time.4 Thus, safety concerns regarding pressure build-up
from nitrogen gas generation at elevated temperatures
would be mitigated by this approach.
With an initial set of reaction parameters in hand, we
commenced our studies in a continuous flow reactor
(Figure 1). The small-footprint setup (60 cm  60 cm)
conveniently fits inside a standard fume hood and consists
of (a) the reagent solutions, (b) two syringe pumps, (c) a
T-shaped mixer (i.d. = 500 μm), (d) a perfluoroalkoxy
(PFA) tubing reactor (i.d. = 750 or 1,000 μm), (e) a back
pressure regulator (bpr, 40 or 100 psi), and (f) the sample
collection flask.11 Elevated temperatures are conveniently
(and safely) achieved with a standard oil bath, and hydrox-
ylamine solutions are prepared directly from commercially
available 50% aqueous solutions. For ease of setup we
typically execute this reaction employing a two-pump
setup with a T-mixer,14 necessitating that the substrate be
premixed with either hydroxylamine or 2. (Both are
equally effective; vide infra.)
During the course of the reaction, nitrogen gas is gen-
erated and a gasÀliquid segmented flow is observed. This
byproduct formation necessitates employing the back-
pressure regulator to not only mitigate unsafe pressure
buildup but also maintain a uniform flow rate throughout
the system. Ultimately, the gas generation may in fact be
advantageous to the overall efficinecy of the reaction
through increased mixing efficiency as a result of the
segmented flow.15
Of the reported conditions for diimide generation, we
considered several, including acid-promoted decarboxyla-
tion of potassium diazodicarboxylate5 and treatment of
sulfonylhydrazines with base.6 Ultimately, inconsistent re-
sults, largely due to low solubility and heterogeneity,
prompted us to design a new reagent combination. Inspired
by diimide formation via elimination of water from
hydrazine oxide 4 (Scheme 1, step C),7 we reasoned that
electrophilic amination of hydroxylamine would provide
a safe, metal-free, alternative to classical hydrazine
oxidation.8 Namely, with an appropriatereagent, selective
O-functionalization (e.g., trifluoroacetylation, Scheme 1,
step A) of hydroxylamine would generate an electrophilic
nitrogen source in the form of an O-trifluoroacetylhy-
droxylamine (3).9 Amination of a second equivalent of
hydroxylamine by 3 (step B) would then form 4.
Paramount to the success ofthis approach was discovery
of a reagent selective for O-functionalization of hy-
droxylamine. Jencks assayed the N- versus O-selectivity
of various acyl electrophiles, finding nearly exclusive
O-acylation with N,O-diacetylhydroxylamine.10 We thus
evaluated a collection of related derivatives for alkene
reduction and identified N,O-bistrifluoroacetylhydroxyla-
mine (2) as the most efficient.11 Originally developed for
conversion of aldehydes to nitriles,12 2 is a stable, commer-
cially available reagent.13 Preliminary studies revealed
1.5 equiv of 2 in 1,4-dioxane with 5 equiv of hydroxylamine
to be sufficient for further reaction development. Of parti-
cular note is the importance of solvent in this reaction; high
solubility of hydroxylamine is necessary, and therefore,
hydrogen bond acceptors were found to be the most
effective solvents.
Hydrogenation of cyclooctene was used for comparison
of conditions (Table 1). The reaction was found to be
sensitive to both residence time (tR) and temperature
(entries 1À4). The optimal conditions were determined to
be heating at 100 °C with a tR of 20 min (entry 5).
Table 1. Continuous Flow Temperature and tR Optimization
entry
temp (°C)
tR (min)
yield (%)a
1
2
3
4
5
80
80
90
90
100
20
30
20
10
20
72
86
92
69
98
a GC yield.
(8) For classical hydrazine oxidation methods, see: (a) Aylward, F.;
Sawistowska, M. Chem. Ind. 1962, 484. (b) Lamani, M.; Ravikumara,
G. S.; Prabhu, K. R. Adv. Synth. Catal. 2012, 354, 1437. For recent use
of flavins for hydrazine oxidation, see: (c) Imada, Y.; Iida, H.; Naota, T.
J. Am. Chem. Soc. 2005, 127, 14544. (d) Smit, C.; Fraaije, M. W.;
Minnaard, A. J. J. Org. Chem. 2008, 73, 9482. (e) Imada, Y.; Kitagawa,
T.; Ohno, T.; Iida, H.; Naota, T. Org. Lett. 2010, 12, 32. (f) Imada, Y.;
Iida, H.; Kitagawa, T.; Naota, T. Chem.;Eur. J. 2011, 17, 5908. (g)
Marsh, B. J.; Heath, E. L.; Carbery, D. R. Chem. Commun. 2011, 47, 280.
(h) Teichert, J. F.; den Hartog, T.; Hanstein, M.; Smit, C.; ter Horst, B.;
Hernandez-Olmos, V.; Feringa, B. L.; Minnaard, A. J. ACS Catal. 2011,
1, 309.
Figure 1. Reaction setup: (a) reagent solutions; (b) syringe
pumps; (c) T-shaped mixer (i.d. = 500 μm); (d) PFA tubing
reactor (i.d. = 760 or 1,000 μm); (e) bpr (40 or 100 psi); (f)
sample collection flask.
(5) Hamersma, J. W.; Snyder, E. I. J. Org. Chem. 1965, 30, 3985.
(6) (a) Haukaas, M. H.; O’Doherty, G. A. Org. Lett. 2002, 4, 1771. (b)
Marsh, B. J.; Carbery, D. R. J. Org. Chem. 2009, 74, 3186.
€
(7) Durckheimer, W. Liebigs Ann. Chem. 1969, 721, 240.
(9) Wade, P. A.; Amin, N. V. Synth. Commun. 1982, 12, 287.
Org. Lett., Vol. 15, No. 3, 2013
711