Polymérisation de diènes conjugués aux hautes pressions. IX. Polymerisation de l'isoprène en presence de HI

Basic assumptions and theoretical approaches concerning the determination of long chain branching by experimental determination of intrinsic viscosities [η], molecular weights M_w and radii of gyration r of statistical coils of polymers with low heterogeneity of molecular weight (U_M ≈ 0,5) are examined. Neither of the two theories which are often applied makes use of the correct relation between the ratios [η]_v/[η]₁ and of branched and linear molecules of a given molecular weight in good solvents. Both theories neglect the fact that the FOX -FLORY -constant Φ of molecules with strong random branching reaches a limiting value and that certain branching mechanisms give abnormal coil shapes and hence abnormal FOX-FLORY-constants Φ. This is verified experimentally by measurements on swollen spherical gel particles and fractions of star shaped as well as statistically branched polymers and of low-density polyethylenes. Both theories furthermore use restrictive assumptions about the molecular weight distribution and partly unsuitable averages for . Model calculations show that for molecularly inhomogeneous polymers approximately the weight average n_w, of the number of branching points per molecule is obtained, if the measurements are evaluated by means of the functions derived for unimolecular polymers. A new method for the evaluation of long chain branching is given, which takes into account the different draining of linear and branched molecules.     

Incréments d'indice de réfraction de polymères en solution: Étude des lois de mélange. Influence de la masse moléculaire et de la structure

The variation of the refractive index increment (dn/dc) with the molecular weight and the structure of series of linear and branched well-defined polystyrenes dissolved in benzene was studied. Testing the Lorenz-Lorentz and Onsager-Böttcher mixture rules, we were able to show that:

• 1 Below a critical molecular weight of about 2·10⁴, the linear correlation dn/dc=f(1/M_n) is well explained by the influence of chemical heterogeneities included in the polymer chain (end-groups, branching points etc.).
• 2 These rules do not account for specific polymer-solvent interactions and are not quite rigorous.
• 3 The variations of (dn/dc) and the partial specific volume of the polymer, v?_p, observed already for molecular weights higher than 2·10⁴ are accompagnied by a change in the partial specific refractivity. These effects are related to the influence of intramolecular segment density in the interior of the linear or branched coil.

A homologous series of polyoxyethylene-glycols (α-hydro-ω-hydroxypoly(oxyethylene)s) in benzene presents the same behaviour.     

Étude viscosimétrique de solutions de polyélectrolytes par dilution isoionique. Effet de la densité de charge sur la conformation du polyion

The reduced viscosity η_sp/c of dilute solutions of carboxymethylcelluloses (CMC) was studied, keeping the total ionic strength of the solution constant; the variation of η_sp/c with concentration is often linear and the Huggins constant k′ increases with decreasing ionic strength. So it is possible to obtain the intrinsic viscosity [η]_{I = 0} at zero ionic strength I and to calculate the corresponding molecular dimensions. The variation of [η]_{I = 0} with the degree of substitution of CMC is explained by a progressive change from a zig-zag conformation to a rod when the linear charge of the polyion increases.     

A general method for the determination of “monomolecular property versus molar mass relationships” from measurements on polymolecular polymers

“Monomolecular property (p) versus molar mass (M) relationships” of the type p = K_p · M^ap as would result from an experimental determination of p and M of a polymer-homologous series of strictly monomolecular polymers can also be obtained from a polymer-homologous series of polymolecular samples or fractions, provided their correct corresponding averages p and M are determined (Principle of Corresponding Averages). In order to obtain the monomolecular intrinsic viscosity [η] vs. M relationship (Mark-Houwink relationship), the corresponding averages weight-average intrinsic viscosity ([η]_w ≡ [η]) and viscosity-average molar mass (M_[η]) of a series of polymer-homologous, polymolecular polymer samples or fractions have to be determined experimentally. Accordingly, the corresponding monomolecular relationships sedimentation coefficient s vs. M, diffusion coefficient D vs. M, and mean-square radius of gyration 〈r²〉 vs. M are obtained by the determination of a series of couples of the corresponding averages s vs. M , D vs. M , and 〈r² 〉 vs. M , respectively. The experimental determination of the corresponding averages presented is possible by different methods, including the determination of the molar mass distribution.     

Polymerization of lactams, 54. Anionic polymerization of 2-pyrrolidone accelerated with CO2, part 3

Intrinsic viscosities in m-cresol and weight average molecular weights, M?_w, were measured for samples of high molecular weight poly(2-pyrrolidone) (poly ( 1 )) prepared by anionic polymerization of 2-pyrrolidone ( 1 ) accelerated with CO₂. It was proved that the earlier found relationship [η] = 4 · 10^?2 · M^0,77 (in cm³ · g^?1) holds for M?M_w up to 8 · 10⁵ g. mol^?1. The probable reason for the formation of poly ( 1 ) with an exceptionally high molecular weight is discussed.     

Volumes spécifiques de polystyrènes en étoile à l'état liquide

Specific volume measurements carried out on linear and star-shaped liquid polystyrene versus the number average molecular weight, M_n scaling between 5000 and 2.10⁶, are in good agreement with a review of the experimental results of others. These results confirm the occurrence of a loss of volume per monomeric unit above a molecular weight of 10000 already observed for polystyrene in solution. We attribute the influence of this specific volume variation against M_n on the thermal dilatometric coefficient α = (1/v)dv/dT and the temperature of vitrous transition. We also attempted to give an explanation for this phenomenon.     

Préparation de polystyrène de masse éalevée à chaîcne aliphatique ou noyaux aromatiques deuteriés

Isotopically substituted styrene, (phenyl-²H₅) styrene ( 3 ), and selectively labled styrene, [alpha;, beta;-²H₂] styrene ( 6 ), were prepared by the classical two step synthesis, starting with the correspondingly labeled compounds. By anionic polymerization in an inert atmosphere labeled high molecular weight polystyrences 10 and 11 were prepared.     

Polymérisation anionique de l'acroléine, 2. Influence de la nature du contre-ion et de la température sur la cinétique de la réaction

In tetrahydrofuran, with Na⁺ and Li⁺ as counter-ions, the kinetic order of the anionic polymerization of acrolein is unity for monomer and for initiator. These results indicate that the living ends are not associated at the studied concentrations of initiator. The polymerization rate depends on the nature of the counter-ion. Transfer reactions to monomer do not affect the polymerization rate but greatly change the molecular weights M _n of polyacroleins. The experimental M are much lower than the theoretical M . The propagation constants k_pr and the transfer constant h_m are determined using these experimental values of M . From these results, we can conclude that with Li⁺ as counter ion, transfer reactions are much more numerous than with Na⁺. Furthermore, the polymerization rate increases with temperature. The activation energies of the propagation (E_a,pr) and transfer reactions (E_a,t) can be determined separately. When the temperature increases the propagation reaction is promoted in comparison to the transfer reaction to monomer, and simultaneously transfer reactions to polymer take place. This last phenomenon restricts the upper value of the polymerization temperature.     

Untersuchungen zum oxidativen Abbau nativer Cellulose, 2. Einfluß der Alkalibehandlung auf die Viskositäts-Molekulargewichts-Beziehungen

For kinetic studies of cellulose degradation by molecular oxygen it is necessary to establish new [η]-M-relationships, since the treatment of cellulose by concentrated alkali solutions changes the intrinsic viscosity [η] without changing the molecular weight M. Two series of degradated celluloses treated for different times with sodium hydroxide solutions were prepared, and the constants of the Kuhn-Mark-Houwink equation were determined for cellulose nitrates in acetone and for celluloses in Cuoxam (Cu(NH₃)₄(OH)₂) by comparison of intrinsic viscosity with light scattering data.     

Bestimmung der langkettenverzweigung von hochdruckpolyäthylen durch streulicht- und viskositätsmessungen

Linear and branched polyethylene were fractionated by solvent gradient chromatography. The fractions were characterized by light scattering and viscosity measurements. The relationship between the molecular weight and the limiting viscosity number and that between the molecular weight and the radius of gyration in α-chloronaphthaline at 127°C were determined. The exponent of the equation between the ratio 〈g_η〉Θ of the limiting viscosity numbers of branched and linear fractions of the same molecular weight M_w and the corresponding ratio 〈g_η〉_Θ,w of the radii of gyration for Θ-conditions was By this equation, the long chain branching of high pressure polyethylene can be determined from 〈g_η〉, the ratio of the limiting viscosity numbers in thermodynamically good solvents, which is within the experimental limits of error identical to 〈g_η〉_Θ in a Θ-solvent. Almost independent of the molecular weight, for the branched polyethylene fractions 1,4 long chain branching points per 1000 C-atoms were obtained.     

Characterization of DEAE-dextran by means of light scattering and combined size-exclusion chromatography/low-angle laser light scattering/viscometry

Ten DEAE (2-(diethylamino)ethyl) dextran samples were investigated by means of static and dynamic light scattering, viscometry and size-exclusion chromatography (SEC) in combination with on-line small-angle laser light scattering (LALLS) and viscometry (VISC). In dilute solution the behavior of DEAE-dextran was compared with that of unsubstituted dextran and the molecular weight M dependences of the radius of gyration R_g, hydrodynamic radius R_h, intrinsic viscosity [η], second virial coefficient A₂ and z-average diffusion coefficient D _z were determined. The relationships for DEAE-dextran dissolved in a 0,8 molar sodium nitrate solution were nearly the same as for dextran dissolved in water with 0,05 wt.-% sodium azide and gave the same exponents. The molecular weight dependence of the intrinsic viscosity cannot be described by a Kuhn-Mark-Houwink relationship with a constant exponent. The slope in the plot of log [η] versus log M decreases with increasing molecular weight which indicates the occurrence of branching. By means of SEC/LALLS/VISC measurements the molecular weight distributions were determined. The distributions were calculated (1) directly from the light scattering signal, (2) from a calibration line obtained by light scattering data of a DEAE-dextran sample with a broad distribution and (3) from the intrinsic viscosity distribution obtained by the on-line viscosity/refractive index detector in combination with the [η]-M relationship. In order to get the correct molecular-weight dependence of the intrinsic viscosity it is necessary to determine the molecular weight distribution directly by LALLS (technique 1) and to combine this with the appropriate intrinsic viscosity data from the viscometer. Only the third technique, which is an extension of technique 1, gave satisfactory results over the whole molecular weight region observed.     

Micellisation de copolymères séquencés polystyrène-polyvinylpyridinium, 3. Influence de la concentration en sel et de la température

The effects of the salt concentration and temperature on the micellization phenomenon of polystyrene-poly(4-vinyl-N-ethylpyridinium bromide) block copolymers in dilute solution are investigated in solvents selective for the polyelectrolytic part (water-methanol-LiBr mixtures). The polystyrene moiety being in non-solvent media remains rather collapsed whatever temperature and salt concentration. On the contrary, the latter parameters modify strongly the conformation of the soluble polyelectrolytic chains and therefore determine the “quality” of the solvent essentially towards the polyvinylpyridinium blocks. Consequently an increase of the temperature or a decrease of the salt concentration have the same effect on the behaviour of the copolymers. In the case of a copolymer with a small molecular weight polystyrene block (M_w = 2 700), the unimer-multimer (single molecules/micelles) equilibrium is easily shifted in favour of the single molecules by increasing the temperature or decreasing the ionic strength. On the contrary, if the polystyrene moiety is longer (M_w = 6 000 – 13 000) the micellar weight of the system is little or even not at all influenced by the ionic strength and the temperature. These results emphasize that, for the investigated systems, it is the “insoluble” blocks which essentially govern the micellization phenomenon.     

Studies of lyotropic liquid crystalline mesophases of some novel regioselective substituted cellulose derivatives, 1. Synthesis and characterization

Novel regioselective substituted cellulose derivatives, suitable for investigations of their liquid crystalline phase behaviour, were synthesized. The derivatives, tricarbanilates with varying substituents at the phenolic residue, were characterized according to their molecular weight and molecular weight distribution by means of size exclusion chromatography coupled with low angle laser light scattering (SEC/LALLS). A simple new method is presented, which allows the calculation of the constants K and a of the Staudinger-Mark-Houwink (SMH) equation, the viscosity-average molecular weight M _η as well as the intrinsic viscosity [η] from the light scattering calibration curve of an SEC experiment. The calculated values of the intrinsic viscosity are in good agreement with those derived from measurements with a low-shear viscometer. A plot [η] vs. weight-average molecular weight M _w shows, that a substitution at the C-2 and C-3 position of the glucosic unit with 3-chlorocarbanilate leads to a well pronounced enlargement of the polymer dimensions which is mirrored by larger values of [η] at the same M _w as compared with other cellulose derivatives.     

Viscosity behaviour of macro-ring polystyrene in solution

A simple treatment for the intrinsic viscosity data of a flexible macro-ring polymer in solution is proposed. The viscosity behaviour of dilute solutions of a macro-ring with molar mass M is considered to be equal to that of a linear chain with molar mass λ_T·M, where the factor λ_T is found to be close to 0,5. Thus the ratio of the intrinsic viscosity of the macro-ring polymer and that of the corresponding (same molar mass) linear chain polymer g_r ≡ [η]_r/[η]_l = λ with α = exponent of Mark-Houwink equation gives the correct excluded volume effect on the value of g_r as verified by the dilute solution viscosity behaviour of a macro-ring polystyrene sample of weight-average molecular weight M_w = 2,72·10⁵ and theta- and thermodynamically good solvents.     

Zur alterung von wäßrigen polyacrylamid-lösungen

For 62 days the aging behaviour of diluted solutions of an extremely pure, linear, noncharged polyacrylamide with M_w = 8400 000 g/mol in water was examined. The solutions were stored in a sterile surrounding at 298 K. The following quantities, all of which give information about the molecular weight or the dimension of the macromolecules, were measured: the intrinsic viscosity [η], the sedimentation cofficient s₀, the translational diffusion coefficient (D_t)_z, the molecular weight, M_w, and the radius of gyration, R_z. Within the accuracy of the measurements all these quantities remained constant. That means, in contrast to other authors, that we could not observe any aging effect in our solutions. It seems that polyacrylamide dissolved in water shows a behavior like a coil molecule in a thermodynamically good solution.     

Etude par diffusion de la lumière du système ternaire solvant/polydiméthylsiloxane/poly(méthacrylate de méthyle)

The ternary system solvent (1)/polydimethylsiloxane (2)/poly(methyl methacrylate) (3) was investigated by light scattering. First the measurements were realized on mixtures containing low-molecular-weight polydimethylsiloxane: M₂ ? M₃. In a second step, high and comparable molecular weights of both polymers were chosen for the measurements. The thermodynamic interaction parameter χ₂₃ was determined from light scattering data (spinodal) and demixtion curves as a function of concentration and molecular weight of the polymers. Also the variation of the angular repartition of the scattered light was studied and the results were compared with calculated values.     

Etude comparée des propriétés mécaniques dynamiques des formes I et II du polybutène-1. Cinetique de transformation II → I

The behaviour of the molecular motion in synthetic cis- and trans-1,4-polyisoprenes was studied by nuclear magnetic resonance. The line-widths (ΔH_pp), second (M₂) and fourth moments (M₄) of the absorption curves were measured at 77, 4.2, and 1.3 K. The obtained values of ΔH_pp and M₂ for cis-polyisoprene indicate that the molecular motions are apparently not completely quenched at 77 K since there is a further increase in the values of M₂ and ΔH_pp when decreasing the temperature from 77 to 4.2 K. It was found also that the value of M₂ at 4.2 and 1.3 K (? 20 G²) was larger than the previously computed one for a rigid lattice (18.5 G²). The temperature dependence of M₂ and ΔH_pp for trans-polyisoprene is different from that for cis-polyisoprene. The values of M₂ are constant in the range from 77 to 1.3 K (? 19.5 G²), which suggests that the molecular motion is presumably stopped in this polymer already at 77 K.     

Molecular characterization of water-soluble,cationic polyelectrolytes

Poly{acrylamide-co-[2-(acryloyloxy)ethyltrimethylammonium chloride]} (1) , Poly{acrylamide-co-[2-(methacryloyloxy)ethyltrimethylammonium chloride]} (2) and Poly{acrylamide-co-[3-(acryloylamino)propyltrimethylammonium chloride]} (3) with cationic monomeric unit mole fractions of 8, 25, and 50 mol-%, and the corresponding homopolymers were characterized with respect to their molecular dimensions in 1 M sodium chloride solution at 25°C in the molecular weight range from 200000 to 16000000, using light-scattering and viscometric measurements. It was possible to establish correlations between molecular weight, intrinsic viscosity and radius of gyration. Whereas a steady increase in the exponents of the [η] vs. M- and 〈R²_G〉^0,5 vs. M-relationships can be observed with increasing charge density, there is an initial decrease in the values of the exponents for 2 and 3 , when 8 mol-% of the cationic monomer is inserted into poly-acrylamide. If the charge density is increased further, these values can also be seen to rise steadily. The findings are explained by the different structures in solution, which are formed by the copolymers and homopolymers investigated due to their chemical structure. The radii of gyration were calculated from the intrinsic viscosities with the aid of the Flory-Fox equation. However, neither the results obtained with the Flory-Fox constant Φ₀= 3,69. 10²⁴ mol^?1, nor those with the value of Φ₀= 2,10. 10²⁴ mol^?1 showed good agreement with the experimentally recorded data.     

Transition Metal‐Catalyzed Ring‐Opening Polymerization of Silicon‐Bridged [1]Ferrocenophanes in the Presence of Functional Silanes: Molecular Weight Control and Synthesis of Telechelic Poly(ferrocenylsilanes)

The transition metal‐catalyzed ring‐opening polymerization of dimethyl[1]silaferrocenophane (fcSiMe₂) ( 1 ), fc = Fe(η‐C₅H₄)₂), in the presence of MePhSiH₂, or the chlorosilanes ClMe₂SiH, ClMePhSiH, or ClPh₂SiH has been shown to allow access to poly(ferrocenylsilane)s R¹R²R³Si? [fcSiMe₂]_n? H ( 4 , 6 – 8 ), and R¹? [Me₂Sifc]_m? SiR²R³? [fcSiMe₂]_n? H ( 5 ) with controlled molecular weights and which are capped by the corresponding R¹R²R³Si and Si? H groups ( 4 , 5 : R¹ = H, R² = Me, R³ = Ph, 6 : R¹ = Cl, R² = R³ = Me, 7 : R¹ = Cl, R² = Me, R³ = Ph, 8 : R¹ = Cl, R² = R³ = Ph). Materials with molecular weights in the range M _n of 3.5 × 10³ to 2.5 × 10⁴ and polydispersities of 1.3–2.2 were prepared. All of the silanes examined in this study were found to be more reactive as capping agents than the previously studied Et₃SiH; the order of reactivity for molecular weight control was determined to be MePhSiH₂ ≈ ClMe₂SiH ≈ ClMePhSiH ≈ ClPh₂SiH > Et₃SiH. In addition, the reactivity of the resulting Si? Cl end‐functionalities of poly(ferrocenylsilane)s 6 and 7 was explored, and reactions with commercial poly(ethylene glycol) methyl ether yielded poly(ethylene oxide)–block–poly(ferrocenylsilane) diblock copolymers, 10 and 11 . The Si? H end group, however, was much less reactive and attempts to utilize this functionality for hydrosilylation of divinyl‐terminated poly(dimethylsiloxane) to form block copolymers was ineffective.

     

Polymérisation de l'ε-caprolactone par des catalyseurs dérivés du bore

The cationic polymerization of ε-caprolactone catalyzed by boron derivatives, namely acyl fluoroborates, triethyloxonium fluoroborate (Meerwein's salt), boron trifluoride etherate and boron trifluoride was investigated. The acyl fluoroborates, especially the acetyl fluoroborate, produced high molecular weight polymers with fairly poor yields, whereas triethyloxonium fluoroborate gave a high yield of a rather low molecular weight polymer. The results were compared with those obtained with the two non-ionic catalysts. It seems that the acylium or ethylium group of the catalyst initiates the molecular chain and that the fluoroborate anion serves as catalyst and keeps the activity of the reaction site at the end of the growing chain. The mechanism is that of a polymerization without termination reaction (living-polymer type). However, according to the results of the kinetic study, this is true only in the case of acyl fluoroborates. The following relation between the viscosity and the number-average molecular weight of our samples was obtained: [η] = 1,25·10^?4M?_n^0,82.