Segmented Block Copolymers with Monodisperse Hard Segments: The Inﬂuence of H-Bonding on Various Properties

The properties of segmented-copolymer-based H-bonding and non-H-bonding crystallisable segments and poly(tetramethylene oxide) segments were studied. The crystallisable segments were monodisperse in length and the non-hydrogen-bonding segments were made of tetraamidepiperazineterephthalamide (TPTPT). The polymers were characterised by DSC, FT-IR, SAXS and DMTA. The mechanical properties were studied by tensile, compression set and tensile set measurements. The TPTPT segmented copolymers displayed low glass transition temperatures (Tg, −70 °C), good low-temperature properties, moderate moduli (G′ ≈ 10–33 MPa) and high melting temperatures (185–220 °C). However, as compared to H-bonded segments, both the modulus and the yield stress were relatively low.


Introduction
Hydrogen bonding has a strong influence on the physical and chemical properties of polymers, such as the melting temperature, solubility and adhesion. [1]Hydrogen bonds are relatively strong secondary forces that increase the chain interaction in both the crystalline and amorphous states.However, hydrogen bonding is not a necessity for crystallisation.For example, polyethylene does not display hydrogen bonding or strong secondary forces but still crystallises easily and to a high degree. [2]This is a result of the polyethylene having a perfectly regular structure, being flexible and demonstrating a tight packing.Nevertheless, hydrogen-bonding plays an important role with regard to numerous polymer properties, and thus, most nylons have higher melting temperatures, yield and fracture stresses as compared to polyethylene. [3][6][7][8][9] In these studies, the urethane segments were monodisperse in length and the repeat length of the alkyl unit was varied from 2 to 4. The polyurethanes containing piperazine exhibited an increased thermal stability as it did not show The properties of segmented-copolymer-based H-bonding and non-H-bonding crystallisable segments and poly(tetramethylene oxide) segments were studied.The crystallisable segments were monodisperse in length and the non-hydrogen-bonding segments were made of tetraamidepiperazineterephthalamide (TPTPT).The polymers were characterised by DSC, FT-IR, SAXS and DMTA.The mechanical properties were studied by tensile, compression set and tensile set measurements.The TPTPT segmented copolymers displayed low glass transition temperatures (T g , À70 8C), good low-temperature properties, moderate moduli (G 0 % 10-33 MPa) and high melting temperatures (185-220 8C).However, as compared to H-bonded segments, both the modulus and the yield stress were relatively low.
trans-urethanisation. As compared to hydrogen-bonded polyurethanes, the piperazine polyurethanes have a glass transition at low temperature (À50 8C), only slightly temperature-dependent rubber modulus, high fracture strains but also high tensile set (TS) values.Such non-Hbonding polyurethanes with monodisperse hard segments show very interesting properties.
As for the polyurethanes discussed above, non-hydrogen-bonding amides can be prepared using the secondary piperazine amine. [17]By choosing to utilize piperazine instead of ethylene diamine, the melting temperature was reduced by 100 8C and the amide chain mobility was found to occur at a lower temperature.The piperazine unit has a non-planar, alicyclic structure that can present either a chair or boat conformation.
The present paper discusses the synthesis and properties of segmented copolymers based on PTMO and tetraamidepiperazineterephthalamide (TPTPT) units of uniform length (Figure 1a).This TPTPT segment had tertiary amide groups and was therefore unable to form hydrogen bonds.The properties of the resulting PTMO x -TPTPT copolymer without H-bonding were compared to those of segmented PTMO x -T6T6T copolymers but with hydrogen bonding.Such a direct comparison between hydrogen-bonding and non-hydrogen-bonding amide systems provided insight into the influence of H-bonding in segmented block copolymers on the properties.

Synthesis of 1-[4-(piperazine-1ylcarbonyl)benzoyl]piperazine
In a first step 1-[4-(piperazine-1-ylcarbonyl)benzoyl]piperazine was synthesised.Piperazine (120 g, 1.4 mol) was melted at 140 8C in a round-bottomed flask with a nitrogen inlet, mechanical stirrer and a reflux condenser, and then reacted with DPT (39 g, 0.12 mol) for 16 h at 140 8C.The product was a transparent liquid that partly solidified during the reaction.Two litres of methanol were added to the reaction mixture to dissolve the product and part of the methanol was subsequently distilled off until 0.5 l of the solution remained.Upon addition of ether (1 l) to the solution, a white product precipitated.This product, denoted PTP, was collected by filtration over a no. 4 glass filter.The yield of the reaction was 67%, and the product had a melting peak, as measured by DSC, of 198 8C with a melting enthalpy of 70 J Á g À1 .The purity of the product, estimated from NMR spectra, was found to be >95%.

Synthesis of Tetraamidepiperazineterephthalamidedimethyl
In a second step, the PTP units were reacted with MPT to give TPTPTdimethyl.In a round-bottomed flask, PTP-piperazine (85 g, 0.28 mol) was dissolved in trichlorobenzene at 120 8C.To this solution, MPT (200 g, 0.8 mol) was added and the reaction was carried out at 140 8C for 16 h under a nitrogen flow.After one hour, a white product started to precipitate from the solution.The product was collected by filtration of the hot solution over a heated no.4 glass filter and washed three times with acetone.TPTPT-dimethyl had a melting temperature of 250 8C and a melting enthalpy of 40 J Á g À1 , as determined by DSC.The purity estimated from NMR spectra was found to be >95%.

H NMR
NMR spectra were recorded with a Bruker AC 300 spectrometer at 300 MHz.Deuterated trifluoroacetic acid (TFA-d) was used as the solvent.

Differential Scanning Calorimetry
DSC thermograms were recorded on a Perkin-Elmer DSC apparatus, equipped with a PE7700 computer and a TAS-7 software.Dried samples of 5-10 mg were heated to approximately 30 8C above their melting temperature, subsequently cooled and heated again.The heating and cooling rates were both 20 8C Á min À1 .The crystallisation temperature (T c ) was taken as the temperature location of the maximum of the crystallisation peak in the first cooling scan.The temperature location of the maximum of the melting peak in the second heating scan was taken as the melting temperature and also the melting enthalpy could be determined from this peak.The degree of crystallinity was calculated from the melting enthalpy of the polymer and the melting enthalpy of the bisester tetraamide according to:

Fourier-Transform Infrared (FT-IR) Spectroscopy
Infrared transmission spectra were recorded using a Nicolet 20SXB FT-IR spectrometer with a resolution of 4 cm À1 .The samples to be measured were prepared by adding a drop of the polymer dissolved in HFIP (1 g Á l À1 ) onto a pressed KBr pellet.Temperature-dependent FT-IR spectra were recorded between 25 and 210 8C.

Atomic Force Microscopy (AFM)
AFM measurements were recorded on a Nanoscope IV controller (Veeco) operating in tapping mode.The AFM was equipped with a Jscanner with a maximum size of 200 mm 2 .Si-cantilevers (Veeco) were used to obtain height and phase images.The amplitude in free oscillation was 5.0 V.The operating set-point value (A/A 0 ) was set to relatively low values of 0.70, the size of the scans was 1 Â 1 mm 2 .Solvent-cast samples with thicknesses of %20 mm were prepared on a silicon wafer from a 3 wt.-%solution in trifluoroacetic acid (TFA).

Synchrotron SAXS
Small angle X-ray scattering (SAXS) experiments were performed on the Dutch-Belgium (DUBBLE) beamline BM26 at the European Synchrotron Radiation Facility (ESRF) in Grenoble, France.The wavelength of the beam was 1.2 A ˚.A two-dimensional SAXS detector was used and the range for the scattering vector (q) was 0-1.8 nm À1 .Temperature-dependent profiles were recorded with a Linkam remote-controlled DSC stage at a heating and cooling rate of 10 8C Á min À1 .The background was subtracted from the intensity.The two dimensional SAXS intensity was azimuthally integrated to obtain the scattering pattern as a function of q ¼ 2 sin u/l and the long spacing (L, in nm) was calculated from

Dynamic Mechanical Thermal Analysis (DMTA)
Samples (70 Â 9 Â 2 mm 3 ) for DMTA were prepared using an Arburg H manual injection moulding machine with a barrel temperature that was set 50 8C above the melting temperature of the copolymer.The mould was kept at room temperature.The test samples were dried in vacuo at 50 8C for 24 h before use.DMTA thermograms were recorded on a Myrenne ATM3 torsion pendulum at a frequency of 1 Hz and 0.1% strain.The storage (G 0 ) and loss (G 00 ) moduli were measured as functions of temperature.The samples were cooled to À100 8C and subsequently heated at a rate of 1 8C Á min À1 .The temperature location of the maximum of the loss modulus peak was taken as the glass transition temperature.The start of the rubber plateau was denoted the flex temperature (T flex ), and the flow temperature (T flow ) was defined as the temperature where the storage modulus reached 1 MPa.The stability of the rubber plateau can be described according to the following expression: Here, DT represents the temperature range: DT ¼ (T flow -50 8C) -T flex .

Tensile Testing
Stress/strain curves were obtained on injection-moulded, 2.2 mm thick dumbbells (ISO37 type 2), using a Zwick Z020 universal tensile machine equipped with a 500 N load cell.The strain was measured with extensometers.The tensile tests were carried out at an initial strain rate of 0.4 s À1 (test speed of 60 mm Á min À1 ).For test temperatures other than room temperature, a temperaturecontrolled environment chamber was used.The E-moduli at each temperature were determined in eight-fold at strains from 0.1-0.25%.The standard deviation of the modulus was 5-8%.Also measured were the yield stress (s y ), the yield strain (e y ), the fracture stress (s b ), the fracture strain (e b ) and the true fracture stress (s true ).The true fracture stress was obtained by multiplying s b by the straining factor [¼1 þ (e b /100)].For each test three samples were used and the average taken.

Compression Set
Samples for CS experiments were cut from injection moulded bars and investigated at room temperature according to the ASTM 395 B standard.A compression was applied and maintained for 24 h at room temperature before being released.After a relaxation period of half an hour, the thickness of the samples was measured.The CS, defined according to Equation (3), was taken as the average of three measurements.

Compression set
In the above expression, d 0 is the thickness before compression (mm), d 1 is the compressed thickness (mm) and d 2 is the thickness 30 min after releasing the compression (mm).

Tensile Set
Cyclic stress/strain experiments were conducted on the injectionmoulded bars cut into dumbbells (ISO 37 type 2).A Zwick Z020 universal tensile machine equipped with a 500 N load cell was used to measure the stress as a function of the strain of each loading and unloading cycle at a strain rate of 60 mm Á min À1 .The strain of each loading-unloading cycle was increased (stair-case loading) and the TS of the strain increment was determined as a function of the applied strain.The incremental TS was calculated from the following relation: where e r, cycle(i) is the remaining strain at the end of cycle i and e r, cycle(i -1) is the remaining strain at the end of the preceding cycle i -1.Directly after the stress was zero, a new cycle was started and for each following cycle, the strain was increased by 20%.

Results and Discussion
Segmented copolymers based on PTMO and monodisperse tetraamidediester units were synthesised (Figure 1).The TPTPT was unable to form hydrogen bonds, while T6T6T was capable of H-bonding.The starting material for the tetraamidediester was either a TPTPT-dimethyl (T m 250 8C, DH m 40 J Á g À1 ) or a hexamethylenediamineterephthalamide (T6T6T-dimethyl) (T m 303, DH m 150 J Á g À1 ). [18]The TPTPT-dimethyl unit compared to T6T6T-dimethyl had a lower melting temperature and a lower heat of fusion.This was a result of two opposing effects, the flexible hexamethylene group being replaced by the cyclic piperazine group and the H-bonding in T6T6T was absent in TPTPT.The low heat of fusion was believed to be due to difficulties in obtaining close-packing caused by the nonplanar piperazine group and mixtures of the chain-boat conformations.The segmented copolymers were made up of PTMO with an M n of either 1 000 or 2 000 g Á mol À1 and denoted PTMO 1 000 and PTMO 2 000 .

PTMO x , TPTPT, T6T6T Segmented Copolymers
The PTMO x -tetra-amide segmented copolymers were prepared in a solution/melt polymerization process.The synthesised copolymers were transparent in the melt and thus melt phasing did not occur.Upon cooling, the polymers were found to be tough, elastic and transparent materials with high inherent viscosities (Table 1).

FT-IR Spectroscopy
The FT-IR spectrum of PTMO 1 000 -TPTPT was very similar to that of TPTPT-dimethyl, apart from the 1 100 cm À1 band of the PTMO (Figure 2).The copolymers displayed the carbonyl bands of esters and amides in the wave length region of 1 600-1 750 cm À1 .For PTMO 1 000 -TPTPT, two peaks could be seen: one at 1 630 cm À1 assigned to the amide carbonyl next to the piperazine ring and a second peak at 1 730 cm À1 attributed to the ester carbonyl group.These peak positions were similar to those of the TPTPT-dimethyl starting material.
The amide carbonyl peak was stronger than that of the ester carbonyl as the ratio of amide to ester was 4:2.Since piperazine is a tertiary amide without a proton at the amide, the (NÀH) amide bands at 3 300 and 1 540 cm À1 were not observed.Upon increasing the temperature of PTMO 1 000 -TPTPT, only small changes in the spectrum were observed and it could be concluded that neither the ester nor amide carbonyl in this copolymer was sensitive to crystalline order.PTMO 1 000 -T6T6T displayed a crystalline amide carbonyl peak at 1 625 cm À1 , an amorphous amide carbonyl peak at 1 670 cm À1 and an ester carbonyl peak at 1 720 cm À1 . [14]pon heating the PTMO 1 000 -T6T6T material, the crystalline amide peak at 1 625 cm À1 decreased in size while the size of the amorphous amide carbonyl peak at 1 670 cm À1 increased.From this change, the T6T6T crystallinity could be determined and was found to be very high (92%). [14]In the FT-IR, the band for the ether group (CÀOÀC stretching) at PTMO 1 110 cm À1 and for the methylene groups of hexamethylene diamine and the polyether were at 1 437, 1 283 and 1 254 cm À1 .

DSC
Also the melting and crystallisation behaviour of the PTMO x -tetra-amide copolymers were studied by DSC.For PTMO 2 000 -TPTPT, no melting or crystallisation transition could be observed.The heating and cooling thermograms for PTMO 1 000 -TPTPT and PTMO 1 000 -T6T6T can be seen in Figure 3.
The cooling curves of both copolymers displayed a small peak next to the main crystallisation peak.A second transition was also observed in the second heating scan of PTMO 1 000 -TPTPT.[14][15] The melting and crystallisation temperature for PTMO 1 000 -TPTPT was approximately 15 8C lower than for PTMO 1 000 -T6T6T (Table 1), and the T m -T c values were very low for both systems.The PTMO 1 000 -TPTPT copolymer demonstrated low melting and crystallisation enthalpies of about 10 J Á g À1 .
The melting enthalpy of the TPTPT-dimethyl unit was 40 J Á g À1 and this value was used to calculate the crystallinity of the TPTPT segment in the copolymer according to Equation (1).The obtained crystallinity was found to be high despite the crystalline packing that might not be all that neat.The crystallinities of PTMO 1 000 -T6T6T and PTMO 2 000 -T6T6T were calculated in a similar fashion and were also found to be high.Moreover, these crystallinities

Sample
Tetraamide corresponded well with the crystallinities as determined by FT-IR.Thus, the TPTPT units in the copolymer displayed lower melting temperatures and lower values of heat of fusion than their T6T6T counterparts.Nonetheless, the crystallinity of the TPTPT segment in the copolymer remained high.

AFM
The morphology of the PTMO 2 000 -TPTPT and PTMO 2 000 -T6T6T copolymers was studied by AFM on cast films.In phase angle mode, white ribbons were observed and these corresponded to the TPTPT and T6T6T crystallites (Figure 4).The lengths of the ribbons were determined to be several hundreds of nanometres.The extended length of the TPTPT and T6T6T units were 2.8 and 3.6 nm, respectively, with a chain direction of the units perpendicular to the ribbon length.It is clear from Figure 4 that the TPTPT ribbons were shorter than their crystalline T6T6T counterparts, and the aspect ratio of the TPTPT ribbons was thus expected to be lower than for T6T6T.

SAXS
The average repeat distance of crystalline segments, the so-called long spacing, can be obtained from SAXS measurements.As was demonstrated above, the copolymers displayed crystallites with a ribbon-like structure.These ribbons had three dimensions: the ribbon length, thickness and width.The ribbon length was a few hundred nanometres and the ribbon thickness, i.e., the extended length of the tetra-amide, was approximately 2.8-3.6 nm.The third dimension, the width of the ribbons, is known to vary with the crystallization conditions.At room temperature, the PTMO 1 000 -TPTPT copolymer had a long spacing of 12.0 nm and PTMO 1 000 -T6T6T had one of 15.3 nm.This suggests that the width of the T6T6T ribbons (i.e., the third direction) was somewhat wider in these samples which indicate that the crystallisation in the width direction occur faster for T6T6T than for TPTPT.Upon heating of semi-crystalline polymers, a gradual melting of the crystallites takes place and the long spacing increased steadily with temperature. [14]The long spacing of both PTMO 1 000 -TPTPT and PTMO 1 000 -T6T6T was studied as a function of temperature in a thermal cycle to temperatures above the melting temperature of the materials (Figure 5).The long spacing of PTMO 1 000 -TPTPT was found to be constant up to near-melting temperatures and PTMO 1 000 -T6T6T demonstrated a similar behaviour.
Upon cooling, the long spacing decreased and this was more gradual than on melting.The fully crystalline state was obtained after some time.This hysteresis in the long spacing was stronger for the TPTPT segments than for the T6T6T segments, indicating that the T6T6T segments crystallised faster.

DMTA
The shear storage and loss moduli of the PTMO x -TPTPT copolymers as functions of temperature are presented in Figure 6 and the DMTA data of all the copolymers is presented in Table 2.
Segmented Block Copolymers with Monodisperse Hard Segments: . . .Two transitions could be observed, a glass transition near À70 8C and melting around 200 8C.The glass transition temperatures were extremely low, suggesting a very small content of TPTPT in the PTMO phase.The glass transition temperatures for TPTPT seemed to be even lower than for T6T6T.This difference was also observed in the piperazinebased polyurethanes.For PTMO 2 000 , a shoulder on the peak representing the glass transition was observed at À10 8C and was believed to be caused by the melting of the PTMO 2 000 .[15][16]19] The start of the rubbery plateau (T flex ) was very low for PTMO 1 000, and was not influenced by the crystallisation of PTMO.The rubber modulus in the plateau region was only little temperature dependent and the melting transition was sharp.Such behaviour is typical for copolymers with crystallisable segments of uniform length.
However, the modulus of the TPTPT copolymers at room temperature, was three times lower than the modulus of their T6T6T counterparts (Table 2).Possible explanations for these lower moduli include a lower crystallinity, a lower aspect ratio of the crystallites and/or less stiff TPTPT crystallites.The crystallinity of the TPTPT segments was determined by DSC not to be lower; however, the heat of fusion value was considerably lower (Table 1).This difference in heat of fusion suggests a less dense packing of the crystallites.The AFM analysis indicated that the TPTPT crystallites had a smaller aspect ratio (Figure 4).
The flow temperatures of the copolymers, obtained by DMTA, corresponded very well with the melting temperatures as measured by DSC (Table 1).The melting temperature of the TPTPT segmented copolymers were, as compared to the T6T6T copolymers, influenced by two opposing factors: the absence of hydrogen bonding and the increased stiffness of the piperazine units as compared to the hexamethylene units.The melting temperature of TPTPT was thus 15-35 8C lower.

Tensile Properties
The stress/strain curves of the segmented block copolymers with and without hydrogen bonding are shown in Figure 7.No necking was observed during the tensile measurement.
The E-modulus for TPTPT segmented polymers was much lower than for their T6T6T counterparts (Table 3).Moreover, the yield stress was a factor of two lower for the TPTPT copolymers than for the T6T6T materials.Thus, the stress necessary to deform the non-hydrogen-bonding TPTPT crystallites was lower than for the H-bonded T6T6T.The underlying reason was the absence of hydrogen bonding   and the poorer crystalline packing.The yield strain, on the other hand, was found to be approximately the same for the non-hydrogen-bonding segmented copolymers as for those with H-bonds.
After the yield point, strain hardening of the PTMO phase occurred, thus increasing the strength of both copolymers.The fracture stresses of the TPTPT polymers were lower than for their T6T6T counterparts, however, the fracture strains were higher.The fracture stress and strain can be combined into a single parameter known as the true fracture stress s true .Surprisingly, the values of true fracture stress increased with the PTMO length (i.e., with a decreasing amide content) (Table 3).Moreover, the true fracture stresses were similar for the TPTPT and the T6T6T copolymers.This suggests that the fracture properties were more dependent on the polyether phase than on the content or aspect ratio of the crystallized amide segments.The higher true fracture stresses for the PTMO 2 000 compared to PTMO 1 000 must have been due to strain hardening occurring more readily for this segment.

Compression Set
A standard means of investigating the elastic behaviour of segmented copolymers is by CS experiments.The CS values of the copolymers in the present study were found to increase with decreasing PTMO length (Table 2).The CS values were somewhat higher for PTMO x -TPTPT as opposed to for PTMO x -T6T6T, and these higher CS values must have been due to more easily deformable crystallites caused by the absence of hydrogen bonding as well as by the inferior packing of TPTPT in the crystalline state.A general trend in copolymers is that the CS values increase with increasing modulus and both the PTMO x -TPTPT and PTMO x -T6T6T copolymers complied with this behaviour (Figure 8).However, the CS-values in the G''/CS graph for the TPTPT copolymers were particularly high as the TPTPT copolymers displayed both lower moduli and higher CS values.

Tensile Set
Another way to study the elastic properties of polymers is by TS experiments; the TS as a function of strain is particularly informative.In these experiments, the strain of each loading-unloading cycle was increased (stair-case loading) and the TS of the strain increment was determined as a function of the applied strain.All of the PTMO x -TPTPT and PTMO x -T6T6T copolymers were studied and the results are presented in Figure 9.The TS curves for TPTPT showed Segmented Block Copolymers with Monodisperse Hard Segments: . . .
the same trend as those for T6T6T.With PTMO 1 000 , the TS values increased up to a strain of about 100% and then flattened out.The curves for the TPTPT materials displayed lower TS values than their T6T6T counterparts.With PTMO 2 000 , the TS increased up to a strain of about 350% before reaching a plateau.The reason why the TS values were ultimately higher for the PTMO 2 000 segments as opposed to PTMO 1 000 was the strain crystallisation of the former.Also in this case the TPTPT materials display lower values than their T6T6T counterparts.The TPTPT copolymers were thus more elastic, however at a lower modulus of the material (Table 2).The TS at 50% strain (TS 50% ) is a typical value.[14] For PTMO x -T6T6T, the TS 50% was in accordance with the results from the PTMO x -T6A6T series, [14] and the TS 50% as a function of the modulus displayed higher values for the PTMO x -TPTPT materials that were thus found to be less elastic at a particular stiffness (Figure 10).

Conclusion
Segmented PTMO x -TPTPT copolymers with non-hydrogenbonding monodisperse crystallisable TPTPT segments were studied.The polymerisation was carried out with a TPTPTdimethyl unit that was synthesised prior to the polycondensation reaction to ensure the uniformity of the segment.This TPTPT-dimethyl unit had a particularly low heat of melting, suggesting a not so tight crystalline packing.This inferior close-packing might be due to the non-planar structure and a mixture of chain/boat conformations of the piperazine unit.The PTMO x -TPTPT copolymers could be melt-processed and the molecular weight of the samples was high.The PTMO x -TPTPT copolymers displayed a melting temperature of about 200 8C as well as a low melting enthalpy.Despite this low melting enthalpy, the crystallinity of the TPTPT segments in the copolymer was high.The TPTPT segments in the copolymers formed nanoribbons but the aspect ratio of these ribbons did not seem to be very high.The SAXS pattern indicated a well-crystallised morphology.
The PTMO x -TPTPT segmented copolymers displayed a low glass transition temperature (À70 8C), suggesting an almost complete crystallisation of the monodisperse amide segments.The copolymers had a temperature-independent rubber modulus in the rubber region, as often observed with monodisperse crystallisable segments, but the moduli of the copolymers were lower than those of the PTMO x -T6T6T copolymers.Also the yield stresses were low for the materials with TPTPT segments.The change in yield stress followed the change in the modulus, and these low modulus and yield stress values for the PTMO x -TPTPT copolymers were thought to be a result of the lower reinforcing effect of the crystallites arising from the absence of hydrogen bonding, a poorer crystalline packing and the lower aspect ratio of the crystallites.
The fracture stresses for the TPTPT copolymers were lower than for the T6T6T copolymer; however, the true fracture stresses were equivalent.However, these true fracture stresses were dominated by the polyether segments.The elastic properties, as measured by CS and TS, as functions of the stiffness were poorer for the TPTPT copolymers as opposed to for the T6T6T materials.This was due to the lower aspect ratio of the crystallites and their poorer packing, probably as a result of the non-planar structure and chair/boat conformations of the piperazine units.

Table 1 .
Thermal properties from DSC analysis of the copolymers based on PTMO 1 000 and PTMO 2 000 .