Regioselective Transition-Metal-Free Allyl–Allyl Cross-Couplings

Article abstract of DOI:10.1002/anie.201603923

Readily prepared allylic zinc halides undergo S_N2-type substitutions with allylic bromides in a 1:1 mixture of THF and DMPU providing 1,5-dienes regioselectively. The allylic zinc species reacts at the most branched end (γ-position) of the allylic system furnishing exclusively γ,α′-allyl–allyl cross-coupling products. Remarkably, the double bond stereochemistry of the allylic halide is maintained during the cross-coupling process. Also several functional groups (ester, nitrile) are tolerated. This cross-coupling of allylic zinc reagents can be extended to propargylic and benzylic halides. DFT calculations show the importance of lithium chloride in this substitution.

Full text of DOI:10.1002/anie.201603923

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201603923
German Edition: DOI: 10.1002/ange.201603923
Synthetic Methods Very Important Paper
Regioselective Transition-Metal-Free Allyl–Allyl Cross-Couplings
Mario Ellwart, Ilya S. Makarov, Florian Achrainer, Hendrik Zipse, and Paul Knochel*
Abstract: Readily prepared allylic zinc halides undergo SN2-
type substitutions with allylic bromides in a 1:1 mixture of THF
and DMPU providing 1,5-dienes regioselectively. The allylic
zinc species reacts at the most branched end (g-position) of the
allylic system furnishing exclusively g,a’-allyl–allyl cross-
coupling products. Remarkably, the double bond stereochem-
istry of the allylic halide is maintained during the cross-
coupling process. Also several functional groups (ester, nitrile)
are tolerated. This cross-coupling of allylic zinc reagents can be
extended to propargylic and benzylic halides. DFT calcula-
tions show the importance of lithium chloride in this substi-
tution.
recently, several transition-metal-catalyzed allyl–allyl cross-
couplings have been reported by the groups of Morken[9] and
others.[10] H. Yamamoto and co-workers have shown that both
a- and g-selective allyl–allyl cross-couplings can be accom-
plished using either allylic barium halides[11] or allylic
magnesium halides.[12]
Recently, we have described convenient mild preparations
of functionalized allylic zinc reagents and demonstrated their
utility in synthesis.[13] In contrast to most reactive allylic
organometallics, these allylic zinc reagents tolerate various
functional groups. Therefore, we envisioned that the allyl–
allyl cross-coupling between such functionalized allylic zinc
reagents and substituted allylic halides may provide access to
a broad range of functionalized 1,5-dienes of type 3. Herein,
we report a highly regioselective head-to-tail cross-coupling
leading to products of type g,a’-3 and tolerating sensitive
functional groups such as esters and nitriles.
In preliminary experiments, we examined the cross-
coupling of prenylzinc bromide (1a), which was generated
by the insertion of zinc dust in the presence of LiCl in THF
(1 h, 258C, 72% yield), with (E)-1-bromonon-2-ene (2a), at
various temperatures and in several solvent mixtures
(Table 1). Thus, the addition of the zinc species 1a to the
allylic bromide 2a in THF at room temperature led to
T
ransition-metal-catalyzed
cross-couplings
represent
a major tool for forming new carbon–carbon bonds.[1]
Although Pd-[2] and Ni-[3] catalyzed cross-couplings have
found numerous applications, the search for alternative
transition-metal catalysts such as Fe and Co salts[4] have
become increasingly important owing to economical and
toxicity issues. Alternatively, the performance of cross-
couplings without transition metals as reported by Hayashi,[5]
Uchiyama,[6] and others[7] opens new perspectives for sustain-
À
able C C bond formations. In this respect, allylic organo-
metallics represent a promising class of organometallic
reagents since the carbon–metal bond in these compounds
is typically highly polarized and therefore highly reactive.
Thus, the cross-coupling between 3-substituted allylic organo-
metallics of type 1 with 3-substituted allylic halides of type 2
may provide up to four regioisomeric coupling products of
type 3 (Scheme 1).
a
mixture of all four regioisomers (a,a’/a,g’/g,a’/g,g’ =
33:25:35:7; entry 1). Selectivity in favor of the a,a’-isomer
was obtained by lowering the reaction temperature to À108C
and À408C (57% and 88% of the a,a’-isomer were obtained,
respectively; entries 2 and 3). This a,a’-regioselectivity was
not further improved as we found that the addition of various
cosolvents leads to a shift to the regioisomer g,a’-3a.
In pioneering work, Y. Yamamoto and co-workers ach-
ieved a regioselective head-to-tail (g,a’) cross-coupling using
allylic boronate complexes and allylic halides.[8] More
Table 1: Optimization of the conditions for the allyl–allyl cross-coupling.
Entry T [8C] Cosolvent[b] a,a’ a,g’ g,a’ g,g’ Yield[c] [%]
1
2
3
4
5
6
7
8
9
25
À10
À40
25
25
25
25
25
25
25
none
none
none
toluene
n-hexane
1,4-dioxane
DMSO
NMP
33
57
88
4
16
5
0
0
0
10
25
27
8
5
2
10
5
5
0
15
35
10
4
89
82
81
91
95
100
71
7
6
0
2
0
4
4
0
0
4
75
65
61
94
82
Scheme 1. Cross-coupling of allylic organometallic 1 with allylic halide
2 leading to four regioisomeric coupling products of type 3.
92
100
100
100 (91)[d]
100
[*] M. Ellwart, Dr. I. S. Makarov, Dr. F. Achrainer, Prof. Dr. H. Zipse,
Prof. Dr. P. Knochel
DMPU
10
DMPU[e]
Ludwig-Maximilians-Universitꢀt Mꢁnchen, Department Chemie
Butenandtstrasse 5-13, Haus F, 81377 Mꢁnchen (Germany)
E-mail: paul.knochel@cup.uni-muenchen.de
[a] LiCl is omitted for clarity. [b] A 1:1 mixture with THF was used.
[c] Determined by GC analysis using undecane as an internal standard.
[d] Yield of isolated product. [e] DMPU was just used as an additive
(3.0 equiv with respect to the organozinc reagent).
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201603923.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 2: Allyl–allyl cross-couplings in a 1:1 mixture of THF and DMPU
after 1–3 h at 258C.
Thus, the use of a 1:1 mixture of THFand unpolar solvents
such as toluene (entry 4) and n-hexane (entry 5) furnished
selectively the cross-coupling product (g,a’-3a) in 89% and
82% selectivity, respectively. Switching the solvent to a 1:1
mixture of 1,4-dioxane and THF led to a similar result (81%
of g,a-3a) with an overall yield of 92% (entry 6). Interest-
ingly, significant regioselectivity improvements were achieved
by using DMSO, NMP, and DMPU as cosolvents (up to 100%
g,a’-selectivity; entries 7–9), allowing us to isolate the pure
coupling product (g,a’-3a) in 91% yield (entry 9 and
Scheme 2). Lowering the amount of DMPU to just 3 equiv-
Entry
Zinc reagent[a]
Electrophile[b]
Product[c]
1
2
1a
1a
2c
2d
3b: 82%
3c: 91%
3
4
1b (25, 2 h, 77)
1c (25, 2 h, 86)
2c
2c
3d: 92%
3e: 96%
Scheme 2. Transition-metal-free allyl–allyl cross-coupling leading to
g,a’-products of type 3 after 3 h at room temperature with very high
selectivity (values in parentheses represent the g,a’/a,a’/a,g’/g,g’
ratio). [a] LiCl is omitted for clarity. Yield determined by titration with
I2.
5
1d (25, 1 h, 83)
2e
3 f: 92%
6
7
1e (25, 1 h, 58)
1 f (50, 8 h, 41)
2c
2 f
3g: 90%
3h: 90%
alents led only to a decrease in selectivity (entry 10).
Furthermore, by inverting the polarity of the reagents, we
were also able to prepare selectively the a,g’-regioisomer
(a,g’-3a). Thus, instead of using prenylzinc bromide (1a), we
have used directly prenyl bromide (2b) and have replaced the
allylic bromide (2a) with the corresponding zinc reagent (1b).
Now, the cross-coupling between 2b and 1b produced only
the regioisomer a,g’-3a with 97% selectivity and 78% yield
(Scheme 2). This g,a’-selectivity was general and the sterically
hindered prenylzinc bromide (1a) reacts smoothly with the
allylic bromides 2c and 2d producing the coupling products
3b and 3c in 82–91% yield (Table 2, entries 1 and 2).[14] (E)-
Non-2-en-1-ylzinc bromide (1b) and cinnamylzinc chloride
(1c) display a similar behavior leading to 3d,e in 92–96%
yield after 3 h (entries 3 and 4). In addition, geranylzinc
bromide (1d) and nerylzinc bromide (1e) react with the
functionalized allylic bromides 2e and 2c furnishing the
branched isomers 3 f and 3g in 90–92% yield (entries 5 and
6).
8
9
1g (25, 1 h, 60)
2b
2g
3i: 83%
3j: 79%
1a
10
11
1d
2b
2h
3k: 83%[d]
3l: 90%
1h[e] (25, 1 h, 66)
[a] LiCl is omitted for clarity. In parentheses: temperature, time, yield [%]
for the insertion. Yields determined by titration with I2. [b] 0.8 equiv of
electrophile was used. [c] Yields refer to isolated, analytically pure
products. [d] 6% of the a,a’-isomer was formed. [e] 2.4 equiv of
organozinc was used.
Remarkably, the substituted allylzinc compounds 1 f[15]
and 1g[16] react smoothly with the allylic bromide 2 f and
prenyl bromide (2b) leading to the polyfunctionalized
products (3h,i) in 83–90% yield (entries 7 and 8). Similarly,
the allylic zinc reagents 1a and 1d were converted into the
corresponding products (g,a’-3j and g,a’-3k) in 79–83% yield
(entries 9 and 10). Finally, the allylzinc reagent 1h reacts
twice with (E)-1,4-dibromobut-2-ene (2h) furnishing the
symmetrical product (E)-3l as the only isomer (E/Z > 99:1)
in 90% yield after 3 h (entry 11).
indicating an SN2-type substitution. Thus, the zinc reagents 1i
and 1b react with geranyl bromide (2i) selectively to give the
corresponding (E)-1,5,9-trienes (3m,n) in 74–84% yield
(Scheme 3). In a control experiment, prenylzinc bromide
(1a) was treated with the (E)- and (Z)-allylic bromides 2j,k
providing stereoselectively the expected g,a’-regioisomers
(E)-g,a’-3o and (Z)-g,a’-3o in 76–77% yield and > 99%
retention of the double bond configuration (Scheme 3).
Remarkably, these cross-couplings proceed with retention
of the double bond configuration of the allylic bromide
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
mixture of 1-bromononane and (E)-1-bromonon-2-ene (2a)
which led to the allyl–allyl cross-coupling product g,a’-3a
exclusively.
In order to support the experimental outcome of this
selective transformation the reaction pathway was investi-
gated using double hybrid density functional theory (DFT).[17]
The addition of LiCl is currently believed to accelerate
reactions of organometallic reagents through increased for-
mation of monomeric species or ate-like complexes.[18] In line
with this rationalization we find that hetero dimers of type 8
represent the most stable structures in solution; this is
indicated by the significant exergonicity of the exchange
reaction of (LiCl)2(sol)4 with (RZnBr)2(sol)2 (DGsol(THF) =
À14.4 kJmolÀ1, DGsol(DMPU) = À27.1 kJmolÀ1) at the
B2PLYP-D3(FC)/def2-TZVPP level[19] including implicit
SMD/B3LYP/6-31G(d) solvation.[20] Indeed, 1H and
13C NMR shifts in THF obtained during the LiCl-mediated
oxidative addition of zinc to the carbon–bromine bond of
prenyl bromide (2b) correlate quite well with theoretically
calculated chemical shifts of lithium zinc dimer 8-THF at the
mPW1K/IGLO-III level of theory[21] (see the Supporting
Information). Substituted allylic organometallics often show
high diastereoselectivity in reactions with electrophiles due to
their rather ordered cyclic or acyclic transition states.[22] How
LiCl facilitates the construction of highly ordered but
unstrained transition states in substitution reactions with
prenyl zinc nucleophiles is shown in an exemplary fashion in
Figure 1 for the formation of g,a’-cross-coupling products of
type 3. Starting from mixed aggregate 8-THF, the reaction is
here assumed to involve initial exchange of THF by DMPU.
Due to the better donor ability of DMPU, as quantified by the
Gutmann donor numbers,[23] this ligand exchange is exergonic
by DGexch = À40.9 kJmolÀ1. Subsequent exchange of one of
the DMPU ligands by the substrate leads to reactant complex
PRC-I and is endergonic by DG298 = 26.0 kJmolÀ1. In the
absence of LiCl this step is significantly more costly, and
formation of product complex PRC-II is endergonic by
DG298 = 54.5 kJmolÀ1. Subsequent reaction barriers are much
lower for LiCl-containing transition state TS-I as compared to
its LiCl-free analogue TS-II, the difference amounting to
DDG° (TS-I/TS-II) = 48.7 kJmolÀ1.[24]
Scheme 3. Allyl–allyl cross-coupling leading to g,a’-products of type 3
with full retention of the double bond configuration. [a] LiCl is omitted
for clarity.
This cross-coupling was also applicable to benzylic and
propargylic halides. Thus, the reaction of prenylzinc bromide
(1a) and (E)-non-2-en-1-ylzinc bromide (1b) with propar-
gylic halides of type 4 produces 1,5-enynes of type 5
(Scheme 4). Accordingly, the cross-coupling of 1b with the
propargylic chloride 4a provides only the 1,5-enyne 5a in
88% yield.
Scheme 4. Cross-coupling of allylic zinc reagents of type 1 with prop-
argylic and benzylic halides leading to g,a’-products of type 5 and 7.
[a] LiCl is omitted for clarity. [b] 0.80 equiv of electrophile were used.
[c] Yields refer to isolated, analytically pure products. [d] The propar-
gylic chloride was used instead of the bromide. [e] 0.40 equiv of
electrophile was used.
The main structural difference between these transition
states concerns the trajectory angle of the backside attack:
while the ideal angle of 1808 is nearly reached in TS-I (1638),
this is not so in TS-II (1348). This indicates that geometrical
factors as well as electronic interactions play an important
role in decreasing the kinetic barrier. Replacement of the
allylic bromide by n-butyl bromide leads to barriers signifi-
cantly higher in energy (DDG° (TS-I) = 29.1 kJmolÀ1, DDG°
(TS-II) = 23.9 kJmolÀ1), which is in agreement with the
results of the competition experiment of 1-bromononane
and (E)-1-bromonon-2-ene (2a). This behavior can be
rationalized by inspection of the HOMO and LUMO levels
which shows a larger energy difference for the respective
aliphatic bromide and an increased substrate deformation
energy in the respective transition state.[25] In conclusion, the
presence of LiCl seems essential for fast and selective cross-
couplings. In fact, complex mixtures of products are obtained
in the absence of LiCl.
The reaction of prenylzinc bromide (1a) with 7-bromo-
hept-5-ynoate (4b) furnishes the corresponding functional-
ized 1,5-enyne 5b in 79% yield. Furthermore, the coupling of
zinc reagent 1a with 1,4-dibromobut-2-yne (4c) selectively
affords the symmetrical product 5c in 85% yield. Interest-
ingly, benzyl bromides (6a,b) reacted under our standard
reaction conditions furnishing the substitution products 7a
and 7b in 63–72% yield. However, prenylzinc bromide did
not react with 1-bromononane. In addition, in a control
experiment, prenylzinc bromide (1a) was added to a 1:1
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 1. Reaction pathway of the g,a’-cross-coupling in THF solution (B2PLYP-D3(FC)/def2-TZVPP+DGsolv(SMD/B3LYP/6-31G(d)), in kJmolÀ1
)
and graphical representation of the B3LYP/631SVP optimized transition states TS-I (] C-Ccrot-Br=163.08) and TS-II (] C-Ccrot-Br=133.98; bond
lengths in pm).
[4] a) T. Tsuji, H. Yorimitsu, K. Oshima, Angew. Chem. Int. Ed.
In summary, we have demonstrated that allylic zinc
reagents undergo highly regioselective cross-couplings with
allylic bromides via an SN2-substitution fashion in a 1:1
mixture of THF and DMPU. Furthermore, unsymmetrical
2002, 41, 4137; Angew. Chem. 2002, 114, 4311; b) T. Nagano, T.
Hayashi, Org. Lett. 2004, 6, 1297; c) R. Martin, A. Fꢀrstner,
Angew. Chem. Int. Ed. 2004, 43, 3955; Angew. Chem. 2004, 116,
4045.
allylic zinc reagents undergo this cross-coupling almost
exclusively from the most branched side of the allylic
system. The stereochemistry of the double bond of the allylic
bromide is maintained during the cross-coupling. This SN2
reaction can be extended to propargylic and benzylic
bromides. Further applications are currently underway in
our laboratories.
[5] E. Shirakawa, Y. Hayashi, K. Itoh, R. Watabe, N. Uchiyama, W.
Konagaya, S. Masui, T. Hayashi, Angew. Chem. Int. Ed. 2012, 51,
218; Angew. Chem. 2012, 124, 222.
[6] a) H. Minami, X. Wang, C. Wang, M. Uchiyama, Eur. J. Org.
Chem. 2013, 7891; b) H. Minami, T. Saito, C. Wang, M.
Uchiyama, Angew. Chem. Int. Ed. 2015, 54, 4665; Angew.
Chem. 2015, 127, 4748.
[7] a) F. Ding, R. William, F. Wang, X.-W. Liu, Chem. Commun.
2012, 48, 8709; b) N. Uchiyama, E. Shirakawa, T. Hayashi, Chem.
Commun. 2013, 49, 364; c) Q. Chen, X. Mollat du Jourdin, P.
Knochel, J. Am. Chem. Soc. 2013, 135, 4958; d) Q. Chen, T.
Leꢁn, P. Knochel, Angew. Chem. Int. Ed. 2014, 53, 8746; Angew.
Chem. 2014, 126, 8891; e) P. Quinio, D. Sustac Roman, T. Leꢁn,
S. William, K. Karaghiosoff, P. Knochel, Org. Lett. 2015, 17, 4396.
[8] Y. Yamamoto, K. Maruyama, J. Am. Chem. Soc. 1978, 100, 6282.
[9] a) L. A. Brozek, M. J. Ardolino, J. P. Morken, J. Am. Chem. Soc.
2011, 133, 16778; b) H. Le, A. Batten, J. P. Morken, Org. Lett.
2014, 16, 2096; c) M. J. Ardolino, J. P. Morken, J. Am. Chem. Soc.
2014, 136, 7092; d) M. J. Ardolino, J. P. Morken, Tetrahedron
2015, 71, 6409.
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft (SFB 749,
B2) for financial support.
Keywords: 1,5-dienes · allylic compounds · cross-coupling ·
SN2 reaction · zinc
[10] For transition-metal-catalyzed allyl–allyl cross-couplings, see
a) K. Sekiya, E. Nakamura, Tetrahedron Lett. 1988, 29, 5155;
b) E. F. Flegeau, U. Schneider, S. Kobayashi, Chem. Eur. J. 2009,
15, 12247; c) V. Hornillos, M. Pꢂrez, M. FaÇanꢃs-Mastral, B. L.
Feringa, J. Am. Chem. Soc. 2013, 135, 2140; d) Q. Yuan, K. Yao,
D. Liu, W. Zhang, Chem. Commun. 2015, 51, 11834; e) K.
Semba, N. Bessho, T. Fujihara, J. Terao, Y. Tsuji, Angew. Chem.
Int. Ed. 2014, 53, 9007; Angew. Chem. 2014, 126, 9153; For
transition-metal-free allyl – allyl cross-couplings, see f) C. Prꢂ-
vost, P. Miginiac, L. Miginiac-Groizeleau, Bull. Soc. Chim. Fr.
[1] a) Metal-Catalyzed Cross-Coupling Reactions (Eds.: A. de Mei-
jere, F. Diederich), Wiley-VCH, Weinheim, 2004; b) Organo-
transition Metal Chemistry (Ed.: J. F. Hartwig), University
Science Books, Sausalito, 2010.
[2] Palladium Reagents and Catalysts: New Perspectives for the 21st
Century (Ed.: J. Tsuji), Wiley, Weinheim, 2004.
[3] R. Jana, T. P. Pathak, M. S. Sigman, Chem. Rev. 2011, 111, 1417.
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
1964, 2485; g) Y. Yamamoto, H. Yatagai, K. Maruyama, J. Chem.
Mulvey, Angew. Chem. Int. Ed. 2011, 50, 6448; Angew. Chem.
Soc. Chem. Commun. 1979, 157; h) Y. Yamamoto, H. Yatagai, K.
Maruyama, J. Am. Chem. Soc. 1981, 103, 1969; i) A. Hosomi, T.
Imai, M. Endo, H. Sakurai, J. Organomet. Chem. 1985, 285, 95;
j) V. Calo, L. Lopez, G. Pesce, J. Chem. Soc. Perkin Trans. 1 1988,
1301; k) S. Araki, T. Shimizu, S.-J. Jin, Y. Butsugan, J. Chem. Soc.
Chem. Commun. 1991, 824; l) S. Araki, S.-J. Jin, Y. Butsugan, J.
Chem. Soc. Perkin Trans. 1 1995, 549.
2011, 123, 6576; d) B. Lecachey, H. Oulyadi, P. Lameiras, A.
Harrison-Marchand, H. Gꢂrard, J. Maddaluno, J. Org. Chem.
2010, 75, 5976; e) D. Stern, N. Finkelmeier, D. Stalke, Chem.
Commun. 2011, 47, 2113; f) A. C. Hoepker, L. Gupta, M. F.
Faggin, D. B. Collum, J. Am. Chem. Soc. 2011, 133, 7135; g) J. E.
Fleckenstein, K. Koszinowski, Organometallics 2011, 30, 5018.
[19] a) L. Goerigk, S. Grimme, J. Chem. Theory Comput. 2011, 7, 291;
b) F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7,
3297; c) F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057.
[20] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B
2009, 113, 6378. DGsol (THF) includes implicit solvation in THF,
DGsol (DMPU) in a DMF continuum.
[11] a) A. Yanagisawa, H. Hibino, S. Habaue, Y. Hisada, K. Yasue, H.
Yamamoto, Bull. Chem. Soc. Jpn. 1995, 68, 1263; b) A.
Yanagisawa, K. Yasue, H. Yamamoto, Synlett 1996, 842.
[12] A. Yanagisawa, H. Hibino, N. Nomura, H. Yamamoto, J. Am.
Chem. Soc. 1993, 115, 5879.
[13] a) H. Ren, G. Dunet, P. Mayer, P. Knochel, J. Am. Chem. Soc.
2007, 129, 5376; b) M. D. Helm, P. Mayer, P. Knochel, Chem.
Commun. 2008, 1916; c) C. Sꢄmann, P. Knochel, Synthesis 2013,
1870; d) M. Ellwart, P. Knochel, Angew. Chem. Int. Ed. 2015, 54,
10662; Angew. Chem. 2015, 127, 10808.
[14] The use of allylic chlorides led to significantly lower yields in
several cases.
[15] J. van der Louw, J. L. van der Baan, F. Bickelhaupt, G. W.
Klumpp, Tetrahedron Lett. 1987, 28, 2889.
[21] P. Quinio, C. FranÅois, A. E. Escribano Cuesta, A. K. Steib, F.
Achrainer, H. Zipse, K. Karaghiosoff, P. Knochel, Org. Lett.
2015, 17, 1010.
[22] M. Yus, J. C. Gonzꢃlez-Gꢁmez, F. Foubelo, Chem. Rev. 2013,
113, 5595.
[23] DN (THF) = 20 kcalmolÀ1, DN (NMP) = 27.3 kcalmolÀ1, DN
(N,N-dimethylacetamide) = 27.8 kcalmolÀ1 taken from V. Gut-
mann, Coord. Chem. Rev. 1976, 18, 225.
[24] The selectivity increase observed by the use of nonpolar solvents
(entries 4 and 5 in Table 1) can be tentatively explained by the
higher Lewis acidity of the zinc cation in these solvents favoring
the TS-I.
[25] I. Fernꢃndez, F. M. Bickelhaupt, Chem. Soc. Rev. 2014, 43, 4953.
DE (TS-I, crotyl bromide) =+ 106.4 kJmolÀ1, DE (TS-I, butyl
bromide) =+ 127.9 kJmolÀ1 at the B3LYP/631SVP level in the
gas phase.
[16] N. El Alami, C. Belaud, J. Villiꢂras, J. Organomet. Chem. 1988,
348, 1.
[17] Geometries of the corresponding organozinc intermediates were
obtained at the B3LYP/631SVP level, followed by single-point
calculations at B2PLYP-D3(FC)/def2-TZVPP level. Solvent
effects were included using Truhlarꢅs solvent model density
(SMD) in combination with B3LYP/6-31G(d) calculations and
THF or DMF as the solvent.
[18] a) K. Koszinowski, P. Bçhrer, Organometallics 2009, 28, 771;
b) G. T. Achonduh, N. Hadei, C. Valente, S. Avola, C. J. OꢅBrien,
M. G. Organ, Chem. Commun. 2010, 46, 4109; c) E. Hevia, R. E.
Received: April 22, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Synthetic Methods
One of four possible products: Cross-
coupling reactions of allylic zinc halides
with allylic bromides provide 1,5-dienes
with high regioselectivity. Various func-
tional groups (ester, nitrile) are tolerated.
Propargylic and benzylic halides can be
used in this transformation. DFT calcu-
lations confirm the importance of LiCl for
the efficient cross-coupling.
M. Ellwart, I. S. Makarov, F. Achrainer,
H. Zipse, P. Knochel*
&&&& — &&&&
Regioselective Transition-Metal-Free
Allyl–Allyl Cross-Couplings
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!