Investigation of the origins of regiochemical control in [4+2] cycloadditions of 2-pyrones and alkynylboronates

Article abstract of DOI:10.1055/s-0031-1291142

The [4+2] cycloaddition of 2-pyrones with substituted alkynylboronates has been studied. In general, the highest yielding cycloadditions were obtained in reactions that employed a trimethylsilyl-substituted alkynylboronate. The highest regioselectivities were obtained using the corresponding phenyl-substituted alkyne, which provided a single regioisomer irrespective of the 2-pyrone used. Mechanistic studies suggest that the high regioselectivity observed is due to stabilization of a zwitterionic transition state. Georg Thieme Verlag Stuttgart · New York.

Full text of DOI:10.1055/s-0031-1291142

1964▌
FEATURE ARTICLE
fIenatuvreeasrtticilge ation of the Origins of Regiochemical Control in [4+2] Cycloadditions
of 2-Pyrones and Alkynylboronates
[4+2] Cycloadditions of 2-Pyrones and Alkynylboronates
James D. Kirkham,a Andrew G. Leach,b Eleanor C. Row,c Joseph P. A. Harrity*a
a
Department of Chemistry, University of Sheffield, Sheffield, S3 7HF, UK
Fax +44(114)2229346; E-mail: j.harrity@sheffield.ac.uk
b
AstraZeneca, Alderley Park, Macclesfield, Cheshire, SK10 4TG, UK
c
Department of Isotope Chemistry and Metabolite Synthesis, sanofi-aventis, Willowburn Avenue, Alnwick, NE66 2JH, UK
Received: 23.03.2012; Accepted after revision: 10.04.2012
react with both electron-poor and -rich dienophiles,6
Abstract: The [4+2] cycloaddition of 2-pyrones with substituted
whilst Kočevar employed 3-aminopyranones in reactions
alkynylboronates has been studied. In general, the highest yielding
with alkynes bearing a range of electronically and sterical-
cycloadditions were obtained in reactions that employed a trimeth-
ylsilyl-substituted alkynylboronate. The highest regioselectivities
ly distinct groups.7 The latter report also proposed that the
were obtained using the corresponding phenyl-substituted alkyne, reaction regioselectivity may be based on the formation of
which provided a single regioisomer irrespective of the 2-pyrone
used. Mechanistic studies suggest that the high regioselectivity ob-
served is due to stabilization of a zwitterionic transition state.
zwitterionic intermediates.
In the context of alkynylboronate cycloadditions, DFT-
led studies have suggested that these dienophiles perform
Key words: cycloaddition, Diels–Alder reaction, alkynes, boron,
regioselectivity
as modestly electron-rich substrates in reactions with tet-
razines and sydnones.8 We wished to further understand
the reactivity of these compounds by exploring their reac-
tions with 2-pyrones, and report our experimental and
Introduction
theoretical results towards this end herein.
Aromatic boronic acids and their derivatives have become
some of the most heavily utilized intermediates in modern
synthetic organic chemistry. The versatility of these com-
pounds makes them attractive for many applications; the
carbon–boron bond can easily be broken allowing the for-
mation of a wide variety of species via various cross-cou-
pling and functional group interconversion processes.1
Synthetic methods towards these substrates classically in-
volve use of organometallic intermediates, but transition-
metal-catalyzed borylation techniques have emerged as
very effective alternatives.2 A complimentary approach to
benzene-based boronic acid derivatives exploit alkynyl-
boronates in cycloaddition processes, and both thermal-
and metal-promoted variants have been documented.3 In
this context, we have recently reported an inverse-elec-
tron-demand [4+2] cycloaddition of alkynylboronates
with 2-pyrones.4 This strategy has proved to be of good
general utility, allowing the formation of highly function-
alized benzene boronic esters in a single step with good
overall yield.
Results and Discussion
Previous studies within our group have shown that 2-py-
rones bearing a methyl ester at C4 or C5 undergo efficient
cycloaddition reactions.4a,b This was presumed to be due
to the LUMO lowering effect of this group, thereby pro-
moting the inverse-electron-demand [4+2] reaction. Inter-
estingly, while the position of the ester group had little
effect on the overall reaction yields, it was found to have
a significant impact on the cycloaddition regioselectivi-
ties (Scheme 1).4b,c
In order to better understand the effect of 2-pyrone sub-
stituents on the reaction efficiency and regioselectivity,
we undertook an extensive study on the cycloaddition re-
actions of a number of functionalized 2-pyrones with al-
kynylboronates. Initially, the synthesis and reactivities of
the readily accessible halo-2-pyrones was investigated
(Table 1). Specifically, we were interested in examining
the reactivities of the isomeric compounds 5-chloro-2-py-
rone (1a) and 4-chloro-2-pyrone (1b), as these would pro-
vide us with a useful insight into the effect of altering the
halogen substitution pattern on reaction regiocontrol. In
the event, both 2-pyrones provided high yields of cycload-
ducts 3b and 4b when reacted with silyl-substituted ethy-
nylboronate 2b, but the products were formed with very
poor regioselectivity (entries 2, 4). In contrast, only one
regioisomer was formed when both 1a,b were reacted
with phenylethynylboronate 2a, although low yields of
cycloadduct were obtained in each case (entries 1, 3).9
2-Pyrone substrates have been widely used in the litera-
ture as the diene partner in [4+2]-cycloaddition reactions.5
Due to the electron-withdrawing nature of the carbonyl
moiety in the ring, 2-pyrones tend to undergo inverse-
electron-demand cycloadditions, usually reacting with
relatively electron-rich dienophiles. However, studies by
Afarinkia have shown that halo-substituted 2-pyrones can
SYNTHESIS 2012, 44, 1964–1973
Advanced online publication: 31.05.2012
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0031-1291142; Art ID: SS-2012-E0303-FA
© Georg Thieme Verlag Stuttgart · New York

FEATURE ARTICLE
[4+2] Cycloadditions of 2-Pyrones and Alkynylboronates
1965
O
O
Ph
MeO2C
Ph
O
Ph
B
+
O
O
O
MeO2C
B
B
MeO2C
O
O
o-DCB, 180 °C, 18 h
75% (14:1)
MeO2C
O
Ph
Ph
MeO2C
O
Ph
B
O
+
O
O
O
MeO2C
B
B
O
O
o-DCB, 180 °C, 18 h
67% (1:1)
Scheme 1 Regiochemical dependence of ester-substituted 2-pyrone cycloadditions of alkynylboronates
Biographical Sketches█
James Kirkham was born field to undertake his Ph.D. search programme under the
in Nottingham, UK, in studies under the guidance EPSRC/University of Shef-
1986. He gained his under- of Prof. J. P. A. Harrity, field Prize Doctoral Fellow-
graduate degree from the which he completed in ship scheme.
University of Sheffield in 2011. He is currently under-
2008. He remained in Shef- taking an independent re-
Andrew Leach grew up in tained an AstraZeneca-Ful- tion energies. Since 2003 he
Chester, UK and gained his bright Award to work with has been a computational
undergraduate and Ph.D. Prof. K. N. Houk at the Uni- chemist with AstraZeneca
degrees from the University versity of California, Los where he has worked in both
of Cambridge where he Angeles studying pericyclic the Cancer and Cardiovas-
worked with Prof. S. V. reactions and the magnitude cular research areas.
Ley. Subsequently, he at- of protein–ligand interac-
Eleanor Row was born in pharma in the design and of antimalarial agents under
Sheffield, and gained her synthesis of CPY3A4 inhib- the supervision of Professor
undergraduate degree from itors. Her subsequent post- P. M. O’Neill. In 2006 Elea-
the University of Hull in doctoral studies were car- nor moved to sanofi–aventis
1999 whilst working for ried out at the University of and is currently working as
Corus. In 2003 she complet- Liverpool into anti-parasitic a Senior Research Investi-
ed her Ph.D. at the Universi- agents under the supervision gator in Isotope Chemistry
ty
of
Sheffield
in of Dr A. Stachulski fol- for Covance Ltd.
collaboration with SAFC lowed by the development
Joseph Harrity is Professor carried out postdoctoral Professor in 2009. His re-
in Organic Chemistry at the studies in Boston College, cent research interests have
University of Sheffield. He USA (1994–1997) with Pro- focused on developing new
received his B.Sc. degree fessor Amir Hoveyda before strategies towards carbon–
from the University of returning to take up a lec- carbon bond forming pro-
Strathclyde in 1991 where tureship in 1997 at the Uni- cesses for which he was
he remained to undertake versity of Sheffield. Here he awarded the Pfizer Discov-
his Ph.D. studies (1991– was promoted to Senior ery Academic Award in
1994) under the guidance of Lecturer and Reader, fol- 2004.
Professor W. J. Kerr. He lowed by a promotion to
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1964–1973

1966
J. D. Kirkham et al.
FEATURE ARTICLE
Table 3 Cycloadditions of 5-Cyano-2-pyrones
Table 1 Cycloadditions of 4-Chloro- and 5-Chloro-2-pyrones
O
O
O
O
B
O
B
O
B
Y
R
B
Y
O
B
R
B
X
Y
O
X
O
Cl
O
2a,b
O
O
O
2a–c
o-DCB
O
+
+
O
mesitylene
155 °C, 16 h
O
R
R
Cl
R
X
X
R
180 °C, 18 h
Y
1a,b
3
4
1e,f
7
8
Entry 2-Pyrone
X
Y
R
Yield (%) Ratio 3/4
Entry 2-Pyrone
X
Y
R
Yield (%) Ratio 7/8
1
2
3
4
1a
1a
1b
1b
H
Cl Ph; 2a
Cl SiMe3; 2b
3a; 21
>98:2
4:3
1
2
3
4
5
6
1e
1e
1e
1f
1f
1f
CN
CN
CN
H
H
H
Ph; 2a
7a; 76
>98:2
1:1
H
3b/4b; 70
4a; 25
SiMe3; 2b 7b/8b; 99
Cl
Cl
H
H
Ph; 2a
< 2:98
3:5
Bu; 2c
7c/8c; 53
7d; 94
5:1
SiMe3; 2b
3b/4b; 70
CN Br Ph; 2a
>98:2
1:1
CN Br SiMe3; 2b 7e/8e; 96
CN Br Bu; 2c 7f/8f; 65
Table 2 Cycloadditions of Chloro- and Bromo-Substituted 2-Py-
rones
11:1
O
Y
Y
O
B
O
B
R
B
tion product 7a regioselectively (entry 1), whereas the
silylethynylboronate 2b afforded 7b/7b with no regiose-
lectivity (entry 2). The butyl-substituted ethynylboronate
provided products in moderate yield, but with good re-
gioselectivity (entry 3). Bromination of 5-cyano-2-pyrone
was also achieved successfully, and the resulting 2-bro-
mo-substituted 2-pyrone 1f reacted with the alkynylboro-
nates 2a–c to afford the corresponding products in
excellent yields. Interestingly, the extra steric bulk pro-
vided by bromide appeared to have little effect on the re-
action regioselectivities, with products 7d, 7e/8e, and
7f/8f forming in similar selectivities to those generated
from pyrone 1e (entries 4–6).11
O
O
X
2a–c
O
O
+
O
mesitylene
155 °C, 16 h
X
X
R
R
Y
1c,d
5
6
Entry 2-Pyrone
X
Y
R
Yield (%) Ratio 5/6
1
2
3
4
5
1c
1c
1d
1d
1d
Cl
Cl
Br
Br
Br
Cl Ph; 2a
Cl SiMe3; 2b
5a; 32
>98:2
1:1
5b/6b; 71
5c; 46
H
H
H
Ph; 2a
>98:2
3:2
SiMe3; 2b
Bu; 2c
5d/6d; 80
5e/6e; 44
3:2
From these studies two trends in the reactivities of al-
kynylboronates have been observed, allowing us to form
the following general conclusions:
Having established the reactivity patterns of 2-pyrones
1a,b towards alkynylboronate cycloadditions, the reac-
tions of the readily available 3,5-dichloro-2-pyrone (1c)
and 5-bromo-2-pyrone (1d) were studied next (Table 2).
Pleasingly, both dienes provided silyl-substituted cyclo-
adducts 5b/6b and 5d/6d in good yields (entries 2, 4).
Once again, however, poor levels of regioselectivity were
observed. Phenyl-substituted aromatics 5a and 5c were
obtained regioselectively, but in poor yield (entries 1, 3).
2-Pyrone 1d was also reacted with an alkyl-substituted
ethynylboronate 2c, which provided products 5e/6e in rea-
sonable yield, but with low levels of regiocontrol (entry
5).10
Aryl-substituted ethynylboronates undergo highly selec-
tive cycloaddition reactions with 2-pyrones, but with
varying yields.
Silyl-substituted ethynylboronates undergo high-yielding
cycloaddition reactions with 2-pyrones, but with little to
no regioselectivity.
In an effort to further understand these trends, quantum
mechanical studies were performed to map the full reac-
tion profile and to provide insight into the factors govern-
ing regioselectivity. The cycloaddition is assumed to
proceed through a Diels–Alder reaction to form a bicyclic
adduct and then a retro-Diels–Alder reaction extruding
carbon dioxide. The calculations employed three different
levels of theory. B3LYP is known to perform reasonably
well in studies of pericyclic reactions,12 whereas although
RHF is expected to over predict barrier heights it often
performs well (given the computational cost) at determin-
ing reaction energies.13 Meanwhile, M06-2X has been
claimed to be effective at computing barrier heights and
transition state geometries.14 All three levels were com-
bined with the 6-31+G** basis set and the results are sum-
The poor yields observed in many of the cycloadditions of
halo-2-pyrones prompted us to explore dienes with elec-
tron-withdrawing groups as these had previously been
found to undergo more efficient reactions (see earlier in
Scheme 1). Therefore, we prepared 5-cyano-2-pyrone
(1e) and explored its reactions with alkynylboronates 2a–c
(Table 3). We were delighted to find that this diene re-
acted with alkynylboronates with far higher efficiency
than simpler halo-2-pyrones. Similar to the previous ex-
amples, phenylethynylboronate 2a provided cycloaddi-
Synthesis 2012, 44, 1964–1973
© Georg Thieme Verlag Stuttgart · New York

FEATURE ARTICLE
[4+2] Cycloadditions of 2-Pyrones and Alkynylboronates
1967
O
+46.2
[+66.3]
(+37.1)
O
O
B
O
Ph
MeO
+15.6
[+31.3]
(+8.3)
+41.2
[+62.3]
(+34.2)
O
Ph
B
+2.3
[+3.5]
(–12.4)
+13.2
[+27.8]
(+4.6)
O
MeO
O
+1.2
[+2.9]
(–12.0)
O
O
O
B
O
O
O
MeO
MeO
–58.7
[–63.2]
(–63.7)
Ph
O
O
Ph
O
O
O
B
–59.5
[–64.1]
(–64.5)
B
MeO
O
O
O
Ph
Figure 1 The free energy profile for the reaction between diene 1g and dienophile 2a. The first step is a Diels–Alder cycloaddition and the
second a retro-Diels–Alder reaction in which carbon dioxide leaves. The energies reported are in kcal/mol relative to reactants and include vi-
brational corrections to free energies in the gas phase at 298 K. They are calculated at the B3LYP/6-31+G** [RHF/6-31+G**] (M062X/6-
31+G**) levels of theory.
marized in Figure 1 for the reaction between diene 1g and
dienophile 2a.
O
O
Ph
B
B
O
O
It can be seen in Figure 1 that all levels of theory agree that
the reaction proceeds through a rate-limiting transition
state for Diels–Alder addition to yield an adduct that is
within a few kcal/mol of the reactants. This is followed by
a retro-Diels–Alder reaction with a lower barrier than the
first step and this yields an aromatic product that is signif-
icantly lower in energy than the reactants or intermedi-
ates. This same pattern is found for all reactions studied
with all three levels of theory. This reaction profile would
have the first step as rate and selectivity determining and
so although all the cycloadditions described have been
studied across the whole profile and these values and
structures are reported in the supporting information, the
focus here will be on the transition states for the first step.
MeO
O
MeO
O
Ph
O
O
O
O
PROXIMAL
DISTAL
Figure 2 The regioisomeric structures and transition states as de-
scribed by the relative positioning of the boronic ester group and the
oxygen of the pyran ring
that the degree of preference changes by several kcal/mol.
The experimentally observed preference can be trans-
formed into an estimate of the energy difference between
the processes assuming the two regioisomers are formed
in processes obeying Arrhenius behaviour and with
It is seen in Figure 1 that although the barrier heights vary
significantly depending upon the level of theory that is
employed, the difference between the barriers leading to
the two regioisomeric adducts are relatively consistent.
B3LYP predicts that the proximal adduct (here denoting
that the boronic ester adds adjacent to the oxygen in the
pyran ring, see Figure 2) is favoured over the distal adduct
(where the boronic ester adds away from the pyran ring
oxygen) by 5 kcal/mol, RHF by 4 kcal/mol, and M06-2X
by 2.9 kcal/mol. All levels of theory are consistent with
the observed regiochemical outcome, although the exper-
imentally observed selectivity (14:1) is lower than would
be expected for any of the differences in free energy com-
puted here.
similar
pre-exponential
factors
such
that
ΔΔG‡ = RT ln(%proximal/%distal). These values are
shown in Table 4 for the reactions that have been studied
experimentally. The computed differences can be com-
pared to those obtained from the experimental regioselec-
tivities and show that the B3LYP calculations are able to
provide a good correlation between the two (Figure 3).15
RHF also provides a reasonable correspondence, but
M06-2X does not. RHF however predicts barriers that are
higher than is likely realistic and so the B3LYP descrip-
tion of the reaction appears to be the one that is most real-
istic and, hence, it is the structures computed at this level
of theory that have been studied to provide insight into the
origins of the regioselectivity. The plot in Figure 3 sug-
gests that the experimental regioselectivity is not perfectly
predicted by the B3LYP gas phase calculations, but that
the experimental ΔΔG‡ is approximately that computed
by B3LYP multiplied by a factor of 0.54, with a small cor-
rection of 0.1 kcal/mol added. Predictions can be made of
the preference for the proximal regioisomer using this cor-
To ascertain if any of the levels of theory are consistent
with the observed outcomes, the differences in free energy
barrier for the two regioisomers has been computed for a
set of the dienes and dienophiles described above. These
are summarized in Table 4 and show that all levels of the-
ory predict that the proximal regioisomer is preferred, but
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1964–1973

1968
J. D. Kirkham et al.
FEATURE ARTICLE
Table 4 Calculated and Experimental Free Energy of Activation Values for Proximal and Distal Regioisomer Formationa
Diene
R
ΔΔG‡
ΔΔG‡
ΔΔG‡
Experimental
Predicted
RHF
B3LYP
M06-2X
proximal/distal ΔΔG‡
proximal/distal
Ph
1.9
0.9
1.9
1.9
0.6
1.8
4.9
0.6
5.3
4.5
0.6
1.6
0.5
0.4
1.0
4.0
0.5
2.7
2.7
0.4
2.3
3.4
0.9
3.2
6.6
0.8
2.4
5.8
0.7
3.0
0.9
0.3
1.7
5.0
0.7
3.0
0.9
0.6
2.6
2.3
0.2
1.2
5.7
2.4
5.9
3.6
1.5
1.3
0.2
1.5
0.2
2.9
0.3
0.6
>49:1
1.3:1
>3.3
0.2
7.0:1
1.5:1
5.2:1
10.5:1
2.1:1
9.3:1
87.3:1
2.0:1
5.6:1
42.1:1
1.8:1
7.3:1
2.1:1
1.4:1
3.2:1
26.1:1
1.7:1
7.6:1
Cl
O
O
O
O
TMS
Bu
Ph
>49:1
1.7:1
>3.3
0.4
Cl
TMS
Bu
Cl
Ph
>49:1
1:1
>3.3
0.0
O
O
TMS
Bu
O
O
Cl
Ph
>49:1
1:1
>3.5
0.0
1.4
0.0
1.0
2.1
2.4
0.0
1.0
NC
TMS
Bu
5:1
MeO2C
O
O
Ph
1:1
O
TMS
Bu
3:1
10:1
14:1
1:1
Ph
O
MeO2C
TMS
Bu
3:1
a The ratios are based on a reaction temperature of 155 °C.
rection and these are included in the final column in Table To investigate the origin of the effects that are manifest in
4. These predictions show that in general the phenylethy- both the experimental outcomes and the computed differ-
nylboronate is expected to be the most regioselective of ences in free energy of activation, the geometries of the
the dienophiles examined and that electron-withdrawing various computed structures have been examined. In Fig-
groups at the 5-position of the diene enhance selectivity, ure 4, the structures of the reactants, intermediates, prod-
but that electron-withdrawing groups at the 4-position di- ucts, and transition states for the profile shown in Figure
minish the regioselectivity.
1 are presented. Of particular interest is that the two tran-
sition states for the initial Diels–Alder reaction differ
markedly in the lengths of the two forming bonds. It can
be seen that the lower energy structure is considerably
more asynchronous than that of the higher energy one. An
asynchronous transition state might be part of a stepwise
process, but even when solvation was included (using the
default parameters for mesitylene in Gaussian09) no inter-
mediate could be located. The two structures are shown in
more detail in Figure 5.
The forming bonds have lengths of 1.867 Å and 2.865 Å
for the proximal regioisomer and 2.160 Å and 2.305 Å for
the distal regioisomer. The same heightened asynchronic-
ity is also observed with the RHF and M06-2X structures,
but it is less pronounced and this may lie at the origin of
why they are less well able to predict the regioselectivity.
DFT studies on 2-pyrone Diels–Alder reactions reported
by Štefane et al. also highlighted a polar and asynchro-
Figure 3 The experimentally observed preference for the proximal re-
gioisomer (transformed to an effective ΔΔG‡) is plotted against the
difference in free energies of activation computed at three different
levels of theory (RHF/6-31+G**, B3LYP/6-31+G**, and M06-2X/6-
31+G**). The red line is the y = x line. All energies are in kcal/mol.
Synthesis 2012, 44, 1964–1973
© Georg Thieme Verlag Stuttgart · New York

FEATURE ARTICLE
[4+2] Cycloadditions of 2-Pyrones and Alkynylboronates
1969
Figure 4 The structures corresponding to the reaction of methyl coumalate (1g) and 2a and the energetic profile shown in Figure 1
nous transition state in reactions with electron-rich dieno-
philes.16 Taken together with the results of our earlier
studies suggesting the electron-rich nature of alkynylbor-
onates,8 the transition states uncovered in this work are
consistent with a mechanism that involves a concerted re-
action with significant polar character, and can be viewed
as an addition of a nucleophilic alkyne with the 2-pyrone
acting as an electrophile emphasizing the inverse-
electron-demand nature of this reaction. The differences
Figure 6 The transition states computed at the B3LYP/6-31+G**
in other bond lengths, including that of the carbonyl, level leading to the two regioisomers for the reaction between 1g and
2b; bond distances are shown in Å
shown in Figure 5, support this view of the transition state.
It was believed that a regioselective reaction could be af-
fected by using an alternative group that could stabilize β-
carbocations. As such, the 3-silylprop-1-ynylboronate 2d
was synthesized (Scheme 2), and reacted with methyl
coumalate (1g).
i) n-BuLi, THF, –78 °C
O
ii)
Figure 5 The rate-limiting transition state structures for the reaction
of methyl coumalate (1g) and 2a leading to proximal (left) and distal
(right) regioisomers; bond distances are shown in Å
i-PrO
B
Me3Si
Me3Si
O
O
O
B
H
iii) 1 M HCl
2d; 40%
When the trimethylsilyl-substituted alkyne transition
states are examined (as shown in Figure 6), it is seen that
both are less asynchronous than the proximal transition
state shown in Figure 5, suggesting that in this case both
ends of the alkyne are approximately equal in terms of nu-
cleophilicity.
Scheme 2 Synthesis of 3-(trimethylsilyl)prop-1-ynylboronate 2d
In the event, the cycloaddition occurred regioselectively
with the preferred isomer being consistent with the one
predicted by theoretical studies (Scheme 3). Although re-
giocontrol was not as high as that noted with the corre-
sponding phenylacetylene-derived substrate 2a (see
earlier in Scheme 1), an enhanced selectivity over the re-
lated butyl-substituted acetylene cycloadditions is ob-
served.18,19
The electronic similarity of the acetylenic carbon atoms in
2b can be rationalized by inspection of the field effect val-
ues for Si- and B-based substituents in the Swain–Lupton
equation [Me3Si; +0.01 and (HO)2B; –0.03].17
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1964–1973

1970
J. D. Kirkham et al.
FEATURE ARTICLE
SiMe3
BPin
SiMe3
BPin
SiMe3
O
MeO2C
o-DCB
175 °C, 18 h
+
+
O
MeO2C
MeO2C
BPin
1g
2d
9
41% (9/10 = 6:1) 10
Scheme 3 Cycloaddition of 3-(trimethylsilyl)prop-1-ynylboronate 2d
Flash chromatography was performed on silica gel (BDH Silica Gel
60 43-60, or Fluorochem Davisil silica gel 43–60); petroleum ether =
PE. TLC was performed on aluminum-backed plates pre-coated
with silica (0.2 mm, Merck DC-alufolien Kieselgel 60 F254), which
were developed using standard visualizing agents: UV light or
KMnO4. 1H/13C NMR spectra were recorded on Bruker AC-250 or
Conclusion
In conclusion, we have found that higher yields for the re-
actions of 2-pyrones with alkynylboronates can be ob-
tained when the silyl-substituted ethynylboronate is used.
The highest regioselectivities can be obtained with the
phenyl-substituted dienophile, where typically only one
regioisomer is formed. 2-Pyrones that have electron-
withdrawing groups incorporated provide cycloadducts in
generally higher yields. Halogens were shown to be es-
sentially electron-neutral functional groups on 2-pyrone
rings, as no enhancement or deterioration of either yields
or regioselectivities were observed for these substrates.
1
Av1-250 instruments or AMX-400 or AV1-400 instruments. H:
solvent resonance as the internal standard (CHCl3:δ = 7.27). 13C
NMR spectra were with complete proton decoupling; solvent reso-
nance as the internal standard (CDCl3: δ = 77.0). FTIR spectra were
recorded on a Perkin Elmer Paragon 100 FTIR spectrophotometer.
Samples were recorded as thin films using NaCl plates, as a CH2Cl2
soln or as a KBr disc. LR-MS were recorded on Micromass Auto-
spec, operating in EI, CI, or FAB mode; or a Perkin-Elmer Tur-
bomass Bench top GC-MS operating in either EI or CI mode.
HRMS recorded for accurate mass analysis, were performed on ei-
ther a MicroMass LCT operating in Electrospray mode (ES+) or a
MicroMass Prospec operating in either FAB (FAB+), EI (EI+) or CI
(CI+) mode. Melting points were performed on recrystallized solids
and recorded on a Gallenkamp melting point apparatus and are un-
corrected. Compounds 1a–f5,6,20 and 2a–c21 were prepared accord-
ing to literature procedures.
Through theoretical studies of the cycloaddition transition
states, we have discovered that 2-pyrone reactions with al-
kynylboronates proceed via asynchronous transition
states. This accounts for the differences in regiochemical
outcomes for phenyl- and silyl-substituted ethynylboro-
nates. Moreover, the significance of zwitterionic transi-
tion states, as proposed by Stefané,16 and implicated here,
may provide a rationale for the different regiochemical
outcomes observed in cycloadditions of phenyl-substitut-
ed acetylene 2a with isomeric ester substituted pyrones
(see earlier in Scheme 1). Specifically, the ester group in
methyl coumalate reinforces stabilization of the zwitter-
ionic transition state, whereas the 4-ester isomer would
provide a pathway for stabilizing both proximal and distal
addition modes and therefore diminishes cycloaddition
regioselectivity (Scheme 4).
Cycloaddition of Halo-2-pyrones with Alkynylboronic Esters;
General Procedure 1
A mixture of the 2-pyrone (0.2 mmol) and alkynylboronate (0.4
mmol) in mesitylene (0.2 mL) was heated at 155 °C and stirred for
16 h under N2. The product was purified by flash column chroma-
tography (PE to 10% EtOAc–PE).22
2-(4-Chlorobiphenyl-2-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lane (3a)
Using general procedure 1, with 2-pyrone 1a (25 mg, 0.19 mmol)
and alkyne 2a (87 mg, 0.38 mmol), gave 3a (13 mg, 21%) as a yel-
low oil.
FTIR (CH2Cl2, thin film): 2977 (s), 1544 (m), 1315 (s), 1141 cm–1
(s).
1H NMR (250 MHz, CDCl3): δ = 1.23 (s, 12 H, CH3), 7.26 (d,
J = 8.0 Hz, 1 H, HAr), 7.37–7.39 (m, 5 H, HAr), 7.59 (dd, J = 8.0, 2.5
Hz, 1 H, HAr), 7.85 (d, J = 2.5 Hz, 1 H, HAr).
13C NMR (100.6 MHz, CDCl3): δ = 24.6, 84.1, 121.1, 127.2, 127.9,
129.0, 130.8, 133.0, 137.0, 142.0, 146.4.
O
O
Ph
14:1
O
MeO2C
MeO2C
O
MeO2C
MeO2C
MeO2C
BPin
H
BPin
reinforcing
HRMS (EI+): m/z [M]+ calcd for C18H2011B35ClO2: 314.1245; found:
314.1238.
O
O
MeO2C
Ph
1:1
O
O
BPin
H
2-[5-Chloro-2-(trimethylsilyl)phenyl-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (3b) and 2-[4-Chloro-2-(trimethylsilyl)phenyl-
4,4,5,5-tetramethyl-1,3,2-dioxaborolane (4b)
Using general procedure 1, with 2-pyrone 1a (25 mg, 0.19 mmol)
and alkyne 2b (85 mg, 0.38 mmol), gave the product (41 mg, 70%)
as a clear oil; inseparable mixture of 3b/4b (4:3).
BPin
diminishing
BPin
O
H
O
MeO
O
FTIR (CH2Cl2, thin film): 2980 (s), 1570 (m), 1388 (s), 1340 (s),
1145 (s), 845 cm–1 (s).
1H NMR (250 MHz, CDCl3): δ (3b) = 0.35 (s, 9 H, SiCH3), 1.38 (s,
12 H, CH3), 7.38 (dd, J = 8.0, 2.0 Hz, 1 H, HAr), 7.55 (d, J = 8.0 Hz,
1 H, HAr), 7.89 (d, J = 2.0 Hz, 1 H, HAr); δ (4b) = 0.37 (s, 9 H,
SiCH3), 1.38 (s, 12 H, CH3), 7.34 (dd, J = 8.0, 2.0 Hz, 1 H, HAr),
7.57 (d, J = 2.0 Hz, 1 H, HAr), 7.87 (d, J = 8.0 Hz, 1 H, HAr).
Scheme 4 Regiochemistry reinforcing and diminishing effects of es-
ter substitution
Synthesis 2012, 44, 1964–1973
© Georg Thieme Verlag Stuttgart · New York

FEATURE ARTICLE
[4+2] Cycloadditions of 2-Pyrones and Alkynylboronates
FTIR (CH2Cl2, thin film): 2981 (w), 1459 cm–1 (w).
1971
13C NMR (62.9 MHz, CDCl3): δ (3b/4b) = 0.0, 0.1, 24.6 (2 ×), 83.6,
86.9, 127.4, 129.2, 130.7, 132.3, 133.8, 135.0, 135.4, 135.5, 136.3,
137.3.
1H NMR (250 MHz, CDCl3): δ = 1.23 (s, 12 H, CH3), 7.26 (d,
J = 8.0 Hz, 1 H, HAr), 7.37–7.39 (m, 5 H, HAr), 7.59 (dd, J = 8.0, 2.5
Hz, 1 H, HAr), 7.85 (d, J = 2.5 Hz, 1 H, HAr).
HRMS (EI+): m/z [M]+ calcd for C15H2411B35ClO2Si: 310.1327;
found: 310.1335.
13C NMR (100.6 MHz, CDCl3): δ = 24.6, 84.1, 121.1, 127.2, 127.9,
129.0, 130.8, 133.0, 137.0, 142.0, 146.4.
HRMS (EI+): m/z [M]+ calcd for C18H2011B79BrO2: 359.0652; found:
359.0650.
2-(5-Chlorobiphenyl-2-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lane (4a)
Using general procedure 1, with 2-pyrone 1b (25 mg, 0.19 mmol)
and alkyne 2a (87 mg, 0.38 mmol), gave 3a (15 mg, 25%) as a yel-
low oil.
FTIR (CH2Cl2, thin film): 2977 (s), 1544 (m), 1315 (s), 1141 cm–1
(s).
1H NMR (250 MHz, CDCl3): δ = 1.21 (s, 12 H, CH3), 7.30–7.45 (m,
7 H, HAr), 7.67 (d, J = 8.0 Hz, 1 H, HAr).
2-[5-Bromo-2-(trimethylsilyl)phenyl-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (5d) and 2-[4-Bromo-2-(trimethylsilyl)phenyl-
4,4,5,5-tetramethyl-1,3,2-dioxaborolane (6d)
Using general procedure 1, with 2-pyrone 1d (50 mg, 0.29 mmol)
and alkyne 2b (130 mg, 0.58 mmol), gave the product (82 mg, 80%)
as a clear oil; inseparable mixture of 5d/6d (3:2).
FTIR (CH2Cl2, thin film): 2980 (s), 2977 (w), 1454 cm–1 (w).
13C NMR (62.9 MHz, CDCl3): δ = 24.6, 83.9, 126.4, 127.2, 127.4,
1H NMR (400 MHz, CDCl3): δ (5d or 6d) = 0.36 (s, 9 H, SiCH3),
1.38 (s, 12 H, CH3), 7.49–7.51 (m, 1 H, HAr), 7.53–7.56 (m, 1 H,
HAr), 8.06 (d, J = 2.0 Hz, 1 H, HAr); δ (5d or 6d) = 0.38 (s, 9 H,
SiCH3), 1.38 (s, 12 H, CH3), 7.46–7.48 (m, 1 H, HAr), 7.74 (d,
J = 2.0 Hz, 1 H, HAr), 7.80 (d, J = 8.0 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (5d/6d) = 0.5 (2 ×), 25.0 (2 ×),
84.0, 84.2, 123.4, 125.9, 130.8, 132.6, 136.0, 137.0, 137.9, 138.7,
145.6, 150.3.
127.9, 129.0, 129.1, 130.0, 134.2, 135.9.
HRMS (EI+): m/z [M]+ calcd for C18H2011B35ClO2: 314.1245; found:
314.1238.
2-[5-Chloro-2-(trimethylsilyl)phenyl-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (3b) and 2-[4-Chloro-2-(trimethylsilyl)phenyl-
4,4,5,5-tetramethyl-1,3,2-dioxaborolane (4b)
Using general procedure 1, with 2-pyrone 1b (25 mg, 0.19 mmol)
and alkyne 2b (85 mg, 0.38 mmol), gave the product (40 mg, 70%)
as a clear oil; inseparable mixture of 3b/4b (3:5).
HRMS (EI+): m/z [M]+ calcd for C15H2411B79BrO2Si: 355.1504;
found: 355.1507.
The mixture provided the same analytical data as for 3b and 4b
above.
2-(5-Bromo-2-butylphenyl)-4,4,5,5-tetramethyl-1,3,2-dioxabor-
olane (5e) and 2-(4-Bromo-2-butylphenyl)-4,4,5,5-tetramethyl-
1,3,2-dioxaborolane (6e)
Using general procedure 1, with 2-pyrone 1d (50 mg, 0.29 mmol)
and alkyne 2c (121 mg, 0.58 mmol), gave the product (43 mg, 44%)
as a brown oil; inseparable mixture of 5e/6e (3:2).
2-(4,6-Dichlorobiphenyl-2-yl)-4,4,5,5-tetramethyl-1,3,2-dioxa-
borolane (5a)
Using general procedure 1, with 2-pyrone 1c (25 mg, 0.15 mmol)
and alkyne 2a (68 mg, 0.30 mmol), gave 5a (17 mg, 32%) as a yel-
low oil.
FTIR (CH2Cl2, thin film): 2979 (s), 1547 (m), 1328 (s), 1144 cm–1
(s).
FTIR (CH2Cl2, thin film): 2958 (s), 2871 (m), 1715 (m), 1584 (m),
1345 (s), 1145 (s), 865 cm–1 (m).
1H NMR (250 MHz, CDCl3): δ (5e) = 0.94 (t, J = 7.0 Hz, 3 H, CH3),
1.23–1.60 (m, 4 H, CH2), 1.36 (s, 12 H, CH3), 2.80–2.88 (m, 2 H,
CH2), 7.05 (d, J = 8.0 Hz, 1 H, HAr), 7.45 (dd, J = 8.0, 2.5 Hz, 1 H,
HAr), 7.88 (d, J = 2.5 Hz, 1 H, HAr); δ (6e) = 0.94 (t, J = 7.0 Hz, 3 H,
CH3), 1.23–1.60 (m, 4 H, CH2), 1.35 (s, 12 H, CH3), 2.80–2.88 (m,
2 H, CH2), 7.28–7.34 (m, 2 H, HAr), 7.63 (d, J = 8.0, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (5e/6e) = 13.9 (2 ×), 22.7 (2 ×),
24.8 (2 ×), 34.9, 35.3, 35.4 (2 ×), 83.6, 83.7, 119.1, 125.6, 128.0,
131.0, 132.1, 133.5, 137.6, 138.4, 148.9, 152.4.
1H NMR (250 MHz, CDCl3): δ (5a) = 1.10 (s, 12 H, CH3), 7.24–
7.28 (m, 2 H, HAr), 7.36–7.42 (m, 3 H, HAr), 7.53 (d, J = 2.0 Hz, 1
H, HAr), 7.56 (d, J = 2.0 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (5a) = 24.5, 84.1, 127.5, 127.6,
129.7, 130.7, 132.0, 132.9, 133.3, 134.0, 139.2.
HRMS (EI+): m/z [M]+ calcd for C18H1911B35Cl2O2: 348.0855;
found: 348.0869.
2-[3,5-Dichloro-2-(trimethylsilyl)phenyl]-4,4,5,5-tetramethyl-
1,3,2-dioxaborolane (5b) and 2-[2,4-Dichloro-6-(trimethylsi-
lyl)phenyl]-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (6b)
Using general procedure 1, with 2-pyrone 1c (25 mg, 0.15 mmol)
and alkyne 2b (67 mg, 0.30 mmol), gave the product (37 mg, 71%)
as a clear oil; inseparable mixture of 5b/6b (1:1).
HRMS (EI+): m/z [M]+ calcd for C16H2411B79BrO2: 338.1053; found:
338.1066.
Cycloaddition of Nitrile-2-pyrones with Alkynylboronic Esters;
General Procedure 2
A mixture of the 2-pyrone (0.2 mmol) and the alkynylboronate (0.4
mmol) in o-dichlorobenzene (0.2 mL) was heated at 175 °C and
stirred for 18 h under N2. The product was purified by flash column
chromatography (PE to 10% EtOAc–PE).22
FTIR (CH2Cl2, thin film): 2981 (s), 1562 (m), 1318 (s), 1142 (s),
1050 (m), 846 cm–1 (s).
1H NMR (250 MHz, CDCl3): δ (5b/6b) = 0.36 (s, 9 H, SiCH3), 0.43
(s, 9 H, SiCH3), 1.39 (s, 12 H, CH3), 1.45 (s, 12 H, CH3), 7.34–7.44
(m, 4 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (5b/6b) = 0.0, 1.7, 25.3, 25.8, 84.6,
84.9, 113.2, 115.6, 119.1, 128.7, 130.3, 131.9 (2 ×), 135.0, 135.3,
138.7.
2-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)biphenyl-4-car-
bonitrile (7a)
Using general procedure 2, with 2-pyrone 1e (25 mg, 0.21 mmol)
and alkyne 2a (96 mg, 0.42 mmol), gave 7a (48 mg, 76%) as a clear
oil.
HRMS (EI+): m/z [M]+ calcd for C15H2311B35Cl2O2Si: 344.0937;
found: 344.0932.
FTIR (CH2Cl2, thin film): 2979 (m), 2228 (m), 1598 (m), 1346
cm–1 (s).
1H NMR (250 MHz, CDCl3): δ = 1.24 (s, 12 H, CH3), 7.36–7.44 (m,
5 H, HAr), 7.49 (d, J = 8.0 Hz, 1 H, HAr), 7.74 (dd, J = 8.0, 2.0 Hz,
1 H, HAr), 8.02 (d, J = 2.0 Hz, 1 H, HAr).
2-(4-Bromobiphenyl-2-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lane (5c)
Using general procedure 1, with 2-pyrone 1d (50 mg, 0.29 mmol)
and alkyne 2a (132 mg, 0.58 mmol), gave 5c (47 mg, 46%) as a yel-
low oil.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1964–1973

1972
J. D. Kirkham et al.
FEATURE ARTICLE
13C NMR (100.6 MHz, CDCl3): 24.6, 84.5, 109.9, 119.0, 128.0,
128.1, 128.9, 129.6, 133.4, 138.3, 141.4, 151.9.
FTIR (CH2Cl2, thin film): 2981 (s), 2232 (m), 1332 (s), 1140 (s),
1048 (m), 847 cm–1 (s).
HRMS (ES+): m/z [M]+ calcd for C19H2011BNO2: 306.1665; found:
306.1664.
1H NMR (250 MHz, CDCl3): δ (7e/8e) = 0.38 (s, 9 H, SiCH3), 0.48
(s, 9 H, SiCH3), 1.39 (s, 12 H, CH3), 1.48 (s, 12 H, CH3), 7.71 (d,
J = 1.5 Hz, 1 H, HAr), 7.75 (d, J = 1.5 Hz, 1 H, HAr), 7.77 (d, J = 1.5
Hz, 1 H, HAr), 7.81 (d, J = 1.5 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (7e/8e) = 0.0, 1.9, 25.5, 26.1, 85.1,
85.6, 113.5, 113.8, 117.6, 117.9, 127.9, 131.3, 131.5, 134.8, 135.2,
135.3, 136.5, 149.0.
3-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)-4-(trimethylsi-
lyl)benzonitrile (7b) and 4-(4,4,5,5-Tetramethyl-1,3,2-dioxa-
borolan-2-yl)-3-(trimethylsilyl)benzonitrile (8b)
Using general procedure 2, with 2-pyrone 1e (25 mg, 0.21 mmol)
and alkyne 2b (94 mg, 0.42 mmol), gave the product (62 mg, 99%)
as a clear oil; inseparable mixture of 7b/8b (1:1).
FTIR (CH2Cl2, thin film): 2980 (s), 2229 (s), 1342 (s), 1143 cm–1
(s).
1H NMR (250 MHz, CDCl3): δ (7b/8b) = 0.37 (s, 18 H, SiCH3),
1.38 (s, 24 H, CH3), 7.60–7.73 (m, 3 H, HAr), 7.85 (d, J = 1.0 Hz, 1
H, HAr), 7.98 (d, J = 7.5 Hz, 1 H, HAr), 8.17 (d, J = 1.0 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (7b/8b) = 0.0, 0.1, 24.7 (2 ×), 84.3
(2 ×), 111.6, 113.1, 118.8, 119.1, 130.6, 132.1, 134.3, 135.8, 137.0,
138.7, 148.4, 153.4.
HRMS (EI+): m/z [M]+ calcd for C16H2411BNO2Si: 302.1748; found:
302.1735.
HRMS (EI+): m/z [M]+ calcd for C16H2311B79BrNO2Si: 379.0774;
found: 379.0777.
3-Bromo-4-butyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)benzonitrile (7f) and 3-Bromo-5-butyl-4-(4,4,5,5-tetrameth-
yl-1,3,2-dioxaborolan-2-yl)benzonitrile (8f)
Using general procedure 2, with 2-pyrone 1f (25 mg, 0.13 mmol)
and alkyne 2c (54 mg, 0.26 mmol), gave the product (31 mg, 65%)
as a yellow oil; inseparable mixture of 7f/8f (11:1).
FTIR (CH2Cl2, thin film): 2977 (s), 2232 (m), 1337 (s), 1140 (s),
849 cm–1 (s).
1H NMR (250 MHz, CDCl3): δ (7f) = 0.97 (t, J = 7.0 Hz, 3 H,
CH2CH3), 1.28–1.30 (m, 2 H, CH2), 1.37 (s, 12 H, CH3), 1.45–1.49
(m, 4 H, CH2CH2CH3), 7.87 (d, J = 1.5 Hz, 1 H, HAr), 8.00 (d,
J = 1.5 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (7f) = 13.8, 22.9, 24.8, 33.1, 35.2,
84.4, 110.6, 117.5, 125.5, 137.8, 138.6, 154.2.
HRMS (EI+): m/z [M]+ calcd for C17H2311B79BrNO2: 363.1005;
4-Butyl-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzo-
nitrile (7c) and 3-Butyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lan-2-yl)benzonitrile (8c)
Using general procedure 2, with 2-pyrone 1e (25 mg, 0.21 mmol)
and alkyne 2c (87 mg, 0.42 mmol), gave the product (32 mg, 53%)
as a yellow oil; inseparable mixture of 7c/8c (5:1).
found: 363.1003.
FTIR (CH2Cl2, thin film): 2959 (s), 2228 (s), 1600 (s), 1347 (s),
1144 cm–1 (s).
Trimethyl[3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)prop-2-ynyl]silane (2d)
1H NMR (250 MHz, CDCl3): δ (7c) = 0.94 (t, J = 7.0 Hz, 3 H, CH3),
1.23–1.60 (m, 4 H, CH2), 1.36 (s, 12 H, CH3), 2.80–2.88 (m, 2 H,
CH2), 7.27 (d, J = 8.0 Hz, 1 H, HAr), 7.61 (dd, J = 8.0, 2.0 Hz, 1 H,
HAr), 8.07 (d, J = 2.0 Hz, 1 H, HAr); δ (8c) = 0.94 (t, J = 7.0 Hz, 3 H,
CH3), 1.23–1.60 (m, 4 H, CH2), 1.35 (s, 12 H, CH3), 2.80–2.88 (m,
2 H, CH2), 7.41–7.48 (m, 2 H, HAr), 7.84 (d, J = 8.0 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ (7c/8c) = 13.9 (2 ×), 22.6, 22.7,
24.9 (2 ×), 35.1, 35.2, 35.8 (2 ×), 84.1 (2 ×), 109.0 (2 ×), 119.1,
119.2, 128.1, 129.8, 132.2, 133.8, 136.3, 139.9, 155.5 (2 ×).
A 2.5 M soln of BuLi in hexanes (4.25 mL, 8.93 mmol), was added
dropwise to trimethylpropargylsilane (1.00 g, 8.93 mmol) in Et2O,
at –78 °C under N2. The resulting mixture was stirred at this temper-
ature for 1 h, then warmed to r.t. and stirred for 30 min. 2-Iso-
propoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (1.83 mL, 8.93
mmol), was added dropwise, and the resulting suspension stirred for
3 h. 1 M HCl in Et2O (11.61 mL, 11.61 mmol) was added dropwise,
and the reaction stirred for 1 h. Lithium salts were removed via fil-
tration through Celite under N2, and the product purified via recrys-
tallization (PE, bp 40–60 °C), affording 2d (840 mg, 40%) as a
colourless solid; mp 43–45 °C.
HRMS (ES+): m/z [M]+ calcd for C17H2411BNO2: 286.1978; found:
286.1965.
FTIR (CH2Cl2, thin film): 2982 (s), 2955 (s), 2197 (s), 1381 (s),
1312 cm–1 (s).
1H NMR (400 MHz, CDCl3): δ = 0.14 (s, 9 H, SiCH3), 1.28 (s, 12
H, CCH3), 1.62 (s, 2 H, SiCH2).
13C NMR (100.6 MHz, CDCl3): δ = –2.0, 8.6, 24.7, 83.8, 103.7.
2-Bromo-6-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)biphe-
nyl-4-carbonitrile (7d)
Using general procedure 2, with 2-pyrone 1f (25 mg, 0.13 mmol)
and alkyne 2a (59 mg, 0.26 mmol), gave 7d (45 mg, 94%) as a yel-
low oil.
FTIR (CH2Cl2, thin film): 2979 (s), 2229 (w), 1590 (m), 1339 (s),
1142 cm–1 (s).
1H NMR (250 MHz, CDCl3): δ = 1.10 (s, 12 H, CH3), 7.19–7.26 (m,
2 H, HAr), 7.39–7.45 (m, 3 H, HAr), 7.91 (d, J = 1.5 Hz, 1 H, HAr),
7.99 (d, J = 1.5 Hz, 1 H, HAr).
13C NMR (62.9 MHz, CDCl3): δ = 24.5, 84.5, 112.4, 117.4, 124.3,
127.7, 128.1, 129.1, 136.2, 137.0, 140.6, 151.9.
HRMS (AP+): m/z [M]+ calcd for C12H23BO2Si: 239.1639; found:
239.1644.
Methyl 3-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)-4-[(tri-
methylsilyl)methyl]benzoate (9) and Methyl 4-(4,4,5,5-Tetra-
methyl-1,3,2-dioxaborolan-2-yl)-3-[(trimethylsilyl)methyl]ben-
zoate (10)
Using general procedure 2, with 2-pyrone 1g (25 mg, 0.16 mmol)
and alkyne 2d (77 mg, 0.32 mmol), gave the product (22 mg, 41%)
as a yellow oil; inseparable mixture of 9/10 (6:1).
HRMS (EI+): m/z [M]+ calcd for C19H1911B79BrNO2: 383.0692;
found: 383.0680.
FTIR (CH2Cl2, thin film): 2976 (m), 2903 (w), 1722 (s), 1591 cm–1
(w).
1H NMR (250 MHz, CDCl3): δ (9) = 0.03 (s, 9 H, CH3), 1.36 (s, 12
H, SiCH3), 2.65 (s, 2 H, CH2SiCH3), 3.91 (s, 3 H, OCH3), 7.05 (d,
J = 8.0 Hz, 1 H, HAr), 7.95 (dd, J = 8.0, 2.0 Hz, 1 H, HAr), 8.44 (d,
J = 2.0 Hz, 1 H, HAr).
3-Bromo-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-4-
(trimethylsilyl)benzonitrile (7e) and 3-Bromo-4-(4,4,5,5-tetra-
methyl-1,3,2-dioxaborolan-2-yl)-5-(trimethylsilyl)benzonitrile
(8e)
Using general procedure 2, with 2-pyrone 1f (25 mg, 0.13 mmol)
and alkyne 2b (58 mg, 0.26 mmol), gave the product (46 mg, 96%)
as a clear oil; inseparable mixture of 7e/8e (1:1).
13C NMR (100.6 MHz, CDCl3): δ (9) = –1.6, 25.0, 27.5, 51.8, 83.6,
125.1, 128.6, 131.6, 137.8, 154.3, 164.6.
Synthesis 2012, 44, 1964–1973
© Georg Thieme Verlag Stuttgart · New York

FEATURE ARTICLE
[4+2] Cycloadditions of 2-Pyrones and Alkynylboronates
1973
HRMS (ES+): m/z [M]+ calcd for C18H2911BO4Si: 349.2002; found:
349.2006.
Vivat, J. F.; Plant, A.; Gomez-Bengoa, E.; Harrity, J. P. A.
J. Am. Chem. Soc. 2009, 131, 7762.
(9) The regiochemistry of compound 3a was elucidated by NOE
NMR spectroscopy, and that of compound 3b/4b was carried
out by chemical derivatization to the corresponding p- and
m-chlorophenols.
(10) The regiochemistry of 5a was assigned by inference in
comparison to the analogous cycloaddition reaction of 3,5-
dibromopyran-2-one.4a The regiochemistry of compound 5c
was elucidated by NOE NMR spectroscopy, but that of
compounds 5b/6b, 5d/6d, and 5e/6e was not established due
to the low levels of selectivity observed in these cases.
(11) The regiochemistry of compounds 7a, 7c/8c, 7d, and 7e/8e
was elucidated by NOE NMR spectroscopy.
(12) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter 1988,
37, 785.
(13) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.;
Bartberger, M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107,
11445.
Acknowledgment
We are grateful to Sanofi-Aventis, AstraZeneca and the EPSRC for
supporting this work.
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synthesis.SnuIomprifgtooatrinSugpIoi
nnfomartit
References
(1) Hall, D. G. Boronic Acids; Wiley-VCH: Weinheim, 2005.
(2) (a) Gilman, H.; Moore, L. O. J. Am. Chem. Soc. 1958, 80,
3609. (b) Ishiyama, T.; Murata, M.; Miyaura, N. J. Org.
Chem. 1995, 60, 7508. (c) Mkhalid, I. A. I.; Barnard, J. H.;
Marder, T. B.; Murphy, J. M.; Hartwig, J. F. Chem. Rev.
2010, 110, 890. (d) Del Grosso, A.; Singleton, P. J.; Muryn,
C. A.; Ingleson, M. J. Angew. Chem. Int. Ed. 2011, 50, 2102.
(3) (a) Davies, M. W.; Johnson, C. N.; Harrity, J. P. A. Chem.
Commun. 1999, 2107. (b) Ester, C.; Maderna, A.; Pritzkow,
H.; Siebert, W. Eur. J. Inorg. Chem. 2000, 1177. (c) Davies,
M. W.; Johnson, C. N.; Harrity, J. P. A. J. Org. Chem. 2001,
66, 3525. (d) Moore, J. E.; York, M.; Harrity, J. P. A. Synlett
2005, 860. (e) Gandon, V.; Leboeuf, D.; Amslinger, S.;
Vollhardt, K. P. C.; Malacria, M.; Aubert, C. Angew. Chem.
Int. Ed. 2005, 44, 7114. (f) Yamamoto, Y.; Ishii, J.-i.;
Nishiyama, H.; Itoh, K. J. Am. Chem. Soc. 2005, 127, 9625.
(g) Auvinet, A.-L.; Harrity, J. P. A.; Hilt, G. J. Org. Chem.
2010, 75, 3893. (h) Auvinet, A.-L.; Harrity, J. P. A. Angew.
Chem. Int. Ed. 2011, 50, 2769. (i) Iannazzo, L.; Vollhardt,
K. P. C.; Malacria, M.; Aubert, C.; Gandon, V. Eur. J. Org.
Chem. 2011, 3283.
(4) (a) Delaney, P. M.; Moore, J. E.; Harrity, J. P. A. Chem.
Commun. 2006, 3323. (b) Delaney, P. M.; Browne, D. L.;
Adams, H.; Plant, A.; Harrity, J. P. A. Tetrahedron 2008, 64,
866. (c) Kirkham, J. D.; Delaney, P. M.; Ellames, G. J.;
Row, E. C.; Harrity, J. P. A. Chem. Commun. 2010, 46,
5154.
(5) Afarinkia, K.; Vinader, V.; Nelson, T. D.; Posner, G. H.
Tetrahedron 1992, 48, 9111.
(6) Afarinkia, K.; Bearpark, M. J.; Ndibwami J. Org. Chem.
2005, 70, 1122.
(14) (a) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120,
215. (b) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41,
157. (c) Xu, X.; Alecu, I. M.; Truhlar, D. G. J. Chem. Theory
Comput. 2011, 7, 1667.
(15) R2 values computed including the out of range values
corresponding to >49:1 are 0.68 for B3LYP, 0.58 for RHF
and 0.35 for M06-2X and RMSEs are 0.82, 0.93 and 1.16
respectively. When the out-of-range values are excluded, the
R2 values are 0.59 (B3LYP), 0.49 (RHF), and 0.08 (M06)
with RMSEs of 0.59, 0.65, and 0.88, respectively.
(16) Štefane, B.; Perdih, A.; Pevec, A.; Šolmajer, T.; Kočevar, M.
Eur. J. Org. Chem. 2010, 5870.
(17) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(18) The cycloaddition of hex-1-yne derived alkyne 2c and
methyl coumalate (1g) provides a 3:1 regioselectivity for the
same major boronate isomer; see ref. 4b.
(19) The regiochemistry of compounds 9/10 was elucidated by
NOE NMR spectroscopy.
(20) (a) Cho, C. G.; Kim, Y. W.; Lim, Y. W.; Park, J. S.; Lee, H.;
Koo, S. J. Org. Chem. 2002, 67, 290. (b) Kvita, V.; Sauter,
H. Helv. Chim. Acta 1990, 73, 883. (c) Ashworth, I. W.;
Bowden, M. C.; Dembofsky, B.; Levin, D.; Moss, W.;
Robinson, E.; Szczur, N.; Virica, J. Org. Process Res. Dev.
2003, 7, 74.
(21) Brown, H. C.; Bhat, N. G.; Srebnik, M. Tetrahedron Lett.
1988, 29, 2631.
(7) (a) Kranjc, K.; Štefane, B.; Polanc, S.; Kočevar, M. J. Org.
Chem. 2004, 69, 3190. (b) Kranjc, K.; Kočevar, M. New J.
Chem. 2005, 29, 1027.
(8) (a) Gomez-Bengoa, E.; Helm, M. D.; Plant, A.; Harrity, J. P.
A. J. Am. Chem. Soc. 2007, 129, 2691. (b) Browne, D. L.;
(22) The carbon attached to boron is not visible in the 13C NMR
spectrum because of the quadrupole effect at boron: Brown,
C.; Cragg, R. H.; Miller, T.; Smith, D. O’N.; Steltner, A.
J. Organomet. Chem. 1978, 149, C34.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1964–1973