The Reaction Mechanism Between Tetraarylammonium Salts and Hydroxide

Article abstract of DOI:10.1002/ejoc.202000435

The mechanism of the reaction between tetraaryl ammonium salts and hydroxide is studied experimentally for different N,N-diaryl carbazolium salts. The N,N-diarylcarbazolium salts are designed, synthesized, characterized, and reacted with hydroxide under different conditions. The products of the reactions were directly characterized or isolated when possible and, using different substituents, the reaction mechanisms were compared. An unexpected H/D exchange observed in these salts helped to discard the classical S_NAr mechanisms, supporting instead a radical mechanism initiated by a single-electron transfer from the hydroxide. By understanding the preferred reaction pathways, better quaternary ammonium salts can be designed to withstand aggressive alkaline environments, critical for many practical applications such as anion-exchange membrane fuel cells.

Full text of DOI:10.1002/ejoc.202000435

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: The Reaction Mechanism Between Tetraarylammonium Salts
and Hydroxide
Authors: Nansi Gjineci, Sinai Aharonovich, Sapir Willdorf-Cohen, Dario
R. Dekel, and Charles Diesendruck
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.202000435
Link to VoR: https://doi.org/10.1002/ejoc.202000435
01/2020

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
The Reaction Mechanism Between Tetraarylammonium Salts and
Hydroxide
Nansi Gjineci[a], Sinai Aharonovich[a], Sapir Willdorf-Cohen[b], Dario R. Dekel*[b][c], Charles E.
Diesendruck*[a][c]
[a]
N. Gjineci, Dr. S. Aharonovich, Prof. Dr. C.E. Diesendruck
Schulich Faculty of Chemistry
Technion - Israel Institute of Technology
Haifa 3200008, Israel
http://diesendrucklab.com/
[b]
[c]
S. Willdorf-Cohen, Prof. Dr. D.R. Dekel
The Wolfson Department of Chemical Engineering, Technion - Israel Institute of Technology
Haifa 3200003, Israel
https://dekel.technion.ac.il/
Prof. Dr. D.R. Dekel, Prof. Dr. C.E. Diesendruck
The Nancy & Stephan Grand Technion Energy Program (GTEP)
Technion - Israel Institute of Technology
Haifa 3200003, Israel
E-mails: charles@technion.ac.il; dario@technion.ac.il
Supporting Information for this article is given via a link at the end of the document.
Abstract: The mechanism of the reaction between tetraaryl
ammonium salts and hydroxide is studied experimentally for different
N,N-diaryl carbazolium salts. The N,N-diarylcarbazolium salts are
designed, synthesized, characterized, and reacted with hydroxide
under different conditions. The products of the reactions were directly
characterized or isolated when possible and, using different
substituents, the reaction mechanisms were compared. An
unexpected H/D exchange observed in these salts helped to discard
mechanism - Scheme 1a).[11] In this case, hydroxide acts as a
base, abstracting the β proton while a tertiary amine is formed as
a good leaving group (LG), neutralizing the molecules’ positive
charges, thus ceasing anion conductance. In the absence of β
hydrogens, the positive charge distributed at the carbons
connected to the nitrogen, leads to classical nucleophilic attack at
an α carbon, with the amine again acting as a good LG (SN2,
Scheme 1b), again, producing neutral molecules.[12,13]
the classical SNAr mechanisms, supporting instead
a
radical
Recently, some QAs have been produced that showed
remarkable stability in highly alkaline aqueous conditions even at
high temperatures, including substituted imidazolium salts,[14–16]
spirocyclic ammonium salts,[12,17–19] and others. Still, when tested
under strictly dry alkaline conditions, in which hydroxide
microsolvation is limited,[20] these ammonium salts decomposed
very rapidly by the expected mechanisms.[21] This is another
example in which solvation environment, and particularly water
molecules, can affect reaction mechanisms heterogeneously.[22–
mechanism initiated by a single-electron transfer from the hydroxide.
By understanding the preferred reaction pathways, better quaternary
ammonium salts can be designed to withstand aggressive alkaline
environments, critical for many practical applications such as anion-
exchange membrane fuel cells.
Introduction
24]
Therefore, new paradigms of QAs which may decompose
through different reaction mechanisms is of supreme importance,
as it opens new avenues for the development of more stable
cations.
Quaternary ammonium (QAs) salts are of foremost importance for
the chemical industry, as they are used in a very wide spectrum
of applications including surfactants, disinfectants, antistatic
agents, phase transfer catalysts, and more.[1–4] In recent years,
the synthesis of new QA salts has burst, given their potential use
in anion-exchange membrane fuel cells (AEMFCs).[5] AEMFC is a
promising technology for affordable and efficient energy
conversion but is currently limited by the chemical stability of the
cationic functional groups that decompose under fuel cell
operation.[6][7] Currently, no organic cation has proved to be stable
enough to withstand the harsh combined alkaline and low
hydration environment for the required lifetime of an AEMFC.[8]
Yet, among organic cations, QAs have shown the most promising
results.[9]
QAs are typically prepared by the alkylation of a tertiary amine,
and as a consequence, most QAs have at least one alkyl
substituent.[10] As a general rule, the presence of β-hydrogens
leads to fast degradation through Hofmann elimination (E2
Scheme 1. Typical decomposition mechanisms of quaternary ammonium salts
in the presence of hydroxide: E2 (a) and SN2 (b).
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
A
reasonable strategy to suppress both E2 and SN2
synthetic steps and significantly better yields.[28] Driven by
previous studies that were able to better simulate AEMFCs
conditions by creating non-aqueous alkaline and low hydration
conditions,[21,33,36] we expected that making such molecules and
testing them under such conditions would open new pathways to
developing more stable QAs, needed for durable AEMFCs.[7]
Importantly, by changing the mechanism of reaction with
hydroxide, different molecular optimization can be used to further
tackle the stability of transition states and intermediates, towards
the production of kinetically stable QAs which will be able to
survive the aggressive conditions present in AEMFC for the
lifetime of the fuel cell.
mechanisms is to change the α carbons hybridization from sp3 to
sp2.[25] However, the fact that the intermolecular arylation of
triarylamines has not yet been demonstrated[10,26] hampers the
testing of simple tetraphenyl ammonium (TPA) salts. TPA salts,
in the presence of hydroxide, could still decompose through
known decomposition mechanisms, but these typically require
harsher reaction conditions for the same nucleophile. Hydroxide
could still attack the electrophilic α carbon leading to SNAr
mechanism through addition-elimination. However, recently, this
pathway has been shown to be much more rarer than described
in organic chemistry textbooks (Scheme 2a).[27] A concerted SN2-
like mechanism is still possible, but compared to aliphatic SN2, it
is kinetically sluggish due to steric hindrance imposed by the
aromatic ring on the nucleophile. In addition, carbon-LG bonds
are shorter as a result of the higher s-character of the carbon
(Scheme 2b). Finally, TPA can decompose through β hydrogen
abstraction, producing a strained benzyne intermediate (Scheme
2c). While intermolecular arylation of triarylamines has not yet
been demonstrated, the intramolecular arylation using diazonium
salts[28,29] or C-F activation by sillyl cation[30] to produce N,N-
diarylcarbazolium salts has proven possible, generating QAs with
a similar general structure compared to TPA salts, given all the
nitrogen substituents are aromatic carbons.
Nansi Gjineci obtained her degree in
Chemical Engineering from National
Technical University of Athens, Greece. She
receives a M.Sc. after investigating the
effect of ionic liquids on the separation of
azeotropic mixtures under the supervision of
Professor Epaminondas Voutsas in 2015. In
October 2016, Nansi joined the Technion-
Israel Institute of Technology under the
supervision of Professor Charles
Diesendruck in Schulich Faculty of
Chemistry and Dario Dekel in Wolfson Department of Chemical
Engineering in a joint project as a PhD student. Nansi works on the
development and testing of novel functional groups and ionomers for
alkaline fuel cell applications.
HO
Ph3N
a
OH
b
Ph3N
NPh3
+
Sinai Aharonovich studied Chemical
a/b
H
Engineering at the Wolfson Department of
c
Chemical Engineering at the Technion -
c
OH
Israel Institute of Technology, graduating in
1997. After working as process engineer, he
joined the group of Professor Moris S.Eisen
at the Schulich Faculty of Chemistry, where
he got a Ph.D. in 2010. He then did a post-
NPh3
Ph3N
OH
Scheme 2. Possible decomposition mechanisms of TPA hydroxide: (a) addi-
tion-elimination, (b) concerted SN2 and (c) benzyne (c).
doc with Prof. Robert M. Waymouth at
Stanford University, working on
homogenous Ziegler-Natta catalysis. In
2012, he returned to Israel to serve as CSO in the development of
organocatalytic oxidation of sulfur and nitrogen oxides. Since 2016, he
returned to the Technion and joined the group of Prof. Diesendruck where
he works as a senior researcher, focusing on polymer chemistry, organic
and organometallic syntheses.
In his pioneering work, Nesmeyanov et al. noticed that the N,N
diphenyl carbazolium salt was quite stable and decomposed only
upon heating with 50% aqueous KOH.[29] In their study, it was
proposed that hydroxide preferably attacks the freely rotating
phenyl rings and not the carbazole ring, given they identified the
reaction products as N-phenyl carbazole, phenol and diphenyl
ether; yet, the prevailing reaction mechanism was not described.
A few years later, Hellwinkel,[31,32] in an attempt to make
pentacoordinated nitrogen species, added strong carbon
nucleophiles to a 9,9-spiro-biscarbazolium salt. In his manuscript,
decomposition through benzyne mechanism was suggested, but
given the analytical limitations at the time, this proposed
mechanism could not be supported by experimental evidence.
Our group has been interested in the synthesis[26,28] and
decomposition[21,33–35] of QAs in order to try to open new directions
in the development of stable cations for AEMFCs. We have
recently developed an improved procedure for the synthesis of
N,N-diaryl carbazolium salts, which allows for the synthesis of a
variety of derivatives with different substituents in very few
Sapir Willdorf-Cohen studied Chemical
Engineering at the Wolfson Department of
Chemical Engineering at the Technion -
Israel Institute of Technology, graduating in
2016. She then started graduate studies
working with Prof. Diesendruck at the
Schulich Faculty of Chemistry and Prof.
Dekel at the Wolfson School of Chemical
Engineering. Her research focuses on the
stability studies of anion exchange
membranes for fuel cells applications.
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
regioselectivity of the hydroxide attack, in order to understand the
dominating decomposition mechanism.
Dario R. Dekel received his BSc in Chemical
engineering in the National Technology
University in Argentina. He immigrated to
Israel in 1990, and joined the Technion –
Israel Institute of Technology, from where he
received his M.Sc. in Chemical Engineering,
Ph.D. in the field of Membrane Science, and
his M.B.A. After managing a power source
unit in Rafael Ltd., he co-found CellEra
Technologies Ltd. where he led the
development of the Anion-Exchange
Scheme 3. Synthesis of 9,9-di-p-tolylcarbazolium acetate.
Membrane Fuel Cell technology. He joined Technion in 2015, as an
Associate Professor in the department of Chemical Engineering. His
current research interests include anion conducting polymers, anion-
exchange membranes, and electrochemical devices based on them.
N, N di(p-tolyl) carbazolium salt was synthesized similarly to
its phenyl congener, following the synthetic strategy previously
developed by us, using 4-iodotoluene instead of iodobenzene
(Scheme 3).[28] The synthesis begins with the hydrogenation of
the 2,2-dinitrobiphenyl (1) using zinc under acidic conditions.
Then, a CuI catalyzed Ullman reaction of diamine 2 with 2
equivalents of 4-iodotoluene provides amine (3) in 35% isolated
yield. This Ullman reaction is regioselective and no additional
isomers were observed by NMR (remaining product is the mono-
arylated 2, which can be recycled in the next batch). The final step
is the formation and decomposition of the diazonium salt leading
to cyclization, which forms the desired carbazolium salt (4). The
QA salt in the acetate form was purified and single crystals were
successfully grown after anion exchange to hexafluorophosphate
(Figure 1).
Charles E. Diesendruck immigrated to Israel
in 1999 from Brazil. He carried out all his
studies in Chemistry at Ben-Gurion
University, receiving his Ph.D. working
under Prof. N. Gabriel Lemcoff in 2011. He
then carried out post-doctoral studies with
Prof. Jeffrey S. Moore at the University of
Illinois at Urbana-Champaign. Since 2014,
he is part of the Schulich Faculty of
Chemistry in the Technion – Israel Institute
of Technology, where he recently became
Associate Professor. His research interests include Polymer Chemistry,
Self-healing polymers, Mechanochemistry, Materials for Energy
Applications, and Physical-Organic Chemistry.
Results and Discussion
Our recently developed synthesis of N,N-diarylcarbazolium was
initially used to synthesize large quantities of N,N-
diphenylcarbazolium hexafluorophosphate,[28] which was tested
under the most aggressive media following the protocol
previously developed in our group, in which dry KOH is prepared
in molten 18-crown-6-ether (CE) by titration of potassium metal
by water.[21] After dissolution in dry DMSO, dry hydroxide solution
has in average ca. 0.1 water molecules per hydroxide and
therefore, acts as a superbase/supernucleophile.[37] In fact, we
have shown that by using dry hydroxide, the rate constant for SN2
and E2 reactions is increased by a few orders of magnitude.[21]
Unfortunately, the NMR spectrum of N,N-diphenylcarbazolium
hexafluorophosphate and CE/KOH (0.5 M) with mesitylene as an
internal standard in DMSO-d6 showed complete decomposition of
the QA happens almost immediately. The products were isolated
by extraction and as previously shown, the main decomposition
product observed was N-phenyl carbazole[29] (see the Supporting
Information). This result supports the hypothesis that hydroxide
attacks occur more rapidly at the freely rotating aryls and not at
the carbazole system. However, it does not allow identification of
the reaction mechanism, which can allow tuning design
parameters to improve the stability of these QAs, at least
kinetically. Therefore, we synthesized a new carbazolium salt
containing substituents in the phenyl rings, which would reveal the
Figure 1. X-ray structure of 9,9-di–p-tolylcarbazolium hexafluorophosphate.
In the solid state, the addition of para methyl groups did not
significantly change the structural features, with the C-N bond
distances being comparable to those of the N,N-
diphenylcarbazolium
hexafluorophosphate,[28]
and
still
significantly longer than those in common arylammoniums
salts.[38] The longest C-N bonds are to the carbon of the freely
rotating phenyl rings (1.516 Å), possibly indicating the cause of
the selectivity of the hydroxide attack. In the carbazole moiety, the
C-N distances are shorter (1.512 Å and 1.508 Å).
As with the N,N-diphenylcarbazolium salt, the addition of the
unsolvated hydroxide leads to rapid degradation. However, in this
case, through the identification of the isomeric phenol
decomposition products, the dominating SNAr mechanism could
be differentiated (Scheme 4). If the main mechanism is hydroxide
attack on the α carbon to the nitrogen, we expect to obtain only 9-
p-tolyl carbazole and para-cresol as products (Scheme 4a). If, on
the other hand, the hydroxide abstracts the β hydrogen, then the
decomposition would follow a benzyne formation, which leads to
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
three expected degradation products: 9-p-tolyl carbazole, and a
mixture of para- and meta-cresols (Scheme 4b).
As seen in Figure 3a good separation and clear peaks were
obtained, with meta cresol coming at ca. 7 min whereas para
cresol presents a retention time of ca. 8 min. The injected extract
(Figure 3b) showed a complete dominance of the para cresol as
a product (also identified by comparison to the NIST library),
together with a very small peak of the meta cresol.
The NMR spectrum of the reaction can be directly compared
to the spectra of the possible products, after their dissolution in a
similar solution (DMSO in the presence of 0.5 M CE/KOH). As
seen in Figure 2, the signals of 9-tolyl carbazole are quite easily
identified, but the resolution was not good enough to distinguish
between the cresol isomers.
Scheme 4. The two possible reaction pathways for 9,9-di-p-tolylcarbazolium
with hydroxide: (a) concerted SN2 and (b) benzyne mechanism.
Figure 3. Chromatogram of (a) mixture of para and meta cresol and (b)
CH2Cl2 extract from the reaction mixture after work-up.
The GC analysis clearly indicates that the predominant
mechanism seems, unexpectedly, to be a direct attack on the ipso
carbon by an SN2-like mechanism (or addition-elimination)
(Scheme 4a), both of which seemed unlikely for the reasons
described above. Additional experiments further lead us to
question these mechanisms. Given that the hydroxide reacts
selectively with the freely rotating phenyls, we decided to prepare
a spiro-biscarbazolium salt, without any such rings. The 9,9-spiro-
biscarbazolium acetate was synthesized following a similar
synthetic strategy (Scheme 5), but in this case, 2,2’-
diiodobiphenyl was used instead of iodobenzene, providing amine
5 in 20% yield after column chromatography. This arylation
reaction was quite challenging due to steric hindrance at the ortho
position of the iodine, as reflected in the 7 days (as opposed to
3.5 hours) needed for its completion.
Figure 2. 1H-NMR in DMSO-d6 of (a) decomposition product from 9,9-di-p-
tolylcarbazolium acetate after addition of CE/KOH, (b) 9-p-tolyl-carbazole, (c)
meta and (d) para cresol with CE/KOH, and (e) 9,9-di-p-tolylcarbazolium
acetate.The aromatic protons of 9-p-tolyl carbazole are shown in the green
box. The aromatic peaks of p-cresol and m-cresol are shown in the black box.
Therefore, we decided to analyze the reaction mixture and the
two isomeric phenols by GC-MS. To that end, the content of the
NMR tube after the reaction was acidified by concentrated sulfuric
acid which neutralized the strong base and turns the phenoxides
into phenols. A few mL of water were then added and the organic
decomposition products were extracted using CH2Cl2, which was
dried and injected into the more sensitive GC-MS (Figure 3). We
also injected a mixture of para and meta cresols as references,
given their EI fingerprint in standard conditions can be misleading.
Scheme 5. Synthesis of 9,9-spiro-biscarbazolium acetate.
The final cyclization was obtained through the formation of the
diazonium salt in acetic acid followed by addition of urea and
heating to 40 oC for 1 h. The spiro-biscarbazolium 6 was obtained
as a pure compound and single crystals could be obtained after a
salt metathesis with ammonium hexafluorophosphate (Figure 4).
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
The C-N bonds in 6 are slightly shorter than in 4, with values
ranging from 1.498 Å to 1.506 Å. Interestingly, 6-acetate was
soluble in water, which motivated us to test its stability in aqueous
(D2O) KOH first. As expected, the molecule seemed perfectly
stable in these conditions. To accelerate the decomposition
reaction, we increased the temperature to 60oC. Surprisingly, the
most upfield aromatic hydrogen signals, which correspond to the
ortho hydrogens started to disappear, while the other hydrogens
remained almost unchanged (Figure 5). After 7 days, this
hydrogen signal completely disappeared, as did its splitting of the
meta carbon, but surprisingly, the compound remains pure – no
additional peaks were observed.
not occur if the hydroxide preferably attacks the ipso carbon, as
this would lead to irreversible C-N bond scission. We, therefore,
went to the literature to look for an alternative mechanism that
would agree with all the experimental observations.
An interesting possibility described in many literature examples,
is the reactivity of hydroxide as a reducing agent. Based on
previous fundamental studies carried out by Sawyer,[42] Schreiner
et al. have shown that hydroxide, especially with low water
microsolvation, is capable of triggering radical halogenations[43]
through single-electron transfer (SET) reactions. SET to
carbazolium molecules would provide similar products as a
nucleophilic attack on the ipso carbon, following a mechanism
similar to a Birch reduction (Scheme 6).[44]
Scheme 6. Proposed Birch-like single-electron transfer (SET) reaction
between hydroxide and 9,9-di-p-tolylcarbazolium.
To further support this SET reaction mechanism, we carried
out cyclic voltammetry studies on N,N-diphenylcarbazolium
acetate in acetonitrile using Ag/AgNO3 as the reference electrode.
The cyclic voltammetry was performed in the absence of
supporting electrolyte at a sweep rate of 10 mV/s (Figure 6) and
the potential scale was fixed based on literature data, against
Normal Hydrogen Electrode (NHE).[45] The diphenyl carbazolium
shows a clear reduction peak at -0.882 V. However, combined
with the +0.92 V of the OH-/HO● redox couple potential,[42] the
SET shown here is 1.80 V endergonic. Therefore, a discrete SET
reaction between diphenyl carbazolium with OH- is improbable.
Actually, considering the many examples of radical organic
transformations using OH- for SET, it is quite interesting that most
of them, when looking only at the initial steps, are quite
inaccessible. The reason these reactions do occur was nicely
discussed by Sawyer:[42] “The redox potentials for the electron
acceptors that react with HO- are such that a pure outersphere
SET step would be endergonic. Hence, the observed net
reactions must be driven by coupled chemical reactions,
particularly bond formation by the HO● to the electrophilic atom
of the acceptor molecule, that accompany a single-electron shift.”
Figure 4. X-ray structure of 9,9-spiro-biscarbazolium hexafluorophosphate (6).
Figure 5. 1H-NMR in D2O of (a) pure 9,9-spiro-biscarbazolium acetate, (b)
after addition of 0.5 M KOH, (c) after 16 h at room temperature, (d) after
additional 24 h at 60oC, (e) after 4 days at 60oC, and (f) after 7 days at 60oC.
The crude reaction mixture was sent to mass spectrometry
and returned showing a mass gain of 4 mass units (see the
Supporting Information), indicating that the ortho hydrogens
exchanged with deuterium, without any decomposition of the
molecule. H/D exchange of acidic hydrogens under harsh alkaline
conditions has been mentioned in the literature for other QAs. It
is known that the benzylic hydrogen position in benzyl
trimethylammonium rapidly is deuterated when the salts are
exposed to alkaline environment in MeOD or D2O[39–41]. This same
effect has been more recently observed also in imidazolium
cations.[16] While the hydroxide is fully hydrated in this case, this
H/D exchange, which typically has a high energy barrier, should
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
Conclusion
Using our new approach for the synthesis of N,N-diaryl carbazoli-
um salts, we were able to study the reaction mechanism between
tetraarylammonium salts and hydroxide. Given their possible use
in ionic polymers for AEMFCs, we studied their reactions with
hydroxide under different severe alkaline conditions. Identification
of N-aryl carbazole as the major decomposition product confirmed
that degradation occurs through the attack on the freely rotating
phenyl rings, while the carbazole remains intact. The classical
nucleophilic reaction mechanisms were put forward and, in order
to decide on the reaction pathway, tetraarylammonium salts
having para methyl substituents were used to look at the reaction
regioselectivity, leading to observations which did not fit with the
classical SNAr mechanisms. Therefore, inspired by previous work
with dry hydroxide, a radical decomposition mechanism was
proposed. However, cyclic voltammetry measurements revealed
that the reduction potential of the N,N diphenylcarbazolium
acetate, in combination with that of hydroxide, could not support
Figure 6. Cyclic voltammogram of N,N-diphenylcarbazolium acetate in
CH3CN.
Therefore, based on our results and the literature, we propose
that a plausible reaction mechanism between the carbazolium salt
and hydroxide consists of an inner sphere single electron transfer
in the ion pair followed by radical coupling.[46] The zwitterionic
sigma complex intermediate undergoes spontaneous elimination
of the carbazole group, restoring aromaticity. The formation of the
bond between the radicals provides the driving force to make the
reaction exergonic (Scheme 7).
a
discrete, outer sphere single-electron transfer (SET)
mechanism. Therefore, inspired by Sawyer’s seminar work,[42] we
proposed the inner sphere coupling of SET with a polar group
transfer reaction as a major operative mechanism. The free
radical pair generated from the SET reaction is followed by the
elimination of the leaving group and the energy of the formed C-
O
bond provides sufficient driving force to overcome the
unfavorable electron transfer reaction. This SET mechanism is
not seen in other QAs used in AEMs, which typically show a good
electrochemical range and decompose through classical SN2 and
E2 mechanisms. Still, with this knowledge at hand, new
tetraarylammonium salts can be judiciously designed to address
this radical mechanism reaction pathway, in order to increase
their stability to match the harsh conditions of ultra-dry hydroxide,
providing a new pathway for the development of highly stable
ionic polymers for AEMFCs and other applications.
Experimental Section
Materials and General Methods
Commercially available materials were used as received unless noted.
Xylene used for synthesis was dried before use by filtration through an
alumina column. The complex of 18-crown-6/KOH was prepared as
previously described,[33] stored and handled in a glovebox. For thin-layer
chromatography, silica gel GF254 plates were used and visualized with a
UV lamp (254nm). Silica gel 60 (230-400 mesh) from Merck was used in
column chromatography. NMR spectra were recorded in a Bruker Avance
III 400 MHz or Bruker Avance 300 MHz spectrometer. The chemical shifts
are referenced to signal at δ 0.00 (TMS) or partially undeuterated solvent
peaks. The coupling constants (J) are reported in Hz. Peak multiplicity is
indicated as follows: s (singlet), d (doublet), t (triplet), q (quartet), dd
(double doublet), dt (double triplet), br (broad) and m (multiplet). High-
resolution mass spectrometry was performed in a Waters LCT Premier
Mass Spectrometer (ESI) and a Bruker maxis impact with APCI solid probe.
Gas chromatography data were obtained using an Agilent 6850 GC
equipped with an Agilent 5973 MSD working under standard conditions
and an Agilent HP5-MS (30*0.25*0.25) column. X-ray diffraction data was
collected on a Nonius Kappa CCD diffractometer with monochromated Mo
Kα radiation (ꢀ=0.71073 Å). Accurate cell parameters were acquired by
analyzing the amount of indicated reflections. The diffraction image was
processed by software Mercury. The structure was solved by direct
methods (SHELXS-97) and refined against F2 by full-matrix least squares
methods (SHELXL-97). All non-hydrogen atoms were refined
Sceme 7. SET reaction pathway including all the reaction potentials and
estimated free energies.[42,47]
Naturally, given the reaction is done in DMSO, there is also a
considerable amount of dimsyl anions, which are slightly better
reducing agents than dry OH- (by ca. 0.3 mV).[48] Therefore, this
full mechanism may also be occurring with dimsyl counter anions
instead of OH-, but the innersphere process is still a requirement
in this case.
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
anisotropically. Hydrogen atoms were refined isotropically on calculated
positions using a riding model. The isotropic displacement parameters Uiso
were constrained to 1.5 times the Ueq value of their pivot atoms for terminal
sp3 carbon atoms, and 1.2 times for all other carbon atoms.
(182 mg, 20% yield).1H NMR (400 MHz, CDCl3) δ 7.91 (s, 2H), 7.63 – 7.54
(m, 1H), 7.47 (dt, J = 9.3, 4.0 Hz, 3H), 7.13 (t, J = 40.4 Hz, 6H), 6.72 – 6.63
(m, 1H), 6.57 (dd, J = 7.6, 1.2 Hz, 1H), 6.33 (d, J = 7.9 Hz, 1H), 6.18 (t, J
= 7.4 Hz, 1H). 13C NMR (101 MHz, CDCl3) δ 143.26, 138.50, 136.15,
132.22, 130.39, 129.78, 129.09, 128.83, 128.72, 125.65, 123.66, 123.22,
120.16, 119.61, 118.21, 115.69, 110.22, 109.70. HRMS (APCI Solid Probe
MS+): [M+]calculated: 335.1543. [M+]found: 335.1516. m.p. 156-1580C.[49]
Experimental Procedures
9,9-spirobiscarbazolium acetate (6):
N,N-di-p-tolyl-[1,1'-biphenyl]-2,2'-diamine (3)
The compound was prepared following the procedure described in the
preparation of compound 4. The product was obtained as an off-white solid
(100mg, 49% yield). 1H NMR (400 MHz, DMSO) δ 8.53 – 8.46 (m, 4H),
7.87 (dd, J = 11.1, 4.1 Hz, 4H), 7.59 – 7.48 (m, 4H), 7.31 (d, J = 8.3 Hz,
4H), 1.72 (s, 3H).13C NMR (101 MHz, DMSO) δ 159.76, 147.90, 132.71,
131.56 (d, J = 16.4 Hz), 123.88, 118.57, 23.66. HRMS (TOF MS ES+):
In an oven-dried Schlenk flask (50 mL), amine 2 (1.00 g, 5.42 mmol) and
4-iodotoluene (2.367 g, 10.84 mmol) were dissolved in xylenes (15 mL)
and the flask deoxygenated using 3 freeze-pump-thaw cycles. The flask
was backfilled with argon. KOtBu (1.33 g, 11.92mmol) was then added and
the mixture left for 10 min stirring at room temperature. Then, CuI (0.201
g, 1.084 mmol) and 1,10-phenanthroline (0.195 g, 1.084 mmol) were
added and the mixture was stirred for 3.5 h at 125oC. The mixture was
allowed to cool to room temperature and was filtered. The solids were
washed with chloroform and then dissolved in 25% NH4OH (30 mL). The
aqueous phase was then extracted with CHCl3. The organic phases were
combined, and the solvents evaporated. The concentrated paste was re-
dissolved in CHCl3 and extracted with NH4OH until no blue color was
observed. To separate the unreacted amine 2, HCl (10 wt% in water) was
added to the organic phase and extracted. The organic phase was then
washed with saturated NaHCO3, dried over Na2SO4, filtered and
evaporated. At this stage, the mixture could be used for the next step
without further purification. Alternatively, the pure amine 3 could be
isolated by column chromatography using CHCl3 as an eluent (Rf = 0.85).
In this case, it is obtained as a brown solid (695 mg, 35% yield).1H NMR
(300 MHz, CDCl3) δ 7.39 (dd, J = 20.4, 7.7 Hz, 3H), 7.29 – 7.19 (m, 2H),
6.94 – 6.86 (m, 5H), 6.73 (d, J = 7.6 Hz, 4H), 6.58 (t, J = 7.5 Hz, 1H), 6.45
(d, J = 7.9 Hz, 1H), 3.26 (s, 2H), 2.26 (s, 6H). 13C NMR (75 MHz, CDCl3)
δ 147.16, 145.90 (2C), 144.08, 137.24 (2C), 133.13, 131.34 (2C), 130.96
(2C), 130.31, 129.75, 129.16, 128.47, 126.32, 125.49, 122.76, 122.76 ,
118.80, 115.99, 115.43, 21.19 (2C). HRMS (APCI Solid Probe MS+)
[M+]calculated: 365.2012 . [M+]found: 365.2021. m.p.123-125oC.
[M]calculated
(decomp.)[49]
: 318.1283. [M]found:
318.1295. m.p. (I- form) 285-288oC
Acknowledgments
This work was partially funded by the Nancy & Stephan Grand
Technion Energy Program (GTEP); by the Ministry of Science,
Technology & Space of Israel through the Israel-Germany
Batteries Collaboration Call 2017 [German grant No. 2675], by the
Israel Science Foundation (ISF) [grant No. 1481/17], by the
Ministry of National Infrastructure, Energy and Water Resources
of Israel [grant No. 3-15564], and by the Planning & Budgeting
Committee / ISRAEL Council for Higher Education (CHE) and
Fuel Choice Initiative (Prime Minister Office of Israel), within the
framework of “Israel National Research Center for
Electrochemical Propulsion (INREP)”.
9,9-di–p–tolyl carbazolium hexafluorophosphate(4)
Conflict of interest
Amine 3 (0.3g, 0.824 mmol) was dissolved in glacial acetic acid (4 mL) in
an Erlenmeyer (25 mL). The solution was cooled to 0oC in an ice bath. The
frozen acetic acid is crushed with a metallic spatula before NaNO2 (0.3 g,
4.35 mmol) in water (0.4 mL) was added, and the slurry mechanically
stirred for 20 min. Urea (0.23 g, 3.8 mmol) was added next, and the mixture
stirred for 1 h at 40oC. The solvents were evaporated, and the residue
dissolved in CHCl3. The non-soluble part was filtered and washed with
CHCl3. The filtrate was concentrated in vacuo and separated using
extraction with water and ether. Water and acetic acid were removed in
vacuo by freeze-drying. The acetate produced is highly hygroscopic and
difficult to be quantified. To quantify the reaction yield and obtain single
crystals, 4 was converted to the hexafluorophosphate by addition to a
saturated solution of NH4PF6. 4 hexafluorophosphate precipitates and the
solids are filtered and washed with water and ether. The red-brown solids
are then dried under vacuum (220 mg, 54% yield).1H NMR (300 MHz, D2O)
δ 8.04 (dd, J = 7.8, 0.9 Hz, 2H), 7.71 – 7.62 (m, 2H), 7.60 – 7.42 (m, 4H),
7.08 (dt, J = 6.7, 5.8 Hz, 8H), 2.21 (s, 6H). 13C NMR (101 MHz, D2O) δ
181.31, 149.98, 143.93, 142.09, 132.06, 130.75, 129.99, 123.00, 121.84,
121.25, 23.18, 19.90. HRMS (TOF MS ES+): [M]calculated: 348.1752.
[M]found: 348.1793. m.p. 195-200 oC (decomp.)
The authors declare no conflict of interest.
Keywords: fuel cell • quaternary ammonium • nucleophilic
substitution • single electron transfer • tetraaryl ammonium
[1]
[2]
[3]
[4]
R. S. Boethling, D. G. Lynch, in Detergents (Ed.: N.T. de Oude),
Springer Berlin Heidelberg, Berlin, Heidelberg, 1992, pp. 145–177.
M. Tischer, G. Pradel, K. Ohlsen, U. Holzgrabe, ChemMedChem
2012, 7, 22–31.
Y. Kugimoto, A. Wakabayashi, T. Dobashi, O. Ohnishi, T. K. Doi, S.
Kurokawa, Prog. Org. Coatings 2016, 92, 80–84.
M. Lissel, D. Feldman, M. Nir, M. Rabinovitz, Tetrahedron Lett.
1989, 30, 1683–1686.
[5]
[6]
G. Merle, M. Wessling, K. Nijmeijer, J. Memb. Sci. 2011, 377, 1–35.
J. R. Varcoe, P. Atanassov, D. R. Dekel, A. M. Herring, M. A.
Hickner, P. A. Kohl, A. R. Kucernak, W. E. Mustain, K. Nijmeijer, K.
Scott, et al., Energy Environ. Sci. 2014, 7, 3135–3191.
D. R. Dekel, I. G. Rasin, S. Brandon, J. Power Sources 2019, 420,
118–123.
[7]
[8]
[9]
2'-(9-carbazolyl)-[1,1'-biphenyl]-2-amine (5):
5 was prepared using the same procedure used for compound 4. In this
case, 2,2’-diiodobiphenyl (500 mg, 1.23 mmol) was reacted with 2 (225
mg, 1.22 mmol). The reaction was performed under argon, at 125 oC for 7
days. Following workup, 5 could be isolated by column chromatography
using a mixture of CHCl3/Hexane (7:3) as eluent (Rf = 0.8). Yellow solid
C. E. Diesendruck, D. R. Dekel, Curr. Opin. Electrochem. 2018, DOI
10.1016/j.coelec.2018.03.019.
S. Gottesfeld, D. R. Dekel, M. Page, C. Bae, Y. Yan, P. Zelenay, Y.
S. Kim, J. Power Sources 2018, 375, 170–184.
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
[10]
[11]
S. Arava, C. E. Diesendruck, Synth. 2017, 49, 3535–3545.
A. W. von Hofmann, Ann. der Chemie und Pharm. 1851, 78, 253–
286.
Memb. Sci. 2012, 399–400, 49–59.
[42]
[43]
D. T. Sawyer, J. L. Roberts, Acc. Chem. Res. 1988, 21, 469–476.
P. R. Schreiner, O. Lauenstein, I. V Kolomitsyn, S. Nadi, A. A.
Fokin, Angew. Chem. Int. Ed. 1998, 37, 1895–1897.
A. J. Birch, S. M. Mukherji, J. Chem. Soc. 1949, 2531–2536.
V. V. Pavlishchuk, A. W. Addison, Inorganica Chim. Acta 2000, 298,
97–102.
[12]
[13]
[14]
M. G. Marino, K. D. Kreuer, ChemSusChem 2015, 8, 513–523.
Z. Sun, J. Pan, J. Guo, F. Yan, Adv. Sci. 2018, 1800065.
K. M. Hugar, H. A. Kostalik, G. W. Coates, J. Am. Chem. Soc. 2015,
137, 8730–8737.
[44]
[45]
[15]
[16]
O. D. Thomas, K. J. W. Y. Soo, T. J. Peckham, M. P. Kulkarni, S.
Holdcroft, Polym. Chem. 2011, 2, 1641–1643.
[46]
[47]
[48]
[49]
R. A. Rossi, A. B. Pierini, A. B. Peñéñory, Chem. Rev. 2003, 103,
71–167.
J. Fan, S. Willdorf-Cohen, E. M. Schibli, Z. Paula, W. Li, T. J. G.
Skalski, A. T. Sergeenko, A. Hohenadel, B. J. Frisken, E. Magliocca,
et al., Nat. Commun. 2019, 10, 2306.
M. Lucarini, P. Pedrielli, G. F. Pedulli, S. Cabiddu, C. Fattuoni, J.
Org. Chem. 1996, 61, 9259–9263.
H. Lund, H. Svith, S. U. Pedersen, K. Daasbjerg, Electrochim. Acta
2005, 51, 655–664.
[17]
[18]
H.-S. Dang, P. Jannasch, Macromolecules 2015, 48, 5742–5751.
J. Xue, X. Liu, J. Zhang, Y. Yin, M. D. Guiver, J. Memb. Sci. 2020,
595, DOI 10.1016/j.memsci.2019.117507.
D. Hellwinkel, H. Seifert, Chem. Ber. 1972, 105, 880–906.
[19]
X. Chu, L. Liu, Y. Huang, M. D. Guiver, N. Li, J. Memb. Sci. 2019,
578, 239–250.
[20]
[21]
S. Pusara, S. Srebnik, D. R. Dekel, J. Phys. Chem. C 2018, 122, 10.
D. R. Dekel, S. Willdorf, U. Ash, M. Amar, S. Pusara, S. Dhara, S.
Srebnik, C. E. Diesendruck, J. Power Sources 2018, 375, 351–360.
J. Staszak-Jirkovský, R. Subbaraman, D. Strmcnik, K. L. Harrison,
C. E. Diesendruck, R. Assary, O. Frank, L. Kobr, G. K. H. Wiberg, B.
Genorio, et al., ACS Catal. 2015, 5, 6600–6607.
R. Otto, J. Brox, S. Trippel, M. Stei, T. Best, R. Wester, Nat. Chem.
2012, 4, 534–538.
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
M. Gazitúa, R. A. Tapia, R. Contreras, P. R. Campodónico, New J.
Chem. 2018, 42, 260–264.
M. N. Glukhovtsev, A. Pross, L. Radom, J. Am. Chem. Soc. 1994,
116, 5961–5962.
M. Hirsch, S. Dhara, C. E. Diesendruck, Org. Lett. 2016, 18, 980–
983.
E. E. Kwan, Y. Zeng, H. A. Besser, E. N. Jacobsen, Nat. Chem.
2018, 1.
S. Aharonovich, N. Gjineci, D. R. Dekel, C. E. Diesendruck, Synlett
2018, DOI 10.1055/s-0036-1591848.
A. N. Nesmeyanov, T. P. Tolstaia, A. V Grib, Dokl. Akad. Nauk
SSSR 1963, 153, 608–611.
M. Lu, O. Allemann, J. Xu, A. Linden, K. K. Baldridge, J. S. Siegel,
Org. Chem. Front. 2019, 6, 2640–2646.
[31]
[32]
[33]
D. H. H. Seifert, V. D. Hellwinkel, 1972, 54, 29–54.
D. Hellwinkel, H. Seifert, Chem. Ber. 1972, 105, 880–906.
D. R. Dekel, M. Amar, S. Willdorf, M. Kosa, S. Dhara, C. E.
Diesendruck, Chem. Mater. 2017, 29, 4425–4431.
A. Zhegur, N. Gjineci, S. Willdorf-Cohen, A. N. Mondal, C. E.
Diesendruck, N. Gavish, D. R. Dekel, Cite This ACS Appl. Polym.
Mater 2020, 2020, 360–367.
[34]
[35]
[36]
S. Srebnik, S. Pusara, D. R. Dekel, J. Phys. Chem. C 2019, 123, 2.
S. Willdorf-Cohen, A. N. Mondal, D. R. Dekel, C. E. Diesendruck, J.
Mater. Chem. A 2018, 6, 22234–22239.
[37]
[38]
[39]
D. Marx, A. Chandra, M. E. Tuckerman, Chem. Rev. 2010, 110,
2174–2216.
F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen, R.
Taylor, J. Chem. Soc. Perkin Trans. 2 1987, S1.
K. M. Hugar, W. You, G. W. Coates, ACS Energy Lett. 2019, 4,
1681–1686.
[40]
[41]
A. D. Mohanty, C. Bae, J. Mater. Chem. A 2014, 2, 17314–17320.
J. B. Edson, C. S. Macomber, B. S. Pivovar, J. M. Boncella, J.
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000435
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
Reaction Mechanism
The reaction between different N,N-diarylcarbazolium salts and hydroxide are probed to understand the reaction mechanism. The
regioselectivity of the reaction and an unexpected H/D exchange, indicate the reaction does not follow expected SNAr, but follows a
radical pathway.
9
This article is protected by copyright. All rights reserved.