Reactivity of cyclooligophosphanes: Synthesis and structural characterisation of cyclo-1,4-(BH3)2(P4Ph 4CH2) and cyclo-1,2-(BH3)2(P 5Ph5)

Article abstract of DOI:10.1039/b511833f

The borane complexes cyclo-1,4-(BH₃)₂(P ₄Ph₄CH₂) (3) and cyclo-1,2-(BH ₃)₂(P₅Ph₅) (4) were prepared by reaction of cyclo-(P₄Ph₄CH₂) and cyclo-(P ₅Ph₅) with BH₃(SMe₂). Only the 2: 1 complexes 3 and 4 were isolated, even when an excess of the borane source was used. In solution, 3 exists as a mixture of the two diastereomers (R _P*,S_P*,S_P*,R _P*)-(±)-3 and (R_P*,R _P*,R_P*,R_P*)-(±)-3. However, in the solid state the (R_P*,S_P*, S_P*,R_P*)-(±) diastereomer is the major stereoisomer. Similarly, while only one isomer of 4 is observed in its X-ray structure, NMR spectroscopic investigations reveal that it forms a complex mixture of isomers in solution. 3 may be deprotonated with ^tBuLi to give the lithium salt cyclo-1,4-(BH₃)₂(P ₄Ph₄CHLi) (3·Li), though this could not be isolated in pure form. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b511833f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Reactivity of cyclooligophosphanes: synthesis and structural characterisation
of cyclo-1,4-(BH3)2(P4Ph4CH2) and cyclo-1,2-(BH3)2(P5Ph5)†
Robert Wolf,a Markus Finger,‡b Carolin Limburg,a Anthony C. Willis,c S. Bruce Wildc and
Evamarie Hey-Hawkins*a
Received 19th August 2005, Accepted 11th October 2005
First published as an Advance Article on the web 1st November 2005
DOI: 10.1039/b511833f
The borane complexes cyclo-1,4-(BH3)2(P4Ph4CH2) (3) and cyclo-1,2-(BH3)2(P5Ph5) (4) were prepared
by reaction of cyclo-(P4Ph4CH2) and cyclo-(P5Ph5) with BH3(SMe2). Only the 2 : 1 complexes 3 and 4
were isolated, even when an excess of the borane source was used. In solution, 3 exists as a mixture of
the two diastereomers (RP*,SP*,SP*,RP*)-( )-3 and (RP*,RP*,RP*,RP*)-( )-3. However, in the solid
state the (RP*,SP*,SP*,RP*)-( ) diastereomer is the major stereoisomer. Similarly, while only one
isomer of 4 is observed in its X-ray structure, NMR spectroscopic investigations reveal that it forms a
complex mixture of isomers in solution. 3 may be deprotonated with tBuLi to give the lithium salt
cyclo-1,4-(BH3)2(P4Ph4CHLi) (3·Li), though this could not be isolated in pure form.
such complexes have not yet been described in the literature,
Introduction
except for an early report by Cowley and Pinell, who treated
n
n
While the chemistry of linear and cyclic oligophosphanes has
been an active field of research for many decades, this area
has attracted increased attention in recent years. Among other
examples, this is illustrated by our investigations of the cyclopen-
cyclooligophosphanes cyclo-(PnRn) (R = Et, Pr, Bu, Ph) with
boron trihalides.5 No crystallographic or NMR data were given
at the time, and the structural identity of the isolated compounds
could not be firmly established. Second, complexation of boranes
is often used to modify the properties of tertiary phosphanes.6
Therefore, we reasoned that the formation of a borane complex of
cyclo-(P4Ph4CH2) might facilitate the formation of the respective
tetraphosphacyclopentanide anion.
t
taphosphanide anion cyclo-(P5 Bu4)−, which have unravelled its
fascinating coordination chemistry.1,2 Motivated by these results
we also showed that the carbon analogue of this anion, the
tetraphosphacyclopentanide anion cyclo-(P4 Bu4CH)− (1), can be
t
t
prepared by deprotonation of cyclo-(P4 Bu4CH2) with lithium
Interestingly, while BH3 complexes of cyclooligophosphanes
have remained elusive, Baudler et al. reported reactions of the
cyclooligophosphanes cyclo-(P4Ph4CH2) (2) and cyclo-(P5Ph5) (5)
with elemental sulfur to give the sulfides cyclo-(S2P4Ph4CH2)
(2·2S)7 and cyclo-(SP4Ph4).8 While 2·2S has an intact P4C ring,
the P4S heterocycle in cyclo-(SP4Ph4) formally originates from the
insertion of a sulfur atom into cyclo-(P4Ph4). Recently, Burford
et al. showed that phosphonium cations of cyclooligophosphanes
such as [cyclo-(P3tBu3Me)]OTf, [cyclo-(P4tBu3Me2)]OTf, [cyclo-
(P4Cy4Me)]OTf and [cyclo-(P5Ph4RRꢀ)]OTf (R, Rꢀ = Me, Ph) can
be prepared by reaction of cyclooligosphanes with phosphenium
triflates or methyl triflate.9 These reports are of particular relevance
to the present study, as the BH3 complexes presented here may
be regarded as neutral, isoelectronic analogues of these cationic
compounds.
alkyls.3 In contrast, ring cleavage is observed in the same reaction
with cyclo-(P4Ph4CH2) (2).
In the context of our ongoing investigations on oligophos-
phanide chemistry,1–4 we now report the synthesis and structural
characterisation of two BH3 complexes of cyclooligophosphanes,
cyclo-1,4-(BH3)2(P4Ph4CH2) (3) and cyclo-1,2-(BH3)2(P5Ph5) (4).
The purpose of this study was twofold: first, borane complexes
of cyclooligophosphanes are interesting objects of study, as their
structures give insights into the reactivity of cyclooligophosphanes
and particularly into the relative nucleophilicity of the coordi-
nated and uncoordinated phosphorus atoms. To our knowledge,
aInstitut fu¨r Anorganische Chemie der Universita¨t Leipzig, Johannisallee 29,
04103 Leipzig, Germany. E-mail: hey@rz.uni-leipzig.de; Fax: (+49)341-
9739319; Tel: (+49)341-9736151
bWilhelm-Ostwald-Institut fu¨r Physikalische und Theoretische Chemie der
Universita¨t Leipzig, Johannisallee 29, 04103, Leipzig, Germany
Results and discussion
cResearch School of Chemistry, Australian National University, Canberra,
Australian Capital Territory 0200, Australia
Complexes 3 and 4 were prepared (Scheme 1) by reaction of two
equivalents of BH3(SMe2) with cyclo-(P4Ph4CH2) (2) and cyclo-
(P5Ph5) (5). Significantly, formation of BH3 complexes with a larger
number of BH3 groups in the molecule was not observed when an
† Electronic supplementary information (ESI) available: (a) Graphical
representations of the experimental and simulated 31P{ H,11B} NMR
1
spectra of 3; (b) graphical representation of the 31P MAS spectrum of
3; (c) graphical representation of the low-temperature NMR experiment
1
excess of the BH3 source was used. This was indicated by 31P{ H}
on 3; (d) graphical representation of the 31P{ H}–31P{ H} COSY NMR
spectrum of 4; (e) graphical representations of the optimised B3LYP
structures of the symmetrical diastereomers of cyclo-(P4Ph4CH2) (2) and
cyclo-1,4-(BH3)2(P4Ph4CH2) (3). See DOI: 10.1039/b511833f
1
1
NMR spectra of the reaction mixtures, which remained practically
unchanged, and by the fact that 3 and 4 were isolated in moderate
to good yields regardless of whether two or more equivalents of
BH3 were used.
‡ Present address: Organisch-Chemisches Institut der Universita¨t
Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg, Germany.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 831–837 | 831
©

View Article Online
Scheme 1 Synthesis of 3, 3·Li, 2·CH2Ph and 4.
Both complexes were studied in detail by multinuclear NMR
spectroscopy and X-ray crystallography (Table 1). 3 crystallises in
¯
the space group P1 with two molecules in the unit cell (Fig. 1).
The asymmetric unit contains the (SP,RP,RP,SP) enantiomer, while
the opposite enantiomer (RP,SP,SP,RP)-3 is generated by the
inversion centre, so that the compound exists as a racemate. The
four symmetric diastereomers of 3 in which the pairs of borane
groups are mutually cis or trans to one another are depicted
in Fig. 2.10 During refinement, minor, but significant, disorder
of the atoms P2 and P3 and their attached phenyl groups was
observed. These atoms were refined with occupancy factors of
94.3 : 5.7%. The disorder can be interpreted in terms of an
inversion of the respective phosphorus atoms, which corresponds
to the presence of a minor amount of the (RP*,RP*,RP*,RP*)-( )
diastereomer of 3 in the solid state (Fig. 3). Significantly, the same
diastereomer is also observed in solutions of 3 (vide infra). Due to
the presence of this diastereomer in the crystal and partial overlap
of the two components, the disorder could not be fully resolved,
and the following structural discussion is restricted to the major
component (RP*,SP*,SP*,RP*)-( )-3.
Fig. 1 Solid-state molecular structure of cyclo-1,4-(BH3)2(P4Ph4CH2)
(3) (disorder and H atoms except those on C1, B1 and B4 omitted
◦
˚
for clarity). Selected bond lengths (A) and angles ( ): P1–P2 2.229(1),
P2–P3 2.2184(8), P3–P4 2.2291(9), P1–B1 1.928(3), P4–B4 1.911(3),
P1–C1 1.839(2), P4–C1 1.835(2), mean P–C(ipso) 1.818; P1–C1–P4
114.6(1), C1–P1–P2 101.74(8), P1–P2–P3 91.73(3), P2–P3–P4 91.23(3),
P3–P4–C1 102.83(7), C1–P1–B1 105.9(2), C1–P4–B4 108.3(2), B1–P1–P2
124.3(1), B4–P4–P3 118.8(1), C11–P1–C1 107.0(1), C11–P1–P2 100.85(7),
C11–P1–B1 105.9(2), C41–P4–C1 107.8(1), C41–P4–P3 101.91(8),
C41–P4–B4 116.0(2), C21–P2–P1 103.25(8), C21–P2–P3 102.00(9),
C31–P3–P2 103.15(8), C31–P3–P4 97.64(8).
In this isomer, the BH3 groups are connected in a transoid
arrangement to the two phosphorus atoms (P1 and P4) adjacent
Table 1 Crystal data for cyclo-1,4-(BH3)2(P4Ph4CH2) (3), cyclo-1,2-
(BH3)2(P5Ph5) (4) and [cyclo-1,2-(BH3)2(P5Ph5)]·CH2Cl2 (4·CH2Cl2)
3
4
4·CH2Cl2
to the ring carbon atom (C1). The phenyl groups are in an all-
Empirical formula
M
C25H28B2P4
473.97
223(2)
Triclinic
¯
P1
10.717(2)
10.717(2)
12.340(2)
78.798(3)
75.012(3)
72.274(3)
1293.7(3)
2
1.217
30.80
10634
7547 (0.019)
C30H31B2P5
568.02
207(2)
Monoclinic
P21/n
10.775(5)
22.123(5)
13.275(5)
90
108.435(5)
90
3002.0(19)
4
1.257
C31H33B2Cl2P5
653.00
200(2)
Monoclinic
C2/c
19.6208(3)
10.1953(1)
33.8163(5)
90
96.7138(6)
90
6718.2(2)
8
1.291
trans arrangement with respect to each other. The B–P and P–P
T/K
˚
distances are in the expected ranges [d(P1–B1) = 1.928(3) A; d(P4–
Crystal system
Space group
˚
B4) = 1.911(3) A; d(P1–P2, P2–P3, P3–P4) = 2.218(1)–2.229(1)
A].11,12 In contrast to the starting material cyclo-(P4Ph4CH2), which
˚
˚
a/A
has an envelope conformation in the solid state,13 the P4C ring in
˚
b/A
˚
c/A
(RP*,SP*,SP*,RP*)-( )-3 has a twist conformation in which the
a/◦
b/◦
c /◦
˚
two b-P atoms P2 and P3 are displaced by 0.789 and 0.661 A,
respectively, from the plane formed by P1, P4 and C1. Interestingly,
the disufilde 2·2S shows a very similar arrangement:13 the sulfur
atoms are bound to the two a-P atoms in a transoid fashion and
the P4C heterocycle has a similar twist conformation.
Crystals of complex 4 could be obtained in an unsolvated
form (4, space group P21/n) and in a solvated form (4·CH2Cl2,
space group C2/c) with one molecule of CH2Cl2 per formula unit
(Table 1). As the structural features of 4 and 4·CH2Cl2 are rather
similar, we restrict our discussion to the unsolvated form 4. Only
3
˚
V/A
Z
Dc/Mg m−3
◦
hmax
/
29.46
19111
7494 (0.027)
1.037
25.01
31775
5896 (0.04)
1.0432
Total data
Unique data (Rint
)
Goodness of fit on F2 0.996
R1, wR2 [I > 2r(I)]
R1, wR2 (all data)
0.0502, 0.1094 0.0382, 0.0900 0.0356, 0.0394
0.0905, 0.1239 0.0640, 0.1005 0.0686, 0.0582
832 | Dalton Trans., 2006, 831–837
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Fig. 4 Solid-state molecular structure of cyclo-1,2-(BH3)2(P5Ph5) (4)
Fig. 2 The four symmetric diastereomers of 3. Only one enantiomer of
the chiral diastereomers is depicted.
(H atoms except those on B1 and B2 omitted for clarity). Se-
lected bond lengths (A) and angles (◦): P1–P2 2.2221(8), P2–P3
˚
2.2234(8), P3–P4 2.2430(8), P4–P5 2.2207(8), P1–P5 2.206(1), P1–B1
1.941(2), P2–B2 1.956(2), mean P–C(ipso) 1.828; P1–P2–P3 103.69(3),
P2–P3–P4 101.53(3), P3–P4–P5 96.22(4), P4–P5–P1 92.78(2), C1–P1–B1
114.6(1), C1–P1–P2 104.11(6), C1–P1–P5 105.34(6), B1–P1–P2 111.56(9),
B1–P1–P5 120.76(9), C7–P2–B2 114.3(1), C7–P2–P1 101.12(6), C7–P2–P3
105.50(6), B2–P2–P1 111.90(9), B2–P2–P3 118.46(9), C13–P3–P2
101.69(7), C13–P3–P4 99.16(7), C19–P4–P3 98.06(6), C19–P4–P5
99.44(6), C25–P5–P4 102.88(7), C25–P5–P1 100.12(6).
(Fig. 2), the major component presumably corresponds to the
energetically favoured all-trans diastereomer (RP*,SP*,SP*,RP*)-
( )-3. The second component may be another diastereomer of
3 with a different relative orientation of the phenyl substituents
(Fig. 2). The 31P coupling parameters obtained by analysis of
Fig. 3 Core structure of 3 (only P, B and their directly attached carbon
atoms are shown; atoms of the minor component◦ are indicated by
1
the 31P{ H,11B} NMR spectra (see Experimental section), differ
˚
dashed bonds). Selected bond lengths (A) and angles ( ): P5–P6 2.195(2);
significantly from those of the related compounds 2 and 2·2S; in
P1–P5–P6 84.3(5), P4–P6–P5 84.1(5), C21f–P5–P1 95(1), C21f–P5–P6
109(2), C33f–P6–P4 98(2), C33f–P6–P5 106(2).
1
particular, the JPP coupling constants are rather small. Another
interesting difference is that only one diastereomer is observed in
the NMR spectra of 2 and 2·2S, although the solid-state structure
2·2S is very similar to that of 3. Unfortunately, more precise
structural information could not be obtained from the 31P NMR
parameters of 3, as more closely related reference compounds
are not available for comparison. Attempts to separate the two
components by TLC or HPLC failed. Similarly, the ratio of the two
isomers remained practically unchanged even after six consecutive
recrystallisations of 3 from CH2Cl2–n-hexane (2 : 1).
one diastereomer of 4 is present in the unit cell, corresponding
to two enantiomers related by the crystallographic centre of
symmetry (Fig. 4). The two BH3 groups are coordinated to two
adjacent phosphorus atoms of the cyclo-(P5Ph5) unit. The resulting
P2Ph2(BH3)2 moiety is in a syn conformation (torsion angle B1–
P1–P2–B2: 35.6◦). The remaining phenyl substituents display an
all-trans arrangement. As in 5,14 the P5 ring is in an envelope
conformation. The P–P bond lengths are normal [d(P1–P2, P2–
12
˚
P3, P3–P4, P4–P5, P5–P1) = 2.206(1)−2.243(1) A] and the B–P
With these observations in mind, quantum chemical calcula-
tions were carried out on the four stereoisomers which agree
bonds compare well with those in 3 [d(B1–P1) = 1.941(2), d(B2–
1
˚
P2) = 1.956(2) A].
in their symmetry with the 31P{ H} NMR spectra, both for 3
Multinuclear NMR spectroscopic investigations were carried
and the parent compound 2. Full geometry optimisations were
performed on pre-optimised AM1 geometries at the DFT-B3LYP
level with the 6-311G* basis set. Except for minor deviations
of the bond lengths, the geometric parameters of the optimised
structures of (RP*,SP*,SP*,RP*)-( )-3 and (RP*,SP*,SP*,RP*)-( )-
2 correspond well to their solid-state structures (for details see
also ESI†). Table 2 contains the relative energies of the different
isomers calculated at the B3LYP level and for comparison at
the Hartree–Fock level (using the optimised B3LYP geometries).
While the stereoisomers of 2 adhere to the expected trend by
becoming increasingly energetically disfavoured with increasing
out on 3 and 4 in solution. Interestingly, the two AAꢀBBꢀ spin
1
systems observed in the 31P{ H} NMR spectrum of 3 indicate
the presence of two symmetric species in a ratio of approximately
1
1 : 2.5. This situation is also reflected in the H NMR spectrum
by two multiplets at 2.80 und 3.02 ppm and by signals in the
1
13C{ H} NMR spectrum at 21.69 and 27.11 ppm corresponding
to two distinct methylene groups. This indicates the presence of
two separate species in solution. Most likely, these are different
stereoisomers of 3. Of the four possible diastereomers which
are in agreement with the symmetry of the 31P NMR spectrum
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 831–837 | 833
©

View Article Online
Table 2 Relative energy differences DE/kJ mol−1 between the stereoiso-
mers of cyclo-(P4Ph4CH2) (2) and cyclo-1,4-(BH3)2(P4Ph4CH2) (3) (see
Fig. 2) as determined by Hartree–Fock (HF) and DFT calculations (basis
set: 6-311G*)
different local environments in the crystal. A low-temperature
NMR experiment gives additional credence to the suggested
1
mechanism. Thus, the low-temperature 31P{ H} NMR spectrum
of a freshly prepared sample of 3 showed only the signals of
(RP*,SP*,SP*,RP*)-( )-3. Only after the sample had been warmed
to room temperature was the u◦sual picture again observed: the
2
3
Isomer
DEB3LYP
DEHF
DEB3LYP
DEHF
1
31P{ H} NMR spectrum at −80 C showed resolved multiplets of
cis-cis-cis
51.02
19.25
13.55
0
58.33
18.66
11.52
0
57.98
5.88
27.79
0
71.85
8.53
32.55
0
both isomers in the usual ratio of approximately 2.5 : 1 (see ESI†
for details).
cis-trans-cis
trans-cis-trans
trans-trans-trans
The underlying reason for the small energy gap between the
cis-trans-cis isomer and the sterically favoured all-trans isomer
of 3 may be that the repulsion between the phosphorus lone
pairs is diminished by the coordinating BH3 moieties. Preliminary
DFT calculations on the related compound 2·2S reveal that the
energies of its symmetric stereoisomers follow the same order
as in 3. However, the energy difference between the all-trans
isomer (RP*,SP*,SP*,RP*)-( )-2·2S and the cis-trans-cis isomer
(RP*,RP*,RP*,RP*)-( )-2·2S is significantly larger than in 3.17 This
may explain why a second isomer of 2·2S was not observed by
Baudler and co-workers in solution.7
number of phenyl groups in a cis orientation, the same order is not
observed for 3. In this case, the all-trans isomer (RP*,SP*,SP*,RP*)-
( )-2 is still lowest in energy but, depending on the method
employed, the cis-trans-cis isomer (RP*,RP*,RP*,RP*)-( )-3 is only
6–8 kJ mol−1 higher in energy. Significantly, the energy difference
to the trans-cis-trans isomer (SP*,RP*,SP*,RP*)-3 is several times
larger. Hence, it appears that the second isomer in solution is
indeed (RP*,RP*,RP*,RP*)-( )-3. The presence of the all-trans
and the cis-trans-cis isomers in solution also becomes plausible
if one considers that both species differ only in the relative
configurations of the uncoordinated b-phosphorus atoms. The
low inversion barrier of the b-phosphorus atoms would effect the
room-temperature interconversion between the two diastereomers.
In fact, it is well know that the inversion barriers of catenated
phosphorus atoms are often significantly reduced compared to
ordinary tertiary phosphanes.15
Thus, in solution an equilibrium is assumed between the
major component (RP*,SP*,SP*,RP*)-( )-3 and the minor
stereoisomer (RP*,RP*,RP*,RP*)-( )-3. The overwhelming part
of the minor isomer is then converted to the all-trans form
(RP*,SP*,SP*,RP*)-( )-3 on crystallisation in a second-order
asymmetric transformation16 (Scheme 2). The 31P MAS NMR
spectrum of the bulk solid confirms the presence of only one isomer
(see Experimental section and ESI†) with three resonances in the
approximate ratio of 2 : 1 : 1. While the low-field resonance is
attributed to the P atoms coordinated to boron, the two peaks at
high field are assigned to the b-phosphorus atoms, which have
Although the solution behaviour of 3 can certainly not be
1
regarded as trivial, the 31P{ H} NMR spectrum of 4 reveals a
much more complicated situation (Fig. 5). Numerous multiplets
are observed in the regions 0 to +25 ppm and −50 to −20 ppm.
1
Note that the 31P{ H} NMR spectrum of 4 is reproducible, and
recrystallisation of the product does not lead to a significant
alteration of the spectrum. Again, the relative deshielding and
the line broadening caused by coupling with the 11B nuclei
indicates that the low-field resonances can be attributed to the P
atoms coordinated to BH3. Interestingly, integration of inversely
1
decoupled 31P{ H} NMR gives a ratio of 1 : 2.1 between the
low-field and the high-field signals, whereas a ratio of 2 : 3
would be expected for 4. Unfortunately, individual spin systems
could not be identified due to severe signal overlap. Therefore, a
1
1
highly resolved 31P{ H} 31P{ H} COSY spectrum was recorded.
Although it would be inappropriate to identify individual spin
systems solely on the basis of this spectrum, it is interesting to
note the presence of only two rather weak cross-peaks in the low-
field region of the spectrum (corresponding to the phosphorus
atoms with coordinated BH3). The observation of seven diagonal
peaks between 0 and +25 ppm therefore seems to indicate the
presence of at least six distinct species in solution (see ESI†).
Obviously, complex isomerisation processes are taking place in
1
Fig. 5 31P{ H} NMR spectrum of cyclo-1,2-(BH3)2(P5Ph5) (4) (CDCl3,
Scheme
formation.
2 Isomerisation of 3 by second-order asymmetric trans-
161.98 MHz).
834 | Dalton Trans., 2006, 831–837
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
solution. These may possibly result in different configurations,
constitutions (different positions of the BH3 groups in the ring),
different ring sizes (as observed for the parent compound 5, which
is in equilibrium with the cyclotetra- and cyclohexaphosphane) or
even loss of BH3 moieties.
indicates that the respective P atoms possess a relatively high
nucleophilicity. In solution, 3 forms a mixture of two diastereomers
related by inversion of the uncoordinated phosphorus atoms,
while a complex mixture of isomers is observed in the case of 4.
Attempts to deprotonate 3 so far have failed to generate the desired
tetraphosphacyclopentanide anion cyclo-1,4-(BH3)2(P4Ph4CH)−
in a clean manner.
The deprotonation of 3 was systematically studied with a
variety of bases [e.g. nBuLi, lithium diisopropylamide (LDA),
Na{N(SiMe3)2}] with the aim of converting it to a tetraphos-
phacyclopentanide anion (Scheme 1). In most cases, intractable
mixtures of products were obtained from these reactions. However,
Experimental
t
the heterogeneous reaction of 3 with BuLi in n-pentane initially
General methods
appeared to be quite promising. An orange, extremely air-sensitive
powder was isolated which showed four symmetric multiplets in
All procedures were performed under an inert atmosphere of pure
nitrogen or argon with rigorous exclusion of air and moisture. The
solvents were dried by distillation over sodium/benzophenone or
LiAlH4 and saturated with argon prior to use. cyclo-(P4Ph4CH2)
(2)7,18 and cyclo-(P5Ph5) (5)19 were prepared by literature methods.
All other reagents were obtained from commercial sources.
The NMR spectra were recorded with a Bruker AVANCE
DRX 400 spectrometer (1H NMR, 400.13 MHz, internal standard
solvent, external standard TMS; 13C NMR: 100.16 MHz, internal
standard solvent, external standard TMS; 31P NMR: 161.9 MHz,
external standard 85% H3PO4; 11B NMR: 128.4 MHz, external
1
the 31P{ H} NMR spectrum, corresponding to two symmetric
AAꢀBBꢀ spin systems in the expected ratio of approximately 2 :
1 (Fig. 6). Such a spectrum would indeed be expected for the
target compound cyclo-1,4-(BH3)2(P4Ph4CHLi) (3·Li) provided
that inversion of the b-phosphorus atoms is slow on the 31P
NMR timescale and rapid inversion of the anionic ring carbon
atom occurs [as observed in the related compound [Li(thf)3{cyclo-
(P4 Bu4CH)}] (1)].3 However, in subsequent experiments it was
t
found that the synthesis of 3·Li was only poorly reproducible,
1
additional signals in the 31P{ H} NMR spectrum of the product in-
7
standard BF3(OEt2); Li NMR: 155.50 MHz, external standard
dicating the presence of significant impurities. Moreover, attempts
to purify the compound by crystallisation failed to give pure 3·Li.
To confirm its identity, a relatively pure sample of 3·Li was treated
with benzyl chloride (Scheme 1). After removal of the BH3 groups
1
1 M LiCl in MeOH). The 31P{ H,11B} NMR spectra were
recorded with a Bruker AMX 300 NMR spectrometer (1H NMR:
300.13 MHz; 31P NMR: 121.50 MHz; 11B NMR: 96.29 MHz). The
1
1
31P{ H} 31P{ H} COSY spectrum was recorded at 283.43 MHz
on a Bruker AVANCE 700 MHz spectrometer equipped with a
5 mm tbi probe. The 31P MAS spectrum was recorded on a Bruker
MSL 500 spectrometer at 202.45 MHz. Elemental analyses were
performed by staff members of the Research School of Chemistry
at the ANU, Canberra. The melting points were determined in
sealed capillaries under argon and are uncorrected.
1
with morpholine the 31P{ H} NMR spectrum showed the presence
of an ABCD spin system as major product. This may indicate the
formation of the C-substituted species cyclo-{P4Ph4CH(CH2Ph)}
(2·CH2Ph). However, this compound could also not be isolated in
pure form. Thus, although in principle the deprotonation of 3 is
feasible, it remains of limited synthetic value at present.
Quantum chemical calculations
All calculations were carried out with the Gaussian98 program
package.20 Full geometry optimisation was performed at the DFT-
B3LYP level using the standard triple-f 6-311G* basis set. For
comparison, single-point Hartree–Fock energies were calculated
as well.
Data collection and structure refinement of 3, 4 and 4·CH2Cl2
˚
Data [k(Mo-Ka) = 0.71073 A] were collected with a Siemens CCD
(SMART) diffractometer (for 3, 4) or on a Nonius Kappa CCD
(for 4·CH2Cl2). All observed reflections were used for refinement
of the unit-cell parameters. Empirical absorption corrections
were applied to the data of 3 and 4 (SADABS)21 and analytical
absorption corrections were applied for 4·CH2Cl2 (NUMABS).22
The structures were solved by direct methods (3, 4: SHELXTL
PLUS,23 4·CH2Cl2: SIR9224). Refinement was on F2 for 3 and 4
(SHELXTL23) and on F for 4·CH2Cl2 (CRYSTALS25). P, B and
C atoms were refined anisotropically; some C atoms of disordered
parts (P atoms, phenyl groups in 3) were refined isotropically; H
atoms were located by difference maps and refined isotropically.
Table 1 lists crystallographic details. The thermal ellipsoids of the
molecular structures in Fig. 1–4 are shown at the 30% probability
level.
1
Fig. 6 Section of the 31P{ H} NMR spectrum (C7D8–THF) of the pro-
duct isolated from the reaction of tBuLi with cyclo-1,4-(BH3)2(P4Ph4CH2)
(3).
Conclusions
The synthesis of the borane complexes cyclo-1,4-(BH3)2-
(P4Ph4CH2) (3) and cyclo-1,2-(BH3)2(P5Ph5) (4) demonstrates that
the cyclooligophosphanes 2 and 5 readily form 2 : 1 adducts
with BH3, whereas the formation of adducts with a higher ratio
of BH3 per molecule appears to be disfavoured. The position
of the BH3 moieties in the cyclophosphane rings of 3 and 4
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 831–837 | 835
©

View Article Online
CCDC reference numbers 280235 (3), 280236 (4) and 280237
(4·CH2Cl2).
3 days. After the morpholine had been removed in vacuo 25 ml
of degassed MeOH–H2O (1 : 3) was added, and the mixture
heated briefly to reflux and stirred for two days. Subsequently,
the mixture was filtered, dissolved in CH2Cl2 and the solution
dried with molecular sieves. The solvent was removed completely
and the colourless residue extracted into 10 ml of diethyl ether.
A colourless powder was obtained on storage of the solution
at −20 ◦C for several weeks. This was isolated and contained
2·CH2Ph as a major product according to 31P NMR spectroscopy.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b511833f
Synthesis of 3
1.0 ml (10 mmol) of BH3(SMe2) was added to a solution of 1.92 g
(4.3 mmol) of 2 in 40 ml of toluene. The solution was heated
to reflux briefly and stirred overnight. 120 ml of n-hexane was
added very quickly and the slightly turbid mixture was stored
at room temperature. Large, colourless crystals formed within a
Attempts to purify the crude product by recrystallisation failed.
1
31P{ H} NMR (161.9 MHz, C6D6), ABCD spin system: dA
=
+34.81(1) (m), dB = +29.15(1) (m), dC = −2.65(1) (m), dD = −8.22
few days, which were isolated and dried in vacuo. Yield: 1.22–
(m), 1JAC
Hz, 2JAD
=
=
232.2(2) Hz, 1JBD
13.5(2) Hz, 2JAB
=
250.8(2) Hz, 1JCD
19.1(2) Hz, 2JBC
=
260.4(2)
17.8(2) Hz.
◦
1
31
1.73 g (60–84%). Melting point: 153–155 C. H{ P,11B} NMR
(300.13 MHz, C6D6), d 1.76 [s, BH3 of (RP*,SP*,SP*,RP*)-( )-
3], 2.08 [s, BH3 of (RP*,RP*,RP*,RP*)-( )-3], 2.79 [s, CH2 of
(RP*,SP*,SP*,RP*)-( )-3], 3.01 [s, CH2 of (RP*,RP*,RP*,RP*)-( )-
=
=
Synthesis of 4
1
3], 6.7–8.3 (m, Ph); 13C{ H} NMR (100.16 MHz, C6D6), d 21.69
0.95 ml (9.5 mmol) of BH3(SMe2) was added to a solution of
2.63 g (4.9 mmol) of 5 in 50 ml of toluene. The mixture was
heated to reflux briefly and stirred for 1.5 h. Subsequently, the
solvent was removed completely, and the remaining colourless
solid dissolved in 10 ml of CH2Cl2. 20 ml of n-hexane was added
quickly, and the slightly turbid mixture was stored at −20 ◦C
for several days. Colourless crystals formed and were isolated
and dried in vacuo. The crystals contained approximately one
molecule of CH2Cl2 per formula unit according to 1H NMR
and elemental analysis. Yield: 1.40 g (44% ref. to 4·CH2Cl2 and
the amount of 5 supplied). Melting point: >180 ◦C (decomp.).
1H NMR (400.13 MHz, CDCl3), d 0.4–1.7 (br, 6H, BH3), 6.6–
1
2
[tt, JPC = 18.6 Hz, JPC = 5.4 Hz, CH2 of (RP*,RP*,RP*,RP*)-
1
2
( )-3], 27.11 [tt, JPC = ca. 17.8 Hz, JPC = ca. 3.3 Hz, CH2
of (RP*,SP*,SP*,RP*)-( )-3], 125–135 (several s and m, Ph);
1
31P{ H,11B} NMR (121.50 MHz, C6D6), (RP*,SP*,SP*,RP*)-( )-
1
1
ꢀ
ꢀ
ꢀ ꢀ
3: dA,A = +25.13(1) (m), dB,B = −35.08(1) (m), JAB = JA B
=
−188.5(3) Hz, 1JBB = −146.5(2) Hz, 2JAB = JA B = +35.1(1) Hz,
2
ꢀ
ꢀ
ꢀ
2
ꢀ
ꢀ
JAA = −15.3(2) Hz, (RP*,RP*,RP*,RP*)-( )-3: dA,A = +33.65(1)
(m), dB,B = −38.99(1) (m), 1JAB = JA B = −266.6(2) Hz, 1JBB
=
1
ꢀ
ꢀ
ꢀ
ꢀ
2
2
2
ꢀ
ꢀ
ꢀ
−172.6(2) Hz, JAB = JA B = +1.8(1) Hz, JAA = −32.0(2) Hz;
11B{ H} (96.29 MHz, C6D6), d 35.3 (br s, BH3); 31P MAS NMR
1
(202.45 MHz): d +21.3 (br, 2P, P–BH3), −23.1 [br, 1P, P–P(BH3)],
−39.0 [br, 1P, P–P(BH3)]. EI-MS, m/z (%): 473.9 (0.3) [M]+, 459.9
(0.7) [M − BH3]+, 368.8 (20) [M − 2BH3 − Ph]+, 229.9 (27) [M −
2BH3 − P2Ph2]+, 215.9 (7) [(P2Ph2)]+. Elemental analysis: found:
C 63.35, H 5.95, P 27.00; calc.: C 63.04; H 6.23; P 26.14%.
1
8.2 (m, 25H, Ph); 31P{ H} NMR (161.9 MHz, CDCl3), d 2–23
(overlapping m, P–BH3), −19 to −46 [overlapping m, P–P(BH3)
1
or P–P–P(BH3)]; 11B{ H} (128.4 MHz, CDCl3), +35 (br m, BH3).
Elemental analysis: found: C 56.96; H 5.07; P 24.84. calc.: C 57.02;
H 5.10; P 23.72%.
Synthesis of 3·Li
Acknowledgements
t
1.47 ml (1.7 mmol) of BuLi (1.14 M in n-hexane) was added to
0.79 g (1.7 mmol) of 3 in 40 ml of n-pentane. Then the mixture
was refluxed for 3.5 h. The resulting yellow–orange suspension
was filtered. The residue was dried in vacuo. The orange, very
air-sensitive powder was insoluble in hydrocarbons, but soluble
in THF. It was investigated by NMR spectroscopy. 1H NMR
(400.13 MHz, C7D8–THF), d 0.5–2.2 (br m, BH3), 2.51 (br m, CH),
We thank Dr Eberhard Matern (Karlsruhe University) for record-
ing the 11B-decoupled NMR spectra and valuable discussions.
Dr Winfried Bo¨hlmann is thanked for solid-state NMR mea-
surements, Dr Christina Thiele for recording the 2D 31P NMR
spectrum and Prof. Dr Stefan Berger for valuable discussions and
spectrometer time on the Bruker AVANCE 700 MHz spectrometer
(all University of Leipzig). Dr Alison J. Edwards (ANU, Canberra)
is thanked for assistance with the X-ray measurements of 3.
Support from the DAAD (PPP, DAAD/ARC, project number
313-ARC-XIV-00/37) and the Studienstiftung des Deutschen
Volkes (PhD grant for R. W.) is gratefully acknowledged.
1
6.3–8.5 (m, Ph); 31P{ H} NMR (161.9 MHz, C6D6–THF), d +90.2
ꢀ
ꢀ
[br m, PA,A of (RP*,RP*,RP*,RP*)-( )-3·Li], +39.1 [br m, PA,A of
ꢀ
(RP*,SP*,SP*,RP*)-( )-3·Li], +2.7 [m, PB,B of (RP*,RP*,RP*,RP*)-
ꢀ
( )-3·Li], −32.9 [m, PB,B of (RP*,SP*,SP*,RP*)-( )-3·Li], further
ꢀ
ꢀ
components: d +47.6 (m, PA,A of 2), +19.9 (m, PB,B of 2), −3.8 (m,
5), −48.2 [s, cyclo-(P4Ph4)], −62.1 (br s, unidentified species); 7Li
NMR (155.50 MHz, C7D8–THF), d 0.02 (s).
References
1 A. Schisler, PhD Dissertation, Universita¨t Leipzig, 2003.
Synthesis of 2·CH2Ph
2 (a) A. Schisler, U. Huniar, P. Lo¨nnecke, R. Ahlrichs and E. Hey-
Hawkins, Angew. Chem., 2001, 113, 4345; A. Schisler, U. Huniar,
P. L o¨nnecke, R. Ahlrichs and E. Hey-Hawkins, Angew. Chem., Int.
Ed., 2001, 40, 4217; (b) A. Schisler, P. Lo¨nnecke, T. Gelbrich and E.
Hey-Hawkins, Dalton Trans., 2004, 2895; (c) A. Schisler, P. Lo¨nnecke
and E. Hey-Hawkins, Inorg. Chem., 2005, 44, 461; (d) R. Wolf, A.
Schisler, P. Lo¨nnecke, C. Jones and E. Hey-Hawkins, Eur. J. Inorg.
Chem., 2004, 3277; (e) A. Schisler, P. Lo¨nnecke, E. Hey-Hawkins,
Chem.–Eur. J., manuscript in preparation.
0.53 ml (4.6 mmol) of benzyl chloride was added to a clear,
t
ice-cooled solution of 3·Li (freshly prepared from 3 and BuLi
according to the above procedure) in 15 ml toluene which
contained a few drops of THF in order to dissolve the lithium
salt. The solution turned pale and slightly turbid on addition
and was stirred overnight. The solvent was removed in vacuo
and 5 ml of morpholine was added. This mixture was stirred for
3 R. Wolf and E. Hey-Hawkins, Chem. Commun., 2004, 2626.
836 | Dalton Trans., 2006, 831–837
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
4 R. Wolf, PhD Dissertation, Universita¨t Leipzig, 2005.
5 A. H. Cowley and R. P. Pinell, Inorg. Chem., 1966, 5, 1463.
6 For example, they play a pivotal role in the synthesis of the well-known
ligand DIPAMP: (a) T. Imamoto, T. Kusumoto, N. Suzuki and K. Sato,
J. Am. Chem. Soc., 1985, 107, 5301; (b) W. S. Knowles, M. J. Sabacky,
B. D. Vineyard and D. J. Weinkauff, J. Am. Chem. Soc., 1975, 97,
2567.
7 M. Baudler, J. Vesper, B. Kloth and H. Sandmann, Z. Anorg. Allg.
Chem., 1977, 431, 39.
8 M. Baudler, D. Koch, Th. Vakratsas, E. Tolls and K. Kipker, Z. Anorg.
Allg. Chem., 1974, 408, 225.
(RP*,RP*,SP*,SP*)-2·2S: +65.3 kJ mol−1;M. Finger, unpublished
results.
18 Baudler, J. Vesper, P. Junkes and H. Sandmann, Angew. Chem., 1971,
83, 1019; Baudler, J. Vesper, P. Junkes and H. Sandmann, Angew. Chem.,
Int. Ed. Engl., 1971, 10, 940.
19 W. A. Henderson, M. Epstein and F. S. Seichter, J. Am. Chem. Soc.,
1963, 85, 2462.
20 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E.
Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels,
K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V.
Ortiz, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi,
R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W.
Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M.
Head-Gordon, E. S. Replogle and J. A. Pople, Gaussian 98, Revision
A.11.3, Gaussian, Inc., Pittsburgh, PA, 1998.
9 (a) N. Burford, C. A. Dyker and A. Decken, Angew. Chem., 2005, 117,
2416; N. Burford, C. A. Dyker and A. Decken, Angew. Chem., Int.
Ed., 2005, 44, 2364; (b) N. Burford, C. A. Dyker, M. Lumsden and A.
Decken, Angew. Chem., 2005, 117, 6352; N. Burford, C. A. Dyker, M.
Lumsden and A. Decken, Angew. Chem., Int. Ed., 2005, 44, 6196.
10 In accordance with IUPAC and current Chemical Abstracts indexing
practice, the diastereomers are designated by their relative configura-
tional descriptors; enantiomers have been given simplified descriptors.
11 A CCD search (November 2003) gave an average P–B distance of
˚
1.918 A for 60 adducts of the type BH3(PR3).
21 G. M. Sheldrick, SADABS: a Program for Empirical Absorption
Correction, Go¨ttingen, 1998.
12 R. Blom and A. Haaland, J. Mol. Struct., 1985, 128, 21.
13 J. Lex and M. Baudler, Z. Anorg. Allg. Chem., 1977, 431, 49.
14 J. J. Daly, J. Chem. Soc., 1964, 6147.
22 P. Coppens, in Crystallographic Computing, ed. F. R. Ahmed, S. R. Hall
and C. P. Huber, Munksgaard, Copenhagen, 1970, pp. 255–270.
23 (a) SHELXTL PLUS, Siemens Analytical X-Ray Instruments Inc.,
XS: Program for Crystal Structure Solution, XL: Program for Crystal
Structure Determination, XP: Interactive Molecular Graphics, 1990;
(b) G. M. Sheldrick, SHELXL-97, University of Go¨ttingen, 1997.
24 A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M. C.
Burla, G. Polidori and M. Camalli, SIR92 - A Program for Automatic
Solution of Crystal Structures by Direct Methods, J. Appl. Crystallogr.,
1994, 27, 435.
15 For example, tetraorganodiphosphanes (R’RP)2 (Rꢀ = Me, R = Ph) can
not be isolated in diastereomerically pure form: J. B. Lambert, G. F.
Jackson III and D. C. Mueller, J. Am. Chem. Soc., 1970, 92, 3093, and
references therein.
16 J. Jaques, A. Collet and S. H. Wilen, Enantiomers, Racemates and
Resolutions, Wiley-Interscience, New York, 1981, pp. 371–377.
17 Relative energies obtained from full structure optimisations, followed
by single-point calculations of 2·2S (BP86/SV(P)//BP86/TZV(P)):
(RP*,SP*,SP*,RP*)-( )-2·2S:
0
kJ mol−1
,
(RP*,RP*,RP*,RP*)-( )-
25 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and D. J.
Watkins, J. Appl. Crystallogr., 2003, 36, 1487.
2·2S: +15.2 kJ mol−1
,
(RP*,SP*,RP*,SP*)-2·2S: +27.9 kJ mol−1
,
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 831–837 | 837
©