byMann and co-workers in1984.5 The photoactiveexcited
state was identified as the distorted a3E1 ligand field state,
characterized by an elongated rutheniumꢀarene bond and
a hole in a low-lying d-orbital that make it susceptible to
nucleophilic attack. Consistent with these data is a me-
chanism in which absorption followed by nonradiative
decay and intersystem crossing afford the excited a3E1
state (4, Scheme 1). The rate-limiting step is nucleophilic
attack by a molecule of solvent or an anion in the first
solvation sphere, in which the strength of the nucleophile is
a crucial factor.
Figure 1. Continuous flow photochemical reactor.
standard 450 W medium-pressure mercury lamp.17 The
reaction is pushed through the tubing by a peristaltic
pump. The UV output was safely housed within an
aluminum foil-lined container. The entire setup requires
an area less than1 ꢁ 0.3 m2 and fitssafely and conveniently
inside a standard laboratory fume hood.
Scheme 1. Desired Reactivity: Intercepting Intermediate 4
Of the multitude of reactions catalyzed by 1,7,8 we chose to
examine an intramolecular eneꢀyne cycloisomerization.18,19
Specifically, we began by examining the cycloisomeriza-
tionofenyne 5 todiene6 (Scheme 2). Atthe outset, acetone
was used as a solvent for the cycloisomerization of enyne 5
to cyclopentene 6. Under standard batch conditions, no
conversion was observed in the presence of catalyst 2
(5 mol %).20 However, under conditions of continuous
flow, in the presence of catalyst 2 (5 mol %), complete
conversion to the desired product (6) was rapidly obtained
in a residence time (tR) of only 2 min. In order to probe
flexibility in the type of solvent employed, a variety of
solvents were examined with generous (10 mol %) catalyst
loadings; no conversion to the desired diene was observed
in ethyl acetate, dichloromethane, or in mixtures of di-
chloromethane and Lewis basic cosolvents such as DMF,
ethyl acetate, or acetone.21
Prior to discussing additional details about our investi-
gation, the operational aspects of the photochemical reac-
tion deserve comment. Continuous flow micro- and
macroreactors have significant benefits on the yield,
reproducibility, efficiency, and throughput of many
reactions, particularly photochemical.10ꢀ12 In “batch”
photochemistry, efficient irradiation of a given molecule in
solution drops as its distance from the light source in-
creases. The feasible substrate concentration is typically
also compromised, to the significant detriment of the
associated throughput (mol Lꢀ1 minꢀ1). Taken together,
these limitations make large-scale photochemical reactions
especially challenging. In contrast, the relatively shallow
fluidic channels of continuous flow reactors enhance light
penetration, which increases both the efficiency and the
homogeneity of the illumination per unit surface area.13
The photochemical apparatus constructed for our stu-
dies is depicted in Figure 1.14 Flexible and chemically inert
high purity perfluoro alkoxy alkane (HPFA) tubing15,16
was wrapped around the quartz immersion well of a
Having identified acetone as the solvent for the desired
reaction,22 we turned our attention toward optimization of
other reaction parameters. We began with residence time
(tR), which is easily examined by variation of the flow rate.
Thus, decreasing the residence time to 30 s resulted in a
(17) The length of the reactor volume is defined herein as the length of
the tubing in physical contact with the surface of the immersion well.
Additional tubing (required to connect this volume to the rest of the
reactor) extends out of the reactor cabinet, although this length is not
counted as part of the formal reaction volume. See Supporting Informa-
tion for details.
(10) Albini, A.; Germani, L. Photochemical Methods. In Handbook
of Synthetic Photochemistry; Albini, A., Fagnoni, M., Eds.; Wiley-VCH:
Weinhheim, 2010; p 13 ff and references therein.
(18) For reviews on enyne cycloisomerizations, see: (a) Trost, B. M.;
^
Krische, M. J. Synlett 1998, 1–16. (b) Michelet, V.; Toullec, P. Y.; Genet,
(11) For a review of photochemistry conducted in flow microreac-
€
J.ꢀP. Angew. Chem., Int. Ed. 2008, 47, 426–4315.
tors, see: Coyle, E. E.; Oelgemoller, M. Photochem. Photobiol. Sci. 2008,
7, 1313–1322.
(19) For studies on enyne cycloisomerizations catalyzed by 1, see:
(a) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 9728–9729.
(b) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 2002, 124, 5025–5036.
(c) Trost, B. M.; Gutierrez, A. C.; Ferreira, E. M. J. Am. Chem. Soc.
2010, 132, 9206–9218.
(12) For an example of a photochemical reaction conducted under a
continuous flow in a macroreactor, see: Hook, B. D.; Dohle, W.; Hirst,
P. A.; Pickworth, M.; Berry, M. B.; Booker-Milburn, K. I. J. Org. Chem.
2005, 70, 7558–7564.
(13) Carofiglio, T.; Donnola, P.; Maggini, M.; Rossetto, M.; Rossi,
E. Adv. Synth. Catal. 2008, 350, 2815–2822.
(14) A similar setup was reported by Hook and co-workers (ref 12).
(15) Idex Health & Science Materials and Tools, Chemical Compat-
(accessed Mar 1, 2011).
(16) In our initial studies, we used 0.762 mm (0.03 in) i.d. tubing and a
reactor volume of 250 μL (tubing dimensions 1.59 mm o.d. ꢁ 0.762 mm
i.d. ꢁ 54.8 cm).
(20) Batch conditions were carried out in a quartz round-bottom
flask, in the absense of a filter. No conversion was observed after
exposure to the light source for 30 min (15 times longer than the exposure
time required for full conversion in flow).
(21) The application of a Pyrex filter did not have any effect on the
conversion in any of these cases.
(22) To verify that the reaction was indeed being catalyzed by the
ruthenium complex and not by some other photochemical process, a
control reaction was run in the absence of catalyst: no conversion was
observed.
Org. Lett., Vol. 13, No. 24, 2011
6415