Contribution to the chemistry of metal complexes with stereogenic metal centers: Diastereoselective formation of ruthenium half-sandwich complexes

Article abstract of DOI:10.1021/om0010262

The chiral half-sandwich complexes 3 and 4 are formed with high diastereoselectivities in the reaction of cyclopentadienides 2a,b, bearing tethered phosphaferrocene donor moieties with planar chirality, and [(PPh₃)₃RuCl₂] in toluene at 90°C. The diastereoisomers of the Cp complex 3 are obtained in a 95:5 ratio, whereas for the Cp* derivative 4 only one isomer is detectable, the structure of which has been determined by X-ray diffraction. Substitution of the chloride ligand in 4 by other anionic (H^-, I^-) or neutral (H₂, py) ligands proceeds stereospecifically in all cases. In contrast, conversion of the chlorides 3a,b (95:5) to the respective hydrides 9a,b proceeds with complete epimerization at Ru. In CHCl₃ the 1:1 mixture of hydrides 9a,b is reconverted to the chlorides to give a kinetically controlled 4:1 mixture of isomers 3a,b. Equilibration of this mixture in toluene at 90°C restores the original ratio of isomers of 95:5, which we therefore believe to reflect the thermodynamically controlled value. The cationic H₂ complex 7, generated via Cl abstraction from 4 in the presence of H₂, was characterized to be a η²-dihydrogen complex by measuring the T₁ value of the coordinated H₂ ligand.

Full text of DOI:10.1021/om0010262

1614
Organometallics 2001, 20, 1614-1619
Con tr ibu tion to th e Ch em istr y of Meta l Com p lexes w ith
Ster eogen ic Meta l Cen ter s: Dia ster eoselective
F or m a tion of Ru th en iu m Ha lf-Sa n d w ich Com p lexes
Corinna Kaulen, Corinne Pala, Chunhua Hu, and Christian Ganter*
Institut fu¨r Anorganische Chemie, Technische Hochschule Aachen, D-52056 Aachen, Germany
Received December 4, 2000
The chiral half-sandwich complexes 3 and 4 are formed with high diastereoselectivities
in the reaction of cyclopentadienides 2a ,b, bearing tethered phosphaferrocene donor moieties
with planar chirality, and [(PPh3)3RuCl2] in toluene at 90 °C. The diastereoisomers of the
Cp complex 3 are obtained in a 95:5 ratio, whereas for the Cp* derivative 4 only one isomer
is detectable, the structure of which has been determined by X-ray diffraction. Substitution
of the chloride ligand in 4 by other anionic (H-, I-) or neutral (H2, py) ligands proceeds
stereospecifically in all cases. In contrast, conversion of the chlorides 3a ,b (95:5) to the
respective hydrides 9a ,b proceeds with complete epimerization at Ru. In CHCl3 the 1:1
mixture of hydrides 9a ,b is reconverted to the chlorides to give a kinetically controlled 4:1
mixture of isomers 3a ,b. Equilibration of this mixture in toluene at 90 °C restores the original
ratio of isomers of 95:5, which we therefore believe to reflect the thermodynamically controlled
value. The cationic H2 complex 7, generated via Cl abstraction from 4 in the presence of H2,
was characterized to be a η2-dihydrogen complex by measuring the T1 value of the coordinated
H2 ligand.
Sch em e 1
In tr od u ction
The selective formation of chiral nonracemic metal
complexes is a challenging goal in coordination chem-
istry with potential implications for enantioselective
catalysis. The proper choice of chiral ligands plays a
crucial role in the “predetermination of metal chirality”,
a term recently coined by von Zelewsky.1 One topic that
has received increasing attention over the last few years
is the formation of chiral half-sandwich complexes, in
which a cyclopentadienyl ligand is tethered to an
additional donor function.2 Several attempts have been
made to introduce an element of chirality into this type
of Cp-L chelate ligand in order to enable a diastereo-
selective complexation reaction, and ruthenium com-
plexes Cp-LRu(PR3)X have been studied in particular
detail.3 In this contribution we wish to report our results
in the stereoselective formation of ruthenium complexes
using the chiral Cp-P chelate ligands 2 (Scheme 1).
project to explore the coordination chemistry of the
cyclopentadienyl anion 2a , which is easily prepared
as its Na salt in a two-step procedure starting from
aldehyde 1a , and the formation of the homoleptic fer-
rocene from 2a and FeCl2 was recently reported.5 For
the preparation of half-sandwich complexes we chose
[(PPh3)3RuCl2] as a starting material in order to obtain
the half-sandwich complex [Ru(η5:η1-2a )(PPh3)Cl] (3).
An equimolar mixture of 2a and [(PPh3)3RuCl2] was
Resu lts a n d Discu ssion
(3) (a) Kataoka, Y.; Saito, Y.; Nagata, K.; Kitamura, K.; Shibahara,
A.; Tani, K. Chem. Lett. 1995, 833-834. (b) Nishibayashi, Y.; Takei,
I.; Hidai, M. Organometallics 1997, 16, 3091-3093. (c) van der Zeijden,
A. A. H.; J imenez, J .; Mattheis, C.; Wagner, C.; Merzweiler, K. Eur.
J . Inorg. Chem. 1999, 1919-1930. (d) Kataoka, Y.; Iwato, Y.; Yama-
gata, T.; Tani, K. Organometallics 1999, 18, 5423-5425. (e) Trost, B.
M.; Vidal, B.; Thommen, M. Chem. Eur. J . 1999, 5, 1055-1069. (f)
Brunner, H.; Valerio, C.; Zabel, M. New J . Chem. 2000, 24, 275-279.
(g) Dodo, N.; Matsushima, Y.; Uno, M.; Onitsuka, K.; Takahashi, S. J .
Chem. Soc., Dalton Trans. 2000, 35-41. (h) For a general review on
chiral Ru half-sandwich complexes see: Consiglio, G.; Morandini, F.
Chem. Rev. 1987, 87, 761-778.
(4) (a) Ganter, C.; Brassat, L.; Glinsbo¨ckel, C.; Ganter, B. Organo-
metallics 1997, 16, 2862-2867. (b) Ganter, C.; Brassat, L.; Ganter, B.
Chem. Ber. 1997, 130, 1771-1776. (c) Ganter, C.; Glinsbo¨ckel, C.;
Ganter, B. Eur. J . Inorg. Chem. 1998, 1163-1168. (d) Brassat, L.;
Ganter, B.; Ganter, C. Chem. Eur. J . 1998, 4, 2148-2153.
(5) Ganter, C.; Kaulen, C.; Englert, U. Organometallics 1999, 18,
5444-5446.
Syn th esis of Ha lf-Sa n d w ich Com p lexes 3 a n d 4.
On the basis of our previous work on chiral phospha-
ferrocene chelate ligands,4 we have launched a research
* To whom correspondence should be addressed. E-mail:
christian.ganter@ac.rwth-aachen.de.
(1) (a) Knof, U.; von Zelewsky, A. Angew. Chem. 1999, 111, 312-
333; Angew. Chem., Int. Ed. 1999, 38, 302-32. (b) von Zelewsky, A.;
Mamula, O. J . Chem. Soc., Dalton Trans. 2000, 219-231. (c) See also:
Brunner, H. Angew. Chem. 1999, 111, 1248-1263; Angew. Chem., Int.
Ed. 1999, 38, 1194-1208.
(2) (a) For a recent comprehensive review focused on P-donors see:
Butenscho¨n, H. Chem. Rev. 2000, 100, 1527-1564. (b) Siemeling, U.
Chem. Rev. 2000, 100, 1495-1526. (c) Mu¨ller, C.; Vos, D.; J utzi, P. J .
Organomet. Chem. 2000, 600, 127-143. (d) Okuda, J . Comments Inorg.
Chem. 1994, 16, 185-205.
10.1021/om0010262 CCC: $20.00 © 2001 American Chemical Society
Publication on Web 03/17/2001

Ruthenium Half-Sandwich Complexes
Organometallics, Vol. 20, No. 8, 2001 1615
to the thermodynamically more stable diastereomer,
because models indicate a severely unfavorable interac-
tion between the CpFe fragment and the PPh3 ligand
in the diastereomeric complex 3b.
Next, we turned our attention to possible modifica-
tions of our ligand system to further improve the
stereoselectivity of the complexation reaction. On the
basis of the steric argument given above, substitution
of the Cp for a sterically more demanding Cp* ligand
in the phosphaferrocene seemed promising for this
purpose. The Cp*-modified cyclopentadienide 2b was
prepared in a similar manner from the corresponding
aldehyde 1b, which was obtained according to a proce-
dure closely related to those reported by Mathey7 and
Fu.8 The complexation reaction was carried out under
the same conditions as reported above, and the 31P NMR
spectrum of the crude mixture indicated the formation
of only one diastereomeric complex, which was obtained
in crystalline form in 68% isolated yield after chroma-
tography. The diastereomeric purity of complex 4 can
be regarded as de > 98%. Thus, a virtually complete
control of the metal configuration is exerted by the
ligand-based chirality. To the best of our knowledge, this
is the highest selectivity observed so far for this kind of
complexation reaction, leading to complexes of the type
[Cp-LRu(PR3)X]. Similar complexation experiments
with the ligand systems reported in ref 3 gave de values
in the range of 18-83%. Although Hidai and co-workers
reported the synthesis of a diastereomerically pure
complex,3b this result was obtained after recrystalliza-
tion of the crude product and does therefore not allow
conclusions to be drawn regarding the selectivity of the
complexation reaction itself.
F igu r e 1. Molecular structure of complex 4. Selected bond
lengths (Å) and angles (deg): Ru(1)-P(1) ) 2.2549(14), Ru-
(1)-P(2) ) 2.2912(14), Ru(1)-Cl(1) ) 2.4285(12); P(1)-Ru-
(1)-P(2) ) 94.45(5), P(1)-Ru(1)-Cl(1) ) 99.37(5), P(2)-
Ru(1)-Cl(1) ) 90.41(5), C(1)-P(1)-C(4) ) 90.4(3), C(1)-
P(1)-Ru(1) ) 108.3(2), C(4)-P(1)-Ru(1) ) 155.21(17),
C(6)-C(5)-C(1) ) 110.4(5).
Sch em e 2
An X-ray diffraction study allowed the unambiguous
assignment of the relative configuration of the two
elements of chirality. An ORTEP plot of the structure
is given in Figure 1, together with selected distances
and angles. Complex 4 crystallizes in the monoclinic
space group Pn with Z ) 2. The orientation of the PPh3
ligand and the Cp*Fe moiety are qualitatively the same
as in complex 3a , in accord with the above assumption
that increasing the steric demand of the CpFe fragment
leads to a higher selectivity in the complexation reac-
tion. The basic conformational feature of the structure
is that the phospholyl and Cp* rings of the phospha-
ferrocene moiety are almost perpendicular to the sub-
stituted Cp ring coordinated to the Ru atom. The Ru-P
bonds differ slightly in length, the distance Ru-P1
(225.5(1) pm) being shorter than the distance Ru-P2
(229.1(1) pm). This is in accord with our earlier observa-
tions that metal-phosphaferrocene bonds generally
tend to be shorter than metal-PR3 bonds in trialkyl-
or triarylphosphine complexes, although the difference
of ca. 4 pm in the present case is rather small.4b In the
related complex [CpCH2CH2PPh2Ru(PPh3)Cl] the bonds
between Ru and the electronically quite similar phos-
phorus atoms are 230.7(2) and 231.1(1) pm; i.e., they
are identical within the experimental error.9 For com-
parison, the Ru-P distances in the complex [CpRu-
(PPh3)2Cl] are 233.7(1) and 233.5(1) pm, respectively.10
heated in toluene at 90 °C overnight (Scheme 2). The
31P NMR spectrum of the crude reaction mixture
indicated complete conversion and showed a signal for
uncoordinated PPh3 and two pairs of doublets in a 95:5
ratio, attributable to the two diastereomeric complexes
3a ,b, which were isolated as red crystals in 63% yield
from the reaction mixture. The isomer ratio of 95:5 was
unaffected by the workup procedure; moreover, attempts
to separate the isomers by chromatography or fractional
crystallization were unsuccessful. The anticipated con-
nectivity of complex 3 was proven by an X-ray diffrac-
tion study carried out on a crystal selected from the 95:5
mixture of isomers.6 The observed relative configuration
of the central and planar elements of chirality in the
examined crystal is that depicted in formula 3a , which
results in a minimal interference of the PPh3 ligand and
the CpFe moiety of the phosphaferrocene. (The relative
configuration is identical with that found for the Cp*
analogue in complex 4, as depicted in Figure 1; see
below.) We believe that this arrangement corresponds
(6) The X-ray structure determination suffered from poor crystal
quality, leading to a low precision data set. However, the structure
solution is significant as far as the constitution and relative configu-
ration of the complex are concerned. Since the related complex 4 could
be characterized by X-ray diffraction with high accuracy, no details of
the structure determination for complex 3 will be discussed. The
assumption that the crystal of 3 investigated by X-ray diffraction
belongs to the major diastereomer served to us as a plausible working
hypothesis based on probability. The examined crystal was not
characterized by NMR spectroscopy.
(7) Mathey, F.; de Lauzon, G. In Organometallic Syntheses; King,
R. B., Eisch, J .-J ., Eds.; Elsevier: Amsterdam, 1986; pp 259-261.
(8) Garrett, C. E.; Fu, G. C. J . Org. Chem. 1997, 62, 4534-4535.
(9) Slawin, A. M. Z.; Williams, D. J .; Crosby, J .; Ramsden, J . A.;
White, C. J . Chem. Soc., Dalton Trans. 1988, 2491-2494.

1616 Organometallics, Vol. 20, No. 8, 2001
Kaulen et al.
Sch em e 3
Sch em e 4
Cl2, followed by addition of a ligand suitable to occupy
the vacant coordination site on the Ru atom. For
example, treatment of a solution of the cation with
hydrogen gas at normal pressure led to the formation
of the cationic dihydride complex 7 in quantitative yield,
Rea ctivity of Com p lexes 3 a n d 4. We have started
to explore the reactivity of complexes 3 and 4 with
respect to the configurational stability of the Ru center.
Results for the Cp* complex 4 are described first. The
exchange of the chloride ligand for iodide occurred
smoothly within a period of 36 h on stirring complex 4
with an excess of solid NaI in a THF/acetone mixture
(Scheme 3). The NMR spectra indicated a clean quan-
titative conversion with the formation of the single
diastereomeric iodide complex 5. Similarly, reaction of
the chloride 4 with an excess of NaOMe in refluxing
MeOH leads within 1 h to the quantitative formation
of the Ru-H complex 6, obtained again as a single
diastereomer. This reaction protocol is a known proce-
dure in Ru chemistry, and the reaction is believed to
occur by successive substitution of the Cl ligand by
methoxide, followed by â-hydrogen elimination and
release of formaldehyde.11 The reaction is accompanied
by a color change from orange-red to yellow. The most
1
as monitored by NMR spectroscopy (Scheme 4). The H
NMR spectrum features one broad resonance for the
metal-bound hydrogen atoms at δ -6.8 ppm, whereas
two doublets are observed in the 31P NMR spectrum.
Thus, again, complex 7 is formed as a single diastere-
omer. When solutions are evaporated in vacuo, complex
7 reversibly loses dihydrogen and the NMR spectra of
the residue in CD2Cl2 show broad and unspecific reso-
nances. When dihydrogen is bubbled through this CD2-
Cl2 solution, the dihydride complex 7 is quantitatively
regenerated. However, once the dihydride 7 is obtained
as a powdery solid by cooling a CH2Cl2 solution layered
with hexane, the H2 ligand is not liberated under
vacuum. Because of the sensitivity of complex 7 a
satisfactory elemental analysis could not be obtained;
hence, the compound was characterized only by NMR
methods in solution. Different results were observed for
the reaction of the cationic dihydride with triethylamine
and pyridine, respectively: the alkylamine acts as a
Brønsted base, abstracting a proton from the complex,
leading to the neutral monohydride complex 6 (formed
as a single diastereomer with the same NMR resonances
as described above). On the other hand, addition of
pyridine leads to the loss of H2 and formation of the
cationic pyridine complex 8 (single diastereomer) (Scheme
5). The different behaviors of amines and pyridine
toward cationic dihydride complexes have been de-
scribed in the literature.13 With the synthesis of complex
7 the question arose as to whether this compound has
to be considered as a classical dihydride or a η2-
dihydrogen complex, and the T1 relaxation time of the
hydride resonance was determined by the inversion
recovery technique in order to address this question. The
value of T1 ) 15 ms measured at room temperature in
CD2Cl2 is characteristic of dihydrogen complexes, which
generally feature short T1 values.13 The observation of
1
significant feature of the hydride complex 6 is the H
NMR resonance for the hydride proton, which appears
as a doublet of doublets due to the coupling to the two
inequivalent P nuclei at δ -9.54 ppm. The hydride
complex 6 turned out to be rather reactive, as workup
manipulations with chlorinated solvents or chromatog-
raphy led partially to the re-formation of the starting
chloride complex 4.12 The hydride to chloride transfor-
mation occurred quantitatively within 1 h when a C6D6
solution of the hydride complex was treated with a few
drops of CDCl3. The chloride complex 4 was formed as
a single diastereomer, which showed the same NMR
resonances as the starting material. Thus, the cycle
chloride-hydride-chloride proceeds with overall reten-
tion of configuration at the Ru center. This is in accord
with literature reports that most substitution reactions
of CpRuLL′Cl proceed with retention.3h
Further substitution reactions could be carried out
by treating the chloride complex 4 with TlPF6 in CH2-
(10) Bruce, M. I.; Wong, F. S.; Skelton, B. W.; White, A. H. J . Chem.
Soc., Dalton Trans. 1981, 1398-1405.
(11) (a) J ia, G.; Morris, R. H. J . Am. Chem. Soc. 1991, 113, 875-
883. (b) Wilczewski, T.; Bochenska, M.; Biernat, J . F. J . Organomet.
Chem. 1981, 215, 87. (c) Davies, S. G.; McNally, J . P.; Smallridge, A.
J . Adv. Organomet. Chem. 1990, 30, 1-76.
(12) For the reaction of hydride complexes with chlorinated solvents
see: (a) Kochi, J . K. Organometallic Mechanisms and Catalysis;
Academic Press: New York, 1978; p 190. (b) Chanon, M. Bull. Soc.
Chim. Fr. 1982, II-197.
1
1
a broad H NMR resonance with an unresolved H-31P
coupling is in accord with this classification, as is the
easy and reversible loss of the coordinated H2 molecule
from the complex in solution, although the latter
(13) Heinekey, D. M.; Oldham, W. J . Chem. Rev. 1993, 93, 913-
926.

Ruthenium Half-Sandwich Complexes
Organometallics, Vol. 20, No. 8, 2001 1617
Sch em e 5
Sch em e 6
criterion is of less diagnostic value.13 Quite a number
of complexes of the type CpRuL2H2+ have been prepared
and characterized. The nature of the RuH2 fragment,
i.e., the dihydride vs the η2-dihydrogen character, is
strongly dependent on the nature of the ligand L, and
many cases are known where both forms coexist in an
equilibrium. Strongly basic ligands such as alkylphos-
phines render the Ru center electron rich, and the
equilibrium usually lies on the dihydride side. As the
phosphines become less basic, the equilibrium is shifted
in favor of the η2-H2 complex. If one phosphine is
replaced by the strong π-acceptor ligand CO, the η2-H2
form is usually the only observed species.11a,14 Thus, the
fact that only this form is observed in the case of our
complex 7 is another nice confirmation of the π-acidic
character of the phosphaferrocene ligand, which has
been described earlier to behave more like a phosphite
P(OR)3 than a phosphine PR3 in terms of donor/acceptor
properties.15
All reactions of the Cp* derivative 4 described in the
preceding section have been carried out as well with the
sterically less demanding Cp complex 3. While consti-
tutionally similar products were obtained, the stereo-
chemical course of the reactions was different and
deserves a brief discussion. Specifically, when a 95:5
mixture of chlorides 3a ,b was treated with NaOMe in
MeOH as described above, the formation of the diaster-
eomeric hydrides 9a ,b was observed in a ratio of 1:1,
indicating complete epimerization at the Ru center. This
result is in striking contrast to the selectivity of the
analogous reaction of the Cp* derivative 4 (see above)
and to the literature report of Consiglio et al., who
observed a stereospecific formation of hydride complexes
from diastereomeric CpRuP*PCl complexes with reten-
tion of configuration at the Ru atom.16 The hydrides
9a ,b were converted back to the chlorides 3 with
interesting results (Scheme 6): a solution of the 1:1
mixture of hydrides in CDCl3 is transformed to the
chloride smoothly at room temperature within 2 h.
However, the ratio of diastereomers 3a :3b is 4:1 under
these conditions, as determined from the integrated 31P
NMR spectra. When the CDCl3 solution is evaporated
to dryness and the residue is dissolved in toluene and
heated to 90 °C for 2 h, the isomer ratio is shifted to
95:5, the value which was observed after the synthesis
of complex 3 according to Scheme 2. We therefore
conclude that the 95:5 ratio reflects the thermodynamic
equilibrium of the diastereomers, while the 4:1 result
of the hydride-chloride transformation is under kinetic
control. Consiglio et al. have reported the preparation
and configurational stability of CpRu(P*P)Cl complexes,
where P*P represents enantiopure, C1-symmetric diphos-
phines.16,17 The diastereomers were formed in almost
equal amounts under kinetic control, and isomerization
of the Ru centers to approach the thermodynamic ratio
of isomers was brought about by heating the mixture
in chlorobenzene to 80 °C for a period of several hours.
However, the highest thermodynamic ratio observed
was 2.4:1, which is in pronounced contrast to the high
degree of thermodynamic preference of the major dia-
(14) (a) Chinn, M. S.; Heinekey, D. M. J . Am. Chem. Soc. 1987, 109,
5865-5867. (b) Conroy-Lewis, F. M.; Simpson, S. J . J . Chem. Soc.,
Chem. Commun. 1986, 506-507.
(16) Morandini, F.; Consiglio, G.; Lucchini, V. Organometallics 1985,
4, 1202-1208.
(17) Morandini, F.; Consiglio, G.; Straub, B.; Ciani, G.; Sironi, A. J .
Chem. Soc., Dalton Trans. 1983, 2293-2298.
(15) Mathey, F. Coord. Chem. Rev. 1994, 137, 1.

1618 Organometallics, Vol. 20, No. 8, 2001
Kaulen et al.
CH), 83.7 (d, J (C,P) ) 8.2 Hz, Cp CH), 84.1 (s, Cp* C(q)), 86.5
(s, Cp C(q)), 87.3 (s, phospholyl R-C(q)), 104.9 (d, J (C,P) ) 7.7
Hz, phospholyl â-C), 116.3 (d, J (C,P) ) 9.0 Hz, phospholyl â-C),
127-137 (PPh3). 31P NMR (202 MHz, CDCl3): 3a , 23.4 (d,
2J (P,P) ) 64 Hz, phospholyl P), 48.0 (d, 2J (P,P) ) 64 Hz, PPh3);
stereomers of the chloro complexes 3 and 4 described
in this work.
On the basis of the result of the X-ray diffraction
analysis of complex 4, we anticipate the same relative
configuration at the Ru atom for all complexes of the
Cp* series, i.e., 5-8, as indicated in Schemes 3-5,
respectively. This assumption is corroborated by two
observations (vide supra): (1) the sequence 4-6-4 leads
back to the same diastereomer of 4 as the starting
material; (2) the same diastereomer of hydride 6 is
obtained either by deprotonation of the dihydrogen
complex 7 or by chloride-hydride exchange from com-
plex 4. We are trying to grow crystals of the respective
complexes suitable for X-ray diffraction in order to
substantiate this assumption. Furthermore, we are
currently exploring further reactions of the chloride
complexes 3a ,b and 4 which will hopefully allow us to
exploit the complex based chirality to induce some
stereoselective transformations of suitable substrates in
the coordination sphere of the Ru atom.
2
2
3b, 21.2 (d, J (P,P) ) 62 Hz, phospholyl P), 50.3 (d, J (P,P) )
2
62 Hz, PPh3); 4, 31.3 (d, J (P,P) ) 64 Hz, phospholyl P), 47.7
(d, 2J (P,P) ) 64 Hz, PPh3). SIMS for 3: m/e 708 (M+), 673 (M+
- Cl), 411 (M+ - Cl - PPh3). Anal. Calcd for C35H33ClP2FeRu
(708.0): C, 59.38, H, 4.70. Found: C, 59.28, H, 4.79. MS for 4:
m/e 778 (M+), 516 (M+ - PPh3). Anal. Calcd for C40H43ClP2-
FeRu (778.1): C, 61.75, H, 5.57. Found: C, 61.55, H, 5.62.
Cr ysta l Da ta a n d Str u ctu r e Refin em en t Deta ils for 4:
C40H43ClFeP2Ru, Mr ) 778.05, red block, 0.2 × 0.2 × 0.15 mm3,
T ) 293 K, Bruker SMART CCD, Mo KR radiation, λ )
0.710 73 Å, monoclinic, space group Pn, a ) 13.7601(13) Å, b
) 9.2059(9) Å, c ) 15.3591(14) Å, â ) 115.160(2)°, V ) 1761.0-
(3) Å3, Z ) 2, Dc ) 1.467 Mg/m3, µ(Mo KR) ) 1.035 mm-1
,
F(000) ) 800, total of 10 335 reflections, 5650 unique reflec-
tions, R(int) ) 0.0386, 4916 reflections with I > 2σ(I), full-
matrix least-squares refinement on F2, 406 parameters, final
results R1 ) 0.0374, wR2 ) 0.0787 (I > 2σ(I)), R1 ) 0.0433,
wR2 ) 0.0814 (all data), GOF ) 0.832, maximum/minimum
residual electron density +0.73/-0.35 eÅ-3. Crystallographic
data (excluding structure factors) for the structure reported
in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication
no. CCDC-154123. Copies of the data can be obtained free of
charge on application to The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, U.K. (fax, int. +1223/336-033; E-mail,
teched@chemcrys.cam.ac.uk).
Exp er im en ta l Section
Gen er a l P r oced u r es. Reactions were carried out under
an atmosphere of nitrogen by means of conventional Schlenk
techniques. Solvents were purified and deoxygenated by
conventional methods. Silica was heated at 220 °C for 12 h,
cooled to room temperature under high vacuum, and stored
under dinitrogen.
Syn th esis of Iod id e 5. A solution of 4 (50 mg, 0.06 mmol)
in THF/acetone was stirred in the presence of excess solid NaI
for 36 h at room temperature. Reaction progress is easily
monitored by measuring 31P NMR spectra of the reaction
mixture. The solvent was removed in a vacuum and the
residue was chromatographed on silica with toluene/ether 4:1
to give 53 mg (95%) of 5 as a red solid after removal of the
NMR spectra were recorded on a Varian Unity 500 spec-
trometer (1H, 500 MHz; 31P{1H}, 202 MHz; 13C{1H}, 126 MHz)
and a Varian Mercury 200 (1H, 200 MHz; 31P, 81 MHz). 1H
spectra are referenced to the residual solvent signal and 31P
spectra to external H3PO4 (85%). Mass spectra were recorded
on a Finnigan MAT-95 spectrometer (EI, 70 eV nominal
electron energy). The anions 2a ,b5 and [Ru(PPh3)3Cl2]18 were
prepared as described in the literature.
1
2
solvent. H NMR (500 MHz, CDCl3): 1.53 (d, J (H,P) ) 30.5
Hz, 1 H, R-H), 1.86 (s, Me), 1.89 (s, Cp*), 1.96 (s, Me), 2.29-
2.38 (m, 2 H, CH2), 2.42 (m, 1 H, C5H4), 4.40 (m, 1 H, C5H4),
5.07 (m, 1 H, C5H4), 5.31 (m, 1 H, C5H4), 7.2-7.5 (m, 15 H,
PPh3). 13C NMR (126 MHz, CDCl3): 10.6 (s, Cp* CH3), 11.7
Syn th esis of Ch lor id e Com p lexes 3 a n d 4. A mixture of
Na-2a or Na-2b (0.92 mmol) and [Ru(PPh3)3Cl2] (0.92 mmol)
in dry toluene (25 mL) was heated to 90 °C overnight. Most of
the solvent was removed under vacuum, and the residue was
purified by chromatography on silica. PPh3 was removed first
with toluene, and the complexes were eluted with 2:1 toluene/
ether. Recrystallization gave red crystals of 3 (from toluene,
0.58 mmol, 63%) or 4 (from CH2Cl2/Et2O, 0.63 mmol, 68%).
1H NMR (500 MHz, CDCl3): 3a (major isomer), 2.03 (s, Me),
2.14 (s, Me), 2.47 (m, 1 H, C5H4), 2.49-2.67 (m, 2 H, CH2),
2
(s, phospholyl CH3), 13.9 (s, phospholyl CH3), 21.7 (d, J (C,P)
) 16.5 Hz, CH2), 60.2 (d, 1J (C,P) ) 19.0 Hz, phospholyl R-CH),
69.5 (s, Cp CH), 80.8 (s, Cp CH), 81.1 (d, J (C,P) ) 11.0 Hz, Cp
CH), 84.3 (s, Cp* C(q)), 86.9 (s, Cp C(q)), 88.0 (s, phospholyl
R-C(q)), 104.9 (d, J (C,P) ) 9.3 Hz, phospholyl â-C), 116.4 (s
(br), phospholyl â-C), 127-138 (PPh3). 31P NMR (202 MHz,
CDCl3): 23.7 (d, 2J (P,P) ) 65 Hz, phospholyl P), 45.1 (d, 2J (P,P)
) 65 Hz, PPh3). MS: m/e 870 (M+), 608 (M+ - PPh3), 481 (M+
- PPh3 - I). Anal. Calcd for C40H43IP2FeRu (869.5): C, 55.25;
H, 4.99. Found: C, 54.57; H, 5.04.
2
2.76 (d, J (H,P) ) 31.7 Hz, 1 H, R-H), 4.43 (s, Cp), 4.43 (m, 1
H, C5H4), 5.10 (m, 1 H, C5H4), 5.19 (m, 1 H, C5H4), 7.28-7.60
(m, 15 H, PPh3); 3b (minor isomer), 2.04 (s, Me), 2.07 (s, Me),
2
3.75 (s, Cp), 2.91 (d, J (H,P) ) 31.5 Hz, 1 H, R-H); 4, 1.59 (d,
2J (H,P) ) 30.8 Hz, 1 H, R-H), 1.88 (s, Me), 1.90 (s, Cp*), 1.92
(s, Me), 2.11 (m, 1 H, C5H4), 2.31 (dd, 2J (H,H) ) 15.2 Hz,
3J (H,P) ) 23.5 Hz, 1 H, CH2), 2.43 (dd, 2J (H,H) ) 15.2 Hz,
3J (H,P) ) 6.2 Hz, 1 H, CH2), 4.35, 5.04, 5.27 (3 m, 3 H, C5H4),
7.24-7.52 (m, 15 H, PPh3). 13C NMR (126 MHz, CDCl3): 3a
Syn th esis of Hyd r id es 6 a n d 9a ,b. Sodium methoxide was
prepared by dissolving 30 mg of Na (1.3 mmol) in 3 mL of
MeOH. To this solution was added a solution of 4 (106 mg,
0.14 mmol) in 2 mL of toluene, and the mixture was heated to
reflux for 2 h, during which time a color change from red to
light orange was observed. The mixture was evaporated to
dryness and the residue chromatographed on silica with 10:1
hexane/ether. A total of 86 mg (84%) of 6 was obtained as a
yellow powder after removal of the solvent. 1H NMR (500 MHz,
2
(major isomer), 14.6 (s, CH3), 16.5 (s, CH3), 24.6 (d, J (C,P) )
1
15.9 Hz, CH2), 61.2 (d, J (C,P) ) 29.1 Hz, phospholyl R-CH),
65.2 (s, Cp CH), 73.8 (s, C5H5), 80.3 (s, Cp CH), 82.4 (d, J (C,P)
) 12.1 Hz, Cp CH), 82.8 (d, J (C,P) ) 11.0 Hz, Cp CH), 86.2 (s,
phospholyl R-C(q)), 88.9 (s, Cp C(q)), 110.3 (d, 1J (C,P) ) 9.8
Hz, phospholyl â-C), 116.5 (s, phospholyl â-C), 128-137 (PPh3);
4, 10.4 (s, Cp* CH3), 11.8 (d, 3J (C,P) ) 4.4 Hz, phospholyl CH3),
2
C6D6): -9.54 (dd, J (H,P) ) 40.2, 30.6 Hz, 1 H, Ru-H), 1.69
(d, 2J (H,P) ) 30.7 Hz, 1 H, R-H), 1.60 (s, Me), 1.68 (s, Me),
1.74 (s, Cp*), 2.14 (dd, 2J (H,H) ) 15.3 Hz, 3J (H,P) ) 9.5 Hz, 1
2
3
3
2
H, CH2), 2.41 (dd, J (H,H) ) 15.3 Hz, J (H,P) ) 18.4 Hz, 1 H,
CH2), 4.47 (m, 1 H, C5H4), 4.56 (m, 1 H, C5H4), 4.88 (m, 1 H,
C5H4), 5.52 (m, 1 H, C5H4), 6.8-7.07 (m, PPh3), 7.73-7.84 (m,
PPh3). 13C NMR (126 MHz, C6D6): 10.2 (s, Cp* CH3), 11.8 (d,
13.9 (d, J (C,P) ) 3.8 Hz, phospholyl CH3), 22.1 (d, J (C,P) )
1
17.0 Hz, CH2), 61.3 (d, J (C,P) ) 22.0 Hz, phospholyl R-CH),
62.9 (s, Cp CH), 80.3 (d, J (C,P) ) 7.1 Hz, Cp CH), 81.6 (s, Cp
3
3J (C,P) ) 4.9 Hz, phospholyl CH3), 14.4 (d, J (C,P) ) 4.4 Hz,
(18) Hallman, P. S.; Stephenson, T. A.; Wilkinson, G. Inorg. Synth.
1970, 12, 237-241.
phospholyl CH3), 22.5 (d, 2J (C,P) ) 17.0 Hz, CH2), 61.8 (d,

Ruthenium Half-Sandwich Complexes
Organometallics, Vol. 20, No. 8, 2001 1619
1J (C,P) ) 13.1 Hz, phospholyl R-CH), 72.5 (d, J (C,P) ) 9.9 Hz,
Cp CH), 77.7 (s, Cp CH), 79.2 (d, J (C,P) ) 6.6 Hz, Cp CH),
81.9 (s, Cp* C(q)), 84.1 (d, J (C,P) ) 4.4 Hz, Cp CH), 84.2 (s,
1.90 (s, Me), 1.94 (s, Me), 2.34 (dd, 2J (H,H) ) 15.7 Hz, 3J (H,P)
2
3
) 6.4 Hz, 1 H, CH2), 2.62 (dd, J (H,H) ) 15.7 Hz, J (H,P) )
25.0 Hz, 1 H, CH2), 3.37 (m, 1 H, C5H4), 4.95 (m, 1 H, C5H4),
5.39 (m, 1 H, C5H4), 5.45 (m, 1 H, C5H4), 7.17-7.54 (m, PPh3).
13C NMR (126 MHz, CD2Cl2): 10.1 (s, Cp* CH3), 11.7 (d,
2
Cp C(q)), 85.4 (d, J (C,P) ) 4.9 Hz, phospholyl R-C(q)), 109.1
(s, phospholyl â-C), 117.0 (s, phospholyl â-C), 127-143 (PPh3).
31P NMR (202 MHz, C6D6): 58.1 (d, 2J (P,P) ) 43 Hz, phos-
3J (C,P) ) 5.0 Hz, phospholyl CH3), 13.9 (d, J (C,P) ) 4.5 Hz,
3
2
pholyl P), 70.2 (d, J (P,P) ) 43 Hz, PPh3). HRMS: m/e calcd
phospholyl CH3), 21.5 (d, 2J (C,P) ) 15.4 Hz, CH2), 60.2 (s,
phospholyl R-CH), 79.6 (d, J (C,P) ) 6.0 Hz, Cp CH), 77.9 (s,
Cp CH), 83.6 (d, J (C,P) ) 6.0 Hz, Cp CH), 84.8 (s, Cp* C(q)),
88.6 (s, Cp CH), 88.8 (s, Cp C(q)), 90.7 (d, 2J (C,P) ) 6.5 Hz,
R-C(q)), 107.5 (d, 2J (C,P) ) 12.6 Hz, phospholyl â-C), 123.2 (d,
2J (C,P) ) 9.4, Hz, phospholyl â-C), 129-136 (PPh3). 31P NMR
(202 MHz, CD2Cl2): -144.3 (sept, 1J (P,F) ) 710 Hz, PF6-), 35.7
for C40H44P2FeRu (743.7), 744.1305; found, 744.1295.
9a ,b was obtained in a completely analogous manner by
treatment of 192 mg (0.27 mmol) of 3a ,b with NaOMe (31 mg,
1.36 mmol) in 10 mL of MeOH at reflux for 3 h. The solvent
was removed and the residue purified by chromatography on
silica with 5:1 hexane/ether to give a yellowish solid of 9a ,b
(152 mg, 84%, isomer ratio 1:1) after removal of the solvent.
2
2
(d, J (P,P) ) 49 Hz, phospholyl P), 51.5 (d, J (P,P) ) 49 Hz,
PPh3).
1
Signal assignments to either 9a or 9b were deduced from H-
1H-COSY and 1H-13C-HMQC spectra. 1H NMR (500 MHz,
C6D6): 9a , -9.52 (dd, 2J (H,P) ) 40.6, 29.7 Hz, 1 H, Ru-H),
1.75 (s, Me), 1.76 (s, Me), 2.22 (dd, 2J (H,H) ) 15.2 Hz, 3J (H,P)
Syn th esis of Ca tion ic P yr id in e Com p lex 8. To a solution
of 7 in CH2Cl2sprepared in the manner described above from
88 mg (0.11 mmol) of 4 and 43 mg (0.11 mmol) of TlPF6swas
added 8.9 mg (0.11 mmol, 1 equiv) of pyridine via syringe. The
solvent was removed from the red solution, and the residue
was purified by chromatography on silica with 1:1 CH2Cl2/THF
to give 79.2 mg (72%) of 8 as a red oil after removal of the
solvent. Crystalline material was obtained by layering a CH2-
2
3
) 7.3 Hz, 1 H, CH2), 2.46 (dd, J (H,H) ) 15.2 Hz, J (H,P) )
2
11.9 Hz, 1 H, CH2), 2.57 (d, J (H,P) ) 31 Hz, 1 H, R-H), 3.98
(s, 5 H, Cp), 4.59 (m, 1 H, C5H4), 4.64 (m, 1 H, C5H4), 4.99 (m,
1 H, C5H4), 5.33 (m, 1 H, C5H4), 7.07-7.79 (m, PPh3); 9b,
-10.16 (dd, 2J (H,P) ) 41.5, 36.2 Hz, 1 H, Ru-H), 1.79 (s, Me),
2
3
1
1.83 (s, Me), 2.25 (dd, J (H,H) ) 15.3 Hz, J (H,P) ) 5.9 Hz, 1
Cl2 solution with ether. H NMR (500 MHz, CD2Cl2): 1.76 (s,
2
3
2
H, CH2), 2.33 (dd, J (H,H) ) 15.3 Hz, J (H,P) ) 12.9 Hz, 1 H,
CH2), 2.60 (d, 2J (H,P) ) 31 Hz, 1 H, R-H), 3.76 (s, 5 H, Cp),
4.70 (m, 2 H, C5H4), 4.81 (m, 1 H, C5H4), 5.02 (m, 1 H, C5H4),
5.16 (m, 1 H, C5H4), 7.06-7.83 (m, PPh3). 31P NMR (202 MHz,
C6D6): 9a , 47.3 (d, 2J (P,P) ) 44 Hz, phospholyl P), 68.7 (d,
2J (P,P) ) 44 Hz, PPh3); 9b, 46.7 (d, 2J (P,P) ) 40 Hz, phospholyl
P), 68.5 (d, 2J (P,P) ) 40 Hz, PPh3). 13C NMR (126 MHz,
C6D6): 9a , 74.0 (d, J (C,P) ) 8.9 Hz, C5H4), 77.7 (d, J (C,P) )
8.8 Hz, C5H4), 79.1 (s, C5H4), 82.5 (d, J (C,P) ) 4.9 Hz, C5H4),
74.8 (s, Cp CH); 9b, 72.9 (d, J (C,P) ) 7.6 Hz, C5H4), 75.6 (d,
J (C,P) ) 7.7 Hz, C5H4), 82.7 (s, C5H4), 84.0 (d, J (C,P) ) 5.5
Hz, C5H4), 73.8 (s, Cp CH); signals attributable to 9a or 9b,
Cp*), 1.95 (s, Me), 2.00 (s, Me), 2.06 (d, J (H,P) ) 32.2 Hz, 1
2
3
H, R-H), 2.35 (dd, J (H,H) ) 15.4 Hz, J (H,P) ) 9.2 Hz, 1 H,
CH2), 2.41 (dd, 2J (H,H) ) 15.4 Hz, 3J (H,P) ) 20.2 Hz, 1 H,
CH2), 3.13 (m, 1 H, C5H4), 4.69 (m, 1 H, C5H4), 5.17 (m, 1 H,
C5H4), 5.33 (m, 1 H, C5H4), 7.00-7.69 (m, py + PPh3), 8.81 (d,
J ) 5.6 Hz, 2 H, NCH). 13C NMR (126 MHz, CD2Cl2): 10.6 (s,
3
Cp* CH3), 11.7 (d, J (C,P) ) 4.4 Hz, phospholyl CH3), 14.2 (d,
3J (C,P) ) 4.4 Hz, phospholyl CH3), 21.3 (d, 2J (C,P) ) 16.4 Hz,
CH2), 61.4 (d, 1J (C,P) ) 14.8 Hz, phospholyl R-CH), 67.9 (s,
Cp CH), 79.5 (d, J (C,P) ) 9.8 Hz, Cp CH), 82.8 (d, J (C,P) )
6.6 Hz, Cp CH), 84.6 (s, Cp* C(q)), 88.7 (d, J (C,P) ) 6.6 Hz,
Cp CH), 90.1 (s, Cp C(q)), 90.2 (s, phospholyl R-C(q)), 105.5
(d, 1J (C,P) ) 13.2 Hz, â-C), 115.1 (d, 2J (C,P) ) 7.1 Hz,
phospholyl â-C), 125.3 (s, py m-C), 129-135 (PPh3), 137.9 (s,
py p-C), 158.5 (d, J (C,P) ) 9.3 Hz, py o-C). 31P NMR (202 MHz,
CD2Cl2): -144.3 (sept, 1J (P,F) ) 710 Hz, PF6-), 23.5 (d, 2J (P,P)
) 55 Hz, phospholyl P), 50.4 (d, 2J (P,P) ) 55 Hz, PPh3). Anal.
3
3
14.2 (d, J (C,P) ) 6.0 Hz, Me), 14.3 (d, J (C,P) ) 4.4 Hz, Me),
3
3
16.3 (d, J (C,P) ) 4.4 Hz, Me), 16.4 (d, J (C,P) ) 4.9 Hz, Me),
2
2
25.1 (d, J (C,P) ) 17.0 Hz, CH2), 25.2 (d, J (C,P) ) 17.5 Hz,
CH2), 60.0 (d, 1J (C,P) ) 12.6 Hz, phospholyl R-CH), 61.1 (d,
1J (C,P) ) 14.2 Hz, phospholyl R-CH), 84.7 (s, C(q)), 85.0 (s,
C(q)), 87.4 (d, J (C,P) ) 5.0 Hz, C(q)), 87.5 (d, J (C,P) ) 4.9 Hz,
C(q)), 111.9 (d, J (C,P) ) 5.5 Hz, C(q)), 112.8 (d, J (C,P) ) 10.4
Hz, C(q)), 117.6 (d, J (C,P) ) 7.7 Hz, C(q)), 117.9 (d, J (C,P) )
7.1 Hz, C(q)). HRMS: m/e calcd for C35H34P2FeRu, 674.0522;
found, 674.0518.
Calcd for
C45H48P3NF6FeRu (966.7): C, 55.91; H, 5.00.
Found: C, 55.45; H, 5.41.
Ack n ow led gm en t. We are grateful to the Deutsche
Forschungsgemeinschaft (DFG; SFB 380 and Gra-
duiertenkolleg 440) and the Fonds der Chemischen
Industrie for financial support.
Syn th esis of Ca tion ic Dih yd r ogen Com p lex 7. Solid
TlPF6 (68 mg, 0.19 mmol) was added to a solution of 4 (151
mg, 0.19 mmol) in 3 mL of CH2Cl2, and the mixture was stirred
for 3 h under an atmosphere of hydrogen at room temperature.
TlCl was removed by filtration, and 7‚PF6 was precipitated
from the filtrate by addition of ether to give 111 mg (79%) of
a yellow powder. 1H NMR (500 MHz, CD2Cl2): -6.80 (b s, 2
H, Ru-H2), 1.52 (d, 2J (H,P) ) 32.0 Hz, 1 H, R-H), 1.73 (s, Cp*),
Su p p or tin g In for m a tion Ava ila ble: X-ray structural
information for compound 4. This material is available free of
charge via the Internet at http://www.acs.org.
OM0010262