Systematic Investigation into the Matsuda-Heck Reaction of α-Methylene Lactones: How Conformational Constraints Direct the β-H-Elimination Step

Article abstract of DOI:10.1021/acs.joc.6b02207

α-Methylene-γ-butyrolactone and α-methylene-δ-valerolactone undergo Pd-catalyzed Matsuda-Heck couplings with arene diazonium salts to α-benzyl butenolides or pentenolides, respectively, or to α-benzylidene lactones. The observed regioselectivity is strongly ring size dependent, with six-membered rings giving exclusively α-benzyl pentenolides, whereas the five-membered α-methylene lactone reacts to mixtures of regioisomers with a high proportion of (E)-α-benzylidene-γ-butyrolactones. DFT calculations suggest that the reasons for these differences are not thermodynamic but kinetic in nature. The relative energies of the conformers of the Pd σ-complexes resulting from insertion into the Pd-aryl bond were correlated with the dihedral angles between Pd and endo-β-H. This correlation revealed that in the case of the six-membered lactone an energetically favorable conformer adopts a nearly synperiplanar Pd/endo-β-H arrangement, whereas for the analogous Pd σ-complex of the five-membered lactone the smallest Pd/endo-β-H dihedral angle is observed for a conformer with a comparatively high potential energy. The optimized conditions for Matsuda-Heck arylations of exo-methylene lactones were eventually applied to the synthesis of the natural product anemarcoumarin A.

Full text of DOI:10.1021/acs.joc.6b02207

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Systematic investigation into the Matsuda-Heck reaction of alpha-methylene
lactones: how conformational constraints direct the beta-H-elimination step
Bernd Schmidt, Felix Wolf, and Christopher Ehlert
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.6b02207 • Publication Date (Web): 17 Oct 2016
Downloaded from http://pubs.acs.org on October 21, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
Systematic investigation into the Matsuda-Heck reaction of α-methylene lactones:
how conformational constraints direct the β-H-elimination step
7
8
9
Bernd Schmidt,* Felix Wolf and Christopher Ehlert.
Universitaet Potsdam, Institut fuer Chemie, KarlꢀLiebknechtꢀStraße 24ꢀ25, Dꢀ14476 Potsdamꢀ
Golm, Germany.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
eꢀmail: bernd.schmidt@uniꢀpotsdam.de
Table of contents graphic:
Abstract: αꢀMethyleneꢀγꢀbutyrolactone and αꢀmethyleneꢀδꢀvalerolactone undergo Pdꢀ
catalyzed MatsudaꢀHeck couplings with arene diazonium salts to αꢀbenzyl butenolides or
pentenolides, respectively, or to αꢀbenzylidene lactones. The observed regioselectivity is
strongly ring size dependent, with sixꢀmembered rings giving exclusively αꢀbenzyl
pentenolides, whereas the fiveꢀmembered αꢀmethylene lactone reacts to mixtures of
regioisomers with a high proportion of Eꢀαꢀbenzylideneꢀγꢀbutyrolactones. DFTꢀcalculations
suggest that the reasons for these differences are not thermodynamic, but kinetic in nature.
The relative energies of the conformers of the Pdꢀσꢀcomplexes resulting from insertion into
the Pdꢀaryl bond were correlated with the dihedral angles between Pd and endoꢀβꢀH. This
correlation revealed that in case of the sixꢀmembered lactone an energetically favorable
conformer adopts a nearly synperiplanar Pd/endoꢀβꢀH arrangement, whereas for the
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 48
1
2
3
4
5
6
7
8
9
analogous Pdꢀσꢀcomplex of the fiveꢀmembered lactone the smallest Pd/endoꢀβꢀH dihedral
angle is observed for a conformer with a comparatively high potential energy. The optimized
conditions for MatsudaꢀHeck arylations of exoꢀmethylene lactones were eventually applied to
the synthesis of the natural product anemarcoumarin A.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Introduction
Acrylates 1 are highly reactive benchmark olefins for evaluating novel catalyst systems or
reaction conditions for MizorokiꢀHeckꢀcouplings.1 The high πꢀacceptor and poor σꢀdonor
reactivity of acrylates leads to an electronically biased insertion of the olefin into the Pdꢀσꢀ
aryl bond of 2, which results in the formation of a Pdꢀσꢀcomplex 3 with the new CꢀCꢀbond in
βꢀposition, and a PdꢀCαꢀσꢀbond.2 The catalytic cycle proceeds with a βꢀhydride elimination,
which requires a synꢀperiplanar or nearly synꢀperiplanar arrangement of the Pd at Cα and one
βꢀhydrogen. To adopt the energetically least unfavorable synꢀperiplanar conformation 3’, a
rotation around the CβꢀCαꢀbond must occur. The stereospecificity of the βꢀHꢀelimination and
the preferred antiꢀorientation of aryl and ester substituent explain the very high Eꢀselectivity
observed for the cinnamates 5 (Scheme 1a).3 While a plethora of examples for Mizorokiꢀ
Heck arylations of simple acrylates exists, comparatively little is known about the behaviour
of conformationally constrained enoates in this transformation. One type of conformationally
restricted enoates are α,βꢀunsaturated lactones 6. If the generally accepted mechanism for
Heckꢀtype arylations as outlined in scheme 1a is applied to these substrates, a Pdꢀσꢀcomplex
7 results, which can not undergo the synꢀβꢀHꢀelimination required to close the catalytic cycle.
Nevertheless some examples for successful MizorokiꢀHeckꢀreactions of cycloalkenes,4
including α,βꢀunsaturated sixꢀmembered lactones,5 lactams,5,6 and cycloketones,7,8 have been
described in the literature. For lack of a plausible synꢀβꢀH elimination pathway, base induced
2
ACS Paragon Plus Environment

Page 3 of 48
The Journal of Organic Chemistry
1
2
3
4
5
transꢀβꢀHꢀelimination4 or epimerization7 of the Pdꢀσꢀcomplex 7 to epiꢀ7 via a Pdꢀenolate 7’
have been proposed to explain the formation of arylated products 8 (Scheme 1b).
6
7
8
Scheme 1. Heckꢀtype arylations of acrylates, unsaturated lactones and αꢀmethylene lactones.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Another, even less investigated type of conformationally restricted enoates are exoꢀmethylene
lactones 9. The Pdꢀσꢀcomplex 10 resulting from insertion into the Pdꢀσꢀaryl bond of 2 can in
principle undergo βꢀHꢀelimination with one exoꢀβꢀH and the synꢀendoꢀβꢀH, leading to αꢀ
benzylatedꢀα,βꢀunsaturated lactones endoꢀ11, or αꢀbenzylidene lactones exoꢀ11 (Scheme 1c).
For example, Genêt and coworkers reported that the Pdꢀcatalyzed coupling of an arene
diazonium salt with αꢀmethyleneꢀγꢀbutyrolactone (9a) gives exclusively the exoꢀbenzylidene
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 48
1
2
3
4
5
6
7
8
9
lactone (exoꢀ11) as a mixture of Eꢀ and Zꢀisomers.9 Shortly afterwards, Arcadi et al.
investigated the coupling of 9a with various aryl iodides and found that mixtures of endoꢀ and
exoꢀproducts were formed, and that the exoꢀproducts are exclusively Zꢀconfigured.10,11 For
more densely substituted αꢀmethyleneꢀγꢀbutyrolactones Kim et al. found that both via
MizorokiꢀHeck arylation or via oxidative CꢀHꢀactivation mixtures of endoꢀ and Eꢀexoꢀ
arylation products were formed.12 The exclusive formation of exoꢀarylation products as E/Zꢀ
mixtures was reported for guaianolideꢀtype sesquiterpene lactones.13 To date, the above
mentioned studies by Arcadi et al. remain the only systematic investigations into Mizorokiꢀ
Heck reactions of αꢀmethylene lactones.10,11 From their observations these authors concluded
that the electronic effects of paraꢀsubstituents at the aryl iodide and the nature of the base
employed in the coupling reaction steer the selectivity. Thus, particularly high endoꢀ
selectivities were observed for aryl iodides with an electron withdrawing paraꢀsubstituent and
with acetates rather than amines as a base. As a rationale for the beneficial effect of acetate
the authors proposed that Pdꢀbound acetate participates in the βꢀHꢀabstraction through a
sevenmembered transition state.10,11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Over the past few years we14ꢀ17 and others18ꢀ23 reported a considerable number of examples for
Heckꢀtype coupling reactions with arene diazonium salts. This variant of the MizorokiꢀHeck
coupling, often referred to as MatsudaꢀHeckꢀreaction, was originally discovered in the late
1970’s by Matsuda and coꢀworkers, but lay dormant for many years.24,25 Over the past 15
years MatsudaꢀHeckꢀ and other Pdꢀcatalyzed cross couplings of arene diazonium salts have
attracted renewed and still growing interest due to a number of beneficial features.26ꢀ28 For
instance, diazonium salts are conveniently synthesized from amines or acetamides29 (or can be
generated in situ,30 e. g. under flow conditions31), are highly reactive, allow for monitoring the
kinetics by quantifying the nitrogen evolution over time32 and do not require elaborate ligands
for catalyst tuning (Pd(OAc)2 or Pd2(dba)3 are the most commonly used precatalysts). Finally,
MatsudaꢀHeck reactions can be conducted both under basic and under baseꢀfree
4
ACS Paragon Plus Environment

Page 5 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
conditions,32,33 which – in light of the assumed vital role of acetate bases in Arcadi’s studies –
prompted us to revisit Heckꢀtype arylations of exoꢀmethylene lactone 9a using arene
diazonium salts. Surprisingly, the homologous sixꢀmembered exoꢀmethylene lactone 9b or
substituted derivatives of the same ring size have, to the best of our knowledge, never been
investigated in Heckꢀtype reactions. As we expected that the ring size affects the PdꢀCαꢀ
CβendoꢀHꢀdihedral angle and hence the regioselectivity of the βꢀHꢀelimination step, we
included 9b, its δꢀmethyl derivative 9c and the Nꢀbenzyl δꢀvalerolactam 9d in our study.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results and discussion
Syntheses of exo-methylene carbonyl compounds 9 and diazonium salt 13m. Exoꢀ
methylene γꢀbutyrolactone (9a) is widely commercially available from a number of suppliers.
It has, like its less conveniently accessible sixꢀmembered homologue 9b, been synthesized
from the corresponding lactone by deprotonation, Claisenꢀcondensation with ethyl formate
and addition of formaldehyde.34ꢀ36 Several 2ꢀmethyleneꢀδꢀvalerolactones, including 9c, have
been synthesized by conjugate addition of carbonyl compounds to αꢀphosphono acrylates,
subsequent lactonization and olefination with formaldehyde.37 For the synthesis of 9b and 9c
we used a modification of the former method which was adapted from a synthesis of
malyngolide38 but had not been applied to these particular derivatives (Scheme 2).
Scheme 2. Syntheses of αꢀmethyleneꢀδꢀvalerolactones 9b,c.
NꢀBenzylꢀαꢀmethyleneꢀδꢀvalerolactam (9d) was recently synthesized from δꢀvalerolactam in
three steps by aldol condensation with formaldehyde, but we decided to use a twoꢀstep
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 48
1
2
3
4
5
6
7
8
synthesis starting from ethyl nipecotate (12d). This compound was Nꢀbenzylated to N-Bn-12d
and then subjected to a baseꢀinduced αꢀmethylenelactam rearrangement39,40 to furnish 9d
(Scheme 3).
9
Scheme 3. Synthesis of αꢀmethyleneꢀδꢀvalerolactam (9d).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These syntheses can be routinely performed on a 20 mmol scale and provide reliably gram
quantities of the required exoꢀmethylene lactones and lactams 9.
The arene diazonium salts used in this study were obtained either by diazotation of anilines or
via deacetylationꢀdiazotation of acetanilides following previously published
procedures.14,29,32,41,42 The hitherto unknown phenol diazonium salt 13l was synthesized from
methylꢀ4ꢀhydroxy anthranilate (14l) through diazotation with NaNO2 in aqueous HBF4
(Scheme 4).
Scheme 4. Synthesis of the hitherto unknown diazonium salt 13l.
Matsuda Heck arylation of α-methylene-δ-valerolacton (9b). For optimization purpose we
investigated the MatsudaꢀHeck coupling of 9b and 4ꢀmethoxybenzene diazonium salt 13a.
Although other solvents, in particular water,43ꢀ45 have been tested for Heckꢀtype coupling
reactions of arene diazonium salts, methanol32,41,46 and acetonitrile,47,48 either under basic or
under baseꢀfree conditions, are still the most commonly used solvents (Table 1). Baseꢀfree
conditions in methanol (entries 1 and 2) were chosen for comparison of the catalysts.
6
ACS Paragon Plus Environment

Page 7 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Pd(OAc)2 was found to perform substantially better than Pd2dba3•CHCl3, and was therefore
used in all other experiments. By addition of NaOAc as a base (entry 3) the yield of the
MatsudaꢀHeck coupling product could be further increased to 80%. In marked contrast, the
reaction fails completely in acetonitrile in the absence of NaOAc (entry 4). This observation is
in line with previous findings by us14,32 and Correia’s mechanistic studies,47 which suggest
that acetonitrile serves as a stabilizing ligand for the Pdꢀhydride species resulting from the βꢀ
hydride elimination step. Consequently, added base should facilitate the regeneration of the
catalytically active Pd(0) species. Upon addition of NaOAc, we obtained indeed a nearly
quantitative yield of the coupling product (entry 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Optimization of MatsudaꢀHeck conditions for 9b.
entry Ratio Precatalyst (catalyst loading)
Basea)
solvent Product (yield, %)b)
9b:13a
1
2
3
4
5
6
7
8
9
2.0:1.0 Pd2(dba)3•CHCl3 (2.5 mol %) ꢀꢀ
CH3OH endoꢀ11ba (54)
CH3OH endoꢀ11ba (64)
CH3OH endoꢀ11ba (80)
CH3CN ꢀꢀc)
2.0:1.0 Pd(OAc)2 (5.0 mol %)
2.0:1.0 Pd(OAc)2 (5.0 mol %)
2.0:1.0 Pd(OAc)2 (5.0 mol%)
2.0:1.0 Pd(OAc)2 (5.0 mol %)
2.0:1.0 Pd(OAc)2 (2.5 mol %)
1.2:1.0 Pd(OAc)2 (5.0 mol %)
1.0:1.1 Pd(OAc)2 (5.0 mol %)
1.0:2.0 Pd(OAc)2 (5.0 mol %)
ꢀꢀ
NaOAc
ꢀꢀ
NaOAc
NaOAc
NaOAc
NaOAc
NaOAc
CH3CN endoꢀ11ba (98)
CH3CN endoꢀ11ba (61)
CH3CN endoꢀ11ba (50)
CH3CN endoꢀ11ba (38)
CH3CN endoꢀ11ba (71)
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 48
1
2
3
4
5
6
7
8
9
a) 3.0 equiv. b) only endoꢀ11ba was observed in all experiments. c) no conversion (TLC).
We then investigated a reduced catalyst loading (entry 6), which leads to a dramatically
diminished yield. The same was observed when the excess of alkene 9b was reduced (entry
7). Using the diazonium salt 13a with an excess of one equivalent (entry 9) led to an
improved yield compared to a 10% excess (entry 8), but application of the exoꢀmethylene
lactone 9b in excess appears to be more suitable and more convenient to ensure high yields of
the desired coupling product. Careful examination of the crude reaction mixtures revealed that
in all cases endoꢀ11ba was exclusively formed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scope and limitations of MatsudaꢀHeck arylations of 9b were explored for a number of other
arene diazonium salts 13 using 5 mol % of Pd(OAc)2 in all experiments and a ratio of
reactants of 2 : 1 (9b : 13). As solvents either methanol or acetonitrile (with and without
added base) were tested for each diazonium salt. For most diazonium salts at least one set of
conditions was identified to obtain the respective MatsudaꢀHeck products endoꢀ11b# in
synthetically useful yields and without the formation of any exoꢀisomers (Table 2). These
results show that methanol without added base are usually the optimal conditions, and that
acetonitrile with added base is only in exceptional cases superior. This includes the example
chosen for the optimization study, which underlines once again that the optimum reaction
conditions of MatsudaꢀHeck reactions are strongly substrate dependent and that
generalizations must be made with care.
A perspective application of these MatsudaꢀHeck products in stereoselective synthesis is their
conversion to 2ꢀsubstituted 2,4ꢀdienoic acids through baseꢀinduced elimination. This
transformation was originally discovered as early as 1859 for the conversion of the rowan
berry oil constituent parasorbic acid to sorbic acid.49 More recently the reaction was applied to
the total synthesis of complex natural products with conjugated diene moieties50ꢀ53 and very
recently we demonstrated that a baseꢀmediated eliminative ring opening of β,γꢀunsaturated
lactones can be incorporated in a tethered ring closing metathesis sequence.54ꢀ56
8
ACS Paragon Plus Environment

Page 9 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
Table 2. Scope and limitations of MatsudaꢀHeck couplings with 9b.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
entry
1
R1
R2
H
R3
solvent base
CH3CN NaOAc endoꢀ11ba
CH3OH ꢀꢀ endoꢀ11bb
CH3CN NaOAc endoꢀ11bc
CH3OH ꢀꢀ endoꢀ11bd
CH3OH NaOAc endoꢀ11be
Product
Yield (%)
98
13
H
OMe
OH
13a
13b
13c
13d
13e
13f
13g
13h
13i
2
H
H
74
3
H
H
OBn
OMe
OH
66
4
H
Br
Br
NO2
NO2
H
82
5
H
68
6
H
OMe
OH
CH3OH ꢀꢀ
CH3OH ꢀꢀ
CH3OH ꢀꢀ
CH3OH ꢀꢀ
CH3OH ꢀꢀ
CH3OH ꢀꢀ
CH3OH ꢀꢀ
endoꢀ11bf
endoꢀ11bg
endoꢀ11bh
endoꢀ11bi
endoꢀ11bj
endoꢀ11bk
endoꢀ11bl
82
7
H
99
8
NO2
OMe
H
48
9
H
CF3
H
ꢀꢀa)
67
10
11
12
13
H
NO2
13j
13k
13l
H
CO2Me OH
ꢀꢀa)
49
CO2Me
H
H
H
OH
NHAc
CH3CN NaOAc endoꢀ11bm 31
13m
a)No conversion in methanol or acetonitrile under either basic or baseꢀfree conditions.
We are not aware of examples for the eliminative ring opening of αꢀbenzylated pentenolides
and investigated therefore the baseꢀinduced cleavage of endoꢀ11ba. Treatment of this lactone
with KOH in DMSO at slightly elevated temperature resulted in the clean formation of the 2ꢀ
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 48
1
2
3
4
5
6
7
8
9
benzylated 2,4ꢀdienoic acid Zꢀ15ba within one hour. Assignment of the Zꢀconfiguration as
shown in scheme 5 is based on an NOEꢀinteraction between H3 and the CH2ꢀgroup of the
benzyl substituent. To check whether a stereoselective formation of the C4ꢀC5ꢀdouble bond is
also possible we synthesized the pentenolide endoꢀ11ca from αꢀmethylene lactone 9c under
the same conditions used for endoꢀ11ba (see table 1, entry 5) and subjected this compound to
the baseꢀinduced eliminative ring opening. The conjugated diene (2Z, 4E)ꢀ15ca was isolated
in a comparable yield as a single isomer. Assignment of the 2Zꢀconfiguration was again
achieved by NOE interactions between H3 and the benzylic CH2ꢀgroup, and the 4Eꢀ
configuration was proved by an NOEꢀinteraction between the terminal methyl group and H4,
as well as a 3J(H4,H5) value of 15.0 Hz (Scheme 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Baseꢀinduced eliminative ring opening of αꢀbenzyl pentenolides.
Matsuda Heck arylation of N-benzyl α-methylene-δ-valerolactam (9d). When we applied
the optimized conditions for the MatsudaꢀHeck coupling of phenoldiazonium salt 13b and αꢀ
methylene lactone 9b (baseꢀfree methanol, 2.0 : 1.0 ratio of alkene and diazonium salt) to the
coupling of this diazonium salt and αꢀmethylene lactam 9d we could isolate endoꢀ11db in
61% yield. As for the analogous lactones no exoꢀisomers could be detected in the reaction
mixture. In contrast to the results observed during the optimization study for MatsudaꢀHeck
couplings with lactone 9b, we noted an increased yield of 75% when the ratio of 9d to 13b
was reduced to 1.2 : 1.0. After testing some other ratios of reactants, we eventually found that
10
ACS Paragon Plus Environment

Page 11 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
the highest yield was obtained for a 2.0 : 1.0 ratio of diazonium salt 13 and exoꢀmethylene
lactam 9d. Scope and limitations of MatsudaꢀHeck arylations of 9d with various arene
diazonium salts 13 were next evaluated, routinely using a 1.0 : 2.0 ratio of reactants. Both
methanol and acetonitrile were tested with and without added base for each diazonium salt.
Apart from baseꢀfree acetonitrile, which failed to give a substantial yield of coupling products
in all cases, most diazonium salts could be coupled with 9d in synthetically useful yields
under at least one set of conditions (Table 3). Exceptions are the mꢀtrifluoromethyl
substituted arene diazonium salt 13i, which failed to undergo MatsudaꢀHeck couplings with
both lactone 9b and lactam 9d (entry 9) and diazonium salts 13h and 13m. Both underwent
coupling reactions with 9b in moderate yields, but not with 9d (entries 8, 13). On the other
hand, methylcarboxylate substituted diazonium salt 13k, which did not react with lactone 9b,
reacted in a satisfying yield with the lactam (entry 11). In general, the isolated yields of
MatsudaꢀHeck coupling products are lower for the lactam 9d compared to the analogous
lactone 9b. Although the arene diazonium salts were used in excess, we did not notice the
formation of symmetrical biaryls. This transformation has very recently been described when
diazonium salts were exposed to catalytic amounts of Pd(OAc)2 in ionic liquids.57
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3. Scope and limitations of MatsudaꢀHeck couplings with 9d.
entry
R1
H
R2
H
R3
OMe CH3CN
OH CH3OH
solvent
base
NaOAc
ꢀꢀ
Product
Yield (%)
13
1
2
endoꢀ11da
endoꢀ11db
72
86
13a
13b
H
H
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 48
1
2
3
H
H
OBn
OMe CH3OH
OBn CH3OH
OMe CH3OH
OH CH3OH
OMe CH3OH
CH3CN
NaOAc
endoꢀ11dc
endoꢀ11dd
endoꢀ11dn
endoꢀ11df
endoꢀ11dg
endoꢀ11dh
endoꢀ11di
endoꢀ11dj
endoꢀ11dk
65
76
66
80
96
ꢀꢀa)
ꢀꢀa)
34
66
3
4
5
6
7
8
9
13c
13d
13n
13f
13g
13h
13i
4
H
Br
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
5
H
Br
6
H
NO2
NO2
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
H
8
NO2
H
9
CF3
H
H
CH3OH
CH3OH
CH3OH
10
11
H
NO2
13j
13k
H
CO2M OH
e
12
13
CO2Me
H
H
H
OH
CH3OH
ꢀꢀ
endoꢀ11dl
65
13l
NHAc CH3CN
NaOAc
endoꢀ11bm ꢀꢀa)
13m
a)No conversion in methanol or acetonitrile under either basic or baseꢀfree conditions.
Matsuda Heck arylation of α-methylene-γ-butyrolacton (9a). In contrast to the Matsudaꢀ
Heck couplings investigated so far, all arylations of the fiveꢀmembered exoꢀmethylene lactone
9a proceed with formation of both endoꢀ and exoꢀisomers in varying ratios (Table 4). For
1
each experiment the endo : exoꢀratio was determined by HꢀNMRꢀspectroscopy of the crude
reaction mixture prior to chromatographic purification. Endoꢀ and exoꢀisomers are
conveniently distinguished by the chemical shift values for the olefinic proton and by the
presence of a triplet integrating for two protons at ca. 3 ppm for the H3ꢀprotons of the exoꢀ
isomers. A NOEꢀinteraction between this methylene group and the orthoꢀprotons of the
aromatic substituent is indicative for the assigned Eꢀconfiguration. Our configurational
assignment differs from that made by Arcadi et al. in their pioneering study,10,11 who stated
that the double bonds of their exoꢀproducts are exclusively Zꢀconfigured. However, a
comparison of the NMRꢀspectroscopical data obtained by us for the paraꢀ
12
ACS Paragon Plus Environment

Page 13 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
methoxybenzylidene lactone Eꢀexoꢀ11aa and those reported by Arcadi et al. for Zꢀexoꢀ11aa
revealed that both data sets are identical, which suggests that the original assignment of a Zꢀ
configuration might be erroneous. As mentioned in the introduction, Arcadi et al. proposed
that acetate bases steer the βꢀhydride elimination towards the endoꢀβꢀH leading to a preferred
formation of endoꢀbutenolides. We found that the MatsudaꢀHeck coupling of 9a and 13a in
acetonitrile in the presence of acetate resulted in the formation of 11aa in a 2.7 : 1.0 ratio of
endoꢀ and exoꢀproducts (entry 1). In baseꢀfree methanol the ratio is inverted to 1.0 : 6.6 (endoꢀ
11aa : exoꢀ11aa), which seems to corroborate Arcadi’s hypothesis that the base plays a
crucial role for the regioselectivity (entry 3). Replacing acetonitrile by toluene under basic
conditions (entry 2) gives 11aa in a comparable yield and very similar endo/exoꢀratio, which
suggests that the base, rather than the solvent, is the dominant selectivityꢀcontrolling factor.
Table 4. Scope and limitations of MatsudaꢀHeck couplings with 9a.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
entry
R2
H
R3
solvent
base
NaOAc
NaOAc
ꢀꢀ
ratioa) endo : exo Product Yieldb) (%)
13
1
2
3
4
5
6
7
8
OMe CH3CN
OMe toluene
OMe CH3OH
2.7 : 1.0
2.0 : 1.0
1.0 : 6.6
1.0 : 6.0
1.0 : 1.4
10.0 : 1.0
10.0 : 1.0
2.0 : 1.0
85
84
73
76
96
44
32
94
13a
13a
13a
13b
13d
13e
13e
13f
11aa
11aa
11aa
11ab
11ad
11ae
11ae
11af
H
H
H
OH
CH3OH
ꢀꢀ
Br
Br
Br
OMe CH3CN
NaOAc
NaOAc
ꢀꢀ
OH
OH
CH3OH
CH3OH
NO2 OMe CH3OH
ꢀꢀ
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 48
1
2
9
NO2 OH
CH3OH
ꢀꢀ
2.0 : 1.0
99
3
4
5
6
13g
11ag
a)Determined by H NMRꢀspectroscopy of the crude reaction mixture. Combined isolated
1
b)
7
8
yield of endoꢀ and exoꢀproducts.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We found a comparable selectivity towards the exoꢀisomer for the phenoldiazonium salt 13b
in methanol under base free conditions (entry 4). However, as pointed out by Arcadi et al., the
electronic effects of substituents at the arene moiety have a strong influence on the
regioselectivity, with electron withdrawing groups favouring endoꢀβꢀHꢀelimination.11 This
might explain why even in the absence of acetate no pronounced selectivity towards the exoꢀ
isomer was observed for the diazonium salts 13dꢀg (bearing an additional bromoꢀ or nitro
substituent), although one should not overestimate the electronic effects of these substituents
as they are located in a metaꢀposition relative to the oxidative addition site.
Arcadi et al. showed that a homogeneous product can be obtained by hydrogenation of the
crude endo/exoꢀmixtures with Pd/C.11 While Arcadi’s twoꢀstep sequence of Heck reaction and
hydrogenation requires orthogonal solvents for each step (DMF and ethyl acetate,
respectively), we seeked to develop a one pot – single catalyst sequence14,19 by exploiting the
advantages of arene diazonium salts as coupling partners. In particular, MatsudaꢀHeck
reactions can be run in the absence of bases or ligands that might have detrimental effects on a
subsequent hydrogenation step, they proceed normally at ambient temperature, which should
be beneficial for catalyst longevity, and methanol is a suitable solvent for both coupling and
hydrogenation step. After running MatsudaꢀHeck couplings of 13a and 13b with αꢀ
methyleneꢀγꢀbutyrolactone 9a in methanol under either basic or baseꢀfree conditions,
activated charcoal was added once the evolution of nitrogen had ceased, and the reaction
mixture was flushed with hydrogen and stirred under hydrogen at atmospheric pressure for 12
h. The expected αꢀbenzylated γꢀbutyrolactones 16aa and 16ab were isolated in high yields
(Scheme 6).
14
ACS Paragon Plus Environment

Page 15 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Scheme 6. One pot – single catalyst MatsudaꢀHeck coupling/hydrogenation sequence.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Analysis of the regioselectivity of the β-H-elimination step based on DFT calculations.
The strikingly different selectivities observed for MatsudaꢀHeck arylations of fiveꢀ and sixꢀ
membered exoꢀmethylene lactones might be of thermodynamic or kinetic origin.
Thermodynamic control would require a feasible “postꢀMizorokiꢀHeck”ꢀdouble bond
isomerization mechanism,58 e. g. via a sufficiently stable Pdꢀhydride species [H−Pd−X] (4,
scheme 1) originating from the βꢀHꢀelimination step. This Pdꢀhydride might be capable of
isomerizing the kinetic product into the thermodynamic one by hydropalladationꢀβꢀHꢀ
elimination steps. Destabilization of Pdꢀhydrides, e. g. by trapping stabilizing iodide ligands
with silver59 or thallium salts,60 is hence a measure to suppress unwanted postꢀMizorokiꢀHeck
double bond migration reactions. With arene diazonium tetrafluoroborates as arylating agents
strongly coordinating halide ligands are absent from the outset and the catalytically relevant
intermediates are supposedly cationic,47,61 which means that rather unstable Pdꢀhydride
species have to be expected. It is for instance in line with this assumption that no subsequent
isomerization of the double bond has been observed in MatsudaꢀHeck reactions with cyclic
enol ethers, while the analogous reactions with aryl iodides were found to be prone to postꢀ
MizorokiꢀHeck reactions.62 In light of these considerations it appears unlikely that the
different regioselectivities observed for fiveꢀ and sixꢀmembered exoꢀmethylene lactones result
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 48
1
2
3
4
5
6
7
8
9
from a Pdꢀhydride catalyzed double bond migration following the MatsudaꢀHeck reaction. To
corroborate this supposition, the relative stabilities of all possible MatsudaꢀHeck products
resulting from the reaction of phenol diazonium salt 13b with 9a and 9b, respectively, were
determined by calculating their Gibbs free energy values using DFTꢀmethods (Figure 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Calculated relative Gibbs free energies of isomers of 11ab and 11bb.
Both exoꢀZꢀisomers were found to be clearly the least stable products. In figure 1 their Gibbs
free energy is set to 0, and the free energies of the isomeric endoꢀ and exoꢀEꢀproducts are
denoted relative to exoꢀZꢀ11ab and exoꢀZꢀ11bb, respectively. For both ring sizes the order of
stability is identical: exoꢀE > endo >> exoꢀZ. This underlines that the strikingly different
product distributions observed for MatsudaꢀHeck reactions of fiveꢀ and sixꢀmembered αꢀ
methylene lactones most likely do not originate from thermodynamic, but from kinetic
control. It should be noted in this context that a single report exists which describes selective
Rhꢀcatalyzed exoꢀtoꢀendoꢀdouble bond migrations, including conversions of exoꢀEꢀ11aa and
16
ACS Paragon Plus Environment

Page 17 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
exoꢀEꢀ11ba to endoꢀ11aa and endoꢀ11ba, respectively, which apparently contradicts our
assumption. Although alternative βꢀHꢀelimination scenarios such as transꢀβꢀHꢀelimination
have been proposed in cases of conformationally constrained Pdꢀσꢀcomplexes, we assumed
that a βꢀHꢀelimination will preferably proceed via the synꢀmechanism, provided that a synꢀβꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H is available. To gain a deeper understanding of the different βꢀHꢀelimination pathways, the
cationic Pdꢀσꢀcomplexes 10 (scheme 1) resulting from the migratory insertion into the exoꢀ
double bonds of 9a and 9b, respectively, were analyzed using DFT methods. For the Pd an
oxidation state of +II and the coordination of two molecules of methanol was assumed. The
PdꢀCαꢀCβꢀAr dihedral angles were then locked to 0 ° and geometry optimizations were
performed. The structures shown in Figure 2 mirror the situation immediately after the
migratory insertion step (Figure 2).
Figure 2. Geometry optimized structures of Pdꢀσꢀcomplexes 10ab and 10bb with a
dihedral angle [PdꢀCαꢀCβexoꢀAr] = 0 °.
Starting from these eclipsedꢀconformations, which are highest in energy, the aryl moiety was
rotated stepwise around the CαꢀCβexoꢀbond between 0 ° and 360 °. Geometry optimizations
were performed for 36 conformers and the differences in energy relative to the starting
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 48
1
2
3
4
5
6
7
8
9
conformation and the dihedral angles [PdꢀCαꢀCβendoꢀH] were calculated for these 36
conformers for each Pdꢀσꢀcomplex 10ab and 10bb. The results for the fiveꢀmembered lactone
structure 10ab are shown in Figure 3. The first relevant conformation arises after rotation of
120 ° around the CαꢀCβexoꢀbond. A βꢀHexoꢀelimination from this conformer would lead to exoꢀ
Zꢀ11ab, but this conformer is as high in energy as the starting conformer. ExoꢀZꢀ11ab is
obviously not only thermodynamically (see figure 1) but also kinetically strongly disfavoured.
Rotating the aryl substituent by a further 120 ° (dihedral angle [PdꢀCαꢀCβexoꢀAr] = 240 °)
leads to another synꢀβꢀHexoꢀarrangement, which is energetically much more favourable than
the structure at 120 °. SynꢀβꢀHꢀelimination from this conformer would lead to exoꢀEꢀ11ab,
which is indeed the main product observed in this MatsudaꢀHeck reaction (see table 4). As a
measure for the probability of a synꢀβꢀHendoꢀelimination we considered the endoꢀdihedral
angle [PdꢀCαꢀCβendoꢀH], which varies from 37 ° to 44 °, depending on the exoꢀdihedral angle
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[PdꢀCαꢀCβexoꢀAr] (blue curve in figure 3). We reasoned that for a synꢀβꢀHendoꢀelimination to
proceed efficiently two conditions have to be met: the endoꢀdihedral angle needs to be
sufficiently small and the respective conformer should be energetically favourable. The
smallest endoꢀdihedral angle (36.5 °) coincides with the gaucheꢀconformation at 60 ° (exoꢀ
dihedral angle), but this conformer is – although at a local minimum – still rather high in
energy. On the other hand, the two energetically most favourable conformers at exoꢀdihedral
angles of 180 ° and 300 ° have larger endoꢀdihedral angles of ca. 39 ° (Figure 3).
18
ACS Paragon Plus Environment

Page 19 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. [PdꢀCαꢀCβendoꢀH] dihedral angle (blue curve) and relative energy (red curve) as a
function of the [PdꢀCαꢀCβexoꢀAr] dihedral angle for Pdꢀσꢀcomplex 10ab.
In the case of the sixꢀmembered lactone structure 10bb the analogous analysis revealed some
striking differences (Figure 4). Both conformers of Pdꢀσꢀcomplex 10bb with a synꢀβꢀHexoꢀ
arrangement (at 120 ° and at 240 °, which would hypothetically give exoꢀZꢀ and exoꢀEꢀ11bb,
respectively, upon βꢀHꢀelimination) are equally high in energy and only marginally more
stable than the starting conformer. In contrast, the conformer at 240 ° of the fiveꢀmembered
lactone structure 10ab, which is the precursor for the exoꢀEꢀproduct, is ca 10 kJ•mol−1 lower
in energy than the respective starting conformer. This might explain why no exoꢀEꢀproducts
were observed for MatsudaꢀHeck reactions with sixꢀmembered αꢀmethylene lactones. The
pronounced preference for the formation of endoꢀproducts can be explained by considering
the endoꢀdihedral angle (blue curve in figure 4). Compared to the fiveꢀmembered lactone
structures this crucial dihedral angle is generally smaller and an energetically favourable
conformer exists at 180 °, which nearly coincides with the smallest calculated endoꢀdihedral
angle of ca 22 °. For the two other local energy minima endoꢀdihedral angles larger than 40 °
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 48
1
2
3
4
were calculated. It appears unlikely that a synꢀβꢀHendoꢀelimination will occur from these
5
6
conformers.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. [PdꢀCαꢀCβendoꢀH] dihedral angle (blue curve) and relative energy (red curve) as a
function of the [PdꢀCαꢀCβexoꢀAr] dihedral angle for Pdꢀσꢀcomplex 10bb.
Total synthesis of anemarcumarin A. Several natural products with a 3ꢀbenzylated
coumarin pattern have been isolated from plants.63,64 Examples are sarcandracoumarin, a
weakly active cytotoxic compound isolated from the plant Sarcandra glabra,65 or
anemarcoumarin A, which has been isolated from the rhizomes of Anemarrhena
asphodeloides, a medicinal plant native in Korea, China and Japan.66 Synthetic 3ꢀbenzylated
coumarins have recently been evaluated for various medicinal chemistry purposes, e. g. as
selective CB2 agonists,67ꢀ69 as inhibitors of TNFꢀα,70 or as inhibitors of estrogen receptors
(Figure 5).71 A previously described synthetic route to these 3ꢀbenzyl coumarins proceeds
through umpolungsꢀdomino reactions of salicylic aldehydes and enals.72
20
ACS Paragon Plus Environment

Page 21 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Examples for 3ꢀbenzylated coumarins as potential target structures.
We applied the MatsudaꢀHeck conditions developed herein to the synthesis of
anemarcoumarin A from umbelliferone (17). Umbelliferone was first hydrogenated and then
protected as TBS ether 19. Application of the conditions used for exoꢀmethylenation of
lactones 12 as described in scheme 2 failed in this particular case and resulted in a complex
mixture of products. However, deprotonation of 19 with in situ generated LDA, followed by
addition of Eschenmoser’s salt73 and methyl iodide furnished the required exoꢀmethylene
coumarine 20.74 MatsudaꢀHeck arylation of 20 with phenol diazonium salt 13b led to an
inseparable mixture of products, which contained only minor amounts of the desired coupling
product. Interestingly, the MatsudaꢀHeck coupling worked satisfactorily with 21, obtained
after desilylation of 20, under otherwise identical conditions and furnished anemarcoumarin A
in six steps from umbelliferone in an overall yield of 28% (Scheme 7).
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 48
1
2
3
Scheme 7. Total synthesis of anemarcoumarin A from umbelliferone.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusions
In summary, we report conditions for the baseꢀ and ligandꢀfree Heckꢀtype arylation of αꢀ
methylene lactones with arene diazonium salts. While mostly poor levels of regioselectivity
were observed for αꢀmethyleneꢀγꢀbutyrolactone, the homologous sixꢀmembered αꢀmethylene
lactone and the analogous lactam react to the MatsudaꢀHeck products with perfect endoꢀ
selectivity. Based on DFTꢀcalculations we suggest that the high levels of endoꢀselectivity
observed for sixꢀmembered substrates result from kinetic rather than thermodynamic control.
We found that for the Pdꢀσꢀcomplex an energetically favourable conformation coincides with
a comparatively small [PdꢀCαꢀCβendoꢀH] dihedral angle, which is not the case for the fiveꢀ
membered lactone. On the other hand, a conformer with a PdꢀHβexoꢀsyn arrangement which is
lower in energy and leads to an exoꢀEꢀproduct was observed for the fiveꢀmembered lactone.
An endoꢀselective MatsudaꢀHeck arylation of an αꢀmethylene coumarin was finally applied to
the synthesis of the plant natural product anemarcoumarin A.
22
ACS Paragon Plus Environment

Page 23 of 48
The Journal of Organic Chemistry
1
2
3
Experimental Section
4
5
6
7
8
General methods. All experiments were conducted in dry reaction vessels under an
1
atmosphere of dry nitrogen. Solvents were purified by standard procedures. H NMR spectra
9
were obtained at 300 MHz in CDCl3 with CHCl3 (δ = 7.26 ppm) as an internal standard.
Coupling constants are given in Hz. 13C NMR spectra were recorded at 75 MHz in CDCl3
with CDCl3 (δ = 77.1 ppm) as an internal standard. Whenever the solubility of the sample or
signal separation were insufficient in CDCl3, it was replaced by one of the following solvents:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
DMSOꢀd6 (DMSOꢀd5 as internal standard for H NMR spectroscopy, δ = 2.50 ppm, DMSOꢀ
d6 as internal standard for 13C NMR spectroscopy, δ = 39.5 ppm); C6D6 (C6D5 as internal
1
13
standard for H NMR spectroscopy, δ = 7.16 ppm, C6D6 as internal standard for C NMR
1
spectroscopy, δ = 128.1 ppm); acetoneꢀd6 (acetoneꢀd5 as internal standard for H NMR
13
spectroscopy, δ = 2.05 ppm, CD3COCD3 as internal standard for C NMR spectroscopy, δ =
1
29.8 ppm); methanolꢀd4 (CD2HOD as internal standard for H NMR spectroscopy, δ = 3.31
ppm, CD3OD as internal standard for 13C NMR spectroscopy, δ = 49.0 ppm). IR spectra were
recorded as ATRꢀFTIR spectra. Wavenumbers (ν
) are given in cm−1. The peak intensities are
defined as strong (s), medium (m) or weak (w). Lowꢀ and high resolution mass spectra were
obtained by EIꢀTOF or ESIꢀTOF. The following arene diazonium salts were synthesized from
anilines by diazotation or from acetanilides through the deacetylationꢀdiazotation sequence,
according to previously published procedures: 13a,29 13b,14 13c,29 13d,29 13e,14 13f,29 13g,14
13h,29 13i,32 13j,42 13k,14 13m,46 13n.29
Computational methods
All DFT calculations were performed with the B3LYP density functional75ꢀ77 as it is
implemented in the ORCA78 program package. We used the def2ꢀTZVP basis set79 for all
atoms and a StuttgartꢀDresden effective core potential (SD(28, MWB))80 for palladium. To
further account for the methanol solvent effects, the conductor like screening model
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 48
1
2
3
4
5
(COSMO)81 was used. A frequency calculation was performed for the stationary points to
calculate the thermochemical properties.
6
7
8
9
3-Methylenetetrahydro-2H
-pyran-2one (9b).36 A suspension of NaH (60 wtꢀ% dispersion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
in mineral oil, 1.12 g, 28.2 mmol) in THF (100 mL) was cooled to 0 °C and a solution of 12b
(1.70 g, 17.1 mmol) and diethyl oxalate (5.10 g, 35.1 mmol) in THF (100 mL) was added
dropwise. After completed addition, ethanol (4.5 mL) was added, the reaction was warmed to
ambient temperature and stirred for 4 h. After cooling to 0 °C a solution of K2CO3 (9.60 g,
70.0 mmol) in water (14.1 mL) and formaldehyde (37 wtꢀ% aq. solution, 18.9 g) were added
and the reaction mixture was stirred for 0.25 h at this temperature. After this time the mixture
was diluted with brine (50 mL). The layers were separated and the aqueous layer was
extracted twice with diethyl ether (50 mL each). The combined organic extracts were washed
with brine, dried with MgSO4, filtered and evaporated. The residue was purified by column
chromatography on silica, using hexane/MTBE mixtures of increasing polarity as eluent, to
1
furnish 9b (1.50 g, 13.4 mmol, 78%). Colorless liquid; H NMR (300 MHz, CDCl3) δ 6.41 –
6.37 (m, 1H), 5.55 – 5.51 (m, 1H), 4.35 (t, J = 5.4 Hz, 2H), 2.68 – 2.58 (m, 2H), 1.99 – 1.84
(m, 2H); 13C NMR (75 MHz, CDCl3) δ 165.6, 134.2, 128.3, 69.8, 28.2, 23.3; IR (ATR) ν 1714
(s), 1624 (w), 1399 (m), 1294 (s), 1143 (s); HRMS (EI) calcd for C6H8O2 [M+] 112.0524,
found 112.0528.
6-Methyl-3-methylenetetrahydro-2
H
-pyran-2one (9c).37 Following the procedure for 9b,
1
12c (2.00 g, 17.1 mmol) was converted to 9c (1.50 g, 12.0 mmol, 70%). Colorless liquid; H
NMR (300 MHz, C6D6) δ 6.41 – 6.37 (m, 1H), 5.10 – 5.07 (m, 1H), 3.91 – 3.78 (m, 1H), 2.10
(dm, J = 16.3 Hz, 1H), 1.96 (ddm, J = 16.4, 12.2 Hz, 1H), 1.20 (dm, J = 13.9 Hz, 1H), 1.10 –
0.95 (m, 1H), 0.96 (d, J = 6.4 Hz, 3H); 13C NMR (75 MHz, C6D6) δ 164.4, 134.5, 126.4, 76.0,
29.8, 27.0, 21.2; IR (ATR) ν 1714 (s), 1387 (m), 1295 (s), 1170 (m), 1129 (s).
24
ACS Paragon Plus Environment

Page 25 of 48
The Journal of Organic Chemistry
1
2
3
Ethyl-1-benzylpiperdine-3-carboxylate (
N-Bn-12d). A solution of ethyl nipecotate (12d,
4
5
6
7
8
7.90 g, 50.0 mmol), NEt3 (14.10 mL, 100.0 mmol) and benzyl bromide (6.60 mL, 55.0 mmol)
in acetonitrile (30 mL) was heated to 80 °C until full conversion of the starting material. The
solvent was evaporated and aq. NaOH (1 M) was added (pH = 12). The residue was extracted
three times with CH2Cl2 (50 mL each). The combined organic extracts dried with MgSO4,
filtered and evaporated. The residue was purified by column chromatography on silica, using
hexane/MTBE mixtures as eluent, to furnish N-Bn12d (12.30 g, 50.0 mmol, quant.).
Colorless liquid; 1H NMR (300 MHz, CDCl3) δ 7.43 – 7.12 (m, 5H), 4.13 (q, J = 7.1 Hz, 2H),
3.58 (d, J = 13.2 Hz, 1H), 3.50 (d, J = 13.2 Hz, 1H), 2.97 (dm, J = 11.0 Hz, 1H), 2.74 (dm, J
= 11.3 Hz, 1H), 2.66 – 2.54 (m, 1H), 2.26 (dd, J = 10.5, 10.5 Hz, 1H), 2.07 (ddd, J = 10.8,
10.8, 2.8 Hz, 1H), 2.00 – 1.88 (m, 1H), 1.82 – 1.70 (m, 1H), 1.70 – 1.40 (m, 2H), 1.25 (t, J =
7.1 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 174.3, 138.4, 129.0, 128.2, 127.0, 63.3, 60.2, 55.4,
53.6, 41.9, 27.0, 24.6, 14.2; IR (ATR) ν 2941 (w), 1729 (s), 1189 (m), 1154 (m); HRMS (EI)
calcd for C15H21NO2 [M+] 247.1572, found 247.1576.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1-Benzyl-3-methylenepiperidin-2-one (9d).82 A solution of
N-Bn-12d (9.50 g, 38.4 mmol)
and NaOH (1.70 g, 42.5 mmol) in water (18 mL) and methanol (350 mL) was stirred at
ambient temperature for 12 h. After this time the solvent was evaporated, the residue was
suspended in toluene and evaporated again. To the dry residue was added acetanhydride (370
mL) and NEt3 (54 mL) and the reaction mixture was heated to 90 °C for 12 h. All volatiles
were evaporated and water (100 mL) and CH2Cl2 (100 mL) were added. The aqueous layer
was extracted three times with CH2Cl2 (100 mL). The combined organic extracts were washed
with brine (50 mL), dried with MgSO4, filtered and evaporated. The residue was purified by
column chromatography on silica, using hexane/MTBE mixtures of increasing polarity as
eluent, to furnish 9d (5.40 g, 26.8 mmol, 70%).Yellowish liquid; 1H NMR (300 MHz, CDCl3)
δ 7.37 – 7.21 (m, 5H), 6.28 (ddd, J = 1.8, 1.8, 1.8 Hz, 1H), 5.31 (dd, J = 1.8, 1.8, 1.8 Hz, 1H),
13
4.66 (s, 2H), 3.32 – 3.24 (m, 2H), 2.67 – 2.49 (m, 2H), 1.99 – 1.74 (m, 2H); C NMR (75
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 48
1
2
3
4
5
6
7
8
MHz, CDCl3) δ 164.4, 137.9, 137.4, 128.7, 128.2, 127.4, 122.0, 50.8, 47.9, 30.3, 23.2; IR
(ATR) ν 2932 (w), 1656 (s), 1610 (s), 1486 (m), 1451 (m), 1222 (m); HRMS (EI) calcd for
C13H15NO [M+] 201.1154, found 201.1158.
9
4-Hydroxy-2-(methoxycarbonyl)benzenediazoniumtetrafluoroborate (13l). A suspension
of 14l (1.00 g, 6.0 mmol), tetrafluoroboric acid (50 wtꢀ% in water, 1.20 mL, 9.6 mmol), water
(0.74 mL) and 2ꢀpropanol (1.00 mL) was stirred for 0.5 h at ambient temperature. The
resulting solution was then cooled to 0 °C and NaNO2 (0.82 g, 12.0 mmol) was added in small
portions. The suspension was stirred for 0.5 h at 0 °C and filtered through a Büchner funnel.
The solid was washed washed with cold water (15 mL), ethanol and then diethylether and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
dried in vacuum to give 13l (0.71 g, 2.7 mmol, 42%). Greyish solid; H NMR (300 MHz,
DMSOꢀd6) δ 8.32 (d, J = 9.4 Hz, 1H), 7.16 (d, J = 2.3 Hz, 1H), 6.90 (dd, J = 9.4, 2.3 Hz,
13
1H), 3.93 (s, 3H); C NMR (75 MHz, DMSOꢀd6) δ 174.6, 162.4, 136.7, 132.9, 123.9, 123.6,
87.4, 53.6; IR (KBrꢀdisc) ν 3320 (w), 2181 (s), 1735 (s), 1709 (s), 1437 (m), 1292 (s); HRMS
(ESI) calcd for C8H7N2O3 [M+] 179.0457, found 179.0454.
General procedure 1: Matsuda-Heck couplings of 9a,b under base-free conditions. The
appropriate arene diazonium salt 13 (0.40 mmol) and Pd(OAc)2 (4.5 mg, 5.0 mol %) were
suspended in the corresponding solvent as indicated in table 2 (4.0 mL). αꢀMethyleneꢀγꢀ
butyrolactone 9a (78 mg, 0.80 mmol) or αꢀmethyleneꢀδꢀvalerolactone 9b (90 mg, 0.80 mmol)
was added and the reaction mixture was stirred at ambient temperature for 12 h. The solvent
was evaporated and the residue was purified by column chromatography on silicia using
hexaneꢀMTBE mixtures of increasing polarity as eluent.
General procedure 2: Matsuda-Heck couplings of 9a,b under basic conditions. The
appropriate arene diazonium salt 13 (0.40 mmol), Pd(OAc)2 (4.5 mg, 5.0 mol %) and NaOAc
(98 mg, 1.20 mmol) were suspended in the corresponding solvent as indicated in table 2 (4.0
mL). αꢀMethyleneꢀγꢀbutyrolactone 9a (78 mg, 0.80 mmol) or αꢀmethyleneꢀδꢀvalerolactone 9b
26
ACS Paragon Plus Environment

Page 27 of 48
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(90 mg, 0.80 mmol) was added and the reaction mixture was stirred at ambient temperature
for 12 h. All volatiles were evaporated and the residue was partitioned between water (10 mL)
and MTBE (10 mL). The aqueous layer was separated and extracted with MTBE (3 times, 10
mL for each extraction). The combined organic extracts were dried with MgSO4, filtered and
evaporated. The residue was purified by column chromatography on silica using hexaneꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MTBE mixtures of increasing polarity as eluent.
11
3ꢀ(4ꢀMethoxybenzyl)furanꢀ2(5H)ꢀone (endoꢀ11aa
)
and (E)ꢀ3ꢀ(4ꢀmethoxybenzylidene)ꢀ
dihydrofuranꢀ2(3H)ꢀone (exoꢀ11aa).83 Following the general procedure 2 (solvent:
acetonitrile), 9a (78 mg, 0.80 mmol) and 13a (89 mg, 0.40 mmol) were converted to a 2.7 :
1
1.0 mixture (determined by H NMR analysis of the crude mixture) of endoꢀ and exoꢀ11aa
(70 mg, 0.34 mmol, 85%). Following the general procedure 1 (solvent: methanol), 9a (78 mg,
0.80 mmol) and 13a (89 mg, 0.40 mmol) were converted to a 1.0 : 6.6 mixture (determined by
1H NMR analysis of the crude mixture) of endoꢀ and exoꢀ11aa (60 mg, 0.29 mmol, 73%).
NMRꢀdata of endoꢀ11aa: 1H NMR (300 MHz, CDCl3) δ 7.13 (d, J = 8.5 Hz, 2H), 6.92 (pent.,
J = 1.8 Hz, 1H), 6.83 (d, J = 8.5 Hz, 2H), 4.72 (q, J = 1.9 Hz, 2H), 3.76 (s, 3H), 3.50 (q, J =
13
1.9 Hz, 2H); C NMR (75 MHz, CDCl3) δ 174.0, 158.5, 145.5, 134.5, 129.9, 129.4, 114.1,
1
70.3, 55.3, 31.0. NMRꢀdata of exoꢀ11aa: H NMR (300 MHz, CDCl3) δ 7.47 (t, J = 2.9 Hz,
1H), 7.42 (d, J = 8.7 Hz, 2H), 6.93 (d, J = 8.7 Hz, 2H), 4.41 (t, J = 7.3 Hz, 2H), 3.82 (s, 3H),
3.17 (td, J = 7.3, 2.9 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 172.9, 160.9, 136.3, 131.8,
127.4, 120.7, 114.4, 65.4, 55.4, 27.4; HRMS (ESI) calcd for C12H13O3 [M+H]+ 205.0865,
found 205.0865.
3ꢀ(4ꢀHydroxybenzyl)furanꢀ2(5H)ꢀone (endoꢀ11ab
)
and (E)ꢀ3ꢀ(4ꢀhydroxybenzylidene)ꢀ
dihydrofuranꢀ2(3H)ꢀone (exoꢀ11ab).83 Following the general procedure 1 (solvent: methanol),
9a (78 mg, 0.80 mmol) and 13b (83 mg, 0.40 mmol) were converted to a 1.0 : 6.0 mixture
1
(determined by H NMR analysis of the crude mixture) of endoꢀ and exoꢀ11ab (58 mg, 0.30
1
mmol, 76%). NMRꢀdata of endoꢀ11ab: H NMR (300 MHz, DMSOꢀd6) δ 9.32 (s (br.), 1H),
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 48
1
2
3
4
5
6
7
8
7.28 (pent., J = 1.6 Hz, 1H), 7.03 (d, J = 8.4 Hz, 2H), 6.69 (d, J = 8.4 Hz, 2H), 4.80 (q, J =
1.8 Hz, 2H), 3.40 (s, 2H); 13C NMR (75 MHz, DMSOꢀd6) δ 175.0, 156.0, 148.0, 132.3, 129.8,
1
128.1, 115.4, 70.7, 30.2. NMRꢀdata of exoꢀ11ab: H NMR (300 MHz, DMSOꢀd6) δ 10.14 (s
9
(br.), 1H), 7.46 (d, J = 8.6 Hz, 2H), 7.31 (t, J = 2.6 Hz, 1H), 6.87 (d, J = 8.6 Hz, 2H), 4.37 (t,
J = 7.3 Hz, 2H), 3.15 (td, J = 7.2, 2.6 Hz, 2H); 13C NMR (75 MHz, DMSOꢀd6) δ 172.6,
159.3, 135.2, 132.3, 125.8, 120.9, 116.1, 65.5, 27.0; HRMS (ESI) calcd for C11H11O3 [M+H]+
191.0708, found 191.0709.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3ꢀ(3ꢀBromoꢀ4ꢀmethoxybenzyl)furanꢀ2(5H)ꢀone (endoꢀ11ad) and (E)ꢀ3ꢀ(3ꢀbromoꢀ4ꢀmethoxyꢀ
benzylidene)dihydrofuranꢀ2(3H)ꢀone 3ꢀ(3ꢀbromoꢀ4ꢀmethoxybenzyl)furanꢀ2(5H)ꢀone (exoꢀ
11ad). Following the general procedure 2 (solvent: acetonitrile), 9a (78 mg, 0.80 mmol) and
1
13d (120 mg, 0.40 mmol) were converted to a 1.0 : 1.4 mixture (determined by H NMR
analysis of the crude mixture) of endoꢀ and exoꢀ11ad (108 mg, 0.38 mmol, 96%). NMRꢀdata
1
of endoꢀ11ad: H NMR (300 MHz, CDCl3) δ 7.43 – 7.34 (m, 1H, signal overlap with endoꢀ
isomer), 7.14 (dd, J = 8.3, 1.8 Hz, 1H), 7.00 (pent., J = 1.7 Hz, 1H), 6.83 (d, J = 8.4 Hz, 1H),
4.75 (q, J = 1.8 Hz, 2H), 3.84 (s, 3H), 3.51 – 3.47 (m, 2H); 13C NMR (75 MHz, CDCl3) δ
173.8, 154.7, 145.9, 133.7, 133.5, 130.9, 129.0, 112.1, 111.6, 70.3, 56.3, 30.5. NMRꢀdata of
exoꢀ11ad: 1H NMR (300 MHz, CDCl3) δ 7.64 (d, J = 1.8 Hz, 1H), 7.43 – 7.34 (m, 2H, signal
overlap with endoꢀisomer), 6.94 (d, J = 8.3 Hz, 1H), 4.43 (t, J = 7.2 Hz, 2H), 3.91 (s, 3H),
3.18 (td, J = 7.3, 2.9 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 172.4, 156.9, 134.6, 134.3,
131.1, 128.6, 122.3, 112.2, 112.0, 65.4, 56.4, 27.2; HRMS (ESI) calcd for C12H1279BrO3
[M+H]+ 282.9970, found 282.9957.
3ꢀ(3ꢀBromoꢀ4ꢀhydroxybenzyl)furanꢀ2(5H)ꢀone
(endoꢀ11ae
)
and
(E)ꢀ3ꢀ(3ꢀbromoꢀ4ꢀ
hydroxybenzylidene)dihydrofuranꢀ2(3H)ꢀone (exoꢀ11ae). Following the general procedure 2
(solvent: methanol), 9a (78 mg, 0.80 mmol) and 13e (115 mg, 0.40 mmol) were converted to
a 10.0 : 1.0 mixture (determined by 1H NMR analysis of the crude mixture) of endoꢀ and exoꢀ
11ae (47 mg, 0.18 mmol, 44%). Following the general procedure 1 (solvent: methanol), 9a
28
ACS Paragon Plus Environment

Page 29 of 48
The Journal of Organic Chemistry
1
2
3
(78 mg, 0.80 mmol) and 13e (115 mg, 0.40 mmol) were converted to a 10.0 : 1.0 mixture
4
5
6
1
(determined by H NMR analysis of the crude mixture) of endoꢀ and exoꢀ11ae (34 mg, 0.13
mmol, 32%). Selected 1HꢀNMRꢀdata of exoꢀ11ae: 1H NMR (300 MHz, DMSOꢀd6) δ 10.14 (s
(br.), 1H), 7.77 (d, J = 2.0 Hz, 1H), 7.48 (dd, J = 8.6, 2.0 Hz, 1H), 7.31 (t, J = 2.7 Hz, 1H),
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
4.40 (t, J = 7.2 Hz, 2H), 3.20 (td, J = 7.2, 2.7 Hz, 2H). NMRꢀdata of endoꢀ11ae: H NMR
(300 MHz, DMSOꢀd6) δ 10.09 (s, 1H), 7.38 – 7.34 (m, 2H), 7.05 (dd, J = 8.3, 2.0 Hz, 1H),
6.88 (d, J = 8.3 Hz, 1H), 4.83 (q, J = 1.7 Hz, 2H), 3.43 (s, 2H); 13C NMR (75 MHz, DMSOꢀ
d6) δ 173.7, 152.6, 148.3, 132.8, 131.6, 130.0, 129.0, 116.3, 109.1, 70.6, 29.6; HRMS (EI)
79
calcd for C11H9 BrO3 [M+] 267.9735, found 267.9728.
3ꢀ(4ꢀMethoxyꢀ3ꢀnitrobenzyl)furanꢀ2(5H)ꢀone
(endoꢀ11af
)
and
(E)ꢀ3ꢀ(4ꢀmethoxyꢀ3ꢀ
nitrobenzylidene)dihydrofuranꢀ2(3H)ꢀone (exoꢀ11af). Following the general procedure 1
(solvent: methanol), 9a (78 mg, 0.80 mmol) and 13f (108 mg, 0.40 mmol) were converted to a
1
2.0 : 1.0 mixture (determined by H NMR analysis of the crude mixture) of endoꢀ and exoꢀ
11af (93 mg, 0.38 mmol, 94%). NMRꢀdata of endoꢀ11af: 1H NMR (300 MHz, CDCl3) δ 7.74
(d, J = 1.9 Hz, 1H), 7.49 (dd, J = 8.9, 2.0 Hz, 1H), 7.09 – 7.05 (m, 2H), 4.82 (q, J = 1.7 Hz,
2H), 3.97 (s, 3H), 3.63 (s, 2H); 13C NMR (75 MHz, CDCl3) δ 173.5, 152.0, 146.0, 134.8,
125.8, 113.9, 70.3, 56.6, 30.6, other signals can not be unambigously assigned due to signal
overlap.
NMRꢀdata of exoꢀ11af: 1H NMR (300 MHz, CDCl3) δ 8.03 (d, J = 1.8 Hz, 1H), 7.69 (dd, J =
8.8, 1.9 Hz, 1H), 7.51 – 7.48 (m, 1H), 7.19 (d, J = 8.8 Hz, 1H), 4.52 (t, J = 7.2 Hz, 2H), 4.04
13
(s, 3H), 3.28 (td, J = 7.2, 2.8 Hz, 2H); C NMR (75 MHz, CDCl3) δ 65.4, 56.8, 27.2, other
signals can not be unambigously assigned due to signal overlap. HRMS (ESI) calcd for
C12H12NO5 [M+H]+ 250.0715, found 250.0702.
3ꢀ(4ꢀHydroxyꢀ3ꢀnitrobenzyl)furanꢀ2(5H)ꢀone
(endoꢀ11ag
)
and
(E)ꢀ3ꢀ(4ꢀhydroxyꢀ3ꢀ
nitrobenzylidene)dihydrofuranꢀ2(3H)ꢀone (exoꢀ11ag). Following the general procedure 1
(solvent: methanol), 9a (78 mg, 0.80 mmol) and 13g (101 mg, 0.40 mmol) were converted to
29
ACS Paragon Plus Environment