Ligand stereoelectronic effects in complexes of phospholanes, phosphinanes, and phosphepanes and their implications for hydroformylation catalysis

Article abstract of DOI:10.1021/om060912v

Convenient syntheses are described for the five-, six-, and seven-membered phosphacycles PhP(CH₂)_x-1, where x = 5 (L₅ ^a), 6 (L₆^a), 7 (L₇^a), and Bu^tP(CH

Full text of DOI:10.1021/om060912v

Organometallics 2007, 26, 713-725
713
Ligand Stereoelectronic Effects in Complexes of Phospholanes,
Phosphinanes, and Phosphepanes and Their Implications for
Hydroformylation Catalysis
R. Angharad Baber, Mairi F. Haddow, Ann J. Middleton, A. Guy Orpen, and
Paul G. Pringle*,†
School of Chemistry, UniVersity of Bristol, Cantocks Close, Bristol, U.K. BS8 1TS
Anthony Haynes* and Gary L. Williams
Department of Chemistry, UniVersity of Sheffield, Dainton Building, Brook Hill, Sheffield, U.K. S3 7HF
Rainer Papp
BASF Aktiengesellschaft, D-67056 Ludwigshafen, Germany
ReceiVed October 5, 2006
Convenient syntheses are described for the five-, six-, and seven-membered phosphacycles PhP(CH2)x-1
,
where x ) 5 (La ), 6 (La ), 7 (La ), and ButP(CH2)x-1 where x ) 5 (Lb ), 6 (Lb ), 7 (Lb ). Treatment of
5
6
7
5
6
7
[PtCl2(cod)] with La
gives cis-[PtCl2(La5-7)2] (1a5-7), whereas with Lb
a mixture of cis-[PtCl2-
5-7
5-7
(Lb5-7)2] (1b5-7) and trans-[PtCl2(Lb5-7)2] (2b5-7) is obtained. Metathesis of 1a7 with NaI gives a mixture
of cis-[PtI2(La ) ] (3a7) and trans-[PtI2(La ) ] (4a7). The crystal structures of 1a5, 1a6, 1a7, and 4a7 have
7 2
7 2
been determined. Comparison of the structures of 1a7 and 4a7 reveals that La has variable steric bulk,
7
with the crystallographically determined cone angle ranging from 137° (smaller than La ) to 172° (larger
5
than La ), depending on the particular twist-chair seven-membered-ring conformations adopted. The
6
complex cis-[PtCl2(Lb )] (1b6) is fluxional on the NMR time scale at ambient temperatures, as a result
6
of restricted PtP rotation. Treatment of [Rh2Cl2(CO)4] with La
or Lb
gives the expected trans-
5-7
5-7
[RhCl(CO)(La5-7)2] (5a5-7) or trans-[RhCl(CO)(Lb5-7)2] (5b5-7), and from the νCO values, it is deduced
that the donor strengths to rhodium(I) are in the order Lb > La and, within the La and Lb series,
5-7
5-7
L7, L6 > L5. An investigation into the kinetics of the oxidative addition of MeI to 5a5-7 showed that the
rate of reaction is in the order 5a5 > 5a7 > 5a6; i.e., the smallest ligand gives the highest rate. It is
postulated that the flexible La7 ligand adopts a lower bulk conformation and the order of decreasing rate
is then in the order of increasing bulk. The rate of reaction of MeI with 5b5 is 4 times faster than with
5a5, but oxidative addition was not observed with 5b6 or 5b7, perhaps because steric congestion destabilizes
the rhodium(III) product. A study of the rhodium-catalyzed hydroformylation of 1-octene is reported
using ligands La
and Lb5-7, but no systematic trends were observed, and the results for La
were
5-7
5-7
similar to those for the acyclic analogue PhPEt2. An anomalous but reproducible result is that the catalyst
derived from Lb shows negligible hydroformylation activity but rapid octene isomerization activity.
7
The overall conclusion is that, with the rhodium complexes of the simple phosphacycles described here,
no special effect of the rings was observed in the hydroformylation catalysis. An R-substituent effect is
identified as a common feature in several high-activity hydroformylation catalysts derived from
phosphinanes.
complexes, as exemplified in this article, in which a comparative
study of the ligand effects of two homologous series of cyclic
phosphines is described.
The chemistry of phosphorus heterocycles is much less
developed than the analogous nitrogen chemistry, and nowhere
is this more true than in the area of coordination chemistry,2
even though cyclic phosphines have been shown to be excellent
ligands for catalysis and particularly hydroformylation catalysis.3-6
Introduction
The correlation of stereoelectronic effects in tertiary phos-
phine ligands with the reactivity of their coordination complexes
is key to understanding their efficacy in homogeneous catalysis.
Tolman’s parameters (the cone angle, θ, and electronic param-
eter, νCO) remain the most widely used guides to the gross
features of stereoelectronic effects for monodentate phosphorus-
(III) ligands.1 However, subtle ligand structural effects can have
significant consequences for the macroscopic properties of the
(2) (a) Mathey, F. In Phosphorus-Carbon Heterocyclic Chemistry;
Pergamon Press: Oxford, U.K., 2001. (b) Doherty, R.; Haddow, M. F.;
Harrison, Z. A.; Orpen, A. G.; Pringle, P. G.; Turner, A.; Wingad, R. L.
Dalton Trans. 2006, 4310 and references therein.
(3) (a) Mason, R. F.; van Winkle, J. L. U.S. Patent 3 400 163, 1968 (to
Shell); (b) Steynberg J. P.; Govender, K.; Steynberg, P. J. World Patent
WO 14248, 2002 (to Sasol).
* To whom correspondence should be addressed.
† E-mail: paul.pringle@bristol.ac.uk. Fax: +44 (0)117 929 0509. Tel:
+44 (0)117 928 8114.
(1) (a) Tolman, C. A. Chem. ReV. 1977, 77, 313. (b) Bunten, K. A.;
Chen, L.; Fernandez, A. L.; Poe¨, A. J. Coord. Chem. ReV. 2002, 233-234,
41 and references therein.
10.1021/om060912v CCC: $37.00 © 2007 American Chemical Society
Publication on Web 12/29/2006

714 Organometallics, Vol. 26, No. 3, 2007
Baber et al.
Table 1. 31P{1H} NMR Data for the Complexes [PtCl2(L)2]a
In this article, the complexes of the two series of phosphacycles
La5-7 and Lb5-7 with platinum(II) and rhodium(I) are reported
ligand
complex
δ/ppm
J(PtP)/Hz
∆δ
La
1a5
1a6
1a7
1b5
2b5
1b6
2b6
1b7
2b7
3a7
4a7
14.7
-10.9
3.1
3524
3533
3559
3481
2386
3601
2433
3591
2407
3400
2307
30.0
23.7
29.2
33.6
38.4
23.2
29.2
29.2
29.8
28.1
22.8
5
La
6
La
7
Lb
35.7b
40.5b
10.6 b
16.6
5
Lb
5
Lb
6
Lb
6
Lb
22.3b
22.9b
2.0
7
Lb
7
La
7
La
-3.8
7
a Spectra measured in CDCl3 unless stated otherwise. b Spectrum mea-
sured in CD2Cl2.
phacycles La5-7 and tert-butylphosphacycles Lb5-7 by the route
shown in eq 1, which is an extension of Jolly’s synthesis14 of
and their spectroscopic and crystallographic properties are used
to probe their σ-/π-bonding characteristics. The rates of oxida-
tive addition of MeI to rhodium(I) complexes of La5-7 and Lb
5-7
have been measured, and the results are interpreted in terms of
ligand stereoelectronic effects. Finally, the hydroformylation of
1-octene catalyzed by rhodium complexes of La
has been investigated, in the hope that these studies would shed
light on the remarkably high hydroformylation activity of
rhodium complexes of Lc reported by BASF.4
and Lb
5-7
5-7
1-tert-butylphospholane (Lb ). An improvement to the procedure
5
was that the dilithiated alkanes were generated in situ by scaling
up Negishi’s method16 (eq 2). These one-pot syntheses are
Results and Discussion
Ligand Synthesis. The phosphacycles La were first syn-
5,6
thesized over 90 years ago7 by the reaction of the corresponding
BrMgCH2(CH2)nCH2MgBr (n ) 2, 3) with PhPCl2, and more
convenient, and pure products are readily obtained, but the yields
are at best modest.
recently8 all three La
compounds have been made by
5-7
intramolecular ring closure reactions of PhP(H)(CH2)nCHdCH2
Platinum(II) Complexes. Addition of 2 equiv of each of the
(n ) 2-4). The NMR spectroscopic properties of La
have
5-7
phosphacycles La
and Lb
to [PtCl2(cod)] afforded the
been extensively discussed,9 but their coordination chemistry
5-7
5-7
complexes [PtCl2(L)2] (1a5-7, 1b5-7, 2b5-7) as air-stable solids.
These complexes were characterized by a combination of 31P,
1H, and 13C NMR spectroscopy, mass spectrometry, elemental
analysis, and X-ray crystallography. The geometries of the
complexes were ascertained from the 1J(PtP) coupling constants
(see Table 1) and in some cases were confirmed by crystal
structure determinations (see below). The geometry depended
has been little studied.10,11,12 The tert-butylphosphacycles Lb
5,6
have been previously made by substitution reactions of But-
13
PCl2 or ButP(OMe)2,14 but the seven-membered 1-tert-
butylphosphepane (Lb ) is new. Rhodium complexes of Lb
7
5,6
have been shown to catalyze the production of ethylene glycol
from CO/H2 mixtures.15 We prepared all the phenylphos-
on the P substituent; the phenylphosphacycles La
gave cis
(4) Mackewitz, T.; Ahlers, W.; Zeller, E.; Ro¨per, M.; Paciello, R.; Knoll,
K.; Papp, R. World Patent WO 00669, 2002 (to BASF).
(5) Breit, B.; Fuchs, E. Chem. Commun. 2004, 694.
(6) Baber, R. A.; Clarke, M. L.; Heslop, K.; Marr, A. C.; Orpen, A. G.;
Pringle, P. G.; Ward, A.; Zambrano-Williams, D. E. Dalton Trans. 2005,
1079.
5-7
complexes 1a5-7, whereas the more bulky ligands Lb
gave
5-7
a mixture of cis complexes 1b5-7 and trans complexes 2b5-7
(eq 3). From the data given in Table 1, the five-membered
(7) (a) Gruttner, G.; Wiernik, M. Chem. Ber. 1915, 48, 1473. (b) Gruttner,
G.; Krause, E. Chem. Ber. 1916, 49, 437.
(8) Davies, J. H.; Downer, J. D.; Kirby, P. J. Chem. Soc. C 1966, 245.
(9) (a) Gray, G. A.; Cramer, S. E.; Marsi, K. L. J. Am. Chem. Soc. 1976,
98, 2109. (b) Chestnut, D. B.; Quin, L. D.; Wild, S. B. Heteroat. Chem.
1997, 8, 451.
(10) (a) Coles, S. J.; Edwards, P. G.; Hursthouse, M. B.; Abdul Malik,
K. M.; Thick, J. L.; Tooze, R. P. J. Chem. Soc., Dalton Trans. 1997, 1821.
(b) Bader, A.; Kang, Y. B.; Pabel, M.; Pathak, D. D.; Willis, A. C.; Wild,
S. B. Organometallics 1995, 14, 1434.
(11) Bodner, G. M.; May, M. P.; McKinney, L. E. Inorg. Chem. 1980,
19, 1951.
(12) Shima, T.; Bauer, E. B.; Hampel, F.; Gladysz, J. A. Dalton Trans.
2004, 1012.
(13) Featherman, S. I.; Lee, S. O.; Quin, L. D. J. Org. Chem. 1974, 39,
2899.
(14) Emrich, R.; Jolly, P. W. Synthesis 1993, 39.
phosphacycles have consistently the highest δP and ∆δ values
and the lowest J(PtP) values, which may be associated with
the hybridization required at P to accommodate the smaller
C-P-C angles (see further discussion below).
1
(15) (a) Yoshida, S.; Ogomori, Y; Watanabe, Y.; Honda, K.; Goto, M.;
Kurahashi, M. J. Chem. Soc., Dalton Trans. 1988, 895. (b) Ogomori, Y;
Yoshida, S.; Watanabe, Y. J. Mol. Catal. 1987, 43, 249.
(16) Negishi, E.; Swanson, D. R.; Rousset, C. J. J. Org. Chem. 1990,
55, 5406.

Phospholane, Phosphinane, and Phosphepane Complexes
Organometallics, Vol. 26, No. 3, 2007 715
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
cis-[PtCl2(La ) ] (1a5)
5 2
Pt1-P1
Pt1-Cl1
P1-C1
2.2397(9)
2.3686(10)
1.823(3)
P1-C10
P1-C7
1.837(3)
1.850(3)
C1-P1-C10
C1-P1-C7
108.87(16)
106.05(16)
C10-P1-C7
P1-Pt1-P1-C7
94.42(16)
148.67(14)
Table 3. Selected Bond Lengths (Å) and Angles (deg) for
cis-[PtCl2(La ) ] (1a6)
6 2
Pt1-P2
Pt1-P1
Pt1-Cl1
Pt1-Cl2
P1-C5
2.2479(9)
2.2541(9)
2.3566(9)
2.3595(9)
1.820(4)
P1-C1
P1-C6
P2-C12
P2-C17
P2-C16
1.824(4)
1.826(4)
1.818(4)
1.825(4)
1.827(4)
Figure 1. ORTEP plot of 1a5. All hydrogen atoms have been
omitted for clarity.
C5-P1-C1
C5-P1-C6
C1-P1-C6
97.69(18)
105.87(18)
104.90(17)
Cl2-Pt1-P1-C6
P1-Pt1-P2-C17
-65.17(13)
110.10(12)
Table 4. Selected Bond Lengths (Å) and Angles (deg) for
cis-[PtCl2(La ) ] (1a7)
7 2
Pt1-P1
Pt1-Cl1
P1-C11
2.2454(11)
2.3631(12)
1.825(4)
P1-C26
P1-C21
1.833(4)
1.837(5)
C11-P1-C26
C11-P1-C21
108.3(2)
99.2(2)
C26-P1-C21
P1A-Pt1-P1-C11
109.0(2)
-26.23(15)
between the C6H5 rings of 3.145 Å (C11-C11′).18 In order to
investigate the conformations of ligand La in the absence of
7
the π-π interactions present in complex 1a7, the diiodoplatinum
analogue was made, in anticipation that the bulky iodo ligands
would promote a trans geometry. In fact, treatment of 1a7 with
NaI gave a mixture of cis- and trans-[PtI2(La ) ] (3a7 and 4a7)
(see the Experimental Section and Table 1 for the data) but the
trans isomer 4a7 readily crystallized and its structure was
determined (see Figure 4 and Table 5).
7 2
Figure 2. ORTEP plot of 1a6. All hydrogen atoms have been
omitted for clarity.
A comparison of selected bond length and angles and ligand
cone angles for the platinum complexes [PtX2(La5-7)2] can be
made from the data collected in Table 6. Inspection of the
intracyclic C-P-C angles reveals that this angle is compres-
sed by reduction in the size of the phosphacycle, as expected.
There is a considerable difference between the intracyclic
C-P-C angles of 1a7 and 4a7, and this variation is presum-
ably a consequence of the flexibility of the seven-membered
ring.
Seven-membered rings adopt four low-energy conforma-
tions,19 and calculations on their relative energies have shown20
that the twist chair is the minimum energy form. Indeed, the
seven-membered rings in 1a7 and 4a7 both adopt twist-chair
conformations but the position of the phosphorus atom within
the twist chair gives the ligand quite different steric demands,
which can be seen more clearly by viewing the Pt-ligand
fragments in 1a7 and 4a7, as shown in Figure 5. The conforma-
tion adopted by La7 in 4a7 gives it much greater steric demands
Figure 3. ORTEP plot of 1a7. All hydrogen atoms have been
omitted for clarity.
Single crystals of 1a5-7 were grown by various methods (see
the Experimental Section) and their structures determined by
X-ray crystallography (see Figures 1-3 and Tables 2-4). The
phenyl substituents in 1a5,6 are anti with respect to each other,
and the six-membered rings in 1a6 adopt chair conformations
with the phenyl groups in equatorial positions.
than the La in 1a7 and this is reflected in the calculated cone
7
angles (Table 6). Notably, the cone angle for La in 4a7 is
7
significantly larger than that for La in 1a6 and the cone angle
6
for La in 1a7 is smaller than that for La in 1a5. A further
7
5
There is an intramolecular offset parallel π-π interaction17
between the phenyl substituents in 1a7 with a close distance
(18) Janiak, J. Dalton Trans. 2000, 3885.
(19) (a) Testa, B. In Principles of Organic Stereochemistry; Marcel
Dekker: New York, 1979. (b) Anet, F. A. L. In Conformational Analysis
of Medium-Sized Heterocycles; Glass, R. S., Ed.; VCH: New York, 1988.
(20) Favini, G. J. Mol. Struct. 1983, 93, 139.
(17) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112,
5525.

716 Organometallics, Vol. 26, No. 3, 2007
Baber et al.
Figure 5. Pt-ligand fragments of the crystal structures, showing
the seven-membered-ring conformations.
Figure 4. ORTEP plot of 4a7. All hydrogen atoms have been
omitted for clarity.
Table 5. Selected Bond Lengths (Å) and Angles (deg) for
trans-[PtI2(La ) ] (4a7)
7 2
Pt1-P1
Pt1-P2
Pt1-I2
Pt1-I1
P1-C6
2.3229(19)
2.3225(19)
2.6140(6)
2.6201(6)
1.816(8)
P1-C1
P1-C7
P2-C18
P2-C13
P2-C19
1.829(8)
1.818(9)
1.817(8)
1.832(8)
1.820(9)
Figure 6. Effect of orientation of the phenyl substituent on the
size of R, which contributes to the cone angle.
C6-P1-C1
C6-P1-C7
101.4(4)
105.2(4)
C1-P1-C7
100.4(4)
Table 6. Selected Bond Lengths (Å) and Angles (deg) for
Complexes [PtX2(La5-7)2] (X ) Cl, I)
1a5
1a6
1a7
4a7
Pt-P
Pt-X
2.239(1)
2.367(1)
2.253(1)
2.359(1)
2.246(1)
2.363(1)
2.3229(19)
2.6140(6)
C-P-C (cyclic)
P-Pt-P
X-Pt-X
94.6(2)
92.60(6)
88.30(7)
97.5(2)
103.11(4)
87.32(4)
109.0(2)
99.67(6)
88.29(7)
101.4(4)
176.54(7)
171.20(2)
Figure 7. ORTEP plot of 1b6. All hydrogen atoms have been
omitted for clarity.
cone angle
140
164, 161
137
160, 173
complication in assessing the bulk of the phenylphosphacycles
is that the value of the cone angle (given by eq 4) is highly
rH
2
3
180
θ )
R +
sin-1
(4)
∑
( )
π
d
R ) P-M-H angle, d ) M-H
dependent on the orientation of the phenyl ring; i.e., the closer
the torsion angle Pt-P-C-C is to 90°, the smaller the
contribution the phenyl group makes to the cone angle (as shown
by angle R in Figure 6). The orientation of the phenyl ring
affects the value of R by up to 30°, which thereby affects the
cone angle by up to 20°, all other atoms being consistent. For
complexes 1a5-7, 4a7, and 5a5 (see below), the torsions of the
phenyl rings are 132, 0/159, 131, 0/57, and 100°, respectively.
The 31P NMR spectra (see Table 1) of the products of the
reactions of [PtCl2(cod)] with the bulky ligands Lb5-7 in CH2-
Cl2 showed that both cis and trans isomers were formed, which
Figure 8. ORTEP plot of 2b6. All hydrogen atoms have been
omitted for clarity.
angles for the platinum complexes cis- and trans-[PtCl2(Lb ) ]
6 2
can be made from the data in Table 9. Both structures show
is consistent with the bulk of Lb
being greater than that of
Lb in a chair conformation with the tert-butyl substituents in
5-7
6
La5-7. When the reactions were carried out in toluene, the cis
isomers 1b5-7 precipitated from the reaction mixture, while the
trans isomers 2b remained in solution. There was no evidence
of cis/trans isomerization when either isomer was redissolved
in CD2Cl2 for periods of weeks.
Crystals of 1b6 and 2b6 were grown by slow diffusion of
Et2O into their saturated CH2Cl2 solutions and their crystal
structures determined (see Figures 7 and 8 and Tables 7 and
8). A comparison of selected bond lengths and angles and cone
pseudoequatorial positions, and there is little difference in the
intracyclic C-P-C angles or the calculated cone angles for
Lb .
6
The room-temperature 31P{1H} NMR spectrum of 1b6 showed
a broad singlet with platinum satellites. Cooling the solution to
-40 °C revealed the presence of two species in the ratio ca.
50:1, and the1J(PtP) values showed that, in both species, the
phosphines are cis to each other (Figure 9). The free energy
barrier to the process was estimated from the coalescence

Phospholane, Phosphinane, and Phosphepane Complexes
Organometallics, Vol. 26, No. 3, 2007 717
Figure 9. Variable-temperature 31P NMR of 1b6.
Table 7. Selected Bond Lengths (Å) and Angles (deg) for
P substituents are anti to each other and there is pseudo (though
not crystallographic) C2 symmetry; this is represented schemati-
cally in Figure 10a. Another possible rotamer (which is pre-
sumably more crowded and therefore less stable) is one where
the substituents are syn to each other and the structure has Cs
symmetry (depicted in Figure 10b). The fluxionality in solution
may be due to the interconversion of these C2 and Cs isomers.
cis-[PtCl2(Lb ) ] (1b6)
6 2
Pt1-P2
Pt1-P1
Pt1-Cl1
Pt1-Cl2
P1-C1
2.2656(15)
2.2754(15)
2.3573(15)
2.3629(15)
1.817(6)
P1-C5
P1-C6
P2-C14
P2-C10
P2-C15
1.828(6)
1.889(6)
1.824(6)
1.836(6)
1.897(7)
C1-P1-C5
C1-P1-C6
C5-P1-C6
C14-P2-C10
97.0(3)
106.5(3)
108.2(3)
98.0(3)
C14-P2-C15
105.9(3)
C10-P2-C15
106.0(3)
P2-Pt1-P1-C6
P1-Pt1-P2-C15
-114.4(2)
-109.8(2)
Table 8. Selected Bond Lengths (Å) and Angles (deg) for
cis-[PtCl2(Lb ) ] (2b6)
6 2
Pt1-Cl1
Pt1-P1
P1-C5
2.3161(6)
2.3185(5)
1.824(2)
P1-C1
P1-C6
1.824(2)
1.858(2)
C5-P1-C1
C5-P1-C6
99.46(11)
C1-P1-C6
106.98(10)
108.48(10)
Figure 10. Suggested rotomers present in solution with (a) C2
symmetry and (b) Cs symmetry.
Table 9. Selected Bond Lengths (Å) and Angles (deg) for
Complexes cis- and trans-[PtCl2(Lb ) ]
6 2
Rhodium(I) Complexes. Addition of 4 equiv of phos-
phacycles La5-7 and Lb5-7 to [Rh2Cl2(CO)4] in hexane afforded
the complexes trans-[RhCl(CO)(L)2] as yellow solids which
precipitated from solution (eq 5). These complexes were
1b6
2b6
Pt-P
2.2754(15)
2.3629(15)
2.3185(5)
2.3161(6)
Pt-Cl
C-P-C (cyclic)
P-Pt-P
Cl-Pt-Cl
97.0(3)
104.12(5)
84.77(6)
99.46(11)
180
180
cone angle
165, 158
163
temperature21 (Tc 283 K) to be 52 ((1) kJ mol-1. We tentatively
suggest that these two isomers are rotamers and the fluxionality
is due to restricted rotation about the Pt-P bonds. Barriers to
M-P rotation in complexes containing trans-M(PR3)2 moieties
have been measured22 to be in the range 36-75 kJ mol-1, but
to the best of our knowledge, barriers to rotation in complexes
containing a cis-M(PR3)2 moiety have not been previously
reported. The crystal structure of 1b6 shows that the exocyclic
(22) (a) Baber, R. A.; Orpen, A. G.; Pringle, P. G.; Wilkinson, M. J.;
Wingad, R. L. Dalton Trans. 2005, 659 and references therein. (b)
Bushweller, C. H.; Hoogasian, S.; English, A. D.; Miller, J. S.; Lourandos,
M. Z. Inorg. Chem. 1981, 20, 3448. (c) Dwyer, C. L.; Kirk, M. M.; Meyer,
W. H.; van Rensburg, W. J.; Forman, G. S. Organometallics 2006, 25,
3806.
(21) Sandstro¨m, J. In Dynamic NMR Spectroscopy; Academic Press:
London, 1982; p 96.

718 Organometallics, Vol. 26, No. 3, 2007
Baber et al.
characterized by a combination of 31P, 1H, and 13C NMR
spectroscopy, mass spectrometry, elemental analysis, IR spec-
troscopy, and X-ray crystallography. The 31P NMR spectra for
all the rhodium complexes showed a doublet, consistent with a
trans geometry, and these data, together with the CO stretching
frequencies for the complexes, are collected in Table 10. The
values of ν(CO) indicate23 that the tert-butylphosphacycles are
stronger σ-donors than their phenylphosphacycle analogues. In
addition, the data are consistent with the five-membered rings
being weaker σ-donors/stronger π-acceptors than the six- and
seven-membered analogues, a conclusion that is also supported
by the1J(PtP) values given in Table 1 and is in agreement with
the prediction from the Walsh diagram of a lower energy
HOMO and lower energy LUMO as the C-P-C angle is
compressed.24
Figure 11. ORTEP plot of 5a5. All hydrogen atoms have been
omitted for clarity.
Table 10. 31P NMRa and IR Datab for the Complexes
[RhCl(CO)(L)2]
ligand
complex
δ/ppm
J(RhP)/Hz
ν(CO)/cm-1
La
5a5
5a6
5a7
5b5
5b6
5b7
28.2
6.4
18.1
49.0c
27.8c
34.8c
118
117
118
117
118
118
1968
1960
1962
1955
1950
1951
5
La
6
La
7
Lb
5
Lb
6
Lb
7
a Spectra measured in CDCl3 unless otherwise stated. b Spectra measured
in CH2Cl2. c In CD2Cl2.
Single crystals of 5a5 and 5b5-7 were grown by various
methods (see the Experimental Section) and the crystal structures
determined (see Figures 11-14 and Tables 11-14).
A comparison of selected bond lengths and angles and cone
angles for the rhodium complexes [RhCl(CO)(L)2] can be made
from the data given in Table 15. The intracyclic C-P-C angles
show the expected trend, with the smaller rings having smaller
angles at phosphorus. The cone angles for the tert-butylphos-
phines are larger than those for the corresponding phenylphos-
phines of the same ring size. In this series of rhodium complexes
there is a monotonic trend with the larger rings having a larger
cone angle.
Figure 12. ORTEP plot of 5b5. All hydrogen atoms and the image
of the disordered CO and Cl ligands have been omitted for clarity.
With the exception of 4a7 and 5a5, the trans complexes of
Pt(II) and Rh(I) reported here have crystallographic inversion
symmetry (or pseudo inversion symmetry for the Rh(I) com-
plexes, where the chlorine and carbonyl ligands are disordered).
None of the complexes adopt Cs symmetry, although it would
certainly be plausible sterically and the C2 conformation adopted
by 5a5 is very close to C2V symmetry. Also, with the exception
of 5a5, all the complexes adopt a conformation such that the
torsion angles Cl-Pt-P-R (where R is the exocyclic substitu-
ent on the phosphine) are close to 90°. The torsion angles R-P-
P-R are close to (or exactly) 180°, except for 4a7, where R-P-
P-R is close to 0°.
Figure 13. ORTEP plot of 5b6. All hydrogen atoms and the image
of the disordered CO and Cl ligands have been omitted for clarity.
Oxidative Addition of Methyl Iodide to the Rhodium(I)
Complexes. The spectroscopic and crystallographic data for the
Table 11. Selected Bond Lengths (Å) and Angles (deg) for
trans-[RhCl(CO)(La ) ] (5a5)
5 2
platinum(II) and rhodium(I) complexes of La
and Lb
5-7
5-7
Rh1-C8
Rh1-P1
Rh1-Cl2
1.819(4)
2.3133(6)
2.3675(10)
P1-C2
P1-C12
P1-C9
1.824(2)
1.839(2)
1.852(2)
discussed above lead to the following generalizations about the
stereoelectronic properties of the phosphacyclic ligands.
(a) The tert-butylphosphacycles Lb
are stronger donors
5-7
C2-P1-C12
C2-P1-C9
104.44(11)
103.65(11)
C12-P1-C9
93.82(11)
than the analogous phenylphosphacycles La
.
5-7
(b) The five-membered phospholanes are weaker donors than
the analogous six-membered phosphinanes or seven-membered
phosphepanes (which are similar in their donor properties).
(c) The phospholanes are smaller than the phosphinanes, but
phosphepanes are flexible and so allow conformations that can
be larger than the chair conformation of the six-membered
phosphacycle or smaller than the envelope conformation of
phospholanes.
(23) Vastag, S.; Heil, B.; Marko´, L. J. Mol. Catal. 1979, 5, 189.
(24) (a) Orpen, A. G.; Connelly, N. G. Chem. Commun. 1985, 1310. (b)
Orpen, A. G.; Connelly, N. G. Organometallics 1990, 9, 1206.

Phospholane, Phosphinane, and Phosphepane Complexes
Organometallics, Vol. 26, No. 3, 2007 719
Table 12. Selected Bond Lengths (Å) and Angles (deg) for
trans-[RhCl(CO)(La ) ] (5b5)
5 2
Rh1-C1
Rh1-P1
Rh1-Cl1
1.734(8)
2.3311(6)
2.414(3)
P1-C16
P1-C5
P1-C9
1.850(3)
1.855(3)
1.861(3)
C16-P1-C5
C16-P1-C9
93.77(13)
106.86(13)
C5-P1-C9
107.54(12)
Table 13. Selected Bond Lengths (Å) and Angles (deg) for
trans-[RhCl(CO)(La ) ] (5b6)
5 2
Rh1-C1
Rh1-P1
Rh1-Cl1
1.729(5)
2.3333(5)
2.429(2)
P1-C2
P1-C6
P1-C7
1.8301(18)
1.8332(17)
1.8621(19)
C2-P1-C6
C2-P1-C7
99.08(8)
107.80(8)
C6-P1-C7
106.27(8)
Table 14. Selected Bond Lengths (Å) and Angles (deg) for
Figure 14. ORTEP plot of 5b7. All hydrogen atoms and the
image of the disordered CO and Cl ligands have been omitted for
clarity.
trans-[RhCl(CO)(La ) ] (5b7)
5 2
Rh1-C11
Rh1-P1
Rh1-Cl1
1.757(8)
2.3207(7)
2.388(3)
P1-C6
P1-C1
P1-C7
1.830(2)
1.837(2)
1.872(2)
Table 16. IRa and NMRb Spectroscopic Data for Products
[RhX2(Me)(CO)(L)2] (X ) Cl, I) Arising from Oxidative
Addition of MeI to 5a5-7 and 5b5-7
C6-P1-C1
C6-P1-C7
101.27(12)
104.39(12)
C1-P1-C7
103.90(11)
ν(CO)/
δ(1H)/ppm
L
cm-1
(Rh-CH3)c
δ(31P)/ppm (1J(RhP)/Hz)
Table 15. Selected Bond Lengths (Å) and Angles (deg) for
Complexes [RhCl(CO)(L)2]
La
2059
2050
2055
1.05, 0.72 (8:1)
33.4 (85), 28.5 (85), 19.3 (76),
27.9 (74)
14.5 (82), 17.6 (82),4.1 (72),
10.8 (71)
4.0 (84), 8.7 (83), -6.5 (74),
-4.4 (84)
57.6 (84), 62.7 (83)
5
6
7
5a5
5b5
5b6
5b7
La
La
0.59, 0.45, 0.80
(5.5:1:0.4)
0.75, 0.47 (vw),
0.83 (vw)
Rh-P
Rh-C
C-O
2.3133(5) 2.3311(6) 2.3335(6)
2.3207(7)
1.757(8)
1.167(11)
2.388(3)
1.816(3)
1.143(4)
1.734(8)
1.169(8)
1.721(7)
1.168(7)
2.428(3)
Rh-Cl
2.3673(9) 2.414(3)
Lb
Lb
Lb
2036
2030
2029
1.23, 0.95 (vw)
5
6
7
C-P-C (cyclic) 93.58(11) 93.77(13) 99.22(10)
101.27(12)
180
P-Rh-P
173.17(3) 180
180
C-Rh-Cl
180
177.2(2)
178.97(18) 177.5(4)
a Spectra measured in CH2Cl2. b Spectra measured in CDCl3. c All triplets
2
3
cone angle
124
146
161 172
of doublets with J(RhH) and J(PH) ca. 2 and 5 Hz, respectively.
The oxidative addition of MeI to the complexes trans-[RhCl-
(CO)(La5-7)2] and trans-[RhCl(CO)(Lb5-7)2] was studied in
detail to explore whether the reactivity trends could be
interpreted in terms of these ligand stereoelectronic effects.
Previous investigations25 have shown that the kinetics of MeI
oxidative addition to square-planar Rh(I) complexes (and the
tendency for migratory CO insertion to occur in the Rh(III)-
methyl product) are very sensitive to the nature of the coligands.
scrambling of halide ligands in the product to give a mixture
containing [RhCl2(Me)(CO)(L)2] and [RhI2(Me)(CO)(L)2] as
well as isomers of [RhCl(I)(Me)(CO)(L)2] with methyl trans to
Cl or I (eq 6).
The reactions of MeI with 5a5-7 and 5b5-7 were monitored
by IR spectroscopy under pseudo-first-order conditions (excess
MeI). Reactions were observed at relatively low [MeI] (0.08-
0.32 mol dm-3) for the three phenyl phosphacycle complexes,
5a5-7, the ν(CO) band of the Rh(I) reactant being replaced in
each case by an absorption at higher frequency (Table 16),
consistent with the oxidative addition product [RhCl(I)(Me)-
(CO)(L)2]. The products recovered after the kinetic runs were
also analyzed by NMR spectroscopy, and the data are given in
Table 16. The 1H NMR spectra showed triplets of doublets due
to coupling of the methyl protons to 103Rh and to two equivalent
31P nuclei. As well as the principal methyl resonance for each
product, some weaker triplets of doublets were also observed,
indicating the presence of additional Rh methyl complexes.
Similarly, the 31P NMR spectra exhibited up to four doublets
with Rh-P coupling. The NMR data can be explained by
Despite the mixtures of products indicated by NMR spec-
troscopy, plots of IR absorbance versus time for the Rh(I)
reactants were very well fitted by exponential decays, showing
that the reactions are first order with respect to the Rh(I)
complex. Pseudo-first-order rate constants (kobs) were measured
as a function of [MeI] and temperature and are reported in the
Supporting Information. Plots of kobs vs [MeI] were linear,
indicating a first-order dependence on the alkyl halide and
therefore a second-order reaction overall. Values of the second-
order rate constants obtained from the slopes of these plots are
given in Table 17. Eyring plots of the variable-temperature data
also showed good linearity and yielded the activation parameters
(25) (a) Gonsalvi, L.; Adams, H.; Sunley, G. J.; Ditzel, E.; Haynes, A.
J. Am. Chem. Soc. 1999, 121, 11233. (b) Gonsalvi, L.; Adams, H.; Sunley,
G. J.; Ditzel, E.; Haynes, A. J. Am. Chem. Soc. 2002, 124, 13597. (c)
Gonsalvi, L.; Gaunt, J. A.; Adams, H.; Castro, A.; Sunley, G. J.; Haynes,
A. Organometallics 2003, 22, 1047. (d) Martin, H. C.; James, N. H.; Aitken,
J.; Gaunt, J. A.; Adams, H.; Haynes, A. Organometallics 2003, 22, 4451.
listed in Table 17. These are similar in magnitude for 5a5-7
,
and the large negative ∆Sq values are typical of MeI oxidative

720 Organometallics, Vol. 26, No. 3, 2007
Baber et al.
Table 17. Second-Order Rate Constants (k1, 25 °C, CH2Cl2)
and Activation Parameters for Oxidative Addition of MeI to
5a5-7 and 5b5-7
halide exchange with MeI to give the iodide analogue, [RhI-
(CO)(Lb ) ]. This shift was accompanied by a decrease in ν-
7 2
(CO) band intensity for the MeI addition product, indicating
that the equilibrium constant for oxidative addition to [RhI-
reactant
103 k1/dm3 mol-1 s-1
∆Hq/kJ mol-1
∆Sq/J K-1 mol-1
(CO)(Lb ) ] is smaller than for the chloride, 5b7. Oxidative
5a5
5a6
5a7
5b5
5b6
5b7
11.6
2.83
8.59
42.8
45 ( 3
43 ( 1
41 ( 1
37 ( 2
-132 ( 9
-149 ( 2
-147 ( 2
-147 ( 7
7 2
addition of MeI was found to be even less favorable for 5b6. In
neat MeI, a weak IR band appeared at 2030 cm-1, consistent
with formation of a small amount of [RhCl(I)(Me)(CO)(Lb ) ].
6 2
small
0.015a
On the basis of the IR intensities, only about 4% of 5b6 is
converted to the product in neat MeI, suggesting that the
equilibrium constant for oxidative addition is an order of
magnitude smaller for 5b6 than for 5b7. Over extended time
periods, the ν(CO) band for 5b6 shifted ca. 2 cm-1 to higher
frequency, consistent with halide exchange, giving [RhI(CO)-
a Calculated from kinetics for approach to equilibrium and estimated
equilibrium constant.
addition proceeding via an SN2 mechanism.26 The observed
order of reactivity for the phenyl phosphacycle complexes is
5a5 > 5a7 > 5a6 (krel ca. 4:3:1), which appears to be determined
predominantly by ligand steric properties. Thus, the complex
of the smallest phosphacycle gives the highest rate, despite this
being the least electron rich (on the basis of ν(CO) values).
The higher reactivity of 5a7 compared to that of 5a6 can be
explained if the flexible phosphepane La7 adopts a conformation
(Lb ) ].
6 2
The dramatically lower reactivity of 5b6,7 compared with that
of 5b5 and 5a5-7 is an example of the operation of a steric
threshold.28 Severe steric congestion by ligands Lb and Lb
6
7
hinders the oxidative addition of methyl iodide and makes the
six-coordinate rhodium(III) products thermodynamically un-
stable. Interestingly, no evidence was found in these systems
for migratory CO insertion. In principle, conversion of an
octahedral methyl carbonyl complex into a square-pyramidal
acetyl complex can provide an alternative route by which steric
congestion can be relieved, as observed in a number of related
systems. Methyl migration to CO in the products arising from
MeI addition to 5b6,7 is probably inhibited by the relatively
electron rich metal center, which imparts considerable back-
donation to the CO ligand. In other Rh(III) systems where steric
promotion of methyl migration has been observed, ν(CO) values
were typically 30-50 cm-1 higher than for [RhCl(I)(Me)(CO)-
(Lb6,7)2].25 The instability conferred by steric congestion in the
latter is therefore relieved by reductive elimination of MeI rather
than by methyl migration.
of lower bulk than the phosphinane La , thereby accommodating
6
the crowding in the SN2 transition state.
A more dramatic effect of ligand steric bulk on reactivity
was found for the tert-butyl phosphacycle complexes, 5b5-7
.
The least hindered complex of this series, 5b5, reacted with MeI
in a manner similar to that described above for 5a5-7. NMR
spectroscopy again indicated a mixture of Rh(III) methyl
products resulting from halide scrambling in [RhCl(I)(Me)(CO)-
(Lb ) ] (Table 16). Kinetic measurements showed that 5b5 is
5 2
more reactive than all the phenyl phosphacycle complexes, its
second-order rate constant for MeI addition (Table 17) being
nearly 4 times larger than that for 5a5. This rate enhancement,
which arises largely from a lowering of ∆Hq, is attributed to
an electronic effect, the stronger donor Lb5 making the rhodium
complex more nucleophilic. It is notable that complexes 5a5-7
and 5b5 are more reactive toward MeI (by factors of 2-30)
than trans-[RhI(CO)(PEt3)2] (ν(CO) 1961 cm-1; k1 ) 1.37 ×
10-3 dm3 mol-1 s-1), studied by Cole-Hamilton and co-
workers.27
The main conclusions from the kinetics are as follows.
(a) The rates of oxidative addition of MeI to the complexes
trans-[RhCl(CO)(La5-7)2] (5a5-7) are in the order 5a5 > 5a7 >
5a6, which correlates better with the steric properties of the
ligands than with the electronic properties. That is, the complex
of the smallest phosphacycle gives the highest rate and it is
postulated that the flexible phosphepane La7 adopts a conforma-
In contrast to the relatively high reactivity of 5b5, the two
most sterically congested complexes, 5b6 and 5b7, did not give
stable oxidative addition products when treated with MeI. Only
at high MeI concentrations was evidence found for formation
of Rh(III) methyl products. When 5b7 was dissolved in neat
MeI, IR spectroscopy indicated initial attainment of an equi-
librium between the reactant and a species with ν(CO) at 2029
tion of lower bulk than the phosphinane La to accommodate
6
the crowding in the rhodium(III) product.
(b) The tert-butylphospholane complex trans-[RhCl(CO)-
(Lb ) ] (5b5) undergoes oxidative addition of MeI 4 times faster
5 2
than trans-[RhCl(CO)(La ) ], consistent with the stronger donor
5 2
Lb making the rhodium complex more nucleophilic.
cm-1, assigned as [RhCl(I)(Me)(CO)(Lb ) ]. At equilibrium
5
7 2
(c) In contrast to 5b5, only when neat MeI was used was
evidence obtained for very slow oxidative addition of MeI to
the complexes of the more bulky phosphacycles trans-[RhCl-
(CO)(Lb ) ] (5b6) and trans-[RhCl(CO)(Lb ) ] (5b7). It is likely
(attained after ca. 1 h at 25 °C) approximately 25% of 5b7 was
converted into product, corresponding to an equilibrium constant
of ca. 0.02 dm3 mol-1 for the oxidative addition reaction. The
approach to equilibrium followed first-order kinetics with a rate
constant of 1 × 10-3 s-1, which will contain contributions from
both forward and reverse rates (kobs ) k1[MeI] + k-1). This
leads to estimates of k1 ) 1.5 × 10-5 dm3 mol-1 s-1 and k-1
) 7.6 × 10-4 s-1. Thus, the oxidative addition rate constant
for 5b7 is ca. 3000 times smaller than that for 5b5. When it was
monitored for longer time periods, the ν(CO) band due to 5b7
shifted slightly (1-2 cm-1) to higher frequency, indicating
6 2
7 2
that this is a consequence of steric congestion destabilizing the
rhodium(III) product; once again the evidence suggests that the
seven-membered phosphacycle (Lb ) behaves as if it were
7
smaller than the six-membered analogue (Lb ).
6
Hydroformylation Catalysis. The ligands La
and Lb
5-7
5-7
were tested for the rhodium-catalyzed hydroformylation of
1-octene, and the results are shown in Table 18. For the phenyl
phosphacycles La5-7 (entries 1-3) and the noncyclic analogue
PPhEt2 (entry 4), essentially the same n-aldehyde selectivity is
obtained. In terms of activity La5 and La6 showed lower activity
(26) (a) Griffin, T. R.; Cook, D. B.; Haynes, A.; Pearson, J. M.; Monti,
D.; Morris, G. E. J. Am. Chem. Soc. 1996, 118, 3029. (b) Rendina, L. M.;
Puddephatt, R. J. Chem. ReV. 1997, 97, 1735. (c) Labinger, J. A.; Osborn,
J. A. Inorg. Chem. 1980, 19, 3230.
than La , which had activity similar to that of PPhEt2; it would
7
(27) (a) Rankin, J.; Poole, A. D.; Benyei, A. C.; Cole-Hamilton, D. J.
Chem. Commun. 1997, 1835. (b) Rankin, J.; Benyei, A. C.; Poole, A. D.;
Cole-Hamilton, D. J. J. Chem. Soc., Dalton Trans. 1999, 3771.
(28) Eriks, K.; Giering, W. P.; Liu, H. Y.; Prock, A. Inorg. Chem. 1989,
28, 1759.

Phospholane, Phosphinane, and Phosphepane Complexes
Organometallics, Vol. 26, No. 3, 2007 721
Table 18. Hydroformylation of 1-Octenea
recorded by The Mass Spectrometry Service, University of Bristol,
on MD800 and Autospec instruments. Infrared spectroscopy was
carried out on a Perkin-Elmer 1600 Series FTIR spectrometer. NMR
spectra were measured on a JEOL GX 300, JEOL Eclipse 400, or
JEOL GX 400 spectrometer. 31P{1H}, 13C{1H}, and 1H NMR
spectra were recorded at ambient temperature of the probe at 300,
100, and 121 MHz, respectively, using deuterated solvent to provide
the field/frequency lock.
all C8 olefins/
mol %
nonanals/
mol %
n-selectivity/
entry
ligand
%
1
2
3
4
5
6
7
8
La
0.9
0.6
1.8
1.8
2.7
3.7
82.8
1.2
26.3
19.5
88.0
88.0
88.0
88.0
4.7
66.5
67.7
67.0
67.0
62.0
54.0
4.3
5
La
6
La
7
PPhEt2
Lb
5
Lb
6
Synthesis of 1-Phenylphospholane (La ). To a solution of
Lb
5
7
t-BuLi (100 cm3, 1.7 M in pentane, 0.17 mol) in Et2O (50 cm3)
was added 1,4-diiodobutane (5.34 cm3, 12.544 g, 0.041 mol) over
5 min at -78 °C. The reaction mixture was stirred at -78 °C for
100 min, before being warmed to room temperature and added
dropwise to a solution of phenyldimethoxyphosphine (6.20 cm3,
0.036 mol) in hexane (150 cm3) at -10 °C. The resultant solution
was stirred at -15 °C for 5 h, washed with water (3 × 50 cm3),
and dried over MgSO4. The solution was filtered and the solvent
removed under reduced pressure to leave a yellow oil, which was
distilled under reduced pressure (ca. 0.1 mmHg, bp 57-59 °C) using
PPh3
89.7
67.8
a Products obtained after 4 h reaction in toluene/1-octene at 90 °C and
10 bar of H2/CO with [Rh] ) 5 µM, 10/1 ratio of L to Rh, and ca. 7000/1
ratio of 1-octene to Rh. See the Experimental Section for full details.
appear that constraining the P donor in a ring reduces the activity
of the catalyst for this series.
The results for the tert-butyl phosphacycles Lb
(entries
5-7
5-7) show that the tert-butyl group leads to more active
catalysts for the five- and six-membered phosphacycles (com-
pare entries 5 and 6 with entries 1 and 2). The anomalous result
with Lb7 (entry 7) showing negligible hydroformylation activity
but rapid olefin isomerization activity was reproducible with
different batches of ligands. This may be another manifestation
of a threshold effect in which apparently very small differences
in ligand structure produce a sharply different catalytic profile.
The hydroformylation study has not detected any systematic
trends in catalyst performance related to the structure of the
phosphacycles La5-7 and Lb5-7. This leads us to conclude that
the source of the extraordinarily high efficiency of the BASF
catalyst4 derived from Lc is more likely to be the position and
bulk of the substituents on the R-carbons rather than a
stereoelectronic effect associated with the C-P-C angle in the
phosphinane ring. The R-substituent effect could be a more
general phenomenon; for example, the high activity of the
rhodium hydroformylation catalysts derived from cyclic phos-
phines5,6 Ld and Le may also be a consequence of the ligands
having bulky R-substituents.
a 20 cm3 Vigreux column to give La as a colorless liquid (3.0
5
cm3, 3.50 g, 59% yield). 31P NMR (CDCl3): δP -15.3 (s).
Synthesis of 1-Phenylphosphinane (La ). To a solution of
6
t-BuLi (200 cm3, 1.7 M in pentane, 0.34 mol) in Et2O (100 cm3)
was added 1,5-diiodopentane (6.02 cm3, 26.22 g, 0.081 mol) over
30 min at -78 °C. The reaction mixture was stirred at -78 °C for
16 h, before being warmed to room temperature and added dropwise
to a solution of phenyldimethoxyphosphine (11.65 cm3, 0.072 mol)
in hexane (250 cm3) at -10 °C. The resultant solution was stirred
at -15 °C for 5 h, washed with water (3 × 100 cm3), and dried
over MgSO4. The solution was filtered and the solvent removed
under reduced pressure to leave a yellow oil, which was distilled
under reduced pressure (ca. 0.1 mmHg, bp 50-54 °C) using a 20
cm3 Vigreux column to give La as a colorless liquid (2.5 cm3,
6
2.58 g, 19% yield). 31P NMR (CDCl3): δP -34.3 (s).
Synthesis of 1-Phenylphosphepane (La ). To a solution of
7
t-BuLi (200 cm3, 1.7 M in pentane, 0.34 mol) in Et2O (100 cm3)
was added 1,6-diiodohexane (11.4 cm3, 24.8 g, 0.146 mol) over
10 min at -78 °C. The reaction mixture was stirred at -78 °C for
16 h, before being warmed to room temperature and added dropwise
to a solution of phenyldimethoxyphosphine (11.4 cm3, 12.4 g, 0.073
mol) in hexane (250 cm3) at -10 °C. The resultant solution was
stirred at -15 °C for 3 h, washed with water (3 × 100 cm3), and
dried over MgSO4. The solution was filtered and the solvent
removed under reduced pressure to leave a yellow oil, which was
distilled under reduced pressure (ca. 0.1 mmHg, bp 63-73 °C) using
a 20 cm3 Vigreux column to give La as a colorless liquid (0.8
7
cm3, 0.824 g, 6% yield). 31P NMR (CDCl3): δP -26.1 (s).
Synthesis of 1-tert-Butylphospholane (Lb ). To a solution of
5
Experimental Section
t-BuLi (183 cm3, 1.5 M in pentane, 0.275 mol) in Et2O (100 cm3)
was added 1,4-diiodobutane (8.74 cm3, 20.1 g, 0.065 mol) over 15
min at -78 °C. The reaction mixture was stirred at -78 °C for 4
h, before being warmed to room temperature overnight. To a
solution of trimethyl phosphite (6.9 cm3, 7.3 g, 0.059 mol) in hexane
(200 cm3) was added t-BuLi (39.2 cm3, 1.5 M in pentane, 0.059
mol) dropwise over 15 min at -78 °C, and then the mixture was
warmed to room temperature overnight. A solution of 1,4-
dilithiobutane was added dropwise to a solution of tert-bu-
tyldimethoxyphosphine at -10 °C over 1 h. The resultant solution
was stirred at -10 °C for 2 h and at room temperature for 4 days
and then washed with water (3 × 100 cm3) and dried over MgSO4.
The solution was filtered and the solvent removed under reduced
pressure to leave a yellow oil, which was distilled under reduced
pressure (53 mmHg, bp 76-80 °C) using a 10 cm3 Vigreux column
to give Lb5 as a colorless liquid (1.702 g, 20% yield). Anal. Found
(calcd): C, 66.48 (66.64); H, 12.41 (11.88). 31P NMR (CDCl3):
General Procedures. Unless otherwise stated, all reactions were
carried out under a dry nitrogen atmosphere using standard Schlenk
line techniques. Dry N2-saturated solvents were collected from a
Grubbs solvent system29 in flame- and vacuum-dried glassware.
MeOH was dried over 3 Å molecular sieves and deoxygenated by
N2 saturation. Commercial reagents were used as supplied unless
otherwise stated. All phosphines were stored under nitrogen at room
temperature. Most complexes were stable to air in the solid state
and were stored in air at room temperature. The starting materials
31
[PtCl2(cod)]30 and [RhCl(CO)2]2 were prepared by literature
methods. Elemental analyses were carried out by The Microana-
lytical Laboratory of the School of Chemistry, University of Bristol.
Electron impact and fast atom bombardment mass spectra were
(29) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.
(30) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem.
Soc. 1976, 98, 6521.
1
δP 2.1 (s). H NMR (CDCl3): δH 1.75-1.56 (m, 6H), 1.54-1.42
(31) McCleverty, J. A.; Wilkinson, G. Inorg. Synth. 1990, 28, 212.
(m, 2H), 0.96 (d, CH3, 9H, 3J(PH) ) 11.7 Hz). 13C NMR (CDCl3):

722 Organometallics, Vol. 26, No. 3, 2007
Baber et al.
2
2
δC 28.5 (d, CH2CH2P, J(PC) ) 1.5 Hz), 27.4 (d, CH3, J(PC) )
26.5 (s, CH2 J(PtC) ) 25 Hz). FAB mass spectrum: m/z 594 (M+),
559 (M+ - Cl).
1
13.8 Hz), 21.9 (d, CH2P, J(PC) ) 15.4 Hz), tertiary carbon not
Preparation of cis-[PtCl2(La ) ] (1a6). To a solution of La
seen.
6 2
6
Synthesis of 1-tert-Butylphosphinane (Lb ). To a solution of
(0.204 g, 1.14 mmol) in CH2Cl2 (2 cm3) was added [PtCl2(cod)]
(0.214 g, 0.572 mmol). After the reaction mixture was stirred for
12 h, the solvent was reduced to ca. 1 cm3 and Et2O (40 cm3) was
added. The resulting white precipitate was filtered off, washed with
Et2O, and dried under reduced pressure to afford the desired product
as a white powder (0.260 g, 0.41 mmol, 73% yield). X-ray-quality
crystals were grown by slow diffusion of Et2O into a saturated CH2-
Cl2 solution. Anal. Found (calcd): C, 42.49 (42.45); H, 4.88 (4.86).
6
t-BuLi (200 cm3, 1.7 M in pentane, 0.34 mol) in Et2O (100 cm3)
was added 1,5-diiodopentane (12.05 cm3, 26.2 g, 0.081 mol) over
15 min at -78 °C. The reaction mixture was stirred at -78 °C for
4 h, before being warmed to room temperature overnight. To a
solution of trimethyl phosphite (8.6 cm3, 9.0 g, 0.073 mol) in hexane
(200 cm3) was added t-BuLi (42.7 cm3, 1.7 M in pentane, 0.073
mol) dropwise over 15 min at -78 °C, and then the mixture was
warmed to room temperature overnight. A solution of 1,5-
dilithiopentane was added dropwise to a solution of tert-bu-
tyldimethoxyphosphine at 5 °C over 30 min. The resultant solution
was stirred at room temperature for 72 h and then washed with
water (3 × 100 cm3) and dried over MgSO4. The solution was
filtered and the solvent removed under reduced pressure to leave a
yellow oil, which was distilled under reduced pressure (20 mmHg,
1
31P NMR (CDCl3): δP -10.9 (s, J(PtP) ) 3533 Hz). H NMR
(CDCl3): δH 7.6-7.7 (m, 2H, Ar H), 7.4-7.5 (m, 3H, Ar H), 2.5-
2.7 (m, 2H), 1.8-2.0 (m, 2H), 1.6-1.8 (m, 2H), 1.2-1.6 (m, 4H).
13C NMR (CDCl3): δC 131.4 (t, Ar C, J(PC) ) 5 Hz), 130.9 (s,
Ar C), 129.8 (m, Ar C), 128.7 (t, Ar C, J(PC) ) 5 Hz), 26.1 (s,
CH2), 24.4 (m, CH2), 22.3 (s, CH2). FAB mass spectrum: m/z 622
(M+), 587 (M+ - Cl).
bp 62-66 °C) using a 10 cm3 Vigreux column to give Lb as a
Preparation of cis-[PtCl2(La ) ] (1a7). To a solution of La
6
7 2
7
colorless liquid (4 cm3, 3.9 g, 38% yield). Satisfactory elemental
analyses for this compound were not obtained despite clean NMR
spectra, presumably because of its extreme air sensitivity. Anal.
Found (calcd): C, 66.68 (68.32); H, 12.57 (12.10). 31P NMR
(CDCl3): δP -12.6 (s). 1H NMR (CDCl3): δH 2.05-1.90 (m, 2H),
1.72-1.63 (m, 1H), 1.62-1.43 (m, 4H), 1.32-1.00 (m, 3H), 0.97
(0.179 g, 0.93 mmol) in CH2Cl2 (2 cm3) was added [PtCl2(cod)]
(0.174 g, 0.47 mmol). After the reaction mixture was stirred for
12 h, the solvent was reduced to ca. 1 cm3 and Et2O (40 cm3) was
added. The resulting white precipitate was filtered off, washed with
Et2O, and dried under reduced pressure to afford the desired product
as a white powder (0.201 g, 0.31 mmol, 66% yield). X-ray-quality
crystals were grown by slow diffusion of Et2O into a saturated
dichloromethane solution. Anal. Found (calcd): C, 44.81 (44.77);
H, 5.39 (6.27). 31P NMR (CDCl3): δP 3.1 (s, J(PtP) ) 3559 Hz).
1H NMR (CDCl3): δH 7.4-7.5 (m, 3H, Ar H), 7.2-7.3 (m, 2H,
Ar H), 2.5-2.7 (m, 2H), 1.9-2.1 (m, 2H), 1.6-1.9 (m, 4H), 1.3-
1.6 (m, 4H). 13C NMR (CDCl3): δC 131.7 (t, Ar C, J(PC) ) 5
Hz), 131.1 (m, Ar C), 130.9 (s, Ar C), 128.6 (t, Ar C, J(PC) ) 5
Hz), 28.1 (s), 27.3 (m), 23.2 (s). FAB mass spectrum: m/z 650
(M+), 615 (M+ - Cl).
(d, 9H, CH3, J(PH) ) 11.5 Hz). 13C NMR (CDCl3): δC 28.2 (d,
3
2
CH2, J(PC) ) 3.1 Hz), 26.8 (d, CH3, J(PC) ) 13.1 Hz), 24.9 (d,
CH2, J(PC) ) 5.4 Hz), 20.8 (d, CH2P, 1J(PC) ) 14.6 Hz), tertiary
carbon not seen.
Synthesis of 1-tert-Butylphosphepane (Lb ). To a solution of
7
t-BuLi (200 cm3, 1.7 M in pentane, 0.34 mol) in Et2O (100 cm3)
was added 1,6-diiodohexane (13.53 cm3, 27.4 g, 0.081 mol) over
15 min at -78 °C. The reaction mixture was stirred at -78 °C for
4 h, before being warmed to room temperature overnight. To a
solution of trimethyl phosphite (8.6 cm3, 9.0 g, 0.073 mol) in hexane
(200 cm3) was added t-BuLi (42.7 cm3, 1.7 M in pentane, 0.073
mol) dropwise over 15 min at -78 °C, and then the mixture was
warmed to room temperature overnight. The solution of 1,6-
dilithiohexane was added dropwise to the solution of tert-bu-
tyldimethoxyphosphine at room temperature over 1 h. The resultant
solution was stirred at room temperature for 48 h and then at 50
°C for 48 h before being washed with water (3 × 100 cm3) and
then dried over MgSO4. The solution was filtered and the solvent
removed under reduced pressure to leave a yellow oil, which was
distilled under reduced pressure (20 mmHg, bp 82-88 °C) using
Preparation of cis- and trans-[PtCl2(Lb ) ] (1b5 and 2b5). To
5 2
a solution of Lb5 (0.132 g, 0.92 mmol) in toluene (5 cm3) was added
[PtCl2(cod)] (0.163 g, 0.43 mmol) as a solid. The resulting light
yellow solution was stirred for 4 days, after which time a white
solid had precipitated. The solid was filtered off, washed with
toluene (2 × 2 cm3), and dried under reduced pressure to give cis-
[PtCl2(Lb ) ] as a fine white powder (0.117 g, 0.21 mmol, 49%).
5 2
Anal. Found (calcd): C, 35.11 (34.76); H, 6.13 (6.18). 31P NMR
(CD2Cl2): δP 35.7 (s, J(PtP) ) 3481 Hz). 1H NMR (CD2Cl2): δH
3.15-2.70 (m, 1H), 2.15-1.20 (br m, 9H), 1.40 (d, 9H, CH3,3J(PH)
14.6 Hz). 13C NMR (CD2Cl2): δC 34.6-34.1 (m, C(CH3)3), 28.6-
28.2 (m, CH3), 27.2-26.5 (m), 26.9-26.1 (m). CI mass spectrum:
m/z 555 (M+), 520 (M+ - Cl), 484 (M+ - 2Cl). The solvent was
removed from the filtrate under reduced pressure to leave trans-
a 10 cm3 Vigreux column to give Lb as a colorless liquid (2.6
7
cm3, 2.6 g, 21% yield). Satisfactory elemental analyses for this
compound were not obtained despite clean NMR spectra, presum-
ably because of its extreme air sensitivity. Anal. Found (calcd):
C, 68.83 (69.73); H, 12.25 (12.29). 31P NMR (CDCl3): δP -6.9
(s). 1H NMR (CDCl3): δH 2.00-1.77 (m, 2H), 1.75-1.53 (m, 6H),
[PtCl2(Lb ) ] as a light yellow solid (0.064 g, 0.12 mmol, 27%
5 2
yield). Anal. Found (calcd): C, 34.90 (34.66); H, 6.28 (6.18). 31P
NMR (CDCl3): δP 40.5 (s, J(PtP) ) 2386 Hz). 1H NMR (CDCl3):
δH 2.56-2.37 (m, 2H), 1.93-1.65 (m, 6H), 1.23 (t, 9H, CH3, J(PH)
) 7.2 Hz). 13C NMR (CDCl3): δC 32.6-32.0 (br m), 28.0-27.4
(br m, CH3), 19.6-18.8 (br m). CI mass spectrum: m/z 554 (M+),
519 (M+ - Cl), 483 (M+ - 2Cl).
3
1.53-1.30 (m, 4H), 0.98 (d, CH3, 9H, CH3, J(PH) ) 11.2 Hz).
13C NMR (CDCl3): δC 27.9 (d, CH2, J(PC) ) 4.6 Hz), 27.1 (d,
CH3, 2J(PC) ) 13.1 Hz), 26.1 (d, CH2, J(PC) ) 10.0 Hz), 23.9 (d,
1
CH2P, J(PC) ) 16.9 Hz), tertiary carbon not seen.
Preparation of cis-[PtCl2(La ) ] (1a5). To a suspension of [PtCl2-
Preparation of cis- and trans-[PtCl2(Lb ) ] (1b6 and 2b6). To
5 2
6 2
(cod)] (0.130 g, 0.35 mmol) in toluene (3 cm3) was added a solution
of La5 (0.114 g, 0.69 mmol) in toluene (2 cm3). The resulting light
yellow solution was stirred for 12 h. The solvent was removed under
reduced pressure to leave a cream-colored solid, which was
recrystallized from CH2Cl2/Et2O to give a fine white powder (0.105
g, 0.18 mmol, 51% yield). X-ray-quality crystals were grown from
a saturated methanol solution. Anal. Found (calcd): C, 40.44
(40.42); H, 4.68 (4.41). 31P NMR (CDCl3): δP 14.7 (s, J(PtP) )
3524 Hz). 1H NMR (CDCl3): δH 7.3-7.8 (m, 5H, Ar H), 2.4-2.7
(m, 2H), 2.1-2.2 (m, 2H), 1.6-2.0 (m, 4H). 13C NMR (CDCl3):
δC 131.7 (m), 131.4 (m), 131.0 (s, Ar C), 128.6 (m), 28.1 (m, CH2),
a solution of Lb (0.139 g, 0.879 mmol) in toluene (5 cm3) was
6
added [PtCl2(cod)] (0.164 g, 0.439 mmol) as a solid. The resulting
light yellow solution was stirred for 4 days, after which time a
white solid had precipitated. The solid was filtered off, washed
with toluene (2 × 2 cm3), and dried under reduced pressure to give
cis-[PtCl2(Lb ) ] as a fine white powder (0.178 g, 0.306 mmol,
6 2
82%). X-ray-quality crystals were grown by slow diffusion of Et2O
into a saturated dichloromethane solution. Anal. Found (calcd): C,
37.02 (37.12); H, 6.58 (6.58). 31P NMR (CD2Cl2): δP 10.6 (s, J(PtP)
) 3601 Hz). 1H NMR (CD2Cl2): δH 2.05-1.50 (br m, 10H), 1.34
3
(d, 9H, CH3, J(PH) ) 14.6 Hz). 13C NMR (CD2Cl2): δC 34.5-

Phospholane, Phosphinane, and Phosphepane Complexes
Organometallics, Vol. 26, No. 3, 2007 723
33.8 (m), 30.8-30.2 (m), 26.9 (s), 24.9-23.1 (m). CI mass
spectrum: m/z 583 (M+), 548 (M+ - Cl), 511 (M+ - 2Cl). The
solvent was removed from the filtrate under reduced pressure to
Cl, 8.07 (8.77). 31P NMR (CDCl3): δP 28.2 (d, J(RhP) ) 118 Hz).
1H NMR (CDCl3): δH 7.5-7.6 (m, 2H, Ar H), 7.2-7.3 (m, 3H,
Ar H), 2.4-2.6 (m, 2H), 2.1-2.3 (m, 2H), 1.7-1.9 (m, 4H). 13C
NMR (CDCl3): δC 136.6 (d, Ar C, J(PC) ) 18 Hz), 131.1 (t, Ar
C, J(PC) ) 6 Hz), 129.3 (s, Ar C), 128.4 (t, Ar C, J(PC) ) 5 Hz),
27.5 (t, CH2, J(PC) ) 14 Hz), 27.0 (s, CH2), CO carbon not seen.
IR νCO (CH2Cl2): 1968 cm-1. FAB mass spectrum: m/z 494 (M+),
466 (M+ - CO), 459 (M+ - Cl), 431 (M+ - CO - Cl).
leave trans-[PtCl2(Lb ) ] as a light yellow solid (0.047 g, 0.081
6 2
mmol, 18% yield). X-ray-quality crystals were grown by slow
diffusion of Et2O into a saturated CH2Cl2 solution. 1H NMR showed
the presence of residual CH2Cl2 and hence elemental analysis was
as follows. Anal. Found (calcd for 2b6‚0.25CH2Cl2): C, 36.31
(36.30); H, 5.92 (6.38). 31P NMR (CDCl3): δP 16.6 (s, J(PtP) )
Preparation of [RhCl(CO)(La ) ] (5a6). To [RhCl(CO)2]2 (0.073
6 2
1
g, 0.19 mmol) was added a solution of La (0.137 g, 0.77 mmol)
2433 Hz). H NMR (CDCl3): δH 3.76-3.69 (m, 2H), 2.76-2.65
6
in hexane (4 cm3). The solution was stirred for 1 h, during which
time the red [RhCl(CO)2]2 was consumed and a fine yellow
precipitate formed. The solid was filtered in air, washed with hexane
(3 × 1 cm3), and dried under reduced pressure for 16 h to afford
the desired product as a fine light yellow powder (0.168 g, 0.32
mmol, 85% yield). Anal. Found (calcd): C, 53.05 (52.84); H, 5.80
(5.78). 31P NMR (CDCl3): δP 6.4 (d, J(RhP) ) 117 Hz). 1H NMR
(CDCl3): δH 7.6-7.7 (m, 2H, Ar H), 7.3-7.4 (m, 3H, Ar H), 2.4-
2.5 (m, 2H), 2.1-2.2 (m, 2H), 1.7-2.1 (m, 4H), 1.4-1.7 (m, 2H).
13C NMR (CDCl3): δC 187.2 (dt, CO, 1J(RhC) ) 75 Hz, 2J(PC) )
17 Hz), 134 (t, Ar C, J(PC) ) 20 Hz), 131.3 (t, Ar C, J(PC) ) 5
Hz), 129.3 (s, Ar C), 128.4 (t, Ar C, J(PC) ) 5 Hz), 27.0 (s, CH2),
24.3 (t, CH2, J(PC) ) 13 Hz), 22.8 (s, CH2). IR νCO (CH2Cl2):
1960 cm-1. FAB mass spectrum: m/z 522 (M+), 494 (M+ - CO),
487 (M+ - Cl), 179 (P+).
(m, 2H), 2.05-1.89 (m, 4H), 1.87-1.78 (m, 2.5H), 1.24 (t, 9H,
CH3, J(PH) ) 7 Hz). 13C NMR (CDCl3): δC 31.6-31.1 (m,
C(CH3)3), 27.6-27.2 (m, CH2), 27.2-26.9 (m, CH3), 23.5-23.1
(m, CH2), 15.6-15.0 (m, CH2). FAB mass spectrum: m/z 582 (M+),
510 (M+ - 2Cl).
Preparation of cis- and trans-[PtCl2(Lb ) ] (1b7 and 2b7). To
7 2
a solution of Lb7 (0.121 g, 0.70 mmol) in toluene (5 cm3) was added
[PtCl2(cod)] (0.125 g, 0.33 mmol) as a solid. The resulting light
yellow solution was stirred for 4 days, after which time a white
solid had precipitated. The solid was filtered off, washed with
toluene (2 × 2 cm3), and dried under reduced pressure to give cis-
[PtCl2(Lb ) ] as a fine white powder (0.140 g, 0.23 mmol, 69%
7 2
yield). Anal. Found (calcd): C, 39.29 (39.35); H, 7.05 (6.99). 31P
1
NMR (CD2Cl2): δP 22.3 (s, J(PtP) ) 3591 Hz). H NMR (CD2-
Cl2): δH 3.74-3.58 (m, 1H), 2.43-2.28 (m, 1H), 2.20-1.45 (m,
8H), 1.25 (d, 9H, CH3, 3J(PC) ) 14.6 Hz), 1.44-1.2 (m, 2H). 13
C
Preparation of [RhCl(CO)(La ) ] (5a7). To [RhCl(CO)2]2 (0.059
7 2
g, 0.16 mmol) was added a solution of La (0.124 g, 0.65 mmol)
NMR (CD2Cl2): δC 35.5-34.9 (m, C(CH3)3), 30.5 (s), 28.6-28.2
(m, CH3), 26.4-25.8 (m), 24.3-23.8 (m). EI mass spectrum: m/z
611 (M+), 538 (M+ - 2Cl). The solvent was removed from the
7
in hexane (4 cm3). The solution was stirred for 1 h, during which
time the red [RhCl(CO)2]2 was consumed and a fine yellow
precipitate formed. The solid was filtered in air, washed with hexane
(3 × 1 cm3), and dried under reduced pressure for 16 h to afford
the desired product as a fine light yellow powder (0.125 g, 0.23
mmol, 71% yield). Anal. Found (calcd): C, 54.76 (54.51); H, 5.88
(6.22). 31P NMR (CDCl3): δP 18.1 (d, J(RhP) ) 118 Hz). 1H NMR
(CDCl3): δH 7.9-7.8 (m, 2H, Ar H), 7.4-7.3 (m, 3H, Ar H), 2.7-
2.6 (m, 2H), 2.3-2.1 (m, 2H), 2.1-1.8 (m, 6H), 1.7-1.6 (m, 2H).
13C NMR (CDCl3): δC 187.9 (dt, CO, 1J(RhC) ) 74 Hz, 2J(PC) )
19 Hz), 136.2 (t, Ar C, J(PC) ) 20 Hz), 132.2 (t, Ar C, J(PC) )
6 Hz), 129.6 (s, Ar C), 128.3 (t, Ar C, J(PC) ) 5 Hz), 28.3 (d,
CH2, J(PC) ) 25 Hz), 28.3 (s, CH2), 24.4 (s, CH2). IR νCO (CH2-
Cl2): 1962 cm-1. FAB mass spectrum: m/z 193 (P+).
filtrate under reduced pressure to leave trans-[PtCl2(Lb ) ] as a light
7 2
yellow solid (0.034 g, 0.06 mmol, 17% yield). Anal. Found
(calcd): C, 39.20 (39.35); H, 7.14 (6.99). 31P NMR (CD2Cl2): δP
1
22.9 (s, J(PtP) ) 2407 Hz). H NMR (CD2Cl2): δH 2.63-2.53
(m, 1H), 2.20-1.05 (m, 11H), 1.24 (t, 9H, CH3, J(PH) ) 6.7 Hz).
13C NMR (CD2Cl2): δC 32.4 (t, C(CH3)3, J(PC) ) 15.4 Hz), 28.5
(s, CH2), 27.3 (t, CH3, J(PC) ) 1.9 Hz), 24.3 (m, CH2), 18.4 (t,
J(PC) ) 14.2 Hz). CI mass spectrum: m/z 610 (M+), 575 (M+
Cl), 538 (M+ - 2Cl).
-
Preparation of cis- and trans-[PtI2(Lb ) ] (3a7 and 4a7). To a
7 2
solution of 1a7 (0.050 g, 0.077 mmol) in CH2Cl2 (5 cm3) was added
NaI (0.230 g, 1.537 mmol) in acetone (5 cm3). The resulting yellow
solution was stirred for 30 min, and the solvent was removed under
reduced pressure. The residue was taken up in dichloromethane,
and this solution was washed with water (3 × 10 cm3) and dried
over MgSO4. The solution was filtered and the solvent removed to
Preparation of [RhCl(CO)(Lb ) ] (5b5). To a solution of Lb
5 2
5
(0.172 g, 1.19 mmol) in hexane (10 cm3) was added [RhCl(CO)2]2
(0.106 g, 0.27 mmol) as a solid. The solution was stirred for 12 h,
during which time the red [RhCl(CO)2]2 was consumed and a fine
yellow precipitate formed. The solid was filtered in air, washed
with hexane (3 × 1 cm3), and dried under reduced pressure for 16
h to afford the desired product as a fine light yellow powder (0.175
g, 0.39 mmol, 71% yield). X-ray-quality crystals were grown by
slow diffusion of hexane into a saturated CH2Cl2 solution. Anal.
Found (calcd): C, 44.55 (44.90); H, 7.82 (7.54). 31P NMR (CD2-
Cl2): δP 49.0 (d, J(RhP) ) 117 Hz). 1H NMR (CD2Cl2): δH 2.37-
2.26 (m, 2H), 1.95-1.71 (m, 6H), 1.20 (t, 9H, CH3, J(PH) ) 7.1
give a mixture of cis- and trans-[PtI2(Lb ) ] as a yellow powder
7 2
(0.061 g, 0.073 mmol, 95% yield). X-ray-quality crystals were
grown by slow diffusion of Et2O into a saturated CH2Cl2 solution.
31P NMR (CDCl3): δP 2.0 (s, J(PtP) ) 3400 Hz, 72%), -3.8 (s,
1
J(PtP) ) 2307 Hz, 28%). H NMR (CDCl3): δH 7.86-7.78 (m,
2H, Ar H), 7.56-7.48 (m, 2H, Ar H), 2.43-2.36 (m, 5H, Ar H),
7.22-7.28 (m, 1H, Ar H), 3.29-3.16 (m, 2H), 2.90-2.77 (m, 2H),
2.55-2.33 (m, 2H), 2.32-2.12 (m, 4H), 1.97-1.60 (m, 12H),
1.58-1.40 (m, 2H). 13C NMR (CDCl3): δC 132.1 (m, Ar C), 131.9
(m, Ar C), 130.6 (s, Ar C), 129.9 (s, Ar C), 128.5 (m, Ar C), 128.2
(m, Ar C), 31.6 (s), 31.5 (s), 31.2 (m), 30.4 (m), 29.7 (s), 29.0 (s),
27.7 (s), 27.6 (s), 24.7 (s), 23.9 (s). EI mass spectrum: m/z 833
(M+), 706 (M+ - I), 577 (M+ - 2I).
1
Hz). 13C NMR (CD2Cl2): δC 188.6 (dt, CO, J(RhC) ) 75 Hz,
2J(PC) ) 16 Hz), 32.3 (t, J(PC) ) 11 Hz), 28.1 (t, J(PC) ) 3.1
Hz), 28 (s) 23.2 (td, J(PC) ) 12 Hz, J(RhC) ) 2 Hz). IR νCO (CH2-
Cl2): 1955 cm-1. FAB mass spectrum: m/z 454 (M+), 426 (M+ -
CO).
Preparation of [RhCl(CO)(La ) ] (5a5). To a solution of [RhCl-
Preparation of [RhCl(CO)(Lb ) ] (5b6). To a solution of Lb
5 2
6 2
6
(CO)2]2 (0.099 g, 0.255 mmol) in toluene (5 cm3) was added a
(0.110 g, 0.70 mmol) in hexane (10 cm3) was added [RhCl(CO)2]2
(0.065 g, 0.17 mmol) as a solid. The solution was stirred for 12 h,
during which time the red [RhCl(CO)2]2 was consumed and a fine
yellow precipitate formed. The solid was filtered in air, washed
with hexane (3 × 1 cm3), and dried under reduced pressure for 16
h to afford the desired product as a fine light yellow powder (0.124
g, 0.26 mmol, 77% yield). X-ray-quality crystals were grown by
slow diffusion of hexane into a saturated CH2Cl2 solution. Anal.
solution of La (0.186 g, 1.023 mmol) in toluene (2 cm3). The
5
resulting solution was stirred for 30 min, after which time a sandy
brown precipitate had formed. The solvent was removed under
reduced pressure to leave the crude product as a brown oil.
Recrystallization from a 50/50 dichloromethane/hexane mixture
afforded the desired product as yellow crystals suitable for X-ray
diffraction. Anal. Found (calcd): C, 51.47 (51.08); H, 5.56 (5.30);

724 Organometallics, Vol. 26, No. 3, 2007
Baber et al.
Table 19. Crystallographic Data
1a5
1a6
1a7
1b6
2b6
color, habit
size/mm
empirical formula
Mr
colorless, block
0.05 × 0.05 × 0.05
C20H26Cl2P2Pt
594.34
colorless, flat block
0.10 × 0.05 × 0.01
C22H30Cl2P2Pt
622.39
colorless, needle
0.3 × 0.06 × 0.05
C24H34Cl2P2Pt
650.44
colorless, stalk
0.50 × 0.04 × 0.02
C18H38Cl2P2Pt
582.41
yellow, block
0.42 × 0.40 × 0.32
C18H38Cl2P2Pt
582.41
cryst syst
space group
a/Å
b/Å
c/Å
monoclinic
C2/c
16.511(4)
8.266(2)
16.101(4)
90
106.441(4)
90
2107.5(9)
4
orthorhombic
Pna21
17.157(2)
9.1406(10)
14.2928(13)
90
90
90
2241.5(4)
4
6.647
monoclinic
C2/c
20.282(5)
10.717(2)
11.804(3)
90
109.229(4)
90
2422.6(10)
4
orthorhombic
Fdd2
22.400(5)
27.229(5)
15.047(3)
90
90
90
9178(3)
16
6.487
monoclinic
P21/n
9.0392(6)
13.0671(8)
9.4358(6)
90
97.346(1)
90
1105.37(12)
2
R/deg
â/deg
γ/deg
V/Å3
Z
µ/mm-1
T
7.065
173
6.154
173
6.732
173
100
173
no. of rflns:
total/indep (Rint
final R1
10 703/2411
(0.0358)
0.0232
24 187/5124
(0.0436)
0.0203
6136/2769
(0.0409)
0.0329
14 536/5018
(0.0433)
0.0261
11 461/2531
(0.0277)
0.0148
)
largest peak, hole/e Å-3
1.282, -1.112
1.051, -0.467
1.819, -1.528
0.954, -0.706
0.519, -1.177
4a7
5a5
5b5
5b6
5b7
color, habit
size/mm
empirical formula
Mr
yellow plate
0.36 × 0.28 × 0.02
C24H34I2P2Pt
833.34
yellow prism
0.30 × 0.10 × 0.1
C21H26ClOP2Rh
494.72
yellow block
0.10 × 0.10 × 0.05
C17H34ClOP2Rh
454.74
yellow block
0.30 × 0.20 × 0.05
C19H38ClOP2Rh
482.79
yellow plate
0.20 × 0.20 × 0.05
C21H42ClOP2Rh
510.85
cryst syst
space group
a/Å
b/Å
c/Å
orthorhombic
Pbca
9.1251(10)
16.8748(18)
33.956(4)
90
orthorhombic
Pbcn
9.8155(11)
17.606(2)
12.2382(14)
90
90
90
2115.0(4)
4
1.093
triclinic
P1h
monoclinic
P21/n
9.0930(14)
13.206(2)
9.4945(14)
90
97.722(2)
90
1129.8(3)
2
monoclinic
P21/n
8.5201(17)
12.871(3)
11.488(2)
90
108.41(3)
90
1195.3(5)
2
7.6928(6)
8.0602(7)
8.9062(7)
89.406(1)
88.992(1)
68.205(1)
512.68(7)
1
R/deg
â/deg
90
90
γ/deg
V/Å3
5228.7(10)
8
7.861
Z
µ/mm-1
T
1.119
173
1.02
173
0.968
100
173
173
no. of rflns:
32 046/5991
(0.0622)
0.0446
12 764/2439
(0.0640)
0.0273
4940/2289
(0.0274)
0.0316
11 705/2582
(0.0304)
0.0226
3744/2314
(0.0190)
0.0264
total/indep (Rint
final R1
largest peak, hole/e Å-3
)
1.735, -2.872
0.394, -0.518
0.678, -1.039
0.422, -0.280
0.578, -0.518
Found (calcd): C, 47.70 (47.37); H, 8.21 (7.93). 31P NMR (CD2-
Cl2): δP 27.8 (d, J(RhP) ) 118 Hz). 1H NMR (CD2Cl2): δH 2.36-
2.26 (m, 2H), 2.06-1.90 (m, 4H), 1.84-1.74 (m, 1H), 1.63-1.50
(m, 2H), 1.28-1.15 (t, 9H, CH3, 3J(PH) ) 7.1 Hz), 1.28-1.15 (m,
1H). 13C NMR (CD2Cl2): δC 188.5 (dt, CO, J(RhC) ) 74.6 Hz,
J(PC) ) 16.0 Hz), 31.5 (t, J(PC) ) 12.3), 27.9 (s), 27.6 (s), 24.2
(s), 20.1 (t, J(PC) ) 11.1 Hz). IR νCO (CH2Cl2): 1950 cm-1. EI
mass spectrum: m/z 482 (M+), 454 (M+ - CO).
conditions were employed, with at least a 10-fold excess of MeI,
relative to the Rh complex. A solution containing the required
concentration of MeI in CH2Cl2 was prepared in a 5 cm3 graduated
flask. A portion of this solution was used to record a background
spectrum. Another portion (typically 500 µL) was added to the solid
Rh complex in a sample vial to give a reaction solution containing
typically 5-10 mM [Rh]. A portion of the reaction solution was
quickly transferred to the IR cell, and data collection was started.
The IR cell (0.5 mm path length, CaF2 windows) was maintained
at constant temperature by a thermostated jacket. Spectra (2200-
1600 cm-1) were scanned and saved at regular time intervals under
computer control. Absorbance vs time data for the appropriate ν-
(CO) frequencies were extracted and analyzed off-line using
Kaleidagraph curve-fitting software. For each experiment, the decay
of the reactant ν(CO) band was fitted to an exponential curve, with
correlation coefficient g0.999, to give the pseudo-first-order rate
constant. Each kinetic run was repeated at least twice to check
reproducibility, the kobs data given being averaged values with
component measurements deviating from each other by e5%.
Hydroformylation Catalysis. The phosphine (0.112 mmol) and
[Rh(CO)2(acac)] (3.0 mg, 0.012 mmol) were dissolved in toluene
(10 cm3) under nitrogen in a Schlenk tube. The resulting solution
was transferred by cannula to a 100 mL autoclave, which had been
flushed three times with 3 bar of CO/H2 (1/1). The autoclave was
then pressurized with 2 bar of CO/H2 (1/1) at room temperature.
The reaction mixture was then stirred vigorously with a sparging
stirrer and heated to 90 °C over a period of 30 min. After this
preactivation of the catalyst, 1-octene (10 g) was introduced into
Preparation of [RhCl(CO)(Lb ) ] (5b7). To a solution of Lb
7 2
7
(0.149 g, 0.87 mmol) in hexane (10 cm3) was added [RhCl(CO)2]2
(0.084 g, 0.22 mmol) as a solid. The solution was stirred for 12 h,
during which time the red [RhCl(CO)2]2 was consumed and a fine
yellow precipitate formed. The solid was filtered in air, washed
with hexane (3 × 1 cm3), and dried under reduced pressure for 16
h to afford the desired product as a fine light yellow powder (0.138
g, 0.27 mmol, 63% yield). X-ray-quality crystals were grown by
slow diffusion of hexane into a saturated CH2Cl2 solution. Anal.
Found (calcd): C, 49.16 (49.37); H, 8.24 (8.29). 31P NMR (CD2-
Cl2): δP 34.8 (d, J(RhP) ) 118 Hz). 1H NMR (CD2Cl2): δH 2.38-
2.29 (m, 2H), 2.06-1.64 (m, 8H), 1.58-1.46 (m, 2H), 1.22 (t, CH3,
9H, J(PC) ) 6.6 Hz). 13C NMR (CD2Cl2): δC 188.3 (dt, CO,
J(RhC) ) 74.6 Hz, J(PC) ) 16.1 Hz), 32.1 (t, C(CH3), J(PC) )
12.7 Hz), 28.3 (s), 27.8 (t, J(PC) ) 2.7 Hz), 25.0 (s, CH3), 22.5
(dt, CH2P, J(RhC) ) 1.5 Hz, J(PC) ) 10.8 Hz). IR νCO (CH2Cl2):
1951 cm-1. EI mass spectrum: m/z 510 (M+).
Kinetic Experiments. Reaction monitoring for kinetic experi-
ments was achieved using a Perkin-Elmer GX FTIR spectrometer
controlled by Spectrum TimeBase software. Pseudo-first-order

Phospholane, Phosphinane, and Phosphepane Complexes
Organometallics, Vol. 26, No. 3, 2007 725
the autoclave via a lock by means of CO/H2 pressure. The pressure
was then immediately raised to 10 bar and maintained at this
pressure throughout the catalysis by introduction of further CO/H2
(1/1) via a pressure regulator. After the reactions were run for 4 h,
the autoclave was cooled, vented, and emptied. Analysis of the
reaction products was carried out by GC.
been modeled as sitting over two positions, with occupancies of
0.33 and 0.67. In compounds 5b5-7, each complex contains a
crystallographic center of inversion, meaning that the carbonyl and
chloride ligands are disordered over two sites, each with occupancies
of 0.5. The chloride and carbonyl ligands are not disordered in 5a5.
X-ray Crystal Structure Analyses. X-ray diffraction experi-
ments on 1a5, 1a7, 1b6, 2b6, 4a7, 5a5, 5b5, and 5b6 were carried
out at 173 K on a Bruker SMART diffractometer, and experiments
on 1a6 and 5b7 were carried out at 100 K on a Bruker SMART
APEX diffractometer, both using Mo KR X-radiation (λ ) 0.710 73
Å) and a CCD area detector, from a single crystal coated in paraffin
oil mounted on a glass fiber. Intensities were integrated32 from
several series of exposures, each exposure covering 0.3° in ω.
Absorption corrections were based on equivalent reflections using
SADABS V2.10,33 and structures were refined against all Fo2 data
with hydrogen atoms riding in calculated positions using SHELX-
TL.34 Crystal and refinement data are given in Table 19. For 1b6,
the position of one of the tert-butyl groups is disordered and has
Acknowledgment. We thank Dr. Paul Elliott (University
of Sheffield) for assistance with the kinetic experiments, Qiang
Miao (BASF) for assistance with the Experimental Section,
BASF and EPSRC for financial support, The Leverhulme Trust
for a Fellowship (to P.G.P.), and Johnson-Matthey for the loan
of precious metal compounds.
Supporting Information Available: CIF files giving crystal-
lographic data for 1a5-7, 4a7, 1b6, 2b6, 5a5, and 5b5-7 and tables
and figures giving kinetic data, including plots of kobs against [MeI]
and Eyring plots. This material is available free of charge via the
Internet at http://pubs.acs.org.
OM060912V
(32) SAINT Integration Software; Siemens Analytical X-ray Instruments
Inc., Madison, WI, 1994.
(33) Sheldrick, G. M. SADABS V2.10; University of Go¨ttingen, Go¨t-
tingen, Germany, 2003.
(34) SHELXTL Program System, Version 5.1; Bruker Analytical X-ray
Instruments Inc., Madison, WI, 1998.