The interplay of thio(seleno)amide/vinylogous thio(seleno) amide "resonance" and the anisotropic effect of thiocarbonyl and selenocarbonyl functional groups

Article abstract of DOI:10.1021/jo0503097

Amino-substituted thio(seleno)acrylamides 1-4 were synthesized and their ¹H and ¹³C NMR spectra assigned. Both the NMR data and the results of theoretical calculations at the ab initio level of theory were employed to elucidate the a

Full text of DOI:10.1021/jo0503097

The Interplay of Thio(seleno)amide/Vinylogous Thio(seleno)amide
“Resonance” and the Anisotropic Effect of Thiocarbonyl and
Selenocarbonyl Functional Groups
Erich Kleinpeter,*,† Anja Schulenburg,† Ines Zug,‡ and Horst Hartmann‡
Universita¨t Potsdam, Chemisches Institut, P.O. Box 60 15 53, D-14415 Potsdam, Germany, and
Fachhochschule Merseburg, Fachbereich Chemie, Geusaer Str., D-06217 Merseburg, Germany
kp@chem.uni-potsdam.de
Received February 18, 2005
Amino-substituted thio(seleno)acrylamides 1-4 were synthesized and their 1H and 13C NMR spectra
assigned. Both the NMR data and the results of theoretical calculations at the ab initio level of
theory were employed to elucidate the adopted structures of the compounds in terms of E/Z
isomerism and s-cis/s-trans configuration. In the case of the asymmetrically N(Me)Ph-substituted
compounds, ab initio GIAO-calculated ring current effects of the N-phenyl group were applied to
successfully determine the preferred conformer bias. The restricted rotations about the two C-N
partial double bonds were studied by DNMR and the barriers to rotation (∆Gcq) determined at the
coalescence temperatures, and these were discussed with respect to the structural differences
between the compounds. The barriers to rotation were also calculated at the ab initio level of theory
where the best results (R2 ) 0.8746) were obtained only with inclusion of the solvent at the SCIPCM-
HF/6-31G* level of theory. The calculations also provided means of assessing structural influences
which were not available due to inaccessible rotation barriers. By means of natural bond orbital
(NBO) analysis of 1-4, the occupation numbers of nitrogen lone pairs and bonding/antibonding
π/π* orbitals were shown to quantitatively describe thio(seleno)amide/vinylogous thio(seleno)amide
“resonance”. Finally, the thio(seleno)carbonyl anisotropic effect was quantitatively calculated by
the GIAO method and visualized by isochemical shielding surfaces (ICSS). Only marginal differences
between the two anisotropic effects were calculated and are therefore of questionable utility for
previous and future applications with respect to stereochemical assignments.
SCHEME 1
Introduction
The restricted rotation about the partial C-N double
bond in amides, as simple models for the peptide bond
in proteins, has been extensively studied.1 The traditional
notion of “amide resonance” A T B (see Scheme 1) as
the cause of the barrier to rotation about this bond has
been questioned2 by Wiberg et al., who determined by
theoretical ab initio studies that the major contributor
to the barrier arises not from charge transfer to the
oxygen, but rather by charge transfer between nitrogen
† Universita¨t Potsdam.
‡ Fachhochschule Merseburg.
(1) (a) Steward, W. E.; Siddal, T. H. Chem. Rev. 1970, 70, 517. (b)
Jackman, L. M. Dynamic NMR Spectroscopy; Academic Press: San
Francisco, 1975. (c) Lambert, J. B., Takeuchi, Y., Eds. Acyclic Organo-
nitrogen Stereodynamics; VCH: New York, 1992.
10.1021/jo0503097 CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/19/2005
6592
J. Org. Chem. 2005, 70, 6592-6602

Thio(seleno)amide/Vinylogous Thio(seleno)amide “Resonance”
SCHEME 2
SCHEME 3
and carbon (see also resonance structure C in Scheme
1). The barrier to rotation in the corresponding thio-
amides, however, is larger, and thus, the traditional
concept of charge transfer to the oxygen is more ap-
propriate,3,4 though the situation is more consistent with
the view that thioamides behave as thioformylamines.5
Based on X-ray analyses and more sophisticated ab initio
calculations, however, the resonance model is indeed
quite adequate to explain the properties of amides,6,7
thioamides,8,9 and also selenoamides.10 When studied
under similar conditions, the larger rotation barriers of
thioamides11,12 when compared with those of the corre-
sponding amides were shown to be linearly dependent,13
thus implying that the same mechanism and electronic
source is the underlying cause for the dynamic process
in both cases. The charge-transfer nN f π*CdX (X ) O, S,
Se) has been identified as the main component respon-
sible for the height of the barrier which linearly increases
(O < S < Se) with participation of resonance structure
B.14,15 This is also accompanied by opposing migration
of σ charge density, thus rendering the integrated atomic
charges smaller than expected from pure π delocaliza-
tion.16
barrier was shown to be higher in the vinylogous thio-
amides than in the corresponding amides.20,24 When both
thioamide and vinylogous thioamide moieties are in the
same structure, the two rotation barriers were shown to
be of approximately equal size (G and H in Scheme 3).25
Recently, however, this vinylogous amide resonance was
questioned when N,N-dimethylaminoacrylnitrile was
better regarded as a vinylamine substituted with electron-
withdrawing substituents,26 even if the rotation barrier
about the C2-C3 bond could not be measured (only two
substituents on the CdC double bond are insufficient to
reduce the barrier enough for experimental measure-
ment).27 Addition of another cyano group, (NC)2Cd
C(Ph)NMe2, further increases the barrier to rotation
about the C-N bond and corroborates the participation
of a resonance structure similar to G (cf. Scheme 3).28
Because of the equivocal ties mentioned, a larger
variety of amino-substituted thioacrylamides 1, seleno-
acrylamides 2, and their azaanalogues 3 and 4 (cf.
Scheme 4; it should be mentioned that 1-4 could be also
classified as vinylogous urea and azavinylogous urea
derivatives, respectively) were synthesized, and their
dynamic behavior was studied experimentally by dy-
namic NMR spectroscopy and theoretically by ab initio
MO calculations at various levels of theory. To investigate
the electronic interactions between the various molecular
orbitals in the compounds studied, natural bond orbital
(NBO) analysis29 was performed. The investigation of
1-4 provides a quantitative understanding of the sub-
stituent effects on the rotational barriers as well as the
rotating vinylogous thio- and selenoamide systems them-
selves by providing valuable information on bond ener-
gies, electron occupancies, and bond/antibond interac-
tions.
Similarly, the corresponding barrier to rotation in
vinylogous amides17-19 and thioamides20-24 (see Scheme
2) has been discussed in terms of the resonance structure
E participating in the ground state to a large degree with
respect to the restricted rotations about the C1-C2, C2-
C3, and C3-N partial double bonds. Again, the rotation
(2) (a) Wiberg, K. B.; Bereneman, C. M. J. Am. Chem. Soc. 1992,
114, 831. (b) Wiberg, K. B.; Hadad, C. M.; Rablen, P. R.; Cioslowski,
J. J. Am. Chem. Soc. 1992, 114, 8644.
(3) Wiberg, K. B.; Rablen, P. R. J. Am. Chem. Soc. 1995, 117, 2201.
(4) Wiberg, K. B.; Rush, D. J. J. Am. Chem. Soc. 2001, 123, 2038.
(5) Laidig, E. K.; Cameron, L. M. J. Am. Chem. Soc. 1996, 118, 1737.
(6) Quinonero, D.; Frontera, A.; Capo, M.; Ballester, P.; Suner, G.
A.; Garau, C.; Deya, P. M. New J. Chem. 2001, 25, 259.
(7) Fogarasi, G.; Szalay, P. G. J. Phys. Chem. A 1997, 101, 1400.
(8) Ferretti, V.; Bertolasi, V.; Gilli, P.; Gilli, G. J. Phys. Chem. 1993,
97, 13568.
(9) Hambley, T. W.; Hibbs, D. E.; Turner, P.; Howard, S. T.;
Hursthouse, M. B. J. Chem. Soc., Perkin Trans. 2 2002, 235.
(10) Prasad, B. V.; Uppal, P.; Bassi, P. S. Chem. Phys. Lett. 1997,
276, 31.
(11) Kim, W.; Lee, H.-J.; Choi, J.-H.; Yoon, C.-J. J. Chem. Soc.,
Faraday Trans. 1998, 2663.
(12) Ou, M.-C.; Chu, S.-Y. J. Phys. Chem. 1995, 99, 556.
(13) Piccinni, C.; Fabre, O.; Zimmermann, D.; Reisse, J.; Cornea,
F.; Fulea, C. Can. J. Chem. 1977, 55, 2649.
(14) Glendening, E. D.; Hrabal, J. A., II. J. Am. Chem. Soc. 1997,
117, 12940.
(15) Bharatam, P. V.; Moudgil, R.; Kaur, D. J. Phys. Chem. A 2003,
107, 1627.
(16) Mo, Y.; von Rague Schleyer, P.; Wo, W.; Lin, M.; Zhang, Q.;
Gao, J. J. Phys. Chem. A 2003, 107, 10011.
(17) (a) Dabrowski, J.; Kozerski, L. J. Chem. Soc., Chem. Commun.
1968, 586. (b) Dabrowski, J.; Kozerski, L. J. Chem. Phys. B 1971, 345.
(18) Dabrowski, J.; Tencer, M. Bull. Chem. Soc. Jpn. 1975, 48, 1310
and references therein.
(19) Siebert, H.-Ch.; Tajkhorsshid, E.; Dabrowski, J. J. Phys. Chem.
A 2001, 105, 8488.
Results and Discussion
1H and 13C NMR Spectra. The signals in the 1H NMR
spectra of 1-4 were generally well dispersed, and con-
(20) Dabrowski, J.; Kamienska-Trela, K. Org. Magn. Reson. 1972,
4, 421.
(21) Kleinpeter, E.; Schroth, W.; Pihlaja, K. Magn. Reson. Chem.
1991, 29, 223.
(22) (a) Kleinpeter, E.; Spitzner, R.; Schroth, W.; Pihlaja, K. J. Prakt.
Chem. 1990, 332, 841. (b) Kleinpeter, E.; Spitzner, R.; Schroth, W.
Magn. Reson. Chem. 1987, 25, 688. (c) Kleinpeter, E.; Pulst, M. J.
Prakt. Chem. 1980, 322, 575. (d) Kleinpeter, E.; Heinrich, B.; Pulst,
M. Z. Chem. 1978, 18, 220. (e) Kleinpeter, E.; Pulst, M. J. Prakt. Chem.
1977, 319, 1003. (f) Kleinpeter, E.; Pulst, M. Z. Chem. 1977, 17, 339.
(23) Pulst, M.; Greif, D.; Kleinpeter, E. Z. Chem. 1988, 28, 345 and
references therein.
(25) Kleinpeter, E.; Tadesse, T.; Abegaz, B. J. Prakt. Chem. 1986,
328, 120.
(26) Rablen, P. R.; Miller, D. A.; Bullock, V. R.; Hutchinson, P. H.;
Gorman, J. A. J. Am. Chem. Soc. 1999, 121, 218.
(27) Kleinpeter, E.; Klod, S.; Rudorf, W.-D. J. Org. Chem. 2004, 69,
4317.
(28) Kaur, S.; Eberhardt, E. S.; Doucette, A.; Chase, A.; Dalby, C.
J. Org. Chem. 2002, 67, 3937.
(29) Forster, J. B.; Weinhold, F. J. Am. Chem. Soc. 1980, 102, 7211.
(24) Kleinpeter, E.; Pulst, M. Org. Magn. Reson. 1977, 9, 308 and
references therein.
J. Org. Chem, Vol. 70, No. 17, 2005 6593

Kleinpeter et al.
SCHEME 4
sequently, in such cases they were easily assigned (Table
S1 in the Supporting Information). Due to the push-pull
effect,27 the chemical shifts of H-1 and H-2 in 1 and 2
are extremely deshielded (δH-1 ) 8.01-8.43 ppm) and
shielded (δH-2 ) 4.88-5.55 ppm), respectively. The
vicinal coupling constant between these two protons lies
in the range of 11.5-11.9 Hz; hence, an E configuration
for 1 and 2 can be assigned25 since in the corresponding
Z configuration this coupling constant has been shown
to be less than 9 Hz.25 In the case of the imino-substituted
6594 J. Org. Chem., Vol. 70, No. 17, 2005

Thio(seleno)amide/Vinylogous Thio(seleno)amide “Resonance”
SCHEME 5
SCHEME 6
FIGURE 1. Possible isomers of the amino-substituted thio-
acrylamides and their relative energies.
TABLE 1. Relative Energy Differences for the Isomers
of Amino-Substituted Thio(seleno)acrylamides 1-3 As
Obtained by ab Initio Quantum Chemical Calculation at
the HF/6-31G* Level of Theory
compd
∆E(s-cis,s-trans) (kcal/mol)
1a
1b
1c
1d
1e
1f
1g
2
3b
3n
4.6
5.3
4.8
analogues 3 and 4, H-1 (δH-1 ) 8.70-9.14 ppm) is
deshielded even more due to additional substituent
effects. Also readily assignable were the proton signals
of the N-alkyl substituents. Coupling information was
corroborated by COSY spectra and spatial information
facilitated by NOESY spectra.30 The aromatic protons in
the N(Ph)2-substituted thio(seleno)acrylamides 1e and 3n
resonate in a very narrow range, and consequently, the
ortho, meta, and para protons could not be discriminated.
The same effect was observed for the N(Me)Ph analogue
4f; in the case of 3h, 3k, 3m, and 4m, the meta protons
could be differentiated from the multiplets of the ortho
and para protons. Finally, in the case of 1f and 3o, all
three proton types were well separated and therefore
assignable.
8.9
6.2
6.9
7.2
8.3
24.6
20.0
configuration was identified as the preferred configura-
tion in similar compounds. To corroborate this previous
assignment25,31 and to clarify the open structural ques-
tions, the amino-substituted thio(seleno)acrylamides 1-4
were also studied quantum chemically: the four isomers/
rotamers (cf. Scheme 6) were calculated and examined
with respect to their relative energies.
The 13C NMR spectra of 1-4 were assigned by the
standard application of APT, HMQC and HMBC spectra
(Table S2, Supporting Information).30 The ipso and CdS/
CdSe carbons were readily discerned as they displayed
vicinal/geminal C,H connectivities to nearby protons (e.g.,
ortho or H-1/H-2 protons, respectively). The 13C reso-
nances for all compounds examined resonated well within
the typical ranges anticipated; only C-1 and C-2, the
carbons of the central CdC partial double bond in 1 and
2, resonated at extreme positions for sp2-hybridized
carbon atoms due to the push-pull effect whereby C-1
is strongly deshielded and C-2, conversely, is strongly
shielded (cf. Scheme 5).27
Structural Study. The scalar vicinal H,H trans
coupling in 1 and 2 (vide supra) was the only structurally
discriminating NMR parameter in the compounds stud-
ied. Whereas the exchange phenomena observed for the
NR2 substituents could be interpreted as a result of
restricted rotation about the partial C1-N and C3-N
double bonds (vide infra), additional information by NMR
was not obtained concerning E/Z isomerism in 3 and 4,
nor regarding s-cis/s-trans isomerism in 1-4 (cf. Scheme
6); the corresponding equilibria are evidently too one-
sided and only single anancomers are present. Previously,
by a lanthanide-induced shift study,25,31 the E(s-cis)
In the amino-substituted thio(seleno)acrylamides 1 and
2, only the E(s-cis)/E(s-trans) isomers were calculated,
the Z isomers having been unequivocally excluded on the
basis of the observed value of the vicinal H,H trans
coupling constants. As expected, the E(s-cis) isomers
proved to be strongly preferred and should represent the
structures as given experimentally by NMR (cf. Table 1
and Figure 1). The same result was obtained for the
imino analogues 3 and 4, but since appropriate scalar
coupling was lacking, additionally the corresponding Z(s-
cis) and Z(s-trans) isomers were also examined. However,
local minima for both Z(s-cis) and Z(s-trans) isomers
could not even be obtained due to steric reasons and
again the same result (cf. Table 1) was obtained, viz. the
E(s-cis) configuration proved to be the preferred isomer
(see Figure S1, Supporting Information).
In the case of N(Me)Ph asymmetrical substitution,
compounds 1f, 3h, 3k, 3m, 3o, 4f, and 4m were obtained;
in addition, there are another two conformers as a result
of the restricted rotation about the C1-N and C3-N
partial double bonds (see Scheme 7). On the NMR time
scale within the temperature range of the present system,
exchange phenomena were not observed (vide infra) and
because the second C-N barrier to rotation is likely to
be very similar to the rotation barriers in the other
compounds studied, there was only one conclusion pos-
sible: the presence again of only one preferred conformer.
The ab initio calculation of the corresponding conformers
provided N1,Ph(s-trans) conformers for compounds 3h,
(30) Pihlaja, K.; Kleinpeter, E. In Carbon-13 NMR Chemical Shifts
in Structural and Stereochemical Analysis; VCH Publishers: New
York, 1994.
(31) Kleinpeter, E.; Pulst, M. Z. Chem. 1985, 25, 336.
J. Org. Chem, Vol. 70, No. 17, 2005 6595

Kleinpeter et al.
SCHEME 7
TABLE 2. Relative Energies of Conformers as a Result
of Restricted Rotation about the C-N Partial Double
Bonds As Obtained by ab Initio Quantum Chemical
Calculations at the HF/6-31G* Level of Theory
N1/3,Ph(s-trans)
N1/3,Ph(s-cis)
∆E(s-trans,s-cis)
compd
(a.u.)
(a.u.)
(kcal/mol)
1f
-970.114
-986.136
-1137.898
-1063.052
-1137.912
-3063.116
-2986.199
-970.108
-986.132
-1137.891
-1063.048
-1137.909
-3063.113
-2986.196
4.063
2.283
4.425
2.166
1.739
2.105
1.641
3h
3k
3m
3o
4m
4f
3k, 3m, and 4m (see Figure S2, Supporting Information)
and N3,Ph(s-trans) conformers for compounds 1f, 3o, and
4f (see Figure S3, Supporting Information). Obviously,
the phenyl, due to rotation about the N-Cipso bond, can
elude steric hindrance better than methyl. However,
except for 1f and 3k, the energy differences, at the ab
initio level of theory (cf. Table 2), between the two
conformers proved to be only a few kcal/mol, and thus a
fast conformational equilibrium consisting of the two
conformers could not be excluded completely. Therefore,
another NMR method was employed to corroborate these
conclusions.
Ab Initio MO Calculation of the Ring Current
Effect of N-Phenyl Substituents. Recently, Klod and
Kleinpeter32 reported the ab initio MO calculations of the
anisotropic effects of a number of functional groups and
the ring current effect of aromatic/heteroaromatic moi-
eties. The ring current effect of phenyl can be visualized
as isochemical shielding surfaces (ICSS) and was em-
ployed to distinguish the conformers depicted in Figures
S2 and S3 in the Supporting Information (e.g., in Figure
2 the orange ICSS represents shielding of -0.1 ppm at
ca. 9 Å from the center of the molecule). The conformers
of 1f and 3h are depicted in Figure 2 with various ICSS
emanating from the N-phenyl group and their resultant
effects on the protons H-1 and H-2 and of the N-1(3)
methyl. The corresponding chemical shift variations due
to this ring current effect were compared with the
experimentally obtained chemical shift differences for the
corresponding alkyl-substituted analogues (e.g., 1f is
compared with 1b, etc.). Unequivocally, the agreement
of the phenyl ring current effects in N1,Ph(s-trans) and
FIGURE 2. Conformers of 1f and 3h and calculated ring
current effects of the N-phenyl rings on H-1, H-2, and N-Me
protons (shielding ICSS of -0.1 ppm (orange), -0.2 ppm
(yellow), -0.3 ppm (green) and -0.5 ppm (light blue)).
N3,Ph(s-trans) and the experimental chemical shift dif-
ferences of H-1 and H-2 and of the N-methyl protons
corroborate the above assertion that these conformers are
the preferred ones in the N(Me)Ph-substituted thio-
(seleno)acrylamides.
Structural Study of N,N-Diaryl-Substituted Thio-
(seleno)acrylamides. Similarly, in the N(Ph)2-substi-
tuted thio(seleno)acrylamides 1e, 3n, and 4e, exchange
phenomena due to restricted C3-N rotation were not
observed within the experimental temperature range
(vide infra). The ab initio calculations provided global
minima structures with completely planar sp2-hybridized
nitrogens of the corresponding N-3 diphenyl substituents
(cf. Figure S4, Supporting Information); thus, the corre-
sponding C-N rotation should be restricted in the
temperature interval accessed but cannot be so measured
due to rotation of the N-phenyls. Obviously, concerted
dynamic processes, which are still fast on the NMR time
scale and which equilibrate ortho and meta protons, are
in effect, thereby blocking dynamic effects of the re-
stricted C3-N rotations which are, in principle, ready for
examination. In the case of 1g, however, with different
aromatic moieties (phenyl and naphthyl) on N-3, besides
(32) Klod, S.; Kleinpeter, E. J. Chem. Soc., Perkin Trans. 2 2001,
1893.
6596 J. Org. Chem., Vol. 70, No. 17, 2005

Thio(seleno)amide/Vinylogous Thio(seleno)amide “Resonance”
TABLE 3. Barriers to Rotation (Solvent, Tc, ∆νc, ∆Gcq) about the C1-N Partial Double Bond in Compounds 1-4
compd
solvent
Tc (K)
∆νc (Hz)
∆Gcq (kcal/mol)
signal studied
1a
1b
1c
1d
1e
1f
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
C2D2Cl4
CD2Cl2
CD2Cl2
C2D2Cl4
CD2Cl2
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
269
86.54
88.56
12.9
13.2
13.5
13.3
14.4
13.6
14.6
16
18.5
20.2
18.9
17.7
19
19.1/19.0
>20.2
19.2
>19.1
>19.5
19.3
N(Me)2
276
282
N(Me)2
87.56
N(Me)2
278
87.71
152.34
162.99
N(Me)2
306
N(Me)2
290
N(Me)2
1g
2
207/228
337
5.11/26.76
N(Me)2/N(Me)2
N(Me)2
129.73
217.34
27.93
3b
3c
3d
3i
395
N(Me)2
398
N(Me)2
370
22.89
N(Me)2
360
63.99
N(CH2)2
N(CH2)2
O(CH2)2/N(CH2)2
N(Me)2
3l
387
66.62
3n
4b
4d
4e
4f
386/376
>395
400
53.66/33.63
∼24.0
116.69
∼105.1
∼56.8
N(Me)2
>395
>395
391
N(Me)2
N(Me)2
4i
80.86
N(CH2)2
the restricted rotation about C1-N, the restricted rotation
about the C3-N partial double bond could also be
examined as the dimethyl group on N-1 group first
decoalesces into two singlets (C1-N rotation) and then
further, at still lower temperatures, into four singlets
(C3-N rotation, vide infra). In the global minima of 1e,
1g, 3o, and 4e, the two aromatic moieties are positioned
perpendicular to each other (cf. Figure S4, Supporting
Information), but the barrier to the concerted N-Cipso
rotations, however, was too low to be measured. The same
result was obtained for compound 2; the global minimum
conformer of which is presented in Figure S4 (Supporting
Information).
In the case of the N-3 asymmetrically substituted
thioacrylamide 1g, two conformers were also obtained as
global minima by ab initio calculations: the cis position
of the thiocarbonyl group to phenyl but trans to naphthyl
was found to be much more stable (∆E ) 2.45 kcal/mol,
see Figure S5 in the Supporting Information). This
conformational preference was again verified by GIAO
ring current calculations of the two aromatic moieties and
evaluating their effects on the chemical shift differences
of the N-1 dimethyl group in 1g: the qualitative agree-
ment of theoretically calculated and experimental values
in case of N3,naph(s-trans) is in agreement with the
aforementioned energy difference of the two conformers
and thus implicated this conformer as the anancomer of
the conformational system.
Barriers to Rotation about the C-N Partial
Double Bonds. In 1-4, one donor and one acceptor
substituent are attached to the central CdC or CdN
double bonds; thus, due to the push-pull effect, the
barrier to rotation about this bond should be lowered,
but the lowering was insufficient to enable examination
by NMR.27 However, the corresponding partial double
bond character of the bonds of these olefinic carbons to
the donor/acceptor substituents are sufficiently high to
be studied within the range set by the NMR time scale,
and the barrier to rotations were evaluated by DNMR
as free energies of activation (∆Gcq)33 at the coalescence
temperature (Tc). The Tc values, the chemical shift
differences (∆ν) of the N(CHn)2 protons at Tc (by extrapo-
lation from slow exchange) and the corresponding rota-
tion barriers (∆Gcq) for the amino-substituted thio(seleno)-
acrylamides 1-4 are presented in Tables 3 and 4,
conditions permitting [barriers to rotation could not be
determined in 1f, 3h, 3k, 3m, 3o, 4m and 4f, this was
due to anancomerism (vide supra); in 2e, 1n and 4e,
because of concerted rotations of the NPh2 substituents
that are too fast on the NMR time scale (vide supra); in
3o and 4l, because of serious signal overlap; and, finally,
in 2, 3l, 4c and 4p due to the rotation barrier being too
high to be studied on the NMR time scale (i.e., no
exchange phenomena observed at the highest tempera-
tures attained)]. The following points are noted concern-
ing the experimentally determined barriers to rotation:
(i) With respect to the influence on the rotation barrier
of the chalcogen elements sulfur and selenium, a number
of analogous structures can be compared, e.g., 3c, 3d, 3i,
3l, 3m, and 3p and 4c, 4d, 4i, 4l, 4m, and 4p, respec-
tively. Both barriers to rotation about the C1-N and
C3-N partial double bonds are higher in the case of
CdSe. One reason for this may be the larger polarizabil-
ity of selenium (R ) 4.56 × 10-30 m3) compared with
sulfur (R ) 3.44 × 10-30 m3), as is generally the case with
increasing atomic weight along the chalcogen ele-
ments,10,34,35 but negative hyperconjugation was also
found to play an important role (vide infra).15
(ii) The rotational barriers about the C-N partial
double bonds in the amino-substituted thio(seleno)-
acrylamides 1-4 as a function of the nitrogen substitu-
ents follow the order: pyrrolidino > NMe2 > piperidino
> morpholino, thus reflecting the basicity of the corre-
sponding N atoms as provided by reactivity indices.36,37
The difference in the six-membered rings resulted from
the additional -I inductive effect of the oxygen atom in
morpholino.24
(iii) Because the barrier to rotation about the central
HC1dC2H partial double bond in push-pull olefins was
(34) Behrendt, S.; Borsdorf, R.; Kleinpeter, E.; Gru¨ ndel, D.;
Hantschmann, A. Z. Chem. 1976, 16, 405.
(35) Rae, I. D.; Wade, M. J. Int. J. Sulfur Chem. 1976, 8, 519.
(36) Kleinpeter, E. J. Mol. Struct. 1996, 380, 139.
(37) Kleinpeter, E.; Koch, A.; Taddei, F. J. Mol. Struct. 2001, 535,
257.
(33) Sandstro¨m, J. In Dynamic NMR Spectroscopy; Academic
Press: London, 1982.
J. Org. Chem, Vol. 70, No. 17, 2005 6597

Kleinpeter et al.
TABLE 4. Barriers to Rotation (solvent, Tc, ∆νc, ∆G) about the C3-N Partial Double Bond in Compounds 1-4
compd
solvent
Tc (K)
∆νc (Hz)
∆Gcq (kcal/mol)
signal studied
1a
1b
1c
1d
1g
3b
3c
3d
3h
3i
3k
3m
3p
4b
4c
4d
4i
CD2Cl2
CD2Cl2
CD2Cl2
C2D2Cl4
C2D2Cl4
CD2Cl2
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
C2D2Cl4
277
296
263
180.63
86.8
200.43
121.05/38.05
158.52
48.9
24.46/68.55
36.96
41.07
55.81
12.9
14.2
N(CH2)2
N(Me)2
12.1
N(CH2)2
N(CH2)2/(CH2)2
N(Me)2
N(Me)2
O(CH2)2/N(CH2)2
N(CH2)2
N(Me)2
N(Me)2
372/356
312
398
356/376
384
395
374
350/364
>395
388
>395
396
>395
>395
.395
>395
17.8/17.8
11.0/11.4
19.8
18.1/18.4
19.3
19.8
18.5
17.9/17.9
>20.2
18.9
>19.2
19.1
21.57/61.38
∼25.7
O(CH2)2/N(CH2)2
N(CH2)2
80.63
N(CH2)2
∼88.8
N(Me)2
111.7
N(CH2)2
∼68.5
>19.4
>19.1
.19.5
>18.9
N(CH2)2
∼93.7
N(Me)2
4m
4p
∼62.2
N(CH2)2
∼123.0
N(CH2)2
previously found to be 2-3 kcal/mol higher than the
HC1dN analogues,43 the corresponding C1-N and C3-N
barriers 1 and 2 compared with 3 and 4 were found to
be 5-6 kcal higher in the latter compounds due to the
greater “vinylogous thioamide resonance”.
compounds 1-4 for further and more detailed informa-
tion concerning their dynamic behavior. The various TS
were developed by the TRIPOS force field and energy
optimized using the semiempirical PM3 method; the
structures, thus obtained (and also the corresponding GS
found by the same procedure), were employed as starting
structures for geometry optimization at the ab initio level
of theory (HF/6-31G*, HF/6-311G**, SCIPCM-HF/6-
31G*, and B3LYP/6-31G*). As an example, in Figure 3,
the TS of the restricted rotations about the C1-N and
C3-N partial double bonds of 1b are depicted [two TS
were obtained having the lone pair of the now sp3-
hybridized nitrogen in cis and trans positions to C1d
C2(N2)] and in Table S4 (Supporting Information) both
the corresponding GS and TS energies and barriers to
C-N rotation as obtained at the various levels of theory
are given.
The barriers calculated at the semiempirical PM3 level
were far too low (due to unavailable parameters for
selenium, compounds 2 and 4 could not even be calcu-
lated), and the relative differences between C1-N and
C3-N were not reproduced. The higher barriers in 3
compared with 1, on the other hand, were correctly found,
but otherwise did not correspond well. The calculational
results improved at the ab initio level of theory [for two
TS (cf. Figure 3 and Table S4 in Supporting Information),
only the more stable with the nitrogen lone pairs trans
to C1dC2(N2) was considered further], though the calcu-
lated rotation barriers were still too low. In addition,
characteristic structural influences on the two barriers
(CdS versus CdSe, NR2 variation and HC1dC2H versus
HC1dN2) are tenuously correctly reproduced. Due to
these shortcomings, additional calculations were per-
formed either where the basis set was extended to HF/
6-311G**, the solvent (CD2Cl2) was included (SCIPCM-
HF/6-31G*) in the calculations or DFT calculations
(B3LYP/6-31G*) were performed (see Table S4, Support-
ing Information). Both the extended triple-ú-split-valence
basis set, which better represents sulfur and selenium
by employing polarization functions, and the inclusion
of the solvent into the calculations improved the agree-
ment of the calculated rotation barriers with those
measured experimentally (cf. Table S4 and Table S5,
However, the conclusions to be drawn from these
experimental results are somewhat limited. In many
cases, the barriers could not be measured on the NMR
time scale for various reasons, thus leaving only a limited
number of cases experimentally accessible. In addition,
the relative energies of both the ground states (GS) and
the transition states (TS) of the rotations about the C-N
partial double bonds are not available, only the difference
of the two are, but both are characteristically contributing
to the electron delocalization. For this reason, the global
minimum energies and structures of both the GS and TS
of the dynamic processes in 1-4 were determined by
theoretical calculations at the ab initio level of theory.
Theoretical Calculation of the Barriers to Rota-
tion About the C1-N and C3-N Partial Double
Bonds in Amino-Substituted Thio(seleno)acryl-
amides 1-4. In the GS, the substituents on N-1 and N-3
are in the plane of the molecule and the global minimum
results from the lowest steric hindrance. In the TS, the
N(1)R2 and N(3)R′ moieties are at an angle of 90°,
2
respectively [i.e., the planes of the two substituents NR2
and NR′ sthe nitrogens are sp3-hybridizedsare at an
2
angle of 90° to the planes of the HC1dC2H(N)C3d
S(Se)NR′2 and R2N-C1HdC2H(N)C3dS(Se) segments,
respectively]. The characteristic feature that is most
distinctive between the geometries of the two states is
the length of the HC1-N and C3-N partial double bonds,
which are truly single bonds in the TS (see Table S3,
Supporting Information). Thus, it is necessary, indeed
crucial, to calculate accurate structures for the TS of the
(38) Benassi, R.; Bertarini, C.; Hilfert, L.; Kempter, G.; Kleinpeter,
E.; Spindler, J.; Taddei, F.; Thomas, St. J. Mol. Struct. 2000, 520, 273.
(39) Foresman, J. B.; Keith, T. A.; Wiberg, K. B.; Snoonnian, J.;
Frisch, M. J. Phys. Chem. 1996, 100, 16098.
(40) Forster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102, 7211.
(41) Balasubrahmanyam, S. N.; Bharathi, S. N.; Usha, G. Org.
Magn. Reson. 1983, 21, 474.
(42) Szalontai, G.; Dudas, J. Acta Chim. Hung. 1985, 119, 7.
(43) Kalinowski, O.; Lubosch, W.; Seebach, D. Chem. Ber. 1977, 110,
3733.
6598 J. Org. Chem., Vol. 70, No. 17, 2005

Thio(seleno)amide/Vinylogous Thio(seleno)amide “Resonance”
FIGURE 3. Transition states of the restricted rotations about the C1-N (I) and C3-N (K) partial double bonds of the amino-
substituted thioacrylamide 1b (cis and trans saddle point rotamers of C1-N and C3-N shown).
Supporting Information). The best results were obtained
employing the double-ú-split-valence basis set and con-
sidering the solvent-SCIPCM44 (at the HF/6-31G* level
of theory, R2 ) 0.8356 for C1-N and R2 ) 0.7766 for C3-
N; at the SCIPCM-HF/6-31G* level of theory, R2 ) 0.8126
for C1-N and R2 ) 0.8746 for C3-N; cf. Figure S6 and
Table S6, Supporting Information). For the DFT results,
the calculated rotation barriers further approach the
experimental values, and obviously electron correlation
which is included in these calculations improves the
representation of the electronic situation (at the B3LYP/
6-31G* level of theory, R2 ) 0.8534 for C1-N and R2 )
0.8799 for C3-N; cf. Figure S6 and Table S6, Supporting
Information). The theoretical calculations help filling in
the voids in the experimental rotational barriers (Tables
3 and 4) and complete sets of barriers to rotation about
the two C-N partial double bonds are thus available. In
addition to conclusions (i)-(iii), which were all confirmed,
the following additional points are worth noting:
(iv) For the amino-substituted selenoacrylamide 2,
exchange phenomena for the C3-N substituent were not
obtained within the temperature range available (cf.
Table 4). The calculated barrier to C3-N rotation, only
4.7 kcal/mol, corroborates this experimental finding and
thus the rotation barrier is too low to be studied on the
NMR time scale.
(v) The relative size of the two barriers to rotation in
1g are now comparable: the rotation barrier about C1-N
(experimental, 14.6; calculated, 13.3 kcal/mol) proved to
be larger than the rotation barrier about C3-N (experi-
mental, 11.4 kcal/mol; calculated, 10.5 kcal/mol).
result is explicitly corroborated by the present calcula-
tions; barriers to rotation above 19 kcal/mol were calcu-
lated. The entire sets of amino-substituted thioacryl-
amides 3 and amino-substituted selenoacrylamides 4 can
now be compared with respect to the influence on the
barrier of the chalcogen element, however, the larger
barriers in the case of selenium cannot be fully substan-
tiated by the ab initio calculations as the actual differ-
ences were calculated to be only negligible.
(vii) The rotation barrier for the C1-N and C3-N
partial double bonds in the case of N(Me)Ph substituents
could not be studied for several reasons (vide supra). The
barriers as calculated are very similar to the N-dialkyl
analogues, corroborating the former observation that the
aryl twist is dramatic, ca. 90°, completely preventing
aromatic resonance influence on the barriers to rotation.
(viii) In the case of 3l and 3m (for N-3, R2 ) pyrroli-
dino), the C3-N barrier to rotation could not be obtained
(cf. Table 4) and this in complete agreement with the ab
initio result, calculated as being in excess of 22 kcal/mol.
The same is true for the C1-N barriers of 3k and 3o.
To further theoretically understand the structural
influences (i)-(viii) on the two C-N barriers to rotation
and to draw firmer conclusions regarding these influences
in terms of “resonance” along the amino-substituted thio-
(seleno)acrylamides, the compounds were further studied
by NBO analysis.40
NBO Analysis of Amino-Substituted Thio(seleno)-
acrylamides 1-4. At the HF/6-31G* level of theory, the
occupation of the various molecular orbitals of both
bonding electron pairs and lone electron pairs in the
compounds were assessed and the structural differences
compared accordingly. These occupancies are provided
in Table S3 (Supporting Information). If the compounds
are compared with respect to “resonance” determining
the two barriers to rotation about the C-N bonds, the
following molecular orbital interactions are revealed as
the most significant: (i) the lone pair of N-1 with the
antibonding π* orbital of the central partial C1dC2 (C1d
(vi) In the case of the amino-substituted selenoacryl-
amides 4, only the C1-N barriers of 4d, i and the C3-N
barrier of 4c could be experimentally obtained (cf. Tables
3 and 4); the other corresponding barriers proved to be
too high to be studied by DNMR. This experimental
(44) (a) Hartmann, H.; Heyde, C.; Zug, I. Synthesis 2000, 805. (b)
Noack, A.; Hartmann, H. Tetrahedron 2002, 58, 2137. (c) Zug, I.;
Hartmann, H. Z. Naturforsch. 2002, 57b, 420.
J. Org. Chem, Vol. 70, No. 17, 2005 6599

Kleinpeter et al.
SCHEME 8a
SCHEME 10a
a Key: (a) C3,N barrier to rotation; (b) orbital occupation
number; (c) bond length.
a Key: (a) C1-N barrier; (b) C3-N barrier; (c) C1-N barrier, as
calculated; (d) C3-N barrier, as calculated; (e) orbital occupation
number; (f) bond length.
increases the two barriers to C-N rotation dramatically
(vide supra); accordingly, the occupations of the two
nitrogen lone pairs are strongly reduced, the occupation
of the π*C(1),C(2)(N-2) orbital and the π*CdS raised and that
of the πC(1),C(2)(N-2) orbital reduced. The corresponding
bond length variations are: C1-N1, C2(N2)-C3, and C3-N
are shortened, whereas CdS is lengthened.
SCHEME 9a
(iii) Finally, the effect of the NR2 substituents on the
C-N barriers to rotation and “resonance”, respectively,
is correctly reproduced: in line with the C3-N barrier
decrease the occupation of π*CdS also decreases while the
occupation of the N-3 lone pair increases. The corre-
sponding bond length change, C3-N shortens (CdS
remains almost constant), corroborates the corresponding
“resonance” along the present system.
From this NBO study, not even the slightest hint was
obtained that the dynamic behavior of the amino-
substituted thio(seleno)acrylamides 1-4 could not be
discussed within the traditional thio(seleno)amide (vi-
nylogous) “resonance” concept. However, the previous
statement25 that approximately the same amount of
thioamide/vinylogous thioamide resonance is in effect
cannot hold because C3-N/CdS(Se) proved to be stronger
than C1-N/CdS(Se) resonance via C1dC2(N), however,
the difference is simply balanced by the N-1 lone pair/
π*C(1)dC(2)(N-2) resonance (cf. Schemes 8-10 and Table S3,
Supporting Information).
a Key: (a) C1-N barrier; (b) C3-N barrier; (c) C1-N barrier, as
calculated; (d) C3-N barrier, as calculated; (e) orbital occupation
number; (f) bond length.
N2) double bond and also the π orbital of the C1dC2 bond
with the antibonding π* orbital of the CdS(Se) bonds for
the C1-N barrier and (ii) the lone pair of N-3 with the
π* orbital of the CdS(Se) bonds for the C3-N barrier (cf.
Scheme 3). These molecular interactions and the oc-
cupancies of the corresponding orbitals were employed
to quantify the “resonance” as the source for the rotation
barriers and have been compared with respect to struc-
tural variations (i)-(iii) in the thio(seleno)acrylamides
1-4. In respect to this, the comparison of thio (3) and
seleno analogues (4) (cf. Scheme 8), of olefino (1) and
imino analogues (3) (cf. Scheme 9), as well as the
structural variations due to the NR2 substituents (cf.
Scheme 10) proved most illuminating:
(i) Replacement of sulfur by selenium increases the two
C-N barriers to rotation (vide supra); accordingly, the
occupations of the two nitrogen lone pairs are reduced,
the occupation of the π*C(1)dN(2) orbital raised and that of
the πC(1)dN(2) orbital reduced. The corresponding bond
length variations C1-N1, N2-C3, and C3-N3 are short-
ened, whereas C1dN2 is lengthened.
Anisotropic Effect of the CdS and CdSe Groups.
In addition to the ring current effect of N-phenyl (for
differentiating the conformers with respect to C1-N and
C3-N restricted rotation), the anisotropic effects of the
thiocarbonyl and the selenocarbonyl groups were also
calculated by the same method.32 The two effects, visual-
ized in Figure 4, proved to be rather similar and not too
far reaching (ICSS for (0.1 ppm at ca. 5 Å). This is about
the same, in distance terms, as that available from NOE
measurements used for stereochemical determinations.
However, because for distances <3.5 Å the calculated
anisotropic effects are not very reliable,27,32 this approach
is therefore not very useful for stereochemical applica-
tions when comparing thiocarbonyl and the selenocar-
bonyl. This is in contrast to the case where carbonyl
anisotropy was compared with thione32,41 and could be
successfully employed for assignments in 1H NMR spec-
troscopy.42,43 Thus, as the differences are only marginal
when comparing CdS with CdSe, the consequences for
(ii) The same is true when comparing the correspond-
ing 1 and 3 analogues. Replacement of C1dC2 by C1dN2
1
correct assignment in H NMR35 are obvious.
6600 J. Org. Chem., Vol. 70, No. 17, 2005

Thio(seleno)amide/Vinylogous Thio(seleno)amide “Resonance”
FIGURE 4. Anisotropic effects of thiocarbonyl (inner ICSS) and selenocarbonyl groups (outer ICSS) in thioformaldehyde and
selenoformaldehyde as calculated by NICS analysis (shielding surfaces at 0.1 ppm in yellow; deshielding surfaces at 0.1 ppm in
red). View from perpendicular to the molecules (left) and in the plane of the molecules (right).
Conclusions
SCHEME 11
A series of amino-substituted thio(seleno)acrylamides
1-4 was synthesized and studied with respect to π
electron distribution. The compounds exist as preferred
E(s-cis) isomers; when N(Me)Ph substituents are present,
further preferred conformations [N3,Ph(s-trans) and
N1,Ph(s-trans) to CdS(Se)] were observed. Of the various
methodologies applied for the assignment procedure, the
application of the GIAO ab initio-calculated ring current
effect of N-phenyl was found to be extremely useful.
The restricted rotations about the C1-N and C3-N
partial double bonds were studied by experimentally
DNMR spectroscopy, the barriers to rotation (∆Gcq)
determined and discussed with respect to structure: ∆Gcq
values in the amino-substituted selenoacrylamides 2 and
4 are larger than in thio analogues 1 and 3 due to the
higher polarizability of selenium; ∆Gcq in 1 are 3-5 kcal/
mol smaller than in imino analogues 3 due to the higher
bond order of C1dC2 compared with C1dN2 and the
corresponding lower “vinylogous resonance”; finally, the
barrier dependence on NR2 substitution for ∆Gcq proved
to be: pyrrolidino > NMe2 > piperidino > morpholino.
In addition, further conclusions were obtained after
calculating the C-N barriers to rotation at the ab initio
level of theory: experimentally inaccessible rotation
barriers (too low in the case of 2 and too high in the case
of 4) or not assessable by NMR due to heavily biased
conformations [in the case of N-1(N-3) methyl phenyl]
or concerted mechanisms (in the case of N-3 diaryl) are
now theoretically available and can be discussed within
the context of structure.
By means of the NBO analysis of 1-4, the occupation
numbers of the lone electron pairs of N-1/N-3, of the
bonding/antibonding π/π* orbitals of the central C1dC2/
C1dN2 partial double bonds and of the antibonding π*
orbitals of the CdS(Se) bonds were calculated and shown
to quantitatively describe thio(seleno)amide/vinylogous
thio(seleno)amide “resonance”. Thus, similar ∆Gcq values
for C1-N and C3-N restricted rotations do not indicate
the same amount of the two “resonance interactions” as
thio(seleno)amide resonance proved to be much stronger.
However, the difference to the “vinylogous resonance” is
balanced by additional N-1 lone pair/π*(C1,C2/N2) orbital
interactions.
The thiocarbonyl and the selenocarbonyl anisotropic
effects were quantitatively calculated by the GIAO
method and visualized by ICSS. They proved to be very
similar with (0.01 ppm isochemical shielding surfaces
running at about the same distances from the respective
nuclei. This equidistance of CdS and CdSe anisotropic
effects ought to be considered if the effects are to be
employed for assignment purposes in 1H NMR spectros-
copy when comparing CdS and CdSe analogues.
Experimental Section
The compounds studied were synthesized according to
reported methods (cf. Scheme 11). Thus, amino-substituted
thioacrylamides 1 and their seleno analogues 2 were pre-
pared44 by the reaction of sodium sulfide or sodium selenide,
respectively, with 1-chlorovinamidinium salts 6 which were
available by reaction of acetamides 5 with Vilsmeier reagent45
(45) Jones, G.; Stanforth, S. P. Org. React. 1997, 49, 1 and references
therein.
J. Org. Chem, Vol. 70, No. 17, 2005 6601

Kleinpeter et al.
prepared from N,N-disubstituted formamides and POCl3.46 For
the synthesis of the analogous aza thioacrylamides 3 and
selenoacrylamides 4, two different routes were used. The first
route (method A) started with N,N-disubstituted cyanamides
of transition states of the rotation about the C1-N and C3-N
partial double bonds using opt ) ts and calcfc. The NBO 5.0
population analysis53 was used linked to the Gaussian 98
program package54 with the keywords nlmo for NLMO analysis
and print for graphical evaluation. NRT analysis was per-
formed within the NBO 5.0 population analysis with nrt and
nrtthr ) 10. The results were illustrated using the program
SYBYL.55
The chemical shieldings in the surroundings of the mol-
ecules were calculated as described in ref 32. Within the
SYBYL contour file, the anisotropy/ring current effects of the
functional groups under investigation were visualized as ICSS
enabling appreciation of the spatial extension of the anisot-
ropy/ring current effects to particular protons.
7
47 which were transformed by reaction with Vilsmeier reagent
into the 1-chlor-2-azavinamidinium salts 8,48 their subsequent
reaction with sodium sulfide or sodium selenide gave rise to
the formation of the desired products 3 and 4, respectively. In
an alternative route (method B), the same compounds 3 and
4 were prepared by reaction of N,N-disubstituted thioureas
9
49 or selenoureas 10,50 respectively, with a mixture of triethyl
orthoformate and a secondary amine according to ref 51. The
sodium selenide reagent required was prepared by the reaction
of sodium borohydride and selenium in methanol.52 In a few
cases, the analogous aza selenoacrylamides 4 were synthesized
by the reaction of sodium selenide with 1-methylmercapto-2-
azavinamidinium iodides which were prepared from the
amino-substituted 2-azathioacrylamides 3 by reaction with
methyl iodide (method A1). The preparative results 1-4
(method of synthesis, yirlds, melting points, elemental analy-
ses) and general procedures are given in the Supporting
Information.
Acknowledgment. Dr. Karel D. Klika is thanked
for language correction.
Supporting Information Available: General procedures
for the synthesis of the compounds under study; tables of
experimental 1H and 13C chemical shifts; tables of the results
from NBO analysis and ab initio calculations for the geom-
etries and energies of both ground and transition states for
the restricted rotations about the C1-N and C3-N partial
double bonds; figures of isomers and conformers; and correla-
tions of experimental values with theoretically calculated C,N
rotation barriers. This material is available free of charge via
the Internet at http://pubs.acs.org.
The 1H and 13C NMR spectra were acquired at 11.75
1
(operating at 500 and 125 MHz for H and 13C, respectively)
and 7.05 T (operating at 300 and 75 MHz for 1H and 13C,
respectively) at room temperature in CD2Cl2 except for the
DNMR experiments where both higher and lower tempera-
tures were used (in the case of higher temperatures, C2D2Cl4
was substituted for CD2Cl2 as the solvent). Determination of
the actual probe temperature was made using standard NMR
thermometers. 13C NMR spectra were acquired using a 10 s
repetition time with proton broadband decoupling. All 1H and
13C chemical shifts were referenced using TMS as an internal
standard () 0 ppm for both nuclei). NMR spectra were
assigned by the application of APT and 2-D techniques such
as HMQC, HMQC-TOCSY and HMBC using standard vendor-
supplied software.
JO0503097
(47) Keil, D.; Hartmann, H. Phosphorus, Sulfur, Silicon 1999, 152,
169.
(48) (a) Liebscher, J.; Hartmann, H. Z. Chem. 1974, 14, 358. (b)
Liebscher, J.; Hartmann, H. Z. Chem. 1975, 15, 16.
(49) Hartmann, H.; Reuther, I. J. Prakt. Chem. 1973, 315, 144.
(50) Keil, D.; Hartmann, H. Synthesis 2004, 15.
(51) (a) Knoll, A.; Meissner, G.; Reist, K.; Liebscher, J. Z. Chem.
1983, 23, 19. (b) Liebscher, J.; Mitzner, E. Synthesis 1985, 414.
(52) Klayman, D. L.; Griffins, T. S. J. Am. Chem. Soc. 1973, 95, 197.
(53) Glending, E. D.; Baadenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, J. A.; Morales C. M.; Weinhold, F. NBO 5.0; Theoretical
Chemistry Institute, University of Wisconson: Madison, WI, 2001.
(54) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M.; Farkas, C. O.; Tomasi, J.; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P.; Cui, Y. Q.; Morokuma,
K. D.; Malick, K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, L.; Gomperts, R.; Martin, L.
R.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.;
Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian, Inc.:
Pittsburgh, PA, 1998.
The free energies of activation were calculated by means of
eqs 1 and 2:
π∆ν
kc )
(1)
(2)
x
2
Tc
∆Gcq ) 19.14Tc 10.32 + log
( )
[
]
kc
Quantum chemical calculations were performed on SGI
Octane R12000 and SGI Origin workstations using the Gauss-
ian 98 software package. The molecules were optimized at
different levels of theory using the keyword opt, optimization
(46) Arnold, Z. Collect. Czech. Chem. Commun. 1961, 26, 3031.
(55) SYBYL 6.9; Tripos, Inc.: St. Louis, MO, 2003.
6602 J. Org. Chem., Vol. 70, No. 17, 2005