Stereoisomeric 2-butylphenylsulfoxides and their binding modes in the adduct formation with an enantiopure dirhodium tetracarboxylate complex

Article abstract of DOI:10.1016/j.molstruc.2010.01.060

All stereoisomers of the 2-butylphenylsulfoxides 1a and 2a and their p-substituted derivatives 1b-1eand 2a-2e (X = F, Br, NO₂, and OCH₃) were synthesized. Absolute configurations were derived from commercial enantiopure 2-butanols used as starting materials, by X-ray diffraction and by polarimetry. Preferred conformations were determined by density functional and second-order M?ller-Plesset calculations. Oxygen atoms dominate in the adduct formation equilibria of 2-butylsubstituted sulfoxides and the chiral dirhodium complex Rh^* although the sulfur atom is, in principle, the stronger donor. This is due to steric shielding of the sulfur atom produced by the aromatic ring and the secondary 2-butyl substituent. Enantiodifferentiation of sulfoxides is easily accomplished by the dirhodium experiment, i.e. recording NMR spectra in the presence of an equimolar amount of Rh^*. Complex formation shifts (Δδ) and diastereomeric dispersion effects (Δδ) differ in the dirhodium experiment for nonracemic mixtures of sulfoxides as compared to pure enantiomers. This, however, does not affect the efficiency of the dirhodium experiment at all.

Full text of DOI:10.1016/j.molstruc.2010.01.060

Journal of Molecular Structure 978 (2010) 86–96
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Stereoisomeric 2-butylphenylsulfoxides and their binding modes in the adduct
formation with an enantiopure dirhodium tetracarboxylate complex
*
Jens T. Mattiza, Vera J. Meyer, Helmut Duddeck
Institute of Organic Chemistry, Leibniz University Hannover, Schneiderberg 1B, D-30167 Hannover, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
All stereoisomers of the 2-butylphenylsulfoxides 1a and 2a and their p-substituted derivatives 1b–1e and
2a–2e (X = F, Br, NO2, and OCH3) were synthesized. Absolute configurations were derived from commercial
enantiopure 2-butanols used as starting materials, by X-ray diffraction and by polarimetry. Preferred con-
formations were determined by density functional and second-order Møller–Plesset calculations. Oxygen
atoms dominate in the adduct formation equilibria of 2-butylsubstituted sulfoxides and the chiral dirho-
dium complex Rh* although the sulfur atom is, in principle, the stronger donor. This is due to steric shielding
of the sulfur atom produced by the aromatic ring and the secondary 2-butyl substituent. Enantiodifferenti-
ation of sulfoxides is easily accomplished by the dirhodium experiment, i.e. recording NMR spectra in the
Received 16 November 2009
Received in revised form 21 January 2010
Accepted 21 January 2010
Available online 1 February 2010
This paper is dedicated to Prof. Dr. Heinz
Oberhammer, Eberhard Karls Universität
Tübingen, Tübingen, Germany, on the
occasion of his 70th birthday.
presence of an equimolar amount of Rh*. Complex formation shifts (
effects ( ) differ in the dirhodium experiment for nonracemic mixtures of sulfoxides as compared to pure
enantiomers. This, however, does not affect the efficiency of the dirhodium experiment at all.
Ó 2010 Elsevier B.V. All rights reserved.
Dd) and diastereomeric dispersion
Dm
Keywords:
Chiral sulfoxides
Dirhodium complex
Conformational analysis
1H and 13C NMR
Adduct formation
Chiral differentiation
1. Introduction
ergy is based primarily on electrostatic interaction [9]. As expected,
we found that the inductive effect of the oxygen on the aliphatic
Dirhodium complexes and their adducts have been the focus of
interest for many years [1]. They were introduced as homogeneous
catalysts in various reactions [2] and found even medicinal applica-
tion [3]. During the last decade, we have shown that the enantiomers
of many chiral ligands, particularly those of soft Lewis bases, can be
differentiated easily by adding an equimolar amount of the dirho-
dium complex Rhð2IIÞ [(R)-(+)-MTPA]4 (Rh*, MTPA-H = methoxytriflu-
oromethylphenylacetic acid ꢀ Mosher’s acid; see Scheme 1) [4] to
their CDCl3 solution and monitoring the diastereomeric dispersion
a
-carbons is enhanced when it is complexed to the rho atom
d > 0). However, deshielding complexation shifts d at the aro-
matic ipso-carbons (also -positioned) turned out to be minute
(D
D
a
whereas ortho- and para-carbon signals are influenced significantly
by the resonance effect of oxygen and its interaction with substitu-
ents in para-position [8].
In order to gain further insight into the complexation mecha-
nisms of chalcogen ligands, we extended this study to structurally
analogous thioethers 4 (Scheme 2) [10]. Whereas ethers are hard li-
gands, thioethers are soft [11] and represent a different ligand cate-
gory in the dirhodium experiment as compared to ethers [5]; i.e.,
thioethers can make use of an additional orbital interaction
(HOMO–LUMO) for complexation. Indeed, this can be monitored
Dm
of their 1H or 13C NMR signals at room-temperature (dirhodium
method) [5–7].
The complexation site of ligand molecules L in dirhodium com-
plex adducts (Rh* L) can be identified by moderate deshieldings
of nearby 1H and – particularly – 13C nuclei (complexation shifts
by inspecting the dependence of complexation shifts at the aliphatic
a
0
´
D
d). In a qualitative interpretation of positive
D
d values, one can as-
carbon (C-2) and the ortho-carbon (C-2 ) on Hammetts inductive
sume that an adduct formation induces an increase of the electron-
acceptor properties of the binding atom [5]. Recently, we investi-
gated some ethers 3 (Scheme 2) with oxygen binding sites attached
to aromatic and aliphatic ring systems [8]. In ethers, the binding en-
and resonance parameters, respectively [10]. For each carbon site
and each parameter the correlations are opposite as compared to
those of ethers, and this can be rationalized semi-quantitatively by
effects of HOMO–LUMO energy changes on 13C chemical shifts [10].
Then, we extended these studies to sulfoxides offering both
types of donors simultaneously, soft sulfur and hard oxygen atoms;
the results are presented here.
* Corresponding author. Tel.: +49 511 762 4615; fax: +49 511 762 4616.
E-mail address: duddeck@mbox.oci.uni-hannover.de (H. Duddeck).
0022-2860/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2010.01.060

J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
87
and the mixture extracted three times with 5 ml chloroform each.
The combined organic phases were dried over MgSO4 and the sol-
vent evaporated in vacuo. The residue was purified by column
chromatography at normal pressure and temperature using silica
gel as stationary and mixtures of petrol ether and ethyl acetate
as mobile phase (see Section 2). Yields are given below. Synthetic
procedures were not optimized. p-Nitrophenyl-methylsulfoxide
(5) was prepared analogously from the respective commercially
available thioether [13].
The parent sulfoxides 1a/2a has been described in literature
[14]; all others are new. We collected the 1H and 13C NMR data
of all enantiopure 2-butylphenylsulfoxides in Table 2; yields, phys-
ical data, infra-red (IR) and mass spectral data (ESI-MS) are listed
below. IR spectra were recorded on a Bruker Vector 22 using neat
samples and positive ESI-MS spectra on a Micromass LCT.
Scheme 1. Structure of the dirhodium complex Rh*.
2. Experiment
2.1. Materials and synthesis
2.1.1. (RS,RC)- or (SS,SC)-(1-methylpropylthio)benzene-S-oxide (1a)ꢁ1
~
m
Yield: 72%, viscous, slightly yellowish liquid. IR (liquid)
(cm
)
The syntheses of Rh* [4], the ethers 3 [8] and the thioethers 4
[10] have been described by us earlier. The sulfoxides 1a–1e and
2a–2e were prepared by H2O2 oxidation of the thioethers 4a–4e,
respectively [12]. General procedure: the respective 2-buty-
lphenylthioether (2.2 mmol) was dissolved in 5 ml methanol. A
solution of H2SO4 (16%) in tert-butanol and 35% H2O2 (0.18 g,
5.1 mmol) was added dropwise. Then, the mixture was stirred for
24 h at room-temperature. Aqueous NaCl solution (5%) was added
3046, 2966, 2932, 1583, 1443, 1381, 1035, 748, 692. ESI-MS calcu-
lated for C10H15OS: 183.0844 [M+H]+, found: 183.0840.
2.1.2. (RS,SC)- or (SS,RC)-(1-methylpropylthio)benzene-S-oxide (2a)ꢁ1
~
m
Yield: 73%, viscous, slightly yellowish liquid. IR (liquid)
(cm
)
3040, 2963, 2932, 1581, 1441, 1384, 1031, 748, 691. ESI-MS calcu-
lated for C10H15OS: 183.0844 [M+H]+, found: 183.0843.
Scheme 2. Structures of stereoisomeric 2-butylphenylsulfoxides 1 (like) and 2 (unlike); structurally related ethers 3 and thioethers 4 for comparison. Although the atom
numbering is not in total agreement with the IUPAC nomenclature, it has been chosen to for a better comparability of NMR data. Correct IUPAC names are given in the
Experimental Part.

88
J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
Table 1
Crystallographic data and structure refinement for (RS,SC)-2d (ul) and selected bond lengths (in pm), bond angles (in °) and torsion angles (in °).a
(RS,SC)-2d
10H13NO3S
227.27 g/mol
297 K
Selected parameters
Chemical formula
Formula weight
Temperature
Crystal system, space group
Unit cell parameters
a = 9.049(3)
b = 10.448(4)
c = 12.547(8)
Cell volume
Z
Calculated density
Absorption coefficient
Crystal color and size
Reflections collected
Independent reflections
C
S–O
1.496(3)
1.798(5)
1.779(4)
107.7(2)
106.4(7)
113.5(4)
107.1(4)
119.8(3)
120.9(3)
ꢁ51.94
S–C2
S–C10
Orthorhombic, P212121
O–S–C2
O–S–C10
a
= b =
c
= 90°
C1–C2–S
C3–C2–S
C20–C10 –S
C60–C10 –S
O–S–C2–C1
O–S–C2–C3
O–S–C10 –C20
C2–S–C10–C20
C3–C2–S–C10
O–N–C40 –C30
1186.2(10)
4
1.273 g/cm3
+71.56
0.260 mmꢁ1
ꢁ9.68
White, 0.31 ꢂ 0.29 ꢂ 0.11 mm
ꢁ121.11
ꢁ178.53
+167.80
16,664
2329
Reflections with I > 2
r(I)
1358
Minimum and maximum transmission
hmin = 2.5°, hmax = 26.1°
R1 = 0.0501, wR2 = 0.1125
R1 = 0.0879, wR2 = 0.1280
1.005
Final R indices [I > 2
R indices (all data)
r(I)]
Goodness-of-fit on F2
Largest difference map peak and hole
Flack parameter
0.372 eÅꢁ3, ꢁ0.202 eÅꢁ3
ꢁ0.01(19)
a
The geometric parameters of the enantiomeric (SS,RC)-2d are the same within the experimental error limits; of course, torsion angles have opposite signs.
Table 2
1H and 13C chemical shifts (d) of the sulfoxides 1 and 2 (Scheme 2); in ppm, solvent: CDCl3, relative to internal tetramethylsilane d = 0 ppm.a,b
1a/2a
X = H
(l)/(u)
1b/2bc
X = F
(l)/(u)
1c/2c
X = Br
(l)/(u)
1d/2d
X = NO2
(l)/(u)
1e/2e
X = OCH3
(l)/(u)
H-1
H-2
H-3d
1.09/1.06
2.72/2.55
a: 1.46/1.48
b: 1.81/1.94
1.01/1.09
7.62/7.57
7.51/7.50
7.51/7.52
–
11.8/10.2
61.0/61.1
22.0/23.8
10.9/11.6
141.5/141.7
125.3/124.7
128.8/128.9
131.0/130.7
–
1.07/1.05
2.70/2.52
a: 1.45/1.45
b: 1.80/1.92
1.02/1.09
7.62/7.56
7.22/7.21
–
1.11/1.06
2.71/2.51
a: 1.45/1.46
b: 1.77/1.94
1.01/1.11
7.49/7.44
7.66/7.65
–
1.19/1.03
2.78/2.60
a: 1.48/1.57
b: 1.72/2.01
1.01/1.15
7.80/7.76
8.38/8.38
–
1.06/1.03
2.70/2.67
a: 1.43/1.45
b: 1.86/1.84
1.04/1.05
7.54/7.55
7.01/7.02
–
3.86/3.86
11.6/10.5
61.0/61.1
22.3/23.3
10.9/11.5
132.3/132.8
127.2/126.5
114.4/114.4
162.0/161.7
55.4/55.5
H-4
H-20/H-60
H-30/H-50
H-40
OCH3
C-1
C-2
C-3
C-4
C-10
C-20/C-60
C-30/C-50
C-40
–
–
–
11.7 /10.1
61.1/61.2
22.1/23.5
10.9/11.5
136.9/137.4
127.6/126.8
116.2/116.2
164.4/164.1
–
11.8/10.1
61.1/61.3
22.0/23.8
11.0/11.6
140.6/141.3
126.9/126.3
132.1/132.1
125.6/125.1
–
12.1/10.0
61.4/61.7
21.7/24.2
11.0/11.7
149.3/149.3
126.3/125.7
123.9/124.0
149.6/150.2
–
OCH3
a
Values left of the slash: like-isomer (l) (RS,RC)-1 or (SS,SR)-1; values right of the slash: unlike-isomer (u) (RS,SC)-2 or (SS,RC)-2.
b
1H, 1H coupling constants in the 2-butyl residue are uniform for diastereomers regardless of the substituent X: 1a–1e (like), 2J(H-3a,H-3b) = 13.9 0.1 Hz, 3J(H-1,H-
2) = 6.9 0.1 Hz, 3J(H-2,H-3a) = 9.1 0.1 Hz, 3J(H-2,H-3b) = 4.1 0.1 Hz, 3J(H-3a,H-4) = 3J(H-3b,H-4) = 7.5 0.1 Hz; 2a-2e (unlike), 2J(H-3a,H-3b) = 14.1 0.1 Hz, 3J(H-1,H-
2) = 6.9 0.1 Hz, 3J(H-2,H-3a) = 6.9–8.1 Hz, 3J(H-2,H-3b) = 5.8–6.6 Hz, 3J(H-3a,H-4) = 3J(H-3b,H-4) = 7.5 0.1 Hz.
Coupling constants involving 19F: 1b, 3J(19F,1H) = 8.8 Hz (H-30 /50 ), 4J(19F,1H) = 8.5 Hz (H-20/60), 1J(19F,13C) = 245.6 Hz (C-40 ), 2J(19F,13C) = 21.2 Hz (C-30 /50 ), 3J(19F,13C) = 8.3 Hz
c
(C-20 /60 ), 4J(19F,13C) = 3.1 Hz (C-10 ), 6J(19F,13C) = 1.0 Hz (C-2); 2b, 3J(19F,1H) = 9.0 Hz (H-30 /50 ), 4J(19F,1H) = 8.8 Hz (H-20/60), 1J(19F,13C) = 243.1 Hz (C-40 ), 2J(19F,13C) = 21.6 Hz (C-30 /
50 ), 3J(19F,13C) = 8.0 Hz (C-20 /60 ), 4J(19F,13C) = 2.9 Hz (C-10).
d
Diastereotopic atoms H-3a and H-3b with d(H-3a) < d(H-3b) in all derivatives; for the stereochemical assignment see text.
2.1.3. (RS,RC)- or (SS,SC)-4-fluoro-1(1-methylpropylthio)benzene-S-
2.1.5. (RS,RC)- or (SS,SC)-1-bromo-4(1-methylpropylthio)benzene-S-
oxide (1b)
oxide (1c)
ꢁ1
(cmꢁ1) 3050, 2968, 2933,
Yield: 62%, viscous, slightly yellowish liquid. IR (liquid)
m
~
m
~
(cm
Yield, 51%, m.p. 71–72 °C. IR (solid)
)
1589, 1491, 1460, 1382, 1220, 834. ESI-MS calculated for
C
3042, 2966, 2931, 1571, 1471, 1385, 1063, 1036, 817. ESI-MS cal-
10H14FOS: 201.0749 [M+H]+, found: 201.0747.
culated for C10H14BrOS: 260.9949 [M+H]+, found: 260.9954.
2.1.4. (RS,SC)- or (SS,RC)-4-fluoro-1(1-methylpropylthio)benzene-S-
2.1.6. (RS,SC)- or (SS,RC)-1-bromo-4(1-methylpropylthio)benzene-S-
oxide (2b)
oxide (2c)
ꢁ1
Yield, 47%, m.p. 70–71 °C. IR (solid)
(cmꢁ1) 3046, 2966, 2937,
Yield: 62%, viscous, slightly yellowish liquid. IR (liquid)
(cm
)
~
m
~
m
1592, 1487, 1460, 1380, 834. ESI-MS calculated for C10H14FOS:
3038, 2966, 2924, 1570, 1461, 1385, 1061, 1036, 814. ESI-MS cal-
201.0749 [M+H]+, found: 201.0756.
culated for C10H14BrOS: 260.9949 [M+H]+, found: 260.9953.

J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
89
2.1.7. (RS,RC)- or (SS,SC)-4-nitro-1(1-methylpropylthio)benzene-S-
ever, should be avoided when hard-base ligands like oxygen func-
tionalities are involved because acetone-d6 is a serious competitor
in the adduct formation in such cases.
oxide (1d)
~
Yield: 74%, m.p. 82–83 °C. IR (solid)
m
(cmꢁ1) 3050, 2968, 2932,
1602, 1521, 1343, 1038, 852. ESI-MS calculated for C10H14NO3S:
Note that
signs because nonracemic mixtures with known enantiomeric
composition have been used; the definitions are: (RSRC) –
(SSSC) for like-sulfoxides 1 and (RSSC) – (SSRC) for unlike-
Dm values are B0-dependent. They are quoted with
228.0694 [M+H]+, found: 228.0689.
Dm = m
2.1.8. (RS,SC)- or (SS,RC)-4-nitro-1(1-methylpropylthio)benzene-S-
m
D
m
=
m
m
oxide (2d)
sulfoxides 2. All dispersion values are given as integers in Hz as
determined at B0 = 9.4 Tesla corresponding to 400 MHz 1H and
100.6 MHz 13C.
Yield: 76%, m.p. 79–81 °C. IR (solid)
(cmꢁ1) 3050, 2968, 2928,
~
m
1600, 1523, 1343, 1037, 850. ESI-MS calculated for C10H14NO3S:
228.0694 [M+H]+, found: 228.0698.
2.4. Polarimetry
2.1.9. (RS,RC)- or (SS,SC)-4-methoxy-1(1-methylpropylthio)benzene-S-
oxide (1e)
Using a Perkin–Elmer Polarimeter 341, specific rotations (at
589 nm) of the enantiomers with (RC)-configuration were mea-
sured in methanol at room-temperature; concentrations (in g/ml)
ꢁ1
~
m
Yield: 94%, viscous, slightly yellowish liquid. IR (liquid)
(cm
)
3042, 2968, 1602, 1583, 1459, 1038, 850. ESI-MS calculated for
C
10H14NO3S: 213.0949 [M+H]+, found: 213.0947.
are given in parentheses; [a]D: 1a,+106.8 (0.0082); 1b,+82.3
(0.0074); 1c,+91.6 (0.0154); 1d,+73.1 (0.0036); 1e,+146.2
(0.0099); 2a,ꢁ215.0 (0.0120); 2b,ꢁ188.2 (0.0079); 2c,ꢁ165.8
(0.0147); 2d,ꢁ167.1 (0.0084); 2e,ꢁ168.7 (0.0120).
2.1.10. (RS,SC)- or (SS,RC)-4-methoxy-1(1-methylpropylthio)benzene-
S-oxide (2e)
ꢁ1
~
Yield: 96%, viscous, slightly yellowish liquid. IR (liquid)
m
(cm
)
3051, 2970, 1606, 1582, 1449, 1040, 850. ESI-MS calculated for
C
2.5. Crystallography
10H14NO3S: 213.0949 [M+H]+, found: 213.0952.
X-ray diffraction experiments were performed using a Stoe IPDS
single crystal diffractometer. Crystallographic data, the table of
atomic coordinates and thermal parameters, and the full list of
bond lengths and angles of (SS,RC)-2d have been deposited with
the Cambridge Crystallographic Data Centre, CCDC 746478; email:
deposit@ccdc.cam.ac.uk; see also Table 1. The enantiomer (RS,SC)-
2d has also been examined. All data are the same except signs of
torsion angles which have opposite signs.
2.1.11. rac-4-Nitro-1-methylthiobenzene-S-oxide (5)
Yield: 24%, yellowish solid, m.p. 147–149 °C. IR (solid)
ꢁ1
~
m
(cm )
3099, 3022, 2390, 1514, 1474, 1338, 1306, 1086, 1044, 850, 740.
ESI-MS calculated for C7H7NO3S: 186.0225 [M+H]+, found:
186.0223.
NMR, in CDCl3, d (ppm): 2.80 (H-1, methyl); 7.84 (H-20/60); 8.40
(H-30/50); 43.9 (C-1, methyl); 149.5 (C-10, ipso); 124.7 (C-20/60,
ortho); 124.5 (C-30/50, meta); 153.2 (C-40, para);
Dd (ppm)/Dm
(Hz) + 1 eq. Rh*: +0.62/4 (H-1, methyl); +0.21/not detectable (H-
20/60); ꢁ0.30/not detectable (H-30/50); ꢁ3.2/51 (C-1, methyl);
ꢁ1.5/11 (C-10, ipso); +1.4/6 (C-20/60, ortho); ꢁ0.5/5 (C-30/50, meta);
methyl); +0.28/9 (H-20/60); ꢁ0.45/35 (H-30/50); ꢁ4.7/62 (C-1,
methyl); ꢁ3.3/11 (C-10, ipso); +1.7/3 (C-20/60, ortho); ꢁ0.3/12 (C-
30/50, meta); ꢁ3.9/12 (C-40, para).
2.6. Powder diffraction
Powder diffraction measurements were performed on a Stoe
powder diffractometry system Stadi P with PSD Cu Ka irradiation,
anode current 30 mA, 40 kV anode voltage, range 5–100° 2#.
ꢁ3.8/6 (C-40, para);
Dd (ppm)/Dm (Hz) + 2 eq. Rh*: +0.82/9 (H-1,
2.7. Theoretical calculations
2.2. Liquid chromatography
All molecular structures were calculated by density functional
methods (B3LYP 6-31Gꢃ and 6-311++Gꢃꢃ) and second-order
Møller–Plesset (MP2 6-31Gꢃ) using the SPARTAN 008 program
package, version 1.0.0 [15].
Column chromatography was performed at atmospheric pres-
sure using silica gel 60 M, 230–400 mesh (Merck) as stationary
and a mixture of petrol ether and ethyl acetate (1:1) as mobile
phase.
3. Results and discussion
2.3. NMR spectroscopy
Sulfoxides offer both types of ligand categories simultaneously,
a soft sulfur atom with a free electron pair and a hard oxygen
atom; the S(+)–O(ꢁ) bond is predominantly a single bond [16]. The
question is which of these two atoms, O or S, is the binding site.
Moreover, sulfoxides possess a configurationally stable chirality
centre at the sulfur atom, (RS) or (SS). If they are substituted by a
chiral hydrocarbon side chain, such as (R)- and (S)-2-butyl, two
diastereomeric pairs of enantiomers exist which, in the presence
of the enantiopure Rh*, are all diastereomeric and display different
NMR spectral parameters. We report here our results on such
molecular systems 1 and 2 (Scheme 2).
All NMR measurements were performed in analogy to those de-
scribed for the corresponding ethers 3 [8]; details of the one- and
two-dimensional NMR experiments (DEPT 90 and DEPT 135, gradi-
ent-selected COSY, HMQC and HMBC spectra) can be found in this
reference.
1H (400.1 MHz) and 13C (100.6 MHz) NMR measurements were
recorded at room-temperature on a Bruker Avance DPX-400 spec-
trometer. Samples were ca 0.01–0.025 mmol in CDCl3. The chemi-
cal shift reference was internal tetramethylsilane (d = 0 ppm).
In the standard dirhodium experiment, Rh* and equimolar
amounts of the ligands 1a–1e, 2a–2e and 5 (Scheme 2), respec-
tively, were dissolved in 0.7 ml CDCl3; quantities of 10–25 mg of
Rh* (ca. 0.01–0.025 mmol concentration) were employed. If neces-
sary, the dissolution process can be accelerated by exposing NMR
sample tubes to an ultrasonic bath for a few minutes. In earlier re-
ports on soft-base ligands, the use of acetone-d6 for increasing the
solubility of Rh* had been recommended [5]. This auxiliary, how-
3.1. Syntheses, isolation and characterization of the four stereoisomers
of 2-butylphenylsulfoxide 1a and 2a as well as their p-substituted
derivatives (1b–1e and 2b–2e)
In order to synthesize all enantiomerically pure isomers of 1a–
1e and 2a–2e, we started from the two commercially available

90
J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
enantiopure 2-butanols and prepared the respective 2-butylphenyl
and their para-substituted derivatives 4a–4e [10]. Oxidation [12]
of those thioethers resulted in mixtures of two diastereomeric,
enantiopure sulfoxides in each case (x = a, b, c, d, and e; Scheme 2):
phenylsulfoxide (1d and 2d) were isolated and characterized
unequivocally.
Inspecting the 1H and 13C NMR chemical shift data of 1d and 2d
(Table 2), it is evident that they differ significantly in their d values
for H-1 (1.19 vs. 1.03 ppm), H-2 (2.78 vs. 2.60 ppm), H-3 (1.48/1.72
vs. 1.57/2.01 ppm), H-4 (1.01 vs. 1.15 ppm), C-1 (12.1 vs.
10.0 ppm), and C-3 (21.7 vs. 24.2 ppm). Interestingly, analogous
chemical shift differences are observed for all other derivatives,
too. The reason for this systematic chemical shift divergence is,
of course, the stereochemistry of the sulfoxides. Some 13C and 1H
chemical shift trends exerted by S(+)–O(ꢁ) groups on neighbouring
hydrogen and carbon atoms have been reported [16,18] but they
were derived from compounds containing rigid six-membered
rings in chair-conformation whereas the sulfoxides 1 and 2 are
mixtures of more than one open-chain conformers (see below).
Therefore, empirical correlations between those molecular systems
with respect to effects on chemical shifts failed or appeared to be
speculative.
ðRÞ-4x ! ðRS; RCÞ-1x þ ðSS; RCÞ-2x
ðSÞ-4x ! ðSS; SCÞ-1x þ ðRS; SCÞ-2x
The two diastereomers in these mixtures were isolated by li-
quid chromatography (for details see Section 2). In the case of
the nitro derivatives 1d and 2d, the less polar isomer was eluted
first from each of the reaction mixtures and recrystallized for X-
ray diffraction experiments (Fig. 1 and Table 1; see also Section
3.2).
It turned out from those measurements that, in both cases, the
less polar isomer was the unlike-configurated, (SS,RC)-2d and
(RS,SC)-2d (Fig. 1), respectively [17]; i.e., those compounds were
enantiomers. Thus, the other one in each of the mixtures obtained
from the oxidation reaction was the like-stereoisomer, (RS,RC)-1d
and (SS,SC)-1d, respectively, again enantiomers. Thereby, the abso-
lute configurations of all four stereoisomers of 2-butyl-para-nitro-
The specific rotations [a]D provided an independent tool for the
differentiation of diastereomers. For the (RC)-isomers, they are po-
sitive for the like- but negative for the unlike-diastereomers (see
Section 2); vice versa for the (SC)-isomers.
As a result, the stereochemistry of all sulfoxides could be as-
signed in analogy to 1d and 2d with a very high level of reliability.
3.2. Conformational analysis of the sulfoxides 1d and 2d
Theoretical density functional (B3LYP/6-21Gꢃ and 6-311++Gꢃꢃ)
as well as second-order Møller–Plesset (MP2/6-31++Gꢃꢃ) calcula-
tions have been performed for the sulfoxides 1a/2a (X = H), 1b/
2b (X = F) and 1d/2d (X = NO2) as isolated molecules without sol-
vent shell. Overall, very similar energy differences for the most sta-
ble conformers resulted from these calculations regardless of the
substituents so that we restrict the following discussion to the ni-
tro derivatives 1d and 2d (Table 3 as well as Figs. 2 and 3, respec-
tively) in order to compare the calculated structures with the X-ray
diffraction geometries (Table 4 and Fig. 1). It should be noted that
one has to accept energy error limits of 1–2 kJ/mol in such calcula-
tions so that a quantitative evaluation of the balances in the equi-
libria of the most stable conformers is doubtful. In the following
interpretation, we excluded conformers with energies higher than
4 kJ/mol relative to that with the lowest energy.
Fig. 1. ORTEP plot from the X-ray diffraction measurement of 4-nitro-1(1-meth-
ylpropylthio)benzene-S-oxide [(RS,SC)-2d]; for discussion see text.
Table 3
Molecular geometry parameters of 2d-conformer B (see Fig. 2) as calculated by
density functional BYLYP and Møller–Plesset MP2 methods [15] compared with data
from an X-ray diffraction measurement.
A comparison of the calculated geometry parameters with those
of the X-ray structure shows that the MP2 calculations reproduce
the experimental values better than any of the density functional
calculations (Table 3). This, however, does not necessarily mean
that MP2-calculated energies are more reliable. Anyway, the most
stable conformers which play the major role in the equilibria were
reproduced satisfactorily by all calculation methods (Table 4).
We start the discussion with the unlike-isomer 2d because we
have an X-ray structure of that diastereomer: Fig. 2 shows the
two calculated conformers of (RS,SC)-2d with the lowest calculated
Geometry
Parameters
Density functional B3LYP
MP2
6-31Gꢃ
X-ray
6-31Gꢃ
6-311++Gꢃꢃ
d(S–O) [pm]
151.6
188.3
182.7
107.2
106.6
ꢁ45.3
ꢁ78.7
ꢁ177.6
151.6
188.8
183.1
107.3
106.1
ꢁ45.3
ꢁ79.5
ꢁ178.1
151.7
184.2
180.6
106.8
106.7
ꢁ46.5
ꢁ76.9
ꢁ177.5
149.6
177.9
179.8
106.4
107.7
ꢁ51.9
ꢁ71.6
ꢁ175.3
d(S–C-2) [pm]
d(S–C-10 ) [pm]
a
a
(O–S–C-2) [0]
(O–S–C-10 ) [°]
/(O–S–C-2–C-1) [°]
/(O–S–C-2–C-3) [°]
/(O–S–C-10 –C-20 ) [°]
Fig. 2. Low-energy conformers A and B of (RS,SC)-2d; as calculated [15]; energy value of conformer B relative to that of A. Color code: black, carbon; white, hydrogen; red,
oxygen; green-yellow, sulfur; blue, nitrogen. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
91
Fig. 3. Low-energy conformers C, D and E of (RS,RC)-1d; as calculated by MP2 methods [15]; energy values relative to that of conformer C. Color code: black, carbon; white,
hydrogen; red, oxygen; green-yellow, sulfur; blue, nitrogen. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
Table 4
Relative conformational energies of some conformers A–F of 1d and 2d (see Figs. 2
and 3) as calculated by density functional (B3LYP) and Møller–Plesset MP2 methods
[15]; energy of B relative to that of A (unlike-conformers) and energies of D–F relative
to that of C (like-conformers).
Conformational
Density functional B3LYP
Møller–Plesset MP2
Energies (in kJ/mol)
6-31Gꢃ
6-311++Gꢃꢃ
6-31Gꢃ
1d (unlike)
2d (like)
A
B
C
D
E
F
0
1.4
0
1.4
2.2
3.6
0
1.3
0
1.7
2.4
3.8
0
1.9
0
0.4
4.3
1.7
energies: A (0 kJ/mol) and B (ca. +1.5 to +2 kJ/mol higher than A).
Conformer B corresponds excellently to that obtained from the
X-ray diffraction experiment (Fig. 1); conformer A differs from B
basically by a rotation of the terminal ethyl group around the C-
2–C-3 bond forming an antiperiplanar orientation of the butyl
chain.
The fact that we found two conformations A and B with quite
similar calculated energies (isolated molecules) is not in contrast
to the existence of one sole conformer in the solid state where only
B was observed. Stoe powder diffraction measurements were per-
formed with both enantiomerically pure unlike-2d samples. In one
case, SS,RC-configuration, a large number of reflections occurred so
that we assume that this sample was composed of mixed crystals.
On the other hand, only very few reflections were detected for the
sample with RS,SC-configuration suggesting that only those crystals
exist where the molecules adopt conformation B (Fig. 2). This may
be explained by the recrystallization process of the two samples
which may have been different with respect to external factors like
solution concentrations, temperatures and, eventually, marginal
impurities.
Fig. 4. Newman projections and spatial orientation of the H-2 and H-3 atoms in the
conformations C and D of (RS,RC)-1d (cf. Fig. 3) as well as A and B of (RS,SC)-2d (cf.
Fig. 2).
A test for the validity of the geometry calculations of the diaste-
reomeric sulfoxides is the interpretation of the H-2/H-3 NMR sig-
nals coupled to the methyl protons H-1 and H-4. These protons
form a first-order spin system in the 400 MHz 1H NMR spectra
which can be interpreted easily in terms of 1H,1H coupling con-
stants [19]. In each series of diastereomers the coupling constants
are remarkably uniform (see footnote ‘b’ in Table 2) indicating that
the conformational behavior is rather independent of the nature of
the substituent X.
The NMR data of the relevant hydrogen atoms in (RS,RC)-1d are:
d(H-2) = 2.78 ppm, d(H-3a) = 1.48 ppm and d(H-3b) = 1.72 ppm (Ta-
ble 2); 3J(H-2,H-3a) = 9.2 Hz and 3J(H-2,H-3b) = 4.1 Hz indicating
that the first coupling parameter is dominated by an antiperiplanar
orientation of H-2 and H-3a whereas H-2 and H-3b are basically
gauche-oriented [19]. By inspecting the calculated major conformers
Four conformers have to be inspected in the like-series of dia-
stereomers (RS,RC)-1d: C (0 kJ/mol), D (+0.4 – +1.7 kJ/mol), E and
F (both ca. +2 to +4 kJ/mol) (Fig. 3, Table 3) but only C and D seem
to be significant.

92
J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
C and D (Fig. 3), it turns out that the (pro-R)-H-3 is gauche to H-2 in
both conformations (calculated torsion angle in C: u = 54.3°, in D:
u = 65.8°); on the other hand, (pro-S)-H-3 is antiperiplanar to H-2
(calculated torsion angle in C: u = 170.0°, in D: u = ꢁ177.2°); see
Fig. 4. Thus, a stereochemical assignment is possible: (pro-R)-H-3
is H-3b and (pro-S)-H-3 is H-3a. Notabene: this pro-R/pro-S assign-
ment is opposite for the other enantiomer (SS,SC)-1d.
The situation is different for the unlike-isomer. For (RS,SC)-2d,
the (pro-R)-H-3 is gauche with respect to H-2 in conformer A (cal-
culated torsion angle u = 73.9°) but antiperiplanar in conformer B
(calculated torsion angle u = ꢁ176.9°); for the (pro-S)-H-3 atom
it is opposite (calculated torsion angle in A: u = ꢁ171.6°, in B: u
= ꢁ61.2°); see Fig. 4. Consequently, the two three-bond coupling
constants 3J(H-2,H-3a) and 3J(H-2,H-3b) are quite similar in their
values: 7.3 and 6.6 Hz, respectively. So, a stereochemical H-3a/H-
3b assignment is not possible here.
each stereoisomer in two different ways (Table 5). Enantiomers
were mixed to form nonracemic mixtures so that the sets of indi-
vidual NMR signals could be assigned to each enantiomer by their
different signal intensities (Experiments A); this corresponds to the
routine dirhodium method [5–8]. In the present study, however,
we had all stereoisomers at hand so that it was of interest to per-
form dirhodium experiments with each of the enantiomers sepa-
rately (Experiments B). Of course, we paid attention to keep
external influences (concentration, molar ratios etc.) constant for
all these measurements.
It turned out that differences in
Dd values between Experiments
A and B were, indeed, found which are significant for some of the
1H and 13C nuclei; for details see Section 3.4. In general, however,
the evaluation of
Dd values in terms of distances to binding sites is
not seriously affected by the different experimental set-ups.
The largest deshieldings are observed for the H-2 and C-2 sig-
nals indicating – as expected – that the complexation site is the
SO group attached to this methine. Apparently, no other hetero
atom or the phenyl group can compete in the complexation, even
not the methoxy oxygen atom in 1e/2e. Neighbouring atoms may
show significant signal shifts as well but these are smaller and
not always consistent for all four stereoisomers of a given deriva-
tive. For example, C-1 is deshielded (+0.2 to +0.7 ppm) in all un-
like-isomers but shielded (ꢁ0.2 to ꢁ0.5 ppm) in all like-isomers.
3.3. Dirhodium experiments – complexation shifts (
modes
Dd) and binding
Ligand exchange in Rh*-adducts is fast on the NMR time-scale
[5]. Therefore, NMR signals are averages of those of the complexed
and the free ligands and signal shifts are defined as
D
d = d (com-
plexed) – d (free). Complexation shifts
Dd have been recorded for
Table 5
1H and 13C complexation shifts
D
d of the sulfoxides 1 and 2 after addition of an equimolar amount of Rh*; in ppm.a
1a/2a
1a/2a
1b/2b
1b
1c/2c
1c/2c
1d/2d
1d/2d
1e/2e
1e/2e
X = H
X = H
X = F
X = F
X = Br
X = Br
X = NO2
X = NO2
X = OCH3
X = OCH3
(RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC) (RS,RC)/(SS,SC)
(RS,SC)/(SS,RC) (RS,SC)/SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC) (RS,SC)/(SS,RC)
Experiment A Experiment B Experiment A Experiment B Experiment A Experiment B Experiment A Experiment B Experiment A Experiment B
H-1
+0.20/+0.21
+0.19/+0.20
+0.65/+0.65
+0.70/+0.70
+0.04/+0.07
+0.13/+0.09
+0.26/+0.23
+0.34/+0.34
+0.20/+0.24
+0.23/+0.12
+0.57/+0.53
+0.57/+0.41
+0.06/+0.11
+0.13/+0.09
+0.23/+0.21
+0.22/+0.19
+0.25/+0.22
+0.21/+0.19
+0.75/+0.75
+0.67/+0.71
+0.11/+0.10
+0.14/+0.12
+0.28/+0.28
+0.30/+0.30
+0.12/+0.15
+0.22/+0.15
+0.25/+0.28
+0.51/+0.48
+0.10/+0.09
+0.12/+0.11
+0.33/+0.39
+0.31/+0.23
+0.21/+0.35
+0.34/+0.25
+0.69/+0.69
+0.75/+0.75
+0.11/+0.14
+0.13/+0.20
+0.26/+0.24
+0.33/+0.33
+0.23/+0.24
+0.18/+0.20
+0.74/+0.62
+0.68/+0.73
+0.12/+0.13
+0.13/+0.21
+0.27/+0.24
+0.30/+0.31
+0.22/+0.28
+0.27/+0.20
+0.18/+0.23
+0.19/+0.22
+0.26/+0.29
+0.19/+0.23
+0.75/+0.75
+0.79/+0.79
+0.06/+0.13
+0.16/+0.18
+0.28/+0.28
+0.33/+0.33
+0.20/+0.25
+0.28/+0.21
+0.48/+0.32
+0.52/+0.45
+0.09/+0.14
+0.08/+0.04
+0.31/+0.26
+0.39/+0.23
ꢁ0.09/ꢁ0.10
ꢁ0.11/ꢁ0.13
+0.15/+0.23
+0.28/+0.25
ꢁ0.18/ꢁ0.16
H-2
+0.78/+0.78b +0.55/+0.58
+0.68/+0.68b +0.56/+0.64
+0.18/+0.18b +0.11/+0.16
+0.18/+0.18b +0.11/+0.16
+0.29/+0.29b +0.22/+0.24
+0.30/+0.30b +0.25/+0.30
H-3ac
H-3bc
H-4
ꢁ0.09/ꢁ0.10 ꢁ0.09/ꢁ0.06 ꢁ0.10/ꢁ0.20 +0.01/ꢁ0.04
ꢁ0.12/ꢁ0.07 ꢁ0.10/ꢁ0.06 ꢁ0.08/ꢁ0.03 ꢁ0.06/ꢁ0.03 ꢁ0.12/ꢁ0.08
ꢁ0.14/ꢁ0.09 ꢁ0.13/ꢁ0.06 ꢁ0.10/ꢁ0.12 ꢁ0.12/ꢁ0.07 ꢁ0.14/ꢁ0.09 ꢁ0.13/ꢁ0.11 –0.12/ꢁ0.08
ꢁ0.11/ꢁ0.09 ꢁ0.12/ꢁ0.09
H-20/H-60 +0.31/+0.28
+0.37/+0.33
+0.25/+0.17
+0.37/+0.18
+0.31/+0.46
+0.28/+0.34
+0.08/+0.17
+0.29/+0.15
0.00/ꢁ0.03
–0.03/+0.01
–
+0.02/ꢁ0.07
+0.06/ꢁ0.02
+0.08/+0.01
+0.08/ꢁ0.01
–
+0.04/ꢁ0.06
+0.07/ꢁ0.04
+0.11/ꢁ0.01
+0.04/ꢁ0.04
–
+0.28/+0.21
+0.25/+0.19
+0.20/+0.16
+0.18/+0.17
+0.31/+0.22
+0.32/+0.24
H-30/H-50 ꢁ0.04/ꢁ0.07 ꢁ0.08/ꢁ0.13 +0.34/+0.43
ꢁ0.04/ꢁ0.08 ꢁ0.08/ꢁ0.07 +0.18/+0.24
ꢁ0.19/ꢁ0.27 ꢁ0.17/ꢁ0.23 ꢁ0.10/ꢁ0.17
–0.19/ꢁ0.27
ꢁ0.17/ꢁ0.21 ꢁ0.11/ꢁ0.18/ ꢁ0.11/ꢁ0.12
H-40
+0.02/+0.02
ꢁ0.03/0.00
–
ꢁ0.01/ꢁ0.07
ꢁ0.05/ꢁ0.09
–
–
–
–
–
–
OCH3
C-1
–
–
–
–
–
–
ꢁ0.04/ꢁ0.07
ꢁ0.05/ꢁ0.07
ꢁ0.6/ꢁ0.3
+0.5/+0.6
+1.0/+0.8
+0.8/+0.7
ꢁ0.1/ꢁ0.2
–1.1/ꢁ1.1
ꢁ0.1/ꢁ0.1
ꢁ0.5/ꢁ0.4
ꢁ2.6/ꢁ2.6
ꢁ3.3/ꢁ3.3
+1.1/+1.0
+1.8/+1.8
0.0/0.0
ꢁ0.08/ꢁ0.06
ꢁ0.04/ꢁ0.04
ꢁ0.5/ꢁ0.4
+0.7/+0.4
+0.6/+0.7
+0.8/+0.7
0.0/ꢁ0.1
ꢁ0.6/ꢁ0.5
+0.6/+0.6
+0.8/+0.7
+0.9/+0.9
+0.3/+0.3
ꢁ1.3/ꢁ1.4
ꢁ0.1/ꢁ0.1
ꢁ0.4/ꢁ0.4
ꢁ2.6/ꢁ2.6
ꢁ3.0/ꢁ3.0
+1.0/+0.9
+1.6/+1.2
0.0/0.0
ꢁ0.5/ꢁ0.2
+0.7/+0.2
+0.7/+0.4
+0.8/+0.3
+0.1/0.0
ꢁ0.5/ꢁ0.6
+0.6/+0.6
+1.0/+1.1
+1.1/+1.0
+0.5/+0.0
ꢁ0.9/ꢁ0.7
ꢁ0.1/ꢁ0.1
ꢁ0.4/ꢁ0.4
+0.4/+0.2
+0.1/+0.1
+1.2/+0.9
+1.2/+1.3
0.0/+0.4
ꢁ0.5/ꢁ0.3
+0.7/+0.3
+0.4/+0.6
+0.7/+0.3
+0.8/0.0
ꢁ0.5/ꢁ0.7
+0.7/+0.5
+1.0/+0.9
+0.8/+0.7
+0.1/ꢁ0.1
–1.2/ꢁ1.0
ꢁ0.1/ꢁ0.2
ꢁ0.5/ꢁ0.5
ꢁ3.4/ꢁ3.5
ꢁ3.6/ꢁ3.6
+1.0/+0.8
+1.1/+1.0
ꢁ0.1/0.0
0.0/ꢁ0.1
+0.8/+0.7
+0.9/+0./8
–
ꢁ0.5/ꢁ0.4
+0.5/+0.5
+0.9/+0.6
+0.5/+0.5
+0.1/+0.7
–1.2/ꢁ0.8
+0.2/+0.1
ꢁ0.5/ꢁ0.5
ꢁ3.7/ꢁ2.7
ꢁ2.8/ꢁ3.9
+1.0/+0.7
+0.9/+0.9
0.0/ꢁ0.2
ꢁ0.1/ꢁ0.6
+0.9/+0.3
+0.6/+0.8
–
ꢁ0.5/ꢁ0.4
+0.6/+0.6
+1.0/+1.0
+0.7/+0.6
+0.2/+0.1
–0.9/ꢁ1.0
ꢁ0.2/ꢁ0.1
ꢁ0.6/ꢁ0.5
ꢁ3.9/ꢁ4.0
ꢁ3.1/ꢁ3.1
+1.0/+0.9
+1.0/+1.0
ꢁ0.1/ꢁ0.1
ꢁ0.1/ꢁ0.1
+0.4/+0.3
ꢁ0.7/ꢁ0.8
–
ꢁ0.4/ꢁ0.4
+0.4/+0.4
+0.7/+0.7
+0.4/+0.4
+0.1/+0.1
–0.9/ꢁ0.9
ꢁ0.2/ꢁ0.2
ꢁ0.5/ꢁ0.5
ꢁ2.6/ꢁ2.7
ꢁ1.8/ꢁ2.1
+0.8/+0.7
+0.8/+0.8
ꢁ0.2/ꢁ0.3
ꢁ0.2/ꢁ0.3
ꢁ0.1/ꢁ0.2
ꢁ0.7/ꢁ0.7
–
C-2
C-3
–1.2/ꢁ0.7
–1.1/ꢁ0.6
–0.6/ꢁ1.1
0.0/ꢁ0.1
C-4
0.0/0.0
+0.5/0.0
ꢁ0.5/ꢁ0.3
ꢁ2.4/ꢁ2.3
ꢁ3.3/ꢁ4.0
+0.9/+0.7
+1.2/+0.5
ꢁ0.1/ꢁ0.1
0.0/ꢁ0.2
+0.3/+0.2
+0.6/+0.1
–
ꢁ0.4/ꢁ0.3
ꢁ2.6/ꢁ2.7
ꢁ3.0/ꢁ2.9
ꢁ0.1/+0.5
+1.4/+0.8
ꢁ0.2/ꢁ0.1
ꢁ0.1/ꢁ0.2
ꢁ0.2/+0.3
+0.4/+0.1
–
ꢁ0.5/ꢁ0.5
ꢁ4.0/ꢁ4.1
ꢁ4.6/ꢁ4.5
+0.3/+1.0
+1.4/+1.8
ꢁ0.2/+0.1
0.0/ꢁ0.2
C-10
C-20/C-60
C-30/C-50
C-40
ꢁ0.1/0.0
ꢁ2.9/ꢁ2.9
ꢁ3.0/ꢁ3.0
–
0.0/ꢁ0.1
ꢁ3.8/ꢁ3.9
ꢁ3.4/ꢁ3.1
–
0.0/0.0
+0.4/+0.3
+0.5/+0.4
+0.1/ꢁ0.1
+0.1/0.0
ꢁ0.2/0.4
+0.5/0.4
OCH3
+0.1/+0.2
+0.1/ꢁ0.1
a
Experiment A: measurement with nonracemic mixtures, Experiment B: separate measurements with pure enantiomers. Upper row in each entry are data of like-isomers (l)
(RS,RC)-1i and (SS,SR)-1i, lower row in each entry those of unlike-isomers (u) (RS,SC)-2i and (SS,RC)-2i (i = a, b, c, d or e, respectively); compare R/S symbols in the heading.
b
Signal overlap. No safe individual signal assignments; values are averaged.
Diastereotopic atoms H-3a and H-3b with d(H-3a) < d(H-3b) in all derivatives; for stereochemical assignments see text.
c

J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
93
Presumably, Rh*-phenyl group anisotropies overrule these small
inductive deshieldings produced by the complexation of the latter
isomers.
In previous studies on related ethers [8] and thioethers [10], we
have shown that the correlation between the Hammett resonance
constants r0R and the
D
d(C-20/60) values is indicative of the absence
(negative slope) or the presence of HOMO–LUMO orbital interac-
tion (positive slope). The sulfoxides in the present study offer
two potential binding sites, oxygen as a hard Lewis acid and sulfur
as a soft Lewis acid, and it has been shown in X-ray reports that
both atoms may serve as binding sites [20]. Thus, a comparison
of the sulfoxide correlation with that of the ethers 3 (black dia-
monds in Fig. 5) and that of the thioethers 4 (magenta squares)
is expected to show which of these two chalcogen atoms is favored
in the adduct formation in solution.
Fig. 5 shows the dependence of C-20/60complexation shifts of the
substituted sulfoxides on the nature of X as a function of the
respective r0R values of X (green triangles for like- and blue ones
for unlike-sulfoxides). In general, the complexation shifts are small,
and the graphs show slightly negative slopes suggesting that the
behavior of the sulfoxides is quite similar to that of the hard ether
but totally different from that of the soft thioethers. Thus, the sulf-
oxides discussed here are hard donors barely able to form orbital
(HOMO–LUMO) interactions; i.e., oxygen is the primary binding
site. Fig. 6 shows the HOMO of the conformation A of (RS,SC)-2d;
those of all other conformers and derivatives are similar. Basically,
it is a p*-orbital of the S-O bond with a contribution of aromatic p-
conjugation which varies a little with the nature of X: stronger for
X = NO2 and weaker for X = F. Moreover, this orbital is sterically
shielded by the adjacent phenyl and 2-butyl groups so that an ap-
proach for efficient orbital interaction with rhodium is hindered.
Thus, the HOMO orbital at sulfur (Lewis base) and the LUMO of
the Rh–Rh bond (Lewis acid) are ‘‘frustrated” orbitals [21].
This interpretation is corroborated by the data of rac-p-nitro-
phenylmethylsulfoxide (5; NMR data in the Experimental Part)
with its slimmer methyl group, the
is significantly larger than that of the corresponding 2-butyl ana-
D
d(C-20/60) value (+1.4 ppm)
Fig. 6. HOMO of conformation A of (RS,SC)-2d.
logues 1d and 2d (+0.7 to +1 ppm; data point ‘‘x” in Fig. 5 and Table
5); one mol equivalent Rh* added in each experiment. Adding a
second mol equivalent of Rh* to the equimolar solution of 5 and
Rh* increases this parameter further to +2.2 ppm (data point ‘‘+”
in Fig. 5). These enhanced resonance effects suggest a shift in the
adduct formation equilibrium (Fig. 7) towards to S-adduct. Analo-
gous enhancements of the
D
d(C-20/60) values are observed when
1d/2d are recorded in the presence of two mol equivalents of
Rh* (filled orange circle in Fig. 5).
The complexation shift of
ative:
d = ꢁ3.2 (1 mol equivalent Rh*) and ꢁ3.3 ppm (2 mol
equivalent Rh*) whereas it is weak and positive for the -posi-
a-positioned C-1 in 5 is strongly neg-
D
a
tioned C-2 in 1d and 2d: +0.6 to +1.0 ppm. The negative value in
5, vanishing for 1d/2d, points to a larger contribution of the S-ad-
duct in the equilibrium, too. Analogous negative
carbons have been observed in phosphanes [22]. These shielding
effects (negative d-value) may be a consequence of binding rho-
Dd values for a-
D
dium to a third-row elements, phosphorus or sulfur; it is reason-
able to assume that a back-donation exists from the HOMO of
the Rh–Rh bond to an unoccupied 3d orbital of the ligating
Fig. 5. Complexation shifts
Dd of ortho-carbon atoms as a function of Hammett
resonance parameters r0R of X for sulfoxides 1, 2 and 5 as well structurally related
ethers [8] and thioethers [10].

94
J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
Fig. 7. Adduct formation equilibria of a sulfoxide and Rh*.
phosphorus or sulfur, respectively, a mechanism which is associ-
ated with a reduction of the inductive effect [23].
Depending on the individual atom/group orientations, the ob-
served values may have positive or negative signs. As men-
tioned above (Section 3.3), we determined dispersion effects
D
m
D
m
in two different ways: in nonracemic mixtures (Experiment A)
and for single enantiomers (Experiment B). All these data are listed
in Table 6.
3.4. Differentiation of enantiomers by 1H and 13C NMR signal
dispersion effects (Dm)
The
types of experiments; even the signs of these parameters may
invert when going from Experiment A to and vice versa.
Fig. 8 shows an example of enantiodifferentiation by NMR signal
duplication of like-1e. In addition, a remarkable difference in
is visible here: in Experiment A [corresponding to the spectrum
section (d)] the dispersion (line distance) is +11 Hz whereas
Dm values of some nuclei differ considerably in the two
Chiral ligands form diastereomeric adducts with Rh*, and there-
fore, their complexations shifts are different; this difference is
called diastereomeric dispersion; data are given as integers in Hz
recorded at B0 = 9.4 Tesla. We use the following definitions in this
study:
B
D
m
D
m
D
D
m
m
¼
¼
m
m
ðRSRCÞ ꢁ
ðRSSCÞ ꢁ
m
ðSSSCÞ forlike-sulfoxides 1
in Experiment B [corresponding to the two spectrum sections
mðSSRCÞ forunlike-sulfoxides 2
(b) and (c)] it is ꢁ7 Hz.
A similar observation is made for most of the fluorine-substi-
tuted ligand compounds where some still unidentified effects seem
to reduce the complexation shifts when the ligands are enantio-
pure but not when they exist in mixtures [25].
Basically, dispersion effects are provoked by the anisotropy of
the aromatic groups in Rh* (ring-current effect) [24] and the differ-
ence of conformational equilibria in diastereomeric Rh*-ligand ad-
ducts. Thereby, the averaged orientation of the Mosher residue
phenyl groups is different in diastereomeric sulfoxide – Rh* ad-
ducts exerting diverging shielding effects to nearby nuclei.
The reason for these surprising differences is not entirely clear.
It may originate from accidental differences in concentrations,
Table 6
1H and 13C diastereomeric dispersion effects (
D
m
) of the like-sulfoxides 1 [
D
m
=
m
(RSRC) – m(SSSC)] and the unlike-sulfoxides 2 [Dm = m(RSSC) – m(SSRC)] after addition of an equimolar
amount of Rh*; in Hz at 9.4 Tesla (400.1 MHz 1H and 100.6 MHz 13C).a
1a/2a
X = H
(l)/(u)
1a/2a
X = H
(l)/(u)
1b/2b
X = F
(l)/(u)
1b/2b
X = F
(l)/(u)
1c/2c
X = Br
(l)/(u)
1c/2c
X = Br
(l)/(u)
1d/2d
X = NO2
(l)/(u)
1d/2d
X = NO2
(l)/(u)
1e/2e
X = OCH3
(l)/(u)
1e/2e
X = OCH3
(l)/(u)
Experiment A Experiment B Experiment A Experiment B Experiment A Experiment B Experiment A Experiment B Experiment A Experiment B
H-1
H-2
ꢁ5/ꢁ8
ꢁ4/+44
+16/+64
ꢁ10/+16
+8/+22
ꢁ12/ꢁ28
+32/+76
+20/ꢁ4
+24/+16
–
ꢁ27/+46
+26/+54
+11/ꢁ51
+2/ꢁ19
ꢁ4/+73
+20/+63
0/+14
+15/ꢁ4
n.d.b/n.d.b
n.d.b/n.d.b
n.d.b/n.d.b
+26/+13
ꢁ44/+49
ꢁ32/ꢁ38
–
ꢁ12/+28
ꢁ12/+12
+4/+4
ꢁ24/+32
+20/ꢁ20
ꢁ36/+56
+12/ꢁ12
–
ꢁ21/ꢁ36
n.d.b/n.d.b
ꢁ12/+24
+7/n.d.b
ꢁ21/+20
+36/ꢁ8
+36/ꢁ14
–
ꢁ4/ꢁ8
+48/ꢁ20
ꢁ4/ꢁ32
+12/ꢁ4
ꢁ16/ꢁ8
+40/+44
+48/+32
–
ꢁ23/+6
n.d.b/n.d.b
n.d.b/n.d.b
n.d.b/n.d.b
ꢁ20/+18
+27/ꢁ23
+34/ꢁ32
–
ꢁ20/ꢁ12
ꢁ12/ꢁ32
ꢁ20/ꢁ20
ꢁ8/ꢁ20
ꢁ12/ꢁ8
+16/+4
+24/+16
–
ꢁ14/+15
n.d.b/n.d.b
n.d.b/+12
n.d.b/n.d.b
ꢁ24/+12
ꢁ34/ꢁ33
+26/ꢁ27
–
ꢁ20/+28
+64/+28
ꢁ20/ꢁ16
+20/+64
+4/+8
n.d.b/n.d.b
ꢁ12/+14
+13/n.d.b
ꢁ5/+17
H-3ac
H-3bc
H-4
H-20/H-60 +18/ꢁ13
H-30/H-50 +14/ꢁ18
ꢁ32/+12
ꢁ8/+4
H-40
0/0
–
–
OCH3
C-1
C-2
C-3
C-4
C-10
C-20/C-60
C-30/C-50
C-40
–
–
–
–
–
–
+11/+10
ꢁ29/ꢁ10
+22/ꢁ14
+10/ꢁ3
ꢁ4/+4
ꢁ7/+1
ꢁ37/0
+15/0
0/ꢁ11
0/ꢁ5
ꢁ2//0
+6/ꢁ49
0/+7
0/0
+9/ꢁ7
ꢁ19/+43
ꢁ21/+45
+81/ꢁ45
ꢁ50/ꢁ7
+10/ꢁ9
ꢁ61/+61
ꢁ6/+13
ꢁ48/+28
–
ꢁ22/ꢁ13
+15/ꢁ9
+12/+20
ꢁ5/+3
ꢁ16/ꢁ3
+30/ꢁ4
ꢁ57/ꢁ35
ꢁ10/+3
ꢁ99/+107
+29/+2
+21/+49
+54/ꢁ28
–
ꢁ13/ꢁ5
+7/ꢁ2
+6/+8
ꢁ7/+4
+15/ꢁ8
+12/ꢁ6
+3/ꢁ5
+9/ꢁ7
–
ꢁ9/+1
+5/ꢁ2
+6/ꢁ3
0/0
ꢁ20/+1
+9/ꢁ4
+4/+4
+9/0
ꢁ13/+34
ꢁ13/+9
+10/+29
+4/+6
ꢁ8/+6
+54/ꢁ31
ꢁ9/+2
+16/ꢁ4
+34/ꢁ26
ꢁ33/+7
ꢁ8/+5
ꢁ15/n.d.b
+18/ꢁ10
+9/+9
0/0
+7/ꢁ10
ꢁ73/ꢁ13
ꢁ25/+21
ꢁ53/+7
ꢁ23/+15
+16/ꢁ1
0/0
+12/+50
–
+12/ꢁ10
+8/ꢁ8
OCH3
–
–
–
–
+4/ꢁ3
a
b
c
Experiment A: measurement with nonracemic mixtures, Experiment B: individual measurements of pure enantiomers.
Not detectable (n.d.) due to signal overlap; no safe individual signal assignments.
Diastereotopic atoms H-3a and H-3b with d(H-3a) < d(H-3b) in all derivatives; for stereochemical assignments see text.

J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
95
and specific optical rotations [
a
]D as well as by X-ray crystal-
lography of some of the products. By the combination of
these methods, an unequivocal differentiation between the
diastereomers (like and unlike) is easily possible.
(b) Equilibria of O- and S-adducts are formed when sulfoxides
are mixed with the dirhodium complex Rh*. If the alkyl res-
idue R is bulky (2-butyl), the equilibria are shifted towards
the O-adduct side but they move towards the S-adduct side
if R is small (methyl).
(c) The dirhodium method (enantiodifferentiation by adding
Rh*) is very effective in the case of sulfoxides whatever
the adduct formation mechanism may be. It can be applied
by comparing
Dm values of enantiomers in mix (Experiment
A) or, alternatively, in separate NMR experiments with each
of the stereoisomers (Experiment B). Comparing these exper-
iments, some dispersion effects may differ considerably in
magnitude and sign but this has no depriving effect on the
capability of the system to discriminate the enantiomers.
(d) The differences in the results from Experiments A and B may
be explained by the admixture of 2:1-adducts in the adduct
formation equilibria. In the case of Experiment B only one
stereoisomeric 2:1 adduct exists whereas a mixture of sev-
eral ones is formed in Experiment A.
Fig. 8. Methoxy 1H NMR signals of 4-methoxy-1(1-methylpropylthio)benzene-S-
oxide [(l)-1e]; (a) signal of the free ligand 1e, (b) enantiopure (RS,RC)-1e + Rh*
(equimolar), (c) enantiopure (SS,SC)-1e + Rh* (equimolar), and (d) nonracemic
mixture of (RS,RC)- and (SS,SC)-1e (2.6:1) + Rh* (equimolar).
(e) Due to the high conformational flexibility of the sulfoxides 1
and 2 and the variation in the equilibria of adducts (O- vs. S-
adducts), the dirhodium method is not safe for deriving
absolute configurations of sulfoxides from the signs of their
temperatures and magnetic susceptibilities of the solutions. A
more plausible source of this incongruity, however, is the stereo-
chemistry of the adduct formation equilibria. Although a descrip-
tion of equilibria involving only 1:1 adducts (L ? Rh–Rh; cf.
Fig. 7) is good enough for a semi-quantitative description of the
underlying phenomena, it is over-simplified since it ignores any
adducts with more than two components, as for example Rh*
bearing two ligand molecules L. In reality, such adducts (L ? Rh–
Rh L) exist even when there is an equimolar ratio of the compo-
nents Rh* and L [26]. If the ligand is chiral and a mixture of both
enantiomers is present – as in the case of Experiment A – several
diastereomeric 2:1-adducts are formed due to different (+)-L and
(ꢁ)-L combinations. These may differ in the averaged geometries
within the Mosher acid residues from those in Experiment B where
only one single diastereomer of 2:1-adduct exists, either two (+)-L
or two (ꢁ)-L ligands [22].
Dm values.
Acknowledgements
The authors gratefully acknowledge valuable help provided by
Dr. Michael Wiebcke and by Dr. S. Locmelis, Institute of Inorganic
Chemistry of the Leibniz University Hannover, in the X-ray diffrac-
tion experiments and the powder diffraction measurements,
respectively. This work was supported by the Deutsche Fors-
chungsgemeinschaft (Grant Du 98/34).
References
In summary, however, applying the dirhodium method for
enantiodifferentiation can be performed by both procedures, either
using mixtures of enantiomers or by measuring both pure enanti-
omers separately although the magnitudes and even the signs of
[1] (a) E.B. Boyar, S.D. Robinson, Coord. Chem. Rev. 50 (1983) 109;
(b) F.A. Cotton, R.A. Walton (Eds.), Multiple Bonds between Metal Atoms,
second ed., Clarendon, Oxford, 1993.
[2] (a) C. Mertis, M. Kravaritoy, M. Chorianopoulou, S. Koinis, N. Psaroudakis, Top.
Mol. Org. Eng. 11 (1994) 321;
dispersion effects
D
m
may vary for some 1H and 13C nuclei.
) can be recognized at
(b) M.P. Doyle, M.A. McKervey, T. Ye, Modern Catalytic Methods for Organic
Synthesis with Diazo Compounds: From Cyclopropanes to Ylides, Wiley, New
York, 1998;
(c) A. Endres, G. Maas, Tetrahedron 58 (2002) 3999;
(d) D.T. Nolan III, D.A. Singleton, J. Am. Chem. Soc. 127 (2005) 6190. and
references cited therein.
Remarkably large dispersions effects (
D
m
many 1H and 13C atoms (Table 6) indicating that chiral differentia-
tion of enantiomeric sulfoxides by NMR spectroscopy in the pres-
ence of Rh* (dirhodium method) is easy and safe. However, in
contrast to some other families of structurally related compounds
[3] M.J. Clarke, F. Zhu, D.R. Frasca, Chem. Rev. 99 (1999) 2511.
[4] K. Wypchlo, H. Duddeck, Tetrahedron: Asymm. 5 (1994) 27.
[5] H. Duddeck, Chem. Rec. 5 (2005) 396. and references cited therein.
[6] E. Díaz Gómez, D. Albert, H. Duddeck, S.I. Kozhushkov, A. de Meijer, Eur. J. Org.
Chem. (2006) 2278.
[7] H. Duddeck, E. Díaz Gómez, Chirality 21 (2009) 51.
[8] E. Díaz Gómez, H. Duddeck, Magn. Reson. Chem. 46 (2008) 23.
[9] D.V. Deubel, Organometallics 21 (2002) 4303.
studied by us earlier [23,27], the pattern of
Dm signs for corre-
sponding nuclei are not consistent so that these effects – despite
of their oftenly large magnitudes – cannot be used for correlating
absolute configurations of related sulfoxide derivatives. Appar-
ently, it is the conformational flexibility of these molecules which
leads to diverging atomic arrangements in the diastereomeric
adducts.
[10] J.T. Mattiza, V.J. Meyer, H. Duddeck, Magn. Reson. Chem. 48 (2010), in press.
[11] J.A. Davies, F.R. Hartley, Chem. Rev. 81 (1981) 79. and references cited therein.
_
[12] J. Drabowicz, P. Łyzwa, M. Popielarczyk, M. Mikołajczyk, Synthesis 10 (1990)
937.
[13] C.O. Kinen, L.I. Rossi, R.H. de Rossi, J. Org. Chem. 74 (2009) 7132.
[14] See for example: (a) B. Karimi, M. Ghoreishi-Nezhad, J.H. Clark, Org. Lett. 7
(2005) 625;
4. Conclusions
(b) M. Hirano, S. Yakabe, J.H. Clark, T. Morimoto, J. Chem. Soc., Perkin Trans. 1
(1996) 2693;
(a) Complete absolute configuration assignments of the diaste-
reomeric S-chiral sulfoxides with a chiral 2-butyl residue
were made by syntheses starting from enantiomerically
pure thioethers, by inspecting 1H and 13C chemical shifts
(c) A. Yamagishi, J. Chem. Soc. Chem. Commun. (1986) 290.
[15] Density functional calculations (B3LYP 6-31Gꢃ) were performed using the
´
SPARTAN 08 program package, version 1.0.0, Wavefunction Inc., Irvine, CA; see
also: Y. Shao, L.F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S.T. Brown, A.T.B.

96
J.T. Mattiza et al. / Journal of Molecular Structure 978 (2010) 86–96
Gilbert, L.V. Slipchenko, S.V. Levchenko, D.P. O’Neill, R.A. DiStasio Jr., R.C.
[21] See for example: J. Nyhlén, T. Privalov, Eur. J. Inorg. Chem. 19 (2009) 2759.
[22] D. Magiera, J. Omelanczuk, K. Dziuba, K.M. Pietrusiewicz, H. Duddeck,
Organometallics 22 (2003) 2464.
[23] S. Moeller, Z. Drzazga, Z. Pakulski, K.M. Pietrusiewicz, H. Duddeck, Chirality 18
(2006) 395.
[24] J.B. Lambert, E.P. Mazzola, Nuclear Magnetic Resonance Spectroscopy – An
Introduction to Principles, Applications, and Experimental Methods, Pearson
Education Inc., Upper Saddle River, NJ, 2004, pp. 62–66.
[25] Interestingly, the fluorinated ligands form insoluble complexes with Rh* when
the CDCl3 solution is stored for a few hours at room temperature, occasionally
even during the NMR measurements. This is not observed for any other
sulfoxide.
Lochan, T. Wang, G.J.O. Beran, N.A. Besley, J.M. Herbert, C.Y. Lin, T. Van Voorhis,
S.H. Chien, A. Sodt, R.P. Steele, V.A. Rassolov, P.E. Maslen, P.P. Korambath, R.D.
Adamson, B. Austin, J. Baker, E.F.C. Byrd, H. Dachsel, R.J. Doerksen, A. Dreuw,
B.D. Dunietz, A.D. Dutoi, T.R. Furlani, S.R. Gwaltney, A. Heyden, S. Hirata, C.-P.
Hsu, G. Kedziora, R.Z. Khalliulin, P. Klunzinger, A.M. Lee, M.S. Lee, W.Z. Liang, I.
Lotan, N. Nair, B. Peters, E.I. Proynov, P.A. Pieniazek, Y.M. Rhee, J. Ritchie, E.
Rosta, C.D. Sherrill, A.C. Simmonett, J.E. Subotnik, H.L. Woodcock III, W. Zhang,
A.T. Bell, A.K. Chakraborty, D.M. Chipman, F.J. Keil, A. Warshel, W.J. Hehre, H.F.
Schaefer, J. Kong, A.I. Krylov, P.M.W. Gill, M. Head-Gordon, Phys. Chem. Chem.
Phys. 8 (2006) 3172.
[16] R.J. Abraham, J.J. Byrne, L. Griffiths, Magn. Reson. Chem. 46 (2008) 667.
[17] For the nomenclature see: D. Seebach, V. Prelog, Angew. Chem. Int. Ed. Engl. 21
(1982) 654.
[26] Z. Rozwadowski, S. Malik, G. Tóth, T. Gáti, H. Duddeck, Dalton Trans. (2003)
375.
[18] H. Duddeck, Top. Stereochem. 16 (1986) 219. see pp. 252f.
[19] C.A.G. Haasnoot, F.A.A.M. de Leeuw, C. Altona, Tetrahedron 36 (1980) 2783.
[20] (a) F.A. Cotton, E.V. Dikarev, M.A. Petrukhina, S.-E. Stiriba, Inorg. Chem. 39
(2000) 1748;
[27] (a) T. Gáti, G. Tóth, J. Drabowicz, S. Moeller, E. Hofer, P.L. Polavarapu, H.
Duddeck, Chirality 17 (2005) S40;
(b) A.G. Petrovic, P.L. Polavarapu, J. Drabowicz, Y. Zhang, O.J. McConnell, H.
Duddeck, Chem. Eur. J. 11 (2005) 4257.
(b) C. Meyer, H. Duddeck, Magn. Reson. Chem. 38 (2000) 29.