Influence of the side-chain structure and molecular weight on the re-entrant behaviors of mesogen-jacketed liquid crystalline polymers

Abstract

Three series of mesogen-jacketed liquid crystalline polymers (MJLCPs) containing different terminal groups (phenmethyl, diphenylmethyl and triphenylmethyl) in the side chains, abbreviated as Pv-m-Bn, Pv-m-DPM, Pv-m-Tr (m = 2, 4, 6, 8, 10, and 12, which are the number of methylene units between the terephthalate core and terminal groups in the side chains), were designed and successfully synthesized via free-radical polymerization. Molecular characterization of the polymers was performed by ¹H NMR, GPC and TG analysis. The phase structures and transitions of the polymers were investigated by a combination of techniques including DSC, POM and 1D/2D WAXD. The experimental results revealed that all the polymers exhibited excellent thermal stabilities and the re-entrant behaviors of the MJLCPs were found to be strongly dependent on the structure of the side-chain, i.e., the spacer length increased with the volume of the terminal groups when the polymers exhibited the re-entrant isotropic phase. On the other hand, a series of MJLCPs, poly{2,5-bis[(diphenylmethoxy-ethyl)oxycarbonyl]-styrenes} (Pv-2-DPMs), with different molecular weights (M_n) and narrow M_n distributions have been successfully synthesized via ATRP. The results indicated that when the M_n was below 1.73 × 10⁴ g mol⁻¹, only the isotropic phase was observed. When M_n was between 3.40 × 10⁴ g mol⁻¹ and 8.48 × 10⁴ g mol⁻¹, a re-entrant isotropic phase was formed at low temperatures and a columnar nematic phase at high temperatures. By further increasing the M_n to exceed 9.71 × 10⁴ g mol⁻¹, a stable columnar nematic phase was developed. This work provides two effective ways to design and synthesize MJLCPs with re-entrant behaviors; moreover, it is meaningful to deeply understand the structure–property relationships of MJLCPs.

---

1. Introduction

In the past decades, liquid crystalline polymers (LCPs) have attracted considerable attention. This is a consequence not only of their scientific interest but also, from an industrial point of view, of their potential applications in fields such as engineering plastics, optical data storage, optics, electro-optics, nonlinear optic devices, and photomechanical devices.^1–10 These applications strongly depend on the phase structures and phase transitions of LCPs. Therefore, many researchers have investigated the structure–property relationships of LCPs in order to understand the principles of structure formation and structure manipulation.

In general, LCPs have two main categories: main-chain liquid crystalline polymers (MCLCPs) with mesogen units located in the main chain^11–13 and side-chain liquid crystalline polymers (SCLCPs) with mesogens attached to the main chain as side groups,^14–16 which can be readily obtained by attaching anisotropic mesogens to the backbone via a flexible spacer,¹⁷ with an excellent combination of liquid crystalline (LC) and polymer characteristics. The flexible spacers between the side-chain mesogens and the backbone decouple the motions of mesogens and main chains, facilitating the formation of LC phases. Later, Zhou et al.¹⁸ reported a special type of side-on SCLCPs called mesogen-jacketed liquid crystalline polymers (MJLCPs), which had a short spacer with a single covalent bond connecting the mesogen to the polymer backbone. The “jacketing effect,” owing to steric hindrance, forces the main chain to take an extended-chain conformation, leading to the MJLCPs displaying many thermotropic properties characteristic of MCLCPs, although they are chemically SCLCPs, such as a high glass transition temperature, a broad LC temperature range, and the formation of a banded texture after mechanical shearing in the LC state. Based on the concept of the “jacketing effect”, many researchers have investigated the structure–property relationships of MJLCPs to understand the principles of structure formation and structure manipulation.^19,20 For example, our group has researched the re-entrant isotropic phase behavior of MJLCPs.^21,22

The re-entrant phase is an exceptional case in LC physics, as it violates the second law of thermodynamics, which states that molecular order should decrease as temperature increases;²³ this was first discovered by Yu et al. when they studied a lauric acid potassium solution in 1980. Currently, the re-entrant phase behavior can be basically divided into three categories, namely the re-entrant isotropic phase, the re-entrant nematic phase and the re-entrant columnar phase. At present, two main classes of thermotropic LC polymers with re-entrant phase behavior have been studied. One is end-on SCLCPs based on mesogenic units containing a cyano group, with five or six atoms in the flexible spacer and a polyacrylate or poly(vinyl ether) backbone.^24–27 These polymers exhibit a re-entrant nematic phase due to the frustration induced by competing order parameters that favor different periodicities. The other is dendritic LCPs and MJLCPs, which exhibit a re-entrant isotropic phase.^{21,22,28–37} For example, Percec et al.²⁸ studied poly{4-[3,4,5-tris(n-dodecanyloxy)benzoyloxy]-4-(2-vinyloxyethoxy)biphenyl} and found that, as the DP was about 4 to 5, the polymer exhibited a re-entrant isotropic phase. Zhu et al.^32,38 have synthesized an MJLCP containing triphenylene (Tp) moieties in the side chains, with 12 methylene units as spacers. The experimental results showed that the low-temperature phase of the polymer was a hexagonal columnar phase, self-organized by Tp discotic mesogens. The high-temperature phase was a nematic columnar phase with a larger dimension developed by the rod-like supramolecular mesogen—the MJLCP chain as a whole. A re-entrant isotropic phase was found in the medium temperature range. Yin et al.³³ and Chen et al.²¹ have reported the synthesis and characterization of a series of novel MJLCPs, with different length alkyl tails, named poly[di(alkyl)vinylterephthalates] (P-m) and poly{2,5-bis[(4-alkoxyphenyl)oxycarbonyl]styrenes} (P-OCm). When m = 6, 8, and 10, P-m exhibited the re-entrant isotropic phase, and P-OCm showed the re-entrant isotropic phase when m > 4. Yu et al.³⁰ and Zhu et al.³¹ also found that poly[2,5-bis(4′-alkoxycarbonylphenyl)styrene]s and poly[2,5-bis(4′-alkoxyphenyl)-styrene] both exhibited the re-entrant isotropic phase when the alkyl tail length exceeded a minimum critical length.

Beside the structural factors, the re-entrant behavior of MJLCPs is also influenced by the M_n. Zhao et al.²⁹ reported re-entrant-phase behavior of MJLCPs based on [poly-{2,5-bis[(4-butoxyphenyl)oxycarbonyl]styrene}] (PBPCS), which were synthesized by ATRP (Atom Transfer Radical Polymerization). The samples with M_n below 2.42 × 10⁴ g mol⁻¹ were isotropic. The samples with M_n exceeding 3.36 × 10⁴ g mol⁻¹ displayed a thermodynamically stable isotropic phase at lower temperatures and an LC phase at higher temperatures. Our group has synthesized a series of MJLCPs via ATRP, poly{2,5-bis[(4-octa-decyloxyphenyl)oxycarbonyl]-styrenes} (P-OC18s).²² The results indicated that the polymers formed a smectic phase at low temperature and an isotropic phase at high temperature when the M_n was lower than a critical M_n of approximately 4.6 × 10⁴ g mol⁻¹. The samples with M_n ≥ 5.2 × 10⁴ g mol⁻¹ displayed a re-entrant isotropic phase.

In this study, in order to deepen understanding of the relationship between the chemical structure and M_n and the re-entrant behavior of the MJLCPs, a series of MJLCPs has been synthesized by changing the length of the spacer and the volume of the side group, named poly{2,5-bis[(phenmethyl-alkyl)oxycarbonyl]-styrenes}, Pv-m-Bn, and poly{2,5-bis[(diphenylmethoxy-alkyl)oxycarbonyl]-styrenes}, Pv-m-DPM (m = 2, 4, 6, 8, 10, and 12, which is the number of methylene units between the terephthalate core and the terminal groups in the side chains). Combined with the results for the poly{2,5-bis[(triphenylmethoxy-alkyl)oxycarbonyl]-styrenes} (Pv-m-Tr),³⁹ it was revealed that all the polymers exhibited excellent thermal stabilities, and the re-entrant behaviors of the MJLCPs were found to be strongly dependent on the side-chain structure and M_n. The spacer length required for re-entrant behavior in the polymers increases with the volume of the terminal groups.

Meanwhile, a series of MJLCPs with different M_ns has been prepared by ATRP polymerization, named poly{2,5-bis[(diphenylmethoxy-ethyl)oxycarbonyl]-styrenes} (Pv-2-DPMs). It was discovered that the MJLCPs presented isotropic phase → re-entrant isotropic phase → columnar phases as M_n increased. All the chemical structures of the polymers are shown in Scheme 1.

Scheme 1 Chemical structures of the polymers.

2. Experimental

2.1. Materials

Anhydrous tetrahydrofuran (THF) was distilled from sodium benzophenone ketyl under argon and used immediately. Triethylamine (TEA) and dichloromethane (CH₂Cl₂) were dried over anhydrous magnesium sulfate. Chlorobenzene (Acros 99%) was purified by washing with concentrated sulfuric acid to remove residual thiophenes, followed by washing twice, first with 5% sodium carbonate solution and then with water, before drying with anhydrous calcium chloride and distilling. Cuprous bromide (CuBr) was synthesized from CuBr₂, purified by stirring in acetic acid, and washed with methanol, before drying under vacuum just before use; 2,2-azobisisobutyronitrile (AIBN) was freshly recrystallized from methanol. N,N,N′,N′′,N′′-pentamethyldiethylenetriamine (PMDETA) (99.5%) and 2-bromoisobutyryl bromide were used as received without further purification. All other reagents and solvents were used as received without further purification.

2.2. Instruments and measurements

Hydrogen-1 NMR spectroscopy was performed on a Bruker ARX400 MHz spectrometer, using CDCl₃ as solvent and tetramethylsilane (TMS) as the internal standard at ambient temperature. Chemical shifts were reported on the ppm scale.

The apparent number average molecular weight (M_n) and polydispersity index (PDI = M_w/M_n) were measured on a GPC (WATERS 1515) instrument with a set of HT3, HT4, and HT5 μ-Styragel columns, using THF as eluent with a flow rate of 1.0 mL min⁻¹ at 38 °C. All the GPC data were calibrated with polystyrene standards.

TGA was performed on a TA SDT 2960 instrument at a heating rate of 20 °C min⁻¹ in a nitrogen atmosphere.

DSC traces of the polymer were obtained using a TA Q10 DSC instrument. The temperature and heat flow were calibrated using standard materials (indium and zinc) at cooling and heating rates of 10 °C min⁻¹. Samples with a typical mass of about 5 mg were encapsulated in sealed aluminum pans.

The LC texture of the polymers was examined under POM (Leica DM-LM-P), using a Mettler Toledo hot stage (FP82HT).

One-dimensional wide-angle X-ray diffraction (1D WAXD) experiments were performed on a BRUKER AXS D8 Advance diffractometer with 40 kV FL tubes as the X-ray source (Cu Kα) and a LYNXEYE_XE detector. Background scattering was recorded and subtracted from the sample patterns. The heating and cooling rates in the 1D WAXD experiments were 10 °C min⁻¹.

Two-dimensional wide-angle X-ray diffraction (2D WAXD) was carried out using a BRUKER AXS D8 Discover diffractometer with 40 kV FL tubes as the X-ray source (Cu Kα) and a VANTEC 500 detector. The point-focused X-ray beam was aligned either perpendicularly or parallel to the mechanical shearing direction. For both the 1D and 2D WAXD experiments, the background scattering was recorded and subtracted from the sample patterns.

2.3. Synthesis of monomers

The synthetic routes for Mv-m-Bn and Mv-m-DPM are outlined in Schemes 2 and 3.^21,40,41 Detailed information about the intermediate and monomers is shown in the ESI.†

Scheme 2 Synthetic route for the monomers (Mv-m-Bn) and the corresponding polymers (Pv-m-Bn).

Scheme 3 Synthetic route for the monomers (Mv-m-DPM) and the corresponding polymers (Pv-m-DPM).

2.4. Synthesis of polymers

The polymers [Pv-m-Bn and Pv-m-DPM in Schemes 2 and 3] were synthesized by conventional solution radical polymerization. For example, 0.30 g (0.41 mmol) of Pv-6-DPM, 40 μL of THF solution of 10 mg mL⁻¹ AIBN, 0.66 mL of THF and a magnetic stir bar were added into a polymerization tube, and the tube was purged with nitrogen and subjected to four freeze–thaw cycles to remove any dissolved oxygen, then sealed off under vacuum. Polymerization was carried out at 70 °C for 12 h. The tube was then opened and the reaction mixture was diluted with 9 mL of THF. Then, the resultant polymer was dropped slowly into a mixture of methanol/THF (4/1). Dissolution and precipitation were repeated three times. After drying under vacuum, 0.21 g of polymer was obtained. Yield: 70%.

In addition, Pv-2-DPM polymers with different M_ns were successfully synthesized by ATRP, and detailed synthetic procedures are shown in S2.†

3. Results and discussion

3.1. Synthesis and characterization of monomers and corresponding polymers

As shown in Schemes 2 and 3, all the monomers were synthesized via only two steps. The structures of the monomers were confirmed through ¹H NMR, IR (see S3†) and mass spectrometry measurements.

All the monomers could be easily polymerized via free-radical polymerization. Herein, we use Pv-6-DPM as an example to elucidate the process. Fig. 1(a) and (b) give the ¹H NMR spectra (CDCl₃^−d) of the monomer Mv-6-DPM and the polymer Pv-6-DPM, respectively. Mv-6-DPM showed characteristic resonances of the vinyl group at 5.39–5.76 ppm. After polymerization, the signals disappeared completely. The absorption peaks of Pv-6-DPM were quite broad and consistent with the expected polymer structure. In addition, the IR spectra of Mv-6-DPM and Pv-6-DPM are shown in Fig. S2.† Obviously, the out-of-plane vibration of the vinyl group disappeared after polymerization. The apparent number-average molecular weights of the polymers determined by GPC were higher than 1 × 10⁵ g mol⁻¹, demonstrating good polymerizability of the monomers. The molecular characterization of the polymers is summarized in Tables 1 and 2.

Fig. 1 ¹H NMR spectra of the monomer Mv-6-DPM (a) and the polymer Pv-6-DPM (b) in CDCl₃.

Table 1 Molecular characteristics and properties of the series of Pv-m-Bn

Sample	Mn^a (× 10^−5)	PDI^a	Td^b (°C)	Tm^c (°C)	Tg^c (°C)	Tg′^c (°C)	T1^d (°C)	Liquid crystallinity^d
a The apparent number-average molecular weight (Mn) and polydispersity index (PDI) were measured by GPC using PS standards.b The temperatures of 5% weight loss under nitrogen were measured by TGA heating experiments at a rate of 20 °C min^−1.c The melting temperatures (Tm) and the glass transition temperatures (Tg) during the second heating process and the glass transition temperatures (Tg′) during the first cooling process were all measured by DSC at a rate of 10 °C min^−1 under a nitrogen atmosphere.d Evaluated by POM at a heating and cooling rate of 10 °C min^−1.
Pv-2-Bn	1.85	1.87	365			−8.0		No
Pv-4-Bn	1.05	1.83	350		−32.4	−36.5	160	Yes
Pv-6-Bn	1.14	1.91	359		−44.0	−46.4	139	Yes
Pv-8-Bn	1.53	1.83	369				237	Yes
Pv-10-Bn	1.24	1.81	344	−46.4				No
Pv-12-Bn	1.01	1.80	358	−31.9				No

---

Table 2 Molecular characteristics and properties of the series of Pv-m-DPM

Sample	Mn^a (× 10^−5)	PDI^a	Td^b (°C)	Tg^c (°C)	T1^d (°C)	Liquid crystallinity^d
a The apparent number-average molecular weight (Mn) and polydispersity index (PDI) were measured by GPC using PS standards.b The temperatures of 5% weight loss under nitrogen were measured by TGA heating experiments at a rate of 20 °C min^−1.c The melting temperatures (Tm) and the glass transition temperatures (Tg) during the second heating process and the glass transition temperatures (Tg′) during the first cooling process were all measured by DSC at a rate of 10 °C min^−1 under a nitrogen atmosphere.d Evaluated by POM at a heating and cooling rate of 10 °C min^−1.
Pv-2-DPM	1.32	1.95	361	45.2		Yes
Pv-4-DPM	1.21	2.16	355	10.8		Yes
Pv-6-DPM	1.93	2.01	349	−5.0		Yes
Pv-8-DPM	1.27	1.81	358	−19.2	126	Yes
Pv-10-DPM	1.42	1.75	358	−27.5		No
Pv-12-DPM	1.13	1.85	346	−33.0		No

---

Polymers (Pv-2-DPMs) with different M_n were successfully synthesized by ATRP. The GPC curves are shown in Fig. S3 and the results are summarized in Table S1.† The M_n was gradually increased from 1.73 × 10⁴ to 13.65 × 10⁴ g mol⁻¹. Furthermore, the polydispersity of the polymers was even lower than the theoretical value of 1.30 for the controlled “living” free-radical polymerization.

3.2. Thermal and liquid-crystalline properties of the polymers

The thermal and liquid-crystalline properties of the polymers (Pv-m-Bn, Pv-m-DPM and Pv-2-DPMs) were investigated by DSC and POM. The polymers (Pv-m-Bn and Pv-m-DPM) were previously investigated by TGA. As shown in Tables 1 and 2, all the polymers exhibited excellent thermal stabilities, with the 5% weight loss temperatures being about 350 °C in nitrogen.

The phase-transition behaviors of the polymers were investigated by DSC. The samples were firstly heated from ambient temperature to 300 °C at a rate of 20 °C min⁻¹ under nitrogen atmosphere to eliminate the thermal history. Fig. 2, 3, and S5† show the DSC curves of Pv-m-Bn, Pv-m-DPM, and Pv-2-DPMs at a rate of 10 °C min⁻¹, respectively.

Fig. 2 DSC curves of polymers Pv-m-Bn during the second heating scan (a) and the first cooling scan (b) at a rate of 10 °C min⁻¹ under nitrogen atmosphere.

Fig. 3 DSC curves of polymers Pv-m-DPM during the second heating scan (a) and the first cooling scan (b) at a rate of 10 °C min⁻¹ under nitrogen atmosphere.

For the series of Pv-m-Bn (m = 2, 4, 6, 8, 10, 12), Fig. 2(a) and (b) show the second heating and the first cooling DSC curves at a rate of 10 °C min⁻¹, respectively. The polymer Pv-2-Bn showed a clear glass transition (T_g′) temperature during the first cooling at −8.0 °C; however, no peaks were detected during the second heating. For the polymers Pv-4-Bn and Pv-6-Bn, a T_g was observed for both heating and cooling. For Pv-8-Bn, we could not find any peaks, but the polymers Pv-10-Bn and Pv-12-Bn showed significant transition peaks during heating and cooling at −46.4 °C and −31.9 °C, possibly as a result of melting and crystallization of the alkyl chain. The results are shown in Table 1.

For the series of Pv-m-DPM (m = 2, 4, 6, 8, 10, 12), Fig. 3(a) and (b) show the second heating and the first cooling DSC curves of Pv-m-DPM at a rate of 10 °C min⁻¹, respectively. All the samples exhibited a single and obvious T_g when heating or cooling. As expected, T_g decreases from 45.2 °C to −33.0 °C as the spacer length increases, owing to the strong internal plasticization effect of the longer alkyl spacers, and the results are listed in Table 2. No other transition peak was observed except T_g, and this phenomenon was also found in the other MJLCPs.

Fig. S5(a) and (b)† show the second heating and the first cooling DSC curves of Pv-2-DPM with different M_ns at a rate of 10 °C min⁻¹, respectively. A single and obvious T_g was observed under heating–cooling cycles, and the T_g increased slightly from 44.4 °C to 45.8 °C as the M_n increased. This implied that, when the M_n of the polymer exceeds 1.73 × 10⁴ g mol⁻¹, the M_n of the polymer has no effect on its T_g.

Polarized optical microscopy (POM) was used to investigate the liquid crystallinity of the polymers. To allow consistency with the DSC result (the second heating), the testing conditions were the same as for DSC. For the series of Pv-m-Bn, the results showed that the polymers can be divided into two types. The first was Pv-m-Bn (m = 2, 10 and 12). No birefringence was detected by POM, implying the absence of the thermotropic mesophase. The second was Pv-m-Bn (m = 4, 6 and 8). On heating, the sample became soft above the T_g, but no birefringence was observed, indicating the formation of an isotropic phase. Then, a granular LC texture was observed when the temperature reached about 160, 139, and 237 °C, respectively. Fig. 4 shows the POM images of the representative sample, Pv-6-Bn, at low temperature [Fig. 4(a)] and high temperature [Fig. 4(b)]. With further heating, no visible change in the birefringence was observed before decomposition (onset temperature > 300 °C). On subsequent cooling, the original scene was re-established, that is, the birefringence disappeared. Therefore, based on our previous research,²¹ the former polymers were amorphous polymers, but the latter were the first type of MJLCPs, with a re-entrant phase.^29,31

Fig. 4 Representative POM images of the texture of Pv-6-Bn maintained at 100 °C (a) and 180 °C (b) (magnification: ×200).

In order to observe the phase transition temperature intuitively, we performed reflected light intensity (birefringence) measurements. The phase transition from the isotropic to the LC phase could not be detected by DSC but only by reflected light intensity measurements, because the thermal enthalpy change was extremely small. Fig. 5 shows the recorded changes in reflected light intensity for Pv-4-Bn, Pv-6-Bn, and Pv-8-Bn by POM at a rate of 10 °C min⁻¹. As can been seen, obvious transition peaks appeared at 160 °C, 139 °C, and 237 °C (T₁), respectively, in which the reflected light intensity sharply increased to 56% from 11%, showing a transition from an isotropic to an ordered state.⁴² When cooling, the change in intensity was slower than during heating, indicating that after the polymer formed the columnar phase, it could retain the ordered structure to some degree to reduce the influence of the entropic driving force (see Fig. S4†).

Fig. 5 The change in reflected light intensity for Pv-4-Bn, Pv-6-Bn, and Pv-8-Bn at a heating rate of 10 °C min⁻¹ in POM.

For the series of Pv-m-DPM, the results showed that the polymers can be divided into three types. The first is Pv-m-DPM (m = 2, 4 and 6). These exhibited strong birefringence that invariably remained even when heated to 300 °C or subsequently cooled to room temperature. All these polymers showed stable needle textures and the POM image of polymer Pv-6-DPM is shown in Fig. 6(a), suggesting the formation of a columnar phase. The second was Pv-8-DPM, in which the phase transition was the same as for Pv-m-Bn (m = 4, 6 and 8), and the transition temperature from the isotropic phase to the LC phase was about 126 °C. Fig. 6(b) and (c) show POM images of the polymer at the isotropic and LC phases, respectively. The last one was Pv-m-DPM (m = 10 and 12). No birefringence was detected by POM, indicating the absence of the thermotropic mesophase. In general, Pv-m-DPMs can be classified into three types. The first type (m = 2, 4 and 6) were transitional MJLCPs, while the second (m = 8) was the first type of MJLCP with a re-entrant phase,^29,31 and the last (m = 10 and 12) were amorphous polymers.

Fig. 6 Representative POM images of the texture of Pv-6-DPM maintained at 180 °C (a), and representative POM images of the texture of Pv-8-DPM maintained at 100 °C (b), and 180 °C (c) (magnification: ×200).

Fig. 7 shows the recorded changes in reflected light intensity for Pv-6-DPM and Pv-8-DPM by POM at a rate of 10 °C min⁻¹. From the curves, there was little change in the intensity of Pv-6-DPM and the reflected light intensity remained at 59%. However, for Pv-8-DPM, an obvious transition peak appeared at 126 °C (T₁), in which the reflected light intensity sharply increased to 55% from 11%, showing a transition from an isotropic state to an ordered state.⁴² When cooling, the reflected light intensity returned to the original state (see Fig. S4†).

Fig. 7 The change in reflected light intensity for Pv-6-DPM and Pv-8-DPM at a heating rate of 10 °C min⁻¹ in POM.

For the series of Pv-2-DPMs, the POM results showed that these polymers can also be divided into three types. The first one was P1. No birefringence was detected by POM. The second type was P2–P5. The observed results were similar to those for Pv-m-Bn (m = 4, 6 and 8) and Pv-8-DPM. Taking P3 as an example, the POM images are shown in Fig. 8(a) and (b). The transition temperature from the isotropic to the LC phase was about 162 °C. The third type was P6–P7. The samples exhibited a stable texture and the LC birefringence remained unchanged even when heated to 300 °C. While cooling to room temperature from 300 °C, the birefringence of the sample remained unchanged, showing that the ordered structure remained unchanged upon cooling. So, the polymers were amorphous when M_n was below 1.73 × 10⁴ g mol⁻¹. The polymers belonged to the first kind of MJLCPs, with a re-entrant phase when M_n was between 3.40 × 10⁴ g mol⁻¹ and 8.48 × 10⁴ g mol⁻¹.^29,31 Lastly, the polymers were transitional MJLCPs when M_n exceeded 9.71 × 10⁴ g mol⁻¹.

Fig. 8 Representative POM images of the texture of P3, maintained at 100 °C (a) and 180 °C (b) (magnification: ×200).

Fig. 9 shows the recorded changes in reflected light intensity for P1–P6 by POM at a rate of 10 °C min⁻¹. For P1 and P6, there was little change in the intensity and the reflected light intensity when kept at 7% and 72%, respectively. However, P2–P5 showed obvious transition peaks at 224 °C, 162 °C, 143 °C, and 137 °C (T₁), in which the reflected light intensity sharply increased to 64% from 7%, showing a transition from an isotropic state to an ordered state,⁴² and the transition temperature decreased as M_n increased.

Fig. 9 The change in reflected light intensity for P1–P6 at a heating rate of 10 °C min⁻¹ in POM.

3.3. Phase structure identification of the polymers

To further characterize the mesomorphic structure of Pv-m-Bn, Pv-m-DPM, and Pv-2-DPMs, temperature-dependent powder 1D WAXD was used. About 60 mg of the polymer was added into an aluminum foil substrate. The testing conditions were consistent with the DSC and POM measurements.

For the series of Pv-m-Bn and Pv-m-DPM, the 1D WAXD results were consistent with the POM results. Fig. 10(a) and (b) illustrate the temperature-variable 1D WAXD patterns of Pv-6-DPM during the second heating (30–250 °C) and subsequent cooling (250–30 °C). And the test region of 2θ was from 1.5° to 35°. In the low-angle region, one narrow reflection peak was observed at 2θ = 4.22° (d = 2.09 nm), demonstrating that ordered structures formed at the nanometer scale. The intensity of the halo basically remained the same at an elevated temperature, i.e., 250 °C. When cooling down, the diffraction peak intensity remained unchanged, suggesting that the polymer retained the ordered structures over the whole temperature range. In the high-angle region, only an amorphous halo around 20° could be recognized. This indicated that no long-range ordered structure formed via molecular packing was detected over the entire temperature region studied. In addition, no noticeable change could be observed during the second heating and cooling processes in 1D WAXD experiments, implying that there was no isotropic phase for Pv-6-DPM in the temperature region from 30 to 250 °C. The 1D WAXD patterns of Pv-2-DPM and Pv-4-DPM were similar to those of Pv-6-DPM. Therefore, Pv-m-DPM (m = 2, 4 and 6) probably forms a stable columnar phase.

Fig. 10 1D WAXD patterns of Pv-6-DPM during the second heating (a) and subsequent cooling (b).

Fig. 11(a) and (b) show the structurally sensitive 1D WAXD patterns of Pv-8-DPM. When the temperature was below 100 °C, the low-angle diffraction peak was diffuse and weak. When temperatures were above 100 °C, a sharp and intense peak at 2θ = 3.81° (d = 2.32 nm) appeared. As the temperature was further increased, the intensity of the peak increased. The peak became weak and disappeared at 100 °C on the subsequent cooling scan [see Fig. 11(b)]. The distinct discontinuous intensities corresponded with an endothermic transition at about 130 °C during the second heating. The observation of discontinuities during cooling clearly indicated that the LC phase formed at higher temperatures disappeared during cooling. This implied that Pv-8-DPM formed a re-entrant phase, and similar results were obtained from Pv-m-Bn (m = 4, 6, 8).

Fig. 11 1D WAXD patterns of Pv-8-DPM during the second heating (a) and subsequent cooling (b).

However, for Pv-m-Bn (m = 2, 10, 12) and Pv-m-DPM (m = 10, 12), no low-angle scattering peaks were observed in the 1D WAXD patterns during heating and cooling, suggesting that these were amorphous throughout the temperature region.

Fig. 12(a) and (b) depict two sets of 1D WAXD patterns for Pv-m-Bn and Pv-m-DPM (m = 2, 4, 6, 8, 10, 12) samples obtained at 250 °C, and the data for all the polymers at 250 °C are summarized in Tables 3 and 4. The results showed that the d-spacing value increases with m.

Fig. 12 1D WAXD patterns of Pv-m-Bn (a) and Pv-m-DPM (b) (m = 2, 4, 6, 8, 10 and 12) at 250 °C.

Table 3 The 2θ d-spacing values and calculated length of the side-chain of Pv-m-Bn

Sample	2θ^a (°)	d-Spacing^a (nm)	Calculated length of the side-chain^b (nm)
a Obtained from one-dimensional WAXD experiments.b Assuming that the n-alkyl spacers in the side chains have an all-trans conformation.
Pv-2-Bn	—	—	2.58
Pv-4-Bn	4.65	1.90	3.08
Pv-6-Bn	4.34	2.04	3.59
Pv-8-Bn	4.05	2.18	4.12
Pv-10-Bn	—	—	4.54
Pv-12-Bn	—	—	5.16

---

Table 4 The 2θ d-spacing values and calculated length of the side-chain of Pv-m-DPM

Sample	2θ^a (°)	d-Spacing^a (nm)	Calculated length of the side-chain^b (nm)
a Obtained from one-dimensional WAXD experiments.b Assuming that the n-alkyl spacers in the side chains have an all-trans conformation.
Pv-2-DPM	4.51	1.96	2.59
Pv-4-DPM	4.20	2.10	3.10
Pv-6-DPM	4.02	2.20	3.60
Pv-8-DPM	3.81	2.32	4.10
Pv-10-DPM	—	—	4.52
Pv-12-DPM	—	—	5.14

---

1D WAXD experimental results for the Pv-2-DPMs showed that these samples could also be divided into three types. The first type was P1; no low-angle scattering peaks were seen in the 1D WAXD patterns during heating and cooling [as shown in Fig. S6(a) and S6(b)†]. The second was P2–P5; taking P4 as the example, the observation of 1D WAXD is shown in Fig. 13. In Fig. 13(a), below 130 °C, the low-angle diffraction peak is diffuse and weak. When temperatures were above 130 °C, a sharp and intense peak at 2θ = 4.42° (d = 2.00 nm) appeared. This peak became weak and disappeared at 100 °C during the following cooling scan [see Fig. 13(b)], which was similar to Pv-8-DPM [see Fig. 11], indicating that the LC phase formed at higher temperatures disappeared during cooling. This suggested that P4 formed a re-entrant phase, and the polymers (P2, P3, and P5) showed similar results. The transition temperatures detected by 1D WAXD decreased as M_n increased, which was consistent with the POM results.

Fig. 13 1D WAXD patterns of P4 during the second heating (a) and subsequent cooling (b).

The third type was P6–P7; taking P6 as an example, Fig. 14(a) and (b) show structurally sensitive 1D WAXD patterns of P6 from 30 to 180 °C and from 180 to 30 °C during the second heating and subsequent cooling. This result was similar to that for Pv-6-DPM [see Fig. 10]; only one narrow diffraction peak was observed at a low angle and an amorphous halo at a high angle throughout heating and cooling, indicating that P6 presented a stable columnar phase. The 1D WAXD patterns for P1–P7 obtained at 250 °C are shown in Fig. 15. As can be seen, the 2θ value of the narrow reflection peak for P2–P7 at the low angle remained the same, implying that the rod diameter of Pv-2-DPMs was not changed as M_n increased.

Fig. 14 1D patterns of P6 during the second heating (a) and subsequent cooling (b).

Fig. 15 1D WAXD patterns of P1–P7 at 250 °C.

Two-dimensional WAXD was used to further identify the phase structures. All fiber samples were drawn with a pair of tweezers at a temperature above the glass-transition temperature. Fig. 16 shows the 2D WAXD patterns of the polymer Pv-6-DPM at room temperature with the X-ray incident beam perpendicular and parallel to the fiber direction. In Fig. 16(a), the fiber axis was perpendicular to the meridian direction. A pair of strong diffraction arcs could be seen on the equator at 2θ = 4.22° (d = 2.09 nm), which was consistent with the 1D WAXD results, indicating that the ordered structures had developed along the direction perpendicular to the fiber axis on the nanometer scale. When the X-ray incident beam was parallel to the shear direction, a ring pattern at 2θ = 4.30° (d = 2.05 nm) was detected, which exhibited an isotropic intensity distribution [Fig. 16(b)]. Meanwhile, diffusion scattering halos at the high 2θ angle were observed, suggesting that only short-range order exists along the fiber direction. Considering the similar X-ray results reported previously,^20,43 we proposed that the polymer Pv-6-DPM exhibited a columnar nematic phase. The polymers Pv-m-DPM (m = 2 and 4) and P6–P7 showed similar 2D WAXD results, and were also considered to form a columnar nematic phase as each cylinder is attributed to a single polymer chain with the side groups tightly jacketing the backbone.

Fig. 16 2D WAXD fiber patterns of Pv-6-DPM. The X-ray incident beam was perpendicular (a) and parallel (b) to the fiber axis.

For the polymers Pv-m-Bn (m = 4, 6, 8), Pv-8-DPM, and P2–P5, we could obtain 2D WAXD patterns from a sample heated to 250 °C, then quickly put into the liquid nitrogen to keep the LC phase structure. The 2D WAXD patterns of Pv-6-Bn and Pv-8-DPM are shown in Fig. 17. From the patterns, a ring pattern at a low 2θ angle and diffusion scattering halos at the high 2θ angle were presented when the X-ray incident beam was perpendicular to the fiber direction; however, these could not be orientated because of the extremely low T_g of these polymers. Combined with the results of POM and 1D WAXD, it was considered that these polymers formed a columnar nematic phase at high temperatures.

Fig. 17 2D WAXD patterns of Pv-6-Bn (a) and Pv-8-DPM (b) with the X-ray incident beam perpendicular to the fiber direction.

3.4. Influence of the side-chain structure on the phase behaviors

Based on the results of DSC, POM, 1D WAXD and 2D WAXD experiments, the phase behaviors of Pv-m-Bn, Pv-m-DPM, and Pv-m-Tr (m = 2, 4, 6, 8, 10, 12) are summarized in Fig. 18. For series 1 (Pv-m-Bn) and series 2 (Pv-m-DPM), the polymers show an isotropic phase when m is 10 or 12. However, Pv-10-Tr exhibits a stable columnar nematic phase and Pv-12-Tr forms a re-entrant phase. On the other hand, Pv-2-Bn cannot form an LC phase, while Pv-2-DPM and Pv-2-Tr can both present a stable LC phase. These results can be explained by the steric effect of the side group. The strength of the steric effect is determined by the “static steric effect” and the “dynamic steric effect”. The strength of the “static steric effect” relies on the side-chain structure occupation volume at low temperatures, which is related to the rigidity, size and shape of the side chain. Yet, the strength of the “dynamic steric effect” relies on the dynamic volume of the side-chain at high temperatures, which is determined by the chemical structure of the side-chain and the temperature. As for Pv-m-Bn, Pv-m-DPM and Pv-m-Tr, the change of the terminal group structure from phenmethyl to triphenylmethyl means that the static space volume of the side-chain increases. Therefore, it is easy to understand that Pv-2-Bn cannot display an LC phase due to the weak static steric effect. On the other hand, with an increasing number of benzene rings in the terminal group, the critical value of the flexible spacer length (m) increases from 4 to 12 when the polymer displays the re-entrant phase (Pv-m-Bn, Pv-m-DPM, and Pv-m-Tr). This is caused by the dynamic steric effect. When the spacer (m) increases, the static steric effect becomes weak, leading to the polymer forming an isotropic phase at low temperatures. However, the dynamic steric effect becomes strong at high temperatures, resulting in the formation of an LC phase. Further increasing the spacer length causes the dynamic steric effect to also become weak.

Fig. 18 Schematic drawing of the thermotropic phase behavior of Pv-m-Bn, Pv-m-DPM, and Pv-m-Tr (m = 2, 4, 6, 8, 10, 12).

Previously, our group has researched the influence of the alkyl tail length on the phase behavior and phase structure of poly{2,5-bis[(4-alkoxyphenyl)oxycarbonyl]styrenes} (P-OCm, m = 1, 2, 4, 6, 8, 10, 12, 14, 16 and 18). It was found that the polymers could show a re-entrant isotropic phase when the tail length exceeded a critical value. According to the results of this article, the alkyl spacer length and the alkyl tail length play a very important role in the formation of a re-entrant phase by the MJLCPs.

3.5. Influence of the molecular weight on the phase behavior

The evolution of the phase behavior as M_n is increased in this series of Pv-2-DPMs can be seen in Fig. 19. The phase transitions of the Pv-2-DPMs with moderate M_n follow the sequence, re-entrant isotropic phase ↔ columnar nematic phase ↔ isotropic phase (this is not true). In the past, Zhao et al.²⁹ synthesized a series of PBPCS with different M_ns by ATRP and found that the phase behavior depended on the M_n. When M_n was below 2.42 × 10⁴ g mol⁻¹, the polymers presented an isotropic phase; as M_n increased, the polymer exhibited a re-entrant isotropic phase at low temperatures and a hexagonal columnar phase at high temperatures. However, the PBPCS did not display a stable columnar phase, even when M_n was 35.6 × 10⁴ g mol⁻¹, while the Pv-2-DPMs showed a stable columnar phase when the M_n exceeded 9.71 × 10⁴ g mol⁻¹. Therefore, the entropy effect of the side-chain depends strongly on the M_n. The temperature of the isotropic to LC phase transition decreases as the M_n is further increased and is even lower than room temperature.

Fig. 19 The relationship between the transition temperature and the M_n of Pv-2-DPMs. (I: isotropic phase, Φ_N: columnar nematic phase).

4. Conclusions

In summary, we have successfully synthesized two series of MJLCPs (Pv-m-Bn and Pv-m-DPM, m = 2, 4, 6, 8, 10, 12) with different spacer lengths and terminal groups by free-radical polymerization, and a series of MJLCPs (Pv-2-DPMs) with different M_ns by ATRP. The phase structures and transitions of these polymers were investigated by a combination of DSC, POM, and 1D/2D WAXD.

For the series of Pv-m-Bn and Pv-m-DPM, the T_g decreased as m increased due to alkyl plastification. The polymers Pv-m-Bn (m = 2, 10 and 12) and Pv-m-DPM (m = 10 and 12) presented an isotropic phase because of a poor jacketing effect.⁴⁴ Pv-m-DPM (m = 2, 4 and 6) showed a stable columnar nematic phase because of a strong static steric effect. Pv-m-Bn (m = 4, 6, 8) and Pv-8-DPM showed an isotropic phase at low temperatures and a columnar nematic phase at high temperatures, because the backbone adopts a somewhat extended conformation due to the consecutively tempestuous motion of side-chains around the main-chain,⁴⁵ leading to the production of a strong dynamic steric effect as the temperature increased. Combined with the results for Pv-m-Tr (m = 2, 4, 6, 8, 10 and 12), we understand that, for the polymers to form the re-entrant phase, they need longer flexible spacers when they have larger terminal groups.

For the series of Pv-2-DPMs, when M_n was below 1.73 × 10⁴ g mol⁻¹, only the amorphous state was observed. When M_n was between 3.40 × 10⁴ g mol⁻¹ and 8.48 × 10⁴ g mol⁻¹, a re-entrant isotropic phase at low temperatures and a columnar nematic phase at high temperatures formed. By further increasing the M_n to exceed 9.71 × 10⁴ g mol⁻¹, a stable columnar nematic phase was developed.

In all, this work provided two facile ways to discover and synthesize polymers with re-entrant behaviors.

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (21504075) and the Research Foundation of Education College of Hunan Province, China (Grant No. 14C1090).

Notes and references

1. Z. Q. Yu, T. T. Li, Z. Zhang, J. H. Liu, W. Z. Yuan, J. W. Y. Lam, S. Yang, E. Q. Chen and B. Z. Tang, Macromolecules, 2015, 48, 2886–2893.
2. J. F. Ban, L. N. Mu, L. Chen, S. J. Chen and H. L. Zhang, RSC Adv., 2016, 6, 38790–38796.
3. M. Tian, H. H. Sun, X. W. Tian, Z. X. Xu, F. Q. Wang, D. S. Yao and J. S. Hu, Liq. Cryst., 2015, 42, 298–308.
4. M. Wang, J. Wang, H. Yang, B. P. Lin, E. Q. Chen, P. Keller, X. Q. Zhang and Y. Sun, Chem. Commun., 2016, 52, 4313–4316.
5. S. F. Zhang, C. X. Zhang, J. C. Wang, F. Hong, X. T. Hao, A. Zhang, Y. F. Wang, H. Wu, W. Y. Zhang and J. L. Pu, Polym. Chem., 2016, 7, 3013–3025.
6. S. Hassan, R. Anandakathir, M. J. Sobkowicz and B. M. Budhlall, Polym. Chem., 2016, 7, 1452–1460.
7. Z. Y. Zhang, Q. K. Zhang, Z. H. Shen, J. P. Yu, Y. X. Wu and X. H. Fan, Macromolecules, 2016, 49, 475–482.
8. S. Q. Yang, W. Qu, H. B. Pan, Y. D. Zhang, S. J. Zheng, X. H. Fan and Z. H. Shen, Polymer, 2016, 84, 355–364.
9. S. Bronnikov, S. Kostromin, V. Musteaţa and V. Cozan, J. Polym. Res., 2016, 23, 1–8.
10. H. N. Gayathri, B. Kumar, K. A. Suresh, H. K. Bisoyi and S. Kumar, Phys. Chem. Chem. Phys., 2016, 18, 12101–12107.
11. M. Tokita, S. Okuda, S. Yoshihara, C. Takahashi, S. Kang, K. Sakajiri and J. Watanabe, Polymer, 2012, 53, 5596–5599.
12. T. S. Jo, A. K. Nedeltchev, B. Biswas, H. Han and P. K. Bhowmik, Polymer, 2012, 53, 1063–1071.
13. C. Ruan, L. Chen, R. Yang, H. Y. Zhong and Y. Z. Wang, RSC Adv., 2015, 5, 48541–48550.
14. B. Geng, L. X. Guo, B. P. Lin, P. Keller, X. Q. Zhang, Y. Sun and H. Yang, Polym. Chem., 2015, 6, 5281–5287.
15. Y. S. Xu, D. Shi, J. Gu, Z. Lei, H. L. Xie, T. P. Zhao, S. Yang and E. Q. Chen, Polym. Chem., 2016, 7, 462–473.
16. X. Q. Liu, J. Wang, S. Yang and E. Q. Chen, ACS Macro Lett., 2014, 3, 834–838.
17. H. Finkelmann, H. Ringsdorf and J. H. Wendorff, Makromol. Chem., 1978, 179, 273.
18. Q. F. Zhou, H. M. Li and X. D. Feng, Macromolecules, 1987, 20, 233–234.
19. Y. Y. Qiao, J. Ping, H. J. Tian, Q. K. Zhang, S. Zhou, Z. H. Shen, S. J. Zheng and X. H. Fan, J. Polym. Sci., Part A: Polym. Chem., 2015, 53, 2116–2123.
20. L. Y. Zhang, H. L. Wu, Z. H. Shen, X. H. Fan and Q. F. Zhou, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 3207–3217.
21. S. Chen, C. K. Jie, H. L. Xie and H. L. Zhang, J. Polym. Sci., Part A: Polym. Chem., 2012, 50, 3923–3935.
22. S. Chen, H. Luo, H. L. Xie and H. L. Zhang, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 924–935.
23. P. E. Cladis, Mol. Cryst. Liq. Cryst., 1988, 165, 85–121.
24. W. Renz, Mol. Cryst. Liq. Cryst., 1988, 155, 549–558.
25. B. I. Ostrovskii, S. N. Sulyanov, N. I. Boiko, V. P. Shibaev and W. H. D. Jeu, Eur. Phys. J. E, 2001, 6, 277–285.
26. J. C. Dubois, G. Decobert, P. L. Barny, S. Esselin, C. Friedrich and C. Noël, Mol. Cryst. Liq. Cryst., 1986, 137, 349–364.
27. N. Lacoudre, A. L. Borgne, N. Spassky, J. P. Vairon, P. L. Barny, J. C. Dubois, S. Esselin, C. Friedrich and C. Noël, Mol. Cryst. Liq. Cryst., 1988, 155, 113–127.
28. V. Percec, M. Lee, J. Heck, H. E. Blackwell, G. Ungar and A. Alvarez-Castillo, J. Mater. Chem., 1992, 2, 931–938.
29. Y. F. Zhao, X. H. Fan, X. H. Wan, X. F. Chen, Y. Yi, L. S. Wang, X. Dong and Q. F. Zhou, Macromolecules, 2006, 39, 948–956.
30. Z. N. Yu, H. L. Tu, X. H. Wan, X. F. Chen and Q. F. Zhou, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 1454–1464.
31. Z. G. Zhu, J. G. Zhi, A. H. Liu, J. X. Cui, X. H. Wan and Q. F. Zhou, J. Polym. Sci., Part A: Polym. Chem., 2007, 45, 830–847.
32. Y. F. Zhu, H. J. Tian, H. W. Wu, D. Z. Hao, Y. Zhou, Z. H. Shen, D. C. Zou, P. C. Sun, X. H. Fan and Q. F. Zhou, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 295–304.
33. X. Y. Yin, C. Ye, X. Ma, E. Q. Chen, X. Y. Qi, X. F. Duan, X. H. Wan, S. Z. Cheng and Q. F. Zhou, J. Am. Chem. Soc., 2003, 125, 6854–6855.
34. D. Zhang, Y. X. Liu, X. H. Wan and Q. F. Zhou, Macromolecules, 1999, 32, 5183–5185.
35. S. Chen, J. Qiu, H. L. Xie and H. L. Zhang, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 2804–2816.
36. L. Y. Zhang and Y. F. Zhang, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 2545–2554.
37. X. Q. Liu, J. Wang, S. Yang, Y. F. Men, P. C. Sun and E. Q. Chen, Polymer, 2016, 87, 260–267.
38. Y. F. Zhu, X. L. Guan, Z. H. Shen, X. H. Fan and Q. F. Zhou, Macromolecules, 2012, 45, 3346–3355.
39. Y. B. Luo, S. Chen and H. L. Zhang, RSC Adv., 2015, 5, 54920–54928.
40. S. George and A. Sudalai, Tetrahedron: Asymmetry, 2007, 18, 975–981.
41. Y. Bikard, J. M. Weibel, C. Sirlin, L. Dupuisc, J. P. Loefflerc and P. Palea, Tetrahedron Lett., 2007, 48, 8895–8899.
42. S. Chen, X. Shu, H. L. Xie and H. L. Zhang, Polymer, 2013, 54, 3556–3565.
43. G. L. Jiang, H. H. Cai, Z. H. Shen, X. H. Fan and Q. F. Zhou, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 557–564.
44. X. F. Chen, Z. H. Shen, X. H. Wan, X. H. Fan, E. Q. Chen, Y. G. Ma and Q. F. Zhou, Chem. Soc. Rev., 2010, 39, 3072–3101.
45. S. T. Sun, H. Tang, P. Y. Wu and X. H. Wan, Phys. Chem. Chem. Phys., 2009, 11, 9861–9870.

---

Footnote

† Electronic supplementary information (ESI) available: More detailed synthesis and characterization. See DOI: 10.1039/c6ra13363k

---