Gold(i)-catalysed alcohol additions to cyclopropenes

Article abstract of DOI:10.1039/c0ob00085j

Gold(i)-catalysed addition of alcohols to 3,3-disubstituted cyclopropenes occurs in a highly regioselective and facile manner to produce alkyl tert-allylic ethers in good yields. The reaction is tolerant of sterically hindered substituents on the cyclopropene as well as primary and secondary alcohols as nucleophiles. In this full article, we report on the substrate scope and plausible mechanism, as well as the regioselectivity issues arising from subsequent gold(i)-catalysed isomerisation of tertiary to primary allylic ethers.

Full text of DOI:10.1039/c0ob00085j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Gold(I)-catalysed alcohol additions to cyclopropenes†
Maximillian S. Hadfield, Ju¨rgen T. Bauer, Pauline E. Glen and Ai-Lan Lee*
Received 5th May 2010, Accepted 11th June 2010
First published as an Advance Article on the web 9th July 2010
DOI: 10.1039/c0ob00085j
Gold(I)-catalysed addition of alcohols to 3,3-disubstituted cyclopropenes occurs in a highly
regioselective and facile manner to produce alkyl tert-allylic ethers in good yields. The reaction is
tolerant of sterically hindered substituents on the cyclopropene as well as primary and secondary
alcohols as nucleophiles. In this full article, we report on the substrate scope and plausible mechanism,
as well as the regioselectivity issues arising from subsequent gold(I)-catalysed isomerisation of tertiary
to primary allylic ethers.
Table 1 The regioselective gold(I) catalysed ring-opening addition of
cyclopropene 1 with a variety of alcohols to form tert-allylic ethers
Introduction
Ethers are ubiquitous in organic chemistry and their preparation
is one of the most fundamental reactions in organic synthesis.
Alkyl allylic ethers, for example, are found in natural products and
are versatile substrates and building blocks in organic synthesis.1
However, the most widely used method for formation of ethers, the
Williamson ether synthesis, is seldom useful for preparing tertiary
ethers as elimination reactions tend to be favored when tertiary
alkoxides or tertiary halides are used.2 Despite recent advances,
a mild, efficient and general method for the synthesis of alkyl
tert-allylic ethers still remains a challenging topic for synthetic
chemists.3,4
We recently disclosed our preliminary results on two separate
methods of forming alkyl tert-allylic ethers: regioselective gold(I)-
catalysed5 addition of alcohols to cyclopropenes6 (Scheme 1) and
regioselective gold-catalysed hydroalkoxylation of allenes.7 In this
full article, we expand on the substrate scope, possible mechanism
and regioselectivity issues of the gold-catalysed addition of
alcohols to cyclopropenes as well as studies into the gold-catalysed
isomerisation of tert-allylic ethers to primary allylic ethers.
Entrya ROH
Method Yield of 2b Product Ratio 2:3c
1
2
MeOH
EtOH
B
A
B
B
A
B
B
A
B
B
B
B
86%
64%
83%
88%
78%
88%
77%
N/Ad
70%
traces
34%
—
2a
2b
2b
2c
2d
2e
2f
2g
2g
2h
2i
>99 : 1
>99 : 1
>99 : 1
>99 : 1
>99 : 1
>99 : 1
>99 : 1
2.5 : 1
97 : 3
N/A
>99 : 1
N/A
3
4
5
EtOH
Allyl alcohol
Benzyl alcohol
=
6
7
8
9
10
11e
12
HOCH2CH2CH CH2
HOCH2CH2Ph
i-PrOH
i-PrOH
t-BuOH
H2O
4-methoxyphenol
a All reactions were carried out at 20 ◦C in CH2Cl2 for 1–2 h unless
otherwise stated. b Isolated yield, unless otherwise stated. c Determined by
1H-NMR analysis of the crude mixture. d Reaction was allowed to stir for
4 days, after which ~50% conversion was observed to 2g and 3g (~30%),
4 and 5 (~20%). e 15 eq. of t-BuOH was added as a co-solvent and the
reaction was allowed to stir for 24 h.
propene 1 in a highly regioselective manner to produce alkyl tert-
allylic ethers in good yields (Table 1).5,10,11
Primary alcohols add in excellent regioselectivities (>99 : 1
2:3, Entries 1–7, Table 1) and secondary alcohols in very good
selectivity (97 : 3 2:3, Entry 9). It is of interest to note that
the employment of air stable PPh3AuNTf2 catalyst12 (Entry 9)
produces superior activity, regioselectivity and yield in this case
compared to PPh3AuOTf (formed in situ with PPh3AuCl and
AgOTf, Entry 8). Tertiary alcohols, however, do not react under
these conditions (<5% conversion, Entry 10). The steric bulk of
the tertiary alcohol is very likely to blame for the diminished
activity. Evenwater cansuccessfully actas a nucleophiletoproduce
the corresponding tertiary alcohol 2i, albeit in low conversion
and yield (34%, Entry 11).13 An attempt to replace the alcohol
nucleophile with phenol, however, was not successful (Entry 12),
presumably due to the reduced nucleophilicity of phenol.
Scheme 1 Preliminary results on the regioselective gold(I) catalysed
ring-opening addition of cyclopropene 1 to form tert-allylic ethers 2.
Results and discussion
We recently initiated a programme to investigate gold(I) catalysed
reactions of cyclopropenes8 in the presence of nucleophiles.9
During our initial studies, we found that Au(I) can catalyse the
intermolecular addition of alcohols to 3,3-disubstituted cyclo-
School of Engineering and Physical Sciences, Chemistry–William
H. Perkin Building, Heriot-Watt University, Edinburgh, UK EH14 4AS.
E-mail: A.Lee@hw.ac.uk; Tel: +44(0)131-4518030
† Electronic supplementary information (ESI) available: Experimental
procedures, characterisation of all new compounds and copies of 1H-NMR
and 13C-NMR spectra. See DOI: 10.1039/c0ob00085j
Upon repeating the reaction shown in Entry 3 with a reduced
Au(I) catalyst loading (1 mol%), complete conversion to 2b
(>99% regioselectivity) is still observed within 1.5 h. Indeed,
4090 | Org. Biomol. Chem., 2010, 8, 4090–4095
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
reactions with primary alcohol nucleophiles such as EtOH (Entry
3) are facile and often complete in <10 min with 5 mol%
catalyst.
AgOTf results in excellent selectivity for 2b and moderately good
isolated yield (64% 2b, Entry 1).17 Replacing the OTf- counterion
with SbF6 does not affect the selectivity, but the isolated yield is
-
Having investigated the scope of the alcohol nucleophile, we
then set out to study the scope of the cyclopropene substrates
employed (Table 2).14,15 A range of cyclopropenes reacts smoothly
with alcohols to produce tert-allylic ethers in good regioselectivity
and yields. Changing a substituent from alkyl (e.g. Entries 1–2) to
benzyl appears to have no detrimental effect on the regioselectivity
or yield (Entries 3–4). To our delight, even more sterically
encumbered substituents (i-Pr and Bn in 9) are tolerated with
a primary alcohol nucleophile to deliver product 10 in excellent
regioselectivity and good yield (>99 : 1, 73%, Entry 5). Remark-
ably, even the more sterically hindered cyclopropene 11 provides
good regioselectivity (87 : 1) with a primary alcohol nucleophile
at room temperature (Entry 6). Excellent regioselectivity (>99 : 1)
can be achieved with cyclopropene 11 by cooling the reaction
mixture down to 0 ◦C to provide 12 in 64% yield. When combining
the sterically hindered cyclopropene 9 with a hindered secondary
alcohol, however, the regioselectivity and yield begins to drop
(92 : 8, 45%, Entry 7).
Finally, even an aryl substituent is tolerated (Entries 8 and 9).
Since there is literature precedence for intramolecular rearrange-
ments of aryl substituted cyclopropenes to form indenes in the
presence of gold(I),6,9 it is pleasing that intermolecular alcohol
additions are viable even with substrate 14. The regioselectivity
with 14, however, is rather surprising. Reaction with nBuOH under
standard conditions produces a non-regioselective 1 : 1 mixture◦of
primary and tertiary ethers. Reducing the temperature to 10 C
and increasing the alcohol to 15 equivalents successfully yields
the tertiary ether 15 regioselectively with nBuOH (65%, Entry
8). Surprisingly, changing the alcohol nucleophile to phenethyl
alcohol completely switches the regioselectivity to the primary
ether 16 (Entry 9, 16 formed regio- and stereoselectively, 65%)!
These results suggest that with aryl substituted cyclopropene 14,
nBuOH might be more effective in inhibiting the isomerisation of
the tert-allylic ether product 15 to the primary allylic ether (vide
infra).16
slightly lower (55% 2b, Entry 2). Changing from PPh3 to an N-
heterocyclic carbene (NHC) ligand on Au(I) [(IPr)AuCl/AgOTf]
also provides 2b exclusively (69%, Entry 3). In all of the above
cases, a hygroscopic silver salt is required as co-catalyst to generate
the active catalyst in situ, resulting in the possibility of there being
slight traces of TfOH, HSbF6 or [LAu–OH2]+ present during the
reaction. Reaction with the air stable PPh3AuNTf2, which does
not require any hygroscopic co-catalyst, produces an even better
yield of the product 2b (83%, entry 4).
In an effort to ascertain whether the reaction is truly gold-
catalysed, we carried out some control reactions (Table 3). The
control reaction employing 5 mol% of TfOH as catalyst results
in no reaction, suggesting that traces of acid are not catalytically
active (Entry 5). Reaction with AgOTf as catalyst is also greatly
inferior to gold(I), resulting in incomplete consumption of the
starting material, along with a complex mixture of products (Entry
6). Next, we wanted to ascertain if Rh(OAc)2, which is believed
to ring-open related cyclopropenes to form the corresponding
rhodium carbene intermediates, could similarly catalyse this
reaction.18 In stark contrast to Au(I), employment of Rh(OAc)2
as catalyst, produces a mixture of 25, along with traces of both 2b
and 3b (Entry 7). Interestingly, the use of Au(III) instead of Au(I)
catalyst also completely changes the outcome of the reaction, with
the aldehyde 25 being the major product (Entry 8). This difference
in reactivity further exemplifies the differences between Au(I) and
Au(III) catalysts.19,20
Our proposed mechanism for the regioselective gold(I) catalyzed
ring-opening addition of cyclopropene 1 with alcohols is shown
in Scheme 2. Activation of the strained cyclopropene double
bond by gold(I) results in ring-opening to produce the proposed
intermediate I, which can be represented as mesomeric structures
Ia, Ib or Ic.21,22 Attack of the alcohol at the C-3 position followed by
protodemetallation thus furnishes the tert-allylic ether 2. In order
to probe the validity of our proposed mechanism, the reaction was
carried out with CD3OD as the nucleophilic alcohol (Scheme 3).
Deuterium is indeed incorporated at the C-1 position (90%),
lending support to our proposed mechanism.
Since tertiary alcohols do not react under these conditions, we
postulated that the unprotected diol 17 would react chemose-
lectively at the primary alcohol. Indeed, the reaction proceeds
smoothly and chemoselectively to furnish 18 in 58% yield (Entry
10). Neopentyl glycol 19, however, behaves differently and forms a
1 : 1 mixture of primary and tertiary ether products under standard
conditions (room temperature, Entry 11). We postulate that the
proximity of a pendant alcohol promotes the gold(I)-catalysed
isomerisation of the tertiary to primary ether (vide infra). Cooling
the reaction mixture increases the regioselectivity (>99 : 1) but the
yield of 20 remains modest (33%) as oligomeric by-products are
also formed.
When optically pure (R)-PhMeCHCH2OH 21 is employed
as the nucleophile, the reaction is still regioselective, but not
diastereoselective (Entry 12). Employment of a secondary chiral
alcohol (R)-PhMeCHOH 23 with cyclopropene 6 also results in
good regioselectivity (96 : 4), but poor diastereoselectivity (d.r.
~1 : 1, Entry 13).
A small catalyst screen shows gold(I) catalysts to be unique in
their selectivity for the tertiary allylic ether product 2 (Table 3). The
gold(I) catalyst PPh3AuOTf, formed in situ from PPh3AuCl and
Scheme 2 Proposed mechanism for the regioselective gold(I)-catalysed
ring-opening addition of 1 with alcohols.
It is of interest to note that an excess of alcohol is necessary
to ensure good regioselectivity (Scheme 4). When the alcohol
nucleophile is reduced from excess to 1 equivalent, the regiose-
lectivity of 2b:3b drops from >99 : 1 to 2 : 1. However, when the
reaction is carried out with 1 equiv. of EtOH and 5 equiv. of
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4090–4095 | 4091
©

View Article Online
Table 2 The regioselective gold(I) catalysed ring-opening addition of alcohols to a variety of cyclopropenesa
Entry
1
Cyclopropene
ROH
Product
Yield (%)b
83%
Regioselectivityc
>99 : 1
2
87%
>99 : 1
3
4
80%
86%
>99 : 1
>99 : 1
5
6
7
73%
64%d
45%
>99 : 1
>99 : 1 (0 ◦C)
87 : 1 (20 ◦C)
92 : 8
8e
65%
>99 : 1
9e
65%
58%
>1 : 99
>99 : 1
10f
11
16%, 16%
33%
1 : 1 (20 ◦C)
>99 : 1 (10 ◦C)
12
13
65%
46%
>99 : 1
dr~1 : 1
96 : 4
dr~1 : 1
a All reactions carried out with PPh3AuNTf2 (5 mol%) and 6 equiv. ROH at 20 ◦C in CH2Cl2 for 1–2 h unless otherwise stated. b Isolated yield, unless
otherwise stated. c Determined by 1H-NMR analysis of the crude mixture. Regioselectivity of the tertiary:primary allylic ether product. d Reaction was
carried out at 0 ◦C for 16 h. e 15 Equiv. ROH, 10 ◦C, 4 h. f Reaction carried out with 2 equiv. diol for 16 h.
4092 | Org. Biomol. Chem., 2010, 8, 4090–4095
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 3 Transition metal-catalysed reaction of 1 with EtOH
Entry
Catalyst
Time/h
Result
1
2
3
4
5
6
7
8
PPh3AuCl/AgOTf
PPh3AuCl/AgSbF6
(IPr)AuCl/AgOTf
PPh3AuNTf2
HOTf
AgOTf
Rh(OAc)2
1.5a
1.5
1.5
1.5
24
24
24
24
2b bonly, 64% yield
2b bonly, 55% yield
2b bonly, 69% yield
2b bonly, 83% yield
No reactionc
7 : 4 ratio of 1:2b along with traces of 3b, 25 and other unidentified by-productsc
Major product: 25 (with traces of 2b and 3b and other unidentified by-products)c
Complex mixture of products: 25 (with traces of 2b and 3b and other unidentified
by-products)c
AuCl3
a No change in yield or selectivity is observed if the reaction is allowed to stir for 24 h. b Isolated yield; 3b and 25 were not detected by 1H-NMR analysis
of the crude mixture. c By 1H-NMR analysis of the crude mixture. (IPr = NHC ligand bis-2,6-diisopropylphenyl imidazolylidinene) .
Scheme 3 The regioselective gold(I) catalyzed ring-opening addition of 1
with CD3OD.
t-BuOH, as an additive, the regioselectivity is retained (>99% 2b,
64% yield). Thus the alcohol nucleophile need not be in excess,
as long as a non-reactive alcohol such as t-BuOH is present in
excess to help maintain good regioselectivities. This approach of
having a cheap, non-reacting alcohol additive may be useful if the
nucleophilic alcohol employed is expensive.23
Scheme 5 Gold(I)-catalysed isomersation of tert-allylic ethers to primary
allylic ethers26
in a mixture of 2b, 3b and other unidentified by-products after
1 h at room temperature. Possible explanations for the inability of
(IPr)AuOTf to isomerise tert-allylic ethers are the steric bulk of
the IPr ligand,24 or the less electrophilic gold(I) centre caused by
the more s-donating NHC catalyst.25
With these results in hand, we propose that the gold(I)- catalysed
addition of alcohols to 3,3-disubstituted cyclopropenes 26 occurs
regioselectively to produce the kinetic tert-allylic ether product
27. In the absence of excess alcohol, this kinetic product can be
isomerised by gold(I) to the more stable primary allylic ether 28
(Scheme 6).27 In the presence of excess alcohol, we postulate that
Scheme 4 Dependence of regioselectivity on excess alcohol
the gold(I) catalyst PPh3AuNTf2 is deactivated such that it is no
Next, we sought to explain the need for excess alcohol to ensure
longer able to isomerise 27→28,28 thus excellent regioselectivity
good regioselectivity. As shown in Scheme 5, the tert-allylic ether
for the tertiary ether 27 is observed.
product 2a was isolated and resubjected to the reaction conditions
in the absence and presence of excess methanol. The tertiary
ether 2a isomerises to the primary ether 3a under gold(I)-catalysis,
but addition of excess alcohol appears to stop this isomerisation
(Scheme 5). The isomerisation 2a→3a is not reversible (both in
the absence and presence of excess MeOH). The isomerisation
appears catalyst dependent: subjection of 2b to NHC-gold catalyst
(IPr)AuCl/AgOTf (5 mol%) in CH2Cl2 at room temperature
results in no reaction whereas PPh3AuCl/AgOTf (5 mol%) results
Scheme 6 Effect of excess alcohol on the gold(I)-catalysed addition of
alcohols to cyclopropenes
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4090–4095 | 4093
©

View Article Online
Since the NHC catalyst (IPr)AuOTf does not isomerise tertiary
ether 2b to primary ether 3b even in the absence of excess alcohol
(vide supra) excess alcohol should in principle not be necessary
in order to achieve good selectivities with this catalyst system.25
The alcohol addition to cyclopropene 1 was thus repeated with
(IPr)AuCl/AgOTf (5 mol%) with only 1 equiv. of alcohol and
indeed, only the tertiary ether 2b is observed (Scheme 7). The
isolated yield, however, is not as high as with PPh3AuNTf2 as
catalyst under our standard conditions (51% vs. 83%).
Acknowledgements
We would like to thank ScotChem (MSH), Erasmus (JTB),
EPSRC (PEG; EP/G006695/1), The Royal Society, and Heriot-
Watt University for funding and the EPSRC Mass Spectrometry
services at Swansea for analytical support. We thank Paul Young
for initial experiments with cyclopropene 4.
References
1 For selected examples of biologically relevant molecules and natural
products with alkyl tert-allylic ether moieties, see: (a) A. Evidente,
A. Andolfi, M. Fiore, A. Boari and M. Vurro, Phytochemistry, 2006,
67, 19; (b) F. Matsumoto, H. Idetsuki, K. Harada, I. Nohara and Y.
Toyoda, J. Essent. Oil Res., 1993, 5, 123; (c) F. Bohlmann, P. Singh and
J. Jakupovic, Phytochemistry, 1982, 21, 157; (d) D. Olagnier, P. Costes,
Philippe, A. Berry, M.-D. Linas, M. Urrutigoity, O. Dechy-Cabaret
and F. Benoit-Vical, Bioorg. Med. Chem. Lett., 2007, 17, 6075; (e) A. L.
Giannouli and S. E. Kintzios, Medicinal and Aromatic Plants–Industrial
Profiles, 2000, 14, 69.
Scheme 7 Excess alcohol is not required for good regioselectivity with
the NHC–Au catalyst (IPr)AuOTf, as the catalyst does not isomerise 2b to
3b.
2 (a) S. Assabumrungrat, W. Kiatkittipong, N. Sevitoon, P. Praserthdam
and S. Goto, Int. J. Chem. Kinet., 2002, 34, 292; (b) B. Shi and B. H.
Davis, J. Catal., 1995, 157, 359.
3 For example, see: (a) P. A. Evans and D. K. Leahy, J. Am. Chem. Soc.,
2002, 124, 7882; (b) B. M. Trost, E. J. McEachern and F. D. Toste,
J. Am. Chem. Soc., 1998, 120, 12702; (c) T. Shintou and T. Mukaiyama,
J. Am. Chem. Soc., 2004, 126, 7359.
4 For examples of recent advances in ether synthesis (but not specifically
tert-allylic ethers), see: (a) T. A. Mitchell and J. W. Bode, J. Am. Chem.
Soc., 2009, 131, 18057; (b) T. Shintou and T. Mukaiyama, J. Am. Chem.
Soc., 2004, 126, 7359; (c) A. Corma and M. Renz, Angew. Chem., Int.
Ed., 2007, 46, 298 and references cited therein.
5 For selected recent reviews on gold-catalysis, see: (a) Z. Li, C. Brouwer
and C. He, Chem. Rev., 2008, 108, 3239; (b) N. Bongers and N. Krause,
Angew. Chem., Int. Ed., 2008, 47, 2178; (c) D. J. Gorin and F. D. Toste,
Nature, 2007, 446, 395; (d) A. Fu¨rstner and P. W. Davies, Angew. Chem.,
Int. Ed., 2007, 46, 3410; (e) E. Jime´nez-Nu´nez and A. M. Echavarren,
Chem. Commun., 2007, 333; (f) A. S. K. Hashmi, Chem. Rev., 2007,
107, 3180; (g) H. C. Shen, Tetrahedron, 2008, 64, 3885; (h) H. C. Shen,
Tetrahedron, 2008, 64, 7847; (i) N. Marion and S. P. Nolan, Chem. Soc.
Rev., 2008, 37, 1776.
These observations have recently helped us to switch the re-
gioselectivity of the gold(I)-catalysed hydroalkoxylation of allenes
to form tert-allylic ethers,7 compared to the previously reported
primary allylic ethers (Scheme 8).29 The major benefit of utilising
the hydroalkoxylation method shown in Scheme 8 is that 1,1-
disubstituted allenes are more readily accessible [commercial
=
=
(R1 R2 Me), or two steps from commercially available material
vs. three steps for 3,3-disubstituted cyclopropenes].30 On the other
hand, the hydroalkoxylation method is much more sensitive to
steric hindrance than the gold(I)-catalysed addition of alcohols
to cyclopropenes described in this paper. For example, secondary
alcohols provide poor regioselectivity (27:28 3 : 1) in the allene
hydroalkoxylation reaction but are still excellent nucleophiles in
the cyclopropene addition reaction (27:28 97 : 3, Entry 9, Table 1).
We believe the two methods are thus complementary approaches
towards alkyl tert-allylic ethers. Future work will focus on the
extension of these reactions to enantioselective methods.
6 J. T. Bauer, M. S. Hadfield and A.-L. Lee, Chem. Commun., 2008, 6405.
7 M. S. Hadfield and A.-L. Lee, Org. Lett., 2010, 12, 484.
8 For recent reviews on cyclopropenes, see: (a) M. Rubin, M. Rubina and
V. G evo rg ya n , Synthesis, 2006, 1221; (b) M. Rubin, M. Rubina and V.
Gevorgyan, Chem. Rev., 2007, 107, 3117; (c) I. Marek, S. Simaan and
A. Masarwa, Angew. Chem., Int. Ed., 2007, 46, 7364.
9 For gold(I)-catalysed intramolecular rearrangements of cyclopropenes,
see: (a) Z.-B. Zhu and M. Shi, Chem.–Eur. J., 2008, 14, 10219; (b) C.
Li, Y. Zeng and J. Wang, Tetrahedron Lett., 2009, 50, 2956.
10 For examples of ring-opening reactions of cyclopropenes, see: (a) R. E.
Giudici and A. H. Hoveyda, J. Am. Chem. Soc., 2007, 129, 3824;
(b) M. A. Smith and H. G. Richey, Organometallics, 2007, 26, 609;
(c) I. Nakamura, C. B. Bajracharya and Y. Yamamoto, J. Org. Chem.,
2003, 68, 2297; (d) P. Binger and B. Biedenbach, Chem. Ber., 1987, 120,
601; (e) T. Shibata, S. Maekawa and K. Tamura, Heterocycles, 2008, 76,
1261; (f) Y. Wang and H. W. Lam, J. Org. Chem., 2009, 74, 1353; (g) Y.
Wang, E. A. F. Fordyce, F. Y. Chen and H. W. Lam, Angew. Chem., Int.
Ed., 2009, 48, 7350.
Scheme 8 Gold(I)-catalysed hydroalkoxylations of allenes7
Conclusions
11 For a review on gold-catalyzed reaction of alcohols, see: Muzart,
Tetrahedron, 2008, 64, 5815.
12 N. Me´zailles, L. Richard and F. Gagosz, Org. Lett., 2005, 7, 4133.
13 Formation of tert-allylic alcohol from 3,3-dimethylcyclopropene-1,2-
dicarboxylate and water is known with Pd(0), albeit in low selectivity.
See: A. S. K. Hashmi, M. A. Grundl and J. W. Bats, Organometallics,
2000, 19, 4217.
14 The cyclopropenes were synthesised following a general literature
procedure: M. Rubin and V. Gevorgyan, Synthesis, 2004, 5, 796.
15 We have previously shown that if one of the substituents on 3,3-
disubstituted cyclopropenes is an ester, gold(I)-catalysed intramolec-
ular rearrangement to furanone occurs. When the reaction is carried
out in the presence of EtOH, at best a ~1 : 1 ratio of intramolecular
rearrangement vs. intermolecular alcohol addition results. See ref. 6.
Gold(I)-catalysed addition of alcohols to 3,3-disubstituted cyclo-
propenes occurs in a highly regioselective manner to produce alkyl
tert-allylic ethers in good yields. The reaction is facile (as quick
as <10 min), mild (20 ◦C), efficient (as low as 1 mol% catalyst
loading can be used to no detrimental effect), and inert atmosphere
and distilled solvents are not required. The reaction is tolerant of
sterically hindered substituents on the cyclopropene as well as
primary and secondary alcohols as nucleophiles. Excess alcohol
is crucial for achieving high regioselectivities as it retards any
subsequent isomerisation of the tertiary allylic ether products to
primary allylic ethers.
4094 | Org. Biomol. Chem., 2010, 8, 4090–4095
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
16 One possible explanation kindly suggested by a referee is that the
O-bound Au(I) complex with phenethyl alcohol (see ref. 28) is
in equilibrium with the Au(I)-arene complex, which might still be
catalytically active for isomerisation of 29 to 16. Examples of stable
Au(I)-arene complexes are known: E. Jime´nez-Nu´n˜ez, C. K. Claverie,
C. Nieto-Oberhuber and A. M. Echavarren, Angew. Chem., Int. Ed.,
2006, 45, 5452. It should be noted that phenethyl alcohol is not a
problematic reagent for 3,3-dialkyl-substituted cyclopropene 1 (Entry
7, Table 1), suggesting that the presence of aryl substituents facilitates
allylic isomerisation.
D. J. Gorin, S. T. Staben and F. D. Toste, J. Am. Chem. Soc., 2005,
127, 18002; (b) C. A. Witham, P. Mauleo´n, N. D. Shapiro, B. D.
Sherry and F. D. Toste, J. Am. Chem. Soc., 2007, 129, 5838; (c) P. W.
Davies and S. J.-C. Albrecht, Chem. Commun., 2008, 238; (d) N. D.
Shapiro and F. D. Toste, J. Am. Chem. Soc., 2007, 129, 4160; (e) G.
Li and L. Zhang, Angew. Chem., Int. Ed., 2007, 46, 5156; (f) S. Lo´pez,
E. Herrero-Go´mez, P. Pe´rez-Gala´n, C. Nieto-Oberhuber and A. M.
Echavarren, Angew. Chem., Int. Ed., 2006, 45, 6029; (g) C. H. M.
Amjis, V. Lo´pez-Carrillo and A. M. Echavarren, Org. Lett., 2007, 9,
4021; (h) G. Lemie`re, V. Gandonm, K. Cariou, A. Hours, T. Fukuyama,
A.-L. Dhimane, L. Festerbank and M. Malacria, J. Am. Chem. Soc.,
2009, 131, 2993; (i) P. W. Davies and S. J.-C. Albrecht, Angew. Chem.,
Int. Ed., 2009, 48, 8372; (j) O. N. Faza, C. S. Lo´pez, R. Alvarez and A. R.
de Lera, J. Am. Chem. Soc., 2006, 128, 2434; (k) G. Li, G. Zhang and L.
Zhang, J. Am. Chem. Soc., 2008, 130, 3740; (l) A. Correa, N. Marion,
L. Festerbank, M. Malacria, S. P. Nolan and L. Cavallo, Angew. Chem.,
Int. Ed., 2008, 47, 718.
23 It should also be noted that the regioselectivity is sensitive to
temperature. Carrying out the reaction shown in Scheme 4 at ~25 ◦C
rather than 20 ◦C resulted in a 96 : 4 ratio of 2b:3b (cf. >99 : 1 at 20 ◦C).
When the ambient temperature is >20 ◦C, the reaction should thus be
cooled to 15–20 ◦C to maintain good regioselectivities.
17 (a) D. C. Rosenfeld, S. Shekhar, A. Takemiya, M. Utsonomiya and J. F.
Hartwig, Org. Lett., 2006, 8, 4179; (b) Z. Li, J. Zhang, C. Brouwer,
C.-G. Yang, N. W. Reich and C. He, Org. Lett., 2006, 8, 4175.
18 For example, see: A. Padwa, J. M. Kassir and S. L. Xu, J. Org. Chem.,
1991, 56, 6971.
24 H. Clavier and S. P. Nolan, Chem. Commun., 2010, 46, 841.
25 For a review on ligand effects in homogenous gold catalysis, see: D. J.
Gorin, B. D. Sherry and F. D. Toste, Chem. Rev., 2008, 108, 3351.
26 For a proposed mechanism of Au(I)-catalysed allylic isomersation
based on DFT calculations, see: R. S. Paton and F. Maseras, Org.
Lett., 2009, 11, 2237.
27 A related conclusion has been proposed (via DFT studies) for gold(I)-
catalysed hydroalkoxylation of allenes to form primary allylic ethers,
see ref. 26.
19 (a) For other examples, see: A. S. K. Hashmi, R. Salathe´ and W. Frey,
Chem.–Eur. J., 2006, 12, 6991; (b) G. Lemie´re, V. Gandom, N. Agenet,
J.-P. Goddard, A. de Kozak, C. Aubert, L. Fensterbank and M.
Malacria, Angew. Chem., Int. Ed., 2006, 45, 7596; (c) A. W. Sromek,
M. Rubina and V. Gevorgyan, J. Am. Chem. Soc., 2005, 127, 10500;
(d) M. R. Fructos, P. de Fre´mont, S. P. Nolan, M. M. Diaz-Requejo
and P. J. Pe´rez, Organometallics, 2006, 25, 2237.
20 For oxidative potential of gold(III), see: A. S. K. Hashmi, M. C. Blanco,
D. Fischer and J. W. Bats, Eur. J. Org. Chem., 2006, 1387.
28 Deactivation of the catalyst through formation of [LAu-O(H)R]+
complexes in the presence of excess alcohol is a possible explanation.
29 (a) Z. Zhang and R. Widenhoefer, Org. Lett., 2008, 10, 2079; (b) N.
Nishina and Y. Yamamoto, Tetrahedron, 2009, 65, 1799; (c) N. Nishina
and Y. Yamamoto, Tetrahedron Lett., 2008, 49, 4908; (d) D.-M. Cui,
K. R. Yu and C. Zhang, Synlett, 2009, 1103; (e) See also: D.-M. Cui,
Z.-L. Zheng and C. Zhang, J. Org. Chem., 2009, 74, 1426.
30 For example, see: (a) J. Takaya and N. Iwasawa, J. Am. Chem. Soc.,
2008, 130, 15254; (b) M. Y. Ngai, E. Skucas and M. Krische, Org. Lett.,
2008, 10, 2705. For cyclopropene synthesis, see ref. 14.
21 For discussion and debate into the nature of gold carbene vs. cationic
intermediates, see: (a) D. Benitez, N. D. Shapiro, E. Tkatchouk, Y.
Wang, W. A. Goddard, III and F. D. Toste, Nat. Chem., 2009, 1, 482;
(b) G. Seidel, R. Mynott and A. Fu¨rstner, Angew. Chem., Int. Ed., 2009,
48, 2510; (c) A. S. K. Hashmi, Angew. Chem., Int. Ed., 2008, 47, 6754;
(d) A. Fedorov, M. E. Moret and P. Chen, J. Am. Chem. Soc., 2008,
130, 8880; (e) A. Fu¨rstner and L. Morency, Angew. Chem., Int. Ed.,
2008, 47, 5030 and references cited therein.
22 For representative examples of other methods of forming gold species
of type I and their corresponding reactivity, see: (a) M. J. Johansson,
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4090–4095 | 4095
©