Article pubs.acs.org/Biomac
Reliable dn/dc Values of Cellulose, Chitin, and Cellulose Triacetate Dissolved in LiCl/N,N-Dimethylacetamide for Molecular Mass Analysis Yuko Ono, Takashi Ishida, Hiroto Soeta, Tsuguyuki Saito, and Akira Isogai* Department of Biomaterials Science, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo 113-8657, Japan S Supporting Information *
ABSTRACT: Freeze-dried microfibrillated cellulose (MFC) was directly dissolved in 8.0% w/w lithium chloride/N,Ndimethylacetamide (LiCl/DMAc), and MFC/LiCl/DMAc solutions with accurate MFC concentrations were prepared. The different MFC solutions were diluted to 1.0% and 0.5% w/v LiCl/DMAc, and subjected to size-exclusion chromatography with multiangle laser-light scattering and refractive index analyses (SEC/MALLS/RI), and off-line RI analysis to determine their refractive index increments (dn/dc). Chitin, cellulose triacetate, a poly(styrene) standard, and cellobiose were used for comparison. Each of the two determination methods gave different dn/dc values for MFC and chitin but similar dn/dc values for cellulose triacetate and poly(styrene). The anomalously small dn/dc values of MFC and chitin were explainable in terms of stable cellulose-LiCl and chitin-LiCl structures (i.e., formation of apparent covalent bonds between hydroxyl groups and LiCl) in the solutions. Thus, the SEC/MALLS/RI method provides reliable molecular mass parameters for cellulose and chitin.
■
INTRODUCTION Native plant celluloses are highly abundant and annually accumulating on earth, and cellulose-related research and development has been carried out intensively and extensively toward the creation of a sustainable society. As is the case for other natural and petroleum-based polymers, the molecular masses and molecular mass distributions of celluloses are significant measures important to their application. The lithium chloride/N,N-dimethylacetamide (LiCl/DMAc) solvent system has been thought to be the most suitable for evaluating the molecular mass parameters of celluloses by size-exclusion chromatography combined with multiangle laser-light scattering and refractive index analyses (SEC/MALLS/RI). The detailed theory and related equations used to measure and calculate the absolute molecular mass and root-mean-square (rms) radius of a polymer in each slice of the SEC elution patterns, and the number- and weight-average molecular masses (Mn and Mw, respectively) of the whole polymer molecules from SEC/MALLS/RI data have been described elsewhere.1 In SEC/MALLS/RI analysis, accurate refractive index increments (dn/dc) must be used when calculating the molecular mass parameters. Polymer dn/dc values are generally determined using polymer solutions in the same solvent as that used in the SEC/MALLS analysis, and using an off-line RI analyzer with the same laser-light wavelength as that of the MALLS analyzer. The dn/dc values of some polymers dissolved in common solvents have been already reported in the © XXXX American Chemical Society
literature and in data books, and can be used in SEC/ MALLS analysis. Most cellulose samples are soluble in the colorless 8% w/w LiCl/DMAc or 8% w/w LiCl/N,N-dimethyl-2-imidazoridinone (DMI) solvent systems without suffering from significant depolymerization during the dissolution process as well as during storage at room temperature for long periods. Thus, after dilution to 0.5−1.0% w/v LiCl concentration, cellulose solutions in LiCl/DMAc and LiCl/DMI have been used in SEC/MALLS analyses to determine the molecular mass parameters of various cellulose samples. However, it is well-known that the dn/dc values of cellulose dissolved in LiCl/DMAc and LiCl/DMI, determined using the off-line RI method,2 are remarkably different from those determined by the SEC/MALLS/RI method, assuming 100% mass recovery.3,4 Especially, the masses of cellulose injected during SEC analysis, calculated using the dn/dc values determined by the off-line RI method, were clearly higher than those of the actually injected cellulose. Berggren et al. hypothesized that the large dn/dc value of LiCl in DMAc (0.324 mL/g) and higher concentrations of LiCl than cellulose in LiCl/DMAc cause errors when measuring the dn/dc values of cellulose using the off-line RI method.5 Potthast et al. proposed that a solvent gradient between cellulose molecules Received: September 26, 2015 Revised: November 25, 2015
A
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
Article
Biomacromolecules
temperature for 1 h and then at 30 °C for overnight to prepare a clear and colorless 8.0% w/w (or 8.2% w/v) LiCl/DMAc solution. Preparation of Cellulose, Chitin, Cellulose Triacetate, and Cellobiose Solutions in LiCl/DMAc. The F-MFC (160 mg), Achitin (170 mg), cellulose triacetate (400 mg), poly(styrene) (200 mg), and cellobiose (400 mg) samples were further vacuum-dried for 1 day at room temperature, and were then dissolved in 8.0% w/w LiCl/ DMAc (5 mL) by stirring at room temperature for 1 week.6 The conventional solvent-exchange procedure from a MCC/water slurry to a MCC/DMAc slurry through a MCC/acetone slurry (1 day-soaking for each step) was applied to prepare a 4 × 10−3 g/mL MCC solution in 8.0% w/w LiCl/DMAc.15−18 The obtained polymer or cellobiose solution (5 mL) in 8.0% w/w LiCl/DMAc was diluted with fresh DMAc (35 mL) to 1.0% w/v LiCl/DMAc,2,4−11 and then further diluted with various amounts of 1.0% w/v LiCl/DMAc to prepare polymer or cellobiose solutions of various concentrations for the following SEC/MALLS/RI and off-line RI analyses. The dried cellulose triacetate (430 mg) and LiCl (600 mg) were dissolved in DMAc (40 mL), while the cellobiose (400 mg) was dissolved in water (40 mL) by stirring the mixture at room temperature for 1 day. Preparation of F-MFC Solution in LiCl/DMI. The freeze-dried FMFC (1.6 mg) prepared in the previous section was directly dissolved in 8.0% w/w LiCl/DMI (0.5 mL) by stirring the mixture at room temperature for 1.5 months. The F-MFC solution was diluted with fresh DMI to prepare a F-MFC solution in 1.0% w/v LiCl/DMI for SEC/MALLS/RI analysis.3,15,17,19 SEC/MALLS/RI Analysis. The SEC/MALLS/RI analysis system comprising a SEC column (KD-806M, Shodex, Japan), a MALLS detector (DAWN HELEOS-II, λ = 658 nm, Wyatt Technologies, USA), and an RI detector (RID-10A, Shimadzu, Japan), and 1.0% w/v LiCl/DMAc, 0.5% w/v LiCl/DMAc, or 1.0% w/v LiCl/DMI was used as eluents. Other SEC attachments and the operation conditions are described elsewhere.19,20 The dn/dc value of each polymer or cellobiose was obtained from the SEC/MALLS/RI data using the attached software according to eq 1, assuming 100% mass recovery. This method requires accurate injection volumes and sample concentrations, and also needs all sample molecules to be eluted off the column.
and the surrounding solution is generated when LiCl/DMAc is used as the solvent, which results in inaccurate dn/dc values measured by the off-line RI method.4 Therefore, in most cases, the dn/dc values reported for cellulose and used to calculate the molecular mass parameters of various celluloses have been obtained by the SEC/MALLS/ RI method instead of the off-line RI method,4−11 and LiCl/ DMAc or LiCl/DMI has been used as the cellulose solvent and eluent in SEC/MALLS analysis. In this case, the precise concentration of cellulose in such LiCl/DMAc and LiCl/DMI solutions should be known for accurate calculation of the dn/dc values. However, the sequential solvent exchange from cellulose/water slurry to cellulose/DMAc or cellulose/DMI slurry through cellulose/acetone slurry to activate the cellulose often causes a small amount of weight loss during handling. Additionally, complete removal of the solvent from these solvent-exchanged cellulose samples is difficult to achieve, even after vacuum drying of the DMAc- or DMI-treated cellulose at 60 °C for a few days.12 In this paper, we used a freeze-dried microfibrillated cellulose (MFC), which is directly soluble in 8% w/w LiCl/DMAc and 8% w/w LiCl/DMI, to determine the dn/dc values of cellulose by the SEC/MALLS/RI and off-line RI methods. The reason for the differences between the dn/dc values obtained by the two methods are studied using cellulose triacetate, chitin, a poly(styrene) standard sample, and cellobiose for comparison. The validity of the obtained dn/dc value for cellulose was then examined for a microcrystalline cellulose/LiCl/DMAc solution, and MFC/LiCl/DMAc and MFC/LiCl/DMI solutions with different MFC and LiCl concentrations.
■
MATERIALS AND METHODS
Materials. A commercial MFC produced from softwood dissolving pulp by repeated microfluidization treatment in water (10% solid content, Celish KY100G, Daicel FineChem, Ltd., Japan) was used as the starting cellulose sample. The MFC consisted of 96.2% glucose, 1.2% mannose, 2.2% xylose, and 0.3% arabinose, as determined by neutral sugar analysis using high-pressure liquid chromatography.13,14 The MFC was suspended in water at a solid content of 0.5%, and the suspension was passed 10 times through a high-pressure homogenizer (Star Burst Mini, HJP-25001S, Sugino Machine, Ltd., Japan). The additionally fibrillated MFC slurry was placed in a plastic centrifugation tube with a screw cap, and the excess water was removed by centrifugation at 12 000g for 10 min and successive decantation. The MFC precipitate was washed with ethanol three times (30 mL each) and then with tert-butyl alcohol three times (30 mL each) by centrifugation, followed by freeze-drying. The obtained fibrillated and freeze-dried MFC is hereafter abbreviated as F-MFC. The chitin sample was a commercial product (Chitin P, Dainichiseika Color & Chemicals Mfg. Co., Ltd., Japan.). The as-received chitin (100 mg) was treated in 1 M HCl (100 mL) at 85 °C for 2 h, after which the mixture was centrifuged to remove the supernatant. The solid chitin was first washed with 0.05 M NaOH and then repeatedly with water by centrifugation at 12 000g for 5 min each. The acid-treated chitin (A-Chitin) was obtained as a freeze-dried sample using the same procedure as that for the F-MFC. Cellulose triacetate (LT-35, Daicel FineChem, Ltd., Japan), a polystyrene standard (Shodex Standard P82 No. P30, Shoko Co., Ltd., Japan), and microcrystalline cellulose (MCC) (Cellulose powder C, Advantec Toyo, Co., Ltd., Japan) were commercial products and used as received. Other chemicals and solvents were of laboratory grade, purchased from Wako Pure Chemical Ind., Ltd., Japan, and were used as received. Preparation of 8.0% w/w LiCl/DMAc. Commercially available LiCl was vacuum-dried at 105 °C overnight, and the dried LiCl (16.4 g) was added to DMAc (200 mL). The LiCl/DMAc was cooled at 4 °C in a refrigerator for 1 h, and the mixture was stirred at room
dn α = dc Winjected
∑ Δvi(Vi − Vbaseline) peak
(1)
where α is a RI detector instrument constant, Winjected is the injected mass, ∑Δvi is the volume of the slice, and Vi and Vbaseline are the RI signal and baseline voltages, respectively. Off-Line RI Analysis. The off-line RI analysis of the polymer, cellobiose, and LiCl solutions of various concentrations was carried out using an interferometric refractometer (Optilab, λ = 658 nm, Wyatt Technologies, USA) at 25 °C. The flow rate was adjusted to 0.5 mL/ min using a syringe pump (Econoflo, Harvard Apparatus, Physio-Tech Co., Ltd., Tokyo, Japan) and a 0.45 μm hydrophilic PTFE disposable membrane filter (Millex-LG, Merck Millipore Co., Germany) directly connected to the RI analyzer. The dn/dc value was determined from the slope of a plot of differential refractive index (αV−αVbaseline) against concentration, where α is a constant specific to the RI analyzer used, V is the polymer solution voltage, and Vbaseline is the basis solution voltage.
■
RESULTS AND DISCUSSION Solubility of Polymers and Cellobiose in Various Solvents and Their dn/dc Values. Native cellulose must be pretreated in DMAc or DMI, through either sequential solvent exchange from water to the solvent or heating in the solvent at its boiling point, to activate and completely dissolve it in LiCl/ DMAc or LiCl/DMI. However, in the case of heating pretreatment, clear depolymerization of cellulose is unavoidable,21,22 which is not suitable to determine cellulose molecular mass parameters. We found that the MFC, produced from softwood dissolving pulp with a high α-cellulose content, was B
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
Article
Biomacromolecules
determined by the SEC/MALLS/RI and off-line RI methods are summarized in Table 2. The different solute concentration ranges were used to obtain reliable dn/dc values. A clear discrepancy in the dn/dc value determined by the two different methods was observed for F-MFC and A-chitin,4 whereas quite similar values were obtained by the two methods for cellulose triacetate and poly(styrene). The dn/dc value of cellobiose determined for its 1.0% w/v LiCl/DMAc solution was similar to that for F-MFC. These results indicate that F-MFC, A-chitin, and cellobiose, which have some interactions between their hydroxyl groups and LiCl in DMAc, were the causes of the large differences between the dn/dc values obtained using the two methods. The SEC elution patterns of F-MFC solutions with various F-MFC concentrations in 1.0% w/v LiCl/DMAc and their molecular mass plots obtained using the dn/dc value of 0.131 mL/g are shown in Figure 1a. The relationship between the FMFC concentration and the differential refractive index of the solution determined by the off-line RI method is shown in Figure 1b. The SEC elution patterns were similar, and the SEC elution intensity detected by RI decreased with F-MFC concentration. The molecular mass plots were linear and similar to those of the elution volume for all F-MFC solutions, indicating that the dn/dc value of 0.131 mL/g in Table 2 obtained by the SEC/MALLS/RI method was reasonable. A linear relationship with a very high correlation coefficient was also obtained from Figure 1b. Based on these results, the slope of 0.065 mL/g for the dn/dc value of F-MFC obtained by the off-line RI method seems to be a reliable value for F-MFC concentration range studied. However, the two dn/dc values, i.e., 0.131 and 0.065 mL/g, are clearly quite different. Similar results were obtained for F-MFC solutions in 0.5% w/v LiCl/ DMAc (Figure S2 in the Supporting Information). When A-chitin was dissolved in 1.0% w/v LiCl/DMAc, the solutions with A-chitin concentrations of 0.7 × 10−3 and 1.1 × 10−3 g/mL exhibited anomalous SEC elution patterns with nonlinear molecular mass versus elution volume plots (Figure S3 in the Supporting Information). These results indicate that A-chitin concentrations 0.5 × 10−3 g/mL or lower in 1.0% w/v LiCl/DMAc are favorable for SEC/MALLS analysis to obtain reliable data. Gel-like precipitates were formed when the 1.0% w/v LiCl/DMAc A-chitin solution was diluted to 0.5% w/v LiCl/DMAc (Figure S1 in the Supporting Information). Thus, 1.0% w/v LiCl/DMAc is required as the eluent for A-chitin. The SEC elution patterns of the 0.5−2 × 10−3 g/mL cellulose triacetate solutions in 1.0% w/v LiCl/DMAc and their
directly soluble in both 8.0% w/w LiCl/DMAc without any activation treatment, and gave highly viscous solutions. In this study, the MFC was repeatedly homogenized in water in our laboratory, solvent-exchanged from water to TBA, and then freeze-dried to prepare F-MFC. The repeated homogenization treatment was applied to reduce the viscosities of the MFC/ LiCl/DMAc solutions to make solutions with various MFC concentrations easier to prepare. Thus, the precise weights of dried F-MFC could be measured, and the concentration of FMFC dissolved in 8.0% w/w LiCl/DMAc were accurately controllable using the freeze-dried F-MFC, which is required for determination of dn/dc values. The conventional solventexchange process from aqueous cellulose slurry to cellulose/ DMAc or cellulose/DMI slurry through acetone for activation and complete dissolution in 8.0% w/w LiCl/DMAc or 8.0% w/ w LiCl/DMI causes partial weight loss of the cellulose samples during the multiple solvent-exchange process. Although F-MFC was completely dissolved in 8.0% w/w LiCl/DMAc after stirring at room temperature for 1 week, it took approximately 1 month to completely dissolve F-MFC in 8.0% w/w LiCl/DMI. In this study, acid-hydrolyzed chitin (Achitin) was used to determine the dn/dc values, because the original chitin solutions in 8.0% w/v LiCl/DMAc had very high viscosities and were difficult to smoothly stir, especially for the solutions with high chitin concentrations. The A-chitin was also directly dissolved in 8.0% w/w LiCl/DMAc without solventexchange. Commercially available and dried cellulose triacetate, poly(styrene), and cellobiose were directly soluble in 8.0% w/w LiCl/DMAc (Table 1). Table 1. Solubility of Various Polymers and Cellobiose in Water, DMAc, and 8.0% w/w LiCl/DMAc solvent solute
water
DMAc
8.0% w/w LiCl/DMAc
F-MFC A-chitin cellulose triacetate poly(styrene) cellobiose
insoluble insoluble insoluble insoluble soluble
insoluble insoluble soluble soluble insoluble
soluble soluble soluble soluble soluble
The dn/dc values of F-MFC, A-chitin, cellulose triacetate, poly(styrene), and cellobiose in the solutions of different concentrations were determined by the SEC/MALLS/RI and off-line RI methods using the same laser light wavelength of 658 nm. The obtained dn/dc values and that of LiCl in DMAc
Table 2. dn/dc Values of Polymers, Cellobiose, and LiCl in Various Solvents dn/dc, mL/g (concentration range, g/mL)
a
solute
solvent
SEC/MALLS/RI
off-line RI
F-MFC F-MFC F-MFC A-chitin cellulose triacetate cellulose triacetate poly(styrene) cellobiose cellobiose LiCl
1.0% w/v LiCl/DMAc 0.5% w/v LiCl/DMAc 1.0% w/v LiCl/DMI 1.0% w/v LiCl/DMAc 1.0% w/v LiCl/DMAc DMAc 1.0% w/v LiCl/DMAc 1.0% w/v LiCl/DMAc water DMAc
0.131 ± 0.005a (0.125−1 × 10−3) 0.131 (0.25−5 × 10−4) 0.104 (0.125−1 × 10−3) 0.138 ± 0.004a (0.13−1.1 × 10−3) 0.040 ± 0.001a (0.5−2 × 10−3) − 0.150b − − −
0.065 (0.5−4 × 10−3) 0.084 (0.125−2 × 10−3) − 0.091 (0.13−4 × 10−3) 0.042 (0.4−1 × 10−3) 0.042 (0.01−1.1 × 10−2) 0.148 (0.63−5 × 10−3) 0.067 (0.125−1 × 10−2) 0.146 (1.25−1 × 10−2) 0.390 (0.025−1.5 × 10−2)
Standard deviation. bRefs 17 and 23. C
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
Article
Biomacromolecules
Figure 2. SEC elution patterns and corresponding molecular mass plots of cellulose triacetate dissolved in 1.0% w/v LiCl/DMAc (a), and relationship between cellulose triacetate concentration and differential refractive index of the cellulose triacetate solutions in 1.0% w/v LiCl/ DMAc and DMAc, determined by the off-line RI method (b).
Figure 1. SEC elution patterns and corresponding molecular mass plots of F-MFC in 1.0% w/v LiCl/DMAc (a), and relationship between F-MFC concentration and differential refractive index of the F-MFC solutions in 1.0% w/v LiCl/DMAc, determined by the off-line RI method (b).
in the solvent.4 This LiCl concentration gradient may result in an inaccurate dn/dc value, when measured using the off-line RI method. As shown in Figures 1a, S2a, and S3a (in the Supporting Information), the maximum concentration of FMFC or A-chitin at each elution volume detected by RI in the SEC/MALLS/RI analysis was lower than 6 × 10−5 g/mL, whereas the total concentration of all molecules in the off-line RI analysis was more than 10−4 g/mL to better observe the linear relationship between the F-MFC or A-chitin concentration and the differential refractive index of the solutions (Figures 1b, S2b, and S3b). The dn/dc measurements obtained by the off-line RI method were applied to the F-MFC solutions in 1.0% w/v LiCl/DMAc, cellulose triacetate solutions in 1.0% w/v LiCl/DMAc, and cellulose triacetate solutions in DMAc with polymer concentrations of 10 −4 g/mL, some consumption of the LiCl to form covalent-like bonds between OH groups and LiCl was expected to decrease in LiCl concentration in the surrounding DMAc solution (at positions far from each cellulose or chitin molecule). This gradient of LiCl concentration in the solutions may not be ignored, and probably resulted in the decreased dn/dc value obtained by the off-line RI method. When cellulose triacetate or poly(styrene) with no hydroxyl groups are used, no such consumption of LiCl forming the covalent-like bonds occurs, thus resulting in almost the same dn/dc values determined between the SEC/MALLS/ RI and off-line RI methods (Table 2). The amount of LiCl bound to F-MFC molecules in 1.0% w/v LiCl/DMAc and 0.5% w/v LiCl/DMAc and A-chitin molecules in 1.0% w/v LiCl/DMAc can be calculated from the dn/dc values obtained by the SEC/MALLS/RI and off-line RI methods and the dn/dc value (0.390 mL/g) of LiCl in DMAc, according to the following equations: doff − lineRI d + dLiCl × a = MALLS 1+a 1+a a × 162 (for cellulose); mLiCl = 42.4 α mLiCl = × 203 (for chitin) 42.4
polysaccharide
solvent
F-MFC
1.0% w/v LiCl/ DMAc 0.5% w/v LiCl/ DMAc 1.0% w/v LiCl/ DMAc
F-MFC A-chitin
LiCl (mol) bound to one repeating unit 0.55 0.41 0.51
imately one mole of LiCl salt was bound to two glucosyl units of F-MFC in 1.0% w/v LiCl/DMAc, and about one-quarter of the apparently bound LiCl was removed when the LiCl concentration of the solution was decreased from 1.0% to 0.5% w/v. Thus, when the A-chitin solution in 1.0% w/v LiCl/DMAc was diluted to 0.5% w/v LiCl/DMAc, the number of moles of LiCl salt apparently bound to one repeating unit of chitin may have been lower than 0.51, resulting in the formation of gel-like precipitate of the chitin (Figure S1 in the Supporting Information), because its dissolved state could not be maintained in 0.5% w/v LiCl/DMAc. Chitin molecules have two OH groups in each repeating unit, compared with three in each repeating unit of cellulose, so sufficient interactions between OH groups and LiCl may not be formed in such a dilute LiCl/DMAc solution. SEC/MALLS/RI Analysis of Celluloses. There has been no standard cellulose sample with a definite and known molecular mass. However, MCCs prepared from higher plant celluloses by dilute acid hydrolysis and commercially available MCCs are known to have almost constant degrees of polymerization (DP) or leveling-off DPs of 200−300. Thus, the molecular mass parameters of a MCC sample were determined from a 1.0% w/ v LiCl/DMAc solution, using the two dn/dc values, 0.131 and 0.065 mL/g, obtained by the two methods (Table 2). When the dn/dc value of 0.131 mL/g was used, the obtained calculated mass and weight-average DP (DPw) value were reasonable, while unacceptable DPw and calculated mass values were obtained for the MCC when the dn/dc value of 0.065 mL/g was used (Table 4). Thus, the dn/dc value of 0.131 mL/g obtained for the F-MFC solutions by the SEC/MALLS/RI method should be used. The SEC elution patterns and the corresponding molecular mass plots of 5 × 10−4 and 2.5 × 10−4 g/mL F-MFC dissolved in 1.0% and 0.5% w/v LiCl/DMAc are shown in Figure 4a. The molecular mass plots of the four samples followed almost the same line, indicating that the cellulose molecules of F-MFC were correctly separated according to their molecular mass by
(2)
(3)
where doff‑line RI is the dn/dc value for F-MFC or A-chitin in 1.0% or 0.5% w/v LiCl/DMAc solution obtained by the off-line E
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
Article
Biomacromolecules Table 4. Obtained Molecular-Mass Parameters of MCC
a
dn/dc (mL/g)
Mn
Mw
DPw
polydispersity (Mw/Mn)
injected mass (μg)
calculated mass (μg)
slope of conformation plota
0.131 0.065
19400 39000
29800 60100
184 371
1.5 1.5
50 50
46.0 92.6
0.60 0.60
Conformation plot: double logarithmic plot of molecular mass versus rms radius.
the SEC column used in this study. The corresponding relationships between the SEC elution volume and root mean square (rms) radius of the cellulose molecules of F-MFC are depicted in Figure 4b. Although small differences in the rms radius were observed between the solutions in 0.5% and 1.0% w/v LiCl/DMAc at the same elution volume, the lines were quite similar. It is likely that the cellulose molecules had slightly more expanded conformations or larger rms radii in 1.0% w/v LiCl/DMAc than in 0.5% w/v LiCl/DMAc. These results indicate that 1.0% w/v LiCl/DMAc is a better solvent for the cellulose molecules of F-MFC than in 0.5% w/v LiCl/DMAc. The lower the LiCl concentration in LiCl/DMAc, the lower the solubility of cellulose molecules in the solvent. The molecular mass parameters of F-MFC determined by SEC/MALLS analysis of the solutions with different F-MFC or LiCl concentrations are summarized in Table 5. The results show that most of the molecular mass parameters of the FMFC except Mn were quite similar to each other as long as LiCl/DMAc was used as the SEC eluent. The molecular mass parameters of F-MFC determined using 1.0% w/v LiCl/DMI were somewhat different from those determined using 1.0% w/ v LiCl/DMAc, as shown in Figure 5. The peak positions and distribution patterns were somewhat different between the two solvent systems. Consequently, the molecular mass parameters of cellulose determined using 1.0% or 0.5% w/v LiCl/DMAc as the eluent by SEC/MALLS/RI analysis using the dn/dc value of 0.131 mL/g are reliable, as long as the present measurement apparatus and conditions are used.
■
CONCLUSIONS
Because F-MFC was directly dissolvable in 8% w/w LiCl/ DMAc, the dn/dc values of F-MFC solutions with precisely measured F-MFC concentrations could be determined by the SEC/MALLS/RI method after dilution to 0.5% and 1.0% w/v LiCl/DMAc. The obtained dn/dc value of F-MFC was 0.131 mL/g. When the off-line RI method was used, the dn/dc values obtained were clearly lower than the above value. This
Figure 4. Effect of LiCl or F-MFC concentration on SEC elution patterns and corresponding molecular mass plots of different solutions (a), and relationship between SEC elution volume and rms radius of cellulose molecules determined by SEC/MALLS analysis (b).
Table 5. Molecular Mass Parameters of F-MFC in 1.0% w/v LiCl/DMAc, 0.5% w/v LiCl/DMAc, and 1.0% w/v LiCl/DMI F-MFC conc. (g/mL)
solvent
Mn
Mw
DPw
polydispersity (Mw/ Mn)
injected mass (μg)
calculated mass (μg)
slope of conformation plot
1.0% w/v LiCl/ DMAc 1.0% w/v LiCl/ DMAc 1.0% w/v LiCl/ DMAc 0.5% w/v LiCl/ DMAc 0.5% w/v LiCl/ DMAc
105000
227000
1400
2.2
100
92.5
0.60
68400
224000
1380
3.2
50
48.6
0.60
80100
231000
1420
2.9
25
24.1
0.60
63600
227000
1400
3.6
50
50.0
0.56
67700
226000
1390
3.3
25
25.0
0.57
average standard deviation
77000 16800
227000 2550
1400 15
3.0 0.5
− −
− −
0.59 0.02
1 × 10−3
146700
246000
1520
1.7
100
99.2
0.61
1 × 10−3 5 × 10−4 2.5 × 10−4 5 × 10−4 −4
2.5 × 10
1.0% w/v LiCl/DMI
F
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
Article
Biomacromolecules
Technology Agency (JST). We thank Mr. Masahide Nakamura of Shoko Co., Ltd., Japan, for helpful advice concerning dn/dc measurements using off-line RI.
■
(1) Wyatt, P. J. Light scattering and the absolute characterization of macromolecules. Anal. Chim. Acta 1993, 272, 1−40. (2) Dupont, A.-L.; Harrison, G. Conformation and dn/dc determination of cellulose in N,N-dimethylacetamide containing lithium chloride. Carbohydr. Polym. 2004, 58, 233−243. (3) Yanagisawa, M.; Isogai, A. SEC-MALS-QELS Study on the Molecular Conformation of Cellulose in LiCl/Amide Solutions. Biomacromolecules 2005, 6, 1258−1265. (4) Potthast, A.; Radosta, S.; Saake, B.; Lebioda, S.; Heinze, T.; Henniges, U.; Isogai, A.; Koschella, A.; Kosma, P.; Rosenau, T.; Schiehser, S.; Sixta, H.; Strlić, M.; Strobin, G.; Vorwerg, W.; Wetzel, H. Comparison testing of methods for gel permeation chromatography of cellulose: coming closer to a standard protocol. Cellulose 2015, 22, 1591−1613. (5) Berggren, R.; Berthold, F.; Sjöholm, E.; Lindström, M. Improved methods for evaluating the molar mass distributions of cellulose in kraft pulp. J. Appl. Polym. Sci. 2003, 88, 1170−1179. (6) McCormick, C. L.; Callais, P. A.; Hutchinson, B. H. R. Solution studies of cellulose in lithium chloride and N,N-dimethylacetamide. Macromolecules 1985, 18, 2394−2401. (7) Terbojevich, M.; Cosani, A.; Conio, G.; Ciferri, A.; Bianchi, E. Mesophase formation and chain rigidity in cellulose and derivatives. 3. Aggregation of cellulose in N,N-dimethylacetamide-lithium chloride. Macromolecules 1985, 18, 640−646. (8) Sjöholm, E.; Gustafsson, K.; Eriksson, B.; Brown, W.; Colmsjö, A. Aggregation of cellulose in lithium chloride/N,N-dimethylacetamide. Carbohydr. Polym. 2000, 41, 153−161. (9) Schult, T.; Hjerde, T.; Optun, O. I.; Kleppe, P. J.; Moe, S. Characterization of cellulose by SEC-MALLS. Cellulose 2002, 9, 149− 158. (10) Röhrling, J.; Potthast, A.; Rosenau, T.; Lange, T.; Ebner, G.; Sixta, H.; Kosma, P. A novel method for the determination of carbonyl groups in cellulosics by fluorescence labeling. 1. Method development. Biomacromolecules 2002, 3, 959−968. (11) Siller, M.; Ahn, K.; Pircher, N.; Rosenau, T.; Potthast, A. Dissolution of rayon fibers for size exclusion chromatography: a challenge. Cellulose 2014, 21, 3291−3301. (12) Ishii, D.; Isogai, A. The residual amide content of cellulose sequentially solvent-exchanged and then vacuum-dried. Cellulose 2008, 15, 547−553. (13) Kuramae, R.; Saito, T.; Isogai, A. TEMPO-oxidized cellulose nanofibrils prepared from various plant holocelluloses. React. Funct. Polym. 2014, 85, 126−133. (14) Shi, Z.; Yang, Q.; Ono, Y.; Funahashi, R.; Saito, T.; Isogai, A. Creation of a new material stream from Japanese cedar resources to cellulose nanofibrils. React. Funct. Polym. 2015, 95, 19−24. (15) Yanagisawa, M.; Shibata, I.; Isogai, A. SEC−MALLS analysis of cellulose using LiCl/1,3-dimethyl-2-imidazolidinone as an eluent. Cellulose 2004, 11, 169−176. (16) Matsumoto, T.; Tatsumi, D.; Tamai, N.; Takaki, T. Solution properties of celluloses from different biological origins in LiCl DMAc. Cellulose 2001, 8, 275−282. (17) Yanagisawa, M.; Shibata, I.; Isogai, A. SEC-MALLS analysis of softwood kraft pulp using LiCl/1,3-dimethyl-2-imidazolidinone as an eluent. Cellulose 2005, 12, 151−158. (18) Ishii, D.; Tatsumi, D.; Matsumoto, T. Effect of solvent exchange on the solid structure and dissolution behavior of cellulose. Biomacromolecules 2003, 4, 1238−1243. (19) Yamamoto, M.; Kuramae, R.; Yanagisawa, M.; Ishii, D.; Isogai, A. Light-scattering analysis of native wood holocelluloses totally dissolved in LiCl−DMI solutions: high probability of branched structures in inherent cellulose. Biomacromolecules 2011, 12, 3982− 3988.
Figure 5. Differential molecular mass distribution of F-MFC dissolved in 1.0% w/v LiCl/DMAc and 1.0% w/v LiCl/DMI, obtained by SEC/ MALLS analysis.
discrepancy is probably because part of LiCl acted as a component of cellulose-LiCl structures that formed between hydroxyl groups of cellulose and LiCl in the F-MFC/LiCl/ DMAc solutions, resulting in the lowering of the LiCl concentration surrounding each F-MFC molecule in the FMFC/LiCl/DMAc solutions, particularly for F-MFC solutions with high F-MFC concentrations (or those with large amounts of cellulose hydroxyl groups). In contrast, because the F-MFC concentration in each SEC elution volume in the SEC/MALLS analysis was sufficiently low, the LiCl concentration gradient in the SEC eluent had a negligible effect on the determination of the dn/dc value. Thus, when the dn/dc value of 0.131 mL/g determined by the SEC/MALLS/RI method was used to determine the molecular mass parameters of various F-MFC solutions and a MCC solution, reasonable and reliable data were obtained.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biomac.5b01302. A photograph of formation of gel-like precipitate of Achitin in 0.5% w/v LiCl/DMAc, SEC elution patterns and corresponding molecular mass plots of F-MFC in 0.5% w/v LiCl/DMAc and those of A-chitin in 1.0% w/v LiCl/DMAc, relationships between F-MFC, A-chitin, cellobiose, and LiCl concentration and differential refractive index of the solutions determined by the offline RI method, and details on equations to calculate the amount of LiCl apparently bound to cellulose or chitin (PDF)
■
REFERENCES
AUTHOR INFORMATION
Corresponding Author
* Tel.: +81 3 5841 5538. Fax: +81 3 5841 5269. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported by Core Research for Evolutional Science and Technology (CREST) of the Japan Science and G
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
Article
Biomacromolecules (20) Ono, Y.; Hiraoki, R.; Fujisawa, S.; Saito, T.; Isogai, A. SEC− MALLS analysis of wood holocelluloses dissolved in 8% LiCl/1,3dimethyl-2-imidazolidinone: challenges and suitable analytical conditions. Cellulose 2015, 22, 3347−3357. (21) Potthast, A.; Rosenau, T.; Sixta, H.; Kosma, P. Degradation of cellulosic materials by heating in DMAc/LiCl. Tetrahedron Lett. 2002, 43, 7757−7759. (22) Potthast, A.; Rosenau, T.; Sartori, J.; Sixta, H.; Kosma, P. Hydrolytic processes and condensation reactions in the cellulose solvent system N,N-dimethylacetamide/lithium chloride. Part 2: degradation of cellulose. Polymer 2003, 44, 7−17. (23) Isogai, T.; Yanagisawa, M.; Isogai, A. Degrees of polymerization (DP) and DP distribution of dilute acid-hydrolyzed products of alkalitreated native and regenerated celluloses. Cellulose 2008, 15, 815−823. (24) McCormick, C. L.; Callais, P. A.; Hutchinson, B. H., Jr. Solution studies on cellulose in lithium chloride and N,N-dimethylacetamide. Macromolecules 1985, 18, 2394−2401. (25) Spange, S.; Reuter, A.; Vilsmeier, E.; Heinze, T.; Keutel, D.; Linert, W. Determination of empirical polarity parameters of the cellulose solvent N,N-dimethylacetamide/LiCl by means of the solvatochromic technique. J. Polym. Sci., Part A: Polym. Chem. 1998, 36, 1945−1955. (26) Striegel, A. M. Advances in the understanding of the dissolution mechanism of cellulose in DMAc/LiCl. Journal of the Chilean Chemical Society 2003, 48, 73−77. (27) Yanagisawa, M.; Shibata, I.; Isogai, A. SEC-MALLS analysis of cellulose using LiCl/1,3-dimethyl-2-imidazolidinone as an eluent. Cellulose 2004, 11, 169−176.
H
DOI: 10.1021/acs.biomac.5b01302 Biomacromolecules XXXX, XXX, XXX−XXX
