Effect of charged and ortho substituents on 17O NMR chemical shifts of substituted phenyl tosylates in DMSO

Article abstract of DOI:10.1002/poc.3870

The ¹⁷O nuclear magnetic resonance chemical shifts for 31 ortho-substituted, para-substituted, and meta-substituted phenyl esters of 4-methylbenzenesulfonic acid (phenyl tosylates, 4-CH₃-C₆H₄SO₂OC₆H₄-X) with the neutral and charged substituents in phenoxy part were measured in DMSO-d₆ at 50°C. The chemical shifts, δ(¹⁷O), of the sulfonyl (SO₂) and single-bonded phenoxy (O─Ph) oxygens for para and meta derivatives in the case of neutral substituents showed a good correlation with the Taft σ_I and σ°_R substituent constants. The influence of the ortho substituents on the δ(¹⁷O) chemical shifts of the sulfonyl oxygens and single-bonded phenoxy oxygens was found to be described well with the Charton equation: δ(¹⁷O)_ortho?=?δ(¹⁷O)_H?+?ρ_I σ_I?+?ρ_Rσ°_R?+?δE^B_s when the data treatment was performed separately for electron-donating +R substituents and electron-withdrawing ?R substituents. The substituent constants, σ°_X±, for the charged substituents (4-N⁺(CH₃)₃I^?, 3-N⁺(CH₃)₃I^?, 2-N⁺(CH₃)₃I^?, 4-COO^?N⁺Bu₄, 3-COO^?N⁺Bu₄, 2-COO^?N⁺Bu₄) calculated from the δ(¹⁷O) chemical shifts for phenyl tosylates in DMSO prove that in DMSO, the compounds involving charged substituents are predominantly in ion pair form and not dissociated into free ions. The same conclusion was drawn recently from the carbonyl IR stretching frequencies, ν_CO, and the carbonyl carbon ¹³C chemical shifts for substituted phenyl benzoates, C₆H₅CO₂C₆H₄-X, in DMSO. It was shown that for ion pairs of the charged substituents in DMSO and concentrated aqueous salt solution, the same substituent constant, σ°_X±, could be used.

Full text of DOI:10.1002/poc.3870

Received: 22 November 2017
DOI: 10.1002/poc.3870
Revised: 22 March 2018
Accepted: 1 May 2018
R E S E A R C H A R T I C L E
17
Effect of charged and ortho substituents on O NMR
chemical shifts of substituted phenyl tosylates in DMSO
Vilve Nummert | Mare Piirsalu
| Lauri Toom | Tõnu Kesvatera | Ivo Leito | Ilmar A. Koppel
Institute of Chemistry, University of
Tartu, Tartu, Estonia
Abstract
The 17O nuclear magnetic resonance chemical shifts for 31 ortho‐
substituted, para‐substituted, and meta‐substituted phenyl esters of 4‐
methylbenzenesulfonic acid (phenyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X) with
the neutral and charged substituents in phenoxy part were measured in
DMSO‐d6 at 50°C. The chemical shifts, δ(17O), of the sulfonyl (SO2) and sin-
gle‐bonded phenoxy (O─Ph) oxygens for para and meta derivatives in the case
of neutral substituents showed a good correlation with the Taft σI and σ°R sub-
stituent constants. The influence of the ortho substituents on the δ(17O) chem-
ical shifts of the sulfonyl oxygens and single‐bonded phenoxy oxygens was
found to be described well with the Charton equation: δ(17O)ortho = δ(17O)
+ ρI σI + ρRσ°R + δEB when the data treatment was performed separately
Correspondence
Mare Piirsalu, Institute of Chemistry,
University of Tartu, Ravila 14a, 50411
Tartu, Estonia.
Email: mare.piirsalu@ut.ee
Funding information
EU, Grant/Award Number: TK141;
Ministry of Education and Research of
Estonia, Grant/Award Numbers: IUT20‐
14 and IUT20‐15
H
s
for electron‐donating +R substituents and electron‐withdrawing −R substitu-
ents. The substituent constants, σ°X , for the charged substituents (4‐N
+(CH3)3I−, 3‐N+(CH3)3I−, 2‐N+(CH3)3I−, 4‐COO−N+Bu4, 3‐COO−N+Bu4, 2‐
COO−N+Bu4) calculated from the δ(17O) chemical shifts for phenyl tosylates
in DMSO prove that in DMSO, the compounds involving charged substituents
are predominantly in ion pair form and not dissociated into free ions. The same
conclusion was drawn recently from the carbonyl IR stretching frequencies,
ν
CO, and the carbonyl carbon 13C chemical shifts for substituted phenyl benzo-
ates, C6H5CO2C6H4‐X, in DMSO. It was shown that for ion pairs of the charged
substituents in DMSO and concentrated aqueous salt solution, the same sub-
stituent constant, σ°X , could be used.
KEYWORDS
17O NMR in phenyl tosylates, charged substituents, correlation equations, ortho effect, substituent
effects
1 | INTRODUCTION
processes considered, the points for the charged substitu-
ents in correlations with the Hammett equation usually
The effects of the charged substituents on different pro-
cesses in the nonaqueous solutions have been studied
considerably less as compared with the neutral substitu-
ents. Because the effects of the charged substituents were
found to be dependent on the solvent effects and the
deviate and are often excluded from the correlations. It
has been found[1,2] that in aqueous solutions, the charged
substituents have dualistic behavior. In diluted aqueous
solutions, the charged substituents were considered to
be completely ionized and an electrostatic Coulombic
J Phys Org Chem. 2018;e3870.
wileyonlinelibrary.com/journal/poc
© 2018 John Wiley & Sons, Ltd.
1 of 13
https://doi.org/10.1002/poc.3870

2 of 13
NUMMERT ET AL.
term, δB,[3] should be taken into account besides the polar
influence of substituents described by the Hammett equa-
tion. For processes containing ionic substituents in the
diluted aqueous solutions, an extended Hammett equa-
tion (Equation 1) could be used[1]
present paper, DMSO was used as solvent because the
solubility of phenyl tosylates with charged substituents
in CH3CN is too low for recording the 17O NMR spectra.
The main goal of this work was the re‐evaluation of
the substituent constants for charged substituents, σ°X
,
using the chemical shifts, δ(17O), of substituted phenyl
tosylates, 4‐CH3‐C6H4SO2OC6H4‐X, in DMSO, and com-
parison of the σ°X values obtained in the present
work with those estimated recently[4] from the infrared
stretching frequencies, νCO, and carbonyl carbon 13C
NMR chemical shifts, δCO, for substituted phenyl benzo-
ate series, C6H5CO2C6H4‐X, in DMSO.
Δ ¼ ρσL þ δB
(1)
In Equation 1 for neutral substituents, σL is a “nor-
mal” σ value, and for charged substituents, σL is the σ
value without electrostatic term, ie, the σ value for
dipoles equal to that for ion pairs. In the concentrated
aqueous salt solutions, the molecules involving charged
substituents were found to behave as ion pairs form
where the electrostatic Coulombic term, δB,[3] turns zero,
and the influence of the charged substituents is deter-
mined by the substituent constant σL values for ion pairs
only.[1,2] The σ values for charged substituents estimated
from the processes considered prove whether the charged
substituents in this medium are in ion pairs form or they
are dissociated into the free ions and the σ value contains
an electrostatic component as well.
Another goal of this work was to compare the substit-
uent effects of electroneutral substituents on the chemical
shifts, δ(17O), of ortho‐substituted, para‐substituted, and
meta‐substituted phenyl tosylates in DMSO with the
same substituent effects in CH3CN.[5]
In esters of benzenesulfonic acids, the influence of
ortho, meta, and para substituents on the chemical shifts,
δ(17O), has received considerably less attention as com-
pared to oxygen 17O chemical shifts in the phenyl esters
of benzoic acids.[6–14] In the substituted phenyl tosylates,
4‐CH3‐C6H4SO2OC6H4‐X, the chemical shifts, δ(17O),
dependent on the ortho, meta, and para substituent effects
in the case of electroneutral substituents has been studied
in CH3CN only.[5] To the best of our knowledge, data on
the chemical shifts, δ(17O), for substituted phenyl tosylates
containing charged substituents in DMSO are absent in
the literature. In the previous work,[5] the chemical shifts,
δ(17O), of the sulfonyl (SO2) oxygen and single‐bonded
phenoxy (PhO) oxygen have been determined for 35
ortho‐substituted, para‐substituted, and meta‐substituted
phenyl tosylates (4‐CH3‐C6H4SO2OC6H4‐X) in the case
of electroneutral substituents in CH3CN. The δ(17O)
values of the sulfonyl (SO2) and single‐bonded phenoxy
(O─Ph) oxygens for para and meta derivatives showed a
good correlation with the Taft equation using the σI and
σ°R substituent constants. Introduction of the electron‐
withdrawing −R para and meta substituents produced
deshielding, while electron‐donating +R substituents
caused the shielding effect on the sulfonyl (SO2) and the
single‐bonded phenoxy oxygens.
Recently,[4] we estimated the substituent constants,
σ°X , for charged substituents (X = 3‐N+(CH3)3I−, 4‐N
+(CH3)3I−, 4‐SO3−Na+, 3‐COO−N+Bu4, 4‐COO−N+Bu4,
3‐O−N+Bu4, 4‐O−N+Bu4, 2‐N+(CH3)3I−) for DMSO solu-
tion on the basis of the infrared stretching frequencies,
ν
CO, and the carbonyl carbon nuclear magnetic resonance
(NMR) chemical shifts, δCO, for substituted phenyl
benzoates, C6H5CO2C6H4‐X, in DMSO. The substituent
constants for charged substituents, σ° , for DMSO coin-
cided well with the substituent constants of ion pairs
reported by Hoefnagel and Wepster[1] for aqueous solu-
tions and those estimated from the log k values of the
alkaline hydrolysis of esters in aqueous 4.4 M NaCl solu-
tion at 50°C.[2,4]
In the present work, we measured the 17O NMR
chemical shifts, δ(17O), of the sulfonyl (SO2) oxygen and
single‐bonded phenoxy oxygen (O‐Ph) in 31 ortho‐
substituted, meta‐substituted, and para‐substituted phe-
nyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X,
In the previous works[4,6,7,15,16] in the case of
electroneutral substituents in substituted phenyl and
alkyl benzoates (X‐C6H4CO2C6H5, C6H5CO2C6H4‐X,
C6H5CO2R), thoroughly the inductive, resonance, and
the steric effects of ortho, para, and meta substituents
on the chemical shifts, δ(17O), in CH3CN, the carbonyl
carbon 13C NMR chemical shifts, δCO, in CHCl3 and
DMSO, as well as on the infrared stretching frequen-
cies of the carbonyl group, νCO, in CCl4 and DMSO
have been studied. In esters of benzoic acid, the polar
substituent effect on the carbonyl oxygen chemical
containing the neutral substituents and the charged sub-
stituents (3‐N+(CH3)3I−, 4‐N+(CH3)3I−, 2‐N+(CH3)3I−, 3‐
COO−N+Bu4, 4‐COO−N+Bu4, 2‐COO−N+Bu4) in DMSO.
Previously,[5] the influence of substituent effects on the
17O NMR chemical shifts, δ(17O), of the sulfonyl (SO2) oxy-
gen and single‐bonded phenoxy (PhO) oxygen in
substituted phenyl tosylates has been studied in the case
of neutral (uncharged) substituents in CH3CN. In the

NUMMERT ET AL.
3 of 13
shifts, δ(17O), was found to be 3 to 4 times stronger as
compared with the substituent effects in the carbonyl
carbon 13C NMR chemical shifts, δCO. The chemical
shifts, δ(17O), in substituted phenyl tosylates and in phe-
nyl benzoates as well as the carbonyl carbon 13C NMR
chemical shifts, δCO, in phenyl benzoates have been suc-
cessfully correlated with the corresponding substituent
effects in the infrared stretching frequencies of the sulfo-
linewidth of the measured 17O spectral lines varied
depending on the 17O nucleus position (═O) in the tosyl-
ate. In the DMSO, the S═O oxygen line was 100 to 180 Hz
wide. Linewidth for the water standard, without applica-
tion of the exponential multiplication, was around 25 Hz.
The error of the chemical shifts measured was estimated
on the basis of repetitive measurements and was
0.4 ppm.
ꢀ
ꢁ
When the 17O NMR spectra for substituted phenyl
tosylates were recorded on a Bruker Avance III 700 spec-
trometer in DMSO‐d6, the signals for both oxygens in the
sulfonyl SO2 group and single‐bonded phenoxy O─Ph
oxygen were detected.
nyl group Δvso ¼ ðvso Þx−ðvso
Þ
and carbonyl group,
2
2
2
H
ΔνCO [ΔνCO = (νCO X
)
− (νCO)H], as well as the rates of
the alkaline hydrolysis Δlog k values (Δlog k = log kX − log
kH).[5–7,15–20]
In the beginning of this work, we recorded the 17O
NMR spectra at natural abundance in deuterated DMSO
(DMSO‐d6) on a Bruker Avance II 200 spectrometer
(Brucer BioSpin GmbH, Rheinstetten, Germany)
equipped with 10‐mm BBO probe (17O frequency
27.13 MHz) for 44 ortho‐substituted, meta‐substituted,
and para‐substituted phenyl tosylates, including deriva-
tives with electroneutral and charged substituents.
Approximately 1 molal solution of ortho‐substituted,
meta‐substituted, and para‐substituted phenyl tosylates
in mixture of DMSO (2 mL) and DMSO‐d6 (0.75 mL,
99.9 atom% D) was used. When 17O NMR spectra for
substituted phenyl tosylates were recorded on a Bruker
Avance II 200 spectrometer, only signals for the sulfonyl
(SO2) oxygen were detected; the signals for the single‐
bonded phenoxy OPh oxygen line was not detected in
the case of all tosylates studied (see Supporting Informa-
tion). The δ(17O) values obtained for 44 ortho‐substituted,
meta‐substituted, and para‐substituted phenyl tosylates,
4‐CH3‐C6H4SO2OC6H4‐X, using a Bruker Avance II 200
spectrometer are listed in Table S1; the results of the
δ(17O) data correlations with substituent constants are
shown in Table S2.
2 | EXPERIMENTAL
2.1 | NMR measurements
In the present paper, the 17O NMR spectra for 31 ortho‐
substituted, meta‐substituted, and para‐substituted phe-
nyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X (phenyl esters of
4‐methylbenzenesulfonic acid), including electroneutral
and charged substituents (X = H, 4‐N+(CH3)3I−, 3‐N
+(CH3)3I−, 2‐N+(CH3)3I−, 4‐COO−N+Bu4, 3‐COO−N
+Bu4, 2‐COO−N+Bu4, 4‐NO2, 4‐CN, 4‐COOH, 4‐F, 4‐Br,
4‐OCH3, 4‐OH, 3‐NO2, 3‐COOH, 3‐F, 3‐Br, 3‐NH2, 2‐
NO2, 2‐CN, 2‐CF3, 2‐COOH, 2‐COCH3, 2‐COC6H5, 2‐F,
2‐Cl, 2‐Br, 2‐OCH3, 2‐CH3, 2‐CH(CH3)2), were recorded
at natural abundance in DMSO‐d6 (see Table 1 and
Supporting Information).
The 17O NMR spectra were recorded on a Bruker
Avance III 700 spectrometer equipped with 5‐mm BBO
probe (17O frequency 94.9 MHz). Approximately 1 molal
solution of ortho‐substituted, meta‐substituted, and
para‐substituted phenyl tosylates in 0.75 mL DMSO‐d6
(99.9 atom% D) was used. DMSO‐d6 was purchased from
Sigma‐Aldrich. Temperature of the 17O NMR measure-
ments was 50°C. The 17O NMR chemical shifts were
referenced to H2O (δ = 0.0 ppm) in a sealed capillary
placed inside each NMR tube.
Depending on the sample, about 0.6 to 4.6 million
scans were acquired at the spectral width 21739 Hz, and
its size was 2048 data points. The 11 microsecond
exitation pulse with prescan delay of 64 microseconds
and with the relaxation delay of 0 millisecond was
applied. The acquisition time was 47 milliseconds. Data
were processed with Bruker TopSpin 2.1 software pack-
age (Bruker BioSpin GmbH, Rheinstetten, Germany).
During of the data processing, an exponential multiplica-
tion of the FID with LB factor value 10 and the zero fill-
ing up to 32768 data points were applied. For correction
of the spectral baseline rolling, cubic spline baseline cor-
rection was applied also during the data processing. The
1
For compound characterization, the H, 13C, and 19F
NMR spectra were recorded at 700.1 or 200 MHz (1H),
176.0 or 50.3 MHz (13C), and 658.7 MHz (19F), using
Bruker Avance II 200 or Bruker Avance III 700 spectrom-
eters. All these NMR measurements were carried out at
25°C. Chemical shifts (1H and 13C) were indirectly refer-
enced to TMS via the residual solvent signal (DMSO‐d6,
2.50 and 349.52 ppm, respectively). 19F chemical shifts
were referenced to external CFCl3 at 0.0 ppm. The NMR
signals were assigned from HSQC, HMBC, NOESY, and
TOCSY spectra. The reported 13C NMR data correspond
1
to the H‐decoupled 13C NMR data.
2.2 | Synthesis of compounds
The preparation procedure and characteristics for
the most ortho‐substituted, meta‐substituted, and para‐

4 of 13
NUMMERT ET AL.
TABLE 1 The 17O NMR chemical shifts, δ(17O), of the sulfonyl oxygen and single‐bonded phenoxy oxygen for ortho‐substituted, para‐
substituted, and meta‐substituted phenyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X (X‐substituted phenyl esters of 4‐methyl‐benzenesulfonic acid),
in DMSO at 50°C
δ(17O)a
δ(13C)
b
Substituent
S═O
OC6H4X
In substituent
ipso‐CO
4‐CH3‐C6H4SO2OC6H4‐X
H
165.8
167.8
167.7
167.7
167.2
165.9
165.0
164.3
164.1
164.9
167.0
166.4
166.6
165.8
167.1
165.9
165.5
167.5
174.6e
168.3
165.9
167.4
167.8
169.9
166.6
167.0
166.2
170.4
165.0
165.9
166.3
191.8
197.1
197.6
–
149.1 (150.7)
153.2 (155.6)
152.0
4‐NO2
576.2 (NO2)
4‐CN
4‐N+(CH3)3I−
4‐COOH
4‐Br
149.0 (151.2)
152.2 (154.1)
148.2 (149.4)c
145.1
–
254.0 (COOH)
191.4
188.4
187.1
–
4‐F
4‐OCH3
50.7 (OCH3)
88.0 (OH)
282.8 (COO−N+Bu4)
142.4 (144)
141.3 (142.8)
149.0 (150.6)
148.9
4‐OH
4‐COO−N+Bu4
3‐NO2
182.8
193.8
192.4
193.2
190.8
–
577.7 (NO2)
3‐F
149.6
3‐Br
149.4
3‐NH2
150.4
3‐N+(CH3)3I−
3‐COOH
3‐COO−N+Bu4
2‐NO2
149.2 (151.1)
149.0 (150.6)
148.6 (149.9)
140.2
–
265.8 (COOH)d
283.1 (COO−N+Bu4)
608.1 (NO2)
–
184.4
–
2‐N+(CH3)3I−
142.1
2‐CN
194.3
177.2
189.1
194.1
185.6
188.5
–
149.4
2‐F
136.1
2‐Cl
144.8
2‐Br
146.1
2‐CF3
146.3
2‐COCH3
2‐COC6H5
2‐COOH
2‐COO−N+Bu4
2‐OCH3
2‐CH3
580.4 (COCH3)
563.4 (COC6H5)
267.7 (COOH)d
305.8 (COO−N+Bu4)
41.8 (OCH3)
145.9
145.8
–
147.0
182.7
176.9
187.3
184.5
145.6
137.6
147.7
2‐CH (CH3)2
146.3
aApproximately 1 molal solution of ortho‐substituted, meta‐substituted, and para‐substituted phenyl tosylates in DMSO‐d6 (0.75 mL) was used. 17O NMR spectra
were recorded on a Bruker Avance III 700 spectrometer.
bThe δ(13C) values for ipso‐CO carbon in substituted phenyl benzoates, C6H5CO2C6H4‐X, in DMSO‐d6 are shown in brackets.[6] (See Supporting Information).
cThe δ(13C) value of ipso‐CO for 4‐cholophenyl benzoate.
dThe δ(17O) value in a mixture of DMSO and DMSO‐d6.
ePartly decomposes.

NUMMERT ET AL.
5 of 13
ꢂ
ꢃ
ꢂ
ꢃ
substituted phenyl esters of para‐toluenesulfonic acid,
4‐CH3‐C6H4SO2OC6H4‐X, used in the present work have
been described previously.[5,21–27]
δ
17O ¼ δ 17O þ ðρIÞσI þ ðρRÞσ°R
(2)
H
ꢂ
ꢃ
ꢂ
ꢃ
δ
17O ¼ δ 17O H þ ðρÞσ°
(3)
The phenyl esters of para‐toluenesulfonic acid, 4‐
CH3‐C6H4SO2OC6H4‐X (X = 4‐Br, 3‐Br, 3‐F, 2,6‐(CH3)2)
were prepared by the addition of p‐toluenesulfonyl
chloride to the corresponding phenol in pyridine with
stirring at 0°C (the Einhorn method[28]). 4‐Bromophenyl
4‐toluenesulfonate: m.p. 95°C to 96°C (Hazlet[29] m.p.
93°C‐95°C). 3‐Bromophenyl 4‐toluenesulfonate: m.p.
61°C to 62°C (Hazlet[29] m.p. 55°C‐57°C). 3‐Fluorophenyl
4‐toluenesulfonate: m.p. 55°C to 56°C (Stang and
Anderson[30] m.p. 47.5°C‐48.5°C). 2,6‐Dimethylphenyl 4‐
toluenesulfonate: m.p. 53°C to 54°C (Rottendorf and
Sternhell[31] m.p. 53.5°C, Table S1).
Tetrabutylammonium salts of the [(4‐methylphenyl)
sulfonyl]oxybenzoic acids (4‐CH3‐C6H4SO2OC6H4‐X,
X = 4‐COO−N+Bu4, 3‐COO−N+Bu4, 2‐COO−N+Bu4)
were prepared similar to the synthesis of the
tetrabutylammonium (benzoyloxy)benzoates.[4] The equi-
molar amount of tetrabutylammonium hydroxide in
methanol was added dropwise to the [(4‐methylphenyl)
sulfonyl]oxybenzoic acid in ethanol with intensive stir-
ring similar to preparation of the tetrabutylammonium
(benzoyloxy)benzoates.[4] The solvent evaporated to
dryness in vacuo. The tetrabutylammonium [(4‐
methylphenyl)sulfonyl]oxybenzoates were obtained as
colorless very viscous liquids.
In the case of ortho derivatives, the following Charton
equations were used[36]
:
ꢂ
ꢂ
ꢃ
ꢃ
ꢂ
ꢃ
δ
δ
17O
¼ δ 17O þ ðρIÞorthoσI þ ðρRÞorthoσ°R
ortho
ortho
H
þ δorthoEB
(4)
(5)
s
ꢂ
ꢃ
17O
¼ δ 17O þ ðρIÞorthoσI þ ðρRÞorthoσ°R
þ δorthoυ H
Two steric scales were used for ortho substituents: the
EB constants[37–39] and the Charton scale of υ.[40,41] For
s
the data processing, a multiple‐parameter linear least‐
squares procedure[42]was used.
The results of the statistical data treatment of the sul-
fonyl SO2 oxygen and the single‐bonded bridging oxygen
δ(17O) values for substituted phenyl tosylates, 4‐CH3‐
C6H4SO2OC6H4‐X, in DMSO (in Table 1) with the Taft
and Charton Equations 2 to 5 are shown in Table 2.
The substituent constants used in correlations are shown
in Table 3.
4 | DISCUSSION
Purity of synthesized substituted phenyl 4‐
4.1 | Influence of para and meta
substituents in 4‐CH3‐C6H4SO2OC6H4‐X
toluenesulfonates was confirmed by 1H and 13C NMR spec-
1
troscopy in deuterated DMSO at 25°C. H and 13C NMR
spectra for all studied phenyl 4‐toluenesulfonates, 4‐CH3‐
C6H4SO2OC6H4‐X, are shown in Supporting Information.
In the case of tetrabutylammonium [(4‐methylphenyl)sul-
fonyl]oxybenzoates (4‐CH3‐C6H4SO2OC6H4‐X, X = 2‐
COO−N+Bu4, X = 3‐COO−N+Bu4, X = 4‐COO−N+Bu4)
The total NMR chemical shift σtot is considered to be
dependent upon a local diamagnetic, σd, and a paramag-
netic, σp, components.[8,9,43] The paramagnetic term σp
as mainly responsible for the structure variability of 17O
NMR chemical shifts is usually described by Karplus‐
1
Pople Equation 6[8,43]
:
in H NMR spectra, the band for H in COOH group (δ:
13.19, 13.33, and 13.22) was not detected (Supporting
Information).
ꢂ
ꢃꢂ
ꢃ
σp ¼ –const: ΔE–1 r–3 ð∑QÞ
(6)
In Equation 6, ΔE is the mean electronic excitation
energy, r−3 is the average inverse cube of the radius for
the 2p orbitals, and ∑Q describes the charge density on
oxygen and the extent of multiple bond order.
3 | DATA PROCESSING AND
RESULTS
The values of the chemical shifts, δ(17O), of the sulfonyl
oxygen (SO2) and single‐bonded phenoxy oxygen (OPh)
in para‐substituted and meta‐substituted phenyl tosylates,
4‐CH3‐C6H4SO2OC6H4‐X, listed in Table 1 were correlated
according to the Taft Equations 2 and 3 using the inductive
σI[32,33] and the resonance σ°R (σ°R = σ°p − σI)[32,34] and
Taft σ°[32,35] substituent constants. The data treatment
with Equations 2 and 3 was performed separately for para
and meta derivatives.
In the present work, the NMR δ(17O) shift values for
the sulfonyl S═O oxygen in substituted phenyl tosylates,
4‐CH3‐C6H4SO2OC6H4‐X, in DMSO (from 163.2 to
172 ppm, Table 1) were detected at much lower chemical
shift values than the 17O signals of the carbonyl C═O
group in substituted phenyl benzoates, C6H5CO2C6H4‐X,
in CH3CN (from 344 to 350 ppm).[7] As it follows from
the Karplus‐Pople Equation 6,[8,43] the δ(17O) values at
lower ppm values demonstrate the higher charge density

6 of 13
NUMMERT ET AL.
TABLE 2 Correlation of the 17O NMR chemical shifts, δ(17O), of the sulfonyl oxygen and single‐bonded phenoxy oxygen for ortho‐
substituted, para‐substituted, and meta‐substituted phenyl tosylates, 4‐CH3‐C6H4SO3C6H4‐X, in DMSO (Table 1) with Equations (2) to (5)
Ra
sb
n/n0
c
ρΙ or ρ °
ρ °R
δ
Scales
δ(17Ο)H
ortho‐Substituted Phenyl Tosylates
+R substituents
S═17O
σI, σoR, EB
σI, σoR, υ
165.8 0.2
165.8 0.2
4.94 0.34
4.56 0.28
−R substituents
10.8 0.4
8.59 0.63
7.53 0.50
−4.43 0.60
0.987
0.991
0.153
0.132
7/7d
7/7d
s
1.85 0.21
σI, σoR, EB
σI, σoR, υ
165.9 0.4
165.9 0.8
165.9 0.4
165.9 2.0
17OC6H4X
190.6 6.8
192.3 4.3
190.5 6.0
−38.2 1.7
−37.6 11.2
−38.0 2.4
−37.5 11.1
−5.80 0.3
−5.09 1.7
2.69 0.18
2.61 0.29
0.997
0.857
0.994
0.858
0.109
0.717
0.154
0.717
6/7e
7/7f
6/7e
7/7f
s
11.0 2.5
11.0 0.5
11.1 2.5
+R substituents
18.2 9.4
σI, σoR, EB
σI, σoR, υ
σI, σoR, EB
σ °para
57.7 17.5
53.6 14.4
55.7 16.0
−9.33 16.6
–
0.777
0.820
0.781
4.24
3.86
4.20
7/7d
7/7d
7/7d
s
16.7 8.2
17.5 9.0
4.14 6.8
−R substituents
−12.4 1.8
191.6 0.8
S = 17O
–
–
0.970
0.804
4/4g
s
para‐substituted phenyl tosylates
165.2 0.2
165.8 0.2
17OC6H4X
189.4 1.1
193.2 1.1
S═17O
3.56 0.47
2.05 0.40
–
–
–
0.943
0.988
0.456
0.219
8/8h
8/8h
σI, σo
4.87 0.34
R
σ °para
10.34 30
2.52 2.12
–
–
–
0.890
0.948
1.98
1.47
6/6i
6/6i
σI, σo
16.4 2.6
R
meta‐substituted phenyl tosylates
σ °meta
165.9 0.1
165.8 0.1
165.6 0.1
17OC6H4X
191.5 0.2
191.7 0.5
1.37 0.23
1.60 0.29
2.29 0.09
–
–
–
–
0.933
0.937
0.998
0.158
0.154
0.033
6/6j
6/6j
4/6k
σI, σo
0.34 0.26
R
σI
–
σ °meta
3.46 0.52
3.05 0.88
–
–
0.956
0.924
0.353
0.460
5/5l
5/5l
σI, σo
2.20 0.90
R
aR—correlation coefficient.
bs—standard deviation.
cn0 reflects the total number of data involved in the correlation; n—the number of points remaining after exclusion of significantly deviating points.
dThe δ(17O) values for 2‐F, 2‐Cl, 2‐Br, 2‐OCH3, 2‐CH3, 2‐CH (CH3)2, and unsubstituted derivative were included.
eThe δ(17O) values for 2‐NO2, 2‐CN, 2‐CF3, 2‐COCH3, 2‐COPh, 2‐COOH, and unsubstituted derivative were included. 2‐COOH derivative was excluded during
the data processing.
fThe δ(17O) values for 2‐NO2, 2‐CN, 2‐CF3, 2‐COCH3, 2‐COPh, 2‐COOH, and unsubstituted derivative were included.
gThe δ(17O) values for 2‐NO2, 2‐CF3, 2‐COCH3, and unsubstituted derivative were included.
hThe δ(17O) values for 4‐NO2, 4‐CN, 4‐COOH, 4‐F, 4‐Br, 4‐OCH3, 4‐OH, and unsubstituted derivative were included.
iThe δ(17O) values for 4‐NO2, 4‐CN, 4‐F, 4‐Br, 4‐OCH3, and unsubstituted derivative were included.
jThe δ(17O) values for 3‐NO2, 3‐Br, 3‐COOH, 3‐F, 3‐NH2, and unsubstituted derivative were included.
kThe δ(17O) values for 3‐NO2, 3‐Br, 3‐COOH, and 3‐NH2 derivatives were included. 3‐F and unsubstituted derivatives were excluded.
lThe δ(17O) values for 3‐NO2, 3‐Br, 3‐F, 3‐NH2, and unsubstituted derivative were included.

NUMMERT ET AL.
7 of 13
TABLE 3 Substituent constants used in the correlations
DMSO were found to be linearly correlated with the
corresponding δ(17O) values in CH3CN[5] as follows
(Equation 7):
b
Bc
X
σIa
σ°R
Es
υd
σ°e
H
0
0
0
0
–
–
–
0
ꢂ
ꢃ
X
4‐NO2
4‐CN
0.63
0.19
0.11
–
0.81
17
δ
O
¼ ð2:5 5:7Þ
þ ð0:993 0:033Þδ
R ¼ 0:997; s ¼ 0:126; n=n0 ¼ 6=6
S═O
DMSO
ꢂ
ꢃ
0.58
0.45
0.52
0.30
0.25
0.25
0.63
0.52
0.45
0.10
0.30
0.63
0.58
0.52
0.47
0.45
0.41
0.28
0.28
0.30
0.25
−0.05
−0.05
–
0.72
0.30
X
17
(7)
O
S═O
CH3CN
4‐Br
−0.19
−0.35
0.14
–
4‐F
–
0.21
4‐COOH
4‐OCH3
4‐OH
–
0.46
When in Equation 7, the intercept was considered as
insignificant (=0), then the slope of the dependence (7)
was 1.007 0.003, s = 0.116, n/n0 = 6/6.
−0.41
−0.41
0.19
–
−0.15
−0.16
0.71
–
We found in substituted phenyl tosylates the para and
meta electron‐withdrawing substituents to cause high‐fre-
quency shift for the 17O signal of the sulfonyl S═O oxygen
showing deshielding effect of the O atom (Table 1). The
electron‐donating substituents have an opposite effect
resulting in low‐frequency shift and shielding of the sul-
fonyl S═O oxygen atom. The similar influence of the sub-
stituents on the δ(17O) values of the carbonyl oxygen was
found in substituted phenyl benzoates.[6,7] The increased
double bond character of the carbonyl group produced
by an electron‐withdrawing groups in substituted phenyl
esters of benzoic acid was proved by the carbonyl carbon
13C NMR chemical low‐frequency, δCO, shifts and the
3‐NO2
3‐F
–
−0.35
−0.19
−0.48
0.14
–
–
0.34
3‐Br
–
–
0.39
3‐NH2
3‐COOH
2‐NO2
2‐CN
–
–
−0.14
0.36
–
–
0.19
−0.374
−0.03
−0.155
−0.243
−0.27
−0.593
−0.665
−0.665
−0.665
−0.308
−0.264
−0.341
0.76
0
0.11
–
–
–
–
–
–
–
–
–
–
–
2‐F
−0.35
−0.20
−0.19
0.10
0.27
0.55
0.65
1.24
1.39
1.39
1.39
0.56
0.52
0.76
2‐Cl
2‐Br
2‐CF3
2‐COCH3
2‐COC6H5
2‐COOH
2‐OCH3
2‐CH3
2‐CH (CH3)2
increased infrared frequencies of the carbonyl group,
0.16
[4,15,44]
νCO
.
0.15
In the case of electroneutral substituents, the sulfonyl
oxygen chemical shifts, δ(17O), for the para‐substituted
phenyl tosylates (Table 1) correlated well with the dual
substituent parameter method using the Taft σI and σ°R
(Equation 8, Table 2)
0.14
−0.41
−0.10
−0.08
aThe σI values from Exner[32] and Taft and Lewis[33] were used.
bThe σ°R values from Exner[32] and Taft et al[34] were used.
cThe steric constant EsB values from Nummert et al,[5] Nummert and
Piirsalu,[37] Nummert et al,[38] and Nummert et al[39] were used.
dThe Charton steric constant υ from Nummert et al,[5] Nummert et al,[38]
Aslam et al,[40] and Charton[41] was used.
eThe σ° values in Exner[32] and Palm[35] were used.
ꢂ
ꢃ
δ17OS═O
¼ ð165:8 0:2Þ
þ ð2:05 0:40ÞσI þ ð4:87 0:34Þσ°R
R ¼ 0:988; s ¼ 0:219; n=n0 ¼ 8=8
para
(8)
and the single parameter treatment using the Taft σ° sub-
stituent constants (R = 0.943, Table 2). In the case of para
derivatives, the influence of the resonance term appeared
to be stronger in comparison with that of the inductive
term (Equation 8). In the case of meta derivatives, the
effect of resonance term was found to be insignificant
(Equation 9, Table 2):
at the sulfonyl SO2 oxygens and the increased ionic bond
character of the sulfonyl SO2 group (S+─O−) as compared
to that of the carbonyl C═O oxygen in phenyl
benzoates.[44]
The values of the chemical shifts, δ(17O), of the sulfo-
nyl oxygen for para‐substituted and meta‐substituted
phenyl tosylates in DMSO (Table 1) in the case of neutral
substituents were ca. 0.5 to 1.5 ppm higher as compared
to the corresponding δ(17O) values measured earlier in
CH3CN.[5] In the case of electroneutral para substituents
(H, NO2, CN, F, OCH3, OH), the δ(17O) values of the sul-
fonyl S═O oxygen for substituted phenyl tosylates in
ꢂ
ꢃ
δ17OS═O
¼ ð165:8 0:1Þ
meta
(9)
þ ð1:60 0:29ÞσI þ ð0:34 0:26Þσ°R
R ¼ 0:937; s ¼ 0:154; n=n0 ¼ 6=6
In phenyl tosylates, the susceptibility of the SO2
chemical shifts, δ(17O), in DMSO toward the para polar
substituent effect from the phenyl side (ρI = 2.05,

8 of 13
NUMMERT ET AL.
ρR = 4.87, Equation 8, Table 2) was approximately the
same as was found earlier for the SO2 chemical shifts,
δ(17O), in phenyl tosylates (ρI = 1.56, ρR = 4.77[5]) and
for the carbonyl CO oxygen chemical shifts, δ(17O) in
substituted phenyl benzoates, C6H5CO2C6H4‐X, (ρI = 3.25,
ρR = 4.34[7]) in CH3CN.
charged substituents, σ°X
Equation 10:
were calculated with
h
i
ꢂ
ꢃ
ꢂ
ꢃ
17
17
ðσ°X Þp ¼ δ
O
–δ
O
=ρp
(10)
S═O
S═O
X
H
Using the chemical shifts, δ(17O), of the sulfonyl oxy-
gen for para‐substituted and meta‐substituted phenyl
tosylates, 4‐CH3‐C6H4SO2OC6H4‐X, in DMSO (Table 1)
and the results of the statistical data treatment in
Table 2 for the charged substituents, the substituent con-
stants, σX , which could be applicable to correlate the
influence of the charged substituent effects simulta-
neously with those of electroneutral substituents,
were calculated. To compare the values of the substituent
constants of charged substituents established in the
present work with those σ°X values obtained previously
by different measurements given in Table 4, for para
For meta charged substituents with Equation 10, the
inductive substituent constants (σI)X were calculated
using the corresponding values of δ(17OS═O X
)
in
Table 1 and (ρI)m in Table 2. It could be mentioned that
σI = σ° in case σ°R = 0.
The substituent constants (σ°X )p and (σI)X for para
and meta charged substituents estimated with Equa-
tion 10 using the corresponding sulfonyl oxygen (SO2)
chemical shifts, δ(17OS═O), in DMSO (Table 1) and the
values of ρp and (ρI)m for neutral para‐substituted and
meta‐substituted phenyl tosylates in DMSO (Table 2) are
listed in Table 4.
TABLE 4 The values of substituent constants, σ°X , for charged substituents estimated from 17O NMR chemical shifts, δ(17O), in DMSO for
substituted phenyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X, the infrared stretching frequencies, νCO
13C NMR carbonyl chemical shifts, δCO, in
DMSO, and rates of alkaline hydrolysis for phenyl benzoates, C6H5CO2C6H4‐X, in 4.4 M aqueous NaCl solution at 50°C
,
σ°
σ° with Electrostatic
σ°p, (σI)
δ(17O)a
σ°
νCO
σ°
δCO
4.4 M
NaCld
Recommended
σ° for
Component
(in Aq. Solutions)g
Ion Pairsf
b
c
e
Substituent X
(σ°X
)
av
4‐N+(CH3)3I−
3‐N+(CH3)3I−
2‐N+(CH3)3I−
4‐CO2−N+Bu4
3‐CO2−N+Bu4
2‐CO2−N+Bu4
4‐O−N+Bu4
0.70 (0.56)
0.65 (0.56)
0.77
0.65
0.46
0.78
0.08
0.07
0.44
0.44
0.53
0.46
0.58
0.53
0.59
–
1.025, 0.88, 0.96
1.094, 1.04, 0.99
1.15
0.50
0.77
0.69
0.83
−0.01 (0.18)
−0.04 (0.03)
0.14
0.06
0.09
0.26
0.24
−0.078, −0.122
−0.194, −0.195
−0.02
0.04
−0.01
0.07h
−0.48i
−0.46i
–
–
−0.44i
−0.44i
3‐O−N+Bu4
aThe values of the σ° substituent constants for charged substituent X in DMSO were calculated with Equation (10) using the 17O NMR chemical shifts, δ(17O),
for substituted phenyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X, in Table 1. For para derivatives ρp = 3.56, δ(17Ο)H = 165.2, for meta derivatives ρm = 2.29, δ(17Ο)
H = 165.6, and for ortho derivatives ρortho = 10.8, δ(17Ο)H = 165.9 (Table 2, for 2‐N+(CH3)3I− substituent the resonance and steric factors appeared to be insig-
nificant[4]). In brackets are shown the σ°X values calculated with the chemical shifts, δ(17O), in Table S1 and using δ(17Ο)H = 165.4, and ρp,m = 3.79 (Table S2).
bThe σ° substituent constants calculated using infrared carbonyl stretching frequencies, νCO, in DMSO for substituted phenyl benzoates, C6H5CO2C6H4‐X.[4]
cThe σ° values calculated from the carbonyl carbon 13C NMR chemical shifts, δCO, for substituted phenyl benzoates, C6H5CO2C6H4‐X, in DMSO.[4]
dThe σ° values calculated from the rate constants for the alkaline hydrolysis of substituted phenyl benzoates, C6H5CO2C6H4‐X, in aqueous 4.4 M NaCl solution
at 50°C.[4]
eThe averaged values of substituent constants, (σ°X av, for ion pairs of charged substituents calculated on the bases of the following experimental data: the 17
) O
NMR chemical shifts, δ(17O), for substituted phenyl tosylates in DMSO (Table 1), the carbonyl stretching frequencies, νCO, in DMSO, the carbonyl carbon 13C
NMR chemical shifts, δCO, in DMSO and log k values for the alkaline hydrolysis in aqueous 4.4 M NaCl solution for substituted phenyl benzoates.[4]
fThe σ° values for ion pairs calculated by Wepster as free from the electrostatic interaction term.[1]
gThe σ° constants containing the electrostatic component for reactions between substrate containing charged substituent and the ionic reagent. To calculate the
σ° values, the rate constants or pK values extrapolated to pure water were used.[32,48]
hUsing the δ(17O)S═O = 170.4 ppm for ortho COO−N+Bu4‐phenyl tosylate in DMSO (Table 1) and σI = 0.1 and σ°R = 0 with Equation (12), the steric constant
EBs = −0.60 for ortho COO−N+Bu4 substituent was estimated. The value of σ° = 0.07 for ortho COO−Na+ substituent could be calculated from the kinetic data of
the alkaline hydrolysis of 2‐COO−Na+‐phenyl benzoate in 4.4 M aqueous NaCl at 50°C (log k = −0.55,[48] log k0 = 0.070,[49,50] δortho = 1.2[50]) using the steric
constant EB = −0.60 for ortho COO−Na+ substituent.
s
iData from Nummert et al.[4]

NUMMERT ET AL.
9 of 13
The σ°X values for charged substituents estimated
with Equation 10 in Table 4 (σ°X = 0.70 for 4‐N
+(CH3)3I−, −0.01 for 4‐COO−N+Bu4 and (σI)X = 0.65
for 3‐N+(CH3)3I−, −0.04 for 3‐COO−N+Bu4) are in good
agreement with the constants, σ°X , obtained recently
from the IR carbonyl stretching frequencies, νCO, and
the carbonyl carbon chemical shifts, δCO, for substituted
phenyl benzoates series, C6H5CO2C6H4‐X, in DMSO,[4]
as well as with the constants, σ°X , for ion pairs proposed
by Hoefnagel and Wepster[1] and those calculated from
the kinetic data of the alkaline hydrolysis of substituted
phenyl benzoates in concentrated aqueous NaCl solution
(Table 4).[4] On the basis of the constants, σ°X , for
charged substituents estimated from the δ(17O), values
of phenyl tosylates in DMSO (Table 3), we can conclude
that in substituted phenyl tosylates, the charged substitu-
ent X = 3‐N+(CH3)3I−, 4‐N+(CH3)3I−, 3‐COO−N+Bu4,
and 4‐COO−N+Bu4, in DMSO are not completely ionized
and are in ion pair form. Consequently, when correlating
data set containing simultaneously derivatives with
charged substituents and electroneutral substituents, for
charged substituents in DMSO, the substituent constants
for ion pairs could be used. The substituent constants
for ion pairs do not include the electrostatic component
which was proved essential in the case of pure aqueous
solutions (Equation 1).[1] A good relationship between
σ°X values for charged substituents estimated from the
δ(17O) values of phenyl tosylates in DMSO, the IR car-
bonyl stretching frequencies, νCO, and the carbonyl car-
FIGURE 1 Relationship between the σ°X values for ion pairs of
the charged substituents estimated from the various processes in
DMSO and the σ°X values of ion pairs of the charged substituents
calculated on the basis of the alkaline hydrolysis of substituted
phenyl benzoates, C6H5CO2C6H4‐X, in 4.4 M aqueous NaCl
solution[4]
σ°X ¼ ð0:102 0:292Þ þ ð0:973 0:061Þσ°X ð4:4 M aq:NaClÞ
R ¼ 0:971; s ¼ 0:097; n=n0 ¼ 17=17:
○ The σ°X and (σI)X values calculated from the 17O NMR
chemical shifts of substituted phenyl tosylates, 4‐CH3‐
C6H4SO2OC6H4‐X, in DMSO.
♦ The σ°X values calculated from the infrared stretching
frequencies, νCO, for substituted phenyl benzoates, C6H5CO2C6H4‐
X, in DMSO.
▲ The σ°X values calculated from the carbonyl carbon 13C NMR
chemical shifts for substituted phenyl benzoates, C6H5CO2C6H4‐X,
in DMSO.
bon chemical shifts, δCO
, for substituted phenyl
benzoates in DMSO, and the σ°X values for charged sub-
stituents calculated from the kinetic data of the alkaline
hydrolysis of substituted phenyl benzoates in concen-
trated 4.4 M aqueous NaCl solution shows that for ion
pairs of charged substituents in DMSO and aqueous con-
centrated salt solution, the same substituent constants,
σ°X , could be used (Figure 1).
In substituted phenyl tosylates, para and meta substit-
uents in phenyl moiety exert on the electron density at
the phenyl ipso carbon (CO‐1, Table 1) nearly in the same
way as in substituted phenyl benzoates. For derivatives
with para and meta electroneutral substituents and
charged substituents (3‐N+(CH3)3I−, 4‐N+(CH3)3I−, 3‐
COO−N+Bu4, 4‐COO−N+Bu4) in DMSO, we obtained
Equation 11:
resonance effect of substituents. For para‐substituted
phenyl tosylates, 4‐CH3‐C6H4SO2OC6H4‐X (X = H, 4‐
NO2, 4‐CN, 4‐COOH, 4‐F, 4‐Br, 4‐OH, 4‐OCH3), in
DMSO, we found (Equation 12):
ꢂ
ꢃ
13
δ
C
¼ ð148:7 0:8Þ þ ð3:01 1:73ÞσI
þ ð16:7 1:4Þσ°R
O−1
Tos
(12)
R ¼ 0:979; s ¼ 0:939; n=n0 ¼ 8=8
During the data treatment, the inductive component
in Equation 12 was excluded as insignificant, and we
obtained δ(13CO‐1 Tos
= (149.9 0.4) + (17.4 1.6)σ°R,
)
R = 0.971, s = 1.09, n/n0 = 8/8. In substituted phenyl
tosylates and phenyl benzoates, the δ(13CO‐1) value for
4‐N+(CH3)3I− derivative (149.1, 151.2) and 4‐COO−N
+Bu4 derivative (149.0, 150.6) was found to be practically
equal to that for unsubstituted derivative (149.1, 150.7)
(Table 1). Therefore, for N+(CH3)3I− and COO−N+Bu4
substituents, σ°R = 0 was used.
ꢂ
ꢃ
13
δ
C
¼ ð6:66 3:92Þ
þ ð0:944 0:026Þδ
R ¼ 0:996; s ¼ 0:310; n=n0 ¼ 11=11
O−1
Tos
ꢂ
ꢃ
13
C
(11)
O−1
Benz
The chemical shift δ(13CO‐1 Tos
)
values for the phenyl
The sulfonyl oxygen chemical shifts, δ(17O)S═O, in
para‐substituted and meta‐substituted phenyl tosylates
ipso carbon were found to be dependent mainly on the

10 of 13
NUMMERT ET AL.
(Table 1) were found to be correlated well with the corre-
sponding infrared stretching frequencies of the carbonyl
group, νCO, in substituted phenyl benzoates,[4] including
data for derivatives with both the charged and
electroneutral substituents in DMSO (Equation 13):
4.1.1 | Influence of para and meta substit-
uents on the δ(17O) values of single‐bonded
phenoxy oxygen
The influence of the inductive and resonance effects of
para and meta substituents from the phenyl side on the
single‐bonded phenoxy oxygen δ(17O) values in
substituted phenyl tosylates in DMSO similar to that in
CH3CN was ca. 3.0 times stronger than on the δ(17O)
values of the sulfonyl oxygen (Table 2):
ꢂ
ꢃ
17
δ
O
S═O
¼ ð–294:0 72:7Þ
Tos
þ ð0:265 0:042ÞðνCOÞBenz
(13)
þ ð1:45 0:76Þσ°R
R ¼ 0:970; s ¼ 0:335; n=n0 ¼ 13=13
ꢂ
ꢃ
17
δ
O
¼ –ð304:0 32:9Þ
þ ð2:99 0:20Þδ
R ¼ 0:981; s ¼ 0:65; n=n0 ¼ 10=10:
OPh
m;p
ꢂ
ꢃ
17
When the data treatment with Equation 13 δ(17OS═O
Tos = a + b(νCO Benz + cσ°R was carried out only for para
)
O
S═O
(14)
m;p
)
derivatives (X = H, 4‐NO2, 4‐OH, 4‐OCH3, 4‐Br, 4‐NH2,
4‐N+(CH3)3I−, 4‐COO−N+Bu4), the excellent correlation
was obtained: a = −255.0
c = 2.23
49.4, b = 0.243
0.029,
4.2 | Influence of ortho substituents
0.51, s = 0.155, R = 0.996, n/n0 = 7/8 (the
4‐COO−N+Bu4 derivative was excluded). In Equation 13,
the parameter b shows the relation of the inductive effects
For ortho‐substituted phenyl tosylates in the case of neu-
tral substituents, a good correlation between the sulfonyl
oxygen chemical shifts, δ(17O), in DMSO (Table 1) and in
CH3CN[5] was found (Equation 15):
in 2 processes compared: b = ρI(δ(17OS═O))Tos/ρI(νCO Benz
)
and c = ρR(δ(17OS═O))Tos − b ρR(νCO Benz
)
.
[15]
In substituted phenyl tosylates,
4‐CH3‐
ꢂ
ꢃ
17
X
C6H4SO2OC6H4‐X, the oxygen chemical shifts, δ(17O), in
substituents 3‐CO2−N+Bu4, 4‐CO2−N+Bu4, and 2‐CO2−N
+Bu4 were observed by ca. 20 to 30 ppm at the higher
frequencies (the δ(17O) values at 283.1, 282.8, and
305.4 ppm, respectively, Table 1) as compared to that in
the corresponding carboxylic acid groups (for X = 3‐
COOH δ(17O) = 265.8 ppm, 4‐COOH δ(17O) = 254.0,
X = 2‐COOH δ(17O) = 267.7 ppm, Table 1). Similarly,
the 13C NMR chemical shifts, δ(13C)COO, of the carboxyl-
ate group in tetrabutylammonium carboxylate substitu-
ents were higher as compared to that in the carboxylic
acid substituent: for X = 3‐CO2−N+Bu4 δ(13C)COO = 166.4,
4‐CO2−N+Bu4 δ(13C)COO = 166.9, 2‐CO2−N+Bu4 δ(13C)
δ
O
¼ ð28:3 8:9Þ
(15)
S═O DMSO
ꢂ
ꢃ
17
X
þð0:833 0:048Þδ
O
S═O CH3CN
R ¼ 0:978; s ¼ 0:277; n=n0 ¼ 12=12
In ortho‐substituted phenyl tosylates, the sulfonyl oxy-
gen chemical shifts, δ(17O), in DMSO (Table 1) in the case
of neutral substituents showed a good correlation with
the Charton equation (Equations 4 and 5, Table 2) when
the data treatment was performed separately for deriva-
tives with electron‐donating (+R) substituents and elec-
tron‐withdrawing (−R) substituents. In the case of ortho
+R substituents (X = H, 2‐F, 2‐Cl, 2‐Br, 2‐OCH3, 2‐CH3,
2‐CH(CH3)2), we obtained (Equation 16, Table 2):
= 166.4. For COOH substituents, the corresponding
ꢂ
ꢃ
COO
17
δ(13C)COOH values were 166.1 for 3‐COOH, 166.4 for 4‐
COOH, and 165.5 for 2‐COOH (Supporting Information).
Earlier[4] in substituted phenyl benzoates, the COO−car-
rbon chemical shifts δ(13C)COO in tetrabutylammonium
carboxylate substituents were found also higher as com-
pared to that in the COOH substituent (for X = 3‐
CO2−N+Bu4 δ(13C)COO = 167.0, and for 3‐COOH δ(13C)
δ
O
¼ ð165:8 0:2Þ þ ð4:94 0:34ÞσI
þ ð8:59 0:63Þσ°R–ð4:43 0:60ÞEB
S═O
ortho
s
R ¼ 0:987; s ¼ 0:153; n=n0 ¼ 7=7
(16)
In ortho‐substituted phenyl tosylates in the case of +R
substituents in DMSO, the influence of the inductive, res-
onance, and steric terms (Equation 16) was approxi-
mately the same as was previously found for the
sulfonyl oxygen chemical shifts, δ(17O), for ortho‐
substituted phenyl tosylates in CH3CN[5] (ρI = 5.23,
ρR = 9.20, δ = −6.74).
= 166.5, for 4‐CO2−N+Bu4 δ(13C)COO = 167.5, and
COOH
for 4‐COOH δ(13C)COOH = 166.6).
The deshielding effect of carbonyl oxygens and the
carbonyl carbons in tetrabutylammonium carboxylate
substituents compared to that in the case of COOH sub-
stituents give evidence that in the tetrabutylammonium
carboxylate substituents in ion pair form, the electrons
are more delocalized (high‐frequency shift) as compared
to that in the case of COOH substituents.
The chemical shifts, δ(17O), for phenyl tosylates con-
taining electron‐withdrawing (−R) substituents (X = H,
2‐NO2, 2‐CN, 2‐CF3, 2‐COCH3, 2‐COPh, 2‐COOH)
showed the following correlation (Equation 17, Table 2):

NUMMERT ET AL.
11 of 13
ꢂ
ꢃ
17
The electron‐withdrawing (−R) substituents in phenyl
moiety in ortho‐substituted phenyl tosylates increase the
shielding effect of the single‐bonded oxygen by the sub-
stituent inductive effect similarly to that in substituted
phenyl benzoates.[7]
δ
O
¼ ð165:9 0:4Þ þ ð10:8 0:4ÞσI
–ð38:2 1:7Þσ°R –ð5:80 0:30ÞEB
S═O
ortho
(17)
s
R ¼ 0:997; s ¼ 0:109; n=n0 ¼ 6=7
In substituted phenyl tosylates, containing ortho
In the case of phenyl tosylates, approximately the
same correlation as shown in Table 2 was obtained when
the δ(17O) values of the sulfonyl oxygen of tosylates in
DMSO were determined on a Bruker Avance II 200 spec-
trometer (Tables S1 and S2).
−R substituents in the phenyl moiety, the sulfonyl oxygen
appeared to be deshielded by an inductive effect but
shielded by a resonance effect similar to that in the
case of +R substituents. The oxygens in ortho COOH
group revealed
a deshielding effect (for 2‐COOH
δ(17O) = 267.7 ppm, see Table 1) as compared to oxygens
in para COOH group (for 4‐COOH δ(17O) = 254.0 ppm,
Table 1). The similar deshielding effect of oxygens in
ortho NO2 substituent was detected earlier in substituted
phenyl tosylates and phenyl benzoates in CH3CN.[5,7]
From Equation 17 for ortho charged substituent 2‐N
+(CH3)3I−, the substituent constant (σI)X = 0.77 was
estimated (Table 4). The estimated (σI)X substituent con-
stant value for 2‐N+(CH3)3I− is in good agreement with
the (σI)X , constants for 2‐N+(CH3)3I−) obtained from
the IR carbonyl stretching frequencies, νCO ((σI)X
= 0.78[4]), the carbonyl carbon chemical shifts, δCO
((σI)X = 0.69[4]) for substituted phenyl benzoate series,
C6H5CO2C6H4‐X, in DMSO (Table 4).
5 | CONCLUSIONS
The 17O NMR spectra for 31 ortho‐substituted, para‐
substituted, and meta‐substituted phenyl esters of p‐
toluenesulfonic acid, 4‐CH3‐C6H4SO2OC6H4‐X, (phenyl
tosylates) with electroneutral and charged substituents
(X = 4‐N+(CH3)3I−, 3‐N+(CH3)3I−, 2‐N+(CH3)3I−, 4‐
COO−N+Bu4, 3‐COO−N+Bu4, and 2‐COO−N+Bu4) in
DMSO were recorded. The chemical shifts, δ(17O), for
sulfonyl oxygen (SO2) in DMSO were found at much
lower frequencies than the 17O signals of the carbonyl
(C═O) group in substituted phenyl benzoates. This
shows that in substituted phenyl tosylates in DMSO,
similar to that in CH3CN, the oxygen nuclei in SO2 group
are more shielded as compared to oxygen nucleus in
carbonyl (C═O) group in phenyl benzoates, and the
extent of the double bond contribution in SO2 group in
benzenesulfonates is reduced as compared to that in
the carbonyl C═O bond in phenyl benzoates,
C6H5CO2C6H4‐X. The substituent constants σ°X and
(σI)X for charged substituents (4‐N+(CH3)3I−, 3‐N
+(CH3)3I−, 2‐N+(CH3)3I−, 4‐COO−N+Bu4, 3‐COO−N
+Bu4) estimated in the present work using 17O NMR
chemical shifts, δ(17O), of the sulfonyl group for phenyl
tosylates in DMSO showed a good agreement with
the σ°X substituent constants for charged substituent
constants obtained earlier[4] using infrared carbonyl
stretching frequencies, νCO, the carbonyl carbon 13C
NMR chemical shifts, δCO, for substituted phenyl benzo-
ates, C6H5CO2C6H4‐X, in DMSO and with constants,
σ°X , for ion pairs proposed by Hoefnagel and Wepster[1]
and those calculated from the kinetic data of the alkaline
hydrolysis of substituted phenyl benzoates in concen-
trated aqueous NaCl solution (Table 4).[2] The substituent
constants, σ°X , for charged substituents obtained from
the chemical shifts, δ(17O), for substituted phenyl
tosylates in DMSO similarly to those obtained previously
The correlation using the modified Charton steric
constants, υ, gave nearly the same results as with the ste-
ric constants, EBs (Table 2). In the case of the Charton ste-
ric constants, υ, the susceptibility to the steric effect was
positive and the magnitude of δortho was ca. twice smaller
as compared to the δortho value obtained with the EBs ste-
ric constants (EB < 0, Table 2). Because of the steric
s
effect of substituents in the phenyl moiety (a van der
Waals deshielding effect[45–47]), the sulfonyl oxygens are
deshielded more than the unsubstituted derivative.
4.2.1 | Influence of ortho substituents on
the δ(17O) values of single‐bonded phenoxy
oxygen
The δ(17O) values for single‐bonded phenoxy oxygens in
ortho‐substituted phenyl tosylates in DMSO similar to
the δ(17O) values for sulfonyl oxygen (Equations 16 and
17, Table 2) showed a good correlation with the Charton
equation (Equations 4 and 5) when the data treatment
was performed separately for derivatives with +R and
−R substituents. Similar to para substituents in the case
of ortho +R substituents in the phenyl moiety, the sin-
gle‐bonded phenoxy oxygen is deshielded by −I inductive
effect and shielded by the inductive +I and +R resonance
effect of ortho substituents more than 3 times greater
(ρI = 18.2, ρR = 57.7, Table 2) than that on the double‐
bonded sulfonyl oxygen (ρI = 4.94, ρR = 8.59, Table 2).
using the infrared carbonyl stretching frequencies, νCO
and the carbonyl carbon 13C NMR chemical shifts, δCO
,
,
for substituted phenyl benzoates in DMSO[4] prove that
in DMSO, the charged substituents are predominantly

12 of 13
NUMMERT ET AL.
[16] V. Nummert, O. Travnikova, S. Vahur, I. Leito, M. Piirsalu, V.
Mäemets, I. Koppel, I. A. Koppel, J. Phys. Org. Chem. 2006, 19,
654.
in ion pair form and are not dissociated into free ions. It
was shown that for ion pairs of the charged substituents
in DMSO and concentrated aqueous salt solution, the
same substituent constants, σ°X , could be used.
[17] V. Nummert, M. Piirsalu, S. Vahur, O. Travnikova, I. A.
Koppel, Collect. Czech. Chem. Commun. 2009, 74, 29.
[18] V. Nummert, M. Piirsalu, I. A. Koppel, J. Phys. Org. Chem.
2013, 26, 352.
ACKNOWLEDGEMENT
[19] V. Nummert, M. Piirsalu, I. A. Koppel, Cent. Eur. J. Chem.
2013, 11, 1964.
This work was supported by the EU through the
European Regional Development Fund (TK141
“Advanced materials and high‐technology devices for
energy recuperation systems”) and by institutional
funding projects IUT20‐14 and IUT20‐15 from the Minis-
try of Education and Research of Estonia.
[20] V. M. Maremäe, Reakts. Sposobnost Org. Soedin. (Tartu) 1967,
4, 87.
[21] V. Nummert, M. Piirsalu, M. Lepp, V. Mäemets, I. Koppel,
Collect. Czech. Chem. Commun. 2005, 70, 198.
[22] V. M. Maremäe, V. A. Palm, Reakts. Sposobnost Org. Soedin.
(Tartu) 1964, 1, 85.
ORCID
[23] V. M. Maremäe, J. B. Asenbush, Reakts. Sposobnost Org. Soedin.
(Tartu) 1965, 2, 83.
Mare Piirsalu http://orcid.org/0000-0001-5836-2790
[24] V. M. Maremäe, T. O. Püssa, Reakts. Sposobnost Org. Soedin.
(Tartu) 1969, 6, 732.
REFERENCES
[25] V. M. Maremäe, T. O. Püssa, V. A. Palm, Reakts. Sposobnost
Org. Soedin. (Tartu) 1971, 8, 127.
[1] A. J. Hoefnagel, M. A. Hoefnagel, B. M. Wepster, J. Org. Chem.
1978, 43, 4720.
[26] V. M. Nummert (Maremäe), M. K. Uudam, Reakts. Sposobnost
Org. Soedin. (Tartu) 1973, 10, 863‐874.
[2] V. A. Palm, V. M. Nummert, T. O. Püssa, M. M. Karelson, I. A.
Koppel, Reakts. Sposobnost Org. Soedin. (Tartu) 1973, 10, 223.
[27] V. M. Nummert, V. A. Palm, Reakts. Sposobnost Org. Soedin.
(Tartu) 1975, 11, 579.
[3] N. Bjerrum, Z. Phys. Chem. 1923, 106, 219.
[28] C. Weygand, in Weygand‐Hilgetag. Organish‐Chemische
Experimentierkunst, 4th ed. (Eds: G. Hilgetag, A. Martini),
Johann Ambrosius Barth, Leipzig 1970.
[4] V. Nummert, M. Piirsalu, S. Vahur, L. Toom, I. Leito, I. A.
Koppel, J. Phys. Org. Chem. 2017, 30, e3608. https://doi.org/
10.1002/poc.3608
[29] S. E. Hazlet, J. Am. Chem. Soc. 1937, 59, 287.
[5] V. Nummert, V. Mäemets, M. Piirsalu, I. A. Koppel, Magn.
Reson. Chem. 2012, 50, 696.
[30] P. J. Stang, A. G. Anderson, J. Org. Chem. 1976, 41, 781.
[31] H. Rottendorf, S. Sternhell, Aust. J. Chem. 1963, 16, 647.
[6] V. Nummert, V. Mäemets, M. Piirsalu, I. A. Koppel, J. Phys.
Org. Chem. 2011, 24, 539.
[32] O. Exner, in Correlation Analysis in Chemistry, (Eds: N. B.
Chapman, J. Shorter), Plenum Press, New York, London 1978
439.
[7] V. Nummert, V. Mäemets, M. Piirsalu, S. Vahur, I. A. Koppel,
Collect. Czech. Chem. Commun. 2011, 76, 1737.
[33] R. W. Taft Jr., I. C. Lewis, J. Am. Chem. Soc. 1958, 80, 2436.
[8] D. W. Boykin (Ed), 17O NMR Spectroscopy in Organic Chemis-
try, CRC Press, Boca Raton, FL 1990.
[34] R. W. Taft Jr., S. Ehrenson, I. C. Lewis, R. E. Glick, J. Am.
Chem. Soc. 1959, 81, 5352.
[9] H. Duddeck, G. Tóth, A. Simon, in Nuclear Magnetic Resonance
(NMR) Data: Chemical Shifts for Oxygen‐17, (Eds: R. R. Gupta,
M. D. Lechner), Springer‐Verlag, Berlin, Heidelberg, New York
2002 III/35E.
[35] Tables of Rate and Equilibrium Constants of Heterolytic
Organic Reactions, (Ed: V. A. Palm), publishing house of
VINITI, Moscow, 1979, 5(2), 164‐165.
[36] M. Charton, J. Am. Chem. Soc. 1969, 91, 624.
[10] D. Monti, F. Orsini, G. S. Ricca, Spectrosc. Lett. 1986, 19, 91.
[37] V. M. Nummert, M. V. Piirsalu, Org. React. (Tartu) 1975, 11,
[11] A. L. Baumstark, P. Balakrishnan, M. Dotrong, C. J.
McCloskey, M. G. Oakley, D. W. Boykin, J. Am. Chem. Soc.
1987, 109, 1059.
921.
[38] V. Nummert, M. Piirsalu, V. Mäemets, I. Koppel, Collect. Czech.
Chem. Commun. 2006, 71, 107.
[12] M. Maccagno, A. Mele, R. Musio, G. Petrillo, F. Sancassan, O.
[39] V. Nummert, K. Ojassalu, A. Bogdanov, Org. React. (Tartu)
1989, 26, 92.
Sciacovelli, D. Spinelli, ARKIVOC 2009, viii, 212.
[13] O. Exner, H. Dahn, P. Péchy, Magn. Reson. Chem. 1992, 30, 381.
[40] M. H. Aslam, A. G. Burden, N. B. Chapman, J. Shorter, M.
Charton, J. Chem. Soc. Perkin Trans. 1981, 2, 500.
[14] P. Balakrishnan, A. L. Baumstark, D. W. Boykin, Org. Magn.
Reson. 1984, 22, 753.
[41] M. Charton, in Steric Effects in Drug Design, (Eds: M. Charton,
I. Motoc), Academie‐Verlag, Berlin 1983 107.
[15] V. Nummert, M. Piirsalu, V. Mäemets, S. Vahur, I. A. Koppel,
J. Phys. Org. Chem. 2009, 22, 1155.
[42] V. Palm, J. Chem. Inf. Comput. Sci. 1990, 30, 409.

NUMMERT ET AL.
13 of 13
[43] M. Karplus, J. A. Pople, J. Chem. Phys. 1963, 38, 2803.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
[44] M.‐T. Béraldin, E. Vauthier, S. Fliszár, Can. J. Chem. 1982,
60, 106.
[45] S. Li, D. B. Chesnut, Magn. Reson. Chem. 1985, 23, 625.
[46] S. Li, D. B. Chesnut, Magn. Reson. Chem. 1986, 24, 93.
How to cite this article: Nummert V, Piirsalu M,
Toom L, Kesvatera T, Leito I, Koppel IA. Effect of
charged and ortho substituents on 17O NMR
chemical shifts of substituted phenyl tosylates in
DMSO. J Phys Org Chem. 2018;e3870. https://doi.
org/10.1002/poc.3870
[47] R. G. Guy, R. Lau, A. U. Rahman, F. J. Swinbourne,
Spectrochim. Acta A 1997, 53, 361.
[48] V. Nummert, Org. Reactiv. (Tartu) 1975, 11, 621.
[49] V. Nummert, M. Piirsalu, Org. Reactiv. (Tartu) 1977, 14, 263.
[50] V. Nummert, M. Piirsalu, J. Chem. Soc., Perkin Trans. 2
2000, 583.