A simple two steps ytterbium triflate-catalysed preparation of 2,2-dimethyl-2H-chromenes from salicylaldehydes and 2-methylpropene

Article abstract of DOI:10.1002/jhet.5570430626

The preparation of various 2,2-dimethyl-2H-chromenes was achieved in two steps via an ytterbium triflate-catalysed reaction between salicylaldehydes, trimethylorthoformate and 2-methylpropene. From salicylaldehyde, two reaction products were characterised

Full text of DOI:10.1002/jhet.5570430626

Nov-Dec 2006
1605
A Simple Two Steps Ytterbium Triflate-Catalysed Preparation of 2,2-
Dimethyl-2H-chromenes from Salicylaldehydes and 2-Methylpropene
Soizic Prado, Yves L. Janin*, Pierre-Etienne Bost
URA 2128 CNRS-Institut Pasteur, 28 rue du Dr. Roux, 75724 Paris cedex 15, France
Email: yljanin@pasteur.fr
Received March 3, 2006
1) HC(OMe)3,
Yb(OTf)3, 25°C
2) H+ cat., 110°C
0-67 %
O
OH
R
R
+
O
The preparation of various 2,2-dimethyl-2H-chromenes was achieved in two steps via an ytterbium
triflate-catalysed reaction between salicylaldehydes, trimethylorthoformate and 2-methylpropene. From
salicylaldehyde, two reaction products were characterised: 4-methoxy-2,2-dimethylchroman and 2-(1,3-
dimethoxy-3-methylbutyl)phenol. The former compound probably results from a Lewis acid-catalysed
[2+4] cycloaddition between the intermediate quinonemethide and 2-methylpropene whereas the latter may
occur via a reaction related to a carbonyl-ene reaction between the quinonemethide and 2-methylpropene.
Both compounds were subjected to a catalytic acidic treatment leading to 2,2-dimethyl-2H-chromene.
Starting from various salicylaldehydes, the scope of this method was investigated.
J. Heterocyclic Chem., 43, 1605 (2006).
In the course of synthesis of biologically active
We wish to report here the synthesis of various 2,2-
dimethylchromenes via an ytterbium triflate-catalysed
reaction between salicylaldehydes, trimethylorthoformate
and 2-methylpropene.
chromene derivatives, [1] our attention was drawn [2] to a
patent [3] reporting a simple preparation of 2,2-dimethyl-
2H-chromene (2) using a reaction between salicylalde-
hyde (1) and 2-methylpropene catalysed by acidic silica-
alumina catalysts at 150 °C or zinc chloride at 75°C. More
recently [4], the preparation of furo[2,3-b]benzopyran 5
was achieved in excellent yield using the ytterbium
triflate-catalysed reaction between salicylaldehyde (1),
trimethylorthoformate and dihydrofuran 4. As illustrated
in Scheme 1, this reaction is an example [5-11] of a [2+4]
cycloaddition between the intermediate o-quinonemethide
3 and, in the present case, compound 4. Moreover, a 20%
yield preparation of chromene 2 was reported from the
reaction between salicylaldehyde diethyl acetal and 2-
methylpropene in the presence of borontrifluoride etherate
[12].
From the overnight reaction between salicylaldehyde
(1), trimethylorthoformate, 2-methylpropene and ytter-
bium triflate in a sealed flask at room temperature, we
could isolate and characterise compounds 6 and 7 in 27
and 14% yield respectively. In a control experiment,
compound 6 was subjected to the reaction conditions
but no ring opening into 7 took place. Moreover, from
salicylaldehyde (1), no reaction was seen if trimethyl-
orthoformate was omitted. From the results of these
negative control experiments, we suggest that
compound 7 occurs via a reaction related to a carbonyl-
ene reaction between the o-quinonemethide inter-
mediate 3 and 2-methylpropene followed by a
methanol addition on the resulting double bond.
Carbonyl-ene reactions catalysed by lanthanide triflate
have been reported [13,14] and an example of a
competition between this reaction and a Diels-Alder
cycloaddition was also observed [13]. An excellent
review describes all the reactions that can be catalysed
by lanthanides triflates [15]. As related intramolecular
cycloadditions have been reported previously [9],
including an iodine-catalysed example [10], we also
tried iodine on our substrates but without success. In
order to obtain the corresponding chromene,
compounds 6 and 7 were heated to reflux in toluene in
the presence of a catalytic amount of p-toluenesulfonic
acid (PTSA). It is noteworthy that allowing the
resulting methanol to distil from the reaction mixture
from time to time was necessary to obtain a complete
Scheme 1
OH
O
i
O
1
ii
2
O
O
O
O
O
4
3
O
5
i: 2-methylpropene, silica-alumina catalyst, 150 °C or ZnCl2, 75 °C, 6 h.
ii: Yb(OTf)3, HC(OMe)3, CH2Cl2, 25 °C, 12h.

1606
S. Prado, Y. L. Janin, P.-E. Bost
Vol 43
were obtained in a better yield. A steric hindrance of the
aldehyde function of 8d could explain the lack of
occurrence of chromene 9d, but not the extensive
decomposition observed. Moreover, as the methoxy-
bearing salicylaldehyde 8e also led to extensive
decomposition, we suggest that a demethylation takes
place upon the Lewis acid aldehyde activation thus
leading to a para-quinonemethide prone to undergo other
reactions such as polymerisations [16,17]. A related
demethylation could also take place with compound 8d
and the resulting unsaturated ꢀ-ketone is likely to undergo
many types of side reactions also. Such concerted
demethylation cannot be written in the case of the salicyl-
aldehyde 8f and we obtained the corresponding chromene
9f in acceptable yield. A similar phenomenon was
reported for salicylaldehyde 8d (but not for 8e) in a
related intramolecular study in which p-tolunenesulfonic
acid was used [9]. We also undertook trials with the
hydroxy-bearing aldehydes 8h-i. However, although the
corresponding chromenes 9h-i could be detected [18] in
1H NMR spectra of sample of the reaction mixture in 10-
20% yield, the occurrence of other unidentified com-
pounds of similar polarity has so far prevented a proper
purification.
conversion of either compounds 6 or 7 into 2,2-
dimethylchromene (2).
Scheme 2
OH
O
OH
i
+
O
O
O
O
7
6
1
ii
ii
O
2
i: 2-methylpropene, Yb(OTf)3, HC(OMe)3, CH2Cl2, 25 °C, 12 h. ii:
PTSA, toluene, reflux, 1h.
Using an optimised two steps procedure (a 60 hours-
long reaction for the first step), without characterising the
intermediate cycloadducts or the corresponding acyclic
derivatives, we applied this method to salicylaldehydes 1
and 8a-i and, depending on the starting material, obtained
2,2-dimethylchromenes as reported in Scheme 3.
One of the most used methods to prepare chromenes [2]
is to build the pyran ring via an alkylation of the corre-
sponding phenol with 3-chloro-3-methylbut-1-yne fol-
lowed by a Claisen rearrangement of the resulting
propargyl ether into the target compound [19-21]. If the
introduction of a copper-catalysed alkylation method did
wonder in yield improvement of this first step [22,23], the
following thermal Claisen rearrangement/cyclization is, in
some cases, problematic [23,24]. The acid-catalyzed for-
mation of chromenes from few phenols and 3-hydroxy-3-
methylbut-1-yne was also reported [25]. A potentially
more general method, involves a condensation between
phenoxide salts and ꢁ,ꢀ-unsaturated carbonyl compounds
[26]. Remarkably, this reaction also take place between
phenols and ꢁ,ꢀ-unsaturated aldehydes under acid
conditions. [27,28] Condensation between salicylalde-
hydes and unsaturated esters followed by a subsequent
decarboxylation was also used [29-31] as well as
oxidative cyclization of 2-allylphenols [32,33]. The
present work illustrates the versatility of ytterbium triflate
and its usefulness in improving the synthetic method
mentioned above [3], which required the use of a pressure
reactor and heating, into a room temperature experiment
only requiring a stoppered glass tube. We hope that the
present procedure, in combination with, for example, an
ortho formylation [34] reaction of the corresponding
phenols, will sometime provide a useful alternative for the
preparation of benzopyran-containing compounds of
biological interest [1,31,35-40]. As a conclusion, we
could also prepare, in a rather small 10% yield, the
Scheme 3
R1
R1
R2
R3
OH
R2
R3
O
i
O
R4
R4
8a-i
9a-i
R1
R2
R3
R4
yield
H
H
H
H
H
H
H
H
H
H
Br
H
H
H
H
42%
67%
64%
64%
0%
1
8a
8b
8c
8d
8e
8f
H
I
H
NO2
H
H
OMe
H
OMe
H
H
0%
OMe
H
59%
H
H
Me
H
H
Me
H
53%
8g
8h
8i
OH
H
ni
ni
OH
H
H
i: 2-methylpropene, Yb(OTf)3, HC(OMe)3, CH2Cl2, 25 °C, 60 h. then
PTSA, toluene, reflux, 1h. ni: not fully isolated.
From salicylaldehyde 1, the chromene 2 was isolated in
a 42% yield only, probably because of its volatility. On
the other hand, the less volatile chromenes 9a-c and 9f-g

Nov-Dec 2006
A Simple Two Steps Ytterbium Triflate-Catalysed Preparation
1607
of 2,2- Dimethyl-2H-chromenes
1H); 7.40 (m, 1H). 13C (100 MHz, CDCl3): ꢀ = 27.2; 28.8; 37.8;
56.2; 72.4; 75.3; 117.6; 120.4; 122.3; 129.0; 129.6; 153.9. A
HRMS spectrum could not be obtained as this compound readily
eliminates methanol to give the ion corresponding to chromene
(m/z = 161).
spirochromene 11 from 5-bromosalicylaldehyde (8a) and
methylenecyclohexane (10).
Scheme 4
2-(1,3-Dimethoxy-3-methylbutyl)phenol (7).
OH
i
O
1
A fraction (0.36 g, 14%) was obtained as an oil. H (400
+
O
MHz, CDCl3): ꢀ = 1.17 (s, 3H); 1.26 (s,3H); 1.89 (dd, 1H, J =
3.0 Hz and 15.1 Hz); 2.14 (dd, 1H, J = 15.1 Hz and 7.7 Hz); 3.23
(s, 3H); 3.36 (s, 3H); 4.53 (dd, 1H, J = 3.0 Hz and 7.7 Hz); 6.89
(m, 2H); 7.07 (dd, 1H, J = 1.7 Hz and 7.3 Hz); 7.18 (m, 1H);
7.90 (s, 1H). 13C (100 MHz, CDCl3): ꢀ = 22.1; 26.0; 47.5; 49.7;
57.4; 74.9; 80.8; 117.3; 120.3; 127.4; 127.6; 129.1; 155.4.
HRMS-FAB: m/z [M+Na]+ Calcd. for C13H20O3Na: 247.1310.
Found: 247,1290.
Br
Br
8a
10
11
i: Yb(OTf)3, HC(OMe)3, CH2Cl2, 25 °C, 60h. then p-tolunenesulfonic
acid, toluene, reflux, 1h.
Further investigations are in progress in attempt to
improve this synthetic access.
Representative Procedure for the Synthesis of the 2,2-Dimethyl-
2H-chromenes, Preparation of Compound 2.
EXPERIMENTAL
General Methods.
In a 60 mL round-bottomed thick glass tube fitted with a
PTFE-faced screw-cap, salicylaldehyde (1) (1 g, 0.0082 mol) was
dissolved in CH2Cl2 (40 mL). Trimethylorthoformate (1.57 mL,
0.012 mol) and ytterbium triflate hydrate (0.25 g, 0.4 mmol) were
then added. The resulting solution was cooled to 0°C using an ice
bath and 2-methylpropene was bubbled through the solution. The
mass added was monitored by periodic weighting and between 2
and 4 equivalents of 2-methylpropene was thus added. The glass
tube was tightly closed and the solution was stirred at room
temperature for 60 hours. The resulting solution was cooled at
0°C this was diluted in CH2Cl2 and washed three times with a 1 N
sodium hydrogen carbonate solution. The organic layer was dried
over sodium sulfate and concentrated to dryness. The residue was
dissolved in toluene (60 mL) and p-toluenesulfonic acid (0.03 g,
0.15 mmol) was added. This was heated to reflux for one hour
while allowing the methanol to distil from the reaction mixture
by removing the water condenser four times (we suggest the
much safer use of a distillation apparatus for larger reaction
scale). After concentration to dryness, the residue was purified by
a chromatography over silica gel to yield the 2,2-dimethyl-2H-
chromenes as described below.
1H NMR and 13C NMR spectra were recorded on a Bruker
spectrometer at 400 MHz and 100 MHz, respectively. Shifts (ꢀ)
are given in ppm with respect to the TMS signal and coupling
constants (J) are given in Hertz. Column chromatography were
performed over Merck silica gel 60 (0.035 - 0.070 mm), using a
solvent pump operating at pressure between 2 and 7 bar (25-50
mL/mn) and an automated collecting system driven by a UV
detector set to 254 nm unless stated otherwise. Mass spectra were
obtained on an Agilent 1100 serie LC/MSD system using an
atmospheric electrospray ionisation system. High resolution mass
spectra (HRMS) were obtained by the laboratoire de spectro-
métrie de masse, CNRS-ICSN, 91198 Gif/Yvette, France.
Isolation of compounds 6 and 7
In a 60 mL round-bottomed thick glass tube fitted with a
PTFE-faced screw-cap, salicylaldehyde (1) (1.34 g, 0.011 mol)
was dissolved in CH2Cl2 (40 mL). Trimethylorthoformate (1.8
mL, 0.016 mol) and ytterbium triflate hydrate (0.34 g, 0.54
mmol) were then added. The resulting solution was cooled to
0°C using an ice bath and 2-methylpropene was bubbled through
the solution. The mass added was monitored by periodic
weighting and, in the present case, 2.5 g (0.044 mol) of 2-
methylpropene was thus added. The glass tube was tightly
closed and the solution was stirred at room temperature
overnight. The resulting solution was cooled at 0°C prior
opening the tube; this was diluted in CH2Cl2 and washed three
times with a 1 N sodium hydrogen carbonate solution. The
organic layer was dried over sodium sulfate and concentrated to
dryness. The residue was purified by a chromatography over
silica gel eluting with a mixture of cyclohexane/CH2Cl2 varying
from 3/2 to pure CH2Cl2. Concentration of the chromatographic
fraction gave compound 6 and 7 as described below.
2,2-Dimethyl-2H-chromene (2).
Obtained as an oil (42%) eluting with
1
a mixture of
cyclohexane/ CH2Cl2 9-1. H (CDCl3) and 13C (CDCl3) identical
with reported data [23].
6-Bromo-2,2-dimethyl-2H-chromene (9a).
Obtained as an oil (67%) eluting with
a mixture of
1
cyclohexane/ CH2Cl2 6-4. H (CDCl3) and 13C (CDCl3) identical
with reported data [23].
6-Iodo-2,2-dimethyl-2H-chromene (9b).
Obtained as an oil (64%) eluting with
1
a mixture of
cyclohexane/ CH2Cl2 5-5. H (CDCl3) and 13C (CDCl3) identical
with reported data [23].
4-Methoxy-2,2-dimethylchroman (6).
6-Nitro-2,2-dimethyl-2H-chromene (9c).
A pure fraction (0.58 g, 27%) was obtained as an oil. 1H (400
MHz, CDCl3; data previously reported in CCl4 [32]): ꢀ = 1.3 (s,
3H); 1.46 (s,3H); 1.98 (dd, 1H, J = 7.5 Hz and 13.5 Hz); 2.12
(dd, 1H, J = 13.5 Hz and 2.4 Hz); 3.49 (s, 3H); 4.46 (m, 1H);
6.79 (dd, 1H, J = 1.0 Hz and 8.2 Hz); 6.93 (m, 1H); 7.19 (m,
The methanol elimination of the intermediate compound was
very slow in refluxing toluene and was better achieved neat in
the presence of catalytic amount of TSA at 150 °C for 3 hours.
The resulting chromene was obtained as a solid (64%; m.p. =

1608
S. Prado, Y. L. Janin, P.-E. Bost
Vol 43
1
[11] M. Ashram, D. O. Miller and P. E. Georghiou, J. Chem. Soc.
Perkin. Trans. I, 1470 (2002).
69°C) eluting with a mixture of cyclohexane/ CH2Cl2 1-1. H
(CDCl3) and 13C (CDCl3) identical with reported data [23].
[12] R. Merten and G. Müller, Chem. Ber., 97, 682 (1964).
[13] C. Qian and T. Huang, Tetrahedron Lett., 38, 6721 (1997).
[14] V. K. Aggarwal, G. P. Vennall, P. N. Davey and C. Newman,
Tetrahedron Lett., 39, 1997 (1998).
[15] S. Kobayashi, M. Sugiura, H. Kitagawa and W. W.-L. Lam,
Chem. Rev., 102, 2227 (2002).
6-Methoxy-2,2-dimethyl-2H-chromene (9f).
Obtained as an oil (59%) eluting with
1
a mixture of
cyclohexane/ CH2Cl2 1-1. H (CDCl3) and 13C (CDCl3) identical
with reported data [23].
[16] R. Gompper, The Chemistry of Quinonoid Compounds. ed.;
Wiley: New York, 1974; part 2, p 1145.
2,2,5,7-Tetramethyl-2H-chromene (9g).
Obtained from salicylaldehyde 8g [41] as an oil (53%) eluting
1
[17] R. W. Van De Water and T. R. R. Pettus, Tetrahedron, 58,
5367 (2002).
with a mixture of cyclohexane/CH2Cl2 9/1. H (CDCl3) and 13C
[18] B. A. Chauder, A. V. Kalinin and V. Snieckus, Synthesis, 140
(2001).
(CDCl3) identical with reported data [27].
6-Bromospiro[chromene-2,1'-cyclohexane] (11).
[19] I. Iwai and J. Ide, Chem. Pharm. Bull., 10, 926 (1963).
[20] I. Iwai and J. Ide, Chem. Pharm. Bull., 11, 1042 (1963).
[21] J. Hlubucek, E. Ritchie and W. C. Taylor, Aust. J. Chem., 24,
2347 (1971).
[22] J. D. Godfrey, R. H. Mueller, T. C. Sedergran, N.
Soundararajan and V. J. Colandrea, Tetrahedron Lett., 35, 6405 (1994).
[23] D. Bell, M.R. Davies, G. Green and I. S. Mann, Synthesis,
707 (1995).
[24] J. M. Evans, C. S. Fake, T. C. Hamilton, R. H. Poyser and E.
A. Watts, J. Med. Chem., 26, 1582 (1983).
[25] G. P. Kalena, A. Jain and A. Banerji, Molecules, 2, 100 (1997).
[26] G. Sartori, G. Casiraghi, L. Bolzoni and G. Casnati, J. Org.
Chem., 44, 803 (1979).
This compound was prepared under the same reaction
conditions described above, using 1.5 equivalent of methylene-
cyclohexane instead of 2-methylpropene and isolated as an oil
(10%) after a chromatography eluting with cyclohexane. 1H (400
MHz, CDCl3): ꢀ = 1.3-1.56 (m, 1H); 1.5-1.6 (m, 5H); 1.74 (m,
2H); 1.9 (m, 2H); 5.69 (d, 1H, J = 9.8 Hz); 6.28 (d, 1H, J = 9.8
Hz); 6.72 (dd, 1H, J = 0.4 Hz and 8.3 Hz); 7.11 (d, 1H, J = 2.4
Hz); 7.19 (ddd, 1H, J = 0.4 Hz, 2.4 Hz and 8.3 Hz). 13C (100
MHz, CDCl3): ꢀ = 21.7; 25.6; 32.3; 77.6; 113.0; 118.6; 122.1;
124.4; 129.1; 131.8; 132.2; 152.3. HRMS-FAB: m/z [M+H]+
Calcd. for C14H16O79Br: 279.0385. Found: 279.0380. Calcd. for
C14H16O81Br: 281.0364. Found: 281.0360.
[27] B. A. Chauder, C. C. Lopes, R. S. C. Lopes, A. J. M. da Silva
and V. Snieckus, Synthesis, 279 (1998).
[28] S. L. MacNeil, B. J. Wilson and V. Snieckus, Org. Lett., 8,
1133 (2006).
Acknowledgment.
We thank Sanofi-Aventis as well as Pfizer for very generous
donations of scientific equipment. Dr Emile Bisagni is gratefully
acknowledged for his interest and suggestions.
[29] Y. Kawase, S. Yamaguchi, H. Horita, J. Takeno and H.
Kameyama, Bull. Chem. Soc. Jpn., 55, 1153 (1982).
[30] S. Yamaguchi, T. Saitoh, M. Kamiumezawa, H. Enomoto
and Y. Kawase, J. Heterocyclic. Chem., 29, 755 (1992).
[31] S. Yamaguchi, M. Nedachi, M. Maekawa, Y. Murayama, M.
Miyazawa and Y. Hirai, J. Heterocyclic. Chem., 43, 29 (2006).
[32] T. Hosokawa, S. Yamashita, S.-I. Murahashi and A. Sonoda,
Bull. Chem. Soc. Jpn., 49, 3662 (1976).
[33] K. C. Nicolaou, J. A. Pfefferkorn, A. J. Roecker, G. Q. Cao,
S. Barluenga and S. A. Mitchell, J. Am. Chem. Soc., 122, 9939 (2000).
[34] T. Suresh, T. Dhanabal, R. N. Kumar and P. S. Mohan,
Heterocyclic Commun., 11, 79 (2005).
REFERENCES
[1] S. Prado, H. Ledeit, S. Michel, M. Koch, J.C. Darbord, S.T.
Cole, F. Tillequin and P. Brodin, Bioorg. Med. Chem., 14, 5423 (2006).
[2] E. E. Schweizer and D. Meeder-Nycz, Chromenes,
Chromanones, and Chromones. In The Chemistry of Heterocyclic
Compounds; Ellis, G.P. Ed John Wiley & Sons: 1977; 31, p 49.
[3] J. T. Arrigo Polycyclic Heterocyclic Compounds. US
2,987,525, 1961.
[35] V. Kumar, J. Reisch and A. Wickramasinghe, Aust. J. Chem.,
42, 1375 (1989).
[36] J. Razafimbelo, G. Baudouin, F. Tillequin, P. Renard, S.
Leonce, A. Pierre and G. Atassi, Chem. Pharm. Bull., 46, 34 (1998).
[37] N. Costes, H. Le Deit, S. Michel, F. Tillequin, M. Koch, B.
Pfeiffer, P. Renard, S. Leonce, N. Guilbaud, L. Kraus-Berthier, A. Pierre
and G. Atassi, J. Med. Chem., 43, 2395 (2000).
[4] J. S. Yadav, B. V. S. Reddy, M. Aruna, C. Venugopal, T.
Ramalingam, S. K. Kumar and A. C. Kunwar, J. Chem. Soc. Perkin.
Trans. I, 165 (2002).
[5] J. Brugidou and H. Christol, C. R. Hebd. Seances Acad. Sci.,
256, 3323 (1963).
[38] H. D. Thi Mai, T. Gaslonde, S. Michel, F. Tillequin, M.
Koch, J. B. Bongui, A. Elomri, E. Seguin, B. Pfeiffer, P. Renard, M. H.
David-Cordonnier, W. Laine, C. Bailly, L. Kraus-Berthier, S. Leonce, J.
A. Hickman and A. Pierre, J. Med. Chem., 46, 3072 (2004).
[39] V. A. Ashwood, R. E. Buckingham, F. Cassidy, J. M. Evans,
E. A. Faruk, T. C. Hamilton, D. J. Nash, G. Stemp and K. Willcocks, J.
Med. Chem., 29, 2194 (1986).
[6] O. L. Chapman and C. L. McIntosh, Chem. Commun., 383
(1971).
[7] J. P. Marino and S. L. Dax, J. Org. Chem., 49, 3671 (1984).
[8] Y. Génisson, P. C. Tyler and R. N. Young, J. Am. Chem.
Soc., 116, 759 (1994).
[9] H. Miyazaki, K. Honda, M. Asami and S. Inoue, J. Org.
Chem., 64, 9507 (1999).
[10] J. S. Yadav, B. V. S. Reddy, C. V. Rao and K. V. Rao, J.
Chem. Soc. Perkin. Trans. I, 1401 (2002).
[40] R. Bergmann and R. Gericke, J. Med. Chem., 33, 492 (1990).
[41] K. v Auwers and E. Borsche, Chem. Ber., 1698 (1915).