- Email: [email protected]
Influence of TiO2 and Fe2O3 fillers on the thermal properties of poly(methyl methacrylate) (PMMA)
Materials Letters 59 (2005) 36 – 39 www.elsevier.com/locate/matlet
Influence of TiO2 and Fe2O3 fillers on the thermal properties of poly(methyl methacrylate) (PMMA) A. Laachachia, M. Cocheza, M. Ferriola,*, J.M. Lopez-Cuestab, E. Leroyb a
Laboratoire de Chimie et Applications, E.A. 3471, Universite´ de Metz, rue Victor Demange-57500 Saint-Avold, France Centre des Mate´riaux de Grande Diffusion, Ecole des Mines d’Ale`s, 6 avenue de Clavie`res, 30319-Ale`s Ce´dex, France
b
Received 24 March 2004; received in revised form 20 August 2004; accepted 13 September 2004 Available online 28 September 2004
Abstract The thermal properties of pure poly(methyl methacrylate) (PMMA) and PMMA filled with 5%, 10%, 15% and 20% of nanometric particles of titanium oxide (TiO2) and ferric oxide (Fe2O3) were investigated under air atmosphere by DSC, TGA and LOI measurements on samples prepared by solvent casting method. In the presence of the filler, the thermal stability of the polymer appeared to be significantly improved. A linear relationship between LOI and glass transition temperature (T g) suggests that the restriction of mobility of the polymer chains is directly linked to the increase of stability. D 2004 Elsevier B.V. All rights reserved. Keywords: Polymers; Nanomaterials; Thermal properties
1. Introduction
2. Experiments and results
The improvement of thermal and flammability properties of polymer materials is a major concern, particularly in the domains of transportation, building, civil and electrical engineering. Numerous and recent studies have shown the interest of the use of lamellar nanofillers (above all modified montmorillonites) as flame retardants [1,2] or as components of flame retardants systems [2]. Moreover, the use of oxide particles in the submicronic or nanometric range as synergistic agents in addition to usual fire retardant additives seems to be very promising. Since a few years ago, we are involved in the search of strongly fire retarded poly(methyl methacrylate) (PMMA), and, at present, we have undertaken researches with such an approach. In a first step, we are investigating the influence of oxide fillers alone on the thermal properties of PMMA before to incorporate them as synergistic agents of phosphorous flame retardants. This paper presents the first results obtained with Fe2O3 and TiO2 fillers.
2.1. Sample preparation
* Corresponding author. Tel.: +33 387939185; fax: +33 387939101. E-mail address: [email protected] (M. Ferriol). 0167-577X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2004.09.014
The poly(methyl methacrylate) was from Aldrich with a weight-average molecular weight of 350,000 g mol 1 (based on GPC analysis) and obtained by free radical synthesis. Commercial TiO2 from Degussa (P-25 99.5%) with a mean particle size of 21 nm was used. The surface area was checked by the BET method and found equal to 48 m2 g 1. Submicron Fe2O3 particles were prepared according to the procedure described by Deb et al. [3]. The method is based on the precipitation of ferric oxide hydroxide (FeOOH) from a ferric nitrate solution, followed by heat treatment to completely decompose the organic layer and FeOOH into Fe2O3. X-ray diffraction analysis after heat treatment at 400 8C showed the co-existence of a- and g-Fe2O3 (Fig. 1) even if the samples were treated during a few hours. However, heat treatment at 500 8C during 2 h resulted in only the a phase (Fig. 1). The surface area was measured by BET and was equal to 52 m2 g 1. Under the scanning electron microscope, the iron oxide powder appeared in the form of
A. Laachachi et al. / Materials Letters 59 (2005) 36–39
37
Fig. 1. X-ray powder diffraction patterns of the synthesized Fe2O3 after heat treatment at 400 and 500 8C.
agglomerates (Fig. 2), whereas transmission electron microscopy showed that these agglomerates were formed from plate-like particles (estimated minimum size: (100– 200)(7–10)nm). The PMMA/oxide-filled polymer samples were prepared by dissolving PMMA into chloroform (typically 1 g for 50 cm3). The solution was stirred at 50 8C during 30 min. Then, the appropriate amount of TiO2 or Fe2O3 was added and ultrasonically dispersed during a few hours at room temperature in order to obtain a good dispersion of the oxide particles into the polymer matrix. Afterwards, the samples were dried during 4 h to complete the evaporation of chloroform. Treatment at about 70–80 8C as commonly performed in the literature was found to be quite noneffective as shown by TGA experiments where an unex-
Fig. 2. Typical SEM micrographs of the Fe2O3 powder after heat treatment at 500 8C.
pected weight loss was recorded at about 170 8C which was found to correspond to chloroform volatilization by Py-GCMS analysis. The temperature of 170 8C was then checked to provide samples free of chloroform. 2.2. Experimental techniques Glass transition temperatures (T g) were measured within 0.5 8C with a Setaram DSC-92 apparatus with an heating rate of 10 8C min 1. The samples (20–25 mg) were placed into aluminum crucibles. Thermogravimetric studies were performed with a Mettler-Toledo TGA/SDTA 851e apparatus operating in air atmosphere. The runs were carried out in dynamic conditions at 10 8C min 1 between 30 and 600 8C. Alumina crucibles were used and the sample weight was about 20–25 mg. The limiting oxygen index (LOI), defined as the minimum oxygen amount, in an oxygen–nitrogen flow, required to completely burn down the material during at least 30 s [4] was determined with an apparatus described in NFT 51-07 using a modified procedure [5]: Pellets of the sample (200 mg) were placed on an aluminum sample holder (height: 25 mm, length: 55 mm, width: 18 mm) located in the middle of the chimney (diameter: 80 mm, height: 540 mm). The values never represented the absolute LOI (ASTM D 2863-77) of the samples but were found to be quite reproducible compared to those obtained for six usual polymers [6]. The oxygen amount was measured with a thermal flow meter Calibrage DGM (mass flow meter Brooks 5850 TR at room temperature, inpressure: 2 bar and out-pressure: atmosphere). The measurements were performed with an oxygen amount step equal to 0.5.
38
A. Laachachi et al. / Materials Letters 59 (2005) 36–39
2.3. Thermogravimetric analysis Fig. 3 gives the TG curves obtained under air for PMMA/Fe2O3-filled polymer and PMMA/TiO2-filled polymer for various amounts of oxides compared to pure PMMA. The thermal stability of the polymer is largely improved (by about 70 8C) even for the lowest oxide content. Fig. 4 gives the DTG curves for pure PMMA and for the highest content in both oxides. It can be seen that PMMA apparently degrades according to three different weight losses. When alone, the first two are the major ones whereas when filled with oxides, these two stages are inhibited and PMMA degrades mainly by the third one which explains the observed shift for the degradation temperature. 2.4. T g and LOI measurements Fig. 5 shows the evolution of the LOI as a function of the glass transition temperature T g for the PMMA/oxide-filled polymer for different amounts of Fe2O3 and TiO2. Both T g and LOI increase as the oxide content increases. T g increases because of the restriction of mobility of the polymer chains.
Fig. 4. DTG curves of pure and filled PMMA polymer under air.
The increase of LOI suggests an enhancement of the fire retardant properties of the filled polymer to be confirmed by flammability tests as cone calorimetry which are now in progress. For the two oxides, an approximately linear relationship between LOI and T g can be observed as found earlier for PMMA filled with Sb2O3 [7].
Fig. 3. TG curves of pure and filled PMMA polymer under air.
Fig. 5. Relation between LOI and T g for pure, Fe2O3-filled and TiO2-filled polymer under air.
A. Laachachi et al. / Materials Letters 59 (2005) 36–39
2.5. Discussion In our previous paper devoted to the investigation of PMMA/Sb2O3-filled polymer [7], the reasons why the thermal stability increases in the presence of the filler was discussed on the basis of the restriction of mobility of the polymer chains resulting from: (i) the bonds due to the adsorption of polymer on the oxide surface via the methoxycarbonyl group, and (ii) the steric hindrance due to the presence of solid particles of the additive [8]. For TiO2 and Fe2O3 fillers, as the same behaviour is observed, it can be assumed that the restriction of mobility by oxide particles is also responsible for the increase of thermal stability, probably by inhibiting a key step of the reaction mechanism involved, that key step occurring during the two first weight losses. However, the mechanism of the oxidative degradation of PMMA is not well known as shown by the literature results [9–12] and the link with our observations requires further investigations. The linear relationship between T g and LOI confirms the role of the mobility restriction on the increase of thermal stability.
3. Conclusion The thermal properties of PMMA filled with titanium oxide (TiO2) and ferric oxide (Fe2O3) were investigated
39
under air atmosphere by DSC, TGA and LOI measurements and compared to those of PMMA alone. From the results obtained, an improvement of the thermal stability was observed increasing with the amount of oxide. A direct influence of the restriction of mobility of the polymer chains on the thermal oxidative degradation mechanism is suggested to explain the observed phenomena.
References [1] J.W. Gilman, Appl. Clay Sci. 15 (1999) 31. [2] M. Le Bras, C. Wilkie, S. Bourbigot (Eds.), Fire Retardancy of Polymers: New Applications of Mineral Fillers, Royal Society of Chemistry, Cambridge, ISBN 0-85404-582-1, in press. [3] P. Deb, A. Basumallick, P. Chaterjee, S.P. Sengupta, Scr. Mater. 45 (2001) 341. [4] C.P. Fenimore, F.T. Martin, Mod. Plast. 44 (1966) 141. [5] C.P. Reghunadhan Nair, G. Clouet, Y. Guilbert, Polym. Degrad. Stab. 26 (1989) 305. [6] M. Cochez, M. Ferriol, J.V. Weber, P. Chaudron, N. Oget, J.L. Mieloszynski, Polym. Degrad. Stab. 70 (2000) 455. [7] A. Laachachi, M. Cochez, M. Ferriol, E. Leroy, J.M. Lopez-Cuesta, N. Oget, Polym. Degrad. Stab. 85 (2004) 641. [8] S. Zulfigar, K. Masud, Polym. Degrad. Stab. 78 (2002) 305. [9] T. Hirata, T. Kashiwagi, J.E. Brown, Macromolecules 18 (1985) 1410. [10] T. Kashiwagi, A. Inaba, E. Brown, Macromolecules 19 (1986) 2160. [11] J.D. Peterson, S. Vyazovkin, C.A. Wight, Macromol. Rapid Comm. 20 (1999) 480. [12] J.D. Peterson, S. Vyazovkin, C.A. Wight, J. Phys. Chem., B 103 (1999) 8087.
Influence of TiO2 and Fe2O3 fillers on the thermal properties of poly(methyl methacrylate) (PMMA) A. Laachachia, M. Cocheza, M. Ferriola,*, J.M. Lopez-Cuestab, E. Leroyb a
Laboratoire de Chimie et Applications, E.A. 3471, Universite´ de Metz, rue Victor Demange-57500 Saint-Avold, France Centre des Mate´riaux de Grande Diffusion, Ecole des Mines d’Ale`s, 6 avenue de Clavie`res, 30319-Ale`s Ce´dex, France
b
Received 24 March 2004; received in revised form 20 August 2004; accepted 13 September 2004 Available online 28 September 2004
Abstract The thermal properties of pure poly(methyl methacrylate) (PMMA) and PMMA filled with 5%, 10%, 15% and 20% of nanometric particles of titanium oxide (TiO2) and ferric oxide (Fe2O3) were investigated under air atmosphere by DSC, TGA and LOI measurements on samples prepared by solvent casting method. In the presence of the filler, the thermal stability of the polymer appeared to be significantly improved. A linear relationship between LOI and glass transition temperature (T g) suggests that the restriction of mobility of the polymer chains is directly linked to the increase of stability. D 2004 Elsevier B.V. All rights reserved. Keywords: Polymers; Nanomaterials; Thermal properties
1. Introduction
2. Experiments and results
The improvement of thermal and flammability properties of polymer materials is a major concern, particularly in the domains of transportation, building, civil and electrical engineering. Numerous and recent studies have shown the interest of the use of lamellar nanofillers (above all modified montmorillonites) as flame retardants [1,2] or as components of flame retardants systems [2]. Moreover, the use of oxide particles in the submicronic or nanometric range as synergistic agents in addition to usual fire retardant additives seems to be very promising. Since a few years ago, we are involved in the search of strongly fire retarded poly(methyl methacrylate) (PMMA), and, at present, we have undertaken researches with such an approach. In a first step, we are investigating the influence of oxide fillers alone on the thermal properties of PMMA before to incorporate them as synergistic agents of phosphorous flame retardants. This paper presents the first results obtained with Fe2O3 and TiO2 fillers.
2.1. Sample preparation
* Corresponding author. Tel.: +33 387939185; fax: +33 387939101. E-mail address: [email protected] (M. Ferriol). 0167-577X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2004.09.014
The poly(methyl methacrylate) was from Aldrich with a weight-average molecular weight of 350,000 g mol 1 (based on GPC analysis) and obtained by free radical synthesis. Commercial TiO2 from Degussa (P-25 99.5%) with a mean particle size of 21 nm was used. The surface area was checked by the BET method and found equal to 48 m2 g 1. Submicron Fe2O3 particles were prepared according to the procedure described by Deb et al. [3]. The method is based on the precipitation of ferric oxide hydroxide (FeOOH) from a ferric nitrate solution, followed by heat treatment to completely decompose the organic layer and FeOOH into Fe2O3. X-ray diffraction analysis after heat treatment at 400 8C showed the co-existence of a- and g-Fe2O3 (Fig. 1) even if the samples were treated during a few hours. However, heat treatment at 500 8C during 2 h resulted in only the a phase (Fig. 1). The surface area was measured by BET and was equal to 52 m2 g 1. Under the scanning electron microscope, the iron oxide powder appeared in the form of
A. Laachachi et al. / Materials Letters 59 (2005) 36–39
37
Fig. 1. X-ray powder diffraction patterns of the synthesized Fe2O3 after heat treatment at 400 and 500 8C.
agglomerates (Fig. 2), whereas transmission electron microscopy showed that these agglomerates were formed from plate-like particles (estimated minimum size: (100– 200)(7–10)nm). The PMMA/oxide-filled polymer samples were prepared by dissolving PMMA into chloroform (typically 1 g for 50 cm3). The solution was stirred at 50 8C during 30 min. Then, the appropriate amount of TiO2 or Fe2O3 was added and ultrasonically dispersed during a few hours at room temperature in order to obtain a good dispersion of the oxide particles into the polymer matrix. Afterwards, the samples were dried during 4 h to complete the evaporation of chloroform. Treatment at about 70–80 8C as commonly performed in the literature was found to be quite noneffective as shown by TGA experiments where an unex-
Fig. 2. Typical SEM micrographs of the Fe2O3 powder after heat treatment at 500 8C.
pected weight loss was recorded at about 170 8C which was found to correspond to chloroform volatilization by Py-GCMS analysis. The temperature of 170 8C was then checked to provide samples free of chloroform. 2.2. Experimental techniques Glass transition temperatures (T g) were measured within 0.5 8C with a Setaram DSC-92 apparatus with an heating rate of 10 8C min 1. The samples (20–25 mg) were placed into aluminum crucibles. Thermogravimetric studies were performed with a Mettler-Toledo TGA/SDTA 851e apparatus operating in air atmosphere. The runs were carried out in dynamic conditions at 10 8C min 1 between 30 and 600 8C. Alumina crucibles were used and the sample weight was about 20–25 mg. The limiting oxygen index (LOI), defined as the minimum oxygen amount, in an oxygen–nitrogen flow, required to completely burn down the material during at least 30 s [4] was determined with an apparatus described in NFT 51-07 using a modified procedure [5]: Pellets of the sample (200 mg) were placed on an aluminum sample holder (height: 25 mm, length: 55 mm, width: 18 mm) located in the middle of the chimney (diameter: 80 mm, height: 540 mm). The values never represented the absolute LOI (ASTM D 2863-77) of the samples but were found to be quite reproducible compared to those obtained for six usual polymers [6]. The oxygen amount was measured with a thermal flow meter Calibrage DGM (mass flow meter Brooks 5850 TR at room temperature, inpressure: 2 bar and out-pressure: atmosphere). The measurements were performed with an oxygen amount step equal to 0.5.
38
A. Laachachi et al. / Materials Letters 59 (2005) 36–39
2.3. Thermogravimetric analysis Fig. 3 gives the TG curves obtained under air for PMMA/Fe2O3-filled polymer and PMMA/TiO2-filled polymer for various amounts of oxides compared to pure PMMA. The thermal stability of the polymer is largely improved (by about 70 8C) even for the lowest oxide content. Fig. 4 gives the DTG curves for pure PMMA and for the highest content in both oxides. It can be seen that PMMA apparently degrades according to three different weight losses. When alone, the first two are the major ones whereas when filled with oxides, these two stages are inhibited and PMMA degrades mainly by the third one which explains the observed shift for the degradation temperature. 2.4. T g and LOI measurements Fig. 5 shows the evolution of the LOI as a function of the glass transition temperature T g for the PMMA/oxide-filled polymer for different amounts of Fe2O3 and TiO2. Both T g and LOI increase as the oxide content increases. T g increases because of the restriction of mobility of the polymer chains.
Fig. 4. DTG curves of pure and filled PMMA polymer under air.
The increase of LOI suggests an enhancement of the fire retardant properties of the filled polymer to be confirmed by flammability tests as cone calorimetry which are now in progress. For the two oxides, an approximately linear relationship between LOI and T g can be observed as found earlier for PMMA filled with Sb2O3 [7].
Fig. 3. TG curves of pure and filled PMMA polymer under air.
Fig. 5. Relation between LOI and T g for pure, Fe2O3-filled and TiO2-filled polymer under air.
A. Laachachi et al. / Materials Letters 59 (2005) 36–39
2.5. Discussion In our previous paper devoted to the investigation of PMMA/Sb2O3-filled polymer [7], the reasons why the thermal stability increases in the presence of the filler was discussed on the basis of the restriction of mobility of the polymer chains resulting from: (i) the bonds due to the adsorption of polymer on the oxide surface via the methoxycarbonyl group, and (ii) the steric hindrance due to the presence of solid particles of the additive [8]. For TiO2 and Fe2O3 fillers, as the same behaviour is observed, it can be assumed that the restriction of mobility by oxide particles is also responsible for the increase of thermal stability, probably by inhibiting a key step of the reaction mechanism involved, that key step occurring during the two first weight losses. However, the mechanism of the oxidative degradation of PMMA is not well known as shown by the literature results [9–12] and the link with our observations requires further investigations. The linear relationship between T g and LOI confirms the role of the mobility restriction on the increase of thermal stability.
3. Conclusion The thermal properties of PMMA filled with titanium oxide (TiO2) and ferric oxide (Fe2O3) were investigated
39
under air atmosphere by DSC, TGA and LOI measurements and compared to those of PMMA alone. From the results obtained, an improvement of the thermal stability was observed increasing with the amount of oxide. A direct influence of the restriction of mobility of the polymer chains on the thermal oxidative degradation mechanism is suggested to explain the observed phenomena.
References [1] J.W. Gilman, Appl. Clay Sci. 15 (1999) 31. [2] M. Le Bras, C. Wilkie, S. Bourbigot (Eds.), Fire Retardancy of Polymers: New Applications of Mineral Fillers, Royal Society of Chemistry, Cambridge, ISBN 0-85404-582-1, in press. [3] P. Deb, A. Basumallick, P. Chaterjee, S.P. Sengupta, Scr. Mater. 45 (2001) 341. [4] C.P. Fenimore, F.T. Martin, Mod. Plast. 44 (1966) 141. [5] C.P. Reghunadhan Nair, G. Clouet, Y. Guilbert, Polym. Degrad. Stab. 26 (1989) 305. [6] M. Cochez, M. Ferriol, J.V. Weber, P. Chaudron, N. Oget, J.L. Mieloszynski, Polym. Degrad. Stab. 70 (2000) 455. [7] A. Laachachi, M. Cochez, M. Ferriol, E. Leroy, J.M. Lopez-Cuesta, N. Oget, Polym. Degrad. Stab. 85 (2004) 641. [8] S. Zulfigar, K. Masud, Polym. Degrad. Stab. 78 (2002) 305. [9] T. Hirata, T. Kashiwagi, J.E. Brown, Macromolecules 18 (1985) 1410. [10] T. Kashiwagi, A. Inaba, E. Brown, Macromolecules 19 (1986) 2160. [11] J.D. Peterson, S. Vyazovkin, C.A. Wight, Macromol. Rapid Comm. 20 (1999) 480. [12] J.D. Peterson, S. Vyazovkin, C.A. Wight, J. Phys. Chem., B 103 (1999) 8087.