Melt derived blocky copolyesters: New design features for polycondensation

Article abstract of DOI:10.1021/ma3011075

Melt polycondensation was utilized to chain extend polytrimethylene terephthalate with 1,3-propanediol based fluorinated isophthalic oligomers, resulting in copolymers with retained microstructure. Our findings point toward the formation of a blocky type

Full text of DOI:10.1021/ma3011075

Article
pubs.acs.org/Macromolecules
Melt Derived Blocky Copolyesters: New Design Features for
Polycondensation
Neville E. Drysdale, Yefim Brun, Elizabeth F. McCord, and Fredrik Nederberg*
DuPont Central Research & Development, Experimental Station, Wilmington, Delaware 19880, United States
S
* Supporting Information
ABSTRACT: Melt polycondensation was utilized to chain
extend polytrimethylene terephthalate with 1,3-propanediol
based fluorinated isophthalic oligomers, resulting in copoly-
mers with retained microstructure. Our findings point toward
the formation of a blocky type copolymer. In general,
formation of block or segmented copolymers from melt
derived polycondensation is a very challenging task due to the
propensity for adverse randomization reactions. Supported by
size exclusion chromatography, our copolymers are fully chain
extended, with no presence of the initial components.
Furthermore, thermal differential scanning calorimetry has
confirmed that the melt characteristics of the starting components are retained. In addition, interaction polymer chromatography
and sequence distribution analysis using 13C NMR supports a blocky backbone microstructure. Seemingly, intermolecular chain
end condensation occurs, whereas transesterification is dormant. While these findings open up new doors for polymer/materials
development, we are particularly interested in these structures as melt additives to address oil repellency of polyester blends.
When used in blends these blocky additives show an improvement in oil repellency compared with random additives of identical
molar composition, i.e., they are more fluorine efficient.
new line of building blocks fills a shortcoming of fluorinated
monomers available for step growth polymerization, and
importantly allows pendant trifluoromethyl functional groups
to be readily introduced to polymers. Trifluoromethyl is a very
useful functional group, known to provide even lower surface
tension than PTFE.1,3 When used in polyesters, polyamides,
and polyaramids, the resulting fluorinated (co)polymers were
found to possess increased repellency to water and oil.4−6 We
have recently built on the synthetic design capability of these
materials to prepare blocky fluorinated/nonfluorinated aro-
matic copolyesters derived from a commercially attractive melt
process, a feature not typical for polycondensation polymers.
As background, Wallace Carother’s pioneering research on
step growth polycondensation was the onset for the develop-
ment of engineering polymers in the 20th century.7,8 Whereas
Carothers and co-workers outlined synthetic routes to
polyesters, polyamides, and others, Flory later coupled
experimental observations to theoretical considerations. Indeed,
one of the fundamental features of polycondensation is that of
equal reactivity between functional groups at all stages of the
polymerization.9 There are several implications. For one, it is
difficult to control the molecular weight or the molecular
weight distribution of the formed polymer (PDI ∼ 2.0 for a
polycondensation process). This can be understood by the fact
INTRODUCTION
■
Fluorinated man-made materials have provided consumer
benefits for a wide range of applications during the last 50
years. Fluorinated molecules, for example, constitute the
primary component in refrigerants, and fluorinated polymers
are used in a vast array of products including textiles, carpets,
cookware, pipes/tubes, construction materials, and electronic
materials to provide, for example, repellency, lubricity, chemical
inertness, and low dielectric constant materials.1 In polymers,
the utilization of fluorine can be separated into three main
categories: (i) polytetrafluoroethylene (PTFE), (ii) melt
processable fluorinated (co)polymers including fluorinated
ethylene propylene (FEP), perfluoroalkoxys (PFA), and
Krytox, (iii) tetrafluoroethylene telomers. While these classes
of fluorinated materials provide specific benefits for the
application, they share a common feature in being derived
from radical chain growth polymerization (except for Krytox
that is made from ring-opening polymerization of hexafluor-
opropylene oxide). We have recently modified fluorinated
monomers available for chain growth polymerization to be
compatible for step growth polycondensation, thus opening the
scope to a new range of industrially relevant polymers including
polyesters, polyamides, and polyaramids. The essence of this
finding was the preparation of fluorinated aromatic ethers, via
the addition of hydroxyl aromatic diesters to fluorinated vinyl
ethers, under base catalyzed conditions.2 The reaction is mild,
efficient, and scalable, and the obtained monomer is compatible
with standard polycondensation protocols. Furthermore, this
Received: June 1, 2012
Revised: September 12, 2012
Published: October 1, 2012
© 2012 American Chemical Society
8245
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
PDO) [1,3-propanediol used in polyesters aka 3G], 1,1,1,2,2,3,3-
heptafluoro-3-(1,2,2-trifluorovinyloxy) propane (PPVE), Poly-
(trimethylene terephthalate) (PTT aka 3-GT), bright 1.02 IV (Mn
25800 g/mol, Mw 49700, PDI 1.92, DP ∼ 125). Purchased from
SynQuest Laboratories, and used as received: 1,1,1,2,2,3,3-heptafluoro-
3-(1,1,1,2,3,3-hexafluoro-3-(1,2,2-trifluorovinyloxy)propan-2-yloxy)-
propane (PPPVE).
Methods. Surface Analysis. The purpose of the analysis was to
study various blends and functional polymers ability to repel
hexadecane (oil) and to measure surface tensions. Surface contact
angles of hexadecane (advancing and receding) on polymer film were
that monomers, without discrimination, readily react with other
monomers, oligomers, or polymers. Another implication of
equal reactivity is that a selected molecular structure, e.g.,
sequences of specific monomer units along the chain, is hard to
achieve. If one, for example, would polymerize isophthalic and
terephthalic acid in equal proportions together with one
particular glycol, for example ethylene glycol, this would render
a random copolymer structure in which isophthalate and
terephthalate units would randomly appear along the polymer
backbone. Published data reports that all crystallinity seen in
PET is removed by replacing ∼20% of terephthalic acid with
isophthalic acid, i.e. a fully amorphous copolymer is formed.10
Similar results are found if one would target a 50/50 mol %
incorporation of two different glycols, for example ethylene
glycol and 1,3-propanediol, in combination with dimethylter-
ephthalate; this too renders a dominantly amorphous
copolymer.11 As most commodity polycondensation polymers
are homopolymers from the same set of diol/diamine and
diacid, for example nylon-6,6, poly(ethylene terephthalate), and
poly(butylene terephthalate), this feature perhaps is of less
concern. However it does construct a major hurdle if one would
consider the manufacturing of structurally challenging
compositions like segmented or block copolymers, based solely
on a polycondensation process. The literature entails numerous
examples of the use of disparate polymerization techniques to
provide block copolymers. A chain growth polymerization
technique could be used followed by a step growth polymer-
ization to render a blocky type copolymer, or vice versa. One
example is the Hytrel thermoplastic elastomer which consists of
a soft poly(tetramethylene glycol) segment made from ring-
opening polymerization of tetrahydrofuran and a hard poly-
(butylene terephthalate) segment prepared through step
growth polymerization.12 Similarly, fluorinated elastomers
were introduced by Tonelli et al.13 Another noteworthy
achievement is the work by Yokozawa et al. They have
developed a so-called chain growth polycondensation process
for the manufacturing of condensation polymers with defined
molecular weights, narrow molecular weight distributions and
selective compositions.14,15 It is achieved by the incorporation
of monomers with substituents of various electron donating
ability, such that the monomer−monomer reactivity is
suppressed and that monomers instead selectively add to
propagating polymer chain-ends. This leads to a living
polycondensation process in which controlled molecular
weights and narrow polydispersities (i.e., PDI’s below 1.2) are
achieved. Furthermore, it allows one to add monomers in
sequence for the construction of block copolymers, star- and
graft polymers.
́ ́
recorded on a Rame-Hart Model 100−25-A goniometer (Rame-Hart
Instrument Co) with an integrated DROPimage Advanced v2.3
software system. Four μL of hexadecane was dispensed using a micro
syringe dispensing system. Surface tensions were calculated using the
mean-harmonic method utilizing methyl iodide and water as probe
liquids.26
Size Exclusion Chromatography. A size exclusion chromatography
system Alliance 2695 from Waters Corporation (Milford, MA), was
provided with a Waters 414 differential refractive index detector, a
multiangle light scattering photometer DAWN Heleos II (Wyatt
Technologies, Santa Barbara, CA), and a ViscoStar differential capillary
viscometer detector (Wyatt). The software for data acquisition and
reduction was Astra version 5.4 by Wyatt. The columns used were two
Shodex GPC HFIP−806 Mstyrene-divinyl benzene columns; and one
Shodex GPC HFIP−804 Mstyrene-divinyl benzene column. The
specimen was dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP)
containing 0.01 M sodium trifluoroacetate by mixing at 50 °C with
moderate agitation for 4 h followed by filtration through a 0.45 μm
PTFE filter. Concentration of the solution was 2 mg/mL. Data was
taken with the chromatograph set at 35 °C, with a flow rate of 0.5 mL/
min. The injection volume was 100 μL. The run time was 80 min.
Data reduction was performed incorporating data from all three
detectors described above. Eight scattering angles were employed with
the light scattering detector. No standard for column calibration was
involved in the data processing
Interaction Polymer Chromatography. The purpose of interaction
polymer chromatography was to separate macromolecules based on
functionality. The experiments were performed using chromatography
system Alliance 2695 coupled with UV/vis 287 detector from Waters
at a 290 nm wavelength. FluoroFlash F8 HPLC 4.6 × 150 mm column
from Fluorous Technologies (PA, USA) has been used for separation
in linear water−HFIP (65−100%) 20 min gradient at flow rate 0.5
mL/min.
Thermal Analysis. Glass transition temperature (Tg) and melting
point (Tm) were determined by differential scanning calorimetry
(DSC) performed according to ASTM D3418−08. Specifically a
heat−cool−heat protocol was used, under a protecting nitrogen
atmosphere, that heated (from 0 to 250 °C), cooled (from 250 to 0
°C), and reheated (from 0 to 250 °C) samples at 10°/min, and
thermal transitions were recorded.
NMR Analysis. 13C NMR data was acquired on a 700 MHz NMR
Varian direct drive with a 10 mm probe, on 310 mg of polymer and 30
mg of chromium acetyl acetonate (CrAcAc) dissolved in deuterated
1,1,2,2 tetrachloroethylene (tce-d2) to 3.1 mL total volume with
minimal heating. Specifically only the CH2O ester carbon was analyzed
for sequence distribution determination. Four CH2O ester carbons
were analyzed in the region of 62.6−62.0 ppm of the copolymers made
and 13C NMR data is only given for those compositions. NMR spectra
were acquired using an acquisition time of 1 s, 90 deg pulse of about
11 μs, spectral width of 44.6 kHz, recycle delay of 5 s, temperature of
120 °C, 2500−4500 transients averaged. Data processed typically with
exponential line broadening of 0.5−2 Hz and zero fill of 512k. Spectra
were referenced to the tce-d2 carbon at 74.2 ppm.
Blends of PTT and Fluorinated Additives Using a DSM
Microcompounder. Blends of neat PTT with fluorinated homo or
copolymers were made in a DSM microcompounder. In general,
blends were made with copolymer additives targeting a total additive
concentration of ∼1−50 wt %. The DSM system is a PC controlled 15
cubic centimeter (cc), co-rotating, intermeshing (self-wiping), 2-
In this report we outline new and selective routes to blocky
copolyesters utilizing polycondensation entirely in the melt.
This is demonstrated using polytrimethylene terephthalate
(PTT) and oligomers based on novel fluorinated aromatic
ether monomers. This new family of fluorinated monomers
provides the ability to address oil repellency of inherently
lipophilic aromatic polyesters, like PTT, and others, while at the
same time introducing new macromolecular design capabilities.
EXPERIMENTAL SECTION
■
Materials. Purchased from Aldrich Chemical Co., and used as
received, were: dimethyl terephthalate, titanium(IV) isopropoxide
(TyzorTPT), tetrahydrofuran, dimethyl 5-hydroxyisophthalate, potas-
sium carbonate. Obtained from the DuPont Company and used as
received, unless otherwise noted, were: Biobased 1,3-propanediol (Bio-
8246
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
tipped, conical twin-screw machine with a recirculation loop, discharge
valve, nitrogen purge system, and with three different heating zones.
Typically, a temperature of 250 °C was used for all three heat zones.
Under nitrogen, PTT and the additive were charged and stirred with a
speed of 150 rpm for a total mixing time of 5 min. The stirring speed
and mixing may be varied, for example, speeds of 245 rpm with 5 min
mixing time could alternatively be used. Following the given mixing
time the discharge valve was opened and an extruded ribbon collected.
Synthesis of Dimethyl 5-(1,1,2-trifluoro-2-(perfluoropropoxy)-
ethoxy) Isophthalate (F10-iso). In a nitrogen gas supplied drybox,
anhydrous THF (500 mL) and dimethyl 5-hydroxy-isophthalate (42 g,
0.20 mol) were added to an oven dry 1000 mL round-bottom flask
equipped with a stir bar. Potassium carbonate (6.9 g, 50.4 mmol) was
added and an addition funnel connected. 1,1,1,2,2,3,3-Heptafluoro-3-
(1,2,2-trifluorovinyloxy)propane (79.80 g, 0.30 mol) was added via the
addition funnel and the reaction mixture was heated to reflux and
stirred overnight (t ∼ 16 h), oil bath temperature ∼80 °C. The
following morning the potassium carbonate catalyst was removed via
filtration through a bed of silica gel. The supernatant was concentrated
under vacuum (roto-vap) and then vacuum distilled, collecting the
fraction boiling at 139−145 °C (1.8−1.5 Torr), yielding 81.04 g (85%)
condenser. The reactants were stirred under a nitrogen purge at a
speed of 50 rpm while the condenser was kept at 23 °C. The contents
were degassed three times by evacuating down to a pressure of 100
Torr and refilling back to atmospheric pressure with N2 gas.
TyzorTPT catalyst (45 mg) was added after the first evacuation.
The flask was immersed into a preheated metal bath after the three
degassing/repressurization cycles set at 210 °C and held for 90 min
while stirring speed was increased from 50 to 180 rpm. Following the
90 min hold, the nitrogen purge was discontinued and a vacuum ramp
was started such that after about 60 min the vacuum reached a value of
50−60 mTorr. The reaction was held under vacuum at 50−60 mTorr
for an additional 90 min with stirring at 180 rpm. The reaction vessel
was then removed from the heat source. The overhead stirrer was
stopped and elevated from the floor of the reaction vessel. The vacuum
was then turned off and the system was purged with N2 gas at
atmospheric pressure. The thus formed product mixture was allowed
to cool to ambient temperature. The product was recovered after
1
carefully breaking the glass with a hammer. Yield = 88%. H NMR
(CDCl3) δ: 8.60 (ArH, s, 1H), 8.00 (ArH-, s, 2H), 7.70 (ArH, s, 4H),
6.15 (−CF2−CFH−O−, d, 1H), 4.70−4.50 (COO−CH2−, m, 4H),
3.95 (−CH2−OH, t, 2H), 3.85 (−CH2−O−CH2−, t, 4H), 2.45−2.30
(−CH2−, m, 2H), 2.10 (−CH2−CH2−O−CH2−CH2−, m, 4H).
Preparation of PTT Homopolymer from Dimethylterephthalate
and 1,3-Propanediol. Dimethylterephthalate (150 g), and 1,3-
propanediol (105.9 g) were charged to an oven-dried 500 mL three
necked round-bottom flask equipped with an overhead stirrer and a
distillation condenser. The reactants were stirred under a nitrogen
purge at a speed of 10 rpm while the condenser was kept at 23 °C. The
contents of the flask were degassed three times by evacuating down to
500 mTorr and refilling back to atmospheric pressure with N2 gas.
TyzorTPT catalyst (94 mg) was added after the first evacuation.
Following the three degassing cycles, the flask was immersed into a
preheated metal bath set at 160 °C. The solids were allowed to
completely melt at 160 °C for 20 min while the stirring speed was
slowly increased to 180 rpm. The temperature was increased to 210 °C
and was held at 210 °C for 90 min. After 90 min at 210 °C, the
temperature was increased to 250 °C after which the nitrogen purge
was discontinued, and a vacuum ramp was started such that after about
60 min the vacuum reached a value of about 60 mTorr. After an
additional 30 min at 250 °C and 60 mTorr, the heat source was
removed. The overhead stirrer was stopped and elevated from the
floor of the reaction vessel. The vacuum was then turned off and the
system purged with N2 gas at atmospheric pressure. The thus formed
product was allowed to cool to ambient temperature. The product was
recovered after carefully breaking the glass with a hammer. Yield =
85% of PTT polymer. 1H NMR (CDCl3/TFA-d), δ: 8.25−7.90
(ArH−, m, backbone), 7.65 (ArH, s, cyclic dimer), 4.75−4.45 (COO−
CH2−, m, backbone), 3.97 (HO−CH2−R, t-broad, end group), 3.82
(−CH2−O−CH2−, t, backbone DPG), 2.45−2.05 (−CH2−, m,
backbone).
Copolymerization of Dimethylterephthalate, F10-iso, and 1,3-
Propanediol. A 12.2 g sample of dimethylterephtalate, 30 g of the F10-
iso prepared, and 17.25 g of 1,3-propanediol were charged to an oven-
dried 500 mL three necked round-bottom flask equipped with an
overhead stirrer and a distillation condenser kept at 23 °C. The
reactants were stirred under a nitrogen purge at a speed of 50 rpm.
The contents were degassed three times by evacuating down to 100
Torr and refilling back to atmospheric pressure with N2 gas.
TyzorTPT catalyst (13 mg) was added after the first evacuation.
The reaction flask was immersed into a preheated metal bath set at 160
°C. The solids were allowed to completely melt at 160 °C for 20 min,
after which the stirring speed was slowly increased to 180 rpm. The
temperature was increased to 210 °C and maintained for 60 min. After
60 min, the nitrogen purge was discontinued, and a vacuum ramp was
started such that after an additional 60 min the vacuum reached 50−60
mTorr. As the vacuum stabilized, the stirring speed was increased to
225 rpm and the reaction held for 3 h. The heat source was then
removed. The overhead stirrer was stopped and elevated from the
floor of the reaction vessel. The vacuum was then turned off and the
system was purged with N2 gas at atmospheric pressure. The thus
1
of a clear liquid product. H NMR (CDCl3), δ: 8.60 (s, 1H), 8.03 (s,
2H), 6.22 (d, 1H), 3.95 (s, 6H).
Synthesis of (Dimethyl 5-(1,1,2-trifluoro-2-(1,1,2,3,3,3-hexa-
fluoro-2-(perfluoropropoxy)propoxy)ethoxy) Isophthalate (F16-iso).
In a nitrogen gas supplied drybox, anhydrous THF (500 mL) and
dimethyl 5-hydroxy-isophthalate (42 g, 0.20 mol) were added to an
oven dry 1000 mL round-bottom flask equipped with a stir bar.
Potassium carbonate (6.95 g, 50.4 mmol) was added and an additional
funnel connected. 1,1,1,2,2,3,3-Heptafluoro-3-(1,1,1,2,3,3-hexafluoro-
3-(1,2,2-trifluorovinyloxy)propan-2-yloxy)propane (129.60 g, 0.30
mol) was added via the addition funnel and the reaction mixture
was heated to reflux and stirred overnight (t∼16hrs), oil bath
temperature ∼80 °C. The following morning the potassium carbonate
catalyst was removed via filtration through a bed of silica gel. The
supernatant was concentrated under vacuum (roto-vap) and then
vacuum distilled, collecting the fraction boiling at 141−148 °C (1.1−
0.95 Torr), yielding 123.3 g (96%) of a clear liquid product. 1H NMR
(CDCl3), δ: 8.60 (s, 1H), 8.03 (s, 2H), 6.22 (d, 1H), 3.95 (s, 6H).
Preparation of 3-GF10-iso Homopolymer from F10-iso and 1,3-
Propanediol. A 150 g sample of the F10-iso prepared and 43.1 g of 1,3-
propanediol were charged to an oven-dried 500 mL three necked
round-bottom flask equipped with an overhead stirrer and a distillation
condenser kept at 23 °C. The reactants were stirred under a nitrogen
purge at a speed of 50 rpm. The contents were degassed three times by
evacuating down to 100 Torr and refilling back to atmospheric
pressure with N2 gas. TyzorTPT catalyst (45 mg) was added after the
first evacuation. The flask was then immersed into a preheated metal
bath set at 160 °C and held for 20 min while slowly increasing the
stirring speed to 180 rpm after which the temperature was increased to
210 °C and the reaction flask was held for an additional 90 min still at
180 rpm. Following the 90 min hold, the nitrogen purge was
discontinued and a vacuum ramp was started such that after about 60
min the vacuum reached a value of 50−60 mTorr. The reaction was
held for an additional 90 min with stirring at 180 rpm. The heat source
was then removed. The overhead stirrer was then stopped and
elevated from the floor of the reaction vessel. The vacuum was then
turned off, and the system was purged with N2 gas. The thus formed
product was allowed to cool to ambient temperature. The product was
recovered after carefully breaking the glass with a hammer. Yield =
82.6%.
1H NMR (CDCl3), δ: 8.60 (ArH, s, 1H), 8.00 (ArH−, s, 2H), 7.70
(ArH, s, 4H), 6.15 (−CF2−CFH−O−, d, 1H), 4.70−4.50 (COO−
CH2−, m, 4H), 3.95 (−CH2−OH, t, 2H), 3.85 (−CH2−O−CH2−, t,
4H), 2.45−2.30 (−CH2−, m, 2H), 2.10 (−CH2−CH2−O−CH2−
CH2−, m, 4H).
Preparation of 3-GF16-iso Homopolymer from F16-iso and 1,3-
Propanediol. A 150 g sample of the F16-iso prepared and 32 g of 1,3-
propanediol were charged to an oven-dried 500 mL three necked
round-bottom flask equipped with an overhead stirrer and a distillation
8247
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
Scheme 1. General Route to Fluorinated Aromatic Ethers from Dimethyl 5-Hydroxyisophthalate
1
Figure 1. H NMR (CDCl3) of fluorinated aromatic ethers from dimethyl 5-hydroxyisophthalate.
formed product was allowed to cool to ambient temperature. The
product was recovered after carefully breaking the glass with a
hammer. Yield = 90% of clear product. H NMR (CDCl3), δ: 8.60
(ArH, s, 1H), 8.15−8.00 (ArH-, m, 2 + 4H), 7.65 (ArH, s, 4H), 6.15
(−CF2−CFH−O−, d, 1H), 4.70−4.50 (COO−CH2−, m, 4H), 3.95
(−CH2−OH, t, 2H), 3.85 (−CH2−O−CH2−, t, 4H), 2.45−2.30
(−CH2−, m, 2H), 2.10 (−CH2−CH2−O−CH2−CH2−, m, 4H). 13C
NMR (CDCl3), δ: 62.6, 62.4, 62.1, 62.0.
CFH−O−, d, 1H), 4.70−4.50 (COO−CH2−, m, 4H), 3.95 (−CH2−
OH, t, 2H), 3.85 (−CH2−O−CH2−, t, 4H), 2.45−2.30 (−CH2−, m,
2H), 2.10 (−CH2−CH2−O−CH2−CH2−, m, 4H). 13C NMR
(CDCl3), δ: 62.6, 62.4, 62.1, 62.0.
1
Preparation of Blockcopolymers from 3-GF10-iso and PTT. A 20 g
sample of PTT, and 46 g of 3-GF10-iso were charged to an oven-dried
250 mL three necked round-bottom flask equipped with an overhead
stirrer and a distillation condenser kept at 23 °C. The reaction mass
was kept under N2 purge atmosphere. The contents were degassed
once by evacuating the reaction flask down to 150 mTorr and refilling
back to atmospheric pressure with N2 gas. TyzorTPT catalyst (20 mg)
was added after the evacuation and repressurization. The nitrogen
purge was then discontinued, and a vacuum ramp was started such that
after about 30 min the vacuum reached a value of about 60 mTorr.
The reaction flask was then immersed into a preheated metal bath set
at 250 °C and the contents of the reaction flask were allowed to melt
and equilibrate for 10 min. Stirring was initiated and speed was slowly
increased to 180 rpm. The thus formed melt was left under vacuum
with stirring for 15 min. The heat source was then removed. The
overhead stirrer was then stopped and elevated from the floor of the
reaction vessel. The vacuum was turned off, and the system was purged
with N2 gas. The thus formed product was allowed to cool to ambient
temperature. The product was recovered after carefully breaking the
glass with a hammer. Yield: 91.2% of turbid product. 13C NMR (tce-
d2), δ: 62.9 (E), 62.7 (D), 62.4 (G), 62.2 (F).
Copolymerization of Dimethylterephthalate, F16-iso, and 1,3-
Propanediol. Dimethylterephthalate (30.1 g), F16-iso (100 g), and 1,3-
propanediol (42.6 g) were charged to an oven-dried 500 mL three
necked round-bottom flask equipped with an overhead stirrer and a
distillation condenser kept at 23 °C. The reactants were stirred under a
nitrogen purge at a speed of 50 rpm. The contents were degassed three
times by evacuating down to 100 Torr and refilling back to
atmospheric pressure with N2 gas. TyzorTPT catalyst [40 mg] was
added after the first evacuation. The flask was immersed into a
preheated metal bath set at 160 °C. The solids were allowed to
completely melt at 160 °C for 20 min after which the stirring speed
was slowly increased to 180 rpm. The temperature was increased to
210 °C and maintained for 90 min. After 90 min at 210 °C, the
nitrogen purge was discontinued, and a vacuum ramp was started such
that after an additional 60 min the vacuum reached 50−60 mTorr. The
reaction was held under stirring 180 rpm for 3 h still at 210 °C after
which the reaction vessel was removed from the heat source. The
overhead stirrer was stopped and elevated from the floor of the
reaction vessel. The vacuum was then turned off and the system
purged with N2 gas at atmospheric pressure. The thus formed product
was allowed to cool to ambient temperature. The product was
recovered after carefully breaking the glass with a hammer. Yield =
Preparation of Block Copolymers from 3-GF16-iso and PTT. A
15.3 g sample of PTT, and 46 g of 3-GF16-iso were charged to an
oven-dried 250 mL three necked round-bottom flask equipped with an
overhead stirrer and a distillation condenser kept at 23 °C. The
reaction mass was kept under nitrogen purge. The contents were
degassed once by evacuating down to 150 mTorr and refilling back to
atmospheric pressure with N2 gas. TyzorTPT catalyst (18 mg) was
1
90% of a clear product. H NMR (CDCl3), δ: 8.60 (ArH, s, 1H),
8.15−8.00 (ArH−, m, 2 + 4H), 7.65 (ArH, s, 4H), 6.15 (−CF2−
8248
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
added after the evacuation and repressurization. The nitrogen purge
was then discontinued, and a vacuum ramp was started such that after
about 30 min the vacuum reached a value of about 60 mTorr. The
flask was then immersed into a preheated metal bath set at 250 °C, and
the contents of the flask were allowed to melt and equilibrate for 10
min. Stirring was initiated and the speed was slowly increased to 180
rpm, and the molten contents of the flask was left under stirring for 60
min in the 250 °C bath. The heat source was then removed. The
overhead stirrer was stopped and elevated from the floor of the
reaction vessel. The vacuum was then turned off, and the system
purged with N2 gas. The thus formed product was allowed to cool to
ambient temperature. The product was recovered after carefully
breaking the glass with a hammer. Yield = 95.7% of an opaque
product. 13C NMR (tce-d2), δ: 62.9 (E), 62.7 (D), 62.4 (G), 62.2 (F).
conversion is reached (typically within 16 h). Upon cooling,
the base is filtered off, the solvent evaporated, and the product
purified by distillation and obtained in high purity and yield.
Mechanistically, the hydroxyl aromatic diester, upon
deprotonation, acts as nucleophile attacking the fluorinated
double bond. Subsequent protonation of the carbanion
completes the addition. Halogenated solvents, like carbon
tetrachloride, or carbon tetrabromide may be employed,
allowing formation of a functional halide, i.e., chloride or
bromide, over the protonated product.4,17 In this paper, we will
only discuss the protonated form of the monomer. The product
is fully supported by NMR. The 1H NMR spectra is provided in
Figure 1, for 19F NMR see Supporting Information Figures SI 1
and SI 2. A characteristic tag for the NMR characterization is
the H−F resonance that appears at ∼6.25 ppm with a JH−F
coupling constant of ∼54 Hz.18
The described chemistry is suitable for a variety of
fluorinated vinyl ethers and hydroxyl aromatic diesters. For
example, both dimethyl 2-hydroxyterephthalate and dimethyl 5-
hydroxyisophthalate are suitable nucleophiles, and 1,1,1,2,2,3,3-
heptafluoro-3-(1,2,2-trifluorovinyloxy) propane (PPVE) and
1,1,1,2,2,3,3-heptafluoro-3-(1,1,1,2,3,3-hexafluoro-3-(1,2,2-
trifluorovinyloxy)propan-2-yloxy)propane (PPPVE) were both
employed electrophiles. In general, dimethyl 5-hydroxyisoph-
thalate is preferred, since it provides a symmetric monomer for
subsequent polymerizations.
DISCUSSION
■
Fluorinated Aromatic Ethers. Commercially available
fluorinated monomers for step growth polymerization have
The structures and abbreviations (F10-iso and F16-iso) of two
highlighted monomers are depicted in Figure 2. The integers 10
and 16 simply refer to the number of fluorine atoms, and the
iso refers to the dimethylester configuration on the aromatic
ring. Although the dimethylester form of the monomer was
used in subsequent polymerizations the fluorine side chain is
robust toward several chemical transformations. For example,
treatment of the diester in water/potassium hydroxide readily
hydrolyzes the diester to the diacid and the monomer may be
isolated after protonation. The diacid may be reacted with
oxalyl or thionyl chloride to make the corresponding diacid
chloride. Both these functional groups are suitable for either
polyamide or polyaramid formation, importantly with retained
fluorine functionalities.4
Fluorinated Aromatic Homo- and Random Copolyest-
ers. The fluorinated monomers described above are suitable for
a range of polycondensation protocols. We have been
particularly interested in the use of these in polytrimethylene
terephthalate (PTT) to impart oil repellency. PTT is the
polymer of terephthalic acid (TPA) or dimethylterephthalate
Figure 2. Structure and abbreviation of two fluorinated monomers,
F10-iso (i) and F16-iso (ii).
traditionally been limited to tetrafluoroterephthalic acid. We
have recently demonstrated the synthesis of fluorinated
aromatic ethers via the addition of hydroxyl aromatic diesters
to fluorinated vinyl ethers according to Scheme 1. Similar
reactions with phenols were previously reported by Feiring.16
We found that base-catalyzed conditions (enough to activate
the hydroxyl aromatic diester) could be employed to drive the
reaction. A range of strong bases was initially employed,
including potassium tert butoxide, but it was found that milder
bases, for example, sodium carbonate, could be used equally
effective and with high practical importance. In a typical
reaction the base, fluorinated vinyl ether, and the hydroxyl
aromatic diester are combined in a polar solvent like THF,
brought to reflux and left under stirring until complete
Scheme 2. Diacid (1, Top) and Diester (2, Bottom) Routes to Polytrimethylene Terephthalate (PTT)
8249
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
1
Figure 3. H NMR (CDCl3) of a polymeric fluorinated aromatic ether.
Table 1. Thermal Characteristics of 3-GF10-iso and 3-GF16-
iso Homo- and Copolyesters
mol %
comonomer feed
mol % comonomer
T
T
m b
a
g b
comonomer
obtained
(°C)
(°C)
PTT control
F16-iso
n/a
50
n/a
50
55
23
34
5
229
n/a
n/a
n/a
n/a
F10-iso
50
50
3-GF16-iso
100
100
100
100
3-GF10-iso
23
a
b
1
From H NMR (ratio of terephthalate to isophthalate). From DSC
(10°/min, 2nd heating scan).
(DMT) with 1,3-propanediol (PDO).19 DuPont currently
manufactures bio 1,3-propanediol (bio-PDO) from a fermenta-
tion process using corn based glucose as feedstock. The total
biocontent in PTT is ∼37 wt %. Standard polymerization
protocols (diacid or diester routes) are depicted in Scheme 2
and the diester route will be explained in more detail.20
In the diester route an ester interchange reaction is initially
conducted in which the dimethylester is exchanged for the
glycol, typically employed in ∼2-fold excess relative DMT. For
PTT this means formation of a bishydroxypropyl intermediate
of the corresponding aromatic diester. A typical ester
interchange reaction temperature is ∼200 °C in the presence
of suitable transesterification catalyst, for example titanium(IV)
isopropoxide. The ester interchange leads to condensation of
methanol and can be easily monitored. Following ester
interchange, which is completed within 1−2 h, the reaction
temperature is increased and vacuum applied. Transesterifica-
tion liberates the excess glycol and an increase in molecular
weight is achieved after glycol distillation. A high molecular
weight polymer is typically formed following 3−4 h at adequate
vacuum and continuous agitation. The build of molecular
weight may be monitored by a measurable increase in melt
viscosity. Fluorinated homo and copolyesters were made by this
Figure 4. DSC comparison of PTT (top) and a fluorinated PTT
copolymer containing 50 mol % F16-iso (below).
method in which DMT partially, or fully, was replaced with F10-
iso or F16-iso and polymerized with bio-PDO resulting in 3-
GF10-iso or 3-GF16-iso homo- or/and copolymers. Initial
characterization of the fluorinated polymers were made from
a combination of NMR techniques and size exclusion
1
chromatography (SEC). H NMR and 19F-NMR confirmed
1
the molecular structure, a general H NMR of a polymeric
fluorinated aromatic ether is depicted in Figure 3 (for 19F NMR
see Supporting Information Figures SI 3 and SI 4). From
Figure 3, the resonance at ∼6.15 ppm (JH−F ∼ 54 Hz) is
8250
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
Figure 5. Comparison between copolymers and blends and the impact
on fluorine efficiency.
Figure 6. DSC thermogram of PTT and 3-GF16-iso (50 mol %)
copolymer blend at 87.5/12.5 wt %.
indicative of an intact fluorine side chain. Additional backbone
and end-group structures were similar to that of the PTT
control, i.e., hydroxyl, methyl, and dipropylene glycol ends.
Moreover, the polymer fluorine content corresponded with the
monomer feed ratio. SEC traces are monomodal with Mw/Mn
ratios ∼2.0 indicative of a successful polycondensation
(Supporting Information, Figure SI 5).
Figure 7. IPC chromatograms of calibration series (top, integer relates
mol % F16-iso in copolymer), corresponding elution curve (middle),
and chemical composition calibration curve (below).
Thermal analysis has shown that the thermal characteristic of
PTT is preserved if the fluorinated comonomer concentration
is low (<3 mol %). However, increasing feed ratios of
fluorinated monomer lead to depression of the PTT melting
point, above a critical concentration of comonomers (>25 mol
%) no crystallization is achieved indicating the formation of an
amorphous copolymer. Table 1 summarizes the thermal
characteristics of copolymers with similar levels of comonomer,
and Figure 4 shows a typical DSC trace of PTT compared with
a fluorinated copolymer containing 50 mol % F16-iso.
Surface Properties of Fluorinated Aromatic Homo-
and Copolyesters. The surface properties of fluorinated
homo- and copolymers were investigated using hexadecane as
probe liquid. PTT completely wets hexadecane (i.e., contact
angle <10° or close to 0°) whereas the 3-GF10-iso and 3-GF16-
iso homopolymers repel hexadecane. The highest contact
angles measured were obtained after coating the homopolymers
onto a PTT film substrate. The homopolymer was dissolved in
Figure 8. IPC chromatograms of extruder derived melt blends (50/50
wt %) of PTT with a fluorinated copolymer (50 mol % 3-GF16-iso,
FNMB-91). Curves represent IPC chromatograms from dynamic
extruder sampling at residence times 1−90 min.
8251
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
Scheme 3. “Monomer First” (Top) vs “Oligomer First”
(Bottom) Approach for 3-GF16-iso Based Copolymers
Figure 10. SEC overlay of oligomers and chain extended copolymer.
fluorinated homopolymer was made and subsequently melt
blended with neat PTT. When comparing these two routes, the
blend approach showed an enhancement in hexadecane
repellency compared with the copolymer approach. This can
be illustrated by comparing the use of 1 wt % of F16-iso (1 wt %
total addition, ∼ 4640 ppm fluorine) utilized either in a
copolymer or in a blend (99 wt % PTT + 1 wt % 3-GF16-iso).
The net amount of fluorine is the same. However, blends
provide an improvement of ∼30° when measuring advancing
hexadecane contact angles, as depicted in Figure 5. An initial
clue to the improved repellency was obtained during a design of
experiment (DOE) study performed to elucidate the important
parameters influencing repellency during the melt compound-
ing process. The DOE was performed on a DSM micro-
compounder in which the additive concentration (1 and 5 wt
%), fluorine monomer (F10-iso and F16-iso), mixing speed
(extruder rpm 150 and 245), mixing time (extruder residence
time 5 and 10 min), and temperature (kept constant at T = 260
°C) were the in-going parameters. Interestingly, the main
effects plot for hexadecane advancing contact angle showed that
the extruder parameters (i.e., residence time and extruder
speed) had a minor effect on repellency and that additive
concentration and choice of monomer solely influenced
repellency (Supporting Information, Figure SI 6). The result
from the DOE suggests additive-matrix phase separation.
Subsequent blend characterizations have been able to confirm
this phenomenon and will be outlined in detail below.
Blend Characterization. Thermal DSC analysis is a robust
method for checking compatibility of different polymer
components. Miscible polymer blends follow the Fox equation
and allow the prediction of a single blend glass transition
temperature (Tg) from the overall blend composition and the
Tg of the individual polymer components.21 The presence of
two individual Tg’s for a two component blend is evidence of
immiscibility and phase separation between the two compo-
nents. For our study, blends were made with a sufficiently high
loading of the fluorinated copolymer to have adequate
resolution for the DSC analysis. PTT copolymers based on
F16-iso (50 mol %) were blended with neat PTT (DSM
conditions: T = 260 °C, tmixing = 10 minutes, extruder
rpm=150) to provide final concentrations of 12.5, 25, and 50
wt % of the copolymer. Shown in Figure 6 below is the DSC
thermogram of the blended material at 12.5 wt % incorporation
(25 and 50 wt % identical but with more distinct transitions, see
Supporting Information Figures SI 7 and SI 8). The two Tg’s
are distinct at ∼23 °C and ∼55 °C, indicative of phase
separation between the fluorinated additive and PTT.
Figure 9. Visual difference between random (FNMB-43) and blocky
copolyester (FNMB-105), copolymer composition is the same, i.e. 50
mol % PTT and 50 mol % 3-GF16-iso.
chloroform, the substrate coated, dried, and the resulting
coating was analyzed for its hexadecane repellency. The average
3-GF10-iso and 3-GF16-iso hexadecane contact angles were
∼63° and ∼76° respectively.
Various routes were considered for a built-in fluorine
approach. The first route involved copolymers in which a
minor portion of DMT was replaced with the fluorinated
monomer. The second route utilized blends in which a
8252
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
Table 2. Molecular Weight Information of Oligomers and Chain Extended Copolymers
a
a
a
sample name
1. 3-GF16-iso (FNMB-102)
Mn (g/mol)
Mw (g/mol)
PDI
9100
8500
16600
16100
1.82
1.89
2.01
2. PTT (FNMB-99)
3. Copolymer (FNMB-105, from FNMB-99 and FNMB-102)
59000
118500
a
From SEC.
Figure 11. DSC thermogram (1st heat) and structure of the blocky
polyester, Tg(1) is from the fluorine component and Tg(2) and Tm
from the PTT component.
Table 3. Thermal Characterization Summary
Figure 14. Available dyads in the fluorinated copolycondensate.
sample
Tg (°C)
Tm (°C)
3-GT
3-GF16-iso
55
230
−
−
8
a
copolymer, random
23
a
copolymer, blocky
18, 54
219
a
Composition, 50 mol % 3-GT, 50 mol % 3-GF16-iso.
Figure 15. Different distributions of aliphatic carbon signal in random
(top) vs blocky (below) copolymer composition.
Figure 12. UV (290 nm) IPC traces of random and blocky
copolymers.
SEC in combination with interaction polymer chromatog-
raphy (IPC) was also used to characterize the blended
materials. SEC provides molecular weight distributions by
separating macromolecules by size, IPC complements SEC as it
separates macromolecules based on functionality, independent
of size. Typically, SEC is performed with a negligible enthalpic
contribution, i.e., the column material (stationary phase) is
made neutral to the polymer present in the eluent (mobile
phase) so that only steric (entropic) interactions between the
macromolecule and the stationary phase occur. In IPC,
however, column materials are used which allow for a specific
stationary phase-macromolecule attractive (adsorption) inter-
action, which depends on the chemical structure of the
polymer. In IPC, steric and adsorption interactions compete
with each other and at the so-called critical point of adsorption
(CPA), elution independent of molecular weight is achieved.
Figure 13. Different structural units in the fluorinated copolymer.
8253
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
Table 4. Degree of Randomness/Blockyness (B) for Various Copolymers
sample name
mol % 3-GF16-iso
Mn hard segment (g/mol)
Mn soft segment (g/mol)
condensation time (min)
B-value
FNMB-43 (random, control)
FNMB-105
50
50
50
50
−
8500
−
180
180
90
∼1
0.8
9100
9100
9100
17100
FNMB-109
8500
0.63
0.56
0.56
a
FNMB-126
25800
90
b
FNMB-132
50
8500
30
a
b
PTT bright 1.02 I.V. Using F10-iso as monomer.
chromatograms from the UV absorbance detector (top), the
chemical composition elution curve (middle), and the resulting
IPC calibration curve (below). Notice that PTT does not
exhibit any retention at these conditions and elutes with the
initial solvent band (ca. 2 min retention time).
IPC was now performed on PTT/3-GF16-iso (50 mol %)
melt blends, Figure 8. Only two components are present, the
additive (retention time ∼25 min) and PTT (retention time ∼2
min). These retention times well match the expected values
from the calibration curve, indicative of low levels of
transesterification. Remarkably, the retention profile remained
similar at long extruder residence times (1−90 min). From the
dynamic extruder sampling study a minor peak shift following
90 min was observed suggesting a minor level of trans-
esterification had occurred. One control experiment was
additionally made and this time the melt generated blend was
overlaid with the solution mixture of the same components
(50/50 wt %). As expected these two materials display the exact
same IPC chromatogram as the one viewed in Figure 8.
Additional support for the observed increased hexadecane
repellency has come from surface tension analysis. Using the
mean harmonic method26 (water and methyl iodide as probe
liquids) surface tensions between 12 and 26 dyn/cm have been
measured utilizing 1−5 wt % of either the 3-GF10-iso (50 mol
%) or the 3-GF16-iso (50 mol %) additive in blends with PTT.
As control, the surface tension of PTT was measured at
∼41dyn/cm. Seemingly the reduced surface tension provided
by the fluorinated additives creates a driving force for both
surface migration and phase separation. The overall blend
characterization suggests that PTT can be blended with
fluorinated additives with limited intermolecular interactions
occurring in the melt. Lack of transesterification explains the
observed increase in hexadecane contact angle, since the low
surface tension and higher additive mobility, relative to PTT,
facilitates easier migration of fluorine to the material-air
interface.
Development of Fluorinated Aromatic Blocky Copo-
lyesters. One opportunity arising from the lack of melt
induced transesterification is to control the microstructure of
the fluorinated additive. The hypothesis was that by utilizing
prepolymers/oligomers instead of monomers, with retained
lack of transesterification during the condensation stage, the
fluorinated additive form a blocky copolymer structure. To test
this hypothesis, we performed a direct comparison between a
“monomer first” approach with that of an “oligomer first”
approach utilizing oligomers of PTT and the fluorinated
homopolymer 3-GF16-iso. In both routes the overall
composition was maintained the same, i.e. 50 mol % 3-GT
and 50 mol % 3-GF16-iso. A schematic depicted below (Scheme
3) shows the difference between these two approaches.
Practically, the “monomer first” approach utilized standard
polycondensation techniques as outlined above. However, in
the “oligomer first” approach, the two starting materials were
Figure 16. Advancing hexadecane contact angles comparing a random
melt additive with a blocky, both components containing 50 mol %
F16-iso.
Figure 17. Receding hexadecane contact angles comparing a random
melt additive with a blocky, both components containing 50 mol %
F16-iso.
The CPA depends on the polymer chemical structure and not
on the molecular weight, and continuous change of the mobile
phase composition (gradient elution) provides the tool of
separation.22 To evaluate the effect of chemical composition on
IPC retention, several statistical (random) copolymers of bio-
PDO, DMT and F16-iso at various mol % compositions, i.e., 10
mol %, 25 mol %, 50 mol %, 90 mol %, and 100% F16-iso, were
synthesized. These copolymers as measured by SEC (in
hexafluoroisopropanol), have monomodal molecular weight
distributions and PDI’s ∼2.0 (Supporting Information, Figure
SI 9). IPC was conducted on the same series using a
hydrophobic silica-gel bonded Si(CH2)2C8F17 stationary
phase. The mobile phase was switched from neat HFIP to a
gradient of water and HFIP (running from 60 vol % HFIP to
neat HFIP). Shown in Figure 7 is an overlay of the IPC
8254
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
made individually, subsequently combined, and copolymerized
in the presence of a transesterification catalyst. The interesting
result is that the “oligomer first” approach rendered an opaque
material that appeared elastic. The difference in appearance is
viewed in Figure 9.
seen in Figure 12 this trend is confirmed and the blocky
copolymer elutes noticeably later than the statistical copolymer
of the same chemical composition, i.e. 50 mol % F16-iso.
We used 13C NMR spectroscopy to determine sequence
distributions in these polymers, and thus determine degree of
randomness/blockiness. NMR is a well established method for
studying polymer microstructures. For example sequence
Characterization verified chain extension of the two
oligomers. Figure 10 shows an overlay of the molecular weight
distributions (MWD) of the two starting oligomers (PTT and
3-GF16-iso) and the copolymer, as measured by SEC. The
interesting observation is that no oligomer remains and that
only one molecular weight distribution (of higher molecular
weight) is obtained. Moreover, the copolymer is monomodal
and the PDI is ∼2.0 (Table 2). The expected PDI for a high
molecular weight product of a regular polycondensation
process is 2.0. Thus, chain extension of oligomers does not
alter the behavior of a traditional polycondensation process, i.e.,
“monomer first” type. The two starting oligomers possessed Mn
values close to 9 000 g/mol whereas the measured copolymer
Mn was ∼60 000 g/mol (Table 2). This indicates that the
copolymer is a multiblock of the two starting materials.
Further support for chain extension came from hydroxyl end-
group characterization using 1H NMR (tce-d2). Here the
hydroxyl end-group concentration of the copolymer was
compared with the oligomers. Typically hydroxyl end-group
concentrations between 2 and 5 mol % were found in the
oligomers whereas for the copolymer levels below 0.6 mol %
was measured. These estimates was made by comparing the 1H
NMR signal of the hydroxyl group (∼3.80−3.85 ppm) with the
combined signals of the fluorinated aromatic ether (∼8.6 ppm),
terephthalate (∼8.0 ppm), and diester (4.35−4.65 ppm) group.
It is to point out that the SEC characterization technique is
1
determination using H NMR has been used to study stereo
sequences in polylactides,23 and 13C NMR spectroscopy has
been used to study the sequence distribution in copolyester/
carbonate systems.24 Quantitative 13C NMR was used to
determine the exact composition.
The blockiness index, B, is defined by Devaux,25 op. cit., as
F
12
2 ∑i2=1
B =
F
i
wherein F12 represents the total mole fraction of diads of first
and second repeat units, in either sequence. Fi represents the
mole fraction of structural repeat units (Figure 13). F12, F21, F11,
and F22 are the molar fractions of dyad repeat units in our
copolymer structure, Figure 14. The blockiness index B is 0 for
a diblock copolymer and 1 for a random copolymer.
The designation “G′′ represents the NMR peak of the two
OCH2 carbons when two trimethylene terephthalate moieties
are adjacent to one another; this dyad is designated TT; its
mole fraction is F11. The designation “D” represents the NMR
peak of the two OCH2 carbons when two 3-GF16-iso (or two 3-
GF10-iso) moieties are adjacent to one another; this dyad is
denoted FF; its mole fraction F22. The designation “E” and “F”
represent the two NMR peaks of the two different OCH2
groups in the dyad which contains both a 3-GF16-iso (or 3-
GF10-iso) moiety and a trimethylene terephthalate moiety.
There are two statistically possible arrangements of this dyad,
which are equivalent by NMR, designated FT and TF, with
mole fractions F12 and F21. The relative amount of the TT dyad
is determined by the area of peak G/2, of the FF dyad by the
area of D/2, and of the sum FT and TF dyads by the area of (E
+ F)/2. These dyad amounts can be normalized to 100% to
give the mole fraction of each type of dyad. Each of the dyad
mole fractions is thus determined as follows:
1
preferred, over H NMR, for high molecular weight polymers
due to the relatively low end-group concentration in these
systems.
Additional characterization of the chain extended copolymer
was obtained by thermal DSC analysis (Figure 11). Results
indicate that the thermal characteristics of the two starting
components are retained in the copolymer, i.e. the Tg from the
fluorinated component is observed at ∼18 °C and Tg and Tm
from the PTT component are observed at 54 and 219 °C,
respectively. The Tg of the fluorinated component (∼18 °C) is
about 10 °C higher compared to the homo 3-GF16-iso oligomer
(Tg∼8 °C). The increase in Tg is ascribed to the fact that the
chain ends now are incorporated into the copolymer and that
this restriction in mobility leads to a slightly higher onset in
translational mobility. Similarly, the Tm of the PTT component
is ∼10 °C lower compared to the neat polymer, due to the
incorporation of the softer fluorinated component. As already
described, the fully amorphous or random/statistical copolymer
only shows a single Tg ∼ 23 °C (Table 1). Thus, overall
findings suggest a blocky type of copolymer. In fact, the thermal
characteristics are typical of thermoplastic elastomers (TPE’s),
which may explain the observed elastic properties of the
resulting material. In Table 3 are summarized the thermal
characteristics of the homo- and copolymers, either random or
blocky.
X
∫
i
F =
i
∑
X
j
∫
j=1−4
In a random copolymer the statistical ratio of the dyad is
1:2:1 for TT:TF + FT:FF. In this case the areas of peaks D, E,
F, G will be 1:1:1:1. Interestingly, the distribution of the
aliphatic carbon signals G, D, E, and F are different depending
on the composition of the copolymer as viewed in Figure 15.
For a random copolymer the four signals, as expected, are equal
in intensity whereas for a blocky copolymer signals D and G are
enhanced relative to signals E and F. This information, in
conjunction with the molar composition, can now be used to
determine the degree of randomness/blockiness.
To demonstrate the ability to tune the copolymer blockiness
the obtained B-values for a few selective copolymer
compositions are summarized in Table 4. One interesting
observation is that the overall condensation time impacts the
final B-value. In the initial reaction (FNMB-105) the
condensation time was kept deliberately long (180 min), yet
the copolymer retains a blocky structure. By lowering the
condensation time (FNMB 109) it appears as if the degree of
We also analyzed the copolymer microstructure. We
compared the chain extended copolymer with the random
copolymer using IPC. As outlined above, IPC separates
polymers based on their microstructure and we have recently
shown that blockiness increases IPC retention, i.e. multiblock
copolymers always elute later than the corresponding statistical
counterpart with the same chemical composition.22 As can be
8255
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256

Macromolecules
Article
blockiness increases, i.e., lower B-value. Another route to an
increased blockiness is to utilize a higher molecular weight PTT
hard segment. Sample FNMB-126 utilized a commercial grade
PTT (bright, 1.02 IV) as starting material with a higher
molecular weight. Blocky copolycondensates were also achieved
utilizing the F10-iso monomer, FNMB-132 in Table 4.
Importantly the B-value for the random copolymer control
(FNMB-43) was ∼1. Likely other factors may influence the B-
value, i.e. soft segment length and hard/soft segment ratio.
Further aspects however will not be covered since the primary
focus, of this paper, has been to demonstrate the initial proof of
principle.
REFERENCES
■
(1) Hougham, G.; Cassidy, P. E.; Johns, K.; Davidson, T.
Fluoropolymers 2, Properties; Kluwer Academic: Dordrecht, The
Netherlands, ISBN: 0-306-46061-0
(2) Drysdale, E. N.; Mahajan, S.; Moloy, K. G.; Nederberg, F.;
Pollino, J.; Ritter, J. C. Fluoroether functionalized aromatic diesters and
derivatives thereof, 2011, WO2011028767.
(3) Bunn, C. W.; Howells, E. R. Nature 1954, 549.
(4) Drysdale, E. N.; Mahajan, S.; Moloy, K. G.; Nederberg, F.;
Pollino, J.; Ritter, J. C. Synthesis of polyesters comprising fluorovinylether
functionalized aromatic moities for use as soil resistant fibers, 2011,
WO2011028778.
(5) Drysdale, E. N.; Mahajan, S.; Moloy, K. G.; Nederberg, F.;
Pollino, J.; Ritter, J. C. Polyester films with improved oil repellency, 2011,
WO2011028771
(6) Drysdale, E. N.; Moloy, K. G.; Nederberg, F.; Pollino, J.; Ritter, J.
C. Aramid comprising fluorovinylether functionalized moities, 2011,
WO2011028791
(7) Carothers, W. H. J. Am. Chem. Soc. 1929, 51, 2548.
(8) Carothers, W. H. Chem. Rev. 1931, 8, 353.
(9) Flory, P. J. Principles of Polymer Chemistry; Cornell University
Press: Ithaca, NY, 1953.
(10) (a) Yu, J.; Li, B.; Lee, S.; Ree, M. J. Appl. Polym. Sci. 1999, 73,
1191. (b) Li, B.; Yu., J.; Lee, S.; Ree, M. Polymer 1999, 40, 5371.
(c) Ubach, J.; Ilarduya, A. M.; Quintana, R.; Alla, A.; Rude, E.; Munoz-
Guerra, S. J. Appl. Polym. Sci. 2010, 115, 1823.
Repellency of Blocky Copolyesters. The blocky
polyester (FNMB-105, Mn ∼ 59000 g/mol) was melt blended
with neat PTT bright (Mn ∼ 25800 g/mol, I.V. 1.02 mL/g) and
evaluated for its hexadecane repellency. A comparison was
made with a melt blend from an amorphous melt additive
(FNMB-43, Mn ∼ 72900 g/mol) of identical molar
composition, i.e., 50 mol % PTT and 50 mol % 3-GF16-iso.
The melt blends were prepared in a DSM micro compounder
outlined above and extruded into ribbons (1 in. wide). The
blocky melt additive seems to outperform the random additive
when comparing advancing hexadecane contact angles, Figure
16. Moreover, receding hexadecane contact angles (Figure 17)
are consistently higher for the blocky additive. In general,
advancing contact angles represent the materials low surface
energy component, whereas, the receding contact angle
(11) (a) Shyr, T.-W.; Lo, C.-M.; Ye, S.-R. Polymer 2005, 5284.
(b) Wei, G.; Wang, L.; Chen, G.; Gu, L. J. Appl. Polym. Sci. 2006, 100,
1511. (c) Zou, H.; Li, G.; Jianming, J.; Shenglin, Y. Polym. Eng. Sci.
2008, 48 (3), 511. (d) Castellano, M.; Marsano, E.; Turturro, A.;
Canetti, M. J. Appl. Polym. Sci. 2011, 122 (1), 698.
(12) http://www2.dupont.com/Plastics/en_US/Products/Hytrel/
Hytrel.html
(13) Tonelli, C.; Trombetta, T.; Scicchitano, M.; Simeone, G.;
Ajroldi, G. J. Appl. Polym. Sci. 1996, 59, 311.
represents the materials high surface energy component.26
A
smaller contact angle hysteresis (difference between advancing
and receding) is correlated with an improved surface
homogeneity. Seemingly, our data suggests that the blocky
additive provides a more homogeneous fluorinated surface as
compared with the random additive.
In summary, melt-derived polycondensation was utilized to
prepare block copolymers of two seemingly immiscible
polymers. The forward condensation reaction is active although
transesterification during normal melt blending is limited.
Blockiness was found to be retained even at 180 min in the
melt, under vacuum, and in the presence of a transesterification
catalyst.27 Melt additives based on blocky copolymers were
shown to be more fluorine efficient compared to random
copolymers and demonstrate an improvement in hexadecane
repellency; in addition, they reduce the contact angle hysteresis.
(14) Iwashita, K.-I.; Yokoyama, A.; Yokozawa, T. J. Polym. Sci., Part
A: Polym. Chem. 2005, 43, 4109.
(15) Yokozawa, T.; Yokoyama, A. Prog. Polym. Sci. 2007, 147.
(16) Feiring, A. A.; Wonchoba, E. R. J. Org. Chem. 1992, 57, 7014.
(17) Koch, H. F.; Kielbania, A. J. J. Am. Chem. Soc. 1970, 92 (3), 729.
(18) Ihrig, A. M.; Smith, S. L. J. Am. Chem. Soc. 1972, 94 (1), 34.
(19) Bhatia, S. K.; Kurian, J. Biotechnol. Lett. 2008, 30, 619.
(20) Scheirs, J.; Long, T. E. Modern Polyesters: Chemistry and
Technology of Polyesters and Copolyesters; Wiley: New York, ISBN
9780471498568, 2003.
(21) (a) Fox, T. G. Bull. Am. Phys. Soc. 1956, 1, 123. (b) Gedde, U.
Polymer physics; Kluwer Academic Publishers: Dordrecht, The
Netherlands, ISBN 0-412-62649-3, 1999.
(22) (a) Brun, Y. J. Liq. Chrom. Rel. Technol. 1999, 22 (20), 3027.
(b) Brun, Y. J. Liq.Chrom. Rel. Technol. 1999, 22 (20), 3067. (c) Brun,
Y.; Alden, P. J. Chrom. A 2002, 966, 25. (d) Chang, T. J. Polym. Sci.,
Part B: Polym. Phys. 2005, 43, 1591. (e) Brun, Y.; Foster, P. J. Sep. Sci.
2010, 33, 3501.
(23) (a) Kricheldorf, H. R.; Boettcher, C.; Tonnes, K.-U. Polymer
1992, 33, 2817. (b) Kasperczyk, J. E. Macromolecules 1995, 28, 3937.
(c) Zhang, L.; Nederberg, F.; Messman, J. M.; Pratt, R. C.; Hedrick, J.
L.; Wadem, C. G. J. Am. Chem. Soc. 2007, 129, 12610.
(24) (a) Pego, A. P.; Zhong, Z.; Dijkstra, P. J.; Grijpma, D. W.;
Feijen, J. Biomacromolecules 2003, 204, 747. (b) Kricheldorf, H. R.;
Stricker, A. Macromol. Chem. Phys. 1999, 200, 1726. (c) Kricheldorf,
H. R. Macromol. Chem. Phys. 1979, 2133.
ASSOCIATED CONTENT
■
S
* Supporting Information
Various analytical results including 19F NMR data, SEC data,
DSC data, and main effect plots. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: fredrik.nederberg@usa.dupont.com.
Notes
(25) Devaux, J.; Godard, P.; Mercier, J. P. J. Pol. Sci. Pol. Phys. 1982,
20, 1875.
The authors declare no competing financial interest.
(26) Wu, S. Polymer Interface and Adhesion; ISBN 0-8247-1533-0;
1982.
ACKNOWLEDGMENTS
■
(27) Note: General concept transferred into blocky copolyesters
based on PET and PBT.
The authors would like to acknowledge J. Michael Barker,
Yvonne B. Bethel, Matthew A. Page, and Wayne E. Marsh for
support in this work.
8256
dx.doi.org/10.1021/ma3011075 | Macromolecules 2012, 45, 8245−8256