Simple gold-catalyzed synthesis of benzofulvenes-gem-Diaurated species as "instant dual-activation" precatalysts

Article abstract of DOI:10.1002/anie.201109183

Alkyl-substituted diynes deliver benzofulvenes in a unique gold-catalyzed reaction. The catalytic cycle involves the formation of gold acetylides by alkynyl C-H activation, the formation of vinylidene gold(I) intermediates by dual activation, and alkyl C-H activation by the vinylidene gold(I) species. gem-Diaurated species obtained from the catalysis reactions prove to be highly active catalysts for these conversions. Copyright

Full text of DOI:10.1002/anie.201109183

.
Angewandte
Communications
DOI: 10.1002/anie.201109183
Gold Catalysis
Simple Gold-Catalyzed Synthesis of Benzofulvenes—gem-Diaurated
Species as “Instant Dual-Activation” Precatalysts**
A. Stephen K. Hashmi,* Ingo Braun, Pascal Nçsel, Johannes Schꢀdlich, Marcel Wieteck,
Matthias Rudolph, and Frank Rominger
In the field of homogeneous transition-metal chemistry, gold
plays a major role in the discovery of new reactions.[1] Most of
the transformations are based on the electrophilic activation
of a multiple bond. The decrease of electron density upon
p coordination of a carbophilic gold center allows nucleo-
philic attack. Now, in independent and parallel work the
group of Zhang and our team have discovered a different
reactivity which was so far unprecedented in the field of gold
chemistry.[2] Initial activation of an alkyne by s coordination
to gold increases the nucleophilicity of the b-carbon atom of
this alkyne. This first alkyne combines with a second alkyne,
which is activated by p coordination, and highly reactive gold
vinylidene intermediates are formed by this dual s/p activa-
tion. The Zhang groupꢀs publication[2b] on the synthesis of
benzofulvenes now prompts us to publish our additional
findings on that specific reaction.
Information); the well-established catalyst [(IPr)AuCl],[4] in
combination with AgNTf2 gave the best results.
To evaluate the substrate scope, a small library of
substrates was transformed to the corresponding products
under the optimized conditions (Table 1). One advantage of
substrates 1 is their easy two-step synthesis (Sonogashira
reaction, Seyferth–Gilbert homologation) from commercially
available 2-bromobenzaldehydes. With our test substrate 1a,
Table 1: Gold-catalyzed synthesis of benzofulvene derivatives.
Entry Substrate
Product
Yield
Since alkyl-substituted alkynes are readily available,
a new fascinating reaction pathway was opened for the gold
vinylidene intermediates. The first two elementary steps of
this reaction are identical to the previous reactions,[2a,c] which
seem to be general for this new sector of gold catalysis. Herein
we report on the use of tertiary alkyl groups, the isolation of
gem-diaurated species which prove to be ideal precatalysts for
this type of transformation, the dynamics of the equilibrium
involving these diaurated species, and the catalyst transfer in
the context of a detailed mechanistic discussion.
1[a]
1a
1b
2a 73%
2
2b 68%
We investigated several diyne systems 1 with a terminal
alkynyl group and a tert-alkyl-substituted alkynyl group as
potential substrates for the gold(I)-catalyzed intermolecular
arene addition.[2a] A new reaction was observed, the clean
formation of a benzofulvene derivative, a class of compounds
that usually is not easily available.[3] No incorporation of the
solvent benzene was observed. To optimize the reaction, we
performed reactions of substrate 1a (Table 1) with different
gold catalysts and different counterions (see the Supporting
3
4
5
6
7
8
1c
1d
1e
1 f
2c 41%
2d 43%
2e 92%
2 f 73%
2g 55%
2h 30%
[*] Prof. Dr. A. S. K. Hashmi, Dipl.-Chem. I. Braun,
Dipl.-Chem. P. Nçsel, Dipl.-Chem. J. Schꢀdlich, M. Sc. M. Wieteck,
Dr. M. Rudolph, Dr. F. Rominger[+]
Organisch-Chemisches Institut
Ruprecht-Karls-Universitꢀt Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
E-mail: hashmi@hashmi.de
Homepage: http://www.hashmi.de
1g
1h
[+] Crystallographic investigation
[**] We thank Umicore AG & Co. KG for the generous donation of gold
salts.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201109183.
4456
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4456 –4460

Angewandte
Chemie
Table 1: (Continued)
Entry Substrate
Product
Yield
49%
2i
9
1i
1j
Scheme 1. Synthesis of the vinyl monogold compound 5. IPr=1,3-
bis(diisopropylphenyl)imidazolylidene.
3a 9%
3b 4%
10
[a] Only 2 mol% [(IPr)AuNTf2], 30 h.
no problems occurred during column chromatography and
the yield of isolated 2a was in the range of the yield
determined by gas chromatography (Table 1, entry 1). Similar
reactivity was observed for benzodioxole derivative 1b
(Table 1, entry 2). Reduced yields were observed for sub-
strates 1c and 1d with electron-donating substituents in meta
and para position, respectively, relative to the terminal
alkynyl group (Table 1, entries 3 and 4). The reason for this
is most probably the instability of the products under the
reaction conditions. Consistent with this observation, in
substrates 1e, 1 f, and 1g electronically neutral (Table 1,
entry 5) and electron-withdrawing substituents (entries 6 and
7) led to a stabilization of the products; the yields were good
to excellent. In the case of the dimethoxy-substituted product
2a, crystals suitable for X-ray crystal structure analysis could
be obtained.[5] The results unambiguously prove the presence
of the benzofulvene core. Changing the tert-butyl group at the
nonterminal alkyne to isopropyl led to a drop in yield, but still
benzofulvene derivative 2h could be isolated in 30% yield
(Table 1, entry 8). A cyclopentyl substituent was also toler-
ated (Table 1, entry 9). In this case, in addition to the expected
product 2i, minor amounts of constitutional isomer 3a were
isolated. The unambiguous structural assignment of 3a was
possible by an X-ray crystal structure analysis.[5] The sensitive
triene substructure likely arises from a cyclization that is
initiated by attack of the nonterminal alkyne onto the
terminal triple bond. Unfortunately, the cyclohexyl-substi-
tuted substrate delivered a product that was unstable during
the workup and isolation; hence only traces of triene 3b could
be obtained.
Further studies focused on the isolation of the reaction
intermediates. We prepared gold acetylide 4. In a first experi-
ment 4 was subjected to catalytic amounts of activated
catalyst. A clean conversion was indicated by thin-layer
chromatography, but purification on silica delivered the fairly
unstable compound 5 in only 40% yield (Scheme 1). Never-
theless, single crystals suitable for X-ray structure analysis
could be grown with extremely careful handling.[5] The
structure analysis proves the formation of the vinyl monogold
complex 5 (Figure 1). The gold center is directly bound to the
alkene double bond, which is derived from the nonterminal
tert-butyl-substituted alkyne. Despite the steric bulk of the
tert-butyl substituent of the substrate, coordination of the gold
Figure 1. X-ray crystal structure of 5 (thermal ellipsoids at 50%
probability).
catalyst at the b-carbon atom of the alkyne must still be
possible.
A second reaction was conducted with gold acetylide 4
and stoichiometric amounts of the activated catalyst in
dichloromethane (Scheme 2). Even at room temperature an
immediate conversion of the starting material took place and
Scheme 2. Synthesis of the IPr/IPr-gem-diaurated compound 6.
precipitation of a crystalline solid was observed upon the
addition of hexane. X-ray crystal structure analysis[5] proved
the formation of a gem-diaurated species (Figure 2). In this
compound the two metal atoms are situated at the same
position of the benzofulvene core as the single gold atom in
compound 5. There are several examples in the literature of
gem-diaurated species that are prepared by the reaction of
monogold organometallic compounds and stoichiometric
amounts of a cationic gold complex.[6] Therefore it is
reasonable that compound 6 is derived from monogold
complex 5 (which itself forms in a previous gold-catalyzed
cyclization reaction). The relevance of the gem-diaurated
compounds in gold-catalyzed reactions so far is unclear.[7]
Besides our findings with arene–diyne substrates,[2a,c] only one
other example for the formation of a gem-diaurated species
out of a catalysis reaction has been reported (derived from an
allene precursor).[8]
Angew. Chem. Int. Ed. 2012, 51, 4456 –4460
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4457

.
Angewandte
Communications
makes complexes of type 6 ideal instant dual-activation
precatalysts, not needing any additives as activators (unlike
the system in Ref. [2b]). In an experiment with 1e and
5 mol% of 4 indeed no conversion was observed; however,
when 10 mol% of HNTf2 was added to generate
[(IPr)AuNTf2] from 4 by protodeauration, conversion to 2e
was immediately detected (72% yield of isolated product).
Thus, shifting the equilibrium depicted in Scheme 4
towards the right side (by means of bulky ligands and/or
elevated temperatures) is crucial not only for this reaction but
also for other gold-catalyzed processes in which gem-diau-
rated compounds are involved.
To our surprise, gold acetylide 4 could also be cyclized in
the presence of stoichiometric amounts of an activated
phosphane complex (Scheme 5). An X-ray crystal structure
analysis of the resulting crystals showed the first example of
Figure 2. X-ray crystal structure of 6 (thermal ellipsoids at 50%
probability).
To determine whether the formation of compound 6 is
only a dead end for the reaction and responsible for catalyst
consumption, catalytic amounts of 6 were added to diyne 1e in
the absence of any additional catalyst, co-catalyst, or activator
(Scheme 3). Under the normal reaction conditions
a 2.5 mol% loading of 6 not only resulted in complete
conversion of the starting material but the reaction was even
Scheme 5. Synthesis of the PPh3/IPr-gem-diaurated compound 7.
a gem-diaurated species containing two different ligands at
the two gold centers, in this case an N-heterocyclic carbene
(NHC) and a phosphane ligand (Figure 3).[5] Interestingly,
1
only traces of other byproducts were evident in the H and
31P NMR spectra and no phosphane/phosphane diaurated
species could be detected. To investigate whether a dynamic
Scheme 3. Threefold acceleration of the formation of benzofulvene 2e
through the use of 2.5 mol% of the gem-diaurated compound 6.
three times faster than with our usual catalyst system (2 h
instead of 6 h)! This observation provides significant insights
into the catalytic cycle and the interesting question of catalyst
transfer. Starting from the normal catalyst [LAu+XÀ], the
formation of the gold acetylide is slow,[9] and only after some
time are the gold acetylide and free [LAu+XÀ] present in
a ratio that is suitable for signficant reaction rates. This leads
to a prolonged reaction time and increases the likelihood of
undesired side reactions. Compound 6, on the other hand, is in
direct equilibrium (Scheme 4) with two species needed for the
catalytic conversion, the cationic catalyst (necessary for
p activation of the starting material) and the vinyl monogold
compound 5 (which directly delivers the s-activated gold
acetylide by the crucial catalyst-transfer step of the catalytic
cycle). Providing both species in a well-defined 1:1 ratio
Figure 3. X-ray crystal structure of 7 (thermal ellipsoids at 50%
probability).
equilibrium exists at room temperature between the mono-
and digold organometallic compounds, we subjected IPr/IPr-
gem-diaurated compound 6 to stoichiometric amounts of
[(PPh3)AuNTf2] (Scheme 6). In perfect agreement with the
findings described above, only the phosphane/carbene gem-
diaurated species 7 was detected by 1H and 31P NMR
spectroscopy (31P NMR (CDCl3): d = 36.7 ppm, clearly differ-
ent from [(PPh3)AuNTf2] with 30.4 ppm). Apparently, there is
a dynamic equilibrium between the digold compound, the
monogold compound, and the activated catalyst (as depicted
Scheme 4. Equilibrium between the gem-diaurated compound 6 and
the vinyl monogold compound 5 as well as the activated gold catalyst.
4458
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4456 –4460

Angewandte
Chemie
Scheme 7. Isotopic-labeling experiment.
detected (Scheme 7), which shows the importance of the
catalyst transfer.
Our proposed mechanism for the reaction is summarized
in Scheme 8. As in the related reaction pathways,[2a,c] the
formation of a gold acetylide 4 must take place as the first
reaction step. One option is simple formation of an acetylide
and liberation of the corresponding acid,[14] the other is the
catalyst transfer. Once the acetylide is formed, dual s/
p activation takes place, leading to the fast formation of
reactive gold vinylidene II. At this stage, in the case of the
alkyl-substituted substrates, instead of the nucleophilic addi-
tion of an sp2 nucleophile,[2a] a 1,5-hydride shift to the
electrophilic vinylidene carbon takes place. This represents
a fascinating C–H activation of an unactivated alkyl group.
This leads to cation III which is stabilized by the gold atom.
(That is the vinylogous case of the well-known discussion of
“carbenes or carbocations” in gold chemistry.)[15] The next
step consists of the addition of the vinyl gold double bond
onto the cation, followed by elimination of the gold complex.
Then the similarities between the reactions recur. A monog-
old species 5 is formed, which exists in an equilibrium with the
Scheme 6. Exchange between compound 6 and [(PPh3)AuNTf2].
in Scheme 4) even at room temperature in dichloromethane.
Interestingly, the mixed phosphane/carbene gem-diaurated
species 7 is thermodynamically favored.
A comparison of the key structural parameters of com-
pounds 6 and 7 can be obtained from the X-ray crystallo-
graphic data.[10] The Au Au bond lengths of gem-diaurated
À
compounds 6 (282.9 pm) and 7 (284.7 pm) show strong
aurophilic interactions.[11] However, they are slightly weaker
than those in other reported diaurated compounds.[2a,c,12]
À
À
Remarkably, the Au1 C and the Au2 C bond lengths in
both gem-diaurated compounds are not equivalent. In the IPr/
IPr compound 6 these are 208.6 pm and 215.0 pm. In IPr/PPh3
À
compound 7, the NHC complex (Au1 C: 209.9 pm) is bonded
stronger to the benzofulvene moiety than the phosphane
[13]
À
complex is (Au2 C: 216.6 pm).
Noteworthy, the shorter
gold–carbon bond of the gem-diaurated compounds is only
slightly longer than in the
corresponding
monogold
compound 5. Altogether,
auration has only a minor
effect on the C–C bond
lengths in the benzofulvene
moiety. While the length of
the C–C single bond to the
benzene ring remains almost
unchanged (147.4 pm in 2a,
148.2 pm in 5, 151.3 pm in 6,
and 148.2 pm in 7), the
double bond in the gem-di-
aurated
compounds
6
(138.3 pm) and 7 (137.5 pm)
is somewhat longer than in
benzofulvene 2a (132.3 pm).
In summary, the gem-diau-
rated species obtained in this
study show a bonding mode
that can be considered to be
a three-center two-electron
bond.
The conversion of the
deuterium-labeled terminal
alkyne [D]-1e was also inves-
tigated. A nearly quantita-
tive incorporation of the
deuterium in [D]-2e at the
former gold position was Scheme 8. Mechanistic rationale.
Angew. Chem. Int. Ed. 2012, 51, 4456 –4460
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4459

.
Angewandte
Communications
[5] CCDC 859289 (2a), 859290 (3a), 859291 (5), 859292 (6), and
859293 (7) 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.
corresponding gem-diaurated species 6. Finally, catalyst
transfer takes place (responsible for the deuterium labeling
in the product) and another molecule of the starting material
is activated through acetylide formation.
The reaction reported in this contribution offers a signifi-
cantly simplified entry to the benzofulvene skeleton from
simple starting materials. In addition, the role of gem-
diaurated species from gold-catalyzed reactions could be
clarified; they are excellent instant dual-activation catalysts
for reactions involving s,p-type dual activation of the
substrate. Based on these species, the full catalytic cycle,
including the catalyst transfer, is now understood. In our
opinion, this general reactivity pattern and this new type of
catalyst will trigger the discovery of a series of new reactions.
[6] a) A. N. Nesmeyanov, E. G. Perevalova, K. I. Grandberg, D. A.
Lemenovskii, T. V. Baukova, O. B. Afanassova, J. Organomet.
Chem. 1974, 65, 131 – 144; b) V. G. Andrianov, Y. T. Struchkov,
E. R. Rossinskaya, J. Struct. Chem. 1974, 15, 65 – 72; c) H.
Schmidbaur, Y. Inoguchi, Chem. Ber. 1980, 113, 1646 – 1653;
d) R. Usón, A. Laguna, P. Brun, J. Organomet. Chem. 1980, 197,
369 – 373; e) R. Usón, A. Laguna, E. J. Fernµndez, A. Mendia,
P. G. Jones, J. Organomet. Chem. 1988, 350, 129 – 138; f) R.
Usón, A. Laguna, E. J. Fernµndez, M. E. Ruiz-Romero, P. G.
Jones, J. Lautner, J. Chem. Soc. Dalton Trans. 1989, 2127 – 2131;
g) T. V. Baukova, L. G. Kuzꢀmina, N. A. Oleinikova, D. A.
Lemenovskii, A. L. Blumenfelꢀd, J. Organomet. Chem. 1997,
530, 27 – 38; h) O. Schuster, A. Schier, H. Schmidbaur, Organo-
metallics 2003, 22, 4079 – 4083; i) K. A. Porter, A. Schier, H.
Schmidbaur, Organometallics 2003, 22, 4922 – 4927; j) M. Osawa,
M. Hoshino, D. Hashizume, Dalton Trans. 2008, 2248 – 2252;
k) G. Seidel, C. W. Lehmann, A. Fürstner, Angew. Chem. 2010,
122, 8644 – 8648; Angew. Chem. Int. Ed. 2010, 49, 8466 – 8470;
l) P. H.-Y. Cheong, P. Morganelli, M. R. Luzung, K. N. Houk,
F. D. Toste, J. Am. Chem. Soc. 2008, 130, 4517 – 4526; m) T. N.
Hooper, M. Green, C. A. Russell, Chem. Commun. 2010, 46,
2313 – 2315.
Received: December 27, 2011
Revised: January 19, 2012
Published online: March 19, 2012
Keywords: alkynes · benzofulvenes · gold ·
.
homogeneous catalysis · vinylgold intermediates
[1] a) A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118,
8064 – 8105; Angew. Chem. Int. Ed. 2006, 45, 7896 – 7936; b) E.
Jimꢁnez-Nfflnez, A. M. Echavarren, Chem. Commun. 2007, 333 –
346; c) R. Skouta, C.-J. Li, Tetrahedron 2008, 64, 4917 – 4938;
d) Z. Li, C. Brouwer, C. He, Chem. Rev. 2008, 108, 3239 – 3265;
e) A. Arcadi, Chem. Rev. 2008, 108, 3266 – 3325; f) D. J. Gorin,
B. D. Sherry, F. D. Toste, Chem. Rev. 2008, 108, 3351 – 3378;
g) A. S. K. Hashmi, M. Rudolph, Chem. Soc. Rev. 2008, 37,
1766 – 1775; h) S. Sengupta, X. Shi, ChemCatChem 2010, 2, 609 –
619; i) C. Nevado, Chimia 2010, 64, 247 – 251; j) A. Corma, A.
Leyva-Pꢁrez, M. J. Sabater, Chem. Rev. 2011, 111, 1657 – 1712.
[2] a) A. S. K. Hashmi, I. Braun, M. Rudolph, F. Rominger,
Organometallics 2012, 31, 644 – 661; b) L. Ye, Y. Wang, D. H.
Aue, L. Zhang, J. Am. Chem. Soc. 2012, 134, 31 – 34; c) A. S. K.
Hashmi, M. Wieteck, I. Braun, P. Nçsel, L. Jongbloed, M.
Rudolph, F. Rominger, Adv. Synth. Catal. 2012, 354, 555 – 562.
[3] a) W. Baker, F. Glockling, J. F. W. McOmie, J. Chem. Soc. 1951,
3357 – 3361; b) R. N. Warrener, P. A. Harrison, R. A. Russell, J.
Chem. Soc. Chem. Commun. 1982, 1134 – 1136; c) M. Schmittel,
M. Strittmatter, Tetrahedron 1998, 54, 13751 – 13760; d) S. Datta,
R.-S. Liu, Tetrahedron Lett. 2005, 46, 7985 – 7988; e) J.-J. Lian, P.-
C. Chen, Y.-P. Lin, H.-C. Ting, R.-S. Liu, J. Am. Chem. Soc. 2006,
128, 11372 – 11373; f) D. Aburano, F. Inagaki, S. Tomonaga, C.
Mukai, J. Org. Chem. 2009, 74, 5590 – 5594; g) C. Werner, H.
Hopf, J. Grunenberg, P. G. Jones, Eur. J. Org. Chem. 2010, 4027 –
4034; h) B. Y. Yao, Y. Li, Z. Liang, Y. Zhang, Org. Lett. 2011, 13,
640 – 643.
[7] D. Weber, T. D. Jones, L. L. Adduci, M. R. Gagnꢁ, Angew.
Chem. 2012, 124, 2502 – 2506; Angew. Chem. Int. Ed. 2012, 51,
2452 – 2456; see also Ref. [6k].
[8] a) D. Weber, M. A. Tarselli, M. R. Gagnꢁ, Angew. Chem. 2009,
121, 5843 – 5846; Angew. Chem. Int. Ed. 2009, 48, 5733 – 5736;
b) D. Weber, M. R. Gagnꢁ, Org. Lett. 2009, 11, 4962 – 4965.
[9] This initiation period has been observed in detail in Ref. [2a].
[10] A summary of the relevant data gained from the X-ray
crystallographic analyses of the organogold intermediates 5, 6,
and 7 is given in the Supporting Information.
[11] Selected examples of “aurophilicity”: a) H. Schmidbaur, Gold
Bull. 1990, 23, 11 – 21; b) H. Schmidbaur, Chem. Soc. Rev. 1995,
24, 391 – 400.
À
[12] Au Au distances for gem-diaurated compounds found in the
literature: 272.7–284.6 pm. Average value (seven structures):
277.8 pm; see Refs. [6a–c], [6e], and [6i–k].
À
[13] Average value from Ref. [6k]: Au C 212 pm.
[14] Y. Odabachian, X. F. Le Goff, F. Gagosz, Chem. Eur. J. 2009, 15,
8966 – 8970.
[15] Selected examples of the “carbene or carbocation?” discussion:
a) A. Fürstner, L. Morency, Angew. Chem. 2008, 120, 5108 –
5111; Angew. Chem. Int. Ed. 2008, 47, 5030 – 5033; b) A. S. K.
Hashmi, Angew. Chem. 2008, 120, 6856 – 6858; Angew. Chem.
Int. Ed. 2008, 47, 6754 – 6756; c) S. Bhunia, R.-S. Liu, J. Am.
Chem. Soc. 2008, 130, 16488 – 16489; d) D. Benitez, N. D.
Shapiro, E. Tkatchouk, Y. Wang, W. A. Goddard III, F. D.
Toste, Nat. Chem. 2009, 1, 482 – 486; e) V. V. Pagar, A. M.
Jadhav, R.-S. Liu, J. Am. Chem. Soc. 2011, 133, 20728 – 20731.
[4] M. R. Fructos, T. R. Belderrain, P. de Frꢁmont, N. M. Scott, S. P.
Nolan, M. M. Díaz-Requejo, P. J. Pꢁrez, Angew. Chem. 2005,
117, 5418 – 5422; Angew. Chem. Int. Ed. 2005, 44, 5284 – 5288.
4460 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4456 –4460