Ring-expanded bicyclic β-lactams: A structure-chiroptical properties relationship investigation by experiment and calculations

Article abstract of DOI:10.1021/jo200171w

In the present work, the validity of the helicity rule relating the absolute configuration of the bridgehead carbon atom in bicyclic β-lactams to the sign of the 220 nm band observed in their electronic circular dichroism (ECD) spectra is examined for rin

Full text of DOI:10.1021/jo200171w

ARTICLE
pubs.acs.org/joc
Ring-Expanded Bicyclic β-Lactams: A StructureꢀChiroptical Properties
Relationship Investigation by Experiment and Calculations
Magdalena Woꢀznica,† Aleksandra Butkiewicz,† Aneta Grzywacz,† Patrycja Kowalska,† Marek Masnyk,†
Karol Michalak,† Roman Luboradzki,‡ Filipp Furche,*,§ Holger Kruse,^ Stefan Grimme,*,^ and Jadwiga Frelek*,†
†Institute of Organic Chemistry of the Polish Academy of Sciences, Kasprzaka 44, 01-224 Warsaw, Poland
‡Institute of Physical Chemistry of the Polish Academy of Sciences, Kasprzaka 44, 01-224 Warsaw, Poland
§Department of Chemistry, University of California, Irvine, Natural Sciences II, Irvine, California 92697-2025, United States
^Organisch-Chemisches Institut der Universit€at M€unster, Corrensstrasse 40, 48149 M€unster, Germany
S Supporting Information
b
ABSTRACT: In the present work, the validity of the helicity
rule relating the absolute configuration of the bridgehead
carbon atom in bicyclic β-lactams to the sign of the 220 nm
band observed in their electronic circular dichroism (ECD)
spectra is examined for ring-expanded cephalosporin analogues.
To this end, a series of model compounds with a seven-
membered ring condensed with the β-lactam unit was synthe-
sized. A key step of their synthesis was either the ring-closing
metathesis (RCM) or the free radical cyclization leading to the seven-membered ring with an S, O, or C atom at the 6 position in the
bicyclic skeleton. To investigate the scope and limitations of the simple, empirically established helicity rule, a combination of ECD
spectroscopy, variable-temperature ECD measurements, X-ray analysis, and time-dependent density functional theory (TD-DFT)
calculations was used. A comparison of the experimental ECD spectra with the spectra simulated by TD-DFT calculations gives a
reasonable interpretation of the Cotton effects observed in the 240ꢀ215 nm spectral range. The results suggest that the helicity rule
does not apply to the investigated compounds because of the planarity of their amide chromophore. Thus, these compounds do not
constitute an exception to the rule that was established for bi- and polycyclic β-lactams with the nonplanar amide chromophore only.
structure shared by all compounds examined so far is the
nonplanarity of amide chromophoric system.5,6b,6c,7,8 This non-
planarity of the chromophore causes its intrinsic dissymmetry
and is expressed in a right- or left-handed helicity depending upon
the absolute configuration at the bridgehead carbon atom. Thus, on
the basis of the helicity rule we were able to correlate the R absolute
configuration at the ring junction with a positive sign of the Cotton
effect (CE) attributed to the amide n f π* transition and the S
absolute configuration with a negative sign of the same CE.
To validate the helicity rule established under the assumption
of conformational rigidity of the bicyclic system and the realiza-
tion that most of the molecular excitations are localized within
the amide chromophore, we conducted a detailed theoretical
study.7 As the method of choice we applied the time-dependent
density functional theory (TD-DFT) as it has become an
important tool in theoretical CD spectroscopy due to a reason-
able balance between accuracy and computational efficiency.10
The results of our study revealed a high sensitivity of the ECD
spectra to the configuration at the bridgehead carbon atom and a
surprisingly high sensitivity to the molecular conformation and
substitution of the five- or six-membered rings.7,8 The molecular
1. INTRODUCTION
The β-lactam family of antibiotics represents one of the most
clinically relevant families of antibiotics known.1 This is attribu-
table to their broad spectrum of antibacterial activity and a
relatively low level of toxicity.2 It has also been proved that only
β-lactam antibiotics with an R absolute configuration at the ring
junction carbon atom display strong antibacterial activity.3
Therefore, the relationship between an R absolute configuration
at the bridgehead carbon atom and the sign of the lowest energy
Cotton effect (CE) attributed to the amide n f π* transition,
which occurs in penicillins and cephalosporins at around 230 and
260 nm, respectively, was extensively investigated.3a,c,4 Frelek at
al. contributed to the studies on this subject with investigation of
chiroptical properties of 5-dethia-5-oxacephams.5 As a result, a
simple helicity rule was proposed that permits the assignment of
absolute configuration at the ring junction carbon atom based on
the sign of the ECD band at around 220 nm arising from the
amide n f π* excitation. It was also found that the same rule is
valid for clavams,6 cephams,7 carbacephams,7 natural penicillins,
and cephalosphorins and their saturated and unsaturated deri-
vatives as well.8 Very recently, we demonstrated the successful
applicability of the rule to carbapenams.9
On the basis of X-ray analysis and computational studies, it was
demonstrated that the most important feature of the molecular
Received: January 28, 2011
Published: March 25, 2011
r
2011 American Chemical Society
3306
dx.doi.org/10.1021/jo200171w J. Org. Chem. 2011, 76, 3306–3319
|

The Journal of Organic Chemistry
ARTICLE
dynamics (MD) simulations carried out for carbacephams re-
vealed a considerable flexibility of the saturated six-membered
ring and a high dependency of the ECD spectra on the con-
formation of both the four- and the six-membered rings. In some
cases, this dependency resulted in the sign inversions for both
n f π* and π f π* bands at various points of the MD
simulation. Thus, in some oxacephams a breakdown of the
helicity rule occurred, likely caused by a change of the conforma-
tion of the pyranose ring from a chair to a boat.7 Therefore, the
question arose about the scope and limitations of the helicity rule.
To answer this question several model compounds with a β-
lactam ring fused to the seven-membered ring were synthesized
(Chart 1). These compounds were selected as models because of
their greater conformational flexibility in comparison with the
investigated bicyclic β-lactams with the five- or six-membered
rings condensed with an azetidinone ring. Therefore, a different
set of CEs for different conformations present in the equilibrium
can be expected for these compounds. Moreover, a flattening
of the amide chromophore can occur in β-lactams 1ꢀ10, due to
the lower strain energy of the seven-membered ring. For this
reason, a noncompliance with the helicity rule is conceivable
because the rule was developed for the skewed bi- and polycyclic
β-lactams only.
bination of experimental and theoretical ECD results should
allow us to draw more definite conclusions within the scope of
present research. This combination has already proven efficient
and reliable for the assignment of the absolute configuration of
various chiral organic molecules.11
2. RESULTS AND DISCUSSION
2.1. Synthesis. Compounds 1a, 2a, 3a, and 4a were synthe-
sized according to Scheme 1 in which the key step is the ring-
closing metathesis (RCM) of 1,8-nonanadiene system leading to
a seven-membered ring.
As a starting material in this synthesis the commercially available
azetidinone 11 was used. In the first step the acetoxy group was
substituted in the reaction with allyl bromide, allyl alcohol, or
allyl thioalcohol, yielding compounds 12a, 12b, and 12c, respec-
tively. Compound 12a was subjected to alkylation reaction with
4-bromo-1-butene in the presence of sodium hydride, and
compounds 12b and 12c were reacted analogously with 1-bro-
mo-3-propene. Resulting compounds 13aꢀ13c were ready for
RCM, which was performed in boiling methylene chloride in the
presence of catalytic amounts of Grubbs’ first-generation catalyst,
providing compounds 2a, 3a, or 4a. The double bond in
compound 2a was then hydrogenated with H2 over palladium
on charcoal to yield 1a.
Compound 5a was prepared according to Scheme 2. In the
first step compound 12a was alkylated at the nitrogen atom with
2,3-dibromo-1-propene giving product 14, which was converted
into 5a in the free radical cyclization reaction with tributyltin
hydride in the presence of catalytic amounts of AIBN.
Compounds 6 and 7 were also obtained using the RCM of
1,8-nonanadiene system; however, readily obtainable compound
15 was used as the starting material (Scheme 3).12 As a first step,
To get more insight in the structureꢀECD spectra relation-
ship, we will apply a combined treatment of experimental ECD
spectroscopy and TD-DFT. Furthermore, we will address the
question about the effect of Boltzmann averaging the theoretical
spectra of several conformers and corresponding theoretical re-
quirements on computed conformational energies. Since ECD
spectroscopy is very sensitive to smallest changes in the geometry
and electronic structure of the investigated molecules, the com-
Chart 1
Scheme 2a
a Legend: (a) 2,3-dibromopropene, NaH, DMF, rt; (b) AIBN, Bu3SnH,
C6H6, 80 ꢀC.
Scheme 1a
a Legend: (a) allyl bromide, KI, In, DMF, rt; (b) allyl alcohol, Pd(OAc)2, Et3N, C6H6, rt; (c) allyl thioalcohol, BuLi, THF, ꢀ78 ꢀC, rt; (d) 4-bromo-1-butene,
NaH, rt; (e) allyl bromide, NaH, DMF, rt; (f) Grubbs’ I cat. (5 mol%), CH2Cl2, 40 ꢀC; (g) Grubbs’ I cat. (20 mol %), CH2Cl2, 40 ꢀC; (h) H2, Pd/C, EtOH, rt.
3307
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Scheme 3a
a Legend: (a) allyl bromide, Zn, NH4Cl(sat.), THF, 0 ꢀC, rt; (b) Ac2O, Py; (c) HoveydaꢀGrubbs’ II cat. (10 mol %).
band occurring between 212 and 238 nm is of a particular interest
because this band is the subject of the helicity rule. This band can
be attributed to the amide n(O) f π* transition, and its
corresponding electronic UV absorption is hidden by the neigh-
boring absorption bands (see Table 1). In the case of β-lactams
1aꢀ5a, the decisive ECD band occurs at about 215 nm, whereas
in the case of 6ꢀ10 its position is shifted to the lower energy
region by around 10ꢀ15 nm. This is likely related to the presence
of an interfering diene chromophore that absorbs in a similar energy
range as the amide chromophore.
To follow the helicity rule, the sign of the decisive ECD band
should be positive for β-lactams belonging to the 7R-configura-
tional series, that is, compounds 1ꢀ5, 8, and 10, whereas it
should be negative for β-lactams 6, 7, and 9, representing the 7S-
configurational series. In fact, in the ECD spectra of β-lactams
with the 7R-configuration a negative sign of the CE at around
215 nm is present and a positive one of the same CE for β-
lactams with 7S-configuration (Table 1, Figure 2). This result
contradicts the prediction made by the helicity rule and, in fact,
calls for the assessment of the validity of the rule for this class of
compounds. It is therefore necessary to examine whether com-
pounds 1ꢀ10 are the exception to the rule or, more generally, the
rule itself is imperfect.
The origin of the inconsistency between the experimental
results and predictions made applying the rule is undoubtedly
very important for further studies on similar systems. Conforma-
tional flexibility of the investigated systems and/or a planarity of
the chromophore might explain these unexpected results. The
planarity of the chromophore will prevent the use of the helicity
rule because the rule was developed for β-lactams with a non-
planar amide chromophore only.5,8 Conformational flexibility of
the seven-membered ring may also lead to a sign change as
different conformers can make different contributions to the total
ensemble-averaged spectrum.
To find reasons for possible breakdown of the helicity rule, the
TD-DFT calculations and variable low-temperature ECD mea-
surements were carried out for selected β-lactams. The results
should give more insight into the conformational and chiroptical
behavior of compounds 1ꢀ10.
Figure 1. 1H difference NOE experiments of compounds 6 and 7 and
diagnostic coupling constants for 9.
the aldehyde 15 was transformed into a chromatographically
nonseparable mixture of alcohols 16. The alcohols were acetylated
with acetic anhydride, and acetates 17 were subjected to RCM
reaction yielding a mixture of 6 and 7 in the ratio of 1: 1.
A sample of the mixture of 6 and 7 was separated into pure
diastereomers using HPLC technique. The absolute configura-
tion at C6 carbon atom was assigned using NOE experiment, as
shown in Figure 1.
Compound 8 was prepared using an enyne variant of metath-
esis according to Scheme 4. First, the acetoxy group in azetidi-
none 11 was substituted by the propargyl chain in the reaction
with propargyl bromide and indium in the presence of potassium
iodide.13 The alkylation product 18 was then alkylated at nitrogen
atom with 4-bromo-1-butene under the PTC condition with
pulverized potassium hydroxide, yielding enyne 19, which was
subjected to RCM using HoveydaꢀGrubbs’ second-generation
catalyst (HoveydaꢀGrubbs’ II). As a result the bicyclic product 8
was formed.
Compound 9 was prepared according to Scheme 5 starting
from aldehyde 15, which was reacted with propargyl bromide and
zinc to produce alcohol 20. Acetylation of alcohol 20 followed by
the metathesis reaction with HoveydaꢀGrubbs’ second-genera-
tion catalyst provided bicyclic product 9. Interestingly, com-
pound 9 was the only diastereomer obtained in this reaction. The
indicated absolute configuration at C6 carbon atom was deter-
mined by comparing the appropriate coupling constants in the
1H NMR spectra of compounds 6, 7, and 9 as shown in Figure 1.
Compound 10 was prepared according to Scheme 6 starting from
azetidinone 11. The acetoxy group was replaced by propargyl ether
leading to compound 22. Alkylation of 22 at the nitrogen atom
with allyl bromide and sodium hydride as a base gave product 23,
which was converted into the bicyclic system 10 in the metathesis
reaction with Grubbs’ first-generation catalyst (Grubbs’ I).
The thiono analogous 1b, 2b, 3b, 4b, and 5b were obtained by
direct thionation of 1a, 2a, 3a, 4a, and 5a in the reaction with
Lawesson’s reagent.14
2.2. Experimental ECD Results. The ECD and UV data are
collected in Table 1. On the basis of the presented data com-
pounds 1ꢀ10 can be classified by R or S absolute configuration at
the ring junction and the presence of an additional chromophore
interfering with the amide absorption bands. As evident from
Table 1, up to four absorption bands within the spectral range of
185ꢀ300 nm are present in the ECD spectra of investigated
compounds 1ꢀ10. For the purpose of our studies, however, the
2.3. Theoretical ECD Results and Variable Low-Tempera-
ture ECD Measurements. To confirm and assign the experimental
and computational ECD spectra, first the thermally accessible
conformational species of the respective molecule must be
obtained. For this prerequisite the MMX force field from the
PCModel 9.115 software was used. The structures thus obtained
were subsequently reoptimized at the dispersion corrected16
BLYP-D17/def2-TZVP18 level of theory. For all conformers found
within a 3 kcal/mol energy window, more accurate B2PLYP-D19/
def2-QZVP18 single point calculations were performed to obtain
a reliable Boltzmann distribution. The COSMO solvation model20
was used to simulate the effects of the solvent (acetonitrile; dielectric
constant of ε = 36.64) and was included in all TD-DFT calculations.
All TD-DFT calculations were carried out with the TURBO-
MOLE suite of programs.21 For BLYP-D the resolution of
3308
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Scheme 4a
a Legend: (a) propargyl bromide, KI, In, DMF, rt; (b) 4-bromo-1-butene, PTC conditions; (c) HoveydaꢀGrubbs’ II cat. (10 mol%), CH2Cl2, 40 ꢀC.
Scheme 5a
a Legend: (a) propargyl bromide, Zn, NH4Cl (sat), THF, 0 ꢀC; (b) Ac2O, Py, rt; HoveydaꢀGrubbs’ II cat. (10 mol %), CH2Cl2, 40 ꢀC.
Scheme 6a
a Legend: (a) propargyl alcohol, Zn(OAc)2, toluene, 80 ꢀC; (b) allyl bromide, NaH, DMF, rt; (c) Grubbs’ I cat. (10 mol %), CH2Cl2, 40 ꢀC.
Table 1. UV and ECD Data of β-Lactams 1ꢀ10 Recorded in Acetonitrile in the Range of 186ꢀ300 nm
compd
UV ε (λmax); [Mꢀ1 cmꢀ1 (nm)]
CD Δε (λmax); [Mꢀ1 cmꢀ1 (nm)]
1a
2a
3a
4a
5a
6
þ2.4 (199.0)
þ4.9 (193.0)
þ9.4 (187.5)
þ33.0 (189.0)
ꢀ3.6 (216.0)
ꢀ9.3 (211.5)
ꢀ12.1 (214.0)
ꢀ11.2 (220.5)
ꢀ7.0 (219.5)
þ2.5 (228.5)
þ3.5 (229.5)
ꢀ4.8 (238.0)
þ14.6 (231.0)
ꢀ10.8 (223.0)
þ0.7 (238.0)
þ0.2 (234.0)
920 (196)
500 (237sh)
520 (244sh)
ꢀ7.3 (205.5)
þ9.2 (201.5)
þ10.1 (202.0)
þ25.1 (202.0)
þ2.0 (212.5)
þ2.4 (216.5sh)
þ1.3 (248.5)
32000 (192)
44500 (192)
13000 (196)
43800 (192)
6500 (218sh)
9900 (218sh)
800 (269)
1200 (268)
13700 (234)
3100 (269)
1100 (276)
ꢀ2.4 (187.0)
ꢀ9.9 (187.5)
ꢀ10.3 (197.0)
ꢀ8.3 (202.0)b
þ17.5 (192.0)
ꢀ0.2 (266.0)a
ꢀ0.2 (265.0)a
7
8
9
19250 (220)
12700 (227)
ꢀ0.5 (268.0)a
þ0.4 (261.5)
10
a An additional small positive ECD band is observed at around 280 nm. b An additional negative ECD band is observed at around 188 nm.
identity22 (RI, density fitting) of the Coulomb integrals has been
applied. Likewise, B2PLYP-D makes use of the RI-MP2 code
within the ricc2 module.23 The necessary auxiliary basis sets were
taken from the TURBOMOLE library.20 The convergence criterion
for the optimizations regarding the change of total energy between
two subsequent optimization cycles was set to 10ꢀ7 Eh and the
numerical quadrature grid of m4 quality was used throughout.
The vertical excitation energies and rotational strengths in dipole
lengths form were calculated with the escf module24 according to
the TD-PBE0 formalisms with the def2-TZVPP18 basis set.
Because a fully consistent COSMO/TD-DFT treatment of elec-
tronically excited states is technically not possible at the mo-
ment, we applied the COSMO model to the calculation of the
ground state and solely used the COSMO modified orbitals and
eigenvalues in a subsequent standard gas-phase TD-DFT treat-
ment. All theoretical CD spectra were simulated by overlapping
Gaussian functions for each transition. A value of σ = 0.2 eV for
the width of the absorption band at a height of 1/e was chosen.
For compound 5a a modified approach was chosen, which was
extensively validated in ref 5, where the reader is referred for
further details. The structure was optimized using a split-valence
basis set with a set of polarization functions on non-hydrogen
atoms [SV(P)].18 SV(P) energetics were validated with polarized
triple-ζ valence basis sets (def2-TZVP).25 Fine quadrature grids
(size m4) were employed.25b,26 All stationary points were con-
firmed to be minima by the force-constant calculations. Excita-
tion energies and rotatory strengths of the lowest 20 singlet states
were computed using TD-DFT26 and SV(P) basis sets. CD spectra
were simulated by superposition of Gaussians with a uniform line
width of 0.16 eV. Solvent effects were taken into account by the
COSMO model21 using the dielectric constant of acetonitrile
(37.5). The slightly different dielectric constant used here is neglect-
able in the COSMO model. The fast response of the screening
potential was neglected in the TD-DFT response calculations.
The one-parameter hybrid density functional of Perdew, Burke,
and Ernzerhof (PBE0)26 was used throughout.
3309
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Figure 2. Experimental ECD spectra of β-lactams. Left: 3a (red line), 6 (blue line), and 5a (green line). Right: 8 (green line), 9 (blue line), and 10 (red
line) recorded in acetonitrile. The symbol ε denotes the molar decadic absorption coefficient and λ the wavelength.
Table 2. Overview of the Conformer Analysis for Compound 1aa
conformer
ΔΕ
population
10-9-1-2
10-9-1-7
8-9-1-2
8-9-1-7
7-8-11-12
pyramid.
1
2
3
4
5
6
7
0.00
0.43
0.75
1.23
1.73
2.25
2.38
49.9
24.1
14.0
6.2
ꢀ2.5
þ0.1
þ179.8
þ179.5
ꢀ177.2
ꢀ177.3
ꢀ178.9
ꢀ178.8
ꢀ179.3
þ178.1
ꢀ179.6
þ158.0
þ158.4
þ177.0
þ177.2
þ176.6
þ0.4
ꢀ0.2
þ1.9
þ1.9
þ1.2
þ1.4
þ0.6
ꢀ69.1
þ48.0
ꢀ69.6
þ49.3
þ48.1
þ179.2
ꢀ179.2
ꢀ1.7
ꢀ0.5
ꢀ21.1
ꢀ20.1
ꢀ3.2
þ18.2
þ17.3
þ3.2
2.7
1.1
ꢀ3.0
þ3.3
þ3.2
þ2.9
0.8
a Selected torsion angles in degrees and pyramidal heights (pyramid.) for the N-atom in pm determined on the basis of COSMO-BLYP-D/def-TZVP
structures. Pyramidal height defined as a deviation from planarity according to Radhakrishnan et al.27 Relative energies ΔE (kcal/mol), computed at
COSMO-B2LYP-D/def2-QZVP level of theory. Conformer population at 25ꢀC in %.
For β-lactam 1a seven conformers in the energy range of 2.4
kcal/mol were found. To simplify the calculation, the large sub-
stituent at C8 (OSitBuMe2) has been replaced by smaller OSiMe3
group. Both of these groups have very similar electronic properties,
and interchanging them does not influence the electronic spectra
significantly. Furthermore, the effect on the conformational space is
also expected to be small. The calculations were done in a similar
fashion for the other compounds. The only exception is β-lactam 5a,
for which the tert-butyl group was taken into account.
As can be seen in Table 2, two out of seven conformers obtained
for 1a, namely, 3 and 4, possess slightly skewed amide chromo-
phores, whereas in the remaining five the chromophoric system is
planar. The seven-membered ring in conformers of 1a is in a
chair or a twist-chair conformation, and the azetin-2-one ring is in
an equatorial position at C7 (Figure 3). Beyond that, the individual
conformers show conformational differences mostly in the substi-
tuent on C8 carbon atom.
For the negative sign of the O10ꢀC9ꢀN1ꢀC2 and O10ꢀ
C9ꢀN1ꢀC7 torsion angles, the helicity rule predicts a positive
sign of the crucial ECD band. In accordance with this, a positive
sign of the lowest energy excitations is calculated for conformers
3 and 4 (Figure 3f and g). The sign of the same band in con-
formers with a planar amide chromophore, i.e., conformers 1, 2,
5, 6, and 7 (see Table 2), is negative and therefore not subject to
the helicity rule. The population ratio of conformers with a planar
and nonplanar chromophore in β-lactam 1a is found to be ap-
proximately 4:1. Therefore, in the average ECD spectrum com-
prising, by definition, the sum of contributions from all populated
conformers, the negative sign of the decisive band should pre-
dominate, which is indeed the case (Figure 3h). However, a small
positive ECD band at around 240 nm, originating from the
twisted conformers 3 and 4, is present in both experimental and
simulated spectra. The band at around 220 nm has primarily the
character of an amide n f π* transition. The average ECD
spectrum shows very close agreement between experiment and
theory, thus providing strong evidence that these conformers are
present in solution under these conditions (Figure 3h).
Only two chair conformers have been calculated for β-lactam
5a in the energy range of 3 kcal/mol. In agreement with the ex-
periment, the simulated ECD spectrum of the lowest energy chair
conformer (chair 1) with a planar amide chromophore exhibits a
negative sign of the 220 nm CE and therefore does not conform
to the helicity rule. The simulated ECD spectrum of the second
chair conformer of 5a (chair 2) displays a positive sign of the
same CE in agreement with the helicity rule. The amide chromo-
phore in this conformer is twisted as demonstrated by the torsion
angle O10ꢀC9ꢀN1-C2 equal to ꢀ12.24ꢀ. Since this conformer
is computed to be 2.4 kcal/mol (PBE0/def2-TZVP) higher in
energy than the lowest energy chair 1 conformer, its population is
negligible at room temperature. Therefore, there is apparently no
difference between the simulated ECD spectrum of chair 1 con-
former and the Boltzmann-averaged spectrum (Figure 4). To
fully account for different orientations of the flexible TBDMS
side chain and quantitatively predict the ECD spectrum, espe-
cially at higher energies, molecular dynamics simulations would
be necessary.
3310
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Figure 3. (aꢀg) COSMO-BLYP-D/def-TZVP structures of conformers of β-lactam 1a and their computed ECD spectra at the TD-PBE0/def2-
TZVPP level of theory. (h) Boltzmann-averaged spectrum compared to experiment. The symbol ε denotes the molar decadic absorption coefficient, λ is
the excitation wavelength, and R is the computed rotatory strength.
Figure 4. COSMO-PBE0/SV(P) structures of conformers of β-lactam 5a and their computed ECD spectra at the TD-PBE0/def2-TZVP level of theory
together with the Boltzmann-averaged spectrum compared to experiment. The symbol ε denotes the molar decadic absorption coefficient, λ is the
excitation wavelength, and R is the computed rotatory strength.
The agreement between the experimental and Boltzmann-
averaged ECD spectra is very satisfactory also for β-lactams
2aꢀ4a (Figure 5). In these cases, depending on the compound,
the main ECD band is a mixture of excitations out of the amide
n f π* transition and transitions out of the sulfur or oxygen lone
pairs into the same acceptor orbitals. In addition, the transitions
3311
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Figure 5. Boltzmann-averaged TD-PBE0/def2-TZVPP spectra compared to experiment of β-lactams 2aꢀ4a. The symbol ε denotes the molar decadic
absorption coefficient, and λ is the excitation wavelength.
Table 3. Overview of the Conformer Analysis for Compound 8a
conformer
ΔΕ
population
10-9-1-2
10-9-1-7
8-9-1-2
8-9-1-7
7-8-11-12
pyramid.
1
2
3
4
5
0.00
0.76
1.66
2.06
3.00
72.24
20.15
4.41
ꢀ7.9
ꢀ6.1
ꢀ7.1
ꢀ4.1
ꢀ16.3
ꢀ178.5
ꢀ178.7
ꢀ179.3
ꢀ179.2
ꢀ177.5
þ172.0
þ172.4
þ172.4
þ174.8
þ162.8
þ1.4
ꢀ0.2
þ0.2
ꢀ0.3
þ1.6
ꢀ68.6
þ48.3
ꢀ88.9
þ47.0
ꢀ69.5
þ6.9
þ5.4
þ5.6
þ3.5
þ14.4
2.24
0.46
a Selected torsional angles in degrees and pyramidal height (pyramid.) for the N-atom in pm determined on the basis of COSMO-BLYP-D/def-TZVP
structures. Relative energies ΔE (kcal/mol) computed at COSMO-B2PLYP-D/def2-QZVP level of theory. Conformer population at 25ꢀC in %.
out of the double bond orbitals mix strongly with the amide n f π*
transition. Some of the structures of conformers of β-lactams
2aꢀ4a, calculated in the range of 3 kcal/mol, demonstrate a small
deviation of the amide chromophore from planarity manifested
by a slight pyramidality of the amide nitrogen. Nevertheless, the
negative sign of the decisive ECD band in both experimental and
Boltzmann-averaged ECD spectra clearly demonstrates the break-
down of the helicity rule for these compounds.
A more complex situation occurs in compounds 6ꢀ10 due to
the presence of groups strongly interfering with the amide chro-
mophore. The phenoxy and acetoxy substituents at C8 and C6,
respectively, present in β-lactams 6, 7, and 9 absorb at similar
wavelengths as the amide chromophore, which considerably
complicates the ECD spectra. In the case of β-lactams 8ꢀ10 the
additional presence of a highly absorbing and strongly interfering
diene system introduces a further complication.
As shown in Table 3, five conformers have been found for β-
lactam 8 in the energy range of 3 kcal/mol. Two of them, namely,
conformers 1 and 2, constitute more than 90% of the conforma-
tional mixture. Thus, these conformers predominate at room
temperature. The simulated ECD spectra of both conformers,
despite the negative sign of the O10ꢀC9ꢀN1ꢀC2 torsion angle,
show a negative sign of the CE at 238 nm (Figure 6). It is worth
adding that in the case of 8, the CE at 238 nm is due mainly to the
electronic excitation within the diene system, absorption of
which predominates the net ECD spectrum.
Six contributing conformers were found for compound 9 within a
1.6 kcal/mol energy range (see Supporting Information). The two
lowest energy conformers, populated with 58% and 24%, respec-
tively, dominate the net ECD spectrum. The O10ꢀC9ꢀN1ꢀC2
torsion angle in 9 is positive, just as is the sign of the CE at 238 nm.
Nevertheless, the positive/negative sign of the ECD band at
around 230 nm, decisive in terms of the helicity rule, in the case of
compounds studied is in contradiction to the rule because for the
S/R configuration at the bridge-head carbon atom the rule pre-
dicts a negative or a positive sign of this band, respectively. To
provide further evidence and to find experimental confirmation
for the computational results, the variable low-temperature ECD
spectra of selected β-lactams were recorded. However, to achieve
a better resolution of the ECD bands, the β-lactams in question
were converted into their respective thiono equivalents 1bꢀ5b,
by thionation reaction with Lawesson’s reagent.14a,28 It is well-
known that in the thiocarbonyl compounds, in general, the UV
and ECD bands are shifted appreciably to the red.28b,29 The sign
of the n f π* band for amides and thioamides is the same
providing that their chiral environment remains identical. The
UV spectra of compounds 1bꢀ5b show a weak low-energy band
around 330 nm corresponding to the n f π* transition, the
intense band at around 260 nm arising from the π f π* ex-
citation, and a moderately intense band at ca. 210 nm (Table 4).4i
The polarization direction of this short-wavelength transition,
most probably of an nσ f π* origin28b,29 is nearly orthogonal to
the polarization direction of the π f π* band.
In variable low-temperature ECD spectra the intensity of the
diagnostic band is greatly dependent on the temperature of the
solution (Figure 7). In most cases, the 340 nm band, associated
with n f π* thioamide transition, shows increase in intensity
with lowering temperature (Figure 7). This strongly indicates
that most of the molecules are frozen in the preferred conforma-
tion. Much smaller temperature dependence is observed in the
case of compound 4b. The results, however, are in line with the
calculations, because the population of the lowest energy con-
former with computed CE in the corresponding spectral regions
should increase with decreasing temperature.
Additional evidence comes from the X-ray diffraction results of
compound 1b, the only one forming suitable crystals for such
3312
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Figure 6. COSMO-BLYP-D/def-TZVP structures of conformers of β-lactam 8 and their computed ECD spectra at the TD-PBE0/def2-TZVPP level of
theory, together with the Boltzmann-averaged spectrum compared to experiment. The symbol ε denotes the molar decadic absorption coefficient, and λ
is the excitation wavelength.
Table 4. UV and ECD Data of β-Lactams 1bꢀ5b Recorded in Acetonitrile at Room Temperature
compd
UV ε (λmax); [Mꢀ1cmꢀ1(nm)]
CD Δε (λmax); [Mꢀ1cmꢀ1(nm)]
1b
2b
3b
4b
5b
5500 (212)
15000 (265)
17900 (266)
16100 (262)
14000 (270)
14100 (265)
60 (326)
90 (325)
70 (333)
75 (329)
70 (325)
ꢀ2.3 (208.0)a
þ17.4 (210.0)
þ20.5 (200.5)
þ49.7 (198.0)b
ꢀ4.5 (193.5)c
þ8.87 (265.0)
ꢀ1.1 (280.5)
ꢀ1.1 (277.5)
ꢀ4.6 (266.0)
ꢀ9.2 (272.5)
þ6.8 (264.0)
ꢀ1.8 (328.0)
ꢀ2.6 (330.5)
ꢀ5.7 (335.0)
ꢀ2.2 (336.0)
ꢀ3.7 (331.0)
7100 (210)
6700 (206sh)
6900 (207sh)
7000 (211)
ꢀ6.1 (252.0)
þ9.2 (201.5)
a An additional small positive ECD band is observed at around 220 nm. b An additional small negative ECD band is observed at around 220 nm. c An
additional small positive ECD band is observed at around 212 nm.
Figure 7. Variable low-temperature ECD spectra of β-lactams 1b (a), 3b (b), 4b (c), and 5b (d) recorded in EPA (ether/isopentane/ethanol, 5:5:2 by
vol).
analysis (Figure 8). The structural data clearly demonstrate the
planarity of the amide chromophore and the chair form of seven-
membered ring. Additionally, the dihedral angle C9ꢀC7ꢀC2ꢀN
equals to ꢀ2.0ꢀ, which quantifies the sp2 hybridization of the
amide nitrogen atom.
The excellent agreement between the solid-state structure and
the computed one (Table 5), as well as the close similarity of the
ECD spectra of 1b recorded in both solid-state and solution
(Figure 8), indicates the presence of the same molecular species
in both states. This means that the soluteꢀsolvent interactions,
which may considerably affect the CD spectra owing to both
conformational and vicinal effects, are negligible in these cases
and indicate that the observed ECD is largely a molecular
property. Therefore, the analysis of the ECD data can be
performed on the basis of chiroptical data obtained for solutions.
In addition, the literature data for tricyclic β-lactam analogues,
structurally very similar to our model compound 5a, confirm the
planarity of amide chromophore in such cyclic systems.30 Tor-
sion angles S10ꢀC9ꢀNꢀC2 and S10ꢀC9ꢀNꢀC7, taken from
ref 31, equal to 4.3ꢀ and 176.4ꢀ, respectively, demonstrate the
3313
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
Figure 8. (Left) ORTEP diagram of compound 1b. Thermal ellipsoids are shown at 50% probability level. Hydrogen atoms are omitted for clarity.
(Right) ECD spectra of β-lactam 1b recorded in MI13 (methylcyclohexane/isopentane, 1:3, by vol) (green line) and nujol mull (red line) given in μdeg.
Table 5. Torsion Angles in Degrees and Pyramidal Height in pm of Compound 1b in Comparison with the Lowest Energy
COSMO-BLYP-D/def2-TZVP Conformer Structure
torsion anglea
1
2
3
4
5
6
7
8
9
10
pyramidal height
X-ray
ꢀ0.7
ꢀ1.9
þ176.4
þ179.9
þ178.8
ꢀ179.1
þ61.1
þ62.9
ꢀ84.8
ꢀ85.3
þ67.8
ꢀ62.9
ꢀ64.5
þ72.1
þ73.0
ꢀ61.8
ꢀ59.3
þ2.6
ꢀ1.5
ꢀ2.1
BLYP-D
68.2
þ0.01
a Torsion angles: 1 = S10ꢀC9ꢀNꢀC2; 2 = S10ꢀC9ꢀNꢀC7; 3 = C9ꢀNꢀC2ꢀC3; 4 = NꢀC2ꢀC3ꢀC4; 5 = C2ꢀC3ꢀC4ꢀC5;
6 = C3ꢀC4ꢀC5ꢀC6; 7 = C4ꢀC5ꢀC6ꢀC7; 8 = C5ꢀC6ꢀC7ꢀN; 9 = C6ꢀC7ꢀNꢀC2; 10 = C7ꢀNꢀC2ꢀC3.
planarity of the chromophoric system in these tricyclic β-lactams
and are in full agreement with our results for 1b. Thus, we can
conclude that the nonclassical β-lactams with a seven-membered
ring fused together with an azetidinone ring do not obey the
helicity rule.
substitution of the seven-membered ring and the side-chain flex-
ibility. The presence of an additional, interfering chromophore in
the molecule complicates interpretation of results. Nevertheless,
it is demonstrated that even in such cases the experimental ECD
spectroscopy supported by computational results can be success-
fully utilized in context of the configurational and conformational
analysis of two- and multifunctional bi- and polycyclic β-lactams
with one or more stereogenic centers. We strongly recommend
such a combined analysis of chiroptical properties because it can
provide detailed insight in the three-dimensional structure and
conformational dynamics of chiral molecules.
3. CONCLUSIONS
The relationship between the molecular structure and the chir-
optical properties of nonclassical β-lactams 1ꢀ10 was investigated
by electronic circular dichroism spectroscopy, variable low-temp-
erature CD measurements, time-dependent density functional theo-
ry, and X-ray diffraction analysis. The computed O10ꢀC9ꢀN1ꢀC2
and O10ꢀC9ꢀN1ꢀC7 torsion angles provide corroborating
evidence for the planarity of amide chromophore. Thus, the β-
lactam analogues examined in this work do not conform to the
previously established helicity rule. However, this fact does not
constitute an exception to the rule because this rule was formulated
strictly for β-lactam analogues with a nonplanar amide chromo-
phore only, whereas in the studied systems the chromophore is
planar.
TD-PBE0 calculations of ECD spectra are in good agreement
with the corresponding experimental data and allow an assignment
of the absolute configuration at the ring junction in β-lactams
1ꢀ10. In calculations it was necessary to consider five to seven
conformers of each compound. The obtained results have shown
that the sign of ECD band at 220 nm, characteristic for β-lactams,
is closely related to the stereostructure of the chromophoric system.
This band has mainly the character of an amide n(O) f π* tran-
sition.
4. EXPERIMENTAL SECTION
4.1. General Information. 1H and 13C NMR spectra were recorded
on 200 or 500 MHz spectrometers in CDCl3 at ambient temperatures.
Chemical shifts (δ) are reported in parts per million, using residual solvents
as internal standard. Optical rotations were measured in CH2Cl2 solutions
at ambient temperature and are quoted in units of 10ꢀ1 deg cm2 gꢀ1
.
Infrared spectra were obtained using neat samples on FTIR (ATR)
spectrometer. Electrospray ionization (ESI) mass spectrometry (MS)
experiments were performed on a spectrometer under normal condi-
tions. PFK solution was used as calibrant for HRMS measurements. UV
spectra were measured using acetonitrile as a solvent. CD spectra were
recorded between 180 and 400 nm at room temperature using acetoni-
trile solutions. Solutions with concentrations in the range 0.8 ꢁ 10ꢀ4 to
1.2 ꢁ 10ꢀ3 mol dmꢀ3 were examined in cells with path length 0.1 or 1 cm.
Low-temperature ECD measurements were recorded with a spectro-
polarimeter equipped with a cryostat using MI-13 (methylcyclopentane/
'isooctane 1:3 by vol) or EPA (ether/isopentane/ethanol 5:5:2 by vol)
as solvents.
The study demonstrated a high sensitivity of the ECD spectra
for even the smallest conformational differences caused by the
3314
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
HPLC separation of mixture of lactams 6 and 7 was performed on
Kromasil 100 Si (5 μ) column, with hexane/isopropanol (9:1) as a
mobile phase.
hexane, colorless oil, yield 79%. [R]D 29.3 (c 0.35, CH2Cl2). IR (film) ν/
1
cmꢀ1: 1495, 1442, 1271, 1257, 1139, 836, 777. H NMR (500 MHz,
CDCl3) δ/ppm: 0.08 (s, 3H), 0.09 (s, 3H), 0.86 (s, 9H), 1.29 (d, J = 6.4
Hz, 3H), 2.29ꢀ2.47 (m, 2H), 2.71 (d, J = 4.6 Hz, 1H), 3.67 (bd, J = 18.9
Hz, 1H), 3.99 (dd, J = 9.6, 5.7 Hz, 1H), 4.23 (qd, J = 6.2, 4.9 Hz, 1H),
4.32ꢀ4.41 (m, 1H), 5.74ꢀ5.80 (m, 1H), 5.83ꢀ5.90 (m, 1H). 13C NMR
(125 MHz, CDCl3) δ/ppm: ꢀ4.7, ꢀ4.0, 17.9, 22.4, 25.8(3C), 28.0,
41.0þ, 54.4, 66.1, 67.8, 122.0, 123.8, 200.7. HR MS (ESI) calcd for [M þ
Na] C16H29NONaSiS 334.16314, found 334.16349.
(7R,8R)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-6-oxa-1-aza-
bicyclo[5.2.0]non-3-en-9-thione (3b). Chromatographed using 5%
AcOEt in hexane, colorless oil, yield 87%. [R]D ꢀ92.24 (c 0.56, CH2Cl2).
IR (film) ν/cmꢀ1: 1490. 1H NMR (500 MHz, CDCl3) δ/ppm: 0.10 (s,
3H), 0.11 (s, 3H), 0.88 (s, 9H), 1.31 (d, J = 6.4 Hz, 3H), 3.02 (bs, 1H),
3.84ꢀ3.91 (m, 1H), 4.25ꢀ4.33 (m, 2H), 4.43ꢀ4.50 (m, 1H), 4.71ꢀ4.79
(m, 1H), 5.60 (s, 1H), 5.63ꢀ5.73 (m, 2H). 13C NMR (125 MHz, CDCl3)
δ/ppm: ꢀ4.7, ꢀ4.0, 17.9, 22.2, 25.8(3C), 44.8, 64.8, 65.4, 66.6, 90.4, 124.2,
128.6, 200.5. HR MS (LSI) calcd for [M þ Na]þ C15H27NO2NaSiS:
336.1424, found 336.1410. Elemental Analysis calcd for C15H27NO2SiS: C,
57.46; H, 8.68; N, 4.47. Found: C, 57.60; H, 8.89; N, 4.27.
All reactions were monitored by thin-layer chromatography using gel
plates 60 F254; visualization was accomplished with UV light and/or
staining with appropriate stains (KMnO4, anisaldehyde, vaniline, ninhy-
drin, or phosphomolybdic acid). Standard flash chromatography proce-
dures were followed (particle size 40ꢀ63 μm).
4.2. Materials. Commercially available starting materials were used
without further purification. Solvents were purified according to the literature
methods before use.31 All of the spectral data for known compounds
either match those reported or by comparison to the literature report.
4.3. Synthesis of Investigated Compounds.
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-1-aza-bicyclo-
[5.2.0]nonan-9-one (1a). A mixture of 2a (50 mg, 0.17 mmol), 10% Pd
on charcoal (5 mg), and ethanol (10 mL) was placed in a 50 mL round-
bottom flask equipped with a magnetic stirring bar, and the flask was
attached to an atmospheric pressure hydrogenation apparatus. Then the
flask was filled with hydrogen, and the components were vigorously stirred
until uptake of hydrogen had been ceased (ca. 2.1 mL). Then the mixture
was filtered through Celite, and the solvent was evaporated to yield analytically
pure 1a (47 mg, 95%) as a colorless oil. [R]D ꢀ11.2 (c 0.54, CH2Cl2). IR
(film) ν/cmꢀ1: 1750, 1472, 1398, 1256. 1H NMR (500 MHz, CDCl3)
δ/ppm: 0.058 (s, 3H), 0.061 (s, 3H), 0.86 (s, 9H), 1.20 (d, J = 6.3 Hz,
3H), 1.29ꢀ1.39 (m, 2H), 1.40ꢀ1.55 (m, 2H), 1.82ꢀ1.92 (m, 3H), 2.60
(d, J = 5.1 Hz, 1H), 3.25 (dt, J = 13.0, 4.3 Hz, 1H), 3.33 (tdd, J = 13.0, 3.5,
1.2 Hz, 1H), 3.66 (dt, J = 9.9, 2.2 Hz, 1H), 4.14 (dq, J = 6.3, 5.2 Hz, 1H).
13C NMR (125 MHz, CDCl3) δ/ppm: ꢀ5.0, ꢀ4.2, 17.9, 22.6, 25.7
(3C), 26.7, 28.7, 29.4, 35.3, 42.6, 55.65, 62.2, 65.7, 167.4. HR MS (ESI)
calcd for [M þ Na]þ C16H31NO2NaSi 320.2016, found 320.2023.
General Procedure for Thionation of β-Lactams with Law-
esson’s Reagent. A solution of β-lactam (1a, 2a, 3a, 4a, or 5a) (0.1
mmol) and Lawesson’s reagent (0.05 mmol) in benzene (10 mL) was
refluxed for 2 h. Then the solvent was removed and the crude product
was purified by chromatography on silica gel to produce thionation
product (1b, 2b, 3b, 4b, or 5b).
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-6-thia-1-aza-
bicyclo[5.2.0]non-3-en-9-thione (4b). Chromatographed using 5%
AcOEt in hexane, colorless oil, yield 90%. [R]D ꢀ20.6 (c 0.50, CDCl3).
IR (film) ν/cmꢀ1: 1481. 1H NMR (500 MHz, CDCl3) δ/ppm: 0.08 (s,
3H), 0.09 (s, 3H), 0.85 (s, 9H), 1.27 (d, J = 6.3 Hz, 3H), 2.99 (dd, J =
16.7, 8.3 Hz, 1H), 3.07 (dd, J = 3.1, 1.4 Hz, 1H), 3.39ꢀ3.45 (m, 1H),
3.85 (dq, J = 17.8, 2.7 Hz, 1H), 4.32 (qd, J = 6.3, 3.1 Hz, 1H), 4.80 (ddt, J
= 17.8, 5.1, 1.6 Hz, 1H), 5.38 (s, 1H), 5.56 (ddt, J = 11.1, 5.1, 2.3 Hz,
1H), 5.90ꢀ5.96 (m, 1H). 13C NMR (125 MHz, CDCl3) δ/ppm: 17.9,
22.0, 24.3, 25.8(3C), 44.1, 64.9, 65.3, 67.3, 126.2, 129.7, 200.8. HR MS
(ESI) calcd for [M þ Na]þ C15H27NONaSiS2 352.11956, found 352.12118.
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-3-methylene-1-
aza-bicyclo[5.2.0]nonan-9-thione (5b). Chromatographed using 5%
AcOEt in hexane, colorless oil, yield 80%. [R]D 25.4 (c 0.50, CH2Cl2).
IR (film) ν/cmꢀ1: 1485, 1442, 1258, 1248, 1140, 1093, 836, 776.1H
NMR (500 MHz, CDCl3) δ/ppm: 0.07 (s, 3H), 0.08 (s, 3H), 0.85 (s,
9H), 1.25 (d, J = 6.3 Hz, 3H), 1.33ꢀ1.44 (m, 1H), 1.49ꢀ1.61 (m, 2H),
1.90 (td, J = 12.9, 2.7 Hz, 1H), 1.99ꢀ2.07 (m, 2H), 2.59ꢀ2.66 (m, 2H),
3.86 (d, J = 15.1 Hz, 1H), 4.11 (bd, J = 11.6 Hz, 1H), 4.18 (qd, J = 6.3, 4.3
Hz, 1H), 4.29 (bd, J = 15.1 Hz, 1H), 5.02 (t, J = 1.6 Hz, 1H), 5.05 (bs,
1H). 13C NMR (125 MHz, CDCl3) δ/ppm: 17.8, 22.3, 25.8(3C), 29.9,
34.6, 36.7, 52.8, 63.9, 64.3, 66.2, 118.0, 143.1, 200.2. HR MS (ESI) calcd
for [M þ Na]þ C17H31NONaSiS 348.17879, found 348.1803.
General Procedure for Metathesis Reaction. A mixture of lactam
(13a, 13b, 13c, 19, 21, or 23) (0.62 mmol), anhydrous dichloromethane
(25 mL), and Grubbs’ first or second ruthenium catalyst (0.03 mmol)
was refluxed under an argon atmosphere until the substrate had
disappeared (according to TLC). Then the solvent was removed, and
the residue was purified by chromatography on silica gel to produce
cyclic product (2a, 3a, 4a, 8, 9, or 10).
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-1-aza-bicyclo-
[5.2.0]nonane-9-thione (1b). Chromatographed using 5% AcOEt in
hexane, colorless crystals, mp 29ꢀ32 ꢀC, yield 80%. [R]D ꢀ10.3 (c 0.51,
1
CH2Cl2). IR (film) ν/cmꢀ1: 1492, 1461, 1449, 1373, 1291. H NMR
(500 MHz, CDCl3) δ/ppm: 0.081 (s, 3H), 0.084 (s, 3H), 0.86 (s, 9H),
1.26 (d, J = 6.3 Hz, 3H), 1.30ꢀ1.39 (m, 2H), 1.52ꢀ1.62 (m, 2H),
1.88ꢀ1.94 (m, 3H), 1.96ꢀ2.05 (m, 3H), 2.56ꢀ2.59 (m, 1H),
3.45ꢀ3.57 (m, 2H), 4.15ꢀ4.22 (m, 2H). 13C NMR (125 MHz, CDCl3)
δ/ppm: ꢀ4.7, ꢀ3.9, 17.9, 22.3, 25.8(3C), 25.9, 29.0, 29.4, 34.4, 46.5,
63.1, 64.6, 66.3, 199.83 HR MS (ESI) calcd for [M þ Na]þ C16H31NO-
NaSiS 336.1788, found 336.1783.
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-1-aza-bicyclo-
[5.2.0]non-4-en-9-thione (2b). Chromatographed using 5% AcOEt in
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-1-aza-bicyclo-
[5.2.0]non-4-en-9-one (2a). Catalyst - Grubbs’ first, reaction time 1 h,
3315
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
chromatographed using 10% AcOEt in hexane, colorless oil, yield 70%.
[R]D ꢀ12.2 (c 0.62, CH2Cl2). IR (film) ν/cmꢀ1: 1753, 142, 835, 777.
1H NMR (500 MHz, CDCl3) δ/ppm: 0.05 (s, 3H), 0.06 (s, 3H), 0.86 (s,
9H), 1.21 (d, J = 6.2 Hz, 3H), 2.15ꢀ2.21 (m, 1H), 2.33ꢀ2.42 (m, 3H),
2.74ꢀ2.81 (m, 2H), 3.55ꢀ3.58 (m, 1H), 3.76 (ddd, J = 13.3, 5.3, 3.5 Hz,
1H), 4.14 (dq, J = 6.2, 5.3 Hz, 1H), 5.85ꢀ5.95 (m, 2H). 13C NMR (125
MHz, CDCl3) δ/ppm: ꢀ5.0, ꢀ4.2, 17.9, 22.6, 25.7(3C), 27.7, 33.8,
39.2þ, 52.9, 63.9, 65.6, 129.8, 131.5, 166.3. HR MS (ESI) calcd for [M þ
Na] C16H29NO2NaSi 318.1860, found 318.1875.
(ddd, J = 13.3, 5.1, 3.6 Hz, 1H), 4.11 (qui, J = 6.2 Hz, 1H), 5.05 (d, J =
10.8 Hz, 1H), 5.15 (d, J = 17.4 Hz, 1H), 5.96 (t, J = 7.2 Hz, 1H), 6.35 (dd,
J = 17.4, 10.8 Hz, 1H). 13C NMR (125 MHz, CDCl3, δ/ppm) ꢀ5.0,
ꢀ4.2, 17.9, 22.7, 25.7(3C), 27.6, 31.5, 38.7, 52.9, 64.1, 66.1, 111.8, 131.7,
139.4, 140.5, 166.4. HR MS (ESI) calcd for [M þ Na]þ C18H31NO2-
NaSi 344.2009, found 344.2031.
Acetic Acid (6S,7S,8R)-9-oxo-8-phenoxy-4-vinyl-1-aza-bicyclo[5.2.0]non-
3-en-6-yl Ester (9). Catalyst - Grubbs’ second, reaction time 1.5 h, chro-
matographed using 15ꢀ20% AcOEt in hexane, colorless oil, yield 92%.
[R]D 74.1 (c 0.84, CH2Cl2). IR (film) ν/cmꢀ1: 1764, 1743, 1599, 1495,
1244. 1H NMR (500 MHz, CDCl3) δ/ppm: 1.84 (s, 3H), 2.54 (dd, J=16.8,
1.7 Hz, 1H), 3.15 (ddt, J = 16.8, 7.9, 1.5 Hz, 1H), 3.77ꢀ3.83 (m, 1H), 4.23
(dd, J=4.4, 2.8Hz, 1H), 4.52ꢀ4.58 (m, 1H), 5.01 (d, J= 10.9 Hz, 1H), 5.15
(d, J= 17.3 Hz, 1H), 5.35 (dd, J= 4.4, 1.1 Hz, 1H), 5.68ꢀ5.71 (m, 1H), 6.32
(dd, J = 17.3, 10.9 Hz, 1H), 6.97ꢀ7.03 (m, 3H), 7.27ꢀ7.31 (m, 2H). 13C
NMR (125 MHz, CDCl3) δ/ppm: 20.8, 30.2, 40.4, 59.9, 67.4, 80.8, 112.5,
115.2(2C), 122.2, 126.4, 129.6(2C), 135.4, 140.9, 157.6, 165.6, 170.2. HR
MS (ESI) calcd for [M þ Na]þ C18H19NO4Na 336.1206, found 336.1198.
(3R,4R)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-4-vinyl-6-oxa-
1-aza-bicyclo[5.2.0]non-3-en-9-one (10). Catalyst - Grubbs’ first, reac-
tion time 1.5 h, chromatographed using 15% AcOEt in hexane, colorless
(7R,8R)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-6-oxa-1-aza-
bicyclo[5.2.0]non-3-en-9-one (3a). Catalyst - Grubbs’ first, reaction time
2 h, chromatographed using 10% AcOEt in hexane, colorless oil, yield
1
81%. [R]D 62.4 (c 0.50, CHCl3). IR (film) ν/cmꢀ1: 1768. H NMR
(500 MHz, CDCl3) δ/ppm: 0.06 (s, 3H), 0.08 (s, 3H), 0.87 (s, 9H),
1.26 (d, J = 6.3 Hz, 3H), 3.05 (bd, J = 3.4 Hz, 1H), 3.66ꢀ3.73 (m, 1H),
4.18ꢀ4.24 (m, 2H), 4.36ꢀ4.46 (m, 2H), 5.23 (s, 1H), 5.56ꢀ5.61 (m,
1H), 5.63ꢀ5.68 (m, 1H). 13C NMR (125 MHz, CDCl3) δ/ppm: ꢀ5.2,
ꢀ4.2, 17.9, 22.4, 24.3, 25.7(3C), 41.1, 57.9, 64.6, 66.7, 127.5, 133.2,
166.3. Elemental Analysis calcd for C15H27NO3Si: C, 57.46; H, 8.68; N,
4.47. Found: C, 57.60; H, 8.89; N, 4.27.
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-6-thia-1-aza-
bicyclo[5.2.0]non-3-en-9-one (4a). Catalyst - Grubbs’ first, reaction
time 5 h, chromatographed using 10% AcOEt in hexane, colorless oil,
yield 81%. [R]D 31.2 (c 0.51, CH2Cl2). IR (film) ν/cmꢀ1: 1762, 836,
778. 1H NMR (500 MHz, CDCl3) δ/ppm: 0.05 (s, 3H), 0.08 (s, 3H),
0.87 (s, 9H), 1.22 (d, J = 6.30 Hz, 3H), 2.99 (dd, J = 16.39, 8.04 Hz, 1H),
3.15 (dd, J = 3.80, 1.74 Hz, 1H), 3.45ꢀ3.50 (m, 1H), 3.63ꢀ3.68 (m,
1H), 4.20ꢀ4.25 (m, 1H), 4.33 (dd, J = 18.1, 5.29 Hz, 1H), 4.90 (s, 1H),
5.50ꢀ5.56 (m, 1H), 5.93ꢀ5.96 (m, 1H). 13C NMR (125 MHz, CDCl3)
δ/ppm: ꢀ5.0, ꢀ4.2, 18.2, 22.5, 24.7, 25.8(3C), 41.3, 5þ8.4, 65.2, 67.1,
128.0, 130.6, 166.3. HR MS (ESI) calcd for [M þ Na] C15H27NO2-
NaSiS 336.1424, found 336.14233.
1
oil, yield 80%. IR (film) ν/cmꢀ1: 1767, 1737, 1645, 1472, 1252. H
NMR (500 MHz, CDCl3) δ/ppm: 0.06 (s, 3H), 0.08 (s, 3H), 0.86 (s,
9H), 1.27 (d, J = 6.3 Hz, 3H), 3.07 (d, J = 3.5 Hz, 1H), 3.77 (d, J = 1958
Hz, 1H), 4.22 (dq, J = 6.3, 3.5 Hz, 1H), 4.37 (dd, J = 15.6, 1.2 Hz, 1H),
4.46 (dd, J = 19.5, 4.8 Hz, 1H), 4.62 (d, J = 15.6 Hz, 1H), 5.03 (dd, J =
11.2, 0.6 Hz, 1H), 5.10 (d, J = 17.7 Hz, 1H), 5.27 (s, 1H), 5.63 (dd, J =
4.0, 3.3 Hz, 1H), 6.25 (dd, J = 17.7, 11.2 Hz). 13C NMR (125 MHz,
CDCl3) δ/ppm: ꢀ5.2, ꢀ4.3, 17.9, 22.5, 25.6(3C), 40.9, 63.6, 64.0, 64.3,
84.5þ, 112.3, 127.4, 138.3, 138.4, 165.9. HR MS (ESI) calcd for [M þ
Na] C17H29NO3NaSi 346.1809, found 346.1822.
Acetic Acid (6R,7S,8R)-9-Oxo-8-phenoxy-1-aza-bicyclo[5.2.0]non-
3-en-6-yl Ester (6) and Acetic Acid (6S,7S,8R)- 9-Oxo-8-phenoxy-1-
aza-bicyclo[5.2.0]non-3-en-6-yl ester (7). Catalyst - Grubb’s first,
reaction time 2.5 h, chromatographed using 10% acetone in hexane,
colorless oil, yield 86%. Sample of the mixture (5 mg) was separated
using HPLC (detection at 202 nm, column Kromasil 100 Si (5 μ) 25 ꢁ
0.4 cm, 10% iPrOH in hexane, flow 1 mL/min) yielding compound 6 (tR
= 13.1 min, 2.94 mg) and 7 (tR = 11.1 min, 0.98 mg). Data for 6: [R]D
63.9 (c 0.64, CH2Cl2). IR (film) ν/cmꢀ1: 1764, 1737, 1597, 1489, 1241.
1H NMR (500 MHz, CDCl3) δ/ppm: 1.80 (s, 3H), 2.49ꢀ2.51 (m, 1H),
2.71ꢀ2.81 (m, 1H), 3.63 (d, J = 17.5 Hz, 1H), 4.12 (dd, J = 5.9, 4.2 Hz,
1H), 4.46 (dd, J = 17.5, 4.2 Hz, 1H), 5.19 (ddd, J = 8.3, 5.9, 2.4 Hz, 1H),
5.39 (d, J = 4.2 Hz, 1H), 5.62ꢀ5.70 (m, 2H), 6.98ꢀ7.05 (m, 3H),
7.24ꢀ7.32 (m, 2H). 13C NMR (125 MHz, CDCl3) δ/ppm: 20.9, 32.5,
41.6, 60.1, 68.5, 80.3, 115.2(2C), 122.2, 124.7, 125.2, 129.6(2C), 157.6,
165.8, 170.2. HR MS (ESI) calcd for [M þ Na]þ C16H17NO4Na
310.1049, found 310.1037. Data for 7: [R]D 60.5 (c 0.45, CH2Cl2). IR
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-3-methylene-1-
aza-bicyclo[5.2.0]nonan-9-one (5a). A solution of tributyltinhydride (107
mg, 0.37 mmol) in benzene (1 mL) was added within 1 h to boiling solution
of β-lactam 14 (80 mg, 0.21 mmol) and AIBN (23 mg, 0.14 mmol) in
benzene (9 mL). Then the benzene was evaporated, and the residue was
chromatographed on silica gel (gradient 5ꢀ15% AcOEt in hexane) to
produce cyclic product as colorless oil (51 mg, 80%). [R]D 31.3 (c 0.57,
CH2Cl2). IR (film) ν/cmꢀ1: 1754. 1H NMR (500 MHz, CDCl3) δ/ppm:
0.05 (s, 6H), 0.85 (s, 9H), 1.18 (d, J = 6.3 Hz, 3H), 1.31ꢀ1.45 (m, 2H),
1.87ꢀ1.94 (m, 1H), 1.94ꢀ2.04 (m, 2H), 3.55 (dt, J = 10.6, 2.1 Hz, 1H),
3.69 (d, J = 13.9 Hz, 1H), 4.09 (bd, J = 13.9 Hz, 1H), 4.15 (qd, J = 6.2, 4.7
Hz, 1H), 4.91 (bs, 1H), 4.94 (bs, 1H). 13C NMR (125 MHz, CDCl3) δ/
ppm: ꢀ5.0, ꢀ4.2, 17.84, 22.6, 25.7(3C), 29.9, 35.7, 36.5, 49.2, 55.3, 63.2,
65.5, 116.2, 144.6, 167.5. HR MS (LSI) calcd for [M þ Na]þ C17H31NO2-
NaSi 332.2016, found 332. 2023.
1
(film) ν/cmꢀ1: 1764, 1737, 1597, 1489, 1241. H NMR (500 MHz,
CDCl3) δ/ppm: 1.89 (s, 3H), 2.43 (d, J = 17.7 Hz, 1H), 2.88ꢀ2.94 (m,
1H), 3.73 (d, J = 19.1 Hz, 1H), 4.22 (dd, J = 4.2, 2.5 Hz, 1H), 4.53 (dd, J
= 19.1, 2.4 Hz, 1H), 5.33 (d, J = 4.2 Hz, 1H), 5.41 (ddd, J = 6.6, 2.5, 2.4
Hz, 1H), 5.56ꢀ5.59 (m, 2H), 6.97ꢀ7.03 (m, 3H), 7.25ꢀ7.33 (m, 2H).
13C NMR (125 MHz, CDCl3) δ/ppm: 20.6, 31.1, 40.6, 62.2, 70.7, 81.1,
115.4(2C), 122.3, 125.3, 1þ27.1, 129.5(2C), 157.4, 164.5, 169.7. HR MS
(ESI) calcd for [M þ Na] C16H17NO4Na 310.1049, found 310.1037.
(7R,8S)-8-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-5-vinyl-1-aza-
bicyclo[5.2.0]non-4-en-9-one (8). Catalyst - Grubbs’ first, reaction time
9 h, chromatographed using 10% AcOEt in hexane, colorless oil, yield
1
60%. IR (film) ν/cmꢀ1: 1751, 1410, 1256, 1143, 835. H NMR (500
MHz, CDCl3) δ/ppm: 0.05 (s, 3H), 0.07 (s, 3H), 0.91 (s, 9H), 1.25 (d, J
= 6.2 Hz, 3H), 2.22ꢀ2.29 (m, 1H), 2.40ꢀ2.47 (m, 2H), 2.72ꢀ2.82 (m,
2H), 2.83 (dd, J = 6.2 Hz, 1.8, 1H), 3.45 (dt, J = 10.9, 2.1 Hz, 1H), 3.80
(3S,4R)-4-Allyl-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-ethyl]-azeti-
din-2-one (12a)13. A mixture of potassium iodide (1.74 g, 10.5 mmol),
3316
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
allyl bromide (0.88 mL, 10.5 mmol) and powdered indium (0.79 g,
6.94 mmol) in DMF (40 mL) was stirred for 1 h. Then the solution of
compound 11 (1 g, 3.46 mmol) in DMF (10 mL) was added and the
stirring was continued at room temperature for 24 h. Then the insoluble
material was filtered off and washed with diethyl ether. The filtrate was
poured into ammonium chloride and extracted with diethyl ether. The
organic layer was washed with brine and dried over MgSO4. Then the
solvent was evaporated, and the residue was purified by chromatography
on silica gel (10ꢀ20% AcOEt in hexane) to produce 12a (807 mg, 86%).
[R]D þ8.67 (c 0.67, CH2Cl2). IR (film) ν/cmꢀ1: 1760. Elemental
Analysis calcd for C14H27NO2SiS: C, 62.40; H, 10.10; N, 5.20, Found:
C, 62.60; H, 10.22; N, 5.32. HR MS (LSI) calcd for [M þ Na]þ
C14H27NO2NaSi 292.17037, found 292.17026.
into the mixture and the stirring was continued for 1 h. The mixture was
poured into water and extracted with diethyl ether. The organic layer was
washed with brine and dried over MgSO4. Then the solvent was
removed, and the residue was purified by chromatography on silica gel
to produce 13b, 13c, 14, or 23.
General Procedure for N-Alkylation of β-Lactams with 4-Bro-
mo-1-butene. A mixture of compound 12a or 18 (1.56 mmol), potas-
sium iodide (0.13 mmol), Bu4NHSO4 (0.58 mmol), THF (15 mL),
powdered potassium hydroxide (3.6 mmol), and 4-bromo-1-butene (3.6
mmol) was stirred at room temperature in the argon atmosphere for 24
h. Then the insoluble material was filtered off and washed with THF.
The filtrate was washed with ammonium chloride, brine and dried over
MgSO4. Then the solvent was evaporated, and the residue was purified
by chromatography on silica gel (5% acetone in hexane) to produce 13a
or 19.
(3R,4R)-1-But-3-enyl-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-ethyl]-4-
vinyloxy-azetidin-2-one (13b). Chromatographed using 10% AcOEt in
hexane, colorless oil, yield 87%. [R]D 12.0 (c 0.56, CH2Cl2). IR (film) ν/
cmꢀ1: 1769, 1092, 837, 778. 1H NMR (500 MHz, CDCl3) δ/ppm: 0.05
(s, 3H), 0.08 (s,3H), 0.87 (s, 9H), 1.25 (d, J = 6.3 Hz, 3H), 2.99 (d, J =
3.8 Hz, 1H), 3.72 (ddt, J = 15.7, 6.8, 1.2 Hz, 1H), 3.95 (ddt, J = 15.7, 5.4,
1.6 Hz, 1H), 4.04ꢀ4.11 (m, 2H), 4.14ꢀ4.18 (m, 1H), 5.04 (bd, J = 1.2
Hz, 1H), 5.18ꢀ5.22 (m, 2H), 5.25ꢀ5.33 (m, 2H), 5.75ꢀ5.83 (m, 1H),
5.86ꢀ5.94 (m, 1H). 13C NMR (125 MHz, CDCl3) δ/ppm: ꢀ4.9, ꢀ4.6,
17.9, 22.6, 25.7(3C), 43.0, 64.2, 64.6, 68.8, 83.3, 117.1, 118.2, 132.2,
133.9, 166.8. Elemental Analysis calcd for C17H31NO3Si: C, 62.73; H,
9.60; N, 4.30. Found: C, 62.78; H, 9.55; N, 4.35. HR MS (LSI) calcd for
[M þ Na]þ C17H31NO3NaSi 348.1965, found 348.1978.
(3S,4R)-4-Allyl-1-but-3-enyl-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-
ethyl]-azetidin-2-one (13a). Colorless oil, yield 60%. [R]D ꢀ27.39 (c
1
0.81, CH2Cl2). IR (film) ν/cmꢀ1: 1754, 1642, 1472, 1405, 1256. H
NMR (500 MHz, CDCl3) δ/ppm: 0.06 (s, 3H), 0.07 (s, 3H), 0.88 (s,
9H), 1.19 (d, J = 6.2 Hz, 3H), 2.25ꢀ2.36 (m, 3H), 2.46ꢀ2.52 (m, 1H),
2.74 (dd, J = 5.2, 1.9 Hz, 1H), 3.03ꢀ3.09 (m, 1H), 3.41 (ddd, J = 14.4, 8.0,
6.9 Hz, 1H), 3.66 (ddd, J = 7.2, 5.3, 2.0 Hz, 1H), 4.12 (quintet, J = 6.2 Hz,
1H), 5.04ꢀ5.19 (m, 4H), 5.73ꢀ5.84 (m, 2H). 13C NMR (125 MHz,
CDCl3) δ/ppm: ꢀ4.8, ꢀ4.4, 17.9, 22.5, 25.8(3C), 32.7, 37.32, 4.8, 54,0,
62.4þ, 65.9, 116.9, 118.2, 133.5, 135.2, 167.7. HR MS (LSI) calcd for [M þ
Na] C18H33NO2NaSi 346.21728, found 346. 21567.
(3S,4R)-1-But-3-enyl-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-ethyl]-4-
prop-2-ynyl-azetidin-2-one (19). Colorless oil, yield 58%. IR (film) ν/
cmꢀ1: 1755, 1642, 1472, 1257, 1142. 1H NMR (500 MHz, CDCl3) δ/
ppm: 0.07 (s, 3H), 0.08 (s, 3H), 0.88 (s, 9H), 1.23 (d, J = 6.3 Hz, 3H),
2.05 (t, J = 2.6 Hz, 1H), 2.29ꢀ2.39 (m, 2H), 2.53ꢀ2.56 (m, 2H), 2.88
(dd, J = 4.8, 1.9 Hz, 1H), 3.12ꢀ3.18 (m, 1H), 3.42 (ddd, J = 14.5, 7.9, 6.9
Hz, 1H), 3.78 (dt, J = 5.4, 1.9 Hz, 1H), 4.16 (dq, J = 6.2, 6.1 Hz, 1H), 5.06
(dd, J = 10.2, 1.3 Hz, 1H), 5.11 (dd, J = 17.1, 1.6 Hz, 1H), 5.78 (ddt, J =
17.1, 10.2, 6.9 Hz, 1H). 13C NMR (125 MHz, CDCl3) δ/ppm: ꢀ4.8,
ꢀ4.4, 17.9, 22.5, 22.8, 25.8(3C), 32.7, 40.0, 52.4, 62.6, 65.4, 71.2, 79.5,
116.9, 135.1, 167.4. HR MS (TOF FDþ) calcd for [M]þ C18H31NO2Si
321.2124, found 321.2114.
(3S,4R)-1-But-3-enyl-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-
ethyl]-4-vinylsulfanyl-azetidin-2-one (13c). Chromatographed using
5% AcOEt in hexane, colorless oil, yield 82%. [R]D 52.9 (c 0.50, CH2Cl2).
IR (film) ν/cmꢀ1: 1763, 1722, 1251, 1146, 1136, 1061, 962, 836, 778.
1H NMR (500 MHz, CDCl3) δ/ppm: 0.07 (s, 3H), 0.09 (s, 3H), 0.89 (s,
9H), 1.23 (d, J = 6.3 Hz, 3H), 3.16 (t, J = 2.3 Hz, 1H), 3.22 (d, J = 7.6 Hz,
2H), 3.60 (ddd, J = 15.8, 7.0, 0.8 Hz, 1H), 4.06 (ddt, J = 15.8, 5.0, 1.5 Hz,
1H), 4.27 (qd, J = 6.3, 3.1 Hz, 1H), 4.80 (d, J = 2.2 Hz, 1H), 5.13ꢀ5.24
(m, 2H), 5.30 (dd, J = 17.1, 1.2 Hz, 1H), 5.76ꢀ5.89 (m, 2H). 13C NMR
(125 MHz, CDCl3) δ/ppm: ꢀ4.8, ꢀ4.6, 17.9, 22.3, 25.7(3C), 33.1,
42.4, 57.6, 64.5, 65.0, 117.8, 118.4, 132.1, 134.1, 166.4. HR MS (ESI)
calcd for [M þ Na]þ C17H31NO3NaSiS 364.1737, found 364.1729.
(3S,4R)-4-Allyl-1-(2-bromo-allyl)-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-
ethyl]-azetidin-2-one (14). Chromatographed using 5% AcOEt in hexane,
colorless oil, yield 47%. [R]D ꢀ12.4 (c 0.66, CH2Cl2). IR (film) ν/cmꢀ1
:
1760. 1H NMR (500 MHz, CDCl3) δ/ppm: 0.06 (s, 3H), 0.08 (s, 3H),
0.88 (s, 9H), 1.22 (d, J = 6.2 Hz, 3H), 2.39ꢀ2.34 (m, 1H), 2.53ꢀ2.49 (m,
1H), 2.84 (dd, J = 5.7, 2.1 Hz, 1H), 3.77 (m, 1H), 3.85 (d, J = 16.3 Hz, 1H),
4.18 (p, J= 6.1 Hz, 1H), 4.24 (d, J = 16.3 Hz, 1H), 5.18ꢀ5.11 (m, 2H), 5.60
(m, 1H), 5.82ꢀ5.74 (m, 1H), 5.86 (m, 1H). 13C NMR (125 MHz, CDCl3)
δ/ppm: ꢀ4.3, ꢀ4.1, 18.3, 23.3, 26.2, 37.48, 48.9, 54.9, 63.4, 66.2, 118.7, 119.6,
General Procedure for N-Alkylation of β-Lactams with
Allyl Bromide. A mixture of lactam (12b, 12c, 12a, or 22) (0.70
mmol), anhydrous DMF (12 mL) and sodium hydride (1.3 mmol) was
stirred at room temperature in the argon atmosphere until evolution of
hydrogen had ceased. Then the allyl bromide (3.45 mmol) was dropped
3317
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
127.9, 133.6, 168.0. HR MS (LSI) calcd for [M þ Na]þ C17H30NO2NaSiBr
410.1121, found 410.1118.
Hz, 1H), 4.16 (ddt, J = 15.8, 5.3, 1.6 Hz, 1.2H), 4.24 (ddt, J = 15.4, 5.2,
1.6 Hz, 1H), 5.06ꢀ5.11 (m, 2.2H), 5.23ꢀ5.29 (m, 8.8H), 5.36ꢀ5.40
(m, 2.2H), 5.71ꢀ5.82 (m, 4.4H), 7.00ꢀ7.06 (m, 2.2H), 7.07ꢀ7.10 (m,
4.4H), 7.28ꢀ7.34 (m, 4.4H). Data for 17a: 13C NMR (125 MHz,
CDCl3) δ/ppm: 21.00, 36.56, 44.26, 59.05, 71.26, 79.99, 115.54(2C),
118.64, 122.47, 129.64(2C), 131.12, 133.06, 157.05, 166.12, 170.11.
Data for 17b: 13C NMR (125 MHz, CDCl3) δ/ppm: 21.2, 35.2, 44.3,
58.3, 72.0, 80.0, 115.8(2C), 118.7, 129.6(2C), 131.4, 132.5, 157.5, 122.4,
166.2, 169.8. HR MS (ESI) calcd for [M þ Na]þ C18H21NO4Na
338.1263, found 338.1351.
(3R,4R)-1-Allyl-3-[(1R)-(tert-butyl-dimethyl-silanyloxy)-ethyl]-4-prop-
2-ynyloxy-azetidin-2-one (23). Chromatographed using 15% AcOEt in
hexane, colorless oil, yield 93%. [R]D 9.1 (c 0.87, CH2Cl2). IR (film)
ν/cmꢀ1: 1767, 1645, 1472, 1258. 1H NMR (500 MHz, CDCl3) δ/ppm:
0.06 (s, 3H), 0.08 (s, 3H), 0.88 (s, 9H), 1.27 (d, J = 6.3 Hz, 3H), 2.48 (t, J
= 2.4 Hz, 1H), 3.05 (dd, J = 3.8, 1.2 Hz, 1H), 3.74 (ddd, J = 15.7, 6.8, 1.1
Hz, 1H), 3.95 (ddt, J = 15.7, 5.5, 1.5 Hz, 1H), 4.17 (dq, J = 6.3, 3.8 Hz,
1H), 4.22 (dd, J = 2.6, 0.5 Hz, 2H), 5.18 (d, J = 1.0 Hz, 1H), 5.21 (dq, J =
10.2, 1.0 Hz, 1H), 5.29 (dq, J = 17.1, 1.4 Hz, 1H), 5.80 (dddd, J = 17.1,
10.2, 6.8, 5.5 Hz, 1H). 13C NMR (125 MHz, CDCl3) δ/ppm: ꢀ4.8,
ꢀ4.5, 17.90, 22.6, 25.8(3C), 40.2, 55.5, 64.2, 64.7, 75.1, 79.2, 83.2, 118.5,
132.1, 166.6. HR MS (ESI) calcd for [M þ Na]þ C17H29NO3NaSi
346.1809, found 346.1822.
Acetic Acid (1S)-1-[(2S,3R)-(1-Allyl-4-oxo-3-phenoxy-azetidin-2-yl)-
but-3-ynyl Ester (21). Colorless oil, yield 95%. [R]D 91.7 (c 1.06,
CH2Cl2). IR (film) ν/cmꢀ1: 3291, 1762, 1495, 1230, 1043, 756, 691.
1H NMR (500 MHz, CDCl3) δ/ppm: 2.04 (t, J = 2.7 Hz, 1H), 2.10 (s,
3H), 2.67 (ddd, J = 17.6, 4.6, 2.7 Hz, 1H), 2.78 (ddd, J = 17.6, 5.3, 2.7 Hz,
1H), 3.79 (ddt, J = 15.8, 6.3, 1.2 Hz, 1H), 4.11 (ddt, J = 15.8, 5.3, 1.4 Hz,
1H), 4.36 (dd, J = 7.8, 5.1 Hz, 1H), 5.30 (d, J = 5.1 Hz, 1H), 5.78ꢀ5.85
(m, 1H), 7.01ꢀ7.05 (m, 1H), 7.08ꢀ7.11 (m, 2H), 7.28ꢀ7.32 (m, 2H).
13C NMR (125 MHz, CDCl3) δ/ppm: 21.3, 21.5, 44.4, 57.6, 70.7, 71.5,
78.3, 80.1, 115.8(2C), 118.6, 122.6, 129.7(2C), 131.3, 157.4, 165.9,
169.7. HR MS (ESI) calcd for [M þ Na]þ C16H17NO4Na 310.1049,
found 310.1037.
(3R,4S)-1-Allyl-4-[(1S)-hydroxy-but-3-ynyl)]-3-phenoxy-azetidin-2-
one (20). A mixture of compound 15 (1 g, 4.3 mmol), zinc powder (1.7
g, 26.0 mmol), THF (10 mL), and saturated solution of ammonium
chloride (25 mL) was cooled to ꢀ15 ꢀC. Then propargyl bromide was
added (1.4 mL, 13.0 mmol), and the mixture was stirred at ꢀ15 ꢀC for 4
h. Then the insoluble material was filtered off and washed with THF.
The filtrate was washed with ethyl acetate and dried over MgSO4. Then
the solvent was evaporated, and the residue was purified by chromatog-
raphy on silica gel (15% AcOEt in hexane) to produce 16 (860 mg, 73%)
as colorless oil. [R]D 149.6 (c 0.71, CH2Cl2). IR (film) ν/cmꢀ1: 1753,
1644, 1598, 1495, 1238. 1H NMR (500 MHz, CDCl3) δ/ppm: 2.04 (t, J
= 2.7 Hz, 1H), 2.11 (s, 1H), 2.67 (ddd, J = 17.6, 4.6, 2.7 Hz, 1H), 2.78
(ddd, J = 17.6, 5.3, 2.7 Hz, 1H), 3.78 (ddt, J = 15.8, 6.3, 1.2 Hz, 1H), 4.11
(ddt, J = 15.8, 5.3, 1.4 Hz, 1H), 4.36 (dd, J = 7.8, 5.1 Hz, 1H), 5.24ꢀ5.29
(m, 1H), 5.30 (d, J = 5.1 Hz, 1H), 5.78ꢀ5.85 (m, 1H), 7.01ꢀ7.05 (m,
1H), 7.08ꢀ7.11 (m, 2H), 7.28ꢀ7.32 (m, 2H). 13C NMR (125 MHz,
CDCl3) δ/ppm: 24.3, 44.6, 59.6, 69.3, 71.7, 79.4, 80.4, 115.9(2C),
118.9, 122.8, 129.7(2C), 131.8, 157.3, 165.9. HR MS (ESI) calcd for [M
þ Na]þ C16H17NO3Na 294.1100, found 294.1091.
(3R,4R)-3-[(1R)-(tert-Butyl-dimethyl-silanyloxy)-ethyl]-4-prop-2-yny-
loxy-azetidin-2-one (22). A mixture of acetoxyazetidinone 11 (2 g,
6.96 mmol), powdered Zn(OAc)2 2H2O (0.92 g, 4.18 mmol), toluene
3
(25 mL), and propargyl alcohol (1.34 mL, 22.97 mmol) was refluxed for 6
h. Then the insoluble material was filtered off and washed with THF.
Then the solvent was evaporated from the filtrate, and the residue was
purified by chromatography on silica gel (15% AcOEt in hexane) to
produce 22 (1.36 g, 69%) as colorless oil. [R]D ꢀ19.3 (c 0.94, CH2Cl2).
IR (film) ν/cmꢀ1: 1767, 1726, 1472, 1256. 1H NMR (500 MHz, CDCl3)
δ/ppm: 0.07 (s, 3H), 0.08 (s, 3H), 0.88 (s, 9H), 1.27 (d, J = 6.3 Hz, 3H),
2.49 (t, J = 2.4 Hz, 1H), 3.09 (dd, J = 4.0, 1.2 Hz, 1H), 4.17 (dq, J = 6.3, 4.0
Hz, 1H), 4.24 (dd, J = 8.2, 2.4 Hz, 2H), 5.24 (d, J = 1.0 Hz, 1H), 6.44 (bs,
1H). 13C NMR (125 MHz, CDCl3) δ/ppm: ꢀ5.1, ꢀ4.3, 17.9, 22.4,
25.7(3C), 55.8, 64.2, 66.0, 75.3, 79.1, 80.3, 167.2. HR MS (ESI) calcd for
[M þ Na]þ C14H25NO3NaSi 306.1496, found 306.1495. Elemental
Analysis calcd for C14H25NO3Si: C, 59.32; H, 8.89; N, 4.94. Found: C,
59.25; H, 8.86; N, 4.94.
Acetylation of Alcohols 16 and 20. A solution of lactams 16 or
20 (0.73 mmol) in pyridine (6 mL) and acetic anhydride (3 mL) was
kept at room temperature for 15 h. Then the reaction mixture was
poured into water and extracted with diethyl ether. The organic layer was
washed consecutively with 10% HCl and saturated solution of NaHCO3
and dried over MgSO4. Then the solvent was removed, and the residue
was chromatographed on silica gel (5% AcOEt in hexane) to produce 17
or 21.
’ ASSOCIATED CONTENT
Supporting Information. 1H and 13C NMR spectra of all
S
b
Acetic Acid (2S,3R)-1-(1-Allyl-4-oxo-3-phenoxy-azetidin-2-yl)-but-
3-enyl Ester (17). Colorless oil, yield 90%. IR (film) ν/cmꢀ1: 1765,
1643, 1598, 1495, 1231. 1H NMR (500 MHz, CDCl3) δ/ppm: 1.93 (s,
3H), 2.05 (s, 3.6H), 2.34ꢀ2.44 (m, 2.2H), 2.55 (ddt, J = 6.3, 4.9, 1.4 Hz,
1.2H), 2.58 (ddt, J = 6.3, 4.9, 3.3 Hz, 1H), 3.63ꢀ3.68 (m, 1.2H),
3.69ꢀ3.73 (m, 1H), 4.05 (dd, J = 7.3, 5.3 Hz, 1.2H), 4.10 (dd, J = 5.2, 4.1
new compounds, crystallographic data (CIF) of compound 1b,
information about details of calculated UV and CD spectra, total
energies, relative energies, and Cartesian coordinates for all
optimized structures used in this work. This material is available
free of charge via the Internet at http://pubs.acs.org.
3318
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319

The Journal of Organic Chemistry
ARTICLE
’ AUTHOR INFORMATION
(13) Kang, S.;Baik, T.;Jiao, X.;Lee, K.;Lee, C.Synlett 1999, 4, 447–449.
(14) (a) Pedersen, B. S.; Scheibye, S.; Nilsson, N. H.; Lawesson, S.-O.
Bull. Soc. Chim. Belg. 1978, 87, 223–228. (b) Scheibye, S.; Kristensen, J.;
Lawesson, S.-O. Tetrahedron 1978, 35, 1339–1343.
(15) PCModel 9.1; Serena Software: Bloomington, 2005.
(16) Grimme, S. J. Comput. Chem. 2006, 27, 1787–1799.
(17) (a) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988,
37, 785–789. (b) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100.
(18) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005,
7, 3297–3305.
Corresponding Author
*E-mail: jadwiga.frelek@icho.edu.pl; filipp.furche@uci.edu; grimmes@
uni-muenster.de.
’ ACKNOWLEDGMENT
This work was supported by the Ministry of Science and
Higher Education, grant nos. N N204 123537 (M.W.) and N
N204 092935(J.F., A.B., K.M.), and by the DFG in the frame-
work of the International Research Training Group 1444 with a
scholarship to H.K. F.F. acknowledges support by the Center for
Chemical Innovation “Chemistry at the Space-Time Limit”
(CaSTL) of the NSF, grant no. CHE-0802913.
(19) Grimme, S. J. Chem. Phys. 2006, 124, 034108.
(20) Ahlrichs, R.; Armbruster, M. K.; B€ar, M., Baron, H.-P.;
Bauernschmitt, R.; Crawford, N.; Deglman, P.; Ehrig, M.; Eichkorn,
K.; Elliott, S.; Furche, F.; Haase, F.; H€aster, M.; H€attig, C.; Hellweg, H.;
Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.; K€omel, C.; Kollwitz,
€
M.; May, P.; Nava, P.; Ochsenfeld, C.; Ohm, H.; Patzelt, D.; Rappoport,
D.; Rubner, O.; Sch€afer, A.; Schneider, U.; Sierka, M.; Treutler, O.;
Unterreiner, B.; von Arnim, M.; Weigand, F.; Weis, P.; Weiss, H.
TURBOMOLE Vers. 6.0; University of Karlsruhe: Karlsruhe, 2009. See
also: www.turbomole.com.
(21) Vahtras, O.; Almlof, J.; Feyereisen, M. W. Chem. Phys. Lett.
1993, 213, 514–518.
’ REFERENCES
(1) (a) Mann, J.; Crabbe, M. J. C. Bacteria and Antibacterial Agents;
Oxford University Press: New York, 1996. (b) Cooper, R. D. G. Topics in
Antibiotics Chemistry; Ellis Horwood Ltd.: Chichester, 1980. (c) Brown,
A. G.; Butterworth, D.; Cole, M.; Hanscomb, G.; Hood, J. D.; Reading,
C.; Rolinson, G. N. J. Antibiot. 1976, 29, 668–669.
(22) Feyereisen, M.; Fitzgerald, G.; Komornicki, A. Chem. Phys. Lett.
1993, 359–363.
(23) Hattig, C.; Weigend, F. J. Chem. Phys. 2000, 113, 5154–5161.
(24) Furche, F.; Rappoport, D. In Density Functional Methods for
Excited States:Equilibrium Structure and Electronic Spectra; Elsevier:
Amsterdam, 2005.
(25) (a) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346–354.
(b) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chem.
Acc. 1997, 97, 119–124.
(26) Perdew, J. P.; Ernzerhof, M.; Burke, K. J. Chem. Phys. 1996,
105, 9982–9985.
(27) Radhakrishnan, T. P.; Agranat, I. Struct. Chem. 1991, 2, 107–115.
(28) (a) Yde, B.; Yousif, N. M.; Pedersen, U.; Thomsen, I.; Lawesson,
S.-O. Tetrahedron 1984, 40, 2047–2052. (b) Maciejewski, A.; Steer, R. P.
Chem. Rev. 1993, 93, 67–98.
(29) Kajtar, M.; Kajtar, J.; Maier, Z.; Zewdu, M.; Hollosi, M.
Spectrochim. Acta 1993, 48B, 87–91.
(30) Penfold, D. J.; Pike, K.; Genge, A.; Anson, M.; Kitteringham, J.;
Kilburn, J. D. Tetrahedron Lett. 2000, 41, 10347–10351.
(31) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purifications of
Laboratory Chemicals; Pergamon Press: Oxford, 1980.
(2) Sweet, R. M.; Dahl, L. I. J. Am. Chem. Soc. 1970, 92, 5489–5507.
(3) (a) Boyd, D. B. J. Am. Chem. Soc. 1972, 94, 6513–6519. (b)
Toome, V.; Wegrzynski, B.; Reymond, G. Biochem. Biophys. Res. Com-
mun. 1976, 69, 206–211. (c) Busson, R.; Roets, E.; Vanderhaeghe, H.
J. Org. Chem. 1978, 43, 4438–4441.
(4) (a) Boyd, D. B. J. Med. Chem. 1973, 16, 1195–1199. (b) Rehling,
H.; Jensen, H. Tetrahedron Lett. 1972, 27, 2793–2796. (c) Ogura, H.;
Takayanagi, H.; Kubo, K.; Furuhata, K. J. Am. Chem. Soc. 1973,
95, 8056–8059. (d) Ogura, H.; Takeda, K.; Takahashi, H. Chem. Pharm.
Bull. 1975, 23, 2469–2473. (e) McCann, J.; Rauk, A.; Shustov, G. V.;
Wieser, H.; Yang, D. Appl. Spectrosc. 1996, 50, 630–641. (f) Galle, D.;
Tolksdorf, M.; Braun, M. Tetrahedron Lett. 1995, 36, 4217–4720. (g)
Wolf, H. Tetrahedron Lett. 1966, 5151–5156. (h) Barbaro, G.; Battaglia,
A.; Guerrini, A.; Bertucci, C.; Geremia, S. Tetrahedron: Asymmetry 1998,
9, 3401–3409. (i) Polonski, T.; Milewska, M. J. Croat. Chim. Acta 1989,
62, 129–134.
(5) yysek, R.; Borsuk, K.; Chmielewski, M.; Kazuz_a, Z.; Urbaꢀnczyk-
Lipkowska, Z.; Klimek, A.; Frelek, J. J. Org. Chem. 2002, 67, 1472–1479.
(6) (a) Cierpucha, M.; Solecka, J.; Frelek, J.; Szczukiewicz, P.;
Chmielewski, M. Biorg. Med. Chem. 2004, 12, 405–416. (b) Frelek, J.;
yysek, R.; Borsuk, K.; Jagodzinꢀski, J.; Furman, B.; Klimek, A.;
Chmielewski, M. Enantiomer 2002, 7, 107–114. (c) Chmielewski, M.;
Cierpucha, M.; Kowalska, P.; Kwit, M.; Frelek, J. Chirality 2008,
20, 621–627.
(7) Frelek, J.; Kowalska, P.; Masnyk, M.; Kazimierski, A.; Korda, A.;
Woꢀznica, M.; Chmielewski, M.; Furche, F. Chem.—Eur. J. 2007,
13, 6732–6744.
(8) Woꢀznica, M.; Kowalska, P.; yysek, R.; Masnyk, M.; Gꢀorecki, M.;
Kwit, M.; Furche, F.; Frelek, J. Curr. Org. Chem. 2010, 14, 1022–1036.
(9) Woꢀznica, M.; Masnyk, M.; Stecko, S.; Mames, A.; Furman, B.;
Chmielewski, M.; Frelek, J. J. Org. Chem. 2010, 75, 7219–7226.
(10) Elliot, P.; Furche, F.; Burke, K. Rev. Comp. Chem. 2009,
26, 91–165.
(11) (a) Vintonyak, V. V.; Warburg, K.; Kruse, H.; Grimme, S.;
Hubel, K.; Rauh, D.; Waldmann, H. Angew. Chem., Int. Ed. 2010,
49, 5902–5905. (b) Goerigk, L.; Grimme, S. J. Phys. Chem. A 2009,
113, 767–776. (c) Mori, T.; Grimme, S.; Inoue, Y. J. Org. Chem. 2007,
72, 6998–7010. (d) Bringmann, G.; Bruhn, T.; Maksimenka, K.;
Hemberger, Y. Eur. J. Org. Chem. 2009, 2717–2727. (e) Stephens,
P. J.; Pan, J. J.; Devlin, F. J.; Urbanova, M.; Julinek, O.; Hajicek,
J. Chirality 2008, 20, 454–470. (f) Stephens, P. J.; Pan, J. J.; Devlin,
F. J.; Urbanova, M.; Hajicek, J. J. Org. Chem. 2007, 72, 2508–2524.
(12) Alcaide, B.; Sierra, M.; Polanco, C. J. Org. Chem. 1998,
63, 6786–6796.
3319
dx.doi.org/10.1021/jo200171w |J. Org. Chem. 2011, 76, 3306–3319