An optically reversible switching membrane surface

Article abstract of DOI:10.1002/anie.200600581

(Figure Presented) Lighting the way: Through the combination of a UV-grafting process with the photoresponsive properties of spiropyran molecules, an optically reversible switching membrane surface was developed. This process can be used in preference to

Full text of DOI:10.1002/anie.200600581

Communications
DOI: 10.1002/anie.200600581
Membranes
An Optically Reversible Switching Membrane Surface**
Arpan Nayak, Hongwei Liu, and Georges Belfort*
Angewandte
Chemie
4
094
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4094 –4098

Angewandte
Chemie
T
he surface chemistry of synthetic membranes is of great
interest to the bioprocessing of protein-containing solutions,
to the filtration of medically relevant solutions, and to the
[
1]
performance of membrane-based sensors. For biosensors,
controlling the polarity of the membrane and reducing
nonspecific protein adsorption would significantly improve
signal to noise ratios. This may also be useful as a switch to
control diffusive transport to biosensor elements (enzymes,
chemoreceptors, etc)by changing the polarity of the
protective membrane when desired. For bioprocessing, it
has direct bearing on the efficacy of producing and delivering
desirable proteins and other biologically derived molecules
[
2]
at high yield and purity.
The performance of synthetic polymer membranes when
filtering bioprocessing and biomedical fluids is severely
limited by the interaction of proteins with membrane surfaces.
Currently, expensive and labor-intensive methods such as
controlling the pH and ionic strength of the feed solution,
periodic cleaning, and back flushing are used to deal with the
loss of filtration performance resulting from protein adhesion.
It would therefore be extremely attractive to have an in situ
cleaning procedure in which filtration performance could be
re-established with minimum need for expensive cleaning
chemicals and down time owing to back flushing. Synthesizing
membranes with polymers that reduce such interactions has
[
3,4]
resulted in improved filtration performance.
We suggest herein that grafting environmentally respon-
sive polymers onto synthetic membranes by using simple
inexpensive methods such as photograft-induced polymeri-
zation, which can be easily incorporated into the membrane
synthesis process, may offer an alternative method to the
approaches that use traditional protein-adhesion-resistant
monomers such as polyethylene glycol and other hydrophilic
Figure 1. Schematic of graft polymerization and switching: A) Rensse-
laer’s patented photografting process in which a synthetic ultrafiltra-
tion membrane formed from PES (photo sensitive polymer) is first
dipped into a vinyl (spiropyran) monomer solution (1% w/v) in ethyl
acetate for 1 h and then exposed to 300-nm UV irradiation to induce
grating and polymerization. B) Exposure of the grafted PES membrane
to visible light for 1 h resulting in the formation of the “closed” apolar
white form of spiropyran. C) Exposure of the grafted “closed” form of
the PES membrane to UV irradiation at 254 nm for 1 h formed the
“open” polar red form of spiropyran. D) The chemical structure of the
vinyl spiropyrans in two configurations as a function of UV and Vis
irradiation. E) A schematic of the chemical structure of the vinyl
spiropyrans in two configurations as a function of UV and visible (Vis)
irradiation.
[
3,5]
monomers.
Specifically, we combine two known technol-
ogies, a patented UV-grafting process with the well-known
photoresponsive properties of spiropyran molecules, to pro-
duce an optically reversible switching membrane surface.
Vinyl monomers with desirable functionality are easily
grafted and polymerized onto UV-sensitive polymers such as
poly(ether sulfone)(PES)membranes by using a UV-induced
[
6]
graft polymerization method (Figure 1A). Exposure of a
PES synthetic membrane to 300-nm UV radiation produces
radical sites onto which vinyl monomers such as vinyl
spiropyran can graft and polymerize. The spiropyrans are
comprised of a group of light-switchable photochromic
organic molecules that are colorless, nonpolar, and in a
[
7,8]
“closed” form in the visible light. When exposed to UVand
visible light, spiropyrans isomerize and produce a colored
polar “open“ merocyanine form and a white nonpolar
”closed“ form, respectively (Figure 1B–E, Figure 2).
[
*] A. Nayak, H. Liu, Prof. G. Belfort
Howard P. Isermann Department of
Chemical and Biological Engineering
Rensselaer Polytechnic Institute
Troy, NewYork 12180 (USA)
[
9–11]
Light-switching of monolayers of azobenzene
and
[12]
spiropyran has been studied on solid substrates. Previously,
binding of photochromic moieties to solid substrates required
Fax: (+1)518-276-4030
E-mail: belfog@rpi.edu
[
13,14]
complicated and multi-step procedures,
whereas herein
we use a relatively simple two-step dip and UV-irradiation
process that could be easily incorporated into a commercial
[
**] We acknowledge the support of the US Department of Energy
(
Grant No. DE-FG02-90ER14114) and the National Science Foun-
[
15]
dation (Grant No. CTS-94-00610).
manufacturing process.
The challenge was to synthesize
optically active vinyl spiropyran (1’-(2-propylcarbamylmetha-
crylamide)ethyl)-3’,3’-dimethyl-6-nitrospiro[2H-1]benzo-
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 4094 –4098
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4095

Communications
increased. If the ratio of peak heights, R = (h1720/h1487)/(h1663
/
h1487), is defined as a measure of the ratio of the relative
chemical surface condition under ultraviolet radiation versus
that under visible-light exposure, we can follow this ratio with
changing irradiation exposure in Figure 4A. The reference
À1
peak at 1487 cm is a measure of the carbonÀcarbon
Figure 2. Color change of spiropyran-grafted PES membrane after
exposure to UV irradiation at 254 nm for 1 h and visible light for 5 min
in air.
aromatic-ring stretch and is used as an internal standard as
it is independent of the grafting process.
Similarly, measuring the sessile contact angle of the
grafted dry PES membrane with a water droplet after
alternating exposure to 254-nm UV (1 h)and visible light
(5 min)demonstrates changes in surface wettability (Fig-
ure 4B). About a 168 drop in contact angle was obtained
through this process. This is similar in magnitude to that
reported by Lahann and co-workers who used electrical
potential to induce a polarity change in a self-assembled polar
pyran-2,2’-indoline)and determine whether the photografting
process that uses UV light at 300 nm would allow the vinyl
spiropyran molecules to graft and retain their switchable
properties. A commercial PES 30-kDa synthetic membrane
filter was chosen as a model membrane surface because of its
high sensitivity to 300-nm UV radiation and because it is one
of the most widely used polymeric materials that is used to
produce commercial ultrafiltration membranes and support
[
17]
molecule on a gold surface. As protein–membrane inter-
actions can determine success or failure for many processes
including those mentioned above, we demonstrate, through
two related experiments, that the switchable hydrophilic (on
exposure to 254-nm UV light)and hydrophobic (with visible
light)surfaces 1)exhibit low and high protein adsorbabilities,
respectively, and 2)display high and low buffer-permeations
rates after protein adsorption, respectively. Results for these
two experiments are shown in Figure 4C,D. Bovine serum
albumin (BSA; 67 kDa, pI 4.7)was used as a model protein
for the adsorption experiments. Clearly, from the data shown
in Figure 4C, the as-received PES membrane exhibited the
highest adsorbed amount of BSA in 10 mm phosphate-
buffered saline (PBS)buffer solution at pH 7.4 and 22 Æ
18C followed by the grafted vinyl spiropyran in the “closed”
(visible light)and “open” (254-nm UV light)configuration.
The “closed” configuration of the vinyl spiropyran surface
adsorbed about 26% more protein than the “open” config-
uration of the surface. These surfaces, with adsorbed protein
on them, were then tested to see if this difference in protein
adsorption translated into higher permeation flux. The
modified membranes were then placed into a filtration test
cell and the permeation rate of freshly prepared PBS at
pH 7.4 and 22 Æ 18C measured (Figure 4D). As expected, the
[
16]
layers for reverse osmosis and nanofiltration membranes.
The synthesized vinyl spiropyran was photografted onto
the 30-kDa PES ultrafiltration membranes by using our
standard grafting method. Attenuated total reflection Fourier
transform infrared spectroscopy (ATR/FTIR)was used to
demonstrate that the vinyl monomers were grafted onto the
PES membranes. In Figure 3, ATR/FTIR spectra are com-
pared for the unmodified PES membrane with those of the
grafted-spiropyran-modified PES membranes. The peaks at
À1
1
410, 1487, 1579, and 1663 cm are owing to the -SO group,
2
the aromatic-ring stretch of C=C, the aromatic CÀH stretch,
and the aromatic C=O stretch, respectively. The last peak may
shift depending on the chemical and charge environment.
Under exposed visible light for 5 min and on a dry membrane,
À1
the peak at ꢀ 1663 cm is maximized whereas the peak at
À1
ꢀ
1720–5 cm
is reduced. Similarly, under 254-nm UV
radiation for 1 h on
1663 cm is reduced whereas that at ꢀ 1720–5 cm is
a
dry membrane, the peak at
À1
À1
ꢀ
Poly (ether sulfone )
SO2
O
n
“
closed” configuration of the vinyl spiropyran gave about a
1
487: Aromatic
17% lower permeation flux as compared with the “open”
configuration of the vinyl spiropyran surface. The contact-
angle measurements of the base PES membrane, of the base
PES membrane after washing with ethyl acetate solvent, and
of the base PES membrane irradiated with 254-nm UV light
for 1 h were also obtained (see the Supporting Information).
No noticeable change in contact angle was observed in this
control experiment. However, irradiation with UV light
under wet rather than dry conditions reduces the switching
time from “closed” to “open” conformation by about one
third (data not shown).
1
579
ring stretch (C=C)
1
663
Unmodified
PES
A
~
1720-5 ~1663
Vis
UV
Vis
UV
1
800 1750 1700 1650 1600 1550 1500 1450 1400
Wavenumbers / cm-1
The results presented herein demonstrate the reversible
conversion of a photografted, photoresponsive, polymeric
synthetic membrane surface from polar to nonpolar charac-
ter. We have clearly shown that a UV/Vis-switchable moiety
such as spiropyran can be grafted onto a commercial photo-
active polymer such as PES by using a very simple photograft
polymerization method. Both molecular (ATR/FTIR)and
Figure 3. Five ATR/FTIR spectra of the unmodified (as received) PES
ultrafiltration membrane and the spiropyran-grafted PES membrane
after two cycles of exposure to UV irradiation at 254 nm for 1 h and
À1
visible light for 5 min. Note that the peak at ꢀ1663 cm decreases
À1
whereas the peak at ꢀ1720–1725 cm increases in intensity during
exposure to UV at 254 nm.
4
096 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4094 –4098

Angewandte
Chemie
A)
B)
of long-chain SAMs toward an electrode by
changing the surface potential of a gold sub-
2
1
1
0
0
.0
.5
.0
.5
.0
60
[
17,18]
5
5
5
0
strate,
we have synthesized and grafted a
photoresponsive molecule (vinyl spiropyran)onto
a widely used industrial polymer. This was accom-
plished by combining two technologies, our
patented UV grafting process with the well-
known photoresponsive properties of spiropyran
molecules, to produce an optically reversible
switching membrane surface. The relative ease of
this grafting and switching process should suggest
many industrial opportunities including those
dependent on surface wettability and molecular
adhesion.
Future work will involve further optimization
of the approach presented herein. Further research
with triggers other than UV/Vis radiation, such as
temperature, pH, and ionic strength, can be
pursued with vinyl functional groups and photo-
sensitive PES. Besides membrane filtration and
sensors, there is a need to consider surface proper-
45
4
3
3
0
5
0
0
1
2
3
4
5
0
1
2
3
4
5
Irradiation cycle number
Irradiation cycle number
C)
D)
0
0
0
0
.08
.06
.04
.02
0
5.0
4
3
2
1
0
.0
.0
.0
.0
.0
Modified,
Vis treated UV treated
Modified,
Unmodified Modified, Modified,
PES
Vis treated UV treated
Figure 4. Summary data demonstrating the optically reversible switching of the
wettability of the spiropyran grafted PES membrane surface after exposure to 254-nm
UV light and visible light. A) Ratio of peak heights, R=(h1720/h1487)/(h1663/h1487), from ties of materials for catheters, surgical instruments,
ATR/FTIR spectra from Figure 3 as a function of alternating exposure to 254-nm UV
light (1 h) and visible light (5 min) or cycle number. B) Surface wettability changes as
measured by the sessile contact angle of the grafted PES membrane with a water
droplet after alternating exposure to 254-nm UV light (1 h) and visible light (5 min)
or cycle number. C) Adsorption of BSA in PBS (10 mm; pH 7.4) at 22Æ18C for 1 h
on the as-received PES membrane and on modified PES membranes with grafted
vinyl spiropyran in the “closed” (visible light) and “open” (254-nm UV light)
configuration. D) Mass flux of PBS at pH 7.4 and 22Æ18C permeating through
modified PES membranes with grafted vinyl spiropyran in the “closed” (visible light)
and “open” (254-nm UV light) configuration after exposure to BSA adsorption from
and pulmonary breathing tubes that are exposed to
body fluids, and for spot detection for genomic
analysis and nanoprocessing in micro-fluidic devi-
ces.
Experimental Section
Vinyl spiropyran (1’-(2-(propylcarbamylmethacrylami-
de)ethyl)-3’,3’-dimethyl-6-nitrospiro[2H-1]benzo- pyran-
(
C).
2,2’-indoline): 3-aminopropyl methacrylamide hydro-
chloride was obtained from Polysciences Inc. All other
chemicals were purchased from Aldrich Chemical Co. and used
without further purification. 1-(2-carboxyethyl)-2,3,3-trimethylindo-
lenine iodide (1) , 1’ -(2-carboxyethyl) À3’,3’-dimethyl À6- nitro spiro-
classical (color, contact angle, protein adsorption, and sub-
sequent buffer ultrafiltration)measurements were used to
confirm the reversible polar–apolar reaction of grafted
spiropyran owing to exposure of UV/Vis radiation. Rather
than inducing a physical movement (conformational change)
[
2H-1]benzopyran-2,2’-indoline (2), and 1’-(2-(carbosuccinimidyl-
oxy)ethyl)-3’,3’-dimethyl-6-nitrospiro[2H-1]benzopyran-2,2’-indoline
[19]
(3)were synthesized according to literature methods
(Scheme 1).
Scheme 1. Synthesis of the spiropyran-containing monomer compound 4. DCC=dicyclohexyl carbodiimide.
Angew. Chem. Int. Ed. 2006, 45, 4094 –4098 ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4097

Communications
NMR spectra were measured on a Varian spectrometer by using
tetramethylsilane (TMS)as the internal standard. Mass spectra were
recorded by using electrospray ionization.
colored solutions was measured at 515 nm. Protein amounts were
calculated based on a calibration which was performed with known
BSA amounts (mBSA = 0.001–1 mg)onto clean unmodified PES
2
1’-(2-(Propylcarbamylmethacrylamide)ethyl)-3’,3’-dimethyl-6-
membranes (A515 nm = 0.6799mBSA + 0.0754; R = 0.9944).
nitrospiro[2H-1]benzo- pyran-2,2’-indoline (4): The mixture of 3
PBS filtration: PBS buffer solution (10 mm; pH 7.4)was used as
the feed. The PBS buffer solution was composed of NaCl (137 mm)
and KCl (2.7 mm) in deionized water. Membranes were immersed in
(
(
3.26 g, 6.83 mmol), 3-aminopropyl methacrylamide hydrochloride
1.34 g, 7.50 mmol), and triethyl amine (1 mL, 7.19 mmol) was stirred
À1
in N,N-dimethylformamide (DMF; 50 mL)at room temperature for
0 h. After the evaporation of DMF, the residue was dissolved in
the BSA solution (1 mgmL in PBS buffer)for 5 min to induce
2
protein adsorption. Then, PBS permeation flux through the mem-
branes with adsorbed BSA was measured. A dead-end stirred cell
(Model 8010, Millipore Corp., Bedford, MA)filtration system was
used for PBS flux measurements through the membranes. The active
chloroform, washed with water and then purified by silica-gel column
chromatography (eluent, ethyl acetate/hexane; 1:1 v/v)to give 4 as a
1
yellow solid (2.6 g, 76%). H NMR (500 MHz, CDCl ): d = 7.99 (m,
3
2
2
4
H; 5-H and 7-H), 7.18 (t, 1H; 6’-H), 7.08 (d, 1H; 4’-H), 6.89 (m, 2H;
-H and 5’-H), 6.75 (d, 1H; 8-H), 6.67 (d, 1H; 7’-H), 6.51 (m, 1H;
membrane area was 3.8 cm . All filtration experiments were con-
ducted at a constant transmembrane pressure of 69 kPa, a stirring rate
of 500 rpm, and a system temperature of 22 Æ 18C.
NH), 6.39 (m, 1H; NH), 5.87 (d, 1H; 3-H), 5.74 (s, 1H; CH ), 5.35 (s,
2
1
H; CH ), 3.68 (m, 1H; CH N), 3.51 (m, 1H; CH N), 3.20 (m, 4H;
2
2
2
CH ), 2.56 (m, 1H; CH CO), 2.44 (m, 1H; CH CO), 1.93 (s, 3H;
Received: February 13, 2006
Published online: May 16, 2006
2
2
2
CH ), 1.58 (m, 2H; CH ), 1.27 (s, 3H; CH ), 1.17 ppm (s, 3H; CH );
3
2
3
3
1
3
C NMR (300 MHz, CDCl ): d = 18.9, 20.1, 26.0, 29.8, 35.9, 36.3, 40.2,
3
5
1
3.1, 107.0, 115.7, 118.9, 120.0, 120.2, 122.0, 122.2, 123.0, 126.1, 128.0,
28.5, 136.1, 139.9, 141.2, 146.7, 159.7, 169.2, 171.9 ppm.
Keywords: membranes · optical properties · polymerization ·
surface chemistry · UV/Vis spectroscopy
.
+
+
MS calcd C H N O : 504.24. Found [M+H] : 505.2, [M+Na] :
2
8
32
4
5
5
27.1.
Preparation of modified membranes: 30-kDa PES membranes
were modified by using a UV-induced graft polymerization method as
[
1] ”Interactions of proteins with polymeric synthetic membranes”:
G. Belfort, A. L. Zydney, Biopolymers at Interfaces, Marcel
Dekker, New York, 2003.
[
3,6,15,16]
described in detail in Taniguchi, Belfort and co-workers.
Rayonet photochemical chamber reactor system (Model RPR-100,
Southern New England, Ultraviolet Co., Branford, CT)containing
A
[
2] L. J. Zeman, A. L. Zydney, Microfiltration and Ultrafiltration
Principles and Applications, Marcel Dekker, New York, 1996.
3] M. Taniguchi, G. Belfort, J. Membr. Sci. 2004, 231, 147 – 157.
4] M. J. Steuck, US Patent 4618533, 1986.
5] M. Ulbricht, H. Matuschewski, A. Oechel, H. G. Hicke, J.
Membr. Sci. 1996, 115, 31 – 47.
3
00-nm UV lamps ( ꢀ 15% of the energy was at < 280 nm)was used.
[
[
[
The membranes were dipped in spyropyran monomer solution (1%
w/v in ethyl acetate)for 1 h with stirring at 22 Æ 18C, removed from
the monomer solution, purged with N for 10 min, and irradiated with
2
3
00-nm UV light in water-saturated N for 4 min. After photografting,
2
[
[
6] G. Belfort, M. Taniguchi, J. Pieracci, US Patent 6852769, 2005.
7] ”Biological applications-Supramolecular chemistry“: M. Inouye,
Organi c, Photo ch romicand Thermo ch romicCompounds , Vol. 2,
Kluwer Academic, New York, 1999, chap. 9.
the membranes were again cleaned with ethyl acetate by shaking
overnight to remove homopolymer and unreacted monomer from the
membrane. Then, the modified membranes were vacuum dried for
use.
ATR/FTIR: ATR/IR (Magna-IR 550 Series II, Thermo Nicolet
Instruments Corp., Madison, WI)was used to obtain a measure of the
degree of grafting. By using an incident angle of 458, the penetration
or sampling depth was approximately 0.1–1.0 mm. Spectra were
collected at a gain of 8 and resolution of 2 cm with 512 scans for
each sample.
Synthetic membrane: The PES membrane with a 30-kDa
molecular-weight cut off from lot 9140E was obtained from Pall
Corp. (East Hills, NY). These Omega series have been slightly
[
8] H. Bouas-Laurent, H. Dürr, Pure Appl. Chem. 2001, 73, 639 –
665.
[
9] W. Jiang, G. Wang, Y. He, X. Wang, Y. An, Y. Song, L. Jiang,
Chem. Commun. 2005, 3550 – 3552.
[
[
[
[
[
[
10] H. Ge, G. Wang, Y. He, X. Wang, Y. Song, L. Jiang, D. Zhu,
ChemPhysChem 2006, 7, 575 – 578.
11] K. Ichimura, S.-K. Oh, M. Nakagawa, Science 2000, 288, 1624 –
À1
1626.
12] B. C. Bunker, B. I. Kim, J. E. Houston, R. Rosario, A. A. Garcia,
M. Hayes, D. Gust, S. T. Picraux, Nano Lett. 2003, 3, 1723 – 1727.
13] R. Rosario, D. Gust, M. Hayes, F. Jahnke, J. Springer, A. A.
Garcia, Langmuir 2002, 18, 8062 – 8069.
hydrophilized by the manufacturer by an undisclosed process as
À1
evidenced by the small carbonyl peak at ꢀ 1663 cm
.
Contact angle: The sessile contact angle of water in air on the
membrane substrates was measured by using an optical system (SIT
camera, SIT66, Dage-MTI, Michigan, IN)converted to a video
display. Water droplets of 2.5 mL were placed on the membrane
substrates at different positions and the contact angles were
measured. At least five measurements were made and the average
reported.
14] S. Abbott, J. Ralston, G. Reynolds, R. Hayes, Langmuir 1999, 15,
8923 – 8928.
15] J. Pieracci, D. W. Wood, J. V. Crivello, G. Belfort, Chem. Mater.
2000, 12, 2123 – 2133.
[16] B. Kaeslev, J. Pieracci, G. Belfort, J. Membr. Sci. 2001, 194, 245 –
261.
[17] J. Lahann, S. Mitragotri, T. -N. Tran, H. Kaido, J. Sundaram, I. S.
Choi, S. Hoffer, G. A. Somorjai, R. Langer, Science 2003, 299,
371 – 374.
[18] U. Rant, K. Arinaga, S. Fujita, N. Yokoyama, G. Abstreiter, M.
Tornow, Nano Lett. 2004, 4, 2441 – 2445.
[19] A. Fissi, O. Pieroni, G. Ruggeri, F. Ciardelli, Macromolecules
1995, 28, 302 – 309.
BSA adsorption: BSA was dissolved in PBS buffer solution
À1
(10 mm; pH 7.4)to prepare a 1 mgmL protein solution. Membrane
2
swatches (3 cm )were immersed in the BSA solution for 2 h at 22 Æ
8C. The amount of adsorbed protein was determined by staining
with Ponceau S solution (Ponceau S (2%), trichloroacetic acid
30%), and sulfosalicylic acid (30%)). Membranes with adsorbed
1
(
protein were immersed for 1 h into a solution of Ponceau S, washed
thoroughly with deionized water, immersed for 1 h in acetic acid (5%
v/v), and again washed with deionized water. Then the protein–dye
complex was quantitatively eluted with NaOH solution (3 mL,
1
00 mm) for 1 h. Next, the membranes were removed, the solutions
neutralized with HCl (50 mL, 6m), and the absorbance of the red-
4
098 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4094 –4098