- Email: [email protected]
Curing properties of furfuryl alcohol condensate with carbonaceous fine particles under ultrasonication
Ultrasonics Sonochemistry 8 (2001) 89±92
www.elsevier.nl/locate/ultsonch
Curing properties of furfuryl alcohol condensate with carbonaceous ®ne particles under ultrasonication Kazuhito Hoshi *, Takashi Akatsu, Yasuhiro Tanabe, Eiichi Yasuda Materials and Structures Laboratory, Tokyo Institute of Technology, 4259, Nagatsuta, Midori-ku, Yokohama 226-8503, Japan Received 20 October 1999; received in revised form 20 January 2000; accepted 27 January 2000
Abstract Ultrasonic treatment (sonication) was carried out through the curing process of furan resin by using an ultrasonic homogenizer at the frequency of 20 kHz and the various intensities (0±90 W). Various carbonaceous ®ne particles were added to furan resin to investigate the change of polymerization degree. The curing rate of furan resin was accelerated by sonication, which increased the polymerization degree with an increase in ultrasound intensity. The increase of curing rate was also observed by small additions of carbonaceous ®ne particles. In this case, the curing rate was increased with an increase in the speci®c surface area on additives. Ó 2001 Elsevier Science B.V. All rights reserved. Keywords: Furan resin; Curing process; Carbonaceous ®ne particles; Ultrasound
1. Introduction As carbon has attractive characteristics such as light weight, chemical stability, etc. carbon materials have been used as a refractory, electrode, moderator for nuclear reactors, rocket nozzle and in various applications [1,2]. Among various carbon materials, carbon ®ber reinforced carbon composites (C/C composites), which will be one of the most important materials in the next century, maintain their excellent mechanical properties at above 2000°C [3]. The C/C composites are generally prepared by impregnating a preform of carbon ®bers with a matrix precursor, e.g., a thermoplastic or thermosetting resin. Properties and appearances of the C/C composites can be controlled in a wide range depending on the starting materials or the heat treatment condition. For example, when a thermosetting resin, such as furfuryl alcohol condensate (furan resin), is used as a matrix [4±6], the direct curing process is usually initiated with presence of acid catalyst such as p-toluenesulfonic acid and progressed by polycondensation reaction. In the ®nal stage of this reaction, the resin is cured by cross-linkage [7]. The cured resin gets converted into carbon via the solid state [8]. However, its curing and *
Corresponding author. Tel.: +81-45-924-5346; fax: +81-45-9245345. E-mail address: [email protected] (K. Hoshi).
carbonization processes take a long period to prevent the formation of cracks and to remove the included pores. Increase in the reaction temperature and the amount of curing catalyst to furan resin does not usually result in homogeneous and crack-free carbon materials. To solve these problems, the authors applied high power ultrasound for the curing process of furan resin. The purpose of the present work is to report the eect of ultrasonic treatment (sonication) on the curing process of furan resin with various carbonaceous ®ne particles having dierent surface properties, and to show the potential method for the homogeneous preparation of cured resin with additives. 2. Experimental Hitafuran 302 (Hitachi Chemical Co., Ltd.) was used as the starting material. Carbonaceous ®ne particles (heat treatment temperature: 1000°C, particles A), and the oxidized ones (particles B) were added to furan resin to compare with the change of polymerization degree on neat resin. The oxidation treatment of the particles was carried out by 80°C HNO3 for 24 h. A part of the oxidized particles was heat treated at 500°C for 1 h to eliminate surface acidic groups (particles C). The number of surface acidic groups (±OH, ±COOH) introduced on the particles was determined by acidimetry. The
1350-4177/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 5 0 - 4 1 7 7 ( 0 0 ) 0 0 0 2 8 - 6
90
K. Hoshi et al. / Ultrasonics Sonochemistry 8 (2001) 89±92
Fig. 2. Change of polymerization degree with and without sonication.
Fig. 1. Schematic illustration of the apparatus for sonication curing process.
speci®c surface area of each particle was determined with nitrogen at liquid nitrogen temperature and the particle size was measured by a particle size distribution analyzer (COULTER, LS230). The amount of additives was ®xed at a 1 mass% of furan resin. The curing catalyst (p-toluenesulfonic acid) was added to the furan resin in the proportion of 0.3 mass%. The catalyst was always added after the addition of particles. After stirred at room temperature for 30 min, the temperature of the furan resin was controlled at 50°C. In this condition, a high power ultrasound was applied through the curing process by using an ultrasonic homogenizer (Branson, model 450 Soni®er, Fig. 1) at frequency of 20 kHz and various intensities (0±90 W). The value of ultrasonic intensity was determined by using a power output chart of the ultrasonic homogenizer. The viscosity change of furan resin in the curing process was determined at a constant temperature (50°C) by a viscometer (Brook®eld, DV-II+). The rotation speed of the viscometer was 100 rpm. Polymer molecular weights were measured by a gel permeation chromatograph (GPC: JASCO, Gulliver series, column; Shodex GPC K802+K803+K804+K805, eluent; chloroform), with polystyrene as a standard material. The polymerization degree of furan resin was estimated based on the linear relationship between the logarithms of the viscosity and the average molecular weight [9]. After the sonication, these samples were hardened at 50°C for 48 h. The hardened samples were post-cured at 160°C for 6 h.
sonication. The carbonaceous ®ne particles were not added in this experiment. The rate constants of curing reaction with and without sonication were 1.4 ´ 10 1 and 2.2 ´ 10 2 s 1 , respectively [9]. When the sonication was stopped after 10 min, the curing rate decreased to the same as that of the non-sonication. Fig. 3 shows the change in polymerization degree at various ultrasound intensities. Under our experimental condition, up to 90 W, the curing rate of furan resin increased with an increase in ultrasound intensity. As for the 90 W run, the solid line drawn above t 10 min had higher slopes than that below t 10 min. This was attributable to the gelation of furan resin. The increase of curing rate was considered to result from cavitation [10]. The cavitation process is attributed to the behavior of small bubbles. The local pressures and temperatures generated by collapse of these bubbles are enormous. Consequently, the rate acceleration resulted directly from this phenomenon. At this point, it should be stressed that the curing reaction proceeded little in the absence of cavitation and that stirring (without sonication) fails to produce such a marked increase in the rate of reaction [11±13]. This acceleration
3. Results and discussion Fig. 2 shows the degree of polymerization (number of furan ring) plotted against curing time with and without
Fig. 3. Change of polymerization degree at various ultrasound intensities.
K. Hoshi et al. / Ultrasonics Sonochemistry 8 (2001) 89±92 Table 1 Characteristics of various carbonaceous ®ne particles Additives
Surface acidic groups (mmol g 1 )
BET surface area (m2 g 1 )
Particle size (lm)
Particle A Particle B Particle C
0.04 1.34 0.04
0.7 507 734
15.5 13.5 13.5
in the curing rate is not primarily a mixing eect (the mixing means the mixing of furan resin and catalyst). Conversely, changing the ultrasound intensity was presumed to alter the volume of furan resin that can be forced to cavitate and dramatically aected the observed curing rate [14]. Table 1 summarizes the characteristics of the additives. The acidimetry measurement indicated that the non-oxidized particles (particles A) had few surface acidic groups, but oxidation treatment introduced 1.34 mmol g 1 of surface acidic groups, such as ±OH and ± COOH, onto the oxidized particles (particles B). On the other hand, particles C, which were made by annealing particles B, have the same low amount of the surface acidic groups as particles A. As for the speci®c surface area, particles C were the largest, but the mean particle size was almost unchanged in each particle. The authors have examined the eect of sonication on the curing process in furan resin with the above three types of the particles having dierent surface properties. Fig. 4 shows a change in the polymerization degree of furan resin with various carbonaceous ®ne particles. The ultrasound intensity of 50 W was used in this series of the experiment. The increase in the curing rate was also observed by addition of various carbonaceous ®ne particles. These particles were not broken by sonication. The curing rate of furan resin increased rapidly in the vicinity of the gelation point. In the non-addition and particles A addition, the gelation time was about 40±45 min, whereas those of particles B and C addition were
91
about 32 and 30 min. respectively. In this case, the curing rate was increased with an increase in the speci®c surface area of the additives. However, the curing rate for stirring (without sonication) showed no dierence among these additives. This phenomenon is closely connected with nucleation of cavitation bubbles during sonication. The generally accepted mechanism for nucleation of cavitation bubbles suggests that the gas trapped in small angle pores of solid additives expands and contracts during acoustic cycles [15]. Such gases are expected to act as nucleation sites. Accordingly, the increase in the speci®c surface area in the additives was considered to derive the increase of cavitation bubbles.
4. Conclusions The results are summarized as follows: In the present work, a high power ultrasound was applied to the curing process of furan resin. The curing rate of furan resin increased with an increase in ultrasound intensity. The increase in the curing rate was also observed by the addition of carbonaceous ®ne particles, and its curing rate was increased with an increase in the speci®c surface area of additives.
Acknowledgements This research was partially supported by the Ministry of Education, Science, Sports and Culture, Grant-in-Aid for Scienti®c Research. The authors wish to thank Prof. Ikeda and Ms. Ogiri of the Research Laboratory of Resource Utilization for their help in the GPC measurements. The authors would also like to thank Dr. Enomoto of the Materials and Structures Laboratory for their help in the sonication. Finally, the authors are grateful to Hitachi Chemical Co. and UNITIKA Co. for providing furan resin and phenol resin beads, respectively. References
Fig. 4. Change of polymerization degree of furan resin with various particles (ultrasonic intensity: 50 W).
[1] E. Fitzer, Carbon 25 (1987) 163. [2] M. Inagaki, Tanso Zairyou Kougaku, The Nikkan Kogyo Shimbun, 1985, p. 174. [3] J.D. Buckley, D.D. Edie, Carbon±Carbon Materials and Composites, Noyes Publications, 1993, p. 12. [4] E. Yasuda, Y. Tanabe, L.M. Manocha, S. Kimura, Carbon 26 (1988) 333. [5] Y. Hishiyama, M. Inagaki, S. Kimura, S. Yamada, Carbon 12 (1974) 249. [6] L.M. Manocha, O.P. Bahl, Y.K. Singh, Carbon 27 (1989) 381. [7] G. Savage, Carbon±Carbon Composites, Chapman & Hall, London, 1993, p. 130. [8] K. Kobayashi, S. Sugawara, S. Toyoda, N. Honda, Carbon 6 (1968) 359.
92
K. Hoshi et al. / Ultrasonics Sonochemistry 8 (2001) 89±92
[9] K. Hoshi, Y. Tanabe, E. Yasuda, TANSO 184 (1998) 199. [10] G.J. Price, Ultrasonics Sonochemistry 3 (1996) 229. [11] S. Moon, L. Duchin, J.V. Cooney, Tetrahedron Letters 19 (1979) 3917. [12] B.H. Han, P. Boudjouk, Tetrahedron Letters 23 (1982) 1643.
[13] E. Lukevics, V.N. Gevorgyan, Y.S. Goldberg, Tetrahedron Letters 25 (1984) 1415. [14] Sonochemistry group, J.L. Luche, Ultrasonics 25 (1987) 40. [15] S. Ley, C. Low, Ultrasound in Synthesis, Springer, Berlin, 1989, p. 11.
www.elsevier.nl/locate/ultsonch
Curing properties of furfuryl alcohol condensate with carbonaceous ®ne particles under ultrasonication Kazuhito Hoshi *, Takashi Akatsu, Yasuhiro Tanabe, Eiichi Yasuda Materials and Structures Laboratory, Tokyo Institute of Technology, 4259, Nagatsuta, Midori-ku, Yokohama 226-8503, Japan Received 20 October 1999; received in revised form 20 January 2000; accepted 27 January 2000
Abstract Ultrasonic treatment (sonication) was carried out through the curing process of furan resin by using an ultrasonic homogenizer at the frequency of 20 kHz and the various intensities (0±90 W). Various carbonaceous ®ne particles were added to furan resin to investigate the change of polymerization degree. The curing rate of furan resin was accelerated by sonication, which increased the polymerization degree with an increase in ultrasound intensity. The increase of curing rate was also observed by small additions of carbonaceous ®ne particles. In this case, the curing rate was increased with an increase in the speci®c surface area on additives. Ó 2001 Elsevier Science B.V. All rights reserved. Keywords: Furan resin; Curing process; Carbonaceous ®ne particles; Ultrasound
1. Introduction As carbon has attractive characteristics such as light weight, chemical stability, etc. carbon materials have been used as a refractory, electrode, moderator for nuclear reactors, rocket nozzle and in various applications [1,2]. Among various carbon materials, carbon ®ber reinforced carbon composites (C/C composites), which will be one of the most important materials in the next century, maintain their excellent mechanical properties at above 2000°C [3]. The C/C composites are generally prepared by impregnating a preform of carbon ®bers with a matrix precursor, e.g., a thermoplastic or thermosetting resin. Properties and appearances of the C/C composites can be controlled in a wide range depending on the starting materials or the heat treatment condition. For example, when a thermosetting resin, such as furfuryl alcohol condensate (furan resin), is used as a matrix [4±6], the direct curing process is usually initiated with presence of acid catalyst such as p-toluenesulfonic acid and progressed by polycondensation reaction. In the ®nal stage of this reaction, the resin is cured by cross-linkage [7]. The cured resin gets converted into carbon via the solid state [8]. However, its curing and *
Corresponding author. Tel.: +81-45-924-5346; fax: +81-45-9245345. E-mail address: [email protected] (K. Hoshi).
carbonization processes take a long period to prevent the formation of cracks and to remove the included pores. Increase in the reaction temperature and the amount of curing catalyst to furan resin does not usually result in homogeneous and crack-free carbon materials. To solve these problems, the authors applied high power ultrasound for the curing process of furan resin. The purpose of the present work is to report the eect of ultrasonic treatment (sonication) on the curing process of furan resin with various carbonaceous ®ne particles having dierent surface properties, and to show the potential method for the homogeneous preparation of cured resin with additives. 2. Experimental Hitafuran 302 (Hitachi Chemical Co., Ltd.) was used as the starting material. Carbonaceous ®ne particles (heat treatment temperature: 1000°C, particles A), and the oxidized ones (particles B) were added to furan resin to compare with the change of polymerization degree on neat resin. The oxidation treatment of the particles was carried out by 80°C HNO3 for 24 h. A part of the oxidized particles was heat treated at 500°C for 1 h to eliminate surface acidic groups (particles C). The number of surface acidic groups (±OH, ±COOH) introduced on the particles was determined by acidimetry. The
1350-4177/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 5 0 - 4 1 7 7 ( 0 0 ) 0 0 0 2 8 - 6
90
K. Hoshi et al. / Ultrasonics Sonochemistry 8 (2001) 89±92
Fig. 2. Change of polymerization degree with and without sonication.
Fig. 1. Schematic illustration of the apparatus for sonication curing process.
speci®c surface area of each particle was determined with nitrogen at liquid nitrogen temperature and the particle size was measured by a particle size distribution analyzer (COULTER, LS230). The amount of additives was ®xed at a 1 mass% of furan resin. The curing catalyst (p-toluenesulfonic acid) was added to the furan resin in the proportion of 0.3 mass%. The catalyst was always added after the addition of particles. After stirred at room temperature for 30 min, the temperature of the furan resin was controlled at 50°C. In this condition, a high power ultrasound was applied through the curing process by using an ultrasonic homogenizer (Branson, model 450 Soni®er, Fig. 1) at frequency of 20 kHz and various intensities (0±90 W). The value of ultrasonic intensity was determined by using a power output chart of the ultrasonic homogenizer. The viscosity change of furan resin in the curing process was determined at a constant temperature (50°C) by a viscometer (Brook®eld, DV-II+). The rotation speed of the viscometer was 100 rpm. Polymer molecular weights were measured by a gel permeation chromatograph (GPC: JASCO, Gulliver series, column; Shodex GPC K802+K803+K804+K805, eluent; chloroform), with polystyrene as a standard material. The polymerization degree of furan resin was estimated based on the linear relationship between the logarithms of the viscosity and the average molecular weight [9]. After the sonication, these samples were hardened at 50°C for 48 h. The hardened samples were post-cured at 160°C for 6 h.
sonication. The carbonaceous ®ne particles were not added in this experiment. The rate constants of curing reaction with and without sonication were 1.4 ´ 10 1 and 2.2 ´ 10 2 s 1 , respectively [9]. When the sonication was stopped after 10 min, the curing rate decreased to the same as that of the non-sonication. Fig. 3 shows the change in polymerization degree at various ultrasound intensities. Under our experimental condition, up to 90 W, the curing rate of furan resin increased with an increase in ultrasound intensity. As for the 90 W run, the solid line drawn above t 10 min had higher slopes than that below t 10 min. This was attributable to the gelation of furan resin. The increase of curing rate was considered to result from cavitation [10]. The cavitation process is attributed to the behavior of small bubbles. The local pressures and temperatures generated by collapse of these bubbles are enormous. Consequently, the rate acceleration resulted directly from this phenomenon. At this point, it should be stressed that the curing reaction proceeded little in the absence of cavitation and that stirring (without sonication) fails to produce such a marked increase in the rate of reaction [11±13]. This acceleration
3. Results and discussion Fig. 2 shows the degree of polymerization (number of furan ring) plotted against curing time with and without
Fig. 3. Change of polymerization degree at various ultrasound intensities.
K. Hoshi et al. / Ultrasonics Sonochemistry 8 (2001) 89±92 Table 1 Characteristics of various carbonaceous ®ne particles Additives
Surface acidic groups (mmol g 1 )
BET surface area (m2 g 1 )
Particle size (lm)
Particle A Particle B Particle C
0.04 1.34 0.04
0.7 507 734
15.5 13.5 13.5
in the curing rate is not primarily a mixing eect (the mixing means the mixing of furan resin and catalyst). Conversely, changing the ultrasound intensity was presumed to alter the volume of furan resin that can be forced to cavitate and dramatically aected the observed curing rate [14]. Table 1 summarizes the characteristics of the additives. The acidimetry measurement indicated that the non-oxidized particles (particles A) had few surface acidic groups, but oxidation treatment introduced 1.34 mmol g 1 of surface acidic groups, such as ±OH and ± COOH, onto the oxidized particles (particles B). On the other hand, particles C, which were made by annealing particles B, have the same low amount of the surface acidic groups as particles A. As for the speci®c surface area, particles C were the largest, but the mean particle size was almost unchanged in each particle. The authors have examined the eect of sonication on the curing process in furan resin with the above three types of the particles having dierent surface properties. Fig. 4 shows a change in the polymerization degree of furan resin with various carbonaceous ®ne particles. The ultrasound intensity of 50 W was used in this series of the experiment. The increase in the curing rate was also observed by addition of various carbonaceous ®ne particles. These particles were not broken by sonication. The curing rate of furan resin increased rapidly in the vicinity of the gelation point. In the non-addition and particles A addition, the gelation time was about 40±45 min, whereas those of particles B and C addition were
91
about 32 and 30 min. respectively. In this case, the curing rate was increased with an increase in the speci®c surface area of the additives. However, the curing rate for stirring (without sonication) showed no dierence among these additives. This phenomenon is closely connected with nucleation of cavitation bubbles during sonication. The generally accepted mechanism for nucleation of cavitation bubbles suggests that the gas trapped in small angle pores of solid additives expands and contracts during acoustic cycles [15]. Such gases are expected to act as nucleation sites. Accordingly, the increase in the speci®c surface area in the additives was considered to derive the increase of cavitation bubbles.
4. Conclusions The results are summarized as follows: In the present work, a high power ultrasound was applied to the curing process of furan resin. The curing rate of furan resin increased with an increase in ultrasound intensity. The increase in the curing rate was also observed by the addition of carbonaceous ®ne particles, and its curing rate was increased with an increase in the speci®c surface area of additives.
Acknowledgements This research was partially supported by the Ministry of Education, Science, Sports and Culture, Grant-in-Aid for Scienti®c Research. The authors wish to thank Prof. Ikeda and Ms. Ogiri of the Research Laboratory of Resource Utilization for their help in the GPC measurements. The authors would also like to thank Dr. Enomoto of the Materials and Structures Laboratory for their help in the sonication. Finally, the authors are grateful to Hitachi Chemical Co. and UNITIKA Co. for providing furan resin and phenol resin beads, respectively. References
Fig. 4. Change of polymerization degree of furan resin with various particles (ultrasonic intensity: 50 W).
[1] E. Fitzer, Carbon 25 (1987) 163. [2] M. Inagaki, Tanso Zairyou Kougaku, The Nikkan Kogyo Shimbun, 1985, p. 174. [3] J.D. Buckley, D.D. Edie, Carbon±Carbon Materials and Composites, Noyes Publications, 1993, p. 12. [4] E. Yasuda, Y. Tanabe, L.M. Manocha, S. Kimura, Carbon 26 (1988) 333. [5] Y. Hishiyama, M. Inagaki, S. Kimura, S. Yamada, Carbon 12 (1974) 249. [6] L.M. Manocha, O.P. Bahl, Y.K. Singh, Carbon 27 (1989) 381. [7] G. Savage, Carbon±Carbon Composites, Chapman & Hall, London, 1993, p. 130. [8] K. Kobayashi, S. Sugawara, S. Toyoda, N. Honda, Carbon 6 (1968) 359.
92
K. Hoshi et al. / Ultrasonics Sonochemistry 8 (2001) 89±92
[9] K. Hoshi, Y. Tanabe, E. Yasuda, TANSO 184 (1998) 199. [10] G.J. Price, Ultrasonics Sonochemistry 3 (1996) 229. [11] S. Moon, L. Duchin, J.V. Cooney, Tetrahedron Letters 19 (1979) 3917. [12] B.H. Han, P. Boudjouk, Tetrahedron Letters 23 (1982) 1643.
[13] E. Lukevics, V.N. Gevorgyan, Y.S. Goldberg, Tetrahedron Letters 25 (1984) 1415. [14] Sonochemistry group, J.L. Luche, Ultrasonics 25 (1987) 40. [15] S. Ley, C. Low, Ultrasound in Synthesis, Springer, Berlin, 1989, p. 11.