Origin of 13C complexation shifts in the adduct formation of 2-butyl phenyl ethers with a dirhodium tetracarboxylate complex

Article abstract of DOI:10.1002/mrc.2116

Complexation of the oxygen atom in 2-butyl phenyl ethers to a rhodium atom of the dirhodium tetracarboxylate Rh^(II)₂[(R)-(+)-MTPA] ₄(Rh*, MTPA-H = methoxytrifluoromethylphenylaceticacid = Mosher's acid) deshields an sp³-hybridized ¹³C nucleus directly bonded to the ether oxygen; apparently, the inductive effect of the oxygen is enhanced when it is complexed to the rhodium atom. On the other hand, deshielding complexation shifts of aromatic ipso-carbons (α-positioned) are minute but ortho- and para-carbon signals are influenced by the resonance effect of oxygen. This effect can be modulated by further substituents at the benzene ring. In turn, this modulation of the resonance correlates linearly ith the magnitude of the inductive effect exerted on the aliphatic α-carbon atoms. Diastereomeric dispersion effects at ¹³C signals can be observed for most compounds, indicating that enantiodifferentiation is possible in this class of ethers. Copyright

Full text of DOI:10.1002/mrc.2116

Research Article
Received: 4 September 2007
Revised: 1 October 2007
Accepted: 1 October 2007
Published online in Wiley Interscience:
(www.interscience.com) DOI 10.1002/mrc.2116
Origin of 13C complexation shifts in the adduct
formation of 2-butyl phenyl ethers with a
dirhodium tetracarboxylate complex
∗
´
Edison Díaz Gomez and Helmut Duddeck
Complexation of the oxygen atom in 2-butyl phenyl ethers to a rhodium atom of the dirhodium tetracarboxylate Rh(II)2[(R)-(+)-
MTPA]4(Rh∗, MTPA-H = methoxytrifluoromethylphenylacetic acid ≡ Mosher’s acid) deshields an sp3-hybridized 13C nucleus
directly bonded to the ether oxygen; apparently, the inductive effect of the oxygen is enhanced when it is complexed to the
rhodium atom. On the other hand, deshielding complexation shifts of aromatic ipso-carbons (α-positioned) are minute but
ortho- and para-carbon signals are influenced by the resonance effect of oxygen. This effect can be modulated by further
substituents at the benzene ring. In turn, this modulation of the resonance correlates linearly ith the magnitude of the inductive
effect exerted on the aliphatic α–carbon atoms. Diastereomeric dispersion effects at 13C signals can be observed for most
c
compounds, indicating that enantiodifferentiation is possible in this class of ethers. Copyright ꢀ 2008 John Wiley & Sons, Ltd.
1
Keywords: 2-butyl phenyl ethers; dirhodium complexes; complexation; 13C NMR; H NMR; inductive effect; resonance effect; chiral
differentiation
Introduction
not contribute significantly if oxygen as a second-row element
is involved; the HOMO-LUMO gap is too large.[6] We chose the
2-butyl group as the aliphatic part and the phenyl group as the
aromatic residue, both attached to oxygen. The chiral 2-butyl
group (in racemic form) offers carbon-hydrogen fragments of
Dirhodium complexes as well as their adducts are in the focus of
interest for many years.[1] They were introduced as homogeneous
catalysts in various reactions[2] and have found even medicinal
application.[3] During the last decade, we have shown that the
enantiomers of many chiral ligands, particularly those of soft
Lewis bases, can easily be differentiated by adding an equimolar
amount of the dirhodium complex Rh(II)2[(R)-(+)-MTPA]4 ([Rh∗],
(MTPA-H) = methoxytrifluoromethylphenylacetic acid ≡ Mosher’s
acid; see Scheme 1) to their CDCl3 solution and monitoring the
diastereomeric dispersion ꢀν of their 1H (or 13C) NMR signals at
room temperature (dirhodium method).[4]
1
different compositions, and their H and 13C signals are easy to
identify. Moreover, the chirality allows monitoring the potential
for enantiodifferentiation in these ethers, in addition. On the
other hand, the electronic interaction inside the phenyl group can
easily be modulated by attaching substituents X. Thus, a series of
2-butyl phenyl ethers (Scheme 3) were subjected to the dirhodium
experiment.
Thecomplexationsiteoftheligandmoleculecanbedetermined
by moderate deshieldings of nearby 1H and – particularly – 13
C
Experimental
nuclei (complexation shifts ꢀδ). In a qualitative interpretation of
positive ꢀδ values, one can assume an increase of the electron-
acceptorpropertiesoftheatomwhichisthebindingsite(inductive
effect).[4] Recently, however, when investigating oxygen ligands
which were attached to aromatic ring systems, we encountered
the observation that significant complexation shifts may be found
on aromatic atoms beyond the ipso-carbon bound to oxygen. For
example, complexation shifts in cyclotriveratrylenes containing
three methoxy groups (Scheme 2) are much stronger at the two
ortho- and the at the para-carbon atoms than at the methoxylated
carbon.[5] This, however, is not compatible with an explanation
based on an inductive effect of the oxygen but rather reminds of
a resonance effect.
Substances
The synthesis of Rh∗ has been described by us earlier.[7]
The 2-butyl phenyl ethers 1–5d were prepared as racemates
by a nucleophilic substitution reaction of 2-bromobutane and
the respective phenolates (Williamson’s ether synthesis). General
procedure: 20 mmol K2CO3 (2.76 g) was suspended in 20 ml
acetone,andthen22 mmol2-iodobutane(4.0 g,2.5 ml)wasadded
followed by 20 mmol of the respective, commercially available
phenol. The mixture was refluxed under stirring for 48 h, acetone
distilled off with a rotary evaporator and the residue poured into
20 ml water. The aqueous layer was extracted twice with 2 ml
So, we decided to conduct a more detailed study on this
phenomenon, i.e. the fact that complexations shifts seem
to be entirely different if aliphatic or aromatic carbons are
involved. Ethers were chosen as suitable candidates because – as
hard Lewis bases – they form rather weak adducts with Rh∗.[4]
The binding energy is expected to be based primarily on
electrostatic interaction; orbital interaction (HOMO-LUMO) should
∗
Correspondence to: Helmut Duddeck, Leibniz University Hannover, Institute of
Organic Chemistry, Schneiderberg 1B, D-30167 Hannover, Germany.
E-mail: duddeck@mbox.oci.uni-hannover.de
Leibniz University Hannover, Institute of Organic Chemistry, Schneiderberg 1B,
D-30167 Hannover, Germany
c
Magn. Reson. Chem. 2008; 46: 23–29
Copyright ꢀ 2008 John Wiley & Sons, Ltd.

E. D. Go´mez and H. Duddeck
2-Butyl phenyl ether, (1-methylpropoxy)benzene (1)
1H chemical shifts δ (ppm), signal multiplicities m and complex-
ation shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.29, d (+0.02);
H-2, 4.29, tq (+0.03); H-3a, 1.61, ddq (+0.01); H-3b, 1.75, ddq
(+0.03); H-4, 0.97, t (+0.01); H-2ꢁ/6ꢁ, 6.89, m (+0.03); H-3ꢁ/5ꢁ, 7.26,
m (+0.01); H-4ꢁ, 6.90, m (+0.04). IR (liquid) ν˜ (cm−1) 2971, 2934,
1598, 1586, 1492, 1290, 1239, 923, 749, 691. EI-MS (70 eV, rel. int. %)
m/z 150 (19, M+), 121 (20), 95 (12), 94 (100), 77 (15), 66 (13), 65 (10).
1-Fluoro-4(1-methylpropoxy)benzene (2a)
1H chemical shifts δ (ppm), signal multiplicities m and complex-
ation shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.27, d (+0.04);
H-2, 4.19, tq (+0.06); H-3a, 1.60, ddq (+0.01); H-3b, 1.73, ddq
(+0.05); H-4, 0.97, t (+0.01); H-2ꢁ/6ꢁ, 6.82, m (+0.06); H-3ꢁ/5ꢁ, 6.95,
m (−0.01); IR (liquid) ν˜ (cm−1) 2973, 2935, 2880, 1600, 1501, 1378,
1245, 1205, 1104, 825, 746, 730. EI-MS (70 eV, rel. int. %) m/z 168
(13, M+), 139 (6), 112 (100), 95 (9), 83 (12), 75 (7), 57 (13).
Scheme 1. Structure of the dirhodium complex Rh∗.
1-Chloro-4(1-methylpropoxy)benzene(2b)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.27, d (0); H-2, 4.24,
tq (+0.03); H-3a, 1.60, ddq (+0.01); H-3b, 1.73, ddq (0); H-4, 0.96,
t (+0.01); H-2ꢁ/6ꢁ, 6.81, m (+0.03); H-3ꢁ/5ꢁ, 7.21, m (0); IR (liquid)
ν˜ (cm−1) 2973, 2935, 1595, 1579, 1487, 1378, 1238, 1091, 823, 664.
EI-MS (70 eV, rel. int. %) m/z 184/186 (10/3, M+), 128/130 (100/31),
111/113 (6/2), 100/102 (4/1), 99/101 (13/4), 65 (9).
Scheme 2. Structure of a cyclotriveratrylene derivative containing three
methoxy groups.
toluene, and the combined organic layers were washed twice with
aqueous sodium hydroxide (10%) and with water. After removal
of the solvent, the crude product was distilled in vacuo. The ethers
were obtained as colorless or slightly yellow liquids; yields were
60–75% (not optimized). Purification by chromatography on silica
gel should be avoided because it leads to massive losses probably
due to a H+-catalyzed phenol elimination.
Most of the ethers, namely 1,[8] 2a,[9] 2b,[10] 2c,[11] 2d,[12] 2e,[13]
3a,[9] 3b,[14] 3c,[15] and 5d,[16] are described in the literature; 3d,
4c and 5c are new. In nearly all cases, spectral data are not well
documented in the literature. Therefore, we collect here 1H NMR
chemical shifts, infrared and electron-impact mass spectral data
of all derivatives. IR spectra were recorded on a Bruker Vector 22
spectrometer and EI-MS spectra on a Micromass LCT instrument.
1-Bromo-4(1-methylpropoxy)benzene (2c)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.27, d (+0.02); H-2,
4.24, tq (+0.03); H-3a, 1.60, ddq (+0.01); H-3b, 1.73, ddq (+0.03);
H-4, 0.96, t (0); H-2ꢁ/6ꢁ, 6.77, m (+0.02); H-3ꢁ/5ꢁ, 7,35, m (0); IR
(liquid) ν˜ (cm−1) 2972, 2934, 2878, 1589, 1485, 1282, 1237, 1072,
820. EI-MS (70 eV, rel. int. %) m/z 228/230 (11/11, M+), 172/174
(100/93), 143/145 (9/9), 93 (14), 75 (7), 65 (16), 63 (14), 57 (8).
1-Iodo-4(1-methylpropoxy)benzene (2d)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.27, d (0); H-2, 4.24,
Scheme 3. Structures of the 2-butyl phenyl ethers studied.(The nomenclature and atom numbering of the substituted 2-butyl phenyl ethers differs
from that of the unsubstituted 1; e.g. 2a is 1-flouro-4(1-methylpropoxy)benzene. For a better comparability of NMR data, however, we used the atom
numbering of 1 for all derivatives 2a to 5d.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2008; 46: 23–29

Origin of 13C complexation shifts
tq (−0.01); H-3a, 1.60, ddq (−0.01); H-3b, 1.72, ddq (+0.01); H-4,
0.96, t (−0.01); H-2ꢁ/6ꢁ, 6.66, m (−0.03); H-3ꢁ/5ꢁ, 7.53, m (+0.04);
IR (liquid) ν˜ (cm−1) 2971, 2932, 2877, 1583, 1482, 1279, 1237, 998,
818. EI-MS (70 eV, rel. int. %) m/z 276 (23, M+), 220 (100), 112 (39),
93 (27), 76 (10), 65 (23), 57 (9).
1,3-Dibromo-6(1-methylpropoxy)benzene(4c)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.32, d (+0.01); H-2,
4.29, tq (0); H-3a, 1.67, ddq (+0.01); H-3b, 1.78, ddq (+0.02); H-4,
1.00, t (0); H-3ꢁ, 7.66, dd (+0.01), H-5ꢁ, 7.33, dd (0); H-6ꢁ, 6.76, d
(+0.01). IR (liquid) ν˜ (cm−1) 2972, 2933, 2878, 1577, 1467, 1379,
1282, 1260, 1243, 1094, 1042, 801, 690. EI-MS (70 eV, rel. int. %) m/z
306/308/310 (4/8/4, M+), 250/252/254 (55/100/56), 221/223/225
(4/6/4), 128 (12), 63 (15), 57 (8).
1-Nitro-4(1-methylpropoxy)benzene(2e)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.34, d (0); H-2, 4.43,
tq (0); H-3a, 1.68, ddq (0); H-3b, 1.79, ddq (0); H-4, 0.99, t (0); H-2ꢁ/6ꢁ,
6.92, m (0); H-3ꢁ/5ꢁ, 8.18, m (+0.01); IR (liquid) ν˜ (cm−1) 2974, 2936,
1591, 1509, 1493, 1338, 1255, 1177, 1108, 920, 844, 752, 691. EI-MS
(70 eV, rel. int. %) m/z 195 (29, M+), 166 (10), 140 (100), 139 (86),
123 (31), 109 (100), 93 (26), 81 (19), 76 (22), 65 (44).
1,3,5-Tribromo-6(1-methylpropoxy)benzene(5c)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.28, d (0); H-2, 4.54,
tq (0); H-3a, 1.72, ddq (0); H-3b, 1.85, ddq (0); H-4, 1.00, t (0); H-3ꢁ/5ꢁ,
7.65, s (0). IR (liquid) ν˜ (cm−1) 2968, 2934, 2876, 1559, 1535, 1436,
1372, 1246, 1088, 899, 857, 739, 727. EI-MS (70 eV, rel. int. %) m/z
384/386/388/390 (1/2/2/1, M+), 369/371/373/375 (0.5/1/1/0.5),
355/357/359/361 (2/4/4/2), 328/330/332/334 (41/100/98/38),
229/301/303/305 (3/6/6/3), 248/250/252/254 (3/6/6/3), 143 (13),
141 (13), 74 (6), 62 (16), 57 (12).
1-Fluoro-2(1-methylpropoxy)benzene(3a)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.31, d (+0.06); H-2,
4.28, tq (+0.09); H-3a, 1.64, ddq (+0.04); H-3b, 1.80, ddq (+0.09);
H-4, 1.00, t (0); H-3ꢁ, 7.06, m (0.0), H-4ꢁ, 6.88, m (+0.06); H-5ꢁ, 7.03, m
(+0.03); H-6ꢁ, 6.97, m (+0.07). IR (liquid) ν˜ (cm−1) 2973, 2936, 2879,
1612, 1501, 1457, 1257, 1110, 744. EI-MS (70 eV, rel. int. %) m/z 168
(7, M+), 139 (13), 121 (6), 112 (100), 94 (26), 83 (9), 64 (12), 57 (12).
1,3,5-Triiodo-6(1-methylpropoxy)benzene(5d)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.30, d (−0.01); H-2,
4.62, tq (0); H-3a, 1.76, ddq (0); H-3b, 1.88, ddq (−0.01); H-4, 1.00,
t (−0.01); H-3ꢁ/5ꢁ, 8.06, s (+0.02). IR (liquid) ν˜ (cm−1) 2964, 2930,
1517, 1413, 1377, 1238, 1085, 895, 859, 721, 693. EI-MS (70 eV, rel.
int. %) m/z 528 (8, M+), 472 (100), 346 (49), 218 (27), 189 (20), 127
(5), 91 (10), 62 (26).
1-Chloro-2(1-methylpropoxy)benzene(3b)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.33, d (0); H-2, 4.32,
tq (+0.01); H-3a, 1.67, ddq (+0.01); H-3b, 1.80, ddq (+0.02); H-4,
1.01, t (0); H-3ꢁ, 7.35, m (+0.01), H-4ꢁ, 6.86, m (+0.02); H-5ꢁ, 7.18, m
(+0.01); H-6ꢁ, 6.93, m (+0.02). IR (liquid) ν˜ (cm−1) 2973, 2936, 2879,
1612, 1501, 1457, 1257, 1110, 744.IR (liquid) (cm−1) 2973, 2935,
2878, 1588, 1478, 1445, 1378, 1273, 1245, 1059, 924, 744, 690.
EI-MS (70 eV, rel. int. %) m/z 184/186 (6/2, M+), 128/130 (100/34),
112 (7), 92 (6), 75 (6), 64 (9).
NMR spectroscopy
Room-temperature 1H (400.1 MHz) and 13C (100.6 MHz) NMR
measurements were performed on a Bruker Avance DPX-400
spectrometer. Samples were ca 0.01–0.025 mmolar in CDCl3.
Chemical shift standard was internal tetramethylsilane (δ = 0).
Following parameters have been used for all one-dimensional
NMR spectra. 1H: acquisition time 4.0 s, relaxation delay 0.5 s, pulse
duration 2.6 µs for a 30◦ flip angle, and spectral width 8224 Hz
(20.6 ppm); 64K points were used for data acquisition, 64K points
for FT transformation and digital resolution was 0.12 Hz/point. 13C:
acquisition time 2.6 s for a 30◦ flip angle, relaxation delay 0.5 s,
pulse duration 2.3 µs, and spectral width 25.629 Hz (250 ppm);
128K points were used for data acquisition, 128K points for FT
transformation and digital resolution was 0.19 Hz/point.
Signal assignments were assisted by DEPT90 and DEPT135,
COSY, HMQC and HMBC spectra (standard Bruker software and
parameters):
1-Bromo-2(1-methylpropoxy)benzene(3c)
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.33, d (0); H-2, 4.33,
tq (+0.01); H-3a, 1.68, ddq (+0.01); H-3b, 1.80, ddq (+0.02); H-4,
1.01, t (0); H-3ꢁ, 7.53, m (+0.01), H-4ꢁ, 6.80, m (+0.01); H-5ꢁ, 7.22, m
(+0.01); H-6ꢁ, 6.90, m (+0.02). IR (liquid) ν˜ (cm−1) 2973, 2936, 2879,
1612, 1501, 1457, 1257, 1110, 744.IR (liquid) (cm−1) 2972, 2934,
2878, 1585, 1474, 1441, 1378, 1272, 1244, 1029, 924, 743, 663.
EI-MS (70 eV, rel. int. %) m/z 228/230 (8/7, M+), 172/174 (100/95),
143/145 (9/8), 92/94 (5/5), 63 (16), 57 (10).
1
1
Gradient-selected H, H COSY spectra: relaxation delay D1
=
1-Iodo-2(1-methylpropoxy)benzene (3d)
1.2 s; 90◦ pulse for 1H: 9.6 µs; 1024 points in t2; 256 experiments in
t1, linear prediction to 512 points, zero-filling up to 1K.
Gradient-selected HMQC spectra: relaxation delay D1 = 1.5 s,
evolution delay D2 = 3.45 ms, 90◦ pulse: 9.6 µs for 1H, 12.1 µs for
13C hard pulses and 66.0 µs for 13C GARP decoupling; 1K points in
t2; 256 experiments in t1, linear prediction to 512 and zero-filling
up to 1K.
1H chemical shifts δ (ppm), signal multiplicities m and complexa-
tion shifts ꢀδ (in parentheses), in CDCl3: H-1, 1.34, d (0); H-2, 4.35,
tq (+0.01); H-3a, 1.69, ddq (+0.01); H-3b, 1.79, ddq (+0.01); H-4,
1.02, t (−0.01); H-3ꢁ, 7.77, m (+0.05), H-4ꢁ, 6.67, m (−0.06); H-5ꢁ,
7.26, m (0); H-6ꢁ, 6.80, m (+0.01). IR (liquid) ν˜ (cm−1) 2973, 2936,
2879, 1612, 1501, 1457, 1257, 1110, 744. IR (liquid) (cm−1) 2971,
2932, 2876, 1580, 1467, 1270, 1243, 1126, 1016, 923, 743. EI-MS
(70 eV, rel. int. %) m/z 276 (18, M+), 220 (100), 112 (20), 93 (16), 76
(7), 65 (20).
Gradient-selected HMBC spectra: relaxation delay D1 = 1.5 s;
evolution delay D2 = 3.45 ms; 90◦ pulse: 9.6 µs for 1H, 12.1 µs
for 13C hard pulses, delay for evolution of long-range coupling
c
Magn. Reson. Chem. 2008; 46: 23–29
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

E. D. Go´mez and H. Duddeck
D6 = 70 ms (J = 7 Hz); 1K points in t2; 256 experiments in t1, linear
prediction to 512 and zero-filling up to 1K.
are not reported here because they are very close to those
listed in a comprehensive review[17] and in standard spectroscopy
textbooks.) As expected, most calculated values correspond well
to the experimental ones (within 1 ppm). Exceptions (underlined
italicized values) appear for the substituted carbons if there is at
least one halogen in ortho-position with respect to oxygen; this
is due to steric repulsion and, eventually, electronic through-
space interaction between O and X. Two more exceptions are
observed, e.g. for C-4ꢁ in the para-F- and para-I derivatives 2a and
2d. Altogether, there is a good additivity for the unsubstituted
carbons proving that the individual electronic properties of the
alkoxy and the X substituents are not modified significantly by
double substitution (OR and X). This additivity is the basis for
the later discussion about how to identify the two mechanisms
(induction and resonance) and about their interdependence.
If, however, three halogen atoms are introduced at the
aromatic ring (5c and 5d), two of them in ortho-position, severe
nonadditivity effects occur (Table 1) but this is of no importance in
the context of this study because both compounds do not show
any change in their 1H- and 13C-chemical shifts and no measurable
ꢀν values. Apparently, a double ortho-substitution averts any
approach of the oxygen atom to Rh∗ close enough for adduct
formation. Therefore, these two compounds are ignored in the
following.
In the standard dirhodium experiment, Rh∗ and equimolar
amountsoftheligands1–5d,respectively,weredissolvedin0.7 ml
CDCl3; quantities of 10–25 mg of Rh∗ (ca 0.01–0.025 mmolar
concentration) were employed. If necessary, the dissolution
process was accelerated by exposing the NMR sample tubes
to an ultrasonic bath for a couple of minutes. In earlier reports on
soft-baseligands,theuseofacetone-d6 forincreasingthesolubility
of Rh∗ has been recommended.[4] This auxiliary, however, should
be avoided when hard-base ligands such as ethers are involved
because acetone-d6 is a competitor in the adduct formation in
such cases.
Note that ꢀν values are B0 dependent and have no signs
here because racemates have been investigated. In this work, all
dispersion values are given as integers in hertz as determined at
B0 = 9.4 T corresponding to 400 MHz 1H and 100.6 MHz 13C.
Results and Discussion
The complete and unambiguous NMR signal assignment of the
free ligands 1–5d (Scheme 3) is straightforward when routine
NMR methods such as DEPT, COSY, HMQC and HMBC techniques
are applied. For the 1H chemical shifts see ‘Experimental’, and
for 13C see Table 1. In the case of Rh∗ adducts the identification
of some ligand signals is hampered by overlapping Mosher acid
signals when aromatic signals are involved.
13C complexation shifts (ꢀδ) of the 2-butyl phenyl ethers 1–4c
Although ethers are known to be weak ligands in an adduct
formation equilibrium (Scheme 4; equilibrium constant K ≈ 1),[4]
significant complexation shifts ꢀδ [ꢀδ = δ(Rh∗-adduct) −
δ(free ligand)] and if the ether molecule is chiral, signal dispersions
ꢀν due to the existence of diastereomeric adducts can be
observed under standard dirhodium method conditions (1 : 1
molar ratio of Rh∗ and the ligand).[5,18] As tested for 1,
other molar ratios, e.g. 2.5 : 1, afford practically the same NMR
spectroscopic results. Apparently, the concentration of the adduct
in the adduct formation equilibrium (Scheme 4) is – within certain
13C chemical shifts (δ) of the 2-butyl phenyl ethers 1–5d
All 13C NMR signals of all free ligands 1–5d are collected in the
Table 1.Itshouldbenotedthatthevaluesinitalics(forthearomatic
carbons) were obtained from calculations using a simple additivity
rule: increments for the substituents X have been extracted from
the 13C NMR spectra of the corresponding monosubstituted
benzenes recorded under identical conditions. (These increments
Table 1. 13C chemical shifts (δ) of the 2-butyl phenyl ethers 1–5d in ppm, recorded in CDCl3. Values in italics are calculated by the incremental rule
discussed in the text
C
1
2aa
2b
2c
2d
2e
3ab
3b
3c
3d
4c
5c
5d
1
2
3
4
1ꢁ
19.26
74.92
29.19
9.80
158.22
–
19.19
76.09
29.12
9.75
19.12
75.52
29.07
9.72
19.10
75.42
29.05
9.72
19.09
75.26
29.05
9.71
18.93
75.94
28.93
9.57
19.31
77.48
29.19
9.73
19.26
77.00
29.26
9.70
19.22
76.92
29.15
9.69
19.19
76.76
29.16
9.77
19.08
77.34
29.05
9.61
19.09
81.93
29.45
9.85
19.24
82.10
29.55
10.02
157.14
174.9
87.97
85.2
154.29
153.7
156.83
156.1
157.33
156.5
158.12
157.1
118.26
117.6
138.20
138.3
82.25
86.3
163.51
164.2
146.14
145.0
153.93
158.3
154.77
161.2
113.56
99.9
156.88
167.1
88.25
81.8
154.11
158.2
151.84
162.5
2ꢁ
3ꢁ
4ꢁ
5ꢁ
6ꢁ
115.90
–
117.21
117.3
117.19
117.1
117.68
117.4
115.14
116.6
153.80
150.3
124.17
121.6
112.65
111.4
116.37
112.9
129.41
–
115.73
116.2
129.28
129.5
132.22
132.4
125.90
124.3
116.37
116.2
130.40
129.5
133.46
132.4
121.62
121.9
128.21
127.7
115.29
117.4
139.55
138.3
122.18
122.1
129.21
128.3
113.75
117.6
135.59
135.4
135.14
133.7
147.52
146.1
93.42
89.7
120.39
–
157.11
154.8
125.18
126.1
112.43
114.4
140.93
140.0
121.25
121.8
121.24
121.6
114.42
115.9
119.65
117.4
129.41
–
115.73
116.2
129.28
129.5
132.22
132.4
138.20
138.3
118.26
117.6
125.90
124.3
124.14
124.9
127.48
126.3
131.02
130.7
135.14
133.7
147.52
146.1
87.97
85.2
115.90
–
117.21
117.3
117.19
117.1
117.68
117.4
115.14
116.6
117.75
117.3
115.72
117.1
116.31
118.9
116.37
112.9
a 1J(19F, 13C) = 238.0 Hz, 2J(19F, 13C) = 22.9 Hz, 3J(19F, 13C) = 7.9 Hz, 4J(19F, 13C) = 2.2 Hz.
b 1J(19F, 13C) = 245.2 Hz, 2J(19F, 13C) = 18.8 Hz (C-3ꢁ) and 10.5 (C-1ꢁ), 3J(19F, 13C) = 7.1 Hz (C-4ꢁ) and 3.8 (C-6ꢁ), 4J(19F, 13C) = 2.1 Hz.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2008; 46: 23–29

Origin of 13C complexation shifts
Scheme 4. Equilibrium between free components and the 1 : 1-adduct; O
symbolizes the ether ligands 1–5d and ‘[Rh-Rh]’ the binuclear complex
Rh∗.
Figure 2. 13C complexation shifts (ꢀδ) of the aromatic carbon atoms C-
2ꢁ/6ꢁ in the compounds 2a–2e (•) and C-6ꢁ in the compounds 3a–3d
(ꢀ) (in ppm) plotted against the 13C complexation shifts of the aliphatic
carbon atoms C-2 in the respective derivatives.
Figure 1. 13C complexation shifts (ꢀδ) of the atoms C-2ꢁ/6ꢁ in the
compounds 2a–2e (•) and C-6ꢁ in the compounds 3a–3d (ꢀ) (in ppm)
plotted against the σR0 parameters of the substituents X.
small; iodine and the nitro group are even less effective, i.e. they
weaken the donor property of oxygen as compared to 1.
Unexpectedly, C-1ꢁ nuclei, the oxygenated aromatic carbons, do
no suffer significantly from electron-acceptor property changes
of oxygen; all ꢀδ-values are only between −0.15 (2a) and +0.14
(2d). On the other hand, carbons in ortho- and para-position with
respect to oxygen are markedly deshielded by up to 1.38 ppm
(3a), and – as for C-2 (see above) – there is a clear dependence on
the nature of the substituents X. This evidence suggests that it is
the resonance effect of X that modulates the complexation shifts
of oxygen via the resonance properties of oxygen.
A great number of substituent parameters can be found in
the literature to quantify such effects: Hammett σ constants,
parameters from dual or even triple substituent parameter
analyses and others.[19] The parameter σR0 describing essentially
the resonance properties of substituents attached to benzene is
most suitable in the context of this study. Figure 1 shows that
there is a fair correlation when the ꢀδ values of unsubstituted
aromatic carbons in ortho-position to oxygen in the series 2a–3d
are plotted against the σR0 values of X; the C-6ꢁ value of 4c with its
two bromine atoms at C-2ꢁ and C-4ꢁ (+0.24 ppm) is very close to
the respective one of the 2ꢁ-monobromo derivative 3c.
limits – independent of changes of the relative concentrations of
thecomponents.Thisprovesthattheequilibrium,indeed,contains
a considerable amount of the free components and is not at all
biased in favor of the adduct.
All ꢀδ(13C) values obtained for the ethers 1–4c are collected
in Table 2; the data of the trihalo derivatives 5c and 5d
have been omitted because their spectra give no indication
of adduct formation. It is apparent that complexation shifts
larger than +0.1 ppm appear at selected carbon atoms only.
Among the aliphatic carbons, it is only C-2 which is affected
(ꢀδ = +0.16 − +2.05 ppm), whereas all others do not show any
significantchange.Obviously,thiscanbeinterpretedintermsofan
increase of the electron-acceptor properties of oxygen by rhodium
complexation but it is striking that substituents X at the aromatic
ring have an influence as well: fluorine (2a and 3a) enhances
the group electronegativity considerably – and thereby the donor
strength in the adduct – as compared to the unsubstituted parent
compound 1. The effects of chlorine and bromine substitution are
Table 2. 13C complexation shifts (ꢀδ) of the 2-butyl phenyl ethers in ppm in the presence of an equimolar amount of Rh∗, recorded in CDCl3
C
1
2a
2b
2c
2d
2e
3a
3b
3c
3d
4c
1
+0.02
+0.65
−0.02
+0.03
+0.03
+0.42
+0.05
+0.04
+0.05
+0.42
+0.04
+1.30
−0.08
+0.09
−0.15
+0.84
−0.03
+0.27
−0.03
+0.84
+0.02
+0.77
−0.04
+0.05
−0.07
+0.46
−0.01
+0.27
−0.01
+0.46
+0.03
+0.72
−0.03
+0.04
−0.03
+0.44
+0.01
+0.28
+0.01
+0.44
−0.02
+0.49
−0.05
+0.01
+0.14
+0.31
−0.05
+0.53
−0.05
+0.31
+0.06
+0.16
+0.04
+0.05
+0.07
+0.11
+0.10
+0.10
+0.10
+0.11
−0.05
+2.05
−0.14
+0.09
−0.09
+0.45
+0.15
+0.50
−0.08
+1.38
−0.05
+0.39
−0.15
0.00
−0.04
+0.27
−0.04
−0.01
+0.06
+0.12
+0.06
+0.02
+0.01
+0.22
−0.06
+0.45
−0.04
−0.04
−0.05
+0.45
+0.01
+0.07
+0.19
+0.18
−0.03
+0.32
−0.04
0.00
2
3
4
1ꢁ
2ꢁ
3ꢁ
4ꢁ
5ꢁ
6ꢁ
+0.05
+0.14
+0.04
+0.04
−0.01
+0.29
+0.04
+0.10
+0.01
+0.10
+0.02
+0.24
c
Magn. Reson. Chem. 2008; 46: 23–29
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

E. D. Go´mez and H. Duddeck
Figure 3. HOMO of the O-methyl analogue of 2c calculated by density
functional methods (B3LYP 6–31G∗) using the SPARTAN ’06 package,
version 1.1.0.[20]
.
This indicates that, indeed, the resonance effect of oxygen
dominates its complexations shifts in ortho- and para-positions,
and this effect is modulated by resonance effects of substituents X
attached to these carbons. Moreover, the extent of this resonance
modulation, exerted from X on O, is transferred to the inductive
effects of oxygen onto the directly bonded C-2. Figure 2 shows an
excellent correlation for the para-substituted ethers of the series
2a–2e; the correlation of the series 3a–3d is not as good but still
satisfactory.
Figure 4. Signal dispersion ꢀν of the H-1 signal of 1-fluoro-4(1-
methylpropoxy)benzene (2a); bottom: free ligand, doublet due to 3J(H-
1,H-2); top: in the presence of one mole equivalent Rh∗.
(Scheme 4) is somewhat increased by fluorine so that splittings
of the time-averaged NMR signals are less ‘diluted’. This is
in line with the above-mentioned observation that these two
derivatives also show the strongest complexation shifts (see
above). Both evidences support the enhancement of the oxygen
donor properties by fluorine.
Some 1H NMR signals display minute dispersions but this is
basically confined to the fluorinated derivatives 2a and 3a again;
see for example Fig. 4.
Other aromatic carbon atoms substituted by X, e.g. C-4ꢁ of
2a–2e or C-2ꢁ of 3, show similar tendencies but the correlations
are much less clear.
As a conclusion, one can deduce from these results that
complexation shifts of oxygen ligands, produced by adduct
formation with dirhodium complexes such as Rh∗, are essentially
inductive in nature if directly attached aliphatic carbons are
involved. In contrast, it is the resonance effect that rules the
complexation shifts in the aromatic part of the ethers. Both
mechanisms are not independent; there is an influence of the
resonance effect on the inductive one via the intervening oxygen
atom. Such interaction is plausible; consider that the HOMOs
of para-substituted alkyl phenyl ethers, as shown in Fig. 3 for
the O-methyl analogue of 2c, are extended from the nπ -orbital
of O to the nπ -orbital of X.[20] Remember that the HOMO-
LUMO gap is dominating the mean excitation energy in the
expression describing the paramagnetic term σp of the 13C nuclear
shielding.[21]
References
[1] (a) Boyar EB, Robinson SD. Coord. Chem. Rev. 1983; 50: 109;
(b) Cotton FA, Walton RA (eds). Multiple Bonds between Metal Atoms
(2nd edn). Clarendon: Oxford, 1993.
[2] (a) Mertis C,
Kravaritoy M,
Chorianopoulou M,
Koinis S,
Psaroudakis N. Top. Mol. Organ. Eng. 1994; 11: 321; (b)Doyle MP,
McKervey MA, Ye T (eds). Modern Catalytic Methods for Organic
Synthesis with Diazo Compounds: From Cyclopropanes to Ylides.
Wiley: New York, 1998; (c) Endres A, Maas G. Tetrahedron 2002; 58:
3999; (d) Nolan DT III, Singleton DA. J. Am. Chem. Soc. 2005; 127:
6190 and references cited therein.
[3] Clarke MJ, Zhu F, Frasca DR. Chem. Rev. 1999; 99: 2511.
[4] Duddeck H. Chem. Rec. 2005; 5: 396 andreferences cited therein.
[5] Díaz-Go´mez E, Brotin T, Duddeck H. Tetrahedron: Asymmetry 2007;
18: 2155.
We are interested to see whether analogous relations exist
for atoms other than oxygen when they act as binding sites to
rhodium. Further studies to investigate soft Lewis-base atoms, as
forexamplesulfurorseleniumbindingmuchstrongerthanoxygen
and involving HOMO-LUMO interaction in the complexation, are
currently in progress in our laboratory.
[6] Deubel DV. Organometallics 2002; 21: 4303.
[7] Wypchlo K, Duddeck H. Tetrahedron: Asymmetry 1994; 5: 27.
[8] Bergman J, Norrby P-O, Sand P. Tetrahedron 1990; 46: 6113.
[9] Stavber S, Zupan M. J. Chem. Soc., Chem. Commun. 1981; 148.
[10] Hirano M, Yakabe S, Monobe H, Clark JH, Takashi M. J. Chem. Soc.,
Perkin Trans. 1 1997; 3081.
[11] Mori T, Inoue Y. J. Phys. Chem. A 2005; 109: 2728.
[12] Takeuchi K, Sasaki M, Takatsu H. Chem. Abstr. 1987; 109: 46346, Eur.
Pat. Appl., 38EP 249933 A1 19871223.
[13] Paradisi C, Quintily U, Scorrano G. J. Org. Chem. 1983; 48: 3022.
[14] Hirano M, Yakabe S, Monobe H, Morimoto T. Can. J. Chem. 1997; 75:
1905.
[15] (a) Hundscheid FJA, Tandon VK, Rouwette PHFM, Van Leusen AM.
Tetrahedron 1987; 43: 5073; (b) Tandon VK, Singh KA, Rai S, Van
Leusen AM. Heterocycles 2004; 64: 347.
Enantiodifferentiation by 13C and 1H signal dispersion (ꢀν)
Although the ethers are weak ligands in their ability to bind to
dirhodium tetracarboxylates,[5,18] some 13C signal splittings can
be observed for C-2 (ꢀν = 3–6 Hz) and C-2ꢁ/6ꢁ (ꢀν = 1–3 Hz) of
1 and 2a–2d due to the formation of diastereomeric adducts with
Rh∗, respectively. If, however, the substituent is in ortho-position
(3a–3d), diastereomericdispersioneffectsaresignificantlyweaker
if observable at all. Interestingly, both fluorinated derivatives 2a
and 3a show the strongest effects in their respective compound
series; apparently, the proportion of the adduct in the equilibrium
[16] (a) Ruddy SB, Mcintire GL, Roberts ME, Toner JL.Chem. Abstr. 1995;
124: 127168; U.S. patent, 9, U.S. 5476646 A 19951219, cont.-in-
part of U.S. 5,443,814; (b) Illig CR, Toner JL, Lee RW, Baker EJ. Chem.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2008; 46: 23–29

Origin of 13C complexation shifts
Abstr. 1995; 123: 250203; U.S. patent, 8, U.S. 5443814 A 19950822,
cont.-in-part of U.S. 5,405,600.
[17] Ewing DF. Org. Magn. Reson. 1979; 12: 499.
[18] Diaz Go´mez E, Albert D, Duddeck H, Kozhushkov SI, de Meijere A.
Eur. J. Org. Chem. 2006; 2278, equilibrium constants have been
estimated from variable-temperature 1H and 13C NMR experiments
conducted after publication.
DiStasio Jr RA, Lochan RC, Wang T, Beran GJO, Besley NA,
Herbert JM, Lin CY, Van Voorhis T, Chien SH, Sodt A, Steele RP,
Rassolov VA, Maslen PE, Korambath PP, Adamson RD, Austin B,
Baker J, Byrd EFC, Dachsel H, Doerksen RJ, Dreuw A, Dunietz BD,
Dutoi AD, Furlani TR, Gwaltney SR, Heyden A, Hirata S, Hsu C-
P, Kedziora G, Khalliulin RZ, Klunzinger P, Lee AM, Lee MS,
Liang WZ, Lotan I, Nair N, Peters B, Proynov EI, Pieniazek PA,
Rhee YM, Ritchie J, Rosta E, Sherrill CD, Simmonett AC, Subotnik JE,
Woodcock HL III, Zhang W, Bell AT, Chakraborty AK, Chipman DM,
Keil FJ, Warshel A, Hehre WJ, Schaefer HF, Kong J, Krylov AI,
Gill PMW, Head-Gordon M. Phys. Chem. Chem. Phys. 2006; 8: 3172.
[21] Karplus M, Pople JA. J. Chem. Phys. 1963; 38: 2803.
[19] Exner O. In Correlation Analysis in Chemistry, Chapman NB, Shorter J
(eds). Plenum Press: New York, London, 1978; 439.
[20] Density functional calculations (B3LYP 6–31G∗) using SPARTAN
06 version 1.1.0, Wavefunction Inc., Irvine, CA 92612;
see also: Shao Y, Molnar LF, Jung Y, Kussmann J, Ochsenfeld C,
Brown ST, Gilbert ATB, Slipchenko LV, Levchenko SV, O’Neill DP,
c
Magn. Reson. Chem. 2008; 46: 23–29
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc