Spherulitic growth rates and coalescence phenomena in polypropylene/polyethylene-based ionomer blends

Publisher’s version / Version de l'éditeur:

Macromolecules, 23, 1, pp. 338-345, 1990-01-01

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

https://nrc-publications.canada.ca/eng/copyright

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à [email protected].

Questions? Contact the NRC Publications Archive team at

[email protected]. If you wish to email the authors directly, please see the first page of the publication for their contact information.

NRC Publications Archive

Archives des publications du CNRC

This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

https://doi.org/10.1021/ma00203a058

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Spherulitic growth rates and coalescence phenomena in

polypropylene/polyethylene-based ionomer blends

Caldas, V.; Brown, G. Ronald; Willis, J. M.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site

LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

NRC Publications Record / Notice d'Archives des publications de CNRC:

https://nrc-publications.canada.ca/eng/view/object/?id=4a4d2bca-abe7-4d15-ace1-79392b78484e https://publications-cnrc.canada.ca/fra/voir/objet/?id=4a4d2bca-abe7-4d15-ace1-79392b78484e

1990, 23,

Spherulitic Growth

Rates

and

Coalescence

Phenomena

in

Polypropylene/Polyethylene-Based Ionomer

Blends

V. Caldas and G.R.

Brown*

Department ofChemistry,Polymer McGill,McGill University,801 SherbrookeStreet West, Montreal, Quebec,CanadaH3A-2K6

J, M. Willis

Industrial Materials ResearchInstitute, NationalResearch Council ofCanada, Boucherville, Quebec, CanadaJ4B-6Y4. ReceivedDecember19,1988;

Revised_ManuscriptReceived_May23, 1989

ABSTRACT: Theisothermalradial_{growth rates}ofspherulitesofpolypropyleneinblendswithan essen-tiallynoncrystallizable ionomer havebeen_{determined by photomicroscopy. Blends}containing 5and 10

wt % ionomer, crystallizedfromthe melt,havea _{grainyappearance due}toionomerentrapmentandshow

an increasednucleationdensityas _{compared to}pure polypropylene. At large supercooling,theseblends

also show depressedradial_{growth rates, which}are attributedtoa _{significant dissipation ofenergy by the}

crystallizingfrontto _{affect the rejection}oftheionomer domains. Upon repeatedcrystallization ofa_given

sample,an increasein theradialgrowth ratewith melt timeisobserved_{initiallygiving rise to}a _maximum,

withgeneral behaviorthatis_{dependent upon the blendcomposition and thecrystallizationtemperature.} Beyondthis maximum thespherulitic growth rate decreasesuponfurther crystallization. Theobserved effectson the_{spherulitic radialgrowth rates}withrespect to repeatedcrystallization are attributedtothe

coalescenceofthe _{ionomer dispersed}phase. Thisisconsistentwith aconcomitantincreaseinthe_average

domainsize.

Introduction

The blending ofpolymerstoproduce alloyswith prop-ertiesselected for given applicationsis an area of poly-mer sciencethatispresentlyreceivinggreatattention.1"3 Blending is _{not only economically viable} but is also a

versatile wayof_makingnew materials_havinga_{wide range}

ofproperties.

For polymerblendsin whichone of_{the components}is

crystallizable, thepresenceofthe secondcomponentcan strongly affect the crystallizationprocess. Both the

crys-tallizationkineticsandthe finalcrystal morphologycan

beverydifferentfromthat ofthepurecrystallizing

com-ponent. As reviewedrecently by Martuscelli,4such changes have been _{found not only}in blendsof_compatible

poly-mers butalsoinblendsofimmiscibleor _partially misci-ble polymers.

Inarecent_study, Willisetal.5demonstrated the

abil-ity

ofa_{polyethylene-based ionomer to}enhancethe

com-patibility

of polyolefin/polyamideblends. The

morphol-ogyoftheternary polyolefin /ionomer /polyamideblends

was _{significantly dependent upon ionomer} concentra-tion. Infact, for fixedprocessingconditions, the size of thedispersedphasewas reduced_byasmuchasfour times

dueto theadditionofaslittleas0.5% _{ionomer byweight.}

From the physicochemicalstudiesofthe_{binary blend of}

the ionomerwith _{polyamide,6specific interactions in the} form of hydrogen bonding between the blend compo-nentswere demonstrated. Suchinteractionswere absent in the polypropylene or _{polyethylene/ionomer} blends. Moreover, the ionomerseverely impedesthe

crystalliza-tion ofthe_polyamidebutenhancesthenucleationofthe

polyolefins,particularly in thecase of polypropylene. The_{presentinvestigation}concerns the effectofa

poly-ethylene-based ionomeron theradial_{growth rates}of

spher-ulitesofpolypropylene. In the lastfew years many

stu-dies7"14 have been done on the effect ofelastomers on

the crystallization kinetics ofpolypropylene. However,

*_{Authorto whom correspondence shouldbeaddressed.}

0024-9297/90/2223-0338$02.50/0

to theknowledgeofthe authors, thereare no _reportsof

studies ofthe effect ofan _{essentially noncrystallizable} ionomer on the radial growth rates of _spherulites of a crystallizable polymer.

The main goal of this work is _{to study the effect} of

composition,crystallization temperature,and melt resi-dencetimeon the spherulitic growthratesof

polypropy-lene _{in binary polypropylene/ionomer} blends. A study

ofthe_{crystallization}ofblendscan _{ultimatelyyield}

infor-mationconcerning the interactionsatthe interface between

the polymers in the blend. The effectof_{crystallization}

on the rate ofcoalescenceof_dispersedionomerphaseis alsoconsidered. Thus, theresults presentedin thispaper

will_{contribute to the understanding}oftheeffectiveness

of the ionomer at enhancing the

compatibility

of

polyolefin/polyamideblends.

Experimental

Section

Materials. The polymers used were _{polypropylene}

(Pro-Fax6501)originallyobtainedin_powderform from Himontand

Surlyn9020from DuPont. The polypropylene (PP) resinwith 0.1% _{of Irganox} 1076 antioxidant was _subsequently

com-pounded with a Werner and Pfleiderer _{(ZSK30) twin-screw}

extruderand converted toa_pelletform. The ionomeric resin

(Du Pont, Surlyn 9020) is a randomterpolymerconsisting of approximately80% ofpolyethylene, 10% of methacrylicacid, approximately70% neutralized withzinc, and10% _{of isobutyl} acrylate. Somepropertiesoftheseresinsare _giveninTable I. Prior_{to blending, the resins}were dried undervacuum at95°C for 24h.

Preparation of the _{Polypropylene/Ionomer} Blends. Blends having 5%, 10%, 20%, and50% byweightofionomer

were prepared according to thefollowingprocedure: The

res-inswere meltblendedina _{Brabender chamber,initially}setat 250_{°C, using}rollerbladesthatare recommendedfor_highshear applications. The resinswere allowed to mixat 50rpm under

a _{constant nitrogen}flowfor5min. Themeltwas then_rapidly

transferredtoa mold and placedinahydraulicpressunder2.5

MPa ofpressureuntilthe samplecooledtoroom _temperature.

Theresultantmolded sampleswere cutinto specimenshaving dimensionsof_roughly1 X0.5X0.5cm. Thin films_(5-10Mm)

©1990American Chemical Society

Macromolecules, Vol. _23, No. 1, 1990 _{Polypropylene/Polyethylene-Based Ionomer Blends} 339 0.40

I

o at o 0.35 0.30

TIMEINTHE MELT (min)

Figure 1. _{Reproducibility of}theradial_{growth rates}for

poly-propylene at 124 °C.

TableI

GeneralProperties of PolypropyleneandSurlyn

meltindex, density, torque(250°C), melting

resin g/10 min g/mL N/m point,°C

PP 4.0 0.905 8.6 165

Surlyn 1.0 0.965 14.3 62

were then cut fromthese specimensusingaReichert Jung

Super-cut2050microtome.

Isothermal CrystallizationProcedure. ALeitz Wetzlar optical polarizing microscope fitted with a MettlerFP52 hot

stage was used to observe isothermal crystallization at 100x

magnification. The hotstagecouldbeheldataconstant tem-perature to ±0.3 °C byaMettlerFP5temperature control unit. For the microscope study, microtomedfilmswere _placed on a_{microscope slideglass}and meltedon ahot_{plate while}

apply-ing pressureon the cover _glass. The thicknessofthemelted

filmwas fixedbya 12-#mialuminumspacerbetween thecover

glassand the microscope slide. Filmthicknessofthe orderof

12-15 nm were obtained andused in the microscope study. The followingprocedurewas usedto view isothermal crys-tallization: Thesampleswere meltedon thehotstageat200°C for 10minto remove _{crystallinity. Different distributions of}

spheruliteswere observedinsuccessive_{crystallizations,} indicat-ingthat the melt residencetime as well as melt _temperature was sufficientto remove residualcrystallinity. Thehot stage

was thensettoan intermediate_temperatureof135°C, and the samplewas allowed tocome to_{temperatureequilibrium,}which

took approximately2min. Whentemperatureequilibriumwas

achieved, the hot-stagecontrolwas settothe desired crystalli-zation temperature, andequilibriumwas achievedrapidly(<30 s). Photomicrographswere takenof_{the growing spherulites at}

appropriate timeintervals, depending on the rateof crystalli-zation.

Radial GrowthRateMeasurements. The isothermalradial growthrates,G= _dr/dt(r= _radiusof_the

spherulites, t= _time),

were determined by measuring the sizesofthe spherulitesas seen _{in the photomicrographs}as afunction oftime. Theradii

ofthese_spheruliteswere measuredwitha_{semiautomatic}

digi-tizer, describedelsewhere.15

The radius ofeach spherulite was taken as the average of 15-20 individual radii. From the _average radii, plotted as a

function of time, theradial growth rateofthespherulitewas

determinedastheslopeoftheresultingstraightline. Foreach composition andcrystallizationtemperature, theaverageradial growth rate was determined from at leastfour radial growth rate measurements. _{The experimental} error in these growth

rates was determinedto be lessthan 5%.

Results and Discussion

Crystallization

of Polypropylene. When polypro-pylene is crystallized from themelt in the temperature

rangeof118-126 °C,the predominant spherulitic

appear-ance observedmicroscopically is_type I, as _{classified by}

Padden and Keith.16 The various results _presented in

this _{investigation} will _mainly concern this _morphology.

CRYSTALLIZATIONTEMPERATURE(°C)

Figure2. _{Radial growth rate dependence}of_{polypropylene and} the 50% PP/50% ionomer blend on _{crystallization}

tempera-ture,for amelt time of10min.

Asecondtypeofspheruliticappearance(type

III),

intro-duced later in the paper, is shown tobehave somewhat

differentlyfrom thefirst.

Inthe_regionof_{15-60 ^m,} theradii of_the

polypropy-lene spherulites crystallized from the melt increase

lin-earlywithtime. To verify the reproducibilityoftheradial growthrates,a sample ofpolypropylene was _repeatedly

melted, for 10 _{min, and crystallized.} As shown in Fig-ure 1, at 124 °C theradial growth rates fluctuate about 0.335/um/s andare within_{approximately}±0.02_^m/sof

eachother.

The_dependence ofthe _{radial growth rate}of

polypro-pylenespheruliteson _{the crystallization temperature}shown

in Figure2 is inaccordance with that predicted by the

Fischer-Turnbullequation17

G =

G0expj-A£D*/kT]

expj-A$*//jTj

(1)

where, the term exp(-AED*/kT\isreferredtoasthe

trans-portterm and_{exp{-A$*/&Tj,} as the nucleation term, G is_{the growth}rate of_{polymers, Ga}is _{the preexponential}

term, AEB* is the _{free energy} of activation for move-ment across the_{embryo-meltinterface, A$*} is the min-imum_{free energyrequired}fortheformationofanucleus

of criticalsize, k isthe Boltzmann constant,andTisthe crystallization temperature. For crystallizationabovethe

optimum crystallization temperature, the growth rate is

controlled by the nucleation process so that the activa-tionfree energy isnegative and theexpectedinverse depen-dence ofthe _{radial growth rate} on the temperature is observed.

Crystallization

of

_{Polypropylene/Ionomer}

Blends.The effectofcomposition on theradial growth

rates _{of the polypropylene spherulites obtained for the} blendsthathave beenheldinthemelt at200 °Cfor 10

minisshownin_Figure3. A_{small depression}ofthe_growth rate is seen at low ionomer content and low

tempera-tures. At_{highertemperatures,consequently much slower}

crystallization, the growth rate is _essentially

indepen-dentoftheionomerconcentration.

As shownin Figure 2,theradial growthratesobtained

for the polypropylene (PP) spherulites in the50% _PP/ 50% ionomer blend are _{identical, within experimental}

error, to those obtained for pure polypropylene. This

generalpatternhas_{also been observed by}Martuscelliet

al.14,18 for_{the spherulitic growth}of_{polypropylene} with ethylene/propylene/diene terpolymer (EPDM).

How-ever,the growthratesofisotacticpolypropylene inblends

with other_types ofrubbersshoweda _significant

depen-denceon _compositionand_{crystallizationtemperature,18} whichwas attributed to partial miscibility.

23, ° 118"C • 120*C i °_ • 0 10 10 20 30 40 50 50 PERCENT IONOMER

Figure3. _{Composition andcrystallizationtemperature}

depen-denceofthe_{radial growth}rates_{in polypropylene/ionomer}blends.

It hasbeendemonstrated thatthe_{presenceof foreign}

inclusionsinthemelt,asin thecase ofimmiscible_blends, can disturb the crystallizingfront.4,14,18 Ifthereare

inter-facial interactions between the blend components, the growing crystal front must do work against the

dis-persedphasein the melt. Inthecase ofa_{one-phasemelt,} some _energyis_{required to effect the rejection of}the sec-ond component from the crystallizing polymer. In the

case ofa _{two-phase melt,}where thesecond component

hastaken the_{form of droplike}domains, energymustalso bedissipated for therejection, engulfment,and/or defor-mationofthe second_component.

These energy dissipations constitute energy barriers

which influence the growth rateof_{the spherulites in the}

melt. Bartczak,Galeski, andMartuscelli19 modified the

classical Fischer-Turnbull equation to account for the

energy dissipations which result during the

crystalliza-tion of a _{polymer in the presence} of_{noncrystallizable,} sphericaldomains. The modified equation is

G =

G0exp|-AED*/kT) exp{-A$*/fcT) X

expHEj

+_{E2 + E3}+ _E4+ _E5)/kT) (2) where_{E1 is}the_{energydissipated by the crystal to effect} the rejection ofthe second component into interlamel-larregions,E2 istheenergydissipated by thespherulite

front to _{effect the rejection} of_droplike domains inthe

melt,E3 is the kineticenergy required to overcome the

inertia of the _{drops, £4} is the energy required to form

new interfaces between the spherulite and drop when

engulfmentoccurs, and£5istheenergydissipatedwhen the engulfed dropsare _{deformed by thecrystallizing front.} The equationsthatdescribethe growth rate, takinginto

account these energy dissipations, derived byBartczak

et _al.,19are summarized inTable II.

The crystallization of polymer blends whose compo-nents havephaseseparatedinthe meltisinfluencedboth

by thesizeofthe dispersed phase and theinterfacial

ener-gies. The drivingforcefor therejection,engulfment,and/ or deformation_processesliesin thedifferenceofthe

inter-facial free energy betweenthe crystallizingfrontand

inclu-sions, 7„s, and the interfacial free energy between the meltana inclusions, _7pl:20-22

AE=

7ps

-TPi (3)

ForAF< 0,the dispersed domainsinthemeltare more

likelyto be_{engulfed than rejected by the growing}front.

However, when AE > 0, the situation is more

compli-cated. At slowrates _{of crystallization,} the domains are

TableII

Growth RateExpressions for Spherulites in an

IncompatiblePolymer-PolymerSystem

description expression11 rejection ofthe ^ G1 G: dispersedphase G= i+_EJRT,~

3p^y^R'

1+ 2Pmr*RTc

engulfment ofthe G= _{Gj exp{E3/i?Tc|}

dispersedphase = Gjexp|-(3c^mAF/(pmrRTc)j deformationofthe G= Gtexp\EJRTc\ dispersed phase = G[ exp{-C/(*)(3cMm7pa/(p„rRTc)\ a

Gj istheundisturbed growthrate;pmisthemolecularweight

ofthe repetitive unit;c isthevolumeconcentrationof the second

componentinthe blend;ymisthe viscosity ofthematrix; R'isthe

temporary radius ofthespherulite;pmisthedensity of the matrix;

ristheradiusoftheionomer domains; U(k)isadeformation

func-tion;R isthegas constant; and Tc isthecrystallization tempera-ture.

pushed alongby the growingfront,while at higherrates

theyare _{simplyengulfed.}

At

some intermediaterate,the domains maybe_pushedfor shortdistances before being engulfedby the crystallizing front. Engulfment occurs

whentheviscoushindrancecausedby the motion ofthe

domainsoverwhelmsthe forcesofrepulsion. This implies

a critical rate at which the pushing of the domains no

longeroccurs and _engulfment takes precedence. _Using

these arguments,fora_{given AE,}there must beacritical

domainsizesuch_{that, at}aconstant growth rate,domains

ofthis size or _larger are _{simply engulfed and} no _longer

pushedby the growingfront. Engulfmentoccurs above

thiscriticalsize because the hydrodynamicforceis

pro-portionalto the domain size _{while the repulsive} forces

are determined by factorssuchas the shapeofthe

inter-facebehindthe_particle.19 Thecriticalsizeitselfis

depen-denton the_{system, i.e.,} on the interfacial _energies,

vis-cosities, and heatconductivities.

In this study, only marginal differences in the radial growthrates _{of polypropylene spherulites resulted from}

theadditionofionomer. This is as_expected for immis-cible blendswithminimal interfacialinteractions. Indeed, the freeze-fractured surfaceofa_{compression-molded} sam-pleof80%_{PP/20%.ionomer blend(Figure}4),asobserved

by scanning electron microscopy,6 showsthe ionomeric

phaseas _sphericaldomains embeddedin the

polypropy-lene matrix with very little adhesion between the two phases. Onthis basis,it is_{unlikely that}thereis

signifi-cant adhesion between the _phases in the melt, indicat-ing that AEis _{approximately} equal to zero or _positive.

Thus, the relative extent of rejectionand_{engulfment of}

the ionomer domainswill be determined by the rate of crystallization andthe sizeofthe_dispersedphase.

At lowionomerconcentrationsandlowcrystallization temperatures, significant energy dissipation must be involvedin_rejectionofthedomainsas_{the spherulitegrows.} This energy loss retardsthe radial growthofthe spher-ulites, as seen in Figure 3. _{Therefore, the radial growth} rate is _depressed as _{compared to} that of_pure

polypro-pylene.

The binaryblends alsoconsistentlyshowa_greater

pri-mary nucleation density than pure polypropylene at all compositions andcrystallization temperatures(Figure5).

poly-Macromolecules, Vol. 23, No. 1, 1990 _{Polypropylene/Polyethylene-Based Ionomer Blends} 341

Figure 4. _{SEM photomicrograph}ofthe freeze-fractured

sur-face of the compression-molded sample ofthe 80% _PP/20% ionomerblend6as obtained from the Brabender_mixing cham-ber, The ionomericphaseisseen asthe spherical entities.

40

min

50

min

60

min

70

min

Figure6. Photomicrographsillustratingthecoalescenceof the ionomer domainsas aresultof_{repeatedmelting(10-min period)}

and _{crystallization for} the 95% _PP/5% ionomer blend, at a

crystallizationtemperatureof124°C.

Figure5. _{Nucleation density, after}10min melttime,as a

func-tion of_{the percent ionomer in the blend,}at_different crystalli-zation temperatures.

propylene spherulites,as _{suggestedpreviouslyby} Willis

et al.6 based on DSC observations. This effect is most pronounced for the 95% PP/5% ionomer blend,

possi-bly reflecting thesmallerionomerdomain size and con-comitantly larger surface area available for nucleation.

Asimilar_{synergistic effect upon}nucleationwas observed

by Martuscelliet al.18for

_iPP/PIB

blends. The effect,

as wellasthe observed reductioninthe_growthrate,was attributedto the_preferentialdissolutionof_imperfect

crys-tals (molecular fractionation). It is possible that these processesare alsoresponsibleforthe observations shown

in Figure5 for the _{polypropylene/ionomer}blends.

Effect of Coalescence on the Growth Rate. The overall morphology of the blends containing 5% iono-mer,as viewedwithoutcross _{polarizers at}a_temperature of124 _°C, is illustrated in Figure6. It shows the

pres-ence ofthe ionomer _phase as _{approximately spherical}

domains embeddedinthe_{polypropylenespherulites.} Ini-tially, a linear increase in the _average domain size with

melting timeisobserved,forthosedomainsof sufficient

sizeto be_{detected microscopically,}witha_changein_slope

at _{approximately}45 min as seen in Figure 7. _By

com-parison,forthe50% _PP/50% ionomer blend, the domains

are _{much larger and}a_simplelinearincreaseinthe

aver-agesizewith melting timeis_observed,as seen in Figure

8. It should be noted that, in the 50% _PP/50%

iono-Figure 7. Increase in the size ofthe ionomer domains with repeated melting (10-min period) and crystallization, for the 95% PP/5% ionomer blend, at a _{crystallization temperature}

of124°C. Thetimeindicatesthe cumulativemelttime.

Figure 8. Increase in thesize of the ionomer domains with repeated melting (10-min period) and crystallization, for the 50% PP/50% ionomer blend,ata _{crystallizationtemperature}

of124 °C.

mer _{blend, the measured diameters}haveextended beyond the thicknessofthefilm. Thus, the domainsare no _longer sphericalas inthe 95% _PP/5% ionomerblend buthave

takentheform ofdisks. If volumes,rather than diame-ters, are _plotted as a function of melt time, calculated for_{spheresin Figure}7 and disksin Figure8, then

iden-tical trends are observed. The _{important point} is that

thesizeincreaseswithmelt time,regardlessoftheshape

DOMAIN SIZE _(pm)

Figure 9. Ionomer domainsize distribution inthe 95% PP/

5%ionomer blend and 80% PP/20% ionomerblend obtained from_{cryogenically}fracturedsurfaces.

is much _greater for thisblend than for that containing

5% ionomer.

The increase in ionomer domainsize could be due to either or both of_{two factors, namely, phase separation} during the time spent in the melt and/or rejection (or

pushingofthe domains)duringrepeatedcrystallization.

AsshowninFigure7,for the95%PP/5% ionomerblend, the _average size of the ionomer domains increases

rap-idly until thefourth_{crystallization. Upon}further

crys-tallization,therate ofincreasein theaveragesizeofthe

domains isdiminished. Thissuggeststhat,at the point

wherethe_change in _{slope occurs,}thereis a_{depletion in}

thenumberofdomainsofradiuslessthan acriticalsize.

Ionomerdomainsofradiuslessthan the criticalsize are

beingrejected bythe crystallizing

front

andas a conse-quence are _coalescing. Therefore, coalescence_{not only} is_{occurring while thepolypropylene}isin the meltphase

butalsoisinduced_{bycrystallization.} Inthecase ofthe

50% _PP/50% ionomer blend, wheredomains are _larger than thecritical size,coalescence occurs _{predominantly}

in the melt. This was confirmed by _separate

experi-ments that showed increases in domain size with melt

time when the crystallization steps were _omitted, also shown _{in Figure}8 (filledsquare).

The

initial

size distribution in the ionomer _phase in theblendswas determined from cryogenically fractured

surfaces_{observed using scanning electron microscopy and} isshownin_Figure9.6 Thesizedistributions afterrepeated

crystallizations at124°C as determined_using

photomi-croscopyare shownin Figure 10. The_{sharpdrop-off}in thenumberofdomainsforthefirst _{crystallization}

(Fig-ure 10) _{probably reflects}limitations in detection at the magnificationlevelused. It isevidentthat

_initially

it is

predominantly the smaller domains(lessthan1_pm)that are _{beingrejectedby the crystallizing}frontand

coalesc-ing into the larger ones. In the case of the 50% _PP/ 50% ionomer _{blend (Figure} 8), the size ofthe domains iswell abovethat critical size. _{Even though} some iono-mer _maybepresentwithin interlamellarregions, which may give rise to some _{rejection, domain} size increase is

occurringpredominantly as a resultofcoalescence dur-ing the timethat the polypropyleneis in the melt.

As mentioned before, it is predicted that the radial growthrates _{of spherulites} in _{phase-separated} polymer

blendscan be_{influenced by the}sizeof_{the dispersed}

par-ticles. Estimateswere madeoftheeffectofdomainsize

on _{energy terms Ely Z?3,} and E4, using the equations of

Bartczakand co-workers19 shownin Table_{II, for}the_PP/ ionomerblends. The_{results, assuming} AF= ₁

erg/cm2

and

_7^

= ₁

erg/cm2,forthe95% _PP/5% ionomerblend

at a_spheruliticradialsizeof30 pm anda _growthrate of

Macromolecules, 23,

Figure 10. Ionomer domainsizedistributionin the95% _PP/ 5% ionomer blend attotal melttimesof(a) 20min, (b)40_min,

(c) 50min, and (d)80minanda_{crystallizationtemperature}of

124 °C. The melt times correspondto repeated melting and crystallizationat 10-minintervals.

0 2 4 6 8 10

RADIUS (pm)

Figure 11. Estimated energies ofrejection, engulfment, and deformationforthe 95% _PP/5% ionomerblendas a function

ofionomer domainsizeatacrystallizationtemperatureof118 °C.

0.92_pm/s,are shown_{in Figure} 11for the crystallization

temperatureof118°C. Itshouldbenotedthatthe inter-facial_{free energies}were choseninaccordancewithresults

previouslyreported for polypropylene/polyethylene23 at

140°C. The viscosityof_{the blend at the crystallization}

temperaturewas estimatedfrommeasuredviscositiesat higher temperatures (~200°C)24andlowshearrate (17 s-1) assuming Newtonianflow andusing the conceptof

activation _{energyfrom the Arrhenius equation.} These

equationspredict that, at small domainsize,theenergy

of rejectionhasa considerable influenceon the

spheru-litic

growth rate. Theenergies due to deformation and occlusion make much smallercontributions to thetotal

energydissipation. Increasing the interfacialfree energy between_{the polymers}inthe blendhasthe effectof

increas-ing the energy of deformation and engulfment.

How-ever, for reasonable changes (AFand_7p8 from0.1 to 10

erg/cm2), these energies remain much smallerthan the

energyofrejection. As the ionomerdomainsize increases,

thetotal of_{these energies}decreasesin magnitudeuntil,

for domain size _greater than4 pm, they are _essentially negligible.

Theinclusionofthese energy terms ineq2allowsthe predictionofthe_{growth rate at}acertainionomerdomain

size. As shown in Figure 12_{for the crystallization}

tem-perature of118°C,an increaseinthe_{radial growth}rate

Macromolecules, Vol. 23, No. 1,1990

RADIUS(Atm)

Figure 12. _{Theoretical growthrate dependence}on the

iono-mer domainsizeasestimatedforthe95%_PP/5%ionomer blend ata _{crystallization temperature of}118°C.

Figure13. Growthrate dependenceof the95% _PP/5%

ion-omer blendwithrepeatedmelting(10-min period) and crystal-lization atvariouscrystallizationtemperatures.

size_{leading toward thegrowth rate}of_{the pure polymer}

at the same _temperature. Abovea certain domain size (4 Mm), where the_energy ofrejectionis _essentially

neg-ligible, the growth rateofthe crystallizing polymerisno

longer affected by the presence of the second

compo-nent.

The effect of domain size on the growth rate ofthe

polypropylene spherulites can be observed

experimen-tallywhen _{blends containing}5% and 10% ionomer are repeatedly meltedfor10minandcrystallized. Asshown

inFigures13and14, an initialincreasein the radial growth

rate is observed,followed bya decrease when the

sam-ple has been crystallized several times. This gives rise toa maximum growthrate whichis_dependentupon the blend composition and _{the crystallization temperature.}

It is_interestingto notethatif themaximum_{radial growth}

rates, ratherthan the values for 10 _min, are _{plotted in} Figure3,the growthratesare _{essentiallyindependent}of ionomer concentration at all temperatures. When the

50% _PP/50% ionomer blend is _{repeatedly crystallized,}

Polypropylene/Polyethylene-Based Ionomer Blends 343

TIME IN THE MELT (min)

Figure14. Growthrate dependenceofthe 90%_PP/10%

ion-omer blendwithrepeatedmelting(10-min period) and crystal-lizationat variouscrystallizationtemperatures.

Figure15. Growth_{rate dependence}ofthe 50% _PP/50%

ion-omer blendwith_{repeatedmelting(10-min period) and}

crystal-lizationata _{crystallization temperature of}124°C.

onlyadecrease _{in the radial growth rate} is _observed,as

shown_{in Figure} 15.

The

initial

increasein_{the growth rate (Figures}13and 14) ismainlydueto_{the rejection}ofthe smaller ionomer domains(lessthan thecriticalsize)into interspherulitic regions. Since thesizeofthe domains in the95%

_PP/

5% ionomerblendis

_initially

muchsmallerthanthat of

the90%_{PP/10% ionomer blend}asdemonstratedin

Fig-ure _{9, the growing} front _requires more _energy to effect

their_{rejection. Consequently, the} initial _{growth rate} is reduced _by a _greater extent than in the 90% _PP/10%

ionomer blend.

Through coalescence_{during crystallization and} melt-ing, thedomains increase insizesuch thata _significant

fraction is above that critical size. _{During subsequent}

crystallization, the growing front dissipatesless _energy

in rejecting these domains, due to their largersize, and

a radial_{growth rate} increase is observed _leadingto the

maximum. As mentioned previously, themaximumgrowth rate corresponds closely to theradial growthrateobtained

for pure polypropylene at the same _temperature. Fur-thermore, ata_temperatureof124_{°C, the}maximum_growth

rate for the95% _PP/5% ionomer blend occurs at

rela-tivelythesame timeas the_changein _{slope observed}for the domain size increase with meltingtime (Figure 7).

Macromolecules, 23,

This strongly supportsthe ideathatrejectionofthe

ion-omer domains affects thecoalescence_mechanism,which inturn _directlyaffectsthe_{radial growth}rateofthe

poly-propylenespherulites.

The _{speed of the crystallizing front, at any constant} small domain size, also influences the perturbation

observed inthe _growth rate of the spherulites. At high crystallizationtemperatures, where thegrowthrateis

rel-ativelyslow, lessenergyis_dissipated to effectthe

rejec-tionoftheionomer domains. Asa_{result, the}initial_growth

rate is not _decreased, with _{respect to pure}

polypropy-lene, tothe extent that it isfor a _samplethatis crystal-lized at lower _{temperatures.}

Althoughthe increasesinthe_{radial growth}rateswith

melt time, observed at short timesforthesamples con-taining 5% and 10% ionomer, are in accordance with

the equationsderived _byBartczak,the maximum inthe

growth rates _{and subsequent} decreases are not consis-tent withthe_{theoretical predictions. Two explanations}

can be offered: (1) Viscosity increases with melt time.

Since _{the transport}term in the classical Fischer-Turn-bull_equationcan bedescribedin terms oftheWLF

equa-tion, increases _{in viscosity after} several crystallizations would be expected to decrease the radial growth rate. However,an increase_{in viscosity after}coalescenceof ion-omericdomainsseems rather unlikely. Furthermore,the direct verificationofsuchan effectis_{verydifficult,}since

it _requiresameasurement oftheviscosity ofthe matrix

polymer in the blend, andthis is not a trivial task. (2)

Limited miscibility oflow_{molecular weight fractions of}

ionomer inthe _{polypropylene would}seem tooffera

bet-ter _{explanation. Increasingquantities}ofthisfraction of

ionomerduringrepeatedmelting would result ina

dimin-ished _{radial growth} rate. Oncethe effects of_rejection/

engulfment/deformationbecomenegligible and thegrowth ratehasreachedits_maximum,this_{ionomer, present}within

thepolypropylene, wouldberesponsiblefor theobserved

hindranceto _{crystallization. Therefore, throughout} the

crystallization study, there is _competition between the influencesof_{phase size}andtheinfluenceoflimited

sol-ubilityon the_growthrate. Theprocess isnotofa

sequen-tial nature but a direct competition until such time as

the domainsare _{large enough} that_theyno _longerplaya

role in the_{crystallization}process (Figure 15).

Crystallization

Behavior of the TypeIII

Spheru-lite. The _{complex crystalline} structure of

polypropy-lenemanifestsitself not only at _{the crystal lattice} level

butalso at the level ofthe polarizingmicroscope where

a_{variety of spherulitetypes have}been_{classified by}their

appearance incrossed _polarizedfields.16

Inthe_{temperaturerange}of118-126°C,asecond

spher-ulitic _{morphology, appearing highly} luminous amidst darker _spherulites,was noted andcan beseen in _Figure 16. The grainystructure observedin this photographis due to _{the ionomer trapped} within the _{polypropylene} spherulites. This spheruliticappearance was classified as_typeIII _byPaddenand Keith.16 It was latershown25"26

that_{this spheruliticmorphology crystallizes}withthe hex-agonal,fJ,crystal structure whilethe predominant form (type I,inthis study) crystallizeswiththemonoclinic, a, structure. It has also been found16,25,27 that once _type

III _{spherulitesnucleate,}theirrate of_{radial growth}is

20-70% fasterthan that oftype I.

During this study,thetype III spheruliteswere found

to behave_similarlyto the_{predominantspherulitic form}

with respect to physical interactions with the ionomer domains. In _{Figure 17,} a _{plot of}the _growth rate as a function oftime in the melt shows a _{very rapid} initial

100 pm

Figure 16. Type III spherulitic morphology (bright) amidst the muchdarker typeIas observedforthe90% PP/10%

ion-omer blend ata_{crystallizationtemperature of}120°C.

Figure 17. _{Growth rate dependence}of_{the type}III _spherulite

withrepeatedmelting (10-min period) andcrystallizationin the 90% PP/10% ionomerblend ata _{crystallization temperature}

of120°C.

radial growthrate increase,followed_byadecreasetoward

longertotalmelttimes. However, whenthis plotis

com-pared tothatoftypeIatthesame _compositionand tem-perature (Figure 13), the depression of the growth rate atashort melting time isfoundto be_{much larger. The} radial growth rateofthetypeIII spherulitesis30%greater

thanthat ofthe typeI spheruliteat the same

crystalli-zation_temperature(120°C)andisconsistentwith obser-vations previously reported byNortonandKeller.27 There-fore, themore _significantdepressionoftheinitial growth

rate is due to _{the larger quantity ofenergy required to}

effectthe rejectionofthe ionomerdomains.

Conclusion

The _{spherulitic growth} of _{polypropylene in binary}

polypropylene/ionomerblends isdisturbed to a certain

extent by the presence of the ionomer component inthe

melt. Infact, for the blendthat contains5% ionomer, a smalldepressionoftheradial growthrate is observedat

a _temperature where the crystallization _proceeds rap-idly. Thisdepressionis due_{to energydissipated} by the growingpolypropylenespherulitic front in rejecting the ionomerdomains. The dispersed ionomeris alsofound

to actas a nucleation site forthe polypropylene spheru-lites. Thenucleation effectismore _{pronounced at lower} ionomer contents _possibly due to the smaller ionomer domains beingmore efficient nucleationsites due to the

Macromolecules 1990, 23, 345-348 345 Coalescenceoftheionomer domainsoccurs _duringthe

time spent in the viscous melt. However, the coarsen-ing ofthe dispersed ionomer appears to be accelerated when the _sample is _{repeatedly melted} and crystallized.

This suggeststhat a_{crystallization-induced type} of

coa-lescence is also _{occurring. A combination} ofthese two effectsresultsinalinearincreaseofthedomainsizewith

time as a _sample is _{repeatedly melted} for 10 min and

crystallized. Thiscoalescencephenomenonaswellasthe

“speed”ofthe_{crystallizing}frontisshowntohavean effect

on the radial growth rate of_{the polypropylene}

spheru-lites. A radial growthrate depressionoccurs as the

spher-ulite is _{rejecting the} second _{component, the} effective-ness ofwhichis_dependenton the crystallization

temper-ature. _{Maximum growth} ratesare observed atthepoint

where_{the rejection}of_{the dispersed domains}isata min-imum or the size ofthe dispersed phase is well above

that criticalsize_{required for rejection.} Oncethe domain

size is _{greater than the} critical size, coalescence occurs predominantly in the melt. Afterthis point, the radial growth rate decreases as the ionomer domain size is increased. _{This radial growth rate}decrease_beyondthat

maximum cannot be explained using the equations for rejection, engulfment, and deformation.

Future workwillinvolvestudiesoftheinfluenceofthe

chemical characteristics ofthe ionomer (such as func-tionalities, neutralization,etc.)on the crystallizationand coalescence phenomenaof

polypropylene/polyethylene-basedionomerblends. _{Bymodifying the}chemical struc-ture ofthe ionomer, the interfacial free energy between

the matrix _{polymer and the dispersed} phase will be changed, which would yieldmore informationaboutthe

importance of interfacial energieson _{the crystallization} kinetics ofan immiscible blend.

Acknowledgment. We_{sincerely appreciate the} assis-tance ofF. Hamel_{in preparation} oftheblends and Dr.

B. D. Favis,

IMRI,

andProf. _{A. Eisenberg, Department}

ofChemistry,McGillUniversity, forseveralfruitful dis-cussions. _{Financial support from the} Natural Sciences

and_EngineeringResearch CouncilofCanada(NSERC)

and the Quebec Government (Fonds FCAR) is

grate-fully

acknowledged.

RegistryNo. Pro-Fax6501, 25085-53-4;Surlyn 9020,61843-71-8.

References and Notes

(1) Utracki, L. A. Polyblends-87\ NRC/IMRI Symposium

Pro-ceedings,1987.

(2) Robeson,L. M. Polym.Eng. Sci. 1984,24, 587.

(3) Shaw,M._{T. Polym.}Eng. Sci. 1982,22, 115.

(4) Martuscelli,E._{Polym.Eng. Sci.} 1984,24, 563.

(5) Willis,J.M.;Favis B. D.Polym.Eng. Sci. 1988,28,1416.

(6) Willis,J.M.;Caldas,V.;Favis, B. D.4th Annual Polymer

Pro-cessingSocietyMeeting, Orlando, FL, May 1988,submitted toPolymer.

(7) Martuscelli,E.;Silvestre,C.;Abate,G._Polymer1982,23, 23. (8) Kumbhoni, K. Polym.Prog. 1974,3.

(9) Spenadel,L. J. Appl. Polym.Sci. 1972,16,2375.

(10) Karger-Kocsis, J.; Kallo, A.;Szufner, A.; Bordor,G.; Senyei,

Zs._Polymer1979,20, 37.

(11) Danesi,S.;Porter,R. S.Polymer1978,19,448.

(12) Ermilowa, G. _{A.; Ragozina,} I. A.; Leont'eva, N. M. Plast.

Massy1969, 5, 52.

(13) Friedrich, K.Prog.Colloid Polym.Sci.1949,64, 103. (14) Martuscelli,E._{Polym.Eng. Sci.} 1984,24, 563.

(15) Favis, B.D.;ChalifouxJ. P.; Van Gheluwe P. Tech.Pap.-Soc.

Plast._Eng. 1987, 33, 1326.

(16) Padden,F. J.;Keith, H.D.J.Appl.Phys. 1959,30, 1479.

(17) Turnbull,D.; Fischer, J.C.J. Chem. Phys. 1949,17,71. (18) Martuscelli, E.; Silvestre, C.;Bianchi, L. Polymer 1983, 24,

1458.

(19) Bartczak, Z.; Galeski, A.; Martuscelli, E. Polym. Eng. Sci.

1984, 24, 1155.

(20) Omenyi,S._{N.;Neumann, A. W.;Lespinard,}G.M.; Smith,R.

P._{J. Appl.Phys.} 1981,52, 789.

(21) Smith,R. P._{J. Appl.Phys.} 1981,52, 796.

(22) Omenyi,S.N.;Neumann, A. W.J. Appl.Phys. 1976,47,3956.

(23) Oda,Y.;Hata T. Prepr. Annu. Meeting HighPolym.Soc.Jpn.

1968,May,267.

(24) Kamal, M.R.;LePoutre,P.PersonalCommunication,

Depart-mentofChemical Engineering,McGillUniversity.

(25) Keith, H. D.;Padden, F.D.;Walter,N. M.; Wyckoff, M.W. J._Appl.Phys. 1959,30, 1485.

(26) Padden,F.D.;Keith,H. D. J.Appl.Phys.1973,44, 1217.

(27) Norton,D. R.;Keller,A.Polymer 1985,26, 704.

Notes

Variable-Temperature

FT-IR Studies of Organic Carbonate

Solutions

J.A.KINGJR.*AND PETERJ. CODELLA

GeneralElectric Corporate Research and Development, P.O. Box8, Schenectady, New York12301.

Received_February 15, 1989;

Revised_ManuscriptReceivedJune15, 1989

Introduction

Although the intermolecular structure ofamorphous BisphenolApolycarbonate(BPAPC)is_ill-defined,a

num-berofsmaller, well-characterized BPA-based carbonate

systems have been examined extensively.1 These past effortshave beenoriented toward_{understanding how}

vari-ationsinfundamental conformationandstructural

pack-ingat the molecular level are manifested in the

macro-scopicpropertiesofthesematerials. Suchintra-and inter-molecular interactions in BPAPC analogues have been

studied by a _variety of_{techniques (principally, NMR,}

X-ray spectroscopy, and quantum mechanical

calcula-tions). Todate,no _{experimental work}hasbeenreported on BPAPC _{usinglow-temperature solution}IR methods. Thispaper summarizesour observationsofboth

mono-meric and _{polycarbonate solutions using}

variable-temperature infrared spectral analysis (VT-IR). These