Synthesis of novel biobased polyimides derived from isomannide with good optical transparency, solubility and thermal stability

Gaili Yanga, Rui Zhanga, Huahua Huang*a, Lixin Liua, Lei Wang*b and Yongming Chen*a
aKey Laboratory for Polymeric Composite and Functional Materials of Ministry of Education, School of Chemistry and Chemical Engineering, Sun Yat-sen University, No. 135, Xingang Xi Road, Guangzhou 510275, China. E-mail: huanghh27@mail.sysu.edu.cn; chenym35@mail.sysu.edu.cn
bShenzhen Key Laboratory of Special Functional Materials, College of Material Science, Shenzhen University, Shenzhen 518060, China. E-mail: wl@szu.edu.cn

Received 24th June 2015 , Accepted 29th July 2015

First published on 29th July 2015


Abstract

Novel biobased polyimides (PIs) with good optical transparency and comprehensive properties were synthesized from isomannide-derived diamine and dianhydride monomers. Three kinds of diamines including 2,5-diamino-2,5-dideoxy-1,4:3,6-dianhydroiditol (M1), 1,4:3,6-dianhydro-2,5-di-O-(4-aminophenyl)-D-mannitol (M2), and 1,4:3,6-dianhydro-2,5-di-O-(2-trifluoromethyl-4-aminophenyl)-D-mannitol (M3), as well as 1,4:3,6-dianhydro-2,5-di-O-(3,4-dicarboxyphenyl)-D-mannitol dianhydride (M4), were prepared based on isomannide. These diamines M1–M3 were reacted with M4 and a commercial dianhydride, 4,4′-oxydiphthalic anhydride (ODPA), via a two-step polymerization method, respectively, to yield a series of biobased PI films, PI-1 to PI-6. The resultant PIs had a high content of biomass up to 48 wt%, and they can be readily soluble in various non-proton polar solvents at room temperature. Most of the biobased PIs showed good optical transparency (transmittances at 450 nm over than 80%), along with a cut-off wavelength of 343–364 nm. Furthermore, due to the existence of rigid alicyclic isomannide among the polymeric backbone, biobased PIs maintained fairly high thermal stability with a glass transition temperature of 227–264 °C, and temperature at 5% weight loss over 400 °C in nitrogen. Meanwhile, these PIs exhibited outstanding mechanical properties with tensile strengths greater than 90 MPa and elongation at break higher than 6.0%. It was also found that the biobased PI series with alicyclic M1 possessed higher thermal stability than PIs with semi-aromatic diamines M2 and M3. Thereof, the introduction of biomass building blocks into PIs can offer a great opportunity to develop new sustainable materials with high performance for microelectronic and optoelectronic applications.


Introduction

Polyimides (PIs), as one of the most important classes of high-temperature polymers, have been widely applied in engineering fields for several decades, due to their excellent combined properties such as outstanding thermal stability, high mechanical strength, and good dielectric property.1 It is also known that conventional aromatic PIs generally exhibit deep color and low optical transparency, as well as poor solution processability, caused by intra- or intermolecular charge transfer complex (CTC) formation in their conjugated molecular structures. However, in recent years, with the rapid development of advanced optoelectronic engineering, PIs (glass transition temperature, Tg > 200 °C) with good optical transparency, are of more and more importance, and they have been considered as one of the most promising flexible substrates for prospective electronic and optical devices.2,3 The common transparent polymers, such as polyethylene naphthalate (PEN) and polyethylene terephthalate (PET) cannot support the high-temperature processing of those devices, e.g., barrier coating, electrode deposition, and patterning, due to their low Tg values (80–150 °C). Thus, more and more interests have been drawn to erase the color of PIs. Considerable progresses have been achieved to synthesize new optically transparent PIs by introducing alicyclic units,4–6 fluorinated,7–9 chlorinated10 or sulfurated11,12 groups, and bulky or twisted rigid moieties13,14 into PIs, so as to inhibit or reduce the CTC formation.

Currently, the common and effective approach to the synthesis of optically transparent PIs is to use alicyclic diamine and/or dianhydride monomers. There are two synthetic strategies to obtain alicyclic diamines or dianhydrides: one is that alicyclic moieties are formed during the preparation of monomers;15 the other is that the existing alicyclic units are directly introduced into monomers through chemical modification.16 For the former one, several typical alicyclic dianhydrides, such as 1,2,3,4-cyclobutane tetracarboxylic dianhydride,15 2,3,5,6-bicyclooctane tetracarboxylic dianhydride17 and 1,2,4,5-cyclohexane tetracarboxylic dianhydride18 have been widely prepared to generate various transparent PIs. Some of the monomers are commercially available. However, it is difficult to modify the structures of these alicyclic monomers, thus to tune the performance of the resultant PIs. Moreover, it is mostly limited to the synthesis of dianhydrides. For the second strategy, whereas, alicyclic dianhydrides or diamines can be facilely prepared, based on alicyclic moieties containing reactive groups; furthermore, it can combine different functional units such as trifluoromethyl group (CF3), to form more desirable monomers. Nevertheless, only a few alicyclic moieties including cyclohexane,19,20 adamantane21 and recently reported heterocyclic piperazine,22 were utilized to obtain new monomers. Adamantane, a rigid aliphatic compound composed of three cyclohexane rings, have been introduced into various alicyclic diamines. Although adamantine-based PIs exhibited colorless and transparent with high thermal stability, they often possessed low solubility.21 Recently, a heterocyclic moiety, piperazine, was reported to form a new kind of alicyclic dianhydride, and fully aliphatic PIs with good solubility and transparency can be prepared. Unfortunately, thermal properties of such PIs are not well enough and some of their Tgs are even lower than 200 °C.22 Thereof, the synthesis of optically transparent PIs with excellent combined properties is still a great challenge. The key issue is how to design and develop new suitable monomers, so as to realize the balance among optical transparency, thermal stability, solubility, and other properties of the resultant PIs. Currently reported alicyclic diamines or dianhydrides are still limited; meanwhile, the monomers of PIs are almost confined to petroleum chemicals.

In recent decade, the research on the biobased polymers from abundant biomass building blocks is being in bloom stage, which is driven by the shortage of oil resources and their escalating prices, as well as the deterioration of global environment.23,24 One type of biomass building blocks that has been gained increasing attention is the isohexide compound derived from natural cereal through reduction of sugars and subsequent dehydration reaction. Due to its rigid heterocyclic structure and chirality, as well as the commercial product of its isomers, isosorbide and isomannide, considerable progresses have been achieved in synthesis of various isohexide-based polymers, e.g., polyesters, polyamides, polycarbonates and polyurethanes.25–27 It was also found that the incorporation of rigid isohexides or their diamino derivatives can effectively improve the thermal stability of biobased polymers.28,29 Recently, biobased PIs composed of petroleum dianhydrides and isohexide diamines were synthesized, and they had good optical transparency in a patent.30 However, the solubility of these PIs was not investigated, which was a key factor to determine their application. In addition, isohexides could be easily transformed into different kinds of diamines and dianhydrides by using their hydroxyl groups. It means that optically transparent biobased PIs with good combined properties such as solubility and thermal properties would be prepared by tuning the structures of isohexide-derived monomers. Up to now, although a great deal of studies have been reported on the synthesis of biobased polymers, the research on biobased PIs is still scarce.31,32

In this article, we demonstrated the synthesis of novel biobased PIs derived from isomannide biomass with good optical transparency and combined performance. For detailed study on the structure–property relationship of biobased PIs, three kinds of isomannide-based diamines, M1–M3 as shown in Scheme 1, were designed and prepared, including aliphatic, semi-aromatic, and semi-aromatic diamine containing CF3 group. Meanwhile, an isomannide-based dianhydride monomer M4 was also synthesized, to further increase the biomass content of the resultant PIs. These biobased diamines were reacted with a biobased dianhydride M4 and a commercial dianhydride ODPA, respectively, via a two-step method, to yield a series of new tough PI films. The solubility, thermal property, optical transparency and mechanical properties of these biobased PIs were investigated.


image file: c5ra14526k-s1.tif
Scheme 1 The synthetic routes of diamines M1–M3 and dianhydride M4 derived from isomannide.

Experimental

Materials

1,4:3,6-Dianhydro-D-mannitol (isomannide, 96%) was purchased from Accela Company. Phthalimide potassium salt (99%), triphenylphosphine (PPh3, 99%), 1-fluoro-4-nitrobenzene (98%), 4-chloro-1-nitro-3-(trifluoromethyl) benzene (99%), ODPA (97%), p-toluenesulfonyl chloride (TsCl, 99%), palladium on activated charcoal (Pd/C catalyst, 5%), hydrazine hydrate (80%), 4-nitrophthalonitrile (98%), and 1-methyl-2-pyrrolidinone (NMP) were purchased from Aladdin Company. All solvents including N,N′-dimethylacetamide (DMAc), N,N′-dimethylformamide (DMF), tetrahydrofuran (THF), and dichloromethane (CH2Cl2) are analytical grade from Guangzhou Chemical Reagent Factory. All the materials were used directly except DMF and NMP, which were dried over CaH2 and distilled prior to use.

Characterization

The attenuated total reflection Fourier-transform infrared spectra (ATR FT-IR) were measured on a Bruker Vector 22 spectrometer in the range of 400–4000 cm−1. 1H/13C NMR spectra were recorded on a Bruker Avance 400 spectrometer with DMSO-d6, CDCl3, or D2O as the solvents. Inherent viscosities were measured with an Ubbelohde viscometer placed in water at 35 °C. Molecular weights were tested by gel permeation chromatography (GPC) on a Waters equipped with a set of Waters Styragel HT-2, HT-4 and HT-6 columns, and a 2414 refractive index detector. The columns were maintained at 50 °C in an oven. DMF containing 0.01 M lithium bromide was used as an eluent with 1.0 mL min−1. Calibration was performed with narrow-molecular-weight polystyrene as standards. The thermal decomposition behaviors were recorded by thermogravimetric analysis (TGA) using a Pyris TG 2000 equipment with about 5 mg of film samples at a heating rate of 10 °C min−1 in N2 atmosphere. Measurements of Tgs were performed by differential scanning calorimetry (DSC) using a Netzsch DSC-204 instrument with a heating rate of 20 °C min−1 under N2. The coefficients of thermal expansion (CTEs) were conducted by thermo mechanical analysis (TMA) using TA Q400 equipment with a heating rate of 10 °C min−1, ranging from 50 to 200 °C. Ultraviolet-visible (UV-vis) spectra of the films (about 20 μm) were evaluated by a Thermo Evolution 201 spectrometer in transmittance mode, with spectrometer width of 300–800 nm at room temperature. Tensile properties were measured with Universal Testing Measurement (UTM) of Xinsansi Company. Materials were tested with 4 mm × 25 mm at a strain rate of 10 mm min−1 at room temperature and the maximum strength was 50 N. Wide-angle X-ray diffraction (WXRD) patterns were obtained with a RIGAKU D-MAX 2200 VPC, using graphite-monochromated Cu Kα radiation (λ = 1.5418 Å) and a 0.5 mm collimator at room temperature.

Synthesis of biobased monomers

1,4:3,6-Dianhydro-2,5-di-O-p-tosyl-D-mannitol (compound 1)33. To a solution of isomannide (24.01 g, 0.16 mol) in pyridine (30 mL) and CH2Cl2 (100 mL) at 0 °C was added a solution of TsCl (62.62 g, 0.33 mol) in the mixture solvents of pyridine (10 mL) and CH2Cl2 (150 mL) dropwise. The mixture was stirred under N2. After 10 h, the reaction product was poured into 0.5 L of an ice–water mixture, stirred and warmed up to room temperature. The resulting precipitate was filtered and the product was recrystallized from ethanol to afford isomannide bis-tosylate as white crystals (52.7 g, 84%). 1H NMR (CDCl3, ppm): δ = 7.78 (d, 4H, Ph), 7.32 (d, 4H, Ph), 4.83 (q, 2H, SO3CH), 4.46 (dd, 2H, OCH2), 3.91 (dd, 2H, OCH2), 3.71 (dd, 2H, OCH2), 2.46 (s, 6H, PhCH3).
2,5-Diazido-2,5-dideoxy-1,4:3,6-dianhydroiditol (compound 2). 10.12 g (22.12 mmol) of compound 1 was dissolved in 110 mL of DMF. The mixture was treated with stirring at 80 °C for 36 h after 1.57 g (22.0 mmol) of sodium azide was added. As soon as the temperature came down, water (60 mL) was added and the mixture was extracted with diethyl ether (15 mL × 7). The combined layers were dried over anhydrous MgSO4, and then concentrated under vacuum. The crude diazide was purified by column chromatography (cyclohexane/EtOAc, 5[thin space (1/6-em)]:[thin space (1/6-em)]1) to afford the compound 2 (2.0 g, 90.4%) as a colorless liquid. 1H NMR (CDCl3, ppm): δ = 4.65 (s, 2H, CH), 4.09 (q, 2H, N3CH), 3.96 (dd, 2H, CH2), 3.90 (dd, 2H, CH2).
2,5-Diamino-2,5-dideoxy-1,4:3,6-dianhydroiditol (M1)34. To the colorless liquid compound 2 (0.95 g, 77.76 mmol), was added PPh3 (5.10 g, 19.44 mmol), followed by a 10[thin space (1/6-em)]:[thin space (1/6-em)]1 solution of water in THF (total 110 mL). The reaction was conducted in 66 °C for 6 h. After the reaction, the THF was removed under reduced pressure. The residue was dissolved in the mixture of diethyl ether (25 mL) and 0.2 M aq. HCl (25 mL). After stirred thoroughly, the water was washed with diethyl ether (25 mL × 7), then using aq NaOH to neutralize the water layer. After removing water, most of the residue were dissolved with CH2Cl2 (70 mL × 3), and filtered. CH2Cl2 solvent was collected and dried with MgSO4 and then evaporated to yield the product M1 (0.5 g, 73%). 1H NMR (D2O, ppm): δ = 4.67 (s, 4H, NH2), 3.94 (dd, 4H, CH2), 3.77 (dd, 2H, CH), 3.67 (dd, 2H, CH). Elemental analysis: found: C, 50.0; H, 8.4; N, 19.4%. Calc. for C6H12N2O2: C, 50.1; H, 8.3; N, 19.6%.
1,4:3,6-Dianhydro-2,5-di-O-(4-nitrophenyl)-D-mannitol (compound 3)35. In a round-bottom flask fitted to the oil and water separation device, 1.46 g (9.99 mmol) of isomannide was mixed with 1.68 g (29.94 mmol) of KOH and 0.53 g (2.01 mmol) of 18-crown-6 were dissolved in DMF 15 mL. We next added 4.23 g (30.00 mmol) of p-fluoronitrobenzene and toluene 10 mL. The reaction mixture was maintained under magnetic stirring in 110 °C for 12 h. After cooled to room temperature, the reaction mixture was washed by K2CO3 saturated solution, then the residue was recrystallized by THF. The product was yellow granular crystal (2.6 g, 67%). 1H NMR (CDCl3, ppm): δ = 8.04 (d, 2H, Ph), 7.07 (d, 2H, Ph), 4.97 (dd, 2H, CH), 4.91 (q, 2H, CH), 4.13 (dd, 2H, CH2), 4.14 (dd, 2H, CH2).
1,4:3,6-Dianhydro-2,5-di-O-(4-aminophenyl)-D-mannitol (M2). In a flask, 1.94 g (5.00 mmol) of dinitro compound 3 was dissolved in 100 mL of THF with 1.12 mL of hydrazine monohydrate. Under stirring, 0.05 g of palladium catalyst was added. The reaction mixture was maintained under magnetic stirring in 66 °C for 12 h. Then the residue was recrystallized by THF. After purification, the compound M2 was obtained in a 91% (1.5 g) yield. 1H NMR (DMSO-d6, ppm): δ = 6.72 (m, 2H, Ph), 6.56 (m, 2H, Ph), 4.68 (m, 2H, CH), 4.65 (s, 4H, NH2), 4.63 (m, 2H, CH), 4.00 (dd, 2H, CH2), 3.68 (dd, 2H, CH2). Elemental analysis: found: C, 65.1; H, 5.9; N, 8.2%. Calc. for C18H20N2O4: C, 65.6; H, 6.1; N, 8.5%.
1,4:3,6-Dianhydro-2,5-di-O-(2-trifluoromethyl-4-nitrophenyl)-D-mannitol (compound 4). Compound 4 was prepared with the similar experimental procedure of compound 3, except that the temperature here was 130 °C. The product was obtained through recrystallization with acetone to yield 64%. 1H NMR (CDCl3, ppm): δ = 8.51 (dd, 2H, Ph), 8.50 (d, 2H, Ph), 7.69 (d, 2H, Ph), 5.35 (m, 2H, CH), 5.24 (m, 2H, CH), 4.22 (dd, 2H, CH2), 4.05 (dd, 2H, CH2).
1,4:3,6-Dianhydro-2,5-di-O-(2-trifluoromethyl-4-aminophenyl)-D-mannitol (M3). M3 diammine was prepared via the same procedure for M2, except that the reaction solvent was ethanol and the temperature was 78 °C. After crystallized from ethanol, the final M3 was obtained as white needles in a 90% yield. 1H NMR (DMSO-d6, ppm): δ = 7.05 (d, 2H, Ph), 6.82 (d, 2H, Ph), 6.76 (dd, 2H, Ph), 5.07 (s, 4H, NH2), 4.79 (m, 2H, CH), 4.71 (m, 2H, CH), 4.01 (m, 2H, CH2), 3.70 (m, 2H, CH2). Elemental analysis: found: C, 51.7; H, 3.7; N, 5.6%. Calc. for C20H18F6N2O4: C, 51.7; H, 3.9; N, 6.0%.
1,4:3,6-Dianhydro-2,5-di-O-(3,4-dinitrile-phenyl)-D-mannitol (compound 5). 1.46 g (9.99 mmol) isomannide, 1.68 g (29.94 mmol) KOH, 0.53 g (2.01 mmol) 18-crown-6 and 3.9 g (22.50 mmol) 4-nitrophthalonitrile was dissolved in 70 mL DMF and 12.5 mL toluene. Then the mixture was stirred and refluxed at 140 °C for 12 h. After the mixture cooled down to room temperature, it was added in 400 mL pure water drop by drop and brown solid products were precipitated at the same time. The brown precipitation was recrystallized from acetone and water to afford light yellow solid products. After filtered, it was dried on vacuum for 4 h. Yield: 2.9 g, 73%. 1H NMR (CDCl3, ppm): δ = 7.76 (d, 2H, Ph), 7.39 (d, 2H, Ph), 7.32 (dd, 2H, Ph), 4.99 (d, 2H, CH), 4.89 (q, 2H, CH), 4.12 (dd, 2H, CH2), 4.09 (dd, 2H, CH2).
1,4:3,6-Dianhydro-2,5-di-O-(3,4-dicarboxyphenyl)-D-mannitol dianhydride (M4)36. 2.89 g (7.25 mmol) compound 5 and 5.14 g (90.0 mmol) KOH was dissolved in 80 mL pure water and refluxed for 24 h. Then the mixture was drop into 130 mL cold dilute HCl aqueous to afford white precipitation. The solid product was filtered and dried on vacuum for the next reaction. Yield: 2.3 g, 67%. 1H NMR (CDCl3, ppm): δ = 7.72 (d, 2H, Ph), 7.18 (dd, 2H, Ph), 7.16 (s, 2H, Ph), 5.04 (q, 2H, CH), 4.91 (d, 2H, CH), 4.00 (dd, 2H, CH2), 3.81 (dd, 2H, CH2). Then, 2.21 g (4.66 mmol) of the solid product was dissolved in 15 mL anhydrous THF, and 3 mL acetic anhydride was added into the solution. After reflux at 70 °C for 8 h, the yellow solution was dropped into 250 mL pure water and solid product was precipitated. The crude product was recrystallized from acetone and water to afford light yellow product. The final product M4 was filtered and dried on vacuum. Yield: 1.5 g, 74%. 1H NMR (DMSO-d6, ppm): δ = 8.00 (d, 2H, Ph), 7.75 (d, 2H, Ph), 7.58 (dd, 2H, Ph), 5.22 (dd, 2H, CH), 5.22 (q, 2H, CH), 5.02 (q, 2H, CH), 4.01 (dd, 2H, CH2), 3.88 (dd, 2H, CH2). Elemental analysis: found: C, 60.8; H, 3.5%. Calc. for C24H18O8: C, 60.3; H, 3.2%.

Preparation of biobased PI films

Three kinds of diamine monomers (M1, M2, M3) were reacted with two kinds of dianhydrides, M4 and ODPA, to give the corresponding PIs (Scheme 2). To illustrate the general synthetic route, the synthesis of PI-1 (M1/ODPA) is used as an example. 0.6796 g ODPA (2.19 mmol) was added to a solution of 0.3158 g M1 (2.19 mmol) in 2.53 g NMP. Extra 3.11 g NMP was added to adjust the solid content of the reaction system to 15 wt%. The mixture was stirred at room temperature for about 12 h to form a viscous poly(amic acid) (PAA) solution.
image file: c5ra14526k-s2.tif
Scheme 2 The synthetic routes of all the isomannide-based PIs.

PI-1 films were obtained by thermal imidization of PAA solutions casting onto glass plates, followed by a preheating program (60 °C/1 h, 80 °C/3 h, 120 °C/0.5 h, 150 °C/1 h) and an imidization procedure under vacuum (180 °C/1 h, 200 °C/36 h, 300 °C/1 h) to produce a fully imidized PI film. PI-1 film was stripped through placing the glass plates into hot water.

Results and discussion

Synthesis and characterization of biobased diamine and dianhydride

Three kinds of isomannide-derived diamines, M1, M2 and M3 were designed to investigate the effect of molecular structure of monomers on their polymerizability and physical properties of PIs. Their synthetic routes are illustrated in Scheme 1.

The preparation of M1 monomer was reported by a couple of groups, and three synthetic routes have been summarized by Thiyagarajan et al.37 Yet, there is no best way to synthesize M1. We utilized the strategy involved of tosylation of isomannide, nucleophilic substitution with sodium azide, followed by reduction of diazide. Here, during reduction reaction of diazide, triphenylphosphine, not H2 atmosphere, was used as a reducing agent via the mild Staudinger reaction, to obtain M1 in 73% yield. For two semi-aromatic diamines M2 and M3, they were prepared via the simple two-step reaction through nucleophilic substitution and subsequent reduction of the nitro group into the amino catalyzed with Pd/C. Additionally, in order to further increase the biomass content in final PI materials, a new dianhydride M4 derived from isomannide was prepared (Scheme 1). Its synthetic route is involved with a procedure of nucleophilic substitution of isomannide with 4-nitrophthalonitrile and alkaline hydrolysis of bis(ether dinitrile) followed by dehydration of the corresponding tetra-acid.

The structure of isomannide-based diamines and dianhydride were confirmed by 1H/13C NMR spectroscopy, and elemental analysis (the data in Experimental part). Fig. 1 shows the 1H NMR spectra of all the diamine and dianhydride monomers where all peaks are clearly assigned. The signals of amino groups of M1 and M2 monomers appear at around 4.6 ppm as a singlet, while the corresponding signal of M3 obviously shifts to down field, at 5.0 ppm, because of the strong electro-withdrawing CF3 group in benzene ring of M3. Their 13C NMR spectra are provided in ESI. All the spectra data of monomers are in accordance with their designed structures.


image file: c5ra14526k-f1.tif
Fig. 1 1H NMR spectra of diamines M1–M3 and dianhydride M4 derived from isomannide (DMSO-d6 for M2–M4, and D2O for M1, 400 MHz).

Preparation of biobased PIs

Three resultant diamines, M1–M3, were reacted with M4 and ODPA, respectively, to yield six kinds of biobased PIs as shown in Scheme 2. The biobased PIs were prepared using a two-step procedure through the ring-opening polyaddition to form PAA intermediates, and subsequent thermal imidization. In general, when aliphatic and alicyclic diamines with high basicity were used, the insoluble salts were often formed during the polymerization of PAA, thus leading to a prolong reaction period up to several days, or even completely inhibition of the polymerization.38 However, in our case, when alicyclic diamine M1 was reacted with ODPA or M4 at room temperature, carboxylic acid salts occurred at the very beginning of the reaction, and then the system became homogeneous within few minutes. As a result, the polymerization of M1 only needed 12 hours to obtain its corresponding PAA with relatively high molecular weight. That should be attributed to the presence of ether bonds in the five-membered ring of M1, thus reducing its basicity. In addition, when M3, containing CF3 group, was used to polymerize, we tried different reaction temperature and time, and found that the suitable reaction temperature was below 0 °C, and polymerization time extended up to 24 hours, to yield PAA precursors of high molecular weight.

The inherent viscosities of all PAAs were over than 0.40 dL g−1 measured in NMP solvent at 35 °C as listed in Table 1, indicating that the PAAs with high molecular weight were obtained, thus yielding tough PI films after thermal imidization. Meanwhile, for some of PIs (PI-4, PI-5 and PI-6) soluble in DMF solvent at room temperature, their molecular weights were further determined by GPC measurement, and it was found that they had high molecular weights (Mn) ranging from 2.14 × 104 to 4.04 × 104 g mol−1, with narrow-molecular-weight polystyrene as standards, as well as relatively narrow molecular weight distribution (PDI) between 1.33 and 1.53.

Table 1 Inherent viscosities of all the isomannide-based PAAs and molecular weights of isomannide-based PI films including PI-4, PI-5 and PI-6
PIs PAA/ηinha (dL g−1) Molecular weights of PIsc
Mn (104 g mol−1) Mw (104 g mol−1) PDI
a Measured at polymer concentration of 0.5 g dL−1 in NMP at 35 °C.b Not completely dissolved in DMF.c Measured by GPC in DMF; polystyrene was used as standard.
PI-1 0.68 b b b
PI-2 0.40 b b b
PI-3 0.73 b b b
PI-4 0.71 3.19 4.87 1.53
PI-5 0.82 4.04 5.36 1.33
PI-6 0.40 2.14 3.17 1.48


The chemical structures of biobased PIs were confirmed by 1H NMR and ATR FT-IR spectroscopy. Fig. 2 shows ATR FT-IR spectra of the resultant PIs. All the PIs exhibit characteristic imide absorptions at 1780–1772 and 1720–1710 cm−1, attributed to the asymmetrical and symmetrical stretching vibrations of carbonyl units, respectively. Meanwhile, the characteristic absorptions of the amide group at the region of 3200–3500 cm−1 disappeared. These results clearly indicate that all the PIs have been fully imidized.


image file: c5ra14526k-f2.tif
Fig. 2 ATR FT-IR spectra of all the isomannide-based PI films.

The typical 1H NMR spectrum of PI film is also shown in Fig. 3, where the proton signals of the alicyclic isomannide ring can be seen in the upfield region from 3.7 to 5.2 ppm, while the signals of the phenyl protons appear between 7.4 and 8.0 ppm. Furthermore, the ratio of such two sets of protons matched with the expected structure of PI polymer. In addition, it is mentioned that the biomass isomannide contents of PIs can be tuned by using different monomers from 17–48 wt%, and PI-2 (M1/M4) possesses the highest isomannide content.


image file: c5ra14526k-f3.tif
Fig. 3 1H NMR spectrum of PI-6 polymer (DMSO-d6, 400 MHz).

Morphology of the biobased PIs were also characterized by WXRD, and their X-ray diffraction patterns were presented in Fig. 4. All the isomannide-based PIs possess a certain degree of crystallization, as evidenced by the appearance of two sharp scattering signals in the 2θ range of 14–17°; moreover, these polymers have the same d spacing of 6.23 Å (2θ = 14.2°) and 5.27 Å (2θ = 16.8°). This might be ascribed to the rigid isomannide structure and its ordered arrangement.33 It is noted that the scattering signals of PI-3 is quite weak, and it should be related to its crystallization.


image file: c5ra14526k-f4.tif
Fig. 4 X-ray diffraction profiles of all the isomannide-based PI films.

Solubility of biobased PIs

The solubility of the biobased PIs was measured by dissolving the PI films in various solvents with the same concentration of 10 mg mL−1, and the results are listed in Table 2. Most of the PIs exhibit good solubility in non-proton polar solvents, such as DMF, NMP, DMAc, DMSO and m-cresol at room temperature or upon heating at 100 °C. Among them, DMSO is the best solvent for dissolving the isomannide-based PIs. The good solubility of PIs is mainly ascribed to the presence of flexible ether bonds in polymeric backbone, which can reduce the interaction between polymer chains. Besides, the introduction of bulky CF3 group into isomannide-derived diamines, leading to less efficient chain packing, can further improve the solubility of PIs. Thus, the ability of diamines to enhance the solubility of PIs follows this order: M3 > M2 > M1. As thus, PI-5 polymer composed of M3 and ODPA exhibits the best solubility in our systems, which can be soluble even in nonpolar solvents of low boiling point, such as CH2Cl2, 1,4-dioxane, and pyridine, at room temperature. These results indicate that isomannide-based PIs possess good solution processability, and it is highly desirable for the preparation of polymeric films or coatings in advanced microelectronic applications.
Table 2 Solubility of all the isomannide-based PI filmsa
Solvent PI-1 PI-2 PI-3 PI-4 PI-5 PI-6
a ++, soluble at room temperature; +, soluble upon heating at 100 °C; +−, partially soluble upon heating at 100 °C or the boiling point; −, insoluble upon heating. Solubility was measured with the concentration of 10 mg mL−1 for 24 h.
DMF + +− + ++ ++ ++
DMAc ++ +− ++ ++ ++ ++
DMSO ++ ++ + ++ ++ ++
NMP ++ +− ++ ++ ++ ++
m-Cresol ++ ++ + + ++ ++
THF +− +− +−
CH2Cl2 +− ++ +−
1,4-Dioxane ++
Pyridine +− +− ++ ++


Thermal properties of biobased PIs

For alicyclic PI materials, thermal property is one of the crucial factors to determine their practical applications. Here, we evaluate thermal properties of biobased PI films including Tgs, thermal decomposition, and CTEs by DSC, TGA and TMA, respectively, and their data are summarized in Table 3. Fig. 5 presents DSC second heating scans of PIs.
Table 3 Thermal properties of the isomannide-based PI films
PIs Tga (°C) T5%b (°C) T10%b (°C) Rwc (%) CTEd (ppm °C−1)
a Measured by DSC with 20 °C min−1 in nitrogen.b Temperature at 5% and 10% weight loss determined by TGA with 10 °C min−1 in nitrogen.c Residual weight percentage at 700 °C in nitrogen.d Measured by TMA from 50 to 200 °C with 10 °C min−1.
PI-1 264 440 460 23 58.7
PI-2 228 435 445 2 56.9
PI-3 255 406 439 5 61.1
PI-4 227 416 433 22 50.4
PI-5 246 428 450 13 68.4
PI-6 229 424 440 45 77.7



image file: c5ra14526k-f5.tif
Fig. 5 DSC second heating scan curves of the isomannide-based PI films with 20 °C min−1.

In general, the Tg values of polymers would increase along with the chain rigidity. As shown in Table 3, PI-1 sample composed of alicyclic diamine M1 and dianhydride ODPA, which possesses the most rigid polymer backbone in our system, shows the highest Tg value of 264 °C. However, the PI series with M4 as dianhydride monomer exhibit lower Tgs (about 228 °C) than those of PIs based on ODPA (over than 245 °C), due to the presence of two flexible ether linkages in M4 monomer. Also, it is found that the introduction of ether linkages into diamines plays minor influence in reducing Tgs than those into dianhydrides, which is supported by comparison M1-based PI series, e.g., PI-1(M1/ODPA) with M2-based PIs, e.g., PI-3 (M2/ODPA). What is more, when the same or similar commercial dianhydrides were used, PIs with isomannide-derived diamine M1 possess obviously higher Tgs than those of reported PIs with alicyclic diamines such as 1,3-diaminoadamantance (DAD),39 1,4-diaminocyclohexane (DACH)40 and its derivative 1,4-bis(4-aminophenoxy)cyclohexane (HDA). For examples, in the literature, the Tg values of PIs composed of DACH/biphenyltetracarboxylic dianhydride40 and HDA/ODPA19 were lower than 230 °C. Thus, the incorporation of isomannide unit into PIs is favorable to improve the thermal performance of current PI materials based on petroleum aliphatic diamine. In addition, it is mentioned that there is no observable melting peaks in the DSC heating curves of PIs even when heating up to 390 °C, and the melting temperatures of biobased PIs may be higher than their decomposition temperatures.

The thermal stability of biobased PI films was estimated with the decomposition temperatures at 5% and 10% weight loss (Td,5% and Td,10%) by using TGA in nitrogen, and the data are summarized in Table 3. Obviously, the incorporation of biomass isomannide unit, ether linkage, and CF3 group into PI polymeric chains would affect their thermal decomposition behaviors, thus leading to the two-step decomposition in most of the resultant PIs (Fig. 6). Nevertheless, these PIs still maintain a fairly high thermal stability, as evidenced by their Td,5% values (>400 °C) and Td,10% ones (>430 °C). Among them, PI-1 polymer (M1/ODPA) possesses the best thermal stability, which is accordance with the above DSC results of PIs. The Td data of isomannide-based PIs can be comparative with those of semi-aromatic PIs18,21 and are also high enough for practical applications of less harsh thermal requirement, such as the substrates for flexible display devices. Besides, the weight residues of PIs, except PI-6, are rather low, ranged from 2 to 23% at 700 °C. PI-6 film has high char yield of 45%, it should be due to the presence of CF3 group and high weight content of isomannide unit in this polymer.


image file: c5ra14526k-f6.tif
Fig. 6 TGA curves of the isomannide-based PI films with 10 °C min−1 in nitrogen.

In addition, as listed in Table 3, the CTEs of biobased PIs range from 50.4 to 68.4 ppm °C−1. The CTE value is associated with the rigidity and linearity of the polymer chain. As thus, the PI series based on M3 containing CF3 group exhibit higher CTEs, compared with PIs with monomer M1 or M2, while PI-4 (M2/M4) film shows the lowest CTE value. That is probably because the diamine M2 and dianhydride M4 have the same isomannide unit and their structure is similar, which would be beneficial for the formation of in-plane chain orientation during thermal imidization, thus increasing the linear structure among PI-4 polymer.

Transparency of biobased PIs

Optical transparency of PI films is of more and more importance with the rapid development of advanced optoelectronic technology. Conventional aromatic PI films exhibit deep color from yellow to brown, due to the strong CTC. Fig. 7 shows UV-vis absorption spectra of PI films and the data are presented in Table 4. All of these biobased PI films exhibit very light color or nearly colorless. The cut wavelengths (λ0%) of the isomannide-based PIs with ODPA as dianhydride, including PI-1, PI-3 and PI-5, are shorter than 350 nm, and their transmittances at 450 nm (T450 nm) exceed 80% (Table 4). These results indicate that the isomannide-derived PIs possess good optical transparency. It is also found that the introduction of CF3 group in diamines would improve optical transparency of biobased PIs, in accordance with the researches on fluorinated PI films.7,8 It is reasonable that the CF3 group in the diamines can effectively inhibit the formation of CTC, due to the bulky steric hindrance, as well as its electron-withdrawing effect to weaken the electron-donating ability of diamine units. Additionally, PI series with M4 do not exhibit better optical transparent as expected, although the content of alicyclic unit is higher than that of PI series with ODPA. It may be related to the crystallinity of biobased PIs.
image file: c5ra14526k-f7.tif
Fig. 7 UV-vis absorption spectra of the isomannide-based PI films.
Table 4 Optical and mechanical properties of the isomannide-based PI films
PIs TSa (MPa) EBb (%) TMc (GPa) T450 nmd (%) λ0%e (nm)
a TS, tensile strength.b EB, elongation at break.c TM, tensile modulus.d T450 nm, transmittance at 450 nm.e λ0%, cut-off wavelength.
PI-1 124 8.2 1.3 84 347
PI-2 98 6.5 2.1 68 364
PI-3 184 6.0 1.5 82 350
PI-4 100 6.5 1.9 75 360
PI-5 93 10.0 1.6 85 343
PI-6 112 6.3 2.3 79 357


Mechanical properties of biobased PIs

The mechanical properties of the PI films, including the tensile strength, tensile modulus, as well as elongation at break are summarized in Table 4. All the isomannide-based PIs have good tensile modulus of 1.3–2.3 GPa, and elongation at break of 6.0–10.0%. Especially, these PIs exhibit high tensile strength in the range of 93–184 MPa, while most of PI films reported in the literature were generally between 70 and 100 MPa.1 The outstanding mechanical properties of biobased PIs mainly result from the rigid alicyclic ring of isomannide unit in the polymeric backbone. Thus, the result clearly indicates that the introduction of suitable biomass units into petroleum-based polymers can effectively improve the mechanical performance of the materials.

Conclusions

In conclusion, this work demonstrates the design of a new kind of biobased transparent PIs with good combined properties, which were prepared by directly introducing biomass alicyclic moiety into diamine or dianhydride monomers. Three kinds of isomannide-derived diamines, including aliphatic M1, semi-aromatic M2, and semi-aromatic M3 containing CF3 group, were successfully synthesized, and they showed high polymerizability with a commercial dianhydride ODPA or an isomannide-based dianhydride M4, to yield a series of biobased PI films with a fairly high content of biomass up to 48 wt%. Most of these PIs possessed good optical transparency (T450 nm > 80%), and excellent solution processability (soluble in various non-proton polar solvents at room temperature). Furthermore, because of the incorporation of the rigid isomannide alicyclic ring into diamines, the biobased PIs exhibited improved thermal stability in comparison with fully petroleum-based PIs with aliphatic diamines; at the same time, they possessed outstanding mechanical properties, especially high tensile strength ranging from 93 to 184 MPa. Thereof, the utilization of biomass building blocks to construct biobased PIs is beneficial to develop environmental-friendly and sustainable materials for future engineering and optical applications. Moreover, it can offer a great opportunity to improve the properties of the existing petroleum-based PIs, thus leading to exploit novel high-performance materials.

Acknowledgements

We gratefully acknowledge the financial support from National Basic Research Program of China (no. 2014CB643600), Guangdong Innovative and Entrepreneurial Research Team Program (no. 2013S086), and Natural Science Foundation of Guangdong Province (2014A030312018).

Notes and references

  1. D.-J. Liaw, K.-L. Wang, Y.-C. Huang, K.-R. Lee, J.-Y. Lai and C.-S. Ha, Prog. Polym. Sci., 2012, 37, 907–974 CrossRef CAS PubMed .
  2. M.-C. Choi, Y. Kim and C.-S. Ha, Prog. Polym. Sci., 2008, 33, 581–630 CrossRef CAS PubMed .
  3. H.-J. Ni, J.-G. Liu, Z.-H. Wang and S.-Y. Yang, J. Ind. Eng. Chem., 2015, 28, 16–27 CrossRef CAS PubMed .
  4. M. Hasegawa, M. Horiuchi, K. Kumakura and J. Koyama, Polym. Int., 2014, 63, 486–500 CrossRef CAS PubMed .
  5. A. S. Mathews, I. Kim and C. S. Ha, Macromol. Res., 2007, 15, 114–128 CrossRef CAS .
  6. Y.-Z. Guo, D.-X. Shen, H.-J. Ni, J.-G. Liu and S.-Y. Yang, Prog. Org. Coat., 2013, 76, 768–777 CrossRef CAS PubMed .
  7. C.-Y. Wang, G. Li and J.-M. Jiang, Polymer, 2009, 50, 1709–1716 CrossRef CAS PubMed .
  8. L. Tao, H. Yang, J. Liu, L. Fan and S. Yang, Polymer, 2009, 50, 6009–6018 CrossRef CAS PubMed .
  9. G.-S. Liou, P.-H. Lin, H.-J. Yen, Y.-Y. Yu, T.-W. Tsai and W.-C. Chen, J. Mater. Chem., 2010, 20, 531–536 RSC .
  10. M.-C. Choi, J. Wakita, C.-S. Ha and S. Ando, Macromolecules, 2009, 42, 5112–5120 CrossRef CAS .
  11. N.-H. You, Y. Suzuki, D. Yorifuji, S. Ando and M. Ueda, Macromolecules, 2008, 41, 6361–6366 CrossRef CAS .
  12. R. Lei, L.-X. Gao, R.-Z. Jin and X.-P. Qiu, Chin. J. Polym. Sci., 2014, 32, 941–952 CrossRef CAS .
  13. S. Zhang, Y. Li, T. Ma, J. Zhao, X. Xu, F. Yang and X.-Y. Xiang, Polym. Chem., 2010, 1, 485–493 RSC .
  14. P. K. Tapaswi, M.-C. Choi, K.-M. Jeong, S. Ando and C.-S. Ha, Macromolecules, 2015, 48, 3462–3474 CrossRef CAS .
  15. H. Suzuki, T. Abe, K. Takaishi, M. Narita and F. Hamada, J. Polym. Sci., Part A: Polym. Chem., 2000, 38, 108–116 CrossRef CAS .
  16. D. J. Liaw, C. C. Huang and W. H. Chen, Macromol. Chem. Phys., 2006, 207, 434–443 CrossRef CAS PubMed .
  17. T. Matsumoto and T. Kurosaki, Macromolecules, 1997, 30, 993–1000 CrossRef CAS .
  18. M. Hasegawa, M. Fujii, J. Ishii, S. Yamaguchi, E. Takezawa, T. Kagayama and A. Ishikawa, Polymer, 2014, 55, 4693–4708 CrossRef CAS PubMed .
  19. Y. Hou, G. Chen, X. Pei and X. Fang, J. Polym. Res., 2012, 19, 9955 CrossRef .
  20. Y. Zhou, G. Chen, H. Zhao, L. Song and X. Fang, RSC Adv., 2015, 5, 53926–53934 RSC .
  21. Y. Watanabe, Y. Shibasaki, S. Ando and M. Ueda, J. Polym. Sci., Part A: Polym. Chem., 2004, 42, 144–150 CrossRef CAS PubMed .
  22. P. K. Tapaswi, M.-C. Choi, Y. S. Jung, H. J. Cho, D. J. Seo and C.-S. Ha, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 2316–2328 CrossRef CAS PubMed .
  23. K. Yao and C. Tang, Macromolecules, 2013, 46, 1689–1712 CrossRef CAS .
  24. C. Vilela, A. F. Sousa, A. C. Fonseca, A. C. Serra, J. F. J. Coelho, C. S. R. Freire and A. J. D. Silvestre, Polym. Chem., 2014, 5, 3119–3141 RSC .
  25. F. Fenouillot, A. Rousseau, G. Colomines, R. Saint-Loup and J. P. Pascault, Prog. Polym. Sci., 2010, 35, 578–622 CrossRef CAS PubMed .
  26. X. Feng, A. J. East, W. B. Hammond, Y. Zhang and M. Jaffe, Polym. Adv. Technol., 2011, 22, 139–150 CrossRef CAS PubMed .
  27. A. Wroblewska, A. Zych, S. Thiyagarajan, D. Dudenko, D. van Es, M. R. Hansen, C. Koning, R. Duchateau and L. Jasinska-Walc, Polym. Chem., 2015, 6, 4133–4143 RSC .
  28. J. Wu, P. Eduard, L. Jasinska-Walc, A. Rozanski, B. A. J. Noordover, D. S. van Es and C. E. Koning, Macromolecules, 2013, 46, 384–394 CrossRef CAS .
  29. L. Feng, W. Zhu, C. Li, G. Guan, D. Zhang, Y. Xiao and L. Zheng, Polym. Chem., 2015, 6, 633–642 RSC .
  30. J. Yan, X. Ji, Z. Wang, J. Liu, W. Fan and X. Meng, Chinese Pat., CN104231268A, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 2014 .
  31. J. S. Mathew, S. P. Vernekar, R. Mercier and R. Kerbour, US Pat., 2002142110-A1, Council of Scientific and Industrial Research, New Delhi, 2002 .
  32. H. Ren, Y. Guan, Y. He, H. He and X. Wang, Acta Polym. Sin., 2006, 248–252 CrossRef CAS .
  33. L. Jasinska, M. Villani, J. Wu, D. van Es, E. Klop, S. Rastogi and C. E. Koning, Macromolecules, 2011, 44, 3458–3466 CrossRef CAS .
  34. E. Anslyn, K. J. Wallace, R. Hanes, J. Morey, K. V. Kilway and J. Siegel, Synthesis, 2005, 2080–2083 CrossRef .
  35. A. A. Caouthar, A. Loupy, M. Bortolussi, J. C. Blais, L. Dubreucq and A. Meddour, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 6480–6491 CrossRef CAS PubMed .
  36. X. Q. Liu, M. Jikei and M. Kakimoto, Macromolecules, 2001, 34, 3146–3154 CrossRef CAS .
  37. S. Thiyagarajan, L. Gootjes, W. Vogelzang, J. Wu, J. van Haveren and D. S. van Es, Tetrahedron, 2011, 67, 383–389 CrossRef CAS PubMed .
  38. M. Hasegawa, M. Horiuchi and Y. Wada, High Perform. Polym., 2007, 19, 175–193 CrossRef CAS PubMed .
  39. Y.-T. Chern and W.-h. Chung, J. Polym. Sci., Part A: Polym. Chem., 1996, 34, 117–124 CrossRef CAS .
  40. Q. Jin, T. Yamashita, K. Horie, R. Yokota and I. Mita, J. Polym. Sci., Part A: Polym. Chem., 1993, 31, 2345–2351 CrossRef CAS PubMed .

Footnote

Electronic supplementary information (ESI) available: 13C NMR spectra of four isomannide-based monomers M1–M4, and 1H NMR spectra of all the intermediates of monomers. See DOI: 10.1039/c5ra14526k

This journal is © The Royal Society of Chemistry 2015
Click here to see how this site uses Cookies. View our privacy policy here.