Solid-state extrusion of chain-extended polyethylene

Solid-state Extrusion of Chain-Extended Pol ye thylene

HOE HIN CHUAH and ROGER S. PORTER, Polymer Science and Engineering Department, Materials Research Laboratory, University of

Massachusetts, Amherst, Massachusetts 01003

Synopsis

Billets of chainextended polyethylene were prepared from Alathon 7050 (M, 59,000, M, 19,OOO) in an Instron capillary rheometer by crystallization at a constant pressure of 460 MPa, at a series of temperatures from 198 to 221°C corresponding to varying degrees of undercooling. This gives chain-extended morphologies with a range of crystallinities and lamellar thick- nesses. The billets were then solid-state extruded at 100°C through a conical die with 20" entrance angle up to an extrusion draw ratio 23.4. Thermal behavior was studied with differential scanning calorimetry. The orientation function measured by wide-angle x-ray diffraction showed higher orientation at equivalent draw ratio when the initial billets were crystallized at lower temperatures. Drawing efficiency, defined as the ratio of molecular draw ratio (from shrinkage) to extrusion draw ratio correspondingly increases, reaching a maximum of 0.71 in our solid-state extrusion. These studies show that highly chainextended polyethylene, i.e., with few chain entanglements, draws poorly. Drawability was improved by increasing chain entanglements by lowering the crystallization temperature. Electron micrographs of fracture surface replicas of extrudates revealed the coexistence of undeformed, tilted, partially drawn lamellae and fibrillar structure consistent with the change of morphologies in Peterlin's model of plastic deformation.

INTRODUCTION The drawing of polyethylene is a subject of continuing interest. Over the

last decade, rapid advances in orienting flexiblechain polymers to achieve high modulus and strength have resulted from several new drawing tech- niques such as solid-state extrusion,' fibrillar crystal growing,2 and gel fiber d r a ~ i n g . ~ Substantial work has been done in this and other laboratories on the solid-state extrusion of polyethylene, but there have been few reports on the extrusion drawing of polyethylene from a chain-extended morphol- om.

When polyethylene is melt crystallized at pressure greater than 350 MPa and at high temperature, chain-extended morphologies are ~bta ined .~ High- temperature crystallization is a result of the increase in melting point with increasing pressure, following the Clausius-Clapeyron equation at ca. 2WC/ 1000 atm for p~lyethylene.~ Depending on crystallization conditions, crystal thickness can be varied from a few hundred nanometers to few microns. Generally some chain folding in crystals is still observed to occur in chain- extended lamellae, with the exception of low-molecular-weight polyethylene, which can be crystallized into fully extended chains. By definition chain-extended lamellae have been proposed as those having thickness of more than 200 nm.6 This thickness corresponds to a polyethylene molecular

Journal of Polymer Science: Polymer Physics Edition, Vol. 22, 1353-1365 (1984) @ 1984 John Wiley & Sons, Inc. CCC 00981273/84/081353-13$04.00

1354 CHUAH AND PORTER

weight of ca. 20,000 and to conditions under which crystal surfaces and chain ends have negligible effect on melting behavior. Crystallization at an intermediate pressure range, 200-350 MPa, gives a mixture of chain-extended and chain-folded morphologies, while at pressure below 200 MPa, only chain-folded lamellae are ~b ta ined .~

Mead and Porter' crystallized high-density polyethylene at 490 MPa and at temperatures between 134 and 220°C. The resulting billets were solid- state extruded at 120°C to a high draw ratio, >30. Although the initial morphology was not fully examined, differential scanning calorimetry showed that some of the billets gave very high melting points, presumably due to the presence of chain-extended crystals. Lupton and Regestel.8 solid- state-extruded ultra high-molecular-weight (ca. 2,000,000) chain-extended polyethylene (CEPE) into a mold. The product was rigid and tough compared to the brittle CEPE crystallized from lower-molecular-weight polyethylenes used in commercial molding. Because of its brittleness, tensile drawing of CEPE failed at draw ratios below 6.9 The solid-state extrusion technique, with the compression and lateral constraints imposed on the billet by the capillary wall, allows drawing to a higher ratio, >23.4, without fracture.

As lamellar thickness of this morphology is comparable to the molecular chain length, chain entanglements must be drastically reduced. This initial morphology is also similar to the chain-extended matrix in drawn fibril models.1° It is therefore of interest to study its drawing behavior.

EXPERIMENTAL An Instron capillary rheometer was used as a temperature-pressure vessel

by blocking the exit for crystallization. High-density polyethylene pellets (Alathon 7050, M, 59,000, M,, 19,000) were compacted in the rheometer under cyclic loading of ca. 200 MPa at 120°C for 30 min, then melted and brought to the desired crystallization temperature T, after release of the pressure. Crystallization was initiated by raising the pressure from atmos- pheric to 460 MPa in about 80 s. The increase of pressure caused compressive heating, raising the temperature of the system initially by ca. 2°C. No correction was made as T, was rapidly reestablished. Crystallization conditions were maintained for 3 h followed by cooling in an air jet while the pressure was held constant. The initial cooling rate was ca. 4Wmin. At the constant pressure of 460 MPa, the four T,'s chosen were 221, 216, 207, and 198"C, corresponding to a range of undercoolings, to give billets of different densities and crystallinities (Table I).

The resultant CEPE billets were solid-state extruded at 100°C in the rheometer using a conical brass die with entrance angle of 20" at a constant plunger speedof 0.05 cm/min. Details of this extrusion technique have been described elsewhere." The initial 20 cm of extrudate was discarded to avoid imperfection and to ensure steady-state extrusion conditions. The extrusion draw ratio EDR was measured from the reduction of cross section before and after extrusion.

Densities were measured in a density-gradient column using a mixture of ethanol and 1% glycerol solution in water at 23°C. The melting behavior was studied using a Perkin-Elmer DSC-2 differential scanning calorimeter

TABL

E I

Cry

stal

lizat

ion

Con

ditio

ns a

nd C

hara

cter

izat

ion o

f In

itial

Bill

et M

orph

olog

ies

for

the

Hig

h-D

ensi

ty P

olye

thyl

ene

Cry

stal

lizat

ion

Cry

stal

linity

C

ryst

allin

ity

Lam

ella

r th

ickn

ess

from

AH

f di

stri

butio

n L

, (nm

) T

empe

ratu

re

Pres

sure

D

ensi

ty

AH

f B

illet

(“

C)

(MPa

) (g

/cm

3)

(cal

l@

from

den

sity

1 22

1 460

0.99

0 68

.5

0.94

2

216

460

0.98

9 67

.5

0.93

3

207

460

0.98

2 63

.3

0.89

4

198

460

0.98

1 61

.3

0.89

0.99

0.

98

0.92

0.

89

480

350

270

280

Y


at a heating rate of 2.5”C/min. A constant sample weight of ca. 4 mg was used throughout.

The molecular draw ratio (MDR) was measured by a thermal shrinkage experiment,12 conducted by shaving extrudates to a thickness < 0.5 mm, followed by immersion in a silicon oil bath at 180°C for 2 min.

MDR is defined as

MDR = AL/L, + 1

where A L is the change in length before and after shrinkage and L, is the original length before extrusion.

Orientation was measured by wide-angle x-ray diffraction (WAXD) from scanning of the 110 and 200 diffractions at varying azimuthal angle using a Siemens D-500 x-ray diffractometer equipped with a pulse-height scintil- lation counter, operating at 27 mA and 45 kV with Ni-filtered Cu Ka radiation. Wilchinsky’s method13 was used to evaluate the orientation pa- rameter < cosz+ > j , where + is the angle between diffraction plane normal and the uniaxial draw axis; j = a, b,c are the crystallographic axes of the polyethylene unit cell. The crystal orientation functions f, along the three axes were then calculated using the HermanStein equation,

f, = 1/2 (3<cos2+> j -1)

A JEOL 100-kV transmission electron microscope was used to examine the fracture surfaces. The billets and extrudates were fractured in the center along the axial direction under liquid nitrogen. The fracture surfaces were shadowed with CPt at 30” and coated with C at 90” angles. The replicas were stripped with polyacrylic acid, which was subsequently dissolved off in water. Micrographs were taken to measure the lamellar thickness distribution of the starting billets. A statistical method similar to that described by Rees and Bassett14 was used to compute the number-average thickness.

RESULTS AND DISCUSSION Table I lists the crystallization conditions and the characterization of the

initial billet morphologies. The billets are listed in order of decreasing T,. Crystallinities as measured from densities are high, 0.94, when T, is 221°C and decreases to 0.89 with decreasing T,. The heat of fusion, however, gives consistently higher crystallinities.

The examination of fracture surfaces with the electron microscope had a dual purpose. It revealed the characteristic feature of a chain-extended morphology by exposing fractured (hk0) surfaces, showing long, striated strips of thick lamellae. This confirmed that the initial billets were indeed of a chain-extended morphology. From the large number of micrographs taken, quantitative measurements of the lamellar thickness distribution were made. The number-average thickness shown in the last column of Table I decreases from 480 to 280 nm with decreasing T,. The lamellae also become more fragmented and disordered, resulting in lower crystallinity.

SOLID-STATE EXTRUSION 1357

A parallel trend of forming more disordered lamellae and lower crystallinity is also observed when a polyethylene of higher molecular weight is cry~tallized'~; the fracture surfaces usually showed large amounts of fibrous overlay which are pulled out from the lamellae during the fracture process. The extent of fibrous overlay was attributed to the high concentration of tie molecules between lamellae. The micrographs of our billets showed a clean fracture, suggesting few interlamellar tie molecules. This conclusion is also reached from the consideration of molecular chain lengths and the lamellar thickness distribution for Alathon 7050, the full chain lengths corresponding to M, and M,, are 540 and 170 nm, respectively. These values are of the same magnitude as the measured number-average lamellar thickness. At the highest T, of 221"C, this arithmetic would mean there is only a single fold per chain on the average; the number increases toward two with decreasing T, A large portion of the chains must be fully extended and thicker lamellae could have more than one fully extended chain to make up the thickness. Conversely, long chains crystallized into thinner lamellae could have more folds. Interlamellar connection is related to the number of folds possible.I6 A chain can either fold back or incorporate into a neighboring lamella, forming interlamellar tie molecules. The small number of folds per chain suggests that these thick lamellae exist as near individual entities with few interlamellar connections. As T, was decreased,

~

BILLET 1

130 140 TMPERATWK "C

Fig. 1. Melting endotherms of initial billets and their extrudates in increasing draw ratio; 22rc.


n 1 216°C BILLET 2

I I 130 140

TEMPERATURE 'C Fig. 2. Melting endotherms of initial billets and their extrudates in increasing draw ratio;

2 m c .

Tc 207'C BILLET 3 n

11 4

91

72

41

I I L o

120 130 140 TEMPERATURE "C

Fig. 3. Melting endotherms of initial billets and their extrudates in increasing draw ratio; 207°C.


the “amorphous” content increased, as loci where chain entanglements may likely exist within the fold loops between lamellae. These trapped entanglements have a profound effect on the drawing behavior of CEPE, as dis- cussed below.

Thermal analysis of the extrudates is shown in Figures 1-3. CEPE is known to superheat at high heating rates, resulting in higher apparent peak melting points. The small sample size used and the low heating rate of 2.5”C/min minimize this effect. Three or four overlapping endotherms are generally observed for the undrawn billets. With billets 1 and 2 (Figs. 1 and 21, the highest peak melting point occurred at 141”C, corresponding to the melting of large, chain-extended lamellae. Much smaller endotherms are noted at lower temperatures of 129 and 134“C, and these have been attributed to chain-folded fractions17 or low-molecular-weight polyethylene fractions that may have segregated during crystallization and subsequently crystallized into chain-extended lamellae during cooling.la It has been reported that fractions with MW < lo4 are rejected in the growth process.lg The billet described in Figure 3 was crystallized at a lower temperature: it also shows three endotherms, but the main melting peak centers at 137°C with a high-temperature shoulder at 140°C and a small endotherm at 131°C. Maeda and Kanetsuna,20 using nitric acid etching followed by gel permea- tion chromatography, detected two fractions of chain-extended lamellae, one fraction, which they termed “ordinary chain-extended” crystals, having a thickness one-third to one-half that of the other fraction of “highly chain- extended” crystals. The lamellar thickness distribution of the billet crystallized at 207”C, as measured by electron microscopy, showed a bimodal distribution, with 3% of lamellae having a thickness of ca. 1 pm. The majority are between 150 and 400 nm thick. The endotherm at 137°C is thus likely due to the ordinary chain-extended crystals, while the high-

1 5 10 15 EXTRUSION DRAW RATIO

Fig. 4. Change of crystallinity as a function of extrusion draw ratio as measured by DSC for the four billets.


temperature shoulder is due to the small fraction of very thick lamellae. The use of the terms “ordinary” and “highly chainextended” is arbitrary but is useful in discussing the change of the different endotherms with drawing.

Figure 1 shows the change of melting behavior on extrusion drawing of billet 1. The highest peak melting point remains virtually unchanged, decreasing by < 1°C with increasing EDR up to 16.6. The lower-temperature endothermic maxima decrease in magnitude and merge to become a single smooth endotherm at high draw. The slight decrease in melting point indicates that there is some lamellar disruption during drawing. However, the disruption is not extensive and substantial amounts of lamellae have been observed undeformed in electron micrographs of fracture surface replicas up to EDR 7.4. At higher drawing, isolated, individual lamellae could be found embedded in fibrillar structure, which counteracts the drop in melting point of the disrupted lamellae to minimum.

The extrudates drawn from billet 2 showed different thermal behavior (Fig. 2). At EDR 4.1, the highest melting peak decreases by 3°C and remains almost constant on further drawing. The low-temperature endotherms are replaced by a peak at 136°C that increases in magnitude with drawing, while the high-temperature peak reduces to a shoulder at EDR 16.4. Here the decrease in the high melting peak is associated with the destruction of chain-extended lamellae and the formation of the fibrillar structure indi- cated by the simultaneous appearance of the 136°C endotherm. The associated crystallinity decreasing on draw is ca. 7% (Fig. 4).

Drawing of billets 3 and 4 with progressively more disorder and smaller lamellae shows similar thermal behavior. Only the thermogram of billet 3 is shown (Fig. 31, which shows a rapid reduction in both the high- and low- temperature endotherms, leaving a single peak at EDR 2 9.1. This occurs earlier at EDR 7.2 for billet 4. The peak melting point at 138°C for the “ordinary” chain-extended lamellae decreases initially and increases again, reaching 137°C as an EDR of 16.4 is reached. Billet 4 shows better drawability and can be readily drawn to an EDR 23.4 without noticeable fracture.

In the drawing of conventional chain-folded polyethylene, a single endotherm is usually observed and the peak melting point increases with drawing, reaching a plateau of 139°C at high EDR, contrary to the observed decrease on drawing of CEPE. This opposite trend of decreasing melting point is to be expected as chain-extended morphology is the one approaching the thermodynamic equilibrium state, and the equilibrium melting point. Any disruption of the lamellae will likely result only in decrease of the melting point. It can be seen in Figures 1-3 that there is a tendency to form a single endotherm and for the melting peak to approach 137°C at high drawing. Although a high-temperature shoulder is still observed at EDR 16.4 for billet 2, its magnitude decreases and a single endotherm should be observed if higher drawing can be achieved. The approach to single endotherm and plateau melting point must therefore indicate an approach to a common morphology at high drawing. This can be an indication of complete transformation of lamellae to a fibrillar structure, which occurs at an EDR of 7 to >16 for the four different initial-morphology CEPEs studied. Transformation is not complete for billets 1 and 2 before fracture occurs.


The DSC measurements were made on samples that are free to shrink on melting. Relaxation and reorganization during partial melting of unrestrained samples causes narrowing of the peaks.21 The entropic shrinkage has an exothermic component and can account for the reduction of measured heat of fusion by <3% in unrestrained samples. Therefore the crystallinities from drawn samples shown in Figure 4 are in actuality a few percent higher.

Orientation of the extrudates is evaluated by WAXD and expressed by the crystal orientation function f (unity for perfect orientation, zero for random orientation, and - 0.5 for perpendicular orientation). Figure 5 shows the changes off for the three crystallographic axes a, b, and c with EDR for drawn billets 1 and 4, which have widely different lamellar thickness distribution in the series. We see that f , increases slowly with drawing for billet 1 but very rapidly for billet 4, in a way comparable to the development of orientation in the drawing of chain-folded polyethylene.22 Similar behavior is observed in the development of the a- and &axis orientations; a-axes orient faster than b-axes. This difference is clearly shown in the drawing of billet 1, and persists even at EDR 16.4.

The effect of initial morphology on drawing is further shown by comparing f , for the four billets at the same EDR of 16.4 (Fig. 5); f , is lowest for billet 1 and increases as the crystallization temperature of the initial billet is decreased.

MDR from thermal shrinkage is an alternative way of showing the effect of initial morphology on drawing. When the sample is heated above the PE melting point at a high rate, the molecular extension produced by drawing can be elastically recovered. The amount of recovery as compared to the initial length gives MDR. In defining draw efficiency as the ratio of MDR

I 15 I 20 I 25

EXTRUSION DRAW RATIO

Fig. 5. Change of crystal orientation function ffor the three crystallographic axes a, b, and c for drawn billets 1 ( T, = 221°C) (A, 0, 0) and 4 (T, = 198oC) (A, 0, m); f i of billets 2 (T, = 216°C (v) and 3 (T, = 207°C) (v) drawn to EDR 16.4 are inserted for comparison.


to EDR, Figure 6 shows that it increased as the crystallization temperature of initial billets was decreased, reaching a maximum of 0.71 in solid-state extrusion.

The above results show that highly chain-extended morphology orients ineffectively on drawing. The drawability and drawing efficiency appear to be aided by increasing chain entanglements in the amorphous region in this case. The effect of entanglements on drawing has received much at- tention. Their reduction enhances chain mobility and can result in highly drawn PE as in gel fiber drawing: in which the entanglement density is reduced by dilution with solvent. However, the reduction is not without limit. Severe reduction results in an insufficient transient networkz3 that is required to effectively transfer stresses during deformation. In melt- crystallized PE, the entanglement density is high. Fischerz4 studied the drawing of melt-crystallized low- and medium-molecular-weight PE and found that higher drawing was obtained for samples crystallized at the higher pressure of 2.85 kbar. The increase in drawability was attributed to the reduction in entanglements as a more chain-extended crystal structure is approached.

In the present study, our approach is from the top down. Highly crystalline CEPE has lamellar thickness comparable to the molecular chain length, and therefore there are few interlamellar links and trapped entanglements. In the extreme, this approaches the situation of a single-crystal mat. For billet 1 crystallized at the highest temperature of 221"C, the drastic reduction in entanglements in the amorphous phase aids slipping of lamellae past each other as whole entities during drawing, resulting in poor orientation. The ductility, despite the few entanglements, must be due in part to the compression and constraint imposed by the solid-state extrusion process, compacting the lamellae, shearing and forcing chains to unravel on passing through the conical die during extrusion. For decreasing T, and a consequent increase of initial amorphous content and entanglements, as in billets 3 and 4, drawability and draw efficiency increase. An increase of

Q

3

c $15 - 4 B 510

- _I 3

1 - I I I I

1 5 10 15 2 0 25 EXTRUSION DRAW RATIO

Fig. 6. Molecular draw ratio (MDR) versus extrusion draw ratio (EDR) showing effect of initial morphology on efficiency of drawing (MDR/EDR). T,: ([7)221"C, &216T, (0)207"C, W198"C.


amorphous content by 5% and lowering of the number-average lamellar thickness by half markedly alters drawing behavior, almost doubling the drawing efficiency.

Figure 7 is an electron micrograph of a fracture surface replica of billet 1 drawn to EDR 4.9. It shows the coexistence of undrawn and partially drawn lamellae, and fibrillar structure. This observation provides an ad- ditional clue to the plastic deformation mechanism. A now widely accepted mechanism for crystalline polymers was proposed by Peterlin.= It consists of three deformation stages involving shear, rotation, chain tilt, and slip in lamellae before necking; the transformation of lamellae into bundles of microfibrils; and the sliding of microfibrils and fibrils past each other in the strain hardening step. The deformation in solid-state extrusion is ad- equately described by this model with the neck in the die region and with substantial admixture of shear fracture of stacks of lamellae.=

Chain-extended lamellae are extremely large and easily observed by electron microscope; they show characteristic striations parallel to the caxis direction. These striations are a surface effect resulting from fracturing. Thick lamellae have been reported to survive drawing with individual lamellae embedded in a matrix of fibrils? Owing to inhomogeneity in deformation of bulk polymer and low drawing efficiency of highly chainextended polyethylene, even at EDR 4.9 (MDR 1.71, which would normally have given highly fibrillar structure, features like those in Figure 7 could still be

Fig. 7. Transmission electron micrograph of fracture surface replica of billet 1 drawn 4.9 x showing coexistence of undeformed, tilted, and partially drawn lamellae and generated fibrillar structure. Arrow indicates drawing direction.


observed. The undrawn lamellae had their c axis perpendicular to the draw direction. This corresponds to the observation of deforming PE spherulites,n in which the polar zones are deformed only at higher elongation. Therefore their survival can be observed. The lamellae with tilted chains are distinguished as those surrounding the fibrillar structure with striations tilted toward the draw direction. The tilt angle as measured by the angle between the striations and lamellar surface normal is 32-38". Higher tilt angles are not observed and transformation into fibrillar structure is discontinuous. The chain tilting observed here must be distinguished from that observed in roof-shaped polyethylene crystals crystallized at high temperatures, which have tilt angles varying from 18" to 46", corresponding to some simple, low-index planes.28 In the present case the chains, as shown by the striations above and below the fibrillar structure in the destruction zone, tilt toward the draw direction.

In the study of polyethylenes drawn at low draw ratio, Peterlin and Meinelm showed from the development of the four-point small-angle x-ray pattern and the splitting of the 110 and 200 wide-angle x-ray diffractions that the rotated lamellae also have chains tilted up to 48". The lamellae rotated at 45" give maximum plastic compliance for chain tilt and slip for its subsequent destruction. The observation in Figure 7 shows discontinuous transformation of lamellae with tilted chains into fibrillar structure and is consistent with Peterlin's model of plastic deformation.

CONCLUSION Highly crystalline CEPE with lamellar thickness comparable to the mo-

lecular chain length contains few trapped entanglements and interlamellar links to provide continuity and stress transfer for effective orientation during drawing. This is shown by the slow development on drawing of the crystal orientation functions. The efficiency of drawing increases with billets crystallized with higher undercoolings and increasing amorphous content as a result of increasing chain entanglements and interlamellar links. The efficiency reaches a maxima of 0.71 in solid-state extrusion. Electron micrographs of fracture-surface replicas show the survival of lamellae at low drawing. Regions of undeformed, tilted, and partially drawn lamellae are found to coexist with the generated fibrillar structure. This is consistent with Peterlin's model of plastic deformation.

The authors express appreciation for support from the National Science Foundation, Division of Chemical and Process Engineering, Grant CPE8005348.

References 1. A. E. Zachariades, W. T. Mead, and R. S. Porter, Chem. Rew.80, 351 (1980). 2. A. J. Pennings and K. E. Meihuizen, in Ultra-High Modulus Polymers, A. Ciferri and I.

3. P. Smith and P. J. Lemstra, J. Polym Sci Polym Phys. Ed., 19, 1007 (1981). 4. Y. Maeda et al., J. Polym. Sci., 19, 1313 (1981); D. C. Bassett, Polymer, 17, 460 (1976); T.

5. B. Wunderlich, J. Polym. Sci. Part A, 2, 3697 (1964). 6. B. Wunderlich and T. Davidson, J. Polym. Sci. A-2, 1, 2043 (1969).

M. Ward, Eds., Applied Science, London, 1979.

Davidson and B. Wunderlich, J. Polym. Sci. A-2, 7 , 2051 (1969).


7. W. T. Mead and R. S. Porter, Int. J. Polym. Muter., 1, 29 (1979). 8. J. M. Lupton and J. W. Regester, J. Polym. Sci., 18, 2407 (1974). 9. G. E. Attenburrow and D. C. Bassett, J. Muter. Sci., 12, 192 (1977). 10. E. S. Clark and L. S. Scott, Polym. Eng. Sci., 14, 682 (1978). 11. N. Capiati, S. Kojima, W. Perkins, and R. S. Porter, J. Mater. Sci., 12, 334 (1977). 12. M. P. C. Watts, A. E. Zachariades, and R. S. Porter, J. Muter. Sci., 15, 426 (1980); M.

13. Z. W. Wilchinsky, J. Polym. Sci. A-2, 6, 281 (1968). 14. D. V. Rees and D. C. Bassett, J. Polym. Sci. A-2, 9, 385 (1971). 15. H. H. Hoehn, R. C. Ferguson, and R. R. Hebert, Polym. Eng. Sci, 18,457 (1978). 16. D. C. Bassett, Principles of Polymer Morphobgy, Cambridge University, Cambridge, 1981. 17. D. C. Bassett and B. Turner, Philos. Mag., 29,285 (1974). 18. R. B. Prime and B. Wunderlich, J. Polym. Sci. A-2, 1, 2061 (1969). 19. B. Wunderlich and L. Melillo, Makromol. Chem., 118, 250 (1968). 20. Y. Maeda and H. Kanetsuna, Polym. J., 13,357 (1981). 21. K. H. Illers, Angew, Makromol. Chem., 12, 89 (1970). 22. T. Kanamoto, S. Fujimatsu, A. Tsuruta, K. Tanaka, and R. S. Porter, Rep. h g . Polym.

23. J. deBoer and A. J. Pennings, Polym. Bull., 7, 309 (1982). 24. L. Fischer, Ph. D. Thesis, University of Marburg, 1982. 25. A. Peterlin, J. Muter. Sci., 6, 490 (1971). 26. A. Peterlin, Polym. Eng. Sci., 17, 183 (1977). 27. P. M. Tarin and E. L. Thomas, Polym. Eng. Sci., 18, 472 (1978). 28. I. G. Voight-Martin and L. Mandelkern, J. Polym. Sci. Polym. Phys. Ed., 19, 1769 (1981). 29. A. Peterlin and G. Meinel, Makmmol. Chem., 142, 227 (1971).

Daniels et al., J. Mater. Sci., 16, 1134 (1981).

Phys. Jpn., 24, 185 (1981).

Received June 11, 1983 Accepted January 13, 1984

Solid-state extrusion of chain-extended polyethylene

Documents

Transcript of Solid-state extrusion of chain-extended polyethylene