Orientational effect of aryl groups on77Se NMR chemical shifts: Experimental and theoretical investigations

FULL PAPER

DOI: 10.1002/chem.200500927

77

Orientational Effect of Aryl Groups on Se NMR Chemical Shifts:

Experimental and Theoretical Investigations

Waro Nakanishi,*^[a]Satoko Hayashi,^[a]Daisuke Shimizu,^[a]and Masahiko Hada^[b]

¯

Dedicated to Prof. Michinori Oki on occasion of his 77th birthday

Abstract: The orientational effect of p-

YC₆H₄(Ar) on d(Se) is elucidated for

ArSeR, based on experimental and

theoretical investigations. The effect is

calculated for ArSeR (R=H, Me, and

under a magnetic field. The main inter-

action in pd is the s(CSeX)–p(C₆H₄)–

Ph), assumingpl and pd, based on the

DFT-GIAO method. Observed charac-

ters are well explained by the total

s(Se). Paramagnetic terms (s^p(Se)) are

governed by (s^p(Se)_xx+s^p(Se)_yy), in

which the direction of n_p(Se) (con-

structed by 4p_z(Se)) is set to the z axis.

The main interaction in pl is the

G

ACHTREUNG

ꢀ

p_x(Y) type, in which Se X is nearly on

the x axis. The Y dependence in pd

mainly arises from admixtures of

4p_z(Se) in n_p(Se) with 4p_x(Se) and

ꢀ

examined in the cases in which Se C_R

in ArSeR is either in the Ar plane (pl)

or is perpendicular to the plane (pd).

9-(Arylselanyl)anthracenes (1) and 1-

(arylselanyl)anthraquionones (2) are

employed to establish the effect in pl

and pd, respectively. Large upfield

shifts are observed for Y=NMe₂,

OMe, and Me, and large downfield

shifts for Y=COOEt, CN, and NO₂in

1, relative to Y=H, as is expected.

Large upfield shifts are brought by Y=

NMe₂, OMe, Me, F, Cl, and Br, and

downfield shifts by Y=CN and NO₂in

2, relative to Y=H, with a negligible

shift by Y=COOEt. Absolute magnet-

ic shieldingtensors of Se ( s(Se)) are

4p_y(Se) in modified s*(CSeX), since

A

n_p(Se) is filled with electrons. It is dem-

onstrated that the effect of Y on s^p(Se)

in the pl conformation is the same re-

gardless of whether Y is an electron-

donor or electron-acceptor, whereas

for pd conformations the effect is

greater when Y is an electron donor, as

observed in 1 and 2, respectively. Con-

tributions of each molecular orbital

and each transition on s^p(Se) are eval-

uated, which enables us to recognize

and visualize the effect clearly.

n_p(Se)–p

pendence in pl occurs through admix-

tures of 4p_z(Se) in p(SeC₆H₄Y) and p*-

(SeC₆H₄Y), modified by the conjuga-

tion, with 4p_x(Se) and 4p_y(Se) in s-

(CSeX) and s*(CSeX) (X=H or C)

(C₆H₄)–p_z(Y) type. The Y de-

AHCTREUNG

A

ACHTREUNG

Keywords: ab initio calculations ·

aryl selenides · NMR spectroscopy ·

orientational effects · selenium ·

substituent effects

Introduction

⁷⁷Se NMR spectroscopy has been established as a powerful

tools in the study of the selenium chemistry.^{[1] 77}Se NMR

chemical shifts (d(Se)) are sharply sensitive to the structural

changes in selenium compounds. Therefore, they are widely

applied to analyze the chemical bonds around Se atoms and/

or to determine the structures.^[2–5]While some empirical

rules between structures and d(Se) have been proposed,^[2–5]

it is not so easy to predict d(Se) from the structures with

substantial accuracy. The measurements of d(Se) are neces-

sary to understand based on plain rules founded on the the-

oretical background. A charge-transfer mechanism has been

proposed to explain the downfield shift by the neighboring

oxygen atom^[6]and it is demonstrated that the mechanism

plays an important role in weak interactions.^[4]Other mecha-

[a] Prof. W. Nakanishi, Dr. S. Hayashi, D. Shimizu

Department of Material Science and Chemistry

Faculty of Systems Engineering, Wakayama University

930 Sakaedani, Wakayama 640–8510 (Japan)

Fax : (+81)73-457-8253

E-mail: nakanisi@sys.wakayama-u.ac.jp

[b] Prof. M. Hada

Department of Chemistry

Faculty of Graduate School of Science

Tokyo Metropolitan University, 1–1 Minami-osawa

Hachioji-shi, Tokyo 192–0397 (Japan)

Supportinginformation for this article is available on the WWW

under http://www.chemeurj.org/ or from the author.

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

3829

nisms must also be important when the structures are dis-

cussed based on the observed values. We have been much

interested in the orientational effect on d(Se) in p-

YC₆H₄SeR (ArSeR), together with the mechanism. We have

already pointed out the importance of the effect for the

better interpretation of d(Se) of ArSeR in a uniform

manner.^[5]A reliable guideline is necessary to determine the

structures of selenium compounds based on d(Se).

On the other hand, calculated absolute magnetic shielding

tensors (s) become reliable.^[7]Although the contribution of

relativistic terms has been pointed out for heavier atoms,^[8]

the perturbation would be small for the selenium nucleus.

Therefore, the calculated tensors are useful for Se nuclei

(s(Se)) in common organic selenium compounds. As shown

in Equation (1), the total absolute magnetic shielding tensor

(s^t) is decomposed into diamagnetic (s^d) and paramagnetic

(s^p) contributions.^[9]This decomposition includes small arbi-

trariness due to the coordinate origin dependence, though it

does not damage our chemical analyses and insights into

⁷⁷Se NMR spectroscopy. The s^pparameter contributes pre-

dominantly to s^tin the structural changes seen in the seleni-

um compounds. The magnetic shielding tensors consist of

three components, which are exemplified by s^pin Equa-

tion (2).

will discuss s^pwith an approximated image derived from

Equation (5). Since s^p_zz,Ncontains the L_z,Noperator, s^p_zz,N

arises from admixtures between atomic p_xand p_yorbitals of

N in various molecular orbitals. When a magnetic field is ap-

plied on a selenium compound, mixingof unoccupied mo-

lecular orbitals (y_j) into occupied molecular orbitals (y_i)

will occur. Such admixtures generate s^p_zz,N, if y_iand y_jcon-

tain p_xand p_yof N, for example. The parameters s^p_xx,Nand

s^p_yy,Nare similarly understood.

It is useful to supply a series of d(Se) of typical conform-

ers in ArSeR, although conformers may change successively

dependingon the electronic and/or steric properties of R

and Y.^[11–13]Planar (pl) and perpendicular (pd) conformers

ꢀ

will be discussed here, in which the Se C_Rbond in ArSeR is

in the Ar plane in pl and perpendicular to the plane in pd.

To clarify the relationship between the structures (conform-

ers) and d(Se), we tried to fix the conformers of all Ar

groups in ArSeR examined. 9-(Arylselanyl)anthracenes (1:

p-YC₆H₄SeAtc) and 1-(arylselanyl)anthraquinones (2: p-

YC₆H₄SeAtq) were chosen as the candidates for pl and pd,

respectively: Y in 1 and 2 are H (a), NMe₂(b), OMe (c),

Me (d), F (e), Cl (f), Br (g), COOEt (h), CN (i), and NO₂

(j). For the 9-anthryl (9-Atc) and 1-anthraquinonyl (1-Atq)

s^t¼ s^dþ s^p

ð1Þ

ð2Þ

s^p¼ ðs_x^p_xþ s^p_yyþ s_z^p_zÞ=3

The parameters s^pand s^dare exactly expressed by the

Ramseyꢁs Equation,^[10]and they are approximately calculat-

ed in the framework of the Hartree–Fock (HF) or DFT

theory. Since s^pis evaluated by the coupled Hartree–Fock

(CPHF) method, they can be decomposed into the contribu-

tion of the occupied orbitals or the orbital–orbital transi-

tions and are shown in Equation (3). The parameter s^dis ex-

pressed simply as the sum of contributions over the occu-

pied orbitals as shown in Equation (4).

groups in 1 and 2, respectively, as well as those proposed for

1-(arylselanyl)naphthalenes (3: p-YC₆H₄SeNap),^[3e,4,11]the

notation with A (perpendicular) B (parallel) and C (inter-

mediate) conformations is used. The structure of 1 is A for

X

occ

i

unocc

j

p

i!j

s^p¼

s^d¼

s

¼

^occs^p_i

i

ð3Þ

ð4Þ

9-Atc and pl for Ar, which is denoted by 1

A

is B for the 1-Atq and pd for Ar (2(B:pd)). The series of

AHCTREUNG

X

^occs^d_i

d(Se) in 1 and 2 are typical for pl and pd, respectively.

To understand the orientational effect based on the theo-

retical background, quantum chemical (QC) calculations

were performed on ArSeH (4), ArSeMe (5), and ArSePh

(6). The conformations were fixed to pl and pd in the calcu-

lations. The gauge-independent

i

Based on the second-order perturbation theory at the

level of the HF and single-excitation CI approximation, s_i^p_!j

on a resonance nucleus N is shown to be proportional to re-

ciprocal orbital energy gap (e_jꢀe_i), and expressed in Equa-

tion (5), in which y_kis the kth orbital function, L_z,Nis orbi-

tal angular momentum around the resonance nucleus, and

r_Nis the distance from the nucleus N.

atomic

orbital

(GIAO)

method^[14]was applied to evalu-

ate s(Se) at the DFT (B3LYP)

level. Although the term s_z^p_z,Se

is used in Equation (5), s^p(Se)_zz

X

occ

i

unocc

j

s_z^p_z,N¼ ꢀðm_oe²=2 m_e²Þ

ðe_jꢀe_iÞ^ꢀ1

will be used here on the analo-

g y tod(Se). Mechanisms of the

orientational effect were ex-

plored for pl and pd, based on the magnetic perturbation

theory. A utility program (NMRANAL-NH98G) was used

ð5Þ

Âfhy_ijL_zjy_jihy_jjL_z,Nr^ꢀ_N³jy_ii þ hy_ijL_z,Nr_N^ꢀ3jy_jihy_jjL_zjy_iig

Consequently, while s^pis evaluated accurately by the

CPHF method and can be calculated by Equation (3), we

3830

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

to carry out the decomposition of the magnetic shieldings,

based on the Gaussian 98.^[15]The program was also applied

to evaluate the contributions separately from each molecu-

lar orbital (y_i) and each y_i!y_jtransition, in which y_iand

y_jdenote occupied and unoccupied molecular orbitals, re-

spectively. These results enabled us to evaluate and visualize

the contributions.

selanyl (p-AnSe) group in 1c

Atc plane (A) with the Se(1) C(1) bond of 9-Atc beingin

the p-An plane (pl). The torsional angles of C(2)-C(1)-

Se(1)-C(15) and C(1)-Se(1)-C(15)-C(16) in 1c

95.6(4) and 174.8(4)8, respectively. The structure of 1c

is very close to that of pure 1

ACHTREUNG

ꢀ

ACHTREUNG

ꢀ

Se(1) C(15) bond of the PhSe group in 2a is in the 1-Atq

ꢀ

plane (B) with the Se(1) C(1) bond of 1-Atq beingperpen-

dicular to the Ph plane (pd). The torsional angles of C(14)-

C(1)-Se(1)-C(15) and C(1)-Se(1)-C(15)-C(16) are 175.1(3)

and 87.7(3)8, respectively. The structure of 2a is very close

Results and Discussion

Structures of 1c and 2a: Single crystals of 1c and 2a were

obtained through slow evaporation of the samples in di-

chloromethane–hexane or benzene–hexane solvent mixtures.

One of suitable crystals was subjected to X-ray crystallo-

graphic analysis for each compound. There are two types of

structure in the crystal of 1c (structure A and B). Only one

type of structure corresponds to 2a in the crystal. Figures 1

to that of pure 2

The structures of 1

Scheme 1, together with those of 3. Why does 1c have an

(A:pl) structure? It must be the steric requirement of 9-Atc.

ACHTREUNG

N

A

and 2 show structure A of 1c (1c(S-A)) and the structure of

U

2a, respectively.^[16]The crystallographic data are collected in

[17]

the SupportingInformation.

Scheme 1. Structures of 1 and 2, together with those of 3.

The structure of 1

is unlikely to be stable.^[18]The contribution of 1

be considered, although 3(A:pd) has not yet been observed.

Compound 1(A:pd) will be a transition state. QC calcula-

tions were performed on 1a; as a result 1a(A:pd) is predict-

ed to be a transition state between 1a(A:pl) and its dupli-

cate (Scheme 2).^[19]The activation energy of the internal ro-

A

N

AHCTREUNG

Figure 1. Structure of 1c

A

probability levels). Selected bond lengths (), angles (8), and torsional

ꢀ

angles (8): Se(1) C(1) 1.931(5), Se(1) C(15) 1.925(5), C(1)-Se(1)-C(15)

99.4(2), C(2)-C(1)-Se(1)-(15) 95.6(4), C(1)-Se(1)-C(15)-C(16) 174.8(4).

Scheme 2. Transition state nature of 1(A:pd).

G

ꢀ

tation around the Se C_Phbond is evaluated to be

13.7 kJmol^ꢀ1at 298 K. These results support that the global

minimum of 1 is 1

A

Figure 2. Structure of 2a (thermal ellipsoids are shown at 40% probabili-

ty levels). Selected bond lengths (), angles (8), and torsional angles (8):

substantially as 1(A:pl) in the solution.

U

ꢀ

Se(1) C(1) 1.921(3), Se(1) C(15) 1.935(3), C(1)-Se(1)-C(15) 99.1(1),

C(14)-C(1)-Se(1)-(15) 175.1(3), C(1)-Se(1)-C(15)-C(16) 87.7(3).

ꢀ

s*

N

operates in 2

G

2

ACHTREUNG

The planarity of the 9-Atc, 1-Atq, and Ar planes in 1c

A

ACHTREUNG

ꢀ

and 2a is very good. The Se(1) C(15) bond of the p-anisyl-

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3831

W. Nakanishi et al.

y ¼ ax þ b ðr : correlation coefficientÞ

ð6Þ

teraction, together with the n_p(Se)–p(Atq) conjugation. The

G

structure of all members of compounds of the 2 series is pre-

dicted to be (B:pd).^[24]

The temperature dependence of 1 must be the reflection

of the shallow energy surface in 1(A:pl), resultingin the ex-

tended population to larger torsional angles of f(C₉SeC_iC_o)

AHCTREUNG

⁷⁷Se NMR chemical shifts of 1 and 2: Table 1 shows the

d(Se) values for 1 and 2^[25]measured in [D]chloroform

(0.050m)^[26]at 213, 297, and 333 K. The d(Se) values of 1a

and 2a are given relative to MeSeMe and those of 1b–j and

AHCTREUNG

around the energy minimum of f=08 at higher tempera-

tures. The energy surfaces of Y with electron-donating

groups must be shallower those with electron-accepting

groups. The p–p conjugation of

the n_p(Se)–p

fects on the stability of 1

ACHTREUNG

Table 1. Observed d(Se)_SCSof 1 and 2 and calculated s^t_rel(Se)_SCSfor 4–6 in pl and pd conformations.^[a,b]

ACHTREUNG

T

[K]

NMe₂OMe

Me

(d)

H

(a)

F

(e)

Cl

(f)

Br

(g)

CO₂R^[c]CN

NO₂

(j)

in addition to the steric require-

(b)

(c)

(h)^[c]

(i)

ment of 9-Atc in 1:^[20]Y accept-

1

2

213

297

333

213

297

333

ꢀ22.7

ꢀ21.0

ꢀ21.3

ꢀ20.6

ꢀ19.6

ꢀ19.5

ꢀ36.4

ꢀ23.9

ꢀ20.5

ꢀ35.9

ꢀ34.9

ꢀ34.2

ꢀ12.7

ꢀ12.2

ꢀ12.7

ꢀ15.5

ꢀ15.0

ꢀ18.0

ꢀ8.2

ꢀ6.3 0.0

N

ꢀ3.3

1.9

1.5

1.0

ꢀ7.1

ꢀ7.2

1.7

4.7

1.9

ꢀ9.1

ꢀ11.8

ꢀ13.3

2.4

1.6

1.2

17.4

16.2

15.2

0.1

27.7

26.2

24.8

8.5

8.2

7.9

29.8

29.7

20.2

16.8

13.4

7.0

32.7

30.4

29.0

2.7

2.5

2.2

33.7

43.8

28.6

10.0

6.6

ors stabilize 1

tively than Y donors. The tem-

perature dependence in

(A:pl) more effec-

ꢀ6.6 0.0

ꢀ6.8 0.0

ꢀ9.2 0.0

ꢀ9.0 0.0

ꢀ9.1 0.0

ꢀ8.2 0.0

ꢀ8.0 0.0

ꢀ3.7 0.0

ꢀ3.6

ꢀ3.9

ꢀ6.4

0.0

d(Se)_SCSfor Y=OMe and

NMe₂seems complex, at a first

glance. However, if we examine

d(Se) of 1b and 1c, relative to

1a, we find them quite simple

and monotonical. The charac-

teristic temperature depend-

ence of d(Se) in 1a may be re-

sponsible for the phenomenon.

Characteristic points of 1 are

summarized as follows:

ꢀ6.7 ꢀ0.3

4 (pl)

5 (pl)

6 (pl)

4 (pd)

5 (pd)

6 (pd)

G

ꢀ1.6

2.1

1.1

ꢀ1.8

7.2

14.3

24.6

13.1

1.0

3.0

ꢀ3.4

ꢀ9.0

2.3

ꢀ23.0

ꢀ21.2

ꢀ25.8

ꢀ15.6

ꢀ16.7

ꢀ14.6

0.0

G

ꢀ8.7

ꢀ12.6

0.5

[a] d(Se)_SCSare given for

1 and 2, together with d(Se) for 1a and 2a in parenthesis, measured in

[D]chloroform. [b] s^t_rel(Se)_SCSare given for 4–6, together with s_r^t_el(Se) for 4a–6a in parenthesis, calculated ac-

cordingto Equation (7) employing s^t(Se:MeSeMe)=1650.4 ppm. [c] R=Et for 1 and 2 and R=Me for 4–6.

1) Large upfield shifts (ꢀ23 to ꢀ6 ppm) are observed for

Y=NMe₂, OMe, and Me and large downfield shifts (17–

33 ppm) for Y=COOEt, CN and NO₂, relative to Y=H.

2) A moderate upfield shift (ꢀ3 ppm) is observed for Y=F

and small downfield shifts (2 ppm) for Y=Cl and Br.

2b–j are given relative to 1a and 2a, respectively (d(Se)_SCS).

The values of d(Se) of MeSeMe in [D]chloroform (10%

v/v) shifts downfield by 5.6 ppm, if the frequency of the

spectrometer is taken as the standard, when the temperature

changes from 213 K to 333 K under the conditions.^[27]The

d(Se) values of 1a, 1b, and 1j

move downfield by 5.3, 6.7, and

Table 2. Correlations of d(Se) for 1 and 2 and s(Se) for 4–6.^[a]

correlation

(Se:1)_SCS,297Kvs. d

(Se:1)_SCS,333Kvs. d

(Se:2)_SCS,297Kvs. d

(Se:2)_SCS,333Kvs. d

(Se:1)_SCS,213Kvs. s^t

_relA

(Se:1)_SCS,213Kvs. s^t

_relA(Se:6(pl))_SCS

(Se:2)_SCS,213Kvs. s^t

_relA(Se:4(pd))_SCS

(Se:2)_SCS,213Kvs. s^t

_relA(Se:5(pd))_SCS

(Se:2)_SCS,213Kvs. s^t

_relA(Se:6(pd))_SCS

1.6 ppm, respectively, relative

to MeSeMe, as the temperature

changes from 213 K to 333 K.

The results show that the tem-

perature dependence of d(Se)

in 1a and 1b is roughly twice as

large as that of MeSeMe,

whereas the dependence in 1j

is comparable to that of

MeSeMe.

To examine the temperature

dependence in 1, the d(Se)

values of 1 at 293 K and 333 K

were plotted versus those at

213 K; excellent correlations

(r>0.999) were observed, see

Table 2 (entries 1 and 2), in

which the correlation constants

(a) and the correlation coeffi-

cients (r) are given by Equa-

tion (6).

a

b

r

n

1

2

3

4

5

6

7

8

d

G

A

0.940

0.916

0.957

0.946

0.823

0.845

1.218

0.562

0.599

0.691

0.325

0.357

0.335

0.307

0.374

0.345

0.309

0.335

0.23

ꢀ0.3

ꢀ0.8

ꢀ0.1

ꢀ0.3

2.6

ꢀ2.1

ꢀ0.4

ꢀ1.5

ꢀ0.5

1.9

1.000

0.986

0.990

0.991

0.990

0.988

0.990

1.000

0.985

1.000

0.998

0.994

0.999

0.85

10

d

G

d

E

d

E

d

H

N

d

H

G

d

G

R

d

E

N

9

d

R

N

10

11

12

13

14

15

16

17

18

19

20

21

22

d

E

s^p(Se) vs. (s^p(Se)_xx+s^p(Se)_yy) in 4(pl)

s^p(Se) vs. (s^p(Se)_xx+s^p(Se)_yy) in 4(pd)

s^p(Se) vs. (s^p(Se)_xx+s^p(Se)_yy) in 5(pl)

s^p(Se) vs. (s^p(Se)_xx+s^p(Se)_yy) in 5(pd)

s^p(Se) vs. (s^p(Se)_xx+s^p(Se)_yy) in 6(pl)

s^p(Se) vs. (s^p(Se)_xx+s^p(Se)_yy) in 6(pd)

s^p(Se)_yyvs. s^p(Se)_xxin 4 (pd)

ꢀ583.7

ꢀ500.6

ꢀ487.5

ꢀ553.4

ꢀ442.0

ꢀ520.2

ꢀ689.0

ꢀ598.8

ꢀ455

22

16^[b]

22

16^[b]

14

11^[b]

13

10^[b]

16^[b]

14

s^p(Se)_yyvs. s^p(Se)_xxin 5 (pd)

ꢀ0.33

ꢀ0.36

ꢀ0.65

ꢀ1722

ꢀ2443

ꢀ2721

0.93

s^p(Se)_yyvs. s^p(Se)_xxin 6 (pl)

0.92

13

s^p(Se)_yyvs. s^p(Se)_xxin 6 (pd)

0.95

13

[a] The constants a, b, r are defined in Equation (6) in the text. [b] For non-ionic species.

3832

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

These charactistics must be the result of the 1

(A:pl) struc-

1) Large upfield shifts (ꢀ21 to ꢀ6 ppm) are observed for

Y=NMe₂, OMe, Me, F, Cl, and Br, relative to Y=H.

2) Downfield shifts (3–9 ppm) are observed for Y=CN and

NO₂, for which the magnitude with Y=CN is larger than

that of NO₂, together with a negligible shift by Y=

COOEt, relative to Y=H.

ture, in which n_p(Se) is parallel to the p(C₆H₄Y-p). The set

G

of d(Se) of 1 can be used as a standard of pl.

In the case of 2, the temperature dependence of d(Se) is

very small. The magnitude in compound 2a is 1.1 ppm, rela-

tive to MeSeMe, under the temperature range from 213 K

to 333 K. The plots of d(Se) of 2 at 293 K and 333 K versus

those at 213 K also give excellent correlations (r>0.999).

Table 2 collects the correlations (entries 3 and 4). The re-

The characteristics must be the reflection of the 2

A

structure, in which n_p(Se) is perpendicular to p(C₆H₄Y-p).

AHCTREUNG

sults show that 2

A

The set of d(Se) of 2 can be used as a typical standard of

conformers are negligible in the solution for all Y examined.

The character of d(Se) in 2 is very different from that of

1. Characteristic points of 2 are as follows:

pd.

While d(Se)_SCSof 1 is in a range of ꢀ23<d(Se)_SCS

<

33 ppm, that of 2 is ꢀ21<d(Se)_SCS<9 ppm. The Y donor

and acceptor groups have an effect on d(Se)_SCSof 1, whereas

only Y donor groups have an effect on the d(Se)_SCSof 2.

The values of d(Se)_SCSof 2 are plotted versus those of 1 in

Figure 3a. Indeed, it emphasizes the difference in characters

between 1 and 2, but most of d(Se)_SCSvalues of 2 seem to

correlate well with those of 1, as shown by a dotted line (a=

0.55). Two points correspondingto Y =H and NO₂deviate

from the line. Namely, d(Se) of 2 with Y°H are more up-

field than those expected from d(Se) of 1a and 2a, especial-

ly for 2j.

Why is such peculiar behavior observed in 1 and 2? Is it

caused by the orientational effect of the aryl group? QC cal-

culations were performed on 4–6, assumingpl and pd for

each, to elucidate the mechanism of the effect.

Observed d(Se) of 1 and 2 versus calculated s^t(Se) of 4–6:

Scheme 3 shows axes and some orbitals of 4–6, together

with SeH₂. While the x axis in SeH₂is in the bisected direc-

ꢀ

tion of aHSeH, the Se H and Se C bonds of MeSeH are

almost on the x and y axes, respectively, although not

shown. Axes of 4–6 are close to those in MeSeH in most

cases. Since aCSeX (X=H or C) in 4–6 are close to 95, 98,

and 1018, respectively, the bonds deviate inevitably from the

axes to some extent. Axes are similar to those of SeH₂for

Scheme 3. Axes and some orbitals of 4–6, together with those of SeH₂.

[a] Axes with Y=Se^ꢀ, Br, and COOMe are close to those for SeH₂.

[b] Axes with Y=Me, CN, and NH₃are close to those for SeH₂.

Figure 3. Plots of a) d

A

G

U

+

versus s^t_rel

A

G

C

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3833

W. Nakanishi et al.

4(pl) with Y=Se^ꢀ, Br, and COOMe and 5(pl) with Y=Me,

nate from the planar structure and those in 2 from the char-

+

CN, and NH₃(Scheme 3).^[28]

acteristic structure in which the Se C_Atqbond in p-

ꢀ

Structures of 4–6 in pl and pd and SeH₂were optimized

YC₆H₄SeAtq is perpendicular to the p-YC₆H₄plane. The re-

sults are satisfactory, if we consider the limited experimental

conditions and/or the different selenides employed in exper-

imental study and in the calculations.

by employingthe 6–311 +G

A

311+G(3d,2p) basis sets for other nuclei in the Gaussian 03

ACHTREUNG

program.^[29]Calculations are performed at the density func-

tional theory (DFT) level of the Becke three-parameter

hybrid functionals with the Lee–Yang–Parr correlation func-

tional (B3LYP). Absolute magnetic shielding tensors of Se

(s(Se)) were calculated by the same method based on the

DFT-GIAO method,^[14]applyingit to the optimized struc-

tures. Tables 3–5 give the s^t(Se), s^d(Se), s^p(Se) parameters

and the components, s^p(Se)_xx,

How does such an orientational effect arise from the

structures? How does the electronic property of Y affect the

d(Se) values of 1 and 2? The s^p(Se) values of 4–6 are ana-

lyzed next.

s^p(Se)_yy, and s^p(Se)_zz, for 4–6

[a]

bearingvarious substituents Y

Table 3. Calculated absolute shieldingtensors ( s(Se)) of 4, containingvarious Y.

in pl and pd,^[30]respectively.

Y

s^d(Se)

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

s^p(Se)

s^t(Se)

1563.4

It is instructive if s^t(Se) of 4–

6 in pl and pd can be compared

directly with the observed d(Se)

of 1 and 2. Relative shielding

tensors of selenium compounds

S, that is, 4–6 in pl and pd, s^t_rel-

4(pl)

H

2999.5

3001.4

3004.1

3007.3

3000.9

3006.4

3001.0

3004.7

3002.4

3001.4

3003.8

3008.7

3005.3

3007.4

3010.0

3001.5

3002.1

3005.3

3004.9

3006.6

3003.7

3000.5

ꢀ1571.7

ꢀ1537.7

ꢀ1560.7

ꢀ1745.3

ꢀ1741.2

ꢀ1676.1

ꢀ1777.1

ꢀ1823.5

ꢀ1760.2

ꢀ1800.4

ꢀ1777.8

ꢀ1883.4

ꢀ1849.6

ꢀ1812.7

ꢀ1469.6

ꢀ1846.4

ꢀ1829.1

ꢀ1889.0

ꢀ1836.6

ꢀ2563.0

ꢀ2579.3

ꢀ2680.8

ꢀ1042.3

ꢀ803.3

ꢀ843.6

ꢀ648.1

ꢀ828.0

ꢀ862.4

ꢀ824.9

ꢀ757.0

ꢀ848.1

ꢀ833.2

ꢀ868.7

ꢀ745.3

ꢀ786.8

ꢀ882.1

ꢀ1197.4

ꢀ866.9

ꢀ889.8

ꢀ819.9

ꢀ905.8

ꢀ976.0

ꢀ1292.6

ꢀ1600.0

ꢀ1694.2

ꢀ1674.3

ꢀ1680.5

ꢀ1704.1

ꢀ1671.4

ꢀ1681.4

ꢀ1673.0

ꢀ1689.5

ꢀ1684.2

ꢀ1675.7

ꢀ1680.0

ꢀ1701.6

ꢀ1696.4

ꢀ1699.3

ꢀ1715.7

ꢀ1691.1

ꢀ1686.5

ꢀ1697.2

ꢀ1683.2

ꢀ1659.3

ꢀ1650.6

ꢀ1647.7

ꢀ1436.1

ꢀ1338.4

ꢀ1361.6

ꢀ1365.8

ꢀ1413.5

ꢀ1406.7

ꢀ1425.0

ꢀ1423.3

ꢀ1430.8

ꢀ1436.4

ꢀ1442.2

ꢀ1443.4

ꢀ1444.3

ꢀ1464.7

ꢀ1460.9

ꢀ1468.2

ꢀ1468.5

ꢀ1468.7

ꢀ1475.2

ꢀ1732.8

ꢀ1840.9

ꢀ1976.2

O^ꢀ

1662.9

1642.4

1641.4

1587.3

1599.8

1576.1

1581.4

1571.6

1565.0

1561.7

1565.2

1561.0

1542.7

1549.1

1533.3

1533.6

1536.6

1529.7

1273.9

1162.9

1024.4

S^ꢀ

Se^ꢀ

NH₂

NMe₂

OH

OMe

Me

F

Cl

Br

CHCH₂

COOH

COOMe

BH₂

CN

A

ingto Equation (7), using

s^t(Se:MeSeMe) of 1650.4 ppm.

The s^t_rel

ACHTREUNG

AHCTREUNG

ACHTREUNG

CHO

NO₂

Se⁺

s^t_relðSe : SÞ ¼ ꢀfs^tðSe : SÞ

ꢀs^tðSe : MeSeMeÞg

S⁺

ð7Þ

O⁺

4(pd)

H

3001.9

3002.6

3002.4

2999.4

3000.0

3004.1

3001.4

3005.4

3002.2

3000.8

3000.5

3003.6

3006.1

3004.2

3001.5

2999.9

3003.3

3000.7

2998.7

3000.2

2999.2

ꢀ1870.9

ꢀ1547.6

ꢀ1590.4

ꢀ1603.2

ꢀ1775.8

ꢀ1782.2

ꢀ1793.9

ꢀ1805.2

ꢀ1821.7

ꢀ1829.8

ꢀ1834.5

ꢀ1835.5

ꢀ1843.1

ꢀ1880.4

ꢀ1872.6

ꢀ1889.4

ꢀ1901.0

ꢀ1895.1

ꢀ1877.7

ꢀ1482.5

ꢀ2941.4

ꢀ2821.1

ꢀ869.9

ꢀ875.7

ꢀ868.2

ꢀ866.2

ꢀ861.7

ꢀ842.4

ꢀ863.7

ꢀ871.3

ꢀ871.0

ꢀ866.2

ꢀ870.2

ꢀ870.5

ꢀ876.7

ꢀ880.5

ꢀ879.2

ꢀ884.3

ꢀ881.6

ꢀ884.4

ꢀ45333.3

295.6

ꢀ1437.6

ꢀ1541.6

ꢀ1507.8

ꢀ1495.2

ꢀ1451.5

ꢀ1452.8

ꢀ1449.1

ꢀ1443.6

ꢀ1439.8

ꢀ1443.7

ꢀ1442.8

ꢀ1442.1

ꢀ1438.7

ꢀ1440.2

ꢀ1436.5

ꢀ1439.2

ꢀ1440.1

ꢀ1438.1

ꢀ1442.8

ꢀ1649.8

ꢀ1657.9

ꢀ1713.4

ꢀ1392.8

ꢀ1321.7

ꢀ1322.1

ꢀ1321.5

ꢀ1363.0

ꢀ1359.1

ꢀ1368.9

ꢀ1373.4

ꢀ1377.5

ꢀ1379.9

ꢀ1382.5

ꢀ1382.7

ꢀ1386.2

ꢀ1400.4

ꢀ1396.1

ꢀ1404.3

ꢀ1407.6

ꢀ1405.9

ꢀ1401.6

ꢀ16155.2

ꢀ1434.6

ꢀ2827.2

1609.1

1680.9

1680.3

1677.9

1637.0

1645.0

1632.5

1632.1

1624.7

1620.9

1618.2

1617.8

1617.4

1605.7

1608.1

1597.2

1592.3

1597.4

1599.1

ꢀ13156.5

1565.7

172.0

O^ꢀ

The characteristics observed

S^ꢀ

for 1 and 2 are well explained

Se^ꢀ

by s^t_rel(Se)_SCS

.

The d(Se)_SCS

NH₂

NMe₂

OH

OMe

Me

F

Cl

Br

CHCH₂

COOH

COOMe

BH₂

CN

values of 1 and 2 were plotted

versus s^t_rel(Se)_SCSof n(pl) and

n(pd) (n=4–6). Good correla-

tions were obtained and the re-

sults are given in Table 2 (en-

tries 5–10). The r value for 1

becomes larger in an order of

4(pl)<5(pl)ꢁ6(pl) with

a of

4(pl)<5(pl)<6(pl). In the case

of 2, r becomes larger in an

order of 5(pd)<4(pd)ꢁ6(pd)

and a of 4(pd)<5(pd)<6(pd).

Figures 3b and c show the plots

for 1 versus 6(pl) and 2 versus

6(pd), respectively. The results

demonstrate that the characters

CHO

NO₂

Se⁺

S⁺

O⁺

ꢀ3947.0

[a] Structures were optimized with the 6–311+G

other nuclei at the DFT (B3LYP) level, assumingpl and pd for each of Y. s(Se) were calculated based on the

A

N

of d(Se)_SCSobserved in 1 origi- DFT-GIAO method with the same basis sets.^[29]

3834

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

[a]

Table 4. Calculated absolute shieldingtensors ( s(Se)) of 5, containingvarious Y.

of each y_i!y_jtransition for 4a,

together with the energy differ-

ence of each transition and

Y

s^d(Se)

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

s^p(Se)

s^t(Se)

5(pl)

H

3006.5

3005.4

3008.7

3007.7

3007.4

3008.0

3006.2

3006.7

3008.1

3009.6

3006.6

3006.9

3007.0

3006.9

ꢀ1893.4

ꢀ1685.1

ꢀ1544.2

ꢀ1645.4

ꢀ1741.5

ꢀ1815.2

ꢀ1911.7

ꢀ1639.8

ꢀ1768.8

ꢀ1840.5

ꢀ1601.6

ꢀ1892.9

ꢀ1800.0

ꢀ1771.0

ꢀ999.0

ꢀ1056.5

ꢀ1236.7

ꢀ1194.5

ꢀ1136.8

ꢀ1064.7

ꢀ990.8

ꢀ1684.9

ꢀ1640.3

ꢀ1676.3

ꢀ1669.5

ꢀ1677.1

ꢀ1678.0

ꢀ1680.6

ꢀ1682.4

ꢀ1679.0

ꢀ1687.1

ꢀ1688.1

ꢀ1683.2

ꢀ1690.0

ꢀ1709.6

ꢀ1525.8

ꢀ1460.7

ꢀ1485.7

ꢀ1503.1

ꢀ1518.4

ꢀ1519.3

ꢀ1527.7

ꢀ1530.7

ꢀ1534.7

ꢀ1553.5

ꢀ1555.6

ꢀ1562.1

ꢀ1570.1

ꢀ1627.2

1480.7 main character of Se in the mo-

O^ꢀ

1544.7

1523.0

1504.6

1488.9

lecular orbital y_j.

Contributions of the p_z(Se)

COO^ꢀ

NMe₂

OMe

Me

F

Cl

Br

COOMe

CN

character in occupied y_iorbi-

tals were evaluated.^[33]The

1488.7

1478.6

1476.0

1473.5

1456.1

1451.0

method is explained exempli-

fied by SeH₂. The molecular or-

bitals y₁₈, y₁₃, y₉, and y₅of

SeH₂have B1 symmetry. While

ꢀ1269.8

ꢀ1156.2

ꢀ1132.8

ꢀ1377.0

ꢀ1110.2

ꢀ1220.1

ꢀ1401.0

BH₂

NO₂

NH₃

1444.8 y₁₈, y₉, and y₅are mainly con-

1436.9

1379.7

structed from 4p_z(Se), 3p_z(Se),

and 2p_z(Se), respectively, the

character of y₁₃is 3d_xz(Se).^[34]

The four orbitals are assumed

to have the p_z(Se) character

here and are denoted collec-

tively as “y_i”. The s^p(Se)_xx,

s^p(Se)_yy, s^p(Se)_zz, and s^p(Se) pa-

+

5(pd)

H

2998.0

3000.3

3002.2

3003.5

3004.1

2999.8

2998.1

2999.3

3001.0

3006.4

2998.0

2998.2

2999.5

2996.9

ꢀ1956.8

ꢀ1606.3

ꢀ1766.1

ꢀ1889.2

ꢀ1938.6

ꢀ1908.0

ꢀ1916.6

ꢀ1925.8

ꢀ1930.0

ꢀ2017.8

ꢀ1995.5

ꢀ1979.9

ꢀ1977.4

ꢀ2131.1

ꢀ1086.4

ꢀ1203.9

ꢀ1152.4

ꢀ1062.0

ꢀ1059.6

ꢀ1090.1

ꢀ1077.7

ꢀ1078.6

ꢀ1077.7

ꢀ1057.8

ꢀ1076.6

ꢀ1089.7

ꢀ1075.9

ꢀ1028.0

ꢀ1656.9

ꢀ1716.8

ꢀ1640.9

ꢀ1660.9

ꢀ1656.6

ꢀ1657.2

ꢀ1663.9

ꢀ1664.3

ꢀ1663.5

ꢀ1658.7

ꢀ1668.2

ꢀ1662.8

ꢀ1671.0

ꢀ1731.9

ꢀ1566.7

ꢀ1509.0

ꢀ1519.8

ꢀ1537.3

ꢀ1551.6

ꢀ1551.8

ꢀ1552.8

ꢀ1556.2

ꢀ1557.1

ꢀ1578.1

ꢀ1580.1

ꢀ1577.5

ꢀ1574.7

ꢀ1630.4

1431.3

1491.3

1482.4

1466.2

1452.5

1448.0

1445.4

O^ꢀ

COO^ꢀ

NMe₂

OMe

Me

F

Cl

Br

COOMe

CN

1443.1 rameters are summed over

1443.9

1428.3

1417.9

1420.7

1424.7

1366.5

“y_i”. They are represented by

s^p(Se)_xx(zi), s^p(Se)_yy(zi), s^p(Se)_zz(zi)

and s^p(Se)_(zi)

respectively, in

,

BH₂

NO₂

NH₃

which the suffix (zi) means the

contribution of the p_z(Se) char-

acter in occupied “y_i”. The re-

sults are given in Table 10,

alongwith those for 4a–6a.^[35,36]

The parameters s^p(Se)_xx(zi)and

s^p(Se)_yy(zi)will be good approxi-

+

[a] Structures were optimized with the 6–311+G

A

G

other nuclei at the DFT (B3LYP) level, assumingpl and pd for each of Y. s(Se) were calculated based on the

DFT-GIAO method with the same basis sets.^[29]

Evaluation of magnetic tensors based on molecular orbitals:

Contributions of each molecular orbital (y_i) and each y_i!

y_jtransition on s^p(Se) and the components, s^p(Se)_xx,

s^p(Se)_yy, and s^p(Se)_zz, are evaluated for 4a, 5a, 6a (Y=H),

and SeH₂, by usinga utility prorgam (NMRANAL-

NH98G). The contributions were calculated in a similar

manner as the tensors given in Tables 3–5, by the same

method. The structures optimized with the Gaussian 03 pro-

gram were employed for the evaluation. The directions of

p_x(Se), p_y(Se), and p_z(Se) are shown in Scheme 3. The pa-

rameteres s^p(Se)_xx, s^p(Se)_yy, and s^p(Se)_zzarise from [p_y(Se);

p_z(Se)], [p_z(Se); p_x(Se)], and [p_x(Se); p_y(Se)], respectively,

in which [A; B] shows admixtures between atomic orbitals

A and B, under a magnetic field [compare with Eq. (5)].

Table 6 contains the values of s^p(Se), the components,

and the contribution of each y_i, together with orbital ener-

gies and main character of Se in SeH₂.^[31]Table 7 shows the

contribution of each y_i!y_jtransition on the s^p(Se)_xx,

s^p(Se)_yy, and s^p(Se)_zzparameters, together with the energy

differences of the transition and main character of Se in the

molecular orbital y_jin SeH₂. Table 8 exhibits s^p(Se), the

components, and contributions of each y_i, together with the

energies and main character of Se in 4a–6a,^[32]for which

those of 4p(Se) and 4s(Se) are mainly shown for 4a and

those of 4p(Se) for 5a and 6a. Table 9 lists the contributions

mations of the values from [p_z(Se) in y_i; p_y(Se) in y_j] and

[p_z(Se) in y_i; p_x(Se) in y_j], respectively. The s^p(Se)_zz(zi)

values are essentially zero, since they are related to [p_z(Se)

in “y_i”; p_z(Se) in y_j]. Those from [p_x(Se) in “y_i”; p_y(Se) in

y_j]+[p_y(Se) in “y_i”; p_x(Se) in y_j] are also essentially zero,

since the p_x(Se) and p_y(Se) characters are negligible in “y_i”.

Equation (8) defines s^p(Se)_{AA(xi’+yi’)}(A=x, y, z, and zero),

in which i’ comes from “y_i’” that are y_iother than “y_i”.^[35]

They correspond to the contributions of p_xand p_yin y_ito

s^p(Se)_AA

.

s^pðSeÞ_AAðxi₀_þyi₀_Þ¼ s^pðSeÞ_AAꢀs^pðSeÞ_AAðziÞ

ð8Þ

The values are also shown in Table 10. The s^p(Se)_{xx(xi’+yi’)}

values result from the admixtures [p_x(Se) in y_i; p_x(Se) in

y_j]+[p_y(Se) in y_i; p_z(Se) in y_j] and s^p(Se)_{yy(xi’+yi’)}to [p_x(Se)

in y_i; p_z(Se) in y_j]+[p_y(Se) in y_i; p_y(Se) in y_j]. The admix-

tures [p_x(Se) in y_i; p_x(Se) in y_j] and [p_y(Se) in y_i; p_y(Se) in

y_j] are denoted by s^p(Se)_xx(xi’)and s^p(Se)_yy(yi’), respectively,

which are essentially zero. Therefore, s^p(Se)_{xx(xi’+yi’)}and

s^p(Se)_{yy(xi’+yi’)}would reduce to s^p(Se)_xx(yi’)and s^p(Se)_yy(xi’), re-

spectively. The s^p(Se)_{zz(xi’+yi’)}parameter is related to [p_x(Se)

in y_i; p_y(Se) in y_j]+[p_y(Se) in y_i; p_x(Se) in y_j] admixtures.

After evaluation of s^p(Se) and the components under the

conditions, the next step is to elucidate the orientational

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3835

W. Nakanishi et al.

[a]

Analysis of s^p(Se) in SeH₂:

What admixtures cause the

shifts in SeH₂? Contributions of

Table 5. Calculated absolute shieldingtensors ( s(Se)) of 6, containingvarious Y.

Y

s^d(Se)

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

s^p(Se)

s^t(Se)

6(pl)

H

2995.1

2999.0

2999.3

2997.7

2995.5

2995.6

2994.5

2994.1

2996.5

2997.2

2994.8

2994.4

2992.1

ꢀ1527.4

ꢀ1282.9

ꢀ1379.1

ꢀ1462.8

ꢀ1504.1

ꢀ1517.7

ꢀ1544.4

ꢀ1550.2

ꢀ1553.1

ꢀ1574.5

ꢀ1605.6

ꢀ1630.8

ꢀ1786.8

ꢀ1887.5

ꢀ2021.7

ꢀ1953.6

ꢀ1902.3

ꢀ1887.4

ꢀ1888.8

ꢀ1879.4

ꢀ1873.8

ꢀ1871.1

ꢀ1871.7

ꢀ1869.2

ꢀ1867.8

ꢀ1835.6

ꢀ1815.6

ꢀ1829.9

ꢀ1840.6

ꢀ1811.6

ꢀ1813.2

ꢀ1814.2

ꢀ1808.1

ꢀ1809.2

ꢀ1817.4

ꢀ1830.0

ꢀ1815.6

ꢀ1815.7

ꢀ1804.9

ꢀ1743.5

ꢀ1711.5

ꢀ1724.5

ꢀ1725.5

ꢀ1734.9

ꢀ1740.3

ꢀ1743.9

ꢀ1744.4

ꢀ1747.2

ꢀ1758.7

ꢀ1763.5

ꢀ1771.4

ꢀ1809.1

1251.6 each y_iand each y_i!y_jtransi-

O^ꢀ

1287.5

1274.9

1272.1

1260.6

1255.3

1250.5

1249.7

1249.3

1238.5

1231.4

tion will answer the question.

COO^ꢀ

NMe₂

OMe

Me

F

Cl

Br

COOMe

CN

The y₁₈orbital (HOMO:

n_p(Se)) of 4p_z(Se) contributes

ꢀ641 ppm to s^p(Se) (s^p(Se)_xx=

ꢀ1335 ppm

and

s^p(Se)_yy=

ꢀ587 ppm), which is 69% of

the total s^p(Se) (Table 6). The

y₁₈!y₂₀(s*

C

NO₂

NH₃

+

s^p(Se)_xx(ꢀ1250 ppm) and y₁₈!

1183.0

6(pd)

H

y₁₉(s*

C

2995.1

2997.2

2999.3

2998.3

3002.2

2996.4

2994.8

2995.1

2997.2

3003.2

2998.5

2995.5

2992.2

ꢀ1887.5

ꢀ1592.1

ꢀ1733.0

ꢀ1787.3

ꢀ2044.1

ꢀ1843.1

ꢀ1851.2

ꢀ1859.5

ꢀ1871.4

ꢀ2085.4

ꢀ2132.1

ꢀ1914.6

ꢀ2043.9

ꢀ1527.4

ꢀ1641.6

ꢀ1601.5

ꢀ1531.6

ꢀ1313.9

ꢀ1532.7

ꢀ1517.7

ꢀ1519.1

ꢀ1514.8

ꢀ1341.9

ꢀ1310.6

ꢀ1502.6

ꢀ1437.0

ꢀ1815.6

ꢀ1882.9

ꢀ1815.2

ꢀ1818.6

ꢀ1816.4

ꢀ1814.8

ꢀ1815.0

ꢀ1812.1

ꢀ1812.8

ꢀ1817.2

ꢀ1818.8

ꢀ1816.1

ꢀ1848.7

ꢀ1743.5

ꢀ1705.5

ꢀ1716.6

ꢀ1712.5

ꢀ1724.8

ꢀ1730.2

ꢀ1728.0

ꢀ1730.2

ꢀ1733.0

ꢀ1748.2

ꢀ1753.8

ꢀ1744.4

ꢀ1776.5

1251.6

1291.6

1282.7

1285.8

1277.4

1266.2

1266.8

1264.9

O^ꢀ

COO^ꢀ

NMe₂

OMe

Me

ꢀ334 and ꢀ521 ppm, respec-

F

Cl

Br

COOMe

CN

tively, to s^p(Se)_zz. While the

1264.2 y₁₇!y₂₀transition contributes

ꢀ580 ppm to s^p(Se)_zz; suitable

1255.1

1244.6

1251.1

1215.7

transitions responsible for y₁₆

NO₂

NH₃

+

cannot be found, other than the

y₁₆!y₂₄transition (s^p(Se)_zz=

ꢀ164 ppm). They must be

[a] Structures were optimized with the 6–311+G

T

N

other nuclei at the DFT (B3LYP) level, assumingpl and pd for each of Y. s(Se) are calculated based on the

DFT-GIAO method with the same basis sets.^[29]

spread over a wide range of y_j.

Contributions

from

the

p_z(Se) character were also ex-

amined

(Table 10).

The

Table 6. Contributions of each y_ion s^p(Se) and the components in SeH₂, together with the energies and the

s^p(Se)_(zi)value of ꢀ646 ppm

corresponds to 69% of total

s^p(Se) and 99% of s^p(Se)_(zi)

comes from y₁₈. Although

main characters.^[a]

i in y_i

e [eV]

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

s^p(Se)

sym

character

4p_z(Se)

18

17

16

15

10–14

9

8

7

6

5

ꢀ6.91

ꢀ1334.9

ꢀ4.4

88.7

ꢀ1.3

1.6

ꢀ586.8

104.2

ꢀ15.1

1.7

ꢀ1.2

ꢀ7.8

2.1

ꢀ0.4

ꢀ333.9

ꢀ521.4

ꢀ16.2

6.6

ꢀ640.7

ꢀ78.0

ꢀ149.3

ꢀ5.3

2.3

ꢀ7.1

ꢀ61.3

ꢀ3.3

0.1

B1

A1

B2

ꢀ9.86

s

ACHTREUNG

s^p(Se)_xx(zi)

(ꢀ1344 ppm),

ꢀ11.53

ACHTREUNG

s^p(Se)_yy(zi)(ꢀ595 ppm), and

s^p(Se)_{zz(xi’+yi’)}(ꢀ1031 ppm) con-

tribute much to the downfield

ꢀ19.58

A1

ACHTREUNG

[b]

[c]

3d(Se)

3p_z(Se)

ꢀ156.63

ꢀ156.85

ꢀ156.95

ꢀ212.85

ꢀ1418.53

ꢀ1418.60

ꢀ13.3

0.0

B1

A1

B2

A1

B1

A1

B2

A1

0.0

ꢀ186.1

s

ACHTREUNG

shift,

s^p(Se)_{xx(xi’+yi’)}

and

ꢀ1.9

0.0

4.3

0.0

5.9

-8.1

s

ACHTREUNG

s^p(Se)_{yy(xi’+yi’)}contribute to the

upfield shift. The results show

that vacant orbitals containing

the p_z(Se) character do not con-

tribute to the downfield shift in

SeH₂. After survey of SeH₂,

0.2

0.0

25.4

2.4

3 s(Se)

2p_z(Se)

3.4

8.2

0.0

4

3

1, 2

1–18

0.0

ꢀ0.7

s

ACHTREUNG

ꢀ1418.63

ꢀ2.3

0.0

s

ACHTREUNG

[d]

0.0

ꢀ1031.5

1 s(Se), 2 s(Se)

ꢀ1263.6

ꢀ497.7

ꢀ930.9

[a] Calculated usinga utility program (NMRANAL-NH98G). [b] ꢀ57.65 to ꢀ57.33 eV. [c] Symmetries are B2,

A1, A2, B1, and A1 for i=10–14, respectively. [d] ꢀ12350.82 and ꢀ1580.42 eV for i=1 and 2, respectively. next extension is to elucidate

the orientational effect of 4–6.

effect in 4–6. Contributions by p_z(Se) in y_iwill be discussed

separately from those by p_x(Se) and p_y(Se) in y_i. The sepa-

ration of contributions of p_x(Se) from those of p_y(Se) is also

attempted, if possible. However, before we discuss the re-

sults for 4–6, the s^p(Se) values of SeH₂will be surveyed

first.^[37,38]

Analysis of orientational effect in 4a–6a: The s(Se) param-

eters of 4–6 are collected in Tables 3–5, respectively. The

s^p(Se) and s^t(Se) of 4a(pd) (Y=H) were evaluated to be

larger (more upfield) than those of 4a(pl) by 43 and

46 ppm, respectively; these shifts correspond to the orienta-

tional effect caused by Ph in 4a.^[39]The inverse orientational

effect is predicted for 5a (Y=H). The values of s^p(Se) and

s^t(Se) of 5a(pd) are smaller than those of 5a(pl) by 41 and

3836

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

Table 7. Contributions of each y_i!y_jtransition on s^p(Se)_xx, s^p(Se)_yy, and s^p(Se)_zzin SeH₂, together with the

energy differences and characters of y_j.^[a,b]

and

s^p(Se)_yy(zi)=ꢀ602 ppm

(Table 10). The value of

s^p(Se)_{xx(yi’+xi’)}is 9 ppm, which is

mainly produced from [p_y(Se)

in y_i; p_z(Se) in y_j]. The results

show that p_z(Se) in vacant y_j

contributes little in 4a(pd), al-

y_i!y_j

De [eV]

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

character of y_j

18(B1)!19(A1)

18(B1)!20(B2)

18(B1)!24(A1)

18(B1)!32(B2)

17(A1)!20(B2)

17(A1)!32(B2)

16(B2)!24(A1)

16(B2)!31(B1)

6.28

6.67

8.69

15.79

9.62

18.74

13.31

20.18

0

ꢀ254

0

s*

ACHTREUNG

ꢀ1250

0

ACHTREUNG

0

ꢀ138

0

ꢀ257

s*(SeH₂: a₁’)

[c]

0

ꢀ151

0

ꢀ580

s*(SeH₂: b₂)

G

though

s^p(Se)_{yy(yi’+xi’)}

is

[c]

0

ꢀ316 ppm. Instead, s^p(Se)_{zz(yi’+xi’)}

(ꢀ1443 ppm) contributes to a

large downfield shift, which

must come from the various

contributions of 4p_x(Se) and

ꢀ164

0

s*(SeH₂: a₁’)

s*(SeH₂: b₂’)

ꢀ124

0

[a] Calculated using a utility program (NMRANAL-NH98G). [b] Values are shown if the magnitude of the

total contribution, js^p(Se)_xx+s^p(Se)_yy+s^p(Se)_zzj, is larger than 120 ppm. [c] Would be 5p_z(Se).

4p_y(Se) in s

ACHTREUNG

A

s

ACHTREUNG

49 ppm, respectively. While the values of s^p(Se) and s^t(Se)

for 5a(pl) are smaller than those of 4a(pl) by 90 and

83 ppm, respectively, the values of 5a(pd) are smaller than

those of 4a(pd) by 174 and 178 ppm, respectively. The dif-

ferences are ꢀ84 and ꢀ95 ppm, respectively, which also cor-

respond to the orientational effect of the Ph group in 5a

and 4a. The more effective contribution to downfield shifts

p(Ph) interaction could contribute to the values in 4a(pd).

Typical transitions in 4a(pd) are depicted in Figure 4b.

The s^p(Se) values of 5a and 6a are analyzed similarly. For

5a(pl), the y₄₂(HOMO: 4p_z(Se)-p₂) and y₃₉(s(C_PhSeC_Me

:

a₁)) orbitals contribute large downfield shifts to s^p(Se),

ꢀ363 and ꢀ857 ppm, respectively. The y₄₂(HOMO:

4p_z(Se)) and y₃₉(s(C_PhSeC_Me: a₁)) orbitals of 5a (pd) con-

tribute ꢀ591 and ꢀ593 ppm, respectively, to s^p(Se). The

s^p(Se)_(zi)values of 5a (pl) and 5a (pd) are ꢀ769 and

ꢀ833 ppm, respectively. The difference contributes to the

orientational effect of 5a. The more effective contribution

ꢀ

by the Se C_Mebond in 5a(pd), relative to 5a(pl), must be

responsible for the results.

The y₃₈orbital (HOMO) of 4p_z(Se)-p₂in 4a(pl) contrib-

utes ꢀ220 ppm to s^p(Se) (s^p(Se)_xx=ꢀ454 ppm and

s^p(Se)_yy=ꢀ200 ppm). The magnitude seems smaller than

that expected from the p–p conjugation. The y₃₈!y₄₁

(s*(C_PhSeH:b₂) transition is mainly responsible for s^p(Se)_xx

(ꢀ325 ppm) and s^p(Se)_yy(ꢀ429 ppm). Although y₃₆and y₃₂

also contain 4p_z(Se) and p(Ph) character, they contribute to

ꢀ

by the Se C_Mebond in 5a(pd) to the downfield shifts than

that in 5a(pl) must be responsible for the effect.

In the case of 6a, y₅₈(HOMO: 4p_z(Se)-p_2(pl)) and y₅₃

(s(C_PhSeC_Ph: a₁)) contribute large downfield shifts to s^p(Se)

with ꢀ631 and ꢀ1084 ppm, respectively. The s^p(Se)_yy(zi)

(ꢀ1800 ppm), s^p(Se)_{xx(xi’+yi’)}(ꢀ1309 ppm), and s^p(Se)_{zz(xi’+yi’)}

(ꢀ1805 ppm) components contribute greatly to downfield

shifts, whereas s^p(Se)_xx(zi)(ꢀ6 ppm) is very small. The axes

of 6a were bisected in the Gaussian 98 program.^[32]

upfield shifts. Instead, large downfield shifts are brought

p

ꢀ

about by y₃₅of s(C_PhSeH:a₁) and y₃₃of s

A

value for y₃₅is ꢀ568 ppm (s^p(Se)_xx=ꢀ367 ppm, s^p(Se)_yy=

ꢀ217 ppm, and s^p(Se)_zz=ꢀ1121 ppm) and the s^p(Se) value

for y₃₃is ꢀ515 ppm (s^p(Se)_yy=ꢀ767 ppm and s^p(Se)_zz=

ꢀ765 ppm). The y₃₅!y₄₁and y₃₃!y₄₁transitions mainly

contribute to s^p(Se)_zzwith values of ꢀ1001 and ꢀ749 ppm,

respectively. Typical transitions in 4a(pl) are depicted in Fig-

ure 4a, which clarifies and visualizes the discussion.

What mechanism is operatingin the Y dependence? The

s^p(Se) of 4–6 in pl and pd are analyzed next.

Y dependence in 4–6: The Y dependence in 4–6 was exam-

ined employing s^p(Se) and the components in Tables 3–5.

While s^p(Se)_xxand s^p(Se)_yyin 4(pl) shift downfield and up-

field by 200–300 ppm, respectively, when Y=H is replaced

by Y=non-H,^[41]those in 5(pl) shift upfield and downfield

by 100–200 ppm, respectively.^[42]Such trend is not clear in

4(pd), 5(pd), and 6.

As shown in Table 10, s^p(Se)_xx(zi), s^p(Se)_yy(zi), and s^p(Se)_(zi)

of 4a(pl) are ꢀ1305, 507, and ꢀ270 ppm, respectively. Al-

though the main interaction in 4a(pl) is the p_z(Se)–p(Ph)

conjugation, s^p(Se)_(zi)of ꢀ270 ppm corresponds to only

19% of total s^p(Se), which is a strikingcontrast to the case

of SeH₂. It must be the reflection of s^p(Se)_yy(zi)=507 ppm.

The large negative value of s^p(Se)_{yy(xi’+yi’)}=ꢀ1553 ppm dem-

To clarify the behavior of s^p(Se) in 4(pl) and 4(pd), s^p(Se)

was plotted versus (s^p(Se)_xx+s^p(Se)_yy). Figures 5a and b ex-

hibit the plots of s^p(Se) in 4(pl) for all Y examined (contain-

inganionic and cationic substituents) and that for Y of non-

ionic groups, respectively. The correlation with all Y is very

good and that with Y of non-ionic groups is also good.^[43]

The results are given in Table 2 (entries 11 and 12). The plot

of s^p(Se) versus (s^p(Se)_xx+s^p(Se)_yy) for 4(pd) gives a very

good correlation for all Y;^[43]this result may be due to a

very wide range of s^p(Se) (entry 13).^[44]Figure 5c shows a

similar plot for 4(pd) with Y beingnon-ionic groups. The

correlation is very good and is given in Table 2 (entry 14).

onstrates the substantial contribution of p_z(Se) in p*

to the downfield shift in 4a(pl).

A

In the case of 4a(pd), a very large downfield shift is

brought about by y₃₈(HOMO: n_p(Se)), which is constructed

from almost pure 4p_z(Se).^[40]Since y₃₈is filled with elec-

trons, both 4p_x(Se) and 4p_y(Se) in s*(C_PhSeH) are expected

R

to play a predominant role in the downfield shift. The

expectation is demonstrated by the large negative values

of s^p(Se)_xx(ꢀ1514 ppm) and s^p(Se)_yy(ꢀ468 ppm) in

y₃₈(Table 8), together with s^p(Se)_xx(zi)=ꢀ1831 ppm

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3837

W. Nakanishi et al.

Table 8. Contributions of each y_ion s^p(Se) and the components in 4a–6a, together with the energies and the

good correlations both for 5(pl)

with all Y and 5(pd) with Y of

non-ionic groups (entries 15

and 16). Those of s^p(Se) versus

(s^p(Se)_xx+s^p(Se)_yy) in 6(pl)

with all Y and in 6 (pd) with Y

of non-ionic groups are also

very good (entries 17 and 18).

The a values of 0.31–0.34 show

that (s^p(Se)_xx+s^p(Se)_yy) control

s^p(Se) in 5 and 6.^[43]

main characters.^[a]

i in y_i

e [eV]

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

s^p(Se)

main character

4a(pl)

38^[b]

37

36

35

34

33

32

31

1–38

D^[c]

4a (pd)

38^[b]

37

36

35

34

33

ꢀ6.08

ꢀ7.29

ꢀ8.05

ꢀ9.02

ꢀ9.80

ꢀ10.28

ꢀ10.53

ꢀ11.41

ꢀ454.4

3.5

ꢀ509.3

ꢀ366.9

0.2

ꢀ12.6

ꢀ310.6

ꢀ32.9

ꢀ1572.3

0.0

ꢀ199.6

ꢀ4.9

ꢀ3.3

ꢀ219.6

ꢀ1.6

4p_z(Se)-p₂

p₂’

4p_z(Se)+p₂

s(C_PhSeH: a₁)

s(Ph)

643.9

ꢀ4.0

43.5

ꢀ217.0

ꢀ25.5

ꢀ766.9

596.0

ꢀ251.1

ꢀ1046.1

0.0

ꢀ1121.1

ꢀ568.3

34.0

2.9

ꢀ765.0

0.1

358.0

ꢀ1696.8

0.0

ꢀ514.8

95.2

24.7

s(Se-H)

T

4p_z(Se)+p₁

s(C_PhSeH: b₂)

ꢀ1438.4

The mechanism of the Y de-

pendence can be elucidated in

more detail, if s^p(Se)_xxand

s^p(Se)_yycan be analyzed sepa-

rately. To get an image in the

behavior of s^p(Se)_xxand

s^p(Se)_yy, together with s^p(Se)_zz,

they are plotted versus s^p(Se).

Figure 6 shows the results for

4(pd), 5(pd), and 6(pl). The cor-

relations for 4(pd) are linear

and both s^p(Se)_xxand s^p(Se)_yy

0.0

ꢀ6.53

ꢀ7.28

ꢀ7.34

ꢀ8.98

ꢀ9.85

ꢀ10.04

ꢀ10.76

ꢀ11.53

ꢀ1514.4

3.1

ꢀ468.3

ꢀ33.8

ꢀ5.8

ꢀ0.2

ꢀ263.8

ꢀ0.5

ꢀ661.0

ꢀ98.2

ꢀ3.6

4p_z(Se)

s(Se-H)+p₂

C

ꢀ4.5

p₂’

ꢀ124.1

ꢀ79.7

ꢀ36.5

60.4

ꢀ16.9

ꢀ5.6

ꢀ791.6

ꢀ0.2

ꢀ310.8

ꢀ28.5

ꢀ87.4

ꢀ77.9

52.0

s(C_PhSeH: a₁)

s(Ph)

ꢀ104.0

ꢀ48.7

ꢀ68.8

ꢀ917.9

128.2

ꢀ121.8

ꢀ245.5

198.1

s

ACHTREUNG

32

31

s

ACHTREUNG

26.8

s(C_PhSeH: b₂)

1–38

D^[c]

5a(pl)

42^[b]

41

40

39

38

1–42

D^[e]

5a(pd)

42^[b]

41

40

39

38

1–42

D^[e]

6a^[f]

58^[b]

57

56

55

54

53

ꢀ1821.8

ꢀ1439.8

ꢀ1393.2

ꢀ249.5

257.0

45.2

ꢀ5.75

ꢀ7.17

ꢀ7.78

ꢀ8.60

ꢀ9.46

ꢀ862.5

ꢀ4.9

ꢀ222.3

2.0

ꢀ3.4

ꢀ362.7

ꢀ2.1

4p_z(Se)-p₂

p₂’

increase alongwith

s^p(Se).

ꢀ596.5

ꢀ56.2

ꢀ143.8

ꢀ1893.4

0.0

104.4

ꢀ490.2

ꢀ45.1

ꢀ999.0

0.0

ꢀ4.0

ꢀ165.4

ꢀ856.7

165.5

4p_z(Se)+p₂

s(C_PhSeC_Me: a₁)

While s^p(Se)_xxand s^p(Se)_yyof

5(pd) and 6(pl) are on smooth

lines, each slope for s^p(Se)_yyis

the inverse of that for s^p(Se)_xx.

The slopes for s^p(Se)_zzare very

small for all cases.^[43]

ꢀ2023.6

685.3

s

(Se-C_Me)+s(Ph)^[d]

G

ꢀ1684.9

ꢀ1525.8

0.0

ꢀ6.03

ꢀ7.14

ꢀ7.26

ꢀ8.56

ꢀ9.48

ꢀ1245.5

ꢀ26.3

ꢀ525.8

ꢀ26.1

ꢀ5.8

ꢀ0.1

ꢀ381.6

ꢀ0.2

ꢀ590.5

ꢀ144.7

ꢀ5.7

4p_z(Se)

(Se-C_Me)+p₂

s

ACHTREUNG

ꢀ11.0

p₂’

s^p(Se)_yywas plotted versus

s^p(Se)_xxfor 4–6. A fairly good

correlation is obtained in the

plot of 4(pd) for Y of non-ionic

groups with a=0.23 and r=0.85

(entry 19 in Table 2). The a

value of 0.23 for 4(pd) may in-

dicate that the 4p_y(Se) orbital

ꢀ25.0

ꢀ200.2

ꢀ1553.7

ꢀ593.0

s(C_PhSeC_Me: a₁)

ꢀ310.8

ꢀ1954.2

ꢀ60.8

42.6

917.5

216.4

s

(Se-C_Me)+p(Ph)^[d]

U

ꢀ1110.5

ꢀ111.5

ꢀ1657.8

27.1

ꢀ1574.2

ꢀ48.4

ꢀ5.82

ꢀ7.11

ꢀ7.32

ꢀ7.41

ꢀ7.80

ꢀ8.73

ꢀ9.29

ꢀ611.1

5.4

ꢀ234.0

ꢀ36.4

638.1

ꢀ1278.2

ꢀ2.9

ꢀ3.2

ꢀ4.9

ꢀ630.9

ꢀ0.8

4p_z(Se)-p_2(pl)

p_2(pl)

(SeC_Ph(pl))-p_2(pd)

p_2(pd)

4p_z(Se)+p_2(pl)

’

ꢀ16.3

ꢀ278.5

ꢀ176.2

ꢀ18.8

121.7

s

ACHTREUNG

ꢀ21.1

1.2

’

ꢀ

involved in the Se C_Arbond is

ꢀ267.0

ꢀ36.6

ꢀ6.1

ꢀ954.5

ꢀ2260.1

524.4

ꢀ1083.7

s

N

52

1–58

24.7

ꢀ454.6

31.5

s

to the Y dependence than the

4p_x(Se) orbital involved in the

ꢀ1314.7

ꢀ2120.3

ꢀ1816.4

ꢀ1750.5

[a] Calculated usinga utility program (NMRANAL-NH98G). [b] Correspondingto HOMO. [c]

pd))_AAꢀs

^p(Se:4a(pl))_AA, in which A=x, y, z, and zero. [d] Containingthe s(C_PhSeC_Me: b₂) character. [e] s^p-

^p(Se:5a(pl))_AA, in which A=x, y, z, and zero. [f] Axes are in the bisected form in Gaussi-

s

A

U

ꢀ

Se H bond. The plot for 5(pd)

ACHTREUNG

with all Y also gives a good cor-

relation, but with a negative

correlation constant, a=ꢀ0.33

and r=0.93 (entry 20). While

the 4p_y(Se) orbital involved in

A

(pl or pd))_AAꢀs

ACHTREUNG

an 98.^[32]

Points correspondingto Y =O^ꢀ, S^ꢀ, and Se^ꢀwere also

added in Figure 5c, although they deviate from the correla-

tion. The correlation constants in entries 11–14 are 0.31–

0.36, which are very close to one third (cf: Equation (2)).

The results exhibit that the Y dependence of s^p(Se)_zzis neg-

ligible in 4 (see, Figure 6).^[43]Namely, (s^p(Se)_xx+s^p(Se)_yy) ef-

fectively controls s^p(Se) of 4.

the Se C_Arbond of 5(pd) contributes to the regular direc-

ꢀ

p

ꢀ

tion to s (Se), the 4p_x(Se) orbital involved in the Se C_Me

bond would work to shift to the inverse direction. Similar

plots for 6(pl) and 6(pd) with all Y give good correlations,

with a=ꢀ0.36 and r=0.92 (entry 21) and a=ꢀ0.65 and r=

0.95 (entry 22), respectively. The 4p_y(Se) orbital involved in

ꢀ

the Se C_Arbond contributes to the regular direction in

p

ꢀ

Similar trends in the Y dependence are also observed in 5

s (Se), but the 4p_x(Se) orbital in the Se C_Phbond seems to

and 6. The plots of s^p(Se) versus (s^p(Se)_xx+s^p(Se)_yy) give

shift to the inverse direction both in 6(pl) and 6(pd). The re-

3838

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

Table 9. Contributions of each y_i!y_jtransition on s^p(Se)_xx, s^p(Se)_yy, and s^p(Se)_zzin 4a(pl) and 4a(pl), togeth-

er with the energy differences and the main characters of y_j.^[a,b]

Mechanism of Y dependence:

The fact that s^p(Se) of 4–6 in pl

and pd are effectively control-

led by (s^p(Se)_xx+s^p(Se)_yy) lead

us to elucidate the mechanisms

of the Y dependence in the ori-

entational effect, based on the

magnetic perturbation theory.

The mechanisms are explained

exemplified by 4(pl) and 4(pd).

The main interaction be-

tween 4p_z(Se) and p_z(Y) in

y_i!y_j

4a(pl)

De [eV]

s^p(Se)_xx

s^p(Se)_yy

s^p(Se)_zz

character of y_j

38

36

U

5.43

5.97

6.29

8.06

7.40

ꢀ325

ꢀ178

ꢀ41

ꢀ264

ꢀ124

0

ꢀ429

11

ꢀ116

ꢀ17

ꢀ163

0

s*(C_PhSeH: b₂)

s*(Se-C_Ph)

A

s*(C_PhSeH: b₂’)

s*(C_PhSeH: a₁)

s*(C_PhSeH: b₂)

35(A’)!41(A’)

35(A’)!43(A’)

8.37

9.23

ꢀ1001

ꢀ172

0

s*(C_PhSeH: b₂’)

[c]

35(A’)!47

35(A’)!81

G

10.11

17.20

9.63

10.16

11.37

12.25

12.91

17.46

ꢀ46

ꢀ135

0

ꢀ142

ꢀ33

0

[d]

G

0

33(A’)!41(A’)

33(A’)!42(A’)

ꢀ749

137

0

ꢀ127

151

ꢀ128

s*(C_PhSeH: b₂)

0

s*

A

)

4(pl) is the 4p_z(Se)–p(C₆H₄)–

G

33(A’)!47

A

ꢀ20

ꢀ438

5p_z(Se)

p_z(Y) type of interaction, which

modifies the contributions of

33(A’)!51(A’)

33(A’)!54(A’)

33(A’)!79(A’)

4a(pd)

0

s*(C_PhSeH: a₁)

[d]

0

[d]

4p_z(Se) in the p

A

p*(SeC₆H₄Y) molecular orbi-

AHCTREUNG

38

C

6.18

6.43

6.80

7.37

7.85

8.90

8.63

9.25

ꢀ433

ꢀ344

ꢀ121

ꢀ13

ꢀ277

ꢀ146

0

ꢀ264

ꢀ48

ꢀ200

ꢀ116

0

s*(C_PhSeH: b₂)

tals.

4p_z(Se)

Admixtures

in

between

s*

A

modified

p-

s*(C_PhSeH: b₂)

N

(SeC₆H₄Y)

s*

ACHTREUNG

s*

A

)

[d]

ꢀ16

0

ACHTREUNG

35(A’)!41(A’)

0

ꢀ284

s*(C_PhSeH: b₂)

AHCTREUNG

35(A’)!43(A’)

0

ꢀ175

give rise to the Y dependence,

when a magnetic field is ap-

plied. Consequently, when the

Y subsitiutent is an electron

donor or an electron acceptor,

dSe of 4(pl) depend on the sub-

stituent. The mechanism seems

applicable to a wide range of Y

(see, Figure 5a). In the case of

4(pd), the 4p_z(Se) orbital re-

mains in n_p(Se) in the almost

[a] Calculated using a utility program (NMRANAL-NH98G). [b] Values are shown if the magnitude of the

total contribution, js^p(Se)_xx+s^p(Se)_yy+s^p(Se)_zzj, is larger than 120 ppm. [c] Would be 5p_z(Se). [d] Difficult to

specify.

Table 10. Contributions of “y_i” and “y_i’” to s^p(Se)_AA(A=x, y, z, and zero) in 4a–6a, together with SeH₂.^[a]

y

s^p(Se)_AA

x

y

z

zero

SeH₂

y₁–y₁₈

“y_i”^[b]

“y_i’”

4a(pl)

y₁–y₃₈

“y_i”^[c]

“y_i’”

4a(pd)

y₁–y₃₈

“y_i”^[c]

“y_i’”

5a(pl)

y₁–y₄₂

“y_i”^[c]

“y_i’”

s^p(Se)_AA

ꢀ1264

ꢀ1344

80

ꢀ498

ꢀ595

97

ꢀ1031

0

ꢀ1031

ꢀ931

s^p(Se)_AA(zi)

s^p(Se)_{AA(xi’+yi’)}

ꢀ646

ꢀ285

pure form.^[40]The s

(C₆H₄)–p_x(Y) interaction occurs

A

s^p(Se)_AA

ꢀ1572

ꢀ1305

ꢀ267

ꢀ1046

507

ꢀ1553

ꢀ1697

ꢀ11

ꢀ1438

ꢀ270

s^p(Se)_AA(zi)

s^p(Se)_{AA(xi’+yi’)}

AHCTREUNG

ꢀ1686

ꢀ1168

instead; this interaction modi-

fies the contributions of 4p_x(Se)

and 4p_y(Se) orbitals in the s-

s^p(Se)_AA

ꢀ1822

ꢀ1831

9

ꢀ918

ꢀ602

ꢀ316

ꢀ1440

3

ꢀ1443

ꢀ1393

ꢀ810

ꢀ583

s^p(Se)_AA(zi)

s^p(Se)_{AA(xi’+yi’)}

G

E

lecular orbitals. Admixtures of

4p_z(Se) in n_p(Se) with 4p_y(Se)

and 4p_x(Se) in modified s*-

s^p(Se)_AA

ꢀ1893

ꢀ1879

ꢀ14

ꢀ999

ꢀ413

ꢀ586

ꢀ1685

ꢀ14

ꢀ1526

ꢀ769

ꢀ757

s^p(Se)_AA(zi)

s^p(Se)_{AA(xi’+yi’)}

ꢀ1671

5a(pd)

y₁–y₄₂

“y_i”^[c]

“y_i’”

A

s^p(Se)_AA

ꢀ1954

ꢀ1632

ꢀ322

ꢀ1110

ꢀ865

ꢀ246

ꢀ1658

ꢀ2

ꢀ1574

ꢀ833

ꢀ741

4(pd), since n_p(Se) is filled with

electrons. Namely, the Y de-

pendence in 4 (pd) must be

more sensitive to an electron-

donatingY group; this result is

in strikingcontrast to that of

4(pl).

s^p(Se)_AA(zi)

s^p(Se)_{AA(xi’+yi’)}

ꢀ1656

6a^[d]

y₁–y₅₈

“y_i”^[c]

“y_i’”

s^p(Se)_AA

ꢀ1315

ꢀ6

ꢀ2120

ꢀ1800

ꢀ320

ꢀ1816

ꢀ11

ꢀ1750

ꢀ606

s^p(Se)_AA(zi)

s^p(Se)_{AA(xi’+yi’)}

ꢀ1309

ꢀ1805

ꢀ1144

[a] Calculated using a utility program (NMRANAL-NH98G). [b] Belonging to the B1 symmetry. [c] Belonging

to the A’’ symmetry. [d] Axes are in the bisected form in Gaussian 98.^[32]

The mechanisms proposed

for 4(pl) and 4(pd) are support-

sults are summarized in Scheme 4. The correlations are poor

for 4(pl) and 5(pl).

ed by the values found for s^p(Se)_SCS. The s^p(Se)_SCSvalues

for 4(pl) are in an order of Y=NO₂(ꢀ39.1)<CN (ꢀ32.4)<

F (ꢀ0.3)ꢁH (0.0)<Me (5.3)<NMe₂(29.4 ppm) and those

for 4(pd) are Y=CN (ꢀ14.8)<NO₂(ꢀ8.8)<H (0.0)<F

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3839

W. Nakanishi et al.

Figure 4. Contribution of each y_i!y_jtransition to s^p(Se): a) for 4a(pl)

and b) for 4a(pd).

(12.9)<Me (15.3)<NMe₂(33.7 ppm). It is demonstrated

that both electron-donatingand electon-acceptingY moiet-

ies contribute effectively to s^p(Se)_SCSin 4(pl), whereas elec-

tron-donatingY groups are more effective than electron-ac-

ceptinggroups in 4(pd). The predictions made for 4(pl) and

4(pd) are observed in 1 and 2, respectively. Similar mecha-

nisms must operate for 5 and 6.

Details of the mechanism seem complex, since the contri-

butions of Y=NO₂and CN are different in 4(pl) and 4(pd).

To clarify the mechanisms in more detail, the s^p(Se)_(zi)SCS

values for 4(pl) and 4(pd) were evaluated for Y=NH₂, Me,

F, CN, and NO₂, in addition to Y=H, by summarizing

s^p(Se) over “y_i”.^[35]The s^p(Se)_{(xi’+yi’)SCS}values were also cal-

culated accordingto Equation (8). The results are collected

in Table 11. The signs of the s^p(Se)_(zi)SCSvalues are for most

cases the same as those found for s^p(Se)_SCS. The s^p(Se)_SCS

parameters are roughly controlled by s^p(Se)_(zi)SCSand im-

proved numerically by s^p(Se)_{(xi’+yi’)SCS}both for 4(pl) and

4(pd). The magnitudes of s^p(Se)_(zi)SCSand s^p(Se)_{(xi’+yi’)SCS}are

less than 90 ppm, except for 4(pl: Y=NO₂) for which values

of 500 ppm are found; this result must be a reflection of the

good electron-accepting character and the ability to extend

p-systems. However, it is noteworthy that the total effect by

NO₂fits almost in the range of the substituents. The orienta-

tional effect, together with the mechanism of Y dependence,

Figure 5. Plots of s^p(Se) versus s^p(Se)_xx+s^p(Se)_yy: a) For 4(pl) with all Y,

b) for 4(pl) with Y of non-ionic groups, and c) for 4(pd) with Y of non-

&

ionic groups ( ), together with Y of anionic groups (&).

is well established based on the experimental and theoreti-

cal investigations.

Conclusion

The orientational effect is empirically established by the Y

dependence on d(Se) of 1 and 2. The sets of d(Se) of 1 and

2 can be used as the standards for pl and pd, respectively,

when d(Se) of aryl selenides are analyzed. The Y depen-

dence observed in 1 and 2 is well explained by analyzingthe

3840

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

Scheme 4. Contributions of s^p(Se)_yyrelative to s^p(Se)_xxin 4–6, in which

the contribution of each s^p(Se)_xxis taken to be 1.00. Contributions are

depicted by reducingto p (Se) and p_y(Se), respectively.

x

we have demonstrated that (s^p(Se)_xx+s^p(Se)_yy) effectively

control s^p(Se) of 4–6 in pl and pd. The results allowed us to

elucidate the mechanisms of the Y dependence in the orien-

tational effect, based on the magnetic perturbation theory.

The main interaction in pl is the n_p(Se)–p

jugation. Therefore, the Y dependence in pl occurs through

admixtures of 4p_z(Se) in modified p(SeC₆H₄Y) and p*-

(SeC₆H₄Y) molecular orbitals with 4p_x(Se) and 4p_y(Se) in s-

(CSeX) and s*(CSeX) molecular orbitals (X=H or C). The

main interaction in pd is the s(CSeX)–p(C₆H₄)–p_x(Y) inter-

action, which modifies both s(C_ArSeH) and s*(C_ArSeH) mo-

(C₆H₄)–p_z(Y) con-

AHCTREUNG

A

ACHTREUNG

A

ACHTREUNG

A

ACHTREUNG

lecular orbitals. The Y dependence in pd mainly originates

from admixtures of 4p_z(Se) in the n_p(Se) orbital with

4p_x(Se) and 4p_y(Se) in modified s*(CSeX) molecular orbi-

N

tals, since n_p(Se) is filled with electrons. Consequently, both

elctron-donatingand electron-acceptingY moieties are ef-

fective in pl, whereas electron-donatingY moieties must be

more effective in pd. The predictions for pl and pd are ob-

served in 1 and 2, respectively. Contributions of each molec-

ular orbital and each transition on s^p(Se) have been evaluat-

ed, which enables us to recognize and visualize the effect

clearly.

The effect of R in ArSeR is also important, which will be

discussed elsewhere, together with the applications of the

method.

p

Figure 6. Plots of s^p(Se)_xx

), s (Se)_yy( ), and s (Se)_zz( ) versus s (Se):

~

&

*

(

a) for 4(pd) with Y of non-ionic groups, b) for 5(pd) with all Y, and c) for

6(pl) with all Y.

Experimental Section

s^t(Se) values of 6(pl) and 6(pd), respectively, calculated with

the DFT-GIAO method. While s^t(Se) of 4a(pl) is predicted

to be more negative than that of 4a(pd) by 46 ppm, s^t(Se)

of 5a(pl) is evaluated to be larger than that of 5a(pd) by

49 ppm. The differences correspond to the orientational

effect of the Ph group in 4a and 5a.

General considerations: Manipulations were performed under a nitrogen

or an argon atmosphere with standard vacuum-line techniques. Glassware

was dried at 1308C overnight. Solvents and reagents were purified by

standard procedures as necessary. Meltingpoints were uncorrected.

NMR spectra were recorded at 258C on a JEOL JNM-AL 300 spectrom-

eter (¹H, 300 MHz; ¹³C, 75.45 MHz; ⁷⁷Se, 57.25 MHz). The ¹H, ¹³C, and

⁷⁷Se chemical shifts are given in ppm relative to those of Me₄Si, internal

CDCl₃in the solvent, and external MeSeMe, respectively. Column chro-

matography was performed on silica gel (Fuji Silysia BW-300), acidic alu-

mina, and basic alumina (E. Merk). Flash column chromatography was

performed with 300–400 mesh silica gel, acidic alumina, and basic alumi-

na and analytical thin-layer chromatography was performed on precoated

Very good correlations are obtained in the plots of s^p(Se)

versus (s^p(Se)_xx+s^p(Se)_yy) for 4–6 in pl and pd. Since the

correlation constants (a) are very close to one third, s^p(Se)_zz

must be almost constant when Y is changed. Consequently,

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3841

W. Nakanishi et al.

Table 11. Contributions of “y_i” and “y_i’” to s^p(Se)_SCSin some 4(pl) and 4(pd)for different Y substituents.^[a]

MeSeMe): d=242.4 ppm; elemental

analysis calcd (%) for C₂₁H₁₆Se:

72.62, H 4.64; found: C 72.76, H 4.57.

C

NO₂

CN

H

F

Me

NH₂

4(pl)

Preparation of 9-(p-fluorophenylsela-

nyl)anthracene (1e): A similar method

to that desribed for the preparation of

1a was used. Compound 1e was isolat-

ed in 26% yield as yellow needles.

s^p(Se)_SCS

ꢀ33^[b]

ꢀ34

ꢀ85

51

0

N

0

8

21

ꢀ13

23

87

ꢀ64

s^p(Se)_(zi)SCS

s^p(Se)_{(xi’+yi’)SCS}

4(pd)

ꢀ478

N

ꢀ3

445

3

s^p(Se)_SCS

ꢀ9

ꢀ64

55

ꢀ15

ꢀ42

27

0

11

ꢀ10

21

15

6

9

18

53

ꢀ35

M.p.

112.5–113.48C;

¹H NMR

s^p(Se)_(zi)SCS

s^p(Se)_{(xi’+yi’)SCS}

(300 MHz, CDCl₃, TMS): d=6.75 (dd,

J=4.5, 9.0 Hz, 2H), 7.06 (dd, J=4.5,

6.0 Hz, 2H), 7.45–7.56 (m, 4H), 8.01

(d, J=7.8 Hz, 2H), 8.55 (s, 1H),

[a] Calculated usinga utility program (NMRANAL-NH98G). [b] About 6 ppm upfield relative to that derived

from Table 3.

8.85 ppm

¹³C NMR (75 MHz, CDCl₃, TMS): d=

(F,C)=21.7 Hz), 125.5, 127.0, 127.2, 127.8 (⁴J

(F,C)=3.3 Hz),

128.9, 129.3 (³J(Se,C)=9.5 Hz), 130.3, 131.0 (³J(F,C)=7.6, ²J

(Se,C)=

13.6 Hz), 131.9, 134.9 (³J(Se,C)=5.6 Hz), 161.5 ppm (¹J

(F,C)=245.2 Hz);

⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=245.4 ppm (⁵J

(Se,F)=6.2 Hz);

(d,

J=8.7 Hz,

2H);

116.2 (²J

A

ACHTREUNG

silica gel plates (60F-254) with the systems (v/v) indicated. Elemental

analyses were performed usinga J-Science Lab Co., JM10 Micro Corder.

A

G

ACHTREUNG

G

ACHTREUNG

Preparation of 9-(phenylselanyl)anthracene (1a): Under an argon atmos-

phere, 9-bromoanthracene (1.00 g, 3.89 mmol) was dissolved in dry dieth-

yl ether (70 mL) and the solution was added to a flask that contained

magnesium (0.10 g, 4.11 mmol) and dry diethyl ether (10 mL). The solu-

tion was refluxed for 2 h. A solution of diphenyl diselenide (1.21 g,

3.89 mmol) in diethyl ether (30 mL) was then added. Then the reaction

mixture was refluxed for 1 h. Then, 5% hydrochloric acid (20 mL) and

benzene (100 mL) were added. The organic layer was separated and then

washed with brine, 10% aqueous solution of sodium hydroxide, saturated

aqueous solution of sodium bicarbonate, and brine again. Then the solu-

tion was dried over sodium sulfate, evaporated, and dried in vacuo. The

crude product was purified by column chromatography (SiO₂, hexane).

Compound 1a was isolated in 65% yield as yellow needles. M.p. 100.8–

AHCTREUNG

elemental analysis calcd (%) for C₂₀H₁₃FSe: C 68.38, H 3.73; found: C

68.23, H 3.70.

Preparation of 9-(p-chlorophenylselanyl)anthracene (1 f):

A similar

method to that desribed for the preparation of 1a was used. Compound

1 f was isolated in 76% yield as yellow needles: M.p. 167.6–168.38C;

¹H NMR (300 MHz, CDCl₃, TMS): d=6.96–7.02 (m, 4H), 7.46–7.56 (m,

4H), 8.03 (dd, J=2.8, 7.2 Hz, 2H), 8.59 (s, 1H), 8.81 ppm (dd, J=2.3,

7.5 Hz, 2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=122.4, 125.6, 127.1,

127.2, 127.6, 128.6, 128.8, 129.4, 130.6, 132.2, 133.3 (²J

(Se,C)=11.6 Hz),

A

134.3 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=250.5 ppm; ele-

mental analysis calcd (%) for C₂₀H₁₃ClSe: C 65.32, H 3.56; found: C

65.43, H 3.49.

1

101.38C; H NMR (300 MHz, CDCl₃, TMS): d=7.01–7.09 (m, 5H), 7.45–

7.54 (m, 4H), 8.02 (dd, J=2.6, 7.1 Hz, 2H), 8.56 (s, 1H), 8.87 ppm (dd,

Preparation of 9-(p-Bromophenylselanyl)anthracene (1g):

A similar

J=0.8, 7.6 Hz, 2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=125.5, 125.7,

method to that desribed for the preparation of 1a was used. Compound

1g was isolated in 69% yield as yellow needles. M.p. 180.4–181.08C;

¹H NMR (300 MHz, CDCl₃, TMS): d=6.91 (d, J=8.5 Hz, 2H), 7.15 (d,

J=8.5 Hz, 2H), 7.47–7.57 (m, 4H), 8.04 (dd, J=2.7, 6.9 Hz, 2H), 8.60 (s,

1H), 8.81 ppm (dd, J=2.7, 6.9 Hz, 2H); ¹³C NMR (75 MHz, CDCl₃,

TMS): d=119.6, 125.6, 126.0, 127.4, 129.0, 129.1, 130.5, 130.6, 131.9,

132.1, 132.5, 135.0 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=

250.6 ppm; elemental analysis calcd (%) for C₂₀H₁₃BrSe: C 58.28, H 3.18;

found: C 58.35, H 3.09.

127.1, 128.3, 128.8, 129.0, 129.1 (³J

C

ACHTREUNG

(57 MHz, CDCl₃, MeSeMe): d=249.0 ppm; elemental analysis calcd (%)

for C₂₀H₁₄Se: C 72.07, H 4.23; found: C 72.33, H 4.15.

Preparation of 9-[(N,N’-dimethylamino)phenylselanyl]anthracene (1b):

A similar method to that desribed for the preparation of 1a was used.

Compound 1b was isolated in 82% yield as yellow needles. M.p. 168.3–

1

170.78C; H NMR (300 MHz, CDCl₃, TMS): d=2.80 (s, 6H), 6.45 (d, J=

9.0 Hz, 2H), 7.15 (d, J=9.0 Hz, 2H), 7.47 (dt, J=1.3, 6.6 Hz, 2H), 7.53

(dt, J=1.6, 6.6 Hz, 2H), 7.99 (dd, J=1.5, 8.1 Hz, 2H), 8.51 (s, 1H),

9.01 ppm (dd, J=1.1, 9.0 Hz, 2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=

Preparation of 9-[p-(ethoxycarbonyl)phenylselanyl]anthracene (1h):

Under an argon atmosphere, 9-bromoanthracene (1.00 g, 3.89 mmol) was

dissolved in dry diethyl ether (50 mL) and the solution was added to a

flask that contained magnesium (0.10 g, 4.11 mmol) and dry diethyl ether

(5 mL). The resultingsolution was refluxed for 2 h. Selenium powder

(0.30 g, 3.80 mmol) was then added and the mixture was refluxed for 2 h.

Then a solution of diazonium, which was prepared from p-ethoxycarbon-

yl aniline (1.93 g, 11.7 mmol) in water (30 mL), was added dropwise . The

mixture was heated to 408C for 30 min, and benzene (100 mL) was

added. The organic layer was separated and was washed with water, 10%

aqueous solution of sodium hydroxide, saturated aqueous solution of

sodium bicarbonate, and brine. Then the solution was dried over sodium

sulfate, evaporated, and dried in vacuo. The crude product was purified

by column chromatography (SiO₂, hexane) to give 1h as yellow needles.

Yield: 5%, m.p. 148.0–149.08C; ¹H NMR (300 MHz, CDCl₃, TMS): d=

1.33 (t, J=7.2 Hz, 3H), 4.30 (q, J=7.2 Hz, 2H), 7.22 (d, J=8.6 Hz, 2H),

40.4, 113.4, 118.4, 125.3, 126.8, 128.8, 129.0, 129.5, 129.8 (³J

ACHTREUNG

10.0 Hz), 131.9 (²J

A

ACHTREUNG

⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=228.0 ppm; elemental analysis

calcd (%) for C₂₂H₁₉NSe: C 70.21, H 5.09, N 3.72; found: C 70.26, H

5.10, N 3.70.

Preparation of 9-(anisylselanyl)anthracene (1c): A similar method to

that desribed for the preparation of 1a was used. Compound 1c was iso-

lated in 67% yield as yellow needles. M.p. 121.4–122.08C; ¹H NMR

(300 MHz, CDCl₃, TMS): d=3.66 (s, 3H), 6.63 (d, J=8.8 Hz, 2H), 7.11

(d, J=9.0 Hz, 2H), 7.49 (dt, J=1.5, 6.6 Hz, 2H), 7.46–7.58 (m, 2H), 8.02

(dd, J=1.8, 7.9 Hz, 2H), 8.55 (s, 1H), 8.95 ppm (ddd, J=1.3, 2.6, 8.8 Hz,

2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=55.2, 114.9, 123.5, 125.4,

127.0, 128.0, 128.9, 129.6 (³J

12.4 Hz), 131.9, 134.9 (³J(Se,C)=5.2 Hz), 158.3 ppm; ⁷⁷Se NMR (57 MHz,

A

N

7.79 (d, J=8.6 Hz, 2H), 7.49–7.59 (m, 4H), 8.04–8.10 (m, 2H), 8.65(s,

A

CDCl₃, MeSeMe): d=236.8 ppm; elemental analysis calcd (%) for

C₂₁H₁₆OSe: C 69.42, H 4.44; found: C 69.49, H 4.39.

1H), 8.69–8.76 ppm (m, 2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=14.8,

60.8, 125.5, 127.0, 127.2, 128.2, 128.9, 129.1, 129.3, 130.3, 130.6, 131.9,

134.9, 139.8, 165.6 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=

265.2 ppm; elemental analysis calcd (%) for C₂₃H₁₈O₂Se: C 68.15, H 4.48;

found: C 68.35, H 4.53.

Preparation of 9-(p-tolylselanyl)anthracene (1d): A similar method to

that desribed for the preparation of 1a was used. Compound 1d was iso-

lated in 75% yield as yellow needles. M.p. 144.8–145.78C; ¹H NMR

(300 MHz, CDCl₃, TMS): d=2.17 (s, 3H), 6.85 (d, J=7.9 Hz, 2H), 6.99

(d, J=8.3 Hz, 2H), 7.54–7.55 (m, 4H), 8.00 (dd, J=1.8, 7.6 Hz, 2H), 8.54-

Preparation of 9-[p-cyanophenylselanyl]anthracene (1i):

A similar

method to that desribed for the preparation of 1a was used. Compound

1i was isolated in 83% yield as yellow needles. M.p. 163.1–164.08C;

¹H NMR (300 MHz, CDCl₃, TMS): d=7.05 (d, J=8.6 Hz, 2H), 7.28 (d,

J=8.6 Hz, 2H), 7.49–7.59 (m, 4H), 8.04–8.10 (m, 2H), 8.65 (s, 1H), 8.69–

(s, 1H), 8.89 ppm (ddd, J=1.0, 2.3, 7.9 Hz, 2H); ¹³C NMR (75 MHz,

CDCl₃, TMS): d=20.9, 125.4, 127.0, 128.9, 129.3, 129.5, 129.7, 129.9,

130.0 131.9, 132.3, 135.1, 135.6 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃,

3842

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

8.76 ppm (m, 2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=108.8, 118.9,

124.3, 125.8, 127.8, 128.7, 129.1, 130.5, 131.1, 131.9, 132.3, 135.1,

141.5 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=275.2 ppm; ele-

mental analysis calcd (%) for C₂₁H₁₃NSe: C 70.40, H 3.66, N 3.91; found:

C 70.38, H 3.69, N 3.89.

Preparation of 1-(p-tolylselanyl)anthraquinone (2d): A similar method to

that desribed for the preparation of 2a was used. Compound 2d was iso-

lated in 76% yield as dark red prisms. M.p. 243.2–244.18C; ¹H NMR

(300 MHz, CDCl₃, TMS): d=2.45 (s, 3H), 7.23–7.32 (m, 3H), 7.41 (t, J=

8.6 Hz, 1H), 7.62 (d, J=8.3 Hz, 2H), 7.81 (dd, J=1.8, 7.2 Hz, 1H), 7.82

(dd, J=1.8, 7.5 Hz, 1H), 8.13 (dd, J=1.3, 7.2 Hz, 1H), 8.30 (dd, J=2.2,

6.8 Hz, 1H), 8.39 ppm (dd, J=2.6, 6.8 Hz, 1H); ¹³C NMR (75 MHz,

CDCl₃, TMS): d=21.4, 124.8, 125.5, 127.0, 127.5, 129.7, 130.8, 132.7,

Preparation of 9-[p-nitrophenylselanyl]anthracene (1j): Under an argon

atmosphere, 9-bromoanthracene (1.00 g, 3.89 mmol) was dissolved in of

dry diethyl ether (50 mL) and the solution was added to a flask that con-

tained magnesium (0.10 g, 4.11 mmol) and dry diethyl ether (5 mL). The

solution was refluxed for 2 h. Selenium powder (0.30 g, 3.80 mmol) was

then added. Then the reaction mixture was refluxed for 2 h. Then p-iodo-

nitrobenzen (0.97 g, 3.90 mmol) and ethanol (50 mL) was added, and the

resultingmixture was refluxed for 2 h. Then the mixture was poured into

ice water. The precipitated solid was flitted and dried in vacuo. The

crude product was purified by column chromatography (SiO₂, hexane) to

give 1j as yellow needles. Yield: 72%, m.p. 150.0–150.88C; ¹H NMR

(300 MHz, CDCl₃, TMS): d=7.09 (d, J=9.1 Hz, 2H), 7.51–7.58 (m, 4H),

7.88 (d, J=9.1 Hz, 2H), 8.08 (dd, J=2.2, 9.7 Hz, 2H), 8.67 (s, 1H),

8.72 ppm (dd, J=2.2, 9.6 Hz, 2H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=

124.0, 124.2, 125.8, 127.8, 128.4, 128.6, 129.2, 131.3, 132.0, 135.1, 144.3,

145.8 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=279.3; elemental

analysis calcd (%) for C₂₀H₁₃NO₂Se: C 63.50, H 3.46, N 3.70; found: C

63.36, H 3.42, N 3.74.

132.8, 133.7, 133.9, 134.3, 134.3, 135.5, 137.4 (²J

(Se,C)=9.9 Hz), 139.7,

G

143.5, 183.0, 183.7 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=

503.4 ppm; elemental analysis calcd (%) for C₂₁H₁₄O₂Se: C 66.85, H 3.74;

found: C 66.82, H 3.61.

Preparation of 1-(p-Fluorophenylselanyl)anthraquinone (2e): A similar

method to that desribed for the preparation of 2a was used. Compound

2e was isolated in 76% yield as orange prisms. M.p. 228.4–229.28C;

¹H NMR (300 MHz, CDCl₃, TMS): d=7.16 (t, J=5.5, 8.8 Hz, 2H), 7.22

(dd, J=1.1, 8.3 Hz, 1H), 7.44 (t, J=7.8 Hz, 1H), 7.72 (dd, J=5.5, 8.8 Hz,

2H), 7.77–7.87 (m, 2H), 8.14 (dd, J=1.3, 7.5 Hz, 1H), 8.30 (dd, J=2.4,

6.8 Hz, 1H), 8.38 ppm (dd, J=2.2, 6.8 Hz, 1H); ¹³C NMR (75 MHz,

CDCl₃, TMS): d=117.3 (²J

125.0, 127.1, 127.5, 129.6, 132.7, 132.9, 133.6, 134.0, 134.4, 135.5, 139.6 (³J-

(F,C)=8.3 Hz, ²J(Se,C)=8.1 Hz), 140.0, 142.9, 163.8 (¹J

(F,C)=249.8 Hz),

182.9, 183.8 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=502.2 ppm

(⁵J

(Se,F)=4.0 Hz); elemental analysis calcd (%) for C₂₀H₁₁FO₂Se: C

A

G

A

N

ACHTREUNG

AHCTREUNG

Preparation of 1-(phenylselanyl)anthraquinone (2a): Sodium hydride

(0.07 g, 3.00 mmol) was added under an argon atmosphere to a solution

of diphenyl diselenide (0.31 g, 1.00 mmol) in DMF (50 mL); the resulting

micxture was then heated to 119.0–120.08C. A solution of 1-chloroan-

thraquinone (0.24 g, 1.00 mmol) in DMF (30 mL) and CuI (1.14 g,

6.00 mmol) were added to the solution and stirringwas continued for 2 h

at 1008C. After pouringinto ice water, the precipitate was filtrated.

After usual workup, the crude product was subjected to chromatography

on silica gel that was covered with a basic alumina layer on the top and

recrystallized from ethanol/chloroform. Compound 2a was isolated in

88% yield as dark red prisms. M.p. 181.2–182.98C; ¹H NMR (300 MHz,

CDCl₃, TMS): d=7.26 (dd, J=1.1, 8.1 Hz, 1H), 7.41 (d, J=7.8 Hz, 1H),

7.43–7.54 (m, 3H), 7.75 (dd, J=1.7, 7.6 Hz, 2H), 7.80 (dd, J=1.8, 7.2 Hz,

1H), 7.82 (dd, J=1.8, 7.3 Hz, 1H), 8.19 (dd, J=1.1, 7.5 Hz, 1H), 8.28

(dd, J=2.2, 7.0 Hz, 1H), 8.38 ppm (dd, J=2.2, 7.0 Hz, 1H); ¹³C NMR

(100 MHz, CDCl₃, TMS): d=124.8, 127.0, 127.4, 129.1, 129.5, 129.6,

129.9, 132.7, 132.8, 133.6, 133.9, 134.2, 135.4, 137.5, 143.0, 182.9,

183.7 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃, MeSeMe): d=512.3 ppm; ele-

mental analysis calcd (%) for C₂₀H₁₂O₂Se: C 66.13, H 3.33, found: C

66.32, H 3.15.

63.01, H 2.91; found: C 62.83, H 2.94.

Preparation of 1-(p-Chlorophenylselanyl)anthraquinone (2 f). A similar

method to that desribed for the preparation of 2a was used. Compound

2 f was isolated in 76% yield as orange prisms. M.p. 247.0–247.98C;

¹H NMR (300 MHz, CDCl₃/TMS): d=7.25 (dd, J=1.1, 8.3 Hz, 1H), 7.44

(d, J=8.4 Hz, 2H), 7.45 (t, J=6.5 Hz, 1H), 7.68 (d, J=8.3 Hz, 2H), 7.77–

7.87 (m, 2H), 8.15 (dd, J=1.1, 7.5 Hz, 1H), 8.30 (dd, J=2.4, 6.8 Hz, 1H),

8.39 ppm (dd, J=2.4, 6.8 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃/TMS): d=

125.1, 127.1, 127.4, 127.5, 129.7, 130.2, 132.8, 133.0, 133.0, 133.6, 134.1,

134.4, 135.5 (²J

A

G

182.9, 183.8 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃/MeSeMe): d=505.3 ppm;

elemental analysis calcd (%) for C₂₀H₁₁ClO₂Se: C 60.40, H 2.79; found:

C 60.42, H 2.70.

Preparation of 1-(p-Bromophenylselanyl)anthraquinone (2g). A similar

method to that desribed for the preparation of 2a was used. Compound

2g was isolated in 76% yield as orange prisms. M.p. 243.9–244.88C;

¹H NMR (300 MHz, CDCl₃/TMS): d=7.25 (dd, J=1.1, 7.9 Hz, 1H), 7.45

(t, J=7.9 Hz, 1H), 7.57–7.63 (m, 4H), 7.78–7.88 (m, 2H), 8.15 (dd, J=

1.3, 7.6 Hz, 1H), 8.30 (dd, J=2.6, 6.6 Hz, 1H), 8.39 ppm (dd, J=2.4,

6.8 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃/TMS): d=124.4, 125.1, 127.1,

127.5, 128.0, 129.7, 132.8, 133.0, 133.2, 133.6, 134.0, 134.0, 134.3, 135.8,

Preparation of 1-[(N,N’-dimethylamino)phenylselanyl]anthraquinone

(2b): A similar method to that desribed for the preparation of 2a was

used. Compound 2b was isolated in 89% yield as dark violet prisms.

M.p. 288.5–289.58C; ¹H NMR (300 MHz, CDCl₃, TMS): d=3.03 (s, 6H),

6.76 (d, J=9.0 Hz, 2H), 7.36–7.40 (m, 2H), 7.54 (d, J=8.8 Hz, 2H), 7.77

(dd, J=1.7 and 7.3 Hz, 1H), 7.78 (dd, J=1.8, 7.3 Hz, 1H), 8.10 (dd, J=

3.0, 5.9 Hz, 1H), 8.27 (dd, J=2.0, 7.0 Hz, 1H), 8.38 ppm (dd, J=2.0,

7.0 Hz, 1H); ¹³C NMR (100 MHz, CDCl₃, TMS): d=40.2, 100.7, 113.6,

114.2, 124.6, 127.0, 127.5, 132.5, 133.0, 133.7, 134.1, 134.2, 134.5, 135.6,

138.6, 145.1, 151.3, 183.2, 183.7 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃,

MeSeMe): d=492.8 ppm; elemental analysis calcd (%) for C₂₂H₁₇NO₂Se:

C 65.03, H 4.22, N 3.45; found: C 65.12, H 4.15, N 3.33.

139.1 (²J(Se,C)=10.2 Hz), 142.3, 182.8, 183.8 ppm; ⁷⁷Se NMR (57 MHz,

N

CDCl₃/MeSeMe): d=505.9 ppm; elemental analysis calcd (%) for

C₂₀H₁₁BrO₂Se: C 54.33, H 2.51; found: C 54.07, H 2.43.

Preparation of 1-[p-(Ethoxycarbonyl)phenylselanyl]anthraquinone (2h):

A similar method to that desribed for the preparation of 2a was used.

Compound 2h was isolated in 66% yield as orange prisms. M.p. 234.0–

234.98C; ¹H NMR (300 MHz, CDCl₃/TMS): d=1.33 (t, J=7.2 Hz, 3H),

4.30 (q, J=7.2 Hz, 2H), 7.21 (dd, J=1.1, 8.1 Hz, 1H), 7.22 (d, J=8.6 Hz,

2H), 7.48 (t, J=7.9 Hz, 1H), 7.79 (d, J=8.6 Hz, 2H),7.82–7.87 (m, 2H),

8.19 (dd, J=1.1, 8.4 Hz, 1H), 8.29–8.34 (m, 1H), 8.36–8.41 ppm (m, 1H);

¹³C NMR (75 MHz, CDCl₃, TMS): d=14.2, 61.2, 125.2, 127.2, 127.5,

128.2, 129.4, 132.7, 133.1, 133.2, 133.4, 134.0, 134.3, 134.5, 135.5, 135.8,

137.9, 141.1, 166.2, 182.7, 183.9 ppm; ⁷⁷Se NMR (57 MHz, CDCl₃/

MeSeMe): d=512.3 ppm; elemental analysis calcd (%) for C₂₃H₁₆O₄Se:

C 63.46, H 3.70; found: C 63.48, H 3.65.

Preparation of 1-(p-anisyl)anthraquinone (2c): A similar method to that

desribed for the preparation of 2a was used. Compound 2c was isolated

in 72% yield as dark red prisms. M.p. 240.6–241.58C; ¹H NMR

(300 MHz, CDCl₃, TMS): d=3.89 (s, 3H), 7.00 (d, J=8.8 Hz, 2H), 7.28

(dd, J=1.1, 8.3 Hz, 1H), 7.43 (t, J=8.3 Hz, 1H), 7.64 (d, J=8.8 Hz, 2H),

7.81 (dd, J=1.8, 7.2 Hz, 1H), 7.83 (dd, J=1.7, 7.2 Hz, 1H), 8.13 (dd, J=

1.3, 7.4 Hz, 1H), 8.30 (dd, J=2.2, 6.8 Hz, 1H), 8.40 ppm (dd, J=2.0,

7.0 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=55.4, 115.7, 119.5,

124.8, 127.0, 127.5, 129.7, 132.7, 132.8, 133.7, 133.9, 134.2, 134.3, 135.5,

Preparation of 1-(p-Cyanophenylselanyl)anthraquinone (2i): A similar

method to that desribed for the preparation of 2a was used. Compound

2i was isolated in 40% yield as orange prisms. M.p. 279.6–280.98C;

¹H NMR (300 MHz, CDCl₃/TMS): d=7.20 (dd, J=1.1, 8.1 Hz, 1H), 7.48

(t, J=7.9 Hz, 1H), 7.74 (d, J=8.4 Hz, 2H), 7.82–7.87 (m, 2H), 7.88 (d,

J=8.4 Hz, 2H), 8.19 (dd, J=1.1, 8.4 Hz, 1H), 8.29–8.34 (m, 1H), 8.36–

8.41 ppm (m, 1H); ¹³C NMR (75 MHz, CDCl₃/TMS): d=113.3, 118.3,

139.0 (²J(Se,C)=10.3 Hz), 143.9, 160.9, 183.1, 183.8 ppm; ⁷⁷Se NMR

A

(57 MHz, CDCl₃, MeSeMe): d=497.3 ppm; elemental analysis calcd (%)

for C₂₁H₁₄O₃Se: C 64.13, H 3.59; found: C 64.32, H 3.65.

125.4, 127.2, 127.5, 129.8 (²J

(Se,C)=10.6 Hz), 132.7, 133.1, 133.2, 133.4,

U

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3843

W. Nakanishi et al.

134.0, 134.3, 134.5, 135.5, 135.8, 137.9, 141.1, 182.7, 183.9 ppm; ⁷⁷Se NMR

(57 MHz, CDCl₃/MeSeMe): d=520.5 ppm; elemental analysis calcd (%)

for C₂₁H₁₁NO₂Se: C 64.96, H 2.86, N 3.61; found: C 64.67, H 2.95, N 3.63.

c) Organic Selenium Chemistry (Ed.: D. Liotta), Wiley-Interscience,

New York, 1987; d) Organoselenium Chemistry, A practical Ap-

proach (Ed.: T. G. Back), Oxford University Press, Oxford, 1999;

e) “Organoselenium Chemistry: Modern Developments in Organic

Synthesis”: Top. Curr. Chem. 2000, 2 08, whole volume.

Preparation of 1-(p-Nitrophenylselanyl)anthraquinone (2j). A similar

method to that desribed for the preparation of 2a was used. Compound

2j was isolated in 70% yield as orange prisms. M.p. 298.0–299.08C;

¹H NMR (300 MHz, CDCl₃/TMS): d=7.23 (dd, J=1.1, 8.6 Hz, 1H), 7.48

(t, J=7.9 Hz, 1H), 7.85 (m, 2H), 7.94 (d, J=8.6 Hz, 2H), 8.20 (dd, J=

1.1, 7.7 Hz, 1H), 8.30 (d, J=8.6 Hz, 2H), 8.32 (dd, J=1.1, 7.4 Hz, 1H),

8.39 ppm (dd, J=1.1, 7.4 Hz, 1H); ¹³C NMR (75 MHz, CDCl₃, TMS): d=

123.6, 125.2, 127.1, 127.5, 129.4, 133.1, 133.2, 133.4, 134.0, 134.3, 134.5,

135.5, 135.8, 137.9, 143.8, 146.1, 182.7, 183.9 ppm; ⁷⁷Se NMR (57 MHz,

CDCl₃/MeSeMe): d=514.8 ppm; elemental analysis calcd (%) for

C₂₀H₁₁NO₄Se: C 58.84, H 2.72, N 3.43; found: C 58.99, H 2.89, N 3.61.

[2] a) W. MacFarlane, R. J. Wood, J. Chem. Soc. Dalton Trans. 1972, 13,

1397–1401; b) H. Iwamura, W. Nakanishi, Yuki Gosei Kagaku Kyo-

kaishi (J. Synth. Org. Chem. Jpn.) 1981, 39, 795–804; c) The Chemis-

try of Organic Selenium and Tellurium Compounds, Vol. 1 (Eds.: S.

Patai, Z. Rappoport), Wiley, New York, 1986, Chapter 6; d) Compi-

lation of Reported ⁷⁷Se NMR Chemical Shifts (Eds.: T. M. Klapotke,

M. Broschag), Wiley, New York, 1996; e) H. Duddeck, Prog. Nucl.

Magn. Reson. Spectrosc. 1995, 27, 1–323.

[3] a) S. Gronowitz, A. Konar, A.-B. Hçrnfeldt, Org. Magn. Reson.

1977, 9, 213–217; b) G. P. Mullen, N. P. Luthra, R. B. Dunlap, J. D.

Odom, J. Org. Chem. 1985, 50, 811–816; c) G. A. Kalabin, D. F.

Kushnarev, V. M. Bzesovsky, G. A. Tschmutova, Org. Magn. Reson.

1979, 12, 598–604; d) G. A. Kalabin, D. F. Kushnarev, T. G. Manna-

fov, Zh. Org. Khim. 1980, 16, 505–512; e) W. Nakanishi, S. Hayashi,

T. Uehara, Eur. J. Org. Chem. 2001, 2001, 3933–3943.

[4] a) S. Hayashi, W. Nakanishi, J. Org. Chem. 1999, 64, 6688–6696;

b) W. Nakanishi, S. Hayashi, H. Yamaguchi, Chem. Lett. 1996, 947–

948; c) W. Nakanishi, S. Hayashi, A. Sakaue, G. Ono, Y. Kawada, J.

Am. Chem. Soc. 1998, 120, 3635–3640; d) W. Nakanishi, S. Hayashi,

J. Org. Chem. 2002, 67, 38–48.

[5] a) W. Nakanishi, S. Hayashi, Chem. Lett. 1998, 523–524; b) W. Na-

kanishi, S. Hayashi, J. Phys. Chem. A 1999, 103, 6074–6081.

[6] D. H. R. Barton, M. B. Hall, Z. Lin, S. I. Parekh, J. Reibenspies, J.

Am. Chem. Soc. 1993, 115, 5056–5059.

[7] a) Encyclopedia of Nuclear Magnetic Resonance (Eds.: D. M. Grant,

R. K. Harris), Wiley, New York, 1996; b) Nuclear Magnetic Shield-

ings and Molecular Structure (Ed.: J. A. Tossell), Kluwer Academic,

Dordrecht, Boston, London, 1993; c) Calculation of NMR and EPR

Parameters; Theory and Applications (Eds.: M. Kaupp, M. Bühl,

V. G. Malkin), Wiley-VCH, Weinheim, 2004; d) Spins in Chemistry

(Ed.: R. McWeeny), Academic Press, New York, 1970; e) “A Tool

for Chemistry”: V. G. Markin, O. L. Malkina, L. A. Eriksson, in

Modern Density Functional Theory (Eds.: J. M. Seminario, P. Polit-

zer), Elsever, Amsterdam, 1994; f) Density Functional Methods in

Chemistry and Material Science (Ed.: M. Springborg), Wiley, New

York, 1977.

[8] a) R. Fukuda, M. Hada, H. Nakatsuji, J. Chem. Phys. 2003, 118,

1015–1026; b) R. Fukuda, M. Hada, H. Nakatsuji, J. Chem. Phys.

2003, 118, 1027–1035; c) S. Tanaka, M. Sugimoto, H. Takashima, M.

Hada, H. Nakatsuji, Bull. Chem. Soc. Jpn. 1996, 69, 953–959;

d) C. C. Ballard, M. Hada, H. Kaneko, H. Nakatsuji, Chem. Phys.

Lett. 1996, 254, 170–178; e) H. Nakatsuji, M. Hada, H. Kaneko,

C. C. Ballard, Chem. Phys. Lett. 1996, 255, 195–202; f) M. Hada, H.

Kaneko, H. Nakatsuji, Chem. Phys. Lett. 1996, 261, 7–12.

[9] a) K. Kanda, H. Nakatsuji, T. Yonezawa, J. Am. Chem. Soc. 1984,

106, 5888–5892; b) Molecular Quantum Mechanics, 3rd ed. (Eds.:

P. W. Atkins, R. S. Friedman), Oxford University Press, Oxford,

New York, 1997, Chapter 13.

[10] a) N. F. Ramsey, Phys. Rev. 1949, 77, 567; b) N. F. Ramsey, Phys.

Rev. 1950, 78, 699–703; c) N. F. Ramsey, Phys. Rev. 1951, 83, 540–

541; d) N. F. Ramsey, Phys. Rev. 1956, 86, 243–246; e) N. F. Ramsey,

Phys. Rev. 1952, 85, 143–144; f) N. F. Ramsey, Phys. Rev. 1953, 91,

303–307.

X-ray structural determination of 1c and 2a: The yellow crystals of 1c

and the orange crystals of 2a were grown by slow evaporation of a solu-

tion of the sample in dichloromethane/benzene or benzene/hexane sol-

vent mixtures at room temperature. The intensity data were collected on

a Rigaku AFC5R four-circle diffractometer with graphite-monochromat-

ed Mo_Karadiation (l=0.71069 ) for 1c and 2a. The structures of 1c

and 2a were solved by heavy-atom Patterson methods, PATTY,^[45]and

[46]

expanded by usingFourier techniques, DIRDIF94.

All the non-hydro-

gen atoms were refined anisotropically. Hydrogen atoms were included

but not refined. The final cycle of full-matrix least-squares refinement

was based on a total of 4232 reflections for 1c and on 2560 for 2a with

415 and 212 [I>1.50s(I)] observed reflections for 1c and 2a, respectively.

Variable parameters and converged with unweighted and weighted agree-

ment factors of R=(ꢀjF_ojꢀjF_cjj)/ꢀjF_oj and R_w={ꢀw(jF_ojꢀjF_cj)²/

ꢀwF²_o}^1/2were used. For least squares, the function minimized was

2

ꢀw(jF_ojꢀjF_cj)², in which w=(s₂^cjF_oj+p²jF_oj /4)^ꢀ1. Crystallographic de-

tails are given in the Supporting Information. CCDC-283618 and CCDC-

283619 contain the supplementary crystallographic data for this paper.

These data can be obtained free of charge from The Cambridge Crystal-

lographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

MO Calculations: Quantum chemical (QC) calculations were performed

by usinga Silent-SCC T2 (Itanium2) computer with the 6–311 +G

A

basis sets for Se and 6–311+G(3d,2p) for other nuclei of the Gaussian 03

AHCTREUNG

program.^[29]Calculations were performed on 4–6 in pl and pd confoma-

tions at the density functional theory (DFT) level of the Becke three-pa-

rameter hybrid functionals combined with the Lee-Yang-Parr correlation

functional (B3LYP). Absolute magnetic shielding tensors of Se nuclei

(s(Se)) were calculated based on the gauge-independent atomic orbital

(GIAO) method, applyingon the optimized structures with the same

method. A utility program (NMRANAL-NH98G) was prepared to carry

out decomposing the magnetic shieldings, based on the Gaussian 98.^[15]

The program was applied to SeH₂, 4(pl), 4(pd) (Y=H, NH₂, Me, F, CN,

and NO₂) 5 (pl), 5(pd), and 6a (Y=H) to evaluate the contributions sep-

arately from each molecular orbital (y_i) and each y_i!y_jtransition, in

which y_iand y_jdenote occupied and unoccupied molecular orbitals, re-

spectively.

Structures of 1a–3a in various conformers were also optimized with the

B3LYP/6–311+G(d,p) method. The frequency analysis was also per-

R

formed.

Acknowledgements

[11] W. Nakanishi, S. Hayashi, T. Uehara, J. Phys. Chem. A 1999, 103,

9906–9912.

This work was partially supported by a Grant-in-Aid for Scientific Re-

search (No. 16550038) from the Ministry of Education, Culture, Sports,

Science, and Technology, Japan.

[12] The nonplanar and nonperpendicular conformer (np) is also impor-

tant in some cases, for example, CC in 1-MeSe-8-PhSeC₁₀H₆.^[11]

[13] The importance of relative conformations in the substituent effects

between substituents and probe sites is pointed out. See for exam-

ple, “Angular Dependence of Dipolar Substituent Effects”: K. Bord-

wen, E. J. Grubbs, Prog. Phys. Org. Chem. 1993, 19, 183–224, and

references therein.

[1] a) Organic Selenium Compounds: Their Chemistry and Biology

(Eds.: D. L. Klayman, W. H. H. Günther), Wiley, New York, 1973;

b) The Chemistry of Organic Selenium and Tellurium Compounds,

Vols. 1 and 2 (Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1986;

[14] a) K. Wolinski, J. F. Hinton, P. Pulay, J. Am. Chem. Soc. 1990, 112,

8251–8260; b) K. Wolinski, A. Sadlej, Mol. Phys. 1980, 41, 1419–

3844

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

NMR Analysis

FULL PAPER

1430; c) R. Ditchfield, Mol. Phys. 1974, 27, 789–807; d) R. McWee-

ny, Phys. Rev. 1962, 126, 1028–1034; e) F. London, J. Phys. Radium

1937, 8, 397–409.

ter 8; f) Chemistry of Hypervalent Compounds (Ed.: K.-y. Akiba),

Wiley-VCH, Weinhaim, 1999.

ꢀ

[23] If two nonbonded n_p(O)!s*

A

operate at the both sides of the n_p(O), extended hypervalent 5c-6e

interactions are constructed. See, W. Nakanishi, S. Hayashi, N. Itoh,

Chem. Commun. 2003, 2003, 124–125; W. Nakanishi, S. Hayashi, N.

Itoh, J. Org. Chem. 2004, 69, 1676–1684; W. Nakanishi, S. Hayashi,

T. Furuta, N. Itoh, Y. Nishina, M. Yamashita, Y. Yamamoto, Phos-

phorus Sulfur Silicon Relat. Elem. 2005, 180, 1351–1355.

[15] Gaussian 98 (Revision A.11), M. J. Frisch, G. W. Trucks, H. B. Schle-

gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,

J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich,

J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.

Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli,

C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q.

Cui, K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick,

A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V.

Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,

I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A.

Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W.

Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez,

M. Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian, Pittsburgh

PA, 2001.

[24] Details of the structure of 2 will be discussed with respect to the

ꢀ

O···Se C 3c-4e interaction elsewhere.

[25] The 1

A

U

tallographic analysis and QC calculations. These structures in crys-

tals and in the gas phase will also exist in solution. The equilibrium

with other conformers would not to be so severe in the solution for

1

2

ACHTREUNG

AHCTREUNG

d(Se) values support the expectation.

[16] While the torsional angle of C(1)-Se(1)-C(15)-C(16) in 1c

ACHTREUNG

[26] The 0.050m CDCl₃solutions were used for NMR measurements.

However, the concentrations would be lower for the compounds of

low solubility, such as 1j and 2j at 213 K.

[27] The temperature dependence of d(Se) of MeSeMe in [D]chloroform

(60% v/v) is reported to be about 2.5 ppm over the temperature

range of 222–323 K. See N. P. Luthra, R. B. Dunlap, J. D. Odom, J.

Magn. Reson. 1983, 52, 318–322.

174.8(4)8, that of C(22)-Se(2)-C(36)-C(37) in 1c

ACHTREUNG

ꢀ155.2(5)8.^[17]The latter must be deformed by the crystal packing

effect. The main intermolecular short contacts around p-An in 1c(S-

A

A) (O···H_mand H_m···O) occur in the directions on the p-An plane

with one of out of plane (H_Me···H_o). As the result, the structure of

1c

A

G

N

(C_m···H_Me, H_m···O, and H_o···H₂) occur mainly out of the plane direc-

tion with one of in plane (H_o’···H_Me), which deforms the torsional

[28] When the axes from the Gaussian 03^[29]calulations were not the

same as those shown in Scheme 3, they were interchanged so as to

be those in Scheme 3 for convenience of discussion, if possible.

[29] Gaussian 03 (Revision B.05), M. J. Frisch, G. W. Trucks, H. B. Schle-

gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-

y, Jr. T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-

gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.

Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.

Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,

O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,

J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,

O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.

Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,

V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.

Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-

man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,

B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.

Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-

nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,

M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Pittsburgh

PA, 2003.

[30] The torsional angle of f=C_oC_iSeH in 4(pd) is fixed at 90.08, if

4(pd) is not the C_ssymmetry (e.g., Y=COOMe and OMe). Similar-

ly, those of f=C_oC_iSeC_Meand f=C_iSeC_MeH in 5(pd) are fixed at

90.0 and 1808, respectively, and those of f=C_oC_iSeC_i’and f=

C_iSeC_i’C_o’in 6(pd) are at 90.0 and 08, respectively, when p-YC₆H₄Se

is not the C_ssymmetry.

[31] MolStudio R3.2 Rev 1.0, NEC Corporation, 1977–2003.

[32] Evaluated values of s^p(Se) and the components by the Gaussian 98

are sometimes substantially different from those by the Gaussian 03,

although the structures are the same. The differences in s^p(Se)_xxand

s^p(Se)_yyamount to 50 ppm or less for 4a(pd) and 5a(pd). They are

very large for s^p(Se)_xx(ꢀ232.8 ppm) and s^p(Se)_yy(212.7 ppm) in 6a.

Axes of 6a were bisected in Gaussian 98. Axes were chosen to be

more close to those in 6(pl).

angle from that in pure 1c(A:pl) by about 258 in the crystal.

A

[17] The crystallographic data of 1c and 2a are collected in Table S1 of

the SupportingInformation and their selected interatomic distances,

angles, and torsional angles in Table S2.

[18] The energies of structures 3a

evaluated to be ꢀ3018.6354, ꢀ3018.6350, and ꢀ3018.6262 au, respec-

tively, by the B3LYP/6–311+G

(d,p) method.^[29]The relative ener-

A

(B:pd), and 3a

A

AHCTREUNG

gies are 0.0, 1.1, and 24.2 kJmol^ꢀ1, respectively. Structure 3a

(B’:pd)

G

has an imaginary frequency (n₁=ꢀ38.9 cm^ꢀ1), which corresponds to

the motion to 3a

N

transition state. The relative energies for the sum of electronic and

thermal (Gibbs) free energies at 298 K of 3a(A:pl), 3a(B:pd), and

3a

(B’:pd) are 0.0, 0.0, and 29.1 kJmol^ꢀ1, respectively.

[19] The energies of structures 1a(A:pl) and 1a(A:pd) were evaluated to

be ꢀ3172.3034 and ꢀ3172.3005 au, respectively, by the B3LYP/6–

311+G (A:pl) is predicted to be more

(d,p) method.^[29]Structure 1a

stable than 1a

(A:pd) by 7.6 kJmol^ꢀ1on the energy surface. While

all frequencies of 1a

A

ACHTREUNG

A

ACHTREUNG

A

ACHTREUNG

A

N

one imaginary frequency of n₁=ꢀ20.6 cm^ꢀ1. Structure 1a

C

(Gibbs) free energies at 298 K of 1a

ꢀ3172.0769 and ꢀ3172.0717 au, respectively. The difference is

13.7 kJmol^ꢀ1

[20] The p–p conjugation of the n_p(Se)–p(Ar) type plays an additional

role to affect the stability of 1(A:pl); cases in which Y is an acceptor

(A:pl) more effectively than cases in which Y is a donor

(A:pl) versus

(B:pd)).^[3e]Since Y=OMe is one of

A

G

.

AHCTREUNG

stabilize 1

ACHTREUNG

(compare

3

A

3

ACHTREUNG

strongdonors, the structure of 1c

minimum of (A:pl) for 1.

A

[21] The evaluated energies of 2a

and ꢀ3321.6015 au, respectively, by the B3LYP/6–311+G

method.^[29]Structure 2a

(B:pd) is more stable than 2a(A:pl) by

31.8 kJmol^ꢀ1. Structure 2a (A:np),^[12]in which

(A:pl) is close to 2a

the Ph group is close to the O=C group.

A

ACHTREUNG

AHCTREUNG

A

ACHTREUNG

A

ACHTREUNG

[33] Chemical shifts have been discussed with parallel and perpendicular

components. See reference [7].

[22] For hypervalent bonds, see, a) G. C. Pimentel, J. Chem. Phys. 1951,

19, 446–448 b) J. I. Musher, Angew. Chem. 1969, 81, 68–83; Angew.

Chem. Int. Ed. Engl. 1969, 8, 54–68; c) M. M. L. Chen, R. Hoff-

mann, J. Am. Chem. Soc. 1976, 98, 1647–1653; d) P. A. Cahill, C. E.

Dykstra, J. C. Martin, J. Am. Chem. Soc. 1985, 107, 6359–6362;

e) R. A. Hayes, J. C. Martin, Sulfurane Chemistry, in Organic Sulfur

Chemistry: Theoretical and Experimental Advances (Eds.: F. Bernar-

di, I. G. Csizmadia, A. Mangini), Elsevier, Amsterdam, 1985, Chap-

[34] s^p(Se)_xx, s^p(Se)_yy, s^p(Se)_zz, and s^p(Se) of y₁₃are 0.1, ꢀ6.2, ꢀ0.2, and

ꢀ2.1 ppm, respectively.

[35] The values are summed over y_iof the A’’ symmetry for 4a–6a of pl

and pd, which are constructed by p_z(Se) and 3d(Se).

[36] Whereas s^p(Se)_zz(zi)of 4a(pd) is almost zero, that of 4a(pl) is

ꢀ11 ppm. The non-zero nature of s^p(Se)_zz(zi)seems common in 4(pl)

(6–10 ppm), although not shown.

Chem. Eur. J. 2006, 12, 3829 – 3846

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

www.chemeurj.org

3845

W. Nakanishi et al.

[37] Calculations of s(Se) for H₂Se, see, a) W. Kutzelnigg, U. Fleischer,

C. van Wüllen, Encyclopedia of Nuclear Magnetic Resonance, Vol. 7

(Eds.: D. M. Grant, R. K. Harris), Wiley, New York, 1996, pp. 4284–

4291; b) P. D. Ellis, J. D. Odom, A. S. Lipton, Q. Chen, J. M. Gulick,

Nuclear Magnetic Shieldings and Molecular Structure (Ed.: J. A. Tos-

sell), Kluwer Academic, Dordrecht, Boston, London, 1993, pp. 539–

555; c) P. J. Wilson, Molecular Physics 2001, 99, 363–367. Those for

4(pl) and 4(pd), see reference [5b].

[38] Details of the analysis of d(Se) in SeH₂and MeSeH will be reported

elsewhere.

[39] DFT shieldings are deshielded in general, due to the underestima-

tion of orbital energy differences, which leads to the overestimation

of s^p(Se).^[37c]Therefore, s^t(Se) of 4a(pl), 4a(pd), 5a(pl), and 5a(pd)

[42] A similar trend is observed in 5(pl) as in the case of 4(pl), although

the direction is reversed.

[43] s^p(Se)_zzis almost constant in the change of Y for both pl and pd of

4–6. The small Y dependence of s^p(Se)_zzis reasonably explained by

the main interaction of the 4p_z(Se)–p

which 4p_x(Se) and 4p_y(Se) do not take part in the interaction direct-

ly. The main interaction in pd is the s(C_ArSeX)–p(C₆H₄)–p_x(Y) (X=

(C₆H₄)–p_z(Y) type in pl, in

A

ACHTREUNG

H or C) type, which modifies the contribution of 4p_x(Se) and

4p_y(Se) in the C_ArSeX moiety. However, the results show that the in-

teraction in pd affect the s^p(Se)_xxand s^p(Se)_yyparameters, but not

the s^p(Se)_zz.

[44] A very wide range of s^p(Se) for 4(pd) must be responsible for the

results: The s^p(Se) values for 4(pd) were evaluated to be ꢀ2827,

ꢀ1435, and ꢀ16155 ppm for Y=O⁺, S⁺, and Se⁺, respectively. The

correspondingvalues for 4(pl) are ꢀ1976, ꢀ1841, and ꢀ1733 ppm,

respectively.

were evaluated with the MP2/6-31+G

GIAO method on the structures optimized with the MP2/6-31+G-

(3d,2p) method. The values are 1827.3, 1865.5, 1761.0, and

A

ACHTREUNG

1708.7 ppm, respectively: s^tof 4a(pd) is predicted to be larger than

that of 4a(pl) by 38 ppm, whereas s^t(Se) of 5a(pd) is smaller than

that of 5a(pl) by 46 ppm. The results supports the orientational ef-

fects evaluated at the DFT level for 4a and 5a, although the basis

sets are not the same.

[45] P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, S.

Garcia-Granda, R. O. Gould, J. M. M. Smits, C. Smykalla, The

DIRDIF program system, Technical Report of the Crystallography

Laboratory, University of Nijmegen, The Netherlands, 1992.

[46] P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, R. de

Gelder, R. Israel, J. M. M. Smits, The DIRDIF-94 program system,

Technical Report of the Crystallography Laboratory, University of

Nijmegen, The Netherlands, 1994.

[40] Interactions between n_p(Se) of 4p_z(Se) and phenyl s orbitals in

4a(pd) must be weak due to large energy differences between

4p_z(Se) and the s orbitals. Longdistances between them are also

disadvantageous.

ꢀ

[41] The results may show that the 4p_y(Se) orbital involved in the Se C

bond works effectively to shift downfield, whereas the 4p_x(Se) orbi-

Received: August 1, 2005

Revised: December 12, 2005

Published online: March 3, 2006

ꢀ

tal involved in the Se H bond causes an upfield shift, when Y=H

changes to non-H in 4(pl).

3846

www.chemeurj.org

ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2006, 12, 3829 – 3846

Article Doi

Orientational effect of aryl groups on77Se NMR chemical shifts: Experimental and theoretical investigations

DOI: 10.1002/chem.200500927

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.200500927

Full text of DOI:10.1002/chem.200500927

Products guided by the article

R&D Labs maybe for 128973-26-2

Relevant to this article

Hot Product