Temperature dependent X-ray observations of TTF-TCNQ

Volume 47A, number 4 PHYSICS LETTERS 8 April 1974 TEMPERATURE DEPENDENT X-RAY OBSERVATIONS OF TTF-TCNQ E.F. SKELTON Naval Research Laboratory, Washington, D.C 20375, USA and A.N. BLOCH, J.P. FERRARIS and D.O. COWAN The Johns Hopkins University, Baltimore, Maryland 21218, USA Received 15 February 1974 Single crystal X-ray oscillation photographs of the organic solid TTF-TCNQ have been recorded at temperatures above and below the conductivity maximum. No evidence of a multiplicative enlargement of the unit cell in a direc- tion parallel to the conducting chains has been observed. The organic charge-transfer salt tetrathiofulvalinium tetracyanoquinodimethane (TTF-TCNQ), the best or- ganic conductor known, exhibits a transition from metallic to non-metallic conduction as the temperature is decreased below 60 K [1]. At room temperature the structure of TTF-TCNQ consists of parallel columns of separately stacked TTF (electron donor) and TCNQ (electron aeceptor) molecules [2]. Electrical conduc- tion occurs largely along these chains, and it has been suggested [3-6] that the metal-insulator transition is associated with the Peierls instability [7] of a simple one-dimensional metal. Added interest in the problem has been generated by theoretical predictions of giant conductivities associated with such a transition [4, 5]. Such conductivities have been provisionally reported [3] for a few crystals of TTF-TCNQ, although other experiments [1,8,9] have failed to confirm these find- ings. The crystallographic structure of TTF-TCNQ is monoclinic with alternating TTF and TCNQ chains running parallel to the b-axis, each unit cell contains two TTF and two TCNQ molecules, one on each of four chains. The Peierls distortion would consist of a multiplication of the period along the chains at low temperature. In search of such an effect we have carried out temperature-dependent X-ray measurements on single crystals of this material. Four different single crystals were studied in nine separate low temperature experiments. The crystals, in the form of rectangular platelets with approximate dimensions of 2 X 0.2 X 0.03 mm 3 , were mounted in a Joule-Thomson cryogenic refrigerator. The refriger- ator, equipped with a low absorbing Be tail, was in turn mounted on a Picker biplane X-ray diffracto- meter. The samples were oriented with their b-axes approximately parallel to the co[20.rotation axis of the diffractometer and irradiated with ftltered radia- tion from a Mo X-ray tube. The crystals were oscil- lated through co angles ranging from +10 ° to +20 ° with the c* -reciprocal lattice vector approximately tangent to the Ewald sphere at co = 0 °. Up to five layer lines were recorded on oscillation photographs made with a flat-plate film cassette mounted on the 20-arm of the diffractometer. Measurements were made at room temperature (~ 294 K), 77 K, and about 10 K on all samples. The exposure times were generally about an hour duration below 60 K; in two specific instances however, low temperature exposure was continued for more than 20 and 24 hours, respectively. In addition, two photographs were recorded while the sample was slowly cooled from 77 to 10 K and subsequently heated back to 77 K. In all cases, no evidence was found for a gross distortion of the unit cell in a direction parallel to the conducting chains. The oscillation photographs were all similar, in that no reflections were unexplain- ably lost, nor were any new reflections observed over the temperature range studied. Visual estimates of the relative intensities appeared comparable in all eases. Particular attention was directed to the regions 313 Volume 47A, number 4 PHYSICS LETTERS 8 April 1974 between the layer lines since a multiplicative enlarge- ment of the unit cell could give rise to additional reflections in this region. Basically the data indicate that the b-axis contracts by about 3% over the thermal range studied (294 - 10 K)~. One curious fact is noted however: although the equilibrium temperature dif- ference between the ends of the sample is estimated to be substantially less than 80 mK and thermal changes were always carried out over a period of several hours, the crystals would often split normal to the b-axis after repeated thermal cycling. This splitting was evidenced by an obvious division of the diffraction peaks and subsequently confirmed by microscopic examination of the sample. In summary, we find no evidence for a lattice distortion in the low-temperature X-ray structure of TTF-TCNQ.f This is consistent with out previous observation [8] that the band structure of TTF-TCNQ is likely to be more complicated than that of a simple * This result is in excellent agreement with very recent thermal expansion measurements communicated to us by J.W. Cook, D.A. Glocker, and M.J. Skove (Clemson U.); a paper des- cribing their work has been submitted to J. Appl. Phys. t Similar and independent conclusions have been commun- icated to us privately by A.C. Lawson (U. of California, San Diego) on the basis of powder photographs, and by G.D. Stukey (U. of Illinois) and P. Coppens (SUNY, Buffalo) on the basis of their current study of the full low- temperature X-ray structure. one-dimensional Peiefls system. We emphasize, how- ever, that our present results do not preclude the possibility of a static or dynamic Peierls supedattice [10]: we simply fred that if any such effects do occur in TTF-TCNQ, then the distortions are very small. References [1] J. Ferraris, et al.,J. Amer. Chem. Soc., 95 (1973) 948. [2] T.E. Phillips et al., Chem. Comm. (1973) 471; T.J. Kistenmacher, T.E. Phillips and D.O. Cowan, Acta Cryst., to be published. [3] L.B. Coleman et al., Solid State Comm. 12 (1973) 1125. [4] J. Bardeen, Sol. St. Comm. 13 (1973) 357. [5] P.A. Lee, T.M. Rice, and P.W. Anderson, Phys. Rev. Lett. 31 (1973) 462, to be published; P.W. Anderson, P.A. Lee and M. Saitoh, Sol. St. Comm. 13 (1973) 595. [6] B.R. Patton and L.J. Sham, Phys. Rev. Lett. 31 (1973) 631. [7] R.E. Peierls, in Quantum theory of solids (Oxford Univ. Press, Oxford, England, 1955), p.108. [8] A.N. Bloch, J.P. Ferraris, D.O. Cowan and T.O. Poehler, Sol. St. Comm., to be published. [9] D.E. Schafer, F. Wudl, G.A. Thomas, J.P. Ferraris and D.O. Cowan, preprint. [10] R. Comes, M. Lambert, H. Launois, H.R. Zeiler, Phys. Rev. B 8 (1973) 571; R. Comes, M. Lambert, H.R. Zeller, Phys. Status Solidi B 58 (1973) 587; B. Renker, H. Rietschel, L. Pintschovius, W. Glaser, P. Bruesch, M.J. Rice, S. Strassler and H.R. Zeller, preprint. 314