Synthesis, characterization and thermolysis of phosphinite-borane adducts: Investigation of an unusual thermally-induced phenol elimination reaction

Article abstract of DOI:10.1039/b202361j

The synthesis, characterization and thermolytic behavior of a new class of phosphinite-borane adducts containing the bulky 2,4,6-t-Bu₃C₆H₂O (Mes*O) group is reported, including the description of the first example of a primary phosphinite-borane adduct isolable at room temperature (Mes*O)PH₂·BH₃ 1. Adduct 1 and the phenylated derivative (Mes*O)PhPH·BH₃ 3 formed highly cross-linked solids, 7 and 8, when heated to 100 and 140 °C, respectively. Characterization of these solids by solid state and solution NMR spectroscopy suggest that both dehydrocoupling (H₂ elimination) and Mes*OH elimination occur simultaneously to give the complex products [(Mes*O)_xRPH_y-BH_{(2x + 2y + z)}]_n (R = H or Ph, x + y + z = 1). The tertiary adduct (Mes*O)Ph₂P·BH₃ 5 and the phosphite-borane (PhO)₃P·BH₃ 6 were found to decompose at high temperatures via P-B bond cleavage without formation of the corresponding phenol. These results suggest that the thermally-induced formation of Mes*OH from 1 and 3 occurs via α-elimination from a phosphorus center. Adducts 1, 3, 5 and 6 were also structurally characterized by single-crystal X-ray diffraction.

Full text of DOI:10.1039/b202361j

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, characterization and thermolysis of phosphinite–borane
adducts: investigation of an unusual thermally-induced phenol
elimination reaction
Eric Rivard, Alan J. Lough and Ian Manners*
Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario,
M5S 3H6, Canada. E-mail: imanners@chem.utoronto.ca
Received 6th March 2002, Accepted 30th May 2002
First published as an Advance Article on the web 9th July 2002
The synthesis, characterization and thermolytic behavior of a new class of phosphinite–borane adducts containing
the bulky 2,4,6-t-Bu3C6H2O (Mes*O) group is reported, including the description of the first example of a primary
phosphinite–borane adduct isolable at room temperature (Mes*O)PH2ؒBH3 1. Adduct 1 and the phenylated
derivative (Mes*O)PhPHؒBH3 3 formed highly cross-linked solids, 7 and 8, when heated to 100 and 140 ЊC,
respectively. Characterization of these solids by solid state and solution NMR spectroscopy suggest that both
dehydrocoupling (H2 elimination) and Mes*OH elimination occur simultaneously to give the complex products
[(Mes*O)xRPHy–BH(2x ϩ 2y ϩ z)]n (R = H or Ph, x ϩ y ϩ z = 1). The tertiary adduct (Mes*O)Ph2PؒBH3 5 and the
phosphite–borane (PhO)3PؒBH3 6 were found to decompose at high temperatures via P–B bond cleavage without
formation of the corresponding phenol. These results suggest that the thermally-induced formation of Mes*OH from
1 and 3 occurs via α-elimination from a phosphorus center. Adducts 1, 3, 5 and 6 were also structurally characterized
by single-crystal X-ray diffraction.
Introduction
Results and discussion
The development of new bond-forming processes continues to
be an important challenge in main group chemistry.1 Significant
advances in this area have recently occurred as a result of
the discovery of various transition metal-catalyzed dehydro-
coupling methodologies.1 For example, catalytic formation of
homonuclear Si–Si,2 Ge–Ge,3 Sn–Sn4 and P–P bonds,5 and
heteronuclear Sn–Te,6 Si–P,7 Si–N8 and Si–O9 bonds have been
reported.
Before this work was initiated, to our knowledge, no examples
of stable primary phosphinite–borane adducts, (RO)PH2ؒBH3,
had been described. For example, Centofanti reported that the
phosphonite–borane adduct, (MeO)2PHؒBH3 is stable to 40 ЊC,
while the primary phosphinite–borane adduct, (MeO)PH2ؒ
BH3, decomposed spontaneously at room temperature to give
an insoluble yellow solid, H2 and PH3ؒBH3.16 More recently,
Roper and co-workers have shown that a variety of unstable
alkyl phosphinites [e.g. (RO)PH2, R = Me, Et] can be stabil-
ized by forming coordination complexes with late transition
metals.17 As a consequence, we decided to explore the synthesis
and potential dehydrocoupling chemistry of a new class of
bulky phosphinite–borane adducts containing the bulky 2,4,6-
t-Bu3C6H2O (Mes*O) group.
Recently, we discovered that a range of species with B–N
and B–P skeletons are readily accessible by transition metal-
mediated dehydrocoupling of appropriate amine– or phosphine–
borane adducts. For example, alkyl-substituted secondary
amine–boranes adducts, R2NHؒBH3, dehydrocouple efficiently
under mild conditions (< 45 ЊC) in the presence of RhI catalysts
to give four-membered amino–borane rings [R2N–BH2]2,
whereas the primary amine–boranes RNH2ؒBH3 eliminate two
equivalents of H2 to form borazines [(RNH–BH)3, R = H, Me,
Ph].10 In addition, primary and secondary phosphine–borane
adducts participate in dehydrocoupling chemistry, however
higher temperatures are generally required (> 90 ЊC).11 Second-
ary phosphine–boranes such as Ph2PHؒBH3 dehydrocouple in
the presence of RhI catalysts to give mixtures of six- and eight-
membered cyclic phosphinoboranes [Ph2P–BH2]x (x = 3 or 4) at
high temperatures, and the unusual linear oligomer R2PH–
BH2–PR2–BH3 when milder reaction conditions are employed.
The less sterically congested primary phosphine–boranes
dehydrocouple cleanly to give high molecular weight poly-
phosphinoboranes [RPH–BH2]n, an interesting class of stable
inorganic polymers.11–15
Synthesis and characterization of (Mes*O)PH2ؒBH3 1
The reaction of Mes*OPCl2, synthesized from Mes*OLiؒOEt2
and PCl3 in Et2O,18 with 2 equivalents of Li[BH4] in THF
produced the novel primary phosphinite–borane adduct 1
(quantitative yield) as a white solid which is stable at room
temperature.
The 31P{1H} NMR spectrum of 1 consisted of a broad
singlet at δ 78.9 which split into a 1 : 2 : 1 triplet in the proton-
coupled spectrum (1JPH ≈ 390 Hz), which indicated the presence
of two hydrogen atoms bound directly to a phosphorus center.
1
Furthermore, the H NMR spectrum of 1 showed a doublet
of quartets (δ 6.73) assigned to coupling between the above
mentioned PH2 protons with a neighboring phosphorus atom
(doublet, 1JPH = 399 Hz) and BH3 protons (1 : 3 : 3 : 1 quartet,
3JHH = 6.6 Hz). The observation of well-resolved coupling
between P–H and B–H protons was somewhat surprising given
the quadrupolar nature of the 11B nucleus, however similar
splitting patterns have been previously observed in the 1H NMR
spectra of PH3ؒBH3 19 and the thermally unstable phosphinite–
borane adduct, (MeO)PH2ؒBH3.16 Signals arising from the
BH3 group were also detected as a broad halo (δ 0.5–1.5) in
the 1H NMR spectrum, and as a broad singlet (δ Ϫ41.9) in the
As part of our continuing research concerning the polymer-
ization of phosphorus–boron precursors, we have now explored
the chemistry of the related phosphinite–borane adducts,
(RO)RЈPHؒBH3, with the aim of synthesizing new rings, chains
or macromolecules of the form [(RO)RЈP–BH2]n using dehydro-
coupling strategies. This paper reports our investigations in this
area, and the discovery of a rare bond-forming process via the
α-elimination of phenols.
2966
J. Chem. Soc., Dalton Trans., 2002, 2966–2972
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b202361j

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for adducts 1, 3, 5
and 6 with estimated standard deviations in parentheses
1
3
5
6
P–B
P–O
1.882(6)
1.597(3)
1.894(3)
1.6117(15)
1.917(5)
1.629(3)
1.878(3)
1.5832(15)
1.5845(15)
1.5757(14)
P–C(phenyl)
P–H
1.799(2)
1.804(4)
1.807(3)
1.33(4)
1.40(4)
1.05(4)
1.07(5)
1.12(4)
1.391(10)
B–H
1.06(3)
1.17(4)
1.18(4)
1.10(4)
1.12(4)
1.27(4)
1.08(2)
1.10(2)
1.11(3)
O–P–B
110.5(2)
111.89(12)
115.2(2)
110.32(12)
117.56(11)
118.32(11)
Fig. 1 Molecular structure of 1 with thermal ellipsoids at the 30%
H–P–H
H–P–C(phenyl)
H–P–B
100(2)
probability level (carbon bound hydrogen atoms omitted for clarity).
95.5(9)
100.0(9)
114.0(14)
116.9(14)
106.5(13)
108.2(14)
and refinement of the hydrogen atoms at phosphorus and
boron. The geometry about the phosphorus and boron atoms
in 1 can be described as distorted tetrahedral with bond angles
ranging from the acute 100(2)Њ for the H–P–H angle to the
wider H–P–B angle of 116.9(14)Њ. The B–P bond length in 1
was determined to be 1.882(6) Å and is significantly shorter
than the B–P bonds observed in most structurally characterized
phosphine–borane adducts [e.g. B–P bond length in PhPH2ؒ
BH3 is 1.924(4) Å12]. The P–H [1.37(6) Å avg.] and B–H [1.07(7)
Å avg.] bond lengths are similar to those observed in other
phosphorus–borane adducts. As both hydridic (B–HδϪ) and
acidic (P–Hδϩ) hydrogen atoms are present in 1, there exists the
possibility of intermolecular hydrogen bonding. In fact, such
H–H interactions have been observed in PhPH2ؒBH3 with the
shortest intermolecular H–H distance (2.38 Å) being similar to
the sum of the van der Waals radii for P and B (2.4 Å).12 By
contrast, the closest intermolecular H–H contact in 1 was
found to be > 3.2 Å, despite the more acidic nature of the P–H
protons (vide infra) when compared to PhPH2ؒBH3. It is pos-
sible that the greater steric bulk in 1 hinders intermolecular
H-bonding and thus only discrete monomers are observed in
the solid state structure of 1.
H–P–O
C–P–B
122.0(9)
116.81(13)
115.7(2)
111.5(2)
11B NMR spectrum. This 11B NMR chemical shift is consistent
with the presence of a phosphorus()–borane adduct as typical
chemical shifts for these adducts are in the range of δ Ϫ40 5.20
All relevant signals associated with the pendant Mes*O group
were located in both the H and 13C{1H} NMR spectra of 1.
1
Further characterization of 1 was provided by melting point
determination, mass spectrometry and a single crystal X-ray
diffraction study (discussed below).
Before this study, only a few reports of structurally
characterized phosphinite–borane adducts could be found in
the literature,21 and of these, none involved species with P–H
bonds. In addition 1 represents, to our knowledge, the first
primary phosphinite–borane adduct which is isolable as a stable
compound at room temperature. To further characterize 1, a
single-crystal X-ray diffraction study was undertaken on small
colorless crystals of 1 grown from toluene at Ϫ30 ЊC. Data
collection and refinement parameters for 1 are summarized in
Table 2 while relevant bond lengths and angles for 1 are listed in
Table 1. The molecular structure of 1 is depicted in Fig. 1.
Despite some disorder in the para-t-Bu groups within the
Mes*O moiety, the quality of the data permitted the location
Synthesis and characterization of (Mes*O)PhPCl 2 and
(Mes*O)PhPHؒBH3 3
The previously reported synthesis of (Mes*O)PhPCl 2 by
Pastor et al. involved the reaction of Mes*OH with PhPCl2 in
Table 2 Crystal data, data and intensity collection parameters, and least squares parameters for 1, 3, 5 and 6
1
3
5
6
Empirical formula
Formula weight
C18H34BOP
308.23
C24H38BOP
384.32
C30H42BOP
460.42
C18H18BO3P
324.10
Crystal system
Space group
Monoclinic
P21/c
Monoclinic
P21/c
Orthorhombic
Pca21
Triclinic
P1
¯
a/Å
b/Å
c/Å
α/Њ
9.9490(12)
15.6210(19)
12.7660(11)
90
101.303(7)
90
1945.5(4)
4
0.139
150(1)
2.61 to 22.50
13449
2533/0.162
224
0.0654
18.0012(4)
8.0595(2)
18.4874(4)
90
98.7360(10)
90
2651.05(11)
4
0.113
150(1)
2.66 to 27.50
23956
6062/0.067
259
0.0620
18.2421(6)
12.3496(6)
12.3712(10)
90
90
90
2787.0(3)
4
0.118
150(1)
2.58 to 25.03
14945
9.4189(3)
9.9295(4)
9.9979(3)
76.568(2)
84.550(2)
72.735(2)
868.16(5)
2
0.169
150(1)
2.69 to 25.03
6801
3054/0.039
221
β/Њ
γ/Њ
V/Å3
Z
µ(Mo-Kα)/mmϪ1
Temperature/K
θ Range/Њ
No. of reflections collcected
No. of independent reflections/Rint
No. of refined parameters
R (I > 2σ(I )a
4606/0.099
321
0.0592
0.0410
0.0925
b
Rw
0.01548
0.01651
0.01035
2
a R = Σ||Fo| Ϫ |Fc||/Σ|Fo|. b Rw = {Σ[w(Fo2 Ϫ Fc2)2]/Σ[w(Fo )2]}1/2
.
J. Chem. Soc., Dalton Trans., 2002, 2966–2972
2967

View Article Online
the presence of KOH.22 A yield of 68% was reported, however
little characterization (IR and melting point) was provided. We
prepared pure 2 in an improved yield of 89% by reacting the
lithium salt Mes*OLiؒOEt2 with PhPCl2 in diethyl ether; a
similar reaction between Mes*OLiؒOEt2 and PCl3 was used by
Power and co-workers to synthesize Mes*OPCl2.18 In addition,
compound 2 was further characterized by 1H, 13C{1H} and
31P{1H} NMR spectroscopy, and high-resolution mass spectro-
metry (HR-MS) which were consistent with the assigned struc-
ture. Reaction of 2 with one equivalent of Li[BH4] gave the
phenyl substituted analogue of 1, (Mes*O)PhPHؒBH3 3, in
good yield (72%, Scheme 1) as a white solid that was character-
Fig. 2 Molecular structure of 3 with thermal ellipsoids at the 30%
probability level (carbon bound hydrogen atoms omitted for clarity).
Compound 5 was characterized by NMR spectroscopy, mass
spectrometry, melting point, elemental analysis and a single-
crystal X-ray diffraction study.
The formation of 5 was monitored by 31P{1H} NMR
spectroscopy which showed the complete conversion of 4
(δ 127) to a new product with a chemical shift of 110 ppm. 1H,
13C{1H} and 11B{1H} NMR spectroscopy also confirmed the
presence of 5. To better understand how the B–P bonding in
this class of phosphinite–borane adducts relies upon the steric
bulk at phosphorus, we also examined the structure of this
compound by single-crystal X-ray diffraction. The molecular
structure of 5 can be found in Fig. 3, while relevant data collec-
tion and refinement parameters are found in Table 2. Table 1
contains selected bond lengths and angles for 5.
Scheme 1 Synthesis of adducts 1, 3, and 5.
ized by NMR spectroscopy, mass spectrometry, melting point,
elemental analysis and a single-crystal X-ray diffraction study.
The 31P{1H} NMR spectrum of 3 gave a broad signal at
δ 99.1 that split into a doublet (1JPH ≈ 410 Hz) in the proton-
coupled spectrum, which confirmed the presence of a PH
group. The associated doublet of quartets was observed in the
3
1H NMR spectrum at δ 7.51 (1JPH = 403 Hz; JHH = 5.4 Hz,
coupling of PH to BH3). This resonance was significantly
downfield from the PH resonance in Ph2PHؒBH3 (δ 5.51),23
which suggests a greater acidity for the PH proton in 3 (due to
the increased electronegativity of the neighboring aryloxy
1
groups in comparison to their aryl counterparts). The H and
13C{1H} NMR spectra of 3 were similar to those observed for 1
except for the presence of added resonances due to the phenyl
group; the 11B{1H} NMR spectrum of 3 consisted of a doublet
at δ Ϫ40.7 (1JBP = 47 Hz).
Chiral (Mes*O)PhPHؒBH3
3 crystallized from toluene
(Ϫ30 ЊC) in the achiral space group P21/c with both enantiomers
of 3 present in the crystal lattice. Fig. 2 depicts the molecular
structure of one of these enantiomers (S) while Table 2 con-
tains relevant collection and refinement parameters for 3;
selected bond lengths and angles for 3 can be found in Table 1.
The geometry about the phosphorus and boron atoms in 3 is
similar to that in 1 (distorted tetrahedral), however the O–P–B
angle is somewhat wider [111.89(12)Њ] in 3 than in 1 [110.5(2)Њ].
The B–P bond length in 3 is similar [1.894(3) Å] to the
B–P bond lengths in related substituted phosphinite–borane
adducts.21 As with 1, no significant intermolecular interactions
could be observed in 3. The pendant Mes*O and Ph groups in 3
showed no unusual features in the solved structure with typical
C–C and C–O bond lengths and angles observed.
Fig. 3 Molecular structure of 5 with thermal ellipsoids at the 30%
probability level (carbon bound hydrogen atoms omitted for clarity).
The molecular structure of 5 revealed a highly crowded
phosphorus center. A much wider O–P–B angle is present in 5
[115.2(2)Њ] and the P–O bond is significantly longer [1.629(3) Å]
than in either adducts 1 [1.597(3) Å] or 3 [1.6117(15) Å]. The
B–P bond length [1.917(5) Å] seems less influenced by steric
effects and has a similar value to those in the previously men-
tioned phosphinite–borane adducts. Of note, the bond lengths
and angles within 5 also compare closely to those within the
phosphite–boraneadduct(PhO)3PؒBH3 6(seeFig.4,Tables1and
2) which has a P–B bond length of 1.878(3) Å.
Synthesis and characterization of (Mes*O)Ph2P 4 and
(Mes*O)Ph2PؒBH3 5
The diphenyl substituted phosphinite–borane adduct 5, a color-
less solid, was prepared in good yield (73%) by the reaction of
the previously unknown (Mes*O)Ph2P 4 (synthesized from
Ph2PCl and Mes*OLiؒOEt2) with a slight excess of BH3ؒTHF.
2968
J. Chem. Soc., Dalton Trans., 2002, 2966–2972

View Article Online
detected by IR with B–H stretching modes detected at 2267 and
2384 cmϪ1
.
In order to further characterize 7, solid state NMR
spectroscopy was performed. The 13C CP-MAS NMR spec-
trum of 7 also supported the presence of Mes*O groups as the
relevant aromatic and t-Bu resonances were observed. The
31P CP-MAS NMR spectrum of 7 was highly complex with a
large number of signals between δ 100 and Ϫ200. The low field
signals above δ 60 have similar chemical shift values as 1 (δ 80)
and imply the presence of a similar chemical environment
(namely a Mes*O group bonded to phosphorus). Resonances
observed in the high field region of the spectrum, around
δ Ϫ100, suggest the presence of B–PH2–B environments (e.g. 31
P
chemical shift of [PhPH–BH2]n is Ϫ34.7 ppm).11,12 The signals
which lie intermediate to the above mentioned resonances could
represent P–P (or B–P) environments resulting from cross-
linking reactions, however it is difficult to speculate further
about the exact assignments of these signals as P–P bonded
species exhibit widely varying chemical shifts24,25 [e.g. P4,
δ Ϫ549; (PhP)5, δ Ϫ4]. The 11B CP-MAS NMR spectrum of
7 was relatively straightforward with two clearly identifiable
boron resonances at δ 3 and Ϫ42 respectively, consistent with
four-coordinate boron environments. The signal at δ Ϫ42 is
similar to previously reported 11B NMR chemical shifts belong-
ing to PBH2P environments within polyphosphinoborane
polymers ([PhPH–BH2]n, δ Ϫ34.0; [iBuPH–BH2]n, δ Ϫ36.312),
while the signal at δ 3 suggests the presence of an electro-
negative atom such as oxygen at boron (possibly Mes*O–B).20
Despite the complexity of 7, the thermolysis of 1 remains sig-
nificant as it represents an attempt to understand a potentially
new inorganic bond-forming reaction involving the elimination
of a phenol. All attempts to induce dehydrogenative coupling
of 1 in toluene using RhI and IrI dehydrocoupling catalysts12
only gave similar phenol elimination reactions as described
above with equally complex and insoluble byproducts.
Fig. 4 Molecular structure of 6 with thermal ellipsoids at the 30%
probability level (carbon bound hydrogen atoms omitted for clarity).
Attempted dehydrocoupling of (Mes*O)PH2ؒBH3 1 and
(Mes*O)PhPHؒBH3 3: discovery of an unusual thermally-
induced phenol elimination reaction
In an attempt to evaluate the ability of 1 to participate in
dehydrogenative B–P bond coupling, a sample was heated
above its melting point (92–3 ЊC). Instead of the anticipated
elimination of hydrogen, large white needles formed on the
colder portion of the reaction vessel. 1H NMR and mass spec-
tral analysis surprisingly identified the pure product as the
phenol, 2,4,6-t-Bu3C6H2OH (m/z = 262). After 1 was heated
under nitrogen for a further 3 d the production of the phenol
ceased (∼ 0.9 mole equivalents) and a yellow involatile solid
remained. This residue (7) was found to be completely insoluble
in common organic solvents (including polar solvents such
as DMSO and DMF) therefore further characterization of
this material was accomplished by solid state NMR and IR
spectroscopy, and elemental analysis (C and H).
In light of the surprisingly facile phenol elimination reaction
involving 1, we also investigated the thermal stability of the
analogous phenylated derivative, (Mes*O)PhPHؒBH3 3. We
anticipated that if phenol elimination occurred in preference
to H2 elimination (dehydrocoupling), then the anticipated
product, [PhPH2–BH2]n, could be easily identified as this
polymer has been characterized in detail by our group.11,12
When a sample of 3 was heated to 140 ЊC (mp of 3: 128–9 ЊC)
in a sublimation apparatus, large white needles were observed
on the water-cooled probe. As with 1, these crystals were
identified as Mes*OH by 1H NMR spectroscopy and mass
spectrometry. Prolonged heating (5 d) produced more phenol
(∼ 0.9 mole equivalents, x = 0.1 according to Scheme 2) along
with an involatile glassy yellow solid, 8, at the bottom of the
sublimation apparatus. In this case, the glassy residue was
found to be highly soluble in most organic solvents. Analysis of
8 by 31P{1H} NMR spectroscopy was consistent with the pres-
ence of a highly complex material with a wide range of signals
Based on the quantity of Mes*OH produced when 1 was
heated, and the results from elemental analysis, we formulated
the insoluble product 7 as [{(Mes*O)xRPHy–BH(2x
}
n
ϩ
2y
ϩ
z)
x = 0.1, y ϩ z = 0.9] (see Scheme 2). If Mes*OH elimination
occurred exclusively (x = z = 0; y = 1) the resulting material
[PH2–BH2]n would be completely inorganic. However signifi-
cant amounts of carbon and a deficiency of hydrogen was
detected by elemental analysis, suggesting the formation of a
more complex, cross-linked product resulting from simul-
taneous H2 and Mes*OH elimination [Anal. Calc. for
{(Mes*O)xRPHy–BH(2x ϩ 2y ϩ z)}n, x = 0.085, y = 0, z = 0.915: C,
28.85. H, 7.21. Found: C, 28.75. H, 6.62%]. Moreover, IR spec-
troscopy confirmed the presence of carbon within 7 (in the
form of Mes*O groups) as many bands were observed between
900 and 1200 cmϪ1. The presence of B–H groups was also
1
detected from δ 100 to Ϫ120. The H NMR spectrum of 8
showed the presence of a broadened signal from δ 6 to 8 (phenyl
groups), a series of broad resonances (δ 1.3–1.5) due to the t-Bu
groups of the Mes*O moiety and a broad halo at δ 0.5–1.5 due
to BH2 protons. This product therefore seemed to be equally
Scheme 2 Thermolysis of 1 and 3.
J. Chem. Soc., Dalton Trans., 2002, 2966–2972
2969

View Article Online
complex as 7, and consequently further exploration of this
reaction was not performed.
Conclusion
A new class of phosphinite–borane adducts containing the
bulky 2,4,6-t-Bu3C6H2O (Mes*O) group was prepared and we
describe the structural characterization of the first primary
phosphinite–borane adduct 1 which is stable at ambient temp-
erature. Attempts to induce dehydrocoupling of 1 and the
phenylated analogue 3, resulted in an unexpected phenol elim-
ination reaction to produce Mes*OH along with cross-linked
materials with complex structures (7 and 8). These results
indicate that dehydrocoupling of phosphinite–borane adducts
does not proceed as cleanly as with phosphine–borane species.
Attempts to initiate phenol elimination from the tertiary
adducts 5 and 6 produced exclusively P–B bond cleavage.
These results suggest that phenol formation occurs via an
α-elimination pathway.
Thermolysis of (Mes*O)Ph2PؒBH3 5 and (PhO)3PؒBH3 6 at
elevated temperatures
In the experiments described above, elimination of Mes*OH
was found to be less facile for the more sterically encumbered
phenyl-substituted derivative, 3, (5 d at 140 ЊC, ca. 90% con-
version) compared to the less sterically congested primary
phosphinite–borane adduct 1 (3 d at 100 ЊC, ca. 90% con-
version). To further examine whether phenol elimination was
dependant on steric effects and to determine if the presence of
a P–H bond was required, thermolysis experiments involving
the sterically encumbered diphenyl-substituted phosphinite–
borane adduct 5 and phosphite–borane 6 were also performed.
When a sample of 5 was heated to 110 ЊC (melting point of
1
5: 103–4 ЊC) for one week, no change was detected by H and
Experimental
31P NMR spectroscopy. Increasing the temperature of the
thermolysis to 160 ЊC led to the decomposition of 5, however
only P–B bond cleavage occurred to form the free phosphinite 4
(1H and 31P NMR) and BH3 (by inference), with no trace of
Mes*OH detected by 1H NMR. Again, as with the thermolysis
of 1 and 3, no catalytic effect was observed when 5 was heated
in the presence of a rhodium catalyst, [{Rh(µ-Cl)(1,5-cod)}2].
We also investigated the thermolysis of 6 as Chopard and
co-workers reported the production of one equivalent of
phenol when a sample was heated in triglyme (130–140 ЊC).26
However, our attempts to induce phenol elimination from 6 in
tetraglyme (140 ЊC) gave exclusively P–B bond cleavage with no
phenol formation. We subsequently heated a neat sample of
6 to 140 ЊC for eight days, and again only observed complete
dissociation of the adduct to the free phosphite, (PhO)3P, with-
out formation of phenol. As there are previously reported
examples of α-elimination of HX (where X = electronegative
element/group) from halophosphine precursors, R(X)PH, to
give cyclophosphanes (RP)n,27 it appears that such a mechanism
could be responsible for the elimination of Mes*OH from 1 and
3 (Scheme 3). However in the case of 1 and 3, many more
General procedures
All reactions and manipulations were carried out under an
atmosphere of prepurified nitrogen (Matheson) using either
Schlenk techniques or an inert atmosphere glove box (Innov-
ative Technology). All solvents were dried and distilled using
standard methods prior to use. Solution 1H NMR spectra were
obtained on a Varian Gemini 300 spectrometer (300.1 MHz)
and referenced to protio impurities in the solvent. Solution
31P, 13C{1H} and 11B NMR spectra were also obtained using
a Gemini 300 spectrometer (122.1, 75.5 and 96.6 MHz) and
referenced externally to 85% H3PO4, SiMe4 (TMS) and BF3ؒ
OEt2 respectively in either CDCl3 or D2O (insert). Solid-state
CP-MAS NMR spectra were obtained on a Bruker DSX-400
(13C at 100.6 MHz) or Bruker DSX-200 (11B at 64.2 MHz, 31P at
81.0 MHz) spectrometers and referenced to adamantane (13C,
δ 38.4 vs. TMS), 85% H3PO4 (31P) or NaBH4 (11B, δ Ϫ42.0 vs.
BF3ؒOEt2).29 Solid-state 11B NMR spectra of 7 were also run as
90Њ–τ–180Њ spin-echo experiments, where τ is one rotor period.
Mass spectra were obtained with the use of a VG 70–250S mass
spectrometer using a 70 eV electron impact ionization source.
Melting points (uncorrected) were obtained under a nitrogen
atmosphere in flame sealed capillaries (0.5 mm o.d.). Infrared
spectra were obtained as Nujol mulls between KBr plates with a
Nicolet Magna-IR 550 spectrometer. Elemental analyses were
performed by Quantitative Technologies Inc., Whitehouse, NJ.
PhPCl2 (Aldrich) was vacuum distilled prior to use. Ph2PCl,
BH3 (1.0 M solution in THF), (PhO)3P and [Ir(1,5-cod)2][OTf]
(1,5-cod = cycloocta-1,5-diene) were obtained from Aldrich and
used as received. 2,4,6-t-Bu3C6H2OLiؒOEt2 (Mes*OLiؒOEt2),30
Mes*OPCl2,18 (PhO)3PؒBH3,31 and [{Rh(µ-Cl)(1,5-cod)}2]32
were prepared according to literature methods.
Preparation of (Mes*O)PH2ؒBH3 1. A pale yellow solution
of Mes*OPCl2 (5.01 g, 13.8 mmol) in THF (25 mL) was added
dropwise to a suspension of Li[BH4] (0.61 g, 28 mmol) in
100 mL of THF at 0 ЊC. The reaction was warmed to room
temperature and stirred for 16 h giving a clear, colorless sol-
ution. The volatiles were removed under reduced pressure and
the residue was extracted with hexanes (100 mL) and filtered
through Celite. Removal of the hexanes afforded a white solid
when vacuum dried (4.24 g, 100%). Dissolution of the product
in a minimum amount of toluene (ca. 0.2 g mLϪ1) and cooling
to Ϫ30 ЊC for 3 weeks produced small, colorless crystals
(blocks) suitable for a single crystal X-ray diffraction study.
1H NMR (CDCl3): δ 0.5–1.5 (br, BH3, 3H), 1.36 (s, para t-Bu,
9H), 1.50 (s, ortho t-Bu, 18H), 6.73 (d of q, PH2, 1JPH = 399 Hz,
3JHH = 6.6 Hz, 2H) and 7.33 (s, Ar–H, 2H). 31P NMR (CDCl3):
δ 78.9 (t, PH2, 1JPH = 389 Hz). 11B{1H} NMR (CDCl3): δ Ϫ41.9
(br). 13C{1H} NMR (CDCl3): δ 31.6 (s), 31.8 (s), 32.7 (s), 36.4
(s), 124.8 (d, J = 2 Hz), 140.5 (d, J = 4 Hz), 147.0 (d, J = 1 Hz)
and 151.7 (d, J = 23 Hz). MS 70 eV, EI (m/z, %): 294
Scheme 3 Potential thermolysis pathways for 1 and 3.
decomposition pathways are potentially available as the
phosphorus–boron intermediate [RP–BH3] could undergo
rearrangement reactions to give polyphosphinoboranes (path
1) or simply decompose to form (RP)n and free borane (path 2).
By CP-MAS NMR, it appears that pathways 1 and 2 occur
simultaneously to give insoluble (in the case of 7), cross-linked
products with a complex structure.28
2970
J. Chem. Soc., Dalton Trans., 2002, 2966–2972

View Article Online
(Mϩ Ϫ BH3, 26), 279 (Mϩ Ϫ BH3 Ϫ Me, 40), 57 (t-Buϩ, 100).
mp 92–3 ЊC (dec).
in the 1H NMR spectrum. 31P{1H} NMR (CDCl3): δ 111.0 (s).
11B{1H} NMR (CDCl3): δ Ϫ38.6 (br). 13C{1H} NMR (CDCl3):
δ 31.8 (s), 33.0 (s), 34.6 (s), 36.7 (s), 124.0 (d, J = 2 Hz), 128.5
(d, J = 11 Hz), 132.0 (d, J = 2 Hz), 132.9 (d, J = 11 Hz), 142.4
(d, J = 4 Hz), and 145.7 (d, J = 2 Hz); Ispo C were not detected.
MS 70 eV, EI (m/z, %): 446 (Mϩ Ϫ BH3, 37), 389 (Mϩ Ϫ BH3 Ϫ
t-Bu, 90), 431 (Mϩ Ϫ BH3 Ϫ Me, 5), 355 (Mϩ Ϫ BH3 Ϫ C7H7,
40), 185 (Ph2Pϩ, 100). mp 103–4 ЊC. Anal. Calc. for C30H42BOP
(460.4): C, 78.26. H, 9.19. Found: C, 78.35. H, 8.88%.
Preparation of (Mes*O)PhPCl 2. A clear solution of PhPCl2
(5.0 g, 28 mmol) in diethyl ether (10 mL) was added dropwise
to a cooled (0 ЊC) pale yellow suspension of Mes*OLiؒOEt2
(9.00 g, 26.3 mmol) in 125 mL of diethyl ether. The reaction
mixture was warmed to room temperature and stirred for 16 h
to give a clear yellow solution over a white precipitate. The
volatiles were removed, and the resulting residue was extracted
with hexanes (100 mL) and filtered through Celite. Removal of
the hexanes afforded 2 as a pure white solid when vacuum dried
(9.52 g, 89%). 1H NMR (CDCl3): δ 1.38 (s, para t-Bu, 9H), 1.50
(s, ortho t-Bu, 18 H), 7.35 (s, Ar–H, 2H), 7.5–7.9 (m, Ph–H,
5H). 31P{1H} NMR (CDCl3): δ 177.1 (s). 13C{1H} NMR
(CDCl3): δ 31.8 (s), 33.9 (d, J = 5 Hz), 34.7 (s), 36.7 (s), 124.9 (d,
J = 2.0 Hz), 128.9 (d, J = 6.0 Hz), 130.1 (d, J = 26 Hz), 131.4 (s),
141.5 (d, J = 4 Hz), 141.8 (s), 142.3 (s), and 145.4 (d, J = 2 Hz).
MS 70 eV, EI (m/z, %): 404 (Mϩ, 49), 347 (Mϩ Ϫ t-Bu, 10), 57
(t-Buϩ). HR-MS: Calc.: 404.2044. Found: 404.2036.
Preparation of (PhO)3PؒBH3 6. The adduct was prepared in
quantitative yields by a similar manner as described by Pelter
et al.31 Crystals (colorless plates) suitable for a single-crystal
X-ray diffraction study were grown from a concentrated solution
1
of the adduct in THF (1 g mLϪ1) at Ϫ30 ЊC (2 d). H NMR
1
(CDCl3): δ 0.3–1.2 (q, BH3, JBH = ∼ 100 Hz, 3H), 7.2–7.4 (m,
Ph–H, 15 H). 31P{1H} NMR (CDCl3): δ 108.0 (pseudo-quartet,
1JPB = 90 Hz). 11B{1H} NMR (CDCl3): δ Ϫ42.7 (d, 1JBP = 82 Hz.
13C{1H} NMR (CDCl3): δ 121.1 (d, J = 8 Hz), 126.1 (s)
and 130.1 (s); Ipso C not detected. MS 70 eV, EI (m/z, %): 310
(Mϩ Ϫ BH3, 37), 217 (Mϩ ϪBH3 Ϫ PhOH, 100), 94 (PhOH, 7).
Preparation of (Mes*O)PhPHؒBH3 3. A clear yellow solution
of 2 (6.13 g, 15.1 mmol) in THF (20 mL) was added dropwise to
a cold (0 ЊC) suspension of Li[BH4] (0.33 g, 15 mmol) in 100 mL
of THF. The reaction mixture was warmed to room temper-
ature and stirred for 16 h to give a clear colorless solution. The
volatiles were removed, and the resulting residue was extracted
with hexanes (75 mL) and filtered through Celite. Removal of
the hexanes afforded a white solid when vacuum dried (4.16 g,
72%). Crystals of 3 (large colorless plates) suitable for a single-
crystal X-ray diffraction study were obtained by cooling a sol-
ution in toluene (0.3 g mLϪ1) to Ϫ30 ЊC for one week. 1H NMR
(CDCl3): δ 1.31 (s, ortho t-Bu, 18H), 1.34 (s, para t-Bu, 9H), 7.22
(s, Ar–H, 2H), 7.51 (d of q, PH, 1JPH = 403 Hz, 3JHH = 5.4 Hz,
1H), 7.3–7.5 (m, PhH, 5H). 31P NMR (CDCl3): δ 99.1 (d, PH,
Thermolysis of 1: elimination of Mes*OH. Compound 1
(5.05 g, 16.4 mmol) was heated to 100 ЊC (under N2) in
a sublimation apparatus. Within minutes of heating, white
needles of Mes*OH could be seen on the water-cooled cold
finger. A white liquid was seen at the bottom of the sublim-
ation apparatus, which slowly became yellow and solidified
upon prolonged heating. After 24 h at 100 ЊC, the apparatus
was cooled and the yellow residue was ground in a mortar
and heated again at 100 ЊC for a further 2 d. This produced a
further crop of Mes*OH. The yellow material which
remained was placed in a Soxlet apparatus (under nitrogen)
and extracted continuously with hexanes (250 mL) for 3 d to
give a pale yellow insoluble powder, 7, (0.73 g, 54% yield
based on amount of Mes*OH produced) along with more
Mes*OH as part of the soluble fraction (total amount of
Mes*OH recovered = 3.7 g; ca. 0.9 mole equivalents, identi-
1JPH = 413 Hz). 11B{1H} NMR (CDCl3): δ Ϫ40.7 (d, BH3, 1JBP
=
47 Hz). 13C{1H} NMR (CDCl3): δ 31.8 (s), 32.7 (s), 34.8 (s),
36.3 (s), 124.4 (d, J = 2 Hz), 128.6 (d, J = 10 Hz), 133.4 (d, J =
1 Hz), 134.1 (d, J = 31 Hz), 141.3 (d, J = 3 Hz), 146.5 (d, J =
1 Hz), 148.5 (s), and 148.7 (s). MS 70 eV, EI (m/z, %): 370
(Mϩ Ϫ BH3, 7), 355 (Mϩ Ϫ Me Ϫ BH3, 5), 299 (Mϩ Ϫ BH3 Ϫ
Me Ϫ t-Bu, 22), 257 (Mϩ Ϫ Me Ϫ 2 t-Bu, 38), 57 (t-Buϩ, 100).
mp 128–9 ЊC (dec). Anal. Calc. for C24H38BOP (384.3): C,
75.00. H, 9.97. Found: C, 74.85. H, 9.88%.
1
fied by H and 13C{1H} NMR spectroscopy and mass spec-
trometry). The insoluble product (7) was characterized by
solid-state CP-MAS NMR, IR and elemental analysis (C and H).
Data for 7: Solid-state 31P CP-MAS (200 MHz, 8.5 KHz
spin rate): δ Ϫ183, Ϫ119, Ϫ82, 0.7, 67 and 91. Solid-state
11B CP-MAS NMR (200 MHz, 8.5 KHz spin rate): δ Ϫ42 and
3 (four-coordinate boron environments). Solid-state 13C CP-
MAS NMR 400 MHz, 6.5 KHz spin rate): δ 31, 34 and 36
(t-Bu, Mes*O), 124, 128, 141, 146 and 150 (aromatic, Mes*O).
IR (νmax, cmϪ1): 3209s, 2384m (ν B–H), 2267w (ν B–H), 1262w,
1102m, 1025w, 967vw, 917w, 881w, 797s, 788s, 729m,
696w, 677w, 649m, 550w. Anal. Calc. for [{(Mes*O)xRPHy–
Preparation of (Mes*O)Ph2P 4. In a similar procedure as for
2, a pale yellow solution of Ph2PCl (8.1 mL, 45 mmol) in 25 mL
of Et2O was added dropwise to a suspension (0 ЊC) of
Mes*OLiؒOEt2 (15.35 g, 44.8 mmol) in 150 mL of Et2O. Work-
up of the reaction mixture after 16 h gave a colorless solid in
1
high yield (18.23 g, 91%). H NMR (CDCl3): δ 1.25 (s, ortho
BH(2x
z)}n, x = 0.085, y =0, z = 0.915]: C, 28.85. H, 7.21.
ϩ
2y ϩ
t-Bu, 18H), 1.36 (s, para t-Bu, 9H), 7.21 (s, Ar–H, 2H), 7.35 (m,
Ph–H, 10H). 31P{1H} NMR (CDCl3): δ 127.3 (s). 13C{1H}
NMR (CDCl3): δ 31.8 (s), 33.0 (d, J = 5 Hz), 34.6 (s), 38.4 (s),
123.9 (s), 128.2 (d, J = 6 Hz), 129.8 (s), 132.5 (d, J = 22 Hz),
139.3 (d, J = 27 Hz), 141.6 (d, J = 4 Hz), 143.7 (d, J = 2 Hz),
150.8 (d, J = 8 Hz). MS 70 eV, EI (m/z, %): 446 (Mϩ, 35), 431
(Mϩ Ϫ Me, 4), 389 (Mϩ Ϫ t-Bu, 85), 355 (Mϩ Ϫ C7H7, 40), 185
(Ph2Pϩ, 100). mp 95–7 ЊC. HR-MS: Calc.: 446.2739. Found:
446.2732.
Found: C, 28.75. H, 6.62%.
1
Data for Mes*OH: H NMR (CDCl3): δ 1.38 (s, para t-Bu,
9H), 1.49 (s, ortho t-Bu, 18H), 5.09 (s, OH, 1H), 7.26 (Ar–H,
2H). 13C{1H} NMR (CDCl3): δ 30.7 (s), 32.0 (s), 34.8 (s), 34.7
(s), 122.1 (s), 135.2 (s), 141.7 (s) and 151.6 (s). MS 70 eV, EI
(m/z, %): 262 (Mϩ, 100).
Attempts to induce dehydrocoupling of 1 (150–200 mg) were
also conducted in toluene (100 ЊC, 2–3 mL) in the presence
of ca. 5 mol% [{Rh(µ-Cl)(1,5-cod)}2], ca. 5 mol% [Ir(1,5-cod)2]-
[OTf] and in the absence of transition metal complexes. In
all cases, 1 eliminated Mes*OH (0.8–0.9 equivalents) to give
insoluble yellow or brown (in the presence of catalyst) solids
with no evidence of clean dehydrocoupling.
Preparation of (Mes*O)Ph2PؒBH3 5. BH3 (19.0 mL, 1.0 M
solution in THF, 19.0 mmol) was added dropwise to a cold
(0 ЊC), pale yellow solution of 4 (8.34 g, 18.7 mmol) in 100 mL
of THF. The reaction was warmed to room temperature and
stirred for 24 h. The resulting clear and colorless solution was
concentrated to 20 mL and cooled to Ϫ30 ЊC for 6 d affording
large colorless blocks suitable for a single-crystal X-ray struc-
Thermolysis of 3: elimination of Mes*OH. Compound 3
(4.95 g, 12.9 mmol) was heated to 140 ЊC for 5 d (under N2) in a
sublimation apparatus. Upon heating the adduct melted to give
a colorless liquid and white needles of Mes*OH could be seen
on the cold finger (water-cooled). As heating was continued, the
1
ture determination (6.27 g, 73%). H NMR (CDCl3): δ 1.15 (s,
ortho t-Bu, 18H), 1.34 (s, para t-Bu, 9H), 7.19 (s, Ar–H, 2H),
7.45 and 7.78 (m, Ph–H, 10H). No BH3 protons were detected
J. Chem. Soc., Dalton Trans., 2002, 2966–2972
2971

View Article Online
3 (a) C. Aitken, J. F. Harrod, A. Malek and E. Samuel, J. Organomet.
Chem., 1988, 349, 285; (b) K. Mochida and H. Chiba, J. Organomet.
Chem., 1994, 473, 45; (c) For a demethanative coupling route, see:
S. M. Katz, J. A. Reichl and D. H. Berry, J. Am. Chem. Soc., 1998,
120, 9844.
4 (a) T. Imori and T. D. Tilley, J. Chem. Soc., Chem. Commun., 1993,
1607; (b) T. Imori, V. Lu, H. Cai and T. D. Tilley, J. Am. Chem. Soc.,
1995, 117, 9931; (c) J. R. Babcock and L. R. Sita, J. Am. Chem. Soc.,
1996, 118, 12481.
5 (a) N. Etkin, M. C. Fermin and D. W. Stephan, J. Am. Chem. Soc.,
1997, 119, 2954; (b) V. P. W. Böhm and M. Brookhart, Angew.
Chem., Int. Ed., 2001, 40, 4694.
6 J. M. Fischer, W. E. Piers, S. D. Pearce Batchilder and M. J.
Zaworotko, J. Am. Chem. Soc., 1996, 118, 283.
colorless liquid gradually became pale yellow and eventually
solidified to give a glassy residue 8 (1.52 g, 79% yield based on
the amount of Mes*OH isolated). Mes*OH (3.03 g, ∼ 0.9 mole
equivalents) was also recovered and characterized by 1H NMR
spectroscopy. Compound 8 was found to be highly soluble in
chlorinated solvents, THF, toluene and sparingly soluble in
hexanes. 1H NMR (CDCl3): δ 0.5–1.5 (br, BH2), 1.32, 1.34 and
1.48 (t-Bu, Mes*O), 6.8–7.6 (br, Ph). 31P{1H} NMR (CDCl3):
many signals from δ Ϫ120 to 100. 11B{1H} NMR (CDCl3):
δ Ϫ30 (very broad). 13C{1H} NMR (CDCl3): δ 30.9, 32.0, 32.2,
33.0, 35.0 and 36.5 (t-Bu, Mes*O), 122.0, 124.2, 127.5–129 (many
signals), 133.1, 133.8, 134.0 and 135.0 (aromatic, Mes*O).
7 R. Shu, L. Hao, J. F. Harrod, H.-G. Woo and E. Samuel, J. Am.
Chem. Soc., 1998, 120, 12988.
8 J. He, H. Q. Liu, J. F. Harrod and R. Hynes, Organometallics, 1994,
13, 336.
9 (a) Y. Li and Y. Kawakami, Macromolecules, 1999, 32, 6871; (b) R.
Zhang, J. E. Mark and A. R. Pinhas, Macromolecules, 2000, 33, 3508.
10 C. A. Jaska, K. Temple, A. J. Lough and I. Manners, Chem.
Commun., 2001, 962.
11 H. Dorn, R. A. Singh, J. A. Massey, A. J. Lough and I. Manners,
Angew. Chem., Int. Ed., 1999, 38, 3321.
12 H. Dorn, R. A. Singh, J. A. Massey, J. M. Nelson, C. A. Jaska, A. J.
Lough and I. Manners, J. Am. Chem. Soc., 2000, 122, 6669.
13 H. Dorn, E. Vejzovic, A. J. Lough and I. Manners, Inorg. Chem.,
2001, 40, 4327.
14 H. Dorn, C. A. Jaska, R. A. Singh, A. J. Lough and I. Manners,
Chem. Commun., 2000, 1041.
15 For a review on inorganic polymers, see: I. Manners, Angew. Chem.,
Int. Ed. Engl., 1996, 35, 1602.
16 L. F. Centofanti, Inorg. Chem., 1973, 12, 1131.
Thermolysis of 5 and 6: P–B bond cleavage without phenol
formation. Adducts 5 and 6 were heated for extended periods of
time under nitrogen, and the reaction progress was followed by
31P{1H} NMR. Representative examples are described below.
A sample of 5 was heated to 160 ЊC for 7 d. Analysis of the
remaining solid indicated complete dissociation of 5 [31P{1H}
NMR: δ 111 (s)] into the free phosphinite 4 [31P{1H} NMR:
1
δ 127 (s)] and volatile diborane (by inference). The H NMR
spectrum consisted exclusively of resonances attributed to the
formation of 4. Similar results were obtained when the experi-
ment was repeated in the presence of ca. 5 mol% [{Rh(µ-Cl)-
(1,5-cod)}2].
A sample of 6 was heated to 140 ЊC for 8 d. Analysis of the
remaining pale beige oil indicated complete dissociation of 6
[31P{1H} NMR: δ 108 (pseudo-quartet)] into the free phosphite,
(PhO)3P [31P{1H} NMR: δ 129 (s)], and presumably volatile
diborane. The 1H NMR spectrum consisted exclusively of
resonances attributed to the formation of (PhO)3P. Thermolysis
experiments in tetraglyme (140 ЊC) and in the presence of ca.
5 mol% [{Rh(µ-Cl)(1,5-cod)}2] yielded similar results after 5 d.
17 (a) D. S. Bohle, G. R. Clark, C. E. F. Rickard and W. R. Roper,
J. Organomet. Chem., 1990, 393, 243 . For related examples, see;
(b) G. Huttner and H.-D. Müller, Angew. Chem., Int. Ed. Engl.,
1975, 14, 571; (c) A. Marinetti and F. Mathey, Organometallics,
1982, 1, 1488; (d ) A. Marinetti, F. Mathey, J. Fischer and A.
Mitschler, J. Am. Chem. Soc., 1982, 104, 4484.
18 R. A. Bartlett, H. V. Rasika Dias, K. M. Flynn, M. M. Olmstead
and P. P. Power, J. Am. Chem. Soc., 1987, 109, 5699.
Single-crystal X-ray structural determination of 1, 3, 5 and 6.
Diffraction data were collected on a Nonius Kappa-CCD dif-
fractometer using graphite-monochromated Mo-Kα radiation
(λ = 0.71073 Å). A combination of 1Њ ꢀ and ω (with κ offsets)
scans were used to collect sufficient data. The data frames were
integrated and scaled using the Denzo-SMN package.33 The
structures were solved and refined with the SHELXTL-PC v5.1
software package.34 Refinement was by full-matrix least squares
on F 2 using all data (including negative intensities). In all struc-
tures, hydrogen atoms bonded to carbon atoms were included
in calculated positions and treated as riding atoms. The dis-
order within the para t-Bu-groups in 1 were modeled with an
87/13 occupancy. Data collection and refinement parameters
are summarized in Table 2.
19 R. W. Rudolph and R. W. Parry, J. Am. Chem. Soc., 1967, 89, 1621.
20 H. Nöth and B. Wrackmeyer, Nuclear Magnetic Resonance
Spectroscopy of Boron Compounds; Springer, Berlin, 1978.
21 (a) S. Jugé, M. Stephan, J.-P. Genet, S. Halut-Desportes and S.
Jeannin, Acta Crystallogr., Sect. C, 1990, 46, 1869; (b) T. Imamoto,
T. Yoshizawa, K. Hirose, Y. Wada, H. Masuda, K. Yamaguchi
and H. Seki, Heteroatom. Chem., 1995, 6, 99; (c) A. Dubourg, J.-P.
DeClercq, R. Contreras, A. Murillo and A. Klaebe, Acta
Crystallogr., Sect. C, 1985, 41, 1314.
22 S. D. Pastor, J. D. Spivack and L. P. Steinhuebel, Phosphorus, Sulfur
Silicon Relat. Elem., 1985, 22, 169.
23 K. Bourumeau, A.-C. Gaumont and J.-M. Denis, J. Organomet.
Chem., 1997, 529, 205.
24 M. Peruzzini, L. Marvelli, A. Romerosa, R. Rossi, F. Vizza and
F. Zanobini, Eur. J. Inorg. Chem., 1999, 931.
CCDC reference numbers 181265–181268.
See http://www.rsc.org/suppdata/dt/b2/b202361j/ for crystal-
lographic data in CIF or other electronic format.
25 J. P. Albrand and D. Gagnaire, J. Am. Chem. Soc., 1972, 94, 8630.
26 P. A. Chopard and R. F. Hudson, J. Inorg. Nucl. Chem., 1963, 25, 801.
27 (a) C. Couret, J. Escudie, H. Ranaivonjatovo and J. Satgé,
Organometallics, 1986, 5, 113; (b) A. H. Cowley, J. E. Kilduff,
N. C. Norman and M. Pakulski, J. Am. Chem. Soc., 1983, 105, 4845 .
For an example of a surprisingly stable fluorophosphine, see; (c)
R. C. Dobbie, P. D. Gosling and B. P. Straughan, J. Chem. Soc.,
Dalton Trans., 1975, 2368.
Acknowledgements
This research was supported by the Natural Sciences and
Engineering Research Council of Canada (NSERC). E. R.
thanks NSERC for a postgraduate fellowship (1999–2003),
I. M. thanks the University of Toronto for a McLean fellow-
ship (1997–2003), the Ontario Government for a PREA award
(1999–2003) and the Canadian Government for a Canada
Research Chair (2001). The authors also thank Mr. Cory Jaska
for obtaining solution 11B NMR spectra, Dr. Alex Young for
mass spectrometry data and Dr. Hiltrud Grondey for obtaining
CP-MAS NMR spectra.
28 An attempt to trap the potential phosphinidene intermediate
[PhP–BH3] by thermolysis of 3 in the presence of various alkene
solvents [cyclohexene, 65 ЊC; 2,3-dimethylbutadiene, 65 ЊC;
1-phenylcyclohexene, 140 ЊC] gave exclusively the free phosphine,
1
(Mes*O)PhPH (31P NMR: ca. δ 96.5, JPH = 190 Hz), and
hydroboration of the corresponding alkene (by 11B NMR). This
suggests that P–B bond cleavage is occurring in these solvents
without further elimination of the phenol.
29 R. K. Harris, J. Bowles, I. R. Stephenson and E. H. Wong,
Spectrochim. Acta, Part A, 1988, 44, 273.
30 B. Çetinkaya, I. Gümrükçü, M. F. Lappert, J. L. Atwood and
R. Shakir, J. Am. Chem. Soc., 1980, 102, 2086.
31 A. Pelter, R. Rosser and S. Mills, J. Chem. Soc., Chem. Commun.,
1981, 1014.
32 G. Giordano and R. H. Crabtree, Inorg. Synth., 1979, 19, 218.
33 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276, 307.
34 G. M. Sheldrick, SHELXTL-PC V5.1, Bruker Analytical X-Ray
Systems Inc., Madison, WI, 1997.
References and notes
1 F. Gauvin, J. F. Harrod and H. G. Woo, Adv. Organomet. Chem.,
1998, 42, 363.
2 (a) C. T. Aitken, J. F. Harrod and E. Samuel, J. Am. Chem. Soc.,
1986, 108, 4059; (b) T. D. Tilley, Acc. Chem. Res., 1993, 26, 22.
2972
J. Chem. Soc., Dalton Trans., 2002, 2966–2972