Properties of decalin as a solvent in thermal field-flow fractionation

Summary Cis-trans-decahydronaphthalene (67.85% trans+ 32.15% cis, here briefly referred to ascis-trans-decalin) as a possible solvent for thermal field-flow fractionation is proposed. Different features such as solvent properties with respect to low polarity polymers, availability of basic physicochemical data (viscosity and thermal conductivity in a convenient temperature range), low toxicity and low cost are emphasized. Retention data over a wide range of cold wall temperatures and thermal gradients were collected. Thev correction factor and λ retention data were determined for polystyrene samples over a relative molecular mass range of 11,300–3,950,000 g mol⁻¹. From these data, selectivity values were determined and compared to the commonly found values for polystyrene in different solvent systems. The average value of −0.6 found here forcis-trans-decalin falls in the optimum selectivity solvent-domain.Cis-trans-decalin is thus fully proven as an optimum solvent for ThFFF.     

Magnitude and direction of thermal diffusion of colloidal particles measured by thermal field-flow fractionation

In this paper we provide experimental evidence showing that various types of submicrometer-sized particles (latexes, inorganic, and metallic), suspended in either aqueous or nonaqueous carrier liquids to which a temperature gradient dT/dx is applied, experience a force in the direction opposite to that of dT/dx. This behavior is similar to that of small particles such as soot, aerosols, and small bubbles suspended in stagnant gases across which temperature gradients are applied, a phenomenon known as "thermophoresis in gases." We report the use of a thermal field-flow fractionation (ThFFF) apparatus in two different configurations to establish the direction of particle motion subject to a temperature gradient. The first approach employed the conventional horizontal ThFFF channel orientation. In this case, small electrical potentials were applied across the narrow channel thickness either to augment or to act in opposition to the applied thermal gradient, depending on whether the accumulation wall was maintained at a positive or negative potential relative to the depletion wall. Thus, by observing the changes in the retention behavior of surface-charged latices or silica particles with changes in potential difference across the channel thickness, we were able to ascertain the direction of migration of the particles in the thermal gradient. The second approach involved the use of a ThFFF column oriented vertically in an implementation of a technique known as thermogravitational FFF. In this approach, the convective flow along the channel length (due to density gradients associated with the temperature gradient) couples with the thermal diffusion effect across the channel thickness to result in a combined particle retention mechanism. A retarded upward migration rate is indicative of accumulation of particles at the cold wall, while enhanced upward migration would indicate a hot-wall accumulation. From the results of our investigations, we conclude that submicrometer-sized particles suspended in either aqueous or nonaqueous carrier liquids and subjected to a temperature gradient migrate from the hot wall toward the cold wall of a ThFFF channel.     

Comparison of polymer resolution in thermal field-flow fractionation and size-exclusion chromatography

A comparison of the resolving power of themal field-flow fractionation (thermal FFF) and size-exclusion chromatography (SEC) has been made by the experimental operation of a high-resolution system from each category. For the experimental systems used, the measured resolution for thermal FFF was found to be higher than that for SEC for three different binary polymer mixtures. Only for a single low-molecular-weight mixture falling outside the optimal operating range of thermal FFF did SEC show better resolution. The experimental resolution values were broken down into selectivity and column efficiency parameters. While selectivity was easily obtained, column efficiency required a correction for polydispersity effects. With the polydispersity correction made on the basis of thermal FFF data, true resolution values were calculated to replace the apparent or experimental resolution levels. Overall, the corrected resolution values showed that thermal FFF had a significant advantage over SEC. Prospects for future advances in these two techniques are discussed.     

Study of temperature dependence of thermal diffusion in polystyrene/ethylbenzene by thermal field-flow fractionation

Using a special thermal field-flow fractionation apparatus capable of working over a broad temperature range, we have observed that retention in the polystyrene/ethylbenzene system decreases substantially as the cold wall temperature increases from 360 to 424 K. Polymers of four different molecular weights ranging from 20,000 to 160,000 were used to verify this conclusion. Based on our earlier work showing that thermal diffusion parameters could be calculated from retention data, we have used the present measurements along with earlier values to generate a compilation of thermal diffusion data over the temperature range from 270 to 424 K. These results are used to develop empirical expressions for the thermal diffusion factor and the thermal diffusion coefficient for polystyrene in ethylbenzene as a function of temperature and molecular weight. It is shown that these results have potential usefulness both in terms of the new physicochemical data obtained and in terms of the guidelines they provide for choosing experimental conditions for thermal field-flow fractionation experiments.     

Correlation between thermal diffusion and solvent self-diffusion in semidilute and concentrated polymer solutions

We have performed measurements of thermal diffusion coefficients DT and solvent self-diffusion coefficients Dss in semidilute to concentrated polymer solutions. Solutes of different glass transition temperatures and solvents of different solvent qualities have been used. The investigated systems are in detail: poly(dimethyl-siloxane) in toluene, tristyrene in toluene, polystyrene in toluene, polystyrene in tetrahydrofuran, polystyrene in benzene, and polystyrene in cyclohexane. The thermal diffusion data are compared to our data and literature data for solvent self-diffusion coefficients. In all systems the concentration dependence of DT closely parallels the one of Dss which may be viewed as a local probe for friction on a length scale of the size of one polymer segment. This identifies local friction as the dominating parameter determining the concentration dependence of DT. Solvent quality, in contrast, has no influence on DT.     

High-resolution polymer separations in a four-pass hairpin thermal field-flow fractionation column

A thin-channel four-pass hairpin thermal field-flow fractionation (FFF) column is described, and the advantages of its unique dimensional characteristics are explained. The problem of isolating the performance of this and other separations columns from the ubiquitous polydispersity effects are discussed and treated theoretically. The discussion is extended to size exclusion chromatography, and it is shown that the 2 to 6 times lower selectivity of the latter compared to FFF leads to the requirement for 4 to 36 times more theoretical plates to encounter polydispersity effects and thereby obtain information on polymer molecular weight distribution. A fractogram of six narrow polystyrene samples obtained from the hairpin system is shown to imitate closely the fractograms obtained from two totally different thermal FFF columns, showing that polydispersity dominates and that molecular weight information is revealed for these samples with only a few hundred theoretical plates. Various experimental and theoretical attempts are made to isolate the polydispersity and column contributions to plate height, including cut-and-recycle experiments, the observation of plate height versus velocity curves, and the direct calculation of the contributing effects. The various methods are subject to moderate errors, but are in rough agreement. The plate height plots show that the polydispersity effect contributed 53% and 68% to the measured plate height for 51,000 and 160,000 molecular weight samples, respectively. The latter polymer is shown to emerge with ca. 1300 true column plates. It is suggested that much higher column efficiency will be observed in the future if higher retention levels can be experimentally realized.     

Characterization of functionalized styrene-butadiene rubber by flow field-flow fractionation/light scattering in organic solvent

Flow field-flow fractionation (FlFFF) using an organic solvent as mobile phase has been effectively utilized for the separation and characterization of functionalized styrene-butadiene rubbers (SBR) that are polymerized and followed by coupling reaction in solution. Separation of broad molecular weight SBR was accomplished by an asymmetrical FlFFF channel in THF under field programming and the molecular weight distribution (MWD) of the SBR sample was determined by on-line measurement of light scattering. In this study, FlFFF has been utilized to characterize high-MW functionalized SBR from the low-MW non-functionalized molecules which were used for coupling reaction to produce high-MW functionalized SBRs, and to determine the coupling number of the functionalized SBRs depending on the type of the coupling reagents. The resulting MWD of the SBR samples prepared by the different coupling reagents (SnCl(4) and a polydimethylsiloxane compound) were compared.     

Lowering the molecular mass limit of thermal field-flow fractionation for polymer separations

Thermal field-flow fractionation (ThFFF) is capable of separating a wide molecular mass range of polymers by their molecular mass (Mr) and chemical composition. However, retention and resolution decrease significantly for polymers with Mr<20 kDa. Various approaches for increasing the retention of low Mr (<15 kDa) polymers were investigated. Our results showed that temperature conditions and single-component solvents had a limited effect on polymer retention and that certain binary solvent mixtures caused a dramatic increase in retention. The binary solvents approach has enabled the use of a ThFFF system and temperature conditions to separate 2.6 kDa PS from 4.4 kDa PS, thereby extending the applicability of ThFFF to lower molecular masses. The effect of binary solvent mixtures on polymer retention is correlated with the mixture viscosity.     

Characterization of humic materials by flow field-flow fractionation

Several humic materials are characterized by flow field-flow fractionation, including humic acids, a fulvic acid, and aqueous leachates from compost. Hydrophilic and hydrophobic fractions of a compost leachate were also examined. After characterizing molecular weight distributions, the effect of pH and salt concentration on hydrodynamic size is studied. In general, the hydrodynamic size decreases as the pH is lowered. However, humic acids form large aggregates below pH 5. Small amounts of sodium chloride have little effect on the size distributions. In contrast, a little calcium chloride reduces the hydrodynamic size of individual molecules while inducing the formation of oligomers, although severe aggregation is absent. With further additions of calcium chloride, the decrease in hydrodynamic size continues but oligomer formation subsides. Precise characterization of the unaggregated material is hindered by sample penetration through the channel membrane.     

Influence of operational parameters on retention of ultra-high molecular weight polystyrenes in thermal field-flow fractionation

Summary The influence of various parameters (concentration of the injected polymer solution, flow rate, temperature gradient, relaxation conditions) on the retention and shape of the fractogram of ultra-high molecular weight polystyrenes in thermal field-flow fractionation was investigated. Under the operating conditions adopted, reproducible oscillations in the peak shape are observed for molecular weights larger than a few millions, especially at relatively high polymer concentration. They are attributed to some hydrodynamic instabilities. The retention of ultra-high molecular weight polystyrenes at high flow rate is strongly dependent on the initial relaxation period. All of the investigated operational variables have a complex effect on the resulting shape of the fractogram. Consequently, the confirmation of the shear-induced focusing of macromolecules across the channel thickness requires further study. Experiments on reinjection of fractions collected after an initial high speed pass through the separation system leads to the conclusion that shear degradation of the ultra-high molecular weight polymers did not occur under the investigated experimental conditions even at the highest flow rates.     

Thermal diffusion and molecular weight calibration of poly(ethylene-co-vinyl acetate)s by thermal field-flow fractionation

Copolymer characterization is accomplished with respect to measurement of thermal diffusion coefficient (D_T) and molecular weight determination by thermal field-flow fractionation. The examined copolymers are the eight poly(ethylene-co-vinyl acetate)s [P(E-V)] having different compositions of vinyl acetate ranging from 25 to 70% and the molecular weight from 110,000 to 285,000, and three polyvinyl acetate standards as component homopolymer. The carrier solvents are tetrahydrofuran, toluene, and chlorobenzene which have different viscosities and thermal conductivities. Measured D_T values vary from 1.36 × 10^?8 to 5.97 × 10_?8 cm²/(s . K) which are dependent on the composition of copolymers and types of carriers. These values increase linearly with the increase of weight percent of vinyl acetate. It is possible to estimate D_T values of polyethylene from the extrapolated intercept in the plots of D_T vs. vinyl acetate wt % of copolymer. Tetrahydrofuran is found to be the appropriate carrier solvent for the separation of P(E-V) copolymers since D_T varies greatly with the increase of wt % in THF. Attempts are made to correlate the measured retention data with molecular sizes of copolymers for the construction of the molecular weight calibration curve. Good correlations (r² ≥ 0.931) are found in which D/D_T values of polymers vary inversely with the product of hydrodynamic volume by weight ratio of vinyl acetate. Based on this relationship, the unknown molecular weight of copolymer sample can be determined from component homopolymers for which standards are readily available. © 1995 John Wiley & Sons, Inc.     

Quantitative characterization by asymmetrical flow field-flow fractionation of IgG thermal aggregation with and without polymer protective agents

Complexes formed between poly(acrylates) and polyclonal immunoglobulin G (IgG) in its native conformation and after heat stress were characterized using asymmetric flow field-flow fractionation (AF4) coupled with on-line UV-Vis spectroscopy and multi-angle light-scattering detection (MALS). Mixtures of IgG and poly(acrylates) of increasing structural complexity, sodium poly(acrylate) (PAA), a sodium poly(acrylate) bearing at random 3 mol % n-octadecyl groups, and a random copolymer of sodium acrylate (35 mol %), N-n-octylacrylamide (25 mol %) and N-isopropylacrylamide (40 mol %), were fractionated in a sodium phosphate buffer (0.02 M, pH 6.8) in the presence, or not, of 0.1 M NaCl. The AF4 protocol developed allowed the fractionation of solutions containing free poly(acrylates), native IgG monomer and dimer, poly(acrylates)/IgG complexes made up of one IgG molecule and a few polymer chains, and/or larger poly(acrylates)/IgG aggregates. The molar mass and recovery of the soluble analytes were obtained for mixed solutions of poly(acrylates) and native IgG and for the same solutions incubated at 65 °C for 10 min. From the combined AF4 results, we concluded that in solutions of low ionic strength, the presence of PAA increased the recovery ratio of IgG after thermal stress because of the formation of electrostatically-driven PAA/IgG complexes, but PAA had no protective effect in the presence of 0.1 M NaCl. Poly(acrylates) bearing hydrophobic groups significantly increased IgG recovery after stress, independently of NaCl concentration, because of the synergistic effect of hydrophobic and electrostatic interactions. The AF4 results corroborate conclusions drawn from a previous study combining four analytical techniques. This study demonstrates that AF4 is an efficient tool for the analysis of protein formulations subjected to stress, an important achievement given the anticipated important role of proteins in near-future human therapies. ?      

On void time determination in thermal field-flow fractionation

Because of the temperature dependence of the carrier liquid density, the mass of carrier which is contained in a thermal field-flow fractionation channel depends on the cold wall temperature and on the temperature difference across the channel thickness. It is observed that the void time of the solvent peak decreases when increasing the average temperature in the channel. The void time is found to be directly proportional to the average carrier density in the channel. The determination of the void time from the knowledge of the channel geometrical volume and the measurement of the volumetric flow-rate leads to significant errors if the thermal expansion of the carrier between the temperature of the measurement and the average channel temperature is not taken into account. Recommendations are given for proper void time determinations in thermal FFF.     

High temperature thermal field-flow fractionation of polyethylene and polystyrene

In this paper the high-temperature thermal field flow fractionation method is exploited for the analysis of polyethylene (PE). The experimental apparatus set-up, obtained by simply modifying a commercial instrument, is presented. The numerical procedure for deriving retention calibration plot versus molecular weight is discussed with reference to the specific polymer-solvent pair, PE-o-dichlorobenzene (ODCB), here employed. Different methods for computing the physicochemical data set of the solvent, necessary for calibration, are compared. The selectivity of the checked PE-ODCB system proves comparable with respect to the values currently found in thermal field-flow fractionation (ThFFF) analysis. Differences are found between PE and polystyrene (PS) analysis in the same solvent. The conditions for high temperature ThFFF operation in PE analysis and their advantages are discussed with respect to the standard SEC technique for PE, PS, and PE-PS copolymer analysis. Molecular weight distributions obtained by ThFFF of two PE commercial samples agree with those obtained by SEC. © 1995 John Wiley & Sons, Inc.     

Star polymer diffusion in concentrated solutions: Dependence on functionality via the coupling model

The coupling model is formulated in terms of the dynamic constraint entropy, S_c(t), and subsequently compared with measured tracer diffusion coefficients D for linear, 3-arm star, and 12arm star polystyrenes in solutions of poly (vinyl methyl ether) (PVME) /o-fluorotoluene. Inequalities between the dynamic constraint entropies, S_cf and S_cf′(t), respectively, for f-arm and f′-arm stars, enable us to explain the experimental observation that at constant arm molecular weight, the ratio D_star/D_linear decreases substantially with increasing concentration above the entanglement concentration for the PVME. This work indicates that not only the reptation model, but also the coupling model, can account for the experimental observation that in entangled solutions the mechanism for diffusion depends on diffusant architecture.     

Thermal diffusion of dilute polymer solutions: the role of solvent viscosity

We have performed measurements of the thermal diffusion coefficient D(T) in the dilute limit on polystyrene in cyclo-octane, cyclohexane, benzene, toluene, tetrahydrofuran, ethyl acetate, and methyl ethyl ketone and of poly(dimethyl-siloxane) in toluene. These data have been combined with literature data to test various theoretical predictions. The viscosity is identified as the dominating and only relevant solvent parameter. On the polymer side, the size or mass of an effective correlated segment determines the strength of the Soret effect. Large and heavy effective segments, as found in stiffer chains, lead to higher D(T).     

Separation and composition distribution determination of triblock copolymers by thermal field-flow fractionation

Thermal field-flow fractionation (ThFFF) is used to separate a linear triblock copolymer of polystyrene, poly(tert-butyl acrylate), and poly(methyl methacrylate) by composition. Fractions were collected and subjected to off-line NMR analysis. The resultant mole fraction versus retention time plots for each of the three polymer components confirmed the success of the separation and yielded the composition distribution of the copolymer. The composition distribution was also obtained using a second approach that involved solving a series of equations comprised of polymer thermal diffusion coefficients and quasi-elastic light scattering, differential refractometry, and UV detector responses. Both sets of data showed similar trends of composition variations in each polymer component as a function of retention time. However, discrepancies were observed in the mole fraction values. The ability to compositionally separate and to determine composition distribution of copolymers is important as demonstrated by the presence of diblock impurities in the ThFFF with off-line NMR results.