulting XPS
spectrum before and after thermal treatment is shown
in Figure 10b. After thermal treatment the fluorine-
containing groups at binding energies from 289 to 295
eV are nearly absent. Also, a 10 wt % blend of MH-26/
5-F7 in APEC 35 reveals a dramatic decrease in the
fluorine signal from 19 at. % to lower than 1 at. %. After
thermal treatment the corresponding XPS spectrum (not
shown here) indicates only pure polycarbonate. These
dramatic changes in the surface composition as sum-
marized in Table 6 agree well with the TGA and TDMS
studies which implied a complete removal of the fluori-
nated compounds from our block copolymer backbones.
To get more detailed information about the structure
of the remaining block copolymer, 200 nm thick films
were prepared by spin-coating onto double side polished
silicon wafers. This sample support allows for FT-IR
measurements at different times during the degradation
process. As can be seen in Figure 12, the polymer SI-
46/11-F8 shows a significant decrease in the CFn (1140-
1260 cm-1) and CdO (1786 cm-1) vibrations without a
significant change in the aromatic CdC (1600 cm-1) or
aromatic and aliphatic C-H (2800-3100 cm-1) vibra-
tions of the styrene block and backbone. Therefore, we
can assume that the fluorinated side chain is cleaved

Macromolecules, Vol. 35, No. 13, 2002
Selective Alteration of Polymer Surfaces 4937
Furthermore, we designed a block copolymer/homopoly-
mer blend system (10 wt % MH-26/5-F7 in APEC 35)
which allows to create hydrophobic surface patterns in
thin polymer films.
Possible applications of these chemically structured
surfaces range from printing plates using aqueous inks
to fundamental studies related to wetting phenomena
on defined chemically heterogeneous surfaces.
Ack n ow led gm en t. We gratefully acknowledge the
support by M. Voetz, S. Greb, U. Itter, and D. Ru¨hle
(Bayer AG) during the XPS, TGA, and contact angle
measurements. We also thank W. Ebert (Bayer AG) for
the polymer-analogous synthesis of the carbonate side
chain block copolymer and C. K. Ober (Cornell Univer-
sity) and his group for the synthesis of the block
copolymer SI-26/7. In addition to that, the support by
K. Dierksen (Agfa Gevaert, Mortsel) with the laser
experiments is gratefully acknowledged.
F igu r e 12. FT-IR of SI-46/11-F8 before (s) and after heating
(‚ ‚ ‚) at 330 °C for 15 min (thin film cast on double side
polished silicon wafer).
off selectively, leaving the remaining polymer backbone
completely intact. Additionally, we found that, as the
fluorine content in the surface region decreases dra-
matically, the contact angles also reveal a significant
change in hydrophilicity. For SI-46/11-F3, the static
contact angle of water decreases from 100° to 51°, and
for SI-46/11-F8 it drops from 113° to 86°, leading to a
significantly increased hydrophilicity.
Refer en ces a n d Notes
(1) Pittman, A. G. In Fluoropolymers; Wall, L. A., Ed.; Wiley:
New York, 1972; Vol. 25.
(2) Scheirs, J ., Ed. Modern Fluoropolymers; Wiley: Chichester,
1997.
(3) Katano, Y.; Tomono, H.; Nakajima, T. Macromolecules 1994,
27, 2342.
(4) Hopken, J .; Mo¨ller, M. Macromolecules 1992, 25, 1461.
(5) Iyengar, D. R.; Perutz, S. M.; Dai, C. A.; Ober, C. K.; Kramer
E. J . Macromolecules 1996, 29, 1229.
(6) Shull, K. R.; Kramer, E. J . Macromolecules 1990, 23, 4769.
(7) Dai, K. H.; Shull, K. R.; Kramer, E. J . Macromolecules 1992,
25, 220.
(8) Bo¨ker, A.; Wang, J .; Reihs, K.; Stadler, R.; Ober, C. K.
Macromolecules 2000, 33, 1310.
(9) DePuy, C. H.; King, R. W. Chem. Rev. 1960, 60, 431.
(10) Fukunaga, K.; Elbs, H.; Krausch, G. Langmuir 2000, 16,
3474.
(11) Bo¨ker, A.; Mu¨ller, A. H. E.; Krausch, G. Macromolecules, in
press.
(12) Singhvi, R.; Kumar, A.; Lopez, G. P.; Stephanopoulos, G. N.;
Wang, D. I. C.; Whitesides, G. M.; Ingber, D. E. Science 1994,
264, 696.
(13) Foti, S.; Guiffrida, M.; Maravigna, P.; Montaudo, G. J . Polym.
Sci., Polym. Chem. Ed. 1983, 21, 1567.
(14) Montaudo, G.; Puglisi, C.; Scamporrino, E.; Vitalini, D.
Macromolecules 1984, 17, 1605.
(15) Foti, S.; Maravigna, P.; Montaudo, G. J . Polym. Sci., Polym.
Chem. Ed. 1981, 19, 1679.
(16) Hirao, A.; Kato, H.; Yamaguchi, K.; Nakahama, S. Macro-
molecules 1986, 19, 1294.
(17) Wang, J .; Mao, G.; Kramer, E. J .; Ober, C. K. Macromolecules
1997, 30, 1906.
(18) Elias, H. G. Von Monomeren und Makromoleku¨len zu Werk-
stoffen; Hu¨thig & Wepf: Zug, Switzerland, 1996; p 111.
(19) Durairaj, B.; Dimock, A. W.; Samulski, E. T. J . Polym. Sci.,
Polym. Chem. Ed. 1989, 27, 3211.
(20) Razga, J .; Petranek, J . Eur. Polym. J . 1975, 11, 805.
(21) Manring, L. E. Macromolecules 1991, 24, 3304.
(22) Denq, B. L.; Hu, Y. S.; Chiu, W. Y.; Chen, L. W.; Chiu, Y. S.
Polym. Degrad. Stab. 1997, 57, 269.
(b) La ser -In d u ced Clea va ge of F lu or in a ted Sid e
Ch a in s. Thermal cleavage of fluorinated side chains can
be similarly obtained by IR laser treatment of the
polymer film surfaces. A film of a blend of 10 wt % MH-
26/5-F7 in APEC 35 was irradiated at 1064 nm by a
neodymium glass laser. As APEC 35 shows sufficient
absorption, no additional laser dye had to be added. The
laser-induced cleavage of the fluorinated side chains was
investigated by XPS before and after laser treatment.
To prevent surface contamination by fluorinated frag-
ments, the samples were washed with methanol and
ethanol after the laser treatment. These solvents do not
alter the block copolymer film surfaces. Several repro-
ducible XPS measurements revealed a significant de-
crease in the fluorine content on the surface of the
polymer films (Figure 10c). The resulting XPS spectra
of films that were heated by a furnace and laser-treated
at 1064 nm give similar results that indicate only
polycarbonate after the treatment. The amount of
fluorine in the uppermost 50 Å decreases from nearly
20 at. % to values around 3 at. %, which is a change as
remarkable as observed for the thermal heating process
in a vacuum oven. In addition, wetting experiments
reveal that the hydrophobicity of the surface (static
contact angle: 103°) is altered to complete wetting.
These first laser printing experiments indicate that the
polymers may potentially be used for printing processes
using water-based ink.
(23) Awad, M. K. Polym. Degrad. Stab. 1995, 49, 339.
(24) Grassie, N.; MacCallum, J . R. J . Polym. Sci. 1964, A2, 983.
(25) Lee, M. Y.; Lee, T. S.; Park, W. H. Macromol. Chem. Phys.
2001, 202, 1257.
Con clu sion
We have demonstrated that a variety of different
fluorinated side groups (perfluorinated esters, fluori-
nated urethanes, and carbonates) incorporated into a
hydroxy functional block copolymer can be cleaved
selectively by means of thermal or laser-induced py-
rolysis. Especially the latter technique has been proven
to be able to selectively alter the surface composition
in a chemically controlled way, i.e., to change the surface
composition from one defined composition to another by
cleavage of the respective fluorinated side chains.
(26) Ehrlich, G. J . Appl. Phys. 1961, 32, 4.
(27) Menzel, D., Vanslow, R., Eds.; Chemistry and Physics of Solid
Surfaces IV; Springer: Berlin, 1982.
(28) Redhead, P. A. Vacuum 1962, 12, 203.
(29) Dawson, P. T.; Walker, P. C. Experimental Methods in
Catalytic Research; Academic Press: New York, 1976; Vol.
II.
(30) Beamson, G.; Briggs, D. High-Resolution XPS of Organic
Polymers; The Scientia ESCA 300 Database; Wiley: New
York, 1992.
MA011938V