Tandem [1,5]-H shift/6π-electrocyclizations of ketenimines bearing 1,3-oxathiane units. Computational assessment of the experimental diastereoselection

Article abstract of DOI:10.1016/j.tet.2012.04.021

N-Aryl ketenimines bearing a 1,3-oxathiane function at the ortho position of the keteniminic nitrogen atom convert into spiro[1,3-oxathiane-2, 4′(3′H)quinolines] under mild thermal treatment. These cyclization processes are interpreted in terms of a two-s

Full text of DOI:10.1016/j.tet.2012.04.021

Tetrahedron 68 (2012) 4672e4681
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Tandem [1,5]-H shift/6p-electrocyclizations of ketenimines bearing 1,3-oxathiane
units. Computational assessment of the experimental diastereoselection
,
a,
*
Mateo Alajarin a, Baltasar Bonillo a, Marta Marin-Luna a, Pilar Sanchez-Andrada b , Angel Vidal
,
*
Raul-Angel Orenes c
a
ꢀ
Departamento de Química Organica, Universidad de Murcia, Facultad de Química, Regional Campus of International Excellence “Campus Mare Nostrum”, Espinardo, 30100 Murcia,
Spain
b
ꢀ
ꢀ
~
University Centre of Defence, at the Spanish Air Force Academy, Base Aerea de San Javier, C/Coronel Lopez Pena s/n, 30720, Santiago de la Ribera, Murcia, Spain
c
ꢀ
Servicio Universitario de Instrumentacion Científica, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
N-Aryl ketenimines bearing a 1,3-oxathiane function at the ortho position of the keteniminic nitrogen
atom convert into spiro[1,3-oxathiane-2,40(30H)quinolines] under mild thermal treatment. These cycli-
zation processes are interpreted in terms of a two-step tandem sequence involving a [1,5]-H migration
followed by a 6p-electrocyclic ring closure. Moreover, the cyclization of 1,3-oxathiane-ketenimines
having two different substituents at the terminal carbon atom of the ketenimine moiety provided spi-
roquinolines bearing two stereocenters, the C3 and C4 atoms, with moderate diastereoselectivity. A DFT
Received 7 December 2011
Received in revised form 3 April 2012
Accepted 4 April 2012
Available online 14 April 2012
Keywords:
1,3-Oxathiane
Ketenimine
Hydride migration
electrocyclization
DFT study
study support that the mechanism of these conversions consists of a [1,5]-H shift/6p-electrocyclization
sequence, in which the [1,5]-H shift is the rate-limiting step. A quantitative kinetic analysis of the cy-
clization of an oxathiane-ketenimine with a prochiral ketenimine function explains the sense and degree
of the experimentally observed diastereoselectivity.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
n
n
Y
Cycloaddition reactions, electrocyclizations and sigmatropic rear-
rangements represent important types of pericyclic transformations
of great utility in synthetic organic chemistry.1 Otherwise, tandem or
domino reactions,defined as thecombinationof twoor moredifferent
chemical transformations occurring in quick succession in the same
reaction flask,2 have emerged as synthetic strategies for the efficient
construction of complex carbon and heterocyclic skeletons from
simple precursors.3 Thus, tandem processes composed by several
consecutive pericyclic events have been used to construct complex
organic molecules, including some natural products.4
X
Y
H
R1
X
R1
R3
Ph
H
1,5-H/6π-ERC
X = Y = O, n = 1, 2
X = Y = S, n = 1, 2
N
C
C
N
R2
R2
2
1
Ph
R3
X = O, Y = S, n = 1
1,5-H/1,5 electrocyclization/
cycloreversion [3+2]
Recently, we have reported a new tandem sequence in acetalic
ketenimines 1,5 also applicable to similar heterocumulenes bearing
triarylmethane fragments,6 that involves a [1,5]-H shift,7 with
O
R1
hydride-like characteristics, followed by a 6p-electron electrocyclic
ring closure, en route to quinolines 2 (Scheme 1). The C(sp3)eH
activation step occurring in these ketenimines is facilitated by the
hydricity (hydride donor ability) of the acetalic functions
S
CH2 CH2
+
N
H
R3
Ph
R2
3
Scheme 1. Previously studied tandem processes in acetalic ketenimines.
* Corresponding authors. Tel.: þ34 868 887418; fax: þ34 868 884149 (A.V.); tel.:
þ34 968 189923; fax: þ34 968 189970 (P.S.-A); e-mail addresses: pilar.sanchez@
cud.upct.es (P. Sanchez-Andrada), vidal@um.es (A. Vidal).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.04.021

M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
4673
(1,3-dioxolane, X¼Y¼O, n¼1; 1,3-dithiolane, X¼Y¼S, n¼1; 1,3-
dioxane, X¼Y¼O, n¼2; 1,3-dithiane, X¼Y¼S, n¼2).
S
O
H
R1
R1
CHO
N3
In contrast, the use of a 1,3-oxathiolane ring as the acetalic
functionality in the corresponding oxathiolane-ketenimines 1
(X¼O, Y¼S, n¼1) promoted a different tandem process via a se-
quence of three pericyclic reactions: [1,5]-H shift/1,5-
electrocyclization/[3þ2] cycloreversion with concomitant elimi-
nation of ethylene, thus furnishing 2,1-benzisothiazol-3-ones 3.8
Stimulated by these results we became interested in exploring
the thermal tandem process that could undergo ketenimines of
general structure A (Fig. 1), bearing a 1,3-oxathiane function for
facilitating the CeH activation step.
a
N3
b
5
4a R1 = H
4b R1 = Br
4c R1 = Cl
S
O
H
R1
S
O
H
R1
c
Ph
Ph
N PPh3
N C C
6
7
e
d
S
O
H
S
O
R1
R1
Ph
CH3
Ph
Fig. 1. General structure of 1,3-oxathiane-ketenimines.
N C C
N
H
d
8
9
Ph
Herein we disclose that the thermal treatment of several exam-
ples of such oxathiane-ketenimines A converts these compounds
into3,4-dihydroquinolines, most likelybythe tandem sequence that
involves as the first mechanistic step a 1,5-H migration and as the
S
O
R1
S
O
R1
CH3
Ph
H
Ph
CH3
H
+
second step a 6p-electrocyclization. We also show that because of
N
cis-10
N
the presence of two different heteroatoms, oxygen and sulfur, at the
acetalic function, the cyclization of 1,3-oxathiane-ketenimines A
with R2sPh generates two stereocenters, the C3 and C4 atoms, at
the reaction products 3,4-dihydroquinolines, thus giving rise to the
expected two diastereoisomers with moderate diastereoselectivity.
We also disclose the results ofa DFTstudyshowing the most relevant
mechanistic aspects of these transformations and accounting for the
sense and degree of the experimentally observed stereocontrol.
trans-10
Scheme 2. Reagents and conditions: (a) 3-mercaptopropanol, p-TsOH, benzene, reflux,
3 h; (b) PPh3, Et2O, rt, 12 h; (c) Ph2C]C]O, ortho-xylene, rt, 30 min; (d) ortho-xylene,
reflux, 12 h; (e) PhCH3C]C]O, dichloromethane, rt, 30 min.
Table 1
Spiroquinolines 8 and 10
Compound
R1
Yield (%)
dra
8a
8b
8c
H
73
71
70
40
70
41
2. Results and discussion
Br
Cl
H
Br
Cl
The treatment of 2-azidobenzaldehydes
4
with 3-
cis-10aþtrans-10a
cis-10bþtrans-10b
cis-10cþtrans-10c
28:72
27:73
28:72
mercaptopropanol in the presence of a catalytic amount of para-
toluenesulfonic acid, in benzene solution at reflux temperature
with azeotropical removal of water, and the further Staudinger
reaction of the resulting 2-(2-azidophenyl)-1,3-oxathianes 5 with
triphenylphosphine, in diethyl ether at room temperature, pro-
vided the iminophosphoranes 6. Aza-Wittig reaction of compounds
6 with diphenylketene, in anhydrous ortho-xylene at room tem-
perature, afforded the 1,3-oxathiane-ketenimines 7, whose forma-
tion was verified by recording FT-IR spectra of the final reaction
mixtures, which showed a very strong absorption band around
2000 cmꢀ1. The heating of ortho-xylene solutions of 1,3-oxathiane-
a
As calculated by integration in the 1H NMR spectra of the final reaction mixtures.
purified by column chromatography on a short silica gel column,
and immediately subjected to the thermal treatment to give the
spiroquinolines 10 (Scheme 2, Table 1). Compounds 10 were
obtained as mixtures of two racemic diastereoisomers in relative
ratios close to 7:3, as calculated by integration of the signals cor-
responding to C(20)H and CH3eC(30) protons in the 1H NMR spectra
of the mixtures. Although these mixtures could not be resolved by
ketenimines
7 at reflux temperature promoted their trans-
formation into the 30,30-diphenylspiro[1,3-oxathiane-2,40(30H)
quinolines] 8 in moderate yield (Scheme 2 and Table 1).
column chromatography,
a small amount of the major di-
astereoisomer precipitated in a pure state by addition of diethyl
ether to each one of the mixtures. The minor isomers could not be
isolated in a pure form. The crystallization of the major di-
astereoisomer of compound 10b allowed us to obtain X-ray quality
crystals from dichloromethane/n-hexane. The X-ray structure
revealed that this major isomer of 10b is the one having the sulfur
atom at C(40) and the phenyl group at C(30) in a relative trans dis-
position (Fig. 2). By extension, a relative trans configuration was
assigned to the major diastereoisomers of spiroquinolines 10a and
10c, as the comparison of their 1H and 13C NMR data with those of
trans-10b afforded congruent results. Relevant NMR data for all
spiroquinolines cis-10aec and trans-10aec are given in Table 2.
The structural determination of the spiroquinolines 8 was car-
ried out following their spectral data: IR, 1H and 13C NMR and mass
spectrometry. In their 1H NMR spectra the C(20)H proton resonates
as a singlet at
d
¼8.54e8.61 ppm. Relevant 13C NMR data of these
spiroheterocycles are the chemical shifts of the aliphatic quater-
nary carbons C(30) at
d
¼59.3e59.7 ppm and C(40) at
d
¼85.4e85.9 ppm, and that of the C(20) methine carbon at
d¼166.1e167.2 ppm.
The aza-Wittig reaction of iminophosphoranes 6 with methyl-
phenylketene, in dichloromethane solution at room temperature,
furnished the C-methyl-C-phenyl ketenimines 9,9 which were

4674
M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
3. Computational study
We have carried out a computational study at the B3LYP/6-
31þG** theoretical level of the transformations of 1,3-oxathiane-
ketenimines into the corresponding spiroquinolines with the aim
of evaluating the ability of the 1,3-oxathiane functionality for pro-
moting the proposed tandem [1,5]-H shift/6p-ERC sequence, and to
compare it with those of the previously investigated 1,3-dioxane
and 1,3-dithiane functions. We also aimed to find out if the calcu-
lations reasonably explain the sense and degree of the stereo-
control operating in the conversions of oxathiane-ketenimines
9aec into spiroquinolines 10aec.
For these tasks we have selected the reaction sequences shown
in Scheme 3 as model transformations, involving the simple C,C-
unsubstituted oxathiane-ketenimine 11a and that substituted by
one methyl and one phenyl groups at the terminal carbon atom of
the ketenimine function (designed as 11b in this computational
study, 9a in the experiments).
Fig. 2. ORTEP representation of the crystal structure of trans-10b.
As expected, an intensive exploration of the potential energy
surface associated to these transformations showed that the two
reaction paths consist of an initial [1,5]-H shift, via transition
structures TS1a,b, leading to o-azaxylylene intermediates INTa,b.
Table 2
1H NMR (
d
, ppm) and 13C NMR (
d
, ppm) data of the spiroquinolines cis-10a and trans-10
cis-10a trans-10a cis-10b trans-10b cis-10c trans-10c
These intermediates experience a further 6
p electrocyclization
C(20)H
7.83
1.85
8.14
1.40
7.86
1.81
8.15
1.40
7.84
1.78
8.12
1.43
CH3eC(30)
C(20)
(6 -ERC) as the second mechanistic step driving to the final spi-
p
roquinolines 12a,b. The computed energy barriers are gathered in
Tables 3 and 4. We will comment only the values of the electronic
energy barriers, otherwise stated.10 Interestingly, even in the sim-
plest case, structure 11a, the first mechanistic step, the [1,5]-H shift,
may give rise to two different azaxylylene diastereoisomers, INTa
and INT0a, respectively, cis and trans at the newly formed C5eC6
double bond, as depicted in Scheme 4. Accordingly, we have located
two transition structures TS1a and TS10a connecting the oxathiane-
ketenimine 11a with these diastereoisomeric intermediates. Both
intermediates cyclize to spiroquinoline 12a through the respective
transition states TS2a and TS20a (Scheme 4). This second step re-
quires an initial s-trans to s-cis conformational equilibration around
the C2eN3 single bond, for placing the two extremes of the con-
jugated 3-azahexatriene fragment in the appropriate vicinity for
allowing the electrocyclization to occur. However, in the calcula-
tions we could not locate the putative s-cis conformers as stationary
points in the potential energy surfaces. Instead, the reaction co-
ordinate directly connects both intermediates with 12a via the cited
transition structures. Nevertheless, TS2a and TS20a are easily
identified as transition structures corresponding to disrotatory
169.0
171.1
169.5
171.7
169.4
171.3
C(30)
C(40)
51.4
86.0
18.4
26.4
23.9
62.4
52.1
87.0
16.8
24.9
23.9
60.6
51.4
85.4
18.1
26.4
23.9
62.6
51.5
86.5
16.8
24.6
23.7
60.7
51.6
85.9
18.2
25.8
23.7
62.6
52.1
87.0
17.1
24.5
23.8
60.7
CH3-(C30)
C(4)
C(5)
C(6)
a
Obtained from the mixture cis-10þtrans-10 after the purification step.
Next, we investigated if the experimental ratio of cis and trans
spiroquinolines 10 is the result of kinetic or thermodynamic control
of the reactions. In other words, if thermal equilibrium between cis
and trans 10 can be established in solution. With this aim, a sample
of 10c enriched in the cis isomer (66:34 cis/trans) was subjected to
thermal treatment at 140 ꢁC for 24 h, and the composition of that
mixture remained unchanged. Consequently, the 3:7 cis/trans ratio
resulting in the experimental reactions must be a consequence of
the kinetic control of these processes.
The conversions 7/8 and 9/10 are reasonably viewed as
tandem processes consisting of the hydride-like [1,5]-H shift of the
acetalic H atom, followed by the 6p-electrocyclization of the
electrocyclization modes, in accord with the selection rules for 6
p-
electrocyclic reactions.
resulting ortho-azaxylylene intermediates (see the mechanistic
sequence in Scheme 3).
Table 3
Energy barriersa in kcal/mol for the conversions of acetal-ketenimines into spi-
roquinolines calculated at the B3LYP/6-31þG**
þD
ZPVE theoretical level
Entry
1
Acetal ring
D
E1s E2s
D
DErxn
11a/INTa/12a
(1,3-Oxathiane)
11a/INT0a/12a
(1,3-Oxathiane)
1,3-Dioxane
35.1
17.1
ꢀ19.4
2
37.0
21.5
O
O
S
H
S
H
3b
4c
34.4
36.2
19.0
18.2
ꢀ22.3
ꢀ18.9
TS2a,b
TS1a,b
1,3-Dithiane
R2
[1,5]-H
6 -ERC
π
N
C
C
N
a
See Scheme 4 for the notation of the energy barriers.
Conversion of 2-(1,3-dioxan-2-yl)-N-phenylketenimine into spiro[1,3-dioxane-
b
R3
INTa,b
2,40(30H)quinoline].
c
Conversion of 2-(1,3-dithian-2-yl)-N-phenyl-ketenimine into spiro[1,3-
R3
dithiane-2,40(30H)quinoline].
R2
O
S
R2
R3
H
11a, R2 = R3 = H
The computed values for the energy barriers of both two-step
processes represented in Scheme 4 are summarized in Table 3.
Also indicated in that table are the barriers of similar processes
involving the 1,3-dioxane and 1,3-dithiane acetalic functions, as
previously calculated by us.5b As presumed, the differences
11b, R2 = Me, R3 = Ph
N
12a,b
Scheme 3. The two reaction steps analyzed in the computational study.

M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
4675
Table 4
alkyl and S-alkyl, as strong donors groups, must exhibit outward
rotational preference. The eOR fragment of 11a most probably
possesses stronger donor character than the eSR one, as sug-
gested by the respective soR substituent constants of the eOMe
and eSMe groups, with values of ꢀ0.41 and ꢀ0.17, respectively.12
This assumption would serve to explain why TS1a, with oxygen
placed outward, is lower in energy than TS10a, which features an
oxygen atom oriented inward.
In relation to the second step transforming the o-azaxylylene
intermediates INTa and INT0a into spiroquinoline 12a, as we men-
tioned above TS2a and TS20a have been identified as corresponding
to disrotatory electrocyclizations by means of IRC (intrinsic reaction
coordinate) calculations and the animation of their first frequencies.
Energy barriersa in kcal/mol for the conversions of 1,3-oxathiano-ketenimines 11a,b
into spiroquinolines 12a,b calculated at the B3LYP/6-31þG**
þDZPVE theoretical
level
Entry
11/12
D
E1s
D
E2s
DErxn
1
2
3
4
5
6
11a/INTa/12a
35.1
37.0
27.9
28.7
26.1
29.8
17.1
21.5
21.6
26.5
24.5
22.9
ꢀ19.4
ꢀ7.7
ꢀ6.3
11a/INT0a/12a
11b/Z-INTb/cis-12b
11b/E-INT0b/cis-12b
11b/E-INTb/trans-12b
11b/Z-INT0b/trans-12b
a
See Scheme 4 for the notation of the energy barriers.
This step, the 6p-electrocyclic ring closure, exhibits a clear prefer-
encefor theout-O, in-Srotation(TS2a)versustheout-S, in-O rotation
(TS20a), TS2a being 3.1 kcal/mol lower in energy than TS20a. How-
ever, as a result of the different relative energies of intermediates
INTa and INT0a, the difference between the E2s energy barriers is
D
higher, 4.4 kcal/mol (compare entries 1 and 2 in Table 3). Overall,
again the beneficial electronic influence of placing O-out, S-in (TS2a)
when compared with O-in, S-out (TS20a) prevails over the higher
steric requirements of the sulfur versus the oxygen atoms, the
stereoelectronic control being higher in these
cyclizations than in the precedent [1,5]-H shifts.
6p-electro-
Moreover, as a reviewer pointed out, the geminal bond partici-
pation theory13 (the participation of germinal bonds in some re-
actions by controlling their selectivity via the sigma-bond donating
effect) may also contribute to explain the observed stereocontrol,
synergistically with our interpretation above. Following this theory,
and taking into consideration that generally a CeS bond is more
donating than a CeO bond because oxygen is more electronegative
than sulfur, the preference for the inward position in the 1,5-H
migration, as well in the electrocyclization step, is higher for the
CeS bond.
As far as the geometries of the transition structures TS2a and
TS20a are concerned, we noted their considerable deviations from
the boatlike conformation shown by the TS of the prototypical
electrocyclization of hexatriene to cyclohexadiene14 (TS-6
p) as
calculated at the same level of theory (see Fig. 3d). As a matter of
fact, the helical geometries of transition structures TS2a and TS20a
were already present in the respective o-azaxylylene intermediates
INTa and INT0a, as result of the steric interferences between the
oxathiane ring at C6 and the N-vinyl group at the other extreme of
the 3-azahexatriene fragment.
Once we have commented the most relevant aspects of the
transformation 11a/12a, we next will deal with the conversion of
the more substituted oxathiane-ketenimine 11b. The [1,5]-H step in
11b causes the generation of two stereogenic double bonds in the
corresponding o-azaxylylene intermediates, which are now the
four different ones Z-INTb, Z-INT0b, E-INTb, E-INT0b depicted in
Scheme 5. Obviously, the whole tandem two-step process is now
potentially diastereoselective, as it can lead either to cis or trans
spiroquinolines 12b.
In the potential energy surface of this process we could locate the
expected four transition states, Z-TS1b, Z-TS10b, E-TS1b, E-TS10b
connecting 11b with the respective INT intermediates. The two el-
ements of each pair of intermediates Z-INTb/E-INTb and Z-INT0b/E-
INT0b differ only in the configuration of the C5eC6 double bond,
whereas those of the pairs Z-INTb/Z-INT0b and E-INTb/E-INT0b differ
on the configuration around the C1eC2 double bond (see atomic
numbering in Scheme 5). Intermediates Z-INTb and E-INT0b con-
nect, through transition states Z-TS2b and E-TS20b, respectively,
with the cis-12b spiroquinoline, whereas intermediates Z-INT0b and
E-INTb connect, through transition states Z-TS20b and E-TS2b, re-
spectively, with the trans-12b spiroquinoline, slightly higher in en-
ergy than cis-12b (see Scheme 5, as well as Tables 4 and 5 and Fig. 4).
Scheme 4. Computed pathways for the conversion of oxathiane-ketenimine 11a into
spiroquinoline 12a.
between the four
D
E1s barriers in Table 3 are not high. In all cases,
the first reaction step of the sequence, the 1,5-H shift, is predicted
to be the rate-limiting one ð
D
E1s
[
D
E2sÞ.
Interestingly, there is a difference close to 2 kcal/mol between
TS1a and TS10a, the first one being more stable. This difference can
be only attributed to the alternative relative positions of the oxygen
and sulfur atoms. In TS1a the O and S atoms are placed outward
(out) and inward (in), respectively, whereas TS10a shows in-O and
out-S. If any, less steric interference between the in-heteroatom and
the ketenimine fragment is a priori conceivable in the in-O, out-S
structure (TS10a) than in the out-O, in-S one (TS1a), taking into
account the respective atomic radii of both heteroatoms. In fact, the
H12eC6 and C2eC6 bond distances are slightly longer in TS1a than
in TS10a (see Fig. 3).
Most probably some electronic effects should be more in-
fluential than the steric ones in order to account for the higher
energy computed for TS10a in comparison with TS1a. In accor-
dance with the well-studied electronic effects of the substituents
in the stereocontrol of the conrotatory ring opening of cyclo-
butenes,11 in the present [1,5]-H shift mechanistic step both O-

4676
M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
rotating away (outward) from the adjacent, breaking sigma CeH
bond. For the same configuration around the C5eC6 bond, those
transition states showing E configuration at the C1eC2 double bond
are lower in energy than their Z-isomers (compare entry 10 with
entry 13 and entry 16 with entry 19). Accordingly, these transition
structures can be ordered by increasing energies as follows:
E-TS1b<Z-TS1b<E-TS10b<Z-TS10b.
In contrast, a similar order of the energies of the INTb o-azax-
ylylene intermediates, in which only the steric interferences are
relevant for their differentiation, gives rise to the following result:
E-INT0b<E-INTb<Z-INT0b<Z-INTb.
In relation with the second step, the 6p-electrocyclization of the
different INTb to cis and trans-12b via the four diastereoisomeric
TS2b transition structures, it is worth pointing out the great dis-
tortion of the geometry of these later stationary points due to the
steric interferences between the inward substituents at the ex-
tremes of the forming new CeC bond. The animation of the
imaginary frequency computed for each transition structure cor-
responds to the formation of the C1eC6 bond accompanied by the
disrotatory rotation around the C5eC6 and C1eC2 bonds. Note that,
as commented above, the C2eN3 bond also rotates for allowing the
required s-trans to s-cis conformational equilibration.
The following remarks should be noted by comparing transition
structures TS2b with those of the previously discussed TS2a
structures, and also with the prototypical TS-6
p
(see Table 6):
ꢁ
(1) the larger C1eC6 bond distances (2.85e3.08 A), as well as the
increased interaction distances between the two inward sub-
ꢁ
stituents at C1 and C6 (2.99e3.67 A).
(2) the marked distortions of the
XouteC1eC6, YineC1eC6,
ZouteC6eC1 and VineC6eC1 bond angles following the in-
troduction of the phenyl and methyl substituents at C1.
(3) the notable twist of the forming six-membered ring, as
reflected in the XouteC1eC6eZout and YineC1eC6eVin dihedral
angles (see also several views of the four transition structures
TS2b in Supplementary data).
Fig. 3. Geometries of transition structures found in the conversion of 11a into spi-
roquinoline 12a, as well that corresponding to the electrocyclization of hexatriene into
As summarized inTable 5, the four transition states TS2b are very
close in energy. The energies of those placing inward the phenyl ring
and outward the methyl group (Z-TS20b and Z-TS2b) are slightly
lower than those with in-Me and out-Ph (E-TS20b and E-TS2b). In
contrast, the energy differences between structures differing only in
the out or in disposition of the O and S substituents at C6 are smaller.
These data seem to indicate the prevalence of the steric effects of the
substituents at the extremes of the electrocyclization over the
cyclohexadiene (TS-6
relevant geometrical parameters are shown.
p
), optimized at the B3LYP/6-31þG** theoretical level. Some
The geometries of transition structures Z-TS1b, Z-TS10b, E-TS1b, E-
TS10b are similar to those found for TS1a and TS10a, showing the
suprafacial transfer of the hydrogen atom. The
D
E1s energy barriers of
these [1,5]-H shifts (entries 3e6 inTable 4) are lower than those of the
two simplerexamples previouslyanalyzed and represented in entries
1 and 2 of the same table. Thus the substitution of hydrogen atoms at
the terminalcarbonatomof theketeniminemoietybythemethyl and
phenyl groups facilitates the H shift. This facilitation can be ratio-
nalized as due to the decrease of the HOMOeLUMO energy gap
resulting from the extension of the conjugation caused by the new
phenyl group at the terminal C atom of the heterocumulenic frag-
ment.5b Most probably, this phenyl group stabilizes not only the TS1b
transition structures but also the resulting INTb intermediates.15
The steric effects caused by the substituents at the two extremes
between, which the H shift is taking place complicate the analysis
of the relative energies of the four TS1b transition structures, in
comparison with that we reasoned above in the simpler case in-
volving the two TS1a ones. First, the transition structures with out-
O, in-S (Z-TS1b, E-TS1b, entries 10 and 13 in Table 5) are lower in
energy than those with out-S, in-O (Z-TS10b, E-TS10b, entries 16 and
19). Again, the stronger donor O-alkyl group shows preference for
electronic ones in these 6p-electrocyclic processes.
In summary, the values of the energy barriers in Table 5 show
the energetic preference for the conversions of intermediates Z-
INT0b and Z-INTb through transition structures Z-TS20b and Z-TS2b,
when compared with those of E-INT0b and E-INTb through E-TS20b
and E-TS2b.
Once we have computed the values of the different energy
barriers, we next approached their utilization for explaining the
sense and degree of the stereocontrol operating in the experi-
mental work, more specifically in the conversions of oxathiane-
ketenimines 9 into spiroquinolines 10. It is not easy to find an ex-
planation for the experimentally observed 7:3 trans/cis ratio by
simply looking at the values of the energy barriers (see Table 5 and
the qualitative reaction profiles at Fig. 4). Consequently, and taking
into account our experimental results indicating that the obtained
diastereoisomeric ratio of compounds 10 is the result of the kinetic
control of the reactions (see above), we decided to carry out

M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
4677
Scheme 5. Mechanistic path found for the conversion of oxathiane-ketenimine 11b into spiroquinolines cis-12b and trans-12b.
a quantitative kinetic analysis with the aim of predicting the ratio of
spiroquinolines by using the energy barriers resulting from the
computations. To this end we considered the mechanistic scheme
represented below, and solved the system of differential equations
outlined in Eqs. 1e9. For simplicity, the steady-state approxima-
tion16 was applied to intermediates (Z-INTb, E-INTb, Z-INT0b and E-
INT0b). Besides, the reversibility of the reaction formation of spi-
roquinolines from intermediates has been discarded due to the low
values of the corresponding rate constants when compared with
those of the other processes. The solution for the ratio cis-12b/
trans-12b is outlined in Eq. 10 (see Supplementary data), which
results to be independent of the reaction time and to be a combi-
nation of the different rate constants (ki) implied in the mechanistic
paths. The ki values were calculated according to Eq. 11. For this task
Table 5
Relative electronic (E) and free energies (G) in kcal/mol computed at the B3LYP/6-
31þG**
þ
DZPVE theoretical level of the stationary points found in the conversions of
1,3-oxathiane-ketenimines 11a,b into spiroquinolines 12a,b
Entry
E
G
G-410a
1
2
3
4
5
6
7
8
11a
0.0
35.1
18.9
35.9
37.0
17.6
39.0
0.00
36.8
TS1a
INTa
TS2a
TS10a
INT0a
TS20a
12a
19.4
37.7
38.5
18.3
40.3
we previously computed the
DGi values at 410 K with the aim of
simulating more precisely the reaction conditions of the experi-
ments. Thus, the calculated value of the ratio cis-12b/trans-12b is
0.49, which corresponds to 33% of cis-12b, and 67% of trans-12b.
These values approximate quite well to those obtained experi-
mentally for the mixtures of cis and trans-10.
ꢀ19.4
ꢀ16.9
9
11b
0.0
27.9
11.9
33.5
26.1
9.5
34.0
29.8
10.5
33.4
28.7
8.3
0.0
31.1
13.9
37.2
29.3
11.4
37.6
32.8
12.6
36.7
32.0
10.5
38.1
ꢀ3.1
ꢀ1.8
0.0
32.6
14.7
38.8
30.7
12.1
39.3
34.2
13.5
38.2
33.5
11.4
39.7
ꢀ0.9
0.4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Z-TS1b
Z-INTb
Z-TS2b
E-TS1b
E-INTb
E-TS2b
Z-TS10b
Z-INT0b
Z-TS20b
E-TS10b
E-INT0b
E-TS20b
cis-12b
trans-12b
34.7
ꢀ7.7
ꢀ6.3
a
Free energies with thermal corrections computed at 410 K.

4678
M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
Table 6
Some relevant geometrical parameters measured in transition structures TS2a, TS2a0, Z-TS2b, E-TS2b, Z-TS20b, E-TS20b and TS-6
p
optimized at the B3LYP/6-31þG**
þDZPVE
theoretical levela
4
3
N
5
Z
X
2
out
out
6
1
Y
V
in
in
Entry
dC1eC6
dYineVin
X
outeC1eC6
Y
ineC1eC6
Z
outeC6eC1
V
ineC6eC1
X
outeC1eC6eZout
YineC1eC6eVin
1
TS2a
2.61
2.44
123.4
121.5
111.2
113.8
114.9
114.7
116.9
70.1
69.5
95.6
89.3
93.0
93.4
79.2
113.1
113.1
101.9
111.9
113.1
102.9
116.9
91.9
86.5
98.2
78.4
88.9
98.3
79.2
21.8
16.4
66.9
45.3
40.1
60.3
0.0
10.4
25.5
64.6
44.0
40.9
55.1
0.0
Xout¼H, Yin¼H, Zout¼O, Vin¼S
2
3
4
5
6
7
TS20a
2.62
2.85
3.08
2.97
2.87
2.27
2.22
Xout¼H, Yin¼H, Zout¼S, Vin¼O
Z-TS2b
3.67 (PheS)
Xout¼Me, Yin¼Ph, Zout¼O, Vin¼S
E-TS20b
2.99 (MeeO)
3.23 (MeeS)
3.41 (PheO)
1.88
Xout¼Ph, Yin¼Me, Zout¼S, Vin¼O
E-TS2b
Xout¼Ph, Yin¼Me, Zout¼O, Vin¼S
Z-TS20b
Xout¼Me, Yin¼Ph, Zout¼S, Vin¼O
TS-6
p
Xout¼H, Yin¼Y, Zout¼H, Vin¼H
a
See Schemes 4 and 5 and Fig. 3 for atomic numeration.
d½Z ꢀ INT0bꢂ
dt
d½11bꢂ
¼ ka3½11bꢂ ꢀ ðkb3 þ kc3Þ½Z ꢀ INT0bꢂ ¼ 0
¼ ka4½11bꢂ ꢀ ðkb4 þ kc4Þ½E ꢀ INT0bꢂ ¼ 0
¼ kb1½Z ꢀ INTbꢂ þ kb4½E ꢀ INT0bꢂ
(4)
(5)
(6)
(7)
¼ ꢀðka1 þ ka2 þ ka3 þ ka4Þ½11bꢂ þ kc1½Z ꢀ INTbꢂ
dt
þ kc2½E ꢀ INTbꢂ þ kc3½Z ꢀ INT0bꢂ þ kc4½E ꢀ INT0bꢂ
(1)
(2)
d½E ꢀ INT0bꢂ
dt
d½Z ꢀ INTbꢂ
dt
¼ ka1½11bꢂ ꢀ ðkb1 þ kc1Þ½Z ꢀ INTbꢂ ¼ 0
¼ ka2½11bꢂ ꢀ ðkb2 þ kc2Þ½E ꢀ INTbꢂ ¼ 0
d½cis ꢀ 12bꢂ
dt
d½E ꢀ INTbꢂ
dt
(3)
d½trans ꢀ 12bꢂ
dt
¼ kb2½E ꢀ INTbꢂ þ kb3½Z ꢀ INT0bꢂ
½11bꢂ0 ¼ ½11bꢂ þ ½Z ꢀ INTbꢂ þ ½E ꢀ INTbꢂ þ ½Z ꢀ INT0bꢂ
þ ½E ꢀ INT0bꢂ þ ½cis ꢀ 12bꢂ þ ½trans ꢀ 12bꢂ
B3LYP/6-31+G** Energy (kcal/mol)
35
(8)
(9)
E-TS2'b
34.7
34.0
E-TS2b
Z-TS2b 33.5
33.4
29.8
Z-TS1'b
Z-TS2'b
30
E-TS1'b 28.7
½Z ꢀ INTbꢂ0 ¼ ½Z ꢀ INTbꢂ0 ¼ ½E ꢀ INTbꢂ0 ¼ ½Z ꢀ INT0bꢂ
¼ ½E ꢀ INT0bꢂ0 ¼ 0
Z-TS1b 27.9
25
20
15
26.1
E-TS1b
8
9
Z-INTb 11.9
Z-INT'b
ꢀ
ꢀ
ꢁ
ꢁ
>
>
>
>
>
>
k
a1$kb1
k
a4$kb4
10.5
>
>
<
>
>
=
þ
þ
10
5
E-INTb 9.5
cis ꢀ 12b
ðkb1 þ kc1
Þ
Þ
ðkb4 þ kc4
Þ
Þ
8.3
E-INT'b
¼
(10)
(11)
>
>
trans ꢀ 12b
ka2$kb2
ka3$kb3
>
>
>
>
>
>
:
>
>
;
ðkb2 þ kc2
ðkb3 þ kc3
0
11b
trans-12b
–6.3
ꢀ
ꢁ
–5
–10
D
Gi
ꢀ
KB$T
RT
ki ¼
$e
cis-12b –7.7
Reaction Coordinate
h
4. Conclusion
Fig. 4. Qualitative reaction profile computed for the transformation of 1,3-oxathiano-
In summary, we have disclosed that 1,3-oxathiane-ketenimines,
under thermal conditions, undergo cyclization to spiro[1,3-
ketenimine 11b into spiroquinolines 12b calculated at the B3LYP/6-31þG**
þ
DZPVE
theoretical level.

M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
4679
oxathiane-2,40(30H)quinolines] in moderate yields. We have also
described that oxathiane-ketenimines bearing a prostereogenic
terminal carbon atom at the ketenimine fragment convert into the
corresponding spiroquinolines with a moderate degree of stereo-
control, obtaining the reaction products with trans/cis di-
astereomeric ratios close to 7:3. The computational calculations
carried out in this work have established the mechanism of these
cyclizations as a sequence of two pericyclic events: a [1,5]-H mi-
(300 MHz, CDCl3) 7.63 (1H, s), 7.27 (1H, d, J 8.4 Hz), 7.01 (1H, d, J
8.4 Hz), 4.28 (1H, d, J 10.2 Hz), 3.77 (1H, t, J 12.3 Hz), 3.22 (1H, t, J
12.9 Hz), 2.80 (1H, d, J 13.2 Hz), 2.11e2.03 (1H, m), 1.75 (1H, d, J
13.5 Hz); dC (75 MHz, CDCl3) 134.8 (s), 132.5 (s), 130.8 (s), 129.6,
128.8, 119.2, 78.2, 70.8, 29.2, 25.7; HRMS (ESI): MHþꢀN2, found
228.0245. C10H11ClNOS requires 228.0244.
5.3. Preparation of 2-(2-triphenylphosphoranylideneaminoph-
enyl)-1,3-oxathianes 6
gration followed by a 6p-electrocyclization. The quantitative ki-
netic analysis for estimating the relative ratio of diastereomers in
the cyclization of oxathiane-ketenimines with a prostereogenic
terminal carbon atom at the ketenimine fragment resulted in good
agreement with the experimental results.
Triphenylphosphine (1.05 g, 4 mmol) was added in portions of
0.21 g every 2 min to a solution of the corresponding 2-(2-
azidophenyl)-1,3-oxathiane
5 (4 mmol) in anhydrous diethyl
ether (25 mL). The reaction mixture was stirred at room tempera-
ture under nitrogen for 12 h. The precipitated compound 6 was
isolated by filtration and washed with anhydrous diethyl ether
(10 mL). Compounds 6 were used in the following step without
further purification. For analytical samples, iminophosphoranes 6
were recrystallized from diethyl ether.
5. Experimental section
5.1. General methods
All melting points are uncorrected. Infrared (IR) spectra were
recorded as liquid film or Nujol emulsions. 1H NMR spectra were
recorded in CDCl3 or DMSO-d6 at 300 or 400 MHz. 13C NMR spectra
were recorded in CDCl3 or DMSO-d6 at 75 or 100 MHz. The chemical
shifts are expressed in parts per million, relative to Me4Si at
5.3.1. 2-(2-Triphenylphosphoranylideneaminophenyl)-1,3-oxathiane
(6a). Yield 89% (1.62 g); mp 141e142 ꢁC (colourless prisms from
diethyl ether); nmax (Nujol) 1454, 1108 cmꢀ1
; dH (300 MHz, CDCl3)
d
¼0.00 ppm for 1H, while the chemical shifts for 13C are reported
7.81e7.75 (6H, m), 7.52e7.40 (10H, m), 6.80 (1H, t, J 7.5 Hz), 6.68
(1H, t, J 7.4 Hz), 6.66 (1H, s), 6.39 (1H, d, J 7.8 Hz), 4.33 (1H, dd, J 10.0,
1.8 Hz), 3.84 (1H, td, J 12.3, 1.4 Hz), 3.24 (1H, td, J 12.9, 2.5 Hz), 2.83
(1H, d, J 13.1 Hz), 2.15 (1H, qt, J 12.9, 3.9 Hz), 1.75 (1H, d, J 13.6 Hz);
dC (75 MHz, CDCl3) 147.3 (s), 133.3 (d, J 21.1 Hz) (s), 132.6 (d, J
9.7 Hz), 131.6 (d, J 2.8 Hz), 131.5 (d, J 100.0 Hz) (s), 128.6 (d, J
12.0 Hz), 128.2, 127.0 (d, J 1.8 Hz), 121.2 (d, J 9.7 Hz), 117.7, 82.1, 71.1,
29.6, 26.4; dP (121.4 MHz, CDCl3, H3PO4) 1.9; HRMS (ESI): MHþ,
found 456.1559. C28H27NOPS requires 456.1545.
relative to the resonance of CDCl3
d¼77.1 ppm or DMSO-d6
d
¼39.5 ppm.
2-Azidobenzaldehyde 4a,17 2-azido-5-bromobenzaldehyde 4b18
and 2-azido-5-chlorobenzaldehyde 4c,18 diphenylketene19 and
methylphenylketene20 were prepared following published experi-
mental procedures.
5.2. Preparation of 2-(2-azidophenyl)-1,3-oxathianes 5
3-Mercaptopropanol (0.93 g, 10 mmol) and p-toluenesulfonic
acid (0.1 g) were added to a solution of the 2-azidobenzaldehyde 4
(10 mmol) in benzene (50 mL). The reaction mixture was heated at
reflux temperature for 3 h, with azeotropic removal of water with
a DeaneStark trap. After the reaction mixture was cooled at room
temperature, diethyl ether (50 mL) was added and the resulting
mixture was washed with 0.1 M KOH solution (50 mL) and brine
(50 mL). The organic phase was dried with anhydrous magnesium
sulfate. The solvent was evaporated in vacuo and the resulting
material was purified by column chromatography on silica gel, with
hexanes/diethyl ether (9:1, v/v) as eluent.
5.3.2. 2-(5-Bromo-2-triphenylphosphoranylideneaminophenyl)-1,3-
oxathiane (6b). Yield 77% (1.65 g); mp 102e103 ꢁC (colourless
prisms from diethyl ether); nmax (Nujol) 1581, 1436, 1359,
1118 cmꢀ1
; dH (300 MHz, CDCl3) 7.80e7.72 (6H, m), 7.60 (1H, t, J
2.5 Hz), 7.55e7.41 (9H, m), 6.86 (1H, dd, J 8.5, 2.6 Hz), 6.56 (1H, m),
6.22 (1H, dd, J 8.5, 1.3 Hz), 4.36e4.30 (1H, m), 3.83 (1H, td, J 12.3,
1.9 Hz), 3.22 (1H, td, J 12.9, 2.8 Hz), 2.85e2.80 (1H, m), 2.14 (1H, qt, J
12.7, 3.9 Hz), 1.78e1.72 (1H, m); dC (75 MHz, CDCl3) 146.6 (s), 135.3
(d, J 21.5 Hz) (s), 132.6 (d, J 9.7 Hz), 131.8 (d, J 2.7 Hz), 130.9 (d, J
99.6 Hz) (s), 130.8, 129.8 (d, J 1.8 Hz), 128.7 (d, J 11.9 Hz), 122.4 (d, J
9.7 Hz), 109.6 (s), 81.5, 71.0, 29.5, 26.3; dP (121.4 MHz, CDCl3, H3PO4)
3.1; HRMS (ESI): MHþ, found 534.0652. C28H26BrNOPS requires
534.0651.
5.2.1. 2-(2-Azidophenyl)-1,3-oxathiane (5a). Yield 88% (1.95 g);
colourless oil; nmax (liquid film) 2127, 2094 cmꢀ1
; dH (400 MHz,
CDCl3) 7.65 (1H, dd, J 7.7, 1.6 Hz), 7.33 (1H, td, J 7.7, 1.6 Hz), 7.18e7.14
(1H, m), 7.12e7.10 (1H, m), 5.99 (1H, s), 4.32e4.27 (1H, m), 3.79 (1H,
td, J 12.3, 2.0 Hz), 3.23 (1H, td, J 13.0, 2.8 Hz), 2.83e2.78 (1H, m),
2.15e2.03 (1H, m), 1.74 (1H, dq, J 13.0, 1.4 Hz); dC (100 MHz, CDCl3)
135.9 (s), 130.5 (s), 129.7, 128.3, 125.1, 117.8, 78.9, 70.8, 29.2, 25.7;
HRMS (ESI): MHþꢀN2, found 194.0636. C10H12NOS requires
194.0634.
5.3.3. 2-(5-Chloro-2-triphenylphosphoranylideneaminophenyl)-1,3-
oxathiane (6c). Yield 70% (1.37 g); mp 108e111 ꢁC (colourless
prisms from diethyl ether); nmax (Nujol) 1438, 1356, 1188 cmꢀ1
; dH
(300 MHz, CDCl3) 7.80e7.74 (6H, m), 7.54e7.43 (10H, m), 6.74 (1H,
dd, J 8.7, 2.7 Hz), 6.57 (1H, s), 6.29 (1H, d, J 8.4 Hz), 4.35 (1H, d, J
9.9 Hz), 3.85 (1H, t, J 12.3 Hz), 3.24 (1H, td, J 12.9, 2.4 Hz), 2.85 (1H,
d, J 13.2 Hz), 2.22e2.09 (1H, m), 1.77 (1H, d, J 13.8 Hz); dC (75 MHz,
CDCl3) 146.0 (s), 134.7 (d, J 21.6 Hz) (s), 132.6 (d, J 9.6 Hz), 131.8 (d, J
2.4 Hz), 131.0 (d, J 98.0 Hz) (s), 128.7 (d, J 12.0 Hz), 127.9, 127.0, 122.3
(s), 121.9 (d, J 9.8 Hz), 81.5, 71.0, 29.5, 26.3; dP (121.4 MHz, CDCl3,
H3PO4) 0.1; HRMS (ESI): MHþ, found 490.1157. C28H26ClNOPS re-
quires 490.1146.
5.2.2. 2-(2-Azido-5-bromophenyl)-1,3-oxathiane (5b). Yield 99%
(2.97 g); colourless oil; nmax (liquid film) 2127, 2094 cmꢀ1
; dH
(400 MHz, CDCl3) 7.77 (1H, d, J 1.8 Hz), 7.43e7.41 (1H, m),
6.98e6.95 (1H, m), 5.90 (1H, s), 4.30 (1H, d, J 11.9 Hz), 3.77 (1H, t, J
12.3 Hz), 3.22 (1H, t, J 13.0 Hz), 2.81 (1H, d, J 13.5 Hz), 2.14e2.03
(1H, m), 1.76 (1H, d, J 13.9 Hz); dC (100 MHz, CDCl3) 135.1 (s), 132.5,
132.4 (s), 131.4, 119.4, 118.2 (s), 78.3, 70.7, 29.2, 25.6; HRMS (ESI):
MHþꢀN2, found 271.9742. C10H11BrNOS requires 271.9739.
5.4. Preparation of spiro[1,3-oxathiane-2,40(30H)-quinolines] 8
Diphenylketene (0.19 g, 1 mmol) in anhydrous ortho-xylene
(5 mL) was added to
a
solution of the 2-(2-triphenyl-
5.2.3. 2-(2-Azido-5-chlorophenyl)-1,3-oxathiane (5c). Yield 92%
phosphoranylideneaminophenyl)-1,3-oxathiane 6 (1 mmol) in the
same solvent (15 mL). The reaction mixture was stirred at room
(2.35 g); colourless oil; nmax (liquid film) 2129, 2097 cmꢀ1
;
dH

4680
M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
temperature for 30 min and next heated at reflux temperature for
12 h. After the reaction mixture has cooled, the solvent was removed
under reduced pressure. The resulting material was purified by col-
umn chromatography on silica gel.
60.6, 52.1 (s), 24.9, 23.9, 16.8; HRMS (ESI): MHþ, found 310.1261.
C19H20NOS requires 310.1260.
5.5.3. cis-10bþtrans-10b. Yield 70% (0.27 g).
5.4.1. 30,30-Diphenylspiro[1,3-oxathiane-2,40(30H)quinoline]
(8a). Eluent for column chromatography: hexanes/diethyl ether (9:1,
v/v); yield 73% (0.27 g); mp 158e159 ꢁC (colourless prisms from
5.5.4. trans-60-Bromo-30-methyl-30-phenylspiro[1,3-oxathiane-
2,40(30H)quinoline] (10b). Mp 141e142 ꢁC (colourless prisms from
diethyl ether); nmax (Nujol) 1621, 1076 cmꢀ1
; dH (300 MHz, CDCl3)
diethyl ether); nmax (Nujol) 1618, 1068, 1022 cmꢀ1
;
dH (400 MHz,
8.15 (1H, s), 8.13 (1H, d, J 2.2 Hz), 7.58e7.51 (3H, m), 7.41e7.32 (4H,
m), 3.58e3.49 (1H, m), 3.47e3.36 (1H, m), 2.63e2.54 (1H, m),
1.96e1.87 (1H, m), 1.62e1.54 (2H, m), 1.40 (3H, s); dC (75 MHz,
CDCl3) 171.7, 140.8 (s), 139.1 (s), 132.9, 131.3, 130.4, 129.5, 127.5,
127.3, 120.9 (s), 86.5 (s), 60.7, 52.0 (s), 24.6, 23.7, 16.9; HRMS (ESI):
MHþ, found 388.0369. C19H19BrNOS requires 388.0365.
DMSO-d6, 125 ꢁC) 8.54 (1H, s), 8.01 (1H, d, J 5.9 Hz), 7.41e7.19 (13H,
m), 3.62 (1H, br s), 3.51e3.48 (1H, m), 2.50e2.48 (1H, m), 2.17e2.10
(1H, m),1.73e1.70 (1H, m),1.50e1.40 (1H, m); dC (100 MHz, DMSO-d6,
125 ꢁC) 166.3, 140.8 (s), 140.7 (s), 139.6 (s), 130.2, 129.4, 128.1, 126.8,
126.7, 126.6, 126.4, 126.1, 125.9, 124.9 (s), 85.9 (s), 60.9, 59.7 (s), 22.8,
21.9; HRMS (ESI): MHþ, found 372.1416. C24H22NOS requires 372.1417.
5.5.5. cis-10cþtrans-10c. Yield 41% (0.14 g).
5.4.2. 60-Bromo-30,30-diphenylspiro[1,3-oxathiane-2,40(30H)quino-
line] (8b). Eluent for column chromatography: hexanes/diethyl
ether (4:1, v/v); yield 71% (0.32 g); mp 107e109 ꢁC (colourless
5.5.6. trans-60-Chloro-30-methyl-30-phenylspiro[1,3-oxathiane-
2,40(30H)quinoline] (10c). Mp 158e160 ꢁC (colourless prisms from
prisms from diethyl ether); nmax (Nujol) 1580, 1108, 1023 cmꢀ1
;
dH
diethyl ether); nmax (Nujol) 1626, 1088 cmꢀ1
; dH (400 MHz, CDCl3,
(300 MHz, DMSO-d6, 100 ꢁC) 8.61 (1H, m), 8.09 (1H, d, J 1.8 Hz), 7.51
(1H, dd, J 8.4, 2.4 Hz), 7.36e7.12 (11H, m), 3.66e3.64 (1H, m),
3.54e3.46 (1H, m), 2.59e2.53 (1H, m), 2.21e2.12 (1H, m), 1.77e1.71
(1H, m), 1.52e1.45 (1H, m); dC (75 MHz, DMSO-d6, 100 ꢁC) 167.2,
140.3 (s), 139.9 (s), 139.2 (s), 133.7 (s), 130.9, 129.9, 129.4, 128.6,
127.1, 126.7, 126.6, 126.2, 126.1, 119.5 (s), 85.4 (s), 61.2, 59.3 (s), 22.6,
21.7; HRMS (ESI): MHþ, found 450.0524. C24H21BrNOS requires
450.0522.
55 ꢁC) 8.12 (1H, s), 8.00e7.98 (1H, m), 7.58e7.55 (2H, m), 7.41e7.33
(5H, m), 3.58e3.53 (1H, m), 3.49e3.41 (1H, m), 2.60e2.54 (1H, m),
1.97e1.91 (1H, m), 1.63e1.57 (1H, m), 1.43 (3H, s); dC (100 MHz,
CDCl3, 55 ꢁC) 171.3, 140.8 (s), 139.3 (s), 132.9 (s), 130.5 (s), 130.4,
129.7,129.3,128.4,127.5,127.4, 87.0 (s), 60.7, 52.1 (s), 24.5, 23.8,17.1;
HRMS (ESI): MHþ, found 344.0867. C19H19ClNOS requires 344.0870.
Acknowledgements
5.4.3. 60-Chloro-30,30-diphenylspiro[1,3-oxathiane-2,40(30H)quino-
line] (8c). Eluent for column chromatography: hexanes/diethyl
ether (9:1, v/v); yield 70% (0.28 g); mp 173e175 ꢁC (colourless
This work was supported by the Ministerio de Ciencia e Inno-
vacion of Spain (Project CTQ2008-05827/BQU) and Fundacion
Seneca-CARM (Project 08661/PI/08). B.B. thanks Fundacion Seneca-
prisms from diethyl ether); nmax (Nujol) 1608, 1093 cmꢀ1
; dH
ꢀ
CARM for a fellowship. We also thank Prof. Eliseo Chacon (U. Sevilla,
(300 MHz, DMSO-d6, 100 ꢁC) 8.60 (1H, m), 7.95 (1H, d, J 1.8 Hz),
7.37e7.18 (11H, m), 3.69e3.63 (1H, m), 3.55e3.47 (1H, m),
2.63e2.54 (1H, m), 2.21e2.12 (1H, m), 1.80e1.69 (1H, m), 1.55e1.43
(1H, m); dC (75 MHz, DMSO-d6, 100 ꢁC) 167.1, 140.3 (s), 139.6 (s),
139.2 (s), 131.3 (s), 129.9, 129.4, 128.2, 127.9, 126.7, 126.2, 126.3,
126.1, 124.2, 85.5 (s), 61.2, 59.3 (s), 22.6, 21.7; HRMS (ESI): MHþ,
found 406.1026. C24H21ClNOS requires 406.1027.
Spain) for their useful comments on the kinetic analysis.
Supplementary data
Details of computational procedures. Geometries, Cartesian
coordinates and energies for all the stationary points. ORTEP rep-
resentation of the crystal structure of trans-10b. Copy of 1H and 13
C
5.5. Preparation of spiro[1,3-oxathiane-2,40(30H)-quinolines] 10
NMR spectra of compounds 8 and trans-10. Cif file of compound
trans-10b. Crystallographic data (excluding structure factors) for
the structure in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication no.
856166. Copies of the data can be obtained, free of charge, on ap-
plication to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: þ44
(0)1233 336033 or e-mail: deposit@ccdc.cam.ac.UK). Supplemen-
tary data associated with this article can be found in the online
version, at doi:10.1016/j.tet.2012.04.021.
Methylphenylketene (0.13 g, 1 mmol) in anhydrous dichloro-
methane (5 mL) was added to
a solution of the 2-(2-
triphenylphosphoranylideneaminophenyl)-1,3-oxathiane
6
(1 mmol) in the same solvent (15 mL). The reaction mixture was
stirred at room temperature for 30 min. The solvent was removed
under reduced pressure and the resulting material was purified by
column chromatography on silica gel, using hexanes/diethyl ether
(9:1, v/v) as eluent, to provide ketenimine 9. A solution of this
ketenimine 9 in anhydrous ortho-xylene (15 mL) was heated at
reflux temperature for 12 h. After the reaction mixture has cooled,
the solvent was removed under reduced pressure. The resulting
material was purified by column chromatography on silica gel,
using hexanes/diethyl ether (7:3, v/v) as eluent.
References and notes
1. (a) Kobayashi, S.; Jorgensen, K. A. Cycloaddition Reactions in Organic Synthesis;
Wiley-VCH: Weinheim, 2002; (b) Enders, D.; Knopp, M.; Schiffers, R. Tetrahe-
dron: Asymmetry 1996, 7, 1847e1882; (c) Nakai, T.; Mikami, K. Chem. Rev. 1986,
86, 885e902; (d) Llardi, E. A.; Stivala, C. E.; Zakarian, A. Chem. Soc. Rev. 2009, 38,
3133e3148; (e) Neuschuetz, K.; Velker, J.; Neier, R. Synthesis 1998, 227e255.
2. Tietze, L. F.; Beifuss, U. Angew. Chem., Int. Ed. Engl. 1993, 32, 131e163.
3. (a) Ho, T.-L. Tandem Organic Reactions; Wiley: New York, NY, 1992; (b) Tietze, L.
F.; Brasche, G.; Gericke, K. Domino Reactions in Organic Synthesis; Wiley-VCH:
Weinheim, 2006; (c) Parsons, P. J.; Penkett, C. S.; Shell, A. J. Chem. Rev. 1996, 96,
195e206; (d) Pellisier, H. Tetrahedron 2006, 62, 1619e1665; (e) Mak, X. Y.;
Crombie, A. L.; Danheiser, R. L. J. Org. Chem. 2011, 76, 1852e1873; (f) Robinson, J.
M.; Sakai, T.; Okano, K.; Kitawaki, T.; Danheiser, R. L. J. Am. Chem. Soc. 2010, 132,
11039e11041; (g) Beck, A.; Knothe, L.; Hunkler, D.; Prinzbach, H. Tetrahedron
Lett. 1984, 25, 1785e1788; (h) Raw, S. A.; Taylor, R. J. K. J. Am. Chem. Soc. 2004,
126, 12260e12261; (i) Denmark, S. E.; Gomez, L. J. Org. Chem. 2003, 68,
8015e8024; (j) Hopf, H.; Wolff, J. Eur. J. Org. Chem. 2001, 4009e4030; (k)
Denmark, S. E.; Gomez, L. Org. Lett. 2001, 3, 2907e2910; (l) Swarbrick, T. M.;
5.5.1. cis-10aþtrans-10a. Yield 40% (0.12 g).
5.5.2. trans-30-Methyl-30-phenylspiro[1,3-oxathiane-2,40(30H)quino-
line] (10a). Mp 158e159 ꢁC (colourless prisms from diethyl ether);
nmax (Nujol) 1618, 1035 cmꢀ1
; dH (300 MHz, CDCl3) 8.14 (1H, s), 8.00
(1H, d, J 7.4 Hz), 7.61e7.58 (2H, m), 7.48e7.28 (6H, m), 3.55e3.47
(1H, m), 3.44e3.35 (1H, m), 2.69e2.60 (1H, m), 2.00e1.92 (1H, m),
1.60e1.52 (2H, m), 1.40 (3H, s); dC (75 MHz, CDCl3) 171.1, 141.7 (s),
139.6 (s), 130.4, 129.9, 128.4, 128.2 (s), 127.9, 127.3, 127.2, 87.0 (s),

M. Alajarin et al. / Tetrahedron 68 (2012) 4672e4681
4681
Marko, I.; Kennard, L. Tetrahedron Lett. 1991, 32, 2549e2552; (m) Skropeta, D.;
Rickardd, R. W. Tetrahedron Lett. 2007, 48, 3281e3284.
Spellmeyer, D. C.; Jefford, C. W.; Rimbault, C. G.; Wang, Y.; Miller, R. D. J. Org.
Chem. 1988, 53, 2125e2127; (e) Buda, A. B.; Wang, Y.; Houk, K. N. J. Org. Chem.
1989, 54, 2264e2266; (f) Kallel, E. A.; Wang, Y.; Spellmeyer, D. C.; Houk, K. N. J.
Am. Chem. Soc. 1990, 112, 6759e6763.
4. (a) Nicolaou, K. C.; Edmonds, D. J.; Bulger, P. G. Angew. Chem., Int. Ed. 2006, 45,
7134e7186; (b) Arns, S.; Barriault, L. Chem. Commun. 2007, 2211e2221.
5. (a) Alajarin, M.; Bonillo, B.; Ortin, M.-M.; Sanchez-Andrada, P.; Vidal, A. Org.
Lett. 2006, 8, 5645e5648; (b) Alajarin, M.; Bonillo, B.; Ortin, M.-M.; Sanchez-
Andrada, P.; Vidal, A. Eur. J. Org. Chem. 2011, 1896e1913.
12. Isaacs, N. S. Physical Organic Chemistry; Longman Scientific & Technical: Belfast,
1987.
13. (a) Naruse, Y.; Inagaki, S. Chem. Lett. 2007, 36, 820e825; (b) Inagaki, S. Top. Curr.
Chem. 2009, 289, 83e127; (c) Ikeda, H.; Ushioda, N.; Inagaki, S. Chem. Lett. 2001, 30,
270e271;(d)Yasui, M.;Naruse, Y.;Inagaki, S. J.Org. Chem. 2004, 69, 7246e7249;(e)
Naruse, Y.; Hayashi, Y.; Inagaki, S. Tetrahedron Lett. 2003, 44, 8509e8512.
14. The transition state for the electrocyclization of hexatriene to cyclohexadiene
has been studied in depth. However, for the sake of comparison we include
here this structure calculated at the same level of theory. (a) Rodriguez-Otero, J.
J. Org. Chem. 1999, 64, 6842e6848; (b) Jiao, H.; Schleyer, P. v. R. J. Am. Chem. Soc.
1995, 117, 11529e11535; (c) Evanseck, J. D.; Thomas, B. E., IV; Spellmeyer, D. C.;
Houk, K. N. J. Org. Chem. 1995, 60, 7134e7141; (d) Jefford, C. W.; Bernardinelli,
G.; Wang, Y.; Spellmeyer, D. C.; Buda, A.; Houk, K. N. J. Am. Chem. Soc. 1992, 114,
1157e1165; (e) Baldwin, J. E.; Reddy, V. P.; Schaad, L. J.; Hess, B. A. J. Am. Chem.
Soc. 1988, 110, 8554e8555.
6. Alajarin, M.; Bonillo, B.; Ortin, M.-M.; Sanchez-Andrada, P.; Vidal, A.; Orenes, R.-
A. Org. Biomol. Chem. 2010, 8, 4690e4700.
7. For some representative reports on tandem processes initiated by 1,5-hydride
shifts see: (a) Nijhuis, W. H. N.; Verboom, W.; Reinhoudt, D. N.; Harkema, S. J.
Am. Chem. Soc. 1987, 109, 3136e3138; (b) Mori, K.; Sueoka, S.; Akiyama, T. J. Am.
Chem. Soc. 2011, 133, 2424e2426; (c) Zhou, G.; Zhang, J. Chem. Commun. 2010,
~
6593e6595; (d) Barluenga, J.; Fananas-Mastral, M.; Aznar, F.; Valdes, C. Angew.
Chem., Int. Ed. 2008, 47, 6594e6597; (e) Bolte, B.; Gagosz, F. J. Am. Chem. Soc.
2011, 133, 7696e7699; (f) Haibach, M. C.; Deb, I.; De, C. K.; Seidel, D. J. Am. Chem.
Soc. 2011, 133, 2100e2103; (g) Vadola, P. A.; Sames, D. J. Am. Chem. Soc. 2009,
131, 16525e16528.
8. (a) Alajarin, M.; Bonillo, B.; Sanchez-Andrada, P.; Vidal, A.; Bautista, D. Org. Lett.
2009, 11, 1365e1368; (b) Alajarin, M.; Bonillo, B.; Sanchez-Andrada, P.; Vidal, A.
J. Org. Chem. 2010, 75, 3737e3750.
15. It is known that conjugating substituents produce large accelerations of the
rate of nucleophilic additions to heterocumulenes such as ketenes, see Tidwel,
T. T. Ketenes; Jonh Wiley & Sons: New York, NY, 1995.
9. Although C-methyl-C-phenyl ketenimines 9 could exist in diastereoisomeric
forms, as
a
consequence of their two stereogenic elements, chiral centre
16. Espenson, J. H. Chemical Kinetics and Reaction Mechanisms; McGraw-Hill: New
York, NY, 1981.
17. Ardakani, M. A.; Smalley, R. K.; Smith, R. H. J. Chem. Soc., Perkin Trans. 1 1983,
2501e2506.
(acetalic carbon atom) and chiral axis (that of the non symmetrical ketenimine
function), their interconversion is very rapid by virtue of the rapid nitrogen
atom inversion.
10. The energy barriers are referred to the most stable conformer of the reactant
ketenimines and the o-azaxylylene intermediates.
11. (a) Rondan, N. G.; Houk, K. N. J. Am. Chem. Soc. 1985, 107, 2099e2111; (b) Ruldolf,
K.; Spellmeyer, D. C.; Houk, K. N. J. Org. Chem. 1987, 52, 3708e3710; (c) Spell-
meyer, D. C.; Houk, K. N. J. Am. Chem. Soc. 1988, 110, 3412e3416; (d) Houk, K. N.;
18. Prepared following the experimental procedure reported for the preparation of
similar compounds in Cuevas, J. C.; De Mendoza, J.; Prados, P. J. Org. Chem. 1988,
53, 2055e2066.
19. Taylor, E. C.; McKillop, A.; Hawks, G. H. Org. Synth. 1972, 52, 36e38.
20. Pracejus, H.; Wallura, G. J. Prakt. Chem. 1962, 19, 33e36.