Pergamon Carbon Vol. 35, No. 5, 703-713,1997 pp.

Copyright 0 1997 Elsevier Science Ltd Printed in Great Britain. All rights reserved

0008-6223/97 $17.00 + 0.00

LETTERS TO THE EDITOR

Effect of reactive atmosphere and maximum heat treatment temperature on char characteristics of pyrolyzed rayon cloth

K. KUMAR, R. KOTHARI and J.N. BOHRA Material Characterization Division, National Physical Laboratory

Dr. K.S. Krishnan Road New Delhi-1 10 012 India

(Received 21 June 1996; accepted in revised form 28 February 1997)

Key words -A. activated carbon, B. pyrolysis, C. adsorption, D. mlcroporosity

Viscose rayon in fibrous form is now widely used as one of the precursors for the production of activated carbon cloth [l]. The pore structure of activated carbon cloth depends on the nature of the precursor used, the impregnants used in the pre-treatment of the precursor, flow rate of the reacting gas, maximum heat treatment temperature (MHTT) and heating rate. The chars prepared by carbonizing pretreated viscose rayon cloth are highly microporous (pore width < 2 nm) with small external surface area and very little mesoporosity [2-61. Badami etal. [7] studied the structural changes that occur during carbonization and graphitization of viscose rayon fibres and observed that heat treatment appeared to increase the microporosity. A series of charcoal cloth samples from viscose rayon cloth were prepared by Carrot and Freeman [8]. They carbonized pretreated cloth samples under nitrogen up to 873°C and then activated these samples to various extents of bum-off under carbon dioxide at the carbonization temperature. They reported that at a burn-off of about 50% the micropore volume reached a maximum. At higher levels of bum-off both pore widening and pore narrowing were found to occur. In the case of untreated rayon fibres

carbonized in nitrogen, Bohra et. al. [9] observed that the micropore volume increased with heat treatment temperature (HTT) and reached a maximum at 7OO’C. Above 700°C, the micropore volume was found to decline. Brunner and Roberts [lo] carbonized powdered cellulose and found that the pores and pore-openings grow with increasing pyrolysis temperature due to weight loss in the material.

In the present work an attempt has been made to investigate the properties of the char obtained by the pyrolysis of phosphoric acid pretreated viscose rayon cloth in an inert as well as in a reactive atmosphere.

Samples of viscose rayon cloth were treated with a 4% (v/v) solution of phosphoric acid. The treated samples were then dried in a hot air oven at 110°C and pyrolysed in a vertical furnace. Two series of samples were prepared at increasing MH’IT to a maximum temperature of 95O’C. A residence time of 30 min was given at MI-ITT to each sample. One series of samples was prepared under carbon dioxide atmosphere and a second series was prepared under nitrogen atmosphere. The flow rate of gas and the heating schedule were kept the same for both series of samples. The furnace was

Table 1. Maximum heat treatment temperature, yield, shrinkage and DR-micropore volume measured by carbon dioxide adsorption at 273 K.

Fig. 1. Variation of DR-micropore volume (CO, at 27W with maximum heat treatment temperature.

controlled with a microprocessor based temperature programmer and controller. Details of the samples prepared are summarized in Table 1. Adsorption measurements were carried out on an Indian Standard Volumetric adsorption apparatus [l l] using ultra high purity nitrogen at 77K and carbon dioxide at 273K as the adsorbates. All samples were outgassed at 200°C for 4 hours under high vacuum before the adsorption measurements. BET, a,, and DR methods have been used for the analysis of adsorption data. Standard isotherm used for the application of cxs method was obtained with nitrogen adsorption at 77K on graphitized carbon black, Vulcan 3G (2700).

The results obtained for the two series of samples prepared under inert and reactive atmospheres are discussed separately. First we shall consider samples prepared under a carbon dioxide atmosphere

Figure 1 shows a continuous increase in the DR-micropore volume (determined by CO, adsorption at 273K) up to MHTT 800°C. This increase is the result of decomposition of the precursor leading to pore creation. The material tends to shrink with heat treatment but the shrinkage eventually lags behind the process of pore creation. At MHTT 850°C the somewhat steep rise in the micropore volume is a consequence of preferential activation due to the reaction of CO, with the already carbonized material. The decrease in DR-micropore volume above MHTT 85O’C usually indicates pore closure due to sintering at high temperatures as observed by Bohra etal.[9] in the case of untreated viscose rayon yarn pyrolysed in a nitrogen atmosphere. Further investigation shows a quite different interpretation. These results are discussed in the following paragraphs.

Fig. 2. Adsorption-desorption isotherms of nitrogen at 77K for samples prepared in carbon dioxide: I(SSC), II(9OC) and III(95C).

Fig. 3. a plots for nitrogen at 77K on samples prepared in carbonsdioxide: 1(85C), II(9OC) and III(95C).

The widening of the knee in the adsorption- desorption isotherms of nitrogen at 77K with increasing MHTf (Figure 2) indicates that a gradual change is taking place in the pore structure of the samples. This distortion in the shape of the isotherms signifies a widening of the micropores as the MHTT is increased from 85O’C to 950°C. The widening of the pores above MHTT 85O’C is further supported by the systematic decrease in the extent of linearity in the CC, plots for the samples heat treated to 850, 900 and 950°C (Fig. 3). Moreover the shifting of the relative pressure range of linearity in the BET plots to higher values of relative pressure (Table 2) is indicative of pore widening. It should be emphasized here that in spite of pore widening, the materials are still microporous in nature even after heat treatment at 950°C under carbon dioxide. This is strongly supported by the fact that the isotherms are reversible and do not show any hysteresis in the higher relative pressure region. In addition, the a, plots of these samples (Figure 3) do not show any upward deviation at higher relative pressures corresponding to the capillary condensation region [ 121. Thus the possibility of the presence of mesopores in the samples investigated is ruled out.

In the case of nitrogen adsorption (77K), the calculated values of micropore volume using DR and a, methods show significant increase when the MHTT is raised from 850 to 950°C (Table 2). This result may seem contrary to the decrease in DR-micropore volume of the same samples when determined by CO, adsorption at 273K (Table 1). The reason for this discrepancy may be that the application of the DR-equation to adsorption isotherms of CO, at 273K measures the narrowest microporosity of activated carbons [ 13-151. Above 850°C, widening of the existing pores may lead to slightly lower values of DR-micropore volume as determined by CO, adsorption at 273K. The authors feel that when MHTT of the sample is increased above 850°C a fraction of the micropores is converted to super micropores. Probably when DR-micro-pore volume is calculated from CO, adsorption at 273K, the fraction of super micropores in the samples is excluded. Perhaps at higher relative pressures (nitrogen adsorption at 77K), the calculated values of micropore volume may include a contribution of super-micropores and external surface also. The increasing values of the BET surface area, Snn~, may correspond to both surface coverage and micropore filling [ 161. Overall increase in the pore volume is shown in Table 2 by the values of Gurvitsch volume VGp.

The values of percentage yield of char for

706 The following conclusions can be drawn from

the above study: (1) Pyrolysis of phosphoric acid pretreated viscose rayon cloth under a carbon dioxide atomosphere leads to widening of pores above MHTf 850°C. (2) In the case of chars prepared under a nitrogen atmosphere, pre-treatement of the precursor with phosphoric acid tends to inhibit sintering of the pores even at high temperatures.

Acknowledgments - The authors are thankful to Dr. 10. Krishan Lal. Head. Material Characterization Division and Prof. E.S.R. Gopal, Director, National Physical 11. Laboratory, New Delhi,India, for their kind permission to publish this paper. The authors are also thankful to Mr. R.K. Saxena for his keen interest in the experimental work. 12.

REFERENCES 13.

Suzuki, M., Carbon. 1994.32, 577. Bailey, A. and Maggs, F.A.P., Br. Patent No. 1310101. 1971. 14. Bailey, A., Maggs, F.A.P. and Wiliams, J.H., Br. Patent No. 1310011, 1973. Maggs, F.A.P., Smith, M.E. and Robins, GA., Proc. 13th Biennial Conf. Carbon, 1977, Irvine, CA., pp 3- 15. 4 Djuricic, Lj., Polovina, M. and Marinkovic, S., Proc. 16. Carbon’80, Baden-Baden, pp. 193-6,

17.

Capon, A., Alves, B.R. , Smith, M.E and Whlte, M.P., Proc. 15th Biennial Conf. Carbon, 1981, Philadelphia, PA, pp. 259-9. Badami, D.V., Compbell, C., Davy, A.D., and Lindsey, M.J., Proc. Second Conf: Indust. Carbon and Graphite, 1966, Sot. Chem. Bid., p.48. Carrot, P.J.M. and Freeman, J.J., Carbon, 1991,29, 499. Bohra, J.N. , Awasthy, B.R. and Chari, S.S., Fibre Sci. and Technol., 1981, 14, 221. Brunner. P. H. and Roberts, P. V. Carbon. 1980. 18, 217. Indian Standard: Method for determination of specific surface area of powders and porous particles, using low temperature gas adsorption techniques, IS:1 1578. 1986. Gregg, S.J. and Sing, K.S.W., Adsorption, Surface Area and Porosity, Academic Press Inc. London, 1982, p. 214. Rodriguez-Reinoso, F., Linares-Solano, A., Martin- Martinez, J.M. and Lonez-Gonzalez, J.D., Carbon,

. 1984. 22. 123. Garrldo, J., Linares-Solano, A., Martin-Martinez, J.M., Molina-Sabio, M.. Rodriguez-Reinoso, F. and Torreerosa. R.. J. Chem. Sot.. Faradav Trans. 1987. 183, 1081: Martin-Martinez, J.M. Molina-Sabio, M., Rodriguez -Reinoso, F. and Torregrosa, R., Fuel, 1989,68, 204. Mlttelmeijer-Hazeleger, MC. and Martin-Martinez, J.M., Carbon, 1992, 30, 695. Molina-Sabio, M., Rodriguez-Reinoso, F., Caturla, F., and Selles, M.J., Carbon, 1995, 33, 1105.

Electromechanical behavior of carbon fiber

XIAOJUN WANG and D.D.L. CHUNG Composite Materials Research Laboratory, State University of New York at Buffalo

Buffalo, NY 142604400

(Received I July 1996: accepted in revised form 24 February 1997)

Key Words - A. carbon fiber, D. electrical properties, elastic properties, mechanical properties

Due to the importance of carbon fibers for structural composites, much attention has been given to the mechamcal properties of carbon fibers. Considerable attention has also been given to the electrical properties of carbon fibers, because the electrical properties shed light on the electronic structure and are relevant to the electromagnetic interference shielding effectiveness and lightning protection of carbon fiber composites. However, little attention has been given to the electromechanical behavior, which refers to the electrical behavior as affected by mechanical stress [l-5]. The electromechanical behavior is relevant to the use of a carbon fiber as a strain/stress sensor and to the ability of a carbon fiber composite to sense its own strain/stress and damage [6]. Due to the importance of sensing in smart structures and the close relationship between strain/stress sensing and the electromechanical behavior, the electromechanical behavior of materials is of much current interest. Furthermore, study of the electro- mechanical behavior provides fundamental information on the effect of stress on the structure of the material. For these reasons, this paper addresses the electromechanical behavior of carbon fibers.

Previous electromechanical study of carbon fibers reported that, for low-modulus carbon fibers, the electrical resistance increases reversibly with tensile

strain and decreases reversibly with compressive strain, mainly due to dimensional change rather than resistivity change [l-5]. However, fiber damage must occur at high strains prior to fracture and damage is expected to cause the fiber resistivity to increase irreversibly, though the irreversible effect was not investigated in previous work. The practical implication is that an irreversible reststivity increase in the fiber may be used as an indicator of fiber damage. The objective of this paper is to investigate the correlation between irreversible resistivity increase, irreversible strain and elastic modulus decrease, so as to evaluate the effect of fiber damage on the fiber resistivity.

The carbon fiber used was lOE-Torayca T-300 (unsized, PAN-based), diam. 7 mm, density 1.76 g/cm3, tensile modulus 221 + 4 GPa, tensile strength 3.1 f 0.2 GPa and ultimate elongation 1.4%. The electrical resistivity was (2.2 f. 0.5) x 1O‘3 Q.cm, as measured by using the four-probe method and silver paint electrical contacts on single fibers. Single fiber electromechanical testing was conducted by measuring the electrical resistance during static and cyclic tension. The DC resistance was measured by using the four-probe method, using silver paint for the electrical contacts. The outer two contacts (50 mm apart) were for passing a current; the inner two contacts (40 mm apart) were for