Microreflectance Infrared Study of Bis(ethylenedithio)- tetrathiafulvalene (BEDT-TTF or ET) Salts

JOHN R. FERRARO,* H. H A U W A N G , M Y U N G - H W A N W H A N G B O , * and P H I L S T O U T Chemistry and Materials Science Divisions, Argonne National Laboratory, Argonne, Illinois 60439 (J.R.F., H.H. W.) ; Department o[ Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204 (M.-H. W.); and Bio-Rad, Digilab Division, Cambridge, Massachusetts 02137 (P.S.)

For several B- and ~-phase salts of bis(ethylenedithio)tetrathiafulvalene (BEDT-TTF or simply ET) and its deuterium analog ds-ET, microre- flectance infrared spectra were obtained by employing polarized and unpolarized light, and their vibronic regions were examined. These salts exhibit a strong vibronic absorption under polarized light. The vibronie absorption of the B-phase salts has a much stronger polarization-depen- dency than that found for the ~-phase salts. For the B-phase salts, the optimum vibronic absorption occurs when the polarized light vector is parallel to their donor-molecule stacking direction. Among the ET salts without structural disorder, the highest C-C-H bending frequency for the superconductors is lower than ~1320 cm ' while that for the non- superconductors is higher than ~ 1320 em-l .

Index Headings: Polarized infrared; Microreflectanee; Organic con- ducting salts.

I N T R O D U C T I O N

Reflection infrared spectroscopy is an important tool for characterizing organic conducting salts. 1-7 The sam- ples of such salts are small, for the most part, and opaque. Therefore, infrared properties of organic salts are best obtained by use of a microreflectance infrared (MR-IR) technique, in which a microscope is interfaced with an F T - I R spect rometer . In the salts of b is(e thylene- dithio)tetrathiafulvalene (BEDT-TTF, CloHsS8, or ET), the C-H bonds of ET form numerous C-H. . .anion con- tacts, as shown in Fig. 1 for ~-(ET)2IBr2. The C-H. . . anion interactions of ET salts have been studied with the MR-IR technique. 8-1° The MR-IR spectra of ET salts provide "fingerprints" of their C-H. . .anion contact en- vironments, and have been employed to distinguish be- tween a- and ~-phase salts of ET 9 and to follow the thermal conversion of a-(ET)213 to at-(ET)213 .10 In the present work we obtain room-temperature MR-IR spec- tra of several ~- and K-phase ET salts by employing po- larized and unpolarized light. On the basis of the MR- IR spectra we discuss the electron-molecular vibration coupling u-2° and the lattice softness 21-25 of these salts.

EXPERIMENTAL

Room-temperature MR-IR measurements on ET salts were obtained with a Digilab FTS-40 purged spectrom- eter interfaced (typically 256 scans per spectrum) with a UMA-300-A microscope and a cadmium-mercury-tel- luride detector (resolution at 4 cm-1). The reflectance mode was used because the single crystals are opaque. A Kramers-Kronig transformation was applied to all re- flectance spectra, and the results are reported in arbi-

Received 15 May 1992. * Authors to whom correspondence should be sent.

trary absorbance units. Care is taken to use a crystal that demonstrates predominately reflectance characteristics and little, if any, transmission. The polarized light is reflected off smooth surfaces of the crystals, and striated surfaces were avoided. Measurements made on two ad- jacent smooth surfaces gave the same data, thus obvi- ating any artifacts occurring in the measurement.

STRUCTURE OF D O N O R - M O L E C U L E LAYER

Figure 2(a-c) shows the packing patterns of ET mol- ecules in the donor molecule layers of B- and K-phase salts. In both phases, the 7r-framework of each ET mol- ecule is not perpendicular but inclined to the donor layer. In/3-(ET)2X (X = Is, AuI2, IBr2) the ET molecules cant along the donor-stack direction (i.e., approximately along the a + c direction of Fig. 2a). 26 In K-(ET)2Cu[N(CN)2]X (X = C1, Br, I) 25,27,28 the donor layer defines the ac-plane (Fig. 2b), and in K-(ET)2Cu(NCS)2 the bc-plane (Fig. 2c) 29 The donor molecules of K-(ET)2Cu[N(CN)2]X cant along the a-direction, which is parallel to the Cu[N(CN)2]X- anion chain direction, 23,24,27,28 while those of K-(ET)2Cu- ( N C S ) 2 cant along the c-direction, which is perpendicular to the C u ( N C S ) 2 - chain directionY ,29 As far as donor- molecule layers are concerned, the b- and c-axes of K-(ET)2Cu(NCS) 2 correspond to the c- anda-axes of K-(ET)2Cu[N(CN)2]X, respectively.

In our MR-IR study of ET salts with polarized light, it is necessary to specify the alignment of the light po- larization vector with respect to the crystallographic axes of the salts. For ~-(ET)2X, K-(ET)2Cu[N(CN)2]X, and the K-(ET)2Cu(NCS)2 we define the alignment angle 0 to be zero when the light polarization vector is parallel to the a-, c-, and b-axis directions, respectively. Therefore, the donor molecules cant approximately along the 0 = 90 ° direction in ~-(ET)2X, K-(ET)2Cu[N(CN)2]X, and K-(ET)2Cu(NCS)2.

RESULTS AND DISCUSSION

Vibronic Envelope. As representative examples, Figs. 3a and 3b show the MR-IR spectra of K-(ET)2Cu(NCS)2 and K-(d8-ET)2Cu(NCS)2, respectively. The MR-IR spec- tra of ET salts have two characteristic features. One is a broad band extending from the infrared to ~ 3000 cm-1 region (i.e., plasma region), and is attributed to inter- and intramolecular electronic transitions 2°,3°-~2 associated with the ET molecules. The other feature is an envelope in the 1200-1450 cm -1 region (i.e., vibronic region), which is attributed to the coupling of conduction electrons to totally symmetrical vibrations of ET molecules, u-2° The

1520 Volume 46, Number 10, 1992 0003-7028/92/4610-152052.00/0 APPLIED SPECTROSCOPY © 1992 Society for Applied Spectroscopy

FIG. 1. Arrangement of ET molecules around the IBr 2- anions in ~-(ET)2IBr2 (taken from Ref. 7).

vibronic region is more diagnostic than the plasma region in characterizing ET salts.

The C=C stretching (vc=¢) vibrations of ET, being to- tally symmetric, are Raman-active but infrared-silent. For symmetrical neutral donor molecules with a central C=C bond, the vc=c vibrations are observed at 1500-1550 cm -1. Upon forming salts, the ~cffic modes of the donor molecules interact with the conduction electrons so that the ~¢=c modes become infrared-active and are shifted to lower frequencies, thereby leading to the vibronic en- velopes. Table I lists the v¢ffi¢ frequencies observed for several neutral donor molecules and other center posi- tions of the vibronic envelopes (hereafter referred to as the vibronic frequencies) observed for their salts. With respect to the ~¢=¢ values, the vibronic frequencies are considerably shifted to lower values in the salts. The magnitude of this shift is a measure of the strength of the electron-molecular vibration coupling in a given salt, ~7-~9 although some of this shift may be also due to the change in charge of the donor.

When an ET salt has C-C-H bending vibrations (see below) that overlap with the vibronic envelope, it is dif- ficult to determine the position of the vibronic frequency. In such a case, the latter can be determined from the MR-IR spectrum of the corresponding ds-ET salt, be- cause the C-C-H bending vibrations shift beyond the vibronic envelope (see Fig. 3a and 3b). s7 Table II lists the vibronic frequencies observed for several /~- and ~-phase salts of ds-ET obtained with unpolarized light.

Figure 4 shows the vibronic envelopes of ~-(ds-ET)2I~ observed by use of polarized light with the O = 0 °, 45 °, and 90 ° alignments. The vibronic frequency is lowest, and the vibronic absorption is strongest, when 0 = 90 °. This may be attributed to the fact that, for the ~-phase salt, which consists of donor-molecule stacks, the light polarization vector is approximately perpendicular to the donor-molecular plane when 0 = 90 °. According to the

reflectance s tudy of linear chain semiconductor TEA(TCNQ)2,16 a strong infrared absorption is observed for the totally symmetric vibrational modes of TCNQ when the light polarization vector is parallel to the TCNQ chain direction, i.e., nearly perpendicular to the plane of TCNQ. The latter is understood in terms of the inter- action of the conduction electrons of the TCNQ chains with the totally symmetric intramolecular vibrational modes of TCNQ. 16b However, it is noted that ~-(ds-ET)213 is a two-dimensional metal, while TEA(TCNQ)2 is a one- dimensional semiconductor. TEA(TCNQ)2 shows 16a a negligible reflectance when the light polarization vector is perpendicular to the TCNQ chain but a strong reflec- tance when the light polarization vector is parallel to the chain direction, i.e., the direction of the primary elec- trical conduction. Thus, our finding that the reflectance of ~-(ds-ET)2I 3 is considerably stronger in the O = 90 ° direction than in any other direction implies that the electrical conductivity is stronger along the stacking di- rection than along the interstack direction. This obser- vation is consistent with Fermi surface calculations. 4

The K-phase salts examined in the present study are two-dimensional metals. These salts do not have donor-

T A B L E I. Comparison of the voffie frequencies of neutral donor mole- cules a with the vibronic frequencies of their salts (em-~).

Vibronic Donor ~offic Salt frequency

ET 1511 K-(ET)2Cu(NCS)2 1290 TTF b 1518 (TTF)Br ° 1368 TMTTF d 1538 (TMTTF)Br d 1340 TMTSF d 1539 (TMTSF)2ReO4 ~ 1415 a TMTTF refers to tetramethyltetrathiafulvalene, and TMSF to tetra- methyltetraselenafulvalene.

b Reference 33. c Reference 34. d Reference 35. e Reference 36.

APPLIED SPECTROSCOPY 1521

(a)

0--_.0 °

T c

(c)

(b) ~:-(ET)2Cu(NCS)2

l 4000 3000 2000 I000

Wavenumbers

FIG. 3. MR-IR spectra of (a) K-(ET)2Cu(NCS)2, and (b) K-(ds- ET)2Cu(NCS)2 obtained with unpolarized light.

molecule stacks but have an orthogonal arrangement of donor-molecule stack repeated dimers. From the polar- ized reflectance behavior of /~-(ds-ET)213 and TEA- (TCNQ)2, the optimal vibronic absorption for the K-phase salts might be expected to occur when the light polar- ization vector is perpendicular to the donor-molecule dimers. As presented below, however, this is not the case. Figure 5 shows the vibronic envelopes of K-(ds- ET)2Cu[N(CN)2]Br observed by the use of polarized light. With respect to the case of K-(ds-ET)2I 3, the polarization dependence of the vibronic absorption is much reduced in K-(ds-ET)2Cu[N(CN)2]Br, probably because the latter is a nearly isotropic two-dimensional metal in the plane of the donor-molecule layer. Nevertheless, the vibronic frequency is lowest, and the vibronic absorption is strongest, when 0 = 0 °. These findings are common to all other K-(ds-ET)2Cu[N(CN)2]X and K-(ds-ET)2Cu- (NCS)2 salts examined in our work. The 0 = 0 ° direction of the K-phase salts is not perpendicular to the donor- molecule plane of any of the donor-molecule dimers but is perpendicular to the donor molecule canting direction. Incidently, the 0 = 90 ° direction of/~-(ds-ET)213, for which the vibronic absorption is optimal, coincides with the

TABLE II. Vibronic frequencies of several ~- and K-phase salts of ds- ET obtained from their MR-IR spectra with unpolarized light. ~

Salt Vibronic frequency (cm 1)

j3-(ds-ET)2AuI2 1290 /~-(ds-ET)213 1263 ~-(ds-ET)212Br 1288 ~'-(ds-ET)2IC12 1317 K-(d8-ET)2Cu[N(CN)~]C1 1278 K- (d~-ET)2Cu[N(CN)2]Br 1288 K- (ds-ET)2Cu[N(CN)2]I 1276 K-(ds-ET)2Cu(NCS)2 1283

a Unpolarized light was used because the reflectance of these salts is sensitive to polarized light, both in absorption and frequency position.

(b)

FIG. 2. Packing pat tern of ET molecules in (a) /~-(ET)2X (X = I3, AuI2, IBr2); (b) K-(ET)2Cu[N(CN)2]X (X = C1, Br, I); and (e) K-(ET)~Cu(NCS)z. The definition for the 0 = 0 ° alignment for the po- larized light reflectance measurements is also shown.

1522 Volume 46, Number 10, 1992

14-

12- A b

10- S

0 p 8 -

b a 6 - R C 4-- e

2 -

0 -

~-(d8-ET)213

1275 O = 90 °

1291 O=O o

l I I I l l I 1600 1500 1400 1300 1200 ii00 i000 900

Wavenumbers

FIG. 4. MR-IR spectra of ~-(ET)2I~ obtained by using polarized light with the alignment angles 0 = 0 °, 45 °, and 90 °.

donor-molecule canting direction. Table III summarizes the vibronic frequencies of the/3- and K-phase salts of ds-ET obtained by use of polarized light with 0 = 0 °, 45 °, and 90 ° alignments. Compared with the case of the fl-phase salts, the vibronic absorptions of the K-phase salts have a weak polarization dependency due to the orthogonal arrangement of the donor dimers in their cation layers.

C-C-H B e n d i n g V i b r a t i o n s . In the vibronic envelopes of ET salts, vibrational structure often occurs due to the ethylene group C-C-H bending vibrations (See Fig. 3a). Normally, the C-H stretching (Vc.H) vibrations are more intense than the C-C-H bending vibrations.T However, the bending vibrations occur at the same energy as the vibronic envelope and are resonance-enhanced, 5,s so that they are more intense than the Vc.H stretching vibrations. In the MR-IR spectra of ET salts, the ~C-H vibrations are very weak and obscured by the broad band of the plasma region.

A secondary effect in lowering the intensities of C-H stretching vi- brations is the lower penetration of IR radiation in reflectance mea- surements in the higher-frequency region.

i~24 S=O °

I B- e=45 ° 1z61 --__7

)c-(d8-ET)2Cu [N(CN)2]Br |

9 = 9 o ° 1~95 ~ I 6-

4-

0-

[ [ I [ I 1 I 1600 1500 1400 1300 1200 1100 1000 900

Navenumbers

Fro. 5. MR- IR spectra of K-(ET)2Cu[N(CN)2]Br obtained by using polarized lights with the alignment angles 0 = 0 °, 45 °, and 90 °.

TABLE III. Polarization dependence of the vibronic frequencies ob- served for/~- and ~-phase salts of ds-ET.

Vibronic frequency (cm-0

Salt 0 = 0 ° 0 = 45 ° 0 = 90 °

fl-(ds-ET)~I3 1291 (1.15) a fl-(ds-ET)2IBr2 1302 (0.955) K-(ds-ET)2Cu[N(CN)2]C1 1285 (1.72) K-(ds-ET)2Cu[N(CN)2]Br 1224 (8.33) K-(ds-ET)2Cu[N(CN)2]I 1218 (3.71) K-(ds-ET)2Cu(NCS)2 1214 (7.43)

1288 (1.62) 1275 (14.58) 1281 (1.24) 1275 (23.91) 1309 (1.31) 1322 (1.53) 1261 (5.39) 1295 (5.17) 1255 (3.46) 1308 (2.84) 1238 (5.22) 1297 (5.70)

a Numbers in parentheses are in absorption units.

The C-C-H bending vibrational features may be absent when the C-H.. • anion contact interactions are such that the bending frequencies occur outside the vibronic en- velope or when the bending frequencies are lowered below the vibronic region by isotope substitution, as in ds-ET salts. 37 In such a case, there is no resonance-enhancement of the C-C-H bending vibrational modes, and thus their intensities are weak.

Because the C-C-H bending vibrations in ET salts occur under the constraint of the C-H.--anion and the C-H. • - donor contacts 25 (See Fig. 1), they may reflect the lattice softness. 21-25 When a C-H-. .anion interaction is less strained (i.e., softer), its C-C-H bending frequency is expected to be lower. Since each ET molecule has eight C-H bonds, there are a number of C-C-H bending fre- quencies to consider. Table IV lists the C-C-H bending frequencies of fl- and K-phase salts of ET observed from their vibronic envelopes by use of unpolarized light at room temperature. Of several C-C-H bending frequen- cies observed for a given ET salt, the most relevant with respect to the lattice softness would be the highest-fre- quency vibration because it represents the "most- strained" C-C-H bending mode. The absence of super- conductivity in fl-(ET)212BP s and K-(ET)2Cu[N(CN)2]P 4 is attributed to the random potentials associated with their structural disorder. For the ET salts without such random potentials, Table IV shows that the highest C-C-H bending frequency is lower than ~ 1320 cm -1 for the superconductors, but higher than ~1320 cm -1 for the nonsuperconductors. Thus, the superconductors ap- pear to have a softer lattice than do the nonsupercon- ductors.

TABLE IV. C-C-H bending frequencies of some hydrogenic ET salts observed from their MR-IR spectra by employing unpolarized light and their superconductivity behavior.

C-C-H bending Salt frequency (cm -1) Superconductivity

/~'-(ET)2IC12 1343, 1269 No ~-(ET)2AuC12 1338, 1265 No fl-(ET)2AuBr 2 1324, 1271 No /3-(ET)212Br 1315, 1255 No

K-(ET)2Cu[N(CN)2]I Not available a No ~-(ET)2IBr 2 1317, 1260 Yes (To less

than 4 K) K-(ET)2Cu(NCS) 2 1316, 1267, 1250 Yes

~-(ET)2Cu[N(CN)2]C1 1313, 1268, 1250 Yes (under 0.3 kbar)

fl-(ET)2AuI2 1310, 1265 Yes fl-(ET)2I 3 1300, 1260 Yes K-(ET)2Cu[N(CN)2]Br 1297, 1259, 1247 Yes K-(ET)2Cu2(CN) 3 1297, 1258, 1229 Yes (1.5 kbar)

a Due to poor crystal quality.

APPLIED SPECTROSCOPY 1523

CONCLUDING REMARKS

The/3- and K-phase ET salts examined in our study are all two-dimensional metals and exhibit strong vi- bronic absorption under polarized light. The polariza- tion-dependency of the vibronic absorption is much stronger in the ~-phase salts than in the d-phase salts. For the/~-phase salts, the optimum vibronic absorption is obtained when the polarized light vector is parallel to their donor-molecule stacking direction. Among the ET salts which do not possess any structural disorder, the highest C-C-H bending frequency is lower than ~1320 cm -1 for the superconductors but is higher than ~ 1320 cm-1 for the nonsuperconductors. The mid-infrared elec- tronic band absorption of the two-dimensional metal ET salts falls in the vibronic region so that, to fully under- stand their vibronic absorption, one needs to examine the polarization-dependency of their electronic density of states. Valuable information about the extent of the electron-molecular vibration coupling and the lattice softness of/3- and d-phase ET salts can be gained from their MR-IR spectra in the vibronic region.

Note added in proof:

The CH2 stretching vibration in ET salts has now been observed, 38 above the strong background in the 3000- cm-' region, with the use of FT-IR absorption techniques and a fast and sensitive InSb detector. Up to 10,000 cumulative interferograms were necessary to obtain a reasonable signal-to-noise ratio.

ACKNOWLEDGMENTS

Work at Argonne National Laboratory and North Carolina State University is supported by the Office of Basic Energy Sciences, Division of Materials Sciences, U.S. Department of Energy, under Contract W-31-109-ENG-38 and Grant DE-FG05-86ER45259, respectively. J.R.F. wishes to thank Dr. K Krishnan and Dr. Richard Crocombe of Bio- Rad Digilab Division for their encouragement of this work. The authors also wish to thank Dr. U. Geiser for his help in determining the crys- talline axes for some of the samples, and Dr. A. M. Kini and Dr. Jack M. Williams for their comments.

1. J. R. Ferraro and J. M. Williams, Introduction to Synthetic Elec- trical Conductors (Academic Press, Orlando, Florida, 1987), pp. 1-337 and references therein.

2. J. R. Ferraro and J. M. Williams, Appl. Spectrosc. 44, 200 (1990); idem, Mol. Cryst. Liq. Cryst. 181, 253 (1990).

3. J. R. Ferraro and J. M. Williams, in Practical Fourier Transform Infrared Spectroscopy: Industrial and Laboratory Chemical Analysis, J. R. Ferraro and K. Krishnan, Eds. (Academic Press, San Diego, California, 1990), pp. 41-101.

4. (a) J. M. Williams, J. R. Ferraro, R. J. Thorn, K. D. Carlson, U. Geiser, H. H. Wang, A. M. Kini, and M.-H. Whangbo, Organic Superconductors (Including Fullerenes): Synthesis, Structure, Properties and Theory (Prentice-Hall, Englewood Cliffs, New Jer- sey, 1992), pp. 210-228 and references therein; (b) ibid., pp. 258- 265.

5. K. Kornelsen, J. E. Eldridge, C. C. Homes, H. H. Wang, and J. M. Williams, Solid State Commun. 72, 475 (1989).

6. (a) K. Kornelsen, J. E. Eldridge, C. C. Homes, H. H. Wang, and J. M. Williams, Solid State Commun. 74, 501 (1990); (b) A. Ugawa, G. Ojima, K. Yukushi, and H. Kuroda, Phys. Rev. B, 38, 5122 (1988).

7. P. C. W. Leung, T. J. Emge, A. J. Schultz, M. A. Beno, K. D. Carlson, H. H. Wang, M. A. Firestone, and J. M. Williams, Solid State Commun. 57, 93 (1986).

8. J. R. Ferraro, H. H. Wang, J. Ryan, and J. M. Williams, AppL Spectrosc. 41, 1377 (1987).

9. J. R. Ferraro, H. H. Wang, U. Geiser, A. M. Kini, M. A. Beno, and J. M. Williams, Solid State Commun. 68, 917 (1988).

10. H. H. Wang, J. R. Ferraro, K. D. Carlson, L. K. Montgomery, U. Geiser, J. R. Whitworth, J. A. Schlueter, S. Hill, M.-H. Whangbo, M. Evain, and J. A. Novoa, Inorg. Chem. 28, 2267 (1989).

11. M. G. Kaplunov, E. B. Yagubskii, L. P. Rosenberg, and Y. G. Borodko, Phys. Stat. Sol. (a), 89, 509 (1985).

12. M. G. Kaplunov, E. B. Yagubskii, M. Z. Aldoshima, and Y. G. Borodko, Phys. Stat. Sol. 104, 833 (1987).

13. M. J. Rice, C. B. Duke, and N. O. Lipari, Solid State Commun. 17, 1089 (1975).

14. M. J. Rice, Phys. Rev. Lett. 37, 36 (1976). 15. C. B. Duke, Ann. N. Y. Acad. Sci. 313, 166 (1978). 16. (a) A. Brau, P. Br/iesch, J. P. Farges, W. Hinz, and D. Kuse, Phys.

Stat. Sol. (b), 62, 615 (1974); (b) M. J. Rice, L. Pietronero, and P. Br/iesch, Solid State Commun. 21, 757 (1977).

17. R. Bozio, M. Meneghetti, and C. Pecile, J. Chem. Phys. 76, 5785 (1982).

18. R. Zamboni, D. Schweitzer, H. J. Keller, and C. Taliani, Z. Na- turforseh 44a, 295 (1989).

19. R. Zamboni, D. Schweitzer, and H. J. Keller, in Lower-Dimensional Systems and Molecular Electronics, Proc. of NATO ASI, R. M. Metzger, P. Day, and G. Papavassilliou, Eds. (Plenum Press, New York, 1989).

20. M. Meneghetti, R. Bozio, and C. Pecile, J. de Physique 47, 1377 (1986).

21. M.-H. Whangbo, J. M. Williams, A. J. Schultz, T. J. Emge, and M. A. Beno, J. Am. Chem. Soc. 109, 90 (1987).

22. M.-H. Whangbo, D. Jung, J. Ren, M. Evaln, J. J. Novoa, F. Mota, S. Alvarez, J. M. Williams, M. A. Beno, A. M. Kini, H. H. Wang, and J. R. Ferraro, in The Physics and Chemistry of Organic Su- perconductors, G. Saito and S. S. Kagoshina, Eds. (Springer-Vet- lag, 1990), p. 262.

23. M.-H. Whangbo, J. J. Novoa, D. Jung, J. M. Williams, A. M. Kini, H. H. Wang, U. Geiser, M. A. Beno, and K. D. Carlson, in Organic Superconductivity, V. Z. Kresin and W. A. Little, Eds. (Plenum, New York, 1990), p. 243.

24. U. Geiser, A. J. Sehultz, H. H. Wang, D. M. Watkins, D. L. Stupka, J. M. Williams, J. E. Schirher, D. L. Overmyer, D. Jung, J. J. Novoa, and M.-H. Whangbo, Physica C, 174, 475 (1991).

25. J. M. Williams, A. J. Schultz, U. Geiser, K. D. Carlson, A. M. Kini, H. H. Wang, W. K. Kwok, M.-H. Whangbo, and J. E. Schirber, Science 252, 1501 (1991).

26. J. M. Williams, H. H. Wang, T. J. Emge, U. Geiser, M. A. Beno, P. C. W. Leung, K. D. Carlson, R. J. Thorn, A. J. Schultz, and M.-H. Whangbo, in Progress in Inorganic Chemistry, S. J. Lippard, Ed. (J. Wiley & Sons, New York, 1987), pp. 51-258.

27 J.M. Williams, A. M. Kini, H. H. Wang, K. D. Carlson, U. Geiser, L. K. Montgomery, G. J. Pyrka, D. M. Watkins, J. M. Kommers, S. J. Boryschuk, A. V. S. Crouch, W. K. Kwok, J. E. Schirber, D. L. Overmyer, D. Jung, and M.-H. Whangbo, Inorg. Chem. 29, 3272 (1990).

28. A. M. Kini, U. Geiser, H. H. Wang, K. D. Carlson, J. M. Williams, W. K. Kwok, K. G. Vandervoort, J. E. Thompson, D. Stupka, D. Jung, and M.-H. Whangbo, Inorg. Chem. 29, 2555 (1990).

29. K. D. Carlson, U. Geiser, A. M. Kini, H. H. Wang, L. K. Mont- gomery, W. K. Kwok, M. A. Beno, J. M. Williams, C. S. Cariss, G. W. Crabtree, M.-H. Whangbo, and M. Evain, Inorg. Chem. 27, 965, 2904 (1988).

30. C. S. Jacobsen, J. M. Williams, and H. H. Wang, Solid State Com- mun. 54, 937 (1985).

31. A. Ugawa, K. Yakushi, and H. Kuroda, Mol. Cryst. Liq. Cryst. 181, 269 (1990).

32. M. E. Kozlov, K. I. Pokhodnia, and A. A. Yurehenko, Spectrochim. Acta 43, 323 (1987).

33. R. Bozio, A. Girlando, and D. Pecile, Chem. Phys. Lett. 52, 503 (1977).

34. R. Bozio, I. Zanon, A. Girlando, and C. Pecile, J. Chem. Phys. 71, 2282 (1979).

35. M. Meneghetti, R. Bozio, I. Zanon, C. Pecile, E. Ricotta, and M. Zanetti, J. Chem. Phys. 80, 6210 (1984).

36. R. Bozio, C. Pecile, K. Beckgaard, F. Wudl, and D. Nakwajek, Solid State Commun. 41, 905 (1982).

37. H. H. Wang, P. E. Reed, and J. M. Williams, Syn. Metals 14, 165 (1986).

38. S. Giirtner, D. Schweitzer, and H. J. Keller, Syn. Metals 44, 227 (1991).

1524 Volume 46, Number 10, 1992