Polycarbonate–acrylonitrile-butadiene-styrene blends with simultaneously improved compatibility and flame retardancy

Abstract

Tri(2,6,7-trioxa-1-phosphabicyclo[2.2.2]octane-1-oxo-4-hydroxymethyl) phenylsilane (TPPSi) was employed as a flame retardant and polydimethylsiloxane-g-styrene-g-methyl methacrylate copolymer (PSM) as a compatibilizer in polycarbonate–acrylonitrile-butadiene-styrene (PC–ABS) blends. The effect of TPPSi and PSM on the flame retardancy and compatibilization of PC–ABS was investigated by UL-94 vertical burning tests, DMA and SEM. The results indicated that the PC–ABS blend with 8 wt% TPPSi and 5 wt% PSM possessed both high flame retardancy and impact strength. It was found that PSM played an effective compatibilization role in the PC–ABS–TPPSi blend and elevated the thermal stability of the blend. A synergistic flame-retardant mechanism between phosphorus and silicon was presented by TGA, SEM/EDX analysis, FT-IR and Raman spectra of the residue.

---

1. Introduction

Polycarbonate (PC) possesses good mechanical properties and heat resistance whereas acrylonitrile-butadiene-styrene (ABS) provides good processability and reliable notch insensitivity. PC–ABS blends have received considerable attention in patents and technical applications due to the complementary properties of the two components and are well-known commercial products.^1–4 With the continuously growing market segment of electronic engineering products, the need for the flame retardancy of PC–ABS is increasing.^5–9 With regard to the gradual prohibition of halogen-containing flame retardants,¹⁰ some environment-friendly phosphorus or silicon-containing flame retardants such as aryl phosphates have received increasing attention.^11–15 Triphenyl phosphate (TPP) and bisphenol-A bis(diphenyl phosphate) (BDP) exhibited an efficient flame retardancy in PC–ABS and the use of 12–18 wt% TPP¹¹ or 12 wt% BDP¹² gained a UL-94 V-0 classification in PC–ABS. The flame retardant mechanism of these aromatic phosphates has been reported in a series of papers.¹³ The combination of phosphorus and silicon results in a superior flame-retardant effect on PC–ABS blends. It was found that 10 wt% loading of polyhedral oligomeric silsesquioxane containing 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO-POSS) endowed PC–ABS blends with a V-0 in the UL-94 test.¹⁵

Nevertheless, the addition of flame retardants usually decreases the impact strength of PC–ABS and causes some problems with processability. For example, some flame retardants with low decomposition temperatures degrade under the processing temperature, and adding too much flame retardant to PC–ABS reduces its fluidity. The use of 10 wt% bisphenol A bis(diethyl phosphate) (BEP) in PC–ABS gained a UL-94 V-0 classification, but lowered the Izod notched impact strength of PC–ABS from 65.3 kJ m⁻² to 7.5 kJ m⁻².¹⁶ It is necessary to add a compatibilizer to flame-retardant PC–ABS blends.¹⁷ Commonly used compatibilizers such as methacrylate-butadiene-styrene (MBS), ethylene-vinyl acetate (EVA), and styrene-maleic anhydride (SMA) can enhance the compatibility between the PC–ABS matrix and flame retardant, but always deteriorate the flame retardancy of PC–ABS blends due to their combustibility. Thus, the addition of a small quantity of MBS increased the Izod notched impact strength significantly, but decreased the limiting oxygen index (LOI) value of the PC–ABS blend.¹⁷

Generally more than 10 wt% loading of the flame retardant in PC–ABS blends is required to achieve a UL-94 V-0 classification, whereas too much loading will deteriorate the mechanical properties, in particular the impact strength of PC–ABS blends. The design and synthesis of novel and efficient flame retardants and compatibilizers is always necessary to endow PC–ABS blends with both high flame retardancies and good mechanical properties. A flame retardant containing silicon and caged bicyclic phosphate groups, tri(2,6,7-trioxa-1-phosphabicyclo[2.2.2]octane-1-oxo-4-hydroxymethyl) phenylsilane (TPPSi), synthesized in our laboratory, has been shown to possess excellent flame retardancy and smoke suppression at low loading in polyamide 6.¹⁸ TPPSi (see Fig. 1a) is expected to be effective at flame-retarding PC–ABS blends. Meanwhile, compared with conventional compatibilizers, polydimethylsiloxane-g-styrene-g-methyl methacrylate copolymer (PSM, see Fig. 1b) contains fire-inert silicon and possesses excellent thermal stability.^19,20 It is expected that PSM does not reduce the flame-retardant properties when employed as a compatibilizer in PC–ABS blends flame-retarded by TPPSi.

Fig. 1 Schematic chemical structures of (a) TPPSi and (b) PSM.

In this paper TPPSi was employed as a flame retardant in PC–ABS blends with PSM as a compatibilizer. The effect of PSM and TPPSi on the compatibilization and flame retardancy of PC–ABS was investigated by dynamic mechanical analysis (DMA), using scanning electron microscopy (SEM), etc. This work may provide a foundation for outlining a new direction in the development of high performance PC–ABS blends.

2. Experimental section

2.1. Materials

PC (type 201) was produced by LG-DOW Company (Korea), and ABS (XR401) was purchased from LG Chem (Korea) with a melt flow rate of 6.0 g per 10 min and a notched impact strength of 160 J m⁻¹. Methacrylate-butadiene-styrene (MBS) was purchased from Sunny Company (China). Octamethyl cyclotetrasiloxane (D4), tetramethyl tetravinyl cyclotetrasiloxane (VD4) and hexamethyldisiloxane (MM) were provided by Tianci Organic Silicon Company (China) and were used as received. Styrene, methyl methacrylate (MMA), dodecylbenzene sulfonic acid (DBSA) and poly(ethylene glycol) monooctylphenyl ether (OP-10) were purchased from Aladdin Industrial Corporation (China). Styrene and MMA were purified by distillation and K₂S₂O₈ was recrystallized from water just before use.

2.2. Synthesis of TPPSi

TPPSi was prepared in our laboratory according to the reference literature.¹⁸ It was synthesized by the reaction of 1-oxo-4-hydroxymethyl-2,6,7-trioxa-1-phosphabicyclo[2.2.2] octane and phenyltrichlorosilane, then washed with acetonitrile and distilled water.

2.3. Synthesis of PSM

PSM was also prepared in our laboratory according to the reference literature¹⁹ by emulsion polymerization. Distilled water, surfactants (DBSA and OP-10) and monomers (D4, VD4 and MM) were added into a four-necked flask. The reaction was carried out for 8 h at 85 °C with stirring. Styrene and MMA were then added in turn and reacted at 85 °C using a K₂S₂O₈ initiator. PSM (white powder) with 50 wt% polydimethylsiloxane was obtained after precipitation in ethanol and dried under reduced pressure.

2.4. Specimen preparation

The mixture of PC, ABS, TPPSi, PSM and other additives was melt-kneaded and extruded into pellets in a 30 mm twin-screw extruder with a cylinder temperature of 250–260 °C at a screw speed of 100 revolutions per minute. The resultant pellets were dried at 95 °C for 5 h and then injection-molded at an injection temperature of 260–270 °C into test pieces to measure the flame-retardancy and mechanical properties.

2.5. Characterization

The tensile strength of the specimens was determined according to ISO 527:1996 and the flexural strength according to ISO 178:2003 using a Zwick/Roell Z010 (Zwick, Germany) universal electronic tensile testing machine. The Izod notched impact strength of the specimens was determined using a Z5113 (Zwick, Germany) radial-boom impact tester according to the ISO 180:2001 standard. The LOI values were measured on a FTT0080 instrument (FTT, England) with specimen dimensions of 150 × 10 × 4 mm³, according to the ISO 4589:1999 standard. The UL-94 vertical burning tests were carried out with a FTT UL94 flame chamber (FTT, England) with specimen dimensions of 120 × 12.7 × 3.2/1.6 mm³ according to the UL94-2009 standard. DMA of the specimens was conducted with a DMA242C (Netzsch, Germany) dynamic mechanical analyzer at a fixed frequency of 1 Hz and an oscillation amplitude of 0.05 mm, and was studied from 30 °C to 180 °C with a heating rate of 4 °C min⁻¹. Thermogravimetric analysis (TGA) was performed on a TG 209 F1 (Netzsch, Germany) thermogravimetric analyzer and dynamic experiments run from 30 °C to 700 °C at a heating rate of 20 °C min⁻¹ under a nitrogen flow of 20 mL min⁻¹. The errors in temperature and mass are ± 2 °C and ± 0.2%, respectively. SEM micrographs of the blends and char residues were observed using a Nova NanoSEM 430 (Fei, Holland). The blends were cryo-fractured in liquid nitrogen prior to SEM examination, and all the samples were coated with a conductive gold layer. Elemental compositions of the char surfaces were investigated by EDX using an ADD350 + HKL Fast EBSD system (HKL, England) attached to the SEM. Raman spectra were obtained on a Lab RAM Aramis Micro-Raman spectroscope (Horiba, France) using He laser beam excitation (wavelength: 632.8 nm). Fourier transform infrared spectra (FT-IR) from 400 to 4000 cm⁻¹ were recorded using a VERTEX70 spectrometer (Bruker, Germany).

3. Results and discussion

3.1. Influence of PSM and TPPSi on properties of PC–ABS blends

PC is self-extinguishing in air on account of its strong tendency to char, exhibiting a V-2 rating by itself in the UL-94 test.²¹ ABS is a highly combustible material with an 18% LOI value. The flame-retardant performances of PC–ABS blends are markedly influenced by the ABS components. A PC–ABS blend at a 90/10 weight ratio (named as PC–ABS0) exhibits a V-2 rating with 3.2 mm specimens. The addition of 8 wt% TPPSi can endow PC–ABS0 with a V-0 rating, but cause the Izod notched impact strength to decrease severely, possibly due to poor compatibility between the matrix of PC–ABS and TPPSi. The functionalized polymer PSM is employed as a compatibilizer for PC–ABS0 and PC–ABS–TPPSi (PAPSi for short) blends. It is found that 5 wt% PSM is enough to provide substantial compatibility for the blends. MBS is a compatibilizer commonly used in PC–ABS blends which is currently on the market.¹⁷ A comparison between MBS and PSM is performed for use in PAPSi. Five samples of PC–ABS blends were investigated and all blends maintained a weight ratio of PC to ABS of 90/10. Their formulation, mechanical properties and flame-retardant performances are listed in Table 1.

Table 1 Mechanical properties and flame-retardant performances of PC–ABS blends

Blends	PC	ABS	PSM	MBS	TPPSi	Tensile strength (MPa)	Flexural strength (MPa)	Izod notched impact strength (J m^−1)	UL-94 rating		LOI (%)
	wt%								3.2 mm	1.6 mm
a NR = No rating.b NB = No break.
PC–ABS0	90	10	0	0	0	58.9 ± 0.5	92.0 ± 0.6	527 ± 25	V-2	NR^a	22
PAPSi	82.8	9.2	0	0	8	64.3 ± 0.4	99.4 ± 0.5	95 ± 5	V-0	V-0	30
PC–ABS–PSM	85.5	9.5	5	0	0	57.6 ± 1.1	86.3 ± 0.8	NB^b	V-2	NR	23
PAPSi–MBS	78.3	8.7	0	5	8	57.2 ± 1.2	89.8 ± 1.3	NB	V-0	V-1	28
PAPSi–PSM	78.3	8.7	5	0	8	55.6 ± 1.1	93.2 ± 0.9	NB	V-0	V-0	31

---

As seen in Table 1, TPPSi is an effective flame retardant in PC–ABS0. PAPSi containing 8 wt% TPPSi has an observable improvement in its flame-retardant performance. The UL-94 rating reaches a V-0 rating for 1.6 mm specimens, and the LOI value increases from 22% for PC–ABS0 to 30%. However, the Izod notched impact strength decreases sharply from 527 J m⁻¹ to 95 J m⁻¹. It is found that the incorporation of 5 wt% PSM into PAPSi leads to a remarkable increase in the impact strength, almost without sacrificing tensile and flexural strength. The addition of PSM does not reduce the UL-94 rating of PAPSi and the PAPSi–PSM still reaches a V-0 rating for 1.6 mm specimens with a 31% LOI value. Thus 5 wt% PSM mainly plays an effective compatibilization role in both PC–ABS0 and PAPSi.

It is noticed that the impact strength of PAPSi is significantly improved by the aid of MBS. However, the UL-94 rating of PAPSi–MBS only reaches a V-1 rating for 1.6 mm specimens and the LOI value decreases from 30% to 28%, indicating that MBS reduces the flame retardancy of PAPSi. In conclusion, compared to MBS, PSM is a more appropriate compatibilizer for flame-retardant PAPSi.

3.2. Compatibility characterization

Dynamic mechanical analysis (DMA) was used to determine the thermal transition and analyze the miscibility of PC–ABS blends. Fig. 2 shows the tan δ spectra for five blends. The glass transition temperature (T_g) of the PC phase in PC–ABS0 is located at 136.5 °C (as shown in Fig. 2a). The T_g of the ABS phase should be located at around 110 °C, but is difficult to discern in the figure due to a low content of ABS in the blend. The improvement in compatibility is characterized by the T_g shift of the PC phase to the ABS phase. Apparently, the addition of 5 wt% PSM to the blend results in a downward shift of the T_g of the PC phase to 132.0 °C, indicating an improved compatibility between PC and ABS.

Fig. 2 The tan δ spectra for five blends.

The tan δ spectra for PAPSi, PAPSi–PSM and PAPSi–MBS are compared in Fig. 2b. The T_g of the PC phase in PAPSi shifts to 138.4 °C, slightly higher than that in PC–ABS0. With the addition of 5 wt% MBS or PSM into PAPSi, the T_g of the PC phase shifts downwards. Furthermore, the shift in the T_g of the PC phase with PSM is almost 8 °C, compared to a shift of 4 °C with MBS, illustrating that PSM is a more efficient compatibilizer for PAPSi.

In addition to DMA, SEM is also an effective method to examine the compatibility of blends. SEM morphologies of the impact fractured sections of PC–ABS0 and PC–ABS–PSM are compared in Fig. 3. The fractured surface of uncompatibilized PC–ABS0 is uneven owing to the limited compatibility between PC and ABS.²² The sizes of the dispersed ABS domains are reduced with the addition of PSM.

Fig. 3 SEM micrographs of (a) PC–ABS0, (b) a PC–ABS–PSM blend, (c) PAPSi and (d) a PAPSi–PSM blend.

SEM morphologies of the impact fractured sections of PAPSi and PAPSi–PSM are also compared in Fig. 3. As shown in the figure, some TPPSi pits have dropped out of the PC–ABS matrix, possibly due to a poor affinity with the matrix. The pits are imbedded in the matrix and the sizes of the dispersed ABS domains are reduced by virtue of PSM. On the basis of DMA and SEM examinations, we can conclude that PSM improves compatibility among PC, ABS and TPPSi components to a great extent.

3.3. Thermogravimetric analysis

TGA and differential thermogravimetry (DTG) curves of PC–ABS0 and three kinds of flame-retarded PC–ABS blends are compared in Fig. 4. The relevant thermal degradation data, including T_5% (defined as the temperature at 5% weight loss), T_max1 andT_max2 (the temperatures at maximum weight loss rates in step 1 and step 2, relating to the degradation stages of ABS and PC, respectively), and the char residue at 700 °C are given in Table 2.

Fig. 4 TGA and DTG curves of four PC–ABS blends.

Table 2 TGA data of four PC–ABS blends

Sample	Temperature/°C			Residue at 700 °C/%
	T5%	Tmax1	Tmax2
PC–ABS0	403	404	505	17.3
PAPSi	393	418	539	23.1
PAPSi–MBS	389	411	536	20.2
PAPSi–PSM	416	417	533	20.8

---

Two-stage thermal degradation behavior for PC–ABS0 is observed. The first degradation stage occurs in the range 350–430 °C with a 10.4% weight loss and T_max1 of 404 °C. The mass loss is almost equal to the ABS content of 10 wt% in the formulation, corresponding mainly to the degradation of ABS. A second degradation stage, attributed to the main degradation of PC, appears at 430–570 °C with T_max2 at 505 °C and a weight loss of 70.1%. The degradation process of PC–ABS0 is in good agreement with other reported results on PC–ABS.^14,23–25

When TPPSi is added to PC–ABS, thermal degradation still happens in a two-stage process, but the degradation temperature varies compared with PC–ABS0. The degradation of PAPSi starts 10 °C earlier than degradation in PC–ABS0 and originates from the degradation of TPPSi in the blend at 380 °C.¹⁸ Nevertheless, the decomposition product of TPPSi promotes elevation in the thermal stability of PC and ABS, with both T_max1 and T_max2 shifting to higher temperatures. In particular, T_max2, which corresponds to the degradation of PC, goes up from 505 °C to 539 °C. In addition, the char residue at 700 °C increases from 17.3% to 23.1%, meaning that TPPSi promotes the charring of PC during degradation. TPPSi contains flame-retardant phosphorus and silicon elements. The caged bycyclic phosphate groups in TPPSi are probably converted by thermal decomposition into long chain phosphoric acids (polymeric phosphates) which in the condensed phase extract water from the pyrolysing substrates of the polymers, causing them to char rather than burn. These non-volatile, polymeric phosphoric acids inhibit pyrolysis reactions by providing the simultaneously forming carbonaceous layer with a glassy coating. This protective layer is resistant to even high temperatures and shields the underlying polymers from further attack by oxygen and radiant heat.¹³ The presence of Si in the TPPSi structure might be further contributing towards the thermal oxidative stability of the char layer, probably due to the formation of SiO₂ after burning.²⁶ It may be that a synergistic effect between phosphorus and silicon in TPPSi primarily in the condensed phase affords acceptable flame-retardancy in PC–ABS blends.

When MBS or PSM is added to PAPSi, the degradation process of the blend is changed to some degree. The degradation of PAPSi–MBS shifts to a lower temperature and T_5% occurs at 389 °C while the degradation of PAPSi–PSM shifts to a higher temperature and T_5% happens at 416 °C, which is 23 °C higher than with PAPSi. A high onset degradation temperature relates to a high thermal stability and an excellent capability to resist the flames, which is beneficial for an improvement in flame retardancy. The weight loss for PAPSi is 12.1% in the temperature range 350–430 °C and the weight loss for PAPSi–MBS rises to 14.3%, while that of PAPSi–PSM declines to 8.0%. The addition of PSM elevates the onset degradation temperature and considerably decreases the weight loss in the first degradation stage while MBS exerts the opposite effect. PSM containing silicon possesses a higher thermal stability than MBS, and this distinction may be caused by the difference in thermal stability between PSM and MBS. In addition, T_max2 and the char residue at 700 °C for PAPSi–PSM and PAPSi–MBS are almost identical and lower than those of PAPSi. It is conceivable that the superior flame-retardant rating of PC blends with other combustible polymeric materials chiefly depends on high thermal stability in the first stage and has little relation to the second stage of degradation.²⁷ Thus the improved thermal stability by PSM in the first stage of degradation plays a key role in the attainment of a V-0 rating by PAPSi–PSM.

3.4. SEM/EDX analysis of the residues’ surfaces

SEM morphologies of PAPSi–MBS and PAPSi–PSM residues from the extinguished specimens formed via vertical burning tests are displayed in Fig. 5. The difference is clearly reflected in the two SEM micrographs. The multiporous residue of PAPSi–MBS is replaced by a coherent vitreous layer, which may be non-porous SiO₂, along with a carbonaceous layer, formed after burning in PAPSi–PSM. Clearly, the layer of PAPSi–PSM can shield the underlying polymers from attack by oxygen and radiant heat, resulting in better flame retardancy.

Fig. 5 SEM morphologies of (a) PAPSi–MBS and (b) PAPSi–PSM residues.

With 8.0 wt% TPPSi, the silicon percentage is 0.35 wt% and the phosphorus percentage is 1.16 wt% in PAPSi and PAPSi–MBS blends before burning. The addition of PSM causes the silicon percentage of the PAPSi–PSM blend to increase up to 0.95 wt%. The elemental compositions on the surfaces of PAPSi, PAPSi–MBS and PAPSi–PSM residues are obtained using SEM/EDX and the concentrations of carbon, oxygen, silicon and phosphorus are listed in Table 3. It is well-known that silicon can migrate to and accumulate on the surface of char.²⁸ In PAPSi, silicon accumulates to a considerable extent on the surface (an increase from 0.35 wt% to 1.09 wt%) whereas in PAPSi–MBS the same accumulation is negligibly small (only from 0.35 wt% to 0.49 wt%), possibly because MBS prevents silicon (from TPPSi) migrating to the surface during or after burning. For PAPSi–PSM, the compatibilizer is itself responsible for silicon migration to the surface and hence for the considerable increase in silicon content in the residual char layer. Moreover, the higher percentage of silicon in the char layer is partly due to the higher silicon percentage in the original PAPSi–PSM. Higher oxygen content in PAPSi–PSM (21.80 wt%) than in PAPSi (9.91 wt%) indicates the formation of SiO₂ in the char layer. However, the presence of MBS or PSM (in PAPSi–MBS or PAPSi–PSM) prevents phosphorus (from TPPSi) migrating to the surface and leads to a considerable decrease in phosphorus content in the residual char layer in comparison to PAPSi. Therefore, containing-silicon PSM is responsible for the char shield and solid residue, which induces the high flame retardancy of PAPSi–PSM.

Table 3 Elemental compositions of the residual chars measured by SEM/EDX

Element/%	PAPSi	PAPSi–MBS	PAPSi–PSM
C	87.72	84.67	75.20
O	9.91	14.26	21.80
Si	1.09	0.49	2.34
P	1.28	0.58	0.66

---

3.5. FT-IR analysis of the residues

The FT-IR spectra of the PAPSi and PAPSi–PSM residues from the extinguished specimens formed via vertical burning tests are compared in Fig. 6. The characteristic Si–O stretching absorptions at 1029 and 1125 cm⁻¹ and P–O–C stretching absorption in phosphorus compounds at 940 cm⁻¹ are observed in the spectrum of PAPSi. The absorption bands at 1000–1200 cm⁻¹ become quite strong, and the Si–O bending vibration peaks at 807 cm⁻¹ and 530 cm⁻¹ appear in the spectrum of PAPSi–PSM, indicating the formation of more SiO₂ in the char layer. In addition, the absorption at 1598 cm⁻¹, attributed to the polyaromatic structure, is strengthened while the P–O–C stretching absorption at 940 cm⁻¹ becomes weaker.²⁷ All these results are in reasonable agreement with SEM/EDX analysis.

Fig. 6 FT-IR spectra of the residues’ surfaces.

3.6. Raman analysis of residues

Raman spectra of PC–ABS0, PAPSi and PAPSi–PSM residues are shown in Fig. 7. For all samples, there are two prominent Raman peaks: G (1580–1593 cm⁻¹) and D (1327–1340 cm⁻¹). The intensity ratio value I_D/I_G is called R and is inversely proportional to the in-plane microcrystalline size.²⁹ A larger R value indicates a smaller crystalline region, which is proportional to the pore diameter of the intumescent char layer. The barrier properties to gas and heat rise with decreasing pore diameter of the char residue.³⁰ As seen in Fig. 7, the R value of PAPSi is higher than PC–ABS0, indicating that TPPSi promotes char formation during burning. PAPSi–PSM exhibits the largest R value among the three systems, implying that the combination of TPPSi and PSM further decreases the pore diameter to form the smallest carbonaceous micro-structures during burning, resulting in sufficiently compact char layers during combustion. The results are in reasonable agreement with UL-94 tests and SEM morphologies.

Fig. 7 Raman spectra of the residues.

4. Conclusions

A flame retardant containing silicon and caged bicyclic phosphate groups, tri(2,6,7-trioxa-1-phosphabicyclo[2.2.2]octane-1-oxo-4-hydroxymethyl) phenylsilane (TPPSi), was used for flame-retarding PC–ABS (90/10) blends. The addition of 8 wt% TPPSi endowed the PC–ABS blend with a UL-94 V-0 rating and a LOI of 30%. Accompanied by the addition of 5 wt% polydimethylsiloxane-g-styrene-g-methyl methacrylate copolymer (PSM) as a compatibilizer, the Izod notched impact strength of PC–ABS–TPPSi (PAPSi) was obviously improved without deteriorating its flame retardancy. PSM is a more appropriate compatibilizer for flame-retardant PAPSi than MBS. On the basis of DMA and SEM examinations, PSM improved the compatibility among PC, ABS and TPPSi components to a great extent. TGA analysis showed that the addition of PSM elevated the degradation temperature and considerably decreased the weight loss in the first degradation stage of PAPSi. A higher silicon percentage on the residue surface of PAPSi–PSM was found by SEM/EDX and FT-IR, evidence of more SiO₂ in the char layer. According to the Raman spectra, the combination of TPPSi and PSM decreases the pore diameter to form smaller sized carbonaceous micro-structures during burning of PAPSi–PSM. PC–ABS blends using TPPSi as flame retardants and PSM as compatibilizers showed both high flame retardancies and impact strengths.

Acknowledgements

This research work was supported by a grant from the Cultivation Fund of the Key Scientific and Technical Innovation Project, Department of Education, Guangdong province (cxzd1008), the Project for Construction of the Guangzhou Key Laboratory (no. 2012-224-7), and the Cleaner Production Engineering Technology Research Center of Guangdong College's Light and Chemical industry (31117004).

Notes and references

1. R. Greco and A. Sorrentino, Adv. Polym. Technol., 1994, 13, 249–258.
2. J. S. Wu, S. C. Shen and F. C. Chang, J. Appl. Polym. Sci., 1993, 50, 1379–1389.
3. G. Wildes, H. Keskkula and D. R. Paul, Polymer, 1999, 40, 7089–7107.
4. B. S. Lombardo, H. Keskkula and D. R. Paul, J. Appl. Polym. Sci., 1994, 54, 1697–1720.
5. B. Perret and B. Schartel, Polym. Degrad. Stab., 2009, 94, 2204–2212.
6. S. V. Levchik and E. D. Weil, J. Fire Sci., 2006, 24, 137–151.
7. E. D. Weil, S. V. Levchik, M. Ravey and W. M. Zhu, Phosphorus, Sulfur Silicon Relat. Elem., 1999, 146, 17–20.
8. H. Vothi, C. Nguyen, K. Lee and J. Kim, Polym. Degrad. Stab., 2010, 95, 1092–1098.
9. P. Wei, G. Tian, H. Yu and Y. Qian, Polym. Degrad. Stab., 2013, 98, 1022–1029.
10. L. Becker, D. Lenoir, G. Matuschek and A. Kettrup, J. Anal. Appl. Pyrolysis, 2001, 60, 55–67.
11. M. Katayama and H. Hamano, US Pat. 6, 417, 257, July 9, 2002.
12. M. Bodiger, T. Eckel, D. Wittmann and H. Alberts, US Pat. 5, 849, 827, December 15, 1998.
13. M. C. Despinasse and B. Schartel, Polym. Degrad. Stab., 2012, 97, 2571–2580.
14. B. Perret, K. H. Pawlowski and B. Schartel, J. Therm. Anal. Calorim., 2009, 97, 949–958.
15. W. Zhang, X. Li and R. Yang, Polym. Adv. Technol., 2012, 23, 588–595.
16. E. Wawrzyn, B. Schartel, M. Ciesielski, B. Kretzschmar, U. Braun and M. Döring, Eur. Polym. J., 2012, 48, 1561–1574.
17. W. Y. Chiang and G. L. Tzeng, J. Appl. Polym. Sci., 1997, 65, 795–805.
18. J. Chen, S. Liu and J. Zhao, Polym. Degrad. Stab., 2011, 96, 1508–1515.
19. H. Li, S. Liu, J. Zhao, D. Li and Y. Yuan, Thermochim. Acta, 2013, 573, 32–38.
20. X. Z. Kong and E. Ruckenstein, J. Appl. Polym. Sci., 1999, 73, 2235–2245.
21. S. V. Levchik and E. D. Weil, Polym. Int., 2005, 54, 981–998.
22. S. C. Tjong and Y. Z. Meng, Eur. Polym. J., 2000, 36, 123–129.
23. K. H. Pawlowski and B. Schartel, Polym. Int., 2007, 56, 1404–1414.
24. D. A. Bright, S. Dashevsky, P. Moy and B. Williams, J. Vinyl Addit. Technol., 1997, 3, 170–174.
25. B. Perret and B. Schartel, Polym. Degrad. Stab., 2009, 94, 2194–2203.
26. I. Ravadits, A. Toth, G. Marosi, A. Marton and A. Szep, Polym. Degrad. Stab., 2001, 74, 419–422.
27. S. Liu, L. Jiang, Z. Jiang, J. Zhao and Y. Fu, Polym. Adv. Technol., 2011, 22, 2392–2402.
28. X. Wang, Y. Hu, L. Song, W. Y. Xing and H. D. Lu, J. Polym. Sci., Polym. Phys. Ed., 2010, 48, 693–705.
29. P. Wei, G. Tian, H. Yu and Y. Qian, Polym. Degrad. Stab., 2013, 98, 1022–1029.
30. X. Zhao, P. Wei, Y. Qian, H. Yu and J. Liu, J. Appl. Polym. Sci., 2012, 125, 3167–3174.

---