Comparative study of the synergistic effect of binary and ternary LDH with intumescent flame retardant on the properties of polypropylene composites

Abstract

Organo-modified binary (MgAl-) and ternary (MgZnAl-) layered double hydroxides (LDHs) were synthesized via a co-precipitation method, and subsequently dispersed into polypropylene (PP) in combination with an intumescent flame retardant (IFR) by a melt blending process. A comparative study on the effect of binary and ternary LDHs on the mechanical, thermal properties and flammability of flame retardant PP composites was undertaken. The incorporation of either binary or ternary LDH resulted in a slightly decreased tensile strength compared to neat PP matrix; however, the Young's modulus of PP/IFR/LDH composites was improved. The synergism between either binary or ternary LDH and IFR occurred during the combustion, but the ternary LDH showed superior char-formation ability and smoke suppression over the binary one due to the presence of the element zinc. In contrast to the binary LDH, the ternary LDH produced better quality char that effectively suppressed the spread of the flame and finally extinguished the fire.

---

1. Introduction

Layered double hydroxides (LDHs), also known as anionic clays, are a promising class of layered materials for preparing multifunctional polymer–matrix composites (PMCs).¹ The general chemical formula of an LDH can be illustrated as [M_1−x²⁺M_x³⁺(OH)₂]A_x/nⁿ⁻ mH₂O, where M²⁺ is a divalent cation, M³⁺ is a trivalent cation and Aⁿ⁻ is an interlayer anion.² The species M²⁺, M³⁺, and Aⁿ⁻ together with the value of x are adjustable, and thereby the structure and properties of LDHs could be easily tailored. Besides the fascinating properties aforementioned, the layered structure of LDHs can be used as a versatile intercalation host which can accommodate a wide variety of organic guest species. This procedure can be employed for the fabrication of polymer nanocomposites.

Over the past few decades, incorporating LDHs into polymers as flame retardant additives have received considerable interests due to their high water content, non-toxicity and layered structure.³ Chen and Qu reported that the addition of Mg–Al LDHs significantly improved the thermal stability in polyethylene-graft-maleic anhydride nanocomposites.⁴ Additionally, the exfoliated Mg–Al LDH layers were found to promote the charring process of linear low-density polyethylene.⁵ The flame retardant effect of LDHs on other polymers, such as polypropylene (PP),⁶ ethylene vinyl acetate,⁷ polymethyl methacrylate and polystyrene,⁸ polyvinyl alcohol,⁹ polyvinyl chloride,¹⁰ polylactic acid,¹¹ and epoxy resin¹² has also been investigated. However, these investigations focused on binary metal based LDH, and the flame retardant effect of ternary metal LDH is rarely reported up to now.

In addition to improving resistance to ignition, another key important issue is the ability to suppress smoke production of polymeric materials when a fire occurs, since toxic smoke is responsible for the majority of deaths in fire accidents.¹³ Zinc-based compounds have been considered to be effective in improving the char formation and smoke suppression of polymers.^14,15 Therefore, taking advantage of the versatility in the composition of LDHs, zinc ions could introduced into the layers of LDHs, in order to improve the fire resistance and smoke suppression simultaneously.

In this work, we synthesized two types of organo-modified LDHs, binary MgAl-LDH (b-LDH) and ternary MgZnAl-LDH (t-LDH) through one-step co-precipitation method. Subsequently, b-LDH or t-LDH in combination with one commercially available intumescent flame retardant (IFR) additive was melt blend with PP to prepare flame retardant PP composite. The comparative study on the influence of b-LDH and t-LDH on the mechanical, thermal properties and fire resistance of the resultant PP composites was investigated. Furthermore, the flame retardant mechanism of the PP/IFR/b-LDH and the PP/IFR/t-LDH composites was proposed.

2. Experimental

2.1. Materials

Mg(NO₃)₂·6H₂O, Zn(NO₃)₂·6H₂O, Al(NO₃)₃·9H₂O, sodium hydroxide and sodium dodecylbenzenesulfonate (SDBS) were purchased from Sigma Aldrich Chemical Company without further purification. Polypropylene (commercial type: HD120MO, density: 0.908 g cm⁻³) was supplied by the Borealis Incorporation (Austria). Coated intumescent combination based on ammonium polyphosphate (Budit® 3167) was provided by the Budenheim Incorporation (Germany) and abbreviated as IFR. Deionized water is used for all experiments unless otherwise stated.

2.2. Synthesis of organo-modified MgZnAl-LDH via co-precipitation method

Organo-modified MgZnAl-LDH (m-LDH) was synthesized via co-precipitation method, using SDBS as the modifier. Mg(NO₃)₂·6H₂O (25.64 g), Zn(NO₃)₂·6H₂O (89.25 g), and Al(NO₃)₃·9H₂O (75.03 g) were dissolved in the 500 ml of deionized water. The mixed metal salt solution was slowly added to the 500 ml of aqueous solution containing 62.73 g SDBS under continuous stirring with magnetic stirrer at 50 °C in a three-necked round bottom flask. The precipitate appeared by adding 0.7 M NaOH solution, with the pH value kept at 10 ± 0.2. The resulting slurry was stirred for 0.5 h and aged for 18 h at 75 °C. Finally, the precipitate was filtrated and washed with deionized water and ethanol until the pH was about 7. The product was dried at 90 °C under vacuum for 6 h. For comparison, organo-modified MgAl-LDH was also prepared using the same route without the addition of Zn(NO₃)₂·6H₂O.

2.3. Compounding of intumescent flame retardant PP composites

The PP, IFR, b-LDH and t-LDH was dried under vacuum at 90 °C for 12 h before use. The samples were compounded using a Lab compounder of Brabender KETSE 20/40 D EC twin-screw system. The melt compounding temperature was set at 210 °C, while the screw speed was 130 rpm. The formulations of various samples are given in Table 1.

Table 1 The formulations of PP-based composite samples

Samples	Formulations/wt%
	PP	IFR	b-LDH	t-LDH
PP	100	0	0	0
PP/IFR20	80	20	0	0
PP/IFR18/b-LDH2	80	18	2	0
PP/IFR16/b-LDH4	80	16	4	0
PP/IFR18/t-LDH2	80	18	0	2
PP/IFR16/t-LDH4	80	16	0	4

---

2.4. Characterization

The powder X-ray diffraction (XRD) patterns were recorded with a Philips X'Pert PRO diffractometer (Cu K_α radiation, λ = 0.15405 nm), operating at 45 kV and 40 mA. Fourier transform infrared (FTIR) spectra were carried out on a NICOLET iS50 FTIR spectrometer using KBr disc method. Thermogravimetric analysis (TGA) was conducted on a Q50 thermal analyzer (TA Instruments, USA) under nitrogen atmosphere with a ramp of 10 K min⁻¹ from room temperature to 800 °C. The limiting oxygen index (LOI) was measured according to ASTM D2863 by an oxygen index meter (Fire Testing Technology, UK). The specimens used for the test were of dimensions 100 mm × 6.5 mm × 3 mm. For each sample, five parallels were carried out to obtain the average value. The UL-94 vertical burning test was performed using a vertical burning instrument (Fire Testing Technology, UK), and the specimens for testing were of dimensions 130 mm × 13 mm × 3 mm. For each sample, five parallels were carried out to obtain the average value. Flammability of the samples was also characterized by dual cone calorimeter (Fire Testing Technology, UK) according to ISO 5660. Samples were mounted into aluminum foil (without frame and grid) and irradiated horizontally at a heat flux of 50 kW m⁻². The dimensions for the samples were 100 mm × 100 mm × 3 mm. Measurement was done in triplicate and the average data were reported. The tensile tests were carried out by INSTRON 5966 dual column tabletop universal testing system according to ASTM D3039-08 method, at a crosshead speed of 10 mm min⁻¹. Five parallels for each sample were tested and the average value was reported. Morphology and element analysis of the chars was investigated by scanning electron microscopy (SEM; EVO MA15, Zeiss) with energy-dispersive X-ray (EDX; Oxford INCA 350) analyzer at accelerating voltages of 18 kV. Samples were coated with a conductive layer of gold before observation.

3. Results and discussion

3.1. Structural characterization

XRD patterns of MgZnAl-LDH and organo-modified MgZnAl-LDH (t-LDH) are displayed in Fig. 1. As can be seen, MgZnAl-LDH exhibits the typical profile of LDH materials with sharp intense peaks at low theta values, whereas they become weaker and less defined at higher angular values. The characteristics bands at 2 theta = 10.1°, 20.1° and 60.5° are ascribed to the (003), (006) and (110) diffraction peaks, respectively, which are in good agreement with the previous literature.¹⁶ From these parameters, the basal spacing of MgZnAl-LDH is estimated to be 0.88 nm according to the Bragg equation. The (003) basal reflection of organo-modified MgZnAl-LDH shifts to 2 theta = 3.2° and the higher order reflections also shift to lower angle, indicating that SDBS anions have been intercalated into the interlayer galleries giving an increased interlayer spacing (d = 2.76 nm). Additionally, no residual peaks have been observed, thus confirming the high purity of the products. Similar phenomenon is observed in the XRD profile of MgAl-LDH and organo-modified MgAl-LDH (b-LDH) (see Fig. S1 in ESI†).

Fig. 1 Powder XRD profiles of unmodified MgZnAl-LDH and organo-modified MgZnAl-LDH.

FTIR spectra of MgZnAl-LDH and organo-modified MgZnAl-LDH are shown in Fig. 2. MgZnAl-LDH shows some similar absorption peaks to organo-modified MgZnAl-LDH: the broad peak around 3490 cm⁻¹ can be ascribed to the stretching of OH groups attached to Al, Mg and Zn ions in the layers;¹⁷ the peak at 1626 cm⁻¹ is assigned to the bending vibration of interlayer water; the strong band at 1385 cm⁻¹ is attributed to the asymmetric stretching of the carbonate anion. However, compared with MgZnAl-LDH, some new peaks appear in the FTIR spectra of organo-modified MgZnAl-LDH. The appearance of the –CH₃ and –CH₂– stretching peaks (2930 and 2860 cm⁻¹) together with the sulfonate stretching bands (1185 and 1036 cm⁻¹)¹⁸ confirms that SDBS has been exchanged into the interlayer space of MgZnAl-LDH. Similar phenomenon is observed in the FTIR spectra of MgAl-LDH and organo-modified MgAl-LDH (see Fig. S2 in ESI†).

Fig. 2 FTIR spectra of (a) MgZnAl-LDH and (b) organo-modified MgZnAl-LDH.

3.2. Mechanical properties

Fig. 3a shows typical tensile stress versus strain curves for the control sample, PP/IFR20 and PP/IFR/LDH composites. Fig. 3b compares the Young's modulus and the ultimate tensile strength of the neat PP and the composite samples. As can be observed, the neat PP exhibits a tensile strength of 38.1 MPa and a Young's modulus of 1.12 GPa. After incorporating 20 wt% IFR into PP matrix, the tensile strength of the resultant PP composite is decreased to 31.9 MPa, but the Young's modulus is increased to 1.47 GPa. In general, additives used in relatively high loadings lead to the reduction in tensile strength of the polymer matrix.^19,20 The reduction in the tensile strength is due mainly to poor additive-matrix compatibility as a consequence of the high amount of additives. The Young's modulus of the composite samples is enhanced compared to that of pure PP, due to the high stiffness of additives. The optimum enhancement in the Young's modulus is observed for PP/IFR18/t-LDH2, with a 31% increment relative to the pure PP. Conversely, at the higher loading fractions, the LDH will likely form agglomerates, resulting in poorer interfacial interactions and therefore the Young's modulus slightly decreases.

Fig. 3 (a) Static stress versus strain plots of the PP and its composites; (b) ultimate tensile strength and Young's modulus of the PP and its composites.

3.3. Thermal degradation behaviors

The typical TG profiles for neat PP and its composites under nitrogen atmosphere are presented in Fig. 4. The relative thermal stability of the samples is evaluated by the temperature at 5% weight loss (T_−5%), the temperature at maximum mass loss rate (T_max) and the char residual percentage at 800 °C, as listed in Table 2. From Fig. 4, it can be observed that pure PP shows a single degradation stage ranging from 390 to 490 °C, corresponding to a T_max at 466 °C. In contrast to neat PP, the PP/IFR and PP/IFR/LDH composites exhibit a reduction in T_−5%, which is probably attributed to the earlier decomposition of the thermally instable additives (including IFR, and SDBS in b-LDH and t-LDH). This earlier thermal degradation of PP composites is believed to be due to the char formation catalyzed by the intumescent flame retardant system. The presence of char layer can slow down the heat release upon the ignition of PP, and thus protect the underlying matrix from being consumed by flame. Therefore, the decreased thermal stability is likely essential rather than a drawback of the intumescent flame retardant system. The mass loss of neat PP is nearly 100%, with no char yield left. The addition of IFR improves the char yield to 8.6% for PP/IFR20, and the simultaneous addition of IFR and t-LDH (or b-LDH) further enhances the char yield. Meanwhile, t-LDH exhibits superior char-formation ability over b-LDH at the equivalent filler loading. PP/IFR18/t-LDH2 composite shows the highest char yield (14.0%), suggesting the best synergistic effect at this formulation.

Fig. 4 Thermogravimetric analysis curves of PP, PP/IFR, PP/IFR/b-LDH and PP/IFR/t-LDH composites.

Table 2 Thermal analysis data of PP and its composite samples

Samples	T−5%/°C	Tmax/°C	Char yield at 800 °C/wt%
			Calculated	Measured
PP	421	466	—	0.2
PP/IFR20	381	470	4.8	8.6
PP/IFR18/b-LDH2	365	470	5.3	6.4
PP/IFR16/b-LDH4	346	468	5.8	8.8
PP/IFR18/t-LDH2	356	474	4.9	14.0
PP/IFR16/t-LDH4	348	474	5.0	9.9

---

To confirm the synergism between IFR and t-LDH (or b-LDH), the theoretical char yield was calculated by linear combination between the TG curves of neat PP, IFR and LDH.²¹ The formula is given as follows:

W_th(T)_PP/IFR/LDH = x × W_exp(T)_pp + y × W_exp(T)_IFR + z × W_exp(T)_LDH, x + y + z = 1

where W_exp(T)_PP: experimental TG curve of the pure PP; W_exp(T)_IFR: experimental TG curve of IFR; W_exp(T)_LDH: experimental TG curve of t-LDH (or b-LDH); x, y, z are the mass fractions of the PP, IFR and t-LDH (or b-LDH) in the composites, respectively. The experimental and theoretical char yield for all the samples is listed in Table 2. Obviously, the calculated char yields are lower than the measured ones, indicating the synergistic effect of IFR and t-LDH (or b-LDH) on the char formation of the PP composites.

3.4. Flame retardant properties

Cone calorimeter was widely used to evaluate the fire retardant behaviors of the polymer composites. The important parameters, including time-to-ignition (TTI), peak heat release rate (PHRR), total heat release (THR) and fire growth rate index (FIGRA), are summarized in Table 3. Fig. 5 displays the heat release rate versus time curves of all the samples. It can be observed that pure PP burns very rapidly after ignition, showing a single sharp peak with the PHRR value of 1450 kW m⁻². As expected, incorporating 20 wt% flame retardant additives into PP significantly decreases the PHRR to 300 kW m⁻², an approximately 80% reduction in PHRR. Meanwhile, the shape of HRR curve exhibits two peaks, suggesting the degradation behavior of PP/IFR20 has changed. Substituting a portion of IFR by t-LDH (or b-LDH), a slight reduction in the first PHRR value is observed compared to PP/IFR20, but the second PHRR value is obviously suppressed. In detail, the combustion process for PP/IFR/LDH composites can be divided into four stages based on the HRR curves: (1) pre-degradation of PP; (2) main burning process; (3) char formation; (4) oxidation of char residues. The incorporation of both binary and ternary LDH has a significant influence on the fourth stage which enhances the thermal oxidative resistance of the char layer, answering for the disappearance of the second peak of HRR curves for PP/IFR/LDH composites.

Table 3 Cone calorimeter data of PP and its composites under the heat flux of 50 kW m⁻²

Samples	TTI/s	PHRR/kW m^−2	THR/MJ m^−2	TSP/m^2	FIGRA/kW m^−2 s^−1
PP	36 ± 1	1373 ± 36	174.8 ± 0.7	18.61 ± 0.11	7.15
PP/IFR20	22 ± 2	326 ± 2	149.9 ± 4.2	21.75 ± 1.39	1.19
PP/IFR18/b-LDH2	18 ± 1	286 ± 6	108.8 ± 3.1	14.25 ± 0.48	0.85
PP/IFR16/b-LDH4	15 ± 1	326 ± 3	120.3 ± 5.3	9.34 ± 0.45	1.17
PP/IFR18/t-LDH2	16 ± 1	306 ± 2	136.0 ± 1.6	14.76 ± 0.76	0.99
PP/IFR16/t-LDH4	14 ± 2	265 ± 5	90.4 ± 0.9	7.21 ± 0.35	0.79

---

Fig. 5 Heat release rate versus time curves of PP and its flame retardant composites.

The total heat release versus time curves of all the samples are displayed in Fig. 6. As can be observed, pure PP releases heat very fast and the total heat release reaches the maximum value (174.8 MJ m⁻²) after 300 s. The addition of 20 wt% IFR inhibits the heat release and the total heat release of PP/IFR20 is reduced to 149.9 MJ m⁻² after burning. Simultaneous addition of IFR and LDH further reduces the total heat release. The reduction in the total heat release indicates that more PP chains participate in the carbonization process and therefore, less volatile products that serve as “fuel” go into the gas phase. As a result, the THR values are dramatically reduced.

Fig. 6 Total heat release versus time curves of PP and its flame retardant composites.

From Table 3, it can be found that incorporation of the flame retardant additives has a notable influence on the TTI of the PP composites. The TTIs of PP/IFR20 and PP/IFR/LDH composites decrease compared to that of pure PP. The reduced TTI can be attributed to the thermally unstable IFR and/or m-LDH catalyzes the earlier ignition of the PP composites, which is consistent with the lowered onset thermal stability as discussed in TG results.

As far as the TSP is concerned, all the PP composites containing flame retardant additives show lower TSP compared to pure PP except PP/IFR20. The increase in smoke production in PP/IFR20 is probably attributed to incomplete combustion of PP induced by intumescent flame retardant. The co-addition of IFR and t-LDH (or b-LDH) suppresses the smoke production. This reduced smoke production can be explained by that the presence of LDH inhibits PP molecules from converting into the organic volatiles, as the organic volatiles are the major source of smoke particles.²² The ternary LDH exhibits more effective smoke suppression than the binary one due to the introduction of zinc element.

The FIGRA calculated from the ratio of PHRR and time to PHRR is regarded as the growth rate of the burning intensity. With regard to the FIGRA, all the PP composites containing flame retardant additives show decreased FIGRA compared to pure PP. The most striking result is observed in the case of PP/IFR16/t-LDH4, an 89% reduction in FIGRA relative to pure PP. This notable reduction in FIGRA indicates the significantly suppressed fire hazard of the material.

The effect of b-IFR and t-LDH on the LOI values and UL-94 results of the flame retardant PP composites is presented in Fig. 7. Pure PP exhibits a LOI value of 18.5% and is not classified in the UL-94 vertical test. When 20 wt% of IFR is added, the LOI value goes up to 29.5%, and this formulation can pass the UL-94 V0 rating. When IFR is partly substituted by t-LDH, LOI values are slightly higher than that with IFR alone. An optimum is observed at 2.0 wt% of t-LDH and 18.0 wt% of IFR, exhibiting the highest LOI of 32.5% and UL-94 V0 rating in vertical burning test. It is noted that PP/IFR16/t-LDH4 shows UL-94 V0 rating whereas PP/IFR16/b-LDH4 no rating. In the vertical burning test of PP/IFR16/t-LDH4, the intumescent char can activate quickly after flame exposure, which extinguish the flame propagation. In contrast, PP/IFR16/b-LDH4 cannot form thermally stable char layer so that flame spreads quickly along the sample (see Fig. S3 in ESI†). The morphology of the char residues is also investigated by SEM to further understand the flame retardant mechanism which will be discussed in the following section.

Fig. 7 The LOI values and UL-94 results of PP and its flame retardant composites.

3.5. Flame retardant mechanism

Flame retardant additives in polymers may function either in the condensed phase or in the gas phase or in both phases at the same time. The condensed phase action plays a predominant mechanism in the intumescent flame retardant system, so the flame retardant efficiency depends strongly on the structure and composition of the char during burning.²³ Therefore, investigating the morphology of the resulting carbonaceous layers will provide an insight into understanding the flame retardant mechanism.

Morphologies and elemental analysis of the chars of PP/IFR16/b-LDH4 and PP/IFR16/t-LDH4 obtained from LOI tests were investigated by SEM, as shown in Fig. 8. As can be observed, the char of PP/IFR16/b-LDH4 presents a cracked and porous surface, indicating the occurrence of fierce combustion. Such a broken char layer cannot resist the attack by flame. In contrast, the char of PP/IFR16/t-LDH4 shows a compact and continuous surface, implying that the char is thermally stable enough to inhibit the flame spread. The change in the chemical components of the chars between PP/IFR16/b-LDH4 and PP/IFR16/t-LDH4 is also impressive. The C/O ratio of the residual chars of PP/IFR16/t-LDH4 is 5.12, which is much higher than that of PP/IFR16/b-LDH4 (1.99), implying that the incorporation of t-LDH improves the thermal oxidative resistance of the char. These findings correspond well with the great difference between PP/IFR16/b-LDH4 (no rating) and PP/IFR16/t-LDH4 (V0 rating) observed from UL-94 vertical burning tests. Undoubtedly, this char layer with high thermal resistance and compact surface can effectively suppress the spread of the flame and finally extinguish the fire.

Fig. 8 SEM images and elemental analysis of the residual chars of (a) PP/IFR16/b-LDH4; and (b) PP/IFR16/t-LDH4.

4. Conclusion

Organo-modified binary MgAl-LDH (b-LDH) and ternary MgZnAl-LDH (t-LDH) were successfully synthesized via one-step co-precipitation method, and their structures were confirmed by XRD and FTIR. Intumescent flame retardant PP composites based on IFR and organo-modified LDH were prepared by melt blending process. The incorporation of either b-LDH or t-LDH enhanced the Young's modulus of the flame retardant PP composites due to the rigidity of the fillers, despite of a slight decrease in the tensile strength. TGA results showed that the earlier degradation of IFR catalyzed the char formation, and the addition of either b-LDH or t-LDH further improved the char yield. Keeping the total filler loading at 20 wt%, the PHRR of all the PP composite samples was reduced by more than 80% compared to pure PP, and the incorporation of t-LDH showed more effective in lowering heat release and smoke production than b-LDH owing to the introduction of zinc element. In the UL-94 vertical burning test, PP/IFR16/b-LDH4 (no rating) exhibited a significantly different behavior from PP/IFR16/t-LDH4 (V0 rating). This notable difference in the fire retardant behaviors of PP/IFR/LDH composites was probably attributed to that t-LDH was able to form thermally stable and compact char layer which effectively suppressed the spread of the flame and finally extinguished the fire. Due to the adjustable metal ions and intercalation guest species, the ternary LDHs can be used as a versatile intercalation host which can accommodate a wide variety of organic guest species. This procedure can be employed for the fabrication of high performance flame retardant polymer composites.

Acknowledgements

The authors want to acknowledge “AMAROUT II EUROPE” Marie Curie COFUND action, partially funded by the European Union's 7th Framework Programme under Grant Agreement n°291803. Also, this work is partly funded by the European Project COST Action MP1105 “FLARETEX”, and Ramón y Cajal grant (RYC-2012-10737).

Notes and references

1. Z. Matusinovic and C. A. Wilkie, J. Mater. Chem., 2012, 22, 18701–18704.
2. N. J. Kang, D. Y. Wang, B. Kutlu, P. C. Zhao, A. Leuteritz, U. Wagenknecht and G. Heinrich, ACS Appl. Mater. Interfaces, 2013, 5, 8991–8997.
3. C. Coluccini, Y. Sporer, A. Leuteritz, I. Kuehnert and D. Y. Wang, Nanomater. Energy, 2014, 3, 177–191.
4. W. Chen and B. J. Qu, Chem. Mater., 2003, 15, 3208–3213.
5. L. C. Du and B. J. Qu, J. Mater. Chem., 2006, 16, 1549–1554.
6. D. Y. Wang, A. Leuteritz, B. Kutlu, M. A. d. Landwehr, D. Jehnichen, U. Wagenknecht and G. Heinrich, J. Alloys Compd., 2011, 509, 3497–3501.
7. C. M. Jiao and X. L. Chen, J. Therm. Anal. Calorim., 2009, 98, 813–818.
8. C. Nyambo, P. Songtipya, E. Manias, M. M. Jimenez-Gasco and C. A. Wilkie, J. Mater. Chem., 2008, 18, 4827–4838.
9. S. Huang, X. Cen, H. Peng, S. Guo, W. Wang and T. Liu, J. Phys. Chem. B, 2009, 113, 15225–15230.
10. Z. P. Xu, S. K. Saha, P. S. Braterman and N. D'Souza, Polym. Degrad. Stab., 2006, 91, 3237–3244.
11. D. Y. Wang, A. Leuteritz, Y. Z. Wang, U. Wagenknecht and G. Heinrich, Polym. Degrad. Stab., 2010, 95, 2474–2480.
12. E. N. Kalali, X. Wang and D. Y. Wang, J. Mater. Chem. A, 2015, 3, 6819–6826.
13. X. Wang, W. Xing, X. Feng, B. Yu, H. Lu, L. Song and Y. Hu, Chem. Eng. J., 2014, 250, 214–221.
14. Y. Ning and S. Guo, J. Appl. Polym. Sci., 2000, 77, 3119–3127.
15. S. D. Jiang, Z. M. Bai, G. Tang, Y. Hu and L. Song, Ind. Eng. Chem. Res., 2014, 53, 6708–6717.
16. Y. Lin, J. Wang, D. G. Evans and D. Li, J. Phys. Chem. Solids, 2006, 67, 998–1001.
17. F. Malherbe and J. P. Besse, J. Solid State Chem., 2000, 155, 332–341.
18. V. Deimede, G. Voyiatzis, J. Kallitsis, L. Qingfeng and N. Bjerrum, Macromolecules, 2000, 33, 7609–7617.
19. N. A. Isitman and C. Kaynak, Polym. Degrad. Stab., 2010, 95, 1523–1532.
20. J. Dai and B. Li, J. Appl. Polym. Sci., 2010, 116, 2157–2165.
21. S. Giraud, S. Bourbigot, M. Rochery, I. Vroman, L. Tighzert and R. Delobel, Polym. Degrad. Stab., 2002, 77, 285–297.
22. Y. Y. Dong, Z. Gui, Y. Hu, Y. Wu and S. H. Jiang, J. Hazard. Mater., 2012, 209, 34–39.
23. X. Wang, S. Zhou, W. Y. Xing, B. Yu, X. M. Feng, L. Song and Y. Hu, J. Mater. Chem. A, 2013, 1, 4383–4390.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra15565g

---