ALBERT

Notes: In the rolling of Plastic sheet the viscous work gives rise to heating of the plastic sheet. The maximum temperature reached in the interior of the sheet during shearing is of the interest, both in the theoretical calculation of roll pressure, and from the point of view of possible thermal decomposition effects. A calculation is given for a simple, some-what artificial, model, which leads to an upper value for this maximum temperature if a measurement of the average temperature of the emergent sheet is given. The model taken is that of an infinite sheet, of thickness 2δ, in which heat is continuously liberated at the rate of 4ηφ2 cal. cc. -1 sec. -1, where φ is the relevant value of the rate of compression (assumed constant across the sheet). The curve giving the temperature distribution across the sheet is calculated numerically for two cases; it is relatively flat, as would be expected on physical grounds, since the viscosity, η, decreases as the temperature rises. It is concluded that these thermal effects may be neglected in the calculation of roll pressure in Part 1.

Notes: This and a previous article (p. 127) review the literature from 1910-1945 on polymerization of olefins and diolefins in suspension and emulsion, and present a number of new measurements not published to date. The subject is considered mainly from the point of view of scientific information on the mechanism of polymerization in aqueous suspensions and emulsions, but brief mention is also made of the more important disclosures in the patent literature. The new data presented in this article refer to (1) initial rates of polymerization of styrene, methyl methacrylate, vinyl acelate and acrylonitrile in aqueous suspensions and in soap emulsions as a function of catalyst concentration, temperature, and soap concentration; (2) influence of water-soluble activators, such as sodium bisulfite; (3) influence of initial size of monomer droplets on initial rates of monomer consumption; (4) study, with the aid of the electron microscope, of size of monomer droplets and polymer particles throughout polymerization; and (5) a few experiments on side reactions in the domain of higher conversions. No attempt is made in this paper to review and appraise the very large number of recent patents (from about 1930 on), which protect special procedures on the use of various promoting, regulating, or modifying ingredients. A complete digest of this practice does not exist at present, but reference may be made to the excellent chapter on emulsion polymerization in the book of Talalay and Magat, to the very valuable compilation of patents by Hoseh, and to the enumeration of a selected number of patents in the book of Scheiber.1

Notes: Equations are derived for stress-strain curves for a hypothetical model substance containing a large number of like systems, each assumed to be in equilibrium between two states having different arrangements of the atoms. With this model one can study the dependence of the initial elastic modulus, the limiting strain (for infinite stress), and the over-all shape of the stress-strain curve on characteristics of the rearranging systems - their number, the initial energy difference between the two states, the shift of atomic positions, etc. Introduction of an assumed proportionality between the shift per rearrangement and the square of the relative length of the sample (in the direction of the shift) leads to stress-strain curves similar to those determined experimentally for natural rubber and other rubberlike materials.

Notes: A review of theoretical progress in copolymerization has shown that the results of previous authors are consistent and equivalent. A general scheme of copolymerization has been stated and its kinetic consequences worked out. The rate expressions have been derived for percentage polymerization against time, and initial rate against initial monomer concentration. Some experimental considerations have been given. The advantages of photochemical technique have been stressed, physical means of following the reaction outlined, and new methods of determining σ and μ described. Means of verifying the kinetics have been suggested, in particular the obtaining of the individual velocity constants. It is important to note in this connection, that further experiments to determine the lifetime of the growing polymer chain are not required.

Notes: Tetrachlorophthalic anhydride (TCPA) was reacted with cellulose diacetate and with cellulose diethyl ether in dioxane solution to produce esters with rather low tetrachlorophthalyl contents. Phthalic anhydride was found to be considerably more reactive than was TCPA with these cellulose derivatives, especially if pyridine was used as the reaction solvent. Acylations could not be executed with TCPA in the presence of pyridine as solvent, however, because this material decomposed the anhydride, removing chlorine from it. The TCPA derivatives were more limited in solubility than were the corresponding esters made from phthalic anhydride.

Notes: An attempt has been made in this communication to formulate explicitly the dependence of the frictional coefficients of threadlike molecules in solution upon the centrifugal field applied to them in the usual velocity ultracentrifuges. This quantitative treatment, using the Kuhn and Kuhn odel for randomly coiled free-draining thread molecules, verifies the assumption generally made that the effect considered is negligible. Since the molecular weight of the particles and the magnitude of the centrifugal field applied to them vary inversely with each other, the constant, \documentclass{article}\pagestyle{empty}\begin{document}$\overline {\beta h_0^2 }$\end{document} rarely attains values which are significant, and the effect produced by orientation of the threads is balanced by the effect due to stretching them. Moreover, by a suitable change of parameters, this treatment could be extended to account for the action of other tensile forces acting on thread molecules in solution

Notes: The variation of the x-ray reflection to the interchain separation in sodium pectate has been determined as a function of moisture content. The crystalline regions in sodium pectate have been found to sorb about 24% water (dry basis) at a relative humidity of 95% at 25°C. The sample as a whole under these conditions has a water content of 64%. At high water contents, therefore, a large difference in water content must exist between the crystalline and amorphous regions of sodium pectate. A marked change in degree of crystallinity occurs, as determined by the number and sharpness of the x-ray reflections, at an average water content of about 6% on desorption and on sorption at an average water content of about 14%.

Notes: A description is given of the preparation, birefringence, and swelling capacity of a series of microscopically homogeneous, partially acetylated cellulose model filaments with an acetyl content varying from 0 to 2.3 acetyl groups per glucose residue, and obtained either by acetylation or by deacetylation. A higher degree of acetylation of these artificial fibers could not be attained without destruction of the fiber. The same holds true for native ramie fibers. The swelling was studied in water, acetone, and methanol. The degree of swelling of the filaments plotted against acetyl content yields entirely different curves for the acetylation and the deacetylation series. These curves can be readily explained by the hypothesis that the intercrystalline (so-called amorphous) portion of the fiber substance is more accessible to the esterification and saponification reactions and is affected earlier than the crystalline portion. The birefringence of the acetylation series shows but a small - although apparently real - difference from that of the saponification series.

Notes: The tensile properties of a range of pure gum natural rubbers has been reviewed, and it has been shown that their principal features can be understood on the assumption that the tensile strength measured in a given test depends directly on the amount of crystallization at break. The most important single factor in determining tensile strength is the degree of cross linking. Cross linking is only needed in order to prevent plastic flow, thus permitting the molecules to be aligned by stretching, and hence to crystallize. A very highly cross-linked rubber is weak because the load required to stretch it is so high, that the rubber is broken before the elongation becomes large enough to produce crystallization. In general, vulcanization also involves reactions, e.g., the combination of sulfur with the rubber, which inhibit crystallization by producing structural modifications of the rubber. These reduce the tensile strength, especially when the degree of cross linking is large. These ideas readily explain the effects of swelling and of the temperature of test. They are also used in a brief discussion of the phenomena of overcure, reversion, and aging.

Notes: The emulsion polymerizations of butadiene and styrene and of butadiene alone have been studied at temperature of 90-160°. The effects of changes in catalysts, modifiers, times of polymerization, and other variables have been investigated. The rate of polymerization at 110° is markedly increased over that at 50° and the final product is a less modified and less processable polymer. However, the standard evaluation and road tests indicate this polymer may be a useful one.

Notes: From the results of initial acid-catalyzed hydrolysis of cellulose, and from certain data in the literature it is concluded that (1) native cellulose possesses an indefinitely high molecular weight; (2) as a consequence of common procedure adhered to by most investigators, native cellulose is unintentionally degraded to about DP 3000; (3) mild, initial acidic degradation, intentional or accidental, causes further decrease in the DP values down to the probable limiting value of 128 or 256, with complete loss of the tensile strength of the cellulose fiber; and (4) the few covalent bonds (about 0.3%) which become cleaved during these mild reactions are equally spaced, acid-sensitive, and entirely different from the regular 1,4-glycosidic bonds which are responsible for the random hydrolysis of cellulose in strongly acidic media. It is suggested that these acid-sensitive linkages represent hemiacetal or acetal bonds which originate from openchain glucose anhydride residues of “limit hydrocellulose” units and connect the latter along the a and b axes of the fiber. The length of the “limit hydrocellulose” is supposed to correspond to that of 27 = 128 glucose anhydride residues - that is, 660 Å., or twice these figures. Extension of this concept to starch leads to the postulate of a three-dimensional network of glucose anhydride units of indefinite number. The latter are held together by regular 1,4-α-glucopyranosidic bonds and, at short intervals, probably by acetal bonds of open-chain glucose anhydride residues. On mild, acid-catalyzed hy-drolysis these acid-sensitive linkages rupture rapidly and completely, giving rise to a material which consists of linear molecules only, containing 1,4-, 1,6-, and possibly 1,3-, glycosidic bonds between two adjacent glucose anhydride residues.

Notes: A substance may have a precipitating action on other dissolved substances. The concentration at which this action becomes apparent depends not only on the chemical nature of the constituents, but also on their molecular weights. In a homologous series the precipitating power increases with increasing molecular weight until it attains a very high value, as in macromolecules. The incompatibility of high polymers, which is manifested at very great dilutions, is due to this increase in the precipitating power with an increase of the molecular weight.

Notes: The orientation of model filaments stretched to various extents in three different states of swelling was investigated by quantitative x-ray analysis before and after drying. The change in orientation upon drying was found to be either very small or zero. This shows that the crystallites do not even approximately follow the rules of affined transformation which would require a very considerable increase in orientation on drying. There must be an entirely different mechanism which prevents the crystallites from being dragged along with the matrix in which they are embedded.

Notes: Part I traces fine structural features of polythene, beginning with the arrangement of carbon atoms disclosed by x-ray diffraction, and including the incorporation of crystalline segments into crystallites, and finally the combination of crystallites and amorphous regions to form spherulites-the largest structural elements of polythene. Attention is called to the frequent occurrence of side chains in the molecules. The structures of spherulitic and cold-drawn polythenes are compared. Part II describes an experimental study of the cold-drawing of polythene which discloses an interesting variant of this process, designated as “redrawing,” in which dimensional changes differ greatly from those of ordinary cold-drawing. On the basis of these observations and structural considerations presented in Part I, a mechanism of cold-drawing as a rheological process is outlined.

Notes: Experimental methods for measuring the mechanical properties of concentrated polymer solutions and other viscoelastic systems under periodically varying stresses are reviewed. These include direct measurements of stress and strain, measurements of response at resonance frequencies determined by the inertia of the apparatus, propagation of waves in extended media, and measurements of resonance vibrations. The conditions for the suitability of each method are discussed, and equations for the interrelation of the various quantities measured are given. Examples are shown to illustrate the fitting of experimental data to the calculated behavior of mechanical models, in order to obtain frequency-independent mechanical constants which can be interpreted in terms of molecular behavior.

Notes: The same point in the space cannot be occupied simultaneously by two molecules. In a solution of rigid rodlike macromolecules, each molecule prevents its neighbors from occupying certain positions, the whole of which form the forbidden angle. Calculation shows that in the case of rigid, filiform molecules, this angle is large even at very low concentrations. Isotropy is not possible at moderate concentrations. The consideration of the forbidden angle is necessary in the study of viscosity, osmotic pressure, flow birefringence, sedimentation, and molecular light scattering. In the solid state, the substance can be put under great tension, and the return to the normal state is extremely slow. This tension accounts for the abnormal solubility of disordered high polymers.

Notes: During recent years attempts have been made to formulate a viscosity-concentration relationship. The method of Huggins, in introducing a factor k′ as a hydrodynamic correction factor is examined, and in particular the attempts to draw conclusions relating k′ to the thermodynamic properties of the solution. An explanation is offered to correlate the experimental viscosity-concentration relationship and the change of solvent. A possible method for the actual calculation of concentration effects has been applied to the case of a suspension of spherical particles, for which, as a first approximation, a value of 10.6 for the coefficient of the c2 term has been obtained.

Notes: Tensile strengths of natural rubber samples quantitatively crosslinked with decamethylene dis-methyl azodicarboxylate have been determined. The proportion of the azo vulcanizing agent was varied 80-fold, the fraction p of cross-linked units in the products ranging from 0.10x 10-2 to 8.0 x 10-2. In spite of the small dimensions of the test specimen (cross section 0.10 in. by about 0.005 in.), the average error for a single observation is similar to that obtained in conventional testing procedures applied to specimens vulcanized with sulfur and accelerators. Deviations from the mean for 262 tests on 15 different cures are well represented by a Gaussian error function; a slight intimation of skewness is evident only at the extremities of the distribution of deviations. These results are incompatible with the commonly held view that rupture of the sample takes place at the “weakest link” in the series of flaws inevitably occurring along the specimen. It would appear that the ability to withstand tensile stress (applied in a specified manner) is an inherent characteristic of the bulk material. The rapid increase in tensile strength with the degree of crosslinking p for small values of this quantity is consistent with the conclusion, previously reached in the investigation of the physical properties of butyl rubber (6) that the tensile strength is directly related to the fraction of the structure which is permanently oriented by stretching. At higher degrees of cross-linking (at ρ x 100 = 1.0 to 1.5, depending on the molecular weight of the rubber) the tensile strength passes through a maximum and then declines steadily to quite low values for higher ρ's. This adverse effect of higher degrees of cross-linking is believed to result from the diminished elongation at which crystallization sets in and, hence, the smaller fraction of the network elements sufficiently oriented to participate in crystallite formation. Modification of up to 7% of the isoprene units of the rubber with the monofunctional compound, ethyl azodicarboxylate, depresses the tensile strength of the disazo vulcanized rubber relatively little. Evidently, the limited extent to which these modified (or copolymer) chains may enter into crystallization is adequate to bring about high tensile strength. Lack of direct correlation between modulus and tensile strength for rubbers vulcanized with various sulfur-accelerator combinations can only be explained by assuming that conventional vulcanization processes are accompanied to varying degrees by chain scission reactions. When vulcanized rubber is “over-cured,” or heat aged, chain scission becomes excessive.

Notes: The infrared and ultraviolet absorption spectra of Koresin are very similar to those of an acid-catalyzed p-tert-butylphenol-acetal- dehyde polymer. This is believed to be evidence that Koresin has Structure I. A variety of phenol-aldehyde resins have been prepared for comparison with Koresin. The ratings of some of these materials as tackifiers have been reported. These furnish further evidence that the molecular weight of a phenol-aldehyde resin is important in its tackifying action. An alkaline-catalyzed condensation product of acetylene and p-tert-butylphenol shows no tackifying action and has no free hydroxyl in the polymer as judged by its infrared pattern.

Notes: The paper deals with a tentative plan to extend quantitative x-ray measurements for determining the degree of crystallinity of polymers (as used earlier in cellulose determinations), to the investigation of rubber, polythene, and a polyamide. Strictly monochromatized copper radiation was used. The results show that further work along these lines may be promising. The raw rubber sample used was identical with one of those investigated by Goppel and by Arlman. It yielded a crystallinity figure higher than that reported by Goppel, but conformed with that found by Arlman. The figures obtained for polythene at different temperatures are compared with those derived from density and heat capacity measurements by other workers. The degree of crystallinity of polythene at room temperature is found to be ∼ 55%, and in samples of polyhexamethylene sebacamide, previously treated in different ways, figures between ∼50 and 70% are obtained. It is shown that the possibility of using the background intensity as a measure of the disordered fraction is confined to not too high angles of diffraction. A method is described (and experimentally checked) allowing for reducing the curves of intensity distribution as a function of 2θ as observed in various polymers to a comparable scale. It is further emphasized that, in polymers where hydrogen bonds play a decisive role in intermolecular cohesion, the x-ray results suggest that the “crystalline” or ordered fraction may to a considerable extent involve states of two-dimensional order.

Notes: o-Hydroxystyrene did not copolymerize with styrene and butadiene in GR-S-type systems. A terpolymer of these three monomers was obtained by copolymerizing o-acetoxystyrene, styrene and butadiene in carbon tetrachloride solution by the action of stannic chloride followed by hydrolysis of the acetate groups in the polymer with dilute hydrochloric acid in dioxane solution. The polymer thus obtained did not improve the tack of GR-S.

Notes: During a definite polymerization time the rate of uncatalyzed polymerization of styrene may be decreased by the use of activated carbon. The intrinsic viscosities of the polymers formed may be either increased or decreased by the addition of active carbon, depending upon the activity of the carbon sample used.

Notes: A comparison has been made of the ultraviolet spectrophotometric and chlorine analysis methods for the composition of copolymers of butadiene and p-chlorostyrene. The results of the two methods are in good agreement.

Notes: The density-gradient tube is a special glass vessel in which a vertical linear density gradient of almost any desired magnitude can be maintained in a stable condition for prolonged periods by using liquids or solutions of the proper density. This article describes the application of such a device to three typical problems appearing in the field of high polymers: crystallization-rate data, inhomogeneities in composition of copolymers, and rates of polymerization. Possible application to still other problems is indicated.

Notes: The technique of obtaining polystyrene silica replicas from surfaces and cross sections of fibers is described and electron micrographs obtained from replicas of several fibers are shown.

Notes: By embedding submicroscopic silver rodlets in micellar systems with fibrous structures, unexpectedly strongly anisotropic rodlet compounds bodies are formed. The observed dichroism, indigo blue-straw yellow, suggests the possibility that a particular anisotropic modification of silver is being dealt with, instead of cubic silver. On applying Wiener's compound body formulas, it appears, however, that in the short-wave region the observed anisotropic effects (negative rodlet double refraction and dichroism) can be computed from the optical properties of isotropic silver. In the spectral region of the average and long waves, differences are found between the observed anisotropy and the value calculated by means of Wiener's formulas. These differences are, however, not of such a nature that it is necessary to assume an unknown allotropic modification of silver with particular optically anisotropic properties in order to explain them.

Notes: Mixtures of butadiene and dimethylbutadiene with varying quantities of styrene and acrylonitrile have been copolymerized under various catalytic conditions at 100°C. The formation of cyclic dimers from dienes and vinyl compounds is, in some cases, much more pronounced than the dimerization of the dienes alone. Since the latter is also a kind of Diels-Alder reaction, a series with increasing dienophilic properties can be established from a comparison of the yields of cyclic dimers. From solubility and swelling experiments it is evident that the polymers prepared from butadiene and the vinyl compounds and those from dimethylbutadiene and acrylonitrile are true copolymers. The mechanical properties of the copolymers from butadiene and increasing amounts of a vinyl compound change gradually from those of a rubber to those of the polyvinyl compound. The copolymers from a mixture containing 30% styrene or acrylonitrile are very similar, respectively, to Buna S and Perbunan. Copolymers of the same composition from dimethylbutadiene and acrylonitrile, on the other hand, form a new class of thermoplastics with a much higher temperature of elasticity.

Notes: The copolymerization in emulsion of butadiene with varying amounts of styrene and acrylonitrile and of dimethylbutadiene with the latter vinyl compound has been studied on a laboratory scale. An essential difference between polymeric vinyl compounds and polydienes is the olefinic character of the latter, which leads to typical secondary changes in the polymers originally formed during the completion of the reaction. Such changes are of fundamental importance for the ultimate mechanical properties of polymers and copolymers based on dienes. After a brief discussion of the general theoretical line of thought serving as a base for our experiments, the influence of various factors on the rate of polymerization is pointed out and demonstrated by experiments. The changes in the nitrogen content of copolymers from acrylonitrile and the yields as a function of the reaction time have also been studied. From an investigation of some physical properties, such as solubility and swelling, the existence of true copolymers (of unknown irregularity) has been established. Copolymers from dimethylbutadiene and acrylonitrile represent a new class of olefinic thermoplastics resembling gutta-percha. Abbreviations used are BP (benzoyl peroxide) and DAB (diazoaminobenzene).

Notes: The transversal dimension of the statistical coil formed by a long-chain molecule in solution or in bulk rubber is independent of the longitudinal dimension, h1, of the coil as long as h1 is small compared with the length of the stretched molecule. On the other hand, the lateral dimension of a coil whose longitudinal vector, h1, is oriented in the direction of stretch undergoes a forced contraction during the deformation of bulk rubber. The forced contraction of the transversal dimension is therefore supplying an extra contribution to the entropy change. Its amount has been estimated in earlier work and it influences the numerical factor occurring in the formula connecting the strain-stretch relation for rubber with the molecular weight of the net fragments.

Notes: A brief outline is given of the present status of research on the problem of the deformation mechanism of cellulose gels. Emphasis is placed on the prevailing idea of a molecular network structure and recognition of the existence of an intimate relationship between the phenomena of deformation and those of swelling. In these respects the problems involved in cellulose research seem to show a marked convergence with those which emerge from recent developments in rubber research. The principle of Kratky's theory of “affined transformation” and the experiments carried out by this author in order to verify his theory are discussed briefly.

Notes: Measurements of the viscosity of polyvinyl alcohol solutions may be correlated with the chemical nature of the polymer. In particular it is shown that polyvinyl alcohol contains carbonyl groups in the polymer chain, which are to some extent present as ketols. The presence of these keto groups has been demonstrated by a combination of spectrophotometric and chemical techniques. The carbonyl groups appear to be randomly distributed along the polymer chain and also seem to be independent of chain length. The presence of keto groups along the chain allows an explanation for the behavior of aqueous polyvinyl alcohol solutions with acid or alkali.

Notes: The paper offers a verification by x-ray experiments of Kratky's theory of affined transformation (affine Raumverzerrung), aiming at an explanation of the mechanism of deformation of swollen cellulose gels upon stretching. Isotropic model filaments of various degrees of swelling were stretched to various extents and x-ray photographs were then taken. Following Kratky, the intensity distribution along the sickles of two paratropic planes of the ribbon-shaped crystallites, the lamellar plane, A0 and the A3 plane (perpendicular to the latter) were measured and the average orientation, expressed in terms of the orientation factor, fx, was calculated. The superposition of the (021) interference on that of the A3 sickle, which had been neglected by Kratky, was accounted for. The experimental results are in conformity with certain characteristic features of the theory; (a) the orientation of the A0 planes advances more rapidly than that of the A3 planes; (b) regardless of the initial degree of swelling of the isotropic filament, the average orientation is actually a univocal function of the elongation, va. On the other hand, the rate of orientation appears to be much greater than that required by theory; the average orientation, expressed in terms of the orientation factor, increases almost twice as rapidly. It is shown, moreover, that, apart from low degrees of orientation, Kratky's theory fails to explain the velocity function of the relative rotation of the crystallites, as derived from the experimental data according to a procedure proposed by Kratky. It is concluded that further work is required to elucidate the mechanism of deformation.

Notes: A convenient procedure for the determination of polystyrene in GR-S is given. The method depends upon degradation of the GR-S by tert-butyl hydroperoxide in the presence of a trace of osmium tetroxide. Polystyrene is not attacked and is separated by precipitation with ethanol.

Notes: By suitable alteration of the exponent the fit of viscosity-concentration data by a Philippoff type equation may be substantially improved. For a series of cellulose acetate solutions the exponent of best fit appears to be related to the physical properties and chemical type of the solvent.

Notes: A new method for the determination of the activation energy for the setting of silicic acid solutions is described. Experimental results point to the conclusion that setting to a gel occurs at a definite stoichiometric point which is independent of temperature.

Notes: In using the velocity ultracentrifuge in conjunction with free diffusion measurements for the determination of molecular weights of polymolecular systems, it is desirable to know what average molecular weights are obtainable. These average values are not simple moments of the molecular weight frequency curve, but depend on the methods of experimental observation, and on the hydrodynamic character of the molecules in solution. Most often, however, the average molecular weight falls between the number and weight-average values.

Notes: The kinetics of the polymerization of methyl methacrylate in emulsions has been examined. The initial rate of polymerization is independent of the amount of discontinuous monomer phase present and it is concluded that initiation occurs in the aqueous phase. Monomer is observed to disappear from the discontinuous phase more quickly than it polymerizes, and probably forms a monomer-polymer mixture with the polymer produced. An increase in the subsequent rate of polymerization is attributed to the fact that propagation of the growing polymer chains occurs in this monomer-polymer mixture and this entails a decrease in the velocity of termination due to the high viscosity of such mixtures.

Notes: The elastic and plastic properties of high polymeric materials are functions of the forces holding the molecules together and the ability of small segments of the chains to move about, relative to other segments. Internal plasticization is accomplished by introducing structural features which reduce intermolecular and intramolecular forces and which, in general, decrease the order of the polymer. Physical techniques used to determine the effect of such structural features have been discussed. The results of numerous investigations into the relations between structure and properties have been considered in an attempt to present a simple and consistent physical picture of these relations.

Notes: Refractive index-temperature and volume-temperature data for polyvinyl acetate and polybutyl acrylate have been determined and used to calculate the specific refractivity as a function of temperature by the Lorentz-Lorentz and the Eyckman equations. The results show a variation of specific refractivity with temperature which exceeds experimental error. The variation is evident as an increase of specific refractivity with temperature above Tm and suggests experiments which will show the increase more clearly. A theoretical justification for the increase of specific refractivity with temperature in polymers is discussed.

Notes: Various mechanisms which have been proposed for the reaction between ferrous iron and hydrogen peroxide are reviewed. Two of these mechanisms appear equally satisfactory on the basis of present evidence: that of Bray and Gorin, involving formation of tetravalent (ferryl) iron, and that of Haber and Weiss, involving formation ofhydroxyl and perhydroxyl radicals. Mechanisms involving these intermediates have been developed for the ferric iron-catalyzed decomposition of hydrogen peroxide; the mechanism of Bray and Gorin appears satisfactory, while that of Haber and Weiss leads to certain inconsistencies. A mechanism with ferrous iron and hydroxyl and perhydroxyl radicals as intermediates is presented, which accounts for the experimental observations in the ferric iron-catalyzed decomposition of hydrogen peroxide. The reaction between hydrogen peroxide and ferrous iron induces the reaction between hydrogen peroxide and various organic compounds; this induced reaction (termed a Fenton reaction) is a chain reaction which is initiated by the active intermediates (ferryl iron or hydroxyl radical) formed in the primary reaction. The determination of hydrogen peroxide by reaction with ferrous iron, if carried out in the presence of various organic compounds (termed promoters), gives incorrect results, due to induced oxidation of the promoters; other compounds, termed suppressors, reduce the extent of this induced oxidation of promoters. It appears that both types of compounds react with ferryl iron or hydroxyl radical to form radicals, which if formed from promoters, reduce ferric iron or hydrogen peroxide, while the radicals formed from suppressors oxidize ferrous iron. These considerations are of particular importance in the determination of organic hydroperoxides. The reaction between hydrogen peroxide or organic peroxides and ferrous iron can induce, or initiate, the polymerization of vinyl monomers; this is an example of redox polymerization, which is of considerable practical importance at present.

Notes: Cellobiose wird als Glykolsäure-glykosid nach der Azid-methode mit Protein kondensiert. Ferner werden aus Lacto-bionsäure und Proteinen nach der Azid-methode Lacto-bionyl-proteine synthetisiert.

Notes: 2-Chlorbutadien-1,3 (Chloropren) bildet schon bei niedrigen Temperaturen mit molekularem Sauerstoff ein Peroxyd. Die Bildungsgeschwindigkeit dieses Peroxyds ist so groß, daß es schwierig ist, peroxydfreies Chloropren herzustellen. Dies gelingt durch sorgfältige Destillation unter reinstem Stickstoff.Die Versuche zur Isolierung des Chloroprenperoxyds ergaben wegen seiner leichten Zersetzlichkeit keine befriedigenden Ergebnisse. Die quantitative Bestimmung des Peroxyds in Chloropren gelingt nach den üblichen Methoden nicht oder nur sehr schwierig. Chloroprenperoxyd oxydiert Antioxydantien, wie z. B. Phenyl-β-naphthylamin, zu gefärbten Produkten. Da Phenyl-β-naphthylamin die weitere Autoxydation des Chloroprens unterbindet, kann auf dieser Grundlage eine photometrische Bestimmung des Peroxyds durchgeführt werden.Die Geschwindigkeit der Sauerstoffaufnahme von Chloropren bei 0°C wurde volumetrisch gemessen. Die Sauerstoffaufnahme ist ein autokatalytischer, gleichförmig beschleunigter Vorgang. Er wird von Tetralinperoxyd stark beschleunigt, nicht aber von Benzoylperoxyde. Es wurden andere katalytisch wirkende Stoffe und der Einfluß des Lichtes untersucht.Da Chloroprenperoxyd die Polymerisation des Chloroprens auslöst und stark beschleunigt, ist es zweckmäßig, für die Stabilisierung dieser ungesättigten Verbindung solche Stoffe anzuwenden, die nicht nur die Polymerisation, sondern auch die Autoxydation hemmen. Die stabilisierende Wirkung der Antioxydantien wird bei Anwesenheit von Chloroprenperoxyd und Chlorwasserstoff nach einer Induktionsperiode durchbrochen. Dies kann dadurch erklärt werden, daß Phenyl-β-naphthylamin im Falle des Chloroprenperoxyds durch Oxydation, im Falle des Chlorwasserstoffs als Salz unwirksam gemacht wird.

Notes: Es wird ein einfaches Verfahren angegeben, den Transport von Wasser durch Textilfaserschichten zu messen, die einerseits in Wasser tauchen und andererseits an Luft mittlerer relativer Feuchtigkeit grenzen. Proteinfasern transportieren das Wasser langsam, Baumwolle rascher und Viskosekunstfasern sehr rasch.

Notes: Es wird der Einfluß von Seitenketten auf die Höhe der Viskositätszahl Zη von niedermolekularen Estern untersucht. Bei Dicetylmalonestern und Dicetylessigestern hat die Länge der Estergruppe keinen Einfluß auf die Höhe der Viskositätszahl. Bei substituierten Malonsäuredicetylestern resp. Disterylestern ändert sich dagegen die Viskositätszahl mit der Länge der in die CH2-Gruppe substituierten Alkylreste, und zwar ist sie bei kurzkettigen Substituenten kleiner, bei langgliedrigen größer als bei den unsubstituierten Verbindungen gleicher Kettenlänge. Das gleiche Verhalten zeigen β-alkylsubstituierte Stearinsäureester. Es wird weiter versucht, inwieweit das Einsteinsche Viskositätsgesetz in veränderter Fassung zur Klärung dieses Viskositätsverhaltens herangezogen werden kann.

Notes: Da nach den üblichen technischen Verfahren die Isolierung von Pektinen unvermeidlich mit einem gewissen Abbau verbunden ist, wurden die schonendsten Bedingungen festgestellt, unter denen es noch möglich ist, Pektin aus dem Fruchtfleisch von Äpfeln in ausreichender Menge zu gewinnen. Durch Wahl des günstigsten Reifezustandes, Ausschaltung einer Enzymtätigkeit, Einhaltung möglichst neidriger Extraktionstemperaturen und eines geeigneten pH konnte ein wenig abgebautes Pektin crhalten werden, dessen Methoxylgehalt dem für eine vollständig veresterte Polygalakturonsäure berechneten sehr nahe kommt. Die Molekulargröße wurde viskosimetrisch an einem aus diesem Material gewonnenen Pektinnitrat bestimmt, das nach einer schonenderen Nitrierungsmethode als der bisher üblichen hergestellt war. Die erhaltenen Werte liegen über den bisher beobachteten. Das im Apfelsaft vorhandene „freie Pektin“ wurde als niedriger molekular befunden.

Notes: Aus einer Zusammenstellung von Viskositätsmessungen an niedermolekularen Kettenmolekülen mit Doppelbindungen geht hervor, daß die trans-Formen eine höhere und die cis-Formen eine niedrigere Viskosität haben als die analogen gesättigten Verbindungen. Die verschiedene Viskosität von trans-, cis- und gesättigten Verbindungen kann mit der verschiedenen Form und Elastizität der Kettenmoleküle erklärt werden. Maleinsäureester und Maleinsäurepolyester lassen sich mit Brom als Katalysator in die entsprechenden Fumarester umlagern.

Notes: Es wurde die Polymerisationsgeschwindigkeit des Vinylchlorids bei 50° unter Zusatz verschiedener Mengen Benzoylperoxyd als Katalysator untersucht, indem der Umsatz das Molekulargewicht und die Zeitdauer beginnender Trübung durch Ausfallen des Polymerisats bestimmt wurde.Der Umsatz steigt zunächst langsam, dann schneller mit der Zeit an; über 80 - 90% nimmt er nur noch sehr langsam zu, so daß die letzten 1 - 2% praktisch nicht mehr polymerisieren. Mit sinkender Katalysatorkonzentration wird die Reaktion langsamer, derart, daß sich die Umsatzzeitkurven durch Dehnung der Zeitachse proportional der Wurzel aus der ursprünglichen Katalysatorkonzentration zur Deckung bringen lassen. Bis zu 70 - 80% läßt sich der Umsatz durch die Gl. u = (eat  -  1)b gut darstellen, wobei b = 0,1 und a = 0,240 · √c20 zu setzen ist.  -  Gegen Ende der Reaktion ließ sich im restlichen Monomeren Salzsäure nachweisen.Das Molekulargewicht nimmt mit steigender Menge Benzoylperoxyd und Sauerstoff deutlich ab; bei mäßigen Peroxydkonzentrationen wird dieser Effekt in der Streuung der Versuchswerte nicht sichtbar.  -  Die Zeit bis zur beginnenden Trübung ändert sich umgekehrt proportional der Wurzel aus der Katalysatorkonzentration und ist ferner unter reinem Sauerstoff größer, unter Stickstoff kleiner als unter Luft.Der Katalysator verschwindet während der Polymerisation, und zwar in um so stärkerem Maße, je höher seine ursprüngliche Konzentration, und um so weniger, je größer die Menge an Sauerstoff ist.  -  Mit Kaliumpersulfat erfolgt die Polymerisation nur in der wäßrigen Phase.Mit den Beobachtungen im Einklang steht das folgende Reaktionsschema (B = Benzoylperoxyd, M = monomeres Vinylchlorid). Die Hemmung der Polymerisation durch eine Anzahl von dritten Stoffen wurde orientierend, etwas eingehender die durch Styrol untersucht. Schwache Reduktionsmittel, Vinylderivate und einige anorganische Stoffe hemmen die Polymerisation, während chemisch indifferente Stoff sie nicht beeinflussen. Die Hemmung beruht vermutlich auf vorzeitigem Kettenabbruch.

Notes: Der bei Pektinpräparaten festzustellende P2O5-Gehalt in der Asche stammt teils von anorganischen und organischen Begleitstoffen (Phosphatiden), teils aber auch von Veresterungen der Pektinketten durch H3PO4; im Protopektin der Pflanzen hat man Vernetzungen der Pektinketten durch H3PO4-Brücken zu „Übermolekülen“ anzunehmen, die zahlenmäßig jedoch gegenüber den Vernetzungen durch mehrwertige Metallionen über die Karboxylgruppen zurücktreten.

Notes: 1Die Untersuchungen von Ohara (1933) über das polarisationsoptische Verhalten des Seidenfibroins wurden einer kritischen Betrachtung unterzogen und dabei Abweichungen von der Wienerschen Theorie festgestellt.2Die Hauptbrechungsindizes von Fibroinfäden, die 40 Stunden bei 100°C über P2O5 getrocknet worden waren, wurden nach der Immersionsmethode von Becke gemessen und zu für eine Wellenlänge von 589 mμ gefunden. Daraus resultiert eine Doppelbrechung von 0,0506 und für den Schwerpunkt des weißen Lichtes bei 550 mμ eine solche von 0,0514.3Imbibitionsversuche an trockenem und feuchtem Fibroin ergeben eindeutig das Fehlen eines Stäbchendoppelbrechungseffektes. Dies steht im Gegensatz zu den Untersuchungen von Ohara. Der mit Hilfe der Imbibitionsmethode gefundene Wert der Doppelbrechung stimmt mit dem der Immersionsmethode gut überein.4Die Abweichungen von diesem Wert in Wasser und wäßrigen Gemischen als Imbibitionsmedien lassen sich weitgehend mit Hilfe der Quellung erklären.5Durch Quellung und nachfolgende Imbibition kann keine Verstärkung eines etwa nur schwach entwickelten Stäbchendoppelbrechungseffektes erreicht werden.6Elektronenoptische Aufnahmen vermitteln Einblick in den Feinbau des Fibroins, welches aus feinen Fibrillen mit ganz verschiedenen Durchmessern besteht, die sehr fest und dicht gepackt sind.

Notes: Die molekulare Ultraschallgeschwindigkeit ist eine additive Eigenschaft für substanzen mit niedrigem Molekulargewicht und ergibt sich mit großer Annäherung aus der Summe der Geschwindigkeiten der verschiedenen Radikale, die das Molekül bilden.Die Übereinstimmung zwischen der experimentell festgestellten Geschwindigkeit und der auf Grund der Geschwindigkeiten der Radikale berechneten scheint besser zu sein als die auf Grund der Bindungsgeschwindigkeiten nach Lagemann und Corry berechneten. Wie schon von diesen Forschern angegeben, liefern die Bindungsgeschwindigkeiten zufriedenstellende Ergebnisse nur für unverzweigte Verbindungen.In den Verbindungen mit niedrigen Molekulargewichten führt die Gegenwart von Verzweigungen in den Molekülen zu einer Erniedrigung der molekularen Ultraschallgeschwindigkeit, die jedoch sehr geringfügig ist (12 Einheiten für die Methyl- und 20 für die Äthylgruppe in anderen Stellungen als  - 1 und  - 2).Für Verbindungen mit hohem Molekulargewicht und unverzweigter Kette stimmt der Wert der grundmolekularen Ultraschallgeschwindigkeit, d. h. bezogen auf die Grundgruppe, mit den auf Grund der Bindungsgeschwindigkeiten und auch mit den auf Grund der Radikalgeschwindigkeiten berechneten überein. Für verzweigte Moleküle ist die Übereinstimmung für die Radikalgeschwindigkeit besser, wobei die Erniedrigung der grundmolekularen Geschwindigkeit infolge der Verzweigungen höher ist im Falle von Verbindungen mit hohem als mit niedrigem Molekulargewicht.Die Bestimmung des Verhältnisses zwischen der experimentellen und berechneten Geschwindigkeit kann ein Mittel darstellen, um Anzahl und Größe der Verzweigungen zu bestimmen. Diesem Verhältnis ist die Bezeichnung „Formfaktor“ gegeben worden.Der Formfaktor ist gleich 1 oder etwas höher für die unverzweigten oder wenig verzweigten Verbindungen mit hohem Molekulargewicht (Paraffine, Polyäthylene, Polyäthylenoxyd, Nylon), niedriger als 1 für die verzweigten Verbindungen: 0,90 im Naturgummi, der eine Methylverzweigung an jedem vierten Kohlenstoffatom in den langen Ketten trägt, und 0,79-0,80 in den Polyisobutylenen mit zwei Methylverzweigungen an jedem zweiten Kohlenstoffatom in der Kette.Niedrigere Werte als der Naturgummi und unwesentlich höhere als die Polyisobutylene weisen die Formfaktoren des Polyalfabutylens und des Polystyrols auf (0,82), die eine äthylische, bzw, eine phenylische Gruppe auf jedes zweite Kohlenstoffatom in der langen Kette besitzen; dies beweist auch einen Einfluß der Länge der Verzweigungen auf den Formfaktor in dem Sinne, daß letzterer um so mehr erniedrigt wird, je größer die Verzweigung ist.Der Einfluß der Häufigkeit der Verzweigungen und ihrer Länge auf den Formfaktor wird beim Polymethylmetakrylat bestätigt, da diese Verbindung den niedrigsten von uns festgestellten Formfaktor aufweist; diese Verbindung besitzt nämlich zwei Verzweigungen an jedem zweiten Kohlenstoffatom der Kette, und zwar eine Methylgruppe und eine methylierte Karboxylgruppe.Die Übereinstimmung der Formfaktoren des Buna und des hydrierten Buna mit dem des Polyalfabutylens ist ein weiterer Beweis, daß die nicht bei niedriger Temperatur ausgeführte Polymerisierung des Butadiens in den Stellungen 1 - 2 oder 2 - 3 und nicht in der Stellung 1 - 4 stattfindet. Dies steht im Einklang mit den kürzlich mitgeteilten Versuchsergebnissen von Hanson und Halverson bei gestrecktem Buna.Der niedrige Formfaktor des Polyvinylisobutyläthers (0,79) kann ein Zeichen für den Einfluß der sekundären Verzweigungen auf den Formfaktor sein.Schließlich bestätigt die Unveränderlichkeit des Formfaktors für alle untersuchten Polyisobutylene mit verschiedenen Molekulargewichten, daß für hochmolekulare Verbindungen die grundmolekulare Ultraschallgeschwindigkeit und demzufolge der Formfaktor vom Molekulargewicht unabhängig ist.

Notes: Durch Untersuchung der Kettenlängenverteilung von oxydative in Kupferoxydammoniaklösung abgebauten Buchen- und Fichtencellulosen wird gezeigt, daß ebenso wie bei Baumwoll- und Ramiecellulosen eine Vereinheitlichung bei Polymerisationsgraden zwischen 400 und 500 eintritt. Auch bei hydrolytisch abgebauten Produkten ist dieser Effekt zu bemerken, läßt sich jedoch wegen der Viskositätsanomalien nicht quantitativ auswerten. Aus den Befunden wird geschlossen, daß auch in Holzcellulosen im Abstand von ca. 500 Glucoseresten Lockerstellen vorhanden sind, die durch Oxydationsmittel und Säuren erheblich schneller gespalten werden als die β-glucosidischen Bindungen. Diese periodische Struktur ist für alle bisher untersuchten Cellulosearten im Abstand der Lockerstellen identisch, jedoch unterschieden durch das Verhältnis der Spaltungsgeschwindigkeiten. Aus elektronenmikroskopischen Aufnahmen wird geschlossen, daß die Lockerstellen in Ebenen senkrecht zur Faserrichtung angeordnet sind.

Notes: The relationships which exist between the solvation and shape of colloid particles, as well as between solvation and viscosity, are discussed. From the most important experiments on these subjects the following conclusions may be drawn. Solvation is dependent on the size and shape of the particles and on the chemical relationship of the dispersion medium to the dispersed portion. True or chemical solvation is generally very small and there is no reason to speak of a thick and tightly bound solvent layer. The solvation of compactly formed particles of spherocolloids is greater, the smaller the particles (all other conditions being equal). Solvation is greater, the more loosely the particles are formed (for dispersion mediums having similar chemical properties and for similar dispersed or dissolved substances). Solvation increases when a corpuscular particle is transformed into a rodlike or threadlike particle; the viscosity also increases; if the threads are further broken down into smaller fragments, then the viscosity decreases, but a further increase in the chemical solvationis supposed to occur. The stability of many lyophilic sols, especially the proteins, depends very little on the solvation; it is mainly dependent on the chemical characteristics of the atomic groups which are found on the surfaces of the particles; the least stability is found for surfaces which contain heteropolar groups - that is, surfaces which have the same number of positive and negative ions.

Notes: The conductance in diphenyl ether at 35°C. and 60 cycles of the picrate of a 4-vinylpyridine-styrene copolymer (10:90) has been measured. The equivalent conductance, based on the nitrogen content, is of the same order as that of picoline picrate, and varies with concentration in accordance with the law of ion association.

Notes: A polyvinyl acetate polymer of medium viscosity was fractionated into sixteen fractions (first series) and three of the larger of these further separated into twenty subfractions (second series). The intrinsic viscosities and the osmotic pressure molecular weights were determined at 25°C. and the relation between them was found to be expressed by the equations: first series: [η] = (1.88 × 10-4) M0.69; second series: [η] = (1.76 × 10-4) M0.68. The data indicate that little, if any, increase in homogeneity is to be expected by further successive fractionations and that the equations applicable to the second fractionation series are representative of essentially homogeneous polyvinyl acetates in acetone. An equation applicable to fractionated and unfractionated vinyl acetate polymers is described that is useful in obtaining the intrinsic viscosity from a single viscosity measurement. Several unfractionated materials from different sources were also studied and the calculated ratios of the viscosity-average to the number-average molecular weight indicate that the degree of heterogeneity of chain-length distribution increases with increasing average molecular weight.

Notes: The rate of emulsion copolymerization of butadiene and styrene, with soap as emulsifier and potassium persulfate as catalyst, is extremely small at 50°C. The presence of very small amounts of high-molecular mercaptans promotes the copolymerization reaction. The promoting effect is at a maximum for primary, secondary, and tertiary dodecyl mercaptans and decreases for mercaptans of either higher or lower molecular weight. The promoting effect is independent within wide limits of the amount of mercaptan added after the minimum quantity has been exceeded. Mercaptans which are poor promoters may be so because they fail to bring about chain initiation or because they aid in chain termination. The low-molecular mercaptans which are poor promoters prevent the high-molecular mercaptans from exerting their good promoting effect when a mixture of both types of mercaptans is used. The mechanism of the promoting effect of mercaptans upon the emulsion copolymerization of butadiene (75 parts) and styrene (25 parts) or upon the polymerization of butadiene alone is not yet clear.

Notes: An empirical equation is described which predicts the intrinsic viscosities of 75:25 butadiene-styrene emulsion copolymers from data on mercaptan consumption. The form of the equation is as follows: \documentclass{article}\pagestyle{empty}\begin{document}$$\left[ \eta \right] = \left[ {\frac{{aP}}{{R_0 R}} + \frac{b}{{R_0 }}\int_0^P {\frac{{dP}}{{dR}}dP} } \right]^c $$\end{document} where [η] is the intrinsic viscosity, P is the conversion, R is the fraction of mercaptan consumed, R0 is the mercaptan charged, and a, b, and c are constants equal to 0.22, 1.12, and 0.66, respectively. Evidence is given that the intrinsic viscosity of mutual-recipe polymers can be described satisfactorily in terms of mercaptan consumption.

Notes: The calculation of the “theoretical” minimum modifier requirement, TMMR, for butadiene-styrene (75:25) copolymers of given intrinsic viscosities at various conversions is described. The TMMR of 75% conversion polymer of intrinsic viscosity 2.0 is R0 = 0.28, which is only about 60% of the value of R0 for commercial primary dodecyl mercaptan (C.M.) in large-size reactors. The efficiency, E, of a modifier for 75:25 butadiene-styrene copolymerization using soap as emulsifier and persulfate as catalyst has been defined by and calculated from the equation: \documentclass{article}\pagestyle{empty}\begin{document}$$E = \frac{{P\left( {0.2 + P} \right)}}{{R_0 \left( {M_v \times 10^{ - 5} } \right)}}$$\end{document} where Mv is the intrinsic viscosity molecular weight of the polymer at conversion, P, with an amount of mercaptan, R0, charged with the monomers. For all modifiers the efficiency is low at low conversions and gradually increases to a maximum with increasing conversion. For different modifiers this maximum may be as at low as 10% conversion or at higher than 80% conversion. After the maximum the efficiency decreases with further increase of conversion. The most efficient modifier for the production of polymers of intrinsic viscosity, [η], at conversion, P, will meet the following conditions, R = 1 at conversion, P, and dP/dR is constant. Inefficient modifiers may be made more efficient by changing the conditions or procedure of polymerization in such a way that the modifier more nearly complies with the above conditions.

Notes: The swelling of vulcanized natural rubber, Buna S, neoprene, Perbunan, and Perbunan Extra was measured in various organic liquids. The essential influence of polarizability and polarity on the swelling equilibrium is illustrated by comparing series of solvents and polymers with gradually changing geometrical and chemical structure. Experiments on the influence of temperature and of strain on the swelling equilibrium are reported. The swelling properties of ebonites and similar derivatives allow some conclusions on the amount of cross linking in these polymers.

Notes: This and a following article review the literature from 1910-1945 on polymerization of olefins and diolefins in suspension and emulsion, and present a number of new measurements not published to date. The subject is considered mainly from the point of view of scientific information on the mechanism of polymerization in aqueous suspensions and emulsions, but brief mention is also made of the more important disclosures in the patent literature. The new data presented in this article refer to (1) initial rates of polymerization of styrene, methyl methacrylate, vinyl acetate and acrylonitrile in aqueous suspensions and in soap emulsions as a function of catalyst concentration, temperature, and soap concentration; (2) influence of water-soluble activators, such as sodium bisulfite; (3) influence of initial size of monomer droplets on initial rates of monomer consumption; (4) study, with the aid of the electron microscope, of size of monomer droplets and polymer particles throughout polymerization; and (5) a few experiments on side reactions in the domain of higher conversions. No attempt is made in this paper to review and appraise the very large number of recent patents (from about 1930 on), which protect special procedures on the use of various promoting, regulating, or modifying ingredients. A complete digest of this practice does not exist at present, but reference may be made to the excellent chapter on emulsion polymerization in the book of Talalay and Magat (55), to the very valuable compilation of patents by Hosch in “India Rubber World” (27), and to the enumeration of a selected number of patents in the book of Scheiber (50) on pages 210-213.

Notes: In this paper it is emphasized that density is a typically macroscopic concept which cannot be transferred to small particles or bodies with small pores. This fact has not been duly accounted for in earlier work on the density of cellulose fibers. The usual interpretation of density determinations in various media is criticized and the point of view that cellulose fibers should be considered as porous bodies is dropped. The pore system occurring in native fibers is coarse enough not to interfere with density determinations in indifferent organic liquids.It is shown experimentally that even substances like water and glycerol penetrate very slowly into dry cellulose and that indifferent organic liquids do not penetrate at all. In the experiments with water vapor a new phenomenon, the heterogeneous course of absorption, is observed, giving rise to sharp microscopically visible boundary lines between the wet and the dry portion of the fiber.

Notes: Accurate determinations of the density and refractive power of various types of model filaments prepared from viscose in the bone-dry state were made. The molecular refraction, according to the formula of Gladstone and Dale, was calculated and shown to correspond within three parts per thousand with the refraction calculated from atomic refractions. It is concluded that the liquid used for the density determinations (mixtures of carbon tetrachloride and nitrobenzene) does not penetrate into the fiber substance. The technique of the experiments is described in detail. The immersion liquids used in the optical measurements were standardized mixtures of butyl stearate and tricresyl phosphate; contamination with moisture during the measurements was definitely excluded.

Notes: Nitrogen was introduced into cellulose acetate by the incorporation of amino acid groups. Some of the mixed esters were soluble in organic solvents. Cast films were examined for dyeing properties with wool-type dyes. Many of the mixed esters containing above 1% nitrogen dyed well with xylene brilliant blue, a typical wool-type dye, Mixed cellulose acetate esters were synthesized containing glycine, dl-α-alanine, β-alanine, and α-isoaminobutyric, d-glutamic, and p-aminobenzoic acids. The amino groups were protected by acetylation prior to use. The N-acetyl amino acids were treated with cellulose diacetate in dioxane solution containing chloroacetic anhydride and magnesium perchlorate as impellor and catalyst, respectively. Cellulose acetate metanilate was prepared by use of the aminosulfo acid chloride. An attempt was made to prepare 6-aminocellulose by the reaction of aqueous ammonia, and also sodamide in liquid ammonia on 6-iodotosylcellulose acetate. Only traces of nitrogen were introduced by this procedure and the recovered products showed signs of extreme degradation.

Notes: A micromethod is described for density determinations of artificial cellulose fibers by the floating method, in carbon tetrachloride in a dry and air-free atmosphere, permitting an accuracy of one part per thousand. The density is derived from the temperature at which the samples neither sink nor float. This temperature varies, with the kind of fiber, between 45° and 65°C. Measurements of the refractive power were carried out simultaneously. The refractivity calculated according to Gladstone and Dale yields slightly lower figures than those found for model fibers in the preceding communication (Part II). The average difference amounts to seven parts per thousand and is ascribed to a systematic error in the density determinations, due to a slight absorption of carbon tetrachloride by the fibers at elevated temperatures. This explanation is supported by experiments. The density of native ramie, of contracted and reoriented ramie, and of several series of rayon fibers spun with increasing stretch, is given. The conclusion is reached that the density is a measure of the percentage of crystalline substances in the fiber and it is shown that there is a correlation between density and sorptive capacity toward water vapor. The percentage of crystalline matter is estimated to be 55-60% in native ramie and 20-25% in rayon.

Notes: In the following paper the crystallization of submicroscopic cellular strands from multimolecular, cellulose films is illustrated by electron micrographs.

Notes: The ultraviolet absorption of polystyrene, with maximum absorption at 262 mμ, is due to the presence of phenyl residues in the polymer. The specific extinction coefficient is constant, i.e., independent of the molecular weight of the polymer. This shows that the extinction of the phenyl residues is additive. On the basis of this fact, it is shown that the styrene content of a butadiene-styrene copolymer (such as GR-S rubber) can be determined by ultraviolet spectrophotometry. The relative precision of the determination is about 1%, the probable relative accuracy is about 3%.

Notes: The chief result of irradiation of cellulose acetate-butyrate with ultraviolet light is cleavage of the cellulose chains so that the average molecular weight is lowered. The exposed material dissolves more slowly in common solvents than the unexposed. The acyl content and carbon and hydrogen contents are not altered appreciably. The rate of the reaction is independent of the type of common mineral impurities, the concentration of expected decomposition products, such as acetic and butyric acids, and the molecular weight of the cellulose ester. Ultraviolet light is necessary for appreciable reaction. Wave lengths shorter than 3500 Å. are most effective; longer wave lenghts are much less active.

Notes: The polymerization of dienes under the influence of heat, of diazoaminobenzene, and of benzoyl peroxide, has been studied. Diazoaminobenzene probably acts as a slow, but continuous, source of free radicals giving quantitative yields of methyl rubber of good quality from dimethylbutadiene after a few days to a few weeks at 100-125°C. The effect of diazoaminobenzene on butadiene is the same, but the rate of the uncatalyzed dimerization of butadiene at 100°C. is as fast as that of dimethylbutadiene at 150°C. and, therefore, dimerization interferes much more strongly with the polymerization of butadiene. Only 60% of polybutadiene has been obtained. A few experiments with isoprene showed its position between the two other dienes. The effect of benzoyl peroxide on the polymerization of the dienes appeared to be much smaller than was expected from its known catalytic action on the polymerization of vinyl compounds, e.g., styrene, for in this case its activity is about 100-1000 times greater than that of diazoaminobenzene. It is assumed that a larger number of radicals, produced by the fast decomposition of benzoyl peroxide, causes this difference. Diazoaminobenzene has about the same effect on the polymerization of styrene and acrylonitrile as on that of the dienes. The catalytic effect of benzoyl peroxide on the polymerization of the dienes is, on the contrary, 10,000-100,000 times smaller than on that of the vinyl compounds. This difference can be understood by the assumption that the catalyst is quickly used up by oxidizing the dienes and their polymers. While methyl rubber and the polyisoprene resemble raw natural rubber, polybutadiene, prepared under comparable conditions, is hard, swells moderately, and has a tendency to become brittle as a result of oxidation (aging). Abbreviations used are DAB (= diazoaminobenzene) and BP (= benzoyl peroxide).