nylon-6 blends with dispersion of clay at the interface

Accepted Manuscript Flame retardancy of polystyrene/nylon-6 blends with dispersion of clay at the interface Chang Lu , Xi-ping Gao , Dian yang , Qing-qing Cao , Xin-hui Huang , Ji-cun Liu , Yuqing Zhang PII:

S0141-3910(14)00183-9

DOI:

10.1016/j.polymdegradstab.2014.04.028

Reference:

PDST 7333

To appear in:

Polymer Degradation and Stability

Received Date: 9 December 2013 Revised Date:

10 April 2014

Accepted Date: 25 April 2014

Please cite this article as: Lu C, Gao X-p, yang D, Cao Q-q, Huang X-h, Liu J-c, Zhang Y-q, Flame retardancy of polystyrene/nylon-6 blends with dispersion of clay at the interface, Polymer Degradation and Stability (2014), doi: 10.1016/j.polymdegradstab.2014.04.028. 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.

ACCEPTED MANUSCRIPT

Flame retardancy of polystyrene/nylon-6 blends with dispersion of clay at the interface

RI PT

Chang Lu∗, Xi-ping Gao, Dian yang, Qing-qing Cao, Xin-hui Huang, Ji-cun Liu, Yu-qing Zhang

SC

Key Lab of Polymer Science and Nanotechnology, Chemical Engineering & Pharmaceutics

ABSTRACT Ammonium

polyphosphate

M AN U

School, Henan University of Science and Technology, Luoyang 471003, China

(APP)

and

clay

were

utilized

to

flame-retard

polystyrene/nylon-6 (PS/PA6) blends. The results of FTIR spectra and transmission electron

TE D

microscopy (TEM) indicated that APP and clay were exclusively dispersed in the PA6 phase. Selective localization of clay at the interface of polymer blends was achieved by the method

EP

that poly(styrene-co-maleic anhydride) (SMA) was first reacted with clay, and then blended with PA6/PS. The influences of the distribution of clay and the morphology of PS/PA6 blends

AC C

on flame retardancy were investigated. The flame retardancy was evaluated by limiting oxygen index (LOI), vertical flammability test, and cone calorimeter tests. For blends with a dispersed PA6 phase, the dispersion of clay in blends has an insignificant effect on the flame retardancy. However, in blends with a continuous PA6 phase, the flame retardancy of blends with clay dispersed at the interface was better than that of blends with clay dispersed in PA6 phase. An investigation of thermo-gravimetric (TG) analysis revealed that the thermal ∗

Corresponding author. Tel./fax: +86 379 64237053 E-mail address: [email protected] (C lu)

1

ACCEPTED MANUSCRIPT stability of blends with clay dispersed at the interface showed obvious change with blends in which clay dispersed in PA6 phase. Scanning electron microscopy (SEM) characterization showed that the dispersion of clay at the interface had the benefit of forming a compact

RI PT

residue char, which augmented the flame retardancy. A completely loose residue char observed in the blends with continuous PA6 phase in which contained clay platelets should be responsible for the deterioration of flame retardant properties. Energy dispersive spectrometry

SC

(EDS) analysis revealed that the reassembling of the clay platelets on the surface of

M AN U

intumescent char caused the formation of loose residue char. On the contrary, the dispersion of clay platelets on whole intumescent char was more favorable to the improvement of flame retardancy. Keywords

Flame retardancy; Ammonium polyphosphate; Clay; PS/PA6 blends; Dispersion;

1. Introduction

TE D

Morphology

EP

Intumescent flame retardant additives have attracted much attention from those interested

AC C

in the development of flame retardation in polymers not only because they are more environmentally friendly than traditional, halogen-containing, flame retardants but also as they have high flame retardant efficiency. On heating, intumescent flame retardant additives form a foamed cellular charred layer on the surface of the product, which decelerates heat and mass transfer between the gaseous and condensed phases

[1-3]

. A typical intumescent

formulation consists of three ingredients: an acid source (phosphates, borates), a carbonising compound (pentaerythritol), and a blowing agent (melamine, isocyanurate)[4-6]. The association of PA6 charring polymers and ammonium polyphosphate (APP) as flame 2

ACCEPTED MANUSCRIPT retardants for homopolymers has already been reported. It was shown that incorporating APP in PA6 enables fire-performance related properties of interest to be obtained. Moreover, using PA6 as a carbonisation agent in association with APP was shown to be successful in PP, ABS,

RI PT

EVA, etc [7-15]. Clay is a promising material for improving the performance of polymers against fire. Addition of clay alone in polymers mainly decreases the peak heat release rate (PHRR) as one

SC

of the flame retardant property in cone calorimeter study. However, when flame retardancy is

M AN U

evaluated using LOI and UL 94 test, clay does not show any enhancement in improving flame retardant properties [16]. Therefore, clay should be used in synergistic combinations with other traditional flame retardants to achieve better flame retardant properties. Clays in combination with APP were found capable of further improving the fire resistance of thermoplastic

TE D

polymers [17-20].

For the polymer clay nanocomposite, it was found that the dispersion of the nanoparticles in the polymer flame retarded nanocomposites was a key factor to obtain the better flame

EP

retardant properties[21, 22]. However, in the case of immiscible polymer blends, clay tends to

AC C

segregate into one of the phases or at the interface, resulting in heterogeneous dispersion[23]. The selective distribution of clay in immiscible polymer blends is mainly due to the different affinity of the clay for each component of the blends. In fact, the selective localization of clay at the interface was seldom reported due to the affinity of clay for polymer matrix more than for interface. Si and Ray

[24, 25]

found that when functionalized clays were introduced into a

polymer blend, in-situ grafts were formed, as polymer chains from the blend absorbed onto the clay surfaces. These grafts can cause the platelets to segregate to the phase interfaces. This

3

ACCEPTED MANUSCRIPT method was found successfully in the blends of PS/PMMA, PC/SAN24, and PMMA/EVA. Pack et al.[26, 27] employed the same method to obtain PS/PMMA and PC/SAN blends in which clay platelets localize at the interface of those blends. The effect of clay on the

RI PT

improvement of the flame retardancy in polymer blends and the respective homopolymers was studied. They found that the effect is particularly efficient in polymer blends due to the dispersion of clay at the interface of polymer blends.

SC

The PS/PA6 blend is a typical immiscible polymer blend and its poor fire performance

M AN U

during combustion limits its use. As far as we know, how to localize clay at the interface of PS/PA6 blend is seldom reported in public literatures. The authors have employed an innovational method that the compatibilizer was first reacted with nanoparticles and then blended with immiscible polymer blends to control carbon black (CB) or carbon nanotubes [28- 29]

. Therefore, in this paper, the association of

TE D

(CNTs) at the interface of polymer blends

clay and APP as flame retardants was used for improving the fire resistance of PS/PA6 blends. The method that poly (styrene-co-maleic anhydride) (SMA), the compatibilizer of PS and

EP

PA6, was first reacted with clay and then blended with PS/PA6 should be employed to control

AC C

clay at the interface. The influence of the distribution of clay in PS/PA6 blends on the flame retardancy was studied. The effect of morphology of PS/PA6 blends on flame retardancy was also investigated.

2. Experimental

2.1 Materials The materials used in this study were polystyrene (PG-383M, MI = 8.5 g/10 min, d =1.05

4

ACCEPTED MANUSCRIPT g/cm3) supplied by Zhenjiang Chi Mei Chemicals Co., Ltd and polyamide-6 (PA6) (33500, relative viscosity of 3.50, d = 1.14 g/cm3), supplied by Xinhui Meida-DSM Nylon Chips Co., Ltd. Ammonium polyphosphate [(NH4PO3)n, n =1500, purity level＞90 %] was supplied by

RI PT

Zhejiang Longyou Gede Chemical Factory (China). The compatibilizer employed in this study was SMA (MPC1501, Shanghai SUNNY New Technology Development Co., Ltd.). The amount of maleic anhydride in SMA was 18 wt%.The organically modified clay (purity

SC

level＞95 %), coded as DK2, was supplied by Zhejiang Fenghong Clay Products, which was

2.2 Preparation of composites

M AN U

ion-exchanged with dioctadecyl dimethyl ammonium chloride.

Based on the condition of the reaction between hydroxyl group at clay surface and maleic

TE D

anhydride in the SMA, the SMA/clay composites were prepared as follows: SMA and clay with the proportion of 20/80 (w/w) were dissolved in dimethylbenzene and then reacted at 120°C for 24 h. The reactant was deposited with acetone at room temperature and then dried

EP

in a vacuum oven at 90°C for 24 h to remove the solvent.

AC C

PA6, clay and APP were dried before blending at 85°C for 12h to remove any moisture. The composites were prepared by passing them through a co-rotating twin screw extruder with a barrel length to diameter ratio of 28 at a barrel diameter of 34 mm; the temperatures from hopper to die were 180, 225, 225, and 230°C. The screw-speed and throughput were 300 rpm and 10 kg/h, respectively. The composites were dried at 85°C for 12h and injection moulded into sheets of suitable thickness in an injection-moulding machine with a hydraulic injection pressure of 50MPa. The injection and mould temperatures were 230 and 30°C, respectively; the injection and cooling times were 5 and 20s respectively. The different blends 5

ACCEPTED MANUSCRIPT prepared are listed in Table 1.

2.3 Measurement and characterisation

RI PT

The LOI value was measured using a JF-3 instrument (Chengde, China) on sheets measuring 120 × 6 × 3 mm according to the standard oxygen index test (ASTM D2863-77). The vertical flammability test was undertaken on sheets measuring 127 × 12.7 × 3 mm

SC

according to the UL-94 test in ASTM D635-77.

The flammability of the samples was characterised by a cone calorimeter (Fire Testing

M AN U

Technology Ltd, United Kingdom) according to ISO 5660 at an incident flux of 35 kW/m2 with a cone-shaped heater. All sample plates, with dimensions of 100 ×100 × 3 mm, were placed in aluminium foil and then in the frame’s sample holder with the same dimensions in

TE D

the horizontal position.

FTIR spectra were recorded on a Bruker Vector 33 spectrometer. PS and SMA were prepared into film for testing. PS/PA6/APP, SMA/clay composites and clay were pressed into

EP

disks with KBr. Before pressed, the samples of PS/PA6/APP or SMA/clay were extracted by

AC C

formic acid or acetone for 48 h, respectively. XRD spectra of clay and composites were obtained with a X’Perl MPD Pro instrument

using Cu Ka radiation (λ=1.54Å) to determine d-spacing between the clay layers. The voltage and current of the X-ray tube were 40 kV and 40 mA, respectively. Basal spacing was estimated from the position of the (001) peak in the XRD pattern. Transmission electron microscopy (TEM) method was used to examine the localization of clay in blends by a FEI TECNAI-G20 microscope at an acceleration voltage of 300 kV. Ultra thin sections of 70 nm in thickness were cryogenically cut with a diamond knife at -100°C. 6

ACCEPTED MANUSCRIPT Thermo-gravimetric analysis (TGA) was carried out at a heating rate of 20°C/min under a nitrogen flow of 50ml/min by a thermo-gravimetric analyser STA409PC (NETZSCH, Germany).

RI PT

The morphology of the residual charred layer obtained after cone calorimeter test, and the fractured surface of the specimen, which were coated with a conductive gold layer, was examined using a JEOL 6301F scanning electron microscope (SEM). The moulded specimens

SC

were fractured in liquid nitrogen. To obtain better contrast, the fractured surfaces of the

M AN U

specimens were etched by formic acid or dimethylbenzene before coating. Energy dispersive spectrometry (EDS) analysis of the residual charred layer obtained after cone calorimeter test was done by JEOL JSM-5900LV EDS analyzer.

TE D

3． ．Results and discussion

3.1 SMA/clay composites

XRD spectra of clay and SMA/clay composite are shown in Figure 1. The XRD spectrum

EP

of clay gave an intense peak at an angle 2θ=3.5 and a weak peak at 2θ value of 6.8o,

AC C

corresponding to the interlayer spacing (d) of 2.5 and 1.3 nm, respectively. The diffraction peaks of SMA/clay were appeared at 2θ value of 2.9o, corresponding to the interlayer spacing (d) of 3.05 nm. The results showed that SMA/clay was intercalation nanocomposite. Figure 2 shows the FTIR spectra of remainder of SMA/clay extracted by acetone, SMA and clay. The bands at 1858 and 1782 cm−1 are the characteristic bands of SMA, which corresponds to the carbonyl absorption of anhydride groups in the five-membered rings. In the spectrum of clay, the characteristic peak of Si-O stretching mode of clay was observed at

7

ACCEPTED MANUSCRIPT absorption bands of 1060-1030 cm-1. The strong absorption band was observed at 3624 cm-1 corresponding to the vibration of –OH bond. The absorption bands at 2851 and 2920 cm-1 were assigned to the –CH2– stretching vibration of organic modifier. In the spectrum of

RI PT

SMA/clay, the characteristic bands of SMA were observed. Moreover, the additional weak absorption peak at 1734 cm-1 assigned to the ester carbonyl groups was found in the spectra. This observation indicated that the esterification reaction between the oxhydryl at clay surface

SC

and anhydride groups in SMA has occurred. Before using FTIR to measure, clay modified

M AN U

with SMA was extracted for 48h by acetone to remove SMA of which has not reacted with clay. XRD spectra showed that SMA molecular chain intercalated into the clay layers. Therefore, the appearance of SMA characteristic peak in composites should be due to the SMA remainder in the clay layers.

TE D

3.2 Dispersion of APP and clay in PS/PA6 blends FTIR spectra were used to investigate the dispersion of APP in PS/PA6 blends. The blind

EP

experiments were carried out to test the dissolubility of PA6 and APP in formic acid or

AC C

dimethylbenzene. The results showed that PA6 and APP were readily dissolved in formic acid and could not be dissolved in dimethylbenzene. So formic acid was chosen as the extraction solvent to remove PA6 and APP. Figure 3 shows the FTIR spectra of pure PS and remainder of the PS/PA6/APP (12/68/20) extracted by formic acid. The 1451 to 1601 cm-1 spectral region contains the characteristic absorption peaks of PS. The peaks at 699 and 756 cm-1 were assigned to the bending vibration of benzene. The peaks at 2851 and 2924 cm-1 were assigned to the –CH2- stretching vibration. The peaks at 3026 and 3060 cm-1 were assigned to the stretching vibration of aromatics. The FTIR spectrum of the remainder was similar to PS and 8

ACCEPTED MANUSCRIPT almost no characteristic peaks of APP and PA6 were found therein. In the composite of PS/PA6/APP (12/68/20), the PA6 phase was continuous because the PA6 content was much higher than that of PS. The continuous PA6 phase was readily dissolved in formic acid,

RI PT

meaning that the PA6 was undetected in the remaining material. This phenomenon whereby APP cannot be detected in the residual material indicated that APP localizes in the PA6 phase. The dispersion of APP in the PA6 phase showed that the affinity of APP is higher for PA6

SC

than for PS.

M AN U

The dispersion of clay in PA56-clay or PA56-(SMA/clay) is shown in Figure 4. It was clearly observed that clay exhibited heterogeneous dispersion in PS/PA6/APP/clay blends. The clays were segregated primarily into continuous PA6 phase and no clay was observed in dispersed PS domains (Figure 4a). In the PS/PA6/APP/(SMA-clay) blends, clay platelets were

TE D

localized at the interface between the PS and PA6 phases (Figure 4b). Combining with Figure 3 and 4, it can be conclusion that APP and clay platelets were both localized in PA6 phase of PS/PA6/APP/clay blends, but in PS/PA6/APP/(SMA-clay), clay platelets were localized at

EP

the interface and APP dispersed in PA6 phase.

AC C

The localization of clay platelets at the interface of PS/PA6 can be attributed to the induced effect of SMA on clay platelets. SMA copolymer, as the compatibilizer of PS/PA6 blends, should congregate at the interface of PS/PA6 blends due to the fact that the compatibilizer will selectively localize at the interface of immiscible polymer blends to reduce the interfacial tension. In this paper, SMA molecular chains had grafted on the surface of clay platelets before being introduced into the PS/PA6 blends. The driving forces originated from SMA molecular chains can induce clay platelets to aggregate at the interface.

9

ACCEPTED MANUSCRIPT In addition, SMA molecular chains intercalated into the laminates of clay can also produce driving forces. Therefore, the clay platelets can be induced to localize at the interface by SMA spontaneously during the processing.

RI PT

3.3 Structural properties of clay in composites and their morphology

XRD spectra of clay, PS/PA6/APP/clay and PS/PA6/APP/(SMA-clay) are shown in

SC

Figure 5. The intense diffraction peaks of clay were absent up to the lowest measurable angle (i.e. 2θ = 2o) in the diffraction spectrum of PA24-clay and the weak peak decreased to 6.2 o,

M AN U

corresponding to the interlayer spacing (d) of 1.4 nm. In the other samples, the diffraction peaks of clay were all vanished. Diffraction peaks in the low angle region indicate the d-spacing (basal spacing) of ordered intercalated and ordered delaminated nanocomposites. Disordered nanocomposites show no peak in this region due to loss of structural registry of [16]

TE D

the layers and/or the large d-spacing

. Therefore, the absence of intense peak and the

reduction in diffraction angle indicated that ordered exfoliation has taken place in the sample

EP

of PA24-clay. The absence of diffraction peaks in the other samples provided an indication of

AC C

disruption in the coherent layer stacking and of formation of exfoliated structures. TEM showed that ordered exfoliation was occurred in PA32-clay (Figure 6a) and disordered exfoliation was occurred in PA56-clay (Figure 6b) and PA56-(SMA/clay) (Figure 4b), which were consistent with previous results of XRD. SEM was used to investigate the morphology of blends, as shown in Figure 7. To enhance the contrast, samples with PA6 content of 24wt % were etched with formic acid to remove the (PA6+APP) phase and dimethylbenzene was used to etch the PS phase for PA6 contents of 56wt %. Figure 7 (a, b, c) showed that blends with 56wt% PA6 content have a sea-island 10

ACCEPTED MANUSCRIPT morphology in which spherical PS particles were dispersed in a continuous (PA6+APP) phase. When the PA6 content was 24wt %, the micrograph showed a sea-island morphology in which (PA6+APP) was the spherical dispersed phase, as shown in Figure 7 (d, e, f). It can be

RI PT

observed that dispersed domain size of PS/PA6/APP/(SMA-clay) was smaller than that of PS/PA6/APP and PS/PA6/APP/(SMA-clay). This result was due to the compatibilization of

SC

SMA-clay dispersed at the interface.

3.4 Flame retardancy

M AN U

The LOI values and UL-94 rating data for blends are shown in Table 2. In the samples with the PA6 content of 24wt%, the samples of PS/PA6/APP had the lowest LOI value and the samples failed the vertical UL-94 test. The clay resulted in the increase of flame retardant

TE D

property. LOI values jumped from 27.4 to 31.5 (PA24-clay) or 30.9 (PA24-(SMA/clay)). Meanwhile, flame retardant property was able to reach UL-94-V1 grade. In comparing with the flammability characteristics of PA24-(SMA/clay) and PA24-(SMA/clay), it can be

EP

observed that the dispersion of clay in blends has an insignificant effect on the flame

AC C

retardancy when the PA phase formed dispersed phase. For samples with 56wt% PA6 content, the LOI values of PA56-clay were close to PA56 and both of the blends failed the vertical UL-94 test. However, for samples of PA56-(SMA/clay), a sharp increase of LOI values was observed and the samples reached UL-94-V1 grade, indicating a remarkably better flame retardancy than that of PA56-clay. The main difference between blends of PS/PA6/APP/clay and PS/PA6/APP/(SMA-clay) was the dispersion of clay in blends except the presence of 1wt% SMA. The LOI value showed that introduction of 1wt% SMA to the blends of PS/PA6/APP/clay had a weak 11

ACCEPTED MANUSCRIPT influence on their flame retardancy. Therefore, it can be conclusive that the dispersion of clay platelets at the interface was more favorable to improved flame retardancy than that of blends with the dispersion of clay platelets in PA6 phase, when PA6 phase formed a continuous

RI PT

structure. The cone calorimeter was used to study the flammability properties of PS/PA6/APP, PS/PA6/APP/clay and PS/PA6/APP/(SMA-clay). A comparison of the HRR data for the

SC

samples with PA6 content of 24wt% is shown in Figure 8. HRR confirmed the good flame

M AN U

retardant property of clay in PS/PA6 blends: PHRR values of PA32-clay and PA32-(SMA/clay) were decreased about 60% and the time to ignition was improved. The advantage of clay appeared in teams of reduction of PHRR, which was in accordance with the results reported in literature

[30, 31]

. The PHRR of PA32-clay slightly lower than that of

TE D

PA32-(SMA/clay) indicated that PA32-clay exhibited the better flame retardancy, which corresponded with the results from LOI. The HRR curves for the samples with PA6 content of 56wt% is shown in Figure 9. Introduction of clay to blends decreased PHRR values about

EP

30% and the time to ignition was improved, which was similar to the results obtained from

AC C

the blends with PA6 content of 24wt%. However, HRR results shown in Figure 9 were inconsistent with the results of LOI and

UL-94 test. Although HRR values of PA56-clay were lower than that of PA56, their LOI values were very close. Moreover, the HRR values of PA56-(SMA/clay) were slightly higher than that of PA56-clay even though the results of LOI and vertical flammability test showed the flame retardancy of PA56-(SMA/clay) remarkably better than that of PA56-clay. The most important contribution of clay to polymer’s flame retardancy was the reduction in PHRR.

12

ACCEPTED MANUSCRIPT However, flame retardant effect of clay vanished when the fire scenario of concern was changed from a developing fire (replicated in the cone calorimeter) to an ignition scenario, like in UL 94 or LOI test

[32-34]

. These phenomena indicated that the intumescent chars,

RI PT

formed in the blends with the localization of clay at interface, were more efficient in improving flame retardancy at an ignition scenario than the intumescent chars formed in the blends with dispersion of clay in PA6 phase.

SC

XRD and TEM results showed that clay formed exfoliated structures in PS/PA6/APP/clay

in

PS/PA6

blends:

the

clay

M AN U

and PS/PA6/APP/(SMA-clay). The main difference in those blends was the dispersion of clay platelets

were

aggregated

at

the

interface

of

PS/PA6/APP/(SMA-clay) or in PA6 phase of PS/PA6/APP/clay. The dispersion of clay in blends had an insignificant effect on flame retardant property when PA phase formed a

TE D

dispersed phase. However, when the PA phase formed a continuous phase, the better flame retardancy of PA56-(SMA/clay) than that of PA56-clay indicated that the dispersion of clay platelets at the interface was benefited to improved flame retardancy. Therefore, the

EP

dispersion of clay at the interface may be a key factor to improve the flame retardancy. In

AC C

addition, the improved effect of clay dispersion at the interface on flame retardancy was depending on the morphology of PS/PA6 blends.

3.5 Characterizations of residue char In order to understand the relationship between the microstructure of intumescent chars and flame retardant properties of composites, the intumescent char residue of outer layer and inner layer after cone calorimeter tests was imaged by SEM, as shown in Figure 10 and 11, respectively. As shown in Figure 10 (a, b, c), samples with dispersed PA6 phase produced a 13

ACCEPTED MANUSCRIPT continuous intumescent chars on the surface, which could act as an insulating barrier to prevent oxygen and feedback of heat from reaching the underlying material. A relatively loose structure on the surface was observed in samples of PA24 and PA24-clay, however, the

RI PT

residue char of PA24-(SMA/clay) was more homogenous and compact. For blends with continuous PA6 phase, the intumescent chars of PA56 and PA56-clay were completely loose structure, as shown in Figure 10 (d, e). Meanwhile, a tight and compact intumescent char

SC

including some small holes on the surface of PA24-(SMA/clay) was observed (Figure 10 (f)).

M AN U

The results suggested that the dispersion of clay at the interface has the benefit of forming a compact residue char.

In Figure 11, it was observed that intumescent char morphology of PA24-(SMA/clay) at the inner surface was different from other samples. The pore size in the char layer of

TE D

PA24-(SMA/clay) was the smallest (Figure 11(c)), corresponding to the dense structure, while the structures of the char layers in other composites were loose and large porosity. The formation of pores in the char layer was due to the pyrolysis gas and NH3 volatilization in the

EP

interior of the material, which made the mixture of the carbonaceous residue and

AC C

phosphocarbonaceous materials to swell. A macroporous and loose structure should have a low thermal conductivity and high temperature gradient in the char layer. So the heat transfer towards the interior of the material was reduced, which in turn slowed down the decomposition of the composite, resulting in an improvement of flame retardancy. So the dense and small pores observed in PA24-(SMA/clay) should lead to a negative effect in the fire properties. In figure 10(c) it was observed that a homogenous and compact residue char was formed on the outer surface of PA24-(SMA/clay). Thus, the compact residue char should

14

ACCEPTED MANUSCRIPT hinder its expanding, leading to the formation of dense and small pores in the inner surface of residue char. The aspect of the crust of samples after the cone calorimeter test is showed in Figure 12.

RI PT

Height of intumescent residue char of samples PA24-(SMA/clay) was remarkably lower than that of samples else. The results also confirmed that the residue char formed in PA24-(SMA/clay) produced the smallest swell, which was in accordance with the results of

SC

SEM.

M AN U

Lewin[35, 36] found that an ablative reassembling of the clay platelets may occur on the surface of the burning nanocomposite creating a physical protective barrier on the surface of material. In this paper, the energy dispersive spectrometry (EDS) was employed to characterize the dispersion of clay in final char residue after the cone calorimeter test. The

TE D

element of clay in the outer and inner char surface is listed in Table 3. It was found that silicon and aluminum element almost did not exist in the inner char surface of PA56-clay while they were enriched on the outer surface, indicating that reassembling of the clay

EP

platelets occurred on the outer surface of the burning PA56-clay samples. Moreover, the

AC C

percentage of phosphorus element in inner char surface of PA56-clay was much less than in outer char surface. APP can react with clay to form an aluminophosphate structure and a ceramic-like structure[7, 8, 11]. Therefore, the very low content of phosphorus element in inner char surface of PA56-clay can be attributed to the reassembling of the clay platelets on the outer surface. The accumulation of clay only on the outer surface did not occur in other composites: Silicon and aluminum element were both existed in the inner and outer char surface. The

15

ACCEPTED MANUSCRIPT intumescent char which was produced by the decomposition of APP and reaction between APP and PA6 during combustion can be reinforced by clay platelets, creating an excellent physical barrier which protected the substrate from heat and oxygen, and slowed down the

RI PT

escape of flammable volatiles generated during polymer degradation. The existence of clay in the outer and inner char surface indicated that the whole intumescent char can be reinforced by clay platelets. Meanwhile, only the outer char surface in PA56-clay was reinforced by clay

SC

platelets due to the reassembling of the clay platelets on the outer surface. Based on the fact

M AN U

that the flame retardancy of PA24-clay, PA24-(SMA/clay) and PA56-(SMA/clay) were better than that of PA56-clay, it was suggested that the reinforcement of clay platelets on whole intumescent char was more favorable to the improvement of flame retardancy than on the surface intumescent char. A completely loose structure on the surface of PA56-clay (Figure

TE D

10 (e)) also indicated that the reinforced surface intumescent char hardly retarded the diffusion of pyrolysis gas from the interior of the material to the environment, resulting in the rupture of pores.

EP

Comparing with PA24-clay, it was found that the content of silicon and aluminum element

AC C

in PA24-(SMA/clay) was higher on the outer surface and lower on the inner surface, respectively. The results indicated that the concentration of clay on the outer surface of PA24-(SMA/clay) was higher than that of PA24-clay. The higher concentration of clay on the outer surface of PA24-(SMA/clay) indicated that the reinforcement of clay platelets on intumescent char was stronger than that of PA24-clay, resulting in a homogenous and compact residue char formed on the outer surface. In order to obtain clay particle structure in the residues, XRD measurements were

16

ACCEPTED MANUSCRIPT conducted for the intumescent char residue, as shown in Figure 13. The absence of diffraction peaks in the char residues showed that clay platelets formed disordered exfoliated structures. XRD spectra in Figure 5 had shown that clay platelets formed ordered exfoliation in

RI PT

PA24-clay. Therefore, the formation of disordered exfoliated structures of clay platelets in the char residues of PA24-clay indicated that the thermal degradation of organic materials trapped in the space between the clay platelets can broaden the spacing of clay platelets in the

SC

char residues. Disordered exfoliated structures of clay platelets in the char residues indicated

3.6 Thermo-gravimetric (TG) analysis

M AN U

that the influence of clay structure in intumescent char on flame retardancy can be ignored.

The DTA and TG curves of pure PS, PA6 and their blends are shown in Figures 14~16. In

TE D

the DTA plot of blends (Figure 14) two main exothermic bands were observed, the first one between 280~380°C, while the second step occurring in the 380~480°C temperature range. The TG curves of blends (Figures 15, 16) also showed two significant changes in their slopes,

EP

which also proved that their degradation was at least a two-step process. It can be seen that

AC C

the thermal degradation of PS or PA6 began at approximately 390 °C or 375 °C, respectively. So it can be said that the thermal degradation of APP and the reaction between APP and PA6 occurred at the first step. Levchik et al.[37, 38] have reported that the reaction between APP and PA6 began at around 300 °C. APP catalyzed the degradation of PA6 and interacts with it to form 5-amidopenthyl polyphosphate. Decomposition of 5-amidopenthyl polyphosphate by further heating liberates polyphosphoric acid and produces the char. Meanwhile, APP can also react with clay to form an aluminophosphate structure and a ceramic-like structure in the 310~560 °C temperature range

[7, 8, 11]

. Obviously, intumescent material which acted as a 17

ACCEPTED MANUSCRIPT protective shield was formed at the first step, which was an essential step for flame retardancy. TG curves of the blends showed that the second step began at approximately 380 °C. The TG curves of PS and PA6 also showed that the thermal degradation of PS and PA6 began at

RI PT

around 380 °C. Therefore, the weight loss at the second step should be mainly assigned to the degradation of PS and PA6.

From Figures 15, 16 it can be observed that the TG curves of PS/PA6/APP and

SC

PS/PA6/APP/clay were similar and different from PS/PA6/APP/(SMA-clay). For blends with

M AN U

dispersed PA6 morphology, in which PA6 content was 24wt%, PA24-(SMA/clay) exhibited a lower degradation rate at the first step and a higher degradation rate at the second step (Figure 15). However, in blends with continuous PA6 phase, the degradation rate of PA56-(SMA/clay) was higher at the first step and lower at the second step (Figure 16). Moreover, the highest

TE D

intumescent char residue in PA56-(SMA/clay) was exhibited. The lower degradation rate at the first step indicated a faster formation of intumescent shield, which was more effective at delaying thermal degradation. Correspondingly, the lower degradation rate at the second step

EP

suggested that the intumescent char residue formed at the first step exhibited higher efficiency

AC C

at delaying thermal degradation of PS and PA6. For the blends with 24wt% PA6, the characterizations of residue char showed that the

abundant aggregation of clay platelets at the interface was benefited to the formation of homogenous and compact residue char, which was more effective at delaying thermal degradation. Meanwhile, a relatively loose intumescent char was formed in the outer surface of PA24-clay. So the thermal stability of PA24-(SMA/clay) was better than that of PA24-clay at the first step. However, the residue char of the outer surface in PA24-(SMA/clay) was so

18

ACCEPTED MANUSCRIPT dense that its expanding was hindered, leading to a negative effect in the fire properties at the second step. On the contrary, in samples of PA24-clay, a loose and large porosity was formed in the inner surface of intumescent char, which was helpful to impede the heat transfer to the

RI PT

interior of the material and lower the temperature of the decomposition zone. As a result, the higher degradation rate of PA24-(SMA/clay) at the second step than that of PA24-clay was exhibited.

SC

Therefore, in blends with a dispersed PA6 phase, the phenomenon that dispersion of clay

M AN U

at the interface has an insignificant effect on the improvement of flame retardancy should be due to the positive and negative effect of dispersion of clay at the interface on flame retardancy. Abundant aggregation of clay platelets at the interface was benefited to a faster formation of a protective shield on the surface of PA6 phase, which produced a positive effect

TE D

on the fire properties. Meanwhile, abundant aggregation of clay platelets at the interface caused that the residue char was so dense that the expanding of residue char was hindered by clay platelets, leading to a negative effect in the fire properties.

EP

In samples of PA56-clay, clay and APP were both dispersed in PA6 phase and PA6 phase

AC C

formed continuous morphology. During combustion, decomposition of APP formed intumescent char and clay platelets dispersed in PA6 phase were reassembled on the intumescent char surface, which was confirmed by EDS results. Therefore, the surface of intumescent char was reinforced by clay platelets, creating a physical barrier which can reduce the degradation rate. However, for composites of PA56-(SMA/clay), abundant clay platelets were aggregated at the surface of dispersed PS phase, while APP was dispersed in PA6 phase. When the samples began to burn, the reinforcement of clay platelets on

19

ACCEPTED MANUSCRIPT intumescent char surface was hard to occur due to the different distribution position of clay platelets and APP. As a result, the degradation rate of PA56-(SMA/clay) was higher than that of PA56-clay at the first step. As the burning duration was extended, the enrichment of clay at

RI PT

the interface should reinforce whole intumescent char, resulting in the formation of a tight and compact intumescent chars, as shown in Figure 10 (f). A compact intumescent chars can protect the substrate from heat and oxygen, and slow down the escape of flammable volatiles.

SC

Meanwhile, the aggregation of clay platelets at the surface of dispersed PS phase can also

M AN U

provide a very efficient effect to delay the decomposition of PS. Correspondingly, the degradation rate at the second step was reduced and the highest intumescent char residue in PA56-(SMA/clay) was exhibited. For the composites of PA56-clay, as the burning duration was extended, the accumulation of pyrolysis gas in the interior of the material should cause

TE D

the rupture of the superficial intumescent char, resulting in the diffusion of pyrolysis gas form the bulk to the flame, even the surface of intumescent char was reinforced by clay platelets. As a result, the degradation at the second step was accelerated. A completely loose

AC C

conclusion.

EP

intumescent char on the outer surface of PA56-clay (Figure 10 (e)) also confirmed this

Therefore, in the blends with a continuous PA6 phase, the flame retardancy of

PA56-(SMA/clay) was better than that of PA56-clay should be that the dispersion of clay at interface has a significant effect on the formation of a tight and compact intumescent chars.

4. Conclusions APP and clay were utilized to flame-retard PS/PA6 blends. In blends of PS/PA6/APP/clay, APP and clay were selectively dispersed in the PA6 phase during melt mixing. Selective 20

ACCEPTED MANUSCRIPT localization of clay at the interface of polymer blends can be achieved by the method that SMA was first reacted with clay, and then blended with PA6/PS. The influences of the distribution of clay in PS/PA6 blends and the morphology on flame retardancy were

RI PT

investigated. In blends with a dispersed PA6 phase and the dispersion of clay at interface, the aggregation of clay platelets at interface was benefited to the formation of compact residue char in outer and inner surface. Although a compact residue char exhibited a positive effect

SC

on flame retardancy, the expanding of intumescent char was also hindered, leading to a

M AN U

negative effect in the fire properties. As a result, the dispersion of clay at the interface has an insignificant effect on the improvement of flame retardancy. In the blends with a continuous PA6 phase and the dispersion of clay at interface, the aggregation of clay platelets at interface was benefited to the formation of compact residue char in outer surface and loose and large

TE D

porosity in inner surface, which were more effective at delaying thermal degradation, resulting in the improvement of thermal stability. But in blends with continuous PA6 phase in which contained clay platelets, a completely loose residue char was formed due to the

EP

reassembling of the clay platelets on the surface of intumescent char, which was responsible

AC C

for the deterioration of flame retardant properties. Accordingly, the flame retardancy of blends with clay dispersed at the interface was better than that of blends with clay dispersed in PA6 phase when PA6 phase formed continuous state. Acknowledgements

The authors gratefully acknowledge the financial support of this work by the National Natural Science Foundation of China (Contract Number: 51003024).

21

ACCEPTED MANUSCRIPT References [1] Bourbigot S, Le BM, Delobel R. Fire degradation of an intumescent flame retardant polypropylene

using the cone calorimeter. J Fire Sci 1995; 13: 3-22.

RI PT

[2] Bourbigot S, Le BM, Siat CI, Lewin M, editors. Recent advances in flame retardancy of polymeric

materials, vol. 7. Norwalk: BCC Pub, 1997. p. 146.

[3] Wang XY, Li Y, Liao WW, Gu J, Li D. A new intumescent flame-retardant: preparation, surface

R, Le

BM, Ouassou N,

polyphosphate-Pentaerythritol

and

Alistiqsa F. Thermal behaviours of ammonium

M AN U

[4] Delobel

SC

modification, and its application in polypropylene. Polym Adv Technol 2008; 19: 1055-1061.

ammonium

pyrophosphate-pentaerythritol

intumescent

additives in polypropylene formulations. J Fire Sci 1990; 8(2): 85-108, [5] Almeras X, Renaut N, Jama C, Le BM, Tóth A, Bourbigot S, Marosi G, Poutch F. Structure and

TE D

morphology of an intumescent polypropylene blend. J Appl Polym Sci 2004; 93(1): 402-411. [6] Chiu SH, Wang WK. Dynamic flame retardancy of polypropylene filled with ammonium

polyphosphate, pentaerythritol and melamine additives. Polymer 1998; 39(10): 1951-1955.

EP

[7] Bourbigot S, Le BM, Dabrowski F, Gilman JW, Kashiwagi T. PA-6 clay nanocomposite hybrid as

AC C

char forming agent in intumescent formulations. Fire Mater 2000; 24(4):201-208. [8] Dabrowski F, LeBras M, Cartier L, Bourbigot S. The use of clay in an EVA-based intumescent

formulation. Comparison with the intumescent formulation using polyamide-6 clay nanocomposite as carbonisation agent. J Fire Sci 2001; 19(3): 219-241.

[9] Almeras X, Bras ML, Hornsby P, Bourbigot S, Marosi G, Keszei S, Poutch F. Effect of fillers on

the fire retardancy of intumescent polypropylene compounds. Polym Degrad Stab 2003; 82: 325-331.

22

ACCEPTED MANUSCRIPT [10] Riva A, Camino G, Fomperie L, Amigoue P. Fire retardant mechanism in intumescent ethylene

vinyl acetate compositions. Polym Degrad Stab 2003; 82: 341-346. [11] Tang Y, Hu Y, Song L, Zong RW, Gui Z, Fan WC. Preparation and combustion properties of

RI PT

flame retarded polypropylene/polyamide-6 alloys. Polym Degrad Stab 2006; 91: 234-241. [12] Zhang Q, Chen YH. Synergistic effects of ammonium polyphosphate/melamine intumescent

system with macromolecular char former in flame-retarding polyoxymethylene. J Polym Res 2011;

SC

18: 293-303.

M AN U

[13] Dahiya JB, Muller-Hagedorn M, Bockhorn H, Kandola. BK. Synthesis and thermal behaviour of

polyamide 6/bentonite/ammonium polyphosphate composites. Polym Degrad Stab 2008; 93: 2038-2041.

[14] Almeras X, Dabrowski F, Bras Le M, Poutch F, Bourbigot S, Marosi G, Anna P. Using

TE D

polyamide-6 as charring agent in intumescent polypropylene formulations. (I) Effect of the compatibilising agent on the fire retardancy performance. Polym Degrad Stab 2002; 77: 305-313. [15] Almeras X, Dabrowski F, Bras Le M, Delobel R, Bourbigot S, Marosi G, Anna P. Using

EP

polyamide-6 as charring agent in intumescent polypropylene formulations. (II) Thermal

AC C

degradation. Polym Degrad Stab 2002; 77: 315-323. [16] Dahiya JB, Rathi S, Bockhorn H, Haußmann M, Kandola BK. The combined effect of organic

phoshphinate/ammonium polyphosphate and pentaerythritol on thermal and fire properties of polyamide 6-clay nanocomposites. Polym Degrad Stab 2012; 97: 1458-1465.

[17] Bourbigot S, Le BM, Duquesne S, Rochery M. Recent advances for intumescent polymers.

Macromol Mater Eng 2004; 289: 499-511. [18] Ma ZL, Zhang WY, Liu XY. Using PA6 as a charring agent in intumescent polypropylene

23

ACCEPTED MANUSCRIPT formulations based on carboxylated polypropylene compatibilizer and nano-montmorillonite synergistic agent. J Appl Polym Sci 2006; 101: 739-746. [19] Tang Y, Hu Y, Li B, Liu L, Wang Z, Chen Z. Polypropylene/montmorillonite nanocomposites and

RI PT

intumescent, flameretardant montmorillonite synergism in polypropylene nanocomposites. J Polym Sci A Polym Chem 2004; 42: 6163-6173.

[20] Zhang S, Horrocks AR, Hull R, Kandola BK. Flammability, degradation and structural

SC

characterization of fibre-forming polypropylene containing nanoclay-flame retardant combinations.

M AN U

Polym Degrad Stab 2006; 91:719-725.

[21] Bourbigot S, Samyn F, Turf T, Duquesne S. Nanomorphology and reaction to fire of polyurethane

and polyamide nanocomposites containing flame retardants. Polym Degrad Stab 2010; 95(3): 320-326.

TE D

[22] Bourbigot S, Turf T, Bellayer S, Duquesne S. Polyhedral oligomeric silsesquioxane as flame

retardant for thermoplastic polyurethane. Polym Degrad Stab 2009; 94(8): 1230-1237. [23] Zou H, Zhang Q, Tan H, Wang K, Du RN, Fu Q. Clay locked phase morphology in the

EP

PPS/PA66/clay blends during compounding in an internal mixer. Polymer 2006; 47: 6-11.

AC C

[24] Si M, Araki T, Ade H, Kilcoyne ALD, Fisher R, Sokolov JC, et al. Compatibilizing bulk polymer

blends by using organoclays. Macromolecules 2006; 39: 4793-4801.

[25] Ray SS, Pouliot S, Bousmina M, Utracki LA. Role of organically modified layered silicate as an

active interfacial modifier in immiscible polystyrene/polypropylene blends. Polymer 2004; 45 (25): 8403-8413.

[26] Pack S, Si MY, Koo J, Jonathan CS, Koga T, Kashiwagi T, Miriam HR. Mode-of-action of

self-extinguishing polymer blends containing organoclays. Polym Degrad Stab 2009; 94: 306-326.

24

ACCEPTED MANUSCRIPT [27] Pack S, Kashiwagi T, Stemp D, Koo J, Si MY, Jonathan CS, Miriam HR. Segregation of Carbon

Nanotubes/Organoclays Rendering Polymer Blends Self-Extinguishing. Macromolecules 2009; 42: 6698-6709

RI PT

[28] Lu C, Cao QQ, Huang XH, Hu XN, He YX, Liu CY, Zhang YQ. Influence of morphology on

positive temperature coefficient effect for conductive polymer composites with carbon black dispersed at interface. Polym Engin Sci 2013; 53(12): 2640-2649.

SC

[29] Lu C, Wang R, Huang XH, Cao QQ, Hu XN, He YX, Zhang YQ. Influence of morphology on

M AN U

PTC effect for poly (ethylene-co-butyl acrylate)/nylon6 blends with multiwall carbon nanotubes dispersed at interface and in matrix. Polym Bull DOI: 10.1007/s00289-013-1076-z. [30] Kiliaris P, Papaspyrides CD. Polymer/layered silicate (clay) nanocomposites: An overview of

flame retardancy. Prog Polymer Sci 2010; 35: 902-958.

TE D

[31] Kashiwagi T, Richard H, Zhang X, Briber RM, Cipriano BH, Raghavan SR, Awad W H, Shields

JR. Flame retardant mechanism of polyamide 6–clay nanocomposite. Polymer 2004; 45: 881–891. [32] Schartel B, Bartholmai M, Knoll U. Some comments on the main fire retardancy mechanisms in

EP

polymer nanocomposites. Polym Adv Technol 2006; 17: 772-777.

AC C

[33] Cui W, Guo F, Chen JF. Preparation and properties of flame retardant high impact polystyrene.

Fire Saf J 2007; 42: 232-239.

[34] Si M, Zaitsev V, Goldman M, Frenkel A, Peiffer DG, Weil E, et al. Selfextinguishing

polymer/organoclay nanocomposites. Polym Degrad Stab 2007; 92: 86-93.

[35] Lewin M. Some comments on the modes of action of nanocomposites in the flame retardancy of

polymers. Fire Mater 2003; 27: 1-7. [36] Lewin M. Reflections on migration of clay and structural changes in nanocomposites. Polym Adv

25

ACCEPTED MANUSCRIPT Technol 2006; 17: 758-763. [37] Levchik SV, Costa L, Camino G. Effect of the fire-retardant, ammonium polyphosphate, on the

thermal decomposition of aliphatic polyamides: Part II—polyamide 6. Polym Degrad Stab 1992;

RI PT

36(3): 229-237. [38] Levchik SV, Levchik GF, Balabanovich AI and Caminob G. Mechanistic study of combustion

AC C

EP

TE D

M AN U

retardants. Polym Degrad Stab 1996; 54: 217-222

SC

performance and thermal decomposition behaviour of nylon 6 with added halogen-free fire

26

ACCEPTED MANUSCRIPT Figure captions

XRD spectra of clay and SMA/clay

Fig. 2

FT-IR spectra of SMA, clay and the remainder of SMA/clay extracted by acetone.

Fig. 3

FT-IR spectra of (a) PS and (b) the remainder of PS/PA6/APP (12/68/20) extracted

by formic acid.

RI PT

Fig. 1

TEM of (a) PA56-clay and (b) PA56-(SMA/clay)

Fig. 5

XRD spectra of clay, PS/PA6/APP/clay and PS/PA6/APP/(SMA-clay) blends.

Fig. 6

TEM of (a) PA24-clay and (b) PA56-clay

Fig. 7

SEM of blends; (a) PA56, (b) PA56-clay, (c) PA56-(SMA/clay), (d) PA24, (e)

M AN U

PA24-clay and (f) PA24-(SMA/clay)

SC

Fig. 4

Heat release rate curves of blends with PA6 content of 24wt%.

Fig. 9

Heat release rate curves of blends with PA6 content of 56wt%.

Fig. 10

TE D

Fig. 8

SEM of intumescent char residue of outer layer after cone calorimeter; (a) PA24,

Fig. 11

EP

(b)PA24-clay, (c) PA24-(SMA/clay), (d) PA56, (e) PA56-clay and (f) PA56-(SMA/clay) SEM of intumescent char residue of inner layer after cone calorimeter; (a)

AC C

PA24-clay, (b) PA56-clay, (c) PA24-(SMA/clay), (d) PA56-(SMA/clay) Fig. 12

The photos of the aspect of the crust of blends after cone calorimeter test; (a)

PA24-clay, (b) PA56-clay, (c) PA24-(SMA/clay), (d) PA56-(SMA/clay) Fig. 13

XRD spectra of intumescent char residue

Fig. 14

DTA curves of PS/PA6/APP, PS/PA6/APP/clay and PS/PA6/ APP/(SMA-clay)

Fig. 15 TGA curves of blends with PA6 content of 24wt%, pure PS and PA6 Fig. 16

TGA curves of blends with PA6 content of 56wt% 27

ACCEPTED MANUSCRIPT Table 1. Formulation of blends

PS

PA6

APP

clay

SMA/clay

SMA

PA24

56

24

20

0

0

0

PA56

24

56

20

0

0

0

PA24-clay

56

24

20

5

PA56-clay

24

56

20

5

PA24-(SMA/clay)

24

56

20

0

PA56-(SMA/clay)

56

24

20

PA24-clay-SMA

56

24

20

PA56-clay-SMA

24

56

20

0

0

0

5

0

5

0

4

0

1

4

0

1

SC

0

M AN U

0

AC C

EP

TE D

SMA/ clay ratio is 20/80 (w/w)

RI PT

Sample code

28

ACCEPTED MANUSCRIPT Table 2. Flammability characteristics of blends

Flammability Samples

PA56

26.0

PA24-clay

31.5

PA56-clay

PA24-(SMA/clay)

No rating

No rating

No rating

30.9

V-1

30.3

V-1

30.9

V-1

26.3

No rating

EP

AC C

PA56-clay-SMA

V-1

25.9

TE D

PA56-(SMA/clay)

PA24-clay-SMA

RI PT

27.4

M AN U

PA24

UL-94 rating

SC

LOI (%)

29

ACCEPTED MANUSCRIPT Table 3

Element composition on the inner and outer surfaces of char layers for different

blends. Samples

C (wt%)

O (wt%)

Al(wt%)

Si (wt%)

P (wt%)

Inner surface

34.71

28.47

02.96

06.11

27.74

Outer surface

43.10

36.25

01.46

03.09

16.10

Inner surface

59.87

33.62

00.00

Outer surface

43.45

30.98

01.84

04.10

19.63

Inner surface

44.88

25.75

02.20

04.30

22.87

Outer surface

52.76

20.81

02.09

03.68

20.65

PA56-(SMA/clay)

TE D

PA24-(SMA/clay)

SC 00.72

M AN U

PA56-clay

RI PT

PA24-clay

05.79

35.53

28.14

03.10

04.93

28.31

Outer surface

45.77

22.30

02.28

03.73

25.92

AC C

EP

Inner surface

30

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 1

31

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 2

32

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 3

33

ACCEPTED MANUSCRIPT Fig. 4

PA6

PA6

PS

AC C

EP

TE D

M AN U

SC

PS

RI PT

PS

34

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 5

35

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 6

36

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 7

37

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 8

38

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 9

39

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 10

40

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 11

41

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 12

42

ACCEPTED MANUSCRIPT Fig. 13

RI PT

PA24-clay

PA24-SMA/clay PA56-clay

2

3

4

5

6

M AN U

SC

PA56-SMA/clay

7

8

9

10

o

AC C

EP

TE D

2 theta ( )

43

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 14

44

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 15

45

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 16

46