Use of an aromatic polyimide as a non-cross-linked molecular imprinting material

Article abstract of DOI:10.1021/ma035199p

The use of a rigid non-cross-linked polymer in molecular imprinting was explored. Aromatic polyimides have excellent thermal stability and mechanical strength, which have contributed to their successful application in several areas, such as films, molding

Full text of DOI:10.1021/ma035199p

6
Macromolecules 2004, 37, 6-8
expected that an aromatic polyimide would produce a
satisfactory non-cross-linked imprinted matrix.
Use of a n Ar om a tic P olyim id e a s a
Non -Cr oss-Lin k ed Molecu la r Im p r in tin g
Ma ter ia l
The template molecule, estrone, was attached to the
polyimide chain by means of a urethane bond. The
urethane bond was stable at room temperature but was
cleaved at elevated temperature, as shown in Scheme
1.11 The use of a thermally reversible bond has certain
advantages in that it is possible to easily remove the
template molecule from the matrix and to introduce
various functional groups into the cavities.12 Estrone is
one of several naturally occurring estrogens. It influ-
ences the normal development and maturation of the
female. Estrone has been suspected of having carcino-
genic properties and adverse environmental effects.13-15
Ch a e Hyu n Lim , Ch a n g Do Ki, Ta e Hoon Kim , a n d
J i You n g Ch a n g*
School of Materials Science and Engineering, and
Hyperstructured Organic Materials Research Center,
College of Engineering, Seoul National University,
ENG445, Seoul 151-744, Korea
Received August 14, 2003
Revised Manuscript Received November 10, 2003
Sch em e 1
In tr od u ction . Mechanically stable materials that
can selectively recognize target molecules have potential
uses in applications such as chemical sensors, stationary
phases for high-performance chromatography, catalysts,
and membranes for separating toxic chemicals.1-6 Mo-
lecular imprinting constitutes a valuable method of
preparing such materials, in which a specific binding
site is introduced into an organic or inorganic polymer
matrix. In the process, a target molecule (template) is
first complexed with a functional monomer and then
frozen into a matrix by polymerization. The removal of
the template from the matrix generates a cavity, which
can recognize the template.
The recognition ability of the cavity is greatly influ-
enced by the structure of the matrix and the nature of
the bond used for the template-monomer complexation.
Most imprinted polymeric matrices reported so far are
vinyl polymers, having cross-linked network structures,
so as to prevent structural changes of the cavity occur-
ring by relaxation or diffusion of the matrix molecules.
Their microscopic structures depend primarily on the
nature of the functional monomer and cross-linking
agent as well as on the polymerization conditions. An
excess of cross-linking agent is generally used to in-
crease the rigidity of the imprinted matrix, but which
sometimes has an adverse effect on the interaction
between the template and the matrix, resulting in low
recognition capacity.7 There are only a few examples for
non-cross-linked molecularly imprinted polymers. Re-
cently, Kobayashi et al. reported a phase inversion
molecular imprinting method, in which linear polymers
were used as a matrix and template molecule informa-
tion was encoded by the polymer phase inversion
process.8,9
Mon om er Syn th esis.16 Diamine 3 having estrone
was synthesized according to Scheme 2. 3,5-Dinitroben-
zoyl chloride in methylene chloride was reacted with
sodium azide in water at 0 °C to yield azide 1.17
A
solution of 1 in toluene was refluxed in the presence of
estrone and dibutyltin dilaurate. The reaction occurred
between an isocyanato group formed by thermal rear-
rangement of an azide and a phenol moiety of estrone,
resulting in urethane 2.18 The nitro groups of 2 were
reduced to diamino groups by catalytic hydrogenation
in tetrahydrofuran.19 3,5-Diaminobenzoyl azide (4) was
synthesized according to the literature.20
Sch em e 2
In this study, we explored the use of an aromatic
polyimide as the imprinted matrix and a thermally
reversible bond between the template and the monomer
in the process of template-monomer complexation.
Aromatic polyimides are prepared by condensation
polymerization of aromatic dianhydrides and aromatic
diamines. Depending on the selection of a dianhydride-
diamine pair, a wide variety of physical properties can
be conferred on the polyimide matrix. Aromatic poly-
imides are known as well-packed materials due to the
strong interactions between the polymer chains, and
most of them are insoluble and infusible.10 Thus, we
P r ep a r a tion of P olyim id e. Estrone imprinted poly-
imide was prepared according to Scheme 3. Polymeri-
zation was carried out in N,N-dimethylacetamide (DMAc)
at room temperature using a stoichiometric amount of
the diamine (1:19 mole ratio of 3 to 4,4′-oxydianiline)
and pyromellitic dianhydride. The highly viscous solu-
tion was cast on a glass plate and dried in a vacuum
oven for 48 h at room temperature. The poly(amic acid)
film was thermally imidized for 0.5 h each at 80, 130,
170, 220, and 270 °C to yield a polyimide film (thick-
ness: ca. 20 µm). A control polyimide film was fabricated
in the same manner as the imprinted polyimide film.
* Corresponding author: Tel +82-2-880-7190; Fax +82-2-885-
1748; e-mail jichang@snu.ac.kr.
10.1021/ma035199p CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/13/2003

Macromolecules, Vol. 37, No. 1, 2004
Communications to the Editor
7
Sch em e 3
For the control polyimide, 3,5-diaminobenzoyl azide (4)
was used as a diamine in place of compound 3.
ated estrone, showing that the urethane bond had been
thermally cleaved.
The thermal behaviors of the polyimide films were
evaluated by thermogravimetric analysis (Figure 1).21
In the thermogram of the estrone-containing film, about
3% weight loss was observed between 220 and 320 °C,
corresponding to the decomposition of estrone. This
value coincides with the calculated weight percent of
estrone contained in the films (3.4%).
F igu r e 2. 1H NMR spectra of compound 3 (a) at 25 °C and
(b) at 110 °C.
It is believed that part of the template molecules were
dissociated during the process of imidization and re-
mained in the matrix. The template molecules in the
film were extracted by refluxing in 1,4-dioxane in the
presence of a small amount of water. In this process,
undissociated template molecules were removed as well.
The isocyanato groups formed by the dissociation were
converted to amino groups through their reaction with
water.12 The extraction process was continued until the
UV absorbance at 282 nm, corresponding to estrone in
the solution, became constant.
Rebin d in g Exp er im en t. Polyimide films (0.1 g)
were added to the solutions of estrone or its structural
analogues (testosterone, testosterone propionate, 98%,
purchased from TCI) in chloroform (20 mL) at various
concentrations (1, 2, and 3 mM). After incubating for
12 h at room temperature, the polyimide films were
isolated by filtration. The filtrate was concentrated to
dryness by evaporation of the solvent before HPLC
analysis.22
Figure 3 shows the result of the rebinding experiment
on the imprinted and control polyimide films. The
concentrations of estrone were 1, 2, and 3 mM in 20 mL
of chloroform. For the same estrone concentration, more
estrone was bound to the imprinted polymer than the
control. Figure 4 shows the results of selectivity tests
on the imprinted polymer and control polyimide. A 0.1
F igu r e 1. TGA thermograms of the estrone-containing and
control polyimide films.
Extr a ction of Estr on e. The thermal cleavage of the
1
urethane bond was investigated by H NMR spectros-
copy and FT-IR spectroscopy. Compound 3 was dis-
solved in DMSO-d6 and the 1H NMR spectra were
measured at 25 and 110 °C (Figure 2). The 1H NMR
spectrum obtained at room temperature showed the
N-H peak of the urethane bond at 9.5 ppm and
aromatic ring proton peaks of the estrone group at 7.3
and 6.9 ppm. After increasing the sample temperature
to 110 °C, the N-H peak decreased remarkably while
new peaks appeared at around 6.5 and 7.1 ppm, corre-
sponding to the aromatic ring protons from the dissoci-

8
Communications to the Editor
Macromolecules, Vol. 37, No. 1, 2004
control. Our results are quite encouraging as regards
the use of non-cross-linked high-performance polymers,
such as aromatic polyimides, in molecular imprinting.
Ack n ow led gm en t. This work was supported by the
National R&D Project for Nano Science and Technology.
Refer en ces a n d Notes
(1) Wulff, G. Chem. Rev. 2002, 102, 1.
(2) Haupt, K.; Mosbach, K. Chem. Rev. 2000, 100, 2495.
(3) Bystrom, S. E.; Borje, A.; Akermark, B. J . Am. Chem. Soc.
1993, 115, 2081.
(4) Leung, K. P.; Chow, C. F.; Lam, H. W. J . Mater. Chem. 2001,
11, 2985.
(5) Dai, S.; Burleigh, M. C.; Shin, Y. S.; Morrow, C. C.; Barnes,
C. E.; Xue, Z. Angew. Chem., Int. Ed. 1999, 38, 1235.
(6) Vlatakis, G.; Andersson, L. I.; Mosbach, K. Nature (London)
1993, 361, 645.
(7) Marty, J .-D.; Tizra, M.; Mauzac, M.; Rico-Lattes, I.; Lattes,
A. Macromolecules 1999, 32, 8674.
(8) Kobayashi, T.; Reddy, P. S.; Fujii, N. Eur. Polym. J . 2002,
38, 779.
(9) Kobayashi, T.; Fukaya, T.; Abe, M.; Fujii, N. Langmuir 2002,
18, 2866.
F igu r e 3. Amount of bound estrone by the imprinted and
control polyimide films. In all experiments, the amount was
analyzed by HPLC after 0.1 g of the polymers were added to
20 mL of the 1, 2, and 3 mM estrone solutions in chloroform
for 12 h at room temperature. All experiments were repeated
three times.
g of polyimide film was added to the 3 mM sample
solution in 20 mL of chloroform. The tests were con-
ducted using testosterone and testosterone propionate,
which are the structural analogues of estrone but do
not have a phenolic group. The imprinted film showed
higher specific recognition ability for estrone than for
its structural analogues. The results indicate that
template-shaped cavities were imprinted in the poly-
imide matrix. The amino group inside the cavity would
form a hydrogen bond with the phenol moiety of Estron
in the rebinding process.
(10) Ghosh, M. K.; Mittal. K. L. Polyimides; Marcel Dekker: New
York, 1996; pp 279-303.
(11) Chang, J . Y.; Kim, T. J .; Han, M. J .; Chae, K. H. Polymer
1999, 40, 4049.
(12) Ki, C. D.; Oh, C.; Oh, S. G.; Chang, J . Y. J . Am. Chem. Soc.
2002, 124, 14838.
(13) Tremblay, M. R.; Lin, S. X.; Poirier, D. Steroids 2001, 66,
821.
(14) Pasqualini, J . R.; Cortes-Prieto, J .; Chetrite, G.; Talbi, M.;
Ruiz, A. Int. J . Cancer 1997, 70, 639.
(15) Xiao, X.; McCally, D. Rapid Commun. Mass Spectrom. 2000,
12, 1991.
(16) Sodium azide (99%) and dibutyltin dilaurate (95%) were
purchased from Aldrich Chemical Co. Estrone (98%) and 3,5-
dinitrobenzoyl chloride (99%) were purchased from TCI.
Methylene chloride and tetrahydrofuran were used after
purification by standard methods. Other chemicals were
used as received without further purification.
(17) Compound 1: Anal. Calcd for C7H3N5O5: C, 35.46; H, 1.28;
N, 29.53. Found: C, 35.56; H, 1.29; N, 29.66. 1H NMR
(DMSO-d6): δ 8.89 (s, 2H benzene ring protons), δ 9.02 (s,
1H benzene ring proton). 13C NMR (CDCl3, 500 MHz):
123.0, 129.2, 134.3, 149.0, 168.9. IR (KBR pellet, cm-1):
3168 (aromatic C-H stretching), 2157 (N3), 1691 (CdO),
1548, 1358 (NdO stretching).
(18) Compound 2: Anal. Calcd for C25H25N3O7: C, 62.62; H, 5.26;
N, 8.76. Found: C, 62.80; H, 5.32; N, 8.45. 1H NMR (DMSO-
d6): 0.85 (s, 3H), 6.99 (s, 1H), 7.01 (d, 1H), 7.34 (d, 1H),
8.50 (s, 1H), 8.73 (s, 2H), 11.17 (s, 1H). 13C NMR (CDCl3,
500 MHz): 13.4, 21.0, 25.4, 26.0, 28.9, 31.2, 35.2, 37.8, 43.3,
47.2, 49.4, 112.7, 114.9, 117.7, 118.8, 121.5, 126.0, 137.3,
137.9, 141.1, 148.4, 219.6.
(19) Compound 3: Anal. Calcd for C25H29N3O3: C, 71.57; H, 6.97;
N, 10.02. Found: C, 71.62; H, 7.14; N, 9.63. 1H NMR
(DMSO-d6): 0.84 (s, 3H), 4.68 (s, 4H), 5.53 (s, 1H), 6.00 (s,
2H) 6.86 (s, 1H) 6.91 (d, 1H), 7.28 (d, 1H), 9.5 (s, 1H). 13C
NMR (CDCl3, 500 MHz): 14.0, 21.8, 26.0, 26.5, 29.6, 30.5,
31.7, 36.1, 38.3, 44.3, 50.7, 96.5, 97.7, 119.1, 121.9, 125.7,
126.6, 137.4, 138.2, 139.4, 148.4, 148.67, 221.0.
(20) Ambade, A. V.; Kulmar, A. J . Polym. Sci., Polym. Chem.
Ed. 2001, 39, 1295.
(21) The thermogravimetric analysis was performed with a TGA
2050 thermogravimetric analyzer (TA instruments, Inc.) at
a heating rate of 10 °C/min under a dried N2 atmosphere.
(22) Reverse phase HPLC analysis was carried out using a M930
solvent delivery system, a M720 UV-vis detector (YOUNG
LIN Instrument Co. Ltd., Korea), a MetaSil 5u ODS column
from Metachem (Torrance, Canada) with methanol as an
eluent at a rate of 1.0 mL/min at room temperature. For
each analysis 20 µL of sample was injected.
F igu r e 4. Amount of bound molecules by the imprinted (MIP)
and control polyimide films. In all experiments, the amount
was analyzed by HPLC after 0.1 g of the polymers were added
to 20 mL of the 3 mM sample solutions in chloroform for 12 h
at room temperature. All experiments were repeated three
times.
Con clu sion . We explored the use of a rigid non-cross-
linked polymer in molecular imprinting. Aromatic poly-
imides have excellent thermal stability and mechanical
strength, which have contributed to their successful
application in several areas, such as films, moldings,
coatings, adhesives, and resin matrices. The template
molecule was attached to the polyimide chain by means
of a thermally reversible urethane bond. The removal
of the template was accomplished by a simple thermal
reaction. Most imprinted materials obtain their rigidity
through a cross-linking reaction, which it is not easy to
MA035199P