A new approach designed for improving flame retardancy of intumescent polypropylene via continuous extrusion with supercritical CO₂

Abstract

The objective of this study was to explore a new approach to enhance the flame retardancy of polypropylene/intumescent flame retardant (PP/IFR) composites through improving IFR dispersion via continuous extrusion with supercritical carbon dioxide (scCO₂) as the processing medium. The IFR used in this study consisted of ammonium polyphosphate (APP) and pentaerythritol (PER) at a weight ratio of 3 : 1. Scanning electron microscopy (SEM) was applied to study the dispersion of IFR in the PP composites prepared in the presence of different contents of scCO₂. Thermogravimetric analysis (TGA) was carried out to study the effect of IFR dispersion on the thermal stability and the yield of char residues of the PP composites. A plastic smoke generation tester was applied to study the specific optical density under a thermal irradiance. SEM was then used to study the morphologies of the intumescent char formed after the smoke testing. Limiting oxygen index (LOI) and vertical burning test (UL-94) ratings were evaluated to study the effect of IFR dispersion on the flame retardancy of the PP composites. It was found that scCO₂ as the processing medium could significantly improve IFR dispersion and thus increase the efficiency of the flame retardant and enhance the flame retardancy of the PP composites. This new approach could allow PP composites to pass the UL-94 V-0 rating at a lower IFR content. Possible mechanisms were proposed for the scCO₂-aided IFR dispersion and the improved flame retardancy of the PP composites.

---

1. Introduction

Polypropylene (PP) has been widely used in electrical appliances, automobiles, and buildings due to its good mechanical properties, chemical stability, easy processing and low cost, etc. In the fields above, fire safety regulations and measures are strictly enforced, which require PP to meet certain flame retardant ratings. However, PP shows burning characteristics of easy flammability accompanied by severe dripping, which would cause catastrophic disasters when a fire is initiated.^1,2 Hence, flame retardant modification of PP is of great importance to meet the fire safety standards and thus to expand its applications. Brominated flame retardants are traditionally used as additives in PP owing to the high flame retardancy and cost effectiveness. However, in recent years, due to the rising environmental and health concerns, applications of halogen-free flame retardants have been demanded in many sectors.^3,4

In recent years, intumescent flame retardant (IFR) has been extensively investigated on flame retardant PP due to its low smoke, low toxicity and anti-dripping, etc.^5,6 Generally, an IFR system is composed of three parts: an acid source, e.g. ammonium polyphosphate, a charring source, e.g. pentaerythrotol, and a blowing source, e.g. melamine. The flame retarding mechanism of intumescent char formation has been reported for IFR elsewhere.⁷ During heating, the acid source could decompose to generate inorganic acid, which could dehydrate the charring source in esterification reactions to form carbonaceous char. Subsequently, the gas produced by the blowing source could swell the carbonaceous char to form intumescent char layers, which covered on the sample surface as barriers to protect the underlying materials. However, the dispersion of IFR in PP is usually poor due to the bad compatibility, which results in a low efficiency of the flame retardant. Based on the previous reports,^8,9 an addition of more than 30 wt% of IFR is usually required for the PP composites to pass the UL-94 V-0 rating. As such, how to improve the flame retardancy of intumescent PP has been an important and meaningful task. Incorporating synergistic agents has proven one feasible means, which includes metallic compounds, carbon nanotube and layered double hydroxides, etc.^10–13

Another approach to enhance the flame retardancy of polymer composites could be through improving the dispersion of flame retardants in the polymer matrix.^14–17 Gui et al.¹⁴ fabricated a ternary composite of polymer/crosslinked rubber/nano-magnesium hydroxide (MH) via first preparation of a special compound powder of crosslinked rubber/nano-MH by a co-spray technique, and found that the better dispersion of nano-MH in the composite favored the better flame retardancy. Huang et al.¹⁵ found that the functionalization of graphene by grafting a synthesized intumescent flame retardant could improve the dispersion of graphene in the polymer matrix and hence improved the flame retardancy of the nanocomposites. Peeterbroeck et al.¹⁶ reported that a better dispersion of the high-density polyethylene coated multiwalled carbon nanotubes (c-MWNTs) in EVA relative to that of the uncoated nanotubes could explain the better flame retardancy. Till now, the approaches for improving the dispersion of flame retardants usually involve no less than two steps of synthesis or modification as well as processing. Hence, an efficient and practical approach that could directly process flame retardant polymer composites with good dispersion is still demanded.

Recently, supercritical carbon dioxide (scCO₂) has usually applied in synthesis, processing, foaming and particle formation due to its gas-like diffusivity, liquid-like density and mild supercritical condition (31.1 °C and 7.38 MPa) as well as environmental friendliness, chemical inertness and low cost.^18–22 According to the previous literature,^18–22 scCO₂ could lower the viscosity of polymer melts and increase the mass transfer rate due to its plasticizing effect on polymers. Some studies^19,22,23 showed that thermal processing in the presence of scCO₂ facilitated the intercalation and exfoliation of polymer/clay nanocomposites. It could be partially due to the larger mobility of macromolecules upon the plasticizing effect of scCO₂ and partially due to the increase of interlayer space upon the expansion of scCO₂ during pressure release. Ma et al.²⁴ reported that the presence of scCO₂ as the thermal processing medium favored the dispersion of PP/sepiolite composites. Wu et al.²⁵ proposed an in situ bubble-stretching model to explain the improved dispersion of inorganic additives in polymers. It was believed that the stretching of the surrounding polymers and oscillation during bubble expansion of foaming could facilitate the particle dispersion. Goren et al.²⁶ reported that the foaming with CO₂ from supercritical state to gas could break the aggregates of SiO₂ and thus improved SiO₂ dispersion in polymer matrix. Till now, scCO₂ has seldom been applied as a processing medium during thermal processing to improve the dispersion of flame retardants and thus to enhance the flame retardancy. If this could be proved feasible, a new practical and efficient approach could be established for improving the flame retardant properties of intumescent PP. Recently, with utilization of scCO₂ as the processing medium or foaming agent, our group has made a series of progresses in improving phase dispersion, mechanical properties, long-chain branching, in situ compatibilization, and fabricating bimodal-cell and open-cell foams.^27–32 The current study would further expand the application of scCO₂ as a green processing medium to flame retardant materials.

This study aimed to explore a new practical approach to enhance the flame retardancy of PP/IFR composites through improving the dispersion of IFR via a thermal process of continuous extrusion with scCO₂ as the processing medium. A traditional IFR system applied was ammonium polyphosphate (APP) and pentaerythritol (PER) at a weight ratio of 3 : 1. Many characterizations were carried out such as scanning electron microscope (SEM), thermogravimetric analysis (TGA), smoke generation, limiting oxygen index (LOI) and UL-94 tests to study the relationship between the IFR dispersion and the flame retardancy of the PP/IFR composites. Possible mechanisms were proposed for the scCO₂-aided IFR dispersion and the consequent improved flame retardancy.

2. Experimental

2.1. Materials

Polypropylene (PPB-MO2) used in this study was purchased from Sinopec Zhenhai Refining & Chemical Company (China), with an MFI of 1.6 g/10 min at 230 °C/2.16 kg. Ammonium polyphosphate (APP, EPFR-APP231) and pentaerythritol (PER, CAS 115-77-5) were supplied by Polyrocks Chemical co. LTD and Aladdin, respectively. APP and PER were used at a mass ratio of 3 : 1. Carbon dioxide (CO₂) with a purity of 99.9% was provided by Ningbo Wanli Gas Corporation.

2.2. Samples preparation

PP, APP and PER were vacuum-dried at 60 °C for 8 h before use. The sample preparation included the following four steps: first, PP/APP/PER premix was prepared using a TSE-35A/600-15-48 twin screw extruder (Nanjing Yarui, China) with a screw diameter of 35 mm and a length diameter ratio of 25, with a rotating speed at 200 rpm and a processing temperature at 195 °C; second, the PP/APP/PER premix was fed into a tandem single-screw extruder to prepare the flame retardant PP composites with different contents of scCO₂ as a processing medium; third, the PP composites were compression-molded into sheets with a XLB50-D compression molder (Huzhou Shuangli Automation Technology, China) at 190 °C under a pressure of 10 MPa for 5 min; fourth, the sheets were cut into standard samples for different tests. It is noted that scCO₂ was used only as a processing medium for promoting the dispersion of IFR in PP and could easily totally be removed during the compression molding. The formulations of the PP/IFR composites are shown in Table 1.

Table 1 PP/IFR composites prepared with different contents of scCO₂

Samples	Composition of PP/IFR			scCO2 (wt%)
	PP	APP	PER
PP/23IFR/0scCO2	77.0	17.3	5.7	0
PP/23IFR/3scCO2	77.0	17.3	5.7	3
PP/23IFR/5scCO2	77.0	17.3	5.7	5
PP/23IFR/7scCO2	77.0	17.3	5.7	7
PP/25IFR/0scCO2	75.0	18.7	6.3	0
PP/25IFR/3scCO2	75.0	18.7	6.3	3
PP/25IFR/5scCO2	75.0	18.7	6.3	5
PP/25IFR/7scCO2	75.0	18.7	6.3	7
PP/30IFR/0scCO2	70.0	22.5	7.5	0
PP/30IFR/3scCO2	70.0	22.5	7.5	3
PP/30IFR/5scCO2	70.0	22.5	7.5	5
PP/30IFR/7scCO2	70.0	22.5	7.5	7

---

The description about the tandem extrusion system used in this study can be found in the previous literature.^27–32 It mainly consists of two extruders, two drive motors, an ISCO syringe pump, a gear pump, a heat exchanger and a nozzle, etc. There are several thermocouples and pressure transducers located in the two extruders and the nozzle, which can provide the temperature and pressure online during the process. The scCO₂ was injected at the three quarter point of the first extruder by the syringe pump under a pressure of 18 MPa. The concentration of scCO₂ was regulated by adjusting its flow rate through the syringe pump referred to the materials feed rate^31,33 and was kept at the ratio of 0 wt%, 3 wt%, 5 wt% and 7 wt% of the mixture, respectively. The first extruder was used to plasticize the polymers and preliminarily mix the scCO₂ with the intumescent flame retardants and polymers, with a rotating speed at 36 rpm and a processing temperature at 200 °C. The second extruder was used to completely and uniformly dissolve the scCO₂ in the polymer melts, with a rotating speed at 5 rpm and a processing temperature set at 190 °C. The pressure of the whole system was ca. 10–15 MPa. The processing condition in this study was above the supercritical condition of CO₂ (31.1 °C and 7.38 MPa), and thereby, it was guaranteed that CO₂ was in supercritical state during the process.

Actually, based on the previous literature,³⁴ the concentration of scCO₂ used in this study was below its solubility in PP at the processing condition to make sure that it can completely dissolve in PP melts. In addition, according to Park et al.³⁵ and Ohshima et al.,³⁶ the diffusivity of scCO₂ in PP at 188 °C and 200 °C was ca. 4.2 × 10⁻⁵ cm² s⁻¹ and 8.0 × 10⁻⁵ cm² s⁻¹, respectively. Upon the injection of scCO₂ into the polymer melts, it took only several seconds for scCO₂ to be completely dissolved in PP melts and formed a one-phase polymer/scCO₂ solution. In this study, the mixing and residence time for scCO₂ in the extruders was ca. 15 min. Thereby, scCO₂ was definitely able to be completely and uniformly dissolved in PP melts, which was considered favorable for improving the dispersion of IFR in PP.

2.3. Thermogravimetry analysis (TGA)

Thermogravimetry analysis of the PP/IFR composites was carried out with a Mettler Toledo TGA/DSC1 thermal analyser from 50 °C to 800 °C at a heating rate of 10 °C min in nitrogen atmosphere with a flow rate of 50 ml min⁻¹. A calculated TGA curve was obtained for PP/25IFR composite in the same temperature range, assuming that no interaction happened among PP, APP and PER. The calculated curve was generated based on the eqn (1), which was the summation of the TGA curves of each individual component in the mixture multiplied by their weight fractions:³⁷

M = 0.75m_PP + 0.25(0.75m_APP + 0.25m_PER) (1)

2.4. Limiting oxygen index (LOI) measurement

The limiting oxygen index values were measured with a Vouch 5801A digital oxygen index tester (Suzhou Yangyi, China) according to GB/T 2406.2-2009. The specimens used were 100 × 6.5 × 3.2 mm³ in dimensions.

2.5. Vertical burning (UL-94) testing

The vertical burning test was carried out with a Vouch 5402 H-V burning tester (Suzhou Yangyi, China) according to GB/T2408-2008. The specimens used were 100 × 13 × 3.2 mm³ in dimensions.

2.6. Smoke density testing

The smoke density test was carried out with a Vouch 5920 plastic smoke generation tester (Suzhou Yangyi, China) according to GB/T8323.2-2008. Sheet samples with dimensions of 70 × 70 × 1 mm³ were placed on a metallic grid under an irradiance of 50 kW m⁻². The optical density was measured continuously with an optical system. By measuring the light transmission, the specific optical density (D_s) was obtained based on eqn (2):³⁸

D_s(t) = 132 × log(100/T(t)) (2)

where D_s(t) is the specific optical density at time t, T(t) is the percent light transmission at time t, and the factor of 132 is obtained based on the volume of the smoke chamber, the exposed sample surface area and the optical path length of the light beam.

2.7. Scanning electron microscope (SEM) observation

The effect of scCO₂ as a processing medium on the dispersion of IFR in PP was investigated with a Zesis EVO18 scanning electron microscope at an acceleration voltage of 20 kV after the specimens were freeze-fractured in liquid nitrogen and a layer of gold was sputtered on the surface. Morphologies of the intumescent char layers of the PP composites after smoke density testing was also characterized with SEM.

3. Results and discussion

3.1. Effect of scCO₂ on IFR dispersion

The effect of scCO₂ as a processing medium on the dispersion of the flame retardants in PP/25IFR composites was investigated with SEM and shown in Fig. 1 (two magnifications). For PP/25IFR/0scCO₂, big IFR agglomerates were observed in PP, indicating bad dispersion of IFR in the polymer matrix at a relatively higher content. For PP/25IFR/3scCO₂, the IFR agglomerates were decreased in size. For PP/25IFR/5scCO₂ and PP/25IFR/7scCO₂, the dispersion of IFR was even better. The results indicated that scCO₂ as a processing medium had a positive effect on the dispersion improvement of IFR.

Fig. 1 SEM images of the dispersion of flame retardants in PP/25IFR composites prepared with different contents of scCO₂.

The reason that scCO₂ as a processing medium enhanced IFR dispersion could be interpreted based on its effect during polymer processing and upon pressure release as follows. First, numerous studies^18–22 have already reported that scCO₂ could plasticize polymers and thus lower the viscosity of polymer melts and increase the mass transfer rate. This technology has been applied to in situ polymerization and reactive extrusion, etc. In this study, the scCO₂ was completely and homogeneously dissolved in the PP melts, and thereby, it was acceptable to infer that the presence of scCO₂ could increase the mass transfer rate of IFR during processing and thus improve the dispersion. Second, Wu et al.²⁵ proposed that the stretching of polymer melts during bubble expansion upon pressure release improved the dispersion of inorganic additives in polymers. Goren et al.²⁶ also reported that the expansion of CO₂ during foaming could break the SiO₂ aggregates and then improved the dispersion of SiO₂. In this study, although the temperature at the die was relatively too high for the polymer melts to hold the gas to be well foamed, the force generated during gas escape upon pressure drop could also help improve the dispersion of IFR in PP.

3.2. Effect of IFR dispersion on thermal stability

Thermogravimetric analysis was characterized in nitrogen atmosphere to study the thermal stability and the yield of char residues of the PP composites. The TGA curves of the PP/25IFR composites prepared in the presence of different contents of scCO₂ and the related data are shown in Fig. 2 and Table 2, respectively. It was found that all the PP/25IFR composites showed similar thermal decomposition behavior. The thermal decomposition occurred in the temperature range of 230–400 °C was mainly due to the decomposition of APP and PER and the subsequent esterification reactions. The thermal decomposition of APP could generate phosphoric acid, which could dehydrate PER in an esterification reaction. The thermal decomposition above 400 °C was mainly about the decomposition of PP and some formed products during the heating.³⁹ A calculated TGA curve was obtained for the PP/25IFR composite, assuming that PP, APP and PER were independent with each other and no interaction occurred during the heating. The inconsistency in TGA curve as well as the yield of the char residues between the experimental and the calculated data confirmed otherwise the interaction among PP, APP and PER during the heating. Thereby, it could imagine that the improved dispersion of IFR could favor their interactions. For the PP/25IFR composites prepared with scCO₂, with the increasing content of scCO₂, T_{1 wt%} (decomposition temperature at 1 wt%) was decreased, T_{50 wt%} (decomposition temperature at 50 wt%) was increased, and the char residues at 800 °C was increased, indicating that better dispersion of IFR endorsed better thermal stability and larger yield of char residues of the PP/25IFR composites.

Fig. 2 TGA curves of PP/25IFR composites in nitrogen atmosphere.

Table 2 TGA data of PP/25IFR composites in nitrogen atmosphere

Samples	T1 wt% (°C)	T50 wt% (°C)	Residues (800 °C, %)
PP/25IFR-calculated	239.8	458.2	4.4
PP/25IFR/0scCO2	244.3	465.7	8.5
PP/25IFR/3scCO2	237.3	471.0	9.0
PP/25IFR/5scCO2	234.6	471.2	11.8
PP/25IFR/7scCO2	234.5	471.7	13.5

---

3.3. Effect of IFR dispersion on smoke generation

The specific optical density (D_s) can reflect the opacity due to smoke formation, and thus usually be used to evaluate the generation of smoke under heat irradiance. The smoke density curves of the PP/25IFR composites prepared with different contents of scCO₂ are shown in Fig. 3. For PP/25IFR/0scCO₂, under the thermal irradiance, the specific optical density started to increase rapidly at about 7 min and reached a maximum value of 16.5 at about 16 min. With the increase of the content of scCO₂, the specific optical density was retarded to increase, and increased relatively slower to a lower value. The larger char yield favored by the scCO₂-aided better dispersion of IFR could well explain the decrease in smoke generation during the heating, as the previous literature³⁸ reported that a higher char yield could contribute to a reduction in smoke density. The result could also be attributed to the fact that IFR could work as smoke suppressants.^40,41

Fig. 3 Specific optical density of PP/25IFR composites prepared with different contents of scCO₂.

3.4. Effect of IFR dispersion on intumescent char formation

The char quality plays an important role in determining the flame retardancy in a condensed phase action. The formation of an intumescent char layer covering on the whole sample surface could efficiently slow down the transfer of heat and oxygen and protect the underlying polymer matrix from thermal decomposition. Morphologies of the char residues of PP/25IFR composites after the smoke density tests were studied, and the digital photos and the SEM images are shown in Fig. 4. The results showed that for PP/25IFR/0scCO₂, the intumescent char layer formed was not in good quality, with incomplete char indicated by an arrow in the digital photo and collapse and cracks in the char in the SEM images. For PP/25IFR/3scCO₂, the char quality was improved relatively, but the defects and incompleteness were still observed. For PP/25IFR/5scCO₂ and PP/25IFR/7scCO₂, intumescent char layers were observed covering the whole sample surface. The results indicated that the scCO₂-aided better dispersion of IFR facilitated formation of char layers in better quality. Hence, it was highly expected that the PP/25IFR composites prepared with 5 wt% and 7 wt% scCO₂ would show better flame retardancy compared to those prepared with 3 wt% and 0 wt% scCO₂.

Fig. 4 Digital photos and SEM images of morphologies of the char residues of PP/25IFR composites prepared with different contents of scCO₂.

3.5. Effect of IFR dispersion on flame retardancy

The effect of scCO₂-aided IFR dispersion on the flame retardancy of PP/25IFR composites was studied with the limited oxygen index and vertical burning tests (Table 3). With the increase of scCO₂, LOI values were increased, indicating that the minimum concentration of oxygen required for combustion in the mixture of oxygen and nitrogen was increased. Also, with the increase of scCO₂, the UL-94 rating was increased and the dripping was effectively suppressed. To be in details, PP/25IFR/0scCO₂ could only pass the UL-94 V-2 rating, while PP/25IFR/3scCO₂ passed the V-1 rating, and PP/25IFR/5scCO₂ and PP/25IFR/7scCO₂ passed the V-0 rating. These results demonstrated that the better dispersion of IFR facilitated the improvement of flame retardancy of the PP/25IFR composites. The findings were consistent with the previous reports documenting that the improved the dispersion of nano-MH, graphene and multiwalled carbon nanotubes could endorsed better flame retardancy of the polymer composites.^14–16

Table 3 Flame retardant properties of PP/25IFR composites prepared with different contents of scCO₂

Samples	LOI (%)	(t1 + t2)^a	Dripping	UL-94
a (t1 + t2) represent the averaged total afterflame time of 5 specimens.
PP/25IFR/0scCO2	32.8	16.5	Y	V-2
PP/25IFR/3scCO2	34.1	11.3	N	V-1
PP/25IFR/5scCO2	34.5	2.8	N	V-0
PP/25IFR/7scCO2	35.8	1.9	N	V-0

---

3.6. Effectiveness of scCO₂-aided improvement of flame retardancy of PP/IFR composites

The results above showed that scCO₂ as a processing medium could improve IFR dispersion in PP matrix and thus enhanced the flame retardancy of PP/25IFR composites. It was considered that if through using scCO₂ to improve the dispersion of IFR in PP a lower loading of IFR could allow the PP composite to pass UL-94 V-0 rating, less IFR additives would then be required for the new processing approach. Thereby, PP/23IFR and PP/30IFR composites were prepared in the presence of different contents of scCO₂ as well as PP/25IFR composites to verify the predication above and the feasibility of the new approach.

The char residues of the various PP/IFR composites at 800 °C after TGA tests as a function of scCO₂ are shown in Fig. 5. The residues of PP/23IFR, PP/25IFR and PP/30IFR composites were all increased with scCO₂, indicating that the scCO₂-aided IFR dispersion facilitated the yield of char residues of all the PP composites, which would favor the improvement of flame retardancy.

Fig. 5 The effect of the contents of scCO₂ on the yield of char residues of various PP/IFR composites at 800 °C after TGA tests.

The LOI values of the various PP/IFR composites as a function of scCO₂ are shown in Fig. 6. It was found that the LOI values of PP/23IFR, PP/25IFR and PP/30IFR composites were all increased with scCO₂, revealing that the improved IFR dispersion increased the oxygen percentage required to support the combustion. It was also observed that with more contents of scCO₂ as the processing medium, the PP/23IFR composite had a LOI value as high as that of PP/30IFR prepared with no or less contents of scCO₂. Similar findings were also observed in Table 4 about the UL-94 ratings for the PP/IFR composites. With 5 wt% or 7 wt% scCO₂ as the processing medium, PP/23IFR passed the UL-94 V-0 rating that PP/30IFR could pass without scCO₂. It meant that in this study, with scCO₂ as the processing medium, addition of 23 wt% IFR allowed the PP composite to pass the UL-94 V-0 rating, and such a low loading required of IFR has seldom been reported in the previous literature without incorporating a synergistic agent. A previous study⁴² reported that the online application of ultrasound to a static mixer die coupled to a single-screw extruder improved the dispersion of IFR and hence enhanced the flame retardancy of PP composites. In their study, the formulation of 7 wt% APP, 7 wt% PER and 7 wt% melamine combined with 1 wt% clay and 2 wt% anhydride-grafted PP allowed the PP composite to pass the UL-94 V-0 rating. The low loading of 21 wt% IFR required in their study could be attributed to the improved dispersion of both the IFR additives and clay, and also the synergistic effect between them under the heating. In this study, without incorporating any synergistic agent, only through utilizing scCO₂ as the processing medium to improve IFR dispersion in PP could significantly increase the efficiency of the flame retardant and allowed the PP composite to pass the UL-94 rating at a low loading of 23 wt% IFR. Thus, it has demonstrated a feasible approach to fabricate high flame retardant PP/IFR composites through improving the dispersion of IFR with scCO₂ as a processing medium, and this new approach could probably be applied to many other flame retardant systems.

Fig. 6 The effect of the contents of scCO₂ on the LOI values of various PP/IFR composites.

Table 4 The effect of the contents of scCO₂ on the UL-94 ratings of various PP/IFR composites

Samples	scCO2 (wt%)
	0	3	5	7
PP/23IFR	NO	V-2	V-0	V-0
PP/25IFR	V-2	V-1	V-0	V-0
PP/30IFR	V-0	V-0	V-0	V-0

---

3.7. Possible mechanisms proposed for scCO₂-aided IFR dispersion and improved flame retardancy

To help more clearly understand the improved IFR dispersion and the enhanced flame retardancy of the PP/IFR composites prepared in the presence of scCO₂ as the processing medium, possible mechanisms were proposed. The schematics in Fig. 7a (as comparison) and Fig. 7b were illustrated for PP/IFR composites prepared without and with scCO₂, respectively. For PP/IFR composites prepared in the absence of scCO₂ (Fig. 7a), big agglomerates would be formed due to the poor compatibility between the flame retardants and the polymer matrix. Under heat flux, no uniform intumescent char layer could be formed covering the whole sample surface due to the bad dispersion of IFR, which thus could not protect the underlying polymer matrix from thermal decomposition and generation of inflammable substances to feed the combustion. In comparison, for PP/IFR composites prepared in the presence of scCO₂ (Fig. 7b), the dispersion of IFR could be well improved, possibly partially due to the high mass transfer rates arising from the plasticizing effect of scCO₂ on the polymers during thermal processing^18–22 and partially due to the in situ stretching effect during bubble expansion²⁵ (at suitable foaming temperatures) or during gas escape (at relatively higher temperatures) upon pressure release. Thereby, under heat flux, a uniform intumescent char layer could be formed covering the whole sample surface, which could keep the heat and oxygen out and prevent the underlying polymer matrix from further thermal decomposition. Hence, better flame retardancy was endorsed by the scCO₂-aided IFR dispersion. Due to the complexity arising from too many experimental factors that could affect the processing of polymer composites processing with scCO₂ as the processing medium, further studies are still needed to more specifically elucidate the mechanism for the scCO₂-aided IFR dispersion and also to explore the optimized processing conditions for improving IFR dispersion in PP.

Fig. 7 Schematics for IFR dispersion and the corresponding flame retardancy of PP/IFR composites prepared via different approaches: (a) conventional processing without scCO₂, and (b) continuous extrusion with scCO₂.

4. Conclusions

This study aimed to explore a new approach to enhance the flame retardancy of PP/IFR composites through improving IFR dispersion via continuous extrusion in the presence of scCO₂ as a green processing medium. It was found that the dispersion of IFR was significantly improved with scCO₂ and that the improved IFR dispersion endorsed the PP composites better thermal stability, more char residues, better intumescent char layers, less smoke density, and enhanced flame retardancy. The improved IFR dispersion increased the efficiency of the flame retardant and thereby allowed the PP composites to pass UL-94 rating at a relatively lower content of flame retardants. In this study, with 5 wt% scCO₂ as the processing medium, PP/23IFR composite passed UL-94 V-0 rating, which usually required a loading of IFR no less than 30 wt% for the conventional processing means. The processing in this study combines the characteristics of capacity of pilot-scale production of continuous extrusion and environmental friendliness, easy handling and removal, and low costs of scCO₂. Therefore, it has demonstrated potential to establish a new approach to improve the flame retardancy of PP/IFR composites, and this approach is highly expected to be applied to many other flame retardant systems.

Acknowledgements

Financial supports from National Natural Science Foundation of China (51603222 and 51473181), Natural Science Foundation of Zhejiang Province (LQ14E030006) and Natural Science Foundation of Ningbo (2014A610131) are gratefully acknowledged.

References

1. T. Zhang, H. Yan, L. Shen, Z. Fang, X. Zhang, J. Wang and B. Zhang, RSC Adv., 2014, 4, 48285–48292.
2. Y. H. Guan, W. Liao, Z. Z. Xu, M. J. Chen, J. Q. Huang and Y. Z. Wang, RSC Adv., 2015, 5, 59865–59873.
3. A. B. Morgan and J. W. Gilman, Fire Mater., 2013, 37, 259–279.
4. A. R. Horrocks, Polym. Degrad. Stab., 2011, 96, 377–392.
5. K. S. Lim, S. T. Bee, L. T. Sin, T. T. Tee, C. T. Ratnam, D. Hui and A. R. Rahmat, Composites, Part B, 2016, 84, 155–174.
6. M. Doğan, A. Yılmaz and E. Bayramlı, Polym. Degrad. Stab., 2010, 95, 2584–2588.
7. Z. Jiang and G. Liu, RSC Adv., 2015, 5, 88445–88455.
8. S. Bourbigot, M. LeBras, S. Duquesne and M. Rochery, Macromol. Mater. Eng., 2004, 289, 499–511.
9. T. Mariappan and C. A. Wilkie, Macromol. Chem. Phys., 2012, 213, 1987–1995.
10. Y. Shen, W. Gong, B. Zheng, X. Meng and L. Gao, Polym. Degrad. Stab., 2016, 129, 114–124.
11. H. Y. Ma, L. F. Tong, Z. B. Xu and Z. P. Fang, Adv. Funct. Mater., 2008, 18, 414–421.
12. X. Wang, Y. Sporer, A. Leuteritz, I. Kuehnert, U. Wagenknecht, G. Heinrich and D. Y. Wang, RSC Adv., 2015, 5, 78979–78985.
13. Y. Q. Ma and Y. Y. Pang, Chin. J. Polym. Sci., 2015, 33, 772–782.
14. H. Gui, X. H. Zhang, Y. Q. Liu, W. F. Dong, Q. G. Wang, J. M. Gao, Z. H. Song, J. M. Lai and J. L. Qiao, Compos. Sci. Technol., 2007, 67, 974–980.
15. G. B. Huang, S. Q. Chen, S. W. Tang and J. R. Gao, Mater. Chem. Phys., 2012, 135, 938–947.
16. S. Peeterbroeck, F. Laoutid, J. M. Taulemesse, F. Monteverde, C. J. M. Lopez, J. B. Nagy, M. Alexandre and P. Dubois, Adv. Funct. Mater., 2007, 17, 2787–2791.
17. P. A. Song, L. H. Xu, Z. H. Guo, Y. Zhang and Z. P. Fang, J. Mater. Chem., 2008, 18, 5083–5091.
18. S. P. Nalawade, F. Picchioni and L. P. B. M. Janssen, Prog. Polym. Sci., 2006, 31, 19–43.
19. M. Sauceau, J. Fages, A. Common, C. Nikitine and E. Rodier, Prog. Polym. Sci., 2011, 36, 749–766.
20. S. D. Yeo and E. Kiran, J. Supercrit. Fluids, 2005, 34, 287–308.
21. O. R. Davies, A. L. Lewis, M. J. Whitaker, H. Tai, K. M. Shakesheff and S. M. Howdle, Adv. Drug Delivery Rev., 2008, 60, 373–387.
22. D. L. Tomasko, X. M. Han, D. H. Liu and W. H. Gao, Curr. Opin. Solid State Mater. Sci., 2003, 7, 407–412.
23. S. Horsch, G. Serhatkulu, E. Gulari and R. M. Kannan, Polymer, 2006, 47, 7485–7496.
24. J. Ma, E. Bilotti, T. Peijs and J. A. Darr, Eur. Polym. J., 2007, 43, 4931–4939.
25. D. M. Wu, Q. Y. Meng, Y. Liu, Y. M. Ding, W. H. Chen, H. Xu and D. Y. Ren, J. Polym. Sci., Part B: Polym. Phys., 2003, 41, 1051–1058.
26. K. Goren, O. B. Okan, L. Chen, L. S. Schadler and R. Ozisik, J. Supercrit. Fluids, 2012, 67, 108–113.
27. K. Wang, S. S. Wang, F. Wu, Y. Y. Pang, W. T. Zhai and W. G. Zheng, Polym. Bull., 2016, 73, 941–957.
28. K. Wang, Y. Y. Pang, P. K. Huang, L. H. Zhang, W. Liu and W. G. Zheng, RSC Adv., 2016, 6, 106347–106354.
29. K. Wang, S. S. Wang, F. Wu, Y. Y. Pang, W. Liu, W. T. Zhai and W. G. Zheng, J. Mater. Sci., 2016, 51, 2705–2715.
30. K. Wang, Y. Y. Pang, W. Liu, F. Wu and W. G. Zheng, J. Supercrit. Fluids, 2016, 118, 203–209.
31. K. Wang, Y. Y. Pang, F. Wu, W. T. Zhai and W. G. Zheng, J. Supercrit. Fluids, 2016, 110, 65–74.
32. S. S. Wang, K. Wang, Y. Y. Pang, Y. Li, F. Wu, S. Wang and W. G. Zheng, J. Appl. Polym. Sci., 2016, 133, 43812.
33. M. Lee, C. B. Park and C. Tzoganakis, Polym. Eng. Sci., 1999, 39, 99–109.
34. G. Li, J. Wang, C. B. Park and R. Simha, J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 2497–2508.
35. C. B. Park and N. P. Suh, Polym. Eng. Sci., 1996, 36, 34–48.
36. S. Areerat, E. Funami, Y. Hayata, D. Nakagawa and M. Ohshima, Polym. Eng. Sci., 2004, 44, 1915–1924.
37. F. Gao, L. F. Tong and Z. P. Fang, Polym. Degrad. Stab., 2006, 91, 1295–1299.
38. L. G. Imhof and K. C. Stueben, Polym. Eng. Sci., 1973, 13, 146–152.
39. M. Zhang, P. Ding and B. J. Qu, Polym. Compos., 2009, 30, 1000–1006.
40. Y. W. Yan, L. Chen, R. K. Jian, S. Kong and Y. Z. Wang, Polym. Degrad. Stab., 2012, 97, 1423–1431.
41. Y. Li, B. Li, J. Dai, H. Jia and S. Gao, Polym. Degrad. Stab., 2008, 93, 9–16.
42. O. G. Sanchez, S. A. Sanchez, F. Calderas, T. L. Medina, V. E. E. Herrera, G. A. Rivera and O. Manero, Polym. Eng. Sci., 2013, 53, 2018–2026.

---