Reaction discovery using microfluidic-based multidimensional screening of polycyclic Iminium Ethers

Article abstract of DOI:10.1021/jo100087h

"Chemical Equation Presented" Polycyclic iminium ethers are ambident electrophilic intermediates that react with a range of nucleophiles in a highly condition-dependent manner to afford densely functionalized lactams. In an effort to expand the scope of reactivity and assist in the generation of new chemotypes from these intermediates, several iminium, ethers were subjected to reaction screening using an automated microfluidics reaction platform. Application of this approach led to the discovery of several interesting reaction pathways involving the iminium ether intermediates that, will be described.

Full text of DOI:10.1021/jo100087h

pubs.acs.org/joc
Reaction Discovery Using Microfluidic-Based Multidimensional
Screening of Polycyclic Iminium Ethers
Jennifer L. Treece,† John R. Goodell,‡ David Vander Velde,† John A. Porco, Jr.,‡ and
,†
ꢀ
Jeffrey Aube*
†Department of Medicinal Chemistry, The University of Kansas, 1251 Wescoe Hall Drive, 4070 Malott Hall,
Lawrence, Kansas 66045, and KU Center for Chemical Methodology and Library Development
(KU-CMLD), The University of Kansas, 2121 Simons Drive, 2070 Structural Biology Center, Lawrence,
Kansas 66045, and ‡Department of Chemistry and Center for Chemical Methodology and Library
Development (CMLD-BU), Boston University, 590 Commonwealth Avenue, Boston Massachusetts 02215
jaube@ku.edu
Received January 19, 2010
Polycyclic iminium ethers are ambident electrophilic intermediates that react with a range of
nucleophiles in a highly condition-dependent manner to afford densely functionalized lactams. In
an effort to expand the scope of reactivity and assist in the generation of new chemotypes from these
intermediates, several iminium ethers were subjected to reaction screening using an automated
microfluidics reaction platform. Application of this approach led to the discovery of several
interesting reaction pathways involving the iminium ether intermediates that will be described.
Introduction
attractive substrates for reaction screening, as there is a
possibility for multiple chemotypes to arise from a single
substrate. In addition, the availability of a wide range of
structurally diverse iminium ethers and their ability to react
with a wide selection of nucleophiles further increases the
potential to generate unique sets of highly functionalized
products.
Iminium ethers have shown great utility in the synthesis of
functionalized heterocycles.1 Acyclic iminium ethers are
typically formed through O-alkylation of amides, while
cyclic and polycyclic versions can be formed through the
direct N-alkylation of oxazolines and dihydrooxazines2 or by
reaction of ketones with hydroxyalkyl azides (Figure 1).3 The
ambident electrophilic nature of these species gives
access to different reaction pathways within a single sub-
strate.1,4 This feature makes iminium ethers particularly
(1) Pittman, C. U., Jr.; McManus, S. P.; Larsen, J. W. Chem. Rev. 1972,
72, 357–438.
(2) (a) Frump, J. A. Chem. Rev. 1971, 71, 483–506. (b) Wiley, R. H.;
Bennett, L. L., Jr. Chem. Rev. 1949, 44, 447–476.
(3) (a) Boyer, J. H.; Hamer, J. J. Am. Chem. Soc. 1956, 78, 325–327. (b)
Boyer, J. H.; Hamer, J. J. Am. Chem. Soc. 1955, 77, 951–954. (c) Gracias, V.;
FIGURE 1. Synthesis and reactivity of cyclic iminium ethers.
ꢀ
Milligan, G. L.; Aube, J. J. Am. Chem. Soc. 1995, 117, 8047–8048. (d)
Gracias, V.; Milligan, G. L.; Aube, J. J. Org. Chem. 1996, 61, 10–11. (e)
ꢀ
Gracias, V.; Frank, K. E.; Milligan, G. L; Aube, J. Tetrahedron 1997, 53,
An ongoing goal of the Kansas laboratory is to increase
the variety of chemotypes that can be accessed from iminium
ethers. As transformations of iminium ethers have been
shown to be highly condition-dependent, we considered that
multidimensional reaction screening5 developed by the BU
ꢀ
16241–16252.
€
(4) (a) Hunig, S. Angew. Chem., Int. Ed. 1964, 3, 548–560. (b) Fazio, M. J.
J. Org. Chem. 1984, 49, 4889–4893. (c) Meyers, A. I.; Collington, E. W.
J. Am. Chem. Soc. 1970, 92, 6676–6678. (d) Meyers, A. I.; Temple, D. L., Jr.
J. Am. Chem. Soc. 1970, 92, 6644–6646. (e) Meyers, A. I.; Temple, D. L., Jr.
J. Am. Chem. Soc. 1970, 92, 6646–6647. (f ) Zhou, A.; Pittman, C. U., Jr.
J. Comb. Chem. 2006, 8, 262–267.
(5) Beeler, A. B.; Su, S.; Singleton, C. A.; Porco, J. A., Jr. J. Am. Chem.
Soc. 2007, 129, 1413–1419.
2028 J. Org. Chem. 2010, 75, 2028–2038
Published on Web 02/17/2010
DOI: 10.1021/jo100087h
r
2010 American Chemical Society

Treece et al.
JOCArticle
laboratory would provide an attractive strategy to address
this need. Nucleophile selection, iminium ether stability,
reaction solvent, time, and temperature, among other fac-
tors, have been shown to influence the outcome of reactions
of iminium ethers.4a,6 Accordingly, a systematic screening of
iminium ether substrates, reaction partners, additives, and
reaction conditions in an array format would facilitate access
to the greatest breadth of substrate reactivity. Moreover, use
of an automated microfluidic reaction screening platform
allows for the reactions to be completed on an analytical
scale in continuous succession, resulting in considerable
savings of both time and material. Herein, we report a series
of multidimensional reaction screens completed utilizing
both a relatively simple bicyclic iminium ether and a densely
functionalized bicyclo[3.2.2] scaffold.7
FIGURE 2. Substrates surveyed in the azido-Schmidt reaction.
product was only observed with promoters in which triflic
acid was released; triflic acid itself afforded full conversion.14
A marked increase in the regioselectivity was observed when
2-azido-1-ethanol was substituted for 3-azido-1-propanol.
Further, the regioselectivity also showed notable tempera-
ture dependence when 2-azido-1-ethanol was employed.
Conversely, this effect was not observed in reactions using
3-azido-1-propanol. Changes in solvent appeared to have
little, if any, effect on the regioselectivity.
Unfortunately, iminium ethers arising from reaction of 5
and 10 (Table 2, entry 1-6) proved to be unstable even as a
mixture of regioisomers and were not used further in this
study. Iminium ethers 14a and 14b were obtained as a 3:1
mixture of regioisomers by carrying out the hydroxyalkyl
azide addition step according to the conditions noted in entry
12 of Table 2 followed by a water wash. It has been pre-
viously shown that iminium ethers can be purified by column
chromatography, but the compounds under investigation
proved unstable to silica gel chromatography. Thus, the
regioisomeric mixture of 14a/14b was used without further
purification in multidimensional screening experiments
(Scheme 1). The regioisomeric ratio was determined by
spectroscopic analysis and by conversion to 13a/13b upon
treatment with mild aqueous base. In all but a few cases, the
products derived from the 14a/14b mixture were separable
by silica gel chromatography.
Previous work has shown that the regiochemical outcome
of ring expansion reactions of hydroxyalkyl azides and
R-substituted cyclic ketones is dependent on both the influ-
ence of the steric environment and the electronic nature of
the migrating centers.13 Thus, although reactions of methyl
and ethyl R-substituted ketones were generally nonselective,
migration of the more highly substituted carbon was favored
when bulkier substituents were used. In contrast, when
inductively electron-withdrawing substituents were placed
R to the ketone, selective (up to 9:1) migration was observed
for the less-substituted carbon. However, there are no
obvious steric differences in the two potentially migrating
carbons of bicyclo[3.2.1]octanoid (5), leaving open the ques-
tion of what forces are responsible for the modest but
systematic regioselectivity of its reactions with hydroxyalkyl
azides. Assuming antiperiplanar migration relative to the N2
leaving group, two mechanistic scenarios arising from the
intermediate oxonium ion are possible (Figure 3). Thus,
addition of azide can occur from either face of the oxonium
Results and Discussion
Selection and Synthesis of Iminium Ethers. We initially
sought to identify a complex scaffold upon which to examine
iminium ether reactivity. To this end, a set of five multi-
functional ketones 1-57-11 were reacted with 3-azido-1-
propanol under acidic conditions, followed by hydrolysis
and spectroscopic analysis (Figure 2).3c Of these, only
bicyclo[3.2.1]octanoid 5 led to an appreciable quantity of
the desired lactam product. Characterization of the resulting
lactams revealed formation of a mixture of regioisomers
arising from differential migration of the carbon centers
adjacent to the bridgehead ketone. Modification of the
workup allowed for isolation of the iminium ether inter-
mediates as a mixture of regioisomers. Thus, the iminium
ethers derived from 5 were deemed suitable complex sub-
strates for the present study.
As previously reported, bicyclo[3.2.1]octanoid 5 was
derived from a [5 þ 2] cycloaddition between a quinone
monoketal and a styrene component.7,12 Use of this metho-
dology allowed for the synthesis of four additional bicyclo-
[3.2.1]octanoid scaffolds, which were also evaluated for
their performance in the azido-Schmidt reaction with
3-azido-1-propanol (Table 1). Bicyclo[3.2.1]octanoid scaf-
folds bearing an allyl group in the R1 position failed to react,
but when R1 was a hydrogen atom, a mixture of regioisomers
could be isolated.13 Substrate 5, synthesized from trans-β-
methylstyrene, gave both the greatest overall yield (88%)
and the best regioselectivity (3:1) in this reaction.
The reaction of 5 was further investigated to determine the
effect of the hydroxyalkyl azide, solvent, and temperature on
regioselectivity of the reaction (Table 2). Preliminary acid
promoter screening revealed that significant conversion to
€
(6) Hunig, S.; Geldern, L. J. Prakt. Chem. 1964, 24, 246–268.
(7) Goodell, J. R.; McMullen, J. P.; Zaborenko, N.; Maloney, J. R.; Ho,
C.-X.; Jensen, K. F.; Porco, J. A., Jr.; Beeler, A. B. J. Org. Chem. 2009, 74,
6169–6180.
(8) Goldfarb, D. S. U.S. patent 2009163545.
(9) McDougal, N. T.; Schaus, S. E. J. Am. Chem. Soc. 2003, 125, 12094–
12095.
(10) Smissman, E. E.; Diebold, J. L. J. Org. Chem. 1965, 30, 4002–4005.
ꢀ
(11) Grecian, S.; Wrobleski, A. D.; Aube, J. Org. Lett. 2005, 7, 3167–
3170.
€
(12) (a) Buchi, G.; Mak, C.-P. J. Am. Chem. Soc. 1977, 99, 8073–8075. (b)
Collins, J. L.; Grieco, P. A.; Walker, J. K. Tetrahedron Lett. 1977, 38, 1321–
1324. (c) Engler, T. A.; Letavic, M. A.; Combrink, K. D.; Takusagawa, F.
J. Org. Chem. 1990, 55, 5810–5812. (d) Harmata, M.; Rashatasakhon, P.
Tetrahedron 2003, 59, 2371–2395.
(14) Acids tested in the azido-Schmidt reaction with substrate 6 include
Sc(OTf )3, La(OTf )3, Zn(OTf )3, Yb(OTf )3/PhCO2H, La(OTf )3/PhCO2H,
ꢀ
(13) Smith, B. T.; Gracias, V.; Aube, J. J. Org. Chem. 2000, 65, 3771–
3774.
Zn(OTf )3/PhCO2H, pTSA, CSA(þ), TFA, BF3 OEt2, HBF4, Tf2NH,
3
TMSOTf, and TfOH.
J. Org. Chem. Vol. 75, No. 6, 2010 2029

Treece et al.
JOCArticle
TABLE 1. Survey of Bicyclo[3.2.1]octanoids as Azido-Schmidt Substrates
aIsolated yield of regioisomeric mixture. bRegioisomeric ratio estimated by crude 1H NMR.
ion species (paths a and b as depicted), in each case leading to
a mixture of N-diazoniumoxazinane intermediates. We
speculate that path a should be favored in this case as a
result of the presence of potentially interfering steric inter-
action with the methyl group in path b. Previous work
suggests that azide adducts formed in this way can equili-
brate via chair-chair interconversion prior to loss of nitro-
gen gas and formation of the iminium ether,13 at which point
the reaction is complete and regiochemistry fully established.
Given the relatively low energy differences leading to 14a
over 14b and lack of obvious preference between the two
chair forms of the oxazinanes shown to the left of the
diagram in Figure 3, it would be imprudent to speculate
further on the cause of the preference for 14a.
the iminium ether (path b) were observed in accordance with
previously reported examples.
Iminium ether 19 was also selected for screening and prepared
according to our previously reported protocol (Scheme 2).15 We
selected this compound on the basis of its simplicity, similarity to
previously studied examples,3e,15 and the presence of a phenyl
substituent, which provided a chromophore to aid in product
detection and analysis. As usual, this iminium ether could be
subjected to base hydrolysis to afford lactam 20.
Multidimensional Screening of Iminium Ether Substrates.
multidimensional screen was performed using an
A
automated microfluidic reaction screening platform capable
of combining specified reagents in a sequential format.7 The
microfluidics platform consists of a multihead syringe pump,
liquid handler, a microreactor, and a UV-triggered fraction
collector. The liquid handler systematically injects pulses of
The suitability of these multifunctional iminium ethers for
ring-opening screens was validated by treatment of 14 with
nucleophilic reagents previously shown to react with simple
iminium ethers (Table 3). Products resulting from reactions
at both the cationic center (path a) and the O-alkyl portion of
ꢀ
(15) Fenster, E.; Smith, B. T.; Gracias, V.; Milligan, G. L.; Aube, J.
J. Org. Chem. 2008, 73, 201–205.
2030 J. Org. Chem. Vol. 75, No. 6, 2010

Treece et al.
JOCArticle
TABLE 2. Optimization of Azido-Schmidt Reaction of Bicyclo[3.2.1]octanoid 5
entry
hydroxyalkyl azide
solvent
CH3CN
1:1 CH3CN/CH2Cl2
CH2Cl2
CH3CN
1:1 CH3CN/CH2Cl2
CH2Cl2
CH3CN
1:1 CH3CN/CH2Cl2
CH2Cl2
CH3CN
temp
ratio (a:b)a
1
2
3
4
5
6
7
8
10
10
10
10
10
10
11
11
11
11
11
11
-20 °C
7.3:1.0
7.6:1.0
7.6:1.0
6.1:1.0
5.8:1.0
4.8:1.0
2.8:1.0
3.1:1.0
3.2:1.0
2.4:1.0
2.2:1.0
3.1:1.0
-20 °C
-20 °C
0 °C f rt
0 °C f rt
0 °C f rt
-20 °C
-20 °C
-20 °C
0 °C f rt
0 °C f rt
0 °C f rt
9
10
11
12
1:1 CH3CN/CH2Cl2
CH2Cl2
aRatios determined by 1H NMR. Only those reactions with full conversion were considered. Yields were not determined.
FIGURE 3. Regioselectivity analysis of the azido-Schmidt reaction of 5.
SCHEME 1
reactants into a continuously flowing solvent stream driven
by the multihead syringe pump. The three reagent pulses
converge in the microreactor and undergo mixing to become
a single reaction pulse. The reactions are quenched in the
reactor by a continuous flowing stream of water to give each
reaction a finite end point. The UV-triggered collection
allows for the collection of the most concentrated region of
the reaction pulse by setting an acceptable absorbance
threshold, thus selectively extracting individual reaction
mixtures from the pulse stream.
Using scaffolds 14 and 19, a total of 23 nucleophilic
reaction partners were evaluated (Figure 4). Reaction part-
ners were chosen for their potential to participate in both
condensations and cycloadditions. Since there are only a few
examples of carbon-carbon bond formations employing
reactions of iminium ethers, we focused particular attention
J. Org. Chem. Vol. 75, No. 6, 2010 2031

Treece et al.
JOCArticle
TABLE 3. Heteroatom Nucleophilic Addition Reactions of a Bicyclo[3.2.1]octanoid Derived Iminium Ether
entry
reagent
X
path
product
yield (%)a
1
2
3
4
satd aq NaHCO3
NaBH4
NaSPh
O
a
a
b
b
13
15
16
17
67
59b
62
70
H,H
SPh
N3
NaN3
aIsolated yield corresponding to major regioisomer. bBicyclo[3.2.2]enone reduced to allylic alcohol.
SCHEME 2
These reactions were quenched with excess water, trans-
ferred to 96-well plates, and analyzed by UPLC/MS/ELSD.
Productive reactions were defined as those affording
>20% conversion to a major product according to crude
UPLC/MS/ELSD analysis. Positive outcomes were obser-
ved with five reaction partners for iminium ether 19 and with
10 reaction partners for iminium ether 14. These reactions
were subsequently scaled up, and the products were isolated
and characterized using traditional spectroscopic methods.
For comparison purposes, we elected to conduct scale-up
reactions of both 14 and 19 with each reaction partner that
was found to be productive in the initial screens regardless of
which substrate indicated the positive outcome. Addition-
ally, scale-up reactions were carried out in acetonitrile
instead of DMSO as solubility of the iminium ether was no
longer a primary concern. The change to acetonitrile not
only increased the ease of product isolation but also in-
creased the overall yield. DIEA was selected as a base for the
scale-up studies, since DBU was found to react with the
iminium ether substrates leading to adduct formation.
Products Derived from Reaction Screening and Proposed
Mechanisms. All products isolated from the scale-up process
were the result of nucleophilic addition to the iminium ether
position distal to the nitrogen (Tables 4 and 5) via path b
(Table 3). The preference for product formation via one path
over another is dependent upon several factors, including
nucleophile size and the nature of the anion-stabilizing
group.15 Since all of the reaction partners used in this study
were larger than those previously shown to result in path a
products, it is not surprising that we observed a preference
for addition to the less hindered site (path b in Table 3). Prior
to this study, the use of carbon nucleophiles has been limited
to preformed sodium salts of stabilized carbon species. In
previous studies, addition of carbon nucleophiles to iminium
ethers was shown to proceed through reversible addition via
path a followed by rapid elimination to afford enamine
products.3d,15 However, this mode of reactivity was not
observed here. The feasibility of elimination is dependent
upon both the acidity and the steric environment of the R-
proton. It is possible that the reaction conditions employed
in this screen were not basic enough to induce elimination
through this pathway, but this point was not thoroughly
examined.
on expansion of these nucleophilic reaction partners. The
reactions were screened under neutral conditions as well as in
the presence of non-nucleophilic bases, 1,8-diazabicyclo-
[5.4.0]undec-7-ene (DBU, pKa = 12) and N,N-diisopropy-
lethylamine (DIEA, pKa=9.5) in order to generate reactive
nucleophilic species.
Reactions completed in the microfluidic system were
performed using 2 μmol (ca. 1 mg of 14 or 0.6 mg of 19) of
substrate with 3.0 equiv of reaction partner and 3.0 equiv of
base (if used). Reagents were prepared as stock solutions in
DMSO (ca. 0.25 M for 14 and 19, 0.75 M for reaction
partners and bases) with minimal exposure to air and stored
in oven-dried glass sleeves housed in the inert reagent storage
block. Each reaction used 8 μL from each desired stock
solution for a total reaction volume of 24 μL. Flow rates for
the system were set at either 4.5 or 1.2 μL/min to achieve
reaction times of 5 or 20 min, respectively. To ensure
comparable reaction times for each addition, reactions were
quenched by the addition of water (excess) prior to exiting
the microreactor. The resulting solutions were analyzed
without further treatment using ultra performance liquid
chromatography (UPLC)/MS/evaporative light scattering
detector (ELSD) analysis.16
Previous reports have shown that nucleophilic additions
to iminium ethers, such as those attempted within the screen,
may require longer reaction times than is possible using the
microfluidic apparatus. To ensure that no productive reac-
tions were overlooked on the basis of insufficient reaction
time, reactions were also completed under similar conditions
in capped glass vials placed in an orbital shaker for 24 h.
(16) Mazzeo, J. R.; Neue, U. D.; Kele, M.; Plumb, R. S. Anal. Chem.
2005, 77, 460A–470A.
2032 J. Org. Chem. Vol. 75, No. 6, 2010

Treece et al.
JOCArticle
FIGURE 4. Multidimensional screening reaction partners and parameters.
Reaction partners that underwent initial enolate forma-
tion frequently favored O-alkylation affording compounds
47, 48, 49, and 50 derived from substrate 19, and compounds
57, 58, 59, and 60 derived from substrate 14. This mode of
addition was noted in previous studies when diethylmalonate
sodium salt was used as the nucleophile.3d,15 Interestingly,
the reaction between 3-phenyl-5-isoxazolone (33) and sub-
strate 19 gave exclusive N-alkylation, whereas the reaction
with substrate 14 produced a ca. 1:1 mixture of both N-
alkylated 56 and O-alkylated 57. This apparent lack of
chemoselectivity may be attributed to increased steric inter-
action between 14 and the 3-phenyl of 33 as compared to the
relatively unhindered 19.
Esters 45 and 55 may also arise from an O-alkylation of the
iminium ether that is preceded by the pseudodimerization of
homophthalic anhydride to first produce acid 63 (Scheme 3).
Similar dimerization processes where one or both of the
anhydride molecules are replaced by the corresponding free
acid have been reported.17 In addition, approximately equi-
molar amounts of proposed intermediate 63 and product 45
were isolated when the reaction shown in Scheme 3 was
carried out preparatively. Since the process is only catalytic
in water, a trace amount of water may be sufficient to initiate
the dimerization of the nucleophile. Upon completion, the
resulting acid may then open the iminium ether under the
basic reaction conditions to provide the observed products.
An alternative mechanism may also be possible in which the
homophthalic anhydride first opens the iminium ether which
may be followed by subsequent dimerization and hydrolysis
upon work up. Further mechanistic investigation will be
required to differentiate between these two possibilities.
Treatment of the densely functionalized iminium ether 14
with ethyl bromoacetate 38 afforded the ring-opened alkyl
bromide 61. The formation of 61 is most likely a result of
bromide addition via path b. The formation of bromide
anion under the reaction conditions may be attributed to
the long reaction times and higher temperatures under which
these reactions were completed. Such conditions may induce
displacement or elimination of bromide anion, which may
then readd to the iminium ether giving the observed bromo-
lactam. Interestingly, when iminium ether 19 was treated
with ethyl bromoacetate 38 under the same conditions, the
same reactivity was not observed. In this case, the only
isolated product was lactam 51, albeit in low yield. Lactam
51 is likely the product of initial iminium ether hydrolysis
followed by displacement of R-halogenated 38. At this time it
(17) (a) Ayyangar, N. R.; Srinivasan, K. V. Indian J. Chem., Sect. B 1983,
22B, 1108–1115. (b) Ozcan, S.; Sahin, E.; Balci, M. Tetrahedron Lett. 2007,
48, 2151–2154. (c) Duddeck, H.; Kaiser, M. Spectrochim. Acta, Part A 1985,
41A, 913–924.
J. Org. Chem. Vol. 75, No. 6, 2010 2033

Treece et al.
JOCArticle
TABLE 4. Reactions of Iminium Ether 19
aConditions: reaction partner (3.0 equiv), DIEA (3.0 equiv), MeCN, 80 °C, 24 h. bConditions: reaction partner (3.0 equiv), MeCN, 135 °C, 24 h.
cConditions: reaction partner (3.0 equiv), KH (3.0 equiv), DMF, 80 °C, 24 h.
is unclear as to why ethyl bromoacetate afforded bromina-
tion upon reaction with 14 yet gave little to no reaction with
19. In this instance, the stabilizing ability of the counterion
may also have a notable influence on the reaction. The
formation of 19 as a tetrafluoroborate salt imparts greater
stability to the complex as compared to the triflate salt of 14.
Attempts made to exchange the triflate anion of 14 for the
tetrafluoroborate anion were unsuccessful.
Reactions using isonitrile 42 (TosMIC) afforded lactams
53 and 62 from 19 and 14, respectively. The structural
assignment of 53 was verified by comparison to the
same product synthesized using an alternative route
(Scheme 4).
TosMIC (42) is widely used in the formation of a diverse
range of heterocycles.18 Often in this route p-toluenesulfinate
64 serves as a leaving group to enable product aromatization,
thereby giving a positive driving force to the overall process.
In the absence of an adequate cyclization partner, however,
decomposition of TosMIC is known to occur.19 Early re-
ports proposed that, in TosMIC decomposition, the p-
toluenesulfinate may be released by direct substitution of
the TosMIC anion.20 More recent studies refute this propo-
sal, suggesting a more complex decomposition route is
responsible for p-toluenesulfinate release.19a Rather than
direct substitution, solvent participation19a or the self-cycli-
zation of TosMIC may lead to the release of 64 along with a
complex mixture of byproducts (Scheme 5). The reactivity
and instability of the resulting byproduct make the isolation
and identification of these species difficult. Once free in
solution, anion 64 can directly add to the electrophilic
iminium ethers 19 and 14 to produce sulfones 53 and 62,
respectively. Although the stabilized p-toluenesulfinate an-
ion is commonly used in the preparation of sulfones,19c,20 the
reaction of tosyl with an electrophile following ejection from
TosMIC to our knowledge has not been reported.
Ethyl diazoacetate (39) was the only reaction partner
examined that formed a new carbon-carbon bond upon
addition to the iminium ether, albeit in low yield. Interest-
ingly, the product isolated from the scale-up reaction did not
correspond to the product observed in the original screen.
This was apparent as the elucidated structure indicated the
incorporation of 1 equiv of acetonitrile that was present only
in the scale-up reactions. Attempts to reproduce the original
reaction in DMSO were unsuccessful. The proposed me-
chanism of this transformation involves the initial decom-
position of ethyl diazoacetate in the presence of excess
acetonitrile to furnish nitrilium ion 65.21 Nucleophilic addi-
tion of 65 to iminium ether 19 completes the carbon-carbon
bond formation to generate amide 52. Hydrolysis upon
(18) Van Leusen, D.; Van Leusen, A. M. Org. React. 2001, 57, 417–666.
(19) (a) Van Leusen, A. M.; Wildeman, J.; Oldenziel, O. H. J. Org. Chem.
1977, 42, 1153–1159. (b) Larionov, O. V.; de Meijere, A. Angew. Chem., Int.
Ed. 2005, 44, 5664–5667. (c) Grigg, R.; Lansdell, M. I.; Thorton-Pett, M.
Tetrahedron 1999, 55, 2025–2044.
(21) (a) Ibata, T.; Sato, R. Chem. Lett. 1978, 10, 1129–1130. (b) Doyle,
M. P.; Buhro, W. E.; Davidson, J. G.; Elliott, R. C.; Heokstra, J. W.;
Oppenhuizen, M. J. Org. Chem. 1980, 45, 3657–3664. (c) Saegusa, T.;
Ito, Y.; Shimizu, T.; Kobayashi, S. Bull. Chem. Soc. Jpn. 1969, 42, 3535–
3538.
(20) Bull, J. R.; Tuinman, A. Tetrahedron 1975, 31, 2151–2155.
2034 J. Org. Chem. Vol. 75, No. 6, 2010

Treece et al.
JOCArticle
TABLE 5. Reactions of Iminium Ethers 14
aConditions: reaction partner (3.0 equiv), DIEA (3.0 equiv), MeCN, 80 °C, 24 h. bIsolated yield corresponding to single major regioisomer. cIsolated
yield corresponding to a mixture of regioisomers. Only major regioisomer shown.
SCHEME 3
SCHEME 4
workup provides the observed product, 52 (Scheme 6). It is
likely that iminium ether 19 solely participated in this reaction
due to the presence of residual BF3 contaminating substrate 19,
leftover from the initial iminium ether formation which would
not have been present in substrate 14 (which used TfOH in
iminium ether formation as per Schemes 1 and 2).
Although reaction partners 28 and 41 were also classified
as productive in the original reaction screen, the scaled-up
reactions afforded complex mixtures from which products
could not be identified or isolated. Attempts to improve the
reaction by modifying the reaction solvent, time, and tempe-
rature were unsuccessful.
J. Org. Chem. Vol. 75, No. 6, 2010 2035

Treece et al.
JOCArticle
SCHEME 5
SCHEME 6
Conclusion
according to a UV trigger into 96-well plates and analyzed by
UPLC/MS/ELSD (10-90% CH3CN), 2 min.
This work exemplifies the ability of microfluidics technol-
ogy and reaction screening to discover new reactions of
polycyclic iminium ethers. The regioselective addition of
various carbon nucleophiles was examined under several
reaction conditions in an array format to multidimensionally
screen the dependent nature of these transformations. Over
400 reactions were completed and analyzed on an analytical
scale, resulting in the discovery of both interesting reagent
behavior (e.g., homophthalic anhydride pseudodimerization
prior to iminium ether addition) and new reaction products
(e.g., sulfone formation from reactions with TosMIC). Ad-
ditional work will focus on the use of the reaction manifolds
uncovered in this study for the construction of complex
chemical libraries.
General Procedure for Reaction Screening and Profiling with
Orbital Shaker. Stock solutions of substrate (0.25 M), reaction
partner (0.70 M), and base (0.70 M) were prepared in DMSO.
Individual reactions were prepared in oven-dried glass vials
under inert atmosphere by combining 0.50 mL of each stock
solution. The glass vials were individually capped and placed in
a 96-well holding bock. Reaction blocks were mixed in an orbital
mixer for 24 h at 90 °C. The reactions were quenched by the
addition of water and analyzed by UPLC/MS/ELSD (10-90%
CH3CN), 2 min.
7-Phenyl-3,5,6,7,8,9-hexahydro-2H-oxazolo[3,2-a]azepin-4-ium
Tetrafluoroborate Salt (19). BF3 OEt2 (2.87 mL, 22.9 mmol)
3
was added dropwise over 5 min to a solution of 4-phenyl
cyclohexanone (2.00 g, 11.4 mmol) and 2-azido-1-ethanol
(1.50 g, 17.2 mmol) in CH2Cl2 (60 mL) at 0 °C. Gas evolution
appeared upon acid addition. The reaction was warmed to rt,
stirred for 12 h, and then evaporated in vacuo to yield a colorless
oil. The crude oil was purified by silica gel column chromato-
graphy (1:9 MeOH-CH2Cl2) to afford 3.38 g (97%) of 19 as a
white solid. Rf 0.33 (1:9 MeOH-CH2Cl2); mp 127.0-128.5 °C;
1H NMR (400 MHz, DMSO-d6) δH 7.35-7.31 (m, 2H),
7.24-7.22 (m, 3H), 4.97-4.86 (m, 2H), 4.28 (app q, J=10.0
Hz, 1H), 4.16 (app q, J = 10.0 Hz, 1H), 3.88-3.78 (m, 2H),
3.03-2.91 (m, 3H), 2.03-1.83 (m, 4H); 13C NMR (400 MHz,
DMSO-d6) δC 179.1, 146.2, 128.5, 126.4 (2), 72.1, 52.4, 46.4,
45.5, 31.4, 27.6, 25.8; IR (neat) 2985, 1739, 1246, 1047 cm-1; MS
(ESþ) m/z 216.1 (Mþ); HRMS calcd for C14H18NO (Mþ)
216.1388, found 216.1377.
Experimental Section
General Procedure for Reaction Screening and Profiling with
the Microfluidics System. Stock solutions of iminium ether sub-
strate (0.25 M), reaction partner (0.70 M), and base (0.70 M) were
prepared in DMSO and placed in oven-dried glass sleeves that
were fitted into a custom-designed 96-well aluminum holding
block. The holding block was sealed with a cover containing a
continuous flow inert gas chamber and attached to the micro-
fluidics platform. System parameters were set to perform reactions
by mixing 8 μL of each reagent at 30 °C for 5 min or at 90 °C for 20
min. The reactions were quenched in the microreactor by flowing
water into the quench port. Individual reactions were collected
2036 J. Org. Chem. Vol. 75, No. 6, 2010

Treece et al.
JOCArticle
1-(2-Hydroxyethyl)-5-phenylazepan-2-one (20). A 1 M NaOH
solution (1.0 mL) was added to a solution of 7-phenyl-3,5,6,
7,8,9-hexahydro-2H-oxazolo[3,2-a]azepin-4-ium tetrafluoroborate
salt (20) (100 mg, 0.330 mmol) in CH2Cl2 (1 mL), and the
biphasic mixture was stirred at rt for 2 h. The reaction
was diluted with water (5 mL) and extracted with CH2Cl2 (3 ꢀ
5 mL). The organic extracts were combined, washed with brine
(10 mL), dried over anhydrous sodium sulfate, and evaporated
in vacuo to afford 72.4 mg (94%) of 20 as a white solid. Rf 0.43
(1:9 MeOH-CH2Cl2); mp 109.5-110.0 °C; 1H NMR (400
MHz, CDCl3) δH 7.29-7.26 (m, 2H), 7.20-7.14 (m, 3H),
3.76-3.63 (m, 5H), 3.51 (dt, J=13.8, 5.2 Hz, 1H), 3.38 (dd,
J=15.2, 5.6 Hz, 1H), 2.78-2.59 (m, 3H), 2.00-1.97 (m, 2H),
1.74 (q, J=12.0 Hz, 2H); 13C NMR (400 MHz, CDCl3) δC 176.7,
145.9, 128.4, 126.5, 126.4, 61.3, 51.5, 50.1, 48.0, 36.3, 35.8, 30.5;
IR (neat) 3428, 1616, 1056, 701 cm-1; MS (ESþ) m/z 234.1 (Mþ
þ 1); HRMS calcd for C14H20NO2 (Mþ þ 1) 234.1494, found
234.1452.
14 (100 mg, 2.17 mmol) in CH2Cl2 (5 mL), and the biphasic
mixture was stirred at rt for 2 h. The reaction was diluted with
water (5 mL) and was extracted with CH2Cl2 (3 ꢀ 10 mL). The
organic extracts were combined, washed with brine, dried over
anhydrous Na2SO4, and evaporated in vacuo to give a yellow
oil. The crude material was purified by SiO2 preparative
thin layer chromatography (5:95 MeOH-CH2Cl2) to afford
33.1 mg (67%) of 13a and 11.0 mg (22%) of 13b as a yellow
oils. Characterization data for 13a: Rf 0.76 (1:9 MeOH-
CH2Cl2); 1H NMR (400 MHz, CDCl3) δH 7.31-7.22 (m,
3H), 7.10-7.07 (m, 2H), 6.34 (d, J = 10.4 Hz, 1H), 4.10 (d,
J = 5.6 Hz, 1H), 3.69 (s, 3H), 3.66-3.47 (m, 4H), 3.33 (dd,
J=10.0, 1.2 Hz, 1H), 3.10 (t, J=6.0 Hz, 1H), 2.61 (t, J=6.8 Hz,
1H), 1.68 (m, 2H), 1.22 (d, J=6.8 Hz, 3H); 13C NMR (400
MHz, CDCl3) δC 188.9, 172.9, 153.3, 137.4, 129.0, 128.3,
127.9, 116.9, 73.0, 58.3, 55.5, 52.7, 48.9, 43.4, 41.2, 30.5,
22.9; IR (neat) 3421, 2958, 1650, 1623, 1456, 1126 cm-1; MS
(ESþ) m/z 330.1 (Mþ þ 1); HRMS calcd for C19H24NO4 (Mþ
þ 1): 330.1705, found 330.1548. Characterization data for 13b:
General Procedure for the Addition of Nucleophiles to 7-
Phenyl-3,5,6,7,8,9-hexahydro-2H-oxazolo[3,2-a]azepin-4-ium
Tetrafluoroborate Salt (45-51 and 53). Diisopropylethylamine
(86.0 μL, 0.495 mmol) was added to a solution of homophthalic
anhydride (80.1 mg, 0.495 mmol) in anhydrous acetonitrile
(1 mL), and the resulting solution was stirred at rt for 5 min.
Iminium ether 7-phenyl-3,5,6,7,8,9-hexahydro-2-H-oxazolo-
[3,2-a]azepin-4-ium tetrafluoroborate salt (19) (50.0 mg, 0.164
mmol) was added to the solution, and the reaction was heated to
80 °C and stirred for 24 h. After cooling to rt, the crude solution
was directly analyzed by LCMS and purified by mass-directed
fractionation. Additional purification for characterization (if
necessary) was completed by SiO2 preparative thin layer chro-
matography (5:95 MeOH-CH2Cl2).
1
(Rf 0.72 (1:9 MeOH-CH2Cl2); H NMR (400 MHz, CDCl3)
δ
H 7.30-7.21 (m, 3H), 7.04-7.02 (m, 2H), 6.37 (d, J=9.2 Hz,
1H), 3.95 (d, J=4.8 Hz, 1H), 3.89 (dd, J=9.2, 0.8 Hz, 1H), 3.82
(m, 1H), 3.67 (s, 3H), 3.59 (m, 1H), 3.51-3.41 (m, 2H), 3.24 (m,
1H), 2.98 (dd, J=6.4, 4.8 Hz, 1H), 2.64 (pent, J=6.8 Hz, 1H),
1.82-1.70 (m, 2H), 1.23 (d, J=6.8 Hz, 3H); 13C NMR (400
MHz, CDCl3) δC 186.3, 169.5, 154.9, 138.4, 129.0, 128.0,
127.7, 118.2, 67.9, 59.7, 58.1, 55.5, 48.2, 44.4, 42.7, 30.6,
21.5; IR (neat) 3421, 2960, 1681, 1649, 1457, 1124 cm-1; MS
(ESþ) m/z 330.1 (Mþ þ 1); HRMS calcd for C19H24NO4 (Mþ
þ 1) 330.1705, found 330.1538.
Synthesis of Polycyclic Iminium Ether Mixture (14a,b). To
a solution of racemic 3-methoxy-6-methyl-7-phenylbicyclo-
[3.2.1]oct-3-ene-2,8-dione (300 mg, 1.17 mmol) and 3-azido-1-
propanol (118 mg, 1.17 mmol) in CH2Cl2 (15 mL) at 0 °C was
added a 1.17 M stock solution of TfOH (3.11 mL, 3.51 mmol)
dropwise over 5 min. The acid stock solution was freshly
prepared by the addition of TfOH (0.40 mL, 4.51 mmol) to a
mixture of CH3CN (0.7 mL) and CH2Cl2 (2.9 mL). After 1 h at
0 °C the reaction was warmed to rt and stirred for 12 h. Upon
dilution with water (10 mL) the solution was extracted with
CH2Cl2 (3 ꢀ 20 mL), and the organic extracts were com-
bined, washed with brine, dried over anhydrous Na2SO4, and
evaporated in vacuo to afford 527 mg (98%) of 14a,b as a fluffy
yellow solid and a mixture of regioisomers (3.5:1.0 regioisomeric
2-(2-Oxo-5-phenylazepan-1-yl)ethyl-2-((1-oxo-1H-isochromen-
3-yl)methyl)benzoate (45). Yellow oil (35.0 mg, 43%); Rf 0.34,
1
(5:95 MeOH-CH2Cl2); H NMR (400 MHz, CDCl3) δH 8.23
(d, J=8.0 Hz, 1H), 8.03 (dd, J=8.0, 1.2 Hz, 1H), 7.64 (dt, J=7.6,
1.6 Hz, 1H), 7.54 (dt, J=7.6, 1.6 Hz, 1H), 7.46- 7.37 (m, 3H),
7.29-7.26 (m, 3H), 7.21 (m, 1H), 7.20-7.08 (m, 2H), 6.16 (s,
1H), 4.47 (t, J = 5.8 Hz, 2H), 4.33 (s, 2H), 3.88 (m, 1H),
3.78-3.66 (m, 2H), 3.39 (dd, J=6.2, 1.2 Hz, 1H), 2.72 (tt, J=
12.4, 3.4 Hz, 1H), 2.67-2.61 (m, 2H), 2.01-1.92 (m, 2H),
1.80-1.63 (m, 2H); 13C NMR (400 MHz, CDCl3) δC 175.5,
166.7, 162.7, 156.6, 145.9, 137.5, 137.4, 134.7, 132.6, 132.1,
131.1, 129.4, 129.3, 128.5, 127.7, 127.5, 126.6, 126.5, 125.3,
120.1, 103.9, 62.9, 49.9, 48.1, 47.4, 37.6, 36.4, 36.1, 30.7; IR
(neat) 1724, 1652, 1487, 1257, 1134 cm-1; MS (ESþ) m/z 496.2
(Mþ þ 1); HRMS calcd for C31H30NO5 (Mþ þ 1) 496.2124,
found 496.2140.
1
1
ratio determined by NMR). Mp 88.5-91.0 °C; NMR (400
MHz, CDCl3) δH 7.32-7.19 (m, 3.9H), 7.15-7.13 (m, 2H),
7.07-7.05 (m, 0.6H), 6.70 (d, J=9.0 Hz, 0.3H), 6.38 (d, J=10.0
Hz, 1H), 4.94-4.85 (m, 1.3H), 4.78-4.72 (m, 1.3H), 4.66 (m,
0.3H), 4.45, (d, J=5.0 Hz, 1H), 4.31 (m, 1H), 4.12-4.06 (m,
0.6H), 3.89 (m, 0.3H), 3.80 (m, 1H), 3.70 (s, 3H), 3.69 (s, 0.9H),
3.64 (d, J=11.0 Hz, 1H), 3.61 (m, 1H), 3.43 (m, 0.3H), 2.80-2.74
(m, 1.3H), 2.43-2.39 (m, 1.3H), 2.30-2.16 (m, 1.3H), 1.30 (d,
J=7.0 Hz, 0.9H), 1.25 (d, J=7.0 Hz, 3H); 13C NMR (400 MHz,
CDCl3) δC 183.7, 181.3*, 177.4, 171.9*, 154.2*, 153.1, 135.9*,
135.0, 129.2*, 129.1, 128.7, 128.4, 128.3*, 127.3*, 117.2*, 114.1,
76.3, 71.0, 70.8*, 64.8*, 63.0*, 55.9*, 55.8, 48.8, 47.6*, 46.7,
46.5*, 45.1, 42.0, 41.9*, 21.5, 20.5*, 19.7, 19.7*; IR (neat) 3055,
1700, 1662, 1265, 1031, 738, 703, 639 cm-1; MS (ESþ) m/z 312.1
(Mþ); HRMS calcd for C19H22NO3 (Mþ) 312.1600, found
312.1582.
2-(2-(2-Oxo-5-phenylcycloheptyl)ethyl)isoxazole-5(2H)-one
(46). Synthesized according to the general procedure where 3-
phenyl-5-isoxazolone (79.7 mg, 0.495 mmol) was substituted for
homophthalic anhydride. Clear oil (47.5 mg, 87%); Rf 0.41,
(5:95 MeOH-CH2Cl2); 1H NMR (400 MHz, CDCl3) δH
7.54-7.42 (m, 5H), 7.27-7.24 (m, 2H), 7.17 (m, 1H),
7.12-7.10 (m, 2H), 5.44 (s, 1H), 3.90-3.79 (m, 2H),
3.73-3.60 (m, 2H), 3.48 (m, 1H), 3.33 (dd, J = 15.1, 5.4 Hz,
1H), 2.71 (tt, J=12.0, 3.6 Hz, 1H), 2.54 (m, 2H), 1.97 (m, 2H),
1.73-1.62 (m, 2H); 13C NMR (400 MHz, CDCl3) δC 175.4,
170.5, 169.2, 145.7, 131.7, 129.3, 128.5, 127.5, 126.9, 126.5,
126.4, 90.9, 51.8, 50.8, 47.9, 47.1, 36.3, 35.9, 30.5; IR (neat)
2927, 1737, 1643, 1490, 1450, 761, 700 cm-1; MS (ESþ) m/z
377.1 (Mþ þ 1); HRMS calcd for C23H25N2O3 (Mþ þ 1)
377.1865, found 377.1827.
(E)-3-Methoxy-8-methyl-9-phenyl-6-(3-(phenylthio)propyl)-
6-azabicyclo[3.2.2]non-2-ene-4,7-dione (16). To a solution of
potassium hydride (13.0 mg, 0.325 mmol) in anhydrous DMF
(0.5 mL) was added a solution of benzenethiol (33.0 μL, 0.325
mmol) in anhydrous DMF (0.5 mL). The mixture was stirred at
rt for 5 min, and bicyclic iminium ether 19 was added to the
flask. The reaction mixture was stirred at 80 °C for 24 h. After
cooling, the solution was diluted with water and extracted with
(E)-6-(3-Hydroxypropyl)-3-methoxy-8-methyl-9-phenyl-6-
azabicyclo[3.2.2]non-2-ene-4,7-dione (13a) and (E)-6-(3-Hydroxy-
propyl)-3-methoxy-9-methyl-8-phenyl-6-azabicyclo[3.2.2]non-3-
ene-2,7-dione (13b). Saturated aqueous NaHCO3 solution
(5 mL) was added to a solution of polycyclic iminium ether
J. Org. Chem. Vol. 75, No. 6, 2010 2037

Treece et al.
JOCArticle
diethyl ether (3 ꢀ 10 mL). The organic extracts were combined,
washed with brine, dried over anhydrous Na2SO4, and evapo-
rated in vacuo to give a yellow oil. The crude material was
purified by SiO2 preparative thin layer chromatography (3:2
EtOAc-Hex) to afford 21.2 mg (62% based on the major
regioisomer in reactant) of 16 as a yellow oil. Rf 0.46 (3:2
EtOAc-Hex); 1H NMR (400 MHz, CDCl3) δH 7.35-7.17 (m,
8H), 7.05-7.02 (m, 2H), 6.32 (d, J=10.4 Hz, 1H), 4.07 (d, J=5.6
Hz, 1H), 3.67 (s, 3H), 3.53 (t, J=6.8 Hz, 2H), 3.30 (dd, J=10.0,
1.2 Hz, 1H), 3.04 (t, J=6.2 Hz, 1H), 2.88 (t, J=7.2 Hz, 2H), 2.57
(pent, J=6.8 Hz, 1H), 1.84 (pent, J=7.2 Hz, 2H), 1.20 (d, J=6.8
Hz, 3H); 13C NMR (400 MHz, CDCl3) δC 189.1, 171.4, 153.3,
137.6, 135.7, 129.8, 129.1, 129.0, 128.4, 127.8, 126.3, 116.9, 72.8,
55.4, 52.9, 49.3, 45.9, 41.1, 31.3, 27.8, 22.9; IR (neat) 2923, 1672,
1620, 1456, 1126, 700 cm-1; MS (ESþ) m/z 422.1 (Mþ þ 1);
HRMS calcd for C25H28NO3S (Mþ þ 1) 422.1790, found
422.1785.
General Procedure for the Addition of Nucleophiles to Poly-
cyclic Iminium Ether Triflate Salt (55-62). Diisopropylethyla-
mine (112.0 μL, 0.650 mmol) was added to a solution of
homophthalic anhydride (105 mg, 0.650 mmol) in anhydrous
acetonitrile (2 mL), and the resulting solution was stirred
at rt for 5 min. Polycyclic iminium ether 14 (100.0 mg,
0.216 mmol) was added to the solution, and the reaction was
heated to 80 °C and stirred for 24 h. After cooling to rt, the
crude solution was directly analyzed by LCMS and purified by
mass-directed fractionation. Additional purification for
characterization (if necessary) was completed by SiO2 prepara-
tive thin layer chromatography (5:95 MeOH-CH2Cl2). Alter-
natively, some reactions were similarly completed on a
0.216 mmol scale using proportional amounts of the necessary
reagents.
(m, 0.23H), 3.77* (m, 0.23H), 3.68 (s, 3H), 3.65* (s, 0.69H), 3.63
(m, 1H), 3.51 (m, 1H), 3.38* (m, 0.23H), 3.29 (dd, J=10.0, 1.2
Hz, 1H), 3.15 (t, J=6.4 Hz, 1H), 2.94* (t, J=5.4 Hz, 0.23H), 2.58
(p, J=6.8 Hz, 1H), 2.03-1.89 (m, 2H), 1.20 (d, J=7.2 Hz, 3H),
1.18* (d, J=6.8 Hz, 0.69H); 13C NMR (400 MHz, CDCl3) δC
189.2, 186.6*, 171.3, 169.4*, 166.9*, 166.8, 162.7, 160.5*, 156.8,
153.3, 152.2*, 138.6*, 138.9*, 137.3, 137.4, 137.2, 136.9*, 134.8*,
134.6, 132.3*, 132.5, 132.0*, 131.9, 131.2, 129.6, 129.4, 129.3*,
129.0, 128.9*, 128.4, 128.0*, 127.8*, 127.7, 127.6, 127.5*, 127.4,
125.3*, 125.3, 120.0, 118.6*, 117.5*, 117.0, 103.8, 103.8*, 72.7,
68.2*, 62.6*, 62.0, 59.4*, 55.4, 55.4*, 52.7, 49.2, 48.1*, 44.9*,
43.7, 42.7*, 41.2, 37.8*, 37.6, 27.7*, 27.4, 22.9, 21.4*; IR (neat)
1720, 1670, 1622, 1456, 1259, 1130, 730 cm-1; MS (ESþ) m/z
592.2 (Mþ þ 1); HRMS calcd for C36H34NO7 (Mþ þ 1)
592.2335, found 592.2310.
(E)-3-Methoxy-8-methyl-6-(3-(3-oxocyclohex-1-enyloxy)-
propyl)-9-phenyl-6-azabicyclo[3.2.2]non-2-ene-4,7-dione (60).
Synthesized according to the general procedure where
1,3-cyclohexadione (36.4 mg, 0.325 mmol) was substituted for
homophthalic anhydride. Clear oil (21.6 mg, 63% based on the
major regioisomer in reactant); Rf 0.42 (5:95 MeOH-CH2Cl2);
1NMR (400 MHz, CDCl3) δH 7.31-7.24 (m, 3H), 7.09-7.06 (m,
2H), 6.33 (d, J=10 Hz, 1H), 5.29, (s, 1H), 4.08 (d, J=5.2 Hz,
1H), 3.77 (t, J=6.0 Hz, 2H), 3.68 (s, 3H), 3.74 (m, 1H), 3.44 (m,
1H), 3.30 (dd, J=10.0, 1.2 Hz, 1H), 3.08 (t, J=6.0 Hz, 1H), 2.58
(t, J=6.8 Hz, 1H), 2.39 (t, J=6.4 Hz, 2H), 2.33 (t, J=6.4 Hz,
2H), 1.94 (m, 4H), 1.20 (d, J=7.2 Hz, 3H); 13C NMR (400 MHz,
CDCl3) δC 199.7, 189.2, 177.6, 171.3, 153.2, 137.6, 129.0, 128.3,
127.8, 117.0, 102.9, 73.0, 65.3, 55.5, 52.9, 49.2, 43.9. 41.2, 36.7,
28.8, 27.3, 22.8, 21.1; IR (neat) 1643, 1600, 1218, 1184 cm-1; MS
(ESþ) m/z 424.2 (Mþ þ 1); HRMS calcd for C25H30NO5 (Mþ þ
1) 424.2124, found 424.2112.
(E)-3-(3-Methoxy-8-methyl-4,7-dioxo-9-phenyl-6-azabicyclo-
[3.2.2]non2-en-6-yl)propyl 2-((1-oxo-1H-isochromen-3-yl)methyl)-
benzoate (55a) and (E)-3-(3-Methoxy-9-methyl-2,7-dioxo-8-phenyl-
6-azabicyclo[3.2.2]non-3-en-6-yl)propyl-2-((1-oxo-1H-isochro-
men-3-yl)mehtyl)benzoate (55b). Clear oil (32.6 mg, 25% as an
inseparable mixture of regioisomers); Rf 0.25 (3/2 EtOAc-
Hex); 1NMR (400 MHz, CDCl3; regioisomers indicated by
asterisks) δH 8.21 (dd, J=8.0, 0.4 Hz, 1H), 8.02 (dd, J=8.0,
1.2 Hz, 1H), 7.61 (dt, J=7.6, 1.2 Hz, 1H), 7.51 (dt, J=7.6, 1.2
Hz, 1H), 7.43-7.36 (m, 3H), 7.29-7.21 (m, 4H), 7.10 (m, 2H),
7.02* (m, 0.46), 6.42* (d, J=9.2 Hz, 0.23H) 6.33 (d, J=10.4 Hz,
1H), 6.11 (s, 1H), 6.09* (s, 0.23H), 4.29 (s, 2H), 4.32-4.21 (m,
2H), 4.11 (d, J=5.6 Hz, 1H), 3.91* (d, J=4.80 Hz, 0.23H), 3.88*
Acknowledgment. Financial support of this work was
provided by the National Institute of General Medical
Sciences (KU-CMLD (P50 GM-069663) and the CMLD-
BU (P50 GM067041)). J.L.T. gratefully acknowledges sup-
port through a NIH Dynamic Aspects of Chemical Biology
Training Grant. We thank Professor Aaron Beeler (Boston
University) for assistance with microfluidics experiments.
Supporting Information Available: Additional experimental
procedures and spectral data for all compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
2038 J. Org. Chem. Vol. 75, No. 6, 2010