Photochemical attachment of polymer films to solid surfaces via monolayers of benzophenone derivatives

Article abstract of DOI:10.1021/ja990962+

We report a simple and yet effective way to photochemically attach thin polymeric layers to solid surfaces. The system is based on a photoreactive benzophenone derivative that is bound to SiO₂ surfaces via a silane anchor. This substrate is the

Full text of DOI:10.1021/ja990962+

8766
J. Am. Chem. Soc. 1999, 121, 8766-8770
Photochemical Attachment of Polymer Films to Solid Surfaces via
Monolayers of Benzophenone Derivatives
†
‡
†
†,‡
Oswald Prucker, Christoph A. Naumann, J u1 rgen R u1 he, Wolfgang Knoll, and
Curtis W. Frank*,‡
Contribution from the Max-Planck-Institute for Polymer Research, Ackermannweg 10,
D-55128 Mainz, Germany, and Center on Polymer Interfaces and Macromolecular Assemblies (CPIMA),
Department of Chemical Engineering, Stanford UniVersity, Stanford, California 94305-5025
ReceiVed March 23, 1999
Abstract: We report a simple and yet effective way to photochemically attach thin polymeric layers to solid
surfaces. The system is based on a photoreactive benzophenone derivative that is bound to SiO2 surfaces via
a silane anchor. This substrate is then covered with a polymer film that is reacted with the benzophenone
moieties by illumination with UV light (λ > 340 nm). As a result of the photochemical reaction, a thin layer
of the polymer is covalently bound to the surface. Nonattached polymer is removed by extraction. As examples,
we have successfully attached thin layers of poly(styrene) and poly(ethyloxazoline). The thickness of the layer
is a function of the illumination time and the molecular weight of the polymer. The film thickness increases
linearly with the radius of gyration of the polymers used for attachment. Using this system, we were able to
photochemically attach up to 16 nm thick films of poly(styrene).
Introduction
Based on these observations, we were looking for a fast and
simple way to covalently attach ultrathin physisorbed layers of
polymers by subsequent chemical reactions. The procedure of
choice would ideally be independent of the chemical nature of
the polymer. This goal rules out any chemistry based on
functional groups as these procedures would inherently restrict
the number of polymers suitable for attachment. Hence, we
designed a system that is based on a light-induced reaction
between a photoreactive group on a surface and C-H bonds in
the backbone or in side groups of a polymeric overcoat. Our
system is schematically depicted in Figure 1. The photoreactive
group used here is the benzophenone moiety. This system was
chosen because the photochemistry of this compound is well-
known and because it attaches to C-H bonds in a wide range
The covalent attachment of ultrathin films of polymers to a
solid substrate is often desirable to enhance the stability of the
films against solvents and displacement reagents (e.g., water
for hydrophilic surfaces). As a consequence of this need,
1-13
numerous grafting protocols have been developed.
The most
common procedures use polymers that carry functional groups
at one or both chain ends or as pendent side groups that are
suitable for binding to the surface. All these strategies, however,
sometimes require extensive synthetic efforts as the preparation
of functionalized polymers is not an easy task. Also, one cannot
easily introduce other functional groups to the polymer film
because these groups must be inert against the often very
reactive anchor groups.
1
4-20
of different chemical environments.
Another advantage of
*
To whom correspondence should be addressed.
Max-Planck-Institute for Polymer Research.
Stanford University.
this and other photoreactive groups is the chemical inertness of
these materials in the absence of light. A detailed description
of the photochemistry of benzophenone can be found in the
†
‡
(
1) Ben Ouada, H.; Hommel, H.; Legrand, A. P.; Balard, H.; Papirer, E.
J. Colloid Interface Sci. 1988, 122, 441.
2) Bridger, K.; Fairhurst, D.; Vincent, B. J. Colloid Interface Sci. 1979,
8, 190.
2
0
general literature on photochemistry, but the reader is also
referred to a review article by Dorm a´ n and Prestwich19 where
its use for biochemical problems is also discussed. To our
knowledge, only two comparable systems exist in the literature.
One uses monolayers of nitrene that can be photochemically
(
6
(
(
3) Bridger, K.; Vincent, B. Eur. Polym. J. 1980, 16, 1017.
4) Dimitrenko, A. V.; Shadrina, N. E.; Ivanchev, S. S.; Ulinskaya, N.
N.; Volkov, A. M. J. Chromatogr. 1990, 520, 21.
5) Hashimoto, K.; Fujisawa, T.; Kobayashi, M.; Yosomiya, R. J.
Macromol. Sci. Chem. 1982, A18, 173.
6) Hashimoto, K.; Fujisawa, T.; Kobayashi, M.; Yosomiya, R. J. Appl.
Polym. Sci. 1982, 27, 4529.
7) Horn, J.; Hoene, R.; Hamann, K. Makromol. Chem. Suppl. 1975,
5, 329.
(
21
linked to the hydroxyl groups of dextrans. The other one uses
benzophenone modified polymers for the photochemical at-
tachment of these materials to polymeric substrates.22 Although
this technique might be considered the “reversed approach” as
(
(
3
(
(
(
8) Krenkler, K. P.; Laible, R. Angew. Makromol. Chem. 1953, 53, 101.
9) Laible, R.; Hamann, K. AdV. Colloid Interface Sci. 1980, 13, 65.
10) Papirer, E.; Nguyen, V. T.; Donnet, J.-B. J. Polym. Sci. 1979, 17,
(15) Smets, G. J.; Hamouly, S.; Oh, T. J. Pure Appl. Chem. 1984, 56,
439.
(16) Horie, K.; Morishita, K.; Mita, I. Macromolecules 1984, 17, 1746.
(17) Horie, K.; Mita, I. Eur. Polym. J. 1984, 20, 1037.
(18) Br a¨ uchle, C.; Burland, D. M.; Bjorklund, G. C. J. Phys. Chem. 1981,
85, 123.
(19) Dorm a´ n, G.; Prestwich, G. D. Biochemistry 1994, 33, 5661.
(20) Turro, N. J. Modern Molecular Photochemistry; University Science
Books: Mill Valley, 1991.
(21) Elender, G.; K u¨ hner, M.; Sackmann, E. Biosens. Bioelectron. 1996,
11, 565.
1
015.
11) Trachenko, V. I.; Zil’berman, Y. N.; Shatskaya, T. F.; Pomerantseva,
E. G. Polym. Sci. USSR 1986, 28, 646.
12) Tsubokawa, N.; Kuroda, A.; Sone, Y. J. Polym. Sci. 1989, A27,
701.
13) Tsubokawa, N.; Hosaya, M.; Yanadori, K.; Sone, Y. J. Macromol.
Sci. Chem. 1990, A27, 445.
14) Horie, K.; Ando, H.; Mita, I. Macromolecules 1987, 20, 54.
(
(
1
(
(
1
0.1021/ja990962+ CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/10/1999

Attachment of Polymer Films to Solid Surfaces
J. Am. Chem. Soc., Vol. 121, No. 38, 1999 8767
synthesis of this compound. Typically, 2 g of 1 was suspended in ca.
2
0 mL of freshly distilled dimethyl chlorosilane. Pt-C (10 mg, 10%
Pt) was added and the mixture was refluxed for 5 h. The excess
chlorosilane was removed in vacuo yielding the desired product 2 in
quantitative yields as an oil. FTIR (film): 3059, 3024, 2940, 2875,
-
1 1
1
0
6
(
C
651, 1600 cm . H NMR (CDCl
3
, δ in ppm): 0.3 (s, 6H, SiCH
3
),
),
.9 (m, 2H, SiCH ), 1.9 (m, 2H, CH
2
2
3
CH
2
CH
2
), 3.9 (t, 2H, OCH
2
1
.9-7.9 (various m, 9H, C-Harom). C NMR (CDCl
3
, δ in ppm): 1
), 112-137 (6 peaks,
arom), 161 (OCarom), 194 (CdO). The catalyst was removed by filtration
3 2 2 2 2 2
SiCH ), 13 (SiCH ), 21 (CH CH CH ), 68 (OCH
of a solution of 2 in toluene. This solution was then directly used for
the surface modification.
Immobilization of 2 on SiO
mobilized on the SiO surface of a silicon wafer at room temperature
from toluene solutions using Et N as catalyst and acid scavenger. The
solutions with the substrates were left to stand overnight. The samples
were then cleaned by rinsing extensively with chloroform. Film
thicknesses were determined by ellipsometry. Typically, values of 1.0
( 0.1 nm were obtained when a refractive index of the layer of n )
1.5 was assumed.
2
Surfaces. The silane 2 was im-
2
Figure 1. Schematic representation of the photochemical attachment
of polymer chains to solid surfaces via illumination of polymer-covered
monolayers of a benzophenone derivative (BP-ML).
3
compared to our system, it also differs significantly as the
method with the benzophenone-modified polymer yields surface-
attached polymer networks: Many photoreactive groups of the
polymer do not bind to the surface but to other chains of the
film-forming material.
This paper describes the synthesis of the photocoupling agent,
its immobilization on SiO2 surfaces, and the subsequent attach-
ment of poly(styrene) (PS) and poly(ethyloxazoline) (PEOX)
by illumination of spin-cast layers of these polymers on
benzophenone-modified surfaces followed by thorough extrac-
tion of nonbound chains. The two polymers were chosen because
they represent examples of very hydrophobic (PS) and hydro-
philic (PEOX) polymers and, therefore, demonstrate the ver-
satility of the system. In addition, we describe a number of
reference experiments that support the reaction pathway pro-
posed in Figure 1.
Preparation of the Polymer Layers. Thick overcoats (>100 nm)
of the polymers were prepared by spin-casting solutions of the polymers
at a typical spin speed of 2000 rpm for 1 min. Typical solvents were
toluene for PS and methanol for PEOX, and concentrations of 10 mg
-
1
mL were employed. The samples were dried in air and used directly
for illumination experiments. These experiments were performed at
room temperature using a high-pressure mercury UV lamp (500W,
Oriel). A water filter (8 cm) was used to remove IR light from the
beam, and a dichroic mirror eliminated short wavelengths with λ <
3
40 nm. The integral light intensity at the sample location was 100
-
2
mW cm . After illumination for the desired period of time, we
extracted the samples in a Soxhlet apparatus with good solvent (PS:
toluene; PEOX: methanol) for at least 10 h to remove nonbonded
polymer. The thicknesses of the resulting polymer layers were again
determined by ellipsometry assuming the respective bulk refractive
indices of the polymers (PS: n ) 1.59; PEOX: n ) 1.52).
Experimental Section
Results and Discussion
Materials. Toluene was distilled from molten sodium using ben-
zophenone as indicator and triethylamine was distilled from calcium
hydride. Both were stored under a nitrogen atmosphere. Dimethyl
chlorosilane was distilled and then immediately used for hydrosilation.
All other chemicals and solvents (HPLC grade) were used as received.
All reactions that involved chlorosilanes were performed under an
atmosphere of dry argon.
The benzophenone silane 2 can be synthesized in high yields
following simple procedures as in Figure 2. The first step is a
Williamson ether synthesis of 4-allyloxybenzophenone 1 from
4-hydroxybenzophenone and allyl bromide. Compound 1 can
then be hydrosilated with dimethyl chlorosilane using platinum
24,25
on charcoal (Pt-C, 10% Pt) as catalyst.
The more commonly
Synthesis of 4-Allyloxybenzophenone (1). This compound was
used catalyst for hydrosilation reactions, hexachloroplatinic acid
H2PtCl6), is not suitable for this compound as it also leads to
23
synthesized by standard procedures. In a typical run, 39.6 g (0.2 mol)
-hydroxybenzophenone and 26.6 g (0.22 mol) allyl bromide were
(
4
the reduction of the benzophenone carbonyl unit to a methylene
moiety. Such reactions of carbonyl groups are known in the
dissolved in 120 mL of acetone and 28 g of potassium carbonate was
added. The mixture was heated to reflux for 8 h and then cooled to
room temperature. Water (80 mL) was added and the resulting solution
was extracted twice with 100 mL of diethyl ether. The combined organic
phases were washed twice with 100 mL of aqueous NaOH (10%) and
2
4
literature.
The silane 2 can be immobilized on SiO2 surfaces by
-3
immersing an appropriate substrate into a dilute (typically 10
dried over Na
2
SO
4
, and the solvent was evaporated. The resulting,
M) solution of 2 in toluene. A few drops of dry Et N are added
3
slightly yellowish raw product was recrystallized from methanol to yield
26-28
to bind the resulting HCl and to act as a catalyst.
This
4
1
2 g (90%) of 1. FTIR (KBr): 3081, 3059, 3022, 2939, 2865, 1650,
procedure yields monolayers of the benzophenone silane (BP-
ML) of a typical thickness of ca. 1 nm.
-
1 1
600 cm . H NMR (CDCl
d), 6.1 (m, 1H, dCHs), 6.9-7.9 (various m, 9H,
3
, δ in ppm): 4.6 (m, 2H, OCH ), 5.3-5.5
2
(m, 2H, CH
2
The polymer overcoats were then deposited by spin-coating.
We typically coat rather thick films of h > 100 nm to avoid
any influence of the thickness of this film on the final thickness
of the covalently bound layer. Also, it is obvious that thick
overcoats should be more homogeneous and essentially pinhole
free. These layers were then illuminated with UV light.
Benzophenone and many of its derivatives show an absorption
of UV light around 345 nm caused by a n,π* transition in the
1
3
C-Harom.). C NMR (CDCl
95 (CdO), 110-140 (9 peaks, remaining Carom and sCHd).
Synthesis of 4-(3′-Chlorodimethylsilyl)propyloxybenzophenone
3 2
, δ in ppm): 69 (OCH ), 162 (OCarom),
1
2
4,25
(2). Standard hydrosilation procedures
were employed for the
(22) Amos, R. A.; Anderson, A. B.; Clapper, D. L.; Duquette, P. H.;
Duran, L. W.; Hohle, S. G.; Sogard, D. J.; Swanson, M. J.; Guire P. E. In
Encyclopedic Handbook of Biomaterials and Bioengineering, Part A.
Materials; Wise, D. L., Trantolo, D. J., Altobelli, D. E., Yaszemski, M. J.,
Gresser, J. D., Schwartz, E. R., Eds.; Marcel Dekker: New York, 1995;
Vol. 1, p 895.
(26) Boksanyi, L.; Liardon, O.; Kovats, E. s. AdV. Colloid Interface Sci.
1976, 6, 95.
(23) Claisen, L.; Eisleb, O. Liebigs Ann. Chem. 1913, 401, 21.
(
24) Marciniec, B. ComprehensiVe Handbook of Hydrosilylation; Per-
(27) Kallury, K. M. R.; Macdonald, P. M.; Thompson, M. Langmuir
1994, 10, 492.
gamon Press: Oxford, 1992.
(25) Speier, J. L. AdV. Organomet. Chem. 1979, 17, 407.
(28) Tripp, C. P.; Hair, M. L. J. Phys. Chem. 1993, 97, 5693.

8768 J. Am. Chem. Soc., Vol. 121, No. 38, 1999
Prucker et al.
Figure 2. Synthesis of the benzophenone silane 2 and its immobiliza-
tion on SiO surfaces.
2
Figure 4. Influence of the thickness, dsc, of the spin-cast layer on the
saturation thickness dmax after 30 min of illumination followed by
6
-1
extraction; PS, M
w
) 1.3 × 10 g mol , integral light intensity 100
-
2
mW cm . The dotted line indicates the reference case of complete
attachment.
saturation values are reached after about 10-20 min of
illumination under the described conditions. The differences
between the polymers may be due to the different molecular
weights used in this study, although other factors may also be
of some importance. One could, for example, envisage different
conformations of the two types of polymers directly at the
interface with the substrate due to different interactions between
the benzophenone layers and the polymers. Also, there could
be differences in the photochemistry for these polymers, as it
is known, for example, that hydrogen atoms on aromatic rings
Figure 3. Influence of the illumination time on the photochemical
attachment of PS and PEOX to benzophenone monolayers (BP-ML).
The integral light intensity was adjusted to 100 mW cm-2
.
(
as those of the phenyl rings of PS) are less easily abstracted
18
carbonyl group. The resulting triplet is biradical in nature and
will therefore react with groups in its vicinity or return to the
electronic ground state mainly by phosphorescence.20 The most
prominent reaction of a benzophenone triplet is schematically
depicted in Figure 1. In the first step of this reaction the oxygen
of the carbonyl group abstracts a hydrogen atom from the
reaction partner leaving behind two radicals , at the previous
carbonyl carbon and the carbon of the other molecule, that
eventually recombine. Figure 1 demonstrates that, in our case,
the attachment of the benzophenone or its carbonyl group to a
polymer chain also leads to the covalent attachment of this chain
to the surface.
by benzophenone than those in aliphatic environments. Ad-
ditionally, in the case of PEOX it should be noted that electron-
transfer reactions of the benzophenone triplet often occur in
1
9,20
the presence of nitrogen atoms.
These reactions will not
necessarily lead to the attachment of the polymer chain to the
benzophenone monolayer.
To ensure that this film formation is indeed due to the light-
induced coupling of the surface benzophenone groups to the
polymer chains of the physisorbed films, we performed two sets
of reference experiments. In the first set, we simply illuminated
spin-cast PS and PEOX layers deposited directly on unmodified
silicon wafers. In the absence of the benzophenone layer, no
coupling should occur. Indeed, even after prolonged illumination
times (2 h) we were able to wash all polymer off the surfaces
by means of Soxhlet extraction with good solvent (toluene for
PS and methanol for PEOX). In most cases, even simple rinsing
procedures were sufficient to remove the polymer layers from
the surfaces. This clearly demonstrates that the observed film
formation based on this benzophenone attachment chemistry is
due to the light-induced coupling between the two layers and
cannot be caused by side reactions such as cross-linking. Also,
spin-cast layers of both polymers could be easily washed away
from benzophenone modified silicon wafers that were not
illuminated.
We have performed various sets of experiments each for PS
and PEOX to study this procedure for the photochemical
attachment of thin polymer coatings to solid surfaces. All
experiments were carried out at comparable light intensities. In
the first set of experiments, we varied the illumination times of
PS and PEOX layers deposited on benzophenone-modified
silicon wafers. The molecular weights of the polymers used in
-
1
-1
these studies were 233000 g mol (PS) and 50000 g mol
PEOX). The illumination times were varied between 5 min
(
and 2 h. After that the samples were rigorously extracted with
toluene (PS) or methanol (PEOX) in a Soxhlet apparatus for at
least 10 h to remove nonbonded chains. We found that the
extractions were necessary because simple rinsing procedures
are not sufficient to wash away all nonbound chains. The
thicknesses determined after only a few minutes of rinsing were
higher than the respective values obtained after extraction. Also,
the scattering of the data was much more pronounced if the
samples were only rinsed. This observation may essentially be
due to entanglements of nonbound chains with those that are
attached to the surfaces.
Figure 3 shows the results of the ellipsometric measurements
of the film thicknesses after extraction as a function of
illumination time. For each polymer the thickness of the bound
layer first increases and then levels off to reach a constant value
of 5 nm for PS and 3 nm for PEOX. In both cases these
In another set of experiments, we investigated the influence
of the thickness of the spin-cast layer on the thickness of the
resulting bound layer after illumination and extraction. Here,
6
-1
we used poly(styrene) with Mw ) 1.3 × 10 g mol and
prepared spin-cast layers with thicknesses ranging from 6 to
140 nm. The results are shown in Figure 4, where dsc denotes
the thickness of the spin-cast layer and dmax stands for the
thickness of the covalently attached PS layer after 30 min of
illumination followed by Soxhlet extraction. The dotted line
indicates the reference case of complete attachment of the film
which is not expected. The thickness of the attached layer is
always smaller than the starting thickness after spin-coating even
for very thin films. Initially about half of the polymer overcoat

Attachment of Polymer Films to Solid Surfaces
J. Am. Chem. Soc., Vol. 121, No. 38, 1999 8769
Figure 6. Possible methods for the photochemical immobilization of
ultrathin organic or polymeric films using monolayers of benzophenone
derivatives.
Figure 5. Influence of the chain dimensions on the photochemical
attachment of PS to benzophenone monolayers. Plot A shows the
dependency of the maximum film thickness on the molecular weight
of the polymers. Plot B shows the same data as in plot A but
recalculated for the bulk radii of gyration, R
samples were illuminated for 30 min at an integral light intensity of
g
, according to eq 1. The
in their unperturbed Gaussian conformation. The bound thick-
ness will be restricted to coils having contact with the ben-
zophenone-modified substrate and, hence, chains with their
center of gravity being further away from the surface than Rg
will not be attached. This argument is further substantiated if
we reinvestigate the dependence of the bound thickness dmax
on the thickness dsc of the spun-cast layer as presented earlier
in Figure 4. For the PS used for these experiments with Mw )
-
2
1
00 mW cm .
can be attached via the benzophenone units on the surface. Soon,
however, the curve levels off at about 11.5 nm for this particular
poly(styrene) with a rather high molecular weight.
It is rather obvious from the experiments described so far
that the molecular weight of the polymers used for film
preformation is a crucial parameter for this attachment chem-
istry. Therefore, we studied this influence and again selected
an illumination time of 30 min which was found to be fully
sufficient to reach the saturation thickness dmax. For PS we varied
the molecular weight over a range from 30000 to 1500000 g
6
-1
1
.3 × 10 g mol we found that dmax no longer increases when
spun-cast films with thicknesses of dsc higher than about 30-
0 nm are illuminated. This range compares rather well with
4
the Rg ) 31 nm of this particular PS. Again, we recognize that
chains inside the spun-cast layer that are not in contact with
the surface cannot be attached. Hence, it is not surprising that
the curve shown in Figure 4 levels off at a value for dsc that is
related to the coil dimensions of the polymer used for attach-
ment.
-1
-1
mol ; for PEOX two samples with 50000 and 500000 g mol
were used. Again, the samples were extracted after illumination,
and the thicknesses were determined by ellipsometry.
The results for PS are plotted in Figure 5A. The values clearly
indicate that the thickness dmax that can be obtained by this
method increases with increasing molecular weight of the
polymers used for film formation. This becomes even more
obvious if the data are plotted as a function of the bulk radius
of gyration (Rg) of the polymers, as shown in Figure 5B. Here,
we find a linear relationship between dmax and Rg. The Rg values
given there were calculated from the weight average molecular
weight Mw according to eq 1:2
Conclusions
In this paper we introduce a new way to stabilize thin polymer
films by photochemical attachment via monolayers of a pho-
toreactive group. The silane that carries this group can be readily
synthesized from simple chemicals and forms monolayers that
are typical for structurally comparable compounds. Molecules
of subsequently deposited layers of polymers, which are in direct
contact with the surface, can be bound to the benzophenone-
modified substrate by illumination with UV light of appropriate
wavelength. Typically, several nanometers of the polymeric
overcoat can be attached. The thickness dmax is a function of
the molecular weight or, more precisely, of the chain dimensions
of the polymer used for film formation: dmax increases linearly
with the bulk radius of gyration of the polymer. Other factors
9,30
Rg ) 0.0275 M (in nm)
(1)
x
w
For the two PEOX molecular weights that we tested, we
qualitatively observed a similar molecular weight dependence;
films prepared from Mn ) 50000 g mol-1 yielded saturation
thicknesses of 3 nm whereas 7.2 nm was obtained from samples
-
1
made from PEOX with Mn ) 500000 g mol .
(
29) Ballard, D. G. H.; Wignall, G. D.; Schelten, J. Eur. Polym. J. 1973,
, 965.
30) Cotton, J. P.; Decker, D.; Benoit, H.; Farnoux, B.; Higgins, J.;
Jannick, G.; Ober, R.; Picot, C.; Cloizeaux, J. d. Macromolecules 1974, 7.
The linear relationship between the obtained thickness and
the coil dimensions of the polymers is, of course, not surprising
as it is entirely consistent with most of the polymer coils being
9
(

8770 J. Am. Chem. Soc., Vol. 121, No. 38, 1999
Prucker et al.
that influence the conformation of polymer chains near interfaces
may also be of some importance, e.g., attractive forces between
the polymer and the surface like hydrogen bonding, but were
not checked explicitly in this study. As first examples, we have
chosen PS and PEOX, the first being a rather hydrophobic and
the latter a hydrophilic polymer, to demonstrate the versatility
of the system.
Finally, we want to point out that this approach is not limited
to the attachment of spin-cast layers of polymers. Other methods
of film preformation are also suitable. A clear advantage of this
photochemical mode of attachment is the chemical inertness of
the benzophenone moiety in the absence of light. This makes it
suitable for the subsequent attachment of Langmuir-Blodgett-
Kuhn (LBK) layers deposited onto benzophenone-modified
surfaces as the contact with the subphase water during the
dipping process will not cause any side reactions in the
benzophenone monolayer. Besides that, one can also envisage
direct chemisorption processes from solutions of polymers or
low molecular weight compounds that can otherwise not easily
be immobilized. Figure 6 summarizes this outlook. Some of
these examples are currently being investigated in our groups.
Altogether, the thicknesses that can be realized with this type
of chemistry are of the same order or even higher as for other
-13
“
grafting-to” techniques.1 The advantage of this photochemi-
cal route to covalently attached polymer films clearly is the
simplicity of the silane synthesis and the versatility with regard
to the polymers that can be used for attachment. If, however,
thicker films are to be prepared, one has to use a different
approach where the polymers are formed in situ on the surface
by means of monolayers of initiators (“grafting-from” tech-
nique).3
Acknowledgment. Oswald Prucker and Christoph A. Nau-
mann want to thank the Deutsche Forschungsgemeinschaft
(DFG) for their fellowships (“Forschungsstipendien”). We would
1-34
also like to thank Martina Knecht for valuable technical support.
This work was supported in part by the Center on Polymer
Interfaces and Macromolecular Assemblies (CPIMA) through
the NSF-MRSEC program under DMR 94-00354.
(
(
(
31) Prucker, O.; R u¨ he, J. Macromolecules 1998, 31, 602.
32) Prucker, O.; R u¨ he, J. Macromolecules 1998, 31, 592.
33) Prucker, O.; Schimmel, M.; Tovar, G.; Knoll, W.; R u¨ he, J. AdV.
Mater. 1998, 10, 1073.
34) Prucker, O.; R u¨ he, J. Langmuir 1998, 14, 6893.
(
JA990962+