Tandem Carbonylation Reactions: Hydroformylation and Hydroaminomethylation of Alkenes Catalyzed by Cationic [(H2C(3,5-Me2pz) 2)Rh(CO)L]+ Complexes

Article abstract of DOI:10.1021/om030351x

Addition of H₂C(3,5-Me₂pz)₂ (Bpm*) to [(η⁴-1,5-COD)Rh(acetone)₂]BF₄ affords the complex [Bpm*Rh(COD)]BF₄, which is carbonylated under mild conditions into [Bpm*Rh(CO)₂]BF₄. One of the CO ligands can be easily displaced by PPh₃, PMePh₂, or P(OMe)₃. The X-ray structures of [Bpm*Rh(CO) ₂]BF₄ and [Bpm*Rh(CO)(PPh₃)]BF ₄ have been determined and show a rhodium center in a square-planar environment. These complexes have been fully characterized by infrared and ¹H, ¹³C, and ³¹P NMR spectroscopy. They are rather active precursors for the hydroformylation of oct-1-ene. They also promote the hydroaminomethylation reaction of terminal alkenes: i.e., the direct production of alkylamines from successive hydroformylation, condensation of the resulting aldehydes with diethylamine, and hydrogenation of the enamines. (2-Methyloctyl)diethylamine and nonyldiethylamine are obtained from oct-1-ene/H₂/CO/NHEt₂ building blocks. The reaction rate can be significantly increased on addition of [RuH₂(PPh ₃)₄], which acts presumably as a hydride-donating species.

Full text of DOI:10.1021/om030351x

Organometallics 2003, 22, 5261-5267
5261
Ta n d em Ca r bon yla tion Rea ction s: Hyd r ofor m yla tion
a n d Hyd r oa m in om eth yla tion of Alk en es Ca ta lyzed by
Ca tion ic [(H2C(3,5-Me2p z)2)Rh (CO)L]+ Com p lexes
Emmanuelle Teuma,† Maxime Loy,† Carole Le Berre,† Michel Etienne,‡
J ean-Claude Daran,‡ and Philippe Kalck*,†
Laboratoire de Catalyse, Chimie Fine et Polyme`res, Ecole Nationale Supe´rieure
des Inge´nieurs en Arts Chimiques et Technologiques, 118 route de Narbonne,
31077 Toulouse Cedex 4, France, and Laboratoire de Chimie de Coordination du CNRS,
UPR 8241, 205 route de Narbonne, 31077 Toulouse Cedex 4, France
Received May 9, 2003
Addition of H2C(3,5-Me2pz)2 (Bpm*) to [(η4-1,5-COD)Rh(acetone)2]BF4 affords the complex
[Bpm*Rh(COD)]BF4, which is carbonylated under mild conditions into [Bpm*Rh(CO)2]BF4.
One of the CO ligands can be easily displaced by PPh3, PMePh2, or P(OMe)3. The X-ray
structures of [Bpm*Rh(CO)2]BF4 and [Bpm*Rh(CO)(PPh3)]BF4 have been determined and
show a rhodium center in a square-planar environment. These complexes have been fully
1
characterized by infrared and H, 13C, and 31P NMR spectroscopy. They are rather active
precursors for the hydroformylation of oct-1-ene. They also promote the hydroaminomethy-
lation reaction of terminal alkenes: i.e., the direct production of alkylamines from successive
hydroformylation, condensation of the resulting aldehydes with diethylamine, and hydro-
genation of the enamines. (2-Methyloctyl)diethylamine and nonyldiethylamine are obtained
from oct-1-ene/H2/CO/NHEt2 building blocks. The reaction rate can be significantly increased
on addition of [RuH2(PPh3)4], which acts presumably as a hydride-donating species.
In tr od u ction
This tandem reaction couples the hydroformylation of
an olefin with amine condensation of the resulting
aldehyde, most of the time followed by the hydrogena-
tion of the resulting enamines or imines. Rhodium
complexes are successful in promoting these types of
reactions. In addition to simple catalytic mixtures,
systems based on more sophisticated nitrogen-contain-
ing ligands have become popular.6 For example, cationic
bis(imidazolyl)methane rhodium complexes promote an
intramolecular (cyclization) version of the hydroamina-
tion reaction.7 In a somewhat related system, we
recently described how [TpMe2,ClRh(CO)2] preferentially
activates photochemically a primary C-H bond as
opposed to the N-H bond of NHiPr2, yielding a single
diastereomer of [TpMe2,ClRhH(CH2CH(CH3)NHiPr)].8 Bis-
(pyrazolyl)methanes can serve as a bridge between bis-
(imidazolyl)methane and tris(pyrazolyl)borate ligands.
On one hand, bis(pyrazolyl)methanes offer two nitrogen
atoms for coordination to rhodium, just like bis(imida-
zolyl)methanes, whereas the Tp′ ligand is usually
coordinated in the κ3-mode in [Tp′Rh(CO)2]. On the
other hand, cationic rhodium precursors such as [(η4-
1,5-COD)Rh(solv)2] are known to catalyze the hydroge-
nation of alkenes in the presence of PPh3.
Catalytic couplings of amines with olefins or with
alkyl and aryl halides to produce more elaborate amines
are some of the most important reactions that remain
to be solved, not only from a conceptual point of view
but also at an economical level, since very often a
pharmaceutical activity is relevant to a nitrogen-
containing framework.1 Three main strategies have
been explored in the literature. First, the overall olefin
insertion into an N-H bond, namely olefin hydroami-
nation, is recognized as a highly challenging process.1,2
Late-transition-metal complexes have been found to
catalyze this reaction, and mechanistic and synthetic
issues have been addressed.3 The catalytic amination
of aryl halides is one of the alternatives which is now
becoming more efficient.4 However, another reaction
which promotes a different type of coupling, namely
hydroaminomethylation of olefins, is worth considering.5
* To whom correspondence should be addressed. E-mail:
Philippe.Kalck@ensiacet.fr.
† Ecole Nationale Supe´rieure des Inge´nieurs en Arts Chimiques et
Technologiques.
‡ Laboratoire de Chimie de Coordination du CNRS.
(1) Applied Homogeneous Catalysis with Organometallic Com-
pounds; Cornils, B., Herrmann, W. A., Eds.; VCH: Weinheim, Ger-
many, 1996.
(2) Recent reviews: (a) Beller, M.; Mu¨ller, T. E. Chem. Rev. 1998,
98, 675-703. (b) Roundhill, D. M. Chem. Rev. 1992, 92, 1-27.
(3) (a) Fulton, J . R.; Holland, A. W.; Fox, D. J .; Bergman, R. G. Acc.
Chem. Res. 2002, 35, 44-56. (b) Senn, H. M.; Blo¨chl, P. E.; Togni, A.
J . Am. Chem. Soc. 2000, 122, 4098-4107. (c) Beller, M.; Trautwein,
H.; Eichberger, M.; Breindl, C.; Herwig, J .; Mu¨ller, T. E.; Thiel, O. R.
Chem. Eur. J . 1999, 5, 1306-1319. (d) Driver, M. S.; Hartwig, J . F. J .
Am. Chem. Soc. 1997, 119, 8232-8245.
(5) Eilbracht, P.; Ba¨rfacker, L.; Buss, C.; Hollmann, C.; Kitsos-
Rzychon, B. E.; Kranemann, C. L.; Rische, T.; Roggenbuck, R.; Schmidt,
A. Chem. Rev. 1999, 99, 3329-3365.
(6) (a) Fache, F.; Schultz, E.; Tommasino, M. L.; Lemaire, M. Chem.
Rev. 2000, 100, 2159-2231. (b) Togni, A.; Venanzi, L. M. Angew.
Chem., Int. Ed. Engl. 1994, 31, 497-526.
(7) Burling, S.; Field, L. D.; Messerle, B. A. Organometallics 2000,
19, 87-90.
(8) Teuma, E.; Malbosc, F.; Pons, V.; Leberre, C.; J aud, J .; Etienne,
M.; Kalck, P. Dalton 2001, 2225-2227.
(4) (a) Wolfe, J . P.; Wagaw, S.; Buchwald, S. L. J . Am. Chem. Soc.
1996, 118, 7215-7216. (b) Driver, M. S.; Hartwig, J . F. J . Am. Chem.
Soc. 1996, 118, 7217-7218.
10.1021/om030351x CCC: $25.00 © 2003 American Chemical Society
Publication on Web 11/06/2003

5262 Organometallics, Vol. 22, No. 25, 2003
Teuma et al.
Ta ble 1. Cr ysta llogr a p h ic Da ta for I-BF 4 a n d
III-BF 4
I-BF4
III-BF4
mol formula
fw
C13H16BF4N4O2Rh C36H43BF4N4O3PRh
450.04
800.43
shape (color)
cryst size, mm
cryst syst
space group
a, Å
flat (yellow)
cubic (orange)
0.52 × 0.28 × 0.08 0.50 × 0.50 × 0.40
monoclinic
P21/n
triclinic
P1h
7.7361(9)
11.8804(9)
18.5455(19)
90.0
93.67(1)
90.0
10.099(1)
13.346(2)
14.955(2)
76.36(1)
75.70(1)
74.83(1)
1853.4(8)
2
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
1701.0(5)
4
Z
F(000)
891
1.757
10.595
824
1.434
5.64
F (calcd), g cm-3
µ(Mo KR) cm-1
diffractometer
radiation
Stoe IPDS
Mo KR (λ ) 0.710 73)
temp, K
180(2)
60
detector dist, mm
scan mode
φ range, deg
φ incr, deg
exposure time, min
2θ range, deg
no. of rflns collected
no. of unique rflns
70
φ (oscillation)
0.0 < φ < 200.2
1.4
φ (rotation)
0.0 < φ < 250.8
1.2
4
2
F igu r e 1. ORTEPIII view of I-BF4 with the atom-labeling
scheme. Ellipsoids represent 50% probability. Selected bond
lengths (Å) and angles (deg): C(1)-Rh(1), 1.865(2); C(2)-
Rh(1), 1.860(3); N(11)-Rh(1), 2.0834(18); N(21)-Rh(1),
2.0888(17); C(1)-O(1), 1.122(3); C(2)-O(2), 1.127(3); N(11)-
N(12), 1.365(2); N(21)-N(22), 1.373(2); N(12)-C(3), 1.445-
(3); N(22)-C(3), 1.450(3); C(2)-Rh(1)-C(1), 87.82(11);
N(21)-Rh(1)-C(2), 93.95(9); N(21)-Rh(1)-N(11), 85.76-
(7); N(11)-Rh(1)-C(1), 92.53(9); N(21)-Rh(1)-C(1), 176.79-
(9); N(11)-Rh(1)-C(2), 178.6(1).
4.4 < 2θ < 52.2
13 449
3283
5.9 < 2θ < 56.3
22 423
8227
merging factor R(int) 0.0277
0.0399
no. of rflns used
refinement
2695 (I > 2σ(I))
F (CRYSTALS)
6400 (I > 2σ(I))
F2 (SHELXL97)
0.0247, 0.0647
0.0284, 0.0665
0.012
R1, wR2 (I > 2σ(I))
R1, wR2 (all data)
(∆/σ)max
0.0250, 0.0136
0.0006
-0.42/0.42
1.030
∆Fmin/∆Fmax
-0.39/0.36
1.033
GOF
no. of variable
params
226
479
To clarify these apparent discrepancies, crystals of
I-BF4 were grown from an acetone/heptane/diethyl ether
solution and analyzed by X-ray diffraction. The X-ray
data collection and processing parameters for I-BF4 are
given in Table 1. Bond distances and angles of interest
are in the classical range and are reported in the caption
of Figure 1.
I-BF4 crystallizes in the space group P21/n, and the
cell contains four asymmetric units consisting of one
mononuclear cationic entity and one BF4 anion. For the
discrete cation, the rhodium atom is in a square-planar
environment, as shown in Figure 1. The Bpm* entity
acts as a chelating ligand, and the four nitrogen atoms
and the rhodium and carbon (CH2) atoms adopt a boat
conformation, reminiscent of that observed in the previ-
ously described [Bpm*Rh(COD)]ClO4.9 The angles be-
tween the two CO ligands is 87.82(11)°, in accord with
the nearly equal intensities of the two νCO bands in the
solution IR spectra.
The four νCO bands observed in the solid-state infrared
spectra are thus best accounted for by the so-called
correlation field approximation, in which the four
cations of the unit cell are treated as the vibrating unit,
subject to symmetry restrictions arising from the sym-
metry of the entire unit cell. Due to the presence of the
inversion center, only the four antisymmetric C-O
stretching vibrations should be observed in the crystal
from the eight CO’s present in the unit cell.10 Recall that
the importance of this effect is virtually unpredictable.
We report herein
a new synthetic pathway to
[Bpm*Rh(CO)2]BF4 which avoids the use of the perchlo-
rate counterion9 and the synthesis of [Bpm*Rh(CO)-
(PR3)]BF4 complexes with their structural studies. In
addition, we present a first approach to the utilization
of such compounds in the hydroformylation and hy-
droaminomethylation of olefins.
Resu lts a n d Discu ssion
Stu d y of [Bp m *Rh (CO)2]BF 4 (I-BF 4). Oro and co-
workers9 have described the preparation of the cation
[Bpm*Rh(CO)2]+, isolated as its [RhCl2(CO)2]- salt,
which shows its own νCO bands at 2075 and 1985 cm-1
in the infrared spectrum, via addition of H2C(3,5-Me2-
pz)2 (Bpm*) to [Rh2(µ-Cl)2(CO)4]. The complex [Bpm*Rh-
(CO)2]ClO4 (I-ClO4) has also been isolated, and two νCO
stretching frequencies at 2100 and 2035 cm-1 in Nujol
mull have been reported.9 We have found that I-BF4 can
be prepared by carbonylation of [Bpm*Rh(COD)]BF4,
under ambient conditions, the latter being obtained by
addition of AgBF4 to an acetone solution of [Rh2(µ-Cl)2-
(COD)2]. I-BF4 shows four strong νCO stretching bands
at 2092, 2065, 2032 and 1985 cm-1 in KBr pellets and
two bands in dichloromethane solutions (2101 and 2042
cm-1). Similar results are observed in reflectance spec-
tra.
(9) Oro, L. A.; Esteban, M.; Claramunt, R. M.; Elguero, J .; Foces-
Foces, C.; Cano, F. H. J . Organomet. Chem. 1984, 276, 79-97.
(10) Chemical Applications of Group Theory; Cotton, F. A., Ed.;
Wiley: New York, 1990; p 342.

Tandem Carbonylation Reactions
Organometallics, Vol. 22, No. 25, 2003 5263
Thus, carbonylation of [Bpm*Rh(COD)]BF4 provides
exclusively the mononuclear complex [Bpm*Rh(CO)2]-
BF4, and the four νCO bands observed in the solid state
are not explained by the presence of [RhCl2(CO)2]- but
are due to solid-state effects.
CO Su bstitu tion Rea ction s by a n Am in e or a
P h osp h in e. I-BF4 does not react cleanly with amines.
Monitoring the reaction of I-BF4 with NHiPr2, morpho-
line, NHEt2, or piperidine by infrared spectroscopy
shows that 1 equiv of amine does not displace any CO
ligand at room temperature. In the presence of excess
amine, the yellow solution turns rapidly to brown, the
CO ligands being evolved. No well-defined species could
be isolated from these solutions. NMR data show the
presence of free Bpm*, and amine chemical shifts are
consistent with their coordination to rhodium. Since no
hydride type signals are observed, N-H bond activation
remains elusive in these systems. With a large excess
of diethylamine, deposition of abundant quantities of
black rhodium accompanies CO and Bpm* decoordina-
tion.
F igu r e 2. ORTEPIII view of III-BF4 with the atom-
labeling scheme. Ellipsoids are drawn at 50% probability.
Selected bond lengths (Å) and angles (deg): C(1)-Rh(1),
1.814(2); P(1)-Rh(1), 2.2542(9); N(11)-Rh(1), 2.1110(18);
N(21)-Rh(1), 2.1142(17); C(1)-O(1), 1.139(3); C(11a)-P(1),
1.821(2); C(11b)-P(1), 1.836(2); C(11c)-P(1), 1.816(2);
N(11)-N(12), 1.365(3); N(21)-N(22), 1.363(3); C(1)-Rh-
(1)-N(11), 94.58(8); N(11)-Rh(1)-N(21), 82.62(7); C(1)-
Rh(1)-P(1), 87.87(7); N(21)-Rh(1)-P(1), 94.89(5).
Trimethylphosphine or dimethylphenylphosphine does
not substitute any CO ligand in the complex I-BF4, even
under prolonged reaction times, as ascertained by in-
1
frared and H NMR spectroscopy. In contrast, methyl-
diphenylphosphine, triphenylphosphine, and trimethyl
phosphite as well lead to the monosubstituted complexes
[Bpm*Rh(CO)(PMePh2)]BF4 (II-BF4), [Bpm*Rh(CO)-
(PPh3)]BF4 (III-BF4), and [Bpm*Rh(CO){P(OMe)3}]BF4
(IV-BF4), respectively. In CH2Cl2 solution they present
Me2pz)2 ligand, to the C atom of the CO group, and to
the phosphorus atom of the PPh3 ligand.
The Rh distances to the coordinated N atoms (2.1110-
(18) and 2.1142(17) Å) are close to those in the literature
(2.111(8) and 2.097(7) Å) for [BpmRh(COD)]ClO4.9 The
Rh(1)-C(1) and Rh(1)-P(1) (1.814(2) and 2.2542(9) Å)
distances are comparable to those found for the square-
planar derivative [Tp*Rh(CO)(PPh3)] (Rh-C ) 1.824-
(6) Å, Rh-P ) 2.272(2) Å). The two Rh-N distances of
the pyrazolyl groups bonded to the metal center of 2.108-
(4) and 2.119(5) Å compare well with those determined
here.12 Similarly, [Tp*Rh(CO)(PR3)] complexes in which
the phosphine ligands are PMe3 and PMePh2 present
bond distances near those of III-BF4.13 There is almost
no distortion around the rhodium atom from the square
plane, N(11) being 0.21 Å away from the mean plane
calculated from the five atoms C(1), P(1), N(21), N(11),
and Rh(1). The six-membered cycle containing the Rh-
(1), N(11), N(12), C(3), N(22), and N(21) atoms adopts
a boat conformation, with the rhodium atom and the
CH2 group being in the upper positions.
NMR Stu d ies of Com p lexes I-BF 4-IV-BF 4. 1H
NMR spectra of acetone solutions of the dicarbonyl
complex I-BF4 show two methyl signals which are
consistent with a Cs symmetry, each signal being
integrated for six protons at δ 2.47 and 2.59. {1H,1H}
and {1H,13C} 2D spectra show no coupling between the
methyls at the 5-position and the CH2 group, so that
the signals cannot be definitely ascribed. The CH2 group
shows a broad, poorly resolved signal centered at δ 6.70,
indicating an AB system in dynamic exchange. The two
CH’s of the pyrazoles present a singlet at δ 6.31. 13C
NMR spectrum is in agreement with the Cs symmetry
a single νCO band at 2003, 2008, and 2022 cm-1
,
respectively, in accord with a mononuclear geometry
and in agreement with the decreasing basicity of the
phosphorus ligands. These observations suggest that
substitution could occur through a pentacoordinated
transition state; the Bpm* ligand should occupy both
an axial and an equatorial position of a TBP geometry.
Considering for instance that the incoming ligand
approaches the axial position trans to the other axial
pyrazolyl ligand, then as long as its π-back-donation
properties accommodate the σ trans-influence of the
pyrazolyl group, the substitution can occur. This is the
case for less basic ligands such as P(OMe)3, PPh3, and
PMePh2. However, when the ligand is a strong σ-donor
such as an amine or PMe2Ph and PMe3, coordination
does not occur and CO substitution is not observed.
II-BF4, III-BF4, and IV-BF4 are inert toward the
substitution of the second CO ligand. In the [Tp′Rh-
(CO)L] case (L ) phosphines, phosphites) an excess of
phosphorus ligand was not able to displace the second
CO, the pyrazolyl groups being progressively removed
from the coordination sphere.11 In the present case, both
pyrazolyl groups of Bpm* remain coordinated to the
rhodium center.
X-r ay Str u ctu r al Stu dies of [Bpm *Rh (CO)(P P h 3)]-
BF 4‚2(a ceton e) (III-BF 4). The X-ray data collection
and processing parameters for III-BF4 are given in
Table 1. Relevant bond distances and angles, and a view
of the molecule, are shown in Figure 2 and its caption.
The rhodium atom is in its usual square-planar ar-
rangement and is bonded to two N atoms of the H2C(3,5-
(12) Connelly, N. G.; Emslie, D. J . H.; Metz, B.; Orpen, A. G.; Quayle,
M. J . J . Chem. Soc., Chem. Commun. 1996, 2289-2290.
(13) Malbosc, F.; Chauby, V.; Leberre, C.; Etienne, M.; Daran, J .
C.; Kalck, P. Eur. J . Inorg. Chem. 2001, 2689-2697.
(11) Paneque, M.; Sirol, S.; Trujillo, M.; Carmona, E.; Gutie´rrez-
Puebla, E.; Monge, M. A.; Ruiz, C.; Malbosc, F.; Leberre, C.; Kalck, P.;
Etienne, M.; Daran, J . C. Chem. Eur. J . 2001, 7, 3868-3879.

5264 Organometallics, Vol. 22, No. 25, 2003
Teuma et al.
Ta ble 2. Ca ta lyzed Hyd r ofor m yla tion of Oct-1-en ea
oct-1-ene
internal
yield in
conversn, %
octene, %
aldehyde, % (L:B)
time
[Bpm*Rh(CO)(PPh3)]BF4 (III-BF4)
[Bpm*Rh(CO)(PPh3)]BF4 + [RuH2(PPh3)4] (III-BF4 + [Ru])
[Bpm*Rh(CO)2]BF4 (I-BF4)
[Bpm*Rh(CO)(PMePh2)]BF4 (II-BF4)
[Bpm*Rh(CO){P(OMe)3}]BF4 (IV-BF4)
98
95
74
92
7
5
5
75
20
82
95 (83:17)
95 (75:25)
25 (60:40)
80 (65:35)
18 (65:35)
3 h
40 min
3 h
3 h
3 h
a
Experimental conditions: 0.25 mmol of [Rh]BF4, excess of PMePh2 for II-BF4, PPh3 for III-BF4, and P(OMe)3 for IV-BF4 (P:Rh ) 5),
40 mmol of oct-1-ene, 40 mL of THF, 80 °C, 5 bar.
for I-BF4 and the two CO ligands give a doublet at δ
183.1 (1J Rh-C ) 69 Hz).
The 1H NMR spectrum of III-BF4, recorded at 400
MHz, clearly shows the presence of three methyls in
almost the same magnetic environment at δ 2.46, 2.53,
and 2.56. The fourth methyl group at δ 1.54 is presum-
ably shielded by one phenyl group of the PPh3 ligand.
Following the numbering scheme of Figure 2, this
shielded signal could be ascribed to C(211). The CH2
2
F igu r e 3. Consumption of CO/H2 over time.
group leads to an AB system at δ 6.94 with J H-H ) 15
Hz. A 31P{1H} NMR doublet has been detected at δ 44.4
(1J Rh-P ) 159 Hz). The rhodium shift is at δ -91.4 with
the same Rh-P coupling constant. The CO ligand shows
18% aldehydes (65:35)). [Bpm*Rh(CO)(PMePh2)]BF4 (II-
BF4) + 4PMePh2 displays a catalytic activity slightly
lower than that of III-BF4 and provides a 92% conver-
sion of oct-1-ene, 20% of internal octene, and 80% of
aldehydes with a L:B ratio of 65:35.
a
13C NMR doublet at δ 188.6 (1J Rh-C ) 71 Hz).
Complexes II-BF4 and IV-BF4, which contain PMePh2
and P(OMe)3, respectively, have also been characterized
by 1H, 13C, and 31P NMR. The signals are in agreement
with a structure analogous to that of III-BF4. As
observed in the X-ray structure, the two hydrogen atoms
on C(3) are in different environments. One is strongly
influenced by the close vicinity to the rhodium metal
(Rh(1)‚‚‚H ) 2.80 Å), which explains the AB system
observed in the 1H NMR spectrum. In the 13C NMR
spectrum of complex II-BF4, coupling between the
13CO carbon atom and the phosphorus atom gives a
doublet of doublets centered at δ 189.0 (1J Rh-C ) 73;
2J P-C ) 23 Hz) for the CO ligand. The 103Rh shifts in
II-BF4 and IV-BF4 have been measured at δ -175.4 and
-258.9, respectively.
Hyd r ofor m yla tion Rea ction of Oct-1-en e. Neutral
rhodium complexes, particularly [HRh(CO)(PPh3)3],14
are frequently involved in catalytic hydroformylation
reactions. We have explored this reaction under low-
pressure conditions in order to determine the activity
and selectivity of complexes I-BF4-IV-BF4 (Table 2).
Under 5 bar of a 1:1 mixture of H2 and CO and at 80
°C, oct-1-ene is consumed within 3 h in the presence
of the system [Bpm*Rh(CO)(PPh3)]BF4/4PPh3. Some
isomerization of the starting material occurs, since the
resulting mixture contains 5% of internal octenes, in
addition to 95% of aldehydes, with a 83:17 ratio of linear
to branched (L:B) isomers (GC). We did not observe any
hydrogenation of the alkene or of the aldehydes.
Under identical experimental conditions, and with no
additional phosphorus ligand, the dicarbonyl complex
I-BF4 is somewhat less active (74% conversion of oct-
1-ene, internal octenes (75%), nonanals (25%)). Longer
reaction times or higher pressures have little influence
on the course of the reaction. Similarly the activity of
IV-BF4, in the presence of 4 equiv of P(OMe)3, has been
considered. This system presents a poor catalytic activ-
ity (7% conversion of oct-1-ene, 82% internal octenes,
Thus, for these cationic complexes, the optimum
activity as a function of the ligand basicity is rather
sharp. The best results are obtained with the classical
PPh3 ligand. High yields of isomerized product indicate
that CO migratory insertion is not fast enough as
compared to the â-elimination reaction, which restores
an alkene from the alkylrhodium species.
Plots of conversion versus time for the complex III-
BF4 (Figure 3) present a relatively long induction period
of ca. 30 min. Presumably, the likely formation of a
dihydride species is a slow process. Another explanation
could have been the formation of [HRh(CO)(PPh3)3],
which would have then been responsible for the catalytic
activity. However, we have never detected any trace
of this hydrido species after reaction. It is worth
mentioning that the reaction requires an excess of PPh3
to prevent the formation of the dicarbonyl complex
[Bpm*Rh(CO)2]BF4, which would have shown its own
catalytic activity, and particularly an extended isomer-
ization process. In addition, in the case of IV-BF4, the
classical hydrido complex should not be formed, since
the complexes [HRh(CO){P(OR)3}3] are known to be
very active at low pressure and to provide an L:B ratio
of greater than 6.15
[RuH2(PPh3)4], which is known to be a good hydro-
genation catalyst, has been used as a potential hydride
donor (Rh:Ru ) 2) in the present hydroformylation
reaction. Under otherwise identical experimental condi-
tions, the ruthenium complex alone was seen to be
inactive. As shown in Figure 3, the kinetic curve
presents a significantly shorter induction period (ca. 4
min) and 95% of oct-1-ene is converted within 40
min.
(15) (a) Pruett, R. L.; Smith, J . A. J . Org. Chem. 1969, 34, 327-
330. (b) van Leeuwen, P. W. N. M.; Roobeck, C. F. J . Organomet. Chem.
1983, 258, 343-350. (c) For an overview on hydroformylation catalyzed
by rhodium phosphite complexes see: Rhodium Catalyzed Hydro-
formylation; van Leeuwen, P. W. N. M., Claver, C., Eds.; Kluwer
Academic: Dordrecht, The Netherlands, 2000; Chapter 3.
(14) J ardine, F. H. Polyhedron 1982, 1, 569 and references therein.

Tandem Carbonylation Reactions
Organometallics, Vol. 22, No. 25, 2003 5265
Ta ble 3. Ca ta lyzed Hyd r oa m in om eth yla tion of Oct-1-en e by Dieth yla m in ea
oct-1-ene yield in yield in enamine, yield in amine, yield in internal
conversn, % aldehydes (L:B)
% (L:B)
% (L:B)
octene, %
[Bpm*Rh(CO)(PPh3)]BF4 (III-BF4)
[Bpm*Rh(CO)2]BF4 (I-BF4)
[Bpm*Rh(CO)2]BF4 + [RuH2(PPh3)4] (I-BF4 + [Ru])
95
67
99
35 (52:48)
35 (39:61)
24 (90:10)
37 (72:28)
85 (62:38)
30 (72:28)
4
15
2
33 (94:6)
a
Experimental conditions: 0.13 mmol of [Rh]BF4, excess of PPh3 (P:Rh ) 5) for I-BF4, 37.7 mmol of oct-1-ene, 56.55 mmol of NHEt2,
35 mL of THF, 80 °C, 12 bar, time 6 h.
(C6H6)]2,17 or ruthenium-rhodium mixed species gen-
Sch em e 1. Su ccessive Step s of th e
Hyd r oa m in om eth yla tion Rea ction
erated in situ.18 These systems work at high tempera-
tures and pressures (120-150 °C, 55-150 bar of CO:
2H2, respectively). We have previously obtained signifi-
cant rates in the same reaction at 18 bar and 80 °C
starting from a dinuclear rhodium precursor.19
The results of our present studies are summarized
in Table 3.
Under 12 bar of an equimolar mixture of CO and H2
at 80 °C and for 6 h in THF, the complex III-BF4
(substrate to catalyst molar ratio 290) converts 95% of
the oct-1-ene. Longer reaction times give the same
general trends: 4% of internal octenes, 35% of aldehydes
(52:48), 24% of enamines (90:10), and 37% of the
saturated amines (72:28) are formed. As expected, if
condensation of NHEt2 occurs faster with nonanal than
with 2-methyloctanal, the L to B ratio for the amines is
rather high. Thus, the hydrogenation of the enamines,
step 3 of Scheme 1, is slow, but curiously such a kinetics
would give low amounts of the aldehyde, since the
condensation (step 2) is generally considered as easy.20
To improve the hydrogenation step of the enamines,
we have tested the reactivity of the dicarbonyl complex
I-BF4, even if its performances in the hydroformylation
reaction were lower than those of III-BF4. Pleasingly,
the catalytic reaction with I-BF4 in THF results in the
exclusive formation of the saturated amines. The rate
The main role of the ruthenium complex is presum-
ably to activate H2, providing a RuH2(PPh3)x species (or
a RuH2(CO)y(PPh3)z species under a CO atmosphere),
followed by a hydride transfer to rhodium. It is interest-
of oct-1-ene conversion is indeed lower than with III-
ing to compare the first two experiments of Table 2,
BF4, but as shown in Table 3, the yield in amines is
since the isomerization of oct-1-ene is the same (5%) and
much higher (85%), even though we observed 15% of
the reaction was analyzed at roughly the same conver-
isomerized octene. We conclude that precursor I-BF4
sion (98 vs 95%). Similarly, the yields in aldehydes were
accelerates the hydrogenation of the two enamines as
in both cases 95%. The main difference stems from the
compared to complex III-BF4.
regioselectivity. With rhodium alone it is 83:17, whereas
Previous results in this laboratory have shown that
for the [Rh-Ru] system it decreases to 75:25. Thus, the
the addition of ethanol improves significantly the yields
role of cocatalyst of the ruthenium complex is essentially
in amines.19 For I-BF4, this effect has not been observed.
revealed in the activation of dihydrogen and presumably
The slight increase in the conversion of oct-1-ene is
for the formation of the alkyl species through a hydride
accompanied by isomerization (25% of internal octenes).
transfer to rhodium. Turnover frequencies are signifi-
Ethanol addition on III-BF4 has no effect on the yields
cantly higher with the mixed [Rh-Ru] system, for
or selectivities.
instance 228 h-1 at 40 min as compared to 52 h-1 at
[RuH2(PPh3)4] assists the catalytic activity of I-BF4
180 min for III-BF4 alone, at the same conversion. We
and influences the selectivity in amines. The effect is
can reasonably suppose that rhodium alone drives the
somewhat more difficult to analyze, since it introduces
last catalytic steps.
a PPh3 ligand. After 80 min, 99% of oct-1-ene is
Hyd r oa m in om eth yla tion Rea ction of Oct-1-en e
transformed with little isomerization (1% internal
w ith Dieth yla m in e. We have examined the activity
of III-BF4 in the cascade reaction of hydroformylation,
condensation of the resulting aldehydes with diethyl-
amine, and hydrogenation of the enamines into the
saturated amines, usually called hydroaminomethyla-
tion (Scheme 1).
octenes). The mixture of products contains 45% of the
aldehydes, 32% of the enamines, and 22% of the
expected amines. Increasing the reaction time to 6 h
does not significantly modify this breakdown, since 35%
of aldehydes are present, 33% of enamines, and 30% of
(17) Schaffrath, H.; Keim, W. J . Mol. Catal. A 1999, 140, 107-113.
(18) Schulte, M. M.; Herwig, J .; Fischer, R. W.; Kohlpaintner, C.
W. J . Mol. Catal. A 1999, 150, 147-153.
This reaction, discovered by Reppe in 1949, usually
requires high pressures.5,16 Two recent papers sum-
marize the activity of ruthenium complexes, i.e., [RuCl2-
(19) Baig, T.; Molinier, J .; Kalck, P. J . Organomet. Chem. 1993, 455,
219-224.
(20) Advanced Organic Chemistry, 3rd ed.; March, J ., Ed.; Wiley-
Interscience: New York, 1985.
(16) Reppe, W. Experientia 1949, 5, 93.

5266 Organometallics, Vol. 22, No. 25, 2003
Teuma et al.
a capillary column (DB-5 J and W Scientific, 30 m, i.d. 0.32,
film 0.25), with the following temperature program: 50 °C (1
°C/min) then 10 °C/min up to 200 °C.
amines. We propose that the â-H elimination reaction,
which accounts for the primary formation of oct-2-ene,
is slower than the CO migratory insertion. We have no
information to explain why this new system hydroge-
nates the enamines so slowly. We have checked that the
reaction medium still contains diethylamine. Thus, we
suspect that the bimetallic catalytic system inhibits the
aldehyde-amine condensation.
The role of ruthenium, either in hydroformylation or
in hydroaminomethylation, appears very interesting,
and separate studies are in progress to understand the
exact implication of both metals.
Typical oct-1-ene hydroformylation conditions were as fol-
lows: THF (40 mL), oct-1-ene (4.0 × 10-2 mol, 6.3 mL), catalyst
(2.50 × 10-4 mol), 5 bar CO:H2 ) 1:1, 80 °C. In all the kinetic
runs with II-BF4 to IV-BF4, the amount of added phosphine
or phosphite was adjusted so as to operate at a P:Rh molar
ratio equal to 5. For the catalytic run performed with the
addition of the ruthenium hydride complex, 0.5 equiv of the
latter was used.
Typical hydroaminomethylation conditions were as fol-
lows: THF (35 mL), oct-1-ene (3.77 × 10-2 mol, 5.9 mL),
diethylamine (5.65 × 10-2 mol, 5.9 mL), catalyst (1.33 × 10-4
mol), 12 bar CO:H2 ) 1:1, 80 °C. Bis(3,5-dimethylpyrazolyl)-
methane (Bpm*),21,22 [Rh2(µ-Cl)2(COD)2],23 and [RuH2(PPh3)4]
Gen er a l Con clu sion
24
were prepared as described earlier.
The complex [Bpm*Rh(COD)]BF4 can be prepared
in high yields from [Rh2(µ-Cl)2(COD)2], and its carbo-
nylation under mild conditions provides the complex
[Bpm*Rh(CO)2]BF4, which is a mononuclear dicarbonyl
cationic species both in solution and in the solid state.
Monocarbonyl complexes [Bpm*Rh(CO)(PR3)]BF4 have
been synthesized with moderately donating ligands (PR3
) PMePh2, PPh3 P(OMe)3) but not with more electron
releasing ligands.
These complexes are precursors for the low-pressure
hydroformylation of oct-1-ene (particularly [Bpm*Rh-
(CO)(PPh3)]BF4) and, quite interestingly, for the hy-
droaminomethylation of oct-1-ene in the presence of
diethylamine (particularly [Bpm*Rh(CO)2]BF4). Addi-
tion of a hydride-donating complex such as [RuH2-
(PPh3)4] dramatically increases the rate of the hydro-
formylation. For the hydroaminomethylation reaction,
the situation is more contrasted. The overall rate is
faster, and the isomerization is almost suppressed, but
the system remains very slow for both the formation of
enamines and their hydrogenation. Presumably ruthe-
nium transfers a hydride ligand to rhodium. Work is in
progress to identify this step and to gain a better
knowledge of the mechanism of this reaction.
Syn th esis of [(H2C(3,5-Me2p z)2)Rh (COD)]BF 4 Addition
of silver tetrafluoroborate (347 mg, 1.78 mmol) in acetone (30
mL) to a suspension of [Rh2(µ-Cl)2(COD)2] (440 mg, 0.89 mmol)
in acetone (55 mL) gave an immediate precipitate of silver
chloride. The suspension was stirred for 30 min. The precipi-
tate was filtered off, and the filtrate was added to H2C(3,5-
Me2pz)2 (364 mg, 1.78 mmol) in acetone. The resulting yellow
solution was stirred for 20 min and concentrated under
reduced pressure to ca. 10 mL. The complex was precipitated
as a yellow solid by slow addition of diethyl ether, filtered,
and dried under vacuum. Yield: 68% (609 mg, 1.21 mmol).
1H NMR (400 MHz, acetone-d6, 25 °C): δ 1.98 (m, 4H, CH2 of
COD), 2.35 (s, 6H, CH3), 2.51 (s, 6H, CH3), 2.59 (m, 4H, CH2
of COD), 4.70 (s, 4H, CH of COD), 6.07 (s, 2H, CH), 7.29 (AB
2
system, J H-H ) 15 Hz, 2H, CH2). 13C{1H} NMR (100 MHz,
acetone-d6, 25 °C): δ 10.5 (CH3), 13.7 (CH3), 30.5 (CH2 of COD),
59.4 (CH2), 83.0 (CH of COD), 109.0 (CH), 143.6 (C-CH3),
152.7 (C-CH3). 103Rh NMR: δ -142.0. Anal. Calcd for C19H28
-
BF4N4Rh: C, 45.43; H, 5.58; N, 11.16. Found: C, 45.75; H,
5.76; N, 11.20.
Syn th esis of [(H2C(3,5-Me2p z)2)Rh (CO)2]BF 4 (I-BF 4).
Carbon monoxide was bubbled through a solution of [(H2C(3,5-
Me2pz)2)Rh(COD)]BF4 (300 mg, 0.60 mmol) in dichloromethane
(15 mL) for 30 min, causing a color change from yellow to pale
yellow. After concentration under vacuum to ca. 1 mL, diethyl
ether was added and a pale yellow solid separated. This solid
was filtered off, washed with diethyl ether, and dried under
vacuum. Crystallization of the residue from acetone/heptane/
diethyl ether at -20 °C gave yellow crystals. Yield: 60% (161
mg, 0.36 mmol). IR (KBr; ν, cm-1): 2092, 2065, 2032, 1985.
1H NMR (400 MHz, acetone-d6, 25 °C): δ 2.47 (s, 6H, CH3),
2.59 (s, 6H, CH3), 6.31 (s, 2H, CH), 6.7 (m, 2H, CH2). 13C{1H}
NMR (100 MHz, acetone-d6, 25 °C): δ 10.6 (CH3), 14.4 (CH3),
58.4 (CH2), 108.7 (CH), 145.4 (C-CH3), 154.1 (C-CH3), 183.1
Exp er im en ta l Section
Gen er a l P r oced u r es. All manipulations were performed
under a dry, oxygen-free nitrogen or argon atmosphere using
standard Schlenk techniques. Dichloromethane was dried and
distilled over CaH2; THF and Et2O were dried over Na/
benzophenone. Diisopropylamine was distilled and dried over
KOH. Microanalyses were by the ENSIACET. Infrared spectra
were recorded on a Perkin-Elmer Model 1710 instrument.
NMR spectra were recorded at 25 °C on a Bruker AMX 400
MHz spectrometer. The 1H and 13C{1H} resonances of the
solvent were used as the internal standards, but the chemical
shifts are reported with respect to TMS. 31P shifts were
referenced to external 85% H3PO4; ¥(103Rh) ) 3.16 MHz.
Kinetic runs for hydroformylation and hydroaminomethylation
reactions were performed at constant pressure and tempera-
ture in a thermostated reactor. Solid reactants were placed in
a 150 mL stainless steel autoclave, which was evacuated by
means of a vacuum pump prior to the introduction of the liquid
mixture under nitrogen. The reactor was then pressurized with
a CO:H2 ) 1:1 mixture and heated to 80 °C, with vigorous
stirring of the reaction solution (800 rpm). The consumption
of the gaseous mixture was monitored at regular intervals by
a pressure gauge until it stopped. The autoclave was then
cooled to room temperature and the pressure released. The
resulting homogeneous solution was analyzed by GC and IR
spectroscopy. Gas chromatography (GC) analysis was per-
formed on a Carlo Erba HRGC 5160 instrument equipped with
1
(d, J Rh-C ) 69 Hz, CO). 103Rh NMR: δ -108.8. Anal. Calcd
for C13H16BF4N4O2Rh: C, 34.70; H, 3.58; N, 12.45. Found: C,
35.09; H, 3.61; N, 12.16.
Syn th esis of [(H2C(3,5-Me2p z)2)Rh (CO)(P P h 3)]BF 4 (III-
BF 4). To a solution of [(H2C(3,5-Me2pz)2)Rh(CO)2]BF4 (100 mg,
0.22 mmol) in dichloromethane was added triphenylphosphine
(58.3 mg, 0.22 mmol). The mixture was stirred for 30 min. The
yellow solution was concentrated under vacuum to ca. 10 mL,
and addition of diethyl ether gave a yellow solid. This solid
was filtered off, washed with diethyl ether, and dried in vacuo.
Crystallization of the residue from acetone/heptane/diethyl
ether at -20 °C gave light orange crystals. Yield: 89% (135
1
mg, 0.20 mmol). IR (KBr; ν, cm-1): 1998, 1966. H NMR (400
MHz, acetone-d6, 25 °C): δ 1.54 (s, 3H, CH3), 2.46 (s, 3H, CH3),
(21) Trofimenko, S. J . Am. Chem. Soc. 1970, 92, 5118-5126.
(22) J ulia, S.; Sala, P.; Del Mazo, J . M.; Sancho, M.; Ochoa, C.;
Elguero, J .; Fayet, J .-P.; Vertut, M.-C. J . Heterocycl. Chem. 1982, 19,
1141-1145.
(23) Giordano, G.; Crabtree, R. H. Inorg. Synth. 1990, 28, 88-90.
(24) Linn, D. E. J . Chem. Educ. 1999, 76, 70-73.

Tandem Carbonylation Reactions
Organometallics, Vol. 22, No. 25, 2003 5267
2.53 (s, 3H, CH3), 2.56 (s, 3H, CH3), 5.85 (s, 1H, CH), 6.25 (s,
Hz, CO). 31P NMR (160 MHz, acetone-d6, 25 °C): δ 135.3 (d,
2
1H, CH), 6.94 (AB system, J H-H ) 15 Hz, 2H, CH2), 7.59 (m,
1J Rh-P ) 245 Hz, P(OMe)3). 103Rh NMR: δ -258.9 (d, 1J Rh-P
)
15H, PPh3). 13C{1H} NMR (100 MHz, acetone-d6, 25 °C): δ
10.6 (CH3), 13.9 (CH3), 14.3 (CH3), 59.0 (CH2), 107.9 (CH),
108.7 (CH), 131.9 (m, PPh3), 144.8 (C-CH3), 153.5 (C-CH3),
245 Hz). Anal. Calcd for C15H25BF4N4O4PRh: C, 32.99; H, 4.61;
N, 10.26. Found: C, 32.51; H, 4.37; N, 9.11.
X-r a y Str u ctu r e Deter m in a tion . Data for I-BF4 and III-
BF4 were collected on a Stoe IPDS diffractometer equipped
with an Oxford Cryostream nitrogen low-temperature device.
The final unit cell parameters were obtained by the least-
squares refinement of 8000 reflections. Only statistical fluc-
tuations were observed in the intensity monitors over the
course of the data collections.
1
188.6 (d, J Rh-C ) 71 Hz, CO). 31P NMR (160 MHz, acetone-
1
d6, 25 °C): δ 44.4 (d, J Rh-P ) 159 Hz, PPh3). 103Rh NMR: δ
1
-91.4 (d, J Rh-P ) 159 Hz). Anal. Calcd for C30H31BF4N4-
OPRh: C, 51.55; H, 3.91; N, 7.63. Found: C, 52.66; H, 4.57;
N, 8.19. Despite several recrystallizations from acetone/
heptane/diethyl ether, more satisfactory elemental analyses
were not obtained.
The structures were solved by direct methods (SIR97)25 and
refined by least-squares procedures on F using CRYSTAL26
for I-BF4 and on F2 using SHELXL-9727 for III-BF4. All H
atoms attached to C were introduced in calculations in
idealized positions (d(CH) ) 0.96 Å) and treated as riding
models. The weighting scheme used in the last refinement
cycles was w ) w′[1 - {∆F/6σ(Fo)}2]2, where w′ ) 1/∑1nArTr(x)
with three coefficients Ar for the Chebyshev polynomial ArTr-
(x) and x was Fc/Fc(max),28 for I-BF4, whereas the weighting
Syn th esis of [(H2C(3,5-Me2p z)2)Rh (CO)(P MeP h 2)]BF 4
(II-BF 4). To a solution of [(H2C(3,5-Me2pz)2)Rh(CO)2]BF4 (100
mg, 0.22 mmol) in dichloromethane was added methyldiphe-
nylphosphine (44.45 mg, 0.22 mmol). The mixture was stirred
for 30 min. The yellow solution was concentrated under
vacuum to ca. 10 mL, and addition of diethyl ether gave a
yellow solid. This solid was filtered off, washed with diethyl
ether, and dried in vacuo. Yield: 86% (118 mg, 0.19 mmol).
IR (KBr; ν, cm-1): 1992, 1977. 1H NMR (400 MHz, acetone-
scheme w ) 1/[σ2(Fo2) + (aP)2 + bP], where P ) (Fo + 2Fc2)/3,
2
2
d6, 25 °C): δ 1.85 (s, 3H, CH3 of Bpm*), 2.21 (d, J P-H ) 10
was used for III-BF4. The drawings of the molecules (Figures
1 and 2) were realized with the help of ORTEP32.29 Further
details of data collection and refinement are listed in Table 1.
Hz, 3H, PMePh2), 2.42 (s, 3H, CH3 of Bpm*), 2.54 (s, 3H, CH3
of Bpm*), 2.56 (s, 3H, CH3 of Bpm*), 6.05 (s, 1H, CH of Bpm*),
2
6.27 (s, 1H, CH of Bpm*), 6.98 (AB system, J H-H ) 15 Hz,
2H, CH2), 7.79 (m, 10H, PPh2). 13C{1H} NMR (100 MHz,
acetone-d6, 25 °C): δ 10.7 (CH3), 10.8 (CH3), 14.0 (CH3), 14.4
(CH3), 14.7 (d, 1J P-C ) 7 Hz, PMePh2), 58.6 (CH2), 107.7 (CH),
108.1 (CH), 132.7 (m, PPh2), 143.7 (C-CH3), 144.4 (C-CH3),
Ack n ow led gm en t. The Ministe`re de la Recherche
et de la Technologie is gratefully acknowledged for a
research grant to E.T. We are indebted to Engelhardt-
CLAL for a generous loan of rhodium trichloride.
1
152.2 (C-CH3), 152.5 (C-CH3), 189.0 (dd, J Rh-C ) 73 Hz,
2J P-C ) 23 Hz, CO). 31P NMR (160 MHz, acetone-d6, 25 °C): δ
Su p p or tin g In for m a tion Ava ila ble: Tables giving crys-
tallographic data (excluding structure factors) for the struc-
tures reported in this paper. This material is available free of
charge via the Internet at http://pubs.acs.org. These data have
also been deposited with the Cambridge Crystallographic Data
Center as Supplementary Publication Nos. CCDC-193814 (I-
BF4) and CCDC-193815 (III-BF4).
1
30.5 (d, J Rh-P ) 153 Hz, PMePh2). 103Rh NMR: δ -175.4 (d,
1J Rh-P ) 153 Hz).
Syn th esis of [(H2C(3,5-Me2p z)2)Rh (CO){P (OMe)3}]BF 4
(IV-BF 4). To a solution of [(H2C(3,5-Me2pz)2)Rh(CO)2]BF4 (100
mg, 0.22 mmol) in dichloromethane was added trimethyl
phosphite (27.54 mg, 0.22 mmol). The mixture was stirred for
30 min. This yellow solution was concentrated under vacuum
to ca. 10 mL, and addition of diethyl ether gave a yellow solid.
This solid was filtered off, washed with diethyl ether, and dried
in vacuo. Yield: 77% (92.83 mg, 0.17 mmol). IR (KBr; ν, cm-1):
OM030351X
(25) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polodori, G.; Spagn,
R. SIR97: A Program for Automatic Solution of Crystal Structures by
Direct Methods. J . Appl. Crystallogr. 1999, 32, 115.
(26) Watkin, D. J .; Prout, C. K.; Carruthers, J . R.; Betteridge, P.
W. CRYSTALS Issue 11; Chemical Crystallography Laboratory, Uni-
versity of Oxford, Oxford, U.K., 2000.
(27) Sheldrick, G. M. SHELXL97: Program for Crystal Structure
Refinement; University of Go¨ttingen, Go¨ttingen, Germany, 1997.
(28) Prince, E. Mathematical Techniques in Crystallography; Springer-
Verlag: Berlin, 1982.
1
2019, 1987. H NMR (400 MHz, acetone-d6, 25 °C): δ 2.35 (s,
3H, CH3 of Bpm*), 2.44 (s, 3H, CH3 of Bpm*), 2.56 (s, 6H, CH3
3
of Bpm*), 3.84 (d, J H-P ) 13 Hz, 9H, CH3 of P(OMe)3), 6.21
(s, 1H, CH of Bpm*), 6.24 (s, 1H, CH of Bpm*), 6.82 (AB
system, 2J H-H ) 15 Hz, 2H, CH2 of Bpm*). 13C{1H} NMR (100
MHz, acetone-d6, 25 °C): δ 10.5 (CH3 of Bpm*), 14.2 (CH3 of
2
Bpm*), 52.5 (d, J P-C ) 2 Hz, CH3 of P(OMe)3), 58.6 (CH2 of
Bpm*), 108.1 (CH of Bpm*), 108.7 (CH of Bpm*), 143.9 (C-
(29) Farrugia, L. J . ORTEP-32 for Windows. J . Appl. Crystallogr.
1997, 30, 565.
1
CH3 of Bpm*), 144.4 (C-CH3 of Bpm*), 249.8 (d, J Rh-C ) 60