Amine attack on the carbonyl ligands of the protonated dicyclopentadienyl-bridged diruthenium complex [{(η5-C5H3)2(SiMe 2)2}Ru2(CO)4(μ-H)]+

Article abstract of DOI:10.1021/om000825h

Complexes [{(η⁵-C₅H₃)₂(SiMe ₂)₂}Ru₂(CO)₄(μ-H)]⁺ (1H⁺BF₄^-, 1D+TfO^-), with a protonated Ru-Ru bond, were prepared by protonation of {(η⁵-C₅H₃)₂(SiMe ₂)₂}Ru₂(CO)₄ (1) with HBF₄·Et₂O or CF₃SO₃D. The bridging proton in 1H⁺ is removed only very slowly by amine bases even though it is thermodynamically acidic (pK_a^AN = 6.5 (±0.2)). This remarkable kinetic inertness of the bridging proton allows amines (NH₃, NH₂CH₃, NH(CH₃)₂, morpholine, piperidine, pyrrolidine) to react with 1H⁺ by attacking the CO ligand to give a formamide (HC(=O)NR₂) and the CO-substituted product {(η⁵-C₅H₃)₂(SiMe ₂)₂}Ru₂(CO)₃(NHR₂) (2). Thus, protonation of the metal-metal bond in 1H⁺ promotes reactions of the CO ligand that are not possible in the unprotonated 1. A proposed mechanism for these reactions is supported by kinetic studies of the reaction of 1D⁺TfO^- with morpholine in nitromethane at 20.0 °C, as well as by deuterium-labeling experiments. The molecular structure of {(η⁵-C₅H₃)₂-(SiMe ₂)₂}Ru₂(CO)₃(NH₂CH ₃) (2f), as determined by an X-ray diffraction investigation, is also presented.

Full text of DOI:10.1021/om000825h

Organometallics 2001, 20, 691-696
691
Am in e Atta ck on th e Ca r bon yl Liga n d s of th e P r oton a ted
Dicyclop en ta d ien yl-Br id ged Dir u th en iu m Com p lex
[{(η5-C5H3)2(SiMe2)2}Ru 2(CO)4(µ-H)]+
Maxim V. Ovchinnikov, Ilia A. Guzei,† and Robert J . Angelici*
Department of Chemistry, Iowa State University, Ames, Iowa 50011
Received September 25, 2000
Complexes [{(η5-C5H3)2(SiMe2)2}Ru2(CO)4(µ-H)]+ (1H+BF4-, 1D+TfO-), with a protonated
Ru-Ru bond, were prepared by protonation of {(η5-C5H3)2(SiMe2)2}Ru2(CO)4 (1) with
HBF4‚Et2O or CF3SO3D. The bridging proton in 1H+ is removed only very slowly by amine
AN
bases even though it is thermodynamically acidic (pKa ) 6.5 ((0.2)). This remarkable
kinetic inertness of the bridging proton allows amines (NH3, NH2CH3, NH(CH3)2, morpholine,
piperidine, pyrrolidine) to react with 1H+ by attacking the CO ligand to give a formamide
(HC(dO)NR2) and the CO-substituted product {(η5-C5H3)2(SiMe2)2}Ru2(CO)3(NHR2) (2). Thus,
protonation of the metal-metal bond in 1H+ promotes reactions of the CO ligand that are
not possible in the unprotonated 1. A proposed mechanism for these reactions is supported
by kinetic studies of the reaction of 1D+TfO- with morpholine in nitromethane at 20.0 °C,
as well as by deuterium-labeling experiments. The molecular structure of {(η5-C5H3)2-
(SiMe2)2}Ru2(CO)3(NH2CH3) (2f), as determined by an X-ray diffraction investigation, is also
presented.
In tr od u ction
complexes either do not react with amines because their
kCO and ν(CO) values are insufficiently high or the
amine bases simply deprotonate the metal to give the
unreactive neutral metal complex. We recently com-
municated7 the synthesis of the cationic dinuclear
complex [{(η5-C5H3)2(SiMe2)2}Ru2(CO)4(µ-H)]+ (1H+),
whose carbon monoxide ligands are activated to attack
by amine nucleophiles because of the positive charge
on the complex, which is sufficiently high to give ν(CO)
Metal-promoted nucleophilic attack on unsaturated
ligands is a reaction common to a number of transition
metal complexes and constitutes a transformation of
synthetic importance.1 Amines and alkoxides attack the
carbon of carbon monoxide ligands in transition metal
complexes if the positive charge on the complexes is
sufficiently high to give CtO stretching force constants,
k
CO, that are higher than 16.5 mdyn/Å (or ν(CO) values
values higher than 2000 cm-1 2
. At the same time,
higher than approximately 2000 cm-1).2 A common
example of such a reaction is the formation of carbamoyl
complexes (eq 1).
complex 1H+ is only slowly deprotonated by amines
despite its high thermodynamic acidity.7 Treatment of
1H+ with 3 equiv of nucleophilic amines resulted in the
formation of {(η5-C5H3)2(SiMe2)2}Ru2(CO)3(η1-NHR1R2)
(2) and the corresponding formamide in a 1:1 ratio (eq
2). This paper provides details of the earlier study7 and
Reactions of metal carbonyl complexes with amines are
also known to give formamides,3 carbamates,4 and
ureas,5 either stoichiometrically or catalytically.
One of the simplest approaches to making a complex
more positive is to add a proton (H+) to the metal
center.6 However, most protonated metal carbonyl
† Iowa State University, Molecular Structure Laboratory.
(1) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mills Valley, CA, 1987; Chapter 7.
(2) Angelici, R. J . Acc. Chem. Res. 1972, 5, 335.
(3) J enner, G.; Bitsi, G.; Schleiffer, E. J . Mol. Catal. 1987, 39, 233.
(4) Valli, V. L. K.; Alper, H. Organometallics 1995, 14, 80.
(5) (a) McCusker, J . E.; Grasso, C. A.; Main, A. D.; McElwee-White,
L. Org. Lett. 1999, 1, 961. (b) McCusker, J . E.; Logan, J .; McElwee-
White, L. Organometallics 1998, 17, 4037. (c) McCusker, J . E.; Abboud,
K. A.; McElwee-White, L. Organometallics 1997, 16, 3863. (d) Gian-
noccaro, P.; Nobile, C. F.; Mastrorilli, P.; Ravasio, N. J . Organomet.
Chem. 1991, 419, 251. (e) Srivastava, S. C.; Shrimal, A. K.; Srivastava,
A. J . Organomet. Chem. 1991, 414, 65. (f) Bassoli, A.; Rindone, B.;
Tollari, S.; Chioccara, F. J . Mol. Catal. 1990, 60, 41.
10.1021/om000825h CCC: $20.00 © 2001 American Chemical Society
Publication on Web 01/12/2001

692 Organometallics, Vol. 20, No. 4, 2001
Ovchinnikov et al.
MHz, CD3NO2): δ 5.73, 8.05, 8.24, 9.50 (Et); 90.13, 97.41,
new results that offer some understanding of the
mechanism and scope of this reaction.
100.48 (Cp); 196.81, 196.86 (CO). IR (CH2Cl2): ν(CO) (cm-1
)
2077 (vs), 2050 (w), 2027 (s). Anal. Calcd for C22H27BF4O4Ru2-
Si2: C, 37.72; H, 3.88. Found: C, 37.65; H, 3.95.
Exp er im en ta l Section
Rea ction s of 1H+BF 4- a n d 1D+TfO- w ith Am in es. In a
typical experiment, amine (3-5 equiv) was added to a mixture
of 1H+BF4- (or 1D+TfO-) (∼10 mg) and triphenylmethane (∼3
mg, internal standard) in deuterated benzene (1 mL) in an
NMR tube (gaseous amines were bubbled through the suspen-
sion for 5 min; then a stream of dry argon was bubbled through
the solution in order to remove the excess amine). During the
addition of amine, the reaction mixture immediately changed
Gen er a l P r oced u r es. All reactions were performed under
an argon atmosphere in reagent grade solvents, using standard
Schlenk or drybox techniques.8 Hexanes, methylene chloride,
and diethyl ether were purified by the Grubbs method9 prior
to use. All other solvents were purified by published methods.10
Gaseous amines were dried by passing them through a BaO
column. Liquid amines were distilled from Na under an Ar
atmosphere. Chemicals were purchased from commercial
sources or prepared by literature methods, as referenced below.
Alumina (neutral, activity I, Aldrich) was degassed under
-
color from colorless to wine-red. The 1H+BF4 or 1D+TfO-
reacted completely, and yields of the formamides were deter-
mined by means of 1H NMR spectroscopy. Yields of the
formamides13 and characterizations of 2a -f are given below.
1
vacuum for 12 h and treated with 7.5% of water. H and 13C
-
Rea ction s of 1H+BF 4 a n d 1D+TfO- w ith NH3: 72%
NMR spectra were recorded on a Bruker DRX-400 spectrom-
eter. Solution infrared spectra were recorded on a Nicolet-560
spectrometer using NaCl cells with 0.1 mm spacers. Elemental
analyses were performed on a Perkin-Elmer 2400 series II
CHNS/O analyzer.
(84%).14 {(η5-C5H3)2(SiMe2)2}Ru2(CO)3(NH3) (2a ): 1H NMR (400
MHz, C6D6): δ 0.29 (s, 6 H, Si(CH3)), 0.36 (s, 6 H, Si(CH3)),
3.54 (bs, 3 H, NH3), 4.67 (d, J ) 2.0 Hz, 2 H), 4.78 (t, J ) 2.0
Hz, 1 H), 5.10 (d, J ) 2.4 Hz, 2 H), 5.73 (t, J ) 2.4 Hz, 1 H).
IR (hexanes): ν(CO) (cm-1) 1974 (vs), 1903 (vs), 1879 (m).
Rea ction s of 1H+BF 4- a n d 1D+TfO- w ith NH2CH3: 84%,
88%. Isolation and characterization of {(η5-C5H3)2(SiMe2)2}-
Ru2(CO)3(NH2CH3) (2b) were reported earlier.7 Crystals of 2b
suitable for X-ray diffraction analysis were obtained by slow
cooling of a saturated solution of 2b in hexanes to -20 °C.
Syn t h e sis
of
[{(η5-C5H 3)2(SiMe 2)2}R u 2(CO)4(µ-
D)]+CF 3SO3 (1D+TfO-). By reacting CF3SO3D (8 µL, 90.4
µmol) (Aldrich) with 1 (50 mg, 89.8 µmol) in CH2Cl2 (30 mL),
1D+TfO- was prepared using the same method as in the
preparation of 1H+BF4-. Complex 1D+TfO- was found to be
-
spectroscopically (1H, 13C NMR, FT-IR) identical to 1H+BF4
-
-
Rea ction s of 1H+BF 4 a n d 1D+TfO- w ith NH(CH3)2:
except for the near absence11 of the µ-H resonance at -19.92
ppm.7 Anal. Calcd for C19H18BDF3O7Ru2SSi2: C, 32.29; H, 2.71;
S, 4.54. Found: C, 32.18; H, 2.65; S, 4.48.
85%, 90%. {(η5-C5H3)2(SiMe2)2}Ru2(CO)3(NH(CH3)2) (2c): 1H
NMR (400 MHz, C6D6): δ 0.30 (s, 6 H, Si(CH3)), 0.36 (s, 6 H,
Si(CH3)), 2.03 (d, J ) 6.4 Hz, 6 H, (CH3)2NH), 2.90 (bs, 1 H,
(CH3)2NH), 4.59 (d, J ) 2.0 Hz, 2 H), 4.85 (t, J ) 2.0 Hz, 1 H),
5.14 (d, J ) 2.4 Hz, 2 H), 5.73 (t, J ) 2.4 Hz, 1 H). IR
Syn th esis of {(η5-C5H3)2(SiEt2)2}Ru 2(CO)4 (3). A solution
of Ru3(CO)12 (200.0 mg, 312.8 µmol), (C5H4)2(SiEt2)212 (143 mg,
475.7 µmol), and methylisobutyl ketone (1.0 mL, 10.0 mmol)
in heptane (100 mL) was heated to reflux for 30 h. The mixture
was cooled to ambient temperature and chromatographed on
an alumina column (1 × 20 cm) first using hexanes as the
eluent and then a 1:5 (v/v) mixture of CH2Cl2 and hexanes,
which eluted a yellow band containing 3 (170 mg, 58%). 1H
NMR (400 MHz, CDCl3): δ 0.73 (m, 4 H, Si(CH2CH3)), 0.91
(m, 4 H, Si(CH2CH3)), 0.97 (m, 6 H, Si(CH2CH3)), 1.10 (m, 6
H, Si(CH2CH3)), 5.39 (d, J ) 2.2 Hz, 4 H, Cp-H), 5.79 (t, J )
2.2 Hz, 2 H, Cp-H). 13C NMR (100 MHz, CDCl3): δ 5.21, 8.20,
8.23, 9.94 (Et); 87.81, 92.21, 96.04 (Cp); 204.58 (CO). IR
(hexanes): ν(CO) (cm-1) 2025 (vs), 1967 (vs). Anal. Calcd for
(hexanes): ν(CO) (cm-1) 1975 (vs), 1907 (vs), 1882 (m).
Rea ction s of 1H+BF 4
a n d 1D+TfO- w ith NH-
-
(CH2CH2)2O: 89%, 91%. {(η5-C5H3)2(SiMe2)2}Ru2(CO)3{NH(CH2-
CH2)2O} (2d ): 1H NMR (400 MHz, C6D6): δ 0.31 (s, 6 H,
Si(CH3)), 0.39 (s, 6 H, Si(CH3)), 2.95 (m, 2 H, {O(CH2CH2)2-
NH}), 3.33 (m, 2 H, {O(CH2CH2)2NH}), 3.66 (m, 4 H, {O(CH2-
CH2)2NH}), 3.71 (bs, 1 H, (NH)), 4.48 (d, J ) 2.0 Hz, 2 H),
4.81 (t, J ) 2.0 Hz, 1 H), 5.11 (d, J ) 2.4 Hz, 2 H), 5.77 (t, J
) 2.4 Hz, 1 H). IR (hexanes): ν(CO) (cm-1) 1981 (vs), 1909
(vs), 1889 (m).
Rea ction s of 1H+BF 4
a n d 1D+TfO- w ith NH-
-
(CH2CH2)2CH2:82%,88%.{(η5-C5H3)2(SiMe2)2}Ru2(CO)3{NH(CH2-
CH2)2CH2} (2e): 1H NMR (400 MHz, C6D6): δ 0.26 (s, 6 H,
Si(CH3)), 0.43 (s, 6 H, Si(CH3)), 1.16 (m, 2 H, {CH2(CH2CH2)2-
NH}), 1.41 (m, 4 H, {CH2(CH2CH2)2NH}), 2.85 (m, 4 H, {CH2-
(CH2CH2)2NH}), 3.20 (bs, 1 H, (NH)), 4.40 (d, J ) 2.1 Hz, 2
H), 4.73 (t, J ) 2.1 Hz, 1 H), 5.01 (d, J ) 2.4 Hz, 2 H), 5.55 (t,
J ) 2.4 Hz, 1 H). IR (hexanes): ν(CO) (cm-1) 1976 (vs), 1909
(vs), 1879 (m).
C
22H26O4Ru2Si2: C, 43.12; H, 4.28. Found: C, 43.09; H, 4.21.
Syn th esis of [{(η5-C5H3)2(SiEt2)2}Ru 2(CO)4(µ-H)]+BF 4
-
(3H+BF 4-). A solution of 3 (100.0 mg, 161.2 µmol) in CH2Cl2
(20 mL) was treated with HBF4‚Et2O (24.0 µL, 174.1 µmol) at
room temperature. A yellow precipitate of 3H+BF4 was
-
obtained in nearly quantitative yield (111.0 mg, 100%) by
diluting the reaction solution with a 10-fold excess of ether
(200 mL). 1H NMR (400 MHz, CD3NO2): δ -19.76 (s, 1 H, Ru-
H-Ru), 0.95-1.30 (m, 20 H, Si(CH2CH3)), 6.13 (t, J ) 2.2 Hz,
2 H, Cp-H), 6.19 (d, J ) 2.2 Hz, 4 H, Cp-H). 13C NMR (100
Rea ction s of 1H+BF 4- a n d 1D+TfO- w ith NH(CH2CH2)2:
77%, 84%. {(η5-C5H3)2(SiMe2)2}Ru2(CO)3{NH(CH2CH2)2} (2f):
1H NMR (400 MHz, C6D6): δ 0.32 (s, 6 H, Si(CH3)), 0.40 (s, 6
H, Si(CH3)), 1.26 (m, 4 H, (CH2CH2)2NH), 2.55 (m, 4 H,
(CH2CH2)2NH), 3.10 (bs, 1 H, (NH)), 4.68 (d, J ) 2.4 Hz, 2 H),
4.84 (t, J ) 2.4 Hz, 1 H), 5.14 (d, J ) 2.4 Hz, 2 H), 5.75 (t, J
) 2.4 Hz, 1 H). IR (hexanes): ν(CO) (cm-1) 1973 (vs), 1907
(vs), 1887 (m).
(6) (a) Angelici, R. J . Acc. Chem. Res. 1995, 28, 52. (b) Kristja´ns-
do´ttir, S. S.; Norton, J . R. In Transition Metal Hydrides: Recent
Advances in Theory and Experiments; Dedieu, A., Ed.; VCH: New York,
1991; Chapter 10. (c) Pearson, R. G. Chem. Rev. 1985, 85, 41. (d)
Martinho Simo˜es, J . A.; Beauchamp, J . L. Chem. Rev. 1990, 90, 629.
(e) Bullock, R. M. Comments Inorg. Chem. 1991, 12, 1.
(7) Ovchinnikov, M. V.; Angelici, R. J . J . Am. Chem. Soc. 2000, 122,
6130.
(8) Errington, R. J . Advanced Practical Inorganic and Metalorganic
Chemistry, 1st ed.; Chapman & Hall: New York, 1997.
(9) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518.
Rea ction betw een 1H+BF 4 a n d P h NH-Li+. A suspen-
-
sion of 1H+BF4- (156 mg, 0.24 mmol) in diethyl ether (50 mL)
was treated with a freshly prepared ether (10 mL) solution of
PhNH-Li+ 15 (0.26 mmol) at -78 °C. The mixture was gradu-
ally warmed to room temperature and filtered through a short
pad of Celite. Removing the solvent under reduced pressure
(10) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals, 2nd ed.; Pergamon: New York, 1980.
(11) The 1H NMR spectrum of 1D+TfO- in CD3NO2 revealed a small
peak for the Ru-H resonance at -19.98 ppm (s, 0.02 H), which
corresponds to 1H+TfO- as a ∼2% impurity.
-
(13) Yield values correspond to the reactions of 1H+BF4 and
1D+TfO-, respectively.
(12) Ko¨hler, F. H.; Schell, A.; Weber, B. J . Organomet. Chem. 1999,
575, 33.
(14) Yield was estimated from the yield of {(η5-C5H3)2(SiMe2)2}Ru2-
(CO)3(NH3).

Amine Attack on Carbonyl Ligands
Organometallics, Vol. 20, No. 4, 2001 693
gave an oily brown residue, which was found to be a mixture
of 1 and formanilide (49% yield) by 1H NMR spectroscopy. The
reaction with 1D+TfO- was conducted in a similar fashion
(61% yield of formanilide).
0.40 × 0.18 × 0.06 mm was selected under oil under ambient
conditions and attached to the tip of a glass capillary. The
crystal was mounted in a stream of cold nitrogen at 173(2) K
and centered in the X-ray beam by using a video camera. The
crystal evaluation and data collection were performed on a
Bruker CCD-1000 diffractometer with Mo KR (λ ) 0.71073 Å)
radiation and a diffractometer-to-crystal distance of 5.08 cm.
The initial cell constants were obtained from three series of ω
scans at different starting angles. Each series consisted of 20
frames collected at intervals of 0.3° in a 6° range about ω with
an exposure time of 10 s per frame. A total of 47 reflections
were obtained. The reflections were successfully indexed by
an automated indexing routine in the SMART program. The
final cell constants were calculated from a set of 4927 strong
reflections from the actual data collection. The data were
collected by using the hemisphere data collection routine.
Reciprocal space was surveyed to the extent of 1.9 hemispheres
to a resolution of 0.80 Å. A total of 10 630 data were harvested
by collecting three sets of frames with 0.3° scans in ω with an
exposure time of 60 s per frame. These highly redundant
datasets were corrected for Lorentz and polarization effects.
The absorption correction was based on fitting a function to
the empirical transmission surface as sampled by multiple
equivalent measurements.17 Systematic absences in the dif-
fraction data were consistent with the space groups P1h and
P1.18 The E-statistics strongly suggested the centrosymmetric
space group P1h that yielded chemically reasonable and com-
putationally stable results of refinement. A successful solution
by the direct method provided most non-hydrogen atoms from
the E-map. The remaining non-hydrogen atoms were located
in an alternating series of least-squares cycles and difference
Fourier maps. All non-hydrogen atoms except for N(1), N(2),
C(7), and C(19) were refined with anisotropic displacement
coefficients. All hydrogen atoms were included in the structure
factor calculation at idealized positions and were allowed to
ride on the neighboring atoms with relative isotropic displace-
ment coefficients. There is occupational disorder present in
the structure. One coordination site of each Ru atom is 50%
occupied by a CO ligand and 50% occupied by a MeNH2 ligand.
The disordered groups were refined with idealized geometries.
There is also half a molecule of solvate hexane per molecule
of complex present in the asymmetric unit. The final least-
squares refinement of 289 parameters against 4878 data
resulted in residuals R (based on F2 for I g 2σ) and wR (based
on F2 for all data) of 0.0464 and 0.1107, respectively (Table
1).
Synthesis of[{(η5-C5H3)2(SiMe2)2}Ru2(CO)3{NH(CH2CH2)2}-
-
-
(µ-H)]+BF 4 (2fH+BF 4-). A suspension of yellow 1H+BF4
(50.0 mg, 77.6 µmol) in hexanes (25 mL) was treated with neat
pyrrolidine (32 µL, 0.4 mmol). The color of the reaction mixture
immediately changed to wine-red. Solvent was removed in a
vacuum, and the red residue was recrystallized from hexanes
(10 mL) at -25 °C to give 41 mg (85%) of 2f as dark red, air-
and moisture-sensitive crystals. A solution of 2f (25 mg, 41.7
µmol) in CH2Cl2 (10 mL) treated with HBF4‚OEt2 (6 µL, 43.5
µmol) immediately gave a bright red solution, which was
diluted with diethyl ether (50 mL) to give 2fH+BF4 (28 mg,
-
98%) as a red crystalline precipitate. 1H NMR (400 MHz, CD2-
Cl2): δ -19.18 (s, 1 H, Ru-H-Ru), 0.38 (s, 3 H, Si(CH3)), 0.47
(s, 3 H, Si(CH3)), 0.51 (s, 3 H, Si(CH3)), 0.53 (s, 3 H, Si(CH3)),
1.71 (m, 2 H, (CH2)), 1.86 (m, 2 H, (CH2)), 2.33 (m, 2 H, (CH2)),
3.19 (m, 1 H, (CH2)), 3.34 (m, 1 H, (CH2)), 4.96 (bs, 1 H, NH),
5.17 (m, 1 H, Cp), 5.64 (m, 1 H, Cp), 5.71 (m, 1 H, Cp), 5.87
(m, 2 H, Cp), 6.01 (m, 1 H, Cp). IR (CH2Cl2): ν(CO) (cm-1
)
2050 (vs), 2002 (s), 1954 (m). Anal. Calcd for C21H28BF4NO3-
Ru2Si2: C, 36.68; H, 4.10; N, 2.04. Found: C, 36.42; H, 3.80;
N, 2.06.
Kin etic Stu d ies of th e Rea ction (eq 2) of 1D+TfO- w ith
Mor p h olin e. In a typical NMR experiment, 1D+TfO- (3-6
mg) and triphenylmethane (∼3 mg, internal standard) were
dissolved in dry CD3NO2 (1 mL) under an argon atmosphere
in an NMR tube. The tube was placed in the NMR spectrom-
eter, and the probe temperature was set at 20.0 ( 0.5 °C. The
temperature of the solution was allowed to equilibrate for at
least 30 min. The tube was taken out, an amount of neat
morpholine to give a 7.72 × 10-2 to 9.40 × 10-2 M solution
was injected into the NMR tube, and the tube was returned
to the NMR spectrometer. In a typical IR experiment, 1D+TfO-
(30-50 mg) was dissolved in dry CH3NO2 (10 mL) under an
argon atmosphere in a Schlenk flask equipped with a constant-
temperature water jacket. The jacket was connected to a
constant-temperature water circulator, and the temperature
of the solution was allowed to equilibrate for at least 30 min.
Then, an amount of neat morpholine to give a 4.10 × 10-1 to
15.1 × 10-1 M solution was added. Samples were periodically
withdrawn from the flask, and their IR spectra were obtained
in a constant-temperature IR cell.
The reactions were monitored by the disappearance of the
2075 cm-1 ν(CO) band in the IR spectra or the 6.11 and 6.17
ppm bands in the 1H NMR spectra of 1D+TfO-. Rate constants
were calculated from the 10-60 spectra taken during the first
two (second-order conditions) or three (pseudo-first-order
conditions) half-lives of the reaction. Rate data, which were
obtained by the IR method under pseudo-first-order conditions,
where a ∼100-fold excess of morpholine was utilized, were fit
to the equation16 rate ) kobs[1D+TfO-] to obtain the pseudo-
first-order rate constants kobs (s-1). Second-order rate constants
Resu lts a n d Discu ssion
Syn th esis an d P r oton ation of {(η5-C5H3)2(SiMe2)2}-
19
Ru 2(CO)4 (1). The reaction of (C5H4)2(SiMe2)2 with
Ru3(CO)12 in the presence of the hydrogen acceptor (1-
dodecene or methylisobutyl ketone) furnished {(η5-
C5H3)2(SiMe2)2}Ru2(CO)4 (1) in 72% yield, as an air- and
moisture-stable yellow solid. When the reaction was
carried out in the absence of a hydrogen acceptor, a
k2 (M-1‚s-1) were calculated using the expression k2 ) kobs
/
20
considerable amount of Ru4(CO)12(µ-H)4 (15-25%
[morpholine]av, where [morpholine]av is the average of the
[morpholine] at the beginning and the end of the reaction.
based on Ru equiv) was formed and lower yields of 1
were obtained.
Rate data, which were obtained by the 1H NMR method,
when [morpholine]0 was less than 30 times larger than
[1D+TfO-], were fit to a second-order equation16 by plotting
ln([morpholine]t/[1D+TfO-]t) vs time. The second-order rate
constants k2 (M-1‚s-1) were determined from the slope of the
best fit straight line.
The hydride-bridged dinuclear Ru complex [{(η5-
C5H3)2(SiMe2)2}Ru2(CO)4(µ-H)]+ (1H+) was prepared in
(16) Espenson, J . Chemical Kinetics and Reaction Mechanisms, 1st
ed.; McGraw-Hill: New York, 1981.
(17) Blessing, R. H. Acta Crystallogr. 1995, A51, 33-38.
(18) All software and sources of the scattering factors are contained
in the SHELXTL (version 5.1) program library (G. Sheldrick, Bruker
Analytical X-ray Systems, Madison, WI).
X-r a y Cr ysta llogr a p h y of {(η5-C5H3)2(SiMe2)2}Ru 2(CO)3-
(NH2CH3) (2b). A red crystal with approximate dimensions
(19) (a) Siemeling, U.; J utzi, P.; Neumann, B.; Stammler, H.-G.;
Hursthouse, M. B. Organometallics 1992, 11, 1328. (b) Hiermeier, J .;
Ko¨hler, F. H.; Mu¨ller, G. Organometallics 1991, 10, 1787.
(20) Wilson, R. D.; Wu, S. M.; Love, R. A.; Bau, R. Inorg. Chem.
1978, 17, 1271.
(15) (a) Dorta, R.; Togni, A. Helv. Chim. Acta 2000, 83, 119. (b)
Matsuzaka, H.; Kamura, T.; Ariga, K.; Watanabe, Y.; Okubo, T.; Ishii,
T.; Yamashita, M.; Kondo, M.; Kitagawa, S. Organometallics 2000, 19,
216.

694 Organometallics, Vol. 20, No. 4, 2001
Ovchinnikov et al.
Ta ble 1. Cr ysta l Da ta a n d Str u ctu r e Refin em en t
for 2b
CH3, NH(CH3)2, morpholine, piperidine, pyrrolidine) at
ambient temperature to yield the complexes {(η5-C5H3)2-
(SiMe2)2}Ru2(CO)3(NHR1R2) (2) and the corresponding
formamides in a 1:1 ratio (eq 2). The only other
Ru-containing product was the deprotonated complex
{(η5-C5H3)2(SiMe2)2}Ru2(CO)4 (1). The yields of forma-
empirical formula
fw
C21H30NO3Ru2Si2
602.78
temperature
wavelength
crystal system
space group
unit cell dimens
173(2) K
0.71073 Å
triclinic
P1h
1
mides, established for most cases by H NMR spectros-
a ) 8.8046(5) Å
b ) 8.9523(5) Å
c ) 16.6067(9) Å
R ) 89.721(1)°
â ) 88.199(1)°
γ ) 66.777(1)°
1202.27(12) Å3
2
copy, vary from 72% for the reaction of ammonia with
1H+ to 91% for morpholine with 1D+. The only other
product of these reactions was 1, which was observed
in 5-20% yields. The least bulky amine (ammonia) gave
the lowest yield of formamide due to the faster rate of
direct deprotonation of 1H+ by amine to give complex
{(η5-C5H3)2(SiMe2)2}Ru2(CO)4 (1). Also, yields of the
formamides were generally higher when the deuterated
1D+ complex was used, since 1D+ presumably under-
goes slower deprotonation compared to 1H+ because of
the deuterium isotope effect. Unfortunately, we were
unable to measure the magnitude of the deuterium
isotope effect for proton transfer from 1H+ because we
did not find a base that will deprotonate 1H+ without
side reactions23 at rates sufficiently fast for kinetic
studies. Experiments using the deuterium-labeled
1D+TfO- and amines (NHMe2, morpholine) gave for-
mamide products (D(CdO)NRR′) that are >95% deu-
terated at the formyl position and no other. Bulky
nucleophilic amines (Bn2NH, i-Pr2NH, Cy2NH) and
weakly nucleophilic amines (aniline) failed to react with
1H+ under the same reaction conditions.
volume
Z
density (calcd)
abs coeff
1.665 Mg/m3
1.377 mm-1
F(000)
606
cryst size
θ range for data collection
index ranges
0.40 × 0.18 × 0.06 mm
2.45-26.37°.
-10 e h e 10, -11 e k e 11,
0 e l e 20
no. of reflns collected
no. of ind reflns
completeness to θ ) 26.37°
abs corr
max. and min. transmn
refinement method
no. of data/restraints/params
goodness-of-fit on F2
final R indices [I>2σ(I)]
R indices (all data)
10 630
4878 [R(int) ) 0.0349]
99.0%
empirical with SADABS
0.9219 and 0.6089
full-matrix least-squares on F2
4878/6/289
1.008
R1 ) 0.0464, wR2 ) 0.1107
R1 ) 0.0662, wR2 ) 0.1189
1.563 and -0.958 e Å-3
largest diff peak and hole
Second-order rate constants (k2) for the reaction (eq
2) of morpholine24 with 1D+TfO- were determined from
rate studies conducted under pseudo-first-order condi-
tions (∼100-fold excess of amine; [1D+TfO-] ) (8.50-
8.95)× 10-3 M, [morpholine] ) (4.10-15.10) × 10-1 M,
by the IR method) or second-order conditions (<30-fold
excess of amine; [1D+TfO-] ) (9.09-9.89) × 10-3 M,
quantitative yield upon addition of 1 equiv of HBF4‚OEt2
or CF3SO3D to a CH2Cl2 solution of complex 1 at room
temperature. The Ru-H resonance in the 1H NMR
spectrum occurs as a singlet at δ -19.92 ppm. The CO
stretching frequencies for 1H+ are approximately 67
cm-1 higher than those for 1 and fall within the range
where amine attack on the CO groups is expected to
occur.2 Complex 1H+ is kinetically inert with respect
to deprotonation by organic bases such as Me3N. Less
than 2% of the complex was deprotonated after 1 h in
CD3NO2 or CD3CN solution in the presence of 10-fold
excesses of Me3N or pyridine, whose conjugate acids
1
[morpholine] ) (7.72-9.40) × 10-2 M, by the H NMR
method) in nitromethane at 20 °C. The reactions were
followed by monitoring the disappearance of the ν(CO)
bands in the IR spectra or the 1H NMR signals of
1D+TfO-. The rate constants k2 (Table S1) were ob-
AN
have pKa values of 17.61 and 12.33, respectively.21
tained either indirectly from the equation k2 ) kobs
/
In contrast, the unbridged22a and monobridged22b
analogues of 1H+, (η5-C5H5)2Ru2(CO)4(µ-H)+ and {(η5-
C5H4)2(SiMe2)}Ru2(CO)4(µ-H)+, undergo fast and quan-
titative deprotonation by bases such as pyridine or
diethylamine. To understand whether the slow rate of
[morpholine]av or directly from linear plot, ln-
a
([morpholine]t/[1D+TfO-]t) vs time, for a second-order
reaction. A plot (Figure 2) of kobs vs [morpholine]av gave
a straight line with a near-zero intercept. Thus, the
reaction follows the second-order rate law, -d[1D+-
TfO-]/dt ) k2[1D+TfO-][morpholine], where k2 ) (2.2
( 0.5) × 10-3 M-1 s-1. Although the rate constant k2
includes the rate of deprotonation of 1D+TfO- by
morpholine, the contribution of this concurrent reaction
(e5%) to the overall rate is smaller than the experi-
mental error in the kinetic studies.
-
deprotonation of 1H+BF4 was caused by kinetic or
AN
thermodynamic factors, we estimated the pKa value
(6.5 ((0.2) in CD3CN) for 1H+BF4- from studies of the
equilibrium constant for the proton-transfer reaction
-
between 1 and HPPh3+BF4 in CD3CN at 25 °C.7 The
AN
pKa value for complex 1H+BF4- clearly indicates that
the above-noted amine bases21 will thermodynamically
deprotonate 1H+BF4- easily. Although it is not obvious
why 1H+ undergoes slow deprotonation, it is perhaps a
combination of the bulkiness of the dimethylsilyl linkers
and the rigidity of the dicyclopentadienyl bridging
ligand.
This rate law is consistent with the mechanism
proposed in Scheme 1. Initial nucleophilic attack by the
amine on a coordinated CO produces the cationic
(22) (a) Ovchinnikov, M. V.; Angelici, R. J . Unpublished results. (b)
Fro¨hlich, R.; Gimeno, J .; Gonza´lez-Cueva, M.; Lastra, E.; Borge, J .;
Garcia-Granda, S. Organometallics 1999, 18, 3008.
Rea ction of [{(η5-C5H3)2(SiMe2)2}Ru 2(CO)4(µ-
(23) NMe3 causes slow decomposition of 1H+ to a mixture of
unidentifiable products at rates that are faster than the rate of direct
deprotonation.
H)]+BF 4 (1H+BF 4-) w ith Am in es. Complex 1H+
-
reacts with 3 equiv of nucleophilic amines (NH3, NH2-
(24) The rate of the reaction of 1D+TfO- with morpholine was the
only rate suitable for kinetic studies by 1H NMR and FT-IR spec-
troscopies. More nucleophilic amines (NH3, NH2CH3, NH(CH3)2, pip-
eridine, pyrrolidine) reacted with 1D+TfO- very fast (t1/2 < 1 s).
(21) Coetzee, J . F.; Padmanabhan, G. R. J . Am. Chem. Soc. 1965,
87, 5005.

Amine Attack on Carbonyl Ligands
Organometallics, Vol. 20, No. 4, 2001 695
Sch em e 1
raethyl analogue of 1H+, [{(η5-C5H3)2(SiEt2)2}Ru2(CO)4-
(µ-H)]+ (3H+), which was successfully characterized by
elemental analysis and 1H NMR, 13C NMR, and IR
spectroscopies. However, the reactions of amines (NH3,
NH2Me, NHMe2, morpholine) with complex 3H+ gave
essentially the same yields of formamides and 3 (the
result of direct deprotonation of 3H+ by amine) as those
for the reactions of 1H+.
F igu r e 1. Thermal ellipsoid drawing of {(η5-C5H3)2(Me2-
Si)2}Ru2(CO)3(NH2CH3) (2b) showing the labeling scheme
at 50% probability ellipsoids; hydrogens are omitted for
clarity. Selected bond distances (Å) and angles (deg) are
as follows: Ru(1)-Ru(2), 2.8955(6); Ru(1)-N(1), 2.148(9);
N(1)-C(8), 1.4691(10); Ru(1)-Cp(centroid), 1.887; Ru(2)-
Cp(centroid), 1.889; C(8)-N(1)-Ru(1), 112.5(7); C(18)-
Ru(2)-C(19), 99.4(4); C(6)-Ru(1)-N(1), 85.3(3); Ru(1)-
Ru(2)-C(18), 88.50(17); C(6)-Ru(1)-Ru(2), 90.94(18);
C(6)-Ru(1)-Ru(2)-C(18), 1.0; C(6)-Ru(1)-Ru(2)-
C(19), 98.4; Cp-Cp fold angle, 125.9.
Weakly nucleophilic amines, such as aniline, with
pKa(H2O) values lower than ∼8 failed to react with
1H+. In contrast to aniline, the corresponding lith-
ium anilide PhNH-Li+ readily reacted with 1H+ and
1D+ to give formanilide (H(CdO)NHPh) in 49% and 61%
isolated yields, respectively. As observed in reactions
of the alkylamines with 1D+, the formanilide product
(D(CdO)NHPh) from the reaction of 1D+ with PhNH-Li+
1
was found by H NMR studies to be almost completely
(>95%) deuterated at the formyl position and no other.
Complex 1, rather than 2, was observed as the only
identifiable ruthenium-containing product of the reac-
tion of PhNH-Li+ with both 1H+ and 1D+. The for-
mation of 1 clearly also requires the formation of other
Ru products, which were not identified. The direct
deprotonation of 1H+ by PhNH-Li+ also may lead to
complex 1.
Ch a r a cter iza tion of Com p lexes {(η5-C5H3)2-
(SiMe2)2}Ru 2(CO)3(NHR1R2) (2). The cyclopentadi-
enyl hydrogens of the (η5-C5H3)Ru(CO)(NHR1R2) frag-
ment in 2 are observed as a well-resolved doublet and
triplet, shifted approximately 0.7 ppm upfield from the
cyclopentadienyl hydrogens of the (η5-C5H3)Ru(CO)2
part of the molecule. It is worth noting that at room
temperature the 1H NMR spectra of the 2 complexes
show only four signals in the cyclopentadienyl region
instead of the expected six and only two singlets for the
Si(CH3)2 groups instead of four, required by the struc-
ture of the molecules. The 13C NMR spectrum of 2b
exhibits only two signals for the CO ligands and six
signals in cyclopentadienyl region.7 Although these
observations suggest that these compounds are flux-
ional, variable-temperature NMR studies of {(η5-C5H3)2-
(SiMe2)2}Ru2(CO)3(NH2CH3) (2b) in CD3NO2 at tem-
peratures down to -25 °C did not reveal any broadening
or splitting of the cyclopentadienyl resonances.
F igu r e 2. Plot of kobs vs [morpholine]av for reaction (eq 2)
of 1D+TfO- with morpholine at 20.0 °C.
intermediate A. Subsequent deprotonation of the nitro-
gen in A is likely to be fast, and reductive elimination
of the formamide from B must be rapid because there
is no spectroscopic evidence (1H NMR or IR) for inter-
mediates in the reaction. Reductive elimination of the
formamide from B gives an unsaturated diruthenium
intermediate that coordinates an amine to give 2. The
observed first-order dependence on amine concentration
and the absence of measurable quantities of intermedi-
ates means that the first step, nucleophilic attack on a
CO ligand, is rate-determining.
Since the bulkiness of the dimethylsilyl linkers in the
(η5-C5H3)2(SiMe2)2 ligand is a possible reason for the
unusually low kinetic acidity of 1H+, it is conceivable
that a more bulky linker would make the proton even
less kinetically acidic and presumably increase yields
of the formamide products in reactions of 1H+ with
amines. We therefore synthesized the more bulky tet-
However, it is known25 that Cp2Ru2(CO)4 is fluxional
even at low temperatures, which suggests that 2 is
fluxional according to the mechanism in Scheme 2,

696 Organometallics, Vol. 20, No. 4, 2001
Ovchinnikov et al.
Sch em e 2
spectrum as five sets of well-resolved multiplets, are
shifted approximately 0.4 ppm downfield compared to
the cyclopentadienyl hydrogens of 2f due to the positive
charge on the molecule. In contrast to 2f, the compli-
cated resonances in the cyclopentadienyl region of 2fH+,
as well as the four separate methyl resonances of the
dimethylsilyl groups, indicate the absence of fluxional
behavior of the type observed for 2 (Scheme 2). The Ru-
which involves the pairwise exchange of two carbonyl
ligands between the two metal atoms via a bridging
carbonyl intermediate. This mechanism requires the
movement of the NHR1R2 ligand from one side of the
molecule to the other, but not from one Ru to the other;
1
H-Ru resonance in the H NMR spectrum occurs as a
singlet at -19.18 ppm. The presence of the strongly
donating pyrrolidine ligand in 2fH+BF4-, as one would
expect, lowers the carbonyl stretching frequencies in
2fH+BF4- (2050, 2002, 1954 cm-1) as compared to those
in 1H+BF4- (2077, 2050, 2027 cm-1), thereby deactivat-
1
this motion accounts for the relatively simple H and
13C NMR spectra of the (η5-C5H3)2(SiMe2)2 ligand. If CO
groups on opposite sides of the Ru-Ru bond are involved
in forming the carbonyl-bridged intermediate, only two
of the three CO groups can participate in the exchange
process, which accounts for the observation of two
13CO signals. The IR spectra in the ν(CO) region for all
complexes of type 2 exhibit a medium band in the 1890
cm-1 region and two strong bands in the 1910 and 1980
cm-1 regions, which are typical for group 8 Cp2M2(CO)3L
complexes without bridging CO ligands.26
-
ing the remaining CO ligands in 2fH+BF4 to nucleo-
philic attack by alkylamines, which confirms the use-
fulness of ν(CO) values for determining the reactivity
of CO groups with amines. Bulky alkylamines (NMe3)
-
deprotonate 2fH+BF4 (t1/2 ≈ 30 min) slowly, although
much faster than the deprotonation of 1H+, whereas less
bulky amines (NH3) deprotonate 2fH+BF4- quickly (t1/2
≈ 1.5 min).
Con clu sion s
The structure of 2b (Figure 1), established by X-ray
crystallography, shows the presence of occupational
disorder. One coordination site of each Ru atom is 50%
occupied by a CO (C(19), O(4)) ligand and 50% by a
NH2Me ligand. The Ru-Ru distance in 2b (2.8955(6)
Å) is longer than that in 1 (2.8180(3) Å).27 The
C(6)-Ru(1)-Ru(2)-C(18) torsion angle (1.0°) confirms
the eclipsed orientation of the CO ligands. Both Ru
atoms of complex 2b have pseudo-octahedral geometry,
with angles of approximately 90° between adjacent
carbonyls, methylamine, and the Ru-Ru bond (Fig-
ure 1). As one would expect, the Cp-Cp fold angle28
(125.9°) is larger than in 1 (120.5(7)°) because of the
longer Ru-Ru distance.
The lability of the amine ligand in complex 2 was
established by observing the substitution of this ligand
(NH3, NH2Me, NHMe2) by CO in a hexanes solution of
complex 2 under a flow of CO gas (1 atm) at ambient
temperature (t1/2 ) 40-75 min) to give {(η5-C5H3)2-
(SiMe2)2}Ru2(CO)4 (1). As expected, the bulky NHMe2
ligand was the most labile (t1/2 ) 40 min) as compared
with the less bulky NH3 and NH2Me. Complexes 2 are
extremely air-sensitive and undergo slow decomposition
in solution under inert atmosphere to give 1 as the only
identifiable product.
We have prepared a cationic complex [{(η5-C5H3)2-
(SiMe2)2}Ru2(CO)4(µ-H)]+ (1H+) in which the bridging
proton is removed only very slowly by amine bases even
AN
though it is thermodynamically acidic (pKa
) 6.5
((0.2)). Although it is not obvious why complex 1H+ is
kinetically stable with respect to deprotonation, it may
be due to a combination of the bulkiness of the dimeth-
ylsilyl linkers and the rigidity of the dicyclopentadienyl
bridging ligand. The low kinetic acidity and the positive
charge in 1H+ allow its CO ligands to undergo attack
by alkylamines. To our knowledge, this is the only sys-
tem in which protonation of metal centers in a complex
activates CO ligands to nucleophilic attack. The amine
reactions lead to the elimination of the µ-H ligand,29
which becomes incorporated into the formamide prod-
uct, as shown in Scheme 1. Weakly nucleophilic amines
(e.g., aniline) with pKa(H2O) values lower than ∼8.0030
and bulky amines with cone angles (Θ) larger than 125°
(e.g., Bn2NH, i-Pr2NH, Cy2NH)31 do not react with 1H+.
Although aniline does not react with 1H+, its amide
(RNH-) does react to give the formanilide.
Ack n ow led gm en t. This work was supported by the
National Science Foundation through Grant No. CHE-
9816342.
P r oton ation of {(η5-C5H3)2(SiMe2)2}Ru 2(CO)3{NH-
(CH2CH2)2} (2f). The Ru-Ru bond in 2f is protonated
with 1 equiv of acid (HBF4‚OEt2) to give the air-stable
cationic complex [{(η5-C5H3)2(SiMe2)2}Ru2(CO)3{NH(CH2-
Su p p or tin g In for m a tion Ava ila ble: Tables giving crys-
tallographic data for 2b including atomic coordinates, bond
lengths and angles, and anisotropic displacement parameters;
rate constants for the reaction (eq 2) of 1D+TfO- with mor-
pholine in nitromethane at 20 °C. This material is available
free of charge via the Internet at http://pubs.acs.org.
CH2)2}(µ-H)]+BF4 (2fH+BF4-). The cyclopentadienyl
-
hydrogens in 2fH+BF4-, appearing in the 1H NMR
(25) (a) Bullitt, J . G.; Cotton, F. A.; Marks, T. J . J . Am. Chem. Soc.
1970, 92, 2155. (b) Gansow, O. A.; Burke, A. R.; Vernon, W. D. J . Am.
Chem. Soc. 1976, 98, 5817.
OM000825H
(26) (η5,η5-C10H8)Ru2(CO)3(PMe3) in: (a) Boese, R.; Tolman, W. B.;
Vollhardt, K. P. C. Organometallics 1986, 5, 582. (b) Boese, R.;
Cammack, J . K.; Matzger, A. J .; Pflug, K.; Tolman, W. B.; Vollhardt,
K. P. C.; Weidman, T. W. J . Am. Chem. Soc. 1997, 119, 6757.
(27) Ovchinnikov, M. V.; Guzei, I. A.; Angelici, R. J . Manuscript in
preparation.
(29) For examples involving reductive eliminations involving a µ-H,
see: Stockland, R. A., J r.; Anderson, G. K.; Rath, N. P. J . Am. Chem.
Soc. 1999, 121, 7945, and references therein.
(30) Perrin, D. D. Dissociation Constants of Organic Bases in
Aqueous Solution; Butterworth: London, 1972.
(31) For amine cone angles, see: Seligson, A. L.; Trogler, W. C. J .
Am. Chem. Soc. 1991, 113, 2520.
(28) The angle between the two cyclopentadienyl planes.