《材料导论》课程教学资源（文献资料）ELSEVIER Thermochimica Acta 276（1996）145-160 Correlation of observations made by DSC and hot-stage optical microscopy of the thermal properties of a monotropic liquid-crystal polyester.pdf_大学文库

therm0chimica acta ELSEVIER Thermochimica Acta 276 (1996) 145-160 Correlation of observations made by DSC and hot-stage optical microscopy of the thermal properties of a monotropic liquid-crystal polyester Z. Bashir *, N. Khan Courtaulds, 72 Lockhurst Lane, Coventry CV6 5RS, UK Received 22 May 1995; accepted 20 September 1995 Abstract The polyester formed by condensation of 4,4'-bis(6-hydroxyhexoxy)biphenyl (BHHBP) and isophthalic acid has been reported to show many interesting and intriguing thermal properties. By optical microscopy, we established that the polymer shows monotropic liquid-crystalline behaviour. Thus, on heating, no mesophases are formed and the crystal melts directly to the isotropic state; on cooling, however, the material transforms from the isotropic melt to a nematic and smectic A mesophase before crystallizing. However, the DSC heatin 9 scan of this polymer was very complex, showing an exotherm and two endotherms; further, unlike the phenomenon of cold crystallization, the exotherm occurred between the two endotherms. If there is only a single phase transition (crystal-isotropic) on heating, then the multiple peaks need to be explained. In this work, we have shown that the multiple peaks in the heating scan are solely related to the thermal history of the sample. The nascent polymer and quench-cooled material showed multiple peaks during heating, whereas slowly crystallized polymer showed a single melting endotherm. By quench cooling from the isotropic melt, it was not possible to obtain the material in an amorphous form, because this polyester had a propensity for fast ordering. In the previous work, on coolin9, we only found two instead of three DSC peaks. Here, we have attempted to see if three peaks (corresponding to I-N, N-S A and S A K) can be resolved by going to very slow cooling rates (0.1°C min- 1). It was found that even at such low cooling rates, I-N and the N-S A transitions could not be separated into distinct peaks. This is in accord with optical microscope observations which indicated that the transformation to the smectic phase from the nematic was never complete and was overtaken eventually by crystallization. This kind of incomplete phase transformation appears to be a feature of this monotropic liquid-crystalline polymer. * Corresponding author. Elsevier Science B.V. SSDI 0040-6031(95)02756-4

146 Z.Bashir,N.Khan/Thermochimica Acta 276(1996)145-160 Keywords:4,4-Bis(6-Hydroxyhexoxy)biphenyl:Isophthalic acid:Monotropic liquid-crystal: DSC;Optical microscopy;Polyester:Thermal history 1.Introduction The polyeste forme condensation of isophthalic acid and 4,4'-bis(6-hy. droxyhexoxy)biphenyl(BHHBP)has interesting thermal properties.The polyester is shown below(Scheme 1). In the earliest reports on various condensation polymers formed with the diol BHHBP [1,2]it was stated that the above polyester formed a smectic A liquid-crystal phase.Subsequently,a more detailed investigation [3]showed that this polyester formed not only a smectic but also a nematic liquid-crystal phase.Moreover,it was shown that the mesophases were monotropic in character and were observed only on cooling and not on heating [3].The phase sequence on cooling was proposed to be I→N-→S+K and K→I on heating (I=isotropic.N=nematic,.S smectic A K=crystal).In another investigation.it was dem nstrated that deformation of this polyester in the smectic melt state led to an extraordinary chain orientation perpen dicular to the flow direction or the fibre axis [4]. f the erential e poly alo imetry(DSC).The DSC principally by hot-sta ould on th ole nterpre e pha equence own abo t there re s g aspects wh d no n resolved For were co plex,sh owing multiple e nermic peaks,even th It was argued that there is only one phase transition(K I).At he ating rates o min or lower,it was found that there were two endothermic peaks s with an exothermic peak lying between them [2].This is reproduced in Fig.1a.Although an exothermic peak before and endothermic peak is quite common on heating semi- crystalline polymers,owing to"cold crystallization",an exothermic peak between two endothermic peaks was more unusual and difficult to explain.We postulated that this may be a result of partial melting.followed by recrystallization to a more perfect form, followed by remelting of the perfected form at a higher temperature.Preliminary annealing experiments were cited as supporting this proposal but sufficient experimen- tation had not been conducted at the time to be definitive about the nature of the multiple peaks observed in the heating scan [27.Further,on cooling.two exothermic peaks were found:these started to overlap as the cooling rate was lowered.However,as the phase sequence cited above shows,on cooling there are three transitions and hence (-0-(CH26-0 -oM Scheme 1.The polymer med and 44-bsydroxyhey)biphenyl

146 Z. Bashir, N. Khan/Thermochimica Acta 276 (1996) 145-160 Keywords: 4,4'-Bis(6-Hydroxyhexoxy)biphenyl; Isophthalic acid; Monotropic liquid-crystal; DSC; Optical microscopy; Polyester; Thermal history 1. Introduction The polyester formed by condensation of isophthalic acid and 4,4'-bis(6-hydroxyhexoxy)biphenyl (BHHBP) has interesting thermal properties. The polyester is shown below (Scheme 1). In the earliest reports on various condensation polymers formed with the diol BHHBP [1,2] it was stated that the above polyester formed a smectic A liquid-crystal phase. Subsequently, a more detailed investigation [3] showed that this polyester formed not only a smectic but also a nematic liquid-crystal phase. Moreover, it was shown that the mesophases were monotropic in character and were observed only on cooling and not on heating [3]. The phase sequence on cooling was proposed to be I ~ N ~ S a ~ K and K ~ I on heating (I = isotropic, N = nematic, S A = smectic A, K = crystal). In another investigation, it was demonstrated that deformation of this polyester in the smectic melt state led to an extraordinary chain orientation perpendicular to the flow direction or the fibre axis [4]. These phases of the polyester were established principally by hot-stage microscopy and differential scanning calorimetry (DSC). The DSC results could on the whole be interpreted according to the phase sequences shown above, but there were still certain intriguing aspects which had not been resolved properly. For example, the heating scans of the polyester were complex, showing multiple endothermic peaks, even though it was argued that there is only one phase transition (K--* I). At heating rates of 5°C min 1 or lower, it was found that there were two endothermic peaks with an exothermic peak lying between them [2]. This is reproduced in Fig. la. Although an exothermic peak before and endothermic peak is quite common on heating semicrystalline polymers, owing to "cold crystallization", an exothermic peak between two endothermic peaks was more unusual and difficult to explain. We postulated that this may be a result of partial melting, followed by recrystallization to a more perfect form, followed by remelting of the perfected form at a higher temperature. Preliminary annealing experiments were cited as supporting this proposal but sufficient experimentation had not been conducted at the time to be definitive about the nature of the multiple peaks observed in the heating scan [2]. Further, on cooling, two exothermic peaks were found; these started to overlap as the cooling rate was lowered. However, as the phase sequence cited above shows, on cooling there are three transitions and hence O O (-- O --(CH2)6-- O ~~~- O -- (CH2)6-- O ~ ) n Scheme 1. The polymer formed by condensation of isophthalic acid and 4,4'-bis(6-hydroxyhexoxy) biphenyl

150 Z. Bashir, N. Khan/ Thermochimica Acta 276 (1996) 145-160 0 0 HO--(CH2)6--O~~~--O_(CH2)6__OH + CI CI O O (-- O-- (CH2)6-- O ~~~- O--(CH2)6-- O/~~ ) n Scheme 3. Reaction of BHHBP and isophthalic acid. polymer (referred to as the nascent polymer in the paper) showed perfect extinction at the clearing point under the polarizing microscope whereas the unpurified polymer showed some birefringent specks in the isotropic melt. Therefore purification of the polymer is important. 2.2. Differential scannin9 calorimetry (DSC) Calorimetric measurements were performed using a Mettler DSC 30 and TC 11 controller. This is a heat-flux type of DSC. Data storage and analysis were performed using an IBM microcomputer running Graph Ware TA-72 under QNX. The DSC cell was calibrated for temperature and enthalpy response according to the melting points and heats of fusion of pure indium. The samples were encapsulated in hermetically sealed pans and the scans were conducted under a static air atmosphere. In order to find if the heating scans are dependent on the thermal history of the polymer, the same sample was used for all the experiments. Degradation was established not to be a problem providing the sample was not heated above about 150°C, as all features could be reproduced for a particular thermal history with the same sample. For both heating and cooling experiments, a 20 mg sample and a 40/~L pan were used. 2.2.1. Heatin9 scan of nascent and slowly crystallized polymer The heating scan of the nascent polymer was recorded. Heat from 30°C ~ 150°C at 5°C min 1 (heating scan of nascent polymer, Fig. la). Cool from 150°C ~ 30°C at - 5°C min 1 (to form slowly-crystallized polymer). Heat from 30°C ~ 150°C at 5°C min - 1 (heating scan of slowly-crystallized polymer, Fig. lb). 2.2.2. Heatin9 scans of isothermally crystallized polymer In order to understand the origin of the multiple peaks found in the heating scan shown in Fig. la, an isothermally crystallized sample was produced. Four isothermal

Z.Bashir.N.Khan/Thermochimica Acta 276(1996)145-160 151 crystallization temperatures were tried (100C,105C,110C and 115C).The pro- cedure used for isothermal crystallization was as follows: Heat 30C-150C at 5C min(destroy thermal history of nascent polymer). Cool from 150C-100C (or 105C,110C.115C)at -5C min-1. Hold isothermal at 100C (or 105C,110C,115C)for 5h (isothermal crystalliza- tion). Ouench to -50C. Reheat-50C-150C at 5C min-(heating scan of isothermally crystallized polymer,.Fig1c-le以 2.2.3.Heating scans of quench-cooled polymer The quench-cooled sample was produced in order to measure the glass transition and compare its heating scan with that of the nascent polymer. Heat from -100C-150C at 5C min-1. Hold isothermal at 150C for 1 min n from DSC cel and immerse in liquid nitrogen(to produce quench- led s -100C and allow 10 min to equilib at I and n -100C for 10 min (heating scan of quench-cooled sample, Fig.10) 2.2.4.Coolin orde g sc mns and e ct of cooling rat her th ons could be resolved on melt was 001 he lowest possib scans were thus performed at cooling rates of 5,-0.7 3,and The following sequence was used for the first cooling rate: Cool from150°C+30°Cat-5°Cmin In the above sequence,the sample was heated very rapidly to the starting tempera- ture of 150C(for isotropization). For the very slow cooling scans (-0.7,-0.3 and -0.1C min),the following procedure was used in order to reduce the length of the experiment.Again,the sample was heated very rapidly to the starting temperature of 150C for isotropization,and was cooled in two stages as shown below. Cool from 150C-120C at-5C min-1(there are no transitions in this interval and a faster rate cuts time). Cool from 120C-70C at -0.7C min (or -0.3,-0.1C min). 2.3.Optical microscopy Optical observations were made with a Zeiss Axioplan polarizing microscope equipped with long working distance objectives.The samples were heated and cooled with a Linkam hot-stage and associated temperature controller.Very small powdered

Z. Bashir, N. K han/Thermochimica Acta 276 (1996) 145-160 151 crystallization temperatures were tried (100°C, 105°C, 110°C and 115°C). The procedure used for isothermal crystallization was as follows: Heat 30°C - 150°C at 5°C min- 1 (destroy thermal history of nascent polymer). Cool from 150°C ---, 100°C (or 105°C, 110°C, 115°C) at - 5°C min- 1. Hold isothermal at IO0°C (or 105°C, 110°C, 115°C) for 5h (isothermal crystallization). Quench to - 50°C. Reheat -50°C ---, 150°C at 5°C min 1 (heating scan of isothermally crystallized polymer, Fig. lc-le). 2.2.3. Heatin9 scans of quench-cooled polymer The quench-cooled sample was produced in order to measure the glass transition and compare its heating scan with that of the nascent polymer. Heat from - 100°C --, 150°C at 5°C min- 1. Hold isothermal at 150°C for 1 min. Remove pan from DSC cell and immerse in liquid nitrogen (to produce quenchcooled sample). Set DSC cell at - 100°C and allow 10 rain to equilibrate• Reinsert pan in DSC cell and hold isothermal at - 100°C for 10 min. Heat from - 100°C --, 150°C at 5°C min- 1 (heating scan of quench-cooled sample, Fig. lf). 2•2.4. Coolin9 scans and effect of coolin9 rate In order to check whether three exothermic transitions could be resolved on cooling, the isotropic melt was cooled at the lowest possible cooling rates attainable. DSC scans were thus performed at cooling rates of -5, -0.7, -0.3, and -0.1°C • --1 rain The following sequence was used for the first cooling rate: Cool from 150°C --, 30°C at - 5°C min- 1. In the above sequence, the sample was heated very rapidly to the starting temperature of 150°C (for isotropization). For the very slow cooling scans (-0.7, -0.3 and -0.1°C min 1), the following procedure was used in order to reduce the length of the experiment. Again, the sample was heated very rapidly to the starting temperature of 150°C for isotropization, and was cooled in two stages as shown below. Cool from 150°C--, 120°C at -5°C min-1 (there are no transitions in this interval and a faster rate cuts time). Cool from 120°C--.70°C at -0.7°C min 1 (or -0.3, -0.1°C min 1). 2.3. Optical microscopy Optical observations were made with a Zeiss Axioplan polarizing microscope equipped with long working distance objectives• The samples were heated and cooled with a Linkam hot-stage and associated temperature controller. Very small powdered

152 Z.Bashir,N.Khan/Thermochimica Acta 276(1996)145-160 fragments of the polymer were placed between two glass cover slips and heated to 150C (i.e.about 20C above the isotropization ter melt was very viscous and did not form a thin lav e r spo ce the melt between the gla as sheared with a c state.The isotr was the -0.7o ni opic The terturalhotorcaltion ofter nperatur to se was meltec 0Cbout Cboveisotroiiotr etween two thin cover slips,and removed wthwrromthe heating block of the Linkam hot-stae nd imrsdq nitrogen.The quenched polymer film that was sandwiched between the coverslips was examined at room temperature by eye for turbidity,and with the polarizing microscope for birefringence. 3.Results and discussion 3.1.Effect of thermal history on heating scan 3.1.1.Nascent polymer and well-crystallized polymer formed by slow cooling As stated earlier sed that the phases in this polvr are monotropicin and he obs ved on cooling [2] the being crystal ith Yet.the e DS of the nase Fi la)do a single ng endotherm,but man between. Wehave herm: Ithis to crysta annealing effects [2].It was suggested that the scan in Fig.la could be interpreted in terms of partial melting recrystallization and remelting [2].It has been proposed by Fischer et al.[1]that chain-folding can occur in polymers containing the monomer BHHBP.hence it is conceivable that during annealing,chain-folded crystals with a lower melting point convert to a higher melting,extended-chain crystal form.Thus,isothermal annealing of the nascent polymer for a prolonged time at a temperature below the crystal melting point should increase crystal size;hence,it should reduce crystal-perfectioning trans- formations occurring during the heating scan.In our initial work [2],one such annealing experiment was reported.The nascent polymer was heated to 115C and annealed for Ih and after cooling to 30C the sample was reheated at 1'C min1.The heating scan of the annealed polymer showed only a maior melting peak at 130 C with a small shoulder at a lower temperature r21 The fact that there is only a cr ystal-isotropic transition on heating can be shown by another experiment demo nstrated here.If instead of ar nnealing the is first ted and slowly at -5C min-1,a well-crystallized material is obtained.When the well-c crystallized polymer is heated at 5c min only a single nelting endotherm is obse ed (Fig Ih are with Fig. la).This clearly indic thethe nal hist of the (ie slowly crystallized polymer)affec s the heating s nascent polyme

152 Z. Bashir, N. Khan/Thermochimica Acta 276 (1996) 145 160 fragments of the polymer were placed between two glass cover slips and heated to 150°C (i.e. about 20°C above the isotropization temperature). The melt was very viscous and did not form a thin layer spontaneously. Hence the melt between the glass plates was sheared with a circular motion in the isotropic state. The isotropic melt (150°C) was then continuously cooled at about -0.7°C min ~ to room temperature (20°C). The textural changes were recorded photographically as a function of temperature. In order to see if an amorphous polymer could be prepared, the polymer was melted at 200°C (about 70°C above isotropization) between two thin cover slips, and removed with tweezers from the heating block of the Linkam hot-stage and immersed in liquid nitrogen. The quenched polymer film that was sandwiched between the coverslips was examined at room temperature by eye for turbidity, and with the polarizing microscope for birefringence. 3. Results and discussion 3.1. Effect of thermal history on heatin9 scan 3.1.1. Nascent polymer and well-crystallized polymer formed by slow coolin9 As stated earlier, we proposed that the mesophases in this polymer are monotropic in character and hence are only observed on cooling [2]. On heating, the crystal transforms directly to the isotropic melt, without any liquid crystal phase being formed. Yet, the DSC heating scan of the nascent reactor polymer (Fig. la) does not show a single melting endotherm, but manifests two endotherms with an exotherm in between. We have tentatively attributed this to crystal annealing effects [2]. It was suggested that the scan in Fig. la could be interpreted in terms of partial melting, recrystallization and remelting [2]. It has been proposed by Fischer et al. [1] that chain-folding can occur in polymers containing the monomer BHHBP, hence it is conceivable that during annealing, chain-folded crystals with a lower melting point convert to a higher melting, extended-chain crystal form. Thus, isothermal annealing of the nascent polymer for a prolonged time at a temperature below the crystal melting point should increase crystal size; hence, it should reduce crystal-perfectioning transformations occurring during the heating scan. In our initial work [2], one such annealing experiment was reported. The nascent polymer was heated to 115°C and annealed for lh and after cooling to 30°C the sample was reheated at I°C min- 1. The heating scan of the annealed polymer showed only a major melting peak at 130°C with a small shoulder at a lower temperature [2]. The fact that there is only a crystal isotropic transition on heating can be shown by another experiment demonstrated here. If instead of annealing the nascent polymer, it is first melted and cooled slowly at -5°C min 1, a well-crystallized material is obtained. When the well-crystallized polymer is reheated at 5°C min- 1, only a single melting endotherm is observed (Fig. lb, compare with Fig. la). This clearly indicates the thermal history of the sample (i.e. nascent polymer versus slowly crystallized polymer) affects the heating scan

Z. Bashir, N. Khan/Thermochimica Acta 276 (1996) 145 160 153 3.1.2. Isothermally crystallized sample Here, a well crystallized sample was produced not by isothermal annealing [-2] or slow cooling from the melt but by an isothermal crystallization from the melt state. This was achieved by cooling from the isotropic state at 150°C to the crystallization temperature (100°C, 105°C, 110°C or 115~C) and holding for 5 h at this temperature. After this, the sample was rapidly cooled and was reheated from - 50°C to 150°C at 5°C min- 1 The heating scan of the sample crystallized at 100°C (Fig. lc) shows a single peak at 120"C with a shoulder at 130°C. The isothermally crystallized sample melts at a slightly higher temperature than the slowly crystallized polymer (Fig. lb) because of its even larger and more perfect crystals. After isothermal crystallization at 105°C, the heating scan in Fig. ld was obtained. This also shows a major endotherm with peak at 120~C; however, a small but distinct peak at 130cC is also noticeable. This at first is puzzling, but it merely means that the true crystalline melting point of the extended chain crystal is probably near 1309C. If the polymer is crystallized at a higher temperature, such as 110°C, then the heating scan shows apparently renewed complexities (Fig. le). The dominant endotherm is now at 130°C (instead of 120°C), but there are several smaller peaks preceding it. The smaller peaks appear to be a sequence of alternating endotherms and exotherms, suggesting that the transformation takes place by a series of almost discrete melting and recrystallization steps. Crystallization at higher temperatures such as 115cC also led to heating scans similar to that in Fig. le (not shown). Thus, it still seems that there is only a single K ~ I transition and the equilibrium crystal melting temperature is just above 130°C. Optical microscope observations were also consistent with the DSC findings. An isothermally crystallized sample (110°C for 10 h) was produced under similar conditions between glass cover slips; this showed a banded spherulitic texture (Fig. 2). On heating, the birefringence diminished at about 125c'C and the spherulites melted and Fig. 2. Banded spherulitic texture of a sample that had been isothermally crystallized at 100'C for 10 h

154 Z. Bashir, N. K han/Thermochimica Acta 276 (1996) 145 160 transformed to the isotropic melt at 133°C; there were no intervening smectic or nematic textures on heating. 3.1.3. Quench-cooled polymer These experiments were conducted to probe further the effect of thermal history on the heating scan of the polymer, and to establish the glass transition temperature. The material was prepared by heating to the isotropic state (150'C) and quenching the molten polymer by dropping the DSC pan in liquid nitrogen. The quenched material was transferred from the liquid nitrogen to the DSC cell which had been equilibrated at -100c'C. It was interesting to compare the heating scan (at 5°C rain 1) of the quench-cooled polymer with the nascent polymer and the slowly or isothermally crystallized material. This is shown in Fig. lf. On heating the quench-cooled material, a weak glass transition was observed between 30 and 40:~C. After the glass transition, again an endotherm occurs before the exotherm and this is followed by the main melting endotherm as well as a very small endothermic peak near 130"~C. Thus, the quench-cooled material is similar to the nascent polymer. It appears that partial melting, reorganization and remelting occur. This behaviour ought to be contrasted with that of quench-cooled poly (ethylene terephthalate) where after the Tg, a cold crystallization exotherm is observed followed by the melting endotherm. Visual observations were also conducted on quench-cooled material. It was found that even material formed by quenching from 200C (i.e. about 70~C above the isotropization temperature) was turbid instead of clear. In contrast, quench-cooled poly(ethylene terephthalate), which is a slowly crystallizing polymer, is generally transparent. Fig. 3 shows that the quenched material was birefringent when viewed between crossed polars, though no recognizable features such as spherulites could be seen. We may contrast the behaviour of this polyester with poly(ethylene terephthlate) Fig. 3. Optical texture of quench-cooled polymer at room temperature, between crossed polars. The birefringence indicates quench cooling in liquid nitrogen does not produce an amorphous polymer. The circular black region is void