lable at ScienceDirect

Polymer 146 (2018) 63e72

Contents lists avai

Polymer

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

Hexa-[4-(glycidyloxycarbonyl) phenoxy]cyclotriphosphazene chainextender for preparing high-performance flame retardant polyamide6 composites

Miaojun Xu a, Kun Ma a, Dawei Jiang a, Jiaoxia Zhang b, Min Zhao b, Xingkui Guo c,Qian Shao c, Evan Wujcik d, Bin Li a, *, Zhanhu Guo b, **

a Heilongjiang Key Laboratory of Molecular Design and Preparation of Flame Retarded Materials, College of Science, Northeast Forestry University, Harbin150040, PR Chinab Integrated Composites Laboratory (ICL), Department of Chemical Engineering, University of Tennessee, Knoxville, TN 37996, USAc College of Chemical and Environmental Engineering, Shandong University of Science and Technology, Qingdao 266590, PR Chinad Materials Engineering and Nanosensor [MEAN] Laboratory, Department of Chemical and Biological Engineering, The University of Alabama, TuscaloosaUSA

a r t i c l e i n f o

Article history:Received 2 March 2018Received in revised form23 April 2018Accepted 5 May 2018Available online 8 May 2018

Keywords:Polyamide 6Flame retardantChain extension effect

* Corresponding author.** Corresponding author.

E-mail addresses: libinzh62@163.com (B. Li), zguo

https://doi.org/10.1016/j.polymer.2018.05.0180032-3861/© 2018 Published by Elsevier Ltd.

a b s t r a c t

The effects of phosphazene based epoxy resin hexa-[4-(glycidyloxycarbonyl) phenoxy]cyclo-triphosphazene (CTP-EP) as a synergistic agent on fire retardancy, thermal and mechanical properties ofpolyamide 6/aluminum diethylphosphinate (PA6/AlPi) composites were investigated. AlPi and CTP-EPpresented an obviously synergistic flame retardant effect for the PA6 matrix. The PA6/AlPi/CTP-EPcomposites with a thickness of 1.6mm successfully passed UL-94 V-0 rating with the limiting oxygenindex value of 31.0% when the loading amount of AlPi/CTP-EP (mass fraction of 97/3) was 11wt%. Theincorporated CTP-EP improved the complex viscosity of PA6/AlPi/CTP-EP composites due to the chainextending reaction between the epoxide groups of CTP-EP and the terminal groups of PA6 matrix. TheCTP-EP stimulated the formation of more sufficient and compact char layer during combustion. Thehigher melt viscosity and more compact char layer of PA6 composites further inhibited the volatilizationof flammable gases, thus the ignition time of PA6/AlPi/CTP-EP composites was prolonged and thereleased heat was reduced. The introduction of CTP-EP simultaneously enhanced the flame retardant andmechanical properties of PA6/AlPi/CTP-EP composites compared with that of PA6/AlPi composites.

© 2018 Published by Elsevier Ltd.

1. Introduction

Polyamide 6 (PA6) has been widely used in automotive, tele-communication, electronic and electrical industries as it has highmechanical strength, self-lubrication and electrical insulation inrecent years [1e3]. However, its applications were limit attributedto inadequate flame retardancy and severe flammable melt drip-ping during combustion [4]. Consequently, the enhancement offlame retardancy of PA6 matrix attracts researchers' interests in thelast decades. Halogen-containing flame retardants are effective forPA6, whereas its application has been restricted due to the releaseof poisonous, corrosive gases and black smoke during combustion

10@utk.edu (Z. Guo).

[5]. In this case, development of halogen-free flame retardant hasbecome an extensive research topic for PA6 matrix. Up to now, awide variety of halogen-free flame retardants have been proposedfor polyamides including melamine polyphosphate [1], metal hy-droxides [2], red phosphorus [6e8], ammonium polyphosphate[9,10], melamine cyanurate [11,12], organoclay nanocomposite[13,14], and multiwall carbon nanotubes (MWCNTs) composites[15e17]. The phosphorus-containing flame retardant is consideredto be an effective additive for flame retardant PA6 matrix. Thealkylphosphinate salts have been used as novel effective flame re-tardants for PA6 matrix in the past few years [18,19] because theyexhibit excellent flame retardant behaviors and thermal stability.Aluminum salt of diethylphosphinic acid (AlPi) has been proved asan effective flame retardant for PA6 [19,20]. Braun's group inves-tigated the mechanism of flame retardant of AlPi. They found thatAlPi mainly acted as flame inhibition [21]. Moreover, PA6 matrix

Scheme 1. The chemical structure of CTP-EP.

Table 1Formulations and flame retardancy results of PA6/AlPi composites.

Sample Components(wt%)

Flame retardancy

PA6 AlPi LOI (%) UL-94(3.2mm)

UL-94(1.6mm)

S0 100 0 21.0 Fail FailS1 90 10 27.5 Fail FailS2 89 11 28.5 V-2 FailS3 88 12 29.5 V-2 FailS4 87 13 30.3 V-0 V-2S5 86 14 30.6 V-0 V-0S6 85 15 31.2 V-0 V-0

M. Xu et al. / Polymer 146 (2018) 63e7264

molecular chain contains plenty of amide groups, and during hightemperature processing the presence of small amount of impuritiesand moisture would lead to their hydrolysis [22]. In consequence,the incorporation of AlPi into PA6 matrix could reduce the molec-ular weight and mechanical performance of the composites, andconceivably limit its application in particular fields [23,24].

The addition of chain extender has been regarded as an alter-native method to solve this problem. The molecular weight andmelt viscosity of composites are increased by the introduction ofthe chain extender which could comprise the coupling of manypolymer chains and the mechanical performance of composites isultimately enhanced [25]. Since the amino and carboxyl groups atthe end of the PA6 molecular chain could react with caprolactam,oxazoline and epoxide [25e27]. The epoxide group has beenrevealed to be highly reactive with carboxyl and amino groups ofPA6, and a small addition of chain extender to PA6 could increase itstensile strength and melt viscosity [28]. Consequently, the methodof flame retardant incorporated with chain extender was supposedto be an acceptable flame retardant system for preparing flameretardant PA6 composites with superior flame retardancy andmechanical performance simultaneously. The reaction of thefunctional groups during composites processing would implementthe extension of molecular chains, meanwhile, the mechanicalperformance was enhanced. In this case, a flame retardant syner-gistic effect could be generated. Moreover, the polymeric materialsincorporated with cyclotriphosphazenes exhibited excellent flameretardancy and self-extinguish ability caused by the flame retar-dant synergistic effect between phosphorus and nitrogen [29,30].However, few reports have been appeared about the effects of acombination of chain extender and flame retardant on the prop-erties of PA6 composites, thus, the mechanism of flame retardantsynergistic effect between chain extender and flame retardantneeds be further investigated.

In this paper, phosphazenes based epoxy resin hexa-[4-(glycidyloxycarbonyl) phenoxy]cyclotriphosphazene (CTP-EP) wasused as chain extender and flame retardant PA6 composites wasprepared by the combination of CTP-EP and AlPi. The thermaldegradation and combustion behavior, flame retardant synergisticmechanism, flame retardancy, rheological and mechanical perfor-mance of the produced flame retardant PA6 composites wereresearched.

2. Experimental section

2.1. Materials

Aluminum diethylphosphinate (AlPi) and Polyamide 6 (PA6)were provided by Qingdao Fusilin Chemicals Co., China andDUPONT Company Ltd., USA, respectively. Hexa-chlorocyclotriphosphazene and 4-hydroxybenzaldehyde werepurchased from Wuhan Yuancheng Chemical Co. Ltd., China.Epichlorohydrin was purchased from Aladdin reagent (Shanghai)Co. Ltd., China. Tetrahydrofuran, triethylamine, ethyl acetate, so-dium hydroxide, potassium permanganate, sodium sulfate, sulfuricacid and ethanol were purchased from Tianjin Kemiou ChemicalReagent Co. Ltd., China. All the materials were used in this paperwithout any treatment.

2.2. Preparation of PA6 composites

According to our reported work [30], the hexa-[4-(glycidyloxycarbonyl) phenoxy]cyclotriphosphazene (CTP-EP) wassynthesized from hexachlorocyclotriphosphazene, 4-hydroxybenzaldehyde and epichlorohydrin, and the chemicalstructure of CTP-EP was illustrated in Scheme 1. The obtained sticky

CTP-EP was firstly diluted with a certain amount of ethanol andthen mixed homogeneously with AlPi following different massfraction, finally ethanol was removed by a rotary evaporator. ThePA6, flame retardants including AlPi and AlPi/CTP-EP compositewere firstly heated in drying oven under vacuum at 70 �C and thenmixed homogeneously with a high-speed mixer. Granules of theflame retardant PA6 composites were obtained by extruding themixtures of PA6, AlPi and CTP-EP with different mass fractions in atwin-screw extruder at a temperature profile of six heating zones(215, 225, 230, 230, 225 and 220 �C). The granules were moldedthrough the injector at a temperature profile of 215, 225, 230, 230and 225 �C as testing samples. Tables 1e3 show the flame re-tardants contents in PA6 composites.

2.3. Characterization

The limiting oxygen index (LOI) values were measured by oxy-gen index meter at room temperature with a specimen dimensionof 130� 6.5� 3mm3 (ISO 4589-2: 2006).

Vertical burning tests (UL-94) were characterized by verticalcombustion apparatus with a specimen dimension of130� 13� 3.2mm3 and 130� 13� 1.6mm3. The best flameretardancy of polymeric materials would pass V-0 rating.

The tensile and flexural performances of all samples were con-ducted by the tensile testing machine with the speed of 5 and2mmmin�1, respectively. The Izod impact strength of all sampleswith the nick depth of 2mm was tested through Notched Izod

Table 2The effect of mass ratio for AlPi/CTP-EP on the flame retardancy of PA6 composites.

Sample Components (wt%) Flame retardancy

PA6 FR AlPi/CTP-EP LOI (%) UL-94 rating(3.2mm)

UL-94 rating(1.6mm)

S2 89 11 100/0 27.5 V-2 FailS7 89 11 99/1 30.0 V-2 V-2S8 89 11 98/2 30.5 V-0 V-2S9 89 11 97/3 31.0 V-0 V-0S10 89 11 96/4 30.5 V-0 V-0S11 89 11 95/5 30.2 V-0 V-2

Table 3Formulations and flame retardancy results of PA6/AlPi/CTP-EP composites.

Sample Components (wt%) Flame retardancy

PA6 FR AlPi/CTP-EP LOI/% UL-94 rating(3.2mm)

UL-94 rating(1.6mm)

S12 92 8 97/3 27.5 Fail FailS13 91 9 97/3 28.7 V-2 FailS14 90 10 97/3 29.8 V-0 V-2S15 89 11 97/3 31.0 V-0 V-0S16 88 12 97/3 31.8 V-0 V-0

M. Xu et al. / Polymer 146 (2018) 63e72 65

impact instrument (ASTM D256). Five specimens were used foreach test at least and the average values were reported.

Rheological properties of melts were measured by usingAR2000ex rheometer (TA Instruments, USA) setting a chambertemperature of 230 �C. The extruded pellets with diameter andthickness of 25 and 1mm were molded into disks.

The thermal gravimetric analysis (TGA) was carried out bythermal gravimetric analyzer with heating rate of 10 �C min�1, andthe 4e5mg samples were heated with a flow rate of 20mLmin�1

under nitrogen (50e800 �C).The combustion was measured by Fire Testing Technology cone

calorimeter behavior under ventilated conditions (ISO 5660-1). Thesamples (100� 100� 4mm3) were exposed to the cone with heatflux of 50 kWm�2. At least three specimens were tested for eachsample.

The morphology of the char was examined by SEM (acceleratingvoltage, 15 kV). A gold layer was sputter-coated on char residuessurface for better imaging.

The compositions of char residues were characterized throughFTIR, XPS and XRD, respectively. The transmittance mode of FTIRwas used (wavenumber from 4000 to 500 cm�1).

The C, N, P, O and Al elements were analyzed by XPS, using amonochromated Al Ka source at a base pressure of 1.0� 10�8m bar.

XRD spectra were recorded in the 5-60� range using a BrukerAXS D8 diffractometer in configuration 2-theta/theta.

3. Results and discussion

3.1. Flame retardancy

The results of UL-94 and LOI tests of PA6 with AlPi are shown inTable 1. The LOI value of neat PA6 was 21.0% and the materials wereeasily flammable along with continual and severe dripping afterignited. However, the flame retardancy of PA6 composites wasobviously enhanced with the introduction of AlPi. The LOI value ofPA6 with 10wt% AlPi composites increased to 27.5% compared withpure PA6 and the value was improved with increasing the AlPicontent. The PA6 composites containing 14wt% AlPi (1.6mm)successfully passed UL-94 V-0 rating and the LOI value reached30.6%. AlPi possessed an efficient flame retardant effect for PA6

matrix. This can be attributed to the formation of phosphinatecompound obtained from the decomposition of AlPi [20], whichwas further degraded and released the PO free radicals. The PO freeradicals could act as radical scavenger and exert quenching effecton the gaseous phases. On the other hand, the phosphate and py-rophosphate of the decomposition product of AlPi remained in thecondensed phase prevent the underlying materials from degrada-tion and combustion [31].

With the development of the electronics towards miniaturiza-tion and ultra-thin thickness, the polymeric materials used inelectronic and electrical areas require excellent flame retardancyand comprehensive performance. For further improving the flameretardancy of PA6/AlPi composites, the cyclotriphosphazene-basedCTP-EP is used as flame retardant synergist and the results of PA6/AlPi/CTP-EP composites are summarized in Table 2. In AlPi/CTP-EPsystem, the loading amount of flame retardant is fixed at 11wt%and themass ratio of AlPi and CTP-EP can be regulated. As shown inTable 2, the samples failed to pass UL-94 tests and the LOI valuewas27.5% when AlPi alone was used in the flame retardant PA6 matrix.However, the flame retardancy of the PA6 composites wasenhanced with the introduction of CTP-EP. The PA6/AlPi/CTP-EPcomposites (1.6mm) passed UL-94 V-0 rating and when the massratio of AlPi/CTP-EP was 97:3, and the LOI value was 31.0%. Theepoxy groups in CTP-EP reacted with amino and carboxyl endgroups in PA6 matrix and extended the polymer chains, thusenhanced the complex viscosity. The higher viscosity benefits toinhibit the volatilization of decomposition products during com-bustion, which declines the combustion intensity of PA6 compos-ites. Moreover, the cyclotriphosphazene-based CTP-EP canstimulate the carbonization of PA6 matrix and benefit the flameretardancy of composites by exerting an isolation effect in thecondensed phase [32]. As a result, AlPi and CTP-EP present superiorsynergistic flame retardant effect for PA6 matrix both in thecondensed and the gaseous phases. Moreover, the flame retardancyof PA6 composites decreases with further increasing the CTP-EPcontent in the composites (Table 2). For example, the LOI value ofPA6/AlPi/CTP-EP composites (S11) was 30.2% when the mass ratioof AlPi/CTP-EP was 95:5, it only passed UL-94 V-2 rating. The factcan be attributed to that excessive CTP-EP in the AlPi/CTP-EP sys-tem destroyed the optimal flame retardant synergistic effect be-tween gaseous and condensed phase.

When the mass ratio of AlPi/CTP-EP was 97:3, AlPi and CTP-EPexerted an optimal synergistic flame retardant effect for the PA6composites. The effects of the total loading amount of AlPi/CTP-EPon the flame retardancy of PA6 composites are investigated and theresults are summarized in Table 3. The samples with a thickness of3.2 and 1.6mm passed UL-94 V-0 rating when the loading amountof AlPi/CTP-EP was 10 and 11wt%, and the LOI values were 29.8 and31.0%, respectively. AlPi/CTP-EP presented a higher flame retardantefficiency for PA6 matrix compared with that of AlPi alone used inPA6 matrix. The introduced CTP-EP strengthened the flame retar-dant effect in the condensed phase by increasing the complexviscosity of PA6 composites. Moreover, the decomposition productsof phosphazene groups promoted the PA6 matrix charring.Consequently, the AlPi/CTP-EP effectively exerts flame retardanteffect both in gaseous and condensed phases.

3.2. Mechanical properties

The mechanical performance of the neat PA6, PA6/AlPi and PA6/AlPi/CTP-EP (mass fraction of AlPi/CTP-EP fixed at 97:3) compositeswith different loading levels of flame retardant were evaluated bythe tensile, flexural and impact strength tests (Table 4). Comparedwith the neat PA6 composites, the tensile, flexural and Izod impactstrength of the PA6/AlPi and PA6/AlPi/CTP-EP composites

Table 4Mechanical properties of PA6 composites with different loading amount of flame retardant.

loading (wt%) Mechanical properties

Tensile strength (MPa) Flexural strength (MPa) Izod impact strength (kJ$m�2)

PA6/AlPi PA6/AlPi/CTP-EP PA6/AlPi PA6/AlPi/CTP-EP PA6/AlPi PA6/AlPi/CTP-EP

0 67.3± 2.1 67.3± 2.1 90.2± 2.6 90.2± 2.6 12.0± 0.4 12.0± 0.411 57.0± 1.3 60.4± 1.2 85.2± 1.7 90.1± 1.6 9.2± 0.2 9.9± 0.212 56.5± 1.1 60.3± 1.0 84.3± 1.2 89.9± 1.0 8.9± 0.2 9.7± 0.213 56.3± 1.0 59.4± 1.1 82.6± 0.9 89.2± 0.8 8.7± 0.1 9.4± 0.214 56.3± 1.1 59.1± 1.2 82.4± 0.8 88.2± 0.8 8.4± 0.2 9.3± 0.115 55.9± 1.2 58.9± 1.0 81.7± 1.0 88.0± 0.9 8.2± 0.2 9.2± 0.2

M. Xu et al. / Polymer 146 (2018) 63e7266

decreased to some extent due to the incorporation of flame re-tardants, and the mechanical performance for PA6 compositesgradually decreased with the increase of the flame retardantloading, as revealed in Table 4. However, it is worth noting that thetensile, flexural and Izod impact strength for PA6/AlPi/CTP-EPcomposites was higher than that of PA6/AlPi composites underthe same loading amount of flame retardant. For example, whenthe loading amount of flame retardants was 11wt%, the tensile,flexural and Izod impact strength was 57.0MPa, 85.2MPa and9.2 kJm�2 for the PA6/AlPi composites and 60.3MPa, 90.1MPa and9.9 kJm�2 for the PA6/AlPi/CTP-EP composites and presented anenhancement of 5.8, 5.8 and 7.6% compared with that of PA6/AlPicomposites, respectively. This fact can be primarily attributed tothat the epoxide groups of CTP-EP reacted with the terminal aminoor carboxyl groups in the PA6 matrix, and the chain extending re-action improved the molecular weight. The increment of molecularweight benefits the increase of the friction of chain segmentalmotion [22], thus the mechanical performance of PA6/AlPi/CTP-EPcomposites was enhanced compared with that of PA6/AlPi com-posites due to the incorporation of small amount of CTP-EP.

Fig. 1. Storage modulus (a) and loss modulus (b) as a function of angular frequency forneat and flame retardant PA6.

3.3. Rheological properties

Rheological measurement is an effective method to evaluate thechain extension on the linear viscoelastic behavior of polymericmaterials [22,33]. The rheological behaviors of pure PA6, PA6/AlPi(S2) and PA6/AlPi/CTP-EP (S9) composites were investigated at230 �C with stress-controlled rheometer. As a function of angularfrequency, Fig. 1 shows the storage modulus (G0) and loss modulus(G00) of the PA6 materials. The G0 and G00 of the three samplesgradually increase over the whole range of angular frequency andpresent a similar tendency. As revealed in Fig. 1, the G0 and G00 forPA6/AlPi composites are higher than that of neat PA6 caused by thehydrogen bonding association effect between the P¼O bonds inAlPi and hydrogen atom in PA6 matrix. With the incorporation ofCTP-EP into PA6/AlPi composites, the G0 and G00 for PA6/AlPi/CTP-EPcomposites were further improved compared with that of PA6/AlPicomposites, which can be attributed to the coupling reaction be-tween CTP-EP and the terminal groups of PA6 matrix. Conse-quently, the molecular size of the PA6 composites is increased [22].

Fig. 2 depicts the complex viscosity (h*) of neat PA6, PA6/AlPi(S2) and PA6/AlPi/CTP-EP (S9) composites over the whole fre-quency range. Similar to the tendency of G0 and G00, the h* for PA6/AlPi composites is higher than that of neat PA6, as shown in Fig. 2. Itcan be attributed to that the hydrogen bonding association effectbetween AlPi and PA6 matrix enhanced the friction effect betweenmolecular chains of PA6. The movement of the PA6 molecularchains is hindered, thus the h* for PA6/AlPi composites is improved.As shown in Fig. 2, the h* for PA6/AlPi/CTP-EP composites is obvi-ously higher than that of neat PA6 and PA6/AlPi composites due tothe introduced small amount of CTP-EP. The relatively higher melt

viscosity of PA6/AlPi/CTP-EP composites could result from thespecial structure of CTP-EP with six epoxy groups in each molecule,which may contribute to form a network and crosslink structurethrough the interaction between the epoxy groups in CTP-EP andthe terminal groups of PA6 matrix during the melting processillustrated in Fig. 3. Consequently, the h* of PA6/AlPi/CTP-EP com-posites is further enhanced. The higher h* of PA6 composites inmelt state will limit the volatilization of degradation productsduring combustion, which may contribute to improve the flameretardancy of the PA6 composites [34].

Fig. 2. Complex viscosity as a function of angular frequency for neat and flameretardant PA6 fluid.

M. Xu et al. / Polymer 146 (2018) 63e72 67

3.4. Thermal degradation behavior

The thermal stability, charring capability and degradationbehavior of neat PA6 (S0), PA6/AlPi (S2) and PA6/AlPi/CTP-EP (S9)composites were evaluated by thermogravimetric analysis (TGA)tests under nitrogen atmosphere. Fig. 4 shows the TGA and DTGcurves. Table 5 shows the corresponding characteristic of thermaldecomposition parameters with the temperature of 5wt% weightlosses (Tinitial), the temperature of maximum weight loss rate(Tpeak), the weight loss rate at maximum weight loss rate (Rpeak)and the char residue at 800 �C. Pure PA6 began to decompose at396.1 �C and the material was almost completely decomposed withonly 0.6 wt% residues at 800 �C. Meanwhile, the thermal degrada-tion presented one step with a maximum loss weight rate of 26.2%min�1 at 450.8 �C associated with the release of H2O, CO, CO2, NH3,its derivatives and hydrocarbon fragments [35] (Fig. 4(b) andTable 5). The initial decomposition temperature of PA6/AlPi com-posites decreased from 396.1 �C for pure PA6 to 383.4 �C with theaddition of AlPi to PA6, which was caused by catalytic activity of themetal hypophosphite. The metal hypophosphite acted as a weakLewis acid-base for the aromatization of the PA6 matrix, interactedwith the matrix and then reduced the thermal stability of the PA6/AlPi composites [36]. Moreover, the final residue of PA6/AlPi com-posites reached 4.9wt% (800 �C). The thermal degradation of thePA6/AlPi composites also presented one degradation step with amaximum loss weight rate of 20.7% min�1 (432.9 �C) shown in

Fig. 3. Schematic illustration of the chain extendi

Fig. 4(b) and Table 5, and the Tpeak appeared earlier and the Rpeakwas lower than that of neat PA6. The fact was caused by thedecomposition products of AlPi that were covered the materialssurface and the formed insulation layer produced the isolation ef-fect, thus the loss weight rate was decreased. With the incorpora-tion of CTP-EP into the PA6/AlPi composites, the Tinitial for PA6/AlPi/CTP-EP composites slightly decreased from 383.4 �C for the PA6/AlPi composites to 368.3 �C because the decomposition products ofphosphazene groups promoted the PA6 matrix degradation andcharring during heating process. As shown in Fig. 4(b), the PA6/AlPi/CTP-EP composites presented one step during the thermaldegradation process and the Tpeak increased from 432.9 �C for thePA6/AlPi composites to 434.2 �C, and the Rpeak decreased from20.7% min�1 for the PA6/AlPi composites to 19.8% min�1. The factcan be attributed to the formed network structure caused by thereaction between CTP-EP and the terminal groups of PA6 matrix.Meanwhile, the decomposition products of phosphazene groupspromote the PA6 matrix charring, and the formed organic charlayer combines with inorganic layer of the decomposition productof AlPi effectively insulated the material from thermal degradation.Consequently, the char residue for PA6/AlPi/CTP-EP composites at800 �C increased from 4.9wt% for the PA6/AlPi composites to 8.4wt% and increase by 71.4%.

3.5. Flammability behavior

The flammability behavior of PA6 composites was characterizedby cone calorimeter tests. The time to ignition (TTI), heat releaserate (HRR), peak heat release rate (PHRR), total heat release (THR)and average of effective heat of combustion (av-EHC) were used toinvestigate the effect of AlPi and AlPi/CTP-EP composite on theburning behavior of PA6 matrix in a real fire environment. Thedetailed information of burning behavior of pure PA6 (S0), PA6/AlPi(S2) and PA6/AlPi/CTP-EP (S9) composites burned by the conecalorimeter at a heat flux of 50 kWm�2 are shown in Fig. 5 andTable 6.

As revealed in Table 6, the TTI decreased from 65 s (pure PA6) to44 s (PA6/AlPi composites), corresponding to that the incorporatedAlPi was decomposed in advance and then stimulated the PA6matrix catalytic-degradation ahead of time compared with that ofpure PA6. However, PA6/AlPi/CTP-EP composites presented a longerTTI and the value increased from 65 s for neat PA6 and 44 s for PA6/AlPi composites to 89 s for composites and delayed by 24 scompared with that of neat PA6, as indicated in Table 6. Theincorporated small amount of CTP-EP possessing a special structureof six epoxy groups effectively enhanced the complex viscosity ofPA6 composites and formed the network structure, then retardedthe heat enter into the inner PA6 and the release of the combustible

ng reaction between CTP-EP and PA6 matrix.

Fig. 4. TGA (a) and DTG (b) curves of neat and flame retardant PA6 composites.

Table 5Thermal degradation data of neat and flame retardant PA6 composites.

Sample Tinitial Rpeak1/Tpeak1 Char residue (wt %)

(�C) (% min�1/�C) 800 �C

Pure PA6 396.1 26.2/450.3 0.6PA6/AlPi 383.4 20.7/432.9 4.9PA6/AlPi/CTP-EP 368.3 19.8/434.2 8.4

Fig. 5. HRR (a) and THR (b) curves of neat and flame retardant PA6 composites.

Table 6Cone calorimetric date of neat and flame retardant PA6 composites.

Sample Neat PA6 PA6/AlPi PA6/AlPi/CTP-EP

TTI (s) 65 44 89Peak1-HRR (kW$m�2) 912.5 598.1 509.2tPeak1-HRR (s) 185 145 164Peak2-HRR (kW$m�2) e 617.1 653.2tPeak2-HRR (s) e 201 240THR (MJ$m�2) 99.2 92 87.7av-EHC (MJ$kg�1) 27.1 23.7 21.3

M. Xu et al. / Polymer 146 (2018) 63e7268

gas [34]. As a result, the accumulation speed of combustible vaporson materials surface became slow and the combustion requiredmore thermal energy, thus the TTI of PA6/AlPi/CTP-EP compositewas prolonged. The increase of TTI is very significant for improvingthe flame retardancy of the PA6 matrix.

Fig. 5(a) presents the HRR of pure PA6, PA6/AlPi and PA6/AlPi/CTP-EP composites. The results indicate that pure PA6 burnsfiercely after ignition, which has just one intensive HRR peak duringthe combustion process with a PHRR of 912.5 kWm-2 at 185 s,however PA6/AlPi composites presents two HRR peaks. One peak at145 s appeared earlier than that of pure PA6 due to that thedecomposition of AlPi occurred at the early stage of the heating andthe decomposition of AlPi stimulated the PA6 to catalytic-degradation in advance. Meanwhile, the PHRR decreased from912.5 kWm�2 (pure PA6) to 598.1 kWm�2 (PA6/AlPi). It can beattributed to that AlPi in the composites decomposed and exertedan important quenching effect in the gaseous phase as free radical

scavenger [20]. Moreover, in the condensed phase, the phosphateand pyrophosphate of the decomposition product of AlPi coveredthe PA6 matrix surface so that the underlying materials were pre-vented from further combustion. The other HRR peak of PA6/AlPicomposites appeared at 201 s with a PHRR of 617.1 kWm�2 andwascaused by the cracking char or the pyrolysis of the remnant AlPi.After incorporating CTP-EP into the PA6/AlPi composites, two peaksare presented in the HRR curve of PA6/AlPi/CTP-EP composites. Onepeak was prolonged by 19 s compared with that of the PA6/AlPi(164 s). The PHRR of PA6/AlPi/CTP-EP composites was decreased by14.9% from 598.1 kWm�2 (PA6/AlPi) to 509.2 kWm�2. It is mainlyattributed to the formed network structure and higher complexviscosity of PA6 composites induced by the reaction between CTP-EP and PA6 matrix. Thus, the formed higher complex viscosityinhibited the heat enter into the inner PA6 matrix and alsoinhibited the release of combustible gas enter into combustionzone, which delayed the degradation process and reduced the

Fig. 6. SEM images of the char residue for PA6/AlPi (a) and PA6/AlPi/CTP-EP (b) composites after cone calorimeter tests.

M. Xu et al. / Polymer 146 (2018) 63e72 69

combustion intensity [34]. The second HRR peak of PA6/AlPi/CTP-EP composite appeared at 240 s and later than that of PA6/AlPicomposites. Moreover, the PHRR increased from617.1 kWm�2 (PA6/AlPi) to 653.2 kWm�2 (PA6/AlPi/CTP-EP com-posites). The introduction of CTP-EP stimulated the formation ofcrosslinking structure and enhanced the complex viscosity ofcomposites. Moreover, the decomposition of the phosphazene inCTP-EP promoted the phosphorus-rich char shield [37], the charshield would impede heat exchange and gas release. Meanwhile,the materials underneath accumulated adequate heat fromcontinuous thermal irradiance and evolved combustible gases topromote the breakage of the polymer melt to some degree.Consequently, the combustion intensity at the second HRR peakwas relatively higher than that of the PA6/AlPi composites. Theincorporated small amount of CTP-EP decreased the THR of com-posites, and the THR declined from 99.2MJm�2 for neat PA6 ma-terial and 92.0MJm�2 for PA6/AlPi composites to 87.7MJm�2 forPA6/AlPi/CTP-EP composites (Fig. 5(b) and Table 6). The combina-tion of CTP-EP with AlPi exerted evidently flame retardant syner-gistic effect for PA6 matrix.

The av-EHC discloses the burning rate of volatile gases duringcombustion. Table 6 shows that the av-EHC is decreased from27.1MJ kg�1 (neat PA6) to 23.7MJ kg�1 (PA6/AlPi composites)indicating that the gaseous phase combustion intensity is sup-pressed due to the introduction of AlPi. It is decomposed to releasePO free radicals and then PO free radicals captures the highlyreactive H and OH radicals formed during the combustion pro-cesses to interrupt the radical action and combustion reaction [20].With the incorporation of CTP-EP, the av-EHC of PA6/AlPi/CTP-EPcomposites further declines from 23.7MJ kg�1 of PA6/AlPi com-posites to 21.3MJ kg�1. The phenomenon can be mainly attributedto the enhanced complex viscosity of the PA6/AlPi/CTP-EP com-posites. It can effectively inhibit the release of flammable gas, thusthe released heat in the corresponding gaseous phase decreasesduring the combustion.

Fig. 7. FTIR spectra of the char residue for PA6/AlPi (a) and PA6/AlPi/CTP-EP (b)composites after cone calorimeter tests.

3.6. Morphologies of the char residue

SEM images show the morphologies of the char residues of PA6/AlPi (S2) and PA6/AlPi/CTP-EP (S9) composites after cone calorim-eter (Fig. 6). The char layer surface morphology of the PA6/AlPicomposite presents an inhomogeneous and relatively loose struc-ture due to the insufficient char formation (Fig. 6(a)). Consequently,the heat and flammable volatiles can transfer between underlyingmaterials and flame zone, which lead to the relatively poor flame

retardancy for PA6 matrix. The morphology of the char layer forPA6/AlPi/CTP-EP composites appears more uniform and compactthan that of PA6/AlPi composites (Fig. 6(b)), which indicates thatthe phosphazene groups in CEP-EP stimulate the PA6 matrixdegradation and char-forming during the combustion and formmore compact char layer on materials surface. Consequently, theformed char layer could effectively produce an isolation effect inthe condensed phase. The flame retardancy of PA6/AlPi/CTP-EPcomposites is enhanced comparing with that of PA6/AlPicomposite.

3.7. Chemical compositions of the char residue

The char residues of PA6/AlPi and PA6/AlPi/CTP-EP compositesafter cone calorimeter are tested by FTIR (Fig. 7). The characteristicabsorption peaks of PA6/AlPi composites char residue (1121 and720 cm�1) indicate the formation of P-O-C and P-O-P structures[38]. The peak at 3445 cm�1 is due to the stretching vibration of N-H. The peak at 1618 cm�1 corresponds to the aromatic ring causedby the decomposition and charring of neat PA6 matrix. The peak at1270 cm�1 is assigned to the P¼N stretching vibration of thephosphazene groups [37]. The peaks at 2860 and 2937 cm�1 areattributed to the stretching vibration of C-H. The peak at 3081 cm�1

Table 7Element relative content of the char layer for flame retardant PA6 by XPS analysis.

Element Char layer of PA6/AlPi Char layer of PA6/AlPi/CTP-EP

Peak BE (ev) Atomic percentage (%) Peak BE (ev) Atomic percentage (%)

C1s 284.55 67.41 284.59 69.74N1s 386.4 1.08 398.77 5.99O1s 531.87 26.82 531.98 15.68P2p 134.61 2.21 134.56 3.89Al2p 75.41 2.55 75.27 4.41

Fig. 9. XRD spectra of the residues of PA6/AlPi and PA6/AlPi/CTP-EP composites aftercone calorimeter tests.

M. Xu et al. / Polymer 146 (2018) 63e7270

is the stretching vibration of O-H induced by the epoxide ring ofCTP-EP that reacted with the amino or carboxyl end groups of PA6matrix, which demonstrate that the chain extension reaction oc-curs during extrusion processing or combustion. The phosphoruscontaining oxides and phosphazene structure remained in the charresidue enhance the strength of char residue, thus the flameretardancy of PA6/AlPi/CTP-EP composites is improved.

The chemical compositions of the outer residual char surface ofPA6/AlPi and PA6/AlPi/CTP-EP composites after cone calorimeterwere tested by XPS (Table 7 and Fig. 8). The relative content ofphosphorus and nitrogen was obviously increased from 2.21 to1.08% for the PA6/AlPi composites to 3.89 and 5.99% for the PA6/AlPi/CTP-EP composites, respectively. The incorporated CTP-EPreacted with the molecular chain of PA6 matrix and the phospha-zene structure of CTP-EP was remained in char residue. The phos-phazene structure possesses excellent thermal stability [29,30] andbenefits to improve the thermal oxidation performance for charlayer. The relative content of carbonwas also increased from 67.41%for PA6/AlPi composites to 69.74% for PA6/AlPi/CTP-EP compositesdue to the catalytic charring effect of the remained phosphazene.Moreover, the relative content of aluminum was also increasedfrom 2.55% for PA6/AlPi composites to 4.41% for PA6/AlPi/CTP-EPcomposites, which can be attributed to the more decompositionproduct of AlPi remained in the condensed phase induced by thehigher complex viscosity of PA6/AlPi/CTP-EP composites during thecombustion. As a result, more amount flame retardant elementsincluding phosphorus, nitrogen, carbon and aluminum benefitedthe formation of sufficient char layer with high quality, whicheffectively prevented the PA6 matrix from further oxidation anddecomposition resulting in the improved flame retardancy for PA6matrix during combustion.

Fig. 8. XPS spectra of char residue of PA6/AlPi and PA6/AlPi/CTP-EP composites aftercone calorimeter tests.

The X-ray diffractions of the residues of PA6/AlPi and PA6/AlPi/CTP-EP composites after cone calorimeter tests were tested (Fig. 9).The two samples were degraded into amorphous phase, which wascharacterized by a broad band between 15� and 35�, as shown inFig. 6. These bands are generally attributed to pregraphitic carbonspecies [39], which reveals the formation of the char. Comparedwith the curve of PA6/AlPi, an additional band (9�) correspondingto the crystallized AlPi [40] appears in the curve of PA6/AlPi/CTP-EP.The introduced CTP-EP effectively enhanced the isolation effect ofchar layer in the condensed phase during combustion, thus AlPi canbe maintained in the residue of PA6/AlPi/CTP-EP.

4. Conclusion

The combination of AlPi and CTP-EP presented obviously syn-ergistic flame retardant effect to the PA6 matrix, and the AlPi/CTP-EP system exerted flame retardant effect both in gaseous andcondensed phases. The specimen (1.6mm) passed UL-94 V-0 ratingand the LOI value reached 31.0% with the addition of 11wt% AlPi/CTP-EP. The TTI of PA6/AlPi/CTP-EP composite was delayed from44 s for PA6/AlPi to 89 s, and the PHRR was also obviously post-poned compared with that of PA6/AlPi composites due to the chainextending and tackifying effect of CTP-EP. The chain extending re-action between CTP-EP and the terminal groups of PA6 matrixgranted a significant enhancement of the storage modulus, lossmodulus and complex viscosity for PA6 composites. The mechani-cal properties of PA6/AlPi/CTP-EP composites were improved withthe introduction of CTP-EP compared with that of PA6/AlPi com-posites. As a result, the flame retardant and mechanical perfor-mance of PA6/AlPi/CTP-EP composites were simultaneouslyenhanced. With adding proper nanofillers, these composites can beused to produce unique nanocomposites that can provide differentproperties and performances for various applications such as

M. Xu et al. / Polymer 146 (2018) 63e72 71

electromagnetic interference (EMI) shielding [41,42], structural andfunctional materials [43e59] and sensing [60e65].

Acknowledgments

We are grateful for funding supported by the FundamentalResearch Funds for the Central Universities (2572016DB02), Na-tional Nature Science Foundation of China (51673035) and Hei-longjiang Major Research Projects (GA15A101).

References

[1] S. Jahromi, W. Gabrielse, A. Braam, Effect of melamine polyphosphate onthermal degradation of polyamides: a combined X-ray diffraction and solid-state NMR study, Polymer 44 (2003) 25e37.

[2] G.X. Fei, Y. Liu, Q. Wang, Synergistic effects of novolac-based char former withmagnesium hydroxide in flame retardant polyamide-6, Polym. Degrad. Stabil.93 (2008) 1351e1356.

[3] Y. Liu, Q. Wang, The investigation on the flame retardancy mechanism ofnitrogen flame retardant melamine cyanurate in polyamide 6, J. Polym. Res.16 (2009) 583e589.

[4] F. Dabrowski, S. Bourbigot, R. Delobel, M.L. Bras, Kinetic modelling of thethermal degradation: of polyamide-6 nanocomposite, Eur. Polym. J. 36 (2000)273e284.

[5] M.O. Milbrath, Y. Wenger, C.W. Chang, C. Emond, D. Garabrant, B.W. Gillespie,O. Jolliet, Apparent half-lives of dioxins, furans, and polychlorinated biphenylsas a function of age, body fat, smoking status, and breast-feeding, Environ.Health Perspect. 117 (2009) 417e425.

[6] A.I. Balabanovich, G.F. Levchik, S.V. Levchik, W. Schnabel, Fire retardance inpolyamide-6: the effects of red phosphorus and radiation-induced cross-links,Fire Mater. 25 (2001) 179e184.

[7] Y. Liu, Q. Wang, Preparation of microencapsulated red phosphorus throughmelamine cyanurate self-assembly and its performance in flame retardantpolyamide 6, Polym. Eng. Sci. 46 (2006) 1548e1553.

[8] W.S. Jou, K.N. Chen, D.Y. Chao, C.Y. Lin, J.T. Yeh, Flame retardant and dielectricproperties of glass fibre reinforced nylon-66 filled with red phosphorous,Polym. Degrad. Stabil. 74 (2001) 239e245.

[9] (a) J.B. Dahiya, M. Muller-Hagedron, H. Bockhorn, B.K. Kandola, Synthesis andthermal behaviour of polyamide 6/bentonite/ammonium polyphosphatecomposites, Polym. Degrad. Stabil. 93 (2008) 2038e2041;(b) C. Wang, Y. Wu, Y. Li, Q. Shao, X. Yan, C. Han, Z. Wang, Z. Liu, Z. Guo, Flameretardant rigid polyurethane foam with a phosphorus-nitrogen single intu-mescent flame retardant, Polym. Adv. Technol. 29 (2018) 668e676.

[10] A. Riva, G. Camino, L. Fomperie, P. Amigou€et, Fire retardant mechanism inintumescent ethylene vinyl acetate compositions, Polym. Degrad. Stabil. 82(2003) 341e346.

[11] P. Gijsman, R. Steenbakkers, C. Furst, J. Kersjes, Differences in the flameretardant mechanism of melamine cyanurate in polyamide 6 and polyamide66, Polym. Degrad. Stabil. 78 (2002) 219e224.

[12] C. Isbasar, J. Hacaloglu, Investigation of thermal degradation characteristics ofpolyamide-6 containing melamine or melamine cyanurate via direct pyrolysismass spectrometry, J. Anal. Appl. Pyrol. 98 (2012) 221e230.

[13] I.S. Zope, A. Dasari, F.L. Guan, Z.Z. Yu, Influence of metal ions on thermo-oxidative stability and combustion response of polyamide 6/clay nano-composites, Polymer 92 (2016) 102e113.

[14] T. Kashiwagi, R.H. Harris Jr., X. Zhang, R.M. Briber, B.H. Cipriano, S.R. Raghavan,W.H. Awad, J.R. Shields, Flame retardant mechanism of polyamide 6eclaynanocomposites, Polymer 45 (2004) 881e891.

[15] J. Ryu, M. Han, Improvement of the mechanical and electrical properties ofpolyamide 6 nanocomposites by non-covalent functionalization of multi-walled carbon nanotubes, Compos. Sci. Technol. 102 (2014) 169e175.

[16] M. Wang, W.Z. Wang, T.X. Liu, W.D. Zhang, Melt rheological properties ofnylon 6/multi-walled carbon nanotube composites, Compos. Sci. Technol. 68(2008) 2498e2502.

[17] H. Meng, G.X. Sui, P.F. Fang, R. Yang, Effects of acid- and diamine-modifiedMWNTs on the mechanical properties and crystallization behavior of Poly-amide 6, Polymer 49 (2008) 610e620.

[18] Q.F. Li, B. Li, S.Q. Zhang, M. Lin, Investigation on effects of aluminum andmagnesium hypophosphites on flame retardancy and thermal degradation ofpolyamide 6, J. Appl. Polym. Sci. 125 (2012) 1782e1789.

[19] Z. Hu, L. Chen, G.P. Lin, Y. Luo, Y.Z. Wang, Flame retardation of glass-fibre-reinforced polyamide 6 by a novel metal salt of alkylphosphinic acid,Polym. Degrad. Stabil. 96 (2011) 1538e1545.

[20] M. Do�gan, E. Bayraml, The flame retardant effect of aluminum phosphinate incombination with zinc borate, borophosphate, and nanoclay in polyamide-6,Fire Mater. 38 (2014) 92e99.

[21] U. Braun, H. Bahr, B. Schartel, Fire retardancy effect of aluminum phosphinateand melamine polyphosphate in glass fibre reinforced polyamide 6, E-Poly-mers 10 (2013) 443e456.

[22] A. Wirasaputra, J.Q. Zhao, Y.M. Zhu, S.M. Liu, Y.C. Yuan, Application of bis(-glycidyloxy) phenylphosphine oxide as a chain extender for polyamide-6, RSC

Adv. 5 (2015) 31878e31885.[23] O. Polat, C. Kaynak, Use of boron oxide and boric acid to improve flame

retardancy of an organophosphorus compound in neat and fiber reinforcedpolyamide-6, Vinyl.Addit. J. Technol 22 (2016) 300e310.

[24] C. Kaynak, O.J. Polat, Influences of nanoclays on the flame retardancy of fiber-filled and unfilled polymer-6 with and without aluminum dieth-ylphosphinate, Fire Sci. 33 (2015) 87e112.

[25] C.X. Lu, R.G. Ye, Y. Yang, X.C. Ren, X. Cai, J. Macromol, Chemical modification ofpolyamide 6 by chain extension with terephthaloyl-biscaprolactam, Sci.Technol. Humanit. Part B: Phys. 50 (2011) 350e362.

[26] L. N�ery, H. Lefebvre, A. Fradet, Chain extension of carboxy-terminatedaliphatic polyamides and polyesters by arylene and pyridtlene bisoxazo-lines, Macromol. Chem. Phys. 205 (2004) 448e455.

[27] R. Bertani, A. Boscolo-Boscoletto, N. Dintcheva, et al., New phosphazene-basedchain extenders containing allyl and epoxide groups, Des. Monomers Polym. 6(2003) 245e266.

[28] M. Villalobos, A. Awojulu, T. Greeley, G. Turco, G. Deeter, Oligomeric chainextenders for economic reprocessing and recycling of condensation plastics,Inside Energy 31 (2006) 3227e3234.

[29] L.J. Qian, L. Ye, Y. Qiu, S. Qu, Thermal degradation behavior of the compoundcontaining phosphaphenanthrene and phosphazene groups and its flameretardant mechanism on epoxy resin, Polymer 52 (2011) 5486e5493.

[30] G.R. Xu, M.J. Xu, B. Li, Synthesis and characterization of a novel epoxy resinbased on cyclotriphosphazene and its thermal degradation and flammabilityperformance, Polym. Degrad. Stabil. 109 (2014) 240e248.

[31] H. Ge, G. Tang, W.Z. Hu, B.B. Wang, Y. Pan, Y. Song, Y. Hu, Aluminum hypo-phosphite microencapsulated to improve its safety and application to flameretardant polyamide 6, J. Hazard Mater. 294 (2015) 186e194.

[32] J. Sun, X.Y. Gu, S. Zhang, M. Coquelle, S. Bourbigot, S. Duquesne, M. Casetta,Improving the flame retradancy of polyamide 6 by incorporating hexa-chlorocyclotriphosphazene modified MWNT, Polym. Adv. Technol. 25 (2014)1099e1107.

[33] M. Benoit, L. Khalid, M. Abderrahim, Improvement of blown film extrusion ofpoly(lactic acid): structureeprocessingeproperties relationships, Polym. Eng.Sci. 54 (2014) 840e857.

[34] P.A. Song, H. Liu, Y. Shen, B.X. Du, Z.P. Fang, Y. Wu, Fabrication of dendrimer-like fullerene (C60)-decorated oligomeric intumescent flame retardant forreducing the thermal oxidation and flammability of polypropylene nano-composites, J. Mater. Chem. 19 (2009) 1305e1313.

[35] P.R. Hornsby, J. Wang, R. Rothon, G. Jackson, G. Wilkinson, K. Cossick, Thermaldecomposition behavior of polyamide fire-retardant compositions containingmagnesium hydroxide filler, Polym. Degrad. Stabil. 51 (1996) 235e249.

[36] M. Afshari, A. Gupta, D. Jung, R. Kotek, A.E. Tonelli, N. Vasanthan, Properties offilms and fibers obtained from Lewis acidebase complexed nylon 6,6, Polymer49 (2008) 1297e1304.

[37] M.J. Xu, G.R. Xu, Y. Leng, B. Li, Synthesis of a novel flame retardant based oncyclotriphosphazene and DOPO groups and its application in epoxy resins,Polym. Degrad. Stabil. 123 (2016) 105e114.

[38] S. Yang, J. Wang, S. Huo, M. Wang, J.P. Wang, B. Zhang, Synergistic flame-retardant effect of expandable graphite and phosphorus-containing com-pounds for epoxy resin: strong bonding of different carbon residues, Polym.Degrad. Stabil. 128 (2016) 89e98.

[39] M.L. Bras, S. Bourbigot, Y.L. Tallec, J. Laureyns, Synergy in intumescenceapplication to b-cyclodextrin carbonisation agent in intumescent additives forfire retardant polyethylene formulations, Polym. Degrad. Stabil. 56 (1997)11e21.

[40] S. Duquesne, G. Fontaine, O. C�erin-Delaval, B. Gardelle, G. Tricot, S. Bourbigot,Study of the thermal degradation of an aluminium phosphinate-aluminiumtrihydrate combination, Thermochim. Acta 551 (2013) 175e183.

[41] K. Zhang, G. Li, L. Feng, N. Wang, J. Guo, K. Sun, K. Yu, J. Zeng, T. Li, Z. Guo,M. Wang, Ultralow percolation threshold and enhanced electromagneticinterference shielding in poly(L-lactide)/multi-walled carbon nanotubesnanocomposites with electrically conductive segregated networks, J. Mater.Chem. C 5 (2017) 9359e9369.

[42] J. Guo, H. Song, H. Liu, C. Luo, Y. Ren, T. Ding, M.A. Khan, D.P. Young, X. Liu,X. Zhang, J. Kong, Z. Guo, Polypyrrole-interface-functionalized nanomagnetiteepoxy nanocomposites as electromagnetic wave absorber with enhancedflame retardancy, J. Mater. Chem. C 5 (2017) 5334e5344.

[43] X. Wang, X. Liu, H. Yuan, et al., Non-covalently functionalized graphenestrengthened poly(vinyl alcohol), Mater. Des. 139 (2018) 372e379.

[44] Y. Guo, G. Xu, X. Yang, et al., Significantly enhanced and precisely modeledthermal conductivity in polyimide nanocomposites by chemically modifiedgraphene via in-situ Polymerization and Electrospinning-hot press Technol-ogy, J. Mater. Chem. C 6 (2018) 3004e3015.

[45] X. Wu, J. Song, J. Li, et al., Reparation of recycled acrylonitrile- butadiene-styrene by pyromellitic dianhydride: reparation performance evaluation andproperty analysis, Polymer 124 (2017) 41e47.

[46] Z. Hu, Q. Shao, Y. Huang, et al., Light triggered interfacial damage self-healingof poly(p- phenylene benzobisoxazole) fiber composites, Nanotechnology 29(2018) 185602.

[47] Y. Zhang, M. Zhao, J. Zhang, et al., Excellent corrosion protection performanceof epoxy composite coatings filled with silane functionalized silicon nitride,J. Polym. Res. 25 (2018) 130, https://doi.org/10.1007/s10965-018-1518-2.

[48] Y. Ma, L. Lyu, Y. Guo, et al., Porous lignin based poly (acrylic acid)/organo-montmorillonite nanocomposites: swelling behaviors and rapid removal of

M. Xu et al. / Polymer 146 (2018) 63e7272

Pb (II) ions, Polymer 128 (2017) 12e23.[49] Z. Wu, S. Gao, L. Chen, et al., Electrically insulated epoxy nanocomposites

reinforced with synergistic core-shell SiO2@MWCNTs and montmorillonitebifillers, Macromol. Chem. Phys. 218 (2017) 1700357.

[50] Y. He, S. Yang, H. Liu, et al., Reinforced carbon fiber laminates with orientedcarbon nanotube epoxy nanocomposites: magnetic field assisted alignmentand cryogenic temperature mechanical properties, J. Colloid Interface Sci. 517(2018) 40e51.

[51] J. Zhao, L. Wu, C. Zhan, Q. Shao, Z. Guo, L. Zhang, Overview of polymernanocomposites: computer simulation understanding of physical properties,Polymer 133 (2017) 272e287.

[52] C. Wang, M. Zhao, J. Li, et al., Silver nanoparticles/graphene oxide decoratedcarbon fiber synergistic reinforcement in epoxy-based composites, Polymer131 (2017) 263e271.

[53] X. Wang, X. Zeng, D. Cao, Biomass-derived nitrogen-doped porous carbons(NPC) and NPC/polyaniline composites as high performance supercapacitormaterials, Eng. Sci. 1 (2018) 55e63, https://doi.org/10.30919/es.180325.http://www.espublisher.com/doi/10.30919/es.180325.

[54] J. Gu, W. Dong, Y. Tang, et al., Ultra-low dielectric, fluoride-containing cyanateester resins combining with prominent mechanical properties and excellentthermal and dimension stabilities, J. Mater. Chem. C 5 (2017) 6929e6936.

[55] K. Sun, R. Fan, X. Zhang, et al., An overview of metamaterials and theirachievements in wireless power transfer, J. Mater. Chem. C 6 (2018)2925e2943.

[56] Z. Hu, D. Zhang, L. Yu, Y. Huang, Light-triggered C60 release from a graphene/cyclodextrin nanoplatform for the protection of cytotoxicity induced by nitricoxide, J. Mater. Chem. B 6 (2018) 518e526.

[57] K. Sun, Z. Xie, T. Wang, et al., Flexible polydimethylsiloxane/multi-walled

carbon nanotubes membranous metacomposites with negative permittivity,Polymer 125 (2017) 50e57.

[58] X. Duan, T. Bai, J. Du, J. Kong, One-pot synthesis of redox-reduction responsiveamphiphlic drug self-delivery micelle of Doxorubicin-disulfide-methoxypolyethylene glycol for tumor therapy, J. Mater. Chem. B 6 (2018) 39e43.

[59] C. Liang, P. Song, H. Gu, et al., Nanopolydopamine coupled fluorescentnanozinc oxide reinforced epoxy nanocomposites, Compos. Appl. Sci. Manuf.102 (2017) 126e136.

[60] Y. Li, B. Zhou, G. Zheng, X. Liu, T. Li, C. Yan, C. Cheng, K. Dai, C. Liu, C. Shen,Z. Guo, Continuously prepared highly conductive and stretchable SWNTs/MWNTs synergistically composited electrospun thermoplastic polyurethaneyarns for wearable sensing, J. Mater. Chem. C 6 (2018) 2258e2269.

[61] X. Guan, G. Zheng, K. Dai, et al., Carbon nanotubes-adsorbed electrospun PA66nanofiber bundles with improved conductivity and robust flexibility, ACSAppl. Mater. Interfaces 8 (2016) 14150e14159.

[62] H. Liu, W. Huang, X. Yang, et al., Organic vapor sensing behaviors ofconductive thermoplastic polyurethane-graphene nanocomposites, J. Mater.Chem. C 4 (2016) 4459e4469.

[63] H. Liu, M. Dong, W. Huang, et al., Lightweight conductive graphene/thermo-plastic polyurethane foams with ultrahigh compressibility for piezoresistivesensing, J. Mater. Chem. C 5 (2017) 73e83.

[64] H. Liu, Y. Li, K. Dai, et al., Electrically conductive thermoplastic elastomernanocomposites at ultralow graphene loading levels for strain sensor appli-cations, J. Mater. Chem. C 4 (2016) 157e166.

[65] C. Hu, Z. Li, Y. Wang, et al., Comparative assessment of the strain-sensingbehaviors of polylactic acid nanocomposites: reduced graphene oxide orcarbon nanotubes, J. Mater. Chem. C 5 (2017) 2318e2328.