Effect of polyhedral oligomeric silsesquioxane (POSS) reinforced polypropylene (PP) nanocomposite on the microstructure and isothermal crystallization kinetics of polyoxymethylene (POM)

Polymer 53 (2012) 5347e5357

Contents lists available at SciVerse ScienceDirect

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

Effect of polyhedral oligomeric silsesquioxane (POSS) reinforced polypropylene (PP) nanocomposite on the microstructure and isothermal crystallization kinetics of polyoxymethylene (POM) Ali Durmus a, *, Alper Kasgoz a, Nevra Ercan a, Dincer Akın a, Selen S¸anlı b a b

Istanbul University, Engineering Faculty, Department of Chemical Engineering, Avcılar, 34320 Istanbul, Turkey EPSAN Plastics, R&D Department, Ali Osman Sönmez Cad. No:16, Bursa, Turkey

a r t i c l e i n f o

a b s t r a c t

Article history: Received 1 August 2012 Received in revised form 9 September 2012 Accepted 11 September 2012 Available online 23 September 2012

In this study, effects of small amount of methyl-polyhedral oligomeric silsesquioxanes (methyl-POSS) on the microstructure and isothermal melt-crystallization behavior of polyoxymethylene (POM) were investigated, in detail. Introducing of methyl-POSS particles in POM phase was achieved via melt blending of methyl-POSS reinforced isotactic polypropylene (i-PP) nanocomposite as POSS carrier material with POM in a twin screw co-rotating extruder. Microstructural features of the POM/PPePOSS compounds were investigated with scanning electron microscopy (SEM) analysis. SEM analysis showed that the POM/PPePOSS compounds exhibited immiscible blend morphology. The POM, continuous matrix, phase includes a significant number of POSS particles due to interfacial interactions between the SieO bonds of POSS and CeO bonds of POM, and resulted POSS migration from PP to POM phase during the melt processing. The kinetic parameters for the isothermal melt-crystallization process of the samples were determined with the Avrami and LauritzeneHoffman models. The crystallization activation energies were determined by the Arrhenius method. It was found that the PPePOSS nanocomposite significantly accelerated the isothermal crystallization rate of POM. Based on the results, it has been highlighted that POM compounds including a small amount of PPePOSS nanocomposite as POSS carrier material can be successfully used in the production of injection molded POM parts because the POM/PPePOSS compounds yield much faster molding cycle thus production rate than the POM. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Polyoxymethylene (POM) Polyhedral oligomeric silsesquioxane (POSS) Crystallization kinetics

1. Introduction Polyoxymethylene (POM) or polyacetal is one the most widely used engineering thermoplastics in many application areas such as automotive industry and parts in mechanical and electromechanical equipments etc. due to its superior physical properties. POM exhibits higher crystallinity than other semi-crystalline polymers, high tensile and flexural modulus, excellent dimensional stability and wear and friction properties [1]. On the other hand, POM is weak for thermal degradation, weathering and environmental conditions since the methyl oxide bonds (eCH2eOe) in the structure are easy to break under heat and oxygen [2]. It is well known that physical properties of a semi-crystalline polymer are governed by the supramolecular structure, which in turn is controlled by the crystallization. Crystallization behavior

* Corresponding author. Tel.: þ90 212 591 24 80; fax: þ90 212 473 70 38. E-mail address: [email protected] (A. Durmus). 0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.polymer.2012.09.026

of POM compounds and composites depends on the compositional and microstructural parameters such as physical properties of the second phase and/or fillers like surface character, loading amount, geometry, dispersion etc. and processing conditions rather than the structural features of matrix because POM homo or copolymer exhibits a quite simple chain structure and relatively higher crystallization rate compared to other industrial thermoplastics. POM compounds are prepared with conventional melt processing methods and end-products are generally used as injection molded parts in many applications. Therefore, comprehensively understanding of melt-crystallization of POM compounds provides important knowledge for managing of injection molding operations in plastic industry and the physical properties of products. Many studies have been reported on the nucleation effects of inorganic fillers such as glass fiber [3,4], various types of minerals, silica, alumina, titanium dioxide (TiO2) and nanoparticles [5e12] and organic additives [13e15] on the crystallization kinetics of semi-crystalline thermoplastics. On the other hand, crystallization

5348

A. Durmus et al. / Polymer 53 (2012) 5347e5357

kinetics of semi-crystalline polymer blends have also been studied [16e19]. Hu and Ye investigated isothermal and non-isothermal crystallization behavior and morphology of POM blended with small amount of polyamide (PA) (0.2e0.6 wt%) [20]. They demonstrated that the addition of PA reduced the spherulite size and improved the crystallization growth rate and the degree of crystallinity of POM due to the nucleation effect of PA as the highmolecular nucleus. Masirek and Piorkowska found that the addition of small amount of submicron poly(tetrafluoro ethylene) (PTFE) particles (0.005e0.5 wt%) enhanced the nucleation of isotactic polypropylene (i-PP) and POM crystallization [21]. They also declared that the PTFE particles improved the elastic modulus of i-PP and POM. Ding et al. studied the effect of polyvinylidene fluoride (PVDF) on the crystallization behavior of POM [22]. They pointed out that addition of PVDF greatly decreased the spherulitic size of POM, but the structure of hexagonal POM crystal was not changed. Furthermore, they also reported that the dispersed PVDF acted as heterogeneous nuclei and accelerated the rate of nucleation, but hindered the crystal growth rate of POM. One of the interesting studies on the crystallization behavior of POM in a multi-component system was the study by Goossens and Groeninckx [23]. They investigated the influence of the curing process of diglycidyl ether of bisphenol A (DGEBA) based epoxy resin and the resulting reaction-induced phase separation (RIPS) on the crystallization and melting behavior of polyoxymethylene (POM) in the POM/epoxy blends. They found that the isothermally crystallized POM was dramatically influenced by the conversion degree of the epoxy resin and more perfect crystals were formed if the epoxy resin could polymerize during the crystallization process. Few papers have also been published about the effects of different fillers and/or organic and inorganic nucleating agents on the crystallization of POM [24e27]. Xu and He investigated the crystallization kinetics of POM including attapulgite and diatomite as low-cost inorganic nucleating agents [26,27]. Polyhedral oligomeric silsesquioxane (POSS) molecule is a rigid, three dimensional, cage-like silicon-oxygen nanostructured skeleton with a general formula of R(SiO1.5)n where R is hydrogen or an organic group (aliphatic or aromatic or any of their derivatives) and n ¼ 8, 10 or 12. Specific geometry and size of POSS molecules (1.5e 3 nm), its chemical versatility for grafting of various types of functional groups and commercial availability of intermediates or many different varieties has attracted great technical attention for application of POSS reinforced and/or modified polymers. Dispersion of POSS nanoparticles in a polymer phase provides many structural and physical advantageous such as higher mechanical properties, thermal and oxidative stability, easy crystallization and higher crystallization rate, molecular sieving and selectivity in gas separation and pervaporation membranes etc [28e30]. Effects of various types of POSS molecules on the microstructure, crystallization behavior and physical properties of semi-crystalline thermoplastics have been studied in recent years [31e37]. To the best of our knowledge, only two papers published on the microstructure and thermo-mechanical properties of POSS-reinforced POM [38,39]. Illescas et al. investigated the morphology and thermomechanical behavior of POM composites filled with monosilanolisobutyl polyhedral oligomeric silsesquioxane (msib-POSS) prepared with direct melt blending at loadings between 0 and 10 wt% of msib-POSS [39]. They reported that formation of hydrogen bonding interactions between the POM and SieOH groups of msib-POSS increased their mutual compatibility and lead to nanometer-size dispersion of some msib-POSS molecules. They also found that such interactions did not prevent aggregation of POSS during melt blending, but lead to micron-scale msib-POSS domains.

In this study, effects of small amount of methyl-polyhedral oligomeric silsesquioxanes (methyl-POSS) on the microstructure and isothermal melt-crystallization behavior of polyoxymethylene (POM) were investigated, in detail. Introducing of methyl-POSS into the POM phase was achieved via a different compounding route from the conventional direct blending method, for the first time. Methyl-POSS reinforced polypropylene (i-PP) nanocomposite was used as POSS carrier material or masterbatch. Hence, introducing of POSS molecules into the POM phase was carried out via blending of two granulated polymers (POM and PP/POSS nanocomposite) by a simple melt processing method. 2. Experimental 2.1. Materials Polyoxymethylene (POM) used in this study is a commercial polyacetal copolymer, KepitalÒ F20, with the molecular formula of ½CH2  On  ½CH2 CH2  Om , the density of 1.41 g cm3 (ASTM D792) and MFI of 9.0 g 10 min1 (under the condition of 10 kg, 220  C). PPePOSS nanocomposite is also a commercial grade material purchased from Aldrich, product number of 565628. It is an isotactic polypropylene (i-PP) based nanocomposite reinforced with 10 wt% of methyl-polyhedral oligomeric silsesquioxanes (methyl-POSS). 2.2. Sample preparation Samples were prepared by melt processing in a lab-scale twin screw extruder (Rondol Micro Lab., UK, D:10 mm, L/D: 20) with a screw speed of 50 rpm. Screws of the extruder were configured as including 3D of 4  60 followed by 2D of 4  90 kneading segments. A temperature profile of 110e160e180e180e180  C was applied throughout the barrel from the feeding zone to die. A rod die was used for the preparing of granulated sample for the DSC and SEM studies. Before the melt processing, all materials were dried in a vacuum oven overnight at 70  C. The POM, i-PP and PPe POSS nanocomposite were also processed at the same conditions. Sample compositions are listed in Table 1. Sample notation of PP-10 and PP-20 indicate the POM/PPePOSS compounds including 10 and 20 wt% of PPePOSS nanocomposite which correspond to nanoPOSS amount of 1 and 2 wt% in the composition, respectively. 2.3. Microstructure and morphology studies by SEM Morphological and microstructural features of the PPePOSS nanocomposite and POM/PPePOSS samples were investigated by a field emission scanning electron microscope (FE-SEM, FEI-Quanta FEG 450). Cryo-fractured cross-sections of the extrudate samples were directly imaged in the electron microscope after a proper sample preparation of sputter-coated with gold. The methyl-POSS particles in the PPePOSS nanocomposite were also characterized by the energy dispersive X-ray spectroscopy (EDS) analysis on the sample images.

Table 1 Sample compositions. Samples

wt%

POM PP-10 PP-20 PPePOSS

100 90 80

POM

PPePOSS

POSS

10 20 100

1.0 2.0 10.0

A. Durmus et al. / Polymer 53 (2012) 5347e5357 Table 2 Isothermal crystallization temperatures employed in the DSC studies. Samples

Tcry ( C)

POM PP-10 PP-20 PPePOSS i-PP

149 152 152 123 121

150 153 153 125 123

151 154 154 127 125

152 155 155 129 127

156 156

2.4. Differential scanning calorimetry (DSC) study Melting and crystallization runs of the samples were carried out in a heat flux type DSC, SII Nanotechnology ExStar 6200. Temperature and heat flow calibration of the instrument were achieved with high purity indium (In), tin (Sn) and zinc (Zn) metals. In non-isothermal crystallization runs, samples weighing about 9e10 mg in an aluminum crucible were heated from 0  C to 220  C with the heating rate of 10  C min1 and kept at this temperature

5349

for 2 min to remove the thermal history then cooled from 220 to 0  C with the cooling rate of 10  C min1 by an electrical cooling device, Thermo Scientific EK90C/SII intracooler. After completion of the melt-crystallization process, samples were kept at 0  C for 2 min. Then the samples were heated again from the 0  Ce220  C with the heating rate of 10  C min1. In isothermal crystallization runs, samples weighing about 9e 10 mg in an aluminum crucible were heated from 0  C to 220  C with the heating rate of 10  C min1 and kept at this temperature for 2 min to remove the thermal history then cooled from 220 to the isothermal crystallization temperature (Tcry) listed in Table 2, with the cooling rate of 80  C min1 by the liquid nitrogen cooling device of the instrument. Samples were isothermally kept at these temperatures for 40 min. Subsequently, isothermally crystallized samples at different temperatures were quenched to 0  C then heated up to 220  C with the heating rate of 10  C min1. Degree of crystallinity (Xc) was determined from the enthalpy values of melting obtained from the second heating run using the following equation:

Fig. 1. SEM images of the cryofractured cross-sections of PPePOSS extrudates.

5350

Xc ð%Þ ¼

A. Durmus et al. / Polymer 53 (2012) 5347e5357

DH m  100 ð1  aÞDHom

(1)

where DHm is the enthalpy of melting in the second heating scan of o is the enthalpy value of melting of a 100% the samples (J g1), DHm crystalline form of matrix polymer and a is the weight fraction of o were taken nano-POSS or PPePOSS nanocomposite. Values of DHm as 326 J g1 for POM [40] and 209 J g1 for i-PP [41]. All runs were carried out under nitrogen (N2) atmosphere at a flow rate of 50 ml min1 to prevent thermal degradation of the samples.

3. Results and discussion 3.1. Microstructure and morphology of samples SEM images of cross-section of PPePOSS nanocomposite are given in Fig. 1. Large cubic or rectangular POSS particles embedded into the PP phase are clearly seen in Fig. 1(a). It was observed that the POSS particles exhibited an average particle size of 0.2e2.0 m although the producer declared the material as nano-POSS reinforced polypropylene. Many submicron and uniformly dispersed POSS particles were also seen in the POSS reinforced PP material as given in Fig. 1(b). EDS analysis on a particle seen in Fig. 1(a) proved that the chemical composition of the rectangular particle were consisted of Si, O and C atoms which corresponds to silsesquioxane structure. SEM micrographs of the POM/PPePOSS compounds are given in Fig. 2. Fig. 2(a, b) and Fig. 2(c, d) show the SEM images of PP-10 and PP-20 samples, respectively at different magnifications. It is seen that the compound samples exhibit characteristic immiscible (matrix-droplet) blend morphology. The average size of PP droplets ranged between 2 and 5 m. It is seen that the number of

droplets in a unit volume of compound increases and the droplet size slightly decreases with doubling of the PPePOSS amount in the composition. It can be observed in Fig. 2(b) and (d) that the many POSS particles present in the POM structure (marked with blue circulars), the continuous matrix phase, possibly due to the migration of the particles from the PP to POM phase during melt processing. But, a significant number of POSS particles embedded in the PP droplets still appear in the PP-20 sample as marked with the red circulars in Fig. 2(d). It is also noticeable that many holes and protuberances appear in the POM phase. These characteristic morphology and structure formation of POM were previously reported [42,43]. Su et al. concluded in their study on the investigation of phase morphology of POM/HDPE blends that such morphology of POM can be used as obvious character for POM to distinguish from other component in a blend [43]. These tear-like holes in the POM phase are more obvious in the PP-20 sample given in Fig. 2(d) compared to the sample PP-10 given in Fig. 2(b). This difference in sample morphology can be qualitatively attributed to faster crystallization of POM in the PP-20 sample than that in the PP-10. 3.2. Melting and non-isothermal crystallization behavior of the samples Enthalpy of melting in the first heating scan and nonisothermal crystallization exotherms of the samples recorded at the heating and cooling rate of 10  C min1 are given in Fig. 3(a) and (b), respectively. As seen in Fig. 3(a), melting behavior and range of POM and i-PP are quite similar. The melting temperatures (Tm) of POM and i-PP used in the study are 171.0 and 166.0  C, respectively. The PP-10 and PP-20 samples showed single melting endotherm which is very similar to the melting peak of POM although these samples include relatively high

Fig. 2. SEM images of the cryofractured cross-sections of PP-10 (a, b) and PP-20 (c, d).

A. Durmus et al. / Polymer 53 (2012) 5347e5357

5351

Table 3 Characteristic crystallization peak temperatures, enthalpy and degree of crystallinity values of the samples crystallized with the cooling rate of 10  C min1. Samples

Tc,o ( C)

Tc,p ( C)

DHm (J g1)a

Xc (%)b

POM PP-10 PP-20 PPePOSS i-PP

146.8 150.0 151.0 120.0 128.0

143.3 147.7 148.2 116.4 112.6

156.5 151.0 145.0 81.3 103.0

48.0 51.5 55.6 43.2 49.3

a Enthalpy of second melting endotherm recorded at the heating rate of 10  C min1. b Degree of crystallinity calculated with the Eq. 1 by using the enthalpy values of second melting.

Fig. 3. (a) First melting endotherms and (b) and (c) non-isothermal crystallization exotherms of the samples recorded at the heating/cooling rate of 10  C min1.

amount of i-PP. The Tm values of POM/PPePOSS compounds are 171.1 and 170.2  C. These results imply that the melting temperatures of samples in the first heating scan are not affected by the composition possibly due to the thermo-mechanical history of the samples during the melt-processing has more influence on the thermal behaviors of the samples. This result also indicates that POSS did not affect the crystalline structure of POM as previously reported [39]. On the other hand, it was observed that the nano-POSS significantly affected the melt crystallization behavior of i-PP and POM, as seen in Fig. 3(b). The crystallization onset (Tc,o) and peak (Tc,p) temperatures of the samples are listed in Table 3. It was found that the crystallization temperature of i-PP shifted to higher temperature by introducing of 10 wt% of nano-POSS. This is due to the wellknown nucleation effect of nano-POSS particles on the melt crystallization of polypropylene which has been previously reported in many papers [31e36]. More interesting result is the fact that the crystallization temperature of POM also shifted to higher

temperatures by introducing of PPePOSS nanocomposite into the composition even though i-PP or PPePOSS nanocomposite crystallize at much lower, approximately 30  C, temperatures than POM and significant amount of molten PP is present in the crystallization medium. This result could be attributed to the nucleation ability of nano-POSS particles for the melt crystallization of POM. Increasing of POSS amount into the POM/PPePOSS compound slightly increased the non-isothermal crystallization temperature of POM. It is also seen in Fig. 3(b) that the non-isothermal crystallization peaks of PP-10 and PP-20 compounds became narrower than that of POM. This result indicates that nano-POSS particles accelerate the crystallization rate of POM. The crystallization peaks of i-PP in the compounds can be clearly seen in Fig. 3(b). One might assume that shifting of crystallization onset and peak temperature of POM could be originated from the presence of i-PP in the crystallization medium by possibly changing the dynamical conditions in the crystallization medium. In order to check possible effects of neat iPP on the melt crystallization of POM, we also prepared two POM/ PP blends with the weight composition of 91/9 (B1) and 82/18 (B2) (w/w) which correspond to i-PP amount in the POM/PPePOSS compounds. In Fig. 3(c), non-isothermal crystallization exotherm of B2 sample is compared to those of POM and PP-20 samples. As obviously seen in Fig. 3(c), B2 and PP-20 samples exhibit the same crystallization behavior which shows that the crystallization improvement effects of PPePOSS nanocomposite is not originated from the presence of i-PP in the composition for the nonisothermal conditions. Similar relationship was observed for the isothermal crystallization at 152  C, but not given and discussed in this study. Enhancement effect of nano-POSS particles which were previously embedded into the PP phase for the nucleation and crystal growth of POM will be quantified in following parts. The enthalpy values of melting in the first heating run (DHm) and corresponding degree of crystallinity (Xc%) values of the samples determined with the Eq. (1) are listed in Table 3. One can notice that these enthalpy values reflect the effects of thermo-mechanical history during melt processing on the melting behavior and formation of crystalline structure of samples. It was found that introducing of PPePOSS nanocomposite increased the Xc values of POM while nano-POSS decreased the Xc of i-PP. This increase probably refers to different crystallization mechanism. 3.3. Kinetics of isothermal crystallization Fig. 4(aee) shows the characteristic isothermal crystallization exotherms of samples. It is seen that the crystallization exotherm becomes flatter and time for the completion of crystallization process increases with the increasing of crystallization temperature for all the samples, as expected. Crystallization exotherms of POM and POM/PPePOSS compounds including different amount of PPe POSS nanocomposite at the crystallization temperature of 152  C are given in Fig. 5(a) to more precisely compare the effect of

5352

A. Durmus et al. / Polymer 53 (2012) 5347e5357

Fig. 4. Isothermal crystallization exotherms of the samples. (a) POM, (b) PP-10, (c) PP-20, (d) PPePOSS and (e) i-PP.

compound composition on the crystallization behavior of samples. This compare for the crystallization of i-PP and PPePOSS nanocomposite at the crystallization temperature of 125  C is also illustrated in Fig. 5(b). It is clearly seen in Fig. 5(a) that the completion of crystallization of POM takes about 15 min at this temperature. But, the POM/PPePOSS compounds crystallize approximately 4e5 min. The PPePOSS nanocomposite dramatically shortens the crystallization time of POM although the nanocomposite cannot crystallize at this temperature. This effect could be attributed to acceleration effect of nano-POSS particles on the crystallization of POM. The Avrami model [44] was used to analyze isothermal crystallization rates of the samples, as given in Eq. (2);

Xt ¼ 1  expðktn Þ

(2)

where Xt is the relative crystallinity, n is the Avrami constant which depends on the mechanism of nucleation and the crystal growth, t is the real time of crystallization and k is the crystallization rate constant involving both nucleation and growth rate parameters. The relative crystallinity can be defined as a function of time in the following form:

Zt  Xt ¼

0

 dHc dt dt

ZN

 dHc dt dt

(3)

0

where Hc is the crystallization enthalpy during the infinitesimal time interval dt, t is the time at the end of the crystallization. Fig. 6(aee) shows the relative crystallinity (Xt)etime (t) plots of the samples. All the Xt curves have the same characteristic sigmoidal shape with time at various crystallization temperatures. The first non-linear part of the S shaped curves is generally considered as the induction time and nucleation step of the crystallization process. Each curve showed a linear part considered as primary crystallization; subsequently a second non-linear part deviated off slightly and is considered to be secondary crystallization, which was caused by the spherulite impingement in the late stage of the crystal growth. As seen in Fig. 6, all characteristic sigmoidal curves shift to right with the increasing of crystallization temperature. One of the most important rate parameter, crystallization halftime (t0.5) which is defined as the time taken the relative

A. Durmus et al. / Polymer 53 (2012) 5347e5357

5353

higher than those of POM and in the range of 2.67e2.96 and 2.56e 2.70 for the PP-10 and PP-20, respectively. In Eq. (4), taking the Xt ¼ 0.5, the relation between the crystallization half-time and crystallization rate constant, k can be written as:

 t0:5 ¼

 ln2 1=n k

(5)

If the t0.5 is taken from the experimental data, theoretical values of k can be calculated by using the Eq. (5). Values of k calculated from the Eq. (5) are also listed in Table 4. As seen, theoretical and experimental values of k are agreed well. It was found that crystallization rate constants of the samples were varied severely depending on the sample compositions. By comparing the k values of i-PP and PPePOSS at a particular crystallization temperature, e.g. 125  C, it is clearly seen that the nano-POSS increased the crystallization rate of i-PP, significantly. By comparing the k values at the crystallization temperature of 152  C, it is also shown that the crystallization rate constant of POM based samples decrease in the order of PP-20 > PP-10 > POM. This indicates that the methyl-POSS particles effectively enhance the nucleation and isothermal crystallization of POM. This can be originated from the fact that interactions between POSS molecules and POM chains. Xu and He concluded that the SieO bonds in the silicate nucleating agents, attapulgite and diatomite, were well compatible with the CeO bonds in POM structure [26,27]. Therefore, it can be concluded that POM chains or segments were interacted with the surfaces of SieO based additives. 3.4. LauritzeneHoffmann analysis

Fig. 5. Comparing of the crystallization exotherms of (a) POM, PP-10 and PP-20 samples at 152  C and (b) i-PP and PPePOSS samples at 125  C.

crystallinity of the sample reaches the value of 50% can be directly obtained from these curves. The values of t0.5 directly indicates the rate of crystallization process and usually the reciprocal of crystallization half-time (s0.5 ¼ 1/t0.5, time1) is simply used to compare crystallization rates of different systems. If the t0.5 is short or s0.5 is high, it means crystallization is fast. The s0.5 values of the samples at the crystallization temperature employed are listed in Table 4. As expected, the s0.5 decreases with the increasing of crystallization temperature for all the samples. By comparing the s0.5 values of POM, PP-10 and PP-20 samples at the crystallization temperature of 152  C, it is seen that the crystallization rate of samples decrease in the order of PP-20 > PP-10 > POM. It was found that the addition of 2 wt% of methyl-POSS into the compound accelerated the crystallization rate of POM about more than 6 times. Taking double logarithmic form of the Eq. (2) as;

ln½  lnð1  Xt Þ ¼ lnk þ nlnt

(4)

and plotting ln[ln(1  Xt)] vs. lnt at a given crystallization temperature, straight line should be obtained to determine Avrami kinetic constants. Slope of the line is equal to n and the intercept is lnk. Avrami plots of the samples are shown in Fig. 7(aee). As seen in Fig. 7, Avrami model was successfully fit the crystallization data for the relative crystallinity value up to 95%. At higher values of Xt, the curves deviate from the linearity which is attributed to the secondary crystallization. Kinetic parameters are listed in Table 4. The n values of PPePOSS nanocomposite were found to be higher than those of i-PP. The n values of POM varied between 2.28 and 2.50. The n values of the compound samples were found to be

Growth rate of spherulites (G) can be calculated with the isothermal crystallization data according to the LauritzeneHoffman secondary nucleation theory [45]. LauritzeneHoffman defined three different crystallization regimes based on the relationships between the rate of secondary nucleation (i) and the lateral growth rate of spherulite (g). Regime I, II and III imply the dependence on the conditions of low, moderate and high undercooling, DT, respectively which is the difference between the equilibrium 0 Þ and the crystallization temperature T , melting temperature ðTm c 0  T Þ. In regime I, i << g whereas i >> g in regime III. On the ðTm c other hand, i is on the order of g in regime II. Temperature or undercooling dependence of the linear growth rate is defined as the following exponential model;

 G ¼ G0 exp

   Kg U* exp Tc DTf RðTc  TN Þ

(6)

where G0 is a pre-exponential factor; R is the universal gas constant; U* is the energy required for the transport of macromolecules in the melt which is universally taken as 6.28 kJ mol1, Tc is the crystallization temperature, TN is a hypothetical temperature where all the motions associated with the viscous flow stop which is commonly defined as Tg-30 K, DT is the undercooling, f is a corrective factor for the decrease of the enthalpy of fusion with 0 Þ. K the crystallization temperature, f ¼ 2Tc =ðTc  Tm g is the nucleation parameter related to the fold and lateral surface energies. LauritzeneHoffman equation can be linearized as the following form by performing the logarithmic transformation;

lnG þ

Kg U* ¼ lnG0  Tc DTf RðTc  TN Þ

(7)

Plot of lnG þ U*/R(Tc  TN) against the term of 1/TcDTf yields Kg and G0 from the slope and intercept of the line, respectively. For the

5354

A. Durmus et al. / Polymer 53 (2012) 5347e5357

Fig. 6. Relative crystallinity (Xt)etime (t) curves of the samples. (a) POM, (b) PP-10, (c) PP-20, (d) PPePOSS and (e) i-PP. Table 4 Crystallization rates and Avrami kinetic parameters of the samples. Samples

Tcry ( C)

s0.5 (min1)

n

k (minn) (103)a

k (minn) (103)b

POM

149 150 151 152 152 153 154 155 156 152 153 154 155 156 123 125 127 129 121 123 125 127

0.559 0.323 0.233 0.133 0.595 0.429 0.287 0.147 0.081 0.813 0.543 0.339 0.195 0.120 0.901 0.546 0.376 0.238 0.490 0.315 0.235 0.139

2.28 2.50 2.35 2.49 2.67 2.82 2.83 2.94 2.96 2.70 2.64 2.60 2.56 2.61 2.71 2.72 2.76 2.73 2.29 2.31 2.19 2.19

209.17 45.82 21.73 4.58 239.84 83.11 22.66 2.34 0.4 523.61 165.10 42.70 9.76 2.18 570.70 143.62 50.71 13.68 120.57 42.08 23.03 8.09

183.79 40.97 22.50 4.56 173.48 63.81 20.17 2.47 0.41 396.35 138.58 41.62 10.59 2.22 521.85 134.77 46.57 13.78 135.45 48.24 29.15 9.16

PP-10

PP-20

PPePOSS

i-PP

a b

Determined from Avrami model. Determined from Eq. (5).

0 Þ ¼ 183  C and T ¼ 50  C [46] calculations, literature values of ðTm g were taken for POM and experimentally determined and G (s0.5) values were used. LauritzeneHoffman plots of the POM and POM/ PPePOSS nanocomposite compounds are given in Fig. 8. Determined values of G0 and Kg are listed in Table 5. As seen in Fig. 8, all the samples yielded no change in slope. This linearity indicates that no regime transition occurs in the crystallization for the temperature range employed. But, it is clearly seen that the addition of PPePOSS nanocomposite decreases the value of Kg. LauritzeneHoffman theory consists of chain mobility and the secondary nucleation terms. Crystal growth process is described as a combination of two processes; the deposition of the first stem on the growing crystal face (secondary nucleation) and the attachment of subsequent stems in the chain of the crystal surface (surface spreading process). In regime III, spherulite growth rate is governed by the rate of secondary nucleation, whereas it is driven by both the rate of secondary nucleation and the rate of surface spreading in regime II. In regime I, the growth rate is the slowest since the crystallization prevails at high temperatures (or under the condition of low undercooling). It is well known that a foreign surface acts as a nucleating center in the system and reduced the nucleus size for crystal growth since the substance create an interface. Formation of such interface might be easier than the creation of

A. Durmus et al. / Polymer 53 (2012) 5347e5357

5355

Fig. 7. Avrami plots of the samples. (a) POM, (b) PP-10, (c) PP-20, (d) PPePOSS and (e) i-PP.

polymer crystal surfaces. The Kg values given in Table 5 confirmed that the PPePOSS nanocomposite reduced the energy need to create a new surface leading to faster crystallization. For a heterogeneous nucleation, Kg, the kinetic term connected with the energy need for the formation of nuclei in critical size, can be calculated from:

Kg ¼

0 nbo sse Tm DHf kB

(8)

Table 5 LauritzeneHoffman parameters and crystallization activation energy values for the isothermal crystallization of the samples. Samples

Fig. 8. LauritzeneHoffman plots for the samples.

Kg (K2) (105) se (J m2) DEA (kJ mol1)

POM

PP-10

PP-20

PPePOSS

i-PP

2.32 0.95 720

2.00 0.82 817

1.92 0.78 783

295

274

5356

A. Durmus et al. / Polymer 53 (2012) 5347e5357

where n is the variable constant that depends on the crystallization regime, bo is the thickness of the stem added on the substrate, s is the lateral surface free energy, se is the surface free energy of folding, DHf is the melting enthalpy per unit volume of crystal and kB is the Boltzman constant. The n is equal to 4 for regimes I (high temperatures) and III (low temperatures) and 2 in regime II (intermediate temperatures). By obtaining the Kg, surface free energy of folding can be determined. Surface free energy terms (se) could be calculated by using the determined Kg values and taking 0, k the crystal parameters of n (4), b (0.447 nm), Tm B 23 1 8 3 2 (1.38  10 J K ), DHf (3.55  10 J m ) and s ¼ 14.7 mJ m [47]. Lateral surface free energy is generally estimated by an empirical equation as follows [48]:

s ¼ 0:11DHf

pffiffiffiffiffiffiffiffiffiffi a0 b0

(9)

where a0 and b0 are the unit cell parameters of POM, respectively. The calculated values of se of the samples are listed in Table 5. The se values of the samples increase in the order of PP-20 < PP10 < POM. The lower the value of se, the less energy need for the chain folding onto the surface of nucleus. As seen in Table 5, PPe POSS nanocomposite reduced the se compared to the POM. According to secondary nucleation theory, the number of effective nuclei (N) can also be calculated by the following equation [49];

k ¼

4 pNG0 3

(10)

The N values were found to be 4.05  1010, 2.12  1012 and 4.63  1012 m3 for the POM, PP-10 and PP-20, respectively which proved that the introducing of methyl-POSS particles via melt blending of PPePOSS nanocomposite with POM increased noticeably the number of nuclei in the system. Doubling of the methylPOSS amount proportionally increases the N. 3.5. Crystallization activation energy The Avrami parameters can be used to calculate the activation energy of isothermal crystallization. The crystallization rate constant, k can be approximately described by the Arrhenius equation as follows;

k1=n ¼ k0 expð  DEA =RTc Þ

(11)

Table 6 Second melting temperature, enthalpy and degree of crystallinity values of the samples crystallized at a particular crystallization temperature. Samples

Tcry ( C)

Tm ( C)

DHm (J g1)a

Xc (%)b

POM PP-10 PP-20 PPePOSS i-PP

152 152 152 125 125

170.4 170.0 169.5 166.0 168.2

157.5 153.4 151.0 78.3 98.4

48.3 52.3 57.9 41.6 47.1

a Enthalpy of second melting endotherm recorded at the heating rate of 10  C min1. b Degree of crystallinity calculated with the Eq. 1 by using the enthalpy values of second melting.

where k0 is a temperature independent pre-exponential factor, DEA is the activation energy, R is the universal gas constant, and Tc is the crystallization temperature [50]. Linearized form of the Eq. (11) is given as:

1=nðlnkÞ ¼ lnko  DEA =RTc

(12)

DEA can be determined by the slope of the linear plot 1/nlnk against 1/Tc. These plots are illustrated in Fig. 9 for the samples. It is seen that the crystallization activation energy for POM is much less than that for i-PP because of the linear chain structure of POM. The DEA values of the samples listed in Table 5 indicate that nano-POSS induced the heterogeneous nucleation in both i-PP and POM by lowering the activation energy. The DEA value of PP-10 sample (817 kJ mol1) is slightly lower than that of PP-20 (783 kJ mol1). Xu and He reported the DEA values of 971.7 kJ mol1, 895.8 kJ mol1 and 977.4 kJ mol1 for the POM, POM nucleated with attapulgite and POM nucleated with diatomite samples, respectively [27]. It is known that the magnitude of activation energy (jDEAj) related to energy need for the motion of polymer chains during the transformation from the melt into the crystalline state. Higher magnitude of DEA means that this transition needs to release more energy suggesting a more difficult motion of polymer chains in the system [51]. Considering the calculated DEA values for the overall crystallization process, the higher magnitude of PP-10 and PP-20 samples than that of POM could be attributed to hindering effect of nano-POSS particles for the chain diffusion throughout to growing crystal face although they dramatically increase the number of effective nuclei in the system. The peak temperatures and enthalpy values of melting in the second heating scan and the corresponding Xc values of samples, isothermally crystallized at a particular crystallization temperature, are listed in Table 6. Similar to previously mentioned results of dynamically crystallized samples under non-isothermal conditions, it was found that the introducing of nano-POSS did not change the melting behavior significantly but improved the degree of crystallinity of POM. On the other hand, nano-POSS addition decreased the Xc of i-PP. This is probably due to presence of relatively high amount of nano-POSS (10 wt%) in the PPePOSS nanocomposite therefore formation of much smaller spherulites which melts slightly lower than neat i-PP and yields lower Xc. 4. Conclusions

Fig. 9. Plots of 1/nlnk versus 1/Tc for the samples.

This study shows that the introducing of PPePOSS nanocomposite dramatically enhances the crystallization rate of POM although the presence of significant amount of molten PP in the crystallization medium. It has been concluded that the rate acceleration effect of PPePOSS nanocomposite for the crystallization of POM is probably due to the tremendous increase of number

A. Durmus et al. / Polymer 53 (2012) 5347e5357

nucleus by the nano-POSS particles. On the other hand, the compatibility of SieO bond existing in the POSS molecule with the CeO bonds in the POM structure could provide a significant interaction between the methyl-POSS particles and POM chains which provides a very fast chain folding onto the POSS surfaces. Rate accelerating effect of PPePOSS nanocomposite for the crystallization of POM compounds yields much faster molding cycle thus production rate for the injection molded POM parts. It was also shown that PPePOSS nanocomposite can be successfully used as a nano-POSS carrier material or masterbatch and a versatile processing additive for the compounding works of POM in the conventional melt processing operations due to granulated form of the PPePOSS nanocomposite. Acknowledgement Authors acknowledged to EPSAN Plastics Co., Ltd. (Bursa, TR) for kindly supplying the POM used in this work and technical support for the study. We thank Dr. Faruk Öksüzömer and Mr. Vedat a from the Department of Chemical Engineering at Istanbul Sarıbog University (Turkey), for their efforts and helps in the SEM study. References [1] Li Y, Tao Z, Chen Z, Hui J, Li L, Zhang A. Polymer 2011;52(9):2059e69. [2] Hu Y, Ye L. Polym Plast Tech Eng 2006;45(7):839e44. [3] Liu T, Yu X, Yu F, Zhao X, Lu A, Wang J, et al. Polym Plast Tech Eng 2012;51(6): 597e604. [4] S¸anlı S, Durmus A, Ercan N. J Mater Sci 2012;47(7):3052e63. lu N, Ercan N, Durmus A, Kas¸göz A. J Macromol Sci Part B: Phys 2012; [5] Oburog 51(5):860e79. [6] Chae DW, Kim BC. J Mater Sci 2007;42(4):1238e44. [7] Jiasheng Q, Pingsheng H. J Mater Sci 2003;38(11):2299e304. [8] Bhimaraj P, Yang H, Siegel RW, Schadler LS. J Appl Polym Sci 2007;106(6): 4233e40. lu N, Ercan N, Durmus A, Kas¸göz A. J Appl Polym Sci 2012;123(1): [9] Oburog 77e91. [10] Durmus A, Ercan N, Soyubol G, Deligöz H, Kas¸göz A. Polym Compos 2010; 31(6):1056e66. [11] Ou B, Ou Y, Li D, Jing B, Gao Y, Zhou Z, et al. Polym Compos 2012;33(7):1054e63. [12] Ge C, Shi L, Yang H, Tang S. Polym Compos 2010;31(9):1504e14. [13] Durmus A, Yalçınyuva T. J Polym Res 2009;16(5):489e98. [14] S¸anlı S, Durmus A, Ercan N. J Appl Polym Sci 2012;125(S1):E268e81.

5357

[15] Nam JY, Okamoto M, Okamoto H, Nakano M, Usuki A, Matsuda M. Polymer 2006;47(4):1340e7. [16] Zhang C, Yi X-S, Asai S, Sumita M. J Mater Sci 2000;35(3):673e83. [17] Wang TC, Li HH, Wang F, Yan SK, Schultz JM. J Phys Chem B 2011;115(24): 7814e22. [18] Zhou H, Ying J, Xie X, Liu F, Li D. Polym Test 2010;29(7):915e23. [19] Kawaguchi K, Nakao H, Masuda E, Tajima Y. J Appl Polym Sci 2008;107(2): 1269e79. [20] Hu Y, Ye L. Polym Eng Sci 2005;45(8):1174e9. [21] Masirek R, Piorkowska E. Euro Polym J 2010;46(7):1436e45. [22] Ding Q, Dai W-L, Zhang P. Polym Eng Sci 2007;47(12):2034e40. [23] Goossens S, Groeninckx G. J Polym Sci Part B Polym Phys 2007;45(17): 2456e69. [24] Xu W, Ge M, He P. J Appl Polym Sci 2001;82(9):2281e9. [25] Kongkhlang T, Kousaka Y, Umemura T, Nakaya D. Polymer 2008;49(6): 1676e84. [26] Xu W-B, He P-S. Polym Eng Sci 2001;41(11):1903e12. [27] Xu W-B, He P-S. J Appl Polym Sci 2001;80(2):304e10. [28] Lin OH, Mohd Ishak ZA, Akil H Md. Adv Comp Lett 2009;18(6):183e90. [29] Iyer P, Iyer G, Coleman M. J Memb Sci 2010;358(1e2):26e32. [30] Le NL, Wang Y, Chung T-S. J Memb Sci 2011;379(1e2):174e83. [31] Guan W, Qui ZB. Ind Eng Chem Res 2012;51(7):3203e8. [32] Zhou ZY, Cui LM, Zhang Y, Zhang YX, Yin NW. J Polym Sci Part B: Polym Phys 2008;46(17):1762e72. [33] Joshi M, Butola BS. J Appl Polym Sci 2007;105(2):978e85. [34] Pracella M, Chionna D, Fina A, Tabuani D, Frache A, Camino G. Macromol Symp 2006;234(1):59e67. [35] Joshi M, Butola BS. Polymer 2004;45(14):4953e68. [36] Lin OH, Mohd Ishak ZA, Akil H Md. Mater Des. 2009;30(3):748e51. [37] Gao JG, Hua DS, Du YG, Li X. Polym Plast Tech Eng 2011;50(14):1429e37. [38] Illescas S, Arostegui A, Schiraldi DA, Sánchez-Soto M, Velasco JI. J Nanosci Nanotech 2010;10(2):1349e60. [39] Illescas S, Sánchez-Soto M, Milliman H, Schiraldi DA, Arostegui A. High Perform Polym 2011;23(6):457e67. [40] Kumar G, Neelakantan NR, Subramanian N. J Mater Sci 1995;30(6):1480e6. [41] Marinelli AL, Bretas RE. J Appl Polym Sci 2003;87(6):916e30. [42] Huang J-M, Cheng H-J, Wu J-S, Chang F-C. J Appl Polym Sci 2003;89(6): 1471e7. [43] Su R, Su J, Wang K, Yang C, Zhang Q, Fu Q. Euro Polym J 2009;45(3):747e56. [44] Avrami MJ. Chem Phys 1939;7(12):1103e12. [45] Hoffman DJ, Davis GT, Lauritzen JI. In: Hannary NB, editor. Treatise on solid state chemistry. New York: Crystalline and Non-crystalline Solids Plenum; 1976. [46] Pelzbauer Z, Galeski A. J Polym Sci Part C 1972;38(1):23e32. [47] Hoffman JD. Polymer 1983;24(1):3e26. [48] Lauritzen JI, Hoffman JD. J Appl Phys 1973;44(10):4340e52. [49] Wunderlich B. Macromolecular physics, vol. 1. New York: Academic Press; 1976. p. 388. [50] Cebe P, Hong SD. Polymer 1986;27(8):1183e92. [51] Li J, Fang Z, Zhu Y, Tong L, Gu A, Liu F. J Appl Polym Sci 2007;105(6):3531e42.