A study of NMR parameters of para-substituted polyfluorinated benzyl cations and their precursors

Article abstract of DOI:10.1016/j.jfluchem.2015.12.003

¹³C and ¹⁹F NMR spectra of a series of para-substituted heptafluorotoluenes and hexafluorobenzyl cations were analyzed. The quantum-chemical calculations of δ_13C and δ_19F chemical shifts (CSs) as well as ¹⁹F-¹⁹F and ¹³C-¹⁹F spin-spin coupling constants (SSCCs) were performed. ¹⁹F and ¹³C CSs were predicted by the GIAO/PBE/L22 method. The accuracy of theoretical CSs values increased in the computation for cations in conjunction with anions as ionic pairs. The computation by the SOPPA(CCSD) method had enough accuracy for the prediction of ¹⁹F-¹⁹F, ¹³C-¹⁹F SSCCs and Ramsey contributions. Changes in various contributions to the spin-spin interaction along with their dependence on substituents at the para-position were also examined on passing from heptafluorotoluenes to the corresponding benzyl carbocations.

Full text of DOI:10.1016/j.jfluchem.2015.12.003

Accepted Manuscript
Title: A study of NMR parameters of para-substituted
polyfluorinated benzyl cations and their precursors
Author: Dmitriy S. Fadeev Igor P. Chuikov Victor I.
Mamatyuk
PII:
S0022-1139(15)30015-4
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.jfluchem.2015.12.003
FLUOR 8700
To appear in:
FLUOR
Received date:
Revised date:
Accepted date:
18-9-2015
11-12-2015
15-12-2015
Please cite this article as: D.S. Fadeev, I.P. Chuikov, V.I. Mamatyuk,
A study of NMR parameters of para-substituted polyfluorinated benzyl
cations and their precursors, Journal of Fluorine Chemistry (2015),
http://dx.doi.org/10.1016/j.jfluchem.2015.12.003
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

A study of NMR parameters of para-substituted polyfluorinated benzyl cations and their
precursors
Dmitriy S. Fadeev, Igor P. Chuikov, Victor I. Mamatyuk
N.N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, Siberian Branch of Russian
Academy of Sciences, Lavrentiev Avenue 9, 630090 Novosibirsk, Russia
Highlights
NMR spectra of a set of polyfluorinated toluenes and benzyl cations were analyzed.
Chemical shifts in NMR spectra were predicted by GIAO/PBE/L22 method.
Accuracy of prediction is higher for cations in conjugation with anions as ionic pairs.
F–F and C–F spin-spin coupling constants were calculated by SOPPA(CCSD) method.
Abstract
13С and 19F NMR spectra of a series of para-substituted heptafluorotoluenes and
hexafluorobenzyl cations were analyzed. The quantum-chemical calculations of δ13C and δ19F
chemical shifts (CSs) as well as 19F–19F and 13C–19F spin-spin coupling constants (SSCCs) were
performed. 19F and 13C CSs were predicted by the GIAO/PBE/L22 method. The accuracy of
theoretical CSs values increased in the computation for cations in conjunction with anions as
ionic pairs. The computation by the SOPPA(CCSD) method had enough accuracy for the
prediction of 19F–19F, 13C–19F SSCCs and Ramsey contributions. Changes in various
contributions to the spin-spin interaction along with their dependence on substituents at the para-
position were also examined on passing from heptafluorotoluenes to the corresponding benzyl
carbocations.
Keywords
NMR spectroscopy, DFT, SOPPA, computation, fluorinated, benzyl carbocation.
1 Introduction
Polyfluorinated benzyl cations are intermediates in electrophilic substitution reactions catalyzed
by acids [1]. However, they can exist as long lived ions in super acidic media: these conditions
are favorable for the NMR study of the characteristic features of these particles. The values of
13C and 19F chemical shifts (CSs) and 19F–19F and 13C–19F spin-spin coupling constants (SSCCs)
can be used as effective tools for the electron structure investigation.
There are some successful studies on polyfluorinated benzyl cations [2], although a detailed
interpretation of the spectral data has not been accomplished. The main difficulty with the
interpretation of such spectra of fluorinated compounds is their complexity which results from
magnetic inequivalence and strong couplings of nuclear spins (when the system is no longer first
order). Therefore, a full quantum-chemical calculation of the NMR spectral parameters is
becoming important for polyfluorinated benzyl cations and their precursors.
The aim of this study was to use the quantum-chemical methods to calculate the NMR
parameters with sufficient accuracy for the iteration procedure to determine real CSs and SSCCs
and explain the obtained data. We also wanted to determine the role of a perturbing action on the
spin system of substituents with various donor-acceptor properties. For this purpose, we
analyzed a series of para-substituted polyfluorinated benzyl cations and their chemical
precursors (toluenes).
1
Page 1 of 21

2 Materials and Methods
2.1 Measurement of NMR parameters.
13C and 19F NMR spectra were recorded on Bruker AV-300, AV-400, and AV-600 (75.5, 100.6,
150.9 MHz for 13C and 282.4, 376.5, 567.7 MHz for 19F respectively). Individual para-
substituted heptafluorotoluenes were dissolved in CDCl3 to a concentration of 1.0 – 2.0 M;
hexafluorobenzene was added as the internal standard (δC C6F6 [TMS] = 138.3 ppm). The cations
were generated by dissolving perfluorotoluenes in double-distilled SbF5 to obtain 15-25 mol %
concentrations. To reduce viscosity the solution was diluted with SO2ClF (20 wt.%.). δ13C and
δ19F of polyfluorobenzyl cations were measured relative to the external standard (a glass capillary
with 1% C6F6 in acetone-d6) at a temperature varied from +10 to +30°C in order to obtain the
best resolution. Magnetic susceptibility corrections were not applied.
Broadband decoupling of protons and selective decoupling of fluorine atoms in both 19F and 13С
spectra allowed us to decompose the nuclear spin systems of the compounds into simpler
subsystems. We also used spectral data obtained from 13С satellites which in the case of a
subsystem of fluorine nuclei of the benzene ring were associated with the removal of symmetry
(AA’BB’ becomes ABCDX, where X is a 13C nuclei). Isotope shift differences are 1∆F(13C) -
3∆F(13C) = 45 ppb for one pair of chemically equivalent fluorine nuclei and 2∆F(13C) - 4∆F(13C)
= 18 ppb for another. The calculated data were used for the interpretation of the observed NMR
spectral parameters and as the initial set of values for the iteration of experimental NMR spectra.
For that purpose the gNMR 5.0 software [3] was used. The experimental SSCC values were
attributed to the signs of the calculated ones. A mismatch of signs for some experimental values
with theoretical ones was mostly observed for small values and probably resulted from
computation errors.
2.2 Computation details
Geometry optimization was performed using the quantum-chemical Priroda software [4] by the
density functional theory (DFT) method with the PBE functional [5] and the L11 basis set [6]
(Λ11, analog of cc-pCVDZ). We used the gauge including/invariant atomic orbital (GIAO)
approximation in combination with DFT and the L22 basis set (Λ22, analog of cc-pCVTZ) for
shielding computation. The values of CSs were calculated relative to the C6F6 shieldings
obtained by the same approach. To get initial approximations of tight ion pairs with random
coordinates, we used the Coalescence Kick software [7]. For each cation-anion pair, we
generated five initial approximations and then optimized the geometric parameters in the Priroda
software by the above method. The calculated CS values were averaged. The quantum-chemical
Dalton 2.0 software package [8] was used for the computation of SSCCs by analogy, as
described previously [9] at the DFT/B3LYP [10], SOPPA [11], and SOPPA(CCSD) [12] levels
of theory in combination with aug-cc-pVTZ-J [9, 13], aug-cc-pCVDZ [14], cc-pVDZ [15] basis
sets as reported below. Also we used aug-cc-pVTZ-Jf (14s5p2d/8s4p2d for F and C) basis set
which was derived from aug-cc-pVTZ-J (15s6p3d1f/9s5p3d1f for F and C) by disregarding
diffuse (1s1p1d) and f-functions in order to reduce the computation costs (see SI for more
details). Equivalent constants were averaged for compounds with asymmetrical geometries. The
Fermi contact (FC) contribution to the 19F–19F SSCCs was visualized using the DeMon software
[16].
3 Results and Discussion
3.1 The investigation of 19F and 13С Chemical Shifts.
Substituents at the para-position affect 13C and 19F CSs of polyfluorinated toluenes and benzyl
cations. In toluenes I (Scheme 1) CSs have changes for the fluorine nuclei located at the meta-
2
Page 2 of 21

position of the aromatic ring and for the carbon nuclei at the para-position (Table 1). In cations
II, fairly large changes were observed simultaneously for the fluorine atoms at the meta- and
alpha-positions and for the carbon atoms at the para-position [2]. In another study [17],
substituted pentafluorobenzenes were used to show that the effects of substituents on a
conjugated system can be reproduced with sufficiently high accuracy by means of a quantum-
chemical computation, namely, by the GIAO/PBE/L22 method. The same approach was used for
the computation of CSs in para-substituted heptafluorotoluenes I and hexafluorobenzyl cations
II.
The predicted 19F and 13C CSs for toluenes I (Figure 1, bars “I”) have the accuracy comparable
with that of pentafluorobenzenes in a nonpolar medium [17]. For cations II in the gas phase,
there is a substantial deviation of the calculated 19F CS values, whereas 13C CSs are predicted
with the accuracy similar to that of toluenes (Figure 1, bars “II”). This is likely to be due to that
the carbon nuclei are shielded from interactions with the medium by fluorine nuclei, whereas 19F
CSs are strongly sensitive to the effect of intermolecular interactions not taken into account in
the computation of the gas phase cations. Therefore, we calculated the cation-anion systems
when geometric parameters of the cations under study were optimized in conjunction with
SbnF5n+1¯ (n = 1, 2, 3, or 4) anions. Sb2F11¯ is more preferable than SbF6¯ due to more accurate
predictions of carbon shifts and more preferable than Sb3F16¯ or Sb4F21¯ because an anion
enlargement requires additional computational resources although it does not improve
significantly the accuracy (Figure 1). The II + Sb2F11¯ case has an increased R-squared parameter
and a reduced correction factor in the regression line correlation for 19F CSs as compared to the
calculation of free cations (Figure 2, a). Equations 1a and 2a describe the regressions and reflect
the improvement of the 19F shift values.
δF (exp.) = 1.071 δF (theor.) – 21.57;
δF (exp.) = 1.050 δF (theor.) – 9.93;
R2 = 0.986
R2 = 0.995
(1a)
(2a)
This change brings about a small improvement of 13С CSs (Figure 2, b).
δC (exp.) = 0.950 δF (theor.) + 1.51;
δC (exp.) = 0.995 δF (theor.) + 2.15;
R2 = 0.976
R2 = 0.985
(1b)
(2b)
In the case of toluenes I, the role of solvation effects seems to be modest; therefore, the
acceptable accuracy of the CS computation is achieved even for the isolated molecules in the gas
phase. Table 1 shows the experimental 19F and 13С CSs in comparison with the calculated ones
for toluenes I and benzyl cations II corresponding to the case of II+Sb2F11¯.
Table 1 data show that on going from toluenes I to positively charged cations II, changes in δF of
fluorine nuclei in the phenyl ring are downfield as follows: Fpara > Fortho > Fmeta. The same order is
observed for the carbon nuclei: Cpara > Cortho > Cmeta. For the Сipso carbon nuclei the ∆δC value is
on average 14 ppm upfield (the calculated value is 13 ppm). The largest ∆δF and ∆δC values
shifted downfield are observed for alpha-nuclei during the detachment of F¯; these changes are
related to the appearance of the positive charge and the alteration of hybridization of a carbon
atom. δF and δC at the alpha-position change symbatically and correlate well (Equation 3):
δFa = 4.82 δCa + 19.25;
R2 = 0.995
(3)
3
Page 3 of 21

This fact suggests that δF and δC are influenced by the same factors. These factors were described
recently by the electronic properties of substituents as Taft constants [18]. Within this theory a
substituent affects the system inductively (σI) and mesomerically (σR+). There are correlations
between δFa, δCa and Taft constants (Equations 4 and 5).
δFa = 30.67σI + 106.47σR+ + 199.69;
δCa = 5.85σI + 21.69σR+ + 37.48;
R2 = 0.918
(4)
(5)
R2 = 0.888
From these equations one can see that the coefficients at σR+ are substantially greater than those
at σI. Therefore, in the context of this approximation we can say that δF and δC at the alpha-
position are predominantly influenced by para-substituents due to the interaction transmitted via
the π-system and to a lesser extent they depend on inductive effects transmitted via space and
through σ-bonds. There are no correlations between other δF, δC and σ-constants for both
toluenes I and cations II.
The investigation of 13C–19F and 19F–19F spin-spin coupling constants.
We observed substantial changes in 19F–19F and 13C–19F SSCCs on going from toluenes I to
corresponding benzyl cations II. According to the comparison of geometric parameters obtained
after the optimization (mainly in the aromatic ring) these changes may be only partially related to
geometric differences (Figure 3).
The double resonance experiments enable the analysis of spin systems as a combination of
simpler subsystems. For example, after decoupling from the fluorine nuclei at the meta-position,
the AA'BB'XX' spin system of the fluorine nuclei in cation IIb transforms to the AA'XX' system
(Figure 4), which has an analytical solution and yields absolute values of JFF SSCCs. These
values have been used for interpretation of more complex IIa spin system due to similarity of JFF
SSCCs (experimental data in Table 11).
Obtained values of 2JFa-Fa', 4JFa-Fo and 4JFa-Fo' SSCCs in cation IIa are compared with results from
SOPPA, SOPPA(CCSD) and DFT calculations. We used three variants of basis sets for F and C
atoms which nuclei involved in the calculated couplings: (A) aug-cc-pVTZ-J, (B) aug-cc-pVTZ-
Jf, (C) aug-cc-pCVDZ. For all other atoms we employed cc-pVDZ basis set (Table 2).
B3LYP was shown to be suitable functional for prediction JFF couplings in neutral
polyfluorinated aromatic compounds [19]. Nevertheless, in the case of cation IIa which was
chosen as a model compound we found that B3LYP in combination with all basis sets applied
provided poor accuracy for 2JFF and 1JCF couplings mostly because of inexact estimation of FC
contribution (see full version of Table 2 in SI). These SSCCs were markedly better reproduced
by SOPPA and SOPPA(CCSD). The analysis of calculated values for JFF couplings showed that
basis sets A and B provided a closer match to the experiment in comparison with C basis set
which gave larger deviations for 2JFa-Fa’ and 4JFa-Fo. For basis set A the data calculated with
SOPPA had no significant differences with that obtained with SOPPA(CCSD). On the other
hand, we found the close results for less-demanding basis set B even in the case of paramagnetic-
spin orbit (PSO) contribution which could be sensitive to the absence of diffuse and f-functions.
SOPPA(CCSD) yielded slightly closer match to the experimental data in comparison with
SOPPA. For this reason, all other nJFF SSCCs for toluenes I and cations II discussed in this work
were calculated at SOPPA(CCSD) level of theory in combination with basis set B (the data of
Tables 5 – 9).
4
Page 4 of 21

In contrast to the case of JFF SSCCs the analysis of calculated 1JCF couplings showed better
accuracy of basis set С over basis sets A and B in accuracy. Similarly as in the case of JFF SSCCs
SOPPA(CCSD) predicted 1JCF couplings with higher accuracy then SOPPA. This was a reason to
calculate all other 1JCF couplings for toluenes I and cations II discussed in present work at
SOPPA(CCSD) level of theory in combination with basis set C (the data of Tables 3 and 4).
Comparing Ramsey contributions in the calculated SSCCs for octafluorotoluene Ia and
heptafluorobenzyl cation IIa, we can say that the main contributions that cause substantial
changes in the 1JCF SSCCs are the PSO and the FC contributions (Table 3).
In cation IIa there is the relation between the 1JCF SSCCs values and the data on δF and δC: the
greater the 19F and 13C nuclei shifted downfield, the greater the absolute value of the 1JCF
constant (Tables 1 and 3). A similar relation was reported elsewhere [20] for fluoro-substituted
carbocations. At the same moment the calculated PSO contribution of 1JCF SSCCs correlates well
with δF and δC in cation IIa, but in precursor Ia it has no dramatic changes and correlates only
with δF and do not with δC, which correlates with the FC contribution. Regarding the spin-dipole
(SD) contribution in IIa, the values change insignificantly for alfa-, ortho- and para-positions
and fall into the range from -5.66 to -4.57 Hz; it is 1.56 Hz at the meta-position and this is
consistent with the concept of charge delocalization through resonance alfa-, ortho-, para-
positions and the nonresonance meta-position. In both toluene and cation the diamagnetic-spin
orbit (DSO) contribution is small and does not undergo noticeable changes; therefore, it will not
be analyzed further. These observations are true for all toluenes I and benzyl cations II
examined.
The 1JCF SSCCs turned out to be almost insensitive to the effect of a substituent in toluenes I.
The greatest range of values is approximately 6 Hz for the meta-position. At the same time, the
greatest constant corresponds to the CH3-substituent and the smallest to the F- and Cl-
substituents (Table 10). Among the cations, the picture for the nonresonance meta-position is
similar, and the range is 12.0 Hz (Table 4 and 11). According to the calculated data, these
changes arise from the FC interaction, which is a scalar value and reflects s-orbital bonding. The
observed picture is most likely to be a consequence of the inductive effect of the substituents
examined, which spreads via σ-bonds, thereby affecting the meta-C-F bond near the substituent.
For the resonance alpha-position, the degree of influence of para-substituents on 1JCF SSCC is as
follows: H < F < Cl < Br < CH3 < OCH3 (Tables 4 and 11), which is in line with the results of
another study [1]; the difference between -H and -OCH3 is 18.2 Hz. According to calculated data,
substantial changes occur due to PSO contribution which correlates well with δCa and δFa. An
increase in the PSO term corresponds to the upfield shifting (Tables 1 and 4).
Due to calculated values of germinal 2JFa –Fa' the decisive role belongs to the PSO term in
comparison with other contributions (Table 5) and it is mostly responsible for the differences
observed for the value of the constant in question in the series of the cations tested; that is, the
effect of a para-substituent manifests itself in the PSO mechanism of transmission of the spin-
spin interaction. Since the PSO mechanism is effective only for orbitals with the non-s-character
[21], it is possible that in the calculated values of PSO the observed changes reflect the influence
of mesomeric effects of para-substituents on 2JFF SSCC via a π-conjugation system. In contrast,
the inductive effects of these substituents can be associated with the change in the FC
contribution.
If we discuss this topic in terms of resonance structures, then the 3JFo-Fm and 3JFm-Fp SSCCs are at
nonresonance positions relative to each other. At the same time, we did not observe substantial
differences in these constants for toluene Ia and cation IIa (Table 6). Similarly, there is no
5
Page 5 of 21

significant difference in the calculated values of the contributions: negative PSO values are
partially compensated by positive SD; the DSO and FC terms are small in comparison with the
total value. Thus, the mechanisms of transmission of a spin-spin interaction between the fluorine
nuclei at the ortho- and meta- or meta- and para-positions in toluene Ia and cation IIa show
insignificant differences. In these pairs the fluorine nuclei are spatially closer together than in the
pair corresponding to 4JFa-Fo SSCC of cations II; however, the FC contribution has small absolute
values for both toluene Ia and cation IIa.
According to computation, the FC contribution to 4JFa-Fo SSCC is 54.90 Hz, and this constant is
related to the resonance fluorine nuclei. The effective orbital overlap of the fluorine atoms
provides the multiple transmission pathways of the spin-spin interaction via space and through
bonds (Figure 5).
We used pentafluoroallyl cation and 1,1,2,3-tetrafluorobutadiene as models to show the changes
in 4JF-F SSCCs during a hypothetical intramolecular rotation along the C-C bond (Figure 6).
As follows from calculation the value of 4JFa-Fb SSCC in pentafluoroallyl cation (corresponds to α
= 0° in Figure 6a) is described by large positive value of FC contribution as well as PSO and SD
(experimental values are from [22]). The rotation along C—C bond to α = 180° transforms this
coupling to 4JFa-Fb’ where only SD contribution dominates. FC contribution has similar behavior
for charged and non-charged compounds under research and in the case of 1,1,2,3-
tetrafluorobutadiene this contribution determines predominantly the value of total 4JF-F SSCC in
a range of α = 0°-120° (Figure 6b). Due to Table 7 and to data of Figure 6c, in the benzyl cation
IIa the spin-spin transition mechanisms of discussed 4JFF are more similar to those in the
pentafluoroallyl cation.
The PSO and SD contributions also have high values for other examined 4JF-F SSCCs
corresponding to the fluorine nuclei at the resonance positions of cation IIa. At the same time,
the values of the FC contribution in all remaining 4JF-F SSCCs in both toluene Ia and cation IIa
are low. In the case of 4JFa-Fo, 4JFa’-Fo, and 4JFo-Fp SSCCs, the values of PSO, SD and FC terms
have the positive sign; this situation results in the high total value of SSCCs. In 4JFo-Fo SSCC, the
PSO and SD contributions have opposite signs and partially compensate each other. In the case
of 4J Fm-Fm SSCC, which corresponds to the fluorine nuclei at nonresonance positions, we
observed small calculated values of the contributions.
The 4JF-F SSCCs where the fluorine nuclei are formally separated by four bonds can occur by the
8
multiple transmission pathways through six or eight bonds (6JFo-Fo, 6JFm-Fm, JFa-Fo, as example).
We did not study the ratio of the effectiveness of different pathways but assume that alternative
pathways can have a sizeable contribution to the spin-spin transmission, especially in the case of
5JFo-Fm’ SSCC where the interaction can occur by two pathways through five bonds. 5JF-F
constants correspond to the nuclei at nonresonance positions and we did not observe there either
strong contributions to the SSCC or significant differences between toluene Ia and cation IIa
(Table 8).
Conversely, in 6JF-F SSCC, which reflects the spin-spin interaction of fluorine nuclei at the
resonance positions of cation IIa, we again observe a clear-cut difference from SSCC across the
6
Page 6 of 21

same number of bonds in toluene Ia. The calculated data predict the SD mechanism as the main
interaction forming 6JFa-Fm SSCC in cation IIa (Table 9).
4 Conclusions
We found that the geometry optimization and computation of δF and δC by the GIAO/PBE/L22
method could be successfully applied to interpret the observed CSs in the 13С and 19F NMR
spectra of para-substituted polyfluorinated toluenes and benzyl cations. To increase the
computation accuracy of 13С and 19F shieldings in cations, it is necessary to calculate them as
tight cation-anion pairs. We found that it was sufficient to take into account the relatively small
Sb2F11¯ anion in the case of cations II. These data indirectly suggest that in the antimony
pentafluoride medium the examined cations exist in the form of complexes as ionic pairs and do
not exist as individual particles.
The calculation of 13C–19F and 19F–19F SSCCs can be performed by the SOPPA and
SOPPA(CCSD) methods with the accuracy sufficient for the subsequent iteration of the spectra
and for the interpretation of the transmission mechanisms of the spin-spin interaction. The
revealed dependence of 13C and 19F CSs, as well as JCF, JFF SSCCs and Ramsey contributions on
the position of nuclei in the structure, especially for cations II, is in good agreement with the
widely accepted theories of electron resonance delocalization characteristic of these compounds.
On the other hand, the quantum-chemical computation provides a more reliable prediction of
NMR spectral parameters of polyfluorinated benzyl cations and their precursors than the
empirical correlations. The results of the quantum-chemical computation make it possible to
discuss 1) the mechanisms of transmission of the spin-spin interaction in the spin systems
analyzed here and 2) their dependence on the electronic structure and the effect of substituents.
Acknowledgments
The NMR spectra were recorded at the Chemical Service Center, SB RAS, Novosibirsk.
Quantum-chemical calculations were performed on the cluster at the Informational-
Computational Center of Novosibirsk State University.
We thank А.М. Genaev, G.Е. Salnikov and V. M. Karpov (Novosibirsk Institute of Organic
Chemistry) for assistance with the research and preparation of the data for publication . This
work was supported by the Russian Foundation for Basic Research (grant # 14-03-31279) and by
the Ministry of Education and Science of the Russian Federation (2014-2016).
References
[1] V. D. Shteingarts, Polyfluorinated carbocations, in: G. A. Olah, G. K. Surya Prakash (Ed.)
Carbocation Chemistry, Wiley, New Jersey, 2004, pp. 159-212.
[2] Yu. V. Pozdnyakovich, V. D. Shteingarts, J. Fluor. Chem. 4 (1974) 283-296.
[3] gNMR 5.0 software, http://home.cc.umanitoba.ca/~budzelaa/gNMR/gNMR.html
[4] D. N. Laikov, Chem. Phys. Lett. 281 (1997) 151-156.
[5] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865-3868.
[6] D. N. Laikov, Chem. Phys. Lett. 416 (2005) 116-120.
[7] A. P. Sergeeva, B. B. Averkiev, H. Zhai, A. I. Boldyrev, L. Wang , J. Chem. Phys. 134
(2011) 224304-224314.
[8] K. Aidas, C. Angeli, K. L. Bak, et al, WIREs Comput. Mol. Sci. 4 (2014) 269-284.
[9] V. Barone, P. F. Provasi, J. E. Peralta, J. P. Snyder, S. P. A. Sauer, R. H. Contreras, J. Phys.
Chem. A. 107 (2003) 4748-4754.
[10] A. D. Becke, J. Chem. Phys. 98 (1993) 5648–5652.
[11] (a) E. S. Nielsen, P. Jørgensen, J. Oddershede, J. Chem. Phys. 73 (1980) 6238-6246.
(b) J. Geertsen, J. Oddershede, J. Chem. Phys. 90 (1984) 301-311.
[12] (a) T. Enevoldsen, J. Oddershede, S. P. A. Sauer , Theor. Chem. Acc. 100 (1998) 275-284.
(b) S. P. A. Sauer, J. Phys. B: At. Mol. Opt. Phys. 20 (1997) 3773-3780.
7
Page 7 of 21

[13] P. F. Provasi, G. A. Aucar, S. P. A. Sauer , J. Chem. Phys.115 (2001) 1324–1334.
[14] R. A. Kendall, T. H. Dunning Jr, R. J. Harrison, J. Chem. Phys. 96 (1992) 6796-6806
[15] T. H. Dunning Jr., J. Chem. Phys. 90 (1989) 1007–1023.
[16] (a) O. L. Malkina, V. G. Malkin, Angewandte Chemie Int. 42 (2003) 4335-4338.
(b) O. L. Malkina, Interpretation of indirect spin-spin coupling constants, in: M. Kaupp, M. Buhl
and V. G. Malkin (Ed.) Calculation of NMR and EPR parameters: theory and applications,
Wiley-VCH, Weinheim, 2006, pp. 307-325.
[17] D. S. Fadeev, I. P. Chuikov, V. I. Mamatyuk , J. Struct. Chem. 54 (2013) Supplement 1,
180-185.
[18] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 91 (1991) 165-195.
[19] V. Barone, P. F. Provasi, J. E. Peralta, J.P. Snyder, S.P.A. Sauer, R.H. Contreras, J. Phys.
Chem. A. 106 (2002) 5607-5612.
[20] V. I. Mamatyuk, Yu.V. Poznyakovich, B.G. Oksenenko, V. I. Buraev, E.V. Malykhin, V. D.
Shteingarts, Izv. Akad. Nauk SSSR Ser. Khim. (1975) 1626-1628.
[21] J. Gräfenstein, D. Cremer, Chem. Phys. Lett. 383 (2004) 332-343.
[22] M. V. Galakhov, V. A. Petrov, V. I. Bakhmutov, et al , Izv. Akad. Nauk SSSR Ser. Khim.
(1985) 306-312.
Scheme 1. Formation of para-substituted benzyl cations.
Fig. 1 Root mean square deviation of calculated chemical shifts (13С – dark bars, 19F – light gray
bars) for toluenes I, cations II and cations II as ionic pairs with SbF6¯, Sb2F11¯, Sb3F16¯,
Sb4F21¯anions.
Fig. 2 The regression line correlations between the theoretical and experimental values. (a) 19F
CSs. (b) 13C CSs. Lines 1 correspond to isolated cations II and lines 2 correspond to cations in
the form of tight ionic pairs with the Sb2F11 ¯anion.
Fig. 3 Bond lengths (Å) and bond angles; these data were obtained from the geometry
optimization of toluene Ia and cation IIa.
Fig. 4 Experimental (a) and simulated (b) 19F-NMR spectra of ortho-F in heptafluorobenzyl
cation IIa (left) and p-Br-hexafluorobenzyl cation IIb with decoupled meta-F (right).
Fig. 5 Pathways of transmission of the Fermi contact interaction in 4JFa-Fo SSCC of the
heptafluorobenzyl cation according to visualization of the difference in the electron densities
between states with parallel and antiparallel nuclear spins (coupling deformation density) [16].
Fig. 6 Calculated 4JFF SSCC vs. dihedral angle in the pentafluoroallyl cation (a), 1,1,2,3-
tetrafluorobutadiene (b) and heptafluorobenzyl cation (c).
Table 1 Experimental vs. calculated values (in parentheses) of δF and δC relative to C6F6 (ppm).
Exp.
(Calc.)
Fortho
Fmeta
Fpara
CF3
CF2
Cipso
Cortho
Cmeta
Cpara
CF3
CF2
Ia
IIa
22.6 (28.1) 2.6 (3.9)
15.3 (18.4) 106.0 (106.4) -32.3 (-36.7) 6.9 (6.7)
0 (-0.2)
5.8 (5.7)
-17.4 (-9.2)
36.7 (32.1)
64.2 (74.1) 16.2 (19.5) 83.4 (96.8) 195.7 (191.4) -45.2 (-48.6) 16.2 (15.3) 1.2 (1.7) 26.3 (25.5)
Ib
IIb
22.6 (28.2) 23.8 (28.8) --
59.5 (69.2) 36.3 (41.5)
105.6 (105.9) -29.5 (-33.7) 6.3 (6.2)
194.4 (192.1) -43.7 (-46.9) 13.7 (13.0) 7.6 (8.0) 14.9 (11.7)
6.5 (6.6) -20.8 (-21.9) -17.5 (-9.3)
36.1 (31.7)
Ic
IIc
23.0 (29.0) 31.6(37.3)
58.4 (67.8) 44.3 (49.4)
--
105.6 (105.7) -28.8 (-33.0) 6.3 (6.2)
193.6 (191.2) -43.1 (-46.5) 12.8 (12.3) 9.0 (8.9) 8.2 (16.5) 35.9 (31.5)
7.4 (7.4) -32.8 (-22.9) -17.3 (-9.3)
Id
IId
21.9 (29.2) 25.7 (32.8) --
62.6 (76.7) 38.9 (46.2)
105.7 (105.4) -27.2 (-30.4) 6.2 (6.1)
202.0 (197.8) -41.4 (-44.3) 14.7 (14.2) 8.9 (9.2) 0.9 (-8.4) 38.0 (33.2)
8.4 (8.2) -28.2 (-37.2) -17.2 (-9.1)
Ie
IIe
19.8 (25.2) 20.7 (24.5) --
58.3 (69.9) 33.9 (38.6)
105.8 (105.7) -30.8 (-33.7) 5.8 (5.5)
189.8 (189.9) -44.4 (-47.5) 13.3 (12.7) 8.1 (8.4) 23.5 (15.1)
7.3 (7.1) -16.9 (-23.2) -17.0 (-8.9)
35.5 (32.0)
If
IIf
20.1 (23.7) 5.3 (0.4)
54.0 (61.2) 15.1 (17.3)
--
106.5 (106.5) -35.5 (-41.2) 6.9 (6.6) 2.5 (0.5) 3.6 (-1.4)
163.3 (174.6) -50.7 (-52.1) 14.0 (13.1) 1.3 (2.6) 30.0 (20.8)
-17.0 (-8.7)
29.9 (28.9)
8
Page 8 of 21

Table 2 nJFF Couplings (in Hz) for IIa in dependence of basis set* and calculation approach.
SOPPA
SOPPA (CCSD)
B3LYP
nJFF
Exp.abs
A
B
C
A
B
C
A
B
C
Fa - Fa'
Fa - Fo
Fa - Fo'
270.3
120.7
27.4
-280.67
122.69
27.19
-288.89 -301.31 -282.41 -286.27 -298.29 -406.55
-414.82
147.79
37.18
-15.25
-479.70
-374.79
-424.16
125.30
32.59
124.85
27.79
113.11
25.59
118.27
26.08
120.43
26.63
107.62
24.03
148.50
37.24
Fo-Fo'
Ca-Fa
14.5
-17.17
-417.52
-304.86
-17.76
-17.46
-16.95
-17.65
-17.16
-15.25
-17.06
-456.88
-356.90
365.4
270.5
-417.64 -399.73 -408.24 -408.20 -390.37 -478.91
-305.69 -296.11 -294.99 -295.52 -285.04 -374.58
Cm-Fm
* Basis set A uses aug-cc-pVTZ-J for F and C involved in the calculated couplings, B – the same
with aug-cc-pVTZ-Jf, C – the same with aug-cc-pCVDZ. For all other atoms cc-pVDZ basis set
was employed.
abs Experimental SSCC values are absolute ones.
Table 3 Experimental vs. theoretical values of 1JCF SSCC as well as Ramsey contributions (in
Hz) for toluene Ia and benzyl cation IIa.
Ia
IIa
DSO
1.55
1.23
1.17
1.15
PSO
SD
FC
Sum
Exp.
DSO
1.19
1.17
PSO
SD
FC
Sum
Exp.
CF3*
-23.54 7.16
-262.90 -277.73 -273.6 Ca-Fa
-269.77 -267.12 -259.4 Co-Fo
-69.66 -5.47
-21.24 -5.66
-316.43 -390.37 -365.4
-290.37 -316.09 -299.9
-284.80 -285.04 -270.5
-291.69 -329.57 -304.8
Co-Fo
Cm-Fm
Cp-Fp
-0.63
5.81
2.06
3.09
2.00
-266.44 -256.37 -252.3 Cm-Fm 1.15
-268.96 -267.00 -259.8 Cp-Fp 1.11
-2.95
1.56
-1.18
-34.43 -4.57
*
averaged
Table 4 Experimental vs. theoretical values of 1JCF (and Ramsey contributions) for the series of
para-substituted hexafluorobenzyl cations II.
F
Cl
Br
H
CH3
OCH3
Ca-Fa
DSO
PSO
SD
1.19
1.20
1.20
1.19
1.20
1.21
-69.66
-5.47
-66.96
-6.22
-68.34
-6.02
-70.78
-5.32
-66.32
-5.99
-55.49
-8.66
FC
-316.43
-390.37
-365.4
1.17
-312.01
-383.99
-365.1
1.17
-313.36
-386.52
-364.8
1.19
-312.54
-387.45
-368.5
1.14
-310.68
-381.79
-363.1
1.16
-313.24
-376.18
-350.3
1.17
Sum
Exp.
DSO
PSO
SD
Co-Fo
-21.24
-5.66
-22.45
-7.77
-22.71
-8.67
-23.55
-10.49
-291.08
-323.98
-300.7
1.06
-21.95
-8.14
-17.41
-3.63
FC
-290.37
-316.09
-299.9
1.15
-290.43
-319.47
-302.9
1.18
-291.55
-321.74
-302.3
1.28
-288.98
-317.92
-299.3
1.12
-290.98
-310.84
-292.1
1.12
Sum
Exp.
DSO
PSO
SD
Cm-Fm
-2.95
-5.63
-6.41
-5.42
-3.05
-0.41
1.56
0.92
0.88
1.46
2.12
1.98
FC
-284.80
-285.04
-270.5
-273.78
-277.31
-265.1
-274.42
-278.66
-262.0
-273.70
-276.61
-263.4
-268.06
-267.87
-258.5
-277.13
-274.44
-264.0
Sum
Exp.
9
Page 9 of 21

Table 5 Calculated 2JFF SSCC and Ramsey contributions (in Hz) in a series of para-substituted
hexafluorobenzyl cations II.
F
Cl
Br
H
CH3
OCH3
DSO
PSO
SD
-0.93
-0.91
-0.89
-0.95
-0.92
-0.89
-333.03
1.55
-326.87
-2.22
-330.48
-0.54
-351.19
0.05
-321.09
-1.27
-250.28
-6.19
FC
46.14
-286.27
-270.3
50.04
49.86
51.48
-300.62
-282.9
49.53
49.03
Sum
Exp.
-279.95
-264.5
-282.05
-263.7
-273.75
-254.4
-208.33
-191.3
Table 6 Experimental vs. theoretical values of 3JFo-Fm and 3JFm-Fp SSCCs (in Hz) in toluene Ia and
cation IIa.
Ia
IIa
DSO
0.40
0.22
PSO
SD
FC
Sum
Exp.
-20.6
-19.6
DSO
0.27
0.20
PSO
SD
FC
Sum
Exp.
-17.7
-22.4
Fo-Fm
Fm-Fp
-26.36
-22.49
12.46
11.76
-0.79
-2.39
-14.30
-12.90
-29.31
-23.12
16.81
7.41
-0.55
-3.81
-12.78
-19.32
Table 7 Theoretical vs. experimental values of 4JF-F SSCC (in Hz) in toluene Ia and cation IIa.
Ia
IIa
4JF-F
4JF-F
DSO
0.17
PSO
2.46
SD
FC
Sum
Exp.
22.4
DSO PSO
1.12 44.83
-1.08 13.73
-1.05 30.22
SD
FC
Sum
Exp.
Fa-Fo
19.59
10.69
12.91
54.90
3.30
-1.06
2.73
120.43
26.63
41.02
-17.65
120.7
27.4
CF3-Fo
0.20
18.94
21.77
Fa-Fo
'
Fo-Fp
-0.99
-0.98
4.90
-1.57
-4.70
-3.60
-3.01
-1.26
-17.4
5.8
Fo-Fp
Fo-Fo
40.7
-8.80
-7.5
-1.01 -30.83 11.46
-1.05 2.69 0.27
-14.5
Fo-Fo
'
'
-1.03
-0.50
-3.22
-4.21
-8.95
0.0
-3.28
-1.37
-1.8
Fm-Fm
'
Fm-Fm
'
Table 8 Theoretical vs. experimental values of 5JFF (in Hz) in toluene Ia and cation IIa.
Ia
IIa
DSO
PSO
1.08
SD
FC
Sum
0.62
Exp.
0.6
DSO
0.67
PSO
SD
FC
Sum
Exp.
Fa-Fm
-1.39
-2.11
-0.97
-5.14
-2.2
5.8
CF3-Fm -0.67
Fo-Fm’ -1.00
0.70
-0.49
-0.91
-1.03
10.48
-6.64
1.92
6.90
-0.52
3.17
10.97
2.41
Fa-Fm
'
-8.88
14.26
6.30
10.86
8.2
2.59
Fo-Fm
'
Table 9 Theoretical vs. experimental values of 6JFF (in Hz) in toluene Ia and cation IIa.
Ia
IIa
DSO
CF3-Fp -0.70
PSO
0.82
SD
FC
Sum
1.47
Exp.
1.4
DSO
-0.80
PSO
2.04
SD
FC
Sum
Exp.
21.8
-0.78
2.12
Fa-Fp
21.25
5.26
27.74
Table 10 The 19F–19F and 13C–19F SSCCs (in Hz) in polyfluorinated para-substituted toluenes I.
SSCC
F
Cl
Br
H
CH3
OCH3
10
Page 10 of 21

CF3 - Fo
CF3 - Fm
CF3 - Fp
Fo - Fo'
Fo - Fm
Fo - Fm'
Fo - Fp
22.4
0.6
21.9
0.2
22.0
0.7
22.3
0.4
--
21.9
1.9
--
21.7
0.7
--
1.4
--
--
-7.5
-20.6
8.2
-6.8
-21.5
10.6
--
-6.4
-22.1
10.8
--
-6.1
-20.7
13.1
--
-6.8
-21.0
12.6
--
-7.3
-20.5
8.3
--
5.7
Fm - Fm'
Fm - Fp
CF3 - CF3
CF3 - Fo
CF3 - Fm
CF3 - Fp
Ci - CF3
Ci - Fo
0
-3.1
--
-3.3
--
0
1.5
--
3.0
--
-19.6
-273.6
1.5
--
-274.8
1.6
-274.8
1.6
-274.6
1.5
3.4
--
-274.0
1.6
3.5
--
-273.5
1.5
3.2
--
3.0
3.2
3.2
1.5
--
--
35.4
13.0
0.7
35.2
12.7
~0.1
--
35.2
12.8
0.5
34.6
12.3
0.7
--
34.7
12.7
0.7
--
35.1
13.1
0.7
--
Ci - Fm
Ci - Fp
4.4
--
Co - CF3
Co - Fo
Co - Fo'
Co - Fm
Co - Fm'
Co - Fp
Cm - CF3
Cm - Fo
Cm - Fo'
Cm - Fm
Cm - Fm'
Cm - Fp
Cp - CF3
Cp - Fo
Cp - Fm
Cp - Fp
1.9
1.8
1.8
1.8
-259.5
3.3
15.9
4.5
--
1.9
-257.5
6.4
14.3
4.3
--
-259.4
6.0
-261.6
4.8
-262.7
4.4
-258.1
4.3
13.1
4.5
15.5
4.8
16.7
4.6
16.7
4.3
3.9
--
--
--
0
~0.2
15.7
4.9
0.2
0.2
13.9
4.8
-250.9
9.5
--
~0.4
13.5
6.5
0.4
15.0
4.6
-248.7
4.1
--
16.5
5.2
15.4
4.8
-252.3
1.8
-252.4
1.5
-250.4
2.4
-246.6
4.6
12.9
0.7
--
--
--
0.8
0.9
0.7
1.2
22.4
--
0.4
0.5
3.7
10.9
--
4.9
2.7
1.7
0.8
13.3
-259.8
18.8
--
22.3
--
19.1
--
Table 11 19F–19F and 13C–19F SSCCs (in Hz) in polyfluorinated para-substituted benzyl cations
II.
SSCC
F
Cl
Br
H
CH3
-254.4
119.0
26.7
-2.1
6.8
OCH3
-191.3
102.4
19.9
-1.7
Fa –Fa'
Fa - Fo
Fa - Fo'
Fa - Fm
Fa - Fm'
Fa - Fp
-270.3
+120.7
+27.4
-2.2
-264.5
119.9
26.9
-3.3
6.5
-263.7
120.2
25.7
-2.1
6.8
-282.9
124.5
29.7
-2.4
5.8
+5.8
19.4
--
+21.8
--
--
--
--
11
Page 11 of 21

Fo - Fo'
Fo - Fm
Fo - Fm'
Fo - Fp
Fm - Fm'
Fm - Fp
Ca - Fa
Ca - Fo
Ca -Fm
Ca - Fp
Ci - Fa
-14.5
-17.7
+2.6
+40.7
-1.8
-22.4
-365.4
-2.1
2.2
-16.3
-18.4
6.6
-17.1
-19.4
6.8
-20.8
-19.9
10.7
--
-15.5
-18.3
6.8
-14.6
-16.1
0.1
--
--
--
--
-1.1
--
-1.3
--
-1.5
--
-6.53
--
0.7
--
-365.1
-2.1
2.2
-364.8
-2.0
2.4
-368.5
-0.9
3.2
-363.1
-1.0
3.9
-350.3
-1.4
5.2
4.8
--
--
--
--
--
7.3
7.3
7.8
6.4
8.0
12.0
11.3
-1.0
--
Ci - Fo
8.2
8.5
8.5
7.7
8.9
Ci - Fm
Ci - Fp
-1.4
-1.0
-2.5
10.9
-299.9
3.0
-1.4
--
-1.4
--
-2.9
--
-3.2
--
Co - Fa
Co - Fa'
Co - Fo
Co - Fo'
Co - Fm
Co - Fm'
Co - Fp
Cm - Fa
Cm - Fa'
Cm - Fo
Cm - Fo'
Cm - Fm
Cm - Fm'
Cm - Fp
Cp - Fa
Cp -Fo
-2.5
11.9
-302.9
5.0
-3.7
10.9
-302.3
2.5
-7.2
13.5
-300.7
2.0
-3.5
14.4
-299.3
4.0
-6.7
10.3
-292.1
6.1
15.8
2.3
13.4
2.5
13.4
0.8
14.8
1.9
17.0
0.4
13.6
-0.2
--
6.5
--
--
--
--
-2.5
2.9
-2.5
2.4
-2.5
2.4
-2.5
2.3
-2.5
2.5
0.8
5.9
10.2
1.8
12.2
0.1
11.9
-1.4
-262.0
0.0
11.2
0.4
10.5
0
13.9
-1.2
-264.0
-4.8
--
-270.5
-1.8
9.3
-265.1
-0.3
--
-263.4
7.9
-258.5
4.6
--
--
--
5.7
4.4
5.1
5.4
5.6
0.6
7.0
7.0
6.4
6.5
6.6
9.5
Cp - Fm
Cp - Fp
17.2
-304.8
17.4
--
20.5
--
21.8
--
18.0
--
7.0
--
Graphical abstract
12
Page 12 of 21

13
Page 13 of 21

Dmitriy S. Fadeev
8(383) 3307864 dsf@nioch.nsc.ru
Igor P. Chuikov
8(383) 3307864 chip@nioch.nsc.ru
Victor I. Mamatyuk
8(383) 3306960 vim@nioch.nsc.ru
N. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry,
Siberian Branch of the Russian Academy of Sciences
Lavrentiev Avenue 9, 630090 Novosibirsk, Russian Federation
Fax: +7 (383) 3309752. E-mail: benzol@nioch.nsc.ru
Page 14 of 21

Page 15 of 21

Page 16 of 21

Page 17 of 21

Page 18 of 21

Page 19 of 21

Page 20 of 21

Page 21 of 21