INTRODUCTION: Cementless implants achieve long-term fixation and consequent clinical stability primarily via biologic means – bony ingrowth into a porous-coated intramedullary implant surface. The adequacy of this biologic fixation depends upon numerous factors, including the initial or short-term fixation of the implant with respect to the adjacent bone (1,5,6). Short-term fixation refers to the post-operative limitation of relative motion between implant surface and the adjacent bone. Micromotion may be limited by utilizing a rough implant coating in concert with a distal stem press-fit. However, the holding power of the press-fit over time is dependent upon the viscoelastic nature of cortical bone (2). The objective of this study was to experimentally evaluate the effect of transverse cortical bone viscoelasticity on initial diaphyseal fixation of an implant for various degrees of press-fit (diametral interference ( )) and surface coating. It was hypothesized that viscoelastic relaxation increases with press-fit amount. It was also hypothesized that if bone was stressed above a certain level during press-fit, implant fixation would be reduced. METHODS: Eleven femoral pairs were utilized in this study (females 63.1 3.9, males 53.6 8.9). The target cortical index range, as calculated from anterior-posterior x-ray films was 40-60%, which was also representative of the hip replacement population (3,4). The right and left femurs were cut into three 4 cm long diaphyseal bone sections that were reamed until “chatter” (resistance to reaming from cortical bone) was felt. After reaming, a cylindrical CoCrMo alloy intramedullary test specimen (DePuy, A Johnson and Johnson Company) was selected that corresponded to line-to-line (same size stem as ream), 0.5 mm press-fit (a stem with a diameter 0.5 mm larger than the reamer) and a 1.0 mm pressfit (a stem with a diameter of 1.0 mm larger than the reamer). The actual press-fit was then calculated form the difference between the reamer diameter and the measured stem diameter. Three stem surface conditions were manufactured including 20 grit, small bead coating (60/80 mesh Porocoat porous coating) and large bead coating (24/35 two layer coating). With the bone held in place on an annular loading platform, the test specimens were inserted into the bone segments using standard operating room tools and procedures. Once inserted, the loading platform was placed between the cross-heads of a servo-mechanical testing machine (Materials Testing System, Minneapolis, MN) and loaded under stroke control at 1 mm/min. Actuator load and deflection were recorded and plotted for determining the push-out load. The push-out load was defined as the maximum load attained prior to slipping of the stem within the canal. The push-out load was determined for specimens at time=0 hours (Pt=0) and after 24 hours (Pt=24) of soaking in a physiological bath. From these two quantities, the viscoelastic Load Radio was found as Load Ratio = P0/P24. The Load Ratio was a measure of the viscoelastic effect on the “holding strength” of the stems. Statistical analysis was performed using JMP (SAS Institute, Cary, NC) and significance was set at p<0.05. RESULTS: Results indicated that there was no significant relationship between push-out loads and press-fit at t=0 and t=24 hours (Figure 1). However, a nonlinear regression analysis of the pooled Load Ratio versus press-fit data indicated a significant (p<0.0002, r = 0.43) quadratic relationship (Figure 2). Therefore, more relaxation occurred between t=0 and t=24 hours resulting in less push-out load as press-fit increased. The analysis also indicated that small bead coated (i.e. Porocoat) stems have significantly greater push-out loads at t=0 hours (p<0.01) and t=24 hours (p<0.04) compared to large bead coated stems. The small bead coated stems also demonstrated greater push-out loads than the 20 grit stems but the difference did not reach statistical significance. DISCUSSION: Results from experiments showed that relaxation significantly increased with press-fit, i.e. there was a greater reduction in push-out loads at higher press-fit as hypothesized. It was found that stability of the stem was not related to press-fit amount, indicating that after achieving press-fit, no additional gains or losses in stability were obtained by increasing press-fit, contradicting our second hypothesis. In finite element analysis conducted in our lab, results suggest an implantbone interference fit threshold ( =0.10mm) above which there would be no advantage in short-term push-out strength. This is indicative of an interference “threshold;” that is, a value of interference above which no significant benefit in terms of holding power is obtained. The results of this study showed that cortical bone viscoelasticity in the transverse plane of the diaphyseal femur has a diminishing effect on the contact pressure between the endosteal bone surface and the press-fit implant. This effect becomes significantly more pronounced as press-fit increases. However, there was no significant relationship between press-fit amount and pushout load. Therefore, for a particular implant size, there is a degree of press-fit beyond which no additional gains or losses in short-term fixation are realized. Additional study into this phenomena is warranted in view of the clinical implications. ACKNOWLEDGMENT: This project was supported by funding from DePuy, A Johnson & Johnson Company. REFERENCES: (1) Bauer TW, Schils J. Skeletal Radiol 28:483-97, 1999. (2) Blaha JD. In The Adult Hip, pp. 1085-1092, 1998. (3) Gruen TA. Acta Orthop Belg 63:20-7, 1997. (4) Gruen TA et al., 23 Annual Meeting of American Society of Biomechanics, 1999. (5) Pilliar RM et al., Clin Orthop 208:108-13, 1986. (6) Rosenberg A. Orthopedics 12:1223-33, 1989. Listing for additional author affiliation **DePuy, A Johnson & Johnson Company, Warsaw, IN ***Zonal Concepts, Wesley Chapel, FL ****University of Rochester, Rochester, NY *****University of Michigan, Ann Arbor, MI

The yield force (or stress) of polycarbonate in simple extension was studied to determine its dependence on deformation history, including the stretching rate, and temperature. At different temperatures, the stretching of a specimen at a constant rate was interrupted both before yielding and during cold drawing to allow the occurrence of stress relaxation and the accompanying physical ageing (hardening). When a test was resumed, progressive erasure of ageing occurred, leading to a yield force whose value compared to that obtained by continuous stretching can be explained qualitatively in terms of the rate of physical ageing, determined earlier at an extension of 2.6% over a wide temperature range. In addition, the present data support the conclusion reached previously that physical ageing occurs during stress relaxation at a fixed deformation. Also determined were the yield and drawing stresses and the yield strain from −90 to 100°C.

The yield force (or stress) of polycarbonate in simple extension was studied to determine its dependence on deformation history, including the stretching rate, and temperature. At different temperatures, the stretching of a specimen at a constant rate was interrupted both before yielding and during cold drawing to allow the occurrence of stress relaxation and the accompanying physical ageing (hardening). When a test was resumed, progressive erasure of ageing occurred, leading to a yield force whose value compared to that obtained by continuous stretching can be explained qualitatively in terms of the rate of physical ageing, determined earlier at an extension of 2.6% over a wide temperature range. In addition, the present data support the conclusion reached previously that physical ageing occurs during stress relaxation at a fixed deformation. Also determined were the yield and drawing stresses and the yield strain from -90 to 100-degrees-C.

Studies were made of the physical properties of the commercially available polyimide Upilex-SGA, which is prepared from biphenyl dianhydride and p-phenylene diamine. Annealing the Upilex-SGA for 2 hr Linder N2 at 400°C gave a film that expanded continuously when heated at a fixed rate, in contrast to the as-received film. The linear expansion showed a change of slope at 84°C and also at 295°C, the later being Tg. The thermal coefficient of linear expansion at all temperatures was very smap, even above 295°C it is 27.8 x 10−6. Its stress-strain curve did not exhibit a yield point, even though its ultimate elongation is ~23%. Similar behavior is shown by the PMDA-ODA polyimide, except its ultimate elongation is ~70’%,. The unusual stress- strain curves exhibited by these polyimides is undoubtedly caused by their liquid-crystalline morphology. The stress-relaxation modulus was measured at 0.5% extension and 12 temperatures from 30 to 330°C. Derived isochrones showed that the 1-s tensile modulus at 20°C is 9.0 GPa, but at 330°C it is 2.0 GPa. Creep curves were also measured at a stress of 30 MPa and at 10 temperatures from 30 to 340°C. Master curves prepared from the relaxation and creep data are discussed briefly and evidence is given which, show that the superposition method is not truly valid for this polyimide, which actually is not surprising.

Studies are discussed which show that the segmental mobility In polycarbonate (Lexan from General Electric Co.) increases when a static deformation in either extension or simple compression (uniaxial stress) is applied to a specimen, but thereafter the mobility decreases progressively. These changes can be termed erasure (partial) of physical aging (strain softening) and physical aging (hardening), respectively. Because the volume of a specimen deformed in simple compression decreases, reversal of aging (de-aging) cannot be attributed to an increase in free volume, as usually defined. A proposed explanation will be mentioned. A study was also made of the yield stress modified by unconventional methods at several temperatures. The results can be explained In terms of the rate of physical aging and its temperature dependence determined on specimens at 2.6% extension over a broad range of temperatures. These results show that yielding results from progressive de-aging. The yield phenomena obtained by special methods results from both physical aging and de-aging.

ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTTime dependence of the storage modulus of polycarbonate following temperature jumps within the glassy stateBassel Haidar and Thor L. SmithCite this: Macromolecules 1990, 23, 15, 3710–3712Publication Date (Print):July 1, 1990Publication History Published online1 May 2002Published inissue 1 July 1990https://doi.org/10.1021/ma00217a029RIGHTS & PERMISSIONSArticle Views116Altmetric-Citations6LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InReddit PDF (394 KB) Get e-Alertsclose Get e-Alerts

AbstractA study was made of the physical aging of an annealed polycarbonate film at a constant extension of 2.6% at 5 temperatures from 30 to 110°C. During stress relaxation at the constant extension, the storage modulus in tension, E′, was determined by imposing a sinusoidal strain of small amplitude at frequencies up to 25 Hz during an aging period, commonly of 5 h. Plots of log E′ against log f, where f is the frequency, gave parallel straight lines, each at a different aging (elapsed) time te. These lines were superposed by shifts along the abscissa. The obtained shift factors log a varied linearly with log te, the slope being the aging rate μ, a quantity introduced by Struik. The results show that μ is about 1.37 at 30°C and that it increases progressively with the temperature and becomes approximately 2.13 at 110°C. Another measure of the aging rate is the rate of increase of E′ with te, expressed as the percent increase per decade of the aging time. This quantity also increases progressively with the temperature from about 2.6% at 30°C to nearly 3.5% per decade of time at 110°C.

AbstractA novel method has been developed and tested to determine accurately the linear coefficient of thermal expansion α of flexible films without subjecting a specimen to a tensile load continuously during the measurement of expansion data. Other methods have invariably required the application of a tensile load which commonly leads to creep during the experimental time. The new method involves mounting an Invar jig in the lower grip in a Dynastat. The closed‐loop servo control in the Dynastat enables the length of a specimen to be determined at a series of decreasing or increasing temperatures without subjecting the specimen to a tensile load, except momentarily on occasion while length data are being determined. Otherwise, a specimen is subjected to a small bending force. To demonstrate that the method gives valid results, α for an aluminum foil was determined and found to agree exactly with literature data. Expansion coefficients and values of the glass transition temperature (Tg) were determined on FEP and PFA Teflon films and also on a commercially available polyimide film, Upilex‐SGA.

Differential storage and loss tensile moduli, E′ and E″, were determined intermittently at 10 Hz on specimens of an annealed polycarbonate film during stress relaxation at static tensile strains from 1.2 to 6.25% at 50°C. It was found that E′ and 1E″ decrease when a static strain is applied but thereafter they increase progressively with time. These changes, which increase with the applied strain until it becomes 4%, are attributed primarily to a rejuvenation of a specimen (an increase in segmental mobility) followed by physical ageing (a progressive decrease in segmental mobility). Measurements at a static strain of 3% at six temperatures from 30°C to 130°C showed, among other things, that the rates of increase of E′ and decrease of E″ are sensibly independent of temperature up to 110°C.

An exploratory study has been undertaken to determine the effect of simple tensile deformations, applied stepwise, on the permeability and diffusion coefficients ( P and D ) of gases in glassy polymers. For nitrogen in a polycarbonate film at 73°C, it was found that ( 1 P ) d P d ϵ ⋍ 13 and ( 1 D ) d D d ϵ ⋍ 5 to 8, where ϵ is the strain. Also, P and D were found to decrease with time at constant strain and to depend somewhat on strain and thermal histories. A calculation of ( 1 D ) d D d ϵ , based primarily on considerations of free volume and its strain dependence, gives 11.8 or 8.5, depending on the assumptions made. Progressive stepwise application of strain to a polyimide (Kapton) film at 72° and 125°C caused P and D for carbon dioxide to increase. But above 2% strain at 125°C, they decreased until at 6% strain they became smaller than those for the undeformed specimen, a reflection of a straininduced relaxation process.

AbstractThe permeability and diffusion coefficients (P and D) for gases in a biaxially oriented polystyrene film have been found to increase when a specimen is stretched in simple tension and to decrease with time when the strain is held constant. These effects are attributed, respectively, to an increase in free volume with strain and to the continuous volume recovery (densification) at constant strain. The strain dependence, at small strains, of P and D for Ar, Kr, N2, CO2, and Xe at 1 atm pressure and 50°C indicates that the size distribution of free‐volume elements is not distorted when a specimen is stretched. At a constant strain of 1.8 percent at 50°C, P and D for xenon decrease about 13.8 and 11.8 percent, respectively, per decade of time—two to threefold faster than for carbon dioxide. These results and those obtained with argon, whose molecular diameter is significantly smaller than that for xenon, suggest that the larger free‐volume elements decrease in size faster than the smaller ones as volume recovery progresses.

The viscosity of flocculated suspensions of rhombohedral and acicular (needle-like) αFe2O3 (nonmagnetic) and acicular γFe2O3 (magnetic) particles in ethylene glycol was measured over broad ranges of concentration at shear rates D from a few sec−1 up to 1340 sec−1. The intrinsic viscosity, [η], derived from the concentration dependence of the viscosity, depends linearly on D−γ, where γ is rm12 and rm13 for suspensions of the acicular and rhombohedral particles, respectively. Intrinsic viscosities are discussed in terms of φpf = 2.5/[η], where φpf is the volume fraction of particles in a floc. At a D of 5.3 sec−1, [η] is about 30 for suspensions of the nonmagnetic particles and is 80 for the magnetic particles. The intrinsic viscosity at infinite shear rate, which increases from 3.3 to 6.0 with the axis ratio of the particles, indicates that small stable flocs exist. From [η] at 5.3 sec−1 and results from a theoretical treatment of floc growth by Vold, the floc size and number of particles in a floc were estimated and are compared with results from sedimentation rates. The dependence of shear stress on D conforms to the Casson equation, which gives the yield stress and viscosity at infinite shear rate. Data are also presented on xylene suspensions of acicular αFe2O3 particles stabilized with a polymeric dispersant.

AbstractThe stress response σ(t) to a constant rate of strain $ \dot \varepsilon $ ε during the period 0 < t ≤ t* and to the constant strain ε* $ ( = \dot \varepsilon t*) $ thereafter is considered in terms of the Boltzmann superposition principle. When t ≤ t*, the data directly give the constant‐rate modulus F (t) ≡ σ(t)/ε(t), which can be converted straightforwardly into the relaxation modulus E(t). Results from illustrative calculations show that a reduction in the relaxation rate effects a decrease in [σ(t*)/ε*]/E(t*) and also in the time at which [σ(t)/ε*]/E(t) becomes essentially unity. To evaluate E(t) at t > t*, F(t) is first obtained from σ(t) and F(t − t*) by using a derived equation similar to that presented by Meissner. Thereafter, F(t) is transformed into E(t). For illustration, E(t) for a rubbery solid is evaluated over some 2.5 decades of time from its response to a strain rate of 0.25 min−1 for 0.40 min and thereafter to the attained strain of 0.10 for 5.4 min.

ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTElectric Birefringence of Dilute Suspensions of Poly(ethylene oxide) Crystals in EthylbenzeneUlrich Leute and Thor L. SmithCite this: Macromolecules 1978, 11, 4, 707–715Publication Date (Print):July 1, 1978Publication History Published online1 May 2002Published inissue 1 July 1978https://pubs.acs.org/doi/10.1021/ma60064a018https://doi.org/10.1021/ma60064a018research-articleACS PublicationsRequest reuse permissionsArticle Views58Altmetric-Citations7LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InRedditEmail Other access optionsGet e-Alertsclose Get e-Alerts