Super-toughening polyamide-612 by controlling dispersed phase domain size: Essential work of fracture assessment

Accepted Manuscript Super-toughening polyamide-612 by controlling dispersed phase domain size: Essential work of fracture assessment Sunil Kumar, Saurindra N. Maiti, Bhabani K. Satapathy PII: DOI: Reference:

S0261-3069(14)00412-9 http://dx.doi.org/10.1016/j.matdes.2014.05.043 JMAD 6521

To appear in:

Materials and Design

Received Date: Accepted Date:

26 February 2014 21 May 2014

Please cite this article as: Kumar, S., Maiti, S.N., Satapathy, B.K., Super-toughening polyamide-612 by controlling dispersed phase domain size: Essential work of fracture assessment, Materials and Design (2014), doi: http:// dx.doi.org/10.1016/j.matdes.2014.05.043

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Super-toughening polyamide-612 by controlling dispersed phase domain size: Essential work of fracture assessment Sunil Kumar, Saurindra N. Maiti, Bhabani K. Satapathy* Centre for Polymer Science and Engineering, Indian Institute of Technology Delhi, Hauz Khas, New Delhi-110016, India *corresponding author: [email protected]; [email protected] Telephone: 0091-11-26596043, Telefax: 0091-11-26591421 Abstract A new pathway to super-toughen polyamide-612 (PA-612) by incorporating domains of soft poly(octene-co-ethylene)-g-maleic anhydride (POE-g-MA) via melt blending leading to more than ~1100% increase in notched Izod impact strength vis-à-vis fracture toughness enhancement is demonstrated. Fourier transform infra red (FTIR) studies showed effective phase interactions between PA-612 and POE-g-MA whereas dynamic mechanical analysis (DMA) revealed a reduction in loss-peak intensity at ~45°C with increase in the soft phase fraction. The optimal dependence of fracture-toughness (in plane-stress) on domain-size (Dn) of dispersed-phase in the form of a reduction in resistance to crack initiation indicated by essential work of fracture (we) and linear increase in resistance to crack propagation indicated by non-essential work of fracture (βwp) of the blends ≥ 10 wt% of POE-g-MA content is correlated to an increase in domain-size ≥ ~0.3 µm. Fracture surface morphology indicated crazing to be responsible for the transition in fracture behavior, i.e. remarkable toughening of PA-612 at the critical rubber phase domain size range of ~0.2-0.3 µm.

Keywords: Polyamide; Toughness; Fracture.

1

1. Introduction Engineering plastic materials that are dimensionally stable and fracture resistant may be attractive to automotive and structural sectors when the toughness criteria at room temperature are ensured. In this regard, polyamide-612 being a matrix with comparably higher dimensional stability and low moisture sensitivity than the other polyamide analogues, the aspects pertaining to toughening may need viable attention both technologically and economically. In this regard melt blending of elastomeric/rubbery soft-phase with the matrix polyamide may lead to many advantages including modifying the morphological requirements at the micron-scale causing an enhancement in resistance to crack initiation and propagation apart from increasing the overall impact strength. Theoretically, the key factors controlling the toughening mechanism of blends are: (a) type and amount of rubbers, (b) size of the micro-domains of dispersed phase, (c) interparticle distance and (d) interaction/adhesion between matrix and rubber [1]. Reportedly, the optimum soft-phase (rubber) particle diameter of ~0.2-0.4 µm leads to maximum toughening of rubber-toughened polymer blends [2]. However, most extensive literature survey though arrives at a general consensus on rubber phase induced toughening of polymers, it does not necessarily conform to any critical or fixed range of rubber/softphase/impact modifier loading that may lead to maximum toughness enhancements. The methods of measurement of toughness of polymers, blends and composites were reported to be widely variable, ranging from notched and un-notched Charpy and Izod tests to area under the uni-axial tensile stress-strain curves. These methods though are practicable their relevance in leading to deeper insights into the aspects responsible for toughening remains debatable because toughness as a material property is testing-geometry dependent and most of the ductile polymer based systems in reality do not undergo complete fracture. Therefore, energy based toughness evaluation methods such as essential work of fracture (EWF) approach was found widely acceptable in characterizing intrinsic toughness of polymeric materials, including that of toughened binary and ternary blends, micro- and nano-structured polymeric systems. The method enables the partitioning of the two energy components that are responsible for fracture initiation and fracture propagation.

2

For example, Yin at al. [3] reported on the largely improved impact toughness due to the formation of melt processed induced core cell particles that effectively act as particles bridges and thereby enabling the absorption of fracture energy via penetration of micro crack propagation in case of PA-6/EPDM-g-MA/HDPE blends. Su et al. [4] studied PET/POE/POE-g-MA (PET: POE = 80:20) blend systems and found that an optimization of toughening could be attained at 3 wt% of POE-g-MA, where the notched impact strength enhanced 15-fold relative to neat PET. The observed toughening behavior was attributed to a combination of good dispersion of elastomer phase (particles) and optimally appropriate interfacial adhesion. Ma et al. [5] have investigated on tough PA6/EPDM-g-MA (80:20) blends which can further be potentially toughened with βnucleated thermoplastic vulcanizates (TPVs). Another similar attempt led to a significantly enhanced level of toughness, i.e. 10-fold increase due to addition of TPV into elastomer modified PP matrix such as PP/EPDM-g-MA [6]. Ozkoc et al. [7] have reported on fracture toughness of olefins based compatibilized PA-6/ABS blends where significant increase in we by > 15-fold and in βwp by > 18-fold were reported by EWF method using SENT specimens. Similarly the fracture toughness of K-g-MA compatibilized PC/K-resin blend was reported where an increment in we upto ~33% could be observed without substantially affecting βwp [8]. Fung et al. [9] investigated fracture characteristics of PET/SEBS-g-MA binary blends under quasi static and impact loading conditions, where an increase in fracture toughness from ~7 kJ/m2 to ~17 kJ/m2 with 10 wt% of EPDM-g-MA incorporation was reported. Furthermore, such an increase was found to have a correspondence to > 60-fold increase in Charpy impact strengths of the blends at identical levels of elastomer loading. Such remarkable increment was attributed to the development of prominent outer plastic deformation zone (OPDZ), which was microscopically characterized as a consequence of de-bonding of rubber particles and subsequent cavitations of PET matrix. Wang et al. have put forth a comparative evaluation of POE-g-MA based blends of semi-crystalline nylon 6 vis-à-vis an amorphous polyamide (a-PA) [10]. The increases in fracture toughness in these systems are characterized by an arrested crack tip which led to the propagation of a critical upper limit and inter particle distance (IPD) with respect to the soft-phase domain size i.e., of elastomeric phase. The increase in fracture energy by > 5-fold and by > 3.53

fold for a-PA and nylon-6 respectively was attributed to resistance of rubber particles followed by crazing and massive shear yielding of matrix. In an effort to investigate the effect of reactive compatibilization of PA-6/EBA blends Balamurugan and Maiti [11] have reported more than 3-fold increase in we along with more than 20% reduction in βwp for the uncompatibilized binary and compatibilized ternary blend systems at identical levels of EBA- contents (EBA or EBA+EBA-g-MA= 10 wt%). Similarly Pegoretti and Ricco [12] have reported on EWF of neat PA-66 and PA-66/rubber blends, where an increase in we and decrease in βwp in 7 wt% rubber filled blend have been attributed to a threshold average particle diameter (dn) of 139 ± 27 nm, above which we and βwp were reported to remain unaffected or follow a decreasing trend. In a unique effort to understand the matrix softening effect, Tang et al. [13] have systematically evaluated the effect of β-phase PP on the fracture behavior of dynamically vulcanized PP/EPDM blends. It was reported that above 0.3 wt% β-nucleating agent incorporation the βcrystallanity of the PP phase appreciably increased to ~15-35% which has a correspondence to an increase in we by more than 15% without significantly affecting the βwp. The study thus reiterates the fact that β-modification of PP matrix in binary PP/EPDM elastomer blend may effectively lead to increase in resistance to stable crack initiation. In contrast to the conventional fracture parametric assessment of semi-crystalline polymer materials such as polyamides with substantial ductility/plasticity the aspects pertaining to the fracture-energy or work of fracture factors in controlling the overall failure response is less understood. Therefore the present paper fundamentally attempts to investigate on the toughening via fracture performance enhancement of polyamide-612 (PA-612; with relatively higher methylene-to-amide (-CH2- to -NH-CO-) ratio compared to the conventional nylon 6 and nylon 66) by mixing POE-g-MA (weight fraction controlled) as an elastomer. 2. Experimental details 2.1. Preparation of PA-612/POE-g-MA blends Commercially available polyamide-612 (PA-612) and poly (octene-co-ethylene)-gmaleic anhydride (POE-g-MA) with trade names Vestamid (DX9300) from Evonik (Germany) and Fusabond (N493) from Dupont (India) respectively, were blended in 4

various compositional ratios. The details of PA-612 and POE-g-MA were already reported by the authors elsewhere [14]. Co-rotating type (Steer Omega-20) twin screw extruder with L/D = 40 has been used for the melt mixing of PA-612 and POE-g-MA at various compositional ratios to give rise to a series of [(wt/wt%: 100/0 (NE-0), 95/5 (NE5), 90/10 (NE-10), 80/20 (NE-20), and 65/35 (NE-35)] blends. The screw speed was kept at 200 rpm, whereas the temperature profile of the extruder barrel was maintained at 220 ºC in the feed zone and at 245 ºC in the die zone for the preparation of the blends. The continuous strands were chopped to obtain granules which were later subjected to drying in the oven at 80 ºC for ~8 h prior to being injection molded (L&T Demag make) at a pressure of 60 bars, injection time of 4 s and cooling time of 25 s to obtain plates of 80 mm x 80 mm squares of ~1 mm thickness. The temperature profile for injection molding was kept at 40 ºC, 222 ºC, 230 ºC, 245 ºC, and 260 ºC corresponding to the feed-zone (Z1), meter-zone (Z-2), compression-zone (Z-3), heating-zone (Z-4) and die-zones (Z-5) respectively. Rectangular strips of dimension 80 mm x 20 mm x ~1 mm were machined from the plates for fracture mechanics investigations. 2.2 Domain size determination of dispersed phase in the blends The morphologies of the cryogenically fractured and etched surfaces of PA-612/POE-gMA blends have been investigated using scanning electron microscopy (SEM) on a Zeiss EVO-50 electron microscope to analyze the dispersion by measuring the domain sizes of rubbery phase in the PA-612/POE-g-MA blends. Prior to being mounted for microscopy, the samples were kept in boiling xylene for 24 h for the extraction of the rubber phase (POE). The surfaces of the specimens were made conductive by gold sputter coating. The micrographs obtained by SEM for the determination of domain sizes are shown in Fig. 1. The domain sizes of the dispersed soft phase were measured using Image J software. The morphological parameter characterized by number average diameter (Dn) was determined by the following equation-1, reported elsewhere [14]. The equation-1 may be given as below.

Dn =

∑ND ∑N i

i

i

5

(1)

where Ni is the number of particles and Di is diameter of particles. The inter-particle (inter-domain) distances have been calculated based on percolation theory following Wu’s equation [1]. The equation-2 may be given as below.

( 6φ )

⎡

τ = Dn ⎢⎢ π ⎢⎣

1

3

⎤ ⎥ −1⎥ ⎥⎦

(2)

where, Dn is the number average domain diameter of the dispersed phase (POE-g-MA) corresponding to a volume fraction φ . The details of the rubber modified PA-612 composition with their respective domain sizes (Dn) and inter-domain distances (τ) are given in Table-1. With the increase in rubber content τ decreased indicating the theoretical possibility of the material to readily undergo ductile failure behavior with the increase in POE-g-MA content, an observation reported earlier [14]. 2.3 Differential scanning calorimetry (DSC) Differential scanning calorimetry (DSC) was carried out at a heating rate of 10 °C /min in a temperature range of 20 ºC to 240 °C in nitrogen atmosphere on a TA instrument, USA (Q 200). The melting behaviour and other thermal transitions in neat PA-612 and PA612/POE-g-MA (NE) blends have been determined from the heat flow versus temperature plot. The samples were pre-heated to remove the residual thermal stress in the range of 20 ºC to 240 °C and then held at 240 °C for 3 minutes prior to getting cooled down to room temperature at the rate of 10 °C/min. Subsequently the second heating run was ensued at a rate of 10 °C /min till 240 °C. 2.4 Fourier Transform Infrared Spectroscopy (FTIR) FTIR spectrometer (Nicolet 200) was used for recording the IR spectra of the films samples of PA-612 and NE blends at room temperature (30±2 ºC) in the range of 4000400 cm-1. The films used for FTIR studies were prepared by compression molding at 260 ºC/ 8000 lbs. 2.5 Quasi-static mechanical properties The notched Izod impact strength, tensile mechanical properties (tensile strength and tensile modulus) and three-point bending flexural properties (flexural strength and flexural modulus) have been determined following ASTM: D256, ASTM: D638 and ASTM: D790 standards [15]. The mechanical properties are detailed in Table-1 as a 6

function of domain size of the dispersed phase (and/or) i.e. POE-g-MA content in the blends. 2.6 Dynamic mechanical analysis (DMA) and rheological behaviour of blends Dynamic mechanical analysis (DMA) measurements of the neat PA-612 and NE blends have been carried out as per ASTM: D4065 test procedure in single cantilever mode on an Q800 (TA Instruments, USA) using specimens with dimensions of (35 x 13 x 3) mm3. In order to qualitatively assess the energy dissipation ability and to quantitatively ascertain shift (if any) in glass transition temperature of the blends, the storage-modulus (E’), loss-modulus (E”) and loss-factor (tan δ) have been plotted as functions of temperature. The experiment was performed in the temperature range of -80 °C and 120 °C at a frequency of 1.0 rad/s and at a heating rate of 5 °C /min. Melt rheological tests of neat PA-612 and their blends were performed on a Malvern rotational rheometer (Bohlin Gemini model) in parallel-plate oscillatory mode. The dynamic frequency (ω) sweep test was conducted in the frequency range of 0.01 to 100 rad/s at a temperature of 250 °C. To ensure the linear viscoelastic region the strain amplitude was maintained at 2% of the value. 2.6 Essential work of fracture (EWF) methodology The essential work of fracture (EWF) measurements have been carried out by applying uni-axial tension (performed on Zwick Z250 Universal Testing Machine) on doubleedge-notched-tension (DENT) specimens, with varying ligament lengths of ~3-10 mm, prepared from rectangular injection moulded bars of 80 mm × 20 mm × 1 mm . The fracture of these pre-notched specimens under tensile loading was conducted at a constant extension speed (5 mm/min at room temperature (30±2 ºC) while clamp distance was fixed at 40 mm) to obtain individual load-displacement curves. The essential work of fracture (EWF) approach methodically enables the distinguishing between two terms representing the resistance to crack initiation (we) and resistance to crack propagation (βwp) corresponding to inner fracture process zone (IFPZ) and outer plastic deformation zone (OPDZ) as was already reported [16]. For each material, at least 8 specimens were tested with different ligament lengths to obtain data of acceptable statistical relevance. Mathematically, the EWF method under plane stress conditions, valid for thin plates and sheets, rests upon the linear equation that accounts for the total work of fracture W 7

(kJ/m2) as the sum of elastic (We) and plastic (Wp) work of fractures characterizing the energy dissipated in the inner fracture process zone (IFPZ) and outer plastic deformation zone (OPDZ) respectively. The schematic of the DENT specimens showing the two fracture process zones is shown in Fig. 2. Therefore W may be expressed as, W = We + Wp = we . B. l + βwp . B . l2

(3)

where B, l and β are specimen thickness, ligament length and shape factor of the plastic zone, respectively. After dividing W by the ligament (notched) area i.e., B. l, the specific work of fracture w is obtained:

w = we + βwp . l

(4)

where, the quantities we (EWF) and βwp (Non-EWF) are in N/mm and N/mm2 units respectively. The we and βwp may be obtained from the intercept and slope of the extrapolation of w as a function of l to zero ligament length [17,18]. 2.7 Fractured surface morphology The fracture-surface morphologies of virgin components of the blends viz. PA-612 and their blend compositions have been investigated using scanning electron microscopy (SEM) on a Zeiss EVO-50 electron microscope to analyze the associated failuremechanisms and structural integrity of such binary blends. The surfaces of the specimens were sputter coated with gold prior to examination to make the surfaces conductive. 3. Results and Discussion 3.1 Differential Scanning Calorimetry (DSC) The DSC heating curves of PA-612 and PA-612/POE-g-MA blends are shown in Fig. 3 (a). The melting endotherm peak remained unaffected at 219 oC while the peaks are decreased with increase in POE-g-MA content. The peaks for the NE blend broadened indicating significant interference of amorphous POE-g-MA phase with the crystalline phase of PA-612. The appearance of the mild exothermic shoulder peak before melting endotherm may be attributed to the unfolding of the polymer chains from the structured crystallites. 3.2 Fourier Transform Infrared Spectroscopy (FTIR) The ATR-FTIR spectra of PA-612 and PA-612/POE-g-MA blends are shown in Fig. 3 (b). The variation in the characteristic IR absorption peaks with composition has been witnessed in terms of (i) increasing intensity of C-H stretching absorption peaks in the 8

range of ~3000-3500 cm-1 with increase in POE-g-MA and (ii) peak splitting accompanied with enhanced intensity of the bending vibration peaks at ~750 cm-1 due to interference of POE-g-MA with PA-612 matrix. 3.3 Quasi-static mechanical properties The relative mechanical properties that are normalized with respect to crystallinity of the polyamide phase are given in Table-1. The crystallinity-normalized tensile modulus (E/χ) and crystallinity-normalized tensile strengths (σ/χ) revealed a ~30% and ~16% drop in the toughened blend with 10 wt% of POE-g-MA (with Dn ~ 0.30µm). The normalized impact strength however, increased 11-fold at the same composition. The increment in impact strength has also been accompanied with ~25% and ~20% reduction in relative flexural moduli and flexural strengths in the blend with Dn ~ 0.30 µm. Thus the critical domain size may be taken as Dn ~0.30 µm, above which the ductility index has also been observed to increase 17-fold (rendering the material much softer and pliable), i.e. corresponding to a domain size Dn ~0.61 µm in 20 wt% of POE-g-MA containing PA612 blend. Typically, for polyamides, such as PA-6 (notched Izod impact strength = ~32117 J/m), super-toughening has been reported [19] to be when the toughness or impact strength increases more than 8-fold (~800% increase). For PA-66 however a much greater enhancement in the impact strength needs to be ensured for super-toughening since the notched Izod impact strength of PA-66 is ~29-53 J/m, which is much lower than that of PA-6. PA-612 being an intrinsically ductile polymer matrix with a notched Izod impact strength value of ~55 J/m, the observed 11-fold increase in the toughness due to the addition of 10 wt% of POE-g-MA, where the dispersed phase domain size remained at ~0.30 µm, may conceptually be attributed to an equivalent super-toughening effect. 3.4 Dynamic mechanical analysis (DMA) The variation of loss modulus (E”) of PA-612 and their blends as a function of temperature is shown in Fig. 3 (c). The loss modulus curve indicated the appearance of two distinct loss-peaks at ~45 °C and at ~-42 °C corresponding to the glass transition temperatures of PA-612 and POE respectively, indicating the phase immiscibility in the blends. An increase in the energy dissipation ability of the blends as observed from the loss peak intensity in the temperature range of -60 °C to -20 °C due to POE-g-MA incorporation has been evident. The relaxation peak temperature remained unaffected, 9

whereas peak width increased with increasing POE-g-MA content. Above 10 wt% POEg-MA content the E’’ peak at -40 oC indicated the relative predominance of the soft POE phase. A singularity in viscoelastic response in temperature -20 oC was observed. Further, at ~45 ºC the E” peak intensity reduction with increase in POE-g-MA content indicated stiffening of the blends and thereby reducing the damping. Such an observation reiterates phase adhesion. 3.5 Essential work of fracture (EWF) measurements 3.5.1 Fracture behaviour of the blends Fig. 4(a) shows typical load-displacement curves of blends. The self-similar nature of load-displacement diagram is clearly indicated in Fig. 4(a), irrespective of the blend compositions. This inevitably indicates the fulfilment of the preconditions for the validity of EWF approach as proposed in the framework of post-yield fracture mechanics (PYFM) to evaluate fracture mechanics of thin plates/films, where the plain-stress condition may easily be attained. Fig. 4(b) show the net section stress (σn) versus ligament length of PA-612 (NE-0) and PA-612/POE-g-MA (NE) blends. It is observed that σn remained nearly constant (uniformly < 1.15 σy) irrespective of the compositions and thereby confirming the plain-stress criterion for the applicability of the PYFM. Quasi-linearity of the specific work of fracture as a function of ligament lengths of PA612 and PA-612/POE-g-MA blends is shown in Fig. 4(c). The slope of each linear plot gives non-essential work of fracture (βwp) while the intercept give essential work of fracture (we). These energy terms expressed as we and βwp have a conceptual correspondence to the resistance against crack initiation and propagation. The variation of

we and βwp as function of domain sizes corresponding to each blend composition is shown in Fig. 5. It is well reported that toughening is more related to the number of particles of rubber that has a direct correspondence to number average domain size (Dn) than the total volume fraction (φ) of rubber in the polymer matrix [20]. It was observed that we increases by the addition of the soft POE-g-MA phase till 10 wt% (Dn ~ 0.3 µm) whereas on further increasing the POE-g-MA content (with a corresponding increase in Dn) we values decrease linearly. A linear increase by (~ 20 %) in the resistance to crack propagation (βwp) has been registered with increase in POE-g-MA content determined Dn from ~0.2 µm to ~0.3 µm. The we and βwp values were found to be complementary to 10

each other in the compositional range of 10-35 wt% of POE-g-MA in the PA-612 based blends. 3.5.2 Correlation of essential work of fracture (we) and non-essential work of fracture (βwp) with domain size (Dn) of blends Fracture toughness remained initiation dominated (indicated by higher magnitude of we) in the critical composition domain of 5-10% i.e. volume fraction (φd) in the range of ~0.06-0.12 as evident from Fig. 5. The observations are well in agreement to the increase in notched Izod impact strength in the said regime (see Table-1 and Fig. 5). In the critical composition domain the toughness may be related to soft phase-dispersed morphology where rubber particles may act as effective stress concentrators and thereby increase the resilience of the blends altogether. The resistance to crack propagation with a correspondence to βwp is known to be dominated by the nature of the interface. It was already reported that the investigated blend systems have a poor phase adhesion parameter [14] at the µm-scale and reduced entanglement efficiency due to topological constraints imposed by the six carbon (C6) side chain of the POE-g-MA with PA-612 chain. Such decrease in entanglement density might reduce the hardness and hence may lead to an increase in Izod impact strength accordingly. However the dynamics of crack in the scenario may get abated significantly due to the inherently higher damping potential of polymer chains that is facilitated by the enhanced free volume space. The theoretical possibility of increase in free volume space is corroborated by the fact that the Tg of the NE blends showed a decrease with increase in POE-g-MA in the blends. In congruence to such a reduced thermal requirement for segmental relaxation with increase in POE-g-MA a consistent increase in βwp has been observed. Since the energy required for the crack-propagation remains leveled-off above a critical soft-phase volume fraction of φd > 0.2 and with the crack propagation phenomena being increasingly less localized, the bulk molecular relaxation process, therefore, starts to recede away from exhibiting this predominance. However, the dynamics of crack may become more controlled by the crack plane (IFPZ) than by the global fracture process zone (OPDZ) around the extending crack for the blends with ≤ 10 wt% POE-g-MA, an aspect which will be discussed in future based on the real-time visualization of strain field images during in-situ deformation of DENT specimens. 11

3.5.4 Fractured surface morphology The morphology of the sample surface at the failure region has been investigated by SEM and the micrographs are shown in Fig. 6. The micrographs clearly reveal three distinct nature of morphological features corresponding to the blends NE-0/NE-5, NE-10 and NE-20/NE-35. Their associated macroscopic deformation characteristics may be attributed to homogeneous deformation in neat PA-612 to non-homogeneous deformation in the blend compositions. In neat polyamide (NE-0) matrix stretching has been uniformly observed giving rise to a smooth surface topography. On incorporation of 5 % of POE-g-MA the surface showed twisted curls like features indicating faster postfracture relaxation process of the matrix polyamide. On further increasing the elastomeric content to 10 wt% the surface appeared uniformly stretched due to dilatation induced by uniform stress-field generation. However on further increasing the POE content to above 20 wt% the surfaces exhibited broad topographical variations due to the faster soft-phase (rubber) stretching rate than the matrix deformation.

The consequences of such

morphological features have manifested in reduction in we of blends with ≥10 wt% of POE-g-MA whereas the matrix-induced stretching effect dominated response of the blends with ≤ 5 wt% POE-g-MA is reflected in the nearly equal level of we and βwp as fracture parameters (Fig. 5). Interestingly these composition ranges correspond to the domain sizes of ≤ 0.15 to ≥ 0.25 µm. This reiterates that a transition in the fracture resistance/toughness response in PA-612/POE-g-MA blends may be attained in the domain size range of ~0.15 to 0.25 µm. These observations are further corroborated by the reported literature [14] that crazing may be a dominant mechanism for maximized toughening in PA-612/POE-g-MA blends instead of shear yielding. 3.6 Discussion on toughening of polyamides In most of the investigated rubber toughened polyamide (PA-6, PA-66, PA-1010 and PA12) based blends the critical domain size for optimal toughening remained in the range of ~0.1- 0.68 µm [10, 17, 21, 22]. However, in aromatic-polyamides the critical weight average domain size (Dw) of the rubber phase was reported to be in the range of ~0.10.85 µm [10, 20]. On the other hand, for polyester like PET the rubber toughening was reported to be achieved when the weight average domain size (Dw) remained in the range of ~0.66-3.75 µm [4]. Interestingly, in the commercially successful systems such as ABS 12

and HIPS the maxima in toughness have been reported to be in the range of ~0.1-1.0 µm and ~1.0-10 µm respectively [20]. Theoretically when two polymers are blended with polar moieties potentially capable of interacting with each other chemically, the crystalline morphology tends to get disturbed and a molecular level reorganization leading to an altered crystalline arrangement may be attained. In polar polymers like polyamide and POE-g-MA the MA-moiety may interact with the terminal -amine /carboxylic end groups of polyamide and may get chemically connected to the main chain of polyamide, on one hand, while on the other, depending on the side-chains/pendant groups the physical entanglement of the side chain of POE-g-MA and the main chains of PA-612 with each other becomes a topological reality. The thermodynamic feasibility of such a scenario leads to an enhancement in the main chain stiffness when polymer blend system is grossly evaluated for its dynamic or quasi-static stress-response. Based on the rheological investigation, as evident from the tan δ versus frequency plot (Fig. 3c, inset), it was observed that the tan δ remained nearly unaffected in the very low frequency (rad/s) range of ~0.1 that is simultaneously accompanied by a significant broadening of the tan δ peak. This indicates that the absolute G” and G’ values tend to complement each other in such a manner that the conventionally expected increase in G” (loss component) due to elastomer incorporation could not effectively offset the effect of G’ (storage component). The observation when compiled vis-à-vis relatively unaffected crystallinity (normalized by the elastomeric content) of the PA-612 reiterates the fact that the interaction between PA-612 and POE-g-MA phase remains confined to the amorphous domains of polyamide matrix chains. Thus while topological constraints reduces the extent of physical entanglements of polyamide chains unlike the virgin-matrix (due to a decrease

in

the

entropic

factors)

the

number

density

of

inter-chain

tie

points/entanglements may increase significantly due to pendant short C6 chains (from 1octene) and the POE main chain leading to an overall increase in the chain stiffness. The relationship between entanglement density and chain stiffness is well reported in literature [23]. For example, from rheological perspectives, the entanglement density (νe) may be expressed as below: νe = ρNA/Me where Me = ρRT/GN0 13

(5)

where, ρ, NA, Me, R, T, and GN0 represent density of the sample, Avogadro’s number, molecular weight between entanglement nodes, universal gas constant, reference temperature and rubber plateau modulus that is equal to the storage modulus at the frequency where tan δ is minimum in the plateau zone of the rheological master curve respectively. Since bulk hardness as a measure of resistance to indentation (i.e. intrinsic response to highly-localized deformation) correlates to entanglement density, the role of segmental stiffness in strengthening of tie-points may be construed as obvious. This may well be supported by the scaling law, reportedly valid for polyolefin homopolymers, explaining the direct correlation of entanglement density (νe) to chain stiffness (C∞) [24]. The scaling law may be given as, νe ∝ C∞ 1.4 (ρ/m0)2.2 l0 3.6

(6)

where, C∞ is a measure of the segmental stiffness, ρ is the density of the monomer, m0 is the mass of the monomer and l0 representing the contour lengths of the monomer. However, for heterogeneous systems like polymer blends the fundamental energy balance equation depicting the fracture of materials creating two new surface may be considered relevant, as has recently been discussed by Deblieck et al. [24] on issues related to the role of crazing, shear yielding and entanglement network on toughness. The equation may be stated as: Γ (γ, νe) = γ + 0.25 νe Ud

(7)

where, Γ is the energy needed to create new surface, γ is the surface energy, d is the endto-end distance between two effective entanglements and U is the energy needed to fracture a covalent chain. The increase in G’ as obtained from rheological investigations in the composition range of 5-20 wt% of POE-g-MA, which may have correspondence to significantly increased we before leveling off at 35 wt% of POE-g-MA blends, may be attributed to an increase in the entanglement density and effective strength of tie-points between POE and PA-612 chain segments. Therefore a semi-qualitative entanglement model may be proposed explaining the possible molecular level segmental entanglement scenario as shown in Fig. 7. The model depicts the entanglement of side chain segments of POE and the main chain ethylene groups of polyamide matrix while the intra/inter

14

molecular hydrogen bonding between two adjacent amide groups remain unaffected and thereby ensuring the crystallinity to remain grossly unaffected. 4. Conclusions The plane-stress fracture toughness behavior of melt-mixed injection molded PA612/POE-g-MA blends has been successfully evaluated. The super-toughening of PA-612 as assessed by work of fracture parameters were correlated to dispersed phase domain sizes corroborated by loss-moduli and damping-response interpretations. The following salient conclusions have emerged from the study. (a) The appearance of relatively broader melting endothermic peaks in blends indicated significant interference of amorphous POE-g-MA phase with that of the crystalline morphology of PA-612. The polar interaction between the two components was confirmed by FTIR spectral data. (b) POE-g-MA induced increase in the energy dissipation ability of the blends could be construed from the loss peak intensity in the temperature range of -60 °C to -20 °C though the relaxation peak temperatures remained unaffected. (c) Super-toughening of PA-612 could convincingly be attained by the incorporation of POE-g-MA that critically corresponds to a domain size of ~0.30 µm, which leads to ~1100% increase in notched Izod impact toughness. (d) The fracture (toughness) behavior remained resistance to crack initiation dominated in the critical composition domain of 5-10% of POE-g-MA in the blends as indicated from the essential work of fracture (we) whereas the increase in POE-g-MA led to consistent increase in resistance to stable crack propagation as indicated from non-essential work of fracture (βwp). (e) Semi-empirical molecular interpretation showed enhanced fracture toughness to be attributed to an overall increase in chain segmental stiffness due to topologically feasible entanglement effects and fracture surface morphology indicated crazing to be responsible for the transition in fracture behavior in the critical domain size range of ~0.15-0.30 µm (i.e. in blends with 5-10 wt% of POE-g-MA).

15

References 1. Wu S. Phase structure and adhesion in polymer blends: A criterion for rubber toughening. Polymer, 1985; 26: 1855-63. 2. Bucknall CB., Paul DR. Notched impact behaviour of polymer blends: Part 2: Dependence of critical particle size on rubber particle volume fraction. Polymer 2013; 54: 320-29. 3. Yin B, Li L, Zhoua Y, Gonga L, Yanga M, Xie B. Largely improved impact toughness of PA6/EPDM-g-MA/HDPE ternary blends:The role of coreeshell particles formed in melt processing on preventing micro-crack propagation. Polymer 2013; 54: 1938-47. 4. Su JJ, Peng F, Gao X, Yang G, Fu Q, Wang K. Superior toughness obtained via tuning the compatibility of poly(ethylene terephthalate)/poly(ethylene–octene) blends. Materials and Design 2014; 53: 673-80. 5. Ma LF, Wei XF, Zhang Q , Wang WK , Gu L , Yang W, Xie BH, Yang MB. Toughening of polyamide 6 with β-nucleated thermoplastic vulcanizates based on polypropylene/ethylene-propylene-diene rubber grafted with maleic anhydride blends. Materials and Design 2012; 33: 104-10. 6. Ma LF, Wei XF, Zhang Q , Wang WK , Gu L , Yang W, Xie BH, Yang MB. Toughening of polypropylene with β -nucleated thermoplastic vulcanizates based on polypropylene/ethylene-propylene-diene rubber blends. Materials and Design 2013; 51: 536-43.

16

7. Ozkoc G, Bayram G, Bayraml E. Impact essential work of fracture toughness of ABS/polyamide-6 blends compatibilized with olefin based copolymers. Journal of Material Science 2008; 43: 2642-52. 8. Jing B, Dai W, Chen S, Hu T, Liu P. Mechanical behavior and fracture toughness evaluation

of

K

resin

grafted

with

maleic

anhydride

compatibilized

polycarbonate/K resin blends. Materials Science and Engineering 2007; A 444: 84-91. 9. Fung KLR, KY Li. A study on the fracture characteristics of rubber toughened poly(ethylene terephthalate) blends. Polymer Testing 2005; 24: 863-72. 10. Huang JJ, Keskkula H, Paul DR. Comparison of the toughening behaviour of nylon 6 versus an amorphous polyamide using various maleated elastomers. Polymer 2006; 47: 639-51. 11. Balamurugan GP, Maiti SN. The influence of reactive compatibilization on uniaxial large strain deformation and fracture behavior of polyamide 6 and poly (ethylene-co-butyl acrylate) blends. Polymer Testing 2008; 27: 752-64. 12. Pegoretti A, Ricco T. On the essential work of fracture of neat and rubber toughened polyamide-66. Engineering Fracture Mechanics 2006; 73: 2486-502. 13. Tang XG, Bao RY,Yang W, Xie B,Yang M, Hou M. Effect of β-phase on the fracture behavior of dynamically vulcanized PP/EPDM blends studied by the essential work of fracture approach. European Polymer Journal 2009; 45: 144853.

17

14. Kumar S, Satapathy BK, Maiti SN. Correlation of morphological parameters and mechanical performance of polyamide-612/poly (ethylene-octene) elastomer blends. Polymers for Advanced Technologies 2013; 24: 511-19. 15. Jaggi HS, Kumar Y, Satapathy BK, Ray AR, Patnaik A. Analytical interpretations of

structural

and

mechanical

response

of

high

density

polyethylene/hydroxyapatite bio-composites. Materials and Design 2012; 36: 75766. 16. Dayma N, Das D , Satapathy BK. Kinetic and structural interpretations of postyield crack resistance behavior of polyamide-6/polyolefin-g-maleic anhydride blends. Journal of Material Science 2012; 47: 4860-75. 17. Barany T, Czigány T, Karger-Kocsis J. Application of the essential work of fracture (EWF) concept for polymers, related blends and composites: A review. Progress in Polymer Science 2010; 35: 1257-87. 18. Dayma N, Jaggi HS, Satapathy BK. Post-yield crack toughness behavior of polyamide-6/polypropylene

grafted

maleic

anhydride/nanoclay

ternary

nanocomposites. Materials and Design 2013; 49: 303-310. 19. Huang JJ, Paul DR. Comparison of fracture behaviour of nylon 6 versus an amorphous

polyamide

toughened

with

maleated

poly(ethylene-1-octene)

elastomers. Polymer 2006; 47: 3505-19. 20. Brydson JA. Plastics Materials. 7th ed. Butterworth Heinemann: 1999. 21. Yu H, Zhang Y, Ren W. Toughening effect of ethylene-vinyl acetate rubber on nylon 1010 compatibilized by maleated ethylene-vinyl acetate copolymers. Journal of Polymer Science: Part B: Polymer Physics, 2009; 47: 434-44. 18

22. Jose S, Thomas S, Lievana E, Kocsis JK. Morphology and mechanical properties of polyamide 12 blends with styrene/ethylene-butylene/styrene rubbers with and without maleation. Journal of Applied Polymer Science 2005; 95: 1376-87. 23. Eckstein A, Suhm J, Friedrich, Maier RD, Sassmannshausen J, Bochmann M, Mulhaupt R. Determination of Plateau Moduli and Entanglement Molecular Weights of Isotactic, Syndiotactic, and Atactic Polypropylenes Synthesized with Metallocene Catalysts. Macromolecules1998; 31: 1335-40. 24. Deblieck RAC, Beek DJMV, Remerie K, Ward IM. Failure mechanisms in polyolefines: The role of crazing, shear yielding and the entanglement network. Polymer 2011; 52: 2979-90.

19

List of Figures Fig.1. SEM micrograph of cryogenically fractured etched surfaces of PA-612/POE-gMA blends Fig.2. Schematic of injection molded DENT specimen used for EWF testing Fig.3. (a) DSC heating curves at a cooling rate of 10oC/min (b) ATR–FTIR spectra (c) Dynamic mechanical analysis plots of the blends loss modulus (E’’) versus temperature for the PA-612/POE-g-MA blends Fig.4. (a) Self-similarity of Load-displacement curves of the investigated blends for various ligament lengths (b) Hill’s analysis plot: Net section stress versus ligament length (c) Variation of specific work of fracture with ligament length for PA-612 and PA612/POE-g-MA blends Fig.5.Variation of essential work of fracture (we) and non-essential work of fracture (βwp) and impact strength with domain size (Dn) of blends Fig.6. SEM micrograph of fractured surfaces of PA-612/POE-g-MA blend Fig.7. Proposed schematic of the PA-612/POE-g-MA chain interaction and reaction between PA-612/POE-g-MA

20

Fig.1. SEM micrograph of cryogenically fractured etched surfaces of PA-612/POE-gMA blends

21

Fig. 2. Schematic of injection molded DENT specimen used for EWF testing

22

(a)

NE-35 NE-20

Heat Flow Exo UP

NE-10 NE-5 NE-0

180

190

200

210

220

230

o

Temperature ( C)

(b)

NE-35

Relative transmittance (%)

NE-20

NE-10 NE-5 NE-0

POE-g-MA

3500

3000

2500

2000

1500 -1

Wavenumber (cm )

23

1000

500

120

(c)

Tg due to POE-g-MA

NE-0 NE-5 NE-10 NE-20 NE-35

Tg due to PA-612

100

60 40 20

NE-0 NE-5 NE-10 NE-20 NE-35

1

10

0 tanδ

Loss modulus (MPa)

80

-20 -40

0

10

-1

0

10

1

10

2

10

10

Frequency (rad/s)

-80

-60

-40

-20

0

20

40

60

80

100

120

o

Temperature ( C)

Fig. 3. (a) DSC heating curves at a cooling rate of 10oC/min (b) ATR–FTIR spectra (c) Dynamic mechanical analysis plots of the blends: loss modulus (E’’) versus temperature for the PA-612/POE-g-MA blends

24

(a) 700

Increasing ligament length

NE-0

Increasing ligament length

NE-5 600

600

500 500

Load (N)

Load (N)

400 400

300

300

200

200

100

100

0

0 0

1

2

3

4

5

0

6

1

2

3

4

5

6

Displacement (mm)

Displacement (mm) 500

Increasing ligament length

NE-10

Increasing ligament length

NE-20

500 400

400

Load (N)

Load (N)

300

300

200

200

100

100

0

0

0

1

2

3

4

5

6

0

1

2

Displacement (mm)

NE-35

400

3

Increasing ligament length

Load (N)

300

200

100

0 0

1

2

3

4

5

Displacement (mm)

25

4

Displacement (mm)

6

7

8

5

6

7

180

(b) Specific work of fracture (N/mm)

70

60

σ n (N/mm 2)

50

40

30

20

10 3

4

5

6

7

8

Ligament length (mm)

9

(c)

160

140

120

100

80

60

2

3

4

5

6

7

8

9

Ligament length (mm)

Fig.4. (a) Self-similarity of Load-displacement curves of the investigated blends for various ligament lengths (b) Hill’s analysis plot: Net section stress versus ligament length (c) Variation of specific work of fracture with ligament length for PA-612 and PA612/POE-g-MA blends

26

(A) 45

(C) (B)

Remarkable enhancement in toughness

700

16 600

40 15

500

35 400 14

30

300

200

25

20

13

A

we (N/mm)

B

β wp (N/mm )

C

Impact strength(J/m)

2

100 12

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Domain size (Dn),μm

Fig.5.Variation of essential work of fracture (we) and non-essential work of fracture (βwp) and impact strength with domain size (Dn) of blends

27

Fig.6. SEM micrograph of fractured surfaces of PA-612/POE-g-MA blends

28

-

-

-

-

-

Chemical

Physical -

-

H-

C

C

O

PA-

N

C

C

-

C

C N

C

PAPhysical

PA-

C

POE-g-

H-

Chemical

Fig.7. Topological schematic of the PA-612/POE-g-MA chain interaction and proposed reaction between PA-612/POE-g-MA

29

Table-1: Morphological parameters and relative mechanical properties* of PA612/POE-g-MA blends Properties Domain size (Dn),µm

NE-0 -

Inter particle distance (IPD) (τ), µm Ductility index Normalized impact strength (Ib/Xb)/(Im/Xm) Normalized tensile modulus (Eb/Xb)/(Em/Xm) Normalized tensile strength (σb/Xm)/(σm/X m) Normalized flexural modulus (Eb/Xb)/(Em/Xm) Normalized flexural strength (σb/Xm)/(σm/X m)

NE-5

NE-10

NE-20

NE-35

0.21

0.30

0.61

0.83

-

0.22

0.20

0.19

0.08

1.1

0.45

5.1

17.5

25.6

1.0

2.8

11.0

14.0

12.3

1.0

0.96

0.72

0.67

0.51

1.0

0.99

0.84

0.74

0.56

1.0

0.98

0.75

0.69

0.52

1.0

0.98

0.82

0.76

0.51

*The standard deviation in all the evaluated mechanical properties remained < 5 % which is well within the acceptable limits. The crystallinity data however, was found to be accurate within ± 2 %.

30

Graphical abstract

31

Highlights

32

•

Super-toughening of PA-612 at a critical domain-size of ~0.30 µm of POE-g-MA

•

~11-fold toughness-enhancement in PA-612 by addition of 10 wt% of POE-g-MA

•

Maxima in essential work of fracture correspond to the critical domain size