Synergistic effect of carbon nanotube on improving thermal stability, flame retardancy, and electrical conductivity of poly(butylene succinate)/piperazine pyrophosphate composites

Biodegradable poly(butylene succinate) (PBS) is considered as promising material to replace conventional nondegradable polymers in various engineering fields, but the applications are limited by its inherent flammability and low electrical conductivity. In this study, the synergistic effect of carbon nanotube (CNTs) on improving thermal stability, flame retardancy, and electrical conductivity of poly(butylene succinate)/piperazine pyrophosphate (PBS/PAPP) composites was investigated. Thermogravimetric analysis (TGA) demonstrated that the addition of CNTs could improve the thermal stability of PBS/PAPP composites, where the maximum mass loss temperature (Tmax) of PBS/18PAPP-2CNT increased by 9.1 °C in comparison with that of PBS/20PAPP. Meanwhile, PBS/18PAPP-2CNT exhibited the optimal flame retardancy with limited oxygen index (LOI) of 30.1%, V0 rating in UL-94 vertical burning test, and 79.6% reduction on the peak of heat release rate (PHRR) in cone calorimeter test. The synergistic effect of CNTs and PAPP was beneficial to construct high-quality char layer, resulting in the enhanced flame retardancy of PBS. Furthermore, the CNTs could improve the electrical conductivity of PBS/PAPP composites with low percolation threshold of 1.75 wt%, which was ascribed to the “volume excluded effect” of PAPP. Thus, the current work provided an efficient strategy to prepare multifunctional PBS composites to meet various engineering applications.

The flame retardancy of PBS can been improved by adding single or combined flame retardants [17][18][19][20][21][22][23][24].For example, Kuan et al. [20] added ammonium polyphosphate (APP) into PBS vis a unique water-crosslinking technique, where PBS composites with 15 wt% APP were classified as UL-94 V0 rating after 0.5-h water-crosslinking reaction.Hu et al. [21] reported the incorporation of ethyl cellulose microencapsulated ammonium polyphosphate (MAPP) and charforming agent (CFA) into PBS.The composite exhibited the limiting oxygen index (LOI) as high as 35.5%, the UL-94 V0 rating, and the decrease on the peak heat release rate (PHRR) by 46.7%.Yue et al. [22] prepared PBS composites via incorporating cassava dregs and intumescent flame retardants (IFR), where the LOI value of 37.3% and UL-94 V0 rating could be achieved.Wang et al. [23] explored the direct use of eucommia residues (ERs) to improve the flame retardancy of PBS via a facile melt-mixing method.The carbon residue of PBS/ER composites could maximally increase by 508%; the PHRR and total heat release (THR) also reduced by 43 and 20%, respectively.In our previous study [24], aluminum diethylphosphinate (AlPi) was added into PBS.With 25 wt% AlPi loading, the PBS composite presented better flame retardancy with 29.5% for LOI, V0 rating for UL-94, and 49.3% reduction for PHRR.
To improve the electrical conductivity of polymers, conductive nanofillers are widely used [25][26][27][28][29], such as carbon nanotubes, carbon fiber, and graphene.For instance, Wang et al. [27] prepared cellulose-PBS films filled with multiwalled carbon nanotubes (MWCNTs).The electrical conductivity of MWCNT/CE-PBS film was improved by 8-9 orders of magnitude from 2.5 × 10 −14 to 1.3 × 10 −5 S m −1 .Huang et al. [28] also prepared PBS/CNT conductive polymer nanocomposites with varied CNT content prepared by a HAAKE torque rheometer.The conductivity of PBS/5 wt% CNT nanocomposite increased to 33.3 S m −1 , which was an increase of 16 orders of magnitude in comparison with pure PBS of 8.23 × 10 -15 S m −1 .Kuang et al. [29] demonstrated a facile and effective way to fabricate PBS/carbon fiber (CF) composite foams with lightweight, high-strength, and improved conductive networks through the combination of solvent mixing, micro-injection molding, and supercritical carbon dioxide (Sc-CO 2 ) foaming methods.The resulting composite foams possessed much higher electrical conductivity (the percolation threshold decreased from 3.6-7.4 to 1.04-2.37vol%), suggesting that the introduction of foaming technique could be beneficial to form effective 3D conductivity networks.
In this study, the synergistic effect of carbon nanotube (CNTs) on improving thermal stability, flame retardancy, and electrical conductivity of poly(butylene succinate)/piperazine pyrophosphate (PBS/PAPP) composites was investigated.Here, CNTs acted not only as synergistic flame retardant of PAPP, but also as nanofillers to construct conductive network in PBS composites.Thermogravimetric analysis (TGA) was used to evaluate their thermal stability in air atmosphere.Furthermore, LOI, UL-94 and cone calorimeter tests were performed to investigate their flame retardancy.Finally, the electrical conductivity was evaluated via calculating the percolation threshold of CNTs in PBS composites.

Materials
PBS (GS PLA, Japan) pellet was provided by Mitsubishi Chemical Corp.(Toyato, Japan) with the average molecular weight (M W ) of 11.8 kg mol −1 and the density of 1.26 g cm −3 .Dicumyl peroxide (DCP) and maleic anhydride (MA) were purchased from Aladdin Chemical Reagent Co., Ltd., China.Multi-walled carbon nanotubes (CNTs) with the diameter of 15 ± 5 nm and the length of approximately 30 μm were purchased from Chengdu Institute of Organic Chemistry, China.Piperazine pyrophosphate (PAPP) was purchased from JITRI Advanced Polymer Material Research Institute Co., Ltd.(Jiangsu, China).
Poly(butylene succinate) grafted maleic anhydride (PBS-g-MA) was synthesized via a melt-grafting process [30], and the resultant grafting amount of MA was about 4.22 wt%.In brief, PBS was physically premixed with 10 wt% MA and 1.5 wt% DCP; then, they were reactively blended in Haake Rheomix 600 mixer (Karlsruhe, Germany) at 135 °C for 7 min.After that, the sample was refluxed in chloroform for 4 h, and the hot solution was filtered into cold methanol.The precipitated polymer was washed with methanol several times, in order to remove any unreacted reagents followed by drying in a vacuum oven at 60 °C for 24 h.

Preparation of PBS composites
The formulations of PBS composites are shown in Table 1.PBS composites were prepared via melt compounding in Haake Rheomix 600 mixer (Karlsruhe, Germany).The processing parameters included the temperature of 135 °C, the rotor speed of 80 rpm, and the mixing time of 6 min for each sample.Here 10 wt% PBS-g-MA was added as compatibilizer, and the total content of flame retardants was kept at 20 wt% in PBS composites.For convenience, the obtained PBS composites were designated as PBS/xPAPP-yCNT, where x and y denote the weight percentage of PAPP and CNTs in PBS composites, respectively.Subsequently, the PBS samples were hot-pressed at 140 °C and 10 MPa to prepare test specimens.

Characterization
The thermal stability was tested by a TA STD Q600 thermal analyzer under air atmosphere.The sample mass was 6.0 ± 0.

Effect of CNTs on thermal stability of P BS/PAPP composites
The effect of CNTs on the thermal stability of PBS/PAPP composites was evaluated in air atmosphere, because it was directly associated with the actual processing of polymer materials.Figure 1(a) and (b) show the TGA and DTG curves of PBS samples with different CNT loadings.Compared to neat PBS, it was found that the TGA curve of PBS/20PAPP decreased to a lower temperature range, which was ascribed to the relatively low decomposition temperature of PAPP [31].However, when CNTs were added into the PBS/PAPP system, the TGA curves gradually shifted to higher temperature zone, indicating the positive effect of CNTs on improving the thermal stability of PBS composites.Table 2 summarizes their relevant thermal decomposition parameters.For PBS/20PAPP, the T 5 wt% and the T 10 wt% decreased 6.0 and 9.2 °C, respectively, in comparison with that of neat PBS.With the increase of CNTs from 0.5 to 2 wt% into PBS/PAPP, both the T 5 wt% and the T 10 wt% gradually increased.The PBS/18PAPP-2CNTs exhibited the highest values for T 5 wt% and T 10 wt% , which was 4.4 and 7.1 °C higher than that of PBS/20PAPP, respectively.Furthermore, the presence of CNTs could improve the maximum mass loss temperature (T max ) of PBS/APP composites, where the T max of PBS/18PAPP-2CNT increased by 9.1 °C in comparison of that of PBS/20PAPP.More interestingly, it was found  that the char residues at 550 °C increased.For PBS/18PAPP-2CNT, its char yield was as high as 18.2%, which is much higher than that of PBS/20PAPP with the value of 11.2 wt%.
In brief, CNTs could simultaneously improve the thermal stability and char-forming ability of PBS/PAPP composites.

Effect of CNTs on flame retardancy of PBS/PAPP composites
LOI and UL-94 tests were firstly employed to evaluate the effect of CNTs on flame retardancy of PBS/PAPP composites, and the detailed combustion parameters were listed in Table 3. Neat PBS was easily combustible with the LOI value of only 21.0%.In contrast, the LOI value increased to 27.6% after the addition of 20% PAPP into PBS.It was noted that the combination of CNTs and PAPP presented the synergistic effect on improving the LOI values, where the highest LOI from PBS/18PAPP-2CNT could reach to 30.1%.According to the results from UL-94 vertical burning test, the addition of 20 wt% PAPP could increase the rating to V2, but it still dripped heavily to ignite the cotton underneath.With the addition of 0.5 wt% CNTs, the dripping of burning droplets was effectively inhibited.The optimal sample was also from PBS/18PAPP-2CNT, which could achieve to V0 rating and no dripping, indicating CNTs as an excellent synergist to efficiently improve flame retardancy and antidripping properties of PBS/PAPP composites.
Subsequently cone calorimeter testing (CCT) was used to investigate the flame retardancy of PBS samples, which is helpful to evaluate the fire risk during combustion [32][33][34][35][36][37][38]. Figure 2(a) presents their heat release rate (HRR) curves, while detailed parameters are summarized in Table 3.With 20 wt% loading of PAPP into PBS, the ignition time (t ign ) became shorter by 16 s, implying the easier ignition of PBS/ PAPP than neat PBS.However, the t ign gradually increased via the incorporation of CNTs as synergist.After ignition, PBS burned rapidly with the PHRR as high as 765 kW m −2 .In contrast, PBS/20PAPP presented the reduction on PHRR to 397 kW m −2 (− 48.1%), indicating good flame retardant effect from PAPP.Notably, with partially substituting PAPP with CNTs, the PHRR values further decreased.For PBS/18PAPP-2CNT, the PHRR significantly decreased to 156 kW m −2 (− 79.6%).Moreover, Fig. 2(b) shows the THR curves of PBS samples.Compared to the THR of  It is statistical that the gas toxicity is mainly responsible for over 70% of people death in fire disasters [39,40], so the total smoke production (TSP) and CO production (COP) of PBS samples were evaluated.As shown in Fig. 3(a), the addition of PAPP resulted in the increase of TSP, mainly due to its trapping radical effect in gas phase.Similar to most phosphorous flame retardants, the decomposition of PAPP produced P• and PO• radicals [41], which could react with H• and HO• radicals to generate some solid smoke particles.However, the TSP gradually decreased with the increase of CNT content, indicating its significantly inhibiting effect on smoke release.Besides, the addition of PAPP into PBS increased the COP, but the combination of PAPP and CNTs together could effectively decrease the COP, suggesting that CNTs can significantly reduce the release of toxic CO gas (Fig. 3(b)).Moreover, as shown in Table 3, the combination of PAPP and CNTs increased the amount of residual char, which is well consistent with above-mentioned TGA results in air.More flammable volatiles were fixed in condensed phase zone to form solid chars, resulting in less TSP and COP in the gas phase.These results indicated that CNTs could synergistically decrease gas toxicity of PBS/ PAPP composite and get more escaping time.
To investigate the synergistic mechanism of CNTs in PBS/PAPP composites, the morphology and microstructure of residual chars after CCT were analyzed.Figure 4 presents their digital photos of char residues.It can be seen that PBS/20PAPP formed intumescent char layer, but the char layer was incompact with poor mechanical strength (Fig. 4(a)).With substituting PAPP with CNTs from 0.5 to 2 wt%, the char layers gradually became thicker without obvious collapse (Fig. 4(b), (c), and (d)).The increase of char amount and char strength could enhance heat-and oxidation-resisting ability of char layer, resulting in better flame retardancy.
Moreover, the char microstructures of PBS composites were investigated by SEM.As shown in Fig. 5(a) and (b), the surface of PBS/20PAPP was sponge-like char, which was composed of porous carbon spheres.In contrast, the surface of PBS/18PAPP-2CNT was a dense block without holes and cracks (Fig. 5(c) and (d)).According to these results, the addition of CNTs was helpful to construct a high-quality char layer with no cracks.
On the basis of above analysis, a possible enhancement mechanism of the CNTs in PBS/PAPP system was proposed.As shown in Fig. 6, PPAP pyrolyzed into metaphosphoric or phosphoric acid and their piperazine salts [41], which can further decomposed into P• and PO• radicals to quench H• and HO• radicals in gas phase region.Meanwhile, the CNTs also could trap radicals due to their graphitized structure [42].More importantly, the CNTs could promote the formation of highquality char in condensed phase region.They migrated onto the outer surface of chars during combustion; then, their aggregations could improve the mechanical strength and supporting ability of char layer.More importantly, the compact char layer could bring about better barrier effect on the isolation of heat and oxygen, delaying the diffusion of flammable gases, then inhibiting the flame propagation and the fire intensity [43].As a result, a substantial improvement on flame retardancy with 30.1% on LOI, V0 in UL-94, and 79.6% reduction on PHRR were achieved in PBS composites.

Effect of CNTs on the electrical conductivity of PBS/ PAPP composites
Electrical conductivity is connected to a conductive filler network within the polymer composite [44].It is noted that the percolation threshold content is defined as the minimum filler loading in polymer composites to achieve the transform from insulating to conductive materials [45].To accurately get the percolation threshold of CNTs in PBS/PAPP composites, some special content gradients were designed.As shown in Fig. 7(a), when the CNTs content reached to 1.75 wt%, the volume resistivity (Ω•cm) of PBS composites suddenly decreased 10 orders of magnitude from 10 18 to 10 8 , indicating that the percolation threshold content of CNTs was 1.75 wt%.However, with the addition of CNTs into neat PBS, the percolation threshold content was as high as 3.75 wt% (Fig. 7(b)).Here the ternary system exhibited lower percolation threshold content than binary system, which was ascribed to the "volume excluded effect" of PAPP, resulting in better conductive network [46].More exactly, the introduction of PAPP changed the distribution state of CNTs, where CNTs bridged each other easily to form conductive network through the spatial volume excluded effect.As a result, the lower filler loading in PBS composites was needed to achieve the insulating-conductive transform.

Fig. 1
Fig. 1 TGA a and DTG b curves of neat PBS and PBS composites in air atmosphere

Fig. 2
Fig. 2 PBS samples measured by cone calorimeter test at a radiant flux of 50 kW m. −2 : a heat release rate curves (HRR) and b total heat release curves (THR)

Fig. 3 Fig. 4
Fig. 3 PBS samples measured by cone calorimeter test at a radiant flux of 50 kW m. −2 : a total smoke production curves (TSP) and b CO production curves (COP)

Fig. 5
Fig. 5 SEM images of the reside chars: a and b PBS/20PAPP and c and d PBS/18PAPP-2CNT

Fig. 6 Fig. 7
Fig. 6 Possible enhancement mechanism of CNTs on flame retardancy in PBS/PAPP system

Table 1
The formulations of PBS composites

Table 2
Thermal decomposition properties of neat PBS and PBS composites in air atmosphere

Table 3
Combustion parameters of PBS samples from LOI, UL-94, and cone calorimeter tests NR: no rating; t ign : the ignition time; PHRR: the peak heat release rate; THR: the total heat release; TSP: the total smoke production