Importance of C-N bond rotation in N-acyl oxazolidinones in their SmI 2-promoted coupling to acrylamides

Article abstract of DOI:10.1021/ja903401y

A detailed mechanistic investigation was undertaken to determine the dominating factors of the postelectron transfer steps in the SmI ₂-promoted carbon-carbon bond forming reaction between N-acyl oxazolidinones and acrylamides. Competition expe

Full text of DOI:10.1021/ja903401y

Published on Web 07/06/2009
Importance of C-N Bond Rotation in N-Acyl Oxazolidinones
in their SmI2-Promoted Coupling to Acrylamides
Rolf H. Taaning, Karl B. Lindsay, Birgit Schiøtt, Kim Daasbjerg,* and
Troels Skrydstrup*
The Center for Insoluble Protein Structures (inSPIN), Department of Chemistry and the
Interdisciplinary Nanoscience Center, Aarhus UniVersity,
Langelandsgade 140, 8000 Aarhus, Denmark
Received April 28, 2009; E-mail: kdaa@chem.au.dk; ts@chem.au.dk
Abstract: A detailed mechanistic investigation was undertaken to determine the dominating factors of the
postelectron transfer steps in the SmI2-promoted carbon-carbon bond forming reaction between N-acyl
oxazolidinones and acrylamides. Competition experiments were performed by reacting two N-acyl
oxazolidinones with a limiting amount of N-t-butyl acrylamide, and from the product distribution, the relative
reactivity values (RV) for a series of N-acyl oxazolidinones were then calculated against N-pivaloyl
oxazolidinone as the reference. An almost linear correlation was obtained for the simple alkyl N-acyl
oxazolidinones when ln RV was plotted against the activation barriers for C-N bond rotation (s-trans to
s-cis) obtained by DFT calculations, implying that C-N bond rotation from the s-trans to s-cis conformation
is one of the essential parameters controlling the reactivity. These results were substantiated by other
competition experiments carried out for the corresponding imide derivatives, where rotation is not necessary
for obtaining bidentate coordination and where no such correlation as described above was observable.
The finding that the reactivity of the simple N-acyl oxazolidinones for these SmI2-mediated transformations
correlates with the activation barriers for C-N bond rotation may have implications for other useful synthetic
organic reactions involving similar substrates. Finally, these studies were extrapolated to understanding
the poor reactivity of N-acyl oxazolidinones, as those derived from Evans chiral auxiliaries, with N-tert-
butyl acrylamide. These couplings appear to be dominated by the activation energy for addition because
of arising syn-pentane interactions in the transition-state for C-C bond formation. We demonstrate that
the addition of Lewis acids can have a beneficial effect on the coupling yields.
groups of Flowers2 and Hoz,3 and by others including our-
selves,4 to understand the nature of this reagent in THF, as well
Introduction
SmI2-mediated reactions have gained substantial prominence
by offering mild and selective alternatives for the creation of
carbon-carbon bonds. This reagent is highly suitable for
promoting a multitude of transformations including ketyl-olefin
radical addition reactions, inter- and intramolecular pinacol
couplings, Reformatsky and Barbier type reactions, as well as
cascade reactions involving a variety of combinations of these
transformations.1 A particular fascinating feature about this
reagent is the ability to fine-tune its electron-donating abilities
and functional group selectivity through the addition of cosol-
vents and additives. Substantial efforts have been made by the
as other solvents, the structure of the metal complexes formed
upon addition of cosolvents/additives, the interaction of such
complexes with the substrates, and the mechanistic intricacies
of the electron transfer step.
Although this reagent has been at the hands of synthetic
chemists for nearly 30 years, relatively little mechanistic
information is available concerning the post electron transfer
(2) (a) Flowers, R. A., II Synlett 2008, 10, 1427. (b) Teprovich, J. A., Jr.;
Balili, M. N.; Pintauer, T.; Flowers, R. A., II Angew. Chem. 2007,
119, 1. (c) Prasad, E.; Flowers, R. A., II J. Am. Chem. Soc. 2005,
127, 18093. (d) Prasad, E.; Knettle, B. W.; Flowers, R. A., II J. Am.
Chem. Soc. 2004, 126, 6891. (e) Chopade, P. R.; Davis, T. A.; Prasad,
E.; Flowers, R. A., II Org. Lett. 2004, 6, 2685. (f) Chopade, P. R.;
Prasad, E.; Flowers, R. A., II J. Am. Chem. Soc. 2004, 126, 44. (g)
Dahlen, A.; Hilmersson, G.; Knettle, B. W.; Flowers, R. A., II J. Am.
Chem. Soc. 2003, 68, 4870. (h) Prasad, E.; Knettle, B. W.; Flowers,
R. A., II J. Am. Chem. Soc. 2002, 124, 14663. (i) Miller, R. S.; Sealy,
J. M.; Shabangi, M.; Kuhlman, M. L.; Fuchs, J. R.; Flowers, R. A., II
J. Am. Chem. Soc. 2000, 122, 7718. (j) Shotwell, J. B.; Sealy, J. M.;
Flowers, R. A., II J. Org. Chem. 1999, 64, 5251. (k) Shotwell, J. B.;
Flowers, R. A., II Tetrahedron Lett. 1998, 39, 8063. (l) Shabangi,
M.; Flowers, R. A., II Tetrahedron Lett. 1997, 38, 1137.
(1) For reviews on the application of SmI2 in organic synthesis, see: (a)
Soderquist, J. A. Aldrichim. Acta 1991, 24, 15. (b) Molander, G. A.
Chem. ReV. 1992, 92, 29. (c) Molander, G. A. Org. React. 1994, 46,
211. (d) Imamoto, T. Lanthanides in Organic Synthesis; Academic
Press: London, 1994; p 160. (e) Molander, G. A.; Harris, C. R. Chem.
ReV. 1996, 96, 307. (f) Skrydstrup, T. Angew. Chem., Int. Ed. Engl.
1997, 36, 345. (g) Molander, G. A.; Harris, C. R. Tetrahedron 1998,
54, 3321. (h) Nomura, R.; Endo, T. Chem. Eur. J. 1998, 4, 1605. (i)
Krief, A.; Laval, A.-M. Chem. ReV. 1999, 99, 745. (j) Steel, P. G.
J. Chem. Soc., Perkin Trans. 1 2001, 2727. (k) Agarwal, S.; Greiner,
A. J. Chem. Soc., Perkin Trans. 1 2002, 2033. (l) Kagan, H. B.
Tetrahedron 2003, 59, 10351. (m) Edmonds, D. J.; Johnston, D.;
Procter, D. J. Chem. ReV. 2004, 104, 3371. (n) Gopalaiah, K.; Kagan,
H. B. New J. Chem. 2008, 32, 607.
(3) (a) Farran, H.; Hoz, S. Org. Lett. 2008, 10, 4875. (b) Tarnopolsky,
A.; Hoz, S. Org. Biomol. Chem. 2007, 5, 3801. (c) Tarnopolsky, A.;
Hoz, S. J. Am. Chem. Soc. 2007, 129, 3402. (d) Hoz, S.; Yacovan,
A.; Bilkis, I. J. Am. Chem. Soc. 1996, 118, 261.
9
10.1021/ja903401y CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 10253–10262 10253

A R T I C L E S
Taaning et al.
Scheme 1. Preparation of Ketomethylene Isosteres of Peptides in
a Convergent Manner
steps for the SmI2-promoted reactions. Such knowledge would
be essential for understanding the reactivity of SmI2 and the
nature of reactive intermediates generated after the electron
transfer step. Furthermore, this information could be exploited
for work directed at increasing selectivity of a given reaction
promoted by this lanthanide reagent or even for the design of
new reactions. The seminal work of Curran5 and Molander6 on
the SmI2-mediated reductive coupling of alkyl iodides with
ketones in the presence of a cosolvent such as HMPA provided
experimental evidence for the intermediacy of organosamarium
reagents. Subsequently, the groups of Kagan7 and Flowers8 came
to similar conclusions on the importance of organosamarium
species.
More recently, the laboratories of Hoz9 and Flowers10 have
reported elegant studies in attempts to understand in more detail
the structures of the intermediates involved in ketone reduction
with and without additives present. In studies on the intramo-
lecular carbonyl-olefin addition, Flowers and Prasad10 were able
to demonstrate, by using stopped-flow spectrophotometry, that
HMPA not only increases the reduction potential of SmI2 but
also enhances the reactivity of the radical anion intermediate
formed after the first electron transfer to the ketone functionality
by promoting the formation of a solvent-separated ion pair. In
continued studies on the reduction of diaryl ketones, Hoz and
Farran9 reported that HMPA also plays an important role after
reduction to the ketyl radical anion. In this case, HMPA slowed
the rate of the bimolecular pinacol coupling through complex-
ation of the trivalent lanthanide ion necessary for bridging the
two ketyl radical anions. Considerable efforts have also been
undertaken by the groups of Hoz, Flowers, and Hilmersson to
understand the role of proton donors in the protonation of ketyl
and ketyl-like intermediates.2a,f,3b-d,4b,9
Scheme 2. Coupling Reactions that Inspired to Further Mechanistic
Studies
In collaboration with the Flowers group, we recently reported
mechanistic information about the C-C bond forming step in
an alternative reaction involving the reductive coupling of N-acyl
oxazoldinones with electron deficient alkenes, such as acrylates,
acrylamides, and acrylonitrile.11 This reaction represents a
general method for the preparation of γ-keto esters and amides
in a rapid and convergent manner. In addition, the process
tolerates a wide range of substituents on both the alkene and
the N-acyl moiety of the oxazolidinone, facilitating access to
products of biological relevance such as ketomethylene isosteres
of peptides (Scheme 1).12 Through a combination of cyclic
voltammetry measurments, stopped-flow spectrophotometry, and
examination of substrate reactivity, we were led to conclude
that these reactions proceed by initial electron transfer and
reduction of the olefin to a radical anion species followed by
radical addition to the exocyclic carbonyl group of the N-acyl
oxazolidinone for C-C bond formation.11
Yet, there are still some features of this reaction that need
further attention. For example, we have performed a competition
experiment with a 1:1:1 ratio of N-acetyl oxazolidinone, the
corresponding N-pivaloyl derivative, and t-butyl acrylamide.11
In this case, the tert-butyl ketone was produced in a 78% yield
with only traces of the corresponding methyl ketone, and the
N-acetyl oxazolidinone was almost completely recovered (Scheme
2a). This result is counterintuitive when considering the relative
bulk of the acyl moieties and their possible influence on the
(4) (a) Dahlen, A.; Hilmersson, G. Eur. J. Inorg. Chem. 2004, 17, 3393.
(b) Vestergren, M.; Gustafsson, B.; Johansson, A.; Håkansson, M. J.
Organomet. Chem. 2004, 689, 1723. Dahlen, A.; Hilmersson, G.
Tetrahedron Lett. 2001, 42, 5565. (c) Enemærke, R. J.; Hertz, T.;
Skrydstrup, T.; Daasbjerg, K. Chem.sEur. J. 2000, 6, 3747. (d)
Enemærke, R. J.; Daasbjerg, K.; Skrydstrup, T. J. Chem. Soc., Chem.
Commun. 1999, 343. (e) Hou, Z.; Zhang, Y.; Wakatsuki, Y. Bull.
Chem. Soc. Jpn. 1997, 70, 149. (f) Evans, W. J.; Gummersheimer,
T. S.; Ziller, J. W. J. Am. Chem. Soc. 1995, 117, 8999. (g) Hou, Z.;
Wakatsuki, Y. J. Chem. Soc., Chem. Commun. 1994, 1205. (h) White,
J. P., III; Deng, H.; Boyd, E. P.; Gallucci, J.; Shore, S. G. Inorg. Chem.
1994, 33, 1685.
(5) (a) Curran, D. P.; Totleben, M. J. J. Am. Chem. Soc. 1992, 114, 6050.
(b) Curran, D. P.; Fevig, T. L.; Totleben, M. J. Synlett 1990, 773. (c)
Curran, D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben, M. J. Synlett
1992, 943. (d) Curran, D. P.; Gu, X.; Zhang, W.; Dowd, P. Tetrahedron
1997, 53, 9023.
(6) Molander, G. A.; McKie, J. A. J. Am. Chem. Soc. 1991, 56, 4112.
(7) Namy, J. L.; Collin, J.; Bied, C.; Kagan, H. B. Synlett 1992, 733.
(8) Prasad, E.; Flowers, R. A. J. Am. Chem. Soc. 2002, 124, 6895.
(9) Farran, H.; Hoz, S. Org. Lett. 2008, 10, 865.
(12) (a) Jensen, C. M.; Lindsay, K. B.; Taaning, R. H.; Karaffa, J.; Hansen,
A. M.; Skrydstrup, T. J. Am. Chem. Soc. 2005, 127, 6544. (b) Karaffa,
J.; Lindsay, K. B.; Skrydstrup, T. J. Org. Chem. 2006, 71, 8219. (c)
Ebran, J.-P.; Jensen, C. M.; Johannesen, S. A.; Karaffa, J.; Taaning,
R. H.; Skrydstrup, T. Org. Biomol. Chem. 2006, 3553. (d) Lindsay,
B. K.; Ferrando, F.; Christensen, K. L.; Overgaard, J.; Roca, T.;
Bennasar, M.-L.; Skrydstrup, T. J. Org. Chem. 2007, 72, 4181. (e)
Mittag, T.; Christensen, K. L.; Lindsay, K. B.; Nielsen, N. C.;
Skrydstrup, T. J. Org. Chem. 2008, 73, 1088.
(10) Sadasivam, D. V.; Sudhadevi Antharjanam, P. K.; Prasad, E.; Flowers,
R. A., II J. Am. Chem. Soc. 2008, 130, 7228.
(11) Hansen, A. M.; Lindsay, K. B.; Sudhadevi Antharjanam, P. K.; Karaffa,
J.; Daasbjerg, K.; Flowers, R. A.; Skrydstrup, T. J. Am. Chem. Soc.
2006, 128, 9616.
9
10254 J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009

C-N Bond Rotation in N-Acyl Oxazolidinones
A R T I C L E S
Scheme 3. Setup of Competition Experiments
addition step to the carbonyl moiety. Although N-acetyl ox-
azolidinone can couple to the acrylamide, dimerization of
reduced acrylamide or a second reduction step followed by
protonation to the corresponding proprionamide is more promi-
nent for this coupling reaction than for the pivaloyl analog, again
supporting a slow cross-coupling for the N-acetyl derivative.
Furthermore, recent experiments in our laboratory have shown
that chiral oxazolidinones can be exploited for these tranfor-
mations as well, allowing products from an Evans asymmetric
alkylation to undergo C-C bond formation directly after the
alkylation step (Scheme 2b).13 Nevertheless, yields of these
systems proved to be lower than those of the simple unsubsti-
tuted oxazolidinone, and again, substantial amounts of the
reduced acrylamide were recovered, witnessing to a slow
coupling reaction.
To gain further mechanistic insight into this SmI2-mediated
coupling of N-acyl oxazolidinones and acrylamides, and to
understand the above-described productivity difference between
the different N-acyl oxazolidinones, we embarked on an
elaborate study to provide more specific knowledge on the
determining factors for reactivity. We chose to expand on the
competition experiments between two structurally related
substrates competing for a common SmI2-reduced reagent,
exploiting the fact that the product distribution will reflect the
relative rates of the reaction, thereby disclosing potentially useful
information about relative activation energies. Combining these
experiments with DFT calculations, we report the unexpected
finding that the relative rate constants of product formation
correlate with the barriers for C-N bond rotation of the simple
alkyl N-acyl oxazolidinones. These findings are supported by
additional coupling reactions with analogous carbonyl substrates
that do not display a difference in reactivity depending on
rotational barriers. Furthermore, we demonstrate a means of
exploiting these results for increasing the reactivity of poorly
reactive N-acyl oxazolidinones such as those derived from Evans
chiral auxiliaries.
nones and determining the product distribution by analysis of
the 1H NMR spectrum of the crude product mixture (eq 1), under
the assumption that the N-acyl oxazolidinones have no alterna-
tive reaction pathways. With the tools available to us, it was
not possible to measure the absolute rates of product formation
as the reaction was observed to take place nearly instantaneously
upon addition of SmI2. For a number of selected experiments,
we were able to recover quantitatively all materials, and
importantly, nearly identical RV’s could be measured as a
function of recovered N-acyl oxazolidinones or as a function
of isolated yields of the corresponding γ-ketoamide products.
Equally important for the validity of the found RV’s is that in
all of the experimental reactions carried out, a large excess of
the N-acyl oxazolidinones was added, thereby minimizing
nonproductive reaction pathways, that is, CdC double reduction
and dimerization of the acrylamide.15
The RV’s for the N-acyl oxazolidinones tested were calculated
against N-pivaloyl oxazolidinone 5 as an arbitrary standard (Rref
) tert-butyl) and the results of these competition experiments
are collected in Table 1. It was interesting to observe, that the
N-acyl oxazolidinones with the more bulky acyl moieties
outcompeted the less bulky equivalents by as much as 10 to 30
times (i.e., entries 1-4). Moreover, it is evident that a
heteroatom in the R- or ꢀ-position has a beneficial effect on
the rate of reaction. As can be seen from the RV’s measured
(entries 8 and 9), 8 and 9 were more reactive than the N-butanoyl
oxazolidinone 3 exhibiting similar bulk properties.
We anticipate that the mechanism for the C-C bond for-
mation involves a Lewis acid coordinated N-acyl oxazolidinone
with either a divalent or trivalent samarium metal cation. One
explanation for the higher reactivity of the N-pivaloyl derivative
5 compared to the corresponding N-acetyl compound 1 might
be the difference in Lewis basicity of the two N-acyl carbonyl
groups. Hence, the greater inductive effect of the tert-butyl group
compared to the methyl group would allow a stronger com-
plexation of 5 with a Lewis acid, that is, Sm(II) or Sm(III).16
However, such an explanation does not account for the high
reactivity of the N-methoxyacetyl 9, which in principle should
Results and Discussion
Competition Studies. To understand the productivity differ-
ence previously observed for the SmI2-promoted reductive
coupling of N-acyl oxazolidinones with acrylamide, competition
experiments were carried out between several N-acyl oxazoli-
dinones with variations in the N-acyl side-chain. In a typical
experiment, the two N-acyl oxazolidinones were led to react
with limiting amounts of N-tert-butyl acrylamide in the presence
of H2O. A 0.1 M solution of SmI2 in THF was added dropwise
over 10 min to produce the reactive intermediate in situ, whereby
the relative reactivity of the two N-acyl oxazolidinones deter-
mines the product distribution (Scheme 3).
The fact that the two oxazolidinones compete in the addition
step for the same reactant, being a reduced acrylamide species,
leads to the following simple kinetic expression (eq 1) as
outlined elsewhere:14
kx
ln(([Oxax,start] - [Prodx,end])/[Oxax,start])
ln(([Oxaref,start] - [Prodref,end])/[Oxaref,start])
)
(1)
kref
The relative reactivity values (RV ) kx/kref) can then be
calculated knowing the initial concentration of the oxazolidi-
(15) Reductive coupling between two alkenes is possible, as reported by
Inanaga: Kikukawa, T.; Hanamoto, T.; Inanaga, J. Tetrahedron Lett.
1999, 40, 7497.
(13) Taaning, R. H.; Thim, L.; Karaffa, J.; Campana, A. G.; Hansen, A. M.;
Skrydstrup, T. Tetrahedron 2008, 64, 11884.
(16) For a comprehensive review of Lewis acids in free radical reactions
and the use of oxazolidinone based auxiliaries: Renaud, P.; Gerster,
M. Angew. Chem., Int. Ed. 1998, 37, 2562.
(14) Zhang, Z.; Ollmann, I. R.; Ye, X-. S.; Wischnat, R.; Baasov, T.; Wong,
C.-H. J. Am. Chem. Soc. 1999, 121, 734.
9
J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009 10255

A R T I C L E S
Taaning et al.
Table 1. Competition Experiments between “Simple” N-Acyl
Oxazolidinones (1-4 and 6-10) and N-Pivaloyl Oxazolidinone 5
as the Referencea
to the carbonyl reactivity but rather to the bulk and conforma-
tional properties of the N-acyl oxazolidinones under investiga-
tion. In particular, we wished to address the possibility of the
rotational barrier for the C-N bond being the controlling
parameter for the reactivity of “simple” N-acyl oxazolidinones.
Rotational Barriers. Imides have been shown to be most
stable in their s-trans conformation.17 N-acyl oxazolidinones,
in general, are more stable in their s-trans conformation, and
for N-acyl oxazolidinones lacking substitution on the ring, the
calculated difference in zero-point energy between the s-cis and
18
the s-trans conformation is approximately 8 kcal mol-1
.
However, in the case of a Lewis acid mediated radical addition,
the energy difference would be smaller, and one might envision
that the reactivity of the “simple” N-acyl oxazolidinones (based
on 2-oxazolidinone) could be governed by the barrier for rotation
around the C-N bond. Sibi et al. reported on diastereoselective
radical allylations with N-R-bromoacyl oxazolidinones and
found an interesting improvement of stereoselectivity at higher
temperatures.19 These observations were explained in part by
the rotamer population around the C-N bond of the N-acyl
oxazolidinone favoring the s-trans-rotamer and the hindered
rotation to the s-cis-rotamer for achieving bidentate Lewis acid
coordination of the two carbonyl groups.
To investigate this possibility in our reactions, we conducted
a theoretical study to obtain activation energies (Ea) for the
restricted rotation around the C-N bond. The two conformations
in question were optimized employing DFT at the B3LYP/6-
31+G(d) level of theory.20 Subsequently, the transition states
(TS’s) for the rotations were identified using the Synchronous
Transit-Guided Quasi-Newton method as implemented in Gauss-
ian03.21 The structure obtained at the energy maximum from a
flexible dihedral scan of the C-N bond was used as starting
structure for the TS search. All minima and TS’s were validated
by frequency analysis at the same level of theory showing only
positive second derivatives for the minimum structures while
TS’s gave exactly one imaginary frequency. Finally, the natural
logarithm of the relative reactivity values (ln RV) was plotted
against the calculated activation energies (Ea ) ETS - Es-trans
as depicted in Figure 1.
)
Interestingly, a reasonable correlation could be established
between the reactivity of the N-acyl oxazolidinone and the
barrier for rotation around the C-N bond. The linear correlation
included in Figure 1 for the nonheteroatom substituted N-acyl
oxazolidinones only serves as a guide to the eye as there
undoubtedly will be other factors contributing to the actual Ea,
such as Lewis acid coordination and solvation. These contribu-
tions were assumed to have an approximately equal effect on
the rate of rotation for the various substances. Nevertheless,
a N-acyl oxazolidinone (1.3 equiv), 5 (1.3 equiv), N-tert-butyl acryl-
amide (1.0 equiv), SmI2 (0.1 M in THF, 3 equiv), H2O (8 equiv), THF,
-78 °C, 1 h. b RV’s were calculated from the product distribution
observed in the 1H NMR spectrum of the crude product mixture using
eq 1. c Value obtained from an average of two experiments. d Product
distribution was obtained from isolated yields of the resulting γ-
ketoamides.
experience an opposite inductive effect resulting in a lower
reactivity relative to the N-butanoyl oxazolidinone 3.
Alternatively, the increased reactivity of 9 compared to 3 by
way of the electron-withdrawing effect on the acyl carbonyl,
thereby resulting in a reactivity enhancement toward nucleo-
philic addition, could account for this reactivity difference. This
is not a convincing explanation for the difference in reactivity
observed between the N-acetyl 1 and the N-pivaloyl 5, as the
electron donating ability (inductive effect) of the tert-butyl group
is greater than that of the methyl group, and this should then
reduce the reactivity of the carbonyl group in 5. In addition, 5
would be expected to be less prone to nucleophilic attack
because of increased steric hindrance around the carbonyl group.
Given the experimental trend observed for the reactivity of
the nonheteroatom substituted acyl moieties, we therefore sought
an alternative mechanistic explanation, which was not linked
(17) (a) Noe, E.; Raban, M. J. Am. Chem. Soc. 1975, 97, 5811. (b)
Steinmetz, W. E. J. Am. Chem. Soc. 1973, 95, 2777. (c) Noe, E.;
Raban, M. J. Am. Chem. Soc. 1973, 95, 6118.
(18) See Supporting Information.
(19) Sibi, M. P.; Rheault, T. R. J. Am. Chem. Soc. 2000, 122, 8873.
(20) For theoretical studies on the barriers for rotation around the C-N bond
of similar compounds such as amides, carbamates and carbamoyl
chlorides, where B3LYP calculations have been employed recently: (a)
Wiberg, K. B.; Rablen, P. R.; Rush, D. J.; Keith, T. A. J. Am. Chem.
Soc. 1995, 117, 4261. (b) Cox, C.; Lectka, T. J. Org. Chem. 1998,
63, 2426. (c) Rablen, P. R. J. Org. Chem. 2000, 65, 7930. (d) Deetz,
M. J.; Forbes, C. C.; Jonas, M.; Malerich, J. P.; Smith, B. D.; Wiest,
O. J. Org. Chem. 2002, 67, 3949. (e) Horwath, M.; Benin, V. J. Mol.
Struct. (Theochem) 2008, 850, 105.
(21) (a) Peng, C.; Schlegel, H. B. Israel J. of Chem. 1993, 33, 449. (b)
Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput. Chem.
1996, 17, 49.
9
10256 J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009

C-N Bond Rotation in N-Acyl Oxazolidinones
A R T I C L E S
Figure 1. Natural logarithm of the relative reactivity values (ln RV) plotted against the activation energy (Ea) for rotation around the C-N bond of the
N-acyl oxazolidinone.
Scheme 4. Modes of C-N Bond Rotation for iso-Butanoyl
Oxazolidinone 4
Scheme 5. Alternative Lewis Acid Activation of the R- and
ꢀ-Heteroatom Substituted Equivalents
A to B to C constitutes a faster pathway for C-N bond rotation
than the direct A to D. The conformational change A to B to C
is most likely concerted, and therefore, the true Ea for iso-
butanoyl oxazolidinone 4 C-N bond rotation is presumed to
be between 11.7 and 15.4 kcal mol-1
.
Another obvious deviation from linearity is the higher than
expected relative reactivity of the R- and ꢀ-heteroatom substi-
tuted equivalents 8-10. We propose that these N-acyl oxazo-
lidinones react through an alternative Lewis acid activation,
namely through coordination of the acyl carbonyl and the
heteroatom in the acyl moiety,23 thereby bypassing rotation
around the C-N bond (Scheme 5).
Kinetic Interpretation and Experimental Validation of the
Relative Reactivity Values (RV). A suggested mechanistic
scheme containing the essential kinetic steps is shown in Scheme
6, where “acrylrad” denotes the samarium diiodide reduced
acrylamide, and “Prod” the product.
there were some discrepancies. The iso-butanoyl oxazolidinone
4 gave a rotational barrier that is higher than what would be
expected considering the relative bulk, despite that the observed
reactivity agreed well with formerly obtained rotational barriers
for N,N-disubstituted amides.22 The activation energy for
rotation from the global minimum A in Scheme 4 to the s-cis
conformation was found to be 15.4 kcal mol-1. Yet rotation
could also be divided into two sequential conformational
changes from the ground state, i.e. for the iso-butanoyl oxazo-
lidinone 4 and other nonspherical acyl moieties, one could
envisage that the conformational change is broken into a rotation
around the C(dO)-CR bond (A-B in Scheme 4) followed by
the actual C-N bond rotation (B-C in Scheme 4). These were
found to be 9.4 and 11.7 kcal mol-1, respectively,18 whereby
As hypothesized in the above section, the first step in the
kinetic scheme is the s-trans to s-cis rotation of the N-acyl
oxazolidinone, which is characterized by the equilibrium
(22) (a) Yoder, C. H.; Meyer, K. D.; Sandberg, J. A.; Leung, L. K.;
Wunderlich, M. D. J. Am. Chem. Soc. 1978, 100, 1500. (b) Martins,
M. A. P.; Zanatta, N.; Caro, M. S. B.; Bonacorso, H. G. Magn. Reson.
Chem. 1993, 31, 451. (c) Neuman, R. C., Jr.; Jonas, V. J. Org. Chem.
1974, 39, 925.
(23) For examples of ꢀ-substituent effects on rate and mechanism of ketone
reduction, see: (a) Molander, G. A.; Etter, J. B.; Zinke, P. W. J. Am.
Chem. Soc. 1987, 109, 453. (b) Prasad, E.; Flowers, R. A., II J. Am.
Chem. Soc. 2002, 124, 6357.
9
J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009 10257

A R T I C L E S
Taaning et al.
Scheme 6. Essential Kinetic Steps for the C-C Bond Formation in
the SmI2-Mediated Reductive Coupling of N-Acyl Oxazolidinones
and Acrylamides
Table 2. Competition Reaction between the N-Pivaloyl
Oxazolidinone 5 and the N-Acetyl Oxazolidinone 1 under Varying
Reaction Conditionsa
[N-tert-butyl
acrylamide]t)0
equiv SmI2
added (n)
entry
addition time
RV
1
2
3
4
5
6
7
8
0.10 M
0.10 M
0.10 M
0.10 M
0.10 M
0.10 M
0.10 M
0.10 M
0.02 M
0.50 M
0.10 M
0.10 M
0.33
0.66
1
2
4
3
3
3
3
∼2 minb
∼2 minb
∼2 minb
∼2 minb
∼2 minb
1 minc
0.16
0.12
0.13
0.16
0.14
0.17
0.087
0.087
0.12
0.091
0.072d
0.12e
10 minc
30 minc
10 minc
10 minc
10 minc
10 minc
9
10
11
12
3
3
3
a 1 (1.3 equiv), 5 (1.3 equiv), N-tert-butyl acrylamide (1.0 equiv),
SmI2 (0.1 M in THF, n equiv), H2O (8 equiv), THF, -78 °C, 1 h.
b Manual addition. c Addition using a syringe pump. d Ratio of the
concentrations of 1 and 5 was 4:1. e Sm(OTf)3 (1.5 equiv) was added to
the reaction mixture.
constant Krot ) krot/k-rot with krot denoting the rate constant for
the s-trans to s-cis rotation and k-rot the rate constant for the
reverse rotation. In the samarium diiodide containing solutions,
the driving force for the s-trans to s-cis rotation is expected to
be enhanced because of the ability of the metal center to
coordinate with the two CdO groups of the N-acyl oxazolidi-
none, resulting in a bidentate coordination of samarium. We
assume that this coordination occurs concertedly with the
rotation process, but even if it should take place in a follow-up
step the kinetic expression will be the same as long as this step
is fast. The final essential reaction is the addition reaction
between the s-cis conformation of the N-acyl oxazolidinone
(coordinated with the samarium center) and a SmI2-reduced form
of N-t-butyl acrylamide. Using the steady-state assumption on
the s-cis conformation the following kinetic expression ensues:24
the limiting first case (eqs 3 and 5) the kinetics becomes more
complex in the sense that the addition step is rate-controlling
with a pre-equilibrium process (s-trans to s-cis rotation)
established first.
To ascertain that the reaction mechanism did not change as
the reaction proceeded, we stopped the reaction prematurely
by the addition of a limited amount of samarium diiodide (Table
2, entries 1-5). Since the product distributions in all cases led
to nearly identical RV’s, the mechanism most likely does not
change during the progress of the reaction. Furthermore, the
observation that the variation in the concentration from 0.02 to
0.5 M of N-tert-butyl acrylamide (Table 2, entries 7 and 9-10)
and a variation in the ratio of the concentration of 1 and 5 (1:1
and 4:1, Table 2, entries 7 and 11, respectively) produced no
significant change in RV’s. We can therefore conclude that the
reaction is first order with respect to the N-acyl oxazolidinone
as assumed for eq 1.
δ[Prod]
δt
-δ[s-trans]
δt
)
) k[s-trans], where
krotkadd[acrylrad]
(k-rot + kadd[acrylrad])
k )
(2)
While the technique of addition, that is, manual or a syringe
pump, exerted no influence on the results (Table 2, entries 5
and 6) a decrease in RV was observed if the addition time was
increased (Table 2, entries 6-8). We have no certain explanation
for this phenomenon but temperature increases during rapid
additions of SmI2, which involve a relatively large volume of
an ethereal samarium diiodide solution at room temperature
compared to the solution in the flask at -78 °C. For addition
times equal to or longer than 10 min consistent results were
obtained. Noteworthy, the addition of 1.5 equivalents of
Sm(OTf)3 produced no significant change in RV (Table 2, entry
12), indicating that even in the presence of high concentrations
of a Lewis acid, the s-trans/s-cis equilibrium is shifted toward
the left.
Additional Experimental Evidence Supporting the Impor-
tance of C-N Bond Rotation. In Figure 1, we observed an
approximately linear relationship between the reactivity of the
simple alkyl N-acyl oxazolidinones toward radical addition and
the barrier for rotation around the C-N bond involving the
exocyclic CdO group. Also from this graph, it can be deduced
that the reactivity of this class of substrates augments with
increasing bulk of the acyl moiety. Nevertheless, these two
observations alone do not exclude the possibility that the driving
force for the reactivity of the acyl carbonyl group could instead
be pyramidalization in the C-C bond forming step. The release
of allylic strain by pyrimidalization at the CdO group upon
Two limiting situations then arise according to the relative
size of k-rot and kadd[acrylrad]:
krotk
For k-rot . kadd[acrylrad]: k )
add[acrylrad] )
k-rot
Krotkadd[acrylrad] (3)
For k-rot , kadd[acrylrad]: k ) krot (4)
Applying the two limiting situations to the expression for
RV (eq 1) gives:
kx
Krot,xkadd,x
For k-rot . kadd[acrylrad]:
)
(5)
(6)
kref
Krot,refkadd,ref
kx
krot,x
For k-rot , kadd[acrylrad]:
)
kref
krot,ref
If the second limiting case (eqs 4 and 6) applies, the
expectation is thus, that the relative reactivity values exhibit a
correlation with the activation barrier for the s-trans to s-cis
rotation as is most likely the case for the “simple” N-acyl
oxazolidinones, as shown in Figure 1. On the other hand, for
(24) Espenson, J. H. in Chemical Kinetics and Reaction Mechanism, 2nd
ed.; McGraw-Hill: New York, 1995; Chapter 4.
9
10258 J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009

C-N Bond Rotation in N-Acyl Oxazolidinones
A R T I C L E S
Scheme 7. Competition Studies Involving Carbonyl Compounds in which the Effect of C-N Bond Rotation is Excluded
addition of the nucleophilic carbon-centered radical would be
expected to be greater with increasing bulk of the acyl moiety.
To investigate whether the observed relative reactivities are
governed by the C-N bond rotation or pyramidalization, we
conducted further competition experiments on analogous car-
bonyl substrates, which inherently would exclude a possible
dependence on the C-N bond rotation (Scheme 7).
formation of one activated N-acyl oxazolidinone over the other
determined by the barrier for rotation around the C-N bond.
Finally, because substrates such as N-acyl succinimide 12
do not require C-N bond rotation for the C-C bond forming
step with the SmI2-reduced acrylamide as for the corresponding
N-acyl oxazolidinones 5, it would be expected that 12 should
display higher reactivity in a competition experiment between
these two substrates for N-tert-butyl acrylamide. As illustrated
in Scheme 7c, this turned out to be the case, where 1.3
equivalents of 5 and 12 were reacted with 1 equivalent of the
acrylamide leading to a 77% isolated yield of the coupling
product and with an 84 and 7% recovery of 5 and 12,
respectively. This corresponds to an RV of 14.4 for the
N-pivaloyl succinimide.
Two types of imide substrates were chosen, which in both
cases allow bidentate coordination with a Lewis acid. In the
first set of experiments, the N-tert-butyl acrylamide was reacted
in the presence of SmI2 with an excess of the N-pivaloyl
succinimide 12 and N-acetyl succinimide 1325,26 in a 1:1 ratio
under identical reaction conditions as described for the competi-
tion experiments with the N-acyl oxazolidinones. This resulted
in the exclusive formation of the methyl ketone 15 in a 75%
isolated yield (Scheme 7a). It was verified in separate experi-
ments that the major product from the coupling reaction with
the individual succinimides was indeed the tert-butyl ketone
11 obtained in a 60% isolated yield, and the methyl ketone 15
in a 75% yield. In a second set of experiments, the mixed imides
14 were reacted with the N-tert-butyl acrylamide and SmI2 to
produce preferentially the methyl ketone 15 (Scheme 7b). In
both these sets of experiments steric encumbrance around the
reacting carbonyl group appears to be the dominant factor for
the product distribution rather than release of strain through
pyramidalization of the CdO carbon in the addition step. Since
this must also be valid for the N-acyl oxazolidinones, it leads
us to conclude, that the relative reactivities observed in Table
1 does not originate from discrimination between two activated
N-acyl oxazolidinones, but rather from the kinetically selective
Investigation of Chiral N-Acyl Oxazolidinones with Ring
Substitution at the 4-Position. As previously stated in the
introductory section, one of our incentives for conducting this
mechanistic study was to understand why the productivity of
the coupling reaction drops for substrates possessing a “simple”
oxazolidin-2-one auxiliary to those with chiral 4-substituted
oxazolidin-2-one auxiliaries. A small set of competition experi-
ments with 4-substituted-N-acyl oxazolidinones was therefore
carried out, as illustrated in Table 3. It was necessary to use a
less reactive reference compound than the N-pivaloyl oxazoli-
dinone 5, as the more substituted oxazolidinone derivatives
proved much less reactive than the previous cases tested. Instead,
the compounds were competed against either the N-acetyl
oxazolidinone 1 or the N-propanoyl oxazolidinone 2 to facilitate
the 1H NMR analysis of the crude product mixture. To validate
that the RV’s were comparable, we conducted a competition
study between 2 and 1 giving k2/k1 ) 1.21 (Table 3; entry 1)
which by multiplying with 0.088 for k1/k5 (Table 1; entry 1)
can be converted to a RV of 0.11 for 2. This value is close to
(25) Heller, G.; Jacobsohn, P. Ber. 1921, 54, 1107. McAlees, A. J.;
McCrindle, R. J. Chem. Soc. (C) 1969, 2425.
(26) For the synthesis of these imides, see Supporting Information.
9
J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009 10259

A R T I C L E S
Taaning et al.
Table 3. Competition Studies with the 4-Substituted
Oxazolidinonesa
Figure 2. Natural logarithm to the relative reactivity values (ln RV) plotted
against the activation energy (Ea) for C-N bond rotation.
Figure 3. Proposed transition-state for addition to N-acyl oxazolidinones
possessing substitution at the 4-position and illustrating the arising syn-
pentane interaction.
a substituent in the 4-position of the oxazolidinone ring and
the R-carbon of the incoming nucleophilic radical in the tran-
sition-state for addition, as depicted in Figure 3. The horizontal
punctured line inserted in Figure 2 illustrates the seemingly equal
reactivity of the 4-methyl substituted compounds indicating that
substitution in the 4-position of the 2-oxazolidinone ring dictates
the reactivity. The reactivity is neither affected by the barrier
of rotation around the C-N bond nor by the bulk properties
presented by the R group of the N-acyl moiety. This is difficult
to explain were the nucleophilic radical to attack the N-acyl
carbonyl on the opposite side of the 4-methyl substituent (Figure
3b), as the rate of addition in that case must be expected to be
influenced by the size of R. Therefore, we suggest that the
radical addition possibly occurs from the same side of the N-acyl
carbonyl as the 4-methyl substituent for (S)-4-methyl-N-pivaloyl
oxazolidinone 17 and possibly also for 20 (Figure 3a). In
conclusion, while the determining factor for product distribution
was shown to be the rate of rotation around the C-N bond for
the “simple” N-acyl oxazolidinones, the reactivities of the
4-substituted oxazolidinones are clearly dominated by the
activation energy for addition due to the arising syn-pentane
interactions in the transition-state for addition. Kinetically this
would then correspond better to the limiting first case presented
(eqs 3 and 5).
a N-acyl-4-substituted oxazolidinone (1.3 equiv), 1 or 2 (1.3 equiv),
N-tert-butyl acrylamide (1.0 equiv), SmI2 (0.1 M in THF, 3 equiv), H2O
(8 equiv), THF, -78 °C, 1 h. b The RV’s are the relative reactivities of
product formation using N-pivaloyl oxazolidinone
5 as reference
compound, calculated from the two relevant sets of kx/kref values given
in Table 3 and 1 (i.e., 0.088 for N-acetyl oxazolidinone 1 and 0.094 for
N-propanoyl oxazolidinone 2). c The designated compound does not
react under the specified conditions.
that of 0.094 obtained in the direct competition experiment with
the N-pivaloyl oxazolidinone 5.
In Figure 2, the relative reactivity values (ln RV) are plotted
against the activation energy (Ea) for C-N bond rotation to
illustrate trends in reactivity for the 4-substituted oxazolidinones.
As we have previously stated, A1,3 strain does not contribute to
a significant extent to the reactivity of the “simple” N-acyl
oxazolidinones. The low reactivity observed for the (S)-4-
methyl-N-pivaloyl oxazolidinone (17), (S)-4-methyl-N-acetyl
(16), and the 4,4-dimethyl-N-acetyl oxazolidinones (18) (Table
3; entries 2-4) compared to the 4-nonsubstitued reference
compounds 1 and 2 leads us to conclude that addition of the
reduced acrylamide species must occur from the s-cis conforma-
tion, as it would otherwise be difficult to explain the profound
effect of the 4-methyl substituent.
Effect of Lewis Acids on the Reactivity of the 4-Substituted
Oxazolidinone Derived Substrates. In an attempt to increase the
reactivity of the 4-substituted-N-acyl oxazolidinones toward
radical addition, we conducted a small screening on the effect
of added Lewis acids, which have previously been used with
success in the stereoselective radical allylations with chiral
oxazolidinone auxiliaries.27 We hypothesized that the addition
of an appropriate Lewis acid could facilitate the rotation around
the C-N bond, as well as increase the reactivity of the CdO
For the 4-substituted cases, pyramidilization of the acyl
carbonyl group will result in a syn-pentane interaction between
9
10260 J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009

C-N Bond Rotation in N-Acyl Oxazolidinones
A R T I C L E S
Table 4. Lewis Acids Promoting the Reductive Coupling of Less Reactive N-Acyl Oxazolidinonesa
a N-acyl oxazolidinone (1.5 equiv), N-tert-butyl acrylamide (1.0 equiv), SmI2 (0.1 M in THF, 2.5 equiv), H2O (8 equiv), THF, -78 °C, 1 h. b Isolated
yields after column chromatography. c Yields in parentheses are conversions of acrylamide based on the ratio of substrate to product obtained from
analysis of the 1H NMR spectrum of the crude product and encompasses both the ketone and the γ-hydroxy product. d The major byproduct was
propionamide. e The major byproduct could not be identified, but neither the reduced acrylamide nor the γ-hydroxy product was observed.
carbon, thereby increasing the rate of the addition step over
reduction/dimerization of the acrylamide.
increase from 61% without additive to a satisfying 87% for 20
and from 33 to 59% for 21 with this Lewis acid (Table 4, entries
1, 6, 10, and 11). No reduced acrylamide (propionamide) was
detected with the use of Zn(OTf)2 despite that this is a major
byproduct for substrates 20 and 21 without the Lewis acid being
present. On the other hand Lewis acids have no effect on the
propensity of the hemiketal to collapse, as is evident from the
near equal amount of γ-hydroxy amide produced when changing
Lewis acid (Table 4, entries 1, 2, 4, and 6). Although Sm(OTf)3
was shown to have no effect on the relative rate of reaction
(Table 2, entry 12), it has a significant effect on yields producing
a conversion of 83% (Table 4, entry 7).
SmI2-mediated reductive coupling of N-tert-butyl acrylamide
with 20 results in the formation of two major products (Table
4, entry 1). One is the desired γ-keto amide 22, and the other
is the corresponding γ-hydroxy amide, presumably originating
from the premature collapse of the resulting hemiketal inter-
mediate after the addition step and subsequent reduction of the
exposed ketone.
Addition of MgBr2, MgI2 and Mg(OTf)2 resulted in a modest
increase of overall yields (γ-keto- and γ-hydroxy amide) and
no significant effect of the counterion was observed (Table 4,
entries 2-4). Mg(OTf)2 also has a beneficial effect on the yield
for the “simple” N-acyl oxazolidinone substrate 4. Adding
Zn(OTf)2, however, resulted in a considerable overall yield
Conclusions
In summary, an elaborate study was undertaken to examine
some of the post electron transfer steps in the SmI2-mediated
coupling of N-acyl oxazolidinones with acrylamides, repre-
senting a novel C-C bond forming reaction for accessing
γ-ketoamides. For the simple alkyl N-acyl oxazolidinones,
our results suggest that the efficiency of the coupling depends
on the propensity for C-N bond rotation from the s-trans to
s-cis conformation. This was illustrated by an almost linear
relationship when ln RV was plotted against the activation
barriers for C-N bond rotation (s-trans to s-cis) obtained
by DFT calculations. In contrast, substrates bearing chiral
(27) For some pertinent references, see: (a) Hein, J. E.; Zimmerman, J.;
Sibi, M. P.; Hultin, P. G. Org. Lett. 2005, 7, 2755. (b) Sibi, M. P.;
Liu, P.; Ji, J.; Hajra, S.; Chen, J. J. Org. Chem. 2002, 67, 1738. (c)
Sibi, M. P.; Ji, J.; Sausker, J. B.; Jasperse, C. P. J. Am. Chem. Soc.
1999, 121, 7517. (d) Sibi, M. P.; Jasperse, C. P.; Ji, J. J. Am. Chem.
Soc. 1995, 117, 10779. (e) Sibi, M. P.; Ji, J.; Wu, J. H.; Gu¨rtler, S.;
Porter, N. A. J. Am. Chem. Soc. 1996, 118, 9200. (f) Sibi, M. P.; Ji,
J. J. Org. Chem. 1997, 62, 3800. (g) Sibi, M. P.; Ji, J. J. Org. Chem.
1996, 61, 6090. (h) Sibi, M. P.; Jasperse, C. P.; Ji, J. J. Am. Chem.
Soc. 1995, 117, 10779. (i) Sibi, M. P.; Porter, N. A. Acc. Chem. Res.
1999, 32, 163.
9
J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009 10261

A R T I C L E S
Taaning et al.
the iNANO and OChem Graduate Schools, the Danish Center for
Scientific Computing, and Aarhus University.
4-substituted oxazolidinones appear to be dominated by the
activation energy for addition because of arising syn-pentane
interactions in the transition-state for C-C bond formation.
What implications such results will have for other synthetic
transformations which exploit this important auxiliary will
be the topic of future investigations.
Supporting Information Available: Experimental details and
1
copies of H NMR and 13C NMR spectra for all the coupling
products and crude 1H NMR spectra of competition experiments.
This material is available free of charge via the Internet at http://
pubs.acs.org.
Acknowledgment. We are deeply appreciative of generous
financial support and resources from the Danish National Research
Foundation, the Lundbeck Foundation, the Carlsberg Foundation,
JA903401Y
9
10262 J. AM. CHEM. SOC. VOL. 131, NO. 29, 2009