Donor-functionalized polydentate pyrylium salts and phosphinines: Synthesis, structural characterization, and photophysical properties

Article abstract of DOI:10.1002/chem.200601650

A series of donor-functionalized pyrylium salts have been prepared by classical condensation reactions which were further converted into the corresponding thienyl- and pyridyl-substituted polydentate λ³- phosphinines by reaction with P(SiMe

Full text of DOI:10.1002/chem.200601650

DOI: 10.1002/chem.200601650
Donor-Functionalized Polydentate Pyrylium Salts and Phosphinines:
Synthesis, Structural Characterization, and Photophysical Properties
Christian Müller,*[a] Dorothee Wasserberg,[b] Jarno J. M. Weemers,[a] Evgeny A. Pidko,[a]
Sandra Hoffmann,[a] Martin Lutz,[c] Anthony L. Spek,[c] Stefan C. J. Meskers,[b]
RenØ A. J. Janssen,[b] Rutger A. van Santen,[a] and Dieter Vogt[a]
Dedicated to Professor Piet W.N.M. van Leeuwen on the occasion of his 65th birthday
Abstract: Aseries of donor-functional-
ized pyrylium salts have been prepared
by classical condensation reactions
which were further converted into the
corresponding thienyl- and pyridyl-sub-
stituted polydentate l3-phosphinines by
series of thienyl- and pyridyl-function-
alized pyrylium salts, l3- and l5-phos-
phinines, were investigated and the re-
sults compared and supported by theo-
retical calculations on the DFT level.
Significant fluorescence was observed
for the pyrylium salts and l5-phosphi-
nines. In contrast, the heteroaromatic
substituted l3-phosphinines show very
little emission which is consistent with
the low oscillator strength predicted by
DFT calculations for this p!p* transi-
tion. Furthermore, all three classes of
compounds show readily observable
phosphorescence in solution, which
was determined by time-gated detec-
tion at low temperature.
reaction with P
ical modification of these phosphorus
heterocycles with Hg(OAc)2 in the
ACHTREUNG(SiMe3)3. Further chem-
AHCTREUNG
Keywords: density functional calcu-
lations · fluorescence · heterocycles ·
phosphorescence · pyrylium salt
presence of methanol resulted in the
formation of l5-phosphinines. The pho-
tophysical properties of
a selected
Introduction
-CH- group of the aryl moiety is substituted by a trivalent
phosphorus atom (Figure 1, structure I), the bonding situa-
tion in l5-phosphinines can best be described as a superim-
position of an ambivalent electronic structure: Anonclassi-
cal 6p delocalized Hückel aromatic system and a cyclic
phosphonium ylide structure, having a four-coordinated
phosphonium center and a negatively charged five-carbon
subunit (Figure 1, structures IIa/b).[3]
l3- and l5-phosphinines (phosphorines, phosphabenzenes)
are known for many decades, due to the pioneering work of
Märkl, Dimroth and Ashe III in the late 1960s.[1,2] While l3-
phosphinines are planar, aromatic heterocycles in which one
[a] Dr. C. Müller, J. J. M. Weemers, E. A. Pidko, S. Hoffmann,
Prof. Dr. R. A. van Santen, Prof. Dr. D. Vogt
Department of Chemical Engineering and
Chemistry, Schuit Institute of Catalysis
Eindhoven University of Technology
5600 MB Eindhoven (The Netherlands)
Fax : (+31)40-245-5054
Figure 1. l3- and l5-phosphinines.
E-mail: c.mueller@tue.nl
[b] Dr. D. Wasserberg, Dr. S. C. J. Meskers, Prof. Dr. R. A. J. Janssen
Department of Chemical Engineering and
So far the organometallic chemistry of monodentate l3-
phosphinines as ligands for transition metals has been devel-
oped considerably, but the introduction of additional donor
functionalities within the phosphinine moiety has been
much less regarded and reports on these systems are rare.
As a matter of fact, additional donor functionalities located
Chemistry, Molecular Materials and Nanosystems
Eindhoven University of Technology
5600 MB Eindhoven (The Netherlands)
[c] Dr. M. Lutz, Prof. Dr. A. L. Spek
Bijvoet Center for Biomolecular Research,
Crystal and Structural Chemistry, Utrecht University
3584CH Utrecht (The Netherlands)
4548
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4548 – 4559

FULL PAPER
at specific positions within the phosphinine framework
might provide interesting properties, such as an enhanced
stability of the corresponding metal complexes due to a che-
late effect, or the generation of secondary interactions
through the additional donor group, with relevance for ho-
mogeneous catalytic reactions. Mathey, Le Floch and co-
workers reported on thienyl-, as well as pyridyl-substituted
phosphinines but their access is limited to certain substitu-
tion patterns due to the complex synthetic procedure.[4–6]
During the course of our investigations on functionalized
phosphinines as ligands for homogeneous catalytic reac-
tions,[7] we started to explore the modularity of the original
phosphinine synthesis via pyrylium salts, which interestingly
offers an easy synthetic access to a whole variety of symmet-
rically as well as non-symmetrically donor-functionalized
phosphinines. We report here on the preparation of thienyl-
and pyridyl-functionalized pyrylium salts which were further
converted into the corresponding l3-phosphinines by reac-
respectively.[10] However, this modular approach has never
been used for a rational design and synthesis of donor-func-
tionalized phosphinines so far. In this respect, we further ex-
plored the incorporation of additional donor groups into the
pyrylium framework and prepared the strongly fluorescent
pyrylium tetrafluoroborates 4–7, containing thienyl and pyr-
idyl functionalities according to Scheme 1.[11]
tion with PACHTREUNG(SiMe3)3 and fully characterized by means of
NMR spectroscopy, elemental analysis and, partially, crystal-
lographically. Because 2,4,6-aryl-substituted l5-phosphinines
are known to show fluorescence in many cases,[2a] we further
extended our investigations of phosphabenzenes with the
preparation of functionalized l5-phosphinines substituted
with additional heteroaromatic systems, which, to the best
of our knowledge, have not been reported in the literature
before. As a matter of fact, the incorporation of phosphorus
centers in polythiophene or polypyridine chains has recently
attracted a lot of interest in the field of p-conjugated materi-
als for molecular electronics and optoelectronic applica-
tions.[8] In this context, two novel l5-phosphinines, substitut-
ed with additional thienyl and pyridyl groups have been pre-
pared from the corresponding l3-phosphinines. The photo-
physical properties of these systems have been studied and
compared with the corresponding l3-phosphinines and pyry-
lium salts. The results have been further compared and sup-
ported by means of DFT calculations.
Scheme 1. Synthetic routes for the preparation of the donor-functional-
ized pyrylium salts 4–7.
The thienyl-substituted chalcone benzal-a-acetothienone
(1) can be prepared in high yields according to the proce-
dure described by Weygand and Strobelt.[12] Reaction of two
equivalents of 1 with p-methyl acetophenone in the presence
of HBF4·Et2O in dichloroethane at 708C gives the thienyl
substituted pyrylium salt 4 as an orange solid in 37% yield
(Scheme 1). Under similar reaction conditions the symmetri-
cally substituted bisACHTRE(UNG thienyl)pyrylium salt 5 was obtained as
a dark red solid in 21% yield from compound 1 and 2-ace-
tylthiophene.
Even though pyrylium salt 5 as well as the analogous
ꢀ
ClO4 salt have been mentioned in the literature some deca-
Results and Discussion
des ago, neither experimental procedures for their synthesis
nor structural information has been described.[9a,11,13] Fur-
thermore, the chalcone route (route 1, Scheme 1) provides a
Pyrylium salts and l3-phosphinines: Triaryl-substituted pyry-
lium salts can generally be synthesized either via the chal-
ꢀ
facile access to the BF4 salt and thus prevents preparation
cone route (a,b-unsaturated aryl ketones, route
Scheme 1) or alternatively by cyclization of 1,5-diketones in
the presence of HBF4·Et2O or BF3·Et2O (route in
1
in
and handling of potentially explosive organic compounds.
Crystals suitable for X-ray diffraction were obtained by
slow recrystallization of 5 from a fluorescent solution in
methanol. Compound 5 crystallizes in the space group P21/c
with one independent molecule in the asymmetric unit. The
molecular structure is depicted in Figure 2a along with se-
lected bond lengths and angles.[14]
2
Scheme 1).[9] Nevertheless, both synthetic protocols usually
have their limitations and the right reaction conditions have
to be selected individually. Interestingly, both methods are
not only restricted to the incorporation of phenyl substitu-
ents into the 2,6 position of the heterocycle (R1 =R2 =Ph,
Scheme 1) but can also be applied for the introduction of
substituted phenyl groups or even heteroaromatic substitu-
ents, such as pyridyl or thienyl groups. As a matter of fact
this procedure has been investigated before to some extend
for the synthesis of pyridines or pyridinium salts by reaction
of substituted pyrylium salts with NH3 or primary amines,
All aromatic rings are essentially coplanar to one another.
ꢀ
Moreover, the bond lengths C5 C16 (1.4357(18) Š) and
ꢀ
C1 C12 (1.4372(18) Š) are considerably shorter than a
ꢀ
carbon carbon single bond, indicating delocalization of the
p systems over the heteroaromatic rings. In the solid state,
the molecules are arranged in a herringbone-like structure
as depicted in the representation of the unit-cell (Figure 2b).
Chem. Eur. J. 2007, 13, 4548 – 4559
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4549

C. Müller et al.
with benzylidene-2-acetophenone in the presence of
BF3·Et2O affords a dark red solid, which after recrystalliza-
tion from methanol, afforded the monotetrafluoroborate
pyrylium salt 7 as a yellow solid containing an unprotonated
pyridyl functionality in 25% yield.[16]
The pyrylium salts 4–7 were further converted into the
donor-functionalized phosphinines 8–11 by reaction with
excess PACTHRE(UNG SiMe3)3 in CH3CN.
Figure 2. a) Molecular structure of 5 in the solid state (ORTEP view,
50% probability level). The thienyl ring at sulfur atom S1 is rotationally
disordered (89:11); only the major isomer is shown. Selected bond
ꢀ
ꢀ
ꢀ
lengths [Š]: O1 C1 1.3583(15), C1 C2 1.3725(18), C2 C3 1.4100(18),
ꢀ
ꢀ
ꢀ
ꢀ
C3 C4 1.4068(18), C4 C5 1.3769(18), O1 C5 1.3554(16), C3 C6
[i]
ꢀ
ꢀ
ꢀ
1.4799(18), C1 C12 1.4372(18), C5 C16 1.4357(18), O1 B1 6.015(2) [i:
1+x, 0.5ꢀy, 0.5ꢀz]. b) Representation of the unit cell of 5 (hydrogen
atoms are omitted for clarity).
The polydentate systems were obtained as yellow, fairly
air- and moisture-stable solids in moderate yields (20–46%)
after column chromatography and were fully characterized
1
The pyridyl-functionalized pyrylium salt 6 cannot be ob-
tained directly by reaction of the chalcone benzylidene-2-
acetophenone with 2-acetylpyridine and HBF4·Et2O, due to
substantial retro-aldol condensation under these reaction
conditions, leading to a product mixture.[10a] However, ac-
cording to a modified experimental procedure described by
Katritzky et al., compound 6 was prepared starting from di-
ketone 3,5-diphenyl-1-(2-pyridyl)pentane-1,5-dione 2 (route
2, Scheme 1).[10b,15] Subsequent oxidation with one equiva-
lent benzylidene-2-acetophenone in the presence of
BF3·Et2O affords the pyrylium salt 6a as a yellow solid in
60% yield, which is most likely the bistetrafluoroborate salt
containing a protonated pyridyl group.[10b] Slow recrystalliza-
tion of 6a from acetonitrile/methanol afforded bright yellow
crystals. Unfortunately, no satisfying X-ray crystal structure
determination of this compound could be performed, never-
theless it is certain from the X-ray data and the elemental
analysis that it crystallizes as the monocationic species 6 in
which the pyridyl functionality remains unprotonated.
Interestingly, we found that the diketone route (route 2,
Scheme 1) is not restricted to the preparation of symmetri-
cally substituted systems, but can also be used for the syn-
thesis of mixed-donor functionalized pyrylium salts, with
two different donor substituents attached to the heterocycle.
Thus, reaction of the thienyl-functionalized chalcone benzal-
a-acetothienone (1) with 2-acetyl-pyridine gave the corre-
sponding diketone 3 in high yields. Subsequent oxidation
by means of H, 13C and 31P NMR spectroscopy. Analytically
pure samples of all phosphinines were obtained by recrystal-
lization from acetonitrile to afford yellow needles. In the
31P NMR spectrum, all four compounds show the character-
istic downfield shift of the phosphorus signal of d
=
ꢁ180 ppm. Figure 3 shows exemplarily the H and 31P NMR
1
spectra of the thienyl-functionalized phosphinine 8.
1
In the H NMR spectrum of 8 the two doublets at d=7.98
and 8.16 ppm with the coupling constants of JH,P =6.0 and
5.6 Hz are distinctive for the P-heterocyclic protons Ha/b of
an asymmetrically substituted phosphinine and were identi-
fied by H-coupled 31P NMR spectroscopy. Similar observa-
1
tions were made for phosphinines 10 and 11. On the other
hand, the H-coupled 31P NMR spectrum of the symmetri-
1
cally substituted bisthienyl-substituted phosphinine 9 (C6D6,
81 MHz) shows a single triplet at d=174.0 ppm with a cou-
pling constant of JP, H =5.8 Hz and a single doublet at d=
8.03 ppm with a coupling constant of JH,P =5.8 Hz (C6D6,
200 MHz) for the P-heterocyclic protons in the 1H NMR
spectrum. Crystals of 9 suitable for X-ray diffraction were
obtained by slow recrystallization from acetonitrile. Com-
pound 9 crystallizes in the space group C2/c with one inde-
pendent molecule in the asymmetric unit and the molecular
structure along with selected bond lengths and angles is de-
picted in Figure 4a.[17] Interestingly, the three heterocycles
are essentially coplanar to one another, which was also ob-
served in the corresponding pyrylium salt 5 (see above). In
4550
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4548 – 4559

Polydentate Pyrylium Salts
FULL PAPER
Figure 3. 1H (400 MHz, C6D6) and 31P NMR (162 MHz, C6D6) spectrum of the thienyl-functionalized phosphinine 8.
contrast, the compound 2,6-bis(2-thienyl)-3-methyl-4-bromo-
phosphinine, reported by Mathey et al., shows only one
thiophene moiety strictly coplanar with the phosphorus ring,
whereas the other one is twisted from the phosphinine
plane. The authors describe this phenomenon as a result of
steric repulsion of the five-membered sulfur heterocycle
with the adjacent methyl group.[4] We further observed, that
ꢀ
ꢀ
both bond lengths C1 C12 (1.470(3) Š) and C5 C16
ꢀ
(1.467(3) Š) in 9 are somewhat shorter than a carbon
carbon single bond, indicating at least some degree of deloc-
alization of the aromatic systems within the three heterocy-
cles. As observed before for the pyrylium salt 5 the mole-
cules are packed in a herringbone structure as illustrated in
Figure 4b.
In case of the pyridyl-substituted phosphinine 10 and the
mixed donor-functionalized phosphinine 11 bright yellow
crystals were obtained by slow recrystallization from aceto-
nitrile. Compound 10 crystallizes in the space group Pna21
with one molecule in the asymmetric unit, which are ar-
ranged in a layer-type structure.
Unfortunately, the X-ray crystal structure analysis re-
vealed a high degree of disorder within the molecule and
therefore bond lengths and angles cannot be discussed ap-
propriately. Nevertheless, the result is illustrated in Fig-
ure 5a, and a representation of the unit cell is depicted in
Figure 5b. The crystals obtained from compound 11 were
not suitable for X-ray diffraction.[18]
Alkyl-substituted bisthienyl- and pyridyl-substituted phos-
phinines were reported earlier by Mathey and Le Floch
et al., who obtained the compounds via Pd-catalyzed Stille
cross-coupling reaction of bromophosphinines with heteroar-
yl–trimethyltin derivatives, or by a multistep synthesis start-
ing from five-membered phospholes. Subsequent ring-ex-
pansion with picolinoyl chloride followed by a Ni-catalyzed
reduction gave the corresponding pyridyl-functionalized
phosphinine as a phosphorus-containing derivative of bipyri-
Figure 4. a) Molecular structure of 9 in the solid state (ORTEP view,
50% probability level). Both thienyl rings are rotationally disordered.
ꢀ
Only one isomer is shown. Selected bond lengths [Š] and angles [8]: P1
ꢀ
ꢀ
ꢀ
ꢀ
C1 1.749(2), C1 C2 1.398(3), C2 C3 1.394(3), C3 C4 1.394(3), C4 C5
ꢀ
ꢀ
ꢀ
ꢀ
1.392(3), P1 C5 1.749(2), C1 C12 1.470(3), C5 C16 1.467(3), C3 C6
1.487(3), C1-P1-C5 101.37(11), C2-C3-C4: 121.5(2). b) Representation of
the unit cell of 9 (hydrogen atoms are omitted for clarity, view along the
crystallographic a,c diagonal).
dine.[4,5] As we could demonstrate here, the described modu-
lar preparation of 8–11 via the pyrylium salt route provides
an interesting, more convenient and less time-consuming al-
Chem. Eur. J. 2007, 13, 4548 – 4559
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4551

C. Müller et al.
resonance at d=59.8 and 62.9 ppm (C6D6), respectively. This
high-field shift of approximately 130 ppm compared with
the values of the corresponding l3-phosphinines is character-
istic for l5-phosphinines.[2f,g] The 1H NMR spectrum of 12
shows a doublet for the P-heterocyclic protons at d=
3
8.12 ppm with a coupling constant of JH,P =36.2 Hz, while
4
two doublet of doublets (d=7.95, 3JH,P =36.0 Hz, JH,H
=
3
4
2.8 Hz; d=8.34, JH,P =38.0 Hz, JH,H =2.6 Hz) are observed
for the asymmetrically substituted phosphinine 13. The ob-
served downfield shifts and large coupling constants com-
pared to l3-phosphinines are characteristic for l5-phosphi-
nines.
Photophysical properties: Because compounds 12 and 13 as
well as the corresponding pyrylium salts 5 and 7 show signif-
icant yellow-green fluorescence in solution, we further re-
corded the electronic absorption and fluorescence spectra of
the 2,6-dithienyl substituted series, 5, 9, 12 (SXS) and the 2-
pyridyl-6-thienyl substituted series 7, 11, 13 (SXN) at room
temperature. The low temperature absorption and phos-
phorescence spectra were obtained for those two series at
80 K in 2-methyltetrahydrofuran (T=176 K in dichlorome-
thane in the case of the pyrylium salts for solubility rea-
sons). Fluorescence lifetimes and fluorescence quantum
yields have further been determined at room temperature,
while the phosphorescence lifetimes were estimated at low
temperatures.
The electronic absorption and emission spectra reveal
broad, featureless bands (Figures 6 and 7). Almost no
changes were observed upon cooling the samples to low
temperature, apart from a slight sharpening of the bands,
which was accompanied by a slight shift of the peaks. The
lowest energy (red-edge) absorption maxima (and shoulders
where discernible), emission maxima as well as all men-
tioned photophysical data are reported in Table 1.
Figure 5. a) Molecular structure of 10 in the solid state (ORTEP view,
50% probability level). All N and P ring systems are disordered. b) Rep-
resentation of the unit cell (hydrogen atoms are omitted for clarity, view
along the crystallographic c axis).
ternative for the synthesis of donor-functionalized phosphi-
nines.
l5 Phosphinines: The l3-phosphinines 9 and 11 were further
converted quantitatively into the corresponding l5-phosphi-
nines by reaction with HgACHTREU(NG OAc)2 and methanol, according
to the procedure described by Dimroth et al. (Scheme 2).[1d]
Within a given series (SXS, SXN) a red-shift can be ob-
served in the electronic spectra by going from l3- to l5-phos-
phinines and subsequently to the pyrylium salts (Figures 6,
7, Table 1). These red-shifts within one series are more pro-
nounced compared with compounds of the same type but of
different series, that is, different substitution patterns. Gen-
erally, l5-phosphinines and pyrylium salts exhibit many simi-
larities in their photophysical properties while the l3-phos-
phinines are rather different (Figures 6, 7, Table 1). The
fluorescence of l5-phosphinines and pyrylium salts is signifi-
cant (quantum yields 5–99%, Table 1) and only slightly red-
shifted (by ꢁ50 nm) compared to the maximum of the
lowest energy (red-edge) absorption band, from which it
seems to arise. Interestingly, both l3-phosphinines exhibit a
weak shoulder at the red-edge of the absorption spectrum.
Furthermore, the two bands in the corresponding fluores-
cence spectra of the l3-phosphinines are of very low intensi-
ty (quantum yields ! 1%, Figure 7, Table 1). They seem to
arise from the state that causes the shoulder at the red-edge
of the absorption spectrum, as it is weak and strongly red-
shifted (> 100 nm) with respect to the maximum of the
lowest energy absorption band (Table 1, Figures 6, 7). How-
Scheme 2. Synthesis of l5-phosphinines 12 and 13.
These compounds were obtained as fairly air- and mois-
ture-stable orange and yellow solids in 80–84% isolated
yield after column chromatography and were characterized
by means of 1H, 13C and 31P NMR spectroscopy as well as el-
emental analysis.[19] It should be mentioned here that ther-
mal 1,1-elimination towards l3-phosphinines has been ob-
served for certain l5-phosphinines and can occur especially
at higher temperature, depending on the substituents.[2f,g] In
the 31P NMR spectrum compounds 12 and 13 show a single
4552
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4548 – 4559

Polydentate Pyrylium Salts
FULL PAPER
!
Table 1. Wavelengths [nm] of the transitions: S1 S0 at room temperature, low temperature and calculated
values, S1!S0 at room temperature and low temperature and T1!S0 at low temperature and calculated values
as well as fluorescence (ns) and phosphorescence [ms] lifetimes [t] and fluorescence quantum yields [%].
The electronic excitation en-
ergies as well as the corre-
sponding oscillator strengths
could be obtained from time-
dependent density functional
theory (TD-DFT) calculations.
Both the theoretical and experi-
mental absorption spectra for
the SXS series are shown in
Figure 6. One can see that the
theoretical UV spectra repro-
duce the major features and
general shape of the experi-
mental spectra, while some dis-
agreement persists for their en-
ergetic positions (Table 1).
However, one should realize
that these data correspond to
electronic excitations within an
isolated molecule, whereas the
experimental results have been
obtained in the presence of a
solvent. Nevertheless, the calcu-
Compound
Pyrylium salt
l3-Phosphinine
l5-Phosphinine
12 13
5
7
9
11
!
l
l
l
l
l
l
l
C
492
499
445
539
535
629
630
4.84
@1
20
457
464
421
507
518
624
595
5.27
>1
99
310/391[g]
312/398[g]
314/376[g]
448/557[h]
450/542[h]
601
306/382[g]
315/381[g]
312/370[g]
448/570[h]
440/547[h]
603
602
–
>1
!1
446 445
!
457
394
503
492
638
602
3.59
ꢁ1
20
455
401
500
496
577
606
5.36
ꢁ1
5
!
(S1 S0)[a] calculations [nm]
(S1!S0)[c] room temp. [nm]
(T1!S0)[d] calculations [nm]
607
–
tFl [ns][e]
tPh [ms][f]
ꢁ1
quantum yield [%]
!1
[a] Wavelengths determined at the maximum of the lowest energy absorption band as no fine-structure was ap-
parent. [b] 176 K for pyrylium salts (dichloromethane) and 80 K for phosphinines (2-methyltetrahydrofuran).
[c] Wavelengths determined at the maximum of the emission band as no fine-structure was apparent. [d] De-
termined at the highest energy maximum, where discernible. [e] Measured on degassed solutions (room tem-
perature) at 400 nm excitation wavelength, which precluded the excitation of the l3-phosphinines. [f] Deter-
mined using varying delay times during phosphorescence measurements. [g] First value corresponds to the
maximum of the high intensity transition at the low-energy edge (red-edge) and the second value to that of
the low intensity transition at the red-edge, where discernible. [h] The two values correspond to two emission
bands, where discernible.
ever, due to the very low fluorescence quantum yields of the
l3-phosphinines, especially of compound 11, their fluores-
cence spectra should be interpreted with caution as effects
of minor impurities could have an influence on the experi-
mental findings, even though l3-phosphinines have been
carefully purified by recrystallization to assure highest possi-
ble purity.
lated absorption spectra for the l3-phosphinines perfectly
coincide with the experimental ones in energy as much as in
shape (Figure 6): the lowest energy absorption band is only
weakly allowed which should give rise to a low intensity
emission band (fluorescence) as indeed observed for the l3-
phosphinines. Moreover, red-edge absorption bands of large
oscillator strengths are predicted for the l5-phosphinines 12
Figure 6. Comparison between experimental and theoretical results on the photophysical properties of dithienyl (SXS) and pyridylthienyl (SXN only in
&
d). Theoretical ( , TD-DFT) versus experimental (D, room temperature) UV/Vis spectra of a) 5, b) 9 and c) 12 as well as d) a comparison between theo-
retical (*, DFT) and experimental (&) T1ꢀS0 energy gaps (designations of compounds as indicated).
Chem. Eur. J. 2007, 13, 4548 – 4559
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4553

C. Müller et al.
~
*
Figure 7. Low temperature UV/Vis ( ), fluorescence (&) and phosphorescence ( ) spectra of the SXS series: a) pyrylium salt 5 in dichloromethane
3
5
~
176 K; b) l -phosphinine (9) and c) l -phosphinine (12) in 2-methyltetrahydrofuran at 80 K; d) normalized room temperature UV/Vis ( ) and fluores-
cence (&) spectra of l5-phosphinine (12, in 2-methyltetrahydrofuran).
and 13 and pyrylium salts 5 and 7 by theoretical calculations,
which again is in agreement with the experimental findings
of large intensities and high fluorescence quantum yields as
well as high extinction coefficients of their lowest energy ab-
sorption bands (Figure 6).
From the theoretical calculations, we further identified
the lowest energy (red-edge) absorption bands of the series
SXS and SXN as p!p* transition. In the case of the pyryli-
um salts and l5-phosphinines these transition (mainly
HOMO!LUMO) are strongly allowed since they are ac-
companied by an apparent change of parity. Figure 8 shows
the frontier orbitals involved in this process for the l5-phos-
phinine 12.[20]
On the other hand, the red-edge absorption bands for the
l3-phosphinines 9 and 11 correspond to p!p* transitions,
rather than n!p* transitions, as originally attributed to l3-
phosphinines.[2a,21] They cause only a partial change of the
orbital symmetry (Figure 9) resulting in the observed low
fluorescence intensity of these compounds. As a matter of
fact, for compound 9 the red-edge absorption band is a com-
bination of the HOMOꢀ1!LUMO and HOMO!
LUMO+1 transition with the strongest contribution of the
first one.
The calculations indicate that this combination results in a
lower energy excitation compared with the HOMO!
LUMO transitions.
The strong absorption band at 314 and 312 nm for com-
pounds 9 and 11, respectively, correspond to the HOMO!
LUMO+1 transitions. Clearly the molecular orbitals in-
volved (Figure 9) are very similar to those involved in the
lowest energy excitations of the l5-phosphinines (Figure 8)
leading to higher oscillator strength of the respective transi-
tions.
Using time-gated detection, delayed luminescence spectra
could readily be obtained for all compounds of the SXS
(Figure 7, Table 1) and SXN (Table 1) series. These delayed
Figure 8. Representation of the frontier orbitals of the l5-phosphinine 12.
4554
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4548 – 4559

Polydentate Pyrylium Salts
FULL PAPER
have a comparatively high extinction coefficient, while for
l3-phosphinines the lowest energy absorption bands exhibit
comparatively low extinction coefficients. These findings
suggest that the energetically lowest transition is partially
forbidden in the case of l3-phosphinines, while it is strongly
allowed for l5-phosphinines and pyrylium salts. The theoret-
ical UV/Vis spectra obtained from time-dependent DFT cal-
culations support the interpretation of a partially forbidden
lowest p!p* transition of the l3-phosphinines as well
(Figure 6, Table 1). In agreement with the experimental
findings the lowest energy transitions of the l5-phosphinines
and pyrylium salts obtained from the calculations are al-
lowed (Figure 6, Table 1).
Figure 9. Representation of the frontier orbitals of the l3-phosphinine 9.
Conclusion
luminescence spectra were interpreted as phosphorescence
due to their large red-shifts > 100 nm compared with their
respective absorption spectra and their very long lifetimes in
the range of milliseconds (Table 1). For none of the com-
pounds a dependence of the delayed emission on the con-
centration could be observed, which excludes the presence
of excimers as the source of this long-lived emission. In con-
trast to the fluorescence measurements, phosphorescence
spectra were recorded with the same ease for all six com-
pounds, including the l3-phosphinines.
The energy difference between the electronic T1 and S0
states was also estimated based on the results of quantum-
chemical calculations. Comparison of experimental and the-
oretical values reveals inaccuracies (those inaccuracies aris-
ing from the neglecting of vibrational effects in the calcula-
tion are deemed small compared with the accuracy of the
measurement) which increase when going from l3- to l5-
phosphinines and pyrylium salts (Figure 6 bottom right,
Table 1). This phenomenon can be understood as the calcu-
lations assume isolated molecules in vacuum, while in the
experiment the molecules are enclosed in solvent cages of
finite dielectric constants. Due to the increasing ionic char-
acter in the order l3-, l5-phosphinine and pyrylium salts the
error introduced by this partially charged character will be
comparatively smaller for l3-phosphinines and comparative-
ly larger for the l5-phosphinines and the salts, in agreement
with the above findings. In addition to this, the low oscillator
strength of the lowest of the l3-phosphininesꢁ transitions
precludes strong interactions with the solvent cage.
From the photophysical data presented above, it can be
concluded that minor chemical changes of one heteroatom
and its substitution pattern in the central ring induces dis-
tinctly different photophysical properties in the two series
SXS and SXN. While considerable fluorescence (quantum
yields 5–99%) and readily detectable phosphorescence is
observed for l5-phosphinines and pyrylium salts, the l3-phos-
phinines show very little fluorescence intensity (quantum
yield !1%) but readily detectable phosphorescence. Fur-
thermore, the absorption spectra corroborate the results ob-
tained from the luminescence experiments. For l5-phosphi-
nines and pyrylium salts the lowest energy absorption bands
We have demonstrated, that the classical l3-phosphinine
synthesis via pyrylium salts is not only suitable for the prep-
aration of aryl-substituted phosphinines, but also for the in-
corporation of additional donor functionalities within the
phosphinine framework. Due to the stepwise assembly of
the phosphorus-containing heterocycle, the additional sub-
stituents can be positioned in a specific manner, allowing
access to a whole variety of polydentate l3-phosphabenzenes
with specific substitution pattern. We further showed that
also donor-functionalized l3-phosphinines can easily be con-
verted into the corresponding l5-phosphinines by simple
chemical modifications. Due to the increasing interest in
phosphorus-containing p-conjugated molecular materials we
started to explore the photophysical properties of a selected
series of thiophene- and pyridyl-substituted pyrylium salts,
l3- and l5-phosphinines. The results were compared and sup-
ported by theoretical calculations on the DFT level. While
the pyrylium salts as well as the l5-phosphinines show signif-
icant fluorescence in solution (quantum yields f=5–99%,
lifetimes t=3.59–5.36 ns), the heteroaromatic substituted l3-
posphinines show very little emission, which is attributed to
a partially forbidden p!p* transition and consistent with
the low oscillator strength predicted by DFT calculations for
this transition. Moreover, all three classes of compounds
show readily observable phosphorescence in solution (t in
the range of milliseconds), which was determined by time-
gated detection at low temperature.
We are currently investigating the influence of the addi-
tional donor functionalities on the stability of the corre-
sponding transition metal complexes as well as the applica-
tion of polydentate phosphinines in homogeneous catalysis.
Due to the modularity of the l3- and l5-phosphinine synthe-
sis we further study the possibility to tune the optoelectronic
properties of the latter systems and to develop this class of
compounds to novel p-conjugated organic materials.
Chem. Eur. J. 2007, 13, 4548 – 4559
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4555

C. Müller et al.
stirred and allowed to cool down to room temperature. The product was
filtered off and dried under vacuum, affording a yellow powder after re-
crystallization from ethanol/chloroform (17.5 g, 52.2 mmol, 63%). M.p.
124.88C; 1H NMR (200 MHz, CDCl3, 258C): d=3.32 (m, 2H; -CH2-),
Experimental Section
General considerations: All manipulations were carried out under an
argon atmosphere, using modified Schlenk techniques unless otherwise
stated. All glassware was dried prior to use by heating under vacuum. All
common solvents and chemicals were commercially available and pur-
3.73 (m, 2H; -CH2-), 4.12 (p, 3JH,H =7.0 Hz, 1H; -CH-), 7.06–7.46 (m,
3
7H), 7.58 (d, 3JH,H =4.0 Hz, 1H), 7.70–7.82 (m, 2H), 7.96 (d, JH,H
=
8.2 Hz, 1H), 8.64 ppm (d, 3JH,H =4.8 Hz, 1H); 13C NMR (50.3 MHz,
CDCl3, 258C): d=36.9 (-CH), 43.7, 45.9 (-CH2-), 121.8, 126.6, 127.2,
127.6, 128.1, 128.5, 132.0, 133.6, 136.9, 143.9 (Cq), 144.4 (Cq), 148.9, 153.3
(Cq), 191.4 (C=O), 199.9 ppm (C=O); elemental analysis calcd (%) for
C20H17NO2S (335.42): C 71.62, H 5.11, N 4.18; found: C 71.73, H 5.12, N
4.12.
chased from Aldrich Chemical Co. and Merck. PACHTRE(UGN SiMe3)3 was prepared
according to the literature.[22] The solvents were taken from custom-made
solvent purification columns filled with Al2O3. The elemental analyses
were obtained from H. Kolbe, Mikroanalytisches Laboratorium, Mülheim
1
a.d. Ruhr (Germany). H, 13C{1H}, 31P{1H} and 19F{1H} NMR spectra were
recorded on a Varian Mercury 200 or 400 spectrometer and all chemical
shifts are reported relative to the residual proton resonance in the deu-
terated solvents or referred to an 85% aqueous solution of H3PO4, re-
spectively.
2-(p-Methylphenyl)-6-(2-thienyl)-4-phenylpyrylium tetrafluoroborate (4):
Benzal-a-acetothienone (1) (8.90 g, 41.35 mmol) and p-methylacetophe-
none (2.80 g, 20.76 mmol) were dissolved in CH2Cl2 (15 mL). The mixture
was heated to 708C and HBF4·Et2O (7.1 g, 41.35 mmol, 52% ethereal so-
lution) was added dropwise. Stirring was continued at 708C for 6 h. Sub-
sequently, Et2O (100 mL) was added and an orange solid precipitated,
which was filtered off, washed with Et2O and dried under vacuum
(3.20 g, 7.7 mmol, 37%). M.p. 2388C; 1H NMR (200 MHz, CD3OD,
Photophysical experiments: 2-Methyltetrahydrofuran was pre-dried over
KOH for 4 d, filtered, distilled from CaH2, and stored under inert nitro-
gen atmosphere.
All sample solutions were prepared in a nitrogen glove box using dry and
deoxygenated 2-methyltetrahydrofuran, toluene, ethanol or dichlorome-
thane unless stated otherwise.
258C): d=2.49 (s, 3H, CH3), 7.47 (t, JH,H =5.0 Hz, 1H), 7.53 (d, JH,H
=
8.6 Hz, 2H), 7.75 (m, 3H), 8.30 (m, 5H), 8.56 (d, JH,H =4.0 Hz, 1H),
8.68 ppm (d, JH,H =5.6 Hz, 2H); 13C NMR (50.3 MHz, CD3OD, 258C):
d=20.5 (CH3), 112.8, 113.0, 126.1, 128.0, 129.1, 129.8, 130.4, 130.5, 132.8,
132.9, 134.6, 135.6, 138.3, 146.9, 165.0, 166.2, 167.8, 169.2 ppm; 19F NMR
(188.1 MHz, [D4]CH3OH, 258C): d=ꢀ154.25, ꢀ154.20 (3:1) ppm; ele-
mental analysis calcd (%) for C22H17BF4OS (416.24): C 63.48, H 4.12;
found: C 63.59, H 4.41.
UV/Vis and photoluminescence spectra were recorded using a Perkin–
Elmer Lambda 900 and an Edinburgh Instruments FS920 spectrophotom-
eter, respectively. Time-correlated single photon counting (TC-SPC) pho-
toluminescence was performed on an Edinburgh Instruments LifeSpec-
PS spectrometer by photoexcitation at 400 nm with a picosecond-pulse
laser (PicoQuant PDL 800B) operated at 2.5 MHz, 400 kHz or 40 kHz
and using a Peltier-cooled Hamamatsu microchannel plate photomultipli-
er (R3809U-50) for detection. During the spectroscopic experiments,
samples were held at room temperature or 80 K/176 K, under nitrogen
atmosphere (using an Oxford Optistat CF continuous flow cryostat). The
quantum yields at room temperature of l5-phosphinines (in 2-methylte-
trahydrofuran) and pyrylium salts (in dichloromethane) were determined
2,6-(2-Thienyl)-4-phenylpyrylium tetrafluoroborate (5): Benzal-a-aceto-
thienone (1) (17.44 g, 81.4 mmol) and 2-acetylthiophene (5.13 g,
40.7 mmol) were dissolved in CH2Cl2 (20 mL). The mixture was heated
to 708C and HBF4·Et2O (13.3 g, 81.4 mmol, 52% ethereal solution) was
added dropwise. Stirring was continued at 708C for 18 h. During that
period a dark precipitate was formed. Et2O (200 mL) was added and the
flask was stored in the refrigerator for 36 h. The precipitate was filtered
using N-(phenyl),N’-(1-ethylpropyl)perylenediimide (in toluene,
f
ꢁ0.99) as a standard. The quantum yields at room temperature of l3-
phosphinines (in 2-methyltetrahydrofuran) were determined using 2-ami-
nopyridine (in ethanol, f ꢁ0.37) as a standard.[23] Excitation wavelengths
were identical for samples and respective standard and all respective
samples had similar O.D. (< 0.1). The integrated emission spectra were
corrected according to common procedure and accuracies amounted to
ꢂ10% due to the very low concentration required to prevent stacking.[24]
Spectral resolution did not exceed 10 nm. Low temperature fluorescence
and phosphorescence was recorded using gated detection. Apulsed Nd/
YAG laser (Surelite II-10, Continuum, FWHM 4 ns, 10 Hz) was used for
excitation combined with an optical parametric oscillator (Panther OPO,
Continuum) to tune the excitation energy. Spectra were recorded with an
intensified CCD camera (PI-MAX:1024HQ, Princeton Instruments) after
dispersion of the emitted light by a spectrograph at gate widths of 2 ns–
20 ms delay times of 2 ns–10 ns for fluorescence and >100 ns for phos-
phorescence to minimize residual fluorescence.
off and recrystallized from hot methanol (6.9 g, 16.9 mmol, 21%). M.p.
3
247.88C; 1H NMR (200 MHz, CD3OD, 258C): d=7.47 (pseudot, JH,H
=
4.6 Hz, 2H), 7.75 (m, 3H), 8.24 (d, 3JH,H =4.8 Hz, 2H), 8.28 (pseudot,
3JH,H =7.0 Hz, 2H), 8.51 (d, 3JH,H =4.0 Hz, 2H), 8.62 ppm (s, 2H; hetero-
arom.-H); 13C NMR (50.3 MHz CD3OD, 258C): d=112.33, 129.06,
129.82, 130.35, 132.70 (Cq), 132.94 (Cq), 134.64 (Cq), 135.25, 138.07,
164.43 (Cq), 164.98 ppm (Cq); 19F NMR (188.1 MHz, [D4]CH3OH, 258C):
d=ꢀ154.70, ꢀ154.66 (3:1) ppm; elemental analysis calcd (%) for
C19H13BF4OS2 (408.24): C 55.90, H 3.21; found: C 56.18, H 3.19.
2-(2-Pyridyl)-4,6-diphenylpyrylium tetrafluoroborate (6): BF3·Et2O
(22.3 g, 157.3 mmol, 8 equiv) was added dropwise to a mixture of 1,3-di-
phenyl-5-(2-pyridyl)-1,5-pentanedione (6.5 g, 19.7 mmol, 1 equiv) and
benzylidene-2-acetophenone (4.1 g, 19.8 mmol, 1 equiv) at room tempera-
ture. The reaction mixture was then heated to 708C for 3 h. After allow-
ing the reaction mixture to cool down to room temperature a yellow
solid precipitated after adding Et2O. The yellow solid was collected on a
glass filter, washed with Et2O and recrystallized from methanol to obtain
the pyrylium salt as yellow needles (4.7 g, 60%). M.p. 2438C (decomp);
1H NMR (200 MHz, [D6]DMSO, 258C): d=7.90–7.76 (m, 7H; pyridyl-H,
pyryliumphenyl-Hm,p, phenyl-Hm,p), 8.26 (dt, JH,H =7.8 Hz, 1H; pyridyl-
H), 8.79–8.49 (m, 5H; pyridyl-H, pyryliumphenyl-Ho, phenyl-Ho), 9.20 (s,
1H; pyrylium-H), 9.00 (d, JH,H =4.4 Hz, 1H; pyridyl-H), 9.28 ppm (s, 1H;
pyrylium-H); 13C NMR (50.3 MHz, [D6]DMSO, 258C): d=115.7, 116.9,
125.0, 129.1, 129.3, 129.6, 130.4, 130.5, 132.7, 136.0, 139.1, 146.7, 151.5,
160.1, 168.2, 171.1 ppm; 19F NMR (188.1 MHz, [D6]DMSO, 258C): d=
ꢀ148.34 ppm; elemental analysis calcd (%) for C22H16BF4NO (397.18): C
66.53, H 4.06, N 3.53; found: C 67.51, H 4.13, N 3.42.
3,5-Diphenyl-1-(2-pyridyl)pentane-1,5-dione (2): The compound was pre-
pared according to a modified literature procedure.[12] Benzylidene-2-ace-
tophenone (8.12 g, 39.2 mmol) and 2-acetylpyridine (4.72 g, 39.2 mmol)
were mixed with NaOH (1.64 g, 39.2 mmol) using a mortar and pestle.
All compounds were stirred for about 10 min until a sticky yellow mix-
ture was obtained. The product was transferred into a flask containing
water/ethanol 1:2 and heated until everything was dissolved. The solution
was stirred and allowed to cool down to room temperature. The pure
product was filtered off and dried under vacuum, affording a white
powder (11.5 g, 34.9 mmol, 89%). M.p. 1048C; for spectroscopic data see
ref. [12].
1-(2-Thienyl)-3-phenyl-5-(2-pyridyl)-1,5-dione (3): Benzal-a-acetothie-
none (1) (17.7 g, 82.6 mmol) and 2-acetylpyridine (10.0 g, 82.6 mmol)
were mixed with NaOH (3.3 g, 82.6 mmol) using a mortar and pestle. All
compounds were stirred for about 10 min until a sticky yellow mixture
was obtained. The product was transferred into a flask containing water/
ethanol 1:2 and heated until everything was dissolved. The solution was
2-(2-Thienyl)-4-phenyl-6-(2-pyridyl)pyrylium tetrafluoroborate (7):
A
mixture of (15.3 g, 45.6 mmol), benzylidene-2-acetophenone (9.5 g,
3
45.6 mmol), trifluoroacetic acid (6 mL) and BF3·Et2O (24 mL) was
heated to 1008C and the dark red solution was stirred for 7 h. Subse-
quently, the mixture was cooled down to room temperature and Et2O
(100 mL) and acetone (100 mL) was added. Stirring was continued for
4556
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4548 – 4559

Polydentate Pyrylium Salts
FULL PAPER
overnight. The red precipitate (7a, bistetrafluoroborate salt) was filtered
off and dried under vacuum (6.5 g, 13.2 mmol, 29%). After recrystalliza-
tion from methanol the monotetrafluoroborate salt 7 was obtained as a
yellow powder (4.5 g, 11.2 mmol, 24.5%). M.p. 270.58C; 1H NMR
(200 MHz, [D6]DMSO, 258C): d=7.58 (t, JH,H =4.2 Hz, 1H), 7.76 (m,
4H), 8.25 (t, JH,H =7.8 Hz, 1H), 8.45 (m, 4H), 8.84 (d, JH,H =3.6 Hz, 1H),
8.95 (d, JH,H =4.2 Hz, 1H), 9.00 (s, 1H; heterocyclic-H), 9.13 ppm (s, 1H;
heterocyclic-H); 19F NMR (188.1 MHz, [D6]DMSO, 258C): d=
ꢀ148.34 ppm; elemental analysis calcd (%) for C20H14BF4NOS (403.20):
C 59.58, H 3.50, N 3.47; found: C 59.60, H 3.60, N 3.44.
2-(2-Thienyl)-4-phenyl-6-(2-pyridyl)-l3-phosphinine
(11):
PACHTREUNG(SiMe3)3
(6.0 g, 24.0 mmol, 2 equiv) was added dropwise at room temperature to a
stirred solution of 7 (4.14 g, 10.3 mmol, 1 equiv), dissolved in acetonitrile
(30 mL) in a 100 mL Schlenk flask. The resulting dark reaction mixture
was heated to 858C and subsequently heated under reflux for 6 h. After
cooling to room temperature, the volatiles were removed under vacuum.
The residue was dissolved in CH2Cl2 and added to an appropriate
amount of alumina (neutral) (ca. 3 g). Evaporation of the solvent was fol-
lowed by flash chromatography with petroleum ether/ethyl acetate 9:1
1
(0.65 g, 2.0 mmol, 20%). M.p. 129.08C; H NMR (200 MHz, C6D6, 258C):
2-(p-Methylphenyl)-6-(2-thienyl)-4-phenyl-l3-phosphinine (8): P
ACHTRE(UNG SiMe3)3
d=6.62 (m, 1H), 6.73 (m, 1H), 6.84 (m, 1H), 7.16 (m + solvent peak,
H), 7.36 (m, 1H), 7.44 (m, 2H), 7.75 (d, 3JH,P =8.0 Hz, 1H, P-heterocy-
clic-H), 8.19 (dd, J=5.8, 1.2 Hz, 1H), 8.52 (d, 3JH,P =4.9 Hz, 1H, P-heter-
ocyclic-H), 8.94 ppm (dd, J=5.8, 1.2 Hz, 1H); 13C NMR (50.3 MHz,
C6D6, 258C): d=120.7, 121.0, 122.5, 123.9, 124.2, 126.0, 126.2, 127.9, 131.3
(d, JC,P =18.5 Hz; C3/5), 132.1 (d, JC,P =21.6 Hz; C3/5), 136.2, 142.0 (d,
(1.70 g, 6.8 mmol, 2.6 equiv) was added dropwise at room temperature to
a stirred solution of 4 (1.1 g, 2.64 mmol, 1 equiv), dissolved in acetonitrile
(10 mL). The resulting dark reaction mixture was heated to 858C and
subsequently heated under reflux for 6 h. After cooling to room tempera-
ture, the volatiles were removed under vacuum. The residue was dis-
solved in CH2Cl2 (20 mL) and added to an appropriate amount of alumi-
na (neutral) (ca. 3–4 g). Evaporation of the solvent was followed by flash
chromatography with petroleum ether/ethyl acetate 19:1 (290 mg,
0.84 mmol, 32%). M.p. 126.38C; 1H NMR (400 MHz, C6D6, 258C): d=
2.09 (s, 3H, CH3), 6.74 (m, 1H), 6.84 (d, JH,H =5.2 Hz, 1H), 7.01 (d,
J
C,P =5.6 Hz; C4), 144.6 (d, JC,P =22.2 Hz; C3), 146.8 (d, JC,P =49.4 Hz;
C2), 149.7, 158.8 (d, JC,P =42.6 Hz; C2), 162.6, 163.6, 169.0, 170.0 ppm;
31P NMR (80.94 MHz, C6D6, 258C): d=182.85 ppm; elemental analysis
calcd (%) for C20H14NPS (331.37): C 72.49, H 4.26, N 4.23; found: C
72.11, H 4.32, N 4.06.
1,1-Dimethoxy-2,6-(2-thienyl)-4-phenyl-l5-phosphinine (12): Hg
ACHTREUNG(OAc)2
J
J
H,H =8.0 Hz, 2H), 7.18 (m + solvent peak, 3H), 7.38 (m, 3H), 7.54 (d,
H,H =6.8 Hz, 2H), 7.98 (d, 3JH,P =6.0 Hz, 1H; P-heterocyclic-H),
(100 mg, 0.31 mmol) was suspended in dry and degassed methanol
(10 mL) and cooled to ꢀ608C. Asolution of recrystallized 9 (100 mg,
0.3 mmol) in toluene (10 mL) was added and stirring was continued at
room temperature for overnight. The solvents were removed in vacuo
and the residue was chromatographed on neutral Al2O3 with toluene.
8.16 ppm (d, 3JH,P =5.6 Hz, 1H; P-heterocyclic-H); 13C NMR (50.3 MHz,
C6D6, 258C): d=20.7 (CH3), 123.9, 124.1, 126.1, 126.2, 128.8, 129.6, 130.1
(d, JC,P =12.0 Hz), 131.7 (d, JC,P =12.4 Hz), 137.6 (d, JC,P =1.9 Hz), 140.4
(d, JC,P =25.1 Hz), 142.1 (d, JC,P =3.1 Hz), 144.4, 144.6, 146.7 (d, JC,P
=
The product was obtained as an orange solid (94 mg, 0.24 mmol, 80%).
29.8 Hz), 163.5 (d, JC,P =51.7 Hz; C2,6), 172.1 ppm (d, JC,P =51.7 Hz, C2/
6); 31P NMR (80.94 MHz, C6D6, 258C): d=178.56 ppm; elemental analy-
sis calcd (%) for C22H17PS (344.41): C 76.72, H 4.98; found: C 76.20, H
4.83.
3
M.p. 135.48C; 1H NMR (200 MHz, C6D6, 258C): d=3.05 (d, JH,P
=
13.8 Hz, 6H, O-CH3), 6.76 (d, JH,H =2.4 Hz, 3H), 7.14 (m, 6H), 7.51 (m,
2H), 8.12 ppm (d, 3JH,P =36.2 Hz, 2H, P-heterocyclic-H); 13C NMR
(50.3 MHz, C6D6, 258C): d=52.1 (OCH3), 87.6, 90.3, 115.8, 116.1, 122.2,
2,6-(2-Thienyl)-4-phenyl-l3-phosphinine (9): P
ACHTRE(UNG SiMe3)3 (5.0 g, 20.0 mmol,
122.3, 125.0, 125.7, 138.1 (d,
JC,P =9.6 Hz), 141.5 (d, JC,P =7.7 Hz),
2 equiv) was added dropwise at room temperature to a stirred solution of
5 (4.0 g, 9.8 mmol, 1 equiv), dissolved in acetonitrile (20 mL) in a 100 mL
Schlenk flask. The resulting dark reaction mixture was heated to 858C
and subsequently heated under reflux for 6 h. After cooling to room tem-
perature, the volatiles were removed under vacuum. The residue was dis-
solved in CH2Cl2 (40 mL) and added to an appropriate amount of alumi-
na (neutral) (ca. 3–4 g). Evaporation of the solvent was followed by flash
chromatography with petroleum ether/ethyl acetate 9:1 (1.5 g, 4.53 mmol,
142.6 ppm; 31P NMR (80.94 MHz, C6D6, 258C): d=59.8 ppm; elemental
analysis calcd (%) for C21H19O2PS2 (398.47): C 63.30, H 4.81, found: C
64.83, H 5.02 (traces of toluene).
1,1-Dimethoxy-2-(2-thienyl)-4-phenyl-6-(2-pyridyl)-l5-phosphinine (13):
HgACHTRE(UNG OAc)2 (194 mg, 0.6 mmol) was suspended in dry and degassed metha-
nol (12 mL) and cooled to ꢀ608C. Asolution of recrystallized 11
(198 mg, 0.5 mmol) in toluene (12 mL) was added and stirring was con-
tinued at room temperature for overnight. The solvents were removed in
vacuo and the residue was chromatographed on neutral Al2O3 with tolu-
ene. The product was obtained as a yellow solid (165 mg, 0.42 mmol,
84%). M.p.: 107.58C; 1H NMR (200 MHz, C6D6, 258C): d=3.59 (d,
3JH,P =13.8 Hz; O-CH3), 7.01 (m, 2H), 7.20 (m, 3H), 7.38 (m, 2H), 7.52
1
46%). M.p. 115.78C; H NMR (200 MHz, C6D6, 258C): d=6.71 (m, 2H),
6.81 (m, 2H), 7.13 (m + solvent peak, 3H), 7.32 (m, 4H), 8.03 ppm (d,
3JH,P =5.8 Hz, 2H; P-heterocyclic-H); 13C NMR (50.3 MHz, C6D6, 258C)
d=124.03, 124.32, 126.20, 126.30, 128.85, 130.3 (d, JC,P =12.3 Hz; C3,5),
141.7 (d,
JC,P =3.1 Hz; C4), 144.8 (d, JC,P =13.4 Hz), 146.2 (d, JC,P =
3
4
30.3 Hz), 163.5 ppm (d, JC,P =50.6 Hz; C2,6); 31P NMR (80.94 MHz, C6D6,
258C): d=174.04 ppm; elemental analysis calcd (%) for C19H13PS2
(336.41): C 67.84, H 3.90; found: C 67.28, H 4.13.
(m, 2H), 7.61 (m, 2H), 7.95 (dd, JH,P =36.0 Hz, JH,H =2.8 Hz, 1H, P-het-
erocyclic-H), 8.34 (dd, 3JH,P =38.0 Hz, 4JH,H =2.6 Hz, 1H; P-heterocyclic-
H), 8.56 ppm (m, 1H); 31P NMR (80.94 MHz, C6D6, 258C): d=62.9 ppm;
elemental analysis calcd (%) for C22H20NO2PS (393.44): C 67.16, H 5.12,
N 3.56; found: C 68.50, H 5.32, N 3.36 (traces of toluene).
2-(2-Pyridyl)-4,6-diphenyl-l3-phosphinine (10): Under an argon atmos-
phere at room temperature, PACHTER(UNG SiMe3)3 (2.8 g, 11.2 mmol, 2.1 equiv) was
Crystal structure determination of 5, 9, 10: Crystals of 5, 9 and 10 suita-
ble for X-ray diffraction were obtained by slow recrystallization from hot
methanol (5) or acetonitrile (9, 10). Reflections were measured on a
Nonius KappaCCD diffractometer with rotating anode and graphite
monochromator (l=0.71073 Š). The reflections were corrected for ab-
sorption and scaled on the basis of multiple measured reflections with
the program SADABS.[25] The structures were solved with SHELXS-97
using Direct Methods[26] and refined with SHELXL-97[27] on F2 of all re-
flections. Non-hydrogen atoms were refined freely with anisotropic dis-
placement parameters. All hydrogen atoms were introduced in geometri-
cally optimized positions and refined with a riding model. Geometry cal-
culations, drawings and checking for higher symmetry were performed
added dropwise to a solution of 2-(2-pyridyl)-4,6-diphenylpyrylium tetra-
fluoroborate (2.0 g, 5.0 mmol, 1 equiv) in acetonitrile (12 mL) in a 50 mL
Schlenk flask. Upon adding a dark reaction mixture was obtained which
was heated under reflux at 858C for 6 h. Subsequently, all volatiles were
removed in vacuo to obtain a dark solid. The crude product was purified
by means of column chromatography over neutral alumina with ethyl
acetate/petroleum ether 1:5 to afford the product as a yellow-orange
solid (0.54 g, 30.7%). M.p. 146.58C; 1H NMR (200 MHz, C6D6, 258C):
d=7.21–7.08 (m, 8H), 7.64 (m, 2H), 7.47 (m, 2H), 7.85 (m, 1H), 8.10
(dd, J=5.7, 1.4 Hz, 1H), 8.55 (m, 1H), 9.10 ppm (dd, J=5.7, 1.4 Hz,
1H); 13C NMR (50.3 MHz, C6D6, 258C): d=120.8, 121.2, 122.4, 128.8,
132.0 (d, JC,P =13.0 Hz; C3/5), 132.9 (d, JC,P =11.9 Hz; C3/5), 136.2, 142.3
(d, JC,P =3.5 Hz; C4), 143.5, 144.0, 144.2 (d, JC,P =22.2 Hz), 149.8, 159.1
[28]
with the PLATON package.
(d, JC,P =26.1 Hz), 169.5 (d, JC,P =50.6 Hz; C2/6), 171.5 ppm (d, JC,P
=
CCDC-627278 (5), -627279 (9) and -627280 (10) contain the supplemen-
tary 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.
50.6 Hz; C2/6); 31P NMR (80.94 MHz, C6D6, 258C): d=187.35 ppm; ele-
mental analysis calcd (%) for C22H16NP (325.35): C 81.22, H 4.96, N 4.31;
found: C 80.84, H 5.24, N 3.97.
Chem. Eur. J. 2007, 13, 4548 – 4559
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4557

C. Müller et al.
Compound 5: [C19H13OS2]
C
phorus Chemistry (Eds.: M. Regitz, O. J. Scherer), Thieme, Stuttgart,
1990, pp. 220–257; d) P. Le Floch, in Phosphorus-Carbon Heterocy-
clic Chemistry: The Rise of a New Domain (Ed.: F. Mathey), Perga-
mon, Palaiseau, 2001, pp. 485–533; e) F. Mathey, Angew. Chem.
2003, 115, 1616–1643; Angew. Chem. Int. Ed. 2003, 42, 1578–1604;
f) K. Dimroth, Acc. Chem. Res. 1982, 15, 58–64; g) R. Streubel, Sci-
ence of Synthesis 2005, 15, 1157–1179; h) K. Dimroth, Houben-
Weyl, Vol. E1, p. 815.
to a resolution of (sinq/l)max =0.65 Š at a temperature of 110 K. 0.78–
0.95 absorption correction range. 4025 Reflections were unique (Rint
0.0227). One thienyl moiety was rotationally disordered and refined with
an occupancy of 0.893:0.107. 258 Parameters were refined with eight re-
straints. R1/wR2 [I
> 2s(I)]: 0.0308/0.0801. R1/wR2 [all refl.]: 0.0357/
[3] a) W. Schäfer, A. Schweig, K. Dimroth, H. Kanter, J. Am. Chem.
Soc. 1976, 98, 4410–4418; b) Z. X. Wang, P. v. R. Schleyer, Helv.
Chim. Acta 2001, 84, 1578–1600.
0.0833. S=1.037. Residual electron density between ꢀ0.31 and
ꢀ3
0.38 eŠ
.
Compound 9: C19H13PS2, Fw =336.38, yellow needle, 0.36”0.06”
0.06 mm3, monoclinic, C2/c (no. 15), a=35.3863(5), b=5.0875(1), c=
[4] P. Le Floch, D. Carmichael, L. Ricard, F. Mathey, J. Am. Chem. Soc.
1993, 115, 10665–10670.
[5] J. M. Alcaraz, A. Br›que, F. Mathey, Tetrahedron Lett. 1982, 23,
1565–1568.
[6] For methoxy- and hydroxy-functionalized phosphinines see: B.
Breit, R. Winde, T. Mackewitz, R. Paciello, K. Harms, Chem. Eur. J.
2001, 7, 3106–3121 and ref. [7].
[7] C. Müller, L. Guarrotxena López, H. Kooijman, A. L. Spek, D.
Vogt, Tetrahedron Lett. 2006, 47, 2017–2020.
19.1892(3) Š, b=111.2405(6)8, V=3219.91(9) Š3, Z=8, 1calcd
=
1.388 gcmꢀ3, m=0.42 mmꢀ1. 24040 Reflections were measured up to a
ꢀ1
resolution of (sinq/l)max =0.62 Š at a temperature of 150 K. 0.92–0.98
absorption correction range. 3265 Reflections were unique (Rint =0.0557).
The thienyl and phenyl moieties were rotationally disordered and refined
with occupancies of 0.64:0.36, 0.57:0.43 and 0.50:0.50, respectively. 271
Parameters were refined with 78 restraints. R1/wR2 [I > 2s(I)]: 0.0425/
0.1066. R1/wR2 [all refl.]: 0.0598/0.1175. S=1.069. Residual electron densi-
[8] a) T. Baumgartner, R. RØau, Chem. Rev. 2006, 106, 4681–4727;
b) H.-C. Su, O. Fadhel, C.-J. Yang, T.-Y. Cho, C. Fave, M. Hissler,
C.-C. Wu, R. RØau, J. Am. Chem. Soc. 2006, 128, 983–995; c) J.
Casado, R. RØau, J. T. López Navarrete, Chem. Eur. J. 2006, 12,
3759–3767; d) T. Baumgartner, W. Wilk, Org. Lett. 2006, 8, 503–
506; e) N. H. T. Huy, E. Perrier, L. Ricard, F. Mathey, Organometal-
lics 2006, 25, 5176–5179; f) T. Baumgartner, W. Bergmans, T.
Kµrpµrti, T. Neumann, M. Nieger, M. Nyulµszi, Chem. Eur. J. 2005,
11, 4687–4699; g) L. Zhang, M. Hissler, H.-B. Bu, P. Bäuerle, C.
Lescop, R. RØau, Organometallics 2005, 24, 5369–5376; h) C. Fave,
M. Hissler, T. Kµrpµti, J. Rault-Berthelot, V. Deborde, L. Toupet, L.
Nyulµszi, R. RØau, J. Am. Chem. Soc. 2004, 126, 6058–6063; i) T.
Baumgartner, T. Neumann, B. Wirges, Angew. Chem. 2004, 116,
6323–6328; Angew. Chem. Int. Ed. 2004, 43, 6197–6201; j) C Fave,
T.-Y. Cho, M. Hissler, C.-W. Chen, T.-Y. Luh, C.-C. Wu, R. RØau, J.
Am. Chem. Soc. 2003, 125, 9254–9255; k) C. Hay, C. Fave, M. Hiss-
ler, J. Rault-Berthelot, R. RØau, Org. Lett. 2003, 5, 3467–3470; l) C.
Hay, M. Hissler, C. Fischmeister, J. Rault-Berthelot, L. Toupet, L.
Nyulµszi, R. RØau, Chem. Eur. J. 2001, 7, 4222–4236; m) C. Hay, C.
Fischmeister, M. Hissler, L. Toupet, R. RØau, Angew. Chem. 2000,
112, 1882–1885; Angew. Chem. Int. Ed. 2000, 39, 1812–1815.
[9] a) G. N. Dorofeenko, G. A. Korol’chenko, S. V. Krivun, Khim. Geter-
otsikl. Soedin. 1965, 6, 817–821; b) J.A. VanAllan, G. A. Reynolds,
J. Org. Chem. 1968, 33, 1102–1105; c) S. S. Lin, C. Y. Li, X. Wang,
Chin. Chem. Lett. 2002, 13, 605–606.
ꢀ3
ty between ꢀ0.41 and 0.66 eŠ
.
Compound 10: C22H16NP, Fw =325.33, orange-red block, 0.36”0.24”
0.15 mm3, orthorhombic, Pna21 (no. 33), a=7.4375(2), b=19.8459(6), c=
11.3381(2) Š, V=1673.55(7) Š3, Z=4, 1calcd =1.291 gcmꢀ3, m=0.17 mmꢀ1
.
33035 Reflections were measured up to a resolution of (sinq/l)max
=
ꢀ1
0.65 Š at a temperature of 150 K. 0.53–0.98 absorption correction
range. 2013 Reflections were unique (Rint =0.0470). The absolute struc-
ture could not be refined reliably; thus, Friedel pairs were merged prior
to the refinement. All ring systems were rotationally disordered and re-
fined with occupancies of 0.67:0.33, 0.79:0.21 and 0.92:0.08, respectively.
244 Parameters were refined with 73 restraints. R1/wR2 [I
0.0504/0.1256. R1/wR2 [all refl.]: 0.0551/0.1291. S=1.129.
> 2s(I)]:
Computational details: The quantum chemical calculations were carried
out within the density functional theory using the Gaussian 03[25] program
at the B3LYP/6-31G(d) level. Full geometry optimization was performed
for compounds 5, 7, 9, 11, 12 and 13 both in a singlet (S0) and triplet (T1)
state. The T1ꢀS0 energy gaps were calculated as l
A
,
where ET1 and ES0 are the total electronic energies of the corresponding
compound in the triplet and singlet state, respectively. The theoretical
UV/Vis spectra were calculated at the same level of theory as the geome-
try optimization using the time-dependent DFT method that was imple-
mented in the Gaussian 03 program package. Compounds 5 and 9 were
modeled as isolated pyrylium cations while the influence of the counter-
ion was neglected. Such an approximation gave better agreement be-
tween the experimental and theoretical results as compared to those ob-
[10] a) A. R. Katritzky, S. S. Thind, J. Chem. Soc. Perkin Trans. 1 1980,
1895–1900; b) A. R. Katritzky, E. M. Elisseou, R. C. Patel, B. Plau,
J. Chem. Soc. Perkin Trans. 1 1982, 125–130; c) K. Dimroth, C.
Reichardt, K. Vogel, Org. Synth. Coll. Vol. V, 1135–1137; d) A. R.
Katritzky, J. M. Lloyd, R. C. Patel, J. Chem. Soc. Perkin Trans. 1
1982, 117–123.
tained for the contact ion pair. Doppler broadening with a band width of
1
8 nm on = height of the band was used for the visualization of the calcu-
2
lated absorption spectra.
[11] Pyrylium salts show in many cases strong fluorescence in solution
and have been used as laser dyes. See for example: D. Basting, F. P.
Schäfer, B. Steyer, Appl. Phys. 1974, 3, 81–88.
Acknowledgements
[12] C. Weygand, F. Strobelt, Chem. Ber. 1935, 68, 1839–1847.
[13] A. R. Katritzky, M. Rezende, J. Chem. Res. 1980, 312–313.
[14] For the crystal and molecular structure of pyrylium salts see also:
a) M. Gdaniec, I. Turowksa-Tyrk, T. M. Krygowski, J. Chem. Soc.
Perkin Trans. 2 1989, 613–616; b) I. Turowska-Tyrk, T. M. Krygow-
ski, P. Milart, G. Butt, R. D. Topsom, J. Mol. Structure 1991, 245,
289–299; c) I. Turowska-Tyrk, T. M. Krygowski, P. Milart, J. Mol.
Structure 1991, 263, 235–245; d) I. Turowska-Tyrk, R. Anulewicz,
T. M. Krygowski, B. Pniewska, P. Milart, Pol. J. Chem. 1992, 66,
1831.
This work was supported in part (M.L., A.L.S.) by the Council for the
Chemical Sciences of the Netherlands Organization for Scientific Re-
search (CW-NWO).
[1] a) G. Märkl, Angew. Chem. 1966, 78, 907–908; Angew. Chem. Int.
Ed. Engl. 1966, 5, 846; b) A. J. Ashe III, J. Am. Chem. Soc. 1971, 93,
3293–3295; c) G. Märkl, F. Lieb, A. Merz, Angew. Chem. 1967, 79,
59; Angew. Chem. Int. Ed. Engl. 1967, 6, 458; d) K. Dimroth, W.
Städe, Angew. Chem. 1968, 80, 966–967; Angew. Chem. Int. Ed.
Engl. 1968, 7, 981.
[2] a) K. Dimroth, Top. Curr. Chem. 1973, 38, 1–147; b) P. Jutzi, Angew.
Chem. 1975, 87, 269–283; Angew. Chem. Int. Ed. Engl. 1975, 14,
232; c) G. Märkl in Multiple Bonds and Low Coordination in Phos-
[15] G. W. V. Cave, C. L. Raston, J. Chem. Soc. Perkin Trans. 1 2001,
3258–3264.
[16] Bright-yellow crystals of 7 were obtained by slow recrystallization
from acetonitrile but no satisfactory X-ray crystal structure determi-
4558
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4548 – 4559

Polydentate Pyrylium Salts
FULL PAPER
nation could be performed. However, the X-ray data show that 7
crystallizes as the monotetrafluoroborate pyrylium salt.
[22] E. Nieke, H. Westermann, Synthesis 1988, 330.
[23] T. Mutai, J.-D. Cheon, G. Tsuchiya, K. Araki, J. Chem. Soc. Perkin
Trans. 2 2002, 862–865.
[24] D. F. Eaton, Pure Appl. Chem. 1988, 60, 1107–1114.
[25] G. M. Sheldrick, SADABS: Area-Detector Absorption Correction,
v2.10, Universität Gçttingen (Germany), 1999.
[26] G. M. Sheldrick, SHELXS-97. Program for crystal structure solution,
University of Gçttingen (Germany), 1997.
[17] For X-ray structures of l3-phosphinines see for example: a) W.
Fischer, E. Hellner, A. Chatzidakis, K. Dimroth, Tetrahedron Lett.
1968, 59, 6227–6230; b) J. C. J. Bart, J. J. Daly, Angew. Chem. 1968,
80, 843–844; Angew. Chem. Int. Ed. Engl. 1968, 7, 811; c) G. Maas,
J. Fink, H. Wingert, K. Blatter, M. Regitz, Chem. Ber. 1987, 120,
819; d) N. Avavari, N. MØzailles, L. Ricard, P. Le Floch, F. Mathey,
Science 1998, 280, 1587–1589; e) S. Choua, C. Dutan, L. Cataldo, T.
Berclaz, M. Geoffroy, N. MØzailles, A. Moores, L. Ricard, P. Le
Floch, Chem. Eur. J. 2004, 10, 4080–4090; f) C. Müller, M. Lutz,
A. L. Spek, D. Vogt, J. Chem. Crystallogr. 2006, 36, 869–874.
[18] Bright yellow crystals of 11 were further obtained by recrystalliza-
tion from hot CH3CN and were subject to an X-ray crystal structure
analysis. However, due to a high degree of disorder the structure
could not be refined sufficiently.
[27] G. M. Sheldrick, SHELXL-97. Program for crystal structure refine-
ment, University of Gçttingen (Germany), 1997.
[28] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–13.
[29] Gaussian 03, Revision B.05, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M.A. Robb, J. R. Cheeseman, J.A. Montgomer-
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J.V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M.A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J.A. Pople, Gaussian, Inc., Pittsburgh
PA, 2003.
[19] For a structural characterization of a l5-phosphinine see: U. The-
walt, C. E. Bugg, Acta Crystallogr. 1972, B28, 871–879.
[20] For the nature of bonding in l5-phosphinines see ref. [2f] and: W.
Schäfer, A. Schweig, K. Dimroth, H. Kanter, J. Am. Chem. Soc.
1976, 98, 4410–4418.
[21] For the nature of bonding in l3-phosphinines see for example: a) H.
Oehling, A. Schweig, Tetrahedron Lett. 1970, 56, 4941–4944; b) H.
Oehling, W. Schäfer, A. Schweig, Angew. Chem. 1971, 83, 723–725;
Angew. Chem. Int. Ed. Engl. 1971, 10, 656–657; c) A. Schweig, W.
Schäfer, Phosphorus 1972, 1, 203; d) A. Schweig, W. Schäfer, K.
Dimroth, Angew. Chem. 1972, 84, 1146–1147; Angew. Chem. Int.
Ed. Engl. 1972, 11, 631–636; e) C. Batich, E. Heilbronner, V. Hor-
nung, A. J. Ashe III, D. T. Clark, U. T. Cobley, D. Kilcast, I. Scanlan,
J. Am. Chem. Soc. 1973, 95, 928–930; f) E. F. DiMauro, M. C. Ko-
zlowski, J. Chem. Soc. Perkin Trans. 1 2002, 439.
Received: November 17, 2006
Published online: March 9, 2007
Chem. Eur. J. 2007, 13, 4548 – 4559
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4559