Following cellulose depolymerization in paper: comparison of size exclusion chromatography techniques

The study focuses on the comparison of the results obtained by size exclusion chromatography (SEC) detection systems using multiangle laser light scattering (MALLS) and/or ultraviolet–visible (UV/VIS) detectors in analyses of cellulose in paper-based heritage objects. The commonly applied parameter for the evaluation of paper degradation stage and kinetics is weight (M _w ) average molar mass measured by SEC. The main problems addressed here are whether the parameters obtained by various techniques and calibration strategies and in various laboratories can be treated universally and what the sources of the discrepancies can be. The SEC-UV/VIS technique proved to give reproducible results as demonstrated by the interlaboratory correlations. Using various data processing methods and SEC techniques it was also shown that molar masses should only be treated relatively. The differences between the kinetic curves based on the M _w values obtained from various techniques and processed with various calibration procedures question the reliability of kinetic equations derived using the SEC results to describe the paper degradation.     

Structure characterization of hyperbranched poly(ether amide)s. I. Preparative fractionation

The focus of our investigation lies on the separation of typically broadly distributed hyperbranched poly(ether amide)s into narrow fractions of various molar masses. Their exact molar mass found via size-exclusion chromatography (SEC) with light uttering detection allows us to use these fractions for sample specific calibration in the SEC investigation of other hyperbranched samples. The analysis of the degree of branching, molar mass and viscosity behavior of the fractions gives a first indication about their molecular shape and the contribution of that shape to the overall viscosity. We determined the Mark-Houwink exponent for a hyperbranched sample using a number of narrow fractions which showed that an increase of molar mass leads to an increased molecular density.     

Degradation of cellulosic insulation in power transformers: a SEC–MALLS study of artificially aged transformer papers

The molecular weight distributions of artificially aged transformer papers were studied using SEC–MALLS with 0.5 % DMAc/LiCl as solvent, providing chain length distributions and averages. The slow dissolution and presence of non-cellulosic contaminants tend in some cases to introduce errors in the DP_w estimates, and a modified data processing procedure was developed to correct for extraneous scattering. Data were compared to the intrinsic viscosities obtained in 0.5 M Cuen and the ‘viscosimetric DP’ calculated thereof according to ISO 5351. DP_visc was 2–3 times lower than the weight average DP (DP_w), but closer to or slightly above the number average DP (DP_n) obtained by SEC–MALLS in the case of pure cellulose. Ageing of transformer papers were in some cases associated with changes in the polydispersity (DP_w/DP_n). The apparent degradation rate (defined as δ(1/DP)/δt) gradually decreased with time, resulting in a tendency for a ‘level-off’ DP in the range 200–300 for DP_visc and DP_n, and roughly 1,000 for DP_w.     

TEMPO-mediated Oxidation of Native Cellulose: SEC–MALLS Analysis of Water-soluble and -Insoluble Fractions in the Oxidized Products

Linter cellulose was suspended in water and oxidized by the NaClO/NaBr/2,2,6,6-tetramehylpiperidine-1-oxy radical (TEMPO) system at pH 10.5 (TEMPO-mediated oxidation), and the oxidized products were separated into several fractions by filtration and centrifugation, depending on their particle sizes and apparent water-solubility. The major fraction (>ca. 80 mass % of the original linter cellulose) is the filter paper-trapped fibers, which can form inter-fiber hemiacetal linkages when handsheets are prepared thereof. Size-exclusion chromatographic analysis with multi-angle laser light scattering detection (SEC–MALLS) of these fibrous fractions dissolved in 0.5% LiCl/N,N-dimethylacetamide (DMAc) showed that some depolymerization occurred on cellulose chains during the TEMPO-mediated oxidation. On the other hand, the apparently water-soluble fractions (<ca. 20 mass % of the original linter cellulose) in the TEMPO-oxidized linter cellulose consisted of small amounts of colloidal particles having the cellulose I crystal structure, which came off from linter cellulose by the TEMPO-mediated oxidation and were mixed in the apparently water-soluble fraction even after filtration using 0.45 μm membrane. The presence of such colloidal cellulose crystals in the water-soluble fractions of the TEMPO-oxidized linter cellulose brings about anomalous bimodal SEC-elution patterns and extremely large molecular-mass values calculated from the SEC–MALLS data. Truly water-soluble cellouronic acid and/or over-oxidized compounds having glucuronic acid and hexeneuronic acid units are also present in the water-soluble fractions.     

Discussion of substantially identical gel points among multiallyl monomers based on characterization of resultant network polymer precursors consisting of oligomeric primary polymer chains

No difference in the actual gel points was substantially observed among three isomeric diallyl phthalates such as diallyl phthalate (DAP), diallyl isophthalate, and diallyl terephthalate (DAT); this interesting gelation behavior was discussed further in terms of the correlation between gelation and the difference in cyclization modes, and also, the difference in reactivity between the uncyclized and cyclized radicals for cross‐linking. In the present work, we tried to extend the preceding discussion to the polymerization of triallyl trimellitate (TAT) because the molecular structure of TAT is presumed to essentially involve the characteristics of three isomeric diallyl phthalates and, therefore, the enhanced gelation was expected in TAT polymerization. However, no enhancement of gelation was observed. For a full understanding of the gelation in multiallyl cross‐linking polymerization, we explored further the polymerizations of DAP, DAT, and TAT, especially focusing on the characterization of resultant network polymer precursors (NPPs) using SEC‐MALLS‐viscometry providing the correlation of [η] versus M_w of fractionated samples. Notably, the structure of NPP consisting of oligomeric primary polymer chains generated from specific allyl polymerization would become core‐shell type dendritic with the progress of polymerization. The correlation between delayed gelation and decreased reactivity of dendritic NPP for intermolecular cross‐linking is discussed. Conclusively, the reactivity for intermolecular cross‐linking between NPPs decreased with the progress of polymerization leading to a delayed gelation. © 2009 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 47: 2871–2881, 2009     

Relationship between Intrinsic Viscosity and Molecular Weight for Fractionated Cellulose

A relationship between intrinsic viscosity and molecular weight for fractionated cellulose is established. The exponent a of the Mark-Houwink equation is 0.80. The unperturbed dimensions of cellulose are discussed on the basis of the Stockmayer-Fixman and Kurata-Stockmayer viscometric theories, and it is determined that cellulose is not a very rigid macromolecule in nature. This conclusion disagrees with the results obtained by statistical mechanics procedures in which the restriction of β-glucose residues to the C1 conformation was kept. The steric hindrance factor, σ is about 2, and it seems to be independent of the substituents in cellulose if it is evaluated in both cellulose and cellulose derivatives from viscometric data. From statistical mechanics data reported in the literature, σ is estimated as 4.4 for cellulose and cellulose derivatives, such as cellulose nitrate, and this means that the steric hindrances of the substituents have little influence on the rigidity of cellulose derivatives. This view disagrees with the results obtained from the non-Newtonian behavior of cellulose and cellulose nitrate in dilute solutions. By applying the Noda-Hearst theory, the conclusion is reached that cellulose nitrate is a more extended macromolecule than cellulose.     

SEC–MALLS analysis of ethylenediamine-pretreated native celluloses in LiCl/<Emphasis Type="Italic">N</Emphasis>,<Emphasis Type="Italic">N</Emphasis>-dimethylacetamide: softwood kraft pulp and highly crystalline bacterial,tunicate, and algal celluloses

Softwood and hardwood bleached kraft pulps (SBKP and HBKP, respectively) and highly crystalline native celluloses such as algal, tunicate, bacterial and cotton lint celluloses were dissolved in 8 % (w/v) LiCl/N,N-dimethylacetamide (DMAc) after ethylenediamine (EDA) pretreatment. Complete dissolution of SBKP and other highly crystalline native celluloses in 8 % LiCl/DMAc was achieved after solvent exchange from EDA to DMAc through methanol. Neutral sugar composition analysis showed no significant differences between the original and EDA-treated pulps. A combination of size-exclusion chromatography and multi-angle laser light scattering (SEC–MALLS) was used to analyze the cellulose solutions after dilution to 1 % (w/v) LiCl/DMAc. The 0.05 % (w/v) solutions of highly crystalline cellulose in 1 % (w/v) LiCl/DMAc contained entangled molecules, and therefore 0.025 % (w/v) cellulose solutions in 1 % (w/v) LiCl/DMAc were used in the SEC–MALLS analysis to obtain reliable conformation plots (or double-logarithmic plots of molecular mass vs. root-mean-square radius). All the cellulose samples except SBKP gave conformation plots with slope values of 0.56–0.57, showing that these cellulose molecules had random-coil conformations. In contrast, SBKP gave a slope value of 0.35, indicating that some branched structures were present in the high-molecular-mass fraction. Double-logarithmic plots of the reduced viscosities of the cellulose solutions in 1 % (w/v) LiCl/DMAc versus the molecular mass were linear, except for SBKP, also suggesting the presence of anomalous cellulose structures in SBKP.     

Characterization of Water-Soluble Cellulose Derivatives in Terms of the Molar Mass and Particle Size as well as Their Distribution

The property profile of cellulose derivatives dissolved in aqueous solvents is not only dependent on the chemical composition (average-, molar- or regiospecific degree of substitution, as well as the substitution along the chain), solvent, temperature and concentration but also on the molar mass and the particle size. All this information can be obtained from the Mark-Houwink-Sakurada-relationship ([;gh]-M-) or the R_G-M-relationship, if these are at hand. These relationships are suitable for a specific degree of substitution. The R_G-M-relationship has only been determined and published for a few water-soluble cellulose derivatives. The prerequisite is the availability of a homologous series of samples with the same chemical composition. In this paper it is shown that only the ultrasonic degradation is able to create such a series. Due to the ability of coupled methods of analysis to acquiring absolute data, molar mass and particle size distributions have been compiled in recent years. Using such methods it was possible to determine molar mass and particle size distributions of several aqueous cellulose derivative solutions by combining a fractionation unit (size exclusion chromatography (SEC) or flow field-flow fractionation (FFFF)) with multi angle laser light scattering (MALLS) for the detection of Mw and R_G and concentration detection (DRI). Results for nonionic cellulose ethers, mixed cellulose ethers, ionic carboxymethyl cellulose, sulfoethyl cellulose, hydrophobically modified hydroxyethyl cellulose were obtained and are partially discussed with focus on the recovery of cellulose derivates after fractionation and the impact on the distribution functions.     

Mesophase formation and intrinsic viscosity for some low-molecular-weight cellulose acetate samples

The intrinsic viscosity [η] of cellulose acetate (CA) has been determined in two solvents: acetic acid (AA) and N,N-dimethylacetamide (DMAC). The samples of this cellulose derivative have a rather weak mean molecular weight MW, with a mean degree of polymerization DP ranging from 213 to 47 and the same origin. We use the theory of Yamakawa for a worm-like chain to obtain the hydrodynamic diameter D and the persistence length q for these samples of CA, and the Mark-Houwink equation to estimate the parameters K and α. These results were compared with the same parameters obtained with previous experimental published results for CA with a high MW. The measurements were also carried out at different temperatures; the intrinsic-viscosity temperature dependence is discussed. The D and q values calculated for the low-MW range allow estimation of the critical concentrations C* for the solutions, which are compared with experimental results.     

紫外光引发硅氢加成反应制备超支化聚碳硅氮烷

在二乙酰丙酮铂存在下, 以紫外光引发AB4型单体双(N,N-二烯丙基胺基)甲基硅烷发生硅氢加成反应, 制备了超支化聚碳硅氮烷. 聚合产物通过FTIR, 1H NMR, 13C NMR和29Si NMR谱和体积排除色谱/激光光散射联用技术进行表征. 与常规加热条件下Karstedt's催化剂催化的硅氢加成聚合相比, 光引发聚合的反应速度快. 波谱分析表明, 聚合过程中以α位硅氢加成反应为主. 光引发聚合制备的超支化聚碳硅氮烷支化度(DB)和平均支化数(ANB)分别为0.46和0.53, 与理论值相近. 其重均分子量为12500 g/mol, 分子量分布系数为2.1, Mark-Houwink方程指数α为0.44, 与热聚合制备的超支化聚碳硅氮烷的参数相近.     

稀土催化聚合的顺式聚丁二烯的单分散粘度方程

对十一个未经分级的多分散稀土顺丁橡胶试样作了渗透压和粘度的测量,计算得到的数均分子量_n及特性粘数[<η>]列于表中。     

A study of the chemical and physical structure of polycaproamide obtained in anionic polymerization of caprolactam in solvent

The chemical and physical structure of polycaproamide obtained by low temperature anionic polymerization of caprolactam in solvent in the presence of the sodium salt of caprolactam and carbon dioxide, has been investigated. Solution behaviour of unfractionated and fractionated samples of polymer in 96% sulphuric acid, m-cresol and mixed solvent (2,2,3,3-tetrafluoropropanol-10% H₂O-0.1 M LiCl) was studied by viscosity measurements. The constancy of the Huggins viscometric coefficient K′ in 96% H₂SO₄ and m-cresol enables determination of limiting viscosity number from measurement at one concentration. On the basis of number-average molecular masses and limiting viscosity numbers, the constants of the Mark-Houwink equation were calculated for $\text{M}{}_{\mathrm{<mtext>n</mtext>}}$ in the range 3 × 10³ to 62 × 10³. They are fairly similar to established values for hydrolytic and anionic polycaproamides. The results confirmed our previous suggestion concerning the linear structure of this polyamide. X-ray analysis indicated high degree of crystallinity. The final, chemical treatment causes changes in chemical and physical structure of this polycaproamide. It differs in some respects from hydrolytic and anionic polycaproamides produced in bulk polymerizations of caprolactam. The findings lead to an understanding of several properties of the polycaproamide.     

Separation and enantioseparation of derivatized amino acids and biogenic amines by high-performance liquid chromatography with reversed and chiral stationary phases

A lignosulfonate sample was fractionated according to the solubility in ethanol-water. The fractions were analysed by aqueous size exclusion chromatography (SEC) combined with in-line multi-angle laser light scattering (MALLS), and by static MALLS. Satisfactory SEC results were obtained with aqueous phosphate buffer containing DMSO and SDS. The refractive index increment (dn/dc) varied from 0.186 to 0.205 ml/g, depending on Mw and the degree of sulfonation. The second viral coefficient (A2) was 7 x 10(-3) ml mol/g2. The weight-average molecular weight (Mw) of the fractions varied from 4600 up to 398 000 g/mol. and the polydispersity (Mw/Mn) varied between 1.3 and 3.5.     

Degrees of polymerization (DP) and DP distribution of cellouronic acids prepared from alkali-treated celluloses and ball-milled native celluloses by TEMPO-mediated oxidation

Various cellulose II samples, ball-milled native celluloses and ball-milled wood saw dust were subjected to 2,2,6,6-tetramethypyperidine-1-oxyl radical (TEMPO)-mediated oxidation to prepare cellouronic acid Na salts (CUAs). The TEMPO-oxidized products obtained were analyzed by ¹³C-NMR and size-exclusion chromatography (SEC). When the cellulose II samples with degrees of polymerization (DP) of 220–680 were used as the starting materials, the CUAs obtained had weight-average DP (DPw) values of only 38–79. Thus, significant depolymerization occurs on cellulose chains during the TEMPO-mediated oxidation. These DP values of CUAs correspond to the cellulose II crystal sizes along the chain direction in the original cellulose II samples, but not necessarily to their leveling-off DP values. CUAs can be obtained also from ball-milled native celluloses in good yields by TEMPO-mediated oxidation, although their DPw values are lower than about 80. On the other hand, CUA with DPw of about 170 was obtained from ball-milled wood saw dust.     

Exploration of cardanol-based phenolated and epoxidized resins by size exclusion chromatography and MALDI mass spectrometry

Cardanol and cardanol derivatives are among the most important biobased materials currently investigated in green chemistry, as renewable and promising building blocks in lieu of traditional raw materials from non renewable resources, in particular owing to the olefinic linkages on the C15 alkyl side-chain. Despite the increasing interest they arouse, analytical chemistry dedicated to cardanol and associated resins has been rarely reported in the literature, found even poorer when dealing with chromatography and mass spectrometry. In this work, a thorough molecular characterization was conducted using matrix assisted laser desorption ionization (MALDI) mass spectrometry, size exclusion chromatography (SEC), and SEC–MALDI coupling to gain insights into the composition of phenolated, epoxidized, and epoxidized phenolated cardanol. A nomenclature was proposed to properly describe the numerous species found in these materials, while simulations of the unsaturation patterns and their comparison with the detected patterns in MALDI-MS gave useful details about the phenolation treatment expected to occur on the polyunsaturated C15 side chain. Finally, the SEC–MALDI off-line coupling allowed SEC peaks to be deconvoluted by mass spectrometry and MALDI artefacts related to matrix adduction to be pointed out.     

Stimuli‐responsive ampholytic terpolymers of N‐acryloyl‐valine,acrylamide, and (3‐acrylamidopropyl)trimethylammonium chloride: Synthesis,characterization, and solution properties

Low‐charge density ampholytic terpolymers composed of acrylamide (AM), (3‐acrylamidopropyl)trimethyl ammonium chloride (APTAC), and N‐acryloyl‐valine were prepared via free‐radical polymerization in 0.5 M NaCl to yield terpolymers with random charge distributions. Sodium formate (NaOOCH) was employed as a chain transfer agent during the polymerization to suppress gel effects and broadening of the molecular weight distribution (MWD). Terpolymer compositions were determined by ¹³C NMR spectroscopy. Terpolymer molecular weights (MWs) and polydispersity indices (PDIs) were obtained via size exclusion chromatography/multi‐angle laser light scattering (SEC‐MALLS). Intrinsic viscosity values determined from SEC‐MALLS data using the Flory–Fox relationship were compared with those determined by low‐shear dilute solution viscometry and found to be in good agreement. SEC‐MALLS experiments allowed examination of radius of gyration‐MW (R_g‐M) relationships and the Mark‐Houwink‐Sakurada intrinsic viscosity‐MW ([η]‐M) relationships for terpolymers. The R_g‐M and [η]‐M relationships indicated little or no excluded volume effects under SEC conditions indicating that the terpolymers were in near theta conditions in an aqueous buffer solution. Potentiometric titration experiments were performed in deionized (DI) water. These studies revealed that the apparent pK_a of the AMVALTAC terpolymers increases with increasing VAL content. The solution properties of low‐charge density ampholytic terpolymers have been studied as functions of solution pH, ionic strength, and polymer concentration. The charge‐balanced terpolymers exhibit polyampholyte behavior at pH values ≥ 6.5. As solution pH is decreased, these charge‐balanced terpolymers become increasingly cationic due to the protonation of the VAL repeat units. Charge‐imbalanced terpolymers generally exhibit polyelectrolyte behavior, although the effects of intramolecular electrostatic interactions (e.g., polyampholyte effects) on the hydrodynamic volume are evident at certain values of solution pH and salt concentration. The solution behavior of the terpolymers in the dilute regime correlates well with that predicted by various polyampholyte solution theories. © 2006 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 44: 3125–3139, 2006     

pH‐Responsive ampholytic terpolymers of acrylamide,sodium 3‐acrylamido‐3‐methylbutanoate,and (3‐acrylamidopropyl)trimethylammonium chloride. I. Synthesis and characterization

Low-charge-density ampholytic terpolymers composed of acrylamide, sodium 3-acrylamido-3-methylbutanoate (NaAMB), and (3-acrylamidopropyl)trimethylammonium chloride were prepared via free-radical polymerization in 0.5 M NaCl to yield terpolymers with random charge distributions. NaOOCH was used as a chain-transfer agent during the polymerization to eliminate the effects of the monomer feed composition on the degree of polymerization (DP) and to suppress gel effects and broadening of the molecular weight distribution. The terpolymer compositions were obtained via ¹³C NMR spectroscopy, and the residual counterion content was determined via elemental analysis for Na⁺ and Cl⁻. The molecular weights (MWs) and polydispersity indices (PDIs) were determined via size exclusion chromatography/multi-angle laser light scattering (SEC–MALLS); the terpolymer MWs ranged from 1.3–1.6 × 10⁶ g/mol, corresponding to DPs of 1.6–1.9 × 10⁴ repeat units, with all terpolymers exhibiting PDIs of less than 2.0. Intrinsic viscosities determined from SEC–MALLS data and the Flory–Fox relationship were compared to intrinsic viscosities determined via low-shear dilute-solution viscometry and were found to agree rather well. Data from the SEC–MALLS analysis were used to analyze the radius of gyration/molecular weight (R_g–M) relationships and the Mark–Houwink–Sakurada intrinsic viscosity/molecular weight ([η]–M) relationships for the terpolymers. The R_g–M and [η]–M relationships revealed that most of the terpolymers exhibited little or no excluded volume effects under size exclusion chromatography conditions. Potentiometric titration of terpolymer solutions in deionized water showed that the apparent pK_a value of the poly[acrylamide-co-sodium 3-acrylamido-3-methylbutanoate-co-(3-acrylamidopropyl)trimethylammonium chloride] terpolymers increased with increasing NaAMB content in the terpolymers and increasing ratios of anionic monomer to cationic monomer at a constant terpolymer charge density. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 3236–3251, 2004     

聚乙酸乙烯酯的溶液性质及长链支化度

用粘度法,GPC和LALLS测定了线型及不同转化率的PVAc分级级份的粘度与分子量。提出了以线型和支化聚合物的K,α计算临界分子量的方法。讨论了表征PVAc长链支化的各种参数与分子量和转化率之间的关系以及不同条件下迭代法计算的支化频率λ的差异。实验结果表明,特性粘数和数均分子量乘积所表示的流体力学体积更适合GPC的普适标定概念。     

A hint on the correlation between cellulose fibers polymerization degree and their thermal and thermo-oxidative degradation

The degree of polymerization is one of the main parameters reflecting cellulose ageing. Viscometry is a method frequently used for determination of cellulose fibers polymerization degree, however, sample preparation and viscosity measurement are demanding, time consuming and do not provide reproducible results. In this study, the relationship between polymerization degree of cellulose fibers obtained by viscosity measurement and their degradation parameters obtained by thermal analysis were investigated. Differential scanning calorimetry provided values of effective combustion heat and thermogravimetry (TG) was used for the determination of temperatures and associated mass losses during the cellulose degradation. Effective combustion heat did not show any correlation with degree of polymerization of investigated cotton fabrics. In contrast, results from TG suggested several promising nonlinear correlations which could be used as a hint to develop a method useful for quick determination of cotton fabrics polymerization degree. The most promising correlations with cellulose polymerization degree were found for both rates of thermal and thermo-oxidative degradations.     

Universal calibration in GPC: A new approach for the calculation of molecular weights

A method has been developed for determining simultaneously the molecular weight of a broad-distribution polymer and the Mark-Houwink coefficients for that polymer type by using only GPC and intrinsic viscosity data. Standardized samples of poly(vinyl chloride), polystyrene, polybutadiene, and an experimental cycloolefin polymer were analyzed by this method. Shear-corrected intrinsic viscosities were used in all cases because of the high molecular weights involved. Molecular weight data for all samples were found to be in good agreement with molecular weight data obtained by membrane osmometry and from other GPC techniques. The proposed technique provides a means for calculating the molecular weight of a single polymer sample through universal calibration of GPC without knowledge of the Mark-Houwink coefficients for that polymer type.