Chemometric analysis of substituent effects. XI. Solvent effects on dissociation of 2,6-disubstituted benzoic acids

Article abstract of DOI:10.1135/cccc19970913

Eleven symmetrically 2,6-disubstituted benzoic acids (with the following substituents: OCH₃, OC₂H₅, OC₃H₇, OCH(CH₃)₂, OC₄H₉, CH₃, F, Cl, Br, I, a

Full text of DOI:10.1135/cccc19970913

Chemometric Analysis of Substituent Effects
913
CHEMOMETRIC ANALYSIS OF SUBSTITUENT EFFECTS. XI. SOLVENT
EFFECTS ON DISSOCIATION OF 2,6-DISUBSTITUTED BENZOIC ACIDS
1
2
Jiri KULHANEK and Oldrich PYTELA
Department of Organic Chemistry, University of Pardubice, 532 10 Pardubice, Czech Republic;
e-mail: 1 kulhanek@hlb.upce.cz, 2 pytela@hlb.upce.cz
Received December 17, 1996
Accepted February 10, 1997
Eleven symmetrically 2,6-disubstituted benzoic acids (with the following substituents: OCH3, OC2H5,
OC3H7, OCH(CH3)2, OC4H9, CH3, F, Cl, Br, I, and NO2) have been synthesized and their dissoci-
ation constants measured potentiometrically in methanol, ethanol, propan-1-ol, propan-2-ol, butan-2-ol,
acetone, dimethyl sulfoxide, dimethylformamide, acetonitrile, pyridine, and 1,2-dichloroethane. The
experimental data obtained have been analyzed from the point of view of solvent effects on acidity
of the individual derivatives. Different behaviour found with benzoic acid and the disubstituted deri-
vatives in protic solvents is due to changes in solvation. The different character of solvation of ben-
zoic acid and the disubstituted derivatives depends on the type of substitution, being manifested only
in 2,6-disubstituted benzoic acids. The graphical analysis has shown a distinct trend in the increase
of magnitude of deviation of the point of benzoic acid in the series: propan-2-ol, butan-2-ol, propan-1-ol,
ethanol, methanol. This order correlates with the steric demands of carbon chain of the alcohols used.
The abnormal behaviour of benzoic acid in the dissociation in these alcohols as compared with that
of its 2,6-disubstituted derivatives is due to the different extent of solvation of the reaction centre
caused by steric hindrance. Against the expectation, benzoic acid appears to be a weaker acid in
protic solvents, whereas its alkoxy derivatives are stronger acids. The solvation also minimizes the
inductive effect of alkoxy groups in the symmetrically 2,6-disubstituted derivatives. In aprotic sol-
vents the acidity of 2,6-dialkoxybenzoic acids is also increased, in this case as a result of sterically
forced deviation of the reaction centre and/or the substituents out of the plane of benzene ring.
Key words: Substituent effects; ortho Effect; Solvation; Dissociation constants.
Our previous contributions dealt with the problem of ortho effect in the dissociation of
monosubstituted1 and disubstituted benzoic acids2. In the disubstituted acids we proved
statistical insignificance of mutual interactions between the substituents and, hence, the
additivity of substituent effects2. On the other hand, the interactions between solvents
and the substituents turned out to be a significant factor. For this reason, the present
paper is focused on the problem of solvent effects on the dissociation of symmetrically
disubstituted benzoic acids. The symmetrically disubstituted derivatives were chosen to
exclude the influence due different effects in ortho positions of the substituents present.
The aim of the present communication is to give an integral self-contained set of pKHA
values of symmetrically 2,6-disubstituted benzoic acids (missing in the literature) and
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

914
Kulhanek, Pytela:
subsequently to analyze the effects of the solvents used on the dissociation of selected
derivatives.
The solvation of carboxylic and/or carboxylate group as a reaction centre belongs
among the most significant effects affecting the acidity of the acids measured. Quanti-
tatively the extent of solvation, and hence of stabilization of molecule, can be estimated
from the magnitude of reaction constant in the Hammett relation: it is involved in this
constant. The extent of solvation can absolutely be quantified e.g. by comparing the
dissociation constants of substituted benzoic acids measured in organic solvents and in
gas phase3–5
.
The proportion of stabilization of the base formed by substituents is generally lower
in amphiprotic (protic) solvents (the ρ constant is near 1 for the dissociation of sub-
stituted benzoic acids), which is predominantly caused by efficient solvation of the
conjugate base formed (neutral and protogenic solvents), or, as the case may be, by a
better solvation of proton than that of the conjugate base (protophilic solvents). The
stabilization of conjugate base of substituted derivatives is also aided by the solvation
of substituents that can form hydrogen bonds6. A considerable role is played by the
solvent in the solvation of substituted salicylic acids7–11 and aminobenzoic acids12.
Whereas in protic solvents the hydroxy groups present are significantly solvated, a
formation of intramolecular hydrogen bond between OH (or NH2) group and the reac-
tion centre was observed in aprotic solvents8,13. The comparison of extent of solvation
of benzoic acid derivatives by selected alcohols14,15 and some aprotic solvents8 is dis-
cussed in the papers quoted.
Aprotic solvents do not solvate well the conjugate base, which increases the effect of
substituents on its stabilization (the ρ constant for dissociation of substituted benzoic
acids16 is greater than 2). In some cases, pyridine can form an exception17, its conjugate
acid being able to stabilize the anion formed. A low sensitivity to substitution was also
observed in the dissociation of 2,6-disubstituted benzoic acids in little polar tetrahydro-
furan8. The different affinity to the proton in protophilic and protophobic aprotic sol-
vents has no substantial effect on the substitution sensitivity18, which indicates a
roughly identical extent of solvation of the proton being dissociated by protophilic and
protophobic solvents. Hence the stabilization of conjugate base proceeds in similar
ways in these two groups of aprotic solvents.
The inability of inert solvents to stabilize the charged particles formed in the disso-
ciation of acids has the consequence of large effect of substituents on the stabilization
of conjugate base manifested in a high value of the reaction constant in the Hammett
equation. Rather surprising in this respect is 1,2-dichloroethane17,19 in which the disso-
ciation of substituted benzoic acids exhibits ρ ≈ 1.7. This anomaly can probably be
explained by the presence of a small amount of methanol, a distinctly more polar and
protic solvent, in the nonpolar and inert 1,2-dichloroethane at the half-value of titration
end point (the titration is carried out with methanolic tetrabutylammonium hydroxide).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Chemometric Analysis of Substituent Effects
915
EXPERIMENTAL
The symmetrically disubstituted benzoic acids containing the substituents OCH3, OC2H5, OC3H7,
OCH(CH3)2, OC4H9, CH3, F, Cl, Br, I, and NO2 were synthesized by known or partly modified pro-
cedures. 2,6-Dichlorobenzoic and 2,6-dimethylbenzoic acids were commercial products (Fluka). The
1
identity of the acids was verified with the help of H and 13C NMR spectra and GC/MS. The purifi-
cation methods after preliminary reprecipitation from the solution of the respective salt, the melting
points and purity of the individual derivatives determined by HPLC are given in Table I.
Synthesis of 2,6-Diiodobenzoic Acid
A 500 ml three-necked flask equipped with a stirrer, separatory funnel and thermometer was charged
with 2-amino-6-nitrobenzoic acid20 (18.2 g, 0.1 mol), water (150 ml) and concentrated sulfuric acid
(55 ml). The mixture was heated to 80 °C with stirring and then quickly cooled. The suspension
formed was diazotized at 5 °C by adding a solution of sodium nitrite (6.9 g, 0.1 mol) in water (40 ml)
within 30 min. After adding the last portion of the nitrite, the reaction mixture was stirred for another
60 min, whereafter the solution of diazonium salt was poured in a mixture of potassium iodide (31 g,
0.18 mol) and water (100 ml) during 10 min. The resulting mixture was boiled 1 h and then the
excess iodide was removed by adding a small amount of dipotassium disulfite and the solution was
cooled. The separated 2-iodo-6-nitrobenzoic acid was filtered off and recrystallized from water to
give 22.4 g (76%) product melting at 184–189 °C (ref.21 gives m.p. 188–189 °C).
TABLE I
Purification procedures, melting points and purity of the symmetrically 2,6-disubstituted benzoic
acids used
Purification
Entry No.
Substituent
M.p., °C
M.p., ref.21, °C
Purityb, %
procedurea
c
1
2
OCH3
W
E
187–189
130–132
55–59
188
132–134
54–56
107–108
82–84
115–116
157.5
100.0
98.9
98.6
99.0
98.2
100.0
99.1
100.0
97.7
99.6
97.3
OC2H5
3
OC3H7
F
4
OCH(CH3)2
C
104–106
81–84
5
OC4H9
P
c
6
CH3
W
W
W
T
115–116
157–158
142–144
145–148
184–187
201–202
7
F
8
Clc
Br
I
144
9
150–151
–
10
11
S
c
NO2
W
202–203
a Recrystallization from: C tetrachloromethane, E aqueous ethanol, P petroleum ether, W water, T toluene,
F flash chromatography (silica gel, petroleum ether–diethyl ether 4 : 1), S reprecipitation. b HPLC (Se-
paron SGW C 18, 150 × 3 mm, 60% aqueous methanol, 1% phosphoric acid, flow rate 1 ml min–1,
detector setting at 230 nm).c Ref.2.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

916
Kulhanek, Pytela:
2-Iodo-6-nitrobenzoic acid (1.5 g, 5 mmol) was dissolved in acetic acid (20 ml) and the solution
was treated with tin(II) chloride (4 g, 21 mmol) and concentrated hydrochloric acid (5 ml). The mix-
ture was stirred at 50 °C 1 h, cooled, and the precipitated yellow amino acid was collected by filtra-
tion, washed with acetic acid, and dried. Yield 0.8 g (59%) 2-amino-6-iodobenzoic acid, m.p.
175–177 °C (ref.21 gives m.p. 162 °C).
2-Amino-6-iodobenzoic acid (2.5 g, 9.5 mmol) was dissolved in a mixture of hot water (15 ml)
and concentrated sulfuric acid (11 ml), cooled, and diazotized by adding a solution of sodium nitrite
(2.5 g, 36 mmol) in water (15 ml) at 5 °C until positive reaction of nitrous acid. The solution of
diazonium salt was stirred 20 min, filtered, the filtrate was poured in a solution of potassium iodide
(6.2 g, 37 mmol) in water (20 ml), and the mixture was heated at 75 °C 30 min. After removing the
excess potassium iodide with disodium disulfite, cooling, and reprecipitation, the yield of 2,6-diiodo-
benzoic acid was 0.15 g (4%), m.p. 184–187 °C. For C7H4I2O2 (373.9) calculated: 22.47% C, 1.08% H;
found: 22.77% C, 1.06% H.
Potentiometric Measurements
The potentiometric titration was carried out on a Radiometer set consisting of an automatic burette
ABU 93, controlling element TIM 90, and stirrer SAM 90. The volume of solution analyzed was 2 ml.
The titration reagent was added dynamically, depending on the change of electric potential. During
the titration, the sample was thoroughly stirred and argon free of carbon dioxide and saturated with
vapours of the corresponding solvent was bubbled through. The samples were titrated in random
order, each acid being titrated three times. The calibration of electrode system using benzoic acid as
the standard was carried out after every 3 measurements. The measuring system involved a glass
electrode G 240 B (Radiometer) and a modified calomel electrode K 4040 (Radiometer), the modifi-
cation consisting in replacement of the inner aqueous solution of potassium chloride by the saturated
methanolic solution of potassium chloride. The glass electrode was hydrated in 0.1 M HCl 12 h be-
fore the measurements in the individual solvents. In the cases of strong dehydration of the membrane
and slow response of the electrode and/or shortening of the potential jump, the electrode was hy-
drated in the same way for a shorter period also during the measurements. If not used in the meas-
urements, the glass electrode was kept in an aqueous buffer of pH 7.00. The titrator was connected
to a PC AT 386 computer and using an evaluation programme the signal obtained was transformed
to values of electric potential E (mV) of the individual samples at the half-value of titration end
point. The pKHA values of the 2,6-disubstituted benzoic acids were calculated from the potential E
values measured for the individual acids and the E0 values of the respective standards using Eq. (1).
pKHA = pK0 + (E – E0)/59.16
(1)
RESULTS AND DISCUSSION
The values of differences ∆pKHA of symmetrically 2,6-disubstituted benzoic acids and
parent benzoic acid together with their standard deviations s measured in methanol
(MeOH), ethanol (EtOH), propan-1-ol (PrOH), propan-2-ol (iPrOH), butan-2-ol
(sBuOH), acetone (Ac), dimethyl sulfoxide (DMSO), N,N-dimethylformamide (DMF),
acetonitrile (AN), pyridine (Py), and 1,2-dichloroethane (DCE) are presented in Table II
(the known pKHA values of benzoic acid: methanol22 9.41, ethanol23 10.25, propan-1-ol24
8.60, propan-2-ol 10.71, acetone25 18.20, dimethyl sulfoxide26 11.00, N,N-dimethylform-
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Chemometric Analysis of Substituent Effects
917
Collect. Czech. Chem. Commun. (vol. 62) (1997)

918
Kulhanek, Pytela:
amide27 12.27, acetonitrile28 20.70, pyridine18 9.80, 1,2-dichloroethane29 20.00). The
standard deviations of repeated measurements varied within the range usual for titra-
tions in non-aqueous media, i.e. 0.1 pKHA units.
Analysis of Solvation Effects by Means of Analysis of Variance
In order to analyze the effects of individual solvents on the substituent–solvent interac-
tion we carried out the analysis of variance separately on the pKHA set of symmetrically
substituted 2,6-dialkoxybenzoic acids determined in protic solvents (MeOH, EtOH,
PrOH, iPrOH, sBuOH) and on the pKHA set of the same derivatives measured in aprotic
media (Ac, DMSO, DMF, AN, Py, DCE). This is a situation involving two factors and
interaction: the first factor is represented by the solvent and the second by the substi-
tuent. Before the calculation, the set was centered by subtracting the pKHA value of
benzoic acid from the pKHA values of substituted derivatives. The decomposition of
variability according to this model is given in Table III. The value of F criterion for
substitution indicates an approximately twofold effect of substituents on the dissoci-
ation in alcohols as compared with that in aprotic solvents at comparable accuracy of
the measurements (the aprotic solvents 0.050 pKHA units, the protic solvents 0.053
pKHA units). This higher sensitivity of dissociation to substituents in protic solvents is
caused by different extent of specific solvation of the individual alkoxy groups by the
alcohols used, which results in a changed magnitude of their inductive effect. The con-
clusion just stated is confirmed also by the increased value of F criterion for the sol-
vent–substituent interaction in the pKHA set of dialkoxy derivatives measured in
alcohols as contrasted by the corresponding quantities obtained in aprotic medium,
TABLE III
The factors monitored (Sol solvent, Sub substitution), sums of squares S, degrees of freedom n,
values of F criterion and critical values of Fisher–Snedecor distribution Fcrit at the significance level
α = 0.05 in the model of analysis of variance with interactions (the pKHA set of symmetrically sub-
stituted 2,6-dialkoxybenzoic acids)
Aprotic solvent
Protic solvent
Factors
S
n
F
Fcrit
S
n
F
Fcrit
Sol
3.97
1.42
0.39
0.15
5.95
5
4
318
144
8
2.37
2.53
1.75
7.42
3.22
4
658
286
9
2.56
2.56
1.85
Sub
4
16
50
74
Sol + Sub X
Residual
Total
20
60
89
0.42
0.14
11.20
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Chemometric Analysis of Substituent Effects
919
which is statistically significant in both the cases. As the value of this factor is statisti-
cally significant also for the other derivatives2, it can be stated that the solvent–substi-
tuent interaction plays a significant part in the dissociation of 2,6-disubstituted benzoic
acids.
Analysis of Solvation Effects of Protic Solvents
In order to find out the effects of the selected solvents on the behaviour of individual
derivatives of benzoic acid in the dissociation, we analyzed the experimental data
graphically. When plotting the values of differences of dissociation constants of sym-
metrically 2,6-disubstituted benzoic acids and parent benzoic acid in aprotic solvents
(Ac, DMSO, DMF, AN, Py, DCE) against the corresponding quantities in methanol
(Fig. 1) we can see a deviation of the point belonging to the parent benzoic acid (solid
circles) from the expected linear dependence. As this is the substance used for the
standardization of the pKHA values of all the other derivatives measured, the result
mentioned cannot be due to a gross experimental error. Since methanol differs in its
properties from aprotic solvents, it can be presumed that the deviation of the point of
benzoic acid is caused by this very solvent. For confirmation of this thesis we plotted
1.5
∆pK
0
0
∆pK
1
2
3
4
5
6
1
2 3
4
5
–2
–2
–4
–4
–2
0
2
4
–4
–2
0
2
4
∆pK
∆pK
MeOH
DMSO
FIG. 1
FIG. 2
Dependences of ∆pKHA values of 2,6-disub-
stituted benzoic acids in aprotic solvents
(straight lines: 1 Ac, 2 DMSO, 3 DMF, 4 AN,
5 Py, 6 DCE) on ∆pKHA in methanol; solid
circles denote ∆pKHA of benzoic acid (to be
better seen the individual straight lines are
shifted by one pKHA unit to the right)
Dependences of ∆pKHA values of 2,6-disub-
stituted benzoic acids in aprotic solvents
(straight lines: 1 Ac, 2 DMF, 3 AN, 4 Py, 5
DCE) on ∆pKHA in dimethyl sulfoxide; solid
circles denote ∆pKHA of benzoic acid (to be
better seen the individual straight lines are
shifted by one pKHA unit to the right)
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

920
Kulhanek, Pytela:
in mutual dependences the ∆pKHA values of 2,6-disubstituted benzoic acids measured
only in aprotic solvents. In contrast to the above-mentioned results concerning meth-
anol, no deviation of the point for disubstitution of 2-H, 6-H takes place. A clear
example is the dependences of ∆pKHA of 2,6-disubstituted benzoic acids in aprotic sol-
vents on the same quantities in dimethyl sulfoxide given in Fig. 2. Thus the presump-
tion of specific effect of methanol as a protic solvent on the behaviour of the individual
derivatives was unambiguously confirmed.
For finding out the effect of substitution on the different character of solvation of the
2,6-disubstituted derivatives and unsubstituted benzoic acid in methanol we analyzed
the data for 3,4- (ref.16) and 3,5-disubstituted benzoic acids19 in a similar way to that
used for the 2,6-disubstituted benzoic acids (the derivatives with combinations of the
substituents CH3, OCH3, Cl, NO2 (ref.2)). As it can be seen from the dependences of
∆pKHA of these derivatives in dimethyl sulfoxide on the same quantities in methanol
(Fig. 3, dependences 1 and 2), no deviation can be observed for the unsubstituted ben-
zoic acid in contrast to the 2,6-derivatives (Fig. 3, dependence 3). This fact agrees with
the findings by Chantooni and Kolthoff15 concerning an analogous behaviour of ben-
zoic acid in the dissociation and of some of its 3,4- and 3,5-disubstituted derivatives.
Hence it can be stated that the different character of solvation of benzoic acid and its
disubstituted derivatives depends on the type of substituents, being only manifested in
the 2,6-disubstituted benzoic acids.
For a more detailed investigation of character of solvation of the diad reaction
centre–substituent in protic solvents, the set of dissociation constants of symmetrically
2,6-disubstituted benzoic acids in the above-mentioned solvents was completed by the
measurements in ethanol, propan-1-ol, butan-2-ol, and propan-2-ol. By plotting the dif-
ferences ∆pKHA of 2,6-disubstituted benzoic acids and parent benzoic acid in acetoni-
trile on the analogous quantities in alcohols we obtained the dependences depicted in
0
∆pK
DMSO
–1
FIG. 3
Dependences of ∆pKHA values of dis-
–2
ubstituted benzoic acids (straight
lines: 1 3,4-disubstituted, 2 3,5-disub-
1
2
3
–3
–4
stituted, 3 2,6-disubstituted) in dimethyl
sulfoxide on ∆pKHA of these derivatives in
methanol; solid circles denote ∆pKHA of
benzoic acid (to be better seen the indi-
vidual straight lines are shifted by one or
two pKHA units to the right)
–2
–1
0
1
2
∆pK
3
MeOH
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Chemometric Analysis of Substituent Effects
921
Fig. 4 (the regression straight lines were only calculated for the substituents unable of
formation of intramolecular hydrogen bond). Practically identical dependences were
obtained by plotting the ∆pKHA obtained also in the remaining aprotic solvents against
the respective quantities in alcohols. From Fig. 4 we can see a clear trend in the mag-
nitude of deviation of the point of benzoic acid in the series: propan-2-ol, butan-2-ol,
propan-1-ol, ethanol, methanol. This order correlates with the steric demands of carbon
chain of the alcohols used. On the basis of the given correlation it can be claimed that
the abnormal behaviour of benzoic acid in dissociation in these alcohols is due to the
different extent of solvation of its reaction centre and the reaction centres of disub-
stituted derivatives, which is a consequence of the steric hindrance to the approach of
solvent forced by the substituents present. Due to this phenomenon, propan-2-ol (as
contrasted with the other alcohols) shows the least specific solvation and its effect on
the acids measured is similar to that of the aprotic solvents. The fact described is also
confirmed by the value of differences of the classical residua eK of benzoic acid from
the regression straight lines in the above-mentioned dependences (Table IV). The re-
gression straight lines were only calculated for the substituents unable of forming an
intramolecular hydrogen bond. Whereas the greatest deviation value is observed for the
measurement in methanol, the deviation for the measurement in propan-2-ol can be
neglected with regard to the accuracy of measurement (0.05 pKHA units). The lower
values of deviations of benzoic acid from the regression straight line obtained for 1,2-
dichloroethane are caused by the application of methanolic solution of titration agent in
the potentiometric determination of the dissociation constants in this solvent17.
It is surprising to find that benzoic acid turns out to be a weaker acid than expected,
whereas their alkoxy derivatives appear to be stronger acids. A probable explanation of
this artefact is a deviation of COO– group out of the plane of benzene ring which is
connected with a more intense solvation. Another possible explanation can be a forma-
0
∆pK
AN
–1
FIG. 4
Dependences of ∆pKHA values of symmetri-
cally 2,6-disubstituted benzoic acids in ace-
tonitrile on ∆pKHA values of these derivatives
in alcohols (straight lines: 1 propan-2-ol, 2
butan-2-ol, 3 propan-1-ol, 4 ethanol, 5 meth-
anol), ● benzoic acid, ❍ 2,6-dialkoxybenzoic
acids, ❐2,6-dimethylbenzoic acid, ▲ 2,6-diha-
logenobenzoic acids, ∆ 2,6-dinitrobenzoic acid
(to be better seen the individual straight lines
are shifted by one pKHA unit to the right)
–2
–3
1
2
3
4
5
–4
–2
0
2
4
∆pK
alcohols
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

922
Kulhanek, Pytela:
tion of intramolecular hydrogen bond between hydrogen atom of alcohol and oxygen
atom of alkoxy group. This could make the alkoxy group a weaker electron donor as
compared with hydrogen substituent, and the subsequent increase in stabilization of
conjugate base leads to increased acidity of the respective dialkoxy derivative.
Differences in the extent of solvation and their influence on substituent effects are
also indicated by the arrangement of points of dialkoxy derivatives in Fig. 4. In methanol
the solvation suppresses the manifestations of inductive effect of individual alkoxy
groups to such an extent that the difference in acidities of individual derivatives is
slight (the points lie at the regression straight line). A quite opposite situation is en-
countered with propan-2-ol, where the solvation is restricted to the minimum by steric
interactions between the substituents and carbon chain of the alcohol with concomitant
more distinct manifestation of inductive effect of alkyls in alkoxyl groups (deviation of
points from the regression straight line). For the remaning alcohols the positions of
points of alkoxy derivatives is a resultant of operation of both effect. These results are
confirmed by the conclusions about the effects of protic solvents on the properties of
alkoxy groups made on the basis of analysis of variance.
Analysis of Solvent Effects of Aprotic Solvents
The above discussion about dialkoxy derivatives concerns the behaviour of these acids
in protic solvents. On the other hand, the behaviour of the same derivatives in aprotic
media is indicated by the arrangement of the points in the coordinate axis. Against
expectations, the alkoxy derivatives do not appear much weaker acids than benzoic acid
despite the positive mesomeric effect of their alkoxy groups. From this fact it can be
concluded that the reaction centre (carboxylic group) and/or the substituents are steri-
TABLE IV
_
Classical residuals eK for benzoic acid and mean values of this characteristic eK in the dependences
of ∆pKHA in aprotic solvents on the corresponding values in alcohols
eK
__
eK
Alcohol
Ac
DMSO
DMF
AN
Py
DCE
MeOH
EtOH
0.60
0.54
0.49
0.31
0.02
0.65
0.62
0.55
0.40
0.14
0.67
0.63
0.58
0.41
0.16
0.60
0.57
0.50
0.34
0.10
0.75
0.71
0.67
0.48
0.21
0.48
0.43
0.63
0.58
0.53
0.35
0.09
PrOH
0.36
sBuOH
iPrOH
0.18
–0.10
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Chemometric Analysis of Substituent Effects
923
cally forced out of the plane of benzene ring14,30. Hence the effect mentioned does not
influence the stability of conjugate base by mesomeric effect of substituent5, i.e. no
decrease in acidity of the derivatives discussed occurs in our case. An analogous effect
can be observed when comparing the pKHA values of 2-methoxybenzoic acid1 and the
sterically demanding 2,6-dimethoxy derivatives in the aprotic solvents mentioned. With
the same values for the standard, the 2,6-disubstituted derivative appears in dimethyl
sulfoxide (pKHA of the 2-substituted and 2,6-disubstituted derivatives are 11.21 and
11.29, respectively), N,N-dimethylformamide (12.66, 12.72), and pyridine (10.10,
10.30) to be an only slightly weaker acid than the monosubstituted derivative. In
acetone (18.94, 18.65), acetonitrile (21.24, 20.87), and 1,2-dichloroethane (20.77,
19.67) 2,6-dimethoxybenzoic acid becomes even more acidic than the monosubstituted
derivative. From the facts given follows such a dominant role of steric effect of methoxy
groups in the 2,6-disubstituted derivative in aprotic medium that the influence of me-
someric effect on the reaction centre is minimized.
The research work was sponsored by the Grant Agency of the Czech Republic, Grant No.
203/94/0122.
REFERENCES
1. Pytela O., Liska J.: Collect. Czech. Chem. Commun. 59, 2005 (1994).
2. Kulhanek J., Pytela O.: Collect. Czech. Chem. Commun. 60, 829 (1995).
3. Caldwell G., Rennebook R., Kebarle P.: Can. J. Chem. 67, 611 (1989).
4. Epshtein L. M., Iogausen A. V.: Usp. Khim. 59, 229 (1990).
5. Bagno A., Scorrano G.: J. Chem. Soc., Perkin Trans. 2 1991, 1601.
6. Niazi M. S. K., Ali J.: Bull. Chem. Soc. Jpn. 1990, 3619.
7. Kolthoff I. M., Chantooni M. K.: J. Am. Chem. Soc. 93, 3843 (1971).
8. Barbosa J., Barron N., Bosch E., Roses M.: Anal. Chim. Acta 265, 157 (1992).
9. Rodante F., Pistoria G.: Thermochim. Acta 1981, 325.
10. Reichardt C.: Solvent and Solvent Effects in Organic Chemistry, 2nd ed. Verlag Chemie,
Weinheim 1988.
11. Mohamed E. S., Mansaur E. S., Kassem A. A., El-Bardan A. A.: Egypt. J. Chem. 31, 105 (1990).
Chem. Abstr. 114, 163422 (1991).
12. Rubino J. T., Berryhill W. S.: J. Pharm. Sci. 1986, 182.
13. Aslam M. H., Chapman N. B., Charton M., Shorter J.: J. Chem. Res. (M) 1981, 2301.
14. Charton M., Charton B. I.: J. Org. Chem. 33, 3872 (1968).
15. Chantooni M. K., Kolthoff I. M.: Anal. Chem. 51, 133 (1979).
16. Pytela O., Kulhanek J., Ludwig M., Riha V.: Collect. Czech. Chem. Commun. 59, 627 (1994).
17. Ludwig M., Baron V., Kalfus K., Vecera M.: Collect. Czech. Chem. Commun. 51, 2135 (1986).
18. Ludwig M.: Ph.D. Thesis. Institute of Chemical Technology, Pardubice 1984.
19. Pytela O., Kulhanek J., Ludwig M.: Collect. Czech. Chem. Commun. 59, 1637 (1994).
20. Kahn R.: Chem. Ber. 35, 3857 (1902).
21. Dictionary of Organic Compounds. Chapman & Hall, New York 1982.
22. Kolthoff I. M., Guss L. S.: J. Am. Chem. Soc. 60, 2516 (1938).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

924
Kulhanek, Pytela:
23. Kolthoff I. M.: J. Phys. Chem. 35, 2732 (1931).
24. Taft R. W.: Prog. Phys. Org. Chem. 13, 135 (1981).
25. Foltin M., Majer P.: Collect. Czech. Chem. Commun. 43, 95 (1978).
26. Ritchie C. D., Uschold R. E.: J. Am. Chem. Soc. 90, 2821 (1968).
27. Barnes K. K., Mann C. K.: Anal. Chem. 36, 2502 (1964).
28. Kolthoff I. M., Chantooni M. K., Bhownik S.: J. Am. Chem. Soc. 88, 5430 (1966).
29. Bos M., Dahmen E. A. M. F.: Anal. Chim. Acta 63, 185 (1973).
30. McDaniel D. H., Brown M. C.: J. Org. Chem. 23, 420 (1958).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)