Improved ceramifiable properties of EVA composites with whitened and capsulized red phosphorus (WCRP)

Abstract

Ceramifiable EVA composites were prepared by addition of WCRP and silicate glass frits (SGFs) to EVA resin. Ceramic residues were obtained by firing ceramifying EVA composites at 600, 800 and 1000 °C for 30 min. The effect of WCRP on the self-supporting property of the composites and the effect of firing temperatures on the flexural strength of the ceramic residues for the composites are investigated. Results show that the addition of WCRP improves the self-supporting property of the composites, and the flexural strength of the ceramic residues containing phosphorus is enhanced with increasing temperature. XRD analysis shows that addition of WCRP promotes the precipitation of cristobalite (SiO₂), which leads to an improvement in the properties. XRF and FT-IR demonstrate the WCRP in the composite could be converted into [PO₄]³⁻ at high temperature. It can be concluded that phase separation between the phosphates and SGFs causes the formation of the crystalline phase. SEM confirms that the number of pores in the residues decreases with increasing temperature, which results in the increase in flexural strength.

---

1. Introduction

Ceramifying composites, possessing the performance features of polymers at room temperature and of ceramics with increasing temperature, have been widely applied in the fire protection field including as fire resistant sheaths for cables,¹ barrier coatings^2,3 and foam sealants.⁴ The idea was first presented by Australian researchers.^5–8 They pointed out that a liquid phase formed at amorphous silica/mica interfaces via a eutectic reaction, and resulted in significant improvements in the strength of residues obtained by firing ceramifiable silicone composites. However, the strength of residues fired at low temperatures (<800 °C) remained poor, which greatly limited the application of the ceramifying composites. Therefore, glass frits with low softening points were introduced to improve the strength of the composites at low temperatures.⁹

In the past few years, the variety of ceramifiable fillers has been further expanded to include surface-modified montmorillonies,¹⁰ kaolin,¹¹ carbon fibers¹² and aluminium hydroxide.¹³ Moreover, in order to lower the cost of matrixes and extend the applied field of ceramifying composites, attention has been shifted to organic polymers as the matrixes. Fire-resistant sheathing was prepared by addition of glass frits and silicate minerals into various polymers including EPDM, HDPE, and PVC.¹⁴ The conclusion that phosphate glass frits can promote the formation of a strong residue was drawn. And then ceramifying poly(vinyl acetate) (PVAc) sealants were prepared by adding talc or kaolin as structural fillers and zinc borate as a flux as well.¹⁵ Ethylene-vinyl acetate (EVA) filled with glass dust, glass fibre, OMMT and melamine cyanurate was developed as a ceramifiable polymer composite for cables and insulated wires.¹⁶

Whitened and capsulized red phosphorus (WCRP) was often added to improve the fire retardancy of polymers,^17–19 but no studies regarding the influence of WCRP on the ceramifiable properties of the composites have been reported so far. The improvement of ceramifiable properties is important so that the polymer composite can continue to provide a barrier to the spread of fire after the polymer has burned off.

In this work, ceramic residues were obtained after ceramifying EVA composites were fired at 600, 800 and 1000 °C for 30 min. The self-supporting property of the composites and the flexural strength and dimensional change of the residues were tested to evaluate the ceramifiable properties of the composites. XRF, FT-IR, XRD and SEM were used to study the ceramifying process. These results show that a new crystalline phase was formed due to the addition of WCRP and SGF over a temperature range where EVA undergoes thermal degradation. Formation of this crystalline phase improved the self-supporting property of the composites.

2. Experimental section

2.1 Materials

EVA (MFI = 3 g/10 min, VA content = 28 wt%) from DuPont Corporation was used in this work. Glass frits (SGFs, grain diameters of 200 mesh, softening point = 645 °C) were manufactured by Donggu New Material Co. Ltd (Foshan, China) and their chemical compositions were estimated by X-ray fluorescence (XRF) and are summarized in Table 1. Whitened and capsulized red phosphorus (WCRP, containing 35 wt% red phosphorus) was supplied by Hualing Chemicals Co. Ltd (Shenzhen, China).

Table 1 Major components of SGF tested by XFR

Element	SiO2	Na2O	TiO2	F	Al2O3	CaO	Others
Content (wt%)	66.32	19.76	6.12	3.71	1.28	0.96	1.85

---

2.2 Sample preparation

Composites were prepared on a two-roll mill at 120 °C for 15 min. Various EVA-based composites were moulded into flat sheets by compression moulding at 150 °C for 10 min under 10 MPa pressure. Symbols and related formulations of compounds used in this study are summarised in Table 2.

Table 2 Compound formulation (g)

Components	Composition code
	A	B
EVA	35	35
SGF	65	50
WCRP		15

---

2.3 Muffle furnace pyrolysis

Pyrolysis of the sheet samples was performed using a muffle furnace, and the samples in Table 2 were held at 600, 800 and 1000 °C for 30 min before naturally cooling to room temperature. Their ceramic residues were defined as A600, A800 A1000, B600, B800 and B1000 respectively.

2.4 Characterisation

2.4.1 Self-supporting measurements. The self-supporting property is included in the ceramifiable properties to evaluate whether the ceramifying composites support weight themselves.^14,20–22 The measuring method was conducted as follows: (1) specimens of dimensions 35–50 mm × 8 mm × 4 mm made from the compositions in Table 2, were placed on a rectangular piece of refractory material so that their long axis was perpendicular to one edge of the supporting refractory blocks and the distance between the two refractory blocks was about 12 mm; (2) all the specimens were held at 600, 800 and 1000 °C for 30 min in a muffle furnace; (3) the bending angles of the residues to the original position were tested.

2.4.2 Flexural strength measurements. The flexural strength of B600, B800 and B1000 with the dimensions of 50 mm × 5 mm × 3 mm was determined by the three-point bend method using a universal testing machine (CMT-6203; MTS System Corporation, China). Loading was applied using a cross-head speed of 0.5 mm min⁻¹.

The flexural strength was calculated using eqn (1):

S = 3PL/(2bd²) (1)

where P is the maximum load (N), L is the outer support span (mm), b is the specimen width (mm) and d is the specimen thickness (mm), to give S in MPa.

2.4.3 Dimensional stability. All the specimens were fired under the same conditions as outlined above. The length of the specimens was precisely measured before and after firing. The changes in dimensional length were calculated using eqn (2):

S = (L_d − L_o)/L_o × 100% (2)

where S is the dimensional change (%), L_d is the dimensional length after firing (mm) and L_o is the original specimen length (mm).

2.4.4 XRD. XRD measurements of SGFs and all the residues were conducted from 5° to 60° at a scan rate of 10° min⁻¹ using a Rigaku X-ray diffractometer using Cu Kα radiation. Relative crystallinity^13,14 was calculated using eqn (3):

Relative crystallinity = I_s/I_r × 100% (3)

where I_s is the strongest diffraction peak of a sample, and I_r is the strongest diffraction peak of the reference sample.

2.4.5 XRF. Chemical compositions of the SGFs were analysed using an X-ray fluorescence method using an ADVANT’ XP XRF spectrometer. Glass sample disks (40 mm in diameter) were prepared using a mixture of flux (borate glass and ammonium nitrate in an approximate ratio of 5.43 : 1). The mixture was heated in a crucible for 15 min at 1050 °C and the melt was then poured into a graphite disc held on a hot plate at about 220 °C. An aluminium plunger was then brought down to gently mould and quench the melt.⁹

2.4.6 FT-IR. The SGFs and the residues from pyrolysis of sample B at different temperatures were analysed by FT-IR.

2.4.7 Morphology analysis. The surfaces of the residues after firing sample B at different temperatures were analysed using a JSM-6510 scanning electron microscope (SEM). Analysis was performed at magnifications of 500× and 5000×.

3. Results and discussion

3.1 The self-supporting property of ceramifying compositions fired at different temperatures

The self-supporting property of the ceramifying EVA-based composites was evaluated by examining the bending angles of the residues to their original position.

It is shown in Table 3 that the bending angle of B600 decreased by 25°, compared with that of A600. With increasing temperature, the difference between A800 and B800 further increased. When the samples in Table 2 were fired at 1000 °C, the bending angle of A1000 cannot be tested because it was melted at this temperature, while B1000 can still support its own weight. Therefore, it is inferred that the addition of WCRP greatly improves the self-supporting property of ceramifying EVA composites.

Table 3 Bending angles of residues

Residue	A600	A800	B600	B800	B1000
Bending angle (°)	34.5	90	9.5	19.7	30.9

---

3.2 Flexural strength of the residues

Fig. 1 shows that the flexural strength and change in the dimensional length of sample A fired at 800 and 1000 °C cannot be tested because their shapes do not meet the testing requirements for these properties.

Fig. 1 Photos of both of the samples (1) and their residues after firing at 600 (2), 800 (3) and 1000 °C (4). The left bar is sample A and the right sample B.

The flexural strength of sample B fired at various temperatures is shown in Fig. 2. A sharp rise in the flexural strength can be seen when the firing temperature increased from 600 °C to 800 °C and a slow rise appeared with a further increase in the temperature.

Fig. 2 Flexural strength and dimensional changes (expansion (+)/shrinkage (−)) of sample B at different temperatures.

As can be seen in Fig. 2, sample B fired at 600 °C shows an expansion, but it shows shrinkage when fired at higher temperatures. Therefore, the shrinkage is raised with increasing temperature.

3.3 X-ray analysis

It is shown in Fig. 3 that the XRD patterns of A600, A800 and A1000 just contain an amorphous peak. This result clearly indicates that no new phase has been found when sample A was fired at various temperatures.

Fig. 3 XRD spectra of a SGF (1) and sample A heated at 600 °C (2), 800 °C (3) and 1000 °C (4) for 30 min.

For sample B, an amorphous hump can be seen in Fig. 4(1). But the XRD patterns for the other residues possess all peaks corresponding to cristobalite (SiO₂).^23,24

Fig. 4 XRD spectra of sample B heated at 600 (1), 800, (2) and 1000 °C (3) for 30 min.

Fig. 5 shows two broad bands for the SGF, one at 1022 cm⁻¹ and a weaker one at 485 cm⁻¹, which can be assigned to vibrations of tetrahedral SiO₄.²⁵ Compared with the FT-IR spectrum of the SGF, there appear to be three new absorption bands at 1262, 896 and 790 cm⁻¹ in the spectra of all the residues containing phosphorous. According to Wang’s study,¹⁷ the peak at 1262 cm⁻¹ has been attributed to the stretching vibrations of the P=O bonds in tetrahedral PO₄. In addition, the bands near 896 and 790 cm⁻¹ can be assigned to the asymmetric stretching modes and the bending modes of the P–O–P bonds in the above units, respectively.²⁶ Thus it can be concluded that WCRP could be transformed into [PO₄]³⁻ when fired at various temperatures.

Fig. 5 IR absorption patterns of a SGF (1) and sample B heated at 600 °C (2), 800 °C (3) and 1000 °C for 30 min.

However, there is no crystalline phase containing phosphate in any of the residues with phosphorus. Therefore, it is inferred that incompatibility between phosphates and silicate glass frits can bring about phase separation, which greatly facilitates precipitation of the cristobalite at temperatures above 600 °C. This result is consistent with the reported literature.^27,28

The relative crystallinity of sample B fired at 600, 800 and 1000 °C is shown in Table 4. The content of the crystalline phase was considerably increased with increasing temperature, though an opposite result appeared when the firing temperature exceeded 800 °C. Meanwhile, according to prior studies,^29,30 it is well known that the nucleation rate and crystal growth rate are the main reasons for the transformation of the glass phase into the crystalline phase. Therefore the phase transformation process has been discussed.

Table 4 Relative crystallinity of sample B fired at 600, 800 and 1000 °C

Residues	Relative crystallinity (%)
B600	0
B800	100
B1000	75

---

When sample B was fired at 600 °C, the viscosity of the SGF in B600 was too high, because the firing temperature was below 650 °C, which is the glass transition temperature of the SGF. The removal of Si⁴⁺ and O²⁻ ions was extremely difficult, which greatly limited the nucleation and the growth of cristobalite. When the firing temperature increased to 800 °C, a distinct decline in the viscosity of the SGF in B800 causes a significant improvement in the nucleation rate and the growth rate. Consequently, the relative crystallinity was improved enormously.

However, with a further increase in the firing temperature, a sharp fall in the viscosity of the SGF in B1000 could result in boundary adhesion among the glass particles before crystallization.³¹ The number of crystal nuclei on the interface between the phosphate and the SGF decreased. In addition, the difference between the free energy of the melted glass and the cristobalite decreased with elevated temperature, which leads to a decrease in the crystal growth rate.³² Thus the amount of precipitated cristobalite was reduced.

3.4 Microstructure analysis

Fig. 6 shows the SEM images of sample B fired for 30 min at 600, 800 and 1000 °C.

Fig. 6 SEM images of B600 (a), B800 (b) and B1000 (c) taken at a magnification of ×500.

As is shown in Fig. 6a, there are a lot of pores and even cracks after firing at 600 °C in B600.

A distinct decrease in the number of pores is shown in Fig. 6b, and no obvious large pores can be seen in B1000 as shown in Fig. 6c. This suggests that B1000 is the densest of all the residues. Therefore, it is concluded that a reduction in the viscosity of the SGF in sample B at elevated temperatures contributes greatly to the inter-infiltration between these liquid phases,⁷ which leads to a decline in the number of pores. This is the reason why the flexural strength was improved with increasing temperature.

4. Conclusions

The self-supporting property, flexural strength and dimensional change of ceramic-like body resulting from pyrolysis of EVA-based composites have been discussed employing XRF, FT-IR, XRD and SEM analysis. These results indicate that cristobalite precipitated at 800 and 1000 °C because of the phase separation between the phosphate and the SGF. The formation of the crystalline phase is expected to enhance the ceramifiable properties of ceramifying EVA composites. SEM images reveal that the improvement in the flexural strength at elevated temperatures is due to the increased density of the residues.

Acknowledgements

This work was supported by the Innovation Foundation for Graduate Students of Jiangsu Province (KYLX15_0778), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and the Natural Science Foundation of Jiangsu Province (Grants No. BK20150956).

References

1. H. Shao, T. Wang and Q. Zhang, Adv. Compos. Lett., 2010, 19, 159.
2. Y. B. Cheng, K. W. Thomson and R. A. Shanks, US Pat., 0599211, 2008.
3. F. Thomas, G. David and M. Vicens, Eur. Coat. J., 2006, 10, 30.
4. P. D. D. Rodrigo, S. W. Y. Wong, Y. B. Cheng, K. W. Thomson, R. A. Shanks and V. Pasanovic, US Pat., 0029103A1, 2012.
5. L. G. Hanu, G. P. Simon, J. Mansouri and R. P. Burford, J. Mater. Process. Technol., 2004, 153, 401.
6. J. Mansouri, R. P. Burford, Y. B. Cheng and L. Hanu, J. Mater. Sci., 2005, 40, 5741.
7. L. G. Hanu, G. P. Simon and Y. B. Cheng, Mater. Sci. Eng., A, 2005, 398, 180.
8. J. Mansouri, R. P. Burford and Y. B. Cheng, Mater. Sci. Eng., A, 2006, 425, 7.
9. J. Mansouri, A. W. Chris, K. Roberts and Y. B. Cheng, J. Mater. Sci., 2007, 42, 6046.
10. R. Anyszka, D. M. Bielinski, Z. Pedzich and M. Szumera, J. Therm. Anal. Calorim., 2015, 119, 111.
11. J. H. Wang, C. T. Ji, Y. T. Yan, D. Zhao and L. Y. Shi, Polym. Degrad. Stab., 2015, 121, 149.
12. M. Imiela, R. Anyszka, D. M. Bielinski and Z. Pedzch, J. Therm. Anal. Calorim., 2016, 124, 197.
13. S. Hu, F. Chen, J. G. Li and Q. Shen, Polym. Degrad. Stab., 2016, 126, 196.
14. P. Rodrigo and Y. B. Cheng, WO Pat., 139011A1, 2010.
15. Z. Al-Hassany, A. Genovese and R. A. Shanks, eXPRESS Polym. Lett., 2010, 4, 79.
16. H. W. Di, C. Deng, R. M. Li, L. P. Dong and Y. Z. Wang, RSC Adv., 2015, 5, 51248.
17. D. K. Wang, H. He and P. Yu, J. Appl. Polym. Sci., 2016, 133, 1.
18. J. C. Liu, S. Peng and Y. B. Zhang, Polym. Degrad. Stab., 2015, 121, 208.
19. M. H. Li, H. J. Cui, Q. Y. Li and Q. Zhang, J. Reinf. Plast. Compos., 2016, 35, 435.
20. G. Alexander, Y. B. Cheng, and R. P. Burford, WO Pat., 088676A1, 2010.
21. G. Alexander, Y. B. Cheng and R. P. Burford, WO Pat., 095545A1, 2005.
22. G. Alexander and I. Ivanov, AU Pat., 200028A1, 2012.
23. H. K. Young, J. G. Yeo and S. C. Choi, Ceram. Int., 2016, 42, 8878.
24. B. Li, Q. Long and D. Duan, J. Mater. Sci.: Mater. Electron., 2016, 27, 2206.
25. B. H. Mussler and M. W. Shafer, Am. Ceram. Soc. Bull., 1985, 64, 1459.
26. C. C. Sascha, S. Marcel, H. Wolfram and R. Volker, J. Non-Cryst. Solids, 2000, 263, 388.
27. G. Wen, X. Zheng and L. Song, Acta Mater., 2007, 55, 3583.
28. S. Agathopoulos, D. U. Tulyaganov and J. M. Ventura, J. Non-Cryst. Solids, 2005, 352, 322.
29. H. Chen, K. Liang, L. Li, R. Duan and Y. Zuo, J. Tsinghua Univ., Sci. Technol., 1999, 39, 15.
30. X. J. Xu and D. E. Day, Phys. Chem. Glasses, 1990, 31, 183.
31. A. Lu, J. Huang and R. Lu, Chin. J. of Nonferrous Met., 1997, 7, 88.
32. H. Chen, K. Liang, S. Gu and Y. Zuo, Ceramic Studies Journal, 1998, 13, 3.

---