Domino catalysis: Palladium-catalyzed carbonylation of allylic alcohols to β,γ-unsaturated esters

Article abstract of DOI:10.1002/anie.201303850

Make it green: The Pd-catalyzed carbonylation of allylic alcohols with aliphatic alcohols represents a cost-effective, salt-free, and atom-economic carbonylation process generating water as the sole by-product. Mechanistic studies uncovered a new domino C-O coupling/carbonylation reaction pathway. Copyright

Full text of DOI:10.1002/anie.201303850

Angewandte
Chemie
Communications
Carbonylation
Q. Liu, L. Wu, H. Jiao, X. Fang, R. Jackstell,
M. Beller*
&&&& — &&&&
Domino Catalysis: Palladium-Catalyzed
Carbonylation of Allylic Alcohols to b,g-
Unsaturated Esters
Make it green: The Pd-catalyzed carbon-
ylation of allylic alcohols with aliphatic
alcohols represents a cost-effective, salt-
free, and atom-economic carbonylation
process generating water as the sole by-
product. Mechanistic studies uncovered
an new domino CÀO coupling/carbon-
ylation reaction pathway.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
DOI: 10.1002/anie.201303850
Carbonylation
Domino Catalysis: Palladium-Catalyzed Carbonylation of Allylic
Alcohols to b,g-Unsaturated Esters**
Qiang Liu, Lipeng Wu, Haijun Jiao, Xianjie Fang, Ralf Jackstell, and Matthias Beller*
Palladium-catalyzed carbonylations of organic (pseudo)ha-
lides are of broad interest for both academic and industrial
p-allylpalladium acetates, which readily undergo reductive
elimination to give the starting acetates rather than CO
[1]
[8]
research. In the last two decades, the majority of the work in
this area focused on the application of new ligand systems and
the extension of the range of nucleophiles for the carbon-
ylation of aryl and vinyl halides (or related pseudohalides).
Consequently, the development of general catalytic protocols
for more challenging substrates remains an important but
challenging goal. Based on our continuous interest in
insertion.
An ideal way to streamline carbonylation of allylic
compounds from an economic and environmental point of
view is to use of allylic alcohols directly as substrates.
Alcohols are more widely available and represent more
environmentally benign reagents, generating water as the sole
by-product. Advantageously, the whole synthetic route is
shortened because most of the above-mentioned substrates
are obtained from the corresponding alcohols. Unfortunately,
the poor leaving ability of the hydroxy group, combined with
the possible side reactions caused by the released water, have
[
1b]
transition-metal-catalyzed reactions with CO, recently we
became attracted by carbonylation of allylic compounds,
which represents a straightforward and economic method for
the synthesis of versatile building blocks, b,g-unsaturated
[
2]
carbonyl compounds. In spite of the tremendous progress in
hindered the application of allylic alcohols in carbonylation
[
3]
[3d]
nucleophilic allylic substitution reactions, carbonylation of
allylic compounds has received much less attention. Most of
the reported examples demand preinstalled leaving groups,
reactions.
As a result, only very few examples were
reported in this field and these protocols generally suffer
from limitations, such as harsh reaction conditions, low yields,
high metal catalyst and ligand loadings, and/or a limited
[
4]
[5]
[6]
such as chlorides,
phates.
carbonates,
acetates,
or phos-
[
6b,e,7]
[9]
Obviously, this creates inherent problems such
substrate scope. In fact, to the best of our knowledge, only
as significant waste generation and requires less efficient
multistep sequences. Moreover, side reactions were observed
for carbonylation of allylic halides owing to the accumulation
of hydrogen halides in this reaction. However, by adding base
to quench the acid, direct reaction with allylic halides is
carbonylations of allylic alcohols with specific nucleophlic
[9g]
substrates, for example, thiols
and one example with
[9f,10]
phenol, are known.
In line with our previous work on the direct amination of
[
11]
allylic alcohols,
we turned our attention to respective
[
5a]
possible. In general, the resulting b,g-unsaturated carbonyl
carbonylation processes. Herein, we present the first general
and efficient catalyst system for the direct carbonylation of
allylic alcohols with a variety of aliphatic alcohols. Our
protocol proceeds through domino catalysis and is atom
economic and environmentally benign (salt free), and pro-
ceeds with excellent linear/branched regioselectivity [Eq. (1);
Ad = adamantyl, TFA = trifluoroacetic acid, Xantphos = 4,5-
bis(diphenylphosphino)-9,9-dimethylxanthene].
compounds are highly susceptible to base-catalyzed isomer-
[4e]
ization, affording a,b-unsaturated carbonyl isomers. In the
well-established carbonylation of allyl alkyl carbonates it is
difficult to introduce various nucleophiles owing to the
preferred reaction of the alkoxides from the substrates.
Besides, allylic carbonates are usually prepared from allylic
alcohols and toxic chloroformates. Thus, the overall process is
also not really halide-free. Notably, carbonylation reactions of
allylic acetates, which are commonly used as electrophiles in
Pd-catalyzed allylic substitution reactions, are much less
effective. For example, high CO pressure or halide additives
are needed. This inefficiency is explained by the formation of
We started our investigation by examining the Pd-
catalyzed carbonylation of cinnamyl alcohol 1a with n-
butanol 2a (see the Supporting Information, Scheme S1 and
Tables S1 and S2). Representative catalyst precursors, phos-
phine ligands, acidic additives, and solvents were studied to
improve the efficiency of this transformation. An optimal
yield (determined by GC) of 75% for product 3aa was
[
*] Dr. Q. Liu, L. Wu, Dr. H. Jiao, X. Fang, Dr. R. Jackstell,
Prof. Dr. M. Beller
Leibniz-Institut fꢀr Katalyse an der Universitꢁt Rostock
Albert-Einstein-Str. 29a, 18059 Rostock (Germany)
E-mail: matthias.beller@catalysis.de
[
**] The research has been funded by the State of Mecklenburg-Western
Pomerania, the BMBF, and the DFG (Leibniz Prize). We thank Dr. W.
Baumann, Dr. C. Fischer, S. Buchholz, S. Schareina, A. Koch, and S.
Rossmeisl (all at LIKAT) for their excellent technical and analytical
support. Q.L. thanks the Alexander von Humboldt Foundation for
financial support.
obtained using 1 mol% of Pd(OAc) , 1.1 mol% of Xantphos,
2
1
0 mol% of TFA, 1.2 equivalents of n-butanol 2a, with
toluene as the solvent at 1058C under 40 bar CO pressure.
The reaction yield was improved to 85% when the reaction
scale was increased from 1 mmol to 10 mmol with only
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201303850.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
0
.5 mol% Pd catalyst loading (Scheme 1). To the best our
knowledge, this is the lowest palladium catalyst loading for
any carbonylation of allylic compounds reported so far.
Scheme 1. Pd-catalyzed carbonylation of cinnamyl alcohol 1a with n-
butanol 2a.
The progress of the carbonylation of cinnamyl alcohol 1a
with n-butanol 2a was examined under the optimized reaction
conditions (Figure 1). Surprisingly, GC analysis of samples
taken from the reaction mixture showed that 1a was initially
converted into the CÀO coupling product 4a within 15 min.
Scheme 2. Investigation of stepwise CÀO coupling and carbonylation
reactions.
Then, the carbonylative product 3aa is generated at lower
rate along with the consumption of the reaction intermediate
4
a.
The presence of the catalytic amount of TFA is essential
to achieve both steps of this domino reaction process. To
assess the role of TFA, 1a and 4a were mixed with TFA in
[
D ]toluene at room temperature, and NMR studies were
8
performed (Figure 2). The formation of trifluoroacetate 5a
from 1a was observed after 6 h. In contrast, 4a was not
converted into 5a upon treatment with TFA under the same
conditions.
Figure 1. Domino CÀO coupling/carbonylation reaction process.
Figure 2. Influence of TFA in the reactions of 1a and 4a.
To gain further insights into this interesting new domino
reaction, stepwise reactions were investigated separately
(
Scheme 2). In the absence of CO gas, the CÀO coupling
We suspected that the carbonylation step should proceed
via typical p-allyl/Pd intermediates because the same product
3aa was obtained when 1-phenylprop-2-en-1-ol 1m, the
isomer of 1a, was employed as the starting material under
the same reaction conditions [Eq. (2)].
reaction proceeded smoothly and produced 4a in 89% yield
after 15 min. Meanwhile, carbonylation of 4a could also be
accomplished under the standard conditions without addi-
tional 2a. Both CÀO coupling and carbonylation steps do not
proceed without the presence of either Pd catalyst or TFA
(
Scheme 2a and 2b). Additionally, the reaction of a 1:1
mixture of 4a and (E)-4-chlorocinnamyl alcohol 1b yielded
a mixture of carbonylated products 3aa and 3ba (Scheme 2c).
These results clearly reveal that the carbonylation of cin-
namyl alcohol 1a with n-butanol 2a occurred through
a sequential CÀO coupling and carbonylation process and
On the basis of our experimental findings, a domino
palladium-catalyzed sequence is proposed for the formation
the reaction rate of the carbonylation step is much slower.
[
12]
of 3aa (Scheme 3).
It is known that Pd(OAc)2 can be
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
observed under the noncatalytic
condition, and this reveals once
again the kinetic influence.
With the optimized reaction
conditions
established,
we
examined the generality and
limitation of this novel domino
process with respect to aliphatic
alcohols (Scheme 5). A variety
of substituted primary aliphatic
alcohols gave the corresponding
carbonylative products in syn-
thetically useful yields with
excellent regioselectivity. Nota-
bly, the branched products were
never observed in any transfor-
Scheme 3. Proposed mechanism.
0
reduced in situ to Pd species 6 in the presence of phosphine
[13]
ligands and CO. Oxidative addition of trifluoroacetate 5 or
0
the protonated allylic alcohol 1 with Pd complex 6 gives the
p-allyl/Pd intermediate 7. This intermediate undergoes a fast
nucleophilic substitution by the aliphatic alcohol 2 at the less
sterically hindered terminal position to afford the CÀO
0
coupling product 4 and regenerates Pd species 6. Subse-
quently, oxidative addition of protonated allylic ether 4 leads
to p-allyl/Pd complex 7, followed by CO coordination and
insertion to give the acyl Pd complex 8. Finally, alcoholysis of
intermediate 8 generates the desired carbonylative product 3
and Pd hydride species 6’, which is in an equilibrium with the
0
Pd complex 6 under acidic condition.
To understand the CÀO coupling process in more detail
(
Scheme 2a and Figure 2), DFT investigations (see the
Scheme 5. Carbonylation of 1a with a variety of aliphatic alcohols.
Supporting Information for details) were carried out. In our
calculations, we used real-size substrates, but methanol
instead of n-butanol as model. Both the reaction free
energy of ether formation and exchange were calculated.
Scheme 4 shows clearly that ether formation (4a’) is favored
Reaction conditions: 1a (1 mmol), 2 (1.2 mmol), Pd(OAc) (1 mol%),
2
Xantphos (1.1 mol%), TFA (10 mol%) in toluene (2 mL) under 40 bar
CO at 1058C. Yields of the isolated products are shown.
mation of the shown substrates. A broad range of functional
groups were tolerated, including reactive chloro (3af), alkene
(3ag), benzyl (3ai), and tetrahydrofuran groups (3aj), which
provide useful handles for further synthetic transformations.
Notably, secondary and tertiary alcohols underwent this
transformation smoothly in moderate to good yields (3ak
and 3al).
Next, we evaluated the scope of allylic alcohols by using 2-
phenylethanol 2e as a standard coupling partner (Table 1).
Both 3-arylprop-2-en-1-ols (Table 1, entries 1–4) and 3-alkyl-
prop-2-en-1-ols (Table 1, entries 5–12) furnished the corre-
Scheme 4. BP86 free energies for the substitution reactions.
À1
(
DG = À2.90 kcalmol ) thermodynamically. Hence, it
sponding products in good yields with Xantphos or PAd nBu
2
reveals that the transformation of 1a into 4a’ under the
same reaction conditions without either Pd catalyst or TFA is
kinetically hindered. This conclusion is supported by the fast
formation of 4a under the catalytic conditions (Figure 1). The
noncatalytic formation of 5a’ is feasible, because the forma-
as ligands, respectively. a,b-Unsaturated esters were not
detected in these reactions despite the facile isomerization of
[4e]
b,g-unsaturated esters to a,b-unsaturated esters. We dem-
onstrated the utility of our carbonylation protocol in the
reaction of 2,7-octadienol 1k, a bulky feedstock and geraniol
1l, an ingredient commonly used in perfumes and flavors
(Table 1, entries 11 and 12). The loss of the stereochemistry of
the double bonds can be explained by syn/anti isomerization
À1
tion of 5a’ is also favored (DG = À2.85 kcalmol ) thermo-
dynamically. According to the calculated Gibbs free energy
À1
(
DG = 0.06 kcalmol ), the exchange of 4a’ and 5a’ will be in
[7]
fair equilibrium, but such equilibrium (4a and 5a’) was not
of the p-allyl Pd intermediate.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[
a]
Table 1: Pd-catalyzed carbonylation of allylic alcohols with 2e.
In conclusion, we developed a general and benign
carbonylation method of allylic alcohols with aliphatic
alcohols to produce a variety of synthetically useful b,g-
unsaturated esters. Interestingly, this reaction proceeds
through a novel sequential C-O coupling/carbonylation path-
way. In view of the easy availability of the substrates, the
efficiency, the excellent regioselectivity, and the salt-free
conditions, this method is expected to complement the
current methods for carbonylations in organic synthesis.
Experimental Section
General procedure for 1 mmol scale reactions: A 25 mL Schlenk flask
was charged with Pd(OAc) (13.4 mg, 1 mol%), Xantphos (38.1 mg,
2
1
2
.1 mol%) or PAd nBu (85.9 mg, 4 mol%), and toluene (12 mL).
mL of this clear yellow stock solution was transferred into each of
2
the six vials (4 mL reaction volume) equipped with a stirring bar
under the protection of argon. Then the allylic alcohol (1 mmol), the
aliphatic alcohol (1.2 mmol), and TFA (8 mL, 10 mol%) were added
into each vial sequentially. The vials were placed in an alloy plate,
which was transferred to a 300 mL autoclave under argon atmos-
phere. After flushing the autoclave three times with N , the pressure
2
was adjusted to 40 bar of CO at ambient temperature and the
mixtures were stirred for 22 h at 1058C. The reaction was cooled to
room temperature and then the pressure was released. Evaporation of
the solvent gave a residue that was purified by silica gel column
chromatography using n-heptane and ethyl acetate as eluent.
Received: May 5, 2013
Published online: && &&, &&&&
Keywords: allylic alcohols · carbonylation · CÀO coupling ·
.
domino reactions · palladium
[
1] a) M. Beller, Applied Homogeneous Catalysis with Organome-
tallic Compounds, Vol. 1, 2nd ed., Wiley-VCH, Weinheim, 2002,
pp. 145 – 156; b) A. Brennfꢀhrer, H. Neumann, M. Beller,
Angew. Chem. 2009, 121, 4176 – 4196; Angew. Chem. Int. Ed.
2
009, 48, 4114 – 4133; c) X.-F. Wu, H. Neumann, M. Beller,
[
a] Reaction conditions: 1 (1 mmol), 2 (1.2 mmol), Pd(OAc) (1 mol%),
2
Chem. Soc. Rev. 2011, 40, 4986 – 5009; d) X.-F. Wu, H. Neumann,
M. Beller, Chem. Rev. 2013, 113, 1 – 35.
Xantphos (1.1 mol%) or PAd nBu (4 mol%), TFA (10 mol%) in toluene
(
2
2 mL) under 40 bar CO at 1058C. [b] Yields of the isolated products.
[
2] a) E. Wenkert, B. L. Buckwalter, S. S. Sathe, Synth. Commun.
1973, 3, 261 – 264; b) K. N. Houk, Chem. Rev. 1976, 76, 1 – 74;
c) D. Armesto, M. J. Ortiz, A. R. Agarrabeitia, CRC Handbook
of Organic Photochemistry and Photobiology, 2nd ed., CRC,
Boca Raton, FL, 2004, pp. 77/71 – 77/11; d) G. B. Bajracharya,
P. S. Koranne, R. N. Nadaf, R. K. M. Gabr, K. Takenaka, S.
Takizawa, H. Sasai, Chem. Commun. 2010, 46, 9064 – 9066; e) M.
Martin-Fontecha, A. R. Agarrabeitia, M. J. Ortiz, D. Armesto,
Org. Lett. 2010, 12, 4082 – 4085.
Finally, to illustrate the synthetic potential of this method,
we performed a gram-scale synthesis of the homoallylic
alcohol 9a from the widely available allylic alcohol 1a in two
[
[
3] a) C. Bruneau, J.-L. Renaud, B. Demerseman, Chem. Eur. J.
2006, 12, 5178 – 5187; b) Z. Lu, S. Ma, Angew. Chem. 2008, 120,
264 – 303; Angew. Chem. Int. Ed. 2008, 47, 258 – 297; c) M.
Bandini, M. Tragni, Org. Biomol. Chem. 2009, 7, 1501 – 1507;
d) M. Bandini, Angew. Chem. 2011, 123, 1026 – 1027; Angew.
Chem. Int. Ed. 2011, 50, 994 – 995; e) B. Sundararaju, M. Achard,
C. Bruneau, Chem. Soc. Rev. 2012, 41, 4467 – 4483.
Scheme 6. Gram-scale synthesis of homoallylic alcohol 9a from allylic
alcohol 1a.
4] a) W. T. Dent, R. Long, G. H. Whitfield, J. Chem. Soc. 1964,
1
588 – 1594; b) J. Tsuji, J. Kiji, S. Imamura, M. Morikawa, J. Am.
steps with 80% overall yield upon isolation (Scheme 6). It
should be noted that this sequence allows for a straightforward
C1 extension of easily available allylic alcohols to the
corresponding homoallylic alcohols.
Chem. Soc. 1964, 86, 4350 – 4353; c) T. Okano, N. Okabe, J. Kiji,
Bull. Chem. Soc. Jpn. 1992, 65, 2589 – 2593; d) A. Yamamoto,
Bull. Chem. Soc. Jpn. 1995, 68, 433 – 446; e) J. Kiji, T. Okano, Y.
Higashimae, Y. Fukui, Bull. Chem. Soc. Jpn. 1996, 69, 1029 –
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
1
031; f) S. Tommasi, S. Perrone, F. Rosato, A. Salomone, L.
Jpn. 1996, 69, 1065 – 1078; f) T. Satoh, M. Ikeda, Y. Kushino, M.
Miura, M. Nomura, J. Org. Chem. 1997, 62, 2662 – 2664; g) W.-J.
Xiao, H. Alper, J. Org. Chem. 1998, 63, 7939 – 7944; h) M. J.
Payne, D. J. Cole-Hamilton, J. Chem. Soc. Dalton Trans. 1997,
3167 – 3175.
Troisi, Synthesis 2012, 423 – 430.
[
[
5] a) J. Tsuji, K. Sato, H. Okumoto, J. Org. Chem. 1984, 49, 1341 –
1344; b) T.-a. Mitsudo, N. Suzuki, T. Kondo, Y. Watanabe, J. Org.
Chem. 1994, 59, 7759 – 7765.
6] a) Y. Koyasu, H. Matsuzaka, Y. Hiroe, Y. Uchida, M. Hidai, J.
Chem. Soc. Chem. Commun. 1987, 575 – 576; b) S. Murahashi, Y.
Imada, Y. Taniguchi, S. Higashiura, Tetrahedron Lett. 1988, 29,
[10] Under the optimized conditions (3–5 mol% of Pd(OAc) , 12–
2
20 mol% of PPh ) reported in Ref. [9f] and [9g], the carbon-
3
ylation reactions of allylic alcohols with aliphatic alchohols
provide very low yields according to the results reported in
Ref. [9f] and our own experimental results. Please see the
Supporting information for details.
4945 – 4948; c) H. Matsuzaka, Y. Hiroe, M. Iwasaki, Y. Ishii, Y.
Koyasu, M. Hidai, J. Org. Chem. 1988, 53, 3832 – 3838; d) M.
Iwasaki, Y. Kobayashi, J. P. Li, H. Matsuzaka, Y. Ishii, M. Hidai,
J. Org. Chem. 1991, 56, 1922 – 1927; e) S. Murahashi, Y. Imada, Y.
Taniguchi, S. Higashiura, J. Org. Chem. 1993, 58, 1538 – 1545.
7] R. Takeuchi, Y. Akiyama, J. Organomet. Chem. 2002, 651, 137 –
[11] a) D. Banerjee, R. V. Jagadeesh, K. Junge, H. Junge, M. Beller,
Angew. Chem. 2012, 124, 11724 – 11728; Angew. Chem. Int. Ed.
2012, 51, 11556 – 11560; b) D. Banerjee, R. V. Jagadeesh, K.
Junge, H. Junge, M. Beller, ChemSusChem 2012, 5, 2039 – 2044.
[12] a) A. Ajamian, J. L. Gleason, Angew. Chem. 2004, 116, 3842 –
3848; Angew. Chem. Int. Ed. 2004, 43, 3754 – 3760; b) D. Enders,
M. R. M. Huettl, C. Grondal, G. Raabe, Nature 2006, 441, 861 –
863; c) A. Fꢀrstner, Angew. Chem. 2009, 121, 1390 – 1393;
Angew. Chem. Int. Ed. 2009, 48, 1364 – 1367.
[
[
145.
8] a) Y. Takahashi, K. Tsukiyama, S. Sakai, Y. Ishii, Tetrahedron
Lett. 1970, 11, 1913 – 1916; b) T. Yamamoto, O. Saito, A.
Yamamoto, J. Am. Chem. Soc. 1981, 103, 5600 – 5602.
[
9] a) J. F. Knifton, J. Organomet. Chem. 1980, 188, 223 – 236; b) H.
Alper, I. Amer, J. Mol. Catal. 1989, 54, L33 – L36; c) K. Itoh, N.
Hamaguchi, M. Miura, M. Nomura, J. Mol. Catal. 1992, 75, 117 –
[13] a) C. Amatore, A. Jutand, M. A. MꢁBarki, Organometallics 1992,
11, 3009 – 3013; b) C. Amatore, E. Carre, A. Jutand, M. A.
MꢁBarki, Organometallics 1995, 14, 1818 – 1826.
122; d) R. Naigre, H. Alper, J. Mol. Catal. A 1996, 111, 11 – 15;
e) M. Sakamoto, I. Shimizu, A. Yamamoto, Bull. Chem. Soc.
6
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!