Elucidating colorization in the functionalization of hydroxyl-containing polymers using unsaturated anhydrides/acyl chlorides in the presence of triethylamine

Full text of DOI:10.1021/bm901237t

3
04
Biomacromolecules 2010, 11, 304–307
Notes
Elucidating Colorization in the Functionalization
of Hydroxyl-Containing Polymers Using
Unsaturated Anhydrides/Acyl Chlorides in the
Presence of Triethylamine
position in the alcoholysis of R,ꢀ-unsaturated acyl chlorides such
as crotonoyl chloride with tertiary amines, such as TEA and
17-27
pyridine.
The reaction of TEA with maleic anhydride (MA)
was first reported by Mayahi and El-Bermani to yield a yellow
2
8
solution first and then a dark brown solid in a few minutes.
Such reaction between tertiary and primary amines with cyclic
Lei Cai and Shanfeng Wang*
2
9,30
anhydrides was further investigated in terms of kinetics.
Department of Materials Science and Engineering, The
University of Tennessee, Knoxville, Tennessee 37996
Though extensive, the above studies were not performed on the
esterification of hydroxyl-containing polymers, and therefore,
little attention was paid to how such complexation would affect
the properties and biological performance of the polymers
resulted.
Received October 30, 2009
Revised Manuscript Received November 26, 2009
Introduction
To demonstrate the formation of complex, we performed the
reactions between TEA and four unsaturated anhydrides or acyl
chlorides, namely, MA, itaconic anhydride (IA), acryloyl
chloride (AC), and FC. The chemical structures of the complexes
produced from the above reactions and reaction mechanisms
were confirmed using Fourier-transform infrared (FTIR) spec-
troscopy, nuclear magnetic resonance (NMR) spectroscopy,
UV-vis spectroscopy, photoluminescence spectroscopy, and
elemental analysis. Using human embryonic kidney (HEK293)
cells, we further examined the cytotoxicity of these complexes
in cell culture medium at different concentrations over one week.
By providing detailed evidence, this Note aims to raise the
concerns about using TEA as a common proton scavenger in
synthesizing cross-linkable polymers for biomedical applications.
It is a widely used method to develop cross-linkable or
injectable materials for biomedical applications by functional-
izing hydroxyl-containing polymers or oligomers with unsatur-
ated anhydrides or acyl chlorides in the presence of a proton
1-6
scavenger, triethylamine (TEA or Et
3
N).
Although numerous
cross-linkable polymers have been synthesized successfully
using this method, yellowish or dark brown color always exists
in the products and purification is practically rather difficult and
1-6
incomplete.
This problem has been intriguing for biomaterial
researchers and could not be ignored in the condensation
between poly(ε-caprolactone) (PCL) diol and fumaryl chloride
(
(
FC) using TEA because the dark color of the PCL fumarate
PCLF) synthesized was detrimental to the efficiency of
photocross-linking and cell/tissue staining as well as product
appearance.
5
Experimental Section
Previously, one of the authors and his colleagues have avoided
Materials. All chemicals were purchased from Aldrich (Milwaukee,
WI) unless otherwise noted. AC from Aldrich and FC, MA, and IA
from Acros Organics (Morris Plains, NJ) were used without further
purification. Methylene chloride and TEA (Fisher Scientific, Fair Lawn,
colorization by using potassium carbonate (K
2 3
CO ) as a
substitute for TEA in synthesizing PCLF, polyethylene glycol
7
2 3
fumarate (PEGF), and their amphiphilic copolymer. K CO was
also used even earlier to neutralize the condensation byproduct
HCl between FC and propylene glycol and prevent premature
loss of the fumarate double bond in synthesizing poly(propylene
2
NJ) were distilled over calcium hydride (CaH ).
Synthesis. Typically, 0.05 mol AC was dissolved in 100 mL of
methylene chloride and stirred at 0 °C under nitrogen flux. A pressure-
equalizing dropping funnel was used for the dropwise addition of a
solution of 0.05 mol TEA (1:1 molar ratio) in 50 mL of methylene
chloride into the AC solution in methylene chloride. The color turned
yellow immediately upon adding the TEA solution. The reaction
proceeded for 24 h. The solution was concentrated using a rotary
evaporator and precipitated in excessive chilled diethyl ether. The
precipitate was filtered and dried at room temperature in vacuum. Great
heat was generated by adding TEA directly into AC, FC, MA, or IA
in a 40 mL glass vial at room temperature. The temperature was
observed to increase 20-50 °C in 1 min and black or yellow smoke
filled the glass vial.
8
fumarate) in the Mikos group. Though this substitute for TEA
has been used to overcome the colorization problem success-
fully, the reason for colorization remains elusive and related
potential concerns are yet to known. In this Note, we feel
compelled to elucidate the origin and mechanism of such
colorization in esterification and to reveal the cytotoxicity of
the complexes formed between unsaturated anhydrides or acyl
chlorides and TEA, which are accused for causing colorization.
Furthermore, we propose principles for selecting proton scav-
engers that will not generate colorization in the functionalization
of hydroxyl-containing polymers using unsaturated anhydrides/
acyl chlorides.
Characterizations. Solubility of the complexes was judged by naked
eye after 0.1 g complex was put in 5 mL of solvent at room temperature
overnight. These solvents included water, ethanol, dimethylsulfoxide
As an electron donor, TEA or pyridine forms n-π type
complexes with electron acceptors such as alkyl halides, sulfonyl
(DMSO), methylene chloride, chloroform, tetrahydrofuran (THF), N,N-
9-13
chlorides, and acyl chlorides, involving proton/charge transfer.
Under photoirradiation, TEA can also react with methylene
dimethylformamide (DMF), acetone, xylene, hexane, ethyl ether,
toluene, and ethyl acetate. FTIR spectra were obtained on a Perkin-
1
4,15
16
1
chloride and C60
.
Known as the Einhorn reaction, much
Elmer Spectrum Spotlight 300 spectrometer with Diamond ATR. H
attention was paid in the past to the formation of intermediate
ketenes and the double bond shift from the R,ꢀ- to the ꢀ,γ-
NMR spectra were acquired on a Varian Mercury 300 spectrometer
using deuterated DMSO (DMSO-d) solutions containing tetramethyl-
1
silane (TMS). H NMR (300 MHz): TEA-AC: δ
H
5.85 (CH
, t, 9H); TEA-FC: δ
, q, 6H), 1.20 (-CH , t, 9H); TEA-
2
dCH, s,
*
To whom correspondence should be addressed. Tel.: 1-865-974-7809.
3H), 3.06 (N-CH
(CHdCH, s, 2H), 3.08 (N-CH
2
, q, 6H), 1.20 (-CH
3
H
5.76
Fax: 1-865-974-4115. E-mail: swang16@utk.edu.
2
3
10.1021/bm901237t
 2010 American Chemical Society
Published on Web 12/09/2009

Notes
Biomacromolecules, Vol. 11, No. 1, 2010 305
MA: δ
t, 9H); IA-TEA: δ
H
5.83, 6.09 (CHdCH, s, 2H), 2.88 (N-CH
5.85 (CH dC, s, 2H), 2.94 (N-CH
, d, 2H), 1.20 (-CH , t, 9H). IR: TEA-AC, TEA-FC: 2977,
2
, q, 6H), 1.20 (-CH
3
,
H
2
2
, q, 6H), 1.92
(
dC-CH
2
3
-
1
-1
+
-1
2
951 cm (N-H, C-H), 2617, 2500 cm (Ct N ), 1720 cm
-1
-1
(CdO), 1580 cm (CdC); TEA-MA: 2984 cm (N-H, C-H), 2657,
-1
+
-1
-1
2487 cm (Ct N ), 1780, 1720 cm (CdO), 1580 cm (CdC); TEA-
-
1
-1
+
IA: 2990, 2920 cm (N-H, C-H), 2657, 2487 cm (Ct N ), 1780,
-
1
-1
1
720 cm (CdO), 1580 cm (CdC).
Wide-angle X-ray diffraction (WAXD) patterns were obtained, in
reflection, using an X’pert diffractometer and CuKR radiation. Elemen-
tal microanalysis was performed by the Complete Analysis Laboratories,
Inc. (Parsippany, NJ). UV-vis spectra were obtained using a BioMate
5
UV-vis spectrophotometer (Thermo Scientific). Photoluminescence
excitation and emission spectra were recorded using a Perkin-Elmer
LS 55 fluorescence spectrometer equipped with a 20 kW Xenon
discharge lamp. The excitation wavelength was 370 nm and photolu-
minescence emission spectra were recorded from 380 to 600 nm. The
emission wavelength was monitored at 450 nm and photoluminescence
excitation spectra were recorded from 250 to 420 nm. The slit width
was 10 nm.
Figure 1. WAXD patterns and digital images of the complexes.
Cytotoxicity. Cytotoxicity evaluation was performed by harvesting
2
HEK293 cells in a 24-well plate at a density of ∼15000 cells/cm in 1
mL of primary medium, which consisted of high glucose Dulbecco’s
modified eagle medium, 10% fetal bovine serum, 1% amino acids, 1%
none-essential amino acids, and 1% sodium pyruvate (Gibco, Grand
Island, NY). The complexes were dissolved in the primary medium at
the concentrations of 0.01, 0.05, 0.1, and 0.5 g/L and exposed to
HEK293 cells for 1, 4, and 7 days. Wells seeded with HEK293 cells
at the same density in the absence of complexes were used as positive
controls and empty wells were used as negative controls. A colorimetric
cell metabolic assay (CellTiter 96 Aqueous One Solution, Promega,
Madison, WI) based on the MTS tetrazolium compound was used to
evaluate the number of viable cells, which could be correlated to the
UV absorbance at 490 nm measured on a microplate reader (SpectraMax
Plus 384, Molecular Devices, Sunnyvale, CA). Cell viability was then
calculated by normalizing the absorption to the average value from
the positive wells. Phase contrast microscopic images of the attached
HEK293 cells in the 24-well plate were taken using a Carl Zeiss
Axiovert 25 fluorescence microscope.
Figure 2. Proposed mechanisms for the reaction between AC and
ROH in the presence of TEA.
final products are theoretically identical (CH
Et N·HCl). Suggested by earlier investigations,
intermediate (CH dCdCdO) formed in situ is in equilibrium
2
dCH-COOR and
17-27
3
a ketene
2
20,23
with the complex, although we and some other researchers
were not able to detect. Unlike R,ꢀ-unsaturated acid chlorides
17-27
such as crotonoyl chloride,
no double bond shift can be
N·HCl in Figure 2 itself
proposed for AC and FC. The salt Et
3
was also prepared by directly reacting TEA with hydrochloride
and it was white. Therefore, the colorization in the resulted
polymers could only be attributed to the residue colored
complex, which was difficult to be removed completely after
5
the purification using anhydrous ethyl acetate. As reported
earlier, TEA did not form a yellow color or any product with
saturated succinic anhydride, as the color was from the
interaction of electrons in TEA and the π-orbitals of both
Results and Discussion
To mimic the condition in the esterification of hydroxyl-
1
,2
28
containing polymers or oligomers, TEA and unsaturated
anhydrides or acyl chlorides were diluted in methylene chloride,
and TEA solution was added dropwise into the other reactant.
If not diluted, reaction would generate great heat immediately
and result in black products. For MA, IA, and FC, two molar
ratios (1:1 and 2:1) were used to prepare complexes with TEA
but only 2:1 is discussed here. As demonstrated in the images
and WAXD patterns in Figure 1, TEA-MA and TEA-IA
complexes were black or dark red muddy compounds with broad
diffraction peaks, while TEA-FC and TEA-AC were brown or
yellow highly crystalline powders with shared sharp diffraction
peaks at 2θ ) 12.4, 17.6, 21.3, 25.3, 27.8, 33.2, and 35.4°. The
complexes demonstrated good solubility in water, ethanol, and
DMSO, while there was no solubility in THF, xylene, hexane,
ethyl ether, toluene, or ethyl acetate. TEA-FC could be partially
dissolved in methylene chloride and DMF and weakly dissolved
in acetone and chloroform, in which TEA-AC had little
solubility.
carbon-carbon double bond and carbonyl bond. Meanwhile,
pyridine was reported not to react with MA because of its lower
2
9,30
pK
a
of 5.20 compared with TEA (pK
a
) 10.85)
although it
1
9
can react with AC. We found that TEA did not react with
diethyl fumarate (DEF) although DEF has both CdC and CdO
bonds, suggesting that both conjugation and a large difference
a
in pK between two reactants are necessary for generating
colored complexes.
The chemical structures of the complexes have been verified
using FTIR and NMR spectra (see Experimental Section and
Supporting Information) and supported by elemental analysis.
The nitrogen content was 7.38 and 8.38% for TEA-AC and
1
TEA-FC, respectively. H NMR spectra demonstrated that all
the resonances could be assigned to the protons in the
complexes. The resonance of the -CH
0.97 in TEA, shifted to lower field, δ
The resonance of -CH - group, which was at δ
shifted to lower field, δ 2.82-2.99, in the complexes.
Carbon-carbon double bond (CdC) was indicated by the band
at δ 5.76-6.09 in the complexes compared with δ 7.05 in
FC, δ 7.10 in MA, δ 6.16-6.63 in AC, and δ 6.90 in IA, a
3
group, which was at δ
1.2, in the complexes.
2.42 in TEA,
H
Η
2
H
H
The formation of complex between TEA and AC, FC, MA,
and IA can be exemplified using AC in Figure 2. The supposed
reaction between AC and ROH (R ) H, CH
H
H
3 2 5
, C H , or polymer
H
H
H
backbone) in the presence of TEA via the bottom two-step route
is parallel to the formation of complex or acylammonium salt
clear shift to higher field. The shift of these bands was due to
the electron withdrawing effect of the positive nitrogen present
in the complexes.
+
-
28
(
CH
2
dCH-CONEt
3
Cl ) between AC and TEA, although the

3
06 Biomacromolecules, Vol. 11, No. 1, 2010
Notes
-
3
Figure 3. UV-vis spectra of the dilute solutions (4 × 10 g/L) of the
complexes in DMSO (solid curves) and ethanol (dotted curves).
-3
UV-vis spectra of the dilute solutions (4 × 10 g/L) of the
complexes in an inert solvent DMSO in Figure 3 indicate a
dominant broad absorption band centered at 260-300 nm, which
is characteristic of charge-transfer transitions associated with
intramolecular donor-acceptor complexes. Consistent with an
2
8
earlier report on TEA-MA, all the complexes had shoulder
peaks at longer wavelengths, suggesting the presence of an
extended conjugated system. In contrast, UV spectra of these
complexes in ethanol (Figure 3) or water (see Supporting
Information) showed a characteristic sharp peak at ∼210 nm
of the carbon-carbon double bond with no TEA attached,
confirming the mixture of CH
CH dCH-COOH and Et N·HCl resulted from alcoholysis or
hydrolysis (Figure 2). Compared with the resolved emission
2
dCH-COOC
2
H
5
or
2
3
Figure 4. (A) Viability of HEK293 cells in the culture medium with
the complexes at 0.01, 0.05, 0.1, 0.5 g/L for 1, 4, and 7 days. (B)
Phase contrast images of the cells cultured with the complexes at
0.01, 0.05, 0.1, and 0.5 g/L for 7 days. The scale bar at the right
bottom corner is 100 µm and is applicable for the images, except the
image of a positive (+) control well that contained no complex at the
left bottom corner, in which the scale bar represents 200 µm.
30,31
spectra of TEA in the gas phase excited at 206.2 nm,
broad,
featureless photoluminescence emission spectra (see Supporting
Information) of the dilute solutions in DMSO in the wavelength
range of 400-550 nm demonstrated the existence of complexes,
in agreement with the emission spectra of TEA-naphthalene
3
2
complexes.
We synthesized PCLF530 and PCL triacrylates (PCLTA300
and 900) using TEA or K CO from PCL diol and triol
precursors with nominal molecular weights of 530, 300, and
proliferation (see Supporting Information) could be found,
except TEA-AC and TEA-FC at 0.5 g/L, in which cell
proliferation was completely prohibited. Correspondingly, phase
contrast microscopic images at day 7 (Figure 4B) demonstrated
attached HEK293 cells with spread-out phenotype when the
concentration was 0.05 g/L or lower, while cells tended to be
round, and less attached cells could be seen at higher concentra-
tions of the complex. The cell viability results clearly indicated
that all these complexes were cytotoxic and their existence in
polymers and degraded products should be avoided. Based on
the contents of nitrogen in the complexes and polymers
synthesized with TEA, the concentration of complex in the
polymers was estimated to be 0.02-0.3 wt %. Although no
investigation has been performed to determine the concentration
of complex released from the polymer disk to the culture
medium in cell studies, this concern should not be ignored
considering that cell viability can be reduced by 10-20%, even
at a very low concentration of complex (0.01 g/L). PEGDA or
2
3
3
3
9
0
00 g/mol. The content of nitrogen was measured to be 0.19,
.86, and 2.35% for PCLF, PCLTA300, and 900 synthesized
using TEA, respectively. Consistently, a noticeable nitrogen
content of ∼0.11% was also reported for PCLF530 by Sharifi
3
4
et al. In contrast, nitrogen was not detectable (<0.02%) in the
2 3
polymers synthesized using K CO . This new proton scavenger
has also been applied in synthesizing PEG diacrylates (PEG-
DAs) to avoid colorization or contamination of the complexes.
The synthesis method using K CO also simplified the purifica-
2 3
tion step, as inorganic particles could be easily removed
completely from organic solution through centrifugation or
2 3
filtration. No detectable solubility of K CO was found in
methylene chloride. These cross-linkable polymers based on
oligomeric PCL and PEG precursors are transparent in solutions
7,35,36
and more efficient in photo-cross-linking.
In the preparation
1
,3
other cross-linkable polymers for preparing hydrogels may
suffer less from the colored complexes because the required
polymer concentration in hydrogels is lower than their hydro-
phobic counterparts and purification steps can remove water-
soluble complexes, at least partially.
of this manuscript, we noticed that propylene oxide was also
used as a proton scavenger to synthesize white PCLFs with a
negligible nitrogen content of 0.012%.
Figure 4A shows the viability of HEK293 cells cultured in
the primary media with four concentrations of complexes (0.01,
3
4
0
.05, 0.1, and 0.5 g/L) 1, 4, and 7 days post-seeding. Cell
Conclusion
viability decreased dramatically with increasing the concentra-
tion of complex, particularly TEA-AC. At a fixed concentration,
cell viability continues to drop over time, although cell
In summary, TEA forms water-soluble, colored complexes
with unsaturated anhydrides or acyl chlorides. Although the final

Notes
Biomacromolecules, Vol. 11, No. 1, 2010 307
products are identical theoretically, the complexation competes
with the esterification of hydroxyl-containing polymers and
resulted in colorization of the products. Furthermore, the
complexes demonstrate cytotoxicity at concentrations as low
as 0.01 g/L in culture medium for HEK293 cells. Inorganic weak
bases such as K CO are recommended to replace TEA in such
2 3
esterification to avoid colorization and simplify the reaction and
purification.
(8) Peter, S. J.; Suggs, L. J.; Yaszemski, M. J.; Engel, P. S.; Mikos, A. G.
J. Biomater. Sci., Polym. Ed. 1999, 10, 363.
9) Brown, H. C.; Eldred, N. R. J. Am. Chem. Soc. 1949, 71, 445.
(
(
(
10) Barrow, G. M.; Yerger, E. A. J. Am. Chem. Soc. 1954, 76, 5247.
11) Collins, R. F. Chem. Ind. 1957, 6, 704.
(12) Nakayama, J.; Tanuma, M.; Honda, Y.; Hoshino, M. Tetrahedron Lett.
984, 25, 4553.
1
(
13) Das, S.; Rajesh, C. S.; Thanulingam, T. L.; Ramaiah, D.; George,
M. V. J. Chem. Soc., Perkin Trans. 2 1994, 7, 1545.
14) Erra-Balsells, R.; Frasca, A. R. Aust. J. Chem. 1988, 41, 103.
15) Lawson, G. E.; Kitaygorodskiy, A.; Sun, Y.-P. J. Org. Chem. 1999,
(
(
Acknowledgment. This work was supported by the start-up
fund of the University of Tennessee, Knoxville. We thank Dr.
James Fleming in the Center of Environmental Biotechnology
for supplying HEK293 cells, Dr. Joseph E. Spruiell in our
department for WAXD measurements, and Royce Dansby-
Sparks and Dr. Ben Z. L. Xue in the Department of Chemistry
for the help with measurements of UV-vis and photolumines-
cence spectra.
6
4, 5913.
(16) Krauch, H.; Kunz, W. Organic Name Reactions; John Wiley & Sons:
New York, 1964; p 141.
17) Sauer, J. C. J. Am. Chem. Soc. 1947, 69, 2444.
(
(
18) Young, W. G.; Webb, I. D.; Goering, H. L. J. Am. Chem. Soc. 1951,
7
3, 1076.
(19) Hickmott, P. W. J. Chem. Soc. 1964, 3, 883.
(20) Payne, G. B. J. Org. Chem. 1966, 31, 718.
21) Ozeki, T.; Kusaka, M. Bull. Chem. Soc. Jpn. 1966, 39, 1995.
22) Ozeki, T.; Kusaka, M. Bull. Chem. Soc. Jpn. 1967, 40, 1232.
23) Ozeki, T.; Kusaka, M. Bull. Chem. Soc. Jpn. 1967, 40, 2686.
24) Gol’dfarb, Ya. L.; Ispiryan, R. M.; Belen’kii, L. I. Russ. Chem. Bull.
(
(
(
(
Supporting Information Available. Solubility, FTIR, NMR,
and photoluminescence spectra of the complexes, and HEK293
cell proliferation data. This material is available free of charge
via the Internet at http://pubs.acs.org.
1
969, 18, 839.
(25) Truce, W. E.; Bailey, P. S, Jr. J. Org. Chem. 1969, 34, 1341.
(26) Hickmott, P. W.; Miles, G. J.; Sheppard, G.; Urbani, R.; Yoxall, C. T.
J. Chem. Soc. Perk. T. 1 1973, 14, 1514.
(
27) Cardillo, G.; De Simone, A.; Mingardi, A.; Tomasini, C. Synlett 1995,
References and Notes
11, 1131.
(28) Mayahi, M. F.; El-Bermani, M. F. Can. J. Chem. 1973, 51, 3539.
(29) Kluger, R.; Hunt, J. C. J. Am. Chem. Soc. 1989, 111, 3325.
(30) Evtushenko, Yu. M.; Zaitsev, B. E.; Ivanoc, V. M. J. Anal. Chem.
(
(
(
(
(
1) Cruise, G. M.; Scharp, D. S.; Hubbell, J. A. Biomaterials 1998, 19,
287.
2) Kim, S.-H.; Won, C.-Y.; Chu, C.-C. J. Biomed. Mater. Res. 1999, 46,
60.
3) Jo, S.; Shin, H.; Shung, A. K.; Fisher, J. P.; Mikos, A. G. Macro-
molecules 2001, 34, 2839.
4) Kweon, H. Y.; Yoo, M. K.; Park, I. K.; Kim, T. H.; Lee, H. C.; Lee,
H. S.; Oh, J. S.; Akaike, T.; Cho, C. S. Biomaterials 2003, 24, 801.
5) Jabbari, E.; Wang, S.; Lu, L.; Gruetzmacher, J. A.; Ameenuddin, S.;
Hefferan, T. E.; Currier, B. L.; Windebank, A. J.; Yaszemski, M. J.
Biomacromolecules 2005, 6, 2503.
1
2
000, 55, 859.
1
(
(
31) Freeman, C. G.; McEwan, M. J.; Claridge, R. F. C.; Phillips, L. F.
Chem. Phys. Lett. 1971, 8, 77.
32) Kuzmin, M. G.; Guseva, L. N. Chem. Phys. Lett. 1969, 3, 71.
(33) Cai, L.; Wang, S. Polymer, in press, doi: 10.1016/j.polymer.2009.11.042.
(
(
(
34) Sharifi, S.; Mirzadeh, H.; Imani, M.; Ziaee, F.; Tajabadi, M.; Jamshidi,
A.; Atai, M. Polym. AdV. Technol. 2008, 19, 1828.
35) Wang, S.; Yaszemski, M. J.; Knight, A. M.; Gruetzmacher, J. A.;
Windebank, A. J.; Lu, L. Acta Biomater. 2009, 5, 1531.
(
6) Ono, T.; Sugimoto, T.; Shinkai, S.; Sada, K. Nat. Mater. 2007, 6,
36) Wang, S.; Yaszemski, M. J.; Gruetzmacher, J. A.; Lu, L. Polymer
4
29.
2
008, 49, 5692.
(
7) Wang, S.; Lu, L.; Gruetzmacher, J. A.; Currier, B. L.; Yaszemski,
M. J. Biomaterials 2006, 27, 832.
BM901237T