lable at ScienceDirect

Polymer 70 (2015) 207e214

Contents lists avai

Polymer

journal homepage: www.elsevier .com/locate/polymer

High melt strength polypropylene by ionic modification: Preparation,rheological properties and foaming behaviors

Yan Li b, c, d, Zhen Yao b, **, Zhen-hua Chen c, d, Shao-long Qiu b, Changchun Zeng c, d, *,Kun Cao a, b, ***

a State Key Laboratory of Chemical Engineering, College of Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, Chinab Institute of Polymerization and Polymer Engineering, Department of Chemical Engineering, Zhejiang University, Hangzhou 310027, Chinac High Performance Materials Institute, Florida State University, Tallahassee, FL 32310, USAd Department of Industrial and Manufacturing Engineering, FAMU-FSU College of Engineering, Tallahassee, FL 32310, USA

a r t i c l e i n f o

Article history:Received 19 April 2015Received in revised form10 June 2015Accepted 15 June 2015Available online 17 June 2015

Keywords:PP ionomerRheologyFoaming

* Corresponding author. High Performance MateUniversity, Tallahassee, FL 32310, USA.** Corresponding author. Institute of PolymerizatioDepartment of Chemical Engineering, Zhejiang Univer*** Corresponding author. State Key Laboratory of Cof Chemical and Biological Engineering, Zhejiang UChina.

E-mail addresses: yaozhen@zju.edu.cn (Z. Yao),kcao@che.zju.edu.cn (K. Cao).

http://dx.doi.org/10.1016/j.polymer.2015.06.0320032-3861/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

In this paper we report a facile process to introduce ionic interactions to polypropylene (PP) based on Zn-neutralization and amine modification reactions of maleic anhydride grafted PP (PP-g-MAH) for thepreparation of high melt strength PP (HMSPP). A series of PP ionomers with different structures and ionicproperties were synthesized via the choice of the pendant groups, and their effects on the rheologicalproperties were investigated. Increased melt viscosity and elasticity resulting from the ionic interactionand aggregation of the ionic domains were confirmed by both shear and extensional rheological mea-surements. The rheological behaviors of the PP ionomers share features similar to those of critical gels.Moreover, foaming behaviors of the PP ionomers were investigated by using CO2 as the foaming agentand compared to that of the original PP-g-MAH. All ionomers show significantly improved foamability. Ofthe group of PP ionomers, the amine-modified PP ionomers (Zn-AM-ionomer) shows the best foam-ability due to the combined effects of optimal rheological properties and the underlying favorable in-teractions with CO2. These findings provide new insights into the design of high melt strengthpolypropylenes for foaming application assisted with CO2.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Polypropylene (PP) foams have been attracting increasing in-terest as lightweight and low cost materials suitable for a variety ofapplications ranging from protective equipments, insulation,cushions, damping to absorbers for acoustic and electromagneticwave [1e4]. Foaming by supercritical carbon dioxide (scCO2) isparticularly appealing because of the environmental friendly na-ture of the process [5]. The key technological constraints in PPfoaming process are the low melt strength and narrow foaming

rials Institute, Florida State

n and Polymer Engineering,sity, Hangzhou 310027, Chinahemical Engineering, Collegeniversity, Hangzhou 310027,

zeng@eng.fsu.edu (C. Zeng),

temperature range [6,7], therefore improving the melt strength ofPP is highly desired [8]. Great efforts in this direction have beenmade and various modification methods have been developed,such as blending PP with high melt viscosity polymers, chemicalcross-linking and long chain branching [9e13].

We recently developed a new strategy to prepare high meltstrength PP by introducing polar groups, such as amine onto thepolymer structure [14]. The disparity in polarity between thepolymer backbone and the pendant groups may lead to phaseseparation, and the phase separated polar domains serve as phys-ical cross-links. This resulted in significant changes in the me-chanical and rheological properties of the PPs whose behaviors inmany ways resemble those of PPs modified by chemical-crosslinking. This physical cross-linking approach presents aviable general strategy to alter the viscoelastic properties of PPs andimprove their foamability, due to the characteristic features ofreversible equilibrium [15e19] and the potential self-healing ability[20,21] of the physical networks.

Physical crosslinking networks by ionic interactions may pre-sent another means to improve the PPs melt strength [22e28]. It

Y. Li et al. / Polymer 70 (2015) 207e214208

has been reported that such interactions in a number of polymersystems lead to significant changes of rheological properties[29e39]. Nevertheless studies on ionically crosslinked PPs (PPionomers) are very limited, and the foaming behaviors of PP ion-omers have not been reported. In this study, ionically modified PPswith different degrees of ionic strength and steric hindrance wereprepared to investigate the influence of ionic crosslinking on therheological and foaming behaviors of these PP ionomers. Their ef-fects on shear and extensional rheological behaviors were firstlyexamined. Moreover, simulation is performed for transient exten-sional viscosity of PP ionomers over a broad range of extensionalstrain rate using the modified WM constitutive equation based onexperiment results. Furthermore, foaming behaviors of the PPionomers with carbon dioxide (CO2) were studied and related tothe rheological properties of the PP ionomers and their interactionwith CO2.

2. Experimental section

2.1. Materials

PP-g-MAH (MAH content, 0.3 wt%) was obtained from NingboNengzhiguang New Material Co., Ltd, China. Ammonia (AM, 30%aqueous solution), methylamine (MAM, 40% aqueous solution) anddimethylamine (DMAM, 40% aqueous solution) were purchasedfrom Aladdin. Zinc acetate was supplied by Shanghai MeixingChemical Reagent Co., Ltd, China. 1,2,4-trichlorobenzene (TCB) and2,6-di-t-butyl-4-methylhenol (BHT) were purchased from Acros.

2.2. Sample synthesis

Fig. 1 shows the schemes for the synthesis of ionically modifiedPPs. Two types of ionomers were prepared: PP ionomers (Zn-ion-omer, structure shown in (a) of Fig. 1) and amine(A)-modified PPionomers (Zn-A-ionomers, structures shown in (b)-(d) of Fig. 1).Using PP-g-MAH as the precursor, the Zn-ionomer was synthesized(route 1) by hydrolyzing the maleic anhydride groups followed bythe carboxylate e zinc ions chelating complex formation. In atypical process ~40 g PP-g-MAH was dissolved in 1500 g of xyleneat 130 �C. A stoichiometric amount of zinc acetate in water wasadded into the solution and the reactionwas allowed to proceed for2 h. The ionomers were then recovered by precipitation in acetoneand repeatedly washed with ethanol. Samples were air-dried for1e2 days first and further dried in a vacuum oven at 80 �C for atleast 24 h.

Several types of Zn-A-ionomers (Zn-AM-ionomer, Zn-MAM-ionomer and Zn-DMAM-ionomer) were synthesized via route 2(Fig. 1) using AM, MAM and DMAM as the amine source, respec-tively. Typically, 40 g PP-g-MAH was dissolved in 1500 g of xyleneat 130 �C. Half of the stoichiometric amount of zinc acetate inwaterwas added into the solution together with 10-fold excess of amines.The reaction conditions and subsequent purification and dryingprocedures were the same as those for the preparation of Zn-ionomer.

2.3. Structure analysis

2.3.1. Size-extrusion chromatography (SEC)SEC was performed at 150 �C using high temperature chroma-

tography (Viscotek 350A, Viscotek Ltd.) with triple detectors (arefractive index detector, a four-capillary differential viscometerdetector and a low angle light scattering detector). The columnused was TSK-gel column (GMHHR-H(S) HT, 300 � 7.8 mm) withTCB as the mobile phase at a flow rate of 1.0 mL/min.

2.3.2. Fourier transform infrared (FTIR) spectroscopyFT-IR (Nicolet 5700, Thermo Ltd) was performed on thin film

samples (100 mm) in a range of 400e4000 cm�1 with a resolution of2 cm�1. Data were collected and averaged over a total 32 scans.

2.4. Rheological measurement

Shear rheomtetry were performed on a stress-controlledrheometer (Rheostress 6000, ThermoHaake co.) in nitrogen envi-ronment using parallel plate geometry (d ¼ 20 mm). Disk speci-mens were compression molded at 180 �C. Frequency sweepmeasurements (0.1e628 rad/s) were conducted at 180 �C using asmall strain amplitude of 1%. Creep tests were also performed at180 �C. Small shear stresses between 2 and 5 Pa were adopted toensure a linear viscoelastic response. At sufficiently long time(2500e4000 s), the deformation rate approaches to steady stateand the zero shear viscosity can be determined.

Uniaxial extensional measurements were performed at 180 �C,with a MARS III rheometer (ThermoHaake co.) equipped with aSentmanat extensional rheometer (SER) universal testing platform(SER-HV-H01 model, Xpansion Instruments). Experiments wererun at constant elongational rates ranging from 0.03 to 3 s�1.

2.5. Solubility measurement

A magnetic suspension balance (MSB) (Rubotherm Prazision-smesstechnik GmbH, Germany) was used to determine the solu-bility of CO2 in the PP-g-MAH and PP ionomers. Measurementswere conducted at 157 �C and 23 MPa. Detailed descriptions of thisdevice and experiment procedures used in this study can be foundin literature [40]. The mass of dissolved CO2, Wg was calculated bythe following equation:

mg ¼ msðP; TÞ �msð0; TÞ þ rgðP; TÞfVsðP; TÞ½1þ SsðP; TÞ� þ VBg(1)

where ms(P,T) and ms(0,T) are the sample weights at the mea-surement temperature T and at pressure P and zero pressure,respectively; r(P,T) is the CO2 density; Vs(P,T) is the sample volumeand VB is the volume of the sample container; S(P,T) is the swellingratio of sample, which can be calculated by Sanchez-Lacombeequation of state [41e45].

2.6. Foaming by supercritical carbon dioxide (scCO2)

Batch foaming process was conducted using a high-pressurestainless steel vessel with an internal volume of 150 mL. Afterplacing the sample and purging the system with CO2, a pre-determined amount of liquid CO2 was added into the vessel. Thevessel was immersed in an oil bath and kept at the targeted tem-perature and pressure (157 �C and 23 MPa) for 8 h. CO2 was thenquickly released (in ~2 s) to induce foam nucleation and growth.The foam structure was fixed by quickly removing the vessel fromthe oil bath and placing it into ice water.

2.7. Foam characterization

The foam bulk foam densities (rf) were determined via waterdisplacement measurements following ASTM D792-08:

rf ¼a

aþw� brwater (2)

where rwater is the density of water, a is the apparent mass ofspecimen in air without the sinker, b is the apparent mass of

Fig. 1. Schematics of the PP ionomers structures and synthetic routes using PP-g-MAH as the precursor.

Table 1Polymer characterization results for PP-g-MAH as the precursor and the modified PPionomers.

Samples Zn/MAH Mw (kg/mol) PI

PP-g-MAH 0 202 2.9Zn-ionomer 1 197 2.8Zn-Am-ionomer 0.5 200 2.7Zn-MAM-ionomer 0.5 190 2.5Zn-DMAM-ionomer 0.5 185 2.6

Y. Li et al. / Polymer 70 (2015) 207e214 209

specimen and sinker completely immersed in water, and w is theapparent mass of the totally immersed sinker. The volume expan-sion ratios (Rv) of the foamed samples are defined as:

Rv ¼ r0rf

(3)

where r0 is the bulk density of the unfoamed materials.The cell size and density were obtained using scanning electron

microscopy (SEM) (Sirion, FEI Ltd). Foamed samples were freeze-fractured in liquid nitrogen and sputter-coated with a thin layerof gold. The cell density (N), defined as the number of cells per unitvolume with respect to the unfoamed polymer, was determinedfrom:

N ¼�nM2

A

�3=2Rv (4)

where n is the number of cells in the SEM image, M is the magni-fication factor and A is the area of the image.

Fig. 2. Molecular weight and its distribution of the PP-g-MAH and PP ionomers.

3. Results and discussion

3.1. Characteristic structure of the modified PP ionomers

Table 1 and Fig. 2 summarize the molecular weight (MW) andmolecular weight distribution (PI) of the PP-g-MAH and PP ion-omers. The MW and PI of all PP ionomers were similar to those of

Y. Li et al. / Polymer 70 (2015) 207e214210

the parent PP-g-MAH, and were not observably affected by theamine-modification/neutralization reactions.

Fig. 3 shows the FTIR spectra of the PP-g-MAH and PP ionomers.The major peak in PP-g-MAH spectrum at 1780 cm�1 correspondedto the anhydride group [26]. This peak was not present in thespectra of any of the PP ionomers, suggesting completion of theneutralization reaction and consumption of the anhydride group.For Zn-ionomer, the bands around 1560e1600 cm�1 were assignedto the carboxylate stretching vibration of the zinc carboxylate. ForZn-A-ionomers, the amide bands (around 1630 cm�1 and1555 cm�1) overlapped with zinc carboxylate bands and weredifficult to resolve independently [46,47].

Fig. 4. t/J(t) as a function of creep time (t) at 180 �C for the PP-g-MAH and PPionomers.

3.2. Rheological behavior

3.2.1. Zero shear viscosityFig. 4 shows the time evolution of the ratio t/J(t) for all samples.

The zero shear viscosity (h0) was determined according to Eq. (5)and is summarized in Table 2.

ho ¼ limt/∞

tJðtÞ (5)

where t is the creep time and J(t) is the compliance of the materials.The zero shear viscosities of the ionomers were several times to

orders of magnitude higher than that of parent PP-g-MAH. Sincetheir molecular weights and polydispersities were similar, thesubstantial viscosity differences can be argued to originate from thetwo main features in these ionomers: presence of inter-chain ionicinteractions and formation of a nanophase-separated structureinduced by the ionic aggregates within the materials[29e36,48,49]. Themagnitudes of the zero shear viscosities were inthe following order: Zn-ionomer > Zn-AM-ionomer > Zn-MAM-ionomer > Zn-DMAM-ionomer. The highest melt viscosity in theZn-ionomer might be reasonably attributed to the highest Znconcentration and ionic association. On the other hand, steric effectplayed a role in the Zn-A-ionomers. An increase in the size andnumber of the pendant groups led to reduced interaction anddecrease of zero shear viscosity.

3.2.2. Linear viscoelasticity by small amplitude oscillatoryrheometry

Fig. 5 compares the complex viscosity jh*j of the PP-g-MAH andPP ionomers. At low frequency regime (terminal zone), the

Fig. 3. FTIR spectra of the PP-g-MAH and PP ionomers.

ionomers showed significantly higher complex viscosity, whosemagnitude followed the same order as that observed in creep test,i.e., Zn-ionomer > Zn-AM-ionomer > Zn-MAM-ionomer > Zn-DMAM-ionomer > PP-g-MAH. In addition, the strong shear thin-ning behavior observed in the ionomers throughout the frequencyrange also followed the same order. These behaviors were consis-tent with the strength of the ionic interactions in these materials,and are typical of percolated structures from the physical crosslinksin the ionomers. By comparison, the PP-g-MAH exhibited lowerviscosity, and a Newtonian plateau in the low frequency regime.

The differences in linear viscoelasticity of the materials werefurthered studied by examining their storage and loss moduli(Fig. 6). The PP-g-MAH showed typical liquid-like terminalbehavior, e.g., G0 ~ u2 and G00~ u at low frequencies. On the otherhand, all PP ionomers showed more solid-like behaviors withsubstantially higher G0 and G00. Moreover, the terminal storage andloss moduli of all ionomer exhibited similar power-law type fre-quency dependency (both G0 and G00 ~ un).

The gel-like behavior and associated network characteristics ofthe ionomers were analyzed using the critical gel theory developedbyWinter et al. [15] According to the linear viscoelastic model for acritical gel, the G0 and G00 can be evaluated as [15,50].

G0ðuÞ ¼ Sun cosðnp=2ÞGð1� nÞ 1�lpg <u<1=l0 (6)

G00 ðuÞ ¼ Sun sinðnp=2ÞGð1� nÞ 1

�lpg <u<1=l0 (7)

where S is the gel stiffness and n (power law index) is the criticalnetwork relaxation exponent (0 < n < 1). n ¼ 1 indicates pre-dominantly viscous behavior, whereas n ¼ 0 indicates elasticbehavior. The characteristic time l0 represents a low cutoff due toglassy behaviors at shorter times or at higher frequencies. lpg in-dicates the lifetime of physical junctions. G(n) is the gammafunction.

The measured pair of G0 and G00 for each ionomer was fittedusing the above model (data at low frequency, <1 rad/s, were used).The model predictions agreed rather well with the experimentalmeasurements (Fig. 6). Table 2 summarizes the gel stiffness S andnetwork relaxation exponent n for each ionomer. The decrease of Sand increase of n followed the same order observed previously: Zn-ionomer, Zn-AM-ionomer, Zn-MAM-ionomer and Zn-DMAM-ionomer. This suggests that as the strength of the ionic associa-tion decreases, the network stiffness and elasticity decreases.

Table 2Rheological properties of PP-g-MAH and PP ionomers.

Samples h0 (Pa$s) ucr (rad/s) n S (Pa$sn) Ge (Pa) Solubility (g/g)a

180 �C

PP-g-MAH 1530 244 e e e 0.062Zn-ionomer 44,000þ 71 0.66 1200 10,000 0.067Zn-Am-ionomer 25,500þ 107 0.73 780 5000 0.082Zn-MAM-ionomer 13,300þ 154 0.75 500 4000 0.081Zn-DMAM-ionomer 8700þ 170 0.8 400 2000 0.082

Notes: þ actual zero shear viscosity is higher than the value reported here.a Solubility of CO2 in the PP-g-MAH and PP ionomers at 157 �C and 24 MPa.

Fig. 5. Complex viscosity (jh*j) vs. angular frequency (u) measured at 180 �C for thePP-g-MAH and PP ionomers.

Y. Li et al. / Polymer 70 (2015) 207e214 211

Furthermore, the chain segmental relaxation of the materialswas investigated by comparing their crossover frequencies (ucr). Alower ucr indicates lower mobility of the chain segment and longerrelaxation time. As shown in Table 2, ucr showed the followingorder: Zn-ionomer < Zn-AM-ionomer < Zn-MAM-ionomer < Zn-DMAM-ionomer < PP-g-MAH. The reduction in the chain segmentmobility and increase in segmental relaxation time in the series ofPP ionomers is consistent with the previous zero shear viscosityobservation and critical-gel analysis.

The series rheological study consistently suggested that theionic domains (from ionic aggregation and phase separation)caused significant changes of the rheological properties of the ionic

Fig. 6. (a) Storage modulus (G0), (b) loss modulus (G00) vs. angular frequ

PPs. The changeswere dictated by the strength of ionic interactions,which were influenced by the ion concentration and steric hin-drance. Due to the highest concentration of zinc ions, the Zn-ionomer exhibited the highest zero shear viscosity, highest gelstiffness and lowest network relaxation exponent. In the Zn-A-ionomers, as the steric hindrance increased, the interactionstrength decreased with reduction of network characteristics.

3.2.3. Uniaxial elongational rheologyThe uniaxial elongational viscosity was investigated. Fig. 7

shows the tensile stress growth coefficients for the PP-g-MAHand ionomers under a series of strain rates at 180 �C. For PP-g-MAH, the tensile stress growth coefficients increased graduallywith no clear “strain hardening”, typical of a linear polymer. Bycontrast, all ionomers showed strong stress hardening with thetensile stress growth coefficients rising rapidly above the linearviscoelasticity limit at all elongational rates. Such stress hardeningphenomena were direct manifestations of the networked micro-structures in the ionomers whose presence and interactionsrestricted the polymer chain mobility and enhanced the systemelasticity.

In light of the similarity on the shear rheological behavior be-tween the ionomers and critical gels, we attempted to predict thetransient elongational flow behavior of the ionomers by using theconstitutive equation developed byWinder and Mours [15,50]. TheWM model is expressed as the following:

sðtÞ ¼Zt

0

Sðt � t0Þ�n v

vt0Bðt; t0Þdt0 (9)

ency (u) measured at 180 �C for the PP-g-MAH and PP ionomers.

Fig. 7. The measured tensile stress growth coefficients (hEþ) of the PP-g-MAH and PP ionomers at rates of 0.03, 0.1, 0.3, 1 and 3 s�1 at 180 �C.

Y. Li et al. / Polymer 70 (2015) 207e214212

where s(t) is the stress tensor, B(t,t΄) is the Finger strain tensor, andS and n are the stiffness and relaxation time exponent of the criticalgel.

The comparison of the experimental data andmodel fit is shownin Fig. 7 (solid line). The model under-predicted the tensile stressgrowth, with larger discrepancy at lower strain rates. The differencemay arise from several characteristics unique to the ionomernetwork structures. In the ionomers, the physical cross-links arethe ionic aggregates from the phase separation. They possess finitesize and have their own distinct viscoelastic properties, both ofwhich are affected by the ionic strength and steric hindrance,therefore may vary among the ionomers. It is plausible that theelasticity and deformation characteristics of the ionic aggregateswould impact the overall behavior of the ionomers. These featuresare absent in critical gels where the cross-links are covalent bondsfrom polymer chain strands.

To account for the contribution of the finite-sized viscoelasticionic aggregates to the ionomers' transient elongation behavior, wemodified the critical gel constitutive equation by adding an addi-tional integral term:

sðtÞ ¼Zt

0

Sðt � t0Þ�n v

vt0Bðt; t0Þdt0 þ

Zt

0

GehðlÞ v

vt0Bðt; t0Þdt0 (10)

where Ge can be considered as the long-term elastic modulus ofphysical cross-links (ionic aggregates); the damping function,h(l) ¼ 1/l [2], represents the extent of destruction of the physicalcross-links by deformation; l is the principal stretch ratios.

Fig. 7 shows the modified model predicted tensile stress growthcoefficients (dotted lines). It captures rather well the transientbehavior of all ionomers albeit somewhat over-prediction of themaximum tensile stress growth coefficients. To our knowledge, thisempirical model for the first time aims to provide a possible meansto take into consideration of the characteristics of the physicalcross-links and evaluate their influences on the extensional flowbehavior. The model also revealed that the physical cross-links inPP ionomers exhibited considerable strength, characterized by themodel parameter Ge (Table 2). The magnitude of Ge followed thesame order of the strength of the ionic association in PP ionomersdiscussed previously.

In addition to cause the critical gel behavior discussed herein,the ionic aggregates (clusters) may also contribute to the systemviscoelastic properties via a “filler effect”, evidenced by the higherstorage moduli of all ionomers than that of PP-g-MAH at high

frequencies in the oscillatory measurement (3.2.2). Thus, both thelinear and nonlinear rheological behaviors of ionomers reflect twoeffects that collectively contribute to the highly heterogeneousdynamics of PP ionomers: a network of junctions that give rise tothe critical gel behavior and the “filler” effect derived from thecluster properties.

3.3. Foamability of the ionomers

From the bubble growth dynamic point of view, rheologicalfactors that impede bubble growth are beneficial in stabilizing thebubble growth [51]. Thus with increased elasticity, the ionomerswould be anticipated to show greatly improved foamability. Toevaluate this, batch foaming was conducted using supercritical CO2at 157 �C and 24 MPa. Fig. 8 shows the cellular structures of thefoams observed by SEM, and Fig. 9 shows the average expansionratio, cell diameter and cell density. Indeed, the PP ionomersgenerally showed significantly improved foamability, displaying ahigher expansion ratio, higher cell density and more uniformcellular structure than PP-g-MAH.

During bubble growth the polymer experiences biaxial exten-sion [52]. While the increase of melt strength helped reduce thematrix failure and foam coalescence and collapse under the stressfield, higher elasticity is not always more beneficial becauseexcessive elasticity would result in a matrix with high stiffness andreduced strain-at-break. This leads to the decreased foamability asbubble growth relies on the matrix polymer to undergo substantialstrain without fracturing [53]. This effect was observed in thefoaming of the PP ionomers. For example, the Zn-AM-ionomer hada lower elasticity and melt strength than Zn-ionomer, yet exhibitedthe best foamability (highest expansion ratio and cell density, andthe most uniform cell morphology). Similar phenomenon havebeen observed in critical gels, in which strong gels are likely to bemore brittle than weak ones and may not be deformed to the sameextent without fracturing [37]. The best overall foamability of Zn-AM-ionomer might be attributed to the good balance betweenstrength and deformability from the moderate enhancement ofelasticity. Such behavior was predicted in our previous foamingsimulation [51].

CO2 solubility might be another contributing factor to theincreased foamability. As shown in Table 2, due to the enhancedpolymer-CO2 interactions [54e56], all three Zn-A-ionomers hadsubstantially higher CO2 solubility than the Zn-ionomers, whichhad solubility comparable to that of the parent PP-g-MAH. The

Fig. 8. Cell morphology of the PP-g-MAH and PP ionomers foamed at 157 �C and 24 MPa.

Fig. 9. (a) Cell expand ratio, (b) cell diameter and (c) cell density of the PP-g-MAH and PP ionomers foamed at 157 �C and 24 MPa.

Y. Li et al. / Polymer 70 (2015) 207e214 213

higher solubility not only enhanced cell nucleation and increasedthe cell density [55], it also facilitated bubble growth.

4. Conclusions

By utilizing Zn-neutralization and amine modification reactionsof maleic anhydride grafted PP (PP-g-MAH), a series of themodifiedPP ionomers with different strength of ionic associations wereprepared. The PP ionomers showed improved melt elasticity, asrevealed by a series of rheological measurements. Through thechoice of the pendant group, both shear and extensional rheolog-ical properties could be conveniently tuned. The shear and exten-sional behaviors of the ionomers were analyzed by resorting to acritical-gel theory, which satisfactory describes the dynamic shearbehavior. Furthermore, a modified Winder - Mours constitutiveequation was proposed to include the effects of the ionic domaincharacteristics of the PP ionomers on the extension flow behavior.The model prediction agreed rather well with experimental results.Finally, foaming of the PP ionomers by supercritical CO2 wasstudied. Zn-AM-ionomer showed the best foamability due to theoptimal rheological properties and the enhanced solubility result-ing from amineeCO2 interaction. These findings suggest that ionicmodification may be a new viable approach to prepare high meltstrength PP for foaming with CO2. Further work should be targetedat controlling the rheological behavior and foaming properties bychanging the types of cations, the degree of neutralization, and

other factors that are expected to influence the interaction betweenion-pairs in polymer chains.

Acknowledgment

This work was supported by the National Natural ScienceFoundation of China through NSFCMajor Project No. 50390097 andNSFC Project Nos. 50773069 and 51173166, and the Program forChangjiang Scholars and Innovative Research Team in University(No. IRT0942). Funding support from Center of Excellence inAdvanced Materials of the State of Florida USA is alsoacknowledged.

References

[1] L.J.M. Jacobs, M.F. Kemmere, J.T.F. Keurentjes, Green. Chem. 10 (2008)731e738.

[2] H.E. Naguib, C.B. Park, S. Song, Ind. Eng. Chem. Res. 44 (2005) 6685e6691.[3] W.T. Zhai, T. Kuboki, L.L. Wang, C.B. Park, E.K. Lee, H.E. Naguib, Ind. Eng. Chem.

Res. 49 (2010) 9834e9845.[4] L.J. Lee, C. Zeng, X. Cao, X.M. Han, J. Shen, G.J. Xu, Compos. Sci. Technol. 65

(2005) 2344e2363.[5] Y. Li, Z. Chen, C. Zeng, in: V. Mittal (Ed.), Polymer Nanocomposite Foams, CRC

Press, Boca Raton, 2014, pp. 1e33.[6] N. Pasquini, Polypropylene Handbook, second ed., Hanser, Munich, 2005, pp.

347e351, 473e488.[7] J. Stange, H. Munstedt, J. Rheol. 50 (2006) 907e923.[8] A.D. Gotsis, B.L.F. Zeevenhoven, J. Rheol. 48 (2004) 895e914.[9] P. Cheung, D. Suwanda, S.T. Balke, Polym. Eng. Sci. 30 (1990) 1063e1072.

Y. Li et al. / Polymer 70 (2015) 207e214214

[10] W. Weng, W. Hu, A.H. Dekmezian, C.J. Ruff, Macromolecules 35 (2002)3838e3843.

[11] Z.B. Ye, F. Alobaiidi, S.P. Zhu, Ind. Eng. Chem. Res. 43 (2004) 2860e2870.[12] K. Cao, Y. Li, S. Wu, Z. Chen, Z. Yao, Z. Huang, J. Appl. Polym. Sci. 121 (2011)

3384e3392.[13] Z. Yao, Z. Lu, X. Zhao, B. Qu, Z. Shen, K. Cao, J. Appl. Polym. Sci. 111 (2009)

2553e2561.[14] Y. Li, Z. Yao, S. Qiu, Z. Chen, F. Zhu, C. Zeng, K. Cao, Ind. Eng. Chem. Res. 52

(2013) 7758e7767.[15] H.H. Winter, M. Mous, Adv. Polym. Sci. 134 (1997) 165e234.[16] S. Sivakova, D.A. Bohnssack, M.E. Mackay, P. Suwanmala, S.J. Rowan, J. Am.

Chem. Soc. 127 (2005) 18202e18211.[17] F.J. Stadler, W. Pyckhout-Hintzen, J.M. Schumers, C.A. Fustin, J.F. Gohy,

C. Bailly, Macromolecules 42 (2009) 6181e6192.[18] F.J. Stadler, T. Still, G. Fytas, C. Bailly, Macromolecules 43 (2010) 7771e7778.[19] C.X. Sun, M.A.J. van der Mee, J.G.P. Goossens, M. van Duin, Macromolecules 39

(2006) 3441e3449.[20] K. Chino, M. Ashiura, Macromolecules 34 (2001) 9201e9204.[21] P. Cordier, F. Tournilhac, C. Souli�e-Ziakovic, L. Leibler, Nature 451 (2008)

977e980.[22] K. Matsumura, O. Fukumoto, J. Polym. Sci. Part A Polym. Chem. 9 (1971)

471e483.[23] L.M. Landoll, D.S. Breslow, Polypropylene Ionomers, J. Polym. Sci. Part A

Polym. Chem. 27 (1989) 2189e2201.[24] R.A. Weiss, P.K. Agarwal, J. Appl. Polym. Sci. 26 (1981) 449e462.[25] M. Fujiyama, M. Kondou, K. Ayama, H. Inata, J. Appl. Polym. Sci. 85 (2002)

762e773.[26] M. Fujiyama, M. Kondou, K. Ayama, H. Inata, J. Appl. Polym. Sci. 86 (2002)

2887e2897.[27] W. Yu, J. He, Polymer 41 (2000) 891e898.[28] D.D.J. Rousseaux, X. Drooghaag, M. Sclavons, P. Godard, V. Carlier,

J. Marchand-Brynaert, Polym. Degrad. Stab. 95 (2010) 363e368.[29] R. Longworth, J. Vaughan, Nature 218 (1968) 85e87.[30] D.J. Yarusso, S.L. Cooper, Macromolecules 16 (1983) 1871e1880.[31] F. Dalmas, E. Leroy, Macromolecules 44 (2011) 8093e8099.[32] A. Eisenberg, J.S. Kim, Introduction to Ionomers, Wiley, New York, 1998, p. 2.[33] B.S. Aitken, C. Francisco Buitrago, J.D. Heffley, M. Lee, H.W. Gibson, K.I. Winey,

K.B. Wagener, Macromolecules 45 (2012) 681e687.

[34] R.A. Register, R.K. Prud'homme, in: M.R. Tant, K.A. Mauritz, G.L. Wilkes (Eds.),Ionomers: Synthesis, Structure, Properties and Applications, Chapman & Hall,New York, 1997, pp. 208e260.

[35] R.A. Weiss, W. Yu, Macromolecules 40 (2007) 3640e3643.[36] K.A. Page, J.K. Park, R.B. Moore, V.G. Sakai, Macromolecules 42 (2009)

2729e2736.[37] R.A. Weiss, H. Zhao, J. Rheol. 53 (2009) 191e213.[38] X. Qiao, R.A. Weiss, Macromolecules 46 (2013) 2417e2424.[39] G.H. Ling, Y. Wang, R.A. Weiss, Macromolecules 45 (2012) 481e490.[40] Y. Sato, T. Takikawa, S. Takishima, H. Masuoka, J. Supercrit. Fluids 19 (2001)

187e198.[41] I. Sanchez, R. Lacombe, J. Phys. Chem. 80 (1976) 2352e2362.[42] I. Sanchez, R. Lacombe, Macromolecules 11 (1978) 1145e1156.[43] Y. Sato, M. Yurugi, K. Fujiwara, S. Takishima, H. Masuoka, Fluid Phase Equi-

libria 125 (1996) 129e138.[44] Z. Lei, H. Ohyabu, Y. Sato, H. Inomata, R. Smith, J. Supercrit. Fluids 40 (2007)

452e461.[45] S. Areerat, E. Funami, Y. Hayata, D. Nakagawa, M. Ohshima, Polym. Eng. Sci. 44

(2004) 1915e1924.[46] D. Lin-Vien, N.N. Colthup, W.G. Fateley, J.G. Grasselli, The Handbook of

Infrared and Raman Characteristic Frequencies of Organic Molecules, Aca-demic Press, Boston, 1991.

[47] B.A. Brozoski, M.M. Coleman, P.C. Painter, Macromolecules 17 (1984)230e234.

[48] A.M. Castagna, W.Q. Wang, K.I. Winey, J. Runt, Macromolecules 44 (2011)5420e5426.

[49] N.K. Tierney, R.A. Register, Macromolecules 35 (2002) 6284e6290.[50] H.H. Winter, F. Chambon, J. Rheol. 30 (1986) 367e382.[51] Y. Li, Z. Yao, Z. Chen, K. Cao, S. Qiu, F. Zhu, C. Zeng, Z. Huang, Chem. Eng. Sci. 66

(2011) 3656e3665.[52] A. Arefmanesh, S.G. Advani, Rheol. Acta 30 (1991) 274e283.[53] R.G. Larson, The Structure and Rheology of Complex Fluids, Oxford University

Press, Oxford, 1999, p. 346.[54] K. Ghosal, R.T. Chern, B.D. Freeman, Macromolecules 29 (1996) 4360e4369.[55] T.B. Low, Y. Xiao, T.S. Chung, Y. Liu, Macromolecules 41 (2008) 1297e1309.[56] M.D. Soutullo, C.I. Odom, B.F. Wicker, C.N. Henderson, A.C. Stenson, J.H. Davis,

Chem. Mater. 19 (2007) 3581e3583.