The stepped reaction of decafluorobiphenyl with thiophenol studied by in situ 19F NMR spectroscopy

Article abstract of DOI:10.1016/j.jfluchem.2013.07.013

Kinetic studies on the reaction of decafluorobiphenyl and di- and tetra(phenylthio)-substituted perfluorobiphenyl with thiophenol in the presence of triethylamine were performed by in situ ¹⁹F NMR spectroscopy to determine rate constants. In we

Full text of DOI:10.1016/j.jfluchem.2013.07.013

Journal of Fluorine Chemistry 156 (2013) 314–321
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
The stepped reaction of decafluorobiphenyl with thiophenol studied
19
by in situ F NMR spectroscopy
Christian Langner a,b, Jochen Meier-Haack a, Brigitte Voit a,b, Hartmut Komber a,
*
a Leibniz-Institut fu¨r Polymerforschung Dresden e.V., Hohe Strasse 6, 01069 Dresden, Germany
b Organische Chemie der Polymere, Technische Universita¨t Dresden, 01062 Dresden, Germany
A R T I C L E I N F O
A B S T R A C T
Article history:
Kinetic studies on the reaction of decafluorobiphenyl and di- and tetra(phenylthio)-substituted
perfluorobiphenyl with thiophenol in the presence of triethylamine were performed by in situ 19F NMR
spectroscopy to determine rate constants. In well-defined consecutive reactions, the even-numbered
phenylthio derivatives are main products whereas the odd-numbered derivatives are much more
reactive and were observed only in low concentrations. Very pronounced differences in rate constants
and reactivity were obtained.
Received 3 May 2013
Received in revised form 3 July 2013
Accepted 20 July 2013
Available online 31 July 2013
Keywords:
ß 2013 Elsevier B.V. All rights reserved.
Nucleophilic substitution
Decafluorobiphenyl
Kinetics
19F NMR
Thiophenol
1. Introduction
persulfurated aromatic compounds came into focus recently.
Gingras et al. [16] reviewed the development of persulfurated
Nucleophilic substitution with thiolates on perfluorinated
aromatics offers routes to tailored molecules, networks and
polymers [1–4]. The reaction is promoted by electron withdrawing
substituents, such as fluorine substituents. Aromatics bearing only
one electron withdrawing moiety show only a moderate reactivity
with thiolates due to weak activation for nucleophilic substitution
under base-mediated conditions [5]. There is noticeable effort to
develop transition metal catalysts that enable the formation of
sulfur-aryl bonds from single-activated aromatics under mild
conditions [6–11]. However, also additional halides or –M
substituents in ortho-/para-position further polarize the carbon-
halide bond making highly or perfluorinated compounds interest-
ing starting materials. Brooke [12] reviewed the broad field of
reactions on perfluorinated aromatic and heteroaromatic com-
pounds. Thus, perfluorinated aromatic compounds are intrinsically
activated and highly reactive toward nucleophilic substitution and
should hence lead to higher sulfurated aromatic compounds.
Adams et al. [13] as well as Robota and Malichenko [14] reported
first on persulfurated arenes. Beck et al. showed that persulfurated
arenes could be obtained also without a catalyst by using a polar
solvent [15]. In the search for materials with tailored properties,
compounds showing the versatility of their application based on
interesting properties like liquid crystallinity, charge capacity,
magnetism, conductivity and complexation capability.
Perfluorinated arenes are for example used to synthesize
partially fluorinated polymers as well as persulfurated com-
pounds. Several groups reported on polymers derived from
oxygen-nucleophiles [17–20]. Kellmann et al. [1] and Masaki
et al. [2] were the first to present partially fluorinated poly(aryl
thioether)s. This is surprising because in terms of polymer
chemistry perhalogenated arenes are multifunctional monomers
and should lead also to branched and/or cross-linked products.
Besides application in polymerization reactions, the nucleophilic
fluorine substitution by thiolates plays also an important role in
the concept of thiol click chemistry since for perfluorinated arenes
the reactions result in high yields under benign conditions [21].
Basic studies on the reaction of thiolate anions as nucleophiles
with fluorinated arenes were reported in the pioneering works of
Peach et al. in the seventies [22] after first studies of Robson et al.
who also reported the formation of perfluoropoly(phenylene
sulfide) as by-product [23]. For the reaction of hexafluorobenzene
or decafluorobiphenyl with an excess of thiolate anion (e.g.
methylthiolate, ethylthiolate or thiophenolate) tetra-substituted
(hexafluorobenzene) or hexa-substituted products (decafluorobi-
phenyl) were isolated as highest substituted reaction products
beside mono- and di-substituted compound [22a]. Interestingly
* Corresponding author. Fax: +49 351 4658565.
E-mail address: komber@ipfdd.de (H. Komber).
0022-1139/$ – see front matter ß 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jfluchem.2013.07.013

C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
315
the substitution pattern of the tetra-substituted benzene is not
symmetric (1,2,4,5 position) but asymmetric (1,2,3,5 position). In
the case of sixfold substitution of decafluorobiphenyl, again an
asymmetric substitution (2,20,4,40,5,50 position) on each ring was
proposed. Contrary to the reaction of hexafluorobenzene with the
aliphatic thiolates, the reaction with a large excess of thiopheno-
late gave presumably the fully substituted benzene. In all reactions
reported in [22] only the major products were isolated and
characterized. However, in order to allow better control in
reactions involving thiophenols and perfluorinated arenes a deeper
understanding of the course of reaction is necessary, which
includes the evaluation and characterization of occurring products
reacted with trimethyl(phenylthio)silane (TMPTS) in the presence
of equimolar amounts of potassium fluoride (Scheme 1). The use of
trimethylsilyl derivatives of nucleophiles has several advantages
over the use of free nucleophiles as for example reported by
Kricheldorf [24].
Compounds 2, 4, 6 and 8 were successfully isolated as main
products from the respective reaction mixture (see Section 4) by
column chromatography. Products with an odd number of
substituents (3, 5) were formed only in small amounts. For
example, the crude reaction product from the synthesis of 4 was a
mixture of 30 mol% 2, 0.5 mol% 3, 62.5 mol% 4, 3.5 mol% 5 and
3.5 mol% 6 as determined by 19F NMR spectroscopy. The synthesis
of the octa-substituted 8 required a higher reaction temperature.
Besides 8, also small amounts of 9 with only one fluorine atom
could be identified in the 19F NMR spectrum. However, its position
could not be determined unambiguously. The 19F NMR spectra of
the even-numbered thiophenyl-substituted fluorobiphenyls 2, 4, 6
and 8 are depicted in Fig. 1. All NMR data including 1H and 13C are
summarized in Supporting information.
The low-content compounds 3 and 5 were not isolated but the
19F chemical shifts and the coupling pattern of the three remaining
fluorine atoms are in accordance with a 3,4-disubstitution of the
disubstituted moiety (see Supporting information). Scheme 2
depicts possible products of the nucleophilic substitution of the
threefold substituted compound 3 based on preferred ortho-/para-
substitution to –SPh. Reaction of the thiophenolate with the higher
fluorinated ring of 3 resulting in 4a seems to be at first glance most
likely due to the expected lower electron density of this benzene
ring. However, only compound 4d (=4) could be identified by NMR
spectroscopy. Although traces of the other compounds cannot be
ruled out, the 2,4,5-substitution pattern is found in all isolated
derivatives with three phenylthio substituents on one ring.
This course of reaction may be explained by a strong activation
of the position para to the phenylthio moiety as proposed by Peach
et al. [22]. This strong activation was confirmed by relative rates of
the nucleophilic aromatic substitution of non-activated aryl
halides by thiolate ions as reported by Tiecco et al. [5]. A low
activation was also observed for the ortho and meta position. Here,
and their reactivity in
a multistep reaction. Such detailed
information is of general interest for applying this reaction in
polymer synthesis because side reactions can result in irreversible
structural defects on the polymer backbone, e.g. in grafting or
crosslinking reactions.
In this study, the conversion of decafluorobiphenyl (DFBP) with
thiophenol is investigated by in situ 19F NMR spectroscopy using
N-methyl-2-pyrrolidone (NMP) as solvent and triethylamine (TEA)
as base. Time-conversion plots were evaluated in order to obtain
rate constants. The application of 19F NMR as in situ technique here
has several advantages. Besides the high NMR sensitivity of the
fluorine nucleus, the chemical shifts of fluorine signals possess
high sensitivity toward local magnetic fields allowing to distin-
guish also very similar compounds by individual signals.
Di-, tetra-, hexa- and octa(phenylthio)-substituted perfluor-
obiphenyls were synthesized and fully characterized as reference
compounds in order to assign the signals observed in the in situ
NMR spectroscopic investigations.
2. Results and discussion
2.1. Synthesis of phenylthio-derivatives of decafluorobiphenyl
In order to assign the 19F NMR signals observed in the course of
reaction, model compounds with different degree of substitution
were prepared but not with focus on optimized yields. DFBP is
Scheme 1. Main products in the preparation of phenylthio derivatives of decafluorobiphenyl. The reactions were carried out in NMP.

316
C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
(=4). This rationalizes the absence (or very low concentration) of
4a–4c as products of an ortho substitution. Probably, the sterical
shielding of reactive sites ortho to phenylthio-substituents as it
occurs in 4b and 4c additionally prefers the formation of 4d. The
need for a higher reaction temperature to obtain 8 points to a high
sterical shielding if six or more phenylthio substituents are
attached to the biphenyl. However, with our experiments we
cannot distinguish the contribution of each effect leading to this
specific substitution pattern.
2.2. Kinetic investigations
For the 19F NMR kinetic investigations thiophenol and TEA as
base were used because the in situ experiments require a
homogenous reaction system and the use of thiophenols is a
more realistic scenario for aimed-for polymerizations. N-methyl-
2-pyrrolidone as solvent meets all requirements for nucleophilic
substitutions (high permittivity, aprotic) and is often applied in
polymerization reactions. An excess of thiol and a 1:2 ratio of thiol
to TEA should ensure that over a broad range of fluorine
substitution the thiolate concentration can be considered as
quasi-stationary. Preliminary studies indicated that the reaction
rates are rather high at 30 8C, and thus the changes of the substance
amounts could barely be followed by 19F NMR spectroscopy. For
that reason, the kinetic investigations had been conducted at 0 8C.
In the first experiment, R1 (Scheme 3), DFBP was reacted with
thiophenol and the kinetics was followed for 60 min. Under these
experimental conditions, DFBP was very quickly converted and
could not be detected anymore after 2 min, as it became evident
from the first spectrum recorded. Therefore, reaction 2 (R2, Scheme
3) was started with 2 to achieve more defined starting conditions.
R2 was followed for 406 min. Then reaction 3 (R3) was conducted
with a doubled amount of TEA as compared to R2 in order to
evaluate the influence of excess base. Finally, reaction 4 (R4)
started from 4 to enable a direct kinetic evaluation of the
conversion of 4 to 6.
Fig. 1. 19F NMR spectra of model compounds (a) 2, (b) 4, (c) 6 and (d) 8 containing
8 mol% 9 (solvent CD2Cl2).
the detection of 3 with ortho-disubstitution proofs for 2 a higher
activation for the ortho position than for the meta position. The
third phenylthio substituent has now a strong activation effect of
its para fluorine promoting nucleophilic attack and resulting in 4d
Fig. 2 depicts selected 19F NMR spectra of R2 in order to
illustrate the development of the characteristic signals. The kinetic
behavior of R1 and R2 for the first 60 min is compared in Fig. 3.
Astonishingly, even at 0 8C no DFBP and no 1 are detectable in the
Scheme 2. Possible tetra(phenylthio)-substituted perfluorobiphenyls as products of the reaction of 3 with thiophenolate.

C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
317
Scheme 3. General scheme for kinetic studies. Reaction 1 (R1): R1–R4 = F (DFBP), i = 11, n = 22; reaction 2 (R2): R1 and R2 = SPh, R3 and R4 = F (2), i = 9, n = 18; reaction 3 (R3):
R1 and R2 = SPh, R3 and R4 = F (2), i = 9, n = 36; reaction 4 (R4): R1–R4 = SPh (4), i = 7, n = 14.
first spectrum representing the first 2 min of reaction time. This
means that both, the reaction of DFBP to 1 and its conversion to 2
are very fast. Although R1 seems to have only slight differences in
the time-conversion curve compared to R2 the fast DFBP
conversion to 2 influences the overall kinetics. Thus, it is not
justified to analyze R1 as a reaction starting with 2.
In all reactions, only the compounds with an even number of
phenylthio-substituents (2, 4, 6) were formed in significant
amounts. Despite 3 and 5 could be identified, their total content
did not exceed 2 mol% at any reaction time. Thus, the conversion of
DFBP and 2, respectively, with an excess of thiophenol appears as
pronounced step-wise reaction in which 2, 4 and 6 are less reactive
compounds compared to the reactive intermediates 1, 3 and 5.
There was no evidence in the NMR spectra, e.g. by signals with
unexpected 19F chemical shifts, that a stable Meisenheimer-like
complex was formed in the course of reaction.
the thiophenol/TEA ratio from 1:2 to 1:4 did not lead to a faster
conversion of 2 and most likely of DFBP under the applied
conditions. The deviation found for longer reaction times might be
caused by small differences in concentrations and hence cannot be
distinguished from experimental errors (dosing, mixing).
Generally, each step in the reaction sequence starting from
DFBP and 2, respectively, to 6 (and higher degrees of substitution)
is an aromatic nucleophilic substitution (SNAr) under base-
mediated conditions [5]. The reaction can occur as a two-step
process involving addition of the thiolate to the aromatic ring to
form an anionic species followed by elimination of the fluoride
anion or as a one-step process with nucleophilic attack and
nucleofuge elimination occurring simultaneously.
The reaction sequences 2 ! 3 ! 4 and 4 ! 5 ! 6 involve a slow
reaction (k23, k45; formation of 3 and 5 due to weak ortho
activation) and a fast reaction (k34, k56; formation of 4 and 6 due to
strong para activation). With k23(45) ꢀ k34(56), the concentrations
of 3 and 5 remain very low as observed and their formation is the
rate-determining step in the reaction sequences. Under the
assumptions that (i) all reactions are irreversible and (ii) the
intermediate steps involving 3 (and 5) are very fast so that their
concentrations are quasi-stationary at almost zero and the steady
state approximation can be applied, the reaction of 2 toward 4 (and
6) can be simplified as outlined in Eqs. (1a) and (1b).
Fig. 4 displays the development of the molar fractions with time
for R2 and R3. In R2 already after 60 min 2 is completely converted
and after 406 min the molar fraction of 6 reaches a value of 59%.
However, even after 5 days of reaction time at room temperature
no higher substitution patterns than 6 could be identified in the 19
F
NMR spectrum. Reaction 3, which was conducted under the same
conditions as R2 but with increased amount of base, runs almost
exactly the same way as R2 up to 100 min. Obviously, increasing
2 þ 2PhSꢁ ! 4 þ 2Fꢁ
4 þ 2PhSꢁ ! 6 þ 2Fꢁ
r ¼ k2½2ꢂ½PhSꢁꢂ2
(1a)
(1b)
(2)
[2] and [PhSꢁ] are the concentrations of the educts and k2 the
rate constant of the rate-determining ortho substitution. Further
assuming, (iii) that the equilibrium concentration [PhSꢁ] of the
thiolate remains constant because an excess of thiol and base was
applied, the rate equation (Eq. (2)) simplifies and an integrated rate
law of first order is obtained (Eq. (3)).
0
2ꢃt
½2ꢂt ¼ ½2ꢂ0 ꢃ eꢁk
(3)
Fig. 2. Selected 19F NMR spectra taken during reaction 2 showing two signal regions
with characteristic signals of 2, 4, and 6. Assignment of signals is given in Section 4.
Fig. 3. Comparison of molar fractions of 2, 4, and 6 for reactions 1 (R1) and 2 (R2)
determined for the first 60 min reaction time.

318
C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
Fig. 6. Development of the molar fractions of 4 and 6 in reaction 4 (R4) as obtained
from 19F NMR spectroscopy.
Fig. 4. The development of the molar fractions of educt and reaction products in R2
and R3 as obtained from 19F NMR spectroscopy. The total content of 3 and 5 does not
exceed 2 mol% at any reaction time.
Eq. (3) contains an apparent rate constant k02 ¼ k2 ꢃ ½PhSꢁꢂ2. ½2ꢂ
t
was calculated from the signal intensity of 2 at time t related
to the signal intensities of all detected species 2–6 (=[2]0). All
integrals were corrected by the number of fluorine atoms causing
these signals.
Thus, the consumption of 2 in both R2 and R3 was evaluated
with a rate law for a first-order reaction. The characteristic kinetic
plot ln(ct/c0) vs time (Fig. 5) gives straight lines with very good
coefficients of determination, R2, confirming the first-order
kinetics for consumption of 2. The k02 values at 0 8C were
determined from the slope to 1.59 ꢄ 10ꢁ3 sꢁ1 (R2) and
1.47 ꢄ 10ꢁ3 sꢁ1 (R3) in good agreement. However, it should be
kept in mind that k02 depends on the thiolate concentration in the
preceding thiol/thiolate equilibrium with TEA as base, i.e. it is an
apparent rate constant. In the same way, the conversion of 4 to 6
(R4) was followed. The experimental data can be interpreted as
described for R2 and R3 but based on the reaction given in Eq. (1b).
The development of the molar fractions is shown in Fig. 6 and the
first order kinetic plot in Fig. 7. The consumption of 4 is in good
agreement with a pseudo first order kinetics whereby the apparent
reaction rate k04 is 8.17 ꢄ 10ꢁ5 sꢁ1. Compound 5 was observed only
in traces as intermediate, i.e, assumption (ii) (concentration is
quasi-stationary at almost zero) is valid. Because 6 was not further
converted, this reaction rate can be applied for calculation of [6].
Both, the apparent rate constants k02 and k40 describe rate-
determining ortho (2ꢁ) substitution in the 2,5-phenylthio substi-
Fig. 7. Ln(c) vs time plot for the consumption of 4 in reaction 4 (R4). [4]t/[4]0
corresponds to the molar fraction of 4 as given in Fig. 6.
tution of a 4-thiophenyl-substituted moiety. However, the values
obtained under similar conditions with respect to temperature,
thiol and base concentration differ by a factor of about 20.
Obviously, a 2,4,5-phenylthio substitution (4) compared to a 4-
phenylthio substitution (2) drastically reduces the reactivity of the
second 4-thiophenyl substituted ring with respect to further
thiolate attack. This applies for the first ortho substitution
(formation of 5) but not for the consecutive reaction in 2-position
resulting in 6. This reaction is still fast as can be concluded from the
negligible concentration of 5. In the 2,4,5-triphenylthio substituted
moieties the positions of remaining fluorine atoms are hardly
activated or sterically shielded as can be concluded from absence
of 7 or 8 under experimental conditions.
3. Conclusion
We have shown that in the reaction of DFBP with phenylthiolate
the formation of even-numbered sulfuryl derivatives is strongly
favored over the formation of odd-numbered ones, which were
observed only in traces. The almost exclusively observed even-
numbered substitution pattern are 4,40 (2), 2,4,40,5 (4) and
2,20,4,40,5,50 (6). The reactions of DFBP, 2 and 4 with thiophenol
and TEA as base could be kinetically studied at 0 8C in NMP using in
situ 19F NMR spectroscopy. Whereas DFBP was converted very fast,
the consumption of 2 and 4 followed a pseudo-first order kinetics
with formation of 3 and 5, respectively, as rate-determining step.
Assuming that the concentrations of 3 and 5 are quasi-stationary at
almost zero, pseudo-first order rate constant of the reaction of 2 to
4 (1.59 ꢄ 10ꢁ3 sꢁ1 for R2) is about 20 times larger than that for the
reaction of 4 to 6 (8.17 ꢄ 10ꢁ5 sꢁ1) under applied experimental
conditions. Further substitution of the 2,4,5-triphenylthiolted
moieties of 6 was not observed. Thus, the efficient reaction also
Fig. 5. Ln(c) vs time plots for the consumption of 2 in reactions 2 (R2) and 3 (R3).
[2]t/[2]0 corresponds to the molar fraction of 2 as given in Fig. 4.

C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
319
at ambient temperature and the very pronounced differences in
reactivity in combination with a precise balance of monomers are
very promising for the synthesis of high molecular or defined
molecular weight poly(aryl thioether)s.
ꢁ134.9 (m, 4F, F-3), ꢁ139.7 (m, 4F, F-2). GC–MS (70 eV, m/z): 516.1
(12), 515.1 (28), 514.1 (100) [M]+, 474.1 (13), 77.1 (25), 51.1 (11).
Anal. calcd for C24H10F8S2: C, 56.03; H, 1.96; S, 12.47; found: C,
56.05; H, 2.16; S, 12.81.
4.2.2. 2,20,30,5,50,6,60-Heptafluoro-3,4,40-tris(phenylthio)biphenyl (3)
3
19F NMR (470 MHz, CD2Cl2)
d
ꢁ129.7 (dd, JFF = 23.1 Hz,
4. Experimental
5JFF = 13.9 Hz, 1F, F-30), ꢁ130.6 (m, 1F, F-20). Other signals are
overlapped by main compound. Signals of the compound were
observed as traces in a chromatographic fraction. Numbering
corresponds to 4 in Fig. 8.
4.1. General remarks
Decafluorobiphenyl (DFBP; 99%, ABCR GmbH) was recrystal-
lized from ethanol/isopropanol (1/1, v/v) and dried in vacuum at
30 8C for 16 h. Dichloromethane (DCM), ethyl acetate, thiophenol
(all Acros), diethyl ether, 1,1,1,3,3,3-hexamethyldisilazane
(HMDS), n-hexane, silical gel (0.063–0.2 mm, all Merck) were
used as received. Potassium fluoride (Fluka) was dried in vacuum
at 150 8C for 7 h and stored in a desiccator prior use. N-methyl-2-
pyrrolidone (NMP; Merck) and triethylamine (TEA; Sigma–
Aldrich) were dried by distillation from CaH2. TMTPS was prepared
following a previously reported procedure [7].
4.2.3. 20,3,30,50,6,60-Hexafluoro-2,4,40,5-tetrakis(phenylthio)biphenyl
(4)
Light yellow solid, mp 140–141 8C; 1H NMR (500 MHz, CD2Cl2)
d
7.40 (m, 2H, H-6 at C6F4), 7.35–7.30 (3H, H-7 and H-8 at C6F4),
7.30–7.20 (10H, H-6, H-60, H-7, H-70, H-8, H-80), 7.19–7.11 (3H, H-
700, H-800), 7.02 (m, 2H, H-600). 13C NMR (125 MHz, CD2Cl2)
d
159.68
1
4
1
00
0
(dd, JCF = 248 Hz, JCF2 = 3.2 Hz, C-3 ), 156.79 (dd, JCF = 250 Hz,
4
1
0
00
JCF3 = 3.2 Hz, C-2 ), 147.16 (m, JCF = 247 Hz, C-2), 144.36 (m,
1JCF = 250 Hz, C-3), 134.86 and 134.69 (two d, 4JCF = 2.3, C-5, C-50),
134.03 (d, 4JCF = 1.6 Hz, C-500), 132.91 (C-5 at C6F4), 131.44 (C-6 at
C6F4), 130.96 (d, JCF = 20.7 Hz, C-40), 129.93, 129.76, 129.67, 129.65,
129.60, 129.58, and 129.49 (C-6, C-60, C-600, C-7, C-7 at C6F4, C-70, C-
700), 129.39 (d, JCF = 19.8 Hz, C-30), 128.43 (C-8 at C6F4), 127.72,
127.65, and 127.58 (C-8, C-80, C-800), 126.08 (dd, JCF = 23.0 Hz,
JCF = 1.6 Hz, C-200), 120.89 (d, J = 21.4 Hz, C-10), 115.76 (t,
The NMR measurements were carried out on a Bruker Avance III
spectrometer (Bruker Biospin, Germany) at 500.13 MHz (1H),
125.76 MHz (13C) and 470.59 MHz (19F). Solvent signals served as
internal chemical shift reference for 1H (CDCl3: 7.26 ppm; CD2Cl2:
5.31 ppm) and 13C (CDCl3: 77.0 ppm; CD2Cl2: 53.7 ppm). For 19F
NMR spectra, C6F6 (ꢁ163.0 ppm) was used as external chemical
shift reference. ATR-FTIR-spectra were recorded on a Bruker
Tensor 27 (aperture 6 mm). Mass spectra were recorded on an
Agilent Technologies 6890N GC/5973N MSD device with electron
impact (2) or on a Bruker Esquire with Ion Trap detector and ESI
source (4, 6, 8).
2
2JCF = 20.0 Hz, C-4), 113.59 (t, JCF = 18.6 Hz, C-1). 19F NMR
5
00
0
(470 MHz, CD2Cl2)
d
ꢁ95.5 (d, JFF2 = 15.3 Hz, 1F, F-3 ), ꢁ103.6
5
(dt, 5JFF3 = 15.3 Hz (d), JFF2 = 4.3 Hz (t), 1F, F-20), ꢁ135.5 (m, 2F, F-
3), ꢁ139.9 (m, 2F, F-2). MS (25 V, m/z): 859.7 (9), 802.3 (7), 733.1
(9) [M+K]+, 717.1 (13) [M+Na]+, 712.2 (86) [M+NH4]+, 695.2 (100)
[M+H]+, 419.3 (14). Anal. calcd for C36H20F6S4: C, 62.23; H, 2.90; S,
18.46; found: C, 62.54; H, 3.14; S, 18.86.
00
4.2. Synthesis of thiophenyl-substituted fluorobiphenyls
Exemplary, the synthesis of 2,20,3,30,5,50,6,60-octafluoro-4,40-
bis(phenylthio)biphenyl (2) is described. 1 mmol (0.3341 g) of
DFBP and 1 mmol (0.0581 g) of potassium fluoride are weighed in a
three-neck flask, equipped with a gas-inlet, gas-outlet and a
septum. Under argon atmosphere DFBP is dissolved in 5 ml NMP at
23 8C. The reaction is started by injection of 2 mmol (0.1824 g)
TMPTS to the stirred solution. The reaction is stopped by
precipitating the reaction product in 75 ml of 0.01 M HCl. The
solid raw product is isolated by filtration, washed with water and
methanol and finally dried at 50 8C in vacuum for 16 h. The
parameters for the preparation of the other compounds are given
in Table 1.
4.2.4. 20,3,50,6,60-Pentafluoro-2,30,4,40,5-pentakis
(phenylthio)biphenyl (5)
19F NMR (470 MHz, CD2Cl2)
d
ꢁ95.0 (d, 5JFF2 = 15.5 Hz, 1F, F-3
0
00
5
0
of
difluoro
ring),
ꢁ103.0
(dt,
JFF3 = 15.5 Hz
(d),
5
5
00
0
00
JFF2 ꢅ JFF2 ꢅ 4 Hz (t), 1F, F-2 of difluoro ring), ꢁ103.1 (dt,
5
4
5
00
00
00
JFF3 = 13.5 Hz (d), JFF ꢅ JFF2 ꢅ 4 Hz (t), 1F, F-2 of trifluoro ring),
2
5
ꢁ130.1 (dd, JFF = 24.0 Hz, JFF = 13.5 Hz, 1F, F-30 of trifluoro ring),
3
4
5
ꢁ131.0 (dt, JFF = 24.0 Hz (d), JFF ꢅ JFF ꢅ 4 Hz (t), 1F, F-20 of
trifluoro ring). Signals of the compound were observed as traces in
a chromatographic fraction. Numbering adopted from 6 in Fig. 8.
4.2.5. 3,30,6,60-Tetrafluoro-,2,20,4,40,5,50-hexakis(phenylthio)biphenyl
The pure products were obtained by column chromatography
on silica gel using the following eluents: 2, 4: ethyl acetate/n-
hexane (1/9, v/v); 6: ethyl acetate/n-hexane (2/8, v/v) and 8: ethyl
acetate/n-hexane (3/7, v/v).
(6)
Yellow solid, mp 45–46 8C; 1H NMR (500 MHz, CD2Cl2)
d 7.23–
7.10 (24H, H-6, H-60, H-7, H-70, H-700, H-8, H-80, H-800), 7.01 (m, 2H,
H-600). 13C NMR (125 MHz, CD2Cl2)
d
159.63 (dd, JCF = 249 Hz,
1
4.2.1. 2,20,3,30,5,50,6,60-Octafluoro-4,40-bis(phenylthio)biphenyl (2)
White solid, mp 105–106 8C (lit: 93–94 [22a]); 1H NMR
4
1
JCF2 = 2.8 Hz, C-3 ), 156.44 (dd, JCF = 248 Hz, 4JCF3 = 3.5 Hz, C-2 ),
00
0
0
00
4
4
135.02 (d, JCF = 1.9 Hz, C-5, C-50), 134.25 (d, JCF = 1.8 Hz, C-500),
129.55 (d, JCF = 20 Hz, C-40), 129.47, 129.45 (2C), 129.44, 129.24,
and 129.22 (C-6, C-60, C-600, C-7, C-70, C-700), 128.90 (d,
2JCF = 20.2 Hz, C-30), 127.43, 127.34, and 127.25 (C-8, C-80, C-800),
(500 MHz, CD2Cl2)
d 7.45 (m, 4H, H-6), 7.38–7.30 (m, 6H, H-7,
1
H-8). 13C NMR (125 MHz, CD2Cl2)
d 147.37 (dd, JCF = 248 Hz,
1
JCF = 13 Hz, C-3), 144.60 (dd, JCF = 255 Hz, JCF = 16.7 Hz, C-2),
2
2
126.93 (d, JCF = 21.3 Hz, C-10), 125.56 (dd, JCF = 22.9 Hz,
132.55 (C-5), 131.43 (C-6), 129.83 (C-7), 128.64 (C-8), 117.43 (t,
2JCF = 20.1 Hz, C-4), 107.83 (m, C-1). 19F NMR (470 MHz, CD2Cl2)
d
3JCF = 1.5 Hz, C-200). 19F NMR (470 MHz, CD2Cl2)
d
ꢁ95.15 (d,
5
5
00
0
0
00
JFF2 = 15.4 Hz, 2F, F-3 ), ꢁ102.43 (d, JFF3 = 15.4 Hz, 2F, F-2 ). MS
Table 1
(10 V, m/z): 897.2 (19) [M+Na]+, 892.3 (100) [M+NH4]+, 875.2 (10)
[M+H]+. Anal. calcd for C48H30F4S6: C, 65.88; H, 3.46; S, 21.98;
found: C, 65.90; H, 3.43; S, 23.52.
Parameters for the synthesis of compounds 2, 4, 6 and 8.
Compound
KF (g)/(mmol)
TMPTS (g)/(mmol)
T (8C)
t (h)
2
4
6
8
0.0581/2
0.1162/4
0.1743/6
0.2905/11
0.1824/2
0.3648/4
0.5472/6
0.9120/11
23
23
23
80
1
2
6
8
4.2.6. 30,60-Difluoro-2,20,3,4,40,5,50,6-octakis(phenylthio)biphenyl (8)
Yellow solid (contains 8 mol%/8.6 wt% 9); 1H NMR (500 MHz,
CD2Cl2)
d d 159.60 (dd,
7.2–6.8. 13C NMR (125 MHz, CD2Cl2)

320
C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
Fig. 8. Chemical formulas of the di- (2), tetra- (4), hexa- (6) and octa-phenylthio-substituted (8) fluorobiphenyls with atom numbering used for NMR signal assignment.
1
4
4
1
00
000
0
JCF = 249 Hz, JCF2 = 2.7 Hz, C-3 ), 156.44 (dd, JCF = 246 Hz,
retransferred to the NMR spectrometer and the measurement was
started after 1 min.
0
000
00
JCF3 = 3.0 Hz, C-2 ), 149.96 (C-4 ), 148.00 (C-3 ), 145.11 (br,
J
CF < 1 Hz, C-100), 143.16 (C-2000), 138.17 (C-5 of PhS at 4000), 138.00
and 136.87 (C-50, C-500 of PhS at 3000, 2000), 135.31 (d, 4JCF = 1.8 Hz, C-5,
Reaction 1 (R1): 20 mg (5.986 ꢄ 10ꢁ5 mol) DFBP, 72.5 mg
(6.585 ꢄ 10ꢁ4 mol) thiophenol, 0.185 ml (1.317 ꢄ 10ꢁ3 mol)
TEA. Stoichiometry: 1:11:22.
4
C-50 of PhS at 40, 30), 134.50 (d, JCF = 2.5 Hz, C-500 of PhS at C-200),
133.89 (dd, JCF = 21.6 Hz, JCF = 1.6 Hz, C-10), 129.38, 129.36,
2
3
129.33, 129.28, 129.24, 128.73, 128.56, 128.46, 128.17, and
Reaction
2
(R2): 20 mg (3.889 ꢄ 10ꢁ5 mol) 2, 38.6 mg
2
128.09 (C-6, C-60, C-600, C-7, C-70, C-700), 128.53 (d, JCF = 20.5 Hz,
(3.5 ꢄ 10ꢁ4 mol) thiophenol, 0.1 ml (7.175 ꢄ 10ꢁ4 mol) TEA.
2
C-30), 127.39 (d, JCF = 20.5 Hz, C-40), 127.18, 127.08, 126.63 (2 C),
Stoichiometry: 1:9:18.
Reaction
126.74, 126.48, and 126.44 (2 C) (C-8, C-80, C-800), 124.96 (dd,
3
(R3): 20 mg (3.889 ꢄ 10ꢁ5 mol) 2, 38.6 mg
2JCF = 22.5 Hz, JCF = 3.0 Hz, 200). 19F NMR (470 MHz, CD2Cl2)
d
(3.5 ꢄ 10ꢁ4 mol) thiophenol, 0.194 ml (1.4 ꢄ 10ꢁ3 mol) TEA.
3
5
ꢁ94.79 ppm (d, JFF = 15.8 Hz, 1F, F-300), ꢁ101.04 ppm (d,
5JFF = 15.8 Hz, 1F, F-20). MS (10 V, m/z): 1163.1 (5) [913C+NH4]+,
1072.2 (43) [M+NH4]+, 1055.2 (100) [M+H]+. Anal. calcd for 92%
Stoichiometry: 1:9:36.
Reaction
4
(R4): 25 mg (3.60 ꢄ 10ꢁ5 mol) 4, 27.8 mg
(2.52 ꢄ 10ꢁ4 mol) thiophenol, 0.07 ml (5.04 ꢄ 10ꢁ4 mol) TEA.
C
60H42F2S8 and 8% C66H47F1S9: C, 68.34; H, 3.84; S, 24.37; found: C,
Stoichiometry: 1:7:14.
68.37; H, 3.78; S, 25.93.
Characteristic signals of the impurity 9: 13C NMR (125 MHz,
CD2Cl2)
ꢁ92.93 ppm (s).
d
160.29 (d, 1JCF = 252 Hz, CF), 19F NMR (470 MHz, CD2Cl2):
Acknowledgments
d
We thank A. Korwitz for assistance in the kinetic measure-
ments, R. Schulze for elemental analysis, I. Bauer (TU Dresden) for
MS measurements and C. Vogel for proofreading the manuscript.
4.3. In situ 19F NMR kinetic investigations
In the in situ measurements, 24 scans were accumulated for
each data point of the conversion-time curve applying 908 19F
pulses. 1H broadband decoupling and a pulse delay of 4 s were
applied. The acquisition time was 1.04 s. Thus, the measuring time
for each spectrum adds up to 2 min. The composition of the
reaction mixture at different reaction times was determined from
signal integrals.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jfluchem.2013.
07.013.
For the in situ experiments, the educt (DFBP, 2 or 4) is weighted
in the NMR tube and dissolved in 0.3 ml NMP. Then triethylamine
was added and filled up with NMP to a total volume of 0.5 ml. The
NMR tube was thermostated in the NMR spectrometer to 0 8C and
the reaction is started by injection of the thiophenol solution in
NMP (total volume 0.2 ml) pre-cooled to ꢁ5 8C to compensate
the warming during the addition. Subsequently, the tube was
References
[1] R. Kellman, J.C. McPheeters, D.J. Gerbi, R.F. Williams, Polym. Prepr. Am. Chem. Soc.
Div. Polym. Chem. 22 (1981) 383–384.
[2] S. Masaki, N. Sato, A. Nishichi, S. Yamazaki, K. Kimura, J. Appl. Polym. Sci. 108
(2008) 498–503.
[3] (a) C.R. Becer, K. Babiuch, D. Pilz, S. Hornig, T. Heinze, M. Gottschaldt, U.S.
Schubert, Macromolecules 42 (2009) 2387–2394;

C. Langner et al. / Journal of Fluorine Chemistry 156 (2013) 314–321
321
(b) C.R. Becer, K. Kokado, C. Weber, A. Can, Y. Chujo, U.S. Schubert, J. Polym. Sci.
Part A: Polym. Chem. 48 (2010) 1278–1286.
[16] M. Gingras, J.-M. Raimundo, Y.M. Chabre, Angew. Chem. Int. Ed. Engl. 45 (2006)
1686–1712.
[4] M. Riedel, J. Stadermann, H. Komber, F. Simon, B. Voit, Eur. Polym. J. 47 (2011)
675–684.
[17] (a) F. Mercer, T. Goodman, J. Wojtowicz, D.J.D. Duff, J. Polym. Sci. Part A: Polym.
Chem. 30 (1992) 1767–1770;
[5] P. Cogillo, F. Maiolo, L. Testaferri, M. Tingoli, M. Tiecco, J. Org. Chem. 44 (1979)
2642–2646.
(b) A.A. Goodwin, F.W. Mercer, M.T. McKenzie, Macromolecules 30 (1997)
2767–2774.
[6] C.G. Bates, R.K. Gujadhur, D. Venkataraman, Org. Lett. 4 (2002) 2803–2806.
[7] Y.-C. Wong, T.T. Jayanth, C.-H. Cheng, Org. Lett. 8 (2006) 5613–5616.
[18] K. Miyatake, K. Oyaizu, E. Tsuchida, A.S. Hay, Macromolecules 34 (2001) 2065–
2071.
´
´
[8] M.A. Fernandez-Rodrıguez, Q. Shen, J.F. Hartwig, Chem. Eur. J. 12 (2006) 7782–
[19] Z. Lu, P. Shao, J. Li, J. Hua, J. Qin, A. Qin, C. Ye, Macromolecules 37 (2004)
7089–7096.
7796.
[9] E. Sperotto, G.P.M. van Klink, J.G. de Vries, G. van Koten, J. Org. Chem. 73 (2008)
5625–5638.
[20] J. Kerres, F. Scho¨nberger, A. Chromik, T. Ha¨ring, Q. Li, J.O. Jensen, C. Pan, P. Noye´,
N.J. Bjerrum, Fuel Cells 8 (2008) 175–187.
[10] H.-J. Xu, X.-Y. Zhao, Y. Fu, Y.-S. Feng, Synlett 19 (2008) 3063–3067.
[11] M.S. Kabir, M. Lorenz, M.L. van Linn, O.A. Namjoshi, S. Ara, J.M.J. Cook, J. Org. Chem.
75 (2010) 3626–3643.
[12] G.M. Brooke, J. Fluorine Chem. 86 (1997) 1–76.
[13] (a) R. Adams, W. Reifschneider, M.D. Nair, Croat. Chem. Acta 29 (1957)
277–281;
(b) R. Adams, A. Ferretti, J. Am. Chem. Soc. 81 (1959) 4927–4931.
[14] L.P. Robota, B.F. Malichenko, J. Org. Chem. USSR 12 (1976) 236.
[15] J.R. Beck, J.A. Yahner, J. Org. Chem. 43 (1978) 2048–2052.
[21] C.E. Hoyle, A.B. Lowe, C.N. Bowman, Chem. Soc. Rev. 39 (2010) 1355–1387.
[22] (a) K. Langille, M.E. Peach, J. Fluorine Chem. 1 (1972) 407–414;
(b) M.E. Peach, A.M. Smith, J. Fluorine Chem. 4 (1974) 341–344;
(c) M.E. Peach, A.M. Smith, J. Fluorine Chem. 4 (1974) 399–408.
[23] (a) P. Robson, M. Stacey, R. Stephens, J.C. Tatlow, J. Chem. Soc. (1960) 4754–4760;
(b) P. Robson, T.A. Smith, R. Stephens, J.C. Tatlow, J. Chem. Soc. (1963)
3692–3703.
[24] H.R. Kricheldorf, Polycondensation of silylated monomers, in: H.R. Kricheldorf
(Ed.), Silicon in Polymer Synthesis, Springer, Heidelberg, 1996, pp. 288–354.