Benzylic Functionalization of Anthrones via the Asymmetric Ring Opening of Oxabicycles Utilizing a Fourth-Generation Rhodium Catalytic System

Article abstract of DOI:10.1002/chem.201502718

While anthrones exist as privileged scaffolds in bioactive molecules, the enantioselective functionalization of anthrones is surprisingly scarce in the literature, with no asymmetric transition metal catalyzed example to date. Herein, we report the first asymmetric transition metal catalyzed benzylic functionalization of anthrones through the rhodium(I) catalyzed desymmetrization of oxabicycles. As previously developed rhodium(I) systems were found to be unsuitable for this substrate, a new robust fourth-generation [Rh(cod)OH]2 based catalytic system was developed to address synthetic challenges in this protocol.

Full text of DOI:10.1002/chem.201502718

DOI: 10.1002/chem.201502718
Communication
&
Asymmetric Catalysis
Benzylic Functionalization of Anthrones via the Asymmetric Ring
Opening of Oxabicycles Utilizing a Fourth-Generation Rhodium
Catalytic System
Charles C. J. Loh,[a] Xiang Fang,[a, b] Brendan Peters,[a] and Mark Lautens*[a]
Abstract: While anthrones exist as privileged scaffolds in
bioactive molecules, the enantioselective functionalization
of anthrones is surprisingly scarce in the literature, with
no asymmetric transition metal catalyzed example to date.
Herein, we report the first asymmetric transition metal cat-
alyzed benzylic functionalization of anthrones through the
rhodium(I) catalyzed desymmetrization of oxabicycles. As
previously developed rhodium(I) systems were found to
be unsuitable for this substrate, a new robust fourth-gen-
eration [Rh(cod)OH]2 based catalytic system was devel-
oped to address synthetic challenges in this protocol.
The anthrone scaffold is an important bioactive motif as it is
present in molecules which possesses anti-cancer properties
by inhibiting tubulin polymerization.[1] It is also well known for
its anti-inflammatory properties in the skin treatment of psoria-
sis,[2] as well as being a key secondary metabolite in fungi bio-
synthetic pathways.[3] Mechanistically, anthrones react through
a unique aromatization/dearomatization pathway via anthrol,
which then functionalizes the benzylic position.[4]
Scheme 1. Comparing known anthrone functionalizations with current work.
However, the development of enantioselective methodolo-
gies which alkylate the anthrone benzylic position is exceed-
ingly rare in the literature. Known protocols generally utilize an
organocatalyst,[5] in most cases those containing a basic
moiety to deprotonate the benzylic proton, which then aroma-
tizes to an anthroxide anion. This anthroxide nucleophile then
dearomatizes analogous to phenols and naphthol dearomatiza-
tion to generate the benzylated product.[6] Asymmetric catalyt-
ic anthrone functionalizations are also highly limited to sub-
strates containing an a,b-unsaturated system, with 1,4-addi-
tions[5a–e] and Diels-Alder reaction catalyzed by cinchona alka-
loids (Kagan et al.)[5a] or chiral guanidines (Tan et al.)[5b] spear-
heading the known reaction in the literature (Scheme 1).[5–6]
Moreover, a survey of the literature revealed a paucity of
transition metal catalyzed protocols to functionalize the ben-
zylic position of anthrones. Examples include zinc catalyzed
protocols for 1,4-addition or Diels-Alder reactions on acrylic
substrates,[7b–c] as well as a ruthenium catalyzed dialkylation of
anthrones with methyl vinyl ketone.[7a] The above mentioned
examples are the sole transition metal catalyzed examples
known in literature, where enantioselective functionalization
was not possible.
The rhodium catalyzed asymmetric ring opening (ARO) reac-
tion of oxabicycles is of long-standing interest in our research
group.[8,9] The first-generation [Rh(cod)Cl]2 based catalytic
system was developed in 2000 for the ARO of alcohols and
amines.[10] Subsequently in 2003, a rhodium iodide based
second-generation system was developed to address short-
comings in enantioselectivities especially with amine and mal-
onate nucleophiles. This protocol was also later extended to
carboxylate and thiol nucleophiles.[11] A third-generation cat-
ionic Rh(cod)2OTf system was later developed in 2010,[12] and
was effective in unreactive oxabicycles possessing quartenary
bridgehead carbons[12a,c] and in ARO reactions with silyl enol
[a] Dr. C. C. J. Loh, Dr. X. Fang, B. Peters, Prof. Dr. M. Lautens
Department of Chemistry, Davenport Chemical Laboratories
University of Toronto, 80 St. George Street
Toronto, Ontario M5S 3H6 (Canada)
E-mail: mlautens@chem.utoronto.ca
[b] Dr. X. Fang
Laboratory for Advanced Material and Institute of Fine Chemicals
School of Chemistry and Molecular Engineering
East China University of Science and Technology
130 Meilong Road, Shanghai 200237 (P. R. China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201502718.
Chem. Eur. J. 2015, 21, 13883 – 13887
13883
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
ethers.[12d] However, a drawback of these systems is the high
reactivity of cationic rhodium which results in undesired oxabi-
cycle decomposition pathways as well as practical handling
challenges. In addition, most ring opening protocols are limit-
ed to nitrogen or oxygen nucleophiles, and the utility of soft
carbon nucleophiles is largely unexplored, with only d2 syn-
thons such as malonates[11a,b] and silyl-enol ethers known.[12d]
Herein, we wish to report the first asymmetric transition
metal catalyzed benzylation of anthrones via the rhodium(I)
catalyzed ARO reaction of oxabicycles utilizing air-stable
[Rh(cod)OH]2 as a robust new generation catalytic system.[13]
Moreover, this protocol features an interesting regioselectivity
where benzylic carbon attacks preferentially over oxygen.[4c–d]
This is significant since the anthroxide nucleophile has the pro-
pensity to react in an analogous fashion to phenol, which is
a well known nucleophile in oxabicycle ring opening reac-
tions.[10b]
Table 1. Optimization of Rh-catalyzed anthrone ARO reaction.[a]
Entry Rh Complex
Solv. (t Cat./Ligand
Yield r.r.
ee
[h])
[mol%/
mol%]
[%][b] [3a/ (maj)
4a][c] [%][d]
1
[Rh(cod)Cl]2
[Rh(cod)Cl]2
[Rh(cod)Cl]2,
AgOTf, TBAI
(1 equiv)
THF (2) 4/8
THF (2) 4/8
THF (2) 4/8
69
20[f]
–
13:1 69
–
–
2[e]
3
N.D.
–
Our investigation was initiated utilizing commercially avail-
able anthrone 1a and subjecting it to the first-generation
asymmetric ring opening conditions using [Rh(cod)Cl]2 com-
plex with (R,S)-PPF-tBu2-Josiphos ligand.[10] While it was encour-
aging that this catalytic system yielded the target ring opening
product in moderate yields (Table 1, entry 1), enantioselectivity
was modest at 69% and the regiomeric ratio (r.r.) was 13:1. We
realized that this methodology provided two regioisomers, the
ring opening product 3a, as well as a minor regiomisomer
4a.[14] Lowering the reaction temperature to room temperature
was detrimental to the system as the reaction was extremely
sluggish, resulting in only 20% yield of the target product
(Table 1, entry 2). Utilizing the second-generation rhodium
iodide catalytic system (Table 1, entry 3) and the third-genera-
tion cationic rhodium systems all gave decomposition of 2a
(Table 1, entry 4, see Supporting Information for complete cata-
lyst screen).[11–12]
4
5
6
7
Rh(cod)2OTf
[Rh(cod)OH]2
[Rh(cod)OH]2
[Rh(cod)OH]2
THF (2) 5/6
THF (2) 4/8
DCE (2) 4/8
–
–
–
74
84
66
7:1
9:1
17:1
97
97
97
MeCN
(2)
4/8
8
9
[Rh(cod)OH]2
[Rh(cod)OH]2
dioxane 4/8
(2)
40
49
96
7:1
8:1
99
99
PhMe
(2)
4/8
4/8
2/4
3/6
10[g] [Rh(cod)OH]2
11[h] [Rh(cod)OH]2
12[h] [Rh(cod)OH]2
MeCN
(4)
20:1 96
N.D.
20:1 97
MeCN
(8)
94[f] 17:1
94
MeCN
(6)
[a] All reactions were conducted using 1a (0.3 mmol) and 2a (0.2 mmol).
[b] Combined isolated yield of 3a+4a after flash column chromatogra-
phy. [c] Determined by 1H NMR analysis of the crude reaction mixture.
[d] Determined by HPLC analysis on a chiral stationary phase. [e] Reaction
was conducted at RT. [f] 1H NMR yield using 1,3,5-trimethoxybenzene as
an internal standard. [g] 2.5 equiv 1a (0.5 mmol) was used. [h] Conducted
using 1a (1.0 mmol) and 2a (0.4 mmol).
The failure of the known rhodium ARO systems spurred the
search other new catalytic systems to address substrate com-
patibility. We were delighted that dimeric [Rh(cod)OH]2 gave
the target product in 74% with an excellent enantioselectivity
of 97% (Table 1, entry 5), albeit with a reduction in regioselec-
tivity. Nonetheless, the discovery of this new catalytic system
gave further impetus to continue the optimization studies.
A subsequent solvent screen provided us with evidence for
solvent effects (Table 1, entries 6–9). While THF has been tradi-
tionally used in most Rh-catalyzed ARO systems to date, it ap-
pears that switching solvents have a profound effect on the re-
gioselectivity of this protocol. The utility of DCE improved the
r.r. to 9:1 with a slightly elevated yield of 84% (Table 1,
entry 6). The use of MeCN significantly elevated the r.r. to 17:1
(Table 1, entry 7), while maintaining the excellent enantoselec-
tivity, albeit with a slight reduction in yield. Other solvents
such as dioxane and toluene gave inferior r.r. values and sup-
pressed yields (Table 1, entries 8–9). It is noteworthy to empha-
size that solvent changes had no deleterious effect on the
enantioselectivity of this new catalytic system.
enantio- and regioselectivity. Finally, an optimization of the cat-
alyst loadings revealed that lower catalyst loadings is detri-
mental for the r.r. At 1 mol% [Rh(cod)OH]2, the r.r. dropped to
11:1 (see Supporting Information). Increasing the catalyst load-
ing to 2 mol% revealed an increase of r.r. to 17:1 (Table 1,
entry 11). Our optimized condition was finalized at 3 mol% cat-
alyst loading (Table 1, entry 12) since all crucial parameters,
yield, r.r. and enantioselectivities were maximal. With these
conditions in hand, we proceeded to examine the substrate
scope of this protocol utilizing the fourth-generation [Rh
(cod)OH]2 catalytic system (Scheme 2).
In general, the anthrone asymmetric ARO reaction tolerates
a wide range of substrates with the fourth-generation system
(Scheme 2). Excellent yields and ee values were observed for
oxabicycles bearing electron-donating substituents or no sub-
stituents (Scheme 2, 3a, 3c, 3e, 3g). The regioselectivities
were mostly excellent throughout the substrate scope (12:1 to
>20:1 r.r.), with the only exceptions being oxabicycles bearing
Further improvements were made by increasing the an-
throne loading to 2.5 equivalents (Table 1, entry 10), which in-
creased the yield to 96%, while maintaining the excellent
Chem. Eur. J. 2015, 21, 13883 – 13887
www.chemeurj.org
13884
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
A slight increase in catalyst loadings (4 mol%) and
reaction times were needed in a modified protocol
(Method 2) to address less reactive oxabicycles which
contain electron-withdrawing groups such as those
bearing fluorine and bromine, which are known to
be less prone to undergo oxidative addition by rho-
dium(I) complexes.[10] This modification allowed us to
access ring opening products (3b, 3d, 3 f, 3i, 3k and
3l) bearing fluorine and bromine moieties, still main-
taining the excellent yields, r.r and ee values of the
protocol.
Anthrones bearing hydroxyl groups (anthralin) and
chlorines were tolerated in this system. The reactivity
of anthralin (3h–l) also serves as evidence of the
preferential reaction of the benzylic carbon in the
presence of phenolic functional groups, thus favoring
carbon over oxygen attack.[10b] Method 2 has to be
utilized when dichlorinated anthrone is used (3m–o),
as the sterically bulky chlorine groups most likely
impede the nucleophilic approach of the benzylic
carbon on the rhodium(III) intermediate after oxida-
tive addition (cf. Scheme 3). The utility of the oppo-
site enantiomer (S,R)-PPF-tBu2 gave the opposite ring
opening product ent-3i with comparable enantiose-
lectivity, allowing both enantiomers to be accessed
by this protocol.
Overall, the methodology gave ring-opening prod-
ucts with excellent enantioselectivities for the entire
reaction scope, further augmenting the generality of
enantio-induction by this protocol. One limitation of
this catalytic system is that non-benzofused oxabicy-
cles and azabicycles are not tolerated in this system.
When peri-dichlorinated anthrones were tested, no
reaction was observed even under Method 2, sug-
gesting a possible electronic effect of the anthrone
on the reactivity. The absolute and relative configura-
tion of the products were also unambiguously deter-
mined by X-ray crystallography on derivatives 3 f, 3h,
3o and ent-3i.[15]
We further extended our methodology to racemic
oxabicycle 2h to test the possibility of obtaining dif-
ferent regioisomers in a regiodivergent strategy pre-
viously pioneered by our research group.[12a] This
strategy provided regioisomers 3p and 3q in close
to their theoretical maximum yields of 50% with ex-
cellent ee values. A 1.15 gram scale (8 mmol) experi-
ment was also conducted which yielded 3h with
very good yields and excellent ee using a 1.5 mol%
catalyst loading, hence demonstrating the scalability
of our developed protocol. Further transformations
of the ARO product were also performed by reducing
the alkene by hydrogenation (98% yield, 95% ee)
and an IBX oxidation of the alcohol to the corre-
sponding ketone (70% yield, 94% ee).[16]
Scheme 2. Scope of anthrone ARO reaction: [a] Method 1: 1 (1.0 mmol), 2 (0.4 mmol),
[Rh(cod)OH]2 (3 mol%), (R,S)-PPF-tBu2 (6 mol%), MeCN (0.1m), 508C, 6 h. [b] Method 2:
1 (1.0 mmol), 2 (0.4 mmol), [Rh(cod)OH]2 (4 mol%), (R,S)-PPF-tBu2 (8 mol%), MeCN (0.1m),
508C, 16 h. [c] Same conditions as [b] at 808C [d] Combined isolated yield of both re-
gioisomers after flash column chromatography. [e] Determined by HPLC analysis on
a chiral stationary phase. [f] Determined by H NMR analysis of the crude reaction mix-
ture. [g] Gram scale (1.15 g, 8 mmol 2a) experiment: [Rh(cod)OH]2 (1.5 mol%), (R,S)-PPF-
tBu2 (3 mol%), MeCN (0.1m), 508C, 16 h.
1
dimethoxyl or dioxolane groups when unsubstituted anthrone
1a was utilized, where the r.r. decreased to approximately 9:1
(Scheme 2, 3c and 3g).
Regarding the mechansim of the anthrone ring opening re-
action, we propose that the rhodium hydroxide/Josiphos
dimer 5 first dissociates to monomeric 6 which then under-
Chem. Eur. J. 2015, 21, 13883 – 13887
www.chemeurj.org
13885
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Acknowledgements
The authors thank Solvias AG, Umicore and Johnson Matthey
for the generous donation of ligands and rhodium catalysts.
The National Sciences and Engineering Research Council
(NSERC), the University of Toronto and Alphora Inc. are grate-
fully acknowledged for financial support. C.C.J.L. thank the
Deutsche Forschungsgemeinschaft (DFG) for a postdoctoral fel-
lowship (LO 2065/1-1). F.X. thank the National Natural Science
Foundation of China (Nos. 21172148, 21202044) and China
Scholarship Council (Nos. 201306745008) for funding. B.P.
thank NSERC for graduate funding. Dr. Alan Lough (University
of Toronto) is gratefully acknowledged for single crystal X-ray
analysis.
Keywords: asymmetric
catalysis
·
dearomatization
·
oxabicycles · rhodium · ring opening reactions
[1] For examples utilizing anthrones as anti-tubulin agents, see: a) H. Prinz,
Y. Ishii, T. Hirano, T. Stoiber, J. A. C. Gomez, P. Schmidt, H. Düssmann,
A. M. Burger, J. H. M. Prehn, E. G. Günther, E. Unger, K. Umezawa, J. Med.
Chem. 2003, 46, 3382–3394; b) H. Prinz, P. Schmidt, K. J. Bçhm, S. Baas-
ner, K. Müller, E. Unger, K. Gerlach, E. G. Günther, J. Med. Chem. 2009,
52, 1284–1294; c) N. Mahindroo, J.-P. Liou, J.-Y. Chang, H.-P. Hsieh,
Expert Opin. Ther. Pat. 2006, 16, 647–691.
[2] For examples utilizing anthrones as anti-inflammatory agents, see: a) K.
Müller, D. Gürster, S. Piwek, W. Wiegrebe, J. Med. Chem. 1993, 36, 4099–
4107; b) K. Müller, Gen. Pharmacol. 1996, 27, 1325–1335; c) K. Müller, H.
Prinz, J. Med. Chem. 1997, 40, 2780–2787; d) K. Müller, Biochem. Phar-
macol. 1997, 53, 1215–1221; e) D. Peus, A. Beyerle, H. L. Rittner, M.
Pott, A. Meves, C. Weyand, M. R. Pittelkow, J. Invest. Dermatol. 2000,
114, 688–692; f) W. M. Lau, C. M. Heard, A. W. White, Pharmaceuticals
2013, 5, 232–245.
[3] For the presence of anthrone metabolites in fungi biosynthetic path-
ways, see: a) J. M. Crawford, P. M. Thomas, J. R. Sheerer, A. L. Vagstad,
N. L. Kelleher, C. A. Townsend, Science 2008, 320, 243–246; b) M. T. Niel-
sen, J. B. Nielsen, D. C. Anyaogu, D. K. Holm, K. F. Nielsen, T. O. Larsen,
U. H. Mortensen, PLoS One 2013, 8, 1–10.
[4] a) B. Freiermuth, B. Hellrung, S. Peterli, M. F. Schultz, D. Wintgens, J.
Wirz, Helv. Chim. Acta 2001, 84, 3796–3809; b) G. M. McCann, C. M.
McDonell, L. Magris, R. A. M. O’Ferrall, J. Chem. Soc. Perkin Trans. 2 2002,
784–795; Electrophiles are also known to be trapped at the anthrone
oxygen forming anthracene products, see: c) H. O. House, N. I. Ghali,
J. L. Haack, D. VanDerveer, J. Org. Chem. 1980, 45, 1807–1817; d) W. H.
Pirkle, J. M. Finn, J. Org. Chem. 1983, 48, 2779–2780.
[5] For asymmetric organocatalytic anthrone functionalization, see: a) O.
Riant, H. B. Kagan, Tetrahedron Lett. 1989, 30, 7403–7406; b) J. Shen,
T. T. Nguyen, Y.-P. Goh, W. Ye, X. Fu, J. Xu, C.-H. Tan, J. Am. Chem. Soc.
2006, 128, 13692–13693; c) Y-H. Liao, H. Zhang, Z-J. Wu, l-F. Cun, X-M.
Zhang, W-C. Yuan, Tetrahedron: Asymmetry 2009, 20, 2397–2402; d) C.
Wu, W. Li, J. Yang, X. Liang, J. Ye, Org. Biomol. Chem. 2010, 8, 3244–
3250; e) A. Zea, A-N. R. Alba, N. Bravo, A. Moyano, R. Rios, Tetrahedron
2011, 67, 2513–2529; f) H.-W. Sun, W.-H. Liao, Z.-J. Wu, H.-Y. Wang, X.-
M. Zhang, W.-C. Yuan, Tetrahedron 2011, 67, 3991–3996.
[6] For selected recent reviews in dearomatizations, see: a) C.-X. Zhuo, W.
Zhang, S.-L. You, Angew. Chem. Int. Ed. 2012, Angew. Chem. 2012, 124,
12834–12858; b) S. P. Roche, J. A. Porco, Angew. Chem. Int. Ed. 2011, 50,
4068–4123; Angew Chem. 2011, 123, 4154–4179; c) Q. Ding, X. Zhou,
R. Fan, Org. Biomol. Chem. 2014, 12, 4807–4815.
[7] For racemic transition metal catalyzed anthrone functionalizations, see:
a) E. Gómez-Bengoa, J. M. Cuerva, C. Mateo, A. M. Echavarren, J. Am.
Chem. Soc. 1996, 118, 8553–8565; b) W. Baik, C. H. Yoon, K. C. Lee, H. J.
Lee, S. Koo, J. Kim, B. Yoon, H. Kim, J. Chem. Res. Synop. 1998, 358–359;
c) W. Baik, C. H. Yoon, S. Koo, H. Kim, J. Kim, J. Kim, S. Hong, Bull. Korean
Chem. Soc. 2004, 25, 491–500.
Scheme 3. Proposed mechanism.
goes oxidative addition to rhodium(III) species 7 (Scheme 3). A
crucial proton transfer from 1 to 7 aromatizes anthrones to an-
throxide in intermediate 8. The major regioisomer 3 is ob-
tained when 8 dearomatizes and attacks the proximal carbon,
followed by ring opening. The minor regioisomer 4 formation
is postulated to originate from the dearomatative attack of the
anthroxide at the distal carbon, as intermediate 8 isomerizes to
the rhodium-allyl species 9.[17] It is also important to emphasize
that regioselection of the benzylic carbon over the oxygen as
nucleophile in anthroxide 10 is significant since ARO reactions
of phenols are known, and anthrones are also known to trap
electrophiles at oxygen to generate anthracene.[10b] Moreover,
ARO reaction still occurred selectively at the benzylic position
in the presence of phenolic functionalities in anthralin. Finally,
reductive elimination regenerates the rhodium(I) species 6.
In conclusion, we developed a new fourth-generation cata-
lytic system which features [Rh(cod)OH]2 as an air-stable cata-
lyst for the asymmetric ring opening reaction of oxabicycles by
anthrones. This protocol represents the first asymmetric exam-
ple of anthrone benzylic functionalization by a transition metal
catalyst. Moreover, our methodology presents a new class of
SN2’-type reaction for functionalizing anthrones, which previ-
ously are not known to react beyond Michael type systems. In
addressing substrate compatibility, we demonstrated that this
fourth-generation catalytic system is superior to previously de-
veloped rhodium systems for the above presented ARO reac-
tion. Further investigations of this fourth-generation system on
other nucleophilic agents are currently underway in our labora-
tories.
Chem. Eur. J. 2015, 21, 13883 – 13887
www.chemeurj.org
13886
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[8] For selected reviews in rhodium catalyzed reactions, see: a) M. Lautens,
K. Fagnou, S. Hiebert, Acc. Chem. Res. 2003, 36, 48–58; b) K. Fagnou, M.
Lautens, Chem. Rev. 2003, 103, 169–196; c) T. Hayashi, K. Yamasaki,
Chem. Rev. 2003, 103, 2829–2844; d) P. A. Evans in Modern Rhodium-
Catalyzed Organic Reactions, Wiley-VCH, Weinheim, 2005.
5421; c) A. Boyer, M. Lautens, Angew. Chem. Int. Ed. 2011, 50, 7346–
7349; Angew. Chem. 2011, 123, 7484–7487; d) L. Zhang, C. M. Le, M.
Lautens, Angew. Chem. Int. Ed. 2014, 53, 5951–5954; Angew. Chem.
2014, 126, 6061–6064.
[13] While [Rh(cod)OH]2 has been utilized by us in ring opening reactions of
diazabicycles, this protocol proceed through a different mechanism in-
volving carborhodation through a RhI isohypsic pathway, see: a) F.
Menard, M. Lautens, Angew. Chem. Int. Ed. 2008, 47, 2085–2088;
Angew. Chem. 2008, 120, 2115–2118. The utility of [Rh(cod)OH]2 has
never been developed as an established catalytic protocol in previous
asymmetric ARO reaction of oxabicycles involving the redox RhI/RhIII ox-
idative C-O insertion pathway.
[14] The structural elucidation of the distal regioisomer 4a is established by
2D NMR techniques and selective 1D NOESY irradiation experiments on
an isolated purified sample in separate 0.4 mmol scale low r.r. yielding
condition (See Supporting Information).
[15] CCDC 1402788 (3o), CCDC 1402789 (3h), CCDC 1402790 (3 f) and
CCDC 1402791 (ent-3i) 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.
[9] Other metals are also known in the ring opening reaction of oxabicy-
cles, for selected examples, see: a) M. Lautens, J-L. Renaud, S. Hiebert, J.
Am. Chem. Soc. 2000, 122, 1804–1805; b) D. K. Rayabarapu, C-F. Chiou,
C-H. Cheng, Org. Lett. 2002, 4, 1679–1682; c) F. Bertozzi, M. Pineschi, F.
Macchia, L. A. Arnold, A. J. Minnaard, B. L. Feringa, Org. Lett. 2002, 4,
2703–2705; d) K. Villeneuve, W. Tam, J. Am. Chem. Soc. 2006, 128,
3514–3515; e) K. Villeneuve, W. Tam, Eur. J. Org. Chem. 2006, 5449–
5453; f) A. Tenaglia, S. Marc, J. Org. Chem. 2006, 71, 3569–3575; g) A.
Tenaglia, S. Marc, L. Giordano, I. De Riggi, Angew. Chem. Int. Ed. 2011,
50, 9062–9065; Angew. Chem. 2011, 123, 9228–9231; h) H. Cheng, D.
Yang, J. Org. Chem. 2012, 77, 9756–9765; i) L. Meng, W. Yang, X. Pan, M.
Tao, G. Cheng, S. Wang, H. Zeng, Y. Long, D. Yang, J. Org. Chem. 2015,
80, 2503–2512; For a non-metal Lewis acid catalyzed protocol, see
j) C. S. Schindler, S. Diethelm, E. M. Carreira, Angew. Chem. Int. Ed. 2009,
48, 6296–6299; Angew. Chem. 2009, 121, 6414–6417; For an interesting
rhodium catalyzed cyclodimerization of oxabicycles, see: k) T. Nishimura,
T. Kawamoto, K. Sasaki, E. Tsurumaki, T. Hayashi, J. Am. Chem. Soc. 2007,
129, 1492–1493.
[10] For utility of the first-generation [Rh(cod)Cl]2 catalyst, see: a) M. Lautens,
K. Fagnou, T. Rovis, J. Am. Chem. Soc. 2000, 122, 5650–5651; b) M. Laut-
ens, K. Fagnou, M. Taylor, Org. Lett. 2000, 2, 1677–1679; c) J. Zhu, G. C.
Tsui, M. Lautens, Angew. Chem. Int. Ed. 2012, 51, 12353–12356; Angew.
Chem. 2012, 124, 12519–12522; d) G. C. Tsui, J. Tsoung, P. Dougan, M.
Lautens, Org. Lett. 2012, 14, 5542–5545; e) G. C. Tsui, M. Lautens,
Angew. Chem. Int. Ed. 2012, 51, 5400–5404; Angew. Chem. 2012, 124,
5496–5500.
[16] See the Supporting Information for details.
[17] In our experiments, regioisomer 4a was only observable when an-
throne substrate 1a is used. Substrates 1b and 1c (hydroxylated and
chlorinated anthrones) generally gave solely ARO product 3. Moreover,
comparative control experiments were performed using both
[Rh(cod)Cl]2 and [Rh(cod)OH]2 catalyst systems on the known ARO of
structurally analogous p-chlorophenol to further understand the origins
of the formation of regioisomer 4 via intermediate 9, yield of the ARO
product were 73% and 96% respectively. However, no regioisomers
can be detected in the phenol ring opening experiments. It is known
that formation of regioisomer 4 is dependant upon the structure of the
nucleophile. The possibility of the presence of rhodium-allyl species 9
was first suggested in our detailed mechanistic studies: a) M. Lautens,
K. Fagnou, Proc. Natl. Acad. Sci. USA 2004, 101, 5455–5460.
[11] For utility of the second-generation Rh iodide catalyst in the ARO reac-
tion, see: a) M. Lautens, K. Fagnou, J. Am. Chem. Soc. 2001, 123, 7170–
7171; b) M. Lautens, K. Fagnou, D. Yang, J. Am. Chem. Soc. 2003, 125,
14884–14892; c) M. Lautens, K. Fagnou, Tetrahedron 2001, 57, 5067–
5072; d) P. Leong, M. Lautens, J. Org. Chem. 2004, 69, 2194–2196.
[12] For utility of the third-generation Rh(cod)2OTf catalyst, see: a) R. Web-
ster, C. Bçing, M. Lautens, J. Am. Chem. Soc. 2009, 131, 444–445; b) R.
Webster, A. Boyer, M. J. Fleming, M. Lautens, Org. Lett. 2010, 12, 5418–
Received: July 10, 2015
Published online on August 27, 2015
Chem. Eur. J. 2015, 21, 13883 – 13887
www.chemeurj.org
13887
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim