5832
Jean-Christophe M. R. Monbaliu et al. / Tetrahedron Letters 51 (2010) 5830–5833
THF, acetonitrile, and dimethylformamide. The optimized condi-
tions and results obtained under standard pressure are listed in Ta-
ble 1 (entries 1, 4, 8, 12, 15, 18, and 21).12
withtotalconversionofthedieneandexcellentyields(87–94%)with
a residence time of 117.5 min (0.4 mL minÀ1 total flow rate). The use
of shorter residence times (i.e., higher flows) significantly decreased
theconversion. Undersimilaroperatingconditions, cyclopentadiene
(4b) underwent cycloaddition with Ac-, Boc-, and Bz-nitroso deriva-
tives (3a–c) in moderate to good yields (70–85%, after extraction).
No dimerization of the diene was observed. Total conversion was ob-
served for the cycloadditions of 9,10-dimethylanthracene (4c) with
3a–c within 117.5 min of residence time. The instability of the corre-
sponding cycloadducts lowers the yield and prevents any purifica-
tion on silica gel column. Similarly to the results obtained in the
cyclic series, 2,3-dimethylbutadiene (4d) reacted under flow condi-
tions with acylnitroso dienophiles 3a and 3b yielding 1,2-oxazines
1718 and 18 in 85% and 87%, respectively. Isoprene (4e) led to a 1/8
and 1/1 mixtures of regioisomers with dienophiles 3a and 3b,
respectively.19 The yields were moderate (65–72%). No cycloaddi-
tions are observed for dienes 4f and 4g, even under 100 bar (X-Cube)
and using an excess of hydroxamic acid.20
In conclusion, the cycloadditions of a selected set of dienes with
representative nitrosodienophiles have been successfully con-
ducted under microflow conditions, leading to the first example
of a continuous lab-scale production of 1,2-oxazines (0.2–1 g h
À1). Using equimolar solutions of reagents, similar and in several
cases higher yields to the corresponding batch reactions were ob-
tained. The continuous process however offers a range of advanta-
ges over batch reactions, namely the efficient heat exchange, the
easy control of the reaction stoichiometry and shorter residence
(reaction) times. Moreover, no metal catalysts or specific additives
were necessary to achieve good yields.
The cycloaddition of 1 was relatively insensitive toward solvent
polarity. For example, using cyclohexadiene (4a) at 55 °C and
1 mL minÀ1 total flow (i.e., 47 min residence time), the conversion
was complete in THF, acetonitrile, and methanol. In DMF, under
the same flow conditions, total conversion was reached at 95 °C
(92% yield). Doubling of the flow rate (i.e., shortening the reaction
time by a factor two) led to 85% conversion. Using the X-Cube
microsystem with a backpressure of 100 bar led to a significant
reduction in residence time (15 min) leading to 95% conversion.
For the reaction of 1 and 4b,13 the dimerization of cyclopentadiene
was the major phenomenon observed. At 0 °C and with
a
0.4 mL minÀ1 total flow, no reaction occurred. As expected, no
cycloaddition was observed for 9,10-dimethylanthracene (4c) and
2-nitrosotoluene.14 2,3-Dimethylbutadiene (4d) gave its corre-
sponding N-2-tolyl-1,2-oxazine cycloadduct 16 in 96% yield (after
extraction) in DMF at 95 °C and 1 mL minÀ1 total flow.15 At lower
temperatures, the conversion of 4d was lower than for cyclohexa-
diene (4a), even with lower flow rates. The reaction of isoprene
(4e) and 1 gave a 1:4 (LC) mixture of regioisomers in 91% yield.
No modification of the regioselectivity is observed in comparison
to batch conditions. Next, the cycloadditions of functionalized
dienes 4f and 4g with 1 led to the corresponding cycloadducts
22 and 25, respectively, in moderate to good yields after extraction.
In both cases, the regioselectivity is total. The longer residence
time needed to complete the reaction of 1 and 4f emphasizes the
lower reactivity of diene 4f.16 These results can be interpreted
within the electrophilicity scale as introduced by Domingo.17 Nitr-
osotoluene has a high electrophilicity power (
x
= 2.82 eV) and will
Acknowledgments
consequently react through polar Diels–Alder reaction toward
electron rich dienes (i.e., nucleophilic ones). For instance, the rela-
This work is supported by the Long Term Structural Methusa-
lem Funding by the Flemish Government, the BOF (Bijzonder Ond-
erzoeksfonds Universiteit Gent, Research Fund Ghent University)
and the FWO (Fonds voor Wetenschappelijk Onderzoek Vlaander-
en, Research Foundation - Flanders). J.M.-B. is a senior research
associate of F.R.S.-FNRS (Fonds de la Recherche Scientique-Belgi-
que, National Funds for Scientific Research-Belgium).
tively low
4d–e emphasize their higher nucleophilic nature. For 9,10-dimeth-
ylanthracene (4c), the computed equals 1.61 eV. This, combined
x values ranging from 0.83 to 1 eV for dienes 4a,b, and
x
with the high reversibility of its Diels–Alder reaction with nitroso
dienophiles, explains the lack of reactivity observed. The presence
of electron-withdrawing substituents onto the butadienyl moiety
(dienes 4f–g) increases their relative electrophilicity (and thus re-
duces their intrinsic nucleophilicity), as showed by their higher
References and notes
electrophilicity power (x4f,g = 1.58 and 1.89 eV, respectively) in
agreement with the observed lower conversions.
1. Yamamoto, H.; Kawasaki, M. Bull. Chem. Soc. Jpn. 2007, 80, 595–607.
2. (a) Crotti, S.; Bertolini, F.; Macchia, F.; Pineschi, M. Tetrahedron Lett. 2010, 51,
2284–2286; (b) Calvet, G.; Coote, S. C.; Blanchard, N.; Kouklovsky, C.
Tetrahedron 2010, 66, 2969–2980; (c) Bodnar, B. S.; Miller, M. J. Tetrahedron
Lett. 2009, 50, 796–798; (d) Yang, B.; Miller, M. J. J. Org. Chem. 2009, 74, 7990–
7993; (e) Bodnar, B. S.; Miller, M. J. J. Org. Chem. 2007, 72, 3929–3938.
3. (a) Cesario, C.; Tardibono, L. P., Jr.; Miller, M. J. Tetrahedron Lett. 2010, 51, 3053–
3056; (b) Yan, S.; Miller, M. J.; Wencewicz, T. A.; Möllmann, U. Bioorg. Med.
Chem. Lett. 2010, 20, 1302–1305; (c) Cesario, C.; Miller, M. J. J. Org. Chem. 2009,
74, 5730–5733; (d) Lin, W.; Virga, K. G.; Kim, K.-H.; Zajicek, J.; Mendel, D.;
Miller, M. J. J. Org. Chem. 2009, 74, 5941–5946; (e) Zhao, D.; Johansson, M.;
Bäckvall, J.-E. Eur. J. Org. Chem. 2007, 4431–4436; (f) Kibayashi, K. Chem. Pharm.
Bull. 2005, 53, 1375–1386; (g) Iwasa, S.; Tajima, K.; Tsushima, S.; Nishiyama, H.
Tetrahedron Lett. 2001, 42, 5897–5899; (h) Quadrelli, P.; Mella, M.; Invernizzi,
A. G.; Caramella, P. Tetrahedron 1997, 55, 10497–10510; (i) Martin, S. F.;
Hartman, M.; Josey, J. A. Tetrahedron Lett. 1992, 33, 3583–3586; (j) Keck, G. E.
Tetrahedron Lett. 1978, 48, 4767–4770.
Further on, the cycloaddition reactions of acylnitroso dienophiles
(3a–c) were studied. The transient acylnitroso species were gener-
ated in situ in the mixing unit of the reactor by pumping separately
theoxidantand thehydroxamic acids (2a–c) (Setup B,Fig. 1). Sodium
metaperiodate is preferred over tetrabutylammonium periodate, as
the latter led to the formation of significant amounts of solid byprod-
ucts that clogged the micrometric channel of the reactor. Several sol-
vent mixtures were investigated, the best results being realized with
a 3/1 water/acetone mixture, and the optimum temperature being
30 °C (except for 9,10-dimethylanthracene (4c), which reacted best
in DMF at 65 °C). By using the ThalesNano X-Cube (total flow rates
ranging from 0.2 to 1 mL minÀ1), no beneficial effect of the increased
pressure has been observed. Several attempts were performed using
the ThalesNano X-Cube using catalyst cartridges filled with sup-
ported NaIO4 to convert in situ the hydroxamic acid into the nitroso
derivative in order to avoid the extraction step; however these
experiments gave poor results. Flow rates ranged from 0.2 to
0.5 mL minÀ1 with the CPC College System from Cytos for 0.1 N equi-
molar solutions of diene/hydroxamic acid (in a 1/1 water/acetone
mixture) and sodium periodate (in water). The optimized conditions
and results are listed below (Table 1).
4. de los Santos, J. M.; Ignacio, R.; Aparicio, D.; Palacios, F. J. Org. Chem. 2007, 72,
5202–5206.
5. Streith, J.; Defoin, A. Synthesis 1994, 1107–1117.
6. Iwasa, S.; Fakhruddin, A.; Nishiyama, H. Mini-Rev. Org. Chem. 2005, 2, 157–175.
7. Adamo, M. F. A.; Bruschi, S. J. Org. Chem. 2007, 72, 2666–2669.
8. (a) Corrie, J.; Kirby, G.; Mackinnon, J. J. Chem. Soc., Perkin Trans. 1 1985, 883–
886; (b) Kirby, G.; Mackinnon, J. J. Chem. Soc., Perkin Trans. 1 1985, 887–892; (c)
Kirby, G.; Mcguigan, H.; Mclean, D. J. Chem. Soc., Perkin Trans. 1 1985, 1961–
1966; (d) Kirby, G.; Mcguigan, H.; Mackinnon, J.; Mclean, D.; Sharma, R. P. J.
Chem. Soc., Perkin Trans. 1 1985, 1437–1442.
9. (a) Wiles, C.; Watts, P. Eur. J. Org. Chem. 2008, 1655–1671; (b) Wiles, C.; Watts,
P. Expert Opin. Drug Disc. 2007, 2, 1487–1503; (c) Acke, D. R. J.; Stevens, C. V.
Green Chem. 2007, 9, 386–390; (d) Acke, D. R. J.; Stevens, C. V.; Roman, B. I. Org.
Process Res. Dev. 2008, 12, 921–928.
The cycloaddition of cyclohexadiene (4a) with acetyl- and tert-
butoxycarbonylnitroso dienophiles (3a and 3b) proceeded very well