5594 Organometallics, Vol. 29, No. 21, 2010
Alcaraz et al.
Table 2. Composition of the NLMO σnlmo(B-H) and LPnlmo(M) Resulting from the Delocalization of the Parent NBO (σ(BH) and
LP(M)) into the Accepting NBO σ*(MH) and π*(BN), Respectively
7a: M = Ru
8a: M = Rhþ
9a: M = Irþ
σnlmo(B-H)
LPnlmo(M)
0.934 σ(BH) þ 0.319 σ*(MH)
0.938 LP(M) þ 0.297 π*(BN)
0.955 σ(BH) þ 0.266 σ*(MH)
0.976 LP(M) þ 0.176 π*(BN)
0.937 σ(BH) þ 0.318 σ*(MH)
0.964 LP(M) þ 0.223 π*(BN)
˚
procedures and stored under argon over 4 A molecular sieves or over
illustrated by the relative weight of the two components in
the resulting natural localized molecular orbital (NLMO,
Table 2). Due to the cationic nature of the rhodium complex
8a, the valence orbitals of the MH2(PCy3)2 fragment are
lower in energy for M = Rhþ than for M = Ru. Conse-
quently larger σ-donation from the borane should have been
observed, contrary to the results presented in Table 2, where
the composition of the σ(BH) NLMO clearly indicates that
σ-donation is larger with Ru. However the orbitals of the
cationic rhodium complex are also more contracted than
those of the neutral ruthenium complex, and this is detri-
mental to an effective overlap between the interacting orbi-
tals in the σ-donation process. A third-row transition metal
such as iridium has more diffuse valence orbitals, and the
composition of the σ(BH) NLMO is very similar to that of
the neutral ruthenium analogue.
The Lewis structure for the three complexes 7a, 8a, and 9a
features a BdN π-bond as the result of the interaction of the
nitrogen lone pair with the empty p AO on boron. Nevertheless,
back-donation in the π*(BN) orbital from the metal nonbond-
ing dxz AO, acting as a lone pair (LP), is still possible. Table 2
gives the expression of the corresponding NLMO, and it is clear
that back-donation varies according to Ru > Irþ > Rhþ. The
lower energy of the valence orbitals for the cationic fragment
results in less efficient interaction with the vacant π*(BN)
orbital. This reduced interaction is partly compensated for by
the more diffuse character of the orbital in the case of iridium,
yielding a situation intermediate between Ru and Rh. In the
case of Rh, both factors, lower energy and more contracted
orbitals, do not favor efficient back-donation.
potassium for d8-THF. C6H5F and 1,2-C6H4F2 were dried over
˚
CaH2, vacuum distilled, and stored over 3 A molecular sieves.
H2BNiPr2,15 [RuH2(η2-H2)2(PCy3)2],16 [RhH2(η2-H2)2(PCy3)2]-
[BArF ],3a and [IrH2(η2-H2)2(PCy3)2][BArF ] (see Supporting In-
4
4
formation) were prepared by literature methods. NMR spectra were
recorded on a Bruker 400 MHz Avance, a Bruker 300 MHz Avance,
or a Varian Unity Plus 500 MHz spectrometer at room temperature,
unless otherwise stated. Chemical shifts are quoted in ppm and
coupling constants in Hz. Microanalyses were performed at the
Laboratoire de Chimie de Coordination (7) or at the London
Metropolitan University (8, 9).
Synthesis of [RuH2(η2:η2-H2BNiPr2)(PCy3)2] (7). At room
temperature, H2BNiPr2 (74 μL) was added to a toluene solution
(1 mL) of RuH2(η2-H2)2(PCy3)2 (0.0315 g, 0.0471 mmol) and the
solution stirred for 15 h. Evaporation of the solvent and washing of
the resulting solid with cold pentane afforded pure complex 7 as a
beige solid (0.024 g, 66%). Crystals suitable for X-ray analysis were
grown from a saturated pentane solution held at -37 ꢀC. 31P{1H}
1
NMR (C6D6, 121.49 MHz): 77.8 (s). H NMR (C6D6, 300.13
MHz): 3.55 (sept, 2H, 3JHH=6.6 Hz, 2H,CH iPr), 1.16-2.38 (m,
66H, Cy), 1.35 (d, 3JHH=6.6 Hz, 12H, CH3 iPr), -6.91 (br, 2H,
RuH2B), -12.37 (t, 2JHP=25.0 Hz, RuH2). 13C{1H} NMR (C6D6,
100.61 MHz): 47.70 (s, CH iPr), 38.97, 30.79, 28.14, and 17.14 (Cy),
23.86, (s, CH3 iPr). 11B{1H} NMR (C6D6, 128.38 MHz): 45 (br).
Anal. Calcd for C42H84BNP2Ru (776.952 g mol-1): C, 64.93; H,
3
10.90; N, 1.80. Found: C, 65.20; H, 10.88; N, 1.61.
Synthesis of [RhH2(η2:η2-H2BNiPr2)(PCy3)2][BArF4] (8). A
yellow solution of [RhH2(η2-H2)2(PCy3)2][BArF4] (0.093 mmol) in
1,2-C6H4F2 (2 mL) was prepared in situ by hydrogenation (4 atm,
20 min) of [Rh(C7H8)(PCy3)2][BArF4] (0.150 g, 0.093 mmol), as
previously described.3a This solution was placed under an Ar
atmosphere, and H2BNiPr2 (50 μL) was added. The resulting pale
yellow solution was layered with pentane and held at 5 ꢀC to afford
the product as colorless crystals. Yield: 0.085 g (56%). 1H NMR
(C6H5F, 500 MHz): 8.37 (br, 8H, [BArF4]-), 7.68(s, 4H, [BArF4]-),
3.45 (sept, 3JHH = 6.7 Hz, 2H, CH iPr), 1.11-1.91 (m, 66H, Cy),
The results from the NBO analysis are in very good agree-
ment with the variations in the geometrical parameters observed
for the calculated structures. The close values for the B-H bond
distances in 7a and 9a (Table 1) agree with the description of the
σ-donation as being of the same magnitude in the two com-
plexes. The significantly lower values for 8a indicate a less
efficient σ-donation from borane. Nevertheless there is transfer
of electron density from B-H in all three cases, as illustrated by
the lengthening of the B-H bond with respect to the value for
the free borane. The trends in metal to borane back-donation
discussed above are nicely illustrated by the variation of the
i
1.20 (d, 3JHH = 6.6 Hz, 12H, CH3 Pr), -2.30 (br, RhH2B), -15.40
(apparent ddt, 2H, RhH2, 1JHRh = 22, 2JHH = 15, 2JHP = 11 Hz,
from selective decoupling experiments). 13C{1H} NMR (d8-THF,
1
126 MHz): 163.0 (q, JBC = 49, [BArF4]-), 135.8 (s, [BArF4]-),
=
130.2 (qq, 2JFC=32 Hz, 3JBC=3 Hz, [BArF4]-), 125.7 (q, 1JFC
272 Hz, [BArF4]-), 118.3 (br, [BArF4]-), 52.4 (s, CH iPr), 37.6, 31.5,
i
28.2, and 27.2 (Cy), 24.7 (s, CH3 Pr). 11B{1H} NMR (C6H5F, 160
MHz): 34 (vbr, 1B, RhH2B), -5.8 (s, 1B, [BArF4]-).31P{1H} NMR
(C6H5F, 202 MHz): 59.5 (d, JRhP = 103 Hz). Anal. Calcd for
1
˚
˚
M-B bond distance: Ru-B = 2.003 A, Ir-B = 2.088 A, and
C74H96B2F24NP2Rh (1642.0 g mol-1): C, 54.13; H, 5.89; N, 0.85.
˚
Rh-B = 2.159 A. This analysis also correlates with the
3
Found: C, 54.22; H, 5.95; N, 0.83. ESI-MS (1,2-C6H4F2, 60 ꢀC, 4.5
observed NMR data: less efficient σ-donation from the B-H
bond to the metal σ*(ΜH) orbital in 8 results in a lower-field
M-H-B chemical shift, whereas the M-H-B chemical shifts
in 7 and 9 are essentially the same and at higher field. Likewise
the 11B chemical shift for the three isoelectronic complexes
generally tracks the involvement of the π*(ΒΝ) orbital in
kV) positive ion: m/z 778.50 [M]þ (calcd 778.52).
Synthesis of [IrH2(η2:η2-H2BNiPr2)(PCy3)2][BArF
]
4
(9).
Complex 9 was produced in a similar manner to 8 starting from
[IrH2(η2-H2)2(PCy3)2][BArF4], which was generated in situ from
addition of H2 to [IrH{P(η2-C6H9)(C6H11)2}{P(η3-C6H8)-
(C6H11)2}][BArF4] (0.030 g, 0.020 mmol) (see Supporting In-
formation for synthesis and solid-state structure). 1H NMR (500
MHz, C6H5F): 8.35 (br, 8H, [BArF4]-), 7.67 (s, 4H, [BArF4]-),
3.46 (sept, 3JHH = 7 Hz, 2H, CH iPr), 2.05-0.95 (m, Cy), 1.21
(d, 3JHH = 7 Hz, CH3 iPr), -6.58 (br, 2H, IrH2B), -15.15 (dt,
M
chemical shift change from free aminoborane.
B bonding: a smaller contribution leads to a smaller
3 3 3
Experimental Section
General Procedures. Manipulations were done using standard
Schlenk and glovebox techniques (O2 level<1ppm;Arasinertgas).
Solvents were dried using an MBraun solvent purification system.
Deuterated solvents were prepared through freeze-pump-thaw
(15) Euzenat, L.; Horhant, D.; Ribourdouille, Y.; Duriez, C.; Alcaraz,
G.; Vaultier, M. Chem. Commun. 2003, 2280–2281.
(16) Borowski, A. F.; Sabo-Etienne, S.; Christ, M. L.; Donnadieu, B.;
Chaudret, B. Organometallics 1996, 15, 1427–1434.