Electrophilic substituent constant σ+ of electron donor substituents in nonpolar media

Article abstract of DOI:10.1021/jo9000383

The electrophilic substituent constant σ⁺ was derived from the relative rate constants of the solvolysis of substituted a-cumyl chlorides in an acetone/water solvent mixture in the original work by Brown and Okamoto (J. Am. Chem. Soc. 1958, 80,

Full text of DOI:10.1021/jo9000383

Electrophilic Substituent Constant σ+ of Electron Donor
Substituents in Nonpolar Media
Alexander Oehlke,† Alexander A. Auer,‡ Katja Schreiter,† Katja Hofmann,†
Franziska Riedel,† and Stefan Spange*,†
Department of Polymer Chemistry, and Department of Theoretical Chemistry, Institute of Chemistry,
Chemnitz UniVersity of Technology, Strasse der Nationen 62, 09111 Chemnitz, Germany
stefan.spange@chemie.tu-chemnitz.de
ReceiVed January 19, 2009
The electrophilic substituent constant σ+ was derived from the relative rate constants of the solvolysis of
substituted R-cumyl chlorides in an acetone/water solvent mixture in the original work by Brown and
Okamoto (J. Am. Chem. Soc. 1958, 80, 4979). As an extension of this procedure, we were looking for
methods to determine σ+ values in nonpolar media. We have calculated the exocyclic charges q(R) for
+
R-cumyl cations (R-C6H4CMe2 ) as a measure of the electron-donating capacity of meta and para
substituents. The UV-vis absorption ν˜max of the analogously substituted nitrobenzene derivatives has
been used for the experimental measurement of the electron donating capacity of substituents R in nonpolar
+
+
media. A linear relationship between ν˜max and ∆q(R) () q(R-C6H4CMe2 ) - q(C6H5CMe2 )) was obtained.
A description of the electron-donating capacity in nonpolar media can be achieved both for substituents
investigated already and for those for which the experimental determination of σ+ has previously been
difficult. Special interest was directed to boron-based substituents in which the electron-donating ability
is dependent on the coordinational environment of the boron atom.
Introduction
or anionic reaction center.2,7 The theoretical description of
substituent effects (σ, σ0, σ+) has presented a challenge for a
long time. However, recently, many attempts to calculate
substituent effects have been described.8-10 Unlike σ and σ-,
whose definition is derived from equilibrium constants, σ+ was
derived from the measured rate constant k of the SN1 solvolyses
of 16 meta- and 21 para-substituted R-cumyl chlorides in 90%
acetone/water at 25 °C (reaction 1).11
The Hammett equation has proved to be a reliable tool for
the determination of quantitative structure-property and
structure-activity relationships in many areas of organic and
physical chemistry.1-6 Besides the established σm and σp values
of the classic Hammett equation, σ+ and σ- values are used to
describe reactions and processes where a significant proportion
of conjugation exists between the substituent and the cationic
(7) Carey, F. A.; Sundberg, R. J. AdVanced Organic Chemistry, Part A:
Structure and Mechanisms, 4th ed.; Plenum Press: New York, 1993.
(8) (a) Hollingsworth, C.; Seybold, P.; Hadad, C. Int. J. Quantum Chem.
2002, 90, 1396–1403. (b) Domingo, L.; Pe´rez, P.; Contreras, R. J. Org. Chem.
2003, 68, 6060–6062. (c) Wiberg, K. J. Org. Chem. 2002, 67, 1613–1617. (d)
Wiberg, K. J. Org. Chem. 2003, 68, 875–882. (e) Wiberg, K. J. Org. Chem.
2003, 67, 4787–4794. (f) Barbour, J.; Karty, J. J. Phys. Org. Chem. 2005, 18,
210–216. (g) Galabov, B.; Ilieva, S.; Schaefer, H. F. J. Org. Chem. 2006, 71,
6382–6387.
† Department of Polymer Chemistry.
‡ Department of Theoretical Chemistry.
(1) Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96–103.
(2) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165–195.
(3) Hansch, C.; Hoekman, D.; Gao, H. Chem. ReV. 1996, 96, 1045–1075.
(4) Krygowski, T.; Stepien´, B. Chem. ReV. 2005, 105, 3482–3512.
(5) (a) Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis
in Chemistry and Biology; Wiley-Interscience: New York, 1979. (b) Hansch,
C.; Leo, A. Exploring QSAR: Fundamentals and Applications in Chemistry and
Biology; American Chemical Society: Washington, DC, 1995.
(6) Exner, O.; Bo¨hm, S. Curr. Org. Chem. 2006, 10, 763–778.
(9) Morao, I.; Hillier, I. Tetrahedron Lett. 2001, 42, 4429–4431.
(10) Nakata, K.; Fujio, M.; Nishimoto, K.; Tsuno, Y. J. Phys. Org. Chem.
2003, 16, 323–335.
3316 J. Org. Chem. 2009, 74, 3316–3322
10.1021/jo9000383 CCC: $40.75 2009 American Chemical Society
Published on Web 04/03/2009

Electron Donor Substituents in Nonpolar Media
effects, as in the biologically important catechol and catecholate
groups,20 is difficult to carry out using the established methods.
The evaluation of the electronic effects of boron-based
substituents is also associated with difficulties. In recent years,
there has been a rapid increase in the number of publications
concerning compounds carrying the B(OH)2 or B(OR)2 groups.21
At the same time, interest in the electronic characteristics of
substituents based on sp2- or sp3-hybridized boron atoms has
also increased. While the electronic character of trigonal planar
boron substituents ranges from electron-withdrawing to neutral,2
tetrahedrally coordinated boron substituents act as electron
donating groups.2,22,23 The hybridization state of a boron atom
and also its surroundings thus affect the electronic properties
of boron-based substituents. In the literature, the Hammett
characteristics for this type of substituents has been described
rarely.2 In particular, the electron-donating strength of fluoro-
adducts of boronic acid esters has not been determined.
As resonance effects play a minor role in the meta position,
kinetic data for meta-substituted compounds were correlated
with the Hammett σm constants according to eq 2. Starting with
eight points (R ) MeO, Me, H, F, Cl, Br, I, NO2) for which
the data was considered reliable, a reaction constant F ) -4.54
was determined with which the remaining σ+m and σp+ constants
were determined. With the knowledge of these values, additional
σ+p constants were determined from the kinetic data of other
electrophilic substitution reactions.11
Theoretical studies and experimental investigations in non-
polar media concerning reactions or intermediates with strong
resonance interactions between substituent and reaction center
have been compared with the solution σ+ values from Brown
and Okamoto11 in several instances.9,10,18,24-26 HBD substitu-
ents, the OH group in particular, tend to deviate from plots
against σp+,9,10,18,24 since solvation effects are excluded in the
gas phase as well as considerably reduced in nonpolar media.
The experimental determination of thermodynamic stabilities
of benzyl cations in the gas phase gives a measure of σ+ free
from solvent effects. However, the set of substituents in such
measurements was limited due to experimental difficulties.27
An approach for the theoretical calculation of σ+ values from
gas-phase stabilities of R-cumyl cations was reported by Nakata
et al. in terms of the Yukawa-Tsuno equation.10 Morao and
Hillier have already stated that there is a linear relationship
between the calculated exocyclic NBO charge of para-substituted
benzyl cations and the σ+p constants.9
In an analogy to Brown’s definition of the σ+ scale, we
investigated substituted R-cumyl cations in this work. These
carbocations are stabilized by delocalization of the positive
exocyclic charge (Scheme 1) through inductive (+I)- and
mesomeric (+M)-effects as well as hyperconjugative effects.27-29
The extent to which the exocyclic charge is delocalized gives
a measure of the electron donating capacity of the corresponding
kR
( )
log
) Fσ
(2)
kH
However, the values determined for the electrophilic sub-
stituent constant σ+ of substituents prone to solvation effects
and pH dependencies are thus subjected to uncertainties. This
fact must be considered especially for hydrogen bond donating
(HBD) and hydrogen bond accepting (HBA) substituents.
Substituents with HBD ability (e.g., OH, NH2) experience
specific solvation in HBA solvents, resulting in an enhancement
of the electron-donating properties.12 This property is explicitly
exploited in solvatochromic dyes (e.g., 4-nitroanilines13,14) and
in acid-base indicators (e.g., 4-nitrophenol15). The reactions,
over which the σ+ scale was originally determined, were carried
out in aqueous media. However, several substituents (e.g., OH,
NH2, NHMe, NMe2) could not be determined experimentally
with the cumyl system, so that the σ+ constants of these
substituents had to be determined using more suitable reaction
series.11,16 Some of these reactions require highly acidic media11
which involve attenuation of the electron-donating capacity of
substituents with HBA ability (e.g., amino substituents) due to
protonation equilibria.17
In the literature, there is a controversy concerning the
classification of the electron-donating capacity of the OH
function. It is assumed that the electron-donating strength of
this substituent is classified as too high with σ+p (OH) ) -0.92.2
Some authors have assumed that the OH group and the OMe
group have the same electron-donating capacity (σ+p (OH) )
σ+p (OMe) ) -0.78).18 Furthermore, the evaluation of the
substituent effects of charged groups is difficult in many cases.19
In particular, the determination of the combined substituent
(19) (a) Binev, I.; Kuzmanova, R.; Kaneti, J.; Juchnowski, I. J. Chem. Soc.,
Perkin Trans. 2 1982, 1533–1536. (b) Hoefnagel, A. J.; Hoefnagel, M. A.;
Wepster, B. M. J. Org. Chem. 1978, 43, 4720–4745.
(20) (a) Schweigert, N.; Zehnder, A.; Eggen, R. EnViron. Microbiol. 2001,
3, 81–91. (b) Avdeef, A.; Sofen, S.; Bregante, T.; Raymond, K. J. Am. Chem.
Soc. 1978, 100, 5362–5370. (c) van de Waterbeemd, H.; Carrupt, P.; Testa, B.
HelV. Chim. Acta 1985, 68, 715–723.
(21) Boronic Acids; Hall, D., Ed.; Wiley-VCH: Weinheim, 2005.
(22) (a) DiCesare, N.; Lakowicz, J. J. Phys Chem. A 2001, 105, 6834–6840.
(b) DiCesare, N.; Lakowicz, J. Anal. Biochem. 2002, 301, 111–116.
(23) Oehlke, A.; Auer, A. A.; Jahre, I.; Walfort, B.; Ru¨ffer, T.; Zoufala´, P.;
Lang, H.; Spange, S. J. Org. Chem. 2007, 72, 4328–4339.
(11) Brown, H. C.; Okamoto, Y. J. Am. Chem. Soc. 1958, 80, 4979–4987.
(12) Epstein, L. M.; Ashkinadze, L. D.; Shubina, E. S.; Kravtsov, D. N.;
Kazitsyna, L. A. J. Organomet. Chem. 1982, 228, 53–59.
(13) (a) Reichardt, C. Chem. ReV. 1994, 94, 2319–2358. (b) Reichardt, C.
SolVents and SolVent Effects in Organic Chemistry, 3rd ed.; Wiley-VCH:
Weinheim, 2003.
(24) (a) DiLabio, G. A.; Pratt, D. A.; LoFaro, A. D.; Wright, J. S. J. Phys.
Chem. A 1999, 103, 1653–1661. (b) DiLabio, G. A.; Ingold, K. U. J. Org. Chem.
2004, 69, 1620–1624.
¨
(25) Ruff, F.; Farkas, O. J. Phys. Org. Chem. 2008, 21, 53–61.
(14) Kamlet, M. J.; Taft, R. W. J. Am. Chem. Soc. 1976, 98, 377–383.
(15) Kolthoff, I. M.; Chantooni, M. K.; Bhowmik, S. J. Am. Chem. Soc.
1966, 88, 5430–5439.
(26) Ferna´ndez, I.; Frenking, G. J. Org. Chem. 2006, 71, 2251–2256.
(27) Tsuno, Y.; Fujio, M. Chem. Soc. ReV. 1996, 25, 129–139.
(28) (a) Reindl, B.; Clark, T.; Schleyer, P. v. R. J. Phys. Chem. A 1998,
102, 8953–8963. (b) van Alem, K.; Lodder, G.; Zuilhof, H. J. Phys. Chem. A
2002, 106, 10681–10690.
(29) (a) Herrero, R.; Quintanilla, E.; Mu¨ller, P.; Abboud, J. Chem. Phys.
Lett. 2006, 420, 493–496. (b) Robbins, A.; Jin, P.; Brinck, T.; Murray, J.; Politzer,
P. Int. J. Quantum Chem. 2006, 106, 2904–2909. (c) Richard, J.; Jagannadham,
V.; Amyes, T.; Mishima, M.; Tsuno, Y. J. Am. Chem. Soc. 1994, 116, 6706–
6712.
(16) The average value from four reaction series was found to be σ+p (NMe2)
)-1.7((0.2).11
(17) Azzaro, M.; Gal, J. F.; Geribaldi, S. J. Chem. Soc., Perkin Trans. 2
1984, 771–774.
(18) (a) Pratt, D. A.; de Heer, M. I.; Mulder, P.; Ingold, K. U. J. Am. Chem.
Soc. 2001, 123, 5518–5526. (b) Pratt, D. A.; DiLabio, G. A.; Valgimigli, L.;
Pedulli, G. F.; Ingold, K. U. J. Am. Chem. Soc. 2002, 124, 11085–11092.
J. Org. Chem. Vol. 74, No. 9, 2009 3317

Oehlke et al.
SCHEME 1. Possible Resonance Structures for the
r-Cumyl Cation with Electron-Donating Substituents in the
Para and Meta Positions
with the exocyclic NBO charges of R-cumyl cations analogously
substituted calculated at the B3LYP/TZVP level35,36 (see
Computational Details).
As revealed by the NBO population analyses, the substituent-
induced charge difference at the R-carbon atom of the cumyl
cation is to a large extent dependent on the population of the
pz-orbital.
The UV-vis absorption maxima of the p-nitrobenzenes
(entries 1-14), measured in cyclohexane, are in agreement with
published values from the literature.32 The 4-nitrophenolate ion
(entry 26) was obtained by the addition of TBAF (0.1 M in
THF) to 4-nitrophenol in dichloromethane solution. The fluo-
roboronate adducts (entries 24 and 25) were generated by
addition of TBAF (0.1 M in THF) to solutions of the corre-
sponding nitrobenzeneboronic acid esters in dichloromethane
(eq 3).23
substituent on the R-cumyl cation. The charge for the exocyclic
carbon atom can be determined easily by a natural bond orbital
(NBO) population analysis,30 which takes the aforementioned
cation-stabilizing effects into account.
To have a facile and independent method for the experimental
determination of the electron-donating capacity in nonpolar
media we used the UV-vis spectroscopy of nitrobenzenes. The
electron-donor-substituted 4-nitrobenzenes used for this purpose
are simple examples of push-pull (D-π-A) chromophores.31
The longest wavelength UV-vis absorption of this class of
compounds represents, in many cases, the lowest energy π-π*
transition and corresponds to a HOMO-LUMO transition. The
energy level of the LUMO is in principle determined by the
acceptor substituent. In the 4-nitrobenzene series, the level of
the LUMO remains approximately constant. The energy level
of the HOMO is determined by the electron-donating capacity
of the substituent R. In the 4-nitrobenzene series, the energy
gap between the HOMO and the LUMO decreases with the
increasing donating capacity of the corresponding substituent.
This is accompanied by a shift of the UV-vis absorption to
lower energies.32 The measurement of the UV-vis absorption
can be carried out either directly in the gas phase33 or very easily
in nonpolar solvents.12,30 In order to minimize spectral shifts,
based on solvatochromic effects, the UV-vis spectra should
be measured in cyclohexane.13,34 UV-vis data of nitrobenzenes,
with highly polar substituents, charged groups in particular,
which are not soluble in cyclohexane, can be obtained, for
instance, from dichloromethane solution. Solvents with a higher
HBD or HBA capacity are less suitable, as solvatochromic
effects have an increasing influence.34
As has been shown in previous work, the UV-vis absorption
of acceptor-substituted arylboronic acid pinacol esters are shifted
to lower energies, following coordination of a fluoride ion to
the boron atom.23 For the nitrobenzeneboronic acid esters
investigated in this work, shifts in the ultraviolet spectral range
were observed. The addition of TBAF was carried out until no
further spectroscopic shift was observed. The generation of the
fluoroboronate adducts was confirmed by heteronuclear NMR
spectroscopy. The meta and para isomers of nitrobenzeneboronic
acid pinacol ester showed a 11B signal at δ ) 30.4 ppm, typical
for a trigonally coordinated boron atom in arylboronic acid
pinacol esters.23,38 After addition of TBAF, a high field shift of
the 11B signal to δ ) 7.0 ppm occurred. This value corresponds
to a tetra-coordinated boron atom and is typical of fluoroboronate
adducts of arylboronic acid pinacol esters.23,38
As has been shown several times in the literature, there is a
general linear relationship between the longest wavelength
UV-vis absorption maxima of electron-donor-substituted 4-ni-
trobenzenes and the σ+p constants.39 However, in such correla-
tions a small number of data points was taken into consider-
ation.40
Figure 1 shows the UV-vis absorption maxima ν˜max of the
4-nitrobenzenes, measured in cyclohexane, plotted against the
literature values for σ+p . The following regression equation was
obtained
ν˜max(R, 103 cm-1) ) 5.842σ+p + 39.046
(4)
Results and Discussion
The longest wavelength UV-vis absorption maxima for
electron-donor-substituted nitrobenzenes, measured in cyclo-
hexane and dichloromethane, are shown in Table 1, together
where r ) 0.9746, sd ) 1.031, n ) 11. Equation 4 is in
agreement with already published data.40b Attention should be
drawn to the following points relating to Figure 1: According
(30) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899–
926.
(31) (a) Exner, O.; Krygowski, T. Chem. Soc. ReV. 1996, 25, 71–75. (b)
Bo¨hm, S.; Exner, O. THEOCHEM 2007, 803, 9–16.
(35) (a) Ko¨hn, A.; Ha¨ttig, C. J. Chem. Phys. 2003, 119, 5021–5036. (b) Ha¨ttig,
C. J. Chem. Phys. 2003, 118, 7751–7761. (c) Ha¨ttig, C.; Weigand, F. J. Chem.
Phys. 2000, 113, 5154–5161.
(36) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(32) Minesinger, R. R.; Kamlet, M. J. J. Phys. Chem. 1974, 78, 494–497.
(33) (a) Schubert, W. M.; Steadly, H.; Craven, J. M. J. Am. Chem. Soc. 1960,
82, 1353–1356. (b) Schubert, W. M.; Murphy, R. B.; Robins, J. J. Org. Chem.
1970, 35, 951–956. (c) Nicolet, P.; Laurence, C. J. Chem. Soc., Perkin Trans. 2
1986, 1071–1079.
(34) (a) Marcus, Y. Chem. Soc. ReV. 1993, 22, 409–416. (b) Catala´n, J.;
Lo´pez, V.; Pe´rez, P.; Mart´ın-Villamil, R.; Rodr´ıguez, J. Liebigs Ann. 1995, 241–
252.
(37) Riedel, F.; Spange, S. J. Phys. Org. Chem. 2008, 22, 203–211.
(38) Neumann, T.; Dienes, Y.; Baumgartner, T. Org. Lett. 2006, 8, 495–
497.
(39) (a) Jaffe´, H. H.; Orchin, M. Theory and Applications of UltraViolet
Spectroscopy; Wiley; New York, 1962. (b) Katritzky, A. R.; Topsom, R. D. In
AdVances in Linear Free Energy Relationships; Chapman, N. B., Shorter, J.,
Ed.; Plenum Press: London, 1972. (c) Rao, C. N. R. UltraViolet and Visible
Spectroscopy, 3rd ed.; Butterworths: London, 1975.
3318 J. Org. Chem. Vol. 74, No. 9, 2009

Electron Donor Substituents in Nonpolar Media
TABLE 1. UV-vis Absorption ν˜max (103 cm-1) of Substituted Nitrobenzenes R-C6H4NO2 in the Gas Phase and in Cyclohexane (CH) or
+
Dichloromethane (DCM) Solution, Calculated Exocyclic NBO Charge q (e) of Substituted r-Cumyl Cations R-C6H4CMe2 and Evaluated σ+
Values
a
b
c
d
e
entry
R
ν˜max,gas
ν˜max,CH
ν˜max,DCM
q
σ+
σ+NBO
σ+CH
σ+DCM
1
2
3
4
5
6
7
8
H
4-F
4-Me
4-Et
4-OH
4-OMe
4-NHAc
4-NH2
4-NHMe
4-NMe2
4-NEt2
41.82
40.80
39.97
39.68
38.76
37.74
37.59
34.97
33.90
33.33
31.06
29.24
28.17
27.47
28.25
27.55
27.78
39.06
38.46
38.17
37.59
36.10
35.97
33.33
32.36
31.85
28.57
26.67
25.45
24.94
25.45
24.94
25.19
37.88
37.45
36.90
35.84
30.77
35.34
33.00
23.15
23.20
35.34
33.00
23.15
23.20
0.4142
0.3971
0.3896
0.3881
0.3606
0.3482
0.3438
0.3185
0.3057
0.2973
0.2869
0.2879
0.2894
0.2834
0.4037
0.3971
0.3909
0.3778
0.3447
0.3881
0.3553
0.2654
0.2494
0.3329
0.2800
0.1583
0.1584
0
0
-0.07
-0.21
-0.37
-0.39
-0.80
-0.96
-1.05
-1.41
-1.69
-1.85
-1.96
-1.84
-1.95
-1.91
-0.16
-0.26
-0.12
-0.20
-0.40
-0.42
-0.79
-0.92
-0.99
-1.44
-1.71
-1.87
-1.94
-1.87
-1.94
-1.91
-0.16
-0.22
-0.29
-0.44
-1.14
-0.51
-0.83
-2.19
-2.18
-0.51
-0.83
-2.19
-2.18
-0.07
-0.31
-0.30
-0.92
-0.78
-0.60
-1.30
-1.81
-1.70
-2.07
-0.26
-0.37
-0.39
-0.81
-1.00
-1.06
-1.45
-1.64
-1.77
-1.92
-1.91
-1.89
-1.98
-0.16
-0.26
-0.35
-0.55
-1.05
-0.39
-0.89
-2.25
-2.49
-1.23
-2.03
-3.87
-3.86
37.98
37.08
34.27
32.61
31.43
30.53
9
10
11
12
13
14
15
16
17
18
19f
20g
21g
22g
23f,g
24
25
26
27f
4-N(CH2)3
4-N(CH2)4
4-N(CH2)5
3-Bpinh
4-Bpinh
3-Bnmeai
4-Bnmeai
3-OH, 4-OH
3-BpinFNa
4-BpinFNa
4-ONa
-
-
-
-
-
-
-
-
k
k
k
k
k
k
-2.30
-4.27l
k
3-OH, 4-ONa
3-BpinF-
4-BpinF-
4-O-
k
k
-
-
k
j
k
3-OH, 4-O-
-
a From ref 33. b From ref 2. c Values from eq 9. d Values from eq 10. e Values from eq 11. f From ref 37. g Experimental values are those without
j
Na+ counterion. h Bpin ) 4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl. i Bnmea ) (NfB)-3-(6-methyl-1,3,6,2-dioxazaborocan-2-yl. With (n-C4H9)4N+
as cation. k Compound is insoluble in cyclohexane. l From ref 19a.
to the literature values (i) the OH group (σ+p (OH) ) -0.92)2 is
classified as a stronger electron donor than the OMe group
(σ+p (OMe) ) -0.78)2 and (ii) the NHMe group (σ+p (NHMe) )
-1.81)2 is a stronger electron donor than the NMe2 group
(σ+p (NMe2) ) -1.70).2 Like other authors, we find that the σp+
values of these substituents are inappropriately evaluated.18
Plots (Figures 2 and 3) of the charge difference ∆q(R) )
q(R) - q(H) between the R-cumyl cations against the energy
equivalent UV-vis absorption maxima ν˜max from Table 1
(entries 1-16) correspond to a single parameter equation of
the Hammett type in the form of eq 5).
ν˜max(R) ) F∆q(R) + ν˜max(H)
(5)
The following regression equation is derived from the
UV-vis absorption data obtained in cyclohexane solution
ν˜max(R, 103 cm-1) ) 97.614∆q(R) + 40.122
(6)
FIGURE 1. Relationship between longest wavelength UV-vis absorp-
tion of 4-nitrobenzenes and the σ+p constants from the literature.2
with r ) 0.9980, sd ) 0.316, n ) 16. The following regression
equation is derived from the UV-vis absorption data obtained
in dichloromethane solution
gously substituted R-cumyl cations than by the σ+p values given
in the literature.2 The OMe group (∆q(OMe) ) -0.066 e)
compared with the OH group (∆q(OH) ) -0.054 e) as well as
the NMe2 group (∆q(NMe2) ) -0.117 e) compared with the
NHMe group (∆q(NHMe) ) -0.109 e) are both characterized
as stronger electron donors according to calculated ∆q and to
the electronic transitions of the corresponding nitrobenzenes in
nonpolar solvents. The suitability of the calculated NBO charge
differences ∆q as a meaningful descriptor of the electrophilic
substituent constant σ+ in nonpolar media is confirmed by linear
correlation with gas-phase UV-vis absorption data of various
nitrobenzenes including HBD substituents33 already published
ν˜max(R, 103 cm-1) ) 109.717∆q(R) + 39.023
(7)
with r ) 0.9965, sd ) 0.469, n ) 16. In both cases (eqs 6 and
7), excellent linear correlations between the experimental
UV-vis absorption and the calculated NBO charge differences
were found for the neutral substituents (entries 1-16). In
comparison to eq 4, regression coefficients r > 0.99 together
with smaller standard deviations were obtained. The UV-vis
absorption of the 4-nitrobenzenes in nonpolar solvents can be
better described by the NBO charge differences ∆q of analo-
J. Org. Chem. Vol. 74, No. 9, 2009 3319

Oehlke et al.
tion data and the calculated NBO charges, it can be concluded
that the 3-Bpin group ranks among the series of electron
donating substituents with para orientation. Thus, the 3-Bpin
group is included in the regression plots resulting in eqs 6
and 7.
Although the UV-vis absorption data and the NBO charge
differences show a very good linear correlation, it should be
noted that the exocyclic charge calculated for the R-cumyl
cations is assigned in each case to a single conformer. The
UV-vis absorption measured for the nitrobenzenes, on the other
hand, comprises the individual UV-vis absorptions of all the
conformers in thermodynamic equilibrium. The lowest energy
conformer therefore gives the largest contribution to the total
UV-vis absorption spectrum. For groups such as R ) Me or
F, this consideration is irrelevant. For larger and more flexible
substituents (e.g., R ) piperidin-1-yl, entry 14), which can exist
in several conformations in solution, differences between
experimental and calculated values can be expected.
FIGURE 2. Relationship between the longest wavelength UV-vis
absorption of nitrobenzenes measured in cyclohexane and the calculated
exocyclic charge of R-cumyl cations (entries 1-16).
In order to be able to compare the NBO charge differences
∆q(R) directly with the known values of σ+ (R) from the
literature, the parameters are plotted against each other. The
following regression equation was obtained
σ+(R) ) 15.10∆q(R)
(9)
with r ) 0.9821, sd ) 0.143, n ) 10 (calculated from the entries
1-6 and 8-11). The σ+ values directly obtained from charge
differences ∆q are denoted as σN+BO
.
Additionally, the donating strength of the substituent R can
be derived directly from the position of the UV-vis absorption
maxima of the nitrobenzenes. For the UV-vis absorption data
measured in cyclohexane or dichloromethane, eqs 10 and 11
are obtained, respectively. The values of σ+ calculated according
to eqs 9-11 are given in Table 1 together with the σ+ values
according to Brown.11
σ+CH(R) ) 0.155(ν˜max,CH(R, 103 cm-1) - 40.122) (10)
FIGURE 3. Relationship between the longest wavelength UV-vis
absorption of substituted nitrobenzenes measured in dichloromethane
and the calculated exocyclic charge of substituted R-cumyl cations.
Closed squares ) neutral substituents included in eq 7 (entries 1-16).
Open squares ) substituents not included in eq 7. Open diamonds )
charged substituents not included in eq 7.
σ+DCM(R) ) 0.138(ν˜max,DCM(R, 103 cm-1) - 39.023)
(11)
The investigation of the NBO charges shows that there is a
disproportionately strong delocalization of the exocyclic charge
concerning cumyl cations with anionic substituents. The result-
ing electron-donating effect of these substituents is more
pronounced in the gas phase than would be expected from the
UV-vis absorption maxima in solution. This observation might
be caused by aggregation effects (ion pairing) in DCM solution.
The influence of ion pairing on the UV-vis absorption behavior
is small, as the UV-vis absorption in strongly dissociating
media, such as DMSO, is slightly different from values
measured in DCM solution. These differences can be attributed
to solvatochromic effects. A more important role is played by
the difference in the polarity between the gas phase and DCM
solution.34b The dipole moment of the betaine-like cumyl cations
is smaller in the gas phase than would be expected in DCM
solution, as charge separation is a process less favored with
diminished polarity of the surrounding medium. Consequently,
the electron donating ability of anionic substituents obtained
either from DFT calculations on cumyl cations or UV-vis
absorption measurements of nitrobenzenes in DCM solution
differ from each other. In our previous study on fluoro-boronate
adducts, we have observed differences between calculated gas
by other working groups (see Table 1) resulting in eq 8
ν˜max(R, 103 cm-1) ) 89.640∆q(R) + 42.342
(8)
with r ) 0.9943, sd ) 0.481, n ) 9.
The 3-Bpin group is particularly interesting (entry 15).
Although this group is attached at the meta position, electron-
donating capacity for the corresponding nitrobenzene was
observed. Calculations on the electronic excitation energies of
3-nitrobenzeneboronic acid pinacol ester at the RI-CC2/SVP
level35,41,42 also show an excitation behavior, similar to para-
substituted nitrobenzenes. From the measured UV-vis absorp-
(40) (a) Brownlee, R. T.; Topsom, R. D. Spectrochim. Acta 1973, 29A, 385–
393. (b) Kamlet, M. J.; Abboud, J. L.; Taft, R. W. J. Am. Chem. Soc. 1977, 99,
6027–6038. (c) Yokoyama, T.; Taft, R. W.; Kamlet, M. J. Spectrochim. Acta
1984, 40A, 669–673.
(41) Christiansen, O.; Koch, H.; Jørgenson, P. Chem. Phys. Lett. 1995, 243,
409–418.
(42) Scha¨fer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571–
2577.
3320 J. Org. Chem. Vol. 74, No. 9, 2009

Electron Donor Substituents in Nonpolar Media
TABLE 2. UV-vis Absorption ν˜max (103 cm-1) of Substituted
4-Nitroanilines Measured in DCM Solution and Derived σ+ Values
phase and measured UV-vis absorption maxima, which also
could be attributed to polarity differences.23
To address this issue, we created formal charge neutrality
by adding a positive charge to the anionic substituents. In this
manner, calculations investigating negatively charged substit-
uents were carried out with Na+ as the counterion. The NBO
charges obtained in this way (entries 20-23) correlate with the
UV-vis absorption data from DCM solution and hold the
regression line from eq 7. On the other hand, the NBO charges
obtained for cumyl cations with anionic substituents without
the Na+ counterion (entries 24-27) show clear deviations from
eq 7.
As the calculated NBO charges and the measured UV-vis
absorption maxima show, the Bpin group exhibits a weak
electron-donating character, both in the meta and the para
positions. This is caused by a weak (+I)-effect, due to the
donation of π-electrons from the oxygen atoms to boron.23 The
electron-donating strength of the Bpin group is less pronounced
in the meta position than in the para position.
For the fluoro adduct of 3-nitrobenzeneboronic acid pinacol
ester, a UV-vis absorption spectrum with a single band is
obtained. Calculations of the electronic excitation energies at
the RI-CC2/SVP level35,41,42 show that superposition of several
electronic transitions to a single band occurs in the ultraviolet
spectral range. Our results confirm that the fluoro adducts
3-BpinF- and 4-BpinF- act as moderate and strong electron-
donating substituents, respectively. As resonance interactions
between the π-system and a tetracoordinated boron atom can
be ruled out (structures similar to 1d are not possible), the effect
of these groups must be due to strong (+I) effects.
As can be concluded from measured UV-vis absorption data,
the electron-donating strength of the 3-BpinF- group in dichlo-
romethane solution is slightly higher than an alkyl substituent
in the para position. Furthermore, in dichloromethane solution
the 4-BpinF- group represents a strong electron donor, with a
σ+ value between that of the 4-OH and the 4-OMe groups. These
findings are supported by the charge differences obtained by
DFT calculations. Previously, we have shown that the UV-vis
absorption of fluoro adducts of nitro-substituted stilbeneboronate
pinacol esters is dependent on the HBD ability of the solvent.
The strongest bathochromic shifts were observed in aprotic
solvents (e.g., DCM) indicating strong electron-donating ability
in such solvents.23
The N-methyldiethanolamine esters of boronic acids (entries
17 and 18) with a intramolecular NfB coordination also act
as electron donating groups. In comparison to the fluoro-adducts
of the boronic acid pinacol esters, the NfB coordinated esters
show a weaker Lewis acid-Lewis base interaction. This is
confirmed by the 11B signal at lower field (δ ) 11.8 ppm) for
the m- and p-nitrobenzeneboronic acid N-methyldiethanolamine
esters. The diminished electron-donating strength is conform
to smaller spectral shifts in the nitrobenzene series and to the
enhanced charge localization at the exocyclic position of the
corresponding R-cumyl cation. For both meta and para orienta-
tion, the electron-donating capacity of the boronic acid N-
methyldiethanolamine ester moiety is between that of the Bpin
and BpinF- group.
a Values from eq 11.
of an intramolecular hydrogen bond between the vicinal OH
groups (4-OH as HBD; 3-OH as HBA), the electron donating
capacity is increased compared to the individual effects. Both,
the sum of the calculated individual σ+ values (σN+BO(4-OH) )
-0.81; σ+NBO(3-OH) ) 0.10; ∑σN+BO ) -0.71) and the sum of
the individual literature σ+ values (σp+(OH) ) -0.92; σm+(OH)
) -0.04; ∑σ+ ) -0.96)2,5 are notably less negative than the
calculated σ+NBO(3-OH, 4-OH).
From the correlations presented in this work, it is possible to
evaluate the electron-donating strength of substituents which
have not yet been reported. In Table 2, the longest wavelength
UV-vis absorptions of various para-substituted nitroanilines are
listed, which have also been investigated in our group.43 Using
eq 11, the positions of the UV-vis absorption maxima (Table
2) can be used to calculate σ+ values for more complex amino
substituents whose electron-donating capacity has not yet been
quantified.
Conclusions
It has been shown that for electron-donating substituents
the values of the electrophilic substituent constants σ+ can
be obtained from the exocyclic charge of substituted R-cumyl
cations. In contrast to kinetic measurements, which define
the σ+ scale, the UV-vis absorption data of electron-donor-
substituted 4-nitrobenzenes in nonpolar solvents were used
as an alternative experimental data set for the quantification
of σ+. The determination of σ+ values, which are valid in
nonpolar media, such as the gas phase and nonpolar solvents,
is possible. An advantage of our study is that a wide range
of electron-donating substituents including HBD and HBA
Combined substituent influences as in catechol and catecho-
late groups can also be examined. Thus, the electron donating
capacity resulting from the two OH groups of the 3,4-
dihydroxyphenyl unit, starting from the exocyclic NBO charge
yielded a σ+NBO(3-OH, 4-OH) ) -1.14. Through the formation
(43) (a) Spange, S.; Hofmann, K.; Walfort, B.; Ru¨ffer, T.; Lang, H. J. Org.
Chem. 2005, 70, 8564–8567. (b) Schreiter, K.; Spange, S. J. Phys. Org. Chem.
2008, 21, 242–250.
J. Org. Chem. Vol. 74, No. 9, 2009 3321

Oehlke et al.
groups is involved. Furthermore, this concept can be extended
to the determination of σ+m values. However, the experimental
confirmation of these values by UV-vis absorption data on
3-substituted nitrobenzenes is not possible by direct com-
parison to the 4-substituted counterparts. Even so, σ+ values
for combined substituent influences (e.g., 3-OH, 4-OH) can
be established. These were previously only available through
summation of the individual σ+ values. The electronic
characteristics of boron-based substituents have also been
evaluated. It has been shown that tetrahedrally coordinated
boron substituents, fluoro-boronate adducts in particular,
exhibit a significant electron-donating character both in the
meta (moderately electron donating) and para position
(strongly electron donating) relative to the electrophilic
center. It was found that the electronic properties are
significantly dependent on the strength of the interaction
between Lewis bases and the Lewis acidic boron center.
Experimental Section
Synthesis of Arylboronic Acid N-Methyldiethanolamine Es-
ters.48 Equimolar amounts of arylboronic acid and N-methyldi-
ethanolamine were mixed in anhydrous toluene and stirred for 1 h
at 100 °C. After azeotropic removal of water, the solvent was
removed under reduced pressure to yield the corresponding esters
as colorless solids.
(NfB)-3-(6-Methyl-1,3,6,2-dioxazaborocan-2-yl)nitroben-
zene (17-NO2). Yield: 0.291 g (97%). Mp (toluene): 124 °C. H
1
NMR (250 MHz, CDCl3): δ 2.33 (s, 3H), 3.04 (ddd, J ) 11.5 Hz,
J ) 8.2 Hz, J ) 6.6 Hz, 2H), 3.25 (ddd, J ) 11.5 Hz, J ) 5.3 Hz,
J ) 4.3 Hz, 2H), 4.09-4.27 (m, 4H), 7.43 (pt, J ) 7.7 Hz, 1H),
7.95 (dpt, J ) 7.3 Hz, J ) 1.1 Hz, 1H), 8.09 (ddd, J ) 8.1 Hz, J
) 2.5 Hz, J ) 1.2 Hz, 1H), 8.44 (m, 1H). 13C NMR (62.5 MHz,
CDCl3): δ 47.8, 60.8, 62.5, 122.7, 127.7, 128.2, 139.5, 147.9. 11B
NMR (80 MHz, CDCl3): δ 11.8 (br). IR (KBr, cm-1): ν 2865, 1607,
1518, 1351. Anal. Calcd for C11H15BN2O4 (250.05): C, 52.83; H,
6.05; N, 11.20. Found: C, 53.01; H, 6.15; N 11.19.
(NfB)-4-(6-Methyl-1,3,6,2-dioxazaborocan-2-yl)nitroben-
zene (18-NO2). Yield: 0.287 g (96%). Mp (toluene): 103-104 °C.
1H NMR (250 MHz, CDCl3): δ 2.32 (s, 3H), 3.03 (ddd, J ) 11.5
Hz, J ) 8.4, J ) 6.6 Hz, 2H), 3.25 (ddd, J ) 11.5 Hz, J ) 5.3 Hz,
J ) 4.3 Hz, 2H), 4.08-4.27 (m, 4H), 7.78 (d, J ) 8.7 Hz, 2H),
8.10 (d, J ) 8.7 Hz, 2H). 13C NMR (62.5 MHz, CDCl3): δ 47.7,
60.8, 62.5, 122.1, 134.0, 147.9. 11B NMR (80 MHz, CDCl3): δ
11.8 (br). IR (KBr, cm-1): ν 2861, 1594, 1513, 1358. Anal. Calcd
for C11H15BN2O4 (250.05): C, 52.83; H, 6.05; N, 11.20. Found: C,
52.80; H, 6.14; N 11.19.
Computational Details
All geometries used for NBO population analyses and RI-
CC2 calculations of the excitation energies were optimized using
the TURBOMOLE program package44 (for a current version see
http://www.turbomole.de) with the B3LYP functional35 in
combination with the TZVP basis set.36 The geometries have
been converged to a residual gradient norm of 10-3 au or better.
The energy was converged to at least 10-6 au, and fine quadrature
grids have been used to calculate the exchange and correlation
potential (size 5).45 The NBO calculations46 have been carried
out at the B3LYP/TZVP level of theory35,36 using the NBO
program as distributed with the NWChem package.47 With the
6-311G** basis set comparable NBO results were obtained. The
RI-CC2 calculations with the SVP35,41,42 basis set were carried
out with TURBOMOLE.44
UV-vis Spectroscopy. UV-vis absorption spectra of freshly
prepared solutions at a concentration of 10-4 to 10-5 M were
measured at 25 °C in quartz cuvettes with an optical path length of
1 cm.
Acknowledgment. Financial Support by the Fonds der
Chemischen Industrie, Frankfurt am Main, is gratefully
acknowledged.
Supporting Information Available: Experimental proce-
1
dures, H and 13C NMR spectra, selected UV-vis absorption
(44) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346–354.
(45) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨lmel, C. Chem. Phys.
Lett. 1989, 162, 165–169.
(46) NBO, Version 5.0; Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.;
Carpenter, J. E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. Theoretical
Chemistry Institute, University of Wisconsin: Madison, WI, 2001.
(47) Apra`, E. et al. NWChem, A Computational Chemistry Package for
Parallel Computers, 4.7; Pacific Northwest National Laboratory: Richland, WA,
2005.
spectra, and Cartesian coordinates for all optimized cumyl
cations. This material is available free of charge via the Internet
at http://pubs.acs.org.
JO9000383
(48) Clough, J.; Diorazio, L.; Widdowson, D. Synlett 1990, 761–762.
3322 J. Org. Chem. Vol. 74, No. 9, 2009