Reactions of lithiated aromatic heterocycles with carbon monoxide

Article abstract of DOI:10.1002/poc.661

The reaction of lithium derivatives of aromatic heterocycles with carbon monoxide was studied under several reaction conditions: 2-furyllithium afforded only one product, the tetrahydro-2-furaldehyde (25% yield). On the other hand, 2-thienyllithium and 5-methyl-2-furyllithium gave two and five products, respectively. Although in relatively low yields, formation of highly functionalized compounds takes place in a fast process under mild conditions. For 1,4-dihydroxy-1,4-bis(5-methyl-2-furanyl)butane-2,3-dione, reaction conditions were achieved to obtain this difficult to prepare derivative in 50% yield. To the best of our knowledge, this is the first report on the reaction of lithiated aromatic heterocycles with CO: evidence is given on the involvement of proton transfers in the reactions of 2-furyllithium and 2-thienyllithium, while the reaction of 5-methyl-2-furyllithium seems to proceed through an electron transfer as the first step. Copyright

Full text of DOI:10.1002/poc.661

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2003; 16: 569–576
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.661
Reactions of lithiated aromatic heterocycles with carbon
monoxide
Guadalupe E. Garc õÂ a Li nÄ ares and Norma S. Nudelman*
Departamento de Qu õ mica Org a nica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Pab. II, P. 3, Ciudad
Universitaria, 1428 Buenos Aires, Argentina
Received 27 December 2002; revised 22 March 2003; accepted 28 March 2003
ABSTRACT: The reaction of lithium derivatives of aromatic heterocycles with carbon monoxide was studied under
several reaction conditions: 2-furyllithium afforded only one product, the tetrahydro-2-furaldehyde (25% yield). On
the other hand, 2-thienyllithium and 5-methyl-2-furyllithium gave two and five products, respectively. Although in
relatively low yields, formation of highly functionalized compounds takes place in a fast process under mild
conditions. For 1,4-dihydroxy-1,4-bis(5-methyl-2-furanyl)butane-2,3-dione, reaction conditions were achieved to
obtain this difficult to prepare derivative in 50% yield. To the best of our knowledge, this is the first report on the
reaction of lithiated aromatic heterocycles with CO: evidence is given on the involvement of proton transfers in the
reactions of 2-furyllithium and 2-thienyllithium, while the reaction of 5-methyl-2-furyllithium seems to proceed
through an electron transfer as the first step. Copyright  2003 John Wiley & Sons, Ltd.
KEYWORDS: mechanism; carbonylation; organolithium; heterocycles; radicals; electron transfer
INTRODUCTION
continued, considerable effort in the development of
novel approaches to this class of compounds: we have
7
The chemistry of five-membered heterocyclic rings is a
field of active research. Some furan derivatives have
recently demonstrated the utility of organolithium and
1
8
organostannane reagents for the preparation of allyl-
achieved considerable technical importance, since it has
been found that the furan ring, in both its native and
reduced forms, occurs in a variety of commercially
furans and -thiophenes.
The reaction of organolithium compounds with carbon
monoxide has found wide application in organic syn-
2
9,10
important pharmaceuticals and flavor and fragrance
thesis,
and the reaction is interesting also from
compounds; they can also serve as pivotal intermediates
in synthetic strategy by virtue of their specific chemistry
and latent functionality. Thiophene has found wide-
mechanistic viewpoints. The synthetic usefulness of the
11
carbonylation of organolithium compounds has been
demonstrated. In addition, we have recently charac-
terized important intermediates in the reaction between
lithium dialkylamides and substituted formamides,
affording evidence for the structures of the lithium
carbamoyl intermediates and the mechanisms of their
3
spread use in the synthesis of biologically active species,
which contain this nucleus either as the central ring or as
1,4
part of a central fused ring system.
Renewed interest in natural product synthesis and for
in synthesis of heterocyclic compounds requires reliable
methods for the preparation of these starting materials.
The usual procedures have the drawback of causing some
resinification and degradation of the heterocyclic rings,
and some substituted furans and thiophenes have been
12
further reactions.
It was then of interest to examine the feasibility of the
reaction of lithiated derivatives of heteroaromatic
compounds with CO. The reaction was examined under
several conditions and evidence for the possible mech-
anisms of carbonylation of 2-furyllithium, 5-methyl-2-
furyllithium and 2-thienyllithium was sought. To our
knowledge, this is the first report about the mechanism of
reactions involving carbanions derived from aromatic
heterocycles.
5,6
obtained only by indirect means. This justifies the
*
Correspondence to: N. S. Nudelman; a) Facultad de Ciencias
Exactas, UBA, Pab II, P.3, Ciudad Universitaria, 1428 Buenos Aires,
Argentina. E-mail: nudelman@qo.fcen.uba.ar
b) Institute for Materials Chemistry and Engineering, Kyushu
University, 6-10-1 Hakozaki, Fukuoka, 812-8581 Japan. E-mail:
nudelman@ms.ifoc.kyushu-u.ac.jp
Contract/grant sponsor: National Agency for the Promotion of Science
and Technology (ANPCyT).
RESULTS AND DISCUSSION
Contract/grant sponsor: National Research Council (CONICET).
Contract/grant sponsor: Universidad de Buenos Aires.
The lithium derivatives of furans were prepared by direct
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

´
˜
G. E. GARCIA LINARES AND N. S. NUDELMAN
5
70
metalation of furan (1a) and of 2-methylfuran (1b) with
n-butyllithium (see Experimental section) [Eqn. (1)].
was observed; a higher yield of 3 in pure solvents was
obtained when the reaction was carried out in THF;
addition of HMPT produced a slight increase in the yield
of 3. The fact that 3 is not formed when the reaction was
carried out in hexane or diethyl ether suggests that a
proton donor such as THF is needed for the reaction to
occur. In fact, some amount of THF in hexane was
sufficient to obtain small yields of 3. This could be
attributed to a direct THF–lithium interaction, which
favors the proton transfer; this type of interaction has
been observed in other reactions of organolithium
ꢀ1†
Reaction of 2-furyllithium (2a) with CO
The carbonylation of 2-furyllithium, carried out in THF
at À78°C, afforded only one product, the tetrahydro-2-
furaldehyde (3) [Eqn. (2)] in low yield (20%). The
structure of 3 was confirmed by independent synthesis
13
compounds in THF.
Lack of formation of any other product in the reaction
suggests that the reaction of 2a with CO, under all
conditions employed, could involve only one mechanistic
pathway. (The mechanism is discussed below).
(
see Experimental section).
ꢀ2†
Reaction of 2-thienyllithium (4) with CO
Usually, the furan nucleus is very reactive and the
Compound 4 was prepared by metal–halogen exchange
between 2-iodothiophene and n-butyllithium. The car-
bonylation of 4, carried out in THF at À78°C, afforded
two main products, 2,5-dihydro-2-thiophencarboxalde-
hyde (5) and 2-oxo-2-(2-thienyl)acetaldehyde (6), to-
gether with small amounts of bithiophene (7) [Eqn. (3)].
reaction conditions must be carefully controlled in order
to avoid polymerization and ring-opening reactions. The
temperature of the reaction was the first variable studied
but a not significant effect on the product yield was found
in the range À78 to 40°C. This result suggests that the
activation energy for this reaction would be not very
high. On the other hand, on increasing the temperature,
more decomposition of the reagent and formation of
polymeric products were observed. Then, the lithium
reagent concentration was varied between 0.1 and 0.3 M
and no observable effect was detected.
ꢀ3†
The presence of 3 indicates that in the reaction mixture
there exists a source of hydrogen; therefore, it was crucial
to analyze the solvent effect; some results are shown in
Table 1. In hexane and diethyl ether no CO absorption
The structure of 5 was confirmed by independent
synthesis (see Experimental section). Variations in the
temperature had no significant effects on product
distribution.
Table 2 shows the results of the solvent effect: for the
reaction carried out in hexane or diethyl ether, 6 was the
only product obtained. Upon addition of THF, 5 was also
formed, suggesting that proton transfer from THF occurs.
On the other hand, the percentages of 5 and 6 were
maximum in pure THF.
a
Table 1. Reaction of 2-furyllithium (2a) with CO: solvent
effects
Yield (%)
Solvent
Hexane
T (°C)
VCO (ml)
1a
3
À78
–
–
100
100
94
–
–
6
5
9
6
0
Table 2. Reaction of 2-thienyllithium (4) with CO at 0°C:
solvent effects
Hexane–THF (1:1)
Hexane–THF (1:10)
Diethyl ether
a
À78
0
2.0
2.2
3.0
3.5
–
92
À78
89
Yield (%)
0
90
À78
0
100
100
78
–
–
Solvent
VCO (ml)
5
6
7
–
THF
À78
6.0
9.0
6.5
9.3
7.2
10.0
20
20
23
22
25
25
Hexane
6
7
10
10
24
25
–
–
12
15
20
25
35
35
–
–
2
2
3
–
0
80
Hexane–THF (1:1)
Hexane–THF (1:10)
Diethyl ether
THF
THF–HMPT (10:1)
THF–HMPT (1:1)
À78
0
75
5
78
–
17
20
À78
75
0
75
THF–HMPT (10:1)
a
2a] = 2 Â 10À3 mol; t = 40 min.
a
[4] = 2 Â 10À3 mol; tabs = 70 min.
[
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

REACTIONS OF LITHIATED AROMATIC HETEROCYCLES WITH CO
571
a
Table 3. Reaction of 5-methyl-2-furyllithium (2b) with CO in
THF: effect of temperature
Table 4. Reaction of 5-methyl-2-furyllithium (2b) with CO at
0°C: solvent effects
a
Yield (%)
Yield (%)
T (°C)
VCO (ml)
8
9
10
11
12
Solvent
V
CO (ml)
8
9
10 11 12
b
À78
12
16
12
10
15
12
10
10
35
50
35
15
5
5
5
–
5
2
–
–
3
2
–
–
Hexane
6
9
10
16
–
20
25
35
–
–
–
5
2
2
2
2
–
–
2
2
c
0
Hexane–THF (1:1)
Diethyl ether
THF
4
c
2
4
5
0
–
d
12 50
a
b
c
d
À3
a
[2b] = 2 Â 10À3 mol; t = 40 min.
[
t
t
t
2b] = 2 Â 10 mol.
abs = 50 min, 1b, 35%.
abs = 40 min, 1b, 28%.
abs = 40 min, 1b, 45%.
increase the yield of 9 by increasing the temperature were
unsuccessful. At higher temperatures the CO absorption
was lower, the organolithium reagent was not totally
recovered and polymeric products were obtained.
Reaction of 5-methyl-2-furyllithium (2b) with CO
The structure of 5-methyl-tetrahydro-2-furaldehyde
(
8) was confirmed by independent reduction of 5-
In THF at À78°C, two main products were formed, 5-
methyltetrahydro-2-furaldehyde (8) and 1,4-dihydroxy-
methyl-2-furaldehyde with H /Pd (C). Table 4 shows
2
the results of the carbonylation of 2b at 0°C in several
pure and binary solvents. In all cases the main product
was 9; the reduction products 8 and 10 were not observed
in pure hexane or diethyl ether. The maximum CO
absorption and the highest yields were obtained in pure
THF, indicating that proton transfer from THF occurs.
1
,4-bis(5-methyl-2-furanyl)butane-2,3-dione (9), to-
gether with small amounts of (5-methyl-tetrahydro-2-
furanyl)-(5-methyl-2-furanyl)methanol (10), 2-hydroxy-
2
-(5-methyl-2-furanyl)acetaldehyde (11) and bis(5-
methyl-2-furanyl)methanol (12) [Eqn. (4)].
Effect of radical traps
ꢀ4†
Earlier work by our group had shown that the car-
bonylation of aryllithium compounds occurs through the
11a,14
intermediacy of paramagnetic species.
In order to
gather information concerning radical involvement in the
reaction of 2b with CO, the reaction was carried out in the
presence of a suitable radical trap (Table 5). Quinones
and hydroquinones can produce stable radicals by one-
Table 3 shows the effect of temperature on the reaction
rate and product distribution of the carbonylation carried
out in THF. It can be observed that the reaction proceeds
slightly faster at 0°C and the CO absorption was higher
than at À78°C. The yield of 9 increased at 0°C, whereas
15
electron uptake or donation and they were previously
used to test the involvement of radicals in the carbonyla-
16
tion of lithium amides; therefore, they were considered
suitable to test the presence of radicals in the carbonyla-
tion of 2b.
8
and 10–12 were obtained in poor yields. Attempts to
a
Table 5. Reaction of 5-methyl-2-furyllithium (2b) with CO in THF in the presence of radical inhibitors
Yield (%)
Inhibitor
None
T (°C)
[Trap]:[2b]
tT (min)
VCO (ml)
8
9
10
11
12
À78
–
–
50
40
25
25
25
25
25
25
12
16
15
12
12
10
13
10
12
10
35
50
15
10
15
10
15
10
5
5
4
3
3
3
4
3
5
2
–
–
–
–
–
–
3
2
–
–
–
–
–
–
0
b
p-Hydroquinone
À78
0
1:10
1:10
1:10
1:10
1:10
1:10
4.0
4.0
3.9
4.0
4.0
4.0
b
Quinhydrone
À78
0
b
Iodine
À78
0
a
À3
[
[
2b] = 1.3 Â 10 mol.
b
1b] = 73%.
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

´
˜
G. E. GARCIA LINARES AND N. S. NUDELMAN
5
72
The reaction was examined in THF at À78 and 0°C in
hydrolysis would produce tetrahydro-2-furaldehyde (3).
The fact that 3 is not obtained when the reaction is carried
out in hexane or diethyl ether suggests that proton
transfer from THF would occur.
To test the hypothesis that THF should be the hydrogen
source, the reaction between 2a and CO was carried out
the presence of radical scavengers, namely p-hydroqui-
none (13), quinhydrone (14), and iodine (15). As shown
in Table 5, a considerable decrease in the CO absorption
in the presence of 13–15 occurs, and 73% of the reagent
was recovered as 1b. It is important to note the drastic
lowering of the yield of 9 and the complete absence of 11
and 12, whereas 8 and 10 were obtained in similar yields
which were found when the reaction was carried out in
the absence of inhibitors. On the other hand, in the
presence of iodine in a ratio of 1:1 (not shown), total
inhibition of carbonylation was observed. These observa-
tions indicate that the reaction of 2b with CO partially
involves a radical pathway and that 13–15 must have
trapped the radical precursors of 9, 11 and 12. The fact
that the yields of 8 and 10 were similar under both
conditions suggests that the pathway leading to these
compounds does not involve radical intermediates.
in THF-d . The determination of the deuterium content in
8
3 revealed approximately 80% deuterium incorporation
(see the Experimental section). This result is consistent
with the proposed mechanism.
Proposed mechanism for the reaction of
2-thienyllithium (4) with CO
The absence of effects in the presence of radical traps
indicates that the reaction does not occur by an electron
transfer mechanism. The reaction should rather lead to
the formation of 5 by a pathway similar to that suggested
for the formation of 3 (Scheme 2), i.e. addition of CO to
the dienic system would give the reduced product; in this
case, probably owing to the lower electronegativity of
sulfur, only partial reduction of the aromatic ring occurs,
giving a thienyl intermediate, 19, similar to 17. Also in
Proposed mechanism for the reaction of
-furyllithium (2a) with CO
2
Interestingly, no reaction products with the intact furan
aromatic ring were found. This, together with the absence
of radical reaction intermediates, suggests that a mech-
anism different from that previously reported for the
carbonylation of aryllithiums could be involved. Furan is
an excellent diene and the [4 ‡ 2] cycloaddition to
dienophiles was one of the earliest described examples of
this case, the reaction was carried out in THF-d , and
8
almost complete deuteration was observed (see Experi-
mental section).
On the other hand, the higher aromaticity of thiophene
4,19
vs furan
would allow an alternative route for
carbonylation, similar to that of aryllithiums, affording
the formation of product 6 with the intact thiophene ring
[see Eqn. (3)]. As outlined in Scheme 3, CO insertion in
the C—Li bond would give first the acyllithium
intermediate 20, which by reaction with another molecule
of CO forms the intermediate 21, precursor of the doubly
carbonylated product. Similar double carbonylation has
been observed in the reaction between lithium amides
and CO giving dialkylglyoxylamides and tetraalkyl-
1,17
the Diels–Alder reaction;
e.g. furans undergo cyclo-
addition with simple dienophiles such as ethylene, singlet
18
oxygen and acrylate. Therefore, addition of CO to 2a
giving the intermediate 16 is proposed as the first step in
Scheme 1. The subsequent proton transfer from the
solvent gives the intermediate 17; formation of 2,5-di-
hydrofuran derivatives has been observed in reactions of
19
diverse electrophilic reagents with furans. Another
11b
proton transfer should give the intermediate 18, which on
tartronamides in excellent yields.
Scheme 1
Scheme 2
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

REACTIONS OF LITHIATED AROMATIC HETEROCYCLES WITH CO
573
electron transfer mechanism was previously proved to
occur in the reaction of phenyllithium with CO to give an
14
acyllithium intermediate.
The different reactivity pattern exhibited by 2b
compared with 2a should be due to the methyl group in
the a-position that prevents the 2,5-addition of CO, favors
the electron transfer to CO and stabilizes the radical
cation 22. Radical cations of aromatic five-membered
heterocycles could be prepared previously by choosing a
Scheme 3
Proposed mechanism for the reaction of
-methyl-2-furyllithium (2b) with CO
22
5
suitable system and, recently, the radical cations of
furans, thiophenes and pyrroles and their oligomers were
easily detected by EPR when the adjacent positions to the
The results obtained in the carbonylation of 2b indicate a
complex reaction scheme through, at least, two different
routes: radical and polar pathways. The observed
decrease in the CO absorption when the reaction was
carried out in the presence of radical inhibitors suggests
the partial involvement of electron transfer in the first
step. Organolithium compounds have been shown to be
2
3
heteroatom were occupied by alkyl substituents.
The intermediate 23 can take up another molecule of
CO to give 24, which can act as an electron acceptor from
another molecule of 2b forming a new radical cation–
radical anion pair, 25. Our previous ab initio calculations
24
on carbamoyllithiums and experimental evidence with
20
14
good one-electron donors to carbonyl compounds and
CO is a known electron-deficient compound and an
effective one-electron oxidant toward a variety of
arylacyl- and carbamoyllithiums indicate that these
intermediates exhibit a strong oxycarbene character. The
radical anion in 25 can partially escape the cage being the
precursor of 11. A resonance structure of 25 is written as
a radical anion; coupling of two molecules would form
26, precursor of 9.
In order to gather more evidence about the mechanism
of the reaction, the carbonylation of 2b was carried out in
THF at 0°C and acetic anhydride was added prior to
work-up. Only one main compound, 29, was obtained,
which was characterized as the tetraacetyl derivative 1,2-
bis-2-(5-methylfuryl)-1,2,3,4-tetraacetylbutene (29). The
isolation and characterization of 29 confirmed the
structure of the intermediate 26 and shed new light on
the mechanism of the reaction of 2b with CO.
2
1
aromatic radical ions. A plausible pathway involving
radical ions is outlined in Scheme 4. As shown, an
electron transfer from 5-methyl-2-furyllithium (2b) to
CO generates the radical cation–radical anion pair 22.
The radical pair 22 can then react within the cage
affording the acyllithium intermediate 23. A similar
On the other hand, the acyllithium intermediate 23 can
react with 2b yielding intermediate 27, which gives 12 on
hydrolysis.
Compound 8 could be formed by the same pathway
that was suggested for the formation of 3 from 2a with no
radical involvement, as indicated by its regular formation
in the presence of radical traps. When the reaction was
carried out in THF-d , 76% deuterium incorporation was
8
observed (see Experimental section), confirming that the
proton transfer occurs from THF. Further reaction with
2
b gives 28, precursor of 10 (Scheme 4).
Very little has been published about the mechanism of
the reaction of furyl- and thienyllithium and to the best
of our knowledge this is the first report on the reaction of
CO with heteroaryllithium compounds.
Scheme 4
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

´
˜
G. E. GARCIA LINARES AND N. S. NUDELMAN
5
74
EXPERIMENTAL
for at least 30 min and then cooled in a dried nitrogen
atmosphere. The vial was capped with a non-air stopper,
it was alternately evacuated and filled with dry nitrogen
several times, then put into an ice-bath with vigorous
magnetic stirring. A lithium reagent solution (0.5 M) in
THF was added by cannula and the stirred solution was
exposed to carbon monoxide at ca 1013 mbar. Carbon
monoxide was generated from 98% formic acid and
concentrated sulphuric acid at 100–110°C. The gas was
routinely passed through a column of potassium hydro-
xide before entering the system. The total amount of gas
was measured by means of a burette attached to the CO
generator provided with a mercury levelling bulb. Once
the reaction was completed, 10% HCl in THF was added
All reactions involving organolithium reagents were
carried out by using standard techniques for the
manipulation of air- and water-sensitive compounds.
GLC analyses were carried out on a Hewlett-Packard
Model 5890 Series II Plus gas chromatograph using an
HP-5 column with temperature programming from 60 to
25
2
50°C. NMR spectra were recorded on a Bruker 200
spectrometer. Mass spectra were recorded on a BG Trio-2
spectrometer. The percentage deuterium incorporation in
products, when applicable, was calculated on the basis of
mass spectrometric (MS) data.
and the dried (MgSO ) organic phase was quantitatively
4
Solvents and reagents
analyzed by GC and GC–MS. 2,5-Dihydro-2-thiophene-
carboxaldehyde (5), 2-oxo-2-(2-thienyl)acetaldehyde (6),
5-methyltetrahydro-2-furaldehyde (8) and 1,4-dihydroxi-
1,4-bis(5-methyl-2-furanyl)butane-2,3-dione (9), ob-
tained by carbonylation of respective reagents, were
isolated by neutral aluminum oxide chromatography
using hexane–ethyl acetate mixtures as eluents.
Materials. Distilled THF, diethyl ether and hexane were
refluxed over sodium benzophenone ketyl until a dark
blue solution was obtained and then distilled immediately
before use under dry oxygen-free nitrogen. n-Butyl-
lithium was prepared as described previously. The
concentration of BuLi was determined by the double
26
12
titration method described previously. All glassware,
syringes and needles were dried in a vacuum oven and
cooled in a desiccator. Furan and 2-methylfuran were left
over sodium strings for 4–5 h, then refluxed and distilled
over sodium and stored in sealed ampoules under
nitrogen, which were opened immediately prior to use.
In the presence of radical inhibitors. The reactions
were carried out in a flask containing weighed amounts of
the inhibitor. The desired volume of the lithium reactant
solution was added by syringe and the CO absorption was
carried out similarly to the general procedure. The rate of
reaction was measured by monitoring the rate of CO
absorption. Once the reaction was complete, 10% HCl
Lithium furans. Cooled (0°C) n-BuLi (10 ml, 1 M in
hexane) was syringed into a non-air stopper-capped tube
under a nitrogen atmosphere, and the corresponding furan
was added. The dried (MgSO ) organic layer was
analyzed by GC.
4
(
15 mmol) was added; it was kept at À18°C. 2-Furyl-
Tetrahydro-2-furaldehyde (3). This compound was
independently synthesized by hydrogenation of 2-
furaldehyde (10 mmol in 20 ml of CH Cl ) with H (2
lithium precipitated after 72 h and 5-methyl-2-furyl-
lithium precipitated after 15 days.
2
2
2
atm) in the presence of 10% Pd(C). The reaction mixture
was stirred for 5 h at room temperature. The catalyst was
separated by filtration and the solvent removed by
distillation at reduced pressure. MS: m/z (relative
2
-Thienyllithium. Cooled (0°C) n-BuLi (10 ml, 1 M in
hexane) was syringed into a non-air stopper-capped tube
under a nitrogen atmosphere and 2-iodothiophene
(
10 mmol) was added; the lithium reagent precipitated
intensity,%), 100 (10.5), 99 (8.2), 71 (100), 43 (88.2).
1
instantly.
H NMR (Cl CD): ꢀ 9.10 (s, 1H), 3.65 (m, 2H), 3.18 (m,
3
13
The organolithium compounds were characterized by
C NMR spectroscopy; the lithiation causes a clear shift
1H), 2.55 (m, 2H), 2.0 (m, 2H). C NMR (Cl CD): ꢀ
3
1
3
198.6, 82.2, 67.2, 26.8, 24.9.
(
5–10 ppm) of the signal of the lithiated carbon and of Ca
When the reaction of 2a with CO was carried out in
THF-d , tetrahydro-2-furaldehyde-d5 (3-d ) was ob-
served. MS: m/z (relative intensity,%), 105 (8.4), 103
relative to the starting material.
8
5
13
2
a: C NMR (Cl CD), ꢀ 155.2 (C-1), 126.7 (C-2),
3
1
11.5 (C-3), 145.4 (C-4).
(5.3), 100 (2.1), 75 (100), 71 (22.2), 45 (65.0), 43 (17.5).
13
2
b: C NMR (Cl CD), ꢀ 157.5 (C-1), 125.9 (C-2),
3
1
10.2 (C-3), 154.3 (C-4), 13.8 (C-5).
2,5-Dihydro-2-thiophenecarboxaldehyde (5). This
compound was independently synthesized by hydrogena-
tion of 2-thiophenecarboxaldehyde (10 mmol in 20 ml of
CH Cl ) with H (2 atm) in the presence of 10% Pd(C).
The reaction mixture was stirred for 2 h at room
temperature. The catalyst was separated by filtration
and the solvent removed by distillation at reduced
pressure. Compound 5 was isolated by neutral aluminum
Reaction of lithiated derivatives of aromatic
heterocycles with CO
2
2
2
General procedure. A vial containing a PTFE-coated
stirring bar was heated in a vacuum oven at 130–150°C
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

REACTIONS OF LITHIATED AROMATIC HETEROCYCLES WITH CO
575
oxide chromatography from the reaction mixture. MS:
The main fraction corresponded to 29; white solid
recrystallized from ethanol, m.p. 117°C. MS: m/z
(relative intensity,%), 446 (10.0), 328 (2.0), 223 (5.2),
m/z (relative intensity,%), 114 (8.61); 113 (2.0), 85
1
(
(
(
45.5), 84 (10.6), 71 (32.2), 57 (45.5), 43 (100). H NMR
1
Cl CD): ꢀ 9.20 (s, 1H), 5.80 (m, 1H), 5.34 (m, 1H), 4.62
d, 1H, J = 3.5 Hz), 3.50 (m, 2H). C NMR (Cl CD): ꢀ
210 (2.3), 71 (34.5), 55 (76.0), 43 (100). H NMR
3
13
(Cl CD): ꢀ 6.25 (d, 1H, J = 3.5 Hz), 5.90 (dd, 1H, J = 3.5,
3
3
13
2
00.0, 130.2, 124.1, 66.5, 30.5.
0.9 Hz), 2.15 (d, 3H, J = 0.9 Hz). C NMR (Cl CD): ꢀ
3
When the reaction of 4 with CO was carried out in
170.0, 165.2, 154.5, 148.7, 143.2, 136.0, 117.6, 98.5,
20.1, 13.6.
THF-d , 2,5-dihydro-2-thiophenecarboxaldehyde-d (5-
8
3
d3) was observed. MS: m/z (relative intensity,%), 117
6.6), 115 (1.5), 114 (2.0), 113 (0.1), 87 (37.2), 85 (0.5),
2 (20.2), 71 (2.3), 58 (37.1), 57 (9.0), 44 (100), 43
81.0).
(
7
(
CONCLUSIONS
This paper reports experimental details about the
carbonylation of lithium derivatives of aromatic hetero-
cycles. The reaction of 2-furyllithium with CO in THF
afforded only one product, 2-tetrahydrofuraldehyde,
which was formed by CO absorption and subsequent
proton transfer from the solvent. 5-Methyl-2-furyllithium
was more reactive and five products were formed by
reaction with CO. Evidence is given that some of them
derive from radical intermediates. The different reactiv-
ity patterns of the two lithium reagents show that the
presence of a methyl group in the a-position of the furan
ring results in a good one-electron donor reacting by an
electron transfer mechanism. In contrast, the reaction of
2
-Oxo-2-(2-thienyl)acetaldehyde (6). This compound
(
oil) was characterized by analogy with dialkylglyoxyl-
11b
amides, and was isolated from the reaction mixture by
neutral aluminum oxide chromatography. MS: m/z
(
relative intensity,%), 140 (31.4), 139 (1.4), 111 (7.2),
1
9
7 (100). H NMR (Cl CD): ꢀ 9.60 (s, 1H), 7.90 (dd, 1H,
3
J = 4.0, 1.6 Hz), 7.60 (dd, 1H, J = 5.1, 1.6 Hz), 6.87 (dd,
13
1
1
H, J = 5.1, 4.0 Hz). C NMR (Cl CD): ꢀ 190.6, 181.0,
45.5, 134.4, 132.7, 128.2.
3
5
-Methyltetrahydro-2-furaldehyde (8). This com-
pound was independently synthesized by hydrogenation
of 5-methyl-2-furaldehyde (10 mmol in 20 ml of CH Cl )
with H (2 atm) in the presence of 10% Pd(C). The
reaction mixture was stirred for 5 h at room temperature.
The catalyst was separated by filtration and the solvent
removed by distillation at reduced pressure. MS: m/z
2
2
2-thienyllithium led to the formation of 2-dihydrothio-
2
phenecarboxaldehyde and a double-carbonylated pro-
duct, with no evidence of radical intermediates. This
work sheds some light on the mechanism of an interesting
set of reactions, reported for the first time.
(
(
relative intensity,%), 114 (6.0), 113 (1.3), 99 (1.5), 85
10.2), 71 (20.1), 57 (46.5), 43 (100). H NMR (Cl CD):
1
3
ꢀ
2
ꢀ
9.08 (s, 1H), 3.73 (m, 1H), 3.22 (m, 1H), 2.55 (m, 2H),
13
Acknowledgements
.2 (m, 2H), 1.37 (d, 3H, J = 5.2 Hz). C NMR (Cl CD):
3
195.6, 80.5, 73.8, 32.6, 25.1, 21.6.
When the reaction of 2b with CO was carried out in
Financial support from the National Agency for the
Promotion of Science and Technology (ANPCyT), from
the National Research Council (CONICET) and from the
Universidad de Buenos Aires, Argentina, is gratefully
acknowledged.
THF-d , 5-methyltetrahydro-2-furaldehyde-d (8-d ) was
observed. MS: m/z (relative intensity,%), 119 (4.5), 117
8
5
5
(
(
0.8), 114 (1.2), 104 (1.8), 99 (0.5), 89 (7.8), 85 (2.2), 74
15.6), 71 (5.0), 59 (30.0), 57 (11.5), 44 (100), 43 (76.4).
1
2
,4-Dihydroxy-1,4-bis(5-methyl-2-furanyl)butane-
,3-dione (9). Compound 9 (oil) was isolated by neutral
REFERENCES
aluminum oxide chromatography from the reaction
mixture. MS: m/z (relative intensity,%): 278 (2.0), 235
1. (a) Benassi R. In Comprehensive Heterocyclic Chemistry II, vol. 2,
Katritzky AR, Rees CW, Scriven EFV (eds). Pergamon Press:
Oxford, 1996; chapt. 5; (b) Padwa A, Bur SK, Danca DM, Ginn
JD, Lynch SM. Synlett 2002; 851–862.
(
(
2.2), 217 (2.4), 183 (5.0), 109 (16.3), 91 (88.5), 71
1
60.6), 55 (95.0), 43 (100). H NMR (Cl CD): ꢀ 6.40 (d,
3
1
1
H, J = 3.8 Hz), 5.3 (s, 1H), 5.25 (dd, 1H, J = 3.8,
.1 Hz), 4.05 (s, 1H), 2.2 (d, 3H, J = 1.1 Hz). C NMR
2. (a) Francesconi I, Wilson WD, Tanious FA, Hall JE, Bender BC,
Tidwell RR, McCurdy D, Boykin DW. J. Med. Chem. 1999; 42:
13
2
260–2265; (b) Boselli E, Grob K, Lercker G. J. Agric. Food
(
Cl CD): ꢀ 197.0, 156.2, 151.1, 107.2, 85.6, 73.2, 14.0.
3
Chem. 2000; 48: 2868–2873; (c) Trost BM, Flygare JA. J. Org.
Chem. 1994; 59: 1078–1082.
. (a) Tietze LF, Lohmann JK. Synthesis 2002; 2083–2085; (b)
Efremov I, Paquette LA. J. Am. Chem. Soc. 2000; 122: 9324–9325;
3
1
,2-Bis-2-(5-methylfuryl)-1,2,3,4-tetraacetylbutene
(
29). The reaction of 1b with CO was carried similarly to
(
c) Grimaldi T, Romero M, Pujol MD. Synlett 2000; 1788–1792.
the general procedure. Once the reaction was complete,
4. Szajda M, Lam JN. In Comprehensive Heterocyclic Chemistry II,
vol. 2, Katritzky AR, Rees CW, Scriven EFV (eds). Pergamon
Press: Oxford, 1996; chapter 9.
. (a) Smitrovich JH, Woerpel KA. Synthesis 2002; 2778–2785; (b)
Ballini R. Bosica G, Fiorini D, Giarlo G. Synthesis 2001; 2003–
acetic anhydride was added. The dried (MgSO ) organic
4
layer was analyzed by GC and the reaction mixture was
separated by neutral aluminum oxide chromatography.
5
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576

´
˜
G. E. GARCIA LINARES AND N. S. NUDELMAN
5
76
2
2
006; (c) Nicola T, Vieser R, Eberbach W. Eur. J. Org. Chem.
000; 527–532.
16. Garc ´ı a Li n˜ ares GE, Nudelman NS. An. Asoc. Quim. Arg. 1996;
619–624.
6
. Trost BM, Van Vranken DL. Chem. Rev. 1996; 96: 395–422.
. Garc ´ı a Li n˜ ares GE, Carro C, Nudelman NS. Att. Phys. Org. Chem.
17. Diels O, Alder K. Chem. Ber. 1929; 62: 554–556.
18. (a) Sargent MV, Dean FM. In Comprehensive Heterocyclic
Chemistry I, vol. 4. Pergamon Press: Oxford, 1984; 599; (b) Joule
AJ, Mills K, Smith GF. In Heterocyclic Chemistry (3rd edn).
Chapman and Hall: London, 1996.
19. Paquette LA. Principles of Modern Heterocyclic Chemistry.
Benjamin/Cumming: Menlo Park, CA, 1987.
7
1
998; 460–477.
8
9
. Carro C, Nudelman NS. Synlett 1999; 1942–1944.
. Nudelman NS. Carbonylation of main-group organometallic
compounds. In The Chemistry of Double-Bonded Functional
Groups, Patai S (ed). Wiley: Chichester, 1989; 799–962.
1
0. Seyferth D, Hui RC, Wang WL. J. Org. Chem. 1993; 58: 5843–
845, and references cited therein.
1. (a) Nudelman NS, Schulz HG. J. Chem. Soc., Perkin Trans. 2
20. (a) Neville AG, Brown CE, Rayner DM, Lusztyk J, Ingold KU. J.
Am. Chem. Soc. 1991; 113: 1869–1870; (b) Reich AJ, Green DP,
Phillips NH. J. Am. Chem. Soc. 1989; 111: 3444–3445; (c)
Maruyama K, Katagiri T. J. Phys. Org. Chem. 1991; 4: 381–385.
21. (a) Ashby EC. Acc. Chem. Res. 1988; 21: 414–421; (b) Ashby EC,
Pham T. J. Org. Chem. 1987; 52: 1291–1300.
5
1
1
999; 2761–2767; (b) Nudelman NS, Lewkowicz ES, P e´ rez DG.
Synthesis 1990; 917–920.
1
2. (a) Nudelman NS, Garc ı´ a Li n˜ ares GE. J. Org. Chem. 2000; 65:
1
629–1635; (b) Nudelman NS, Schulz HG, Garc ı´ a Li n˜ ares GE,
22. Davies AG. Chem. Soc. Rev. 1993; 299–309.
Bonatti A, Boche G. Organometallics 1998; 17: 146–150.
3. Cainelli G, Giacomini D, Galletti P, Marini A. Angew. Chem., Int.
Ed. Engl. 1996; 35: 2849–2853, and references cited therein.
4. Nudelman NS, Doctorovich F. Tetrahedron Lett. 1994; 35: 4651–
23. Fajar ı´ Ll, Brillas E, Alem a´ n C, Juli a´ L. J. Org. Chem. 1998; 63:
5324–5331.
24. Viruela-Martin P, Viruela-Martin R, Tom a´ s-Vert F, Nudelman NS.
J. Am. Chem. Soc. 1994; 116: 10110–10116.
25. Shriver DF, Drezdzon MA. The Manipulation of Air-Sensitive
Compounds (2nd edn). Wiley: New York, 1986.
26. Trzupek LS, Newirth TL, Kelly EG, Nudelman NS, Whitesides
GM. J. Am. Chem. Soc. 1973; 95: 8118–8133.
1
1
1
4
655.
5. (a) Rak SF, Miller LL. J. Am. Chem. Soc. 1992; 114: 1388–1394;
b) Arnoldi C, Citterio A, Minisci F. J. Chem. Soc., Perkin Trans. 2
983; 531–535.
(
1
Copyright  2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 569–576