3062
Langmuir 2006, 22, 3062-3067
In Situ Compatibilizer-Reinforced Interface between a Flexible Polymer (a Functionalized Polypropylene) and a Rodlike Polymer (a Thermotropic Liquid Crystalline Polymer) Yongsok Seo,*,† Tran Hai Ninh,‡ Soon Man Hong,*,‡ Sehyun Kim,§ Tae Jin Kang,† Hansung Kim,† and Jinyeol Kim| School of Materials Science and Engineering and Intellectual Textile System Research Center (ITRC), College of Engineering, Seoul National UniVersity, Shillim9dong 56-1, Kwanakgu, Seoul, Korea 151-744, KIST R&D Academy, Hawolgokdong 39-1, Sungbukku, Seoul, Korea 130-650, Polymer Processing Technology Team, LG Chemical Ltd./Tech Center, Moonjidong 104-1, Yousungku, Daejon, Korea 305-738, and School of AdVanced Materials Engineering, Kookmin UniVersity, Chongnungdong, Sungbukku, Seoul, Korea 133-791 ReceiVed July 15, 2005. In Final Form: January 31, 2006 We present an investigation of the interfacial reinforcement between a flexible folded-chain polymer (functionalized polypropylene-maleic anhydride-grafted polypropylene, MAPP) and a rodlike polymer (a themotropic liquid crystalline polymer, TCLP - poly(ester amide)). Fracture toughness was measured using an asymmetric double-cantilever beam test (ADCB). High fracture toughness at the bonding temperature of 200 °C indicates that a chemical reaction has occurred at the interface to provide a strong interaction between MAPP and TLCP. Despite the higher modulus of TLCP, the fracture was propagated in the TLCP phase because of inherent TLCP domain structure. An analysis on the locus of failure revealed that at constant bonding temperature the fracture toughness between MAPP and TLCP was influenced not only by the bonding temperature but also by the bonding time. The fracture toughness increased with the bonding temperature until 200 °C was reached and then decreased at higher bonding temperature. The fracture toughness increased with annealing time until it reached a plateau value. We ascribe the dependence of the fracture toughness on the bonding time to the progressive occurrence of two different failure mechanisms, adhesive failure and cohesive failure. The adhesive strength increased with bonding temperature whereas the cohesive strength decreased because of weaker adhesion between TLCP crystalline domains. The dependence of fracture toughness on bonding time was explained in terms of the TLCP crystalline domain structure.
Introduction Polymer blends or alloys have been used in many important applications because of improvements in their mechanical and physicochemical properties over those of single-component systems.1-3 The final physical or chemical properties of polymer blends and alloys depend mostly on the interfacial properties as well as on the properties of both component polymers. Because most polymer pairs are thermodynamically immiscible, not enough interdiffusion of molecules exists to create entanglements on both sides of the interface.2,3 The interfacial fracture strength between them is very weak, so interface reinforcement by compatibilization is required to make useful polymer blends.4-6 This is also important for the so-called in situ composites, that is, two-phase polymer blends containing a thermotropic liquid crystalline polymer (TLCP) processed in such a way that the dispersed TLCP domains can be deformed into fibril shapes that †
Seoul National University. KIST R&D Academy. LG Chemical Ltd./Tech Center. | Kookmin University. * To whom correspondence should be addressed. E-mail:
[email protected];
[email protected]. ‡ §
(1) Majumdar, B.; Keskkula, H.; Paul, D. R. Polymer, 1994, 35, 3164. (2) Paul, D. R., Newman, S., Eds.; Polymer Blends; Academic Press: New York, 1978; Vol. 1. (3) Paul, D. R., Bucknall, C. B., Eds.; Polymer Blends; John Wiley & Sons: New York, 2000;,Vols. 1 and 2. (4) Wool, R. P. Polymer Interfaces: Structure and Strength; Hanser Publishers: New York, 1995. (5) Creton, C. In Polymer Surfaces and Interfaces; Richards, R. W., Peace, S. K., Eds.; John Wiley & Sons: Chichester, U.K., 1999. (6) Wu, S. Polymer Interface and Adhesion; Marcel Dekker: New York, 1982.
reinforce the matrix upon solidification.7 The incompatibility between the matrix polymer and the reinforcing TLCP leads to poor interfacial adhesion, which leads to a reinforcing effect that is less than that expected from the law of mixtures. The addition of a compatibilizer, when it is available, is an easy way of diminishing the incompatibility at a polymer/polymer interface.3 Another method, a reactive (or chemical) compatibilization, is prevalently used to control the morphology and to improve the interfacial properties of immiscible polymer blends.8-22 (7) Seo, Y.; Hong, S. M.; Kim, K. U. Macromolecules 1997, 30, 2978. (8) Cho, K.; Seo, K. H.; Ahn, T. O. Polym. J. 1997, 29, 987. (9) Jeon, H. K.; Kim, J. K. Macromol. Res. 1999, 7, 130. (10) Gong, L.; Friend, A. D.; Wool, R.P. Macromolecules 1998, 31, 3706. (11) Dai, C.; Kramer. E. J.; Washiyama, J.; Hui, C. Macromolecules 1996, 29, 7536. (12) Beck Tan, N. C.; Peiffer, D. G.; Briber, R. M. Macromolecules 1996, 29, 4969. (13) Norton, L. J.; Smigolova, V.; Pralle, M. U.; Hubenko, A.; Dai, K. H.; Kramer, E. J.; Hahn, S.; Berglund, C.; DeKoven, B. Macromolecules 1995, 28, 1999. (14) Boucher, E.; Folkers, J. P.; Creton, C.; Hervert, H.; Leger, L. Macromolecules 1997, 30, 2102. (15) Cho, K.; Li. F. Macromolecules 1991, 24, 2752. (16) Scott, C.; Macosko, C. W. J. Polym. Sci., Part B: Polym. Phys. 1994, 32, 205. (17) Kim, H. J.; Lee, K.; Seo, Y.; Kwak, S.; Koh, S. Macromolecules 2001, 34, 2546. (18) Kim, H. J.; Lee, K.; Seo, Y. Macromolecules 2002, 35, 1267. (19) Boucher, E.; Folkers, J. P.; Hervert, H.; Leger, L.; Creton, C. Macromolecules 1996, 29, 774. (20) Plummer, J. G.; Kausch, H.H.; Creton, C.; Kalb, F.; Leger, L. Macromolecules 1998, 31, 6164. (21) Kalb, F. Leger, L.; Creton, C.; Plummer, J. G.; Marcus, P.; Malgalhaes, A. Macromolecules 2001, 34, 2702. (22) Laurens, C.; Ober, R.; Creton, C.; Leger, L. Macromolecules 2001, 34, 2932.
10.1021/la051918c CCC: $33.50 © 2006 American Chemical Society Published on Web 03/07/2006
Adhesion Enhancement between MAPP and a TLCP
Reactive functionalized polymers form in situ block or graft copolymers at the interface during the blend preparation. For amorphous polymer blends, the failure mechanism at the polymer/polymer interface depends on the homopolymer pairs.4,9,11-13 For a semicrystalline polymer interface, Boucher et al.14,19 argued that the mechanism of reinforcement was similar to that observed for glassy polymers. They investigated the reinforcement of the interface between polypropylene (PP) and nylon 6 (Ny6) associated with the incorporation of a small amount of maleic anhydride-functionalized PP (MAPP), which reacts with the NH2 group of the Ny6 to form a copolymer in situ. (This method has also been applied to the in situ composite compatibilization between the matrix polymer and the reinforcing TLCPs.7) In situ block or graft copolymers generated at the interface during blend preparation transfer stress across the interface. Boucher et al.14,19 found that the fracture energy (Gc) for the PP/Ny6 system followed the same dependence on the areal density of the block copolymer as amorphous polymer pairs but deviated seriously from this behavior when the fracture samples were annealed above the melting point of Ny6. They attributed this extraordinary reinforcing effect to the presence of the β form of PP at the interface. However, subsequent experiments invalidated Boucher et al.’s suggestion.20-22 In a recent paper on the effect of crystalline orientation on the adhesion at polypropylene/polyamide 6 interfaces compatibilized with polypropylene-Ny6 diblock copolymers, Laurens et al. concluded that the specific orientation of the polymers in the immediate vicinity of the interface was not sufficient to promote strong adhesion.23,24 In our previous study, the interfacial fracture toughness between MAPP-added PP and Ny 6 increased with the bonding temperature until 220 °C was reached and then decreased at higher bonding temperatures.25 The progressive occurrence of two different failure mechanisms, adhesive failure at the interface and cohesive failure between chains, was given as the reason. A different crystalline PP phase was not identifiable by X-ray diffraction measurement for specimens prepared at different bonding temperatures. Fracture toughness increased with annealing time, passed a peak, and then reached a plateau value. The dependence of fracture toughness on bonding time could also be explained in terms of these two fracture mechanisms. However, the occurrence of high fracture toughness at a peculiar temperature remains to be explained clearly. Entanglement between chains is especially important for strong adhesion at an interface.8,25,26 Even though many studies have been carried out on interfaces between flexible folded-chain polymers, the interface between a thermoplastic folded-chain polymer (flexible coils in the melt state) and a rodlike polymer (a thermotropic liquid crystalline polymer) has not yet been studied. Such a study would be intriguing because unlike flexible polymers, thermotropic liquid crystalline polymers do not form random entanglements with each other. Using TLCP as another adhesion pair can provide some insight into the role of the inherent molecular structure of the polymer. We can probe the influence of the annealing temperature and the annealing time on the change in the interfacial fracture energy. An understanding of the reaction between MAPP and TLCP would also be quite useful for the preparation of compatibilized in situ composites that have physical properties that are better than those of the binary blend.7,27-31 (23) Laurens, C.; Ober, R.; Creton, C.; Leger, L. Macromolecules 2004, 37, 6806. (24) Laurens, C.; Creton, C.; Leger, L. Macromolecules 2004, 37, 6814. (25) Seo, Y.; Tran, H. N. Polymer 2004, 45, 8573. (26) Benkoski, J.J.; Flores, P.; Kramer, E. J. Macromolecules 2003, 36, 3289. (27) Seo, Y.; Kim, H.; Kim, B.; Hong, S. M.; Hwang, S. S.; Kim, K. U. Macromol. Res. 2001, 9, 238. (28) O’Donnel, H.J.; Baird, D. G. Polymer 1995, 36, 3113.
Langmuir, Vol. 22, No. 7, 2006 3063
Figure 1. Schematic diagram of the ADCB test.
Thus, the present work was undertaken to verify the feasibility of correlating the failure mechanisms with the inherent molecular structure of the polymers and with the adhesion temperature and bonding time. Experimental Section Materials. Materials employed in this study were commercially available polymers. Maleic anhydride-grafted polypropylene was from Honam Petrochemicals (Korea). The weight-average molecular weight, Mw, was 8.3 × 104 g/mol, and the polydispersity index was 4.95. The melting temperature (Tm) of MAPP measured by DSC was 165 °C. It has an average of 0.6 wt % succinic anhydride per chain. The TLCP used was a copolyester amide of 6-hydroxy-2naphthoic acid (60%), terephthalic acid (20%), and aminophenol (20%), commercially known as Vectra B950 (VB) manufactured by Celanese Hoechst. This material has been used and characterized by many researchers.27-31 It shows a glass transition at 143 °C and a nematic transition at 290 °C. It was supplied in the form of pellets. Pellets of MAPP and VB were dried in a vacuum oven at 100 °C for 24 h. Samples were then made by compression molding at 190 and 320 °C for MAPP and VB, respectively, with predetermined dimensions. Measurement of Fracture Toughness. The fracture toughness was measured using an asymmetric double-cantilever beam (ADCB) test because it has been shown to be a reliable test of the fracture toughness of a polymer interface.18,32,33 Details of this test are shown in Figure 1; a blade of thickness ∆ was inserted at the interface between MAPP and VB and was pushed into the sample. Although the materials were semicrystalline, the MAPP side remained transparent enough at the experimental thickness to allow a video camera to be used to measure the crack length. An image of the region ahead of the blade was recorded after an hour when there was no further increase in the crack length. Boucher et al.14 reported that the ADCB test yielded reliable values of the energy of adhesion, Gc, if two precautions were taken. First, the samples had to be asymmetric because the different mechanical properties of the two polymers might induce various modes of fracture. They also noted that varying the ratio of thickness changed the amount of the KII mode in the fracture process34 and that if the fracture tended to deviate into the more ductile material the measured energy release rate could increase significantly, leading to substantial errors in the evaluation of Gc.35 To minimize contributions of the second component, all of our samples were made with a thickness ratio hMAPP/htot of 0.67 because Gc had a minimum value at a ratio between 0.55 and 0.714 (However, our experimental data for the MAPP/VB interface did not show any noticeable change in the fracture toughness with a thickness ratio between 0.33 and 0.67.) In this system, the Young’s moduli of MAPP and VB are 1.1 and 80 GPa, respectively. Because the crack length ahead of the blade, a, was less than 10hVB for most of our samples, (29) Seo, Y.; Kim, J.; Kim, H. Polym. Eng. Sci. 2002, 42, 2401. (30) Valenza, A.; La Mantia, F. P.; Paci, M.; Magagni, P. L. Int. Polym. Process. 1991, 6, 247. (31) La Mantia, F. P.; Valenza, A.; Paci, M.; Magagni, P. L. Polym. Eng. Sci. 1990, 30, 7. (32) Creton, C. F.; Kramer, E. J.; Hui, C. Y.; Brown, H. R. Macromolecules 1992, 25, 3075. (33) Brown, H. R. Macromolecules 1991, 24, 2752. (34) Brown, H. R. J. Mater. Sci. 1990, 25, 2791. (35) Kanninen, M. F. Int. J. Fract. 1973, 9, 83.
3064 Langmuir, Vol. 22, No. 7, 2006
Seo et al.
Figure 2. FT Raman spectra of MAPP and the extracts from the binary blend of MAPP and VB. the following equation derived by Boucher et al.14 based on calculations by Kanninen,35 whose assumption was that the finite elasticity of the material ahead of the crack tip required correction factors for small crack lengths, was used Gc )
EMAPPhMAPP3EVBhVB3 3∆2 8a4 EMAPPhMAPP3RVB2 + EVBhVB3RMAPP2
Figure 3. Variation of the fracture toughness of the MAPP/VB interface with bonding time. Annealing temperatures were (4) 180, (9) 200, (O) 220, and (3) 240 °C, respectively. The lines are guides for the eyes. Scheme 1. Possible Reaction between MAPP and VB
(1)
where Ei and hi denote the Young’s modulus and the thickness of material i, respectively, and ∆ is the thickness of the blade. Ri is the correction factor for material i and is given by
Ri )
()
hi hi 1 + 1.92 + 1.22 a a
2
+ 0.39
hi 1 + 0.64 a
() hi a
3
(2)
Surface Characterization. Scanning electron microscopy (SEM) observations of the samples were performed on a Hitachi S-2500C. Fractured surfaces were coated with gold in an SPI sputter coater. The morphology was determined using an acceleration voltage of 15 keV. FT Raman spectra were obtained using a Perkin-Elmer system 2000 instrument equipped with near-IR optics. Spectra were obtained with 4 cm-1 resolution and typically required 10-15 min of acquisition time.
Results and Discussion In our previous report, we proposed some reactions occurring between the maleic anhydride group and the end groups of VB.36 To confirm these reactions, we prepared a binary blend of VB and MAPP (50:50 ratio) in a 42 mm Bravender twin-screw extruder (AEV651) equipped with a pulling unit. The extrusion temperatures of the feeding zone/transporting zone/meshing zone/ die were set at 140 °C/290 °C/290 °C/290 °C, respectively. Using a Soxhlet extractor with boiling dichlorobenzene, we extracted the soluble part (MAPP phase). After 1 week of extraction, the extracted solution was poured into ethanol. The precipitate was dried in a vacuum oven at 60 °C for 2 weeks to remove any solvent residue. An FT Raman spectrum was taken of the remnants. The Raman spectra of MAPP and the extract are shown in Figure 2. Characteristic peaks of an aromatic and an ester that cannot be seen in the MAPP spectrum are observable at 1380, 1620, and 1742 cm-1. This indicates that the VB moiety is included in the extracts. Because there is no aromatic moiety in the neat MAPP polymer and VB is not soluble in dichlorobenzene, we believe that some chemical reactions occurred to graft the (36) Seo, Y.; Kim, B. Y.; Kwak, S.; Kim, K. U.; Kim, J. Polymer 1999, 40, 4441.
VB moiety to MAPP. It may be conjectured that the reacted VB moiety would be from the low-molecular-weight VB formed by thermal depolymerization. However, the TGA thermogram of the TLCP (VB) did not show any signs of thermal degradation until the temperature reached 350 °C.37 It is well known that maleic anhydride can react with the amine groups of nylons.7-9 Hydroxyl end groups can also react with maleic anhydride to produce graft copolymers. (How many amine groups remain in VB is not detectable from IR analysis because of the overlapping of the amine peak with the hydroxyl peak, but the existence of the hydroxyl group can be clearly seen at 3300 cm-1 (Figure 21 of ref 7).) One of the possible reactions is that between the amine end group of VB and the anhydride group of MAPP (Scheme 1) to form a kind of (comb-shaped) block copolymer having different branches that will act as the compatibilizer at the interface.36 The parameters affecting the adhesion and the reaction are the bonding temperature and the initial concentration of functional groups on the polymer surface.18 Because we cannot precisely control the number of functional groups on the polymer surface, the effects of the bonding temperature and bonding time at constant temperature, both of which affect the chemical reaction at the interface, are examined in this study. Fracture toughness versus annealing time is plotted in Figure 3. Some facts are worthy of note. First, the interfacial fracture toughness for each temperature series increased with bonding time and reached a plateau. Second, the fracture toughness increased with bonding temperature, reached its highest value at a temperature of 200 °C, and then decreased with increasing temperature. The adhesion strength at the interface increased more rapidly at higher bonding temperatures. Third, the fracture toughness increased rapidly (37) Seo, Y. J. Appl. Polym. Sci. 1998, 70, 1589.
Adhesion Enhancement between MAPP and a TLCP
Figure 4. Variation of the fracture toughness between MAPP and VB with annealing temperatures (after 180 min of annealing time). The line is a guide for the eyes.
with a short induction time of annealing except when the annealing temperature was 240 °C. The general fracture toughness behavior is similar to that of an MAPP-added PP/Ny6 interface,25 but there are some subtle differences, as will be explained later. Figure 4 shows the fracture toughness as a function of bonding temperature. It shows a maximum at 200°C. At 200 °C, the energy of adhesion is saturated after 180 min near 45 J/m2. This implies that the interaction or the reaction occurring at the interface is a chemical reaction forming a covalent bond between MAPP and VB.36 We proposed some reactions occurring between the maleic anhydride group and the functional groups (mostly the amine group) of VB.37 As we mentioned above, the hydroxyl end groups as well as the amine end groups of VB can react with the maleic anhydride group of MAPP to produce graft copolymers. This interaction is not a physical one such as hydrogen bonding or van der Waals forces because it increases with temperature until 200 °C is reached. (We do not include chain entanglements as a physical interaction here.) Most physical interactions decrease with system temperature. Another reason for chemical interaction is the time-dependent behavior shown in Figure 3. If the interaction is solely physical, then no time dependence of the fracture toughness should exist at a constant bonding temperature.18 The fracture toughness at the interface was almost zero when the bonding temperature was 50 °C. This means that the physical interaction is not significant at all. Raising the temperature to 220 °C accelerates the rate of increase of the fracture toughness but decreases the saturation value. The growth rate increases with temperature because of faster reactions. We speculate that there is a slow increase in adhesion strength at a bonding temperature of 240 °C because of some degradation of MAPP at that temperature, which was checked by TGA (the data are not shown here because of space limitations), hence there are fewer chemical reactions at the interface. As the intermolecular reaction between MAPP molecules and VB molecules proceeds, more graft copolymers are produced at the interface. The fracture toughness also increases with the areal density of the graft copolymers, and the adhesion strength at the interface increases more rapidly at higher bonding temperatures. Because the cooling conditions for each series of samples were the same, the early increase in the interfacial adhesion with bonding time must mainly be due to an increase in the number of intermolecular reactions at the interface. However, it should be pointed out that the areal density of the produced copolymer cannot increase indefinitely.10,11,38 (38) Lo, C.; Laabs, F.; Narasimhan, B. J. Polym. Sci., Part B: Polym. Phys. 2004, 42, 2667.
Langmuir, Vol. 22, No. 7, 2006 3065
In our previous study of enhanced interfacial adhesion between MAPP-added PP and Ny6, we reported that an optimum bonding temperature existed because the adhesion strength at the interface increased with the bonding temperature because of the rapid reaction and the cohesive strength decreased because of fewer entanglements of the produced copolymer with other molecules on the bulk side of PP.18 We expected that a similar fracture mechanism change would be possible for the MAPP/VB interface. Figure 5 shows SEM micrographs of the fractured surface on the MAPP side for a series of typical specimens bonded for 60 min at different temperatures. All of the fractured surfaces show almost the same fractures, with the surfaces being covered with many fibril-shaped domains. (When the bonding temperature was 240 °C, a slightly different look appeared. MAPP degraded somewhat at that temperature.) This shows that the cracks propagated in the VB phase. This result is a little unexpected because the yield stress of VB is much higher than that of MAPP. However, the yield stress of VB is that in the fibril axis direction, not that in the normal direction.29 Once a MAPP molecule reacts with VB molecules at the surface, the whole MAPP chain forms a branched or tethered chain. Increasing the temperature increases the number of reactions and hence the adhesion strength. In the MAPP/VB system, because of VB’s ordered molecular structure, the cracks tend to propagate in the bulk of the VB phase despite the fact that the yield stress of VB is much higher than that of MAPP. Most TLCPs have rodlike molecules that are easily packed in an ordered state. They are in local crystalline domains.39 The adhesion strength between local crystalline domains is not as strong as that between entangled molecules, hence VB crystalline domains have low cohesive strength.29 The interface between crystalline domains and the amorphous phase in VB is a more disordered state at temperatures above Tg. Hence, the adhesive strength decreases at higher temperature because of the increased free volume between them.40 Because the MAPP used in this study has a molecular weight that is much higher than the critical value for effective chain entanglements, the cohesive strength in MAPP is much stronger than that of the VB phase, which has inherent molecular ordering. Surface elemental compositions by XPS analysis provide the elemental compositions of the fractured MAPP side (Table 1). The key elemental difference between MAPP and VB is the presence of nitrogen in VB (ca 2.6%). The composition percent of nitrogen on the MAPP side almost coincides with that in VB molecules, implying that the MAPP surface is mostly covered with VB molecules. At higher bonding temperatures, however, the interfacial adhesion between the crystalline domains of VB became weaker (Figure 7). Hence, VB microdomain separation occurred more easily at higher bonding temperatures. The adhesion strength of the MAPP/VB interface is much lower than that of the functionalized PP/Ny6 interface because of the different molecular structure between VB and Ny6.25 The strength of interfacial adhesion is determined from the adhesive strength at the interface and from the cohesive strength of both polymers, which depends on the states of individual chains and, hence, on the bonding time. Depending on the intrinsic failure energy of the interface, failure prefers the weaker of the two. The locus of failure analyzed by using microscopic examination supports this reasoning. Figure 6 shows SEM (39) Lovinger, A. J.; Amundson, K. R.; Davis, D. D. Chem. Mater. 1994, 6, 1726. (40) The impact strength of the VB phase was measured according to ASTM D256 after annealing the VB sample for 180 min at 180, 200, 220, and 240 °C. Though not shown here because of space limitations, the impact strength of VB decreased with annealing temperature. We speculate that this is because of the increased free volume between microdomains and the free volume in the amorphous liquid phase of the VB molecules.
3066 Langmuir, Vol. 22, No. 7, 2006
Seo et al.
Figure 5. SEM photographs of the cleaved MAPP surface after bonding with VB for 1 h at different temperatures: (a) 180, (b) 200, (c) 220, and (d) 240 °C. Upper row, ×500 magnification; bottom row, ×2000 magnification. Table 1. Elemental Compositions of Fractured PPMA Surfaces Measured by XPS annealing time in min (bonding temperature ) 200 °C) 60 90 120 240
%N
annealing temperature in °C (bonding time ) 180 min)
%N
1.84 1.97 2.16 2.29
180 200 220 230
2.37 2.29 2.09 2.11
micrographs of the fractured surface on the MAPP side for a series of typical specimens bonded at 200 °C. When the bonding time was short (30 min), MAPP molecules reacted less to result in a lower toughness and a less rugged fracture surface. Cleaved MAPP surfaces annealed for a long time (longer than 60 min) were rougher and more rugged, and many fibrils adhered to the MAPP surfaces because of crack propagation in the bulk of the VB phase. In the MAPP/VB system, however, after the adhesive strength at the interface had become larger, the cracks propagated along the weak domain boundary in the VB phase to go with the transfer of VB chains onto the surface of the MAPP (Figure 7). Thus, the MAPP surface is covered with detached VB molecules. In the MAPP/Ny6 system, the cracks tended to propagate in the bulk of the material with the lower yield stress (i.e., polypropylene) when the interface was strong because the yield stress of polypropylene is much lower than that of polyamide. In the MAPP-added PP/Ny6 system, the fracture toughness obviously showed a maximum for a bonding time of around 60 min when the bonding temperature was 220 °C.25 We ascribed the appearance of a maximum fracture toughness at a particular time (60 min at 220 °C) to a change in the fracture mode. For the MAPP-added PP/Ny 6 system, the initial increase in the fracture toughness with annealing time was due to the increased number of reactions between MAPP and Ny6 at the interface at longer bonding times. A longer bonding time also provides more relaxation time for other PP chains to have fewer entanglements with the molecules at the interface. The other PP chains move away from the highly energetic interface.18 The more reactions that occur, the more that other chains move away from the interface and the less that other chains move to the interface. Therefore, the fracture toughness in the MAPP-added PP/Ny 6 interface increased with annealing time because of the increased adhesive
strength caused by interfacial reactions, passed a peak, and then decreased to a plateau value because of both a critical surface coverage by the reacted molecules and reduced cohesive strength. However, no obvious maximum was observed in the MAPP/VB system because of the long relaxation time of VB molecules and hence the unchanging cohesive strength between local VB domains with bonding time (Figure 7). For general polymer blends, as we mentioned in the Introduction, the failure mechanism at the polymer/polymer interface depends on homopolymer pairs. In glassy amorphous polymers, the stress is mainly transferred by entanglements so that the fracture toughness of the interface is negligible until the chains on both sides of the interface are mutually entangled.11 In the case of a diblock copolymer at the interface of glassy amorphous polymer where the full length of a block is entangled with its respective homopolymer, the fracture mechanism is a function of the areal density of the block copolymers and of the length of each segment of the block copolymer.13,14,31 Thus, the fracture mechanism changes from chain pullout to chain scission and then finally to crazing. For an interface between an amorphous polymer and a semicrystalline polymer (e.g., a polystyrene/ polyethylene system), Kramer et al.26 observed that most diblock copolymers were more likely to fail by chain pullout during craze breakdown. Because of the low crazing stress, forces less than that required for scission were sufficient to support wide crazes. They concluded that the major transitions depended on the crystalline microstructure. Kramer and co-workers24 investigated the effect of many variables on the behaviors of interfaces involving semicrystalline polymers. However, for a semicrystalline polymer interface, Boucher et al.14,19 argued that the mechanism of reinforcement was similar to that observed for glassy polymers. For the semicrystalline polymer/thermotropic liquid crystalline polymer interface studied in this article, the fracture behavior is similar to the chain pullout of a semicrysalline polymer/amorphous polymer system or to the chain pullout of an amorphous polymer pair interface. After the long annealing time needed to reach the plateau value at 200 °C, the Gc value had reached 45 J/m2 because of the weak adhesion between VB crystalline domains. Because of the almost constant adhesive strength in the VB domains at constant bonding temperature, no maximum was observed with the bonding time. However, because
Adhesion Enhancement between MAPP and a TLCP
Langmuir, Vol. 22, No. 7, 2006 3067
Figure 6. SEM photographs of the cleaved MAPP surface after bonding with VB at 200 °C: (a) 30, (b) 60, (c) 90, (d) 120, and (e) 180 min. Magnification is ×600.
Figure 7. Schematic representation of the locus of failure. The upper side shows MAPP chains, and lower ones are crystalline VB domains. The empty circles represent the functional groups on the MAPP molecules (succinic anhydride group). When the adhesive strength is lower than the cohesive strength, failure (dotted lines) occurs first through adhesive failure at the interface. After the former becomes strong enough to endure the stress, however, the failure occurs at the weak crystalline domain boundary on the VB side. At low temperatures, the number of reacted molecules is small (hence the adhesion strength is low) whereas at higher bonding temperatures more reactions occur for MAPP molecules. The adhesion strength increases with temperature whereas the cohesive strength between VB crystalline domains decreases with temperature. Therefore, an optimum temperature exists for a maximum interface strength.
of the variation in the VB phase (adhesion strength and reaction rates with temperature), a maximum in the fracture strength was observed at a bonding temperature of 200 °C. Therefore, we can conclude that entanglements of block (or graft) copolymer with the molecules on the bulk side are quite important for achieving high fracture toughness in a semicrystalline polymer pair system, especially when the interfacial adhesion is strong enough to withstand the failure stress.41
Summary In this study, the interfacial fracture toughness between a flexible coiled polymer (semicrystalline polymer) and a rigid (41) Duvall, J.; Sellitti, C.; Myers, C.; Hiltner, A.; Bare, E. J. Appl. Polym. Sci. 1994, 52, 207.
rodlike polymer (a TLCP, poly(ester amide)) was studied. Experimental results verify that some chemical reactions occurred at the interface. The interfacial fracture toughness was substantially enhanced by the reactions at the interface (i.e., the formation of graft copolymers at the interface). However, the fracture toughness value was not as high as that of the interface in the functionalized PP/Ny6 system. This is ascribed to the difference in the molecular structures between the Ny6 and TLCP molecules. Though the yield stress of VB is much higher than that of MAPP, cracks propagated in the TLCP phase because of the ordered molecular structure of the TLCP phase in the local crystalline domains. The adhesion strength between the local crystalline domains was not as strong as that between entangled flexible molecules. The fracture strength showed a maximum at a bonding temperature of 200 °C because of the temperature dependence of both the reactions at the interface and the cohesive strength in TLCP microdomains (increased adhesive strength with the reactions at the interface and reduced cohesive strength between TLCP crystalline domains). For each temperature series, the fracture toughness increased with bonding time and then reached a plateau, but no obvious maximum was observed because of the almost constant adhesive strength in the VB domains at constant temperature. Therefore, the general behavior of the fracture strength at the MAPP/TLCP interface is similar to chain pullout of block copolymer molecules at the interface of semicrystalline polymer pairs (the functionalized PP/Ny6 system). Thus, we can conclude that entanglements of block (or graft) copolymers with molecules on the bulk side are quite important for achieving high fracture toughness in a semicrystalline polymer-pair system, especially when the interfacial adhesion is strong enough to withstand the failure stress. Acknowledgment. This work was supported by the SRC/ ERC program of MOST/KOSEF (R11 - 2005-065) (Y.S. and T.J.K.) and by MOCIE (Y.S., no. RIAMD-AC05-05 and J.K., no. 13-5060). We acknowledge helpful discussions with Dr. Hyong-Jun Kim at the University of Michigan. LA051918C
