Dangling phosphine complexes: Phosphine exchange in pentacarbonyl tungsten complexes of bis(diphenylphosphinomethyl)phenylphosphine

Article abstract of DOI:10.1016/j.jorganchem.2015.07.017

Abstract The reaction of [W(CO)₅NH₂Ph] with Ph₂PCH₂PPhCH₂PPh₂ under mild conditions and stoichiometric control leads to two sets of linkage isomers: [(OC)₅W{κ¹-PPh(CH₂PPh₂)₂}] 6a and [(OC)₅W(κ¹- PPh₂CH₂PPhCH₂PPh₂)] 6b; [(OC)₅W{μ-PPh₂CH₂PPh(CH₂PPh₂)}W(CO)₅] 7a and [(OC)₅W(μ-PPh₂CH₂PPhCH₂PPh₂)W(CO)₅] 7b. Isomers 6a and 6b exist in solution in equilibrium with K = 4.35 for 6b/6a in CDCl₃ at 55 °C. Under the same conditions, the rate of isomerization of 6a to 6b is 4.32 × 10^-6 s^-1. Isomerization is thought to proceed by an initial attack of a carbonyl group by a pendant phosphine followed by ring opening and a 1,2-phosphine shift. At higher temperatures, chelated complexes with four- and six-membered rings are formed: [(OC)₅W{μ-κ¹-, κ²-PPh₂CH₂ (PhPCH₂PPh₂)W(CO)₄ }] 9a, [(OC)₅W{μ-κ¹-, κ²-PPh (CH₂PPh₂)₂}W(CO)₄ }] 9b, and [(OC)₄W{κ²-(PPh₂CH₂)₂PPh}] 10b. X-ray crystal structures of 6a, 7b, and 9b have been determined.

Full text of DOI:10.1016/j.jorganchem.2015.07.017

Journal of Organometallic Chemistry 794 (2015) 258e265
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Dangling phosphine complexes: Phosphine exchange in
pentacarbonyl tungsten complexes of bis(diphenylphosphinomethyl)
phenylphosphine
Chaminda P. Gamage a, Ryan C. Bailey a, Ellen A. Keiter a, John R. Kuczynski a,
,
*
Kraig A. Wheeler a, Charlotte L. Stern c, Douglas E. Brandt b, Richard L. Keiter a
a Department of Chemistry, Eastern Illinois University, Charleston, IL 61920, USA
b Department of Physics, Eastern Illinois University, Charleston, IL 61920, USA
c Department of Chemistry, Northwestern University, Evanston, IL 60208, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The reaction of [W(CO)5NH2Ph] with Ph2PCH2PPhCH2PPh2 under mild conditions and stoichiometric
control leads to two sets of linkage isomers: [(OC)5W{
k
1-PPh(CH2PPh2)2}] 6a and [(OC)5W(k1
7a and [(OC)5W(
-
-
Received 25 May 2015
Received in revised form
13 July 2015
Accepted 14 July 2015
Available online 15 July 2015
PPh2CH2PPhCH2PPh2)] 6b; [(OC)5W{ -PPh2CH2PPh(CH2PPh2)}W(CO)5]
m
m
PPh2CH2PPhCH2PPh2)W(CO)5] 7b. Isomers 6a and 6b exist in solution in equilibrium with K ¼ 4.35 for
6b/6a in CDCl3 at 55 ꢀC. Under the same conditions, the rate of isomerization of 6a to 6b is
4.32 ꢁ 10ꢂ6 sꢂ1. Isomerization is thought to proceed by an initial attack of a carbonyl group by a pendant
phosphine followed by ring opening and a 1,2-phosphine shift. At higher temperatures, chelated com-
Keywords:
plexes with four- and six-membered rings are formed: [(OC)5W{
}] 9a, [(OC)5W{
1-, 2-PPh (CH2PPh2)2}W(CO)4 }] 9b, and [(OC)4W{
crystal structures of 6a, 7b, and 9b have been determined.
m
-
k
1-,
k
2-PPh2CH2 (PhPCH2PPh2)W(CO)4
2-(PPh2CH2)2PPh}] 10b. X-ray
Group 6 carbonyls
Pendant phosphines
Phosphine exchange
Linkage isomers
m
-k
k
k
© 2015 Elsevier B.V. All rights reserved.
bis(diphenylphosphinomethyl)
phenylphosphine
1. Introduction
Carbonyl complexes of dpep, first synthesized in 1961, have
been much more extensively investigated than those of dpmp
[1,16,17]. Whereas no M(CO)5 complexes of dpmp have been re-
ported, all five dpep complexes of W(CO)5 have been prepared
selectively from vinyl phosphines (Fig. 2). These consist of two
sets of linkage isomers (1, 2 and 3, 4) with uncoordinated phos-
phines and one complex in which dpep is fully coordinated (5)
[17].
The stability of metal carbonyl complexes of dpep suggested
that syntheses of similar complexes of dpmp should be possible.
That and our continuing interest in dangling phosphine carbonyl
complexes and their exchange of coordinated and uncoordinated
phosphine groups led us to initiate this study.
In our previous work we have shown that rates of intra-
molecular exchange of coordinated and uncoordinated phos-
phines at 55 ꢀC vary over a wide range for rather similar ligands
[18e20]. In this work we further examined the mechanism of
the exchange process by determining the rates of isomerization
of 6a and 6b (Scheme 1). Two mechanistic possibilities are
discussed.
Complexes
of
the
linear
polydentate
phosphines,
Ph2PCH2PPhCH2PPh2 (dpmp) and Ph2 PCH2CH2PPhCH2CH2PPh2
(dpep), have been known for many years [1]. A convenient syn-
thesis for dpmp first appeared in 1979 [2] but attracted notable
attention only after Balch and coworkers demonstrated its
remarkable capacity to form trimetallic complexes of linked d8 and
d10 metals (Fig. 1) [3,4]. Among the many chelated and bridged
complexes of dpmp are those in which one [5e11] or, in rare in-
stances, two [12,13] of the phosphine groups remain
uncoordinated.
Few studies have been reported for reactions of dpmp with
metal carbonyls other than those of Rh and Ir [3,14]. The one report
of reactions with [M(CO)6] (M ¼ Cr, Mo, W) gave only [cis-
M(CO)4(dpmp)] in which the central phosphorus atom remains
uncoordinated [15].
* Corresponding author.
E-mail address: rlkeiter@eiu.edu (R.L. Keiter).
http://dx.doi.org/10.1016/j.jorganchem.2015.07.017
0022-328X/© 2015 Elsevier B.V. All rights reserved.

C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
259
Fig. 1. Doubly bridged d8 and d10 dpmp complexes.
Fig. 2. Non-chelated W(CO)5 complexes of dpep.
Fig. 3. Targeted products from the reaction of [(OC)5WNH2Ph] with dpmp; 8 and 10a
were not observed.
Scheme 1. Isomerization of isomers 6a and 6b.
2. Results and discussion
2.1. Synthesis and characterization
Although selective methods for synthesizing dpmp complexes
analogous to those for dpep are not known, we have succeeded in
non-selectively preparing the target compounds by traditional
substitution reactions. The reaction of [(OC)5WNH2Ph] with dpmp
under various conditions gave W(CO)5 products 6a, 6b, 7a, and 7b
but not 8. In addition, chelated products 9a, 9b, and 10b, but not
10a, were characterized (Fig. 3).
To maximize the formation of 6a and 6b, an excess of dpmp was
used. Even so, some 7b was seen in these reactions as both 6a and
6b are also ligands competing for vacant tungsten sites. Complex 7a
was seldom observed in these reactions because its precursor 6a is
less abundant than 6b in the reaction mixtures and because 7a is
unstable with respect to 7b. With an excess of [(OC)5WNH2Ph], only
dimetallic complexes (7b, 9a, and 9b) were found. These results are
summarized in Scheme 2.
Scheme 2. A flowchart for the reaction of [(OC)5WNH2Ph] with dpmp.
The P-31 NMR spectrum of linkage isomer 6a shows a downfield
triplet with W-183 satellites for the coordinated central phosphine
and a doublet for the two uncoordinated phosphines, confirming
that the two are equivalent in solution. The P-31 NMR spectrum of
6b shows three nonequivalent phosphorus atoms. That the W(CO)5
is bound to a terminal phosphine rather than the central one of
dpmp is confirmed by the presence of a simple downfield doublet
with W-183 satellites. The two uncoordinated phosphine groups
give an upfield doublet of doublets (central phosphine) and doublet
(terminal phosphine). For 7b, the P-31 spectrum consists of an
upfield triplet corresponding to the unbound central phosphorus
atom and a doublet (with W satellites) for the two equivalent co-
ordinated terminal phosphines.
observed solid state configuration appears to be partially driven by
two sets of intramolecular, face-to-face stacked phenyl groups, each
separated by less than 3.5 Å. The arrangement of these pairs of
phenyl groups is nearly parallel, with a mutual angle of less than 6ꢀ.
Intermolecular stacking is not observed.
For 7b, the crystal structure (Fig. 5) shows that the two coor-
dinated phosphines P(1) and P(3) are nonequivalent as the unco-
ordinated phosphine P(2) is directed toward an equatorial carbonyl
of W(2) and away from the carbonyls of W(1). Just as in 6a, intra-
molecular, face-to-face stacking is observed with, in this case, three
phenyl groups each separated from another by less than 3.5 Å. The
groups are almost parallel with a mutual angle of less than 6ꢀ and,
as in 6a, intermolecular stacking is not observed. A second feature
to note for both 6a and 7b is that the smaller of the two PeCeP
angles is the one in which the uncoordinated phosphine is in close
The crystal structure of 6a (Fig. 4), shows that one of the
dangling phosphines is directed toward an equatorial carbonyl
while the other is directed away from the carbonyl plane. The

260
C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
Table 1
Four-bond phosphorus-carbon coupling constants (4JPC) and solid state phosphor-
usecarbon bond distances in W(CO)5 complexes.
Complex
4JPC(Hz)
3.00
P$$$CO (Å)
Ref
21
3.417
3.89
3.400
5.104
3.488
18a
0.0
18a
Fig. 4. Solid state molecular structure of [(OC)5W[k
1- PPh(CH2PPh2)2] 6a with 50%
probability ellipsoids. Selected bond lengths (Å) and bond angles (deg): W(1)-
P(1) ¼ 2.5194(8), P(3)-C(2) ¼ 3.430(4), P(3)-C(3) ¼ 3.925(4); P(1)-C-P(2) ¼ 115.6(1),
P(1)-C-P(3) ¼ 114.9(1).
2.70
2.50
3.430
6.40
3.203
24
Fig. 5. Solid state molecular structure of [(OC)5W(
with 50% probability ellipsoids. Selected bond lengths (Å) and bond angles (deg):
W(1)-P(1) 2.5424(11), W(2)-P(3) 2.5209(11), P(2)-C(38) 3.488(5), P(2)-
C(42) ¼ 3.794(5); P(1)-C-P(2) ¼ 119.1(2), P(2)-C-P(3) ¼ 117.0(3).
m-PPh2CH2PPhCH2PPh2)W(CO)5] 7b
¼
¼
¼
the complexes listed in Table 1, cis 4JPC coupling has been observed
in [(OC)5W{
1-PPh2CH2P(p-tol)2}] (3.2 Hz), [(OC)5W{ 1-P(p-
tol)2CH2PPh2}] (2.9 Hz), and [(OC)5W{
1-P(p-tol)2CH2P(p-tol)2}]
k
k
k
(2.8 Hz) [20,21]. An examination of four-bond PeC couplings re-
ported in the literature suggests that our values, while comparable
to those involving pi bonds [22], are larger than expected for sys-
tems involving sigma bonds. For example, in the compounds
proximity to a carbonyl group.
The C-13 spectra of the carbonyl region for complexes 6a, 6b,
and 7b show four-bond PeC coupling, JPC (2.5, 1.9 and 2.7 Hz,
4
4
respectively), between the uncoordinated phosphine and a cis
Ph2PCH2CH2CH2CH3 and [BuPh2PCH2CH2CH2CH3]þ, JP-C(Me) cou-
carbonyl group. It should be noted that 4JPC coupling is not observed
plings are measured as
PhP(CH2PPh2)2, the JPC coupling (P-CPh) is 0.6 Hz.
0 and 0.7 Hz respectively [23]. In
4
4
for the trans carbonyl groups. Table 1 shows JPC values and solid
state distances between pendant phosphines and the carbon atoms
of the closest carbonyl group for several complexes. In addition to
4
We believe that the larger JPC values for our pentacarbonyl
complexes result from a favorable structural conformation in

C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
261
solution and it seems likely that there is a “through-space”
component arising from a non-bonded interaction of the uncoor-
dinated central P atom with the cis carbonyl groups. While one
would not expect the solid state structure to be the same as that in
solution, it does seem that the most probable solution arrangement
is similar to that in the solid state with the uncoordinated phos-
phine on average near a cis carbonyl group. The uncoordinated
phosphorus atom in [(OC)5W{k
1-PPh2(C]CH2)PPh2}] lies away
from the carbonyl plane and 4JPC is zero (Table 1). In the solid state
only one of the two uncoordinated phosphines of 6a lies near a
carbonyl group but in solution it can be imagined that each takes a
turn to give an average interaction. Similarly, the central uncoor-
dinated arm of 7b likely alternates between cis carbonyls of the two
W(CO)5 units.
The complex [(OC)5W{
k
1-PPh2C(SiMe3) ¼ PMes}] is of special
4
interest (Table 1). It has two features that enhance JPC coupling.
First, there is a double bond in the coupling pathway and second, its
configuration places the uncoordinated phosphine especially close
to a carbonyl group (PeCO distance ¼ 3.203 Å). Thus for this system
4
we find by far the largest JPC coupling (6.4 Hz) [24].
The reaction of pure 6a with [(OC)5WNH2Ph] led to a mixture of
7a, 7b, and 9a (Scheme 3). Evidence for 7a was given by P-31 NMR
in which the three nonequivalent phosphorus atoms give three sets
of signals, two downfield and one upfield, corresponding to two
coordinated and one uncoordinated phosphine. Coupling constants
are consistent with other complexes in which the phosphorus
atoms are separated by one carbon [21].
Fig. 6. Solid state molecular structure of [(OC)5W{
m
-
k1
,
k
2-PPh (CH2PPh2)2}W(CO)4 ] 9b
with 50% probability ellipsoids. Selected bond lengths (Å) and bond angles (deg):
W(1)-P(2) ¼ 2.5279(11), W(2)-P(3) ¼ 2.5080(11), W(2)-P(1) ¼ 2.4949(11); P(1)-C(22)-
P(2) ¼ 124.3(2); P(2)-C(29)-P(3) ¼ 119.3(3), C(22)-P(2)-C(29) ¼ 107.2(2), P(1)-W(2)-
P(3) ¼ 87.22(4).
Attempts to prepare
8 from the reaction of 7b with
[(OC)5WNH2Ph] led only to formation of 9a. The presence of two
CH2 groups separating phosphorus atoms in 5 as compared with
one in 8 emphasizes the importance of steric requirements for
trimetallic systems. The P-31 NMR spectrum of 9a shows three
nonequivalent phosphorus atoms giving rise to two doublets (ter-
minal phosphorus atoms) and one doublet of doublets (central
phosphorus atom).
appropriate time intervals. The concentration of 6a in CDCl3 at
55 ꢀC as a function of time is shown in Fig. 7. The decrease in the
concentration of 6a is matched by an increase in the concentration
of 6b. At equilibrium the ratio 6b/6a is 4.35 0.51, showing that 6b
is somewhat more stable than 6a under these conditions.
A plot of ln{[6a]-[6a]eq}/{[6ain]-[6a]eq} versus time gave a
straight line with slope equal to -(k1 þ k-1) (Fig. 8) [25]. Rate con-
stants of (4.32 0.66) ꢁ 10ꢂ6 sꢂ1 and (0.99 0.12) ꢁ 10ꢂ6 sꢂ1 were
found for the forward and reverse reactions, respectively.
These values may be compared to those for isomerization of
other complexes containing pendant phosphines (Scheme 4). Rates
of isomerization for 11a, 12a, and 13a vary over four orders of
magnitude.
An excess of [(OC)5WNH2Ph] with dpmp led to formation of 9b.
Its P-31 NMR spectrum consists of a triplet corresponding to a
phosphorus atom bound to W(CO)5 and a doublet corresponding to
two phosphorus atoms coordinated to W(CO)4. The crystal struc-
ture of 9b reveals a six-membered ring with a distorted chair
arrangement (Fig. 6). Two of the five phenyl groups are stacked
^
with less than 3.5 A separation.
2.2. Kinetics and thermodynamics
The isomerization reaction, 6a # 6b, starting with pure 6a or
pure 6b was monitored with 31P{1H} NMR by taking integrals at
Scheme 3. Formation and subsequent reactions of 7a.
Fig. 7. A plot of concentration versus time for the isomerization of 6a.

262
C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
Fig. 8. A plot of ln{[6a]-[6a]eq}/[6a]in-[6a]eq versus t (ksec).
Scheme 5. Pathways for isomerization of 11a.
Previously we have proposed two mechanistic pathways that
account for the observed rate differences. The relatively rapid
isomerization of 11a results because one of the two pendant arms
interacts with a bound CO group, leading to dissociation of the
coordinated phosphine (Scheme 5). When the ring opens, isomer
11b could potentially form either by a 1,2-shift or by coordination of
the second pendant phosphine followed by dissociation of the
carbonyl-attached phosphine. The latter reaction is kinetically
more favorable [18a].
Isomerization of 12a can only proceed by a 1,2-shift because it
does not have a second arm for the faster pathway available to
isomer 11a (Scheme 6).
The isomerization of 13a is faster than that of 12a but slower
than for 11a. Evidence suggests that 13a, also lacking a second
pendant arm, follows a 1,2-shift mechanism with the intermediate
relative rate resulting from the close proximity of its one pendant
arm to the carbonyl plane [20].
Isomer 6a, like 11a, has two pendant arms and so both mecha-
nistic pathways shown in Scheme 7 are theoretically possible for its
isomerization. The rate of isomerization of 6a is approximately the
Scheme 6. Mechanism for isomerization of 12a.
same as for 13a, suggesting that a 1,2-shift is the most plausible
mechanistic route. If the isomerization parallels that of 11a, an in-
termediate seven-membered ring would be required and the rate
of its formation may be slow compared to the 1,2-shift.
Conversion of 7a to 7b was observed with P-31 NMR, but the
position of equilibrium between the two linkage isomers so greatly
favored 7b that it was not possible to determine the equilibrium
constant. Overlapping signals in the mixture also prevented
determination of a reliable rate constant for the conversion.
3. Conclusions
Under mild reaction conditions and stoichiometric control we
have synthesized six new W(CO)5 complexes of dpmp and have
obtained X-ray structures of three. This work has broadened the
scope of linkage isomers examined and revealed additional struc-
tural features that influence isomerization rates.
We have established that the rate of exchange between coor-
dinated and uncoordinated phosphines in W(CO)5L, follows the
order: L ¼ Ph2PCH2CH(PPh2)2 >> PhP(CH2PPh2)2 ~ Ph2PCH2 P(p-
tol)2 >> Ph2PCH2CH2P(p-tol)2. In all cases the first step is the attack
Scheme 4. Rates of isomerization at 55 ꢀC.

C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
263
4.2. Pentacarbonyl[bis(diphenylphosphinomethyl)
phenylphosphine]ditungsten(0), [(OC)5W{
1- PbPh(CH2Pa,cPh2)2} ]
6a and [(OC)5W(
1- PaPh2CH2PbPhCH2PcPh2)] 6b
k
k
To a solution of dpmp (2.00 g, 3.95 mmol) in 30 mL of CH2Cl2
was added [(OC)5WNH2Ph] (0.590 g, 1.41 mmol). The intense yel-
low color of the solution slowly dissipated over 24 h after which the
solvent was removed. The residue consisted of 6a, 6b, unreacted
dpmp and a minor amount of 7b. The mixture was eluted with
petroleum ether (90%)/ethyl acetate (10%) through a 1.2500 x 1800
neutral silica gel (60e200 mesh) column. The free ligand, dpmp,
eluted first followed by a band containing both 6a and 6b. Crys-
tallization of the 6a/6b mixture from CH2Cl2/CH3OH (1:2) at ꢂ5 ꢀC
gave 6a (0.35 g, 30%). IR: 1934, 1979, 2070. 31P NMR: da,c ꢂ24.8, d, db
3.9, t; JPP ¼ 86.8, JWP ¼ 250.4. 13C NMR: dtrans 199.4, d,
2JPC(trans) ¼ 23.6; dcis 197.6, dd, 2JPC(cis) ¼ 7.2 Hz, 4JPC(cis) ¼ 2.5. Anal.
Calc. for C37H29O5P3W: C, 53.52; H, 3.52. Found: C, 53.89; H, 3.61.
The mother liquor was concentrated and at ꢂ5 ꢀC 6b precipitated
(0.25 g, 21%). IR: 1937, 1981, 2070. 31P NMR: da 9.3 d, db ꢂ34.7 dd,
dC ꢂ23.5 d; 2JP(a)P(b) ¼ 96.1, 2JP(b)P(c) ¼ 112.6, 1JWP ¼ 245.0. 13C NMR:
2
1
2
2
dtrans 199.5 d, JPC(trans) ¼ 23.2; dcis 197.0, dd, JPC(cis) ¼ 7.2,
4JPC(cis) ¼ 2.0. Anal. Calcd. for C37H29O5P3W: C, 53.52; H, 3.52.
Found: C, 54.13; H, 3.58. The dimetallic complex 7b was not
recovered from this reaction.
Scheme 7. Mechanistic pathways for the formation of 6b.
of a carbonyl group by a pendant phosphine leading to dissociation
of the coordinated phosphine [26]. For complexes of Ph2PCH2P(p-
tol)2 and Ph2PCH2CH2P(p-tol)2, the isomerization proceeds by a 1,2-
shift while for complexes of Ph2PCH2CH(PPh2)2, substitution of the
bound phosphine with the second pendant phosphine is faster than
a 1,2-shift. Isomerization of the PhP(CH2PPh2)2 complex could go
either way but reaction rates suggest that a 1,2-shift route is more
likely. In our 1998 paper we claimed that “two arms are much
better than one” for the acceleration of phosphine exchange [18b].
It is now clear that arm length must also be considered.
Our exchange studies may have broad implications for reactions
of phosphines with metal carbonyl complexes. Substitution of CO
by a phosphine in group 6 metal carbonyls follows a two-term rate
law. There is general agreement that the dominant first-order term
arises from a dissociative pathway. There is no clear consensus,
however, on the mechanistic origin of the second-order term
[27e31]. Phosphine attack on the metal or on a bound carbonyl
group have both been considered and largely rejected in favor of an
interchange dissociative mechanism in which the incoming PR3
and the departing CO ligand interact prior to full dissociation. Our
work suggests that the phosphine may indeed attack a bound
carbonyl group but not the one that is replaced.
4.3. [bis(diphenylphosphinomethyl)phenylphosphine]
[bis(pentacarbonyl)tungsten(0)], [(OC)5W(m-
PaPh2CH2PbPhCH2PcPh2)W(CO)5] 7b
A
CH2Cl2 solution (30 mL) of [(OC)5WNH2Ph] (1.00 g,
2.40 mmol) and dpmp (0.570 g, 1.11 mmol) was stirred for 24 h. The
residue resulting from removal of solvent was crystallized from a
1:2 volume of CH2Cl2/CH3OH to give a powder consisting princi-
pally of 7b with minor amounts of chelated products, [(OC)5W{
2- PaPh2CH2 (PbPhCH2PcPh2)}W(CO)4] 9a and [(OC)5W{ k1
PbPhCH2 (CH2Pa,cPh2)2}W(CO)4] 9b. No evidence for the formation
of [(OC)5W{ - PaPh2CH2PbPh(CH2PcPh2)}W(CO)5] 7a was observed.
m
,
-
k1
k2
,
-
k
m-
m
The powder was dissolved in a minimum of CH2Cl2 to which an
equal amount of CH3OH was added. At ꢂ5 ꢀC after 120 h, crystals of
7b were obtained (0.38 g, 29%). IR: 1939, 1981, 2071. 31P NMR: dP(a,c)
8.8, d, dP(b) ꢂ37.9, t; 2JPP ¼ 79.3, 1JWP ¼ 245.3. 13C NMR: dtrans 199.4,
2
2
4
d, JPC(trans) ¼ 23.3; dcis 196.9 dd, JPC(cis) ¼ 7.2, JPC(cis) ¼ 2.7. Anal.
Calcd. for C42H29P3W2O10: C, 43.70; H, 2.53. Found: C, 44.12; H, 2.58.
4.4. [bis(diphenylphosphinomethyl)
phenylphosphine](pentacarbonyl) (tetracarbonyl)ditungsten(0)],
[(OC)5W{m- , k
k1 2- PaPh2CH2 (PbPhCH2PcPh2)}W(CO)4] 9a
4. Experimental section
A solution of 7b (0.16 g, 0.14 mmol) and [(OC)5WNH2Ph] (0.10 g,
0.24 mmol) in toluene (25 mL) was heated for 24 h at 60 ꢀC. The
residue resulting after solvent removal was crystallized from a 1:2
solution of CH2Cl2/CH3OH to give 9a (0.090 g, 57%). IR: yCO 1898,
1934, 1978, 2019, 2071. 31P NMR: dP(a) 8.1, d, dP(b) ꢂ33.1 dd,
4.1. General
All reactions were carried out under a dry nitrogen atmosphere
using standard Schlenk techniques. The tridentate ligand,
Ph2PCH2PPhCH2PPh2 (dpmp), was prepared by reaction of
Ph2PCH2SiMe3 (synthesized from NaPPh2 in liquid ammonia [32,33]
with PhPCl2 [2]. Our analysis of the AB2 pattern in the 31P NMR spec-
2
2
1
dP(c) ꢂ17.9 d; JP(a)P(b) ¼ 19.4, JP(b)P(c) ¼ 36.3; JWP(a) ¼ 247.9,
1
1JWP(b) ¼ 226.4, JWP(c) ¼ 201.7. Anal. Calcd. for C41H29O9P3W2: C,
43.72; H, 2.60. Found: C, 43.95; H, 2.72. The goal of this reaction had
been to synthesize [{(OC)5W}3dpmp] 8 but there was no evidence
for it as a product.
trumis inagreement with that reported inthe literature (
d
Pterm ꢂ22.7,
Pcentral ꢂ32.9, JPP ¼ 114.2; lit. ꢂ22.5, ꢂ32.7, JPP ¼ 115.9 [34]).
Mp ¼ 138e139 ꢀC (lit. 122e123 [2]; 140e141 [34]). The aniline com-
plex, [(OC)5WNH2Ph], was prepared as described previously [18a]. A
300 MHz instrument was used to obtain phosphorus-31 and carbon-
13 NMR spectra from CDCl3 solutions, referenced to 85%
phosphoric acid and TMS, respectively. Infrared spectra of the yCO re-
4.5. [bis(diphenylphosphinomethyl)
phenylphosphine](pentacarbonyltunsten 0)
(tetracarbonyltungsten(0)], [(OC)5W{m- , k
k1 2- PbPhCH2
(CH2Pa,cPh2)2}W(CO)4] 9b
gion were obtained from CHCl3 solutions; values are reported in cmꢂ1
.
A solution of [(OC)5WNH2Ph] (1.0 g, 2.4 mmol) and dpmp

264
C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
Table 2
Crystal data and structure refinement for complexes 6a, 7b, and 9b.
6a
7b
9b
Empirical Formula
Formula Weight
Crystal Dimensions
Crystal System
C37H29WO5P3
830.40
C42W2P3H29O10
1154.31
C41W2P3O9H29
1126.30
0.31 ꢁ 0.41 ꢁ 0.21 mm
0.13 ꢁ 0.24 ꢁ 0.26 mm
0.31 ꢁ 0.19 ꢁ 0.26 mm
tetragonal
triclinic
Triclinic
Lattice Type
I-centered
Primitive
Primitive
Lattice Parameters
a ¼ 32.902(3) Å
c ¼ 12.681(2) Å
V ¼ 13727.7(22) Å3
a ¼ 11.7988(12) Å
b ¼ 13.651(1) Å
c ¼ 14.579(2) Å
a ¼ 9.647(1) Å
b ¼ 11.567(2) Å
c ¼ 18.726(3) Å
a
b
g
¼ 102.148(2) Å
¼ 107.335(2) Å
¼ 106.689(2)
a
b
g
¼ 95.383(2) Å
¼ 94.288(2) Å
¼ 102.280(2) Å
V ¼ 2031.4(4) Å3
P-1 (#2)
2
V ¼ 2023.2(5) Å3
P-1 (#2)
2
Space Group
Z value
Dcalc
I41/a (#88)
16
1.607 g/cm3
6560.00
35.51 cmꢂ1
6565
1.887 g/cm3
1108.00
58.41 cmꢂ1
8025
1.849 g/cm3
1080.00
58.60 cmꢂ1
7809
F000
m
(MoKa)
No. Observations (I > 3.00s(I))
No. Variables
415
514
496
Reflection/Parameter Ratio
Residuals: R1; wR2
Goodness of Fit Indicator
15.82
0.030; 0.038
1.34
15.61
0.020; 0.041
1.62
15.74
0.023; 0.048
1.67
Max Shift/Error in Final Cycle
Maximum peak in Final Diff. Map
Minimum peak in Final Diff. Map
0.03
0.00
0.00
1.10 eꢂ/Å3
ꢂ0.79 eꢂ/Å3
1.25 eꢂ/Å3
ꢂ0.97 eꢂ/Å3
1.81 eꢂ/Å3
ꢂ0.92 eꢂ/Å3
(0.38 g, 0.75 mmol) was stirred for 24 h. The solvent was removed
and the residue was crystallized from a 1:2 ratio of CH2Cl2/CH3OH
to give 0.63 g, shown to consist of 9b and 7b. Recrystallization of
this mixture from a 1:1 solution of CDCl3/CH3OH gave light yellow
crystals of 9b (0.090 g, 11%). IR: 1898, 1933, 1982, 2021, 2071. 31P
NMR: dP(ac) 4.7 d, dP(b) ꢂ2.7 t; JP(a,c)P(b) ¼ 12.1, JWP(a,c) ¼ 237.1,
1JWP(b) ¼ 249.1. Anal. Calcd. for C41H29O9P3W2: C, 43.72; H, 2.60.
Found: C, 43.90; H, 2.75.
constant temperature bath of the appropriate temperature. The
NMR probe was brought to the same temperature and 31P NMR
spectra were recorded at selected time intervals until a constant
ratio of isomers resulted. A delay time of 18 s was used to
insure sufficient relaxation to allow meaningful integrations. The
signals from isomers of each spectrum were integrated three
times and the average value was used in subsequent
calculations.
2
1
4.6. Evidence for formation of [bis(diphenylphosphinomethyl)
4.9. X-ray crystallography
phenylphosphine][bis(pentacarbonyl)ditungsten(0)], [(OC)5W{m-
PaPh2CH2PbPh(CH2PcPh2)}W(CO)5] }]7a
X-ray quality crystals of 6a, 7b, and 9b were grown by slow
diffusion of methanol into a solution of CH2Cl2. Colorless block
crystals were mounted on a glass fiber. All measurements were
made on a Bruker SMART-1000 CCD plate detector with graphite
A solution of [(OC)5WNH2Ph] (0.050 g, 0.12 mmol) and 6a
(0.080 g, 0.096 mmol) in 20 mL of CH2Cl2 was stirred for 24 h. The
residue, obtained upon solvent removal, was shown by 31P NMR to
contain 7a, 7b, 9a, and 9b. 31P NMR assigned to 7a: dP(a) 8.3 d, dP(b)
monochromated Mo-K
temperature of ꢂ120
a
1
radiation. The data were collected at a
ꢀC and processed using the SMART-NT
2
2
11.9 dd, dP(c) ꢂ20.7 d; JP(a)P(b) ¼ 57.9, JP(b)P(c) ¼ 137.5. This com-
pound could not be separated from 7b, 9a, and 10b by chroma-
tography and is unstable with respect to its linkage isomer 7b.
Overlap of P-31 signals from 7b, 9a, and 9b prevented determina-
tion of tungsten-phosphorus satellites for 7a.
AND SAINT-NT programs from Bruker. The data were corrected
for Lorentz and polarization effects. The structure was solved using
Direct Methods SHELXS, and refined with the SHELXL refinement
package using Least Squares minimization [35]. The non-hydrogen
atoms were refined anisotropically. The hydrogen atoms were
included in idealized positions but not refined. Neutral atom
scattering factors were taken from Cromer and Waber [36].
Anomalous dispersion effects were included in Fcalc [37]; the
4.7. Formation of tetracarbonyl[bis(diphenylphosphinomethyl)
phenylphosphine]tungsten(0), [(OC)4W(k2
PaPh2CH2PbPhCH2PcPh2)] 10b
-
values for
D D
f1and f2 are those of Creagh and McAuley [38]. The
values for the mass attenuation coefficients are those of Creagh
and Hubbell [39]. Additional crystallographic details are summa-
rized in Table 2.
CCDC 1033877, 1033878, and 1033879 contain the supplemen-
tary crystallographic data for this paper. These data can be obtained
free of charge from the Cambridge Crystallographic Data Center via
www.ccdc.cam.ac.uk/data_request/cif.
A 1:1 mol ratio of [(OC)5WNH2Ph] and dpmp in CH2Cl2 was
stirred for 24 h. As identified in the reaction mixture by P-31 NMR,
major products were 6a, 6b, and 7b. Minor products were 7a, 9a,
and 10b. 31P NMR assigned to 10b: dP(a,c) 3.1, d, d P(b) ꢂ38.4, t; 2JP(a,c,)
¼ 73.6. Complex 10b was previously prepared by refluxing
P(b)
dpmp with W(CO)6 in toluene over 30 h [15].
4.8. Equilibrium and kinetic studies
Acknowledgments
These experiments were performed on a pure sample of each
isomer (30e40 mg) dissolved in 0.4e0.5 mL of CDCl3 and flame-
sealed in a standard NMR tube. The samples were placed in a
We thank the U. S. National Science Foundation (CHE-136423)
and the Camille and Henry Dreyfus Foundation (SF-98-004) for
support of this work.

C.P. Gamage et al. / Journal of Organometallic Chemistry 794 (2015) 258e265
D.E. Brandt, Organometallics 31 (2012) 4619e4622.
265
References
[21] J.W. Benson, R.L. Keiter, E.A. Keiter, A.L. Rheingold, G.P.A. Yap, V.V. Mainz,
Organometallics 17 (1998) 4275e4281.
[22] (a) C. Slugovc, P. Wiede, K. Mereiter, R. Schmid, K. Kirchner, Organometallics
16 (1997) 2768e2769;
[1] (a) F.A. Cotton, B. Hong, Prog. Inorg. Chem. 40 (1992) 179e289;
(b) H.A.Mayer, W.C.Kaska, Chem. Rev. 94(1994) 1239e1272; (c) J.-C. Hierso,
R. Amardeil, E. Bentabet, R. Broussier, B. Gautheron, P. Meunier, P. Kalck,
Coord. Chem. Rev. 236 (2003) 143e206;
(d) A. Pascariu, S. Iliescu, A. Popa, G. Ilia, J. Organomet. Chem. 694 (2009)
3982e4000.
[2] R. Appel, K. Geisler, H.-F. Scholer, Chem. Ber. 112 (1979) 648e653.
[3] A.L. Balch, Prog. Inorg. Chem. 41 (1994) 239e329.
[4] (a) T. Tanase, Bull. Chem. Soc. Jpn. 75 (2002) 1407e1422;
(b) E. Goto, R.A. Begum, C. Ueno, A. Hosokawa, C. Yamamoto, K. Nakamae,
B. Kure, T. Nakajima, T. Kajiwara, T. Tanase, Organometallics 33 (2014)
1893e1904.
[5] A.L. Balch, V.J. Catalano, B.C. Noll, M.M. Olmstead, J. Am. Chem. Soc. 111 (1990)
7558e7566.
[6] T. Tanase, H. Takahata, Y. Yamamoto, Inorg. Chim. Acta 264 (1997) 5e9.
[7] T. Tanase, T. Igoshi, Y. Yamamoto, Inorg. Chim. Acta 256 (1997) 61e67.
[8] T. Tanase, H. Ukaji, H. Takahata, H. Toda, T. Igoshi, Y. Yamamoto, Organome-
tallics 17 (1998) 196e209.
[9] Y. Kosaka, Y. Shinozaki, Y. Tsutsumi, Y. Kaburagi, Y. Yamamoto, Y. Sunada,
K. Tatsumi, J. Organomet. Chem. 671 (2003) 8e12.
[10] Y. Yamamoto, Y. Kosaka, Y. Tsutsumi, Y. Kaburagi, K. Kuge, Y. Sunada,
K. Tatsumi, Eur. J. Inorg. Chem. (2004) 134e142.
[11] Y. Yamamoto, Y. Sinozuka, Y. Tsutsumi, K. Fuse, Kuge, Y. Sunada, K. Tatsumi,
Inorg. Chim. Acta 357 (2004) 1270e1282.
[12] M. Bardaji, A. Laguna, V.M. Orera, M.D. Villacampa, Inorg. Chem. 37 (1998)
5125e5130.
[13] Y. Yamamoto, Y. Kosaka, Y. Tsutsumi, Y. Sunada, K. Tatsumi, T. Fumie,
T. Shigetoshi, Dalton Trans. (2004) 2969e2978.
[14] C. Moreno, M.J. Macazaga, S. Delgado, J. Organomet. Chem. 407 (1991)
125e133.
[15] A.L. Balch, R.R. Guimerans, J. Linehan, Inorg. Chem. 24 (1985) 290e296.
[16] J. Chart, H.R. Watson, J. Chem. Soc. (1961) 4980e4988.
[17] R.L. Keiter, J.W. Brodack, R.D. Borger, L.W. Cary, Inorg. Chem. 21 (1982)
1256e1259.
[18] (a) R.L. Keiter, P. Ye, E.A. Keiter, J.W. Benson, W. Lin, D.E. Brandt, J.S. Southern,
J.A.L. Rheingold, I. Guzei, K.A. Wheeler, L.W. Cary, Inorg. Chim. Acta 364 (2010)
176e184;
(b) D.D. Wick, W.D. Jones, Inorg. Chim. Acta 362 (2009) 4416e4421;
(c) C. Slugovc, K. Mereiter, R. Schmid, K. Kirchner, Organometallics 18 (1999)
1011e1017;
(d) O. Kühl, Phosphorus-31 NMR Spectroscopy: a Concise Introduction for the
Synthetic Organic and Organometallic Chemist (Google eBook), Springer,
Verlag, Berlin, 2008.
€
[23] G.A. Gray, S.E. Cremer, K.L. Marsi, J. Am. Chem. Soc. 98 (1976) 2109e2118.
[24] S. Ito, K. Nishide, M. Yoshifuji, Organometallics 25 (2006) 1424e1430.
[25] J.H. Espenson, Chemical Kinetics and Reaction Mechanisms, second ed.,
McGraw-Hill, New York, 1995. Ch. 3.
[26] Others have shown that pendant ligands may accelerate substitution rates in
metal carbonyls. See (a) D.J. Darensbourg, H.P. Wiegreffe, Inorg. Chem. 29
(1990) 592e597;
(b) D.J. Darensbourg, J.A. Joyce, C.J. Bischoff, J.H. Reibenspies, Inorg. Chem. 30
(1991) 1137e1142;
(c) A.L. Phelps, M.V. Rampersand, S.B. Fitch, M.Y. Darensbourg,
D.J. Darensbourg, Inorg. Chem. 45 (2006) 119e126;
(d) S.P. Jeffery, M.L. Singleton, J.H. Reibenspies, M.Y. Darensbourg, Inorg.
Chem. 46 (2007) 179e185.
[27] J.R. Graham, R.J. Angelici, Inorg. Chem. 6 (1967) 2082e2085.
[28] W.D. Covey, T.L. Brown, Inorg. Chem. 12 (1973) 2820e2825.
[29] D.J. Darensbourg, Adv. Organomet. Chem. 21 (1982) 113e145.
[30] J.D. Atwood, Inorganic and Organometallic Reaction Mechanisms, Wiley-VCH,
New York, 1997, pp. 99e100.
[31] J. Hartwig, Organotransition Metal Chemistry, University Science Books,
Sausalito, CA, 2010, pp. 241e242.
[32] W. Hewertson, H.R. Watson, J. Chem. Soc. (1962) 1490e1494.
[33] B.E. Cooper, W.J. Owen, J. Organomet. Chem. 21 (1970) 329e332.
[34] K.A. Asker, K.R. Seddon, M.C. Grossel, New J. Chem. 13 (1989) 587e594.
[35] G.M. Sheldrick, Acta Cryst. A64 (2008) 112e122.
[36] D.T. Cromer, J.T. Waber, in: J.A. Ibers, W.C. Hamilton (Eds.), International
Tables for Crystallography, vol. IV, Kynoch Press, Birmingham, England, 1974.
Table 2.2 A.
[37] J.A. Ibers, W.C. Hamilton, Acta Crystallogr. 17 (1964) 781.
[38] D.C. Creagh, W.J. McAuley, in: A.J.C. Wilson (Ed.), International Tables for
Crystallography, vol. C, Kluwer Academic Publishers, Boston, 1992, pp.
219e222. Table 4.2.6.8.
[39] D.C. Creagh, J.H. Hubbell, in: A.J.C. Wilson (Ed.), International Tables for
Crystallography, vol. C, Kluwer Academic Publishers, Boston, 1992, pp.
200e206. Table 4.2.4.3.
(b) R.L. Keiter, J.W. Benson, E.A. Keiter, W. Lin, Z. Jia, D.M. Olson, D.E. Brandt,
J.L. Wheeler, Organometallics 17 (1998) 4291e4297;
(c) R.L. Keiter, E.A. Keiter, D.M. Olson, J.R. Bush, W. Lin, J.W. Benson, Organ-
ometallics 13 (1994) 3752e3754.
[19] R.L. Keiter, J.W. Benson, Z. Jia, E.A. Keiter, D.E. Brandt, Organometallics 19
(2000) 4518e4522.
[20] R.L. Keiter, D. Chen, G.A. Holloway, E.A. Keiter, Y. Zang, M.T. Huml, J. Filley,