Rational synthesis and mutual conversion of Bis-N-heterocyclic diphosphanes and secondary N-heterocyclic phosphanes

Article abstract of DOI:10.1002/ejic.201201471

Symmetrical N-heterocyclic 1,1′,3,3′-tetrahydro-2,2′-bi- 1,3,2-diazaphospholes and 2,2′-bi-1,3,2-diazaphospholidines are prepared by time-saving, sequential "one-pot" syntheses starting from 1,4-diazabutadienes or N-alkyl or N-aryl-substituted ethane-1,2-

Full text of DOI:10.1002/ejic.201201471

FULL PAPER
DOI:10.1002/ejic.201201471
Rational Synthesis and Mutual Conversion of
Bis-N-heterocyclic Diphosphanes and Secondary
N-Heterocyclic Phosphanes
Oliver Puntigam,[a] Daniela Förster,[a] Nick A. Giffin,[b]
Sebastian Burck,[a] Johannes Bender,[a] Fabian Ehret,[a]
Arthur D. Hendsbee,[b] Martin Nieger,[c] Jason D. Masuda,*[b] and
Dietrich Gudat*[a]
Keywords: Phosphanes / Diphosphanes / Heterocycles / Radicals / Reduction / Reaction mechanisms
Symmetrical N-heterocyclic 1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-
1,3,2-diazaphospholes and 2,2Ј-bi-1,3,2-diazaphospholidines
are prepared by time-saving, sequential “one-pot” syntheses
starting from 1,4-diazabutadienes or N-alkyl or N-aryl-sub-
stituted ethane-1,2-diamines. This method offers high selec-
tivity and minimizes the loss of products owing to unwanted
hydrolysis, and thus grants high product yields. In some
cases, secondary phosphanes were formed together with or
instead of diphosphanes. This reaction is explained by a fol-
low-up process involving homolytic fission of diphosphanes
to give phosphanyl radicals, which then react with ammo-
nium salts to give a mixture of secondary phosphanes and
chlorophosphanes. Even if its synthetic scope is as yet lim-
ited, this approach seems promising in offering superior
selectivity and higher yields than common synthetic proto-
cols that rely on the use of complex hydrides as reducing
agents. In addition to the reductive conversion of diphos-
phanes into secondary phosphanes, a reverse reaction under
exposure of the reactants to light is also reported.
Recent studies have particularly focused on derivatives like
2,2Ј-bi-1,3,2-diazaphospholidines 3[5,6] or 1,1Ј,3,3Ј-tetra-
hydro-2,2Ј-bi-1,3,2-diazaphospholes 3Ј[7–9] (Scheme 1), in
which the (R2N)2P moieties form parts of five-membered
N-heterocyclic rings.[10] The introduction of bulky N-aryl or
N-alkyl substituents at the rather rigid heterocyclic frame-
works of these molecules offers excellent opportunities for
the specific tuning of steric interactions and P–P bond reac-
tivity.[8,9] In this respect, heterocycles 3Ј received particular
attention as the stabilization of diazaphospholenyl radicals
Introduction
Acyclic tetrakis(dialkylamino)diphosphanes of general
composition (R2N)2P–P(NR2)2 have sporadically been
studied during the past decades.[1,2] The combination of po-
tentially reactive P–N and P–P bonds offers good prospects
for various synthetic applications. Lappert et al. reported[2]
that sterically encumbered bisphosphanes can undergo re-
versible homolytic dissociation to give phosphanyl radi-
cals,[3] which may be subsequently employed to generate
stable phosphanido metal carbonyls. Crossairt and Cum-
mins recently discussed the possibility to use the unique P–
P bond reactivity to activate P4 or AsP3,[4] which are start-
ing materials for a variety of P- or As-containing species.
[a] Institut für Anorganische Chemie, University of Stuttgart,
Pfaffenwaldring 55, 70550 Stuttgart, Germany
Fax: +49-711-685-64241
E-mail: gudat@iac.uni-stuttgart.de
Homepage: http://www.iac.uni-stuttgart.de/arbeitskreise/
akgudat
[b] The Atlantic Centre for Green Chemistry and Dept. of
Chemistry, Saint Mary’s University,
Halifax, Nova Scotia, B3H 3C3, Canada
E-mail: jason.masuda@smu.ca
Homepage: http://www.smu.ca/faculty/jasonmasuda/
[c] Laboratory of Inorganic Chemistry, Department of Chemistry,
University of Helsinki,
Scheme 1. Synthesis of 3/3Ј R = tBu (a), iPr (b), Cy (c), 2,6-dimeth-
ylphenyl (Dmp, d), 2-methylphenyl (o-Tol, e), Mesityl (f), 2,6-di-
isopropylphenyl (Dipp, g), o-tBuC6H4 (h); reactants and condi-
tions: (i) THF, –78 °C, 1 equiv. PCl3, 2.5 equiv. NEt3; (ii) THF, –78
to 0 °C, 2 equiv. Li, 2 equiv. NEt3HCl; (iii) –78 °C, 1 equiv. PCl3;
(iv) THF, 1.5 equiv. Mg.
P. O. Box 55 (A. I. Virtasen aukio 1), 00014 University of
Helsinki, Finland
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201201471.
Eur. J. Inorg. Chem. 2013, 2041–2050
2041
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
by cyclic π-electron delocalization enables P–P bond cleav- in the same reaction vessel. The target diphosphanes were
age processes to occur under extremely mild conditions.[8] isolated by extraction into a nonpolar solvent, separation
Quite interestingly, Förster et al.[9] were able to show that, of the insoluble side products by filtration, and evaporation
regardless of this disposition, specific substrates like elec- of the filtrate. The decision not to purify and crystallize the
tron-poor alkynes may also address P–N bond reactivity. intermediate products results in substantial time saving and
This apparent dualism offers interesting prospects and stim- helps to minimize losses caused by hydrolysis upon poten-
ulates extended studies of the chemical properties of species tial contact with moisture.
like 3Ј.
In contrast to the cases described so far, the reactions of
Up to now, only a few heterocyclic diphosphanes of type derivatives of 1 and 1Ј with very bulky N-aryl substituents
3[5,6,10] or 3Ј[7,8] are known, and common synthetic ap- (R = Dipp, o-tBuC6H4; 1g–h, 1Јg) took a different course
proaches by reductive coupling of 2-chloro-1,3,2-diaza- and yielded secondary phosphanes (4g, 4Јg)[13] or mixtures
phospholenes 2Ј or 2-chloro-1,3,2-diazaphospholidines 2 of secondary phosphanes and diphosphanes (3/4h;
often give only moderate yields. To improve the synthetic Scheme 2). Phosphane 4g was isolated after a similar work-
access to diphosphanes 3 and 3Ј, we wanted to elaborate a up as that employed for 3/3Ј. Spectroscopic studies allowed
time-saving, sequential “one-pot” synthesis from ethane- us to establish that diphosphanes 3g–h/3Јg were the initial
1,2-diamines 1 or 1,4-diazabutadienes 1Ј. This route should reaction products and were later converted into 4g–h/4Јg
grant high product yields and minimize the effort required when the reaction went on. This last conversion could be
for work-up of the reaction mixtures. Surprisingly, we found suppressed when the precipitated solids (LiCl, Et3NHCl),
that some of the reactions studied yielded N-heterocyclic which had been formed as side products with chlorophos-
secondary phosphanes rather than the targeted diphos- phanes 2g–h/2Јg, were removed by filtration before the mag-
phanes. Therefore, we performed more detailed studies, nesium was added. The pure diphosphanes 3g,[6] 3Јg[8,9] and
which allowed us to propose a mechanistic explanation for 3h were readily isolated after appropriate work-up when the
these at first glance puzzling results and to establish condi- reaction was conducted under these conditions.
tions for the mutual conversion between secondary hetero-
cyclic phosphanes and the corresponding diphosphanes.
Results and Discussion
A convenient preparation of bisphosphanes 3 and 3Ј pro-
ceeds by magnesium reduction of 2[6] and 2Ј, respectively.[8]
P-Chloro-substituted precursors are typically generated by
condensation of ethane-1,2-diamines 1 with PCl3 or by re-
duction of 1,4-diazabutadienes 1Ј and subsequent conden-
Scheme 2. One-pot synthesis of phosphanes 4g, 4Јg and 4h; (i)
sation with PCl3 and are isolated prior to the subsequent
THF, –78 °C, 1 equiv. PCl3, 2.5 equiv. Et3N; (ii) 1.5 equiv. Mg, cat.
I2, room temp. (iii) THF, –78 to 0 °C, 2 equiv. Li, 2 equiv. Et3NHCl;
coupling step.[11–13] Having previously developed a high-
yield synthesis of 2Ј that combines the reduction and cou- (iv) –78 °C, 1 equiv. PCl3; (v) Et3NHCl/Mg.
pling steps into a sequential one-pot procedure,[13] we set
out to extend this approach to the preparation of 3 and 3Ј
by performing all stages of the multistep synthesis in a one-
pot manner without purification of the intermediates.
The isolated highly air and moisture sensitive diphos-
The reactions were carried out by using starting materi- phanes and phosphanes were found to be spectroscopically
als 1 and 1Ј with N-alkyl or N-aryl substituents of different pure, and their identity was unequivocally confirmed by
steric bulk (Scheme 1). The condensation of diamines 1 NMR and MS data. The diphosphanes 3a and 3h (Fig-
with PCl3 to give 2 was initiated by the addition of Et3N, ure 1) and the secondary phosphane 4g (Figure 2) were also
whereas 2Ј were prepared by the previously described one- characterized by single-crystal X-ray diffraction studies of
pot procedure, which involves quenching of the initially crystals obtained by recrystallization from THF/hexane at
formed diazabutadiene dianion with an ammonium salt and –20 °C. The molecular structures of 3a and 3h exhibit crys-
the subsequent base-induced condensation of the α-amino- tallographic Ci symmetry; the five-membered rings feature
aldimine intermediate with PCl3.[13] Even if 2Ј is in principle transoid alignment and a twist configuration. The P–P dis-
also accessible by direct metathesis of diazabutadiene di- tances are somewhat shorter than those in the acyclic tetra-
[1]
anions with PCl3,[12] the two-step approach used here has aminodiphosphanes (iPr2N)2P–P(NiPr2)2
and (iPr2N)-
been reported to give superior yields.[13] We found that in [(Me3Si)2N]P–P(NiPr2)[N(SiMe3)2][2] (2.29–2.30 Å) but are
most cases (2/2Јa–f) the LiCl and NEt3HCl formed in the still far longer than the average P–P distance in diphos-
previous steps did not interfere with the subsequent re- phanes (2.217Ϯ0.08 Å[14]). The endocyclic bond lengths
ductive coupling step. The removal of tetrahydrofuran match those in other bis-N-heterocyclic diphosphanes
(THF) and precipitated salts was thus unnecessary, and the (Table 1). The P–P distances are equal for 3a and 3Јa (and
Mg reduction was easily carried out at room temperature likewise for 3g[6] and 3Јg[9]) and increase in both series of
Eur. J. Inorg. Chem. 2013, 2041–2050
2042
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
heterocycles with growing bulk of the N-aryl substituents. Table 1. Selected bond lengths [Å] and angles [°] for 3a, 3d, 3f, 3g,
3h, 3Јa, 3Јf and 3Јg.
These parallel trends suggest that the bond lengthening is
mainly induced by steric factors.
R
P1–P1Ј P1–N2, C3–C4 N2–P1–
P1–N5 N5
Ref.
3a tBu
2.240(1) 1.716(1), 1.509(1) 94.93(1)
this work
1.724(1)
[5]
3d Dmp
3f Mes
2.301(1) 1.706(1), 1.512(2) 90.30(6)
1.710(1)
2.283(1) 1.712(1), 1.498(2) 90.66(6)
1.716(1)
2.321(1) 1.700(2), 1.525(2) 91.74(9)
1.727(2)
2.270(2) 1.716(2), 1.504(4) 91.35(11) this work
1.735(2)
[5]
[6]
3g Dipp
3h o-tBuC6H4
3Јa tBu
[7,8]
2.244(1) 1.739(1), 1.338(1) 94.28(4)
1.739(1)
[8]
3Јf Mes
2.316(2) 1.723(3), 1.328(4) 89.51(1)
1.726(3)
[9]
3Јg Dipp
2.320(1) 1.737(3), 1.330(5) 90.83(15)
1.718(3)
Figure 1. Crystal structure of 3a and 3h; H atoms partially omitted
for clarity; 50% probability ellipsoids. Both molecules exhibit crys-
tallographic Ci symmetry (symmetry-dependent atoms were gener-
ated with the operator –x + 1, –y + 1, –z + 1).
To understand the reason for the special behaviour of 2g,
2h and 2Јg in the one-pot reaction, we attempted to obtain
a consistent explanation for the ongoing processes. Both the
intermediacy of 3g, 3h and 3Јg and the formation of 4g, 4h
and 4Јg is readily explained by the mechanism displayed in
Scheme 3. The in-situ-formed chlorophosphanes 2/2Ј react
as expected with Mg by reductive coupling to give 3/3Ј,
which are stable species if the substituents R are small or
medium sized. However, sterically demanding groups (R =
Dipp, o-tBuC6H4) induce cleavage of 3g, 3h and 3Јg to pho-
sphanyl radicals, which then react with NEt3HCl to give a
combination of 4g, 4h and 4Јg and 2g, 2h and 2Јg. The
chlorophosphanes can re-enter the reaction cycle and are
eventually transformed into secondary phosphanes 4g, 4h
and 4Јg.
Figure 2. Molecular structure of 4g; H atoms in the Dipp substitu-
ents omitted for clarity; 50% probability ellipsoids. Selected bond
lengths [Å]: P(1)–H(1) 1.34(2), P(1)–N(5) 1.688(1), P(1)–N(2)
1.700(1), N(2)–C(3) 1.473(2), N(5)–C(4) 1.466(2), C(3)–C(4)
1.522(2).
Crystalline 4g contains isolated molecules that display a
similar twist conformation of the five-membered ring as
that in 3a and 3h (Figure 2). The P–N bonds are somewhat
shorter but are similar to those in the C–C unsaturated ana-
logue (4Јg, 1.69–1.71 Å).[13] In contrast, the P–H bond in
4g [1.34(2) Å] is much shorter than that in 4gЈ [1.48(1) Å]
and is closer to the “normal” P–H distance in phosphanes
(1.288Ϯ0.09 Å).[13] The bond lengthening in 4Јg was attrib-
uted to the hydride character of the P–H unit and is inti-
Scheme 3. Radical-induced formation of phosphanes 4 and 4Ј by
recursive reductive coupling of 2 and 2Ј with Mg.
Support for this mechanistic proposal was gained from
mately connected with the possibility of stabilization of a further experimental studies, which finally allowed us to
negative formal charge on the hydrogen atom by hypercon- confirm the feasibility of each individual step in the pro-
jugation between the σ*(PH) orbital with the 6π electrons posed reaction sequence. In particular, we found that:
in the heterocycle (σ* aromaticity).[13,15] In this respect, the
(a) Treatment of chlorophosphanes 2g, 2h and 2Јg with
relative shortening (with respect to 4Јg) of the P–H distance magnesium in the absence of Et3NHCl provides isolable
in 4g is in accord with the concept that the interruption of quantities of the expected diphosphanes.
the cyclic π delocalization by the C–C saturated backbone
(b) The dissociation of the diphosphanes by homolytic
renders the hyperconjugative stabilization less efficient and, P–P bond cleavage had already been established for 3Јg,[8]
thus, diminishes the hydride character of the P–H moiety.
and electron paramagnetic resonance (EPR) studies also
Eur. J. Inorg. Chem. 2013, 2041–2050
2043
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
confirmed the formation of radicals at room temperature two 1H nuclei. This, at first glance surprising, finding is con-
for 3g and 3h (Scheme 3; generation of radicals from 3g has veniently explained in the light of a computational analysis
recently been discovered by Masuda et al.).[6]
of the hyperfine interactions in 5g.[6] The computed split-
(c) The isolated diphosphanes 3g, 3h and 3Јg react with ting to protons of axial CH bonds at the twisted ring are
Et3NHCl in THF or toluene in the absence of Mg to give clearly larger than those to protons of equatorial CH
a 1:1 mixture of 2g, 2h and 2Јg and 4g, 4h and 4Јg. Moni- bonds, and the equivalence of all ring protons was ex-
toring of the reaction by 31P NMR spectroscopy revealed plained as resulting from dynamic averaging owing to rapid
further that i) a mixture of chlorophosphanes 2g and 2h ring inversion. If one assumes that 5h exhibits a similar
and secondary phosphanes 4g and 4h also formed after un- alignment of N-aryl rings and central heterocycle as 3h, dy-
reacted magnesium was removed from the appropriate one- namic averaging in this case requires a combination of ring
pot reaction mixtures before the reaction had gone to com- inversion and rotational reorientation of the asymmetrical
pletion, and ii) the consumption of the chlorophosphane substituents. As it is reasonable to assume that this process
and further formation of secondary phosphane resumed af- occurs on a much longer time scale, the observed spectrum
ter magnesium was re-added to the mixture.
is considered to represent the hyperfine interactions in a
(d) Even for diphosphanes such as 3a and 3Јa, which do “quasistatic” five-membered ring and is dominated by the
not undergo homolytic dissociation at room temperature, two large couplings to the axial protons.
formation of secondary phosphanes as side products can be
The generation of secondary aminophosphanes 4g, 4h
enforced at elevated temperature under conditions for and 4Јg during the magnesium reduction of chlorophos-
which the formation of small quantities of phosphanyl radi- phanes in the presence of a weak Brønsted acid can be com-
cals has been established by EPR studies.[8]
pared with common synthetic protocols such as the re-
[17]
The EPR data of 5g [g = 2.013, a(31P) = 61.3 G, a(14N) duction with the complex hydrides LiAlH4[16] or NaBH4
.
= 3.6 G, a(1H) = 3.1 G; values were obtained from spectral The LiAlH4 reduction frequently suffers from the formation
simulations and match the previously reported ones][6] and of substantial amounts of side products, which complicate
5h (Figure 3) reveal that the hyperfine couplings to phos- the work-up protocol and lowers the achievable product
phorus are some 20 G higher than those reported for the yields,[15,16] although species such as 4Јg[13] or 4Јd are ac-
C–C unsaturated diazaphosphanyl radicals 5Јa, 5Јf and cessible in reasonable yield. The serendipitous isolation of
5Јg.[8] This is in accord with the fact that the unpaired elec-
tron can no longer be delocalized around the entire hetero-
cycle as it is in 5Ј[8] and is more or less confined to the NPN
moiety. We further explain the low a(31P) and a(14N) values
for 5g and 5h, which are both close to the bottom end of
the range of known couplings in acyclic diaminophos-
phanyl radicals [a(31P) = 63–108 G, a(14N) = 3.7–5.1 G[3]],
as a consequence of conformational constraints in the het-
erocyclic structures, which improve NPN π delocalization
and, therefore, increase the π character of the unpaired elec-
tron.
Figure 4. Molecular structure of 6g·THF; H atoms in the Dipp
substituents omitted for clarity; 50% probability ellipsoids. Selected
distances [Å]: P(1)–H(1) 1.33(3), P(1)–H(2) 1.36(3), P(1)–N(5)
1.685(2), C(3)–C(4) 1.522(4), N(2)–Al(1) 1.813(2), Al(1)–H(3)
1.51(3), Al(1)–H(4) 1.53(3), Al(1)–O(1) 1.916(2).
Figure 3. Experimental and simulated EPR spectrum of 5h at
390 K in toluene. Simulation parameters: g = 2.014, a(31P) = 63.8 G
(1 P), a(14N) = 4.3 G (2 n), a(1H) = 3.8 G (2 H).
The most striking deviation between the EPR spectra of
5g and 5h concerns the different hyperfine splitting pat-
terns: the signal of 5g appears as a doublet of nonets owing
1
to coupling with one 31P, two 14N and four equivalent H
Scheme 4. Synthesis of 6g and phosphane–boranes 7g and 8g (R =
nuclei, whereas that of 5h is split into a doublet of septets
Dipp): (i) LiAlH4, THF; (ii) NaBH4, THF; (iii) H3B·SMe2, THF;
attributable to interaction with one 31P, two 14N and only (iv) H3B·SMe2 (7g) or B(C6F5)3 (8g), toluene.
Eur. J. Inorg. Chem. 2013, 2041–2050
2044
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
a few crystals of side product 6g (Figure 4) during the at- methyl sulfide in THF initially gave an NMR silent species
tempted synthesis of 4g (Scheme 4) confirms that this low when analyzed by 31P NMR spectroscopy. However, after
selectivity is related to the high chemical activity of the re- several minutes, the broad doublet of 7g (1JP,H = 357 Hz)
actant, which is liable to induce over-reduction and ring appeared, which indicated that the addition of a hydrogen
cleavage. In contrast, the NaBH4 reduction of 2g smoothly atom to the phosphanyl radical to give a P–H bond had
produced phosphane–borane 7g (Scheme 4), which was iso- occurred. The phosphane–borane 7g was isolated
lated in good yield as a stable white solid and characterized (Scheme 4) and unambiguously identified by analytical and
by spectroscopic data and a single-crystal X-ray diffraction spectroscopic data and a single-crystal X-ray diffraction
study (Figure 5). Removal of the coordinated borane is study. Similarly, the reaction of 3g with B(C6F5)3 gave the
feasible by heating a mixture of 7g and Et3N in hexane to secondary phosphane–borane complex 8g as revealed by
reflux for 12 h. Monitoring of the reaction by 31P NMR 31P NMR spectroscopy (1JP,H = 404 Hz) and single-crystal
spectroscopy indicated, however, that this process is not free X-ray crystallography (Figure 6). Once again, a hydrogen
of side reactions. No attempts to separate Et3NBH3 and atom has been added to the phosphanyl to give a closed-
isolate 4g were made. The use of super hydride (Li[BHEt3]) shell species. The P–B distance of 2.106(3) Å in 8g is at the
as an alternative hydride transfer agent[15] allows for the long end of the range for secondary phosphane–B(C6F5)3
formation of the secondary aminophosphane 4g from 2g adducts (2.025–2.098 Å).[18] The P–H distance is similar to
without coordination of the borane. From this route, 4g can those in 4g and 7g, which indicates that coordination of
be isolated in good yield (80%), and this protocol can thus B(C6F5)3 has little to no structural impact. The gradually
be considered as an alternative route to the one-pot method increasing ν(PH) vibrational frequencies of 4g (2025 cm–1),
outlined in Scheme 2.
7g (2255 cm–1) and 8g (2314 cm–1) suggest, however, that
the P–H bond strengthens perceptibly upon coordination
of more and more Lewis acidic boranes.
Figure 5. Molecular structure of 7g; H atoms in the Dipp substitu-
ents omitted for clarity; 50% probability ellipsoids. Selected dis-
tances [Å] and angles [°]: P(1)–H(1) 1.36(3), P(1)–B(1) 1.884(4),
P(1)–N(2) 1.652(2), C(3)–C(3)Ј 1.528(4), B(1)–P(1)–H(1) 109(1),
N(2)–P(1)–N(2)Ј 93.3(1).
Compared to the commonly applied reduction protocols
with complex hydrides,[16,17] the magnesium/ammonium salt
approach proceeded in our hands with much higher selec-
tivity, facilitated the work-up procedure, and thus granted
substantial time-saving and higher product yields. On the
other hand, this method currently seems to be limited to
the synthesis of a few sterically hindered target species (at
least as long as only reactions at room temperature are con-
sidered), and one will have to see if the selectivity can be
Figure 6. Molecular structure of 8g; H atoms (other than on the
phosphorus atom) omitted for clarity; 50% probability ellipsoids.
Selected distances [Å] and angles [°]: P(1)–B(1) 2.106(3), P(1)–
H(99) 1.33(2), N(1)–C(1) 1.482(3), N(1)–P(1) 1.671(2), N(2)–C(2)
1.475(3), N(2)–P(1) 1.677(2), P(1)–B(1) 2.106(3), C(1)–C(2)
1.508(4). C(1)–N(1)–P(1) 111.67(17), C(2)–N(2)–P(1) 111.62(16),
N(1)–P(1)–N(2) 93.74(11), N(1)–P(1)–B(1) 121.16(12), N(2)–P(1)–
B(1) 125.82(12), N(1)–C(1)–C(2) 104.6(2), N(2)–C(2)–C(1)
maintained in reductions of substrates with less bulky sub- 105.3(2).
stituents at elevated temperatures. However, the operational
advantages are clearly evident and should stimulate further
Attempts to ascertain the source of the hydrogen atoms
studies to develop this protocol into a more widely appli- for the formation of 6g and 8g through a number of experi-
cable alternative to commonly used reaction schemes.
ments proved to be fruitless. The reaction of 3g with
Based on the proven formation of phosphanyl radicals at BH3·THF in THF and BD3·THF in THF gave inconclusive
room temperature by homolytic cleavage of the P–P bond results, and BH3·NMe3 did not react at all. With this in
in 3g, we desired to further investigate the reaction of this mind, we studied the reaction of 3g with B(C6F5)3 in C6F6
species with boranes. We were interested in the possibility as a hydrogen-free solvent, thus eliminating the borane and
of trapping a phosphanyl radical by forming a Lewis acid– solvent as hydrogen-atom sources. This reaction resulted
base complex. Thus, the reaction of 3g with borane–di- likewise in the formation of 8g. As we see no evidence of
Eur. J. Inorg. Chem. 2013, 2041–2050
2045
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
ligand decomposition in the NMR spectra, the hydrogen
atom most likely originates from impurities in the reagents
and glassware, or in the starting diphosphane 3g.
Experimental Section
General Remarks: All preparations were done under an atmosphere
of dry, O2-free N2 or Ar, by employing both Schlenk-line tech-
niques and an mBraun Labmaster SP inert atmosphere glove box.
Solvents were purified by employing a Grubbs type solvent purifi-
cation system manufactured by Innovative Technology. Hyflo Su-
per Cel (Celite) was purchased from Sigma–Aldrich and dried for
24 h in an oven prior to use. Molecular sieves (4 Å) were purchased
from Sigma–Aldrich. and dried overnight at 140 °C under vacuum.
EI-MS: Varian MAT 711 instrument, 70 eV. (+)-ESI MS: Bruker
Daltonics MicroTOF Q instrument, methanolic solution. Elemen-
tal analyses: Perkin–Elmer 2400 Series II CHN/O Analyzer. Some
samples could not be prepared under an inert atmosphere, and we
explain the rather large deviations between the found and calcu-
lated element compositions of some diphosphanes 3/3Ј as a conse-
quence of the high oxidation sensitivity of these samples. Melting
points were recorded with a Büchi Point B-545 or a Electrothermal
MEL-Temp 3.0 apparatus with samples in glass capillaries pre-
pared and sealed under inert conditions. Solution NMR spectra
were recorded at 303 K with Bruker Avance 400 (1H 400.1, 13C
100.5, 31P 161.9, 11B 128.4 MHz), Avance 300 (1H 300.1, 13C 75.4,
31P 121.4, 19F 282.4, 11B 96.3 MHz) or Avance 250 (1H 250.1, 13C
62.8, 31P 101.2, 11B, 80.2 MHz) NMR spectrometers. Chemical
shifts are referenced to external Si(CH3)4 (1H, 13C), 85% H3PO4 (Ξ
= 40.480747 MHz, 31P), CFCl3 (Ξ = 94.094011, 19F) or BF3·OEt2
(Ξ = 32.083974 MHz, 11B). Coupling constants are given as abso-
lute values; positive chemical shift values denote shifts to lower
frequencies, i-, o-, m-, p- denote positions in N-aryl rings. EPR
spectra were measured with a Bruker EMX X-band spectrometer.
The EasySpin software[19] was used for spectral simulations. 2-
Chloro-1,3,2-diazaphospholidines and 2-chlorodiazaphospholenes
were synthesized as described elsewhere. The analytical, spectro-
scopic and single-crystal X-ray diffraction data of previously un-
known 2c and 2h are included in the Supporting Information
The readily occurring conversion of some diphosphanes
3/3Ј into appropriate phosphanes 4/4Ј raised the question if
a reverse transformation might also be feasible. Indeed,
during the synthesis of the secondary phosphane 4Јd by Li-
AlH4 reduction of 2Јd, we noticed that the initially formed
product was quantitatively (as monitored by 31P NMR
spectroscopy) converted into diphosphane 3Јd when the re-
action mixture was stirred for a prolonged time with expo-
sure to daylight. Partial conversion of secondary phospha-
nes to the corresponding diphosphanes under similar condi-
tions was also observed for 4c and 3c, and it had previously
been found that prolonged storage of pure 4Јa with expo-
sure to daylight resulted in partial conversion to diphos-
phane 3Јa, which was isolated in crystalline form and char-
acterized by spectroscopy and a single-crystal X-ray diffrac-
tion study.[7] Control experiments confirmed that the phos-
phane–diphosphane conversion did not take place when the
samples were protected from light. The failure to observe
any spectroscopically detectable side products suggests that
the conversion may be accompanied by the formation of
molecular hydrogen, but a positive proof of this hypothesis
is still pending.
Conclusions
Symmetrical
2,2Ј-bi-1,3,2-diazaphospholidines
and
1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-1,3,2-diazaphospholes can be
prepared from the appropriate ethane-1,2-diamines or 1,2-
diimines in overall yields of 70 to 91% in a short period of
time. Most syntheses can be performed as one-pot pro-
cedures without isolation of intermediates, but preparation
of a few products with sterically demanding N-aryl groups
requires removal of the ammonium salt (Et3NHCl), formed
as a by-product of a preceding reaction step, prior to the
final reductive coupling. In these cases, failure to remove
Et3NHCl results in the reduction of the applied chloro-
phosphanes to the corresponding secondary phosphanes.
Mechanistic studies suggest that this reaction is initiated by
homolytic dissociation of the diphosphanes. The phos-
phanyl radicals formed then react with Et3NHCl to pro-
duce 1:1 mixtures of secondary phosphanes and chloro-
phosphane starting material, which may then re-enter the
reaction sequence. Despite its as yet limited scope, the mag-
nesium/ammonium salt reduction protocol offers some
clear advantages over the commonly applied reduction with
complex hydrides and, thus, offers interesting prospects for
its development into a more generally applicable synthetic
method. Preliminary studies reveal that in addition to the
reductive conversion of diphosphanes to appropriate sec-
ondary phosphanes, a reverse transformation is also feas-
ible upon prolonged exposure of the appropriate secondary
phosphanes to daylight. Further studies are needed to es-
tablish if these reactions involve the production of molecu-
lar hydrogen.
General Procedure for the Preparation of 2,2Ј-Bi-1,3,2-diazaphos-
pholidines 3
From Ethane-1,2-diamines (1): The appropriate diamine (5–
15 mmol) was added dropwise at –78 °C to a solution of PCl3 (5–
15 mmol, 1 equiv.) and Et3N (15–40 mmol, 3 equiv.) in THF
(50 mL). A white precipitate of Et3NHCl formed, and the mixture
was stirred for 15 min at –78 °C and another 6 h at room tempera-
ture to allow complete reaction. Magnesium chips (7.5–22 mmol,
1.5 equiv.) were then added, and the mixture was stirred until com-
pletion of the reaction was confirmed by 31P{1H} NMR spec-
troscopy (ca. 24 h; 96 h for 1c). The volatiles were then removed
under reduced pressure, and the residue was suspended in hexane
(50 mL). The solids were removed by filtration and washed with
hexane or toluene to extract precipitated 3. The filtrate was evapo-
rated to dryness. The spectroscopically pure products were ob-
tained as white-to-yellow solids or yellow oils (3b, 3c, and 3e).
Crystals of 3a and 3h suitable for single-crystal X-ray diffraction
studies were obtained by recrystallization from THF/hexane (1:3)
at –20 °C.
From 2-Chloro-1,3,2-diazaphospholidines 2: The appropriate chloro-
diazaphospholidine (2d 500 mg, 1.5 mmol; 2e 1.00 g, 3.2 mmol; 2g
4.40 g, 10 mmol) was dissolved in THF (20–50 mL). Magnesium
turnings (1.5 equiv.) and a crystal of I2 (2e: no I2 added) were added
to the solution. The mixture was stirred until completion of the
reaction was confirmed by 31P NMR spectroscopy (approx. 24 h).
The work-up was carried out as described above.
1,1Ј,3,3Ј-Tetra-tert-butyl-2,2Ј-bi-1,3,2-diazaphospholidine
(3a):
Yield 2.50 g (81%); m.p. 127 °C. 31P{1H} NMR (C6D6): δ = 91.7
Eur. J. Inorg. Chem. 2013, 2041–2050
2046
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
ppm. 1H NMR (C6D6): δ = 1.26 (s, 36 H, CH3), 2.86 (m, 4 H, 1,1Ј,3,3Ј-Tetradiisopropylphenyl-2,2Ј-bi-1,3,2-diazaphospholidine
NCH2), 3.23 (m, 4 H, NCH2) ppm. 13C{1H} NMR (C6D6): δ = (3g): Yield 2.95 g (72%); m.p. 215 °C. 31P{1H} NMR ([D8]THF):
3/4
1
3
30.3 (pseudo-t,
J
= 4.7 Hz, CH3), 48.5 (s, NCH2), 53.5
δ = 145.2 ppm. H NMR ([D8]THF): δ = 0.5 (d, JH,H = 6.5 Hz, 6
P,C
2/3
3
3
[pseudo-t,
J
P,C
= 9.3 Hz, NC(tBu)] ppm. EI-MS (70 eV, 415 K):
H, CH3), 0.91 (d, JH,H = 6.5 Hz, 6 H, CH3), 0.99 (d, JH,H =
m/z (%) = 201.1 (100) [M/2]+, 402.3 (20) [M]+. C20H44N4P2
(402.54): calcd. C 59.67, H 11.02, N 13.92; found C 57.12, H 10.4,
N 12.53.
6.5 Hz, 6 H, CH3), 0.98 (d, JH,H = 6.5 Hz, 6 H, CH3), 1.05 (d,
3
3JH,H = 6.5 Hz, 6 H, CH3), 1.38 (d, JH,H = 6.5 Hz, 6 H, CH3),
3
3
3
1.22 (d, JH,H = 6.5 Hz, 6 H, CH3), 1.29 (d, JH,H = 6.5 Hz, 6 H,
3
CH3), 2.60 (sept, JH,H = 6.5 Hz, 2 H, CH), 3.16 (m, 4 H, NCH2),
1,1Ј,3,3Ј-Tetraisopropyl-2,2Ј-bi-1,3,2-diazaphospholidine (3b): Yield
2.06 g (79%). 31P{1H} NMR (C6D6): δ = 97.4 ppm. 1H NMR
(C6D6): δ = 1.18 (d, JH,H = 6.3 Hz, 12 H, CH3), 1.21 (d, JH,H
3.36 (m, 4 H, NCH2), 3.34 (m, 2 H, CH), 4.02 (sept, 3JH,H = 6.5 Hz,
3
2 H, CH), 4.08 (sept, JH,H = 6.5 Hz, 2 H, CH), 6.72 (m, 2 H, m-
3
3
=
CH), 6.78 (m, 2 H, m-CH), 6.8–6.87 (m, 6 H, m/p-CH), 7.15 (m, 2
H, m-CH) ppm. EI-MS (70 eV, 430 K): m/z (%) = 409.3 (100) [M/
2]+, 367.3 (70) [C25H38N2]+. C52H76N4P2 (819.13): calcd. C 76.25,
H 9.35, N 6.84; found C 73.91, H 9.43, N 6.25.
6.5 Hz, 12 H, CH3), 2.75 (m, 2 H, NCH2), 3.21 (m, 4 H, CH), 3.31
(m, 2 H, NCH2) ppm. 13C{1H} NMR (C6D6): δ = 23.15 (pseudo-
3/4
3/4
t,
J
P,C
= 3.8 Hz, CH3), 23.4 (pseudo-t,
J
P,C
= 6.4 Hz, CH3),
2/3
2/3
50.2 (pseudo-t,
J
= 0.7 Hz, NCH2), 52.2 (pseudo-t,
J
P,C
=
P,C
10.5 Hz, CH) ppm. EI-MS (70 eV): m/z (%) = 333.28 (100) [M –
1,1Ј,3,3Ј-Tetra-2-tert-butylphenyl-2,2Ј-bi-1,3,2-diazaphospholidine
(3h): Yield 3.07 g (87%); m.p. Ͼ180 °C (decomp.). 31P{1H} NMR
CH3]+.
1
(C6D6): δ = 120.8 ppm. H NMR (C6D6): δ = 1.39 (s, 36 H, tBu),
1,1Ј,3,3Ј-Tetracyclohexyl-2,2Ј-bi-1,3,2-diazaphospholidine
(3c):
3.08 (m, 4 H, NCH2), 3.77 (m, 4 H, NCH2), 6.82 (t, 3JH,H = 7.5 Hz,
Yield 920 mg (91%). 31P{1H} NMR (C6D6): δ = 96.1 ppm. 1H
NMR (C6D6): δ = 0.90–1.57 (m, 20 H, C6H11), 1.63–1.80 (m, 12
H, C6H11), 2.03–2.19 (m, 8 H, C6H11), 2.83 (m, 4 H, C6H11), 2.90
(m, 4 H, NCH2), 3.40 (m, 4 H, NCH2) ppm. 13C{1H} NMR
(C6D6): δ = 26.0 (br. s, C6H11), 26.4 (br. s, C6H11), 33.8 (pseudo-t,
3
3
4 H, C6H4), 6.94 (t, JH,H = 7.5 Hz, 4 H, C6H4), 7.35 (d, JH,H
=
3
7.8 Hz, 4 H, C6H4), 7.67 (d, JH,H = 7.5 Hz, 4 H, C6H4) ppm.
13C{1H} NMR (C6D6): δ = 31.8 (pseudo-t, JP,C = 1.5 Hz, CH3),
36.0 (s, i-C), 58.0 (s, NCH2), 125.2 (s, C6H4), 126.8 (s, C6H4), 128.7
(s, C6H4), 130.8 (pseudo-t, JP,C = 8.5 Hz, C6H4), 146.2 (pseudo-t,
JP,C = 2.5 Hz, C6H4), 147.0 (pseudo-t, JP,C = 5.3 Hz, C6H4) ppm.
C44H60N4P2 (706.92): calcd. C 74.76, H 8.55, N 7.93; found C
72.60, H 8.30, N 7.24.
JP,C = 3.8 Hz, C6H11), 34.2 (pseudo-t, JP,C = 6.4 Hz, C6H11), 50.4
2/3
(br. s, NCH2), 60.8 (pseudo-t,
J
P,C
= 9.6 Hz, NCH) ppm. MS
(ESI+): m/z (%) = 539.36 (80) [MHO2]+, 372.33 (30) [M – 2(C6H11)-
HOCH3]+.
1,1Ј,3,3Ј-Tetrahydro-2,2Ј-bi-1,3,2-diazaphospholes 3Ј
1,1Ј,3,3Ј-Tetrakis(2,6-dimethylphenyl)-2,2Ј-bi-1,3,2-diazaphosphol-
idine (3d): Yield 340 mg (77 %); m.p 155 °C. 31P{1H} NMR
From Diazabutadienes 1Ј: Lithium turnings (69 mg, 10 mmol) were
(CDCl3): δ = 98.6 ppm. H NMR (CDCl3): δ = 1.44 (s, 12 H, o- added to a solution of the appropriate diazabutadiene 1Јa–1Јf
1
CH3), 2.50 (s, 12 H, o-CH3), 2.90 (m, 4 H, NCH2) 3.63 (m, 4 H,
(5 mmol) in THF (50 mL). The mixture was stirred for 24 h, and
3
3
NCH2), 6.56 (d, JH,H = 7 Hz, 4 H, m-CH), 6.89 (t, JH,H = 7 Hz, the excess Li was removed by filtration. The red solution was co-
3
4 H, p-CH), 6.98 (d, JH,H = 7 Hz, 4 H, m-CH) ppm. 13C{1H}
oled to –78 °C, NEt3HCl (1.38 g, 10 mmol) was added, and the
stirred solution slowly warmed until the reaction was complete. The
mixture was again cooled to –78 °C, and PCl3 (690 mg, 5 mmol)
was added dropwise. The mixture was warmed to room tempera-
ture and stirred for another 2 h. Magnesium turnings (182 mg,
7.5 mmol) and a few crystals of I2 were added, and the solution
was irradiated in an ultrasonic bath for 2 h. The volatiles were then
removed under reduced pressure, the residue was extracted with
hexane, and the filtrate was evaporated to dryness. The spectro-
scopically pure products were obtained as yellow to light rose sol-
ids. Performing the same reaction sequence with 1Јg as starting
material resulted in quantitative (by 31P NMR spectroscopy) for-
mation of 4Јg. No attempts were made to isolate this product.
4/5
NMR (CDCl3): δ = 18.6 (pseudo-t,
J
P,C
= 2.9 Hz, CH3), 20.7
4/5
2/3
(pseudo-t,
J
P,C
= 5.9 Hz, CH3), 53.1 (pseudo-t,
J
P,C
= 2.3 Hz,
NCH2), 124.3 (br. s, p-C), 128.2 (br. s, m-C), 129.6 (br. s, m-C),
3/4
3/4
135.0 (pseudo-t,
J
P,C
= 1.1 Hz, o-C), 137.3 (pseudo-t,
J
=
P,C
2/3
1.6 Hz, o-C), 141.9 (pseudo-t,
J
P,C
= 5.8 Hz, i-C) ppm. EI-MS
(70 eV, 415 K): m/z (%) = 297.1 (50) [M/2]+, 121.1 (100) [NH2
–
Dmp]+, 314.1 (20) [M/2 + OH]+.
1,1Ј,3,3Ј-Tetra-o-tolyl-2,2Ј-bi-1,3,2-diazaphospholidine (3e): Yield
1.02 g (76%). 31P{1H} NMR (C6D6): δ = 101.2 ppm. 1H NMR
(C6D6): δ = 2.01 (br. s, 12 H, o-CH3), 2.79 (m, 4 H, NCH2), 3.70
(m, 4 H, NCH2), 6.85–7.00 (m, 12 H, C6H4), 7.24–7.29 (m, 4 H,
4/5
C6H4) ppm. 13C{1H} NMR (C6D6): δ = 19.0 (pseudo-t,
J
P,C
=
2.8 Hz, CH3), 54.0 (br. s, NCH2), 123.5 (pseudo-t, JP,C = 1 Hz,
C6H4), 124.5 (pseudo-t, JP,C = 6.4 Hz, C6H4), 126.6 (br. s, C6H4),
131.2 (br. s, C6H4), 133.5 (pseudo-t, JP,C = 1.6 Hz, C6H4), 146.8
(pseudo-t, JP,C = 7.2 Hz, C6H4) ppm. EI-MS (70 eV, 415 K): m/z
From 2-Chloro-2,3-dihydro-1H-diazaphospholes 2Ј: The appropriate
chlorodiazaphospholene (2Јa 1.17 g, 5 mmol; 2Јf 940 mg, 2.6 mmol;
2Јg 2.00 g, 4.5 mmol) was dissolved in THF (10–20 mL). Magne-
sium turnings (1.5 equiv.) and a crystal of I2 were added. The mix-
(%) = 269.1 (70) [M/2]+, 120.1 (100) [C8H10N]+, 121.1 (55) ture was irradiated in an ultrasonic bath for approx. 2 h. The work-
[C8H11N]+.
up was carried out as described above.
1,1Ј,3,3Ј-Tetramesityl-2,2Ј-bi-1,3,2-diazaphospholidine (3f): Yield
From Diazaphospholene 4Јd: Compound 4Јd was prepared in situ
2.6 g (80%); m.p. 255 °C. 31P{1H} NMR (C6D6): δ = 99.4 ppm. 1H by the addition of LiAlH4 (1.25 mmol, 1.25 mL, 1 m in THF) to a
NMR (C6D6): δ = 1.65 (s, 12 H, p-CH3), 2.26 (s, 12 H, o-CH3), cooled (–78 °C) solution of 2Јd (5 mmol, 1.65 g) in THF (50 mL);
2.57 (m, 4 H, NCH2), 2.62 (s, 12 H, o-CH3), 3.44 (m, 4 H, NCH2),
the mixture was stirred for 1 h. The quantitative formation of 4Јd
was established by 31P NMR spectroscopy. The mixture was stirred
for 72 h with exposure to daylight and then evaporated to dryness.
The residue was extracted with toluene (50 mL), the resulting sus-
pension was filtered, and the filtrate was concentrated under re-
duced pressure to a total volume of 10 mL. The 31P NMR spec-
trum of the filtrate revealed the presence of 3Јd as the only phos-
phorus-containing species. Crystallization at –20 °C produced a
6.50 (br. s, 8 H, m-CH), 6.67 (br. s, 8 H, m-CH) ppm. 13C{1H}
4
5
NMR (C6D6): δ = 19.2 (br. s, CH3), 20.7 (dd, JP,C = 1.3 Hz, JP,C
2/3
= 0.8 Hz, o-CH3), 20.8 (br. s, CH3), 53.3 (pseudo-t,
J
P,C
= 2.4 Hz,
= 1.4 Hz, m-C),
4/5
NCH2), 129.2 (br. s, m-C), 130.4 (pseudo-t,
J
P,C
3/4
134.7 (pseudo-t,
J
P,C
= 0.5 Hz, o-C), 137.1 (br. s, o-C), 139.9
= 5.9 Hz, i-C) ppm. EI-MS (70 eV, 420 K): m/z
2/3
(pseudo-t,
J
P,C
(%) = 325.2 (100), [M/2]+, 650.4 (5) [M]+.
Eur. J. Inorg. Chem. 2013, 2041–2050
2047
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
moderate yield of orange crystals, which were isolated by filtration
and dried under vacuum.
From 2g and Super Hydride: Compound 2g (3.00 g, 6.7 mmol) was
dissolved in THF (25 mL). With rapid stirring, Li[BHEt3]
(6.75 mL, 1.0 m in THF, 6.7 mmol) was added dropwise over
10 min. The reaction was allowed to stir for 30 min and remained
colourless throughout. The volatiles were removed in vacuo, and
the resulting white solid was dissolved in pentane (50 mL). The
suspension was filtered through a pad of diatomaceous earth. The
resulting colourless solution was evaporated to dryness to produce
4g as a white powder. Colourless crystals were grown from slow
evaporation of a concentrated toluene solution at ambient tempera-
ture. Yield 2.20 g (80 %), m.p. 139–140 °C. C26H39N2P (410.58):
calcd. C 76.06, H 9.57, N 6.82; found C 76.10, H 9.57, N 6.75. 31P
NMR (C6D6): δ = 71.0 (d, 1JP,H = 141 Hz) ppm. 1H NMR (C6D6):
1,1Ј,3,3Ј-Tetra-tert-butyl-1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-1,3,2-diaza-
phosphole (3Јa): Yield 920 mg (92 %), m.p. 117 °C. C20H40N4P2
(398.51): calcd. C 60.28, H 10.12, N 14.06; found C 59.85, H 10.07,
N 14.05. The spectroscopic data are identical to those reported
elsewhere.[8]
1,1Ј,3,3Ј-Tetrakis(2,6-dimethylphenyl)-1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-
1,3,2-diazaphosphole (3Јd): Yield 540 mg (37%), m.p. 140 °C. 1H
NMR (C6D6): δ = 6.99 (s, 8 H, m-CH), 6.73 (m, 4 H, p-CH), 5.68
(m, 2 H, NCH), 2.39 (s, 12 H, o-CH3), 1.88 (m, 12 H, o-CH3) ppm.
31P{1H} NMR (C6D6): δ = 87.8 ppm. 13C{1H} NMR (C6D6): δ =
141.1 (m, o-C), 139.9 (m, i-C), 136.7 (m, o-C), 131.4 (m, m-CH),
126.3 (m, m-CH), 123.1 (m, p-CH), 114.6 (m, NCH), 20.2 (m, o-
CH3), 19.7 (s, o-CH3) ppm.
3
3
δ = 1.18 (d, JH,H = 6.8 Hz, 6 H, CH3), 1.22 (d, JH,H = 6.8 Hz, 6
H, CH3), 1.3 (d, 3JH,H = 7.0 Hz, 6 H, CH3), 1.32 (d, 3JH,H = 7.0 Hz,
6 H, CH3), 3.32 (m, 4 H, NCH2), 3.41 (m, 2 H, CH), 3.69 (sept,
3JH,H = 7.1 Hz, 2 H, CH), 6.98–7.22 (m, 6 H, m/p-C6H3), 7.04 (d,
JP,H = 141 Hz, 1 H, PH) ppm. 13C{1H} NMR (C6D6): δ = 24.1 (s,
1,1Ј,3,3Ј-Tetramesityl-1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-1,3,2-diazaphos-
phole (3Јf): Yield 640 mg (76%), m. p. Ͼ200 °C (decomp.). 31P{1H}
1
CH3), 24.7 (d, JP,C = 2 Hz, CH3), 24.9 (s, CH3), 25.6 (d, JP,C
=
NMR (CDCl3): δ = 74.1 ppm. H NMR (CDCl3): δ = 2.27 (br. s,
12 H, p-CH3), 2.38 (br. s, 24 H, o-CH3), 5.83 (br. d, 3JP,H = 2.4 Hz,
4 H, NCH), 6.90 (s, 8 H, m-CH) ppm. C40H48N4P2 (646.78): calcd.
C 74.28, H 7.48, N 8.66; found C 73.69, H 7.47, N 8.62.
1.3 Hz, CH3), 28.6 (d, JP,C = 1.5 Hz, CH3), 29.0 (s, CH3), 56.0 (d,
2JP,C = 8.0 Hz, NCH2), 124.0 (d, JP,C = 1.5 Hz, C6H3), 124.3 (s,
C6H3), 128.1 (s, C6H3), 139.2 (d, JP,C = 14.7 Hz, C6H3), 148.9 (d,
JP,C = 2.9 Hz, C6H3), 150.7, (d, JP,C = 2.8 Hz, C6H3) ppm. IR
1,1Ј,3,3Ј-Tetradiisopropylphenyl-1,1Ј,3,3Ј-tetrahydro-2,2Ј-bi-1,3,2-di-
azaphosphole (3Јg): Yield 1.30 g (72%), m.p. 157 °C. C52H72N4P2
(805.12): calcd. C 76.62, H 8.90, N 6.87; found C 75.90, H 8.67,
N 6.78. The spectroscopic data are identical with those reported
elsewhere.[9]
(KBr): ν = 2958 (s), 2924 (m), 2967 (m), 2025 (m, PH), 1464 (m),
˜
1441 (m), 1381 (m), 1360 (m), 1321 (m), 1265 (m), 1205 (m), 1107
(w), 1063 (m), 1040 (m) cm–1. IR (solution in heptane): ν = 2047
˜
(PH) cm–1.
((2,6-Diisopropylphenyl){2-[(2,6-diisopropylphenyl)(phosphanyl)-
amino]ethyl}amino)aluminum Dihydride (6g): LiAlH4 (171 mg,
4.5 mmol) was added to a cooled (–78 °C) solution of 2g (1.33 g,
3 mmol) in THF (50 mL). The mixture was warmed to room tem-
perature and stirred for 15 min. The volatiles were removed under
reduced pressure, hexane (50 mL) was added, and the solids were
separated by filtration. The 31P NMR spectrum of the solution
showed the presence of a mixture of 4g (75%) and 6g (25%). Stor-
age at –20 °C for 2 d produced a few colourless crystals of 6g,
which were manually picked and found to be suitable for a single-
crystal X-ray diffraction study. 31P{1H} NMR (C6D6): δ = –26.3
(s) ppm. 31P NMR (C6D6): δ = –26.3 (tt, 1JP,H = 197, 3JP,H = 6 Hz)
1,3-Bis(2,6-dimethylphenyl)-2,3-dihydro-1H-1,3,2-diazaphosphole
(4Јd): Compound 4Јd was prepared as described above by the ad-
dition of LiAlH4 (1.25 mmol, 1.25 mL, 1 m in THF) to a cooled
(–78 °C) solution of 2Јd (5 mmol, 1.65 g) in THF (50 mL). The
mixture was warmed to ambient temperature and stirred for 1 h.
During this time, the reaction flask was wrapped with aluminium
foil to protect the reaction mixture from light. The solids were re-
moved by filtration and washed with hexane. The filtrates were
evaporated to dryness, and the residue was recrystallized from
MeCN. Yield 830 mg (56%); m.p. 38 °C. 31P NMR (C6D6): δ =
1
1
64.3 (d, JP,H = 138 Hz) ppm. H NMR (C6D6): δ = 2.29 (br. s, 18
3
H, o-CH3), 5.74 (d, JP,H = 2.0 Hz, 2 H, NCH), 6.92 (s, 6 H, m/p-
1
3
ppm. H NMR (C6D6): δ = 1.06 (m, 4 H, THF), 1.13 (d, JH,H
=
1
CH), 7.12 (d, JP,H = 138 Hz, 1 H, PH) ppm. 13C{1H} NMR
3
3
7 Hz, 6 H, CH3), 1.17 (d, JH,H = 7 Hz, 6 H, CH3), 1.28 (d, JH,H
2
(C6D6): δ = 19.2 (br. s, o-CH3), 121.3 (d, JP,C = 6.3 Hz, NCH),
3
= 7 Hz, 6 H, CH3), 1.23 (d, JH,H = 7 Hz, 6 H, CH3), 3.43 (m, 4
4
5
126.3 (d, JP,C = 1.8 Hz, m-CH), 128.8 (d, JP,C = 1.0 Hz, p-CH),
H, NCH2) 3.47–3.61 (m, 4 H, CH), 3.52 (m, 4 H, THF), 5.44 (d,
1JP,H = 197 Hz, 2 H, PH), 7.0–7.22 (m, 6 H, m/p-C6H3) ppm. The
1H NMR signal of the protons attached to the aluminium atom
could not be detected, and further analytical data are not available
as macroscopic samples could not be obtained free from contami-
nation by 4g and other species.
2
137.3 (br. s, o-C), 140.6 (d, JP,C = 13.8 Hz, i-C) ppm. C18H21N2P
(296.35): calcd. C 72.95, H 7.14, N 9.45; found C 72.86, H 7.31, N
9.41.
1,3-Bis(2,6-diisopropylphenyl)-1,3,2-diazaphospholidine 4g
From 1g in a One-Pot Procedure: Compound 1g (761 mg, 2 mmol)
in THF (10mL) was added dropwise to a cooled (–78 °C) solution
of PCl3 (275 mg, 2 mmol) and Et3N (607 mg, 6 mmol) in THF
(50 mL). A precipitate of Et3NHCl formed, and the mixture was
stirred for 15 min at –78 °C and then for 6 h at room temperature.
Magnesium turnings (73 mg, 3 mmol) were added, and the mixture
was stirred for 24 h at room temperature and then for 60 h at 50 °C.
Completion of the reaction was confirmed by 31P{1H} NMR spec-
troscopy. The volatiles were then removed under reduced pressure,
and the residue was suspended in hexane (20 mL). The solids were
removed by filtration and washed with hexane. The filtrates were
evaporated to dryness to give spectroscopically pure 4g as a pale
yellow solid. Analytically pure single crystals were obtained by
recrystallization from hexane. Yield 640 mg (78%), m.p. 138 °C.
C26H39N2P (410.58): calcd. C 76.06, H 9.57, N 6.82; found C 75.89,
H 9.85, N 6.79.
1,3-Bis(2,6-diisopropylphenyl)-1,3,2-diazaphospholidine–Borane (7g)
From 2g and NaBH4: NaBH4 (38 mg, 1 mmol) was added to a solu-
tion of 2g (450 mg, 1 mmol) in THF (50 mL). The mixture was
stirred for 48 h at room temperature, the solvent was removed un-
der reduced pressure, and the residue was extracted with warm hex-
ane. Pure 7g was obtained from the extract after recrystallization
from THF/hexane at –20 °C. Yield 420 mg (93%), m.p. 163.5 °C.
C26H43BN2P (425.42): calcd. C 73.58, H 9.97, N 6.60; found C
71.21, H 9.80, N 6.27.
From 3g and BH3·SMe2: In a 20 mL scintillation vial, 3g (100 mg,
122 μmol) was dissolved in toluene (5 mL) and BH3·SMe2 (122 μL
2.0 m soln. in THF, 244 μmol) was added with rapid stirring. The
orange solution was allowed to stir for 30 min, at which point it
transitioned to yellow. The solution was filtered through a pad of
Eur. J. Inorg. Chem. 2013, 2041–2050
2048
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
diatomaceous earth. Crystallization by slow evaporation of the sol-
vent at ambient temperature yielded X-ray quality, colourless crys-
tals of a solvate of composition 7g·THF [m.p. 156–159 °C.
C26H42BN2P·C4H8O (496.52): calcd. C 72.57, H 10.15, N 5.64;
found C 72.18, H 10.17, N 6.24]. Complete removal of the solvent
yielded 90 mg (yield 87%) of an off-white solid. 31P NMR (C6D6):
ture was solved (direct methods and difference Fourier techniques)
and refined by using SHELXL.[20] All hydrogen-atom positions
were idealized and were allowed to ride on the atom to which they
were attached. The final refinement included anisotropic tempera-
ture factors on all non-hydrogen atoms. The results of the ad-
ditional studies on 4g and 7g are given in the Supporting Infor-
mation
1
δ = 81.7 (br. m, JP,H = 355 Hz) ppm. 11B{1H} NMR (C6D6): δ =
–40.2 (br. d, 1JP,B = 70 Hz) ppm. 1H NMR (C6D6): δ = 7.45 (br. d,
3a: Yellow crystals, C20H44N4P2, M: 402.53 gmol–1, crystal size:
1JP,H = 355 Hz, PH), 7.10–7.17 (m, 4 H, m-CH), 6.96–7.06 (m, 2
¯
0.31ϫ0.30ϫ0.14 mm, triclinic, space group P1, a = 6.2158(6) Å,
3
H, p-CH), 3.73 (sept, JH,H = 6.9 Hz, 2 H, CH), 3.12–3.38 (m, 6
b = 9.9562(9) Å, c = 10.3690(8) Å, α = 104.970(4)°, β = 107.243(4)°,
γ = 94.332(5)°, V = 583.95(9) A3, Z = 1, ρcalcd = 1.145 Mgm–3,
F(000) = 222, θmax = 28.28°, μ = 0.198 mm–1, 9622 reflections mea-
sured, 2879 unique reflections [Rint = 0.035] for structure solution
and refinement with 118 parameters, R1 [IϾ2σ(I)] = 0.038, wR2 =
0.082, largest diff. peak and hole: 0.502 and –0.246 eÅ–3.
3
H, NCH2 and CH), 1.45 (d, JH,H = 6.8 Hz, 6 H, CH3), 1.23 (d,
3
3JH,H = 6.8 Hz, 6 H, CH), 1.21 (d, JH,H = 6.9 Hz, 6 H, CH), 1.17
3
(d, JH,H = 6.9 Hz, 6 H, CH) ppm. 13C{1H} NMR (C6D6): δ =
24.0 (s, CH3), 24.4 (s, CH3), 25.0 (s, CH3), 25.7 (s, CH3), 28.5 (s,
CH), 29.3 (s, CH), 52.1 (s, NCH2), 124.1 (d, JP,C = 1.3 Hz, C6H3),
124.8 (d, JP,C = 0.5 Hz, C6H3), 128.6 (d, JP,C = 1.5 Hz, C6H3), 135.0
(d, JP,C = 6.8 Hz, C6H3), 149.2 (d, JP,C = 1.6 Hz, C6H3), 149.8 (d,
3h: Light yellow crystals, C44H60N4P2, M = 706.90 gmol–1, crystal
size: 0.21ϫ0.12ϫ0.08 mm, monoclinic, space group C2/c, a =
29.019(5) Å, b = 7.0404(15) Å, c = 22.511(4) Å, β = 122.750(5)°, V
= 3868.0(13) Å3, Z = 4, ρcalcd = 1.214 Mgm–3, F(000) = 1528, θmax
= 28.28°, μ = 0.149 mm–1, 15184 reflections measured, 4755 unique
reflections [Rint = 0.110] for structure solution and refinement with
227 parameters, R1 [IϾ2σ(I)] = 0.062, wR2 = 0.115, largest diff.
peak and hole: 0.335 and –0.351 eÅ–3.
JP,C = 2.7 Hz, C H ) ppm. IR (KBr): ν = 2429, 2373, 2346 (BH ),
˜
6
3
3
2255 (PH) cm–1.
1,3-Bis(2,6-diisopropylphenyl)-1,3,2-diazaphospholidine Tris(pen-
tafluorophenyl)borane (8g): In a 20 mL scintillation vial, 3g (30 mg,
37 μmol) was dissolved in toluene (5 mL). With rapid stirring, a
solution of B(C6F5)3 (37.5 mg, 73 μmol) in toluene (5 mL) was
added dropwise over 5 min. The orange solution was allowed to
stir for 30 min, at which point it transitioned to colourless. The
solution was filtered through a pad of diatomaceous earth, and
crystallization by slow evaporation of the solvent at ambient tem-
4g: Colourless crystals, C26H39N2P M = 410.56 gmol–1, crystal size:
0.49ϫ0.26ϫ0.25 mm, monoclinic, space group P21/n,
a =
19.6355(13) Å, b = 6.4031(4) Å, c = 20.3670(14) Å, β = 107.602(3)°,
V = 2440.8(3) A3, Z = 4, ρcalcd = 1.117 Mgm–3, F(000) = 896, θmax
= 28.28°, μ = 0.127 mm–1, 38687 reflections measured, 5991 unique
reflections [Rint = 0.039] for structure solution and refinement with
256 parameters, R1 [IϾ2σ(I)] = 0.043, wR2 = 0.102, largest diff.
peak and hole 0.406 and –0.309 eÅ–3.
perature yielded X-ray quality, colourless crystals. Yield: 55 mg
1
(49%); m.p. 145–148 °C. 31P NMR (C6D6): δ = 66.8 (br. d, JP,H
=
1
404 Hz) ppm. H NMR (C6D6): δ = 8.35 (br. s, 0.5 H, PH) 7.13–
6.80 (br. m, 6 H, Ar-H), 3.54–2.86 (br. m, 8 H, CH/NCH2), 1.16–
0.84 (br. m, 24 H, CH3) ppm. 13C{1H} NMR (C6D6): δ = 150.30,
149.24, 147.34, 146.48, 139.02, 135.82, 129.06, 124.88, 54.68, 52.36,
29.63, 28.54, 28.38, 26.50, 24.17, 22.32 ppm. 11B NMR (C6D6): δ
= 74.3 (br. s) ppm. 19F NMR (C6D6): the signals were prohibitively
6g: Colourless crystals, C30H50AlN2OP, M = 512.35 gmol–1, crystal
size: 0.25ϫ0.21ϫ0.13 mm, monoclinic, space group P21/c, a =
11.4877(14) Å, b = 16.524(2) Å, c = 16.392(2) Å, β = 97.247(5)°, V
= 3086.8(7) A3, Z = 4, ρcalcd = 1.103 Mgm–3, F(000) = 1120, θmax
= 26.35°, μ = 0.141 mm–1, 23075 reflections measured, 6279 unique
reflections [Rint = 0.092] for structure solution and refinement with
328 parameters, R1 [IϾ2σ(I)] = 0.060, wR2 = 0.108, largest diff.
peak and hole: 0.734 and –0.357 eÅ–3.
broad at ambient temperature. IR (KBr): ν = 2314 (PH) cm–1
.
˜
C44H39BF15N2P (922.57): calcd. C 57.28, H 4.26, N 3.04; found C
57.13, H 4.53, N 2.99.
Crystal Structure Determinations: Diffraction studies on 3a, 3h, 4g,
6g and 7g were carried out with a Nonius Kappa-CCD dif-
fractometer at 100(2) K with Mo-Kα radiation (λ = 0.71073 Å).
The structures were solved by direct methods (SHELXS-97[20]) and
refined with a full-matrix least-squares scheme on F2 (SHELXL-
97[20]). Semi-empirical absorption corrections were applied for 3a,
3h and 4g. Non-hydrogen atoms were refined anisotropically, H
atoms on phosphorus, boron, or aluminium atoms in 4g, 6g, and
7g were refined isotropically, and H atoms in CH bonds were re-
fined with a riding model on F2. The hydrogen atoms attached to
the phosphorus atoms in 4g and 7g were located and refined freely
using isotropic thermal displacement parameters.
7g: Colourless crystals, C26H42N2PB, M = 424.4 gmol–1, crystal
size: 0.46ϫ0.22ϫ0.19 mm, orthorhombic, space group Pnma, a =
11.9870(13) Å, b = 20.809(2) Å, c = 10.5990(9) Å, V = 2643.7(5)
A3, Z = 4, ρcalcd = 1.066 Mgm–3, F(000) = 928, θmax = 26.38, μ =
0.118 mm–1, 17895 reflections measured, 2775 unique reflections
[Rint = 0.0795] for structure solution and refinement with 149 pa-
rameters, R1 [IϾ2σ(I)] = 0.046, wR2 = 0.091, largest diff. peak
and hole: 0.223 and –0.265 eÅ–3.
8g: Colourless crystals, C44H39BF15N2P, M = 922.55 gmol–1, crys-
tal size: 0.33ϫ0.31ϫ0.15 mm, monoclinic, space group P21/c, a =
11.410(3) Å, b = 17.962(5) Å, c = 20.961(6) Å, β = 99.656(4)°, V =
Crystals of 4g, 7g and 8g were mounted from Paratone-N oil in a
nylon cryoloop or on a MiTeGen MicroMount. The data were col-
lected at 130(2) K (7g) or 150(2) K (4g, 8g) with a Siemens/Bruker
diffractometer equipped with an APEX II CCD-detector and a
KRYO-FLEX cooling device with Mo-Kα radiation (λ = 0.71073
A). A hemisphere of data was collected with ω scans and with
frame exposures of 5 to 30 seconds and 0.3° frame widths. Data
collection and initial indexing and cell refinement were handled
4235(2) A3, Z = 4, ρcalcd = 1.447 Mgm–3, F(000) = 1888, θmax
=
28.15°, μ = 0.166 mm–1, 41000 reflections measured, 7462 unique
reflections [Rint = 0.037] for structure solution and refinement with
706 parameters (906 restraints), R1 [IϾ2σ(I)] = 0.044, wR2 =
0.113, largest diff. peak and hole: 0.470 and –0.331 eÅ–3.
CCDC-913787 (for 3a), -913803 (for 3h), -913802 (for 4g), -913789
with the APEX II software.[21] Frame integration, including Lo- (for 6g), -913868 (for 7g), and -912900 (for 8g) contain the supple-
rentz-polarization corrections, and final cell parameter calculations
were carried out with the SAINT+ software.[21] The data were cor-
rected for absorption by using the SADABS program.[21] The struc-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Eur. J. Inorg. Chem. 2013, 2041–2050
2049
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[6] N. A. Giffin, A. D. Hendsbee, T. L. Roemmele, M. D.
Lumsden, C. C. Pye, J. D. Masuda, Inorg. Chem. 2012, 51,
11837–11850.
[7] A. Haghverdi, Dissertation, University of Bonn, Germany,
2000.
[8] R. Edge, R. J. Less, E. J. L. McInnes, K. Müther, V. Naseri,
J. M. Rawson, D. S. Wright, Chem. Commun. 2009, 1691–1693.
[9] D. Förster, H. Dilger, F. Ehret, M. Nieger, D. Gudat, Eur. J.
Inorg. Chem. 2011, 3989–3994.
Supporting Information (see footnote on the first page of this arti-
cle): Synthetic procedures and crystallographic and spectroscopic
data for 2c,h; spectroscopic data for 4h; additional crystallographic
data for 4g and 7g, crystallographic data for an oxidation product
of 8g.
Acknowledgments
[10] Some derivatives containing six-membered rings have also been
described, see: a) I. V. Komlev, A. I. Zavalishina, I. P. Chernike-
vich, D. A. Predvoditelev, E. E. Nifant’ev, Zh. Obshch. Khim.
1972, 42, 802; b) E. E. Nifantiev, N. S. Vyazakin, S. F. Sorok-
ina, L. A. Vorobiev, O. A. Vyazankina, D. A. Bravo-Zhivotov-
sky, A. R. Bekker, J. Organomet. Chem. 1984, 277, 211–225.
[11] M. K. Denk, S. Gupta, A. J. Lough, Eur. J. Inorg. Chem. 1999,
41–49.
Financial support by Deutsche Forschungsgemeinschaft (DFG),
Deutscher Akademischer Austauschdienst (DAAD) (to D. F.) and
the Academy of Finland (to M. N.) is gratefully acknowledged. The
authors thank Dr. Wolfgang Frey (Institut für Organische Chemie,
University of Stuttgart) for measuring X-ray diffraction data sets
and J. Trinkner and K. Wohlbold (Institut für Organische Chemie,
University of Stuttgart) for recording the mass spectra.
[12] C. J. Carmalt, V. Lomeli, B. G. McBurnett, A. H. Cowley,
Chem. Commun. 1997, 2095–2096.
[13] S. Burck, D. Gudat, M. Nieger, W.-W. du Mont, J. Am. Chem.
Soc. 2006, 128, 3946–3955.
[1] a) H. Nöth, H.-J. Vetter, Naturwissenschaften 1960, 204; b) H.
Nöth, H.-J. Vetter, Chem. Ber. 1961, 94, 1505; c) H. West-
ermann, M. Nieger, E. Niecke, Chem. Ber. 1991, 124, 13–16; d)
J. Niemann, W. W. Schoeller, V. von der Gönna, E. Niecke,
Chem. Ber. 1991, 124, 1563–1565; e) H. R. G. Bender, E.
Niecke, M. Nieger, H. Westermann, Z. Anorg. Allg. Chem.
1994, 620, 1194–1202; f) W. Domanska-Babul, J. Chojnacki, E.
Matern, J. Pikies, J. Organomet. Chem. 2007, 692, 3640–3648;
g) R. Grubba, L. Ponikiewski, J. Chojnacki, J. Pikies, Acta
Crystallogr., Sect. E 2009, 65, o2214; h) R. Grubba, A. Wis-
niewska, K. Baranowska, E. Matern, J. Pikies, Dalton Trans.
2011, 40, 2017.
[14] S. Burck, D. Gudat, M. Nieger, Angew. Chem. 2004, 116, 4905–
4908; Angew. Chem. Int. Ed. 2004, 43, 4801–4804.
[15] D. Gudat, A. Haghverdi, M. Nieger, Angew. Chem. 2000, 112,
3211–3214; Angew. Chem. Int. Ed. 2004, 39, 3084–3086.
[16] a) R. B. King, N. D. Sadanani, P. M. Sundaram, Phosphorus
Sulfur Relat. Elem. 1983, 18, 125; b) R. B. King, N. D. Sadan-
ani, P. M. Sundaram, J. Chem. Soc. , Chem. Commun. 1983,
477.
[17] R. T. Paine, H. Nöth, Chem. Rev. 1995, 95, 343–479.
[18] a) S. J. Lancaster, A. J. Mountford, D. L. Hughes, M. Schorm-
ann, M. Bochmann, J. Organomet. Chem. 2003, 680, 193–205;
b) G. C. Welch, T. Holtrichter-Roessmann, D. W. Stephan, In-
org. Chem. 2008, 47, 1904–1906; c) G. C. Welch, R. Prieto,
M. A. Dureen, A. J. Lough, O. A. Labeodan, T. Holtrichter-
Rossmann, D. W. Stephan, Dalton Trans. 2009, 1559–1570.
[19] S. Stoll, A. Schweiger, J. Magn. Reson. 2006, 178, 42–55.
[20] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[21] APEX II, SAINT+ and SADABS, Bruker AXS Inc., Madison,
WI 2008.
[2] a) J.-P. Bezombes, P. B. Hitchcock, M. F. Lappert, J. E. Nycz,
Dalton Trans. 2004, 499–501; b) J.-P. Bezombes, K. B. Bori-
senko, P. B. Hitchcock, M. F. Lappert, J. E. Nycz, D. W. H.
Rankin, H. E. Robertson, Dalton Trans. 2004, 1980–1988.
[3] M. J. S. Gynane, A. Hudson, M. F. Lappert, P. P. Power, H.
Goldwhite, J. Chem. Soc., Dalton Trans. 1980, 2428.
[4] B. M. Cossairt, C. C. Cummins, J. Am. Chem. Soc. 2009, 131,
15501.
[5] O. Puntigam, I. Hajdók, M. Nieger, M. Niemeyer, S. Strobel,
D. Gudat, Z. Anorg. Allg. Chem. 2011, 637, 988–994.
Received: December 5, 2012
Published Online: February 11, 2013
Eur. J. Inorg. Chem. 2013, 2041–2050
2050
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim