Ring‐chain interactions in solutions of cyclic and linear poly(methylene) ester oligomers in p‐dioxan and chloroform

The thermodynamic behavior of ternary solutions containing one cyclic (CYC10 and CYC20) and one linear (LIN10 and LIN20) poly(methylene) ester oligomer, respectively, in p‐dioxan and chloroform were investigated through vapor‐pressure measurements. Experiments were carried out at 25, 30, 35, and 40 °C. The ring‐chain interaction parameters χ″ were evaluated using the Leonard's equation for a ternary polymer solution containing one ring polymer. Results indicated that the values of χ″ are concentration and solvent dependent, and they vary with molecular size and slightly with temperature. All χ″ values were positive for both CYC10/LIN10 and CYC20/LIN20 blends, with the higher values for the former system, indicating that both ring‐chain oligomer blends are nearly immiscible, and the difference between cyclic and linear oligomers decreases as molecular size increases. The positive χ″ agrees with differential scanning calorimetric measurements performed on ring‐chain oligomer blends that show two distinct endotherms for the blends. © 2002 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 40: 1425–1433, 2002     

Solvent quality as reflected in concentration‐ and temperature‐dependent Flory–Huggins interaction parameters

Flory–Huggins interaction parameters (χ) between poly(dimethylsiloxane) (weight‐average molecular weight = 152 kg/mol) and various solvents (methyl ethyl ketone, toluene and n‐octane) were determined as a function of composition and temperature with vapor‐pressure measurements. These data, complemented by independent information for dilute and very concentrated solutions, serve as the basis for a discussion of solvent quality via different theoretical relations. Regardless of polymer concentration, the χ values fall from methyl ethyl ketone via toluene to n‐octane, the ketone being the worst solvent and the hydrocarbon being the best solvent. The variation of χ with composition and temperature is complex. Within the range of moderate polymer concentrations, the influences of composition decrease with increasing solvent quality. Additional effects become noticeable at the ends of the composition scale. The enthalpy parts (χ_H) and entropy parts (χ_S) of the Flory–Huggins interaction parameter, obtained from χ(T), vary considerably with composition and change their sign in some cases; these constituents of the Flory–Huggins interaction parameter do not permit a direct assessment of solvent quality. A clear‐cut picture is, however, regained with a comparison of the interdependence of χ_S and χ_H. The elimination of explicit concentration influences re‐establishes the order in the solvent quality setup via χ. © 2001 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 39: 651–662, 2001     

Blends of polycarbonate and poly(ϵ‐caprolactone) and the determination of the polymer–polymer interaction parameter of the two polymers

The thermal properties of blends of polycarbonate (PC) and poly(ε‐caprolactone) (PCL) were investigated by differential scanning calorimetry (DSC). From the thermal analysis of PC‐PCL blends, a single glass‐transition temperature (T_g) was observed for all the blend compositions. These results indicate that there is miscibility between the two components. From the modified Lu and Weiss equation, the polymer–polymer interaction parameter (χ₁₂) of the PC‐PCL blends was calculated and found to range from −0.012 to −0.040 with the compositions. The χ₁₂ values calculated from the T_g method decreased with the increase of PC weight fraction. By taking PC‐PCL blend as a model system, the values of χ₁₂ were compared with two different methods, the T_g method and melting point depression method. The two methods are in reasonably good agreement for the χ₁₂ values of the PC‐PCL blends. © 2000 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 38: 2072–2076, 2000     

Synthesis,characterization, and ring opening polymerization of poly(1,4‐cyclohexylenedimethylene terephthalate) cyclic oligomers

Cyclic oligomers of poly(1,4‐cyclohexylenedimethylene terephthalate) (PCT) were prepared by reaction of 1,4‐cyclohexanedimethanol (CHDM) with terephthaloyl chloride under diluted conditions and separated from the linear products by silica gel column at a yield of 23.7 wt %. Cyclic dimer, trimer, tetramer, pentamer, and hexamer were further separated by high performance liquid chromatography, and found to constitute 98% of the cyclics mixtures. The structures of PCT cyclics were confirmed by means of mass spectrometry, Fourier transform infrared, and ¹H NMR analysis. A series of experiments were carried out to study the effects of catalysts and cis/trans configuration of isomers of CHDM on the yield of cyclic oligomers. Ring opening polymerization of the cyclic oligomers was carried out by heating the sample mixtures at 310 °C for 30 min in the presence of antimony oxide. Polymerization was confirmed by inherent viscosity changes and infrared spectra of the resulting polyesters. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 1828–1833, 2000     

Phase diagrams of poly(siloxane)/liquid crystal blends

Phase diagrams of blends of poly(methylphenylsiloxane) in short PMPS and two low molecular weight liquid crystals (4‐cyano‐4′‐n‐pentyl‐biphenyl and an eutectic mixture of paraphenylenes) are reported. Two polymers with very different weight‐average molar masses are considered in an evaluation of the loss of miscibility resulting from a known increase in the weight‐average molar mass. The experimental diagrams have been constructed via polarized optical microscopy and are rationalized in terms of the Flory–Huggins theory of isotropic mixtures and the Maier–Saupe theory of nematic order. The results show a good agreement between the theory and experiments and reveal a remarkable enhancement of miscibility with respect to similar systems involving poly(dimethylsiloxane). Variations of the interaction parameter with the temperature are compared for different systems of polysiloxanes. The effects of the nature of the liquid crystal and the polymer molar mass on the χ parameter are evaluated. © 2002 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 41: 39–43, 2003     

On what terms and why the thermodynamic properties of polymer solutions depend on chain length up to the melt

Theoretical considerations based on chain connectivity and conformational variability of polymers have led to an uncomplicated relation for the dependence of the Flory–Huggins interaction parameter (χ) on the volume fraction of the polymer (?) and on its number of segments (N). The validity of this expression was tested extensively with vapor‐pressure measurements and inverse gas chromatography (complemented by osmotic and light scattering data from the literature) for solutions of poly(dimethylsiloxane) in thermodynamically vastly different solvents such as n‐octane (n‐C₈), toluene (TL), and methylethylketone (MEK) over the entire range of composition for at least six different molecular masses of the polymer. The new approach is capable of modeling the measured χ (?, N), regardless of the thermodynamic quality of the solvent, in contrast to traditional expressions, which are often restricted to good solvents but fail for bad mixtures and vice versa. At constant polymer concentration, the χ values were lowest for n‐C₈ (best solvent) and highest for MEK (Θ solvent); the data for TL fell between them. The influences of N depended strongly on the thermodynamic quality of the solvent and were not restricted to dilute solutions. For good solvents, χ increased with rising N. The effect was most pronounced for n‐C₈, where the different curves for χ (?) fanned out considerably. The influences of N were less distinct for TL, and for MEK they vanished at the (endothermal) θ temperature. For worse than θ conditions, the χ values of the long chains were less than that of the short ones. This change in the sign of N agreed with this concept of conformational relaxation. © 2004 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 1601–1609, 2004     

Study of the thermal evolution of the cyclic‐oligomer formation in a cyclic‐oligomer‐free PET

A low percentage of cyclic oligomers can be found in poly(ethylene terephthalate) (PET) from its synthesis onward. In this article, a cyclic‐oligomer‐free PET (COFP) obtained by solvent extraction was used to study the thermodynamics of the re‐formation of cyclics from the melt. The cyclic‐oligomer content re‐increased into molten COFP, finally reaching an equilibrium. An analysis of the fraction of the re‐formed cyclic oligomers showed that a majority of cyclic trimer (60–70%) was found at the equilibrium. Before the establishment of the equilibrium, an unusual behavior was observed in the relative proportion of cyclic trimer and tetramer during the first steps of their formation that was probably due to a competition between kinetic and thermodynamic products. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 416–422, 2000     

Intramolecular screening effects in nondilute polymer solutions: An equation‐of‐state approach

In a previous article, we presented a simple modification of the traditional Flory–Huggins theory that took intramolecular screening effects (or same chain contacts) into account. In this article, we present a natural extension of that work, in which free‐volume effects are also explained with an equation‐of‐state model. The predictions of the interaction parameter, χ, for several polymer–solvent systems are presented, over the entire concentration range, in θ solvents and good solvents. A geometric mean assumption is applied to the calculation of an exchange energy interaction term. The predictions of χ are successful to various degrees when internal pressures are used, whereas the use of solubility parameters in most cases produces fairly good agreement with experimental results. © 2003 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 41: 2911–2922, 2003     

Permeation of a solvent mixture through an elastomeric membrane—The case of pervaporation

This article presents a model for the permeation of solvent mixtures through an elastomer in the particular case of pervaporation. An analytical expression for each solvent permeation rate is derived, in the limited case of a membrane that undergoes small swelling, without making any assumptions on the solvent diffusion coefficients and their dependence on solvent concentrations. Applying this analytical expression to different situations, we fitted most of the curves previously published on pervaporation experiments. In particular, we correlated the synergy developed by a mixture of two solvents in the permeation process with the sign of their Flory–Huggins interaction parameter χ_AB. This explains why, in most cases (χ_AB > 0), a molecule permeating easily through a membrane is mixed with a molecule permeating much less easily; the latter can see its permeation flux increase by a factor 10 or 100 because the swelling of the polymer induced by the more permeable molecule “opens the meshes of the network” allowing the less permeable molecule to pass through more easily. Within our analysis, the efficiency of the pervaporation process, expressed through the separation factor, is derived very simply as a function of the interaction coefficients and the viscosities of solvents and exhibits an exponential dependence on the volume fraction of either component as seen in most experiments. © 2002 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 41: 183–193, 2003     

Phase behavior of a water/2‐propanol/poly(methyl methacrylate) cosolvent system

A phase diagram of poly(methyl methacrylate) in mixtures of water and 2‐propanol, individually nonsolvents for the polymer, was studied at 25 °C. For this system, there were two liquid–liquid demixing regions separated by a miscible region. This cosolvent phenomenon was thought to be a joint effect of the nonsolvents. The phase behavior was modeled according to modified Flory–Huggins chemical‐potential equations, which accounted for the possible contribution from a ternary interaction in terms of a lumped parameter, χ₁₂₃. The calculated phase‐equilibrium curves (binodals) agreed well with the measured results. By contrast, if only binary interaction parameters were considered, computations yielded binodals whose compositions departed significantly from the measured data. Using the wet phase inversion method with casting dopes selected on the basis of the phase diagram, we prepared membranes with microporous structures in various coagulation baths. © 2000 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 38: 747–754, 2000     

Cyclic(arylene ether) oligomers containing the phenylphosphine oxide moiety

A series of cyclic(arylene ether) oligomers containing the phenylphosphine oxide moiety has been synthesized by reaction of bis(4-fluorophenyl)phenylphosphineoxide with dihydroxy compounds 1a–d as well as 1,2-dihydro-4-(4-hydroxyphenyl) (2H)phthalazin-1-one in DMF in the presence of anhydrous K₂CO₃ under high dilution conditions. These cyclic oligomers are amorphous and have high solubility in organic solvents. The MALDI-TOF-MS technique has been used as a powerful tool to analyze these cyclic systems. The cyclic(arylene ether) oligomers readily undergo anionic ring-opening polymerization in the melt at 350°C by using potassium 4,4′-biphenoxide as the initiator, affording linear, high molecular weight poly(arylene ether)s containing the phenylphosphine oxide moiety. © 1998 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 36: 519–526, 1998     

Dynamic scattering properties of mixtures of cyclic and linear homopolymers and copolymers

The dynamic scattering properties of mixtures of cyclic copolymers and homopolymers are discussed. These properties are compared with those characterizing linear chain mixtures in similar conditions. The differences between cyclic and linear chains are introduced through the form factors only. The interaction parameter between different monomer species are assumed to be the same whether they belong to cyclic or linear chains. The dynamical model is based upon a generalization of the random phase approximation neglecting hydrodynamic interaction and mode coupling effects. Despite these simplifications, substantial differences are found in the dynamics of mixtures containing cyclic copolymers and homopolymers when compared to those of linear chain systems in similar conditions of temperature, concentration, and molecular weight. © 1997 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 35 : 1191–1200, 1997     

Investigation of polymer–solvent interactions in poly(styrene sulfonate) thin films

The swelling behavior of acid form poly(styrene sulfonate) (PSS‐H) thin films were investigated using in situ spectroscopic ellipsometry (SE) to probe the polymer–solvent interactions of ion‐containing polymers under interfacial confinement. The interaction parameter (χ), related to the polymer and solvent solubility parameters in the Flory–Huggins theory, describes the polymer‐solvent compatibility. In situ SE was used to measure the degree of polymer swelling in various solvent vapor environments, to determine χ for the solvent‐PSS‐H system. The calculated solubility parameter of 40–44 MPa^1/2 for PSS‐H was determined through measured χ values in water, methanol, and formamide environments at a solvent vapor activity of 0.95. Flory–Huggins theory was applied to describe the thickness‐dependent swelling of PSS‐H and to quantify the water‐PSS‐H interactions. Confinement had a significant influence on polymer swelling at low water vapor activities expressed as an increased χ between the water and polymer with decreasing film thickness. As the volume fraction of water approached ～0.3, the measured χ value was ～0.65, indicating the water interacted with the polymer in a similar manner, regardless of thicknesses. © 2017 Wiley Periodicals, Inc. J. Polym. Sci., Part B: Polym. Phys. 2017 , 55, 1365–1372     

Effective interaction parameter between branched and linear polystyrene

Values of the effective interaction parameter (χ) between regular, long‐branched polystyrene chains and their linear analogues were measured with small‐angle neutron scattering for several star‐branched chains and one comb‐type polymer. The contribution to this interaction due to architecture alone increases monotonically with star functionality for the set of polymers studied here. The interaction appears to be less sensitive to variations in arm size than would be expected from fluctuation theory predictions by G. H. Fredrickson, A. Liu, and F. S. Bates (Macromolecules 1994, 27, 2503) for a purely entropic interaction due to architecture. The change in χ with the volume fraction of the star in the blend is in agreement with the theory, however. The magnitudes of the interaction in the star/linear blends are small enough that bulk phase separation is unlikely, whereas that in the comb/linear blend is about 20 times higher for the same number of arms. Thus, bulk phase separation can be readily expected for comb/linear blends at commercially relevant values of molecular weights. © 2001 John Wiley & Sons, Inc. J Polym Sci Part B: Polym Phys 39: 2549–2561, 2001     

Quantifying phase behavior in partially miscible polystyrene/poly(styrene‐co‐4‐bromostyrene) blends

The phase behavior of thin‐film blends of polystyrene (PS) and the random copolymer poly(styrene‐co‐4‐bromostyrene) (PBS) was studied with atomic force microscopy (AFM) and small‐angle X‐ray scattering (SAXS). Phase behavior was studied as a function of the PBS and PS degree of polymerization (N), degree of miscibility [controlled via the volume fraction of bromine in the copolymer (f)], and annealing conditions. The Flory–Huggins interaction parameter χ was measured directly from SAXS as a function of temperature and scaled with f as χ = f²χ_S–BrS [where χ_S–BrS represents the segmental interaction between PS and the homopolymer poly(4‐bromostyrene)] Simulations based on the Flory–Huggins theory and χ measured from SAXS were used to predict phase diagrams for all the systems studied. The PBS/PS system exhibited upper critical solution temperature behavior. The AFM studies showed that increasing f in PBS led to progressively different morphologies, from flat topography (i.e., one phase) to interconnected structures or islands. In the two‐phase region, the morphology was a strong function of N (due to changes in mobility). A comparison of the estimated PBS volume fractions from the AFM images with the PBS bulk volume fraction in the blend suggested the encapsulation of PBS in PS, supporting the work of previous researchers. Excellent agreement between the phase diagram predictions (based on χ measured by SAXS) and the AFM images was observed. These studies were also consistent with interdiffusion measurements of PBS/PS interfaces (with Rutherford backscattering spectroscopy), which indicated that the interdiffusion coefficient decreased with increasing χ in the one‐phase region and dropped to zero deep inside the two‐phase region. © 2001 John Wiley & Sons, Inc. J Polym Sci Part B: Polym Phys 40: 255–271, 2002     

Macrocyclic oligoimides. I. Preparation of macrocyclic oligoetherimides via Kricheldorf substitution polymerization

Macrocyclic oligoetherimides were synthesized by means of Kricheldorf's nucleophilic aromatic substitution polycondensation. A solution containing equimolar quantities of bis(trimethylsilyl ether) of bisphenol and arylenebis(fluorophthalimide) was continuously added into a high boiling solvent containing CsF catalyst under a pseudo-high dilution condition. The resulting reaction mixture contained a high yield of cyclic oligomers which could be isolated by solvent extraction. © 1997 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 35: 759–767, 1997     

Bis[4-monosubstituted-5(4H)oxazolinones] as coupling agents for block copolymer synthesis: Reaction with amine-terminated oligomers and with hydroxy-terminated/amine-terminated oligomer mixtures

Block copolymers were prepared by the bulk reaction of mixtures of amine-terminated aliphatic polyamides and polyethers with bis[4-monosubstituted 5(4H)-oxazolinones] as the coupling agents. The simultaneous chain-coupling reaction of amine-terminated polyethers and hydroxy-terminated polyesters with these coupling agents also was investigated. As indicated by NMR and size exclusion chromatography studies, block copolymers of M_n ≥ 10,000 were obtained without side reactions in a much shorter reaction time than required when the conventional reaction was used between oligomers bearing mutually reactive end-groups. The block copolymers behaved as thermoplastic elastomers. Their thermal and thermomechanical properties were evaluated by differential scanning calorimetry and dynamic mechanical thermal analysis and compared to the block copolymers synthesized in the conventional way. When low molar mass oligomers were used (M_n ≤ 1000), copolymers of low crystallinity or amorphous copolymers were obtained. This was assigned to the disruption of chain regularity induced by the presence of the chain-coupling-agent moieties. However, properties comparable to those of copolymers obtained by using the conventional method were obtained by reacting oligomers of M_n ≥ 2000. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 4412–4421, 1999     

Melting and crystallization behavior of poly(vinylidene fluoride) produced isothermally in the intercrystalline spaces of poly(ethylene terephthalate) from its blends

The melting of isothermally crystallized poly(vinylidene fluoride) (PVF₂), produced in the intercrystalline spaces of poly(ethylene terephthalate) (PET) from its blends, showed a unique behavior: the melting temperature decreased with the increasing crystallinity of PVF₂ (i.e., with increasing crystallization time) for PVF₂ volume fractions of 0.64 and 0.51. The melting temperature of already crystallized PET also decreased as the PVF₂ crystallization progressed and the isothermal crystallization temperature of PVF₂ increased. Separate reasons were proposed to account for these behaviors. The equilibrium melting temperatures of PVF₂ in the blends, measured by the Hoffman–Weeks extrapolation procedure, were used to calculate the polymer–polymer interaction parameter (χ₂₁); only the noncrystallized portion of PET contributing to the mixed amorphous phase was considered. The χ₂₁value (−1.75) was lower than χ₁₂ (−0.14), calculated from the melting temperature depression of PET. However, when they were normalized to the unit volumes of the respective components, the two values were found to be the same. The crystallization rate of PVF₂ decreased with an increasing volume fraction of PET in the blend. The Avrami exponent increased for the volume fraction of PVF₂ (0.77) and then progressively decreased with an increasing volume fraction of PET. A gradual change in the nature of the regime transition from regime II/regime I to regime III/regime II with increasing PET concentration was observed. The value of the chain-extension factor of PVF₂ significantly increased with an increase in the PET concentration in the blends. © 2004 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 2215–2227, 2004     

Isolation,spectra, structure,and ring-opening polymerization of strained macrocyclic aryl(ether ketone) dimer

Macrocyclic arylene ether ketone dimer was isolated from a mixture of cyclic oligomers obtained by the nucleophilic substitution reaction of bisphenol A and 4,4′-difluorobenzophenone and easily polymerized to high molecular weight linear poly-(ether ketone). The cyclic compound was characterized by FTIR, ¹H- and ¹³C-NMR, and single-crystal x-ray diffraction. Analysis of the spectral and crystal structure reveals extreme distortions of the phenyl rings attached to the isopropylidene center and of the turning points of the molecular polygons. The release of the ring strain on ring-opening combined with entropical difference between the linear polymer chain and the more rigid macrocycle at temperatures of polymerization may be the proposed motivating factors in the polymerization of this precursor to high molecular weight poly(ether ketone). © 1997 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 35 : 1753–1761, 1997     

Cyclic and telechelic ladder polymers derived from tetrahydroxytetramethylspirobisindane and 1,4‐dicyanotetrafluorobenzene

5,5′,6,6′‐Tetrahydroxy‐3,3,3′,3′‐tetramethylspirobisindane was polycondensed with 1,4‐dicyanotetrafluorobenzene in four different solvents at 70 °C. In dimethylformamide, N‐methylpyrrolidone, and sulfolane exclusively, cyclic polymers were detectable by matrix‐assisted laser desorption/ionization time‐of‐flight (MALDI‐TOF) mass spectrometry up to masses around 13,000 Da. In dimethyl sulfoxide, linear byproducts were also found. Higher temperatures caused degradation reactions catalyzed by potassium carbonate. Polycondensations performed with the addition of 4‐tert‐butyl catechol or 2,2′‐dihydroxy binaphthyl yielded linear telechelic oligomers. Equimolar mixtures of linear and cyclic ladder polymers were examined by MALDI‐TOF mass spectra to determine how the end groups and the cyclic structure influenced the signal‐to‐noise ratio. The results suggested a preferential detection of the linear chains. © 2006 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 44: 5344–5352, 2006