Search

The academic efficiency and social justice of entry procedures at Oxford and Cambridge Universities are examined over the past quarter of a century. For each major subject the mean A‐level scores of males and females entering from state and independent schools are compared with mean final examination scores in the major subjects. In any comparison of state and independent cohorts of the same gender, within the bounds of normal statistical fluctuation, the difference in A‐level score is a good predictor of the difference in finals score. For example, when between state men and independent men the difference in A‐level score is zero, the difference between mean finals score is zero also. The origin of female under‐achievement is examined. In most subjects there is pronounced gender inequality due to the following chain of circumstances: (1) to break‐even in finals women require at entry better grades at Advanced Level than men; (2) women used to have much the better A‐levels and so, in finals a qua...

The origins of the superior academic performance of men compared with women at Oxford and Cambridge are examined for the years 1987 to 1991, and compared with performance at UK universities excluding Oxford and Cambridge over the same years. A Norrington scale (5 for Class 1, 3 for Class 2:1, 2 for Class 2:2, 1 for Class 3) was used to obtain an average score for four cohorts (men and women educated in state schools, men and women educated in independent schools) in each of the major subjects. For UK universities men and women performed roughly equally (men superior in seven subjects, women in six), in good agreement with the observations of Clarke for the years 1976 to 1979. At Oxford men excelled in History, Engineering, Chemistry, Mathematics, Physics, PPE, Law and English: women excelled in no subject: state‐school men performed better than independent‐school men: independent‐school women performed better than state‐school women. The abrupt drop in the performance of women, which coincided with the mi...

The temperature and stress dependence of creep of human digital tendon was studied in vitro at temperatures between 10 and 60°C and at tensile strains below 2%. The activation energy, Δ H , for the viscoelastic deformation mechanism was determined by imposing small, abrupt changes in temperature on the specimen during creep. The value of Δ H at 28°C was 12 (±1·5) kcal mol −1 for specimens from donors of age 21, 55 and 89 yr. The activation energy for the shrinkage transition in human digital tendon was determined as 250 kcal/mole using similar techniques. The stress dependence of the isochronal strain rate is roughly of the form e α δ n , n being less than unity. The observed viscoelastic phenomena are discussed in terms of the structure of the collagenous tissue. The advantages of the creep T -jump technique for the study of biological polymers are also discussed.

Volume relaxation of isotactic polypropylene was studied at the temperature region of the α-relaxation by the technique of thermal stimulation (or temperature-jump). It is shown that the interpretation of volume relaxation experiments with viscoelastic models fails unless the limiting values of the bulk compliance are assumed to be temperature dependent. Much simplification in the theory of volume relaxation follows if the rigorous formalism of viscoelasticity is adopted, particularly if the treatment by one model is intended for the interpretation of controlled experiments at a single temperature, for both volume and mechanical relaxation. With this model an experiment is designed, based on the Boltzmann superposition principle. The activation enthalpy for volume relaxation, ΔH v , can then be determined if the Boltzmann principle is applied at an appropriate reduced time, t a v (T) , where a v (T) is the Arrhenius shift factor. For polypropylene at 40°C ΔH v = 35 kcal mol 1 , equal to ΔH J the activation enthalpy for creep (at 40°C).

The kinetics of the α relaxation of a crosslinked copolymer of acrylonitrile and butadiene ( T g = −7°C) were studied in the temperature range ( T g + 17°C) down to ( T g − 8°C). The techniques used were shear creep analysed by time-temperature ( t - T ) superposition and thermal sampling (TS) with correction procedures proposed by McCrum. In this range the kinetics do not follow the compensation rule, as had been proposed in the pioneering TS experiments by Zielinski, Swiderski and Kryszewski and by Lacabanne et al. . The McCrum correction removes a discrepancy between the pioneering TS experiments and the conclusions of classical t - T superposition experiments. The methods of TS and t - T superposition are compared. At low temperatures, below ( T g + 3°C), the TS method is superior: t - T superposition is unreliable due to lack of normalization and to the physical ageing perturbation. At temperatures from ( T g + 3°C) to ( T g + 11°C) t - T superposition is highly reliable since normalization is not required and there is no ageing. At temperatures above ( T g + 11°C) the precision of t - T superposition depends on the validity of the normalization procedure. It has yet to be determined whether or not the compensation rule applies at temperatures below ( T g − 8°C): the method most likely to settle this important question is TS, mechanical or dielectric variant, with McCrum correction for the distribution of relaxation times.

It has been reported that the water-swollen crosslinked poly(2-hydroxyethyl methacrylate) (PHEMA) gel displays an increase of entropy on extension at constant volume in a simple tensile experiment. In this paper, the authors report a new technique for measuring the entropy contribution to tensile load in a solvated polymer. Samples immersed in water at a reference temperature (T//0 equals 30 degree C) are extended to a series of lengths at each of which equilibration with the neighboring solvent is permitted. At each of these lengths, the tensile force is measured as a function of temperature with the sample uptake of water constant by changing the temperature of the sample in a much shorter time than that necessary for changes in water uptake to occur to reestablish osmotic equilibrium at the new temperature. The principle of the experiment is thus the measurement of load at different temperatures at thermal but not osmotic equilibrium. Experimental data are presented and evaluated by using mathematical techniques.

It is shown that the Bucci relaxation time, τB, is not in general a true measure of a physical relaxation time in a mechanical or dielectric thermal sampling (TS) experiment. A method is proposed by which τB may be corrected to obtain an effective relaxation time, τB12, which is the relaxation time computed from the Bucci equation at 50% relaxation: τB12 when correctly calculated, depends only on the temperature of observation. The calculation is based on two assumptions: (i) that the shape of the distribution of relaxation times (slope of the ramp) has been assessed within the experimental error; (ii) that the shift factor is the same for all relaxation times within the narrow packet activated in the TS experiment. It is shown by experiment that the plot of log τB versus T−1 for a viscoelastic polymer (isotactic polypropylene in the β-region) is curved: the curvature is explained theoretically. The plot of log τB12versusT−1 yields a straight line, in agreement with the Arrhenius equation. The early TS experiments, which have been interpreted to favour the compensation rule, should be re-analysed since the published analysis is made by Arrhenius plots using uncorrected values of τB. The high temperature TS experiments on polystyrene and polypropylene (stated to be anomalous by Lacabanne and co-workers in yielding curved log τBversusT−1 plots) will be found, when analysed by the methods described in this paper, to conform to the Arrhenius equation.

The kinetics of the two mechanically dominant relaxations in isotactic polypropylene are studied by thermal sampling (TS). The Bucci relaxation times are corrected by the method of McCrum. The α and β relaxation processes are found to be governed by the compensation rule. For the α relaxation the isokinetic point occurs at τc = 19 μs, Tc = 180°C: the β relaxation isokinetic point occurs at τc = 1.3 ms, Tc = 23°C. These large values of τc are anomalous: they imply that the approach frequencies (τ−1c) are extremely low: 50 kHz and 0.8 kHz for the α and β relaxations respectively. The classical view of the approach frequency gives kT h ∼6 × 10 12 Hz . If this anomally is substantiated for relaxations in other polymers it will require a complete re-examination of current theories of polymer relaxation mechanisms. It is shown also that the temperature dependence of the difference between the limiting compliances, JR-JU, plays a role in thermal sampling and in thermally stimulated creep.

We discuss various temperature jump methods for the accurate determination of the activation energy ΔH of creep for a linear viscoelastic material, and carry out a numerical computation in a discrete relaxation model to show that the effect arising from the temperature dependence of the compliances ( αΔT effect) quickly decays after the temperature jump. A general and compact expression is then derived for the creep rate of a viscoelastic solid under constant stress and subject to arbitrary temperature variations. Application of the expression to various experimental situations is discussed, and the double T -jump method is extended to cases in which a temperature pulse of arbitrary shape is imposed on the sample. A systematic analysis shows that the αΔT effect makes a significant contribution to the expression only during a short period immediately after a T -jump and that it is otherwise totally negligible for any reasonable distribution function in relaxation time. Measurements of ΔH for polypropylene at 40°C by single and double T -jump are in good agreement and support this theoretical prediction. The double T -jump technique is so precise that observations of variation of ΔH with retardation time is now possible.

That all conventional thermoelastic experiments on elastomers are founded on an irrational premiss is a striking claim made recently (Polymer 1984, 25 (Commun.) 213) and one which we now examine in detail using tension as well as torsion techniques. In tension it is shown that κ (the temperature coefficient of r 2 0 ) may be obtained from α (the temperature coefficient of JTR, the relaxed compliance) which may be determined by performing two step-function experiments. In the first a T- jump is performed at constant force F, (T − ΔT)→T, and the extension of the specimen Δl(t) observed as a function of time t after the T-jump: in the second a force-jump, F→(F + ΔF), is performed at T, and Δl(t) observed as a function of time after the F-jump. The derived quantities ( Δl(t) ΔT ) F and ( Δl(t) ΔF ) T at different values of F are related by, it is thus possible to find α if β°, the temperature coefficient of the volume, is known. Plots of ( (Δl(t) ΔT ) F − l( (β° 3 ) against F( Δl(t) ΔF ) T were linear for arbitrary F and arbitrary t. In torsion the analogous equation is in which Δθ(t) is the specimen rotation produced at time t after a T-jump or after a torque jump, Γ→(Γ + ΔΓ). Experiments were performed on a copolymer of acrylonitrite and butadiene (Tg = −7°C). Linearity was observed between ( Δθ(t) ΔT ) Γ and Γ( Δθ(t) ΔΓ ) T . The values of α obtained from tension and torsion agree to within 7%; this error is experimental. It is shown that the conventional analysis due to Shen fails. This failure is due to the use of the Gaussian model and to incorrect treatment of the viscoelastic effects: all existing values of κ and f e f in the literature are systematically in error for the same reasons. The magnitude of the error will depend primarily on the temperature, increasing as T approaches the glass transition.

The mechanism of physical aging in isotactic polypropylene was studied in the α-region (15–80°C) by thermally stimulated creep and by creep superposition. Aging for an elapsed time te was observed to have no effect on retardation elements at τ for τ ⪢ te. Thus the pronounced effects observed in aging cannot be due to the movement of the retardation spectrum as a whole to longer times. The method of time-elapsed time (t−te) superposition fails because the retardation spectrum changes shape. This change in shape leads to systematic deficiencies in the superposed curves, an effect which shows up most clearly in superposition of the differentiated creep curves. The possible changes that take place during aging are a change in magnitude of the retardation elements and a shift to longer times. These changes will only occur in retardation elements at τ when τ ≲ te.

The theory of a new method for measuring the temperature dependence of the limiting compliances, JR and JU, is described. The method is based on the observation of secondary creep; this is a small effect generated in a stressed specimen by a T-jump. In the experiment, the specimen is maintained isothermally under a constant stress until the primary creep rate is negligible. A T-jump of several degrees, positive or negative, is then imposed and this generates the secondary creep. If JR and JU are independent of temperature there is no secondary creep. The secondary creep has the same time dependence as primary creep. A comparison of the magnitudes of secondary and primary creep — at equivalent reduced times — leads by means of the theory of thermoviscoelasticity to the determination of α and β, the temperature coefficients of (JR-JU) and JU. For isotactic polypropylene in the α-region at 40°C both coefficients are positive: α = 0.7 × 10−2 and β = 0.1 × 10−2°C−1. It is clear from the positive sign for the coefficient of (JR-JU) that the α-relaxation is of the type in which stress increases the entropy. The magnitudes of the coefficients support the conclusion that the use of un-normalized time-temperature superposition to determine activation energy in polypropylene will be highly erroneous. The magnitude of the coefficient of (JR-JU) is sufficiently large to account partially for physical ageing in crystalline polymers. Quenching from T down to T0 will freeze in a non-equilibrium value of (JR-JU), considerably in excess of the equilibrium value at T0. Storage at T0 permits the equilibrium value to be slowly attained, so that with increasing ageing time the polymer becomes stiffer and exhibits lower creep rates.

New stereospecific polymerization systems for methyl methacrylate are reported. The temperature dependence of the internal friction of the stereoregular forms of polymethyl methacrylate (PMMA) has been determined at frequencies of ca. 1 cycle/sec. Isotactic PMMA does not exhibit a side chain relaxation and its main chain relaxation is 66°C. lower than that of conventional PMMA. Syndiotactic PMMA exhibits exactly the same relaxations as conventional PMMA. Since the density and infrared absorption spectra of syndiotactic and conventional PMMA are also comparable and differ from the isotactic form, it is concluded that conventional radical-initiated polymer is not truly atactic but has considerable syndiotactic character.

The relative merits of time-temperature superposition procedures (those of Leaderman, Tobolsky, Ferry, Kě, McCrum, and Morris) have been assessed for the glass-rubber (a) relaxation of an epoxy resin. Measurements of creep compliance in shear were made at temperatures between 24° and 190°C. The different superposition procedures yield shift factor versus temperature plots which are essentially identical. Differences between the procedures are greatest in the master curves particularly at short and at long times. The McCrum-Morris procedure alone yielded a smooth master curve at all times. At temperatures above the glass transition the shift factor versus temperature plots followed the WLF equation.

Thermal expansion of linear polyethylene in both macroscopically isotropic and single-crystal texture forms was measured between 0 and −190° C. Results are interpreted by treating the semicrystalline polymer as a simple two-phase composite material. The single-crystal texture specimen was prepared by drawing with constant width at 121° C followed by annealing, giving unique preferred orientations of each of the three crystal axesa, b andc (c parallel to the draw directionZ, a parallel to the sheet normalX andb parallel toY, the normal to the thin edge of the sheet). The thermal expansions parallel toX, Y andZ follow approximately the thermal expansions ofa, b andc vectors of the unit cell as determined by X-ray diffraction. An attempt is made to allow for the contribution to thermal expansion from amorphous regions, representing the oriented solid by a simple model in which crystal and amorphous regions are coupled in series in the draw direction and in parallel in the two perpendicular directions. This analysis requires the thermal expansion of amorphous polyethylene which was obtained by extrapolation of thermal expansion measurements on ten isotropic specimens whose crystal volume fractions lay between 0.81 and 0.49. It was observed that the single crystal texture specimen of crystal fraction 0.80 and an isotropic specimen of the same crystal fraction on cooling changed in volume by equal amounts. This fact together with the observation that the thermal expansion parallel toY of the single crystal texture specimen is equal to that of the isotropic specimen leads to important conclusions concerning the existence of microstresses in linear polyethylene caused by a change of temperature.

Alternative theories for the origin of the high temperature,α-relaxation in oriented linear polyethylene are compared with experiment. Theα-mechanism is assumed to be simple shear by all theories which differ in the allocation of the shear plane. The theories take the shear plane to be the interlamellar surface (theory I), a crystal plane of type (hk0) (theory IIA), a crystal plane of type (h00) (theory IIB). The experiments comprise measurements of the creep tensile compliance D(t) on oriented specimens at angles 0°, 45° and 90° to the draw axis. The crucial test involves the examination of specimens with approximately the same crystallographic orientation (determined by wide-angle X-ray diffraction) but differing orientation of lamellar normals. The analysis is based on the Reuss, constant stress hypothesis. It is shown that with this assumption, only theory I agrees with experiment.

A theory is proposed which enables activation energies for creep (Δ H c ), and stress relaxation (Δ H s,r ) to be calculated when the limiting moduli vary with temperature. The theory explains the agreement obtained by Kê between Δ H c , Δ H s.r. and the activation energy for internal friction Δ H i, f. for the grain boundary relaxation in aluminium . The theory rectifies the existing discrepancy between Δ H c (or Δ H s.r. ) for the β -relaxation in polymethyl methacrylate ( PMMA ) and Δ H i.f. . The discrepancy is shown to be due to the assumption that G U and G R , the unrelaxed and relaxed moduli, do not vary with temperature. It is eliminated by assuming that both limiting moduli have the same finite temperature dependence. This temperature dependence is obtained from measurements of the real component of the shear modulus G '(ω) at 3·3 x 10 4 c/s on the assumption that in the β -region ( — 50 to 20 °C) and at this frequency the temperature dependencies of G '(ω) and G U are the same. It is probable that the many previous determinations of Δ H c and Δ H s.r. for rigid polymers exceed the true values by amounts of the order of 70%, the discrepancy in PMMA . The theory is also used to determine Δ H s.r. for a Snoek relaxation for which G U decreases and G R increases with increasing temperature. The result obtained for oxygen in tantalum is Δ H s.r. = 27 kcal/mole in agreement with Δ H i.f. .

A torsion pendulum has been used to measure the internal friction of polytetrafluoroethylene. The crystallinity of the specimens varied form 48 to 92%, and the measurements were made over a wide temperature range. The site of each relaxation was assigned by observing the variation of relaxation strength and peak height with crystallinity. In this way amorphous relaxations were found to occur at 176.5°K. (0.60 cycle/sec.) and 399°K. (0.62 cycle/sec.). The crystallite relaxations occur at 291°K. (0.28 cycle/sec.) and the crystallite melting point, 600°K. It is concluded that the amorphous regions of polytetrafluoroethylene exist in three phases between 4.2°K. and the crystallite melting point. The supercooled liquid which exists below 600°K. is transformed into an amorphous solid at a temperature in the region of 400°K. This solid undergoes another transformation at approximately 176°K. to form a third phase which is stable down to 4.2°K. The crystallite realaxation centered at 291°K. consists of a series of disordering transitions including the well-known first-order transition. There is evidence that hindered rotation occurs within the crystallites at temperatures as low as 220°K.

Mechanical relaxation in polyethylene specimens prepared in various ways has been studied at 0*67 c/s and by creep experiments of 35 min duration in order to elucidate the mechanisms of the α' and α' relaxations. Internal friction measurements were made from 20 to 120 °C and creep measurements from 14 to 85 °C. The α' relaxation has five significant properties. (1) The magnitude of the α' relaxation is highly sensitive to anisotropy in the orientation of lamellae. (2) The α' retardation time can be increased by irradiation which cross links the molecules in between lamellae. (3) The α' retardation time is greater in a slowly crystallized specimen with coarse lamellae than in a quickly crystallized specimen with fine lamellae. (4) The α' creep is entirely recoverable. (5) The superposed creep curves form not one master curve but two. These properties are predicted by a model in which the cl' relaxation is due to slip at lamellar boundaries while the a relaxation occurs within the lamellae. The model predicts full recoverability of the relaxation and the retardation time of the α' relaxation to be a function of the compliance of the lamellae, the viscosity of the inter-lamellar material and the geometry of the lamellae. The experiments also proved that large variations in the spherulitic structure of polyethylene have no measurable effect on any of the properties of the cl relaxation. The α relaxation cannot therefore be due to slip at the spherulite boundary.

The temperature variation of the internal friction of tetrafluoroethylene-hexafluore propylene (TFE-HFP) copolymers was studied using a torsion pendulum at frequencies ca. 1 cps. The compositions varied from 0 to 14 mole% HFP. Increasing HFP content was found to, 1 depress the temperature of the crystal disordering transition. 2 increase the size and depress the temperature of the glass I peak 400° to 348°K. 3 decrease the size of the glass II peak. The temperature remined constant at 177°K. The glass I and II peaks were assigned to relaxation mechanisms involving long side branched and linear segments of the molecule and short linear segments, respectively. The depression of the crystal disordering transition was attributed to a decrease in the crystal lite size with increasing HFP content. An analogous glass I peak was observed in high density polyethylene at 333°K. in addition to the previously reported glass II peak at 156°K. Polypropylene exhibits glass I and II Peaks at 273° and 170°K. and two smaller peaks at 53° and 32°K due to hindered rotation of the methyl groups. Further work is required to elucidate the nature of the motions of the molecular backbone responsible for the glass I and II and crystal disordering peaks in the above Polymers. Die Temperaturabhangigkeit der inneren Reibung von Tetrafluorathylen-Hexafluor propylen-Copolymeren (TFE-HFP) wurde mittels eines Torsionspendels bei Frequenzen von ca. 1 Hz untersucht. Ihre Zusammensetzung wurde von 0 bis 14 Mol-% HFP variiert. Es wurde gefunden, das mit zunehmendem Anteil von HFP 1 die Temperature der Unordnungs-Umwandlung im Kristall herabgesetzt wird, 2 die Hohe des I. amorphen Maximum zunimmt und die zugehorige Temperature von 400° auf 348°K abnimmt, 3 die Hohe des II. amorphen Maximum abnimmt, wahrend die zugehorige Temperature bei 177°K konstant bleibt. Als Ursache des I. und II. amorphen Maximum wurden Relaxationsmechanismen angenommen, die den langen Seitenketten und linearen Molekulsegmenten bzw. den kurzen linearen Segmenten zuzuordnen sind. Die Erniedrigung der Temperatur, bei der die Unordnungs-Umwandlung im Kristall stattfindet, wurde als eine Folge der mit zunehmendem Gehalt an HFP abnehmenden Kristallitgrose angenommen. Bei Niederdruck-Polyathylen wurde ein analoges I. amorphes Maximum bei 333°K gefunden, zusatzlich zu einem 11. amorphen Maximum bei 156°K, uber das fruher schon berichtet wurde. Polypropylen zeigt ein I. und 11. amorphes Maximum bei 273° und 170°K sowie zwei kleinere Maxima bei 53° und 32°K, die durch die gehemmte Rotation der Methylgruppen verursacht werden. Weitere Arbeiten sind erforderlich, um die Art der Bewegung des Molekulskeletts aufzuklaren, die fur den I. und II. Glaspunkt und fur die Unordnungs-umwandlung in den betrachteten Polymeren verantwortlich ist.

The orientations produced by high temperature rolling of polyoxymethylene (126 ‡C) and polytetrafluoroethylene (150 ‡C) and nylon 66 are examined by pole figures. No evidence is found to support the theory of Akahane and Mochizuki that it is the plane of the zig-zag chains that orientates into the rolling plane in nylon 66. The classical rolling texture of Bunn and Garner describes the pole figures except for the (100) pole, which is observed at 28 ‡ to the transverse direction; in the texture of Bunn and Garner it should occur at 24 ‡. In hot-rolled polyoxymethylene the texture observed corresponds to a unique orientation of the hexagonal unit cell: (10¯10) planes parallel to the rolling plane and c-axis parallel to rolling direction. The polyoxymethylene texture differs significantly from the pseudo-fibre textures observed in the cold-rolled polymer. The rolling texture of hot-rolled polytetrafluoroethylene is similar to that of hot-rolled polyoxymethylene. These results show the plastic slip system in polyoxymethylene and polytetrafluoroethylene to be (10¯10) [0001].

Mechanical relaxation was studied in bulk polyoxymethylene by means of a torsion pendulum. The magnitude of the large γ peak at 203°K. (for f = 1.1 cycles/sec.) decreases with increasing density. The magnitudes of the β peak at 260°K, and the α peak at 400°K. do not depend on density. The β peak is small unless the specimen contains water or has just been cooled from a temperature above ca. 370°K. The activation enthalpies of the α and γ relaxations are 24 and 27 keal./mole, respectively. The relaxation time distribution and activation enthalpy of the γ relaxation do not depend on density.

Mechanical relaxation has been studied at 0.67 cps in linear polyethylene (LPE) and polytetrafluoroethylene (PTFE) between −190 and −20°C. Specimens were prepared by use of various thermal treatments to produce in LPE a range of crystalline fractions from 0.690 to 0.825 and in PTFE from 0.615 to 0.870. An empirical theory is proposed relating the modulus of the crystalline–amorphous composite solid to the moduli and the volume fractions of the two phases. The empirical theory is shown to be in accord with the bounds of Hill and of Hashin and Shtrikman. The theory is used to determine the magnitudes of the crystalline and amorphous components of the low temperature relaxations in LPE and PTFE from measurements of logarithmic decrement and shear modulus. In PTFE the γ relaxation occurs in the amorphous fraction alone. In LPE the γ relaxation is a composite one, formed from the superposition of a small crystal relaxation and a large amorphous relaxation. For the crystal relaxation in LPE the ratio of relaxed to unrelaxed modulus equals 0.78; for the amorphous relaxation, the ratio equals 0.23. In a specimen of LPE with crystal fraction 0.69 the crystal contribution to the relaxation is 25% of the total. The magnitude of the unrelaxed modulus of the crystal fraction of LPE (modulus of polycrystalline LPE at −190°C) is in reasonable agreement with theoretical calculations of Odajima and Maeda.

Values of ΔH and τ ∞ for the mechanical relaxation times centred at ∼ 700 s were obtained for isotactic polypropylene at three temperatures in the α-region by double T -jump. The values of ΔH and τ ∞ varied systematically in accordance with the compensation rule. The compensation temperature was observed to be T c ∼240°C. It is shown that the compensation rule is in good agreement with the predictions of theories of Wert and Zener and of Eby.

IF the temperature of a linear viscoelastic solid creeping under a constant stress is changed abruptly from T0 to T then the ratio of the instantaneous creep rates at the time of the T-jump is illustration in which illustration β is a parameter which depends on the viscoelastic properties of the solid1, κ a parameter which accounts for the perturbation due to thermoelastic stresses generated by the T-jump2 and ΔH the activation energy for creep. If β=κ=1, equation (1) reduces to the Dorn equation3, illustration so that ΔH may be determined from a measurement of eT0 and eT. Here we show how ΔH may be determined when κ=1 but β takes values greater or less than unity depending on the direction of the T-jump. Open image in new window Open image in new window Open image in new window

The creep rate of polyethylene following an imposed temperature change is observed to exceed the rate predicted theoretically. The enhancement of creep rate is attributed to thermoelastic stresses generated by the anisotropic thermal expansion of the polymer crystals.

THE purpose of the experiment reported here was to examine the striking observation of Rigby et al.1 that for collagenous tissue the thermally activated process controlling mechanical relaxation changes abruptly at the body temperature. For rat tail tendon it was observed1 that below 37° C the process is entropy activated: the activation energy ΔH=0. Above 37° C, Rigby et al.1 conclude that ΔH is large although their experiment does not yield a quantitative result. The result is paradoxical in two ways. First, entropy activation, although feasible conceptually, is essentially unknown in nature. Second, is it plausible that the rate process should change in such a profound way right at the body temperature? We have now resolved both paradoxes.

Since Hoeve and Flory (1,2) corrected the earlier work of Meyer and Ferri and Wolisch et al. it has been deduced by all later workers (apart from Weis-Fogh and Anderson) (3) that the mechanism of mechanical deformation in elastin is essentially entropy elastic. [If the force f is written f = fe + fs (the sum of the energy and entropy components fe and fs) then this implies fe/f << 1.] The purpose of this paper is to examine critically the methods used in making this deduction. This will lead to a discussion of the reliability of mechanical experiments in understanding the nature of the molecular conformation of elastin in vivo.

The thermoelastic behavior of water solvated elastin has been investigated in simple tension, in the temperature range 0–70°C. Specimens purified from both the ox ligamentum nuchae and pig thoracic aorta have been studied. Force data obtained by cycling the temperature for various constant specimen lengths display a separated variable dependence of the form f = A(T)B(α), where T is absolute temperature and α the extension ratio. For ligament elastin B(α) is a linear function whereas for aortic elastin it is a nonlinear function. The applicability of the rubber elasticity theory to elastin has been tested by setting A(T) equal to the temperature-dependent front factor for simple tension of a homogeneous rubber whilst B(α) is left undefined. In this way it has been possible to take into account the fibrous nonhomogeneity of the polymer, and also to avoid any inconsistency within the theory of attributing a dependence of the variable fe/f upon extension ratio. The behavior of both ligament and aortic elastin agrees well with the conclusion that the dominant deformation mechanism is entropy elastic, fe/f ≪ 1. The linearity of the load isotherm for ligament elastin permits a particularly simple experimental procedure using a single force-temperature plot for one value of interclamp length. Using this procedure high precision has been obtainble, and the data shows a close adherence to the theory with fe/f = 0.1. The relationship between this result and current controversy over the molecular conformation of elastin is discussed.