Carbon nanotube reinforced polylactide/basalt fiber composites containing aluminium hypophosphite: thermal degradation, flame retardancy and mechanical properties

Abstract

This work aims to develop modified carbon nanotube (m-CNT) reinforced polylactide/basalt fiber (PLA/BF) composites with enhanced thermal stability, flame retardancy, and mechanical properties using aluminium hypophosphite (AHP). Morphological observations revealed that both the m-CNT and AHP particles were homogeneously dispersed in the PLA/BF composites. The improvement of thermal stability and tensile strength/modulus strongly depended on the uniform dispersion of the mixed particles and the interactions between m-CNT and PLA. The presence of AHP and m-CNT significantly reduced the peak heat release rate of PLA/BF in microscale combustion calorimetery testing. The combustion behaviors were evaluated by limiting oxygen index (LOI), Underwriters Laboratories 94 (UL 94), and cone calorimetery. For PLA/BF containing 19 wt% AHP and 1 wt% m-CNT, it achieved a V-0 classification in UL 94 testing with a high LOI (31%). Additionally, the peak heat release rate and total heat release were respectively reduced by around 64% and 27% in comparison with PLA/BF. The residue analysis showed that many microspheres formed by the sufficient reaction between AHP and PLA under the physical barrier effect of m-CNT migrated to or accumulated on the surface of residues. The compact char layer composed of these microspheres could effectively reduce the heat conduction of basalt fibers and cut off the mass transfer path resulting in the weakening of the wick effect resulting in the significant improvement of the composites.

---

1. Introduction

Nowadays, petrol-based plastics are widely used in many fields, including automobile, aircraft, naval constructions, ships, buildings and offshore applications. However, there are more and more environmental problems resulting from their incomplete decomposition. Polylactide (PLA) is one of the most promising biopolymers due to its harmless biodegradation capability, abundant renewable sources, easy processing by conventional equipment, low shrinkage, reasonable price and excellent mechanical properties.^1–4 Due to its excellent properties, PLA has been commonly used in biomedical fields, household and packaging industries, etc.^5–9 Although the high strength and stiffness of PLA suggest it can be used as the matrix of durable engineering composite materials, its low heat deflection temperature and low impact strength still retard its use as a high performance composite.⁴

In recent years, more and more reports show that the addition of basalt fiber can significantly improve the tensile, flexural, and impact strength of various polymers, which is a novel reinforcement for composites.^4,10–14 Basalt fibers can be considered as natural fibers, because they are always produced by using common volcanic rock, which can be found in nature and every city on earth. Furthermore, basalt fibers are environmentally friendly and can be used in very harsh environments. As reinforcement materials, basalt fibers can be used from 200 to 600 °C without any significant loss of mechanical properties.^15,16 Owing to these advantages, basalt fibers can seem to be a good alternative to reinforce PLA.^4,17–19 The results from these articles indicate that the presence of basalt fibers can greatly enhance the mechanical properties of PLA. The novel composites composed of PLA and basalt fibers would be one of the high performance and durable engineering materials.

Even though the basalt fiber (BF) reinforced PLA (PLA/BF) composites can seem as one potential candidate to be used in engineering fields, the composites also show flammability which restricts its application in new areas. Recently, metal hypophosphite has been applied as a flame retardant in some polymeric materials.^20–24 Even low amount of aluminium hypophosphite (AHP) added can significantly enhance the fire retardancy of the polymers. In Tang's work,²² the flame retarded PLA was prepared using AHP as additives. It showed that the introduction of AHP could significantly improve the flame retardancy of PLA through the condensed-phase action of AHP in the combustion process. However, the PLA/BF composites are different from neat PLA. The basalt fibers are similar with glass fibers, which can further enhance the flammability of polymer matrix due to their wick effect for the propagation of the combustion by increasing effectiveness of heat transmission in the burning area.²⁵ If achieving an excellent flame retardant classification, PLA/BF thus needs higher loading of AHP to inhibit the wick effect of BF. However, a mass of AHP particles could reduce the mechanical properties of PLA/BF resulting from the discontinuities at inorganic particle/organic matrix interface.

Nanosized additives have been found that they can bring about a large enhancement in the mechanical, thermal and flame retardancy properties of polymers.^26–29 Because of the extraordinary physical and chemical properties, carbon nanotubes (CNTs) have been attractive in the field of research and application. The improvements of mechanical properties in various polymers including general plastics, engineering plastics, and elastomers, have been widely reported with the incorporation of CNTs.^30–32 CNTs provide another candidate as a flame retardant additive due to their highly elongated shape.^33–36 The introduction of CNT can significantly reduce the peak of heat release rate of polymers in cone calorimeter testing due to the formation of continuous network-structured protective layer without cracks. This protective layer consisted mainly of carbon nanotubes and it seemed to act as a heat shield for the virgin polymer below the layer.

Up to now, few paper reports the research of thermal decomposition and combustion behaviors of PLA/BF composites. The aim of this work is to examine the use of CNT and AHP as halogen-free flame retardant and reinforcement for PLA/BF composites. In order to improve the dispersion of the nanotubes and the interfacial adhesion between the nanotubes and PLA matrix, the pristine CNT was modified with inorganic acids to introduce carboxylic acid groups in this work. The modified CNT was blended with PLA to prepare the nanocomposite masterbatch using a twin-screw extruder. BF and AHP were incorporated to the masterbatch to fabricate the flame retardant composites through a twin-roll internal mixer. The structure and fracture morphology of the composites were characterized. The thermal degradation, flame retardant, and mechanical properties of these composites were also detailedly studied.

2. Experimental section

2.1 Materials

In this work, the PLA resin (Nature Works 4032D) in granular form was supplied by Cargill Dow Inc. (America). Silane coated short basalt fiber (fiber diameter of 5.5–25 μm and initial fiber length of 6–100 mm) was supplied by Jiangsu Tianlong Continuous Basalt Fiber Co., Ltd, China. The carbon nanotube used in the current work was multi-walled carbon nanotube (degree of purity > 95%, 10–30 μm length, 20–30 nm diameter) synthesized by a thermal chemical vapor deposition process. It was purchased from Beijing DK Nano technology Co., LTD. The concentrated sulfuric acid (H₂SO₄, 98%) and nitric acid (HNO₃, 68%) were purchased from the Sinopharm Chemical Reagent Co., Ltd. The acids were used without further purification. The aluminium hypophosphite (AHP) particles were supplied by Qingzhou Yichao Chemical Co., Ltd, China.

2.2 Preparation of modified carbon nanotube (m-CNT)

The carbon nanotube (CNT) in this work was modified by acids, which was prepared and characterized in our previous work.^37,38 In a typical experiment,³⁹ the CNT was added to the mixture of concentrated H₂SO₄ and HNO₃ with a volumetric ratio of 3 : 1. The mixtures were in ultrasonic bath at 80 °C for 5 h. They were then diluted and washed with deionized water until the pH value of water was about 7. The filtrated solid was dried in vacuum oven at 80 °C for 24 h. The acid-treated CNT is the m-CNT used in the current work.

2.3 Composite preparation

PLA, BF, m-CNT and AHP were dried at 80 °C overnight before use. The compounding process of masterbatch production (PLA/m-CNT) was performed using a co-rotating twin-screw extruder (LSSJ-20, made in Shanghai, China). The length-diameter ratio of the extruder was 25/1. The masterbatch was prepared at 185 °C with a constant rotation speed of 120 rpm. The mass fraction of m-CNT in the masterbatch was 5 wt%. The PLA, BF and AHP particles were blended with the PLA/m-CNT masterbatch using a twin-roll internal mixer (XK-160, made in Jiangsu, China) at 185 °C for 10 min. The roller speed was 100 rpm for the preparation of all the samples. They were then molded using an injection molding machine (SA600/100, made in Ningbo, China) and a hot press at 190 °C to obtain the composite sheets with different thick for the further measurements. The schematic representation for the whole preparation flow was shown in Fig. 1. In the current work, the composites were investigated with the same basalt fiber content of 20 wt%. The examined formulations of the composites were listed in Table 1.

Fig. 1 Schematic representation for the composite preparation.

Table 1 Examined formulations for each sample

Sample	Composition (wt%)
	PLA	BF	m-CNT	AHP
PLA/BF	80	20	0	0
PLA/BF/m-CNT1	79	20	1	0
PLA/BF/AHP10	70	20	0	10
PLA/BF/AHP15	65	20	0	15
PLA/BF/AHP20	60	20	0	20
PLA/BF/AHP19/m-CNT1	60	20	1	19

---

2.4 Measurement and characterization

2.4.1 Morphology observation. The dispersion of additives in PLA was characterized by SEM with a scanning electron microscope AMRAY1000B. Samples were prepared by immersing the films in liquid nitrogen for 10 min before fracture. The fracture surfaces were previously coated with a conductive layer of gold before analysis. In this work, the specimens of PLA/BF, PLA/BF/AHP20, and PLA/BF/AHP19/m-CNT1 were analyzed.

2.4.2 Thermal degradation behaviors. Thermal degradation behaviors of the composites were carried out using a Q5000 IR thermogravimetric analyzer (TA Instruments Waters, China) at a linear heating rate of 20 °C min⁻¹ under nitrogen condition. The weight of all the samples were kept within 5–10 mg. Composites in an open Pt pan were tested under an gas-flow rate of 6 × 10⁻⁵ m³ per minute at temperature ranging from room temperature to 700 °C.

2.4.3 Thermal combustion properties. Thermal combustion properties of the composites were measured using a microscale combustion calorimeter (MCC, Govmark) according to ASTM D 7309-07. 4–8 mg of each sample was heated at 1 °C s⁻¹ from 90 to 600 °C and held there for 30 seconds. During pyrolysis, the volatilized decomposition products are transferred in the stream of nitrogen to a high-temperature combustion furnace where pure oxygen is added and the decomposition products are completely combusted. The amount of oxygen consumed is measured with an oxygen analyzer and used to calculate a heat release rate (HRR).

2.4.4 Flammability tests. The flammability tests of the composites were determined by limiting oxygen index (LOI) and Underwriters Laboratories (UL) 94 testing according to ASTM D2863 and ANSI/UL 94-2010 respectively. An HC-2 oxygen index meter (Jiangning Analysis Instrument Company, China) was used in the LOI testing. The specimens used for the test were of dimensions 100 × 6.5 × 3.0 mm³. UL-94 testing was carried out on a CFZ-2-type instrument (Jiangning Analysis Instrument Company, China). The specimens used were of dimensions 130 × 13 × 3.0 mm³.

2.4.5 Cone calorimeter tests. The cone calorimeter tests of the composites were performed in a cone calorimeter (Fire Testing Technology) according to ASTM E 1354/ISO 5660. Each specimen (100 × 100 × 3 mm³) was wrapped in an aluminum foil and exposed horizontally to 35 kW m⁻² external heat flux.

2.4.6 Morphology of char residues. The residues of PLA/BF and PLA/BF/AHP20 collected in the cone calorimeter testing were analyzed by means of the scanning electron microscope (SEM). The SEM micrographs were obtained with a scanning electron microscope AMRAY1000B at an accelerating voltage of 20 kV. The specimens were sputter-coated with a conductive layer.

2.4.7 Tensile strength properties. The tensile strength properties of the composites were investigated using a universal testing machine (WD-20D) according to the ASTM D-638 at room temperature. The specimens were prepared by cutting strips 4.0 ± 0.1 mm wide and 2.0 ± 0.3 mm thick. The crosshead speed was 20 mm min⁻¹. An average value of at least fire individual determinations was obtained.

3. Results and discussion

3.1 Morphology characterization

In order to understand the dispersion and morphology, the composites PLA/BF, PLA/BF/AHP20 and PLA/BF/AHP19/m-CNT1 were characterized by SEM. From Fig. 2(a) and (b), it can be seen that the basalt fibers are homogeneously dispersed in the PLA matrix. The average diameter of basalt fibers is about 10 μm. Moreover, no gap is found at the fiber–matrix interface leading to the good compatibility between basalt fibers and PLA, resulting from the modified surface of basalt fibers by coupling agent. The morphology of PLA/BF/AHP20 is shown in Fig. 2(c) and (d). From Fig. 2(c), it is observed that both the basalt fibers and the AHP particles are homogeneously dispersed in the PLA matrix. Fig. 2(d) shows that the size of AHP is less than that of basalt fibers, which can reach micro-nano scale. The good compatibility between AHP and PLA is also observed. From Fig. 2(e) and (f), it can be seen that both AHP and m-CNT are homogeneously dispersed in the PLA matrix with few agglomeration. The good dispersion of m-CNT in PLA is attributed to three reasons: (1) the modification of CNTs by inorganic acids can decrease their agglomeration resulting from the reduction the π–π stacking effect among the aromatic rings of nanotubes.⁴⁰ (2) The carboxylic acid groups formed on the surface of the m-CNT can enhance the interaction between the –COOH groups on m-CNT and the C=O groups of PLA through the hydrogen bonding formation.^37,41 (3) The transesterification reaction between –COOH groups on m-CNT and PLA chains probably possibly happens during melt compounding resulting in the appearance of crosslinking, as shown in Fig. 1.

Fig. 2 SEM images of the fractured composites: (a) and (b) PLA/BF; (c) and (d) PLA/BF/AHP20; (e) and (f) PLA/BF/AHP19/m-CNT1.

3.2 Thermal degradation behaviors

The thermal degradation behaviors of neat PLA, neat CNT, m-CNT and AHP under nitrogen condition are showed in Fig. 3. The onset degradation temperature (T_-1%) of neat PLA is 332 °C, and the final residue weight at 700 °C is only 0.4 wt%. There is no change in the TGA curve of neat CNT. The 1 wt% mass loss (T_-1%) for m-CNT at 100 °C is attributed to the loss of water molecules. The further mass loss corresponds to the decomposition of organic groups formed on the surface of m-CNT.^37,42 The thermal degradation of AHP begins from 288 °C (T_-1%) with two maximal mass loss rates at 332 °C and 450 °C, leaving 74.3 wt% of solid residue at 700 °C.^23,43 The decomposition process of AHP could be represented by the two equations as detailed below:^23,43

Fig. 3 Thermogravimetric curves for neat PLA, neat CNT, m-CNT and AHP.

The TGA and DTG curves of all composite samples are shown in Fig. 4, and the results are summarized in Table 2. The thermal degradation of PLA/BF in nitrogen atmosphere is characterized by a single decomposition step with maximum mass loss rate at 367 °C (Table 2). The onset degradation temperature (T_-1%) of PLA/BF is 305 °C, lower than that of neat PLA. It is probably resulted from the degradation of silane coupling agent on the surface of basalt fibers. The resulting residue is of about 21.2 wt% and hence mainly composed of basalt fibers. From Fig. 4 and Table 2, it is clearly found that the incorporation of m-CNT significantly increases the thermal decomposition temperature and residual yields of PLA/BF. The T_-1% value increases form 305 °C for PLA/BF to 330 °C for PLA/BF/m-CNT1 with the increment of 25 °C. Both the temperatures at 30 wt% mass loss (T_-30%) and maximum mass loss rate (T_max) of PLA/BF/m-CNT1 are higher than those of PLA/BF. Additionally, PLA/BF/m-CNT1 also has a high residual weight. The results show that the presence of m-CNT can significantly enhance the thermal stability of the PLA/BF, because of its excellent thermal conductivity.^44,45

Fig. 4 Thermogravimetric (a) and differential thermogravimetric (b) curves for each sample.

Table 2 TGA and DTG data under nitrogen atmosphere for each sample. (20 °C min⁻¹, 5–15 mg; errors ± 0.5 wt%, ±1 °C)

Sample	T-1% (°C)	T-30% (°C)	Tmax (°C)	Residue (wt%) at 700 °C
PLA/BF	305	360	367	21.2
PLA/BF/m-CNT1	330	369	377	23.9
PLA/BF/AHP10	321	367	377	30.1
PLA/BF/AHP15	321	369	379	33.6
PLA/BF/AHP20	321	369	380	36.9
PLA/BF/AHP19/m-CNT1	323	371	382	38.2

---

The thermal degradation of PLA/BF is also influenced by the addition of AHP. The T_-1% value is increased with increasing AHP content. Both T_-30% and T_max values for various PLA/BF/AHP composites are higher than those of PLA/BF. For example, the T_-1% value increases from 305 °C for PLA/BF to 321 °C for PLA/BF/AHP20 with the increment of 16 °C. The T_max value increases from 367 °C for PLA/BF to 380 °C for PLA/BF/AHP20 with the increment of 13 °C. The results show that the incorporation of AHP improves the thermal stability of PLA/BF. The reduced thermal stability of PLA/BF is due to degradation of silane coupling agent on the surface of basalt fibers. Many AHP particles in PLA/BF play a barrier role to retard the pyrolysis of silane coupling agent and PLA molecular chain leading to the improvement of thermal stability of PLA/BF.^22,23 The residual weight is also improved by the addition of AHP. The high residual weight indicates that AHP probably reacts with PLA matrix. The T_-1%, T_-30% and T_max values are further increased, when the mixed compounds of AHP and m-CNT are added into PLA/BF. In comparison with PLA/BF/AHP composites, PLA/BF/AHP19/m-CNT1 also has higher residual weight.

From the mass loss rate curves, it can be observed that the maximum mass loss rate of the composites is reduced with the incorporation of AHP and m-CNT. The maximum mass loss rate of PLA/BF/AHP19/m-CNT1 is the lowest in all composites. This means that the pyrolysis of PLA/BF is greatly retarded by the loading of AHP and m-CNT. The weak peaks at around 327 °C for PLA/BF/AHP composites should be attributed to the decomposition of AHP. The results demonstrate that the presence of m-CNT effectively acts as physical barriers to prevent the transport of volatile decomposed products out of the composites during the thermal degradation, resulting in the improved solid residual yields.⁴¹ Additionally, AHP or its decomposition products can adequately react with the degradation products of PLA to inhibit the release of volatiles resulting from the strong barrier effect of m-CNT. These trapped volatiles participate in the carbonization leading to the increased residual weight.

3.3 Thermal combustion properties

Thermal combustion properties were characterized by microscale combustion calorimetry (MCC) that directly measures the heat of combustion of the gases evolved during controlled heating of milligram-sized samples, which separately reproduces the gas and condensed phase processes of flaming combustion in a nonflaming laboratory test and forces them to completion to obtain intrinsic/material combustion properties.⁴⁶ In the present work, all the composite samples were screened by MCC. All of the heat release data and the optimized heat release rate curves are summarized in Table 3 and shown in Fig. 5 respectively. PLA/BF is a flammable polymeric composite which has high peak heat release rate (PHRR) and total heat release (THR) (Table 3). The introduction of m-CNT and AHP significantly reduces the PHRR and THR values of PLA/BF. From Fig. 5 and Table 3, it can be seen that the PHRR values decrease from 304 W g⁻¹ for PLA/BF to 256 W g⁻¹ for PLA/BF/m-CNT1 with a reduction of 16%, and to 188 W g⁻¹ for PLA/BF/AHP20 with a reduction of 38%, respectively. The THR values for PLA/BF/m-CNT1 and PLA/BF/AHP20 are reduced by 24% and 40% respectively. As a certain fraction of AHP is replaced by m-CNT, a significant reduction of PHRR value is obtained: it decrease to 167 W g⁻¹ for PLA/BF/AHP19/m-CNT1 with a remarkable reduction of 45% in comparison with PLA/BF. The THR value for PLA/BF/AHP19/m-CNT1 is also reduced by 45%. The results show that the presence of m-CNT further reduces the fire risk of the PLA/BF/AHP composites. The introduction of AHP and m-CNT to PLA/BF improves the temperature at PHRR (T_PHRR). The T_PHRR value increases from 383 °C for PLA/BF to 403 °C for PLA/BF/AHP19/m-CNT1 with the increment of 20 °C. The increased temperature at PHRR indicates that the presence of AHP and m-CNT improves the flame retardancy of PLA/BF composites. From the thermal degradation analysis, the addition of AHP and m-CNT could promote the formation of condensed residues and reduce the release of volatilization products. The barrier composed of condensed residues can prevent the release of flammable substance to the burning area, leading to the good improvement of flame retardant properties.

Table 3 MCC results for each sample. (PHRR: peak heat release rate, ±5 W g⁻¹; THR: total heat release, ±0.1 kJ g⁻¹; T_PHRR: temperature at PHRR, ±2 °C)

Sample	PHRR (W g^−1)	THR (kJ g^−1)	TPHRR (°C)
PLA/BF	304	13.9	383
PLA/BF/m-CNT1	256	10.6	401
PLA/BF/AHP10	246	12.7	393
PLA/BF/AHP15	217	11.6	395
PLA/BF/AHP20	188	8.3	395
PLA/BF/AHP19/m-CNT1	167	7.6	403

---

Fig. 5 HRR curves from microscale combustion calorimeter testing for each sample.

3.4 Flammability behaviors

LOI and UL 94 tests were performed to determine the flame class of PLA/BF and flame retardant PLA/BF composites. The results obtained from the two tests are summarized in Table 4. PLA/BF is a flammable polymeric composite with a LOI of 20%. With increasing AHP concentration, the LOI increases. For the PLA/BF composite that contains 20 wt% of AHP, the LOI value increases to 29.5%. PLA/BF fails in UL 94 testing with a very long burning time. The difficult extinguishment of PLA/BF can be attributed to the “candlewick effect” caused by basalt fibres in the combustion process, which is similar with glass fiber reinforced polymer composites.⁴⁷ Adding 1 wt% of m-CNT to PLA/BF does not influence the flame classification. When PLA/BF/m-CNT is ignited in the first flame application, it can also burn to the clamp in UL 94 test. Meanwhile, the addition of m-CNT has little influence to PLA/BF in the LOI testing.

Table 4 Results of LOI and UL-94 testing for each sample

Sample	LOI (%)	UL-94, 3 mm bar
		t1/t2^a (s)	Dripping	Rating
a t1 and t2, average combustion times after the first and the second applications of the flame.b BC, burns to clamp.c NR, not rated.
PLA/BF	20.0 ± 0.5	BC^b	Yes	NR^c
PLA/BF/m-CNT1	22.0 ± 0.5	BC	Yes	NR
PLA/BF/AHP10	27.5 ± 0.5	BC	Yes	NR
PLA/BF/AHP15	28.0 ± 0.5	3.7/11.6	No	V-1
PLA/BF/AHP20	29.5 ± 0.5	1.0/2.5	No	V-0
PLA/BF/AHP19/m-CNT1	31.0 ± 0.5	1.0/1.8	No	V-0

---

The PLA/BF composites containing low concentration of AHP also fail in UL 94 testing. When the fraction of AHP is 15 wt%, it can extinguish after the first application of the flame within a short time. The average burning time is less than 20 s after the second application of the flame. Moreover, no melt flow or dripping is observed. Therefore, PLA/BF/AHP15 can pass the V-1 classification testing. The PLA/BF composite containing 20 wt% of AHP achieves a V-0 rating with a high LOI (29.5%), indicating the high flame retardant efficiency of AHP. The synergistic flame retardant effect between AHP and m-CNT is observed: PLA/BF/AHP19/m-CNT1 can pass the V-0 classification testing with shorter burning time in comparison with other composites. It also has the highest LOI value (31.5%). In combination with thermal decomposition analysis, it has been known that the addition of AHP and m-CNT could promote the formation of condensed residues and reduce the release of volatilization products. Condensed residues covered on the surface of basalt fibers may be effective to reduce the wick effect to enhance the self-extinguishment of PLA/BF composites.

3.5 Forced flaming combustion

Cone calorimeter is one of the most effective bench-scale methods for the laboratory evaluation of the fire performance of polymeric materials. In cone calorimeter testing, the time to ignition (TTI), peak of heat release rate (PHRR) and total release rate (THR) are the most important parameters to evaluate the combustion behaviors. The cone calorimeter data obtained at a heat flux of 35 kW m⁻² for all composite samples are summarized in Table 5. The HRR and THR curves are shown in Fig. 6. The TTI values of PLA/BF composites are increased with the addition of AHP. At the AHP content of 20 wt%, the TTI value of PLA/BF/AHP20 is 69 s, higher than that of PLA/BF (48 s). An increase in loading of AHP from 10 to 20 wt% significantly reduces the PHRR and THR values. For example, the PHRR value decreases from 395 kW m⁻² for PLA/BF to 177 kW m⁻² for PLA/BF/AHP15 with a reduction of 55%, and to 156 kW m⁻² for PLA/BF/AHP20 with a reduction of 61%. The THR values for PLA/BF/AHP15 and PLA/BF/AHP20 are reduced by 21% and 24% respectively.

Table 5 Cone calorimeter data for each sample at 35 kW m⁻² (TTI: time to ignition, ±2 s; PHRR: peak heat release rate, ±15 kW m⁻²; THR: total heat release, ±0.5 MJ m⁻²; AMLR: average mass loss rate, ±0.1 g s⁻¹ m⁻²; ASEA: average smoke extinction area, ±20 m² kg⁻¹; residue, ±0.5 wt%)

Sample	TTI (s)	PHRR (kW m^−2)	THR (MJ m^−2)	ASEA (m^2 kg^−1)	AMLR (g s^−1 m^−2)	Residue (wt%)
PLA/BF	48	395	48.3	342	10.7	20.9
PLA/BF/m-CNT1	63	244	46.8	348	7.4	23.4
PLA/BF/AHP10	61	217	40.2	295	5.2	29.5
PLA/BF/AHP15	68	177	38.1	274	4.4	35.8
PLA/BF/AHP20	69	156	36.8	263	4.1	39.7
PLA/BF/AHP19/m-CNT1	69	143	35.1	239	4.0	41.2

---

Fig. 6 HRR (a) and THR (b) curves from cone calorimeter testing at 35 kW m⁻² for each sample.

The TTI value of PLA/BF composites is significantly increased with the addition of m-CNT. At the m-CNT content of 1 wt%, the TTI is 63 s, higher than that of PLA/BF/AHP10 (61 s). The delayed TTI should be attributed to the high thermal conductivity of m-CNT, which conducts high temperature on the composites' surface.^38,48 Additionally, the presence of m-CNT can act as barrier to reduce the release of polymer fragments from matrix underneath into the air, which delays time to the critical point of the concentration of combustible gases.^38,48 The introduction of m-CNT also reduces the PHRR and THR values. The PHRR value decreases from 395 kW m⁻² for PLA/BF to 244 kW m⁻² with a reduction of 38%. In the HRR curve of PLA/BF/AHP19/m-CNT1, a positive interaction between m-CNT and AHP can be observed: they show the broad and low peak with a reduction of 64% compared with PLA/BF. Additionally, the THR value is reduced by 27%. The improved flame retardancy shows a strong synergistic effect between m-CNT and AHP, which is in good agreement with the results of LOI and UL 94 testing.

Average smoke extinction area (ASEA) means that smoke is produced per unit mass of material being volatilized, which represents the relationship between volatile property and smoke emission.⁴⁹ From Table 5, it is observed that loading 1 wt% of m-CNT to PLA/BF does not influence the ASEA value. An increasing concentration of AHP from 10 to 20 wt% reduces the ASEA values of PLA/BF. For example, the ASEA value decreases from 342 m² kg⁻¹ for PLA/BF to 263 m² kg⁻¹ for PLA/BF/AHP20 with a reduction of 23%. The introduction of m-CNT into PLA/BF/AHP further reduces the ASEA value. It decreases from 342 m² kg⁻¹ for PLA/BF to 239 m² kg⁻¹ with a reduction of 30%. The results indicate that the mixed compounds composed of m-CNT and AHP reduce the smoke emission of PLA/BF composites.

The fire hazards of PLA/BF, represented by TTI, PHRR, THR, and ASEA in cone calorimeter testing, are remarkably reduced by the addition of AHP and m-CNT. The reduction of fire hazards should be mainly attributed to the remarkable decrease of the mass loss rate and improvement of solid residue content. From the AMLR results, it can be seen that the AMLR value of PLA/BF/AHP19/m-CNT1 is reduced by around 63% in comparison with that of PLA/BF. The reduction of mass loss rate may be resulted from the formation of char layer. The residue results are similar with TGA data. The combined additives significantly improve the formation of solid residues of PLA/BF composites after combustion. The results are in good agreement with PHRR and THR values. The evidence also demonstrates that the combustion of PLA/BF is greatly inhibited by the introduction of AHP and m-CNT based on the condensed-phase action.

3.6 Residue analysis

In order to realize the flame retardancy mechanism of the composites in the current work, the residues after cone calorimeter testing were analyzed. Fig. 7 shows the digital photos of the residues after cone calorimeter testing. It can be seen that the color of the residues of PLA/BF without AHP and m-CNT is green, which are the remained basalt fibers. The color of PLA/BF/m-CNT1 residues is black, but it is lighter than that of other flame retardant PLA/BF composites, because of the low residual weight in the combustion process of PLA/BF/m-CNT1. The residues of PLA/BF/AHP composites are black, as shown in Fig. 7(c)–(e). It indicates that the addition of AHP can promote the formation of char residues of PLA/BF through the reaction between AHP and PLA matrix. However, many holes can be clearly observed on the surface of all the residues of PLA/BF/AHP composites. Interestingly, the holes disappear in PLA/BF/AHP19/m-CNT1 residues as shown in Fig. 7(f). It indicates that the incorporation of m-CNT in PLA/BF/AHP can promote the formation of more compact char residues.

Fig. 7 Digital photographs for the residues after cone calorimeter testing: (a) PLA/BF; (b) PLA/BF/m-CNT1; (c) PLA/BF/AHP10; (d) PLA/BF/AHP15; (e) PLA/BF/AHP20; (f) PLA/BF/AHP19/m-CNT1.

In order to understand the structure and morphology, the residues of PLA/BF, PLA/BF/AHP20 and PLA/BF/AHP19/m-CNT1 were further analyzed by means of SEM. The appearance of PLA/BF residues (Fig. 8(a)) shows that there are few solid chars around basalt fibers. Only some solid chars are covered on the surface of basalt fibers in the high magnification image (Fig. 8(b)). In the SEM images of the PLA/BF/AHP20 residues (Fig. 8(c) and (d)), it is clearly observed that there are a mass of char residues covered on the surface of basalt fibers. The SEM images of the PLA/BF/AHP19/m-CNT1 residues (Fig. 8(e) and (f)) show that there are more char residues covered on the surface of basalt fibers. The residues are more compact. From the TGA and cone calorimeter results, it has been known that the residual weight of PLA/BF is significantly improved by the loading of AHP and m-CNT. Condensed char layer can decrease the mass loss rate of the polymer matrix leading to the reduction of flammable gases' concentration in the burning area.

Fig. 8 SEM images for the residues after cone calorimeter testing: (a) and (b) PLA/BF; (c) and (d) PLA/BF/AHP20; (e) and (f) PLA/BF/AHP19/m-CNT1.

Fig. 9(a) and (b) are the magnified SEM images from PLA/BF/AHP20 and PLA/BF/AHP19/m-CNT1 respectively. It is found that the char residue morphology of PLA/BF/AHP20 on the surface of basalt fibers is irregular, which is in good agreement with the reported articles.^21–24 From the SEM image displayed in Fig. 9(b), it can be seen that there are a great many uniform particles distributed on the residue surface. These particles are closed to sphere with micrometer scale. In our previous study, it has been found that the introduction of nanoclay in glass fiber reinforced PBT/AHP composites can promote the formation of microspheres.²³ Similarly, the microspheres in the current work should be the condensed-phase products caused by the sufficient reaction between AHP and PLA matrix under the physical barrier effect of m-CNT. They migrate to or accumulate on the surface of residues to form a novel protective layer. The compact char layer can effectively reduce the heat conduction of basalt fibers and cut off the mass transfer path resulting in the weakening of the wick effect. The fire resistant properties of PLA/BF composites are thus enhanced.

Fig. 9 SEM images for the residues after cone calorimeter testing: (a) PLA/BF/AHP20; (b) PLA/BF/AHP19/m-CNT1.

3.7 Mechanical properties

The mechanical properties of the composites were characterized according to their tensile properties. Fig. 10 shows the tensile properties of all the composite samples. The tensile strength and elongation at break for PLA/BF with 20 wt% of basalt fibers are respectively 58.5 MPa and 2.52%. The results are consistent with the reported values.⁴ The introduction of m-CNT in PLA/BF leads to an enhancement of mechanical properties such as tensile strength and Young's modulus. It is attributed to the good dispersion of m-CNT in PLA matrix, and the interfacial interactions of the functional groups of m-CNT with PLA chains. As shown in Fig. 10, it is observed that both the tensile strength and tensile modulus increase firstly then decrease with increasing AHP fraction. The elongation at break decreases with increasing AHP content. The tensile strength value increases from 58.5 MPa for PLA/BF to 75.7 MPa for PLA/BF/AHP10. Although the tensile strength and tensile modulus are reduced when the concentration of AHP is above 10 wt%, the two values for PLA/BF composites with various AHP contents are still larger than those for PLA/BF. This means that the low loading of AHP to PLA/BF results in the improvement of tensile properties. The inorganic/organic composites are generally expected to become stiff and more brittle upon incorporation of inorganic fillers.⁵⁰ Because there is no reaction between the interface of PLA matrix and AHP, the compatibility is too poor to improve the elasticity of PLA/BF composite, resulting in the reduction of strain-at-break.^51,52 Additionally, the poor compatibility and interfacial interaction between AHP particles and PLA matrix leads to discontinuities at particle/matrix interface. Therefore, the tensile strength and tensile modulus of PLA/BF are slightly reduced when the concentration of AHP is high. Interestingly, the presence of m-CNT with the reinforcement at nanoscale improves the Young's modulus, tensile strength for the PLA/BF/AHP composites. It should be attributed to the enhanced reinforcement effect of m-CNT and the reduced concentration of AHP.

Fig. 10 Tensile properties of each sample at a crosshead speed of 20 mm min⁻¹: (a) Young's modulus and tensile strength; (b) elongation at break.

4. Conclusions

In this work, halogen-free flame retardant basalt fiber reinforced polylactide (PLA/BF) composites were prepared using aluminium hypophosphite (AHP) and modified carbon nanotube (m-CNT) as additives. Morphological observations revealed that both the m-CNT and AHP particles were homogeneously dispersed in the PLA/BF composites. The improvement of thermal stability and tensile strength/modulus were strongly depended on the uniform dispersion of the mixed particles and the interactions between m-CNT and PLA. The presence of AHP and m-CNT significantly reduced the peak heat release rate of PLA/BF in microscale combustion calorimeter testing. The combustion behaviors were evaluated by limiting oxygen index (LOI), Underwriters Laboratories 94 (UL 94), and cone calorimeter. For PLA/BF containing 19 wt% AHP and 1 wt% m-CNT, it achieved a V-0 classification in UL 94 testing with a high LOI (31%). Additionally, the peak heat release rate and total heat release were respectively reduced by around 64% and 27% in comparison with PLA/BF. The residue analysis showed that many microspheres formed by the sufficient reaction between AHP and PLA under the physical barrier effect of m-CNT migrated to or accumulated on the surface of residues. The compact char layer composed of these microspheres could effectively reduce the heat conduction of basalt fibers and cut off the mass transfer path resulting in the weakening of the wick effect resulting in the significant improvement of the composites.

Acknowledgements

The work was financially supported by National Natural Science Foundation of China (No. 51403048 and No. 51276054), Anhui Provincial Natural Science Foundation (1508085QE111), and Program for Excellent Young Talents in University of Anhui Province.

References

1. O. Martin and L. Avérous, Polymer, 2001, 42, 6209.
2. L. Yu, K. Dean and L. Li, Prog. Polym. Sci., 2006, 31, 576.
3. L.-T. Lim, R. Auras and M. Rubino, Prog. Polym. Sci., 2008, 33, 820.
4. T. Tábi, P. Tamás and J. G. Kovács, eXPRESS Polym. Lett., 2013, 7, 107.
5. J. R. Dorgan, H. J. Lehermeier, L. I. Palade and J. Cicero, Macromol. Symp., 2001, 175, 55.
6. F. Nagata, T. Miyajima and Y. Yokogawa, J. Eur. Ceram. Soc., 2006, 26, 533.
7. Y. Tokiwa and B. P. Calabia, Appl. Microbiol. Biotechnol., 2006, 72, 244.
8. R. G. Sinclair, J. Macromol. Sci., Pure Appl. Chem., 1996, 33, 585.
9. K. Meinander, M. Niemi, J. S. Hakola and J. F. Selin, Macromol. Symp., 1997, 123, 147.
10. T. Czigany, Mater. Sci. Forum, 2005, 473–474, 59.
11. J. S. Szabo and T. Czigany, Polym. Test., 2003, 22, 711.
12. J. M. Park, W. G. Shin and D. J. Yoon, Compos. Sci. Technol., 1999, 59, 355.
13. T. Czigány, K. Pölöskei and J. Karger-Kocsis, J. Mater. Sci., 2005, 40, 5609.
14. V. Manikandan, J. T. W. Jappes, S. M. S. Kumar and P. Amuthakkannan, Composites, Part B, 2012, 43, 812.
15. Q. Liu, M. T. Shaw and R. S. Parnas, Polym. Compos., 2006, 27, 475.
16. T. Bárány, E. Földes and T. Czigány, eXPRESS Polym. Lett., 2007, 1, 180.
17. T. Liu, F. M. Yu, X. J. Yu, X. L. Zhao, A. Lu and J. H. Wang, J. Appl. Polym. Sci., 2012, 125, 1292.
18. D. Kurniawan, B. S. Kim, H. Y. Lee and J. Y. Lim, Composites, Part B, 2012, 43, 1010.
19. X. Chen, Y. Li and N. Gu, Biomed. Mater., 2010, 5, 044104.
20. B. Zhao, Z. Hu, L. Chen, Y. Liu, Y. Liu and Y. Z. Wang, J. Appl. Polym. Sci., 2011, 119, 2379.
21. G. Tang, X. J. Huang, H. C. Ding, X. Wang, S. D. Jiang, K. Q. Zhou, B. B. Wang, W. Yang and Y. Hu, RSC Adv., 2014, 4, 8985.
22. G. Tang, X. Wang, W. Y. Xing, P. Zhang, B. B. Wang, N. N. Hong, W. Yang, Y. Hu and L. Song, Ind. Eng. Chem. Res., 2012, 51, 12009.
23. W. Yang, Y. Hu, Q. L. Tai, H. D. Lu, L. Song and R. K. K. Yuen, Composites, Part A, 2011, 42, 794.
24. W. Yang, L. Song, Y. Hu, H. D. Lu and R. K. K. Yuen, Composites, Part B, 2011, 42, 1057.
25. A. Casu, G. Camino, M. de Giorgi, D. Flath, A. Laudi and V. Morone, Fire Mater., 1998, 22, 7.
26. S. S. Ray and M. Okamoto, Prog. Polym. Sci., 2003, 28, 1539.
27. X. Wang, W. Y. Xing, P. Zhang, L. Song, H. Y. Yang and Y. Hu, Compos. Sci. Technol., 2012, 72, 737.
28. C. Nyambo, P. Songtipya, E. Manias, M. M. Jimenez-Gascoc and C. A. Wilkie, J. Mater. Chem., 2008, 18, 4827.
29. V. Patil, R. V. Dennis, T. K. Rout, S. Banerjee and G. D. Yadav, RSC Adv., 2014, 4, 49264.
30. M. Lahelin, M. Annala, A. Nykänen, J. Ruokolainen and J. Seppälä, Compos. Sci. Technol., 2011, 71, 900.
31. S. Pande, A. Chaudhary, D. Patel, B. P. Singh and R. B. Mathur, RSC Adv., 2014, 4, 13839.
32. J. Fritzsche, H. Lorenz and M. Klppüel, Macromol. Mater. Eng., 2009, 294, 551.
33. T. Kashiwagi, E. Grulke, J. Hilding, R. Harris, W. Awad and J. Douglas, Macromol. Rapid Commun., 2002, 23, 761.
34. T. Kashiwagi, E. Grulke, J. Hilding, K. Groth, R. Harris, K. Butlera, J. Shields, S. Kharchenko and J. Douglas, Polymer, 2004, 45, 4227.
35. T. Kashiwagi, F. M. Du, J. F. Douglas, K. I. Winey, R. H. Harris and J. R. Shields, Nat. Mater., 2005, 4, 928.
36. G. Beyer, Fire Mater., 2002, 26, 291.
37. W. Yang, H. Zhou, B. H. Yang, H. D. Lu, L. Song and Y. Hu, Polym. Compos.,  DOI:10.1002/pc.23354.
38. W. Yang, B. H. Yang, H. D. Lu, L. Song and Y. Hu, Ind. Eng. Chem. Res., 2014, 53, 18489.
39. K. Esumi, M. Ishigami, A. Nakajima, K. Sawada and H. Honda, Carbon, 1996, 34, 279.
40. S. Tzavalas, V. Drakonakis, D. E. Mouzakis, D. Fischer and G. Gregoriou, Macromolecules, 2006, 39, 9150.
41. J. Y. Kim, S. I. Han and S. Hong, Polymer, 2008, 49, 3335.
42. P. C. P. Watts, N. Mureau, Z. Tang, Y. Miyajima, J. D. Carey and S. R. P. Silva, Nanotechnology, 2007, 18, 175701.
43. N. V. Romanova and N. V. Demidenko, Russ. Chem. Rev., 1975, 44, 1036.
44. L. X. Benedict, S. G. Lourie and M. L. Cohen, Solid State Commun., 1996, 100, 177.
45. S. Berber, Y. K. Kwon and D. Tomanek, Phys. Rev. Lett., 2000, 84, 4614.
46. R. E. Lyon, R. N. Walters and S. I. Stoliarov, Polym. Eng. Sci., 2007, 47, 1501.
47. C. S. Zhao, F. L. Huang, W. C. Xiong and Y. Z. Wang, Polym. Degrad. Stab., 2008, 93, 1188.
48. Q. Wu, W. Zhu, C. Zhang, Z. Y. Liang and B. Wang, Carbon, 2010, 48, 1799.
49. B. Li, Polym. Degrad. Stab., 2003, 82, 467.
50. C. H. Tseng, H. B. Hsueh and C. Y. Chen, Compos. Sci. Technol., 2007, 67, 2350.
51. N. A. Isitman, H. O. Gunduz and C. Kaynak, Polym. Degrad. Stab., 2009, 94, 2241.
52. W. Yang, N. N. Hong, L. Song, Y. Hu, R. K. K. Yuen and X. L. Gong, Ind. Eng. Chem. Res., 2012, 51, 8253.

---