Unmodified nano-powder magnetite or iron(III) oxide catalyze the easy and fast synthesis of 4-substituted-4H-pyrans

Article abstract of DOI:10.1055/s-0030-1261162

Inexpensive and commercially available nano-power magnetite or iron(III) oxide can be used as catalyst in a new, straightforward, and fast protocol for the construction of 4-substituted-4H-pyrans. The reaction implies a tandem process, involving an aldol

Full text of DOI:10.1055/s-0030-1261162

LETTER
2017
Unmodified Nano-Powder Magnetite or Iron(III) Oxide Catalyze the Easy and
Fast Synthesis of 4-Substituted-4H-Pyrans
C
atalyze
d
Synthe
a
sis of 4-Su
f
bstitute
d
a
-4
H
-pyransel Cano, Diego J. Ramón,* Miguel Yus
Departamento Química Orgánica and Instituto de Síntesis Orgánica, Universidad de Alicante, Apdo. 99, 03080 Alicante, Spain
Fax +34(96)5903549; E-mail: djramon@ua.es
Received 13 May 2011
Dedicated to Prof. Dr. Alfredo Ricci on the occasion of his retirement
Thus, we report here a simple, mild, and fast protocol to
Abstract: Inexpensive and commercially available nano-power
perform this tandem process using unmodified nanopow-
er magnetite or iron(III) oxide as catalysts.
magnetite or iron(III) oxide can be used as catalyst in a new,
straightforward, and fast protocol for the construction of 4-substi-
tuted-4H-pyrans. The reaction implies a tandem process, involving
an aldol condensation, a Michael-type addition, and a dehydrating
annulation.
In order to optimize the reaction conditions, we studied
the reaction between methyl 3-oxobutanoate (1a) and 4-
bromobenzaldehyde (2a) to give the corresponding com-
pound 4a,12 as depicted in Table 1.
Key words: annulation, heterocycles, heterogeneous catalysis,
iron, pyrans
The uncatalyzed reaction using an excess of acetyl chlo-
ride (300 mol%) only produced a small amount of the al-
dol condensation product 3a, with the expected pyran
Pyrans and their benzo derivatives occupy an important
being not detected. The reaction performed using only na-
area in natural chemistry, especially in plant life. The sig-
nopower magnetite also failed, recovering both reagents
nificant pharmacological and ecological activity shown
unmodified and pointing out the need of both agents to ob-
by these systems has stimulated a great synthetic effort.1
tain the expected product. So, when the reaction was re-
However, the preparation of the parent 4H-pyrans has
peated in the presence of magnetite and acetyl chloride
been less developed, probably due to the presence of the
(600 mol%), the expected pyran 4a was obtained after one
dienol ether functionality and to the absence of aromatic
day, together with the byproduct 3a (compare entries 1–3
character, which makes these types of compounds less
in Table 1). Then, we studied the influence of other pa-
stable. Beside these inconveniences, and since some of
rameters such as the amount of acetyl chloride, with the
these architectures have important biological activities,
presence of two equivalents being optimal (Table 1, entry
including anticoagulant,2 antioxidant,3 and as IKCa chanel
4). The decrease of the amount of catalyst produced a de-
blocker,4 several ways to build these compounds have
crease in the yield (Table 1, entries 5 and 6). The use of
been introduced.5
other solvents such as dioxane, diethyl ether or hexane re-
Among the different synthetic strategies used, the simple
cascade process involving an aldol condensation, a
Michael-type addition, and a final annulation dehydration
has been scarcely used. For instance, the initial conditions
to perform this task involved stoichiometric amounts of
ZnCl2, excess of acetic acid as dehydrating agent, and ace-
tic anhydride as solvent during several weeks at room
temperature rendering the 4-substituted-4H-pyyrans in
30–50% yields.6 A further improvement implied the use
of ultrasound at 50 °C in acetic anhydride, which reduced
the reaction time to several days and increased the yield
up to 75%, avoiding the use of both the catalyst and the
dehydrating agent.7 However, both protocols involve very
harsh reaction conditions.8
duced significantly the yield (Table 1, entries 7–9). The
reaction performed under basic conditions failed (Table 1,
entry 10), and other acid chloride derivatives gave in all
cases worse results than those obtained with acetyl chlo-
ride (Table 1, compare entries 4 and 11–16), even in the
case of acetic anhydride (Table 1, entry 17). Then, we re-
peated the reaction with commercial micropowder mag-
netite, the product being obtained with lower yield
(Table 1, entry 18). Then, we faced the problem which
iron center could be responsible for the reaction, perform-
ing the reaction with iron oxide. In the case of iron(II), the
reaction only produced the aldol condensation product 3a,
with the expected pyran being undetected (Table 1, entry
19). However, in the case of nanopowder iron(III) oxide,
we obtained practically the same result as using magnetite
(Table 1, compare entries 4 and 20). Although in the mag-
netite the amount of iron(III) centers is half the number of
those in pure iron(III) oxide, the reduction of the amount
of catalyst decreased the yield of desired product 4a, even
more than for magnetite (Table 1, compare entries 5, 6,
21, and 22). Then, we thought that the different iron ox-
ides could only play a marginal role as catalyst, being only
the source of the corresponding iron chloride. To prove
In our ongoing project on the use of magnetite9,10 and
metal-impregnated magnetite,11 we realized that iron ox-
ides could be an excellent catalyst for the aforementioned
process avoiding these previous drawbacks.
SYNLETT 2011, No. 14, pp 2017–2020
x
x
.x
x
.2
0
1
1
Advanced online publication: 10.08.2011
DOI: 10.1055/s-0030-1261162; Art ID: S04011ST
© Georg Thieme Verlag Stuttgart · New York

2018
R. Cano et al.
LETTER
Table 1 Optimization of the Reaction Conditionsa
this hypothesis, we repeated the reaction using iron(III)
and iron(II) chloride, and the yields of compound 4a were
significantly lower than using the related oxides, with the
initial formed aldol condensation derivative 3a being the
main isolated product in both cases (Table 1, entries 23
and 24). Finally, it should be pointed out that both iron ox-
ides could be easily removed from the reaction media just
by filtration.
O
O
O
O
MeO
OMe
1a
3a
cat.
Br
O
+
+
Br
dehydrating agent,
solvent, 25 °C
O
Br
After finding that both, magnetite or iron(III) oxide, could
be successfully used in the tandem formation of 4H-pyr-
ans, we tested different impregnated metallic oxides on
magnetite with the hope that these oxides could improve
the obtained results (Table 2). The reactions performed
under similar reaction conditions to those presented in en-
try 4 of Table 1 with different catalysts gave always worse
results, independently of the metallic oxide used (Table 2,
entries 1–7). Only the palladium derivatives gave similar
results.
H
O
2a
MeO
OMe
4a
O
Entry Catalyst
(mol%)
Dehydrating Solvent
agent
Time Yield of
(h)
72
72
24
3
3a/4a (%)
18/0
0/0
1
2
–
MeCOClb
–
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
1,4-dioxane
Et2O
Fe3O4 (65)
Once we realized that both magnetite and iron(III) oxide
were excellent catalysts for this transformation, we stud-
ied the scope of the reaction (Table 3).
3
Fe3O4 (65) MeCOClc
Fe3O4 (65) MeCOCl
Fe3O4 (32) MeCOCl
22/73
3/96
10/85
26/45
30/28
56/9
7/64
0/0
4
The isolated yields of pyrans 4 were similar independently
of the aromatic aldehyde used, with electron-withdrawing
groups (Table 3, entries 1 and 2), unsubstituted rings
(Table 3, entry 3), or electron-donating groups (Table 3,
entries 4 and 5) being well tolerated. It should be pointed
out that in the case of 4-hydroxybenzaldehyde, the corre-
sponding hydroxy pyran 4e was isolated, not detecting the
related acylated pyran. The reaction with a sterically more
congested 2-naphthylcarbaldehyde gave a lower yield
(Table 3, entry 6). The results obtained using aliphatic al-
dehydes were good for cyclic and acyclic derivatives
(Table 3, entries 7 and 8). The change of the ester moiety
in the 1,3-dicarbonyl reagent did not have any influence
on the achieved results (Table 3, compare entries 1, 4, 9,
and 10). Instead of b-keto esters, other 1,3-dicarbonyl
compounds could be used as pentane-2,4-dione, obtaining
the corresponding pyran 4k with similar results (Table 3,
entry 11). The substitution at the 2,6-positions of the pyr-
an could be changed only by using the appropriated sub-
stituted 1,3-dicarbonyl compounds. Thus, the reaction of
methyl 3-oxopentanoate with 4-bromobenzaldehyde (2a)
gave the expected compound 4l with the same practical
results, independently of the catalyst used (Table 3, entry
12).
5
3
6
Fe3O4 (7)
MeCOCl
3
7
Fe3O4 (65) MeCOCl
Fe3O4 (65) MeCOCl
Fe3O4 (65) MeCOCl
Fe3O4 (65) NaOH
Fe3O4 (65) PhCOCl
3
8
3
9
hexane
PhMe
PhMe
3
10
11
12
13
14
15
16
17
18d
19
72
72
72
72
72
72
72
72
3
38/14
11/2
5/0
Fe3O4 (65) ClCH2COCl PhMe
Fe3O4 (65) HCCl2COCl PhMe
Fe3O4 (65) HCBr2COCl PhMe
0/0
Fe3O4 (65) TsCl
PhMe
PhMe
0/0
Fe3O4 (65) TMSCl
8/15
0/0
Fe3O4 (65) (MeCO)2O PhMe
Fe3O4 (65) MeCOCl
FeO (65) MeCOCl
PhMe
PhMe
27/57
63/0
3
Finally, we applied the same protocol to the dialdehyde 6
(Scheme 1), just increasing the corresponding amount of
carbonyl compounds and acid chloride, yielding the ex-
pected product 7 with good yield, with similar results be-
ing obtained by using iron(III) oxide (71%).
20
21
22
23
24
Fe2O3 (65) MeCOCl
Fe2O3 (32) MeCOCl
PhMe
PhMe
PhMe
PhMe
PhMe
3
3
3
3
3
5/94
24/55
39/30
38/13
39/25
Fe2O3 (7)
FeCl3 (65)
FeCl2 (65)
MeCOCl
MeCOCl
MeCOCl
Although the possible mechanism is still unknown, it
should be pointed out that the formation of a catalytic spe-
cies having an Fe(III)–Cl bond should play an important
role as a Lewis acid, with the centers of iron(III) being the
only avail to catalyze the final Michael addition (see
above).
a Reaction carried out using compound 1a (2.5 mmol), 2a (1 mmol),
and MeCOCl (2 mmol) in 3 mL of solvent, unless otherwise stated.
b Reaction performed using 300 mol%.
c Reaction performed using 600 mol%.
d Reaction performed with micromagnetite.
Synlett 2011, No. 14, 2017–2020 © Thieme Stuttgart · New York

LETTER
Catalyzed Synthesis of 4-Substituted-4H-Pyrans
Table 3 Synthesis of 4-Substituted-4H-Pyransa
2019
Table 2 Optimization of the Reaction Catalysta
O
O
R1
O
R2
O
Fe3O4 or Fe2O3
(65 mol%)
O
O
H
O
MeO
+
Y
Y
R2
MeCOCl (200 mol%),
PhMe, 25 °C, 3 h
O
O
R1
O
R1
Y
OMe
Br
1
2
4
1a
3a
cat.
Entry R1
Y
R2
Pyran Yield (%) Mp (°C)b
+
+
MeCOCl (200 mol%),
PhMe, 25 °C, 3 h
O
1
2
Me
OMe 4-BrC6H4
OMe 4-NCC6H4
OMe Ph
4a
96 (94)
79
98
92
Br
Br
H
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
4b
4c4
2a
O
O
3
85 (82)
83
61
OMe 4-MeOC6H4 4d13
112
119
125
65
MeO
OMe
4
O
5
OMe 4-HOC6H4
OMe 2-naphthyl
OMe (CH2)5CH
OMe i-Pr
4e
4f
68
4a
6
57c
Entry
Catalyst (mol%)
Yield of 3a/4a (%)
7
4g
4h
4i7
80c (79c)
72 (67)
95
1
3
4
5
6
7
Co(OH)2·Fe3O4 (1.4)
Ni(OH)2·Fe3O4 (1.2)
Cu(OH)2·Fe3O4 (1.3)
Ru(OH)3·Fe3O4 (1.4)
Pd(OH)2·Fe3O4 (1.2)
Pd/Cu(OH)x·Fe3O4 (1.5/0.8)
18/64
24/56
22/61
14/80
8/86
8
46
9
OEt
OEt
Me
4-BrC6H4
75d
84e
83
10
11
12
4-MeOC6H4 4j7
4-MeOC6H4 4k
63
75c (64)
91 (92)
OMe 4-BrC6H4
4l
68
8/83
a Reaction carried out using compound 1 (2.5 mmol), 2 (1 mmol), in
3 mL of toluene during 3 h, unless otherwise stated. Yields obtained
by Fe2O3 catalysis appeared in parenthesis.
b Recrystallized from mixtures of EtOAc and hexane.
c Reaction performed during 5 h.
a Reaction carried out using compound 1a (2.5 mmol) and 2a (1 mmol)
in 3 mL of toluene during 3 h.
d Mp 81 °C (ref. 7).
In conclusion, we have demonstrated that both magnetite
and iron(III) oxide are good catalysts for the pyran forma-
tion tandem reaction. The process could be performed
with a broad range of substrates, keeping the high level of
e Mp 83–85 °C (ref. 7).
the results, avoiding the use of high temperature, and re- References and Notes
ducing the reaction time from weeks to hours.
(1) (a) Hepworth, J. D.; Gabbutt, C. D.; Heron, B. M. In
Comprehensive Heterocyclic Chemistry II, Vol. 5; Katrizky,
A. R.; Rees, C. W.; Scriven, E. F. V., Eds.; Pergamon:
Oxford, 1996, 351. (b) Martín, N.; Martínez-Grau, A.;
Quinteiro, M.; Seoane, C. Educ. Chem. 1996, 20.
(c) Larrosa, I.; Romea, P.; Urpí, F. Tetrahedron 2008, 64,
2683.
O
O
O
O
H
O
O
H
Fe3O4
O
O
OMe
OMe
OMe
(65 mol%)
+
MeCOCl (400 mol%),
PhMe, 25 °C, 5 h
(2) Manolov, I.; Maichle-Moessmer, C.; Danchev, N. Eur. J.
Med. Chem. 206, 41, 882.
(3) Hamdi, N.; Puerta, M. C.; Valerga, P. Eur. J. Med. Chem.
208, 43, 2541.
(4) Urbahns, K.; Horváth, E.; Stasch, J.-P.; Mauler, F. Bioorg.
Med. Chem. Lett. 2003, 13, 2637.
OMe
O
MeO
1a
500 mol%
6
O
7
75%
(5) For an excellent review, see: Seoane, C.; Soto, J. L.;
Quinteiro, M. Heterocycles 1980, 14, 37.
Scheme 1 Double tandem process
(6) Wolinsky, J.; Hauer, H. S. J. Org. Chem. 1969, 34, 3169.
(7) Ni, C.-L.; Song, H.-H.; Yan, H.; Song, X.-Q.; Zhong, R.-G.
Ultrason. Sonochem. 2010, 17, 367.
(8) Other protocols are restricted to only cyclic 1,3-dicarbonyl
compounds: (a) Gao, Y.; Tu, S.; Shi, F.; Wang, Q.; Zhu, X.;
Shi, D. Synth. Commun. 2007, 37, 1603. (b) Fan, X. S.; Qu,
Y. Y.; Zhang, X. Y.; Wang, X.; Wang, J. J. Chin. Chem. Lett.
2009, 20, 387.
Acknowledgment
This work was financially supported by the Spanich Ministerio de
Ciencia y Tecnología Spain (Projects CTQ2007-65218/BQU and
Consolider Ingenio 2010 CSD2007-00006), Generalitat Valenciana
(G.V. PROMETEO/2009/039 and FEDER). R.C. thanks to G.V. for
a fellowship through the PROMETEO program.
(9) (a) Martínez, R.; Ramón, D. J.; Yus, M. Adv. Synth. Catal.
2008, 350, 1235. (b) Martínez, R.; Ramón, D. J.; Yus, M.
Org. Biomol. Chem. 2009, 7, 2176.
Synlett 2011, No. 14, 2017–2020 © Thieme Stuttgart · New York

2020
R. Cano et al.
LETTER
(10) For results from other groups, see: (a) Mojtahedi, M. M.;
Abaee, M. S.; Eghtedari, M. Appl. Organometal. Chem.
2008, 22, 529. (b) Mojtahedi, M. M.; Abaee, M. S.; Alishiri,
T. Tetrahedron Lett. 2009, 50, 2322. (c) Zhang, Z.-H.; Lü,
H.-Y.; Yang, S.-H.; Gao, J.-W. J. Comb. Chem. 2010, 12,
643. (d) Lü, H.-F.; Yang, S.-H.; Deng, J.; Zhang, Z.-H. Aust.
J. Chem. 2010, 63, 1290. (e) Gooben, L. J.; Mamone, P.;
Oppel, C. Adv. Synth. Catal. 2011, 353, 57. (f) Firouzabadi,
H.; Iranpoor, N.; Gholinejad, M.; Hoseini, J. Adv. Synth.
Catal. 2011, 353, 125.
(11) (a) Aliaga, M. J.; Ramón, D. J.; Yus, M. Org. Biomol. Chem.
2010, 8, 43. (b) Cano, R.; Ramón, D. J.; Yus, M. J. Org.
Chem. 2010, 75, 3458.
(12) Representative Procedure for the Synthesis of Dimethyl
4-(4-Bromophenyl)-2,6-dimethyl-4H-pyran-3,5-dicar-
boxylate (4a)
2 mmol). The resulting mixture was stirred at 25 °C during 3
h. The mixture was quenched with H2O (5 mL) and extracted
with EtOAc (3 × 5 mL). The organic phases were dried over
MgSO4, followed by evaporation under reduced pressure
to remove the solvent. The product was purified by chroma-
tography on silica gel (hexane–EtOAc) to give the titled
product; mp 96–100 °C(hexane). Rf = 0.5 (hexane–EtOAc =
4:1). IR (KBr) : n = 1721, 1671, 1629, 1584, 1296 cm–1.
1H NMR (300 MHz, CDCl3): d = 2.36 (6 H, d, J = 0.6 Hz,
2 × CCH3), 3.64 (6 H, s, 2 × CO2CH3), 4.72 (1 H, s, CH),
7.10 (2 H, d, J = 8.5 Hz, 2 × BrCCHCH), 7.36 (2 H, d,
J = 8.5 Hz, 2 × BrCCH). 13C NMR (75 MHz, CDCl3): d =
18.7 (2 C), 37.8, 51.4 (2 C), 107.7 (2 C), 120.4, 129.9 (2 C),
131.2 (2 C), 144.3, 158.7 (2 C), 166.9 (2 C). CG-MS: m/z
(%) = 382 (11) [M+ + 2], 380 (11) [M+], 367 (14), 365 (13),
323 (14), 321 (14), 226 (11), 225 (100), 193 (11). HRMS:
m/z calcd for C17H17BrO5: 382.0259; found: 380.0256.
(13) Goerlitzer, K.; Trittmacher, J.; Jones, P. G. Pharmazie 2002,
57, 523.
To a stirred solution of 4-bromobenzaldehyde (2a, 1 mmol)
in toluene (3 mL) were added Fe3O4 (50 mg, 65 mol%),
methyl 3-oxobutanoate (1a, 2.5 mmol), and AcCl (0.14 mL,
Synlett 2011, No. 14, 2017–2020 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.