PET-coSIPA
PET-coSIPA
PET-coSIPA
www.elsevier.com/locate/polymer
Abstract
A series of poly(ethylene terephthalate-co-ethylene 5-sodiosulfoisophthalate) copolyesters containing from 1 up to 50 mol% of sulfonated
units was prepared by melt polycondensation from ethylene glycol and mixtures of dimethyl terephthalate and dimethyl 5-
sodiosulfoisophthalate. The resulting copolymers had a random microstructure and contained oligo(ethylene glycol) units in amounts
increasing with the content in sulfonated isophthalate units. Copolyesters with more than 20 mol% of 5-sodiosulfoisophthalic units were
amorphous and easily soluble in water. The hydrodegradability of the copolyesters was very high as compared to poly(ethylene
terephthalate), and increased with the content in sulfonated units. It was demonstrated that the susceptibility to acidic hydrolysis of these
copolymers is mainly due to the presence of the sodium sulfonate groups, the influence of the oligo(ethylene glycol) units in this regard being
noticeable but limited.
q 2003 Published by Elsevier Ltd.
Keywords: Poly(ethylene terephthalate); PET copolymers; Sulfonated copolyesters
titanate (TBT) was supplied by Merck –Schuchardt. All of (wavelength ¼ 1.5418 Å). The patterns were calibrated
them were reagent grades and used as received. All solvents with molybdenum sulfide.
used for purification and characterization were of either
technical or high purity grade and used as received.
1
H and 13C NMR spectra were recorded on a Bruker 2.2. General procedure for the synthesis of the PETSI
AMX-300 spectrometer at 25.0 8C operating at 300.1 and copolyesters
75.5 MHz, respectively. The spectra were internally refer-
enced to tetramethylsilane. The polyesters were dissolved in DMT, Na-DMSI, EG, and TBT catalyst were introduced
a CDCl3/trifluoroacetic acid (TFA) (1.5/1 v/v) mixture and into a three-necked 100-mL round-bottomed flask equipped
64 scans were acquired for 1H and 1000 to 10,000 for 13C with a mechanical stirrer, a nitrogen inlet, and a vacuum
NMR spectra, with 32 K data points and relaxation delays of distillation outlet, using the selected comonomer ratios. A
1 and 2 s, respectively. 13C – 1H heteronuclear shift molar ratio 2.2:1 of glycol to the total amount of the DMT
correlation spectra were recorded by means of the pulse and Na-DMSI monomers was used in all copolymerizations.
sequences (hxco) implemented in the Bruker NMR instru- Transesterification reactions were carried out at 190 8C
ment package. Gel permeation chromatography (GPC) under a nitrogen flow, whereas the polycondensation
analysis was performed on a Waters system equipped with reactions were performed in the range of 220– 270 8C
a refractive index detector and using 1,1,1,3,3,3-hexafluoro- under a 0.5 – 1 mbar vacuum. The polymerization was
2-propanol as the mobile phase. The molecular weights and allowed to proceed isothermally for the required period of
their distribution were calculated against monodisperse time with simultaneous removal of the excess of the glycol
poly(methyl methacrylate) standards using the Millenium by distillation. Finally, the reaction mixture was cooled
820 software. down to room temperature and atmospheric pressure was
The thermal behavior of the polyesters was examined by recovered using nitrogen to prevent degradation. The
differential scanning calorimetry (DSC) using a Perkin– resulting polymers were isolated by dissolving the reaction
Elmer DSC Pyris 1 instrument calibrated with indium. DSC mixture in a CHCl3/TFA (8/1 v/v) mixture, followed by the
data were obtained from 4 to 6 mg samples at a heating rate slow addition of cold diethyl ether under vigorous stirring.
of 10 8C min21 under a nitrogen circulation. Thermogravi- The precipitated polyesters were collected by filtration,
metric analyses (TGA) were performed using a Perkin– washed with diethyl ether, and finally dried under reduced
Elmer TGA-6 thermobalance at a heating rate of pressure.
10 8C min21 under a nitrogen atmosphere. Powder wide- The same procedure, with the corresponding changes in
angle X-ray diffraction patterns were obtained on flat the composition of the feed, was used for the preparation of
photographic films with a modified Statton camera using PET and the poly(ethylene-co-diethylene terephthalate)
nickel-filtered monochromatic Cu Ka radiation (PEDET) copolyester studied in this work.
O. Gaona et al. / Polymer 44 (2003) 7281–7289 7283
Table 1
Characteristics of the PETSI copolyesters
Films of PET, and the PETSI and PEDET copolymers 3.1. Synthesis and characterization
with a thickness of , 300 mm, were prepared by casting at
room temperature from a 10% (w/v) solution in CHCl3/TFA The procedure applied in this work to produce the PETSI
(8/1 v/v) on a silanized Petri dish. The films were cut into copolymers is outlined in Scheme 1. The process was
14-mm diameter, 60 to 90-mg weight disks, which were conducted in two stages with the feed consisting of a
dried in vacuo at 50 8C to constant weight. Samples were dimethyl terephthalate and dimethyl 5-sodiosulfoisophtha-
immersed in Elix-Millipore purified water (pH 4.35) in late mixture with the selected molar composition, and an
parallel experiments at 37 and 80 8C, respectively. After excess of ethylene glycol. A mixture of bis(2-hydroxyethyl)
immersion for the fixed period of time, the samples were terephthalate (BHET) and the sodium salt of bis(2-
rinsed thoroughly in water and dried to constant weight. hydroxyethyl) 5-sulfoisophthalate (BHESI), accompanied
Sample weighting, GPC measurements, and NMR spec- with additional amounts of higher oligoesters and oli-
troscopy were used to follow the evolution of the go(ethylene glycol)s, was the product of the transesterifica-
hydrodegradation. tion reactions happening in the first stage at 190 8C. The
Fig. 2. 13C NMR signal of the ethylene glycol units in PET70SI30, with indication of the dyad assignments and the deconvolution used for microstructure
determination. The peaks on the left without labeling arise from diethylene glycol units.
copolymers were formed in the second stage conducted The complex signals observed in the 3.8 – 4.9 ppm region
between 220 and 270 8C by cross transesterification arise from the oxyethylene units contained in the diol
reactions of the intermediate hydroxyethyl esters. counterpart of the copolymer. In the 13C NMR spectra, the
A series of PETSI copolyesters having a content in 5- aliphatic carbon resonances appear as two complex groups
sodiosulfoisophthalic units ranging from 1 to 50 mol% was of peaks situated in the 65– 75 ppm range. In addition to the
prepared by this procedure. The most relevant character- expected ethylene glycol units, diethylene glycol ( –CH2-
istics of the resulting sulfonated copolyesters are collected CH2OCH2CH2O –), triethylene glycol (– CH2CH2OCH2-
in Table 1. CH2OCH2CH2O –), and even higher oligo(ethylene glycol)
The chemical constitution and comonomer composition units are present in the PETSI copolyesters. An accurate
of PETSI was assessed by NMR spectroscopy. The 1H NMR assignment of the 1H and 13C peaks associated to the
spectra of the series are compared in Fig. 1. The relative area magnetically different ethylene protons and carbons present
of the aromatic proton signals appearing at 8.9 and 8.2 ppm in the copolyesters was achieved by heteronuclear corre-
for the 5-sodiosulfoisophthalic and terephthalic units, lation (HETCOR) analysis. The relative content of the
respectively, was measured to estimate the content of the oligo(ethylene glycol) units increased with the degree of
two aromatic comonomers in the copolyesters, which was sulfonation of the copolymer, the amount of monoethylene
found to be essentially the same as that used in the reaction glycol being reduced to less than 50 mol% in the PET70SI30
feed. and PET50SI 50 copolyesters. Signals arising from
Fig. 3. TGA traces of (a) BHESI and (b) PET80SI20, and their corresponding derivative curves.
O. Gaona et al. / Polymer 44 (2003) 7281–7289 7285
Table 2
Thermal and structural data of the PETSI copolyesters
PET 78 253 (46) 436 5.40 5.06 4.17 3.94 3.47 3.20 2.74 2.14
PET99SI1 74 240 (18) 438 5.48 5.00 4.17 3.99 3.47 3.20 2.78 2.14
PET98SI2 69 216 (7.4) 442 5.48 4.08 3.54 2.81
PET95SI5 66 198 (7.6) 436 5.48 4.08 3.54 2.85
PET90SI10 58 167 (3.5) 439 5.48 4.17 3.54 2.85
PET80SI20 56 – 437 –
PET70SI30 53 – 434 –
PET50SI50 27 – 429 –
a
Glass-transition and melting temperatures (melting enthaply in J g21 in parentheses) measured by DSC, and decomposition temperature measured by TGA
as the peak of the derivative curve.
b
Powder X-ray diffraction spacings recorded from the polymer coming directly from synthesis. Spacings of PET are consistent with a triclinic unit cell of
parameters: a0 ¼ 4:56 A; b0 ¼ 5:94 A;
c0 ¼ 10:75 A; a ¼ 98:58; b ¼ 1188; g ¼ 1128; as reported by Daubeny et al. [11]. Additional spacings appearing upon
annealing of PET99SI1 are given in italics.
hydroxyethyl end-groups were also identified and more 3.2. Thermal properties and structure
noticeably detected as the molecular weight of the polymer
decreased. Thermogravimetric analysis of BHESI revealed that the
The aliphatic region of the 13C NMR spectra thermal decomposition of this compound starts below
contained adequate information for the evaluation of 300 8C with the release of SO2 gas in almost quantitative
the comonomer distribution. In fact the signal arising amounts. Nevertheless, the thermal stability of PETSI
from the ethylene glycol units at 65– 67 ppm appeared to appeared to be not significantly affected by the presence
be resolved in four peaks, which could be assigned to the of the sulfonated units, which is contrary to that reported
three different dyad sequences feasible in PETSI previously by other authors [4c]. The TGA analysis of
copolyesters, i.e. T-EG-T, SI-EG-SI, and T-EG-SI or PETSI revealed decomposition temperatures (Td) around
SI-EG-T (Fig. 2). A quantitative analysis of these peaks 430 8C, such values being perfectly comparable with the
in relation to the composition of the copolyester led to decomposition temperature of PET. The TGA traces of the
the conclusion that the microstructure of the polymer is BHESI monomer and the 20 mol% sulfonated copolymer
almost at random. are shown in Fig. 3 and the Td observed for the whole series
The molecular weight of PETSI was estimated by GPC. of PETSI are compared in Table 2. Furthermore, in this
With the exception of PET90SI10, the overall trend displayed table, it can be seen that both the glass-transition (Tg) and
by the series is that molecular weight decreased with the melting (Tm) temperatures of PETSI, determined by DSC,
content of the copolyester in sulfonated units to a point that decreased as the content in sulfonated units increased. The
an oligomeric product consisting of about ten units in enthalpy of fusion followed the same trend and no sign of
average was the result of the copolymerization of the crystallinity was detected for PETSI containing 20 mol% or
equimolar mixture of DMT and Na-DMSI. In accordance higher of Na-SI units. The strong repressing effect of
with this trend, the homopolymer poly(ethylene 5-sodio- copolymerization on the crystallizability was also evi-
sulfoisophthalate) could not be prepared by this method. denced in the cooling DSC traces registered from molten
Fig. 4. Powder X-ray diffraction patterns of PETSI. The PET99SI1 sample was annealed as explained in the text.
7286 O. Gaona et al. / Polymer 44 (2003) 7281–7289
Fig. 5. The evolution of the remaining weight for PET, PETSI and PEDET samples immersed in water (pH 4.35) at 37 8C (top) and 80 8C (bottom) upon
immersion time.
polymers at a cooling rate of 10 8C min21, which showed no whereas diffuse patterns characteristic of amorphous
crystallization exotherm as soon as the content in Na-SI material were recorded from PET80SI20, PET70SI30, and
reached 5 mol%. It should be noted that the content of this PET50SI50. Comparison of the Bragg spacings observed for
copolymer in DEG and TEG amounts around 20 mol%. The semi-crystalline PETSI to those observed for PET revealed
crystallinity of PETSI could be noticeably increased by that the same crystalline structure must be adopted in both
annealing processes at a temperature 30– 40 8C below the cases. Additional spacings appearing upon annealing of
melting point. PET99SI1 corroborated this similitude. According to what is
Powder X-ray diffraction of PETSI corroborated the known to happen in nitrated PET copolymers [12], the
DSC results. As illustrated in Fig. 4, powder X-ray sulfonated units would remain confined in the amorphous
diffraction patterns consisting of sharp rings were obtained phase of the material, since the bulkiness of the sulfonate
for PET90SI10, as well as for other less sulfonated members, group prevents them from entering the crystal lattice.
O. Gaona et al. / Polymer 44 (2003) 7281–7289 7287
Fig. 6. Change in the number-average molecular weight of PETSI and PEDET samples immersed in water (pH 4.35) at 37 8C (top) and of PET, PETSI and
PEDET samples at 80 8C (bottom) as a function of the immersion time.
Fig. 7. Comparison of the signals in the aromatic region of 1H NMR spectra of PET80SI20: (a) initial sample, (b) degraded disk and (c) residue left by the mother
solution after evaporation to dryness.
polymers. Apparently, ester hydrolysis happened very fast effect exerted by the sodio sulfonate group, and (b) the
in the initial stage to almost cease for a certain chain length, hydrolytic stability of the sulfonated polyester is lower
which is shorter as the content in Na-SI of the copolyester because of the greater hydrophilicity, the lower crystallinity,
increased. and the high polarity of the ionic aggregated microdomains
The 1H NMR analysis of the degraded disks and the of these copolymers, all facilitating the penetrability of
residue left by the incubation water after evaporation, water. Grobe et al. [14] reported that the inductive effect of
revealed an increasing presence of hydroxyethyl end-groups the Na-SI group should be ignored since BHET and BHESI
with degradation time, which is indicative of the occurrence have the same hydrolysis rates under alkaline conditions.
of hydrolysis of the ester groups. As illustrated in Fig. 7, the Chisholm et al. [13] reached to similar conclusions from
changes observed in the aromatic signals of the 1H NMR hydrolysis experiments carried out on DMT and Na-DMSI.
spectra evidenced that the degradation process involved These results lead to favor the interpretation that hydrolysis
mainly the sulfonated units. enhancement is caused by the higher water absorption
Comparison of the spectra from the residual disks upon happening in the sulfonated poly(butylene terephthalate)
incubation revealed that the relative ratio of the terephthal- copolymers. Such a hydrolysis mechanism can be readily
ate to 5-sodiosulfoisophthalate units increased in the postulated for the hydrolytic degradation of the PETSI
residual polymer disk samples, whereas it largely decreased copolymers.
in the residue left by the mother solution, with the formation
of 5-sodiosulfoisophthalic acid taking place in a certain
extent. In view of these results, the hydrolytic degradation Acknowledgements
of PETSI under slightly acidic conditions is thought to
proceed according to the reaction mechanism as depicted in
This work has been supported by CICYT grants no
Scheme 2.
MAT99-0578-CO2-02 and MAT02-04600-CO2. Financial
The effect of sulfonation on the hydrodegradability could
support received from Catalana de Polı́mers, S.L. (Barce-
be in part accounted for by the presence in the copolymer of
lona, Spain), is also gratefully acknowledged.
the oligo(ethylene glycol) units. In order to assess more
accurately the role of the sulfonated units on the
hydrodegradability of PET, the contribution of the
oligo(ethylene glycol) units was evaluated separately. For References
this purpose, copolyester PEDET containing 3.5 mol% of
diethylene glycol (DEG) units was prepared and its [1] Po R. Poly(ethylene terephthalate) modified with ionic groups in
Salamone JC. editor. The polymeric materials encyclopedia, Boca
hydrolytic degradability behavior was evaluated and Raton, Fl: CRC Press, pp 6100–6.1996.
compared to that displayed by PET99SI1, which is the [2] Hong Z. J Appl Polym Sci 1987;34:1353.
PETSI copolyester containing a similar amount of DEG [3] Timm DA, Hsieh YL. J Polym Sci: Polym Phys Ed 1993;31:1873.
units. The results of this study are plotted in both Figs. 5 and [4] (a) Ostrowska-Czubenko J, Ostrowska-Gumkowska B. Eur Polym J
6. Although an appreciable degradation is detected for 1988;24:65. (b) Ostrowska-Czubenko J, Ostrowska-Gumkowska B.
Eur Polym J 1988;24:803. (c) Ostrowska-Czubenko J, Ostrowska-
PEDET, the decay is much more prominent in the case of Gumkowska B. Eur Polym J 1991;27:681. (d) Ostrowska-Czubenko J,
PET99SI1, indicating that the Na-SI units are the main units Ostrowska-Gumkowska B. Eur Polym J 1994;30:875.
responsible for the exceptionally high hydrodegradability [5] Eisenberg A, Hird B, Moore RB. Macromolecules 1990;23:4098.
displayed by the PETSI copolyesters. [6] Greener J, Gillmor JR, Daly RC. Macromolecules 1993;26:6416.
Chisholm et al. [13] have recently carried out a detailed [7] Abrabams DH. Am Dyest Rep 1970;59:36.
[8] Burget L, Prikryl J, Ruzicka J. Faserforsch Textiltech 1978;29:426.
study on the hydrolytic degradation of sulfonated poly-
[9] Ingamells W, Lilou SH, Peters RH. J Appl Polym Sci 1981;26:4087.
(butylene terephthalate) copolymers. In this work they [10] Datye KV, Palan BH. J Appl Polym Sci 1989;38:1447.
discussed thoroughly the different reasons that can be [11] Daubeny RP, Bunn CW, Brown CJ. Proc R Soc London: Ser A 1954;
invoked to explain the strong depressing influence of the 5- A226:531.
sodiosulfoisophthalate units on the hydrolytic stability. Two [12] (a) Kint DPR, Martı́nez de Ilarduya A, Muñoz Guerra S. J Polym Sci:
Polym Phys Ed 2001;39:1553. (b) Kint DPR, Martı́nez de Ilarduya A,
groups of factors, one of chemical nature and the other
Muñoz Guerra S. J Polym Sci: Polym Phys Ed 2002;40:2759.
related to the accessibility of the polymer to water were [13] Chisholm BJ, Sisti L, Soloveichik S, Gillette G. Polymer 2003;44:
considered: (a) the ester hydrolysis reaction rate is higher in 1903.
the sulfonated polyester due to the inductive and/or catalytic [14] Grobe H, Nissen D, Rossbach V. Makromol Chem 1975;176:2839.