Advanced
Structural Study of the Oxidized High Modulus Carbon Fiber using Laser Raman Spectroscopy
Structural Study of the Oxidized High Modulus Carbon Fiber using Laser Raman Spectroscopy
Carbon letters. 2009. Mar, 10(1): 38-42
Copyright ©2009, Korean Carbon Society
  • Received : February 02, 2008
  • Accepted : March 03, 2009
  • Published : March 30, 2009
Download
PDF
e-PUB
PubReader
PPT
Export by style
Share
Article
Author
Metrics
Cited by
TagCloud
About the Authors
Jae-Seung Roh
jsroh@kumoh.ac.kr
Suk-Hwan Kim
Abstract
This study aims to find a correlation between XRD and Raman result of the oxidized high modulus carbon fibers as a function of its oxidation degrees, and compare with the isotropic carbon fiber reported early. La of the high modulus carbon fiber prepared by oxidation in carbon dioxide gas have been observed using laser Raman spectroscopy. The basic structural parameters of the fibers were evaluated by XRD as well. The La of the original high modulus carbon fibers were measured to be 144 Å from Raman analysis and 135 Å from XRD analysis. La of the 92% oxidized fiber were 168 Å by using Raman and 182 Å by using XRD. There was some correlation between the La value obtained from Raman and XRD. However the La value changes of the high modulus carbon fiber through whole oxidation process showed opposite tendency compare with the isotropic carbon fiber because of the fiber structure basically.
Keywords
1. Introduction
The high modulus carbon fiber (HM fiber) shows superior mechanical, electrical and chemical properties because of its highly ordered structure. Therefore HM fiber is considered as the best materials in the field of aerospace applications [1 - 3] . HM fiber has turbostratic structure and the mechanical modulus of fiber is influenced by the degree of orientation along the fiber axis [4] .
A fiber consisted of poorly oriented crystallites and/or very tiny crystallites has many crystallite edges. Since the crystallite edges act as an active site the fiber has low oxidation resistivity [5 , 6] .
The oxidation reactions of carbon based materials take place at the active sites preferentially. In addition to the active sites, the other factors affecting characteristics of oxidation reaction include microstructural characteristics related to the inhomogeneity taking place during the manufacture of the carbon materials [7 - 11] . Also it was reported the homogeneous sheath-core structure of mesophase carbon fiber caused by insufficient carbonization at the middle area of fiber during the manufacture [12] .
Many researchers tried to interpret structural parameters of carbon materials via several analyzing techniques such as X-Ray Diffraction(XRD), High Resolution Transmission Electron Microscopy(HRTEM), High Resolution Scanning Electron Microscopy(HRSEM), Scanning Tunneling Microscopy(STEM) and Raman spectroscopy. However it is not easy to find an article using Raman apparatus for structural analysis of the carbon fiber [13 - 16] .
We reported the results of structural changes of HM fibers due to oxidation using XRD and isotropic carbon fibers using Raman apparatus [17 , 18] . In this study the structural changes of HM fibers prepared by oxidation in carbon dioxide gas have been observed using laser Raman spectroscopy. The basic structural parameters of the fibers were evaluated by XRD as well, and compared with Raman result.
2. Experimental Procedures
- 2.1. Sample preparations and XRD analysis
HM fibers were used as raw materials. Based on assortment of carbon fibers according to crystallites orientation, HM fiber is a highly strengthened carbon fiber with modulus 584 GPa.
The original fibers were oxidized by isothermal heating in the horizontal tube reactor using carbon dioxide gas at high temperatures. Burn-off degree was calculated by dividing the weight loss after oxidation by initial weight of sample.
The detailed physical properties of the fibers, oxidation conditions, and the result of XRD spectra such as 002 interlayer spacing and La value were early reported [17] .
- 2.2. Raman Analysis
For the Raman analysis, a microscope was used to focused
Lager Image
Raman spectrum of the original HM fiber.
Structural Parameters of the HM Fibers Obtained from XRD[13]
Lager Image
Structural Parameters of the HM Fibers Obtained from XRD [13]
laser beam spot on the fiber surface (×500). The wavelength and laser beam size of the apparatus were 514.5 nm of argon laser (Green) with ~2 ㎛ of diameter.
Extended scans from 500 to 4000 cm -1 were performed to obtain both the 1st- and 2nd-order Raman bands of the HM fibers. The band intensity, and band position were obtained from static scans of the first-order Raman spectra using a Lorentzian curve-fitting procedure [18] . Fig. 1 shows scan result of original HM fiber. We discuss only the 1st-order band in this paper.
3. Results and Discussion
- 3.1. XRD analysis[17]
The structural parameters as a function of burn-off degrees obtained from XRD were summarized in Table 1 . 002 and 10ℓ interlayer spacing show slightly changed after oxidation. The Lc and La values of the original fibers were 146 Å and 135 Å, respectively. The 002 interlayer spacing of the original fiber
Lager Image
The first-order D and G-bands of the HM fibers shown as a function of burn-off degrees.
was relatively large as 3.44 Å in comparison with 3.35 Å of graphite. With increasing burn-offs, the La increased up to 182 Å at the burn-off degree of 92%.
It was considered that the increasing of La as the burn-offs increased was caused by removal of atoms located at less ordered area. Therefore, it was assumed that the crystallinity of oxidized fibers became more ordered and the crystallites size of oxidized fibers became smaller than original fiber.
- 3.2. Raman analysis
Fig. 2 shows the 1st-order Raman bands vary as a function of burn-off degrees of the HM fibers. In all cases the spectra exhibited the same appearance, that is, two well-resolved bands, namely D-(~1330 cm -1 ) and G-(~1580 cm -1 ) band. It seems that the bands of oxidized samples were not shifted from original fiber bands.
In Fig. 3 and Table 2 there were no significant shift exist between the each positions of the D- (1355.0 to 1357.0) and
Lager Image
Peak positions of the HM fibers are not changed significantly as a function of burn-off degrees.
Mean Values of the Peak Position Peak Intensity and R(=ID/IG) Values of the First-order D and G-band of the HM Fibers
Lager Image
Mean Values of the Peak Position Peak Intensity and R(=ID/IG) Values of the First-order D and G-band of the HM Fibers
G-bands (1585.4 to 1587.0). The oxidation processes almost have no effect on the band positions and the intensities of the bands.
These results are quite different with results of isotropic fibers as shown in Table 3 [18] . The D-bands of isotropic fibers shifted from 1341.1 to 1358.7. These shifts indicate variation of ordering degrees. Especially it is explained that shifting toward lower wavelength number means the fiber goes to more disordering [19 , 20] .
Even the intensity ratios (R=I D /I G ) of D- and G-band of the HM fibers were varying in the range of 0.44 to 0.77, these were always lower values of isotropic fibers (0.86 to 0.93) as a function of burn-offs is not significant up to 66% burn-off with exception of 16% burn-off. The variation of R values of HM fibers as a function of burn-offs until 66% burn-offed excepting 16%. However the R values slightly decrease at very high burn-offs, 88% and 92%. It could be explained that the reason of non-shifting of Raman bands of HM fibers are caused by well ordered structure of HM fiber.
Structural Data of the Isotropic Fibers[18]
Lager Image
Structural Data of the Isotropic Fibers [18]
Lager Image
Plots of the structural order parameter ID/(ID+IG) of the HM fibers.
Since it has crystalline area, non-crystalline area, and mixed area the structure sould be changed during oxidation process [21 , 22] .
In Fig. 4 structural order parameter of HM fibers oxidized more than 66% burn-off decreased slightly, which means fiber has changed to more ordering. In Table 1 La values obtained from XRD show that La increased over 66% burnoff. Therefore it could be suggested that the initial oxidation take place at less ordered area, the ordered area remains until final oxidation stage.
La values from Raman and XRD summarized in Table 4 . The structural parameter La of the HM fibers obtained from Raman agrees well also with the structural parameters obtained from X-ray diffraction. Even though absolute values from Raman are usually lower than XRD, the La values from Raman analysis were calculated by using L a =C/R, here R=I D /I G and constant C was 80 in this study.
For the isotropic fibers from previous study, it was suggested that each layer consisting of fibers became more disordered as the burn-off proceeded, and the layer stacking became imperfect [18] . From the Raman and the XRD result, it was considered that the decreasing of La as the burn-off increased, was caused by removal of carbon atoms at the edge of crystallites. The Lc and La values of the original fibers were 6.2 Å and 23.6 Å, representing stacks of less than two layer planes. With increasing burn-off, the interlayer spacing increased up to 4.76 Å and Lc increased
Lager Image
Correlation of La values of the HM fiber between Raman and XRD result.
La Values(mean) of the HM Fibers Obtained from Raman and XRD
Lager Image
La Values(mean) of the HM Fibers Obtained from Raman and XRD
up to 6.7 Å. So, the increasing of Lc caused merely widening of interlayer spacing. Therefore, it was cleared that the crystallites of activated fibers became more disordered and the size of crystallites of ACFs became smaller than original fibers. Isotropic fiber has the opened structure originally and could be an ACFs.
For the HM fibers, 002 interlayer spacing and La increased as burn-off increased. This results possibly explained by the oxidation process causing effect of heat treatment to the fiber. The effect of heat treatment during activation for the carbon materials has been reported by many researchers [24 - 27] . It is suggested that increase of La is caused by crystalline growth by heat treating effect during long time oxidation at high temperatures. Furthermore if the crystallinity of growth part was formed to be a poorly ordered, 002 interlayer spacing could be much larger.
The structural parameter La of the HM fibers obtained from Raman compares well also with the structural parameters obtained from X-ray diffraction in Table 4 and Fig. 5 . Fig. 5 shows there are some correlations between the La values obtained from Raman and XRD for the HM fiber. Also, for the isotropic fiber, it could be obtained same result between Raman analysis and XRD analysis.
4. Conclusions
The structural changes of high modulus carbon fibers prepared by oxidation in carbon dioxide gas have been observed using laser Raman spectroscopy. The basic structural parameters of the fibers were evaluated by XRD as well, and compared with Raman results.
No significant shifts measured were observed between the each positions of the D-(1355.0 to 1357.0) and G-bands (1585.4 to 1587.0).
Raman analysis shows that R values slightly decreased at very high burn-offs (88% and 92%), and therefore La increased as burn-off increased. These results were possibly explained that the oxidation process affect the heat treatment to the fiber.
The La of the original HM fibers were measured to be 144 Å from Raman analysis and 135 Å from XRD analysis. La of the 92% oxidized fiber were 168 Å by using Raman and 182 Å by using XRD. The structural parameter La of the HM fibers obtained from Raman compares well also with the structural parameters obtained from X-ray diffraction.
However the La value changes of the high modulus carbon fiber through whole oxidation process showed opposite tendency as compared with the isotropic carbon fiber because of the fiber structure difference basically.
It is suggested there was some correlation between the La value obtained from Raman and XRD for the HM fiber. Also, for the isotropic fiber, same results could be obtained between Raman analysis and XRD analysis.
References
Endo M , Kim C , Kasai T , Mathews M. J , Brown S. D. M , Dresselhaus M. S , Tamaki T , Nishimura Y 1998 Carbon 36 1633 -
Hong S. H , Korai Y , Mochida I 2000 Carbon 38 805 -
Montes-Moran M. A , Young R. J 2002 Carbon 40 845 -
Donnet J. B , Bansal R. C 1990 “Carbon Fibers” Marcel Decker New York 8 -
Dami T. L , Manocha L. M , Bahl O. P 1999 Carbon 29 51 -
Ismail M. K 1991 Carbon 29 777 -
Mahajan O. P , Yarzab R , Walker Jr. P. L 1978 Feul 57 643 -
Sanchez A. R , Elguezabal A. A , Torre Saenz L. L 2001 Carbon 39 1367 -
Kasaoka S , Sakata Y , Kayano S , Masuoka Y 1983 Int. Chem. Eng. 23 477 -
Hu Y. Q , Nikzat H , Nawata M , Kobayashi N , Hasatani M 2001 Feul 80 2111 -
Rafsanjani H. H , Jashidi E , Rostam-Abadi M 2002 Carbon 40 1167 -
Blanco S , Lu C , Rand B 2002 Carbon 40 2002 -
Sharma A , Kyotani T , Tomita A 2000 Carbon 38 1977 -
Kovalevski V. V , Buseck P. R , Cowley J. M 2001 Carbon 39 243 -
Senneca O , Salatino P , Masi S 1998 Fuel 77 1483 -
Busyin R. M , Rouzaud J. N , Ross J. V 1995 Carbon 33 679 -
Roh J. S 2004 Carbon Science 5 27 -
Roh J. S 2008 Carbon Science 9 127 -
Montes-Moran M. A , Young R. J 2002 Carbon 40 845 -
Kuo C. T , Wu J. Y , Lu T. R 2001 Materials Chemistry and Physics 72 251 -
Lespade P , Al-Jishi R , Dresselhaus M. S 1982 Carbon 20 427 -
Escribano R , Sloan J. J , Siddique N , Sze N , Dudev T 2001 Vibrational Spectroscopy 26 179 -
Montes-Moran Miguel A , Young Robert J 2002 Carbon 40 845 -
Kamegawa K , Nishikubo K , Yoshida H 1998 Carbon 36 433 -
Sharma A , Kyotani T , Tomita A 1999 Fuel 78 1203 -
Yoshizawa N , Maruyama K , Yamada Y , Zielinska-Blajet M 2000 Fuel 79 1461 -
Gondy D , Ehrburger P 1997 Carbon 35 1745 -