Terminal phosphinidene complexes CpR(L)M=PAr of the group 9 transition metals cobalt, rhodium, and iridium. Synthesis, structures, and properties

Article abstract of DOI:10.1021/om0208624

Novel terminal rhodium- and cobalt-complexed phosphinidenes, Cp*(PR₃)Rh=PAr (3-5) and Cp(PPh₃)Co=PAr (8), were obtained by dehydrohalogenation of the primary phosphine complexes Cp*RhCl₂(PH₂Ar) (2) and CpCoI₂(PH₂Ar) (7) in the presence of a phosphine. X-ray crystal structures are reported for Cp*(PPh₃)Rh=PMes* (3) and Cp(PPh₃)Co=PMes* (8). A comparative reactivity study and a computational survey were performed on the Co-, Rh-, and Ir-containing phosphinidene complexes. All react with organic dihalides to form phosphaalkenes, with the rhodium congener being far more reactive than the iridium and cobalt complexes. Density functional theory calculations give geometrical parameters and ³¹P NMR chemical shifts in good agreement with experimental data. The rhodium congeners exhibit the most pronounced charge separation of the Rh=P bond, which may explain its higher reactivity. The M-L bond is strong in all Cp(L)M=PH (M = Co, Rh, Ir) complexes and inhibits reactivity at the metal center. Comparisons with the Zr-containing complex Cp₂(PH₃)Zr-PH are made.

Full text of DOI:10.1021/om0208624

Organometallics 2003, 22, 1827-1834
1827
Ter m in a l P h osp h in id en e Com p lexes Cp R(L)MdP Ar of th e
Gr ou p 9 Tr a n sition Meta ls Coba lt, Rh od iu m , a n d
Ir id iu m . Syn th esis, Str u ctu r es, a n d P r op er ties
Arjan T. Termaten,† Halil Aktas,† Marius Schakel,† Andreas W. Ehlers,†
Martin Lutz,‡ Anthony L. Spek,‡ and Koop Lammertsma*,†
Department of Organic and Inorganic Chemistry, Faculty of Sciences, Vrije Universiteit,
De Boelelaan 1083, NL-1081 HV, Amsterdam, The Netherlands, and Bijvoet Center for
Biomolecular Research, Department of Crystal and Structural Chemistry,
Utrecht University, Padualaan 8, NL-3584 CH, Utrecht, The Netherlands
Received October 15, 2002
Novel terminal rhodium- and cobalt-complexed phosphinidenes, Cp*(PR3)RhdPAr (3-5)
and Cp(PPh3)CodPAr (8), were obtained by dehydrohalogenation of the primary phosphine
complexes Cp*RhCl2(PH2Ar) (2) and CpCoI2(PH2Ar) (7) in the presence of a phosphine. X-ray
crystal structures are reported for Cp*(PPh3)RhdPMes* (3) and Cp(PPh3)CodPMes* (8). A
comparative reactivity study and a computational survey were performed on the Co-, Rh-,
and Ir-containing phosphinidene complexes. All react with organic dihalides to form
phosphaalkenes, with the rhodium congener being far more reactive than the iridium and
cobalt complexes. Density functional theory calculations give geometrical parameters and
31P NMR chemical shifts in good agreement with experimental data. The rhodium congeners
exhibit the most pronounced charge separation of the RhdP bond, which may explain its
higher reactivity. The M-L bond is strong in all Cp(L)MdPH (M ) Co, Rh, Ir) complexes
and inhibits reactivity at the metal center. Comparisons with the Zr-containing complex
Cp2(PH3)Zr-PH are made.
In tr od u ction
dppe, RNtC, and CO and Ar ) Mes, Is, and Mes*),6
which were generated by dehydrohalogenation of pri-
mary phosphine complex Cp*IrCl2(PH2Ar) with 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) in the presence of
a stabilizing ligand (L). In this paper we describe the
synthesis and characterization of phosphinidene com-
plexes containing the other transition metals of group
9, namely, rhodium and cobalt. Furthermore, we present
density functional theory (DFT) calculations on model
complexes to demonstrate the role of the triad of metals
(Co, Rh, Ir) and their stabilizing ligands on the philicity
and reactivity of the phosphinidene complexes. In this
analysis comparisons with the well-explored Zr-com-
plexed phosphinidenes are made to highlight differences
in electronic properties.
Transition
metal complexed
phosphinidenes,
LnMdPR, are underdeveloped compared to the analo-
gous carbene and imido complexes.1 Despite the diago-
nal C-P relationship in the periodic table only a few
well-characterized stable phosphinidene complexes are
known. Most of these are nucleophilic and contain an
early to middle transition metal of the second or third
row.2,3 Only few are known with late transition metals.
The cationic complex [Cp*(CO)2RudPN(iPr)2]+, recently
prepared by Carty et al.,4 represents a unique example
of a stable electrophilic complex, and Hillhouse et al.5
reported a nickel complex, (dtbpe)NidP(Dmp) (Dmp )
2,6-dimesitylphenyl; dtbpe ) 1,2-bis(di-tert-butylphos-
phino)ethane), which is the first example of a fully
characterized first-row transition metal complex.
Recently, we described highly stable iridium com-
plexes of the type Cp*(L)IrdPAr (L ) PR3, POR3, AsR3,
Resu lts a n d Discu ssion
The synthesis and characterization of novel rhodium
and cobalt phosphinidene complexes are described first,
followed by a discussion on their reactivity with the
assistance of density functional theory calculations.
A. Syn th esis of Cp *(P R3)Rh dP Ar . The precursors
in our approach are the rhodium-complexed primary
phosphines Cp*RhCl2(PH2Ar) (2a ,b), which are readily
prepared in high yields (86-94%) from commercially
available [Cp*RhCl2]2 (1) and phosphines (Scheme 1)
that contain either the bulky supermesityl (Mes*; 2,4,6-
tri-tert-butylphenyl) group or the less congesting isityl
* Corresponding author. Tel: + 31-20-444-7474. Fax: + 31-20-444-
7488. E-mail: lammert@chem.vu.nl.
† Vrije Universiteit.
‡ Utrecht University.
(1) Mathey, F.; Tran Huy, N. H.; Marinetti, A Helv. Chim. Acta 2001,
84, 2938. (b) Lammertsma, K.; Vlaar, M. J . M. Eur. J . Org. Chem.
2002, 4, 1127.
(2) Cowley, A. H.; Barron, A. R. Acc. Chem. Res. 1988, 21, 81. (b)
Cowley, A. H. Acc. Chem. Res. 1997, 30, 445.
(3) Dillon, K. B.; Mathey, F.; Nixon, J . F. Phosphorus: The Carbon
Copy; Wiley: Chichester, 1998; Chapter 3.
(4) Sterenberg, B. T.; Carty, A. J . J . Organomet. Chem. 2001, 617-
618, 696. (b) Sterenberg, B. T.; Udachin, K. A.; Carty, A. J . Organo-
metallics 2001, 20, 2657. (c) Sterenberg, B. T.; Udachin, K. A.; Carty,
A. J . Organometallics 2001, 20, 4463.
(5) Melenkivitz, R.; Mindiola, D. J .; Hillhouse, G. L. J . Am. Chem.
Soc. 2002, 124, 3846.
(6) Termaten, A. T.; Nijbacker, T.; Schakel, M.; Lutz, M.; Spek, A.
L.; Lammertsma, K. Organometallics 2002, 21, 3196.
10.1021/om0208624 CCC: $25.00 © 2003 American Chemical Society
Publication on Web 04/02/2003

1828 Organometallics, Vol. 22, No. 9, 2003
Termaten et al.
Sch em e 1
The stability of the complexes decreases rapidly with
less crowded aromatic rings such as Mes and with
smaller metal ligands such as CO. For example, dehy-
drohalogenation of Cp*RhCl2(PH2Mes) in the pres-
ence of PPh3 gave in low yield (∼10%) (E)-Cp*(PPh3)-
RhdPMes, which could only be detected by its 31P NMR
chemical shifts at δ 845 and 47 and their characteristic
2J RhP coupling constants of 65 and 232 Hz, respectively.
Introduction of a carbonyl ligand, by leading CO through
the reaction mixture during the dehydrohalogenation
of 2a , led only to intractable material in contrast to the
same procedure that was reported to give the stable
iridium analogue Cp*(CO)IrdPMes*.6 This suggests
that access to rhodium phosphinidene complexes is more
limited than is the case for the iridium analogues. The
imido complexes appear to behave likewise, as iridium
complexes Cp*IrtN-R are stable isolable compounds,
while the rhodium analogues have so far remained
elusive.9,10
(Is; 2,4,6-tri-isopropylphenyl) group. The 31P NMR
spectra of 2a ,b show large diagnostic 1J PH coupling con-
stants of ∼387 Hz and typical 1J RhP values of ∼140 Hz.7
Phosphinidene complexes are then obtained from 2a ,b
in a dehydrohalogenation-ligation sequence. For ex-
ample, addition of a CH2Cl2 solution of 2a to a solution
of DBU and PPh3 in toluene at room temperature
affords dark green colored terminal rhodium complex
3 (Scheme 1) in high yield (89%). The low-field 31P NMR
B. Syn th esis of Cp *(P R3)CodP Ar . The dehydro-
halogenation-ligation sequence that was used for the
Ir and Rh phosphinidene complexes proved to be more
difficult for the cobalt complexes. Even the synthesis of
precursors was problematic, but CpCoI2(Mes*PH2) (7)
could be prepared successfully from [CpCoI2]n (6) and
Mes*PH2 in 78% yield as a dark green solid (Scheme
2), even though it decomposes slowly in solution. Its 31P
NMR spectrum shows a characteristic broad triplet at
1
resonance at δ 868 and the small J RhP coupling con-
stant of 69 Hz are characteristic for a bent phosphini-
dene.2b,3 Its considerable downfield shift with respect
to the heavier iridium analogue (δ 687)6 relates to a
similar difference in chemical shifts between the mo-
lybdenum (δ 800) and tungsten (δ 661) complexes of
Cp2MdPMes*.8 The large 1J RhP coupling constant of 239
Hz for the phosphine resonance at δ 44 suggests a
strong Rh-PPh3 bonding interaction.
The structure of 3, ascertained by an X-ray determi-
nation (Figure 1), is similar to that of the previously
described iridium analogue.6 It has a two-legged piano
stool arrangement with a short Rh(1)-P(1) double bond
of 2.1903(4) Å, a Rh(1)-P(2) single bond of 2.2647(4) Å,
and an acute P(1)-Rh(1)-P(2) angle of 86.524(15)°. The
steric congestion in the system is evident from the
nonplanarity of the Mes* ring with a maximum torsion
angle of 13.8(2)° in the ring. The E configuration for the
RhdP double bond is consistent with the observed large
2J PP coupling constant of 59 Hz; Z isomers have smaller
2J PP values and an even more deshielded δ(31P) for the
phosphinidene group.6
1
δ -49 with a large J PH of 382 Hz.
Addition of a CH2Cl2 solution of 7 to a solution of DBU
and PPh3 in toluene at 0 °C led to the formation of dark
red phosphinidene complex 8 (Scheme 2). The limited
yield (42%) is due to a competing ligand exchange that
leads to the more stable Cp(PPh3)CoI2 11 and Mes*PH2.
This side reaction could not be suppressed and il-
lustrates that Mes*PH2 is only weakly coordinated in
7, which contributes to its instability in solution.
The 31P NMR spectrum of 8 shows resonances at δ
867 and 54 for the phosphinidene and phosphine groups,
respectively. Both are broad due to the high nuclear spin
(7/2) and large electric quadrupole moment of the 59Co
2
nucleus. The J PP coupling constant of 76 Hz suggests
an E configuration for the CodP bond, which was
confirmed by an X-ray crystal structure determination
(Figure 2). The two-legged piano stool geometry has a
short Co(1)-P(1) bond of 2.1102(8) Å, which illu-
strates its double-bond character. The 2.1893(8) Å
Co(1)-P(2) bond distance is normal for a Co-PPh3
single bond, which ranges from 2.15 to 2.23 Å.12 The
P(1)-Co(1)-P(2) angle of 89.96(3)° is less acute than
those of the rhodium and iridium analogues, which may
have its origin in the increased steric congestion that
is caused by the shorter metal-ligand bonds of the
cobalt complex.
The same dehydrohalogenation-ligation sequence,
but using the smaller PMe3 ligand instead of PPh3,
yielded Cp*(PMe3)RhdPMes* (4) as a mixture of E and
Z isomers (35:65) in a combined yield of 79%. The 31P
NMR spectrum of the E isomer exhibits double doublets
at δ 815 (1J RhP ) 66 Hz) and -14 (1J RhP ) 232 Hz), both
2
with a J PP coupling constant of 35 Hz. The Z isomer
shows the expected deshielding for the phosphinidene
resonance (δ 953; 1J RhP ) 86 Hz) and a decrease in J PP
2
(19 Hz), but a similar phosphine chemical shift at δ -7
(1J RhP ) 230 Hz). Dehydrohalogenation of Cp*RhCl2-
(PH2Is) (2b) in the presence of PPh3 afforded in 82%
yield only (E)-Cp*(PPh3)RhdPIs (5), the assignment of
Using CO as stabilizing ligand instead of PPh3
resulted in cobalt phosphinidene complex (Z)-Cp(CO)-
CodPMes*, as indicated by its 31P NMR resonance at δ
which is based on its chemical shifts at δ 859 (1J RhP
)
(9) Glueck, D. S.; Wu, J .; Hollander, F. J .; Bergman, R. G. J . Am.
Chem. Soc. 1991, 113, 2041.
(10) Danopoulos, A. A.; Wilkinson, G.; Sweet, T. K. N.; Hursthouse,
M. B. J . Chem. Soc., Dalton Trans. 1996, 3771.
2
63 Hz) and 45 (1J RhP ) 234 Hz) and J PP coupling
constant of 38 Hz.
(11) King, R. B. Inorg. Chem. 1966, 5, 82. (b) Roe, D. M.; Maitlis, P.
M. J . Chem. Soc. (A) 1971, 3173.
(12) Wakatsuki, Y.; Sakurai, T.; Yamazaki, H. J . Chem. Soc., Dalton
Trans. 1982, 1923. (b) Diversi, P.; Ingrosso, G.; Lucherini, A.; Porzio,
W.; Zocchi, M. Chem. Commun. 1977, 811.
(7) Esteban, M.; Pequerul, A.; Carmona, D.; Lahoz, F. J .; Mart´ın,
A.; Oro, L. A. J . Organomet. Chem. 1991, 402, 421.
(8) Hitchcock, P. B.; Lappert, M. F.; Leung, W. P. J . Chem. Soc.,
Chem. Commun. 1987, 1282.

Terminal Phosphinidene Complexes CpR(L)MdPAr
Organometallics, Vol. 22, No. 9, 2003 1829
F igu r e 1. Displacement ellipsoid plot (50% probability level) of 3. Hydrogen atoms are omitted for clarity. Selected bond
distances (Å), angles (deg), and torsion angles (deg): Rh(1)-P(1) 2.1903(4), Rh(1)-P(2) 2.2647(4), Rh(1)-Cp(cg) 1.9253(9),
P(1)-C(1) 1.8503(16), P(2)-C(29) 1.8374(16), P(2)-C(35) 1.8321(17), P(2)-C(41) 1.8411(16), Rh(1)-P(1)-C(1) 113.94(5),
P(1)-Rh(1)-Cp(cg) 143.25(3), P(1)-Rh(1)-P(2) 86.524(15), P(2)-Rh(1)-Cp(cg) 130.20(3), C(6)-C(1)-C(2)-C(3) 13.8(2),
C(2)-C(1)-C(6)-C(5) -13.2(2).
Sch em e 2
Zr complex, we resorted to DFT calculations on model
systems.
D. Th eor etica l Ca lcu la tion s. In a recent study on
a series of transition metal complexed phosphinidenes,
Ehlers et al.16 showed that their philicity is largely
determined by the ancillary ligands of the metal rather
than by the metal itself. This electronic effect results
from charge transfer between the transition metal
ligands and the phosphorus atom. It can then be argued
that the group 9 phosphinidene complexes are ambi-
philic. Namely, they carry a single π-accepting Cp*
ligand besides an electron-donating phosphine group,
which distinguishes them, for example from the nucleo-
philic Cp2(PMe3)ZrdPMes*14 and the electrophilic
(CO)5WdPR.1,3,17
To evaluate the properties of the group 9 phosphin-
idene complexes, we performed DFT calculations on the
unsubstituted model system, using Cp, CO, and PH3 as
metal ligands. We start by discussing structural and
magnetic properties, followed by an analysis of metal-
ligand bond energies and atomic charges.
Geom etr ies a n d 31P NMR Ch em ica l Sh ifts. As
model Co, Rh, and Ir systems we use (E/ Z)-Cp(PH3)-
MdPH and (E/ Z)-Cp(CO)MdPH. Their optimized ge-
ometries (Cs symmetry) are depicted for M ) Ir in
Figure 3. All E/ Z relative energies, selected bond dis-
tances, and angles are collected in Table 1. The calcu-
lated geometries show good agreement with the X-ray
crystal structures of 8 (Co), 3 (Rh), Cp*(PPh3)IrdPMes*,
1047, but its yield (∼10%) is too low to be isolated. The
more congested Cp* analogue of phosphinidene precur-
sor 7, i.e., Cp*CoI2(PH2Mes*) (9) (δ(31P) -86, 1J PH ) 394
Hz), was too unstable to sustain the dehydrohalogena-
tion-ligation procedure.
C. Rea ctivity of Gr ou p 9 P h osp h in id en e Com -
p lexes. NMR-scale experiments were conducted to
compare the reactivities of Rh complex 3 and Co com-
plex 8 with those of the previously reported Ir com-
plexes.6 All three systems behave similarly, but there
are subtle differences. For example, Rh complex 3 reacts
faster than the others with organic halides, such as
CH2X2 and CHX3 (X ) Cl, Br, I), to give phosphaalkenes
and CpR(PPh3)MX2. Reaction of 3 with CH2I2 in pentane
proceeds within 1 h quantitatively at room temperature,
whereas it takes several days at 60 °C for the iridium
analogue. Moreover, 3 reacts also with CH2Br2 and
CH2Cl2 in decreasing order. The reactivity of 3 increases
by reducing the size of its stabilizing ligand PR3.
Neither of the Co, Rh, and Ir complexed phosphin-
idenes undergo “phospha-Wittig” reactions2b,13 with car-
bonyl groups nor do they give [1 + 2] cycloadditions with
olefins or alkynes. This behavior contrasts that of both
the established nucleophilic Cp2(PMe3)ZrdPMes*14 and
the electrophilic (CO)5WdPR and (CO)4FedPNiPr2.1-3,15
To address the electronic and steric influences of the
group 9 phosphinidene complexes in relationship to the
(13) Shah, S.; Protasiewicz, J . D. Coord. Chem. Rev. 2000, 210, 181.
(14) Breen, T. L.; Stephan, D. W. J . Am. Chem. Soc. 1995, 117,
11914.
(15) Wit, J . B. M.; van Eijkel, G. T.; de Kanter, F. J . J .; Schakel,
M.; Ehlers, A. W.; Lutz, M.; Spek, A. L.; Lammertsma, K. Angew.
Chem. 1999, 111, 2716; Angew. Chem., Int. Ed. 1999, 38, 2596. (b)
Wit, J . B. M.; van Eijkel, G. T.; Schakel, M.; Lammertsma, K.
Tetrahedron 2000, 56, 137.
(16) Ehlers, A. W.; Baerends, E.-J .; Lammertsma, K. J . Am. Chem.
Soc. 2002, 124, 2831.
(17) Mathey, F. Angew. Chem. 1987, 99, 285-296; Angew. Chem.,
Int. Ed. Engl. 1987, 26, 275.

1830 Organometallics, Vol. 22, No. 9, 2003
Termaten et al.
F igu r e 2. Displacement ellipsoid plot (50% probability level) of Cp(PPh3)CodPMes*‚C5H12 (8). n-Pentane and the hydrogen
atoms are omitted for clarity. Selected bond distances (Å), angles (deg), and torsion angles (deg): Co(1)-P(1) 2.1102(8),
Co(1)-P(2) 2.1893(8), Co(1)-Cp(cg) 1.7159(19), P(1)-C(1) 1.861(3) Å, P(2)-C(24) 1.828(3), P(2)-C(30) 1.842(4), P(2)-
C(36) 1.838(3), Co(1)-P(1)-C(1) 109.00(9), P(1)-Co(1)-Cp(cg) 136.74(7), P(1)-Co(1)-P(2) 89.96(3), P(2)-Co(1)-Cp(cg)
132.99(7), C(6)-C(1)-C(2)-C(3) 15.7(4), C(2)-C(1)-C(6)-C(5)-16.4(4).
F igu r e 3. DFT optimized structures (Cs symmetry) for Cp(CO)IrdPH and Cp(PH3)IrdPH.
Ta ble 1. Ca lcu la ted Rela tive En er giesa (k ca l/m ol) a n d Selected Bon d Dista n ces (Å) a n d An gles (d eg) for
(E)- a n d (Z)-Cp (L)MdP H (L ) CO, P H3; M ) Co, Rh , Ir )
complex
energy
MdP
M-L
Cp-M
Cp-MdP
MdP-H
L-MdP
(E)-Cp(CO)CodPH
(Z)-Cp(CO)CodPH
(E)-Cp(PH3)CodPH
(Z)-Cp(PH3)CodPH
(E)-Cp(CO)RhdPH
(Z)-Cp(CO)RhdPH
(E)-Cp(PH3)RhdPH
(Z)-Cp(PH3)RhdPH
(E)-Cp(CO)IrdPH
(Z)-Cp(CO)IrdPH
(E)-Cp(PH3)IrdPH
(Z)-Cp(PH3)IrdPH
+2.75
0.00
+0.34
0.00
+2.10
0.00
+0.43
0.00
+1.30
0.00
-0.46
0.00
2.105
2.090
2.095
2.089
2.215
2.206
2.202
2.195
2.229
2.220
2.218
2.213
1.741
1.723
2.137
2.113
1.869
1.858
2.251
2.230
1.858
1.848
2.239
2.226
1.714
1.724
1.690
1.706
1.949
1.957
1.928
1.949
1.969
1.976
1.942
1.957
134.7
131.9
136.8
132.7
135.6
133.1
137.9
134.4
135.4
132.0
138.4
134.5
99.4
106.7
101.9
107.1
97.9
105.7
100.9
106.8
98.4
90.5
90.7
90.4
91.6
88.2
88.9
86.7
88.8
89.3
91.2
86.8
90.1
105.2
101.4
106.2
a
The energy of the E isomer relative to its Z isomer (0.00 kcal/mol).
and Cp*(CO)IrdPMes*.6 Only the MdP-H angles are
underestimated by 7-13°, which is attributed to the
absence of steric congestion in the model structures.19
The E and Z isomers for each of the model complexes
have similar energies, which agrees with those experi-
ments in which both isomers were obtained. Steric
congestion is considered to be the determining factor
for formation of the E isomer, as it tends to be slightly
less stable.
The E/ Z configuration of the MdP bond hardly effects
the geometrical parameters, but it has a profound
impact on the 31P NMR chemical shifts. This is a well-
known characteristic of phosphaalkenes and is related
(18) Goede, S. J .; van Schaik, H. P.; Bickelhaupt, F. Organometallics
1992, 11, 3844.
(19) Pyykko¨, P. Chem. Rev. 1988, 88, 563.

Terminal Phosphinidene Complexes CpR(L)MdPAr
Organometallics, Vol. 22, No. 9, 2003 1831
Ta ble 2. Th eor etica l 31P NMR Isotr op ic Sh ield in g
Ten sor s, Ch em ica l Sh ifts (p p m ), a n d J P P Cou p lin g
[Cp2ZrO]n and a phosphaalkene.14 We note that the
M-PH3 distance is 0.24 Å longer than the MdP bond
in 10a , whereas this difference amounts to only
0.04-0.07 Å for the group 9 phosphinidene complexes
3, 8, and Cp*(PPh3)IrdPmes*. Evidently, the phosphine
ligand is more tightly bound to the metal in the group
9 complexes, indicating that the intermediate formation
of a 16-electron species is not likely. The calculated
metal-ligand bond dissociation energies (BDEs), listed
in Table 3, seem to support this notion. The BDE for
the M-PH3 bond of Cp2(PH3)ZrdPH amounts to only
22.2 kcal/mol, whereas those of both the E and Z isomers
of the Co and Rh complexes are much higher, with an
average value of 34 kcal/mol, and those of the Ir
complexes are higher still (42 kcal/mol). The ∆EA′ (σ)
and ∆EA′′ (π) contributions to the orbital interaction
energy (∆E°i) confirm PH3 to act as a strong σ-donor/
weak π-acceptor ligand in Cp(PH3)MdPH, as ex-
pected.21,22 The CO ligands are more tightly bound to
the metal with 11.9-18.8 kcal/mol larger BDEs due to
their stronger π-acceptor abilities, i.e., ∆Eσ amounts to
60-70% of ∆Eπ.23,24 The BDEs of the M-CO bonds
follow the Co ≈ Rh < Ir order. The iridium complexes
are the most tightly ligated ones due to stronger
σ-donation, which is caused by relativistic effects that
are known to be more important for the third-row
transition metals.19
The key distinguishing feature of the phosphinidene
complexes, however, is their philicity, which is deter-
mined by the relative charge on the phosphorus atoms
as explored earlier by Ehlers et al.16 In Table 4, the
calculated Hirshfeld charges25 for both the metal and
the phosphinidene phosphorus atoms are listed for all
Co, Rh, and Ir complexes together with those of
Cp2(PH3)ZrdPH.26 It is evident that the negative charge
on phosphorus decreases upon changing the PH3 ligand
in Cp(L)MdPH for Co due to the difference in π-acceptor
abilities.16 Within the nucleophilic PH3-ligated series,
the Rh complexes have a much more polarized MdP
bond than both the Co and Ir congeners, even though
the polarization is far less than the ZrdP bond in
Cp2(PH3)ZrdPH. We presume that this charge separa-
tion is an important factor in the affinity of the phos-
phinidene complexes for other polarized bonds such as
the C-Cl bonds in the dihalides. This would explain
why the Rh-containing complexes are the most reactive
of the group 9 series but still far less reactive than
Cp2(PMe3)ZrdPAr.
2
Con sta n ts for Com p lexed P h osp h in id en es
isotropic
shielding
δ (31P)
2J PP
calcd
δ
31P (2J PP
)
expt
complex
calcda
(E)-Cp(PH3)CodPH
(Z)-Cp(PH3)CodPH
(E)-Cp(PH3)RhdPH
-611.6
-686.3
-603.3
+903
+977
+894
207
66
171
867 (76)b
868 (59)c
815 (35)d
953 (19)d
(Z)-Cp(PH3)RhdPH
(E)-Cp(PH3)IrdPH
-650.8
-437.5
+942
+729
30
216
687 (102)c
629 (84)d
727 (18)d
(Z)-Cp(PH3)IrdPH
(E)-Cp(CO)CodPH
(Z)-Cp(CO)CodPH
(E)-Cp(CO)RhdPH
(Z)-Cp(CO)RhdPH
(E)-Cp(CO)IrdPH
(Z)-Cp(CO)IrdPH
-476.9
-715.4
-752.7
-708.8
-730.4
-538.2
-552.9
+768
+1006
+1044
+1000
+1022
+829
21
1047e
+844
805f
a
b
Relative to PMe3 (δ -62 with respect to 85% H3PO4). (E)-
[Cp(PPh3)CodPMes*]. c (E)-[Cp*(PPh3)MdPMes*] (M ) Rh, Ir).
(E/ Z)-[Cp*(PMe3)MdPMes*] (M ) Rh, Ir). e [Cp(CO)CodPMes*].
d
f [Cp*(CO)IrdPMes*].
to the environment of the lone pair on phosphorus.18
The DFT calculated 31P NMR isotropic shielding tensors
and chemical shifts (referenced against PMe3) and,
2
where applicable, J PP coupling constants are listed in
Table 2. The chemical shifts compare remarkably well
with those observed experimentally, considering the
tremendous difference in steric bulk, and thereby sup-
port the assignment of the more deshielded 31P NMR
chemical shift to the Z isomer. The difference in ∆δ for
the E and Z isomers decreases in the order Co > Rh >
Ir, as is also observed. The phosphinidene chemical
shifts of the Ir complexes are shielded as compared to
the Rh complexes, and these are again more shielded
than those of the Co complexes, which reflects a heavy
metal effect.19 Also the transition metal ligands influ-
ence the chemical shift of the dP-phosphorus, with that
of the CO-ligated structure being the most deshielded
2
one for each of the metal complexes. Last, the J PP
coupling constants for the PH3-ligated complexes are
larger for the E than for the Z forms, although the
applied DFT method appears to significantly overesti-
mate the absolute values.
Bon d En er gies a n d Atom ic Ch a r ges. Stable nu-
cleophilic phosphinidene complexes, including those of
group 9, may well be hampered in their reactivity by
steric congestion of both their ligands and substituents.
However, this is contrasted by the reported reactions
of the bulky Cp2(PMe3)ZrdPAr (Ar ) Mes* (10a ), Dmp
(10b)) with various substrates including carbonyl com-
pounds and organic halides.14,20 This apparent discrep-
ancy requires closer scrutiny.
A distinguishing feature of 10 is the ease by which it
loses or exchanges the phosphine ligand, thereby sug-
gesting that the active species may, in fact, be the 16-
electron complex Cp2ZrdPAr. For example, reaction of
10 with a carbonyl group presumably proceeds via
initial coordination of the oxygen to the metal (with loss
of PMe3), followed by nucleophilic attack of the phos-
phinidene on the carbonyl carbon to form a Zr-P-C-O
cyclic intermediate. Retrocyclization then affords
Con clu sion s
In this paper, the synthesis and characterization of
novel terminal rhodium and cobalt phosphinidene com-
plexes are described. Their synthesis involves dehydro-
(21) Frenking, G.; Wichman, K.; Fro¨hlich, N.; Grobe, J .; Golla, W.;
Le Van, D.; Krebs, B.; La¨ge, M. Organometallics 2002, 21, 2921-2930.
(22) Gonza´les-Blanco, OÅ .; Branchadell, V. Organometallics 1997,
161, 5556-5562.
(23) Frenking, G.; Fro¨hlich, N. Chem. Rev. 2000, 100, 717-774.
(24) Ehlers, A. W.; Ruiz-Morales, Y.; Baerends, E. J .; Ziegler, T.
Inorg. Chem. 1997, 360, 5031-5036.
(25) Hirshfeld, F. L. Theor. Chim. Acta 1977, 44, 129. (b) Wiberg,
K. B.; Rablen, P. R. J . Comput. Chem. 1993, 14, 1504.
(26) Cp2(PH3)ZrdPH] was calculated in
a similar manner (C1
symmetry) as all group 9 phosphinidenes. Some bond distances (Å)
and angles (deg): ZrdP 2.502, Zr-P 2.682, Cp-Zr-Cp 137.1, P-ZrdP
82.4. Calculated 31P NMR isotropic shielding tensor: -530.17 (δ 821
with respect to PMe3; expt δ 792).
(20) Urnezius, E.; Shah, S.; Protasiewicz, J . D. Phosphorus, Sulfur
Silicon 1999, 144-146, 137. (b) Urnezius, E.; Lam, K.-C.; Rheingold,
A. L.; Protasiewicz, J . D. J . Organomet. Chem. 2001, 630, 193.

1832 Organometallics, Vol. 22, No. 9, 2003
Termaten et al.
Ta ble 3. Ca lcu la ted Bon d En er gies a n d Bon d Dissocia tion En er gies (k ca l/m ol)a for M-L in [Cp (L)MdP H]
a n d [Cp 2(P H3)Zr dP H]
∆EA′
∆EA′′
∆Eσ
∆Eπ
∆E°i
∆E°
∆Etït
BDE
(E)-Cp(PH3)CodPH
(Z)-Cp(PH3)CodPH
(E)-Cp(PH3)RhdPH
(Z)-Cp(PH3)RhdPH
(E)-Cp(PH3)IrdPH
(Z)-Cp(PH3)IrdPH
(E)-Cp(CO)CodPH
(Z)-Cp(CO)CodPH
(E)-Cp(CO)RhdPH
(Z)-Cp(CO)RhdPH
(E)-Cp(CO)IrdPH
(Z)-Cp(CO)IrdPH
Cp2(PH3)ZrdPHb
-48.9
-53.3
-51.6
-53.2
-72.1
-76.5
-66.6
-69.9
-72.6
-71.3
-99.2
-100.9
-13.5
-13.9
-13.9
-14.1
-17.6
-17.7
-31.8
-32.7
-30.9
-31.5
-39.1
-39.9
-35.4
-39.4
-37.7
-39.1
-54.5
-58.8
-34.8
-37.2
-41.7
-39.8
-60.1
-61.0
-27.0
-27.8
-27.8
-28.2
-35.2
-35.4
-63.6
-65.4
-61.8
-63.0
-78.2
-79.8
-62.4
-67.2
-65.5
-67.3
-89.7
-94.2
-98.4
-102.6
-103.5
-102.8
-138.3
-140.8
-39.1
23.9
21.4
26.1
21.1
36.4
36.9
41.4
36.3
49.6
41.8
64.6
62.1
15.0
-38.5
-45.8
-39.4
-46.2
-53.3
-57.2
-57.0
-66.2
-53.9
-61.0
-73.7
-78.7
-24.1
33.5
33.8
34.4
34.8
42.2
41.8
49.8
52.6
46.3
48.4
58.3
59.6
22.2
n.d.c
n.d.c
n.d.c
n.d.c
a
b
For definitions, see Computational Section. See ref 21 for structural data. c Not determined due to low symmetry (C1) of the system.
Ta ble 4. Ca lcu la ted Hir sh feld Ch a r ges of M a n d P
the shells of lower energy were treated by the frozen core
approximation. All calculations were performed at the nonlocal
exchange self-consistent field (NL-SCF) level, using the local
density approximation (LDA) in the Vosko-Wilk-Nusair
parametrization28 with nonlocal corrections for exchange
(Becke88)29 and correlation (Perdew86).30 All geometries were
optimized using the analytical gradient method implemented
by Versluis and Ziegler,31 including relativistic effects by the
zero-order regular approximation (ZORA).32
in Cp (L)MdP H a n d Cp 2(P H3)Zr dP H
complex
M
P
(E)-Cp(PH3)CodPH
(Z)-Cp(PH3)CodPH
(E)-Cp(PH3)RhdPH
(Z)-Cp(PH3)RhdPH
(E)-Cp(PH3)IrdPH
(Z)-Cp(PH3)IrdPH
(E)-Cp(CO)CodPH
(Z)-Cp(CO)CodPH
(E)-Cp(CO)RhdPH
(Z)-Cp(CO)RhdPH
(E)-Cp(CO)IrdPH
(Z)-Cp(CO)IrdPH
Cp2(PH3)ZrdPH
-0.086
-0.085
+0.108
+0.107
-0.027
-0.029
-0.038
-0.036
+0.163
+0.162
+0.044
+0.042
+0.333
-0.066
-0.056
-0.111
-0.101
-0.089
-0.087
+0.004
+0.010
-0.046
-0.045
-0.021
-0.024
-0.270
The 31P NMR chemical shift tensors were calculated with
ADF’s NMR program,33 including an all-electron basis set and
relativistic effects (ZORA) for the phosphorus atoms. All other
atoms were treated as mentioned above. All model complexes
were calculated without molecular symmetry. The total iso-
tropic shielding tensors were referenced against PMe3, which
has a value of 353.08. It has an experimental chemical shift
2
at δ -62 with respect to 85% H3PO4. The J PP coupling
halogenation of the primary phosphine complexes
Cp*RhCl2(PH2Ar) (2) and CpCoI2(PH2Mes*) (7) by DBU
in the presence of a stabilizing phosphine ligand.
X-ray crystal structures are presented for Cp*(PPh3)-
RhdPMes* (3) and Cp(PPh3)CodPMes* (8), thereby
completing the first (group 9) transition metal triad of
phosphinidene complexes. All react with organic diha-
lides, with the rhodium congener being by far the most
reactive one. No reactivity toward CdC, CtC, and
CdO bonds was observed.
DFT calculations on the model systems (E/ Z)-Cp-
(PH3)MdPH and (E/ Z)-Cp(CO)MdPH show the geo-
metrical parameters and 31P NMR chemical shifts to be
in good agreement with experimental data. Calculated
Hirshfeld charges show the RhdP bond to be much more
polarized than the iridium and cobalt analogues, which
explains their difference toward organic halides.
The M-L ligand bonds of the group 9 complexes
Cp(L)MdPH are much stronger than that of Cp2(PH3)-
ZrdPH and make ligand exchange or loss an unlikely
process. This, together with the low oxophilicity of the
group 9 transition metals, makes these phosphinidene
complexes unreactive toward carbonyl groups.
constants were calculated with a preliminary version of ADF’s
CCL program.34
The metal-ligand bonds were analyzed with ADF’s estab-
lished energy decomposition35 into an exchange repulsion plus
electrostatic interaction energy part (∆E°) and an orbital
interaction energy (charge transfer, polarization) part (∆E°i).
The energy necessary to convert fragments from their ground-
state equilibrium geometries to the geometry and electronic
state they acquire in the complex is represented by a prepara-
tion energy term (∆Eprep). The overall bond energy (∆Etot) is
formulated as ∆Etot ) ∆E° + ∆E°i + ∆Eprep
.
Note that ∆Etot is defined as the negative of the bond
dissociation energy (BDE), i.e., ∆Etot ) E(molecule) - ∑E(frag-
ments), thereby giving negative values for stable bonds. The
orbital interaction term ∆E°i accounts for charge transfer
(interactions between occupied orbitals on one fragment with
unoccupied orbitals on the other, including HOMO-LUMO
(27) te Velde, G.; Bickelhaupt, F. M.; van Gisbergen, S. J . A.; Fonseca
Guerra, C.; Baerends, E. J .; Snijders, J . G.; Ziegler, T. Chemistry with
ADF. J . Comput. Chem. 2001, 22, 931. (b) Fonseca Guerra, C.; Snijders,
J . G.; te Velde, G.; Baerends, E. J . Theor. Chem. Acc. 1998, 99, 391.
(c) ADF2000.02, SCM, Theoretical Chemistry, Vrije Universiteit,
Amsterdam, The Netherlands, http://www.scm.com.
(28) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J . Phys. 1992, 70, 84.
(29) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
(30) Perdew, J . P. Phys. Rev. B 1986, 33, 8822.
(31) Fan, L.; Versluis, L.; Ziegler, T.; Baerends, E. J .; Raveneck, W.
Int. J . Quantum. Chem.; Quantum. Chem. Symp. 1988, S22, 173. (b)
Versluis, L.; Ziegler, T. Chem. Phys. 1988, 322, 88.
Com p u ta tion a l Section
(32) van Lenthe, E.; Ehlers, A. W.; Baerends, E. J . J . Chem. Phys.
1999, 110, 8943.
All calculations were performed using the parallelized
Amsterdam density functional (ADF) package (version
2000.02).27 All atoms were described by a triple-ú basis set with
polarization functions, corresponding to basis set IV in the
ADF. The 1s core shell of carbon and oxygen and the 1s2s2p
core shells of phosphorus were treated by the frozen core
approximation. The metal centers were described by a triple-ú
basis set for the outer ns, np, nd, and (n + 1)s orbitals, whereas
(33) Schreckenbach, G.; Ziegler, T. J . Phys. Chem. 1995, 99, 606-
611. (b) Schreckenbach, G.; Ziegler, T. Int. J . Quantum Chem. 1997,
61, 899. (c) Wolff, S. K.; Ziegler, T. J . Chem. Phys. 1998, 109, 895.
(34) Autschbach, J .; Ziegler, T J . Chem. Phys. 2000, 113, 936. (b)
Autschbach, J .; Ziegler, T J . Chem. Phys. 2000, 113, 9410.
(35) Morokuma, K. Acc. Chem. Res. 1977, 10, 297. (b) Ziegler, T.;
Rauk, A. Inorg. Chem. 1979, 18, 1755. (c) Ziegler, T.; Rauk, A. Theor.
Chim. Acta 1977, 46, 1.

Terminal Phosphinidene Complexes CpR(L)MdPAr
Organometallics, Vol. 22, No. 9, 2003 1833
3
1
interactions) and polarization (empty/occupied orbital mixing
on the same fragment). The charge transfer part is the result
of both σ-donation from the ligand to the metal and π-back-
donation from the metal into the unoccupied orbitals of the
ligand. We used the extended transition state (ETS) method
developed by Ziegler and Rauk to decompose ∆E°i into
contributions from each irreducible representation of the
interacting system.31 As the M-L bonds analyzed in this case
do not exhibit a clear σ,π-separation, we can only make
Hz, J PC ) 3.3 Hz, C5(CH3)5), 114.9 (d, J PC ) 40.2 Hz, i-Is),
3
4
122.2 (d, J PC ) 8.0 Hz, m-Is), 153.0 (d, J PC ) 2.6 Hz, p-Is),
153.5 (d, 2J PC ) 7.1 Hz, o-Is). IR (KBr, cm-1): ν(PH) 2382, 2396.
Anal. Calcd for C25H40PCl2Rh: C, 55.06; H, 7.40; P, 5.68.
Found: C, 54.68; H, 7.35; P, 5.40.
[Cp *(P P h 3)R h dP Mes*] (3). A red-brown solution of 2a
(117 mg, 0.20 mmol) in CH2Cl2 (2.5 mL) was added dropwise
to a solution of DBU (59.8 µL, 0.40 mmol) and PPh3 (52.5 mg,
0.20 mmol) in toluene (5 mL) and stirred for an additional 5
min. After removal of the solvents the residue was extracted
with n-pentane (25 mL), filtered, and concentrated. Cooling
at -20 °C yielded green-black crystals of 3 (138 mg, 0.178
mmol, 89%). Mp: 162-164 °C. 31P NMR (C6D6): δ 867.6 (dd,
1J RhP ) 68.6 Hz, 2J PP ) 59.4 Hz, RhdP), 44.1 (dd, 1J RhP ) 239.1
Hz, 2J PP ) 59.4 Hz, Rh-PPh3). 1H NMR (C6D6): δ 1.33 (d, 4J PH
) 1.14 Hz, 15H, C5(CH3)5), 1.50 (s, 18H, o-C(CH3)3), 1.51 (s,
9H, p-C(CH3)3), 7.04-7.20 (m, 9H, PPh3), 7.53 (s, 2H, m-Mes*),
7.90 (m, 6H, PPh3). 13C{1H} NMR (C6D6): δ 10.2 (s, C5(CH3)5),
estimations of the contributions of σ-donation (∆EA′ - ∆EA′′
)
and π-back-bonding (2∆EA′′) assuming near degeneracy of the
ligand π* orbitals and negligible ligand to metal π-donation.21
Exp er im en ta l Section
Gen er a l Da ta . All experiments were performed in flame-
dried glassware and under an atmosphere of dry nitrogen or
argon. Solvents were distilled (under N2) from sodium (tolu-
ene), sodium benzophenone (THF), diphosphoruspentoxide
(CH2Cl2, CHCl3), or lithium aluminum hydride (n-pentane).
Deuterated solvents were dried over 4 Å molecular sieves
(CDCl3, C6D6). All solid starting materials were dried in vacuo.
1H, 13C, and 31P NMR spectra were recorded at 300 K on a
Bruker Avance 250 spectrometer at 250.13, 62.90, and 101.25
MHz, respectively. 1H NMR spectra were referenced to CHCl3
(δ 7.27 ppm) or C6D5H (δ 7.17 ppm), 13C NMR spectra to CDCl3
(δ 77.16 ppm) or C6D6 (δ 128.06 ppm), and 31P NMR spectra
to external 85% H3PO4. IR spectra were recorded on a Mattson-
6030 Galaxy FTIR spectrophotometer, and high-resolution
mass spectra (HRMS) on a Finnigan Mat 900 spectrometer.
Elemental analyses were performed by Mikroanalytisches
Labor Pascher, Remagen-Bandorf, Germany. [Cp*RhCl2]2,36
[CpCoI2]2,11b IsPH2,37 and Mes*PH2,38 were prepared according
to literature procedures.
4
32.1 (d, J PC ) 4.7 Hz, o-C(CH3)3), 32.2 (s, p-C(CH3)3), 35.0 (s,
2
p-C(CH3)3), 38.7 (s, o-C(CH3)3), 97.5 (d, J PC ) 4.2 Hz,
C5(CH3)5), 121.0 (s, m-Mes*), 127.8 (d, 3J PC ) 9.9 Hz, m-PPh3),
4
2
129.3 (d, J PC ) 1.8 Hz, p-PPh3), 135.5 (d, J PC ) 11.3 Hz,
o-PPh3), 136.7 (d, J PC ) 40.6 Hz, i-PPh3), 144.5 (s, o-Mes*),
146.4 (s, p-Mes*). HRMS: calcd for C46H59P2Rh 776.31470,
found 776.31745.
[Cp *(P Me3)Rh dP Mes*] (4). In a fashion similar to that
described for 3, 2a (117 mg, 0.20 mmol), DBU (59.8 µL, 0.40
mmol), and PMe3 (0.20 mL of a 1.0 M solution in toluene, 0.20
mmol) were used to give dark green crystals of 4 (93 mg, 0.158
mmol, 79%) as a mixture of isomers. Major isomer (65%)
1
(Z)-4. 31P NMR (C6D6): δ 953 (dd, J RhP ) 86 Hz, J PP ) 19
Hz, RhdP), -6.9 (dd, 1J RhP ) 230 Hz, 2J PP ) 19 Hz, Rh-PPh3).
1H NMR (C6D6): δ 0.78 (dd, 2J PH ) 8.5 Hz, 4J PH ) 1.2 Hz, 9H,
P(CH3)3), 1.44 (s, 9H, p-C(CH3)3), 1.50 (s, 18H, o-C(CH3)3), 1.96
(bs, 15H, C5(CH3)5), 7.46 (bs, 2H, m-Mes*). Minor isomer (35%)
1
2
[Cp *Rh Cl2(P H2Mes*)] (2a ). Mes*PH2 (0.58 g, 2.10 mmol)
was added to a dark red-brown solution of [Cp*RhCl2]2 (0.62
g, 1.00 mmol) in CHCl3 (50 mL), which was stirred for 3 h at
50 °C and filtered to remove insoluble material. After concen-
trating the solution to ∼5 mL, n-pentane (30 mL) was added
slowly to cause precipitation of an orange-red powder, which
was isolated by filtration, washed with n-pentane (25 mL), and
dried in vacuo. Recrystallization from CH2Cl2/n-pentane yielded
2a as red needles (1.10 g, 1.87 mmol, 94%). Mp > 240 °C (dec).
(E)-4. 31P NMR (C6D6): δ 815 (dd, J RhP ) 66 Hz, J PP ) 35
1
2
1
2
Hz, RhdP), -14.5 (dd, J RhP ) 232 Hz, J PP ) 35 Hz,
Rh-PPh3). 1H NMR (C6D6): δ 1.50 (s, 9H, p-C(CH3)3), 1.53 (d,
2J PH ) 8.4 Hz, 9H, P(CH3)3), 1.53 (bs, 15H, C5(CH3)5), 1.63 (s,
18H, o-C(CH3)3), 7.54 (bs, 2H, m-Mes*). HRMS: calcd for
31H53P2Rh 590.26776, found 590.26851.
[Cp *(P P h 3)Rh dP Is] (5). In a fashion similar to that
C
described for 3, 2b (136 mg, 0.25 mmol), DBU (75 µL, 0.50
mmol), and PPh3 (66 mg, 0.25 mmol) were used to give dark
green 5 (151 mg, 0.206 mmol, 82%). The compound could not
be recrystallized properly due to high solubility and therefore
1
31P NMR (CDCl3): δ -43.1 (doublet of triplets, J RhP ) 143
1
Hz, J PH ) 387 Hz). 1H NMR (CDCl3): δ 1.32 (s, 15H,
C5(CH3)5), 1.34 (s, 9H, p-C(CH3)3), 1.55 (s, 18H, o-C(CH3)3), 6.38
1
4
contains some impurities. 31P NMR (C6D6): δ 859.5 (dd, J RhP
1
(d, J PH ) 387 Hz, 2H, PH2), 7.48 (d, J PH ) 2.5 Hz, 2H,
m-Mes*). 13C{1H} NMR (CDCl3): δ 9.56 (s, C5(CH3)5), 31.6
2
1
2
) 63 Hz, J PP ) 38 Hz, RhdP), 45.4 (dd, J RhP ) 234 Hz, J PP
) 38 Hz, Rh-PPh3). 1H NMR (CDCl3): 1.04 (d, 3J HH ) 6.8 Hz,
6H, o-CH(CH3)2), 1.43 (d, 3J HH ) 6.9 Hz, 6H, p-CH(CH3)2), 1.44
4
(s, p-C(CH3)3), 33.6 (d, J PC ) 1.9 Hz, o-C(CH3)3), 35.5 (s,
p-C(CH3)3), 38.6 (d, 3J PC ) 1.6 Hz, o-C(CH3)3), 99.1 (d, 1J RhC
)
1
4
3
7.2 Hz, C5(CH3)5), 116.3 (d, J PC ) 24.8 Hz, i-Mes*), 123.0 (d,
3J PC ) 9.4 Hz, m-Mes*), 152.7 (s, p-Mes*), 156.4 (d, 2J PC ) 3.8
Hz, o-Mes*). IR (KBr, cm-1): ν(PH) 2401, 2435. HRMS: calcd
for C28H46PCl2Rh 586.17694, found 586.174664.
(d, J PH ) 1.2 Hz, 15H, C5(CH3)5), 1.56 (d, J HH ) 6.8 Hz, 6H,
3
o-CH(CH3)2), 3.02 (sep, J HH ) 6.9 Hz, 1H, p-CH(CH3)2), 3.12
(sep, 3J HH ) 6.8 Hz, 2H, o-CH(CH3)2), 7.05-7.17 (m, 9H, PPh3),
7.22 (s, 2H, m-Is), 7.90-7.95 (m, 6H, PPh3). 13C{1H} NMR
(CDCl3): 9.7 (s, C5(CH3)5), 23.2 (s, p-CH(CH3)2), 24.8 (s, o-CH-
(CH3)2), 24.9 (s, o-CH(CH3)2), 31.4 (bs, o-CH(CH3)2), 35.0 (s,
p-CH(CH3)2), 97.0 (dd, 1J RhC ) 4.2 Hz, 2J PC ) 3.0 Hz, C5(CH3)5),
119.7 (s, m-Is), 127.6 (d, 3J PC ) 9.7 Hz, m-PPh3), 129.3 (d, 4J PC
[Cp *Rh Cl2(P H2Is)] (2b). In a fashion similar to that
described for 2a , IsPH2 (0.50 g, 2.10 mmol) was used to give
orange crystals of 2b (0.94 g, 1.72 mmol, 86%). Mp > 190 °C
1
(dec). 31P NMR (CDCl3): δ -59.1 (doublet of triplets, J RhP
)
)
1
1
3
2
141 Hz, J PH ) 386 Hz). H NMR (CDCl3): δ 1.25 (d, J HH
) 2.0 Hz, p-PPh3), 135.5 (d, J PC ) 11.4 Hz, o-PPh3), 137.0 (d,
3
1J PC ) 39.9 Hz, i-PPh3), 143.6 (s, p-Is), 147.2 (s, o-Is). HRMS:
calcd for C43H53P2Rh 734.26776, found 734.26837.
6.9 Hz, 6H, p-CH(CH3)2), 1.29 (d, J HH ) 6.6 Hz, 12H, o-CH-
(CH3)2), 1.55 (d, 4J PH ) 4.2 Hz, 15H, C5(CH3)5), 2.89 (sep, 3J HH
3
) 6.9 Hz, 1H, p-CH(CH3)2), 3.18 (sep, J HH ) 6.6 Hz, 2H,
[Cp CoI2(P H2Mes*)] (7). To a suspension of [CpCoI2]n (1.0
g, 2.65 mmol) in CH2Cl2 (25 mL) was added a solution of
Mes*PH2 (0.75 g, 2.70 mmol) in CH2Cl2 (5 mL). The resulting
dark green mixture was stirred for 1 h at room temperature,
after which it was filtered and concentrated to a few milliliters.
Subsequent addition of n-pentane and cooling to 0 °C caused
formation of a green-black precipitate that was isolated by
filtration, washed with cold n-pentane, and dried in vacuo (1.36
g, 2.07 mmol, 78%). The solid is stable, but in solution, 7 slowly
decomposes, which hampers further purification by crystal-
o-CH(CH3)2), 5.57 (d, 1J PH ) 386 Hz, 2H, PH2), 7.08 (d, 3J PH
)
3.1 Hz, 2H, m-Is). 13C{1H} NMR (CDCl3): δ 9.44 (s, C5(CH3)5),
3
24.2 (s, p-CH(CH3)2), 24.6 (s, o-CH(CH3)2), 33.0 (d, J PC ) 9.4
Hz, o-CH(CH3)2), 34.8 (s, p-CH(CH3)2), 99.1 (dd, J RhC ) 7.1
2
(36) White, C.; Yates, A.; Maitlis, P. M. Inorg. Synth. 1992, 29, 228.
(37) van den Winkel, Y.; Bastiaans, H. M. M.; Bickelhaupt, F. J .
Organomet. Chem. 1991, 405, 183.
(38) Cowley, A. H.; Kilduff, J . E.; Newman, T. H.; Pakulski, M. J .
Am. Chem. Soc. 1982, 104, 5820.

1834 Organometallics, Vol. 22, No. 9, 2003
Termaten et al.
lization. Mp > 140 °C (dec). 31P NMR (CDCl3): δ -48.9 (bt,
direct methods (SIR-97)40 and refined with the program
SHELXL-9741 against F2 of all reflections. Non hydrogen atoms
were refined freely with anisotropic displacement parameters.
Hydrogen atoms were refined freely with isotropic displace-
ment parameters (compound 3) or as rigid groups (compound
8). The drawings, structure calculations, and checking for
higher symmetry was performed with the program PLATON.42
Com p ou n d 3: C46H59P2Rh, fw ) 776.78, dark green prism,
0.25 × 0.25 × 0.13 mm3, triclinic, space group P1h (no. 2). Cell
parameters: a ) 9.7539(1) Å, b ) 14.3652(2) Å, c ) 16.8717-
(2) Å, R ) 64.9564(7)°, â ) 78.5027(7)°, γ ) 72.6990(7)°, V )
2037.47(4) Å3. Z ) 2, F ) 1.266 g cm-3, F000 ) 820. 38 294
reflections were measured up to a resolution of (sin θ/λ) ) 0.65
Å-1; 9302 reflections were unique (Rint ) 0.0637). An absorp-
tion correction based on multiple measured reflections was
applied (µ ) 0.528 mm-1, 0.71-0.93 transmission). 678 refined
parameters, no restraints. R (obs reflns): R1 ) 0.0281, wR2
) 0.0702. R (all data): R1 ) 0.0326, wR2 ) 0.0728. Weighting
1
1J PH ) 382 Hz). H NMR (CDCl3): δ 1.34 (s, 9H, p-C(CH3)3),
1
1.71 (s, 18H, o-C(CH3)3), 4.85 (s, 5H, C5H5), 7.06 (d, J PH
)
382 Hz, 2H, PH2), 7.59 (d, J PH ) 2.9 Hz, 2H, m-Mes*). 13C
NMR (CDCl3): δ 30.9 (s, p-C(CH3)3), 33.6 (s, o-C(CH3)3), 35.2
(s, p-C(CH3)3), 38.4 (d, J CP ) 2.3 Hz, o-C(CH3)3), 85.2 (d, J CP
) 3.3 Hz, C5H5), 122.0 (d, 1J PC ) 30 Hz, i-Mes*), 123.5 (d, 3J PC
) 9.8 Hz, m-Mes*), 152.8 (s, p-Mes*), 154.7 (d, J PC ) 5.3 Hz,
4
3
2
2
o-Mes*). IR (KBr, cm-1): ν(PH) 2375, 2346.
[Cp (P P h 3)CodP Mes*] (8). A dark green solution of
[Cp(PH2Mes*)CoI2] (7; 328 mg, 0.50 mmol) in CH2Cl2 (2.0 mL)
was added slowly to a stirred solution of DBU (150 µL, 1.00
mmol) and PPh3 (131 mg, 0.50 mmol) in toluene (5 mL) at 0
°C. The resulting dark red solution was warmed to room
temperature and stirred for an additional 5 min. After removal
of the solvents, the residue was extracted into n-pentane (25
mL), filtered, and concentrated. Crystallization at -20 °C
yielded dark red crystals of 8‚C5H12 (153 mg, 0.21 mmol, 42%).
scheme w ) 1/[σ2(Fo2) + (0.0310P)2 + 0.9560P], where P ) (Fo
2
2
Mp: 83-84 °C. 31P NMR (C6D6): δ 866.9 (d, J PP ) 76 Hz,
+ 2Fc2)/3. GoF ) 1.030. Residual electron density between
2
1
CodP), 54.0 (bd, J PP ) 76 Hz, PPh3). H NMR (C6D6): δ 1.49
(s, 18H, o-C(CH3)3), 1.55 (s, 9H, p-C(CH3)3), 4.12 (s, 5H, C5H5),
7.07-7.15 (m, 9H, PPh3), 7.64 (s, 2H, m-Mes*), 7.86-7.94 (m,
6H, PPh3). 13C NMR (C6D6): δ 31.8 (s, p-C(CH3)3), 31.9 (bs,
o-C(CH3)3), 35.2 (s, p-C(CH3)3), 38.8 (s, o-C(CH3)3), 82.2 (d, 2J PC
-0.64 and 0.54 e/Å3.
Com p ou n d 8: C41H49CoP2‚C5H12, fw ) 734.82, red plate,
0.27 × 0.27 × 0.09 mm3, monoclinic, space group P21/c (no.
14). Cell parameters: a ) 17.4679(4) Å, b ) 11.2656(3) Å, c )
26.2346(5) Å, â ) 125.8431(10)°, V ) 4184.93(17) Å3. Z ) 4, F
) 1.166 g cm-3, F000 ) 1576. 26 578 reflections were measured
up to a resolution of (sin θ/λ) ) 0.59 Å-1; 7314 reflections were
unique (Rint ) 0.0654). An absorption correction was not
considered necessary. The pentane solvent molecule was
refined with a disorder model and with isotropic displacement
parameters. 438 refined parameters, 43 restraints. R (obs
reflns): R1 ) 0.0443, wR2 ) 0.0984. R (all data): R1 ) 0.0775,
wR2 ) 0.1108. Weighting scheme w ) 1/[σ2(Fo2) + (0.0474P)2
3
) 2.1 Hz, C5H5), 121.0 (s, m-Mes*), 128.8 (d, J PC ) 9.4 Hz,
m-PPh3), 129.4 (d, 4J PC ) 2.1 Hz, p-PPh3), 134.8 (d, 2J PC ) 10.5
Hz, o-PPh3), 137.5 (d, 1J PC ) 40.1 Hz, i-PPh3), 144.4 (d, 2J PC
)
0.6 Hz, o-Mes*), 147.4 (s, p-Mes*), 171.3 (dd, 1J PP ) 105.7 Hz,
3J PP ) 19.4 Hz, i-Mes*). HRMS: calcd for C41H49P2Co 662.26416,
measd 662.26489.
Rea ction of P h osp h in id en e Com p lexes 3 a n d 8 w ith
CH2X2 a n d CHX3 on a n NMR Sca le. A small excess of CH2X2
(0.12 mmol) was added to a dark green solution of 3 (0.10
mmol) in n-pentane (0.5 mL), and the reaction was followed
by 31P NMR spectroscopy for product formation. An orange-
red precipitate of Cp*(PPh3)RhX2 formed quantitatively over
a period of 1 h (I), 12 h (Br), and 48 h (Cl) at room temperature
and was removed by decanting the solution and washing the
residue twice with n-pentane (1 mL). The combined n-pentane
solutions were evaporated to yield Mes*PdCH2 as a yellow
solid. Performing the reaction with CHI3 resulted in the
quantitative formation of orange-colored Mes*PdCHI as a
mixture of E (14%) and Z (86%) isomers. 31P, 1H, and 13C NMR
are as reported in the literature.39 The reactions were per-
formed in a similar fashion for 8.
2
+ 1.8625P], where P ) (Fo + 2Fc2)/3. GoF ) 1.011. Residual
electron density between -0.33 and 0.53 e/Å3.
Ack n ow led gm en t. This work was supported by The
Netherlands Foundation for Chemical Sciences (CW)
with financial aid from the Netherlands Organization
for Scientific Research (NWO). Dr. H. Zappey is ac-
knowledged for measuring high-resolution mass spectra.
Su p p or tin g In for m a tion Ava ila ble: Cartesian coordi-
nates of the optimized structures. Crystallographic data for
compounds 3 and 8. This material is available free of charge
via the Internet at http://pubs.acs.org.
X-r a y Cr ysta l Str u ctu r e Deter m in a tion s of 3 a n d 8.
Intensities were measured on a Nonius KappaCCD diffracto-
meter with a rotating anode (Mo KR, λ ) 0.71073 Å) at a
temperature of 150(2) K. The structures were solved with
OM0208624
(40) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J . Appl. Crystallogr. 1999, 32, 115.
(39) Appel, R.; Casser, C.; Immenkeppel, M.; Knoch, F. Angew.
Chem. 1984, 96, 905; Angew. Chem., Int. Ed. Engl. 1984, 23, 895. (b)
Appel, R.; Immenkeppel, M. Z. Anorg. Allg. Chem. 1987, 553, 7. (c)
Goede, S.; Bickelhaupt, F. Chem. Ber. 1991, 124, 2677.
(41) Sheldrick, G. M. SHELXL-97, Program for crystal structure
refinement; University of Go¨ttingen: Germany, 1997.
(42) Spek, A. L. PLATON, a multipurpose crystallographic tool;
Utrecht University: The Netherlands, 2001.