A biobased photocurable binder for composites with transparency and thermal stability from biomass-derived isosorbide

Seunghan Shin*ab, Byung-Chul Kimac, Eugene Changa, Jin Ku Chobd and Dong Hack Suhc
aIT Convergence Materials R&D group, Korea Institute of Industrial Technology (KITECH), 89 Yangdaegiro-gil, Ipjang-myeon, Cheonan, 331-822, Republic of Korea. E-mail: shshin@kitech.re.kr
bDepartment of Green Process and System Engineering, University of Science and Technology (UST), 89 Yangdaegiro-gil, Ipjang-myeon, Cheonan, 331-822, Republic of Korea
cDepartment of Chemical Engineering, Hanyang University, Seoul 133-010, Republic of Korea
dGreen Process and Materials R&D group, Korea Institute of Industrial Technology (KITECH), 89 Yangdaegiro-gil, Ipjang-myeon, Cheonan, 331-822, Republic of Korea

Received 4th December 2013 , Accepted 19th December 2013

First published on 20th December 2013


Abstract

A biobased photocurable binder was synthesized from isosorbide in a high yield (91%) and used as a binder for transparent glass fabric composites. The photocurable isosorbide binder showed a lower refractive index (n = 1.489) than that of glass (n = 1.560) and low viscosity (48.6 cP at 25 °C) owing to the wedge-shaped fused ring structure of the isosorbide moiety. Transparent glass fabric composites were prepared by refractive index matching using a cardo-type fluorene based binder (n = 1.583) as a co-binder. A glass fabric composite photocured with equal wt% of the cardo- and isosorbide-based binder showed 85% of transmittance at 550 nm by UV-visible spectroscopy. It also showed a considerably good coefficient of thermal expansion (α1 = 16.3 ppm K−1) and glass transition temperature (142 °C by DMA). As the wt% of the isosorbide-based binder increased, the network structure of the binder mixture became tight to give a higher glass transition temperature but the composite's transparency was decreased.


1. Introduction

1,4:3,6-Dianhydrohexitols obtained from biomass-derived carbohydrates attract much attention owing to their potential applications in polymers, pharmaceuticals, and cosmetics.1–3 They are divided into three stereoisomers corresponding to isosorbide, isomannide, and isoidide, respectively, prepared from glucose, mannose, and idose via reduction of the aldehyde group and subsequent double dehydration.

Taking the issues on carbon footprint and depletion of petroleum reservoirs into account, isosorbide is one of the most commercially satisfactory compounds to replace petroleum-based compounds because its starting material, glucose is originated from the most abundant biomass resource on Earth. In addition, isosorbide is the only biobased diol that improves resistance to heat, UV rays and chemicals, and offers excellent optical and mechanical properties thanks to its rigid structure. With this regard, isosorbide has been applied as a monomer for transparent polymers like polyester and polycarbonate.4–7

In recent years, several researches were performed to develop photocurable isosorbide derivatives for radical polymerization.8–12 Łukaszczyk et al. synthesized 2,5-bis(2-hydroxy-3-methacryloyloxypropoxy)-1,4:3,6-dianhydrosorbitol (ISD-GMA) as a feasible substituent of 2,2-bis[4-(2-hydroxy-3-methacryloyloxypropoxy)phenyl]propane (BIS-GMA), a famous dental sealant, and showed it had low viscosity and comparable mechanical properties to BIS-GMA.11 Sadler et al. used isosorbide methacrylate as a biobased low viscosity resin for high performance thermosetting applications.13

In this study, we made a first attempt to apply photocurable isosorbide derivatives to optically transparent films reminding that isosorbide can give excellent thermal and optical properties. Among many fabrication methods for thermally stable transparent films, we used glass fibre composite system and refractive index (RI) matching method. According to RI matching method suggested by Stoffer et al., transparent glass fibre composites can be obtained if RI difference between glass fibre and matrix resin is less than 0.002.14,15 Therefore we synthesized isosorbide dimetharcylate (ISDM) as a matrix resin for transparent glass fibre composites and used cardo-type fluorene based acrylate (CD) for RI matching.

CD was used for not only RI matching with glass fibre but also improving thermal stability. Structural feature of CD having 9,9-diaryl fluorene moieties of which the fluorene unit and aryl groups are positioned similarly to body and wings of butterfly (this is called cardo structure) allows low optical anisotropy as well as high heat resistance. Although CD is very limited to use due to high viscosity,16,17 we expected that V-shaped rigid structure of isosorbide moiety in ISDM could alleviate high viscosity of CD without sacrificing thermal and optical properties of final materials by mixing.

Basic properties of ISDM including RI, viscosity, and thermal properties were investigated. Various composition of binder mixtures with CD were prepared and screened to obtain an optimized transparent glass fibre composite. Glass fibre composites' thermal properties as well as transparencies were also measured and discussed in a viewpoint of ISDM addition.

2. Experimental

Materials

Isosorbide (1,4:3,6-dianhydroglucitol, 98%), methacrylic anhydride (94%), triethylamine (TEA, 99.5%), and 4-dimethylaminopyridine (DMAP, 99%) were purchased from Sigma-Aldrich (St. Louis, MO, USA) and used for synthesizing isosorbide dimethacrylates without further purification. Solvents including hexane, ethyl acetate (EA), dichloromethane (DCM, anhydrous), and tetrahydrofuran (THF, anhydrous) were all reagent grades and purchased from TCI Inc. (Japan). CD was supplied by Miwon Co. (South Korea) as a trade name of MIRAMER HR-6060® and used as received. Its chemical structure was displayed in Fig. 1.
image file: c3ra47287f-f1.tif
Fig. 1 Chemical structure of cardo-type fluorene based diacrylates (n + m = 2 and its refractive index is 1.583).

2,4,6-Trimethylbenzoylphosphine oxide (Darocur TPO) as a photoinitiator, was obtained from BASF Schweiz AG (Switzerland). E-glass fabric supplied by Unitika (Japan) has a plain woven pattern and 70 μm of thickness.

Measurements

Structure of ISDM was analysed with 1H and 13C-NMR (Bruker-400 MHz, Bruker-100 MHz), and mass spectrometers (JEOL, JMS-600W). RI was measured by Abbe refractometer (ATAGO, 4T(1240)). Viscosity was determined using a rotational viscometer (Brookfield, LVDVII+) with cone spindle (CPE-40). Transparencies of composites were evaluated by measuring their transmittance at 550 nm with UV-Vis spectroscopy (Varian, CARY-100). In order to measure the degree of conversion of photocurable binders, attenuated total reflectance (ATR) analysis was performed using FTIR (Perkin Elmer, Spectrum 100) equipped with single reflection ATR accessory. Diamond (n = 2.4) was used as ATR crystal and incident angle was 42°.

Dynamic scanning colorimeter (DSC, Perkin-Elmer, DSC8000), thermomechanical analyser (TMA, TA Instruments, Q400), and dynamic mechanical analyser (DMA, TA Instruments, DMA Q800) were used to evaluate the thermal properties of binders and composites. DSC analysis was performed in the temperature range of 25–300 °C with a 10 K min−1 ramp in N2 atmosphere. TMA measurements were carried out in tension mode with 0.03 N of initial force and a 5 K min−1 ramp. The sample size was 16 × 4 × 0.1 mm3 (length × width × thickness). Meanwhile, DMA measurements were carried out on rectangular specimens (17.3 × 5.3 × 0.1 mm3, length × width × thickness) in flexural mode at 1 Hz. DMA spectra were recorded in the temperature range 20–300 °C with a 5 K min−1 ramp.

Synthesis of isosorbide dimethacrylate (ISDM)

5 g (34.21 mmol, 1.0 eq.) of isosorbide and 0.63 g (5.13 mmol, 0.15 eq.) of DMAP were dissolved in anhydrous DCM under N2 atmosphere. 15.74 ml (112.91 mmol, 3.3 eq.) of TEA was added to the mixture with stirring and 11.21 ml (75.27 mmol, 2.2 eq.) of methacrylic anhydride was added dropwise with ice bath. The mixture was stirred overnight and then washed with DI water to remove DMAP and TEA. DCM was removed by rotary evaporation and the crude product was purified by column chromatography (hexane[thin space (1/6-em)]:[thin space (1/6-em)]ethyl acetate = 4[thin space (1/6-em)]:[thin space (1/6-em)]1) on silica to afford ISDM (8.77 g, 31.1 mmol, 90.8%) as a clear oil. 1H-NMR (400 MHz, CDCl3): δ = 6.15 (brs, 1H), 6.09 (brs, 1H), 5.60 (t, J = 1.6 Hz, 1H), 5.58 (t, J = 1.6 Hz, 1H), 5.24 (brs, 1H), 5.19 (q, J = 5.5 Hz, 1H), 4.89 (t, J = 5.0 Hz, 1H) 4.52 (d, J = 5.2 Hz, 1H), 4.0–3.97 (m, 2H), 3.96 (dd, J = 10.0, 6.0 Hz, 1H), 3.88 (dd, J = 10.0, 5.2 1H), 1.95 (s, 3H), 1.91(s, 3H). 13C-NMR (100 MHz, CDCl3): δ = 166.9, 166.6, 135.9, 135.8, 126.7, 126.5, 86.2, 82.1, 78.4, 74.4, 73.6, 70.8, 18.5, 18.4. Mass spectrum (ESI) m/z calcd for C14H19O6 [M + H]+ 283.1, found 283.1.

Preparation of transparent composites

Photocurable binders for transparent composite fabrication were prepared by mixing ISDM and CD. Their relative content was controlled by referring the transmittance of composites. 2.5 wt% of photoinitiator was added to the binder mixture for photocuring. In order to make a good interface between glass fabric and photocurable binder, E-glass fabric was surface treated with γ-methacryloxypropyltrimethoxy silane (γ-MPS, Dow chemical, USA) according to the ref. 18. The silanized glass fabric soaked in a photocurable binder mixture was transferred between glass-plates separated with 100 μm of spacer and photocured by irradiating 20 J cm−2 (10 J cm−2 per each side) using UV curing apparatus (Lichtzen, LZ-I404, South Korea) equipped with high-pressure Hg lamp.

3. Results and discussion

Preparation and characterization of isosorbide based dimethacrylate (ISDM)

ISDM was synthesized from isosorbide according to Scheme 1. Isosorbide is a biobased chemical obtained from glucose via sorbitol, which is recently highlighted in the aspect of a renewable and sustainable chemical to replace petroleum-based chemicals.
image file: c3ra47287f-s1.tif
Scheme 1 Synthetic scheme of isosorbide dimethacrylate.

Methacrylic functional groups of ISDM were introduced via ester linkage with the hydroxyl groups of isosorbide. Since direct esterification with isosorbide and methacrylic acid is hardly proceeded due to the low reactivity of secondary alcohols at C2 and C5 sites of isosorbide (especially, the hydroxyl group at C2 site is much less reactive than that at C5 site because of internal hydrogen bonding with oxygen atom of the ether ring), esterification of isosorbide was attempted by using methacrylic anhydride in the presence of a catalytic amount of DMAP. Considering the solubility of isosorbide and stability of methacrylic anhydride, DCM and THF were tested as a solvent. Isosorbide was more soluble in THF than in DCM (approximately 4 times) but the yield of ISDM in DCM was greater than in THF (91% for DCM vs.60% for THF). It is believed that DCM could efficiently suppress the undesired breakage of methacrylic anhydride into methacrylic acid.

Basic properties of newly synthesized ISDM such as RI, viscosity, and thermal property were investigated to evaluate ISDM as an optical binder. The RI, n of ISDM measured by Abbe refractometer was 1.489. This is a reasonable value considering that n of (meth)acrylate binders are mainly dependent on their backbone moieties. That is, RIs of (meth)acrylate binders bearing linear aliphatic backbone moieties are less than 1.44, whereas RIs of binders having aromatic backbone moieties are larger than 1.51.

Viscosity of ISDM measured by rotational viscometer with cone spindle was 48.6 cP at 25 °C. Considering that viscosity is governed by the strength of intermolecular forces and the shapes of the molecules, wedge-shaped rigid isosorbide moiety can induce low viscosity.

The glass transition temperature (Tg) of cured ISDM was measured by DSC. ISDM was mixed with 2.5 wt% of Darocur TPO and photocured by irradiating 20 J cm−2 of UV dose. Ten mg of cured ISDM film was moved to DSC cell and its DSC thermogram was plotted at a rate of 5 K min−1. Fig. 2 shows DSC thermogram of ISDM film obtained by 2nd scan and its glass transition was observed at around 142 °C. Considering Tgs of methacrylates are dependent on their carbon number, chain structures, and functionality, high Tg of cured ISDM should be originated from two methacrylates directly attached to a rigid isosorbide core without flexible spacers. As the curing reaction proceeded, ISDM developed a cyclic chain linkage between crosslink points that restrained molecular motions. According to Mays et al.,19 cyclic chain can induce a higher Tg than linear chain, even if the carbon number is the same.


image file: c3ra47287f-f2.tif
Fig. 2 TMA curve of glass fabric reinforced composite prepared with ISDM binder and DSC thermogram of cured ISDM film.

Transparent composites prepared with ISDM

To achieve high optical transparency of a composite, various binder mixtures were prepared by controlling the relative amount of ISDM (n = 1.489) to CD (n = 1.583) for RI matching with glass fibre (n = 1.560 for E-glass). As shown in Fig. 3, RI of a binder mixture was linearly decreased with increasing the ISDM content. Since RI of a photocurable binder increases after curing reaction, it is necessary to measure the RI change of a binder mixture after photocuring. The average difference of RI caused by photocuring at 20 J cm−2 of UV dose was 0.025. Therefore RI of a binder mixture was adjusted to around 1.53. From the Fig. 3, binder mixtures containing 40, 50, 60 wt% of ISDM respectively were prepared and evaluated as matrix resins.
image file: c3ra47287f-f3.tif
Fig. 3 Refractive index changes of CD and ISDM binder mixture as a function of ISDM content (circle: uncured, cross: cured with a dose of 20 J cm−2).

Fig. 4 shows UV-visible transmittance of E-glass fabric reinforced composites prepared with three different binder mixtures. Transmittance of a composite film was improved as expected with RI matching. When the equal amount of CD and ISDM (50 wt% of ISDM) was used, composite film showed 85% of optical transparency at 550 nm. RI of this binder mixture was 1.536 and increased to 1.561 after curing, which resulted in a good transparency.


image file: c3ra47287f-f4.tif
Fig. 4 Transmittance of glass fabric reinforced composites with different binder composition.

Meanwhile, viscosity of binder mixture was remarkably decreased by addition of ISDM (see Fig. 5). This is because wedge-shaped fused ring structure of isosorbide in ISDM is likely to hinder the interaction of fluorene moieties in CD. Actually, CD shows high viscosity due to its bulky 9,9-diaryl fluorene moieties and interaction between fluorene moieties (in case of HR6060, its viscosity is 85[thin space (1/6-em)]000 cP at 25 °C). As the ISDM content increased to 50 wt%, viscosity of binder mixture decreased to 2421 cP at 25 °C and this was suitable for a prepreg fabrication.


image file: c3ra47287f-f5.tif
Fig. 5 Viscosity change of binder mixtures as a function of ISDM content (viscosity of CD was 85[thin space (1/6-em)]000 cP at 25 °C).

Thermal properties of transparent composites

Coefficient of thermal expansion (CTE, α) of a glass fabric composite was measured by TMA in tension mode. As composite's xy plane CTE was known to be seriously dependent on the wt% of inorganic reinforcement,20 glass fabric content was carefully controlled to be 56.0 ± 1.7 wt%. TMA curves showed that CTEs below the Tg (α1) were similar (about 16.3 ppm K−1) irrespective of ISDM content, however CTEs above the Tg (α2) were varied with ISDM content (see Fig. 6).
image file: c3ra47287f-f6.tif
Fig. 6 TMA curves of glass fabric reinforced composites prepared with different binder compositions (TMA curves were measured at 2nd scan in tension mode at a heating rate of 5 K min−1).

This means α2 is more dependent on matrix resin properties than α1. Since matrix resin modulus decreases rapidly after its Tg, composite modulus in z-direction is expected to decrease likewise. This leads to large expansion in z-direction and then relatively small expansion in xy plane. Consequently, TMA data indicate matrix resin modulus above Tg decreases with lowered ISDM content.

To confirm the matrix resin modulus as a function of temperature, dynamic mechanical analysis (DMA) was performed and the results were displayed in Fig. 7. DMA curves show storage moduli (E′) of composites at rubbery state decreased as ISDM content decreased. Using the concept of rubber elasticity,21,22 crosslink density (νe) can be calculated using following eqn (1).

 
νe = G/RT = E/3RT, (1)
where the storage modulus values, G′ or E′ are obtained in the rubbery plateau, T is temperature in degrees K corresponding to the storage modulus value, and R is the gas constant. The calculated νe was 0.380 mol cm−3 for ISDM 40 wt% binder, 0.412 mol cm−3 for 50 wt%, and 0.488 mol cm−3 for 60 wt%, respectively. Taking into account of small molecular weight of ISDM (282.1 for ISDM vs. 546.2 for CD), total number of photocurable groups in binder mixture increases as wt% of ISDM increases. The number of total photocurable groups was calculated as 25.2 mmol for ISDM 40 wt%, 26.9 mmol for ISDM 50 wt%, and 28.6 mmol for ISDM 60 wt%, respectively. Therefore, many photocurable groups are one of the reasons for high crosslink density. In addition, tight network structure developed by ISDM, as previously mentioned, is also believed to be another reason.


image file: c3ra47287f-f7.tif
Fig. 7 DMA curves of glass fabric reinforced composites prepared with different binder mixture compositions (DMA spectra were measured in flexural mode at 1 Hz at a heating rate of 5 K min−1).

Loss tangent (tan[thin space (1/6-em)]δ) curves as a function of temperature showed broad and low shapes (also displayed in Fig. 7). They became broader and their peak points (considered as Tg) shifted to the higher temperature with increasing wt% of ISDM. Since broad and low tan[thin space (1/6-em)]δ (Fig. 7 shows tan[thin space (1/6-em)]δ is less than 0.1) means composites have less damping characteristics by restriction of soft segment mobility, it can be known that increase of ISDM induces restriction of segmental motion not only by increasing crosslink density but also by decreasing the soft segments belonged to CD.

In order to know the effect of ISDM on curing reaction, degree of double bond conversion (DC) was measured by FTIR-ATR method and calculated by eqn (2).

 
image file: c3ra47287f-t1.tif(2)

As given in Fig. 8, monitoring peak for double bond conversion calculation was 1635 cm−1 (assigned to C[double bond, length as m-dash]C stretching of (meth)acrylate) and reference peak was 1720 cm−1 (assigned to C[double bond, length as m-dash]O stretching of (meth)acrylate), respectively. The calculated degree of conversion is 57.9–59.7% and this means degree of conversion is not significantly affected by the increase of ISDM content under our curing conditions although methacrylates have lower reactivity rather than acrylates.


image file: c3ra47287f-f8.tif
Fig. 8 FTIR-ATR spectra of uncured and cured glass fabric reinforced composite; degree of conversion was calculated using peak at 1635 cm−1 as a monitoring peak and peak at 1720 cm−1 as an internal standard.

4. Conclusions

Biobased isosorbide dimethacrylate (ISDM) was synthesized in 91% of yield and used with highly viscous and thermally stable cardo-type fluorene photocurable binder (CD) for the fabrication of transparent glass fabric composites. RI and viscosity (at 25 °C) of ISDM were 1.489 and 48.6 cP, respectively. Photocurable binder mixture composed of ISDM and CD in 1[thin space (1/6-em)]:[thin space (1/6-em)]1 by wt% was applicable to a transparent glass fabric composite. The transparent glass fabric composite showed 85% transmittance at 550 nm and considerably good α1 (16.3 ppm K−1) and Tg (142 °C measured by DMA). Composites showed increased storage modulus (E′) at rubbery state and Tg as wt% of ISDM increased in binder mixture. This is because ISDM induced tight network structure as well as high crosslink density by a rigid isosorbide directly connected to methacrylate and increment of total number of photocurable groups in binder mixture.

Acknowledgements

This research was supported by grants from the Industrial Strategic Technology Development Program (Project no. 10045475) and Cooperative Research Program funded by the Ministry of Trade, Industry & Energy (MOTIE) of Korea.

Notes and references

  1. P. Stoss and R. Hemmer, Adv. Carbohydr. Chem. Biochem., 1991, 49, 93 CrossRef CAS.
  2. F. Fenouillot, A. Rousseau, G. Colomines, R. Saint-Loup and J.-P. Pascault, Prog. Polym. Sci., 2010, 35, 578 CrossRef CAS PubMed.
  3. M. Rose and R. Palkovits, ChemSusChem, 2012, 5, 167 CrossRef CAS PubMed.
  4. J. Wu, P. Eduard, S. Thiyagarajan, L. Jasinsk-Walc, A. Rozanski, C. F. Guerra, B. A. J. Noordover, J. Van Haveren, D. S. van Es and C. E. Koning, Macromolecules, 2012, 45, 5069 CrossRef CAS.
  5. J. Wu, P. Eduard, L. Jasinsk-Walc, A. Rozanski, B. A. J. Noordover, D. S. van Es and C. E. Koning, Macromolecules, 2013, 46, 384 CrossRef CAS.
  6. W. J. Yoon, K. S. Oh, J. M. Koo, J. R. Kim, K. J. Lee and S. S. Im, Macromolecules, 2013, 46, 2930 CrossRef CAS.
  7. W. J. Yoon, S. Y. Hwang, J. M. Koo, Y. J. Lee, S. U. Lee and S. S. Im, Macromolecules, 2013, 46, 7219 CrossRef CAS.
  8. A. Bratov, J. Muñoz, C. Dominguez and J. Bartrolí, Sens. Actuators, B, 1995, 25, 823 CrossRef CAS.
  9. J. Ge, M. Trujillo and J. Stansbury, Dent. Mater., 2005, 21, 1163 CrossRef CAS PubMed.
  10. F. Pion, A. F. Reano, P.-H. Ducrot and F. Allais, RSC Adv., 2013, 3, 8988 RSC.
  11. J. Łukaszczyk, B. Janicki and A. Frick, J. Mater. Sci.: Mater. Med., 2013, 23, 1149 CrossRef PubMed.
  12. J. Łukaszczyk, B. Janicki, J. Kozuch and H. Wojdyla, J. Appl. Polym. Sci., 2013, 130, 2514 CrossRef.
  13. J. M. Sadler, A. T. Nguyen, F. R. Toulan, J. P. Szabo, G. R. Palmese, C. Scheck, S. Lutgen and J. J. Scala, J. Mater. Chem. A, 2013, 1, 12579 CAS.
  14. J. R. Olson, D. E. Day and J. O. Stoffer, J. Compos. Mater., 1992, 26, 1181 CrossRef CAS PubMed.
  15. K. D. Weaver, J. O. Stoffer and D. E. Day, Polym. Compos., 1995, 16, 161 CrossRef CAS.
  16. S. Shibahara, Y. Shimobe and H. Kuramoto, US Patent No. 7250209, 2007.
  17. J. G. Han, C. S. Kang, S. K. Kim, Y. S. Suh and K. K. Yoon, US Patent Pub. No. 2011–0117498, 2011.
  18. Y. Hirai, H. Hamada and J.-K. Kim, Compos. Sci. Technol., 1998, 58, 91 CrossRef CAS.
  19. J. W. Mays, E. Siakali-Kioulafa and N. Hadjichristidis, Macromolecules, 1990, 23, 3530 CrossRef CAS.
  20. M. Nogi, S. Ifuku, K. Abe, K. Handa, A. N. Nakagaito and H. Yano, Appl. Phys. Lett., 2006, 88, 133124 CrossRef PubMed.
  21. J.-P. Pascault, H. Sautereau, J. Verdu, and R. J. J. Williams, Basic physical properties of networks, in thermosetting polymers, Marcel Dekker Inc., New York, 2002, p. 282 Search PubMed.
  22. E. K. Reed, Polym. Compos., 1980, 1, 44 CrossRef.

This journal is © The Royal Society of Chemistry 2014
Click here to see how this site uses Cookies. View our privacy policy here.