# A novel fabrication route for Auxetic polyethylene, part 2: mechanical properties.

INTRODUCTION

Auxetic microporous polymers (i.e., those with a negative Poisson's ratio, [upsilon]) have been studied since 1989, when a form of polytetrafluoroethylene was discovered to have this property due to its complex microstructure [1, 2]. This consisted of nodules interconnected by fibrils. The microstructure was subsequently reproduced in ultra high molecular weight polyethylene (UHMWPE) (3), polypropylene (PP) (4), and nylon (5) using the powder processing techniques of compaction (6), sintering (7) and extrusion (8), resulting in cylindrical rods. The role of each stage was carefully studied to aid understanding, and this indicated that the compaction stage had the sole function of imparting structural integrity to the extrudate. So, the compaction stage was omitted in a systematic study [9, 10], and the result of this was that the extrudate produced was highly fibrillar and very auxetic but had low mechanical properties and low density. The complex microstructural formation for the cylinders was also studied (11), using the no-compaction route to fabricate the specimens, as these can be easily sectioned for detailed microstructural analysis. The main aim of the study was to understand the mechanisms of microstructure formation. The microstructure was produced by surface melting of the material leading to the formation of thick necks, which are drawn into fibrils. The latter was revealed to be not necessarily the result of extrusion, but could also result from particle movement prior to extrusion as the material is forced into close contact.

The cylindrical rods produced in these studies are ideal for laboratory scale testing, but are restricted in their size and geometry by the processing rig in which they are produced. Two methods can be considered to produce more useable and useful shapes, with a move toward applications. The most obvious is to develop routes to fabricate extruded products such as fibers or films, and this strand of the work has been the focus of concerted research at Bolton in recent years. In 2001, auxetic fibers were fabricated using a partial melt spinning technique, with PP, polyester, and nylon fibers being produced (12-14). A slight modification to the process (i.e., by changing the extrusion die to a slit die rather than a cylindrical die) allowed auxetic PP films also to be extruded (15). Once again, the mechanisms for auxetic behavior were based on the microstructure, but in these cases, from microstructural observations, there is evidence of a particulate/granular microstructure as well as low porosity. This is consistent with the starting powder being compacted and partially melted, leading to an interconnected structure (16), (17).

As a result of possessing a negative Poisson's ratio, certain properties are predicted from classical elasticity theory to be enhanced in auxetic materials (18). To date, experimental verification has been carried out on certain of these, with excellent results. For example, enhancements of up to four times over conventional properties have been found in indentation resistance (19-23), low velocity impact resistance (24), plane strain fracture toughness (19), and absorption at sonic and ultrasonic frequencies (25-28). These property enhancements occur across the whole spectrum of auxetic materials. To take the indentation resistance as an example, the work referred to above concerns such diverse materials as polymeric and metallic foams, carbon fiber composite laminates, and microporous polymers. The fabrication of a more useful form of auxetic polymer, that is, the fiber, has allowed more applications centered research into the resistance to fiber pullout in composite materials. Here, an auxetic fiber embedded in a matrix is predicted to lock into place (29). This concept was experimentally verified, and it has been shown that an auxetic fiber is up to three times more difficult to extract from a matrix than an equivalent conventional fiber (30).

However, even the production of films and fibers is still limiting the potential use of auxetic polymers, as their shape is decided by the extrusion die. The second approach, then, is to develop a route for producing auxetic polymers that does not include extrusion at all. The first part of this article (31) reports work to produce auxetic polymers in a more usable and useful form, which does not require an extrusion stage. The advantage of this is the flexibility that it should allow in forming more complex, applications-driven shapes. The process is summarized in Fig. 1 and consists of compaction followed by multiple sintering. The optimum results (31) have been achieved by compaction followed by double sintering, which has given a strain-dependent Poisson's ratio as low as [upsilon] = -0.32.

This article sets out to evaluate the mechanical properties of auxetic UHMWPE produced by compaction, followed by multiple sintering by measuring the flexural properties of the cylinders produced. In addition, the indentation resistance was also measured to see if any enhancements over conventionally processed UHMWPE could be observed as is the case for auxetic UHMWPE produced by the three-stage processing route. Part 1 of this article found a variation in the microstructure and Poisson's ratio along the specimen length and so for the indentation resistance, testing was carried out at various defined locations along the specimen length.

EXPERIMENTAL METHODS

For completeness, this section will detail the fabrication processes for the production of auxetic cylindrical rods by the multisintering route. For the indentation tests undertaken, the comparative material used was compression-molded plaques and the process for producing these is also detailed. This was to provide a direct comparison between UHMWPE produced by conventional means and the auxetic multisintered specimens produced here.

Fabrication of Auxetic Specimens

Specimens were fabricated as detailed in Fig. 1, with further details on the experimental procedures contained in Part 1 of this article (31). For completeness, this is summarized here. The first stage of the process to produce these specimens was to compact the UHMWPE powder as had previously been done for the three-stage processing route. The rig that is shown in Fig. 2 was heated to 110 [degrees]C and fitted with a blank die. The powder was poured into the barrel, which had a diameter of 15 mm and allowed to come to temperature for 10 min. A pressure of 7 kN was applied at a constant displacement rate linearly up to 140 mm/min and maintained for 20 min. The compacted rod was then removed from the extrusion rig and allowed to air cool. A range of cylinders were produced by subjecting them to single, double, triple, or quadruple sintering treatments, each at 160 degree C for 20 min. Between each sintering treatment, the specimens were removed from the barrel and allowed to air cool to room temperature outside barrel and without constraint.

Fabrication of Compression-Molded plaques for Comparative Testing

Plaques of UHMWPE were produced by compression molding using an in-house molder. Temperatures were set to 165 degree C and a pressure of 710 MPa was applied hydraulically for a period of 30 min by a manually operated pump. At the end of the loading period, the pressure was removed using a release valve and the plaque was allowed to cool within the tooling to room temperature. Plaques of dimensions 240 X 115 mm were produced in this manner.

Measurement of poisson's Ratio

Samples from each type of specimen were subjected to a series of single, discrete compression tests in the radial, r, direction. This was so that the Poisson's ratio, [V.sub.rz] could be determined using a photographic technique (1) developed in-house. For each material type, orientation and location, at least five tests were conducted.

Determination of the Flexural Properties of the Specimens

Simple three-point bend tests were conducted, in accordance with the British Standard (32) on specimens representing each of the four types of multisintered materials and for further comparison, on several samples that had been fabricated by the original three-stage processing route and those that had been compacted only. At least three examples of each material type, orientation, and location were tested and the results are presented as an average of the tests in Table 1. The purpose of these tests was to provide a semiquantitative indication of the strengths and stiffnesses of the multisintered materials with respect to those of specimens produced by alternative processing routes.

The support points were set at a spacing equal to five times the specimen diameter and the crosshead speed used was 1 mm/min. A force/displacement curve was produced for each test and a typical example of this is shown in Fig. 3. From this, values of the flexural strength, S, modulus, M and strain to failure, [[epsilon].sub.f] were calculated using the following equations (33):

S=8PL / [pie[D.sup.3]]

M=[4PL.sup.3] / [3.pie]D.sup.4]]y

[epsilan.sub.f]=6Dy / [L.sup.3]

where P is the applied load, L the support separation, D the rod diameter, and y the maximum rod deflection.

Determination of Indentation Resistance

To study the variation along the specimen length of the indentation resistance, sections 4-mm thick were cut from each of the rods undergoing different thermal treatments in both the radial, r, and longitudinal, z, directions. For the former, the rods were thinned into bars with flat parallel sides in the z direction to a thickness of 3-4 mm and for the latter, the rods were sectioned radially into discs 3-4 mm thick in the r direction, as indicated in Fig. 4. These were then indented at a loading rate of 0.5 mm/min using a spherical steel indentor of diameter 5 mm driven by an Instron 1185 mechanical testing machine. Accuracy of the tests was ensured by recalibrating after each test had been carried out. Tests were conducted on three specimens of each type of material, location, and orientation. Furthermore, at least three tests were performed on each section whilst ensuring that the indentations were at least 10-mm apart so as to avoid edge effects from neighboring indentations. This gave a minimum of nine values of strain and corresponding hardness at each load for each type of material, location, and orientation.

A typical example of the force/displacement plot obtained during indentation testing is shown in Fig. 5. For a given depth of indentation into a material, the indentation resistance, H, is equal to the pressure required to cause the indentation, that is,

H=P / [[pie]a.sup.3]

Where a is the indentation radius of the surface. Since the radius of the indentor, R, is very much greater than the indentation depth, h, the approximation:

[a.sup.2]=2Rh

is used and so, H is then given by:

The values of P and h were then obtained from the force/displacement plots produced during the indentation tests (22) and H was calculated. Tests were conducted up to loads of 15, 25, 50, 75, 100, 150, and 200 N. For the test load of 15 N, the data were also analyzed at 5 and 10 N; for the test load of 25 N, analysis was carried out at 5, 10, and 20 N; and for the test load of 50 N at 10 and 25 N. For the remaining test loads considered, analysis was only carried out at the maximum applied load. An average value was calculated for each particular location, type of material, and orientation and these are presented in Tables 2-7. The reason for concentrating data analysis at the lower loads is that previous work (22) has indicated that this is where enhancements are most likely to occur, that is, where the material is still behaving in a primarily elastic fashion.

RESULTS AND DISCUSSION

Measurement of Poisson's Ratio

For completeness, the results of the variation of the Poisson's ratio of the samples are considered here. A much fuller analysis has been undertaken and is reported in Part 1 of this article (31), along with a detailed microstructural examination plus density measurements.

After a single sintering treatment, it is possible that auxetic behavior has been induced on the plunger end of the samples but it is only found at very low strains and at levels where confidence in the measured values is not high. However, after double sintering, the mid span region of the specimen was found to be auxetic at strains of up to 5% before becoming slightly positive as the strain increased. Strain dependent values as low as v = -0.32 were measured here. The die and plunger ends, after double sintering, were not auxetic. After a third sintering treatment, no significant differences were observed and after a fourth sintering treatment, only the die end of the specimen gave any indication of auxetic behavior, but again at too low a strain to be reported with confidence. The regions of auxetic behavior in each case were found to coincide with a nodule-fibril microstructure in the cylinders albeit with shorter fibrils than those found in the microstructure produced in the compaction, sintering, and extrusion three-stage processing route. Thus, the existence of a negative Poisson's ratio was consistent with the same mechanisms occurring to produce the negative Poisson's ratio as for cylinders produced by compaction, sintering, and extrusion, that is, translation of the nodules caused by the fibrils. Simple geometric modeling of the microstructure has been undertaken, giving excellent agreement with experimental results. This has allowed a theoretical curve to be constructed so that if the strain applied to the specimen at any given point is known, the Poisson's ratio can be predicted (31).

Flexural Properties of the Specimens

The results of the three-point bend tests conducted to reveal the flexural strength, modulus, and strain to failure of the compacted and multisintered specimens are shown in Table 1. The most dramatic changes to the flexural properties occur on the first sintering treatment. Here, the strength rises from 5.4[+ or -] 0.5 MPa to 32.1 [+ or -] 0.2 MPa and the strain to failure increases from (1.9 [+ or -] 0.1)% to (28.4 [+ or -] 0.5)%, while the modulus drops from 290 [+ or -] 10 MPa to 113 [+ or -] 3 MPa. On successive sintering treatments, the changes to these properties are, by comparison, practically negligible.

This fits in well with the variations in the overall density of the specimens that were reported in detail in Part 1 of this article (31). A purely compacted rod has a density of 790 [+ or -] 30 kg/[m.sup.3]. This increases to 890 [+ or -] 30 kg/[m.sup.3] with the first sintering treatment. After this, the density continues to increase, but at a much reduced rate until the fourth sintering treatment, where a density of 932 [+ or -] 4 kg/[m.sup.3] is achieved. This represents a density packing factor of 98%, obtained by dividing the density measured by the maximum density quoted for this material of 950 kg/[m.sup.3] (31).

It does appear, thus, that the sintering treatments only and not any applied pressure are causing a densification of the material and this is increasing the mechanical properties. However, it should be noted that the three-stage processing route produced much stronger specimens, with the strength being 55.6 [+ or -] 0.5 MPa, as can be seen in Table 1.

Indentation Resistance Tests

Tables 2-7 show the average indentation resistance at loads of 5, 10, 15, 25, 50, 75, 150, and 200 N on a conventional, compression-molded UHMWPE plaque (fabricated as described above) and on each of the plunger end, mid span, and die end sections of the single, double, and quadruple sintered specimens in both the z and r directions.

The results of the tests performed on the plunger end, mid span, and die end sections of the single sintered material are shown in Tables 2 and 3. It can be seen that in both directions, the single sintered material is always harder than the compression-molded plaque at the plunger end. The mid span material is harder at the lower loads and the die end material always has the lowest indentation resistance, which is particularly marked at the higher loads, for example, at 200 N, H = 9 [+ or -] 1 N/[mm.sup.2] in the z direction for the die end material and H = 15 [+ or -] 1 N/[mm.sup.2] for the compression-molded material.

Tables 4 and 5 show the results of tests performed on the double-sintered specimens. As expected, the hardness values rise in comparison with these obtained after a single sintering treatment. In each orientation, the hardness of both the plunger end and mid span sections is higher than that of the compression-molded plaque. In the z direction, it can also be seen that the hardness of the plunger end and mid span regions after a second sintering treatment were found to be comparable except at the highest loads, where the plunger end material is the hardest. In the r-direction, however, the mid span material was found to be harder than the plunger end material at lower loads. The hardness of the die end material remains substantially lower than that of the compression-molded plaque, particularly at the higher loads.

Tables 6 and 7 show that after the fourth sintering treatment, the rods become more consistent with respect to hardness values. The rise in hardness is particularly marked in the die end material, where it is now comparable to the compression-molded plaque at the higher loads and is considerably higher at the lower loads (e.g., at 5 N, the die end has an indentation resistance of H = 4 [+ or -] 0.6 N/m]m.sup.2] with the compression-molded plaque having H = 1.6 [+ or -] 0.1 N/[[[mm.sup.2] in the r direction).

The most interesting of all the results presented here are those at low loads in the r direction of the double-sintered material. Here, previous work reported in Part 1 of this article (31) has shown that the processing route generates a nodule-fibril microstructure with a resulting strain dependent negative Poisson's ratio as low as [upsilon] = -0.32 in the mid span region. At the strains employed in the indentation test, this corresponds to Poisson's ratios of the value of [upsilon] = -0.23 at an indentation load of 5 N, [upsilon] = -0.19 at an indentation load of 10 N, [upsilon] = -0.17 at an indentation load of 15 N, and [upsilon] = -0.11 at an indentation load of 20 N. At the lowest applied loads, that is, 5 N where the Poisson's ratio is most negative, the hardness in the r direction at the mid span region is 2.5 times that of the compression-molded plaque and 1.5 times that of the plunger end. The density variations along the specimen lengths were determined in Part 1 of this article and are reported here in Table 8 for completeness. It can be seen that the density in the mid span region is comparable to the compression-molded plaque (i.e., 910 kg/[m.sup.3] compared with 908 kg/[m.sup.3]) and less than that of the plunger end material (i.e., 940 kg/[m.sup.3]), so this is clearly not the cause of this increase in hardness. This is a significant finding and agrees with previous work (22) on the auxetic cylinders produced by compaction, sintering, and extrusion. In the latter case, enhanced hardness at low loads of up to three times that of compression-molded UHMWPE of equivalent modulus and density were measured. The mechanism involved in the response was a local densifica-tion of the microstructure with the fibrils pulling the nodular material elastically under the indentor.

It is also interesting to note that a nodule-fibril micro-structure was observed after a single sintering treatment at the plunger end and in the quadruple sintered specimen at the die end. There was some evidence for auxetic behavior, but at such small strains that confidence in asserting this was low. The die end of the samples remains much less dense than the rest of the specimen, even after four sintering treatments, so it is not surprising that this remains the least hard part of the specimen. However, it is at a comparable density to the compression- molded plaque and is, as noted above, harder than this possibly due to the microstructure observed. As far as the plunger end on the first sintering treatment is concerned, it is more difficult to assign the increase in hardness to an auxetic effect as this is a very dense region of the specimen even after just one sintering treatment.

It should be noted at this stage that all the work reported here has been primarily concerned with the manufacture of cylinders. However, the novelty of this work is that a route is proposed that can be easily adapted to the production of far more complex and useful shapes. Compaction of powder followed by double sintering has generated a nodule-fibril microstructure without the need for an extrusion stage. This has resulted in a strain dependent negative Poisson's ratio as low as [upsilon] = -0.32. The indentation resistance of the auxetic material produced here is enhanced at low loads, as it was for the three-stage extrusion route, so no loss in enhancement appears to result from replacing the extrusion stage with multiple sintering.

CONCLUSIONS

A novel processing route based on the powder processing techniques of compaction and multiple sintering has been developed that produces an auxetic form of UHMWPE. The best results to date have been produced by compaction followed by sintering at 160[degrees]C. air cooling to room temperature, and sintering again (i.e., a double sintering treatment) at 160[degrees]C. This has generated a nodule-fibril micro struct are and a strain-dependent negative Poisson's ratio as low as [upsilon] = -0.32 at low strains.

The cylinders produced show a large increase in flex-ural properties with a single sintering treatment accompanied by an increase in the density (31). Both continue to increase but at a reduced rate through successive sintering treatments, up to four.

The indentation resistance of the auxetic double-sintered material was measured and was found to be enhanced by a factor of up to 2.5 times at low loads when compared to conventional compression-molded UHMWPE. The effect is directly linked to the auxetic character of the specimen, as the more fibrillar regions are less dense but as expected still display a greater resistance to indentation.

This processing route marks the first stage to produce a greater range of more useful shapes rather than the cylinders, fibers, and films produced to date. The size and shape of these is severely limited by the need previously to extrude, that is, the dimensions of the extrusion barrel and die govern those of the end product. Removing the need to extrude should overcome this problem. Further development of the compaction with multiple sintering route for more complex shapes is the next step in this work.

ACKNOWLEDGMENTS

RSW thanks Prof. Wesley Cantwell for his assistance with this project.

REFERENCES

(1.) B.D. Caddock and K.E. Evans, J. Phys. D Appl. Phys,, 22. 1877 (1989).

(2.) K.E. Evans and B.D. Caddock, J. Phys. D Appl. Phys., 22. 1883 (1989).

(3.) K.L. Alderson and K.E. Evans. Polymer. 33. 4435 (1992).

(4.) A.P. Pickles, K.L. Alderson. and K.E. Evans, Polym. Eng. Sci., 36, 636 (1996).

(5.) K.L. Alderson, A. Alderson. R.S. Webber, and K.E. Evans, J. Mater. Sci. Lett., 17, 1415 (1998).

(6.) A.P. Pickles, R.S. Webber. K.L. Alderson. P.J. Neale, and K.E. Evans, J Mater. Sci., 30, 4059 (1995).

(7.) K.L. Alderson. A.P. Kettle. P.J. Neale, A.P. Pickles, and K.E. Evans,./. Mater. Sei.. 30. 4069 (1995).

(8.) P.J. Neale, A.P. Pickles. K.L. Alderson, and K.E. Evans, J. Mater. Sci., 30, 4087 (1995).

(9.) R.S. Webber, K.L. Alderson, and K.E. Evans. Polym. Eng. Sci., 40(8), 1894 (2000).

(10.) K.L. Alderson. R.S. Webber, and K.E. Evans. Polym. Eng. Sci., 40(8). 1064 (2000).

(11.) K.L. Alderson. R.S. Webber, and K.E. Evans, Phys. Status Solidi B, 244(3), 828 (2007).

(12.) K.L. Alderson, A. Alderson, G. Smart, V.R. Simkins. and P.J. Davies, Plast. Rubber Compos.. 31(8), 344 (2002).

(13.) N. Ravirala. A. Alderson, K.L. Alderson, and P.J. Davies, Phys. Status Solidi B, 242(3), 653 (2005).

(14.) N. Ravirala, K.L. Alderson. P.J. Davies, V.R. Simkins. and A. Alderson, Text. Res. J., 76(7), 540 (2006).

(15.) N. Ravirala, A. Alderson, K.L. Alderson. and P.J. Davies, Polym. Eng. Set.. 45(4), 517 (2005).

(16.) N. Ravirala, A. Alderson, and K.L. Alderson. J. Mater Set. 42, 7433 (2007).

(17.) K.W. Wojciechowski, J. Phys. A Math. Gen.. 36. 11765 (2003).

(18.) R.S. Lakes, Science. 235, 1038 (1987).

(19.) J.P. Donoghue and K.E. Evans, "Composite Laminates with Enhanced Indentation and Fracture Resistance Dae to Negative Poisson's Ratios," in Proceedings of ICC M8 SAMPE, S.W. Tsai and G.S. Springer, Eds., Sample, Covina, CA, 2-k-1 (1991).

(20.) N. Chan and K.E. Evans,./. Cell. Plast., 34. 231 (1998).

(21.) R.S. Lakes and K.J. Elms, J. Compos. Mater.. 27. 1 193 (1993).

(22.) K.L. Alderson, A.F. Fitzgerald, and K.E. Evans. J. Mater. Sci., 35, 4039 (2000).

(23.) V. Coenen. K. Alderson, P. Myler, and K. Holmes. "The Indentation Response of Auxetic Composite Laminates." in Proceedings of the 6th International Conference on Deformation and Fracture of Composites. Manchester, UK, April 4-5 (2001).

(24.)V. Coenen, K. Alderson, P. Myler. and K. Holmes, "The Low Velocity Impact Response of Auxetic Composite Laminates", in Proceedings of the 8th International Conference on Composites Engineering, Tenerife, Spain, August 5-11 (2001).

(25.) C.P. Chen and R.S. Lakes, Cell. Polym., 8. 343 (1989).

(26.) C.P. Chen and R.S. Lakes, J. Mater. Sci., 28. 4288 (1993).

(27.) B. Howell, P. Prendergast, and L. Hansen. Appl. Acoust., 43, 141 (1994).

(28.) K.L, Alderson, R.S. Webber, U.F. Mohammed. E. Murphy, and K.E. Evans. Appl. Acoust., 43(1), 23 (1997).

(29.) K.E. Evans, Chem. Ind., 20. 654 (1990).

(30.) V.R. Simkins, A. Alderson, P.J. Davies. and K.L, Alderson, J. Mater. Sci., 40, 4355 (2005).

(31.) K.L. Alderson, R.S. Webber. A.P. Kettle, and K.E. Evans. Polym. Eng. Sci., 45(4). 568 (2005).

(32.) BS 2782: Part 3: Methods of Testing Plastics. Mechanical Properties. Method 341A. Determination of Apparent Inter-laminar Shear Strength of Reinforced Plastics (1977).

(33.) H.S. Loveless, "Flexural Test," in Testing of Polymers, Vol. 2, J.V. Schmitz, Ed., Interscience Publishers. New York. 321 (1966).

Correspondence to: Kim Alderson; e-mail: kal@bolton.ac.uk

Contract grant sponsor: EPSRC.

DOI 10.1002/pen.21110

Published online in Wiley InterScience (www.interscience.wiley.com).

[c] 2008 Society of Plastics Engineers

R.S. Webber, (1) K.L. Alderson, (2) K.E. Evans (3)

(1) Department of Materials Science and Engineering, University of Liverpool, Brownlow Hill, Liverpool, UK

(2) Centre for Materials Research and Innovation, The University of Bolton, Bolton, UK

(3) School of Engineering, Computer Science and Mathematics, University of Exeter, Exeter, UK

Auxetic microporous polymers (i.e., those with a negative Poisson's ratio, [upsilon]) have been studied since 1989, when a form of polytetrafluoroethylene was discovered to have this property due to its complex microstructure [1, 2]. This consisted of nodules interconnected by fibrils. The microstructure was subsequently reproduced in ultra high molecular weight polyethylene (UHMWPE) (3), polypropylene (PP) (4), and nylon (5) using the powder processing techniques of compaction (6), sintering (7) and extrusion (8), resulting in cylindrical rods. The role of each stage was carefully studied to aid understanding, and this indicated that the compaction stage had the sole function of imparting structural integrity to the extrudate. So, the compaction stage was omitted in a systematic study [9, 10], and the result of this was that the extrudate produced was highly fibrillar and very auxetic but had low mechanical properties and low density. The complex microstructural formation for the cylinders was also studied (11), using the no-compaction route to fabricate the specimens, as these can be easily sectioned for detailed microstructural analysis. The main aim of the study was to understand the mechanisms of microstructure formation. The microstructure was produced by surface melting of the material leading to the formation of thick necks, which are drawn into fibrils. The latter was revealed to be not necessarily the result of extrusion, but could also result from particle movement prior to extrusion as the material is forced into close contact.

The cylindrical rods produced in these studies are ideal for laboratory scale testing, but are restricted in their size and geometry by the processing rig in which they are produced. Two methods can be considered to produce more useable and useful shapes, with a move toward applications. The most obvious is to develop routes to fabricate extruded products such as fibers or films, and this strand of the work has been the focus of concerted research at Bolton in recent years. In 2001, auxetic fibers were fabricated using a partial melt spinning technique, with PP, polyester, and nylon fibers being produced (12-14). A slight modification to the process (i.e., by changing the extrusion die to a slit die rather than a cylindrical die) allowed auxetic PP films also to be extruded (15). Once again, the mechanisms for auxetic behavior were based on the microstructure, but in these cases, from microstructural observations, there is evidence of a particulate/granular microstructure as well as low porosity. This is consistent with the starting powder being compacted and partially melted, leading to an interconnected structure (16), (17).

As a result of possessing a negative Poisson's ratio, certain properties are predicted from classical elasticity theory to be enhanced in auxetic materials (18). To date, experimental verification has been carried out on certain of these, with excellent results. For example, enhancements of up to four times over conventional properties have been found in indentation resistance (19-23), low velocity impact resistance (24), plane strain fracture toughness (19), and absorption at sonic and ultrasonic frequencies (25-28). These property enhancements occur across the whole spectrum of auxetic materials. To take the indentation resistance as an example, the work referred to above concerns such diverse materials as polymeric and metallic foams, carbon fiber composite laminates, and microporous polymers. The fabrication of a more useful form of auxetic polymer, that is, the fiber, has allowed more applications centered research into the resistance to fiber pullout in composite materials. Here, an auxetic fiber embedded in a matrix is predicted to lock into place (29). This concept was experimentally verified, and it has been shown that an auxetic fiber is up to three times more difficult to extract from a matrix than an equivalent conventional fiber (30).

However, even the production of films and fibers is still limiting the potential use of auxetic polymers, as their shape is decided by the extrusion die. The second approach, then, is to develop a route for producing auxetic polymers that does not include extrusion at all. The first part of this article (31) reports work to produce auxetic polymers in a more usable and useful form, which does not require an extrusion stage. The advantage of this is the flexibility that it should allow in forming more complex, applications-driven shapes. The process is summarized in Fig. 1 and consists of compaction followed by multiple sintering. The optimum results (31) have been achieved by compaction followed by double sintering, which has given a strain-dependent Poisson's ratio as low as [upsilon] = -0.32.

This article sets out to evaluate the mechanical properties of auxetic UHMWPE produced by compaction, followed by multiple sintering by measuring the flexural properties of the cylinders produced. In addition, the indentation resistance was also measured to see if any enhancements over conventionally processed UHMWPE could be observed as is the case for auxetic UHMWPE produced by the three-stage processing route. Part 1 of this article found a variation in the microstructure and Poisson's ratio along the specimen length and so for the indentation resistance, testing was carried out at various defined locations along the specimen length.

EXPERIMENTAL METHODS

For completeness, this section will detail the fabrication processes for the production of auxetic cylindrical rods by the multisintering route. For the indentation tests undertaken, the comparative material used was compression-molded plaques and the process for producing these is also detailed. This was to provide a direct comparison between UHMWPE produced by conventional means and the auxetic multisintered specimens produced here.

Fabrication of Auxetic Specimens

Specimens were fabricated as detailed in Fig. 1, with further details on the experimental procedures contained in Part 1 of this article (31). For completeness, this is summarized here. The first stage of the process to produce these specimens was to compact the UHMWPE powder as had previously been done for the three-stage processing route. The rig that is shown in Fig. 2 was heated to 110 [degrees]C and fitted with a blank die. The powder was poured into the barrel, which had a diameter of 15 mm and allowed to come to temperature for 10 min. A pressure of 7 kN was applied at a constant displacement rate linearly up to 140 mm/min and maintained for 20 min. The compacted rod was then removed from the extrusion rig and allowed to air cool. A range of cylinders were produced by subjecting them to single, double, triple, or quadruple sintering treatments, each at 160 degree C for 20 min. Between each sintering treatment, the specimens were removed from the barrel and allowed to air cool to room temperature outside barrel and without constraint.

Fabrication of Compression-Molded plaques for Comparative Testing

Plaques of UHMWPE were produced by compression molding using an in-house molder. Temperatures were set to 165 degree C and a pressure of 710 MPa was applied hydraulically for a period of 30 min by a manually operated pump. At the end of the loading period, the pressure was removed using a release valve and the plaque was allowed to cool within the tooling to room temperature. Plaques of dimensions 240 X 115 mm were produced in this manner.

Measurement of poisson's Ratio

Samples from each type of specimen were subjected to a series of single, discrete compression tests in the radial, r, direction. This was so that the Poisson's ratio, [V.sub.rz] could be determined using a photographic technique (1) developed in-house. For each material type, orientation and location, at least five tests were conducted.

Determination of the Flexural Properties of the Specimens

Simple three-point bend tests were conducted, in accordance with the British Standard (32) on specimens representing each of the four types of multisintered materials and for further comparison, on several samples that had been fabricated by the original three-stage processing route and those that had been compacted only. At least three examples of each material type, orientation, and location were tested and the results are presented as an average of the tests in Table 1. The purpose of these tests was to provide a semiquantitative indication of the strengths and stiffnesses of the multisintered materials with respect to those of specimens produced by alternative processing routes.

TABLE1. The variation in flexural properties with successive sintering treatments. Number of sintering Strength (MPa) Modulus (MPa) Strain to treatments failure (%) 0 5.4 [+ or -] 0.5 290 [+ or -] 10 1.9 [+ or -] 0.1 1 32.1 [+ or -] 0.2 113 [+ or -] 3 28.4 [+ or -] 0.5 2 32.0 [+ or -] 0.2 106 [+ or -] 2 30.1 [+ or -] 0.3 3 30 [+ or -] 0.2 106 [+ or -] 5 27 [+ or -] 1 4 31.0 [+ or -] 0.4 104 [+ or -] 7 30 [+ or -] 2 Three-stage 55.6 [+ or -] 0.5 360 [+ or -] 10 16 [+ or -] 1 route

The support points were set at a spacing equal to five times the specimen diameter and the crosshead speed used was 1 mm/min. A force/displacement curve was produced for each test and a typical example of this is shown in Fig. 3. From this, values of the flexural strength, S, modulus, M and strain to failure, [[epsilon].sub.f] were calculated using the following equations (33):

S=8PL / [pie[D.sup.3]]

M=[4PL.sup.3] / [3.pie]D.sup.4]]y

[epsilan.sub.f]=6Dy / [L.sup.3]

where P is the applied load, L the support separation, D the rod diameter, and y the maximum rod deflection.

Determination of Indentation Resistance

To study the variation along the specimen length of the indentation resistance, sections 4-mm thick were cut from each of the rods undergoing different thermal treatments in both the radial, r, and longitudinal, z, directions. For the former, the rods were thinned into bars with flat parallel sides in the z direction to a thickness of 3-4 mm and for the latter, the rods were sectioned radially into discs 3-4 mm thick in the r direction, as indicated in Fig. 4. These were then indented at a loading rate of 0.5 mm/min using a spherical steel indentor of diameter 5 mm driven by an Instron 1185 mechanical testing machine. Accuracy of the tests was ensured by recalibrating after each test had been carried out. Tests were conducted on three specimens of each type of material, location, and orientation. Furthermore, at least three tests were performed on each section whilst ensuring that the indentations were at least 10-mm apart so as to avoid edge effects from neighboring indentations. This gave a minimum of nine values of strain and corresponding hardness at each load for each type of material, location, and orientation.

A typical example of the force/displacement plot obtained during indentation testing is shown in Fig. 5. For a given depth of indentation into a material, the indentation resistance, H, is equal to the pressure required to cause the indentation, that is,

H=P / [[pie]a.sup.3]

Where a is the indentation radius of the surface. Since the radius of the indentor, R, is very much greater than the indentation depth, h, the approximation:

[a.sup.2]=2Rh

is used and so, H is then given by:

The values of P and h were then obtained from the force/displacement plots produced during the indentation tests (22) and H was calculated. Tests were conducted up to loads of 15, 25, 50, 75, 100, 150, and 200 N. For the test load of 15 N, the data were also analyzed at 5 and 10 N; for the test load of 25 N, analysis was carried out at 5, 10, and 20 N; and for the test load of 50 N at 10 and 25 N. For the remaining test loads considered, analysis was only carried out at the maximum applied load. An average value was calculated for each particular location, type of material, and orientation and these are presented in Tables 2-7. The reason for concentrating data analysis at the lower loads is that previous work (22) has indicated that this is where enhancements are most likely to occur, that is, where the material is still behaving in a primarily elastic fashion.

TABLE 2. Indentation resistance, H, in the z direction of specimen subjected to a single sintering treatment at the plunger (P), mid span (M), and die (D) end of the specimen. Applied Hardness at P Mardncss at M Hardness at D H of plaque load (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N) 5 4.4[+ or -]0.5 3.2[+ or -]0.4 1.2[+ or -]0.1 1.6[+ or -]0.1 10 6.0[+ or -]0.7 4.4[+ or -]0.5 2.4[+ or -]0.3 3.1[+ or -]0.3 15 7.2[+ or -]0.8 5.8[+ or -]0.6 4.0[+ or -]0.4 4.5[+ or -]0.4 25 9[ +or -]1 7.6[+ or -]0.8 5.6[+ or -]0.6 6.7[+ or -]0.6 50 11[+ or -]1 9[+ or -]1 6.6[+ or -]0.7 9.2[+ or -]0.8 75 12[+ or -]1 10[+ or -]1 7.8[+ or -]0.9 11[+ or -]1 100 14[+ or -]1 11[+ or -]1 9[+ or -]1 13[+ or -]1 150 15[+ or -]2 12[+ or -]1 9[+ or -]1 14[+ or -]1 200 16[+ or -]2 12[+ or -]1 9[+ or -]1 15[+ or -]1 For comparison, the value for a compression-molded plaque is also included. TABLE 3. Indentation resistance, H, in the r direction of specimen subjected to a single sintering treatment at the plunger (P), mid span (M), and die (D) end of the specimen. Applied Hardness at P Mardncss at M Hardness at D H of plaque load (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N) 5 4.4[+ or -]0.5 2.8[+ or -]0.3 1.2[+ or -]0.1 1.6[+ or -]0.1 10 5.6[+ or -]0.6 4.0[+ or -]0.4 2.4[+ or -]0.3 3.1[+ or -]0.3 15 5.8[+ or -]0.7 5.4[+ or -]0.6 3.6[+ or -]0.4 4.5[+ or -]0.4 25 9[+ or -]1 6.7[+ or -]0.7 4.7[+ or -]0.5 6.7[+ or -]0.6 50 11[+ or -]1 9[+ or -]1 5.6[+ or -]0.6 9.2[+ or -]0.8 75 14[+ or -]2 10[+ or -]1 6.0[+ or -]0.7 11[+ or -]1 100 15[+ or -]2 11[+ or -]1 6.5[+ or -]0.7 13[+ or -]1 150 16[+ or -]2 11[+ or -]1 7.2[+ or -]0.8 14[+ or -]1 200 19[+ or -]2 12[+ or -]1 7.7[+ or -]0.8 15[+ or -]1 For comparison, the value for a compression-molded plaque is also included. TABLE 4. Indentation resistance, H, in the z direction of specimen subjected to a double sintering treatment at the plunger (P), mid span (M), and die (D) end of the specimen. Applied Hardness at Hardness al Hardness al H of plaque load (N) P M D (N/[m.sup.2]) (N/[m.sup.2]) (N/[m.sup.2]) (N/[m.sup.2]) 5 5.0 [+ or -] 4.4 [+ or -] 1.2 [+ or -] 1.6 [+ or -] 0.8 0.7 0.2 0.1 10 7 [+ or -] 1 7 [+ or -] 1 2.2 [+ or -] 3.1 [+ or -] 0.4 0.3 15 7 [+ or -] 1 8 [+ or -] 1 3.8 [+ or -] 4.5 [+ or -] 0.6 0.4 25 9 [+ or -] 1 9 [+ or -] 2 5.6 [+ or -] 6.7 [+ or -] 0.9 0.6 50 11 [+ or -] 2 11 [+ or -] 2 7 [+ or -] 1 9.2 [+ or -] 0.8 75 13 [+ or -] 2 13 [+ or -] 2 8 [+ or -] 1 11 [+ or -] 1 100 15 [+ or -] 2 14 [+ or -] 2 9 [+ or -] 1 13 [+ or -] 1 150 16 [+ or -] 3 15 [+ or -] 3 9 [+ or -] 2 14 [+ or -] 1 200 17 [+ or -] 3 16 [+ or -] 3 11 [+ or -] 2 15 [+ or -] 1 For comparison, the value for a compression-molded plaque is also included. TABLE 5. Indentation resistance. H, in the r direction of specimen subjected to a double sintering treatment at the plunger (P), mid span (M), and die (D) end of the specimen. Applied Hardness at P Hardness at M Hardness at D H of plaque load (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N) 5 2.6[+ or -]0.4 4.0[+ or -]0.7 1.0[+ or -]0.2 1.6[+ or -]0.1 10 4.4[+ or -]0.7 5.2[+ or -]0.9 2.6[+ or -]0.4 3.1[+ or -]0.3 15 7[+ or -]1 7[+ or -]1 4.0[+ or -]0.7 4.5[+ or -]0.4 25 8[+ or -]1 8[+ or -]1 5.2[+ or -]0.9 6.7[+ or -]0.6 50 10[+ or -]2 9[+ or -]2 7[+ or -]1 9.2[+ or -]0.8 75 12[+ or -]2 11[+ or -]2 9[+ or -]1 11[+ or -]1 100 13[+ or -]2 12[+ or -]2 10[+ or -]2 13[+ or -]1 150 15[+ or -]3 13[+ or -]2 10[+ or -]2 14[+ or -]1 200 16[+ or -]3 13[+ or -]2 11[+ or -]2 15[+ or -]1 For comparison, the value for a compression-molded plaque is also included. TABLE 6. Indentation resistance, H, in the z direction of specimen subjected to four sintering treatments at the plunger (P), mid span (M), and die (D) end of the specimen. Applied Hardness at P Hardness at M Hardness at D H of plaque load (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N) 5 6.2[+ or -]0.9 6.0[+ or -]0.9 3.5[+ or -]0.5 1.6[+ or -]0.1 10 8[+ or -]1 7[+ or -]1 5.0[+ or -]0.7 3.1[+ or -]0.3 15 10[+ or -]1 9[+ or -]1 7[+ or -]1 4.5[+ or -]0.4 25 11[+ or -]2 10[+ or -]1 8[+ or -]1 6.7[+ or -]0.6 50 13[+ or -]2 13[+ or -]2 11[+ or -]2 9.2[+ or -]0.8 75 14[+ or -]2 14[+ or -]2 12[+ or -]2 11[+ or -]1 100 15[+ or -]2 16[+ or -]2 13[+ or -]2 13[+ or -]1 150 16[+ or -]2 17[+ or -]2 14[+ or -]2 14[+ or -]1 200 16[+ or -]2 18[+ or -]3 14[+ or -]2 15[+ or -]1 For comparison, the value for a compression-molded plaque is also included. TABLE 7. Indentation resistance. H, in the r direction of specimen subjected to four sintering treatments at the plunger (P). mid span (M), and die (D) end of the specimen. Applied Hardness at P Hardness at M Hardness at D H of plaque load (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N/[mm.sup.2]) (N) 5 4.8[+ or -]0.7 6.6[+ or -]0.9 4.0[+ or -]0.6 1.6[+ or -]0.1 10 6.0[+ or -]0.9 9[+ or -]1 5.3[+ or -]0.8 3.1[+ or -]0.3 15 8[+ or -]1 10[+ or -]1 7[+ or -]1 4.5[+ or -]0.4 25 10[+ or -]1 11[+ or -]2 8[+ or -]1 6.7[+ or -]0.6 50 12[+ or -]2 13[+ or -]2 11[+ or -]2 9.2[+ or -]0.8 75 14[+ or -]2 15[+ or -]2 12[+ or -]2 11[+ or -]1 100 16[+ or -]2 16[+ or -]2 14[+ or -]2 13[+ or -]1 150 17[+ or -]2 17[+ or -]2 15[+ or -]2 14[+ or -]1 200 18[+ or -]3 18[+ or -]3 16[+ or -]2 15[+ or -]1 For comparison, the value for a compression-molded plaque is also included.

RESULTS AND DISCUSSION

Measurement of Poisson's Ratio

For completeness, the results of the variation of the Poisson's ratio of the samples are considered here. A much fuller analysis has been undertaken and is reported in Part 1 of this article (31), along with a detailed microstructural examination plus density measurements.

After a single sintering treatment, it is possible that auxetic behavior has been induced on the plunger end of the samples but it is only found at very low strains and at levels where confidence in the measured values is not high. However, after double sintering, the mid span region of the specimen was found to be auxetic at strains of up to 5% before becoming slightly positive as the strain increased. Strain dependent values as low as v = -0.32 were measured here. The die and plunger ends, after double sintering, were not auxetic. After a third sintering treatment, no significant differences were observed and after a fourth sintering treatment, only the die end of the specimen gave any indication of auxetic behavior, but again at too low a strain to be reported with confidence. The regions of auxetic behavior in each case were found to coincide with a nodule-fibril microstructure in the cylinders albeit with shorter fibrils than those found in the microstructure produced in the compaction, sintering, and extrusion three-stage processing route. Thus, the existence of a negative Poisson's ratio was consistent with the same mechanisms occurring to produce the negative Poisson's ratio as for cylinders produced by compaction, sintering, and extrusion, that is, translation of the nodules caused by the fibrils. Simple geometric modeling of the microstructure has been undertaken, giving excellent agreement with experimental results. This has allowed a theoretical curve to be constructed so that if the strain applied to the specimen at any given point is known, the Poisson's ratio can be predicted (31).

Flexural Properties of the Specimens

The results of the three-point bend tests conducted to reveal the flexural strength, modulus, and strain to failure of the compacted and multisintered specimens are shown in Table 1. The most dramatic changes to the flexural properties occur on the first sintering treatment. Here, the strength rises from 5.4[+ or -] 0.5 MPa to 32.1 [+ or -] 0.2 MPa and the strain to failure increases from (1.9 [+ or -] 0.1)% to (28.4 [+ or -] 0.5)%, while the modulus drops from 290 [+ or -] 10 MPa to 113 [+ or -] 3 MPa. On successive sintering treatments, the changes to these properties are, by comparison, practically negligible.

This fits in well with the variations in the overall density of the specimens that were reported in detail in Part 1 of this article (31). A purely compacted rod has a density of 790 [+ or -] 30 kg/[m.sup.3]. This increases to 890 [+ or -] 30 kg/[m.sup.3] with the first sintering treatment. After this, the density continues to increase, but at a much reduced rate until the fourth sintering treatment, where a density of 932 [+ or -] 4 kg/[m.sup.3] is achieved. This represents a density packing factor of 98%, obtained by dividing the density measured by the maximum density quoted for this material of 950 kg/[m.sup.3] (31).

It does appear, thus, that the sintering treatments only and not any applied pressure are causing a densification of the material and this is increasing the mechanical properties. However, it should be noted that the three-stage processing route produced much stronger specimens, with the strength being 55.6 [+ or -] 0.5 MPa, as can be seen in Table 1.

Indentation Resistance Tests

Tables 2-7 show the average indentation resistance at loads of 5, 10, 15, 25, 50, 75, 150, and 200 N on a conventional, compression-molded UHMWPE plaque (fabricated as described above) and on each of the plunger end, mid span, and die end sections of the single, double, and quadruple sintered specimens in both the z and r directions.

The results of the tests performed on the plunger end, mid span, and die end sections of the single sintered material are shown in Tables 2 and 3. It can be seen that in both directions, the single sintered material is always harder than the compression-molded plaque at the plunger end. The mid span material is harder at the lower loads and the die end material always has the lowest indentation resistance, which is particularly marked at the higher loads, for example, at 200 N, H = 9 [+ or -] 1 N/[mm.sup.2] in the z direction for the die end material and H = 15 [+ or -] 1 N/[mm.sup.2] for the compression-molded material.

Tables 4 and 5 show the results of tests performed on the double-sintered specimens. As expected, the hardness values rise in comparison with these obtained after a single sintering treatment. In each orientation, the hardness of both the plunger end and mid span sections is higher than that of the compression-molded plaque. In the z direction, it can also be seen that the hardness of the plunger end and mid span regions after a second sintering treatment were found to be comparable except at the highest loads, where the plunger end material is the hardest. In the r-direction, however, the mid span material was found to be harder than the plunger end material at lower loads. The hardness of the die end material remains substantially lower than that of the compression-molded plaque, particularly at the higher loads.

Tables 6 and 7 show that after the fourth sintering treatment, the rods become more consistent with respect to hardness values. The rise in hardness is particularly marked in the die end material, where it is now comparable to the compression-molded plaque at the higher loads and is considerably higher at the lower loads (e.g., at 5 N, the die end has an indentation resistance of H = 4 [+ or -] 0.6 N/m]m.sup.2] with the compression-molded plaque having H = 1.6 [+ or -] 0.1 N/[[[mm.sup.2] in the r direction).

The most interesting of all the results presented here are those at low loads in the r direction of the double-sintered material. Here, previous work reported in Part 1 of this article (31) has shown that the processing route generates a nodule-fibril microstructure with a resulting strain dependent negative Poisson's ratio as low as [upsilon] = -0.32 in the mid span region. At the strains employed in the indentation test, this corresponds to Poisson's ratios of the value of [upsilon] = -0.23 at an indentation load of 5 N, [upsilon] = -0.19 at an indentation load of 10 N, [upsilon] = -0.17 at an indentation load of 15 N, and [upsilon] = -0.11 at an indentation load of 20 N. At the lowest applied loads, that is, 5 N where the Poisson's ratio is most negative, the hardness in the r direction at the mid span region is 2.5 times that of the compression-molded plaque and 1.5 times that of the plunger end. The density variations along the specimen lengths were determined in Part 1 of this article and are reported here in Table 8 for completeness. It can be seen that the density in the mid span region is comparable to the compression-molded plaque (i.e., 910 kg/[m.sup.3] compared with 908 kg/[m.sup.3]) and less than that of the plunger end material (i.e., 940 kg/[m.sup.3]), so this is clearly not the cause of this increase in hardness. This is a significant finding and agrees with previous work (22) on the auxetic cylinders produced by compaction, sintering, and extrusion. In the latter case, enhanced hardness at low loads of up to three times that of compression-molded UHMWPE of equivalent modulus and density were measured. The mechanism involved in the response was a local densifica-tion of the microstructure with the fibrils pulling the nodular material elastically under the indentor.

TABLE 8. The variation in density of the plunger (P), mid span (M), and die (D) end of the specimen with successive sintering treatments. No. of sintering treatments Position Density (kg/[m.sup.3]) None P 870 [+ or -] 40 M 780 [+ or -] 30 D 720 [+ or -] 30 1 P 920 [+ or -] 30 M 880 [+ or -] 30 D 860 [+ or -] 30 2 P 940 [+ or -] 20 M 910 [+ or -] 20 D 880 [+ or -] 20 3 P 942 [+ or -] 9 M 932 [+ or -] 9 D 897 [+ or -] 9 4 P 947 [+ or -] 4 M 950 [+ or -] 4 D 900 [+ or -] 4

It is also interesting to note that a nodule-fibril micro-structure was observed after a single sintering treatment at the plunger end and in the quadruple sintered specimen at the die end. There was some evidence for auxetic behavior, but at such small strains that confidence in asserting this was low. The die end of the samples remains much less dense than the rest of the specimen, even after four sintering treatments, so it is not surprising that this remains the least hard part of the specimen. However, it is at a comparable density to the compression- molded plaque and is, as noted above, harder than this possibly due to the microstructure observed. As far as the plunger end on the first sintering treatment is concerned, it is more difficult to assign the increase in hardness to an auxetic effect as this is a very dense region of the specimen even after just one sintering treatment.

It should be noted at this stage that all the work reported here has been primarily concerned with the manufacture of cylinders. However, the novelty of this work is that a route is proposed that can be easily adapted to the production of far more complex and useful shapes. Compaction of powder followed by double sintering has generated a nodule-fibril microstructure without the need for an extrusion stage. This has resulted in a strain dependent negative Poisson's ratio as low as [upsilon] = -0.32. The indentation resistance of the auxetic material produced here is enhanced at low loads, as it was for the three-stage extrusion route, so no loss in enhancement appears to result from replacing the extrusion stage with multiple sintering.

CONCLUSIONS

A novel processing route based on the powder processing techniques of compaction and multiple sintering has been developed that produces an auxetic form of UHMWPE. The best results to date have been produced by compaction followed by sintering at 160[degrees]C. air cooling to room temperature, and sintering again (i.e., a double sintering treatment) at 160[degrees]C. This has generated a nodule-fibril micro struct are and a strain-dependent negative Poisson's ratio as low as [upsilon] = -0.32 at low strains.

The cylinders produced show a large increase in flex-ural properties with a single sintering treatment accompanied by an increase in the density (31). Both continue to increase but at a reduced rate through successive sintering treatments, up to four.

The indentation resistance of the auxetic double-sintered material was measured and was found to be enhanced by a factor of up to 2.5 times at low loads when compared to conventional compression-molded UHMWPE. The effect is directly linked to the auxetic character of the specimen, as the more fibrillar regions are less dense but as expected still display a greater resistance to indentation.

This processing route marks the first stage to produce a greater range of more useful shapes rather than the cylinders, fibers, and films produced to date. The size and shape of these is severely limited by the need previously to extrude, that is, the dimensions of the extrusion barrel and die govern those of the end product. Removing the need to extrude should overcome this problem. Further development of the compaction with multiple sintering route for more complex shapes is the next step in this work.

ACKNOWLEDGMENTS

RSW thanks Prof. Wesley Cantwell for his assistance with this project.

REFERENCES

(1.) B.D. Caddock and K.E. Evans, J. Phys. D Appl. Phys,, 22. 1877 (1989).

(2.) K.E. Evans and B.D. Caddock, J. Phys. D Appl. Phys., 22. 1883 (1989).

(3.) K.L. Alderson and K.E. Evans. Polymer. 33. 4435 (1992).

(4.) A.P. Pickles, K.L. Alderson. and K.E. Evans, Polym. Eng. Sci., 36, 636 (1996).

(5.) K.L. Alderson, A. Alderson. R.S. Webber, and K.E. Evans, J. Mater. Sci. Lett., 17, 1415 (1998).

(6.) A.P. Pickles, R.S. Webber. K.L. Alderson. P.J. Neale, and K.E. Evans, J Mater. Sci., 30, 4059 (1995).

(7.) K.L. Alderson. A.P. Kettle. P.J. Neale, A.P. Pickles, and K.E. Evans,./. Mater. Sei.. 30. 4069 (1995).

(8.) P.J. Neale, A.P. Pickles. K.L. Alderson, and K.E. Evans, J. Mater. Sci., 30, 4087 (1995).

(9.) R.S. Webber, K.L. Alderson, and K.E. Evans. Polym. Eng. Sci., 40(8), 1894 (2000).

(10.) K.L. Alderson. R.S. Webber, and K.E. Evans. Polym. Eng. Sci., 40(8). 1064 (2000).

(11.) K.L. Alderson. R.S. Webber, and K.E. Evans, Phys. Status Solidi B, 244(3), 828 (2007).

(12.) K.L. Alderson, A. Alderson, G. Smart, V.R. Simkins. and P.J. Davies, Plast. Rubber Compos.. 31(8), 344 (2002).

(13.) N. Ravirala. A. Alderson, K.L. Alderson, and P.J. Davies, Phys. Status Solidi B, 242(3), 653 (2005).

(14.) N. Ravirala, K.L. Alderson. P.J. Davies, V.R. Simkins. and A. Alderson, Text. Res. J., 76(7), 540 (2006).

(15.) N. Ravirala, A. Alderson, K.L. Alderson. and P.J. Davies, Polym. Eng. Set.. 45(4), 517 (2005).

(16.) N. Ravirala, A. Alderson, and K.L. Alderson. J. Mater Set. 42, 7433 (2007).

(17.) K.W. Wojciechowski, J. Phys. A Math. Gen.. 36. 11765 (2003).

(18.) R.S. Lakes, Science. 235, 1038 (1987).

(19.) J.P. Donoghue and K.E. Evans, "Composite Laminates with Enhanced Indentation and Fracture Resistance Dae to Negative Poisson's Ratios," in Proceedings of ICC M8 SAMPE, S.W. Tsai and G.S. Springer, Eds., Sample, Covina, CA, 2-k-1 (1991).

(20.) N. Chan and K.E. Evans,./. Cell. Plast., 34. 231 (1998).

(21.) R.S. Lakes and K.J. Elms, J. Compos. Mater.. 27. 1 193 (1993).

(22.) K.L. Alderson, A.F. Fitzgerald, and K.E. Evans. J. Mater. Sci., 35, 4039 (2000).

(23.) V. Coenen. K. Alderson, P. Myler, and K. Holmes. "The Indentation Response of Auxetic Composite Laminates." in Proceedings of the 6th International Conference on Deformation and Fracture of Composites. Manchester, UK, April 4-5 (2001).

(24.)V. Coenen, K. Alderson, P. Myler. and K. Holmes, "The Low Velocity Impact Response of Auxetic Composite Laminates", in Proceedings of the 8th International Conference on Composites Engineering, Tenerife, Spain, August 5-11 (2001).

(25.) C.P. Chen and R.S. Lakes, Cell. Polym., 8. 343 (1989).

(26.) C.P. Chen and R.S. Lakes, J. Mater. Sci., 28. 4288 (1993).

(27.) B. Howell, P. Prendergast, and L. Hansen. Appl. Acoust., 43, 141 (1994).

(28.) K.L, Alderson, R.S. Webber, U.F. Mohammed. E. Murphy, and K.E. Evans. Appl. Acoust., 43(1), 23 (1997).

(29.) K.E. Evans, Chem. Ind., 20. 654 (1990).

(30.) V.R. Simkins, A. Alderson, P.J. Davies. and K.L, Alderson, J. Mater. Sci., 40, 4355 (2005).

(31.) K.L. Alderson, R.S. Webber. A.P. Kettle, and K.E. Evans. Polym. Eng. Sci., 45(4). 568 (2005).

(32.) BS 2782: Part 3: Methods of Testing Plastics. Mechanical Properties. Method 341A. Determination of Apparent Inter-laminar Shear Strength of Reinforced Plastics (1977).

(33.) H.S. Loveless, "Flexural Test," in Testing of Polymers, Vol. 2, J.V. Schmitz, Ed., Interscience Publishers. New York. 321 (1966).

Correspondence to: Kim Alderson; e-mail: kal@bolton.ac.uk

Contract grant sponsor: EPSRC.

DOI 10.1002/pen.21110

Published online in Wiley InterScience (www.interscience.wiley.com).

[c] 2008 Society of Plastics Engineers

R.S. Webber, (1) K.L. Alderson, (2) K.E. Evans (3)

(1) Department of Materials Science and Engineering, University of Liverpool, Brownlow Hill, Liverpool, UK

(2) Centre for Materials Research and Innovation, The University of Bolton, Bolton, UK

(3) School of Engineering, Computer Science and Mathematics, University of Exeter, Exeter, UK

Printer friendly Cite/link Email Feedback | |

Author: | Webber, R.S; Alderson, K.L; Evans, K.E. |
---|---|

Publication: | Polymer Engineering and Science |

Article Type: | Technical report |

Geographic Code: | 1USA |

Date: | Jul 1, 2008 |

Words: | 6013 |

Previous Article: | Temperature dependence of electrical resistivity in carbon nanofiber/unsaturated polyester nanocomposites. |

Next Article: | Reactive compatibilization of biodegradable poly(lactic acid)/poly([epsilon]-caprolactone) blends with reactive processing agents. |

Topics: |