Organometallics
Article
procedures. Gas chromatography was performed on Hewlet-Packard
3398 and 6890 Series gas chromatographs equipped with a split-mode
capillary injection system and flame ionization detectors using HP
Ultra-1 (25 m × 0.32 mm), CP-Chirasil-DEX CB (25 m × 0.25 mm),
and Beta Dex 120 (30 m × 0.25 mm) columns. (SM,RC)-[(η5-
C5Me5)M(PROPHOS)(H2O)][SbF6]2 (M = Rh (1), Ir (2)) were
prepared according to published procedures.17
side of the double bond. In this disposition and with the M−O
rotamer fixed by the CH/π interactions, the Cα-si face of the
enone is shielded by the pro-S phenyl ring of the P1Ph2 group,
and therefore, the diene attack would take place preferentially
through the re-face, in good agreement with the catalytic
outcome for the MVK/HCp reaction. The comparison of the
structural parameters of the EVK ligand in 5 to those of its
analogue MVK in 4 sheds light on the different catalytic
behavior of 5. The MVK ligand in 4 is essentially planar. The
maximum deviation from the mean plane, 0.030(5) Å,
corresponds to C(40). However, although the O−C(38)−
C(39)−C(40)−C(41) skeleton of the EVK ligand in 5 is also
essentially planar, the remaining CH3 fragment significantly
deviates from planarity (Figure 3). In fact, the C(41)−C(42)
bond is almost perpendicular to the above-defined plane, the
angle between the C(41)−C(42) vector and the normal to this
plane being only 16(1)°. Probably, this methyl fragment adopts
a similar disposition in solution because we have measured a
NOE relationship between these protons and those of the
C5Me5 ring (see above). In this conformation, this methyl
hinders the approach of the diene through the Cα-re face, and
therefore, both faces are similarly accessible by the diene.
Consequently, even at low temperature, enantioselectivity is
eroded and, according to the catalytic outcome, the attack via
the Cα-si face is slightly preferred for this complex.
Scheme 3. Labeling of the Cation of the Complexes for NMR
Assignments
Preparation of (SM,RC)-[(η5-C5Me5)M(PROPHOS)(MVK)][SbF6]2
(M = Rh (3), Ir (4)) and (SM,RC)-[(η5-C5Me5)Ir(PROPHOS)(EVK)]-
[SbF6]2 (5). At −20 °C, under argon, to a solution of the
corresponding (SM,RC)-[(η5-C5Me5)M(PROPHOS)(H2O)][SbF6]2
(0.09 mmol) complex in CH2Cl2 (4 mL) were added enone (0.9
mmol) and 4 Å molecular sieves (100.0 mg). The resulting suspension
was stirred for 20 min and then was filtered through a cannula. The
filtrate was concentrated to ca. 3 mL. The slow addition of 20 mL of
dry n-hexane afforded yellow crystals, which were filtered off, washed
with n-hexane, and vacuum-dried. Recrystallization from CH2Cl2/n-
hexane yielded pure samples of the complexes.
CONCLUSION
■
In summary, the aqua complexes 1 and 2 generate active
systems that efficiently catalyze the Diels−Alder reaction
between the vinyl ketones MVK and EVK and dienes in
good endo/exo ratio and moderate to good enantioselectivity.
From 1 and 2, the catalyst−substrate intermediates (SM,RC)-
[(η5-C5Me5)M(PROPHOS)(enone)][SbF6]2 can be prepared
in a completely diastereoselective manner. From detailed
structural information about these catalyst−substrate inter-
mediates, in both the solid state and solution, it is possible to
explain the catalytic outcome. The coordinated enone adopts
an s-trans conformation and the configuration around the CO
double bond is Z. Particularly relevant is the existence of CH/π
intramolecular interactions in the solid state that, according to
NMR solution data, most probably remain in solution. These
interactions fix the M−O enone rotamer and conform the
disposition of the enone inside the chiral pocket of the catalyst
defined by the (C5Me5)M(PROPHOS) moiety. As a result, the
Cα-si face of both MVK and EVK intermediates becomes
hindered by a PROPHOS phenyl and, additionally, the CH3
fragment of the enone EVK hampers approach of the diene
through the opposite enantioface. All these structural data are
in good agreement with the experimental catalytic results: while
89% ee's are achieved for the MVK/HCp reaction, only 4% ee
is obtained for the related EVK/HCp system.
(SRh,RC)-[(η5-C5Me5)Rh(PROPHOS)(MVK)][SbF6]2 (3). Yield:
85%. Anal. Calcd for C41H47F12RhOP2Sb2: C, 41.3, H, 3.9. Found:
C, 41.4; H, 3.9. IR (KBr, cm−1): ν(CO) 1664 (m), ν(SbF6) 659 (s).
1H NMR (400.16 MHz, CD2Cl2, −50 °C): δ 7.91−7.26 (m, 20H, Ph),
6.07 (d, J = 17.7 Hz, 1H, Hc), 6.03 (d, J = 10.8 Hz, 1H, Hb), 4.65 (dd, J
= 17.6, 10.8 Hz, 1H, Ha), 3.37 (dt, J = 53.2, 14.2 Hz, 1H, H22), 2.59
(m, 1H, H11), 2.55 (m, 1H, H21), 1.48 (s, 3H, COCH3), 1.42 (m, 15H,
C5Me5), 1.19 ppm (m, 3H, Me). 13C NMR (100.61 MHz, CD2Cl2,
−50 °C): δ 215.57 (CO), 141.87 (C4), 131.83 (C3), 134.44−119.42
(24C, Ph), 99.17 (C5Me5), 31.91 (Me), 31.02 (dd, J(PC) = 37.0, 6.8
Hz, C1), 30.35 (dd, J(PC) = 38.0, 13.5 Hz, C2), 26.20 (COCH3),
14.36 (dd, J(PC) = 18.0, 3.0 Hz, Me), 10.07 ppm (C5Me5). 31P NMR
(161.96 MHz, CD2Cl2, −20 °C): δ 74.51 (dd, J(RhP1) = 130.2 Hz,
J(P1P2) = 39.8 Hz, P1), 50.71 ppm (dd, J(RhP2) = 131.3 Hz, P2).
(SIr,RC)-[(η5-C5Me5)Ir(PROPHOS)(MVK)][SbF6]2 (4). Yield: 79%.
Anal. Calcd for C41H47F12IrOP2Sb2: C, 38.4, H, 3.7. Found: C, 38.3;
1
H, 3.9. IR (KBr, cm−1): ν(CO) 1676 (m) ν(SbF6) 659 (s). H NMR
(400.16 MHz, CD2Cl2, −70 °C): δ 7.91−7.23 (m, 20H, Ph), 6.14 (d, J
= 17.1 Hz, 1H, Hc), 6.07 (d, J = 10.7 Hz, 1H, Hb), 4.62 (dd, J = 18.1,
11.2 Hz, 1H, Ha), 3.26 (dt, J = 53.9, 10.8 Hz, 1H, H22), 2.51 (m, 1H,
H11), 2.42 (m, 1H, H21), 1.56 (s, 3H, COCH3), 1.44 (m, 15H,
C5Me5), 1.20 ppm (m, 3H, Me). 13C NMR (100.61 MHz, CD2Cl2,
−50 °C): δ 216.60 (CO), 141.92 (C4), 132.90 (C3), 134.81−118.43
(24C, Ph), 99.17 (C5Me5), 30.98 (dd, J(PC) = 36.8, 7.7 Hz, C1), 30.35
(m, C2), 26.37 (Me), 14.48 (COCH3), 9.66 ppm (C5Me5). 31P NMR
(161.96 MHz, CD2Cl2, −50 °C): δ 45.74 (d, J(P1P2) = 11.6 Hz, P1)
28.49 ppm (d, P2).
EXPERIMENTAL SECTION
■
General Comments. All solvents were dried over appropriate
drying agents, distilled under argon, and degassed prior to use. Dienes
and dienophiles were distilled prior to use. All preparations have been
carried out under argon. Infrared spectra were obtained as KBr pellets
with a Perkin-Elmer Spectrum One FT-IR spectrophotometer.
Carbon, hydrogen, and nitrogen analyses were performed using a
Perkin-Elmer 240C microanalyzer. 1H, 13C, and 31P NMR spectra were
recorded on a Bruker AV 500 (500.13 MHz), AV-400 (400.16 MHz),
or 300 ARX (300.10 MHz) spectrometer. Chemical shifts are
expressed in ppm upfield from SiMe4 or 85% H3PO4 (31P). NOESY
and 13C, 31P, 1H correlation spectra were obtained using standard
(SIr,RC)-[(η5-C5Me5)Ir(PROPHOS)(EVK)][SbF6]2 (5). Yield: 83%.
Anal. Calcd for C42H49F12IrOP2Sb2: C, 38.9, H, 3.8. Found: C, 39.0;
1
H, 3.3. IR (KBr, cm−1): ν(CO) 1677 (m) ν(SbF6) 652 (s). H NMR
(400.16 MHz, CD2Cl2, −70 °C): δ 7.86−7.20 (m, 20H, Ph), 6.29 (d, J
= 17.7 Hz, 1H, Hc), 6.03 (d, J = 10.8 Hz, 1H, Hb), 4.42 (dd, J = 17.7,
11.1 Hz, 1H, Ha), 3.28 (dt, J = 53.2, 14.2 Hz, 1H, H22), 2.43 (m, 1H,
H11), 2.37 (q, J = 8.7 Hz, 2H, COCH2), 2.34 (m, 1H, H21), 1.41 (m,
15H, C5Me5), 0.83 (t, J = 7.0 Hz, 3H, Me), 0.23 ppm (t, J = 7.3 Hz,
3H, COCH2CH3). 13C NMR (100.61 MHz, CD2Cl2, −50 °C): δ
220.54 (CO), 140.79 (C4), 131.16. (C3), 136.60−118.34 (24C, Ph),
99.68 (C5Me5), 33.12 (dd, J(PC) = 40.6, 7.7 Hz, C1), 33.16
E
dx.doi.org/10.1021/om300346s | Organometallics XXXX, XXX, XXX−XXX