Rheol Acta (2012) 51:143–150 DOI 10.1007/s00397-011-0599-1
ORIGINAL CONTRIBUTION
Design of new HDPE/silica nanocomposite and its enhanced melt strength Hyung Tag Lim · Kyung Hyun Ahn · Seung Jong Lee · Joung Sook Hong
Received: 29 March 2011 / Revised: 28 June 2011 / Accepted: 22 September 2011 / Published online: 21 October 2011 © Springer-Verlag 2011
Abstract Linear polymers are restricted to use in processes that involve severe extensional deformation, such as fiber spinning, film blowing, and thermoforming. To extend their applicability, the extensional properties of polymer melts should be enhanced such that strain hardening, which is defined as an increase in extensional viscosity under a large strain that deviates from the linear viscoelastic curve, is pronounced. In this study, a novel preparation method of linear polymer/inorganic nanocomposites was proposed with a main focus on enhanced melt strength. The design of molecular structure consists of three components— linear polymer, compatibilizer, and surface-modified particles. High-density polyethylene was used as a linear polymer while polyethylene grafted with maleic anhydride was used as a compatibilizer. Silica particles were synthesized and modified on their surfaces by 3-aminopropyltriethoxysilane. The strain hardening behavior of the surface-modified silica composites was pronounced. However, such a result was not observed for the composites of the same composition with puresilica. In addition, the dispersion of the modified silica was much better than that of pure-silica. Keywords Melt strength · Nanocomposites · Silica · Surface modification
H. T. Lim · K. H. Ahn (B) · S. J. Lee School of Chemical and Biological Engineering, Seoul National University, Seoul 151-742, South Korea e-mail:
[email protected] J. S. Hong Department of Chemical Engineering, Soongsil University, Seoul 156-743, South Korea
Introduction Polymer nanocomposites are considered as an important class of organic–inorganic hybrid materials. Compared to conventional composites based on macrosized fillers, the interface between filler particles and matrix constitutes a much larger surface area and therefore affects particle the properties of the composites to a much greater extent, even at a low filler loading (Kim et al. 2008; Nielsen and Landel 1994; Rong et al. 2001; Spencer et al. 2010). However, as particle size decreases down to a nanometer scale, the particles show a strong tendency to aggregate, making it difficult to uniformly disperse the nanoparticles in polymer matrices (Pavlidou and Papaspyrides 2008; Rozenberg and Tenne 2008). Extensional flow is one of the main flows in polymer processing. The extensional properties of a polymer, which are closely related to the melt strength, have been recognized as a key parameter in various polymer processes such as fiber spinning, film blowing, and thermoforming (Ishizuka and Koyama 1980; Munstedt 1980; Vlachopoulos and Sidiropoulos 2001; Cheng et al. 2010; Vlachopoulos and Strutt 2010). This is mainly due to the pronounced extensional flow field that is a characteristic of these processes, although much less is known about the rheological behavior of polymer melts under extensional flow compared to shear flow. In general, strain hardening is defined as the steep increase in extensional viscosity under a large strain that deviates from the linear viscoelastic curve, and it is known to be directly related with melt strength. The strain hardening behavior of polymer composites in which nano- or microparticles are dispersed has also been reported (Chan et al. 1978; Tanaka and White 1980; Takahashi et al. 1997; Le Meins et al. 2003;
144
Rheol Acta (2012) 51:143–150
Song and Youn 2004; Kakuda et al. 2006; D’Avino et al. 2008). In most cases, the particles tend to reduce the hardening effect of the matrix polymer even when it exhibits hardening behavior. One exception is the clay-dispersed polypropylene system. In a claydispersed maleic anhydride-grafted polypropylene system, Okamoto et al. (2001a, b) observed strong strain hardening behavior that is induced by the alignment of exfoliated clay layers. Park et al. (2006) investigated the effect of clay dispersion on the strain hardening behavior of three different composite systems, and the strain hardening behavior was observed only for the exfoliated systems. From an engineering perspective, it is critical to control the interaction between the nanoparticles and polymer resin, as well as to design nanocomposites with increased mechanical properties and processability. In this study, we propose a new method for the design of nanocomposites which show strain hardening behavior. HDPE was selected as a linear polymer, and silica was synthesized and surfacemodified. We first investigate the rheological behavior of the nanocomposites with a special focus on the effect of surface-modified silica on extensional properties, and then discuss the key factors that induce strain hardening behavior.
Experimental Materials High-density polyethylene (B220A, MFI: 3.5 g/min at 2.16 kg and 190◦ C) was supplied from Samsung Total Petrochemicals Co. Ltd., and maleic anhydride-grafted polyethylene (EM-101, MA graft ratio=1 wt.%) was provided by Honam Petrochemical Corp. Tetraethoxysilane (TEOS) and aminopropyltriethoxysilane (APTES) were obtained from Aldrich, ethyl alcohol (99.5% EtOH) and potassium bromide
Fig. 1 FE-SEM images of a pure-silica and b APTES-silica
a
(KBr) from Daejung Chemicals and Metals Co., Ltd., ammonia solution (28 wt.% ammonia) and nitric acid from Duksan Pure Chemical Co., Korea. They were all used as received. Synthesis of silica particles and surface modification Monodisperse silica spheres were synthesized according to the modified Stöber process (Wang et al. 2003; Wu et al. 2006). The method is to rapidly mix Solution A and Solution B, and stirring the solution for 3.5 h at room temperature. One is the mixture of 0.250 mol TEOS and 0.188 mol ammonia in deionized water and EtOH. To modify the surface of silica particles, 4.79 × 10−4 mol of APTES was injected into the stirred solution after 3.5 h of reaction. The reaction was allowed to continue for additional 19 h with stirring after the addition of APTES. The ethoxy groups of the organosilane were readily hydrolyzed and reacted with silanol groups on the silica surface. For the preparation of pure silica, no precursor was added. The particles were centrifuged, washed with EtOH and then with deionized water. Homogenizer was used to disperse the particles, and centrifugation was used for solid–liquid separation. The size of the modified particles could be controlled using “mother” pure silica particles of different size. Using this method, it was easy to obtain uniform particles with different size and surface properties (Wu et al. 2006). SEM micrographs of the prepared particles are shown in Fig. 1, which demonstrates that the modified particles are uniform and spherical. From QELS, the average diameter of pure-silica is 159 ± 21 nm and one of APTES-silica is 165 ± 25 nm. Preparation of HDPE/silica composites All samples were dried in the vacuum oven at 80◦ C for 24 h before mixing. Melt compounding of HDPE/silica
b
Rheol Acta (2012) 51:143–150
composites was carried out in a HAAKE batch mixer (Rheomix 600 & Rheocord 90, the bowl capacity is 120 cm3 ). The rotor speed was 50 rpm and mixing time was 15 min after silica feeding at 190◦ C. MAPE was introduced to facilitate the interaction between HDPE and modified silica. The ratio of HDPE and MAPE was fixed at 9:1 and the composition of silica in the composite was 2, 5, and 8 wt.%, respectively. Figure 2 shows the design concept of this study and shows the expected molecular structure during melt mixing. HDPE and MAPE are compatible to some extent and the maleic group of MAPE reacts with NH3 at the surface of the modified silica. In consequence, HDPE which is mingled with MAPE is able to form a network structure, and silica particle modified by APTES plays a role as a network junction. This network structure is expected to drastically enhance the resistance to extension, thereby enhancing the melt strength and showing the strain hardening behavior.
Characterization Characterization of the particles Particle size was determined by field-emission scanning electron microscope (FE-SEM) using a JEOL JSM6700F operating at an accelerating voltage of 5 kV and the Quasi-Elastic Light Scattering (QELS) using a Nano ZS Zetasizer (ZEN3600; Malvern Instruments Ltd.). Nano ZS Zetasizer was also used to determine the zeta potential of the silica particles. The specific surface area and pore size distribution were obtained by the BET (Brunauer–Emmett–Teller) method using the adsorption/desorption isotherms of nitrogen at 77 K with Micromeritics ASAP 2020 apparatus. All samples Fig. 2 Schematic of chemical reaction between MAPE and surface-modified silica
145
were dried in the vacuum oven at 120◦ C for 2 h before measurement. Morphology The blend morphology was examined by FE-SEM. The samples were fractured in liquid nitrogen and then sputter-coated with palladium to avoid charging during observation. After tensile test, the morphology of the specimen was observed at two positions. One is the breaking region and the other is the elongation region. The specimens for tensile tests were prepared using a hot press (CH4386, Carver) in accordance with ASTM D638 typeVV at 190◦ C. Tensile tests were performed at a crosshead speed of 5.0 mm/min and room temperature using a TAPluse Testing Machine (LLOYD instruments™). Rheological properties For rheological characterization, the pellets were compression-molded into disks with a diameter of 25 mm and a thickness of 1 mm using a hot press at 190◦ C. All the measurements were carried out at 160◦ C under a nitrogen gas flow. Small-amplitude oscillatory shear measurements were carried out on Rheometrics Mechanical Spectrometer (RMS800) with a parallel plate fixture (25 mm diameter). Extensional properties were measured on SER-HVA01 Universal Testing Platform (Xpansion Instruments LLC) specifically designed for use as a detachable extensional rheometer fixture on commercially available rotational rheometer. The dimensions of the solid specimens were approximately 17 mm in length and 10 mm in width, with thicknesses in the range of 0.75∼0.85 mm. The constant extension rates were applied at 0.05, 0.1, and 0.5 s−1 .
146
Rheol Acta (2012) 51:143–150
a
b 105
105
104
0wt% 2wt% 5wt% 8wt%
G'[Pa]
G'[Pa]
Fig. 3 Storage moduli, G of HDPE silica nanocomposites at 160◦ C: a HDPE/MAPE/pure-silica and b HDPE/MAPE/APTES-silica
103
104
0wt% 2wt% 5wt% 8wt%
103
10-1
100
101
102
10-1
Frequency[rad/s]
Results In polymer/nanocomposites, particle–particle and polymer–particle interactions influence linear and nonlinear viscoelastic responses. A typical characteristic of nanocomposites is the transition from a liquid-like to solid-like linear viscoelasticity as the filler content is increased. In other words, the second plateau modulus appears in the storage modulus at low frequency, and this behavior is more pronounced than the loss modulus (Cassagnau 2008). However, the storage modulus for pure silica nanocomposites did not show such a trend as shown in Fig. 3a. As the pure silica content increased, the storage moduli were slightly increased. Though
a
100
101
102
Frequency[rad/s]
the absolute values of the storage modulus slightly increased, the terminal slopes of the storage moduli at low frequency remained constant at about 0.50 ± 0.03. This is because pure silica particles were not dispersed well in the polymer matrix and instead formed large aggregates. In contrast, the appearance of non-terminal behavior becomes more pronounced at a higher APTES-silica content. The terminal slope was decreased from 0.50 to 0.32 upon increased APTES-silica content. This behavior can be ascribed to improved particle dispersion due to the reaction of the APTES amino group and the MAPE maleic group. Though the rheological response was dependent upon the silica content in both systems, it is apparent that the
b 1010
1010 3
8wt%(x10 )
109
109 2
5wt%(x10 )
3
8wt%(x10 )
108
η E [Pa-s]
η E [Pa-s]
108
2
5wt%(x10 )
107 2wt%(x10)
106
106
0wt%
105
2wt%(x10)
107
0wt%
105
3η0 104
103 10-2
3η0
0.05[s-1] 0.1 0.5 10-1
100
101
time [s]
102
-1
0.05 [s ] 0.1 0.5
104
103
103 10-2
10-1
100
101
102
103
time [s]
Fig. 4 Transient extensional viscosity of HDPE silica nanocomposites at 160◦ C: a HDPE/MAPE/pure-silica and b HDPE/MAPE/APTES-silica
Rheol Acta (2012) 51:143–150
147
silica surface and matrix functionalization play more important roles. Understanding the rheological behavior of polymer melts under extensional flow is important to control and optimize polymer processing. The extensional viscosity is defined, η+ ˙) = E (t, ε
σzz (t) − σxx (t) ε˙
(1)
where, ε˙ is the extension rate, σ zz and σ xx are the stresses applied to the direction of extension and compression, respectively. At small deformation, the extensional viscosity is known to be related with the shear viscosity (Petrie 1979) η E (˙ε )˙ε = 3η(γ˙ )γ˙ |ε˙ ,γ˙ →0
(2)
where, γ˙ is the shear rate. The extensional viscosity becomes simply three times the zero-shear viscosity (η E (˙ε ) = 3η0 ) in the limit of vanishing strain rate. Figure 4 shows the uniaxial extensional viscosity of HDPE/silica composites using both pure- and APTESsilica at 160◦ C and at different extensional rates. The solid line represents the linear extensional viscosity, which is three times the shear viscosity as measured
Fig. 5 FE-SEM images; surface at breakup (a), (c) and elongated surface (b), (d); (a) and (b) with 5 wt.% pure-silica, (c) and (d) with 5 wt.% APTES-silica
from the viscosity curve under simple shear flow. All samples were in good agreement with the prediction provided by Eq. 2. HDPE/MAPE blend without silica showed weak strain hardening behavior such that the extensional viscosity increased with time and was higher than 3η0 . When pure-silica particles were added to the polymer matrix, there was no significant increase in extensional viscosity and no indication of strain hardening behavior. Additionally, the breakup of the specimen under extensional flow occurred earlier than the matrix polymer with increased concentration of pure-silica. On the other hand, HDPE/MAPE/APTESsilica nanocomposites showed strong strain hardening behavior under a uniaxial extensional flow field as shown in Fig. 4b. The extensional viscosities increased for short time period but were independent of the extension rate. After a certain amount of time, the curve sharply increased, manifesting strain hardening behavior, and then the strand was broken. In addition, a larger amount of APTES-silica increased the strain hardening behavior, creating a stronger network structure caused by the reaction between the amino group of APTES-silica and the maleic group of MAPE.
a
b
c
d
148
Rheol Acta (2012) 51:143–150
Fig. 6 FE-SEM images of HDPE/silica composites with (a) pure-silica and (b) treated-pure silica with KBr
a
b
Discussion Particle dispersion is one of the key factors that determine the extensional properties as well as shear properties of polymer composites. To investigate the dispersion of pure-silica and APTES-silica in the HDPE/MAPE blend, the morphologies of the surface at breakup and of the elongated surface after a tensile test were observed. Figure 5 shows FE-SEM images of specimens with 5 wt.% pure-silica and APTESsilica after a tensile test. Large aggregates of puresilica were observed on the fracture surface and at the elongated part in Fig. 5a, b, whereas APTES-silica particles were well dispersed in the polymer matrix as seen in Fig. 5c, d. The main reason for this difference could be due to the increased interaction between the particles and polymer matrix via coupling between the maleic groups of MAPE and amine groups of APTES-silica. The strain hardening behavior was observed only in HDPE/MAPE/APTES-silica composites, which may
G'[Pa]
105
104 0wt% pure-silica(5wt%) treated-pure silica(5wt%) 103 10-1
100
101
102
Frequency[rad/s]
Fig. 7 Storage modulus G of HDPE/silica composites at 160◦ C with 5 wt.% pure-silica and treated-pure silica with KBr
be due either to better dispersion of APTES-silica particles or to the network structure formed by the chemical reaction between the maleic group of MAPE and the amine group of modified silica. To separate these effects, we changed the silica dispersion without surface modification by APTES. After the colloidal silica suspension was dried after synthesis, the silica particles had a tendency to aggregate. These aggregates were difficult to disperse again due to strong van der Waals interactions. Based on the DLVO (Derjaguin–Landau–Verwey– Overbeek) theory, an energy barrier can be lowered by reducing the electrostatic double-layer repulsion, which can be achieved by controlling pH or adding large cations (Iler 1979). Tanahashi et al. (2006) succeeded in preparation of poly(ethylene-ran-vinyl alcohol) and polystyrene/silica nanocomposites without surface modification by the melt compounding through, the control of pore structure and fracture strength of the silica agglomerates by controlling pH or by adding large cations to the colloidal solutions of silica nanoparticles. To increase pure-silica dispersion in polymer matrix without silane coupling agent, we used K+ as a large cation and controlled pH by a nitric acid. As the pH of silica solution decreases from 7 to 2, the zeta potential changed from −40 to +2 mV. The isoelectric point (IEP) of pure-silica exists between pH 2 to pH 3. When the ratio of KBr/silica increased from 0 to 3 at pH 3, the zeta potential went to almost IEP at pH 3 and their average pore diameter of the dried silica aggregates changed from 30 to 60 nm. The pure-silica particles treated with KBr at pH 3 were named as treated-pure silica particles. The treated-pure silica particles were uniformly dispersed in the polymer matrix as shown in Fig. 6b, whereas pure-silica particles formed large aggregates as in Fig. 6a. Here, the mechanism for the dispersion of
Rheol Acta (2012) 51:143–150
149
109 treated-pur e silica(5w t%)x10
ηE [Pa-s]
108
dispersion and rheological properties under both smallamplitude oscillatory shear and uniaxial extensional flow, was investigated. The main observations were:
2
107 106
• pure-silica(5w t%)x10
3η 0
105
103 10-2
•
0.05 [s -1] 0.1 0.5
0wt%
104
10-1
100
101
102
103
time [s]
•
Fig. 8 Transient extensional viscosity of HDPE/silica composites at 160◦ C with 5 wt.% pure-silica and treated-pure silica with KBr
• treated-pure silica particles was different from that of APTES-silica. The surface of treated-pure silica did not have any functional group capable of reacting with the maleic group of MAPE. Therefore, the effect of particle dispersion can be determined by comparing the composites of the treated-pure silica particles with those of the pure-silica particles. Figure 7 shows a decrease in the slope of the storage modulus for the 5 wt.% treatedpure silica composite at low frequency. Based on FESEM and rheological results, the treated-pure silica particles were dispersed better than pure-silica on the polymer matrix, but did not display any strain hardening behavior in extensional viscosity as shown in Fig. 8. The growth of extensional viscosity of the treatedpure silica composite was similar to the HDPE/MAPE. Though particle dispersion was improved, no strain hardening behavior was observed. Among the possible causes of strain hardening, dispersion, and network formation, a high level of dispersion was not sufficient to induce strain hardening. Instead, the network structure formed by the chemical reaction between the maleic group of MAPE and the amine group of modified silica by APTES was what really induced the strain hardening behavior.
Conclusions Silica particles modified with ATPES were prepared by a modified Stöber process. HDPE nanocomposites with silica particles were prepared by melt compounding with 10% MAPE, which is partially compatible with HDPE. The effect of surface modification on silica
The rheological properties under small-amplitude oscillatory shear suggested that the surfacemodified silica was dispersed better in polymer matrix than pure silica. Based on the SEM of the specimen taken after the tensile test, the silica particles modified with APTES were well dispersed in the polymer matrix, whereas pure silica formed large aggregates. HDPE/MAPE/APTES-silica nanocomposites showed strong strain hardening behavior upon increased silica concentration. This is due to the network structure formed by the reaction between the amine group of silica and the maleic group of MAPE. Good dispersion alone did not induce strain hardening behavior; instead, formation of the network structure was required.
In summary, we designed HDPE/silica nanocomposites with enhanced melt strength. The polymer nanocomposite made from a linear polymer with surface-modified particles and an appropriate compatibilizer could generate strain hardening behavior. Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant (No. 20100027746) funded by the Korea government (MEST).
References Cassagnau P (2008) Melt rheology of organoclay and fumed silica nanocomposites. Polymer 49:2183–2196 Chan Y, White JL, Oyanagi Y (1978) A fundamental study of the rheological properties of glass-fiber-reinforced polyethylene and polystyrene melts. J Rheol 22:507–524 Cheng S, Phillips E, Parks L (2010) Processability improvement of polyolefins through radiation-induced branching. Radiat Phys Chem 79:329–334 D’Avino G, Maffettone PL, Hulsen MA, Peters GWM (2008) Numerical simulation of planar elongational flow of concentrated rigid particle suspensions in a viscoelastic fluid. J NonNewton Fluid Mech 150:65–79 Iler RK (1979) The chemistry of silica: solubility, polymerization, colloid and surface properties, and biochemistry. Wiley, New York Ishizuka O, Koyama K (1980) Elongational viscosity at a constant elongational strain rate of polypropylene melt. Polymer 21:164–170 Kakuda M, Takahashi T, Koyama K (2006) Elongational viscosity of polymer composite including hydrophilic or hydrophobic silica nano-particles. Nihon Reoroji Gakkaishi 34: 181–184
150 Kim DH, Fasulo PD, Rodgers WR, Paul DR (2008) Effect of the ratio of maleated polypropylene to organoclay on the structure and properties of TPO-based nanocomposites. Part II: Thermal expansion behavior. Polymer 49:2492–2506 Le Meins JF, Moldenaers P, Mewis J (2003) Suspensions of monodisperse spheres in polymer melts: particle size effects in extensional flow. Rheol Acta 42:184–190 Munstedt H (1980) Dependence of the elongational behavior of polystyrene melts on molecular weight and molecular weight distribution. J Rheol 24:847–867 Nielsen LE, Landel RF (1994) Mechanical properties of polymers and composites. Marcel Dekker, New York Okamoto M, Nam PH, Maiti P, Kotaka T, Hasegawa N, Usuki A (2001a) A house of cards structure in polypropylene/clay nanocomposites under elongational flow. Nano Lett 1:295–298 Okamoto M, Nam PH, Maiti P, Kotaka T, Nakayama T, Takada M, Ohshima M, Usuki A, Hasegawa N, Okamoto H (2001b) Biaxial flow-induced alignment of silicate layers in polypropylene/clay nanocomposite foam. Nano Lett 1:503– 505 Park JU, Kim JL, Kim DH, Ahn KH, Lee SJ, Cho KS (2006) Rheological behavior of polymer/layered silicate nanocomposites under uniaxial extensional flow. Macromol Res 14:318–323 Pavlidou S, Papaspyrides CD (2008) A review on polymerlayered silicate nanocomposites. Progr Polym Sci 33:1119– 1198 Petrie CJS (1979) Elongational flows: aspects of the behavior of model elasticoviscous fluids. Pitman, London Rong MZ, Zhang MQ, Zheng YX, Zeng HM, Friedrich K (2001) Improvement of tensile properties of nano-SiO2/PP composites in relation to percolation mechanism. Polymer 42:3301– 3304 Rozenberg BA, Tenne R (2008) Polymer-assisted fabrication of nanoparticles and nanocomposites. Prog Polym Sci 33:40–112
Rheol Acta (2012) 51:143–150 Song YS, Youn JR (2004) Modeling of rheological behavior of nanocomposites by Brownian dynamics simulation. Korea– Australia Rheol J 16:201–212 Spencer MW, Cui L, Yoo Y, Paul DR (2010) Morphology and properties of nanocomposites based on HDPE/HDPE-gMA blends. Polymer 51:1056–1070 Takahashi T, Wu W, Toda H, Takimoto JI, Akatsuka T, Koyama K (1997) Elongational viscosity of ABS polymer melts with soft or hard butadiene particles. J Non-Newton Fluid Mech 68:259–269 Tanahashi M, Hirose M, Lee JC, Takeda K (2006) Organic/inorganic nanocomposites prepared by mechanical smashing of agglomerated silica ultrafine particles in molten thermoplastic resin. Polym Adv Technol 17: 981–990 Tanaka H, White JL (1980) Experimental investigations of shear and elongational flow properties of polystyrene melts reinforced with calcium carbonate, titanium dioxide, and carbon black. Polym Eng Sci 20:949–956 Vlachopoulos J, Sidiropoulos V (2001) Polymer film blowing: modeling. In: Buschow KHJ, Robert WC, Merton CF, Bernard I, Edward JK, Subhash M, Patrick V (eds) Encyclopedia of materials: science and technology. Elsevier, Oxford, pp 7296–7301 Vlachopoulos J, Strutt D (2010) Rheology of molten polymers. In: Wagner JR, Jr (ed) Multilayer flexible packaging. William Andrew, Boston, pp 57–72 Wang W, Gu B, Liang L, Hamilton WA (2003) Fabrication of near-infrared photonic crystals using highly-monodispersed submicrometer SiO2 spheres. J Phys Chem B 107:12113– 12117 Wu Z, Xiang H, Kim TH, Chun MS, Lee KT (2006) Surface properties of submicrometer silica spheres modified with aminopropyltriethoxysilane and phenyltriethoxysilane. J Colloid Interface Sci 304:119–124