Communications
to the cis isomer and is folded, as expected for a bicyclo-
[1.1.0]butane structure; it was found to be 58.3 kcalmolÀ1
lower in energy than I. More surprising is the geometry of
the trans isomer I’’, which adopts a C2h symmetry with a very
unusual planar PBPB skeleton. The latter resembles diradical
I, but the phosphorus atoms are strongly pyramidalized, and
the phosphorus–phosphorus distance is in the usual range for
À
a P P single bond (2.355 ). Derivative I’’ was calculated to
be 8.9 kcalmolÀ1 higher in energy than the cis isomer I’. T o
complete our study, we also investigated the 1,2-diboryldi-
phosphane II, and found that it was the least stable isomer.
The geometric parameters show no interaction between the
phosphorus lone pairs and the boron vacant orbitals; the
phosphorus atoms are strongly pyramidalized and the phos-
phorus–boron distances long.
Figure 3. Molecular structure of 3 in the solid state (hydrogen atoms
are omitted for clarity). Selected bond lengths [] and angles [8]: P1-P2
2.1467(7), P2-B 2.044(2), B-P1 2.042(2), P1-K 3.3039(7); P1-B-P2
63.39(7), P2-P1-B 58.35(6), P1-P2-B 58.26(6).
As derivatives C/C’ were prepared by valence isomer-
ization of 1,2-diphosphino-1,2-diboranes,[12] and as the buta-
diene valence isomer II was calculated to be higher in energy
than I’/I’’, it seemed reasonable to expect that 1,2-diboryl-1,2-
diphosphanes would similarly undergo an isomerization into
the target compounds (Scheme 1). Aiming at introducing
phosphorus. The addition of dicyclohexyl- and chlorodiphe-
nylborane to 1,2-dipotassium-1,2-di(tert-butyl)diphosphane,
and of chlorodi(tert-butyl)borane to 1,2-dipotassium-1,2-
diphenyldiphosphane,[17] led to 2a, 2b, and 2c, which were
isolated in 59, 60, and 55% yield, respectively, as extremely
air-sensitive but thermally stable crystals (Scheme 3). The
Scheme 1. Valence isomerization as a synthetic strategy.
Scheme 3. Preparation of 1,2-diboryl-1,2-diphosphanes 2a–c, and 2,4-
diborata-1,3-diphosphonio[1.1.0]bicyclobutane 1’. c-Hex=cyclohexyl.
bulky groups around the BPBP core, 1,2-dipotassium-1,2-
di(tert-butyl)diphosphane[15] was chosen as a starting material,
and was treated with two equivalents of chlorodi(tert-
butyl)borane. A clean reaction occurred, but the spectro-
scopic data of the resulting compound 3 revealed the presence
of two equivalent phosphorus and only one boron nucleus
(Scheme 2). The upfield 31P (d = À73 ppm) and 11B NMR
31P NMR (2a: d = À3.6; 2b: À5.4; 2c: À23.7 ppm) and
11B NMR (2a: d = + 77; 2b: + 77; 2c: + 87 ppm) chemical
shifts are found relatively downfield, which suggests the
formation of 1,2-diphosphino-1,2-diboranes. In marked con-
trast with the related derivatives reported by Power et al,[18]
with mesityl groups at boron and a 1-adamantyl or a mesityl
substituent at phosphorus, X-ray diffraction studies showed
that compounds 2a–c do not have butadiene-like structures
(Figure 4).
Interestingly, along the series 2a–2b–2c, there is a length-
ening of phosphorus–boron distances (1.89–1.90–1.93 ), and
a pyramidalization of the phosphorus centers (ꢀoP = 341–335–
3288), which indicate decreasing interactions between the
phosphorus and boron in the a position. Consequently, the
phosphorus and boron centers should be more nucleophilic
and electrophilic, respectively, which should favor the desired
1,3-interactions and thus the formation of compounds of types
I/I’/I’’. However, all attempts to thermally (benzene, reflux,
12 h) and photochemically (254 nm) induce the rearrange-
ment of these compounds failed. The size of the boron
substituents was decreased further, and when two equivalents
of chlorodicyclohexylborane were added to 1,2-dipotassium-
1,2-diphenyldiphosphane, a new compound 1’ was isolated in
54% yield (Scheme 3). The 2,4-diborata-1,3-diphosphonio-
Scheme 2. Reaction of a bulky chloroborane with a sterically hindered
diphosphide leading to 3.
(d = + 14 ppm) chemical shifts were consistent with a three-
membered-ring structure with a tetracoordinated boron
atom. A single-crystal X-ray diffraction study[16] of compound
3 confirmed our hypothesis; noteworthy is the extremely long
phosphorus–boron distance (2.04 ) (Figure 3).
Even under forcing conditions, we have not been able to
introduce a second di(tert-butyl)boryl group to 3. As the
excessive steric bulk seemed to be the obstacle, we gradually
decreased the size of the substituents at boron, then at
156
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 155 –159