Effects of medium and substituents on dissociation of 4,4′-disubstituted bis(benzenesulfon)imides

Article abstract of DOI:10.1135/cccc19961205

Ten 4,4′-disubstituted bis(arenesulfon)imides of the general formula XC₆H₄SO₂NHSO₂C₆H ₄X have been synthesized and their structures confirmed by their ¹H NMR spectra. Elemental analyses are presented for the compounds not yet described. The dissociation constants of these model substances have been measured potentiometrically in pyridine, dimethylformamide, methanol, ethanol, propylene carbonate, acetone, acetonitrile, 1,2-dichloroethane and tetramethylene sulfone. The pK_HA values obtained have been correlated with three sets of the Hammett substituent constants and the results have been used to discuss the solvent and substituent effects on the dissociation of the compounds studied. Sulfonimides with electron-acceptor substituents behave as rather strong acids in some solvents (pyridine, dimethylformamide, methanol and ethanol), whereas normal substituent dependences are found in other solvents. The experimental data have also been interpreted with the help of the statistical methods based on latent variables. From the calculations it follows that only the first principal component, which correlates well with the substituent constant sets adopted, is statistically significant in describing the substituent effect on the acid-base process studied.

Full text of DOI:10.1135/cccc19961205

4,4’-Disubstituted Bis(benzenesulfon)imides
1205
EFFECTS OF MEDIUM AND SUBSTITUENTS ON DISSOCIATION
OF 4,4′-DISUBSTITUTED BIS(BENZENESULFON)IMIDES
1
Miroslav LUDWIG and Pavel STVERKA
Department of Organic Chemistry, Faculty of Chemical Technology, University of Pardubice,
532 10 Pardubice, Czech Republic; e-mail: 1 ludwig@hlb.upce.cz
Received March 26, 1996
Accepted July 19, 1996
Ten 4,4′-disubstituted bis(arenesulfon)imides of the general formula XC6H4SO2NHSO2C6H4X have
been synthesized and their structures confirmed by their 1H NMR spectra. Elemental analyses are
presented for the compounds not yet described. The dissociation constants of these model substances
have been measured potentiometrically in pyridine, dimethylformamide, methanol, ethanol, propylene
carbonate, acetone, acetonitrile, 1,2-dichloroethane and tetramethylene sulfone. The pKHA values ob-
tained have been correlated with three sets of the Hammett substituent constants and the results have
been used to discuss the solvent and substituent effects on the dissociation of the compounds studied.
Sulfonimides with electron-acceptor substituents behave as rather strong acids in some solvents (py-
ridine, dimethylformamide, methanol and ethanol), whereas normal substituent dependences are
found in other solvents. The experimental data have also been interpreted with the help of the statis-
tical methods based on latent variables. From the calculations it follows that only the first principal
component, which correlates well with the substituent constant sets adopted, is statistically signifi-
cant in describing the substituent effect on the acid-base process studied.
Key words: Sulfonamides; Dissociation; Substituent; Solvent; Chemometrics.
Hammett1 used substituted benzoic acids and their dissociation in water as the basic
model and model process, respectively, for studies of substituent effects in the benzene
nucleus. It is, of course, possible to observe effects of substituents and media also on
other types of substrates and in other media. The present paper, which continues a
series of papers dealing with dissociation of sulfonamides of various kinds2–7, is fo-
cused on the dissociation of 4,4′-disubstituted bis(arenesulfon)imides 1.
O
S
O
S
X
NH
X
O
O
1
Generally, the compounds having two sulfonyl groups and one hydrogen attached to
a nitrogen atom will behave as relatively strong acids since sulfonyl group is a power-
ful electron acceptor. Obviously, the substituent effects will be transmitted from the
benzene nucleus through sulfonyl group in a similar way as in monosubstituted ben-
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

1206
Ludwig, Stverka:
zenesulfonamides2–4. The combined effects of the two substituents should result (par-
ticularly in aprotic media) in a strong influence upon the stability of particles present in
the equilibrium mixture. The substitution at para position was chosen for the reason of
more complex influence of substituent at this position as compared with that at the
non-alternating meta position8.
Using the Beilstein XFIRE data base we found that there are 39 compounds of this
kind with para–para-disubstitution described in literature (the literature search was
performed for the following substituents: hydrogen, halogen, alkyl, alkoxy, alkylcarbo-
nyl, amino, nitro, and alkylcarbonylamino). No complex study dealing with dissoci-
ation of the said compounds has been found in literature. Dauphin et al.9 focused their
attention only on the pKHA values of unsubstituted (1.45), dimethyl (1.70) and diamino
(2.89) derivatives in water; the same value for the diamino derivative was also found by
other authors10. Although Dauphin’s paper9 presents a calculation of parameters of the
Hammett relation, the number of experimental points is evidently insufficient. Other
1
literature data about the substrates studied by us involve the H NMR spectra of
diamino, dimethyl, dibromo and unsubstituted derivatives11, 13C NMR spectra of di-
methyl and dichloro derivatives12, IR spectra of the parent substance and its dimethyl
derivative13, mass spectrum of the unsubstituted parent substance14, and a study of anti-
bacterial activity of the diamino derivative (sulfanilimide)15.
The chief aim of the present work is to discuss the combined substituent effects from
both nuclei of the model substrates on the stability of the conjugated bases and on
solvation of the conjugated bases by solvents of various types. When formulating the
problem we anticipated that the simultaneous action of both substituents can decide
whether the substance will behave as a weak or a strong acid.
EXPERIMENTAL
Measurements
The 1H NMR spectra of the model compounds prepared were measured with a Bruker AMX 360
apparatus using 5% solutions of the substances in (CD3)2SO and referring the chemical shifts to the
solvent signal. The pKHA values of the A- and B-type imides studied were determined in methanol
(MeOH), ethanol (EtOH), acetonitrile (AN), dimethylformamide (DMF), pyridine (Py), acetone (Ac),
propylene carbonate (PC), 1,2-dichloroethane (DCE), and tetramethylene sulfone (TMS) by means of
potentiometric titration using an automatic titrator Radiometer RTS-622 or Radiometer Titrilab 3
with the same electrode arrangement and titration agent as in our previous works2–7. The solvents
used were purified by standard methods.
Synthesis of p-Substituted Bis(arenesulfon)imides
Except for the diamino derivative, the bis(arenesulfon)imides were obtained by the reaction of the
corresponding benzenesulfonamide with a slight excess of benzenesulfonyl chloride in alkaline me-
dium16. The benzenesulfonamides were obtained by reacting the benzenesulfonyl chlorides with ex-
cess ammonia2. The benzenesulfonyl chlorides were obtained by direct chlorosulfonation of benzene
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

4,4’-Disubstituted Bis(benzenesulfon)imides
1207
derivatives2 or by the Sandmeyer reaction of the respective benzenediazonium chlorides with cuprous
chloride in glacial acetic acid saturated with silver oxide2,17. The starting benzene and aniline deriva-
tives were commercial samples except for fluorobenzene which was prepared from benzenediazo-
nium salt18. The preparation of bis(aminobenzenesulfon)imide started from acetanilide which was
transformed into 4-acetylaminobenzenesulfonyl chloride19 and then (by the above-described method)
into bis(acetylaminobenzenesulfon)imide which was reacted with concentrated NaOH solution20 to
give the required amino derivative. Table I presents the melting points, reaction temperatures and
yields of the bis(arenesulfon)imides, yields of the preparation of the arenesulfonyl chlorides and
arenesulfonamides, and elemental analyses of the methoxy and acetyl derivatives.
RESULTS AND DISCUSSION
1
The H NMR chemical shifts of all the model substituted bis(arenesulfon)imides pre-
pared are presented in Table II. The chemical shifts of the acidic proton are not in-
cluded because they mostly formed very broad signals. A paper by DeChristoper11
gives the shifts of the parent compound and its methyl, nitro, and bromo derivatives
which agree well with our values. However, the paper11 only gives a single value for
the whole multiplet of aromatic section of the molecules. The values of dissociation
constants of the model substrates in the individual solvents are given in Table III in the
___
form of the average value pKHA (together with the respective standard deviations).
When discussing the results one must realize the difficult comparison of pH in the
individual nonaqueous media which follows from the fact that no absolute scale has
been defined so far. The solvents were chosen so as to cover the main solvent types
most frequently used for titrations.
For orientation let us give the pKHA values of the standards used in the solvents
studied by us (the following survey of solvents gives in succession the name of solvent,
pKHA of benzoic acid, and pKHA of picric acid with the corresponding references): am-
phiprotic solvents (methanol – 9.41, 3.67, refs23,24; ethanol – 10.25, 4.10, refs25,26),
dipolar aprotic protophobic solvents (acetonitrile – 20.70, 11.00, refs27,28; acetone –
18.20, 9.20, refs29,30; tetramethylene sulfone – 26.30, 17.40, refs30,31; propylene carbo-
nate – 19.70, 9.30, ref.31), dipolar aprotic protophilic solvents (dimethylformamide –
12.27, 3.65, refs27,32; pyridine – 9.80, 3.50, refs33,34), and inert solvents (1,2-dichloro-
ethane – 20.00, 13.70, ref.35). The substituents were selected so as to cover the whole
spectrum of substituent σ constants. They can be classified as substituents without any
distinct mesomeric effects (CH3, H, Cl, Br), those with positive mesomeric effects
(NH2, OCH3, NHCOCH3, F), and those with negative mesomeric effects (COCH3,
NO2).
The experimental results obtained were treated by several mathematical statistical
methods. The treatment by simple linear regression involved all the experimental points
obtained (each experiment was repeated three or four times). The methods working
with latent variables – the principal component analysis (PCA)36, conjugated deviation
analysis (CDA)37, and projection of latent structures (PLS)38 – adopted the average
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

1208
Ludwig, Stverka:
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

4,4’-Disubstituted Bis(benzenesulfon)imides
1209
___
values pKHA. The calculation by the method of simple linear regression according to the
classical Hammett equation (pKHA = pK0 – ρσ, adapted from ref.1) was carried out
separately for all the media studied using three sets of substituent constants, viz. σP (the
universal constants according to Exner39), σP6 and σF6 (the constants designed for appli-
cations to nonaqueous media40). In all the media, the statistical characteristics of the
results obtained with σP were the worst ones, the results of calculations using the other
two sets of substituent constants being of comparable quality. This difference is due to
the fact that the σP6 and σF6 constants were determined by a statistical calculation using
the data for dissociation of benzoic acids in organic solvents whereas the σP constants
are rather restricted to processes in aqueous media. Table IV presents the regression
parameters and statistical characteristics of the Hammett relation using the σF6 constant
set. With propylene carbonate, acetone, acetonitrile, and tetramethylene sulfone sol-
vents the correlation coefficients are above 0.997 when using the constants for organic
solvents. The lower value observed with 1,2-dichloroethane (0.992) can be caused by
specific effects of this solvent or lower precision of measurements in inert media. The
distinctly worse values of statistical characteristics in pyridine, dimethylformamide,
methanol, and ethanol are caused by the lower value of slope ρ and, in addition to it –
for dimethylformamide and pyridine – by the considerable remoteness of the pKHA
values for the amino derivative. The small slope value is probably due to the fact that
the bis(benzenesulfon)imides studied behave as rather strong acids in methanol, etha-
nol, pyridine and dimethylformamide (except just for the amino derivative in Py and
DMF). The different behaviour of these acids in both groups of solvents also confirms
TABLE II
1H NMR spectra of 4,4′-disubstituted bis(arenesulfon)imides
X
1H NMR, δ, ppm
NH2
4.5 s (H-7, H2O and (ArSO2)2NH); 6.64 m, 4 H (H-3,5); 7.45 m, 4 H (H-2,6)
3.86 s, 6 H (H-7); 7.00 m, 4 H (H-3,5); 7.63 m, 4 H (H-2,6);
2.39 s, 6 H (H-7); 7.28 d, 4 H (H-3,5); 7.61 d, 4 H (H-2,6)
OCH3
CH3
NHCOCH3 2.12 s, 6 H (H-7 (NHCOCH3)); 7.63 m, 8 H, 2 × 2 AA′ BB′ (H-2,6 and H-3,5);
10.26 m, 2 H (H-7 (NHCOCH3))
H
7.48 m, 4 H (H-3,5); 7.52 m, 2 H (H-7); 7.74 m, 4 H (H-2,6)
7.23 m, 4 H (H-3,5); 7.71 m, 4 H (H-2,6)
F
Cl
7.49 m, 4 H (H-3,5); 7.68 m, 4 H (H-2,6)
Br
7.61 m, 8 H, 2 × 2 AA′ BB′ (H-2,6 and H-3,5)
2.64 s, 6 H (H-7); 7.79 m, 4 H (H-2,6); 7.97 m, 4 H (H-3,5)
7.94 m, 4 H (H-2,6); 8.28 m, 4 H (H-3,5)
COCH3
NO2
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

1210
Ludwig, Stverka:
the value of the 2ρBS/ρ ratio (ρBS is the value of reaction constant for substituted ben-
zenesulfonamides2–4). If neglecting the different solvation of the two substrate types
(benzenesulfonamides and bis(benzenesulfon)imides), one should expect the value 1
for this ratio. Values close to 1 are found in Table IV for solvents with high-quality
Hammett correlations (Ac, AN, TMS, DCE), whereas, on the other hand, the values of
the fraction are much higher (as high as 18.72 in the case of pyridine) for solvents with
low correlation coefficients. The fraction confirms that the two types of substrates be-
have differently in these solvents. Logically, the effect of substituents on stabilization
of conjugated base is much lower for strong acids (i.e. the value of reaction constant is
lower too) than for weaker acids. The value of fraction for the solvents in which imides
TABLE III
Average values of pKHA (and their standard deviations s(pKHA)) of model substrates in individual
solvents: Py pyridine, DMF dimethylformamide, MeOH methanol, EtOH ethanol, PC propylene car-
bonate (2-oxo-5-methyl-1,3-dioxolane), Ac acetone, AN acetonitrile, DCE 1,2-dichloroethane, TMS
tetramethylene sulfone (1,1-dioxotetrahydrothiophene, tetrahydrothiophene 1,1-dioxide)
___
___
pKHA/s(pKHA
)
X
Py
DMF MeOH EtOH
PC
Ac
–
AN
DCE
TMS
NH2
4.27
(0.14) (0.05)
6.44
–
–
12.73
(0.10)
14.27
(0.10)
–
20.79
(0.06)
OCH3
CH3
3.63 4.26
4.33
4.82
10.59
11.59
12.52
14.61
18.78
(0.06) (0.05) (0.04) (0.04) (0.06) (0.02) (0.04)
3.61 3.87 4.12 4.84 10.22 11.19 12.03
(0.10) (0.04) (0.02) (0.11) (0.04) (0.03) (0.08)
(0.04) (0.10)
14.35 18.45
(0.05) (0.03)
NHCOCH3
3.65
(0.10) (0.08)
4.10
–
–
–
–
11.98
(0.10)
–
18.33
(0.10)
H
3.56 3.66
3.78
4.45
9.54
10.63
11.44
13.95
18.12
(0.02) (0.05) (0.03) (0.05) (0.10) (0.02) (0.05)
(0.09) (0.07)
13.40 16.96
(0.07) (0.11)
13.16 16.64
(0.14) (0.05)
13.16 16.80
(0.15) (0.07)
12.67 16.27
(0.09) (0.06)
11.79 14.54
(0.07) (0.13)
F
–
3.44
3.49
4.23
8.93
9.82
10.76
(0.05) (0.01) (0.05) (0.07) (0.07) (0.07)
Cl
3.45
3.46 3.24 4.09 8.30 9.38 10.25
(0.06) (0.12) (0.01) (0.05) (0.01) (0.02) (0.05)
3.45 3.46 3.37 4.14 8.33 9.43 10.33
(0.05) (0.07) (0.04) (0.04) (0.05) (0.04) (0.03)
3.48 3.35 3.40 3.63 7.93 9.13 9.84
(0.07) (0.05) (0.05) (0.04) (0.08) (0.03) (0.06)
3.47 3.31 3.02 3.92 6.31 7.71 8.16
(0.03) (0.05) (0.03) (0.13) (0.03) (0.03) (0.04)
Br
COCH3
NO2
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

4,4’-Disubstituted Bis(benzenesulfon)imides
1211
behave as weak acids indicates that e.g. DCE destabilizes bis(benzenesulfon)imide
anion more than benzenesulfonamide anion etc. The effect of medium on pK0 values
can be discussed but roughly because of the limitations connected with non-existence
of a universal pH scale for all media. With regard to the fact that similar stand-
ardization was used for the measurements in all the solvents, it is possible to carry out
such a discussion. The sequence of pK0 values in the individual media agrees well with
that of pKHA values of the standard picric acid in these solvents. It somewhat differs
from that in the earlier published works dealing with the dissociation of benzenesulfon-
amides2–4, benzoic acids29,41,42, and N-phenylsulfonylbenzamides5.
For benzenesulfonamides the lowest pK0 value was observed in Py, and it increased
in the order: MeOH, EtOH, DMF, Ac, AN, DCE, and TMS. In the case of benzoic
acids the respective sequence was: MeOH, Py, EtOH, DMF, Ac, DCE, AN, TMS; for
N-phenylsulfonylbenzamides: Py, MeOH, DMF, AN; for bis(arenesulfon)imides: Py,
DMF, MeOH, EtOH, PC, Ac, AN, DCE, TMS. Generally, the presence of two SO2
groups adjacent to the reaction centre has two effects. One, the distinct electron-ac-
ceptor effect stabilizes the negative charge of conjugated base by extensive delocaliza-
tion; two, the distinct steric effect (similar to that of nitro group in picric acid) prevents
TABLE IV
Regression parameters and statistical characteristics of the Hammett dependence (pKHA = pK0 – ρσF6
)
for individual solvents
Solvent
pK0(spK
)
ρ(sρ)
s(r)
n
2ρBS/ρa
b
MeOH
EtOH
AN
3.86 (0.03)
4.48 (0.05)
11.47 (0.02)
10.61 (0.02)
17.90 (0.02)
9.60 (0.01)
4.08 (0.09)
3.77 (0.04)
3.66 (0.03)
3.57 (0.02)
13.93 (0.02)
1.41 (0.10)
1.20 (0.15)
4.45 (0.05)
4.08 (0.06)
4.46 (0.06)
4.55 (0.04)
2.10 (0.25)
1.01 (0.12)
0.54 (0.08)
0.22 (0.05)
2.93 (0.07)
0.162 (0.932)
0.216 (0.855)
0.082 (0.998)
0.099 (0.997)
0.130 (0.997)
0.076 (0.999)
0.486 (0.851)
0.180 (0.851)
0.162 (0.797)
0.075 (0.678)
0.110 (0.992)
32
24
33
32
36
33
30
27
27
24
31
2.41
0.94
1.09
Ac
TMS
PC
0.99
c
DMF
DMFd
Py
1.93
4.02
7.62
Pyd
18.72
1.19
DCE
a The ρBS value gives the reaction constant for substituted benzenesulfonamides2–4 b The ρBS value
.
0.52 for methanol2 is uncertain, so the (2ρBS/ρ) ratio was not given. c The dissociation of
benzenesulfonamide was not measured in propylene carbonate. d The calculation without the pKHA
values for the amino derivative.
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

1212
Ludwig, Stverka:
establishing of the homoconjugation equilibrium. Hence, the imides studied by us show
similar behaviour to that of picric acid, which explains the good correlation of the pK0
values measured in the individual solvents with the pKHA values of picric acid. In the
case of the substrates studied we can presume that the dissociation equilibrium will be
strongly affected by the solvent ability of stabilization of the proton. This ability is
highest in Py and DMF, which belong among dipolar aprotic protophilic solvents: the
lowest values of pK0 dissociation constants are observed in these solvents. Amphiprotic
methanol and ethanol can solvate well both the conjugated base and the proton. That is
why they have a similar effect on pK0 as have Py and DMF. The solvents PC, Ac, AN
and TMS belong among dipolar aprotic protophobic solvents which solvate the proton
very poorly, and since they solvate the conjugated base poorly too, the equilibrium is
shifted in favour of the undissociated form: the pKHA values of the acids measured
increase. In the inert DCE solvent the stabilization of both the proton and the conju-
gated base is probably very low.
For the analysis of substituent effects by the method using the latent variables we
___
created a matrix of average values pKHA. The rows and columns of the matrix corre-
sponded to the substituents and solvents, respectively (the solvents in which imides
behave as mild acids: PC, Ac, AN, DCE, TMS). The matrix of 10 × 5 magnitude was
filled to 90% and was treated by the principal component analysis (PCA)36 and conju-
gated deviation analysis (CDA)37. According to statistical criteria we found the first
principal latent variable to be statistically significant in both types of calculations, in-
terpreting 99.78% of variability in both the cases (score vector in standardized form
NH2 – 1.00, OCH3 – 0.69, CH3 – 0.62, NHCOCH3 – 0.61, H – 0.53, F – 0.40, Cl – 0.32,
Br – 0.33, COCH3 – 0.25, NO2 – 0.00). The second principal component described only
0.11% of variability and was statistically insignificant. The first principal component
from the PCA calculations was correlated with the vector of σF6 substituent constants.
The correlation coefficient of this regression has the value of 0.999. The values of
loading vectors (PC – 3.45, AN – 3.61, TMS – 3.61, Ac – 4.50, DCE – 4.50) are very
close to each other and indicate a lower difference of Ac and DCE on one hand and that
of the other solvents on the other hand. This distribution can be caused by e.g. a differ-
ent magnitude of reaction constant. The calculations by the method of projection of
latent structures (PLS) using the pKHA values of the substrates studied by us for the
matrix of dependent variables and analogous values of a series of benzoic acids29 or, as
the case may be, benzenesulfonamides2–4 for the matrix of independent variables
showed the same structure of both matrices and brought no new results.
In conclusion we would like to stress the risk connected with the application of the
Hammett correlations to such series of substituted derivatives which, from the point of
view of their acidity, lie at the border between strong and weak acids. This border is
different in different solvents and can lie at a relative high pH value in some solvents.
As it follows from the present study the situation can be similar to that of the change in
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

4,4’-Disubstituted Bis(benzenesulfon)imides
1213
reaction mechanism connected with a change in substituent which is manifested by a
break in the Hammett dependence.
This research work was sponsored by the Grant Agency of the Czech Republic, Grant No. 203 94 0122.
REFERENCES
1. Hammett L. P.: Physical Organic Chemistry, 2nd ed. McGraw–Hill, New York 1970.
2. Ludwig M., Pytela O., Kalfus K., Vecera M.: Collect. Czech. Chem. Commun. 49, 1182 (1984).
3. Ludwig M., Pytela O., Vecera M.: Collect. Czech. Chem. Commun. 49, 2593 (1984).
4. Ludwig M., Pytela O., Vecera M.: Collect. Czech. Chem. Commun. 51, 1984 (1986).
5. Ludwig M., Pytela O., Javurkova H., Vecera M.: Collect. Czech. Chem. Commun. 52, 2900
(1987).
6. Ludwig M., Parik P., Kulhanek J.: Collect. Czech. Chem. Commun. 60, 841 (1995).
7. Ludwig M., Parik P.: Org. React. 29, 75 (1995).
8. Ludwig M., Exner O., Wold S.: Acta Chem. Scand. 46, 549 (1992).
9. Dauphin G., Kergomard A.: Bull. Soc. Chim. Fr. 1961, 486.
10. Bell P. H., Roblin R. O.: J. Am. Chem. Soc. 64, 2905 (1942).
11. DeChristoper P. J., Adamek J. P., Lyon G. D., Klein S. A., Baumgarten R. J.: J. Org. Chem. 39,
3525 (1974).
12. Birchal J. D., Glidewell C.: J. Chem. Soc., Dalton Trans. 1978, 604.
13. Yukido Y., Yoshimasa Y.: Chem. Pharm. Bull. 12, 25461 (1974).
14. Dynesen E., Lawesson S. O., Schroll G., Bowie J. H., Cooks R. G.: J. Chem. Soc., B 1968, 15.
15. Grieco C., Silipo C., Vittoria A.: G. Ital. Chemometr. 26, 233 (1979).
16. Dykhanov N. N.: Zh. Obshch. Khim. 29, 3602 (1952).
17. Huntress E. H., Carten F. H.: J. Am. Chem. Soc. 62, 511 (1940).
18. Food D. T.: Org. Synth. 2, 295 (1966).
19. Vecera M., Panchartek J.: Laboratorni prirucka organicke chemie. SNTL, Praha 1987.
20. Crosley M. L., Northey E. N., Hultquist M. E.: J. Am. Chem. Soc. 60, 2222 (1938).
21. Dykhanov N. N., Roshenko A. I.: Zh. Org. Khim. 1, 270 (1965).
22. Dorn H., Hilgetag G., Rieche A.: Angew. Chem. 73, 560 (1961).
23. Kolthoff I. M., Guss L. S.: J. Am. Chem. Soc. 60, 2516 (1938).
24. Bykova L. N., Petrov S. I.: Usp. Khim. 39, 1631 (1970).
25. Kolthoff I. M.: Treatise on Analytical Chemistry, 2nd ed., Part I., Vol. 2. Wiley, New York 1979.
26. Kolthoff I. M.: J. Phys. Chem. 35, 2732 (1931).
27. Kolthoff I. M., Chantooni M. K.: J. Am. Chem. Soc. 93, 3843 (1971).
28. Kolthoff I. M., Chantooni M. L., Bhowmik S.: J. Am. Chem. Soc. 88, 5430 (1966).
29. Ludwig M., Baron V., Kalfus K., Pytela O., Vecera M.: Collect. Czech. Chem. Commun. 51,
2135 (1986).
30. Foltyn M., Majer P.: Collect. Czech. Chem. Commun. 43, 95 (1978).
31. Isutsu K., Kolthoff I. M., Fujinag T. J., Hattori M., Chantooni M. K.: Anal. Chem. 49, 503
(1977).
32. Julliard J.: J. Chem. Phys. 67, 691 (1970).
33. Mukherjee L. M., Shultz R. S.: Talanta 19, 707 (1972).
34. Bos M., Dahmen E. A. M. F.: Anal. Chim. Acta 53, 39 (1971).
35. Bos M., Dahmen E. A. M. F.: Anal. Chim. Acta 63, 185 (1973).
36. Wold S., Esbensen K., Geladi P.: Chemom. Intell. Lab. Syst. 2, 37 (1987).
37. Pytela O.: Collect. Czech. Chem. Commun. 55, 42 (1990).
Collect. Czech. Chem. Commun. (Vol. 61) (1996)

1214
Ludwig, Stverka:
38. Geladi P., Kowalski B. R.: Anal. Chim. Acta 185, 1 (1986).
39. Exner O.: Correlation Analysis of Chemical Data. Plenum, New York 1988.
40. Pytela O., Ludwig M., Vecera M.: Collect. Czech. Chem. Commun. 51, 2143 (1986).
41. Pytela O., Kulhanek J., Ludwig M., Riha V.: Collect. Czech. Chem. Commun. 59, 627 (1994).
42. Pytela O., Kulhanek J., Ludwig M.: Collect. Czech. Chem. Commun. 59, 1637 (1994).
Collect. Czech. Chem. Commun. (Vol. 61) (1996)