Functionalization reactions characteristic of a robust bicyclic diphosphane framework

Article abstract of DOI:10.1021/ic401052a

The 3,4,8,9-tetramethyl-1,6-diphospha-bicyclo-[4.4.0]deca-3,8-diene (P ₂(C₆H₁₀)₂) framework containing a P-P bond has allowed for an unprecedented selectivity toward functionalization of a single phosphorus lone pair with reference to acyclic diphosphane molecules. Functionalization at the second phosphorus atom was found to proceed at a significantly slower rate, thus opening the pathway for obtaining mixed functional groups for a pair of P-P bonded λ⁵-phosphorus atoms. Reactivity with the chalcogen-atom donors MesCNO (Mes = 2,4,6-C ₆H₂Me₃) and SSbPh₃ has allowed for the selective synthesis of the diphosphane chalcogenides OP₂(C ₆H₁₀)₂ (87%), O₂P₂(C ₆H₁₀)₂ (94%), SP₂(C ₆H₁₀)₂ (56%), and S₂P ₂(C₆H₁₀)₂ (87%). Computational studies indicate that the oxygen-atom transfer reactions involve penta-coordinated phosphorus intermediates that have four-membered {PONC} cycles. The P-E bond dissociation enthalpies in EP₂(C ₆H₁₀)₂ were measured via calorimetric studies to be 134.7 ± 2.1 kcal/mol for P-O, and 93 ± 3 kcal/mol for P-S, respectively, in good agreement with the computed values. Additional reactivity with breaking of the P-P bond and formation of diphosphinate O₃P ₂(C₆H₁₀)₂ was only observed to occur upon heating of dimethylsulfoxide solutions of the precursor. Reactivity of diphosphane P₂(C₆H₁₀)₂ with azides allowed the isolation of monoiminophosphoranes (RN)P₂(C ₆H₁₀)₂(R = Mes, CPh₃, SiMe ₃), and treatment with additional MesN₃ yielded symmetric and unsymmetric diiminodiphosphoranes (RN)(MesN)P₂(C ₆H₁₀)₂ (91% for R = Mes). Metalation reactions with the bulky diiminodiphosphorane ligand (MesN)₂P ₂(C₆H₁₀)₂ (nppn) allowed for the isolation and characterization of (nppn)Mo(η³-C₃H ₅)Cl(CO)₂ (91%), (nppn)NiCl₂ (76%), and [(nppn)Ni(η³-2-C₃H₄Me)][OTf] showing that these ligands provide an attractive preorganized binding pocket for both late and early transition metals.

Full text of DOI:10.1021/ic401052a

Article
pubs.acs.org/IC
Functionalization Reactions Characteristic of a Robust Bicyclic
Diphosphane Framework
Daniel Tofan,† Manuel Temprado,*,‡ Subhojit Majumdar,§ Carl D. Hoff,*,§
and Christopher C. Cummins*,†
†Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139,
United States
‡Department of Physical Chemistry, Universidad de Alcala,
́
Ctra. Madrid-Barcelona Km. 33,600, Madrid 28871, Spain
§Department of Chemistry, University of Miami, 1301 Memorial Drive, Coral Gables, Florida 33021, United States
S
* Supporting Information
ABSTRACT: The 3,4,8,9-tetramethyl-1,6-diphospha-bicyclo-
[4.4.0]deca-3,8-diene (P2(C6H10)2) framework containing a P−P
bond has allowed for an unprecedented selectivity toward
functionalization of a single phosphorus lone pair with reference
to acyclic diphosphane molecules. Functionalization at the second
phosphorus atom was found to proceed at a significantly slower rate,
thus opening the pathway for obtaining mixed functional groups for
a pair of P−P bonded λ5-phosphorus atoms. Reactivity with the
chalcogen-atom donors MesCNO (Mes = 2,4,6-C6H2Me3) and
SSbPh3 has allowed for the selective synthesis of the diphosphane
chalcogenides OP2(C6H10)2 (87%), O2P2(C6H10)2 (94%),
SP2(C6H10)2 (56%), and S2P2(C6H10)2 (87%). Computational
studies indicate that the oxygen-atom transfer reactions involve
penta-coordinated phosphorus intermediates that have four-membered {PONC} cycles. The P−E bond dissociation enthalpies
in EP2(C6H10)2 were measured via calorimetric studies to be 134.7 2.1 kcal/mol for P−O, and 93 3 kcal/mol for P−S,
respectively, in good agreement with the computed values. Additional reactivity with breaking of the P−P bond and formation of
diphosphinate O3P2(C6H10)2 was only observed to occur upon heating of dimethylsulfoxide solutions of the precursor. Reactivity
of diphosphane P2(C6H10)2 with azides allowed the isolation of monoiminophosphoranes (RN)P2(C6H10)2(R = Mes, CPh3,
SiMe3), and treatment with additional MesN3 yielded symmetric and unsymmetric diiminodiphosphoranes (RN)(MesN)-
P2(C6H10)2 (91% for R = Mes). Metalation reactions with the bulky diiminodiphosphorane ligand (MesN)2P2(C6H10)2 (nppn)
allowed for the isolation and characterization of (nppn)Mo(η3-C3H5)Cl(CO)2 (91%), (nppn)NiCl2 (76%), and [(nppn)Ni(η3-2-
C3H4Me)][OTf] showing that these ligands provide an attractive preorganized binding pocket for both late and early transition
metals.
with respect to the formation of bimetallic complexes.4 With
INTRODUCTION
■
the present work, we report our investigations on functionaliza-
tion reactions at the two phosphorus lone pairs of diphosphane
1 together with characterization data for the diphosphorane
(bis-λ5-phosphane) products.
Although the literature of metal complexes with phosphorane
ligands is extensive,5 research on complexes of diphosphoranes
containing P−P bonds (R2P(E)−P(E)R2) has been limited
almost exclusively to diphosphane disulfides (E = S, Chart 1).
In addition, while phosphine sulfides have been reported as
ligands for precursors in homogeneous catalysis,6,7 no such
reports exist for using diphosphorane ligands with an intact P−
P bond. These observations are quite surprising since such
ligands would chelate and presumably form very stable five-
Since the first report on the synthesis of multicyclic
tetraorganodiphosphane molecules by effective sequential
Diels−Alder addition of two 1,3-diene molecules to P2,1 our
group has uncovered a synthetically useful, one-step synthesis
of bicyclic diphosphanes from the co-photolysis of P4 (white
phosphorus) and 1,3-dienes.2 Following the discovery of a
streamlined route to synthesize the 3,4,8,9-tetramethyl-1,6-
diphosphabicyclo[4.4.0]deca-3,8-diene3 (P2(C6H10)2, 1, Figure
1), we have investigated the advantages that stem from having a
fixed dihedral angle between the two phosphorus lone pairs
Figure 1. 3,4,8,9-Tetramethyl-1,6-diphosphabicyclo[4.4.0]deca-3,8-
diene (diphosphane 1).
Received: April 27, 2013
Published: July 10, 2013
© 2013 American Chemical Society
8851
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
into multimetallic complexes (Chart 1, center). In addition,
they could allow for increased solubility in nonpolar solvents,
and thus potentially expand on the applications demonstrated
for less soluble diphosphane dichalcogenides. Iminophosphor-
anes are a good model for this behavior because the imine
group can provide steric bulk in the immediate vicinity of the
metal center. In contrast to α-diimines, which have been widely
used as ligands for homogeneous catalysis,19,20 diiminodiphos-
phorane analogues containing a P−P bond have not been
readily accessible, and only a few have been reported.21,22
Several cyclic diiminodiphosphoranes have been prepared from
Chart 1. Metal-Binding Modes Available to Diphosphorane
Ligands with P−P Bonds
membered {ME2P2} metallacycles, which ought to exhibit
enhanced stability toward decomposition compared to
monodentate phosphorane ligands.
t
iminophosphanes of the type RPNMes* (R = Me, Bu, Ph;
Mes* = 2,4,6-tBu3C6H2) via P−P coupling reactions,21 but they
exhibit trans imino substituents and are not suitable for metal
chelation. There is a single example of direct diiminodiphos-
phorane synthesis from acyclic diphosphanes (R2P−PR2, R =
Me, Et, n-Pr, n-Bu) via treatment with an inorganic azide
(Me3SiN3).22 Nonetheless, the reactivity of these diiminodi-
phosphoranes has only been investigated in relation to P−N
bond cleavage.23
The majority of the reported diphosphane disulfide
complexes involve mono- and divalent late transition metals,8
and a few involve metals from groups 6 and 7.9 Spectroscopic
evidence has also been presented for a diphosphane dioxide
chelating to iron(II) and zinc(II) centers (E = O, R = Ph),10 for
a dioxide bridging two molybdenum(VI) centers (E = O, R = n-
C3H7),11 and a diselenide chelating to a nickel(II) center (E =
Se, R = Ph).12 Several binding modes can be envisioned in
these complexes (Chart 1), yet unambiguous molecular
structures have been determined only for a few selected
examples. The first structure of a binuclear complex with
chelating diphosphane disulfide ligands (Chart 1, center) was
reported by Cotton et al. for copper(I) (R = Me).13 The
structures of mononuclear complexes (Chart 1, left) have only
been reported in four cases: copper(I) (R = Me), silver(I) (R =
Use of a bicyclic framework such as the diphosphane 1 ought
to allow the synthesis of diphosphoranes with substituents
locked in a cis configuration. We report here the synthesis of
diphosphane dichalcogenides and diiminodiphosphoranes with
a conserved bicyclic framework and with substituents that are
confined to an eclipsed conformation. In addition, we
demonstrate that the cis-diimine derivatives offer an attractive,
preorganized binding pocket that accommodates transition-
metal centers.
Me), rhodium(I) (R = Et), and palladium(II) (PR2
=
P(NEt2)(cyclohexyl)).14
A candidate explanation for the scarcity of examples wherein
diphosphoranes have been used for metal chelation is that they
are typically acyclic and thus allow for free rotation about the
P−P bond, rotation which in turn may lead to rearrangement
and formation of bridges between two metal centers instead of
robust metallacycles (Chart 1, right). For example, it was
shown that in the presence of copper(II), Me2P(S)−P(S)Me2
yields a bimetallic complex with chelating ligands in a Gauche
conformation (Chart 1, center),13 as well as a polymer with
disulfide ligands bridging between metal centers in an anti
configuration (Chart 1, right).15 Cyclic diphosphoranes are
much less common, but employing such frameworks would
prevent rotation about the P−P bond and lock the two sulfur
atoms in an eclipsed conformation. However, functionalization
of monocyclic diphosphanes has only allowed for the formation
of trans-disulfides, which are not suitable for metal chelation.16
Cis-diphosphane disulfides (Chart 2) have only been achieved
RESULTS AND DISCUSSION
■
Oxygen-Atom Transfer Reactions. Although methods for
preparing diphosphane monoxides are available, there are very
few examples that illustrate their synthesis directly from the
parent diphosphane. For example, the oxidation of commer-
cially accessible P2Ph4 to Ph2P−P(O)Ph2 has not been achieved
selectively.24 Only for cyclic diphosphanes with bulky
substituents has the oxidation been achieved selectively with
H2O2 or (Me3Si)2O2.25 Mono-oxidation of a 1,1′-biphosphole
was also achieved with (Me3Si)2O2 over a period of 7 days.26
We uncovered a new method for achieving a single oxidation of
a diphosphane selectively by employing the oxygen-atom
transfer (OAT) reagent mesitylnitrile-N-oxide (MesCNO;
Mes = 2,4,6-C6H2Me3).
In the presence of MesCNO, the diphosphane 1 acts as an
oxygen-atom acceptor at the phosphorus lone pairs. Dropwise
treatment of 1 with 1 equiv of MesCNO at room temperature
led to the selective formation of the diphosphane monoxide
OP2(C6H10)2 (1-O) within minutes (Scheme 1). The product
precipitated out of diethyl ether solutions as a white powder
and was isolated in 87% yield. In the solid state, monoxide 1-O
(Figure 2) exhibits a slight diminution of the P−P distance to
2.2034(5) Å as compared to a 2.2218(5) Å distance found in
diphosphane 1,2 consistent with an increased positive charge on
the phosphorus(IV) atom. Natural Bond Orbital (NBO)
analysis27 on 1-O did indeed reveal a partial charge of +1.52
on the phosphorus(IV) atom, while the charge on the
phosphorus(II) atom at +0.43 remains relatively unchanged
from that in diphosphane 1 (Table 3). The pure product
appears to be indefinitely stable in solution at room
temperature and it does not decompose even upon prolonged
heating (120 °C, toluene), as isomerization to a diphosphinite
via O-atom insertion into the P−P bond was predicted through
Chart 2. Diphosphane Dichalcogenides with a Locked cis
Configuration
by functionalizing bicyclic diphosphanes.17 Still, cis bicycles are
required for this conversion, yet there is only a relatively limited
number of reports of diphosphanes with such a locked cis
configuration.18
Additional groups near the chelation pocket may prevent
some of the degradation pathways that are accessible to
diphosphane dichalcogenide complexes, such as aggregation
8852
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
Scheme 1. Oxidation of the Diphosphane 1 with 1 and 2
equiv of MesCNO
density functional theory (DFT) methods to be unfavorable by
6.7 kcal/mol.
The enthalpy of the OAT reaction of MesCNO with
Figure 3. Relative enthalpies for the OAT reactions from 2,6-
Me2C6H3CNO to 1′ (bottom, blue) and to 1′-O (top, red),
respectively. R corresponds to a molecule of 2,6-Me2C6H3CNO and
either 1′ or 1′-O, while P corresponds to a molecule of 2,6-
Me2C6H3CN and either 1′-O or 1′-O2 set at infinite distances.
diphosphane 1 was measured experimentally at −82.4
2.0
kcal/mol by reaction calorimetry (eq 1), which corresponds to
a P−O bond dissociation enthalpy (BDE) in 1-O of 134.7
2.1 kcal/mol (BDE of N−O in MesCNO is 52.3 0.7 kcal/
mol28). The P−O BDE is between those for trialkyl and triaryl
phosphine oxides, as comparable measurements for phosphines
such as OPMe3, OPCy3, and OPPh3 yielded 138.5, 137.6, and
132.2 kcal/mol, respectively.28
94% isolated yield. In solution, the dioxide was stable even after
12 h in refluxing benzene and exhibited a C2v molecular
symmetry (1H NMR). In the solid state, the dioxide 1-O2 is
rigorously Cs-symmetric along the {OPPO} plane (Figure 2)
and adopts an exo-exo configuration which is in contrast to the
exo-endo conformations adopted by the monoxide 1-O and
unfunctionalized bicyclic diphosphanes.2
The enthalpy of this second OAT from MesCNO to the
diphosphane monoxide 1-O could not be measured exper-
imentally because of the slow reaction rate and the low
solubility of the product. It was computed, however, to be
−72.5 kcal/mol (eq 2). This corresponds to a P−O BDE of
124.8 kcal/mol, which is 6.2 kcal/mol lower than the computed
P−O BDE in 1-O. The mechanism of the formation of dioxide
1-O2 is similar to that observed for 1-O, as it involves a four-
membered ring intermediate and an activation barrier ΔH⧧ of
11.5 kcal/mol (Figure 3).
ΔH=−82.4 2.0 kcal/mol
(1)
1 + MesCNO ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 1‐O + MesCN
DFT computations were employed to corroborate the
experimental BDE value, and to understand the mechanism
of the oxidation reaction. The reaction enthalpy was computed
to be −78.7 kcal/mol, corresponding to a P−O BDE of 131.0
kcal/mol, in good agreement with the experimental value (eq
1). The simplified model framework 1,6-diphospha-bicyclo-
[4.4.0]deca-3,8-diene (1′, P2(C4H6)2, with H replacing the Me
groups in 1) and 2,6-Me2C6H3CNO were used for investigating
the mechanism of the OAT. A productive pathway was found
to involve a two-step process with an intermediate containing a
four-membered {PONC} ring similar to those observed for the
oxidation of PR3 phosphines (Figure 3).28 This mechanism is
similar to that of the Staudinger reaction between phosphines
and organic azides,29 and it involves little change in the P−P
distance and an activation barrier ΔH⧧ of 10.0 kcal/mol.
If the monoxide 1-O is allowed to form in a solvent in which
it is soluble, it undergoes further oxidation with MesCNO to
form the symmetric dioxide O2P2(C6H10)2 (1-O2, Scheme 1).
This second OAT is much slower and required stirring a
toluene solution of diphosphane 1 and 2 equiv of MesCNO for
24 h at room temperature to generate the dioxide 1-O2 for a
ΔH=−72.5 kcal/mol
1‐O + MesCNO ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 1‐O2 + MesCN
(2)
A subtle, yet important difference was observed between the
activation enthalpies of the first OAT step of 1′ with 2,6-
Me2C6H3CNO and that of the second step. The slight increase
in activation enthalpy from 10.0 to 11.5 kcal/mol corresponds
to a 12-fold decrease in reaction rate at room temperature from
the first to the second OAT. When removing the ortho groups
(using PhCNO), the activation enthalpies change to 10.2 and
Figure 2. Solid-state structures of the diphosphane oxides with ellipsoids at the 50% probability level and hydrogen atoms omitted for clarity.
Selected interatomic distances [Å] and angles [deg] for monoxide 1-O (left): P1−P2 2.2034(5), P1−O1 1.4959(10), O1−P1−P2 114.65(5); for
dioxide 1-O2 (center): P1−P2 2.2185(7), P1−O1 1.4934(16), P2−O2 1.4935(15), O1−P1−P2 113.88(6), O2−P2−P1 113.72(7), O1−P1−P2−
O2 0.0(0); for trioxide 1-O3 (right): P1−P2 2.9257(13), P1−O1 1.513(3), P2−O2 1.4792(18), P1−O3 1.5972(16), P2−O3 1.609(2), O1−P1−O3
109.10(17), O2−P2−O3 109.67(9), P1−O3−P2 131.73(10), O1−P1−P2−O2 1.0(3).
8853
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
10.8 kcal/mol, respectively. This implies that steric repulsions
of ortho-methyl groups with the already installed chalcogen
atom are responsible for the selectivity toward the singly
functionalized 1-O. Previously reported selective mono-
oxidations of diphosphanes25,26 are likely to involve some
type of steric repulsions, but here however, these are a result of
the bicyclic nature of the diphosphane framework 1 that
prevents the oxo group from rotating away from the ortho
groups of MesCNO.
Sulfur-Atom Transfer (SAT) Reactions. Using appro-
priate chalcogen transfer reagents, heavier analogues of the
oxides were also obtained. Treatment of diphosphane 1 with 1
equiv of SSbPh3 in benzene led to the formation of the
monosulfide SP2(C6H10)2 (1-S) as a precipitate, which could be
isolated pure in 56% yield. The enthalpy of the reaction was
also measured via calorimetry at −25.7 1.0 kcal/mol (eq 4),
which corresponds to a P−S BDE of 93 3 kcal/mol (BDE of
S−Sb in SSbPh3 is 67
3 kcal/mol32). The P−S BDE is
No additional reactivity of dioxide 1-O2 with excess
MesCNO could be observed, as oxidation to the P−P bond-
cleaved diphosphinate O3P2(C6H10)2 (1-O3) was only achieved
with boiling Me2SO solutions of 1-O2. Exposing solutions of
diphosphane 1 at room temperature to air however, led to the
formation of a mixture of all three oxides. The air stability of
various phosphines has been correlated to the energy level of
the singly occupied molecular orbital (SOMO) of their radical
cation.30 We computed that the radicals [1]•+ and [1-O]•+ have
SOMOs localized on the phosphorus atoms at −10.0 and
−10.2 eV, respectively. These values are close to the −10 eV
threshold above which phosphines have been predicted to be
air stable.30 A very slow air oxidation was observed for
diphosphane 1, consistent with the proposed threshold.
Crystals of the trioxide 1-O3 in an exo-endo configuration
(Figure 2) were obtained by aging a solution of 1 in air.
Notably, a minor disorder component consisting of the exo-
endo conformer of dioxide 1-O2 was observed in this crystal
(Supporting Information, Figure S.2), thus provides exper-
imental evidence for a very low energy difference between the
exo-exo and exo-endo isomers of derivatives of diphosphane 1.
Indeed, DFT calculations revealed that for diphosphane 1′, the
exo-endo isomer surmounts a barrier of only 5.9 kcal/mol with a
reaction enthalpy of just 2.5 kcal/mol for converting into the
exo-exo isomer (Figure 4). Solution 1H NMR spectra of 1 show
slightly larger than that measured for SPPh3 (88 kcal/mol), but
slightly smaller than that for SPMe3 (94 kcal/mol).32 DFT
computations predicted the P−S BDE to be 90 kcal/mol,
corroborating well the measured experimental value.
ΔH=−25.7 1.0 kcal/mol
1 + SSbPh3 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 1‐S + SbPh3
(4)
Addition of a second equivalent of SSbPh3 to solutions of 1
led to rapid generation of the disulfide S2P2(C6H10)2 (1-S2,
Scheme 2). Likely because of the less sterically demanding
Scheme 2. Synthesis of the Disulfide 1-S2
nature of SSbPh3 as compared with MesCNO, the second SAT
step occurs at a comparable rate with the formation of 1-S, thus
allowing for calorimetric measurements (eq 5). The reaction
enthalpy is only slightly higher compared to the first
sulfurization, implying a second BDE of 94
3 kcal/mol.
The disulfide 1-S2 was isolated in 87% yield following treatment
of 1 with 2 equiv of SSbPh3 for 2 h. Unlike the dioxide 1-O2,
the disulfide 1-S2 adopts an endo-exo configuration in the solid
state (Figure 5) with a small S−P−P−S angle of about 2°. The
Figure 4. Relative enthalpies for the interconversion of diphosphane 1′
(numbers in black) between the endo-exo and the exo-exo conformers
(eq 3). Very similar values were observed for the conversions of the
monoxide 1′-O (in red) and dioxide 1′-O2 (in blue) conformers.
Figure 5. Solid-state structure of the diphosphane disulfide 1-S2 with
ellipsoids at the 50% probability level and hydrogen atoms omitted for
clarity. Selected interatomic distances [Å] and angles [deg]: P1−P2
2.2192(4), P1−S1 1.9489(4), P2−S2 1.9495(4), S1−P1−P2
114.170(15), S2−P2−P1 114.823(16), S1−P1−P2−S2 1.96(2).
no changes at temperatures as low as −90 °C, and with such a
low calculated barrier, no changes are anticipated above −160
°C.31 Similar values were computed for the interconversion of
conformers of oxides 1′-O and 1′-O2 (Figure 4). This suggests
that the exo-exo configuration observed for the dioxide 1-O2 in
the solid state may be solely a result of favorable crystal packing
interactions. In addition, such small barriers are consistent with
fast exchange between conformers in solution (eq 3), which
monosulfide and disulfide exhibit similar NMR spectroscopic
features to those observed for the monoxide and the dioxide,
respectively (Table 1).
ΔH=−25.0 1.1 kcal/mol
1
1‐S + SSbPh3 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 1‐S2 + SbPh3
results in fast equilibration of the H NMR environments on
(5)
the bicyclic backbone of derivatives of diphosphane 1 and an
apparent σ-symmetry plane containing the P−P bond.
Nitrene-Group Transfer (NGT) Reactions. Similar to
other dichalcogenides that contain a P−P bond, the
dichalcogenides 1-E2 provide well-poised chelating pockets
(3)
exo‐endo ⇌ exo‐exo ⇌ endo‐exo
8854
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
Table 1. Spectroscopic and Structural Parameters of Phosphorane Derivatives of 1
a
a
b
c
d
compound
δλ p
1JP−P
P−P
E−P−P−E
3/5p
δλ
5
1-O
1-S
+75.9
+62.0
+30.7
+36.9
−87.4
−62.0
−97.9
−98.9
217
280
274
296
2.2034(5)
1-NMes
l-NCPh3
1-O2
+11.5
+20.4
−35.7
−30.9
b
2.2185(7)
2.2192(4)
2.2398(5)
2.2805(7)
0
1-S2
1.96(2)
5.57(10)
10.95(11)
1-(NMes)2
1-(NCPh3)(NMes)
−33.0
93
a
d
c
δ31 [ppm] in C6D6 (for 1:2 53.6 ppm). Experimental Jp−p [Hz] (for η1-1:4 344 Hz). Interatomic distances [Å] (for 1:2 2.2218(5) Å).
1
P
Dihedral angles [deg].
Scheme 3. Synthesis of Mono- and Diiminodiphosphoranes 1-NR and 1-(NR)(NMes)
Figure 6. Solid-state structure of the symmetric and the unsymmetric diiminodiphosphoranes 1-(NMes)2 (left and center) and 1-(NCPh3)(NMes)
(right) with ellipsoids at the 50% probability level and hydrogen atoms omitted for clarity. Selected interatomic distances [Å] and angles [deg] for 1-
(NMes)2: P1−P2 2.2398(5), P1−N1 1.5531(13), P2−N2 1.5459(14), N1−P1−P2 120.61(5), N2−P2−P1 120.70(6), C11−N1−P1 140.53(12),
C21−N2−P2 140.72(12), N1−P1−P2−N2 5.57(10); and for 1-(NCPh3)(NMes): P1−P2 2.2805(7), P1−N1 1.5561(15), P2−N2 1.5504(16),
N1−P1−P2 124.25(6), N2−P2−P1 114.39(6), C21−N2−P2 134.51(14), C31−N1−P1 131.48(13), N1−P1−P2−N2 10.95(11).
for accommodating various metal centers. However, because of
their bicyclic nature, they exhibit a rigid structure with small E−
P−P−E dihedral angles. Despite this attractive feature, the low
solubility of these molecules in low-dielectric organic media has
been an impediment in studying them as partners in metalation
reactions. In searching for more soluble variants, we decided to
target the iminophosphorane derivatives. The Staudinger
reaction of phosphines with organic azides has been shown
to have a similar mechanism29 to the one proposed between
phosphines and MesCNO.28 As a result, it was deemed logical
to similarly investigate the reactions of diphosphane 1 with RN3
azide molecules.
Treatment with MesN3 led to a vigorous evolution of
dinitrogen coupled with the formation of the monoiminophos-
phorane (MesN)P2(C6H10)2 (1-NMes). At room temperature,
the diphosphane 1 reacted to completion with 1 equiv of
MesN3 within 30 min (Scheme 3). Further conversion was
observed when additional MesN3 was present but, as with
MesCNO, this second step was significantly slower than the
initial formation of 1-NMes. Upon heating to 80 °C, the
consumption of monoiminophosphorane 1-NMes became fast
and was coupled with vigorous evolution of dinitrogen.
Refluxing the diphosphane 1 in benzene in the presence of 2
equiv of MesN3 led to the clean formation of diiminodiphos-
phorane (MesN)2P2(C6H10)2 (1-(NMes)2, nppn) after 12 h.
Upon cooling, crystals of analytically pure diiminodiphosphor-
ane formed, giving a 91% isolated yield.
The iminophosphoranes 1-NMes and 1-(NMes)2 present
similar features as observed for the chalcogenides 1-E and 1-E2,
respectively. In solution, they exhibit Cs and C2v molecular
symmetries, respectively (1H NMR), but in the solid state, the
diiminodiphosphorane 1-(NMes)2 exhibits a low symmetry
with distinct environments for each of its ten methyl groups
(Figure 6). Steric crowding from the mesityl groups led to a
8855
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
notable increase in P−P distance and N−P−P−N dihedral
angle compared to the dichalcogenides.
Table 2. Tolman Electronic Parameters for Mono-
Functionalized 1-E Derivatives Compared to PMe3, PPh3,
and PF3
Treatment of diphosphane 1 with more electron-rich azides
led to qualitatively slower reaction rates. In the particular case
of trityl azide, heating the reaction mixture in benzene at 70 °C
for 9 h was needed for clean, complete conversion to the
monoiminophosphorane (Ph3CN)P2(C6H10)2 (1-NCPh3). No
further reactivity was observed upon continued heating at 90
°C in the presence of excess Ph3CN3. However, in the presence
of MesN3, this monoiminophosphorane is converted to the
unsymmetric monomesityl, monotrityl diiminodiphosphorane
(MesN)(Ph3CN)P2(C6H10)2 (1-(NCPh3)(NMes)). In the
solid state, the P−P interatomic distance is slightly elongated
from that in 1-(NMes)2, while the N−P−P−N dihedral angle is
above 10° (Figure 6). Inorganic azides react even more slowly
with 1, as the reactions with Ph3SiN3 or Me3SiN3 required
additional heating after 9 h at 70 °C to complete the conversion
to the corresponding monoiminophosphorane products. No
reaction was observed between diphosphane 1 and Ph3SnN3
even after 36 h of heating at 90 °C.
Examples of monoiminophosphoranes with P−P bonds are
plentiful in the literature, yet there has been only one report
describing the synthesis of iminophosphoranes directly from
their parent R2P−PR2 diphosphanes (R = Me, Et, n-Pr, n-Bu).22
The reaction of acyclic diphosphanes with Me3SiN3 neces-
sitated heating of the neat reagents to 100 °C to produce
mixtures of both monoimino- and diimino-functionalized
products. Control over selectivity could only be achieved by
varying the number of azide equivalents used, and isolation of
iminophosphorane-phosphanes and diiminodiphosphoranes
necessitated separation via distillation.22 As reported above,
use of diphosphane 1, however, allowed for complete selectivity
control without requiring tedious purification. In fact, except for
the aryl azide MesN3, no azide underwent a second Staudinger
reaction at the remaining proximal phosphorus atom under the
tested conditions.
a
b
compound
vCO(A1)(cm−l)
ΔνPMe (cm−l)
χP(E)R2 (cm−1
)
3
1
2042.9
2045.6
2051.9
2052.9
2052.2
2050.3
−1.7
1.0
4.0
1′
l-O
6.7
7.3
13.0
14.0
13.3
11.4
1-S
8.3
1-Se
1-NMes
7.6
5.7
c
PMe3
2044.6(2064.1)
2046.3(2068.9)
2090.1(2110.8)
0
1.7
2.6(2.6)
c
PPh3
3.3(4.3)
c
PF3
45.5
b
17.9(18.2)
a
ΔνPMe = νCO(1-E) − νCO(PMe3). χP(E)R = νCO(A1)E − 2036.5 − 2
3
2
× χiBu, where 1-E have been approximated to (iBu)2P−P(E)(iBu)2 (χi
Bu
= 1.2 cm−1).33 Experimental values are shown inside parentheses.33
c
from a methyl to a fluorine substituent.33 This is in contrast to
the minimal change in the NBO s-orbital character (Table 3),
indicating a significant inductive effect with little change in
geometry at the phosphorus atom.
Table 3. NBO Analysis for Mono-Functionalized 1-E
Derivatives Compared to PMe3, PPh3, and PF3
natural charge on
compound s-orbital character of P lone-pair (%) λ3-P
λ5-P
E
1
59.8
60.8
61.3
61.3
61.2
61.5
54.4
49.4
75.2
0.46
0.45
0.43
0.49
0.50
0.44
0.70
0.75
1.59
1′
1-O
1.52
0.94
0.84
1.40
−1.00
−0.54
−0.47
−0.96
1-S
1-Se
1-NMes
PMe3
PPh3
PF3
Such selectivity for monofunctionalization of diphosphane 1
allows facile access to 1-E molecules. These are potential
ligands for binding metal centers in either an κ1 fashion or a
κ2P,E mode with the formation of four-membered {MPPE}
metallacycles. The potential of these monophosphoranes as
ligands has been investigated in silico for the κ1P binding mode.
Using Ni(CO)3 as a metal fragment (Figure 7), the Tolman
Metal Complexes of Diiminodiphosphoranes. Mole-
cules containing two iminophosphorane groups such as
CHx(Ph2PNR)2 have been used as ligands in many metal-
catalyzed reactions.34 In contrast, complexes that incorporate
diiminodiphosphorane ligands and that contain a P−P bond
have not yet been reported. This is quite surprising when
considering that the latter class of ligands could prove to be
valuable counterparts to the α-diimines that have been
employed as ligands for a wide range of applications in
transition-metal catalysis.19,20 The bicyclic diiminodiphosphor-
anes reported here are superficially similar to α-diimine ligands
and offer a restricted rotation about the P−P bond. The
diiminodiphosphorane 1-(NMes)2 (nppn) framework was
considered to constitute an attractive bidentate ligand for
applications in transition-metal chemistry, and metalation
reactions were accordingly carried out with both early and
late transition-metal sources.
Figure 7. NBO analysis of 1-NMes shows a well-defined phosphorus
lone pair (left, in blue) which can be used for metal binding such as in
(OC)3Ni((1-NMes)-κ1P) (right, DFT-optimized structure).
electronic parameter can be extracted from the totally
symmetric stretch of the Ni(CO)3 unit (Table 2).33 The
contribution χi from the λ5-phosphorus substituent (P(=E)R2)
was computed to be 7 to 10 cm−1 higher than in the
unfunctionalized diphosphane 1. For comparison, this sub-
stituent contribution increases by around 15 cm−1 upon going
Molybdenum(II) diimine complexes have been shown to be
good catalyst precursors for epoxidation of alkenes,35 leading us
to target diiminodiphosphorane analogues. Treatment of
36
molybdenum(II) source MoCl(CO)2(η3-C3H5)(NCMe)2
with nppn in dichloromethane (Scheme 4), allowed for the
isolation of the (nppn)MoCl(CO)2(η3-C3H5) complex (2) as a
8856
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
diiminodiphosphorane complexes and isolated the deep-
turquoise, paramagnetic nickel(II) complex (nppn)NiCl2 (3)
in 76% yield from the treatment of nppn with the NiCl2(dme)
precursor (Scheme 5, dme = 1,2-dimethoxyethane). The solid-
state structure of the paramagnetic complex 3 reveals a pseudo-
tetrahedral geometry around the nickel(II) center (Figure 8).
As with complex 2, the metallacycle exhibits an envelope
conformation with almost coplanar P−N bonds (N−P−P−N:
4.91(14)°; P−P−N−Ni: 16.84(14)° and 24.47(14)°) and
contains almost equal Ni−N distances, with a shortened P−P
but elongated P−N distances when compared to free nppn. A
square-planar nickel(II) complex was also obtained by treating
half of an equivalent of the [NiCl(η3-2-C3H4Me)]2 dimer in
dichloromethane with AgOTf in the presence of nppn (Scheme
5, OTf = CF3SO3). The [(nppn)Ni(η3-2-C3H4Me)][OTf]
product (4) was crystallized from the filtrate as a golden
powder, in 59% yield. The nickel complex 4 exhibits
diamagnetism characteristic of square-planar nickel(II) centers
with a low-spin d8 electronic configuration. Although it shows a
sharp 31P NMR singlet resonance (+25 ppm), the complex
Scheme 4. Synthesis of Molybdenum(II) Complex
(nppn)MoCl(CO)2(η3-C3H5)
pale yellow powder in 91% yield. NMR spectroscopy revealed
broad features indicative of a low-symmetry complex engaging
in a slow exchange process, and two broad 31P NMR
resonances near +6 ppm (1JP−P = 180 Hz) that imply two
different substituents trans to the nitrogen atoms of the
chelating nppn ligand. Indeed, the solid-state structure obtained
for the complex 2 reveals a pseudo-octahedral geometry around
molybdenum, with a carbonyl and an allyl group occupying the
positions trans to the nppn chelate (Figure 8). The N−Mo−N
bite angle of 81.54(7)° is similar to the angle in molybdenum-
(II) α-diimine complexes.35 The metallacycle exhibits an
envelope conformation, with almost coplanar P−N bonds
(N−P−P−N: 1.28(11)°; P−P−N−Mo: 21.51(12)° and
25.11(12)°). The longer Mo−N bond observed trans to the
CO ligand is consistent with the strong structural trans
influence of CO.37 Compared to the free nppn, chelation led
to a slight elongation of the P−N distances by about 0.05 Å,
coupled with significant shortening of the P−P distance to
2.1525(9) Å. This distance is almost 0.1 Å shorter than in free
nppn and is the shortest P−P distance of all the compounds
reported herein, this being consistent with the electron-
withdrawing effect of a Lewis-acidic metal center.
1
manifests two distinct H NMR environments for the ortho-
mesityl groups and two for the methyl groups on the bicyclic
backbone. This pseudo-Cs symmetry (with a σ plane through Ni
and the middle of the P−P bond) is indicative of a crowded,
rigid environment around the metal center, where the mesityl
groups are locked on one side of the {NiN2} plane.
CONCLUDING REMARKS
■
The synthesis of diphosphoranes was achieved directly from
diphosphane 1 by taking advantage of the rigidity inherent to
its bicyclic nature. The pronounced stability of 1 toward P−P
bond breaking reactions has allowed for selective functionaliza-
tion reactions which are not available for acyclic diphosphanes.
Reactivity of 1 with the OAT reagent MesCNO has been
investigated in detail using computational and calorimetric
techniques. Sequential OAT reactions allowed for clean
formation of a monoxide and a dioxide; further oxidation
involving the cleavage of the P−P bond is much slower.
Nickel(II) α-diimine complexes are desirable synthetic
targets as many have been shown to exhibit high activities
toward olefin polymerization.20 We targeted analogous
Figure 8. Solid-state structures of the molybdenum(II) 2 and nickel(II) 3 nppn complexes with ellipsoids at the 50% probability level; solvent
molecules and hydrogen atoms were omitted for clarity. Selected interatomic distances [Å] and angles [deg] for 2: P1−P2 2.1525(9), Mo1−N1
2.317(2), Mo1−N2 2.244(2), Mo1−C1 1.918(3), Mo1−C2 1.949(3), Mo1−Cl1 2.6213(7), P1−N1 1.600(2), P2−N2 1.605(2), N2−Mo1−N1
81.54(7), P1−N1−Mo1 117.46(10), P2−N2−Mo1 120.17(11), N1−P1−P2 105.98(8), N2−P2−P1 103.91(8), C4−Mo1−N2 165.37(9), C2−
Mo1−N1 169.23(10), C1−Mo1−Cl1 173.12(10), N1−P1−P2−N2 1.28(11), P2−P1−N1−Mo 21.51(12), P1−P2−N2−Mo 25.11(12); for 3: P1−
P2 2.1676(12), Ni1−N2 2.011(3), Ni1−N1 2.019(3), P1−N1 1.586(3), P2−N2 1.585(3), N2−Ni1−N1 93.88(11), P1−N1−Ni1 113.02(14), P2−
N2−Ni1 114.53(16), N1−P1−P2 104.16(12), N2−P2−P1 104.10(11), Cl1−Ni1−Cl2 122.58(4), N1−P1−P2−N2 4.91(14), P2−P1−N1−Ni
16.84(14), P1−P2−N2−Ni 24.47(14).
8857
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
Scheme 5. Syntheses of Nickel(II) Complexes (nppn)NiCl2 and [(nppn)Ni(η3-2-C3H4Me)][OTf]
1.5 Hz, OPCH2CH0), 124.7 (dd, 2JCP = 9 Hz, 3JCP = 4 Hz, PCH2CH0),
36.7 (d, 1JCP = 41.8 Hz, OPCH2), 28.9 (dd, 1JCP = 26.5 Hz, 2JCP = 4.5
Phosphine sulfides were isolated from treatment of diphos-
phane 1 with SSbPh3, while treatment of 1 with organic azides
led to formation of mono- and diiminodiphosphoranes via
sequential Staudinger reactions. Functionalization at the second
phosphorus atom is significantly slower than for the first one,
permitting the installation of mixed functional groups on the
phosphorus atoms. Metalation reactions with diiminodiphos-
phoranes have shown that these ligands provide an attractive
preorganized binding pocket for both late and early transition
metals. All these chemical transformations have demonstrated
that a stable and versatile platform is provided by the
diphosphane backbone 1.
Hz, PCH2), 21.4 (d, 3JCP = 1.0 Hz, CH3), 21.2 (d, 3JCP = 3.2 Hz, CH3)
1
ppm. 31P{1H} NMR (C6D6, 20 °C, 161.9 MHz) δ: +75.9 (d, JPP
=
217 Hz, 1 P, OPP), −87.4 (d, 1 P, OPP) ppm. 31P NMR (CDCl3, 20
°C, 121.5 MHz) δ: +82.1 (d), −84.9 (d). Elemental analysis [%]
found (and calcd. for C12H20P2O): C 59.37 (59.50); H 8.17 (8.32); N
none (0.00).
Synthesis of Dioxide O2P2(C6H10)2 (1-O2). A solution of
MesCNO (123 mg, 0.76 mmol, 2.0 equiv) in toluene (2 mL) was
added dropwise over 3 min to a solution of 1 (86 mg, 0.38 mmol, 1.0
equiv) in toluene (3 mL). The mixture became cloudy within minutes,
but was allowed to stir for 24 h. After diluting with Et2O (5 mL), the
white precipitate was collected on a fritted glass filter, washed with
Et2O (10 mL), and dried under reduced pressure. A white powder
consisting of pure O2P2(C6H10)2 (92 mg, 0.36 mmol, 94% yield) was
EXPERIMENTAL DETAILS
■
collected. 1H NMR (DMSO-d6, 20 °C, 400 MHz) δ: 3.03 (dd, 2JHH
=
General Synthetic Details. Unless otherwise stated, all
manipulations were performed in a Vacuum Atmospheres model
MO-40 M glovebox under an inert atmosphere of purified N2.
Solvents were obtained anhydrous and oxygen-free by sparging with
N2 and purifying using a Glass Contours Solvent Purification System
(SG Water). Deuterated solvents were purchased from Cambridge
Isotope Laboratories, degassed, and stored over molecular sieves prior
to use. All glassware was oven-dried at temperatures greater than 170
°C prior to use. Celite 435 (EM Science) was dried by heating above
200 °C under a dynamic vacuum for at least 48 h prior to use, while
alumina (activated, basic, Brockman I) was dried similarly for at least 7
days prior to use. NMR spectra were obtained on Varian Mercury 300,
Varian Inova 500, or Bruker Avance 400 instruments equipped with
Oxford Instruments superconducting magnets, and were referenced to
the appropriate solvent resonances.38 31P NMR spectra were
referenced externally to 85% H3PO4 (0 ppm). The following
abbreviations were used to explain the multiplicities: s = singlet, d =
doublet, t = triplet, q = quartet, m = multiplet, br = broad, v = virtual
multiplet. Elemental analyses were performed by Midwest Microlab
LLC, Indianapolis, IN. White phosphorus was acquired from
Thermphos International. Diphosphane 1,4 MesCNO,39 MesN3,40
Ph3EN3,41 and [Mo(CO)2Cl(η3-C3H5)(NCMe)2]36 were prepared
and isolated according to literature protocols, while the other reagents
were acquired from commercial sources.
14.6 Hz, 2JHP = 7.6 Hz, 4 H, OPCHH), 2.68 (d, 4 H, OPCHH), 1.74
(s, 12 H, CH3) ppm. 13C{1H} NMR (DMSO-d6, 20 °C, 100.6 MHz)
δ: 123.8 (s, CH0), 34.8 (vt, 12JCP = 27.7 Hz, CH2), 21.4 (s, CH3) ppm.
31P{1H} NMR (DMSO-d6, 20 °C, 121.5 MHz) δ: +16.2 (s) ppm.
31P{1H} NMR (C6D6, 20 °C, 121.5 MHz) δ: +11.5 (s) ppm. 31P NMR
(CDCl3, 20 °C, 121.5 MHz) δ: +14.1 (s) ppm. Elemental analysis [%]
found (and calcd. for C12H20P2O2): C 56.40 (55.81); H 8.59 (7.81); N
none (0.00).
Generation of Trioxide O3P2(C6H10)2 (1-O3). A solution of
monoxide 1-O (10 mg, 0.76 mmol, 2.0 equiv) in DMSO-d6 (1 mL)
inside an NMR tube was heated via a heat-gun for 5 min until the
solvent started boiling, after which, NMR spectroscopic analysis
1
showed the presence of only trioxide 1-O3. H NMR (CDCl3, 20 °C,
1
400 MHz) δ: 2.93 (br s, 8 H, CH2), 1.87 (s, 12 H, CH3) ppm. H
NMR (DMSO-d6, 20 °C, 400 MHz) δ: 2.85 (vbr s, 8 H, CH2), 1.77 (s,
12 H, CH3) ppm. 13C NMR (CDCl3, 20 °C, 100.6 MHz) δ: 125.5 (s,
12
CH0), 38.8 (vt,
J
= 19.5 Hz, CH2), 22.3 (s, CH3) ppm. 31P{1H}
CP
NMR (CDCl3, 20 °C, 162 MHz) δ: +22.1 (s) ppm. 31P{1H} NMR
(DMSO-d6, 20 °C, 162 MHz) δ: +12 (s) ppm.
Synthesis of Monosulfide SP2(C6H10)2 (1-S). To a solution of 1
(94 mg, 0.41 mmol, 1.0 equiv) in Et2O (5 mL) a suspension of SSbPh3
(160 mg, 0.41 mmol, 1.0 equiv) in Et2O (5 mL) was added dropwise.
After 1 h of vigorous stirring, the suspension was filtered and washed
with Et2O (10 mL), and the white powder was dried under reduced
Synthesis of Monoxide OP2(C6H10)2 (1-O). A solution of
MesCNO (71 mg, 0.44 mmol, 1.0 equiv) in Et2O (3 mL) was
added dropwise, over a period of 5 min, to a solution of 1 (100 mg,
0.44 mmol, 1.0 equiv) in Et2O (6 mL). White precipitate started
forming halfway through the addition; the mixture was stirred for
another 5 min after the addition was complete. The final mixture was
diluted with pentane (10 mL), and the white solids were collected on a
fritted glass filter, washed with pentane (10 mL), and dried under
reduced pressure. A white powder consisting of pure OP2(C6H10)2 was
obtained (93 mg, 0.38 mmol, 87% yield). 1H NMR (C6D6, 20 °C, 400
1
pressure to yield 1-S (60 mg, 0.23 mmol, 56%). H NMR (C6D6, 20
°C, 400 MHz) δ: 2.35 (d, 2JHP = 13.5 Hz, 2 H, SPCHH), 2.05 (m, 2 H,
SPCHH), 1.89 (m, 2 H, PCHH), 1.75 (m, 2 H, PCHH), 1.57 (s, 6 H,
PCH2CCH3), 1.38 (s, 3 H, SPCH2CCH3), 1.36 (s, 3 H, SPCH2CCH3)
ppm. 13C{1H} NMR (C6D6, 20 °C, 100.6 MHz) δ: 128.8, 125.7, 39.9,
31.5, 22.9, 22.7 ppm. 31P{1H} NMR (C6D6, 20 °C, 121.5 MHz) δ:
1
+62.0 (d, JPP = 280 Hz, 1 P, SPP), −62.0 (d, 1 P, SPP) ppm. 31P
1
NMR (CDCl3, 20 °C, 162 MHz) δ: +62.6 (d, JPP = 277 Hz, 1 P,
SPP), −59.1 (d, 1 P, SPP) ppm. Elemental analysis [%] found (and
calcd. for C12H20P2S): C 55.59 (55.80); H 7.96 (7.80); N none (0.00).
Synthesis of Disulfide S2P2(C6H10)2 (1-S2). A 1:1 toluene−THF
solution (8 mL) of SSbPh3 (321 mg, 0.83 mmol, 2.0 equiv) was added
to a toluene (2 mL) solution of 1 (94 mg, 0.41 mmol, 1.0 equiv). After
5 min of vigorous stirring, a white precipitate started forming. After 2 h
of stirring, Et2O (5 mL) was added to precipitate all of the product. A
white powder consisting of analytically pure S2P2(C6H10)2 (105 mg,
2
2
MHz) δ: 2.37 (dd, JHP = 18.2 Hz, JHH = 15.0 Hz, 2 H, OPCHH),
2.14 (pst, 12.2 Hz, 2 H, OPCHH), 2.02 (psq, 14.4 Hz, 2 H, PCHH),
2
2
1.76 (dd, JHP = 22.7 Hz, JHH = 14.0 Hz, 2 H, PCHH), 1.54 (s, 6 H,
PCH2CCH3), 1.39 (s, 3 H, OPCH2CCH3), 1.37 (s, 3 H,
OPCH2CCH3) ppm; assignments were made based on the 1H-1H
COSY NMR spectrum (Supporting Information, Figure S.1). 13C{1H}
2
3
NMR (C6D6, 20 °C, 100.6 MHz) δ: 127.1 (dd, JCP = 12 Hz, JCP
=
8858
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
0.36 mmol, 87%) was collected by filtering on a fritted glass filter,
washing with Et2O (10 mL) and drying under reduced pressure. H
to yield turquoise solids consisting of pure paramagnetic complex
(nppn)NiCl2 (77 mg, 0.12 mmol, 76% yield). H NMR (CDCl3, 20
1
1
NMR (CDCl3, 20 °C, 400 MHz) δ: 2.93 (br s, 8 H, CH2), 1.87 (s, 12
°C, 400.1 MHz) δ: 42.5 (s, 6 H, p-Mes), 28.5 (br s, 12 H, o-Mes), 25.4
(s, 4 H, CHH), 22.5 (s, 4 H, CHH), 5.83 (br s, 4 H, m-Mes), 0.39 (s,
12 H, CH3) ppm. Elemental analysis [%] found (and calcd. for
NiCl2C30H42P2N2): C 58.23 (57.91); H 6.86 (6.80); N 4.64 (4.50).
Synthesis of Nickel(II) Complex [(nppn)Ni(η3-2-C3H4Me)]-
[OTf] (4). A solution of diiminodiphosphorane (98 mg, 0.20 mmol, 1.0
equiv) in dichloromethane (5 mL) was added to a red suspension of
[NiCl(2-C3H4Me)]2 (30 mg, 0.10 mmol, 0.5 equiv) in dichloro-
methane (2 mL). AgOTf (51 mg, 0.20 mmol, 1.0 equiv) was added to
this mixture as a slurry in dichloromethane (5 mL) leading to a color
change from red to yellow-brown. After stirring for 1 h, the mixture
was filtered though a pad of Celite (to remove the AgCl byproduct),
and the solids were washed with dichloromethane until a colorless
filtrate was obtained. The yellow filtrate was concentrated (to 5 mL),
then pentane (1 mL) and diethyl ether (5 mL) were added. The
resulting mixture was stirred for 1 min after which the golden solids
were collected on a fritted glass filter and dried under reduced pressure
to yield pure [(nppn)Ni(η3-2-C3H4Me)][OTf] complex (89 mg, 0.12
mmol, 59% yield). 1H NMR (CDCl3, 20 °C, 400.1 MHz) δ: 6.87 (s, 2
H, m-Mes), 6.85 (s, 2 H, m-Mes), 3.44 (d, 2JHH = 16 Hz, 2 H, PCHH),
3.27 (d, 2 H, PCHH), 2.86 (s, 4 H, PCHH), 2.66 (s, 6 H, o-Mes), 2.40
(s, 6 H, o-Mes), 2.24 (s, 6 H, p-Mes), 2.13 (s, 3 H, CH3allyl), 1.86 (s, 6
H, CH3), 1.81 (s, 6 H, CH3), 1.59 (s, 2 H, CH2allyl), 1.39 (s, 2 H,
H, CH3) ppm. 13C NMR (CDCl3, 20 °C, 100.6 MHz) δ: 125.5 (s,
12
CH0), 38.8 (vt,
J
= 19.5 Hz, CH2), 22.3 (s, CH3) ppm. 31P{1H}
CP
NMR (C6D6, 20 °C, 121.5 MHz) δ: +20.4 (s) ppm. 31P{1H} NMR
(CDCl3, 20 °C, 162 MHz) δ: +22.1 (s) ppm. Elemental analysis [%]
found (and calcd. for C12H20P2S2): C 49.59 (49.64); H 6.96 (6.94); N
none (0.00).
Synthesis of Diiminodiphosphorane (MesN)2P2(C6H10)2 (1-
(NMes)2, nppn). A toluene (5 mL) solution of MesN3 (1.416 g, 8.78
mmol, 2.0 equiv) was added dropwise to a stirring toluene (5 mL)
solution of 1 (0.993 g, 4.39 mmol, 1.0 equiv) inside a 50 mL Schlenk
flask. The reaction mixture was allowed to stir for 15 min to allow N2
evolution before sealing the flask containing the mixture and bringing
it outside of the glovebox. The flask was then connected to a reflux
condenser and N2 bubbler, and the mixture was heated to reflux
overnight. After cooling, the mixture was brought back into the
glovebox, and concentrated by 50%. Et2O (5 mL) and hexane (10 mL)
were added to further precipitate the product, which was collected on
a fritted glass filter, washed with hexane (30 mL), and dried under
reduced pressure. The white crystalline solids consisting of analytically
pure 1-(MesN)2 were collected (1.946 g, 3.85 mmol, 90% yield).
Concentrating the filtrate and collecting the resulting solids yielded an
1
additional amount of product (28 mg, 91% total yield). H NMR
CH2allyl) ppm. H NMR (C6D6, 20 °C, 500 MHz) δ: 6.86 (s, 2 H, m-
1
(CDCl3, 20 °C, 400.1 MHz) δ: 6.89 (s, 4 H, m-Mes), 2.65 (br dd, 2JHH
= 14 Hz, 4 H, CHH), 2.38 (s, 12 H, o-Mes), 2.29 (s, 6 H, p-Mes), 2.08
(br d, 4 H, CHH), 1.39 (s, 12 H, CH3) ppm. 13C{1H} NMR (C6D6, 20
°C, 100.6 MHz) δ: 131.7 (vt, JCP = 6.5 Hz, i-Mes), 129.1 (m-Mes),
129.0 (o-Mes), 127.8 (p-Mes), 124.8 (CC), 38.4 (vtr, JCP = 29 Hz,
CH2), 21.6 (o-Mes), 21.4 (p-Mes), 21.3 (H3C-CC) ppm. 31P{1H}
NMR (C6D6, 20 °C, 121.5 MHz) δ: −35.7 (s) ppm. Elemental
analysis [%] found (and calcd. for C30H42P2N2): C 73.06 (73.14); H
8.46 (8.59); N 5.60 (5.69); P 12.70 (12.58). Allowing the crystals to
age in air for 45 days at room temperature led to less than 10%
Mes), 6.84 (s, 2 H, m-Mes), 3.26 (d, 16 Hz, 2 H, CH2), 2.86 (d, 2 H,
CH2), 2.69 (s, 6 H, o-Mes), 2.44 (m, 4 H, CH2), 2.43 (s, 6 H, o-Mes),
2.19 (s, 6 H, p-Mes), 1.98 (s, 3 H, CH3allyl), 1.75 (s, 1 H, CH2allyl), 1.55
(s, 12 H, CH3), 1.41 (s, 1 H, CH2allyl). ppm 13C{1H} NMR (C6D6, 20
°C, 125.8 MHz) δ: 141.8 (i-Mes), 135.0 (o-Mes), 133.8 (o-Mes),
133.4 (p-Mes), 129.8(m-Mes), 129.6 (m-Mes), 124.7 (CC), 123.8
(CF3), ∼83 (vbr, CH0allyl), 54.8 (CH2allyl), 32.9 (br d, JCP = 18 Hz,
CH2), 22.9 (o-CH3), 22.8 (CHa3llyl), 22.7 (o-CH3), 20.8 (H3CC
CCH3), 20.7(p-CH3), 20.5 (H3CCCCH3) ppm. 31P{1H} NMR
(CDCl3, 20 °C, 162.0 MHz) δ: +25.2 (s) ppm. 19F NMR (CDCl3, 20
°C, 282.4 MHz) δ: −78.6 (s) ppm. Elemental analysis [%] found (and
calcd. for NiC35H49P2SN2O3F3): C 55.19 (55.64); H 5.91 (6.54); N
3.61 (3.71).
Calorimetric Measurements. In the glovebox, a solution of
0.0808 g of 1 (0.36 mmol) was dissolved in 6 mL of C6D6, and 1 mL
of this stock solution was loaded into an NMR tube. The remaining 5
mL of solution were loaded into the Calvet calorimeter cell with
MesCNO (0.0189 g, 0.1172 mmol) as limiting reagent. The
calorimeter cell was sealed, taken from the glovebox, and loaded
into the Setaram C-80 calorimeter. Following temperature equilibra-
tion, the reaction was initiated and the calorimeter rotated to achieve
mixing. Following return to baseline the calorimeter cell was taken into
the glovebox, opened, and 1 mL of the solution loaded into an NMR
tube. 31P NMR spectra of both the stock solution and the calorimetry
solution were then acquired, and the reaction was confirmed as
quantitative. The sole product was 1-O and no detectable amount of
1-O2 was detected. The enthalpy of three measurements done in this
way led to ΔH = −78.1 1.9 kcal/mol based on the reaction
1
decomposition (as assayed by H NMR spectroscopy).
Synthesis of Molybdenum(II) Complex (nppn)MoCl(CO)2(η3-
C3H5) (2). A dichloromethane solution (5 mL) of diiminodiphosphor-
ane 1-(NMes)2 (84 mg, 0.17 mmol, 1.0 equiv) was added to a
dichloromethane solution (5 mL) of MoCl(CO)2(η3-C3H5)(MeCN)2
(53 mg, 0.17 mmol, 1.0 equiv). After 30 min of stirring, a bright yellow
solid had started precipitating from the solution. After another 30 min,
the mixture was layered with pentane (2 mL) and stored at −35 °C
overnight to yield bright yellow solids, which were collected on a
fritted glass filter. After washing with Et2O (10 mL) and drying under
reduced pressure, pale yellow solids were isolated consisting of
analytically pure (nppn)MoCl(CO)2(η3-C3H5) complex (112 mg,
1
0.155 mmol, 91% yield). H NMR (CDCl3, 20 °C, 400.1 MHz) δ:
6.99 (br s, 4 H, m-Mes), 4.83 (t, 2 H, CHHallyl), 3.64 (m, 1 H, CHallyl),
2.69 (dd, 2 H, PCHH), 2.8−2.4 (vbrs, 14 H, o-Mes, PCHH), 2.31 (s, 6
H, p-Mes), 2.06 (d, 2 H, PCHH), 1.77 (s, 6 H, CH3), 1.66 (s, 6 H,
CH3), 0.62 (br s, 2 H, CHHallyl) ppm. 13C{1H} NMR (CDCl3, 20 °C,
125.8 MHz) δ: 227.8 (CO), 135−137 (vbr, o-Mes), 133.6 (i-Mes),
129.9 (m-Mes), 129.8 (o-Mes), 127.3 (p-Mes), 122.8 (CC), 73.5
(CHallyl), ∼63 (vbr, CHa2llyl), 29.3 (br d, CH2P), 24.3 (br, o-CH3), 23.2
(o-CH3), 21.0 (p-CH3), 20.7 (br, H3CCCCH3), 20.0 (H3CC
CCH3) ppm. 31P{1H} NMR (CDCl3, 20 °C, 162.0 MHz) δ: +7.1 (br
d, 1J = 180 Hz), +5.6 (br d) ppm. Elemental analysis [%] found (and
calcd. for MoC35H47ClP2N2O2): C 57.33 (58.29); H 6.03 (6.57); N
3.62 (3.88).
1(tol. sol.) + MesCNO(solid)
ΔH
⎯⎯→ 1‐O(tol. sol.) + MesCN(tol. sol.)
(6)
To these data the enthalpy of solution of MesCNO in toluene (+4.3
0.1 kcal/mol) was subtracted to give ΔHrxn = −82.4 2.0 kcal/mol
with all species in toluene solution. Because of the long reaction time
of the second OAT reaction coupled with the low solubility of the
product, the enthalpy of formation of 1-O2 could not be measured.
Reactions of SSbPh3 and 1 were performed in a similar manner. In
contrast to reaction with MesCNO which showed with limiting
MesCNO that only the monosubstituted product was formed, even
with SSbPh3 as limiting reagent small but detectable levels of the
disubstituted product 1-S2 were identified. The average value for the
first substitution (which included traces of disubstituted product) was
−20.4 0.5 kcal/mol. Because of its apparently greater solubility in
Synthesis of Nickel(II) Complex (nppn)NiCl2 (3). A suspension
of diiminodiphosphorane 1-(NMes)2 (79 mg, 0.16 mmol, 1.0 equiv) in
dichloromethane (5 mL) was added to a yellow suspension of
NiCl2(dme) (35 mg, 0.16 mmol, 1.0 equiv) in dichloromethane (5
mL). After 15 min, the color had changed slowly from yellow to an
intense turquoise color. After 1 h of stirring, the solution was filtered
though a glass paper plug to yield a deep blue filtrate and yellow solids.
The filtrate was concentrated under reduced pressure (to ca. 1 mL)
and then diluted with Et2O (10 mL). The resulting blue solids were
collected on a fritted glass filter, washed with Et2O (5 mL), and dried
8859
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
Table 4. Crystallographic Data for Diphosphane Oxides 1-O, 1-O2, and 1-O3
1-O
1-O2
1-O3
reciprocal net code/CCDC no.
empirical formula, FW (g/mol)
crystal size (mm3)
D8_11005/898603
C12H20OP2, 242.22
0.40 × 0.30 × 0.20
100(2)
D8_11006/898604
C12H20O2P2, 258.22
0.30 × 0.20 × 0.10
100(2)
X8_10087/898605
C12H20O3P2, 274.22
0.12 × 0.10 × 0.05
100(2)
temperature (K)
wavelength (A)
1.54178
1.54178
0.71073
crystal system, space group
unit cell dimensions (Å, deg)
monoclinic, P21/c
a = 10.8769(2), α = 90
b = 6.92120(10), β = 104.9620(10)
c = 17.2230(3), γ = 90
1252.61(4)
orthorhombic, Pnma
a = 9.59800(10), α = 90
b = 20.1851(3), β = 90
c = 6.58950(10), γ = 90
1276.63(3)
trigonal, P1
̅
a = 7.6898(6), α = 93.956(2)
b = 8.1011(6), β = 98.939(2)
c = 10.7689(8), γ = 101.833(2)
645.09(8)
volume (A3)
Z
4
4
2
density (calc, g/cm3)
1.284
1.343
1.412
absorption coefficient (mm−1
)
2.924
2.964
0.331
F(000)
520
552
292
θ range for data collection (deg)
4.21 to 69.82
−13 ≤ h ≤ 13,
−8 ≤ k ≤ 7,
−20 ≤ l ≤ 20
24434
4.38 to 69.89
−11 ≤ h ≤ 11,
−23 ≤ k ≤ 24,
−5 ≤ l ≤ 7
1.92 to 30.51
−10 ≤ h ≤ 10,
−11 ≤ k ≤ 11,
−15 ≤ l ≤ 15
42402
index ranges
reflections collected
11533
independent reflections, Rint
completeness to θmax (%)
max. and min transmission
data/restraints/parameters
2375 (0.0259)
99.6
1229 (0.0269)
99.0
3904 (0.0503)
98.9
0.5924 and 0.3876
2375/0/140
0.7559 and 0.4700
1229/0/81
0.9836 and 0.9613
3904/33/183
1.171
a
goodness-of-fit
1.086
1.077
b
final R indices [I > 2σ(I)]
R1 = 0.0293,
wR2 = 0.0757
R1 = 0.0293,
wR2 = 0.0757
0.287 and −0.276
R1 =0.0291,
R1 =0.0531,
wR2 = 0.0815
R1 = 0.0302,
wR2 = 0.0823
0.359 and −0.269
wR2 = 0.1453
R1 =0.0599,
b
R indices (all data)
wR2 = 0.1502
0.816 and −0.779
largest diff. peak and hole (e·Å−3
)
a
b
GoF = [(∑[w(F2o − F2c)2])/(n − p)]1/2. R1 = (∑||Fo| − |Fc||)/(∑|Fo|); wR2 = [(∑[w(F2o − F2c)2)/(∑[w(Fo2)2])]1/2; w = (1)/(σ2(F2o) + (aP)2 + bP);
P = (2F2c + max(F2o, 0))/(3).
toluene, the enthalpy of conversion of 1-S to 1-S2 could be measured
real minimum energy structures. Neither minima nor transition states
could be located corresponding to a 1,3-cycloaddition of 2,6-
Me2C6H3CNO to the diphosphane 1′ via a concerted mechanism.
Likewise, direct OAT from RCNO to the diphosphane was found to
be repulsive and did not lead to reaction. No transition states were
found to contain a 5-membered {P2ONC} ring presumably because of
the repulsion between the lone pair electrons of the O atom and the
unactivated P atom. The second OAT does not involve a transition
state containing a six-membered {POCNOP} cycle with the nitrile-
oxide C and O atoms binding to the oxo atom and the other P atom of
the substrate, presumably because of the repulsion between the lone
pair electrons of the second P atom and the O atom of the nitrile
oxide.
and was found to be −21.7
enthalpies of substitution 1.3
0.6 kcal/mol. The difference in
1.0 kcal/mol showed the second
enthalpy of sulfurization to be slightly more exothermic than the first.
Because of this small difference, no correction for the small amount of
disubstituted product detected in the first sulfurization was needed. To
evaluate the bond strengths in solution, the enthalpy of solution of
SSbPh3 in toluene (+5.3 0.5 kcal/mol) must be subtracted giving
final enthalpies of sulfurization of −25.7 1.0 kcal/mol for formation
of 1-S from 1 and −27.0 1.1 kcal/mol for formation of 1-S2 from 1-
S.
Computational Details. Electronic structure calculations for
thermodynamic values were carried out using the M05-2X42 density
functional with the 6-311G(3df,2p) basis set as implemented in the
Gaussian 09 suite of programs.43 Minimum energy and transition state
structures were optimized by computing analytical energy gradients.
The obtained stationary points were characterized by performing
energy second derivatives, confirming them as minima or first order
saddle points by the number of negative eigenvalues of the Hessian
matrix of the energy (zero and one negative eigenvalues respectively).
Computed electronic energies were corrected for zero-point energy
and thermal energy to obtain the corresponding thermodynamic
property H0. To derive binding energies, the basis set superposition
error (BSSE) was computed using counterpoise calculations.44 All
structures were derived from the reported crystal structures of 1 and 1-
O2 after truncating the CH3 groups to H.
Using a value of 52.3 kcal/mol for the N−O BDE in MesCNO,28
and 94 kcal/mol for the P−S BDE in SPMe3,32 the computed reaction
enthalpies below yielded the following E−P BDEs:
ΔH=−78.7 kcal/mol
1 + MesCNO ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→⎯ 1‐O + MesCN
P−O in 1‐O: 131.1 kcal/mol
⇒
(7)
(8)
ΔH=−72.5 kcal/mol
1‐O + MesCNO ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→⎯ 1‐O2 + MesCN
P−O in 1‐O2: 124.8 kcal/mol
⇒
ΔH=+3.7 kcal/mol
1 + SPMe3 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 1‐S + PMe3
P−S in 1‐S: 90 kcal/mol
ΔH=+8.9 kcal/mol
⇒
Intrinsic reaction coordinate (IRC)45 calculations were done to
describe the reaction mechanisms for 2,6-Me2C6H3CNO and PhCNO
with 1′ and 1′-O, providing the connection between the minimum
energy points through the different transitions states. Further
optimization of the final points of the IRCs was done to obtain the
(9)
1‐S + SPMe3 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 1‐S2 + PMe3
P−S in 1‐S2: 85 kcal/mol
⇒
(10)
8860
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
Table 5. Crystallographic Data for Disulfide 1-S2 and Diiminodiphosphoranes 1-(NMes)2 and 1-(NCPh3)(NMes)
1-S2
1-(NMes)2
D8_11049/898607
1-(NCPh3)(NMes)
X8_11203/898608
reciprocal net code/CCDC no.
empirical formula, FW (g/mol)
crystal size (mm3)
11224/898606
C12H20P2S2, 290.34
0.30 × 0.10 × 0.06
100(2)
C30H42N2P2, 492.60
0.35 × 0.12 × 0.10
100(2)
C40H46N2P2, 616.73
0.30 × 0.05 × 0.05
100(2)
temperature (K)
wavelength (Å)
0.71073
1.54178
0.71073
crystal system, space group
unit cell dimensions (Å, deg)
monoclinic, P21/c
a = 6.7523(6), α = 90
b = 18.5199(17), β = 90.3740(10)
c = 11.2782(11), γ = 90
1410.3(2)
monoclinic, P21/c
a = 12.43340(10), α = 90
b = 15.8501(2), β = 98.4980(10)
c = 13.8650(2), γ = 90
2702.39(6)
monoclinic, P21/c
a = 18.026(2), α = 90
b = 9.1399(11), β = 102.766(3)
c = 20.939(3), γ = 90
3364.5(7)
volume (Å3)
Z
4
4
4
density (calc, g/cm3)
1.367
1.211
1.218
absorption coefficient (mm−1
)
0.577
1.603
0.160
F(000)
616
1064
1320
θ range for data collection (deg)
2.11 to 30.50
−9 ≤ h ≤ 9,
−26 ≤ k ≤ 26,
−16 ≤ l ≤ 16
38978
3.59 to 69.66
−15 ≤ h ≤ 15,
−19 ≤ k ≤ 19,
−11 ≤ l ≤ 15
55585
1.16 to 27.88
−23 ≤ h ≤ 23,
−12 ≤ k ≤ 10,
−27 ≤ l ≤ 27
103303
index ranges
reflections collected
independent reflections, Rint
completeness to θmax (%)
max. and min transmission
data/restraints/parameters
4311 (0.0383)
100.0
4971 (0.0403)
97.6
8011 (0.0793)
100.0
0.9662 and 0.8459
4311/0/149
0.8561 and 0.6038
4971/0/317
0.9920 and 0.9535
8011/0/404
1.078
a
goodness-of-fit
1.057
1.032
b
final R indices [I > 2σ(I)]
R1 = 0.0227,
wR2 = 0.0589
R1 =0.0272, wR2 = 0.0619
R1 = 0.0372,
wR2 = 0.0990
R1 =0.0448, wR2 = 0.1052
0.370 and −0.325
R1 =0.0446,
wR2 = 0.1100
R1 =0.0627, wR2 = 0.1196
0.527 and −0.317
b
R indices (all data)
largest diff. peak and hole (e·Å−3
)
0.466 and −0.270
a
b
GoF = [∑([w(F2o − F2c)2])/((n − p))]1/2. R1 = (∑||Fo| − |Fc||)/(∑|Fo|); wR2 = [(∑[w(F2o − F2c)2)/(∑[w(Fo2)2])]1/2; w = (1)/(σ2(F2o) + (aP)2 +
bP); P = (2F2c + max(F2o, 0))/(3).
All the other computations were carried out using the ORCA 2.9
quantum chemistry program package from the development team at
the University of Bonn.46 The LDA functional employed was that of
Perdew (PW-LDA)47 while the GGA part was handled using the
functionals of Becke and Perdew (BP86).48 In addition, all calculations
were carried out using the Zero-Order Regular Approximation
(ZORA),49 in conjunction with the SARC-TZV basis set hydrogen,
SARC-TZV(2pf) set for metals, and SARC-TZV(p) set for all other
atoms.50 Spin-restricted Kohn−Sham determinants have been chosen
to describe the closed-shell wave functions, employing the RI
approximation and the tight SCF convergence criteria provided by
ORCA. The structures were derived from the reported crystal
structure of 1-(NMes)2 by removing one imine group, and replacing
the other as necessary. The CO stretching frequencies were obtained
by constructing a Ni(CO)3 fragment on the free phosphorus atom in
MOLDEN51 For the open-shell wave functions of [1]•+ and [1-O]•+
the B3LYP functional, 6-31G basis set, spin-unrestricted Kohn−Sham
determinants were used. The optimized structures were rendered
using PLATON52 while the NBO distributions were computed using
GENNBO.53 The coordinates for all optimized structures are provided
in the Supporting Information.
X-ray Crystallographic Details. Diffraction quality crystals were
obtained typically from saturated, hot solutions by allowing them to
slowly cool down to room temperature. Colorless crystals of 1-O, 1-
(NMes)(NCPh3), and 1-(NMes)2 were obtained from hot benzene;
colorless crystals of 1-S2 and turquoise crystals of 3 from hot THF;
colorless crystals of 1-O2 from hot DMSO. Yellow crystals of 2 were
grown from a chloroform solution which was allowed to slowly
evaporate at room temperature. A colorless crystal of 1-O3 was
obtained by allowing a benzene solution of diphosphane 1 to age in an
NMR tube capped with a plastic cap. Low-temperature (100 K) data
were collected on a Siemens Platform three-circle diffractometer
coupled to a Bruker-AXS Smart Apex CCD detector with graphite-
monochromated Mo Kα radiation (λ = 0.71073 Å) for the structure of
disulfide 1-S2 and complex 3; on a Bruker-AXS X8 Kappa Duo
diffractometer coupled to a Smart Apex2 CCD detector with Mo Kα
radiation (λ = 0.71073 Å) for the structure of the trioxide 1-O3, the
diiminodiphosphorane 1-(NMes)(NCPh3), and complex 2; on a
Bruker D8 three-circle diffractometer coupled to a Bruker-AXS Smart
Apex CCD detector with graphite-monochromated Cu Kα radiation
(λ = 1.54178 Å) for the structures of the oxides 1-O and 1-O2, and the
complex 3 (φ- and ω-scans). General refinement details have been
described elsewhere.4,54 Reported structures were rendered using
PLATON.52 Specific details are provided in the text and tables below,
and in the form of .cif files available from the CCDC.55
The monoxide 1-O, disulfide 1-S2, and diiminodiphosphoranes 1-
(NMes)(NCPh3) and 1-(NMes)2 crystallized in the monoclinic space
group P21/c, each with one molecule of the exo-endo conformer per
asymmetric unit (Tables 4 and 5). The dioxide 1-O2 crystallized in the
orthorhombic space group Pnma, with half of a molecule of the exo-exo
conformer per asymmetric unit, and a symmetry plane containing the
phosphorus and oxygen atoms. All these models contain no solvent,
no disorder, and no restraints.
The trioxide 1-O3 co-crystallized with an isomer of the dioxide 1-O2
in the triclinic space group P1, with one molecule of exo-endo
̅
conformer per asymmetric unit. Only the central part consisting of one
P atom and the O atoms is disordered between the trioxide and the
dioxide, while the remaining atoms overlap between the two
constituents. The hydrogen atoms of the CH2 groups adjacent to
the disordered P atom were each refined over two positions. Similarity
restraints on 1−2 and 1−3 distances and displacement parameters as
well as rigid bond restraints for anisotropic displacement parameters
were applied to the phosphorus and oxygen atoms. The ratio between
8861
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
Table 6. Crystallographic Data for Molybdenum Complex 2 and Nickel Complex 3
2
3
reciprocal net code/CCDC no.
empirical formula, FW (g/mol)
color/morphology
X8_12025/898609
C35H47ClMoN2O2P2·3CHCl3, 1079.18
yellow/block
0.10 × 0.10 × 0.05
100(2)
11228/898610
C30H42Cl2N2NiP2·2C4H80, 766.41
blue/block
crystal size (mm3)
0.30 × 0.20 × 0.10
100(2)
temperature (K)
wavelength (Å)
0.71073
0.71073
crystal system, space group
unit cell dimensions (Å, deg)
monoclinic, P21/n
a = 13.838(2), α = 90
b = 18.902(3), β = 110.417(3)
c = 19.269(3), γ = 90
4723.6(14)
orthorhombic, P212121
a = 9.5206(7), α = 90
b = 16.9847(14), β = 90
c = 23.9758(19), γ = 90
3877.0(5)
volume (Å3)
Z
4
4
density (calc, g/cm3)
1.517
1.313
absorption coefficient (mm−1
)
0.946
0.756
F(000)
2200
1632
θ range for data collection (deg)
1.56 to 29.13
−18 ≤ h ≤ 18,
−25 ≤ k ≤ 25,
−26 ≤ l ≤ 26
111074
1.47 to 28.28
−12 ≤ h ≤ 12,
−22 ≤ k ≤ 22,
−31 ≤ l ≤ 31
93247
index ranges
reflections collected
independent reflections, Rint
completeness to θmax (%)
max. and min transmission
data/restraints/parameters
12696(0.0605)
100.0
9623 (0.0709)
100.0
0.9542 and 0.9113
12696/119/613
1.141
0.9283 and 0.8051
9623/434/520
1.073
a
goodness-of-fit
b
final R indices [I > 2σ(I)]
R1 =0.0431,
R1 = 0.0453,
wR2 = 0.1081
R1 = 0.0591,
wR2 = 0.1180
0.705 and −0.518
wR2 = 0.0832
R1 = 0.0566,
b
R indices (all data)
wR2 = 0.0879
largest diff. peak and hole (e·Å−3
)
0.928 and −0.789
a
b
GooF = [∑([w(Fo2 − F2c)2])/((n − p))]1/2. R1 = (∑||Fo| − |Fc||)/(∑|Fo|); wR2 = [(∑[w(F2o − F2c)2)/(∑[w(Fo2)2])]1/2; w = (1)/(σ2(F2o) + (aP)2 +
bP); P = (2F2c + max(F2o, 0))/(3).
the two components of the disorder was refined freely, and the sum of
the two occupancies for each pair was constrained to unity.
The molybdenum complex 2 crystallized in monoclinic space group
P21/n with one molecule of complex and three molecules of CHCl3,
while the nickel complex 3 crystallized in the orthorhombic space
group P212121 with one molecule of complex and two molecules of
THF (Table 6). In both cases, the complexes contained no disorders,
but all solvent molecules required modeling over two positions. The
occupancy of each position was refined freely, and their sums were
restrained to unity. Similarity restraints on 1−2 and 1−3 distances and
displacement parameters, as well as rigid bond restraints for
anisotropic displacement parameters were applied to the disordered
fragments.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the National
Science Foundation under CHE-1111357 (C.C.C.) and CHE-
0615743 (C.D.H.); and Spanish Ministry of Economy and
Competitiveness (MINECO) under CTQ2012-36966 (M.T.).
We thank Thermphos International for support including a gift
of white phosphorus, Dr. Peter Muller and Dr. Michael Takase
for assistance with crystallography, and Anthony Cozzolino for
assistance with computations.
̈
REFERENCES
■
(1) Piro, N. A.; Figueroa, J. S.; McKellar, J. T.; Cummins, C. C.
Science 2006, 313, 1276−1279.
(2) Tofan, D.; Cummins, C. C. Angew. Chem., Int. Ed. 2010, 49,
7516−7518.
(3) 3,4,8,9-Tetramethyl-1,6-diphosphabicyclo[4.4.0]deca-3,8-diene
may also be found under the CAS #1258870-28-8 or the alternative
name 1,4,6,9-tetrahydro-2,3,7,8-tetramethyl-[1,2]diphosphorino[1,2-
a][1,2]diphosphorin.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional synthetic, spectroscopic, and computational details.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(4) Tofan, D.; Cummins, C. C. Chem. Sci. 2012, 3, 2474−2478.
(5) (a) Johnson, A. W. Ylides and Imines of Phosphorus; John Wiley,
New York, 1993. (b) Collins, S. Handbook of Transition Metal
Polymerization Catalysts; John Wiley & Sons, Inc., 2010; Chapter
Ethylene Polymerization and α-Olefin Oligomerization Using
Catalysts Derived from Phosphoranes and Ni(II) and Ni(0)
Precursors, pp 459−466.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: manuel.temprado@uah.es (M.T.), c.hoff@miami.edu
(C.D.H.), ccummins@mit.edu (C.C.C.).
Notes
(6) Claver, C.; Gili, F.; Vinas, J.; Ruiz, A. Polyhedron 1987, 6, 1329−
̃
The authors declare no competing financial interest.
1335.
8862
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
(7) Li, G. Y.; Marshall, W. J. Organometallics 2002, 21, 590−591.
(8) (a) Meek, D. W.; Nicpon, P. J. Am. Chem. Soc. 1965, 87, 4951−
4952. (b) Ainscough, E. W.; Bergen, H. A.; Brodie, A. M.; Brown, K.
A. J. Chem. Soc., Dalton Trans. 1976, 1649−1656. (c) Schmutzler, R.;
Halfpenny, M. T.; Levason, W.; McAuliffe, C. A. J. Chem. Soc., Dalton
Trans. 1978, 1662−1663. (d) McQuillan, G. P.; Oxton, I. A. J. Chem.
Soc., Dalton Trans. 1979, 895−900. (e) Pomeroy, R. K.; Wijesekera, K.
S. Can. J. Chem. 1980, 58, 206−209. (f) Kuhn, N.; Winter, M. J.
Organomet. Chem. 1982, 239, C31−C34. (g) Et, D. T.; Legros, J.-P.
Phosphorus Sulfur Relat. Elem. 1983, 14, 377−380. (h) Troy, S.; Legros,
J.-P.; McQuillan, G. Inorg. Chim. Acta 1983, 72, 119−126. (i) Subasi,
S.; Nomoto, A.; Sonoda, M.; Ogawa, A. J. Org. Chem. 2008, 73, 7928−
7933.
(25) (a) Mackewitz, T. W.; Peters, C.; Bergstrasser, U.; Leininger, S.;
̈
Regitz, M. J. Org. Chem. 1997, 62, 7605−7613. (b) Keglevich, G.;
Farkas, R.; Ludan
Chem. 2005, 16, 104−110.
(26) Fave, C.; Hissler, M.; Kar
V.; Toupet, L.; Nyulazi, L.; Rea
6058−6063.
́
yi, K.; Kudar, V.; Hanusz, M.; Simon, K. Heteroat.
́
pat
́
i, T.; Rault-Berthelot, J.; Deborde,
́
́
u, R. J. Am. Chem. Soc. 2004, 126,
(27) Glendening, E. D.; Landis, C. R.; Weinhold, F. WIREs Comput.
Mol. Sci. 2012, 2, 1−42, DOI: 10.1002/wcms.51.
(28) Cai, X.; Majumdar, S.; Fortman, G. C.; Frutos, L. M.;
Temprado, M.; Clough, C. R.; Cummins, C. C.; Germain, M. E.;
Palluccio, T.; Rybak-Akimova, E. V.; Captain, B.; Hoff, C. D. Inorg.
Chem. 2011, 50, 9620−9630.
(29) (a) Staudinger, H.; Meyer, J. Helv. Chim. Acta 1919, 2, 635−
646. (b) Gololobov, Y. G.; Zhmurova, I. N.; Kasukhin, L. F.
Tetrahedron 1981, 37, 437−472. (c) Bebbington, M. W.; Bourissou, D.
Coord. Chem. Rev. 2009, 253, 1248−1261.
(30) Stewart, B.; Harriman, A.; Higham, L. J. Organometallics 2011,
30, 5338−5343.
(31) Faller, J. W. In Comprehensive Organometallic Chemistry III;
Crabtree, R. H., Mingos, D. M. P., Eds.; Elsevier: Oxford, U.K., 2007;
Vol. 1: Fundamentals; pp 407−427.
(32) Capps, K. B.; Wixmerten, B.; Bauer, A.; Hoff, C. D. Inorg. Chem.
1998, 37, 2861−2864.
(33) Tolman, C. A. Chem. Rev. 1977, 77, 313−348.
(34) (a) Cavell, G. R.; Aparna, K.; Kamalesh Babu, P. R.; Wang, Y. Q.
J. Mol. Catal. A: Chem. 2002, 189, 137−143. (b) Wei, R. P.; Stephan,
W. D. Organometallics 2002, 21, 1308−1310. (c) Hill, S. M.;
Hitchcock, B. P. J. Chem. Soc., Dalton Trans. 2002, 4694−4702.
(d) Gamer, T. M.; Rastatter, M.; Roesky, W. P.; Steffens, A.; Glanz, M.
Chem.Eur. J. 2005, 11, 3165−3172. (e) Rastatter, M.; Zulys, A.;
Roesky, W. P. Chem. Commun. 2006, 874−876. (f) Wiecko, M.;
Roesky, W. P.; Burlakov, V. V.; Spannenberg, A. Eur. J. Inorg. Chem.
2007, 876−881. (g) Gamer, T. M.; Roesky, W. P.; Palard, I.; Le
Hellaye, M.; Guillaume, M. S. Organometallics 2007, 26, 651−657.
(h) Klemps, C.; Buchard, A.; Houdard, R.; Auffrant, A.; Mezailles, N.;
Le Goff, F. X.; Ricard, L.; Saussine, L.; Magna, L.; Le Floch, P. New J.
Chem. 2009, 33, 1748−1752. (i) Buchard, A.; Platel, H. R.; Auffrant,
A.; Le Goff, F. X.; Le Floch, P.; Williams, K. C. Organometallics 2010,
29, 2892−2900. (j) Jenter, J.; Roesky, W. P.; Ajellal, N.; Guillaume, M.
S.; Susperregui, N.; Maron, L. Chem.Eur. J. 2010, 16, 4629−4638.
(35) Alonso, J. C.; Neves, P.; Pires da Silva, M. J.; Quintal, S.; Vaz, P.
E.; Senturk, O. S. Synth. React. Inorg., Met.-Org., Nano-Met. Chem. 2011,
̈
41, 834−838.
(9) (a) Almond, M. J.; Sarikahya, F.; Senturk, O. S. Polyhedron 1997,
̈
16, 1101−1103. (b) Senturk, O. S.; Shekhel, H. A.; Sterenberg, B. T.;
Udachin, K. A.; Sert, S.; Ozdemir, U.; Sarikahya, F. U. Polyhedron
̈
̈
2003, 22, 1659−1664.
(10) (a) Hieber, W.; Kummer, R. Z. Anorg. Allg. Chem. 1966, 344,
292−305. (b) Roesky, H. W.; Djarrah, H.; Thomas, M.; Krebs, B.;
Henkel, G. Z. Naturforsch. B 1983, 38B, 168−171.
(11) Westland, A. D.; Haque, F.; Bouchard, J. M. Inorg. Chem. 1980,
19, 2255−2259.
(12) Kuhn, N.; Winter, M. J. Organomet. Chem. 1983, 246, C80−
C82.
(13) Cotton, F. A.; Frenz, B. A.; Hunter, D. L.; Mester, Z. C. Inorg.
Chim. Acta 1974, 11, 111−117.
́
(14) (a) Liu, H.; Calhorda, M. J.; Drew, M. G.; Felix, V. Inorg. Chim.
Acta 2003, 347, 175−180. (b) Claver, C.; Ruiz, A.; Masdeu, A. M.;
Vinas, J.; Saballs, T.; Lahoz, F. J.; Plou, F. J. J. Organomet. Chem. 1989,
̃
373, 269−278. (c) Legros, J.-P.; Troy, D. Acta Crystallogr., Sect. B
1983, 39, 337−343.
(15) Cotton, F. A.; Frenz, B. A.; Hunter, D. L.; Mester, Z. C. Inorg.
Chim. Acta 1974, 11, 119−122.
(16) (a) Schmidt, U.; Boie, I. Angew. Chem., Int. Ed. 1966, 5, 1038−
1038. (b) Schmidt, U.; Boie, I.; Osterroht, C.; Schroer, R.;
̈
Grutzmacher, H.-F. Chem. Ber. 1968, 101, 1381−1397. (c) Kawashima,
̈
T.; Shimamura, M.; Inamoto, N. Heterocycles 1984, 21, 751.
(d) Brieden, W.; Kellersohn, T. Chem. Ber. 1993, 126, 845−847.
(17) (a) Sheldrick, W. S.; Pohl, S.; Zamankhan, H.; Banek, M.;
Amirzadeh-Asl, D.; Roesky, H. W. Chem. Ber. 1981, 114, 2132−2137.
(b) Alder, R. W.; Ganter, C.; Gil, M.; Gleiter, R.; Harris, C. J.; Harris,
S. E.; Lange, H.; Guy Orpen, A.; Taylor, P. N. J. Chem. Soc., Perkin
Trans. 1 1998, 1643−1656.
(18) Teramoto, Y.; Kubo, K.; Mizuta, T. J. Organomet. Chem. 2011,
696, 3402−3407.
́
D.; Silva, C.; Valente, A. A.; Ferreira, P.; Calhorda, M. J.; Felix, V.;
Drew, M. G. B. Organometallics 2007, 26, 5548−5556.
(19) (a) Vrieze, K.; Koten, G. V. Inorg. Chim. Acta 1985, 100, 79−96.
(b) Myeongsoon, K.; Ayusman, S.; Lev, Z.; L., R. A. Beyond
Metallocenes; American Chemical Society: Washington, DC, 2003;
Chapter 12, pp 143−153; (c) Klein, A.; Budnikova, Y. H.; Sinyashin,
O. G. J. Organomet. Chem. 2007, 692, 3156−3166. (d) Ivanchev, S. S.
Russ. Chem. Rev. 2007, 76, 617. (e) Anselment, T. M. J.; Vagin, S. I.;
Rieger, B. Dalton Trans. 2008, 4537−4548. (f) Guan, Z.; Popeney, C.
S. In Metal Catalysis in Olefin Polymerization; Guan, Z., Ed.; Topics in
Organometallic Chemistry; Springer-Verlag: London, U.K., 2009; Vol.
26; Chapter Recent Progress in Late Transition Metal α-Diimine
Catalysts for Olefin Polymerization, pp 179−220.
(36) Clark, D. A.; Jones, D. L.; Mawby, R. J. J. Chem. Soc., Dalton
Trans. 1980, 565−569.
(37) Coe, B. J.; Glenwright, S. J. Coord. Chem. Rev. 2000, 203, 5−80.
(38) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.
Organometallics 2010, 29, 2176−2179.
(39) (a) Grundmann, C.; Dean, J. M. J. Org. Chem. 1965, 30, 2809−
2812. (b) Barybin, M. V.; Diaconescu, P. L.; Cummins, C. C. Inorg.
Chem. 2001, 40, 2892−2897.
(40) Ugi, I.; Perlinger, H.; Behringer, L. Chem. Ber. 1958, 91, 2330−
2336.
(20) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. Rev. 2000, 100,
1169−1204.
(21) (a) Link, M.; Niecke, E.; Nieger, M. Chem. Ber. 1994, 127, 313−
319. (b) Lukashev, N. V.; Averin, A. D.; Zhichkin, P. E.; Kazankova,
M. A.; Beletskaya, I. P. Phosphorus Sulfur Relat. Elem. 1996, 109, 609−
612.
(22) Appel, R.; Milker, R. Chem. Ber. 1974, 107, 2658−2670.
(23) (a) Appel, R.; Milker, R.; Ruppert, I. Z. Anorg. Allg. Chem. 1977,
429, 69−73. (b) Appel, R.; Milker, R. Chem. Ber. 1977, 110, 3201−
3204.
(41) (a) Saunders, W. H.; Ware, J. C. J. Am. Chem. Soc. 1958, 80,
3328−3332. (b) Wiberg, N.; Neruda, B. Chem. Ber. 1966, 99, 740−
749. (c) Thayer, J. S.; West, R. Inorg. Chem. 1964, 3, 406−409.
(42) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput.
2006, 2, 364−382.
(43) Frisch, M. J. et al. Gaussian 09, Revision B.01; Gaussian, Inc.:
Wallingford, CT, 2010.
(44) (a) Simon, S.; Duran, M.; Dannenberg, J. J. J. Chem. Phys. 1996,
105, 11024−11031. (b) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19,
553−566.
(45) (a) Fukui, K. Acc. Chem. Res. 1981, 14, 363−368. (b) Hratchian,
H. P.; Schlegel, H. B. J. Chem. Theory Comput. 2005, 1, 61−69.
(24) (a) Arisawa, M.; Onoda, M.; Hori, C.; Yamaguchi, M.
Tetrahedron Lett. 2006, 47, 5211−5213. (b) Kawaguchi, S.; Nagata,
8863
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864

Inorganic Chemistry
Article
(46) Neese, F. ORCA-an ab initio, Density Functional and Semi-
empirical program package, Version 2.9; University of Bonn: Bonn,
Germany, 2012.
(47) Perdew, J. P.; Wang, Y. Phys. Rev. B: Condens. Matter Mater.
Phys. 1992, 45, 13244−13249.
(48) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100. (b) Perdew,
J. P. Phys. Rev. B Condens. Matter 1986, 33, 8822−8824.
(49) (a) Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1993, 99, 4597−4610. (b) Heully, J. L.; Lindgren, I.; Lindroth, E.;
Lundqvist, S.; Maartensson-Pendrill, A. M. J. Phys. B: At. Mol. Phys.
1986, 19, 2799−2815.
(50) Schaefer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
2571−2577.
(51) Schaftenaar, G. MOLDEN-a pre- and post processing program of
molecular and electronic structure, Version 5.0; Centre for Molecular
and Biomolecular Informatics: Nijmegen, The Netherlands, 2010.
(52) Spek, A. L. Acta Crystallogr. 2009, D65, 148−155.
(53) Weinhold, F. GENNBO, Natural Bond Orbital (NBO), 5.0
package.
(54) Tofan, D.; Cossairt, B. M.; Cummins, C. C. Inorg. Chem. 2011,
50, 12349−12358.
(55) These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/data_
request/cif.
8864
dx.doi.org/10.1021/ic401052a | Inorg. Chem. 2013, 52, 8851−8864