Borylene-based functionalization of iron-alkynyl-σ-complexes and stepwise reversible metal-boryl-to-borirene transformation: Synthesis, characterization, and density functional theory studies

Article abstract of DOI:10.1021/ic1010874

Thermally induced chemoselective borylene transfer from [(OC)₅Mo = BN(SiMe₃)₂] (2a) to the carbon-carbon triple bond of an iron dicarbonyl alkynyl complex [(η⁵-C₅Me ₅)Fe(CO)₂C≡CPh] (3) led to the isolation of an iron aminoborirene complex [(η⁵-C₅Me₅)(OC) ₂Fe{μ-BN(SiMe₃)₂C = C}Ph] (4) in satisfactory yield. Room temperature photolysis of 4 resulted in an unprecedented rearrangement and a concurrent decarbonylation, affording the novel C₂ side-on coordinated iron boryl complex [(η⁵- C₅Me₅)(OC)FeBN(SiMe₃)₂(η ²-CC)Ph] (5). Carbonylation of 5 under CO atmosphere at ambient temperature yielded [(η⁵-C₅Me₅)(OC) ₂FeBN(SiMe₃)₂CCPh] (6), which is an isomer of 4. Decarbonylation of 6 at 80 °C led to 5, which could be upon introduction of CO gas further converted into 4 under same conditions. Reaction of 5 with PMe₃ at 80 °C yielded the phosphane complex [(η⁵- C₅Me₅)(OC)(PMe₃)Fe{μ-BN(SiMe ₃)₂C = C}Ph] (7). All above-mentioned iron complexes 4-7 were isolated as air and moisture sensitive crystalline solids in good yields and have been fully characterized in solution and by X-ray crystallography. Quantum chemical calculations using density functional theory (DFT) have been carried out to understand the mechanisms of the experimentally observed reactions and to analyze the bonding situation in the molecules 4-7.

Full text of DOI:10.1021/ic1010874

62 Inorg. Chem. 2011, 50, 62–71
DOI: 10.1021/ic1010874
Borylene-Based Functionalization of Iron-Alkynyl-σ-Complexes and Stepwise
Reversible Metal-Boryl-to-Borirene Transformation: Synthesis, Characterization,
and Density Functional Theory Studies
Holger Braunschweig,* Qing Ye, Krzysztof Radacki, and Peter Brenner
€
€
Institut fu€r Anorganische Chemie, Bayerische Julius-Maximilians-Universitat Wurzburg Am Hubland,
97 074 Wu€rzburg, Germany
Gernot Frenking* and Susmita De
€
Fachbereich Chemie, Philipps-Universitat Marburg Hans-Meerwein-Strasse, 35043 Marburg, Germany
Received June 1, 2010
Thermally induced chemoselective borylene transfer from [(OC)5ModBN(SiMe3)2] (2a) to the carbon-carbon triple
bond of an iron dicarbonyl alkynyl complex [(η5-C5Me5)Fe(CO)2CtCPh] (3) led to the isolation of an iron
aminoborirene complex [(η5-C5Me5)(OC)2Fe{μ-BN(SiMe3)2CdC}Ph] (4) in satisfactory yield. Room temperature
photolysis of 4 resulted in an unprecedented rearrangement and a concurrent decarbonylation, affording the novel C2
side-on coordinated iron boryl complex [(η5-C5Me5)(OC)FeBN(SiMe3)2(η2-CC)Ph] (5). Carbonylation of 5 under CO
atmosphere at ambient temperature yielded [(η5-C5Me5)(OC)2FeBN(SiMe3)2CCPh] (6), which is an isomer of 4.
Decarbonylation of 6 at 80 °C led to 5, which could be upon introduction of CO gas further converted into 4 under same
conditions. Reaction of 5 with PMe3 at 80 °C yielded the phosphane complex [(η5-C5Me5)(OC)(PMe3)Fe{μ-
BN(SiMe3)2CdC}Ph] (7). All above-mentioned iron complexes 4-7 were isolated as air and moisture sensitive
crystalline solids in good yields and have been fully characterized in solution and by X-ray crystallography. Quantum
chemical calculations using density functional theory (DFT) have been carried out to understand the mechanisms of
the experimentally observed reactions and to analyze the bonding situation in the molecules 4-7.
Introduction
case of flouroborylene “:B-F” already in the 1960s.1 Like-
wise, in 1984, West et al. published the photochemical
generation of the silylborylene “:B-SiPh3” in hydrocarbon
matrixes at -196 °C and a number of well characterized trap-
ping products derived thereof.2 During the past decade
borylene chemistry has been put into focus again, as it
became possible to generate and stabilize borylenes as ligands
in the coordination sphere of various transition metals under
standard conditions.3
Free borylenes of the general type “:B-R” constitute hypo-
valent, highly reactive species that can only be obtained by
applying drastic conditions as shown by Timms et al. in the
*To whom correspondence should be addressed. E-mail: h.braunschweig@
mail.uni-wuerzburg.de (H.B.); frenking@see.foot.note (G.F.). Fax: (þ49) 931-
888-4623 (H.B.); þ49-(0)6421-28-25566 (G.F.).
(1) (a) Timms, P. L. J. Am. Chem. Soc. 1967, 89, 1629. (b) Timms, P. L. Acc.
Chem. Res. 1973, 6, 118.
The ligand properties of borylenes have been closely
studied, in particular by computational methods,4,5 thus re-
vealing a close relationship to the isolectronic carbonyls in
terms of bonding pattern and coordination modes. However,
(2) Pachaly, B.; West, R. Angew. Chem. 1984, 96, 444. Pachaly, B.; West, R.
Angew. Chem., Int. Ed. Engl. 1984, 23, 454.
(3) (a) Braunschweig, H. Angew. Chem. 1998, 110, 1882. Braunschweig, H.
Angew. Chem., Int. Ed. 1998, 37, 1786. (b) Wrackmeyer, B. Angew. Chem.
1999, 111, 817. Wrackmeyer, B. Angew. Chem., Int. Ed. 1999, 38, 771.
(c) Braunschweig, H.; Colling, M. J. Organomet. Chem. 2000, 18, 614.
(d) Braunschweig, H.; Colling, M. Coord. Chem. Rev. 2001, 223, 1.
(e) Braunschweig, H.; Colling, M. Eur. J. Inorg. Chem. 2003, 393. (f) Aldridge,
S.; Coombs, D. L. Coord. Chem. Rev. 2004, 248, 535. (g) Braunschweig, H. Adv.
Organomet. Chem. 2004, 51, 163. (h) Braunschweig, H.; Whittell, G. R.
Chem.—Eur. J. 2005, 11, 6128. (i) Braunschweig, H.; Rais, D. Heteroat. Chem.
2005, 16, 566. (j) Braunschweig, H.; Kollann, C.; Rais, D. Angew. Chem. 2006,
118, 5380. Braunschweig, H.; Kollann, C.; Rais, D. Angew. Chem., Int. Ed.
2006, 45, 5252. (k) Anderson, C. E.; Braunschweig, H.; Dewhurst, R. D.
Organometallics 2008, 27, 6381. (l) Vidovic, D.; Pierce, G. A.; Aldridge, S.
Chem. Commun. 2009, 1157.
(4) (a) Radius, U.; Bickelhaupt, F. M.; Ehlers, A. W.; Goldberg, N.;
Hoffmann, R. Inorg. Chem. 1998, 37, 1080. (b) Macdonald, C. L. B.; Cowley,
A. H. J. Am. Chem. Soc. 1999, 121, 12113. (c) Uddin, J.; Boehme, C.; Frenking,
G. Organometallics 2000, 19, 571. (d) Chen, Y.; Frenking, G. Dalton Trans.
2001, 434. (e) Uddin, J.; Frenking, G. J. Am. Chem. Soc. 2001, 123, 1683.
(f) Boehme, C.; Uddin, J.; Frenking, G. Coord. Chem. Rev. 2000, 197, 249.
(g) Xu, L.; Li, Q.-S.; Xie, Y.; King, R. B.; Schaefer, H. F. Inorg. Chem. 2010, 49,
1046. (h) Pandey, K. K.; Musaev, D. G. Organometallics 2010, 29, 142.
(5) Ehlers, A. W.; Baerends, E. J.; Bickelhaupt, F. M.; Radius, U.
Chem.—Eur. J. 1998, 4, 210.
r
pubs.acs.org/IC
Published on Web 08/16/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 50, No. 1, 2011 63
borylenes possess improved σ-donor and π-acceptor proper-
ties, thus leading to increased thermodynamic stability of the
coordination products with respect to homolytic cleavage of
the metal-borylene multiple bond.5
Scheme 1. Photoisomerization of Dimesityl(mesitylethynyl)borane in
Donor Solvent
Since the most general approach to group 6 terminal
borylene complexes [(OC)5MdB=N(SiMe3)2] (M=Cr
2b, W 2c) was reported in 19986 and subsequently extended to
various central metals and boron bound substituents7,3e
various of their reactivity patterns have been investigated.3
Most notably, these compounds have turned out to be con-
venient sources for borylenes B-R (R=N(SiMe3)2) that
could be efficiently transferred to suitable borylene acceptors
under standard conditions. Thus, transmetalation of boryl-
enes has been established,8 as well as borylene transfer to
unsaturated organic substrates such as alkenes9 and alky-
nes10 became possible for the first time. The latter, that is, the
borylene based functionalization of carbon-carbon triple
bonds, provides a facile access to borirenes, which are
isoelectronic with cyclopropenium cations, and thus, consti-
tute the smallest boron heterocycle that might exhibit 2π-
aromatic stabilization.11 Despite considerable fundamental
interest in aromatic and antiaromatic unsaturated boron-
containing heterocycles such as borirenes,11 boroles,12 and
borepins,13 only a limited number of synthetic routes to bori-
renes have been published,2,11d,14 and among these, most are
laborious and low yielding. Eisch et al. reported in 1987 the
first structurally characterized borirene compound 1, which
was synthesized via photoisomerization of diaryl(arylethy-
nyl)boranes (Scheme 1), and thus, provided experimental
evidence, that is, shortened B-C and elongated CdC bond
€
lengths, for the theoretically predicted Huckel aromaticity of
this class of compounds.14b,c Nonetheless, the scope of syn-
thetic approaches to borirenes is strongly limited with respect
to the available substituents at the boron atom.
The borylene based functionalization of carbon-carbon
triple bonds upon photolysis that we reported in 2005 pro-
vides a high yielding, straightforward access to borirene com-
pounds.10a In our further studies, this has turned out to be
applicable to a wide range of substrates such as alkynes,
diynes10b (Scheme 2), and thus a variety of borirene com-
pounds has been synthesized and fully characterized. The
proposed aromaticity of borirenes and extensive π-delocali-
zation over the BCC-ring has been confirmed by altered
endocyclic bond lengths and a significantly decreased barrier
to rotation about the exocyclic B-N double bond. This syn-
thetic strategy gives access to a new class of boron-based π-
conjugated systems with particularly interesting photophysi-
cal properties.15 As the introduction of a metal center into the
π-conjugated chain may introduce a range of different pro-
perties with regard to redox, magnetic, optical and electronic
characteristics,16 we recently reported the borylene based
functionalization of σ-alkynyl complexes of platinum, which
yielded such molecular frameworks involving d-block metals.17
Despite a somewhat broader range of well characterized
borirenes our vision of reactivity of these three-membered
(6) Braunschweig, H.; Kollann, C.; Englert, U. Angew. Chem. 1998, 110,
3355. Braunschweig, H.; Kollann, C.; Englert, U. Angew. Chem., Int. Ed. 1998,
37, 3179.
(7) (a) Braunschweig, H.; Colling, M.; Kollann, C.; Merz, K.; Radacki, K.
Angew. Chem. 2001, 113, 4327. Braunschweig, H.; Colling, M.; Kollann, C.;
Merz, K.; Radacki, K. Angew. Chem., Int. Ed. 2001, 38, 4198. (b) Braunschweig,
H.; Radacki, K.; Scheschkewitz, D.; Whittell, G. R. Angew. Chem. 2005, 117,
1685. Braunschweig, H.; Radacki, K.; Scheschkewitz, D.; Whittell, G. R. Angew.
Chem., Int. Ed. 2005, 44, 1658 (VIP). (c) Blank, B.; Braunschweig, H.; Colling-
Hendelkens, M.; Kollann, C.; Radacki, K.; Rais, D.; Uttinger, K.; Whittel, G.
Chem.-Eur. J. 2007, 13, 4770.
(8) (a) Braunschweig, H.; Colling, M.; Kollann, C.; Neumann, B.;
Stammler, H.-G. Angew. Chem. 2001, 113, 2359. Braunschweig, H.; Colling,
M.; Kollann, C.; Neumann, B.; Stammler, H.-G. Angew. Chem., Int. Ed. 2001,
40, 2298. (b) Braunschweig, H.; Colling, M.; Hu, C.; Radacki, K. Angew. Chem.
2003, 115, 215. Braunschweig, H.; Colling, M.; Hu, C.; Radacki, K. Angew.
Chem., Int. Ed. 2003, 42, 205. (c) Braunschweig, H.; Forster, M.; Radacki, K.
Angew. Chem. 2006, 118, 8036. Braunschweig, H.; Forster, M.; Radacki, K.
Angew. Chem., Int. Ed. 2006, 45, 2132. (d) Braunschweig, H.; Forster, M.;
Radacki, K.; Seeler, F.; Whittell, G. Angew. Chem. 2007, 119, 5304. Braunsch-
weig, H.; Forster, M.; Radacki, K.; Seeler, F.; Whittell, G. Angew. Chem., Int. Ed.
2007, 46, 5212. (e) Braunschweig, H.; Forster, M.; Kupfer, T.; Seeler, F. Angew.
Chem. 2008, 120, 6070. Braunschweig, H.; Forster, M.; Kupfer, T.; Seeler, F.
Angew. Chem., Int. Ed. 2008, 47, 5981.
(9) Braunschweig, H.; Dewhurst, R. D.; Herbst, T.; Radacki, K. Angew.
Chem. 2008, 120, 6067. Braunschweig, H.; Dewhurst, R. D.; Herbst, T.; Radacki,
K. Angew. Chem., Int. Ed. 2008, 47, 5978.
(10) (a) Braunschweig, H.; Herbst, T.; Rais, D.; Seeler, F. Angew. Chem.
2005, 117, 7627. Braunschweig, H.; Herbst, T.; Rais, D.; Seeler, F. Angew.
Chem., Int. Ed. 2005, 44, 7461. (b) Braunschweig, H.; Herbst, T.; Rais, D.;
Ghosh, S.; Kupfer, T.; Radacki, K.; Crawford, A.; Ward, R.; Marder, T.; Fern ndez,
I.; Frenking, G. J. Am. Chem. Soc. 2009, 131, 8989. (c) Braunschweig, H.;
Fernndez, I.; Frenking, G.; Radacki, K.; Seeler, F. Angew. Chem. 2007, 119,
5307. Braunschweig, H.; Fernndez, I.; Frenking, G.; Radacki, K.; Seeler, F.
Angew. Chem., Int. Ed. 2007, 46, 5215.
(11) (a) Volpin, M. E.; Koreshkov, Y. D.; Dulova, V. G.; Kursanov,
D. N. Tetrahedron 1962, 18, 107; For INDO calculations, see:. (b) Pittman,
C. U.; Kress, A.; Patterson, T. B.; Walton, P.; Kispert, L. D. J. Org. Chem. 1974,
39, 373. (c) Allinger, N. L.; Siefert, J. H. J. Am. Chem. Soc. 1975, 97, 752; For ab
initio calculations, see:. (d) Krogh-Jespersen, K.; Cremer, D.; Dill, J. D.; Pople,
J. A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1981, 103, 2589.
(13) (a) Eisch, J. J.; Galle, J. E. J. Am. Chem. Soc. 1975, 97, 4436. (b) Ashe,
A. J., III; Drone, F. J. J. Am. Chem. Soc. 1987, 109, 1879. (c) Sugihara, Y.; Yagi,
T.; Murata, I.; Imamura, A. J. Am. Chem. Soc. 1992, 114, 1479. (d) Ashe, A. J.,
III; Kampf, J. W.; Klein, W.; Rousseau, R. Angew. Chem. 1993, 105, 1112. Ashe,
A. J., III; Kampf, J. W.; Klein, W.; Rousseau, R. Angew. Chem., Int. Ed. 1993,
32, 1065. (e) Schulman, J.; Disch, R. L. Organometallics 2000, 19, 2932.
(f) Mercier, L. G.; Piers, W. E.; Parvez, M. Angew. Chem., Int. Ed. 2009, 48, 6108.
(14) (a) Pues, C.; Berndt, A. Angew. Chem. 1984, 96, 306. Pues, C.; Berndt,
A. Angew. Chem., Int. Ed. Engl. 1984, 23, 313. (b) Eisch, J. J.; Shafii, B.;
Rheingold, A. L. J. Am. Chem. Soc. 1987, 109, 2526. (c) Eisch, J. J.; Shafii, B.;
Odom, J. D.; Rheingold, A. L. J. Am. Chem. Soc. 1990, 112, 1847.
(15) (a) Matsumi, N.; Chujo, Y. In Contemporary Boron Chemistry, Spec.
Publ. No. 253; Davidson, M. G., Hughes, A. K., Marder, T. B., Wade, K., Eds.;
Royal Society of Chemistry: Cambridge, 2000; pp 51-58; (b) Entwistle, C. D.;
Marder, T. B. Angew. Chem. 2002, 114, 3051. Entwistle, C. D.; Marder, T. B.
(12) (a) Eisch, J. J.; Hota, N. K.; Kozima, S. J. J. Am. Chem. Soc. 1969, 91,
4575. (b) Eisch, J. J.; Galle, J. E.; Kozima, S. J. Am. Chem. Soc. 1986, 108, 379.
(c) Schleyer, P. v. R.; Freeman, P. K.; Jiao, H.; Goldfuss, B. Angew. Chem. Int.
€
Angew. Chem. 2002, 41, 2927. (c) Jackle, F. J. Inorg. Organomet. Polym.
Mater. 2005, 15, 293. (d) Gabel, D. In Science of Synthesis: Houben-Weyl
Methods of Molecular Transformation; Kaufmann, D., Matteson, D. S., Eds.;
Verlag G. T. Thieme: Stuttgart, 2005; Vol. 6, p 1277.
(16) Long, N. J.; Williams, C. K. Angew. Chem. 2003, 115, 2690. Long,
N. J.; Williams, C. K. Angew. Chem., Int. Ed. 2003, 42, 2586.
(17) Braunschweig, H.; Ye, Q.; Radacki, K. Chem. Commun. 2009, 6979.
ꢀ
Ed. 1995, 34, 337. (d) Braunschweig, H.; Fernandez, I.; Frenking, G.; Kupfer, T.
ꢀ
Angew. Chem. 2008, 120, 1977. Braunschweig, H.; Fernandez, I.; Frenking, G.;
Kupfer, T. Angew. Chem., Int. Ed. 2008, 47, 1951. (e) Braunschweig, H.; Kupfer,
T. Chem. Commun. 2008, 4487. (f) Braunschweig, H.; Chiu, C.-W.; Radacki, K.;
Brenner, P. Chem. Commun. 2010, 916.

64 Inorganic Chemistry, Vol. 50, No. 1, 2011
Braunschweig et al.
Scheme 2. Borylene Based Functionalization of Alkynes, Diynes, and σ-alkynyl Complexes of Platinum upon Photolysis
Scheme 3. Synthesis of the Iron-Substituted Borirene 4
rings is still very limited and restricted to ring-opening with
cleavage of one endocyclic B-C bond, initiated by either
protic reagents, for example, water, methanol, or ethanol as
reported already by Eisch et al.14b,c or by hydroboration with
9-borabicyclo-[3.3.1]nonane (9-BBN) as recently shown in
our laboratory.18 In addition, the above-mentioned transfer
reaction requires activation by UV-light, thus strongly limit-
ing its scope of application to photostable substrates.
Herein we report a thermally induced chemoselective
borylene transfer from [(OC)5ModBN(SiMe3)2] (2a) to the
carbon-carbon triple bond of an iron dicarbonyl alkynyl
complex [(η5-C5Me5)Fe(CO)2CtCPh] (3), in which the iron
atom acts as a competitive borylene acceptor. In addition an
unprecedented reversible transformation between an iron-
borirene and iron-boryl complex is described.
spectroscopy that revealed a complete conversion of the
borylene complex 2a (δ = 89.1 ppm) into the iron bori-
rene species 4 (δ = 36.1 ppm). This result is in sharp con-
trast to initial experiments that we conducted under
photochemical conditions. Here, we attempted to obtain
4 by irradiating an equimolar mixture of 3 and the chrom-
ium borylene complex [(CO)5CrdBdN(SiMe3)2] (2b) in
toluene or THF in analogy to previously observed bor-
ylene transfer reactions to alkynes.10 However, in con-
trast to these clean and selective reactions, photolysis of
2b with 3 gave intractable mixtures of various boron con-
taining species (vide infra). According to Scheme 3
though, the iron borirene 4 was isolated as a yellow,
moderately air- and moisture sensitive solid material in
sufficient yield (63%) after work up. Multinuclear NMR
spectra of 4 were unobtrusive, displaying all relevant
signals in the expected range, with the exception of those
of the boron bound carbon atoms that were not observed
because of quadrupolar coupling, thus confirming its
constitution in solution. Most notably, 1H NMR spectra
revealed a single resonance at δ = 0.32 ppm for the
nitrogen-bound SiMe3 groups, thus indicating rapid ro-
tation around the B-N bond at room temperature. This
finding implies the presence of 2π-aromatic stabilization
within the BCC-ring that reduces the B-N π-contribu-
tion, as observed before in related systems.10,14b,c
Results and discussions
Synthesis and Characterization of [(η5-C5Me5)(OC)2-
Fe{μ-BN(SiMe3)2CdC}Ph] (4) by Thermally Induced
Borylene Transfer from [(CO)5ModBdN(SiMe3)2] (2a).
As shown in Scheme 3, reaction of a pale yellow solution
of [(OC)5ModBdN(SiMe3)2] (2a) with an equimolar
amount of iron-alkynyl 3 in tetrahydrofuran (THF) at
80 °C yielded, within 30 min, the iron-substituted borirene
complex (4). The reaction was monitored by 11B NMR
Single crystals of 4 suitable for X-ray diffraction anal-
ysis were obtained from a hexane solution at -60 °C.
The molecule crystallizes in the monoclinic space group
(18) Braunschweig, H.; Herbst, T.; Radacki, K.; Frenking, G.; Celik,
M. A. Chem.—Eur. J. 2009, 15, 12099.

Article
Inorganic Chemistry, Vol. 50, No. 1, 2011 65
Scheme 4. Irradiation of the Iron-Substituted Borirene 4
˚
Figure 2. Molecular structure of 5. Relevant bond lengths [A] and
angles [deg] for one of two independent molecules in the asymmetric unit,
which feature very similar structures: Fe1-B1 1.9950(17), B1-N1
1.403(2), B1-C1 1.521(2), C1-C2 1.268(2), C2-C3 1.453(2), Fe1-
B1-N1 153.59(12), C1-B1-N1 135.14(14), Fe1-B1-C1 71.09(9),
B1-C1-C2 133.98(14), C1-C2-C3 147.34(15), B1-N1-Si1 123.20(10),
B1-N1-Si2 114.42(10), Si1-N1-Si2 121.66(7).
˚
Figure 1. Molecular structure of 4. Relevant bond lengths [A] and
angles [deg]: Fe1-C1 1.9826(14), C1-C2 1.3631(19), C1-B1 1.501(2),
C2-B1 1.474(2), C2-C3 1.4699(19), B1-N1 1.4319(19), N1-Si1
1.7517(13), N1-Si2 1.7464(13), Fe1-C1-C2 137.60(10), C1-C2-C3
140.19(13), B1-C1-C2 61.73(10), B1-C2-C1 63.76(10), C1-B1-C2
54.52(9), C1-B1-N1 152.29(14), C2-B1-N1 153.03(15), B1-N1-Si1
112.67(10), B1-N1-Si2 121.10(10).
δ=75.5 and 86.7 ppm respectively. After full character-
ization of these new compounds (vide infra) it became
obvious that the formation of a mixture is due to the
incomplete ejection of CO imposed by the closed reaction
vessel. Thus, we irradiated 4 under similar conditions, but
replaced the atmosphere in the NMR tube every 30 min
by dry argon, which led to complete conversion of 4 into 5
as evidenced by 11B NMR spectroscopy (Scheme 4). After
work up, complex 5 was isolated as a red, moderately air-
and moisture sensitive solid in 71% yield.
P2(1)/c and displays the overall geometry of an aminobori-
rene. Inparticular, the endocyclic distances, thatis, C1-C2
1.3631(19) A, C1-B1 1.501(2) A, C2-B1 1.474(2) A
together with the slightly elongated B-N separation of
1.4319(19) A resemble that of previously reported borirenes
with exocyclic B-N bonds (Figure 1). These data are in
accord with extensive delocalization of the two π-electrons
over a three-center bonding molecular orbital composed of
the p-atomic orbitals of boron and carbon. The phenyl ring
and the boracyclopropene unit are not coplanar, but enclose
a dihedral angle of 67.03°, which is presumably due to steric
congestion imposed by the bulky N(SiMe3)2 groups and the
˚
˚
˚
˚
The significant downfield shift in the 11B NMR spec-
trum of 5 in comparison to 4 of approximately 40 ppm
already indicates a rearrangement of the former borirene
moiety, as such deshielded resonances are typically found
for iron bound boryl groups.21 Additionally, the observa-
tion of two broad signals for the nitrogen-bound tri-
methylsilyl groups at 0.58 and 0.43 ppm in a 1:1 ratio at
ambient temperature in the 1H NMR spectrum indicates
a significantly increased rotational barrier of the nitrogen-
boron bond. These spectroscopic data are in good accor-
dance with the molecular structure of 5 as elucidated in
the crystal, in which the B-N π-contribution is increased
as a result of the BCC-ring-opening.
Single crystals of 5 suitable for X-ray crystallography
were obtained by cooling a hexane solution to -60 °C.
The molecule crystallizes in the triclinic space group P1
with two almost identical molecules in the asymmetric
unit. As the geometry of both subunits is identical within
the experimental error, only one set of data will be dis-
cussed in the following. As shown in Figure 2, the overall
5
˚
Cp*-ligand (Cp*=η -C5Me5). Thedistanceof1.9826(14) A
between Fe1 and the sp2-hybdrized C1 is somewhat greater
as the corresponding bond in the alkynyl precursor 3
(1.924(7) A), where the C1-atom is sp-hybdrized,19 and is
˚
thus comparable with the Fe-C bond in [Cp*Fe(CO)2-
C6F5] (Cp* = η -C5Me5) (1.970(4) A),20 which connects
5
˚
the iron center to an aromatic ring.
Irradiation of [(η5-C5Me5)(OC)2Fe{μ-BN(SiMe3)2CdC}-
Ph] (4). Because of the aforementioned preliminary re-
sults on the attempted photochemical synthesis of 4, we
became interested in the behavior of this borirene com-
plex upon irradiation with UV light. Room temperature
photolysis of a yellow C6D6 solution of 4 was carried out
in a sealed Young-NMR-tube. The reaction was moni-
tored by 11B NMR spectroscopy, which revealed gradual
consumption of the starting material 4 and formation of
two new boron-containing compounds 5 and 6 in a ratio
of approximately 3:1, as indicated by new resonances at
(21) (a) Braunschweig, H.; Ganter, B.; Koster, M.; Wagner, T. Chem. Ber.
1996, 129, 1099. (b) Braunschweig, H.; Kollann, C.; Englert, U. Eur. J. Inorg.
Chem. 1998, 465. (c) Braunschweig, H.; Kollann, C.; M€uller, M. Eur. J. Inorg.
Chem. 1998, 291. (d) Hartwig, J. F.; Huber, S. J. Am. Chem. Soc. 1993, 115,
4908.
(19) Akita, M.; Terada, M.; Oyama, S.; Morooka, Y. Organometallics
1990, 9, 816.
(20) Chukwu, R.; Hunter, A. D.; Santarsiero, B. D.; Bott, S. G.; Atwood,
J. L.; Chassaignac, J. Organometallics 1992, 11, 589.

66 Inorganic Chemistry, Vol. 50, No. 1, 2011
Scheme 5. Synthesis of 6 upon Carbonylation of 5
Braunschweig et al.
appearance of 5 is that of an (alkynyl)boryl complex, in
which the boron bound C-C triple bond coordinates in a
η2-fashion to the iron center, thus constituting a highly
unusual structural motif in boryl chemistry. While the
sum of angles around boron B1 (359.82°) and nitrogen N1
(359.28°) proves planar coordination for both atoms, the
Fe1-B1-C1 angle of 71.09°(9) displays significant de-
viation from ideal trigonal planar geometry commonly
observed for sp2-hybdrized boron centers, thus indicating
a highly strained molecular structure. Likewise, the
˚
Figure 3. Molecular structure of 6. Relevant bond lengths [A] and
angles [deg]: Fe1-B1 2.075(2), B1-C1 1.554(3), B1-N1 1.444(3),
C1-C2 1.208(3), C2-C3 1.439(3), Fe1-B1-C1 111.64(13), Fe1-
B1-N1 131.84(15), C1-B1-N1 115.83(16), B1-N1-Si1 115.97(13),
B1-N1-Si2 125.21(13), Si1-N1-Si2 117.69(9), B1-C1-C2 178.4(2),
C1-C2-C3 177.0(2).
tioned deshielded 11B NMR resonance of δ=86.7 ppm
indicates the presence of a metal bound boryl ligand.
Interestingly, the 1H NMR spectrum at ambient tempera-
ture shows only one resonance at δ=0.46 ppm for the
nitrogen-bound SiMe3 group, which suggests a reduced
rotational barrier of the B-N bond in comparison with
that in 5. This finding is somewhat surprising as hindered
rotation about B-N double bonds is well documented for
both (amino)boryl complexes of the type [(η5-C5R5)(OC)2-
Fe-B(NR2)R0]21b and alkynyl(amino)boranes R(R02N)B-
CtC-R0.24
˚
Fe1-B1 separation of 1.9950(17) A is rather short for a
iron-boryl bond and matches, for example, the one in
5
˚
[(η -C5H5)(OC)2Fe-B(F)-Si(SiMe3)3] (1.983(9) A), de-
spite the fact that the latter is sterically less congested
because of the presence of the parent C5H5 ligand at the
iron center.22 The side-on coordination of the alkynyl
group has the expected effect on the C-C triple bond.
˚
Thus, the C1-C2 distance, which amounts to 1.268(2) A
is significantly elongated in comparison to the non-co-
˚
The formation of 6 was further confirmed by X-ray
diffraction analysis. Suitable single crystals were obtained
from a hexane solution at -70 °C, and the molecule cry-
stallizes in the orthorhombic space group Pbca. The
major difference between 6 and 5 lies with the almost
undistorted geometry of the boryl group in case of the
former. As shown in Figure 3, the sum of the angles
around boron B1 (359.3°) and nitrogen N1 (358.9°) prove
planar coordination geometries for boron atoms. How-
ever, because of the “free” alkynyl group, the angles Fe1-
B1-C1 (111.64(13)°), Fe1-B1-N1 (131.84(15)°), and
C1-B1-N1 (115.83(16)°) indicate a non-strained, sp2 hybri-
ordinated C-C triple bond in 6 (1.208(3) A; vide infra),
but very similar to values typically found for alkynes that
are η2-coordinated to a metal of the iron triad.23 The
overall geometry of the C3-C2-C1-B1 moiety gives
further evidence for the presence of molecular strain, as
the boryl- and the phenyl group adopt a mutual trans
disposition with respect to the C1-C2 multiple bond dis-
playing angles of 147.34(15)° (C1-C2-C3) and 133.98(14)°
(B1-C1-C2), respectively. Finally, the somewhat shor-
˚
tened B-N bond length of 1.403(2) A in comparison to
˚
the value of 1.4319(19) A found for the borirene complex
4 indicates a slightly stronger π-interaction between
boron and the exocyclic nitrogen atom because of clea-
vage of the BCC-ring and canceling of the endocyclic 2 π
electron delocalization.
˚
dized boron atom. The Fe1-B1 separation of 2.075(2) A
is significantly larger than that in 5 and marks the higher
end of Fe-B distances commonly observed for neutral
iron half-sandwich boryl complexes of iron (196-209
ppm).3d Consistent with the results from 1H NMR spec-
[(η5-C5Me5)(OC)2FeBN(SiMe3)2CCPh] (6). To pro-
vide selective access to the dicarbonyl complex 6, the
previously obtained monocarbonyl species 5 was treated
with CO. To this end, a deep red solution of 5 in C6D6 was
kept under an atmosphere of CO at ambient temperature
for a couple of hours (Scheme 5). The progress of reaction
was monitored by multinuclear NMR spectroscopy, which
revealed gradual consumption of the starting material 5,
and quantitative formation of a new boron containing
compound as indicated by the presence of a new reson-
ance at δ = 86.7 ppm in the 11B NMR spectrum. After
work up, 6 could be isolated by crystallization from
hexanes at -60 °C as an analytically pure, tawny crystal-
line solid in 52% yield.
˚
troscopy, the B-N bond length of 1.444(3) A resembles
that of the iron borirene complex 4 (1.4319(19) A), and
˚
thus suggests a comparable B-N π-interaction.
Stepwise Transformation from 6 to 4 under Thermal
Conditions. To evaluate the thermal stability of the un-
usual (alkynyl)boryl complex 6, a tawny solution of this
complex in C6D6 was heated at 80 °C under argon in a
sealed Young-NMR-tube, and the reaction was moni-
tored by multinuclear NMR spectroscopy. 11B NMR spec-
tra revealed a gradual consumption of 6 (δ = 86.7 ppm)
with concomitant formation of 5 (δ=75.5 ppm). How-
ever, as evidenced by a 11B NMR signal at δ = 36.1 ppm
and corresponding resonances in the 1H NMR spectrum,
a small amount of 4 was formed as well. Upon the assum-
ption that part of the thermally dissociated CO was not
The spectroscopic data of 6 in solution are in agreement
with the proposed structure. In particular, the aforemen-
(22) Braunschweig, H.; Colling, M.; Kollann, C.; Englert, U. J. Chem.
Soc., Dalton Trans. 2002, 2289.
(23) Ball, R. G.; Burke, M. R.; Takats, J. Organometallics 1987, 6, 1918.
€
(24) Feulner, H.; Metzler, N.; Noth, H. J. Organomet. Chem. 1995,
489, 51.

Article
Inorganic Chemistry, Vol. 50, No. 1, 2011 67
Scheme 6. Stepwise Transformation from 6 to 4 under Thermal Conditions
Scheme 7. Synthesis of Phosphane Complex 7
distributed into the gas phase and thus, enabled the
formation of the borirene(dicarbonyl) complex 4, we
heated the reaction mixture under a dry atmosphere of
CO. As indicated by the characteristic 11B- (δ = 36.1 ppm)
and 1H NMRresonances(δ= 0.42(s, 18H, Si(CH3)3), 1.42
(s, 15H, C5(CH3)5)) the intermediate 5 was indeed comple-
tely converted into 4 under these conditions (Scheme 6).
The thermal conversion of the iron (alkynyl)boryl 6
into the iron borirene 4 is unprecedented in metal boryl/
borirene chemistry and suggests that 4 is thermodynami-
cally favored over 6 (vide infra). Morevover, 4 and 6
constitute a pair of hitherto unknown iron boron species,
which are fully convertible under either photochemical or
thermal conditions. Likewise interesting is the fact that
the isomerization of an alkynylborane into a borirene (1)
has been reported by Eisch et al.14b,c (Scheme 1), however,
under photochemical conditions. Thus, the thermal reac-
tion depicted in Scheme 6 constitutes a complementary
synthetic approach to these boron heterocycles.
˚
Figure 4. Molecular structure of 7. Relevant bond lengths [A] and
angles [deg]: Fe1-C1 1.9727(16), Fe1-P1 2.1753(5), C1-C2 1.380(2),
C1-B1 1.506(2), C2-B1 1.473(2), C2-C3 1.473(2), B1-N1 1.443(2),
Fe1-C1-C2 142.59(12), N1-Si1 1.7491(13), N1-Si2 1.7475(14),
Fe1-C1-B1 155.98(12), C2-C1-B1 61.21(11), C1-C2-B1 63.62(12),
C1-C2-C3 143.45(15), C3-C2-B1152.57(15), C1-B1-N1156.17(15),
C2-B1-N1 148.63(16), B1-N1-Si1 113.78(11), B1-N1-Si2 111.46(10),
Si1-N1-Si2 134.23(8).
B1-C1, C1-C2, and B1-N1 bonds are slightly longer in
comparison to those in 4. The dihedral angle of 86.69°
between the phenyl ring and the boracyclopropene unit is
larger than in 4, which may be due to the presence of the
phosphane ligand and its higher sterical demand in
comparison to CO.
Quantum Chemical Calculations. To understand the
mechanisms of the experimentally observed reactions
and to analyze the bonding situation in the molecules
particularly in the unusual compounds 5 and 6 we opti-
mized the geometries of 4-7 with quantum chemical
methods using density functional theory (DFT) at the
M052X/def2-SVP level of theory. Improved energies were
obtained using larger basis sets at M052X/def2-TZVPP at
M052X/def2-SVP optimized geometries. Figure 5 shows
the calculated structures with the most important bond
lengths and the relative energies of the molecules. The full
set of geometrical parameters is given in the Supporting
Information.
Synthesis of [(η5-C5Me5)(OC)(PMe3)Fe{μ-BN(SiMe3)2-
CdC}Ph] (7). Finally we investigated, whether the addi-
tion of CO to 5 at elevated temperature with isomeriza-
tion into 4 can be extended to different ligands such as
phosphines. As shown in Scheme 7, reaction of a deep red
solution of 5 with an equimolar amount of PMe3 in
hexane at 80 °C yielded the phosphane complex 7 within
1 h. The reaction was monitored by multinuclear NMR
spectroscopy, which revealed formation of a new boron-
and phosphorus-containing compound, as indicated by
the presence of a new resonance at δ=38.4 ppm in the
11B NMR spectrum, and at δ = 35.39 ppm in the 31P
NMR spectrum. The iron borirene derivative 7 was
isolated in the form of yellow crystals by filtration of
the reaction mixture and subsequent crystallization from
hexanes at -70 °C.
Single crystals of 7 suitable for X-ray diffraction anal-
ysis were obtained from a hexane solution at -70 °C. The
molecule crystallizes in the triclinic space group P1, and
the overall geometry resembles that of (Figure 4), in
accord with extensive delocalization of the two π-elec-
trons over a three-center bonding molecular orbital from
the p-atomic orbitals of boron and carbon. Because of
stronger σ-donor ability of the phosphane ligand, the
The calculated geometry of 4 is in excellent agreement
with the results of the X-ray structure analysis (Figure 1).
The differences between theory and experiment are within
the expected error range of the theoretical method and
may also arise from intermolecular interactions in the
solid. A very good agreement between the calculated and
experimental geometry is also found for the unusual
compound 5. It is interesting to note that the calculated
˚
interatomic distances Fe-C1 (2.088 A) and Fe-C2
(2.087 A) have nearly the same value which poses the
˚
question about the nature of the interactions between Fe
and the C2 moiety in 5. Figure 6 shows the Laplacian dis-
2
tribution r F(r) of the central moiety in 5 which addresses
this question. There are bond paths between Fe-C2,
C2-C1, C1-B, and Fe-B but there is no bond path

68 Inorganic Chemistry, Vol. 50, No. 1, 2011
Braunschweig et al.
Figure 5. Optimized geometries (M052X/def2-SVP) and relative energies in kcal/mol (M052X/def2-TZVPP//M052X/def2-SVP) of the calculated
molecules. The relative energies including ZPVE corrections at M052X/def2-SVP level are given in parentheses.
˚
˚
between Fe-C1. There is ring critical point for the
FeC2C1B moiety which identifies it according to the
AIM (Atoms in Molecules) criteria as a four-membered
ring. Note that the carbon atom C2 has an area of charge
concentration which has the shape of a droplet-like
appendix pointing toward the Fe atom while there is no
such feature in the Laplacian field of C1. We could locate
a gradient path from the ring critical point to the Fe atom
and to the midpoint of the C1-C2 moiety. The trans-
lation of this information into simple bonding schemes
supports the writing with a dashed line as shown in Figure 2.
Figure 5 gives the relative energy of 5 þ CO with respect
to 4 (22.0 kcal/mol) which is the theoretically predicted
bond dissociation energy (BDE) De of one CO ligand in 4
provided that 5 is the global energy minimum. The BDE
after correcting for zero-point vibrational contributions
is Do = 19.3 kcal/mol. Compound 6 which is an isomer of
4 is 12.4 kcal/mol higher in energy. Thus, the BDE of one
CO ligand in 6 is only De = 9.6 kcal/mol (Do = 6.9 kcal/
mol). The carbonylation reaction 5 þ CO f 6 (Scheme 5)
which slowly takes place at room temperature is thus
calculated to be exothermic by 6.9 kcal/mol. We did not
calculate transition states for the rotation about the B-N
bonds in 5 and 6 but the latter compound (Figure 3) has a
the former molecule (1.401 A calc., 1.403(2) A exper.)
which is in agreement with the NMR spectra suggesting
that the rotational barrier in 6 is lower than in 5.
The experimental observations which were described
above indicate that 5 is formed from 6 via decarbonyla-
tion at 80 °C while the carbonylation of 5 under the same
conditions gives 4 (Scheme 6). We were interested in the
question if a direct rearrangement of the isomers 6 f 4
might be possible under thermal conditions. Therefore we
calculated the transition state for the latter isomerization.
The optimized transition state structure TS(4-6) is shown
in Figure 5. The calculations predict that it is 59.2 kcal/
mol higher in energy than 4 which means that the activa-
tion barrier for the reaction 6 f 4 is 46.8 kcal/mol. After
correcting for zero-point vibrations we calculate a barrier
of 44.5 kcal/mol. This is much higher than the decarbo-
nylation/carbonylation reaction via 5.
We also calculated the phosphane complex 7 (Figure 4)
which is the product of the reaction 5 þ PMe3 (Scheme 7).
Figure 5 shows the calculated geometry of 7 which is in
very good agreement with the experimental values. The
phosphane ligand effectuates slightly longer bonds in 7
for C1-C2, C1-B, and B-N while the C2-B bond is a
bit shorter than in 4. This trend reflects the experimental
observations.
˚
˚
clearly longer bond (1.449 A calc., 1.444(3) A exper.) than

Article
Inorganic Chemistry, Vol. 50, No. 1, 2011 69
Table 1. Bond Orders P and Atomic Partial Charges q for Compounds 4-7
no.
P Fe-C1
P Fe-C2
P Fe-B
P C1-C2
P C1-B
P C2-B
P B-N
q Fe
q C1
q C2
q B
4
5
6
7
0.76
0.37
1.68
2.19
2.70
1.65
1.04
1.03
0.88
1.04
1.02
0.98
1.02
0.90
0.97
-1.01
-0.82
-1.30
-0.82
-0.19
-0.38
-0.36
-0.23
-0.32
0.06
-0.02
-0.35
0.67
0.95
0.96
0.65
0.48
0.56
0.64
0.76
1.04
The relative energy of 7 (-2.1 kcal/mol) is given with
respect to 4-CO þ PMe3. The ligand exchange reaction is
thus predicted to be exothermic by 2.1 kcal/mol (1.6 kcal/
mol after ZPE correction) and the reaction 5 þ PMe3 f
7 þ CO is exothermic by 24.1 kcal/mol (20.9 kcal/mol
after ZPE correction) which is in agreement with the
observed reaction course under thermal conditions.
We calculated the charge distribution and bond orders
of compounds 4-7. Table 1 shows that C1-C2 bond
order in 6 clearly suggests a triple bond while 5 has a
weakened triple bond because of the stronger conjugation
with the boron atoms. The C1-B bond order in 5 (1.03) is
clearly bigger than in 6 (0.88). Note that the former
compound has rather small bond orders for Fe-C2
(0.48) and particularly for Fe-C1 (0.37) which according
to the AIM results is not a chemical bond. Both values are
significantly smaller than the Fe-C1 bond orders for 4 and
7 (0.76) which supports the interpretation of a donor-
acceptor bond in 5 between the C2 moiety and the iron
atom. The boron atoms carries large positive charges of
0.67 e in 4 and 0.65 e in 7 which are even larger in 5 (0.95 e)
and 6 (0.96 e). The Fe atom has always a large negative
charge of ∼ -1.0 e. The calculated bond orders for the
B-N bond in 5 (1.02) and 6 (0.90) also support a lower
B-N rotational barrier in 6 compared with 5 (Figure 6).
2
Figure 6. Contour line diagrams r F(r) of the central moiety in 5. Solid
2
lines indicate areas of charge concentration (r F(r)<0) while dashed
2
lines show areas of charge depletion (r F(r) > 0). Solid lines connecting
atomic nuclei give the bond paths. The black dot indicates the ring critical
point (RCP).
acquired on either a Varian Unity 500 (1H: 499.834; 11B:
160.364; 13C: 125.697 MHz) or Bruker Avance 500 (1H:
500.133; 11B: 160.472; 13C: 125.777 MHz) NMR spectrometer.
1H and 13C{1H} NMR spectra were referenced to external TMS
via the residual protio solvent (1H) or the solvent itself (13C).
Conclusion
11B{1H} NMR spectra were referenced to external BF3 OEt2.
Here, we reported the selective synthesis of the iron borirene
complex 4, upon heating of the molybdenum borylene 2a
in the presence of the iron alkynyl complex 3, thus estab-
lishing thermally induced borylene transfer for the synthe-
sis of a metal borirene complex. Subsequently, we investi-
gated the photochemical and thermal stability of the title
compound and disclosed a fully reversible and unprece-
dented borirene-to-boryl transformation, which is sum-
marized in Scheme 8.
In addition to full characterization of all new complexes
4-7 in solution and in the crystal, DFT calculations were
performed, to elucidate the unusual electronic structure of 5
and to provide information about the thermodynamics of the
aforementioned reversible transformations. Indeed, quan-
tum chemical calculations suggest that the reaction 4-CO f
5 is endothermic by 19.3 kcal/mol and that 6 is 12.4 kcal/mol
higher in energy than 4. The formation of the phosphane
complex 7 in the reaction 5 þPMe3 f 7 þ CO is exothermic
by 20.9 kcal/mol.
3
31P{1H} NMR spectra were referenced to 85% H3PO4. Micro-
analyses for C, H, and N were performed on either a Carlo Erba
model 1106 or a Leco CHNS-932 Elemental Analyzer. PMe3
was used as 0.10 mol L-1 solution in hexane. Na2[M(CO)5]
3
(M=Cr, Mo),25 Cl2B(SiMe3)2,26 iron alkynyl complex [(η5-
C5Me5)(OC)2FeCtCPh]19 and borylene complexes, (OC)5Md
BN(SiMe3)2 (M=Cr, Mo)6 were prepared according to the
literature. NMR spectroscopic experiments were performed in
quartz or Young NMR tubes. The light source was a Hg/Xe arc
lamp (400-550 W) equipped with IR filters, irradiating at
210-600 nm. Large-scale experiments were performed in a 150 mL
Schlenk flask equipped with a quartz cooling jacket into which a
Hg lamp (125 W) was inserted vertically.
Synthesis of [(η5-C5Me5)(OC)2Fe{μ-BN(SiMe3)2CdC}Ph]
(4). In a Young NMR tube, a yellow solution of 2 (104 mg,
0.30 mmol) and 3 (120 mg, 0.30 mmol) in 2 mL of THF was
heated in a oil bath at 80 °C for 0.5 h. The volatile components
were removed in vacuo, and the brown residue was extracted
with 5 mL of hexane. The brown hexane solution was stored at
-60 °C overnight to separate Mo(CO)6. The filtrate was stored
at -60 °C for 2 weeks to afford yellow crystals of 4 (98 mg, 63%
1
yield). H NMR: δ = 0.42 (s, 18H, Si(CH3)3), 1.42 (s, 15H,
Experimental Section
C5(CH3)5), 7.09 (m, 1H, CH-p of C6H5), 7.31 (m, 2H, CH-m of
C6H5), 7.47(m, 2H, CH-o of C6H5); 13C{1H}-NMR: (C bonded
to boron not detected), δ = 217.76 (s, CO), 143.67 (s, C of
C6H5), 128.53 (s, CH-m of C6H5), 126.21 (s, CH-o of C6H5),
125.69 (s, CH-p of C6H5), 96.21 (s, C of C5(CH3)5), 9.93 (s, CH3
General Procedures. All manipulations were performed either
under dry argon or in vacuo using standard Schlenk line and
glovebox techniques. Solvents (THF, ether, toluene, benzene
and hexane) were purified by distillation under dry argon from
appropriate drying agents (sodium and sodium wire, re-
spectively) and stored under the same inert gas over molecular
sieves. Carbon monoxide gas was previously passed through
acetone-dry ice traps to remove moisture. NMR spectra were
(25) Maher, J. M.; Beatty, R. P.; Cooper, N. J. Organometallics 1985, 4, 1354.
(26) Haubold, W.; Kraatz, U. Z. Anorg. Allg. Chem. 1976, 421, 105.

70 Inorganic Chemistry, Vol. 50, No. 1, 2011
Braunschweig et al.
Scheme 8. Synthesis of the Iron-Substituted Borirene 4, Stepwise Reversible Metal-boryl(6)-to-borirene(4) Transformation and Synthesis of 7 upon
the Unprecedented Metal-boryl-to-borirene Transformation
of C5(CH3)5), 3.54 (s, Si(CH3)3); 11B{1H}-NMR: δ=36.1 (s).
Elemental analysis calcd. [%] for BC26FeH38NO2Si2: C 60.12, H
7.37, N 2.70; found: C 59.96, H 7.35, N 2.94.
and refilled with dry CO gas, brought again to 80 °C for another
2 h, examined by multinuclear NMR spectroscopy, which
indicated the formation of 4.
Synthesis of [(η5-C5Me5)(OC)FeBN(SiMe3)2(η2-CC)Ph] (5).
In a 5 mm quartz NMR tube, a pale yellow solution of 4 (80 mg,
0.15 mmol) in 1 mL of C6D6 was irradiated for 4 h at room
temperature. During this period the NMR tube was degassed
and filled with fresh argon every half an hour. The volatile
components were removed in vacuo, and the deep red residue
was extracted with hexane. The concentrated filtrate was stored
at -30 °C for several days to afford red crystals of 5 (54 mg,
71%). 1H NMR: δ = 1.59 (s, 15H, C5(CH3)5), 0.58 (bs, 9H,
Si(CH3)3), 0.43 (bs, 9H, Si(CH3)3), 7.08 (m, 1H, CH-p of C6H5),
7.21 (m, 2H, CH-m of C6H5), 7.96(m, 2H, CH-o of C6H5);
13C{1H}-NMR: (C of the alkyne ligand not detected), δ =
220.22 (s, CO), 133.08 (s, C of C6H5), 128.53 (s, CH-m of C6H5),
131.45 (s, CH-o of C6H5), 127.56 (s, CH-p of C6H5), 92.12 (s, C
of C5(CH3)5), 4.31 (bs, Si(CH3)3), 3.51 (bs, Si(CH3)3); 11B{1H}-
NMR: δ = 75.5; Elemental analysis calcd. [%] for BC25FeH38-
NOSi2: C 61.10, H 7.79, N 2.85; found: C 61.23, H 7.92, N 3.20.
Synthesis of [(η5-C5Me5)(OC)2FeBN(SiMe3)2CCPh] (6). In a
Young NMR tube, a deep red solution of 5 (40 mg, 0.08 mmol)
in 1 mL of C6D6 was degassed and refilled with dry CO gas. A
gradual color change from deep red to golden-brown could be
observed. After keeping the reaction mixture under CO for 2 h,
the volatile components were removed in vacuo, and the brown
residue was extracted with 1.5 mL of hexane. The brown hexane
solution was stored at -70 °C for a few days to afford yellow
Synthesis of [(η5-C5Me5)(OC)(PMe3)Fe{μ-BN(SiMe3)2CdC}-
Ph] (7). In a Young NMR tube, a deep red solution of 5 (20 mg,
0.04 mmol) and PMe3 (0.40 mL, c = 0.10 mol L-1, 0.04 mmol)
3
in 1 mL of hexane was heated in a oil bath at 80 °C for 1 h. The
volatile components were removed in vacuo, and the brown
residue was extracted with 1.5 mL of hexane. The brown hexane
solution was stored at -70 °C for several weeks to afford yellow
1
crystals of 7 (15 mg, 65% yield). H NMR: δ = 0.42 (s, 18H,
2
Si(CH3)3), 1.55 (s, 15H, C5(CH3)5), 0.97 (d, 15H, JH-P
=
8.8 Hz, PMe3), 7.06 (t, H, CH-p of C6H5), 7.29 (m, 4H, CH-m
and CH-o of C6H5); 13C{1H}-NMR: (C bonded to boron not
detected), δ = 221.90 (d, 2JC-P = 35.0 Hz, CO), 147.51 (s, C of
C6H5), 127.94 (s, CH-m of C6H5), 125.62 (s, CH-o of C6H5),
124.42 (s, CH-p of C6H5), 92.05 (s, C of C5(CH3)5), 19.30 (d,
2JC-P = 26.1 Hz, C of PMe3), 10.61 (s, CH3 of C5(CH3)5), 3.70
(s, Si(CH3)3); 11B{1H}-NMR: δ = 38.4 (s); 31P{1H}-NMR: δ =
35.39 (s). Elemental analysis calcd. [%] for BC28FeH47NOPSi2:
C 59.26, H 8.35, N 2.47; found: C 59.12, H 8.31, N 2.62.
X-ray Structure Determinations. The crystal data of 4, 5, 6 and
7 were collected on a Bruker X8Apex diffractometer with a
CCD area detector and multilayer mirror monochromated
MoKR radiation. The structure was solved using direct methods,
refined with the Shelx software package (Sheldrick, G. Acta
Crystallogr. 2008, A64, 112-122) and expanded using Fourier
techniques. All non-hydrogen atoms were refined anisotropi-
cally. Hydrogen atoms were assigned to idealized positions and
were included in structure factors calculations.
1
crystals of 6 (22 mg, 52% yield). H NMR: δ = 0.68 (s, 18H,
Si(CH3)3), 1.78 (s, 15H, C5(CH3)5), 7.11 (m, 1H, CH-p of C6H5),
7.17 (m, 2H, CH-m of C6H5), 7.627(m, 2H, CH-o of C6H5);
13C{1H}-NMR: (C of carbon-carbon triple bond not detected),
δ = 218.08 (s, CO), 125.25 (s, C of C6H5), 128.83 (s, CH-m of
C6H5), 130.55 (s, CH-o of C6H5), 128.16 (s, CH-p of C6H5),
96.84 (s, C of C5(CH3)5), 9.99 (s, CH3 of C5(CH3)5), 5.12 (s,
Si(CH3)3); 11B{1H}-NMR: δ = 86.7 (s). Elemental analysis
calcd. [%] for BC26FeH38NO2Si2: C 60.12, H 7.37, N 2.70; found:
C 60.17, H 7.28, N 3.26.
Crystal Data for 4. C26H38BFeNO2Si2, Mr = 519.41, orange
block, 0.27 ꢀ 0.25 ꢀ 0.19 mm3, monoclinic space group P21/c, a =
˚
˚
˚
15.9329(4) A,b= 8.3475(3) A,c= 22.8808(7) A,β= 110.2980(10)°,
V= 2854.17(15) A ,Z=4,Fcalcd = 1.209 g cm-3,μ= 0.634 mm-1
,
3
˚
3
F(000) = 1104, T = 233(2) K, R1 = 0.0377, wR2 = 0.0998, 7105
independent reflections [2θ e 56.66°] and 309 parameters.
Crystal Data for 5. C25H38BFeNOSi2, Mr = 491.40, orange
block, 0.26 ꢀ 0.14 ꢀ 0.12 mm3, triclinic space group P1,
˚
˚
˚
Transformation from 6 to 4. In a Young NMR tube, a tawny
solution of 6 (20 mg, 0.04 mmol) in C6D6 under argon was
heated in a oil bath at 80 °C for 2 h, yielding a deep red solution
constituted by 5 and a small amount of 4, as indicated by 11B-
and 1H NMR spectroscopy. The reaction mixture was degassed
a = 7.2055(8) A, b = 19.729(2) A, c = 19.868(2) A, R =
3
˚
71.520(5)°, β = 83.681(5)°, γ = 85.313(5)°, V = 2659.2(5) A ,
Z = 4, Fcalcd = 1.227 g cm-3, μ = 0.674 mm-1, F(000) = 1048,
3
T = 100(2) K, R1 = 0.0346, wR2 = 0.0826, 11191 independent
reflections [2θ e 53.58°] and 581 parameters.

Article
Inorganic Chemistry, Vol. 50, No. 1, 2011 71
Crystal Data for 6. C26H38BFeNO2Si2, Mr = 519.41, yellow
needle, 0.13 ꢀ 0.05 ꢀ 0.05 mm3, orthorhombic space group
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
˚
˚
˚
Pbca, a = 8.0512(7) A, b = 18.0246(14) A, c = 38.492(3) A, V =
˚
Theoretical Methods. The DFT calculations have been car-
ried out using the hybrid meta functional M05-2X27 in con-
junction with the def2-SVP basis set.28 The optimized structures
have been verified as energy minima by calculating the vibra-
tional frequencies analytically. Single point calculations have
been carried at M05-2X/def2-TZVPP level.29 NBO analysis30
was performed at M05-2X/def2-SVP optimized geometries at
the M05-2X/def2-TZVPP level. The program package Gauss-
ian03 was used throughout.31 The AIM calculations were
carried out using the program AIMPAC.32
3
5586.0(8) A , Z = 8, Fcalcd = 1.235 g cm-3, μ = 0.648 mm-1
,
3
F(000) = 2208, T = 100(2) K, R1 = 0.0478, wR2 = 0.0857, 6425
independent reflections [2θ e 55.28°] and 309 parameters.
Crystal Data for 7. C28H47BFeNOPSi2, Mr = 567.48, yellow
plate, 0.60 ꢀ 0.20 ꢀ 0.10 mm3, triclinic space group P1, a =
˚
˚
˚
8.9941(5) A, b = 11.8876(6) A, c = 15.4367(8) A, R =
92.142(2)°, β = 98.880(2)°, γ = 109.052(2)°, V = 1534.54(14)
A , Z = 2, Fcalcd = 1.228 g cm-3, μ = 0.643 mm-1, F(000) =
3
˚
3
608, T = 100(2) K, R1 = 0.0343, wR2 = 0.0833, 6248 indepen-
dent reflections [2θ e 52.76°] and 330 parameters.
Crystallographicdatahave beendeposited withthe Cambridge
Crystallographic Data Center as supplementary publication no.
CCDC 774455-774458. These data can be obtained free of
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft (DFG).
Supporting Information Available: CIF file for 4, 5, 6, and 7,
Cartesian coordinates and total energies of all calculated com-
pounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
(27) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput.
2006, 2, 364.
€
(28) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
(29) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
(30) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899.
(31) Frisch, M. J. et al. Gaussian 03, Rev. D.01; Gaussian, Inc.: Wallingford,
CT, 2004; For full reference, see the Supporting Information.
(32) Bader, R. F. W.; http://www.chemistry.mcmaster.ca/aimpac/imagemap/
imagemap.htm.