Reactions of (Butadiene)tantalocene Cation with Alkyl Isocyanides

Article abstract of DOI:10.1021/om990215l

Treatment of (η⁴-s-trans-butadiene)tantalocene cation (3, with methyltris(pentafluorophenyl)borane anion) with tert-butyl isocyanide (4 h, at 45°C in bromobenzene) yields the [(η²-butadiene)(C≡NCMe₃)TaCp₂ ⁺] complex 4a (70% isolated, v?(C≡NR) 2164 cm^-1). Likewise treatment with n-butyl or cyclohexyl isocyanide gives the analogous products 4b and 4c, respectively. The cyclohexyl isocyanide adduct 4c was characterized by X-ray diffraction. It shows the η²-butadiene ligand oriented at the front of the Cp₂Ta⁺ bent metallocene wedge with the noncoordinated butadiene vinyl group located in the vinyl-inside position. The Ta-C≡N-R unit is almost linear (angles Ta-C-N = 177.2(5)°, C≡N-R = 174.0(8)°; d(C≡N) = 1.152(6) A?). Photolysis of [(butadiene)TaCp₂⁺] (3) with excess cyclohexyl isocyanide followed by thermal treatment gave rise to the formation of [bis(κC-cyclohexyl isocyanide)-TaCp₂⁺] (8, with [CH₃B(C₆F₅)₃^-] counteranion; >50% isolated, v?(C≡NR) 2134 and 2088 cm^-1). Complex 8 was characterized by an X-ray crystal structure analysis. It shows a pseudotetrahedral [Cp₂Ta(C≡NR)₂⁺] cation (R = C₆H₁₁). The Ta-C≡N-R units are both slightly bent at nitrogen (angles C≡N-R = 169.0(13) and 163.7(8)°; the corresponding Ta-C-N angles are 179.4(9) and 177.4(7)°). The bonding situations of complexes 4 and 8 were analyzed by DFT calculations. The computational study shows that metal to isonitrile back-bonding does not contribute significantly to the CP₂Ta⁺-C≡NR bonding interaction in these complexes. The bonding between Cp₂Ta⁺ and the C≡NR ligands is dominated by electrostatic effects.

Full text of DOI:10.1021/om990215l

3802
Organometallics 1999, 18, 3802-3812
Rea ction s of (Bu ta d ien e)ta n ta locen e Ca tion w ith Alk yl
Isocya n id es
Hans Christian Strauch, Birgit Wibbeling,† Roland Fro¨hlich,† and
Gerhard Erker*
Organisch-Chemisches Institut der Universita¨t Mu¨nster,
Corrensstrasse 40, D-48149 Mu¨nster, Germany
Heiko J acobsen‡ and Heinz Berke*,‡
Anorganisch-Chemisches Institut der Universita¨t Zu¨rich, Winterthurer Strasse 190,
CH-89057 Zu¨rich, Switzerland
Received March 29, 1999
Treatment of (η4-s-trans-butadiene)tantalocene cation (3, with methyltris(pentafluoro-
phenyl)borane anion) with tert-butyl isocyanide (4 h, at 45 °C in bromobenzene) yields the
[(η2-butadiene)(CtNCMe3)TaCp2+] complex 4a (70% isolated, ν˜(CtNR) 2164 cm-1). Likewise
treatment with n-butyl or cyclohexyl isocyanide gives the analogous products 4b and 4c,
respectively. The cyclohexyl isocyanide adduct 4c was characterized by X-ray diffraction. It
shows the η2-butadiene ligand oriented at the front of the Cp2Ta+ bent metallocene wedge
with the noncoordinated butadiene vinyl group located in the “vinyl-inside” position. The
Ta-CtN-R unit is almost linear (angles Ta-C-N ) 177.2(5)°, CtN-R ) 174.0(8)°;
d(CtN) ) 1.152(6) Å). Photolysis of [(butadiene)TaCp2+] (3) with excess cyclohexyl isocyanide
followed by thermal treatment gave rise to the formation of [bis(κC-cyclohexyl isocyanide)-
-
TaCp2+] (8, with [CH3B(C6F5)3 ] counteranion; >50% isolated, ν˜(CtNR) 2134 and 2088 cm-1).
Complex 8 was characterized by an X-ray crystal structure analysis. It shows a pseudo-
+
tetrahedral [Cp2Ta(CtNR)2 ] cation (R ) C6H11). The Ta-CtN-R units are both slightly
bent at nitrogen (angles CtN-R ) 169.0(13) and 163.7(8)°; the corresponding Ta-C-N
angles are 179.4(9) and 177.4(7)°). The bonding situations of complexes 4 and 8 were analyzed
by DFT calculations. The computational study shows that metal to isonitrile back-bonding
does not contribute significantly to the Cp2Ta+-CtNR bonding interaction in these
complexes. The bonding between Cp2Ta+ and the CtNR ligands is dominated by electrostatic
effects.
In tr od u ction
extrusion with formation of the respective Cp2MII
complexes (e.g. Cp2Zr(CO)2) compete.3
(η4-Conjugated diene)zirconocene and -hafnocene com-
plexes have found interesting applications in organic
synthesis and in catalysis.1 Several studies have been
reported where these reagents were treated with the
C1-building blocks carbon monoxide or alkyl or aryl
isocyanides (RNtC). In most of these cases incorpora-
tion of CO/CtNR is observed, leading to carbon-carbon
coupling with the conjugated diene ligand, and ulti-
mately cyclopentenone formation is often achieved.2
Only in a few exceptional examples can 1,3-diene
We have recently prepared the group 5 analogue of
these (1,3-diene) bent-metallocene systems, namely
(butadiene)tantalocene cation (3). We have shown that
only the [(η4-s-trans-C4H6)TaCp2]+ isomer is stable and
is observed at ambient temperature. It inserts a variety
of unsaturated reagents (ketones, nitriles, alkynes) to
form the respective metallacyclic (π-allyl)- or the seven-
membered (σ-allyl)metallocene complexes, similar to the
way their neutral group 4 metallocene analogues react
with these reagents.4 We have now treated (butadiene)-
* To whom correspondence should be addressed. G.E.: FAX, +49
251 83 36 503; E-mail, erker@uni-muenster.de. H.B.: FAX, +41 1 635
4680; E-mail, hberke@aci.unizh.ch.
(2) (a) Erker, G.; Engel, K.; Kru¨ger, C.; Chiang, A.-P. Chem. Ber.
1982, 115, 3311. Hessen, B.; Teuben, J . H. J . Organomet. Chem. 1988,
358, 135. Hessen, B.; Blenkers, J .; Teuben, J . H.; Helgesson, G.; J agner,
S. Organometallics 1989, 8, 830, 2809. Yasuda, H.; Okamoto, T.;
Matsuoka, Y.; Nakamura, A.; Kai, Y.; Kanehisa, N.; Kasai, N. Orga-
nometallics 1989, 8, 1139. (b) For related reactions see e.g.: Hicks, F.
A.; Kablaoui, N. M.; Buchwald, S. L. J . Am. Chem. Soc. 1996, 118,
9450. Hicks, F. A.; Buchwald, S. L. J . Am. Chem. Soc. 1996, 118, 11688.
Takahashi, T.; Xi, Z.; Nishihara, Y.; Huo, S.; Kasai, K.; Aoyagi, K.;
Denisov, V.; Negishi, E. Tetrahedron 1997, 53, 9123.
(3) Erker, G.; Kru¨ger, C.; Mu¨ller, G. Adv. Organomet. Chem. 1985,
24, 1. Beckhaus, R.; Wilbrandt, D.; Flatan, S.; Bo¨hmer, W.-H. J .
Organomet. Chem. 1992, 423, 211.
(4) Strauch, H. C.; Erker, G.; Fro¨hlich, R. Organometallics 1998,
17, 5746.
† X-ray crystal structure analyses.
‡ Quantum-chemical calculations.
(1) (a) Erker, G.; Pfaff, R.; Kru¨ger, C.; Nolte, M.; Goddard, R. Chem.
Ber. 1992, 125, 1669. Erker, G.; Pfaff, R. Organometallics 1993, 12,
1921. Erker, G.; Pfaff, R.; Kowalski, D.; Wu¨rthwein, E.-U.; Kru¨ger,
C.; Goddard, R. J . Org. Chem. 1993, 58, 6771. Berlekamp, M.; Erker,
G.; Scho¨necker, B.; Krieg, R.; Rheingold, A. L. Chem. Ber. 1993, 126,
2119. Erker, G.; Berlekamp, M.; Lo´pez, L.; Grehl, M.; Scho¨necker, B.;
Krieg, R. Synthesis 1994, 212. (b) Temme, B.; Erker, G.; Karl, J .;
Luftmann, H.; Fro¨hlich, R.; Kotila, S. Angew. Chem. 1995, 107, 1867;
Angew. Chem., Int. Ed. Engl. 1995, 34, 1755. Karl, J .; Dahlmann, M.;
Erker, G.; Bergander, K. J . Am. Chem. Soc. 1998, 120, 5643.
10.1021/om990215l CCC: $18.00 © 1999 American Chemical Society
Publication on Web 08/14/1999

Reactions of the (Butadiene)tantalocene Cation
Organometallics, Vol. 18, No. 19, 1999 3803
Sch em e 1
Sch em e 2
1
ligand). Consequently, the H/13C NMR resonances of
a pair of diastereotopic η5-cyclopentadienyl ligands at
the cationic Ta complex 4a are observed (δ 4.69, 4.68/
96.8, 96.5 ppm in bromobenzene-d5 at 298 K and 600/
150 MHz).
tantalocene cation (3) with a small series of isonitrile
reagents RNtC under various reaction conditions and
observed a quite different behavior and outcome of these
reactions. These observations are described in this
article.
The reactions between [(butadiene)TaCp2]+ (3) and
n-butyl isocyanide or cyclohexyl isocyanide proceed
equally well to give high yields of the respective 1:1
addition products 4b and 4c. In each case a single 1:1
reaction product is obtained. In principle, the addition
of the isonitrile ligand to the bent-metallocene cation
complex 3 could have resulted in the formation of two
diastereoisomeric [(η2-butadiene)Ta(L)Cp2]+ cation com-
plexes (L ) RNtC). These would differ in their relative
positions of the uncoordinated butadiene vinyl moiety;
the complexes 4 exhibit the vinyl-substituted η2-buta-
diene carbon C2 in the central position in the σ-ligand
plane at the bent metallocene wedge, whereas the
isomer 4′ would show this in a lateral position (Scheme
2).10 In each case, we have observed the formation of
only a single isomer. The product formed by the addition
of cyclohexyl isocyanide to 3 again shows the 1H/13C
NMR signals of a pair of diastereotopic Cp ligands at
tantalum. Its 13C NMR spectrum is characterized by the
presence of four butadiene carbon resonances, two at
large δ values in the typical olefinic region (δ 141.3,
111.1). These belong to the noncoordinated butadiene
vinyl group (C3, C4). The remaining pair of butadiene
ligand carbon signals are observed at much smaller δ
values (δ 23.1, 43.6 ppm; C1, C2) as is typical for a
metal-coordinated butadiene -CHdCH2 unit.11 A simi-
lar distinction between the butadiene ligand sections
Resu lts a n d Discu ssion
Th er m a lly In d u ced Rea ction s of (Bu ta d ien e)-
ta n ta locen e w ith Alk yl Isocya n id es. (η4-s-trans-
Butadiene)tantalocene cation (3) was prepared accord-
ing to the synthetic route that we have recently
developed,4 starting from (η4-s-cis-butadiene)(η5-cyclo-
pentadienyl)tantalum dichloride (1).5 Treatment with
2 molar equiv of sodium cyclopentadienide gave [(η5-
Cp)2(η1-Cp)(η2-butadiene)Ta] (2), which was subse-
quently treated with [Cp2ZrCH3+][CH3B(C6F5)3-].6
A
clean Cp transfer was achieved to yield the desired
[(C4H6)TaCp2]+ cation (3, with [CH3B(C6F5)3]- anion),
which was easily separated from the stoichiometrically
7
formed neutral coproduct Cp3ZrCH3 (see Scheme 1).
[(butadiene)TaCp2]+ cation (3) was then treated with
a variety of alkyl isocyanides to cleanly yield the
respective 1:1 addition products. The reaction between
3 and tert-butyl isocyanide takes place readily in bro-
mobenzene solution at elevated temperature. The reac-
tion is complete after 4 h at 45 °C, and the product (4a )
was isolated in close to 70% yield as a red solid. The IR
spectrum shows the ν˜[CtN(R)] band at 2164 cm-1
,
which is ca. 25 cm-1 shifted to higher wavenumbers
from the value of the free tert-butyl isocyanide.8,16 The
1H and 13C NMR data show that the butadiene ligand
was shifted into a η2-coordination9 by the attachment
of the additional donor ligand at the central tantalum
atom. This has resulted in the formation of a center of
chirality (at the carbon atom C2 of the η2-butadiene
1
is observed in the H NMR spectra of the [(η2-butadiene)-
(C6H11NtC)TaCp2]+ product. Three signals of the
coordinated -C2HdC1H2 moiety are observed at high
field (δ 1.62, 1.58, 2.60 (1-H′/H; 2-H) with coupling
2
3
constants of J ) 5.5 Hz, J ) 12.9 and 10.1 Hz). The
noncoordinated -C3HdC4H2 moiety exhibits correspond-
ing H NMR resonances at δ 5.39, 4.90, and 4.67 ppm
(3-H, 4-H′/H) with coupling constants of J ) 16.7 and
1
3
(5) Yasuda, H.; Tatsumi, K.; Okamoto, T.; Mashima, K.; Lee, K.;
Nakamura, A.; Kai, Y.; Kanehisa, N.; Kasai, N. J . Am. Chem. Soc.
1985, 107, 2410.
(6) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1994, 116,
10015.
(7) Brackemeyer, T.; Erker, G.; Fro¨hlich, R. Organometallics 1997,
16, 531.
9.9 Hz. The NMR spectra show that in each case a single
product was formed, but they do not allow for a clear
structural assignment as to which of the two, the “vinyl-
inside” (4) or “vinyl-outside” isomer (4′),10 was actually
obtained. Fortunately, we have obtained single crystals
of two of the complexes, namely the n-butyl-NtC and
the cyclohexyl-NtC adducts, which allowed for a
characterization by X-ray crystal structure analyses.
Both clearly showed the presence of the “vinyl-inside”
(8) For related effects in d0-configured metal isonitrile complexes
see e.g.: Guo, Z.; Swenson, D. C.; Guram, A. S.; J ordan, R. F.
Organometallics 1994, 13, 766 and references therein. See also:
Hurlburt, P. K.; Rack, J . J .; Luck, J . S.; Dec, S. F.; Webb, J . D.;
Anderson, O. P.; Strauss, S. H. J . Am. Chem. Soc. 1994, 116, 10003.
Antonelli, D. M.; Tjaden, E. B.; Stryker, J . M. Organometallics 1994,
13, 763. Guram, A. S.; Swenson, D. C.; J ordan, R. F. J . Am. Chem.
Soc. 1992, 114, 8991.
(9) Thomas, J . L. Inorg. Chem. 1978, 17, 1507. Sen, A.; Halpern, J .
Inorg. Chem. 1980, 19, 1073. Beck, W.; Raab, K.; Nagel, U.; Sacher,
W. Angew. Chem. 1985, 97, 498; Angew. Chem., Int. Ed. Engl. 1985,
24, 505. Peng, T.-S.; Wang, Y.; Arif, A. M.; Gladysz, J . A. Organo-
metallics 1993, 12, 4535.
(10) For related situations see e.g.: Erker, G. Acc. Chem. Res. 1984,
17, 103, and references therein. Tatsumi, K.; Nakamura, A.; Hofmann,
P.; Stauffert, P.; Hoffmann, R. J . Am. Chem. Soc. 1985, 107, 4440.
(11) Burger, B. J .; Santarsiero, B. D.; Trimmer, M. S.; Bercaw, J .
E. J . Am. Chem. Soc. 1988, 110, 3134.

3804 Organometallics, Vol. 18, No. 19, 1999
Strauch et al.
IR (CtN(cyclohexyl)) stretching band at ν˜ 2174 cm-1
,
which is slightly increased relative to the free, uncoor-
dinated isonitrile, these data indicate that the Cp2Ta+-
isonitrile coordination is lacking any substantial back-
bonding component.15 The bonding situation of the
metallocene-isonitrile linkage is probably more ad-
equately represented by the description A in eq 1 than
F igu r e 1. Molecular structure of 4c (cation only) with
nonsystematic atom-numbering scheme. Selected bond
lengths (Å) and angles (deg): Ta-C1 ) 2.118(6), Ta-C11
) 2.314(5), Ta-C12 ) 2.392(6), Ta-C13 ) 3.288(8),
C1-N2 ) 1.152(6), N2-C3 ) 1.461(7), C11-C12 )
1.416(9), C12-C13 ) 1.435(10), C13-C14 ) 1.349(10);
Ta-C1-N2 ) 177.2(5), C1-N2-C3 ) 174.0(8), C11-Ta-
C12 ) 35.0(2), Ta-C11-C12 ) 75.5(3), Ta-C12-C11 )
by B. The latter represents a situation where metal to
carbon back-bonding is substantial,16 whereas A char-
acterizes a metal/ligand arrangement that is probably
dominated by electrostatic bonding effects.15
69.5(3), Ta-C12-C13
) 116.2(5), C11-C12-C13 )
122.8(8), C12-C13-C14 ) 124.2(8).
isomers 4b,c. The structure of the n-BuNC adduct
suffered from some disorder problems, and so will not
be discussed here, but the 4c structure was of a
sufficient quality for a characterization of this class of
compounds.
Metal to isonitrile back-bonding would be indicated
by both a decrease of the ν˜(CtN-) IR band of the
coordinated CtNR ligand relative to the free isonitrile
and a substantial bending of the coordinated CtNR unit
at the central nitrogen atom.16 Both features are clearly
absent in the complexes 4. We conclude that tantalum
to isonitrile carbon back-bonding is negligible in the
cation complexes 4 and that the Cp2Ta+-CtNR bond-
ing situation is likely to be characterized by a large
electrostatic contribution. Formally the metal in the
systems 4 has a d2 configuration Ta(III), but the
pronounced metallacyclopropane character of the
(η2-butadiene)Ta subunit may have shifted the system
toward a pronounced d0 Ta(V) character.14 We have,
therefore, tried to substitute the η2-butadiene for a
second isonitrile ligand and characterize the bonding
features of the resulting formally d2-configured [Cp2-
Ta(L)2]+ systems. It turned out that we had to use a
photochemical synthetic route to prepare such a system
starting from the (s-trans-η4-butadiene)tantalocene cat-
ion (3).
P h otoch em ica lly In d u ced Rea ction s of (s-tr a n s-
η4-Bu ta d ien e)ta n ta locen e Ca tion . Scouting experi-
ments have revealed that the butadiene ligand can
rather easily be removed from 3 by photolysis in the
presence of suitable donor ligands. Thus, when a sample
of [(η4-butadiene)TaCp2]+ (3) was irradiated (Philips
HPK 125 lamp, Pyrex filter) in THF-d8 (ca. 80 mg of 4
in 0.8 mL of solvent) at -78 °C for ca. 1 h, a 55:10:5:30
mixture of the products 3 (i.e. residual starting mate-
rial), 5, 5′, and 6 (plus free butadiene) was obtained.
The products 5 and 5′ both contained a η2-butadiene
ligand coordinated to tantalum, as is evident from their
very characteristic NMR data (see Table 1), and each
of them exhibits a pair of diastereotopic Cp ligands. The
nature of the additional donor ligand is not completely
clear. We have observed the low-intensity signals of
The X-ray crystal structure analysis of complex 4c
shows well-separated anion (i.e. CH3B(C6F5)3-) and
cation moieties (Figure 1). The cation of 4c exhibits a
bent-metallocene unit (Cp(centroid)-Ta-Cp(centroid)
angle 134.9°) to which a κ-isonitrile and a η2-butadiene
ligand are coordinated, both at the front side of the bent-
metallocene wedge.12 The η2-butadiene ligand is bonded
such that the free vinyl group is oriented in the central
position (“vinyl-inside” position). The free vinyl group
shows the expected bond lengths and angles: 1.349(10)
Å (C13-C14), 1.435(10) Å (C12-C13), 124.2(8)° (C12-
C13-C14), which are similar to free 1,3-butadiene (C1-
C2 ) 1.334 Å; C2-C3 ) 1.416 Å).13 The coordinated
butadiene double bond is substantially elongated (C11-
C12 ) 1.416(9) Å). The C11-C12-C13 angle is
122.8(8)°, and the C11-C12-C13-C14 torsional angle
is 163.6(7)°. This ca. 16° deviation from the butadiene
θ value already indicates some metallacyclopropane
character of the Ta(η2-C4H6) linkage, as is commonly
observed for many early-transition-metal η2-alkene
complexes.14 Consequently, the Ta-C11 (2.314(5) Å) and
Ta-C12 (2.392(6) Å) bonds are approaching a metal-
carbon σ-bond range.
The cyclohexyl isocyanide ligand is bonded to tanta-
lum at a lateral position in the major coordination plane
at the front sector of the bent-metallocene wedge (C1-
Ta-C11 ) 107.3(3)°). The -NtC substituent is oriented
at the cyclohexyl chair in an equatorial position. The
C3-N2 bond length is 1.461(7) Å, and the Ta-C1 bond
is 2.118(6) Å. The C1-N2 linkage is in the typical range
of a carbon-nitrogen triple bond at 1.152(6) Å. The Ta-
C1-N2 moiety is very close to linear (177.2(5)°). More
importantly, the C1-N2-C3 angle (174.0(8)°) also does
not deviate much from linear. Together with the typical
(15) (a) J acobsen, H.; Berke, H.; Do¨ring, S.; Kehr, G.; Erker, G.;
Fro¨hlich, R.; Meyer, O. Organometallics 1999, 18, 1724. (b) J acobsen,
H.; Berke, H.; Brackemeyer, T.; Eisenbla¨tter, T.; Erker, G.; Fro¨hlich,
R.; Meyer, O.; Bergander, K. Helv. Chim. Acta 1998, 81, 1692. (c)
Goldman, A. S.; Krogh-J espersen, K. J . Am. Chem. Soc. 1996, 118,
12159.
(16) Martinez de Ilarduya, J . M.; Otero, A.; Royo, P. J . Organomet.
Chem. 1988, 340, 187. Go´mez, M.; Martinez de Ilarduya, J . M.; Royo,
P. J . Organomet. Chem. 1989, 369, 197.
(12) Lauher, W.; Hoffmann, R. J . Am. Chem. Soc. 1976, 98, 1729.
(13) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,
A. G.; Taylor, R. J . Chem. Soc., Perkin Trans. 2 1987, S1.
(14) Dewar, M. J . S.; Dougherty, R. C. The PMO Theory of Organic
Chemistry; Plenum Press: New York, 1975; p 300. Dewar, M. J . S.;
Ford, G. P. J . Am. Chem. Soc. 1979, 101, 783 and references therein.
Cremer, D.; Kraka, E. J . Am. Chem. Soc. 1985, 107, 3800.

Reactions of the (Butadiene)tantalocene Cation
Organometallics, Vol. 18, No. 19, 1999 3805
Ta ble 1. Selected NMR a n d IR Da ta of [(η2-bu ta d ien e)Ta (L)Cp 2]+ Ca tion Com p lexes P r ep a r ed or Gen er a ted
in Th is Stu d ya
1H NMR
13C NMR
IR
ν˜(CdC)
1-H′/H
4-H′/H
compd
2-H
3-H
C1
C2
C3
C4
2
1.28b
0.88
1.25
1.11
1.57
1.58
2.34
2.95
2.35
2.11
2.14
2.65
2.60
3.65
3.02
6.15
5.70
4.96
5.40
5.39
6.11
6.22
4.91
4.80
4.68
4.90
4.90
4.60
4.46
4.58
4.64
4.52
4.67
4.67
4.79
4.58
24.0
23.1
23.1
23.1
23.1
31.8
42.3
36.1
35.1
43.9
43.6
43.6
60.9
50.5
149.0
150.9
141.0
142.1
141.3
147.2
148.8
104.9
106.2
111.0
111.5
111.1
112.2
110.1
1595
c
1614
1613
1611
c
2′
4a
4b
4c
5
1.60b
1.34d
1.83
1.62e
1.66f
2.21f
5′
c
a
b
d
For the structures of the complexes see Schemes 2 and 3; IR spectra are in KBr. NMR in benzene-d6 at 298 K. c Not located. NMR
in bromobenzene-d5 at 298 K. e NMR in dichloromethane-d2 at 298 K. f NMR in THF-d8 at 228 K.
Sch em e 3
THF-d7 slightly off the large THF-d7 solvent resonance
in the H (600 MHz) spectrum and the respective THF-
We have then photolyzed 3 in the presence of excess
cyclohexyl isocyanide (ca. 4:1 ratio) in THF-d8 at -78
°C. When the photochemically induced butadiene cleav-
age was complete (ca. 6 h, checked by NMR) the sample
was kept for 4 h at 40 °C. This resulted in a subsequent
thermally induced exchange of ligands at the tanta-
locene cation moiety with formation of the complex [Cp2-
Ta(CtN-cyclohexyl)2+MeB(C6F5)3-] 8. Complex 8 was
isolated in >50% yield. It shows two prominent IR
1
d8 features in the 13C (150 MHz) NMR spectrum of the
sample at 228 K (for details see the Experimental
Section), but the complexes 5 and 5′ were not isolated.
Therefore, we cannot rigorously rule out that other
donor molecules (e.g. the counteranion) have coordi-
nated to the tantalum center in these complexes.
Nevertheless, it is quite clear that here we have
observed both possible [(η2-butadiene)Ta(L)Cp2]+ com-
plexes with the noncoordinated butadiene vinyl group
positioned “inside” (5) and “outside” (5′) at the front of
the bent-metallocene wedge (see Scheme 3). The re-
maining product (6) exhibits only a single Cp resonance
(δ 5.54 (1H); δ 96.9 (13C)) in addition to the anion signals.
Irradiation of a much more dilute solution of 3 in THF-
d8 (at -78 °C, 30 min) resulted in a close to complete
conversion of 3 to 6. Subsequent warming of the sample
to room temperature led to a reconversion of 6 to the
starting material 3. We propose that the newly formed
complex 6 is [Cp2Ta(THF-d8)2]+.
features (ν˜(CtNR)) at 2134 and 2088 cm-1
.
1
Complex 8 shows a single set of H and 13C NMR
spectra at ambient temperature (dichloromethane-d2,
298 K: 1H, δ 5.14 (Cp), 3.99 (CH of C6H11); 13C, δ 89.4
(Cp)). There is a single isonitrile carbon resonance at δ
167.5 ppm that is substantially shifted to high δ values
relative to free cyclohexyl isocyanide (δ 154.2 ppm
(CtN-)). However, these features correspond to aver-
aged values. Lowering the temperature of the NMR
probe rapidly leads to broadening and then splitting of
most of the resonances to eventually give rise to the
observation of three sets of signals of three isomeric
species in a 1:2:1 ratio. A close inspection of the data
has revealed that these correspond to the three isomers
having the isonitrile functionality at the two cyclohexyl
groups in axial and equatorial positions statistically
distributed. Thus, the three observed compounds are to
be described as the stereoisomers 8(eq,eq), 8(eq,ax), and
We then irradiated [(butadiene)TaCp2]+ (3) in THF-
d8 in the presence of an excess of an alkyl isocyanide
RNtC (R ) tert-butyl or cyclohexyl). In each case the
butadiene ligand was cleaved completely with formation
of the respective [Cp2Ta(L)(CtNR)]+ products 7a ,b (L
is probably THF-d8) plus free butadiene.

3806 Organometallics, Vol. 18, No. 19, 1999
Strauch et al.
F igu r e 3. Molecular structure of 8 with nonsystematic
atom-numbering scheme. Selected bond lengths (Å) and
angles (deg): Ta-C1 ) 2.143(8), Ta-C8 ) 2.101(9), C1-
F igu r e 2. Low-temperature 1H NMR spectrum (183 K,
in dichloromethane-d2) of the 8(ax,ax), 8(eq,ax), and
8(eq,eq) mixture of isomers.
N15 ) 1.143(10), C8-N16 ) 1.149(11), N15-C2
)
1.443(8), N16-C9 ) 1.440(8); C1-Ta-C8 ) 87.3(3), Ta-
C1-N15 ) 177.4(7), Ta-C8-N16 ) 179.4(9), C1-N15-
C2 ) 163.7(8), C8-N16-C9 ) 169.0(13).
1
8(ax,ax). Each of them shows a separate H/13C NMR
Cp resonance (δ 5.10/5.07, 5.05 (1:2:1 ratio)/88.8, 88.6,
88.5). The isonitrile (i.e. CtN-) 13C NMR resonances
of all three isomers are clearly distinguished at 150 MHz
and 183 K to give rise to four separate signals at δ 166.3,
165.2, 164.3, and 163.1 ppm. The assignment of the
three compounds as axially and equatorially substituted
CtN-cyclohexyl isomers is illustrated by the very
characteristic appearance of the low-temperature 1H
NMR spectrum (see Figure 2), in conjunction with a 2D
GCOSY experiment. The spectrum shows the 1:2:1 ratio
signals of the Cp ligands. The expanded area of the
spectrum shows the narrow CtN-cyclohexyl CH quin-
tet of 8(ax,ax) at δ 4.19. The 8(eq,ax) isomer exhibits a
similarly shaped narrow quintet of the CH of the axially
substituted cyclohexyl (i.e. exhibiting the methine
hydrogen in an equatorial position, which gives rise to
the rather small 3J coupling with four gauche methylene
hydrogens) at 4.13 ppm. Finally, there are two much
wider [N]CH signals closely overlapping (but GCOSY
separated) at δ 3.65 ppm. These originate from the
equatorially substituted cyclohexyls (methine H axially
positioned, resulting in two large and two small 3J
couplings) of the isomers 8(eq,ax) and 8(eq,eq), respec-
tively.
Diffusion of pentane vapor into a solution of 8 in
tetrahydrofuran gave single crystals that were suited
for an X-ray crystal structure analysis. Again, cation
and anion are well separated in the crystal (Figure 3).
The tantalum center of the cation of 8 is pseudotetra-
hedrally coordinated by two η5-cyclopentadienyls and
two κC-cyclohexyl isocyanide ligands.12 The Cp(cen-
troid)-Ta-Cp(centroid) angle of the bent metallocene
8 is 138.5°. The angle between the σ-ligands (C1-Ta-
C8) is 87.3(3)°. The tantalum-C(isonitrile) bond lengths
were determined to be 2.101(9) Å (Ta-C8) and
2.143(8) Å (Ta-C1) (4c: 2.118(6) Å), and they are
slightly different from each other. Both Ta-CtN-
cyclohexyl units in 8 deviate slightly from linearity. The
bond angles of the Ta-C8-N16-C9 unit are 179.4(9)°
(Ta-C8-N16) and 169.0(13)° (C8-N16-C9). The sec-
ond Ta-isonitrile moiety is slightly more bent at
nitrogen (Ta-C1-N15 ) 177.4(7)°, C1-N15-C2 )
163.7(8)°).
Ta ble 2. Selected Str u ctu r a l a n d IR Da ta for
Meta llocen e Alk yl Isocya n id e Ad d u cts fr om th e
Liter a tu r e for a Com p a r ison a
compd M-CtNb CtN-Rb
CtNc
ν˜(CtNR)d
ref
9
10
11
12
13
14
15
174.7(3)
e
178.4(2)
172.5(5)
e
e
173.1(3)
e
178.9(3)
176.6(6)
e
e
1.133(2)
e
1.145(4)
1.154(6)
e
e
2310
2247
2209
2195
2090
2180
2050f
2070f
2137
2102
2250
2240
2174
2134
2088
1770
1852h
15a
17
7
18
16
16
19
179(1)
175(2)
1.19(2)
16
17
g
172.0(13) 1.143(18)
167.3(13) 1.166(18)
20
21
178.5(15) 176.6(9)
179.4(16) 173.4(16) 1.133(6)
177.2(5)
179.4(9)
177.4(7)
e
1.122(6)
4c
8
174.0(8)
169.0(13) 1.149(11)
163.7(8)
e
139.5(4)
1.152(6)
this work
this work
1.143(10)
e
1.193(4)
18
19
19
22
173.1(3)
a
Structures of the compounds 9-19 are depicted in Chart 1.
Bond angles in deg. c Bond lengths in Å. In KBr; ν˜ in cm-1.e Not
b
d
determined. f CtN(R) and CtN bands. Not listed. In pentane.
g
h
axially substituted cyclohexyl ring at N15, whereas the
C9-C14 cyclohexyl ring at N16 is conformationally
disordered (for details see the Experimental Section).
A comparison with relevant structural data from the
literature (see Table 2 and Chart 1) shows that the
Ta-CtNC6H11 bonding features are in a very typical
range, close to values that have been observed for a
variety of related para- and diamagnetic four-coordinate,
pseudotetrahedral vanadocene or niobocene complexes.
These systems show linear, or at the most slightly
bent, M-CtN-R units, and they are characterized by
a shifting of the ν˜(CtNR) IR bands to higher wavenum-
bers relative to the free alkyl isocyanide ligand. The
structural features of 8 are remarkably similar to those
of the ansa-vanadocene cation 16, recently described by
Brintzinger et al. (see Table 2). In contrast, three-
coordinate Cp2M-CtNR systems such as 18 and 19
show the strong bending at nitrogen and substantial
lowering of the isonitrile IR band that is expected for a
situation characterized by strong metal to ligand back-
bonding. The slight bending at nitrogen in 8 is probably
not caused by an onsetting back-bonding effect but
seems to have other reasons.
The C8-N16 bond length is 1.149(11) Å, and the C1-
N15 distance is 1.143(10) Å. The adjacent N16-C9
(1.440(8) Å) and N15-C2 distances (1.443(8) Å) are also
in the expected range.13 The structure contains an
We conclude from these experimental findings that
metal to ligand back-bonding effects seem to be not of

Reactions of the (Butadiene)tantalocene Cation
Organometallics, Vol. 18, No. 19, 1999 3807
Ch a r t 1
F igu r e 4. Optimized geometries and relative energies for
various isomers of [(η2-butadiene)(CH3NtC)TaCp2]+ (I).
ture of 4c. The isonitrile unit is almost ideally linear,
and the η2-butadiene ligand is bonded to the metal
center in an asymmetric fashion, the terminal C-atom
forming the shorter Ta-C bond. The calculated ener-
getic differences between the four isomers, however, are
small; the second lowest structure tvo-I is only 3 kJ /
mol higher in energy and represents the “vinyl-outside”
isomer. The two s-cis-butadiene structures, cvi-I and
cvo-I, lie 6 and 7 kJ /mol higher in energy, respectively,
again with a slight preference for “vinyl-inside” coordi-
nation.
The substantially elongated coordinated double bond
indicates that the d2 transition-metal fragment is indeed
capable of some back-bonding. There is, however, no
structural evidence that back-bonding to the isonitrile
ligand also takes place. One would expect an elongation
of the (R)NtC triple bond and bending of the central
nitrogen atom of the CNR ligand.16 In all four isomers
the geometry of the isonitrile does not change under
coordination and is essentially the same as calculated
for the free ligand (dMeNtC ) 1.18 Å and dMe-NC ) 1.42
Å).
Changes in the coordination geometry of the RNtC
ligand are found in complexes of the type [(RNC)2-
TaCp2]+, in which the isonitrile ligand does not compete
with an olefinic double bond for back-bonding. The
optimized geometry for [(MeNC)2TaCp2]+ (II) is dis-
played in Figure 5. The two isonitrile ligands are of
different coordination geometries, one being almost
linear and the other being bent at the central N atom
by 17°, and they show slightly different ligand to metal
bond distances. These geometric parameters are in good
agreement with the structural features of the related
compound 8. We also carried out a calculation in which
identical local geometry of the isonitriles was enforced
by applying a C2 symmetry constraint. The resulting
structure, C2-II, is of virtually the same energy as II
and again shows CNMe units which deviate from
linearity by 14° (Figure 5).
major importance for the description of the distorted
tetrahedral (isonitrile)(L)tantalocene cation complexes
(4, 8) investigated in this study. It appears that the
TarCtNR coordination is probably dominated by elec-
trostatic effects15 and thus is very similar to the general
bonding situation that was previously observed and
characterized for a variety of d0-configured [M]rCtNR
(or CtO and NtCR) type complexes.7,10
Bon d in g of Ta n ta locen e Ca tion w ith Alk yl Iso-
cya n id e: Qu a n tu m -Ch em ica l Ca lcu la tion s. To fur-
ther investigate the bonding of isonitriles with cationic
metallocene fragments, as manifested in compounds 4
and 8, we performed density functional calculations23
on a set of representative model compounds. We wanted
to gain a deeper understanding of the interplay between
electrostatic effects and π back-donation in ligand to
metal bonding and to explore the energetics of different
structural isomers which one might encounter.
In connection with the latter problem, the structures
of four different isomers of [(η2-butadiene)(CH3NtC)-
TaCp2]+ (I) have been optimized. As indicated in the
previous sections, there exist several possibilities for the
structure of [(η2-butadiene)Ta(L)Cp2]+, in particular
concerning the orientation of the butadiene ligand. The
results of the calculations are presented in Figure 4. In
the most stable isomer, tvi-I (i.e., trans-vinyl-inside),
the butadiene ligand possesses an s-trans orientation
and is coordinated to the isonitrile in “vinyl-inside”
fashion. Not only this arrangement but also other main
geometric features are in accord with the crystal struc-
(17) Casanova, J .; Schuster, R. E. Tetrahedron Lett. 1964, 405.
(18) Ahlers, W.; Erker, G.; Fro¨hlich, R.; Peuchert, U. J . Organomet.
Chem. 1999, 578, 115.
(19) Gambarotta, S.; Floriani, C.; Chiesi-Villa, A.; Guastini, C. Inorg.
Chem. 1984, 23, 1739.
(20) Dorer, B.; Prosenc, M.-H.; Rief, U.; Brintzinger, H.-H. Collect.
Czech. Chem. Commun. 1997, 62, 265.
(21) Carrondo, M. A. A. F. de C. T.; Morais, J .; Roma˜o, C. C.; Roma˜o,
M. J .; Veiros, L. F. Polyhedron 1993, 12, 765.
(22) Martins, A. M.; Calhorda, M. J .; Roma˜o, C. C.; Vo¨lkl, C.; Kiprof,
P.; Filippou, A. C. J . Organomet. Chem. 1992, 423, 367.
(23) (a) Kohn, W.; Becke, A. D.; Parr, R. G. J . Phys. Chem. 1996,
100, 12974. (b) Baerends, E. J .; Gritsenko, O. V. J . Phys. Chem. A 1997,
101, 5383.
The Ta-CNR separation is calculated to be somewhat
smaller in II than in I, which might be indicative of a

3808 Organometallics, Vol. 18, No. 19, 1999
Strauch et al.
Ta ble 3. Bon d An a lysisa of [(CH3NtC)MCp 2]n +
Com p lexes (III, M ) Ta , n ) 1; IV, M ) Zr , n ) 0)
w ith Lin ea r -Coor d in a ted Ison itr ile
Cs-III
Cs-IV
Cs-III
Cs-IV
∆EPauli
∆Eelstat
∆Eint(A′)
468
-451
-222
363
-305
-122
∆Eint(A′′)
BE
-90
295
-114
178
a
In kJ /mol.
and orbital interactions ∆Eint
:
BE ) -[∆EPauli + ∆Eelstat + ∆Eint(A′) + ∆Eint(A′′)]
(2)
F igu r e 5. Computed structure of [(CH3NtC)2TaCp2]+ (II).
In eq 2, ∆Eint is further split into contributions from
different irreducible representations according to Cs
symmetry. The orbital structure of metallocenes is well
understood,12,25 and the three important frontier orbitals
all lie in the major coordination plane at the front sector
of the metallocene wedge (the xz plane as defined in
Figure 6). The orbital responsible for back-bonding will
be mainly M dxz based (corresponding to b2 in C2v
symmetry), and is antisymmetric with respect to the xy
mirror plane. Thus, the term ∆Eint(A′′) will represent
the main contributions to π back-bonding. The results
of the analysis are collected in Table 3. For the cationic
tantalocene complex Cs -III, we find some contribution
due to π back-bonding, which, however, is more than 2
times smaller than the σ component ∆Eint(A′). Further-
more, we have a strong electrostatic component, which
almost compensates for all the repulsive interaction. In
contrast, for the neutral zirconocene complex, the σ and
π contributions are of almost equal importance and the
absolute value of the electrostatic interaction now is
smaller than that of the Pauli repulsion. This bonding
situation is similar to that encountered in classical π
back-bonding complexes, where the π component is often
stronger than the σ component. An analysis of the first
bond dissociation energy in W(CO)6, for example, results
in values of 150 and 173 kJ /mol for ∆Eσ and ∆Eπ,
respectively.26
F igu r e 6. [(CH3NtC)MCp2]n+ complexes (III, M ) Ta,
n ) 1; IV, M ) Zr, n ) 0) with linear and bent isonitrile
ligands.
When the symmetry constraint is lifted, both struc-
tures Cs -III and Cs -IV undergo a stabilizing bending
of the isonitrile unit (Figure 6). The changes in CNMe
coordination geometry when going from Cs -III to III
are similar to those found when comparing I and II or
4 and 8. The metal to ligand bond shortens by about
0.02 Å, and the isonitrile ligands bend by about 15°, but
no significant changes in CtΝ bond length are noticed.
For IV, the ligand distortion is more drastic. The angle
at the central N atom of 128° approaches the sp2 value,
and the CtN bond elongates by 0.03 Å. The coordina-
tion geometry of the hypothetical 16e- complex [(MeNC)-
ZrCp2] is strongly reminiscent of that observed for
[(tBuNC)MoCp2].22
small increase in metal to ligand bond strength, pre-
sumably because of enhanced back-bonding. This also
serves to explain the slight bending of the coordinated
isonitrile ligand. To address the question of what extent
back-bonding to the isonitrile ligand is actually of
importance, the model compound [(MeNC)TaCp2]+ (III)
and the related isoelectronic, but neutral, complex
[(MeNC)ZrCp2] (IV) have been investigated.
We first look at the complexes with a linearly coor-
dinated isonitrile ligand, which have been restricted to
Cs symmetry in such a way that the mirror plane bisects
the two cyclopentadienyl ligands of the MCp2 unit. The
optimized geometries for Cs-III and Cs-IV are displayed
in Figure 6. It is noticeable that in the neutral complex
Cs-IV the CtΝ bond is elongated by 0.02 Å, compared
to Cs-III or the free ligand.
The bonding analysis becomes less instructive for
complexes III and IV, mostly because the different
interaction terms can no longer be differentiated by
symmetry. One can, however, assess the importance of
the π component by means of a Mulliken population
A detailed analysis of the bond between the metal-
locene fragment M and the isonitrile L clarifies the
nature of the M-L interaction. The energy BE associ-
ated with the bond-forming reaction M + L f M-L can
essentially by broken down24 into contributions due to
(24) Ziegler, T. NATO ASI Ser. 1992, C378, 367. See ref 15a,b for a
more detailed description and further references.
(25) Green, J . C. Chem. Soc. Rev. 1998, 27, 263.
(26) Ehlers, A. W.; Ruiz-Morales, Y.; Baerends, E. J .; Ziegler, T.
Inorg. Chem. 1997, 36, 5031.
Pauli repulsion ∆EPauli, the electrostatic term ∆Eelstat
,

Reactions of the (Butadiene)tantalocene Cation
Organometallics, Vol. 18, No. 19, 1999 3809
F igu r e 8. Energy decomposition of the bond energy (B)
of the M-L bond according to orbital terms and electro-
static terms. The negative values of ∆Eorb (O) and ∆Eelstat
(E) are shown, so that in any case positive energy values
indicate stabilization, and vice versa.
F igu r e 7. Frontier orbital diagram for isonitrile and
[MCp2]n+ fragments (M ) Ta, n ) 1; M ) Zr, n ) 0).
analysis. The total population of the formerly empty
virtual orbitals of the isonitrile ligand in the final
molecule serves as a measure for back-bonding. For both
systems, the value so defined increases under bending
and goes from 0.36 to 0.42 au for Cs -III f III and from
0.48 au to 0.73 au for Cs-IV f IV. Although the absolute
values should not be taken too seriously, the general
trend becomes clear. Back-bonding increases under
ligand distortion, to a small amount for the cationic
tantalocene and to a major extent for the neutral
zirconocene.
orbital term ∆Eorb, in which electronic interaction and
Pauli repulsion are combined,15a,b and write the bond
energy BE as
BE ) -[∆Eorb + ∆Eelstat
]
(3)
The results of this analysis are shown in Figure 8. As
we have found in previous studies,15a,b the isonitrile
ligand undergoes a very strong electrostatic interaction
with the cationic metal fragment, which stabilizes the
metal-ligand bond by 65 kJ /mol, compared to the
acetonitrile ligand. This electrostatic stabilization might
be one of the reasons isonitrile complexes such as 4 are
isolable and do not readily undergo insertion reactions
as do other unsaturated reagents, such as nitriles and
ketones.
In summary, the computational study supports the
notation that back-donation does not represent a major
bonding interaction between isonitriles and the tanta-
locene cation, especially if the isonitrile has to compete
with other strong π acceptors for electron density from
the metal center as in 4. Here, the bonding is dominated
by electrostatic effects, which manifest themselves in a
ν˜[CtN] band which is shifted to higher wavenumbers
from the value of the free ligand.15 In cases where
isonitriles are the only π acceptors present, as in the
case of 8, the amount of back-bonding increases and
induces a bent of the isonitrile ligand. This counters the
changes in the shift of the ν˜[CtN] band due to electro-
statics, and as a consequence ν˜[CtN] in 8 is shifted to
lower wavenumbers by 30-80 cm-1 compared to 4.
The main features of the transition-metal isonitrile
bond are well-understood,27 and the origin of the bend-
ing distortion has been analyzed.28 The degeneracy of
the π* orbitals is lifted, and one of these orbitals is
considerably lowered in energy. This in turn leads to a
more effective back-bonding interaction with a transi-
tion-metal center, and this additional stabilization of
the transition metal to ligand bond outweighs the
destabilization of the ligand framework. In contrast to
the π-acceptor capability of the isonitrile ligand, the
σ-donor ability changes little with the degree of bending.
An illustrative orbital energy diagram for CN-Me and
the ZrCp2 and [TaCp2]+ fragments is presented in
Figure 7. The valence d-orbital set for the Ta fragment
lies about 6 eV lower in energy compared to Zr, which
is mainly caused by the positive charge on the metal
center. Bending of the isonitrile only slightly enhances
the interaction of the orbitals involved in π bonding,
whereas for the Zr a strong bending results in a good
match in energy between the ligand π*-acceptor and the
metal d(π)-donor orbital. The balance between stabiliza-
tion due to enhanced back-bonding and destabilization
due to distortion of the ligand framework thus results
in a small bending for III but in a large bending for IV.
On the other hand, the σ-donor orbital of the isonitrile
is close in energy to the d set of IV but is well-separated
from that of III. We can expect a much stronger donor
interaction in case of [TaCp2]+, which is in accord with
our bond analysis as presented in Table 3.
Exp er im en ta l Section
All reactions were carried out under an inert atmosphere
(argon), using Schlenk-type glassware, or in
a glovebox.
Solvents, including deuterated solvents, were dried and dis-
tilled under argon prior to use. [(s-trans-η4-butadiene)TaCp2
-
+
CH3B(C6F5)3-] (3) was prepared as recently described by us
in the literature.4 See also ref 4 for additional general
information.
To further illustrate the importance of the electro-
static component for the transition-metal-isonitrile
bond, we compared complex Cs-III with the model
nitrile adduct [(MeCN)TaCp2]+ (Cs-V). We introduce the
Rea ction of 3 w ith ter t-Bu tyl Isocya n id e. P r ep a r a tion
of (η2-Bu ta d ien e)(KC-ter t-bu tyl isocya n id e)bis(η5-cyclo-
p en ta d ien yl)ta n ta lu m Meth yltr is(p en ta flu or op h en yl)-
bor a te (4a ). tert-Butyl isocyanide (41.0 mg, 493 µmol) was
added to a solution of 400 mg (452 µmol) of 3 in 10 mL of
bromobenzene. The mixture was stirred at 45 °C for 4 h. At
room temperature 10 mL of pentane was added to precipitate
the product. The supernatant liquid was removed and the
(27) Sarapu, A. C.; Fenske, R. F. Inorg. Chem. 1975, 14, 247.
(28) Howell, J . A. S.; Saillard, J .-Y.; Le Beuze, A.; J aouen, G. J .
Chem. Soc., Dalton Trans. 1982, 2533.

3810 Organometallics, Vol. 18, No. 19, 1999
Strauch et al.
crude product dissolved in 5 mL of dichloromethane. The
product was precipitated again by the addition of pentane (10
mL). The solvent was decanted off and the residue dissolved
in 3 mL of pentane. Diffusion of pentane vapor into this
solution furnished the product 4c as a red solid: yield 302 mg
(69%); mp 131 °C. Anal. Calcd for C38H28NBF15Ta (975.4): C,
46.80; H, 2.89; N, 1.44. Found: C, 46.14; H, 3.21; N, 1.69. IR
(KBr): ν˜ 3126 (w), 2980 (w), 2957 (w), 2164 (s), 1641 (m), 1614
(w), 1511 (vs), 1458 (vs), 1372 (w), 1266 (m), 1188 (m), 1087
Ta, Mr ) 975.37, 0.40 × 0.20 × 0.10 mm, a ) 12.622(1) Å, b )
14.559(1) Å, c ) 19.111(1) Å, V ) 3511.9(4) Å3, Fcalcd ) 1.845 g
cm-3, µ ) 32.41 cm-1, absorption correction via SORTAV (0.357
e T e 0.738), Z ) 4, orthorhombic, space group P212121 (No.
19), λ ) 0.710 73 Å, T ) 198 K, ω and æ scans, 21 190
reflections collected ((h, (k, (l), (sin θ)/λ ) 0.67 Å-1, 8460
independent and 7729 observed reflections (I g 2 σ(I)), 507
refined parameters, R ) 0.039, Rw2 ) 0.091, maximum residual
electron density 2.44 (-1.14) e Å-3 close to Ta, hydrogens
calculated and refined riding. The anisotropic thermal param-
eters indicates large movements or positional disorder, espe-
cially in the n-butyl-NtC and the butadiene units. Refine-
ment with split positions was not possible; therefore the
n-butyl group was refined with the SADI restraint. The
resulting large standard deviations of the geometrical data do
not permit a discussion of this or a comparison with other
structures. Nevertheless, the analysis confirms the chemical
composition of the compound.
(vs), 1010-940 (br m), 840 (s), 802 (w) cm-1 1H NMR
.
(bromobenzene-d5, 599.8 MHz, 298 K): δ 4.96 (ptd, 3J HH ) 16.6
3
Hz, 9.8 Hz, 1H, 3-H), 4.69 (s, 5H, Cp H), 4.68 (d, J HH ) 16.6
Hz, 1H, 4-H′), 4.67 (s, 5H, Cp H), 4.52 (dd, 3J HH ) 9.8 Hz, 2J HH
3
) 1 Hz, 1H, 4-H), 2.13 (m, 1H, 2-H), 1.34 (dd, J HH ) 9.9 Hz,
2J HH ) 5.4 Hz, 1H, 1-H′), 1.22 (s, 9H, CMe3), 1.11-1.08 (m,
4H, 1-H and Me-B(C6F5)3) ppm. 1D TOCSY NMR (bromo-
benzene-d5, 599.8 MHz, 298 K): irradiation at δ 2.14 (2H) ppm;
response at δ 4.96 (3-H), 4.68 (4-H′), 4.52 (4-H), 1.34 (1-H′),
1.09 (1-H) ppm. 13C NMR (bromobenzene-d5, 150.8 MHz, 298
Rea ction of 3 w ith Cycloh exyl Isocya n id e. P r ep a r a -
1
tion of (η2-Bu ta d ien e)(KC-cycloh exyl isocya n id e)bis(η5-
K): δ 156.9 (C5), 148.8 (d, J CF ) 232 Hz, o-B(C6F5)3), 141.0
(C3), 137.8 (d, 1J CF ) 233 Hz, p-B(C6F5)3), 136.7 (d, 1J CF ) 233
Hz, m-B(C6F5)3), 111.0 (C4), 96.8, 96.5 (each C Cp), 59.7
(CMe3), 43.9 (C2), 30.1 C(CH3)3), 23.4 (C1), 11.1 (Me-B(C6F5)3)
ppm; ipso C of C6F5 not observed.
cyclop en t a d ien yl)t a n t a lu m
Met h ylt r is(p en t a flu or o-
p h en yl)bor a te (4c). Treatment of 400 mg (450 µmol) of 3 with
53.8 mg (490 µmol) of cyclohexyl isocyanide in 10 mL of
bromobenzene at 45 °C for 3 h and subsequent workup
analogously as described above gave 334 mg (74%) of 4c as
red crystals that were suited for X-ray crystal structure
analysis; mp 141 °C. Anal. Calcd for C40H30NBF15Ta (1001.4):
C, 47.99; H, 3.02; N, 1.41. Found: C, 47.82; H, 3.13; N, 1.60.
IR (KBr): ν˜ 3123 (vw), 2942 (m), 2863 (w), 2174 (s), 1642 (m),
1611 (w), 1510 (vs), 1458 (vs), 1372 (w), 1365 (w), 1264 (m),
1088 (vs), 1080 (s), 1003-946 (br m), 841 (s), 831 (m), 802 (w)
cm-1. 1H NMR (dichloromethane-d2, 599.8 MHz, 298 K): δ 5.39
Rea ction of 3 w ith n -Bu tyl Isocya n id e. P r ep a r a tion
of (η2-Bu ta d ien e)(KC-n -bu tyl isocya n id e)bis(η5-cyclo-
p en ta d ien yl)ta n ta lu m Meth yltr is(p en ta flu or op h en yl)-
bor a te (4b). Analogously as described above, 35.9 mg (430
µmol) of n-C4H9NtC was reacted with 350 mg (390 µmol) of 3
in 10 mL of bromobenzene at 45 °C for 3 h. Workup with
dichloromethane/pentane analogously as described above gave
289 mg (76%) of 4b as red crystals, mp 143 °C. Anal. Calcd
for C38H28NBF15Ta (975.4): C, 46.80; H, 2.89; N, 1.44. Found:
C, 46.21; H, 3.20; N, 1.76. IR (KBr): ν˜ 3119 (w), 2963 (m), 2938
(m), 2185 (s), 1641 (m), 1613 (m), 1510 (vs), 1459 (vs), 1381
(vw), 1263 (s), 1088 (vs), 1002-948 (br s), 842 (vs), 834 (vs),
3
(ptd, J HH ) 16.6 Hz, 9.8 Hz, 1H, 3-H), 5.29/5.27 (s, each 5H,
3
2
4
Cp H), 4.90 (ddd, J HH ) 16.6 Hz, J HH ) 1.6 Hz, J HH ) 0.9
Hz, 1H, 4-H′), 4.67 (ddd, 3J HH ) 9.8 Hz, 2J HH ) 1.6 Hz, 4J HH
0.4 Hz, 1H, 4-H), 4.19 (m, 1H, [N]-CH), 2.62-2.58 (m, 1H,
2-H), 2.15-2.10 (m, 4H, CH2), 1.82 (dd, J HH ) 10.1 Hz, J HH
) 5.5 Hz, 1H, 1-H′), 1.79-1.70 (m, 6H, CH2), 1.58 (dd, J HH
)
3
2
660 (w), 605 (w) cm-1
.
1H NMR (dichloromethane-d2, 599.8
3
3
)
MHz, 298 K): δ 5.40 (ptd, J HH ) 16.7 Hz, 9.8 Hz, 1H, 3-H),
2
3
2
12.9 Hz, J HH ) 5.5 Hz, 1H, 1-H), 1.74-1.38 (m, 2H, CH2),
0.50 (br s, 3H, Me-B(C6F5)3). 1D TOCSY NMR (dichlo-
romethane-d2, 599.8 MHz, 298 K): irradiation at δ 2.60 (2-
H), response at δ 5.39 (3-H), 4.90 (4-H′), 4.67 (4-H), 1.82 (1-
H′), 1.58 (1H) ppm. 13C NMR (bromobenzene-d5, 150.8 MHz,
298 K): δ 156.5 (C5), 148.7 (d, 1J CF ) 230 Hz, o-B(C6F5)3), 141.3
(C3), 137.6 (d, 1J CF ) 234 Hz, p-B(C6F5)3), 136.6 (d, 1J CF ) 233
Hz, m-B(C6F5)3), 111.1 (C4), 96.7, 96.4 (each C Cp), 56.8 ([N]-
CH), 43.6 (C2), 33.0, 32.8, 24.4, 23.4 (CH2), 23.3 (C1), 10.8
(Me-B(C6F5)3) ppm; ipso C of C6F5 not observed. ESI MS
(dichloromethane-d2): cation, m/z 474 (<1%), 420 (70%), 338
(100%), 311 (68%).
5.30/5.15 (s, each 5H, Cp H), 4.90 (ddd, J HH ) 16.7 Hz, J HH
4
3
) 1.7 Hz, J HH ) 0.9 Hz, 1H, 4-H′), 4.67 (ddd, J HH ) 9.8 Hz,
4
2J HH ) 1.7 Hz, J HH ) 0.4 Hz, 1H, 4-H), 4.11-4.06 (m, 2H,
[N]-CH2-), 2.68-2.62 (m, 1H, 2-H), 1.88-1.83 (m, 2H, CH2),
3
2
1.83 (dd, J HH ) 10.2 Hz, J HH ) 5.4 Hz, 1H, 1-H′), 1.57 (dd,
2
3J HH ) 12.8 Hz, J HH ) 5.4 Hz, 1H, 1-H), 1.52-1.46 (m, 2H,
3
CH2), 1.03 (t, J HH ) 7.4 Hz, 3H, CH3), 0.50 (br s, 3H, Me-
B(C6F5)3) ppm. NOE NMR (dichloromethane-d2, 599.8 MHz,
298 K): (1) irradiated at δ 1.57 (1-H), response at δ 1.83 (1-
H′); (2) irradiated at δ 1.83 (1-H′), response at 2.65 (2-H), 1.57
(1-H); (3) irradiated at δ 2.65 (2-H), response at 4.90 (4-H′),
1.83 (1-H′); (4) irradiated at δ 4.67 (4-H), response at 5.40 (3-
H), 4.90 (4-H′); (5) irradiated at δ 4.90 (4-H′), response at 4.67
(4-H), 2.65 (2-H); (6) irradiated at δ 5.40 (3-H), response at
X-ray crystal structure analysis of 4c: formula C40H30NBF15
-
Ta, Mr ) 1001.41, 0.30 × 0.25 × 0.10 mm, a ) 12.667(1) Å, b
) 14.864(1) Å, c ) 19.263(1) Å, V ) 3626.9(4) Å3, Fcalcd ) 1.834
g cm-3, µ ) 31.41 cm-1, absorption correction via SORTAV
(0.453 e T e 0.744), Z ) 4, orthorhombic, space group P212121
(No. 19), λ ) 0.710 73 Å, T ) 198 K, ω and æ scans, 31 595
reflections collected ((h, (k, (l), (sin θ)/λ ) 0.71 Å-1, 10 530
independent and 8094 observed reflections (I g 2 σ(I)), 525
refined parameters, R ) 0.049, Rw2 ) 0.071, maximum residual
electron density 1.57 (-1.06) e Å-3 close to Ta, hydrogens
calculated and refined riding. The structure could only be
refined with the assumption of a racemic twin; with this crystal
the resulting ratio was [0.45(1)]:[0.55(1)]. We collected data
sets from different attempts on different instruments; all of
them showed this effect, and the resulting ratios were always
about 0.50:0.50.
1
4.67 (4-H). H NMR (bromobenzene-d5, 599.8 MHz, 298 K): δ
3
5.03 (ptd, J HH ) 16.6 Hz, 9.8 Hz, 1H, 3-H), 4.70 (s, 5H, Cp
3
2
4
H), 4.69 (ddd, J HH ) 16.6 Hz, J HH ) 1.6 Hz, J HH ) 0.9 Hz,
1H, 4-H′), 4.64 (s, 5H, Cp H), 4.52 (pdd, 3J HH ) 9.8 Hz, 2J HH
)
1.7 Hz, 1H, 4-H), 3.68-3.60 (m, 2H, [N]-CH2), 2.25-2.20 (m,
3
1H, 2-H), 1.48-1.43 (m, 2H, CH2), 1.33 (dd, J HH ) 9.9 Hz,
2J HH ) 5.4 Hz, 1H, 1-H′), 1.19-1.13 (m, 2H, CH2), 1.10 (br s,
3
2
3H, Me-B(C6F5)3), 1.09 (dd, J HH ) 12.9 Hz, J HH ) 5.4 Hz,
1H, 1-H), 0.79 (t, J HH ) 7.4 Hz, 3H, CH3) ppm. 1D TOCSY
3
NMR (bromobenzene-d5, 599.8 MHz, 298 K): irradiated at δ
2.22 (2H) ppm, response at δ 5.03 (3-H), 4.69 (4-H′), 4.52 (4-
H), 1.33 (1-H′), 1.09 (1-H) ppm. 13C NMR (bromobenzene-d5,
1
150.8 MHz, 298 K): δ 156.6 (C5), 148.8 (d, J CF ) 233 Hz,
1
o-B(C6F5)3), 142.1 (C3), 137.7 (d, J CF ) 233 Hz, p-B(C6F5)3),
1
136.7 (d, J CF ) 234 Hz, m-B(C6F5)3), 111.5 (C4), 96.8, 96.5
P h otolysis of 3 in THF -d 8. Gen er a tion of th e Com -
p lexes 5, 5′, a n d 6. Two NMR experiments were carried out.
In the first, a solution of 80.0 mg (89.6 µmol) of 3 in 0.8 mL of
THF-d8 was photolyzed (Philips HPK 125 lamp, Pyrex filter)
at -78 °C for 1 h. NMR measurement at -45 °C revealed the
(each C Cp), 46.4 ([N]-CH2), 43.6 (C2), 31.3 (CH2), 23.9 (C1),
19.8 (CH2), 13.1 (CH3), 11.1 (Me-B(C6F5)3) ppm; ipso C of C6F5
not observed.
X-ray crystal structure analysis of 4b: formula C38H28NBF15
-

Reactions of the (Butadiene)tantalocene Cation
Organometallics, Vol. 18, No. 19, 1999 3811
presence of a mixture of the complexes 3, 5, 5′, and 6 in a 55:
10:5:30 ratio. In a second NMR experiment a less concentrated
solution, which contained 20 mg (22.4 µmol) of 3 in 0.8 mL of
THF-d8, was irradiated under similar conditions for 30 min.
In this case an almost quantitative generation of 6 and
noncoordinated butadiene was observed. These products were
not isolated but only characterized by NMR spectroscopy. Data
1.50-1.42 (m, 2H), 1.27-1.22 (m, 2H), 1.20-1.13 (m, 1H,
cyclohexyl CH2), 0.49 (br s, 3H, Me-B(C6F5)3) ppm. 1D TOCSY
NMR (THF-d8, 599.8 MHz, 203 K): irradiation at δ 3.65 ([N]-
CH) ppm, response at δ 2.16, 1.80, 1.61, 1.46, 1.25, 1.15 ppm.
1
13C NMR (THF-d8, 150.8 MHz, 203 K): δ 148.9 (d, J CF ) 235
Hz, o-B(C6F5)3), 138.2 (d, 1J CF ) 233 Hz, p-B(C6F5)3), 137.0 (d,
1J CF ) 237 Hz, m-B(C6F5)3), 129.8 (br m, ipso-B(C6F5)3), 95.5
(C Cp), 63.6 ([N]-CH), 35.3, 34.5, 21.1 (cyclohexyl CH2), 10.2
(br m, Me-B(C6F5)3) ppm; coordinated CtN signal not ob-
served; free cyclohexyl isocyanide, δ 154.9 (CtN-), 52.4 ([N]-
CH), 31.4, 25.8, 19.7 (CH2) ppm.
1
for 5 are as follows. H NMR (THF-d8, 599.8 MHz, 228 K): δ
6.11 (ptd, 3J HH ) 9.7 Hz, 17.0 Hz, 1H, 3-H), 5.90, 5.75 (s, each
3
3
5H, Cp H), 4.79 (d, J HH ) 17.0 Hz, 1H, 4-H′), 4.60 (d, J HH
)
3
9.7 Hz, 1H, 4-H), 3.65 (m, 1H, 2-H), 2.84 (dd, J HH ) 10.2 Hz,
2J HH ) 6.2 Hz, 1H, 1-H′), 1.66 (dd, J HH ) 11.6 Hz, J HH ) 6.2
Hz, 1H, 1-H), 0.45 (br s, 3H, Me-B(C6F5)3) ppm. 1D TOCSY
NMR (THF-d8, 599.8 MHz, 228 K): irradiation at δ 4.79
(4-H′), response at δ 6.11 (3-H), 4.60 (4-H), 3.65 (2-H), 2.84
(1-H′), 1.66 (1-H) ppm. 13C NMR (THF-d8, 150.8 MHz, 228 K):
3
2
Syn th esis of Bis(KC-cycloh exyl isocya n id e)bis(η5-cy-
clopen tadien yl)tan talu m Meth yltr is(pen taflu or oph en yl)-
bor a te (8). A solution containing 180 mg (202 µmol) of 3 and
88.1 mg (807 µmol) of cyclohexyl isocyanide in 0.8 mL of THF-
d8 was photolyzed (Philips HPK 125 lamp, Pyrex filter) at -78
δ 149.0 (d, 1J CF ) 235 Hz, o-B(C6F5)3), 147.3 (gated: d, 1J CH
)
1
°C until 3 had completely disappeared (checked by H NMR).
1
153 Hz, C3), 138.2 (d, J CF ) 233 Hz, p-B(C6F5)3), 137.0 (d,
This took ca. 6 h of irradiation time. Then the sample was
kept at +60 °C until the photolytically formed intermediate
(7b) had completely been converted into 8 (ca. 4 h). Pentane
(2 mL) was added to precipitate the product. The supernatant
solvent mixture was removed and the crude product dissolved
in 1 mL of dichloromethane. The product was again precipi-
tated by the addition of 2 mL of pentane and the orange solid
product 8 collected by filtration: yield of 8 115 mg (54%); mp
123 °C. Anal. Calcd for C43H35N2BF15Ta (1056.5): C, 48.89;
H, 3.34; N, 2.65. Found: C, 48.32; H, 3.46; N, 2.76. IR (KBr):
ν˜ 3131 (vw), 2938 (s), 2859 (m), 2134, 2088 (s), 1641 (m), 1510
(vs), 1456 (vs), 1363 (m), 1265 (s), 1084 (vs), 995-935 (br m),
1J CF ) 235 Hz, m-B(C6F5)3), 129.9 (br m, ipso-B(C6F5)3), 112.2
1
(gated: t, J CH ) 155 Hz, C4), 105.3, 105.2 (each C Cp), 60.9
(1J CH ) 151 Hz, C2), 31.8 (1J CH ) 151 Hz, C1), 10.5 (br m,
Me-B(C6F5)3) ppm. Signals of potentially coordinating THF-
d8 were observed at δ 68.2 and 26.1 ppm (free solvent THF-
d8, signals at δ 67.4 and 25.3 ppm). Data for 5′ are as follows.
1H NMR (THF-d8, 599.8 MHz, 228 K): δ 6.22 (ptd, 3J HH ) 11.5
Hz, 17.0 Hz, 1H, 3-H), 5.93, 5.92 (s, each 5H, Cp H), 4.58 (d,
3J HH ) 17.0 Hz, 1H, 4-H′), 4.46 (d, J HH ) 11.5 Hz, 1H, 4-H),
3
3
2
3.08 (m, 1H, 2-H), 2.95 (dd, J HH ) 10.8 Hz, J HH ) 6.2 Hz,
3
2
1H, 1-H′), 2.21 (dd, J HH ) 11.2 Hz, J HH ) 6.2 Hz, 1H, 1-H),
0.45 (br s, 3H, Me-B(C6F5)3 ppm. 1D TOCSY NMR (THF-d8,
599.8 MHz, 228 K): irradiation at δ 4.46 (4-H) ppm, response
at δ 6.22 (3-H), 4.58 (4-H′), 3.08 (2-H), 2.95 (1-H′), 2.21 (1-H)
834 (m), 806 (m) cm-1 1H NMR (dichloromethane-d2, 599.8
.
MHz, 298 K): δ 5.14 (s, 10, Cp H), 3.99-3.92 (m, 2H, [N]-CH),
1.94-1.86 (m, 4H), 1.70-1.60 (m, 10H), 1.47-1.40 (m, 6H,
cyclohexyl CH2), 0.49 (br s, 3H, Me-B(C6F5)3) ppm. 13C NMR
(dichloromethane-d2, 150.8 MHz, 298 K): δ 167.5 (CtN-),
148.3 (d, J CF ) 235 Hz, o-B(C6F5)3), 137.4 (d, J CF ) 234 Hz,
p-B(C6F5)3), 136.5 (d, J CF ) 236 Hz, m-B(C6F5)3), 128.7 (br m,
ipso-B(C6F5)3), 89.4 (C Cp), 56.6 ([N]-CH), 32.9, 24.8, 22.9
(cyclohexyl CH2), 10.1 (Me-B(C6F5)3) ppm. ESI MS (dichlo-
romethane-d2): m/z 530 (100%), 447 (10%), 421 (68%), 365
(6%), 338 (14%), 311 (8%).
ppm. 13C NMR (THF-d8, 150.8 MHz, 228 K): δ 149.0 (d, J CF
1
1
) 235 Hz, o-B(C6F5)3), 148.8 (C3), 138.2 (d, J CF ) 233 Hz,
p-B(C6F5)3), 137.0 (d, 1J CF ) 235 Hz, m-B(C6F5)3), 129.9 (br m,
ipso-B(C6F5)3), 110.1 (gated: t, 1J CH ) 157 Hz, C4), 105.6, 105.4
(each C Cp), 50.5 (1J CH ) 143 Hz, C2), 42.3 (1J CH ) 157 Hz,
C1), 10.5 (br m, Me-B(C6F5)3) ppm. Data for 6 are as follows.
1H NMR (THF-d8, 599.8 MHz, 228 K): δ 6.39-6.31 (m, 2H,
2-H), 5.54 (s, 10H, Cp H), 5.24-5.17 (m, 2H, 1-H′), 5.11-5.05
(m, 2H, 1-H), 0.45 (br s, 3H, Me-B(C6F5)3) ppm. TOCSY NMR
(THF-d8, 599.8 MHz, 228 K): irradiation at δ 6.35 (2-H),
response at δ 5.20 (1-H′), 5.07 (1-H) ppm. 13C NMR (THF-d8,
X-ray crystal structure analysis of 8: single crystals were
obtained by diffusion of pentane vapor into a THF-d8 solution
of 8; formula C43H35N2BF15Ta, Mr ) 1056.49, 0.20 × 0.20 ×
0.10 mm, a ) 11.248(1) Å, b ) 11.481(1) Å, c ) 17.206(1) Å, R
) 108.05(1)°, â ) 103.09(1)°, γ ) 94.61(1)°, V ) 2030.3(4) Å3,
Fcalcd ) 1.728 g cm-3, µ ) 28.11 cm-1, empirical absorption
correction via ψ scan data (0.934 e C e 0.999), Z ) 2, triclinic,
space group P1h (No. 2), λ ) 0.710 73 Å, T ) 223 K, ω/2θ scans,
1
150.8 MHz, 228 K): δ 149.0 (d, J CF ) 235 Hz, o-B(C6F5)3),
138.8 (C2), 138.2 (d, 1J CF ) 233 Hz, p-B(C6F5)3), 137.0 (d, 1J CF
) 235 Hz, m-B(C6F5)3), 129.9 (br m, ipso-B(C6F5)3), 118.2 (C1),
96.9 (C-Cp), 10.5 (br m, Me-B(C6F5)3) ppm.
P h otolysis of 3 in th e P r esen ce of ter t-Bu tyl Isocya -
n id e: Gen er a tion of 7a . A solution containing 30 mg (33.6
µmol) of 3 and 5.5 mg (67.3 µmol) of tert-butyl isocyanide in
0.8 mL of THF-d8 was photolyzed (Philips HPK 125 lamp,
Pyrex filter) at -78 °C for 2 h. The NMR spectra (at -45 °C)
showed the formation of the product 7a besides free butadiene
and excess Me3CNtC. Data for 7a are as follows. 1H NMR
(THF-d8, 599.8 MHz, 223 K): δ 5.58 (s, 10H, Cp H), 1.40 (s,
9H, C(CH3)3), 0.49 (br s, 3H, Me-B(C6F5)3) ppm; free tert-butyl
isocyanide, δ 1.39 (ps, 9-H) ppm. 13C NMR (THF-d8, 150.8
MHz, 223 K): δ 148.9 (d, 1J CF ) 235 Hz, o-B(C6F5)3), 138.2 (d,
1J CF ) 233 Hz, p-B(C6F5)3), 137.0 (d, 1J CF ) 237 Hz, m-B(C6F5)3),
129.8 (br m, ipso-B(C6F5)3), 95.5 (C Cp), 57.3, 30.8 (tert-butyl),
10.2 (br m, Me-B(C6F5)3) ppm.
P h otolysis of 3 in th e P r esen ce of Cycloh exyl Isocya -
n id e: Gen er a tion of 7b. A mixture of 30 mg (33.6 µmol) of
3 with 7.3 mg (67.3 µmol) of cyclohexyl isocyanide in 0.8 mL
of THF-d8 was photolyzed at -78 °C for 2 h analogously as
described above to give 7b and free butadiene (plus excess
RNC). At room temperature 7b was slowly converted into 8
(see below). Complex 7b was not isolated but was characterized
spectroscopically, as follows. 1H NMR (THF-d8, 599.8 MHz,
203 K): δ 5.62 (s, 10H, Cp H), 3.69-3.61 (m, 1H, [N]-CH),
2.18-2.12 (m, 2H), 1.82-1.78 (m, 2H), 1.65-1.58 (m, 1H),
8508 reflections collected ((h, (k, +l), (sin θ)/λ]) 0.62 Å-1
,
8221 independent and 6390 observed reflections (I g 2σ(I)),
2
560 refined parameters, R ) 0.047, Rw ) 0.126, maximum
residual electron density 1.30 (-2.13) e Å-3, hydrogens calcu-
lated and refined riding. The anisotropic thermal parameters
of the cyclohexyl-NtC unit C8-N16 indicate the presence of
several stereoisomers also in the crystal. Resolving the kind
and ratio of these was not possible; for chemically more
meaningful geometrical parameters, restraints were used,
taking the other nondisordered unit as a model. All data sets
were collected with Enraf-Nonius CAD4, MACH3, and
KappaCCD diffractometers. Programs used: data collection,
EXPRESS and COLLECT; data reduction, MolEN and DENZO-
SMN; absorption correction for CCD data, SORTAV; structure
solution, SHELXS-86 and SHELXS-97; structure refinement,
SHELXL-93 and SHELXL-97; graphics (with unsystematical
numbering schemes), SCHAKAL-92 and DIAMOND.36
Com p u ta tion a l Deta ils. Gradient-corrected density func-
tional calculations were carried out, with corrections for
exchange and correlation according to Becke29 and Perdew,30
respectively (BP86). Geometries were optimized using the

3812 Organometallics, Vol. 18, No. 19, 1999
Strauch et al.
program system TURBOMOLE31 within the framework of the
set to 5.0. No symmetry constraints were employed, except
when noted.
RI-J approximation.32 For Ta, effective core potentials (ECP-
60) were applied, which include relativistic corrections.33
A
triple-ú valence basis plus polarization (TZVP) was employed
for all elements.34 For the optimized geometries, a bond
analysis was carried out with the ADF program package
(version 2.3).35 Main-group elements were described with a
double-ú-STO basis plus polarization, whereas for the transi-
tion metal a triple-ú-STO basis was used (ADF databases III
and IV, respectively). The integration accuracy parameter was
Ackn owledgm en t. Financial support from the Fonds
der Chemischen Industrie, the Bundesministerium fu¨r
Bildung und Wissenschaft, Forschung und Technologie,
the Deutsche Forschungsgemeinschaft, and the Swiss
National Science Foundation is gratefully acknowl-
edged.
(29) (a) Becke, A. D. J . Chem. Phys. 1986, 84, 4524. (b) Becke, A. D.
J . Chem. Phys. 1988, 88, 1053. (c) Becke, A. D. Phys. Rev. 1988, A38,
3098.
(30) (a) Perdew, J . P. Phys. Rev. 1986, B33, 8822. (b) Perdew, J . P.
Phys. Rev. 1986, B34, 7406.
(31) (a) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨lmel, C. Chem.
Phys. Lett. 1989, 162, 165. (b) Treutler, O.; Ahlrichs, R. J . Chem. Phys.
1995, 102, 346. (c) Ahlrichs, R.; von Arnim, M. In Methods and
Techniques in Computational Chemistry: METECC-95; Clementi, E.,
Corongiu, G., Eds.; STEF: Cagliari, Italy, 1995; p 509.
(32) (a) Eichkorn, K.; Treutler, O.; O¨ hm, H.; Ha¨ser, M.; Ahlrichs,
R. Chem. Phys. Lett. 1995, 240, 283. (b) Eichkorn, K.; Treutler, O.;
O¨ hm, H.; Ha¨ser, M.; Ahlrichs, R. Chem. Phys. Lett. 1995, 242, 652.
(33) Andrae, D.; Ha¨ussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
Su p p or tin g In for m a tion Ava ila ble: Figures and tables
giving details of the X-ray crystal structure analyses of the
complexes 4b,c and 8, figures and tables giving additional
NMR information, and a listing of Cartesian coordinates of
optimized geometries, together with the corresponding total
energies, for complexes I-V. This material is available free
of charge via the Internet at http://pubs.acs.org.
OM990215L
(34) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J . Chem. Phys. 1994, 100,
5829.
(36) (a) EXPRESS: Nonius BV, 1994. (b) COLLECT: Nonius BV,
1998. (c) MolEN: Fair, K. Enraf-Nonius BV, 1990. (d) DENZO-SMN:
Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307. (e)
SORTAV: Blessing, R. H. Acta Crystallogr. 1995, A51, 33. Blessing,
R. H. J . Appl. Crystallogr. 1997, 30, 421. (f) SHELXS: Sheldrick, G.
M. Acta Crystallogr. 1990, A46, 467. (g) SHELXL: Sheldrick, G. M.
Universita¨t Go¨ttingen, 1997. (h) SCHAKAL: Keller, E. Universita¨t
Freiburg, 1997. (i) Brandenburg, K. Universita¨t Bonn, 1996.
(35) (a) Baerends, E. J .; Ellis, D. E.; Ros, P. E. Chem. Phys. 1973,
2, 41. (b) teVelde, G.; Baerends, E. J . J . Comput. Phys. 1992, 99, 84.
(c) Fonseca Guera, C.; Visser, O.; Snijders, J . G.; te Velde, G.; Baerends,
E. J . In Methods and Techniques in Computational Chemistry: ME-
TECC-95; Clementi, E., Corongiu, G., Eds.; STEF: Cagliari, Italy,
1995; p 305.