Improved synthesis of cyclic tertiary allylic alcohols by asymmetric 1,2-addition of AlMe3 to enones

Article abstract of DOI:10.1002/chem.201303061

The development of an improved protocol for the enantioselective Rh ^I/binap-catalysed 1,2-addition of AlMe₃ to cyclic enones is reported. ³¹P NMR analysis of the reaction revealed that the catalyst in its resting state is a chloride-bridged dimer. This insight led to the use of AgBF₄ as an additive for in situ activation of the dimeric precatalyst. Thus, the catalyst loading can now be reduced to only 1 mol % with respect to rhodium. Various 5-7-membered cyclic enones can be transformed into tertiary allylic alcohols with excellent levels of enantioselectivity and high yields. The obtained products are versatile synthetic building blocks, shown by a highly enantioselective formal total synthesis of the pheromone (-)-frontalin as well as formation of a bicyclic lactone that has the core structure of the natural flavour component "wine lactone". Copyright

Full text of DOI:10.1002/chem.201303061

DOI: 10.1002/chem.201303061
Improved Synthesis of Cyclic Tertiary Allylic Alcohols
by Asymmetric 1,2-Addition of AlMe3 to Enones
Andreas Kolb, Wei Zuo, Jꢀrgen Siewert, Klaus Harms, and Paultheo von Zezschwitz*[a]
Abstract: The development of an im-
proved protocol for the enantioselec-
tive RhI/binap-catalysed 1,2-addition of
AlMe3 to cyclic enones is reported.
31P NMR analysis of the reaction re-
vealed that the catalyst in its resting
state is a chloride-bridged dimer. This
insight led to the use of AgBF4 as an
additive for in situ activation of the di-
meric precatalyst. Thus, the catalyst
loading can now be reduced to only
1 mol% with respect to rhodium. Vari-
ous 5–7-membered cyclic enones can
be transformed into tertiary allylic al-
cohols with excellent levels of enantio-
selectivity and high yields. The ob-
tained products are versatile synthetic
building blocks, shown by a highly
enantioselective formal total synthesis
of the pheromone (ꢀ)-frontalin as well
as formation of a bicyclic lactone that
has the core structure of the natural
flavour component “wine lactone”.
Keywords: alanes · allylic alcohols ·
asymmetric catalysis · enones · nu-
cleophilic addition · rhodium
Introduction
edge, only two protocols exist for the asymmetric prepara-
tion of cyclic tertiary allylic alcohols from cycloalk-2-
enones.[4] The research groups of Walsh and Yus have inde-
a,b-Unsaturated carbonyl compounds are highly versatile
ꢀ
substrates for C C bond formation in organic synthesis. The
pendently developed a TiACTHNGUTERN(UNG OiPr)4-mediated addition of orga-
asymmetric conjugate addition (1,4-addition) of nucleophilic
species to such compounds is one of the most intensively
studied and well-established synthetic methods in organic
chemistry.[1] In contrast, asymmetric 1,2-addition to a,b-un-
saturated carbonyl compounds has been reported to a lesser
extent, even though the products, chiral allylic alcohols, are
very important. They occur as structural motifs in numerous
target molecules and can take part in a broad range of reac-
tions, in which additional stereocentres can be selectively as-
sembled owing to effective stereoinduction by the hydroxy
group.[2]
nozinc reagents to ketones using a chiral bis(sulfonamide)
ligand (hocsac).[5] However, when using cycloalkenones as
substrates, a substituent in the 2-position is usually required
to achieve high levels of selectivity.[5f,g] We have previously
reported a RhI/binap-catalysed 1,2-addition furnishing excel-
lent levels of enantioselectity and high yields of product
even in the case of unsubstituted cycloalkenones
(Scheme 1).[6] The utilisation of organoaluminium reagents
Among the protocols for asymmetric 1,2-addition, only
a handful allow for the transformation of 2-enones.[3] In gen-
eral, the transformation of ketones, in comparison with that
of aldehydes, requires a more reactive nucleophile and more
sophisticated stereodiscrimination by the catalyst. Moreover,
the resulting tertiary allylic alcohols are less accessible than
secondary ones, which can also be obtained through meth-
ods such as stereoselective reduction of C=O bonds.[2] Fo-
cussing on cyclic structures, which account for a large por-
tion of target molecules, the number of available synthetic
methods becomes even smaller. To the best of our knowl-
Scheme 1. RhI/binap-catalysed addition of AlMe3 (n.d.=not deter-
mined).[6]
as nucleophiles is an attractive feature of this method.[7]
Both inexpensive AlMe3 and air-stable adduct DABCO-
[8]
ACHUTNGREN(NUG AlMe3)2 (DABCO=1,4-diazabicycloACHUTNGTRENN[NGU 2.2.2]octane) can be
used for the addition of a methyl group; aryl groups can be
introduced from mixed alanes ArAlMe2. The products of
these addition reactions are either biologically active, for ex-
ample, as pheromones,[9] or they serve as chiral building
blocks (see below). We recently expanded this methodology
to include classic kinetic resolution as well as regiodivergent
reactions on racemic substituted cyclohex-2-enones.[10]
[a] Dipl.-Chem. A. Kolb, W. Zuo, Dr. J. Siewert, Dr. K. Harms,
Prof. Dr. P. von Zezschwitz
Fachbereich Chemie
Philipps-Universitꢀt Marburg
Hans-Meerwein-Straße, 35043 Marburg (Germany)
Fax : (+49)6421-2825362
E-mail: zezschwitz@chemie.uni-marburg.de
A drawback of the methodology described above is the
need for a rather high catalyst loading of 5 mol% rhodium,
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201303061.
16366
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16366 – 16373

FULL PAPER
that is, 2.5 mol% of dimeric precatalysts, such as [{Rh-
ACHTUNGTRENNUNG(cod)OMe}2]. Lowering the amount of rhodium to 1 mol%
For further experiments, 4,4-dimethylcyclohex-2-enone
(1b) was chosen as the substrate because its transformation
occurs almost as selectively as the transformation of 1a, but
the reaction is slower (Table 1).[6] In the absence of silver
salts, the reaction proceeded smoothly using 5 mol% rhodi-
led to a significantly lower yield; almost no product was
formed when the amount of rhodium was reduced to
0.1 mol% (Scheme 1). The need for high catalyst loadings
becomes particularly problematic in the case of multigram
preparations,[11] even though the price of rhodium has drop-
ped over the last 10 years and is now significantly lower
than that of gold or platinum and approximately just 30%
higher than that of palladium.[12] Herein, we report an im-
proved catalytic system that affords excellent results for the
synthesis of cyclic tertiary allylic alcohols using only
1 mol% rhodium. Moreover, applications of the resulting
products in target structure syntheses are presented. These
results broaden the scope of the described method in both
an economic and a synthetic sense.
Table 1. Optimisation of the catalyst system.
Entry
Rhodium
[mol%][a]
Additive
([mol%])
Conv.
[%][b]
Yield
[%][b]
ee
G
N
[%][b]
1
2
3
4
5
6
7
8
9
5
1
1
0.5
0.5
0.5
0[d]
1[e]
1[f]
–
–
91
91
97
73ꢁ2[c]
97ꢁ1
58ꢁ2
73ꢁ16
64
64ꢁ2[c]
88ꢁ2
47ꢁ8
47ꢁ13
23
97ꢁ1
97ꢁ1
89ꢁ1
93ꢁ2
5
AgBF4 (1.5)
–
AgBF4 (0.75)
AgBF4 (2.5)
AgBF4 (1.5)
–
Results and Discussion
To gain some insight into the mechanism of the RhI/binap-
catalysed 1,2-addition reaction, the transformation of cyclo-
hexenone 1a into product 2a was followed by 31P NMR
spectroscopy and simultaneous GC analysis. During the re-
action, the large majority of the detected phosphorous spe-
28
6
n.d.
94ꢁ1
75ꢁ1
96ꢁ1
AgBF4 (1.5)
97ꢁ0
78ꢁ1
93ꢁ1
[a] Actual amount of rhodium. [b] Determined by GC. Values with stan-
dard deviation represent the mean of 3–4 experiments. [c] After 24 h:
97% conv., 68% yield. [d] Performed with 1.2 mol% (R)-binap for 3 h.
[e] [Rh
source.
CAHTUGTNREN(NGU cod)2]BF4 as rhodium source. [f] [{RhAHCTUTGNRNE(NUGN C2H4)2Cl}2] as rhodium
cies was in fact dimeric complex [{RhACTHNUTRGNEG(NU binap)X}2] (X=OMe,
Cl), which is formed upon premixing the respective dimeric
cycloocta-1,5-diene (cod) complex and the chiral phosphane.
This binap complex must be a precatalyst that slowly liber-
ates the catalytically active species. Thus, we speculated that
promoting the cleavage of this dimer would provide the key
to a more active catalyst, even though we previously ob-
um, whereas both the reaction rate and the chemoselectivity
greatly decreased when only 1 mol% of rhodium was used,
especially after a prolonged reaction time (Table 1, entries 1
and 2). With the same catalyst loading (1 mol%), high selec-
tivity was once again obtained upon addition of AgBF4
(Table 1, entry 3). Lowering the amount of rhodium to
0.5 mol% resulted in diminished reproducibility of the 1,2-
addition and, interestingly, the beneficial effect of the silver
salt was no longer observed (Table 1, entries 4 and 5). In the
presence of a larger excess of AgBF4, the product of the re-
action was almost racemic (Table 1, entry 6). In the absence
of any rhodium species, product 2b was formed rather unse-
lectively (Table 1, entry 7). Finally, using 1 mol% of [Rh-
served inferior results with [RhACTHNUTRGEN(UNG cod)2]BF4 as rhodium
source.[6] Fortunately, addition of a slight excess of AgBF4
(based on the amount of chloride) led to a quantitative
yield of the 1,2-addition product 2a with just 0.5 mol% of
dimeric [{RhACHTUNGTRENNUNG(binap)Cl}2].
Abstract in German: Die Entwicklung einer optimierten
Vorschrift fꢀr die RhI/binap-katalysierte enantioselektive 1,2-
Addition von AlMe3 an cyclische Enone wird beschrieben.
Eine Analyse des Reaktionsverlaufs mittels 31P-NMR wies
auf ein Chlorid-verbrꢀcktes Dimer als Resting State des Kata-
lysators hin; tatsꢁchlich gelang eine in-situ-Aktivierung durch
Zusatz von AgBF4 als Additiv. Auf diese Weise kann die Ka-
talysatorbeladung auf nur noch 1 mol% Rhodium reduziert
werden. Die Methode erlaubt die Umsetzung verschiedener
5–7-gliedriger cyclischer Enone mit exzellenten Enantioselek-
tivitꢁten und hohen Ausbeuten. Die erhaltenen tertiꢁren
Allylalkohole sind vielseitige Synthesebausteine und wurden
zu einer hoch enantioselektiven formalen Totalsynthese von
(ꢀ)-Frontalin und zur Bildung eines bicyclischen Lactons
mit dem Grundgerꢀst des natꢀrlichen Duftstoffs Weinlacton
verwendet.
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
consistently led to inferior results (compare Table 1, entry 3
versus entries 8 and 9).
Moreover, the effects of the anion of the silver salts were
examined using 0.5 mol% of rhodium (Table 2). The hexa-
fluoroantimonate, hexafluoroarsenate and perchlorate silver
salts were less suitable than the tetrafluoroborate salt
(Table 2, entries 1–3), and the hexafluorophosphate, trifluor-
oacetate and triflimide salts afforded slightly lower levels of
selectivity than AgBF4 (compare Table 2, entries 4–6 versus
Table 1, entry 5). In the presence of 0.5 mol% of rhodium,
AgOTf appeared to be superior to AgBF4, but when the
amount of rhodium was increased to 1 mol%, AgBF4 again
performed better (compare Table 2, entries 7 and 8 versus
Table 1, entries 3 and 5).
Chem. Eur. J. 2013, 19, 16366 – 16373
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16367

P. von Zezschwitz et al.
precatalyst, as observed by 31P NMR spectroscopy, and, pre-
sumably, also a cationic aluminium species[15] [Eq. (2) and
(3)].
Table 2. Screening of various silver salts.
Entry
Additive
Conv
[%][a]
Yield
[%][a]
ee
[%][a]
1
2
3
4
5
6
AgSbF6
AgAsF6
AgClO4
AgPF6
AgO2CCF3
AgNTf2
AgOTf
AgOTf
54
45
44
84ꢁ2
49ꢁ6
58ꢁ1
56ꢁ3
82ꢁ2
20
17
24
49ꢁ7
33ꢁ6
36ꢁ3
48ꢁ7
79ꢁ2
86
83
68
90ꢁ2
91ꢁ1
89ꢁ2
92ꢁ2
97ꢁ1
Therefore, some amount of the precipitated chloride obvi-
ously becomes available upon addition of AlMe3. If the
solid AgCl was removed by filtration prior to addition of
the alane, the resulting catalyst was less selective.[16,17] Simi-
larly to the effect of the cod ligand, there is obviously
a subtle effect of the chloride on the reaction. On the one
hand, it slows down formation of the reactive catalyst spe-
cies; on the other hand, it may stabilize the catalyst against
decomposition.
7
8[b]
[a] Determined by GC. Values with standard deviation represent the
mean of 3 experiments. [b] With 0.5 mol% [{RhACTHNUTRGEN(UNG cod)Cl}2], 1.2 mol%
(R)-binap, and 1.5 mol% AgOTf.
Notably, GC analysis of these reactions never indicated
complete conversion, regardless of the reaction time, even if
NMR analysis of the crude products after an aqueous
workup showed no remaining enone 1b. These contradictory
observations are due to formation of dimeric compound 3
(in yields below 5%), which undergoes a retro-aldol reac-
tion to form enone 1b in the
Using the optimised reaction conditions, various cycloalk-
2-enones with 5–7-membered rings were converted into the
corresponding tertiary allylic alcohols, all with ꢂ 95% ee
(Table 3). The yields achieved when 0.5 mol% of the dimer-
ic [{Rh
used compare very well to yields previously obtained with
2.5 mol% of [{Rh(cod)OMe}2] in the absence of silver
ACHTUNGRTNE(NUNG cod)Cl}2] precursor and 1.5 mol% of AgBF4 were
AHCTUNGTRENNUNG
GC injector (Figure 1).
The results of the catalyst op-
timisation experiments, de-
scribed above, suggest that the
salts.[6] When 6-membered rings are used as substrates yields
are generally ꢂ 80%, with the single exception of a substrate
with geminal disubstitution at the C5 position (Table 3,
entry 3). Also, the method described does not tolerate sub-
stituents at the C2 or C3 positions. Cyclopentenones, which
were so far unsuitable substrates, now furnish synthetically
useful yields of 37–58%, but still require a higher catalyst
loading (Table 3, entries 6–8). A good yield was also ach-
ieved with cyclohept-2-enone (1i, Table 3, entry 9).[18]
catalytic system is rather sensi-
Figure 1. Side
product
3
tive towards the presence of cy-
cloocta-1,5-diene (cod). The
best results are achieved with
a 1:1 ratio of diene and rhodi-
formed during the synthesis of
compound 2b.
um (Table 1, entry 3), but both a 2:1 ratio (Table 1, entry 8)
or the complete absence of cod (Table 1, entry 9) is unfav-
ourable. Moreover, the transformation is not as successful
when extra cod is added.[13]
The utility of cyclic tertiary allylic alcohols 2 is demon-
strated with a formal total synthesis of (ꢀ)-frontalin (6,
Scheme 2), which is the aggregation pheromone of the bark
beetle[19] and the male sex pheromone of Elephas maximus,
the Asian elephant.[20] More than 30 syntheses of enantioen-
riched frontalin have been performed and in 1998 Mori
et al. reported an outstanding large-scale approach deliver-
ing 10 g of this compound with 89% ee in 7.8% yield over
10 steps.[21] The acetonide-protected triol 5 is a key inter-
mediate and is obtained in six steps in Moriꢂs synthesis.[22]
Starting with an asymmetric 1,2-addition to cyclohex-2-
enone (1a), protected triol 5 is now available in 3 steps by
ozonolysis of compound 2a with a reductive workup, fol-
lowed by protection of the 1,2-diol moiety. Based on the
subsequent transformations by Mori et al., this new ap-
proach represents a formal total synthesis of (ꢀ)-frontalin
(6) in 19% yield over 7 steps with 99% ee.[23] Thus, a signifi-
cant improvement was achieved that is noteworthy because
of the increasing use of frontalin for the regulation of beetle
populations,[24] and because of the observation that only the
(ꢀ)-enantiomer is bioactive.[21]
Inspection of the catalyst mixture used for Table 1,
entry 3 by 31P NMR spectroscopy revealed the presence of
[RhACHTUNGTRENNUNG(binap)ACHTUNGTRENNUNG(cod)]BF4 as the main compound prior to addi-
tion of the enone and AlMe3.[14] Upon addition of AlMe3,
the reaction mixture immediately turned black, thus indicat-
ing the formation of elemental silver; independent treat-
ment of a suspension of AgCl in THF with AlMe3 led to the
same black precipitate. The presence of ethane and
AlMe2Cl was proven by GC-MS analysis of the gas phase
and 1H NMR analysis of the liquid phase, respectively
[Eq. (1)].
The picture becomes rather puzzling because AlMe2Cl
rapidly reacts with the monomeric rhodium complex. This
reaction leads to formation of the chloride-bridged dimeric
16368
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16366 – 16373

Synthesis of Cyclic Tertiary Allylic Alcohols
Table 3. Asymmetric 1,2-addition to cycloalkenones.
FULL PAPER
Entry
Enone
T
[8C]
t
Yield
[%][a]
ee
[h]
[%][b]
1
RT
60
0.5
88(84)
85(86)
34(31)
83
99
97
Scheme 2. Formal total synthesis of (ꢀ)-frontalin (6).
and Mislow–Evans rearrangements furnishing allylic amines
and sulfoxides, respectively, have already been reported
with cyclic allylic alcohol substrates.[25] We decided to
pursue a Claisen rearrangement. At first, the Johnson proto-
col, that is, treatment with triethyl orthoacetate in the pres-
ence of propionic acid, was attempted; however, unsatisfy-
ing results were achieved. This might be caused by the
acidic conditions, which favour dehydration of the starting
material. In contrast, the Ireland protocol proceeds under
alkaline conditions. For this transformation, allylic alcohol
2a was acetylated with Ac2O to afford compound 7 in 80%
yield (Scheme 3).[26] Formation of the respective trimethyl-
2
2
3
RT
RT
3
98
4[c]
1
>99
5
0
2.5
2
90
>99
6[d]
RT
37(10)
95
7[e]
40
4
2
2
58(28)
45
95
99
99
8
RT
RT
Scheme 3. Claisen rearrangement and Tsuji–Trost reaction of acetate 7
(DMAP=4-dimethylaminopyridine; KHMDS=potassium hexamethyl
disilazide; TMSCl=trimethylsilyl chloride).
9
72(74)
silyl enol ether at ꢀ788C, [3,3]-sigmatropic rearrangement
in refluxing toluene and hydrolysis of the initially formed
silyl ester furnished desired carboxylic acid 8 in 66%
yield.[27] Compound 8 is a versatile chiral building block and
racemic derivatives have previously been prepared either by
related rearrangements[28a,b] or by a Tsuji–Trost reaction that
starts from diethyl malonate and 3-methylcyclohex-2-enyl
acetate (the regioisomer of compound 7).[28c] To prove chir-
ality transfer, acetate 7 (98% ee) was similarly transformed,
the product 9 being obtained in 81% yield with 93% ee
(Scheme 3). Thus, transition-metal-catalysed allylic substitu-
[a] Isolated yield. Values in parentheses with 2.5 mol% [{RhACTHUNRGTNEUNG(cod)OMe}2]
without AgBF4.[6] [b] Determined by GC. [c] (S)-binap was used to form
the cis-configured product as a single diastereomer (matched pair, com-
pare ref. [10]). [d] With 2.5 mol% [{RhACHTUNRGTNEUNG(cod)Cl}2], 6.0 mol% (R)-binap,
and 7.5 mol% AgBF4. [e] With 1.0 mol% [{Rh
binap, and 3.0 mol% AgBF4.
ACHTUNGTERN(NUNG cod)Cl}2], 2.4 mol% (R)-
Besides oxidative degradation of the C=C bonds, cyclic al-
lylic alcohols 2 are well-suited for sigmatropic rearrange-
ments that proceed with the transfer of chirality. Overman
Chem. Eur. J. 2013, 19, 16366 – 16373
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16369

P. von Zezschwitz et al.
tions represent further useful applications of allylic alcohols
2.[29]
Carboxylic acid 8 is suitable for cyclization to form the
core structure of bicyclic lactones, which frequently occur
either as natural products themselves or as intermediates in
natural product syntheses (Scheme 4). The famous “wine
misation indicated a significant influence of the amount of
cod used and pointed to a subtle effect of the content of
chloride in the reaction mixture. This unique enantioselec-
tive 1,2-addition of methyl groups to cyclic enones furnishes
high yields of products with just 1 mol% of rhodium catalyst
and is applicable to various unsubstituted, as well as, substi-
tuted cycloalkenones. The only restriction appears to be that
the substrate must not contain a substituent at the C2 or C3
positions. On the basis of the synthetic applications in this
report, the resulting allylic alcohols are highly versatile syn-
thons, especially because oxidative cleavage of the C=C
bond provides access to acyclic scaffolds. Currently, ad-
vanced NMR studies are being performed together with col-
laboration partners to get a better understanding of the re-
action mechanism. This understanding might help to further
increase the substrate scope towards acyclic enones or cyclo-
alkenones with tri- or even tetrasubstituted C=C bonds.
Experimental Section
General: 1H NMR spectra were recorded at 300 MHz on a Bruker AV-
300 or at 500 MHz on a Bruker DRX 500 spectrometer. Chemical shifts
are reported as d values relative to the residual proton signal (CHCl3:
d=7.26 ppm, [D5]DMSO: 2.50 ppm) as an internal reference. 13C NMR
spectra were recorded at 125 MHz on a Bruker DRX 500, at 75.5 MHz
on a Bruker AV-300 or at 62.5 MHz on a Bruker DPX 250 spectrometer.
Scheme 4. Synthesis of bi- and tricyclic lactones (DBU=1,8-diazabicyclo-
ACHTUNGTRENNUNG[5.4.0]undec-7-ene; LDA=lithium diisopropylamide).
Chemical shifts are reported as
d values relative to CDCl3 (d=
77.16 ppm) or [D6]DMSO (d=39.52 ppm). 31P NMR spectra were record-
lactone” (10) is a flavour component with an odour thresh-
old of only 0.01–0.04 pg Lꢀ1[30a] that has been isolated from
Koala urine,[30b] but is also found in wine[30c] black pep-
per,[30d] and fruit juices.[30e] In addition, lactone 11 is an inter-
mediate in the syntheses of (S)-actinidiolide and epiloliolide
by Mori et al.;[31] a synthesis of paeonilactone A by Bꢀckvall
et al. involved lactone 12.[32] In a similar way to the prepara-
tion of compound 10 by Helmchen et al. and Bartlett
et al.,[33] compound 8 was transformed in an iodolactonisa-
tion. Interestingly, this reaction proceeded with 6:1 regiose-
lectivity furnishing iodolactones 13 and 14. Major product
13 was subjected to an elimination reaction, furnishing the
unsaturated lactone 15; the structure of 15 was proven by
X-ray analysis.[34] In addition, determination of the ee value
showed that no racemisation took place along the reaction
sequence starting from alcohol 2a. To finish the preparation
of an isomer of wine lactone 10, a diastereoselective methyl-
ation was performed, compound 16 being obtained in 77%
yield. Finally, the structure of minor product 14 from the io-
dolactonisation was also proven by X-ray analysis;[34] under
the reaction conditions of the elimination, iodolactone 14
was transformed into the tricyclic lactone 17 in an intramo-
lecular alkylation that led to a 3-exo-tet cyclisation.[35]
ed at 101 MHz on a Bruker DPX 250 spectrometer. Chemical shifts are
reported as
d values relative to H3PO4 as internal reference (d=
0.00 ppm). IR spectra were recorded on a Bruker Alpha-P FT-IR spec-
trometer. Electrospray ionisation (ESI) mass spectra were recorded on
a Finnigan LTQ FT spectrometer by the analytical service department of
the Philipps-Universitꢀt Marburg. Optical rotations were measured on
a Perkin–Elmer 241 polarimeter with concentrations in g/100 mL. Melt-
ing points are uncorrected. Gas chromatograms were recorded with a Shi-
madzu GC-2010 Plus with an AOC-20i autosampler. GC yields were de-
termined on a Supelco SPB-1 column (30 m, 0.32 mm internal diameter,
0.25 mm film) against mesitylene as internal standard; ee values were de-
termined on
a Chiral-Separations Cyclodextrin TA (6-TBDMS-2,3-
acetyl-b-cyclodextrin, 50% in PS086, 25 m, 0.25 mm internal diameter,
0.125 mm film) or a Cyclodextrin TE (6-TBDMS-2,3-ethyl-b-cyclodextrin,
30% in PS086, 30 m, 0.25 mm internal diameter, 0.125 mm film) column.
Helium was used as carrier gas and temperature programs are given with
the respective substances. Column chromatography was carried out on
MN Kieselgel 60 M (Machery–Nagel, 0.04–0.063 mm). TLC analysis was
carried out on precoated sheets (Merck DC Kieselgel 60 F254). Medium-
pressure liquid chromatography (MPLC) was carried out on Interchim
puriFlash SI-HP columns. Solvents used for extraction and chromatogra-
phy were of technical grade and distilled prior to use. All moisture-sensi-
tive reactions were carried out under dry nitrogen or argon in oven- or
flame-dried glassware. THF and toluene were distilled from sodium ben-
zophenone ketyl, dichloromethane and triethylamine from CaH2. Sam-
ples of the racemic allylic alcohols for GC analysis were prepared by
MeLi addition to the respective enones. Stock solutions of AlMe3 were
prepared by dissolving the neat reagent in anhydrous hexane.
(S)-1-Methylcyclohex-2-en-1-ol (2a):
A solution of [{RhACHTUNGTRENNUNG(cod)Cl}2]
(2.47 mg, 5.00 mmol) and (R)-binap (7.47 mg, 12.0 mmol) in anhydrous
THF (2.50 mL) was stirred for 1 h at room temperature. AgBF4 (29.2 mL,
15.0 mmol, 100 mgmLꢀ1 in THF) and cyclohex-2-enone (1a, 96.8 mL,
96.1 mg, 1.00 mmol) were added, followed by dropwise addition of
AlMe3 (235 mL, 1.20 mmol, 5.10m in hexane). After 0.5 h, the reaction
mixture was poured into a stirred mixture of pentane (25 mL) and H2O
Conclusion
In summary, preliminary 31P NMR analysis of the RhI/binap-
catalysed 1,2-addition of AlMe3 to enones allowed for the
development of an improved catalytic system. Careful opti-
16370
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16366 – 16373

Synthesis of Cyclic Tertiary Allylic Alcohols
FULL PAPER
(0.02 mL), and stirring was continued for 15 min. Na2SO4 (1.5 g) and
charcoal (0.7 g) were added and the crude product was obtained after fil-
tration and careful removal of the solvent under reduced pressure.
MPLC (pentane, 4 g silica gel 30 mm) yielded 99 mg (88%) of allylic alco-
hol 2a as a colourless oil (Rf =0.33, cyclohexane/EtOAc 2:1). The ee
value was determined by GC analysis (Cyclodextrin TA; 4 min 608C iso-
thermal!2 Kminꢀ1 to 1008C; 45 cmsꢀ1 gas flow), retention times:
12.8 min (S)-enantiomer, 13.2 min (R)-enantiomer, 99% ee; the physical
and spectroscopic data were consistent with those reported in the litera-
ture.[36]
compound 8 as a yellow oil (Rf =0.58). [a]2D2 = +38.8 (c=1.0 in EtOAc);
1H NMR (300 MHz, CDCl3): d=1.16–1.27 (m, 1H), 1.49–1.60 (m, 1H),
1.65 (d, J=0.6 Hz, 3H), 1.68–1.93 (m, 4H), 2.23–2.38 (m, 2H), 2.56 (m,
1H), 5.28 (br s, 1H), 10.34 ppm (br s, 1H); 13C NMR (75.5 MHz, CDCl3):
d=21.5, 24.0, 28.8, 30.1, 32.5, 41.0, 124.1, 135.8, 179.4 ppm; IR: n˜ =2923
(br), 2861, 2671, 1702 (s), 1409, 1338, 1288, 1205, 935, 856, 810, 694, 635,
445, 411 cmꢀ1; HRMS (ESI): m/z calcd for C9H13O2: 153.0921; found
153.0924.
Diethyl (R)-2-(3-methylcyclohex-2-en-1-yl)malonate (9): Sodium hydride
(49.9 mg, 2.08 mmol) was added at room temperature to a solution of di-
ethyl malonate (346 mL, 363 mg, 2.27 mmol) in THF (4.0 mL) and stirred
for 15 min. In a separate flask, compound 7 (100 mg, 648 mmol), [Pd-
(S)-2-Methylhexane-1,2,6-triol (4): (S)-1-Methylcyclohex-2-en-1-ol (2a,
195 mg, 1.74 mmol) was dissolved in anhydrous CH2Cl2 (25 mL), and
a stream of ozone in oxygen was passed through the solution at ꢀ788C
until a blue colour persisted (15 min). Excess ozone was first displaced
with oxygen, then with argon. Anhydrous THF (25 mL) and LiAlH4
(100 mg, 2.64 mmol) were added to the solution, and it was allowed to
warm to room temperature. After 22 h, saturated aqueous potassium
sodium tartrate (9 mL) was added in portions at 08C, the mixture was fil-
tered over Celite, and the filtrate concentrated under reduced pressure.
Flash chromatography (CH2Cl2/MeOH 10:1!7:1) yielded 140 mg (54%)
of triol 4 as a highly viscous, colourless oil (Rf =0.14, CH2Cl2/MeOH
10:1). [a]2D0 =ꢀ2.1 (c=1.0 in EtOAc); 1H NMR (300 MHz, [D6]DMSO):
d=0.97 (s, 3H), 1.23–1.41 (m, 6H), 3.14 (m, 2H), 3.37 (m, 2H), 3.95 (s,
1H), 4.34 (t, J=5.1 Hz, 1H), 4.44 ppm (t, J=5.7 Hz, 1H); 13C NMR
(75.5 MHz, [D6]DMSO): d=19.7, 24.1, 33.5, 38.4, 60.9, 69.0, 71.5 ppm;
HRMS (ESI): m/z calcd for C7H16O3Na: 171.0992; found: 171.0995.
AHCTUNTGRENNG(UN PPh3)4] (22.5 mg, 19.5 mmol) and triphenylphosphine (15.3 mg,
58.3 mmol) were dissolved in THF (1.0 mL) and stirred for 15 min. After-
wards, the solution of sodium malonate was added, and the mixture was
stirred for an additional 15 h. This solution was then poured into Et2O
(20 mL) and water (10 mL) and the aqueous phase was extracted with
Et2O (2ꢃ10 mL). The organic phases were dried over MgSO4 and the
solvent was removed under reduced pressure. Flash chromatography
(hexane/EtOAc 15:1) yielded 134 mg (81%) of malonate 9 as a colourless
oil (Rf =0.28). The ee value was determined by GC analysis (Cyclodextrin
TE; 4 min 608C isothermal!1 Kminꢀ1 to 1008C!20 Kminꢀ1 to
1608C!10 min isothermal; 45 cmsꢀ1 gas flow), retention times: 50.4 min
(S)-enantiomer, 50.6 min (R)-enantiomer, 93% ee; [a]2D4 = +19 (c=1.3 in
EtOAc); 1H NMR (500 MHz, CDCl3): d=1.25–1.33 (m, 1H), 1.26 (t, J=
7.1 Hz, 3H), 1.27 (t, J=7.1 Hz, 3H), 1.51–1.60 (m, 1H), 1.64 (s, 3H),
1.69–1.76 (m, 2H), 1.83–1.96 (m, 2H), 2.87 (m, 1H), 3.20 (d, J=9.6 Hz,
1H), 4.16–4.24 (m, 4H), 5.25 ppm (s, 1H); 13C NMR (126 MHz, CDCl3):
d=14.3 (2C), 21.4, 24.1, 26.6, 30.0, 35.7, 57.6, 61.3 (2C), 121.9, 136.8,
168.75, 168.81 ppm; IR: n˜ =2931, 2866, 1734 (s), 1451, 1371, 1331, 1295,
1257, 1150, 1033, 862, 810 cmꢀ1; HRMS (ESI): m/z calcd for C14H22O4H:
255.1591; found 255.1593; m/z calcd for C14H22O4Na: 277.1410; found
277.1412.
(S)-4-(2,2,4-Trimethyl-1,3-dioxolan-4-yl)butan-1-ol (5): Triol 4 (64.0 mg,
432 mmol), 2,2-dimethoxypropane (52.9 mL, 45.0 mg, 432 mmol), and
pTsOH·H2O (4.11 mg, 21.6 mmol) were dissolved in anhydrous toluene
(7.0 mL) and stirred for 4 h at 608C. The mixture was then concentrated
under reduced pressure. Flash chromatography (cyclohexane/EtOAc 1:1)
yielded 69 mg (85%) of dioxolane 5 as a colourless oil (Rf =0.20, cyclo-
hexane/EtOAc 1:1). [a]1D9 =ꢀ2.6 (c=1.0 in CHCl3); HRMS (ESI): m/z
calcd for C10H20O3Na: 211.1305; found: 211.1304; The NMR spectroscop-
ic data were consistent with those reported in the literature.[22a]
(3aS,7R,7aR)-7-Iodo-7-methylhexahydrobenzofuran-2(3H)-one (13) and
ACHUTNGRENUN(G 1S,5S,9S)-9-Iodo-1-methyl-2-oxabicycloACHTUNTGREN[NUGN 3.3.1]nonan-3-one (14): A solu-
tion of compound 8 (427 mg, 2.77 mmol) in Et2O (9.5 mL) and aqueous
NaHCO3 (9.4 mL, 4.7 mmol, 0.50n) was stirred at room temperature for
20 min. Then, a solution of potassium iodide (2.75 g, 16.6 mmol) and
iodine (1.40 g, 5.52 mmol) in H2O (11.0 mL) was added and the reaction
mixture was stirred for 23 h. Saturated aqueous Na2S2O3 was added until
a clear solution was obtained. The mixture was then extracted with Et2O
(4ꢃ30 mL). The combined organic phases were washed with saturated
aqueous NaHCO3 (100 mL) and brine (100 mL), dried (MgSO4) and con-
centrated under reduced pressure. Flash chromatography (cyclohexane/
EtOAc 1:1) yielded 626 mg (81%) of lactone 13 (Rf =0.61, cyclohexane/
EtOAc 2:1) and 104 mg (13%) of lactone 14 (Rf =0.53, cyclohexane/
EtOAc 2:1) as colourless solids. Lactone 13: Mp: 61–628C; [a]2D2 =ꢀ75.8
(c=1.0 in EtOAc); 1H NMR (300 MHz, CDCl3): d=0.99–1.27 (m, 2H),
1.72–1.84 (m, 3H), 1.95–2.06 (m, 1H), 2.24 (d, J=18.0 Hz, 1H), 2.25 (s,
3H), 2.69 (dd, J=16.7, 6.3 Hz, 1H), 2.86–2.95 (m, 1H), 4.68 ppm (d, J=
3.4 Hz, 1H); 13C NMR (75.5 MHz, CDCl3): d=23.6, 26.6, 33.8, 34.8, 38.0,
39.2, 52.0, 86.0, 176.4 ppm; IR: n˜ =2938, 2865, 1768 (s), 1447, 1419, 1374,
1320, 1288, 1221, 1177, 1143, 1049, 1027, 965, 937, 888, 853, 829, 767, 687,
597, 557, 507, 428 cmꢀ1; HRMS (ESI): calcd for C9H13IO2Na: 302.9852;
found 302.9863. Lactone 14: Mp: 114–1168C; [a]2D2 = +23.4 (c=1.0 in
EtOAc); 1H NMR (300 MHz, CDCl3): d=1.54 (s, 3H), 1.58–1.71 (m,
3H), 1.75–1.86 (m, 1H), 2.13–2.26 (m, 2H), 2.45 (m, 1H), 2.57 (dd, J=
18.5, 1.1 Hz, 1H), 2.95 (ddd, J=18.5, 6.8, 1.1 Hz, 1H), 4.44 ppm (m, 1H);
13C NMR (62.5 MHz, CDCl3): d=17.1, 26.5, 30.5, 32.3, 34.2, 36.4, 37.0,
82.7, 171.1 ppm; IR: n˜ =2981, 2929, 2866, 1779, 1712 (s), 1444, 1409,
1372, 1336, 1304, 1278, 1228 (s), 1176, 1148, 1109, 1081, 1048, 1016, 966,
946, 897, 819, 666, 571, 544, 517, 493, 435 cmꢀ1; HRMS (ESI): m/z calcd
for C9H13IO2Na: 302.9852; found 302.9862; Crystals suitable for X-ray
analysis were obtained by slow evaporation from pentane.
(S)-1-Methylcyclohex-2-enyl acetate (7): (S)-1-Methylcyclohex-2-en-1-ol
(2a, 112 mg, 1.00 mmol), triethylamine (282 mL, 205 mg, 2.00 mmol),
acetic anhydride (190 mL, 205 mg, 2.02 mmol), and 4-dimethylaminopyri-
dine (12.2 mg, 100 mmol) were stirred in CH2Cl2 (2.0 mL) at room tem-
perature for 48 h. The mixture was diluted with CH2Cl2 (70 mL) and
washed with KHSO4 (15 mL, 10% aq), NaHCO3 (10 mL, saturated aq)
and water (2ꢃ10 mL). The organic phase was dried over MgSO4 and con-
centrated under reduced pressure. Distillation (room temperature, 7.8ꢃ
10ꢀ2 mbar) yielded 124 mg (80%) of 7 as a colourless oil. The ee value
was determined by GC analysis (Cyclodextrin TA; 4 min 608C isother-
mal!2 Kminꢀ1 to 1008C; 45 cmsꢀ1 gas flow), retention times: 9.4 min
(S)-enantiomer, 9.9 min (R)-enantiomer, 98% ee; [a]2D2 =ꢀ167.5 (c=1.0
in CH2Cl2); 1H NMR (300 MHz, CDCl3): d=1.53 (s, 3H), 1.57–1.81 (m,
3H), 1.88–2.12 (m, 3H), 1.97 (s, 3H), 5.78–5.84 (m, 1H), 6.03–6.08 ppm
(m, 1H); 13C NMR (75.5 MHz, CDCl3): d=18.9, 22.5, 25.0, 26.1, 35.7,
78.7, 130.4, 130.5, 170.6 ppm; IR: n˜ =3034, 2937, 2872, 2837, 1728 (s),
1439, 1367, 1242 (s), 1200, 1172, 1144, 1091, 1018, 967, 940, 913, 874, 841,
813, 732, 684, 609, 502, 471 cmꢀ1
C9H14O2Na: 177.0886; found 177.0883.
; HRMS (ESI): m/z calcd for
(S)-2-(3-Methylcyclohex-2-enyl)acetic acid (8): At ꢀ788C a solution of 7
(841 mg, 5.45 mmol) in toluene (15 mL) was added dropwise to a solution
of potassium hexamethyldisilazide (9.89 mL, 6.55 mmol, 15% in toluene)
in toluene (10 mL). After 30 min at this temperature, a solution of trime-
thylsilyl chloride (1.11 mL, 950 mg, 8.75 mmol) and triethylamine
(1.21 mL, 880 mg, 8.70 mmol) in toluene (15 mL) was added dropwise.
The cooling bath was removed after 30 min, and the mixture was stirred
for 1 h at room temperature and for 5 h under reflux. After cooling, the
mixture was poured into aqueous NaOH (20 mL, 2n) and stirred for
15 min, followed by dilution with Et2O (40 mL) and extraction with aque-
ous NaOH (3ꢃ40 mL, 2n). The combined aqueous phases were acidified
to pH 2 with hydrochloric acid (6m) and extracted with Et2O (3ꢃ
150 mL). Drying (Na2SO4) and evaporation of the solvent followed by
flash chromatography (cyclohexane/EtOAc 1:1) yielded 557 mg (66%) of
A
ACHTUNGTERN(NGNU 7aH)-one (15): A
AHCTUNGTRENNUNG
Chem. Eur. J. 2013, 19, 16366 – 16373
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16371

P. von Zezschwitz et al.
EtOAc, Et2O and H2O (10 mL each). The aqueous phase was acidified to
pH 2 with hydrochloric acid (6m) and extracted with Et2O (3ꢃ30 mL).
The combined organic phases were dried (MgSO4) and concentrated
under reduced pressure. Flash chromatography (cyclohexane/EtOAc 5:1)
yielded 320 mg (94%) of lactone 15 as an off-white solid (Rf =0.36, cyclo-
hexane/EtOAc 2:1). The ee value was determined by GC analysis (Cyclo-
dextrin TE; 4 min 608C isothermal!2 Kminꢀ1 to 1008C!20 Kminꢀ1 to
1608C!5 min isothermal; 45 cmsꢀ1 gas flow), retention times: 28.6 min
(3aS,7aR)-enantiomer, 29.0 min (3aR,7aS)-enantiomer, 99% ee; mp:
558C; [a]2D2 = +23.7 (c=1.0 in CHCl3); 1H NMR (300 MHz, CDCl3): d=
1.35–1.48 (m, 1H), 1.65–1.74 (m, 1H), 1.82 (d, J=1.6 Hz, 3H), 1.93–2.17
(m, 2H), 2.33 (dd, J=17.2, 3.4 Hz, 1H), 2.50–2.61 (m, 1H), 2.74 (dd, J=
17.2, 8.1 Hz, 1H), 4.62 (d, J=5.9 Hz, 1H), 5.78 ppm (s, 1H); 13C NMR
(75.5 MHz, CDCl3): d=20.8, 23.2, 23.9, 34.1, 35.3, 79.5, 128.4, 130.3,
176.7 ppm; IR: n˜ =2950, 2916, 2851, 1750 (s), 1446, 1409, 1381, 1333,
1291, 1249, 1229, 1201, 1167 (s), 1089, 1071, 1023, 974, 932, 873, 828, 792,
Chem. Soc. Rev. 2009, 38, 1039–1075; c) A. Alexakis, J. E. Bꢀckvall,
N. Krause, O. Pꢅmies, M. Diꢆguez, Chem. Rev. 2008, 108, 2796–
2823; d) K. Yoshida, T. Hayashi, Modern Rhodium-Catalyzed Or-
ganic Reactions (Ed.: P. A. Evans), Wiley-VCH, Weinheim, 2005,
pp. 55–77.
[2] a) A. Lumbroso, M. L. Cooke, B. Breit, Angew. Chem. 2013, 125,
1942–1986; Angew. Chem. Int. Ed. 2013, 52, 1890–1932; b) D. M.
Hodgson, P. G. Humphreys, Science of Synthesis: Houben–Weyl
Methods of Molecular Transformations (Ed.: J. P. Clayden), Thieme,
Stuttgart, 2007, 36, pp. 583–665.
[3] a) A. V. R. Madduri, A. J. Minnaard, S. R. Harutyunyan, Chem.
Commun. 2012, 48, 1478–1480; b) K.-H. Wu, D.-W. Chuang, C.-A.
Chen, H.-M. Gau, Chem. Commun. 2008, 2343–2345; c) M. Hatano,
T. Miyamoto, K. Ishihara, Org. Lett. 2007, 9, 4535–4538.
[4] For an alternative approach through asymmetric epoxidation and
a subsequent Wharton reaction, see: H. Jiang, N. Holub, K. A. Jør-
gensen, Proc. Natl. Acad. Sci. USA 2010, 107, 20630–20635.
[5] a) D. J. Ramꢇn, M. Yus, Synlett 2007, 2309–2320; b) V. J. Forrat, O.
Prieto, D. J. Ramon, M. Yus, Chem. Eur. J. 2006, 12, 4431–4445;
c) S.-J. Jeon, H. Li, P. J. Walsh, J. Am. Chem. Soc. 2005, 127, 16416–
16425; d) H. Li, P. J. Walsh, J. Am. Chem. Soc. 2005, 127, 8355–
8361; e) J. M. Betancort, C. Garcꢈa, P. J. Walsh, Synlett 2004, 749–
760; f) H. Li, C. Garcꢈa, P. J. Walsh, Proc. Natl. Acad. Sci. USA
2004, 101, 5425–5427; g) S.-J. Jeon, P. J. Walsh, J. Am. Chem. Soc.
2003, 125, 9544–9545.
[6] J. Siewert, R. Sandmann, P. von Zezschwitz, Angew. Chem. 2007,
119, 7252–7254; Angew. Chem. Int. Ed. 2007, 46, 7122–7124.
[7] For review articles on the addition of organoaluminium reagents to
carbonyls, see: a) A. Kolb, P. von Zezschwitz, Top. Organomet.
Chem. 2013, 41, 245–276; b) P. von Zezschwitz, Synthesis 2008,
1809–1831; c) M. Oishi, Science of Synthesis: Houben–Weyl Meth-
ods of Molecular Transformations (Ed.: H. Yamamoto), Thieme,
Stuttgart, 2004, 7, pp. 261–385.
694, 617, 547, 495, 443, 418 cmꢀ1
; HRMS (ESI): m/z calcd for
C9H12O2Na: 175.0730; found 175.0729; crystals suitable for X-ray analysis
were obtained by slow evaporation from pentane/Et2O.
(3S,3aS,7aR)-3,7-Dimethyl-3,3a,4,5-tetrahydrobenzofuran-2ACHTNUTGRNEUNG(7aH)-one
(16): A solution of lactone 15 (130 mg, 854 mmol) in THF (3.0 mL) was
slowly added to a solution of lithium diisopropylamide (LDA) (3.51 mL,
926 mmol, 0.264m in THF) at ꢀ788C and stirred for 30 min. Methyl
iodide (304 mL, 693 mg, 4.88 mmol) was added and the mixture was
stirred for another 1.5 h. After addition of saturated aq NH4Cl (7.5 mL)
the reaction mixture was warmed to room temperature and extracted
with Et2O (3ꢃ10 mL). The combined organic phases were dried
(Na2SO4) and concentrated under reduced pressure. Flash chromatogra-
phy (CH2Cl2) yielded 109 mg (77%) of 16 as a colourless oil (Rf =0.42)
that solidified in the freezer (ꢀ288C). [a]2D4 =ꢀ38.5 (c=1.0 in EtOAc);
1H NMR (300 MHz, CDCl3): d=1.27 (d, J=7.1 Hz, 3H), 1.61–1.83 (m,
2H), 1.78 (d, J=1.7 Hz, 3H), 2.00–2.09 (m, 2H), 2.28–2.49 (m, 2H), 4.72
(d, J=6.8 Hz, 1H), 5.67 ppm (br s, 1H); 13C NMR (75.5 MHz, CDCl3):
d=14.2, 20.3, 21.5, 22.3, 37.6, 41.7, 78.0, 127.1, 131.3, 179.8 ppm; IR: n˜ =
2970, 2926, 1764 (s), 1449, 1380, 1328, 1198, 1164, 1083, 1029, 986, 957,
[8] K. Biswas, O. Prieto, P. J. Goldsmith, S. Woodward, Angew. Chem.
2005, 117, 2272–2274; Angew. Chem. Int. Ed. 2005, 44, 2232–2234.
[9] a) A. Gayet, S. Bertilsson, P. G. Andersson, Org. Lett. 2002, 4, 3777–
3779; b) D. P. G. Hamon, K. L. Tuck, Tetrahedron 2000, 56, 4829–
4835.
918, 892, 808, 592, 564, 503 cmꢀ1
C10H14O2Na: 189.0886; found 189.0887.
(2aR,2aS,2bR,5aS)-5a-Methylhexahydrocyclopropa[cd]benzofuran-2-
(2aH)-one (17): solution of 14 (104 mg, 371 mmol) and 1,8-
diazabicyclo[5.4.0]undec-7-ene (66.8 mL, 68.1 mg, 448 mmol) in THF
; HRMS (ESI): m/z calcd for
A
A
[10] A. Kolb, S. Hirner, K. Harms, P. von Zezschwitz, Org. Lett. 2012, 14,
1978–1981.
ACHTUNGTRENNUNG
(3.0 mL) was stirred under reflux for 16.5 h. After cooling, the mixture
was diluted with EtOAc and H2O (20 mL each). The aqueous phase was
acidified to pH 2 with hydrochloric acid (6m) and extracted with Et2O
(3ꢃ40 mL). The combined organic phases were dried (MgSO4), and con-
centrated under reduced pressure. Flash chromatography (cyclohexane/
EtOAc 5:1) yielded 38.9 mg (69%) of lactone 17 as a colourless oil (Rf =
0.39, cyclohexane/EtOAc 2:1). [a]2D6 = +24.0 (c=0.45 in CHCl3);
1H NMR (500 MHz, CDCl3): d=1.43–1.61 (m, 4H), 1.53 (s, 3H), 1.72–
1.77 (m, 1H), 1.80–1.85 (m, 1H), 1.94–2.01 (m, 1H), 2.10 ppm (d, J=
8.4 Hz, 2H); 13C NMR (125 MHz, CDCl3): d=16.4, 18.3, 20.0, 25.3, 27.9,
28.6, 31.9, 80.9, 175.2 ppm; IR: n˜ =2933, 2876, 1748 (s), 1454, 1379, 1354,
1326, 1288, 1255, 1218, 1185, 1156, 1119, 1022, 982, 946, 906, 823, 755,
654, 529, 504, 452 cmꢀ1; HRMS (ESI): m/z calcd for C9H13O2: 153.0910;
found 153.0907; m/z calcd for C9H12O2Na: 175.0730; found 175.0726.
[11] For a large scale (16 g) transformation of 1a, see: P. S. Charifson,
M. P. Clark, U. K. Bandarage, R. S. Bethiel, J. J. Court, H. Deng, I.
Davies, J. P. Duffy, L. J. Farmer, H. Gao, W. Gu, D. H. Jacobs, J. M.
Kennedy, M. W. Ledeboer, B. Ledford, F. Maltais, E. Perola, T.
Wang, M. W. Wannamaker, R. Byrn, Y. Zhou, C. Lin, M. Jiang, S.
Jones, U. A. Germann (Vertex Pharmaceuticals Inc.) US2012/
0171245, 2012 (pp. 197–198).
[12] For daily trading prices of precious metals, see: http://pm-prices.
heraeus.com/Heraeus_CurrentPrices.aspx?Lang=EN.
[13] Using the same conditions as those in Table 1, entry 3 with 3 mol%
of additional cod delivered 61% yield, 95% ee at 95% conversion
after 2 h at 608C.
[14] 31P NMR (101 MHz, THF): d=26.1 ppm (d, J=146.5 Hz). For
31P NMR data of [Rh
ACTHUNGRTNE(NUG binap)cod]BF4 in other solvents, see: a) T. Oh-
shima, H. Tadaoka, K. Hori, N. Sayo, K. Mashima, Chem. Eur. J.
2008, 14, 2060–2066 d=26.8 ppm (d, J=145 Hz) in CDCl3; b) A.
Preetz, H.-J. Drexler, C. Fischer, Z. Dai, A. Bçrner, W. Baumann,
A. Spannenberg, R. Thede, D. Heller, Chem. Eur. J. 2008, 14, 1445–
1451 d=25.7 ppm (d, J=145.8 Hz) in CD3OD; c) A. E. Russell,
N. O. Fuller, S. J. Taylor, P. Aurriset, J. P. Morken, Org. Lett. 2004, 6,
2309–2312 d=25.8 ppm (d, J=145.6 Hz) in ClCD2CD2Cl.
[15] a) W. Gu, M. R. Haneline, C. Douvris, O. V. Ozerov, J. Am. Chem.
Soc. 2009, 131, 11203–11212; b) S. Aldridge, A. J. Blake, A. J.
Downs, R. O. Gould, S. Parsons, C. R. Pulham, J. Chem. Soc. Dalton
Trans. 1997, 1007–1012.
Acknowledgements
We thank the Fonds der Chemischen Industrie for a scholarship for A.K.
and BASF SE, Ludwigshafen, and Symrise AG, Holzminden, for the gen-
erous donation of chemicals. We are grateful to Christian H. Mꢄller for
helpful discussions and to Stefan Lenz for technical assistance.
[16] Using the same reaction conditions as those in Table 1, entry 3, in
the absence of precipitated AgCl, furnished 93% conversion and
72% yield after 2 h at 608C.
[1] a) J.-X. Ji, A. S. C. Chan, Catalytic Asymmetric Synthesis, 3rd ed.
(Ed.: I. Ojima), John Wiley & Sons, Hoboken, 2010, pp. 439–495;
b) T. Jerphagnon, M. G. Pizzuti, A. J. Minnaard, B. L. Feringa,
16372
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16366 – 16373

Synthesis of Cyclic Tertiary Allylic Alcohols
FULL PAPER
[17] For a discussion of the influence of silver ions on gold-catalysed re-
actions, see: D. Wang, R. Cai, S. Sharma, J. Jirak, S. K. Thummana-
pelli, N. G. Akhmedov, H. Zhang, X. Liu, J. L. Petersen, X. Shi, J.
Am. Chem. Soc. 2012, 134, 9012–9019.
[18] As already reported for silver-free conditions, the use of AlEt3 leads
to reduction of the starting material, presumably owing to b-hydride
elimination. With ZnEt2, slow decomposition of the substrate was
observed.
[28] a) B. K. Ghorai, S. Menon, D. L. Johnson, J. W. Herndon, Org. Lett.
2002, 4, 2121–2124; b) P. A. Grieco, J. L. Collins, K. J. Henry, Tetra-
hedron Lett. 1992, 33, 4735–4738; c) V. A. Schmidt, E. J. Alexanian,
Angew. Chem. 2010, 122, 4593–4596; Angew. Chem. Int. Ed. 2010,
49, 4491–4494.
[29] For Mo-catalysed allylic substitutions of rac-2a, see: A. V. Malkov,
I. R. Baxendale, D. Dvorak, D. J. Mansfield, P. Kocovsky, J. Org.
Chem. 1999, 64, 2737–2750.
[19] G. W. Kinzer, A. F. Fentiman, T. F. Page, R. L. Foltz, J. P. Vite, G. B.
Pitman, Nature 1969, 221, 477–478.
[20] D. R. Greenwood, D. Comeskey, M. B. Hunt, L. E. L. Rasmussen,
Nature 2005, 438, 1097–1098.
[30] a) H. Guth, Helv. Chim. Acta 1996, 79, 1559–1571; b) I. A. South-
well, Tetrahedron Lett. 1975, 16, 1885–1888; c) H. Guth, J. Agric.
Food Chem. 1997, 45, 3022–3026; d) T. Jagella, W. Grosch, Eur.
Food Res. Technol. 1999, 209, 16–21; e) A. Buettner, P. Schieberle,
J. Agric. Food Chem. 2001, 49, 2387–2394.
[31] K. Mori, V. Khlebnikov, Liebigs Ann. Chem. 1993, 77–82.
[32] C. Jonasson, M. Rçnn, J.-E. Bꢀckvall, J. Org. Chem. 2000, 65, 2122–
2126.
[33] a) E. J. Bergner, G. Helmchen, Eur. J. Org. Chem. 2000, 419–423;
b) P. A. Bartlett, C. F. Pizzo, J. Org. Chem. 1981, 46, 3896–3900.
[34] CCDC-953012 (14) and CCDC-953013 (15) contain the supplemen-
tary 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. See the Supporting Infor-
mation for ORTEP drawings.
[35] Analogous structures have been obtained in intramolecular carbe-
noid additions, see: a) J.-L. Zhang, C.-M. Che, Org. Lett. 2002, 4,
1911–1914; b) P. Mꢄller, G. Bernardinelli, P. Nury, Tetrahedron:
Asymmetry 2002, 13, 551–558; c) M. P. Doyle, A. B. Dyatkin, A. V.
Kalinin, D. A. Ruppar, J. Am. Chem. Soc. 1995, 117, 11021–11022;
d) D. L. J. Clive, S. Daigneault, J. Org. Chem. 1991, 56, 3801–3814.
[36] V. Pace, L. Castoldi, P. Hoyos, J. V. Sinisterra, M. Pregnolato, J. M.
Sꢅnchez-Montero, Tetrahedron 2011, 67, 2670–2675.
[21] Y. Nishimura, K. Mori, Eur. J. Org. Chem. 1998, 233–236.
[22] For alternative preparations of compound 5 in 8 and 10 steps, re-
spectively, see: a) B. Ortiz, R. Sꢅnchez-Obregꢇn, R. A. Toscano, F.
Yuste, Synthesis 2008, 2105–2109; b) S.-s. Jew, D.-Y. Lim, J.-Y. Kim,
S.-j. Kim, E.-y. Roh, H.-J. Yi, J.-M. Ku, B.-s. Park, B.-s. Jeong, H.-g.
Park, Tetrahedron: Asymmetry 2002, 13, 155–159.
[23] For a synthesis of 6 with 85% ee by asymmetric addition of ZnMe2
to an acyclic enone and subsequent ozonolysis, see: M. Yus, D. J.
Ramꢇn, O. Prieto, Eur. J. Org. Chem. 2003, 2745–2748.
[24] a) Z. Liu, B. Xu, Z. Miao, J. Sun, Chem. Senses 2013, 38, 485–495;
b) B. S. Lingren, D. R. Miller, J. P. LaFontaine, J. Entomol. Soc. B.
C. 2012, 109, 72–74.
[25] For an Overman rearrangement of rac-2a, see: a) V. K. Aggarwal,
G. Y. Fang, Chem. Commun. 2005, 3448–3450; for a Mislow–Evans
rearrangement of a similar substrate, see: b) D. H. Hua, S. Venka-
taraman, R. Chan-Yu-King, J. V. Paukstelis, J. Am. Chem. Soc. 1988,
110, 4741–4748.
[26] Upon storage or chromatography on silica gel, compound 7 under-
goes allylic rearrangement: A. C. Oehlschlager, P. Mishra, S. Dhami,
Can. J. Chem. 1984, 62, 791–797.
[27] Protocol adapted from: P. Beaulieu, W. W. Ogilvie, Tetrahedron
Lett. 2003, 44, 8883–8885.
Received: August 2, 2013
Published online: October 24, 2013
Chem. Eur. J. 2013, 19, 16366 – 16373
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16373