1149
INFRAREID SPBCTRA OF GROUND GRAPH IT^ The non-Newtonian viscosity of dilute solutions of flexible polymers is a very complex phenomenon which is affected by many contributing factors such as excluded volume effect, hydrodynamic interaction rigidity of polymer chains and so on. Therefore, it is difficult to assess definitely various contributing factors from our experimental results alone. Acknowledgments. The authors wish to express their
deep gratitude to Dr. S. Iwayanagi of the Institute of Physical and Chemical Research for many helpful discussions and suggestions during this work. The authors thank Professor M. Nagasawa of Nagoya University for supplying the fractionated polystyrene samples and Dr. Y. Chikahisa of Tokyo University of Agriculture and Technology for supplying the numerical results of Fixman’s theory.
Infrared Spectra of Ground Graphite by R. A. Friedel* U . S. Bureau of Mines, Pittsburgh, Pennsylvania
and G. L. Carlson Carnegie-Mellon University, Pittsburgh, Pennsylvania
16.913 (Received November 3,1970)
Publication coats assisted by the U.S. Bureau of Mines
The infrared spectrum of ground graphite has been observed. After very intense grinding two broad absorption bands appeared at 1587 and 1362 cm-l and weak bands appeared at 830 and 2200 cm-l; 1587 and 1362 cm-1 are practically the same frequencies found by Tuinstra and Koenig in Raman spectra of powdered graphite and various powdered carbons. A shift of about 20 cm-l to higher frequencies was observed for the most intense infrared band as particle sizes decreased; a similar shift has been reported by Tuinstra and Koenig for Raman frequencies. Their Raman lines were obtained on crystalline graphites while the similar infrared bands were observed for noncrystalline graphite; thus both sets of bands probably originate in the “aromatic” structure of the graphite matrix, crystalline or noncrystalline. Infrared investigations of some carbonaceous materials have not been poesible in the past because of the intractability of these materials, e.g., carbon blacks, coal chars, and activated carbons. Recently, Friedel and Hofer succeeded in obtaining an infrared spectrum of one of the most difficult substances, a coal-based activated carbon.’ By the use of appropriate sample preparation and instrumental techniques a transmission spectrum of activated carbon was obtained for the first time. Extensive and efficient grinding were found to be important. Mattson, et al., have published spectra of sorbates on activated carbon using the attentuated total reflectance (ATR) m e t h ~ d . ~ - ~ There are other carbonaceous materials, both more and less tractable than activated carbon; the literature contains infrared spectra of difficult carbonaceous materials such as pyrolytic carbon black,gs10 and coal Coals should also be mentioned here though they are less difficult materials; it is possible to obtain good spectra of coals with absorption intensities up to 80% by the transmission method.8112-15 Until recently it has been difficult to obtain good spectra
of coals by the ATR method. Various people over a period of years have tried; S. Polchlopek succeeded (1) R. A.Friedel and L. J. E. Hofer, J. Phys. Chem., 74, 2921 (1970). (2) J. S.Mattson, H. B. Mark, Jr., M. D. Malbin, W. J. Weber, and J. C. Crittenden, J. Colloid. Interface Sci., 31, 116 (1969). (3) J. 8. Mattson and H. B. Mark, Jr., ibid., 31, 131 (1969). (4) J. S.Mattson and H. B. Mark, Jr., Anal. Chem., 41, 355 (1969). (5) R. A. Friedel and M. G. Pelipetz, J. Opt. SOC.Amer., 43, 1061 (1953). (6) R. A. Friedel, “Proceedings of the 4th Carbon Conference,” S. Mrozowski, Ed., Pergamon Press, New York, N. Y., 1960, pp 321-336. (7) R. A. Friedel, R. A . Durie, and Y. Shewchyk, Carbon, 5 , 559 (1968). (8) J. K.Brown, J. Chem. Soc., 744 (1955). (9) V. A. Garten and D. E. Weiss, Aust. J. Chem., 10, 295 (1957). (10) V. A. Garten and D. E. Weiss, “Proceedings of the 3rd Carbon Conference,” S. Mrozowski, Ed., Pergamon Press, New York, N.Y., 1969,p 295. (11) J. K.Brown, J. Chem. SOC.,752 (1955). (12) R.A. Friedel and J. A . Queiser, Anal. Chem., 28 22 (1956). (13) C. G. Cannon and G. B. B. M. Sutherland, Trans. Faraday Soc., 41, 279 (1945). (14) R. A. Friedel, Brennst. Chem., 44, 23 (1963). (15) R. A. Friedel, “Applied Infrared Spectroscopy,” D. N. Kendall, Ed., Reinhold Pub. Co., New York, N. Y., 1966,pp 319,321.
The Journa2 of Physieal Chemistry, Vol. 76, No. 8, 1971
R. A. FRIEDEL AND G. L. CARLSON
1160 FREQUENCY. wovenumbers
WAVELENGTH, microns
Figure 1. Spectral intensities of ground graphite increased with efficient grinding for: A, 27 hr; B, 48 hr; C, 72 hr; D, 96 hr; E, 120 hr. These spectra are on aliquots removed from the sample after grinding for the times indicated. All spectra are scale-expanded except the one indicated. KBr pellets were used.
in getting a weak ATR spectrum of a Pittsburgh bituminous coal (unpublished work), and Bent and Ladner obtained a spectrum from a solid piece of anthracite coal. l6 Of the anthracite samples investigated only the one was sufficiently flat to produce a spectrum, actually a partial spectrum. However, Mattson has demonstrated recently that a good spectrum of a Pittsburgh coal can be obtained through the use of ATR methods with modern infrared instruments having accurate scale-expansion systems (unpublished work). The ATR method is advantageous because of the simplicity of sample preparation, whereas the transmission method has the advantage of being applicable to very small samples. Good spectra by transmission measurements on KBr pellets of various carbonaceous materials have been recorded successfully; next it seemed appropriate to attempt to obtain spectra of the most difficult carbonaceous substance, graphite. By means of exhaustive, efficient grinding it was possible to demonstrate that infrared spectra of ground graphite could be observed. The infrared absorption of graphite was first looked at by Cannon,” who investigated a thin mineral oil mull of graphite. He ascertained that absorption (and/ or scatter) was essentially the same throughout a wide range of infrared frequencies. No spectral information was obtained from such data. Later, attempts were made to obtain absorption spectra of graphite in the infrared (Friedel, unpublished work). Samples were prepared by rubbing powdered graphite on soft polyethylene which retained the graphite as a film; no specific infrared absorption was found. However, absorption and reflectance bands The Journal of physical Chemistry, Vol. 76, No. 8,1071
were found in the ultraviolet region on examination of powdered graphite on soft polyethy1ene.ls Infrared studies of polycrystalline graphite have been carried out by Foster and Howarthale They determined refractive and absorption indices from 1 to 10 p by the reflection method of AveryS20 One absorption maximum of moderate intensity was observed a t -1300 cm-’; for reasons unknown, this frequency does not coincide with the frequencies of either of the two strong absorption bands observed in our spectra and discussed below.
Experimental Section For the present investigation of possible absorption of graphite in the 4000-250-cm-’ infrared region it was decided to try the sample preparation technique that was successful with activated carbons, namely, very extensive and efficient grinding. Partial spectra were obtained after grinding for 24 hr but for better development of the spectrum many more hours of grinding were required. As noted elsewhere, it was necessary to utilize a very small sample, as a large sample softens the blow of the small ball bearings used for grinding in a steel capsu1e.l It is obvious that much more extensive grinding would be required for graphite because of its good lubricating charac(16) R. Bent and W. R. Ladner, Fuel, 44, 243 (1965). (17) C. G.Cannon, Nature, 171, 308 (1953). (18) R. A. Friedel and H. L. Retcofsky, “Proceedings of the 5th Carbon Conference,” Vol. 11, 8. Mrosowski, Ed., Pergamon Press, London, 1903, p 165. (19) P. J. Foster and C. R. Howarth, Carbon, 6 , 719 (1968). (20) D.G.Avery, Proe. Roy. Soc., Ser. B , 65 1087 (1952).
INFRARED SPECTRA OF GROUND GRAPHITE
1151
teristics. Grinding for 72 hr developed a reasonably good infrared spectrum. Two broad bands were observed a t 1565 and 1382 cm-’. With continued grinding the two bands become stronger and sharper, as shown in Figure 1. Also, the frequencies shift slightly to 1587 and 1362 cm-’ (Table I). I n addition to these two absorption bands of graphite there is a weak absorption band at 830 cm-’ and a combination band at 2200. The spectrum out to 250 cm-l was investigated but no further absorption was observed.
Table I : Infrared and Raman Frequencies of Graphite Raman Infrared
1575 1587
1355 1362
830 (w)
2200 (w)
Discussion of Results The extensive grinding reduces the graphite sample to minute particle sizes on which useful infrared transmission measurements were obtained. The process works also for lamp black, for several activated carbons, and for carborundum.21 The frequencies found for ground graphite are essentially the same as those found for similar carbons. By the criterion of X-ray diffraction patterns, it is apparent that the materials measured are not crystalline graphites. Under extensive grinding graphite loses many of the X-ray diffraction peaks that are characteristic of graphite.22 However, this situation does not alter the fact that the infrared spectra obtained for ground graphite produce information concerning the molecular structures involved. Graphite is altered by the grinding but the disappearance of crystallinity does not mean that the carboncarbon bondings in the original graphite are broken. The grinding operation does not introduce sufficient energy into the system to break many carbon-carbon bonds. Therefore it is considered that the infrared
spectrum of graphite shown is indeed characteristic of the molecular structure of graphite. These results are very similar to the laser-Raman results reported by Tuinstra and K ~ e n i g . Their ~ ~ results indicate for ground graphite one intense scattering band at 1575 cm-’ and a weaker band a t 1355 cm-’. They assigned these frequencies to the EZgand the A’, modes respectively of crystalline graphite with D6h4crystal symmetry. The two infrared bands that we find at about 1587 and 1362 cm-’ compare reasonably well with the Raman bands. Further, the infrared maximum of the strongest band shifts from 1565 to 1587 as particle sizes decrease with grinding. Tuinstra and Koenig found that a closely similar shift occurred in the Raman spectra with decreasing particle sizes. The infrared bands were obtained on ground graphite for which X-ray measurements indicate that the typical crystalline fine structure of graphite has disappeared. And yet, the frequencies observed are practically the same as those of the Raman bands. It would appear that the observed infrared and Raman spectra are not related to the crystallinity of graphite. Perhaps vibrations of the “aromatic” structure of graphite, crystalline or noncrystalline, are responsible for the observed spectra. The weak infrared bands found at 830 and 2200 cm-l are not reported for the Raman spectra. It is likely that the 2200 band is a combination band resulting from the 830- and 1362-cm-’ bands which total 2212 cm-’. This value is reasonably close to the observed 2200 cm-l. The 830-cm-’ band could be due to an aromatic impurity in graphite. If so, this would then remove the possibility of assigning the 2200-cm-1 band as a combination band. (21) R. A. Friedel and L. J. E. Hofer, unpublished work. (22) P. L. Walker, Jr., and 8 . B. Seeley, “Proceedings of the 3rd Carbon Conference,” 8. Mrozowski, Ed., Pergamon Press, New York, N. Y., 1959, p 481. (23) F. Tuinstra and J. L.Koenig, J. Chem. Phys., 53, 1126 (1970).
The Journal of’Physical Chemistry, Vol. 76,No. 8,1071
