Influence of the boron atom on the solvatochromic properties of 4-nitroaniline-functionalized boronate esters

Article abstract of DOI:10.1021/jo300530k

Para-nitroaniline derivatives with peripheral 1,2- and 1,3-diol functionalities [O₂N-C₆H₄-NR ¹-CH₂CH(OH)CH₂OH; O₂N-C ₆H₄-NR¹-CH(CH₂OH)₂; R¹ = -H, -CH₃] covalently bonded to the amino group are esterified with various para-substituted phenylboronic acids [R ²-C₆H₄-B(OH)₂; R² = -OCH₃, -CH₃, -H, -Br, -CHO, -NO₂, -B(OH) ₂], and the solvatochromic properties of these esters are investigated in 33 solvents of different polarity. To interpret the solvent effects, the established linear solvation energy (LSE) multiparameter equations of Kamlet-Taft and the improved Catalan scales are used. Although the boron atom is separated by two or three sp³-hybridized carbon atoms from the actual chromophore, solvation effects have a significant positive solvatochromic effect on the nitroaniline unit (R¹ = -CH₃) as result of the solvent acting as a donor at the boron atom. The influence of the substituent R² on the coefficient b of the LSE relationship according to Kamlet-Taft and Catalan, which reflects the quantitative influence of the hydrogen-bonding acceptor or the electron-pair donor capacity of the solvent on the position of the UV-vis absorption maximum, can be determined via a linear Hammett relationship [b = f(σ_p)]. The interpretation of the effects is based on the electronic influence of the solvated boronic acid ester unit on the 4-nitroaniline group, predominantly through inductive interactions.

Full text of DOI:10.1021/jo300530k

Article
pubs.acs.org/joc
Influence of the Boron Atom on the Solvatochromic Properties of
4‑Nitroaniline-Functionalized Boronate Esters
Katja Hofmann and Stefan Spange*
Department of Polymer Chemistry, Chemnitz University of Technology, Strasse der Nationen 62, D-09117 Chemnitz, Germany
S
* Supporting Information
ABSTRACT: Para-nitroaniline derivatives with peripheral
1,2- and 1,3-diol functionalities [O2N−C6H4−NR1−CH2CH-
(OH)CH2OH; O2N−C6H4−NR1−CH(CH2OH)2; R1 = −H,
−CH3] covalently bonded to the amino group are esterified
with various para-substituted phenylboronic acids [R2−C6H4−
B(OH)2; R2 = −OCH3, −CH3, −H, −Br, −CHO, −NO2,
−B(OH)2], and the solvatochromic properties of these esters
are investigated in 33 solvents of different polarity. To
interpret the solvent effects, the established linear solvation
energy (LSE) multiparameter equations of Kamlet−Taft and
the improved Catalan
́
scales are used. Although the boron atom is separated by two or three sp3-hybridized carbon atoms from
the actual chromophore, solvation effects have a significant positive solvatochromic effect on the nitroaniline unit (R1 = −CH3)
as result of the solvent acting as a donor at the boron atom. The influence of the substituent R2 on the coefficient b of the LSE
́
relationship according to Kamlet−Taft and Catalan, which reflects the quantitative influence of the hydrogen-bonding acceptor
or the electron-pair donor capacity of the solvent on the position of the UV−vis absorption maximum, can be determined via a
linear Hammett relationship [b = f(σp)]. The interpretation of the effects is based on the electronic influence of the solvated
boronic acid ester unit on the 4-nitroaniline group, predominantly through inductive interactions.
Scheme 1. Molecular Structure of a 4-Nitroaniline-
Functionalized Boronic Acid Ester
INTRODUCTION
■
a
The linking of individual molecules to build up coupled
structures is of great interest for the preparation of materials
with specific adaptable properties. Here, classical chemical
methods can be combined with those of supramolecular
chemistry.1 An important criterion is often the reversibility of
the bond formation. Many methods exist, which use non-
covalent bonding, such as metal−ligand, ion−ion, and ion−
dipole interactions, hydrogen bonding, and also π−π-stacking
interactions.2−4 The reversible formation of covalent bonds can
also be used to construct coupled structures. As well as imine
formation, the reversible formation of boronic acid esters has
become an established method.5 Aromatic boronic acids readily
condense with diols to form cyclic esters. By combining
bifunctional molecules in a boronic acid-diol reaction, a large
number of possible combinations can be achieved.2,6,7 It is
therefore necessary to equip functional molecules, such as
chromophores, with peripheral diol functions.
4-Nitroaniline derivatives are widely used as indicators for
the determination of solvent polarity,8−10 as lipophilic
indicators for micelles and biological membranes,9−13 and as
guest molecules in cyclodextrins14 and thus represent very good
solvatochromic probe molecules. In this work, we present new
boronic acid esters, which carry a 4-nitroaniline building block
as the chromophoric unit, together with an available diol
function. The chromophoric system is separated from the
boronic acid by sp3-hybridized carbon and oxygen atoms
(Scheme 1).
a
The 4-nitroaniline chromophore and the Lewis-acidic boron are
separated by σ bonds.
Because of the vacant p-orbital of the boron atom, boronic
acid esters exhibit Lewis acid character and react readily with
specific Lewis bases such as fluoride or cyanide.15−17 We
decided to use these reactions and investigate the influence of
the boron atom and the ring size and thus the number of sp3-
hybridized carbons on the solvatochromic behavior of the 4-
nitroaniline-functionalized ester bonded to it. We tried to find
out if there is an influence on the chromophoric system that
Received: March 12, 2012
Published: May 8, 2012
© 2012 American Chemical Society
5049
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055

The Journal of Organic Chemistry
Article
could be measured using UV−vis spectroscopy in spite of the
separation of chromophore and boron atom by several σ bonds
(Scheme 1).
spite of these critical aspects, the Kamlet−Taft parameters have
been widely used for the interpretation of solute-influences of
physicochemical processes. An advantage of the new Catalan
́
In order to determine the influences of solvents,
solvatochromism has established itself as a simple and readily
applicable tool. It is possible, using suitable linear free energy
relationships (LSER) to describe quantitatively the influence of
solvents on chemical processes. Solute/solvent interactions
between solvatochromic dyes and pure solvents or solvent
mixtures are a combination of several independent influen-
ces.18−21 Multiple intermolecular solute/solvent interactions
can be described using the LSER concept. The most
sophisticated equation, which has been most widely and
successfully used for the quantification of solvent influences was
parameters lies in the fact that they are based on well-defined
single reference processes of carefully selected homomorphic
pairs of probe molecules.
RESULTS AND DISCUSSION
■
Synthesis. 1,2- and 1,3-diols react with boronic acids
forming cyclic five-membered and six-membered esters (1,3,2-
dioxaborolanes and 1,3,2-dioxaborinanes, respectively). Tol-
uene has proven to be a suitable solvent for this reaction,33
because it permits the removal of the water formed by
azeotropic distillation and also because the boronic acid esters
precipitate out on cooling the reaction mixture, while the
starting materials remain in solution. Scheme 2 shows the esters
4a−g, 5a, 5b, 6a, and 6b synthesized by this route.
proposed by Kamlet and Taft in 1976 (eq 1)8,22
.
(XYZ) = (XYZ)0 + a·α + b·β + s·π*
(1)
The hydrogen bond-donor capacity (HBD) α,22 the
hydrogen-bond acceptor capacity (HBA) β,8,23 and the
dipolarity/polarizability π*9,10 of a solvent can be expressed
using the simplified form of the Kamlet−Taft equation. (XYZ)0
corresponds to a standard process referenced to cyclohexane as
nonpolar medium; a, b, and s are solvent-independent
regression coefficients that reflect the contribution of each
parameter to the overall solute/solvent interaction.
Scheme 2. Synthesis of the Boronic Acid Esters 4−6
́
In the last ten years, Catalan et al. have developed three
alternative empirical solvent parameter scales in principle based
on the Kamlet−Taft parameters. Analogous to the Kamlet−
Taft α and β parameters, the solvent acidity, SA,24,25 and
solvent basicity, SB,26 scales have been established as a measure
of the HBD and HBA properties and the electron-pair donor
and acceptor properties (EPD and EPA) of the solvent. The
solvent’s polarizability/dipolarity is reflected by the SPP-value
and is comparable to the π*-value (eq 2).27,28
(XYZ) = (XYZ)0 + a·SA + b·SB + c·SPP
(2)
In the case of the borolanes 4, the substituent R2 of the
building block carrying the boron atom was varied. R2 is
separated from the chromophore by several sp3-hybridized
carbon and oxygen atoms in order to investigate the influence
of these functionalities on the boron and thus on the
solvatochromic behavior of the nitroaniline building block.
Both electron-donating (−OCH3, −CH3) and electron-accept-
ing (−Br, −CHO, −NO2) substituents were used. The
influence of additional hydrogen bonds on the solvatochrom-
ism was investigated by a variation of the substituent R1
(−OCH3, −H: 4a, 4e vs 5a, 5b). As well as the substituents,
the ring size of the ester and thus the number of sp3-hybridized
atoms between the chromophoric nitroaniline building block
and the boron atom can have an effect on the solvatochromic
behavior (5a, 5b vs 6a, 6b).
Until now, empirical polarity scales have always described the
polarity and dipolarity of the solvent in different proportions at
the same time. This requires that a change in the polarity of the
solvent is usually accompanied by a significant change in the
dipole moment of the solvatochromic probe. Abe has already
stated that these scales reach their limits in less polar
probes.29,30 The introduction of a π* correction parameter dδ
by Kamlet and Taft only allows a differentiation of aliphatic (δ
= 0), halogenated (δ = 0.5) and aromatic (δ = 1.0) solvents;
however, the development of alternative scales, such as that of
Buncel (πazo*)29,30 and Abe (π2*)29 were not able to solve this
problem satisfactorily. A general separation of solvent’s polarity
into an independent polarizability and a dipolarity scale was
needed. The first successful equation was suggested by Catalan
́
with the introduction of two independent scales for solvent
polarizability (SP)31 and solvent dipolarity (SdP)32 (eq 3).
Characterization. All boronate esters were obtained in
good yields (71−85%). Because of the stereogenic carbon in
the propyl fragment of the esters 4a−4g and 5a, 5b, the
(XYZ) = (XYZ)0 + a·SA + b·SB + d·SP + e·SdP
(3)
1
For the determination of the α, β, and π*-values, Kamlet and
Taft used a large number of solvent-dependent UV−vis
spectroscopic processes, among them a large number of
solvatochromic probe molecules. Seven reference probe
molecules and more than 40 other compounds were used in
the determination of the π*-value.28 A disadvantage of the
Kamlet−Taft parameters lies in the fact that they are not based
on well-defined single reference process; they represent the
average values from several solvent-dependent processes. In
assignment of the signals of the H NMR spectra is not trivial
and was carried out using 2D spectra (H,H COSY and C,H
1
COSY NMR, see the Supporting Information, Figure S7). H
and 13C NMR spectra were consistent with the molecular
structure of the esters for 4a−4g, 5a, 5b, 6a, and 6b. The signal
of the carbon directly bonded to the boron atom is not
observed because of line broadening as a result of the
quadrupole moment. In the 13C CP MAS NMR spectra, a
carbon signal due to the carbon next to the boron is observed,
5050
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055

The Journal of Organic Chemistry
Article
which is split because of the quadrupole moment. The 11B
NMR spectra show signals in the region of 30.9−33.0 ppm for
all borolanes 4a−4g, 5a, and 5b. These signals are characteristic
for 5-membered ring esters.15,16,34 The 11B NMR signal is
shifted to higher field with an increase in the electron-accepting
character of the substituent R2. The borinanes 6a and 6b show
an 11B NMR signal at lower field at 27.5 and 27.6 ppm,
respectively, in accordance with the literature.15,16
six selected solvents. For all the boronic acid esters investigated,
the strongest hypsochromic shift was observed in tetrachloro-
methane, while the strongest bathochromic shift was obtained
in formamide (Table 1). This corresponds to moderate
solvatochromic ranges (Δν
̃
= ν(hypsochromic)
̃
− ν(bathochromic)
̃
) of
Δν
̃
= 3300−3900 cm−1 in the case of the N-methylated
borolanes 4a−f and somewhat higher values of Δν
̃
= 4000−
4500 cm−1 for the NH-functionalized boronic acid esters 5a,
5b, 6a, and 6b. The solvatochromic range increases with the
increasing electron-withdrawing character of R2 (Scheme 2 and
Table 1).
The thermal stability of the boronic acid esters obtained was
investigated using thermogravimetric analysis. As can be seen in
Figure 1, the borolanes 4a−g are thermally stable up to ca. 300
The results of the multiple linear regression analysis using
the Kamlet−Taft parameters are shown in Table 2. The solvent
parameters α, β, and π* for the correlation are given in the
Supporting Information. The correlation coefficients r observed
are ≥0.95 for all regressions. This indicates that the
multiparameter equations are valid and give significant
information on the solvatochromic behavior. Figure 3 shows,
for example, the linear relationship between the measured and
calculated wavenumber for ester 4a.
Influence of the Substituent R2. The strongest influence on
the solvatochromic behavior of the borolanes 4a−4g is exerted
by the solvent’s dipolarity/polarizability π*, reflected by the
coefficient s. It has a negative value for all compounds and thus
indicates a positive solvatochromism. With increasing dipo-
larity/polarizability of the solvent, a bathochromic shift of the
UV−vis absorption maximum is observed. In the first excited
state, the boronic acid esters have a higher dipole moment than
in the ground state and are therefore better solvated by polar
solvents and thus stabilized.
Figure 1. Thermogravimetric analyses of the borolanes 4a−g and the
diol-functionalized nitroaniline 1. A temperature program of 20−400
°C with a heating rate of 20 K min−1 under a helium atmosphere was
used.
The influence of the hydrogen-bonding capacity α of the
solvent, expressed by the coefficient a, is also always negative.
Protic solvents, which can act as hydrogen-bond donors, mainly
interact with the oxygen atoms of the nitro group. As a result of
this, the (−M)-effect of the nitro group is increased, whereby
the push−pull character of the aromatic system is also
reinforced. The UV−vis absorption maximum is thus shifted
to higher wavelengths. Surprisingly, borolanes 4a−g show that
there is a significant influence of the hydrogen-bonding
acceptor capacity β of the solvent, although the chromophore
does not have any typical HBD groups, such as OH or NH
functions. The investigation of the solvatochromic behavior of
the starting material 1 shows that there is a clear effect with
regard to β from the specific solvation of the diol group,
although this group is separated from the chromophoric system
by the σ-bonds of the propyl group.33 In the case of the ester,
such an interaction with the OH-groups is no longer possible.
Therefore it can be concluded that specific solvation the sp2-
hybridized boron atom takes place, whereby the negative
coefficient b is obtained. Analogous to the diol-functionalized
nitroaniline 1, this is noteworthy, as the boron atom is
separated from the aromatic push−pull system by several σ-
bonds of the propyl group. In addition, it is established that the
contribution of the coefficient b increases with increasing
electron-withdrawing character of the substituent R2. Stronger
electron-withdrawing effects on the aromatic ring lower the
electron density at the boron atom, which increases its Lewis
acidity. The electron deficit can be compensated for by
interaction with the surrounding solvent molecules. The
stronger the electron deficit on the boron, the stronger is its
interaction with the solvent, which is reflected in a higher
contribution by b. Structure−property relationships of
substituents can be described by the σp Hammett constants.35
°C, whereas the decomposition of the corresponding diol-
functionalized nitroaniline 1 begins already at about 200 °C.
Not only is the diol function protected by the reaction with
aromatic boronic acids, but the thermal stability is also
increased. Because of sublimation, investigation of the thermal
stability of the NH-substituted borolanes 5a and 5b and the
borinanes 6a and 6b was not possible.
Solvatochromism. The investigation of the solvatochromic
properties of the boronic acid esters was carried out in 33
solvents of different dipolarities/polarizabilities and hydrogen-
bonding capacities so that the solvents used covered a range as
wide as possible, thus having a large variance. Figure 2 shows,
for example, the UV−vis absorption spectra of 4a measured in
Figure 2. UV−vis spectra of 4a measured in six solvents of different
polarity and hydrogen-bonding capacities.
5051
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055

The Journal of Organic Chemistry
Article
Table 1. UV−vis Absorption Maxima of 4a−g, 5a, 5b, 6a, and 6b in Six Solvents of Different Polarity and Hydrogen Bonding
Capacities
νmax )
10−3 (cm−1
̃
a
comp.
TCM
toluene
DCM
ethanol
DMSO
formamide
̃
Δν )
(cm−1
4a
4b
4c
4d
4e
4f
27.40
27.47
27.40
27.55
27.62
27.78
b
26.67
26.74
26.67
26.88
26.88
26.81
26.67
27.86
27.93
28.01
28.33
25.64
25.77
25.71
25.77
25.77
25.84
25.71
27.03
27.17
27.32
27.40
25.51
25.51
25.38
25.51
25.51
25.58
25.58
25.84
25.97
26.11
26.11
24.57
24.63
24.39
24.33
24.45
24.63
24.45
25.06
25.13
25.32
25.38
24.04
24.15
24.10
24.10
24.10
24.10
24.10
24.78
24.70
24.94
24.81
3360
3320
3300
3450
3590
3680
2640
4010
4380
3960
4510
4g
5a
5b
6a
6b
28.82
29.07
28.90
29.33
a
b
Tetrachloromethane. Sample probe is insoluble.
Table 2. Solvent-Independent Correlation Coefficients a, b, and s of the Kamlet−Taft Parameters α, β, and π*, Solute Property
of the Reference System νmax,0
(Cyclohexane, 103 cm−1), Significance (f), Correlation Coefficient (r), Standard Deviation (SD),
and Number of Solvents (n) Calculated for the Solvatochromism of Compounds 4a−g, 5a, 5b, 6a, and 6b
̃
comp.
νmax,0
̃
a
b
s
n
r
SD
f
4a
4b
4c
4d
4e
4f
27.992
28.126
28.095
28.399
28.440
28.503
28.042
29.502
29.824
29.776
30.159
−0.920
−0.887
−0.852
−0.908
−0.895
−0.919
−0.837
−0.897
−0.952
−0.883
−0.950
−0.636
−0.639
−0.728
−0.877
−0.946
−1.036
−0.729
−1.644
−1.972
−1.785
−2.220
−2.733
−2.878
−2.956
−3.216
−3.230
−3.172
−2.880
−3.160
−3.365
−3.101
−3.366
33
33
33
33
33
33
32
33
33
33
33
0.954
0.959
0.968
0.967
0.966
0.962
0.956
0.955
0.962
0.966
0.961
0.252
0.241
0.217
0.241
0.242
0.282
0.235
0.312
0.314
0.270
0.332
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
4g
5a
5b
6a
6b
Figure 4. Linear relationship between the σp-constants and the b
coefficients of eq 4 determined using the Kamlet−Taft equation (eq
1).
Figure 3. Relationship between calculated and measured νmax
for 4a, according to the Kamlet−Taft equation (eq 1).
̃
values
This gives σp = −0.03 and is of the same order of magnitude as
the σp value for the H-atom (σp = 0) or the B(OCH3)2
substituent (σp = 0.04).35,36
Influence of the Substituent R1. Besides the substituent
influences on the solvatochromic behavior of the boronic acid
esters discussed, the substituents R1 (Scheme 2) directly
bonded to the aromatic push−pull system have a significant
influence as expected. While the coefficients a (HBD capacity
of the solvent), determined for the borolanes 5a and 5b, are of
the same order of magnitude as those of the borolanes 4a−g,
the coefficients s have a somewhat higher value. This is because
in comparison with the N-methylated chromophores, the
influence of the dipolarity/polarizability in the case of the NH-
If the coefficients b in eq 1 are compared with the σp constants
of substituents described in the literature, there is a linear
relationship (eq 4), as shown in Figure 4.
b = −0.418·σp − 0.741
(4)
This confirms the conclusions already drawn about the
influence of the substituents on the boron-substituted aromatic
system on the solvatochromic behavior of the borolanes. Using
the linear eq 4 obtained, a σp constant for the 4-nitroaniline-
functionalized borolane substituent of 4g can be estimated.
5052
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055

The Journal of Organic Chemistry
Article
Table 3. Solvent-Independent Correlation Coefficients a, b, d, and e of the Catalan Parameters SA, SB, SP, and SdP, Solute
́
Property of the Reference System νmax,0
̃
(Gas Phase, 103 cm−1), Significance (f), Correlation Coefficient (r), Standard Deviation
(SD), and Number of Solvents (n) Calculated for the Solvatochromism of Compounds 4a−g, 5a, 5b, 6a, and 6b
comp.
νmax,0
̃
a
b
d
e
n
r
SD
f
4a
4b
4c
4d
4e
4f
29.446
29.734
29.737
30.121
30.078
30.218
29.438
31.590
31.252
31.883
32.063
−1.321
−1.272
−1.221
−1.191
−1.161
−1.118
−1.062
−1.246
−1.007
−1.374
−1.284
−0.469
−0.435
−0.476
−0.601
−0.640
−0.805
−0.483
−1.495
−1.729
−1.636
−2.051
−2.922
−3.209
−3.194
−3.390
−3.272
−3.334
−2.816
−3.750
−2.846
−3.780
−3.490
−1.823
−1.960
−2.045
−2.288
−2.329
−2.273
−2.070
−2.299
−2.627
−2.193
−2.542
32
32
32
32
32
32
32
32
32
32
32
0.953
0.957
0.970
0.971
0.974
0.959
0.968
0.953
0.965
0.960
0.926
0.222
0.217
0.184
0.199
0.190
0.245
0.174
0.296
0.281
0.271
0.301
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
4g
5a
5b
6a
6b
functionalized chromophores increases. A possible explanation
is that because of the C,H-hyperconjugation of the methyl
group, the chromophoric system of the borolanes 4 is extended
further than that of the borolanes 5. The most significant
difference is obtained, however, with regard to the HBA
properties of the solvent. Much higher values of the coefficient
b are found. As well as the specific solute/solvent interaction at
the boron atom that has already been discussed, there is an
interaction of the solvent with the NH-group. The (+M)-effect
of the amino group is thus raised, and consequently, the
aromatic push−pull system is strengthened, which results in a
bathochromic shift (b < 0) and an increase in the influence of
the HBA solvents. Analogously to the borolanes 4a−g, the
contribution of b on changing from the electron-donating
substituent OCH3 to the electron-accepting substituent −CHO
increases, which supports the claims already made.
multiple linear regression analyses according to eq 2. This
suggests that possibly one or more of the Catalan parameters
́
for formamide do not describe the interactions of the solvent
sufficiently precisely. Although correlation coefficients of r ≥
0.93 are obtained, which are thus only slightly lower than those
of the Kamlet−Taft correlation, a clear trend with regard to the
hydrogen-bond acceptor and electron-pair donor capacity
(coefficient b) can be observed. The solvent parameters SA,
SB, and SPP and the results of the multiple regression analysis
are given in the Supporting Information.
Use of the new four-parameter equation (eq 3) given by
́
Catalan, incorporating the separated polarizability scale SP and
the dipolarity scale SdP leads, on the other hand, to a very good
correlation (r ≥ 0.95). The results obtained using eq 3 are
shown in Table 3. The solvent parameters SP and SdP are
given in the Supporting Information.
Thus, not only the influence of the nonspecific polarizability
and dipolarity interactions can be separated and described in
detail, but also the influence of the substituents R2 with regard
to the HBD and EPA capacity of the solvent can be found.
Also, formamide can now be incorporated into the correlation.
Consequently, the SPP-value of this solvent appears to be
Influence of the Ring Size. The comparison of the results of
the multiple linear regression analysis between the borolanes
5a/5b and the borinanes 6a/6b allows conclusions to be drawn
with regard to the influence of the ring size and thus the effect
of the number of sp3-hybridized carbon atoms between the
boron atom and the chromophore of the ester. The coefficients
a and s are found to be of the same order of magnitude for all
four chromophores and reflect the previously mentioned
interaction with regard to the solvent’s HBD capacity and
dipolarity/polarizability. In spite of the greater ring strain in the
5-ring esters 5a/5b and the resulting higher Lewis acidity of the
boron atom,15 a smaller coefficient b is obtained in comparison
with the six-ring esters 6a/6b. In the case of 6a and 6b, the
boron atom is separated symmetrically by two sp3-hybridized
carbon atoms from the chromophoric nitroaniline unit, while in
the esters 5a/5b, the separation is by two and three sp3-
hybridized carbon atoms. This short separation in the case of
the borinanes 6a/6b attenuates the higher Lewis acidity
observed at the boron atom of the borolanes 5a/5b and clearly
shows the strong effect of the distance between the boron atom
and the chromophore. In analogy to the borolanes 4a−f, there
is an increase in the contribution of b on moving from electron-
donating to electron-accepting substituents (5a → 5b; 6a →
6b).
prone to error, or in other words, the SPP scale of Catalan
results that are lower than those of the π* scale of Kamlet−
Taft. Catalan was able to show a relationship between his four
́
gives
́
solvent parameters and the π* value of Kamlet and Taft (eq 5).
π* = (1.48 0.09)SP + (0.74 0.03)SdP
− (0.11 0.03)SB + (0.08 0.04)SA
− (0.89 0.06)
(5)
Consequently, π* is comprised mainly of contributions from
SP and SdP (2:1), while the proportions of SA and SB are
marginal. Comparable results between the Kamlet−Taft
equation established in the literature with π* and the four-
́
parameter equation of Catalan can also be obtained if, on the
one hand, the influence of the polarizability is greater than the
dipolarity and, on the other hand, these nonspecific interactions
dominate in comparison with the specific interactions (SA and
SB). This is the case with the boronic acid esters investigated.
The polarizability of the solvent has the strongest influence
on the solvatochromic behavior of boronic acid esters, reflected
by the coefficients d. The ratio of the influence of the
polarizability and that of the dipolarity of the solvent (d/e) is
between 1.4 and 1.7 and falls off if the boron-substituted
aromatic system has strongly electron-withdrawing substituents.
́
Kamlet−Taft vs Catalan. Analogous to the Kamlet−Taft
equation, the correlation of the UV−vis absorption maxima of
the boronic acid esters according to Catalan uses SA, SB, and
SPP as a three-parameter equation. It is noteworthy that unlike
the Kamlet−Taft equation (eq 1), the values for formamide
cannot be incorporated in the correlation in the case of all
5053
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055

The Journal of Organic Chemistry
Article
For example, in the case of R2 = NO2, strong delocalization of
the electrons at the boron-substituted aromatic system is
possible, whereby the dipolar part (coefficient d) of the
interaction with the solvent increases, reflected in a smaller d/e
ratio.
equation of Catalan
́
, in which the solvent’s polarizability and
dipolarity are treated separately. The strongest influence on the
solvatochromic behavior is exerted by the polarizability and the
dipolarity of the solvent. Positive solvatochromism, i.e., a
bathochromic band shift, is found with an increase of the
polarizability and dipolarity of the solvent. We were able to
show that the boron atom has a clear effect on the position of
the UV−vis absorption maximum, although it is removed from
the chromophore system by σ-bonds, and this is reflected in the
coefficient b (HBD/EPA capacity). In addition, this effect
depends on the electron-donating or -withdrawing ability of the
substituents on the boron-substituted aromatic ring, which can
be determined using the Hammett free-energy equation.
With an increase in the influence of the HBA capacity of the
solvent (e.g., |b4f > b4a|) on the solvatochromic behavior, the
influence of the HBD capacity of the solvent (e.g., |a4f < a4a|) is
lowered. Solvents that act as HBD or as EPA can interact with
the free electron pair of the amino group, as well as the oxygen
atoms of the nitro group (a < 0). Thus, the (+M)-effect of the
NR1 group (R1 = −H, −CH3) is lowered, and the aromatic
push−pull system is weakened, resulting in a coefficient a with
a positive sign. These interactions compete with each other. If
the electron deficiency at boron is increased by an electron-
withdrawing substituent R2, and this is compensated for by
stronger interactions with the surrounding solvent molecules,
the charge on the amino group of the chromophore increases.
Correspondingly, the proportion of the interaction of the EPA
with the solvent increases at this point, resulting in a smaller
contribution of a on going from 4a to 4f. Because of the
different reference systems for determination of the solvent
EXPERIMENTAL SECTION
■
The diol-functionalized 4-nitroanilines 1−3 were synthesized as
previously described.33
General Method for the Synthesis of Boronate Esters.
Equimolar amounts of diol-functionalized 4-nitroaniline and the
corresponding boronic acid were suspended in 80 mL of toluene
and heated to 100 °C. The mixture was stirred for 1 h at this
temperature, whereby a clear solution was formed. The water formed
was removed by azeotropic distillation from the reaction mixture; the
yellow precipitate formed was filtered off, washed with toluene, and
dried in vacuo to afford the corresponding ester of the boronic acid in
yields of 71−85%.
The boronate ester 4a, 4c−e, and 4g were already described in ref
33, but no 11B NMR data are reported. 4g is only sparingly soluble;
therefore, its 11B NMR measurement was not possible.
rac-N-{[2-(4-Methoxyphenyl)-1,3,2-dioxaborolan-4-yl]-
methylene}-N-methyl-4-nitroaniline (4a). 11B NMR (80 MHz,
CDCl3) δ 32.0.
́
parameters, the electron-pair interactions on the Catalan scale
are stronger than those from the Kamlet−Taft equation. As has
already been discussed in the case of Kamlet−Taft, the
contribution of coefficient b increases with the strength of
the electron pull at the boron-substituted aromatic system. A
linear relationship between b and σp has been found (eq 6).
b = −0.347·σp − 0.514
(6)
Using this eq 6, a σp value of −0.08 can be determined for
the nitroaniline-functionalized borolane substituent of 4g. This
value is of the same order of magnitude as that obtained
according to eq 4 using the coefficient b of the Kamlet−Taft
equation. Figure 5 shows the example of compound 5a,
summarizing the possible specific solute/solvent interactions
affecting the solvatochromic behavior.
rac-N-{[2-(4-Phenyl)-1,3,2-dioxaborolan-4-yl]methylene}-N-
methyl-4-nitroaniline (4c). 11B NMR (80 MHz, CDCl3) δ 31.8.
rac-N-{[2-(4-Bromophenyl)-1,3,2-dioxaborolan-4-yl]-
methylene]-N-methyl-4-nitroaniline (4d). 11B NMR (80 MHz,
CDCl3) δ 31.6.
rac-N-{[2-(4-Formylphenyl)-1,3,2-dioxaborolan-4-yl]-
methylene}-N-methyl-4-nitroaniline (4e). 11B NMR (80 MHz,
CDCl3) δ 31.4.
rac-N-{[2-(4-Methylphenyl)-1,3,2-dioxaborolan-4-yl]-
methylene}-N-methyl-4-nitroaniline (4b). Yield 0.25 g (88%):
yellow powders of mp 94−96 °C; 1H NMR (250 MHz, CDCl3) δ 2.38
(s, 3H), 3.22 (s, 3H), 3.63 (dd, J = 15.5 Hz, J = 7.4 Hz, 1H), 3.73 (dd,
J = 15.5 Hz, J = 4.1 Hz, 1H), 4.06 (dd, J = 9.5 Hz, J = 6.6 Hz, 1H),
4.49 (dd, J = 9.5 Hz, J = 8.1, 1H), 4.81−4.92 (m, 1H), 6.70 (d, J = 9.3
Hz, 2H), 7.21 (d, J = 7.9 Hz, 2H), 7.67 (d, J = 7.9 Hz, 2H), 8.13 (d, J
= 9.3 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 21.8, 40.1, 57.1, 68.6,
75.6, 109.6, 110.6, 126.1, 128.8, 134.9, 142.2, 153.5; 11B NMR (80
MHz, CDCl3) δ 32.0; IR (KBr, cm−1) ν
̃
= 3037, 2971, 1595, 1513,
1402, 1362, 1316; Anal. Calcd. for C17H19BN2O4 (326.15) C, 62.60,
H, 5.87, N, 8.59; Found C, 62.96, H, 5.91, N, 8.60; UV/vis εDCM (388
nm) 20450 L mol−1 cm−1.
Figure 5. Possible hydrogen-bond (HB) and electron-pair (EP)
interactions between solute and solvents and their influence on the
UV−vis absorption demonstrated for chromophore 5a.
rac-N-{[2-(4-Nitrophenyl)-1,3,2-dioxaborolan-4-yl]-
methylene}-N-methyl-4-nitroaniline (4f). Yield 0.30 g (75%):
1
yellow powders of mp 169−171 °C; H NMR (250 MHz, CDCl3) δ
3.24 (s, 3H), 3.68 (dd, J = 15.5 Hz, J = 7.4 Hz, 1H), 3.79 (dd, J = 15.5
Hz, J = 4.1 Hz, 1H), 4.13 (dd, J = 9.5 Hz, J = 7.1 Hz, 1H), 4.57 (dd, J
= 9.5 Hz, J = 8.1 Hz, 1H), 4.89−4.99 (m, 1H), 6.71 (d, J = 9.5 Hz,
2H), 7.94 (d, J = 8.2 Hz, 2H), 8.15 (d, J = 9.5 Hz, 2H), 7.94 (d, J = 8.2
Hz, 2H); 13C NMR (100 MHz, CDCl3) δ = 40.2, 56.3, 69.0, 76.3,
110.6, 122.7, 126.2, 135.9, 137.7, 150.2, 153.3; 11B NMR (80 MHz,
CONCLUSION
The solvatochromic behavior of the boronic acid esters 4−6
was used to compare the solvent parameter equations of
■
́
Kamlet−Taft and Catalan. Using the linear solvation energy
CDCl3) δ 30.8; IR (KBr, cm−1) ν
̃
3112, 2909, 1600, 1515, 1475, 1351,
(LSE) relationship of Kamlet−Taft, very good correlations have
1312 Anal. Calcd. for C16H16BN3O6 (357.13) C, 53.81, H, 4.52, N,
11.77; Found C, 54.20, H, 4.65, N, 11.77; UV/vis εDCM (387 nm)
20890 L mol−1 cm−1.
been obtained. The use of the analogous three-parameter
́
equation involving Catalan’s SPP scale is not suitable for
describing the solvatochromic behavior of the chromophores
investigated. However, results that are comparable with the
Kamlet−Taft results were obtained using the four-parameter
(R)-N-{[2-(4-Methoxyphenyl)-1,3,2-dioxaborolan-4-yl]-
methylene}-4-nitroaniline (5a). Yield 0.12 g (77%): yellow crystals
1
of mp 134−136 °C; H NMR (250 MHz, CDCl3) δ 3.34−3.40 (m,
5054
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055

The Journal of Organic Chemistry
Article
1H), 3.51−3.57 (m, 1H), 3.84 (s, 3H), 4.08 (dd, J = 9.4 Hz, J = 7.8
Hz, 1H), 4.50 (dd, J = 9.4 Hz, J = 7.8 Hz, 1H), 4.87 (t, J = 5.5 Hz,
1H), 6.62 (d, J = 9.4 Hz, 2H), 6.92 (d, J = 8.6 Hz, 2H), 7.75 (d, J = 8.6
Hz, 2H), 8.10 (d, J = 9.4 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ
47.9, 55.5, 69.0, 76.1, 111.9, 114.0, 126.8, 137.1, 139.1, 153.4, 163.0;
(4) Grabowski, S. J. Chem. Rev. 2011, 111, 2597.
(5) Christinat, N.; Scopelliti, R.; Severin, K. Angew. Chem. 2008, 120,
1874; Angew. Chem., Int. Ed. 2008, 47, 1848.
(6) Sanchez
Rojas-Lima, S. Chem.Eur. J. 2002, 3, 612.
(7) Christinat, N.; Croisier, E.; Scopelliti, R.; Cascella, M.;
́
, M.; Hopfl, M.; Ochoa, M.-E.; Frafan, N.; Snatillan, R.;
́
̈
11B NMR (80 MHz, CDCl3) δ 31.7; IR (KBr, cm−1) ν
̃
3383, 3166,
Rothlisberger, U.; Severin, K. Eur. J. Inorg. Chem. 2007, 5177.
2901, 1602, 1534, 1479, 1384, 1329, 1215; Anal. Calcd. for
C16H17BN2O5 (328.13) C, 58.52, H, 5.22, N, 8.54; Found C, 58.53,
H, 5.19, N, 8.53; UV/vis εDCM (369 nm) 17730 L mol−1 cm−1.
(R)-N-{[2-(4-Formylphenyl)-1,3,2-dioxaborolan-4-yl]-
methylene}-4-nitroaniline (5b). Yield 0.25 g (85%): yellow
̈
(8) Taft, R. W.; Kamlet, M. J. J. Am. Chem. Soc. 1976, 98, 377.
(9) Kamlet, M. J.; Hall, T. N.; Boykin, J.; Taft, R. W. J. Org. Chem.
1979, 44, 2599.
(10) Kamlet, M. J.; Abboud, J.-L. M.; Taft, R. W. J. Am. Chem. Soc.
1977, 99, 6027.
1
powders of mp 155−157 °C; H NMR (250 MHz, CDCl3) δ 3.42
(dd, J = 13.6 Hz, J = 7.4 Hz, 1H), 3.60 (dd, J = 13.6 Hz, J = 3.8 Hz,
1H), 4.16 (dd, J = 9.5 Hz, J = 7.0 Hz, 1H), 4.57 (dd, J = 9.5 Hz, J = 8.1
Hz, 1H), 4.84−4.95 (m, 1H), 6.64 (d, J = 9.5 Hz, 2H), 7.88 (d, J = 8.2
Hz, 2H), 7.97 (d, J = 8.2, 2H), 8.11 (d, J = 9.5 Hz, 2H), 10.07 (s, 1H,
NH); 13C{1H} CP MAS NMR (400 MHz, 12 kHz) δ 49.0, 71.0, 75.9,
109.4, 114.6, 126.5, 133.1, 137.3, 139.0, 154.1, 194.5; 11B NMR (80
(11) Novaki, L. P.; El Seoud, O. A. Ber. Bunsenges. Phys. Chem. 1996,
100, 648.
(12) Helburn, R.; Dijiba, Y.; Mansour, G.; Maxka, J. Langmuir 1998,
14, 8856..
(13) Mansour, G.; Creedon, W.; Dorrestein, P. C.; Maxka, J.; Mac
Donald, J. C.; Helburn, R. J. Org. Chem. 2001, 66, 4050.
MHz, CDCl3) δ 30.7; IR (KBr, cm−1) ν
̃
3377, 3068, 2944, 1701, 1604,
́
(14) Lo Meo, P.; DAnna, F.; Gruttadauria, M.; Riela, S.; Noto, R.
1535, 1478, 1364, 1324; Anal. Calcd. for C16H15BN2O5 (326.24) C,
58.86, H, 4.63, N, 8.59; Found C, 58.98, H, 4.69, N, 8.58; UV/vis εDCM
(367 nm) 16440 L mol−1 cm−1.
Tetrahedron 2004, 60, 9099..
(15) Hall, D. G. Boronic Acids; Wiley-VCH: Weinheim, 2005.
(16) Oehlke, A.; Auer, A. A.; Jahre, I.; Walfort, B.; Ruffer, T.; Zoufala,
́
̈
P.; Lang, H.; Spange, S. J. Org. Chem. 2007, 72, 4328.
(17) Badugu, R.; Lakowicz, J. R.; Geddes, C. D. Sens. Actuators, B
2005, 104, 103.
N-[2-(4-Methoxyphenyl)-1,3,2-dioxaborinan-5-yl]-4-nitroa-
niline (6a). Yield 0.17 g (76%): yellow crystals of mp 213−215 °C;
1H NMR (250 MHz, CDCl3) δ 3.84 (s, 3H), 3.98−4.03 (m, 1H),
4.12−4.19 (m, 2H), 4.35−4.41 (m, 2H), 6.59 (d, J = 9.3 Hz, 2H), 6.90
(d, J = 8.7, 2H), 7.74 (d, J = 8.7 Hz, 2H), 8.12 (d, J = 9.3 Hz, 2H);
13C{1H} CP MAS NMR (400 MHz, 7 kHz) δ 49.1, 54.6, 64.4, 66.7,
106.1, 111.3, 112.5, 117.0, 123.7, 124.7, 126.1, 128.5, 135.5, 136.0,
(18) Reichardt, C. Chem. Rev. 1994, 94, 2319.
(19) Marcus, Y. Chem. Soc. Rev. 1993, 22, 409.
(20) Reichardt, C.; Welton, T. Solvents and Solvents Effects in Organic
Chemistry, 4th ed.; Wiley-VCH: Weinheim, 2011.
151.8, 160.8; 11B NMR (80 MHz, CDCl3) δ 27.6; IR (KBr, cm−1) ν
̃
́
(21) Catalan, J. In Handbook of Solvents; Wypych, G., Ed.; ChemTech
3368, 3052, 2958, 1604, 1541, 1473, 1345, 1310, 1236; Anal. Calcd. for
C16H17BN2O5 (328.13) C, 58.57, H, 5.22, N, 8.44; Found C, 58.42, H,
5.20, N, 8.49; UV/vis εDCM (366 nm) 17970 L mol−1 cm−1.
Publishing: Toronto, 2001; Chapter 10.3, p 583ff.
(22) Taft, R. W.; Kamlet, M. J. J. Am. Chem. Soc. 1976, 98, 2886.
(23) Yokoyama, T.; Kamlet, M. J; Taft, R. W. J. Am. Chem. Soc. 1976,
98, 3233.
N-[2-(4-Formylphenyl)-1,3,2-dioxaborinan-5-yl]-4-nitroani-
line (6b). Yield 0.13 g (76%): yellow crystals of mp 203−205 °C; 1H
NMR (250 MHz, CDCl3) δ 4.03−4.11 (m, 1H), 4.17−4.23 (m, 2H),
4.40−4.46 (m, 2H), 6.62 (d, J = 9.2 Hz, 2H), 7.86 (d, J = 8.1 Hz, 2H),
7.96 (d, J = 8.1 Hz, 2H), 8.13 (d, J = 9.2 Hz, 2H), 10.06 (s, 1H);
13C{1H} CP MAS NMR (400 MHz, 12 kHz) δ 48.8, 63.6, 64.7, 107.5,
(24) Catalan
(25) Catalan
(26) Catalan
Rodríguez, J.-G. Liebigs Ann. 1996, 1785.
́
́
, J.; Díaz, C. Liebigs Ann. 1997, 1941.
, J.; Díaz, C. Eur. J. Org. Chem. 1999, 885.
, J.; Díaz, C.; Lopez, V.; Perez, P.; de Paz, J.-L. G.;
́ ́
́
(27) Catalan
(28) Catalan, J.; Lop
J.-G. Liebigs Ann. 1995, 241.
́
, J.; Lop
́
ez, V.; Per
́
ez, P. Liebigs Ann. 1995, 793.
115.0, 126.8, 136.5, 138.0, 139.2, 139.9, 153.5, 196.2; 11B NMR (80
́
́
ez, V.; Per
́
ez, P.; Martín-Villamil, R.; Rodríguez,
MHz, CDCl3) δ 27.5; IR (KBr, cm−1) ν
̃
3320, 3069, 2972, 1689, 1604,
1541, 1472, 1364, 1307; Anal. Calcd. for C16H15BN2O5 (326.24) C,
58.86, H, 4.63, N, 8.59; Found C, 58.86, H, 4.70, N, 8.53; UV/vis εDCM
(364 nm) 16970 L mol−1 cm−1.
(29) Lide D. R., Ed.; CRC Handbook of Chemistry and Physics, 75th
ed.; CRC Press: Boca Raton (Florida), 1994; cf, Abe, T. Bull. Chem.
Soc. Jpn. 1990, 63, 2328.
(30) McClellan, A. L. Tables of Experimental Dipole Moments; Rahara
Enterprises: El Cerrito, CA, 1989; Vol. 3.
ASSOCIATED CONTENT
■
(31) Catalan
́
, J.; Hopf, H. Eur. J. Org. Chem. 2004, 4694.
S
* Supporting Information
1
Copies of H and 13C NMR spectra of 4b, 4f, 5a, 5b, 6a, and
(32) Catalan
́
, J. J. Phys. Chem. B 2009, 113, 5951.
(33) Spange, S.; Hofmann, K.; Walfort, B; Ruffer, T; Lang, H. J. Org.
̈
́
6b, Kamlet−Taft and Catalan solvent parameter sets, UV−vis
absorption maxima of all compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
Chem. 2005, 70, 8564.
(34) Noth, H.; Wrackmeyer, B. Nuclear Magnetic Resonance
̈
Spectroscopy of Boron Compounds; Springer Verlag: Berlin, 1978.
(35) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(36) Oehlke, A.; Auer., A. A.; Schreiter, K.; Hofmann, K.; Riedel, F.;
Spange, S. J. Org. Chem. 2009, 74, 3316.
AUTHOR INFORMATION
Corresponding Author
*E-mail: stefan.spange@chemie.tu-chemnitz.de.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Financial support by the DFG and the Fonds der Chemischen
Industrie is gratefully acknowledged.
■
REFERENCES
(1) Hullinger, J. Spektrum Wiss. 2000, 12.
(2) Schmittel, M.; Mahata, K. Angew. Chem. 2008, 120, 5364; Angew.
Chem., Int. Ed. 2008, 47, 5284.
■
(3) Fischer, R. C.; Power, P. P. Chem. Rev. 2010, 110, 3877.
5055
dx.doi.org/10.1021/jo300530k | J. Org. Chem. 2012, 77, 5049−5055