A Systematic Survey of the Reactivity of Chlorinated N2P2, NP3 and P4 Ring Systems

Article abstract of DOI:10.1002/chem.201903410

The reactivity of the four-membered NP₃ ring system [RN(μ-PCl)₂PR] (R=Mes=2,4,6-tri-tert-butylphenyl) towards Lewis acids, Lewis bases, and reducing agents was investigated. Comparisons with the literature-known, analogous cyclic compounds [ClP(μ-NR)]₂ (R=Ter=2,6-dimesitylphenyl) and [ClP(μ-PR)]₂ (R=Mes*) are drawn, to obtain a better systematic understanding of the reactivity of cyclic NP species. Apart from experimental results, DFT computations are discussed to further the insight into bonding and electronic structure of these compounds.

Full text of DOI:10.1002/chem.201903410

A Journal of
Accepted Article
Title: A systematic survey of the reactivity of chlorinated N2P2, NP3
and P4 ring systems
Authors: Jonas Bresien, Liesa Eickhoff, Axel Schulz, Tim Suhrbier, and
Alexander Villinger
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201903410
Link to VoR: http://dx.doi.org/10.1002/chem.201903410
Supported by

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
A systematic survey of the reactivity of chlorinated N2P2, NP3 and
P4 ring systems
Jonas Bresien,*[a] Liesa Eickhoff,[a] Axel Schulz,*[a,b] Tim Suhrbier,[a] and Alexander Villinger[a]
Abstract: The reactivity of the four-membered NP3 ring system
[RN(μ-PCl)2PR] (R = Mes* = 2,4,6-tri-t-butylphenyl) towards Lewis
acids, Lewis bases, and reducing agents was investigated.
Comparisons with the literature-known, analogous cyclic compounds
[ClP(μ-NR)]2 (R = Ter = 2,6-dimesitylphenyl) and [ClP(μ-PR)]2 (R =
Mes*) are drawn, to obtain a better systematic understanding of the
reactivity of cyclic NP species. Apart from experimental results, DFT
computations are discussed to further the insight into bonding and
electronic structure of these compounds.
membered N2P2 ring systems were thoroughly studied, as
detailed in a number of review articles.[16–20] Out of the many
possible substitution patterns at the N2P2 ring system, those
species with halogen substituents (e.g. compounds of the type
[XP(μ-NR)]2 (A, X = halogen, R = sterically demanding group;
Scheme 1) were shown to be easily functionalized by halide
abstraction, substitution reactions, or reduction, rendering them
worthwhile building blocks in phosphorus-nitrogen chemistry
(Scheme 2).[21–26]
Introduction
Phosphorus chemistry nowadays plays an important role in a
variety of research domains, such as biochemistry, organic and
inorganic chemistry, catalysis, and materials science.[1–6] A great
variety of chemical processes rely on phosphorus compounds,
both in nature[7] and in industrial chemistry. For example,
phosphane-based ligands find widespread use in transition
metal complexes that are being applied for large-scale industrial
processes, such as hydroformylation reactions or syntheses
involving C–C and C–N bond formation.[3]
Scheme 1. Four-membered N2P2, NP3 and P4 ring systems (R = sterically
demanding substituent, X = (pseudo)halogen).
It is therefore desirable to further the systematic develop-
ment of phosphorus chemistry, particularly when regarding the
fact that phosphorus is often dubbed a “carbon copy”.[8] The
chemistry of carbon is, of course, well and systematically
investigated, and there is a plethora of different types of
reactions that can be used to synthesize a large variety of
different classes of organic compounds. It is probably without
dispute that the same level of understanding has not yet been
reached in case of phosphorus chemistry, which is why we are
interested in a systematic investigation of phosphorus-based
compounds.
In particular, our group has a long-standing interest in cyclic
oligophosphanes and aminophosphanes. Such ring systems
show a diverse reaction behaviour and have therefore attracted
the interest of many researchers during the past few decades.[9–
15]
Among these classes of compounds, especially four-
Scheme 2. Reactivity of A (R = sterically demanding substituent, LB = Lewis
base; LA = Lewis acid; M = Mg, K; X, Y = Cl, OTf, N3, etc.).
[a]
[b]
Dr. J. Bresien, L. Eickhoff, Prof. Dr. A. Schulz, T. Suhrbier,
Dr. A. Villinger
Institut für Chemie
More recently, we became interested in the reactivity of
cyclic phosphanes of the type [XP(μ-PR)]2 (C),[27–31] which had
barely been investigated prior to our work. In particular, we were
interested in how the reaction behaviour of these species would
compare to the congeneric N2P2 ring systems (A), in view of the
formal replacement of the two N atoms by phosphorus. It was
found that the P4 ring system C displayed a tendency to stabilize
positive charges (induced by halide abstraction or substitution)
by rearrangement reactions associated with the formation of
Universität Rostock
Albert-Einstein-Str. 3a, 18059 Rostock, Germany
E-mail: jonas.bresien@uni-rostock.de, axel.schulz@uni-rostock.de
Prof. Dr. A. Schulz
Abteilung Materialdesign
Leibniz-Institut für Katalyse an der Universität Rostock e.V.
Albert-Einstein-Str. 29a, 18059 Rostock, Germany
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
transannular P–P bonds (Scheme 3).[32–35] The N2P2 ring system
A, on the other hand, was shown to stabilize positive charges by
delocalization of π-electron density within the ring, due to the
p-type character of the lone pair of electrons (LP) at N. In
contrast, the LP at phosphorus in the P4 ring system has a large
s character, which hampers this kind of electronic interaction.
After addition of CH2Cl2 to the suspension, the starting material
slowly dissolved. Letting the mixture rest overnight afforded
large block-shaped crystals of the product in 73 % yield. A series
of 31P NMR spectra was recorded to monitor the progress of the
isomerization. The spectra show that almost no side products
were formed in the MeCN/CH2Cl2 mixture (Figure S3, Supporting
Information). Using this improved protocol, compound 2 could be
synthesized on multi-gram scale.
In a first series of experiments, the reactivity of the NP3 ring
system 2 towards the Lewis base DMAP (4-dimethylamino-
pyridine) was investigated. As previously shown, the reaction of
DMAP with the P4 ring system [ClP(μ-PMes*)]2 (3, type C) led to
rearrangement of the P–P bonding system and elimination of
Mes*PCl2, yielding the tricyclic hexaphosphane 5 in nearly
quantitative yields (Scheme 5).[33] It was therefore of interest to
see if a similar reaction behaviour could be observed in case of
the NP3 ring system 2, possibly yielding an analogous tricyclic
structure with an N-capped P4 ring system. Astonishingly,
though, the isolated product of the reaction of 2 with DMAP was
the same tricyclic P6 system (5), indicating that the N atom had
formally been eliminated from the starting material (Scheme 6).
To shed light on the reaction path, in situ 31P NMR spectra were
collected, which showed that Mes*PCl2 and presumably
Mes*NPCl⋅DMAP (pp. S19ff) were formed as further products of
the reaction. Hence, the formation of 5 can be understood in
terms of a formal cycloreversion of 2, that is, the products can
formally be derived from the monomeric building blocks of 2,
Mes*NPCl and “Mes*PPCl” (i.e. [ClP(μ-PMes*)]2, 3; Scheme 6).
However, as no intermediates besides 4 could be observed in
the 31P NMR spectra (Figure S6), the exact mechanism of the
reaction remains unclear.
Scheme 3. Reactivity of C (R = sterically demanding substituent, LB = Lewis
base; LA = Lewis acid; X = Cl; Y = C6F5, see below).
The apparent differences in reactivity prompted us to
investigate NP3 ring systems (B, Scheme 1),[36] which
incorporate only one nitrogen atom for electronic stabilization
and can formally be regarded as a blend of N2P2 and P4 ring
systems. Particularly, we were interested in the reactivity of ring
system B towards Lewis acids, Lewis bases, nucleophiles, as
well as reducing agents. The synthesis of the NP3 ring system
It is worthy to note that the reaction of the acyclic NP3 compound
1 with DMAP led to a similar outcome, which indicates
conversion of 1 to 2 under reaction conditions.
[Mes*N(μ-PCl)2PMes*] (2; Mes*
=
2,4,6-tri-t-butylphenyl,
Scheme 4) was recently published[36] and served as a starting
point for the investigations reported in this paper.
Scheme 5. Reaction of 3 with DMAP (R = Mes*).[33]
Scheme 4. Synthesis of 2 (R = Mes*) starting from Mes*NPCl[37] (a: Mes*PH2,
NEt3; b: nBuLi, PCl3, −80 °C; c: THF).
Results and Discussion
To begin with, the synthesis of 2 could be optimized using a
MeCN/CH2Cl2 mixture instead of THF for the isomerization step
(Scheme 4c). As reported previously, the isomerization of 1 can
lead to different products depending on the polarity of the
solvent.[36] Since more polar media favour the formation of the
desired ring system 2, MeCN seemed a reasonable choice of
solvent; however, 1 and 2 were only sparingly soluble in MeCN.
Scheme 6. Reaction of 2 with DMAP (R = Mes*).
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
polarized NP double bond (δ(PX) = +446.4 ppm; cf. 7: +358.9,[32]
10: +366.6 ppm[23]). All observed NMR shifts and coupling
constants are in good agreement with theoretical data (Table 1).
Table 1. Experimental 31P NMR data of 6 (AMX spin system). Calculated
values (PBE0-D3/def2-SVP, cf. SI) are given in brackets.
J [Hz]
δ [ppm]
A
M
A
M
X
114.1 (105.7)
175.4 (126.3)
446.4 (442.7)
−317 (−273)
+107 (+69)
−470 (−442)
Scheme 7. Reaction of NP3 ring system 2 with GaCl3 (R
= Mes*), in
comparison with the reactivities of P4 and N2P2 ring systems 3 (R = Mes*)[32]
and 9 (R = Ter = 2,6-dimesitylphenyl).[23]
Contrary to its N2P2 and P4 congeners 7 and 10, the NP3
species 6 decomposed to an unidentified mixture of products
upon warming. At first glance, this might seem unexpected,
especially in view of the fact that the diazadiphosphenium salt
10 is perfectly stable up to well beyond 200 °C.[23] In case of the
NP3 species 6, the single N atom is apparently not sufficient to
stabilize the formal phosphenium centre by donation of π-
electron density into the vacant p-orbital at P. On the other hand,
while the P4 system 7 can stabilize itself by formation of a
transannular P–P bond and concomitant pyramidalization of all
P atoms, the same is not true of the NP3 ring system 6, as this
would entail a high ring strain at the N atom, which is
energetically unfavourable.[45,46] Consequently, the bicyclic NP3
isomer 11 is calculated to be much higher in energy than the
phosphenium isomer 6 (Scheme 8), whereas the isomerization
of the P4 system 7 to the bicyclic phosphino-phosphonium salt 8
is thermodynamically favoured.
In a next set of experiments, the reactivity of the NP3 ring
system 2 towards Lewis acids was investigated. As GaCl3 had
proved to be
a suitable reagent for selective chloride
abstraction,[38–44] it was chosen as a model substrate to
investigate this type of reaction. As expected, low temperature
31P NMR spectroscopy revealed that abstraction of one chloride
ion from the ring system led to the formation of a phosphenium
salt (6; Scheme 7, top), which is analogous to those observed in
case of the congeneric N2P2 and P4 ring systems (7, 10; Scheme
7).
Scheme 8. Calculated Gibbs energies for the isomerization of the cations of
compounds 6 and 7 (PBE0-D3/def2-SVP).
Figure 1. Experimental and simulated 31P NMR spectrum (−70 °C) of the
reaction between 2 and GaCl3 showing the formation of phosphenium salt 6
(R = Mes*; starting material and by-products indicated by asterisks).
To better understand the differences in bonding within the
cations of 6, 7, and 10, Natural Resonance Theory (NRT)
calculations on model cations (with R = H) were performed. For
each species, a total of about 45 resonance structures was
considered, most of which describe bond polarization within the
ring or negative hyperconjugation of the LPs at Cl into the ring
system. To simplify the discussion, only the two most important
contributions to the Lewis resonance scheme, as well as the
Similarly to tetraphosphenium salt 7,[32] a solution of the
azatriphosphenium salt 6 was only stable at low temperatures
below −40 °C, as evidenced by variable temperature NMR
spectra (Figure S8). In the 31P NMR spectrum, 6 was identified
by an AMX spin system (Figure 1) with a characteristic,
downfield-shifted X part, due to the positive charge and strongly
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
Lewis structure with an electron sextet at the formal
phosphenium centre will be considered in the following (Scheme
9). Two trends can be derived, which underline the observed
reactivity: Firstly, the overall contribution of the two most
important Lewis structures (with NP or PP double bonds,
respectively) decreases along the series N2P2, NP3, P4; that is
the formal phosphenium is best stabilized by π-type interactions
in the N2P2 cation, and least stabilized in the P4 derivative.
Secondly, the weight of the resonance structure with an electron
sextet at the formal phosphenium increases from N2P2 to P4,
formally rendering the latter the most reactive derivative. This is
corroborated by the fact that the isomerization of the
tetraphosphenium salt 7 to the bicyclic isomer 8 proceeded even
at −80 °C.[32]
unsuccessful. In two instances, a few crystals of two different
decomposition products could be isolated and studied by single
crystal X-ray diffraction (SC-XRD). In both cases, a tBu group of
the Mes* moiety had been transferred onto another molecular
fragment or solvent molecule, indicating decomposition of the
sterically demanding substituent (Figure 2). Cleavage of tBu
groups from Mes* substituents is not uncommon in the presence
of Lewis acids (or by thermal treatment),[47–51] thus
demonstrating the high Lewis acidity of salt 6.
Next, we were interested in demonstrating nucleophilic
substitution at the NP3 ring system 2. By analogy with the
congeneric N2P2 ring system [ClP(μ-NDipp)]2 (Dipp = diiso-
propylphenyl),[52] the reaction of 2 with AgC6F5 resulted in
precipitation of AgCl and formation of [Mes*N(μ-PC6F5)2PMes*]
(12, Scheme 10). After filtration, the latter could be crystallized
from CH2Cl2/MeCN to give yellow, block shaped crystals suitable
for SC-XRD (Figure 3, left; yield of isolated substance: 48 %).
Scheme 10. Reaction of 2 with AgC6F5 (R = Mes*). Analogous reactions were
observed for congeneric N2P2[52] and P4 species.
Scheme 9. Some Lewis resonance structures of model phosphenium cations
(R = H) as well as their respective weights from NRT analysis.
Figure 3. Molecular structures of 12 (left) and 13 (right) in the crystal.
Ellipsoids are set at 50 % probability (123 K and 173 K, respectively). Selected
bond lengths (Å) and angles (°): 12 P1–P2 2.2461(6), P1–P3 2.2532(6), P2–
N1 1.744(1), P2–C37 1.884(2), P3–N1 1.742(1), P3–C43 1.888(2), P2–P1–P3
76.09(2), N1–P2–P1 88.29(5), N1–P3–P1 88.12(5), P3–N1–P2 105.39(7), P1–
P2–P3–N1 −164.04(8); 13 P1–P2 2.2471(8), P1–P4 2.2611(9), P2–C19
1.853(2), P2–P3 2.1949(9), P3–P4 2.2367(8), P4–C43 1.858(2), P2–P1–P4
84.37(3), P3–P2–P1 88.99(3), P2–P3–P4 86.17(3), P3–P4–P1 87.61(3), P1–
P2–P4–P3 −142.45(4).
Figure 2. Molecular structures of the decomposition products [MeCNtBu]
[GaCl4] (left) and [Mes*N(H)PCl2tBu][GaCl4] (right) in the crystal. Ellipsoids are
set at 50 % probability (123 K). Selected bond lengths (Å) and angles (°):
[CH3CNtBu][GaCl4] N1–C1 1.479(3), N1–C5 1.129(3), C5–N1–C1 178.2(2),
N1–C5–C6 180.0(3), C1–N1–C5–C6 180.0; [Mes*N(H)PCl2tBu][GaCl4] N1–P1
1.604(2), N1–H1 0.77(3), P1–C19 1.815(3), P1–Cl1 1.9798(9), N1–P1–C19
112.1(1), N1–P1–Cl1 111.40(9), C19–P1–Cl1 109.8(1), C1–N1–P1–C19
−179.4(2).
The molecular structure of 12 revealed a flattened NP3 ring
system (fold angle:[29,53] 164.04(8)°) with NP and PP bond
lengths that correspond to somewhat shortened NP and slightly
elongated PP single bonds, respectively (cf. Σrcov(N–P) =
As with the P4 congener 7, all attempts to crystallize the
azatriphosphenium salt
6 at low temperatures remained
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
1.82 Å; . Σrcov(P–P) = 2.22 Å).[54] The nitrogen atom is located in
a nearly planar coordination environment (Σ(∡N) = 352.8(3)°),
whereas the phosphorus atom P1 is strongly pyramidalized
(Σ(∡P) = 277.2(1)°). These structural parameters compare well
with those of the starting material (2).[36]
interesting to explore which type of stabilization (planarization vs.
pyramidalization) would predominate in the NP3 case.
Since the reactivity of the P4 ring system 3 towards AgC6F5
had not been reported previously, we also treated a solution of 3
with AgC6F5, resulting in the corresponding P4 ring system
[F5C6P(μ-PMes*)]2 (13, Figure 3, right). Its crystal structure could
be determined by SC-XRD. While the overall molecular structure
is quite similar to that of its NP3 congener (12), compound 13
exhibits two Mes*-bound P atoms in a pyramidal coordination
environment. Hence, the P4 ring system adopts a much more
puckered conformation with a fold angle of 142.45(4)°, which
corresponds nicely to the experimental fold angle of the P4 ring
system in the starting material 3 (120–143° depending on the
modification).[29]
Both 12 and 13 displayed complex heteronuclear coupling
patterns in the 31P and 19F NMR spectra. Additionally, the NP3
ring system 12 exhibited various broadened signals due to
hindered rotation of the C6F5 substituents (Figure S9). It is
worthy to note that a yellow solution of 12 in CH2Cl2 started to
turn green after one day, which was accompanied by the
appearance of additional signals in the 31P NMR spectrum.
These signals could be assigned to the [2+2] cycloreversion
products of the NP3 ring system 12, i.e. (Z)-Mes*P=PC6F5
(δ(31P) = 381.2, 568.7 ppm; J = −557 Hz) and (Z)-Mes*N=PC6F5
(δ(31P) = 361.7 ppm)[52] (Scheme 11, see also Figure S10).
Computed 31P NMR data corroborate the assignment
Scheme 12. Reduction of N2P2 (9) and P4 ring systems (3) leads to different
types of products, namely the open shell singlet biradical 14 in case of E = N,
and the closed shell [1.1.0]-bicyclic species 15 in case of E = P.
To that end, compound 2 was reduced using activated Mg
chips, by analogy with the reduction of 3 and 9.[26,34,56] This led to
a mixture of products, which – to our surprise – contained the
bicyclic tetraphosphane Mes*P4Mes* (15)[57] as one of the main
components, indicating a formal cycloreversion of the NP3 ring
system into NP and PP fragments. To further investigate the
formation of the bicyclotetraphosphane 15, Cp2Ti(BTMSA)[58]
(BTMSA = bis(trimethylsilyl)acetylene) was employed as a
milder reducing agent. Indeed, this procedure facilitated the
isolation of a few crystals of an unusual N2P6 cage compound
with a bicyclo[1.1.1]pentaphosphane scaffold (16, Scheme 13).
Compound 16 may be regarded as a dimer of the putative
biradical Mes*NP3Mes* (17, p. S62f), which is predicted to be
only slightly less stable than its [1.1.0]-bicyclic isomer (18, ΔG° =
14.2 kJ/mol at DLPNO-CCSD(T)/def2-TZVP//PBE-D3/def2-SVP
level of theory, see below). Formal dimerization of congeneric
N2P2 biradicals was already reported for species with small
substituents R, leading to the formation of α- or β-cage
structures with an N4P4 scaffold (Scheme 14).[59–61]
((Z)-Mes*P=PC6F5:
373.7,
525.8 ppm,
−507 Hz;
(Z)-Mes*N=PC6F5: 372.8 ppm). The same type of reactivity was
previously discussed for “symmetric” N2E2 (E = pnictogen) ring
systems that can be regarded as dimers of iminopnictanes
(RN=ER’).[21,40,55] The equilibrium between monomeric and
dimeric species was shown to depend on the size of the
substituents R and R’.
Scheme 13. Formally, cage compound 16 (R = Mes*) can be derived from a
β-cage-type structure, which itself may be viewed as a dimer of Mes*NP3Mes*.
Scheme 11. Cycloreversion of 12 in solution (R = Mes*). Due to the blue
colour of (Z)-Mes*N=PC6F5, the mixture appears green. The equilibrium ratio
between the ring system 12 and the cycloreversion products is about 5:1:1.
Lastly, the reduction of the NP3 ring system 2 was of
particular interest, especially when considering that the
congeneric N2P2 species 9 and P4 species 3 yielded two very
different reduction products (Scheme 12). While the former can
be reduced to a singlet biradical without a transannular PP bond
(14),[26] reduction of the latter leads selectively to the formation
of a P4 butterfly with a transannular PP bond (15).[34] This can be
understood in terms of π-electron delocalization, as detailed
above, or ring strain at the nitrogen atom which prevents the
N2P2 species from pyramidalization.[46] In any case, it seemed
Scheme 14. With small substituents (e.g. R = tBu), the reduction of N2P2 ring
systems leads to the formation of N4P4 cage compounds (left: α-, right: β-cage)
that can be regarded as dimers of the respective biradicals [P(μ-NR)]2.[59–61]
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
The molecular structure of 16 could be elucidated by SC-
XRD (Figure 4). All PP bond lengths within the bicyclo[1.1.1]-
pentaphosphane scaffold correspond to slightly elongated PP
single bonds (cf. Σrcov(P–P) = 2.22 Å).[54] The N1–P6 distance
corresponds to a polarized NP single bond, whereas the N2–P6
distance is in the range of a typical NP double bond (cf. Σrcov(N–
P) = 1.82, Σrcov(N=P) = 1.62 Å). It is worthy to note that the bond
angles at the P atoms within the P5 cage are quite small (P1, P2,
P3: av. 77.8(2)°; P4, P4: av. 84(1)°), as expected for a polycyclic
Pn structure. When viewed along the P4–P5 axis, all substituents
are bent to the left, thus minimizing the repulsion between the
LPs at P1, P2, and P3. To the best of our knowledge, compound
16 is the first example of a bicyclo[1.1.1]pentaphosphane
derivative.
shifts and coupling constants, which were extracted from the
spectrum by line-shape fitting, correspond very well to calculated
NMR data (Table 2).
Figure 5. 31P NMR spectrum of a solution containing cage compound 16. The
experimental spectrum (up) shows some impurities due to the instability of 16.
The simulated spectrum (down) was fitted to the most intense signals. A
detailed view of all signals is presented at the top (lines caused by impurities
are indicated by an asterisk).
Table 2. Experimental 31P NMR data of 16 (ABGM2X spin system). Calculated
values (PBE0-D3/def2-TZVP, cf. SI) are given in brackets.
Figure 4. Molecular structure of 16 in the crystal. Ellipsoids are set at 50 %
probability (123 K). Selected bond lengths (Å) and angles (°): P1–P4 2.245(1),
P1–P5 2.274(1), P2–P4 2.250(1), P2–P5 2.276(1), P3–P4 2.270(1), P3–P5
2.225(1), P3–N1 1.759(3), P6–N1 1.702(3), P6–N2 1.544(3), P4–P1–P5
77.66(4), P4–P2–P5 77.50(4), P5–P3–P4 78.14(4), P1–P4–P2 84.80(4), P1–
P4–P3 81.23(4), P2–P4–P3 88.27(5), P3–P5–P1 81.58(4), P3–P5–P2
88.72(5), P1–P5–P2 83.52(4), N1–P3–P4 116.2(1), N1–P3–P5 112.2(1), P6–
N1–P3 121.4(2), N2–P6–N1 110.6(2), P1–P4–P5–P2 118.59(6), P1–P4–P5–
P3 −113.97(6), P2–P4–P5–P3 127.44(6), P3–N1–P6–N2 −167.9(2).
J [Hz]
δ [ppm]
A
B
G
M
−25.0
(−19.0)
A
−21.0
(−26.6)
−4
(−14)
B
26.7
(29.9)
−6
(−22)
−3
(−9)
G
M2
X
In our attempts to fully characterize compound 16, it
transpired that this species was rather unstable and only of
intermediary nature. Several NMR experiments were run using
isolated crystals of compound 16 as well as the reaction mixture
of 2 and Cp2Ti(BTMSA) (Figure S13). All experiments eventually
indicated the formation of a similar product mixture as observed
for the reduction of 2 with Mg. Hence, the formation of
Mes*P4Mes* can be rationalized by formal cleavage of the P3–
P4 and P3–P5 bonds in compound 16.
200.7
(189.9)
−185
(−132)
−136
(−83)
−202
(−160)
281.3
(283.7)
−14
(−15)
+224
(+288)
−28
(−36)
+19
(+47)
Natural bond orbital (NBO) and natural localized molecular
Despite its inherent instability, compound 16 could be
identified by 31P NMR spectroscopy in a solution that was freshly
prepared from isolated crystals. The six phosphorus atoms
displayed an ABGM2X spin system (Figure 5), with typical NMR
shifts for the Mes*-substituted P atoms (−26.0, −21.0 ppm)[29]
and a downfield-shifted X part corresponding to the P atom
involved in the NP double bond (281.3 ppm). The experimental
orbital (NLMO) analysis[62,63] of 16 revealed that all P atoms of
the P5 scaffold are connected by localized σ-type bonds, and
that each atom possesses one LP that is mainly localized in an s
orbital (s-character: 65–69 %). The Wiberg bond indices of the
PP bonds range from 0.90 to 0.95, indicating typical PP single
bonds in agreement with experimental structural data. The
exocyclic NPN scaffold comprises an NP double bond (N2=P6)
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
and a p-type LP at N1, which interacts slightly with the anti-
bonding π* orbital (donor-acceptor energy E(2) = 106.6 kJ/mol).
Again, these findings are in line with the experimental structure.
The stability of different P5Hn (n = 0⋯7) structures has been
investigated theoretically and experimentally in a number of
publications.[64–71] Still, the bicyclo[1.1.1]pentaphosphane motif
was considered only in a single publication, which identified the
minimum energy structure of P5H3 as bicyclo[2.1.0]penta-
phosphane (19a).[68] This is in agreement with our own results,
which show that bicyclo[1.1.1]pentaphosphane (19c) is in fact
the least favoured isomer of those previously discussed in the
literature (Scheme 15, see also p. S40).
The biradical character of H2NP3 was computed to be 31 %,
which is slightly larger than the biradical character of its N2P2
congener (23 %; cf. [P(μ-NTer)]2 (14): 27 %[72]). The predicted
biradical character of the P4 species is still smaller (16 %), which
can be attributed to a different through-bond interaction in the
absence of a N atom.[73,74] Overall, the biradical character of all
three species is moderate and compares with other pnictogen-
based biradicals.[75–79]
Conclusions
In conclusion, it was shown that the chemistry of the NP3
ring system 2 systematically expands the known chemistry of
congeneric N2P2 and P4 ring systems. Some similarities with
these known compounds notwithstanding, we could observe
some unexpected reaction behaviour, such as the formation of a
bicyclo[1.1.1]pentaphosphane derivative (16) or the formation of
products with formal [NP]n and [PP]n composition due to formal
cycloreversion of the NP3 ring system (e.g. Mes*P6Mes* (5),
Mes*NPCl, or (Z)-Mes*PPC6F5; cf. Scheme 17).
Scheme 15. Different isomers of P5H3 and their relative Gibbs energies (ΔG°
in kJ/mol, DLPNO-CCSD(T)/def2-TZVP//PBE0-D3/def2-TZVP).
Since we did not find clear experimental evidence whether
the reduction of the NP3 ring system 2 led to intermediary
formation of the biradical (17) or bicyclic isomer (18) of
Mes*NP3Mes* (cf. SI, pp. S35f), DFT and ab-initio calculations
were performed to compare both isomers (for details on
computations, please refer to the SI). As already indicated, the
bicyclic structure 18 is energetically slightly favoured (ΔG° =
14.2 kJ/mol). When disregarding effects of the bulky substituents,
i.e. using H2NP3 as a model compound, the difference in energy
between both isomers is somewhat more pronounced (ΔG° =
40.2 kJ/mol, Scheme 16). By comparison, the energetic
difference between biradical and bicyclic structure of H2P4
amounts to 118.6 kJ/mol, rendering the hypothetical H2P4
biradical very unstable. In contrast, the H2N2P2 biradical is
substantially more stable than the bicyclic structure (ΔG° =
−76.2 kJ/mol), in agreement with earlier considerations and
experimental observations.[26,46]
Scheme 17. Summarized reactivity of the NP3 ring system 2.
In particular, it was demonstrated that chloride abstraction
from 2 using a Lewis acid such as GaCl3 resulted in the
formation of
a highly labile azatriphosphenium salt (6).
Substitution of the Cl atoms with C6F5, on the other hand, gave a
rather stable NP3 ring system (12), which underwent partial
cycloreversion in solution to yield a diphosphene and an
iminophosphane. Most intriguingly, the reduction of 2 afforded a
bicyclo[1.1.1]pentaphoshane (16), an as yet uninvestigated
substance class.
Comprehensive theoretical studies were performed to
understand the differences and similarities between N2P2, NP3
and P4 ring systems. One main factor that governs the stability
of the products is the electronic structure of the lone pairs of
electrons at N vs. P: The p-type lone pair at N can easily
delocalize into the ring system, resulting in resonance
Scheme 16. Comparison of N2P2, NP3 and P4 ring systems with respect to the
relative stabilities of biradical and bicyclic isomers (ΔG° in kJ/mol, DLPNO-
CCSD(T)/def2-TZVP//PBE-D3/def2-SVP).
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
[6]
[7]
[8]
X. Ren, P. Lian, D. Xie, Y. Yang, Y. Mei, X. Huang, Z. Wang, X. Yin, J.
Mater. Sci. 2017, 52, 10364–10386.
stabilization of highly reactive species, whereas the s-type lone
pair at P is rather unsuited for this kind of interaction. Moreover,
the ring strain at N vs. P plays a role in the stability of different
ring systems, as previously detailed elsewhere.[45,46]
In consequence, N2P2 derivatives tend to form planar ring
systems (often involving electron delocalization), while P4
systems may stabilize themselves by intramolecular bond
formation (bicyclic structures). The investigated reactivity of the
NP3 ring system 2 implies that, in case none of the former types
of stabilization predominate, cycloreversion becomes important
as an alternative pathway of energy gain. This is often
associated with a mixture of products, rendering the isolation of
pure substances a challenge.
D. Voet, J. G. Voet, C. W. Pratt, Fundamentals of Biochemistry: Life at
the Molecular Level, John Wiley & Sons, Hoboken, NJ, 2016.
K. B. Dillon, F. Mathey, J. F. Nixon, Phosphorus: The Carbon Copy:
From Organophosphorus to Phospha-Organic Chemistry, John Wiley &
Sons Ltd, West Sussex, 1998.
[9]
O. J. Scherer, Comments Inorg. Chem. 1987, 6, 1–22.
[10] M. Baudler, K. Glinka, Chem. Rev. 1993, 93, 1623–1667.
[11] M. Scheer, G. Balázs, A. Seitz, Chem. Rev. 2010, 110, 4236–4256.
[12] N. A. Giffin, J. D. Masuda, Coord. Chem. Rev. 2011, 255, 1342–1359.
[13] G. He, O. Shynkaruk, M. W. Lui, E. Rivard, Chem. Rev. 2014, 114,
7815–7880.
[14] M. Donath, F. Hennersdorf, J. J. Weigand, Chem. Soc. Rev. 2016, 45,
1145–1172.
[15] J. E. Borger, A. W. Ehlers, J. C. Slootweg, K. Lammertsma, Chem. Eur.
J. 2017, 11738–11746.
[16] M. S. Balakrishna, V. S. Reddy, S. S. Krishnamurthy, J. F. Nixon, J. C.
T. R. B. St. Laurent, Coord. Chem. Rev. 1994, 129, 1–90.
[17] L. Stahl, Coord. Chem. Rev. 2000, 210, 203–250.
[18] G. G. Briand, T. Chivers, M. L. Krahn, Coord. Chem. Rev. 2002, 233–
234, 237–254.
Experimental Section
All manipulations were carried out under oxygen- and moisture-free
conditions in an inert atmosphere of argon, using standard Schlenk or
Drybox techniques. For detailed synthetic protocols, analytic data and
experimental spectra please refer to the Supporting Information.
[19] M. S. Balakrishna, D. J. Eisler, T. Chivers, Chem. Soc. Rev. 2007, 36,
650–664.
[20] M. S. Balakrishna, Dalton Trans. 2016, 45, 12252–12282.
[21] N. Burford, T. S. Cameron, C. L. B. Macdonald, K. N. Robertson, R. W.
Schurko, D. Walsh, R. McDonald, R. E. Wasylishen, Inorg. Chem. 2005,
44, 8058–8064.
Supplementary crystallographic data for this paper can be found online
under CCDC 1939289–1939295. These data are provided free of charge
by the Cambridge Crystallographic Data Centre.
[22] R. J. Davidson, J. J. Weigand, N. Burford, T. S. Cameron, A. Decken, U.
Werner-Zwanziger, Chem. Commun. 2007, 4671–4673.
[23] D. Michalik, A. Schulz, A. Villinger, N. Weding, Angew. Chem. Int. Ed.
2008, 47, 6465–6468.
Computations were performed using the programs Gaussian09[80] and
Orca 4.1.1.[81] Structure optimizations employed the pure density
functional PBE or hybrid functional PBE0,[82–84] in conjunction with
Grimme’s dispersion correction D3(BJ).[85,86] For more accurate single-
point energies, the DLPNO-CCSD(T)[87–89] method was applied. All
calculations used the basis sets def2-SVP or def2-TZVP.[90] Detailed
information on all calculations is given in the Supporting Information.
[24] M. H. Holthausen, J. J. Weigand, J. Am. Chem. Soc. 2009, 131,
14210–14211.
[25] R. Kuzora, A. Schulz, A. Villinger, R. Wustrack, Dalton Trans. 2009,
9304–9311.
[26] T. Beweries, R. Kuzora, U. Rosenthal, A. Schulz, A. Villinger, Angew.
Chem. Int. Ed. 2011, 50, 8974–8978.
[27] N. Wiberg, A. Wörner, H.-W. Lerner, K. Karaghiosoff, Z. Naturforsch. B
2002, 57, 1027–1035.
Acknowledgements
[28] A. Lorbach, A. Nadj, S. Tüllmann, F. Dornhaus, F. Schödel, I. Sänger,
G. Margraf, J. W. Bats, M. Bolte, M. C. Holthausen, et al., Inorg. Chem.
2009, 48, 1005–1017.
We are grateful for financial support by the DFG (SCHU
1170/11-1). We thank the University of Rostock for access to the
cluster computer, and especially Malte Willert for his assistance
with the queueing system and software installations. Dr Fabian
[29] J. Bresien, C. Hering, A. Schulz, A. Villinger, Chem. Eur. J. 2014, 20,
12607–12615.
[30] A. Hinz, A. Schulz, A. Villinger, Inorg. Chem. 2016, 55, 3692–3699.
[31] J. Bresien, A. Schulz, A. Villinger, Phosphorus Sulfur Silicon Relat.
Elem. 2016, 191, 601–604.
Reiß is gratefully acknowledged for
Cp2Ti(BTMSA).
a generous gift of
[32] J. Bresien, K. Faust, A. Schulz, A. Villinger, Angew. Chem. Int. Ed.
2015, 54, 6926–6930.
Keywords: Phosphorus • ring systems • PN chemistry •
biradicals • computational chemistry
[33] J. Bresien, A. Schulz, A. Villinger, Chem. Eur. J. 2015, 21, 18543–
18546.
[34] J. Bresien, K. Faust, C. Hering-Junghans, J. Rothe, A. Schulz, A.
Villinger, Dalton Trans. 2016, 45, 1998–2007.
[1]
[2]
D. E. C. Corbridge, Phosphorus: Chemistry, Biochemistry and
Technology, CRC Press, Boca Raton, FL, 2016.
[35] J. Bresien, A. Schulz, A. Villinger, Dalton Trans. 2016, 45, 498–501.
[36] J. Bresien, A. Hinz, A. Schulz, T. Suhrbier, M. Thomas, A. Villinger,
Chem. Eur. J. 2017, 23, 14738–14742.
T. S. George, C. D. Giles, D. Menezes-Blackburn, L. M. Condron, A. C.
Gama-Rodrigues, D. Jaisi, F. Lang, A. L. Neal, M. I. Stutter, D. S.
Almeida, et al., Plant Soil 2018, 427, 191–208.
[37] E. Niecke, M. Nieger, F. Reichert, Angew. Chem. Int. Ed. Engl. 1988,
27, 1715–1716.
[3]
P. C. J. Kamer, P. W. N. M. van Leeuwen, Eds. , Phosphorus(III)
Ligands in Homogeneous Catalysis: Design and Synthesis, John Wiley
& Sons Ltd, Chichester, UK, 2012.
[38] N. Burford, J. A. C. Clyburne, P. K. Bakshi, T. S. Cameron,
Organometallics 1995, 14, 1578–1585.
[4]
[5]
M. C. Simpson, J. D. Protasiewicz, Pure Appl. Chem. 2013, 85, 801–
815.
[39] N. Burford, T. S. Cameron, J. A. C. Clyburne, K. Eichele, K. N.
Robertson, S. Sereda, R. E. Wasylishen, W. A. Whitla, Inorg. Chem.
1996, 35, 5460–5467.
T. Low, A. S. Rodin, A. Carvalho, Y. Jiang, H. Wang, F. Xia, A. H.
Castro Neto, Phys. Rev. B 2014, 90, 075434.
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
[40] N. Burford, T. S. Cameron, K. D. Conroy, B. Ellis, M. D. Lumsden, C. L.
B. Macdonald, R. McDonald, A. D. Phillips, P. J. Ragogna, R. W.
Schurko, et al., J. Am. Chem. Soc. 2002, 124, 14012–14013.
[41] A. Villinger, P. Mayer, A. Schulz, Chem. Commun. 2006, 1236–1238.
[42] P. Mayer, A. Schulz, A. Villinger, J. Organomet. Chem. 2007, 692,
2839–2842.
[66] W. W. Schoeller, J. Mol. Struct.: THEOCHEM 1993, 284, 61–66.
[67] M. Häser, J. Am. Chem. Soc. 1994, 116, 6925–6926.
[68] M. Häser, Phosphorus Sulfur Silicon Relat. Elem. 1994, 93, 235–239.
[69] M. N. Glukhovtsev, A. Dransfeld, P. von R. Schleyer, J. Phys. Chem.
1996, 100, 13447–13454.
[70] M. D. Chen, R. B. Huang, L. S. Zheng, Q. E. Zhang, C. T. Au, Chem.
Phys. Lett. 2000, 325, 22–28.
[43] J. J. Weigand, M. Holthausen, R. Fröhlich, Angew. Chem. Int. Ed. 2009,
48, 295–298.
[71] T. Xue, J. Luo, S. Shen, F. Li, J. Zhao, Chem. Phys. Lett. 2010, 485,
26–30.
[44] M. Kuprat, A. Schulz, A. Villinger, Angew. Chem. Int. Ed. 2013, 52,
7126–7130.
[72] J. Bresien, A. Hinz, A. Schulz, A. Villinger, Dalton Trans. 2018, 47,
4433–4436.
[45] J. M. Mercero, X. Lopez, J. E. Fowler, J. M. Ugalde, J. Phys. Chem. A
1997, 101, 5574–5579.
[73] M. Abe, Chem. Rev. 2013, 113, 7011–7088.
[74] J. Bresien, A. Schulz, L. Szych, A. Villinger, R. Wustrack, Dalton Trans.
2019, DOI 10.1039/C9DT01654F.
[46] R. Grande-Aztatzi, J. M. Mercero, J. M. Ugalde, Phys. Chem. Chem.
Phys. 2016, 18, 11879–11884.
[47] A. Schulz, A. Villinger, Angew. Chem. Int. Ed. 2008, 47, 603–606.
[48] Y. Okamoto, H. Shimizu, J. Am. Chem. Soc. 1968, 90, 6145–6148.
[49] C. M. D. Komen, F. Bickelhaupt, Synth. Commun. 1996, 26, 1693–1697.
[50] F. Rivière, S. Ito, M. Yoshifuji, Tetrahedron Lett. 2002, 43, 119–121.
[51] C. G. E. Fleming, A. M. Z. Slawin, K. S. Athukorala Arachchige, R.
Randall, M. Bühl, P. Kilian, Dalton Trans. 2013, 42, 1437–1450.
[52] M. Kuprat, M. Lehmann, A. Schulz, A. Villinger, Inorg. Chem. 2011, 50,
5784–5792.
[75] A. Hinz, A. Schulz, A. Villinger, J. Am. Chem. Soc. 2015, 137, 9953–
9962.
[76] A. Hinz, A. Schulz, A. Villinger, Angew. Chem. Int. Ed. 2015, 54, 2776–
2779.
[77] A. Hinz, A. Schulz, A. Villinger, Angew. Chem. Int. Ed. 2015, 54, 668–
672.
[78] A. Hinz, A. Schulz, A. Villinger, J.-M. Wolter, J. Am. Chem. Soc. 2015,
137, 3975–3980.
[53] W. W. Schoeller, T. Busch, Chem. Ber. 1991, 124, 1369–1371.
[54] P. Pyykkö, M. Atsumi, Chem. Eur. J. 2009, 15, 12770–12779.
[55] M. Lehmann, A. Schulz, A. Villinger, Struct. Chem. 2011, 22, 35–43.
[56] A. Hinz, R. Kuzora, U. Rosenthal, A. Schulz, A. Villinger, Chem. Eur. J.
2014, 20, 14659–14673.
[79] J. Bresien, T. Kröger-Badge, S. Lochbrunner, D. Michalik, H. Müller, A.
Schulz, E. Zander, Chem. Sci. 2019, 10, 3486–3493.
[80] Gaussian 09, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Peterson, et al., Gaussian, Inc., Wallingford CT,
2013.
[57] E. Niecke, R. Rüger, B. Krebs, Angew. Chem. Int. Ed. Engl. 1982, 21,
544–545.
[81] F. Neese, WIREs Comput. Mol. Sci. 2018, 8, e1327.
[82] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865–
3868.
[58] U. Rosenthal, V. V. Burlakov, in Titanium and Zirconium in Organic
Synthesis, Wiley-VCH, Weinheim, 2003, pp. 355–389.
[59] D. DuBois, E. N. Duesler, R. T. Paine, J. Chem. Soc., Chem. Commun.
1984, 4, 488–489.
[83] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1997, 78, 1396–
1396.
[60] H. Bladt, S. Gonzalez Calera, J. M. Goodman, R. J. Less, V. Naseri, A.
Steiner, D. S. Wright, Chem. Commun. 2009, 6637–6639.
[61] A. Hinz, R. Kuzora, A.-K. Rölke, A. Schulz, A. Villinger, R. Wustrack,
Eur. J. Inorg. Chem. 2016, 2016, 3611–3619.
[84] C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158–6170.
[85] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[86] S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32, 1456–
1465.
[62] E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter, J. A.
Bohmann, C. M. Morales, C. R. Landis, F. Weinhold, NBO 6.0,
Theoretical Chemistry Institute, University of Wisconsin, Madison, 2013.
[63] F. Weinhold, C. R. Landis, E. D. Glendening, Int. Rev. Phys. Chem.
2016, 35, 399–440.
[87] C. Riplinger, F. Neese, J. Chem. Phys. 2013, 138, 034106.
[88] D. G. Liakos, M. Sparta, M. K. Kesharwani, J. M. L. Martin, F. Neese, J.
Chem. Theory Comput. 2015, 11, 1525–1539.
[89] C. Riplinger, P. Pinski, U. Becker, E. F. Valeev, F. Neese, J. Chem.
Phys. 2016, 144, 024109.
[64] M. Baudler, H. Ständeke, M. Borgardt, H. Strabel, J. Dobbers,
Naturwissenschaften 1966, 53, 106–106.
[90] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–305.
[65] M. Baudler, Z. Chem. 1984, 24, 352–365.
This article is protected by copyright. All rights reserved.

10.1002/chem.201903410
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
News on NP3 ring systems: The
reactivity of chlorinated NP3 ring
systems was investigated by
experimental and computational
methods. Comparisons with known
N2P2 and P4 congeners are drawn to
obtain a systematic understanding of
the reactivity of these types of
compounds.
Jonas Bresien,* Liesa Eickhoff, Axel
Schulz,* Tim Suhrbier, and Alexander
Villinger
Page No. – Page No.
A systematic survey of the reactivity
of chlorinated N2P2, NP3 and P4 ring
systems
This article is protected by copyright. All rights reserved.