# Viscoelastic Behavior and Constitutive Modeling of PP/HNT Composites Prepared by Water-Assisted Extrusion.

INTRODUCTIONPolymer nanocomposites based on unmodified or modified mineral nanofillers have attracted particular interest both in scientific and industrial fields. The interfacial interactions and degree of dispersion of fillers in a polymer matrix are key issues in determining the final performance of polymer nanocomposites [1-4].

To obtain the best properties of nanocomposites, the dispersion of nanofillers in a polymer matrix is essential during extrusion. It has been reported that a well dispersion is affected significantly by processing conditions. To maximize dispersion, extrusion using water is introduced for a polymer nanocomposite system [5-8]. Without any further modification of fillers, exfoliated polymer/filler nanocomposites are obtained by melt mixing in a twin screw extruder attached with a water injection and a degassing system. Recently, a water-assisted mixing extrusion (WAME) has been reported for other polymers, for example, poly(vinylidene fluoride) and poly(propylene) [6, 8, 9].

Halloysite nanotubes (HNTs) are novel one-dimensional natural nanomaterials with predominantly hollow tubular nanostructures and high aspect ratios. Owing to their high mechanical strength, thermal stability, biocompatibility, and abundance, HNTs can be used in exciting applications in polymer nanocomposites [10, 11]; meanwhile, the cost of HNTs is much lower than those of other nanofillers such as carbon nanotubes [12, 13]. Currently, the research of HNT nanocomposites is primarily focused on their crystallization behavior, thermal stability, and mechanical properties [14]. Few have reported the rheological behavior of HNTs in polymers. Akhlaghi et al. investigated the shear flow properties of polyacrylonitrile (PAN) solution in the presence of HNTs at an industrially relevant concentration of the polymer. In this study, the elasticity and relaxation time of PAN solutions and the availability of heterogeneous PAN domains are found to increase with HNT content [15]. Singh et al. investigated the effect of HNT loading and compatibilizer amount on morphology and rheological properties. The effects of HNT loading and compatibilizer amount on morphology and rheological properties were investigated. Rheological studies indicate that both the complex viscosity and storage modulus in oscillatory rheometry are the highest for nanocomposites prepared using 10 wt% HNTs, and reduced slightly after incorporating a compatibilizer [16].

Rheological investigations are considered to be as fundamental as other characterization techniques to understand melt flow properties, which is key in polymer processing [13, 17, 18]. Meanwhile, it provides a convenient method to evaluate the dispersion state of nanofillers in the polymer matrix [19]. To characterize dispersion, many rheological indicators can be used for reference, such as the shear thinning exponent, elastic plateau at low frequencies, and crossover frequencies along the overall pattern of G' and G" that provide a valuable fingerprint for the characterization of global exfoliation, and in determining the percolation threshold [20-24].

The Kaye-Bernstein-Kearsley-Zapas (K-BKZ) rheological constitutive integral equation has been used broadly to investigate the linear and nonlinear rheological characteristics of several different polymer melts [25, 26], Regarding polymer nanocomposites, successful applications of the constitutive equations in general and the K-BKZ constitutive model in particular have been reported by Kagarise et al. [27] for polystyrene/carbon nanofiber composites, for example, Lee et al. [28] for polypropylene/layered silicate nanocomposite, Pasanovic Zujo et al. [29] for organically modified bentonite/ethylenevinyl acetate copolymer (EVA 18) nanocomposites, etc.

In our previous study [30], the rheological response was correlated with the microstructure of PP/HNT nanocomposites. It was found that no percolation network formed in the PP/HNTs nanocomposites, since no solid-like behavior (plateau of G' at low frequencies) is observed considering the low HNTs loadings used in the experiment. Thus, the rheological behavior of PP/HNT nanocomposites is studied by increasing the content of filler. It is known that rheology can potentially offer a method to assess the state of dispersion of nanocomposites directly in the melt state. Using rheology as a screening tool offers an integrated picture of the composite material, with contributions stemming from the effects of particles on the local deformation rate of the polymer, the effect of particulate orientation, exfoliation, and the interaction between the particles [7, 16, 30, 31]. The objective of this work is to extend the investigation of polymer nanocomposite rheology to the case of PP/HNT composites prepared by water injection, to validate the application of the K-BKZ constitutive model on the predictability of the transient shear behaviors, and to determine the unique material parameters of PP/HNT composites. In this study, the effects of dispersion and HNT content on the microstructure are investigated; the linear and nonlinear rheological response, and the experimental rheological behavior of the PP/HNT nanocomposites are examined; the transient shear flow properties using the K-BKZ integral constitutive equation is predicted. The primary objective of this study is to understand the correlation between HNT dispersion and the resulting rheological properties.

EXPERIMENTAL

Materials and Sample Preparation

PP: grade F401 (Sinopec Yangzi Petrochemical Co. Ltd.), powder density 0.9 g/[cm.sup.3], melt index (MI) 2.5 g/10 min (230 DEGC, 2.16 kg). HNTs: brand Premium AD (New Zealand Imerys Asia Plus), density 2.36 g/[cm.sup.3], average diameter -80 nm, length-to-diameter ratio 10-30. Antioxidant: grade 1010 (Basf Company Ltd), white powder. A small amount (0.6 wt%) of antioxidant was added to avoid the thermal degradation of the components during blending and subsequent experiments.

Sample Preparation

The mixture of PP and HNT powder was added to the extruder with a feeding rate of 3 kg/h. The PP/HNT nanocomposite samples with three different weight ratios (95/5, 90/10, and 85/15) were prepared using the WAME. The barrel temperatures were 175[degrees]C, 180[degrees]C, 200[degrees]C, 200[degrees]C, 195[degrees]C, 190[degrees]C, and 185[degrees]C from the hopper to die, and the screw speed was 100 rpm. Water was pumped into the PP melt in the barrel at a 20 mL/min flow rate. For comparison, PP/HNT nanocomposite samples without water injection were prepared at the same processing conditions. The extrudate was pelletized at the die exit. After drying in a vacuum oven at 80[degrees] C for 12 h, the pellets were compression molded into sheets with a thickness of ~2 mm at 190[degrees]C under 15 MPa. The PP samples prepared without and with water injection are denoted as P and P-W, respectively; the PP/HNT nanocomposite samples prepared without and with water injection are denoted as P-Hm and P-Hm-W, respectively, where m represents the weight percentage of the HNTs. The PP/HNT nanocomposite and PP samples prepared are summarized in Table 1. The preparation of various specimens used for characterization is detailed in our previous work [9].

It is noteworthy that the HNTs were not chemically modified, and no compatibilizer was added when preparing the PP/HNT nanocomposite samples in this work.

Characterization

Specimens cut from the aforementioned sheets were cryofractured in liquid nitrogen. The fractured specimens were subsequently gold sputtered and examined using scanning electron microscopy (SEM) to evaluate the dispersion state of the HNTs in the PP matrix.

Rheology measurements were performed on a stress controlled rotational rheometer (Bohlin Gemini 200, Malvern Instruments, England). In these experiments, parallel plates with a diameter of 25 mm and gap size of 1 mm were used. Small-amplitude oscillatory shear (SAOS) measurements were implemented at 200[degrees]C in the oscillatory shear mode using the 25-mm parallel plate geometry; the gap was set at 1 mm, and frequency sweeps ranging from 0.01 to 100 rad/s were performed at 1 % strain to maintain the experiment within the linear viscoelastic domain. Stress relaxation measurements were performed using the controlled strain mode with a parallel plate shear flow device with a maximum strain of 500%. All tests in the linear and nonlinear viscoelastic domains were repeated several times using different batches of the mixture.

RESULTS AND DISCUSSION

Dispersion of HNTs in the Matrix

The SEM micrographs of cryofractured surfaces for the prepared PP/HNT nanocomposite samples are shown in Fig. 1. The HNTs are well dispersed in the matrix in the P-H5-W and P-H5 samples (Fig. 1a and b), but as 15 wt% HNTs was added to the PP matrix, large aggregates with a diameter of ~5 [micro]m appeared in the P-H15 sample, as shown in Fig. 1c. In general, the dispersion of HNTs in the P-Hm-W samples is better than that in the P-Hm samples, as the aggregates are relatively small (Fig. Id). Thus, it can be deduced that the two groups of samples will exhibit different rheological properties.

Dynamic Rheological Behavior

SAOS is a frequently used method to further study the influence of HNT dispersion on its rheological behavior. Figure 2 compares the efficiency of the water injection extrusion preparation method utilized in the present work over the conventional mixing preparation method. As shown in Fig. 2a, the G' of the PHm samples increase with the increase in HNTs, while for P-HmW samples, seen in Fig. 2a', the G' of the P-H5-W sample is higher than those of both the P-H10-W and P-H15-W samples in the entire frequency range. The possible reason is that the high concentration of HNTs in the PP matrix leading to coalescence of the HNT nanotubes [32], meanwhile, HNTs of 5 wt% content dispersed uniformly in the matrix, thus enhancing the interaction between the HNTs and polymer [33]. This comparison reveals the efficiency of the water injection extrusion method for preparing PP-HNT nanocomposites over the conventional preparation approach. It can be concluded that the novel and simple method used in this work led to a good dispersion resulting in large increases in complex viscosity and modulus. The influence of HNTs content on the [eta]* is similar to that on the G' (Fig. 2b and b'). As is shown in Table 2, the [eta]* of the P-H5-W sample is almost six times as high as that of the P-H5 sample at 0.1 rad/s. It is well known that the nanoparticles may exhibit strong physical adsorption or interactions with polymers [34], and that the HNTs can exhibit more contact with the PP matrix after water injection as the HNT dispersion is better. Consequently, the interaction between PP and HNTs enhanced to a large extent, thus resulting in a higher [eta]* value in the low-frequency range. Meanwhile, the [eta]* of the P-H10-W and P-H15-W samples are 7.36 times and 8.32 times higher, respectively, than that of the P-W sample, while the P-H10 and P-H15 samples are 7.22 times and 7.81 times higher, respectively, than that of the P sample at 0.1 rad/s. In general, water molecules in processing can promote the dispersion of HNTs in the PP matrix [5]. VGP plotting [35] can reveal a full mapping on the SAOS flow of two systems at various loading levels, as shown in Fig. 2c and c'. It is seen that the phase angle (S) of P-Hm-W samples are lower than P-Hm samples, which is indicative of a transition of this system from viscous flow to solid-like one. This is because HNTs in P-Hm-W samples have better affinity to the PP chain relative to P-Hm samples.

Figure 3 shows the Cole-cole plots for all prepared samples. As can be seen, semicircle arcs with small radii appear on the Cole-cole plots for the P and P-W samples owing to the linear chain structure of PP. The arcs on the plots for the P-Hm-W samples become broader and shift up to a higher viscosity region. As the loading level achieving up to 10 wt%, the local relaxation arc nearly disappears, indicating that the long-term relaxation of those restrained PP chains becomes the dominant one in the whole relaxation behaviors of the composites. While for the P-H15 sample, the local relaxation arc of PP disappears, indicating that larger aggregates exist and this is consistent with the SEM results.

Preshear is used to further verify the difference in HNT dispersion between the P-H5 and P-H5-W samples; the result is shown in Fig. 4. As SAOS measurements require a relatively long time (-13 min), a 500 s rest was imposed after each preshearing to ensure that changes in the microstructure with time was insignificant to the results. From Fig. 4, a strong dependence of G' on the pre-shear rates can be observed especially at low frequencies. For the P-H5 sample (Fig. 4a), G' increases with the shear rate from 0.01 to 1/s; this can be attributed to the fact that the HNT aggregates are dispersed into smaller aggregates under shear. However, the G' of the P-H5-W sample (Fig. 4b) decreased with the shear rate (0.01 to 1/s). As mentioned earlier, the HNTs are well dispersed in the PP matrix for the P-H5-W sample; thus, its microstructure can be destroyed easily by a slight shear force.

Stress Relaxation Behavior and K-BKZ Model

The stress relaxation behavior of pure PP samples and 5 wt% HNT nanocomposites samples are studied at 1% strain considering the linear relaxation modulus; the results are shown in Fig. 5. Both the P and P-W samples exhibit a relaxation behavior corresponding to the relaxation of the matrix phase. The relaxation behavior of the P-H5 and P-H5-W samples can be divided into two parts, corresponding to the relaxation of the matrix and filler phase. As shown in Fig. 5, the P-H5-W sample exhibits a large shift in the spectra toward a longer relaxation time. It is retarded by the structure created by the HNTs dispersed in the PP matrix as well as by the strong interaction between the PP matrix and HNTs [30, 36].

To study the interaction between the PP matrix and HNTs, a discrete relaxation spectrum is obtained by fitting the stress relaxation experiments to a generalized Maxwell model [37] for the P-H5 and P-H5-W samples. Table 3 shows a set of relaxation parameters determined from linear relaxation measurements. To better understand the rheological response of PP/HNT nanocomposites prepared with and without water injection, especially the PP-HNTs interaction in P-H5 and P-H5-W samples, a truncated K-BKZ constitutive equation was used. This equation exhibits the following form:

[sigma](t) = [integral].sup.t.sub.-[infinity]] m(t-t')h([I.sub.1], [I.sub.2])[C.sup.-1.sub.t](t,t')dt (1)

where [sigma]{t) is a stress tensor, m(t - t') is the memory function that describes the time dependence of the material, and (t - t') is the elapsed time between the remembered past and the present. [C.sup.-1.sub.t] is the Finger tensor, and h is the damping function. Here [I.sub.1] and [I.sub.2] stand for the first and second invariants of the Finger tensor and depend on the specifics of the flow studied. The memory function is related to the linear relaxation modulus, G(t) and expressed as a sum of exponentials:

m(t-t') = [N.summation over (i)] [G.sub.i]/[[lambda].sub.i] exp (- t-t'/[[lambda].sub.i]) (2)

where the relaxation times, [[lambda].sub.i], and moduli, [G.sub.i] is determined from stress relaxation measurements by fitting the experimental data to the generalized Maxwell model (seen in Table 3).

In general, linear responses cannot be used to predict the values of material functions for the nonlinear viscoelastic range. To predict the material function, it is necessary to use a nonlinear constitutive equation, such as a damping function. The K-BKZ model is highly useful because using a proper damping function enables the correct description of various shear flows [38]. To evaluate the damping function, stress relaxation after a sudden step strain was performed in the nonlinear viscoelastic region. In this study, two types of damping functions were chosen for the K-BKZ constitutive equation and the optimal parameters for each damping function were estimated. Using these optimal parameters and discrete relaxation spectra, the transient shear viscosity ([[eta].sup.+]) can be predicted. The shear damping function, h([gamma]), can be obtained graphically by evaluating the amplitude of the vertical shift required to superpose the relaxation curve for each nonlinear strain onto the linear curve on a log-log plot. Precisely, it can be obtained as a function of strain, [gamma], by the vertical shift of the nonlinear relaxation modulus, G(t, [gamma]), onto the linear relaxation modulus, G(t):

h([gamma]) = G(t,[gamma])/G(t)(3)

In the limit of linear viscoelasticity, the linear relaxation modulus is independent of the shear strain. The damping function is unity when the shear strain is relatively small, that is, within the linear viscoelasticity range. As the shear strain increases, the damping function decreases, which can be regarded as the nonlinearity of the relaxation modulus.

Two representative forms exist for the shear damping function in the K-BKZ constitutive equation. One was proposed by Wagner [39] and is expressed as follows:

h(I) = exp (-n[square root of I-3]) (4)

The other is proposed by Papanastasiou et al.:

h(I) = 1/1 + a (I-3) (5)

where a is an adjustable parameter that depends on molecular characteristics and flow type. In this work, two types of damping functions were chosen and fit Eqs. 4 and 5 to the experimental data. The K-BKZ integral constitutive equation for any state of strain may be rewritten as below for the Wagner model (Eq. 6) and for the Papanastasiou-Scriven-Macosko (PSM) model (Eq. 7).

[[sigma].sub.12] = [[integral].sup.t.sub.-[infinity]][N.summation over (i=1)] [G.sub.i]/[[lambda].sub.i] exp (-t- t'/[[lambda].sub.i])[exp(-n[gamma])][gamma]dt' (6)

[[sigma].sub.12] = [[integral].sup.t.sub.-[infinity]] [N.summation over (i=1)] [G.sub.i]/[[lambda].sub.i] exp (-t- t'/[[lambda].sub.i])([gamma]/1 + a [[gamma].sup.2]) dt' (7)

The transient viscosity growth curves under shear flow were calculated by the Wagner model (Eq. 8) and the PSM model (Eq. 9) as follows:

[mathematical expression not reproducible] (8)

[mathematical expression not reproducible] (9)

Figure 6 shows the evolution of the relaxation modulus with respect to the elapsed time after a sudden step shear strain where the amplitude of the strain was varied from linear to nonlinear viscoelasticity ranges. The plot of G(t, [gamma])/h([gamma]) versus time (t) is displayed for the two samples at strains of 1%, 10%, 20%, 50%, 100%, 200%, and 400%. It is evident from the data shown in Fig. 6a and b that the time-strain separability is valid for the range of applied strains [40]. The step shear-strain alters the nonlinear viscoelastic response of the P-H5-W sample more strongly than that of the P-H5 sample as the microstructure of HNTs in the P-H5 sample is more strain dependent. This shows that the interaction between the HNTs and PP matrix in the P-H5-W sample is strong such that its microstructure cannot be deformed easily.

The mathematical forms of the Wagner and PSM models under shear flow were chosen to fit the experimental data obtained from the step strain-shear flow. The first and second strain invariants are identical with each other under shear. The damping function, h([I.sub.1], [I.sub.2]), in shear flow is the same as the shear damping function, h([gamma]). In Fig. 7, the strain dependence of the damping function is fitted adequately by the relationship that has been used frequently to predict the damping behavior in complex polymer systems [41, 42]. Although the PH5-W sample exhibits similar damping behavior, that is, delayed damping behavior, the P-H5 sample demonstrates a strong strain softening behavior with low onset strain amplitude. The strain-softening behavior is related to the strain-induced alignment of the HNTs in the shearing direction because the alignment is important in the damping of nanocomposites. It also shows that the more homogeneous is the HNT dispersion in the matrix, the more difficult it is to orientate under shear.

Damping function of strain obtained from step strain relaxation experiments are presented in Fig. 7. The Wagner and PSM damping functions are described for the two samples. The values of damping coefficients, n and a, are represented in Fig. 7a and b. It can be concluded that the HNTs in the P-H5 sample tend to depress the damping function, specifically rendering the nanocomposites more strain softened as compared to the P-H5-W sample. The decreased damping function also revealed more strain dependence or less elastic behavior for PP/HNT nanocomposites owing to the orientation of HNTs with the shearing direction. This indicates that the damping coefficient is related to the interaction between the PP matrix and HNTs. After water injection, the interaction between the matrix and HNTs increases, thus decreasing the damping coefficient, and vice versa.

The transient shear viscosity [[eta].sup.+]([gamma], t) was calculated by applying the memory function and two types of damping functions in the K-BKZ constitutive Eqs. 8 and 9. The experimental data represents the steady shear viscosity obtained for a stress growth shear experiment at a given shear rate. In general, a close agreement is observed between the experiment and predicted results for P-H5-W sample. In Figs. 8 and 9, the shear viscosity quickly reaches a steady plateau value in -5-10 s after the application of shear. The viscosity growth curves of the P-H5 sample are predicted by the Wagner and PSM models under transient shear flow, as shown in Figs. 8a and 9a. Because the viscosity decreases with increasing shear rate, the shear viscosity approaches successively lower steady-state asymptotes as the shear rate is increased. Figures 8b and 9b show the shear viscosity growth curves of the P-H5-W nanocomposite melts. The t[[eta].sup.+]([gamma], t) exhibits a slight stress overshoot at the highest applied shear rate of 5/s and 10/s. Figure 8b shows obvious overshoots that is attributed to the HNT orientation by shear flow. The [[eta].sup.+]([gamma], t) increase by the orientation of HNTs in the PP matrix enables the HNTs and polymer interactions to occur at the interphase between the HNTs and PP chains [43]. In Fig. 9a, a very small overshoot (barely visible in the figure) is depicted for the P-H5 sample, probably owing to the poor dispersion of the HNTs [44, 45]. The transient rheological properties reported in this section agree well with the SEM results. The PSM models can well predict the dispersion state of HNTs after water injection through the amplitude of the overshoot.

CONCLUSIONS

In this work, the rheological behavior of PP/HNT nanocomposites prepared with and without water injection is studied. The SEM results indicate that a better HNT dispersion in the PP matrix results in a higher value of storage modulus and complex viscosity at low frequencies. Although no plateau of G' at low frequencies is observed, the rheological behavior of composites with low content of HNTs is greatly improved. The results of stress relaxation experiments indicated that the sample prepared by water injection exhibited a longer relaxation time owing to the good dispersion of the HNTs. The transient shear results were predicted satisfactorily using a truncated form of the K-BKZ constitutive equation, and the overshoot predicted by the Wagner model could be used as an indicator of HNT dispersion. The predicted results indicated good agreement with the SAOS results. Further, the nanocomposites prepared by water injection exhibited a strong damping behavior owing to the strong interaction between the HNTs and PP matrix in shear flow. The dispersion state of the HNTs resulted in the change in parameters of the damping function in the constitutive equation, which will affect the transient viscosity.

Yu-Xiao Huang, Han-Xiong Huang

Lab for Micro Molding and Polymer Rheology, South China University of Technology, Guangzhou, 510640, People's Republic of China

Correspondence to: H.-X. Huang; e-mail: mmhuang@scut.edu.cn, hyxscut@163.com

Contract grant sponsor: Guangdong Provincial Natural Science Foundation; contract grant number: 2016A030308018. contract grant sponsor: National Natural Science Foundation of China; contract grant number: 21374033.

DOI 10.1002/pen.25156

Published online in Wiley Online Library (wileyonlinelibrary.com).

ACKNOWLEDGMENTS

Financial support provided by the National Natural Science Foundation of China (21374033) and Guangdong Provincial Natural Science Foundation (2016A030308018) is gratefully acknowledged.

REFERENCES

[1.] K. Prashantha, M.F. Lacrampe, and P. Krawczak, Express Polym. Lett., 5(4), 295 (2011).

[2.] Y. Zhang, A.D. Tang, H. Yang, and J. Ouyang, Appl. Clay Sci., 119, 8 (2016).

[3.] K. Szpilska, K. Czaja, and S. Kudla, Polimery, 60(06), 359 (2015).

[4.] P. Parthajit, K.K. Mrinal, M. Asish, and K.D. Chapai, J. Appl. Polym. Sci., 131, 39587 (2014).

[5.] B. Lecouvet, M. Sclavons, S. Bourbigot, J. Devaux, C. Bailly, and B. Lecouvet, Polymer, 52(19), 4284 (2011).

[6.] D.D.J. Rousseaux, S.I. Naima, A.C. Baudouin, J. Devaux, P. Godard, M.B. Jacqueline, and M. Sclavons, Polymer, 52(2), 443 (2011).

[7.] V. Molajavadi and H. Garmabi, J. Appl. Polym. Sci., 119(2), 736 (2011).

[8.] J. Tong, H.X. Huang, and M. Wu, Compos. Sci. Technol., 129, 183 (2016).

[9.] M. Wu, H.X. Huang, and J. Tong, Mater. Des., 108, 761 (2016).

[10.] E. Oliaei and B. Kaffashi, Polymer, 104, 104 (2016).

[11.] C. Chen and D.G. Baird, Polymer, 47, 7485 (2006).

[12.] C.E. Gonzalez, C.A. Orta, G.M. Colunga, F. Lionetto, and A. Maffezzoli, IEEE Trans. Nanotechnol., 15(5), 731 (2016).

[13.] M.J. Solomon, A.S. Almusallam, K.F. Seefeldt, A. Somwangthanaroj, and P. Varadan, Macromolecules, 34(6), 1864 (2001 ).

[14.] M.X. Liu, Z.X. Jia, D.M. Jia, and C.R. Zhou, Prog. Polym. Sci., 39, 1498 (2014).

[15.] O. Akhlaghi, O. Akbulut, and Y.Z. Menceloglu, Eur. Polym. J., 73(73), 17 (2015).

[16.] V.P. Singh, K.K. Vimal, G.S. Kapur, S. Sharma, and V. Choudhary, J. Polym. Res., 23, 43 (2016).

[17.] S. Ceccia, D. Ferri, D. Tabuani, and P.L. Maffettone, Rheol. Acta, 47(4), 425 (2008).

[18.] Y.S. Song, Rheol. Acta, 46(2), 231 (2006).

[19.] J.H. Zhu, S.Y. Wei, and Ryu, J. Mater. Chem., 20(23), 4937 (2010).

[20.] R. Wagener and T.J.G. Reisinger, Polymer, 44(24), 7513 (2003).

[21.] B. Momani, M. Sen, M. Endoh, X.L. Wang, K. Tadanori, and H. H. Winter, Polymer, 93, 204 (2016).

[22.] T.G. Gopakumar, J.A. Lee, M. Kontopoulou, and J.S. Parent, Polymer, 43(20), 5483 (2002).

[23.] L. Incarnato, P. Scarfato, L. Scatteia, and D. Acierno, Polymer, 45(10), 3487 (2004).

[24.] J. Zhao, A.B. Morgan, and J.D. Harris, Polymer, 46(20), 8641 (2004).

[25.] B. Bernstein, E.A. Kearsley, and L.J. Zapas, Trans. Soc. Rheol., 7, 391 (1963).

[26.] A. Kaye, "Non-Newtonian Flow in Incompressible Fluids," Note No. 134, College of Aeronautics, Cranfeld, UK (1962).

[27.] C. Kagarise, K. Miyazono, M. Mahboob, K.W. Koelling, and S. E. Bechtel, J. Rheol., 55(4), 781 (2011).

[28.] S.H. Lee, S.Y. Kim, and J.R. Youn, Polym. Compos., 30, 10 (2009).

[29.] V. Pasanovic-Zujo, R.K. Gupta, and S.N. Bhattacharya, Int. Polym. Process., 19, 4 (2004).

[30.] B. Wang and H.X. Huang, Polym. Degrad. Stab., 98(9), 1601 (2013).

[31.] W.S. Chow, Z.A.M. Ishak, and K.J.Z. Kocsis, Macromol. Mater. Eng., 290(2), 122 (2005).

[32.] S.H. Karamipour, H. Ebadi-Dehaghani, M. Hajari, M. Akbarzadeh, D. Ashouri, and S. Mousavian, J. Macromol. Sci. Part B: Phys., 53, 1574 (2014).

[33.] Y.K. Wang, D.F. Wu, W.Y. Xie, K. Wang, Q.R. Xia, and H. Yang, Polymer, 140, 167 (2018).

[34.] D.F. Wu, J.H. Wang, M. Zhang, and W.D. Zhou, Ind. Eng. Chem. Res., 51, 6705 (2012).

[35.] L.S. Nielsen, R.F. Landel, and L.L. Faulkner, Mechanical Properties of Polymers and Composites, Marcel Dekker Inc, New York (1994).

[36.] I. Alig, P. Potschke, D. Lellinger, T. Skipa, S. Pegel, G. R. Kasaliwal, and T. Villmow, Polymer, 53, 4 (2012).

[37.] C. Triebel, N. Katsikis, H. Stara, and H. Munstedt, J. Rheol, 54 (2), 407 (2010).

[38.] S.H. Lee and J.R. Youn, Adv. Compos. Mater., 17(3), 191 (2012).

[39.] M.H. Wagner, Rheol. Acta, 18, 33 (1979).

[40.] M. Mudasir, H. Azizi, and R. Ahmed, Iran. Polym. J., 26, 797 (2017).

[41.] K. Osaki, Rheol. Acta, 32, 429 (1993).

[42.] M.T. Islam, J.S. Reyes, and L.A. Archer, J. Rheol., 45, 61 (2001).

[43.] M. Rajabian, M. Samadfam, G. Naderi, and M.H. Beheshty, Rheol. Acta, 51, 1007 (2012).

[44.] D. Bagheriasl, P.J. Carreau, B. Riedl, C. Dubois, and W. Y. Hamad, Cellulose, 23, 1885 (2016).

[45.] D. Bagherias, P.J. Carreau, C. Dubois, and B. Riedl, Compos. Sci. Technol., 117, 357 (2015).

Caption: FIG. 1. SEM micrographs of fractured surfaces for (a) P-H5, (b) P-H5-W, (c) P-H15, and (d) P-H15-W samples. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 2. Comparison of storage modulus (G'), complex viscosity ([eta]*) and vGP plots for (a, b, c) P-Hm and (a', b', c') P-Hm-W samples. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 3. Cole-Cole plot for all prepared samples. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 4. Effect of preshearing on the storage modulus (G') of (a) P-H5 and (b) P-H5-W samples after preshearing under different shear rates. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 5. Comparison of stress relaxation after step strain of 1% showing slow chain relaxation for pure PP sample and 5 wt% HNTs nanocomposites. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 6. Evolution of the linear and nonlinear relaxation modulus of (a) P-H5, (b) P-H5-W samples with respect to the elapsed time after step shear strain. [Color figure can be viewed at wileyonlinelibrary.com!

Caption: FIG. 7. (a) Wagner and (b) PSM shear damping functions for P-H5 and P-H5-W samples obtained from step strain relaxation experiments. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 8. Comparison of the Wagner predictions on shear viscosity growth curves in transient shear flow for (a) P-H5 and (b) P-H5-W samples. [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIG. 9. Comparison of the PSM model predictions on shear viscosity growth curves in transient shear flow for (a) P-H5 and (b) P-H5-W samples. [Color figure can be viewed at wileyonlinelibrary.com]

TABLE 1. PP/HNTs nanocomposite and PP samples prepared in this work. Sample P (wt%) H (wt%) W (mL/min) P 100 0 0 P-W 100 0 20 P-H5 95 5 0 P-H10 90 10 0 P-H15 85 15 0 P-H5-W 95 5 20 P-H10-W 90 10 20 P-H15-W 85 15 20 TABLE 2. [eta]* of the PP/HNTs nanocomposites samples at 0.1 rad/s. Sample P P-W P-H5 P-H10 [eta]* (Pa x s) 389 430 1,070 2,810 Sample P-H15 P-H5-W P-H10-W P-H15-W [eta]* (Pa x s) 3,040 6,130 2,990 3,380 TABLE 3. Discrete relaxation spectrum calculated from generalized Maxwell model for P-H5 and P-H5-W samples. P-H5 [[lambda].sub.i] (s) [G.sub.i] (Pa) 7.234 x [10.sup.-3] 1.423 x [10.sup.4] 1.014 x [10.sup.-1] 1.570 x [10.sup.3] 5.896 x [10.sup.-1] 5.802 x [10.sup.1] 1.098 x [10.sup.0] 1.514 x [10.sup.1] P-H5-W [[lambda].sub.i] (s) [G.sub.i] (Pa) 1.166 x [10.sup.-2] 1.292 x [10.sup.4] 3.401 x [10.sup.-2] 1.542 x [10.sup.4] 9.918 x [10.sup.-2] 2.924 x [10.sup.3] 2.892 x [10.sup.-1] 6.489 x [10.sup.3] 2.459 x [10.sup.0] 1.055 x [10.sup.3] 2.091 x [10.sup.1] 2.589 x [10.sup.1]

Printer friendly Cite/link Email Feedback | |

Title Annotation: | polypropylene halloysite nanotubes |
---|---|

Author: | Huang, Yu-Xiao; Huang, Han-Xiong |

Publication: | Polymer Engineering and Science |

Article Type: | Report |

Geographic Code: | 9CHIN |

Date: | Aug 1, 2019 |

Words: | 5267 |

Previous Article: | Effect of Interfacial Chemistry on Crystallization of Polypropylene/Multiwall Carbon Nanotube Nanocomposites. |

Next Article: | Thermoplastic Polyurethane/Polytetrafluoroethylene/Bridged DOPO Derivative Composites: Flammability, Thermal Stability, and Mechanical Properties. |

Topics: |