Synthetic and reaction chemistry of heteroatom stabilized boryl and cationic borylene complexes

Article abstract of DOI:10.1039/b512275a

The synthesis, spectroscopic and structural characterization of the aryloxy and amino functionalized chloroboryl complexes (η⁵-C ₅R₅)Fe(CO)₂B(OMes)Cl (R = H, 2a; R = Me, 3a) and (η⁵-C₅H₅)Fe(CO)₂B(N ⁱPr₂)Cl (7a) are reported. Compound 2a is shown to be a versatile substrate for further boron-centred substitution chemistry leading to the asymmetric boryl complexes (η⁵-C₅H ₅)Fe(CO)₂B(OMes)ER_n [ER_n = OC ₆H₄^tBu-4, 2c; ER_n = SPh, 2d] with retention of the metal-boron bond. The reactivities of 2a, 3a and 7a towards the halide abstraction agent Na[BAr^f₄] have also been examined, in order to investigate the potential for the generation of cationic heteroatom-stabilized terminal borylene complexes. The application of this methodology to the mesityloxy derivatives 2a and 3a gives rise to B-F containing products, presumably via fluoride abstraction from the [BAr^f ₄]^- counter-ion. By contrast, amino-functionalized complex 7a is more amenable to this approach, and the thermally robust terminal aminoborylene complex [(η⁵-C₅H₅)Fe(CO) ₂B(NⁱPr₂)][BAr^f₄] (9) can be isolated in ca. 50% yield. The reactivity of 9 towards a range of nucleophilic and/or unsaturated reagents has been examined, with examples of addition, protonolysis and metathesis chemistries having been established. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b512275a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthetic and reaction chemistry of heteroatom stabilized boryl and cationic
borylene complexes†
Deborah L. Kays (ne´e Coombs),*‡ Andrea Rossin, Joanne K. Day, Li-Ling Ooi and Simon Aldridge*
Received 30th August 2005, Accepted 17th October 2005
First published as an Advance Article on the web 14th November 2005
DOI: 10.1039/b512275a
The synthesis, spectroscopic and structural characterization of the aryloxy and amino functionalized
5
chloroboryl complexes (g -C5R5)Fe(CO)2B(OMes)Cl (R = H, 2a; R = Me, 3a) and
5
(g -C5H5)Fe(CO)2B(NiPr2)Cl (7a) are reported. Compound 2a is shown to be a versatile substrate for
further boron-centred substitution chemistry leading to the asymmetric boryl complexes
5
t
(g -C5H5)Fe(CO)2B(OMes)ERn [ERn = OC6H4 Bu-4, 2c; ERn = SPh, 2d] with retention of the
metal–boron bond. The reactivities of 2a, 3a and 7a towards the halide abstraction agent Na[BArf ]
4
have also been examined, in order to investigate the potential for the generation of cationic
heteroatom-stabilized terminal borylene complexes. The application of this methodology to the
mesityloxy derivatives 2a and 3a gives rise to B–F containing products, presumably via fluoride
abstraction from the [BArf ]− counter-ion. By contrast, amino-functionalized complex 7a is more
4
amenable to this approach, and the thermally robust terminal aminoborylene complex
5
[(g -C5H5)Fe(CO)2B(NiPr2)][BArf ] (9) can be isolated in ca. 50% yield. The reactivity of 9 towards a
4
range of nucleophilic and/or unsaturated reagents has been examined, with examples of addition,
protonolysis and metathesis chemistries having been established.
metalla-diyl complexes of boron, gallium and indium, and herein
we report on attempts to extend this chemistry to heteroatom (O,
N) stabilized boranediyl complexes,10 with a view to probing the
role of the heteroatom in their structural and reaction chemistry.
1
Introduction
There has been much recent interest in the chemistry of transition
metal boryl complexes (LnM–BX2),1 in part due to their impli-
cation in the hydroboration and diboration of carbon–carbon
multiple bonds,2,3 and also in the highly selective stoichiometric
and catalytic functionalization of hydrocarbons.4 As a result of the
remarkable activity exhibited by such species, the reactivity of bo-
ryl compounds towards a variety of substrates has therefore been
investigated in some depth.1 Notwithstanding this, fundamental
reactions that proceed with retention of the metal–boron linkage
are still relatively rare.1 Generally, to exhibit such reactivity, the
boryl ligand must feature weakly p donor groups such as chloride
or bromide which are labile under conditions which leave the M–
B bond intact. Asymmetric haloboryl species [LnM–B(R)X] (X =
Cl, Br), although relatively rare, are therefore potential precursors
to a range of boron-containing complexes through substitution
and abstraction chemistries.
5
Thus, synthetic routes to precursor complexes of the type (g -
C5Rꢀ5)Fe(CO)2B(ERn)X (ERn = OR, NR2) have been developed
and their reactivities with respect to halide substitution and
abstraction have been probed.
2
Experimental
All manipulations were carried out under a nitrogen or argon
atmosphere using standard Schlenk line or dry-box techniques.
Solvents were pre-dried over sodium wire (hexanes, toluene) or
molecular sieves (dichloromethane) and purged with nitrogen
prior to distillation from the appropriate drying agent (hexanes:
potassium, toluene: sodium, dichloromethane: CaH2). Benzene-d6
and dichloromethane-d2 (both Goss) were degassed and dried over
potassium (benzene-d6) or molecular sieves (dichloromethane-d2)
prior to use. Mesitol (MesOH, 2,4,6-trimethylphenol) (Aldrich)
was sublimed under high vacuum (ca. 100 ◦C at 10−4 Torr) prior
to use; nBuLi (1.6 M solution in hexanes), BCl3 (1.0 M in heptane)
and BBr3 (Aldrich) were used as received. TlF, Na[BPh4] and
[PPN]Cl ([PPN]+ = bis(triphenylphosphoranylidene)ammonium,
[Ph3PNPPh3]+) (all Aldrich) were dried under high vacuum for
In particular, given the elegant chemistry reported by Tilley
and co-workers in the synthesis of a range of base-free silylene
complexes,5 we have been seeking to exploit the abstraction
methodology outlined in Scheme 1 towards the synthesis of
cationic unsaturated group 13 ligand systems.6–10 Such an approach
has previously been shown to be effective for cationic aryl- and
Centre for Fundamental and Applied Main Group Chemistry, School of
Chemistry, Cardiff University, Main Building, Park Place, Cardiff, UK CF10
3AT. E-mail: AldridgeS@cardiff.ac.uk; Fax: (029) 20874030; Tel: (029)
20875495
† Electronic supplementary information (ESI) available: Details of our
characterizing data for the known compounds 11a, 11b, 12a and 12b. See
DOI: 10.1039/b512275a
5
5
12 h prior to use. Na[(g -C5H5)Fe(CO)2], Na[(g -C5Me5)Fe(CO)2],
iPr2NBCl2, tmpBBr2 (tmp = 2,2,6,6-tetramethylpiperidyl) and
Na[BArf ] were prepared according to literature methods.11–14
4
NMR spectra were measured on a Bruker AM-400 or JEOL 300
Eclipse Plus FT-NMR spectrometer. Residual signals of solvent
were used as reference for 1H and 13C NMR, while a sealed
tube containing a solution of [nBu4N][B3H8] in CDCl3 was used
‡ Current address: Central Research Laboratory, Mansfield Road, Oxford,
UK OX1 3TA.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 399–410 | 399
©

View Article Online
Scheme 1 Contrasting reactivity of haloboryl complexes towards nucleophilic and weakly coordinating anion substitution and abstraction chemistries.
as an external reference for 11B NMR; 19F NMR spectra were
referenced to external CFCl3. Infrared spectra were measured
for each compound pressed into a disk with an excess of dried
KBr or as a solution in an appropriate solvent on a Nicolet
500 FT-IR spectrometer. Mass spectra were measured by the
EPSRC National Mass Spectrometry Service Centre, University
of Wales, Swansea. Perfluorotributylamine was used as a standard
for high-resolution EI mass spectra. Despite repeated attempts,
satisfactory elementary microanalysis for new boryl and borylene
complexes was frustrated by their extreme air, moisture and (in
some cases) thermal instability. With the exception of oils 2e and
4, characterization of the new compounds is therefore based upon
multinuclear NMR, IR and mass spectrometry data (including
accurate mass measurement of molecular ions), supplemented by
single-crystal X-ray diffraction studies in the cases of 2a, 2d, 3,
7a, 10 and 13. In all cases the purity of the bulk material was
established by multinuclear NMR to be >95%. Abbreviations:
b = broad, s = singlet, d = doublet, t = triplet, q = quartet, pcq =
partially collapsed quartet, quin = quintet, m = multiplet, fwhm =
frequency width at half maximum.
Yields were typically in the order of 40–50% (e.g. 0.879 g, 47%
1
for 2a). Data for 2a: H NMR (300 MHz, C6D6) d 2.12 (s, 3H,
5
para-CH3), 2.28 (s, 6H, ortho-CH3), 4.24 (s, 5H, g -C5H5), 6.75 (s,
2H, aromatic CH). 13C NMR (76 MHz, C6D6) d 17.0 (ortho-CH3),
5
20.5 (para-CH3), 84.7 (g -C5H5), 129.2 (aromatic CH), 127.6, 132.7
(aromatic quaternary), 152.3 (ipso-aromatic quaternary), 214.1
(CO). 11B NMR (96 MHz, C6D6) d 61.5. IR (KBr disk, cm−1)
m(CO) 2002, 1940. Mass spec. (EI) M+ = 358 (weak), isotope
pattern corresponding to 1 Fe, 1 B and 1 Cl atom. Exact mass:
1
calc. for M+ 358.0225, measd 358.0236. Data for 3a: H NMR
5
(300 MHz, CD2Cl2) d 1.91 (s, 15H, g -C5Me5), 2.13 (s, 3H, para-
CH3), 2.23 (s, 6H, ortho-CH3), 6.83 (s, 2H, aromatic CH). 13C
5
NMR (76 MHz, CD2Cl2) d 9.9 (g -C5Me5), 17.0 (ortho-CH3), 20.4
5
(para-CH3), 96.5 (g -C5Me5), 128.8 (aromatic CH), 127.7, 129.9
(aromatic quaternary), 132.6 (ipso-aromatic quaternary), 217.8
(CO). 11B NMR (96 MHz, C6D6) d 63.7. IR (KBr disk, cm−1)
m(CO) 1971, 1925. Mass spec. (EI) M+ = 428 (weak), isotope
pattern corresponding to 1 Fe, 1 B and 1 Cl atom, fragment ion
peaks at m/z 400 [(M − CO)+, 55%] and 372 [(M − 2CO)+, 100%].
Exact mass: calc. for M+ 428.1007, measd 428.1017. Data for
1
2b: H NMR (300 MHz, C6D6) d 2.11 (s, 3H, para-CH3), 2.27
5
(s, 6H, ortho-CH3), 4.25 (s, 5H, g -C5H5), 6.75 (s, 2H, aromatic
Synthesis of MesOBX2 (1a: X = Cl; 1b: X = Br)
CH). 13C NMR (76 MHz, C6D6) d 17.1 (para-CH3), 20.5 (ortho-
5
The two compounds were prepared in a similar fashion, exem-
plified for 1a. To a solution of mesitol (1.01 g, 7.3 mmol) in
hexanes (40 cm3) at −78 ◦C was added dropwise a solution of
nBuLi in hexanes (5.5 cm3 of a 1.6 M solution, 8.8 mmol); the
resulting mixture was warmed slowly to 20 ◦C and stirred for 16 h.
The supernatant was filtered from the solid product, which was
then washed with hexanes (2 × 20 cm3) and dried in vacuo. To a
suspension of this lithium salt in toluene (40 cm3) was added BCl3
in heptane (7.3 cm3 of a 1.0 M solution, 7.3 mmol) and the mixture
CH3), 85.2 (g -C5H5), 129.3 (aromatic CH), 127.7, 132.9 (aromatic
quaternary), 153.2 (ipso-aromatic quaternary), 213.8 (CO). 11B
NMR (96 MHz, C6D6) d 59.3. IR (KBr disk, cm−1) m(CO) 2015,
1961. Mass spec. (EI) M+ not observed, fragment ion peaks at m/z
374 [(M − CO)+, 30%], 346 [(M − 2CO)+, 30%], isotope patterns
corresponding to 1 Fe, 1 B and 1 Br atom. Exact mass: calc. for
(M − CO)+ 373.9771, measd 373.9781.
◦
stirred at 20 C for a further 16 h. Filtration of the supernatant
Substitution chemistry of 2a: syntheses of
t
(g5-C5H5)Fe(CO)2B(OMes)OC6H4 Bu-4, 2c and
provided a stock solution of 1a in toluene which could be used for
further reactions. Due to redistribution of the OMes groups in the
product under continuous pumping, the pure compound could not
be isolated. 11B NMR (96 MHz, toluene) d 31.1. Stock solutions of
1b in toluene were prepared in a similar manner, with analogous
redistribution of the OMes groups occurring under continuous
pumping. 11B NMR (96 MHz, toluene) d 26.7.
(g5-C5H5)Fe(CO)2B(OMes)SPh 2d
t
To a suspension of NaOC6H4 Bu-4 (0.068 g, 0.40 mmol) in toluene
(10 cm3) was added a solution of 2a (0.142 g, 0.40 mmol) in
toluene (10 cm3) and the resulting mixture was stirred for 72 h.
Filtration, removal of volatiles in vacuo and recrystallization from
hexanes (ca. 20 cm3) at −50 ◦C gave rise to 2c as a colourless
microcrystalline solid (0.147 g, 79%). 2d was prepared in an
analogous fashion from NaSPh (0.041 g, 0.31 mmol) and 2a
(0.112 g, 0.31 mmol) and isolated as colourless crystals suitable for
X-ray diffraction (0.117 g, 83%). Data for 2c: 1H NMR (300 MHz,
Syntheses of (g5-C5R5)Fe(CO)2B(OMes)Cl (R = H, 2a; R = Me,
3a) and (g5-C5H5)Fe(CO)2B(OMes)Br, 2b
The three compounds were prepared in a similar fashion, exem-
5
t
plified for 2a. Typically, to a suspension of Na[(g -C5H5)Fe(CO)2]
C6D6) d 1.17 (s, 9H, Bu), 2.16 (s, 3H, para-CH3), 2.41 (s, 6H,
(1.035 g, 5.18 mmol) in toluene (15 cm3) was added a solution of
1a in toluene (43 cm3 of a 0.12 M solution, 5.18 mmol) and the
resulting mixture stirred at 20 ◦C for 16 h. Filtration, removal of
volatiles in vacuo and recrystallization from hexanes (ca. 20 cm3)
ortho-CH3), 4.19 (s, 5H, g -C5H5), 6.79 (s, 2H, mesityl CH), 6.86
5
(d, J = 8.5 Hz, OC6H4 2H, meta-CH), 7.10 (d, J = 8.5 Hz,
2H, OC6H4 ortho-CH). 13C NMR (76 MHz, C6D6) d 15.4 (ortho-
t
CH3), 18.6 (para-CH3), 29.4 (CH3 of Bu), 32.0 (quaternary of
◦
5
at −30 C yielded 2a and 3a as pale yellow crystals suitable for
tBu), 80.9 (g -C5H5), 121.0, 126.0 (OC6H4 aromatic CH), 128.5,
X-ray diffraction, and 2b as a pale yellow microcrystalline solid.
131.5 (mesityl aromatic quaternary), 129.1 (mesityl aromatic
400 | Dalton Trans., 2006, 399–410
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
5
CH), 145.4, 153.2 (OC6H4 aromatic quaternary), 153.2 (mesityl
aromatic quaternary), 213.3 (CO). 11B NMR (96 MHz, C6D6) d
47.4. IR (KBr disk, cm−1) m(CO) 2006, 1944. Mass spec. (EI): M+
not observed, fragment ion peaks at m/z 444 [(M − CO)+, weak],
isotope pattern corresponding to 1 Fe and 1 B atom. Data for 2d.
1H NMR (300 MHz, C6D6) d 2.08 (s, 3H, para-CH3), 2.24 (s, 6H,
of Na[(g -C5Me5)Fe(CO)2] with BF3·OEt2 these latter resonances
are attributed to (g -C5Me5)Fe(CO)2BF2, 4.15
5
Synthesis of (g5-C5H5)Fe(CO)2B(NiPr2)Cl, 7a
Reaction of iPr2NBCl2 (1.199 g, 6.6 mmol) with Na[(g -
C5H5)Fe(CO)2] (1.318 g, 6.6 mmol) in toluene (40 cm3) at 20 C
5
5
ortho-CH3), 4.20 (s, 5H, g -C5H5), 6.63 (s, 2H, mesityl aromatic
◦
CH), 6.93 (m, 3H, SPh meta- and para-CH), 7.46 (d, J = 7.3 Hz,
for 20 h, followed by filtration, removal of volatiles in vacuo and
extraction with hexanes (ca. 40 cm3) yielded crude 7a as an oily
brown solid. Yellow crystals suitable for X-ray ◦diffraction were
obtained by sublimation under high vacuum (40 C at 10−4 Torr).
Yield (of sublimed material): 0.259 g, 12%. 1H NMR (400 MHz,
C6D6) d 1.11 (d, J = 6.7 Hz, 6H, CH3 of iPr), 1.39 (d, J = 6.7 Hz,
2H, SPh ortho-CH). 13C NMR (76 MHz, C6D6) d 17.0 (ortho-
5
CH3), 20.5 (para-CH3), 83.7 (g -C5H5), 126.7, 129.2, 134.4 (SPh
aromatic CH), 127.6 (mesityl aromatic CH), 128.4, 132.4, 152.8
(mesityl aromatic quaternary), 134.8 (SPh aromatic quaternary),
214.6 (CO). 11B NMR (96 MHz, C6D6) d 69.1. IR (KBr disk, cm−1)
m(CO) 1996, 1930. Mass spec (EI): M+ not observed, fragment ion
peaks at m/z 404 [(M − CO)+, 100%], 376 [(M − 2CO)+, 20%].
Exact mass: calc. for (M − CO)+ 404.0699, measd 404.0700.
i
i
6H, CH3 of Pr), 3.40 (sept, J = 6.7 Hz, 1H, CH of Pr), 4.44
(sept, J = 6.7 Hz, 1H, CH of iPr), 4.69 (s, 5H, g -C5H5). 13C NMR
5
(76 MHz, C6D6) d 21.2 (CH3 of iPr), 23.9 (CH3 of iPr), 47.8 (CH
of Pr), 55.2 (CH of Pr), 88.2 (g -C5H5), 215.6 (CO). 11B NMR
(96 MHz, C6D6) d 55.4. IR (C6D6 soln, cm−1) m(CO) 2001, 1941.
Mass spec. (EI): M+ not observed, fragment ion peaks at m/z 295
[(M − CO)+, 65%], 288 [(M − Cl)+, weak], 267 [(M − 2CO)+,
10%], 223 [(M − NiPr2)+, 100%]. Exact mass: calc. for (M − CO)+
295.0592, measd 295.0595; calc. for (M − Cl)+ 288.0853, measd
i
i
5
Reaction of (g5-C5H5)Fe(CO)2B(OMes)Cl (2a) with Na[BArf4]
A mixture of Na[BArf ] (0.070 g, 0.08 mmol), 2a (0.028 g,
4
0.08 mmol) and dichloromethane-d2 (1.5 cm3) in a J. Young’s
NMR tube was sonicated at 20 ◦C for 16 h. After this time,
the reaction was found to be complete by the conversion of the
NMR signal at dB 61.5 (2a) to a broad doublet at dB 45.7, and by
the appearance of a partially collapsed quartet at dF −9.1. The
reaction was repeated on a preparative scale (0.350 g, 0.39 mmol
5
288.0856. The corresponding (g -C5H4Me) compound (7b) can be
prepared in an analogous manner: 1H NMR (400 MHz, C6D6) d
i
1.07 (d, J = 6.7 Hz, 6H, CH3 of Pr), 1.66 (d, J = 6.9 Hz, 6H,
i
5
CH3 of Pr), 1.72 (s, 3H, CH3 of g -C5H4CH3), 3.34 (sept, J =
of Na[BArf ]), whereupon filtration, removal of volatiles in vacuo
6.7 Hz, 1H, CH of Pr), 4.17 (m, 2H, CH of g -C5H4CH3), 4.30
i
5
4
and recrystallization from hexanes (ca. 20 cm3) at −50 ◦C yielded
(m, 2H, CH of g -C5H4CH3), 4.60 (sept, J = 6.7 Hz, 1H, CH of
5
5
a red oil (yield: 0.064 g, 47%), characterized by multinuclear
iPr). 13C NMR (76 MHz, C6D6) d 12.8 (CH3 of g -C5H4CH3), 21.3
5
i
i
i
NMR and IR spectroscopies as (g -C5H5)Fe(CO)2B(OMes)F, 2e.
(CH3 of Pr), 23.9 (CH3 of Pr), 47.8 (CH of Pr), 55.1 (CH of
1H NMR (300 MHz, C6D6) d 2.17 (s, 3H, para-CH3), 2.29 (s,
iPr), 82.4 (CH of g -C5H4CH3), 84.6 (CH of g -C5H4CH3), 100.9
5
5
5
5
6H, ortho-CH3), 4.23 (s, 5H, g -C5H5), 6.74 (s, 2H, aromatic
(quaternary C of g -C5H4CH3), 216.1 (CO). 11B NMR (96 MHz,
CH). 13C NMR (76 MHz, C6D6) d 17.0 (para-CH3), 20.4 (ortho-
C6D6) d 55.0. IR (C6D6 soln, cm−1) m(CO) 1998, 1938. Mass spec.
(EI): M+ 377 (weak), isotopic pattern corresponding to 1 Fe, 1 Cl
and 1 B atom, strong fragment ion peaks at m/z 309 [(M − CO)+,
65%], 281 [(M − 2CO)+, weak], 237 [(M − NiPr2)+, 100%]. Exact
mass: calc. for (M − CO)+ 309.0749, measd 309.0747.
5
CH3), 83.4 (g -C5H5), 129.3 (aromatic CH), 132.2, 134.7 (aromatic
quaternary), 216.7 (CO), ipso-C of mesityl ring not observed. 11
B
NMR (96 MHz, C6D6) d 45.7 (d, J = 181 Hz). 19F NMR (283 MHz,
C6D6) d −9.1 (pcq, J = 181 Hz). IR (C6D6 soln, cm−1) 2013, 1954.
Reaction of (g5-C5Me5)Fe(CO)2B(OMes)Cl (3a) with Na[BArf4];
identification of (g5-C5Me5)Fe(CO)2BF2
Synthesis of [(g5-C5H5)Fe(CO)]2(l-CO)(l-Btmp), 8
5
To a suspension of Na[(g -C5H5)Fe(CO)2] (0.300 g, 1.5 mmol)
A mixture of Na[BArf ] (0.042 g, 0.05 mmol), 3a (0.020 g,
in toluene (20 cm3) was added a solution of tmpBBr2 (0.47 g,
1.5 mmol) in toluene (5 cm3) and the resulting mixture was stirred
4
0.05 mmol) and dichloromethane-d2 (1.5 cm3) in a J. Young’s
NMR tube was sonicated at 20 ◦C for 30 min. Multinuclear
NMR monitoring of the reaction mixture at this point revealed,
in addition to unreacted 3a (dB 63.7), the appearance of a broad
doublet at dB 50.3 and a quartet at dF −5.8 (J = 160 Hz).
Further sonication was necessary to bring about consumption
of the remaining 3a, with the additional consequence that a
mixture of boron and fluorine-containing products was formed.
In addition to the species responsible for the resonances at
dB 50.3/dF −5.8, a second significant component was observed
giving rise to a well-defined triplet at dB 47.8 and a quartet at
dF −3.4 (J ≈ 200 Hz). The latter resonances were observed to
increase in intensity at the expense of the former on prolonged
sonication, although complete conversion to a single iron/boron
containing product could not be achieved. By comparison of
data with the product obtained from the independent reaction
◦
at 45 C for 192 h. Filtration, removal of volatiles in vacuo and
recrystallization from hexanes (ca. 20 cm3) at −30 ◦C gave rise
1
to 8 as dark red crystals (0.059 g, 8%). H NMR (400 MHz,
C6D6) d 0.96 (quin, J = 6 Hz, 2H, CH2), 1.54 (t, J = 6 Hz,
5
4H, CH2), 1.58 (s, 12H, CH3 of tmp), 4.58 (s, 10H, g -C5H5).
13C NMR (76 MHz, C6D6) d 16.0 (para-CH2), 33.4 (CH3 of
5
tmp), 37.6 (meta-CH2), 55.3 (quaternary of tmp), 88.3 (g -C5H5),
217.6 (terminal CO), 274.7 (bridging CO). 11B NMR (96 MHz,
C6D6) d 115.3. IR (KBr disk, cm−1) m(CO) 1911, 1762. Mass spec
(EI): M+ 477 (15%), isotopic pattern corresponding to 2 Fe and
1 B atom, strong fragment ion peaks at m/z 449 [(M − CO)+,
20%], 421 [(M − 2CO)+, 15%], 393 [(M − 3CO)+, 5%], 328 [(M −
5
5
(g -C5H5)FeCO)+, 10%], 300 [(M − (g -C5H5)Fe(CO)2)+, 5%],
272 [(M − (g -C5H5)Fe(CO)3)+, 25%]. Exact mass: calc. for M+
5
477.0856, measd 477.0864.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 399–410 | 401
©

View Article Online
Synthesis of [(g5-C5H5)Fe(CO)2B(NiPr2)]+[BArf4]−, 9
Reaction of [(g5-C5H5)Fe(CO)2B(NiPr2)]+[BArf4]−, 9, with
=
=
Ph3P S and Ph3As O
Reaction of 7a (0.259 g, 0.80 mmol) and Na[BArf ] (0.710 g,
4
0.80 mmol) in dichloromethane (15 cm3) from −78 to 20 ◦C over
a period of 30 min leads to quantitative conversion (by 11B NMR)
of 7a (dB 55.4) to 9 (dB 93.5). Filtration, and recrystallization from
dichloromethane–hexanes (or fluorobenzene–hexanes) at −30 ◦C
leads to the isolation of 9 as a spectroscopically pure colourless
oil. Isolated yield: 0.465 g, 50%. 1H NMR (400 MHz, CD2Cl2) d
In a typical reaction 9 (0.068 g, 0.06 mmol) and Ph3P S (1.0 equiv.)
were stirred together in dichloromethane for 30 min at 20 ◦C, after
which time the reaction was judged to be complete by 11B and
31P NMR (conversion of signals at dB 93.5 and dP 43.7 to dB 35.6
and dP 60.8). Removal of volatiles in vacuo and extraction into
=
hexanes gives Pr2NB(l-S)2BNiPr2 (12a) which was identified by
i
1.39 (d, J = 6.6 Hz, 12H, CH3 of iPr), 3.32 (sept, J = 6.6 Hz, 2H,
comparison of 1H, 13C and 11B NMR and mass spectrometry data
with those reported previously.17 1H, 13C, 11B, 19F and 31P NMR
5
CH of iPr), 5.33 (s, 5H, g -C5H5), 7.55 (s, 4H, para-CH of BArf −),
4
7.70 (s, 8H, ortho-CH of BArf −). 13C NMR (76 MHz, CD2Cl2) d
and IR of the hexane-insoluble product confirmed it to be [(g -
C5H5)Fe(CO)2(PPh3)]+[BArf4]− (11a).18 Our data for the known
compounds 11a and 12a are included in the ESI.† A similar
procedure was adopted for the reaction of 9 with Ph3As O and
details of the characterizing data for 11b and 12b are also included
5
4
24.4 (CH3 of iPr), 51.0 (CH of iPr), 87.1 (g -C5H5), 117.6 (para-CH
5
of BArf4−), 124.6 (q, J = 272 Hz, CF3 of BArf4−), 128.8 (q, J =
34 Hz, meta-C of BArf4−), 134.8 (ortho-CH of BArf4−), 161.8 (q,
J = 49 Hz, ipso-C of BArf4−), 205.6 (CO). 11B NMR (96 MHz,
=
CD2Cl2) d −7.7 (BArf4−), 93.5 (b, fwhm ca. 615 Hz, BNiPr2). 19
F
in the ESI.†17,18
NMR (283 MHz, CD2Cl2) d −62.6 (CF3). IR (CD2Cl2 soln, cm−1)
m(CO) 2070, 2028. Mass spec. (ES): M+ 288.1 (10%).
Reaction of [(g5-C5H5)Fe(CO)2B(NiPr2)]+[BArf4]−, 9, with
=
Ph3P O
Reaction of [(g5-C5H5)Fe(CO)2B(NiPr2)]+[BArf4]−, 9, with
[PPN]Cl
The reaction of 9 with Ph3PO proceeds along similar lines to
those reported above for Ph3PS and Ph3AsO, with the caveat that
the increased temperature (480 h at 35 C) required to drive the
◦
To a solution of [PPN]Cl (0.010 g, 1.67 equiv.) in dichloromethane-
d2 in a J. Young’s NMR tube was added a solution of 9 (0.012 g)
in dichloromethane-d2 at 20 ◦C. Monitoring of the reaction by
11B NMR spectroscopy revealed that after 10 min the resonance
characteristic of 9 (dB 93.5) had been quantitatively replaced by a
signal at dB 55.4. Further examination of the reaction mixture
by 1H, 11B and 13C NMR and by IR revealed spectroscopic
reaction to completion (i) allows for the isolation of the inter-
mediate species [(g -C5H5)Fe(CO)2{B(NiPr2)(OPPh3)}]+[BArf4]−,
5
13; and (ii) results in appreciable conversion of the iron-
containing product [(g -C5H5)Fe(CO)2(PPh3)]+ into dinuclear (g -
C5H5)2Fe2(CO)3(PPh3).19 Data for 13: Reaction of 9 (0.199 g,
0.17 mmol) and Ph3PO (0.048 g, 0.17 mmol) in dichloromethane
(5 cm3) at 20 ◦C over a period of 30 min, followed by filtration and
recrystallization from dichloromethane–hexanes at −30 ◦C leads
to the isolation of 13 as pale yellow crystals. Isolated yield: 0.105 g,
43%. 1H NMR (400 MHz, CD2Cl2) d 1.06 (d, J = 6 Hz, 6H, CH3
of iPr), 1.20 (d, J = 6 Hz, 6H, CH3 of iPr), 3.25 (sept, J = 6 Hz, 1H,
5
5
5
features characteristic of (g -C5H5)Fe(CO)2B(NiPr2)Cl (7a), which
5
were indistinguishable from samples synthesized from Na[(g -
C5H5)Fe(CO)2] and iPr2NBCl2 (vide supra).
CH of iPr), 4.10 (sept, J = 6 Hz, 1H, CH of iPr), 4.50 (s, 5H, g -
5
Reaction of [(g5-C5H5)Fe(CO)2B(NiPr2)]+[BArf4]−, 9, with H2O
C5H5), 7.38–7.48 (m, 9H, ortho and para-CHs of Ph3PO), 7.55 (s,
4H, para-CH of BArf4−), 7.60–7.78 (m, 6H, meta-CH of Ph3PO),
7.73 (s, 8H, ortho-CH of BArf4−). 13C NMR (76 MHz, C6D6) d 22.4,
To a solution of 9 (0.012 g, 0.01 mmol) in dichloromethane-d2
(2 cm3) in a J. Young’s NMR tube was added H2O (0.01 cm3,
0.0◦1 g, 0.5 mmol) and the resulting mixture was sonicated at
20 C for 2 h. At this point 1H and 11B monitoring of the reaction
revealed complete consumption of 9; filtration, layering of the of
the resulting solution with hexanes and storage at −30 ◦C led
to the formation of [H2NiPr2]+[BArf4]−, 10, as colourless crystals
suitable for X-ray diffraction (see ESI†) Isolated yield: 0.007 g,
72%. Identification of the boron- and iron-containing products as
23.9 (CH3 of iPr), 47.2 (CH of iPr), 84.4 (g -C5H5), 117.5 (para-CH
5
of BArf4−), 122.6 (ipso-C of Ph3PO), 124.6 (q, J = 272 Hz, CF3 of
BArf4−), 128.9 (q, J = 34 Hz, meta-C of BArf4−), 130.1 (meta-CH
of Ph3PO), 133.5 (ortho-C of Ph3PO), 134.9 (ortho-CH of BArf4−),
135.9 (para-C of Ph3PO), 161.8 (q, J = 49 Hz, ipso-C of BArf4−),
214.7 (CO). 11B NMR (96 MHz, CD2Cl2) d −7.7 (BArf4−), 48.9 (b,
fwhm ca. 480 Hz, B(OPPh3)NiPr2). 19F NMR (283 MHz, CD2Cl2)
d −62.7 (CF3). IR (CD2Cl2 soln, cm−1) m(CO) 2004, 1949. Mass
spec. (ES): M+ 566.1 (5%).
5
B(OH)3 and [(g -C5H5)Fe(CO)2]2 was confirmed by comparison of
IR and multinuclear NMR data with reported values.16 Data for
10: 1H NMR (400 MHz, CD2Cl2) d 1.25 (d, J = 6.5 Hz, 12H, CH3
of iPr), 3.37 (sept, J = 6.5 Hz, 2H, CH of iPr), 4.99 (t, J = 52.9 Hz,
2H, NH2), 7.44 (s, 4H, para-CH of BArf4−), 7.63 (s, 8H, ortho-CH
of BArf4−). 13C NMR (76 MHz, CD2Cl2) d 20.0 (CH3 of iPr), 51.9
3
Results and discussion
(i) Synthesis and substitution chemistry of boryl complexes
i
(CH of Pr), 117.6 (para-CH of BArf4−), 124.6 (q, J = 272 Hz,
CF3 of BArf4−), 128.9 (q, J = 32 Hz, meta-C of BArf4−), 134.9
The boranes MesOBX2 (X = Cl 1a, X = Br 1b) can readily
be synthesized by the reaction of MesOLi with BX3 in toluene.
However, isolation of either 1a or 1b as a pure compound
proves impossible as exposure to continuous vacuum facilitates
substituent redistribution, presumably driven by the loss of volatile
BX3. Thus, for example, in the case of 1a attempted removal
of solvent yields an oily mixture containing 1a (dB 31.1) and a
−
(ortho-CH of BArf4−), ipso-C of BArf4 not observed. 11B NMR
(96 MHz, CD2Cl2) d −7.6 (BArf4−). 19F NMR (283 MHz, CD2Cl2)
d −62.6 (CF3). IR (CD2Cl2 soln, cm−1) m(NH) 3340, 3123. Mass
spec. (ES+): [H2NiPr2]+ 102 (100%); (ES−): [BArf4]− 863 (100%).
Exact mass: calc. for [H2NiPr2]+ 102.1277, measd 102.1277; calc.
for [BArf4]− 863.0654, measd 863.0642.
402 | Dalton Trans., 2006, 399–410
This journal is The Royal Society of Chemistry 2006
©

View Article Online
Scheme 2 Synthesis and reactivity of mesityloxy(halo)boryl complexes. Reagents and conditions: (i) BX3 (1 equiv.), toluene, 20 ◦C, 16 h;
◦
(ii) Na[(g -C5R5)Fe(CO)2] (1 equiv.), toluene, 20 C, 16 h, 40–50%; (iii) 2a, [RnE]− (NaOC6H4 Bu-4 or NaSPh) (1 equiv.), toluene, 20 ◦C, 48–72 h,
5
t
79–83%; (iv) 2a, Na[BArf4] (1 equiv.), dichloromethane, 20 ◦C, 16 h, 47%; (v) 3a, Na[BArf4] (1 equiv.), dichloromethane-d2, −78 to 20 ◦C, 1 h. (vi) 2a,
Na[BPh4] (1 equiv.), dichloromethane, 20 ◦C 168 h; (vii) 2a or 2b, excess Na[(g -C5H5)Fe(CO)2], toluene, 50 ◦C, >72 h.
5
species giving rise to a second resonance at dB 22.3, which on
the basis of the known rearrangement chemistry of boranes of
the type (RO)BX2,20 and the reported shift for (PhO)2BCl (dB
22.0),20 is tentatively identified as (MesO)2BCl. Further support
for this assignment comes from the fact that the reactions can be
driven in the reverse direction by the addition of excess BCl3 (in
heptane) with subsequent stirring for 12 h. The rate of substituent
redistribution under vacuum is significantly lower for 1b compared
to 1a, presumably due to the less volatile nature of BBr3 (cf . BCl3).
Despite this redistribution chemistry, both 1a and 1b can con-
veniently be obtained as stock solutions in toluene by filtration of
the BX3–MesOLi reaction mixture, and used for further reaction
chemistry (Scheme 2). Accordingly, the asymmetric haloboryl
Both complexes have been spectroscopically characterized, and
the structure of 2d confirmed by X-ray diffraction. Given the
lower steric requirements of the mesityloxo substituent, it is
somewhat surprising that unlike (g -C5H5)Fe(CO)2B(Mes)Br, 2a
and 2b do not undergo substitution with a second equivalent of
5
an organometallic nucleophile to form the corresponding bridging
5
borylene species {i.e. [(g -C5H5)Fe(CO)2]2B(OMes)}. Presumably,
this reflects p electron release by the aryloxy substituent and
hence the lower electrophilicity of the boron centre in the
B(OMes)X ligand compared to B(Mes)Br. Given its relevance to
halide abstraction chemistry (vide infra), attempts were made to
5
synthesize the fluoroboryl complex (g -C5H5)Fe(CO)2B(OMes)F
(2e) by the reaction of 2a with sources of fluoride (such as
5
complexes (g -C5R5)Fe(CO)2B(OMes)X (2a: R = H, X = Cl; 2b:
TlF), a transformation which has previously been reported by
5
R = H, X = Br; 3a: R = Me, X = Cl) can be synthesized in
moderate yields (39–50%) by the reaction of a toluene solution of
1a or 1b with one equivalent of the appropriate organometallic
anion (Scheme 2). Under these conditions, substitution is selective
for a single halide, with negligible quantities of the disubstituted
Braunschweig for (g -C5H5)Fe(CO)2B(Cl)Si(SiMe3)3.22 In the case
of 2a however this approach was unsuccessful, leading instead to
the decomposition of all Fe–B bonded species.
(ii) Halide abstraction chemistry
5
species [(g -C5R5)Fe(CO)2]2B(OMes) being formed (as determined
by the 11B NMR spectrum of the reaction mixture). 2a, 2b and
3a have been characterized by multinuclear NMR, IR and mass
spectrometry (including exact mass determination) and in the
cases of 2a and 3a by single-crystal X-ray diffraction.
Given the success of halide abstraction methodology in the
synthesis of novel terminal borylene species,6,7 the reactivity of the
boryl complexes 2a and 3a towards Na[BPh4] and Na[BArf4] was
also investigated (Scheme 3). No reaction of 2a with Na[BPh4] in
dichloromethane was observed under any conditions, presumably
due to the low solubility of the tetraphenylborate reagent; 2a does,
however, react with Na[BArf4] in dichloromethane over a period
of 16 h. This reaction does not lead to the isolation of the expected
cationic terminal borylene species, however, yielding instead a red
oil, the identity of which is suggested by spectroscopic data to be
As has been found for the analogous mesityl(bromo)boryl com-
5
5
plex (g -C5H5)Fe(CO)2B(Mes)Br, (g -C5H5)Fe(CO)2B(OMes)Cl
(2a) is a versatile substrate for further subsititution chemistry
with anionic main group nucleophiles.21 Thus, reaction with one
◦
t
equivalent of NaOC6H4 Bu-4 or NaSPh in toluene at 20 C leads to
5
the asymmetric boryl complexes (g -C5H5)Fe(CO)2B(OMes)ERn
(ERn = OC6H4 Bu = 2c; ERn = SPh = 2d) in good yield.
t
5
the asymmetric fluoroboryl complex (g -C5H5)Fe(CO)2B(OMes)F
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 399–410 | 403
©

View Article Online
5
Scheme 3 Synthesis and reactivity of amino(halo)boryl complexes. Reagents and conditions: (i) NR2 = tmp, Na[(g -C5H5)Fe(CO)2] (1 equiv.), toluene,
◦
45 ◦C, 192 h, 8%; (ii) NR2 = NiPr2, Na[(g -C5H5)Fe(CO)2] (1 equiv.), toluene, 20 C, 20 h, 12% (after sublimation); (iii) Na[BArf4] (1 equiv.),
5
dichloromethane-d2, −78 ◦C to 20 ◦C, 30 min, 50%.
(2e). The formation of 2e is assumed to occur by the abstraction
substituents we have therefore sought to exploit this methodology
towards the synthesis of more robust cationic borylene species.
In a similar fashion to its NMe2 substituted counterpart,25a the
of a fluoride anion from [BArf4]− by the highly reactive putative
5
terminal borylene complex [(g -C5H5)Fe(CO)2B(OMes)]+. Such
reactivity towards the weakly coordinating [BArf4]− counterion
has precedent in the literature for highly electrophilic species,23 and
asymmetric chloroboryl complex (g -C5H5)Fe(CO)2B(NiPr2)Cl
5
(7a) can be synthesized by the reaction of iPr2NBCl2 with
5
5
cationic borylene complexes of the type [(g -C5R5)Fe(CO)2BX]+
one equivalent of Na[(g -C5H5)Fe(CO)2] in toluene (Scheme 3).
have previously been shown to abstract both F− and Ph− from
borate anions of the type [BX4]−.7 No trace of any intermediate
was observed by multinuclear NMR monitoring of the reaction,
even at −80 ◦C.
As with aryloxyboryl complexes 2a, 2b and 3a, this reaction
proceeds with the selective substitution of one chloride sub-
stituent, with negligible quantities of the disubstituted species
5
[(g -C5H5)Fe(CO)2]2B(NiPr2) being formed (as determined by 11
B
Given the increased steric shielding and electron donation of-
NMR). Indeed, all attempts to substitute the second chloride to
form the bridging borylene complex, even by using very forcing
5
fered by the g -C5Me5 ligand, the corresponding halide abstraction
5
reaction with the more sterically encumbered boryl complex 3a
was also examined by multinuclear NMR spectroscopy. Slow
warming of 3a with Na[BArf4] in dichloromethane-d2 from −78
to 20 ◦C gives a mixture of two major boron-containing products.
The first of these is characterized by a broad doublet at dB 50.3 and
a quartet at dF −5.8 (1JBF = 160 Hz), which by comparison with
conditions failed to produce [(g -C5H5)Fe(CO)2]2B(NiPr2). Due
to the high solubility of 7a in hexanes, clean samples could not
easily be isolated by crystallization at low temperatures. However,
sublimation of the crude product (10−4 Torr, 45 ◦C) yielded
crystalline samples of pure 7a, which have been characterized by
multinuclear NMR, IR mass spectrometry (including exact mass
determination) and single-crystal X-ray diffraction.
5
the very similar data measured for (g -C5H5)Fe(CO)2B(OMes)F
1
(2e, dB 45.7, dF −9.1, JBF = 181 Hz) suggests an analogous
By contrast, the corresponding reaction of the more
5
5
formulation as (g -C5Me5)Fe(CO)2B(OMes)F (3b). As the reaction
bulky aminoborane tmpBBr2 with one equivalent of Na[(g -
progresses, the resonances associated with 3b diminish, with
C5H5)Fe(CO)2] in toluene does not yield the analogous boryl
1
5
signals at dB 47.8 (triplet) and dF −3.4 (quartet, JBF = 201 Hz)
complex (g -C5H5)Fe(CO)2B(tmp)Br, but leads instead to the
5
being observed to grow in. Comparison of multinuclear NMR and
supported bridging borylene [(g -C5H5)Fe(CO)]2(l-CO)(l-Btmp)
5
IR data with those reported for (g -C5Me5)Fe(CO)2BF2 (prepared
8 (Scheme 4). The structure of 8 in solution is derived from
5
from Na[(g -C5Me5)Fe(CO)2] and BF3·OEt2)15 and for other –
spectroscopic measurements, including mass spectrometry data
BF2 complexes,24 strongly suggests that this second species is the
and exact mass measurements. In addition to the presence of g -
5
5
difluoroboryl complex (g -C5Me5)Fe(CO)2BF2, 4. Although 4 is
C5H5 and 2,2,6,6-tetramethylpiperidyl moieties in the 1H and 13
C
the major boron-containing component of the reaction mixture
after 1 h, complete conversion to a single iron/boron product
proved impossible.
NMR spectra in the expected 2:1 ratio, the 11B NMR spectrum of
8 shows a resonance at dB 115.3, which is characteristically shifted
downfield ca. 90 ppm from the starting material, indicating the re-
5
The unusual reactivity displayed by boryl complexes 2a and 3a
placement of both bromides by (g -C5H5)Fe(CO)2 fragments.21a,26
towards Na[BArf4] contrasts with that observed for the mesityl
This chemical shift value is very similar to that for other bridging
5
5
complex (g -C5Me5)Fe(CO)2B(Mes)Br, which cleanly generates
aminoborylene complexes [e.g., dB 118.4, 119.1 and 105.9 for {(g -
5
[(g -C5Me5)Fe(CO)2BMes]+[BArf4]−.6,7 Presumably, given the dif-
C5R5)Fe(CO)}2(l-CO)(l-BN{SiMe3}2 (R5 = H5 or H4Me) and
5
fering steric and electronic properties of the Mes and OMes
substituents, this behaviour is dominated by the smaller degree
of steric shielding at the highly electrophilic boron centre afforded
by the OMes group, leading to a more reactive putative borylene
{(g -C5H5)Ru(CO)}2(l-CO)(l-BN{SiMe3}2, respectively].25 The
13C NMR data for 8 shows both terminal and bridging carbonyl
signals (dC 217.6 and 274.7, respectively) and this together with the
carbonyl stretches (1911 and 1762 cm−1) provides strong evidence
5
5
species in [(g -C5R5)Fe(CO)2B(OMes)]+.
for the proposed structure, [(g -C5H5)Fe(CO)]2(l-CO)(l-Btmp).
Given the lability demonstrated by the putative aryloxybo-
The carbonyl stretching frequencies for 8 are very similar to those
5
5
rylene complexes [(g -C5R5)Fe(CO)2B(OMes)]+, we have sought
for the bridging aminoborylene complexes {(g -C5R5)Fe(CO)}2(l-
to exploit the greater steric shielding afforded by amino sub-
stituents at the boron centre. We have previously shown that
CO)(l-BN{SiMe3}2 (R5 = H5, 1924 and 1770 cm−1; R5 = H4Me,
1925 and 1770 cm−1), and in common with these species a trans
arrangement of the terminal CO ligands in 8 would be expected
on steric grounds.25 Intriguingly, no evidence of the intermediate
5
the aminoborylene complex [(g -C5Me5)Fe(CO)2B(NMe2)]+ is
accessible by halide abstraction chemistry and stable enough below
−30 ◦C to permit spectroscopic characterization and subsequent
chemical trapping.7 By increasing the steric bulk of the amino
5
boryl complex (g -C5H5)Fe(CO)2B(tmp)Br is seen in the 11B NMR
spectrum of the reaction mixture at any time, indicating that if
404 | Dalton Trans., 2006, 399–410
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Scheme 4 Reactivity of terminal aminoborylene complex 9. Reagents and conditions: (i) [PPN]Cl (1.67 equiv.), dichloromethane-d2, 20 ◦C, 10 min,
◦
quantitative (by NMR); (ii) H2O (ca. 50 equiv.), dichloromethane-d2, 20 C, 2 h, 72%; (iii) Ph3P S or Ph3As O (1 equiv.), dichloromethane, 20 ◦C,
=
=
30 min, quantitative by NMR; (iv) Ph3P O (1 equiv.), dichloromethane, 20 ◦C, 30 min, 43%; (v) 35 ◦C, 480 h.
=
5
5
this species is an intermediate in the formation of 8, it never
achieves measurable concentration. The difference in reactivity
between the two aminoboranes iPr2NBCl2 and tmpBBr2 towards
and 9) mirrors that found for (g -C5Me5)Fe(CO)2B(Mes)Cl/[(g -
C5Me5)Fe(CO)2(BMes)]+ (dB 112.1 and 145.0) and for (g -
5
C5Me5)Fe(CO)2B(NMe2)Cl/[(g -C5Me5)Fe(CO)2(BNMe2)]+ (dB
5
5
Na[(g -C5H5)Fe(CO)2] is difficult to rationalize in terms of the
58.6 and 88.0).6,7 1H and 13C NMR data are consistent with the
5
(similar) electronic properties of the two amino substituents, and
the greater steric demands of tmp over NiPr2, but presumably
reflects the differences in reactivities of B–Cl and B–Br bonds.
presence of (g -C5H5), NiPr2 and [BArf4]− moieties in a 1 : 1 : 1
5
ratio, and the ES+ mass spectrum shows the presence of the [(g -
C5H5)Fe(CO)2(BNiPr2)]+ cation. The observation of equivalent iPr
amino substituents is consistent with the structure of the related
5
Moreover, the lability of the putative tmp(halo)boryl complex (g -
5
28c
=
C5H5)Fe(CO)2B(tmp)Br is somewhat surprising given the likely
degree of steric shielding at the boron centre towards attack by a
second equivalent of the organometallic nucleophile.27
neutral system (g -C5H5)V(CO)3 BN(SiMe3)2, and agrees with
5
the results of DFT calculations for the model compounds [(g -
C5R5)Fe(CO)2(BNMe2)]+ (R = H, Me) for which a minimum
energy structure close to Cs symmetry [∠centroid-Fe–N–C ≈
90◦ (84.6◦ for R = Me)] and a low barrier to rotation about
the Fe–B–N axis (ca. 2.2 kcal mol−1) have been calculated.29
Additionally, the IR spectrum of 9 shows carbonyl stretching
frequencies (2070, 2028 cm−1) which are significantly blue-shifted
with respect to 7a (2001, 1941 cm−1) {cf. Dm ≈ 50 cm−1 for
The reactivities of amino(halo)boryl complexes 5a and 7a
5
towards Na[BArf4] have also been investigated. Reaction of (g -
C5Me◦5)Fe(CO)2B(NMe2)Cl (5a) with Na[BArf4] in CD2Cl2 at
5
−20 C yields the cationic terminal borylene complex [(g -
C5Me5)Fe(CO)2B(NMe2)]+[BArf4]− (6), the synthesis of which has
previously been reported.7 Due to the lack of steric shielding
around the low-coordinate boron centre, 6 is thermally fragile, but
has been characterized by multinuclear NMR and IR spectroscopy
at −20 ◦C and by its reaction with [PPN]Cl in CD2Cl2 to yield
5
5
(g -C5Me5)Fe(CO)2B(Mes)Cl/[(g -C5Me5)Fe(CO)2(BMes)]+} and
which are very similar to those reported for archetypal Fischer
5
+
−
=
carbene systems such as [(g -C5H5)Fe(CO)2 CH(SPh)] [PF6]
5
the known chloroboryl complex (g -C5Me5)Fe(CO)2B(NMe2)Cl
(2073, 2034 cm−1).30 Further evidence for the nature of 9 is
obtained (i) from its reaction with [PPN]Cl which in common
with the analogous reaction for structurally characterized cationic
derivatives,6,7 generates a haloboryl complex (in this case 7a) by
(5b).7 Increased steric shielding was therefore thought to be
paramount in the stabilization of the extremely electrophilic
boron centre found in these cationic B/N vinylidene ana-
logues. Reaction of 7a with Na[BArf4] in dichloromethane re-
=
halide addition at boron; and (ii) from its reaction with Ph3P O
1
5
5
sults in quantitative conversion (by H and 11B NMR) to [(g -
C5H5)Fe(CO)2(BNiPr2)]+[BArf4]− (9). 9 (along with its C5H4Me
counterpart) is a colourless oil at (or close to) 20 ◦C, but its
formulation can be definitively established from spectroscopic and
reactivity data. Thus the 11B chemical shift (dB 93.5) is very close
to those reported by Braunschweig for neutral terminal aminobo-
which proceeds via the structurally characterised adduct [(g -
C5H5)Fe(CO)2{B(NiPr2)(OPPh3)}]+[BArf4]− (vide infra).
(iii) Reactivity of terminal aminoborylene complex 9
=
Reports of the fundamental chemistry of M B double bonds
have been somewhat limited, predominantly to metal–metal
transfer reactions and addition/substitution reactivity towards
rylene systems of the type LnM BN(SiMe3)2 (dB 86.6–98.3),28 and
=
the downfield shift on chloride abstraction (DdB 38.1 between 7a
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 399–410 | 405
©

View Article Online
5
nucleophiles.7,28,31 Thus, for example, the chemistry of [(g -
presumably reflecting the greater strength of the P O bond. In
=
C5Me5)Fe(CO)2(BMes)]+ (Mes = C6H2Me3-2,4,6) is dominated
by electrophilic character at both Fe and B centres, while neutral
aminoborylene complexes of the type (OC)5MBN(SiMe3)2 (M =
Cr or W) act as effective sources of the [BN(SiMe3)2] fragment
towards metal complexes bearing labile ligands, and towards
C≡C triple bonds.7,28,32 Hence it was with a view to further
exploring the scope of reactivity of these highly unsaturated
this case, however, it is possible to identify a reaction intermediate
which is characterized by NMR signals at dB 48.9 and dP 48.3. The
former resonance is consistent with values previously reported for
base-stabilized terminal borylene complexes (e.g. dB 51.7–53.2 for
N-donor adducts of osmium aminoborylenes),31 whereas the 31P
chemical shift is as expected for donor/acceptor adducts of Ph3PO
with boron-centred Lewis acids (dP 43.6–46.7).35 In addition, the
i
complexes, that we set about investigating the synthesis of
observations of inequivalent Pr groups by 1H and 13C NMR
+
=
cationic aminoborylene systems, [LnM BNR2] . The reactivity
and of significantly lower carbonyl stretching frequencies (2004,
1949 cm−1) are both consistent with the formation of a trigonal
planar boron centre by coordination of a Lewis base. Confirmation
that the intermediate species is indeed the B-bound Ph3PO adduct
of [(g -C5H5)Fe(CO)2(BNiPr2)]+ [BArf4]−, 9, towards a range of
5
nucleophilic and/or unsaturated reagents has been examined (see
Scheme 4), with examples of addition, protonolysis and metathesis
chemistries having been established.
5
[(g -C5H5)Fe(CO)2{B(NiPr2)(OPPh3)}]+[BArf4]−(13) has been ob-
The reactivity of 9 towards anionic nucleophiles, exemplified by
tained crystallographically (vide infra). Given the isolation of
13, it is plausible that the first step in the reaction mechanism
5
chloride, mirrors that observed for the arylborylene complex [(g -
C5Me5)Fe(CO)2(BMes)]+[BArf4]−, being characterized by addition
at the boron centre. Thus, reaction of 9 with [PPN]Cl regenerates
chloroboryl complex 7a, the identity of which was confirmed
by spectroscopic data indistinguishable from authentic samples
involves coordination of Ph3E X (E = P, As; X = O, S) at
=
boron, and that the overall metathesis chemistry of 9 therefore
occurs via a combined addition/substitution pathway (Scheme 5).
Such a proposal contrasts with the concerted metallacyclobutane
5
36
synthesized from Na[(g -C5H5)Fe(CO)2] and iPr2NBCl2. The reac-
mechanism known to operate for M C/C C metathesis, but
=
=
=
=
tivity of 9 towards water also displays similarities to the chemistry
observed for related M–B bonds towards protic reagents. Thus,
the reaction proceeds via scission of the Fe–B linkage,1,7 although
somewhat surprisingly this is also accompanied by breakage of
the B–N bond. The nitrogen-containing product isolated from
the reaction has been shown by spectroscopic and crystallo-
graphic studies (see ESI†) to be the diisopropylammonium salt
[H2NiPr2]+[BArf4]− (10); the boron- and iron-containing products
can be shown by comparison with previously reported data to
is plausible given the polar nature of the M B and E X bonds
involved, and the previously demonstrated propensity of cationic
borylenes to undergo addition at boron and substitution at the
metal centre.7
5
be B(OH)3 and [(g -C5H5)Fe(CO)2]2, respectively.16 B–N bond
breakage under these mild conditions is somewhat unexpected;
iPr2NBCl2 for example undergoes hydrolysis to yield a mixture of
(iPr2NBO)3 and (iPr2NBCl)2(l-O) under aqueous basic conditions
after prolonged stirring at room temperature.17b
Scheme 5 Proposed addition/substitution pathway for metathesis reac-
5
i
+
f
−
=
tions of [(g -C5H5)Fe(CO)2B(N Pr2)] [BAr 4] , 9 (exemplified by Ph3P S).
Although the reactivity of cationic aminoborylene complex 9
(iv) Spectroscopic and structural studies
towards Cl− is indicative of the electrophilic character known to
5
dominate the chemistry of [(g -C5Me5)Fe(CO)2(BMes)]+[BArf4]−,
Single-crystal X-ray diffraction studies were undertaken on com-
pounds 2a, 2d, 3a, 7a, 10 and 13; of these, the structure of
[H2NiPr2]+[BArf4]−, 10 was obtained predominantly for compound
verification and shows geometric features very similar to related
derivatives. Hence this structure together with that of 3a (which
closely resembles that of 2a) have been included only in the ESI.†
For the remaining compounds, details of data collection, structure
solution and refinement parameters are given in Table 1; relevant
bond lengths and angles are included in the figure captions.
Complete details of all structures are given in the ESI† and have
been deposited with the Cambridge Structural Database.
its reactivity towards unsaturated substrates suggests a broader
=
scope for its chemistry. Th◦us, the reaction of 9 with Ph3P S
5
in dichloromethane at 20 C leads to the formation of [(g -
C5H5)Fe(CO)2(PPh3)]+[BArf4]− (11a) and iPr2NB(l-S)2BNiPr2
1
(12a) with >95% conversion (as determined by H, 11B and 31P
NMR spectroscopies). The identities of the isolated products
11a and 12a were confirmed by comparison of multinuclear
NMR (1H, 11B, 13C, 19F and 31P), IR and mass spectrome-
try data with those reported for authentic samples.17,18 Simi-
5
=
lar reactivity towards Ph3As O leads to the isolation of [(g -
C5H5)Fe(CO)2(AsPh3)]+[BArf4]− (11b) and (iPr2NBO)3 (12b).17,18
The structures of mesityloxy(chloro)boryl complexes 2a and
3a (see Fig. 1 and ESI†) display the expected half sandwich
geometry at the iron centre with the coordination sphere being
completed by two carbonyls and the B(OMes)Cl ligand. The iron-
=
=
The reactions of 9 with Ph3P S and Ph3As O represent, to
our knowledge, the first examples of net metathesis chemistry
for a terminal borylene complex.33 Although metathesis chem-
5
˚
istry has been reported for isoelectronic vinylidene systems (g -
boron distances for 2a and 3a are relatively short [1.977(4) A for
34
C5H5)M(CO)2 C CH2 (M = Mn, Re), and for B C double
bonds,33 no such chemistry has been reported for neutral aminobo-
rylene complexes.28 The origins of the differing reactivity of 9 and
an idea of the likely mechanism can be gauged by examining the
both compounds];1 however in each case the orientation of the
boryl plane (i.e. that defined by the B, O and Cl atoms) relative
to that defined by the cyclopentadienyl centroid, Fe and B atoms
[torsion angles are 67.1(2) and 87.6(2)◦ for 2a and 3a, respectively]
and the relatively low carbonyl stretching frequencies (2002, 1940
and 1971, 1925 cm−1, respectively) indicate little or no Fe → B p
= =
=
=
analogous reactivity towards Ph3P O. This reaction proceeds at
=
=
a significantly slower rate than those with Ph3P S or Ph3As O,
406 | Dalton Trans., 2006, 399–410
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 399–410 | 407
©

View Article Online
5
Fig. 1 Structure of (g -C5H5)Fe(CO)2B(OMes)Cl, 2a. Hydrogen atoms
have been omitted for clarity and ORTEP ellipsoids set at the
◦
5
˚
50% probability level. Important bond lengths (A) and angles ( )
Fig. 2 Structure of (g -C5H5)Fe(CO)2B(OMes)SPh, 2d. Hydrogen atoms
for 2a: Fe(1)–B(1) 1.977(4), B(1)–Cl(1) 1.816(4), B(1)–O(3) 1.350(4);
B(1)–O(3)–C(8) 124.6(3), centroid-Fe(1)–B(1)–Cl(1) 67.1(2).
have been omitted for clarity and ORTEP ellipsoids se◦t at the 50%
˚
probability level. Important bond lengths (A) and angles ( ): Fe(1)–B(1)
2.034(4), B(1)–S(1) 1.848(4), B(1)–O(3) 1.349(5); B(1)–O(3)–C(8) 128.4(3),
B(1)–S(1)–C(17) 107.8(2), centroid-Fe(1)–B(1)–Cl(1) 57.0(3).
back bonding interaction from the metal–based HOMO.§ The
Fe–B bond length for 2a is similar to that measured for the
5
catecholboryl complex (g -C5H5)Fe(CO)2Bcat [cat = O2C6H4-1,2;
˚
precursor 2a [2.034(4) and 1.977(4) A, respectively], presumably
˚
1.959(6) A] but the higher carbonyl stretching frequencies (2024
reflecting, at least in part, the greater steric requirements of the
SPh substituent.
and 1971 cm−1) and the near zero torsion angle (7.9◦) measured for
the Bcat complex imply that, in contrast to 2a, this species features
a modest but significant HOMO-derived Fe→B p interaction.39
The torsion angle measured for 2a is similar to that ◦found for the
Due to its very high solubility in hexanes, it was not possible
to crystallize 7a by cooling even of very concentrated solutions.
It did prove possible, however, to obtain crystals of X-ray quality
by sublimation of the crude mixture at 40 ◦C under high vacuum
(10−4 Torr). The molecular structure of 7a (Fig. 3 and Table 1) re-
5
dichloroboryl complex (g -C5H5)Fe(CO)2BCl2 (77.7 ),40 implying
possible p overlap with the HOMO−2 molecular orbital of the
5
[(g -C5H5)Fe(CO)2]+ fragment.37 However, the significantly shorter
˚
veals a relatively long iron-boron distance [2.054(4) A] indicative of
˚
Fe–B length [1.942(3) A] for this compound compared with 2a is
little Fe–B p interaction in the solid state; this is further evidenced
by the near 90◦ torsion angle [∠centroid-Fe(1)–B(1)–N(1) =
72.5(5)◦]. Indeed, similar values are reported by Braunschweig for
presumably due to the greater p acceptor capability of the –BCl2
ligand compared to –B(OMes)Cl.40
Substitution of the chloride substituent in 2a with para
5
˚
(g -C5Me5)Fe(CO)2B(NMe2)Cl [Fe–B = 2.027(5) A, interplanar
5
tert-butyl phenoxide leads to the new boryl complex (g -
t
C5H5)Fe(CO)2B(OMes)(OC6H4 Bu-4), 2c, characterized by an
upfield shift in the 11B NMR to 47.4 ppm, indicative of replacement
of the chloride with a more p electron releasing phenoxide
group. In similar fashion, replacement of the chloride with a
thiophenolate group yields 2d, characterized by an 11B NMR
chemical shift of 69.1 ppm. The marked downfield shift of the
sulfur-substituted boryl complex 2d (compared to 2c) reflects
the greater Lewis acidity of the –B(OMes)SPh ligand compared
to –B(OMes)OC6H4 Bu-4 and is consistent with the 11B NMR
t
chemical shifts reported for the platinum(II) bis-boryl complexes
(Ph3P)2Pt(Bcat)2 (dB 47.0) and (Ph3P)2Pt{B(1,2-S2C6H4)}2 (dB
72.0).41 2d is only the second crystallographically characterized
example of a thio-substituted boryl complex,41b and is the first fully
characterized boryl to contain mixed oxygen/sulfur substituents
(Fig. 2). 2d has a significantly longer Fe–B bond distance than
5
§ The HOMO of the [(g -C5H5)Fe(CO)2]+ fragment is an orbital of aꢀꢀ
5
Fig. 3 Structure of (g -C5H5)Fe(CO)2B(NiPr2)Cl, 7a. Hydrogen atoms
symmetry lying approximately co-planar with the cyclopentadienyl ring;
the HOMO−2 is of aꢀ symmetry and is perpendicular to the HOMO.37 For
a recent analysis of p back-bonding in half-sandwich boryl systems of the
have been omitted for clarity and ORTEP ellipsoids set at t◦he
˚
50% probability level. Important bond lengths (A) and angles ( ):
5
Fe(1)–B(1) 2.054(4), B(1)–Cl(1) 1.841(4), B(1)–N(1) 1.389(5), cen-
troid-Fe(1)–B(1)–N(1) 83.7(4).
type (g -C5H5)Fe(CO)2BX2 as a function of the centroid-Fe–B–X torsion
angle, see ref. 38.
408 | Dalton Trans., 2006, 399–410
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
◦
5
angle = 87.4 ] and (g -C5H5)Fe(CO)2B(NMe2)B(NMe2)Cl [Fe–
B(NiPr2)Cl can be formed via salt elimination reactions of
the haloboranes MesOBX2 or iPr2NBCl2 with the appropri-
ate transition metal anion. Conversely, reaction of tmpBBr2
◦
˚
B = 2.090(3) A, interplanar angle = 92.4 ] for which little or no
Fe–B p back bonding is proposed.25,42 The relatively low carbonyl
stretching frequencies for 7a (2001, 1941 cm−1), are similar to
5
with Na[(g -C5H5)Fe(CO)2] yields only the bridging borylene
5
5
those for (g -C5H5)Fe(CO)2B(E{SiMe3}3)Cl (2009, 1956 cm−1
species [(g -C5H5)Fe(CO)]2(l-CO)(l-Btmp) seemingly indepen-
5
5
for both E = Si, Ge), and (g -C5H5)Fe(CO)2B(NMe2)Cl (2006,
dent of reaction conditions. (g -C5H5)Fe(CO)2B(OMes)Cl is
1946 cm−1) and give further evidence that there is little Fe–B
found to be a versatile substrate for substitution chemistry
with retention of the Fe–B bond, leading to the formation of
p back bonding.22,25 As found for related derivatives,25,42 the B–
5
˚
N bond distance for 7a is relatively short [1.389(5) A], revealing
appreciable p bond character in the solid state; this increased B–N
bond order is also apparent in solution—the H and 13C NMR
spectra for 7a show two sets of signals for the isopropyl CH and
CH3 groups, consistent with slow rotation about the B–N bond on
the NMR timescale.
new asymmetric boryl complexes (g -C5H5)Fe(CO)2B(OMes)ERn
t
(ERn = OC6H4 Bu-4 or SPh), the latter complex being the first
1
crystallographically characterized example of a boryl species
containing mixed oxygen/sulfur substituents. The reaction of
5
complexes (g -C5R5)Fe(CO)2B(OMes)Cl (R = H or Me) with
5
Na[BArf4] leads to the formation of fluoroboryl complexes (g -
5
5
The structure of the Ph3PO adduct [(g -C5H5)Fe(CO)2-
C5R5)Fe(CO)2B(OMes)F and (g -C5Me5)Fe(CO)2BF2, presum-
{B(NiPr2)(OPPh3)}]+[BArf4]−(13) has also been determined (see
ably due to halide abstraction from [BArf4]− by the putative
Fig. 4 and Table 1). Consistent with related complexes,31,43 the Fe–
cationic terminal borylene species [(g -C5R5)Fe(CO)2BOMes]+.
5
5
B bond length for 13 [2.057(4) A] is more akin to that expected
Application of halide abstraction to (g -C5H5)Fe(CO)2B(NiPr2)Cl
˚
for a single bond than a double bond {cf. 2.054(4) and 1.792(8)
on the other hand yields the cationic terminal borylene complex
5
+
6,7
5
[(g -C5H5)Fe(CO)2BNiPr2]+[BArf4]−, which displays a range of
˚
A for 7a and [(g -C5Me5)Fe(CO)2(BMes)] , respectively}. Such
a phenomenon has previously been ascribed to significant con-
tributions from resonance forms incorporating a formal M–B
single bond (as in Scheme 5),31 and a description of 13 as an
amino(oxy)boryl species featuring a pendant cationic phosphorus
centre is probably most apt. Consistent with this, the PO distance
novel reaction chemistries including metathesis.
Acknowledgements
We would like to acknowledge financial support from the EPSRC
and assistance from the EPSRC National Mass Spectroscopy
Service Centre, University of Wales. We would also like to thank
Professor Cameron Jones (Cardiff) for assistance in modelling
crystallographic disorder.
in 13 is significantly longer then that found in free Ph3PO [1.540(2)
44
˚
vs. 1.493 A (mean)].
References
1 For recent reviews of boryl and borylene chemistry, see: (a) H.
Wadepohl, Angew. Chem., Int. Ed. Engl., 1997, 36, 2441; (b) G. J. Irvine,
M. J. G. Lesley, T. B. Marder, N. C. Norman, C. R. Rice, E. G. Robins,
W. R. Roper, G. R. Whittell and L. J. Wright, Chem. Rev., 1998, 98, 2685;
(c) H. Braunschweig, Angew. Chem., Int. Ed., 1998, 37, 1786; (d) M. R.
Smith, III, Prog. Inorg. Chem., 1999, 48, 505; (e) B. Wrackmeyer, Angew.
Chem., Int. Ed., 1999, 38, 771; (f) H. Braunschweig and M. Colling,
J. Organomet. Chem., 2000, 614–615, 18; (g) H. Braunschweig and M.
Colling, Coord. Chem. Rev., 2001, 223, 1; (h) H. Braunschweig and
M. Colling, Eur. J. Inorg. Chem., 2003, 393; (i) S. Aldridge and D. L.
Coombs, Coord. Chem. Rev., 2004, 248, 535; (j) H. Braunschweig, Adv.
Organomet. Chem., 2004, 51, 163.
2 For reviews of relevant hydroboration chemistry, see: (a) K. Burgess and
M. J. Ohlmeyer, Chem. Rev., 1991, 91, 1179; (b) I. Beletskaya and A.
Pelter, Tetrahedron, 1997, 53, 4957; (c) C. M. Crudden and D. Edwards,
Eur. J. Org. Chem., 2003, 4695.
3 For reviews of relevant diboration chemistry, see: (a) T. B. Marder and
N. C. Norman, Top. Catal., 1998, 5, 63; (b) T. Ishiyama and N. Miyaura,
J. Organomet. Chem., 2000, 611, 392.
4 For a recent review of relevant C–H functionalization chemistry, see:
T. Ishiyama and N. Miyaura, J. Organomet. Chem., 2003, 680, 3.
5 See, for example: S. K. Grumbine, T. D. Tilley, F. P. Arnold and A. L.
Rheingold, J. Am. Chem. Soc., 1994, 116, 5495.
6 D. L. Coombs, S. Aldridge, C. Jones and D. J. Willock, J. Am. Chem.
Soc., 2003, 125, 6356.
7 D. L. Coombs, S. Aldridge, A. Rossin, C. Jones and D. J. Willock,
Organometallics, 2004, 23, 2911.
8 N. R. Bunn, S. Aldridge, D. L. Coombs, A. Rossin, D. J. Willock, C.
Jones and L.-L. Ooi, Chem. Commun., 2004, 1732.
5
Fig. 4 Structure of the cationic component of [(g -C5H5)Fe(CO)2-
{B(NiPr2)(OPPh3)}]+[BArf4]−, 13. Hydrogen atoms have been omitted
for clarity and ORTEP ellipsoids set at the 50% probability level.
◦
˚
Important bond lengths (A) and angles ( ): Fe(1)–B(1) 2.057(4), B(1)–O(3)
1.469(4), B(1)–N(1) 1.397(5), P(1)–O(3) 1.540(2); P(1)–O(1)–B(1) 148.0(2),
centroid-Fe(1)–B(1)–N(1) 72.5(5).
9 N. R. Bunn, S. Aldridge, D. L. Kays, N. D. Coombs, A. Rossin,
D. J. Willock, J. K. Day, C. Jones and L.-L. Ooi, Organometallics,
DOI: 10.1021/om0506318.
4
Conclusions
10 For a preliminary account of part of this work, see: D. L. Kays,
J. K. Day, L.-L. Ooi and S. Aldridge, Angew. Chem., Int. Ed., 2005,
DOI: 10.1002/anie.200502343.
Heteroatom stabilized asymmetric haloboryl complexes of
the types (g -C5R5)Fe(CO)2B(OMes)X and (g -C5H5)Fe(CO)2-
5
5
This journal is The Royal Society of Chemistry 2006
Dalton Trans., 2006, 399–410 | 409
©

View Article Online
11 R. B. King and M. B. Bisnette, J. Organomet. Chem., 1967, 8, 287; R. B.
King, Acc. Chem. Res., 1970, 3, 417.
12 W. Gerrard, H. R. Hudson and E. F. Mooney, J. Chem. Soc., 1960,
5168.
27 The synthetic viability of group 13 ligand systems containing tmp
substituents (albeit for a larger aluminium donor atom) has previously
5
been demonstrated by the synthesis of (g -C5H5)Fe(CO)2Al(tmp)2:
B. N. Anand, I. Krossing and H. No¨th, Inorg. Chem., 1997, 36, 1979.
28 (a) H. Braunschweig, C. Kollann and U. Englert, Angew. Chem., Int.
Ed., 1998, 37, 3179; (b) H. Braunschweig, M. Colling, C. Kollann,
H. G. Stammler and B. Neumann, Angew. Chem., Int. Ed., 2001, 40,
2298; (c) H. Braunschweig, M. Colling, C. Hu and K. Radacki, Angew.
Chem., Int. Ed., 2003, 42, 205.
29 S. Aldridge, A. Rossin, D. L. Coombs and D. J. Willock, Dalton Trans.,
2004, 2649.
30 C. Knors, G. H. Kuo, J. W. Lauher, C. Eigenbrot and P. Helquist,
Organometallics, 1987, 6, 988.
13 J. Escudie, C. Couret, M. Lazraq and B. Garrigues, Synth. React. Inorg.
Met.-Org. Chem., 1987, 379.
14 D. L. Reger, T. D. Wright, C. A. Little, J. J. S. Lamba and M. D. Smith,
Inorg. Chem., 2001, 40, 3810.
5
15 R. J. Calder, PhD thesis, Cardiff University, 2002. Data for (g -
5
C5Me5)Fe(CO)2BF2, 4: 1H NMR (300 MHz, C6D6) d 1.57 (s, 15H, g -
5
5
C5Me513C NMR (76 MHz, CD2Cl2) d 9.5 (g -C5Me5), 95.9 (g -C5Me5
quaternary). 11B NMR (96 MHz, C6D6) d 47.8 (t, J = 201 Hz). 19F
NMR (283 MHz, C6D6) d −3.4 (q, J = 205 Hz). IR (C6D6 soln, cm−1
1994, 1933.
31 (a) G. J. Irvine, C. E. F. Rickard, W. R. Roper, A. Williamson and L. J.
Wright, Angew. Chem., Int. Ed., 2000, 39, 948; (b) C. E. F. Rickard,
W. R. Roper and A. Williamson, Organometallics, 2002, 21, 4862.
32 H. Braunschweig, personal communication.
16 (a) G. R. Eaton and W. N. Lipscomb, in NMR Studies of Boron Hydrides
and Related Species, Benjamin, 1969; (b) R. F. Bryan, P. T. Greene, M. J.
Newlands and D. S. Field, J. Chem. Soc. A, 1970, 3068.
17 (a) H. No¨th and W. Rattay, J. Organomet. Chem., 1986, 312, 139; (b) W.
Maringgele, M. Noltemeyer and A. Meller, Organometallics, 1997, 16,
2276.
=
33 Metathesis chemistry has been reported for C B double bonds, e.g. P.
Paetzold, U. Englert, R. Finger, T. Schmitz and A. Tapper, Z. Anorg.
Allg. Chem., 2004, 630, 508.
18 H. Schumann and L. Eguren, J. Organomet. Chem., 1991, 403, 183.
34 See, for exampleM. R. Terry, L. A. Mercando, C. Kelley, G. L. Geoffroy,
P. Nombel, N. Lugan, R. Mathieu, R. L. Ostrander, B. E. Owens-
Waltermire and A. L. Rheingold, Organometallics, 1994, 13, 843.
35 (a) N. Burford, B. W. Royan, R. E. v. H. Spence, T. S. Cameron, A.
Linden and R. D. Rogers, J. Chem. Soc., Dalton Trans., 1990, 1521;
(b) G. J. P. Britovsek, J. Ugolotti and A. J. P. White, Organometallics,
2005, 24, 1685.
5
5
19 The conversion of [(g -C5H5)Fe(CO)2(PPh3)]+ into dinuclear (g -
C5H52Fe2(CO)3(PPh3) has previously been reported: S. Byrne, M. G.
Cox and A. R. Manning, J. Organomet. Chem., 2001, 629, 182.
20 (a) H. K. Hofmeister and J. R. Van Wazer, J. Inorg. Nucl. Chem., 1964,
26, 1201; (b) H. K. Hofmeister and J. R. Van Wazer, J. Inorg. Nucl.
Chem., 1964, 26, 1209.
21 (a) D. L. Coombs, S. Aldridge and C. Jones, J. Chem. Soc., Dalton
Trans., 2002, 3851; (b) D. L. Coombs, S. Aldridge and C. Jones, Appl.
Organomet. Chem., 2003, 17, 356.
22 H. Braunschweig, M. Colling, C. Kollann and U. Englert, J. Chem.
Soc., Dalton Trans., 2002, 2289.
23 (a) S. R. Bahr and P. Boudjouk, J. Am. Chem. Soc., 1993, 115, 4514;
(b) D. Ferraris, C. Cox, R. Anand and T. Lectka, J. Am. Chem. Soc.,
1997, 119, 4319; (c) R. P. Hughes, T. L. Husebo and S. M. Maddock,
J. Am. Chem. Soc., 1997, 119, 10231; (d) W. E. Buschmann and J. S.
Miller, Chem.–Eur. J., 1998, 4, 1731.
24 (a) A. Kerr, T. B. Marder, N. C. Norman, A. G. Orpen, M. J.
Quayle, C. R. Rice, P. L. Timms and G. R. Whittell, Chem. Commun.,
1998, 319; (b) N. Lu, N. C. Norman, A. G. Orpen, M. J. Quayle,
P. L. Timms and G. R. Whittell, J. Chem. Soc., Dalton Trans., 2000,
4032.
36 See, for example: T. M. Trnka and R. H. Grubbs, Acc. Chem. Res.,
2002, 34, 18.
37 B. E. R. Schilling, R. Hoffmann and D. Lichtenberger, J. Am. Chem.
Soc., 1979, 101, 585.
38 A. A. Dickinson, D. J. Willock, R. J. Calder and S. Aldridge,
Organometallics, 2002, 21, 1146.
39 J. F. Hartwig and S. Huber, J. Am. Chem. Soc., 1993, 115, 4908.
40 H. Braunschweig, K. Radacki, F. Seeler and G. R. Whittell,
Organometallics, 2004, 23, 4178.
41 (a) G. Lesley, P. Nguyen, N. J. Taylor, T. B. Marder, A. J. Scott, W. Clegg
and N. C. Norman, Organometallics, 1996, 15, 5137; (b) W. Clegg, F. J.
Lawlor, G. Lesley, T. B. Marder, N. C. Norman, A. G. Orpen, M. J.
Quayle, C. R. Rice, A. J. Scott and F. E. S. Souza, J. Organomet. Chem.,
1998, 550, 183.
42 H. Braunschweig, B. Ganter, M. Koster and T. Wagner, Chem. Ber.,
1996, 129, 1099.
25 (a) H. Braunschweig, C. Kollann and U. Englert, Eur. J. Inorg. Chem.,
1998, 465; (b) H. Braunschweig, C. Kollann and K. W. Klinkhammer,
Eur. J. Inorg. Chem., 1999, 1523.
43 H. Braunschweig, D. Rais and K. Uttinger, Angew. Chem., Int. Ed.,
2005, 44, 3763.
26 D. L. Coombs, S. Aldridge, S. J. Coles and M. B. Hursthouse,
Organometallics, 2003, 22, 4213.
44 C. P. Brock, W. B. Schweizer and J. D. Dunitz, J. Am. Chem. Soc., 1985,
107, 6964.
410 | Dalton Trans., 2006, 399–410
This journal is
The Royal Society of Chemistry 2006
©