What is the real steric impact of triphenylphosphite? Solid-state and solution structural studies of cis- and trans-isomers of M(CO) 4[P(OPh)3]2 (M = Mo and W)

Article abstract of DOI:10.1021/om700761q

The steric requirements for the triphenylphosphite ligand in several molybdenum and tungsten carbonyl derivatives have been shown by X-ray crystallography to exceed the original Tolman's cone angle of 128°. That is, due to various accessible conformers possible for P(OPh)₃, solid-state data predict a considerably larger cone angle for the ligand of between 140° and 160°. Importantly, the solution behavior of cis-M(CO)₄[P(OPh)₃]₂ (M = Mo or W), coupled with similarly reported observations on a series of cis-Mo(CO) ₄[PR₃]₂ derivatives, support this conclusion, for these molecules both undergo thermal rearrangement to the more stable trans-isomers. On the other hand, the electronically similar but sterically much smaller cis-Mo(CO)₄[P(OCH₂)₃CEt]₂ complex is thermally stable under much harsher conditions. Furthermore, a comprehensive survey of structural data for transition-metal-triphenylphosphite derivatives available in the Cambridge Crystallographic Database reveals that most molecules display conformations that dictate cone angles much greater than that originally suggested by Tolman.

Full text of DOI:10.1021/om700761q

6832
Organometallics 2007, 26, 6832–6838
What is the Real Steric Impact of Triphenylphosphite? Solid-State
and Solution Structural Studies of cis- and trans-Isomers of
M(CO)4[P(OPh)3]2 (M ) Mo and W)
Donald J. Darensbourg,* Jeremy R. Andreatta, Sarah M. Stranahan, and
Joseph H. Reibenspies
Department of Chemistry, Texas A&M UniVersity, College Station, Texas 77843
ReceiVed July 27, 2007
The steric requirements for the triphenylphosphite ligand in several molybdenum and tungsten carbonyl
derivatives have been shown by X-ray crystallography to exceed the original Tolman’s cone angle of
128°. That is, due to various accessible conformers possible for P(OPh)3, solid-state data predict a
considerably larger cone angle for the ligand of between 140° and 160°. Importantly, the solution behavior
of cis-M(CO)4[P(OPh)3]2 (M ) Mo or W), coupled with similarly reported observations on a series of
cis-Mo(CO)4[PR3]2 derivatives, support this conclusion, for these molecules both undergo thermal
rearrangement to the more stable trans-isomers. On the other hand, the electronically similar but sterically
much smaller cis-Mo(CO)4[P(OCH2)3CEt]2 complex is thermally stable under much harsher conditions.
Furthermore, a comprehensive survey of structural data for transition-metal-triphenylphosphite derivatives
available in the Cambridge Crystallographic Database reveals that most molecules display conformations
that dictate cone angles much greater than that originally suggested by Tolman.
different conformations of the substituents are readily accessible.
Relevant to the ensuing discussion, Tolman realized early on
that phosphine ligands like PPh3 experienced a great deal of
steric strain when its cone angle was compressed beyond 145°,
while phosphite ligands like P(OPh)3 endured no major barrier
when its phenoxide substituents were folded upward to minimize
its spatial requirements.4 Parenthetically, using this argument
there should be no significant barrier to folding the phenoxide
substituents down in the phosphite ligand. Indeed, studies by
Ernst and co-workers have demonstrated that the P(OMe)3
ligand generally adopts configurations where all three methoxy
groups are not bent back away from the metal center.5,6 More
recently, Coville and co-workers have examined several metal
complexes containing (alkoxy)3P ligands and shown these to
exhibit conformation flexibility, with crystallographically de-
termined cone angles significantly greater than their minimized
values reported by Tolman.7,8 Furthermore, these larger steric
requirements for P(OR)3 ligands were substantiated in solution
via equilibrium studies of cis a trans isomers of Mo(η5-
C5H5)(CO)2(P(OR)3)I. Pertinent to our report herein, Yarger and
co-workers have established experimentally that P(OPh)3 at low
temperature exists in two configurations, C3 (all phenoxide
substituents bent toward the lone pair of electrons on the
phosphorus center) and Cs (two toward lone pair and one away).9
These conformations were supported by theoretical studies as
well.
Introduction
Tolman’s cone angles for quantifying the spatial requirements
of phosphine and phosphite ligands have played a prominent
role in organometallic chemistry.1 Indeed, it is not possible to
teach a course in modern inorganic/organometallic chemistry
without discussing this concept, which is well-covered in most
major textbooks.2 Tolman utilized space-filling CPK models to
define the angle of the cone formed having the metal atom as
its apex and encompassing the tertiary phosphine ligand
employing the surface of the van der Waals spheres of the
outermost hydrogen atoms of the substituents as the periphery.
Since Tolman’s landmark contribution, there have been numer-
ous alternative methods employed to refine these steric param-
eters for phosphine ligands, including molecular mechanics
models and X-ray structural data.3
When the substituents on the phosphorus atom were not
spherical, Tolman choose to fold them back to assume the
conformation requiring the least amount of space. From their
inception it has been recognized that the use of Tolman’s cone
angles as indicators of the steric influence of phosphine and
phosphite ligands has limitations and must be employed with
some discretion. This is particularly true when assessing the
steric impact of unsymmetric phosphines, or phosphines where
* Corresponding author. Fax: (979) 845-0158. E-mail: djdarens@
mail.chem.tamu.edu.
(1) Tolman, C. A. Chem. ReV. 1977, 77, 313–348.
(2) (a) See for example: Miessler, G. L.; Tarr, D. A. Inorganic Chemistry,
3rd ed.; Pearson, Prentice Hall, 2004; p 523. (b) Housecroft, C. E.; Sharpe,
A. G.; Inorganic Chemistry, 2nd ed.; Pearson, Prentice Hall, 2005; p 703.
(c) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals,4th ed.; Wiley-Interscience: New York, 2005; p 102.
(3) (a) White, D.; Coville, N. J. AdV. Organomet. Chem. 1994, 36, 95–
158. (b) Brown, T. L.; Lee, K. J. Coord. Chem. ReV. 1993, 128, 89–116.
(c) Alyea, E. C.; Dias, S. A.; Ferguson, G.; Restivo, R. J. Inorg. Chem.
1977, 16, 2329–2334. (d) Mingos, D. M. P.; Muller, T. E. J. Organomet.
Chem. 1995, 500, 251–259. (e) Guzei, I. A.; Wendt, M. Dalton Trans. 2006,
3991–3999.
(4) Tolman, C. A.; Seidel, W. C.; Gosser, L. W. J. Am. Chem. Soc.
1974, 96, 53–60.
(5) Stahl, L.; Ernst, R. D. J. Am. Chem. Soc. 1987, 109, 5673–5680.
(6) Stahl, L.; Trakarnpruk, W.; Freeman, J. W.; Arif, A. M.; Ernst, R. D.
Inorg. Chem. 1995, 34, 1810–1814.
(7) Smith, J. M.; Coville, N. J.; Cook, L. M.; Boeyens, J. C. A.
Organometallics 2000, 19, 5273–5280.
(8) Smith, J. M.; Coville, N. J. Organometallics 2001, 20, 1210–1215.
(9) Ghalsasi, P. S.; Yarger, J. 227th ACS National Meeting, Anaheim,
CA, Abstracts of Papers, 2004. Also see: http://yarger.asu.edu/Presentations/
Prasanna04.ppt (accessed Sept 28, 2007).
10.1021/om700761q CCC: $37.00
2007 American Chemical Society
Publication on Web 11/30/2007

What is the Real Steric Impact of Triphenylphosphite?
Organometallics, Vol. 26, No. 27, 2007 6833
Table 1. Crystal and Refinement Data for Complexes 1, 2, 4, 5, 6, and 7
1
2
4
5
6
7
empirical formula
fw
temp (K)
cryst system
space group
a (Å)
C40H30MoO10P2
828.52
110(2)
monoclinic
P2(1)
11.197(3)
16.033(3)
11.737(3)
90
118.030(4)
90
1860(1)
1.479
C54H52N2W2O14P2
1382.62
110(2)
orthorhombic
Pbca
18.3660(18)
14.5829(15)
19.565(2)
90
90
90
5240.0(9)
1.747
4
C40H30WO10P2•CHCl3
1035.80
110(2)
C16H22MoO10P2
532.22
110(2)
monoclinic
P2(1)/n
13.017(4)
19.538(6)
16.406(5)
90
99.544(5)
90
4115(2)
1.718
C40H30MoO10P2
828.52
110(2)
C23H15MoO8P
546.26
110(2)
triclinic
P1
9.505(4)
11.103(4)
11.161(4)
105.927(6)
94.806(6)
98.846(6)
1109.3(7)
1.635
triclinic
triclinic
j
j
P1
P1
9.170(1)
11.151(1)
11.375(1)
69.755(1)
89.766(1)
68.296(1)
1003.7(3)
1.714
8.483(7)
9.623(4)
12.110(5)
74.843(5)
75.001(2)
76.930(5)
1818(2)
1.515
b (Å)
c (Å)
R (deg)
ꢀ (deg)
γ (deg)
V (Å3)
Dc (Mg/m3)
Z
2
2
8
2
2
µ (mm-1
)
0.497
4.516
3.125
0.844
0.509
0.711
no. of reflns colltd
no. of indep reflns
no. of params
17 547
8191
478
47 056
5987
337
7889
3453
277
38 363
9318
527
8549
3966
241
7624
3464
299
goodness-of-fit
final R indices [I > 2σ(I)]
1.137
1.084
R1 ) 0.0504
Rw ) 0.1237
1.069
R1 ) 0.0200
Rw ) 0.0491
1.039
R1 ) 0.0772
Rw ) 0.1656
1.131
R1 ) 0.0475
Rw ) 0.1160
1.073
R1 ) 0.0488
Rw ) 0.1230
R1 ) 0.0457a
Rw ) 0.1120b
2
2 2
a R1 ) ∑Fo| - Fc|Fo. b Rw ) {[∑w(Fo - Fc ) ]/[∑w(Fo2)2]}1/2
.
What has perked our interest in this subject is our revisiting
earlier solution studies involving intramolecular and inter-
molecular cis–trans rearrangements in Mo(CO)4[PR3]2
derivatives.10–13 That is, we demonstrated that the isomer-
ization of cis-isomers of Mo(CO)4[PR3]2 to trans-isomers or
cis/trans mixtures thereof occurred by an intramolecular
process presumably involving a trigonal-prismatic intermedi-
ate when the tertiary phosphine was small, e.g., n-Bu3P or
Me3P. On the other hand, for the larger PPh3 ligand
cis-Mo(CO)4[PPh3]2 was observed to isomerization to the
trans-isomer via a dissociative, intermolecular process.
Specifically what has prompted this investigation of steric
effects in triphenylphosphite metal derivatives stems from
our observation that cis-Mo(CO)4[P(OPh3]2 was found to
undergo complete rearrangement to trans-Mo(CO)4[P(OPh)3]2
in solution via an intermolecular pathway. This is contrary
to our expectations based on our earlier observations since
Tolman’s cone angle for P(OPh)3 of 128° is less than that of
n-Bu3P (132°). Hence, we wish to report herein the X-ray
structures and solution behavior of triphenylphosphite com-
plexes of molybdenum and tungsten carbonyls. As has been
previously pointed out by others, the number of structurally
characterized complexes of triarylphosphite is extremely
small relative to their triarylphosphine analogues.14 For
example, in the Cambridge Crystallographic Database as of
May 2007 there were 10 687 structures with at least one
transition-metal-PPh3 bond and only 257 structures with one
transition-metal-P(OPh)3 bond.
60 °C over a 24 h period, which led to predominantly the cis-
isomer of W(CO)4[P(OPh)3]2.
(1)
(2)
Crystals of complexes 1, 2, and 4 suitable for X-ray
crystallographic studies were obtained from chloroform
layered with cold methanol. The structure of complex 1,
where data were collected at ambient temperature, has
previously been reported in the literature.16 We have repeated
it here to be consistent with other low-temperature structures
contained in this report. Crystallographic data pertaining to
all complexes structurally characterized herein are provided
in Table 1. A thermal ellipsoid view of complex 1 is depicted
in Figure 1, where the molybdenum center is shown to exhibit
a pseudo-octahedral coordination. The metric parameters
found in Table 2 determined in this study are only slightly
different from those previously reported by Alyea et al.16
That is, the average Mo-P distance is 2.447(10) Å, and
importantly, the P-Mo-P angle is 89.05(3)°. This angle is
of particular interest since the cis-Mo(CO)4[P(OPh)3]2 isomer
Results and Discussion
(10) Darensbourg, D. J. Inorg. Chem. 1979, 18, 14–17.
(11) Darensbourg, D. J.; Graves, A. H. Inorg. Chem. 1979, 18, 1257–
1261.
The ligand substitution reaction between cis-Mo(CO)4-
[NHC5H10]2 and triphenylphosphite in refluxing CH2Cl2 (∼40
°C) readily goes with complete replacement of the piperidine
ligands by P(OPh)3 (see eq 1).15 The resulting molybdenum
tetracarbonyl phosphite complex was 100% cis-Mo(CO)4-
[P(OPh)3]2 (1). On the other hand, the analogous tungsten
derivative under these reaction conditions afforded monosub-
stitution of piperidine by P(OPh)3 even over prolonged reaction
times of 24 h (eq 2). In order to displace the second piperidine
ligand, it was necessary to increase the reaction temperature to
(12) Cotton, F. A.; Darensbourg, D. J.; Klein, S.; Kolthammer, B. W. S.
Inorg. Chem. 1982, 21, 294–299.
(13) Cotton, F. A.; Darensbourg, D. J.; Klein, S.; Kolthammer, B. W. S.
Inorg. Chem. 1982, 21, 2661–2666.
(14) Baker, M. J.; Harrison, K. N.; Orpen, A. G.; Pringle, P. G.; Shaw,
G. J. Chem. Soc., Dalton Trans. 1992, 2607–2617.
(15) Darensbourg, D. J.; Kump, R. L. Inorg. Chem. 1978, 17, 2680–
2682.
(16) Alyea, E. C.; Ferguson, G.; Zwikker, M. Acta Crystallogr. 1994,
C50, 676–678.

6834 Organometallics, Vol. 26, No. 27, 2007
Darensbourg et al.
Figure 1. Thermal ellipsoid representation of complex 1 at 50% probability with atomic numbering scheme, and space-filling model illustrating
the intertwining of the two triphenylphosphite ligands.
Table 2. Selected Bond Distances (Å) and Bond Angles (deg) for
cis-Mo(CO)4[phosphite]2
phosphite
P(OPh)3
P(OCH2)3CC2H5
Mo(1)-C(1)
Mo(1)-C(3)
Mo(1)-C(2)
Mo(1)-C(4)
Mo(1)-P(1)
Mo(1)-P(2)
C-Oav
2.060(4)
2.025(4)
2.035(4)
2.034(4)
2.4318(10)
2.4360(10)
1.134(5)
2.044(5)
2.051(5)
2.032(5)
2.026(5)
2.4247(14)
2.4218(13)
1.137(6)
Figure 3. τ defines the angle that indicates whether the aryl
substituent is gauche (τ ∼60°) or anti (τ ∼180°) to the metal bonded
to the phosphite ligand.
P(1)-Mo(1)-P(2)
P(1)-Mo(1)-C(1)
P(1)-Mo(1)-C(3)
P(1)-Mo(1)-C(2)
P(2)-Mo(1)-C(1)
P(2)-Mo(1)-C(3)
P(2)-Mo(1)-C(4)
P(1)-Mo(1)-C(4)
P(2)-Mo(1)-C(2)
Mo(1)-P(1)-Oav
Mo(1)-P(2)-Oav
O-P(1)-O
89.05(3)
89.36(10)
87.83(11)
94.44(11)
95.90(10)
86.59(11)
88.62(11)
177.65(11)
174.01(11)
116.8(11)
117.9(11)
(96.97–103.61)
(97.30–104.05)
93.76(4)
87.10(13)
89.40(12)
87.97(13)
89.40(13)
85.93(12)
88.51(12)
174.93(12)
177.40(13)
116.5(12)
116.7(13)
(100.96–102.20)
(100.77–101.81)
O-P(2)-O
readily rearranges to the trans-isomer in solution (Vide infra).
Relevant to this issue, the orientation of the phenolate
substituents on the triphenylphosphite ligands in cis-
Mo(CO)4[P(OPh)3]2 are for P(1) one down and two up, and
for P(2) two down and one up. Figure 2 illustrates with stick
drawings the conformers of P(OPh)3, where up and down
are defined. A more precise description of the conformers
of P(OPh)3 entails determining τ, the angle that indicates
whether the aryl substituent is gauche (down) or anti (up) to
the metal bonded to the phosphorus atom of the phosphite
ligand (Figure 3).14 As is apparent in Figure 1, these
Figure 4. Thermal ellipsoid plot of 5 at 50% probability level.
conformers of P(OPh)3 and their spatial orientation readily
allow for a cis arrangement of these phosphite ligands about
the molybdenum octahedral center. This is best illustrated
in the space-filling model also illustrated in Figure 1.
We have also determined the solid-state structure of the ste-
rically nonencumbering phosphite complex cis-Mo(CO)4-
[P(OCH2)3CEt]2 (5). A thermal ellipsoid drawing of complex 5
is depicted in Figure 4. The structural parameters for this
derivative are also listed in Table 2 for an easy comparison with
those of complex 1. As listed in Table 2, although the
metal–ligand distances are quite similar the P-Mo-P angle is
somewhat more open at 93.76(4)°. In this instance only one
conformation of the phosphite ligand exists. The spatial require-
ments of P(OCH2)3CEt are quite small; that is, the calculated
cone angle from crystal structure data was determined to be
between 68° and 72° (Vide infra). Hence, the cis-isomer of
complex 5 is anticipated to be the thermodynamically more
stable isomer on the basis of both electronic (minimize trans
CO ligands) and steric arguments. Indeed, upon heating complex
5 in solution for an extended period of time at 80–95°, no
Figure 2. Conformers of P(OPh)3.

What is the Real Steric Impact of Triphenylphosphite?
Organometallics, Vol. 26, No. 27, 2007 6835
Figure 5. Thermal ellipsoid drawing of complex 2 at 50%
probability level.
Figure 7. Thermal ellipsoid drawing of complex 6 at 50%
probability level.
Figure 6. Ball-and-stick view of complex 2 shown down the trans-
Mo(CO)2 axis.
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) in
cis-W(CO)4[P(OPh)3][NHC5H10]
C(2)-O(2)
1.171(7)
1.157(6)
1.159(7)
1.155(6)
2.039(5)
1.947(6)
2.027(5)
1.995(5)
2.332(5)
2.4465(12)
88.23(10)
178.00(17)
92.58(18)
91.96(19)
94.16(18)
C(2)-W(1)-P(1)
C(4)-W(1)-P(1)
C(3)-W(1)-P(1)
C(1)-W(1)-P(1)
C(2)-W(1)-C(4)
C(2)-W(1)-C(3)
C(4)-W(1)-C(3)
C(2)-W(1)-C(1)
C(4)-W(1)-C(1)
C(3)-W(1)-C(1)
W(1)-P(1)-Oav
O-P(1)-Oav
91.88(14)
176.23(14)
85.90(14)
98.23(15)
87.2(2)
86.1(2)
90.4(2)
87.8(2)
85.4(2)
C(4)-O(4)
Figure 8. Thermal ellipsoid drawing of complex 4 at 50%
probability level.
C(1)-O(1)
C(3)-O(3)
W(1)-C(1)
bond distances and bond angles. There are several interesting
features in this complex worthy of note. As depicted in Figure
6, the phosphite and piperidine ligands are positioned to
minimize steric interaction concomitantly providing H-bonding
capability.17 Furthermore, the Mo-C(2) distance trans to
piperidine is shorter (1.947(6) Å) than the Mo-C(4) distance
trans to P(OPh)3 of 1.995(5) Å, and both are shorter than the
Mo-C distances trans to each other (2.033(5) Å).
W(1)-C(2)
W(1)-C(3)
W(1)-C(4)
W(1)-N(1)
W(1)-P(1)
172.7(2)
117.08(13)
100.7(2)
N(1)-W(1)-P(1)
C(2)-W(1)-N(1)
C(4)-W(1)-N(1)
C(3)-W(1)-N(1)
C(1)-W(1)-N(1)
The trans-M(CO)4[P(OPh)3]2 (M ) W (4) and Mo (6))
complexes were isolated from the thermal isomerization of the
initially afforded cis-isomers obtained via reactions 1 and 2.
Crystals suitable for X-ray analysis were obtained from a
chloroform solution of the respective complex layered with
methanol. Thermal ellipsoid representations of the two mol-
ecules are provided in Figures 7 and 8, with metric parameters
compiled in Table 4. As depicted in Figures 7 and 8, the
phenolate groups in the two phosphite ligands adopt a confor-
mation where two groups are down and one is up. At this point
it should be noted that the average steric requirements of the
P(OPh)3 ligand are greatly increased as the phenolate groups
point downward (Vide infra). A similar conformer of P(OPh)3
was found for the phosphite ligand in Mo(CO)5P(OPh)3 (7), as
indicated in Figure 9.
isomerization to the trans-isomer was observed (Vide infra).
Furthermore, it should be noted that the two different phosphite
ligands in complexes 1 and 5, respectively, exhibit quite similar
electronic properties, as revealed by infrared spectroscopy in
the νCO region; that is, νCO values for 1 are 2047, 1965, and
1947 cm-1 and for 5 are 2046, 1957, and 1937 cm-1
.
As illustrated in eq 2, while attempting to synthesize the cis-
W(CO)4[P(OPh)3]2 derivative, because of the greater kinetic
stability of the W-amine bond as compared to that in
molybdenum, an intermediate, cis-W(CO)4[P(OPh)3]NHC5H10
(2), was isolated. An X-ray structure of complex 2 was
determined, and a thermal ellipsoid drawing of this molecule
is shown in Figure 5. Table 3 contains a compilation of selected

6836 Organometallics, Vol. 26, No. 27, 2007
Darensbourg et al.
Table 4. Selected Bond Distances (Å) and Bond Angles (deg) for
Scheme 1
trans-M(CO)4[P(OPh)3]2, M ) Mo and W
Mo
W
C-Oav
1.143(4)
2.045(3)
2.040(3)
2.4063(7)
1.144(3)
2.036(3)
2.047(3)
2.4017(7)
M(1)-C(1)
M(1)-C(2)
M(1)-P(1)
P(1)-M(1)-C(1)
P(1)-M(1)-C(2)
M(1)-P(1)-Oav
O-P(1)-O
92.74(8)
86.35(8)
118.42(8)
95.79–103.8
87.86(8)
86.50(8)
118.23(7)
96.32–103.43
We have previously established a relationship between the
size of the phosphorus donor, as estimated by the Tolman cone
angle, and the relative thermodynamic stability of the cis/trans
isomers in Mo(CO)4[PR3]2 derivatives (eq 3).10,11 For example,
the Keq value for the reaction defined in eq 3 when R ) n-Bu
(Tolman cone angle ) 132°) was determined to be 5.3 at 60–80
°C. On the other hand, for the more compact PMe3 ligand
(Tolman cone angle ) 118°) the cis-isomer is slightly favored,
with ∆G for reaction 3 in this instance being slightly positive,
i.e., 0.32 kcal mol-1 at 65 °C. In both of these rearrangement
processes, the isomerization mechanism was definitively estab-
lished as being intramolecular. That is, ligand rearrangement
occurred via a trigonal-prismatic intermediate where no ligand
dissociation took place (Scheme 1a). By way of contrast, when
R ) Ph where the PPh3 ligand is sterically more demanding
(Tolman cone angle ) 145°), rearrangement to the highly
favored trans-isomer proceeds by way of phosphine dissociation
with isomerization of the five-coordinate intermediate followed
by phosphine recombination (Scheme 1b).
exist as a thermodynamically stable mixture of cis/trans-isomers.
Furthermore, isomerization of a kinetically isolated cis-
Mo(CO)4[P(OPh)3]2 complex would be expected to rearrange
to the trans-isomer or cis/trans mixture via an intramolecular
mechanism. However, herein quite the opposite situation was
observed for the behavior of cis-Mo(CO)4[P(OPh)3]2. Specifi-
cally, cis-Mo(CO)4[P(OPh)3]2 was found to readily isomerize
100% to the trans-isomer with P(OPh)3 dissociation, i.e., via
Scheme 1b.
The rearrangement of cis-Mo(CO)4[P(OPh)3]2 to trans-
Mo(CO)4[P(OPh)3]2 was monitored by infrared spectroscopy
in toluene at 85 °C over 48 h (eq 4). When this isomerization
process was carried out under an atmosphere of carbon
monoxide, Mo(CO)5P(OPh)3 (7) was afforded quantitatively,
indicative of a process proceeding by way of Scheme 1b.
Importantly, trans-Mo(CO)4[P(OPh)3]2 does not undergo phos-
phite dissociation under these reaction conditions. By way of
contrast, cis-W(CO)4[P(OPh)3]2 (4) was shown via this protocol
to undergo quantitative isomerization to its trans-isomer by an
intramolecular pathway. Hence, the added strength of the W-P
bonds in 4 allows for a trigonal twist rearrangement without
bond disruption. It should be noted that we have previously
reported a ∆Hq value of 31.9 kcal/mol for the dissociation of
P(O-o-tolyl)3 from cis-Mo(CO)4[P(O-o-tolyl)3]2, whereas the
similar process involving the tungsten analogue would be
expected to be on the order of 10 kcal/mol higher in energy.11
We have estimated the steric requirements of P(OPh)3 in the
cis-Mo(CO)4[PR3]2 h trans-Mo(CO)4[PR3]2
(3)
Based on electronic arguments, in the absence of steric
demands made by phosphine or phosphite ligands,
Mo(CO)4[PR3]2 or Mo(CO)4[P(OR)3]2 derivatives would be
anticipated to exhibit cis stereochemistry. As indicated earlier,
this was indeed found to be the case for the sterically compact
P(OCH2)3CEt derivative of molybdenum tetracarbonyl. On the
other hand for the electronically quite similar P(OPh)3 derivative
with a reported Tolman angle of 128°, that is, 4° smaller than
n-Bu3P, our initial thoughts were that Mo(CO)4[P(OPh)3]2 would
(4)
various group 6 metal–carbonyls discussed herein by calculating
the aVerage and maximum cone angles from crystallographic
parameters employing the method of Müller and Mingos.18
Although Tolman’s original cone angles obtained from CPK
models of the most compact conformer of phosphine and
phosphite ligands have served us well when discussing ligand
Figure 9. Thermal ellipsoid view of complex 7 at 50% probability
level. Selected bond distances (Å) and bond angles (deg):
Mo-Ceq(av) ) 2.061(9), Mo-Cax ) 2.066(9), Mo-P ) 2.448(2),
P-Mo-Ceq(av) ) 90.1(2).
(17) (a) Atwood, J. L.; Darensbourg, D. J. Inorg. Chem. 1977, 16, 2314–
2317. (b) Darensbourg, D. J. Inorg. Chem. 1979, 18, 2821–2825.
(18) Müller, T. W.; Mingos, D. M. P. Transition Met. Chem. 1995, 20,
533–539.

What is the Real Steric Impact of Triphenylphosphite?
Organometallics, Vol. 26, No. 27, 2007 6837
Table 5. Calculated Cone Angles of Triphenylphosphite in Group 6
Metal–Carbonyl Derivativesa
in the solid-state cis structure. Furthermore, rearrangement to
the trans structures of both molybdenum and tungsten tetra-
carbonyl (complexes 4 and 6) allows for both phosphites to
possess what is evidently their favored conformation, two phenyl
substituents down. This conformer is also found in the sterically
unencumbered monosubstituted Mo(CO)5P(OPh)3 complex (7)
and as was previously observed in the unbound P(OPh)3
molecule. It is worthy of note that in the sterically more bulky
tris(2-methoxyphenyl)phosphite molecule the C3 conformer was
observed crystallographically, where all three phenyl substituents
are oriented downward as defined herein.14
cone angle (deg)
phenyl orientations
complex
cis-Mo(CO)4[P(OPh)3]2 (1) two down (160, 55, 44)
one down (165, 125, 48) 125
(τ values, deg)
average maximum
140
162
152
154
166
162
trans-Mo(CO)4[P(OPh)3]2 (6) two down (179, 52, 40)
trans-W(CO)4[P(OPh)3]2 (4) two down (168, 53, 37)
139
143
143
Mo(CO)5P(OPh)3 (7)
two down (160, 55, 44)
a θi values calculated from crystallographic data using the method of
Müller and Mingos, where θi ) R + 180/π × sin-1(rH/d).18 The angle R
is defined as the angle between the phosphorus and metal centers and
the center of the outermost hydrogen atom on the phenyl substituent.
The distance d is the crystallographic distance between the metal center
and the outmost hydrogen atom, and rH is the van der Waals radius of
hydrogen.
Conclusions
Although Tolman’s cone angles for tertiary phosphines and
phosphites have served the organometallic community well,
there are instances where their absolute values must be utilized
with some degree of skepticism. This is particularly true when
various conformers of the ligands are readily accessible. Herein,
we have shown that triphenylphosphite displays four different
conformers in the solid state, two of which are dominant,
accounting for 96% of the structures currently contained in the
Cambridge Crystallographic Database. Importantly, these latter
conformers in group 6 carbonyl derivatives are spatially much
larger than the original Tolman cone angle for P(OPh)3 of 128°.
Significantly, the steric requirements of P(OPh)3 found in
these solid-state structures persist in solution as demonstrated
by cis–trans equilibria and isomerization mechanisms. That is,
phosphite ligands with cone angles ∼128° would be expected
to thermodynamically favor cis-M(CO)4[P(OPh)3]2 (M ) Mo
or W) structures and their isomerization pathways to the
analogous trans-isomers would be anticipated to be intramo-
spatial requirements, these values fail to take into account
variations in cone angles with ligand conformation. As depicted
in Figure 2, triphenylphosphite can exist as several conformers.
In the crystallographically defined metal-carbonyls in this study
two of the commonly observed conformers of P(OPh)3 were
observed, i.e., those of Cs symmetry with either two phenyl
groups oriented up (anti) or down (gauche). The latter conformer
is found in the low-temperature structure of triphenylphosphite
derived from a single crystal study at 119 K.19 Furthermore,
no phase changes as a function of temperature were noted from
powder X-ray diffraction measurements of P(OPh)3 over the
temperature range 150–290 K. Indeed, of the 362 triphenylphos-
phite ligands bound to transition metal centers in the Cambridge
Crystallographic Database as of May 2007, only 1.7% (6) of
these displayed the Tolman’s all (up, anti) arrangement,20
whereas, 33.7% (122) and 62.4% (226) were two (up, anti) and
one (down, gauche) or two (down, gauche) and one (up, anti),
respectively.21 The remaining 2.2% (8) of the TM-P(OPh)3
structures exhibit all three aryl substituents down (gauche).22
Table 5 contains the calculated phosphite cone angles for the
seven P(OPh)3 ligands found in complexes 1, 4, 6, and 7, along
with the three τ angles as described in Figure 3. The Müller
and Mingos18 algorithm for calculating cone angles uses the
center of the outermost hydrogen atom in the substituent as
defined by crystallographic data. The aVerage cone angle is
defined as 2/3 ∑iθi, where θi is the half-angle to the van der
Waals radius of the outermost hydrogen atom of the substituent
i, whereas, our maximum cone angle is ascribed to twice the
largest half-angle, θi (see Supporting Information). For the
conformer where two of the phenyls are oriented downward,
the aVerage and maximum cone angles were found to have
average values of 141 ( 2° and 161 ( 5°, respectively. On the
other hand, the P(OPh)3 ligand with two phenyl substituents
oriented upward was found to have an aVerage cone angle of
125° and a maximum cone angle of 152°. Hence, the cis-
Mo(CO)4[P(OPh)3]2 structure contains one phosphite ligand with
two phenyl substituents down and the other phosphite ligand
with two phenyl substituents up, thereby allowing for a meshing
of the two ligands minimizing steric interactions. That is, the
one up and two up areas are in closest proximity to one another
lecular.
However,
cis-Mo(CO)4[P(OPh)3]2
and
cis-
W(CO)4[P(OPh)3]2 isomers undergo essentially complete ther-
mal rearrangement to the trans-isomers, with the former
complex proceeding via an intermolecular process. Current
studies are aimed at more extensively exploring the steric effects
of various triarylphosphite and triarylarsenite ligands in met-
al–carbonyl derivatives employing the X-ray crystallographic
and isomerization reaction probes described herein.
Experimental Section
Methods and Materials. Unless otherwise stated all synthesis
and manipulations were carried out on a double-manifold Schlenk
vacuum line under an argon atmosphere or in an argon-filled
glovebox. Molybdenum hexacarbonyl, tungsten hexacarbonyl, and
triphenylphosphite were purchased from ACROS. Trimethylolpro-
panephosphite was obtained from TCI. All solvents were freshly
purified using an MBraun solvent purification system packed with
Alcoa F200 activated alumina desiccant. Infrared spectra were
recorded on a Mattson 6021 Galaxy series FTIR, and 31P NMR
spectra were collected on an Inova 300 MHz spectrometer. cis-
M(CO)4[NHC5H10]2 (M ) Mo and W) were prepared in 90% yield
by the method previously described in the literature.15
Compound Preparations. (a) cis-Mo(CO)4L2 (L ) P(OPh)3
(1) and P(OCH2)3CEt (5). Derivatives were prepared from cis-
Mo(CO)4[NHC5H10]2 and excess ligand (L) in refluxing dichlo-
romethane as previously reported.15 White crystals of complex 1
suitable for X-ray analysis were obtained from chloroform and cold
methanol in 66.3% yield. The νCO values of 1 in hexane were
observed at 2047, 1965, and 1947 cm-1, with the 31P NMR
resonance found at 152.1 ppm. White crystals of complex 5 were
obtained from dichloromethane/methanol in 57.2% yield. The νCO
frequencies of 5 in toluene were observed at 2046, 1957, and 1937
cm-1, with the 31P NMR signal found at 139.7 ppm.
(19) Senker, J.; Lüdecke, J. Z. Naturforsch. 2001, 56b, 1089–1099.
(20) For example: Guss, J. M.; Mason, R. J. Chem. Soc., Dalton Trans.
1972, 2193–2196.
(21) The three calculated values of τ for each triphenylphosphite ligand
bound to a transition metal structure found in the Cambridge Crystal-
lographic Database as of May 2007 are provided in the Supporting
Information.
(22) For examples, see: (a) Rasmussen, P. G.; Anderson, J. E.; Bayón,
J. C Inorg. Chim. Acta 1984, 87, 159–164. (b) Haumann, M.; Meijboom,
R.; Moss, J. R.; Roodt, A. Dalton Trans. 2004, 1679–1686.

6838 Organometallics, Vol. 26, No. 27, 2007
Darensbourg et al.
(b) cis-W(CO)4[NHC5H10]P(OPh)3 (2). To 2.57 g of cis-
W(CO)4[NHC5H10]2 suspended in 50 mL of dichloromethane was
added 1.47 mL (1 equiv) of P(OPh)3, and the solution was refluxed
overnight. The resultant reaction solution was filtered through Celite,
and the solvent was removed under vacuum. Small yellow crystals
suitable for X-ray analysis were isolated upon recrystallization of
the product from chloroform and cold methanol. The purified
product was obtained in a 44.2% yield, or 1.68 g. The νCO infrared
(c) Complex 5, cis-Mo(CO)4[P(OCH2)3CEt]2, was heated in
toluene solution for 48 h at 80 °C, followed by further heating at
95 °C for 24 h with no change occurring in its infrared νCO bands.
X-ray Crystallography Data Collection and Refinement. X-ray
data for compounds 1, 3, 4, 5, 6, and 7 were collected on a Bruker
Smart 1000 CCD diffractometer and covered more than a hemi-
sphere of reciprocal space by a combination of four sets of
exposures. The structures were solved by direct methods utilizing
SAINT,23 SHELX,24–26 and WinGX27 suite programs. Crystal
details and details of data collection are provided in Table 1.
bands of 2 appeared in hexane at 2029, 1911, 1906, and 1895 cm-1
.
The 31P NMR chemical shift was observed at 136.4 ppm, with JPW
) 419 Hz. Anal. Found: C, 46.48; H, 3.54. Calcd: C, 46.91; H,
3.79.
Acknowledgment. We gratefully acknowledge the finan-
cial support from the National Science Foundation (CHE 05-
43133) and the Robert A. Welch Foundation (A-0923).
(c) cis-W(CO)4[P(OPh)3]2 (3). To 0.174 g of 2 in 12 mL of
toluene was added 0.066 mL of P(OPh)3, and the solution was
heated at 60 °C for 24 h. The product, which consisted mostly of
3 with a small quantity of 2 remaining, was isolated following
removal of solvent under vacuum. The νCO infrared vibrations of 3
Supporting Information Available: CIF files of complexes 1,
2, 4, 5, 6, and 7 giving full details of the X-ray crystallographic
studies. Observed τ angle values for the metal-P(OPh)3 X-ray
structures in the Cambridge Crystallographic Database as of May
2007. Calculated cone angles for the crystallographically defined
metal-P(OPh)3 units in this study. This material is available free
of charge via the Internet at http://pubs.acs.org.
were observed at 2043, 1953, and 1933 cm-1
.
Isomerization Reactions. (a) Complete rearrangement of cis-
Mo(CO)4[P(OPh)3]2 (1) to trans-Mo(CO)4[P(OPh)3]2 (6) occurred
in toluene at 85 °C over a 48 h period. The νCO infrared absorption
of 6 was observed at 1939 cm-1 in toluene, with a 31P NMR signal
appearing at 160.5 ppm. Crystals suitable for X-ray analysis were
obtained from chloroform and cold methanol. When this process
was similarly carried out under an atmosphere of carbon monoxide,
the sole reaction product was Mo(CO)5P(OPh)3 (7).
OM700761Q
(23) SAINT-Plus, version 6.02; Bruker: Madison, WI, 1999.
(24) Sheldrick, G. SHELXS-86: Program for Crystal Structure Solution;
Institut fur Anorganische Chemie der Universitat: Gottingen, Germany,
1986.
(25) Sheldrick, G. SHELXL-97: Program for Crystal Structure Refine-
ment; Institut fur Anorganische Chemie der Universitat: Gottingen, Ger-
many, 1997.
(b) Isomerization of cis-W(CO)4[P(OPh)3]2 (3) in toluene was
performed as described above, leading to trans-W(CO)4[P(OPh)3]2
(4). The νCO infrared absorption of 4 in toluene appeared at 1932
cm-1, with the 31P NMR signal observed at 135.4 ppm (JPW
)
564 Hz). Complex 4 was crystallized from chloroform and cold
methanol. This isomerization reaction was unaffected when carried
out under an atmosphere of carbon monoxide.
(26) SHELXTL, version 5.0; Bruker: Madison, WI, 1999.
(27) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837–838.