Dinitrogen as double lewis acid: Structure and bonding of triphenylphosphinazine N2(PPh3)2

Article abstract of DOI:10.1002/anie.201206305

Making Ns meet: Triphenylphosphinazine N₂(PPh₃) ₂ is a donor-acceptor complex between nitrogen in the highly excited ¹Γ_g state and two anti-periplanar coordinated phosphine ligands (see structure). Although the dissociation into N ₂+2 PPh₃ is calculated to be exergonic by 75 kcal mol ^-1, the compound is kinetically very stable as a result of the very large Lewis acidity of N₂ in the excited ¹Γ _g state. Copyright

Full text of DOI:10.1002/anie.201206305

.
Angewandte
Communications
DOI: 10.1002/anie.201206305
Dinitrogen Species
Dinitrogen as Double Lewis Acid: Structure and Bonding of
Triphenylphosphinazine N2(PPh3)2**
Nicole Holzmann, Deepak Dange, Cameron Jones,* and Gernot Frenking*
Much progress has been made in recent years
in the synthesis and isolation of small low-
valent, low oxidation state main-group mole-
cules M which are stabilized by N-heterocyclic
carbene ligands (NHC) in donor–acceptor
complexes (NHC!)nM.[1] In particular, dia-
tomic species E2 of Group 13–15 elements E
which cannot be isolated as free molecules
!
have been synthesized as NHC!E2 NHC
species. Robinson et al. reported in 2008 the
isolation of the silicon complex Si2(NHC)2
which has a trans-planar arrangement of the
=
CSiSiC moiety featuring
a Si Si double
bond.[2] A similar structure was found by us
for the heavier homologues Ge2(NHC)2[3] and
Sn2(NHC)2.[4] Theoretical studies suggest that
!
the NHC!E2 NHC donation in both these
compexes is into the vacant in-plane 1pu and
1pg valence orbitals of E2 in the excited (1)1Dg
state which yields two lone-pairs at the j E =
E j moiety (Figure 1a).[2–4] An even more
dramatic stabilization is found in the recently
Figure 1. Schematic representation of the electronic reference states of some diatomic
!
species E2 in the complexes L!E2 L. a) (1)1Dg state of Ge2, Sn2. b) (3)1Sg+ state of B2.
c) (1)1Gg state of N2, P2, As2. d) Donation of the plus and minus combination of the lone-
pair donor orbitals of L into the vacant in-plane p and p* orbitals of N2. The orbital
numbering refers to the valence orbitals of E2.
!
synthesized boron complex NHC!B2 NHC
by Braunschweig et al.[5] which has a boron–
boron triple bond and a linear CBBC arrange-
ment. The charge donation of the NHC ligands occurs into the
where the C-E-E-C torsion angle is 1808 or a gauche
conformation where the torsion angle is between 998–
1348.[1,7] The donation of the NHC ligands takes place into
the vacant in-plane 1pu’ (bonding) and 1pg’ (antibonding)
orbitals of E2 in the doubly excited (1)1Gg state (Figure 1c)
which has four lone-pairs at the jE ꢀ Ej acceptor fragment.
Two of them have p symmetry (1pu/1pg) while two have s-
symmetry (1su/2sg).
vacant 1su and 2sg valence orbitals of the highly (double)
excited (3)1Sg state of B2 which leads to an electron
+
configuration that is similar to the ground-state configuration
of N2 (Figure 1b).[6] Theoretical studies predict that the
!
heavier Group 13 homologues NHC!E2 NHC where E =
Al–In have an anti-periplanar structure.[6a] The Al–In com-
plexes could not become synthesized to date.
!
!
Complexes NHC!E2 NHC have also been isolated for
The donor–acceptor interaction NHC!E2 NHC in the
the Group 15 elements E = P, As.[7] Depending on the size of
the substituents at nitrogen, the equilibrium structures have
either an anti-periplanar arrangement of the NHC ligands
complexes of the Group 13–15 atoms E are strong enough to
overcompensate the excitation energies of E2 into the
electronic reference state which leads to thermodynamically
stable species E2(NHC)2. The situation looks less favorable
for dinitrogen species N2(NHC)2. A recent computational
[*] N. Holzmann, Prof. Dr. G. Frenking
!
Fachbereich Chemie, Philipps-Universitꢀt Marburg
Hans-Meerwein-Strasse, 35032 Marburg (Germany)
E-mail: frenking@chemie.uni-marburg.de
study of Group 14 and 15 complexes NHC!E2 NHC by
Wilson et al. showed that all the complexes are stable toward
dissociation of the NHC ligands except for E = N.[8] Depend-
ing on the substituents R at nitrogen of the NHC ligand, the
dissociation reaction N2(NHC)2!N2 + 2NHC was calculated
to be exergonic by between 21.9 kcalmolꢀ1 (R = H) and
33.5 kcalmolꢀ1 (R = methyl). Even lower thermodynamic
stabilities were predicted for the phosphine complexes
D. Dange, Prof. Dr. C. Jones
School of Chemistry, Monash University
PO Box 23, Melbourne, VIC, 3800 (Australia)
E-mail: cameron.jones@monash.edu
[**] Helpful comments by Roald Hoffmann are gratefully acknowledged.
This work was supported by the Deutsche Forschungsgemeinschaft.
C.J. thanks the Australian Research Council.
!
R3P!N2 PR3 for which the calculated dissociation reaction
N2(PR3)2!N2 + 2PR3 is exergonic by between 87.8 kcalmolꢀ1
(R = phenyl) and 129.9 kcalmolꢀ1 (R = H).[8]
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201206305.
3004
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3004 –3008

Angewandte
Chemie
ꢀ
The calculated thermodynamic instabilities of N2(NHC)2
and N2(PR3)2 are in striking contrast to experimental studies.
Himmel and co-workers reported in 2011 that N2(NHC)2 in
which the NHC ligand has methyl groups at nitrogen is
thermally stable and can be purified by sublimation at 908.[9]
The X-ray structure of the compound showed an anti-
value (1.579 ꢀ) for all reported Ph3P N(two-coordinate) separa-
tions.[11]
We analyzed the electronic structure of N2(PPh3)2 with
quantum chemical methods[12] to understand the surprising
thermal stability of the compound. The geometry optimiza-
tion at RI-BP86/def2-TZVPP gave an anti-periplanar
arrangement of the phosphine ligands as it is found in the
solid state. Figure 2 shows that the calculated bond lengths
and angles at RI-BP86/def2-TZVPP are in reasonable agree-
ꢀ
periplanar arrangement of the NHC ligands and a N N
bond length of 1.415 ꢀ which is in good agreement with the
calculated structure of Wilson et al.[8] Even more surprising is
the experimental study by Appel and Schçllhorn who
reported in 1964 that triphenylphosphinazine which was
ꢀ
ment with the experimental values. The theoretical N N bond
length (1.441 ꢀ) is shorter than the experimental value
= ꢀ =
ꢀ
sketched with the formula Ph3P N N PPh3 is a thermally
(1.497 ꢀ) while the calculated P N bonds are a bit longer
stable diamagnetic species that has a melting point of 1848.[10]
However, no structural information about the compound was
given in the work. In the light of the theoretically predicted
free reaction energy of N2(PPh3)2 for decomposition into N2
and two PPh3 molecules of ꢀ87.8 kcalmolꢀ1[8] we decided to
reinvestigate the experimental work and to determine the
geometry of the compound.
(1.606 ꢀ) than the measured value (1.582 ꢀ). The calculated
N–N distance at B3LYP/TZVP which was reported by Wilson
et al. (1.457 ꢀ)[8] is also shorter than the experimental data.
The theoretically predicted reaction energy for the dissocia-
tion N2(PPh3)2!N2 + 2PPh3 at RI-BP86/def2-TZVPP is
ꢀ47.5 kcalmolꢀ1. After correcting for thermal and entropic
contributions we calculate a Gibbs free energy of ꢀ74.5 kcal
molꢀ1. This value is a bit smaller than the MP2/TZVP//
B3LYP/TZVP value of ꢀ87.8 kcalmolꢀ1 which was reported
by Wilson et al.[8] but there is agreement that triphenylphos-
phinazine is a thermodynamically unstable species.
The triphenylphosphinazine, N2(PPh3)2, was prepared
according to the method of Appel and co-workers,[10] that is,
by deprotonation of [Ph3PN(H)N(H)PPh3]Cl2 with KOtBu. It
is a deep red crystalline solid that quantitatively decomposes
in the melt to PPh3 and N2 above 2158C (m.p. 184–1868C).
When exposed to the air it is rapidly and quantitatively
Figure 3 shows the most important occupied orbitals of
N2(PPh3)2 which are relevant for the bonding interactions
between dinitrogen and the phosphine ligands. The HOMO is
the out-of-plane[13] p* orbital of N2 (1pg in Figure 1c) which
has only small contributions at PPh3. The shape of the HOMO
and the anti-periplanar arrangement of the PPh3 ligands
indicate that the electronic reference state of N2 is the highly
excited (1)1Gg state which has the valence configuration
(1sg)2(1su)2(1pu)2(2sg)2(1pg)2 where the 1pu’ and 1pg’ are
vacant (Figure 1c). The 1pu’ and 1pg’ MOs serve as in-plane[13]
acceptor orbitals for the plus and minus combinations of the
PPh3 s lone-pair orbitals. Figure 3 shows that the HOMOꢀ1
=
oxidized to O PPh3, presumably also generating N2. In their
original publication, Appel and co-workers did not report
NMR spectroscopic data for N2(PPh3)2. In the current study
we obtained these data, most notable of which is the
31P{1H} NMR chemical shift of the compound at d =
9.16 ppm. Moreover, the X-ray crystal structure of the
compound was obtained (Figure 2). This structure shows it
to be dimeric with a planar PN2P fragment with PNN angles
of 107.10(11)8. The N–N separation in the compound
(1.497(2) ꢀ) is extremely long for a single bond, and in fact
lies within the longest 0.1% of such interactions in com-
pounds bearing XN–NX fragments (X = any atom, N is two-
!
displays the features of the Ph3P!N2 PPh3 s donation from
the plus combination of the phosphorus lone-pairs while the
coordinate).[11] Conversely, the P N bonds in N2(PPh3)2
ꢀ
(1.5819(12) ꢀ), while short, are close in length to the mean
Figure 2. X-ray crystal structure of N2(PPh3)2. Selected bond lengths [ꢁ]
and angles [8]. Calculated values at RI-BP86/def2-TZVPP are given in
brackets: N1–N1 1.497(2) [1.441], P1–N1 1.5819(12) [1.606], P1-N1-
N1’ 107.10(11) [112.6], P1-N1-N1’-P1’ 180.0 [180.0].
Figure 3. Plot of the relevant occupied orbitals of N2(PPh3)2.
Angew. Chem. Int. Ed. 2013, 52, 3004 –3008
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3005

.
Angewandte
Communications
HOMOꢀ15 can be identified as the minus combination of the
!
Ph3P!N2 PPh3 s donation. The HOMOꢀ15 orbital is much
lower in energy than the HOMOꢀ1, because the donation
takes place into the bonding 1pu’ orbital of N2 while the 1pg’
acceptor orbital of N2 is N–N antibonding. The shape of the
1pg’ MO suggests that there is some mixing with the 2sg
orbital of N2 which has lone-pair character (Figure 1). The
HOMOꢀ14 is the occupied 1pu orbital of N2 which exhibits
some mixing with the p orbitals of the phenyl rings.
The bonding situation in N2(PPh3)2 was further analyzed
with the EDA-NOCV method.[14] Table 1 gives the numerical
Table 1: EDA-NOCV results of N2(PPh3)2 using the fragments N2 in the
(1)1Gg reference state and 2PPh3. Energy values in kcalmolꢀ1
.
DEint
DEPauli
DEelstat
ꢀ300.1
853.4
[a]
ꢀ377.4 (32.7%)
ꢀ776.1 (67.3%)
ꢀ357.9 (46.1%)
ꢀ289.8 (37.3%)
ꢀ46.9 (6.0%)
ꢀ81.5 (10.6%)
[a]
DEorb
[b,c]
[b,c]
[b,c]
[b]
DEs1
DEs2
DEp1
Figure 4. Plot of the deformation densities D1 which are associated
with most relevant orbital interactions in N2(PPh3)2. a) Deformation
DEresr
!
density D11 arising from donation Ph3P!N2 PPh3 into the vacant in-
[a] The percentage values in parentheses give the contribution to the
total attractive interactions DEelstat +DEorb. [b] The percentage values in
parentheses give the contribution to the total orbital interactions DEorb
[c] The notation s1, s2, p1 refers to the orbital pairs which are associated
with the deformation densities D11, D12, D13 that are shown in Figure 4.
plane p orbital of N2(1Gg). b) Deformation density D12 arising from
!
donation Ph3P!N2 PPh3 into the vacant in-plane p* orbital of N2-
.
(1Gg). c) Deformation density D13 arising from p back-donation
!
Ph3P N2!PPh3 from the occupied out-of-plane p orbital of N2(1Gg).
!
Ph3P N2!PPh3 p back-donation. This picture is also sup-
results. The intrinsic interaction energy DEint between the
fragments N2 in the (1)1Gg reference state and PPh3 in the
frozen geometry of N2(PPh3)2 is very large [Eq. (1)]
ported by an NBO analysis of N2(PPh3)2. The calculated
partial charge of the N2 moiety in the complex is ꢀ1.73 e. The
ꢀ
NBO calculations give a N N single bond which has nearly
perfect sp3 hybridization (Table 2). There are two lone-pair
orbitals at each nitrogen, one with s symmetry that has 48%
s contribution and one p lone-pair. The NBO results give two
N2ð11GgÞfr þ 2 ðPPh3Þfr ! N2ðPPh3Þ2
ꢀ300:1 kcal molꢀ1
ð1Þ
The EDA-NOCV data suggest that one third of the
attraction between N2(1Gg) and the two PPh3 molecules comes
from electrostatic interactions DEelstat while two thirds comes
from orbital interactions DEorb. Inspection of the pair
contributions to the DEorb terms reveals that the dominant
interactions come from the plus and minus combinations of
ꢀ
P N single bonds which are clearly localized toward the
nitrogen end (66%). The resulting extended Lewis structure
for triphenylphosphinazine which uses arrows for donor–
acceptor bonds reads as:
!
Ph3P!N2 PPh3 s donation. Figure 4 shows the deformation
densities D1 which are associated with the most important
orbital interactions. Note that the color coding indicates the
direction of the charge flow donor!acceptor (red!blue).
The deformation density D11 can be identified with the charge
!
donation Ph3P!N2 PPh3 into the vacant in-plane[13] p orbi-
The bonding situation may be expressed as double ylid
Ph3P+-Nꢀ-Nꢀ-P+Ph3 as it was done in a later paper by Appel
et al.[15] We prefer the notation with arrows because it directly
reveals the electronic reference state of N2 and the nature of
the phosphorus–nitrogen bonds.
tal of N2(1Gg) while D12 illustrates the charge flow of the
phosphorus lone-pair electrons into the vacant p* orbital of
N2(1Gg). The sum of the two interactions provides 83.4% of
!
DEorb (Table 1). The p back-donation Ph3P N2!PPh3 which
is associated with the deformation
density D13 contributes only 6.0%.
Table 2: NBO results at BP86/def2-TZVPP for the PNNP moiety of N2(PPh3)2.
The EDA-NOCV results and
the shape of the valence orbitals
suggest that the bonding between
dinitrogen and the phosphine
Orbital
Bond order Occupancy. % (N) % s(N) % p(N) % (P) % s(P) % p(P)
ꢀ
N N
1.056
1.184
–
1.97
1.97
1.90
1.63
50.0
66.4
–
23.7
28.3
47.7
0.4
76.1
71.2
52.1
99.3
–
33.6
–
–
29.4
–
–
70.0
–
ꢀ
P N
N (lone pair)
N (lone pair)
ligands comes from strong Ph3P!
–
–
–
–
–
!
N2 PPh3
s donation and less
3006 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3004 –3008

Angewandte
Chemie
The description of N2(PPh3)2 with two lone-pairs at each
nitrogen atom suggests that the compound should be a strong
proton acceptor and that even the second proton affinity,
where the same nitrogen atom is protonated, should be rather
high. This idea is verified by the calculated proton affinities
(PAs). The first PA of N2(PPh3)2 is 245.2 kcalmolꢀ1 which is
larger than the PA of most amines.[16] The second PA yielding
Ph3PN(H+)2NPPh3 is 170.5 kcalmolꢀ1 which is only a bit
smaller than the second PA yielding Ph3PN(H+)N(H+)PPh3
(195.6 kcalmolꢀ1). It has been shown that the second PA is
a sensitive probe for the occurrence of two lone pairs in
carbon bases.[17] The calculated second PA for one nitrogen
atom is a further evidence for the existence of two lone-pair
orbitals at nitrogen.[18]
a RASSCF level which is not possible for us. However, our
calculations do provide information about the dissociation
pathway. We calculated N2(PPh3)2 at RI-BP86/def2-TZVPP
ꢀ
where one P N separation was fixed at longer distances than
ꢀ
the equilibrium values with intervals of 0.1 ꢀ. The second P
ꢀ
N bond becomes longer while the N N bond becomes clearly
shorter up to a value of 2.0 ꢀ for the frozen P N bond. At this
ꢀ
ꢀ
point, the second P N bond length was 1.606 ꢀ, the N–N
separation was 1.379 ꢀ, and the energy was 25.1 kcalmolꢀ1
higher than the optimized structure. When the P–N frozen
ꢀ
separation was further elongated to 2.1 ꢀ, the second P N
bond broke and the associated PPh3 ligand dissociated. A
geometry optimization of a possible intermediate N2(PPh3)
did not give an equilibrium structure. The calculations suggest
that the reaction N2(PPh3)2!N2 + 2PPh3 which proceeds as
Why is triphenylphosphinazine an amazingly stable com-
pound although the dissociation reaction into N2 + 2PPh3 is
strongly exergonic? As shown above, the intrinsic interaction
energy DEint between the fragments N2 in the (1)1Gg reference
state and PPh3 in the frozen geometry of N2(PPh3)2 is very
large (ꢀ300.1 kcalmolꢀ1). Thus, dinitrogen in the (1)1Gg state
is a very powerful Lewis acid which is due to two reasons. One
reason is the rather high electronegativity of nitrogen while
the other reason is that the nitrogen atoms in jN ꢀ Nj have
only an electron sextet. However, the large interaction energy
does not compensate for the relaxation of the fragments into
the equilibrium geometries and electronic ground states
which are calculated to be 8.3 kcalmolꢀ1 for both phosphine
ligands and 341.0 kcalmolꢀ1 for the process N2 (11Gg)fr!
ꢀ
concerted but not necessarily synchronous rupture of the P N
bonds with a rather high barrier. The DFT calculations
suggest that the activation barrier for the concerted but not
necessarily synchronous reaction is higher than 25.1 kcal
molꢀ1 but lower than 67.6 kcalmolꢀ1 which is the bond
ꢀ
dissociation energy for the N N bond of N2(PPh3)2 yielding
NPPh3 in the electronic doublet ground state [Eq. (2)].
ð2Þ
N2ðPPh3Þ2 ! 2 NPPh3
þ 67:6 kcal molꢀ1
The above results suggest that donor–acceptor interac-
tions[20] in main-group compounds of atoms of the first octal-
row may be more important than hitherto realized. In 2006 it
was recognized by one of us[21] that carbodiphosphoranes
C(PR3)2 which were synthesized as early as 1961[22] are
examples for the compound class of carbones CL2 which have
N2(X1Sg ). High-level ab initio calculations predict that the
+
(1)1Gg state of N2 has an equilibrium separation of 1.608 ꢀ and
is 294.3 kcalmolꢀ1 above the X1Sg ground state.[19] The RI-
+
BP86/def2-TZVPP calculations for N2 (1)1Gg give a N–N
separation of 1.661 ꢀ and an excitation energy of 328.6 kcal
molꢀ1. This is a remarkably good agreement with the ab initio
result in light of the approximations of the DFT method. The
N–N separation becomes shorter in N2(PPh3)2 because the
!
donor–acceptor bonds L!C L to a naked carbon atom in
the singlet 1D state that has two lone pairs, and is thus
a double Lewis base. This was the starting point for intensive
studies which led to the synthesis of new carbones and
carbone complexes.[23] Herein we show that that triphenyl-
phosphinazine which was isolated in 1964[10] is a donor–
ꢀ
donation into the N N bonding orbital 1pu’ is likely to affect
the N–N distance more strongly than the donation into the
antibonding 1pg’ acceptor orbital (See Figure 1d and
Figure 3). The 1pg’ orbital mixes with the 2sg orbital of N2
!
acceptor complex of dinitrogen Ph3P!N2 PPh3 which can
be considered as an example of dinitrogen complexes L!
ꢀ
(Figure 1) which decreases the N N antibonding character.
!
N2 L. We think that triphenylphosphinazine is an excep-
The calculations thus suggest that the very strong Ph3P!
tional compound, because it melts at 1848 and it decomposes
into N2 and PPh3 only above 2158 although the associated
bond breaking reaction is calculated to be exergonic by 74.5–
87.7 kcalmolꢀ1. Further studies might lead to the synthesis of
complexes N2(L)2 with other ligands L and to transition-metal
complexes in which the nitrogen atoms of N2(L)2 serve as
double donors to the metal. Also, it has not escaped our
attention that the direct synthesis of N2(PPh3)2 from N2 and
PPh3 would be a significant contribution to the topics of
nitrogen activation and chemical energy storage. The results
which are reported herein are a challenge for experiment.
!
N2(1)1Gg PPh3 attraction makes triphenylphosphinazine
a
kinetically stable compound, because the activation
energy for breaking the donor–acceptor bonds is very high.
We carried out preliminary calculations to estimate the
activation barrier for dissociation of the phosphine ligands
but we experienced severe convergence problems. The
transition state for the dissociation reaction N2(PPh3)2!
N2 + 2PPh3 which involves a change in the configuration of
+
the N2 moiety from the (1)1Gg excited state to the X1Sg
ground state can be optimized only at a multi-reference
level. This is a formidable task which requires substantial
computational resources, because a size reduction of the
molecule to a model system such as N2(PH3)2 is not feasible
because hydrogen migration may take place during the
geometry optimization. A calculation of the transition state
for the molecule which has C1 symmetry involves the
optimization of 3Nꢀ6 = 214 geometry variables at
Received: August 6, 2012
Revised: October 16, 2012
Published online: February 4, 2013
Keywords: bonding analysis · DFT calculations ·
dinitrogen complex · donor–acceptor interactions
.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3004 –3008
www.angewandte.org 3007

.
Angewandte
Communications
[17] R. Tonner, G. Heydenrych, G. Frenking, ChemPhysChem 2008,
9, 1474.
[18] One referee challenged our interpretation of the bonding
situation in N2(PPh3)2 saying that “each N atom has the same
[1] a) Y. Wang, G. H. Robinson, Inorg. Chem. 2011, 50, 12326; b) N.
Kuhn, A. Al-Sheikh, Coord. Chem. Rev. 2005, 249, 829; c) Y.
Yang, G. H. Robinson, Dalton Trans. 2012, 41, 337; d) Y. Yang,
G. H. Robinson, Chem. Commun. 2009, 5201.
=
environment as in any other phosphinimine (R3P NX) type
[2] Y. Wang, Y. Xie, P. Wei, R. B. King, H. F. Schaefer III, P. v. R.
Schleyer, G. H. Robinson, Science 2008, 321, 1069.
[3] A. Sidiropoulos, C. Jones, A. Stasch, S. Klein, G. Frenking,
Angew. Chem. 2009, 121, 9881; Angew. Chem. Int. Ed. 2009, 48,
9701.
molecule.” We strongly disagree! Although the “environment”
of the N atoms may be similar, the bonding analysis give clear
evidence that N2(PPh3)2 is the phosphine-substituted nitrogen
!
homologue of the donor–acceptor complexes L!E2 L where
the species with E = P, As have been identified before.[7]
Phosphinimines have a substantially different bonding situation
which becomes evident by calculating the bond dissociation
[4] C. Jones, A. Sidiropoulos, N. Holzmann, G. Frenking, A. Stasch,
Chem. Commun. 2012, 48, 9855.
[5] H. Braunschweig, R. D. Dewhurst, K. Hammond, J. Mies, K.
Radacki, A. Vargas, Science 2012, 336, 1420.
[6] a) N. Holzmann, A. Stasch, C. Jones, G. Frenking, Chem. Eur. J.
2011, 17, 13517; b) G. Frenking, N. Holzmann, Science 2012, 336,
1394.
[7] a) Y. Wang, Y. Xie, P. Wei, R. B. King, H. F. Schaefer III, P. v. R.
Schleyer, G. H. Robinson, J. Am. Chem. Soc. 2008, 130, 14970;
b) M. Y. Abraham, Y. Wang, Y. Xie, P. Wei, H. F. Schaefer III,
P. v. R. Schleyer, G. H. Robinson, Chem. Eur. J. 2010, 16, 432.
[8] D. J. D. Wilson, S. A. Couchman, J. L. Dutton, Inorg. Chem.
2012, 51, 7657.
=
energy of the R3P NX bond. The theoretically predicted value
=
for the dissociation reaction Ph3P NPh!Ph3P + NPh at RI-
BP86/def2-TZVPP is DG =+ 44.6 kcalmolꢀ1. This is in striking
contrast to the calculated dissociation energy of ꢀ74.5 kcalmolꢀ1
!
for the dissociation reaction of Ph3P!N2 PPh3.
[19] M. Hochlaf, H. Ndome, D. Hammoutꢂne, M. Vervloet, J. Phys. B
2010, 43, 245101.
[20] A. Haaland, Angew. Chem. 1989, 101, 1017; Angew. Chem. Int.
Ed. Engl. 1989, 28, 992.
[21] R. Tonner, F. ꢃxler, B. Neumꢄller, W. Petz, G. Frenking, Angew.
Chem. 2006, 118, 8206; Angew. Chem. Int. Ed. 2006, 45, 8038.
[22] F. Ramirez, N. B. Desai, B. Hansen, N. McKelvie, J. Am. Chem.
Soc. 1961, 83, 3539.
[9] M. Reinmuth, C. Neuhꢁuser, P. Walter, M. Enders, E. Kaifer, H.-
J. Himmel, Eur. J. Inorg. Chem. 2011, 83.
[10] R. Appel, R. Schçllhorn, Angew. Chem. 1964, 76, 991; Angew.
Chem. Int. Ed. Engl. 1964, 3, 805.
[11] As determined from a survey of the Cambridge Crystallographic
Database, July 2012.
[12] Details of the theoretical methods are given in Supporting
Information.
[13] The plane we are referring to is defined by the P-N-N-P
fragment.
[14] a) T. Ziegler, A. Rauk, Inorg. Chem. 1979, 18, 1558; b) A.
Michalak, M. Mitoraj, T. Ziegler, J. Phys. Chem. A 2008, 112,
1933; c) M. P. Mitoraj, A. Michalak, T. Ziegler, J. Chem. Theory
Comput. 2009, 5, 962.
[15] R. Appel, B. Blaser, G. Siegemund, Z. Anorg. Allg. Chem. 1968,
363, 176.
[23] a) C. A. Dyker, V. Lavallo, B. Donnadieu, G. Bertrand, Angew.
Chem. 2008, 120, 3250; Angew. Chem. Int. Ed. 2008, 47, 3206;
b) A. Fꢄrstner, M. Alcarazo, R. Goddard, C. W. Lehmann,
Angew. Chem. 2008, 120, 3254; Angew. Chem. Int. Ed. 2008, 47,
3210; c) M. Alcarazo, C. W. Lehmann, A. Anoop, W. Thiel, A.
Fꢄrstner, Nat. Chem. 2009, 1, 295; d) I. Fernꢅndez, C. Dyker, A.
De Hope, B. Donnadieu, G. Frenking, G. Bertrand, J. Am.
Chem. Soc. 2009, 131, 11875; e) M. Melaimi, P. Parameswaran, B.
Donnadieu, G. Frenking, G. Bertrand, Angew. Chem. 2009, 121,
4886; Angew. Chem. Int. Ed. 2009, 48, 4792; f) W. Petz, F. ꢃxler,
B. Neumꢄller, R. Tonner, G. Frenking, Eur. J. Inorg. Chem. 2009,
4507; g) B. Inꢆs, M. Patil, J. Carreras, R. Goddard, W. Thiel, M.
Alcarazo, Angew. Chem. 2011, 123, 8550; Angew. Chem. Int. Ed.
2011, 50, 8400.
[16] D. H. Aue, M. T. Bowers, Gas Phase Ion Chemistry, Vol. 2, p. 1
(Ed.: M. T. Bowers), Academic Press, New York, 1979, p. 1 f.
3008 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3004 –3008