Bond rotation dynamics of N-cycloalkenyl-N-benzyl α-haloacetamide derivatives

Article abstract of DOI:10.1021/jo900491w

(Chemical Equation Presented) Barriers to rotation of the N-alkenyl bond in a series of N-cycloalkenyl-N-benzyl α-haloacetamide derivatives have been measured by variable-temperature NMR experiments. The barriers range from 10 to 18 kcal/mol, depending on ring size and on substituents on the cycloalkene and the amide. The observed trends aid in the design of substituent combinations that provide resolvable enantiomers or diastereomers at ambient temperature. The compounds undergo 4-exo and 5-endo radical cyclizations at rates that may be faster or slower than the estimated rate of N-alkenyl bond rotation in the derived radicals, depending on the substituents.

Full text of DOI:10.1021/jo900491w

Bond Rotation Dynamics of N-Cycloalkenyl-N-benzyl
r-Haloacetamide Derivatives
David B. Guthrie,† Krishnan Damodaran,† Dennis P. Curran,*,† Paul Wilson,‡ and
Andrew J. Clark*,‡
Department of Chemistry, UniVersity of Pittsburgh, Pittsburgh PennsylVania 15260, and Department of
Chemistry, UniVersity of Warwick, CoVentry, W Midlands, CV4 7AL, U.K.
curran@pitt.edu; a.j.clark@warwick.ac.uk
ReceiVed March 5, 2009
Barriers to rotation of the N-alkenyl bond in a series of N-cycloalkenyl-N-benzyl R-haloacetamide
derivatives have been measured by variable-temperature NMR experiments. The barriers range from 10
to 18 kcal/mol, depending on ring size and on substituents on the cycloalkene and the amide. The observed
trends aid in the design of substituent combinations that provide resolvable enantiomers or diastereomers
at ambient temperature. The compounds undergo 4-exo and 5-endo radical cyclizations at rates that may
be faster or slower than the estimated rate of N-alkenyl bond rotation in the derived radicals, depending
on the substituents.
Introduction
Ishibashi and co-workers (1a f 3)2a and 5-endo (4b f 6) and
4-exo (7 f 9) atom transfer cyclizations by Clark and
co-workers.3a Related cyclizations to make larger rings are
known, and the functionalized lactam products from such
reactions are useful in alkaloid synthesis.6
Rotational features of amides can dictate the success or failure
of their radical cyclizations,7 and we commented on the
importance of the amide N-CO bond in a recent mechanistic
analysis of radicals like 2.5 Figure 2 shows rotations of both
the amide N-CO bond and the N-alkenyl bond of typical radical
precursor 1. Comparable rotations are available to the derived
radical 2. It is evident that radicals like 2 can only cyclize as
the E-rotamer of the amide N-CdO bond. In the Z-rotamer,
the radical cannot reach the alkene. So the amide bond rotation
dynamics of both the radical precursor and the radical are
important in such cyclizations. Fortunately, enamides typically
prefer the E-rotamer,8 so their radicals are formed in a geometry
that is predisposed to cyclize.
Though still usually classified as disfavored, 5-endo radical
cyclizations are increasingly recognized as invaluable synthetic
transformations in certain settings.1 The most widely used class
of 5-endo cyclizations is that of radicals derived from
R-haloenamides.2-5 Such radicals can also provide products of
4-exo cyclization, another disfavored pathway, if substitution
patterns dictate. Reactions have been conducted by tin hydride
and atom transfer methods (among others) and produce reduced,
isomerized, or oxidized products depending on the reaction
conditions and the nature of the precursor.5 Figure 1 shows
representative examples of a 5-endo tin hydride cyclization by
† University of Pittsburgh.
‡ University of Warwick.
(1) (a) Ishibashi, H.; Sato, T.; Ikeda, M. Synthesis 2002, 695–713. (b)
Alabugin, I. V.; Manoharan, M. J. Am. Chem. Soc. 2005, 127, 9534–9545.
(2) (a) Tamura, O.; Matsukida, H.; Toyao, A.; Takeda, Y.; Ishibashi, H. J.
Org. Chem. 2002, 67, 5537–5545. (b) Ishibashi, H.; Ishita, A.; Tamura, O.
Tetrahedron Lett. 2002, 43, 473–475.
(3) (a) Clark, A. J.; Dell, C. P.; Ellard, J. M.; Hunt, N. A.; McDonagh, J. P.
Tetrahedron Lett. 1999, 40, 8619–8623. (b) Clark, A. J.; Dell, C. P.; McDonagh,
J. P. C. R. Acad. Sci. Ser. II C 2001, 4, 575–579. (c) Clark, A. J.; Dell, C. P.;
McDonagh, J. M.; Geden, J.; Mawdsley, P. Org. Lett. 2003, 5, 2063–2066.
(4) Chatgilialoglu, C.; Ferreri, C.; Guerra, M.; Timokhin, V.; Froudakis, G.;
Gimisis, T. J. Am. Chem. Soc. 2002, 124, 10765–10772.
(5) Curran, D. P.; Guthrie, D. B.; Geib, S. J. J. Am. Chem. Soc. 2008, 130,
8437–8445.
(6) Taniguchi, T.; Yonei, D.; Sasaki, M.; Tamura, O.; Ishibashi, H.
Tetrahedron 2008, 64, 2634–2641.
(7) (a) Stork, G.; Mah, R. Heterocycles 1989, 28, 723–727. (b) Curran, D. P.;
Tamine, J. J. Org. Chem. 1991, 56, 2746–2750. (c) Musa, O. M.; Choi, S. Y.;
Horner, J. H.; Newcomb, M. J. Org. Chem. 1998, 63, 786–793. (d) Musa, O. M.;
Horner, J. H.; Newcomb, M. J. Org. Chem. 1999, 64, 1022–1025.
(8) Ahlbrecht, H.; Becher, G.; Blecher, J.; Kalinowski, H. O.; Raab, W.;
Mannschreck, A. Tetrahedron Lett. 1979, 20, 2265–2268.
4262 J. Org. Chem. 2009, 74, 4262–4266
10.1021/jo900491w CCC: $40.75 2009 American Chemical Society
Published on Web 05/04/2009

Bond Rotation Dynamics of Enamides
calculations.4 We suggest that a key reason for the success of
such cyclizations is that the radicals are already twisted in the
ground state, as shown in Figure 2. Indeed, cyclizations of
radicals like 2 occur not by twisting away from planarity, as
suggested by the flat structure in Figure 1, but toward planarity,
as shown by E,P-2 in Figure 2.
In other words, both the radicals and the radical precursors
are axially chiral and exist as atropisomers.9 For example, an
angle of 74° is observed for the N-alkenyl bond in the X-ray
crystal structure of N-cyclohexenyl-R-chloroacetamide 1a (X
) Cl).5 In this respect, N-acylenamides resemble acetanilides,
which are also known to be twisted and reach angles of 70-90°
if one or both of the ortho substituents are not hydrogen.10 If
rotation barriers are high enough, then the axial chirality of
suitable acetanilides can be retained or transferred in a diverse
set of asymmetric reactions.11 By analogy, chirality transfer
could also be possible for enamides, depending on whether
cyclization is faster than N-alkenyl bond rotation.
To better contemplate the prospects of chirality transfer in
radical or other reactions of enamides, a basic understanding
of their rotation dynamics is needed. Ahlbrecht and co-workers
described rotation barriers of several acyclic enamides and
related molecules in 1979.8 Here we report variable-temperature
(VT) NMR experiments that provide rotation barriers of a
complementary series of R-haloenamides with N-cycloalkenyl
substituents. Taken together, the past and new results provide
a quantitative footing that can be used to design axially chiral
enamides that are stable at ambient temperatures or that undergo
onward reactions that are faster (or slower) than bond rotations.
FIGURE 1. Typical 5-endo and 4-exo radical cyclizations of R-ha-
loenamides.
Results and Discussion
The experiments were initiated with N-benzyl-N-cyclohexenyl
R-chloroacetamide 1a and the analogous bromide 1b and iodide
1c as part of a comprehensive mechanistic study of the
cyclizations of the derived radicals.5 Iodide 1c was dissolved
in CDCl3 with a small amount of tetramethylsilane for use as a
1
chemical shift and line width standard. H NMR spectra were
recorded at temperature intervals in the range of 215-297 K.
Portions of these spectra are shown in Figure 3, while Figure 4
shows the rotation process for 1c with the protons relevant for
the analysis.
Consistent with the expectation that 1c was mainly the
E-amide rotamer in solution, its NMR spectrum at 297 K
showed one singlet apiece for the benzyl methylene group and
the methylene group R to the amide. As the sample was cooled,
decoalescence of the benzyl and R-amide proton resonances
occurred. At 215 K, the benzyl protons gave rise to mutually
coupled doublets (J ) 14.6 Hz) at 5.04 and 4.22 ppm. Similarly,
the R-amide protons gave rise to doublets (J ) 9.3 Hz) at 3.99
1
and 3.79 ppm. The variable-temperature H NMR spectra of
chloride 1a and bromide 1b are shown in the Supporting
Information, and these displayed temperature-dependent decoa-
lescences and couplings similar to those of 1c.
FIGURE 2. Rotamers of radical precursor 1 and cyclizations of derived
radical 2.
(9) Wolf, C. Dynamic Stereochemistry of Organic Compounds; RSC Publish-
ing: Cambridge, UK, 2008; pp 94-104; see especially “non-biaryl atropisomers”.
(10) (a) Curran, D. P.; Qi, H.; Geib, S. J.; DeMello, N. C. J. Am. Chem.
Soc. 1994, 116, 3131–3132. (b) Curran, D. P.; Hale, G. R.; Geib, S. J.; Balog,
A.; Cass, Q. B.; Degani, A. L. G.; Hernandes, M. Z.; Freitas, L. C. G.
Tetrahedron: Asymmetry 1997, 8, 3955–3975.
(11) (a) Curran, D. P.; Chen, C. H. T.; Geib, S. J.; Lapierre, A. J. B.
Tetrahedron 2004, 60, 4413–4424. (b) Petit, M.; Geib, S. J.; Curran, D. P.
Tetrahedron 2004, 60, 7543–7552. (c) Lapierre, A. J. B.; Geib, S. J.; Curran,
D. P. J. Am. Chem. Soc. 2007, 129, 494–495.
Less evident but perhaps equally important are the rotational
features of the N-alkenyl bond. Conventional drawings of
enamide radicals like those in Figure 1 convey the mistaken
impression that 5-endo cyclizations may be difficult or even
impossible because the radical and alkene orbitals seem to be
parallel and do not overlap. Intuition suggests that twisting must
occur to reach a transition state, and this is confirmed by
J. Org. Chem. Vol. 74, No. 11, 2009 4263

Guthrie et al.
FIGURE 5. Eyring plot of VT NMR data of 1a (diamonds), 1b
(squares), and 1c (triangles).
1
FIGURE 3. Expansion of the H NMR spectra of 1c at 215-297 K.
singlets at room temperature, so only cooling was required, while
those of 11 were already a well-resolved pair of doublets, so
only heating was needed. The other compounds were in
between, with more or less broad (in some cases, extremely
broad) resonances, so samples were both cooled and heated.
Enamide 12 has an additional stereocenter, so its rotamers are
diastereomers, not enantiomers. The equilibrium constant for
the two diastereomers 12 is about 2, and we did not attempt to
assign the structure of the major rotamer.
To extend the temperature range for data collection in both
directions, we used toluene-d8 as the solvent for this series of
variable-temperature experiments. The spectra are gathered in
the Supporting Information along with the derived Eyring plots.
The benzylic protons were well-resolved from other resonances
in every case, so we made these the focal points of the line-
shape analyses.
Table 1 lists the rate constants at room temperature and the
activation parameters for all the substrates. Among the interest-
ing comparisons, notice that cyclohexenamide 4b with a
quaternary carbon on the acyl group has a barrier of 13.3 kcal/
mol (entry 7). This is more than 1 kcal/mol higher than the
barriers of halides 1a-c with a secondary carbon on the acyl
group (entries 1-5). Varying the ring size of the cycloalkenyl
group (4b-e, entries 7-10) does not have much effect on the
barrier (12.4-13.5 kcal/mol), except when the ring shrinks to
five-membered 4a (entry 6, 10.0 kcal/mol). Presumably, as the
ring is pinned back by its decreasing size, the barrier goes down.
Adding a methyl substituent on the N-benzyl group to give
phenethyl analogue 12 boosts the barrier by about 1 kcal/mol
(entry 13).
With an sp3-hybridized atom (CH2) at the ꢀ′-position of the
enamide, 2-tetralone analogue 7 rather closely resembles cy-
clohexamide 4b, though its rotation barrier at 11.7 kcal/mol is
reduced by about 1.6 kcal/mol (entry 11). In contrast, with an
sp2-hybdrized atom at the ꢀ′-position (aromatic C), 1-tetralone
enamide 11 is more related to an anilide than a cyclohexena-
mide. It has the highest rotation barrier observed at 18 kcal/
mol (entry 12), though this is surely due primarily to the fact
that it is also the only compound in the series to have a non-
hydrogen substituent (the phenyl ring of the tetralone) in the
ꢀ′-position. Indeed, only the beginnings of coalescence were
observed in the high-temperature spectra for 11, so its barrier
measurement may not be as accurate as the others.
FIGURE 4. N-Alkenyl rotation of 1a-c with diastereotopic protons.
This behavior is not consistent with standard amide N-CO
bond rotation since such a process should cause all signals to
double and should not cause geminal protons to become
diastereotopic. Instead, the results are consistent with existence
of largely a single E-amide rotamer in solution. The decoales-
cences are caused by slowing of the N-cyclohexenyl bond
rotation, which reveals the diastereotopicity of the pairs of
geminal protons.
Using the WINDNMR 7.1 line-shape analysis program12 to
analyze the benzyl peaks of 1a-c, we determined the rotational
rate constant krot at each temperature T. The standard Eyring
plots for chloride 1a, bromide 1b, and iodide 1c are shown in
Figure 5. The three lines nearly coincide, showing that the
dynamics of the molecules are similar. The activation barriers
at 298 K (∆Gq) were also similar: 12.1 kcal/mol for 1a, 11.7
kcal/mol for 1b, and 11.9 kcal/mol for 1c. The activation
enthalpies and entropies were calculated in the standard way
and are shown in Table 1. To assess solvent effects, the barriers
for 1b were also determined in toluene-d8 (entry 3, 11.1 kcal/
mol) and methanol-d4 (entry 4, 11.6 kcal/mol).
We next extended the studies to the series of eight R-bro-
moisobutenamides shown in Figure 6. Most of these precursors
undergo 5-endo cyclizations,3 the exception being 7, which
prefers 4-exo cyclization (see Figure 1). The 300 MHz NMR
spectra of 1a-c at room temperature were simple, and peak
broadening and decoalescence were only observed upon cooling.
Only in hindsight did we notice that the benzyl methylene
singlets were slightly broadened at rt. In contrast, the rt spectra
of the amides in Figure 6 exhibited a range of different
behaviors. For example, the benzylic protons of 4a were sharp
Though ꢀ-tetralone-derived enamide 7 is closely related to
cyclohexenamide 4b, its barrier to rotation is about 1.6 kcal/
(12) Reich, H. J. J. Chem. Educ. Software 1996, 3D2.
4264 J. Org. Chem. Vol. 74, No. 11, 2009

Bond Rotation Dynamics of Enamides
TABLE 1. N-Alkenyl Bond Rotation Rates and Activation Parameters from Variable-Temperature NMR Experiments
entry
substrate
temp range
sola
krot 298 K (s-1
)
∆Hq (kcal/mol)
∆Sq (cal/mol K)
∆Gq (kcal/mol)
298
1
2
3
4
5
6
7
8
1a
1b
1b
1b
1c
4a
4b
4c
4d
4e
7
297-215 K
323-213 K
323-193 K
298-213 K
270-216 K
290-186 K
311-233 K
350-250 K
341-230 K
341-230 K
313-203 K
379-341 K
350-271 K
C
C
T
M
C
T
T
T
T
T
T
T
T
8.49 × 103
1.58 × 104
4.54 × 104
1.96 × 104
1.12 × 104
3.11 × 105
1.06 × 103
7.35 × 102
3.09 × 103
4.88 × 103
1.77 × 104
4.08 × 10-1
1.83 × 102
3.64 × 102
8.9
8.5
8.1
8.2
9.2
-10.6
-10.6
-10.0
-11.3
-9.0
-6.6
-6.6
-5.7
-4.3
-3.5
-6.3
-20.4
-5.4
-4.0
12.1
11.7
11.1
11.6
11.9
10.0
13.3
13.5
12.7
12.4
11.7
18.0
14.4
14.0
8.0
11.3
11.8
11.4
11.4
9.8
11.9
12.7
12.7
9
10
11
12
13
11
12b
a C is CDCl3, T is C6D5CD3, M is CD3OD. b Rotamers are diastereomers, so forward and reverse rates were determined.
FIGURE 8. Designing resolvable N-cycloalkenyl amides.
FIGURE 6. Structures of R-haloenamides studied by VT NMR.
FIGURE 7. X-ray crystal structure of 7.
mol lower (compare entries 7 and 11). Seeking insight into this
difference, we crystallized 7 and solved the X-ray structure. This
exhibited several interesting features that are summarized in
Figure 7. The torsion angle of the key N-alkenyl bond is only
56°. This is smaller than usually observed for related anilides9
and for enamide 1a.5 Perhaps the additional overlap of the
nitrogen lone pair through the alkene to the fused phenyl ring
reduces both the twist angle and the rotation barrier compared
to 1b and 4b. In addition, the amide is not completely planar.13
For example, the sum of the bond angles to nitrogen is not 360°
but 356.9°, and the torsion angle of the amide C-N bond to
the cycloalkene is not 180° but is only 152°. So the partial
rehybridization of N from sp2 toward sp3 may also reduce the
rotation barrier.
FIGURE 9. Comparing rate constants for onward reactions of radical
2. Cyclization competes with rotation.
possible to design resolvable enamides simply by changing the
hydrogen substituents on the ꢀ- and ꢀ′-carbons to non-hydrogen
ones. Indeed, Ahlbrecht and co-workers partially resolved
compound 13 (Figure 8) about 30 years ago and measured its
rotation barrier at about 25 kcal/mol. A similarly substituted
cyclic analogue such as 14 will probably have a higher rotation
barrier (because the ꢀ′ substituent is not freely rotating) and be
in the range (g26-27 kcal/mol) that allows for convenient
handling at room temperature.
None of the enamides in this study has a barrier to N-alkenyl
bond rotation that is high enough to provide for separation of
rotamers at room temperature. However, the trends of the past
and present results with enamides show parallels to the more
well studied anilides.9,10 By analogy, it should be generally
Finally, it is also interesting to compare the rate constants
for rotation of radical precursors with the rate constants for
cyclization of the derived radicals to deduce which process is
faster, cyclization or rotation. This is already possible for one
class of radical 2, shown in Figure 9. In round numbers, the
rate constant for cyclization of radical 2 at 298 K is about 104
M-1 s-1.4 Remarkably, this is essentially the same as the rate
(13) Yamada, S. ReV. Heteroatom. Chem. 1999, 19, 203–236.
J. Org. Chem. Vol. 74, No. 11, 2009 4265

Guthrie et al.
constant for rotation of the radical precursors 1a-c.14 Assuming
that the radical 2 has a similar rotation barrier to the precursor
1,15 radical 2 partitions between direct cyclization and rotation
prior to cyclization. Rate constants for cyclization are not known
for the other radicals derived from the halides in this study.
However, the lowest possible rate constants for successful
radical cyclizations under these conditions are in the range of
103 s-1. So the radicals derived from 11 and probably also 12
cyclize faster than they rotate.
was stirred under reflux in a Dean-Stark apparatus for 6 h. The
solvent was removed in vacuo to give the crude imine that was
used in the next step without further purification. The crude imine
(9 mmol) was dissolved in dry toluene (50 mL) and cooled to
0 °C. 2-Bromoisobutyryl bromide (9 mmol) was added dropwise,
followed by the slow addition of N,N-diethylaniline (9 mmol). The
reaction was then stirred for 4 h at room temperature then added
to water (50 mL). The layers were separated, and the organic layer
was washed with 10% HCl (10 mL). After drying over MgSO4,
the solvent was removed in vacuo to give a crude residue that was
purified by column chromatography: yield (60%); clear oil; Rf (3:1
petroleum ether/EtOAc) 0.83; IR (film)/cm-1
υmax 2932, 2850, 1633,
Conclusions
1
1496, 1464, 1453, 1392, 1365, 1173, 1108, 729, 697; H NMR
(300 MHz, CDCl3) δ 7.21 (5H, m), 5.53 (1H, m), 4.62 (2H, s),
2.39 (2H, m), 2.22 (2H, m), 1.94 (6H, s), 1.85 (2H, q, J ) 7.5 Hz);
13C NMR (75.5 MHz, CDCl3) δ 170.9, 142.6, 137.9, 130.3, 128.7,
128.3, 127.5, 58.3, 52.4, 33.6, 32.3, 30.5, 22.4; ESI m/z 344 ([M]
+ Na) 322, 242, 91; found (MNa)+, 344.0620, C16H20BrNO requires
(MNa)+, 344.0626; found C, 57.1; H, 6.1; N, 4.0; C16H20BrNO
requires C, 56.8; H, 6.0; N, 4.1.
Barriers to rotation of the N-alkenyl bond in a series of
N-cycloalkenyl-N-benzyl R-haloacetamide derivatives have been
measured by variable-temperature NMR experiments. The
barriers are all significant and range from 10 to 18 kcal/mol,
depending on ring size and on substituents on the cycloalkene
and the amide. The trends are qualitatively similar to those
observed for anilide derivatives, so it is now possible to design
substituent combinations that will provide resolvable enanti-
omers or diastereomers at ambient temperature. The compounds
undergo 4-exo and 5-endo radical cyclizations at rates that may
be faster or slower than the estimated rate of N-alkenyl bond
rotation in the derived radicals, depending on the nature of the
substituents that are present.
Acknowledgment. The Pitt group thanks the National Science
Foundation for funding and the National Institutes of Health
for a grant to purchase an NMR spectrometer. The Warwick
group thanks the Engineering and Physical Science Research
Council for funding. The Oxford Diffraction Gemini XRD
system was obtained, through the Science City Advanced
Materials project: Creating and Characterizing Next Generation
Advanced Materials, with support from Advantage West
Midlands (AWM) and in part funded by the European Regional
Development Fund (ERDF).
Experimental Section
Representative Procedure for Synthesis of Enamides 4a-e,
7, 11, and 12. Benzyl-2-bromo-N-(cyclopent-1-enyl)-2-methyl-
propionamide (4a): Benzylamine (30 mmol) was added to the
cyclopentanone (30 mmol) in toluene (20 mL), and the mixture
Supporting Information Available: Procedures of synthesis
and characterization data of 4a-e, 7, 11, and 12, general
methods for VT NMR experiments, and copies of NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(14) Musa, O. M.; Horner, J. H.; Newcomb, M. J. Org. Chem. 1999, 64,
1022–1025.
(15) Support for this assumption comes from anilide rotation barriers; see
ref 9, Table 3. Compounds with alkyl substituents on the acyl group have similar
rotation barriers to compounds with alkenyl substituents. Here, the sp2 alkenyl
group is a model for a radical.
JO900491W
4266 J. Org. Chem. Vol. 74, No. 11, 2009