PEG-based hydrogel synthesis via the photodimerization of anthracene groups

Article abstract of DOI:10.1021/ma012263z

Photo-cross-linking has received a considerable attention for the design of intelligent materials in biochemical and biomedical applications. In this report, we describe the synthesis and properties of a novel photoreversible poly(ethylene glycol)- (PEG-) based hydrogel system. 9-Anthracene-carboxylic acid was used to modify the hydroxyl groups of an eight-armed PEG polymer (molecular weight 20 000) and the degree of substitution was determined to be 87.4%. The PEG-anthracene macromers (PEG-AN) exhibited high photosensitivity at wavelengths close to visible light (absorption maxima at 366 and 380 nm) and underwent rapid and reversible photo-cross-linking upon exposure to alternating wavelengths of irradiation (365/254 nm) in the absence of photoinitiators or catalysts. Changes in light exposure and wavelength of irradiation reversibly altered the physicochemical properties of the PEG-AN hydrogel, including swellability, absorption spectrum, and topography.

Full text of DOI:10.1021/ma012263z

5228
Macromolecules 2002, 35, 5228-5234
PEG-Based Hydrogel Synthesis via the Photodimerization of
Anthracene Groups
Yu ju n Zh en g,† Miod r a g Micic,† Sa r ita V. Mello,† Mu sta p h a Ma br ou k i,†
F otios M. An d r eop ou los,*,‡ Veer a n ja n eyu lu Kon k a ,† Si M. P h a m ,‡ a n d
Roger M. Lebla n c*,†
Center for Supramolecular Science and Department of Chemistry, University of Miami,
Coral Gables, Florida 33124-0431; and Department of Surgery, School of Medicine, and
Department of Biomedical Engineering, University of Miami, Miami, Florida 33136
Received December 28, 2001; Revised Manuscript Received March 25, 2002
ABSTRACT: Photo-cross-linking has received a considerable attention for the design of intelligent
materials in biochemical and biomedical applications. In this report, we describe the synthesis and
properties of a novel photoreversible poly(ethylene glycol)- (PEG-) based hydrogel system. 9-Anthracene-
carboxylic acid was used to modify the hydroxyl groups of an eight-armed PEG polymer (molecular weight
20 000) and the degree of substitution was determined to be 87.4%. The PEG-anthracene macromers
(PEG-AN) exhibited high photosensitivity at wavelengths close to visible light (absorption maxima at
366 and 380 nm) and underwent rapid and reversible photo-cross-linking upon exposure to alternating
wavelengths of irradiation (365/254 nm) in the absence of photoinitiators or catalysts. Changes in light
exposure and wavelength of irradiation reversibly altered the physicochemical properties of the PEG-
AN hydrogel, including swellability, absorption spectrum, and topography.
In tr od u ction
cross-linking systems (e.g., photodimerization reactions)
are intrinsically competed by E/Z isomerization, which
greatly compromises the efficiency of photodimerization.
Photosensitive groups such as coumarin,5 stilbazolium,6
thymine,7 cinnamylidene,1d,8 etc. have been used to
prepare photo-cross-linked networks which exhibited
superior photostability and photoefficiency but mini-
mum reversibility. Andreopoulos et al.1d,9 developed a
photoswichable PEG-based hydrogel (PEG-cinnamyl-
idene) and demonstrated that the physical properties
(mesh size and swellability) of the hydrogel membrane
were controlled in a predictable fashion by alternating
the wavelength (>300/254 nm) and sequence of irradia-
tion. The photoreversibility of the PEG-cinnamylidene
hydrogel however was reported later to be strongly
limited due to photoscissive light inefficiency, cin-
namylidene photodegradation, and side polymerization
reactions.10 Our group has recently reported the syn-
thesis of a photo-cross-linked poly(ethylene glycol)-
nitrocinnamate (PEG-NC) hydrogel which demon-
strates a photoscissile behavior upon shortwave UV
irradiation.11 The PEG-NC macromers exhibited fast
gelation times and superior storage stability over their
cinnamylidene and acrylate analogues.
Over the past decade, numerous studies are focused
on the construction of novel polymeric networks by
means of photo-cross-linking.1 Photo-cross-linking ex-
hibits special advantages over conventional physical or
chemical methods of network formation: mild reaction
conditions, minimum side-product formation, fast curing
times, and spatial control of the polymerization reaction.
The physicochemical properties of the photo-cross-linked
polymer network can be easily modulated by adjusting
the sequence of illumination; the swelling ability of
photo-cross-linked hydrogels can be tailored by merely
changing the irradiation time.1,2 Photo-cross-linking
reactions usually involve (a) the free radical polymeri-
zation of pendant acrylate or methacrylate groups along
the polymeric backbone of hydrophilic polymers in the
presence of an initiator or catalyst3 or (b) the photo-
dimerization of cinnamates attached as end groups in
multiarmed polymers.4
We are interested in developing highly efficient pho-
toresponsive materials with tunable properties. We
envision these materials to have potential applications
in developing optical switches, controlled delivery car-
riers or biosensors. Current photoinduced systems are
associated with various shortcomings. For example,
acrylate-based polymers, even though they demonstrate
fast curing times, the rate of polymerization usually
depends on the nature and concentration of potentially
toxic initiators, the photopolymerization leads to ir-
reversible cross-links, and the polymers are unstable for
storage. On the other hand, cinnamates-based photo-
Our continuous research effort is targeted toward the
development of novel, highly photosensitive polymers
that undergo rapid reversible transitions. In the present
work, we report the synthesis of a novel PEG-based
hydrogel based on the reversible photodimerization of
the anthracene functionality. Anthracene and its de-
rivatives undergo [4 + 4] dimerization upon longwave
* Corresponding authors. R.M.L. Telephone: (305) 284-2282.
Fax: (305) 284-4571. E-mail: rml@umiami.ir.miami.edu. F.M.A.
Telephone: (305) 355-5000 ext 5174. Fax: (305) 355-5074. E-
mail: fandreop@med.miami.edu.
† Center for Supramolecular Science and Department of Chem-
istry, University of Miami.
‡ Department of Surgery and School of Medicine, and Depart-
ment of Biomedical Engineering, University of Miami.
UV irradiation (>350 nm).12 The dimer can be restored
10.1021/ma012263z CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/24/2002

Macromolecules, Vol. 35, No. 13, 2002
PEG-Based Hydrogel Synthesis 5229
In the swelling experiments, a 10% w/v aqueous solution of
PEG-AN in a detachable cell (one slide is precoated with
SigmaCote to avoid sample adhering) was treated with light
irradiation (10 min of 365-nm light and 5 min of 254-nm light,
respectively) and then washed with water to remove any un-
cross-linked PEG-AN. The gel was allowed to equilibrate in
water for 1 h. Following equilibration the hydrogel was
removed from water and the exposed surface was carefully
blotted with a Kimwipe (Aldrich, St. Louis, MO). The degree
of swelling (DS) was determined from (Ws - Wd)/Wd, where
Ws is the weight of the swollen gel following 1 h equilibrium
in water and Wd is the weight of the dry gel (air-dried for 24
h). All measurements were performed in triplicates and the
standard deviation is reported (average ( SD).
UV-vis spectra of PEG-AN films were recorded on a
Lambda 900 UV/vis spectrometer (Perkin-Elmer, Inc., Shelton,
CT). For the UV-vis measurement of solid-state PEG-AN,
the sample was prepared by spreading a dilute dichlo-
romethane solution of PEG-AN (1 mg/mL) on a quartz slide.
After solvent evaporation, a thin solid PEG-AN film was left
on the quartz slide.
to the original monomers upon exposure to shortwave
UV light. This photochromism has been extensively
applied to study many intermolecular and intramolecu-
lar processes.13 The incorporation of anthracene and its
derivatives within a polymer matrix is not a new topic.14
For example, Horie and co-workers14b,14c utilized the
anthracene functionality to produce photoemissive probes.
Coursan and Desvergne14d reported the synthesis of
anthracene modified polystyrene polymers that exhib-
ited excellent photoreversibility. Despite these studies,
the use of anthracene moieties as cross-linking sites to
produce photoreversible hydrophilic biomaterials has
rarely been reported.15 Herein we modified a branched
poly(ethylene glycol) polymer with 9-anthracenecar-
boxylic acid to obtain photoreactive poly(ethylene glycol)-
anthracene macromers (PEG-AN). The PEG-AN mac-
romers underwent reversible photo-cross-linking when
exposed to alternating wavelengths of irradiations (365/
254 nm).
1H NMR was measured on a Bruker 300 MHz instrument.
A 5 mm quartz NMR tube was purchased from Wilmad-Glass,
Inc., Buena, NJ . PEG-AN solution in D2O (5% w/v) was placed
in the quartz tube, and the sample was irradiated with high
power UV lamps (100-W 365 nm and 50-W 254 nm, both from
UVP, Inc., Upland, CA).
Exp er im en ta l Section
Ma ter ia ls. Poly(ethylene glycol) (b-PEG, MW ) 20 000,
eight arms) was purchased from Shearwater Polymers Inc.
(Huntsville, AL). It was fully crushed and dried under high
vacuum before use. All other chemicals and organic solvents
were purchased from Sigma-Aldrich (St. Louis, MO) at their
highest purity. The deionized water used was purified by a
Modulab 2020 water purification system (Continental Water
Systems Corp., San Antonio, TX).
Environmental scanning electron microscopy (ESEM) im-
ages were recorded with a FeiCo Phillips-Electroscan FEG XL-
30 field emission gun ESEM microscope. A drop of aqueous
PEG-AN solution (10% w/v) was placed on a specimen holder
and exposed to the indicated irradiation sequence. No ad-
ditional sample treatment was performed, thus avoiding
introduction of possible coating artifacts. All the micrographs
were obtained in the environmental wet mode and collected
with a large-field gaseous secondary electron (GSE) detector.
As a protective atmosphere and imaging gas, a water vapor
pressure of 0.9 Torr was present in the chamber at all times.
To further prevent the sample from dehydration, we kept
water vapor in the saturated condition within the microscope
chamber by cooling the sample down using the Pertiler cold
stage. The sample was kept at a working distance of 7.7 mm
from the large field GSE detector. Images were acquired at a
magnification of 1000× and optimized for maximum sharpness
and contrast, using the charge contrasting technique.
Atomic force microscopy (AFM) images were taken with a
PicoSPM microscope (Molecular Imaging Inc., Phoenix, AZ).
Samples of 10% w/v aqueous solution of PEG-AN were spin-
coated on a mica substrate to form a thin film. The film was
then submitted to UV exposure. Following irradiation, the
sample was scanned in air using the constant force mode with
microfabricated Si3N4 tips attached to rectangular beams
(spring rate 0.09 N/m). The setting point was adjusted to
minimize the force between the tip and the sample. This force
was maintained constant between 1 and 5 nN.
Syn th esis of 9-An th r a cen eca r bon yl Ch lor id e. A 15 mL
aliquot of SOCl2 (0.21 mol) was placed in a N2-purged flask
followed by the addition of 3.0 g of 9-anthracenecarboxylic acid
(13.5 mmol). One drop of anhydrous DMF was added to start
the reaction. After 5 min, the solid dissolved completely.
Temperature was increased to 60 °C, and the reaction mixture
was kept overnight under stirring. Upon reaction completion,
unreacted SOCl2 was evaporated under reduced pressure. The
final product (3.0 g, 93% yield) was used without further
characterization.
Syn th esis of th e An th r a cen e-Mod ified P EG Ma cr om er
(P EG-AN). The synthesis was achieved by reacting PEG with
9-anthracenecarbonyl chloride. In a typical experiment, 1.4 g
(0.56 mmol OH’s) of an eight-branched PEG was dissolved in
10 mL of anhydrous THF. 9-Anthracenecarbonyl chloride (1.35
g, 5.6 mmol) was dissolved in 10 mL of THF and then added
to the PEG solution. The mixture reacted for 20 h at 60 °C.
N2 atmosphere was used at all times. After the reaction was
over, the solution was concentrated by evaporation, and 20
mL of water was added to the residue. The mixture was
filtered and lyophilized in a vacuum. The dried solid was
washed with ether two times and dried in a vacuum. The final
yield of the PEG-AN was 0.43 g (28%). 1H NMR showed that
the hydroxyl groups were converted to ester to the extent of
87.4%. 1H NMR (300 MHz, CDCl3, TMS): δ ) 8.53 (s, 7H),
8.12 (d, 14H), 8.02 (d, 14H), 7.51 (m, 28H), 4.78 (d, 14H) and
3.96-3.39 (m, 1802H). FT-IR (NaCl, film): 772.2, 945.9,
1061.1, 1113.7, 1147.8, 1219.2, 1241.9, 1280.5, 1344.9, 1359.7,
1466.8, 1721.4 (CdO), 2883.2 cm-1. UV-vis (in water):
ꢀ365 nm ) 3.61 × 104 L‚mol-1‚cm-1. ꢀ254 nm ) 6.83 × 105
Resu lts a n d Discu ssion
Syn th esis of P oly(eth ylen e glycol)-An th r a cen e
(P EG-AN). The synthesis of the photosensitive PEG
macromer (PEG-AN) is diagrammatically represented
in Scheme 1. Commonly, esterification reactions are
assisted by HCl scavengers such as tertiary amines.
However, the esterification of 9-anthracenecarboxylic
acid is carried out at efficient yields only in the absence
of bases.16 In the presence of triethylamine or pyridine,
only 20% hydroxyl groups in the PEG was acylated. In
contrast, in the absence of amines, the degree of
modification was greatly improved, i.e., by 87%. In
addition, coupling reagents such as diisopropylcarbodi-
imide, 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide,
bis(2-oxo-3-oxazolidinyl)-phosphinic chloride, etc, were
also used to catalyze the reaction between PEG and
L‚mol-1‚cm-1
.
Meth od s a n d In str u m en ta tion . A concentrated 10% w/v
aqueous solution of PEG-AN macromer was used for the
photochemical studies. Unless indicated otherwise, the sample
was prepared by spreading drops of the aqueous solution to
form a film of 0.1 mm thickness between two quartz slides
(detachable quartz cells, Hellma Worldwide, Inc., USA, Plain-
view, NY).
Photo-cross-linking was performed using a 4-W 365-nm UV
lamp and photocleavage was performed using a 4-W 254-nm
UV lamp (Spectronics Corp., Westbury, NY). In either case,
the lamp was placed above the sample at a distance of 3 cm.

5230 Zheng et al.
Macromolecules, Vol. 35, No. 13, 2002
Sch em e 1. Syn th esis of P EG-An th r a cen e Ma cr om er (P EG-AN)
Sch em e 2. P h oto-Cr oss-Lin k in g/P h otoscission of P EG-AN Ma cr om er s
9-anthracenecarboxylic acid, but none of these reactions
were successful. The reason is probably due to the steric
hindrance of two adjacent R-protons and the poor
solubility of the reactive intermediates in the reaction
system.
fast photogelation are summarized as follows: (i) High
modification of PEG with anthracene groups provides
sufficient photosensitive groups for cross-linking. In our
study, we found that 20% modified PEG of the same
type did not induce gelation even after hours of irradia-
tion. (ii) The flexibility of the PEG polymer chains allows
for optimal aligning of the anthracene groups to undergo
efficient photodimerization reactions.17 (iii) The hydro-
phobic nature of the anthracene groups of the PEG-
AN macromers in an aqueous solution favors their
aggregation in the polymer matrix, which is a prereq-
uisite for [4 + 4] dimerization. Similar hydrophobic
effect was also observed in previous studies.18
The degree of swelling (DS) of the PEG-AN gel was
significantly changed following exposure to alternating
wavelengths of irradiation (Figure 1). After cross-linking
with 365-nm irradiation for 5 min, DS was determined
to be 10.5 ( 2.1. However, if 254-nm irradiation (5 min)
was subsequently applied, the DS increased to 17.0 (
Gela tion by P h oto-Cr oss-Lin k in g. The synthe-
sized PEG-AN macromer underwent photo-cross-link-
ing to form a three-dimensional hydrogel network due
to the [4 + 4] dimerization of anthracene groups
(Scheme 2). A solution of PEG-AN macromer (10% w/v
in water) was cast within a thin quartz cell (0.1 mm
thick). The macromer solution underwent rapid photo-
cross-linking upon irradiation at 365 nm for 1 min and
formed a transparent water-insoluble gel sheet. The
gelation speed was comparable to that of the poly-
(ethylene glycol)-nitrocinnamate (PEG-NC) macromer
that our group has recently demonstrated11 and is much
faster than those for previously reported hydrophilic
photosensitive systems.1d,15 The main reasons for such

Macromolecules, Vol. 35, No. 13, 2002
PEG-Based Hydrogel Synthesis 5231
F igu r e 1. Degrees of swelling of PEG-AN hydrogel at
different irradiation stages: (1) 365 nm (15 min), (2) 365 nm
(15 min) + 254 nm (5 min), (3) 365 nm (15 min) + 254 nm (5
min) + 365 nm (15 min), and (4) 365 nm (15 min) + 254 nm
(5 min) + 365 nm (15 min) + 254 nm (5 min). The degree of
swelling was calculated from the equation (Ws - Wd)/Wd, where
Ws is the weight of the swollen gel following 1 h of equilibration
in H2O and Wd is the weight of the dry gel (air-dried for 24 h
at room temperature).
F igu r e 3. UV-vis spectra of a PEG-AN solid film as a
function of irradiation conditions. The polymer film was first
exposed to 365-nm irradiation (0, 1, 3, 5, 10, 15, 25, 35, 45,
55, 80 min (a)) followed by 254-nm irradiation (1, 3, 6, 12, 20,
23, 30 min (b)).
to its dimers. The PEG-AN gel was then exposed to
254-nm light. Very quickly the absorption increased due
to the regeneration of anthracene groups. At 3 min of
irradiation, the spectral intensity reached its maximum.
Further irradiation with 254 nm changed the spectra
at a negligible level. This indicated that a photostation-
ary state was reached in a short period of time.
The effects of irradiation on the UV-vis absorption
spectra of PEG-AN were also determined in the form
of solid thin films (Figure 3). The results showed a big
difference from the aqueous solution. Upon irradiation
at 365 nm, the rate of decrease of the absorption
intensity is much slower than that in the solution phase.
For the aqueous PEG-AN solution (10% w/v), after 12
min of 365-nm irradiation the absorption reached its
minimum. However, for the solid sample, even after 15
min of irradiation (365 nm), the absorption maximum
at 366 nm is still as high as 40% of the original value.
The photo-cross-linking reaction is not completed even
after 80 min of irradiation. As aforementioned, PEG
exhibits its full flexibility in aqueous solution that
allows the optimal alignment of the anthracene groups
for photodimerization. On the other hand, for the solid
sample of PEG-AN, the solvent-induced advantage does
not exist, and therefore, it takes a longer time of
irradiation to achieve complete gelation. The photoscis-
sion process of the solid PEG-AN film induced by 254-
nm irradiation also demonstrates a slower kinetic than
the dissociation process of PEG-AN in aqueous solution
possibly due to the stability of the 4,4′-photocycloadduct
under different media (Figure 3b). Without the polar
aqueous environment, the 4,4′- photocycloadduct in the
solid state becomes more stable to photoscission.
F igu r e 2. UV-vis spectra of an aqueous PEG-AN solution
as a function of irradiation conditions. The initial concentration
of PEG-AN macromers is 10% w/v in H2O. Cuvette path )
0.1 mm. The PEG-AN solution was first exposed to 365-nm
irradiation (0, 0.5, 1, 4, 6, 8, 10, 12 min (a)) followed by 254-
nm irradiation (1, 1.5, 2, 3, 18 min (b)).
2.2. Continuous irradiation at 365 and/or 254 nm
changed the DS in a similar way. These results indicate
that different irradiation sequences have an enormous
impact on the backbone structure of the formed hydro-
gel. The DS is strongly related to the mesh size of the
polymer network; as the mesh size increases, the DS of
the PEG-AN gel also increases. Here, 365-nm light led
to the formation of a tightly interconnected three-
dimensional gel network with a small mesh size, while
254-nm UV irradiation significantly photocleaved the
gel network, thus increasing its mesh size and degree
of swelling.
UV-Vis Absor p tion . The photochemical process of
PEG-AN can be conveniently monitored by its elec-
tronic absorption spectroscopy. Figure 2 shows the UV-
vis absorption of a 10% w/v aqueous solution of PEG-
AN (optical path length ) 0.1 mm). Upon irradiation
with 365-nm light, the absorption maximum at 366 nm
decreased with the irradiation time, indicating the
undergoing [4 + 4] dimerization of anthracene groups
in the polymer. After 12 min of irradiation, there was
no further decrease in the spectral intensity that
suggested the complete conversion of anthracene group
NMR Stu d ies. 1H NMR spectroscopy was used to
delineate the effect of UV wavelength on the gel
structure. There are two characteristic regions in the
1H NMR spectrum of PEG-AN, one representing the
PEG backbone chain protons (3-5 ppm) and the other
corresponding to the anthracene moiety (7-9 ppm)
(Figure 4a). The 365-nm irradiation has no obvious
effect on the PEG backbone proton peaks (spectral
region not shown). On the other hand, however, 365-
nm exposure causes the peaks of the anthracene protons
to decrease in intensity and their chemical shifts to
move to a lower field (Figure 4b). These changes are
attributed to the disappearance of the anthracene
groups and the formation of [4 + 4] dimers. The

5232 Zheng et al.
Macromolecules, Vol. 35, No. 13, 2002
F igu r e 4. Changes in 1H NMR spectra of PEG-AN (in D2O)
at different irradiation conditions: (a) no irradiation; (b) 365
nm (30 min). Lamp power: 365 nm, 100 W; 254 nm, 50 W.
appearance of new proton peaks at δ ) 6.8 and 5.6,
further validates the formation of 4,4′-cycloadduct caused
by photo-cross-linking. Further exposure of the PEG-
AN gel to 254-nm UV does not change the basic shape
of Figure 4b but causes the area under the anthracene
peaks to increase with respect to the area of the 4,4′-
cycloadduct, illustrating the partial photoscission of the
4,4′-cycloadducts to monomers.
En vir on m en ta l Sca n n in g Electr on Micr oscop y
(ESEM). Microscopic techniques were utilized to pro-
vide a direct method of monitoring the changes on the
topography of the PEG-AN surface caused by photo-
cross-linking and photocleavage. Previously, we have
shown that ESEM and field emission SEM could be used
to image fragile structures, such as Langmuir-Blodgett
films, polymeric self-assembled films, protein coatings,
and hydrogels.11,19 Parts a-c of Figure 5 present typical
topographies of the hydrogel at three different irradia-
tion stages.
PEG-AN gel (prepared by 10 min of 365-nm irradia-
tion, Figure 5a) shows an essentially smooth surface,
with randomly distributed pores of less than 1 µm in
size. These pores are probably caused by a process of
heterogeneous congregation and redistribution during
light irradiation (see the AFM images in next section).
After the gel sample was further irradiated with 254-
nm UV light (5 min), the topography of the gel surface
was significantly altered from smooth to coarse (Figure
5b). The smooth surface was cleaved into many divided
domains with size of 10 µm. These micrographic changes
strongly demonstrated the photoscissive behavior of
254-nm irradiation. Figure 5c depicts the topography
of a PEG-AN hydrogel being treated with alternating
wavelengths of irradiation (365 nm + 254 nm + 365
nm, 10 min of 365 nm, and 5 min of 254 nm, respec-
tively). The gel surface was smooth in nature and only
a minimum amount of coarseness was detected. It
highly resembles the topography of the PEG-AN gel,
which was prepared under only 365-nm irradiation.
Therefore, the final dose of 365 nm exposure restored
the topography of the photocleaved hydrogel. These
F igu r e 5. ESEM micrographs of PEG-AN hydrogel under
different irradiation conditions: (a) 365 nm (10 min), (b) 365
nm (10 min) + 254 nm (5 min), and (c) 365 nm (10 min) + 254
nm (5 min) + 365 nm (10 min). ESEM conditions: wet mode,
0.9 Torr of H2O vapor, and gaseous secondary electron (GSE)
image.
ESEM images clearly demonstrate that the wavelength
of irradiation (365/254 nm) can significantly alter the
physical properties of the PEG-AN gel in a periodic
manner.

Macromolecules, Vol. 35, No. 13, 2002
PEG-Based Hydrogel Synthesis 5233
F igu r e 6. AFM images of PEG-AN hydrogel. (a) Scan size: 3 × 3 µm. Scan rate: 2.8 Hz. (b) Scan size: 190 × 190 nm. Scan
rate: 3 Hz. 365 nm (10 min). (c) Scan size: 3 × 3 µm. Scan rate: 2.8 Hz. (d) Scan size: 190 × 190 nm. scan rate: 3 Hz. 365 nm
(10 min) + 254 nm (5 min).
Atom ic F or ce Micr oscop y (AF M). AFM was used
to investigate surface structure of the PEG-AN hydro-
gel at a nanometric scale. Parts a and b of Figure 6
represent images of the surface of a PEG-AN gel
prepared under 365-nm irradiation (10 min) at different
scan sizes. Both images portray a heterogeneous orien-
tation of microscopic structures aligned in two do-
mains: flat layers and vertical edges. Polymer molecules
aggregate into many stretching, bow-shaped islets
separated by gaps with an average size of 15 nm.
Exposure of the PEG-AN gel to 254-nm light (5 min)
caused a clear change on the topography of the gel
surface as is shown in Figure 6, parts c and d. The
structure of the gel surface now appears cracked and
rough, the polymeric aggregates become smaller and
more irregular, and the gaps between the irregular
islets become wider (20 nm). Both photocleaved images,
compared to the photo-cross-linked ones (Figure 6a,b),
strongly indicate that 254-nm irradiation appreciably
breaks the microscopic structure of the PEG-AN gel
surface.
P EG-AN Hyd r ogel Rever sibility. The reversibility
of the hydrogel was illustrated by monitoring the UV-
vis absorption spectrum of PEG-AN gels as a function
of alternating wavelengths of irradiation (Figure 7).
Following an initial absorbance decrease during the first
photoreversible cycle (365/254 nm), photodimerization
and photoscisision proceed for more than 10 cycles
without additional losses in the photoreversibility ef-
ficiency, indicating that wavelength of irradiation can
be used as a trigger to predictably alter the physico-
chemical properties of these gels. We have shown that
254-nm irradiation reverted the process only at a degree
of 20%, instead of a complete sol-gel-sol process.

5234 Zheng et al.
Macromolecules, Vol. 35, No. 13, 2002
Serres, A.; Hatchell, D. L.; Grinstaff, M. W. J . Macromol.
Sci.sPure Appl. Chem. 1999, A36, 981. (c) Giammona, G.;
Pitarresi, G.; Cavallaro, G.; Buscemi, S.; Saiano, F. Biochim.
Biophys. Acta 1999, 1428, 29. (d) Andreopoulos, F. M.;
Christopher, R. D.; Stauffer, M. T.; Weber, S. G.; Wagner,
W. R.; Beckman, E. J .; Russell, A. J . J . Am. Chem. Soc. 1996,
118, 6235.
(2) Zhang, Y.; Won, C.; Chu, C. J . Polym. Sci., Part A: Polym.
Chem. 2000, 38, 2392.
(3) (a) Subramanian, K.; Krishnasamy, V.; Nanjundan, S.; Reddy,
A. V. R. Eur. Polym. J . 2000, 36, 2343. (b) Kim, S. H.; Won,
C. Y.; Chu, C. C. Carbohyd. Polym. 1999, 40, 183. (c) Kato,
M.; Ohara, H.; Fukuda, T.; Matsuda, H.; Nakanishi, H.
Macromol. Chem. Phys. 1998, 199, 881. (d) ACS Symposium
Series 673; Scranton, A. B., Bowman, C. N., Peiffer, R. W.,
Eds.; American Chemical Society: Washington, DC, 1997.
(4) (a) Laschewsky, A.; Rekai, E. Macromol. Rapid Commun.
2000, 21, 937. (b) Ali, A. H.; Srinivasan, K. S. V. J . Appl.
Polym. Sci. 1998, 67, 441. (c) Huang, J .; Zhu, D.; Lu, G.;
Wang, Z. Acta Chim. Sin. 1995, 53, 1018. (d) Hashidate, S.;
Nagasaki, Y.; Kato, M.; Matsuda, H.; Nakanishi, H. Polym.
Adv. Technol. 1995, 6, 626. (e) Nakayama, Y.; Matsuda, T.
J . Polym. Sci. A.; Polym. Chem. 1993, 31, 977.
(5) (a) Chen, Y.; Chen, K.-H. J . Polym. Sci., Part A: Polym.
Chem. 1997, 35, 613. (b) Chen, Y.; Chou, C.-F. J . Polym. Sci.,
Part A: Polym. Chem. 1995, 33, 2705. (c) Chen, Y.; Wu, J .-
D. J . Polym. Sci., Part A: Polym. Chem. 1994, 32, 1867. (d)
Chujo, Y.; Sada, K.; Saeguss, T. Macromolecules 1990, 23,
2693.
F igu r e 7. Number of reversible cycles of PEG-AN hydrogel
monitored by UV-vis spectroscopy at 366 nm. An aqueous
solution of PEG-AN (10% w/v) was exposed to alternating
wavelengths of irradiation (365 nm for 10 min; 254 nm for 5
min.
(6) (a) Kuckling, D.; Adler, H. J .; Arndt, K. F.; Wolff, T.;
Hoffmann, J .; Fischer, W. J . Macromol. Symp. 1999, 142, 111.
(b) Schinner, R.; Wolff, T.; Kuckling, D. Ber. Bunsen-Ges.
Phys. Chem. Chem. Phys. 1998, 102, 1710. (c) Matsuda, T.;
Moghddam, M. J . Trans. Am. Soc. Artif. Intern. Organs 1991,
37, M437. (d) Ichimura, K. J . Polym. Sci., Part A: Polym.
Chem. 1984, 22, 2817.
(7) (a) Imai, Y.; Ogoshi, T.; Naka, K.; Chujo, Y. Polym. Bull.
(Berlin) 2000, 45, 9. (b) Inaki, Y. J . Synth. Org. Chem. J pn.
1993, 51, 188.
(8) (a) Lippert, T.; Wei, J .; Wokaun, A.; Hoogen, N.; Nuyken, O.
Macromol. Mater. Eng. 2000, 283, 2000. (b) Hoogen, N.;
Nuyken, O. J . Polym. Sci., Part A: Polym. Chem. 2000, 38,
1903.
(9) Andreopoulos, F. M.; Beckman, E. J .; Russell, A. J . Bioma-
terials 1998, 19, 1343.
Interestingly, studies by Coursan and Desvergne14d
demonstrated that anthracene-modified polystyrene
polymers underwent reversible transformations with
minimum loss in photoreversibility efficiency. On their
study, ω-hydroxy-telechelic polystyrene was derivatized
with an anthracene moiety lacking a carbonyl function-
ality, namely 9-chloromethylanthracene. It was sug-
gested that the presence of the carbonyl functionality
may interfere with the photoreversible process. In
subsequent experiments, we also synthesized highly
modified PEG with anthracene groups through an ether
linkage. However, the photoreversible efficiency did not
change in comparison to the PEG-AN macromers.
From a practical standpoint, the partial reversibility is
still a valuable attribute; i.e., light can be used as a
trigger to alter the physicochemical properties in a
continuous and reversible manner, while not changing
the gel nature of the system.
(10) Andreopoulos, F. M.; Beckman, E. J .; Russell, A. J . J . Polym.
Sci., Part A: Polym. Chem. 2000, 38, 1466.
(11) Zheng, Y.; Andreopoulos, F. M.; Micic, M.; Huo, Q.; Pham, S.
M.; Leblanc, R. M. Adv. Funct. Mater. 2001, 11, 37.
(12) See two reviews and references therein: (a) Bouas-Laurent,
H.; Castellan, A.; Desvergne, J .-P.; Lapouyade, R. Chem. Soc.
Rev. 2000, 29, 43. (b) Bouas-Laurent, H.; Desvergne, J .-P.
In Photochromism Molecules and Systems; Du¨rr, H., Bouas-
Laurent, H., Eds.; Elsevier, New York, 1990; Chapter 14, p
561.
Con clu sion
(13) See two recent examples: (a) Cao, D.; Meier, H. Angew.
Chem., Int. Ed. 2001, 40, 186. (b) Mcskimming, G.; Tucker,
J . H. R.; Bouas-Laurent, H.; Desvergne, J .-P. Angew. Chem.,
Int. Ed. 2000, 39, 2167.
(14) For typical examples, see: (a) Ishii, T.; Tezuka, Y.; Kawamoto,
S.; Uno, T. J . Photochem. Photobiol. AsChem. 1994, 83, 55.
(b) Torii, T.; Ushiki, H.; Horie, K. Polym. J . 1993, 25, 173.
(c) Torii, T.; Ushiki, H.; Horie, K. Polym. J . 1992, 24, 1057.
(d) Coursan, M.; Desvergne, J . P. Macromol. Chem. Phys.
1996, 197, 1599.
(15) Chujo, Y.; Sada, K.; Nomura, R.; Naka, A.; Saegusa, T.
Macromolecules 1993, 26, 5611.
(16) Bayliss, M. A.; Homer, R. G.; Shepherd, M. J . Chem. Com-
mun. 1990, 305.
(17) Tazuke, S.; Banba, F. J . Polym. Sci., Part A: Polym. Chem.
1976, 14, 2463.
(18) Nakayama, Y.; Matsuda, T. J . Polym. Sci., Part A: Polym.
Chem. 1992, 30, 2451.
(19) (a) Sui, G.; Micic, M.; Huo, Q.; Leblanc, R. M. Colloids Surf.
A 2000, 171, 15. (b) Sui, G.; Micic, M.; Huo, Q.; Leblanc, R.
M. Langmuir 2000, 16, 7847. (c) Micic, M.; J eremic, M.;
Radotic, K.; Mavers, M.; Leblanc, R. M. Scanning 2000, 22,
288. (d) Micic, M.; J eremic, M.; Radotic, K.; Leblanc, R. M.
J . Colloid Interface Sci. 2000, 231, 190. (e) Micic, M.; Chen,
A.; Leblanc, R. M.; Moy, V. T. Scanning 1999, 21, 394.
In this study, we have synthesized a highly modified
PEG-AN macromer, which undergoes rapid photogel-
ation in the absence of initiators or catalysts and
demonstrates a photoreversible behavior. Physicochem-
ical properties such as swellability, UV-vis absorption,
and topography were all controlled by alternating the
irradiation wavelength (365 and 254 nm). The reversible
behavior of the PEG-AN gel provides a unique method
of fine-tuning the gel’s properties in a stimuli-responsive
manner. We envision potential applications of such
photoresponsive gels in photorecording as light switch
materials and in biomedical engineering as controlled
delivery vehicles.
Ack n ow led gm en t. This research is supported by
the National Science Foundation (CHE-0091390), U.S.
Army Research Office (DAAD19-00-1-0138), and the
Stanley Glaser Foundation.
Refer en ces a n d Notes
(1) (a) J o, S.; Shin, H.; Shung, A. K.; Fisher, J . P.; Mikos, A. G.
MA012263Z
Macromolecules 2001, 34, 2839. (b) Smeds, K. A.; Pfister-