Chelate complexes of functionalized cycloheptatrienyl ligands: Molybdenum complexes with linked cycloheptatrienyl-phosphane ligands and their use in catalytic carbon-carbon bond formation

Article abstract of DOI:10.1002/ejic.200390140

The synthesis of P-functionalized molybdenum chelate complexes incorporating the linked cycloheptatrienyl-phosphane ligand [2-(diisopropylphosphanyl)phenyl]cycloheptatrienyl (o-iPr₂P-C₆H₄-C₇H₆) is described. From the ligand precursor [2-(cyclohepta-2,4,6-trienyl)phenyl]diisopropylphosphane (1) the paramagnetic 17-electron dibromide [(o-iPr₂PC₆H₄-η⁷- C₇H₆MoBr₂W-MO)] (2) can readily be obtained. This is a versatile starting material for the preparation of cycloheptatrienyl molybdenum hydrides [(o-iPr₂PC₆H₄-η⁷- C₇H₆)Mo(η²-BH₄) (P-Mo)] (3) and [(o-iPr₂PC₆H₄-η⁷- C₇H₆)Mo (PPh₃)-H(P-Mo)] (4). Treatment of 3 with dimethylanilinium tetraphenylborate ([PhNMe₂H][BPh₄]) allows the production of the cationic 14-electron complex fragment [(o-iPr₂PC₆H₄-η⁷- C₇H₆)Mo(P-Mo)]⁺ (13), which can be stabilized in the presence of suitable ligands L. The complexes [(o-iPr₂PC₆H₄-η⁷- C₇H₆)MoL₂(P-Mo)]BPh₄ [L = dimethylphenyl isocyanide, (5)BPh₄; L₂ =η⁴ 111-2,5-norbornadiene, (6)BPh₄; L₂ = η²-phenylacetylene, (7)BPh₄; L₂ =η²-tert-butylacetylene, (8)BPh₄] can thus be isolated in good yields. The alkyne complex (7)BPh4 can be used as a single-source catalyst for the oligomerization of phenylacetylene, affording a mixture of triphenylbenzenes and linear oligomers. Quantum chemical calculations reveal an intimate relationship between the catalytically active 14-electron complex fragment [(o-iPr₂PC₆H₄-η⁷- C₇H₆)MO(P-Mo)⁺ (13) and hypothetical isoelectronic and isolobal complexes of the type [(o-iPr₂PC₆H₄-η⁵- C₅H₄)M(P-M)]⁺ (M = Ru, 14; M = Os, 15), indicating that cycloheptatrienyl molybdenum systems might in general be a suitable replacement for cyclopentadienyl ruthenium and osmium catalysts. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200390140

FULL PAPER
Chelate Complexes of Functionalized Cycloheptatrienyl Ligands: Molybdenum
Complexes with Linked Cycloheptatrienyl-Phosphane Ligands and Their Use in
Catalytic Carbon؊Carbon Bond Formation
Matthias Tamm,*[a][‡] Bernd Dreßel,[a][‡] Thomas Lügger,[a] Roland Fröhlich,[b] and
Stefan Grimme[b]
Keywords: Alkyne ligands / CϪC coupling / Cycloheptatrienyl ligands / Density functional calculations / Hydride ligands
The synthesis of P-functionalized molybdenum chelate com-
plexes incorporating the linked cycloheptatrienyl-phosphane
ligand [2-(diisopropylphosphanyl)phenyl]cycloheptatrienyl
(5)BPh4; L2 = η4-2,5-norbornadiene, (6)BPh4; L2 = η2-phenyl-
acetylene, (7)BPh4; L2 = η2-tert-butylacetylene, (8)BPh4] can
thus be isolated in good yields. The alkyne complex (7)BPh4
(o-iPr2P−C6H4−C7H6) is described. From the ligand precursor can be used as a single-source catalyst for the oligomeriz-
[2-(cyclohepta-2,4,6-trienyl)phenyl]diisopropylphosphane (1)
the paramagnetic 17-electron dibromide [(o-iPr2PC6H4-η7-
C7H6)MoBr2(P−Mo)] (2) can readily be obtained. This is a
ation of phenylacetylene, affording a mixture of triphenyl-
benzenes and linear oligomers. Quantum chemical calcula-
tions reveal an intimate relationship between the cata-
versatile starting material for the preparation of cyclohepta- lytically active 14-electron complex fragment [(o-iPr2PC6H4-
trienyl molybdenum hydrides [(o-iPr2PC6H4-η7-C7H6)Mo(η2- η7-C7H6)Mo(P−Mo)]+ (13) and hypothetical isoelectronic and
BH4)(P−Mo)] (3) and [(o-iPr2PC6H4-η7-C7H6)Mo(PPh3)- isolobal complexes of the type [(o-iPr2PC6H4-η5-
H(P−Mo)] (4). Treatment of 3 with dimethylanilinium tetra-
phenylborate ([PhNMe2H][BPh4]) allows the production of
the cationic 14-electron complex fragment [(o-iPr2PC6H4-η7-
C7H6)Mo(P−Mo)]+ (13), which can be stabilized in the pres-
ence of suitable ligands L. The complexes [(o-iPr2PC6H4-η7-
C7H6)MoL2(P−Mo)]BPh4 [L = dimethylphenyl isocyanide,
C5H4)M(P−M)]+ (M = Ru, 14; M = Os, 15), indicating that
cycloheptatrienyl molybdenum systems might in general be
a suitable replacement for cyclopentadienyl ruthenium and
osmium catalysts.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
C5R5)Ru(PRЈ3)]ϩ (I), which allows various organic sub-
strates to be brought together and coupled in its coordina-
tion sphere.[2,3] The similarity between the isoelectronic and
isolobal fragments [(η-C5R5)Ru] and [(η-C7H7)Mo]
prompted us to pursue a general study on the possibility
of replacing the frequently used ruthenium catalysts by the
analogous but much cheaper molybdenum systems. Con-
sequently, similar chemical and physical properties may be
expected in complexes possessing an identical set of co-li-
gands upon substitution of the [(η-C5R5)Ru] unit by [(η-
C7H7)Mo] unit.[4,5]
Our goal of introducing cycloheptatrienyl molybdenum
complexes into homogeneous catalysis is based on the con-
cept of donor functionalization, extremely successful in
cyclopentadienyl chemistry in particular,[6Ϫ9] and we have
already been able to synthesize complexes with novel cyclo-
heptatrienyl ligands bearing pendant O-, N-, S- and P-
donor groups capable of coordinating to a transition metal
center in a chelating η7:η1-fashion.[10,11] Complexes with
linked cycloheptatrienyl-phosphane ligands seemed to be
particularly well suited for applications in catalytic CϪC
coupling reactions, and so we have started to prepare pre-
cursor complexes that should allow us to obtain coordin-
Ruthenium half-sandwich complexes of the type [(η-
C5R5)Ru(PRЈ3)2Cl] (R ϭ H, Me; RЈ ϭ alkyl, aryl) are
among the most important starting materials in organo-
transition metal chemistry and have played key roles in the
stabilization of highly reactive species such as vinylidenes
and allenylidenes by metal coordination.[1] In addition,
extensive studies have shown that these complexes are par-
ticularly useful for effective ruthenium-catalyzed CϪC
bond formation.[2] The actual active catalyst system in this
process is generated by splitting off of the halide and one
phosphane ligand, resulting in the formation of the doubly
coordinatively unsaturated complex fragment [(η-
[a]
Institut für Anorganische und Analytische Chemie,
Westfälische Wilhelms-Universität,
Wilhelm-Klemm-Str. 8, 48149 Münster, Germany
E-mail: matthias.tamm@ch.tum.de
[b]
Organisch-Chemisches Institut, Westfälische Wilhelms-
Universität,
Corrensstr. 40, 48149 Münster, Germany
[‡]
Current Address: Anorganisch-chemisches Institut, Lehrstuhl
für Anorganische Chemie, Technische Universität München,
Lichtenbergstr. 4, 85747 Garching
Fax: (internat.) ϩ 49Ϫ(0)89/289Ϫ13473
1088  2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ1948/03/0306Ϫ1088 $ 20.00ϩ.50/0 Eur. J. Inorg. Chem. 2003, 1088Ϫ1098

Chelate Complexes of Functionalized Cycloheptatrienyl Ligands
FULL PAPER
atively unsaturated molybdenum complex fragments of type
II (Scheme 1).[12] In this contribution we present the syn-
theses of tetrahydroborate complexes of type [(o-R2PC6H4-
η7-C7H6)Mo(η2-BH4)(P-Mo)] (Rϭ iPr, Ph), from which the
cationic 14-electron complex fragments [(o-R2PC6H4-η7-
C7H6)Mo(P-Mo)]ϩ can be generated and stabilized in the
presence of suitable co-ligands.
Scheme 1. Isolobal relationship between cyclopentadienyl ruthe-
nium and cycloheptatrienyl molybdenum complex fragments
Results and Discussion
Preparation and Reactions of Molybdenum Hydride
Complexes
We have recently reported on the preparation of a cyclo-
heptatriene with a diphenylphosphanyl donor group, and
were able to demonstrate that the resulting paramagnetic
17-electron complexes of the type [(o-Ph2PC6H4-η7-
C7H6)MoX2(P-Mo)] (X ϭ Br, CH2SiMe3) can be used in
the ring-opening metathesis polymerization of norbor-
nene.[11] In order to render the donor moiety more sterically
demanding and electron-rich, we aimed at replacement of
the phenyl substituents with isopropyl groups, and the cor-
responding ligand precursor [2-(cyclohepta-2,4,6-trienyl)-
phenyl]diisopropylphosphane (1) used in this contribution
was prepared by addition of 2-(diisopropylphosphanyl)-
phenyllithium to the tropylium cation C7H7ϩ. A three-step
synthesis starting from the air-sensitive 1 and Mo(CO)6 af-
fords the 17-electron dibromo complex 2 (Scheme 2). Full
details of the preparation of 1 and 2, together with the spec-
troscopic and structural properties of all reaction interme-
diates, will be presented elsewhere.[13]
Scheme 2. XyNC ϭ 2,6-dimethylphenyl isocyanide; NBD ϭ 2,5-
norbornadiene; DMAϩ ϭ dimethylanilinium (PhNMe2 H)ϩ
Reduction of paramagnetic 2 can be achieved by use of
NaBH4 in ethanol solution, resulting in the formation of
the tetrahydroborate complex 3 (Scheme 2). In contrast,
Suzuki and co-workers have shown that the corresponding
treatment of isoelectronic ruthenium() complexes of the
type [(η-C5Me5)Ru(PR3)Cl2] (R ϭ Ph, iPr, cyclo-C6H11)
does indeed also give tetrahydroborate complexes [(η-
C5Me5)Ru(PR3)(η2-BH4)], but that these undergo immedi-
ate ethanolysis under these conditions to yield the trihy-
drides [(η-C5Me5)Ru(PR3)H3].[14] In our hands, 2 could not
be converted into a trihydride complex and remained per-
fectly stable toward hydrolysis even in wet ethanol solution.
The molecular structure of 3 is shown in Figure 1 (top),
Figure 1. ORTEP drawings of 3 (top) and 4 (bottom) with thermal
and selected bond lengths and angles are listed in Table 1. ellipsoids drawn at 50% probability
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098
1089

M. Tamm, B. Dreßel, T. Lügger, R. Fröhlich, S. Grimme
Table 1. Selected bond lengths (A) and angles (°) for 3, 4, (5)BPh4·C5H12, (6)BPh4·CH2Cl2, (7)BPh4, and (10)BPh4·3C4H8O
FULL PAPER
˚
3
4
(5)BPh4·C5H12
(6)BPh4·CH2Cl2
(7)BPh4
(10)BPh4·3C4H8O
C1ϪC2
C2ϪC3
C3ϪC4
C4ϪC5
C5ϪC6
C6ϪC7
C1ϪC7
C1ϪC8
C14ϪC15/C26ϪC27
C15ϪC16/C27ϪC28
C20ϪC21
C23ϪC24
C20ϪN1
C29ϪN2
MoϪC1
MoϪC2
MoϪC3
MoϪC4
MoϪC5
1.4218(19)
1.413(2)
1.418(2)
1.402(2)
1.427(2)
1.414(2)
1.419(2)
1.5042(18)
1.425(3)
1.401(3)
1.407(3)
1.406(3)
1.419(3)
1.409(3)
1.409(3)
1.497(3)
1.417(6)
1.376(7)
1.404(7)
1.401(7)
1.401(7)
1.403(6)
1.416(6)
1.509(6)
1.420(5)
1.391(5)
1.410(6)
1.392(6)
1.422(6)
1.392(6)
1.406(5)
1.502(5)
1.408(4)
1.402(4)
1.422(5)
1.380(5)
1.415(5)
1.405(4)
1.410(4)
1.496(4)
1.292(4)
1.457(4)
1.403(5)
1.435(5)
1.411(5)
1.403(6)
1.404(6)
1.413(5)
1.421(5)
1.500(5)
1.279(5)
1.465(5)
1.371(6)
1.368(5)
1.164(5)
1.165(6)
2.264(5)
2.321(5)
2.298(5)
2.313(5)
2.301(5)
2.286(5)
2.301(4)
2.2527(12)
2.2744(14)
2.2557(14)
2.3191(13)
2.3168(14)
2.2550(13)
2.2736(11)
2.286(2)
2.316(2)
2.319(2)
2.323(2)
2.310(2)
2.261(2)
2.294(2)
2.324(3)
2.294(4)
2.272(4)
2.343(4)
2.343(4)
2.270(4)
2.311(4)
2.373(3)
2.292(3)
2.221(3)
2.324(3)
2.343(3)
2.241(3)
2.280(3)
2.023(3)
2.090(3)
2.366(3)
2.325(3)
2.226(3)
2.311(3)
2.351(4)
2.246(3)
2.268(3)
2.035(3)
2.090(3)
MoϪC6
MoϪC7
MoϪC14/C26
MoϪC15/C27
MoϪC20
MoϪC21
MoϪC23
MoϪC24
MoϪC29
MoϪP/P1
MoϪP2
2.076(5)
2.381(4)
2.401(4)
2.400(4)
2.388(4)
2.078(5)
2.4958(13)
2.4824(3)
2.4912(5)
2.4886(6)
2.5863(10)
2.5283(8)
136.6(3)
2.5073(9)
138.4(3)
C14ϪC15ϪC16/
C26ϪC27ϪC28
The positions of all hydrogen atoms were refinable, indicat- hydrides such as 4 can also be efficiently prepared directly
ing that the BH4 anion is coordinated to the molybdenum from the dibromide 2 by use of various hydride sources in
atom through two three-center, two-electron bonds, thereby the presence of Lewis bases. Consequently, treatment of 2
simultaneously blocking two coordination sites. This bi- with sodium hydride in the presence of PPh3 affords 4 in
dentate M(µ2-H2BH2) ligation represents the most common good yield. In 4, the pseudotetrahedrally coordinated mo-
mode of bonding in tetrahydroborate transition metal com- lybdenum center has four different ligands, and so the com-
plexes.[15] In addition, it is possible to establish the angles plex is obtained as a racemic mixture of two enantiomers.
between the centroid of the cycloheptatrienyl ring, the C7 As this representative of so-called ‘‘chiral-at-metal’’ half-
carbon atoms, and the adjacent hydrogen atoms, to reveal sandwich compounds[17Ϫ19] is configurationally stable, the
a significant out-of-plane displacement for the C7 ring hy- diastereotopic cycloheptatrienyl hydrogen atoms give rise to
drogen atoms. The average bending is about 9° toward the six different 1H NMR resonances between 3.8 and 5.5 ppm.
molybdenum center, such deviation having previously been This portion of the 600 MHz spectrum is shown in Fig-
attributed to a reorientation of the large seven-membered ure 2. Correspondingly, seven 13C NMR resonances are ob-
ring for a better metal overlap.[4,16]
served for the C7H6 carbon atoms, together with six reson-
As described above, 3 remained perfectly stable in eth- ances for the phenylene bridge and six resonances for the
anol solution. In the presence of Lewis bases such as PPh3, diastereotopic isopropyl groups. The hydride resonance in
however, rapid ethanolysis and cleavage of the BH4 ligand 4 is observed as a doublet of doublets at Ϫ2.88 ppm with
can be achieved, and formation of the monohydride com- 62 and 41 Hz couplings to the two different phosphorus
plex 4 is observed together with evolution of dihydrogen. nuclei (Figure 2). Recrystallization of 4 from hexane af-
The reaction can be followed by 31P NMR spectroscopy, forded single crystals suitable for structure determination
which reveals two doublets at 74.5 and 53.7 ppm for the by X-ray diffraction; the molecular structure is shown in
two different phosphorus nuclei in 4 (2JP,P ϭ 31 Hz) to- Figure 1 (bottom). The position of the MoϪH hydrogen
gether with the singlet at δ ϭ 67.9 ppm for complex 3. Pure atom could be refined to reveal a molybdenum hydrogen
1090
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098

Chelate Complexes of Functionalized Cycloheptatrienyl Ligands
FULL PAPER
containing only additional phosphane ligands.[24,25] Ac-
cordingly, the molecular structure of the cation in
(5)BPh4·C5H12 (Figure 3, top) has longer Mo-C bond
˚
lengths [2.076(5) and 2.078(5) A] to the isocyanide carbon
atoms, together with slightly shorter CϵN bond lengths
˚
[1.164(5) and 1.165(6) A], in comparison with the values
found for the centrosymmetric structure of trans-
[25]
˚
[Mo(dppe)2(CNPh)2] [2.031(6) and 1.171(6) A].
Only
slight deviation from linearity is observed for the Mo-CϪN
and CϪNϪC bond angles in 5 (168.7(5) Ϫ 174.6(5)°).
Figure 2. Selected parts of the 1H NMR spectrum (600 MHz,
[D8]toluene) of 4
˚
bond length of 1.69(2) A. As would be expected, this dis-
tance is slightly shorter than the MoϪH distances of
˚
1.88(2) and 1.96(2) A in 3.
Compound 4 is a rare example of a cycloheptatrienyl
transition metal hydride, and to the best of our knowledge,
[(η7-C7H7)Mo(dppe)H] represents the only previously re-
ported closely related example,[20] apart from a series of
tungsten complexes of the type [(η7-C7H7)W(PR3)(CO)H]
(R ϭ OMe, OiPr, Ph).[21] Furthermore, chiral hydrides such
as 4 might Ϫ in view of the extensive chemistry of related
cyclopentadienyl iron, ruthenium, and osmium hydrides Ϫ
be interesting starting materials for hydrometalation or hy-
drogenation reactions.[22] For instance, protonation of the
hydride ligand in 4 might result in the formation of a
cationic
dihydrogen
complex
[(o-iPr2PC6H4-η7-
C7H6)Mo(PPh3)(H2)(P-Mo)]ϩ, from which the 14-electron
complex fragment [(o-iPr2PC6H4-η7-C7H6)Mo(P-Mo)]ϩ
could be generated by splitting of the phosphane and the
dihydrogen ligands. To our disappointment, however, de-
composition was mainly observed upon treatment of 4
with HBF4·Et2O.
More conveniently, the tetrahydroborate 3 proved to be
an ideal starting material for the generation of our target
14-electron complex by cleavage of the boranate ligand. In
Figure 3. ORTEP drawings of the cations in (5)BPh4·C5H12 (top)
and (6)BPh4·CH2Cl2 (bottom) with thermal ellipsoids drawn at
50% probability
In
a
similar fashion, treatment of
3
with
fact, this unit can be made available by treatment of 3 with [HNMe2Ph][BPh4] in the presence of norbornadiene gives
dimethylanilinium (DMAϩ) salts, widely used as mild acids the diolefin complex (6)BPh4, which is also a versatile start-
in metallocene chemistry.[23] Treatment of
3
with ing material for further ligand substitution reactions. Treat-
[HNMe2Ph][BPh4] proceeds with evolution of dihydrogen, ment of (6)BPh4 with XyNC thus results in the quantitative
and the formation of stable cationic adducts is observed in formation of (5)BPh4. The norbornadiene complex was also
THF and acetonitrile solutions. The bis(acetonitrile) com- characterizable by X-ray diffraction analysis, and the mo-
plex can even be isolated in crystalline form. The two avail- lecular structure of the cation in (6)BPh4·CH2Cl2 is shown
able coordination sites can be blocked irreversibly by addi- in Figure 3 (bottom). Although the structural parameters
tion of more strongly coordinating ligands. If the pro- about the molybdenum centers again fall in the expected
tonolysis is carried out in the presence of 2,6-dimethyl- ranges (Table 1),[11] the norbornadiene ligand seems to
phenyl isocyanide (XyNC), for instance, the stable cause some steric congestion at the metal center, and so an
˚
diisocyanide complex (5)BPh4 is obtained. The CN stretch- elongated MoϪP bond length of 2.5863(10) A together
ing frequencies of 2099 and 2066 cmϪ1 show that the cat- with a slight puckering of the seven-membered ring can be
ionic [(o-iPr2PC6H4-η7-C7H6)Mo(P-Mo)]ϩ unit has a rela- observed with the longest molybdenum-carbon bonds to C4
tively strong electron-releasing capability, although signific- and C5, which are adjacent to the diolefin ligand. A similar
antly stronger metal-to-ligand backbonding is observed in, observation has been made in the related ruthenium com-
for instance, trans-diisocyanide molybdenum(0) complexes plex [(η-C5Me5)Ru(PPh3)(η4-NBD)]ClO4.[26]
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098
1091

M. Tamm, B. Dreßel, T. Lügger, R. Fröhlich, S. Grimme
FULL PAPER
Preparation of Molybdenum Alkyne Complexes
Cyclopentadienyl ruthenium complexes are extensively
used for CϪC bond formation through the employment of
terminal alkynes, which can either be oligomerized or
coupled with various other substrates such as allyl alcohols
or alkenes.[2,3] We have consequently studied the cleavage of
the tetrahydroborate ligand in the presence of phenyl- and
tert-butylacetylene, resulting in the formation of crystalline
(7)BPh4 and (8)BPh4. Both complexes contain η2-coordin-
ated alkyne ligands, which stabilize both available coordina-
tion sites with their π-electrons and must therefore act as
four-electron donors. This fact is easily deduced from their
1H NMR spectra, which exhibit characteristic resonances
at low field for the acetylenic HC-hydrogen atoms.[27] These
resonances appear as doublets at δ ϭ 10.41 ppm (3JH,P
ϭ
29.4 Hz) for 7 and at δ ϭ 10.03 ppm, (3JH,P ϭ 33.0 Hz) for
8. A similar downfield shift is observed in the 13C NMR
spectra for the acetylenic carbon atoms. For the phenylacet-
ylene complex 7, these resonances are found as doublets at
δ ϭ 178.5 ppm (2JC,P ϭ 28.4 Hz, ϵCH) and at δ ϭ
173.2 ppm (2JC,P ϭ 4.5 Hz, ϵCPh). Surprisingly, the order
of these resonances is reversed for the tert-butylacetylene
complex 8, its 13C NMR spectrum exhibiting a broad sing-
let at δ ϭ 191.2 ppm (ϵCtBu) and a doublet at 177.8 ppm
(2JC,P ϭ 30.1 Hz, ϵCH).
Figure 4. ORTEP drawings of the cations in (7)BPh4 (top) and
(10)BPh4·C6H14 (bottom) with thermal ellipsoids drawn at 50%
probability
plex 9 containing the linked cycloheptatrienyl-phosphane li-
gand with a diphenylphosphanyl donor group, the synthesis
and coordination chemistry of which have been reported
before.[11] Treatment of [(o-Ph2PC6H4-η7-C7H6)MoBr2-
(PϪMo)] with NaBH4 in EtOH does indeed result in the
formation of 9, which could be isolated in relatively low
yield as a green, air-sensitive solid. The crude product was
In addition, the molecular structure of the cation 7 (Fig-
ure 4, top) shows the structural parameters for the alkyne
ligand, with an elongated CϵC (C14ϪC15) bond length of
˚
1.291(4) A, together with
a
small CϵCϪC(Ph)
(C14ϪC15ϪC16) angle of 136.5(3)°, which are clearly in-
dicative of a four-electron donor alkyne ligand.[27] The geo-
metry around the molybdenum atom can be interpreted as
a two-legged piano stool with the acetylenic C14ϪC15 car-
bon bond and the MoϪP axis in the same plane, and the
P-Mo-C14ϪC15 torsion angle is 173.7°. The acetylenic
phenyl and the ortho-phenylene rings are also almost copla-
nar (interplanar angle of 16.7°), so the complex can be re-
garded as being essentially Cs-symmetric if the two isopro-
pyl groups are neglected. The molybdenum distances to the
1
only characterized by H and 31P NMR spectroscopy and
was used without further purification for the preparation of
the alkyne complex (10)BPh4. Thus, treatment of 9 with
[HNMe2Ph][BPh4] in the presence of phenylacetylene af-
forded (10)BPh4 as
a stable, crystalline compound
(Scheme 3). Single crystals of (10)BPh4·3THF could be ob-
tained by recrystallization from THF/hexane solution, and
Figure 4 (bottom) shows an ORTEP presentation of the
cation 10. The alkyne ligand in 10 seems to be somewhat
more weakly coordinated than in 7, as it has a slightly
˚
ring carbon atoms range from 2.221(3) A (Mo-C3) to
˚
˚
2.373(3) A (Mo-C1), indicating a fairly significant deviation
shorter CϵC bond length [1.279(5) vs. 1.292(4) A], together
from planarity (Table 1), and the mean and maximum devi-
with a slightly larger CϵCϪC(Ph) angle [138.4(3) vs.
136.5(3)°]. The overall geometry, though, is similar to that
observed for 7 (Table 1), although the phenylacetylene li-
gand deviates more strongly from a perfectly Cs-symmetric
conformation, as indicated by the torsion angle of 169.8°
between the P-Mo-C26ϪC27 atoms and by the interplanar
angle of 26.6° between the least-squares planes containing
the acetylenic phenyl and phenylene carbon atoms. From
the difficulties encountered in particular with the synthesis
of the tetrahydroborate derivative 9, we were able to ascer-
tain that in our hands the more sterically demanding diiso-
propylphosphane donor seems to be superior to its di-
phenylphosphane analogue. We have generally observed in-
creased stability upon replacement of the phenyl groups
ations from the least-squares plane (C1ϪC7) are 0.075 and
˚
Ϫ0.108 A (for C3), respectively. Nevertheless, these struc-
tural parameters are still strongly indicative of an η7-coord-
inated cycloheptatrienyl ligand. We are not aware of any
related ruthenium complex containing an η2-bonded alkyne
ligand. Very recently, however, the molecular structure
of
a
related osmium complex [(η-C5H5)Os(η2-
HCϵCCPh2OH)(PiPr3)]PF6 containing a coordinated pro-
pargyl alcohol has been reported.[28] Comparison of the
bond lengths and angles with those in 7 demonstrates that
cycloheptatrienyl molybdenum complexes are also strongly
related to cyclopentadienyl osmium systems, with which
they exist in a diagonal relationship.
For purposes of comparison, we also studied the possibil- with isopropyl groups, and so both our experimental work
ity of synthesizing the corresponding tetrahydroborate com- and our catalytic studies have mainly focused on, and
1092
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098

Chelate Complexes of Functionalized Cycloheptatrienyl Ligands
FULL PAPER
mainly will continue to focus on, the use of complexes con- phanylphenyl substituent (Scheme 5). The structures of all
taining the [2-(diisopropylphosphanyl)phenyl]cycloheptatri- three cationic complexes were optimized by DFT methods
enyl ligand.
employing the BP86 functional (Figure 5). The alkyne con-
formations in all three complexes are very similar and are in
agreement with the expected nearly Cs-symmetric geometry
about the metal centers. The phenylacetylene ligand in each
complex exhibits the expected lengthening of the CϵC
Scheme 3. DMAϩ ϭ dimethylanilinium (PhNMe2 H)ϩ
Catalytic Study
Alkyne complexes such as 7, 8, and 10 could be interme-
diates in the catalytic coupling of acetylenes at cationic mo-
lybdenum complex fragments of type II (Scheme 1).[2,3] We
have therefore studied the use of (7)BPh4 (5 mol %) as a
catalyst for the oligomerization of phenylacetylene[29] in
THF solution at 80 °C. Our preliminary results indicate
that this system is indeed efficient at catalyzing the above
reaction, affording a product mixture of cyclotrimers (80%)
and linear oligomers (20%). Surprisingly, cyclotrimerization
to give 1,3,5-triphenylbenzene (52%, unambiguously char-
acterized by X-ray diffraction analysis)[30] and 1,2,4-tri-
phenylbenzene (28%) is predominantly observed, in addi-
tion to the formation of a linear dimer (12%) and a linear
trimer (8%) (Scheme 4). This result contrasts with the reac-
tion involving ruthenium catalysts, in which linear dimers
and trimers are formed almost exclusively.[3b,3e,3f] Further
work will be directed toward the elucidation of the oligo-
merization mechanism[31] effective with our novel molyb-
denum catalyst system. Thus, (7)BPh4 and our related al-
kyne complexes represent ideal starting materials for further
controlled stoichiometric transformations, which will hope-
fully allow additional possible intermediates in the catalytic
cycle to be identified.
Scheme 5. Alkyne bond dissociation energies (De) of phenylacetyl-
ene complexes
Scheme 4. Catalytic oligomerization of phenylacetylene
Theoretical Study
To study the (C7H7)Mo-(C5R5)M-relationship (M ϭ Ru,
Os) by theoretical methods, we chose the molybdenum
phenylacetylene complex 7 for comparison with the hypo-
thetical cyclopentadienyl ruthenium and osmium complexes
11 and 12, each containing an identical diisopropylphos- Figure 5. Calculated structures of the cations 7, 11, and 12
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098
1093

M. Tamm, B. Dreßel, T. Lügger, R. Fröhlich, S. Grimme
FULL PAPER
˚
Scheme 5 shows the theoretically predicted alkyne bond
dissociation energies (De) of 7, 11, and 12, calculated as the
energy difference between the alkyne complex on one hand,
and the unoptimized complex fragments 13 Ϫ15 and the
phenylacetylene ligand at its optimized geometry on the
other (vertical De). The predicted De for the molybdenum
complex 7 (68.7 kcal·molϪ1) takes an intermediate position
between the values calculated for the Ru (De ϭ 63.5
kcal·molϪ1) and Os systems (De ϭ 84.3 kcal·molϪ1). Com-
parative calculations employing the popular B3LYP hybrid
functional result in De values approximately 10 kcal·molϪ1
Table 2. Experimentally measured and calculated bond lengths (A)
and angles (°) for 7, 11, and 12
(7)BPh4
1.408(4)
1.402(4)
1.422(5)
1.380(5)
7
11
1.439
1.442
1.427
1.440
1.440
12
1.429
1.435
1.420
1.432
1.432
C1ϪC2
C2ϪC3
C3ϪC4
C4ϪC5
C5ϪC1
C5ϪC6
C6ϪC7
C1ϪC7
C1ЈϪC2Ј
C2ЈϪC3Ј
MϪC1
MϪC2
MϪC3
MϪC4
MϪC5
MϪC6
MϪC7
1.423
1.426
1.426
1.405
1.415(5)
1.405(4)
1.410(4)
1.292(4)
1.457(4)
2.373(3)
2.292(3)
2.221(3)
2.324(3)
2.343(3)
2.241(3)
2.280(3)
2.023(3)
2.090(3)
2.5283(8)
76.52(9)
112.86(8)
136.6(3)
1.428
1.424
1.421
1.318
1.450
2.403
2.300
2.246
2.356
2.343
2.259
2.338
2.033
2.113
2.580
1.309
1.445
2.220
2.193
2.210
2.215
2.205
1.306 lower, but the differences between the three complexes are
1.446
2.275
2.246
2.258
2.262
2.253
very similar. Figure 6 shows a representation of the frontier
orbitals in the unoptimized 14-electron complexes 13 Ϫ15
formed by splitting of the phenylacetylene ligand.
In each complex, the symmetry properties of the two low-
est unoccupied molecular orbitals (LUMOs), which are es-
sentially vacant d-orbitals, match those of the occupied π||
and πЌ orbitals of an alkyne ligand. Consequently, this can
act as a four-electron donor, providing 18-electron com-
MϪC1Ј
MϪC2Ј
MϪP
PϪMϪC1Ј
PϪMϪC2Ј
C1ЈϪC2ЈϪC3Ј
2.004
2.029
2.388
81.2
119.1
143.8
2.004
2.031
2.416 plexes.[27,28,32] Comparison of the energies and shapes of the
76.8
113.7
138.8
82.3
120.0
142.7
relevant frontier orbitals reveals the close resemblance be-
tween 13, 14, and 15 and confirms the intimate relationship
between the isoelectronic and isolobal [(η-C7H7)Mo] and
[(η-C5R5)M] fragments (M ϭ Ru, Os).
bond length, together with a small CϵCϪC(Ph) angle
(Table 2). For the molybdenum complex 7, the computed
optimized geometry is in excellent agreement with the struc-
ture determined by X-ray diffraction analysis (Table 2, Fig-
ure 3), and the puckering of the seven-membered ring is
Conclusion
In summary, we present a new method for the generation
also accurately reproduced. This deviation from planarity and stabilization of coordinatively unsaturated cyclohepta-
is significantly less pronounced in 11 and 12, the structural trienyl molybdenum complexes, which can be used for cata-
data of which are very similar and in excellent agreement lytic alkyne-alkyne coupling reactions. These results indi-
with the experimentally measured and calculated geometric cate that cycloheptatrienyl complexes can indeed be poten-
parameters of unbridged cyclopentadienyl osmium alkyne tially useful for applications in homogeneous transition
complexes.[28]
metal catalysis, the goal of these studies being the develop-
Figure 6. Frontier orbitals of the 14-electron complexes 13 [M ϭ Mo], 14 [M ϭ Ru], and 15 [M ϭ Os]
1094 Eur. J. Inorg. Chem. 2003, 1088Ϫ1098

Chelate Complexes of Functionalized Cycloheptatrienyl Ligands
FULL PAPER
C7H6), 4.18 (d, 1 H, C7H6), 3.86 (q, 1 H, C7H6), 2.10 (m, 1 H, iPr:
CH), 1.09 (d sept, 1 H, iPr: CH), 0.86 (m, 6 H, iPr: CH3), 0.81 (dd,
3 H, iPr: CH3), 0.68 (dd, 3 H, iPr: CH3), Ϫ2.82 (dd, 1 H, MoH)
ment of novel effective catalyst systems that should allow
the precious metal ruthenium (and also osmium) to be sub-
stituted with the much cheaper molybdenum. This work is
part of our general goal to extend the chemistry of cyclo-
heptatrienyl complexes and to raise their level of signific-
ance in comparison with those of cyclopentadienyl and ben-
zene complexes.
2
ppm. 13C NMR ([D8]toluene, 150.9 MHz): δ ϭ 154.0 (d, JC,P
ϭ
1
25.8 Hz, C-8), 141.4 (d, JC,P ϭ 28.7 Hz, ipso-C6H5), 140.5 (d,
1JC,P ϭ 28.3 Hz, C-9), 134.2 (d, 3JC,P ϭ 11.0 Hz, o-PPh3), 134.0 (d,
3
2JC,P ϭ 19.9 Hz, m-PPh3), 129.1 (m, C-10), 128.3 (d, JC,P
ϭ
ϭ
ϭ
3
1.9 Hz, C-11), 127.2 (d, JC,P ϭ 8.5 Hz, p-PPh3), 126.9 (d, JC,P
2
7.7 Hz, C-13), 126.7 (d, JC,P ϭ 4.1 Hz, C-12), 105.4 (t, JC,P
2
2
2.0 Hz, C-1), 88.3 (d, JC,P ϭ 1.5 Hz, C-7), 87.9 (d, JC,P ϭ 3.1 Hz,
2
2
Experimental Section
C-5), 87.9 (s, C-6), 85.9 (d, JC,P ϭ 5.1 Hz, C-4), 84.6 (dd, JC,P
ϭ
1
8.3 Hz, C-2), 80.6 (s, C-3), 26.5 (d, JC,P ϭ 21.4 Hz, iPr: CH), 25.7
(dd, 1JC,P ϭ 14.4 Hz, iPr: CH), 19.9 (d, 2JC,P ϭ 27.2 Hz, iPr: CH3),
General: All operations were performed under an atmosphere of
dry argon by Schlenk and vacuum techniques. Solvents were dried
by standard methods and distilled prior to use.
2
2
18.9 (d, JC,P ϭ 3.7 Hz, iPr: CH3), 18.7 (d, JC,P ϭ 6.3 Hz, iPr:
CH3), 17.4 (d, JC,P ϭ 3.7 Hz, iPr: CH3) ppm. 31P NMR ([D8]tol-
2
uene, 81 MHz): δ ϭ 74.5 (d, 2JP,P ϭ 31.0 Hz, PiPr2), 53.7 (d, 2JP,P ϭ
31.0 Hz, PPh3) ppm. MS (EI): m/z (%) ϭ 642 (100) [M]ϩ.
C19H28BMoP (642.7): calcd. C 69.14, H 6.27; found C 70.45, H
6.51.
Dimethylanilinium tetrafluoroborate was prepared by published
procedures.[33] Full details of the preparation of 1 and 2, together
with the spectroscopic and structural properties of all reaction in-
termediates, will be presented elsewhere.[13] Elemental analyses (C,
H N) were performed on a Heraeus CHNS-Rapid elemental ana-
lyzer. EI and ESI mass spectra were recorded on a Varian MAT
212 or on a Micromass Quattro LCZ mass spectrometer, respect-
ively. 1H and 13C NMR spectra were measured on Bruker AC 200,
Bruker AMX 400, or Varian U 600 spectrometers with the solvent
as internal standard, whereas 31P NMR measurements were run on
a Bruker AC 200 spectrometer with aqueous H3PO4 (85%) as an
external reference. IR spectra were recorded on a Bruker Vector 22
instrument. The assignment of all 1H and 13C NMR resonances
was supported by two-dimensional NMR spectroscopy (COSY and
NOE experiments). For the atomic numbering schemes used in the
Exp. Sect., see Figure 2.
[(o-iPr2PC6H4-η7-C7H6)Mo(XyNC)2(Mo؊P)]BPh4 (5)BPh4: A so-
lution of 3 (250 mg, 0.63 mmol) in THF (50 mL) was treated at
room temperature with [HNMe2Ph][BPh4] (355 mg, 0.76 mmol).
After addition of an excess of 2,6-dimethylphenyl isocyanide, the
reaction mixture was subsequently stirred at ambient temperature
for 12 h. After evaporation of the solvent, the residue was extracted
with a small amount of dichloromethane and added dropwise to
rapidly stirred diethyl ether. Purification by recrystallization from
dichloromethane/diethyl ether at 0 °C was possible, affording air-
stable, brown crystals. Yield: 472 mg (78%). 1H NMR (CD2Cl2,
600 MHz): δ ϭ 7.81 (dm, 1 H, C6H4), 7.68 (tm, 1 H, C6H4), 7.64
(m, 2 H, C6H4), 7.41 (br. m, 8 H, o-BPh4), 7.28 (t, 2 H, p-
C6H3Me2), 7.22 (d, 4 H, m-C6H3Me2), 7.08 (t, 8 H, m-BPh4), 6.93
(t, 4 H, p-BPh4), 5.50 (m, 4 H, C7H6), 5.04 (dd, 2 H, C7H6), 2.55
(m, 2 H, iPr: CH), 2.44 (s, 12 H, o-CH3), 1.21 (dd, 6 H, iPr: CH3),
1.11 (dd, 6 H, iPr: CH3) ppm. 13C NMR (CD2Cl2, 150.9 MHz):
[(o-iPr2PC6H4-η7-C7H6)Mo(η2-BH4)(Mo؊P)] (3): A solution of 2
(500 mg, 0.93 mmol) in ethanol was treated at 0 °C with NaBH4
(250 mg, 6.52 mmol) and stirred for 3 h at ambient temperature.
The solvent was removed in vacuo, and the residue was extracted
with diethyl ether/hexane (1:1). After evaporation of the solvent, 3
could be isolated as a light green compound. Purification by recrys-
tallization from diethyl ether/hexane was possible, affording green
crystals of 3. Yield: 342 mg (91%). 1H NMR ([D8]toluene,
600 MHz): δ ϭ 7.56 (dm, 1 H, C6H4), 7.31 (tm, 1 H, C6H4), 7.19
(m, 2 H, C6H4), 5.48 (m, 2 H, C7H6), 4.49 (m, 4 H, C7H6), 1.86
(sept, 1 H, iPr: CH), 1.84 (sept, 1 H, iPr: CH), 0.72 (dd, 6 H, iPr:
CH3), 0.64 (dd, 6 H, iPr: CH3), Ϫ5.72 (br. s, 4 H, BH4) ppm. 13C
NMR ([D8]toluene, 150.9 MHz): δ ϭ 155.2 (d, 2JC,P ϭ 24.8 Hz, C-
2
1
δ ϭ 177.8 (d, JC,P ϭ 16.6 Hz, CNR), 164.4 (q, JC,B ϭ 49.4 Hz,
2
1
ipso-BPh4), 150.2 (d, JC,P ϭ 25.9 Hz, C-8), 137.8 (d, JC,P
ϭ
35.9 Hz, C-9), 136.3 (s, o-BPh4), 134.8 (s, o-C6H3Me2), 131.6 (d,
JC,P ϭ 2.6 Hz, C6H4), 130.7 (s, C6H4), 129.5 (s, p-C6H3Me2), 129.4
(d, JC,P ϭ 5.1 Hz, C6H4), 128.7 (s, m-C6H3Me2), 127.5 (d, JC,P
ϭ
3
9.3 Hz, C6H4), 127.3 (s, ipso-C6H3Me2), 125.9 (q, JC,B ϭ 2.6 Hz,
3
m-BPh4), 122.0 (s, p-BPh4), 118.0 (d, JC,P ϭ 2.5 Hz, C-1), 92.9 (s,
2
C7H6), 91.1 (d, JC,P ϭ 3.9 Hz, C7H6), 91.0 (s, C7H6), 26.2 (d,
2
1JC,P ϭ 21.1 Hz, iPr: CH), 19.2 (s, C6H3Me2), 18.6 (d, JC,P
ϭ
1
5.1 Hz, iPr: CH3), 18.1 (s, iPr: CH3) ppm. 31P NMR (CD2Cl2,
81 MHz): δ ϭ 75.7 ppm. MS (ESI): m/z (%) ϭ 641 (100) [M Ϫ
8), 138.7 (d, JC,P ϭ 33.8 Hz, C-9), 130.2 (br. s, C6H4), 129.6 (br.
3
2
s, C6H4), 114.2 (d, JC,P ϭ 2.6 Hz, C-1), 90.3 (d, JC,P ϭ 5.4 Hz,
2
BPh4]ϩ. IR (CH2Cl2): ν˜ ϭ 2099 ν(CN), 2067 ν(CN) cmϪ1
.
C7H6), 85.8 (d, JC,P ϭ 2.0 Hz, C7H6), 76.6 (s, C7H6), 25.0 (d,
2
C61H62BMoN2P (960.9): calcd. C 76.25, H 6.50, N 2.92; found C
75.59, H 6.02, N 2.37.
1JC,P ϭ 17.2 Hz, iPr: CH), 19.1 (d, JC,P ϭ 7.0 Hz, iPr: CH3), 19.0
(d, JC,P ϭ 1.5 Hz, iPr: CH3) ppm. 31P NMR ([D8]toluene,
2
81 MHz): δ ϭ 68.0 ppm. MS (EI): m/z (%) ϭ 394 (100) [M]ϩ, 379
(80) [Mϩ Ϫ BH4]. C19H28BMoP (394.1): calcd. C 57.90, H 7.16;
found C 57.65, H 6.77.
[(o-iPr2PC6H4-η7-C7H6)Mo(η4-NBD)(Mo؊P)]BPh4
(6)BPh4:
Compound (6)BPh4 could be prepared in a manner similar to that
described for (5)BPh4, by treatment of a solution of 3 (157 mg,
0.40 mmol) in THF with [HNMe2Ph][BPh4] (175 mg, 0.40 mmol),
[(o-iPr2PC6H4-η7-C7H6)Mo(PPh3)H(Mo؊P)] (4): A solution of 2
(1.184 g, 2.20 mmol) and PPh3 (576 mg, 2.20 mmol) in THF was followed by addition of norbornadiene (44 mg, 0.48 mmol). Yield:
treated with NaH (421 mg, 17.54 mmol) and stirred for 12 h. The
reaction mixture was filtered, and the solvent was removed by evap-
oration. The crude product was recrystallized from hexane to af-
258 mg (82%). 1H NMR (CD2Cl2, 600 MHz): δ ϭ 7.80 (d, 1 H,
C6H4), 7.74 (dd, 1 H, C6H4), 7.45 (t, 1 H, C6H4), 7.40 (dm, 1 H,
C6H4), 7.34 (br. m, 8 H, o-BPh4), 7.03 (t, 8 H, m-BPh4), 6.88 (t, 4
H, p-BPh4), 5.66 (m, 2 H, C7H6), 5.22 (m, 2 H, CϭCH), 4.93 (m,
1
ford 4 as a brown crystalline solid. Yield: 848 mg (60%). H NMR
([D8]toluene, 600 MHz): δ ϭ 7.74 (t, 3 H, PPh3), 7.52 (dm, 1 H, 2 H, C7H6), 4.85 (dd, 2 H, C7H6), 3.52 (m, 2 H, CϭCH), 3.43 (br.
C6H4), 7.32 (m, 6 H, PPh3), 7.16 (m, 1 H, C6H4), 7.05 (m, 1 H, s, 1 H, CH), 3.27 (br. s, 1 H; CH), 2.92 (m, 2 H, iPr: CH), 1.29
C6H4), 7.04 (m, 6 H, PPh3), 6.98 (t, 1 H, C6H4), 5.57 (m, 1 H, (dd, 6 H, iPr: CH3), 1.23 (dd, 6 H, iPr: CH3), 1.02 (s, 2 H, CH2)
C7H6), 5.29 (m, 1 H, C7H6), 4.78 (t, 1 H, C7H6), 4.52 (q, 1 H, ppm. 13C NMR (CD2Cl2, 150.9 MHz): δ ϭ 164.9 (br. s, ipso-BPh4),
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098
1095

M. Tamm, B. Dreßel, T. Lügger, R. Fröhlich, S. Grimme
FULL PAPER
136.3 (q, 2JC,B ϭ 1.3 Hz, o-BPh4), 135.0 (s, C6H4), 131.6 (d, 2JC,P ϭ
2.0 Hz, C-8), 131.3 (s, C-1), 129.9 (s, C6H4), 129.4 (d, JC,P ϭ 4.6 ethyl ether/hexane was possible, affording green crystals of 9. Yield:
light green compound. Purification by recrystallization from di-
1
1
Hz, C-9), 128.2 (s, C6H4), 126.9 (d, JC,P ϭ 9.1 Hz, C6H4), 126.0 (q, 165 mg (36%). H NMR ([D6]benzene, 200 MHz): δ ϭ 7.34 (m, 1
2
3JC,B ϭ 3.2 Hz, m-BPh4), 122.1 (s, p-BPh4), 93.4 (d, JC,P ϭ 2.0 H, C6H4), 7.12Ϫ7.00 (m, 2 H, C6H4), 6.86Ϫ6.72 (m, 6 H, C6H4 ϩ
2
Hz, C7H6), 90.6 (d, JC,P ϭ 1.3 Hz, C7H6), 88.3 (s, C7H6), 75.6 (s,
C6H5), 6.65 (m, 5 H, C6H4ϩC6H5), 5.12 (dd, 2 H, C7H6), 4.40 (dd,
NBD), 71.3 (s, NBD), 61.1 (s, NBD), 50.5 (d, JC,P ϭ 2.5 Hz, NBD), 2 H, C7H6), 4.18 (tm, 2 H, C7H6), Ϫ5.56 (br. s, 4 H, BH4) ppm.
1
2
47.9 (s, NBD), 28.1 (d, JC,P ϭ 18.2 Hz, iPr: CH), 19.0 (d, JC,P
ϭ
31P NMR ([D6]benzene, 81 MHz): δ ϭ 52.6 ppm.
4.2 Hz, iPr: CH3), 18.2 (s, iPr: CH3) ppm. 31P NMR (CD2Cl2, 81
MHz): δ ϭ 66.5 ppm. MS (ESI): m/z (%) ϭ 471 (100) [M Ϫ BPh4]ϩ.
C50H52BMoP (790.7): calcd. C 75.95, H 6.63; found C 76.30, H
6.24.
[(o-Ph2PC6H4-η7-C7H6)Mo(η2-HCCPh)(Mo؊P)]BPh4 (10)BPh4:
Compound (10)BPh4 could be prepared in a manner similar to that
described for (5)BPh4, by treatment of a solution of 9 (165 mg,
0.36 mmol) in THF with [HNMe2Ph][BPh4] (188 mg, 0.43 mmol),
followed by the addition of large excess of phenylacetylene. Yield:
186 mg (60%). 1H NMR (CD2Cl2, 200 MHz): δ ϭ 10.08 (d, 3JH,P ϭ
34.2 Hz, 1 H, CϵCH), 7.86Ϫ7.41 (br. m, 14 H, C6H4 ϩ PC6H5),
7.31 (br. m, 8 H, o-BPh4), 6.98 (br. t, 8 H, m-BPh4), 6.83 (br. t, 4
H, p-BPh4), 5.86 (m, 2 H, C7H6), 5.60 (m, 2 H, C7H6), 5.25 (m, 2
H, C7H6) ppm.31P NMR ([D6]benzene, 81 MHz): δ ϭ 95.1 ppm.
[(o-iPr2PC6H4-η7-C7H6)Mo(η2-HCCPh)(Mo؊P)]BPh4
(7)BPh4:
Compound (7)BPh4 could be prepared in a manner similar to that
described for (5)BPh4, by treatment of a solution of 3 (397 mg,
1.01 mmol) in THF with [HNMe2Ph][BPh4] (532 mg, 1.21 mmol),
followed by the addition of phenylacetylene. Yield: 576 mg (72%).
1H NMR (CD2Cl2, 600 MHz): δ ϭ 10.41 (d, 3JH,P ϭ 29.4 Hz, 1 H,
CϵC), 7.77Ϫ7.45 (m, 9 H, C6H4 ϩ BPh4), 7.34 (br. m, 8 H, o-
BPh4), 7.01 (t, 8 H, m-BPh4), 6.87 (t, 4 H, p-BPh4), 5.74 (dd, 2 H,
C7H6), 5.54 (dd, 2 H, C7H6), 5.30 (m, 2 H, C7H6, 2.83 (m, 2 H,
iPr: CH), 1.22 (dd, 6 H, iPr: CH3), 1.01 (dd, 6 H, iPr: CH3) ppm.
Catalytic Study: A solution of phenylacetylene (0.2 mL, 3.03 mmol)
in THF (10 mL) was treated with (7)BPh4 (128 mg, 0.16 mmol, 5
mol %), and the reaction mixture was stirred at 80 °C for 12 h.
The solvent was evaporated, and the residue was transferred to
a chromatography column (SiO2, 4% H2O). Elution with hexane
afforded a yellow oil (245 mg, 65%). GC/MS studies revealed that
the mixture contained a linear dimer (12%, m/z ϭ 204, presumably
one regioisomer of diphenyl-1,3-butenyne), 1,3,5-triphenylbenzene
(52%, m/z ϭ 306), a linear trimer (8%, m/z ϭ 306, presumably one
regioisomer of triphenylhex-1-yne-3,5-diene) and 1,2,4-triphenyl-
benzene (28%, m/z ϭ 306). Crystals of 1,3,5-triphenylbenzene were
isolated from the mixture and characterized by X-ray diffraction
analysis.[30] Subjection of these crystals to a GC/MS study con-
firmed the formation of 1,3,5-triphenylbenzene as the main prod-
uct.
2
13C NMR (CD2Cl2, 150.9 MHz): δ ϭ 178.5 (d, JC,P ϭ 28.4 Hz,
ϵCH), 173.2 (d, 2JC,P ϭ 4.5 Hz, ϵCPh), 164.4 (q, 1JC,B ϭ 49.9 Hz,
ipso-BPh4), 149.8 (d, 2JC,P ϭ 21.7 Hz, C-8), 136.3 (s, o-BPh4), 132.4
4
(d, JC,P ϭ 2.0 Hz, C-12), 131.7 (d, 1JC,P ϭ 38.2 Hz, C-9), 131.2 (s,
3
C-10), 130.6 (s, p-C6H5), 130.0 (d, JC,P ϭ 5.1 Hz, C-11), 129.6 (s,
3
m-C6H5), 129.5 (s, o-C6H5), 127.4 (d, JC,P ϭ 8.1 Hz, C-13), 126.0
3
2
(q, JC,B ϭ 2.6 Hz, m-BPh4), 122.1 (s, p-BPh4), 118.4 (d, JC,P
ϭ
2
2
2.6 Hz, C-1), 95.7 (d, JC,P ϭ 2.6 Hz, C-4,5), 94.7 (d, JC,P
ϭ
2.4 Hz, C-2,7), 82.9 (s, C-3,6), 27.6 (d, 23.0 Hz, iPr: CH), 19.8 (d,
2JC,P ϭ 4.0 Hz, iPr: CH3), 19.0 (s, iPr: CH3) ppm. 31P NMR
(CD2Cl2, 81 MHz): δ ϭ 72.4 ppm. MS (ESI): m/z (%) ϭ 481 (100)
[M Ϫ BPh4]ϩ. C51H50BMoP (800.7): calcd. C 76.50, H 6.29; found
C 75.50, H 5.93.
X-ray Crystallography:[34] Data sets were collected at Ϫ75 °C [for
4 and (5)BPh4] with an EnrafϪNonius KappaCCD or at Ϫ120 °C
[for 3, (6)BPh4, (7)BPh4, and (10)BPh4] with a Bruker AXS APEX
diffractometer, both equipped with a rotating anode and both using
[(o-iPr2PC6H4-η7-C7H6)Mo(η2-HCC-tBu)(Mo؊P)]BPh4 (8)BPh4:
Compound (8)BPh4 could be prepared in a manner similar to that
described for (5)BPh4, by treatment of a solution of 3 (235 mg,
0.60 mmol) in THF with [HNMe2Ph][BPh4] (315 mg, 0.72 mmol),
˚
Mo-Kα radiation (λ ϭ 0.71073 A). Empirical absorption correction
with SORTAV[35] [for 4 and (5)BPh4] or SADABS[36] [for 3,
(6)BPh4, (7)BPh4, and (10)BPh4] was applied to the raw data.
Structure solution in all cases with SHELXS[37] and refinement
with SHELXL[38] with anisotropic thermal parameters for all
atoms. Hydrogen atoms were added to the structure models on
calculated positions and were refined as riding atoms [4, (5)BPh4]
or are unrefined (for (10)BPh4). Hydrogen atoms for 3, (6)BPh4,
and (7)BPh4 were located in the difference Fourier map and were
refined with isotropic thermal parameters. ORTEP[39] was used for
all drawings. Additional crystallographic data are listed in Table 3.
followed by the addition of tert-butylacetylene. Yield: 400 mg
3
(86%). 1H NMR (CD2Cl2, 600 MHz): δ ϭ 10.03 (d, JH,P
ϭ
33.0 Hz, 1 H, CϵCH), 7.70 (m, 1 H, C6H4), 7.65 (m, 2 H, C6H4),
7.59 (m, 1 H, C6H4), 7.33 (br. m, 8 H, o-BPh4), 7.01 (t, 8 H, m-
BPh4), 6.88 (t, 4 H, p-BPh4), 6.08 (dd, 2 H, C7H6), 5.49 (dd, 2 H,
C7H6), 5.32 (m, 2 H, C7H6), 2.75 (m, 2 H, iPr: CH), 1.44 (s, 9 H,
tBu), 1.16 (dd, 6 H, iPr: CH3), 0.94 (dd, 6 H, iPr: CH3) ppm. 13C
NMR (CD2Cl2, 150.9 MHz): δ ϭ 191.2 (s, ϵCtBu), 177.8 (d,
1
2JC,P ϭ 30.1 Hz, ϵCH), 164.3 (q, JC,B ϭ 50.5 Hz, ipso-BPh4),
149.7 (d, 2JC,P ϭ 22.4 Hz, C-8), 136.3 (s, o-BPh4), 132.2 (d, 4JC,P ϭ
1.9 Hz, C-12), 131.8 (d, 1JC,P ϭ 39.0 Hz, C-9), 131.1 (s, C-10), 129.9
Computational Details: All quantum chemical calculations were
performed with the TURBOMOLE suite of programs.[40] The
phenylacetylene complexes have been fully optimized at the density
functional (DFT) level employing the BP86 functional[41] and
Gaussian AO basis sets of valence-triple-ξ quality including
polarization functions (TZVP, C:[5s3p1d], H:[3s1p], P:[5s4p1d]).[42]
For the metals the quasi-relativistic pseudo potentials of the
Stuttgart group with a 28 (Mo, Ru) and 60 (Os) electron core, re-
spectively, are used.[43] The corresponding optimized AO basis sets
(d, 3JC,P ϭ 5.3 Hz, C-11), 127.3 (d, 3JC,P ϭ 7.9 Hz, C-13), 125.9 (q,
2
3JC,B ϭ 2.6 Hz, m-BPh4), 122.1 (s, p-BPh4), 118.4 (d, JC,P
ϭ
2
2
2.5 Hz, C-1), 94.6 (d, JC,P ϭ 2.5 Hz, C-4,5), 94.2 (d, JC,P
ϭ
1.9 Hz, C-2,7), 81.6 (s, C-3,6), 31.6 (s, tBu), 27.4 (d, 23.6 Hz, iPr:
2
CH), 19.7 (d, JC,P ϭ 3.9 Hz, iPr: CH3), 18.9 (s, iPr: CH3) ppm.
31P NMR (CD2Cl2, 81 MHz): δ ϭ 72.9 ppm. C49H54BMoP (780.7):
calcd. C 75.38, H 6.97; found C 75.63, H 6.48.
[(o-Ph2PC6H4-η7-C7H6)Mo(η2-BH4)(Mo؊P)] (9): A solution of (Mo, Ru: [5s3p3d1f], Os: [6s3p3d1f]) have been taken from the
[(o-Ph2PC6H4-η7-C7H6)MoBr2(MoϪP)] (500 mg, 0.72 mmol) in TURBOMOLE library.[44] The calculated dissociation energies
ethanol was treated at 0 °C with NaBH4 (250 mg, 6.52 mmol) and have been obtained by freezing the metal fragment without the
stirred for 1 h at ambient temperature. The solvent was removed phenylacetylene ligand in the optimized geometry of the complex
by evaporation, and the residue was extracted with diethyl ether/
hexane (1:1). After evaporation of the solvent, 9 was isolated as a
(vertical De). Comparative calculations have also been performed
with the popular B3LYP hybrid functional.[45]
1096
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098

Chelate Complexes of Functionalized Cycloheptatrienyl Ligands
FULL PAPER
Table 3. Crystallographic data for 3, 4, (5)BPh4·C5H12, (6)BPh4·CH2Cl2, (7)BPh4, and (10)BPh4·C6H14
3
4
(5)BPh4·C5H12
(6)BPh4·CH2Cl2 (7)BPh4
C51H54BCl2MoP C51H50BMoP
(10)BPh4·3C4H8O
Empirical formula C19H28BMoP
Formula mass, amu 394.13
C37H40MoP2
642.57
17.635(1)
8.681(1)
21.649(1)
90.00
107.65(1)
90.00
3158.2(4)
1.351
monoclinic
P21/c
4
0.541
7965
C66H74BMoN2P
1032.99
21.534(1)
13.622(1)
21.592(1)
90.0
118.55(1)
90.0
5563.5(5)
1.233
monoclinic
P21/n
4
0.306
9754
C69H70BMoO3P
1084.97
11.1097(5)
21.2852(10)
23.4095(11)
90.00
96.1750(10)
90.00
5503.6(4)
1.309
monoclinic
Cc
4
0.316
12468
875.56
800.63
19.1772(10)
19.0903(10)
22.2170(12)
90.0
90.0
90.0
˚
a, A
22.2257(11)
22.2257(11)
7.4034(5)
90.0
90.0
90.0
3657.1(4)
1.432
orthorhombic
I4bar
10.0047(8)
14.6763(12)
14.8010(12)
82.037(2)
84.542(2)
82.248(2)
2126.2(3)
1.368
˚
b, A
˚
c, A
α, deg
β, deg
γ, deg
3
˚
V, A
8133.6(7)
1.308
dcalcd., g cmϪ1
Crystal system
Space group
Z
triclinic
P1
orthorhombic
Pbca
¯
8
2
8
µ, mmϪ1
0.799
11271
10309
0.507
5549
4631
0.397
5308
4130
Unique data
observed data
{I Ͼ 2σ(I)}
6685
5162
11347
R1 (obsd. data),% 2.57, wR1 ϭ 5.48 3.23, wR1 ϭ 7.61 5.98, wR1 ϭ 10.42 3.78, wR1 ϭ 8.86 2.89, wR1 ϭ 6.35 4.43, wR1 ϭ 10.97
R2 (all data),%
GoF
2.88, wR2 ϭ 5.53 4.35, wR2 ϭ 8.11 14.48, wR2 ϭ 12.86 5.00, wR2 ϭ 9.48 4.61, wR2 ϭ 7.09 4.96, wR2 ϭ 11.24
0.949
311
1.037
369
0.990
650
1.033
713
1.013
687
1.045
650
No. of variables
Res. electron
0.667 /Ϫ0.873
0.410 /Ϫ0.509
0.560 /Ϫ0.430
0.473 /Ϫ0.435
0.470/Ϫ0.204
1.462/Ϫ1.529
3
˚
density, e/ A
[11]
[12]
M. Tamm, K. Baum, T. Lügger, R. Fröhlich, K. Bergander,
Eur. J. Inorg. Chem. 2002, 918Ϫ928.
M. Tamm, B. Dreßel, V. Urban, T. Lügger, Inorg. Chem. Com-
mun. 2002, 5, 837Ϫ840.
M. Tamm, B. Dreßel, K. Baum, T. Lügger, T. Pape, J. Or-
ganomet. Chem., submitted.
H. Suzuki, D. H. Lee, N. Oshima, Y. Moro-oka, Organometal-
lics 1987, 6, 1569Ϫ1575.
Acknowledgments
This work was financially supported by the Deutsche Forschungs-
gemeinschaft, the Fonds der Chemischen Industrie, and the govern-
ment of North Rhine-Westphalia (Bennigsen-Foerder-Preis 2000 to
M. T.). We are indebted to Dr. Klaus Bergander for carrying out
valuable NMR experiments.
[13]
[14]
[15]
S. L. J. Conway, L. H. Doerrer, M. L. H. Green, M. A. Leech,
Organometallics 2000, 19, 630Ϫ637, and references cited
therein.
[1]
M. I. Bruce, Chem. Rev. 1998, 98, 2797Ϫ2858.
[2]
B. M. Trost, F. D. Toste, A. B. Pinkerton, Chem. Rev. 2001,
[16]
101, 2067Ϫ2096.
M. Elian, M. M. L: Chen, D. M. P. Mingos, R. Hoffmann,
Inorg. Chem. 1976, 15, 1148Ϫ1155.
[3]
For selected publications, see: [3a] B. M. Trost, A. B. Pinkerton,
[17] [17a]
[17b]
F. D. Toste, M. Sperrle, J. Am. Chem. Soc. 2001, 123,
H. Brunner, Eur. J. Inorg. Chem. 2001, 905Ϫ912.
H.
[3b]
12504Ϫ12509.
M. I. Bruce, B. C. Hall, B. W. Skelton, A.
Brunner, Angew. Chem. 1999, 111, 1248Ϫ1263; Angew. Chem.
H. White, N. N. Zaitseva, J. Chem. Soc., Dalton Trans. 2000,
Int. Ed. 1999, 38, 1194Ϫ1209.
[3c]
[18]
[19]
2279Ϫ2286.
E. Rüba, W. Simanko, K. Mauthner, K. M.
H. Brunner, M. Muschiol, I. Bernal, G. M. Reisner, J. Or-
ganomet. Chem. 1980, 198, 169Ϫ178.
For a related chiral ruthenium complex of the type [(η-
C5Me5)Ru(PPh3)(PR3)H] (R ϭ OMe), see: T. Arliguie, B. chau-
dret, F. A. Jalon, A. Otero, J. A. Lopez, F. J. Lahoz, Organo-
metallics 1991, 10, 1888Ϫ1896.
M. L. H. Green, R. B. A. Pardy, Polyhedron 1985, 4,
1035Ϫ1041.
E. Schwed, H.-U. Hund, H. W. Bosch, H. Berke, Helv. Chim.
Acta 1991, 74, 189Ϫ196.
Soldouzi, C. Slugovc, K. Mereiter, R. Schmid, K. Kirchner,
Organometallics 1999, 18, 3843Ϫ3850. [3d] B. M. Tost, B. Vidal,
M. Thommen, Chem. Eur. J. 1999, 1055Ϫ1069. [3e] C. S. Yi, N.
Liu, Organometallics 1998, 17, 3158Ϫ3160. [3f] C. S. Yi, N. Liu,
Organometallics 1996, 15, 3968Ϫ3971. [3g] T. Braun, O. Gevert,
H. Werner, J. Am. Chem. Soc. 1995, 117, 7291Ϫ7292.
[20]
[21]
[22]
[4]
M. L. H. Green, D. K. P. Ng, Chem. Rev. 1995, 95, 439Ϫ473.
[5] [5a]
C. E. Davies, I. M. Gardiner, J. C. Green, M. L. H. Green,
N. J. Hazel, P. D. Grebenik, V. S. B. Mtetwa, K. Prout, J. Chem.
Soc., Dalton Trans. 1985, 669Ϫ683. [5b] J. S. Adams, C. Bitcon,
Comprehensive Organometallic Chemistry II (Eds.: E. W. Abel,
F. G. A. Stone, G. Wilkinson), vol. 7 (Vol. Eds. D. F. Shriver,
M. I. Bruce), Chapman and Hall, Oxford, 1995.
G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew. Chem.
1999, 111, 448Ϫ468; Angew. Chem. Int. Ed. 1999, 38, 428Ϫ447.
J. Chatt, C. M. Elson, A. J. L. Pombeiro, R. L. Richards, G.
H. D. Royston, J. Chem. Soc., Dalton Trans. 1978, 165Ϫ169.
H. Seino, C. Arita, D. Nonokawa, G. Nakamura, Y. Harada,
Y. Mizobe, M. Hidai, Organometallics 1999, 18, 4165Ϫ4173.
J. R. Brown, D. Collison, M. Cunningham, M. W. Whiteley, J.
[5c]
Chem. Soc., Dalton Trans. 1987, 3049Ϫ3053.
C. Bitcon, R.
[23]
[24]
[25]
[26]
Breeze, P. F. Miller, M. W. Whiteley, J. Organomet. Chem. 1989,
[5d]
364, 181Ϫ193.
R. L. Beddoes, R. W. Grime, Z. I. Hussain,
M. W. Whiteley, J. Chem. Soc., Dalton Trans. 1996, 3893Ϫ3895.
[6] [6a]
J. Okuda, T. Eberle in Metallocenes (Eds.: A. Togni, R. L.
Halterman), Wiley-VCH, Weinheim, 1998, Vol. 1, p. 415. [6b] J.
Okuda, Comment, Inorg. Chem. 1994, 16, 185.
[7]
[8]
P. Jutzi, T. Redeker, Eur. J. Inorg. Chem. 1998, 663Ϫ674.
U. Siemeling, Chem. Rev. 2000, 100, 1495Ϫ1526.
H. Butenschön, Chem. Rev. 2000, 100, 1527Ϫ1564.
M. Tamm, T. Bannenberg, B. Dreßel, R. Fröhlich, C. Holst,
Inorg. Chem. 2002, 41, 47Ϫ59.
´
R. Uson, L. A. Oro, M. A. Ciriano, M. M. Naval, M. C. Ap-
reda, C. Foces-Foces, F. H. Cano, S. Garcia-Blanco, J. Or-
ganomet. Chem. 1983, 256, 331Ϫ347.
[9]
[10]
[27]
[28]
J. L. Templeton, Adv. Organomet. Chem. 1989, 29, 1Ϫ100.
´
´
J. J. Carbo, P. Crochet, M. A. Esteruelas, Y. Jean, Agustı
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098
1097

M. Tamm, B. Dreßel, T. Lügger, R. Fröhlich, S. Grimme
FULL PAPER
[36]
[37]
´
´
Lledos, A. M. Lopez, E. On˜ate, Organometallics 2002, 21,
SMART, Bruker AXS, 2000.
SHELXS-97, G. M. Sheldrick, Acta Crystallogr., Sect. A 1990,
46, 467Ϫ473.
305Ϫ314.
[29] [29a]
W. E. Douglas, J. Chem. Soc., Dalton Trans. 2000, 57Ϫ62.
M. G. L. Petrucci, A. K. Kakkar, J. Chem. Soc., Chem.
[29b]
[38]
[39]
[40]
SHELXL-97, G. M. Sheldrick, Universität Göttingen, 1997.
ORTEP-3, L. J. Farrugia, University of Glasgow, 1999.
R. Ahlrichs, M. Bär, H.-P. Baron, R. Bauernschmitt, S. Böcker,
M. Ehrig, K. Eichkorn, S. Elliott, F. Furche, F. Haase, M.
Häser, H. Horn, C. Huber, U. Huniar, M. Kattannek, C.
Kölmel, M. Kollwitz, K. May, C. Ochsenfeld, H. Öhm, A.
Schäfer, U. Schneider, O. Treutler, M. von Arnim, F. Weigend,
P. Weis, H. Weiss, TURBOMOLE (vers. 5.3), Universität
Karlsruhe, 2000.
[29c]
Commun. 1995, 1577Ϫ1578.
ringhausen-Bley, U. Flörke, H.-J. Haupt, Organometallics 1992,
H.-F. Klein, M. Mager, S. Is-
11, 3174Ϫ3175.
Crystal data for 1,3,5-triphenylbenzene: a ϭ 7.4779(3), b ϭ
19.7314(9), c ϭ 11.2217(5) A, Z ϭ 4, orthorhombic, space
group Pna21, final R ϭ 3.79% based on 2137 observed struc-
ture factors [I Ն 2 σ(I)].
L. S. Meriwether, M. F. Leto, E. C. Colthup, G. W. Kennerly,
J. Org. Chem. 1962, 27, 3930Ϫ3941.
G. Frenking, N. Fröhlich, Chem. Rev. 2000, 100, 717Ϫ774.
M. Haufe, R. D. Köhn, R. Weimann, G. Seifert, D. Zeigan, J.
Org. Chem. 1996, 520, 121Ϫ129.
CCDC-CCDC-183826, -183827, -183828, -183829, -192130,
and -192131 contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cam-
bridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB2 1EZ, UK; fax: (internat.) ϩ44-1223/336-033;
E-mail: deposit@ccdc.cam.ac.uk].
[30]
[31]
˚
[41] [41a]
[41b]
A. D. Becke, Phys. Rev. A 1988, 38, 3098Ϫ3100.
Perdew, Phys. Rev. B 1986, 33, 8822Ϫ8824.
J. P.
[32]
[33]
[42]
[43]
[44]
A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100,
5829Ϫ5835.
D. Andrae, U. Häussermann, M. Dolg, H. Stoll, H. Preuss,
Theor. Chim. Acta 1990, 77, 123Ϫ130.
[34]
The AO basis sets used are available at (anonymous)
ftp://ftp.chemie.uni-karlsruhe.de/pub/basen
[45] [45a]
[45b]
A. D. Becke, J. Chem. Phys. 1993, 98, 5648Ϫ5652.
P.
J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623Ϫ11627.
Received August 26, 2002
[35]
SORTAV, R. H. Blessing, J. Appl. Crystallogr. 1997, 30,
421Ϫ426.
[I02478]
1098
Eur. J. Inorg. Chem. 2003, 1088Ϫ1098