Specific and reversible alkynyl transfer reactions of an n-heterocyclic phosphane

Article abstract of DOI:10.1002/ejic.201301141

Reaction of P-chloro-1,3,2-diazaphospholenes with alkynyllithium or Grignard reagents produced N-heterocyclic alkynylphosphanes 3a-c. Terminal alkynylphosphane 3c reacted in the presence of a catalytic amount of CuCl with formation of bis-phosphanylalkyne 4, ethyne and chlorodiazaphospholene 2b, which was converted into the appropriate azide 6 upon addition of LiN₃. Detailed studies allowed us to explain the formal alkynyl/Cl metathesis and alkynyl scrambling as arising from a combination of Cu-catalysed cross-coupling steps and the reverse processes, which involved Cu-induced P-C bond breaking. The superposition of individual steps allows the formulation of an equilibrium reaction: 2 3c [lrarr2] 4 + HCCH. Shift of the equilibrium by enforced release of acetylene into the gas phase enabled quantitative conversion of 3c into 4. Traces of P-C bond-activation products were also detected upon action of CuCl on 3b with an internal triple bond, but alkynylphosphanes with Ph₂P or (iPr₂N)₂P moieties were unreactive. Alkynylphosphanes 3a-c and 4 were characterised by spectroscopic data and single-crystal X-ray diffraction studies. The presence of notably elongated P-C bonds points to a bond-weakening effect, which is responsible for explaining the unusual P-C bond reactivity.

Full text of DOI:10.1002/ejic.201301141

FULL PAPER
DOI:10.1002/ejic.201301141
Specific and Reversible Alkynyl Transfer Reactions of an
N-Heterocyclic Phosphane
CLUSTER
ISSUE
Michael Gediga,[a] Sebastian Burck,[a] Johannes Bender,[a]
Daniela Förster,[a] Martin Nieger,[b] and Dietrich Gudat*[a]
Keywords: Phosphanes / Alkynes / C–P Activation / Cross-coupling / Copper
Reaction of P-chloro-1,3,2-diazaphospholenes with alk-
tion: 2 3c p 4 + HCCH. Shift of the equilibrium by enforced
ynyllithium or Grignard reagents produced N-heterocyclic
release of acetylene into the gas phase enabled quantitative
alkynylphosphanes 3a–c. Terminal alkynylphosphane 3c re- conversion of 3c into 4. Traces of P–C bond-activation prod-
acted in the presence of a catalytic amount of CuCl with for-
mation of bis-phosphanylalkyne 4, ethyne and chlorodiaza-
phospholene 2b, which was converted into the appropriate
azide 6 upon addition of LiN3. Detailed studies allowed us to
explain the formal alkynyl/Cl metathesis and alkynyl scram-
bling as arising from a combination of Cu-catalysed cross-
ucts were also detected upon action of CuCl on 3b with an
internal triple bond, but alkynylphosphanes with Ph2P or
(iPr2N)2P moieties were unreactive. Alkynylphosphanes 3a–
c and 4 were characterised by spectroscopic data and single-
crystal X-ray diffraction studies. The presence of notably
elongated P–C bonds points to a bond-weakening effect,
coupling steps and the reverse processes, which involved which is responsible for explaining the unusual P–C bond
Cu-induced P–C bond breaking. The superposition of indi-
vidual steps allows the formulation of an equilibrium reac-
reactivity.
ible). However, whereas reversal under fission of an initially
formed P–Y bond is easily accomplished with many halo-
gen-, oxygen- or nitrogen-based substituents (e.g., a P–N
bond created in a previous reaction of a halophosphane
with a metal amide or an amine might be cleaved upon
reaction with a hydrogen halide), P–C bonds are usually
considered kinetically inert, and the possibility to reverse a
P–C bond-formation process is often neglected.
Introduction
Metathesis reactions that involve displacement of a nega-
tively polarised substituent X at an electrophilic trivalent
phosphorus atom by an organometallic nucleophile, usually
an organolithium or Grignard reagent (Scheme 1, a), are
extremely versatile and powerful tools for P–C bond cre-
ation and are commonly employed in the syntheses of a
wide range of organophosphanes.[1] The leaving group X is
in most cases a halide, although alkoxy/phenoxy[2] or amino
substituents,[3] for example, have been used as well. For-
mally, these P–C bond-formation reactions can be consid-
ered a special case of a general substituent displacement
reaction at phosphorus (Scheme 1, b), which might techni-
cally be considered an equilibrium process that is in prin-
ciple reversible under suitable conditions (it should be
noted that these considerations remain valid even if one
considers that many P–C bond-formation reactions rely on
catalytic processes and are mechanistically more complex
than a simple nucleophilic substitution, since the individual
steps in a catalytic cycle must likewise be considered revers-
Scheme 1. Schematic representation of selected ligand metathesis
processes at a tricoordinate phosphorus atom (R = hydrocarbyl; X,
Y = halogen, OR, NR2; M = Li, Mg).
Upon closer examination, specific examples of such reac-
tions are well documented. Thus, transition-metal-induced
bond activation plays an important role in the degradation
of phosphane ligands in complexes,[4] and a formal meta-
thesis accomplished by means of reductive P–C bond cleav-
age by a stoichiometric amount of an alkaline metal and
subsequent quenching of the metal phosphides formed by a
suitable electrophile is even of some synthetic significance.[5]
Furthermore, scrambling of halogen and hydrocarbyl sub-
stituents between different phosphorus atoms (Scheme 1, c)
[a] Institut für Anorganische Chemie, University of Stuttgart,
Pfaffenwaldring 55, 70550 Stuttgart, Germany
E-mail: gudat@iac.uni-stuttgart.de
http://www.iac.uni-stuttgart.de/arbeitskreise/akgudat
[b] Laboratory of Inorganic Chemistry, Department of Chemistry,
University of Helsinki,
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.201301141.
Eur. J. Inorg. Chem. 2014, 1818–1825
1818
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
has been employed to access special halophosphanes (e.g., carbon–phosphorus frameworks.[11] Borane adducts of alk-
in the generation of PhPCl2 from Ph3P and PCl3).[6] More ynylphosphanes were recently used as reagents[12] in the
explicitly, both mechanistically unspecified loss of a carbon- CuI-catalysed alkyne/azide cycloaddition reaction generally
based substituent at phosphorus[7] and specific metathesis[8] referred to as a “click reaction”.[13] To explore the reactivity
have in particular been observed for phosphane derivatives of P-alkynylated N-heterocyclic phosphanes, we prepared
that bear cyclopentadienyl substituents (Cp or Cp*). These derivatives 3a–c by means of alkynylation of the appropri-
reactions rely crucially on the ability of the carbon-based ate chloro-diazaphospholene precursors 2a,b with alk-
substituents to form stable cyclopentadienyl radicals or ynyllithium or Grignard reagents (Scheme 3).
anions as good leaving groups. Genuine metathesis pro-
The products were isolated in good to excellent yields
cesses under selective transfer of a Cp moiety are especially after standard workup and characterised by spectroscopic
facilitated if the R2P+ fragment formed by cleavage of Cp– data and single-crystal X-ray diffraction studies. The mo-
can also be efficiently stabilised, as is the case in Cp-substi- lecular structures (a graphical representation is displayed in
tuted N-heterocyclic phosphanes such as 1, which react Figure 1 for 3c and in the Supporting Information for 3a,b;
with complex transition-metal halides under specific trans- important bond lengths and angles of all three compounds
fer of the Cp substituent (Scheme 2).[8b]
are listed in Table 1) contain nonplanar five-membered
PN2C2 rings that show the conformational preferencesϪa
flat envelope conformation with quasi-planar and distinctly
pyramidal coordination at the nitrogen and phosphorus
atoms, and P substituents that adopt a flagpole posi-
tionϪthat have been identified as typical features of these
heterocycles.[9]
Scheme 2. Cp metathesis reactions of an N-heterocyclic phosphane
(Mes = mesityl; Cp = C5H5).[8b]
During extended studies of the chemistry of these
N-heterocyclic phosphanes,[9] we also began to explore the
reactivity of some P-alkynyl-substituted derivatives
3
(Scheme 3). In this context, we came across another exam-
ple of a P–C bond-fission process,[10] which involves the
Cu+-catalysed metathetic exchange of an alkynyl substitu-
ent in a N-heterocyclic phosphane by a halide or another
alkynyl moiety in a reaction that should be considered a
reversible equilibrium process. Transformations of this type
are to the best of our knowledge unprecedented in organo-
phosphorus chemistry but might be considered to have
interesting implications that stimulate the further develop-
ment of new synthetic methodology in this field.
Figure 1. Molecular structure of 3c. Hydrogen atoms (with the ex-
ception of the H atom in the alkynyl group) were omitted for clar-
ity; thermal ellipsoids were drawn at 50% probability level.
Table 1. Selected bond lengths [Å] and angles [°] for 3a–c and 4.
3a
3b[a]
3c
4[a]
P1–Csp[b]
Csp–Csp
P1–N2
1.801(1)
1.817(2)
[1.815(2)]
1.215(3)
[1.206(3)]
1.706(2)
[1.701(2)]
1.709(2)
[1.712(2)]
1.827(2)
1.823(1)
[1.816(1)]
1.213(2)
1.205(2)
1.700(1)
1.718(1)
1.182(3)
1.704(2)
1.712(2)
1.7038(1)
[1.703(1)]
1.712(1)
P1–N5
[1.718(1)]
Scheme 3. Synthesis of P-alkynyl-1,3,2-diazaphospholenes 3a–c.
P1–Csp –Csp
173.1(1)
164.1(2)
[164.7(2)]
177.5(2)
[177.2(2)]
294.4(3)
[294.0(3)]
168.0(2)
176(2)
164.5(1)
[161.9(1)]
–
Csp –Csp–C/H[c] 178.0(1)
ΣЄ(P)[d]
Results and Discussion
291.5(2)
293.1(3)
295.5(4)
[296.3(4)]
The combination of the phosphane donor moiety with
an adjacent, and possibly interacting, reactive triple bond
makes alkynylphosphanes interesting bifunctional sub-
strates[1] that have been employed, for example, in syntheses
[a] Values in brackets denote the data of a second crystallographi-
cally independent molecule (3b) or the second half of the molecule
(4). [b] Csp denotes the formally sp-hybridised carbon atom in a
triple bond. [c] C–C–C for 3a,b and C–C–H for 3c. [d] Sum of bond
of special phosphorus heterocycles and highly unsaturated angles at P1.
Eur. J. Inorg. Chem. 2014, 1818–1825
1819
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
The endocyclic P–N distances [1.700(1) to 1.718(1) Å] mained incomplete even after extended reaction time. Near-
range at the lower end of the span established for derivatives quantitative transformation was achievable, however, by
1
with P-amino, phosphanyl and cyclopentadienyl substitu- conducting the reaction in THF heated at reflux. The H
ents (1.70–1.74 Å). At the same time, the P–C single bond NMR spectroscopic signals of acetylene were found to van-
lengths of 1.801(1) to 1.827(2) Å are longer than usual: re- ish almost completely under these conditions (as a result of
ported P–Csp bonds in alkynylphosphanes range from 1.728 eventual evaporation into the gas phase). Taken together,
to 1.795 Å (with a mean of 1.768 Å[15]), which is close to these findings lead us to infer that acetylene and 4 exist
the computationally predicted P–Csp single-bond length[11] in a dynamic equilibrium with 3c, which obeys the general
of 1.75 Å. Taken together, the combination of P–N bond equation [Equation (1)].
contraction and P–C bond elongation points to the pres-
2 3c p 4 + HCϵCH
(1)
ence of a certain degree of n(N)Ǟσ*(PC) hyperconjugation,
which is expected to exert a weakening effect on the P–C
bond.[9,14]
The observed (near) quantitative conversion of 3c to 4
upon removal of acetylene reflects simply another manifes-
Reaction of 3c with various amounts of CuCl (5 to tation of Le Chatelier’s principle. The reversibility of the
50 mol-%) at room temperature in THF produced mixtures reaction can be deduced from the fact that formation of
that contained unreacted 3c alongside three further phos- both 4 and acetylene from 3c requires that P–C bonds are
phorus-containing species detectable by 31P NMR spec- not only newly formed but also broken under CuI catalysis,
troscopy. The major new reaction product was isolated after which allows the overall reaction to proceed in either direc-
suitable adaptation of the reaction conditions (see below) tion.
and later identified as alkynylbisphosphane 4 (Scheme 4).
Although Equation (1) describes the stoichiometric com-
A second species could neither be isolated nor comprehen- position of the reaction mixture, it does not provide infor-
sively characterised, but exhibited the same 31P chemical mation on the mechanistic aspects of elementary reaction
shift (δ31P = 118 ppm) and an identical cross-peak pattern steps. In light of the report by Beletskaya et al.,[16] it seems
in a 1H,31P HMQC spectrum as the product that arose from reasonable to explain the formation of 4 by means of CuI-
the reaction of 2b with 4-chlorobutanol, and was on this catalysed cross-coupling of chlorophosphane 2b (Scheme 5,
basis assigned as alkoxy-substituted N-heterocyclic phos- reaction b), which forms one of the minor constituents of
phane 5 (Scheme 4). The remaining product was readily the reaction mixture, with 3c. We were able to verify this
identified as known chlorodiazaphospholene 2b. Reaction assumption by establishing that stoichiometric amounts of
1
monitoring by H NMR spectroscopy (the reaction was in 3c and 2b indeed react in the presence of a catalytic quan-
this case carried out by using [D8]THF as solvent) disclosed tity of CuCl with Ͼ75% conversion (determined by 31P
the formation of acetylene as further product, which was NMR spectroscopy) to yield 4 and HCl, which was detected
positively identified by correlation signals in an 1H,13C by precipitation as sparingly soluble [Et3NH]Cl after ad-
HMBC spectrum of the reaction mixture (NMR spectro- dition of Et3N. Action of HCl on THF to give traces of 4-
1
scopic data: δ1H = 2.25 ppm, δ13C = 72.3 ppm, JC,H
249 Hz, JC,H = 50 Hz, JH,H = 10 Hz).
=
chlorobutanol and subsequent quenching of this product
by 2b (Scheme 5, reaction f) is further likely to explain the
occurrence of 5 as byproduct in reactions of 3c with CuCl.
Assuming the validity of this hypothesis, it should be feas-
ible to suppress any formation of 5 by conducting reactions
of 3c with CuCl from the beginning in the presence of Et3N
as acid scavenger, which was indeed observed.
2
3
In light of the preceding information, chlorophosphane
2b represents not merely a byproduct but an essential reac-
tion intermediate. Its formation can be consistently ration-
alised as the outcome of a metathesis reaction of 3c with
CuCl to give the observed chlorodiazaphospholene and
copper acetylide CuCCH, which is known to undergo a
scrambling reaction under generation of Cu2C2 and acetyl-
ene at temperatures above –40 °C (Scheme 5, reaction d).[17]
Alternatively, it is conceivable that 2c might also be formed
together with Cu2C2 through the metathesis of 4 with CuCl
(Scheme 5, reactions a and c). We were able to confirm both
routes in studies of reactions of 4 and 3c with stoichiomet-
ric amounts of CuCl.[18] Reaction monitoring by 31P NMR
spectroscopy showed that 2b was in the first case detectable
Scheme 4. Observed products in reactions of 3c with varying
amounts of CuCl [no quantitative reaction equation; R = 2,6-di-
isopropylphenyl (Dipp)].
Closer investigation of the individual reactions revealed as only phosphorus-containing species alongside unreacted
that although the amount of 5 formed exceeded the quan- starting material and eventually became the main product.
tity of CuCl employed (thus indicating that the reaction is The reaction of 3c with CuCl gave similar results, apart
catalytic in CuCl), conversion of the starting material re- from the fact that in this case also 4 was formed. Both reac-
Eur. J. Inorg. Chem. 2014, 1818–1825
1820
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Scheme 6. Formation of azidophosphane 6 (R = Dipp).
sults of these considerations can be condensed in the reac-
tion sequence shown in Scheme 5, each individual step of
which is either based on literature precedence or has been
explicitly verified by experiments. The key to the under-
standing of the formal alkyne metathesis between 3c, 4 and
HCCH are the combination of CuI-catalysed P–C cross
coupling[16] (Scheme 5, reaction b) and bond-breaking steps
(Scheme 5, reactions a and c) with a proton-transfer reac-
tion (Scheme 5, reaction d), which add up to the equilib-
rium reaction of Equation (1). Hydrogen chloride formed
in (b) is quenched by copper acetylide (Scheme 5, reaction
e) but might promote the (catalytic) conversion of 2b into
side product 5 (Scheme 5, reaction f) unless scavenged by a
suitable additive (Et3N or LiN3). The intermediate 2b can
Scheme 5. Individual steps involved in the CuI-catalysed transfor-
mation of 3c to give 2b, 4 and 5 (R = Dipp).
–
also be intercepted by an added nucleophile (N3 ).
Adopting this mechanistic explanation leads to the pre-
diction that the amount of an interception product (e.g., 6)
depends on the quantity of chloride sources present and
can be deliberately increased by using high loadings of
CuCl and LiCl. On the other hand, optimum conditions for
the specific synthesis of 4 are obtained by conducting the
reaction with low catalyst load (to minimise losses due to
formation of 2b or its interception products), at high tem-
perature (to release HCCH) and in the presence of an acid
scavenger (for further reduction of byproducts). Following
these ideas enabled us to accomplish quantitative conver-
sion of 3c into azidophosphane 6 in a process that adds up
to an unusual P–C/P–N bond metathesis, and to generate 4
from 3c with Ͼ90% selectivity (by integration of the 31P
NMR spectrum, which also revealed the absence of any
other byproducts except a small amount of 6; see Figure 2
and the Supporting Information). The bis-phosphanyl alk-
yne 4 was isolated in decent overall yield (59%) after crys-
tions produced further an amorphous, brownish precipitate
that was isolated from the reaction of 4 with CuCl by fil-
tration. Analytical characterisation proved the presence of
copper (positive qualitative test on Cu) and acetylide ions
(positive identification of acetylene in 1H and 1H,13C
HMBC spectra of an extract in acidified [D8]THF) and led
us to infer the presence of copper acetylides (CuC2H or
Cu2C2).
Extended studies allowed us to establish that the selectiv-
ity of the reaction of 3c with CuCl can also be improved by
using additives other than Et3N. In particular, the addition
of a mixture of NaN3 and LiCl (as a phase-transfer catalyst
to solubilise azide ions through conversion of insoluble
NaN3 into soluble LiN3 and NaCl) proved to be highly ef-
ficient in reducing the number and relative amount of side
products. Apart from scavenging strongly acidic compo-
nents (HCl), the azide ions promoted conversion of 2b into
azidophosphane 6 (Scheme 6) the identity of which was es-
tablished by independent synthesis from 2b and NaN3 (see
the Experimental Section). Further control experiments
indicated that neither LiN3 nor 6 are directly involved in
the interconversion of 3c, 4 and HCCH, and that Cl/N3
metathesis provides the only viable access route to 6. Since
thermal decomposition of azidodiazaphospholenes requires
considerably higher temperatures,[8b] this reaction is highly
selective under the conditions employed. Additional studies
indicated that 3c fails also to undergo any CuI-catalysed
alkyne/azide cycloaddition reactions with LiN3 or various
organic azides.
Putting the results of the individual experiments together
allows us to propose a consistent mechanistic explanation
for both the unusual interconversion between alkynylphos-
phanes 3c and 4 by means of reversible alkynyl scrambling,
and the transformation of 3c into chlorophosphane 2b
Figure 2. Stacked plot showing expansions of the 31P{1H} NMR
spectra of the reaction of 3c with CuCl (5 mol-%) in the presence
of 1 equiv. of NaN3. The displayed region contains the resonances
of the alkynylphosphanes (3c, 4). Traces represent spectra obtained
after reaction times of (from bottom to top) 0, 1, 2, 3 and 4 d at
through unprecedented P–C/P–Cl bond metathesis. The re- 66 °C.
Eur. J. Inorg. Chem. 2014, 1818–1825
1821
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
tallisation. Characterisation was achieved by spectroscopic 28.8) than acetylene (pKa 25), which indicates that destabili-
studies and a single-crystal X-ray diffraction study (Fig- sation of the acetylide anion fragment might also help to
ure 3), which revealed the product to have similar spectro- counterbalance the bond-weakening effect.
scopic and structural features as 3a–c (Table 1).
Conclusion
We have established that P-alkynyl-substituted N-hetero-
cyclic phosphanes, which are readily accessible from reac-
tions of P-chloro-substituted precursors with alkynyllith-
ium and Grignard reagents, respectively, undergo meta-
thetic exchange of alkynyl substituents in the presence of
CuCl. The key to this reactivity is a sequence of P–C bond-
formation and bond-breaking steps that produce a dynami-
cally exchanging equilibrium mixture of mono- and bis-
phosphanyl alkynes and acetylene. Bond formation pro-
ceeds by means of (previously known[16]) Cu-catalysed
cross-coupling of a chlorophosphane with a terminal alk-
yne. The bond-breaking step represents the as-yet unprece-
Figure 3. Molecular structure of 4. Hydrogen atoms were omitted
dented reversal of this process, which illustrates nicely the
for clarity; thermal ellipsoids were drawn at the 50% probability
reversibility of the cross-coupling reaction. Detailed investi-
level.
gations of these reactions enabled us to identify conditions
that allowed specific displacement of the alkynyl by an
After realising that reversible alkynyl transfer between
azido group (which is a rare example for the transformation
different R2P fragments might be of wider synthetic inter-
of a P–C into a P–N bond) and a quantitative conversion
est, we wanted to probe if analogous reactivity can also be
of the starting material into an isolable bis-phosphanyl alk-
observed for other alkynylphosphanes than 3c. In fact, we
were able to establish that formation of a small amount of
2b was spectroscopically detectable upon prolonged reac-
tion of 3b with CuCl (20 mol-%) at elevated temperature
yne by means of trans-alkynylation. The (current) limitation
of the P–Csp bond activation to alkynyldiazaphospholenes
suggests that the proven ability of the N-heterocyclic frame-
work to exert a weakening effect on an adjacent single
(72 h in THF heated to reflux). Although the low degree
bond[9] might induce not only structural distortions (which
of conversion (2–3%) even under more forceful conditions
attests that 3b has a much lower activity than 3c, we con-
become manifest in the distinct elongation of the P–Csp
bonds) but might also be crucial for explaining the unusual
sider this result an important proof of principle that con-
veys that CuI-catalysed P–C bond cleavage is not limited to
phosphanes with terminal alkyne functionalities, but might
in principle also be observed for those with internal triple
bonds. In contrast to 3b,c, alkynylphosphanes RЈ2P–CϵCR
(R = H, Ph) with diphenylphosphanyl (RЈ = Ph) or acyclic
diaminophosphanyl moieties (R = iPr2N) were found to be
inactive and gave no evidence for any CuI-catalysed P–C
bond activation up to temperatures of 110 °C.
reactivity.
We are currently pursuing a search for more active cata-
lysts, which might help to widen the currently quite limited
scope of phosphane transalkynylation reactions, and cocat-
alysts that might permit us to activate the reputedly unreac-
tive copper acetylides for further in situ cross-coupling reac-
tions.
In evaluating these findings, it seems that P–C bond-acti-
vation processes are highly sensitive to effects that result
from substituent-induced tuning of the electronic structure
of the substrates. We relate the high activity of 3b,c in alk-
ynyl scrambling reactions to the P–C bond weakening,
which can be deduced from the observed bond elongation
Experimental Section
General Remarks: All reactions were carried out under an atmo-
sphere of dry argon. Solvents were dried by standard procedures.
NMR spectra: Bruker Avance 250 (1H, 250.0 MHz; 13C, 62.9 MHz;
31P, 101.2 MHz) or Bruker Avance 400 (1H, 400.1 MHz; 13C,
in the crystal structures and has been consistently explained 100.5 MHz; 31P, 161.9 MHz) spectrometers at 303 K if not men-
tioned otherwise. Chemical shifts were referenced to external TMS
(1H, 13C) or 85% H3PO4 (Ξ = 40.480747 MHz, 31P). Coupling con-
stants are given as absolute values. The prefixes i, o, m and p denote
atoms of aryl substituents. Elemental analyses were carried out
with Perkin–Elmer 2400CHSN/O or Elementar Micro Cube ana-
lysers. The rather large deviations between found and calculated
element composition of 3c and 6 are due to the presence of residual
amounts of inorganic chlorides in the bulk samples that could only
be removed by repeated recrystallisation. The presence of these im-
purities is considered to have no influence on the reported results
as a consequence of the unique propensity of the N-hetero-
cyclic framework to induce hyperconjugative destabilisation
and concomitant polarisation of adjacent P–X bonds. This
effect has previously been found to be crucial for the high
chemical activity of the P–C and P–P bonds in Cp- and
phosphanyl-substituted N-heterocyclic phosphanes.[9,14]
The lower activity of 3b than 3c most likely reflects differ-
ences in the Cu–C(alkyne) interaction, which might be con-
sidered a major driving force for the observed reactivity, but
also correlates with a lower acidity of phenyl acetylene (pKa and conclusions. IR spectra were determined with a Nicolet iS5
Eur. J. Inorg. Chem. 2014, 1818–1825
1822
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
3
3
FTIR spectrometer equipped with a iD5 diamond ATR unit. Mass
spectrometry (EI) was conducted with a Varian MAT 711, 70 eV.
Melting points were determined with a Büchi Point B-545 melting
point apparatus. Compound 2b[19] and alkynylphosphanes RЈ2P –
CϵCR (RЈ = Ph, iPr2N; R = H, Ph)[20] were synthesised as de-
scribed elsewhere.
H, CH(CH3)2], 2.85 (d, JP,H = 3.4 Hz, 1 H, CCH), 1.41 [d, JH,H
3
= 6.8 Hz, 6 H, CH(CH3)2], 1.38 [d, JH,H = 6.8 Hz, 6 H, CH-
3
3
(CH3)2], 1.34 [d, JH,H = 6.8 Hz, 6 H, CH(CH3)2], 1.26 [d, JH,H
=
6.8 Hz, 6 H, CH(CH3)2] ppm. 13C{1H} NMR (CDCl3): δ = 128.4
2
4
(d, JP,C = 7.3 Hz, i-C), 127.5 (d, JP,C = 2.0 Hz, m-C), 125.4 (s, p-
C), 123.9 (s, o-C), 120.2 (d, JP,C = 6.8 Hz, CHN), 88.5 (d, JP,C
115 Hz, PCCH), 88.4 (d, JP,C = 18.2 Hz, PCCH), 28.7 [s, CH-
2
1
=
2
2-Phenylethynyl-1,3-bis(2,6-dimethylphenyl)-1,3,2-diazaphosphol-
ene (3a) and 2-Phenylethynyl-1,3-bis(2,6-diisopropylphenyl)-1,3,2-di-
azaphospholene (3b): A solution of lithiophenylacetylene was pre-
pared by adding nBuLi (4.0 mL of a 2.5 m solution in hexane) to a
solution of phenyl acetylene (1.02 g, 10 mmol) in THF (20 mL) at
–78 °C. The mixture was stirred for 1 h and then warmed to room
temperature. This solution was added dropwise to a solution of
2a (3.32 g, 10 mmol) or 2b (4.42 g, 10 mmol), respectively, in THF
(40 mL) at –78 °C. The reaction mixture was warmed to ambient
temperature and stirred for a further 30 min. Volatiles were re-
moved under vacuum, and the residues were suspended in toluene
(20 mL) and filtered. The solvent was removed again under vac-
uum, and the residue was dissolved in hexane (10 mL). Storage at
–21 °C produced orange crystals.
(CH3)2] 25.9 (s, CH3), 25.3 (s, CH3), 25.1 (s, CH3), 23.8 (s, CH3)
ppm. 31P{1H} NMR (CDCl ): δ = 63.8 (s) ppm. IR: ν = 3294 (w)
˜
3
νϵC–H, 2008 (w) ν(CϵC) cm–1. C28H37N2P (432.59): calcd. C
77.74, H 8.62, N 6.48; C 75.98 H 8.78 N 6.15.
1,2-Bis[1,3-bis(2,6-diisopropylphenyl)-1,3,2-diazaphospholene]-
ethyne (4): Compound 3c (440 mg, 1 mmol) was dissolved in THF
(10 mL). NaN3 (65 mg, 1 mmol) and CuCl (5.0 mg, 50 μmol) were
added to the solution. The mixture was heated to reflux and stirred
for 5 d by maintaining the temperature. After cooling to room tem-
perature, the solvent was removed under vacuum, and the residue
was dissolved in hexane (5 mL) and filtered. Storage of the filtrate
at –21 °C produced orange crystals, yield 249 mg (59 %), m.p.
1
177 °C. H NMR (400 MHz, [D8]THF): δ = 7.39–7.29 (m, 4 H, p-
C6H3), 7.14 (br. m, 8 H, m-C6H3), 5.94 (s, 4 H, NCH), 3.76 [br., 4
1,3-(2Ј,6Ј-Dimethylphenyl)-2-phenylethinyl-1,3,2-diazaphospholene
(3a): Yield 87%, m.p. 106 °C. H NMR (C6D6): δ = 7.30–7.17 (m,
3
H, CH(CH3)2], 3.46 [sept, JH,H = 6.8 Hz, 4 H, CH(CH3)2], 1.26
1
3
3
[d, JH,H = 6.8 Hz, 12 H, CH(CH3)2], 1.14 [d, JH,H = 6.8 Hz, 12
H, CH(CH3)2], 1.35–1.05 [br., 24 H, CH(CH3)2] ppm. 1H NMR
(400 MHz, [D8]THF, 183 K): δ = 7.20 (br. m, 4 H, m/p-C6H3), 7.23
(m, 2 H, m-C6H3), 7.16 (m, 2 H, p-C6H3), 7.11 (m, 2 H, m-C6H3),
7.03 (m, 2 H, m-C6H3), 6.13 (m, 2 H, NCH), 6.06 (m, 2 H, NCH),
3.87 [sept, 3JH,H = 7 Hz, 2 H, CH(CH3)2], 3.55 [sept, 3JH,H = 7 Hz,
2 H, o-Ph), 7.02 (s, 6 H, m/p-C6H3) 6.95–6.85 (m, 3 H, m/p-Ph),
3
5.82 (d, JP,H = 2.5 Hz, 2 H, NCH), 2.61 (br. s, 12 H, CH3) ppm.
2
13C{1H} NMR (C6D6): δ = 140.1 (d, JP,C = 13.6 Hz, i-C), 137.7
4
(br., o-C), 132.1 (d, JP,C = 2.0 Hz, i-C), 129.2 (br. s, m-CH), 128.9
(s, p-CH), 128.4 (br., o-CH), 126.7 (d, 6JP,C = 2.4 Hz, p-CH), 123.1
7
2
(d, JP,C = 2.7 Hz, m-CH), 119.0 (d, JP,C = 6.9 Hz, NCH), 100.7
3
2 H, CH(CH3)2], 3.43 [sept, JH,H = 7 Hz, 2 H, CH(CH3)2], 3.39
2
1
(d, JP,C = 14.7 Hz, PCC), 95.7 (d, JP,C = 110 Hz, PCC), 19.7 (br.,
CH3) ppm. 31P{1H} NMR (C6D6): δ = 58.9 (s) ppm. MS (EI, 70 eV,
430 K): m/z (%) = 396.2 (100) [M+], 295.1 (12) [M+ – CCPh], 132.1
3
3
[sept, JH,H = 7 Hz, 2 H, CH(CH3)2], 1.55 [d, JH,H = 7 Hz, 6 H,
3
3
CH(CH3)2], 1.40 [d, JH,H = 7 Hz, 6 H, CH(CH3)2], 1.26 [d, JH,H
= 7 Hz, 6 H, CH(CH3)2], 1.23 [d, JH,H = 7 Hz, 6 H, CH(CH3)2],
1.15 [d, JH,H = 7 Hz, 6 H, CH(CH3)2], 1.09 [d, JH,H = 7 Hz, 6 H,
CH(CH3)2], 0.94 [br., 6 H, CH(CH3)2], 0.49 [br., 6 H, CH(CH3)2]
ppm. 13C{1H} NMR (100.6 MHz, [D8]THF): δ = 149.1 (s, o-C6H3),
3
(43), 105.0 (30). IR: ν = 2142 (m) ν(CϵC) cm–1. C26H25N2P
˜
3
3
(396.47): calcd. C 78.77, H 6.36, N 7.07; found C 78.28, H 6.59, N
6.79.
2
2-Phenylethynyl-1,3-bis(2,6-diisopropylphenyl)-1,3,2-diazaphos-
147.3 (s, o-C6H3), 136.2 (d, JP,C = 13.3 Hz, i-C6H3), 127.1 (s, p-
C6H3), 124.1 (s, m-C6H3), 123.5 (s, m-C6H3), 119.8 (d, JP, C
1
2
pholene (3b): Yield: 4.43 g (87%). H NMR (250 MHz, C6D6): δ =
=
3
7.30–7.10 (br. m, 6 H, C6H3), 6.86 (m, 5 H, Ph) 5.96 (d, JP,H
=
6.8 Hz, NCH), 109.8 (dd, J = 125.1/24.9 Hz, PC), 28.5 [m,
CH(CH3)2], 28.2 [br. s, CH(CH3)2], 24.5 [s, CH(CH3)2], 23.3 [m,
CH(CH3)2] ppm. 31P{1H} NMR (162.9 MHz, [D8]THF): δ = 65.9
(s) ppm. C54H72N4P2 (839.14): calcd. C 77.29, H 8.65, N 6.68;
found C 76.93, H 8.82, N 6.72.
3
2.2 Hz, 2 H, NCH), 3.99 [sept, JH,H = 6.7 Hz, 2 H, CH(CH3)2],
3.98 [sept, 3JH,H = 6.7 Hz, 2 H, CH(CH3)2], 1.38 [d, 3JH,H = 6.8 Hz,
3
6 H, CH(CH3)2], 1.29 [d, JH,H = 6.8 Hz, 6 H, CH(CH3)2], 1.28 [d,
3JH,H = 6.8 Hz, 6 H, CH(CH3)2], 1.24 [d, JH,H = 6.8 Hz, 6 H,
3
CH(CH3)2] ppm. 13C{1H} NMR (62.9 MHz, C6D6): δ = 137.0 (d,
2JP,C = 13.7 Hz, i-C), 132.2 (d, 4JP,C = 2.0 Hz, m-C), 128.7 (s, p-C),
2-Azido-1,3-bis(2,6-diisopropylphenyl)-1,3,2-diazaphospholene (6):
NaN3 (650 mg, 10 mmol) and LiCl (21 mg, 50 μmol) were added
to a solution of 2b (4.42 g, 10 mmol) in THF (40 mL). The mixture
was stirred for 24 h at room temperature before all solvents were
evaporated under vacuum. The residue was dissolved in hexane
(20 mL), and solids were removed by filtration. The filtrate was
concentrated to a volume of 10 mL. Storage at –21 °C gave colour-
3
128.2 (s, o-C), 124.7 (d, JP,C = 2.0 Hz, i-CPh) 124.2 (s, m-CPh),
123.0 (s, p-CPh), 122.9 (s, o-CPh), 120.0 (d, JP,C = 6.8 Hz, CHN),
2
2
1
102.2 (d, JP,C = 14.2 Hz, PCCPh), 94.1 (d, JP, C = 111 Hz,
PCCPh), 28.7 [s, CH(CH3)2], 25.4 (s, CH3), 25.3 (s, CH3), 24.9 (s,
CH3), 23.7 (s, CH3) ppm. 31P{1H} NMR (101.2 MHz, C6D6): δ =
65.3 (s) ppm. IR: ν = 2131 (w) ν(CϵC) cm–1. C H N P·0.5C H
˜
34 41
2
6
14
1
less crystals, m.p. 142 °C, yield 3.82 g (85%). H NMR (250 MHz,
(551.78): calcd. C 80.28, H 8.12, N 5.51; found C 79.19, H 8.27, N
5.12.
3
CDCl3): δ = 7.60–7.30 (m, 6 H, C6H3), 6.26 (d, JP,H = 1.5 Hz, 2
H, NCH), 3.85 [br., 2 H, CH(CH3)2], 3.46 [br., 2 H, CH(CH3)2],
2-Ethynyl-1,3-bis(2,6-diisopropylphenyl)-1,3,2-diazaphospholene
(3c): A solution of ethynylmagnesium chloride in THF (10 mmol,
20 mL of 0.5 m solution) was added to a solution of 2b (4.42 g,
10 mmol) in THF (40 mL) at –78 °C. The mixture was stirred for
1 h and then warmed to ambient temperature. The solvent was re-
moved under vacuum and the residue was dissolved in toluene
(20 mL). The formed suspension was filtered, and the filtrate was
concentrated under reduced pressure to a volume of 10 mL. Stor-
3
1.54 [d, JH,H = 6.8 Hz, 12 H, CH(CH3)2], 1.44 [br., 12 H,
CH(CH3)2] ppm. 13C{1H} NMR (62.9 MHz, CDCl3): δ = 148.6
2
(br., o-C), 135.1 (d, JP,C = 13 Hz, i-C), 128.7 (d, 2JP,C = 2.0 Hz, p-
2
C), 124.4 (br., m-C), 119.5 (d, JP,C = 7.6 Hz, =CH), 28.8 [s,
CH(CH3)2] 25.1 (s, CH3), 24.6 (br., CH3) ppm. 31P{1H} NMR
(101.2 MHz, CDCl ): δ = 120.8 (s) ppm. IR: ν = 2069 (s) ν (NN)
˜
3
as
cm–1. C26H36N5P (449.58): calcd. C 69.46, H 8.07, N 15.58; found
C 68.44, H 7.92, N 14.88.
age at –21 °C produced colourless crystals, yield 2.81 g (65%), m.p.
1
95 °C. H NMR (CDCl3): δ = 7.40–7.20 (m, 6 H, C6H3), 6.04 (d, General Procedure for the Studies of Reactions Between Alkynyl-
3JP,H = 1.3 Hz, 2 H, NCH), 3.85 [br., 2 H, CH(CH3)2], 3.46 [br., 2 phosphanes and CuCl: In a typical reaction, the alkynylphosphane
Eur. J. Inorg. Chem. 2014, 1818–1825
1823
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
(1 mmol) was dissolved in THF (10 mL), or THF/toluene (1:1,
[IϾ2σ(I)] = 0.040, wR2 = 0.095, largest diff. peak and hole: 0.392
10 mL), and the appropriate reagents (CuCl, NaN3, Et3N) were and –0.376 eÅ–3.
added as needed. For monitoring by 1H NMR spectroscopy, ap-
Compound 6: Colourless plates, C26H36N5P, Mr = 449.57 gmol–1,
proximately 100 μmol of sample were placed in an NMR spec-
troscopy tube, and solvent ([D8]THF) and needed reagents were
added. The progress of the reactions was monitored in regular in-
tervals (1 d or less if necessary) by recording 31P NMR spectra. If
needed, the mixture was heated to reflux. Reaction monitoring was
continued in regular intervals by allowing the mixture to cool to
room temperature and recording an 31P NMR spectrum. After con-
trolling the reaction progress by means of 31P NMR spectroscopy,
the mixture was heated at 66 °C for 3 d. The advance of the reac-
tion was again controlled by means of 31P NMR spectroscopy.
Then the solvent was removed under vacuum, the residues were
dissolved in toluene and the solution was stirred for 3 d at 110 °C.
Once again a 31P NMR spectrum was taken to control the reaction
proceedings.
crystal size 0.22ϫ0.17ϫ0.06 mm, monoclinic, space group P21/n,
a = 9.9251(17) Å, b = 17.254(3) Å, c = 15.472(3) Å, β = 104.302
(6)°, V = 2567.3(8) Å3, Z = 4, ρcalcd. = 1.163 mgm3, F(000) = 968,
θmax = 25.1°, μ = 0.129 mm–1, 16899 reflections measured, 4524
unique reflections [Rint = 0.058] for structure solution and refine-
ment with 289 parameters, R1 [IϾ2σ(I)] = 0.049, wR2 = 0.108,
largest diff. peak and hole: 0.420 and –0.296 eÅ–3.
CCDC-957800 (for 3a), -957767 (for 3b), -957768 (for 3c), -957769
(for 4) and -957766 (for 6) contain the supplementary crystallo-
graphic 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.
Supporting Information (see footnote on the first page of this arti-
cle): Graphical representation of the molecular structures of 3a,b
and 6; 31P{1H} NMR spectrum of the reactions of 3c with NaN3/
CuCl/LiCl to give 4 and 6.
Crystal Structure Determinations: Diffraction studies on 3a–c, 4
and 6 were carried out with a Bruker Kappa APEXII Duo dif-
fractometer at T = 100(2) K [T = 110(2) K for 4] or a Nonius
Kappa CCD diffractometer at T = 123(2) K using Mo-Kα radiation
(λ = 0.71073 Å). The structures were solved by using direct meth-
ods (SHELXS-97[21]) and refined with a full-matrix least-squares
scheme on F2 (SHELXL-97[21]). Semiempirical absorption correc-
tions were applied for 4 and 6, and numerical absorption correc-
tions for 3b,c. Non-hydrogen atoms were refined anisotropically,
and hydrogen atoms in CH bonds with a riding model on F2. The
position of the hydrogen atom of the terminal alkynyl group of 3c
was located in a difference density map and refined freely using
isotropic thermal displacement parameters.
Acknowledgments
Financial support by Deutscher Akademischer Austauschdienst
(DAAD) (to D. F., J. B.) 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 meas-
uring X-ray diffraction data sets, and J. Trinkner and K. Wohlbold
(Institut für Organische Chemie, University of Stuttgart, Germany)
for recording the mass spectra.
Compound 3a: Orange blocks, C26H25N2P, Mr = 396.45 gmol–1,
crystal size 0.60ϫ0.45ϫ0.30 mm, monoclinic, space group P21/c,
a = 20.0907(6) Å, b = 14.5135(4) Å, c = 7.4162(2) Å, β = 99.841
(2)°, V = 2130.64(10) Å3, Z = 4, ρcalcd. = 1.236 Mgm3, F(000) =
840, θmax = 27.5°, μ = 0.14 mm–1, 12849 reflections measured, 4748
unique reflections [Rint = 0.030] for structure solution and refine-
ment with 266 parameters, R1 [IϾ2σ(I)] = 0.034, wR2 = 0.096,
largest diff. peak and hole: 0.235 and –0.326 eÅ–3.
[1] a) The Chemistry of Organophosphorus Compounds (Ed.: F. R.
Hartley), Wiley, Chichester, 1990–1996, vol. 1–4; b) L. D. Quin,
in: A Guide to Organophosphorus Chemistry, Wiley, New York,
2000; c) M. L. Clarke, J. M. J. Williams, in: Organophosphorus
Reagents (Ed.: P. J. Murphy), Oxford University Press, Oxford,
UK, 2004, p. 15–50; d) Houben-Weyl, Methoden der Or-
ganischen Chemie, part 2 (Ed.: E. Müller), Thieme, Stuttgart,
Germany, 1977, vol. 5.
Compound 3b: Orange blocks, C37H48N2P, Mr = 551.74 gmol–1,
crystal size 0.50ϫ0.33ϫ0.29 mm, orthorhombic, space group
Pna21, a = 17.3020(11) Å, b = 15.3922(12) Å, c = 25.1447(16) Å, V
= 6696.4(8) Å3, Z = 8, ρcalcd. = 1.095 mgm3, F(000) = 2392, θmax
= 27.5°, μ = 0.108 mm–1, 84440 reflections measured, 15313 unique
reflections [Rint = 0.031] for structure solution and refinement with
711 parameters (1 restraint), R1 [IϾ2σ(I)] = 0.038, wR2 = 0.093,
largest diff. peak and hole: 0.349 and –0.237 eÅ–3.
[2] a) W. Wolfsberger, H. Schmidbaur, Synth. React. Inorg. Met.
Org. Chem. 1974, 4, 149–156; b) J. Keller, C. Schlierf, C. Nolte,
P. Mayer, B. F. Straub, Synthesis 2003, 354–365.
[3] S. Singh, K. N. Nicholas, Chem. Commun. 1998, 149–150.
[4] P. E. Garrou, Chem. Rev. 1985, 85, 171–185.
[5] W. Wolfsberger, Chem. Zeitg. 1991, 115, 7–13.
[6] K. A. Petrov, S. V. Agafonov, V. P. Pokatun, V. M. Chizov, Zh.
Obshch. Khim. 1987, 57, 299–302.
[7] M. Bode, G. Schnakenburg, P. G. Jones, R. Streubel, Organo-
metallics 2008, 27, 2664–2667.
[8] a) R. Pietschnig, J. Ebels, M. Nieger, N. Zoche, M. Jansen, E.
Niecke, Bull. Soc. Chim. Fr. 1997, 134, 1039–1043; b) S. Burck,
D. Gudat, M. Nieger, J. Tirrée, Dalton Trans. 2007, 1891–1897.
[9] Review: D. Gudat, Acc. Chem. Res. 2010, 43, 1307–1316.
[10] For an example of (irreversible) P–Csp bond activation in a
metal cluster, see: F. van Gastel, S. A. Mac Laughlin, M.
Lynch, A. J. Carty, E. Sappa, A. Tiripicchio, M. Tiripicchio-
Camellini, J. Organomet. Chem. 1987, 326, C65–C70.
[11] R. Gleiter, D. B. Werz, Chem. Rev. 2010, 110, 4447–4488.
[12] a) F. Dolhem, M. J. Johansson, T. Antonsson, N. Kann, J.
Comb. Chem. 2007, 9, 477–486; b) M. R. J. Vallée, L. M.
Artner, J. Dernedde, C. P. R. Hackenberger, Angew. Chem.
2013, 125, 9682–9686; Angew. Chem. Int. Ed. 2013, 52, 9504–
9508.
Compound 3c: Yellow blocks, C28H37N2P, Mr = 432.57 gmol–1,
crystal size 0.56ϫ0.30ϫ0.21 mm, orthorhombic, space group
P212121, a = 8.9922(6) Å, b = 10.0936(7) Å, c = 28.5603(19) Å, V
= 2592.2(3) Å3, Z = 4, ρcalcd. = 1.108 mgm3, F(000) = 936, θmax
=
25.1°, μ = 0.123 mm–1, 38799 reflections measured, 4575 unique
reflections [Rint = 0.059] for structure solution and refinement with
283 parameters, R1 [IϾ2σ(I)] = 0.039, wR2 = 0.091, largest diff.
peak and hole: 0.299 and –0.197 eÅ–3.
Compound 4: Yellow prisms, C54H72N4P2, Mr = 839.10 gmol–1,
¯
crystal size 0.84ϫ0.66ϫ0.34 mm, triclinic, space group P1, a =
13.3399(8) Å, b = 14.3748(9) Å, c = 14.9128(8) Å, α = 67.642 (3)°,
β = 89.73 (3)°, γ = 71.643 (3)°, V = 2488.4(3) Å3, Z = 2, ρcalcd.
=
1.120 mgm3, F(000) = 908, θmax = 27.5°, μ = 0.126 mm–1, 47584
reflections measured, 11376 unique reflections [Rint = 0.03] for
structure solution and refinement with 541 parameters, R1
[13] Review: M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108, 2952–
3015.
[14] D. Gudat, Top. Heterocycl. Chem. 2010, 21, 63–102.
Eur. J. Inorg. Chem. 2014, 1818–1825
1824
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[15] Results returned by a query in the CSD data base (CSD version
5.34 including 3rd update from May 2013) for alkynylphos-
phanes with the structural fragment R2P–CϵC– in which R
denotes any non-metal-based substituent.
[16] V. V. Afanasiev, I. P. Beletskaya, M. A. Kazankova, I. V. Efi-
mova, M. U. Antipin, Synthesis 2003, 2835–2838.
[17] R. Nast, W. Pfab, Z. Anorg. Allg. Chem. 1957, 292, 287–292.
[18] In the reaction of 3c, also one equivalent of Et3N was added
in order to avoid any complications arising from possible HCl
formation.
[20]
a) R.-M. Lequan, M.-J. Pouet, M.-P. Simonnin, Org. Magn.
Reson. 1975, 7, 392–400; b) C. Charrier, W. Chodkiewicz, P.
Cadiot, Bull. Soc. Chim. Fr. 1966, 3, 1002–1011; c) R. B. King,
P. M. Sundaram, J. Org. Chem. 1984, 49, 1784–1789; d) A.
Hengefeld, R. Nast, Chem. Ber. 1983, 116, 2035–2036; e)
M. S. M. Timmer, H. Ovaa, D. V. Filippov, G. A. van der Ma-
rel, J. H. van Boom, Tetrahedron Lett. 2001, 42, 8231–8233.
[21] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
Received: September 4, 2013
Published Online: November 19, 2013
[19] S. Burck, D. Gudat, M. Nieger, W.-W. Du Mont, J. Am. Chem.
Soc. 2006, 128, 3946–3955.
Eur. J. Inorg. Chem. 2014, 1818–1825
1825
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim