Conversions of osmabenzyne and isoosmabenzene

Article abstract of DOI:10.1002/chem.201201558

We report herein the first example of the conversion of metallabenzyne II and isometallabenzene III. The osmium hydride vinylidene complex 1 reacts with HC≡CCH(OEt)₂ to give osmabenzyne 3 via isoosmabenzene 2. Compound 3 exhibits high thermal stability in air. Nonetheless, nucleophilic attack at 3 provides isoosmabenzenes 4 a and 4 b, or opens the ring to produce 5 a and 5 b. We propose mechanisms to disclose the intrinsic connection between the six-membered metallacycles, and carry out DFT calculations to rationalize the regioselectivity of the nucleophilic addition reactions. Copyright

Full text of DOI:10.1002/chem.201201558

FULL PAPER
DOI: 10.1002/chem.201201558
Conversions of Osmabenzyne and Isoosmabenzene
Qianyi Zhao, Jun Zhu, Zi-Ao Huang, Xiao-Yu Cao,* and Haiping Xia*[a]
Abstract: We report herein the first example of the conversion of metallabenzyne
II and isometallabenzene III. The osmium hydride vinylidene complex 1 reacts
ꢀ
with HC CCH
(OEt)2 to give osmabenzyne 3 via isoosmabenzene 2. Compound 3
Keywords: isoosmabenzene · met-
allacycles · nucleophilic addition ·
osmabenzyne · osmium
exhibits high thermal stability in air. Nonetheless, nucleophilic attack at 3 provides
isoosmabenzenes 4a and 4b, or opens the ring to produce 5a and 5b. We propose
mechanisms to disclose the intrinsic connection between the six-membered metal-
lacycles, and carry out DFT calculations to rationalize the regioselectivity of the
nucleophilic addition reactions.
Introduction
studies of their chemistry. Herein, we show for the first time
that isometallabenzene can transform into metallabenzyne,
and metallabenzyne can undergo nucleophilic addition reac-
tions to “restore” isometallabenzenes, or to go further and
open the metallacycle.
Metalla-aromatics[1] are derived from the replacement of a
(hydro)carbon unit in aromatic hydrocarbons with a metal
fragment. Considerable effort has been devoted to the study
of metallabenzenes I, and these metal-containing analogues
Results and Discussion
Conversion of isoosmabenzene to osmabenzyne: Treatment
ꢀ
of the osmium hydride vinylidene 1 with HC CCH
G
chloroform produced osmabenzyne 3 through a formal
[3+3] cycloaddition reaction[10a] via the key intermediate iso-
osmabenzene 2 (Scheme 1). The reaction took 8 h to com-
plete at room temperature (RT; Scheme 1, reaction condi-
tions i), or 2 h if heated at reflux.
of benzene (I’) have been found to be isolable, stable, and
aromatic.[2] Synthetic methods and the reactivity of metalla-
benzenes have now been established,[3–5] including some re-
actions that directly relate to classic benzene chemistry.[6–8]
The isolation of stable metallabenzynes II[9] and isometal-
labenzenes III[10] challenged the usual notions of highly
strained rings. In comparison, benzyne (II’)[11,12] and isoben-
zene (III’)[13] are too reactive and short-lived to be isolated.
Among the reactions of metallabenzynes,[9e–h] the intercon-
version of metallabenzyne and metallabenzene is particular-
ly interesting[9g,14] because it resembles the mutual transfor-
mation of benzyne and benzene. On the other hand, isomet-
The structure of osmabenzyne 3 in the solid state has
been verified by X-ray diffraction. As shown in Figure 1, the
six-membered metallacycle in 3 is essentially planar. The
maximum deviation from the least-squares plane through
Os1 and C1–C5 is 0.028 ꢀ for C5. The Os1-C1-C2 angle is
ACHTUNGTRENNUNG
allabenzenes remain rare,[10] thus restricting the in-depth
[a] Q. Zhao, J. Zhu, Z.-A. Huang, Dr. X.-Y. Cao, Prof. Dr. H. Xia
State Key Laboratory of Physical Chemistry of Solid Surfaces
College of Chemistry and Chemical Engineering
Xiamen University, Xiamen, 361005 (P.R. China)
Fax : (+86)592-2186628
E-mail: xcao@xmu.edu.cn
hpxia@xmu.edu.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201558.
Scheme 1. Preparation of osmabenzyne 3 via isoosmabenzene 2.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Figure 1. Molecular structure of the cation of osmabenzyne 3 (ellipsoids
shown at the 50% probability level). The hydrogen atoms of PPh3 are
À
omitted for clarity. Selected bond lengths [ꢀ] and angles [8]: Os1 C1:
À
À
À
À
1.775(6), Os1 C5: 2.011(7), C1 C2: 1.390(9), C2 C3: 1.403(9), C3 C4:
Figure 2. Time-evolved 31P NMR spectra of the formation of isometalla-
benzene 2 and its conversion into metallabenzyne 3 at 293 K in CDCl3.
À
1.381(9), C4 C5: 1.389(9), C1-Os1-C5: 80.7(3), Os1-C1-C2: 148.6(5), C1-
C2-C3: 113.1(6), C2-C3-C4: 123.5(6), C3-C4-C5: 123.7(6), C4-C5-Os1:
130.2(5).
action solution contained a mixture of 2 and 3, with 2 as the
main component. It took another 7 h for 2 to convert com-
pletely into 3.
ꢀ
148.6(5)8, the Os1 C1 bond length is 1.775(6) ꢀ and the
À
À
Os1 C5 bond length is 2.011(7) ꢀ. The C C bond lengths
in the metallacycle in 3 (1.381(9)–1.403(9) ꢀ) suggest a delo-
calized nature. All of these structural features are compara-
ble to the osmabenzynes produced by Jiaꢂs group.[9b–h]
Based on these observations, we proposed a mechanism
for the formation of 3 from 2 (Scheme 1). The mechanism
for the formation of isoosmabenzene 2 is a formal [3+3] cy-
cloaddition that has been discussed previously.[10a] Once 2
has been generated, its ethoxyl group on C3 extracts a
proton (presumably from trace amounts of acids in chloro-
form or from the eliminated ethanol) to form an onium salt
A, which undergoes ethanol elimination to generate 3’, a
resonance form of 3.[9a,d,g]
Thus, we reasoned that the addition of an acid would ac-
celerate the conversion from 2 to 3. Indeed, after the reac-
tion had been carried out at RT for 1 h, the addition of
1 equivalent of HCl caused a fast color change from green
to deep blue. The transformation was complete within
15 min (Scheme 1, reaction conditions ii). In contrast, the
addition of K2CO3 slowed down the conversion, and hence,
facilitated the precipitation of 2 (Scheme 1, reaction condi-
tions iii).
Osmabenzyne 3 and the intermediate isoosmabenzene 2
were characterized by NMR spectroscopy. The ring number-
ing system for osmabenzyne and isoosmabenzene is shown
1
in Scheme 1. In the H and 13C NMR spectrum, compound 3
exhibits characteristic chemical shifts of an osmabenzyne.
H5 has a characteristically low-field resonance at 13.9 ppm,
and the metal-bound carbon atoms C1 and C5 show typical
downfield signals at 287.9 and 246.7 ppm. The 31P NMR
spectrum of 3 consists of two singlets: 14.7 ppm for CPPh3
and À4.4 ppm for OsPPh3. The NMR spectra of 2 resemble
those of previous isoosmabenzenes.[10] In the H NMR spec-
1
trum, two vinyl protons resonate at 8.1 (H5) and 5.0 ppm
(H4) with a coupling constant of 9.0 Hz, thus indicating a cis
geometry. The H3 proton, attached to the saturated carbon
atom C3, was found at 1.7 ppm. In the 31P NMR spectrum of
2, CPPh3 resonates at 4.4 ppm as a singlet, and two OsPPh3
signals appear as a characteristic AB quadruplet at À4.4 and
À7.1 ppm (2JPP =350.3 Hz), due to the presence of a chiral
center at C3.[10] The molecular formulae of 2 and 3 have
been confirmed by high-resolution mass spectrometry
(HRMS: m/z calcd for C61H53OP3ClOs (2): 1121.2613
[MÀCl]+; found: 1121.2609; m/z calcd for C59H48P3Cl2Os
(3): 1111.1961 [M]+; found: 1111.1956).
The formation of osmabenzyne 3 via isoosmabenzene 2
was monitored by in situ NMR spectroscopy at RT. As
shown in Figure 2, isoosmabenzene 2 was formed preferen-
tially in the beginning, and transformed gradually (at a
slower rate than its formation) into osmabenzyne 3. After
1 h, the starting material 1 had been consumed, and the re-
Conversion of osmabenzyne to isoosmabenzene: Osmaben-
zyne 3 is highly stable, even when heated in air at 1208C for
5 h in the solid state or at 808C for 8 h in solution. This un-
usual stability is most likely attributable to the steric hin-
drance and electronic effects induced by the phosphonium
group at C2. The same stabilizing effect in isoosmabenzenes
has been studied by DFT calculations.[10a] According to theo-
retical studies, other p-accepting groups (silyl or boryl) at
C2 or C4 also stabilize osmabenzynes.[9i] The inertness of 3
is likely for the same reasons. It remains intact in the pres-
ence of acids (HOAc, H3PO4, HBF4, and HCl), bases
(K2CO3 and NaOH), terminal alkynes, and nucleophiles
(H2O, MeOH, and NaBH4).
&
2
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Conversions of Osmabenzyne and Isoosmabenzene
FULL PAPER
Scheme 2. Nucleophilic addition of 3: from metallabenzyne to isometalla-
benzene.
However, strong nucleophiles, such as ethyllithium (EtLi)
or sodium methanethiolate (NaSMe), attack osmabenzyne 3
at C3 to generate isoosmabenzene 4a[10a] or 4b (Scheme 2).
The addition of approximately 1 equivalent of EtLi to a sol-
ution of 3 in THF switched the color from deep blue to
brownish red. A red solid was then isolated and identified
as our reported isoosmabenzene 4a.[10a] Excess EtLi decom-
posed 4a. The nucleophilic attack by MeSÀ gave isoosma-
benzene 4b as the main product, as shown by the presence
of characteristic signals for an isoosmabenzene in the 1H
and 31P NMR spectra generated in situ. Nevertheless, 4b de-
composed completely within 4 h at RT. Attempts to isolate
4b were not successful. It would be interesting if treatment
of osmabenzyne 3 with sodium ethoxide (NaOEt) could
generate isoosmabenzene 2, but upon NaOEt addition, 3 re-
mained intact at RT and decomposed at higher tempera-
tures.
Figure 3. Molecular structure of the cation of 5a’ (ellipsoids shown at the
50% probability level). Some hydrogen atoms are omitted for clarity.
À
changed complex 5a’ (BPh4 instead of ClÀ; Figure 3). The
other product 5b is less stable than 5a, and can only be de-
tected by in situ NMR spectroscopy.
Possible mechanisms for the nucleophilic addition to os-
mabenzyne 3 are proposed in Schemes 2 and 3. The alterna-
tive resonance structure 3’ again bridges the transformation
from 3 into 4a (or 4b). Carbon or sulfur nucleophiles attack
3’ at C3, producing isoosmabenzene 4a or 4b (Scheme 2).
Amines take a similar path yet go further (Scheme 3). In
the beginning, an amine replaces a chloride ligand to coordi-
nate with the osmium center. The dicationic osmabenzyne B
is more electrophilic than cationic osmabenzyne 3. Hence,
primary amines (weaker nucleophiles than EtLi and
NaSMe) subsequently attack the C3 of B’ (a resonance form
of B) to give isoosmabenzene C. Besides a phosphonium on
C2, C carries a quaternary ammonium on C3, which differs
from 4a or 4b. The neighboring positively charged substitu-
Ring opening of osmabenzyne 3: Compared with the nucle-
ophiles used above, amines react differently with osmaben-
zyne 3. Primary amines (n-butylamine and 2-propargyl-
ACHTUNGTRENNUNGamine) also attack at C3, but open the metallacycles rather
À
than giving isoosmabenzenes (Scheme 3). Secondary and
tertiary amines produced complicated mixtures. The struc-
ents in C weaken the C2 C3 bond, presumably due to de-
creased overall electron density and increased repulsive
forces, thus driving the metallacycle to rearrange and open
to give 5a or 5b.
1
ture of ring-opening product 5a was inferred by H, 13C, and
31P NMR spectroscopy, and then confirmed by X-ray diffrac-
tion, with red crystals grown from the counteranion-ex-
Note that intermediate A (Scheme 1, structurally similar
to C) tends to eliminate an ethanol molecule instead of
opening the ring, because the cation in A carries only one
positive charge, and ethanol is easier to lose than an amine.
Meanwhile, although the ring opening of six-membered met-
[9h]
À
alla-aromatics through M C (IV, forming a carbene or cy-
clopentadienyl[4c,g] complex through carbene migratory inser-
À
tion) and C1 C2 (V, forming a metallacyclopentadiene
through ring contraction[4j,k]) bond cleavages has been re-
ported, the transformation from osmabenzyne 3 to 5a or 5b
À
represents the first C2 C3 bond cleavage (VI).
DFT calculations were carried out to rationalize the
highly regioselective nucleophilic attack at C3 in osmaben-
Scheme 3. Nucleophilic addition and ring opening of 3.
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

X.-Y. Cao, H. Xia et al.
zyne 3.[15] Dicationic osmabenzyne B* is used as the model
of the cation of B (R=nBu). C* and C# are models of cati-
ons of two possible intermediates (Scheme 4). The natural
population analysis[16] on the full model of B* shows differ-
among metallabenzynes because it is formed via an isome-
tallabenzene precursor, and can undergo regioselective nu-
cleophilic additions to furnish isometallabenzenes. Hence,
we have achieved the first formal interconversion of metal-
labenzyne and isometallabenzene.
Experimental Section
General comments: All manipulations were carried out under an inert at-
mosphere (Ar or N2) by means of standard Schlenk techniques. Solvents
were distilled from sodium/benzophenone (hexane and diethyl ether) or
calcium hydride (dichloromethane, chloroform, and 1,2-dichloroethane)
under N2 prior to use. Methanol was used as received. NMR spectroscop-
ic experiments were carried out on a Bruker 500 MHz spectrometer (1H:
500.2; 13C: 125.8; 31P: 202.5 MHz). High-resolution mass spectra (HRMS)
were recorded on a Bruker En Apex Ultra 7.0T FTMS mass spectrome-
ter. Elemental analyses were performed on a Vario EL III elemental ana-
lyzer.
Scheme 4. Gibbs free energies (DG8) and electronic energies (DE) of the
reaction via two possible intermediates C* and C#.
ꢀ
Compound 2: HC CCH
0.70 mmol) were added to a solution of [OsH
G
E
G
CHTUNGTRENNUNG
ent charges on the metallacycle (from C1 to C5: 0.305,
À0.610, À0.088, À0.317, and À0.188). This analysis indicates
that the order of preferential sites for nucleophilic attack is
C1>C3>C5, in line with previous studies.[9g,i] Because C1
in osmabenzyne 3 is surrounded by three bulky PPh3 groups,
the resulting steric hindrance would make nucleophilic
attack at C1 difficult. Further examination of the contribu-
tion to the lowest unoccupied molecular orbital (LUMO,
Figure 4) of B* from C1 (9.2%), C3 (23.3%), and C5
(1; 150 mg, 0.14 mmol) in CHCl3 (15 mL). The mix-
ture was stirred at room temperature (RT) for 1 h
to give a green solution. After removing excess
K2CO3 by filtration, the filtrate was evaporated
under vacuum. The residue was washed with
hexane (3ꢃ10 mL) and dried under vacuum to give
1
a 3:1 mixture (estimated by H NMR spectroscopy)
of 2 and 3 as a light-green solid (129 mg, 60% for 2). Diagnostic peaks
for 2 are as follows: 1H NMR (500 MHz, CDCl3): d=8.1 (d, JHH
=
3
3
9.0 Hz, 1H; H5), 5.0 (d, JHH =9.0 Hz, 1H; H4), 3.0 (m, 1H; H6), 2.3 (m,
1H; H6), 1.7 (m, 1H; H3), 0.5 ppm (t, 3JHH =7.0 Hz, 3H; H7); 31P NMR
(202 MHz, CDCl3): d=4.4 (s; CPPh3), À4.4 (d, 2JPP =350.3 Hz; OsPPh3),
À7.1 ppm (d, 2JPP =350.3 Hz; OsPPh3); HRMS (ESI): m/z calcd for
C61H53OP3ClOs: 1121.2613 [MÀCl]+; found: 1121.2609.
Compound 3:
ꢀ
Method A: HC CCH
G
of [OsH
A
G
CHTUNGTRENNUNG
(60 mg, 0.36 mmol) in CHCl3 (15 mL). The mixture was stirred at RT for
8 h or under reflux for 2 h to give a blue-green solu-
tion. The solvent was removed under vacuum. The
residue was washed with CHCl3 (10 mL), Et2O (2ꢃ
10 mL), and dried under vacuum to give 3 as a blue
solid (290 mg, 85%).
ꢀ
Method B: HC CCH
U
added to
a
solution of [OsH
N
ACHTUNGTRENNUNG
AHCTUNGTRENN(GUN PPh3)2] (1; 330 mg, 0.31 mmol) in CHCl3 (15 mL). The mixture was stir-
red at RT for 1 h to give a green solution. The addition of concentrated
HCl (14 mL, 0.45 mmol) changed the color from green to deep blue.
After stirring for a further 15 min, the solvent was removed under
vacuum. The residue was washed with CHCl3 (10 mL), Et2O (2ꢃ10 mL),
and dried under vacuum to give 3 as a blue solid (310 mg, 88%).
1H NMR (500 MHz, CD2Cl2): d=13.9 (d, 3JHH =7.5 Hz, 1H; H5), 7.9–6.9
(m, 46H; PPh3 and H3), 6.6 ppm (t, 3JHH =7.5 Hz, 1H; H4); 31P NMR
(202 MHz, CD2Cl2): d=14.7 (s; CPPh3), À4.4 ppm (s; OsPPh3);
Figure 4. The spatial plot of the LUMO (isovalue=0.04) of dicationic
osmabenzyne B*. All hydrogen atoms are omitted for clarity.
(19.7%) suggests that the nucleophilic attack at C3 is most
favorable. Furthermore, the Gibbs free energy of C* is
6.3 kcalmolÀ1 lower than that of C# (Scheme 4). Thus, steric
hindrance and electronic effects contribute synergistically to
the regioselectivity for nucleophilic attack at C3.
13C NMR plus HSQC (126 MHz, CD2Cl2): d=287.9 (m; C1), 246.7 (br;
3
C5), 159.0 (d, 2JPC =8.8 Hz; C3), 135.0–127.2 (m; PPh3), 125.0 (d, JPC
=
7.8 Hz; C4), 94.7 ppm (d, 1JPC =110.7 Hz; C2); HRMS (ESI): m/z calcd
for C59H48P3Cl2Os: 1111.1961 [M]+; found: 1111.1956; elemental analysis
calcd (%) for C59H48P3Cl3Os: C 61.81, H 4.22; found: C 61.41, H 4.46.
Compound 4a:
A
solution of
3
(120 mg,
0.10 mmol) in THF (7 mL) was cooled to À108C.
EtLi (0.5m in benzene and cyclohexane, 240 mL,
0.12 mmol) was added slowly, and the mixture
was allowed to warm to RT and stirred for anoth-
er 30 min, giving a brownish-red solution. The sol-
Conclusion
We have applied the formal [3+3] cycloaddition reaction to
the preparation of metallabenzynes. The product is unique
&
4
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Conversions of Osmabenzyne and Isoosmabenzene
FULL PAPER
vent was removed under vacuum. The residue was washed with Et2O (3ꢃ
10 mL), and dried under vacuum to give 4a as a red solid (81 mg, 68%).
1H NMR (500 MHz, CDCl3): d=8.0 (d, 3JHH =9.5 Hz, 1H; H5), 7.8–6.8
(m, 45H; PPh3), 4.9 (m, 1H; H4), 3.8 (m, 1H; H3), 0.7 (m, 1H; H6), 0.4
vacuum to give 5a’ as a red solid (309 mg, 92%). 1H NMR (500 MHz,
CDCl3): d=13.6 (d, JHH =12.5 Hz, 1H; H3), 7.8–6.8 (m, 45H; PPh3), 5.8
3
(m, 2H; H4 and H5), 4.3 (m, 1H; OsCHCHCHNH), 2.3 (m, 2H; H6),
2.0 (m, 2H; NH2CH2CH2CH2CH3), 1.8 (m, 2H; H6’), 0.8–0.3 ppm (m,
14H; H7, H8, H9, H7’, H8’ and H9’); 31P NMR (202 MHz, CDCl3): d=
4.9 (s; CPPh3), À26.6 ppm (s; OsPPh3); 13C NMR plus HSQC (126 MHz,
CDCl3): d=240.0 (m; C3), 185.0 (s; C1), 164.1 (m; BPh4), 159.2 (br; 2C,
C4 and C5), 122.6 (d, 1JPC =91.8 Hz; C2), 136.2–121.8 (m; PPh3), 43.5 (s;
C6), 43.1 (s; C6’), 34.6 (s; C7), 29.7 (s; C7’), 19.7 (s; C8), 19.5 (s; C8’),
13.7 (s; C9), 13.5 ppm (s; C9’); elemental analysis calcd (%) for
C91H89BN2P3ClOs: C 70.97, H 5.82, N 1.82; found: C 71.06, H 5.64, N
1.76.
(m, 1H; H6), 0.3ppm (m, 3H; H7); 31P NMR (202 MHz, CDCl3): d=2.2
2
(s; CPPh3), À3.0 (d, 2JPP =362.5 Hz; OsPPh3), À6.9 ppm (d, JPP
=
362.5 Hz; OsPPh3); HRMS (ESI): m/z calcd for C61H53P3ClOs: 1105.2663
[MÀCl]+; found: 1105.2665.
Compound 4b: In an NMR tube, compound 3 (13 mg, 0.01 mmol) and
NaSMe (6.4 mg, 0.09 mmol) were mixed in CD2Cl2 (0.4 mL). A brown-
red solution was obtained after 30 min at RT. The
NMR spectra were then collected (NMR yield:
80%). 1H NMR (500 MHz, CD2Cl2): d=7.8 (t,
Compound 5b: In an NMR tube, compound 3 (14 mg, 0.01 mmol) and 2-
propargylamine (4 mL, 0.06 mmol) were mixed in CD2Cl2 (0.4 mL). A red
solution was produced after 1 h at RT. Then, the NMR spectra were col-
3JPH =7.5 Hz, 1H; H5), 4.7 (br, 1H; H4), 4.6 (m,
1H; H3), 1.5 ppm (s, 3H; H6); 31P NMR (202 MHz,
lected (NMR yield: 93%). 1H NMR
2
CD2Cl2): d=4.6 (s; CPPh3), À5.2 (d, JPP =364.5 Hz;
3
2
(500 MHz, CD2Cl2): d=13.7 (d, JHH
=
OsPPh3), À7.6 ppm (d, JPP =364.5 Hz; OsPPh3).
14.0 Hz, 1H; H3), 10.0 (br, 1H;
Compound 5a: nBuNH2 (98 mL, 1.1 mmol) was added to a solution of 3
OsCHCHCHNH), 6.7 (m, 1H; H5), 6.1 (t,
(240 mg, 0.21 mmol) in CH2Cl2 (13 mL). The reaction mixture was stirred
3JHH =14.0 Hz, 1H; H4), 3.2 (br, 2H;
NH2CH2CCH), 2.5 (br, 2H; H6), 2.3 (br,
2H; H6’), 1.9 (s, 1H; H8), 1.8 ppm (s, 1H; H8’); 31P NMR (202 MHz,
CD2Cl2): d=5.8 (s; CPPh3, À22.2 ppm (s; OsPPh3).
X-ray crystallography: All single crystals were mounted on glass fibers
and transferred into a cold stream of nitrogen. Diffraction data for 3
were obtained on an Oxford Gemini-S Ultra charge coupled device
(CCD) diffractometer at 123(2) K, with graphite-monochromated MoKa
radiation (l=0.71073 ꢀ). Diffraction data for 5a’ were collected on a
Bruker CCD diffractometer at 173(2) K, with monochromated MoKa ra-
diation (l=0.71073 ꢀ). Semiempirical or multiscan absorption correc-
tions (SADABS) were applied.[17] Structures were solved by direct meth-
ods or the Patterson function, completed by subsequent difference Fouri-
er map calculations, and refined by full matrix least-squares on F2 with
the SHELXTL program package.[18] Nonhydrogen atoms were refined
anisotropically unless otherwise stated. Hydrogen atoms were placed at
idealized positions and assumed the riding model. Details of crystal data,
data collection, and refinements are summarized in Table 1.
at RT for 1 h to give a red solution (NMR yield: 99%). The solution was
concentrated to approximately 2 mL, and passed through a syringe filter
into a mixed hexane/Et2O solution (40 mL, 3:1 v/v). The precipitate was
collected, washed with Et2O (3ꢃ10 mL) and dried under vacuum to give
5a as a red solid (216 mg, 82%). 1H NMR (500 MHz, CDCl3): d=13.5
(d, 3JHH =13.0 Hz, 1H; H3), 9.5 (br, 1H; OsCHCHCHNH), 7.9–6.8 (m,
45H; PPh3), 5.8 (t, 3JHH =13.0 Hz, 1H; H4), 3.5 (m, 1H; H5), 2.4 (br,
2H; H6), 2.2 (m, 2H; NH2CH2CH2CH2CH3), 1.9 (m, 2H; H6’), 1.0–
0.3 ppm (m, 14H; H7, H8, H9, H7’, H8’ and H9’); 31P NMR (202 MHz,
CDCl3): d=5.7 (s; CPPh3), À26.8 ppm (s; OsPPh3); 13C NMR plus
HSQC (126 MHz, CDCl3): d=235.1 (m; C3), 186.3 (s; C1), 160.6 (br; C4
and C5), 135.0–127.2 (m; PPh3), 123.4 (d, 1JPC =92.0 Hz; C2), 43.8 (s;
C6), 43.0 (s; C6’), 34.6 (s; C7), 30.1
For complex 3, crystals suitable for X-ray diffraction were grown from a
solution in ClCH2CH2Cl layered with hexane. One chlorine atom of a
(s; C7’), 20.1 (s; C8), 19.5 (s; C8’),
Table 1. Crystal data and structure refinement for 3 and 5a’.
13.6 (s; C9), 13.5 ppm (s; C9’); ele-
mental analysis calcd (%) for
C67H69N2P3Cl2Os: C 64.05, H 5.54, N
2.23; found: C 64.04, H 5.61, N 1.86.
3·2C2H4Cl2·2H2O
5a’·0.5C2H4Cl2·0.5H2O
formula
Mr
crystal system
space group
a [ꢀ]
b [ꢀ]
c [ꢀ]
a [8]
b [8]
C59H48Cl3OsP3·2C2H4Cl2·2H2O
C91H89BClN2OsP3·0.5C2H4Cl2·0.5H2O
1380.37
triclinic
1598.50
triclinic
Compound 5a’: nBuNH2 (100 mL,
1.1 mmol) was added to a solution of
¯
¯
P1
P1
11.3929(3)
14.1341(4)
19.6739(5)
104.123(2)
94.344(2)
92.607(2)
3056.76(14)
2
12.052(2)
16.299(3)
23.550(4)
109.731(3)
94.807(4)
104.913(3)
4133.0(12)
2
1.284
1.711
1644
1.36 to 25.00
21194
g [8]
V [ꢀ3]
Z
3
(250 mg, 0.22 mmol) in CH2Cl2
1calcd [gcmÀ3
]
1.500
2.513
1388
m [mmÀ1
F (000)
]
(13 mL). The mixture was stirred at
RT for 1 h to give a red solution. The
solution was concentrated, and the
residue was washed with Et2O and
then redissolved in CH3OH (2 mL).
NaBPh4 (90 mg, 0.26 mmol) was
added to the solution. A red precipi-
tate appeared after stirring for 5 min.
The precipitate was filtered, washed
with CH3OH (2ꢃ2 mL) and Et2O
(2ꢃ10 mL), and then dried under
q range [8]
reflns collected
2.84 to 25.00
25077
independent reflns
observed reflns [Iꢁ2s(I)]
data/restrains/params
GOF on F2
10747
9504
10747/12/713
1.000
14337
12068
14337/30/937
1.006
0.0629/0.1615
0.0748/0.1672
1.604/À1.236
R1/wR2 [Iꢁ2s(I)]
0.0493/0.1104
0.0589/0.1143
1.881/À1.292
R1/wR2 (all data)
largest peak/hole [eꢀÀ3
]
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

X.-Y. Cao, H. Xia et al.
ClCH2CH2Cl solvent molecule was disordered and refined with partial
occupancy factors. Two water solvent molecules were refined with iso-
tropic thermal parameters.
Haley, Angew. Chem. 2002, 114, 3620; Angew. Chem. Int. Ed. 2002,
41, 3470; ruthebenzene: e) H. Zhang, H. Xia, G. He, T. B. Wen, L.
Gong, G. Jia, Angew. Chem. 2006, 118, 2986; Angew. Chem. Int. Ed.
2006, 45, 2920; f) J. Yang, W. M. Jones, J. K. Dixon, N. T. Allison, J.
Am. Chem. Soc. 1995, 117, 9776.
For complex 5a’, crystals suitable for X-ray diffraction were grown from
a solution in ClCH2CH2Cl layered with hexane. The solvent water mole-
cule and carbon atoms of the solvent ClCH2CH2Cl were refined with iso-
tropic thermal parameters.
[6] For electrophilic aromatic substitution reactions of metallabenzenes,
see: a) G. R. Clark, P. M. Johns, W. R. Roper, L. J. Wright, Organo-
metallics 2008, 27, 451; b) C. E. F. Rickard, W. R. Roper, S. D.
Woodgate, L. J. Wright, Angew. Chem. 2000, 112, 766; Angew.
Chem. Int. Ed. 2000, 39, 750.
[7] For nucleophilic aromatic substitution reactions of metallabenzenes,
see: a) R. Lin, H. Zhang, S. Li, J. Wang, H. Xia, Chem. Eur. J. 2011,
17, 4223; b) G. R. Clark, L. A. Ferguson, A. E. McIntosh, T. Sçhnel,
L. J. Wright, J. Am. Chem. Soc. 2010, 132, 13443; c) T. Wang, S. Li,
H. Zhang, R. Lin, F. Han, Y. Lin, T. B. Wen, H. Xia, Angew. Chem.
2009, 121, 6575; Angew. Chem. Int. Ed. 2009, 48, 6453.
[8] For nucleophilic addition reactions of metallabenzenes, see: a) H.
Zhang, R. Lin, G. Hong, T. Wang, T. B. Wen, H. Xia, Chem. Eur. J.
2010, 16, 6999; b) M. Paneque, C. M. Posadas, M. L. Poveda, N.
Rendꢄn, L. L. Santos, E. ꢅlvarez, V. Salazar, K. Mereiter, E. OÇate,
Organometallics 2007, 26, 3403; c) M. Paneque, C. M. Posadas, M. L.
Poveda, N. Rendꢄn, V. Salazar, E. OÇate, K. Mereiter, J. Am.
Chem. Soc. 2003, 125, 9898.
CCDC-857968 (3) and CCDC-857969 (5a’) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Computational details: All of the structures were optimized at the
B3LYP level of density functional theory.[19] Frequency calculations were
also performed to confirm the characteristics of the calculated structures
as minima or transition states. In B3LYP calculations, the effective core
potentials (ECPs) of Hay and Wadt with a double-z valence basis set
(LanL2DZ) were used to describe Os, P, and Cl atoms, and the standard
6–31G(d) basis set was used for C, N, and H.[20] Polarization functions
were added for Os (z(f)=0.886), Cl (z(d)=0.514), and P (z(d)=0.34).[21]
To study the solvation effects, all structures were optimized by the IEF-
PCM model (radii=UAKS) with dichloromethane as the solvent.[22] All
calculations were performed with the Gaussian 03 software package.[23]
[9] For preparation of metallabenzynes, see: a) J. Chen, H. H. Y. Sung,
I. D. Williams, Z. Lin, G. Jia, Angew. Chem. 2011, 123, 10863;
Angew. Chem. Int. Ed. 2011, 50, 10675; b) G. He, J. Zhu, W. Y.
Hung, T. B. Wen, H. H. Y. Sung, I. D. Williams, Z. Lin, G. Jia,
Angew. Chem. 2007, 119, 9223; Angew. Chem. Int. Ed. 2007, 46,
9065; c) T. B. Wen, W. Y. Hung, H. H. Y. Sung, I. D. Williams, G. Jia,
J. Am. Chem. Soc. 2005, 127, 2856; d) T. B. Wen, Z. Y. Zhou, G. Jia,
Angew. Chem. 2001, 113, 2005; Angew. Chem. Int. Ed. 2001, 40,
1951; for electrophilic substitution reactions of metallabenzynes,
see: e) W. Y. Hung, B. Liu, W. Shou, T. B. Wen, C. Shi, H. H. Y.
Sung, I. D. Williams, Z. Lin, G. Jia, J. Am. Chem. Soc. 2011, 133,
18350; f) T. B. Wen, S. M. Ng, W. Y. Hung, Z. Y. Zhou, M. F. Lo,
L. Y. Shek, I. D. Williams, Z. Lin, G. Jia, J. Am. Chem. Soc. 2003,
125, 884; for nucleophilic addition reactions of metallabenzynes,
see: g) W. Y. Hung, J. Zhu, T. B. Wen, K. P. Yu, H. H. Y. Sung, I. D.
Williams, Z. Lin, G. Jia, J. Am. Chem. Soc. 2006, 128, 13742; for the
conversion of metallabenzynes into carbene complexes, see: h) J.
Chen, C. Shi, H. H. Y. Sung, I. D. Williams, Z. Lin, G. Jia, Angew.
Chem. 2011, 123, 7433; Angew. Chem. Int. Ed. 2011, 50, 7295; for
theoretical studies, see: i) S. M. Ng, X. Huang, T. B. Wen, G. Jia, Z.
Lin, Organometallics 2003, 22, 3898; for highlights, see: j) W. R.
Roper, Angew. Chem. 2001, 113, 2506; Angew. Chem. Int. Ed. 2001,
40, 2440.
Acknowledgements
This work was supported by the National Natural Science Foundation of
China (Nos. 20925208 and 21103142) and the National Basic Research
Program of China (973 Program, No. 2012CB821600)
[1] a) J. R. Bleeke, Acc. Chem. Res. 2007, 40, 1035; b) G. Jia, Coord.
Chem. Rev. 2007, 251, 2167; c) G. Jia, Acc. Chem. Res. 2004, 37, 479.
[2] a) C. W. Landorf, M. M. Haley, Angew. Chem. 2006, 118, 4018;
Angew. Chem. Int. Ed. 2006, 45, 3914; b) L. J. Wright, Dalton Trans.
2006, 1821; c) G. He, H. Xia, G. Jia, Chin. Sci. Bull. 2004, 49, 1543;
d) J. R. Bleeke, Chem. Rev. 2001, 101, 1205.
[3] For recent examples of osmabenzenes, see: a) L. Gong, Z. Chen, Y.
Lin, X. He, T. B. Wen, X. Xu, H. Xia, Chem. Eur. J. 2009, 15, 6258;
b) L. Gong, Y. Lin, T. B. Wen, H. Zhang, B. Zeng, H. Xia, Organo-
metallics 2008, 27, 2584; c) G. R. Clark, P. M. Johns, W. R. Roper,
L. J. Wright, Organometallics 2006, 25, 1771; d) H. Xia, G. He, H.
Zhang, T. B. Wen, H. H. Y. Sung, I. D. Williams, G. Jia, J. Am.
Chem. Soc. 2004, 126, 6862.
[10] a) Q. Zhao, L. Gong, C. Xu, J. Zhu, X. He, H. Xia, Angew. Chem.
2011, 123, 1390; Angew. Chem. Int. Ed. 2011, 50, 1354; b) P. Barrio,
M. A. Esteruelas, E. OÇate, J. Am. Chem. Soc. 2004, 126, 1946.
[11] For recent reviews of benzynes, see: a) K. J. Cahill, A. Ajaz, R. P.
Johnson, Aust. J. Chem. 2010, 63, 1007; b) H. H. Wenk, M. Winkler,
W. Sander, Angew. Chem. 2003, 115, 518; Angew. Chem. Int. Ed.
2003, 42, 502; c) W. Sander, Acc. Chem. Res. 1999, 32, 669.
[12] For examples of benzynes, see: a) W. Sander, M. Exner, M. Winkler,
A. Balster, A. Hjerpe, E. Kraka, D. Cremer, J. Am. Chem. Soc.
2002, 124, 13072; b) H. H. Wenk, A. Balster, W. Sander, D. A.
Hrovat, W. T. Borden, Angew. Chem. 2001, 113, 2356; Angew.
Chem. Int. Ed. 2001, 40, 2295; c) R. Marquardt, A. Balster, W.
Sander, E. Kraka, D. Cremer, J. G. Radziszewski, Angew. Chem.
1998, 110, 1001; Angew. Chem. Int. Ed. 1998, 37, 955; d) A. Z. Brad-
ley, R. P. Johnson, J. Am. Chem. Soc. 1997, 119, 9917.
[13] For examples of isobenzenes, see: a) A. Ajaz, A. Z. Bradley, R. C.
Burrell, W. H. H. Li, K. J. Daoust, L. B. Bovee, K. J. DiRico, R. P.
Johnson, J. Org. Chem. 2011, 76, 9320; b) I. D. Mackie, R. P. John-
son, J. Org. Chem. 2009, 74, 499; c) K. J. Daoust, S. M. Hernandez,
K. M. Konrad, I. D. Mackie, J. Winstanley, Jr., R. P. Johnson, J. Org.
Chem. 2006, 71, 5708; d) B. Engels, J. C. Schçneboom, A. F. Mꢆn-
ster, S. Groetsch, M. Christl, J. Am. Chem. Soc. 2002, 124, 287; e) M.
Prall, A. Krꢆger, P. R. Schreiner, H. Hopf, Chem. Eur. J. 2001, 7,
[4] For recent examples of iridabenzenes, see: a) A. F. Dalebrook, L. J.
Wright, Organometallics 2009, 28, 5536; b) M. Paneque, M. L.
Poveda, N. Rendꢄn, E. ꢅlvarez, E. Carmona, Eur. J. Inorg. Chem.
2007, 2711; c) H. P. Wu, D. H. Ess, S. Lanza, T. J. R. Weakley, K. N.
Houk, K. K. Baldridge, M. M. Haley, Organometallics 2007, 26,
3957; d) G. R. Clark, G. L. Lu, W. R. Roper, L. J. Wright, Organo-
metallics 2007, 26, 2167; e) E. ꢅlvarez, M. Paneque, M. L. Poveda,
N. Rendꢄn, Angew. Chem. 2006, 118, 488; Angew. Chem. Int. Ed.
2006, 45, 474; f) K. Ilg, M. Paneque, M. L. Poveda, N. Rendꢄn, L. L.
Santos, E. Camona, K. Mereiter, Organometallics 2006, 25, 2230;
g) H. P. Wu, T. J. R. Weakley, M. M. Haley, Chem. Eur. J. 2005, 11,
1191; h) C. S. Chin, H. Lee, M. S. Eum, Organometallics 2005, 24,
4849; i) C. S. Chin, H. Lee, Chem. Eur. J. 2004, 10, 4518; j) J. R.
Bleeke, R. Behm, J. Am. Chem. Soc. 1997, 119, 8503; k) J. R.
Bleeke, R. Behm, Y. F. Xie, M. Y. Chiang, K. D. Robinson, A. M.
Beatty, Organometallics 1997, 16, 606.
[5] For recent examples of other metallabenzenes, see: rhenabenzene:
a) K. C. Poon, L. Liu, T. Guo, J. Li, H. H. Y. Sung, I. D. Williams, Z.
Lin, G. Jia, Angew. Chem. 2010, 122, 2819; Angew. Chem. Int. Ed.
2010, 49, 2759; platinabenzenes: b) V. Jacob, C. W. Landorf, L. N.
Zakharov, T. J. R. Weakley, M. M. Haley, Organometallics 2009, 28,
5183; c) C. W. Landorf, V. Jacob, T. J. R. Weakley, M. M. Haley, Or-
ganometallics 2004, 23, 1174; d) V. Jacob, T. J. R. Weakley, M. M.
&
6
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Conversions of Osmabenzyne and Isoosmabenzene
FULL PAPER
4386; f) R. C. Burrell, K. J. Daoust, A. Z. Bradley, K. J. DiRico,
R. P. Johnson, J. Am. Chem. Soc. 1996, 118, 4218.
[14] B. Liu, H. Xie, H. Wang, L. Wu, Q. Zhao, J. Chen, T. B. Wen, Z.
Cao, H. Xia, Angew. Chem. 2009, 121, 5569; Angew. Chem. Int. Ed.
2009, 48, 5461.
[23] Gaussian 03, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[15] See the Supporting Information for details.
[16] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899.
[17] G. M. Sheldrick, SADABS, Program for Semiempirical Absorption
Correction, University of Gçttingen, Gçttingen, 1997.
[18] G. M. Sheldrick, SHELXTL, Siemens Analytical X-ray Systems,
Madison, Wisconsin.
[19] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) B. Miehlich, A.
Savin, H. Stoll, H. Preuss, Chem. Phys. Lett. 1989, 157, 200; c) C.
Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[20] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[21] Gaussian Basis Sets for Molecular Calculations (Ed.: S. Huzinaga),
Elsevier Science, Amsterdam, 1984.
[22] a) M. T. Cancꢇs, B. Mennucci, J. Tomasi, J. Chem. Phys. 1997, 107,
3032; b) M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem. Phys.
Lett. 1998, 286, 253; c) B. Mennucci, J. Tomasi, J. Chem. Phys. 1997,
106, 5151; d) M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem.
Phys. 2002, 117, 43.
Received: May 4, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

X.-Y. Cao, H. Xia et al.
Metallacycles
Q. Zhao, J. Zhu, Z.-A. Huang,
X.-Y. Cao,* H. Xia* ......... &&&& — &&&&
Metallacycles under attack: The first
formal interconversion of metallaben-
zyne and isometallabenzene is ach-
ieved. The osmabenzyne (left) is
formed via an isoosmabenzene
philic attack at osmabenzyne restores
the isoosmabenzenes or provides ring-
opened products (right). The regiose-
lectivity of nucleophilic additions is
rationalized by DFT calculations.
Conversions of Osmabenzyne and
Isoosmabenzene
(middle) intermediate, and nucleo-
The interconversion of metallabenzyne…… and
isometallabenzene has been achieved for the first
time. An isoosmabenzene intermediate undergoes
elimination reaction to give an osmabenzyne. Regio-
selective nucleophilic attack at the osmabenzyne
restores isoosmabenzenes. DFT calculation suggests
that steric hindrance and electronic effect contribute
synergistically to this regioselectivity.
&
8
&
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!