Paramagnetic and semiconducting 1:1 salts of 1,1′-disubstituted ferrocenes and [Ni(mnt)2]-. Synthesis, structure, and physical properties

Article abstract of DOI:10.1021/om9603981

The ferrocene-based electron donors 1,1′-bis[2-(4-(methylthio)phenyl)-CE)-ethenyl]ferrocene (2), 1,1′-bis[2-(4-methoxyphenyl)-CE)-ethenyl]ferrocene (3), 1,1′-bis[(1,3-dithiolo[4,5-b][1,3]-dithiol-2-ylidene)methyl]ferrocene (4), and 1,1′-bis[(1,3-benzodithiol-2-ylidene)methyl]ferrocene (5) were found to react with ferrocenium bis(maleonitriledithiolato)nickelate (1-) ([FeCp₂]⁺[Ni(mnt)₂]^-, 6) to afford the corresponding 1:1 paramagnetic salts 7-10, containing 1,1′-disubstituted ferrocenium derivatives. SQUID magnetic susceptibility measurements of these new compounds showed a behavior dominated by antiferromagnetic interactions within pairs of [Ni(mnt)₂]^- ions. Pressed pellets of compounds 7 ([2][Ni(mnt)₂]) and 9 ([4]-[Ni(mnt)₂]) are semiconducting, with a relatively large conductivity activation energy (0.85 and 1.13 eV). Crystals of 7 reveal the monodimensional nature of the compound. Each separate stack of [Ni(mnt)₂]^- ions is flanked by two ferrocenium stacks. Ni-Ni distances alternate between 3.67 and 3.99 ?.

Full text of DOI:10.1021/om9603981

5342
Organometallics 1996, 15, 5342-5346
P a r a m a gn etic a n d Sem icon d u ctin g 1:1 Sa lts of
1,1′-Disu bstitu ted F er r ocen es a n d [Ni(m n t)2]-. Syn th esis,
Str u ctu r e, a n d P h ysica l P r op er ties
Markus Hobi, Stefan Zu¨rcher, Volker Gramlich, Urs Burckhardt,
Christian Mensing, Michael Spahr, and Antonio Togni*
Laboratory of Inorganic Chemistry, ETH-Zentrum, Swiss Federal Institute of Technology,
CH-8092 Zu¨rich, Switzerland
Received May 22, 1996X
The ferrocene-based electron donors 1,1′-bis[2-(4-(methylthio)phenyl)-(E)-ethenyl]ferrocene
(2), 1,1′-bis[2-(4-methoxyphenyl)-(E)-ethenyl]ferrocene (3), 1,1′-bis[(1,3-dithiolo[4,5-b][1,3]-
dithiol-2-ylidene)methyl]ferrocene (4), and 1,1′-bis[(1,3-benzodithiol-2-ylidene)methyl]fer-
rocene (5) were found to react with ferrocenium bis(maleonitriledithiolato)nickelate (1-)
([FeCp2]+[Ni(mnt)2]-, 6) to afford the corresponding 1:1 paramagnetic salts 7-10, containing
1,1′-disubstituted ferrocenium derivatives. SQUID magnetic susceptibility measurements
of these new compounds showed a behavior dominated by antiferromagnetic interactions
within pairs of [Ni(mnt)2]- ions. Pressed pellets of compounds 7 ([2][Ni(mnt)2]) and 9 ([4]-
[Ni(mnt)2]) are semiconducting, with a relatively large conductivity activation energy (0.85
and 1.13 eV). Crystals of 7 reveal the monodimensional nature of the compound. Each
separate stack of [Ni(mnt)2]- ions is flanked by two ferrocenium stacks. Ni-Ni distances
alternate between 3.67 and 3.99 Å.
In tr od u ction
two reversibly accessible oxidation states. The reduced
form present in the CT complex is usually a radical
anion. Such planar molecules favor the formation of
segregated or alternating stacks of donors and acceptors
in the solid state, with corresponding consequences on
the physical properties of the CT complexes.
We note that there are only few reports in the
literature about the use of organometallics for the
preparation of conducting CT complexes.6 An important
exception is represented by the (per)alkylated metal-
locenes and related bis(arene) complexes of various
transition metals. These compounds have been inten-
sively studied in connection with the magnetic proper-
ties of their CT complexes, mainly by Miller and
co-workers.7 These studies led to the discovery of the
first organometallic compound displaying bulk ferro-
magnetic behavior.
Since the discovery in 1973 of the high electrical
conductivity and metallic behavior of the one-dimen-
sional charge-transfer (CT) complex formed by the donor
tetrathiafulvalene (TTF) and the acceptor tetracyano-
p-quinodimethane (TCNQ),1 research in the field has
flourished. As compiled in a number of excellent
reviews,2 synthetic efforts toward electron donors and
acceptors have been concentrated on two main struc-
tural types:3 those containing the C2E4 fragment (E )
S, Se), with the parent TTF donor and the ligand dmit
as the most prominent prototypes, and the quinoid
structural types dominated by TCNQ and dicyano-
quinodiimine (DCNQI).4 Beside these systems, inor-
ganic acceptors such as mnt metal complexes5 have been
used.2d mnt systems (mnt ) maleonitriledithiolato)
often contain a late transition metal and exhibit at least
(5) See, for example: (a) Magnetic Molecular Materials; Gatteschi,
D., Kahn, O., Miller, J . S., Palacio F., Eds.; NATO ASI Series, Series
E: Applied Sciences, Vol. 198; Kluwer: Dordrecht, The Netherlands,
1991. (b) Davison, A.; Edelstein, N.; Holm, R.; Maki, A. H. J . Am. Chem.
Soc. 1963, 85, 2029-2030. (c) Davison, A.; Edelstein, N.; Holm, R. H.;
Maki, A. H. Inorg. Chem. 1963, 2, 1227-1232. For the preparation of
mnt complexes, see: (d) Davison, A.; Holm, R. H. Inorg. Synth. 1967,
10, 8-26. (e) Hoyer, E.; Olk, R.-M.; Olk, B.; Dietzsch, W.; Kirmse, R.
Coord. Chem. Rev. 1992, 117, 99-131.
(6) For a review on ferrocene derivatives, see: (a) Togni, A. In
Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials
Science; Togni, A., Hayashi, T., Eds.; VCH: Weinheim, Germany, 1995;
pp 433-469. For scattered reports, see, for example: (b) Mu¨ller-
Westerhoff, U. T.; Eilbracht, P. J . Am. Chem. Soc. 1972, 94, 9272-
9274. (c) Goldberg, S. Z.; Spivack, B.; Stanley, G.; Eisenberg, R.;
Braitsch, D. M.; Miller, J . S.; Abkowitz, M. J . Am. Chem. Soc. 1977,
99, 110-117. (d) Ueno, Y.; Sano, H.; Okawara, M. J . Chem. Soc., Chem.
Commun. 1980, 28-30. (e) Bolinger, C. M.; Darkwa, J .; Gammie, G.;
Gammon, S. D.; Lyding, J . W.; Rauchfuss, T. B.; Wilson, S. R.
Organometallics 1986, 5, 2386-2388.
X Abstract published in Advance ACS Abstracts, November 15, 1996.
(1) (a) Ferraris, J .; Cowan, D. O.; Walatka, V. V.; Peristein, J . H. J .
Am. Chem. Soc. 1973, 95, 948-950. (b) Ferraris, J . P.; Poehler, T. O.;
Bloch, A. N.; Cowan, D. O. Tetrahedron Lett. 1973, 2553-2556.
(2) (a) Graja, A. Low-dimensional Organic Conductors; World
Scientific: Singapore, 1992. (b) Bryce, M. R. Chem. Soc. Rev. 1991,
20, 355-390. (c) Cassoux, P.; Valade, L. In Inorganic Materials Bruce,
D. W., O’Hare, D., Eds.; Wiley: Chichester, U.K., 1992; pp 2-58. (d)
Cassoux, P.; Valade, L.; Kobayashi, H.; Kobayashi, V.; Clark, R. A.;
Underhill, A. E. Coord. Chem. Rev. 1991, 110, 115-160. (e) Williams,
J . M.; Ferraro, J . R.; Thorn, R. J .; Carlson, K. D.; Geiser, U.; Wang, H.
H.; Kini, A. M.; Whangbo, M.-H. Organic Superconductors. Synthesis,
Structure, Properties and Theory; Prentice Hall: Englewood Cliffs, NJ ,
1992. (f) Williams, J . M.; Wang, H. H.; Emge, T. J .; Geiser, U.; Beno,
M. A.; Leung, P. C. W.; Carlson, K. D.; Thorn, R. J .; Schultz, A. J .
Prog. Inorg. Chem. 1987, 35, 51-218.
(3) For other classes of compounds, see for example: (a) Tiecke, B.;
Wegmann, A.; Fischer, W.; Hilti, B.; Mayer, C. W.; Pfeiffer, J . Thin
Solid Films 1989, 179, 233-238. (b) Gu¨nther, E.; Hu¨nig, S.; von Schu¨tz,
J .-U.; Langohr, U.; Rieder, H.; So¨derholm, S.; Werner, H.-P.; Peters,
K.; von Schnering, H. G.; Lindner, H. J . Chem. Ber. 1992, 125, 1919-
1926 and references cited therein.
(4) See, for example: (a) Aumu¨ller, A.; Erk, P.; Klebe, G.; Hu¨nig,
S.; von Schu¨tz, J .-U.; Werner, H.-P. Angew. Chem. 1986, 98, 759-761;
Angew. Chem., Int. Ed. Engl. 1986, 25, 740-741. (b) Hu¨nig, S.; Erk,
P. Adv. Mater. 1991, 3, 225-236.
(7) For reviews, see: (a) Miller, J . S.; Epstein, A. J . Angew. Chem.,
Int. Ed. Engl. 1994, 33, 385-416. (b) Miller, J . S.; Epstein, A. J .; Reiff,
W. M. Chem. Rev. 1988, 88, 201-220. (c) Miller, J . S.; Epstein, A. J .;
Reiff, W. M. Acc. Chem. Res. 1988, 21, 114-120. (d) Miller, J . S.;
Epstein, A. J .; Reiff, W. M. Science 1988, 240, 40-47. (e) Miller, J . S.;
Epstein, A. J . In Research Frontiers in Magnetochemistry; O’Connor,
C. J ., Ed.; World Scientific: Singapore, 1993; pp 283-302.
S0276-7333(96)00398-6 CCC: $12.00 © 1996 American Chemical Society

Paramagnetic and Semiconducting Fe and Ni Salts
Organometallics, Vol. 15, No. 25, 1996 5343
Ch a r t 1
Sch em e 1
We previously reported the preparation and charac-
terization of conducting and antiferromagnetic CT
complexes containing 1,1′-disubstituted ferrocenes of
types 1 and 2 as donors, respectively (see Chart 1).8 We
argued that the physical properties of the corresponding
CT complexes with TCNQ may depend on the relative
orientation of the substituents on the ferrocene core. An
eclipsed, intramolecularly stacked conformation, e.g. in
1, is thought to favor the formation of separate stacks
of donors and acceptors, thus leading to a relatively high
conductivity. On the other hand, a perfectly anti-
periplanar orientation of the side chains was observed
for the insulating derivative [2][TCNQ]2, displaying a
DADA-type arrangement in the solid state.8
In continuation of these studies, we report herein the
preparation and characterization of CT complexes con-
taining partially the same ferrocene donors and the
inorganic acceptor [Ni(mnt)2]-. The main purpose was
to enforce a 1:1 ratio of donor and acceptor in the CT
complex, since this stoichiometry was not possible to
achieve with TCNQ. Thus, a completely different
structural principle and, hence, different physical prop-
erties were anticipated.
Ta ble 1. Exp er im en ta l Da ta for th e X-r a y
Diffr a ction Stu d y of 7
formula/mol wt
crystal dimens, mm
cryst syst
space group
a (Å)
C36H26FeN4NiS6/821.58
0.1 0.1 0.5
monoclinic
P21/n
7.572(5)
b (Å)
c (Å)
28.647(10)
16.374(11)
93.10(5)
â (deg)
V (Å3)
3547(4)
Z
4
F(calcd) (g cm-3
)
1.539
1.326
µ (mm-1
)
diffractometer
radiation
measured rflns
Syntex P21
Mo KR, λ ) 0.710 73 Å
0 e h e 6, 0 e k e 24,
-13 e l e 13
3.0-40.0
2θ range (deg)
scan type
scan width (deg)
bkgd measurement
max scan speed
ω
1.10
0.30 scan time
1.0-14.0
Resu lts a n d Discu ssion
(deg min-1 in ω)
no. of indep data coll
no. of obsd rflns (no)
no. of params refined (nv)
quantity minimized
weighting scheme
Ra
2234
Syn th esis. The ferrocene-based electron donors il-
lustrated in Chart 1 are easily accessible as powder or
crystalline, air-stable materials in one step from fer-
rocene dicarbaldehyde9 and the appropriate Wittig-
Horner reagents,10 as previously reported from our
laboratory.8
1139 ( Fo > 4.0σ( F 2))
2
233
Σw(Fo - Fc)2
w-1 ) σ2(F) + 0.0544F2
0.0915
0.1046
0.52
b
Rw
GOFc
For the synthesis of the CT complexes with the
a
b
R ) ∑ ( Fo - (1/k) Fc )/∑ Fo
.
Rw ) ∑w( Fo - (1/k) Fc )2/
monoanionic, formally Ni(III) acceptor [Ni(mnt)2]-
a
2
1/2
[∑w Fo
] . .
c GOF ) [Σw( Fo - (1/k) Fc )2/(no - nv)]1/2
novel strategy needed to be developed. Since the 1,1′-
disubstituted ferrocenes which were used in this work
are more readily oxidized than ferrocene itself, [FeCp2]+-
[Ni(mnt)2]- (6) seemed to be a key intermediate fulfilling
two functions: (1) delivering the anion [Ni(mnt)2]- and
(2) oxidizing the desired donor.11 The ferrocenium salt
6 was obtained in moderate yield on adding 2 equiv of
[FeCp2]BF4 to the dianionic Ni(II) form [NR4]2[Ni(mnt)2]
(R ) Et, Bu). Subsequently, when hot solutions of a
ferrocene donor and 6 were combined, oxidation at the
Fe center of the donor took place, affording the desired
CT complex and ferrocene, as illustrated in Scheme 1.
The two products of this redox process were easily
separated by virtue of their drastically different solu-
bilities. By this procedure the dark, microcrystalline
CT complexes 7-10 were obtained in moderate to good
yields.
Solid -Sta te Str u ctu r e of th e CT Com p lex 7. Rele-
vant parameters concerning the X-ray diffraction study
are collected in Table 1. A view of the asymmetric unit
of 7 with its atom-numbering scheme is depicted in
Figure 1. Unfortunately, for all CT complexes of this
(8) Togni, A.; Hobi, M.; Rihs, G.; Rist, G.; Albinati, A.; Zanello, P.;
Zech, D.; Keller, H. Organometallics 1994, 13, 1224-1234.
(9) (a) Mu¨ller-Westerhoff, U. T.; Yang, Z.; Ingram, G. J . Organomet.
Chem. 1993, 463, 163-167. (b) Balavoine, G. G. A.; Doisneau, G.;
Fillebeen-Khan, T. J . Organomet. Chem. 1991, 412, 381-382.
(10) (a) Nakayama, J . J . Chem. Soc., Chem. Commun. 1975, 38-
39. (b) Degani, I.; Fochi, R. J . Chem. Soc., Chem. Commun. 1976, 471-
472. (c) Parg, R. P.; Kilburn, J . D.; Ryan, T. G. Synthesis 1994, 195-
198. (d) Akiba, K.; Ishikawa, K.; Inamoto, N. Bull. Chem. Soc. J pn.
1978, 51, 2674-2682.
(11) The redox properties of ferrocenes of the type 1-5 have been
reported previously from our laboratory.8 Their formal electrode
potentials for the one-electron oxidation step are lower by 0.1-0.2 V
than for ferrocene.

5344 Organometallics, Vol. 15, No. 25, 1996
Hobi et al.
F igu r e 1. ORTEP view and atom-numbering scheme of
the asymmetric unit of 7.
F igu r e 3. Side view of the [Ni(mnt)2]- stacks of 7, showing
the formation of [Ni(mnt)2]- dimers.
periplanar as in the corresponding TCNQ CT complex.8
A reason for this relative arrangement could be the
similar elongation of the anion and the cation in its
intramolecularly stacked, approximately Cs symmetric
conformation. However, alterations of this ideal geom-
etry are almost insignificant, in view of the high
standard deviations observed on the structural param-
eters. Thus, the planes of two phenyl rings span an
angle of 2.4° and are also almost coplanar with the
respective Ni(mnt)2 unit (angles of 2.8 and 3.9°). Fur-
thermore, the Ph groups form angles of 6.0 and 9.7° with
the plane of the respective Cp ring. The last two
fragments deviate from coplanarity by 6.9°, thus show-
ing a slight opening of the ferrocene unit. This distor-
tion is also reflected by the gradually increasing dis-
tances within couples of corresponding atoms of the two
substituents, “moving away” from the ferrocene core,
indicating a slight repulsive interaction between the
substituents (this is reflected by the distances between,
e.g., C6 and C6′ of 3.71 Å and S1 and S1′ of 3.84 Å).
As depicted in Figure 3, in the [Ni(mnt)2]- stacks the
Ni-Ni distances alternate between long (3.99 Å) and
short (3.67 Å), this being an expression of dimer
formation, as observed previously.12a,14 Within a dimer
the two [Ni(mnt)2]- units are nearly eclipsed and the
short Ni-Ni distance is slightly larger than the one
reported by Miller et al. for the corresponding [Fe(Me5-
Cp)2] derivative.14 The two nitrogen atoms of each mnt
unit show relatively short distances (of 4.18 and 4.62
F igu r e 2. Perspective view of 7 down the a crystal-
lographic axis, showing the separate stacks of [Ni(mnt)2]-
ions, each flanked by two stacks of 1,1′-disubstituted
ferrocenium cations.
study, due to their very low solubility in a variety of
organic solvents, it proved very difficult to grow suitable
crystals. Sufficiently large black (flexible) needles of
compound 7 could be obtained directly from the reaction
mixture. However, they were of poor quality, thus
explaining the relatively low accuracy of the structural
parameters.
The most prominent general structural features of
compound 7 are the completely segregated stacks
of [Ni(mnt)2]- anions,12 each flanked by two columns
of disubstituted ferrocenium cations, as revealed by
the projection along the crystallographic a axis in Fig-
ure 2. This is the shortest crystallographic axis but
the longest macroscopic crystal dimension. Besides
[Ni(mnt)2]- salts of diamagnetic cations (e.g. NEt4+ 13),
to our knowledge 7 is the first bimetallic [Ni(mnt)2]-
derivative displaying this kind of three-dimensional
structure. Interestingly, the two substituents of the
ferrocenium unit are almost parallel and not anti-
(12) For a review on stacked metal complexes, see: (a) Ibers, J . A.;
Pace, L. J .; Martinsen, J .; Hoffman, B. M. Struct. Bonding 1982, 50,
1-55. For more recent reports on paramagnetic salts containing
(Ni(mnt)2)-, see: (b) Reith, W.; Polborn, K.; Amberger, E. Angew.
Chem., Int. Ed. Engl. 1988, 27, 699-700. (c) Colpas, G. J .; Kumar,
M.; Day, R. O.; Maroney, M. J . Inorg. Chem. 1990, 29, 4779-4788. (d)
Mahadevan, C. J . Cryst. Spectrosc. 1986, 16, 159-167. (e) Brunn, K.;
Endres, H.; Weiss, J . Z. Naturforsch. 1987, 42B, 1222-1226.
(13) (a) Kobayashi, A.; Yukiyoshi, S. Bull. Chem. Soc. J pn. 1977,
50, 2650-2656. See also: (b) Weiher, J . F.; Melby, L. R.; Benson, R.
E. J . Am. Chem. Soc. 1964, 86, 4329-4333. (c) Davison, A.; Edelstein,
N.; Holm, R. H.; Maki, A. H. Inorg. Chem. 1963, 2, 1227-1232.
(14) Miller, J . S.; Calabrese, J . C.; Epstein, A. J . Inorg. Chem. 1989,
128, 4230-4238.

Paramagnetic and Semiconducting Fe and Ni Salts
Organometallics, Vol. 15, No. 25, 1996 5345
F igu r e 5. Temperature dependence of the electrical
resistivity of 7 (filled circles), 9 (open circles), and 9 after
heating to 480 K (squares).
only one not containing sulfur. Again, assuming a
three-dimensional structure similar to that for 7, this
would indicate an unexpected strong influence of the
peripheral heteroatoms on the overall magnetic behav-
ior. Compound 10 shows a somewhat more complex be-
havior, with a broad maximum of the reciprocal sus-
ceptibility at ca. 60 K, probably due to a phase transi-
tion, this reasonably excluding the presence of para-
magnetic impurities. However, since structural data for
this compound are not available, a more detailed
interpretation of this observation is not possible.
F igu r e 4. SQUID susceptibility data for compounds 7-10
measured in an external magnetic field of 1000 G.
Å) to the closest Fe atom, located only 0.82 Å out of the
[Ni(mnt)2]- plane (however, the shortest Ni-Fe distance
is 9.16 Å).
Con du ctivity Measu r em en ts. Resistivity measure-
ments on polycrystalline pressed pellets of 7 and 9 were
carried out by the two-probe method in the temperature
range between 300 and 410 K. Both samples show an
exponential, reversible temperature dependence of the
resistivity, characteristic of a semiconductor (see Figure
5), whereby the room-temperature conductivity of both
In fr a r ed Sp ectr oscop y. The infrared spectrum of
the acceptor [Ni(mnt)2][NEt4], in which [Ni(mnt)2] dimers
are present, displays a CN stretching vibration at 2208
cm-1
. The corresponding band for the ferrocenium
precursor 6 and for the CT complexes 7-10 is observed
in the range between 2204 and 2207 cm-1. The only
slight lowering of ν(CN) for the new compounds indi-
cates that the state of charge and electronic environ-
ment of the Ni(mnt)2 unit are very similar to those of
the starting material.
compounds is smaller than 10-7 S cm-1
. From the
resistivity data we obtain values of the conductivity
activation energy Ea for 7 and 9 of 0.85 and 1.13 eV,
respectively. These values are higher than the energy
gap reported for single crystals of the parent compound
[NR4][Ni(mnt)2].13 However, it is clear that the [Ni-
(mnt)2] stacks are responsible for the observed conduc-
tivity15 and that the contribution of the ferrocene stacks
is negligible, due to localization of the unpaired electron
on iron, as shown previously for conducting CT com-
plexes containing TCNQ and 1 and 5, respectively.8
Ma gn etic Mea su r em en ts. Magnetic susceptibility
measurements were carried out for the Ni(mnt)2 CT
complexes 7-10 using a SQUID susceptometer. Plots
of the reciprocal susceptibility ømol and of ømolT vs T are
shown in Figure 4. None of the compounds show a clean
Curie-Weiss behavior. The almost linear decrease of
Interestingly, the electrical behavior of derivative 9
changed when the pellet was heated to 480 K (it is
important to note that up to this temperature there was
no indication of annealing or sintering of the pellet).
Thus, the room-temperature conductivity increased by
ca. 2 orders of magnitude, and the material showed
again a semiconductive behavior, now with a lower
activation energy of 0.49 eV. This observation is indica-
tive of an irreversible phase transition occurring in the
solid state at elevated temperature.
ømolT on cooling from room temperature for 7-9 sug-
gests an antiferromagnetic interaction. We ascribe this
behavior to the progressive thermal depopulation of the
triplet excited state along the Ni(mnt)2 chains, due to
the formation of S ) 0 dimers, this behavior being most
pronounced for derivative 9, assuming a solid-state
structure similar to that found for 7. However, this
interpretation does not take into account the ferroce-
nium spin, as well as possible interstack interactions.
The latter possibly dominate the magnetic behavior at
relatively high temperatures. This is particularly so for
compound 8, which shows a reciprocal susceptibility
lower than for all other derivatives, over the entire
temperature range. We note that this compound is the
(15) For a detailed theoretical account on structure and mecha-
nism of conductivity of metal bis(dithiolene) derivatives, see: Alvarez,
S.; Vicente, R.; Hoffmann, R. J . Am. Chem. Soc. 1985, 107, 6253-
6277.

5346 Organometallics, Vol. 15, No. 25, 1996
Hobi et al.
Com p ou n d 8 was obtained in an analogous manner from
50 mg (0.111 mmol) of 3 in 10 mL of 1,2-dichloroethane and
58 mg (0.111 mmol) of [FeCp2][Ni(mnt)2] (6) in 30 mL of 1,2-
dichloroethane and 10 mL of acetone. Yield: 31 mg (35%).
Anal. Calcd for C36H26N4O2S4FeNi: C, 54.77; H, 3.32; N, 7.10;
O, 4.05. Found: C, 54.73; H, 3.37; N, 7.14; O, 4.01. IR (KBr):
3105, 2928, 2835, 2207, 1619, 1595, 1574, 1510, 1472, 1438,
1422, 1383, 1323, 1300, 1288, 1249, 1192, 1170, 1110, 1032,
Con clu sion s
We have shown that 1:1 [Ni(mnt)2]- salts containing
1,1′-disubstituted ferrocenium cations are easily ob-
tained by a simple redox exchange process utilizing
[FeCp2]+[Ni(mnt)2]- (6) as both an oxidant and the [Ni-
(mnt)2]--delivering reagent. The corresponding CT
complexes 7-10 were thus accessible in moderate to
high yields. The structural characterization of deriva-
tive 7 disclosed the columnar nature of the compound
in the solid state, typical for a one-dimensional semi-
conductor. From the similarity in their magnetic and
resistivity properties it is reasonable to assume that
compounds 8-10 should display a very similar struc-
ture, as compared to 7, since the observed structure is
probably dictated by the similar extension of donor and
acceptor in their longest molecular dimension (ca. 11-
13 Å). From this observation it is important to note that
the ferrocenium cation of this study thus serves as an
ideal spacer between stacks of [Ni(mnt)2]- anions.
However, the observed electrical and magnetic behavior
of compounds 7-10 appears to be essentially conveyed
by the [Ni(mnt)2]- stacks.
955, 930, 885, 840, 813, 778, 714, 638, 542, 518, 501, 454 cm-1
.
Com p ou n d 9 was obtained in an analogous manner from
50 mg (0.088 mmol) of 4 in 8 mL of 1,2-dichloroethane and 46
mg (0.088 mmol) of [FeCp2][Ni(mnt)2] (6) in 30 mL of 1,2-
dichloroethane and 5 mL of acetone. Yield: 64 mg (80%). Anal.
Calcd for
C28H14N4S12FeNi: C, 37.13; H, 1.56; N, 6.19.
Found: C, 37.00; H, 1.63; N, 6.27. IR (KBr): 3099, 2205, 1515,
1478, 1415, 1378, 1272, 1156, 1100, 1053, 967, 916, 853, 833,
788, 707, 682, 652, 634, 496, 472, 431, 366, 335 cm-1
.
Com p ou n d 10 was obtained by very slow addition of a
saturated solution of 50 mg (0.088 mmol) of [FeCp2][Ni(mnt)2]
(6) in 1,2-dichloroethane to a solution of 46.3 mg (0.088 mmol)
of 5 in 10 mL of 1,2-dichloroethane at room temperature.
Yield: 26 mg (35%). Anal. Calcd for C34H18N4S8FeNi: C,
47.84; H, 2.13; N, 6.56. Found: C, 47.73; H, 2.15; N, 6.50. IR
(KBr): 3106, 2204, 1566, 1527, 1487, 1454, 1441, 1384, 1338,
1291, 1158, 1134, 1115, 1048, 1035, 957, 918, 832, 796, 741,
680, 663, 641, 516, 501, 460, 416 cm-1
.
Exp er im en ta l Section
X-r a y Cr ysta llogr a p h ic Stu d y of 7. Suitable crystals of
compound 7 were obtained directly from the corresponding
reaction mixture as shiny black needles. Selected crystal-
lographic and relevant data collection parameters are listed
in Table 1. Data were measured at room temperature with
variable scan speed to ensure constant statistical precision on
the collected intensities. One standard reflection was mea-
sured every 120 reflections, and no significant variation was
detected. The structure was solved by direct methods and
refined by full-matrix least squares using anisotropic displace-
ment parameters for all non-hydrogen atoms. Due to the poor
quality of the data, the Cp and phenyl rings were refined as
rigid bodies. The contribution of the hydrogen atoms in their
idealized position (Riding model with fixed isotropic U )
0.080Å2) was taken into account but not refined. All calcula-
tions were carried out by using the Siemens SHELXTL PLUS
(VMS) system.
General experimental techniques were described earlier.8
1,1′-Bis[(1,3-d ith iolo[4,5-b][1,3]-d ith iol-2-ylid en e)m eth -
yl]fer r ocen e (4). A 1.00 g (3.47 mmol) amount of dimethyl
(1,3,4,6-tetrathiapentalen-2-yl)phosphonate was dissolved in
120 mL of THF. A 2.1 mL (3.6 mmol) portion of a 1.68 M
n-BuLi solution in hexane was added over a period of 5 min
at -78 °C. The resulting mixture was stirred for 5 min, and
then 419 mg (1.7 mmol) of 1,1′-ferrocenedicarbaldehyde,
dissolved in 50 mL of THF, was added dropwise over a period
of 10 min. The mixture was stirred for 3 h at -78 °C and was
then warmed slowly to room temperature. After 18 h, 10 mL
of a saturated solution of NH4Cl and 300 mL of CH2Cl2 was
added. The organic phase was washed three times with water
and then dried with MgSO4. The solvent was then evaporated
under reduced pressure until a precipitate began to form.
When the temperature was lowered to -20 °C, a red-brown
microcrystalline material formed and was subsequently fil-
tered off. Yield: 514 mg (54%). 1H NMR: δ 4.22 (t, J ) 1.9
Hz, 4H), 4.31 (t, J ) 1.9 Hz, 4H), 4.93 (s, 2H), 5.98 (s, 2H).
MS: m/ z 566 (M+), 442, 400, 342, 210, 166, 136, 121, 76 (100).
Anal. Calcd for C20H14S8Fe: C, 42.39; H, 2.49. Found: C,
42.35; H, 2.72. IR (KBr): 3081, 2990, 2918, 1633, 1572, 1520,
Resistivity Mea su r em en ts. Samples of the CT complexes
(ca. 80-100 mg) were pressed into pellets, and temperature-
dependent conductivity was measured by the standard two-
probe method.
Ma gn etic Mea su r em en ts. The magnetic susceptibility ø
of CT complexes 7-10 was measured in the temperature range
2-300 K in an external variable magnetic field, by means of
a Quantum Design superconducting quantum interference
device (SQUID) magnetometer. The polycrystalline samples
were mounted in a sample holder tube made of quartz glass
in order to keep the magnetic background as low as possible.
1453, 1390 cm-1
.
P r ep a r a tion of Ch a r ge-Tr a n sfer Com p lexes. The pro-
cedure for the preparation of CT complexes is exemplified by
the synthesis of precursor 6. A hot solution of 1.013 g (1.25
mmol) of [NBu4]2[Ni(mnt)2] in 50 mL of 1,2-dichloroethane was
filtered and then added to a refluxing solution of 683 mg (2.5
mmol) of [FeCp2]BF4 in 1,2-dichloroethane (80 mL). The
resulting black mixture was briefly stirred. Slow cooling to
room temperature led to the formation of very small shiny
black crystals. These were filtered off and dried in vacuo.
Yield: 216 mg (33%). Anal. Calcd for C18H10N4S4FeNi: C,
41.17; H, 1.92; N, 10.67. Found: C, 41.06; H, 1.97; N, 10.25.
IR (KBr): 3100, 2205, 1628, 1481, 1412, 1156, 1106, 1006, 854,
Routine corrections for core diamagnetism (7, -378
10-6 emu/mol;
10-6 emu/mol) and for the sample holder were
10-6
emu/mol; 8, -362
10-6 emu/mol; 9, -420
10, -394
applied.
Ack n ow led gm en t. M.H. and S.Z. are grateful to the
Swiss National Science Foundation for financial support
(Grants 21-36220.92 and 20-41974.94). We thank the
reviewers for useful comments.
854, 512, 499, 393, 365 cm-1
.
Com p ou n d 7 was obtained in an analogous manner from
200 mg (0.415 mmol) of 2 in 20 mL of 1,2-dichloroethane and
218 mg (0.415 mmol) of [FeCp2][Ni(mnt)2] (6) in 140 mL of
1,2-dichloroethane and 20 mL of acetone. Yield: 211 mg
(62%). Anal. Calcd for C36H26N4S6FeNi: C, 52.63; H, 3.19;
N, 6.82; S, 23.42. Found: C, 52.38; H, 3.34; N, 6.83; S, 23.48.
IR (KBr): 3095, 2921, 2201, 1606, 1576, 1542, 1486, 1461,
1402, 1376, 1321, 1294, 1176, 1154, 1119, 1088, 952, 839, 797,
Su p p or tin g In for m a tion Ava ila ble: Tables of crystal
data and refinement details, atomic coordinates, all bond
distances and angles, anisotropic displacement coefficients,
and coordinates of hydrogen atoms for 7 (10 pages). Ordering
information is given on any current masthead page. A table
of calculated and observed structure factors (8 pages) may be
obtained from the authors upon request.
628, 411 cm-1
.
OM9603981