β- and γ-disubstituted olefins: Substrates for copper-catalyzed asymmetric allylic substitution

Article abstract of DOI:10.1002/adsc.200800086

The copper-catalyzed asymmetric allylic alkylation has shown through many examples that it is a powerful means to generate stereogenic centers with mono β- and γ-substituted olefinic substrates. However, little has been reported about more substituted olefinic patterns, such as β-disubstituted allylic electrophiles. In this paper, we show that a simple procedure using easily accessible Grignard reagents and as low as 3 mol% of copper/ligand can promote high to nearly perfect enantioselectivities (up to >99% ee) with very good γ-selectivities on a wide panel of aliphatic or aromatic β-disubstituted substrates.

Full text of DOI:10.1002/adsc.200800086

FULL PAPERS
DOI: 10.1002/adsc.200800086
b- and g-Disubstituted Olefins: Substrates for Copper-Catalyzed
Asymmetric Allylic Substitution
Caroline A. Falciola,a Karine Tissot-Croset,a Hugo Reyneri,a
and Alexandre Alexakisa,*
a
Department of Organic Chemistry, University of Geneva, 30 quai E. Ansermet, 1211 Geneva 4, Switzerland
Fax : (+41)-22-379-3215; e-mail: alexandre.alexakis@chiorg.unige.ch
Received: February 9, 2008; Published online: April 18, 2008
Dedicated to Professor Andreas Pfaltz on the occasion of his 60th birthday.
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The copper-catalyzed asymmetric allylic al- high to nearly perfect enantioselectivities (up to
kylation has shown through many examples that it is >99% ee) with very good g-selectivities on a wide
a powerful means to generate stereogenic centers panel of aliphatic or aromatic b-disubstituted sub-
with mono b- and g-substituted olefinic substrates. strates.
However, little has been reported about more substi-
tuted olefinic patterns, such as b-disubstituted allylic
electrophiles. In this paper, we show that a simple Keywords: allylic alkylation; allylic compounds;
procedure using easily accessible Grignard reagents asymmetric catalysis; copper; nucleophilic substitu-
and as low as 3 mol% of copper/ligand can promote tion
Introduction
The allylic alkylation, also known as an SN2’ reaction,
has become a key reaction for creating carbon-carbon
bonds.[1] In the literature many metals have been
found to catalyze such reactions with the use of a
wide diversity of soft or hard nucleophiles.[2] Catalytic
Scheme 1.
amounts of copper usually enable a highly g-regiose-
lective process using different organometallic reagents
as nucleophilic sources.[3] Numerous enantioselective
variants have been developed over the past years with
a range of easily accessible organometallic reagents
such as Grignard reagents,[4] organozinc[5–6] and more
recently alkylaluminum[7] reagents, all of which have
been used with considerable success.
The olefinic patterns in early studies from our
group or others on the asymmetric Cu-catalyzed SN2’
reaction were limited to simple g- and b- monosubsti-
tuted frameworks (R1 =R2 =H; Scheme 1). Trisubsti-
tuted allylic substrates – such as S1 and S2, if E and Z
isomers are not a problematic issue – can also under-
go asymmetric allylic substitution (Scheme 2) but
have been poorly documented until recently. The for-
mation of quaternary carbon stereogenic centers
could be realized through SN2’ substitution of g-disub-
stituted allylic substrates (S1). However, few groups Scheme 2.
1090
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1090 – 1100

b- and g-Disubstituted Olefins
FULL PAPERS
have managed to exploit this reaction with good L2 (Table 1, entry 2). Similar results were obtained
regio-, chemo- and enantioselectivities, with the nota- with the para-methoxy-substituted ligand L8 of same
ble exception of the work of Hoveyda and co-work- diastereomeric nature, which suggested that there was
ers.[6] Conversely, b-disubstituted olefinic systems (S2- no electronic effect of the methoxy function on the
type allylic substrates) had not yet been covered, stereoselectivity of the reaction. Conversely, the
other than in stereoselective additions to Baylis– ortho-methoxy substituents present on ligand (S,SS)-
Hilman-derived allylic electrophiles as described by L5, otherwise highly effective in different reactions
Woodward and co-workers.[8] In addition, this last ex- with copper or iridium catalysis on unsubstituted cin-
ample was mechanistically closer to a conjugate (Mi- namyl substrates,[12,13] affected negatively the asym-
chael) addition, followed by elimination of the chlo- metric outcome with lower ee values of 80% (Table 1,
ride through the allylic rearrangement of the inter- entry 6). Interestingly, although most equivalent com-
mediate enolate. In 2006, a preliminary report by our binations of (S)-binaphthol unit and (S,S)-amines
group described the first successful attempts on these would enable a similar positive rotation in products
unexplored b-disubstituted target structures.[9]
3a, the later ligand L5 produced the enantiomer (À)-
3a. This observation suggests that the ortho-methoxy
moieties are coordinating/chelating and would modify
the conformation of the intermediate copper catalyst.
To further improve the selectivity in favor of the g
Result and Discussion
With the simplest model of trisubstituted olefins at product, catalyst loading of CuTC/L2 was increased
hand, namely b-methylcinnamyl chloride (1) and g- to an optimal amount of 3 mol% (Table 2, entry 3),
methylcinnamyl chloride (2) (Scheme 3), we conduct- which afforded 92% SN2’-selectivity and in parallel
ed a screening of different biphenol-[10] and binaph- improved the enantioselectivity from 96% to 98% ee.
thol-based[11,12] phosphoramidite ligands (Figure 1).
By accelerating the reaction and thus preventing the
Results for the first study on trans-(3-chloro-2- formation of a dialkyl-cuprate complex in the reaction
methylprop-1-enyl)benzene 1 are listed in Table 1. medium, the catalyst loading is an important parame-
The allylic substitution of 1 by the slow addition of ter that favors the formation of the branched product
ethylmagnesium bromide, catalyzed by copper thio- (3a).[2,12c,14]
phenecarboxylate (CuTC; 1 mol%) and different
Because the formation of enantiomerically enriched
phosphoramidite ligands (1.1 mol%), gave adducts (3) quaternary centers is highly valuable and could easily
with poor-to-good g/a ratios and provided high enan- be accessed through an asymmetric allylic substitu-
tiomeric excesses of up to 96% ee with ligand (S,SS)- tion, the g-methylcinnamyl chloride (2) was subjected
Scheme 3. Enantioselective allylic substitution of b-methylcinnamyl (1) and g-methylcinnamyl chlorides (2).
Figure 1. Ligands used for the stereoselective allylic substitution of trisubstituted allylic halides.
Adv. Synth. Catal. 2008, 350, 1090 – 1100
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1091

Caroline A. Falciola et al.
FULL PAPERS
Table 1. Ligand screening for the copper-catalyzed allylic
substitution of b-methylcinnamyl chloride (1) with ethyl-
magnesium bromide (L*=L1–L12; Figure 1).
Table 3. Ligand screening for the copper-catalyzed allylic
substitution of g-methylcinnamyl chloride (2) with ethylmag-
nesium bromide (L*=L1–L12; Scheme 3)
Entry[a] Substrate
Ligand Yield
[%][b]
SN2’/
ee
Entry[a] Substrate
Ligand Conv.
[%][b]
SN2’/
ee
SN2[c]
[%][d]
SN2[c]
[%][d]
1
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
1 (R1 =H,
R2 =Me)
L1 (SS)
N
65/35
87/13
30/70
89/11
87/13
73/27
71/29
86/14
37/63
56/44
59/41
31/69
53(+)
96(+)
27(À)
81(À)
80(À)
61(À)
52(+)
96(+)
26(À)
15(+)
71(+)
22(À)
1
2 (R1 =Me,
R2 =H)
L1 (SS) >99
70/30
44/56
59/41
86/14
77/23
44/56
80/20
61/39
73/27
46/54
87/13
52/48
29(À)
26(À)
11(À)
20(+)
35(+)
30(À)
26(À)
15(À)
20(À)
20(À)
5(À)
3
L2
(S,SS)
L3
(R,SS)
L4 (SS)
2
2 (R1 =Me,
R2 =H)
L2
(S,SS)
L3
>99
>99
2
3
2 (R1 =Me,
R2 =H)
(R,SS)
L4 (SS) >99
4
4
2 (R1 =Me,
R2 =H)
(54)
6
L5
(S,SS)
L6
(R,SS)
L7 (SS)
5
2 (R1 =Me,
R2 =H)
L5
>99
(S,SS)
5
6
2 (R1 =Me,
R2 =H)
L6
>99
(R,SS)
L7 (SS) >99
7
7
2 (R1 =Me,
R2 =H)
9
L8
(S,SS)
L9
(R,SS)
L10
(SS)
L11
(S,SS)
L12
(R,SS)
8
2 (R1 =Me,
R2 =H)
L8
(S,SS)
L9
(R,SS)
L10
(SS)
L11
(S,SS)
L12
>99
>99
>99
>99
>99
8
9
2 (R1 =Me,
R2 =H)
10
12
11
10
11
12
2 (R1 =Me,
R2 =H)
2 (R1 =Me,
R2 =H)
2 (R1 =Me,
R2 =H)
7(À)
(R,SS)
[a]
[a]
[b]
Conditions: 1 (1 mmol), CuTC (1 mol%), and L* (1.1
mol%) in CH2Cl2 (2 mL) at À788C with addition of
EtMgBr in Et2O(1.2 equiv.) over 1 h.
Conditions: same as Table 1.
Yield of isolated products after purification by column
chromatography on silica gel, (in parentheses, conversion
determined by GC-MS).
[b]
Yield of isolated products after purification by column
chromatography on silica gel, (in parentheses, conversion
determined by GC-MS).
[c]
[d]
1
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on
chiral stationary phase.
[c]
[d]
1
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on
chiral stationary phase.
our group, which afforded excellent enantioselectivi-
ties on less substituted cinnamyl-type substrates, gave
our best result for the formation of 4a with a poor
35% ee and 77% branched product (Table 3, entry 5).
Such a process with ethylmagnesium bromide and cat-
alytic CuTC/L5 being obviously not optimal on the
trisubstituted olefin 2, we additionally attempted the
in situ generation of an ethylzinc reagent (Et2Zn or
EtZnBr) – starting from the corresponding Grignard
reagent and zinc dibromide (0.5 or 1 equivalent) in
THF[13] – using ligand (S,SS)-L5. These last conditions
afforded excellent regioselectivities (branched-to-
linear ratio>99:1), alas as a racemic mixture.
Table 2. Variation of the catalyst loading [CuTC/(S,SS)-L2]
for the addition of ethylmagnesium bromide (1.2 equiv.) to
b-methylcinnamyl chloride (1).
A
Entry Substrate Cat. loading
[mol%]
Conv.
[%]
SN2’/ ee
SN2
[%]
1
2
3
1
1
1
1
2
3
>99
>99
>99(87) 92/8
87/13 96(+)
90/10 96(+)
98(+)
to the same panel of phosphoramidite ligands used Aryl b-Disubstituted Allylic Chlorides
above. As summarized in Table 3, using our devel-
oped methodology with organomagnesium reagents With the set of optimal conditions for b-disubstituted
under CuTC catalysis, only moderate regioselectivities aryl allylic chlorides at hand, we investigated differ-
and poor ee values were obtained. The bis-ortho-me- ently substituted cinnamyl substrates, with electron-
thoxyphosphoramidite ligand (S,SS)-L5 developed by richer or poorer aryl scaffolds and with higher substi-
1092
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1090 – 1100

b- and g-Disubstituted Olefins
FULL PAPERS
Scheme 4. Stereoselective SN2’ alkylation on allylic chlorides 1–8.
tution at the b-position (Scheme 4). Results for the 92% ee (Table 4, entry 4). At last, the 3-(chloro-
addition of ethylmagnesium bromide catalyzed with methyl)-1,2-dihydronaphthalene (8) presented a note-
CuTC (3 mol%) and chiral ligand (S,SS)-L2 are col- worthy advantage to the cinnamyl derivatives, as its
lected in Table 4. Overall results are comparable to structure is configurationally locked as a trans isomer.
the model reaction on 1 (Table 4, entry 1) with very This allowed for a highly asymmetric allylic substitu-
good enantioselectivities for either para-tolyl (5) or tion with an ethylmagnesium reagent reaching up to
para-chlorophenyl (6) substrates, both affording 96% 98% ee for 12a. Conversely, the lower flexibility of
ee (Table 4, entries 2 and 3). Owing to the higher the bicyclic allylic electrophiles or its electronically
steric hindrance generated by the b-ethyl group on richer nature afforded a g-selectivity close to the ones
compound 7, we observed a slight decrease in regio- observed for 9 and 11 with a branched-to-liner ratio
and enantioselectivities with 83% g-selectivity and of 83/17.
Table 4. Enantioselective Cu-catalyzed allylic alkylation on 1–8 with EtMgBr (R2 =Et; Scheme 4).
Entry[a]
Substrate
Product
Yield [%][b]
SN2’/SN2[c]
ee [%][d]
98 (+)
1
87
92:8
2
3
4
85
87
83
87
84:16
92:8
96 (+)
96 (+)
92 (+)
98 (+)
83:17
83:17
7
[a]
Conditions: same as Table 1.
[b]
[c]
[d]
Yield of isolated products after purification by column chromatography on silica gel.
1
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on chiral stationary phase.
Adv. Synth. Catal. 2008, 350, 1090 – 1100
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1093

Caroline A. Falciola et al.
FULL PAPERS
With these encouraging results at hand, other linear um reagents with the CuTC/L2 catalytic mixture to
Grignard reagents were tested, as illustrated in give products 13–18. The lower branched-to-linear
Table 5. Very good enantioselectivities of up to 98% ratios of 79/21 and 82/18 obtained, respectively, with
ee were achieved in the allylic substitution of sub- both but-3-enylmagnesium bromide (Table 5, entry 3)
strates 1 and 8 using linear saturated organomagnesi- and pent-4-enyl bromide (Table 5, entry 5) were im-
Table 5. Enantioselective allylic alkylation of 1 and 8 with R2MgX (Scheme 4).
Entry
Substrate
R2MgX
Product
Yield [%][a]
b/l[b]
ee [%][c]
97 (+)
1
1
n-propylMgCl
85
84/16
2
1
n-hexylMgCl
83
84
83/17
96 (+)
3
1
1
but-3-enylMgBr
but-3-enylMgBr
79/21
89/11
95 (+)
97 (+)
4[d]
15a
16a
A
5
1
1
pent-4-enylMgBr
pent-4-enylMgBr
87
82/18
87/13
93 (+)
96 (+)
6[d]
(>99)
7
8
n-butylMgCl
62
62
71/29
98 (+)[16]
8
8
PhCH2CH2MgBr
78/22
96.5 (+)
[a]
Yield of isolated products after purification by column chromatography on silica gel, (in parentheses, conversion deter-
mined by GC-MS).
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on chiral stationary phase.
Slow addition of the Grignard reagent over a period of four hours.
Ligand (R,SS)-L3 was used.
[b]
[c]
[d]
[e]
1
1094
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1090 – 1100

b- and g-Disubstituted Olefins
FULL PAPERS
proved by a slower addition of the reagent over a ing groups, they obtained SN2’ rearrangement prod-
period of four hours (to 89 and 87% g-selectivity for ucts with low-to-high enantioselectivities (27–90%
15a and 16a, respectively). This also afforded a slight ee). The stoichiometric organocopper reagent pre-
increase in the enantiomeric excess. However such a pared in situ stood out as the choice reagent for a
methodology could not be transposed to secondary highly selective g-substitution (>92% SN2). Gratify-
organomagnesium reagents. Indeed, when using iso- ingly, the addition of different Grignard reagents pro-
propylmagnesium chloride on 1 in the presence of moted by 3 mol% of the copper catalyst proceeded
ligand (S,SS)-L2 – although regioselectivities were in with high selectivity towards the g-substitution prod-
favor of the branched product (g/a 86/14) – the ee uct and delivered adducts 23a–f and 24a–f with excel-
dropped below 28% ee. The diastereomeric ligand lent enantiomeric excess of up to >99% (Scheme 6).
(R,SS)-L3, on the other hand, yielded 52% ee.[15] Both
Although, in our first substitution assays, a short
n-butylmagnesium and phenethylmagnesium halide ligand screening was done for the addition of n-butyl-
reagents would yield excellent enantiomeric excess on magnesium chloride to the five-membered ring sub-
the bicyclic substrate 8, affording 98% ee and 97% ee strate 21 (Table 6, entries 1–3), the best ligand was
for 17a and 18a, respectively (Table 5, entries 7 and found to be (S,SS)-L2 which, as mentioned earlier, af-
8).
forded excellent enantioselectivities on the b-methyl-
Additionally, as described previously by our cinnamyl substrates. We were pleased to observe that
group,[17] chiral compounds 15a and 16a, generated by the chiral adduct (+)-23a was obtained with 98% ee
the addition of the corresponding Grignard reagent and 96% g-selectivity (Table 6, entry 2). In regard to
on 1, could undergo a one-pot ring-closing metathesis the five-membered ring substrate 21, this asymmetric
(19–20) upon treatment with the Grubbs second gen- methodology constantly provided very high selectivi-
eration catalyst in a completely stereoretentive fash- ties of 97–98% ee throughout the array of linear
ion (Scheme 5).[18]
Grignard reagents tested, displayed in Table 6. This
copper-catalyzed process could be used on a prepara-
tive synthesis for 23b on a one gram-scale, allowing a
slower addition time of the n-hexylmagnesium re-
agent (Table 6, entry 5). The latter compound was ob-
Aliphatic b-Disubstituted Allylic Halides
To broaden the scope of the copper-catalyzed proce- tained with equally good stereoselectivities and was
dure employing the phosphoramidite ligand (S,SS)-L2 the starting point for the asymmetric synthesis of In-
and to circumvent the problem of E:Z isomeric ratio dolizidine frameworks.[20] Other useful transforma-
present in the cinnamyl substrates, we looked at ali- tions can be envisioned on the exocyclic double bond
phatic endocyclic allylic chlorides. These allylic elec- (such as oxidation to a ketone or selective epoxida-
trophiles were again configurationally locked as trans tion)[21] or deprotection of the tert-butoxide in 23c–
isomers and looked like a very interesting class of b- 24c. Moreover, the absolute configuration of the exo-
disubstituted allylic substrates in comparison to the methylenic adducts 23a–f and 24a–f was ascertained
results disclosed some years ago by Gais and co-work- by comparison with literature data.[19,22]
ers.[19] Indeed, this group had described the enantiose-
The six-membered ring allylic chloride 22 afforded
lective addition of various organocuprate and organo- equally good ee values and g selectivities whatever
copper reagents to optically active endocyclic allylic the Grignard reagents employed. Once again, a small
sulfoximines. By using these chiral auxiliaries as leav- array of phosphoramidite ligands – L2/L3, L5/L6 and
Scheme 5. One-pot Cu-catalyzed enantioselective SN2’ reaction and ring-closing metathesis.
Adv. Synth. Catal. 2008, 350, 1090 – 1100
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1095

Caroline A. Falciola et al.
FULL PAPERS
Scheme 6.
Table 6. Asymmetric CuTC-catalyzed allylic alkylation on endocyclic 1-(chloromethyl)cyclopent-1-ene 21 with RMgX
(Scheme 6; L*=L1–L6).
Entry
Substrate
RMgX
L*
Product
Conv. [%][a]
g/a[b]
ee [%][c]
1
2
3
21
21
21
21
21
21
21
21
n-butylMgCl
n-butylMgCl
n-butylMgCl
n-hexylMgBr
n-hexylMgBr
L2
L3
ent-L6
L2
L2
L2
L2
L2
23a
23a
23a
23b
23b
23c
23d
23f
>99 (10)
>99
>99
>99 (44)
>99 (91)
96 (60)
>99
96/4
63/37
91/9
97/3
98/2
98/2
97/3
98/2
98 (+,S)
29 (+,S)
29 (À,R)
98 (+)
4
5[d]
6
98 (+)
t-BuO
A
97 (+)[e]
98 (+)
7
8
PhCH2CH2MgCl
4-methylpent-3-enylMgBr
>99
95 (+)
[a]
Conversion determined by GC-MS (in parentheses, yield of isolated products after purification by column chromatogra-
phy on silica gel).
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on chiral stationary phase.
Reaction with 4 mmol of material; addition of RMgX over 4 h.
Determined by GC analysis of 23g [R=(CH2)4OCOCF3].
[b]
[c]
[d]
[e]
1
L12/L13 – were tested for their asymmetric induction the reaction. Indeed, if one considers the six-mem-
of the allylic substitution reaction of the six-mem- bered ring n-butyl adduct (24a) and the closely relat-
bered ring chloride 22. Nevertheless, from analysis of ed 3-butenyl product (24e), the optical purity is signif-
Table 7 it becomes clear that the previously estab- icantly increased from 97% ee to 99.2% ee (Table 7,
lished ligand (S,SS)-L2 was once again the most suc- entries 1 and 11). Furthermore, other reagents, con-
cessful throughout the different magnesium reagents taining a similar unsaturation at related sites, fostered
used. Remarkable and unprecedented enantioselectiv- even better asymmetric outcome. Indeed, phenethyl-
ities up to >99% ee are obtained for compounds magnesium reagent and the terpenic 4-methylpent-3-
24a–f in the Cu-catalyzed asymmetric SN2’ alkylation enyl Grignard reagent afforded each 99.4% ee and
on the six-membered ring chloride 22 (Table 7). The 99.6% ee for 24d and 24f, respectively (Table 7, en-
highest enantioselectivity for such a reaction was re- tries 10 and 12), the highest enantioselectivities re-
corded for the addition of 4-methylpent-3-enylmagne- corded for copper-catalyzed allylic alkylation and
sium bromide to 22 with 3 mol% CuTC/L2 loading, more importantly on aliphatic substrates (Figure 2).
to afford 24f with 99.6% ee and a 98:2 branched-to-
linear ratio (Table 7, entry 12).
The reactivity of the seven-membered ring sub-
strate was subsequently studied to ascertain whether
It is conceivable that a certain degree of p-p stack- it exhibited the same selectivity trends as the six- and
ing or p-cation interactions could be involved in the five-membered ring systems. Reaction of substrate 25
transition state and thus govern the stereocontrol of with n-butyl- and phenethylmagnesium reagent in the
1096
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1090 – 1100

b- and g-Disubstituted Olefins
FULL PAPERS
Table 7. Asymmetric CuTC-catalyzed allylic alkylation on endocyclic 1-(chloromethyl)cyclohex-1-ene 22 with RMgX
(Scheme 6; L*=L1–L4).
Entry
Substrate
R
L*
Product
Conv. [%][a]
g/a[b]
ee [%][c]
1
22
22
22
22
22
22
22
22
22
22
22
22
n-butyl
n-butyl
n-butyl
n-hexyl
L2
24a
24a
24a
24b
24c
24c
24c
24c
24d
24d
24e
24f
73
75
>99
>99 (67)
>99 (58)
94
>99
80 (42)
>99
81/19
44/66
70/30
97/3
97 (+)
2[d]
3
ent-L5
ent-L6
L2
L2
L3
L12
L13
L1
L2
L2
75 (À)
88 (+)
4
5
6
7
8
9
10
11
12
98 (+)
t-BuO
t-BuO
t-BuO
t-BuO
C
91/9
98.9 (+)
74 (À)
73/27
75/25
58/42
72/28
85/15
97/3
94 (À)
67 (+)
PhCH2CH2
PhCH2CH2
3-butenyl
4-methylpent-3-enyl
87 (À,S)
99.4 (+,R)
99.2 (+)
99.6 (+)
>99 (78)
>99 (83)
>99 (99)
L2
98/2
[a]
Conversion determined by GC-MS (in parentheses, yield of isolated products after purification by column chromatogra-
phy on silica gel).
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on chiral stationary phase.
Catalyst loading CuTC/L* 1 mol%.
[b]
[c]
[d]
1
(Table 8). In contrast to our previous results, this al-
lylic substitution afforded only moderate enantiose-
lectivities of 70% ee for 29 with 3 mol% catalyst load-
ing (Table 8, entry 1) and 74% ee when using up to 5
mol% CuTC/(S,SS)-L2 (Table 8, entry 2). We ascribed
A
the poorer chiral results to the lower flexibility of the
substrate 28 and thus ventured in further screening of
ligands for the optimization of the asymmetric induc-
tion.
Figure 2.
presence of 3 mol% of CuTC/
C
We looked into smaller biphenol-based ligands or
other substitutions in this series.
trolled allylic substitution for 29, with an enantiomer-
We then applied the aforementioned highly effi- ic excess of up to 75% ee when increasing the catalyst
cient protocol for the asymmetric allylic substitution loading to an optimal 5 mol% (Table 8, entry 4).
of b-disubstituted allylic substrates to chloride 28 Under similar conditions, products 30 and 31 were ob-
(Scheme 8). Our initial trial was conducted by a slow tained with 71% ee and 84% ee, respectively, albeit
addition of phenethylmagnesium bromide catalyzed with a moderate g-selectivity (Table 8, entries 6 and
by copper(I) thiophenecarboxylate and (S,SS)-L2 8).
Scheme 7. Cu-catalyzed asymmetric allylic substitution of seven-membered ring 25.
Adv. Synth. Catal. 2008, 350, 1090 – 1100
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1097

Caroline A. Falciola et al.
FULL PAPERS
Scheme 8. SN2’-allylic reaction on rigid endocyclic allylic chloride (28).
Table 8. Enantioselective Cu-catalyzed alkylation of 28.
Entry
R
L*
Cu/L* [mol%]
Product
Conv. [%][a]
SN2’/SN2[b]
ee [%][c]
1
2
3
4
5
6
7
8
PhCH2CH2-
PhCH2CH2-
PhCH2CH2-
PhCH2CH2-
n-butyl-
U
3
5
3
5
5
5
5
5
29
29
29
29
30
30
31
31
72 (30)
65
86
>98
82
85
70/30
78/22
86/14
90/10
31/69
57/43
47/53
70/30
72 (+)
74 (+)
74 (+)
75 (+)
55
71
65 (+)
84 (+)
n-butyl-
t-BuO
t-BuO
U
>99
>99 (62)
[a]
1
Conversion determined by H NMR (in parentheses, yield of isolated products after purification by column chromatogra-
phy on silica gel).
[b]
[c]
1
Ratio determined by H NMR spectroscopy and GC-MS.
Enantiomeric excess determined by GC analysis on chiral stationary phase.
Unsymmetrical 1,4-Dibromo-2-methylbut-2-ene
to 86% ee. Under these conditons, the minor quater-
nary coumpound (34) (4%) would only afford 26%
Having achieved promising results in the allylic subsi- ee. Converserly, using the bis-ortho-methoxyphosphor-
tution on 1,4-bishalo-2-butene, (up to 94% ee on 1,4- amidite (R,RR)-L5, up to 20% of the quaternary
dibromo-2-butene),[24] and moreover high asymmetric chiral homo-allylic bromide was formed with 69% ee.
induction on the more substituted endocyclic allylic Although this last result was recorded on the minor
chlorides (21) (>99% ee), we envisioned to combine product 34, this is to our knowledge the best enantio-
these two types of allylic electrophiles in the com- selectivity reported for the copper-catalyzed allylic
pound 1,4-dibromo-2-methylbut-2-ene (32). We antici- substitution using Grignard reagents (Scheme 9).
pated that such a challenging unsymmetrical allylic
electrophile would react regioselectively towards the
incoming nucleophile depending on the selected Conclusions
ligand (Scheme 9). Indeed, when subjecting 32 to n-
butylmagnesium chloride in the presence of ligand In conclusion, we have described the first method for
(S,SS)-L2, the substitution took place preferentially at the copper-catalyzed enantioselective alkylation of b-
the g-position (tertiary-to-quaternary ratio 96/4), disubstituted allylic chlorides (bromides) with organo-
acting as if the compound was a pure b-disubtituted magnesium reagents through the use of a simple
substrate. This produced the chiral adduct 33 with up chiral phosphoramidite ligand (S,SS)-L2 and promo-
Scheme 9. Unsymmetrical substrate 32 used in Cu-catalyzed asymmetric allylic alkylation (AAA).
1098
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1090 – 1100

b- and g-Disubstituted Olefins
FULL PAPERS
tion by copper thiophenecarboxylate (CuTC). This Supporting Information
identified process could be further applied to a varie-
Preparation procedures of starting materials and spectral
analysis of different chiral products are available in the sup-
porting information.
ty of substrates with equally high ranges of stereose-
lectivities (up to >99% ee), with a single exception
on highly rigid frameworks where another ligand was
preferred. Moreover, compound 23b is a precursor for
the asymmetric synthesis of the indolizidine back-
bone, and was thus prepared on a gram-scale. We
have also illustrated that fine tuning of the ligand can
promote regioselectivity upon an unsymmetrical allyl-
ic dibromide 32, which in one case has started to
enable interesting enantioselectivities for the forma-
tion of chiral quaternary centers, moreover on a
highly functionalized substrate.
Acknowledgements
The authors thank the Swiss National Research Foundation
(no. 20–068095.02), and COST action D24/0003/01 (OFES
contract no. C02.0027) for financial support, and BASF for a
generous gift of chiral amines.
References
Experimental Section
[1] a) B. M. Trost, C. Lee, in: Catalytic Asymmetric Synthe-
sis (Ed.: I. Ojima), 2nd edn., Wiley, NewYork, 2000,
p 593; b) A. Pfaltz, M. Lautens, in: Comprehensive
Asymmetric Catalysis, Vols. I–III, (Eds.: E. N. Jacobsen,
A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999, p 833.
[2] For reviews of asymmetric allylic alkylation with vari-
ous metals, see: a) H. Miyabe, Y. Takemoto, Synlett
2005, 1641–1655; b) B. M. Trost, J. Org. Chem. 2004,
69, 5813–5837; c) B. M. Trost, M. L. Crawley, Chem.
Rev. 2003, 103, 2921–2943; d) R. Takeuchi, Synlett
2002, 1954–1965.
[3] For recent reviews of Cu-catalyzed AAA reactions,
see: a) C. A. Falciola, A. Alexakis, Eur. J. Org. Chem.
2008, ASAP; b) A. Alexakis, C. Malan, L. Lea, K.
Tissot-Croset, D. Polet, C. Falciola, Chimia 2006, 60,
124–130; c) H. Yorimitsu, K. Oshima, Angew. Chem.
2005, 117, 4509–4513; Angew. Chem. Int. Ed. 2005, 44,
4435–4439; d) A. Kar, N. P. Argade, Synthesis 2005,
2995–3022; e) S. A. E. Karlstrçm, J.-E. Bꢁckvall, in:
Modern Organocopper Chemistry, (Ed.: N. Krause),
Wiley-VCH, Weinheim, 2002.
[4] a) M. van Klaveren, E. S. M. Persson, D. M. Grove, J.-
E. Bꢁckvall, G. van Koten, Tetrahedron Lett. 1994, 35,
5931–5934; b) A. Alexakis, C. Malan, L. Lea, C. Ben-
haim, X. Fournioux, Synlett 2001, 927–930; c) F. Lopez,
A. W. Van Zijl, A. J. Minnaard, B. L. Feringa, Chem.
Commun. 2006, 409–411.
[5] a) F. Dubner, P. Knochel, Angew. Chem. 1999, 111,
391–393; Angew. Chem. Int. Ed. 1999, 38, 379–381;
b) F. Dubner, P. Knochel, Tetrahedron Lett. 2000, 41,
9233–9237.
[6] Allylic substitution on g-disubstituted allylic substrates,
with peptidic Schiff base: a) C. A. Luchaco-Cullis, H.
Mizutani, K. E. Murphy, A. H. Hoveyda, Angew.
Chem. 2001, 113, 1504–1508; Angew. Chem. Int. Ed.
2001, 40, 1456–1460; with N-heterocyclic diaminocar-
benes: b) A. O. Larsen, W. Leu, C. N. Oberhuber, J. E.
Campbell, A. H. Hoveyda, J. Am. Chem. Soc. 2004,
126, 11130–11131.
[7] a) D. G. Gillingham, A. H. Hoveyda, Angew. Chem.
2007, 119, 3934–3938; Angew. Chem. Int. Ed. 2007, 46,
3860–3864; b) Y. Lee, K. Akiyama, D. G. Gillingham,
M. K. Brown, A. H. Hoveyda, J. Am. Chem. Soc. 2008,
130, 446–447.
Typical Procedure for the Enantioselective Copper
Catalyzed Allylic Substitution with Grignard
Reagents
CuTC (1 mol%) and chiral ligand (1.1 mol%) were charged
in a dried Schlenk tube, under inert gas, and suspended in
dichloromethane (2 mL). The mixture was stirred at room
temperature for 30 min, followed by the addition of the al-
lylic halide (1 mmol) at room temperature before cooling
the mixture to À788C in an ethanol-dry ice cold bath. The
Grignard reagent (3M in diethyl ether, 1.2 equiv.) diluted in
CH2Cl2 (0.6 mL) was added over 60 min via a syringe pump.
Upon completion of the addition, the reaction mixture was
left a further 4 h at À788C. The reaction was quenched by
addition of aqueous HCl (1N, 2 mL) and then Et2O
(10 mL). The aqueous phase was separated and further ex-
tracted with Et2O(3”3 mL). The combined organic frac-
tions were washed with brine (5 mL), dried over anhydrous
magnesium sulfate, filtered and concentrated under vacuum.
The oily residue was purified by flash column chromatogra-
phy. Gas chromatography on a chiral stationary phase gave
the enantiomeric excess of the SN2’ product.
(+)-(R)-1-Methylene-2-(4-methylpent-3-enyl)cyclohexane
(24f): yield: 99%; SiO2, pentane, RF =0.95; IR (neat): n=
3065 (w), 2962 (w), 2925 (s), 2855 (m), 1784 (w), 1645 (m),
1445 (s), 1376 (m), 1107 (w), 981 (w), 888 (s), 832 (m), 629
(w) cmÀ1 1H NMR (400 MHz, CDCl3): d=5.12 (t, J=
;
7.0 Hz, 1H), 4.65 (s, 1H), 4.56 (s, 1H), 2.26–2.20 (m, 1H),
2.05–1.94 (m, 4H), 1.79–1.72 (m, 1H), 1.69 (s, 3H), 1.63–
1.20 (m, 10H); 13C NMR (100 MHz, CDCl3): d=153.2,
131.5, 125.0, 105.6, 42.8, 35.0, 34.0, 32.4, 29.0, 26.0, 25.9, 24.4,
17.8; MS (EI mode): m/z (%)=178 (9), 135 (31), 109 (13),
95 (12), 93 (12), 83 (14), 82 (100), 81 (17), 79 (12), 69 (15),
67 (36), 55 (24), 41 (38); HR-MS (EI mode): m/z=178.1722,
calcd. for C13H22: 178.1719; [a]2D2: +31.30 (c 1.28, CHCl3) for
99.56% ee. The ee was measured by chiral GC with a Chira-
sil Dex CB, helium flow (program: 70–0–1–170–5) RT: 38.80
(À), 38.93 (+).
Adv. Synth. Catal. 2008, 350, 1090 – 1100
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1099

Caroline A. Falciola et al.
FULL PAPERS
[8] a) C. Borner, P. J. Goldsmith, S. Woodward, J. Gimeno,
S. Gladiali, D. Ramazzotti, Chem. Commun. 2000,
2433–2434; b) P. J. Goldsmith, S. J. Teat, S. Woodward,
Angew. Chem. 2005, 117, 2275–2277; Angew. Chem.
Int. Ed. 2005, 44, 2235–2237.
[9] C. A. Falciola, K. Tissot-Croset, A. Alexakis, Angew.
Chem. 2006, 118, 6141–6144; Angew. Chem. Int. Ed.
2006, 45, 5995–5998.
[10] a) A. Alexakis, S. Rosset, J. Allamand, S. March, F.
Guillen, C. Benhaim, Synlett 2001, 1375–1378; b) A.
Alexakis, D. Polet, S. Rosset, S. March, J. Org. Chem.
2004, 69, 5660–5667.
[11] a) B. L. Feringa, Acc. Chem. Res. 2000, 33, 346–353;
b) B. L. Feringa, M. Pineschi, L. A. Arnold, R. Imbos,
A. H. M. De Vries, Angew. Chem. 1997, 109, 2733;
Angew. Chem. Int. Ed. Engl. 1997, 36, 2620.
[12] a) K. Tissot-Croset, D. Polet, A. Alexakis, Angew.
Chem. 2004, 116, 2480–2482; Angew. Chem. Int. Ed.
2004, 43, 2426–2428; b) K. Tissot-Croset, D. Polet, S.
Gille, C. Hawner, A. Alexakis, Synthesis 2004, 2586–
2590; c) D. Polet, A. Alexakis, K. Tissot-Croset, C. Cor-
minboeuf, K. Ditrich, Chem. Eur. J. 2006, 12, 3596–
3609; d) K. Tissot-Croset, A. Alexakis, Tetrahedron
Lett. 2004, 45, 7375–7378.
[16] M. Scommoda, H.-J. Gais, S. Bosshammer, G. Raabe, J.
Org. Chem. 1996, 61, 4379–4390.
[17] A. Alexakis, K. Croset, Org. Lett. 2002, 4, 4147–4149.
[18] The first generation Grubbs catalyst gave poor conver-
sions (<50%).
[19] a) J. Bund, H. J. Gais, I. Erdelmeier, J. Am. Chem. Soc.
1991, 113, 1442; b) H.-J. Gais, H. Mueller, J. Bund, M.
Scommoda, J. Brandt, G. Raabe, J. Am. Chem. Soc.
1995, 117, 2453.
[20] a) P. Schꢁr, P. Renaud, Org. Lett. 2006, 8, 1569; b) E.
Nyfeler, P. Renaud, Abstracts of papers, SCS Fall Meet-
ing 2005, EPFL, Lausanne, Switzerland, October 13,
2005, ORGN-278; c) E. Nyfeler, Ph. D. Thesis, Univer-
sity of Berne, 2005; d) E. Nyfeler, P. Schꢁr, P. Renaud
unpublished results.
[21] P. G. Reddy, S. Baskaran, J. Org. Chem. 2004, 69, 3093–
3101.
[22] Moreover, compound (+)-23b (98% ee) was oxidized
by ozonolysis into the (S)-2-hexylcyclopentanone,
which chiroptical data confirmed the absolute configu-
ration. [a]2D2: + 84.11 (c 0.23, CHCl3); Lit. [a]22: À110
(c 0.7, CHCl3) for (R)-2-hexylcyclopentanoneD; M. M.
Kayser, G. Chen, J. D. Stewart, J. Org. Chem. 1998, 63,
7103–7106.
[23] L. Palais, I. S. Mikhel, C. Bournaud, L. Micouin, C. A.
Falciola, M. Vuagnoux-d’Augustin, S. Rosset, G. Ber-
nardinelli, A. Alexakis, Angew. Chem. 2007, 119, 7606–
7609; Angew. Chem. Int. Ed. 2007, 46, 7462–7465.
[24] C. A. Falciola, A. Alexakis, Angew. Chem. 2007, 119,
2673–2676; Angew. Chem. Int. Ed. 2007, 46, 2619–
2622.
[13] A. Alexakis, S. El Hajjaji, D. Polet, X. Rathgeb, Org.
Lett. 2007, 9, 3393–3395.
[14] J.-E. Bꢁckvall, M. Sellꢂn, B. Grant, J. Am. Chem. Soc.
1990, 112, 6615.
[15] Ligands L3–L12 were also tested but afforded lower ee
values.
1100
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1090 – 1100