J. Phys. Chem. 1990, 94, 5639-5641 Tahir-Kheli and Holzheylo consider a similar model but assume a mutual (repulsive) interaction of the diffusing excitations so that the diffusion properties depend on the temperature-dependent concentration. The changing TI dispersion may generally be considered as a manifestation of a transition between an ergodic behavior at high temperatures to a nonergodic situation connected with frozen-in states in the time scale of the experiments at low temperatures. Thus a certain analogy to the glass transition of glass-forming materials can be stated. Other experiments provide similar findings indicating changes a t 200 K. An N M R example is the temperature dependence of the second moment of the proton line.' The Mirssbauer resonance absorption of the heme iron in myoglobin shows a crossover at 200 K." Flash photolysis experiments for ligand binding indicate the freezing of ligand transport at about 200 K.I2 Recently an interesting absorption and scattering study using Mossbauer ra(IO) Tahir-Kheli, R.A.; Holzhey, D. J. In Proceedings ofthe Sir Roger Elliorr's 60th Eirrhday Symposium; Blackman, J. A., Taguena, J., Eds.; Oxford University Press: Oxford, 1990. (11) Keller, H.; Debrunner, P. Phys. Rev. Lett. 1980, 45, 68. (12) Austin, R. H.; Beeson, K. W.; Einsenstein, L.; Frauenfelder, H.; Gunsales. I. C . Biochemistry 1975, 14, 5355.
5639
diation has been reported." The temperature dependence of the mean square displacements of myoglobin crystals was found to show a clear break at about 200 K. Above this temperature, the displacements were interpreted by intramolecular short-range modes, which could be identical with those that are responsible for the peculiar backbone fluctuation. The fact that findings of N M R and non-NMR experiments can be projected on each other reduces the variety of possible interpretations considerably. The conclusion is in particular that the low-frequency N M R relaxation times of protein protons (in the absence of overall tumbling) are governed by backbone dynamics. The N M R data thus reveal the peculiar spectral density of these fluctuations.
Acknowledgment. We thank Professor Tahir-Kheli for illuminating discussions, and H.-W. Weber and J. Wiringer for the cooperation in the course of this work. Financial support by the Deutsche Forschungsgemeinschaft and the Bundesministerium fur Forschung und Technologie (grant no. 01VF85203) is gratefully acknowledged. ( I 3) Nienhaus, G . U.; Heinzl, J.; Huenges, E.; Parak, F. Nature 1989,338, 665.
Scanning Tunneling Microscopy of Polythiophene, Poly(3-methylthiophene), and Poly(3-bromothiophene) G. Caple,**+B. L. Wheeler: R. Swift,' T. L. Porter,* and S. Jefferst Departments of Chemistry and Physics, Northern Arizona University, Flagstag, Arizona 86001 (Received: March 5, 1990; In Final Form: June 6 , 1990) Scanning tunneling microscope (STM) images were obtained from polythiophene, poly( 3-methylthiophene), and poly(3bromothiophene) doped with tetrafluoroborate. Images of polythiophene contained both helical and chainlike structures. The distance between thiophene units was about 4 A, and imaging may have occurred at the sulfur atoms. The poly(3methylthiophene) had an ordered zigzag structure in 350-A2 images, but less regular structure in 30-A2 images. The poly(3-bromothiophene) surface images gave indications of ridges in 500-A2 images, but an almost regular dislocation in its linear chains in 20-A2 images. It would also appear that the sulfur atoms are involved in the tunneling process. There has been a great deal of interest in the surface structure and electronic structure of conducting organic polymers.'q2 Recently, STM was used to image polypyrrole to give evidence of semicrystalline growth and helical polymer g r ~ w t h . Helical ~ conformations have been predicted for polythiophene by using MNDO for geometrics and EHT for band-gap calculations.' Polypyrrole-coated platinum has been imaged in an acetonitrile solution: which could have a different structure than dry films. The morphology of the conducting films has been shown to be dependent upon both the dopant anion incorporated and the conditions of polymer f o r m a t i ~ n . The ~ insoluble nature of some electrochemically deposited conducting polymers has made their characterization more difficult. CP-MAS NMR experiments have been used to characterize organic conducting films6 as have X-ray diffraction and transmission electron microscopy4 and IR and XPS.7.8 STM images have been cited as direct evidence in support of a helical structure in doped polypyrr~le.~ This type of structure has also been calculated for highly doped 3-meth~lthiophene.~ While helical structures have been imaged on solid surface^,^ rodlike anti planar configurations may be preferred on the neutral molecule in solution. This is supported by SANS measurements, and in solution doping can change the configuration.I0 STM images'l,12 were obtained of electrochemically deposited thick films of polythiophene, poly(3-methylthiophene), and 'Department of Chemistry. *Department of Physics.
poly(3-br0mothiophene).'~ Platinum was used as a substrate material. Images were taken on various parts of the film; in some ~
~~~
~-
~~~
( I ) Cui, C. X.; Kertesz, M. Phys. Rev. 1989, 40, 9661. (2) Stafstrom, S.;Bredas, J. L. Phys. Reu. 1988, 38, 4180. (3) Yang, R.; Dalsin, K. M.; Evans, D. F.;Christensen, L.; Hendrickson, W. A. J. Phys. Chem. 1989, 93, 51 1. (4) Fan, F.-R. F.; Bard, A. J. J. Electrochem. Soc. 1989, 136, 3216. ( 5 ) Warren, L. F.; Walker, J. A.; Anderson, D. P.; Rhodes, C. G.; BucWey, L. J. J . Electrochem. SOC.1989, 136, 2286. (6) Stein, P. C.; Hartzell, C. J.; Jorgensen, B. S.; Earl, W. L. Synth. Mer. 1989, 29, E297. (7) Street, G . B.; Clark, T. C.; Krounbi. M.: Kanazawa. K.: Lee. V.: Pfluegen, P.; Scott, J. C.; Weiser, G . Mol. Cryst. Liq. Cryst. 1982.83, 253. (8) Zeller. M. V.: Hahio. S. J. Surf. Interohuse Anal. 1988. 327. (9) Gamier, F.;Tourillan, G.; Bariud, J. k.; Dexpert, H. J.'Muter. Sci.
--.--,---
1985. 20. 2687. (IO) Aime, J. P.; Borgain, F.; Schott, M.; Eckhardt, H.; Miller, G . G.; Elsenbaumer, R. L. Phys. Reu. Lett. 1989, 62(1), 55-58. ( I 1) The STM used in these studies was a custom ultra high vacuum ~
(UHV) compatible design, employing a UHV inchworm for coarse sample approach and a single-tube element for x , y , and z scanning. Atomic resolution images of highly oriented pyrolytic graphite were used for calibration. (12) Binnig, G.;Smith, D. P. E. Reu. Sci. Instrum. 1986, 57, 1688. (1 3) Deposition was done galvanostatically on a EGG-PAR Model 362 scanning potentiatat. The plating surface was platinum foil 4 mm X IO mm. This size was necessary to make the necessary electrical contacts in the STM. The electroplating solutions contained 3.0 mL of acetonitrile (Aldrich HPLC Grade), 0.1 mL of the thiophene compound, and 50.0 mg of lithium tetrafluoroborate. Five milliamperes of current was passed through for 240 s. Large plates formed this way; covering of the entire surface was somewhat uneven, being thickest at the edges. The foil was mounted for STM imaging by using conducting glue to attach it to the metal strip that fit the contacts of the STM.
0022-3654/90/2094-5639$02.50/00 1990 American Chemical Society
Letters
5640 The Journal of Physical Chemistry, Vol. 94, No. 15, 1990
.,W'
Figure 2. The innagc\ 01' pol! (.l-mcthJ Ithiopticric). Irn;iyc :\ 5 t i o u . s ;i highly structured iigmg-like surfiicc, with about 30 3-methylthiophene units per zig. Image R shows a short section along the top of one of the ridges and looks like an irregular kinked arrangement.
Figure 1 . Three images of polythiophene. Image A is possibly a helical structure, although it appears to be flattened in the image. Image R appears to be a linear arrangement showing no definite cross-linking. but possibly dislocation or binding at the 3-position in thiophene. There is an offset in the left-hand chain, about 1 / 3 of the way of the chain and a kink in the upper part of the right chain. Image C shows an enlarged section from one of the chains in R; tunneling appears to occur about where the sulfurs should be. The structure imaged in 1 R and 1C appear to be consistent with an anti planar polythiophene.I0
places, the film was too "rough" for imaging, and here we show images that we could find in several places on the film. Considerable searching was done to find sections of the film flat enough to allow good STM images; this flat area was usually about 500 A2. Even though repeated images were taken, only about 1 part in IO'*of the total surface was imaged. About 10%of the areas examined produced reproducible images similar to those shown in this report. A11 images were obtained with a bias voltage of 200 mV and a tunneling current of 1 nA. The probe tip was biased negative, and the organic conductor was grounded. The areas used here were considerably smaller than the larger
images presented earlier, but an SEM study on thiophene showed 1A the expected nodular growth, as already r ~ p o r t e d . Figure ~ shows an STM image of an area of polythiophene that resembles the geological formations called hogbacks. This is probably a coil similar to that of polypyrrole,3 and that predicted from a combination of MNDO and EHT crystal orbital calculations.' Cross sections of this image show the height to be 18 f 3 A and the width to be 22 f 2 A. Other areas of the film gave a more linear, stringlike scqucnce of thiophene units as shown in Figure 1 R. This linear arrangcmcnt has a perpendicular spacin of 3.8 f 0.2 A, a dot width of 5 f 2 A, and a height of 4 f 2 tf which supports an anti planar configuration for polythiophene. An anti configuration has been detected by X-ray ~cattering.'~Figure 1 C shows an enlarged portion of the image, taken from the upper portion of the leftmost line of dots in Figure 1 R. Tunneling appears to occur at about 4-A intervals, and this is about wherc the sulfurs might be expected in an anti planar configuration. While the off centering of the tunneling sites may be indicative of an anti conformation,' the image gives no indication as to the electronic structure, such as the quinoid form or the aromatic form. MNDO calculations give quinoid bond length between thiophenes of 1.36 A,' for either the positively or negatively charged polymer. A molecular mechanics calculation using the PC model on the neutral chain gave the same quinoid distance as 1.35 A, and 1.45 A for bond distance between non-quinoid thiophenes, but both calculations indicated the sulfurs were about 1 A off the center line. This is a somewhat larger distancc than the measured 0.5-0.7 A (14) Rruchner. S.; Porzio, W. Makromol. Chon. 1988. 89. 961.
Letters
The Journal of Physical Chemistry, Vol. 94, No. 15, 1990 5641
Figure 3. Four imagcs of poly(3-bromothiophene). Imagc A shows long parallel ridges. Image R shows a linear section atop one of the ridges; the chain appears not to be a continuous anti planar configuration. Image C shows an enlarged view of the chain and image D another section o f a chain enlarged further and tunneling appears to occur about where the sulfur atoms might be expected.
offset in Figure IC; however, this does support the linear anti planar arrangement of the thiophene without distinguishing between quinoid or aromatic structures. The images obtained from poly( 3-methylthiophene) were quite different than those from polythiophene. A highly structured surface was seen in the 350 A X 350 A images, Figure 2A. A zigzag pattern was observed, with approximatcly a 90' kink every 100-1 10 A. or about every 30 or so thiophene units. The height is about 20 f 3 A, the width about 22 f 3 A. An image 20 A X 30 A is shown in Figure 2R. This shows an irregular structure, with an apparent kink at the lower left, its maximum width being about 14 f 3 A, not very different than the 18 A reported for or the 16 A calculated for the coil conformation of polypyrr~le,~ polythiophene.' In imaging poly( 3-methylthiophene), nothing that could bc attributed to an anti planar conformation was detected. The images of poly(3-bromothiophene) proved to be interesting. A rather large image at 500 A X 800 A shows ridges that look 55 f 3 A high and 60 f 3 A wide with pitches that occur every 35 f 5 A (Figure 3A). It even appears that the ridges may be part of a supercoil structure. Super helical coils for polythiophene However, high-resolution images may have pitches of 26 100 A X 100 A taken in the same area as the ridges showed a more linear structure, with a curious triad arrangcment almost regularly spaced along the chain, as shown in Figure 3R. A more detailed image of this linear structure is imaged at 40 A X 40 A in Figure 3C, which is from the upper part of the structure in image Figure 3R. Here there should be 4-5 bromothiophene units between the kinks or dislocation. The dislocation is about 8 A across and might be due to bromine alignments along the chain. One dislocation is "up", and the other is "down". There is a possibility that the bromines come up in 3,3-position and 3.4position combinations (numbering each individual thiophene unit (15) Yang, R.; D. F. Evans, J . Vuc. Si.Technol. In press.
the same way). MNDO calculations for poly(3-methylthiophene) assumed the methyls came up in the same relative positions in thc chains; perhaps this is not so, and there appear to be irregularities in the structure shown for poly(3-methylthiophene) (Figure 2R). Figure 3D is a 9 X 12 A image of poly(3-bromothiophene) but is taken from neither Figure 3B nor 3C areas of the film. Again the image has bright spots, which might be interpreted as tunneling from about where the sulfur atoms might be found. The spacing between peaks is 1 and 2 A along the chain and perpendicular to the chain, respectively, but again there are uncertainties in the lateral measurements. The images presented have indicated several structure types. Some may be in agreement with proposed helical structures1s but also show apparent linear anti planar polythiophene configured structures with almost parallel chains. In some of the images observed, however, the angle between some of the linear chains could be as large as 30'. We did not observe anything that looked like branching, but we could be observing dislocations or polymerization at the 3-position. With poly(3-methylthiophene) we found structures not easily interpreted but as part of a larger ordered structure. The poly(3-bromothiophene) appears to be linear, with kinks in the chain. The largest substituent so far reported, rerr-butyl, gives a purported helical structure.' Scanning tunneling microscopy could become a useful tool in further studies of these conducting surfaces, but there needs to be ways of looking at a statistically valid amount of the surface. As the actual contrasting forming mechanism in all the images is still unknown, STM imaging is still very interpretive, but STM images should give some of the most detailed information about surface configurations in the conducting polymers.
Acknowledgment. We thank the NAU Organized Research Fund for funds to construct the STM in support of this work. We also thank the NIH-MBRS for support for two of us (G.C. and R.S.), and the Research Corp. for support for T.P.
