Nature of the intramolecular Se···N nonbonded interaction of 2-selenobenzylamine derivatives. An experimental evaluation by 1H, 77Se, and 15N NMR spectroscopy

Article abstract of DOI:10.1021/ja953358h

¹H NMR analysis of seven 2-selenobenzylamine derivatives (ArSeX, 1-7) has revealed the existence of attractive nonbonded interaction between the divalent selenium and an unsymmetric amino nitrogen, whose strength significantly depends on the re

Full text of DOI:10.1021/ja953358h

J. Am. Chem. Soc. 1996, 118, 8077-8084
8077
Nature of the Intramolecular Se‚‚‚N Nonbonded Interaction of
2-Selenobenzylamine Derivatives. An Experimental Evaluation
1
77
15
by H, Se, and N NMR Spectroscopy
Michio Iwaoka and Shuji Tomoda*
Contribution from the Department of Chemistry, College of Arts and Sciences, The UniVersity of
Tokyo, Komaba, Meguro-ku, Tokyo 153, Japan
ReceiVed October 2, 1995X
Abstract: 1H NMR analysis of seven 2-selenobenzylamine derivatives (ArSeX, 1-7) has revealed the existence of
attractive nonbonded interaction between the divalent selenium and an unsymmetric amino nitrogen, whose strength
significantly depends on the relative electrophilic reactivity of the selenium moiety. In the intermediate cases (5, X
) CN; 6, X ) SPh), the dissociation energy of the Se‚‚‚N nonbonded interaction was evaluated as 12.4 and 10.8
kcal/mol, respectively, by variable-temperature 1H NMR spectral simulations. The corresponding values were roughly
estimated as >18.8 kcal/mol for the strong cases (2, X ) Br; 3, X ) Cl; 4, X ) OAc) and <7.7 kcal/mol for the
weak cases (1, X ) SeAr; 7, X ) Me). In order to elucidate physicochemical properties of the interaction, 15N-
labeled compounds were synthesized. Downfield shifts of 15N NMR (∆δN) and increment of JSe‚‚‚N (coupling constant
between 77Se and 15N) were observed with increasing Se‚‚‚N interaction. In the case of strong Se‚‚‚N interaction,
saturation of coupling constants was observed. These experimental observations as well as large NBO deletion
energies (3-21G* and LANL1DZ) between the selenium and the nitrogen for model compounds (9-13) strongly
suggested that the observed Se‚‚‚N interaction is mainly caused by the orbital interaction between the nitrogen lone
pair (nN) and the antibonding orbital of the Se-X bond (σ*Se-X).
Introduction
as the high bioactivity of selenium,9 because such an interaction
may occur at an early stage along the reaction coordinate.
Herein we wish to present the first quantitative evaluation of
one of these interactions, i.e., the nonbonded interaction between
a divalent selenium and a tertiary amino nitrogen (Se‚‚‚N
interaction), and its theoretical interpretation.
A pseudo-high-valent state has also been found in other
divalent chalcogen compounds and has been frequently dis-
cussed in the literature.10 The theory that the hypervalent
property of divalent chalcogens increases in the order O < S <
Se < Te is now widely accepted, although the dissociation
energy and the essential force of the nonbonded interaction are
still uncertain. In order to answer these physicochemical
questions, it is necessary to study a series of divalent chalcogen
compounds systematically. Divalent selenium compounds have
advantages for this purpose: (1) they are usually more stable
than organotellurium compounds, (2) they sometimes exhibit
such a strong nonbonded interaction in solution that the dynamic
process can be investigated by 1H NMR spectroscopy, and (3)
they contain an NMR-active isotope (77Se, I ) 1/2) that should
allow invaluable NMR analysis of the interaction.
Selenium in organic compounds usually exists in a divalent
state with two covalently bonded substituents and two lone pairs
of valence electrons. In the solid state and in solution, however,
it has been frequently discovered that the divalent selenium
further interacts with (a) nearby heteroatom(s) (O, N, Se, etc.)
forming a pseudo-high-valent selenium species.1-7 This phe-
nomenon has been interpreted by assuming the nonbonded
interaction between the divalent selenium and a heteroatom. The
nonbonded interaction, which is derived from the hypervalent
nature of the selenium, is expected to be responsible for the
specific redox reactivity of organoselenium compounds8 as well
X Abstract published in AdVance ACS Abstracts, August 1, 1996.
(1) Hargittai, I.; Rozsondai, B. Structural Chemistry of Organic Com-
pounds Containing Selenium or Tellurium. In The Chemistry of Organic
Selenium and Tellurium Compounds; Patai, S., Rappoport, Z., Eds.; John
Wiley & Sons: New York, 1986; Vol. 1.
(2) (a) Baiwir, M.; Llabre`s, G.; Dideberg, O.; Dupont, L.; Piette, J.-L.
Acta Crystallogr. 1975, B31, 2188. (b) Roesky, H. W.; Weber, K.-L.;
Seseke, U.; Pinkert, W.; Noltemeyer, M.; Clegg, W.; Sheldrick, G. M. J.
Chem. Soc., Dalton Trans. 1985, 565. (c) Tomoda, S.; Iwaoka, M.; Yakushi,
K.; Kawamoto, A.; Tanaka, J. J. Phys. Org. Chem. 1988, 1, 179. (d)
Tomoda, S.; Iwaoka, M. J. Chem. Soc., Chem. Commun. 1990, 231. (e)
Kubiniok, S.; du Mont, W.-W.; Pohl, S.; Saak, W. Angew. Chem., Int. Ed.
Engl. 1998, 27, 431. (f) du Mont, W.-W.; Martens, A.; Pohl, S.; Saak, W.
Inorg. Chem. 1990, 29, 4847.
(3) Ramasubbu, N.; Parthasarathy, R. Phosphorus Sulfur 1987, 31, 221.
(4) (a) Goldstein, B. M.; Kennedy, S. D.; Hennen, W. J. J. Am. Chem.
Soc. 1990, 112, 8265. (b) Burling, F. T.; Goldstein, B. M. J. Am. Chem.
Soc. 1992, 114, 2313.
(5) Barton, D. H. R.; Hall, M. B.; Lin, Z.; Parekh, S. I.; Reibenspies, J.
J. Am. Chem. Soc. 1993, 115, 5056.
In the solid state, pseudo-high-valent behavior of a divalent
selenium has been reported in numerous crystals.2 Ramasubbu
and Parthasarathy first analyzed it statistically and found that
there are two types of nonbonded interactions involving a
divalent selenium (Figure 1).3 Type A selenium weakly donates
(9) (a) Flohe´, L. Glutathione Peroxidase Brought into Focus. In Free
Radicals in Biology; Pryor, W. A., Ed.; Academic Press: New York, 1982.
(b) Shamberger, R. J. Biochemistry of Selenium; Plenum Press: New York,
1983. (c) Epp, O.; Ladenstein, R.; Wendel, A. Eur. J. Biochem. 1983, 133,
51.
(6) Iwaoka, M.; Tomoda, S. Phosphorus Sulfur Silicon Relat. Elem. 1992,
67, 125.
(7) Iwaoka, M.; Tomoda, S. J. Org. Chem. 1995, 60, 5299.
(8) (a) Clive, D. L. J. Tetrahedron 1978, 34, 1049. (b) Reich, H. J.
Acc. Chem. Res. 1979, 12, 22. (c) Nicolaou, K. C.; Petasis, N. A. Selenium
in Natural Products Synthesis; CIS: Philadelphia, 1984. (d) Paulmier, C.
Selenium Reagents and Intermediates in Organic Synthesis; Pergamon
Press: Oxford, 1986. (e) Organoselenium Chemistry; Liotta, D., Ed.; John
Wiley & Sons: New York, 1987.
(10) (a) The Chemistry of Organic Selenium and Tellurium Compounds;
Patai, S., Rappoport, Z., Eds.; John Wiley & Sons: New York, 1986; Vol.
1. (b) Schultz, G.; Hargittai, I.; Kapovits, I.; Kucsman, A. J. Chem. Soc.,
Faraday Trans. 2 1984, 80, 1273. (c) Tayor, R.; Kennard, O. Acc. Chem.
Res. 1984, 17, 320. (d) Wiberg, K. B.; Waldron, R. F.; Schulte, G.;
Saunders, M. J. Am. Chem. Soc. 1991, 113, 971. (e) Steiner, T.; Saenger,
W. J. Am. Chem. Soc. 1993, 115, 4540.
S0002-7863(95)03358-0 CCC: $12.00 © 1996 American Chemical Society

8078 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Iwaoka and Tomoda
Scheme 1
using various methods. They include variable temperature (VT)
1
1H NMR of two benzylic protons, dynamic H NMR analysis
of 5 and 6, 77Se and 15N NMR chemical shifts analysis,
measurement of JSe‚‚‚N coupling constants, and ab initio mo-
lecular orbital (MO) calculations and natural bond orbital (NBO)
analysis. On the basis of the results from these analyses, the
mechanism of the Se‚‚‚N nonbonded interaction is discussed.
Figure 1. Nonbonded interaction involving a divalent selenium.
the selenium lone pair to atom X (X ) Pt, Hg, etc.), which is
located just above the divalent selenium plane. This type is
observed in the C-H‚‚‚Se hydrogen bond recently reported by
us.11 On the contrary, type B selenium partially accepts lone
pair electrons of atom Y (Y ) O, N, Cl, etc.), which is located
at the backside of one of the covalent bonds. These variable
structural features can be reasonably attributed to the orbital
interaction (i.e., the charge transfer interaction) between the
selenium and electrophile X or nucleophile Y.
Results and Discussion
1. Synthesis. Six 2-selenobenzylamine derivatives (2-7)
were synthesized from diselenide 16 according to the previously
known methods (Scheme 1).14 All compounds were obtained
in almost quantitative yields except for 7. Similarly, corre-
sponding 15N-labeled compounds were easily synthesized from
15N-labeled 1, which was synthesized from aniline-15N according
to Scheme 2. Selenenyl bromide 2,6 selenenyl chloride 3, and
selenenyl acetate 412 were relatively stable in solution as
compared with the corresponding benzeneselenenyl derivatives
and did not rapidly decompose in CDCl3 in the presence of a
small amount of water and oxygen. This may be due to
significant stabilization by strong Se‚‚‚N nonbonded interaction
as described below. Selenocyanate 5, selenenyl sulfide 6,12 and
methyl selenide 7 were indefinitely stable and could be
characterized by elemental analysis.
2. 1H NMR of Benzylic Protons. Compounds 1-7 have
an unsymmetrical tertiary amino nitrogen at the position
separated by four bonds from a divalent selenium. Therefore,
it is expected that two benzylic protons become magnetically
nonequivalent when the unsymmetric nitrogen inversion or the
rotation around the C-N bond is sufficiently slow within an
NMR time scale due to the intramolecular Se‚‚‚N interaction,
which allows the formation of a five-membered ring.
1H NMR chemical shifts of benzylic protons for 1-7 are
listed in Table 1. According to the NMR spectral properties,
1-7 can be classified into three groups. The first group, which
contains 2-4, shows a strong Se‚‚‚N interaction. The benzylic
protons of these compounds resonate as an AB quartet at room
temperature. At higher temperatures the NMR signals signifi-
cantly broaden, resulting in irreversible decomposition. The
second group, which contains 5 and 6, possesses an intermediate
Se‚‚‚N interaction, showing a dynamic mutual exchange process
between two benzylic protons. The benzylic protons of 5 and
6 resonate as a singlet at room temperature. These signals
broaden and split into an AB quartet upon cooling. Detailed
More advanced analyses were independently made by two
research groups as to an Se‚‚‚O interaction that belongs to type
B. Goldstein et al. showed an electrostatic character of Se‚‚‚O
interaction in selenazofurins.4 On the other hand, Barton et al.
analyzed the Se‚‚‚O interaction of selenoiminoquinones and
suggested a three-center four-electron interaction on the basis
of X-ray structure and Bader electron density analysis.5 These
results clearly suggested that there is a wide spectrum from ionic
to covalent character in the nonbonded interaction involving a
divalent selenium. Thus, the interaction has been thoroughly
studied in some particular cases, but it has never been
systematically analyzed in a series of organoselenium com-
pounds and no quantitative experimental data have been
available as to the dissociation energy of the interaction.
We have recently found Se‚‚‚N nonbonded interaction of
diselenide 1 in the solid state6 and have shown the possibility
that the interaction may play important roles in the catalytic
cycle of glutathione peroxidase, a selenium-containing antioxi-
dant enzyme,12 and that it can be applied to novel synthetic
reactions.13 More recently, we have for the first time succeeded
in the X-ray structural determination of areneselenenyl chloride
strongly stabilized by Se‚‚‚N nonbonded interaction,7 in which
the predominant force is the n-σ*-type orbital interaction
between the nitrogen and the selenenyl chloride moiety. In this
paper, Se‚‚‚N nonbonded interaction of a series of 2-seleno-
benzylamine derivatives (1-7) is analyzed quantitatively by
(11) Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1994, 116, 4463.
(12) Ioaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1994, 116, 2557.
(13) (a) Iwaoka, M.; Tomoda, S. J. Chem. Soc., Chem. Commun. 1992,
1165. (b) Fujita, K.; Iwaoka, M.; Tomoda, S. Chem. Lett. 1994, 923. (c)
Fujita, K.; Murata, K.; Iwaoka, M.; Tomoda, S. Tetrahedron Lett. 1995,
36, 5219. (d) Fujita, K.; Murata, K.; Iwaoka, M.; Tomoda, S. J. Chem.
Soc., Chem. Commun. 1995, 1641.
(14) (a) Schmid, G. H.; Garratt, D. G. J. Org. Chem. 1983, 48, 4169.
(b) Tomoda, S.; Takeuchi, Y.; Nomura, Y. Chem. Lett. 1981, 1069. (c)
Sharpless, K. B.; Young, M. W. J. Org. Chem. 1975, 40, 947.

Se‚‚‚N Interaction of 2-Selenobenzylamine DeriVatiVes
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8079
Scheme 2
Scheme 3
1
Table 1. 90 MHz H NMR of Benzylic Protons for 1-7
lone pair (nN) and the low-lying antibonding orbital (σ*Se-X
)
compd (ArSeX)
δHa/ppm
(2JHH/Hz)
Tcb/K
of the selenium moiety.
1
3. VT H NMR Analysis. The dynamic process of the
mutual exchange between two benzylic protons (Scheme 3) was
observed only for 5 and 6. The barrier of this process should
be related to both the strength of Se‚‚‚N interaction and the
barrier of nitrogen inversion or C-N rotation because the
magnetic nonequivalence arises from the chirality of the
unsymmetrical tertiary amine. The benzylic protons of N-
cyclohexyl-N-methylbenzylamine (8),17 which does not have a
selenium substituent at the 2-position of the phenyl ring, resonate
as sharp singlet even at -90 °C at 500 MHz, indicating that
the nitrogen is magnetically achiral if the Se‚‚‚N interaction is
absent for 1-7.
2: X ) Br
3: X ) Cl
4: X ) OAc
4.15, 3.83
4.16, 3.92
4.05, 3.83
(13.6)
(13.9)
(13.1)
>320
>320
>320
5: X ) CN
6: X ) SPh
3.68
241
217
3.75, 3.54c
3.73
(13.8)
(13.0)
3.82, 3.62d
1: X ) SeAr
7: X ) Me
3.72
3.67
<183
<183
a Chemical shifts measured in CDCl3 at room temperature. b Coa-
lescence temperature in CDCl3 for 2-4 and in CD2Cl2 for 1 and 5-7.
c Measured in CD2Cl2 at 215 K. d Measured in CD2Cl2 at 196 K.
Figure 2 compares 90 MHz 1H NMR spectra of 5 at various
temperatures (left) with the simulated ones (right).18 Similar
spectral changes were observed for 6.19 Coalescence temper-
atures (Tc) were 241 and 217 K for 5 and 6, respectively,
indicating a higher activation barrier of the mutual exchange
(Scheme 3) for 5 than that for 6. According to the Eyring plots
(Figure 3), the activation enthalpy (∆Hq) and the activation
entropy (∆Sq) for 5 were determined as 19.1 ( 1.6 kcal/mol
and +28.9 ( 6.7 eu, and those for 6 were 17.5 ( 1.1 kcal/mol
and +30.8 ( 5.0 eu, respectively. The ∆Hq values showed
that the mutual exchange of two benzylic protons was 1.6 kcal/
mol more difficult for 5 than 6. The observed large positive
∆Sq values were very consistent with the assumption that the
Se‚‚‚N interaction dissociated during the mutual exchange
process.
4. Se‚‚‚N Dissociation Energy. The mutual exchange of
two benzylic protons requires several individual steps. They
should include (1) the dissociation of Se‚‚‚N nonbonded
interaction accompanied by internal rotation around the Ar-C
and Ar-Se bonds, (2) the rotation around the C-N bond, (3)
the inversion of the unsymmetrical tertiary amine, and (4) the
association of the selenium with the nitrogen to regenerate the
Se‚‚‚N interaction. A possible mechanism is shown in Figure
4 with a hypothetical reaction coordinate.
analysis of the dynamic processes is described in the following
two sections. The third group, which contains 1 and 7, has a
1
weak (or no) Se‚‚‚N interaction. The singlet H NMR signals
of the benzylic protons for these compounds broaden at low
temperature but do not split even at -90 °C at 500 MHz. Since
the weak Se‚‚‚N nonbonded interaction has been found in the
solid state of 1,6 a similar interaction may exist in 1 in solution.
However, it is not clear at present whether weak Se‚‚‚N
interaction exists in 7 or not according to the NMR line
broadening of benzylic protons, though in fact a weak interaction
may exist in 7 on the basis of ab initio MO calculation as
described later.
The above classification indicates that the strength of the
Se‚‚‚N interaction correlates with relative electrophilicity of the
selenium moiety (the Se-X bond). In general, selenenyl
bromides, selenenyl chlorides, and selenenyl acetates smoothly
add to simple olefins to produce trans-1,2-adducts,15 whereas
selenocyanates and selenosulfonates can add only to activated
olefins or simple olefins in the presence of a Lewis acid.16
Diselenides and methyl selenides are totally unreactive toward
olefins under the usual reaction conditions. The observed
correlation between the strength of the Se‚‚‚N interaction and
the electrophilic reactivity of the selenium moiety strongly
suggests that the Se‚‚‚N interaction may represent the initial
stage or near-transition-state of the electrophilic reaction of a
divalent selenium with a nitrogen lone pair, suggesting that it
should mainly arise from orbital interaction between the nitrogen
The activation enthalpy (∆Hqexp) determined by the VT NMR
method was approximately equal to the highest barrier of the
reaction coordinate. Since the rotational barrier E1 and the
nitrogen inversion barrier E3 may be small (∼1.5 and 3.8 kcal/
mol, respectively, according to MO calculations20), the ∆Hq
exp
(15) (a) Sharpless, K. B.; Lauer, R. F.; Teranishi, A. Y. J. Am. Chem.
Soc. 1973, 95, 6137. (b) Reich, H. J. J. Org. Chem. 1974, 39, 428. (c)
Sharpless, K. B.; Lauer, R. F. J. Org. Chem. 1974, 39, 429. (d) Reich, H.
J.; Renga, J. M.; Reich, I. L. J. Am. Chem. Soc. 1975, 97, 5434.
(16) (a) Toshimitsu, A.; Uemura, S.; Okano, M. J. Chem. Soc., Chem.
Commun. 1977, 166. (b) Uemura, S.; Toshimitsu, A.; Aoki, T.; Okano,
M. Chem. Lett. 1979, 1359. (c) Tomoda, S.; Takeuchi, Y.; Nomura, Y. J.
Chem. Soc., Chem. Commun. 1982, 871. (d) Back, T. G.; Collins, S.
Tetrahedron Lett. 1980, 21, 2213. (e) Back, T. G.; Collins, S. Tetrahedron
Lett. 1980, 21, 2215.
can be estimated as the sum of Se‚‚‚N dissociation energy (DE)
and the C-N rotational barrier (E2), which may be equal to the
(17) Glover, S. A.; Warkentin, J. J. Org. Chem. 1993, 58, 2115.
(18) DNMR5 program was used for NMR line shape analysis. Stephen-
son, D. S.; Binsch, G. QCPE 1978, 11, 365.
(19) Data are indicated in the supporting information.
(20) The detailed method of the E1, E2, E3 estimation is described in the
Experimental Section.

8080 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Iwaoka and Tomoda
is not unreasonable as compared with the experimental barrier
for N-isopropyl-N-methylbenzylamine (∆Gq ) 6.2 kcal/mol).22
By subtracting this value from the ∆Hq
for the mutual
exp
exchange process of two benzylic protons, the DE of Se‚‚‚N
interaction can be determined as 12.4 ( 1.6 kcal/mol for 5 and
10.8 ( 1.1 kcal/mol for 6. According to 1H NMR analysis for
1-7 (Table 1), it is obvious that the dissociation energy for
2-4 is larger than that for 5 and that those for 1 and 7 are
smaller than that for 6. By assuming that the ∆Sq value is equal
to +30 eu for these compounds and that the chemical shift
difference of two benzylic protons and their coupling constant
are 18 and 13.5 Hz, respectively, the DE can be roughly
estimated as >18.8 kcal/mol for 2-4 and <7.7 kcal/mol for 1
and 7 based on the well-known correlation between ∆Gq and
Tc,23 which is higher than 320 K for 2-4 and lower than 183
K for 1 and 7.
5. 77Se NMR Chemical Shift. In order to elucidate the
mechanism of the formation of Se‚‚‚N nonbonded interaction,
we subsequently analyzed 77Se and 15N NMR of the 15N-labeled
compounds. The observed 77Se NMR chemical shifts (δSe), 15
N
NMR chemical shifts (δN), and JSe‚‚‚N coupling constants of 15N-
labeled 1-7 are listed in Table 2. It is known that the 77Se
NMR chemical shift is susceptible to the electronic structure
around the selenium and has a wide range from -100 ppm to
+1100 ppm in a divalent state with respect to dimethyl
selenide.24 The observed 77Se NMR chemical shifts of 1-7
almost cover this range.
The comparison of 77Se NMR chemical shifts of 1-7 with
those of the corresponding 2-methyl benzeneselenenyl deriva-
tives will be useful for elucidating the Se‚‚‚N interaction.
Unfortunately, however, there was only one piece of data
available in the literature: 2-methylphenyl methyl selenide
shows a 77Se NMR absorption at 161.6 ppm,25 which is shifted
upfield by 19 ppm with respect to methyl selenide 7, suggesting
a small downfield shift due to the Se‚‚‚N interaction. On the
other hand, when the 77Se NMR chemical shifts of 1-3 and
5-7 are compared with those of unsubstituted benzeneselenenyl
derivatives in the literature,24 similar downfield shifts are
observed for 2 (-142 ppm), 3 (-9 ppm), 5 (-41 ppm), and 6
(-46 ppm). The opposite upfield shifts are observed for 1 (+32
ppm) and 7 (+21 ppm). Since an o-methyl substituent results
in a large upfield shift of 77Se NMR (+40 ppm for methyl
selenide), it is strongly suggested that the intramolecular Se‚‚‚N
nonbonded interaction results in an apparent downfield shift of
77Se NMR. Similar phenomena have previously been reported
by Baiwir et al. in an intramolecular Se‚‚‚O interaction.26
However, the values of downfield shifts do not seem to correlate
with the strength of Se‚‚‚N and Se‚‚‚O interaction.
1
Figure 2. Variable-temperature H NMR spectra of benzylic protons
for 5 (left) and the simulated ones (right).
6. 15N-Induced Isotope Shifts in 77Se NMR. We subse-
quently examined isotope shifts in 77Se NMR due to the
substitution of the natural abundant 14N with the 15N isotope.
The observed isotope shifts (listed in Table 2) scatter from -1.2
ppm to +0.2 ppm. Because of the lack of data about this kind
of isotope shifts in the literature, a detailed analysis cannot be
made at this moment. However, it is suggested that the strong
Figure 3. Eyring plots of the mutual exchange process of two benzylic
protons for 5 (circles) and 6 (squares).
(22) Reny, J.; Wang, C. Y.; Bushweller, C. H.; Anderson, W. G.
Tetrahedron Lett. 1975, 503.
corresponding barrier of 8. However, E2 could not be experi-
mentally determined by VT 1H NMR analysis of 8 (∆Gq , 8.9
kcal/mol).21 We therefore estimated it by ab initio MO
calculations.20 The calculated barrier was 6.7 kcal/mol, which
(23) The equations, kc ) π2-1/2(∆δ2 + 6J2)1/2 ) (RTc/Nh) exp(-∆Gq/
RTc), were employed.
(24) (a) Nakanishi, W.; Ikeda, Y.; Iwamura, H. Org. Magnet. Res. 1982,
20, 117. (b) Luthra, N. P.; Odom, J. D. Nuclear Magnetic Resonance and
Electron Spin Resonance Studies of Organic Selenium and Tellurium
Compounds. In The Chemistry of Organic Selenium and Tellurium
Compounds; Patai, S., Rappoport, Z., Eds.; John Wiley & Sons: New York,
1986; Vol. 1.
(25) MacFarlane, W.; Wood, R. J. J. Chem. Soc. A 1972, 1397.
(26) Llabre`s, G.; Baiwir, M.; Piette, J.-L.; Christiaens, L. Org. Magn.
Reson. 1981, 15, 152.
(21) Two benzylic protons of 8 resonated as a sharp singlet even at -90
°C due to rapid nitrogen inversion or C-N rotation. The activation barrier
was estimated as ,8.9 kcal/mol according to the well-known correlations23
by assuming that the chemical shift difference of two benzylic protons and
their coupling constant were 18 and 13.5 Hz, respectively.

Se‚‚‚N Interaction of 2-Selenobenzylamine DeriVatiVes
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8081
Figure 4. Plausible mechanism of the mutual exchange of two benzylic protons with a hypothetical reaction coordinate.
Table 2. 77Se and 15N NMR Chemical Shifts and JSe‚‚‚N Coupling
a spin-dipolar term, a spin-orbital term, and a Fermi-contact
Constants of 15N-Labeled 1-7
term. Among these, the Fermi-contact term is usually a major
component of a coupling constant because the spin state of the
one nucleus transmits to the other nucleus via electrons, among
which s-electrons are most important due to the possibility of
the direct contact (Fermi contact) at the nuclear position. Pople
and Santry29 showed that this term is proportional to the square
of the s-bond character between two atoms. Therefore, JSe‚‚‚N
should approximately depend on the s-bond order of the Se‚‚‚N
interaction if the spin-dipolar and spin-orbital terms are
negligible.
compd
(ArSeX)
77Se NMRa
δSe/ppm
15 NMR
δN/ppm
coupling constantc
JSe‚‚‚N/Hz
2: X ) Br
3: X ) Cl
4: X ) OAc
1011.2 (-1.2)
1051.3 (-1.0)
1158.1 (-1.0)
-298.2
-300.3
-311.2
47.0
48.6
59.3
5: X ) CN
6: X ) SPh
362.2 (+0.2)
571.5 (0.0)
-326.1
-323.1
51.9
53.1
1: X ) SeAr
7: X ) Me
431.9 (0.0)
181.0 (-0.2)
-327.7
-332.2
32.4
13.0
The observed coupling constants (Table 2) are significantly
increased with the increase in the strength of the Se‚‚‚N
interaction as expected, though those for 2 and 3 are saturated.
This saturation may be due to the participation of a spin-dipolar
or a spin-orbital term that will be relatively large in the case
that the Se‚‚‚N interaction possesses a higher bond order.
Again, the reverse order of the magnitude of JSe‚‚‚N is observed
for 5 and 6. This may be also due to the relatively small
covalent character of the Se‚‚‚N nonbonded interaction of 5 with
respect to 6 as suggested by the discussion in the previous
section (the downfield shifts of 15N NMR).
9. MO Calculations. According to the above NMR
analyses, two statements seem reasonable: (1) the strength of
the Se‚‚‚N interaction increases with the increase in the
electrophilic reactivity of the selenium moiety, and (2) the
Se‚‚‚N interaction mostly arises from the orbital interaction
between the selenium and the nitrogen. In order to ascertain
these, ab initio MO calculations and their NBO analysis30 were
performed on 9-13. According to our previous MO calcula-
tions on 9 at various basis set levels,7 it has been found that the
minimum limit of a basis set to treat such pseudo-high-valent
selenium species is 3-21G*31 or LANL1DZ.32 We therefore
employed these basis sets for the MO calculations of 10-13.
The fully optimized geometries at the HF/3-21G* and HF/
a 77Se NMR chemical shifts in CDCl3 with dimethyl selenide as an
external standard. The values in parentheses are isotope shifts in 77Se
NMR due to the substitution of 14N with 15N. A negative value means
a downfield shift. b 15N NMR chemical shifts with nitromethane (δ
+5.0) as an external standard. c Coupling constants measured from
expanded 77Se NMR spectra.
Se‚‚‚N interaction results in a slight downfield 15N-induced
isotope shift in 77Se NMR.
7. 15N NMR Chemical Shifts. The chemical shift of an
amino nitrogen (δN) is significantly influenced by the electron
delocalization of the nitrogen lone pair.27 Therefore, it should
be a good measure for the strength of the Se‚‚‚N interaction
for 1-7 if a covalent interaction is an essential force. Observed
δN of 1-7 (Table 2) is shifted downfield by 30 ppm as the
strength of the Se‚‚‚N interaction increases except for 5, showing
that the lone pair of the nitrogen efficiently delocalizes with
increasing Se‚‚‚N interaction. However, the δN of 5 is slightly
shifted upfield with respect to 6. According to the results from
1
VT H NMR analysis, it is anticipated that the δN of 5 should
be shifted downfield with respect to 6. This reversal can be
reasonably understood by assuming that the Se‚‚‚N interaction
of 5 has a less covalent character than 6, but it has a more ionic
character due to the larger electron-withdrawing ability of the
cyano group. Thus, the Se‚‚‚N interaction of 5 becomes slightly
stronger than that of 6.
8. JSe‚‚‚N Coupling Constants. Another useful measure for
the covalent strength of the Se‚‚‚N interaction is a direct
coupling constant between 77Se and 15N (JSe‚‚‚N). The NMR
coupling constant can be theoretically divided into three terms;28
(29) Pople, J. A.; Santry, D. P. Mol. Phys. 1964, 8, 1.
(30) (a) Reed, A. E.; Weinhold, F. J. Chem. Phys. 1983, 78, 4066. (b)
Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83,
735. (c) Reed, A. E.; Weinhold, F. J. Chem. Phys. 1985, 83, 1736. (d)
Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899.
(31) A standard 3-21G basis set with d polarization functions for heavier
atoms.39
(27) Levy, G. C.; Lichter, R. L. Nitrogen-15 NMR Spectroscopy; John
Wiley: New York, 1979.
(28) Ramsey, N. F. Phys. ReV. 1953, 91, 303.
(32) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. (b)
Wadt, W. R.; Hay, P. J. Ibid. 1985, 82, 284. (c) Hay, P. J.; Wadt, W. R.
Ibid. 1985, 82, 299.

8082 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Iwaoka and Tomoda
Table 3. Molecular Structures of 9-13 Obtained by ab Initio MO
Calculations at HF/3-21G* and HF/LANL1DZ Levels along with
Structural Parameters of 14 and 1 Determined by X-ray Analysis
compd
(ArSeX)
r(Se-X)/ r(Se‚‚‚N)/ N‚‚‚Se-X/ ωb/
basis seta
Å
Å
deg
deg
14: X ) Clc
2.432
2.290
LANL1DZ 2.448
1.826
LANL1DZ 1.863
2.191
2.331
2.359
2.513
2.653
178.4
174.6
174.4
170.4
170.0
26
9: X ) Cld
3-21G*
35.6
32.9
35.1
37.5
10: X ) OH 3-21G*
11: X ) CN 3-21G*
1.881
LANL1DZ 1.889
2.631
2.826
2.752
2.845
170.6
168.2
172.0
169.0
42.9
47.2
47.8
48.1
12: X ) SMe 3-21G*
2.202
LANL1DZ 2.358
LANL1DZ levels are listed in Table 3 along with X-ray
structural parameters of the related compounds (147 and 16).
The global energy minima of 9-13 have the conformation with
an Se‚‚‚N van der Waals contact like X-ray structures of 14
and 1. The obtained Se‚‚‚N interatomic distance increases in
the order 9 (X ) Cl) < 10 (X ) OH) < 11 (X ) CN) < 12
(SMe) < 13 (Me), which is completely in accordance with the
experimental results from 1H NMR of 1-7 without the reversal
of the order. The fact that the averaged Se‚‚‚N distance (2.87
Å) of 1 determined by X-ray analysis is larger than that of 12
and smaller than that of 13 is also consistent, suggesting the
accuracy of these calculations. It should be noted that methyl
selenide 13 also has weak Se‚‚‚N interaction in its global energy
minimum conformation. The results from two levels of
calculations are slightly different from each other, but they are
internally consistent within one basis set level. It is clear that
a strong Se‚‚‚N interaction causes shortening of the Se‚‚‚N
interatomic distance, increase in the linearity of the N‚‚‚Se-X
alignment, and decrease in ω. The N-C-C(Ar) bond angle
narrows by 3-4°.
1: X ) SeAre
2.362
2.78
2.96
2.924
3.185
172
165
168.2
162.4
52
57
53.4
61.0
13: X ) Me 2-21G*
1.968
LANL1DZ 1.986
a 3-21G* is a standard 3-21G basis set with d polarization functions
for Se, S, and Cl.39 LANL1DZ is a Los Alamos ECP+DZ basis set
for Se, S, and Cl and Dunning/Huzinaga valence double-ú basis set
for other atoms. b A dihedral angle of an Ar-C bond. c X-ray structure
from ref 7. d Data from ref 7. e X-ray structure from ref 6.
Table 4. Results of NBO Deletion Analysis on 9-13
compd
(ArSeX)
9: X ) Cle
∆Edel
/
∆q(σ*)c/
electron
dq(n)d/
electron
basis seta
kcal‚mol-1
3-21G*
LANL1DZ
3-21G*
47.0
47.4
22.9
12.9
+0.133
+0.116
+0.006
+0.046
-0.143
-0.141
-0.069
-0.038
10: X ) OH
LANL1DZ
11: X ) CN
12: X ) SMe
3-21G*
LANL1DZ
3-21G*
15.8
6.9
11.7
7.1
+0.045
+0.022
+0.035
+0.031
-0.050
-0.018
-0.038
-0.025
LANL1DZ
13: X ) Me
3-21G*
LANL1DZ
6.3
2.1
+0.018
+0.006
-0.020
-0.006
a 3-21G* is a standard 3-21G basis set with d polarization functions
for Se, S, and Cl.39 LANL1DZ is a Los Alamos ECP+DZ basis set
for Se, S, and Cl and Dunning/Huzinaga valence double-ú basis set
for other atoms. b NBO deletion energy between the divalent selenium
moiety and the amino nitrogen atom. c Charge increase of the σ* orbital
of an Se-X bond due to the orbital interaction between the selenium
moiety and the nitrogen. d Charge decrease of the nN orbital of an amino
nitrogen due to the orbital interaction between the selenium moiety
and the nitrogen. e Data from ref 7.
10. NBO Analysis.30 The results of NBO analysis on 9-13
are listed in Table 4. NBO deletion energy (∆Edel) in the table
means the orbital interaction energy between the divalent
selenium moiety and the tertiary amino nitrogen. The values
∆q(σ*) and ∆q(n) are the charge increase of the σ* orbital of
an Se-X bond due to the Se‚‚‚N interaction formation and the
corresponding charge decrease of the nitrogen lone pair for
9-13, respectively.
The ∆Edel value obtained at a LANL1DZ basis set level
increases in the order 13 (X ) Me) < 11 (X ) CN) ≈ 12 (X
) SMe) < 10 (X ) OH) < 9 (X ) Cl). The order is the same
as the order of the Se‚‚‚N orbital interaction strength suggested
by the 15N NMR downfield shifts (∆δN) and the JSe‚‚‚N coupling
constants for 1-7. It is quite reasonable that the ∆Edel for 12
(X ) SMe) is larger than that for 11 (XCN) in spite of the
larger Se‚‚‚N interatomic distance for 12. This fact clearly
suggests that the δN and JSe‚‚‚N are good indices for the extent
of the Se‚‚‚N orbital interaction though JSe‚‚‚N values for 2 and
3 seem to be saturated. At the 3-21G* basis set level, however,
the ∆Edel increases in the order, 13 < 12 < 11 < 10 < 9, which
is the same with the order of the total Se‚‚‚N interaction energy.
The discrepancy between the results from two different basis
set levels may be due to the incorporated relativistic effects or
the lack of d orbitals for Se in LANL1DZ. In order to elucidate
this point, higher level calculations, such as MP2, which includes
electron correlation effects, using a more accurate basis set will
be required.
The obtained ∆Edel values for 11 (15.8 kcal at 3-21G* and
6.9 kcal at LANL1DZ) and 12 (11.7 kcal at 3-21G* and 7.1
kcal at LANL1DZ) are comparable with the experimentally
determined ∆E′s (12.4 kcal for 5 and 10.8 kcal for 6). This
suggests that the greater part of the Se‚‚‚N nonbonded interaction
should arise from the orbital interaction between the divalent
selenium moiety (the Se-X bond) and the nitrogen atom (Figure
5). The significantly large changes of the orbital occupancies
are only observed for σ*Se-X and nN natural orbitals (∆q(σ*)
and ∆q(n)) with marginal changes for the bonding orbital of
the Se-X bond (σSe-X) and the Rydberg orbitals on Se. These
results clearly shows that the orbital interaction between the
divalent selenium moiety and the tertiary amino nitrogen is
predominant for the Se‚‚‚N nonbonded interaction as well as
that the main origin of the interaction is the electron delocal-
ization from the nitrogen lone pair to the low-lying σ* orbital
of an Se-X bond.

Se‚‚‚N Interaction of 2-Selenobenzylamine DeriVatiVes
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8083
nitrogen. Other organic solvents were used without purification. 1H,
13C, 77Se, and 15N NMR spectra were measured on a JEOL R-500
spectrometer in CDCl3 containing tetramethylsilane as an internal
standard for 1H and 13C NMR. For 77Se or 15N NMR, dimethyl selenide
(δ 0 ppm) or nitromethane (δ 5 ppm) in CDCl3 was used as an external
standard, respectively. 15N-induced isotope shifts in 77Se NMR were
measured at 297 K in CDCl3. The solutions were prepared separately
for a natural abundant sample and the 15N-labeled one. JSe‚‚‚N coupling
constants of 77Se NMR absorptions were measured from the expanded
1H-decoupled 77Se NMR spectra of 15N-labeled 1-7. Variable-
temperature 1H NMR spectra were recorded on a JEOL FX90Q
spectrometer in CHCl3 for 1-7 at room or higher temperature and in
CD2Cl2 for 1 and 5-7 at low temperature. The low-resolution mass
spectrum was measured on a Shimadzu QP1000 spectrometer operating
at 70 eV by using the direct insertion method. Diselenide 1 was
synthesized by the literature method6 and was characterized by 1H, 13C,
and 77Se NMR spectra and elemental analysis: 77Se NMR δ 431.9
(429.8 ppm in the literature6). 2, 4, and 6 were synthesized from 1
according to the literature methods6,12 and were fully characterized by
1H, 13C, and 77Se NMR (see the supporting information) as well as by
elemental analysis for 6 (see below).
2-[(N-Cyclohexyl-N-methylamino)methyl]benzeneselenenyl Bro-
mide (2). 2 was quantitatively synthesized from 1 by the reaction with
an equimolar amount of bromide in CH2Cl2 according to the literature
method.6 1H, 13C, and 77Se NMR spectra of 2 are shown in the
supporting information. 77Se NMR: δ 1010.0 (1019.4 ppm in the
literature6).
2-[(N-Cyclohexyl-N-methylamino)methyl]benzeneselenenyl Chlo-
ride (3). 1 (28.1 mg, 0.05 mmol) was dissolved in dry THF (2 mL)
under nitrogen atmosphere, and SO2Cl2 (4.0 mL, 0.05 mmol) was added
to the solution. After 1 h the reaction mixture was concentrated under
reduced pressure. 3 was quantitatively obtained as a yellow-white
powder. Spectral data for 3: 1H NMR δ 1.3 (m, 5H), 1.9 (m, 5H),
2.61 (s, 3H), 3.08 (m, 1H), 3.92 (d, 1H, J ) 13.9 Hz), 4.16 (d, 1H, J
) 13.9 Hz), 7.1-7.3 (m, 3H), 8.1-8.2 (m, 1H); 13C NMR δ 25.3,
25.4, 25.6, 26.3, 30.9, 38.6, 63.1, 67.2, 125.3, 126.3, 129.0, 129.1, 134.7,
136.8; 77Se NMR δ 1050.3. 1H, 13C, and 77Se NMR spectra of 3 are
shown in the supporting information.
Figure 5. Electron delocalization from the nitrogen lone pair (nN) to
the antibonding orbital of an Se-X bond (σ*).
Conclusions
The strength of the Se‚‚‚N nonbonded interaction was
quantitatively evaluated for 5 (12.4 ( 1.6 kcal/mol) and 6 (10.8
( 1.1 kcal/mol) and was roughly estimated for 2-4 (>18.8
kcal/mol) and 1 and 7 (<7.7 kcal/mol) on the basis of dynamic
1H NMR study of diastereotopic benzylic protons and the
calculated C-N rotational barrier of 8. A strong correlation
between the electrophilic reactivity of the selenium moiety and
the strength of Se‚‚‚N interaction was clear. The estimated
dissociation energy for 2-4 was unexpectedly large as compared
with the bond dissociation energy of the Se-Se covalent bond
in diphenyl diselenide (64.9 ( 1.2 kcal/mol)33 and that of the
Se-Br bond in benzeneselenenyl bromide (61.7 kcal/mol).34
77Se NMR analysis of 15N-labeled 1-7 showed that a small
downfield isotope shift, due to the substitution of the natural
abundant 14N with the 15N isotope, arises from the strong Se‚‚‚N
interaction. A large downfield shift of 15N NMR and a
significant enhancement of JSe‚‚‚N coupling constants were also
observed with an increase in the Se‚‚‚N interaction, though JSe‚‚‚N
seems to be saturated for 2 and 3. These experimental
observations are reasonably understood by assuming a predomi-
nant n-σ* orbital interaction between the nitrogen and the
Se-X bond, because δN and JSe‚‚‚N should make good measures
for a covalent character of the Se‚‚‚N interaction. In seleno-
cyanate 5, however, electrostatic interaction between the
selenium and the nitrogen may also participate to the interaction
formation due to the strong electron-withdrawing ability of a
cyano group.
2-[(N-Cyclohexyl-N-methylamino)methyl]benzeneselenenyl Ac-
etate (4). 4 was quantitatively obtained from 2 by the treatment with
an equimolar amount of silver acetate in CH2Cl2 according to the
literature method.12 1H, 13C, and 77Se NMR spectra of 4 are shown in
the supporting information. 77Se NMR: δ 1157.1 (1158.3 ppm in the
literature12).
2-[(N-Cyclohexyl-N-methylamino)methyl]phenylselenenyl Cyan-
ate (5). 3, quantitatively prepared from 1 (218 mg, 0.5 mmol) by the
above method, was dissolved in dry THF, and cyanotrimethylsilane
(135 mL, 1.0 mmol) was added to the solution. After 10 min the
reaction mixture was concentrated under reduced pressure. 5 was
quantitatively obtained as a pale yellow oil from the residue by column
chromatography on silica gel (CH2Cl2 as eluent). Spectral data for 5:
1H NMR δ 1.2-1.4 (m, 5H), 1.8-1.9 (m, 5H), 2.11 (s, 3H), 2.56 (m,
1H), 3.68 (s, 2H), 7.2-7.3 (m, 3H), 7.8-7.9 (m, 1H); 13C NMR δ
25.4, 25.6, 27.3, 33.8, 58.7, 61.6, 108.9, 126.9, 127.0, 128.3, 128.4,
130.5, 137.6; 77Se NMR δ 362.4; mass spectrum m/e 308 (M+), 265
(base). Anal. Calcd for C15H20N2Se: C, 58.63; H, 6.56; N, 9.12.
Found: C, 58.40; H, 6.36; N, 8.99.
2-[(N-Cyclohexyl-N-methylamino)methyl]benzeneselenenyl Phen-
yl Sulfide (6). 6 was synthesized from 1 by the reaction with 2 molar
equiv of benzenethiol in CH2Cl2 according to the literature method.12
1H, 13C, and 77Se NMR spectra of 6 are shown in the supporting
information. 77Se NMR: δ 571.5 (571.9 ppm in the literature12). Anal.
Calcd for C20H25NSSe; C, 61.52; H, 6.45; N, 3.59. Found: C, 61.29;
H, 6.36; N, 3.72.
Molecular structures of 9-13 obtained by ab initio MO
calculations at HF/3-21G* and HF/LANL1DZ levels are very
consistent with the experimental results, though they are slightly
discrepant within the two basis sets. NBO deletion analysis
clearly indicated that the orbital interaction between a divalent
selenium moiety and a nitrogen atom largely contributes to the
Se‚‚‚N interaction formation as well as that the main component
of the interaction is the mixing of a low-lying σ*Se-X into a
nitrogen lone pair (nN). The electron delocalization from nN to
σ*Se-X may cause kinetic activation, or weakening, of the Se-X
bond whereas it provides overall thermodynamic stabilization
for the X-Se‚‚‚N system. These conclusions are in accordance
with our previous conclusion that the electronic structure of the
Cl-Se‚‚‚N nonbonded interaction of 14 is most likely compared
to the SN2 transition-state at a divalent selenium.7
Experimental Section
General Procedures. Commercially available organic and inorganic
reagents were used without further purification. Tetrahydrofuran (THF)
was dried over sodium wire and was distilled under nitrogen. Dichlo-
romethane (CH2Cl2) was dried over calcium hydride and was distilled
under nitrogen before use. Methanol (MeOH) was distilled under
2-[(N-Cyclohexyl-N-methylamino)methyl]phenyl Methyl Selenide
(7). 1 (169 mg, 0.3 mmol) was suspended in MeOH (10 mL) under
nitrogen atmosphere, and sodium borohydride (22 mg, 0.6 mmol) was
added to the solution. After 10 min, methyl iodide (0.19 mL, 3 mmol)
was added, and the mixture was stirred for 2 h. It was then added to
NaHCO3 solution and was extracted by ether. The crude product was
purified by column chromatography on silica gel (CH2Cl2-MeOH as
eluent). 7 was obtained as a colorless oil (0.122 g, 69%). Spectral
(33) Batt, L. Thermochemistry of Selenium and Tellurium Compounds.
In The Chemistry of Organic Selenium and Tellurium Compounds; Patai,
S., Rappoport, Z., Eds.; John Wiley & Sons: New York, 1986; Vol. 1.
(34) Mortimer, C. T.; Waterhouse, J. Thermochim. Acta 1988, 131, 91.

8084 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Iwaoka and Tomoda
data for 7: 1H NMR δ 1.3 (m, 5H), 1.8 (m, 5H), 2.11 (s, 3H), 2.23 (s,
3H), 2.58 (m, 1H), 3.67 (s, 2H), 7.1-7.3 (m, 4H); 13C NMR δ 6.2,
25.9, 26.2, 28.0, 35.2, 59.3, 62.0, 124.6, 127.2, 128.3, 129.0, 134.5,
139.8; 77Se NMR δ 180.8; mass spectrum m/e 297 (M+), 202 (base).
Anal. Calcd for C15H23NSe; C, 60.80; H, 7.82; N, 4.83. Found: C,
60.54; H, 7.64; N, 5.00.
15N-Labeled 1-7. 15N-labeled 1 was synthesized as follows.
Aniline-15N (>99 atom % 15N, ISOTEC Inc.) was hydrogenated over
ruthenium(III) hydroxide.35 The obtained cyclohexylamine-15N was
formylated by ethyl formate36 and then reduced by lithium aluminum
hydride to give cyclohexylmethylamine-15N.37 The obtained secondary
amine was coupled with 2,2′-diselenobis(benzyl chloride)6 in the
presence of NaHCO3 and excess potassium iodide to give 1-15N. 15N-
labeled 2-7 were obtained from 1-15N according to the methods
described above.
N-Cyclohexyl-N-methylbenzylamine (8). 8 was synthesized in
44% yield by the condensation between benzyl bromide and cyclo-
hexylmethylamine in the presence of excess triethylamine. Spectral
data for 8 were identical with the literature.17
Spectral Simulation. Variable-temperature 1H NMR spectra for 5
and 6 were simulated by using the DNMR5 program.18 The chemical
shifts, coupling constant, and line width were read from the spectrum
measured at low temperature (215 K for 6 and 196 K for 7). Then the
rate constant (k) at each temperature was determined so that the
simulated spectrum became almost identical with the experimental one.
analysis.30 The N inversion barrier (E3) and the C-N rotational barrier
(E2) of 8 were calculated at the HF/3-21G level. The most favorable
path from the global energy minimum conformation of 8 (Ph is
antiperiplanar to c-Hex) to its enantiomer was found to be the C-N
rotation to the rotational isomer (Ph is antiperiplanar to the nitrogen
lone pair) followed by the nitrogen inversion. All the transition states
were fully characterized by frequency calculations, and their energies
were corrected by the zero point energy. The values obtained were
6.7 kcal/mol for E2 and 3.8 kcal/mol for E3. The rotational barrier E1
was estimated by semiempirical PM3 calculations on 9-13 changing
the dihedral angle of the Ar-C bond (ω) from 0° to 120°. The
geometry was fully optimized in each calculation with the constraint
ω. The heat of formation was maximized at 90° and then minimized
at 100°. The energy difference between these two dihedral angles was
∼1.5 kcal/mol for 9-13. It was found that the conformation changes
drastically when the ω increases from 100° to 120°.
Geometries of 9-13 were fully optimized at the Hartree-Fock level
using a 3-21G* basis set (3-21G with d polarization functions for Se,
S, and Cl)39 and a LANL1DZ basis set (Los Alamos ECP+DZ basis
set32 for Se, S, and Cl and Dunning/Huzinaga valence double-ú basis
set for N, O, C, and H). NBO analysis, which is included in Gaussian
92, was then performed on the optimized conformations using the same
basis set. The electron delocalization energy (∆Edel) between Se-X
and N was calculated by using the NBO deletion method.30
Acknowledgment. We thank the Ministry of Education,
Science and Culture of Japan for financial support (Nos.
07854043, 06854029, 06640681, and 04854043).
Calculation Methods. Gaussian 9238 and Spartan 3.039 were used
as source programs for ab initio MO calculations and NBO deletion
Supporting Information Available: 1H and 13C NMR
spectra for 1-7 and a comparison of 90 MHz 1H NMR spectra
of benzylic protons for 6 at various temperatures with simulated
ones (15 pages). This material is contained in many libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
(35) Nishimura, S.; Kono, Y.; Otsuki, Y.; Fukaya, Y. Bull. Chem. Soc.
Jpn. 1971, 44, 240.
(36) Ugi, I.; Mery, R.; Lipinski, M.; Bodesheim, F.; Rosendahl, F.
Organic Synthesis; Wiley: New York, 1973; Collect. Vol. V, p 301.
(37) Blicke, F.-F.; Lu, C.-J. J. Am. Chem. Soc. 1952, 74, 3933.
(38) Frisch, M. J.; Trucks, G. W.; Head-Gordon, M.; Gill, P. M. W.;
Wong, M. W.; Foresman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb, M.
A.; Replogle, E. S.; Gomperts, R.; Andres, J. L.; Raghavachari, K.; Binkley,
J. S.; Gonzales, C.; Martin, R. L.; Fox, D. J.; Defrees, D. J.; Baker, J.;
Stewart, J. J. P.; Pople, J. A. Gaussian, Inc., Pittsburgh, PA, 1992.
(39) Hehre, W. J. et al. Wavefunction, Inc., Irvine, CA.
JA953358H