Article pubs.acs.org/Biomac
Thermal Mobility of β-O-4-Type Artificial Lignin Yasumitsu Uraki,*,† Yusuke Sugiyama,‡,# Keiichi Koda,† Satoshi Kubo,§ Takao Kishimoto,∥ and John F. Kadla⊥ †
Research Faculty of Agriculture and ‡Graduate School of Agriculture, Hokkaido University, Sapporo 060-8589, Japan § Department of Biomass Chemistry, Forestry and Forest Products Research Institute, Tsukuba 305-8687, Japan ∥ Biotechnology, Toyama Prefectural University, Imizu 939-0398, Japan ⊥ Advanced Biomaterials Laboratory, University of British Columbia, Vancouver, BC Canada, V6T1Z4 S Supporting Information *
ABSTRACT: Several lignin model polymers and their derivatives comprised exclusively of β-O-4 or 8-O-4′ interunitary linkages were synthesized to better understand the relation between the thermal mobility of lignin, in particular, thermal fusibility and its chemical structure; an area of critical importance with respect to the biorefining of woody biomass and the future forest products industry. The phenylethane (C6−C2)-type lignin model (polymer 1) exhibited thermal fusibility, transforming into the rubbery/ liquid phase upon exposure to increasing temperature, whereas the phenylpropane (C6−C3)-type model (polymer 2) did not, forming a char at higher temperature. However, modifying the Cγ or 9-carbon in polymer 2 to the corresponding ethyl ester or acetate derivative imparted thermal fusibility into this previously infusible polymer. FT-IR analyses confirmed differences in hydrogen bonding between the two model lignins. Both polymers had weak intramolecular hydrogen bonds, but polymer 2 exhibited stronger intermolecular hydrogen bonding involving the Cγhydroxyl group. This intermolecular interaction is responsible for suppressing the thermal mobility of the C6−C3-type model, resulting in the observed infusibility and charring at high temperatures. In fact, the Cγ-hydroxyl group and the corresponding intermolecular hydrogen bonding interactions likely play a dominant role in the infusibility of most native lignins.
■
INTRODUCTION Lignin is a natural, renewable biopolymer second only to cellulose in natural abundance. A multifunctional aromatic biopolymer, lignin serves as a continuous matrix within plant cell walls, providing mechanical strength and structural support.1 Lignin biosynthesis involves enzymatic radical coupling of coniferyl and sinapyl alcohol monomers, termed monolignols, connected mainly through β-O-4 (8-O-4′) interunitary linkages, as well as other minor linkages, such as β-β (8−8′), β-5 (8−5′), 5−5, and so forth.2,3 As a result, native lignins possess a complex three-dimensional network structure. However, lignin cannot be isolated in its native state; all lignin isolation procedures result in extensive lignin degradation and modification. Of the various isolation methods, milled wood lignin (MWL) is the most widely utilized, as it contains many of the same interunitary linkages as that observed in wood.4,5 An enormous amount of lignin is produced as a coproduct of “wood-free” papermaking, where it is primarily burnt as an energy source and as part of a complex chemical recovery system.6 In fact, less than 2% of technical lignins are utilized for value-added products, traditionally as stabilizers for plastics and rubber, as well as in the formulation of dispersants and surfactants.7 Although this can be attributed to its role in chemical recovery, much of its use has been dictated by its complex irregular macromolecular properties. However, as part © 2012 American Chemical Society
of emerging wood-to-ethanol biorefinery platforms lignin utilization is becoming more important. Here, value-added lignin-based materials and chemicals are needed to offset the low cost of the commodity fuel. There are numerous reports on lignin utilization.7−13 Of particular interest is the conversion of lignin to thermoplastic and fusible materials,14−25 which can be readily processed using conventional thermal and melt processing technologies. Such fusible lignins are being investigated for advanced materials like carbon and activated carbon fibers,25−33 as well as hot-melttype adhesives34,35 and biodegradable plastics.15,36−40 However, most technical lignins do not exhibit thermal fusibility due to native lignin structures and methods of biomass separation,19 as well as which affect lignin solubility and chemical reactivity, although all isolated lignins undergo a glass transition. Typically, technical lignins have been derivatized to enhance their thermal mobility and plasticity. Glasser et al. developed lignin derivatives with melt-flow properties through the introduction of a long alkyl into the lignins.41 Likewise, Li and Sarkanen,21 as well as Kadla and Kubo,17 have produced Kraft lignin-based plastics through the disruption of interReceived: December 15, 2011 Revised: February 10, 2012 Published: February 17, 2012 867
dx.doi.org/10.1021/bm201772v | Biomacromolecules 2012, 13, 867−872
Biomacromolecules
Article
Figure 1. Synthetic route of a C6−C2-type polymeric lignin model (route A; polymer 1) and a C6−C3-type polymeric lignin model (route B; polymer 2). 71.4 (a), 83.6 (b), 111.7 (2), 115.0 (5), 119.1 (6), 133.1 (1′), 135.0 (1), 145.2 (4′), 146.7 (4), 148.9 (3); threo: δ 70.8 (α), 84.4 (β), 111.3 (2), 114.5 (5), 118.7 (6), 134.7 (1), 147.0 (4); 1′,4′: phenolic end units. The erythro/threo ratio of polymer 2 was 4 to 1. The absorption coefficients (E) of this polymer at 280 nm in 90% dioxane and 50% aqueous methylsolv were 14.8 and 15.0 dm3 g−1 cm−1, repectively, and the polymer did not show bathochromic effect. The polymers were acetylated with acetic anhydride in pyridine. A milled wood lignin (MWL) was prepared from Todo-fir (Abies sachalinensis) according to the Bjorkman method.30 Measurement of Molecular Mass. The molecular mass of the acetylated lignin model polymers were determined using high performance−size exclusion chromatography (HP-SEC) and endgroup determination by 1H NMR. The HP-SEC was performed at 40 °C using Shodex KF-802 and KF- 803 columns and UV detection at 280 nm. The eluting solvent was THF, and the flow rate was 0.5 mL/ min. The HP-SEC was calibrated using standard polystyrene samples with a molecular mass range from 580 Da to 106 KDa. The number-average molecular mass (Mn) of polymer 2 acetate was also calculated using the average area ratio of the proton signals for the acetate methyl protons of the aliphatic (Cα and Cγ positions at 1.8 − 2.2 ppm) and end-group aromatic acetyl groups (∼2.4 ppm) relative to the corresponding methylene protons at Cα (6.6−7.2 ppm), Cγ (4.5−4.9 ppm), methoxyl protons (3.6−4.0 ppm), and the methyne proton at Cβ (5.8−6.2 ppm). In the case of polymer 3, the proton signal at 12−12.2 ppm, assigned to the phenolic hydroxyl proton at the polymer end was used to estimate the Mn of polymer 3. Thermal Analyses. Thermal gravimetric analysis (TGA) and themomechanical analysis (TMA) of the lignin models were performed using a TG-DTA 2000s and TMA 4000s (MAC Science System 010, MAC Science, Yokohama, Japan). TGA was performed using 5 mg of polymer and heated from room temperature to 500 °C at a heating rate of 10 °C/min under a N2 stream of 200 mL/min. The thermal decomposition temperature (Td) was defined as the temperature at 5% weight loss.46 TMA of powdered specimens was also measured from room temperature up to 500 at 10 °C/min under a N2 stream of 100 mL/min and a 5 g load.31 The glass transition temperature (Tg) and thermal-flow starting temperature (Tf) were measured as the first and second inflection points, respectively. Differential scanning calorimetric analysis (DSC) was measured using ∼2 mg of sample in hermetic aluminum pans using a MDSC
molecular interactions via alkylation and the addition of small amounts of polyethylene oxide, respectively. Uraki et al.33 first reported on the fusibility of lignin using organosolv lignin produced from atmospheric acetic acid pulping. It was found that hardwood acetic acid lignin was a fusible material, while that produced from softwood was not19 but could be made fusible upon removal of the higher molecular mass, more condensed fraction. Similarly, Kadla and Kubo found that commercial Kraft hardwood lignin,17,42 as well as organosolv lignin (Alcell),20 could be thermally processed, while the softwood Kraft lignin could not. Again, the more condensed interunitary linkages, such as 5−5 and β-5 linkages, and more extensive and stronger intermolecular hydrogen bonding interactions in the softwood lignins being responsible for the decreased thermal mobility.43 To further examine the impact of lignin structure on thermal mobility we have synthesized polymeric lignin models consisting of only β-O-4 interunitary linkages.44 Referred to as an artificial β-O-4-type lignin, these lignin models are linear polymers with no branching, and as such, are expected to have fusibility. Using advanced spectroscopic and thermal analyses, we report on the relation between thermal mobility and lignin structure.
■
EXPERIMENTAL SECTION
Synthesis of Lignin Models and Preparation of Milled Wood Lignin (MWL). The C6−C2-type softwood of lignin model (polymer 1) was synthesized from acetovanillone through three steps as previously published (Figure 1).44 1H NMR (DMSO-d6): δ 1.28 (d, J = 6.4, CH3), 3.76 (s, 3H, OCH3), 3.90−3.98 (m, 2H, Cβ-H), 4.84 (m, 1H, Cα-H), 5.49 (d, 1H, J = 4.7, Cα-OH), 6.90 (s, 2H, C5−H, C6− H), 7.06 (s, 1H, C2−H); 13C NMR (DMSO-d6): δ 55.5 (OCH3), 70.7 (a), 74.1 (b), 110.7 (2), 113.1 (5), 118.3 (6), 135.2 (1), 147.1 (4), 148.5 (3). The C6−C3-type of lignin model (polymer 2) was also synthesized from acetovanillone through ethyl 3-(3′-methoxyphenyl)-3-oxopropanoate (polymer 3) by six steps, in accordance with our previous report (Figure 1).45 13C NMR (DMSO-d6) erythro: δ 55.5 (OCH3), 59.8 (c), 868
dx.doi.org/10.1021/bm201772v | Biomacromolecules 2012, 13, 867−872
Biomacromolecules
Article
(Q1000, TA Instruments, New Castle, DE). The Tg was measured as the midpoint temperature of the heat capacity transition over the temperature range of −50 to 300 °C of the reversing profile using a heating rate of 5 °C/min, amplitude of ±0.66 °C, a period of 50 s. FT-IR and NMR Measurements at Various Temperatures. FTIR analysis of the polymers as a function of temperature was performed using a custom manufactured temperature-controlled IR cell equipped with NaCl cell window plates (manufactured in the machine laboratory of the Research Faculty of Science, Hokkaido University). Lignin models were dissolved in a mixed solution of methanol and chloroform (3:7), dropped onto the NaCl plates, and the solvents removed by evaporation at 80 °C in vacuo. A thermocouple was inserted into the second NaCl plate, placed over the sample plate and set into the heating cylinder. The cell was then introduced into the FT-IR spectrometer (Biorad FTS-7) and the temperature was raised from 50 °C at a 5 °C intervals; 1024 scans were recorded at a resolution of 0.5 cm−1 at each temperature. The recorded spectra were analyzed using Peak Fit software (SPSS Inc., Chicago, IL) as reported previously.43 Deconvolution was performed using Gaussian peak shape and full width at half-maximum (fwhm in cm−1) were varied between 50 and 80% of the peak width to ensure good resolution of peaks without overfitting. All data was analyzed and compared based on peak areas of C−H stretching band. Selection of peaks and calculations of peak areas as a measure of spectral intensity were performed by maximum likelihood peak fitting, and all data were fit with r2 values of greater than 0.995. 1 H NMR spectra were recorded with a JEOL JNM EX-270 FTNMR spectrometer (270 Hz) on samples prepared in DMSO-d6. Spectra were collected from 30 to 150 °C at 5 °C intervals. 13C NMR spectrum for determination of erythro/threo ratio of polymer 2 was recorded on a Bruker AMX500 FT-NMR (125 MHz for carbon) at ambient temperature.
Table 1. Thermal Properties of Lignin Models and Todo-fir MWL
polymer 1 polymer 2 Todo-firMWL
Tda (°C)
residueb (%)
Tgc (°C)
Tgd (°C)
Tfd (°C)
302 263 260
31 40 60
93 111 143
106 140 (111) 154 (128)
153 n.d. (165) n.d. (184)
Td is the temperature at 5% weight loss. bResidual weight at 500 °C. Tg is estimated by a modulated DSC. dTg and Tf are estimated by TMA, and the temperatures of acetylated samples are shown in parentheses. a c
from polymer 2 in the structure of the alkyl side chain; polymer 2 contains a Cγ hydroxyl methyl group, whereas polymer 1 has no Cγ group. Therefore, the thermal instability of polymer 2 must be attributable to the decomposition of the γ-carbon, probably due to the elimination of the methylol group.47 The MWL indicated a similar Td to that of polymer 2. As the phenylpropane structure is a fundamental component of lignin, it appears that the Cγ-hydroxyl group may be one of the more thermally labile lignin moieties. Similarly, the Cγ-hydroxyl group had a large impact on the Tg. Despite the fact that polymer 2 was more thermally labile than polymer 1, it showed a higher Tg, as measured by both modulated DSC and TMA (Table 1); suggesting that the thermal mobility of polymer 2 was lower than polymer 1. MWL showed a substantially higher Tg than both of the lignin models. The decreased thermal mobility of MWL is likely a combination of the phenylpropane structure (Cγ-hydroxyl methyl group) and the more intricate interlinkages such as 5−5′ and β-5′ condensed structures. In all cases, the Tg values measured by DSC were lower than those from TMA, likely due to the differences in the way the Tg is determined. DSC is very sensitive to changes in heat capacity, while TMA detects the deformation of the specimen as it is heated through the glass transition and, thus, a temperature lag is observed. Thermal Fusibility of Polymeric Lignins. TMA is a widely used technique to detect viscoelastic and volumetric changes of polymers under shear force and load with elevating temperature. Because the thermal flow of polymers is not a clear phase transition, this phase change cannot be detected by DSC. In TMA, the Tg and the thermal-flow starting temperature (Tf) can be detected as a first and a second inflection point, respectively, in the thermal response profile. Typically, technical lignins exhibit only a single inflection point corresponding to the Tg.48 Figure 2A shows the TMA profiles for the two lignin models (polymers 1 and 2) along with that of MWL. As expected the MWL only showed a single inflection point, which could be assigned to the Tg. Interestingly, the two synthetic lignin models behaved quite different from each other. Polymer 1 showed two inflection points, the first corresponding to the Tg and the second to the Tf. However, the profile of polymer 2 exhibited only one inflection for the Tg. These results clearly suggest that the C6−C2-type lignin polymer was fusible, whereas the C6−C3-type polymer was not. The thermal transitions for both lignin polymers were also confirmed by visible inspection, as shown in Supporting Information, Figure S1. Since the thermal infusibility of polymer 2 might be attributable to the Cγ-hydroxyl group and its propensity to form intermolecular interactions,43 TMA analysis was also performed on polymer 3, as well as acetylated polymer 2 and
■
RESULTS AND DISCUSSION Fundamental Properties of Lignin Models. The number average molecular mass, (Mn), and polydispersity index (shown in parentheses) of polymers 1, 2, and 3 were estimated by the HP-SEC to be 3100 (2.3), 5700 (1.8), and 2900 g/mol (4.5), respectively. The corresponding value for Todo-fir MWL was 2000 g/mol (2.7). As outlined in Figure 1, polymer 3 is an intermediate in the synthesis of polymer 2. The fact that it showed a smaller Mn and larger polydispersity index implies that the reduction and subsequent purification step for the synthesis of polymer 2 from polymer 3 removes the lower molecular mass fraction. Similar differences in the Mn of polymers 2 and 3 were obtained from end-group determination using 1H NMR. Again the Mn of polymer 2 was double that of polymer 3 at 4500 and 2150 g/mol, respectively, however, both lower than that found using HP-SEC. In the analysis of lignins by SEC, the use of polystyrene standards is often criticized based on the fact that lignins are not linear, but rather complex network polymers. However, the lignin models polymers 2 and 3 are linear and may be better represented using linear polystyrene standards. The observed discrepancy between the two techniques is likely a combination of the fact that the lignin polymers are extensively oxidized and will thus exhibit quite different hydrodynamic radii, and that SEC is relatively insensitive to low-molecular-weight contaminants unlike NMR end-group analysis, both potentially leading to the higher Mn values. Thermal Decomposition and Glass Transition of Lignin Model Compounds. Table 1 shows the decomposition (Td) and glass transition (Tg) temperatures for polymers 1 and 2 as well as the MWL preparation. Polymer 2 is more thermally labile than polymer 1, having the lower decomposition temperature. Structurally, polymer 1 differs 869
dx.doi.org/10.1021/bm201772v | Biomacromolecules 2012, 13, 867−872
Biomacromolecules
Article
Figure 2. TMA profiles of artificial lignin polymers, Todo-fir MWL, and their derivatives: (A) red line, polymer 1; blue line, polymer 2; green line, Todo-fir MWL; (B) red line, acetylated polymer 2; blue line, polymer 3; green line, acetylated MWL.
Figure 3. 1H NMR spectra of polymer 1 (A) and polymer 2 (B) in DMSO-d6 as a function of temperature from 30−150 °C.
they are less involved in intra- and interlignin interactions.49,50 These findings are consistent with our previous work wherein we observed that the Cγ-hydroxyl groups in lignin and lignin model compounds participate in stronger, and in a wider variety of intra- and intermolecular hydrogen bonds than Cαhydroxyl groups.43 To more quantitatively understand the specific intra- and intermolecular hydrogen bonding interactions in polymers 1 and 2, FT-IR spectra were collected on thin films deposited on NaCl windows as a function of temperature (Figure 4). Figure 4A shows the deconvoluted hydroxyl stretching region of the two artificial lignin polymers at 50 °C. The number of bands and the band centers do not vary significantly between the
MWL, as shown in Figure 2B. Both acetylated polymers 2 and 3 showed two inflection points, suggesting thermal fusibility. Similarly, the acetylated MWL also indicated thermal fusibility but over a narrow temperature range. These results suggest that the hydroxyl group at the Cγ-position strongly influences the thermal mobility of lignin, particularly the ability of flow under thermal treatment. Relationship between Thermal Behavior of Hydroxyl Groups and Thermal Mobility of Polymeric Lignins. The thermal infusibility of lignins is presumably caused by strong intermolecular interactions and thermal decomposition of the polymer upon heating. Thermal analysis showed that the Td of polymer 2 (263 °C) was lower than of polymer 1 (302 °C), while its Tg was slightly higher, 111/140 °C versus 93/106 °C, as measured by DSC/TMA, respectively. The Tf of polymer 1 (153 °C) was substantially lower than its Td (302 °C) but significantly higher than its Tg. As the Td of polymer 2 was much higher than the Tf of the acetylated analogue, as well as those observed for all of the lignin samples investigated in this study, it is proposed that the infusibility of polymer 2 is caused by strong intermolecular interactions involving the Cγ-hydroxyl groups. On the basis of this hypothesis, we examined the thermal behavior of the hydroxyl groups in polymers 1 and 2 using 1H NMR and FT-IR. 1 H NMR analysis on the effect of increasing temperature on the chemical shift of the Cα-hydroxyl protons in both polymer 1 and polymer 2, as well as the Cγ-hydroxyl protons in polymer 2 showed a linear shift to higher magnetic field with concomitant line broadening (Figure 3). The amount of change in the chemical shift was slightly greater for polymer 1 than either hydroxyl proton in polymer 2. These results seem to imply that the hydroxyl group in polymer 1 interacts more favorably with the solvent than those in polymer 2, suggesting
Figure 4. Deconvolution spectra (A) and temperature-dependence spectra (B) of FT-IR for polymers 1 (a) and 2 (b) in the hydroxyl band. 870
dx.doi.org/10.1021/bm201772v | Biomacromolecules 2012, 13, 867−872
Biomacromolecules
Article
hydroxyl group. Similarly, a MWL sample was also found to exhibit thermal-flow behavior through acetylation. These results suggest that modification of intermolecular interactions, specifically hydrogen bonding interactions involving the Cγhydroxyl group, could facilitate thermal fusibility and the development of thermoprocessable lignins.
polymers, although a change in the relative intensity of several bands is evident. In polymer 2 the band at ∼3273 cm−1 is more than double the intensity of that in polymer 1, whereas the band at 3486 cm−1 is substantially lower in intensity. Similarly, a new band at a much lower wavenumber, ∼3180 cm−1, is also observed only in polymer 2, indicating the existence of stronger intermolecular hydrogen bonds.43 Increasing temperature decreased the absorption intensity and shifted the maximum of the hydroxyl stretching band envelope to higher wavenumber for both polymers (Figure 4B). The higher wavenumber stretching bands (∼3486 and ∼3491 cm−1 for polymers 1 and 2, respectively) are likely associated with intramolecular hydrogen bonds43 and dominate the hydroxyl stretching region. In polymer 1 the decrease in intensity was accompanied by a linear shift to higher wavenumber with increasing temperature (Figure 5A).
■
ASSOCIATED CONTENT
S Supporting Information *
Photoimages of several lignin models and Todo-fir MWL at various temperatures, FT-IR deconvolution data, SEC chromatograph (polymer 2) and MDSC profiles for polymers 1 and 2, as well as Todo-fir MWL. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Present Address #
Ibiden Co. Ltd., 2-1, Kanda-cho, Ogaki City, Gifu 503-8604, Japan. Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Sederoff, R. R.; Chang, H.-m., Lignin Biosynthesis. In Wood Structure and Composition; Lewin, M., Goldstein, I. S., Eds.; M. Dekker: New York, 1991. (2) Ralph, J.; Lundquist, K.; Brunow, G.; Lu, F.; Kim, H.; Schatz, P. F.; Marita, J. M.; Hatfield, R. D.; Ralph, S. A.; Christensen, J. H.; Boerjan, W. Phytochem. Rev. 2004, 3 (1−2), 29−60. (3) Capanema, E. A.; Balakshin, M. Y.; Kadla, J. F. J. Agric. Food Chem. 2004, 52 (7), 1850−1860. (4) Fujimoto, A.; Matsumoto, Y.; Chang, H. M.; Meshitsuka, G. J. Wood Sci. 2005, 51 (1), 89−91. (5) Bjorkman, A. Nature 1954, 174 (4440), 1057−1058. (6) Kirkman, A. G.; Gratzl, J. S.; Edwards, L. L. Tappi J. 1986, 69 (5), 110−114. (7) Glasser, W. G., Sarkanen, S., Eds. Lignin: Properties and Materials; American Chemical Society: Washington, DC, 1989. (8) Huttermann, A.; Mai, C.; Kharazipour, A. Appl. Microbiol. Biotechnol. 2001, 55 (4), 387−394. (9) Kubo, S.; Gilbert Richard, D.; Kadla John, F. Lignin-based polymer blends and biocomposite materials. In Natural Fibers, Biopolymers and Biocomposites; Mohanty, A. K., Misra, M., Drzal, L. T., Eds.; Taylor & Francis: New York, 2005; pp 671−697. (10) Lora, J. H.; Wu, C. F.; Pye, E. K.; Balatinecz, J. J. ACS Symp. Ser. 1989, 397, 312−323. (11) Sena-Martins, G.; Almeida-Vara, E.; Duarte, J. C. Ind. Crop Prod. 2008, 27 (2), 189−195. (12) Rana, D.; Matsuura, T. Chem. Rev. 2010, 110, 2448−2471. (13) Kleinert, M.; Barth, T. Energy Fuels 2008, 22 (2), 1371−1379. (14) Gandini, A. Macromolecules 2008, 41 (24), 9491−9504. (15) Ghosh, I.; Jain, R. K.; Glasser, W. G. Lignin: Hist., Biol., Mater. Perspect. 2000, 742, 331−350. (16) Inone-Kauffmann, E. R. Int. Sugar J. 2009, 111 (1321), 10−11. (17) Kadla, J. F.; Kubo, S. Macromolecules 2003, 36 (20), 7803−7811. (18) Kadla, J. F.; Kubo, S. Composites, Part A 2004, 35 (3), 395−400. (19) Kubo, S.; Ishikawa, N.; Uraki, Y.; Sano, Y. Mokuzai Gakkaishi 1997, 43 (8), 655−662. (20) Kubo, S.; Kadla, J. F. Macromolecules 2004, 37 (18), 6904−6911. (21) Li, Y.; Sarkanen, S. Macromolecules 2002, 35 (26), 9707−9715. (22) Luo, X. G.; Lin, X. Y.; Liu, F. Mater. Sci. Forum 2009, 620−622. (23) Mansour, O. Y. Polym.-Plast. Technol. 1992, 31 (9−10), 747− 758.
Figure 5. Effect of temperature on the wavenumber for the νOH band of polymers 1 (A) and 2 (B).
However, in polymer 2 the shift in the maximum of the ∼3491 cm−1 hydroxyl stretching band appeared to follow two regression lines; one from 50 to 120 °C and the other from 120 to 150 °C (Figure 5B). The slope of the initial regression line in the lower temperature region was similar to that of polymer 1, whereas that of the second high temperature region was much sharper. Similar temperature behavior was observed for all of the deconvoluted bands in both polymer 1 and polymer 2 (Supporting Information, Figure S3). It has been reported that the Cα-hydroxyl groups in lignins primarily participate in intramolecular hydrogen bonding.51,52 Accordingly, the initial shift of the hydroxyl stretching band in polymer 2, which is comparable to that observed in polymer 1 is attributed to the cleavage of intramolecular hydrogen bonds. By contrast, at higher temperature, the dramatic change in the slope of the regression line for polymer 2 likely corresponds to the cleavage of intermolecular hydrogen bonds involving the Cγ-hydroxyl group. These bonds are stronger and bring about a larger wavenumber shift than the intramolecular hydrogen bonds.
■
CONCLUSION The role of lignin structure on the thermal mobility and fusibility of lignins was investigated using well-defined linear lignin model polymers comprised exclusively of β-O-4 (8-O-4′) interunitary linkages. A C6−C2-type polymeric lignin model (polymer 1) was found to be fusible, while the corresponding C6−C3 analogue (polymer 2) was not. NMR and FT-IR analysis of the two lignin models revealed that the thermal infusibility was due to intermolecular hydrogen bonding involving the Cγ-hydroxyl group. However, the C6−C3-type polymeric lignin model could be made fusible by disrupting the intermolecular hydrogen bonding by acetylating the Cγ871
dx.doi.org/10.1021/bm201772v | Biomacromolecules 2012, 13, 867−872
Biomacromolecules
Article
(24) Sakata, I.; Senju, R. J. Appl. Polym. Sci. 1975, 19 (10), 2799− 2810. (25) Uraki, Y.; Taniwatashi, R.; Kubo, S.; Sano, Y. J. Wood Sci. 2000, 46 (1), 52−58. (26) Lysenko, A. A.; Lysenko, V. A.; Astashkina, O. V.; Gladunova, O. I. Fibre Chem. 2011, 42 (5), 278−286. (27) Ma, X. J.; Zhao, G. J. J. Appl. Polym. Sci. 2011, 121 (6), 3525− 3530. (28) Shen, Q.; Zhang, T.; Zhang, W. X.; Chen, S. A.; Mezgebe, M. J. Appl. Polym. Sci. 2011, 121 (2), 989−994. (29) Ruiz-Rosas, R.; Bedia, J.; Lallave, M.; Loscertales, I. G.; Barrero, A.; Rodriguez-Mirasol, J.; Cordero, T. Carbon 2010, 48 (3), 696−705. (30) Dallmeyer, I.; Ko, F.; Kadla, J. F. J. Wood Chem. Technol. 2010, 30 (4), 315−329. (31) Kadla, J. F.; Kubo, S.; Venditti, R. A.; Gilbert, R. D. J. Appl. Polym. Sci. 2002, 85 (6), 1353−1355. (32) Kadla, J. F.; Kubo, S.; Venditti, R. A.; Gilbert, R. D.; Compere, A. L.; Griffith, W. Carbon 2002, 40 (15), 2913−2920. (33) Uraki, Y.; Kubo, S.; Nigo, N.; Sano, Y.; Sasaya, T. Holzforschung 1995, 49 (4), 343−350. (34) Kleinert, M.; Barth, T. Chem. Eng. Technol. 2008, 31 (5), 736− 745. (35) Uraki, Y.; Nemoto, J.; Yanaga, K.; Koizumi, A.; Hirai, T. J. Wood Sci. 2005, 51 (6), 589−594. (36) Chen, P.; Zhang, L. N.; Peng, S. P.; Liao, B. J. Appl. Polym. Sci. 2006, 101 (1), 334−341. (37) El-Wakil, N. A. J. Appl. Polym. Sci. 2009, 113 (2), 793−801. (38) Lin, N.; Fan, D. K.; Chang, P. R.; Yu, J. H.; Cheng, X. C.; Huang, J. J. Appl. Polym. Sci. 2011, 121 (3), 1717−1724. (39) Rouilly, A.; Rigal, L. J. Macromol. Sci., Part C 2002, 42 (4), 441− 479. (40) Wei, M.; Fan, L. H.; Huang, J.; Chen, Y. Macromol. Mater. Eng. 2006, 291 (5), 524−530. (41) Glasser, W. G.; Jain, R. K. Holzforschung 1993, 47 (3), 225−233. (42) Kadla, J. F.; Kubo, S.; Gilbert, R. D.; Venditti, R. A. Chem. Mod., Prop., Usage Lignin 2002, 121−137. (43) Kubo, S.; Kadla, J. F. Biomacromolecules 2005, 6 (5), 2815− 2821. (44) Kishimoto, T.; Uraki, Y.; Ubukata, M. Org. Biomol. Chem. 2005, 3 (6), 1067−1073. (45) Kishimoto, T.; Uraki, Y.; Ubukata, M. Org. Biomol. Chem. 2006, 4 (7), 1343−1347. (46) Kubo, S.; Kadla, J. F. J. Wood Chem. Technol. 2008, 28 (2), 106− 121. (47) Kawamoto, H.; Horigoshi, S.; Saka, S. J. Wood Sci. 2007, 53 (2), 168−174. (48) Kubo, S.; Uraki, Y.; Sano, Y. Holzforschung 1996, 50 (2), 144− 150. (49) Baxter, N. J.; Williamson, M. P. J. Biomol. NMR 1997, 9 (4), 359−369. (50) Stjacques, M.; Sundararajan, P. R.; Taylor, K. J.; Marchessault, R. H. J. Am. Chem. Soc. 1976, 98 (15), 4386−4391. (51) Langer, V.; Li, S. M.; Lundquist, K. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 2008, 64, O558−O560. (52) Ludwig, C. H.; Mccarthy, J. L.; Nist, B. J. J. Am. Chem. Soc. 1964, 86 (6), 1186−1191.
872
dx.doi.org/10.1021/bm201772v | Biomacromolecules 2012, 13, 867−872
