materials Communication

Influence of Polymer-Clay Interfacial Interactions on the Ignition Time of Polymer/Clay Nanocomposites Indraneel S. Zope 1 , Aravind Dasari 1, * 1 2

*

ID

and Zhong-Zhen Yu 2

School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore; [email protected] State Key Laboratory of Organic-Inorganic Composites, College of Materials Science and Engineering, Beijing University of Chemical Technology, Beijing 100029, China; [email protected] Correspondence: [email protected]; Tel.: +65-6790-6402; Fax: +65-6790-9081

Received: 14 July 2017; Accepted: 8 August 2017; Published: 11 August 2017

Abstract: Metal ions present on smectite clay (montmorillonite) platelets have preferential reactivity towards peroxy/alkoxy groups during polyamide 6 (PA6) thermal decomposition. This changes the decomposition pathway and negatively affects the ignition response of PA6. To restrict these interfacial interactions, high-temperature-resistant polymers such as polyetherimide (PEI) and polyimide (PI) were used to coat clay layers. PEI was deposited on clay by solution-precipitation, whereas PI was deposited through a solution-imidization-precipitation technique before melt blending with PA6. The absence of polymer-clay interfacial interactions has resulted in a similar time-to-ignition of PA6/PEI-clay (133 s) and PA6/PI-clay (139 s) composites as neat PA6 (140 s). On the contrary, PA6 with conventional ammonium-based surfactant modified clay has showed a huge drop in time-to-ignition (81 s), as expected. The experimental evidences provided herein reveal the role of the catalytic activity of clay during the early stages of polymer decomposition. Keywords: time-to-ignition; thermo-oxidation; combustion; polyamide 6; clay

1. Introduction Polymer/clay nanocomposites show a significant reduction in peak heat release rates (up to 70%) during combustion [1,2]. However, with the exception of a few reports [3,4], the majority of the studies on flammability of polymer/clay nanocomposites have reported an earlier ignition behavior compared to their corresponding neat polymers [2]. It is known that the reactivity of smectite clays such as montmorillonite is affected by replacing the relatively mobile and exchangeable cationic species, such as Na+ or Ca2+ , on their surfaces by higher valence metal ions, such as Mg2+ , Zn2+ , Cu2+ , Al3+ or Fe3+ [5–8]. In fact, this attribute has been exploited for applications like catalysis [9,10]. In our recent systematic investigations on this topic [7,11], we showed that by amplifying the content of a specific cation (which are inherently present in the clay structure), thermal decomposition and combustion reactions of PA6 could be affected. For instance, Al3+ -rich PA6/clay nanocomposites ignited early with a lower onset temperature of decomposition under oxidative conditions; while an Mg2+ -rich composite displayed the maximum thermo-oxidation stability and gave the highest residue. This suggested that the chemical interaction of the PA6 matrix and/or its intermediate decomposition products with the clay surface changes the decomposition pathway of PA6. Furthermore, it has been observed that the nature of intermediate products depends on the type of interacting and/or coordinating metal ions. Investigations have also been carried out on the usage of other metal ions like nickel along with clay layers to influence the carbonaceous residue formation [12–14]. Tang et al. [12] have noted that the yield of multiwalled carbon nanotubes increased (up to 41.1% from 5.2%) in the presence of nickel and clay in polypropylene when subjected to catalytic combustion.

Materials 2017, 10, 935; doi:10.3390/ma10080935

www.mdpi.com/journal/materials

Materials 2017, 10, 935

2 of 10

Hu et al. [14] have also used clay supported with nickel cation to enhance char-forming reactions in polypropylene (PP) containing ammonium polyphosphate (APP), melamine formaldehyde (MF), and 1,3,5-triazine-2,4,6-trimorpholinyl (TTM). Though the exact reasons are unknown, it has been shown that even the addition of 3 wt % of Ni2+ /MF modified montmorillonite (MMT) to a PP/APP/TTM system results in a drop of ignition time from 38 s to 31 s. In the presence of a non-swelling clay such as sepiolite (0.5 wt %), Pappalardo et al. [15] have noted a reduction in time-to-ignition and limiting oxygen index (LOI) of PP. However, they have attributed this to the reduced melt flow due to the increase in viscosity of PP in the presence of sepiolite. Based on the above discussions, it is hypothesized that if there is a delay in the direct chemical interactions between the clay surface and the neighboring decomposing organic matrix, the effect on time-to-ignition could be minimized. This forms the basis of the current work, which attempts to validate the hypothesis. A simple approach is to coat the clay layers with another matrix before blending with PA6. In fact, though not directly related to catalytic activity or reactivity, there are some reports on the coating of clay platelets (which are inherently highly hydrophilic) with various water-soluble and soft polymers such as polyvinyl alcohol [16], polypyrole [17], and polyethyleneglycol [18]. For instance, coated and self-assembled clay sheets were synthesized to mimic nacre-like building blocks with intrinsic hard/soft character [16]. Nonetheless, the poor thermal stability and water solubility of these polymers hinder their usage to prove the above-proposed hypothesis. It is essential to use high-temperature-resistant polymers that are compatible with PA6 and maintain the protective coating on clay during compounding as well as the initial stages of thermal decomposition of the PA6 matrix (i.e., polymers with a glass transition temperature above 220 ◦ C). Accordingly, polar polymers such as PEI and PI were chosen. The rationale behind using these polymers is to compare the effectiveness of two coating methodologies, i.e., solution blending-precipitation (for PEI) and solution blending-imidization-precipitation (for PI). Commercially available organically modified clay, especially ammonium-based surfactant modified clay (C30B), has been extensively studied in the PA6 matrix for combustion properties. Hence, the use of C30B instead of unmodified/pristine Na+ MMT provides an opportunity to have a direct comparison with ignition times previously reported for PA6/organoclay systems. This will facilitate a straight-forward evaluation of the proposed hypothesis. 2. Results and Discussion 2.1. Analysis of Clay Coatings Representative scanning electron microscope (SEM) micrographs of organoclay (i.e., C30B), PEI coated clay (PEI-clay) and PI coated clay (PI-clay) platelets are shown in Figure 1, highlighting the differences in the uniformity of coating on organoclay with PEI and PI. The precipitation of PEI from solution on organoclay resulted in a non-uniform and thick coverage, whereas imidization of poly(pyromellitic dianhydride-co-4,40 -oxydianiline) amic acid (PAA) over organoclay resulted in a more uniform coverage. This is due to the good dispersion and distribution of the silicate layers in PAA, providing a high heterogeneous surface area for further reactions. In fact, this uniformity in the coating with PI is expected, based on the investigation by Tyan et al. [19]. They reported that the imidization kinetics of PAA are enhanced with the addition of organically (p-phenylenediamine) modified clay. This further reduced the imidization temperature from 300 ◦ C to 250 ◦ C without affecting the structure of the PI thus formed. The heterogeneous surface provided by the clay platelets was considered as a major parameter for changes in reaction temperature and kinetics, as it promoted dehydration and imide ring closure reactions.

Materials 2017, 10, 935 Materials 2017, 10, 935 Materials 2017, 10, 935

3 of 10 3 of 10 3 of 10

Figure 1. 1. Representative scanning electron microscope microscope micrographs: (a) (a) organoclay; (b) (b) PEI-clay; and and Figure Figure 1.Representative Representativescanning scanningelectron electron microscopemicrographs: micrographs: (a)organoclay; organoclay; (b)PEI-clay; PEI-clay; and (c) PI-clay. (c) (c)PI-clay. PI-clay.

Further, the FTIR spectra of coated clay powders in Figure 2a,b confirm the completion of Further,the theFTIR FTIR spectra of coated clay powders in Figure 2a,b confirm the completion of Further, spectra of coated clay powders in Figure 2a,b confirm the completion of reactions. reactions. Characteristic IR bands for imide I (anti-symmetric C=O stretch) and imide II (symmetric reactions. Characteristic IR bands for imide I (anti-symmetric C=O stretch) and imide II (symmetric Characteristic IR bands for imide −1I (anti-symmetric C=O stretch) and imide II (symmetric C=O stretch) C=O stretch) are seen at 1800 cm −1−and 1730 cm−1−1, respectively. Imide III (C–N stretch) is also present −1 and 1 , respectively. C=O stretch) are seen at 1800 cm and 1730 cm , respectively. Imide III (C–N are seen at 1800 cm 1730 cm Imide III (C–N stretch) is alsostretch) presentisatalso 1385present cm−1 at 1385 cm−1−1 in coated clays, which confirms the deposition of PEI and PI on clay. Detailed peak at coated 1385 cm in which coatedconfirms clays, which confirms of thePEI deposition PEI Detailed and PI on clay. Detailed peak in clays, the deposition and PI onofclay. peak assignments are assignments are listed in Table 1. XRD patterns of the coated clay powders indicate intercalated assignments listedpatterns in Tableof1.the XRD patterns the coated clayintercalated powders indicate intercalated listed in Tableare 1. XRD coated clay of powders indicate structures in both structures in both cases (Figure 2c). This is expected based on the above discussions and SEM structures in2c). both cases (Figure based 2c). This is expected based onand theSEM above discussionsconfirming and SEM cases (Figure This is expected on the above discussions micrographs, micrographs, confirming the rather thick coverage of polymer on clay stacks. Observed d001 spacing micrographs, thepolymer rather thick coverage polymer don stacks. d001PEI-clay, spacing the rather thickconfirming coverage of on clay stacks.of Observed spacing forObserved organoclay, 001clay for organoclay, PEI-clay, and PI-clay are 17.6 Å, 18 Å, and 14.2 Å, respectively. It should be noted that for organoclay, PEI-clay, PI-clay 17.6 Å, 18 Å, and 14.2 Å,berespectively. It should be noted that and PI-clay are 17.6 Å, 18 and Å, and 14.2are Å, respectively. It should noted that equivalent amounts of equivalent amounts of clay and (coating) polymer (i.e., 1:1 wt % ratio) are employed here, which is equivalent amounts of clay and1:1(coating) polymer (i.e., 1:1 here, wt %which ratio) is are employed clay and (coating) polymer (i.e., wt % ratio) are employed possibly one here, of thewhich reasonsis possibly one of the reasons for this behavior. possibly one of the reasons for this behavior. for this behavior.

Figure 2. FTIR spectra of (a) organoclay, PEI, and PEI-coated clay; (b) organoclay, PI, and PI-coated Figure2.2.FTIR FTIRspectra spectra of of (a) (a)organoclay, organoclay,PEI, PEI,and andPEI-coated PEI-coatedclay; clay;(b) (b)organoclay, organoclay,PI, PI,and andPI-coated PI-coated Figure clay; and (c) XRD patterns of organoclay, PEI- and PI-coated clay powders. clay;and and(c) (c)XRD XRDpatterns patternsof oforganoclay, organoclay,PEIPEI-and andPI-coated PI-coatedclay claypowders. powders. clay;

Materials 2017, 10, 935 Materials 2017, 10, 935

4 of 10 4 of 10

Table 1. FTIR bands assignment for organoclay and coating polymers (PI, PEI) [20,21].

Material Absorption Range (cm−1) and coating polymers Assignment Table 1. FTIR bands assignment for organoclay (PI, PEI) [20,21]. Organoclay Material Organoclay Organoclay Organoclay Organoclay PI, PEI Organoclay PI, PEI PI, PEI PI, PEI PI, PEI PI, PEI PI, PEI PI, PEI PEI PI,PEI PEI PEI PI, PEI PI, PEI Organoclay Organoclay PI, PEI PEI

3650 Absorption Range (cm−1 ) 3250–3500

2950–2830 3650 3250–3500 1800 2950–2830 1740–1720 1800 1610–1600 1740–1720 1500–1490 1610–1600 1500–1490 1390–1380 1390–1380 1250 1250 1230–1210 1230–1210 1150–1050 1150–1050 ~730 ~730

Al–Al–OH in MMT, stretch Assignment –OH in MMT, stretch –CH3, –CH 2– aliphatic carbon, Al–Al–OH in MMT, stretchstretch in MMT, stretch Imide–OH I C=O, anti-sym. stretch –CHImide aliphatic carbon, stretch 3 , –CHII 2 –C=O, sym. stretch Imide I C=O, anti-sym. stretch C=C ImideinIIaromatic, C=O, sym.stretch stretch aromatic ring stretch C=C in aromatic, stretch aromatic ring stretch Imide III C–N, stretch Imide III C–N, stretch Tertiary C aliphatic Tertiary C aliphatic C–O–C aromatic ether C–O–C aromatic ether Si–O–Si stretch Si–O–Siin inMMT, MMT, stretch Imide deformation ImideIV IVC–N C–Nring ring deformation

2.2.Thermal ThermalDecomposition DecompositionBehavior BehaviorofofCoated CoatedClays Claysand andTheir TheirPA6 PA6 Nanocomposites 2.2. Nanocomposites Thermalstability stabilityofofthe thedifferent differentclay claypowders powderswas wasanalyzed analyzedusing usingTGA TGA(thermogravimetric (thermogravimetric Thermal analysis)under underoxidative oxidativeconditions conditions(Figure (Figure3).3).With Withorganoclay, organoclay,asasreported reportedininmany manystudies, studies,major major analysis) ◦ mass loss is seen between 200–350 °C, which is attributed to the loss of excess and surface adsorbed mass loss is seen between 200–350 C, which is attributed to the loss of excess and surface adsorbed surfactantmolecules. molecules.Upon Uponcontinued continuedheating, heating,the theintercalated intercalatedsurfactant surfactantdegrades degrades(under (underconfined confined surfactant conditionsofofcollapsed collapsedplatelets) platelets)totoform formcarbonaceous carbonaceousresidue. residue.This Thisresidue residueoxidizes oxidizescompletely completely conditions above550 550◦ C. °C.InInthe thecase caseofofneat neatPEI, PEI,nonomass massloss lossisisseen seenupuptoto535 535◦ C, °C,whereas whereas(synthesized) (synthesized)PIPI above ◦ displaysa amass massloss lossofofaround around14% 14%between between180–300 180–300 C. °C.This Thisisisprimarily primarilydue duetotothe theimidization imidizationofof displays residual PAA with the evolution of water (this is validated by the presence of the IR bandfor forPIPIatat residual PAA with the evolution of water (this is validated by the presence of the IR band − 1 −1 1650cm cm in inFigure Figure2b, 2b,which whichisisassociated associatedwith withcarbonyl carbonylstretch stretchfrom fromamic amicacid). acid). 1650

Figure Thermogravimetriccurves curvesunder underoxidative oxidativeconditions: conditions:(a) (a)organoclay, organoclay,PEI, PEI,and andPEI-coated PEI-coated Figure 3.3.Thermogravimetric clay;and and(b) (b)organoclay, organoclay, and PI-coated clay. clay; PI,PI, and PI-coated clay.

Withboth bothPEIPEI-and and PI-coated silicate platelets/stacks,aagradual gradualmass massloss lossisisseen seenstarting startingfrom from With PI-coated silicate platelets/stacks, 100◦ C°Cuntil until 450 The mass associated is around 16%PEI-coated for PEI-coated clay11% and ◦ C.°C. 100 450 The mass lossloss associated withwith this this stepstep is around 16% for clay and 11% for PI-coated clay. This behavior can be attributed to the continuous loss of fragmented for PI-coated clay. This behavior can be attributed to the continuous loss of fragmented surfactant surfactantasmolecules as a consequence of solution treatmentbyexperienced organoclay. Both PEImolecules a consequence of solution treatment experienced organoclay. by Both PEI- and PI-coated and display PI-coated display an overlap majorwith mass lossrespective peaks with their polymers respectivebeyond coating clays an clays overlap of major mass lossofpeaks their coating polymers beyond 530 °C. The key point to note here is the good thermal stability of PEIand PI◦ 530 C. The key point to note here is the good thermal stability of PEI- and PI-coated clay platelets, coated clay platelets, which effectively retards the direct chemical interactions between the clay which effectively retards the direct chemical interactions between the clay surface and the neighboring surface and PA6 the matrix. neighboring matrix.onset Thisdecomposition is evident from the higher(Tonset decomposing This isdecomposing evident fromPA6 the higher temperatures 5% ) decomposition temperatures (T 5%) for PA6/PEI-clay and PA6/PI-clay composites, as compared to for PA6/PEI-clay and PA6/PI-clay composites, as compared to those of the PA6/organoclay system those 4of(inset) the PA6/organoclay system there (Figure 4 (inset) and difference Table 2). However, there is nosuggesting significant (Figure and Table 2). However, is no significant in their T50% values, difference in their T 50% values, suggesting a similar rate of mass loss post-decomposition onset. a similar rate of mass loss post-decomposition onset.

Materials2017, 2017,10, 10,935 935 Materials

of10 10 55of

Figure 4. Thermogravimetric curves under oxidative conditions for neat PA6 and its clay composites. Figure 4. Thermogravimetric curves under oxidative conditions for neat PA6 and its clay composites. Table , T 50%,,and and TTPPfrom fromthermogravimetric thermogravimetric analysis analysis thermograms thermograms for for PA6 PA6 and Table2.2.TT5% 5%, T50% and its its composites. composites.

Specimen Specimen PA6 PA6 PA6/organoclay PA6/organoclay PA6/PEI-clayPA6/PEI-clay PA6/PI-clay PA6/PI-clay

TT5% (◦ C) 5%(°C) 353 353 360 360 380 380 384 384

◦ T50% (◦ C) T50%T(°C) P ( C)

455 445 452 454

455462 445451 452458 454453

TP (°C) 462 451 458 453

2.3. 2.3. Combustion Combustion Response Response of of PA6 PA6with withCoated CoatedClays Clays PA6/organoclay nanocompositeshows showsearly earlyignition ignitionbyby almost as compared to neat PA6/organoclay nanocomposite almost 59 59 s ass compared to neat PA6 PA6 (Figure 5 and Table 3). Coated clay composites display significant improvement in ignition (Figure 5 and Table 3). Coated clay composites display significant improvement in ignition resistance, resistance, and their ignition times to areneat similar neat PA6. This that the clay-matrix and their ignition times are similar PA6.toThis confirms that confirms the clay-matrix interaction (as interaction (asour proposed in our previous investigation [11]), a combined effect of Brønsted proposed in previous investigation [11]), a combined effect of Brønsted and Lewis and acid Lewis acid characteristics metal ions,aplays a major rolethe in the ignition behavior characteristics associated associated with the with metaltheions, plays major role in ignition behavior of of polymer/clay nanocomposites. That during pre-ignitionstage stageofofPA6 PA6 in in particular, polymer/clay nanocomposites. That is, is, during thethepre-ignition particular, the the chemical and PI prevent clayclay fromfrom interacting with the decomposing PA6. This allows chemicalinertness inertnessofofPEI PEI and PI prevent interacting with the decomposing PA6. This the matrix to continue to follow the peroxy-based decomposition pathway under thermo-oxidative allows the matrix to continue to follow the peroxy-based decomposition pathway under thermoconditions, similar to neat PA6.toAs a result, theatime-to-ignition of PEI- and PI-coated clay composites oxidative conditions, similar neat PA6. As result, the time-to-ignition of PEI- and PI-coated clay is similar to that of neat PA6. In contrast, we have previously shown that when the clay is exposed composites is similar to that of neat PA6. In contrast, we have previously shown that when the clay to the decomposing PA6 matrix,PA6 metal ions metal (fromions the clay) preferential reactivity reactivity towards is exposed to the decomposing matrix, (from have the clay) have preferential peroxy/alkoxy groups. Metal ionsMetal can either oxidize by oxidize reducingbythe hydroperoxide group into agroup more towards peroxy/alkoxy groups. ions can either reducing the hydroperoxide reactive peroxy group, or reduce by oxidizing the hydroperoxide group into another reactive alkoxy into a more reactive peroxy group, or reduce by oxidizing the hydroperoxide group into another group. ultimately result in alternate pathways for the decomposition of PA6. reactiveThis alkoxy group.could This ultimately could result in alternate pathways for the decomposition of

PA6.

Materials 2017, 10, 935

6 of 10

Materials 2017, 10, 935

6 of 10

Figure 5. release Heat release rate (HRR) forand PA6 its composites, as obtained a cone Figure 5. Heat rate (HRR) curvescurves for PA6 itsand composites, as obtained from a from cone calorimeter exposed a heat flux2 .of 35 kW/m2. whencalorimeter exposed towhen a heat flux ofto35 kW/m Table 3. Cone calorimeter results of neat PA6 and its composites.

Table 3. Cone calorimeter results of neat PA6 and its composites. Specimen TTI (s) pHRR (kW/m2) THR (MJ/m2) Char (%) TSP at Ignition (m2) SpecimenPA6 TTI (s)140 pHRR (kW/m Char (%) THR 97 (MJ/m2 ) TSP at Ignition (m2 ) 840 2 ) 0 0.58 PA6/organoclay140 81 338 142 13.1 0 0.70 0.58 PA6 840 97 PA6/PEI-clay 81 133 554 122 7.613.1 0.13 0.70 PA6/organoclay 338 142 PA6/PI-clay 133 139 396 128 9.4 7.6 0.47 0.13 PA6/PEI-clay 554 122 PA6/PI-clay 139 pHRR: peak values 396 of HRR; THR: total 128 heat released; 9.4 0.47 TTI: time-to-ignition; TSP: total smoke produced. TTI: time-to-ignition; pHRR: peak values of HRR; THR: total heat released; TSP: total smoke produced.

Apart from the significant positive effect on the time-to-ignition, other advantages of coated clay systems include the lowerpositive THR and smoke generation in comparison with PA6/organoclay Apart from the significant effect on the time-to-ignition, other advantages of coated clay composite. The PA6/PEI-clay composite also shows a 34% reduction in peak HRR value compared to systems include the lower THR and smoke generation in comparison with PA6/organoclay composite. neat PA6, whereas the PA6/PI-clay composite shows a 52% reduction, very similar to the 60% The PA6/PEI-clay composite also shows a 34% reduction in peak HRR value compared to neat PA6, reduction for the PA6/organoclay nanocomposite. The variations in pHRR follow a pattern that can whereas the PA6/PI-clay composite shows a 52% reduction, very similar to the 60% reduction for the be correlated back to the coating uniformity, dispersion, and distribution of coated clay PA6/organoclay nanocomposite. TheTo variations in pHRR follow a pattern that can be correlated stacks/platelets in the PA6 matrix. support this argument, TEM micrographs showing the extent back to theofcoating uniformity, dispersion, and distribution of coated clay stacks/platelets in the matrix. the dispersion of clay platelets are shown in Figure 6. The micrographs clearly reveal thatPA6 in the To support this argument, showing the extent theindispersion clayfinely platelets PA6/PEI-clay composite,TEM manymicrographs clusters/stacks of clay platelets are of seen addition toof some dispersed clay. As6. discussed in many studies and distribution of high are shown in Figure The micrographs clearlybefore reveal[22–24], that indispersion the PA6/PEI-clay composite, many aspect ratio nanoparticles like clay are key parameters for obtaining significant reductions in the peak clusters/stacks of clay platelets are seen in addition to some finely dispersed clay. As discussed in Thisbefore is due[22–24], to the dependence onand the physical barrier by collapsing clay plateletslike during manyHRR. studies dispersion distribution of formed high aspect ratio nanoparticles clay are combustion. However, it is important to note that the dispersion/distribution of clay in PA6 is also on key parameters for obtaining significant reductions in the peak HRR. This is due to the dependence influenced by the affinity of PEI and PI to PA6. On a similar note, Morgan et al. [25] synthesized the physical barrier formed by collapsing clay platelets during combustion. However, it is important PEI/clay nanocomposites via in situ imidization of PAA (for PEI) and with organically modified clay to note that the dispersion/distribution of clay in PA6 is also influenced by the affinity of PEI and (with an aliphatic chain of twelve carbons). Even with 10 wt % clay loading (as opposed to 50 wt % PI to clay PA6.loading On a similar note, Morgan [25] synthesized PEI/clay nanocomposites viawas in situ in the current system), et theal.magnitude of intercalation observed was poor. It imidization of PAA and with organically modified clay (with anPEI aliphatic concluded that in(for situPEI) imidization produced a mixed dispersion state in the matrix. chain of twelve carbons). Smoke Even with 10 wt % clay loading opposedduring to 50 combustion, wt % clay loading in the current system), generation is another critical(as parameter especially during the early stages that of areintercalation key for safe evacuation. Whenpoor. compared PA6, PA6 composites with PEI-clayproduced and the magnitude observed was It waswith concluded that in situ imidization PI-clay producestate significantly lessmatrix. smoke during the pre-ignition stages, as seen from Table 3. The a mixed dispersion in the PEI PA6/organoclay composite rapidlycritical generates smoke even prior tocombustion, ignition. This further corroborates Smoke generation is another parameter during especially during the the effectiveness of coated clays. early stages that are key for safe evacuation. When compared with PA6, PA6 composites with

PEI-clay and PI-clay produce significantly less smoke during the pre-ignition stages, as seen from Table 3. The PA6/organoclay composite rapidly generates smoke even prior to ignition. This further corroborates the effectiveness of coated clays.

Materials 2017, 10, 935

7 of 10

Materials 2017, 10, 935

7 of 10

Materials 2017, 10, 935

7 of 10

Figure 6. Representative transmission electron microscope images of: (a) PA6/organoclay; (b)

Figure 6. PA6/PEI-clay; Representative transmission electron microscope images of: and (c) PA6/PI-clay. (b) PA6/PEI-clay; and (c) PA6/PI-clay.

(a) PA6/organoclay;

3. Materials and Methods

3. Materials Methods 3.1.and Materials Figure 6. Representative transmission electron microscope images of: (a) PA6/organoclay; (b)

Ultramid B3S (trade name PA6) with a melting point of 220 °C and a V-2 rating in UL94 (vertical 3.1. Materials PA6/PEI-clay; and (c) PA6/PI-clay.

burning test) at a thickness of 1.6 mm was supplied by BASF, Singapore. Prior to melt compounding,

◦ C and PA6 was dried at 90 name °C for 24PA6) h in awith convection oven. Montmorillonite organically Ultramid B3S (trade a melting point of 220(clay), a V-2modified rating in UL94 3. Materials and tallow, Methods with methyl, bis-2-hydroxyethyl quaternary ammonium salt, commercially known as (vertical burning test) at a thickness of 1.6 mm was supplied by BASF, Singapore. Prior to melt Cloisite® 30B or C30B, was procured from BYK-Chemie GmbH (Singapore). Commercially available ◦ compounding, PA6 wasULTEM dried at 90PEI C was for 24 h in afrom convection oven. Montmorillonite (clay), 3.1.solvent-soluble Materials 1000 obtained SABIC Innovative Plastics (Singapore). It has aorganically modified with methyl, tallow, bis-2-hydroxyethyl quaternary ammonium salt, commercially known as reported T g of ~217 °C and a melting range of 320–355 °C. However, for the PI coating, as it is Ultramid B3S (trade name PA6) with a melting point of 220 °C and a V-2 rating in UL94known (vertical ® for its insolubility, poly(pyromellitic dianhydride-co-4,4′-oxydianiline)amic acid (PAA) solution, a Cloisiteburning 30B ortest) C30B, was procured from GmbH (Singapore). Commercially available at a thickness of 1.6 mm wasBYK-Chemie supplied by BASF, Singapore. Prior to melt compounding, precursor for PI, was obtained from Sigma-Aldrich (Singapore) at a concentration of 15 ± 5 wt % in PA6 was dried at 901000 °C forPEI 24 hwas in a obtained convectionfrom oven. SABIC Montmorillonite (clay), organically modified It has solvent-soluble ULTEM Innovative Plastics (Singapore). N-methyl-2-pyrrolidone/aromatic hydrocarbons (80%/20% solvent ratio). This was followed by ◦ C and ◦ C. However, withTmethyl, tallow, bis-2-hydroxyethyl quaternary ammonium salt, commercially as as it is a reported a melting range of 320–355 for the PIknown coating, g of ~217 imidization, resulting in PI. PI, thus formed, had a T g > 300 °C. It does not melt or decompose at Cloisite® 30B or C30B, was procured from BYK-Chemie GmbH 0(Singapore). Commercially available known fortemperatures its insolubility, dianhydride-co-4,4 -oxydianiline)amic (PAA) belowpoly(pyromellitic 540 °C. Other solvents required for coating such as tetrahydrofuranacid (THF) and solution, solvent-soluble ULTEM 1000 PEI was obtained from SABIC Innovative Plastics (Singapore). It has a N, for N-dimethylacetamide (DMAc) were procured from Merck Milliporeat (Singapore). a precursor PI, was obtained from Sigma-Aldrich (Singapore) a concentration of 15 ± 5 wt % reported Tg of ~217 °C and a melting range of 320–355 °C. However, for the PI coating, as it is known in N-methyl-2-pyrrolidone/aromatic hydrocarbons (80%/20% solvent ratio). This was followed by for 3.2. its Synthesis insolubility, poly(pyromellitic acid (PAA) solution, a of Coated Clays: PEI-Claydianhydride-co-4,4′-oxydianiline)amic and PI-Clay ◦ imidization, resulting in PI. PI, thus had a(Singapore) Tg > 300 atC.a concentration It does not melt decompose at precursor for PI, was obtained fromformed, Sigma-Aldrich of 15 ±or5 wt % in A schematic ◦flowsheet of detailed step-by-step procedure that was followed to obtain PEI- and N-methyl-2-pyrrolidone/aromatic hydrocarbons (80%/20% solvent ratio). This was followed by and N, temperatures below 540 C. Other solvents required for coating such as tetrahydrofuran (THF) PI-coated clay powders is given in Figure 7. In general, a suspension of C30B in organic solvent was imidization, resulting in PI. PI, thus formed, had aMerck Tg > 300 °C. It does not melt or decompose at N-dimethylacetamide (DMAc) (Singapore). mixed with dilute solutionwere of PEIprocured (or PAA) tofrom achieve 1:1 wt.Millipore ratio, followed by precipitation in non-

temperatures below 540 °C. Other solvents required for coating such as tetrahydrofuran (THF) and solvent (deionized water). An additional step of imidization was carried out prior to filtration in the

N, case N-dimethylacetamide (DMAc) were procured from Merck (Singapore). of Coated the PI-coating. and PI-clay, thus obtained, wereMillipore used for further characterization. To 3.2. Synthesis of Clays:PEI-clay PEI-Clay and PI-Clay achieve the same degree of physical barrier effect across all composites (as influenced by the loading

3.2.level Synthesis of Coated Clays: PEI-Clay and PI-Clay A schematic flowsheet ofthe detailed procedure that wasAccordingly, followed 5towtobtain of clay platelets), loading step-by-step level of base clay was kept constant. % of PEI- and organoclay and 10 wt % of coated clays were used in the fabrication of PA6/clay nanocomposites. PI-coated clay powders is given in Figure 7. In general, a suspension of C30B in organic A schematic flowsheet of detailed step-by-step procedure that was followed to obtain PEI- and solvent PI-coated powders is given in Figure 7. In general, a suspension of C30Bfollowed in organicby solvent was was mixed withclay dilute solution of PEI (or PAA) to achieve 1:1 wt. ratio, precipitation in mixed(deionized with dilute solution PEIadditional (or PAA) tostep achieve 1:1 wt. ratio, followed by precipitation in nonnon-solvent water).ofAn of imidization was carried out prior to filtration in water). An additional step of imidization was carried out prior to filtration in the the casesolvent of the (deionized PI-coating. PEI-clay and PI-clay, thus obtained, were used for further characterization. case of the PI-coating. PEI-clay and PI-clay, thus obtained, were used for further characterization. To To achieve the same degree of physical barrier effect across all composites (as influenced by the loading achieve the same degree of physical barrier effect across all composites (as influenced by the loading level oflevel clayofplatelets), the the loading level waskept kept constant. Accordingly, clay platelets), loading levelofofbase base clay clay was constant. Accordingly, 5 wt %5ofwt % of organoclay and 10and wt10%wt of%coated clays were thefabrication fabrication of PA6/clay nanocomposites. organoclay of coated clays wereused used in the of PA6/clay nanocomposites.

Figure 7. Cont.

Materials 2017, 10, 935 Materials 2017, 10, 935

8 of 10 8 of 10

Figure 7. Step-by-step synthesis procedure: (a) (a) PEI-clay PEI-clay platelets platelets by by solution-precipitation solution-precipitation technique; technique; and (b) PI-clay platelets by solution-imidization-precipitation solution-imidization-precipitation technique. technique.

3.3. 3.3. Extrusion Extrusion Parameters Parameters Melt was carried carried out out using using the Melt mixing mixing of of coated coated clays clays with with PA6 PA6 was the Leistritz Leistritz Micro Micro 18 18 twin twin screw screw extruder (Singapore) at 120 rpm to assist the uniform dispersion of clay platelets/stacks. The extrusion extruder (Singapore) at 120 rpm to assist the uniform dispersion of clay platelets/stacks. The extrusion ◦ C (die°C temperature profileused used was: (hopper 230–235–240–245–245–250 (diePelletized end). Pelletized temperature profile was: (hopper end)end) 230–235–240–245–245–250 end). material material was compression molded mm × 100 5 mmsamples sized samples a Carver Hot was compression molded into 100 into mm 100 × 100 mm × 5mm mm× sized using using a Carver Hot Press ◦ Press (Wabash, IN, USA.) maintained at 250 °C with a molding pressure of 5 bar and a total molding (Wabash, IN, USA.) maintained at 250 C with a molding pressure of 5 bar and a total molding time of time of 20 min. 20 min. 3.4. Characterization X-ray diffraction studies of clay powders and molded polymer composites were carried out on ◦ /min, a Shimadzhu XRD 6000 (Singapore, 40 kV, 30 30 mA, mA, Cu Cu kα, kα, λλ== 1.542 1.542 Å) Å) with with aa scan scan speed speed of of 22°/min, scan range of 3–45°, 3–45◦ , and and step step size size of of 0.02°. 0.02◦ . Fourier Fouriertransform transforminfrared infraredspectra spectraof ofPA6 PA6 materials, materials, clay, clay, and powdered powderedresidues residues (prepared KBr pellets) were using collected using a Perkin-Elmer (prepared usingusing KBr pellets) were collected a Perkin-Elmer SpectrumGX SpectrumGX (Singapore) and Perkin-Elmer Frontier Spectrometer (Singapore, surface spectrometer spectrometer (Singapore) and Perkin-Elmer Frontier Spectrometer (Singapore, surface Attenuated − 1 −1 Attenuated Total Reflectance (ATR) FTIR). All spectra were acquired in the range of 400–4000 cm32 Total Reflectance (ATR) FTIR). All spectra were acquired in the range of 400–4000 cm using − 1 −1 using 32 scans and resolution 4 cm . To clay observe clay dispersion in the polymer matrix, trimmed scans and resolution of 4 cm of . To observe dispersion in the polymer matrix, trimmed polymer polymer samples were ultra-microtomed 0.2 mm/s usingknife a diamond on Leica compositecomposite samples were ultra-microtomed at 0.2 mm/satusing a diamond on Leicaknife Ultracut UCT ◦ Ultracut UCT in liquid N2at environment at –10070–90-nm °C to obtain 70–90-nm sections.were The (Singapore) in (Singapore) liquid N2 environment –100 C to obtain thick sections.thick The sections sectionsup were picked up using a droplet of 2.3and molplaced sucrose placed on formvar/carbon-coated 400picked using a droplet of 2.3 mol sucrose onand formvar/carbon-coated 400-mesh copper mesh grids. After thorough rinsing water with distilled to wash away the sucrose, sections grids. copper After thorough rinsing with distilled to wash water away the sucrose, sections were observed were using a Carl Zeiss transmission electron microscope (Singapore), LIBRA 120, in bright usingobserved a Carl Zeiss transmission electron microscope (Singapore), LIBRA 120, in bright field mode, at an field mode, atvoltage an accelerating of 120 kV. The morphology uncoated claysusing was accelerating of 120 kV.voltage The morphology of uncoated and of coated claysand wascoated observed observed using a JEOL field emission scanning electron microscope, FESEM 6340F (Singapore) with a JEOL field emission scanning electron microscope, FESEM 6340F (Singapore) with an accelerating an accelerating 10–20 kV and a between working 7–9 distance between 7–9were mm.sputter-coated All samples were voltage of 10–20voltage kV and of a working distance mm. All samples with sputter-coated 10 mm s, at from a distance of 30 and mmwith fromathe source with a voltage of platinum for 10with s, at platinum a distancefor of 30 the source voltage of and 20 kV. 20 kV.

Materials 2017, 10, 935

9 of 10

3.5. Thermo-Oxidation and Combustion Tests Thermogravimetric analysis (TGA) for different clay powders and polymer composites was carried out on TA Instruments Q500 (Singapore) under air atmosphere. Clay samples were pre-dried at 110 ◦ C, while polymer samples were dried at 80 ◦ C for 24 h in a convection oven. Tests were carried out from room temperature to 750 ◦ C employing a heating rate of 20 ◦ C/min. A cone calorimeter from Fire Testing Technology (East Grinstead, West Sussex, UK) was used to determine the combustion response of compression molded PA6 and its composites. The tests were carried out according to ISO 5660 at an irradiant heat flux of 35 kW/m2 . 4. Conclusions To retard the catalytic effect of clay on the time-to-ignition of a PA6 matrix, a methodology that prevented direct chemical interactions between the clay surface and the neighboring decomposing PA6 matrix was employed here. This involved the coating of clay with thermally stable polymers such as PEI and PI by solution blending-precipitation and solution blending-imidization-precipitation techniques, respectively. The thermal decomposition and combustion behavior of the composites confirmed the delayed decomposition onset as well as the time-to-ignition for coated clay composites. Additionally, up to 52% reduction in peak HRR was noted for coated clay composites relative to neat PA6, maintaining the advantage of the presence of clay platelets. This work proved that preventing direct contact between the clay surface and the decomposing PA6 matrix using a physical barrier such as a high-temperature-resistant polymer coating is a simple solution to the major issue of early ignition with polymer/clay nanocomposites. Acknowledgments: The authors acknowledge JTC Corporation and National Research Foundation (NRF), Singapore, for supporting projects that deal with flame retardancy of polymer coatings through NTU-JTC Industrial Infrastructure Innovation Centre (RCA-16/277) and L2NIC grant (Award No.: L2NICCFPI-2013-4), respectively. Author Contributions: Indraneel S. Zope and Aravind Dasari conceived and designed the experiments; Indraneel S. Zope performed the experiments; Indraneel S. Zope and Aravind Dasari analyzed the data; Zhong-Zhen Yu contributed with cone calorimeter tests. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Kiliaris, P.; Papaspyrides, C.D. Polymer/layered silicate (clay) nanocomposites: An overview of flame retardancy. Prog. Polym. Sci. 2010, 35, 902–958. [CrossRef] Dasari, A.; Yu, Z.-Z.; Cai, G.-P.; Mai, Y.-W. Recent developments in the fire retardancy of polymeric materials. Prog. Polym. Sci. 2013, 38, 1357–1387. [CrossRef] Zhu, J.; Start, P.; Mauritz, K.A.; Wilkie, C.A. Thermal stability and flame retardancy of poly(methyl methacrylate)-clay nanocomposites. Polym. Degrad. Stab. 2002, 77, 253–258. [CrossRef] Liang, Z.M.; Wan, C.Y.; Zhang, Y.; Wei, P.; Yin, J. PVC/Montmorillonite nanocomposites based on a thermally stable, rigid-rod aromatic amine modifier. J. Appl. Polym. Sci. 2004, 92, 567–575. [CrossRef] Humphrey, J.; Boyd, D. Clay: Types, Properties and Uses; Nova Science Publishers, Inc.: New York, NY, USA, 2011. Nawani, P.; Gelfer, M.Y.; Hsiao, B.S.; Frenkel, A.; Gilman, J.W.; Khalid, S. Surface Modification of Nanoclays by Catalytically Active Transition Metal Ions. Langmuir 2007, 23, 9808–9815. [CrossRef] [PubMed] Zope, I.S.; Dasari, A.; Camino, G. Elucidating the catalytic effect of metal ions in montmorillonite on thermal degradation of organic modifier. Mater. Chem. Phys. 2015, 157, 69–79. [CrossRef] Reddy, C.R.; Bhat, Y.S.; Nagendrappa, G.; Jai Prakash, B.S. Brønsted and Lewis acidity of modified montmorillonite clay catalysts determined by FT-IR spectroscopy. Catal. Today 2009, 141, 157–160. [CrossRef] Reddy, C.R.; Nagendrappa, G.; Jai Prakash, B.S. Surface acidity study of Mn+ -montmorillonite clay catalysts by FT-IR spectroscopy: Correlation with esterification activity. Catal. Commun. 2007, 8, 241–246. [CrossRef] Breen, C.; Moronta, A. Characterization and Catalytic Activity of Aluminum- and Aluminum/Tetramethylammonium-exchanged Bentonites. J. Phys. Chem. B 2000, 104, 2702–2708. [CrossRef]

Materials 2017, 10, 935

11. 12. 13.

14. 15. 16.

17.

18. 19. 20. 21.

22.

23. 24.

25.

10 of 10

Zope, I.S.; Dasari, A.; Guan, F.-L.; Yu, Z.-Z. Influence of metal ions on thermo-oxidative stability and combustion response of polyamide 6/clay nanocomposites. Polymer 2016, 92, 102–113. [CrossRef] Tang, T.; Chen, X.; Meng, X.; Chen, H.; Ding, Y. Synthesis of Multiwalled Carbon Nanotubes by Catalytic Combustion of Polypropylene. Angew. Chem. Int. Ed. 2005, 44, 1517–1520. [CrossRef] [PubMed] Tang, T.; Chen, X.; Chen, H.; Meng, X.; Jiang, Z.; Bi, W. Catalyzing Carbonization of Polypropylene Itself by Supported Nickel Catalyst during Combustion of Polypropylene/Clay Nanocomposite for Improving Fire Retardancy. Chem. Mater. 2005, 17, 2799–2802. [CrossRef] Hu, Y.; Wang, X.; Li, J. Regulating Effect of Exfoliated Clay on Intumescent Char Structure and Flame Retardancy of Polypropylene Composites. Ind. Eng. Chem. Res. 2016, 55, 5892–5901. [CrossRef] Pappalardo, S.; Russo, P.; Acierno, D.; Rabe, S.; Schartel, B. The synergistic effect of organically modified sepiolite in intumescent flame retardant polypropylene. Eur. Polym. J. 2016, 76, 196–207. [CrossRef] Walther, A.; Bjurhager, I.; Malho, J.-M.; Pere, J.; Ruokolainen, J.; Berglund, L.A.; Ikkala, O. Large-Area, Lightweight and Thick Biomimetic Composites with Superior Material Properties via Fast, Economic, and Green Pathways. Nano Lett. 2010, 10, 2742–2748. [CrossRef] [PubMed] Zang, L.; Qiu, J.; Yang, C.; Sakai, E. Preparation and application of conducting polymer/Ag/clay composite nanoparticles formed by in-situ UV-induced dispersion polymerization. Sci. Rep. 2016, 6, 20470. [CrossRef] [PubMed] Chaiko, D.J. Process for the preparation of organoclays. U.S. Patent 6,521,678, 18 February 2003. Tyan, H.-L.; Liu, Y.-C.; Wei, K.-H. Enhancement of imidization of poly(amic acid) through forming poly(amic acid)/organoclay nanocomposites. Polymer 1999, 40, 4877–4886. [CrossRef] Fan, H.; Yang, R. Flame-Retardant Polyimide Cross-Linked with Polyhedral Oligomeric Octa(amino-phenyl)silsesquioxane. Ind. Eng. Chem. Res. 2013, 52, 2493–2500. [CrossRef] Cervantes-Uc, J.M.; Cauich-Rodríguez, J.V.; Vázquez-Torres, H.; Garfias-Mesías, L.F.; Paul, D.R. Thermal degradation of commercially available organoclays studied by TGA–FTIR. Thermochim. Acta 2007, 457, 92–102. [CrossRef] Liu, M.; Zhang, X.; Zammarano, M.; Gilman, J.W.; Davis, R.D.; Kashiwagi, T. Effect of montmorillonite dispersion on flammability properties of poly(styrene-co-acrylonitrile) nanocomposites. Polymer 2011, 52, 3092–3103. [CrossRef] Samyn, F.; Bourbigot, S.; Jama, C.; Bellayer, S. Fire retardancy of polymer clay nanocomposites: Is there an influence of the nanomorphology? Polym. Degrad. Stab. 2008, 93, 2019–2024. [CrossRef] Wang, Z.; Du, X.; Yu, H.; Jiang, Z.; Liu, J.; Tang, T. Mechanism on flame retardancy of polystyrene/clay composites- The effect of surfactants and aggregate state of organoclay. Polymer 2009, 50, 5794–5802. [CrossRef] Morgan, A.B.; Gilman, J.W.; Jackson, C.L. Characterization of the Dispersion of Clay in a Polyetherimide Nanocomposite. Macromolecules 2001, 34, 2735–2738. [CrossRef] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).