PP Nylon Core

8/4/2019 PP Nylon Core

1/112

PRODUCTION AN D STRUCTURE / PROPERTIES OF NYLON -6 CORE /

ISOTACTIC POLYPROPYLENE SHEATH BICOMPONENT FIBERS

SUITABLE FOR USE IN CARPETING APPLICATIONS.

by

David Godshall

Thesis subm itted to th e faculty of the Virginia Polytechn ic Institute and StateUniversity in partial fu lfillment of the requ irements for the d egree of

MASTERS OF SCIENCE

in

Chemical Engineering

APPROVED:

___________________________G.L. Wilkes, Chairman

___________________________ ___________________________D.G. Baird R.M. Davis

February 1,1999Blacksburg, VA


2/112

ii

PRODUCTION AN D STRUCTURE / PROPERTIES OF NYLON -6 CORE /

ISOTACTIC POLYPROPYLENE SHEATH BICOMPONENT FIBERS

SUITABLE FOR USE IN CARPETING APPLICATIONS.

by

David Godshall

G. L. Wilkes, Chairman

Abstract

Bicomp onent fibers consisting of nylon-6 and isotactic polyprop ylenewere prod uced. In-situ, reactive compatibilization was achieved u sing a maleicanhydride functionalized polypropylene between the materials at the interface.The overall goal of the research was to p rod uce a bicomp onent fiber of thesematerials that would be suitable for use in comm ercial carpet app lications.Carpet samples prod uced using nylon-6 core / polyprop ylene sheathbicomponent fibers d isplayed stain resistance comp arable to a whollypolyprop ylene carpet. The wear characteristics of these fibers were foun d to bestrongly depend ent up on the maleic anhyd ride content and the molecularweight of the maleic anhydr ide fun ctionalized polyprop ylene. Adhesionbetween the nylon-6 and polyprop ylene p hases, and the mechanical propertiesof the polypropylene phase were affected by th e add ition of the fun ctionalizedpolyprop ylene. Add itional information regarding the p rocessing cond itionsnecessary to p rod uce fibers of the d esired cross-section from these materials was

obtained u sing capillary rheometry. A num ber of analytical techn iquesinclud ing DSC, TGA, and SEM were used to better u nd erstand the structure ofthe maleated materials.


3/112

iii

ACKNOWLEDGEMENTS

First and foremost, I am most ind ebted to Charlie White for his assistance, ideas,

long hours, infinite patience, humor , and free lunches. Without Char lies help at CamacI would still be lost to this day . Thankfully Charlie digs deeper than th e surface to seewh at is there, so first impr essions are not everything. When I attempted to bu rn myhand off with molten polymer du ring the first few minu tes of my first day on sight, hedid nt write me off. In reality this thesis should include his name as a co-au thor as thework herein is every bit as mu ch his as it is mine. I hope I have the opp ortun ity towork with Charlie in the futur e, both on a professional and a personal level. I knowthat there is still a lot that h e can teach me.

Secondly I would like to thank my advisor, Dr. Wilkes. He too is blessed withinfinite patience. For every chance I was given to stumble and fail as I tried to find m yway, I was given two chances to find the answers and succeed. It goes without sayingthat his technical advice was always flawless and timed perfectly. I feel that the timethat I have spen t in his lab has been am ong the m ost edu cating of my life, thanks to thepeop le that he has brough t in, and those extra requiremen ts that seem like a hassle,but in r eality are the experiences looking back at w hich I feel I have learned th e mostdu ring. Thank you Dr. Wilkes for giving me the time and th e environmen t to developas a you ng scientific professional. (ok, especially thanks for the generou s amou nt oftime)

And , speaking of the lab environm ent, I am su re that I owe most of theknow ledge that I have acquired over the past two years to my labmates. But, because Ifeel that I have someth ing more than just a professional relationship w ith mycoworkers, it is also necessary that I thank them for their friend ship. Without them Iwou ld have su rely gone insane. Because of them there is always a reason to come into

lab even if I don t feel like wor king at the time. I hope they forgive me for being su ch adistraction to them d uring these times. So, let not the nam es of Chris, Kurt, ChengHon g, Varu n, Brian, Matt O., and Matt J. not go unm entioned, as they have kep t meconstantly entertained an d amu sed these years in the big city of Blacksburg.

There were some peop le behind the scenes outside of Blacksburg that are a lsoresponsible for the comp letion of this work. In Bristol, at Camac, Brendan McSheeh ymad e this project possible, kept us on track, and gave us a d irection when we began towond er what good these fibers could be for. Matthew Studh olme provided hours ofinsightful d iscussions, and helpful indu strial contacts. His scientific and technicalknow ledge continually impr essed me. I sincerely hope tha t Camac finds a way to fullyut ilize all of his talents. Lastly down in Bristol, Diane Wrigh t did a wond erful job

helping us with wear and stain testing. And , I wou ld be remiss if did not acknow ledgethe help and entertainmen t provided by Connie Dixon and the rest of the girls in thetesting lab. They always had time for me, and a few things to say if I was willing tolisten.

Finally, I wou ld like to give a brief acknow ledgemen t to Dr. Scranton, and Dr.Rangarajan at Michigan State for encourag ing, and m aking me believe that I could beaccepted to grad uate school and succeed.


4/112

iv

TABLE OF CONTENTS

LIST OF TABLES....................................................................................................... vii

LIST OF FIGURES .................................................................................................... viii

I. INTRODUCTION .............................................................................................1

II. PROJECT BACKGROUND ..............................................................................3

2-1. Indu strial Carpet Production ...............................................................32-2. Bicomponent Fiber Spinn ing ................................................................52-2-1. Bicomponent fiber configurat ions .......................................................52-2-2. Uniqu e bicomp onent sp inning pr oblems - equip ment......................72-2-3. Uniqu e bicomp onen t fiber spinning p roblems - mechan ics .............82-3. Polym er Rheology ...............................................................................102-3-1. Terminology and representative behavior of polym er melts..........112-3-2. Fun ctional forms of viscosity .............................................................122-3-3. Cap illary rh eom etry - theory..............................................................152-3-4. Cap illary rheom etry - sou rces of error ..............................................162-3-5. Extensional viscosity ...........................................................................182-4. Draw ing and Orientation ...................................................................192-4-1. Processing steps lead ing to orien tation .............................................192-4-2. Effect of orientation on crystallization behavior ..............................202-4-3. Structural models of fiber ...................................................................21

2-5. Thermod ynam ics of Polymer Blend s ................................................222-5-1. Mathematical view of miscibility .......................................................222-5-2. Compatibilization using block and graft copolymers ......................232-5-3. In-situ compatibilization of polyprop ylene, nylon-6 blend s...........242-5-4. In-situ compa tibilization of polypropylene, nylon-6 blends -

chemistry..............................................................................................252-5-5. In-situ comp atibilization of polypropy lene, nylon-6 blend s -

prop erties .............................................................................................262-6. Prod uction of Maleic Anh yd ride Functionalized Polyprop ylene......30

III. EXPERIMENTAL............................................................................................32

3-1. Materials ...............................................................................................323-2. Mater ials Hand ling .............................................................................333-3. Fiber Spinn ing .....................................................................................333-4. Com pound ing ......................................................................................343-5. Op tical Microscopy .............................................................................343-6. Cap illary Rheom etry ...........................................................................35


5/112

v

3-7. Compression Mold ing ........................................................................353-8. Differential Scann ing Calorimetry (DSC) .........................................353-9. Scanning Electron Microscopy (SEM) ...............................................353-10. Tensile Testing .....................................................................................363-11. Therm ogravim etric Analysis (TGA)..................................................36

3-12. Wid e Angle X-Ray...............................................................................363-13. Fourier Transform Infrared Spectroscopy (FTIR) ............................363-14. Carp et Prod uction ...............................................................................373-15. Wear Testing of Carp et Specimens ....................................................373-16. Acid Red Stain Testing........................................................................373-17. Coffee Stain Testing ............................................................................37

IV. RESULTS AN D DISCUSSION ............................................................................38

4-1-1. Prod uction of Bicomponent fibers - side by side fibers...................384-1-2. Production of Bicomponent Fibers - Extrud ate exit angle...............39

4-1-3. Production of Bicompon ent Fibers - Interface movement ...............414-1-4. Ad ditional Rheology ...........................................................................444-1-5. Produ ction of Bicomp onent Fibers - Spin p ack hydrod ynamics ....504-1-6. Prod uction of Bicomponent Fibers - Fiber size distribution ...........514-2-1. Therm al Stability of Mater ials During Processing ...........................524-2-2. Thermal Stability of Materials Dur ing Processing - TGA ...............524-2-3. Thermal Stability of Materials Dur ing Processing - Capillary

Rheometry............................................................................................564-2-4. Thermal Stability of Materials During Processing -

Melt Blend ing ......................................................................................58

4-3-1. Crystallization Behavior of Project Materials an d Blends ...............594-3-2. Crystallization Behavior of Project Materials - Melt functionalizedmaterials...............................................................................................59

4-3-3. Crystallization Behavior of Project Materials - PP-MA-1and PE-MA...........................................................................................61

4-3-4. Crystallization Behavior of Project Materials - Crystal Content .....634-3-5. Crystallization Behavior of Project Mater ials - Blend s ....................644-4-1. Mechan ical Prop erties of the Blend System s ....................................664-4-2. Mechanical Properties of Blend Systems - Small to Intermed iate

Strain.....................................................................................................664-4-3. Mechanical Properties of Blend Systems - Ultimate Properties......69

4-4-4. Mechanical Properties of the Blend Systems - An Aside onMorphology .........................................................................................70

4-5-1. Qu alitative In terfacial Adhesion of Fibers ........................................724-5-2. Qu alitative Interfacial Adhesion of Fibers - PP-MA-1 Blends an d

PE-MA ..................................................................................................724-5-3. Qu alitative Interfacial Adhesion of Fibers - Blends of Melt

fun ctionalized Materials .....................................................................754-5-4. Qu alitative Interfacial Adhesion of Fibers - Mixing of Blends .......78


6/112

vi

4-6-1. Wear Testing of Carpet Samp les........................................................794-6-2. Stain Testing of Carp et Samples ........................................................82

CONCLUSIONS.........................................................................................................86

RECOMEN DATION S FOR FUTURE STUDIES......................................................88

APPENDIX A - ISOTACTIC/ SYNDIOTACTIC POLYPROPYELENEBICOMPONENT FIBERS ..............................................................A1

REFERENCES

VITA


7/112

vii

LIST OF TABLES

Table 2-1. Key prop erties in carpet performance.................................................3Table 2-2. Viscosity va lues of variou s materials ................................................15

Table 4-1. Typical bicomp onent fiber processing cond itions ...........................39Table 4-2. Typical bicomp onent fiber d rawing cond itions ...............................39Table 4-3. Sum mary of flow activation energy data at two shear rates (6,980

s-1and 140 s-1) for p rocess m ater ials ...................................................48Table 4-4. Sum mary of TGA results for PP18, PP-MA-1, and PP-MA-2..........56Table 4-5. Crystalline content of materials using DSC ......................................63Table 4-6. Crystalline content of mechanical testing specimens, prior to

testing, using DSC ...............................................................................68


8/112

viii

LIST OF FIGURES

Figure 2-1. Typical carpet constructions:A - single level pile loop; B -mu ltilevel loop p ile; C - mu ltilevel cut and loop; D - tip sheared; E

velour; F - plush; G - saxony; H - frieze; I - shag................................4Figu re 2-2. 1994 U.S. relative use of comm on face fibers for carp eting ...............5Figure 2-3. Examples of bicomponent fiber cross sections and configurations ..6Figure 2-4. Represen tative bicomponent spin pack...............................................7Figure 2-5. Encapsulation of high viscosity componen t (black) by lower

viscosity comp onent (white) for increasing L/ D ratio. Schematicshow s configu ration at ent rance of cap illary ......................................8

Figure 2-6. Cond itions leading to variation in extrud ate exit angle, , asmater ial exits the spinerette ..................................................................9

Figure 2-7. Simple Shear ........................................................................................11Figu re 2-8. Various shear stress / shear rate behaviors ......................................12Figure 2-9. Typical shear rate and temp erature dependence of polymer

melts .....................................................................................................13Figure 2-10. Effect of weight average molecular weight, M

w, on viscosity for

many polym er system s .......................................................................14Figu re 2-11. Schematic view of a capillary rheometer ..........................................15Figu re 2-12. Sources of error in cap illary rheometry d ata ....................................17Figure 2-13. Comp arison of extensional viscosity and sh ear viscosity

depen dence on deformation rate .......................................................19Figu re 2-14. Buildu p of orientation d uring p rocessing.........................................20Figu re 2-15. Three p hase model of fiber stru cture ................................................21

Figure 2-16. Compatibilization mechanism of block and graft copolymers .......24Figure 2-17. Chemical structures of materials used in blend systems: A)

polypropylene, B) maleic anhydride functionalized polypropylene,C) Nylon-6............................................................................................25

Figu re 2-18. Formation of copolymer and possible imid ization ..........................25Figure 2-19. Hyd rolysis equilibrium leading to chain scission and formation of

additional end groups.........................................................................26Figure 2-20. Effect of ma leic anhyd ride content on extraction residu e content..27Figure 2-21. Viscous respon se of blends of PP/ PA6 with varying maleic

anhyd rid e content ...............................................................................27Figure 2-22. Effect of MA content on tensile pr operties of a PP/ N6-6 blend

using a PP-g-MA with low grafting levels........................................28Figure 2-23. Effect of MA content on tensile pr operties of a PP/ N6-6 blend

using a PP-g-MA with high grafting levels ......................................29Figure 2-24. Effect of MA content on average dom ain size of dispersed PP-g-

MA p hase in a N6 matrix ...................................................................29Figure 2-25. Effects of annealing time and tem perature on fracture toughness,

Gc, of the interface ...............................................................................30


9/112

ix

Figu re 2-26. Change in Mn as a fun ction of peroxide an d MA content ...............31Figure 3-1. Schematic of mold used to moun t fiber bundles in wax prior to

cross-sectioning ...................................................................................34Figure 4-1. Capillary rheometry d ata of various process materials utilized, as

determined at 254C............................................................................40

Figure 4-2. Cross section of a PE-MA / PA6 fiber demonstrating viscositycrossover...............................................................................................41Figure 4-3. Typical level of interfacial rearrangement d uring side by side

spinning. Fiber shown is 90%PP18, 10% PP-MA-1 / PA6..............42Figure 4-4. Typical configuration of a core-sheath fiber spun with the lower

viscosity comp onent as the sheath. The fiber shown is a 60% coreof PA6 / 40% sheath of 33%P18, 67% PP-MA-1 ...............................43

Figure 4-5. Typical configuration resulting from a core-sheath fiber spun w iththe lower viscosity comp onent as the core. The fiber show n is a60% core of 33% PP18, 67% PP-MA-1 / 40% sheath of PA6............44

Figure 4-6. Capillary rheometry of selected m aleated polypropylenes. Note

that d ata for the low viscosity materials was obtained at 179C incomp arison to 243C for the PP-MA-1 material ...............................45

Figure 4-7. Capillary rheometry of selected p rocess materials at 243C ...........46Figure 4-8. Capillary rheometry of process materials at 266C..........................46Figure 4-9. Plot used to determine the activation energy of flow at a shear rate

of 6980 s-1 ..............................................................................................47Figure 4-10. Plot used to d etermine the activation energy of flow at a shear rate

of 140 s-1 ................................................................................................47Figure 4-11. Uneven d istribution of comp onents across the spin pack. The fiber

show n is a PP-MA-1 / PA6 trilobal fiber that

has not been drawn .............................................................................50Figure 4-12. Typical maldistribution of fiber sizes. The fiber show n is a 33%PP18, 67% PP-MA-1 / PA6 fiber ........................................................51

Figure 4-13. Results from taking cross sections of an ind ividu al fiber at differentpoints along its length .........................................................................51

Figure 4-14. Isothermal degradation of materials und er a nitrogenatmosphere...........................................................................................53

Figure 4-15. Isothermal degradation of materials und er an air atmosphere ......53Figure 4-16. Degradation of materials und er a nitrogen atmosphere, ramp ed at

10C/ min..............................................................................................54Figure 4-17. Degradation of materials under an air atmosphere, ramped at

10C/ min..............................................................................................55Figure 4-18. Capillary rheometry d ata for materials initially sheared at 6,980 s-1

and 254C. Data collected at 254C...................................................57Figure 4-19. Capillary rheometry d ata for materials initially sheared at 6,980 s-1

and 179C. Data collected at 179C...................................................57Figure 4-20. Cross-sections of fibers prod uced by hop per blend ing of

polyp ropylene ma terials. A) 5% PP-MA-2, 95% PP18 / PA6, B)50% PP-MA-1, 50% PP18 / PA6.........................................................58


10/112

x

Figure 4-21. Cross-sections of fibers prod uced by m elt mixing of polyprop ylenematerials pr ior to spinning in a twin screw extrud er. A) 5%PP-MA-2, 95% PP18 / PA6, B) 50% PP-MA-1, 50% PP18 / PA6 ....58

Figure 4-22. Melting behavior of melt functionalized m aterials.Degree of functionalization increases mov ing vert ically ................60

Figure 4-23. Crystallization behavior of melt fun ctionalized materials.Degree of functionalization decreases moving vertically................60Figure 4-24. Melting behavior of process materials. Note the presence of

lowering m elting p olyethylene content in the PE-MA and PP-MA-1materials...............................................................................................62

Figure 4-25. Crystallization behavior of selected functionalized materials. PP18included as a reference .......................................................................62

Figure 4-26. Melting behavior of blends of PP-MA-2 and PP18 ..........................64Figure 4-27. Crystallization behav ior from the m elt of blend s of PP-MA-2 and

PP18 ......................................................................................................65Figure 4-28. Melting and crystallization behavior of a 50% PP-MA-1, 50% PP18

blend .....................................................................................................66Figure 4-29. Tensile mod ulus of blend systems. Testing rate 10 mm / min. at

room temperatu re using d og bon e specimens..................................67Figure 4-30. Strain at maximum load of blend systems. Testing rate 10

mm / min. at room temperature u sing d og bone specimens............68Figure 4-31. Toughness of the blend systems. Testing rate 10 mm / min. at room

temperatu re using dog bone sp ecimens ............................................69Figure 4-32. Strain at break of blend systems. Testing rate 10 mm / min. at room

temperatu re using dog bone sp ecimens ............................................70Figure 4-33. SEM micrograph s of blends fractured after cooling in liqu id

nitrogen . A)60% PP-MA-2, 40% PP18: B,C)50% PP-MA-1,50% PP18 ..............................................................................................71Figure 4-34. PE-MA / PA6 fiber, head block temp erature 288C........................72Figure 4-35. PP-MA-1 / PA6 fiber, head block temp erature 288C.....................73Figure 4-36. Blends of PP-MA-1 and PP18 to redu ce the ma terials cost of the

bicomp onent fiber. A)10% PP-MA-1, 90% PP18 / PA6. B) 20%PP-MA-1, 80% PP18 / PA6. Head block tem peratu re 288C ..........74

Figure 4-37. Viscous response of PP-MA-1, PP18 blend systems at temperaturesand shear rates comparable to spinning conditions.........................74

Figure 4-38. Fiber consisting of 67% PP-MA-1, 33% PP18 / PA6 produced w iththe goal of approximating the minimu m level of PP-MA-1

necessary for ad equ ate ad hesion .......................................................75Figure 4-39. Viscous response of blends of PP-MA-3 and PP-MA-2 with PP18 at

temperatures and shear rates comp arable to spinningconditions............................................................................................76

Figure 4-40. Fibers produ ced u sing melt fun ctionalized p olypropylenes. A)10%PP-MA-3, 90% PP18 / PA6 B)10% PP-MA-2, 90% PP18 / PA6. Headblock temperatu re 288C ....................................................................77


11/112

xi

Figure 4-41. 5% PP-MA-2, 95% PP18 / PA6 fiber show ing adhesion similar tothat of the 10% blend ...........................................................................77

Figure 4-42. Comp arison of fibers processed by hopper blend ing versus premelt blend ing in a tw in screw extruder. A) 50% P-MA-1, 50% PP18PA6, hopp er blend ed. B) 50% PP-MA-1, 50% PP18 / PA6, twin

screw blended. C) 5% PP-MA-2, 95% PP18 / PA6, hopp erblended. D) 5% PP-MA-2, 95% PP18 / PA6, twinscrew blended ......................................................................................78

Figure 4-43. Wear tested carpet: 60% core, 40% sheath. Core - PA6, Sheath -PP18 ......................................................................................................79

Figure 4-44. Wear tested samp le show ing good wear resistance. 60% core, 40%sheath. Core - PA6, Sheath - 67% PP-MA-1, 33%PP18.....................80

Figure 4-45. Wear tested carpet: 60% core, 40% sheath. Core - PA6, Sheath - 10%PP-MA-2, 90% PP18 ............................................................................80

Figure 4-46. Effect of sheath th ickness on wear prop erties. A) 40% sheath, 60%core B) 30% sheath, 70% core. Both fibers consist of a sheath of 50%

PP-MA-1, 50% PP18 / core of PA6 ....................................................81Figure 4-47. 10% sheath of PP18 / 90% core of PA6 illustrating the rad ial

thickness of the sheath comp onent at low volum e fractions...........82Figure 4-48. Coffee stain testing of carpet samples. A) after staining B) after

cleaning. Both samples were mad e from a 40% sheath of 67%PP-MA-1, 33% PP18 / 60% core of PA6 ............................................83

Figure 4-49. Acid red stain testing of carpet samples. A)after staining B)aftercleaning. Both sam ples consist of a 40% sheath of 33% PP-MA-1,67% PP18 / 60% core of PA6 ..............................................................84

Figure 4-50. Wear test results for a 40% sheath of 33% PP-MA-1, 67% PP18 /

60% core of PA6...................................................................................84Figure 4-51. Results of stain testing after wear testing for a 40% sheath of 33%PP-MA-1, 67% PP18 / 60% core of PA6 ............................................85

Figure A-1. Capillary rheometry data of synd iotactic polyprop ylene. Isotacticdata at 490F is provided for reference ............................................A2

Figure A-2. Synd iotactic / isotactic polyprop ylene fiber. Dark phase consists ofisotactic material. A)low magnification show ing fiber sym metry. B)higher m agnification show ing fusing of synd iotacticcomponent...........................................................................................A2

Figure A-3. Cross-section of drawn syndiotactic / isotacticpolyp ropylene fiber............................................................................A3

Figure A-4. DSC melting and crystallization behavior of synd iotacticpolyp ropylene. Sample was heated (top scan), cooled (bottomscan), and reheated (middle scan) ....................................................A4

Figure A-5. SEM of isotactic portion of the fiber. Fibers maintain crimp afterremoval of syndiotactic component..................................................A5


12/112

xii

Figure A-6. FTIR spectra comp aring isotactic polyprop ylene standard,syndiotactic polyprop ylene stand ard and syndiotacticpolyprop ylene residue obtained from toluene after solvation.Stand ard s obtained by comp ression molding p ellets into a filmprior to scan ........................................................................................A5

Figure A-7. WAXS patterns of a side by side synd iotactic / isotacticpolypropy lene fiber. A) fiber as spu n B) fiber drawn online, d rawratio of 3, C) fiber hand draw n to maximum extensionbefore failure.......................................................................................A6


13/112

1

1. Introduction

The combination of two materials into a bicomp onent fiber has the potential toexpress the un ique p roperties of each material in a single fiber. For examp le, thefocus of this research was the prod uction of a nylon-6 core / isotacticpolyprop ylene sheath bicomponent fiber. A bicomponent fiber of thisconfiguration shou ld combine the stain resistance properties of a polypropylenefiber w ith the d urability of a nylon-6 fiber.

The processing w indow of a bicomponent fiber system is set by the viscosityratio of the two comp onents as they exit the spinneret. As both materials mu stpass through the same spin pack assembly, it is requ ired that the final stage ofthe extrusion step be performed at the same temperature for both materials.Thus, the location of each components therm al transitions and process stabilityare also key considerations. Proper matching of the viscosities can p rod uce afiber of the desired cross-section.


14/112

2

In a bicomp onent ap plication w here it is desired that the tw o componentsremain as a single fiber, additional care mu st be taken to assu re that su fficientadhesion is obtained betw een the fibers. For the nylon-6, polyprop ylene systemthis requires the use of a compatabilizer to improve the ad hesion between thechemically dissimilar materials. In this research an in-situ form ed

compatabilizer was used . Commercially available ma leic anh ydridefunctionalized p olypropylene was u sed to form a polyprop ylene - nylon-6copolymer at the interface. The process used to prod uce the maleic anh ydridegrafted polyprop ylene has a significant imp act on the prop erties of the startingpolyprop ylene. When a functionalized p olypropylene is used in substantialquantities in a fiber, it can ad versely affect the mechan ical properties of the fiber.Thus to prod uce a bicomponent fiber from chemically dissimilar materialsrequires an understand ing of the p rocessing wind ow of each m aterialindividually and together du ring spinning, and an ap preciation of the role thatthe structure of the compatabilizer plays in determining the qu ality of adhesionobtained between the ph ases.


15/112

3

2. Project Background

2-1. Industrial Carpet Production

The overall quality of a carpet is determined by its performance in several areas,and its ability to retain these char acteristics over the its service life. Carp etingnot only provides a visually pleasing surface, but also contributes to walkingcomfort, helps to dampen out noises, and provides a measure of insulation.Qualitatively a carpets ability to serve these functions over the expected life ofthe carpet can be described by its performance in relation to the followingpr operties listed in table 2-1.Table 2-1. Key p roperties in carpet p erformance.Property Description

Wear Resistance Does the carpet lose face material (fiber mass) du e to wearAppearanceRetention

Does the carpet maintain its original texture and height

Color Retention Does the carpet fade or change color over timeStain Resistance How p rone to staining from anionic compounds (dye,

foodstuffs, etc.)Soil Resistance How prone to staining from oil based compounds (di rt ,

grease, etc.)


16/112

4

Obviously the material from which the carpet fibers are made will greatlyimp act the performance. There is no one material today used in carpetapp lications wh ich is capable of providing the best p erformance in all categories.

Before making a comparison of the materials commonly used today, a very briefmention of a few of the comm only used carpet constructions shou ld be m ade.

Carpet construction refers to the manner in which the face fibers are presentedon the backing. Figure 2-1 shows some of the commonly used constructions inthe ind ustry today. It is read ily app arent from the figure that a single pile loopcarpet may have very different wear characteristics than a velour carpet or ashag carpet even if the same materials are used to make the face fiber. The looppile should resist wear better than the velour carpet all other variables beingequa l. A velour carpet presents fiber ends to the viewer. Any matting action willtend to expose the sides of the fiber, offering different optical characteristics to

the eye, and making w ear pa tterns more noticeable (ref.1).

Figure 2-1. Typical carpet constructions:A - single level pile loop; B - mu ltilevel loop pile; C - multilevel cut and loop; D -tip sh eared ; E velour; F - plush; G - saxony; H - frieze; I - shag (ref.1).

The single level loop pile construction method is the most common constructionstyle used in applications where large amounts of foot traffic are expected.Finally, the method of attachment of the face fiber to the backing and thepacking d ensity (amoun t of material used p er un it area) of the face fiber w ill alsoaffect how the carpet p erforms.


17/112

5

Rough comparisons can be made between the materials which are mostcommonly used in the carpet ind ustry today. Figure 2-2 shows the relativeamounts of material that are being used as face fibers in carpet applications inthe United States, which represents 55 - 60% of the 2 billion kg/ yr. worlddemand (ref.1).

N ylo n P o lyp ro p y le n e P o lye s te r W oo l0

10

20

30

40

50

60

70

PercentUsage

Figu re 2-2. 1994 U.S. relative u se of comm on face fibers for carp eting (ref.1).

In general it can be said that both nylon and p olypropylene fibers possessexcellent wear resistance and can have adequate color stability if the properstabilizing package is used . However, sulfonated nylon fibers typically havebetter appearan ce retention characteristics and soil resistance. Propy lenes

unique strengths come from its excellent stain resistance, low cost, and lowerdensity, allowing less material to be used to cover the same area. Clearly, a fiberwh ich is characterized by the best properties of nylon and polyprop ylene w ouldbe in high d emand by the carpet ind ustry . While no single material exists whichdisplays these properties, it may be possible through a bicomponent fiberstructure to obtain the best of both materials. This is the focus of this researchproject.

2-2. Bicomponent Fiber Spinning

2-2-1. Bicomponent fiber conf igurations

Bicomponent fiber spinning is the process of spinning fibers which arecomposed of two or m ore d istinct regions in th e fibers cross-section, w hich canbe subd ivided into three main categories: side by side, core/ sheath, and islandsin the sea. These d ifferent configurations may be placed into any num ber ofpossible cross-sectional shapes that are common ly used in monocomp onet fibers.


18/112

6

Thus, a wide variety of fibers may be produced, a few examples of which aregiven in Figure 2-3.

Figure 2-3. Examp les of bicomponen t fiber cross sections and configurations

Side by side fibers and eccentric core / sheath fibers show the often desiredproperty of self crimp ing. Differences in contractility upon d raw ing or heatingof the two sides cause the fiber to wrap itself into a helical configuration, theintensity of which will be a function of the materials used and theirarrangement. This self crimp ing behavior produces extra bulk in the yarn ,allowing the same mass of fiber to occupy a larger volum e. The add ition oftexture to a fiber is usually accomplished through an additional crimping step inmon ocompon ent fibers. It is not necessary that each d omain contain d ifferentmaterials. Many bicompon ent fibers have been prod uced in wh ich each sectionis composed of the same material processed under slightly differentlyconditions.

The core / sheath configuration presents the possibility of prod ucing a fiberwhich d isplays certain ind ividu al pr operties of each comp onent. For example,placing a polypropylene sheath around a nylon core may potentially produce afiber with the wear resistance of a nylon fiber, while displaying the stainresistance of a polyprop ylene fiber. A strong bond mu st be made between the

two materials to prevent fiber splitting. A fiber mad e of chemically differentspecies will require a sp ecialized scheme to enhance the strength of the interface.

Finally, the island s in the sea configuration consists of two compon ents whichideally show no adh esion. By making a matrix comp onent consisting of asoluble material, it is possible to remove it at a later time, allowing numerousmicro denier fibers to be produced without the process difficulties involved inhandling very fine fibers. Recently a general review of the types of


19/112

7

configurations possible, current producers of bicomponents, and producers ofbicomp onent sp in equipm ent w as pu blished (ref.2).

2-2-2. Unique bicomponent spinn ing problems - equipment.

As one can imagine, the spinning of bicomponent fibers adds complexity andcreates unique problems not normally associated with monocomponent fiberspinn ing. To produce a mu ltiplicity of fibers from one spinneret, the twopolymer streams mu st be split numerous times and then brought together in theproper configuration prior to exiting the spinneret. An examp le of abicomponent spin pack is shown in Figure 2-4.

Figure 2-4. Representative bicomp onent sp in pack. Separate melt streams enterpack from extrusion at points 17 and 18. Filtering and initial d istribu tion occur atpoin ts 22 and 23. The two thin plates split the streams multiple times andarrange the streams with the proper configuration prior to contacting in thecapillaries, point 41, and exit through the sp inneret face, point 40 (ref. 3).

The two polym er streams enter the top of the pack separately. The screens serveto filter and to help evenly distribute the streams across the width of the pack.From here the streams are split and routed to match the number of spinneret

orifices. Finally the two polymers are brought together just pr ior to exiting thespinneret. Many spin packs are quite versatile, allowing a wid e range of fibersto be mad e. The pack shown in Figure 2-4 has interchangeable plates whichallow various core / sheath and side by side configur ations to be spun.Alternate spinneret faces can also be used to produce any desired overall fibercross section (rou nd , trilobal, etc.).


20/112

8

2-2-3. Unique b icomponent fiber spinning problems - mechanics.

The concept of taking very different polymers and combining them into a singlefiber displaying the best properties of each material is enticing. However thereare some limitations on the combinations of materials that may be used together

in a bicomp onent system. Large differences in viscosity between the materialscan complicate the production of the desired cross section, and in extreme casesmake the system unspinn able. If a splitable fiber is not the end goal, the twopolymers must also be sufficiently compatible at the interface so as to haveadequa te interfacial adhesion.

In the spin pack just prior to exiting the spinneret the materials come togetherand then flow in the desired configuration through a capillary. Over this finitelength rearrangem ent of the polymer streams can occur . The degree to whichthis phenomena occurs, referred to as interface movement, is a strong function ofthe viscosity difference between the two polymers, and a function of the lengthto diameter ratio of the capillary for a given set of spinn ing cond itions. Figure 2-5 show s the phenom ena observed experimentally. The photographs show crosssections of the extrudate whose melt streams were contacted with the initialconfiguration show n in the schematic. The figu re represents a grad uallyincreasing L/ D ratio from 4 to 11 to 18 for a polystyrene (black core), wh ich ismore viscous than the polyethylene (white sheath).

Figure 2-5. Encapsulation of high viscosity component (black) by lower viscositycomponent (white) for increasing L/ D ratio. Schematic show s configuration atentran ce of capillary (ref. 4). L/ D = 4 at top right. L/ D = 18 at bottom right.

Han, as well as several other researchers (refs. 4-7), have shown experimentallythat differences in the elastic nature of the two materials have very little if any


21/112

9

effect on the shape or position of the interface formed relative to the effectsproduced by viscosity d ifferences. The ph enomena of interface movementleading to the encapsulation of the more viscous component has been explainedby the concept of minimum viscous dissipation by several authors (refs. 8-11).Through numerical simulations of Newtonian, power law, and various

viscoelastic mod els, these authors h ave obtained the same resu lts wh ich are seenexperimentally. As the two polymers traverse the capillary, they willspontaneously rearrange themselves until the configuration w hich prod uces theleast amount of resistance to flow is obtained . Thus, the material of lowerviscosity will move to the regions of greater shear, near the walls, leading to theencapsulation effect.

Large viscosity differences between the two materials can also affect thespinability of the fiber as it exits the spinneret. Differences in viscosity andposition in the capillary lead to d iffering velocities for the two materials. Aconstant volum etric flow rate for each compon ent can be achieved for various

combinations of cross sectional area and velocity. Thus, regard less of thevolumetric flow rate ratio, it is likely that one material will be traveling fasterthan the other within the capillary. Upon exiting the spinneret, the conditions ofa shearing flow are removed. The shear free boundary at the surface of the fiberafter exiting leads to a flat velocity profile across the fiber. In ord er to obtain th isprofile, material anterior to the fiber must accelerate while material interior tothe fiber mu st decelerate. The rearrangem ent of the velocity profiles within thefiber can lead to an imbalance of forces. This effect is show n schematically inFigure 2-6 for an initially parallel plate flow configuration.

Figure 2-6. Cond itions leading to variation in extrud ate exit angle, , as materialexits a pa rallel plate geometry (adap ted form ref. 5).


22/112

10

Intuitively if the fiber is spun as a p erfectly symm etric core / sheath fiber thiseffect will not occur . However, in the production of side by side fibers or anyasymmetric fiber, if the viscosity differences are too great, this effect on theextrudate exit angle will be noted. Everage (ref. 5) has studied and confirmedthis phenomena and foun d that as the capillary L/ D ratio increased, the

magnitude of the exit angle decreased for a side by side bicomponent fiberconsisting of two nylon 6s of d iffering viscosity. This logically occurs as aconsequence of the increased w rap around effect for larger L/ D ratios whichwill lead to a more symmetric configuration as the material exits the capillary.In extreme cases, the exit angle may be so great that the extrudate may contactthe face of the spinneret and stick as it exits leading to a situation in which it isimp ossible to sp in fibers.

This phenomenon should not be confused with the processes which lead to dieswell or non-zero exit angles in mono-component fiber spinning. In singlecomponent fiber spinning these effects will be due to the elastic nature of the

material remembering chain conformations present prior to entering thecapillary, and attempting to return to these conformations after exiting thecapillary. These elastic effects can occur in a bicomponen t system. The exactnature of the effects will be a function of the individual components elasticitiesand their arrangement in the fiber.

2-3. Polymer Rheology

The viscoelastic nature of a material may well constitute its most importantproperties in the fiber spinning p rocess. The viscosity of the m aterial will affect

the processing characteristics of the material through extrusion, its separationinto separate fiber streams in the spin pack, its ability to form a fibrous m aterialafter it exits the spinneret, and the final structur e of the material at the end of thefiber draw ing and take-up p rocess. Rheology is the stud y of the deformationand flow of materials. Thus, measurem ents of viscosity and other rheologicalparameters will be very important in the understanding of the fiber spinningprocess. There are three basic types of deformation that a material may un dergo:shear, elongational, and hyd rostatic bu lk deformations. In fiber spinning thecond itions und er wh ich the m aterial is subjected m ay be divided into two m ajorcategories. The first occurs within the extruder and inside the spin pack. Herethe polymer will experience essentially shear d eformations. After exiting the

spinneret the polymer primarily undergoes elongational deformations as it isd rawn d own the stack. What follows is an introdu ction to polymer rheologyunder shear conditions followed by a brief discussion of the elongational flowproperties of the materials relevant to this work.


23/112

11

2-3-1. Terminology and representative behavior of polymer melts.

The most common mode of deformation used when discussing the viscosity of amaterial is tha t of shear . Figu re 2-7 gives a schematic of simp le shear ingconditions for a single fluid elemen t. The up per su rface is subjected to a force

while the lower surface is held stationary, resulting in a deformation of theelement.

Figu re 2-7. Simp le Shear

It is now possible to define some terms w ith respect to Figure 2-7.

A

Fyx = (1)

The shear stress, yx, gives a measure of the amount of force which leads to thedeformation of the element with respect to a surface on the element, app lied in aspecific d irection.

=X

Y(2)

The shear strain, , gives an indication of the amount of deformation takingplace. It is useful in defining a very commonly used quantity, the shear rate.

= =

d

dt

d

d

d

dt

X

Y(3)

The shear rate has units of reciprocal time (sec-1) and gives a measure of thespeed at which the shearing strain is changing. The shear stress and the shearrate are linked through the quantity known as viscosity, . When a shearing

stress is applied, the material will provide a degree of resistance to the shearingaction. This resistance is characterized by the viscosity. The exact math ematicalform of the viscosity function must be determined from experimentalmeasurements.


24/112

12

The simp lest mod el of viscous behavior is termed New tonian..= (4)

For a Newtonian fluid shear stress and shear rate tensors are linearly relatedthrough the Newtonian viscosity, . Figure 2-8 show s some various possibleshear stress, shear rate relations.

S h e a r R a te

ShearStress

D

A

C

F B

E

Figure 2-8. Various shear stress / shear rate behaviors (ref.14)

The behaviors above can be classified as follows: (A) Newtonian fluid, a linearrelationship, in general common to fluids of low molecular weight species, (C)shear thinning or psuedoplastic, deviates negatively from Newtonianbehavior, commonly observed in polymeric systems, (D) shear thickening ordilatant, deviates positively from Newton ian behavior. Curves (B), (E), and(F) represent the equivalent behaviors of (A), (C), and (D) with the addition ofBingham character. A Bingham fluid is one in wh ich a critical stress, termed theyield stress, must be sur passed before any flow w ill begin.

2-3-2. Functional forms of viscosity.

To mathematically describe the observed stress strain behavior of shear thinn ingand thickening materials, a nu mber of viscosity functions, commonly referred toas constitutive equations, have been proposed. The simplest mod el which candescribe a non-linear relationship between and is the power law modelwhich for the deformation show n in fig. 2-7 could be w ritten as:


25/112

13

1= Byxyxyx (5)The value of the constant B, describes the amount of shear thinning (B1) behavior a material shows. The pow er law m odel hasbeen used extensively for polymers because it describes their pronounced shear

thinn ing behavior well. Figure 2-9 show s a typ ical viscosity versus shear rateplot for a shear thinning material at varying temperatures. At low shear ratesmany polymers have viscosities which are independent of , this region isreferred to as the Newtonian p lateau. The viscosity in this region is often termedthe zero shear viscosity, o. At higher shear rates the viscosity begins to drop.Eventually a second Newtonian plateau is theoretically possible, though it ispossible that degradation due to mechanical chain scission m ay occur before thispoint is reached.

Figure 2-9. Typical shear rate and temp erature depend ence of a low d ensitypolyethylene melt (ref.16)

The viscosity of a polymer is not only a strong function of shear rate, it is alsohighly depend ent upon temperature. Two relations have been set forth whichare common ly used to describe the temperature dep end ence of viscosity. Thefirst relation is the Williams-Landell-Ferry (WLF) equation.

log( )

( *)

( *)

( *)10

1

2

o T

o T

C T T

C T T=

+ (6)

where o(T) refers to the zero shear viscosity at temp erature, To(T*) refers to the zero shear viscosity at some reference

temperatu re, usually TgC1 and C2 are constants specific to the polym er though u niversal

values have been found that estimate the behavior for mostpolymers


26/112

14

The WLF equation has been found to describe viscosity best for temperatureswhich range from approximately Tg to Tg+100. At higher temp eratures thetemperature dependence of viscosity is best described by an empirical Arrheniusexpression.

o T AeEa RT( ) /= (7)where o(T) refers to the zero shear viscosity at tem perature, T

A is the p re-expon ential factorEa is the Newtonian activation energy for flowR is the u niversal gas constant

It should be noted that while the pre-exponential factor shows molecular weightdep endence, the flow activation energy d oes not.

The activation energy for flow is a function of local chain chemistry and

structure. Changes in the structure which increase the stiffness of the backbone(i.e. presence of double bonds, aromatic rings, or bulky side grou ps) will tend toraise the flow activation energy. The add ition of branches to the chain will havea similar effect as the branches will also restrict backbone movement.

The pre-exponential factor accounts for the increase in viscosity which resultfrom the increase in entanglements associated with rising molecular weight,observed in macromolecular systems. Figure 2-10 show s the general effect of theweight average molecular weight on v iscosity.

slope

=3.4

slope

=1

increasing .

Mwc

log Mw

log

Figure 2-10. Effect of weight average molecular weight, Mw , on viscosity formany p olymer systems (ref. 15).


27/112

15

From the figure it can be seen that once a critical molecular weight, M wc, isachieved the depend ence becomes mu ch greater. Below Mwc the log - log plotyields a slope of unity. This behavior is characteristic of low m olecular weightspecies. Above Mwc the polymer chains have become long enough to entangle.This greatly increases the viscosity resulting in a viscosity dependence on

molecular weight to the 3.4 pow er for most polymers. For polyprop ylene thiscritical molecular weight occurs at 7,000 g/ mol, while the onset is at 5,000 g/ molfor nylon-6 (ref. 75). These values correspond to 167 and 44 repeat unitsrespectively. Table 2-2 provides viscosity data for various materials forcomp arison to polymer m elts.

Table 2-2. Viscosity values of various materials (ref. 13).Material (Pa-sec)

Air 10-5Water 10-3

Glycerin 100

Syrup 102

Polymer melts 102-106

Pitch 109

Glass 1021

2-3-3. Capillary rheometry - theory.

Several instruments have been developed to measure rheological properties inshear. Each instrum ent has its own ind ividu al strengths and w eaknesses. The

capillary rheometer is capable of measuring rheological prop erties at shear rateswhich are comparable to many polymer processes. Figure 2-11 is a schematicview of a capillary rheom eter.

Figu re 2-11. Schematic view of a capillary rheometer (ref.79).


28/112

16

Solid p olymer is loaded into th e reservoir at the top of the rheom eter where it ismelted at the desired temperature. After this melt time, the plun ger is drivendow n forcing molten polymer through the capillary region. Measurem ents ofthe flow rate of material and the force required to move the plunger, corrected

for frictional effects, allow s the calculation of the materials viscosity. Making abalance of forces (neglecting gravity) on a fluid elemen t w ithin the capillaryyields eq. 8.

rz rr dP

dz( ) =

2

(8)

where rz(r) is the shear stress along a sur face of constan t r in th e zdirection

dP

dzis the pressure grad ient d riving flow

Thus, for a capillary of rad ius R, the shear stress at th e wall (r = R) of thecapillary, w, can be d efined and related to the shear stress as follows.

rz wrr

R( ) = (9)

From eq. 9 it can be seen that the shear stress in th e capillary is a linearly var yingfunction of position, regard less of the material being tested. That is, to this pointno constitutive equations have been u sed. Assuming that fluid only flows in thez d irection and that the no slip boun dary cond ition ap plies (uz=0 @ r=R), andconsidering th at the shear rate is simp ly the first derivative of the fluid velocitywith respect to radial position, the following relations may be derived .

[ ] [ ]u r r dr r

rdrz

R

r

R

r

( ) ( ) ( )( )= = (10)

Equation 10 provides a relationsh ip between , , and . Thus if two of thequantities can be measu red , the third can be calculated.

2-3-4. Capillary rheometry - sources of error.

In a capillary rheom eter an d are obtained at the w all of the capillary throughmeasu rements of force on the plunger and flow rate respectively. Equation 10can be used to obtain the Rabinowitsch equation.

+=

ln

ln w

d

d w

3

4

1

4(11)

wh ere a new term is introd uced, the apparent shear rate, , which can bedefined as follows.

=4

3Q

R(12)


29/112

17

The term in parentheses in eq. 11 is commonly referred to as the Rabinowitschcorrection. For Newton ian fluid s w an d are equivalent because there is nodepend ence of viscosity on shear rate. However, for non-Newtonian fluids,

w and are not equivalent. A shear thinning material will have a steepervelocity gradient near the wall, where viscosities are determined in capillaryrheometry, as compared to a Newtonian fluid . Thus, errors may result if theRabinow itsch correction is not applied. For a power law fluid, the Rabinow itschcorrection factor becomes,

3 1

4

B

B

+

(13)

wh ere B is the pow er law exponent from eq. 5.Additional corrections can be made to capillary rheometry data to further

improve accur acy. Figure 2-12 shows an add itional affect wh ich may cause errorin measurements.

- Low rate of shear

- Acceleration of flow leading

to an extensional flow

- High rate of shear

- Elastic recovery

Figure 2-12. Sources of error in capillary rheom etry d ata (adapted from ref. 20).

At the entrance of the capillary in the reservoir, flowing fluid is rapidly forcedfrom a large cross section into a much smaller cross section. This entran ce effectleads to an additional pressure drop which can be corrected for by using themethod of Bagley (ref.12). The method requ ires that measurements be takenusing two (ideally three or more) capillaries of varying L/ D ratio. Add itional

errors in measurement may a rise from the following: the finite amou nt of kineticenergy added to the system to accelerate the fluid in the reservoir, an additionalpressure drop from friction between the plunger and reservoir walls, and fromthe small pressure depend ence of viscosity. Finally, due to the elastic natu re ofpolym er melts, some energy can be stored in the m aterial as it exits the capillary,resulting in a pressure greater than atmospheric in the material at the outlet.This effect manifests itself visibly as die swell, and is due to normal stresseswithin the material. Normal stress data can be obtained from a capillary


30/112

18

rheometer by measuring the amount of d ie swell. One author, Cogswell, hasshown that elongational viscosity data can be obtained with an analysis ofentrance effects. The entran ce effects can be related to extensional viscositybecause the entrance flow to the capillary will have an elongational component,as sh own in Fig. 2-12(ref. 13).

2-3-5. Extens ional viscosity.

The viscous response of a material is a function of the deformation type. Earlierit was noted that in the fiber spinn ing process after the spinneret region m aterialwill un dergo primar ily extensional deformations. Thus, the processability andthe development of structure within the fiber in the spin line will be highlydependent upon , the elongational viscosity of the material. Troutons rulestates that the elongational viscosity in the Newtonian region is three times theshear viscosity.

0 03

=(14)

Troutons rule has been verified for man y polymers experimentally and it can beshown that in the limit of zero extension rates the relationship holds. It shouldbe emphasized, however, that the rule only applies to the Newtonian region.Based on the above rational it can be seen that the elongational viscosity willlikely p lay a larger role in determining a m aterials behavior in the spin line thanthe shear viscosity. Most materials show a similar temperatu re depend ence inelongation as in shear. The response of to changing extension rates is morevaried. Generally, the depend ence is not as strong as the dep end ence of shearviscosity on shear rate. Figure 2-13 comp ares the shear and elongational viscositybehavior of a polypropylene at varying d eformation rates. From the figure it canbe seen that the depend ence of on extension rate is slightly less than the sheardependence of. Of relevance to p revious d iscussions, the d ata in figure 2-12 atlow extension rates was obtained from uniaxial extension experiments (solidsymbols) while the data at higher extension rates was obtained using entranceloss data from a capillary rheometer (cross hatched data points) usingCogswells method .


31/112

19

10-3

10-2

10-1

100

101

102

10310

2

103

104

105

106

107

shear

elongation

Viscosity

,,,[Pa-s

]

Deformation rate , , , [s-1]

Figure 2-13. Comp arison of extensional viscosity and shear v iscosity dep end ence

on d eformation rate uniaxial extension d ata, capillary entr ance pressure lossdata, O capillary shear d ata. (replotted from ref.17).

Independ ently Ishibashi et. al (ref.18) have shown that for nylon-6 is relatively

insensitive to changes in extension rate. Both ordina ry polypropylene andnylon-6 have been shown to obey Trou tons rule at small extension rates. Littledata exists for extensional properties relative to shear properties due to thedifficulty in obtaining steady state elongational flows experimentally.

2-4. Drawing and Orientation

2-4-1. Process ing steps leading to orientation.

As spun, fibers do not posses the mechanical properties necessary for mostapp lications, includ ing carpeting. Only a small amount of chain orientation is

achieved dur ing the spinn ing process. It is stand ard p ractice to includ e ad rawing step either in line with spinning or as a separate step in the p rodu ctionof fibers. The draw ing step changes the final structure of the fibers, and hencetheir properties, by increasing the orientation of the chains along the fiber axis.The changes which take place during drawing are largely a function of threeparameters: the amount of draw (the draw ratio), the conditions of drawing(temp eratu re, velocity, etc.), and the initial structure of the fiber. Figure 2-14


32/112

20

presents a schematic demonstrating the relative amounts of orientation whichare introd uced at each step of fiber prod uction.

S p in n in g D ra w in g C o n d it io n in g H e a t T re a tm e n t

Orientation

Factor,

f

T i m e , t

u n d e r s t re s s

f r e e

Figure 2-14. Buildup of orientation during processing. (ref.19).

2-4-2. Effect of orientation on crystallization behavior.

Because of the presence of some orientation during spinning and becauseconditions are changing with time (position) as a fluid element moves down thespin line, the crystallization p rocess results in structures wh ich a re d ifferent fromthose obtained from qu iescent crystallization. Sph erulite formation occurs onlyin very thick filaments, commonly referred to as bristles, with very littleorientation. This is not to say that crystallization mu st occur in the sp in line forthe effects of orienta tion to be felt. PET fibers are a classic example of a materialwhich is often spun into the glassy state with subsequent drawing and heattreatmen t lead ing to oriented crystallization.

Nylon-6 does crystallize somewhat during the spinning process, though thiscrystallinity level will be substantially lower than that of the final commercialfiber. To prevent hyd rolysis, nylons mu st be processed with very little watercontent. Rapid moistur e pickup w ill occur once the material comes in contactwith the environm ent. Waters effect on a nylon-6 as a plasticizer is so effectivethat it can decrease Tg from 70C to as low as -10C. Thus, ad ditionalcrystallization will occur on the winder or during storage, increasing the crystalcontent. Analysis of nylon-6 crystallinity is comp licated by the fact that nylon-6

is polym orphic. It is generally believed that two forms exist, the and . The form appears to be the most stable and is formed under standard spinningconditions and d ur ing storage as a result of moisture regain. The form isincreasingly observed u nd er high speed spinning cond itions.

Polyprop ylene too is polymorp hic, the most predom inant and stable crystalform being the form, having a mon oclinic structure. The form, which hashexagonal packing, may in some instances be formed. Ad ditional structures,


33/112

21

an d , have been noted in the literatu re, but are very rare (ref. 77). All forms arebased on the 31 helix, with variations in packing and h and edn ess. The relativeamounts of each of these forms w ill largely depend on the m elt temperature andthe cooling rate. The more stable form is favored at h igher temperatures andslower cooling rates. Polyprop ylene crystallizes mu ch more rapid ly, in general,

than nylon-6, thus its crystallinity will change very little after spinning withoutadditional drawing or heating, at conventional cooling rates.

2-4-3. Structural models of fiber.

Several different models have been put forth in an attempt to describe themorph ological texture of fibers. No one model has been determined to besatisfactory. One basic mod el that has been developed is the three phase mod el.The model is depicted in Figure 2-15.

Microfibril

Crystallites

Tie moleculesExtended

non-crytsalline

molecules

(interfibrillar

phase)

Figu re 2-15. Three phase model of fiber structure (ref. 1).

As the name implies, the fiber consists of three distinct phases: orientedcrystalline regions, amorphous regions also with preferential orientation alongthe fiber axis which contain tie molecules connecting crystallites, and highlyextend ed non-crystalline molecules, called the interfibrillar ph ase. In the threephase model the interfibrillar phase will play a key role in the tensile propertiesof the fiber. More basic mod els do not include the interfibrillar phase, thusmaking the tie molecules the most important factor in determining the observedmechanical properties of the material. It is probable that no one mod el candescribe the texture of all materials pr oduced u nd er all cond itions. However, itcan be definitively stated that the fiber formation process results in amorphology consisting of oriented crystalline and amorphous regions arrangedin structures very different from those obtained under quiescent conditions, forthe same semi-crystalline material.


34/112

22

2-5. Thermodynamics of Polymer Blends

2-5-1. Mathematical view of miscibi lity.

In some instances it is possible to mix different polymers and form a wholly

miscible blend . This behavior often occur s when the two materials havefavorable specific interactions. Miscibility can be analyzed thermodynam icallyby considering the Gibbs free energy of mixing for the system, which can bewritten as a combination of enthalp ic and entrop ic mixing terms.

G H T Sm m m= (15)

For spon taneous mixing to occur , Gm mu st be negative. The entrop y of mixingterm, Sm, will always be positive as mixing increases the d isorder of the system,thus making the second half of eq. 15 negative and favorable to mixing. But, in

polymer systems because of the long chain nature of the molecules, Sm is oftensmaller in magnitud e than the enthalpy of mixing, H m. It is this fact that causepolym er m iscibility to rely on specific interactions betw een the species, leadingto a negative H m, to achieve miscibility. The second therm odynam icrequirement for miscibility is expressed using the second partial derivative ofthe Gibbs free energy of mixing w ith respect to composition as shown in eq. 16.

0

,

22

2

>

PT

mG (16)

The change in Gibbs free energy of mixing for two polymers can be calculatedusing Flory-Hu ggins theory. The theory gives a theoretical found ation formath ematically mod eling polymer-polym er miscibility. The following equationmay be derived from the theory (ref. 72).

( ) G RT N x xm = +1 2 12 1 1 2 2 ln ln (17)

where R is the un iversal gas constantT is the absolute temp eratureN is the total nu mber of lattice sites in th e systemi is the volume fraction of species i

12 is the Flory-Hu ggins interaction param eter between the speciesxi is the mole fraction of species i

Because the terms contained in the parentheses in eq. 17 involve taking thenatu ral log of fractional nu mbers, these terms will always be negative. Theinteraction parameter, 12, is closely associated with the enthalpic interactions ofthe blend sp ecies. The value of12 may be negative, positive, or zero (athermal


35/112

23

mixing). Solubility parameters may be used to mod el values of12 in systemswithou t specific interactions, as in eq. 18.

( )

12

1 2

2

=

RT(18)

where is the molar segmental volum e of the speciesi is the solubility parameter of species i

In this model, 12 cannot have negative values. Therefore, to achieve miscibility,12 should be mad e as small as possible. For materials of interest to thisparticular project, polypropylene and nylon 6-6, the values of the solubilitypa rameter are 9.4 and 13.6 (cal/ cm3) for polypropylene and nylon 6-6respectively (ref. 72). A general ru le of thum b for solubility states that theabsolute value of the d ifference between the solubility parameters of the speciesshould be less than 1 (cal/ cm 3) for solubility. Obviously from these values

polypropylene and nylon-6 (owing to the chemical similarity between nylon-6and nylon6-6) w ill not form a miscible blend . This is not surp rising sincepolyp ropylene is olefinic while nylon-6 is somewhat p olar. On a m olecular scalethis means that there will be little if any diffusion leading to entanglement ofnylon-6 and polyprop ylene chains during bicomp onent spinning. Therefore tospin a bicomponent fiber consisting of these two m aterials some scheme must bedeveloped to p rovide adequate adhesion between the comp onents.

2-5-2. Compatibil ization using block and graft copolymers.

In the blending of polymers, to achieve systems with adequate mechanicalproperties, comp lete miscibility is not necessary. However, the characteristics ofthe material will tend to improve as the two materials become increasinglycompatible. As compatibility increases, the average domain size of thedispersed phase will decrease and the ability to transfer stresses from onedomain to another without separation will increase. This trend can be though tof one in which the interfacial adhesion of the two materials is being increased.Figure 2-16 gives a schematic of each of these types an d a conceptual p icture ofhow they increase compa tibility.


36/112

24

A domain

B domain

graft copolymer

block copolymer

Figure 2-16. Comp atibilization mechanism of block and graft copolymers.

Block copolymers consist of alternating sections of two or more monomers.Figure 2-16 show s a diblock copolym er consisting of repeat units A and B. In agraft copolym er the chain backbone consists of one repeat un it with one or moregrafts of a second repeat unit branching off the backbone. If the composition ofthese copolymers is such that each of the repeat units involved is chemicallycompatible (or identical) with a phase in the blend, the copolymers will tend tocongregate at the domain interfaces with the orientation shown in Figure 2-16.This action allows entanglements between the copolymer and each phase, thusstrengthening the bond between the two phases, lowering interfacial tension,and creating smaller d omain sizes. Nu merous blends of otherwise incomp atiblepolymers have been formed and studied u sing this method. To be effective the

block or graft copolymer m ust be located a t the domain interface.

While thermodynamics may dictate such an arrangement, the process ofmigration to the interface may be limited kinetically. Even in the melt,macromolecules are mu ch less mobile than smaller molecules. Also, consideringthe relatively short time span most materials spend in the melt state when beingprocessed, it easy to imagine that the copolym ers may not have sufficient time tod iffuse to the interfacial region in most app lications. This pr oblem can becircum vented if the copolym er is formed in-situ at the interface.

2-5-3. In-situ compatabilization of polypropylene, nylon-6 blends.

The in-situ formation of a block or graft copolymer in polymer blends is oftenaccomplished by using two mutually reactive species to form the copolymerunder the elevated temperature and mobility cond itions of the melt. To providereactive sites, a functionalized version of one of the components is often added


37/112

25

to the blend . This functionality should be capable of un der going a reaction withthe other species, while not being so heavily functionalized that it becomesincompatible w ith its par ent polymer. An excellent example of such a systemwas used in this study . The compatabilization of nylon-polypropylene blend susing a maleic anhydride functionalized polypropylene has been studied by

nu merous researchers. A portion of this work is summarized below.

2-5-4. In-situ compatabilization of polypropylene , nylon-6 blends - chemistry.

It is generally believed that the reaction between the maleic anhydride (MA)grafts and the nylon-6 (N6) occurs at the amine ends of the N6 chains when in-situ comp atabilization is occurring in the melt. To help visualize this, thechemical structures of the materials involved are given below in Figure 2-17.

O OO O OO

n

N H 2

O

N H

O

O H

n

A)

B)

C)

Figure 2-17. Chemical structures of materials used in blend systems: A)polypropylene, B) maleic anhydride functionalized polypropylene, C) Nylon-6

It should be noted that the structure presented for the maleic anhydridefunctionalized polypropylene (PP-g-MA) represents a possible arrangement ofthe maleic anhyd ride grou ps. The exact location and d istribu tion of the groupsis still the subject of debate. The likely mechanism for copolymer formation w illconsist of the following step s as show n in Figure 2-18.

R NH2 +

O OO

O

OH

O

NH

R

N

R

OO

- H2O

Figure 2-18. Formation of copolymer and possible ring closur e.Accord ing to Al-Malaika (ref. 28), the second reaction in th is sequence is likely tooccur due to the h igh temp eratures (> 200C ) in the melt. Direct evidence of thisreaction has been d ifficult to obtain.

It might be inferred that the maximu m ratio of MA group s to amine end group swhich can react is 1:1. This is not the case. Polyam ides are capable of


38/112

26

undergoing hydrolysis which scissions the chains, reduces molecular weight,and creates new am ine and carboxyl end g roups. The process of hyd rolysis isillustrated in figure 2-19.

NH C

O

+ H2O NH

2

+

HO C

O

Figure 2-19. Hydrolysis equilibrium leading to chain scission and formation ofadd itional end groups.

Through hyd rolysis, the new reactive amine end groups created allow foradditional linking with maleic anhydride groups. Amide groups within thechain are believed to be much less reactive than those at the chain end s. Theproximity of water molecules formed in the compatabilization reaction maymake this process more likely to occur near the interface between the twomaterials (ref. 28).

One final note should be made regarding the different behaviors of asymmetric(N6) versus symm etric nylons (N6,6) dur ing compatibilization w ith PP-g-MA. Ithas been by Paul and coworkers that a symmetric nylon chain may form twografts with PP-g-MA while an asymmetric nylon can only form one (ref. 31).The monomers which are u sed to p rodu ce these polymers explain this behavior.Assymetric chains, which are formed from a single monomer, will always beprod uced with one amine end grou p. Symm etric chains, which are polymerizedfrom diamines and dicarboxylic acids, may have two, one, or no amine endgroups.

2-5-5. In-situ compatabilization of polypropylene, nylon-6 blends - properties.

One of the earliest studies of the compatabilization of polypropylene, nylon-6blend s was done by Ide and H asegawa (ref 21). Blend s were prepared in anextrud er with varying levels of PP, PP-g-MA and N 6. These blends were thenextracted with xylene to remove the PP portion, and the mass of the remainingpor tion determined. Figure 2-20 show s their results. In this plot, [N] representsthe ratio of maleic anhyd ride group s to amino group s. It can be seen that as MAcontent increases ([N] increases) that the non-extractable portion of the blendincreases, suggesting that a chemical link has formed between the PP-g-MA andN6 w hich cannot be solvated.


39/112

27

0

5

10

15

20

25

30

35

0 1 2 3

[N]

Residueofextra

ction(%)

Figure 2-20. Effect of maleic anhydride content on extraction residue content

(adap ted from ref. 21).The plot also shows that above [N] values of 0.5 or higher the amino end groupsof N6 have effectively been saturated, resulting in a maximum in the amount ofresidu e. Ide and H asegawa also showed that the melt indices of thecomp atabilized blends w ere lower than non-compatabilized blend s, suggestingan increase in viscosity.

More extensive characterization of the effect of copolymer formation on theviscosity of these blend s has been done by Marco et. all (ref 22). In this stud yviscosity measurements of 70/ 30, PP/ N6 blends were m ade using a capillary

rheom eter for various levels of MA content. Figure 2-21 gives a sampling of theresults.

1 10 100 1000 10000

10

100

1000

10000

0% PP-g-MA

1%

3%

5%

10%

Viscosity(Pa-s)

Shear rate (s-1

)

Figure 2-21. Viscous response of blend s of PP/ PA6 with varying maleicanh yd ride content (ref 22).


40/112

28

The viscosity of the grafted PP is typically lower than both the PP and N6compon ents of the blend. Therefore, if no comp atabilization was occur ring, adecrease in the viscosity of the blend should result from increasing the graftpolym er content. The opp osite effect is observed. Increasing viscosity with

increasing MA content may be a function of the increased compatibility of thematerials in the blend and the formation of a high molecular copolymerconsisting of PP-g-MA and N6.

The mechanical properties of a blend also give an indication of the degree ofcompatibility between compon ents. One such stud y of these effects has beendone by Duvall et. all (ref 23). A blend of two incompatible materials will tendto be very brittle. A blend which shows a high d egree of compatibility shoulddisplay mechanical properties which are intermediate to the two materials.Figure 2-22 d isplays this trend .

5 0

4 0

3 0

2 0

1 0

00 5 1 0 1 5 2 0 5 0 6 0

0 % 2 0 . 0 %1 5 . 0 %

1 1 . 2 5 %7 . 5 %

3 . 7 5 %2 . 5 %

Stress

(M

Pa)

S t r a i n ( % )

Figure 2-22. Effect of MA content on tensile pr operties of a PP/ N6-6 blend usinga PP-g-MA with low gra fting levels - 0.2 wt% MA(ref 23).

It should be noted that this stud y was d one with a nylon 6-6. The same generaltrends would be expected to hold for blends using N 6 rather than N 6-6. Thisplot clearly shows that as the comp atabilizer content increases, the material failsat larger strain levels, show ing a d ecrease in brittleness. This behavior suggestsincreased compa tibility between the PP and N6-6 phases.

Maleic anhyd ride content is not the only var iable which w ill affect the p ropertiesof these blend s. The molecular w eight of the grafted PP must be sufficient to

allow for entanglements w ith PP chains. Ad ditionally, increasing the number ofgrafts on the PP backbone will change the natu re of the molecule. It is likely thatat high grafting levels the presence of num erous polar m aleic anhyd ride grou pson the chain will cause the grafted polymer to become less compatible withpolyp rop ylene. These effects can be seen in Figu re 2-23.


41/112

29

5 0

4 0

3 0

2 0

1 0

00 5 1 0 1 5 2 0 2 5

0 %

1 1 . 2 5 % 7 . 5 % 3 . 7 5 % 2 . 5 %

S t r a i n ( % )

Stress

(M

Pa)

Figure 2-23. Effect of MA content on tensile pr operties of a PP/ N6-6 blend usinga PP-g-MA with h igh grafting levels - 2.7 wt% MA(ref 23).

From figure 2-23 it can be seen that the strain to break of this blend decreases as

the amount of grafted polypropylene used increases, although thecompatibilized blends still show larger values than the uncompatibilized blend.This trend suggests that while the graft copolymer formed does increasecompatibility, the strength of the interphase between PP and N6-6 in theseblends is affected by the natu re of the PP-g-MA por tion of the copolymer.

To this point much of the discussion has focused on indirect observations of thecompatibilizing effect of the in-situ formed copolymer of PP-g-MA and N6 forcompatibilizing PP/ N6 blend s. More d irect evidence of compatablilzation isprovid ed by microscopy stud ies. Paul and coworkers have done an excellent

stud y of the effect of PP-g-MA on the morphology of PP/ N6 blend s (ref 24).Figure 2-24 shows Pauls results for the effect of MA content on blendmorphology.

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.50.01

0.1

1

10

dn

dw

AverageParticleSize[m]

Weight % Maleic Anhydride in PP-g-MA

Figure 2-24. Effect of MA content on average domain size of dispersed PP-g-MAphase in a N6 matr ix (ref 24).


42/112

30

These results are reminiscent of Ide and Hasegawas results wh ich also showed aplateau at high MA contents. As expected, increasing the amou nt ofcompatibilizer decreases the d omain size.

Studies of the fracture toughness of the interface formed between sheets of PPcontaining PP-g-MA w ith N 6 have been cond ucted by Boucher et. all (ref 25, 26),and Bidau x et. all (ref 27). In these stud ies sheets of PP and PP-g-MA werebond ed to sheets of N6 at various bond ing times and temp eratures. A blade wasthen forced along the interface causing the PP and N6 sheets to separ ate. Theforce required and the size of the plastic deformation zone ahead of the crackallow calculation of the interfacial toughness. Boucher et. all has demonstratedthe effect of chain mobility on copolym er formation u sing this method as shownin Fig. 2-25.

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

R

ooo

o

o

oo

ooo

oooo

o

oo

oR

223C

220C

213C

185C

00

200 400 600 800 1000

10 0

20 0

30 0

40 0

50 0

60 0

70 0

t (min)

Gc

(J/m2)

Figure 2-25. Effects of annealing time an d temp erature on fracture toughness, Gc,of the interface (ref 26).

Examining these results it can be seen that the greatest amount ofcompatibilization occurs at the highest temp erature. In this plot, 223C is theonly temperature at which both the PP-g-MA and the N6 are in the melt state.Some copolymer appears to form even when only the graft polymer is in themelt state, but the formation of copolymer greatly increases when bothcompon ents have increased mobility. These results suggest the importantconcept that compatabilization is brought about by the interdiffusion of the PP-g-MA and N6 leading to the reaction of the amine end groups with the maleic

anhydride grafts and the formation of entanglements between the ends of thecopolymer w ith each p hase.

2-6. Production of Maleic Anhydride Functionalized Polypropylene.

The emphasis to this point has been on the effects of maleic anhydride contenton the properties of blend systems. A brief mention has been m ade regard ing


43/112

31

the effects of PP-g-MA structure on b lend p roperties. A short d iscussion of themethod used to produce PP-g-MA may shed some light on a weakness of thiscompatabilization approach.

There are three reported methods for the production of PP-g-MA: solid state

reaction, solution grafting, and m elt grafting. The most comm only used methodcommercially is melt grafting. In this process polyprop ylene, maleic anhyd rideand an organic peroxide are simultaneously processed through an extruder.Heat inside of the extruder leads to the thermal decomposition of the peroxide,forming rad icals. These rad icals may then abstract a tertiary hyd rogen atomfrom polyp ropylene forming a macro-radical. This macro-radical may reactwith a maleic anhydride or lead to chain scission of the polypropylene.Unfortunately, the chain scission reaction is very fast. It is for this reason thatthe production of PP-g-MA involves a large redu ction in m olecular w eight of thestarting PP. In a study by De Roover et. all (ref. 29) the d egrad ation of PPduring the melt functionalization process was followed as shown in Figure 2-26.

80000

60000

40000

20000

0.00 % MA0.05 % MA

0.25% MA1.00 %MA4.80 % MA

0.0 0.5 1.0 1.5Peroxide weight percent

Mn

(PPequivalent)

Figu re 2-26. Change in Mn as a function of peroxide an d MA content (ref. 29).

Increasing levels of peroxide are necessary to increase the amount of maleicanhyd ride grafted to the chain. Therefore, increasing m aleic levels w ill lead tolarger am oun ts of degradation, resulting in a low molecular w eight m aterial.

The exact mechan ism of graft formation after rad ical prod uction is still being

debated , as is the exact natu re and location of the maleic anh yd ride grafts alongthe backbone. There are conflicting views amon g researchers wh ether or notmaleic anhyd ride may hom opolymerize und er melt reaction cond itions to formshort oligomeric grafts. De Roover et al. has presented evidence suggesting thathom opolymerization is possible (ref. 30). Further work n eeds to be done in thisarea to determine the location and natu re of the grafts as they have a p rofoun deffect on the compatibilization p rod uct.


44/112

32

3. Experimental

3-1. Materials

Materials from several d ifferent sup pliers were used in this stud y. Ny lon-6 waspurchased from BASF, grad e BS300 ([] = 2.7 dL/ g in sulfuric acid, 25C). BASF700D, a high amine end content nylon w as also used in limited amounts. Non -

functionalized polyprop ylene of 18 melt flow index (ASTM D1238 - 230C/ 2.16kg) was obtained from Aristech. In a few instances, 12 and 35 melt flow index(ASTM D1238 - 230C/ 2.16 kg) Aristech polyprop ylenes were also used .Maleated polypropylenes were obtained from a nu mber of ind ustrial sup

Documents

PP Nylon Core