Printer Friendly

Anisotropic Loss of Toughness with Physical Aging of Work Toughened Polycarbonate.

We have studied the response of mechanically toughened and physically aged polycarbonate primarily using Charpy impact and ultrasonic wave speed measurements. The toughening was conducted through plastic compression on as-received PC. The Charpy impact tests showed anisotropic toughening, both in the absorbed energy and in the mode of fracture. The amount of toughening with plastic compression, even though anisotropic, is centered around the response of annealed and quenched samples, which represent the response of an unaged PC. There was an anisotropic drop in the toughness of some samples with aging. The time of this drop was uncorrelated in the different directions and disappeared for the highly toughened samples. This transition was bimodal and statistically distributed between either a fully ductile or a fully brittle failure. As the samples were prepared in a manner to remove induced residual stresses, this drop in toughening may be associated with an intrinsic anisotropic thermal aging of the deformed material. POLYM. ENG. SCI., 54:794-804. 2014. [c]2013 Society of Plastics Engineers

INTRODUCTION

In a series of articles. Broutman and coworkers (1-3) have studied the effects of thennal treatment, cold working, and annealing on the anisotropic toughness of polycarbonate (PC). These studies have focused on the role of residual stresses, induced by cold working, on the changing of the toughness. In a set of new results, we study the effects of cold working ami thermal aging on the anisotropic nature of plastically deformed samples and show that, even wilh the removal of boundary effects, plastic How produces large anisotropic toughening that also thermally ages anisotropically. These results suggest that ihe induced changes mighl be associated with the intimate structure of the material, rather than due to process induced residual stresses.

The extent of anisotropy induced by plastic flow in polymers has been well documented. For example, the elastic anisotropy induced by extension in glassy polymers is documented in Ward (4) for quasi-static deformation and by Goel et al. (5) for ultrasonic wave moduli. In addition, the associated anisotropy induced by plastic How in thermal conduction is reported by Kuraha\ashi (6). While these changes are substantial, nonnally in ihe same order as the induced plastic strain, ihe change in toughness is one to two orders of magnitude larger (1-3), and much more sensitive to physical aging (7-9). In particular, the experimental results indicate thai the aging of physical!; toughened PC is anisotropic. Thai is. experimental coupons cut from the same material, but along different orientations to the plastic deformation, show ductile lo brittle transitions that are orders of magnitude different in their transition time, indicating anisotropy in the thermal aging. Even though Broutman and coworkers (1-3) have studied the development of anisotropy in the toughness and report some of the influences of aging, they attribute these changes to the observed development of residual stresses in their samples, which due lo the method of sample preparation should not exist in Ihe current experimental results.

The current study focuses on the effects of plastic flow anil physical aging of PC" on the dissipated energy during fracture as measured in Charpy tests. PC. a lough transparent polymer, is primarily used for structural and protective applications. As it is one of the glassy polymers in high production, PC has been studied extensively, and its mechanical and thennal response has been of particular interest, particularly in regard to its toughness. An extensive study of mechanical response of PC was done by Boyce et at. (10), the evaluation of the equilibrium stress after plastic defonnation by Goel et al. (11). the influence of plastic How on dynamic mechanical properties by Koman el al. (12) and Liu et al. (13), on conductivity by Kurabayashi (6). and on toughness by Broutman and coworkers (1-3), Numerous other studies by these and other authors, have further studied the mechanical response of PC (Refs. (14. 15)).

Physical aging (16-19). which is the effect of holding a glassy system below its glass-transition temperature, has a substantial influence on the response of polymers, particularly in regard to their fracture. In general, this effect is a manifestation of the non-equilibrium stale of a glassy polymer and its gradual transition toward equilibrium. When cooled through the glass-transition, due to restricted mobility in the glass; state, amorphous polymers gel trapped in a non-equilibrium expanded state, which with lime transitions toward the denser equilibrium state. This transition is associated with a reduction in the free volume of the glass, and results in an increase in the yield stress (7-9), and a transition from ductile to brittle fracture (20-22). The influence of thennal aging in glassy systems is. for the most part, fully recoverable through healing to above the glass-transition temperature (23, 24). normally lenned rejuvenation. Similar effects to rejuvenation can also be induced though plastically deforming a glassy system (25). This deformation induced rejuvenation is sometimes termed work toughening of the glass, as it seemingly removes embriitlemenl resulting from physical aging.

Work toughened PC shows distinctively anisotropic fracture properties (2), and, as will be shown, with a distinctive anisotropic thermal aging characteristic. In particular, the time it takes to transition from ductile to brittle fracture is different for Charpy samples taken along different orientations relative to the plastic defonnation. As the energ; dissipated in the ductile stale is 5-15 limes larger than that dissipated in the brittle state, the lime it lakes for a work toughened PC to lose this toughness can be an important factor. In addition, the results indicate that the pronounced anisotropy seen in the work toughened sample disappears once all transitions associated with all directions have been completed. To the knowledge of the authors, this is the first lime experimental results have been presented that show this anisotropy in the ductile to brittle transition time.

MATERIALS AND METHODS

Polycarbonate (Lexan 9034) sheet stock (0.5 inch thick 4 ft * 8 ft) was used for both the Charpy impact and ultrasonic measurements, while cylindrical bar stock (1 inch diameter) was used for the DSC tests. The ultrasonic tests were done directly on the Charp) samples.

To prepare the samples, coupons were cut from the PC stock and then compressed to final nominal plastic strains of 0, 25, or 50%. After compression, each coupon was isothennally aged at a temperature between 105 and 135[degrees]C. The majority of samples were taken from coupons that were aged up lo 600 h al 125[degrees]C or up to 6000 h at 105[degrees]C. Additional tests were done on 25% compressed coupons that were aged at 120, 130. and 135[degrees]C. These aging temperatures were selected to span between 10 and 40[degrees]C below the glass transition temperature of PC, which was estimated by means of differential scanning calorime-try as 147[degrees]C.

Figure 1 shows the three possible distinct orientations of Charpy samples that can be taken after compressing an initially isotropic sample. From these three possible directions, we only made samples indicated as TA and TT.since the dimension of the coupons did not allow the making of AT samples. In designating the samples, "A" indicates the direction of compression (axial) and "T" indicates the direction perpendicular to the direction of compression (transverse). The first letter in the sample designation indicates the orientation of the longest side of the sample and the second indicates the direction of the width, which is also the direction of crack propagation. As such. TA designates a sample that has its longest dimension transverse lo the direction of compression and its width along the direction of compression, and TT designates a sample that has its longest dimension transverse to the direction of compression and its width also transverse lo the direction of compression. The outer boundary of each coupon was removed, and. depending on the extent of compression, typically two to three samples were cut from the central region of each coupon.

All Charpy samples were machined to the final dimensions of 80 mm * 10 mm * 4 mm. After machining, a 2 mm 45[degrees] impact notch was cut so the sample width at the notch was reduced from 10 mm to a uniform 8 mm. The notch was made with a hand-cranked Charpy sample notching machine (CEAST 6816).

The same Charpy samples were used for ultrasonic measurements. These tests were conducted through the thickness of ihe sample after conducting a Charpy lest. In this case, the TT samples had a thickness along the axis of compression and the ultrasonic measurements provided the longitudinal wave modulus along this axis, and the TA samples had a thickness along the direction perpendicular to the axis of compression and so the ultrasonic tests provided the longitudinal wave modulus along the transverse direction.

All Charpy experiments were performed on a non-instrumented Charpy impact machine (CEAST 6545) with a 15 J Charpy pendulum hammer (CEAST 6545.015) and Charpy supports (CEAST 6545.073). The total energy of the impact hammer was 15 J wilh a speed at impact of 3.7 m/s. The distance between the supports was 40 mm. The energy absorbed ? the impact was measured from the change of kinetic energy of the pendulum, which was calculated from the difference in the starting and ending heights of the impact pendulum by a digital shaft encode!'. Correction for friction losses in the machine was calculated before each set of experiments by running the machine without a sample present.

The ultrasonic tests were perfonned using a square-wave generator (Olympus 5077PR) and a 1 MHz constant longitudinal wave transducer (Panamelrics V103). The pulse-echo method was used to calculate the longitudinal wave speed by measuring the lime of flight between the second and third reflections. A standard density measurement was done using the weight of each sample in and out of water. The signals from the transducer were recorded using a Lab View program running on a PXI base National Instrument oscilloscope (NI PXI-5122). Approximately 10 tests were conducted for each measurement.

Additional information and results can be found in the thesis of Strabala (26). Meagher (27) and Landais (28). General reference to fracture in polymers can be found in Ref. (29).

Positron annihilation lifetime (PAL) spectra s(t) = [SIGMA]([I.sub.i]/[[tau].sub.i]exp(-t/[[tau].sub.i]). approached by a sum of discrete exponentials with [[tau].sub.i], and [I.sub.i], representing, respectively, the average positron lifetime and intensity of ihe i-th positron decay component, were recorded with the fast coincidence system (ORTEC) of 230 ps resolution (FWHM of a single Gaussian, detennined by measuring the [60.sup.Co] isolope) at the temperature 22[degrees]C and relative humidity of 35%. To resolve short-lived components, each PAL spectrum was measured with a channel width of 6.15 ps (total number of channels 8000) and contained 3 * [10.sup.6] coincidences in total. To obtain information about long-lived components, the same samples were measured again with a 61.5 ps channel width, the total number of channels and coincidences being the same. Isotope [ .sup.22]Na (activity Tilde50 kBq) was used as the source of positrons (prepared from an aqueous solution of [ .sup.22]NaCl, wrapped with Kapton[R] foil of 12 [mu]m thickness and sealed), which was sandwiched between two identical samples. All the PAL spectra of the investigated samples were decomposed into four discrete exponentials (i = 4) using the standard LT 9.0 program (30). where additional peaks into the envelope of the fitted curve were added only if they significantly improved goodness of the fit. The uncertainties in the determination of lifetimes ([[tau].sub.i]) and corresponding intensities ([I.sub.i]) are [+ or -]0.005-0.5 ns (increasing with increasing [[tau].sub.i]) and [+ or -]0.5%, respectively.

Assuming an approximately spherical shape for the free volume, the O-PS lifetime ([[tau].sub.0]) can be related to the average radius of holes (R) by the semi-empirical Tao-Eldrup equations (31. 32).

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where [DELTA]R is an empirically detennined parameter (in the classical case [DELTA]R [approximately equal to] 0.1656 nm). describing effective thickness of the electron layer responsible for the pick-off annihilation of o-Ps in the hole. The fraction of free volume associated with o-Ps lifetime can be estimated (31) using equation

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where [I.sub.o] is the intensity of a component in the fit of the PAL spectrum, which corresponds to the o-Ps lifetime. In a general sense, we may relate each [tau] to an elementary cage size and the associated I to the number of cages.

RESULTS

Fig. 3 shows the dissipated energy measured during Charpy impaci tesis on samples without any induced aging. For the undeConned coupons, the as-received sheets provided a dissipated energy oC 10 kJ/[m.sup.2]. while a similar sample thai was rejuvenated, by healing above the glass transition and quenching in an ice-bath, provided a dissipated energy about 10 times larger (103 kJ/[m.sup.2]). Plastically defomied samples showed dissipated energy more in line with the rejuvenated samples, and with a clear anisotropy associated wilh the direction of the sample relative to the direction of compression. The TA samples showed higher dissipated energy and the TT samples showed smaller dissipated energy, when compared to the rejuvenated PC. For 25% compressed samples, the TA value is 133 kJ/[m.sup.2] and the TT value is 95 kJ/[m.sup.2]. For 50% compressed samples, the TA value is 164 kJ/[m.sup.2] and the TT is 56 kJ/[m.sup.2].

Figs. 3 and 4 show the dissipated energy obtained from thermally aged samples, respectively, at 105 and 125[degrees]C. On these figures are also reported the data for the as-received samples aged under similar conditions and the response of the annealed and quenched sample (without aging).

In Fig. 3 for physical aging al 105[degrees]C. there is minimal change in the dissipated energy for the first 300 h of aging, al which time the 25% compressed TT sample-shows a fairly abrupt drop in dissipated energy lo that of the as-received sheet. A similar drop occurs in the 25% compressed TA sample only after about 1500 h of aging. No such drops were observed in the 50% compressed samples during 6000 h of aging.

We accelerated the thennal aging by increasing the aging lemperaiure from 105[degrees]C to 125[degrees]C. In Fig. 4. the accelerated aging produced similar response in ihe samples, but wilh a shift of the ductile to bridle iransiiion to shorter times (about 10 limes faster). We still do not see a transition in the 50% compressed samples for the 600 h duration of aging, while for the 25% compressed coupons, we see a practically instantaneous drop in dissipated energy in the TT samples and a transition at about 20 h of aging for the TA samples.

Fig. 5 shows the typical fracture surfaces observed after Charpy testing. As can be seen, we observed brittle fracture for samples showing a low dissipalion of energy (Tilde10 kJ/[m.sup.2]). we observed ductile fracture for samples showing a high dissipation of cnerg\. and we observed ductile fracture with crazing on the samples with a dissipated energy below 90 kJ/[m.sup.2], but which were above the brittle fracture response of 10 kJ/[m.sup.2]. The surface of the samples that underwent brittle fracture was smooth with small dimensional changes in the shape of the sample cross section. The samples showing ductile fracture had a failure surface that was rough, characteristic of shear failure, and the cross section was substantially distorted

The effect of aging on the fracture surface and the associated dissipated energy was typical of physical aging. This can be seen in Fig. 6. which plots the failure-mode of samples as a function of aging temperature for different aging times. As can be seen, the lime ii lakes to transition from a ductile to brittle failure mode decreases as ihe aging temperature is increased.

Fig. 7 shows ihe results of an extensive study of the transition from ductile lo brittle failure during aging al 125[degrees]C in the 25% compressed samples cul along the TA orientation. As shown in the figure, the samples either failed in a ductile mode with a high dissipated energy or in a brittle mode with a low dissipated energy, never failing al energy between the two. For this case, all samples below 20 h of aging always showed ductile failure, while samples above 300 hours of aging all showed brittle failure. Between 20 and 300 h of aging the samples failed either in a ductile or brittle mode, wilh a typical statistical nature thai somewhal favors one limit or the other.

Figs. 8 and 9 show the longitudinal ultrasonic wave modulus measured along the axial and transverse directions as a function of aging at 105 and 125[degrees]C, respectively. As can be seen, there are small changes in the moduli with aging, and are uncorrected wilh the drops in the dissipated energies. Clearly, the anisotropy in elastic wave moduli is slightl) reduced, but nol totally removed by aging.

Fig. 10 shows the differential scanning calorimetric response for a compressed and slightly aged material showing the development of two endothermic peaks al the glass transition. For comparison, the response after annealing of the sample at the end of the first heating is shown for a second cycle under the same conditions, which shows a glass transition in the same temperature range (140-155[degrees]C). and that the value of [DELTA][C.sub.P] ([T.sub.g]) = [C.sub.pliquid] - [C.sub.pglass] at [T.sub.g] is the same for the two curves. The comparison of this to the response til undeformcd PC" indicates thai the values are similar. This similarity indicates thai there are no irreversible phenomena, such as a chemical reaction, during these transformations and mechanical treatments.

Fig. 11a and b shows the data obtained by means of PAL spectroscopy for a channel width of 6.15 and 61.5 Ps, respectively. The results provided sufficient information to calculate four exponents in the analysis. Table 1 provides the values for ([[tau].sub.3], [I.sub.3]) and ([[tau].sub.4], [I.sub.4]). and the associ ated hole radii and free volumes.

TABLE I. PALs measurements on samples that
were uncompressed and unaged. uncompressed
and aged 1500 h at 115[degrees]C, 25% plastically
compressed but unaged. and 25% plastically
compressed and aged 1500 h at 115[degrees]C.
Sample IPO       ([(tau].sub.3],  [l.sub.3],  [R.sub.3],
                 (ns; [+ or -]     (%; [+ or    (A)
                 0.01)               -] 0.5)
Uncompressed;    2.07                2.1.6         2.943
unaged
Uncompressed;    2.04                24.1          2.915
aged 1500 h at
115[degrees]C
25% compressed:  2.10                22.8          2.961
unaged
25% compressed;  2.07                23.8          2.943
aged 1500 h at
115[degrees]C
Sample IPO       [Tildef.sub.v],  ([(tau].sub.4],  [l.sub.4],
                 (%)            (%; [+ or -]      (%; [+ or
                                1.5)                -] 0.02)
Uncompressed;    5.8            68.6                0.16
unaged
Uncompressed;    5.8            60.0                0.13
aged 1500 h at
115[degrees]C
25% compressed:  5.7            76.7                0.25
unaged
25% compressed;  5.8            71.1                0.15
aged 1500 h at
115[degrees]C
Sample IPO       [R.sub.4],  [Tildef.sub.v],
                 (A)           (%)
Uncompressed;    18.14         9.3
unaged
Uncompressed;    16.60         5.8
aged 1500 h at
115[degrees]C
25% compressed:  19.74         18.3
unaged
25% compressed;  18.63         9.2
aged 1500 h at
115[degrees]C


For [[tau].sub.3], we find values between 2.04 and 2.10 ns and for [I.sub.3] values between 22 and 24%, which are consistent with the reported values for polycarbonate (32-34). For the uncompressed, as well as for compressed, samples, aging leads to a decrease of the values of [[tau].sub.3] wilh insignificant change in [I.sub.3], which is the same as what was observed by Cangialosi el al. (35, 36). This indicates that aging decreases the size of the individual holes in the free volume, hut not the number of holes. On the other hand, the compressed and imaged sample exhibits a value of [[tau].sub.3] thai is higher than ihe one measured for ihe uncompressed and imaged sample. This indicates that compression increases the hole volume, which is also consistent with the expected behavior when an amorphous structure is mechanically loaded (35). The fraction of free volume engaged in ihe processes is found to be close to constant (Tilde6%).

For the longer cage characteristics obtained using ([[tau].sub.4], [I.sub.4]), we observe a volume of 6 * [10.sup.3] [A.sup.3] for uncom pressed and unaged PC that increases by compression io 7.7 * [10.sup.3] [A.sup.3] ([approximately equal to]28% of variation). On the other hand, aging of the undefonned sample acts systematically lo reduce ihese caviiies from 6 * [10.sup.3] [A.sup.3] to 4.6 * [10.sup.3] [A.sup.3] (a decrease of [approximately equal to]23%). Similarly, aging of defonned samples reduces the size of these cavities from 7.7 * [10.sup.3] [A.sup.3] to 6.5 * [10.sup.3] [A.sup.3] (a decrease of [approximately equal to]16%).

DISCUSSION

The Charpy test results clearly indicate thai we have a brittle stale of low dissipated energy during failure for the as-received samples (103 kJ/[m.sup.2]), which transforms lo a state that has a ductile characteristic with high dissipated energy (103 kJ/[m.sup.2]) when the sample is annealed above the [T.sub.g] of 147[degrees]C and quenched in an ice bath. A similar rejuvenation occurs when we plastically compress the as-received samples, with the additional feature of the response becoming anisotropic so the failure characteristics become a function of the orientation of the sample (TA or 11 1. In this case, the response of the samples is approximately centered around the response of the annealed and quenched sample. This is clearly shown in Fig. 2 and. also, reported by others (2). The difference between the current results and those reported in Refs. 11-31 is the fact that we removed the boundaries from the samples so lhat residual stresses will play a smaller role in the results. The 500-1500% increase in the dissipated energy observed relative to the as-received material primarily comes from the change in the mode of failure from a smooth tension driven brittle fracture lo a torturous shear-driven ductile failure that frequently results in a sample that is only partially fractured, ihe characteristics of which are shown in Fig. 5.

Since the TA and TT samples result in nominal failure surfaces that pass through identical amount of identically oriented material, one may explain the observed difference (anisotropy) by the difference of wave speed along the axial and transverse sample directions, or by development of anisotropic flaws. The TT supports a crack that primarily runs along the iransverse direction, which is the direction of larger wave speed, while the TA sample supports a crack that runs in the axial direction of the simple, which has the lower wave speed. The main events influencing the failure are the intrinsic material failure properties and how the loading at the crack lip is influenced by ihe impact. In the Charpy test, for the initiation and propagation of the failure surface one needs waves to transmit the load from the impaci to the crack tip. The observed dissipated energy in ihe lest is thus controlled, in addition to material failure properties, by a complex interaction between the impaci event, the boundaries, and the crack lip. Preliminary results of Sirabala (26) indicate that similar anisotropy exists in quasi-static tests, further indicating lhal the observations are a result of an inherent material characteristic, not a result of wave propagation and sample size.

The defonnation aliers ihe dissipated energy for the PC in a way that is similar to that of removing the aging of the sample. This is seen in Fig. 2 where ihe dissipated energy of the plastically defonned material is centered around the response of the annealed and quenched material, as opposed io the as-received material. This is further indicated by the PAL response shown in Table I where the radii [R.sub.3] and [R.sub.4] of the voids for the deformed sample are larger than those of the undeformed sample. As for ihe undeformed sample. Table 1 shows lhal aging reduces the radii of voids [R.sub.3] and [R.sub.4] indicating a similar event. According to the obtained PAL data, we can assume ihat physical ageing leads lo a decrease in ihe number and size of bigger voids, or. alternatively, their fractioning into voids of smaller dimensions. Plastic compression acts in an opposite manner so that larger voids appear in the fonn of cracks or. alternatively, smaller voids agglomerate together to fonn bigger sizes. In addition, when the material is defonned. the fracture surface after the Charpy test indicates a shear driven ductile failure lhal is similar to thai of an unaged material. These results are in good agreement wilh previous studies which demonstrated that the molecular mobility, and more precisely cooperative motions, in the macromolccular structure arc greatly influenced by the strong intemiolccular interactions occurring between neighbor phenyl rings (37). The thennal aging is obviously promoting inlermolecular interactions or reducing the free volume ratio, leading to a more dense material, while plastic defonnation globally reduces the iniemiolec-ulat interactions by increasing the size of the free volume. Fven if it is impossible to observe the anisotropy in the material during the plastic flow using the PAL technique, one can assume thai during plastic flow alignment of mac-romolecular chains in ihe transverse direction promotes intennolecular interactions and penalizes these interactions in the axial direction.

Bngels et al. (22) clearly saw the drop in the dissipated energy with thennal aging of an undefonned PC, and studied this effect for different molecular weights and aging temperatures, and observed ihe change in the fracture from a ductile lo brittle characteristic. They also noted ihe aging lemperaiure dependence of the transition time and used a lime-lemperalure superposition lo nodd il. They also noted the slatislical nature of ihe transition zone resulting in either duciile or brittle failure, ihe average showing a large standard deviation.

The aging of ihe defonned malerial exhibits the influence of the anisotropy developed by the plastic flow. This is characterized with uneven changes in the wave speeds as shown in Figs. 8 and 9. In addition. Fig. 10 shows that the compressed malerial shows a separation of the glass transiiion into two peaks possibly indicative of the difference in aging associated with the two different directions. An anisotropic difference in ihe effect of physical aging is clearly seen in the 25% compressed samples. As seen in Figs. 3 and 4, samples that are plastically defonned and aged show a drop in their dissipated energy, bui the time of this drop was different for the TT and TA sample directions, indicating ihat the effect of physical aging is different along different directions of the same plasiicalK compressed coupon. The development of the double peaks with aging of the plastically deformed sample in ihe DSC signal shown in Fig. 10 could be an indication of ihe development of this anisotropy in ihe response to aging.

The transition by aging of the Charpy measured dissipated energy is clearly statistical in nature since once we reach the range during which the transition can happen we see either ihe ductile or the brittle response, never seeing a response in between the two. This is clearly seen in Fig. 7 where the individual test results arc plotted. In addition, the range of aging limes, in which we see both modes of failure, changes with ihe aging temperature as indicated in Fig. 6.

We do nol see a transition from ductile lo brittle failure in Ihe samples thai were 50% deformed. Further study might he nccessan lo determine if large deformations suppress this transition or if they still happen, but at a larger aging lime. Unfortunately, aging of highly compressed samples at higher temperatures can result in large recovery events that make it difficult to assess the proper source of a transition and low lemperaiure aging can be prohibitively long. A broader sludy of different defonnations might provide a picture of the relation of this transition to the amount of plastic deformation and aging temperature, from which one might estimate the lime needed lo reach this transitions in highly deformed materials.

CONCLUSIONS

In this study, it was demonstrated lhal samples toughened by compression display a toughening that is anisotropic and approximately centered around ihe behavior seen for the annealed and quenched material. This indicates that ihe plastic How. in part, acts in a way similar to that of rejuvenation of the as-received material, but also introduces a substantial anisotropy that can be persistent and that may anisotropically affect the later aging. The energ) dissipation in ihe tests depended on the orientation thai ihe crack will run in a Charpy sample. In particular, cracks moving along ihe compression axis consistently displayed greater energy dissipalion than those moving perpendicular to the compression direction. Nonetheless, ihe energy dissipalion for both orientations of samples was substantially greater compared to the as-received undefonoed polycarbonate.

The energy dissipation that was gained as a result of the plastic How was lost in the 25% compressed samples after aging at temperatures below the glass-transition temperature. After this loss had occurred due lo aging, the samples broke wilh roughly the same energy dissipation and brittle fracture mode as the as-received uncompressed samples. Samples broken with cracks running along the transverse direction experienced this loss much sooner than the other samples (after 1-2 h at 125[degrees]C). while the samples broken wilh cracks running along the axial direction look longer to lose their increased toughness (after 50 h al 125[degrees]C). Once the drop of toughness occurred, both the TT and TA samples showed exactly the same brittle failure and energy dissipation indicating a transition from anisotropic to isotropic response. This transition occurs at longer aging limes as the aging temperature was decreased and at shorter aging times when the aging temperature was increased, indicating a temperature dependence of these transitions. In the case of the 50% compressed samples, the majority of the energy increase was maintained al all aging limes tested, wilh what seems to be a full suppression of any fracture mode transition.

In samples thai did transition from their toughened stales, the transition was bimodal. The samples would either fail at iheir toughened state or would revert fully to the as-received undeformed state. Furthermore, after performing ultrasonic wave speed measurements on Ihe samples, there was a significant drop in the longitudinal wave speed for certain samples, specifically those with 25% compression, but these drops were uncorrelated lo the fracture transitions.

REFERENCES

(1.) LJ. Broutman and S.M. Krishnakumar. Polvm. Eng. Sci..16(2). 74 (1976).

(2.) B.S. Thakkar. L.J. Broutman. and S. Kalpakjian, Polym. Eng. Sci.,20(11), 756 (1980).

(3.) B.S. Thakkar and LJ. Broutman. Polym. Eng. Sci..21(3), 155 (1981).

(4.) I.M. Ward. Mechanical Properties of Solid Polymers, Wiley. New York (1979).

(5.) A. Goel. K. Sirahala. M. Ncgahban, and J.A. Turner. Polym. Eng. Sci..49(10), 1951 (2009).

(6.) K. Kurahayashi. Int. J. Thcrmopliys..22(1). 277 (2011).

(7.) E.T.J. Klompen. T.A.P. Engels. L.E. Govacrt. and H.E.H. Mcijcr. Macromolecules,38, 6997 (2005).

(8.) CH. Ho and T. Vu-Khanh. Theor. Appl. Fract. Mali..40(1). 75 (2003).

(9.) R.A. Bubcck. S.E. Bales, and H.-D. Lee, Polym. Eng. Sci.,24(14). 1142 (1984).

(10.) M.C. Boyec. E.M. Arruda. and R. Jayachandran, Polym. Eng. Sci.,34(9). 716 (1994).

(11.) A. Goel, K. Strabala. M. Negahban. and R. Feng. Exp. Merit..50(6), 709 (2009).

(12.) H.J. Kolman. K. Ard. and C.L. Beany. Polym. Eng. Set..22(15). 950 (1982).

(13.) L.B. Liu, A.F. Yce. and D.W. Gidley. J. Polym. Sci. Part B: Polym. Phys.,30(3). 221 (1992).

(14.) S. Sarva. A.D. Mulliken. and M.C. Boycc, Int. J. Solids Struct..44(7-8). 2381 (2007).

(15.) D.J.A. Scndcn. S. Krop. J.A.W. van Dommelen, and L.E. Go vac n. J. Polym. Sci. Part B: Polym. Phys..50. 1680 (2012).

(16.) L.C.E. Slruik. Physical Aging in Amorphous Polymers and Other Materials. Elsevier, Amsterdam (1978).

(17.) J.M. Hutchinson. Prog. Polym. Sci.,20, 703 (1995).

(18.) I.M. Hodge. J. Non-Crystalline Solids.169(3). 211 11994).

(19.) R. Rahouadj and C. Cunat, Physical Aging and Glass Transition of Polymers. Handbook of Materials Behavior Models. Academic Press. Burlington. 944-954 (2001).

(20.) CH. Ho and T. Vu-Khanh, Theor. Appl. Fract Mech,41(1-3), 103 (2004).

(21.) F. Rouabah. M. Fois, L. Ibos, A. Boudenne.C. Picard. D. Dad ache, and N. Haddaoui.7. Appl Polym. Sci..109. 1505 120081.

(22.) T.A.P. Engels, L.C.A. van Brccmen. L.E. Govacn, and H.E.H. Mcijcr, Polymer.52(8). 1811 (2011).

(23.) C. Bauwcns-Crowct and J.-C. Bauwens. Polymer.31(4). 646 (1990).

(24.) C. Bauwens-Crowet and J.-C. Bauwens. Polymer.31(2). 248 (1990).

(25.) J. Othniczouri-Dccerf. Polymer.35(22). 4734 (1994).

(26.) K.W. Strabala. "The Effect of Combined Compression and Aging on the Properties of Glassy Polycarbonate." Master of Science Thesis.University of Nebraska-Lincoln. Lincoln. Nebraska, August (2009).

(27.) S.E. Meagher. "Anomalous Loss of Toughening of Work Toughened Polycarbonates." Master of Science Thesis.University of Nebraska-Lincoln. Lincoln. Nebraska. December (2010).

(28.) C. Landais. "Ultrasonic Methods for Ihe Characterization of Complex Materials and Malerial Systems: Polymers. Structured Polymers. Soft Tissue and Rone." Master of Science Thesis.Univcrsily of Nebraska-Lincoln. Lincoln. Nebraska. May (2011).

(29.) W. Grellmann and S. Seidler, Deformation and Fracture Behaviour of Polymers. Springer. Berlin Less than2(X)l l.

(30.) J. Kansy. Nucl. Instr. Methods Phys. PesA374. 235 (1996?.

(31.) D. Cangialosi. H. Sehul. A. van Veen, and S.J. Pieken. Macromolecules.36. 142 (2003).

(32.) A,J, Hill. K.J. Healer, and CM. Agrawal. J. Polym. Sii. Part B: Polym. Phys..28(3), 387 (1990).

(33.) J. Del Rio. J. Sema. F. Plazaola. J.J. Iruin. and J.L. Nazabal. J. Non Crystalline Solids.287. I(X) (2001).

(34.) WJ. Davis and R.A. Pcthrick. Eur. Polym. J..34. 1747 (1998).

(35.) D. Cangialosi. M. Wubbcnhorsi, H. Schui. A. Van Veen, and S.J. Pickcn. J. Chem. Phys..122. 064702 (2005).

(36.) AJ. Hill. P.L. Jones. J.H. Lind. and G.W. Pcarsall J. Polym. Sc. Part A: Polym. Chem.,26(6). 1541 (1988).

(37.) K. Arabeche. L. Delbreilh. R. Adhikari. G.h. Miehlcr, A Hiltner, E. Baer. and J.M. Sailer. Polymer.53(6), 1355 (2012).

Corrrxpomlcme to: M. Ncy;ihtxm: e-mail: iiinegahhatKaunl.eilu CoMMl grant sponsor: US Army Research Laboratory (ARL). RMAC-RTP Cooperative; contract grant number: W911NF-04-2-OOI I; contract pram sponsor; Handle Managed Contratl TCN-MKW. Amiy Research Office: contract gram number: INF-OX-I -04X3: contract grant sponsor: Department of Education FIPSE ATLANTIS Mobility PI I6J07H02X-09, DO! IO.I002/pcn.236I.S Published online in Wiley Online Library (wili-yonlinclibrary.comi. [c]2013 Society of Plastics Engineers

Kyle Strabala, (1), (2) Shawn Meagher,' (1), (2) Charles Landais, (1), (2) Laurent Delbreilh, (2) Mehrdad Negahban, (1) Jean-Marc Salter, (2) Joseph Turner, (1) Adam Ingram, (3) Roman Golovchak (4)

(1) AM ME-ATE AM, Mechanical and Materials Engineering, University of Nebraska-Lincoln, Lincoln, Nebraska 68588-0526

(2) AMME-LECAP, tnstitut des Materiaux de Rouen, Universite de Rouen, Avenue de I'Universite' BP 12, 76801 Saint Etienne du Rouvray, France

(3) Department of Physics of Opole University of Technology, 75, Ozimska str., 45370 Opole, Poland

(4) Department of Materials Science and Engineering, Lehigh University, 5 East Packer Avenue, Bethlehem, Pennsylvania 18015-3195
COPYRIGHT 2014 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Strabala, Kyle; Meagher, Shawn; Landais, Charles; Delbreilh, Laurent; Negahban, Mehrdad; Salter, Jea
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:4EUFR
Date:Apr 1, 2014
Words:6000
Previous Article:Positron annihilation lifetime spectroscopy for miscibility investigations of styrene-butadiene-styrene copolymer/polystyrene blends.
Next Article:Synthesis and Properties of Block and Graft Waterborne Polyurethane Modified With [alpha],[omega]-Bis(3-(1-methoxy-2-hydroxypropoxy)propyl)...
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters