Conformational analysis of p-X-anilino dioxaphosphinanes. Substituent effects on 31P and 15N NMR signals and on negative hyperconjugation (n-σ*)

Article abstract of DOI:10.1016/j.tet.2010.01.034

The conformational analysis of anancomeric cis-ax and cis-eq 2-p-X-anilino-2-thio-4,6-dimethyl-1,3,2λ⁵-dioxaphosphi nanes (X=OCH₃, C₆H₁₁, H, Cl, CN and NO₂) is informed. In accordance with ³J_HH, ³J_HP, ⁴J_HP, and ³J_CP coupling constants the preferred conformation in solution is a chair in both series of compounds. Structural parameters obtained through X-ray diffraction studies of the series of cis-ax and cis-eq diastereomers 1-6, suggest that the stabilization of the axial and equatorial diastereomers in chair conformation rely on stereoelectronic n_πO-σ*_P-N and n_πN-σ*_P-O interactions, respectively. Theoretical Kohn-Sham DFT calculations support the participation of the cited stereoelectronic interactions not only to understand the conformational behavior of these systems but also to give an explanation of the observed substituent-induced chemical shift (SCS) on ³¹P and ¹⁵N NMR signals.

Full text of DOI:10.1016/j.tet.2010.01.034

Tetrahedron 66 (2010) 2066–2076
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Conformational analysis of p-X-anilino dioxaphosphinanes. Substituent effects
31
15
on P and N NMR signals and on negative hyperconjugation (n–
s
*)
b
c
a
b
a
,
1
´
´
´
Zaira Domınguez , Marcelo Galvan , Ma. Teresa Cortez , Magali Salas , Rocıo Meza
,
Marco A. Leyva-Ramirez a, Barbara Gordillo a
,
*
a
´
´
´
´
´
Departamento de Quımica del Centro de Investigacion y Estudios Avanzados del Instituto Politecnico Nacional. Apdo. Postal 14-740, Mexico, D. F., Mexico
Unidad de Servicios de Apoyo en Resolucio´n Anal´ıtica. Universidad Veracruzana. Dr. Rafael Sanchez Altamirano s/n C.P. 91190, Xalapa, Ver. Mexico
Departamento de Qu´ımica, Divisio´n de Ciencias Basicas e Ingenierıa, Universidad Autonoma Metropolitana-Iztapalapa, A. P. 55-534, Mexico, D. F, Mexico
b
c
´
´
´
´
´
´
a r t i c l e i n f o
a b s t r a c t
The conformational analysis of anancomeric cis-ax and cis-eq 2-p-X-anilino-2-thio-4,6-dimethyl-1,3,2l5
-
Article history:
Received 31 July 2009
Received in revised form 8 January 2010
Accepted 8 January 2010
Available online 18 January 2010
3
dioxaphosphinanes (X¼OCH3, C6H11, H, Cl, CN and NO2) is informed. In accordance with JHH
, , ,
3JHP 4JHP
and 3JCP coupling constants the preferred conformation in solution is a chair in both series of compounds.
Structural parameters obtained through X-ray diffraction studies of the series of cis-ax and cis-eq di-
astereomers 1–6, suggest that the stabilization of the axial and equatorial diastereomers in chair con-
formation rely on stereoelectronic npO–
s
*
and npN–
s
*
interactions, respectively. Theoretical
P–O
P–N
Keywords:
Kohn–Sham DFT calculations support the participation of the cited stereoelectronic interactions not only
to understand the conformational behavior of these systems but also to give an explanation of the ob-
served substituent-induced chemical shift (SCS) on 31P and 15N NMR signals.
Ó 2010 Elsevier Ltd. All rights reserved.
Dioxaphosphinanes
Thiophosphoramidates
Conformation
NMR
SCS and LFER correlations
X-ray crystal structures
Negative hyperconjugation
1. Introduction
This behavior has been attributed mainly to the stabilizing
stereoelectronic interactions npO–
s
*
(endo-anomeric
P-w(substituent)
In contrast to the conformational behavior of monosubstituted cy-
clohexanes1 and 2-sustituted-1,3-dioxanes,2 where the steric effects
govern the conformational preference of substituents, electronic ef-
fects play an important role in the prediction of the conformation of
1,3,2-dioxaphosphinanes.3 For instance, in the case of the 2-Z-2-oxo-
effect) for the axial preference and the npY–
anomeric effect) for the equatorial one (Scheme 2).
s
*
(exo-
P–Oendocyclic
3a,6,7
1,3,2l
5-dioxaphosphinanes of Scheme 1, the conformational equilib-
rium is notably shifted to the right when Z is an electron withdrawing
group (W) such as Cl, F or OPh;4 or to the left when Z is an electron
donating group (Y), for example, an amino substituent NR2 (R¼H, Me).5
Scheme 1.
Scheme 2.
Along with the exo-anomeric effect, the equatorial preference of
an amino group at the four-coordinated phosphorus of a 1,3,2-
dioxaphosphinane ring has been ascribed to the repulsive 1,3-syn
axial interaction between the R groups of the amino (NR2)
* Corresponding author. Tel.: þ52 55 5747 3729; fax: þ52 55 5747 3389
E-mail address: ggordill@cinvestav.mx (B. Gordillo).
Present address: FCBITyT. Universidad Auto´noma de Tlaxcala, Apizaco Tlax,
1
Mexico.
0040-4020/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.01.034

Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
2067
substituent and the axial hydrogen atoms at the 4 and 6 positions of
the heterocycle,3b the stabilizing npO–
interaction (endo-
anomeric effect)8 and/or the p
(N)–d (P) interaction between the
s*
P]O(S)
p
p
lone pair at nitrogen atom and one of the empty orbitals of
phosphorus.9 However, the role played by phosphorus d orbitals in
1,3,2-dioxaphosphinanes remains controversial, and theoretical
studies10,11 performed on phosphorus compounds have provided
evidence against d orbital participation even in molecules with low-
lying acceptor orbitals such as those containing PO and PS groups.12
On the other hand, Bentrude et al. found that for analogous 2-
anilino-2-oxo-5,5-dimethyl-1,3,2
lino group prefers an axial orientation in the solid state. In order to
explain this observation, a stabilizing np PN orbital interaction,
l
5-oxazaphosphinanes,13 the ani-
–s*
N
ascribed to the endo-anomeric effect, was proposed. From these
results, it is clear that the nature of R in the amino group may
change the regular tendency of the amino substituent to occupy the
equatorial orientation; a tendency, that is, clearly observed, at least,
in highly hindered thiophosphoramidates.14a
In this work we are reporting the conformational behavior of two
series of anancomeric 2-anilino-1,3,2l
5-dioxaphosphinanes
(Scheme 3) in solution and in solid state. The electronic character-
istics of the substituents (X) at the para position of the aromatic ring
were varied from electron withdrawing to electron donating groups
with the idea of analyzing the participation of the stereoelectronic
effects in the conformation adopted by both diastereomeric series of
compounds. NMR data as 31P and 15N chemical shifts were corre-
lated with Hammett constants within the context of linear free-
energy relationships (LFER) theory, considering that the tendencies
observed are a consequence of the changes in electronic density
caused by the p-X-substituent. Kohn–Sham DFT calculations15 were
Scheme 4. The nomenclature ax and eq of 1–6 was used to mark the position of the
amino substituent, even though nitrogen ranks lower than sulfur by CIP rules.
All 1H and 13C NMR experiments were carried out in CDCl3 at
27 ꢀC. Natural abundance 15N NMR spectra of the cis-eq series were
recorded in CDCl3 at 35 ꢀC, meanwhile the cis-ax series, which are
barely soluble in CDCl3, were recorded in DMSO at 75 ꢀC; the
spectra were acquired with a decoupled INEPT pulse sequence
(J¼90 Hz) using nitromethane as the external reference. The as-
signment of configuration of cis-ax and cis-eq diastereomers was
successfully attained by 31P NMR, taking into account that the
signal of the cis substituted thiophosphoramidates, is shifted to
performed to give theoretical support to the participation of n–s*
anomeric stabilization through the geometry of the optimized
ground state structures.
Table 1
1H, 31P, and 15N chemical shifts (in ppm) for 1–6 ax and 1–6 eq 2-p-X-anilino-2-thio-
4,6-dimethyl-1,3,2l
5-dioxaphosphinanes in CDCl3
Scheme 3.
2. Results
The synthesis of compounds 1–6 was achieved in a tandem
sequence of three steps as it is summarized in Scheme 4, with the
isolation of neither the intermediate 2-chloro-1,3,2-l
3-dioxa-
phosphinane (I, mixture of isomers) nor the phosphoramidite (II,
mixture of isomers). Both intermediates decompose with hu-
midity; therefore, they should be managed carefully and main-
tained under dry atmosphere if needed (see Experimental part for
details). Due to the fact that the starting 2,4-pentanediol is
a mixture of meso and d/l isomers, thiophosphoramidates 1–6
were obtained as a mixture of anancomeric-cis and racemic-trans
isomers. Because the first step of the synthetic pathway is dia-
stereoselective,16 and favors the cis-ax (I) over the trans-ax (I)
isomer (the cis-eq isomer is not observed), the trans phosphor-
amidites (II) and consequently the trans thiophosphoramidates
(1–6) were the minor components (10% at the most) in the mix-
tures cis/trans, and they were not characterized in this work. The
mixture of ax and eq cis-thiophosphoramidates (1–6) was sepa-
rated and purified by column chromatography.
X
H4,6
H5ax
H5eq
H7,8
31P
15Na
OMe ax
OMe eq
C6H11 ax
C6H11 eq
H ax
H eq
Cl ax
Cl eq
CN ax
4.64
4.84
4.65
4.87
4.65
4.88
4.62
4.87
4.62
4.89
4.63
4.89
1.92
1.58
1.92
1.69
1.93
1.71
1.92
1.67
1.90
1.75
1.93
1.78
1.79
1.76
1.79
1.81
1.81
1.82
1.80
1.82
1.81
1.87
1.83
1.91
1.42
1.35
1.41
1.37
1.44
1.38
1.43
1.37
1.39
1.36
1.43
1.41
60.6
63.5
60.0
63.0
59.4
62.9
59.4
62.9
57.3
62.2
57.2
61.9
ꢁ298.6
ꢁ295.3
ꢁ295.5
ꢁ291.4
ꢁ293.8
ꢁ290.1
ꢁ293.7
ꢁ290.9
ꢁ285.8
ꢁ284.5
ꢁ283.8
ꢁ283.6
CN eq
NO2 ax
NO2 eq
a
The 15N spectra of cis-ax compounds were acquired at 75 ꢀC using DMSO-d6 as
solvent.

2068
Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
high frequency of the corresponding trans isomers12,14a (Table 1).
Proton chemical shifts ( ) are reported inTable 1; the assignment of
3. Discussion
d
the signals and the backbone coupling constants (Table 2) were
obtained through first-order analysis of the spectra, along with
homo and heteronuclear decoupling experiments. The complete
assignment of 13C NMR signals (Table 3), was achieved by means of
1H, 13C correlated 2D NMR spectra. The twelve isomers 1–6 are
solids, thus recrystallization from mixtures of AcOEt/Hexane or
CH2Cl2/Hexane rendered good quality crystals for X-ray diffraction
analysis. The molecular structures are presented in Figures 1–12
and the structural parameters in Tables 1–8 (Supplementary data).
The DFT calculations were carried out at B3LYP/6-31G (d,p) level of
theory; such combination of exchange and correlation functional
and basis set provides reliable geometries and acceptable chemical
accuracy for phosphorus containing systems.17
3.1. Conformational analysis in solution
The chemical shift of H4,6 is in the range of 4.62–4.89 ppm
(Table 1) and the corresponding of C4,6 is in the range of
73.8–76.7 ppm (Table 3). Taking into consideration that the C4,6
signal of the cis-ax isomers is shifted to high frequency respect to
the signal of the corresponding cis-eq, but the opposite trend is
found for H4,6, it is concluded that the amino substituent of
dioxaphosphinane ring suffer a reciprocal of a ga gauche effect.18
This effect can be ascribed to a negligible participation of the
steric compression by the anilino substituent on H4,6, very likely
due to the ability of the endocyclic oxygens to change hybrid-
ization from sp3 to sp2 (COP angles w 118ꢀ).14,19
The conformation of the dioxaphosphinane ring was deduced
3
3
from vicinal coupling constants JHH
,
3JHP
,
4JHP, and JCP (Tables 2
3
3
and 3). Determination of JH4,6H5ax and JH4,6H5eq was achieved
through heteronuclear H{31P} decoupling experiments, whereas
3JHP and 4JHP were obtained by direct comparison of the decoupled
H{31P} and the undecoupled 1H NMR spectra. The observed vicinal
3JH4H5ax is around 11 Hz (anti arrangement) for both series of di-
Table 2
1
1H NMR backbone coupling constants and JNP (Hz) for 1–6 ax and 1–6 eq 2-p-X-
anilino-2-thio-4,6-dimethyl-1,3,2
CDCl3
l
5-dioxaphosphinanes. First-order analysis in
X
3JH4,6H5ax
3JH4,6H5eq
3JH4,6P
2JH5axH5eq
4JH7P
1JNP
3
astereomers cis-ax and cis-eq 1–6, whereas JH4H5eq is 2.2–2.6 Hz
OMe ax
OMe eq
C6H11 ax
C6H11 eq
H ax
H eq
Cl ax
Cl eq
CN ax
11.3
11.5
11.4
11.0
11.4
11.3
11.1
11.6
11.4
11.5
11.0
11.5
2.6
2.3
2.4
2.2
2.6
2.6
2.6
2.2
2.6
2.2
2.6
2.4
2.3
2.3
2.5
2.0
2.1
2.7
2.1
2.5
2.0
2.4
2.6
2.4
14.5
14.6
14.4
14.9
14.6
13.9
14.5
13.4
14.6
14.5
14.6
13.0
2.0
2.2
1.8
1.9
2.0
2.0
1.9
2.0
1.8
2.0
2.0
2.2
9.9
25.8
10.9
25.8
10.9
25.2
10.9
24.8
9.9
(gauche arrangement), suggesting that these molecules are in so-
lution in chair conformation. In agreement with this finding, values
3
4
of JH4(6)P and JH7(8)P coupling constants are between 2.0–2.7 Hz
and around 2.0 Hz, respectively.3a,b
On the other hand, JCP coupling constants obtained from 13C
3
NMR signals of the methyl groups at positions C7 and C8 for cis-ax
and cis-eq, are between 8.5 and 10.8 Hz (Table 3), indicating that
the molecules are in locked chair conformations.3a,b,14 As observed
initially by Nifantiev20 for similar phosphorus heterocycles, we also
found that one of the most sensible coupling constant to configu-
CN eq
NO2 ax
NO2 eq
23.8
10.9
23.8
3
ration is JC5P; the values of the cis-ax thiophosphoramidates 1–6
Table 3
13C NMR signal assignments for cis-ax and cis-eq (1–6)a
b
p-X
C-4,6 (2JC4,6P
)
C-5(3JC5P
)
C-7,8(3JC7,8P
)
C-11 (2JC11P
)
C-12 (3JC12P
)
C-13
C-14
OMe ax
OMe eq
OMe Ac
C6H11 ax
C6H11 eq
C6H11-Ac
H ax
76.0 (8.8)
73.8 (4.6)
d
40.5 (6.2)
41.1 (3.1)
d
22.4 (8.8)
22.1 (10.8)
d
132.5 (n.o.)
131.3 (5.4)
140.3
137.0 (n.o.)
136.3 (4.6)
144.4
139.5 (n.o.)
138.7 (5.4)
146.8
138.1 (n.o.)
137.5 (4.6)
145.2
144.3 (n.o.)
143.5 (4.9)
151.1
146.0 (n.o.)
144.9 (4.9)
152.5
120.6 (6.2)
122.6 (5.4)
116.2
118.2 (6.9)
119.3 (6.1)
115.4
118.2 (6.9)
119.4 (6.1)
115.3
119.5 (6.1)
121.0 (6.1)
116.3
118.0 (6.1)
119.1 (7.3)
114.4
117.3 (6.9)
118.2 (7.7)
113.4
114.8
114.3
114.9
127.8
127.5
127.6
129.4
129.1
129.5
129.4
129.0
129.2
133.5
133.8
133.7
125.6
125.3
126.3
155.6
156.2
152.8
142.3
142.9
138.5
122.4
123.0
118.6
127.7
128.2
123.0
105.5
106.0
99.0
75.9 (9.2)
74.0 (4.6)
d
40.5 (6.1)
41.2 (3.1)
d
22.4 (9.2)
22.2 (10.8)
d
76.0 (9.2)
74.0 (5.3)
d
40.5 (5.4)
41.2 (3.1)
d
22.4 (9.2)
22.1 (10.8)
d
H eq
H Ac
Cl ax
76.2 (9.2)
74.1 (4.6)
d
40.4 (6.9)
41.1 (2.3)
d
22.4 (8.5)
22.1 (10.8)
d
Cl eq
Cl Ac
CN ax
CN eq
CN Ac
NO2 ax
NO2 eq
NO2 Ac
75.6 (7.7)
74.9 (4.5)
d
40.4 (6.1)
41.6 (3.1)
d
22.3 (9.2)
22.5 (10.8)
d
76.7 (9.7)
74.5 (4.6)
d
40.4 (6.1)
41.3 (3.1)
d
22.2 (9.2)
22.3 (10.8)
d
142.4
142.8
139.2
a
Chemical shifts (d) in ppm from TMS in CDCl3. In parentheses JCP in Hz.
b
n.o.¼not observed.
c
This work.

Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
2069
are 2–3 Hz larger than the cis-eq ones (Table 3). It is worth men-
corresponding of Rꢁ/
d
15N than to the corresponding of Rꢁ/
d
31P, as
3
tioning that similar values of JC5P, were reported for analogous
expected (Table 4). These findings give support to the participation
of the stabilizing hyperconjugative endo-anomeric effect [npO–
N] for the axial isomers and the corresponding exo-anomeric effect
[npN–
P–O] for the equatorial (see discussion below).3,14
dioxaphosphinanes in chair conformation.14a,21 In consort, values of
2JC4(6)P are around 4.6 Hz for the cis-eq series and between 7.7 and
9.7 Hz for the cis-ax series (see Table 3).
s*
P–
s
*
In addition, it can be observed from Table 4, that 31P chemical
shifts of axial and equatorial diastereomers correlate inversely with
sꢁp and Rꢁ (average slope values: w ꢁ2.58 for cis-ax, and w ꢁ1.04
for cis-eq), which indicate the inverse dependence of the charge
transfer from the p-substituent of the aryl group to phosphorus,
where electron withdrawing substituents provoke a shielding ef-
fect and vice versa. We explained this abnormal behavior in an
earlier study on thiophosphates21 by proposing a charge compen-
sation effect involving the transfer of charge from the endocyclic
oxygens to phosphorus. In consort, Kuivalainen et al.26 found by
using AIM theory that it is very likely that the origin of the 31P
chemical shift of O,O-dialkyl-O-aryl phosphorothionates relies on
a back bonding character (X /P) of the P–O and P]S bonds. The
variation of the chemical shift on the phosphorus nucleus can also
be explained as the result of a geometrical change in the OPO bond
angle of the dioxaphosphinane ring, as demonstrated by Goren-
stein for cyclic and acyclic phosphates;27 however, we did not ob-
serve significant changes in the OPO bond angle, neither within the
axial series nor within the equatorial ones, of the X-ray structures of
thiophosphoramidates 1–6.
3.2. Substituent effects on 15N and 31P NMR signals
The 15N chemical shifts of the dioxaphosphinanes 1–6 are be-
tween ꢁ298.6 and ꢁ283.6 ppm, being the signal of the equatorial
compounds shifted to high frequency of the corresponding axial
epimers (Table 1). As expected, an electron releasing group in the p-
position of the aryl group leads to a shielding effect in nitrogen
nucleus and vice versa.22 Nevertheless, the opposite behavior is
observed for the 31P chemical shifts (Table 1).
In agreement with the data reported in the literature for cyclic
oxazaphospholidines23 and phosphoramidates,24 where changes in
geometry around the nitrogen atom (C–N–P bond angle mainly)
explain the trend observed in 15NMR chemical shifts; we also ob-
served by means of X-ray diffraction analysis (see below) that there
is a flattening of the nitrogen edge in the case of the equatorial
series, with the consequent opening of the C11–N–P bond angle,
that explains the deshielding of the 15N NMR signal. Along with this
geometrical change, we found that the electronic nature of the para
substituent influences the electronic density around the nitrogen,
therefore the chemical shift. This electronic effect is observed for
both axial and equatorial isomers; although it is greater for the axial
series with a difference of almost 15 ppm, between the low-fre-
quency shifted signal (X¼OMe) and the high-frequency shifted
peak (X¼NO2), whereas the corresponding difference for the
equatorial series is around 12 ppm.
3.3. Solid state structural analysis
In order to support the participation of stereoelectronic effects
in the conformation adopted for aromatic thiophosphoramidates,
X-ray diffraction analyses of compounds (1–6) in cis-ax and cis-eq
configurations were performed. The structure drawings are shown
in Figures 1–12, and data collection and refinement parameters for
the axial series and equatorial series in Tables 1 and 2, respectively
(Supplementary data). Seven of the twelve compounds crystallized
in the monoclinic system; the space group for cis-ax (5 and 6) and
cis-eq (1, 3, and 4) is P21/c, whereas for cis-ax-2 and cis-eq-2 is P21/
a, and P21/n, respectively. Four of the other five compounds [cis-ax
(1, 3, 4) and cis-eq-6] crystallized in the triclinic system in P-1 space
group; and the last one, the cis-eq-5, crystallized in the Pbca space
group of the orthorhombic system. Two molecules per unit cell
were found for six compounds: cis-ax and cis-eq (1, 3, and 4), with
basically the same dioxaphosphinane ring conformation in each
pair, but changes in the rotameric conformation of the anilino
substituent. Selected bond lengths, bond angles and torsion angles
are presented in Tables 3–8 (Supplementary data). The N–H hy-
drogen atom of all the structures was located from a difference
Fourier map and refined.
In order to evaluate the nature of the so-called Substituent-In-
duced Chemical Shifts (SCS), several plots of d d d
15N, 31P, and 13C
(C-11) data of both series of compounds against Hammett type
constants (sꢁp and Rꢁ)25 were performed. The regression coefficient
(R) and the slope of the linear correlations for each series is pre-
sented in Table 4; acceptable regression values Rw0.98 (sꢁp and Rꢁ)
were obtained for both 31P and 15N chemical shifts. The slope of the
curves is significantly higher (w four times for the ax series and w
eight times for the eq series) for the correlation of Rꢁ (or spꢁ) vs
d
15N
than the corresponding for Rꢁ (or sꢁp ) vs 31P; these results suggest
d
that the nitrogen atom is more susceptible to sense electronic ef-
fects than the phosphorus atom, this is perhaps due to the fact that
the anilino nitrogen is four bonds apart from the X substituent and
phosphorus is five bonds apart; moreover, the nitrogen lone pair p-
overlaps with the aromatic ring, contributing to polarization and
charge redistribution by resonance effect. The slope of the curves
Rꢁ/ 15N for cis-ax and cis-eq compounds are 12.7 and 9.9, re-
d
spectively, indicating that charge distribution is more important in
the axial series than in the equatorial. Linear correlations were also
performed for aniline carbons; however, only C-11 (Cipso to nitro-
gen) gave a good regression value (R w 0.98) with Rꢁ; the slopes of
In agreement with the conformation adopted by the molecules
in solution, the dioxaphosphinane ring, of the twelve crystalline
structures, is in the chair conformation. In the case of the equatorial
series; regardless of the electronic nature of the p-X-substituent,
the rotameric conformation of the anilino group (P–N bond
the curves Rꢁ/
dC-11 for the ax and eq series are closer to the
Table 4
Linear correlations between NMR parameters and Hammett constants
1–6 cis-ax compounds
1–6 cis-eq compounds
x
y
Ra
Slope
x
y
Ra
Slope
spꢁ
Rꢁ
spꢁ
Rꢁ
spꢁ
Rꢁ
d
d
d
d
d
d
31P
0.984
0.977
0.988
0.981
0.939
0.989
ꢁ2.145
ꢁ3.024
8.989
12.665
7.310
spꢁ
Rꢁ
spꢁ
Rꢁ
spꢁ
Rꢁ
d
d
d
d
d
d
31P
0.976
0.975
0.957
0.993
0.930
0.984
ꢁ0.858
ꢁ1.215
6.709
9.879
7.241
31P
31P
15N
15N
15N
15N
13C (C-11)
13C (C-11)
13C (C-11)
13C (C-11)
10.901
10.881
a
Regression coefficient values.

2070
Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
rotation) is in close proximity in all cases: uO–P–N–C11 spans a range
of approximately 37–71ꢀ and uS–P–N–H10 of 1–22ꢀ (Table 8, Suppl.
Mat.), provided that the hybridization of the nitrogen atom is nearly
sp2 (see below). The plane of the aromatic ring (N–C11 bond rota-
tion) tend to be parallel to the bisector plane of the dioxaphos-
phinane ring (uC12–C11–N–H10 w7–33ꢀ and uC12–C11–N–P w132–
178ꢀ); this rotameric conformation allows some delocalization of
The geometry around the phosphorus atom for the equatorial as
well as for the axial diastereomers, is tetrahedral; the sum of the
four angles around P is in the range of 439.38–440.19 degrees for
the ax series and 434.5–436.33 degrees for the eq series (Tables 5
and 6), giving rise to a mean tetrahedral angle of 109.9ꢀ for the ax
dioxaphosphinanes and108.8ꢀ for the eq series. On the other hand,
the sum of the angles around the nitrogen atom is close to 360ꢀ,
within the interval of 354.05ꢀ to 357.9ꢀ for the axial series (ex-
cluding cpd. cis-ax-2 whose value: 359.71ꢀ is off of the mean value)
and 357.3ꢀ to 360.8ꢀ for the equatorial series (excluding cpd. cis-eq-
6 whose value: 355.3ꢀ is off of the mean value), showing that the
geometry at the nitrogen atom is almost planar (mean trigonal
angle of 118.66ꢀ for the ax and 119.68ꢀ for the eq cpds., Tables 5
and 6). These findings are in agreement to those reported for
analogous thiophosphoramidates,14 where the nitrogen atom of the
equatorial isomers tends to be flatter than the nitrogen of the axial
ones. Consequently, the changes in nitrogen sp2 hybridization be-
tween the two series of compounds, are in favor of an increase in s-
character of the equatorial P–N bond with respect to the axial P–N
bond, that according to Stec rule30 accounts for the observed in-
the nitrogen lone pair, into the aromatic ring by
p-resonance;
considering that the ideal delocalized structure is expected to have
uC12–C11–N–H10¼0ꢀ and uC12–C11–N–P¼180ꢀ. The situation in the axial
series is similar, taking into account that all diastereomers have
close rotameric (P–N) conformations; the hydrogen of the N–H
group is pointing toward one of the endocylic oxygens [uO1(3)–P2–
w21–27ꢀ, uO3(1)–P2–N10–H10 w132–137ꢀ, uS9–P2–N10–H10
N10–H10
w94–105ꢀ] (Table 7, Suppl. Mat.). Even though this conformation
suggests the existence of an intramolecular [N–H/O] hydrogen
bond [N–O bond distance is around 2.51 Å (average value) against
2.75–3.0 Å obtained from the van der Waals radii sum],28 the [N–
H/O] geometry is not proper for the non-covalent interaction (82ꢀ,
average); thus, the H/O distance is 2.45 Å (average), far from the
observed 2.0 Å in the case of a hydrogen bond.29 Here again the
nitrogen has a nearly sp2 hybridization (see below), and the aro-
matic ring is located outside the dioxaphosphinane ring avoiding,
this way, the 1,3-syn axial steric compression with the axial hy-
drogen atoms at the C4 and C6 positions. The preferred rotational
N–C11 conformer of the axial compounds is such, that it is clearly
benefited of delocalization of the N-lone pair within the aromatic
1
crease in JPN coupling constant for the equatorial thiophosphor-
1
amidates 1–6 [i.e., 1JPN (eq) w25 Hz vs JPN (ax) w10 Hz, Table 2].
Interestingly, the geometrical differences in nitrogen hybridization
may also explain the deshielding of the equatorial 15N NMR signal
as compared to the axial [i.e., Dd15N(eq-ax), X: OMe¼3.3;
C6H11¼4.4; H¼3.7; Cl¼2.8; CN¼1.3; NO2¼0.2 Hz] since as it was
pointed out by Gray,24 the change in the pyramidal like (sp3) ni-
ring (uC12–C11–N–H10 w173–179ꢀ and
u
C12–C11–N–P w15–24ꢀ).
trogen to trigonal (sp2), is accompanied by
a
substantial
Table 5
Structural properties of cis-ax 1–6
Molecule A/B
1a
2a
3a
4a
5a
6a
Geometry at phosphorusa
439.89
439.92
356.83
355.91
112.25
112.37
52.16
51.78
0.61
440.01
359.71
112.16
52.50
0.61
439.89
439.96
356.60
357.90
112.49
112.41
51.30
51.68
0.63
440.19
440.08
356.22
356.68
112.54
112.62
51.17
50.68
0.63
439.38
439.46
Geometry at nitrogenb
Baeyer strainc
357.25
112.46
51.52
0.62
354.05
112.47
51.45
0.62
Pitzer straind
cos
u
0.62
0.62
0.63
ꢁcos
q/(1þcos
q)
0.61
0.62
0.61
0.62
0.62
0.62
0.63
0.62
0.62
a
Calculated as the sum of the bond angles (O1P2O3), [OP2N10 (mean)], [OP2S9 (mean)], and (S9P2N10) in deg.
Calculated as the sum of the bond angles (P2N10C11), (P2N10H10), and (C11N10H10) in deg.
Calculated as the average value of the bond angles (O1P2O3), (O1C6C5), (O3C4C5), (C4O3P2), (C6C5C4), and (C6O1P2) in deg.
b
c
d
Calculated as the average absolute value of the torsion angles (O1P2O3C4), (O1C6C5C4), (O3P2O1C6), (O3P2O1C6), (P2O3C4C5), and (P2O1C6C5) in deg. (For single angle
values, see Tables 5 and 7 in supplementary Data).
Table 6
Structural properties of cis-eq 1–6
Molecule A/B
1e
2e
3e
4e
5e
6e
Geometry at phosphorusa
434.50
436.09
358.15
358.63
111.19
111.68
55.50
54.09
0.57
435.77
435.75
435.74
360.80
359.20
110.86
111.56
56.59
54.66
0.55
436.0
435.78
357.50
357.30
111.13
111.80
54.59
54.62
0.58
436.33
435.82
Geometry at nitrogenb
Baeyer strainc
359.90
111.76
53.98
0.59
359.41
111.17
55.76
0.56
355.30
111.94
53.50
0.59
Pitzer straind
cos
u
0.59
0.58
0.58
ꢁcos
q/(1þcos
q)
0.57
0.58
0.59
0.55
0.58
0.56
0.59
0.56
0.60
a
Calculated as the sum of the bond angles (O1P2O3), [OP2N10 (mean)], [OP2S9 (mean)], and (S9P2N10) in deg. .
Calculated as the sum of the bond angles (P2N10C11), (P2N10H10), and (C11N10H10) in deg.
Calculated as the average value of the bond angles (O1P2O3), (O1C6C5), (O3C4C5), (C4O3P2), (C6C5C4), and (C6O1P2) in deg.
b
c
d
Calculated as the average absolute value of the torsion angles (O1P2O3C4), (O1C6C5C4), (O3P2O1C6), (O3P2O1C6), (P2O3C4C5), and (P2O1C6C5) in deg. (For single angle
values, see Tables 6 and 8 in supplementary data).

Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
2071
deshielding, at least in heterocyclic phoshoramidates. Besides; as it
3.4. Stereoelectronic effects
According to Deslongschamps,34 a stereoelectronic effect is ob-
was pointed out earlier, the SCS correlations found for nitrogen
atom, indicates that the chemical shifts are also sensible to elec-
tronic effects of the p-substituent transmitted not only by inductive
but also by resonance effects. This can also be observed by the fact
that the list of values of Dd15N(eq-ax) just mentioned above, is not
a constant value, but it tends to decrease as the electron with-
drawing characteristic of the p-substituent is increased. We will
come back to this point in the discussion below.
served in a molecule when the spatial disposition of a donor
electron pair (bonded or non bonded), is antiperiplanar to the ac-
ceptor * antibonding orbital. This effect is customarily described as
p
s
negative hyperconjugation or as generalized anomeric effect (term
used to denote the axial preference of an electronegative substituent
over the equatorial, in the anomeric position of a pyranose sugar
ring). On the other hand, the term reverse anomeric effect is used
when a cationic substituent at the anomeric position of a pyranose
sugar, shifts the equilibrium toward the equatorial anomer.7c,35
In the conformational analysis of highly crowded 2-dicyclo-
hexylamino-2-thio-1,3,2-dioxaphosphinanes (structure A, Fig. 1),
Summarized in Tables 5 and 6, are also the Baeyer and Pitzer
strain [calculated as the average of the internal bond angles (
torsion angles ( ) of the dioxaphosphorinane ring, respectively],
for all the crystalline compounds. The agreement between the
values obtained through the formulas cos and –cos
q), and
u
u
q
/(1þcos
q)
has been employed in our group,14,21 in analogous systems, as an
indication of the propensity of the dioxaphosphorinane ring in
chair conformation to establish the same compromise between
averaged ring torsional and angular strain that the one existing in
cyclohexane in chair conformation.31 As observed in Table 5, the
values of these parameters, obtained for the axial series are close to
0.62 with discrepancies of not more than ꢂ0.01 unity for each pair.
On the other hand; in the equatorial series (Table 6), both strain
values do not average to a single value, but to an interval of 0.55–
0.60; however, here again the discrepancies between every pair of
values are negligible (ꢂ0.01 to ꢂ0.02).
we have observed that negative hyperconjugation of the type n–s*,
is as important as the van der Waals steric compression, to drive the
conformational preference of the amino substituent, in a mobile
system, to occupy the equatorial position.14 However, in analogous
2-anilino-2-oxo-1,3,2-oxazaphosphinanes (structure B, Fig. 1),
where the amino substituent is clearly more electronegative, the
axial conformer is in preference over the equatorial; Bentrude
et al.,33a have interpreted the axial preference of the anilino sub-
stituent in terms of an endo-anomeric (n–s*) stereoelectronic in-
teraction. These results are in agreement with the stereoelectronic
interactions found in protonated glucosylanilines where a more
localized positive charge on nitrogen leads to a decrease in the
reverse anomeric effect.35 Therefore, we can deduce that negative
hyperconjugation is likely to be in operation not only in the
equatorial but also the axial series of anancomeric thiophosphor-
amidates (1–6). In this section, we examine the influence of the
electronic effect of the p-X-substituent on the geometry and con-
formation adopted by the molecules in solid state and in gas phase
through the geometries obtained by means of theoretical Kohn–
Sham DFT calculations.15 Considering that the stabilization of a ring
Taking into consideration that the Internal (I) strain in a het-
erocyclic molecule is modified by changes from sp3–sp2 hybrid-
ization of one or more than one of the atoms in the ring,32 we
consider pertinent to analyze the differences in Baeyer strain be-
tween the eq and ax compounds taking as reference the tetrahedral
angle (109.5ꢀ) and 60ꢀ for the minimum Pitzer strain. Substituting
these values in the strain formulas cos
u
and ꢁcos
q/(1þcos
q), both
become 0.5; thus we conclude that the equatorial diastereomers
have less geometrical ring strain, than the axial ones, since the
strain values of the eq compounds are closer to 0.5. Thus, Baeyer
and Pitzer cos values by themselves can give a clue on the origin of
the conformational effects. It is important to remark, that when the
Baeyer and Pitzer strain for a molecule compromise within them-
selves, it can be assumed that the strain of the molecule is mini-
mized in chair conformation. However, if this is not the case, the
ring conformation is distorted to a twist or boat as in highly
crowded thiophosphoramidates.14
conformation by anomeric n–s* effect, implies the participation of
a double bond–no–bond structure in resonance with its normal
lewis formula (see examples in Scheme 2), we certainly expect to
have shortening or lengthening of vicinal phosphorus bonds, along
with an opening of the COP bond to almost 120ꢀ.
On the other hand, in some cases an anilino group in phos-
phorus containing heterocycles, prefers an axial conformation;33
therefore, a simple analysis based on ring strain is not sufficient to
explain the observed conformational effects. Indeed, it has been
demonstrated, several times, that the conformation adopted by
1,3,2-dioxaphosphinanes is highly dependent on stereoelectronic
effects as we have already mentioned in the introduction of this
work. We will address this point in the next section.
Figure 1. Negative hyperconjugation in six-membered ring phosphorus compounds.
On the other hand, we found that the dioxaphosphinane ring is
an irregular chair, since the OPO region tend to be flat as a result of
the nearly planar sp2 hybridization of the oxygen atoms [ax qCOP
(average), X: OMe 118.42ꢀ, C6H11 117.88ꢀ, H 118.76ꢀ, Cl 118.92ꢀ, CN
118.43ꢀ, NO2 118.48ꢀ; eq qCOP (average), X: OMe 117.0ꢀ, C6H11
118.29ꢀ, H 116.5ꢀ, Cl 117.15ꢀ, CN 116.46ꢀ, NO2 118.15ꢀ]. As observed,
with the exception of the thiophosphoramidate 2 (X¼C6H11), the
COP bond angles are larger for the axial than for the equatorial
compounds; this tendency could be explained invoking a possible
1,3-syn axial compression of the aniline group with the hydrogens
at C4 and C6 in the axial isomers; however, the fact that the ten-
dency is reverted for the most crowded compound 2, is not sup-
porting this hypothesis. Alternatively, the opening of the COP bond
angle for the ax isomers can be explained through the participation
The chair conformation adopted by the axial and equatorial thio-
phosphoramidates 1–6 in gas phase, was not destabilized by modi-
fying the electronic nature of the p-X-substituent; however, variation
in geometry, specifically bond lengths of vicinal P–O, P–N and P–S
bonds, indeed varied periodically with the change of the X group (see
Table 7); giving support to the participation of hyperconjugation, as
the main mechanism of stabilization of thiophosphoramidates.
As it was mentioned before the hybridization on oxygens is closer
to sp2 than to sp3,36 and as observed in Table 7, the endocyclic O1–P
bond lengths are smaller in the axial series than in the equatorial, but
the opposite trend is found for the N–P bond lengths. These ten-
dencies are in agreement with the shortening and lengthening
expected for the endo-anomeric npO–
the cis-ax series, and the exo-anomeric npN–
erating in cis-eq series, as summarized in Scheme 5.
s*P–N interaction operating in
of the stabilizing stereoelectronic npO–
s
*
effect, as discussed
s*
P–O
interaction op-
P–NAr
below.

2072
Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
Table 7
a resonance effect. From a comparison of the regression values of
Hammett type linear correlations with the constants (sꢁp or Rꢁ) and
Selected bond distances (Å) for cis-ax 1–6 and cis-eq 1–6 at B3LYP level of theory
with the basis set 6-31G (d,p)
d
13C (C-11) (Table 4); we deduced, that the resonance effect (Rꢁ)
O1-P
O3-P
N-P
P-S
N-C11
may have more importance than the field/inductive effect (taken
into consideration in sꢁp , as a combined effect) in the propagation of
the electronic density from a donor substituent to any of these
nuclei; or vice versa, for an attractor substituent. Taking this into
account, and considering that stereoelectronic interactions are
largely dependent on charge distribution of polar groups in the
molecule, we analyze the resonance effect of the p-substituent
(donor or acceptor) in the stabilization by negative hyper-
OMe ax
OMe eq
C6H11 ax
C6H11 eq
H ax
H eq
Cl ax
Cl eq
CN ax
1.6244
1.6395
1.6230
1.6316
1.6226
1.6315
1.6219
1.6272
1.6197
1.6281
1.6194
1.6258
1.6245
1.6292
1.6245
1.6279
1.6238
1.6278
1.6224
1.6272
1.6204
1.6256
1.6196
1.6257
1.7079
1.6641
1.7090
1.6643
1.7093
1.6654
1.7127
1.6685
1.7180
1.6740
1.7208
1.6770
1.9274
1.9452
1.9269
1.9448
1.9264
1.9441
1.9253
1.9426
1.9237
1.9401
1.9224
1.9392
1.4226
1.4262
1.4187
1.4196
1.4172
1.4186
1.4143
1.4163
1.4054
1.4058
1.4022
1.4014
conjugation (n–
1–6, through the transquinoidal mesomeric structures C and D that
are the result of the -conjugation of the nitrogen electronic charge
s*) of axial and equatorial thiophosphoramidates
CN eq
NO2 ax
NO2 eq
p
into the aromatic ring in the two tautomeric structures A4B
depicted in Figures 2 and 3.
Scheme 5.
The averaged values of the P–O and P–N bond distances in the
X-ray structures (Tables 3 and 4, Suppl. Mat.) are smaller than the
theoretical ones; however, they follow the same trend: ax series
[(P–O) 1.5661ꢁ1.5821 Å, and (P–N) 1.6479ꢁ1.6632 Å]; eq series [(P–
O) 1.561ꢁ1.587 Å, and (P–N) 1.619ꢁ1.643 Å]. The small differences
in P–O bond lengths between the ax and eq isomers are perhaps due
to the participation of the second stabilizing anomeric interaction:
npO–s
*
in operation in the eq isomers (Scheme 6). This in-
P]S,
teraction is also supported by the lengthening of the P–S bond of the
equatorial compounds as compared to the axial [(P–S)ax 1.9224
ꢁ1.9274 Å vs (P–S)eq 1.9392 ꢁ1.9452 Å] (Table 7). It is worth of
mention that similar endo-anomeric npO–
s
*
P]x (X¼O, S) effect has
been proposed as plausible, although not strong stabilizing in-
Figure 2. Resonance effects of p-substituents and negative hyperconjugation (axial
series).
teraction, in analogous 2-substituted-1,3,2l
5-dioxaphosphinanes.8
In the axial series (Fig. 2), structures C and D (for an acceptor
group), certainly stabilize the hyperconjugative A and B structures.
However, C and D (for a donor group) do not accomplish the same
goal, because neither structure C nor D are stabilized; in C the
phosphorus atom can not bear the lone pair in axial position be-
cause the vicinal oxygens have lone pairs in antiperiplanar re-
lationship to the first, which is highly unstable; whereas in D there
are two negative charges in neighboring atoms. In short, because of
Scheme 6.
the participation of the endo-anomeric npO–
s
*
interaction, the
P–N
The influence of the p-substituent at the aniline ring, in the
stabilization of the anancomeric thiophosphoramidates 1–6, can be
analyzed through two factors: (a) a field/inductive effect, and (b)
mesomeric C and D structures help to stabilize the axial (p-W-
anilino) substituents but not the axial (p-Y-anilino) substituents;
and this hypothesis can be proved through the extent at which

Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
2073
pair with the
p-electrons of the aromatic aniline ring. This conju-
gation is more important in the ax series than in the eq series, as
confirmed by the larger differences found in N–C11 bond length
between cpd. cis-ax-6 with an acceptor (NO2) group and cpd. cis-
ax-1 with a donor (OMe) substituent as compared with the corre-
sponding cis-eq compounds [ax: (NO2) 1.396 Å vs (OMe) 1.417 Å
(average); eq: (NO2) 1.426 Å (OMe) 1.424 Å (average)] (Tables 11
and 12, Suppl. Mat.).
Besides, the electronic behavior of the aniline ring carbons (C-11
to C-14) was analyzed through the linear correlations between
d
13C
of the thiophosphoramidates 1–6 and the
d
13C of the p-substituted
anilines (Table 9, Suppl. Mat.), whose chemical shifts were de-
termined in this work and are summarized in Table 3. As observed
in Table 9, slight deviations of an ideal linear correlation (slope¼1),
were observed for C-11 and C-13 chemical shifts for both series of
compounds. Carbon-11 of the thiophosphoramidates is shifted to
low frequencies of the corresponding anilines by around 8 ppm
(Table 3), indicating that the dioxaphosphinane ring behaves as an
electron donating inductive group toward the aniline moiety. The
regression coefficient for the curve
series (w 0.85) and the corresponding for
d
13C12/
d
13C12’ is poor for both
13C14’ is not very
d
13C14/
d
good (w 0.96) (Table 9, Suppl. Mat.); the inductive effect of the
substituent (electronegativity of the atom or group directly at-
tached), is perhaps predominant in C-14.
Finally, we observed that the endo-anomeric npO–
s*
in-
P–N
teraction not only explains the structural changes (bond lengths)
observed for the axial thiophosphoramidates but also the COP bond
opening expected for the axial, as compared to the equatorial series
[i. e. averaged COP angles: OMe (ax) 118.41 vs (eq) 117.0; NO2 (ax)
118.48 vs (eq) 118.15]. In addition, this stereoelectronic interaction
also gives support to the compensation of phosphorus charge from
the endocyclic oxygens proposed to explain the reversal tendency
observed for 31P NMR chemical shifts in 1,3,2-dioxaphosphinanes.21
Figure 3. Resonance effects of p-substituents and negative hyperconjugation (equa-
torial series).
shortening of the endocyclic P–O bonds and lengthening of the
exocyclic P–N bond is taking place. Indeed, we found shorter P–O
bonds accompanied with longer P–N bonds for thiophosphor-
amidates with electron withdrawing (W) substituents than for
those with electron releasing (Y) substituents [i.e., P–O bond
length: (NO2) 1.5708 Å (average) vs (OMe) 1.5757 Å (average); P–N
bond length (NO2) 1.6632 Å vs (OMe) 1.6493 Å (average) (Table 3,
Suppl. Mat.), and the same tendency can be observed in theoretical
bond lengths in Table 7]. On the other hand, in the equatorial series
(Fig. 3) an acceptor W group is not of help for the exo-anomeric
4. Conclusions
The conformational analysis of diastereomeric cis-ax and cis-eq
2-p-X-anilino-2-thio-4,6-dimethyl-1,3,2-l
5-dioxaphosphorinanes
(1–6) was carried on in solution and in solid state. Regardless of the
p-X-subtituent, it was found that the preferred conformation was
a chair in both series. The dioxaphosphinane ring is an irregular
chair; the oxygens at the OPO region are almost planar and the
phosphorus atom is a puckering end. The ability of the endocyclic
oxygens to change hybridization from sp3 to sp2 is higher in the
axial series than in the eq series; as a consequence, the steric
compression caused by 1,3-syn axial interaction of the anilino
substituent, with H4,6 is practically negligible. The geometry around
the nitrogen atom, for both series of diastereomers, is trigonal;
however, the nitrogen of the equatorial isomers tends to be flatter
than the nitrogen of the axial ones. This finding accounts for the
npN–
pair on the nitrogen atom needed for the stereoelectronic in-
teraction is distracted in the conjugation with the -electron
s*P–O interaction because of two reasons: one is that the lone
p
density of the aniline ring (structure C); and the other is, that the
mesomeric structure D is not stable because it has two positive
charges in neighboring atoms. Nevertheless, a donor Y group in-
deed stabilizes the structures A and B by resonance with C and D,
respectively; because in C, the lone pair on phosphorus in the
equatorial position is stabilized by gauche effect;37 and in D, the
positive charge is delocalized in the aromatic ring. Here again,
the shortening of the exocyclic P–N bond and lengthening of the
endocyclic P–O bond for the thiophosphoramidates with donor
substituents as compared to the acceptors, can be taken as a proof
of the participation of the exo-anomeric effect [i.e., P–N bond
length: (OMe) 1.621 Å (average) vs (NO2) 1.640 Å P–O bond length
(OMe) 1.5814 Å (average) vs (NO2)1.583 Å (average) (Table 8, Suppl.
Mat.)]. Although the P–O bond length difference in X-ray is not very
clear; the theoretical values sustain the hypothesis [P–N bond
length: (OMe) 1.6641 Å vs (NO2) 1.6770 Å P–O bond length (OMe)
1.6344 (average) Å vs (NO2)1.6258 Å (average) (Table 7)]. From the
above results, one can confirm that the p-X-substituent have the
main electronic influence in the negative hyperconjugation stabi-
lization of the dioxaphosphinane ring (in both ax and eq configu-
rational isomers 1–6), through the conjugation of the nitrogen lone
1
observed increase in JPN coupling constants, for the eq series
(Stec rule).
Substituent-Induced Chemical Shifts were evaluated and
a comparison of the slope values for the correlations of Rꢁ (or spꢁ
)
versus d d d
15N, 31P and 13C led to the conclusion that the electronic
effect is sensed by the nitrogen atom and by C-11 more effectively
than by the phosphorus atom. Meanwhile 15N and 13C NMR shifts
have normal SCS trends, a reversal trend was found for correlations
of d d
31P/sꢁp or 31P/Rꢁ, this reversibility can be explained by charge
density transfer from oxygens toward the phosphorus nuclei.
The participation of the stabilizing hyperconjugative endo-
anomeric effect [npO–
anomeric effect [npN–
s
s
*
*
]
]
for the axial isomers and exo-
for the equatorial isomers was
P–N
P–O
confirmed by the expected shortening and lengthening of vicinal
phosphorus bonds observed in the solid state structures and in the
minimized ground state structures, obtained by DFT calculations.

2074
Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
The preferred rotational N-C11 conformer is such, that the nitrogen
lone pair is conjugated with the -electrons of the aromatic ring,
contributing to polarization and charge redistribution by resonance
effect; this effect is somewhat more important in the axial, than in
the equatorial series.
The electronic influence of the p-substituent in the stabilization
of the molecules by negative hyperconjugation, was analyzed
through the transquinoidal mesomeric structures that result of
a white solid (trietrylammonium chloride). The solid was filtered
off through a filter tipped cannula and the filtrate added to
a 500 mL flask, equipped with stir bar and reflux condenser,
charged previously with 0.77 g (24 mmol) of elemental sulfur. The
mixture was heated under reflux for 12 h. After cooling, the sus-
pension was concentrated under vacuum. Purification of the crude
product was accomplished by flash chromatography.
p
p-conjugation in the Lewis and double bond–no–bond structure.
5.2.1. ax-2-p-Methoxyanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-ax-1). Mp 190–191 ꢀC. Isolated yield (19.5%). 1H
NMR 1.42 (dd, 6H),1.79 (dq,1H),1.92 (dt,1H), 3.77 (s, 3H), 4.64 (m,
l
5-dioxa-
An electron withdrawing substituent contributes to the stabiliza-
tion of the axial isomers by the endo-anomeric effect and an elec-
tron donating group to the stabilization of the equatorial isomers by
the exo-anomeric effect.
d
2
3
2H), 4.74 (d, JHH¼7.7, NH), 6.83 (d, JHH¼8.8, 2H), 7.01 (dd,
3JHH¼8.8, 4JHP¼0.5, 2H); 13C NMR
d 22.4 (d, C7,8), 40.5 (d, C5) 55.6 (s,
OMe), 76.0 (d, C4,6); 114.8 (s, Cmeta), 120.6 (d, Cortho), 132.5 (s, Cipso),
5. Experimental part
5.1. General
155.6 (s, Cpara); 31P NMR
d
60.60; 15N NMR
d
ꢁ298.64 (d). MS (m/z)
287 (Mþ), 255 (Mþꢁ32), 219 (Mþꢁ68), 201 (Mþꢁ86), 165
(Mþꢁ122), 154 (Mþꢁ133), 122 (Mþꢁ165), 95 (Mþꢁ192), 69
(Mþꢁ218), 41 (Mþꢁ246). Anal. Calcd for C12H18O3NPS: C, 49.99%; H,
6.59%. Found: C, 50.20%; H, 6.50%.
Melting points are uncorrected. The 1H, 13C, 31P, and 15N NMR
spectra were acquired on a JEOL Eclipse spectrometer at 400, 100.5,
161.8, and 40.5 MHz correspondingly, using CDCl3 as solvent, at
27 ꢀC, except in the case of the 15N spectra of the ax-cis series, which
were recorded in DMSO-d6 at 75 ꢀC in order to improve their sol-
5.2.2. eq-2-p-Methoxyanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-eq-1). Mp 101–102 ꢀC. Isolated yield (8.5%). 1H
NMR 1.35 (dd, 6H), 1.58 (dt, 1H),1.76 (dq, 1H), 3.78 (s, 3H), 4.84 (m,
l
5-dioxa-
d
2
3
ubility. Chemical shifts (d) are reported in ppm and coupling con-
2H), 5.24 (d, JHH¼14.7, NH), 6.81 (d, JHH¼8.9, 2H), 7.04 (dd,
stants (J) in Hertz. The 31P chemical shifts are reported in parts per
million downfield (þ) from 85% H3PO4 used as external standard.
The 15N spectra were measured with an INEPT pulse sequence and
using CH3NO2 as reference. Mass spectra (EI) were measured on
a Hewlett Packard 5989 A spectrometer using electron impact (EI)
at 70 eV. The stereoisomeric mixture of 2,4-pentanodiol (meso:d/l in
a 55:45 ratio) was obtained from the reduction of 2,4-pentanedione
with sodium borohydride, as reported by Pritchard and Vollmer.38
Compounds cis-ax 1–6 and cis-eq 1–6 were synthesized in oven-
dried glassware under nitrogen atmosphere. Solvents and solutions
were transferred by syringe-septum and cannula techniques. Sol-
vents for the reactions, diethyl ether and toluene, were reagent
grade and were dried and distilled from sodium/benzophenone
immediately before use. Triethylamine was dried and distilled from
LiAlH4. Products were purified by flash column chromatography on
silica gel 230–400 mesh, using a mixture of AcOEt/hexanes (1:9) as
eluent. Yields are given for isolated products. Mixtures of AcOEt/
hexanes or CH2Cl2/hexanes (1:3) were used for recrystallization.
Crystallographic work was performed in an Enraf–Nonius Kappa
CCD difractometer. Data collection Collect (Nonius BV, 1997–2000).
Data reduction: Win GX,39 solved by direct methods SHELXS9740
and refined with SHELXL-9741: Molecular graphics: Diamond 2.1e
(crystal impact 2003) and dihedral angles: Win GX.39 Ab initio
calculations were performed employing Gaussian 98.15
3JHH¼8.9, 4JHP¼0.7, 2H); 13C NMR
d 22.1 (d, C7,8), 41.1 (d, C5) 55.5 (s,
OMe), 73.8 (d, C4,6); 114.3 (s, Cmeta), 122.6 (d, Cortho), 131.4 (s, Cipso),
156.2 (s, Cpara); 31P NMR
d
63.46; 15N NMR
d
ꢁ295.34 (d). MS (m/z)
287 (Mþ), 254 (Mþꢁ33), 219 (Mþꢁ68), 201 (Mþꢁ86), 170
(Mþꢁ117), 154 (Mþꢁ133), 95 (Mþꢁ192), 69 (Mþꢁ218), 41
(Mþꢁ246). Anal. Calcd for C12H18O3NPS. C, 49.99%; H, 6.59%. Found:
C, 50.10%; H, 6.41%.
5.2.3. ax-2-p-Cyclohexylanilino-2-thio-cis-4,6-dimethyl-1,3,2
oxaphosphinane (cis-ax-2). Mp 200 ꢀC. Isolated yield (1.5%). 1H
NMR 1.23 (m, 1H), 1.34–1.48 (m, 4H), 1.41 (dd, 6H), 1.74 (m, 1H),
l
5-di-
d
1.79 (dq, 1H), 1.79–1.96 (m, 4H), 1.92 (dt, 1H), 2.94 (tt, 1H), 4.65 (m,
2H), 4.83 (d, 2JHP¼6.8, NH), 6.97 (d, 3JHH¼8.5, 2H), 7.11 (d, 3JHH¼8.5,
2H); 13C NMR
d 22.4 (d, C7,8), 26.2 (s, CH2), 26.9 (s, CH2), 34.6 (s,
CH2), 40.5 (d, C5), 43.8 (s, CH), 76.0 (d, C4,6), 118.2 (d, Cortho), 127.8 (s,
Cmeta), 137.0 (s, Cipso), 142.3 (s, Cpara); 31P NMR
d
130 (Mþꢁ209), 92 (Mþꢁ247), 69 (Mþꢁ270), 41 (Mþꢁ298). Anal.
Calcd for C17H26O2NSP: C, 60.15%; H, 7.72. Found: C, 60.20%; H,
8.02%.
d
59.95; 15N NMR
ꢁ295.50 (d). MS (m/z): 339 (Mþ), 271 (Mþꢁ68), 228 (Mþꢁ111),
5.2.4. eq-2-p-Cyclohexylanilino-2-thio-cis-4,6-dimethyl-1,3,2
oxaphosphinane (cis-eq-2). Mp 145 ꢀC. Isolated yield (50.2%). 1H
l
5-di-
NMR
d 1.33–1.50 (m, 5H), 1.37 (dd, 6H), 1.69 (dt, 1H), 1.71–1.89 (m,
5H), 1.81 (dq, 1H), 2.44 (tt, 1H), 4.87 (m, 2H), 5.37 (d, 2JHP¼14.9, NH),
3
3
5.2. General procedure for the preparation of 2-p-X-anilino-
6.97 (d, JHH¼8.4, 2H), 7.09 (d, JHH¼8.4, 2H); 13C NMR
d 22.2 (d,
2-thio-4,6-dimethyl-1,3,2
l
5-dioxaphosphorinanes cis-ax (1–6)
C7,8), 26.2 (s, CH2), 26.9 (s, CH2), 34.6 (s, CH2), 41.2 (d, C5), 43.8 (s,
CH), 74.0 (d, C4,6), 119.3 (d, Cortho), 127.5 (s, Cmeta), 136.3 (s, Cipso),
and cis-eq (1–6)
142.9 (s, Cpara); 31P NMR
d
63.02; 15N NMR
d
ꢁ291.38 (d). MS (m/z):
In a three-necked 2.0 L flask, fitted with dropping funnel, stir
bar, and rubber septa, were placed 2.5 g (24 mmol) of 2,4-penta-
nodiol (mixture of stereoisomers) and 100 mL of dry diethyl ether.
The solution was stirred at 0 ꢀC and 2.1 mL (24 mmol) of phos-
phorus trichloride in 30.0 mL of diethyl ether were added slowly
through the dropping funnel. The resulting solution was evapo-
rated in a rotary evaporator and the concentrate transferred via
cannula to a dropping funnel that contained 50 mL of dry toluene.
The funnel was previously set in another 500 mL three-necked
flask, equipped with stir bar and rubber septa, which contained
24 mmol of the corresponding p-X-substituted aniline and 3.4 mL
(24 mmol) of triethylamine in 150 mL of dry toluene. The solution
in the funnel was added dropwise, resulting in precipitation of
339 (Mþ), 306 (Mþꢁ33), 271 (Mþꢁ68), 253 (Mþꢁ86), 228
(Mþꢁ111), 202 (Mþꢁ137), 175 (Mþꢁ164), 140 (Mþꢁ199), 132
(Mþꢁ207), 69 (Mþꢁ270), 43 (Mþꢁ296). Anal. Calcd for
C17H26O2NSP: C, 60.15%; H, 7.72. Found: C, 60.60%; H, 7.95%.
5.2.5. ax-2-Anilino-2-thio-cis-4,6-dimethyl-1,3,2
nane (cis-ax-3). Mp 204–205 ꢀC. Isolated yield (15.0%). 1H NMR
l
5-dioxaphosphi-
d
1.44 (dd, 6H), 1.81 (dq, 1H), 1.93 (dt, 1H), 4.65 (m, 2H), 4.88 (d,
2JHP¼7.5, NH), 7.02 (m, 1H), 7.05 (m, 2H), 7.28 (m, 2H) 13C NMR
d
22.4 (d, C7,8), 40.5 (d, C5), 76.0 (d, C4,6), 118.2 (d, Cortho), 122.4 (s,
Cpara), 129.4 (s, Cmeta), 139.6 (s, Cipso); 31P NMR
d
59.43; 15N NMR
d
ꢁ293.76 (d). MS (m/z) 257 (Mþ), 225 (Mþꢁ32), 216 (Mþꢁ41), 189
(Mþꢁ68), 155 (Mþꢁ102), 124 (Mþꢁ133), 97 (Mþꢁ160), 93

Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
2075
(Mþꢁ164), 69 (Mþꢁ188), 41 (Mþꢁ216). Anal. Calcd for
(Mþꢁ68), 187 (Mþꢁ115), 170 (Mþꢁ132), 138 (Mþꢁ164), 69
(Mþꢁ233), 41 (Mþꢁ261). Anal. Calcd for C11H15O4N2SP: C, 43.70%;
H, 5.00%. Found: C, 43.59%; H, 5.16%.
C11H16O2NPS: C, 51.35%; H, 6.27%. Found: C, 51.17%; H, 6.58%.
5.2.6. eq-2-Anilino-2-thio-cis-4,6-dimethyl-1,3,2
phrinane (cis-eq-3). Mp 114–115 ꢀC. Isolated yield (37.5%). 1H NMR
l
5-dioxaphos-
5.2.12. eq-2-p-Nitroanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-eq-6). Mp 133–136 ꢀC. Isolated yield (13.6%). 1H
l
5-dioxa-
d
1.38 (dd, 6H), 1.71 (dt, 1H), 1.82 (dq, 1H), 4.88 (m, 2H), 5.44 (d,
2JHP¼13.9 Hz, NH), 7.04 (m, 3H), 7.28 (m, 2H); 13C NMR
d
22.2 (d,
NMR d 1.41 (dd, 6H),1.78 (dt,1H),1.91 (dq,1H), 4.89 (m, 2H), 6.04 (d,
3
3
C7,8), 41.2 (d, C5), 74.0 (d, C4,6), 119.4 (d, Cortho), 123.0 (s, Cpara), 129.1
2JHP¼13.0, NH), 7.13 (d, JHH¼9.1, 2H), 8.17 (d, JHH¼8.16, 2H); 13C
(s, Cmeta), 138.7 (d, Cipso); 31P NMR
d
62.94; 15N NMR
d
ꢁ290.06 (d).
NMR d 22.3 (d, C7,8), 41.3 (d,C5), 74.5 (d, C4,6), 118.2 (d, Cortho), 125.3
MS (m/z) 257 (Mþ), 224 (Mþꢁ33), 215 (Mþꢁ42), 189 (Mþꢁ68), 160
(Mþꢁ97), 140 (Mþꢁ117), 94 (Mþꢁ163), 93 (Mþꢁ164), 69
(Mþꢁ188), 41 (Mþꢁ216). Anal. Calcd for C11H16O2NSP: C, 51.35%; H,
6.27%. Found: C, 51.60%; H, 6.33%.
(s, Cmeta), 142.8 (s, Cpara), 144.9 (d, Cipso); 31P NMR 61.92; 15N NMR
d
d
ꢁ283.60 (d). MS (m/z) 302 (Mþ), 269 (Mþꢁ33), 234 (Mþꢁ68), 216
(Mþꢁ86), 185 (Mþꢁ117), 138 (Mþꢁ164), 108 (Mþꢁ194), 69
(Mþꢁ233), 41 (Mþꢁ261). Anal. Calcd for C11H15O4N2SP: C, 43.70%;
H, 5.00%. Found: C, 43.92%; H, 5.02%.
5.2.7. ax-2-p-Chloroanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-ax-4). Mp 245–246 ꢀC. Isolated yield 18.9%. 1H
l
5-dioxa-
Supplementary data
NMR d 1.43 (dd, 6H), 1.80 (dq, 1H), 1.92 (dt, 1H), 4.62 (m, 2H), 4.8 (d,
2JHP¼7.4, NH), 6.92 (d, 3JHH¼8.7, 2H), 7.17 (d, 3JHH¼8.7, 2H), 13C NMR
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tet.2010.01.034.
d
22.4 (d, C7,8), 40.4 (d, C5), 76.2 (d, C4,6), 119.5 (d, Cortho), 129.4 (s,
Cmeta), 138.1 (s, Cipso); 31P NMR
d
59.43; 15N NMR
d
ꢁ293.67 (d). MS
(m/z) 291 (Mþ), 223 (Mþꢁ68), 205 (Mþꢁ86), 170 (Mþꢁ121), 158
(Mþꢁ133), 127 (Mþꢁ164), 99 (Mþꢁ192), 69 (Mþꢁ222), 41
(Mþꢁ250). Anal. Calcd for C11H15O2NSPCl: C, 45.29%; H, 5.18.
Found: C, 45.42%; H, 5.26%.
References and notes
1. (a) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of Organic Compounds;
Wiley: New York, NY, 1994, Chapter 11; (b) Kellie, G. M.; Ridell, F. G. Top. Ster-
eochem. 1974, 8, 225.
2. (a) Eliel, E. L.; Knoeber, M. C. J. Am. Chem. Soc. 1968, 90, 3444; (b) Ridell, F. G.;
Robinson, J. T. Tetrahedron 1967, 23, 3417; (c) Pihlaja, K. Acta Chem. Scand. 1948,
22, 716.
5.2.8. eq-2-p-Chloroanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-eq-4). Mp 97–99 ꢀC. Isolated yield 6.7%. 1H NMR
l
5-dioxa-
3. Excellent reviews about this topic can be found in: (a) Bentrude, W. G. In Steric
and Stereoelectronic Effects in 1,3,2-Dioxaphosphorinanes in Conformational Be-
havior of Six-Membered Rings: Analysis, Dynamics and Stereoelectronic Effects;
Juaristi, E., Ed.; VCH: New York, NY, 1995; (b) Maryanoff, B. E.; Hutchins, R. O.;
Maryanoff, C. A. Top. Stereochem. 1979, 11, 187; (c) Gorenstein, D. G. Chem. Rev.
1987, 87, 1049.
4. (a) Ref. 3a, and 3b and references cited there. (b) Bentrude, W. G.; Tan, H.-W. J.
Am. Chem. Soc. 1973, 95, 4666; (c) Van Nuffel, P.; Van Alsenoy, C.; Lenstra, A. T.
H.; Geise, H. J. J. Mol. Struct. 1984, 125, 1.
5. (a) Majoral, J. P.; Bergounhou, C.; Navech, J. Bull Soc. Chim. Fr. 1973, 3146; (b)
Brault, J.-F.; Majoral, J. P.; Savignac, P.; Navech, J. Bull Soc. Chim. Fr. 1973, 3149.
6. (a) Stewart, E. L.; Nevins, N.; Allinger, N. L.; Bowen, J. P. J. Org. Chem. 1999, 64,
5350; (b) Arshinova, R. P. Russ. Chem. Rev. 1984, 53, 351; (c) Cramer, C. J. J. Mol.
Struct. (THEOCHEM) 1996, 370, 135.
7. For a further discussion on orbital hyperconjugation see: (a) Thatcher, G. R. J.
The Anomeric Effect and Associated Stereoelectronic Effects; American Chemical
Society: Washington, DC, 1992; Vol. 539; (b) Juaristi, E.; Cuevas, G. Tetrahedron
1992, 48, 5019; (c) Kirby, A. J. The Anomeric Effect and Related Stereoelectronic
Effects at Oxygen; Springer: Heidelberg, Germany, 1983.
d
1.37 (dd, 6H), 1.67 (dt, 1H), 1.82 (dq, 1H), 4.87 (m, 2H), 5.44 (d,
3
3
2JHP¼13.4, NH), 7.0 (d, JHH¼8.8, 2H), 7.23 (d, JHH¼8.8, 2H); 13C
NMR
d 22.1 (d, C7,8), 41.1 (d, C5), 74.1 (d, C4,6), 121.0 (d, Cortho), 128.2
(s, Cpara), 129.0 (s, Cmeta), 137.5 (d, Cipso); 31P NMR 62.87; 15N NMR
d
d
ꢁ290.88 (d). MS (m/z) 291 (Mþ), 223 (Mþꢁ68), 205 (Mþꢁ86), 170
(Mþꢁ121), 158 (Mþꢁ133), 127 (Mþꢁ164), 99 (Mþꢁ192), 69
(Mþꢁ222), 41 (Mþꢁ250) Anal. Calcd for C11H15O2NSPCl: C, 45.29%;
H, 5.18. Found: C, 44.96%; H, 6.61%.
5.2.9. ax-2-p-Cyanoanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-ax-5). Mp 113–116 ꢀC. Isolated yield (22.4%). 1H
l
5-dioxa-
NMR
d 1.39 (dd, 6H), 1.81 (dq, 1H), 1.90 (dt, 1H), 4.62 (m, 2H), 5.85
(d, NH), 7.14 (d, 3JHH¼8.4, 2H), 7.54 (d, 3JHH¼8.4, 2H); 13C NMR
d 22.3
(d, C7,8), 40.4 (d, C5), 75.6 (d, C4,6), 105.6 (s, Cpara), 118.0 (d, Cortho),
118.9 (s, CN), 133.5 (s, Cmeta), 144.3 (s, Cipso); 31P NMR
d
57.35; 15N
8. Ref. 3a pp 277–278. (b) Verkade, J. G. Phosphorus Sulfur 1976, 2, 251.
9. Muiphy, P. C.; Tebby, J. C. dp–pp Bonding Effects on 31P NMR Chemical Shift of
N-Aryliminophosphoranes In Phosphorus Chemistry, Proceedings of the 1981
International Conference; Quin, L. D., Verkade, J. G., Eds.; ACS Symposium Series;
ACS: Washington, D.C., 1981; Vol. 171, p 573.
NMR
d
ꢁ285.80 (d). MS (m/z) 282 (Mþ), 255 (Mþꢁ27), 241
(Mþꢁ41), 214 (Mþꢁ68), 179 (Mþꢁ103), 149 (Mþꢁ133), 149
(Mþꢁ133), 118 (Mþꢁ164), 90 (Mþꢁ192), 69 (Mþꢁ213), 41
(Mþꢁ241). Anal. Calcd for C12H16O2N2SP: C, 51.06%; H, 5.36% N,
9.92%. Found: C, 50.53%; H, 5.31%, N, 9.90%.
10. Gilheany, D. G. Chem Rev. 1994, 94, 1339.
´
´
11. Dobado, J. A.; Martınez-Garcıa, H.; Molina-Molina, J.; Sundberg, M. R. J. Am.
Chem. Soc. 1998, 120, 8461.
12. Stec, W. J.; Okrusek, A. J. Chem. Soc. Perkin 1 1975, 1828.
13. Bentrude, W.; Setzer, W. N.; Newton, M.; Meehan, E. J.; Ramli, E.; Khan, M.;
5.2.10. eq-2-p-Cyanoanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-eq-5). Mp 174–175 ꢀC. Isolated yield (1.5%). 1H
l
5-dioxa-
Ealick, S. Phosphorus, Sulfur Silicon 1991, 57, 25.
14. (a) Domı´nguez, Z. J.; Cortez, M. T.; Gordillo, B. Tetrahedron 2001, 57, 9799; (b)
Hommer, H.; Domı´nguez, Z.; Gordillo, B. Acta Crystallogr. 1998, C54, 832.
15. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G.; Robb, M. A.; Cheeseman,
J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Stain, M. C.; Farkas, O.;
Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, A. M.; Peng, A. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnsons, B.;
Chen, W.; Wong, M. W.; Andress, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98; Gaussian,: Pittsburgh PA, 1998.
NMR
d 1.36 (dd, 6H), 1.75 (dt, 1H), 1.87 (dq, 1H), 4.89 (m, 2H), 6.29
(d, 2JHP¼13.4, NH), 7.12 (d, 3JHH¼8.6, 2H), 7.52 (d, 3JHH¼8.6, 2H); 13
C
NMR
d 22.5 (d, C7,8), 41.6 (d, C5), 74.9 (d, C4,6), 105.6 (s, Cpara), 119.1
(d, Cortho), 119.4 (s, CN), 133.8 (s, Cmeta), 143.6 (d, Cipso); 31P NMR
d
62.25; 15N NMR
d
ꢁ284.46 (d). MS (m/z) 282 (Mþ), 249 (Mþꢁ33),
214 (Mþꢁ68), 196 (Mþꢁ86), 165 (Mþꢁ117), 149 (Mþꢁ133), 118
(Mþꢁ164), 85 (Mþꢁ197), 69 (Mþꢁ213), 41 (Mþꢁ241). Anal. Calcd
for C12H15O2NSP: C, 51.06%; H, 5.35%. Found: C, 51.01%; H, 5.06%
5.2.11. ax-2-p-Nitroanilino-2-thio-cis-4,6-dimethyl-1,3,2
phosphinane (cis-ax-6). Mp 226–229 ꢀC. Isolated yield (6.5%). 1H
l
5-dioxa-
16. The cis/ trans diastereoselectivity observed for the phosphochloridites I, is due to
the selective decomposition of the trans- ax isomer by traces of humidity and the
hydrochloric acid evolved during the reaction [see discussion in Gordillo, B.;
NMR
d 1.43 (dd, 6H), 1.83 (dt, 1H), 1.93 (dq, 1H), 4.63 (m, 2H), 4.89
(m, NH), 7.14 (d, JHH¼9.1, 2H), 8.17 (d, JHH¼8.16, 2H); 13C NMR
3
3
´
Hernandez, J. Org. Prep. Proc. Intern. (OPPI) 1997, 29,195 Based on the amount of the
meso diol, the diastereomer I (cis- ax) is obtained in w 85% yield, when isolated.
17. (a) Stewart, E. L.; Nevins, N.; Allinger, N. L.; Bowen, J. P. J. Org. Chem. 1997, 62,
5198; (b) Leyssens, T.; Peeters, D. J. Org. Chem. 2008, 73, 2725.
18. Ref. 1, p. 716.
d
22.2 (d, C7,8), 40.4 (d, C5), 76.8 (d, C4,6), 117.3 (d, Cortho), 125.6 (s,
Cmeta), 142.4 (s, Cpara), 146.0 (s, Cipso); 31P NMR
d
57.24; 15N NMR
d
ꢁ283.79 (d). MS (m/z) 302 (Mþ), 272 (Mþꢁ30), 236 (Mþꢁ66), 234

2076
Z. Dom´ınguez et al. / Tetrahedron 66 (2010) 2066–2076
19. Although we observed that thiophosphoramidates have almost sp2-hybridized
oxygens, the sp3 hybridization for oxygens in P (IV) containing compounds has
also been considered in theoretical studies (Ref. 6a). See also the comment on
Ref. 36 of this work.
20. Nifantiev, E. E.; Sorokina, S. F.; Borisenko, A. A. J. Gen. Chem. 1984, 55,
1481.
31. As stated by Eliel et al. (Ref. 1 p. 687), there exist a compromise between tor-
sional angles and valency angles in which the total strain of the molecule is
minimized. The mathematical cosine equations relating torsional angles and
bond angles in cyclohexane and heterocyclic compounds are proposed in Ro-
mers, C.; Altona, H. R.; Buys, H. R.; Havinga, E. Top. Stereochem. 1969, 4, 39.
32. Ref. 1 p. 770.
´
21. Hernandez, J.; Ramos, R.; Sastre, N.; Meza, R.; Hommer, H.; Salas, M.; Gordillo, B.
33. (a) Bentrude, W. G.; Setzer, W. N.; Ramli, E.; Khan, M.; Sopchik, A. E. J. Org.
Chem. 1991, 56, 6127; (b) Cameron, T. S.; Galdecki, Z.; Karolak-Wojciechowska, J.
Acta Crystallogr., Sect. B 1976, B32, 492.
34. Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry; Pergamon: New
York, NY, 1983.
35. Perrin, C. L.; Kuperman, J. J. Am. Chem. Soc. 2003, 125, 8846.
36. Considering that the non-boding electron pairs on oxygen participate in the
anomeric effect; the discussion on their canonical or sp3 nature, become im-
portant (see Ref 7c, p. 39). In the studied dioxaphosphinanes, the hybridization
on the oxygens is closer to sp2 (COP angles w118ꢀ) than the hybridization of the
oxygens of 1,3-dioxanes (COC angles w115ꢀ) where the anomeric effect is also
present Uehara, F.; Sato, M.; Kaneko, C.; Kurihara, H. J. Org. Chem. 1999, 64, 1436.
37. Wolfe, S. Acc. Chem. Res. 1972, 5, 102.
Tetrahedron 2004, 60, 10927.
22. Jameson, C. J.; Mason, J. In Multinuclear NMR; Mason, J., Ed.; Plenum: New York,
NY, 1987.
23. Modro, A. M.; Modro, T. A.; Bernatowicz, P.; Schilf, W.; Stefaniak, L. Magn. Reson.
Chem. 1997, 35, 774.
24. Gray, G. A.; Buchanan, G. W.; Morin, F. G. J. Org. Chem. 1979, 44, 1768.
25. Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
26. Kuivalainen, T.; El-Bahraoui, J.; Uggula, R.; Kostiainen, R.; Sundberg, M. R. J. Am.
Chem. Soc. 2000, 122, 8073.
27. (a) Gorenstein, D. G. J. Am. Chem. Soc. 1975, 97, 898; (b) Gorenstein, D. G. J. Am.
Chem. Soc. 1977, 99, 2254.
28. Huheey, J.E.; Keiter, E.A.; Keiter, R.L. Inorganic Chemistry. Priciples of Structure
and Reactivity. Harper Collins Collegue: New York, NY, 1993, 4th ed.
38. Pritchard, G. J.; Vollmer, R. J. J. Org. Chem. 1963, 28, 1545.
39. Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
´
29. Salas, M.; Gordillo, B.; Gonzalez, F. J. Arkivoc 2003, Xi, 72; http://www.arkat-
usa.org/get-file/19823/.
30. (a) Stec, W. J. Acc. Chem. Res. 1983, 16, 411; (b) Tomasz, J.; Bottka, S.; Radics, L.
NucleosidesNucleotides 1989, 8, 1297.
40. Sheldrick, G. M. SHELEX97: Program for the solution of Crystal Structures; Uni-
versity of Go¨ttingen: Germany, 1997.
41. Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.