An Effective Synthesis of N, N -Diphenyl Carbazolium Salts

Article abstract of DOI:10.1055/s-0036-1591848

Tetraaryl ammonium salts are a synthetic challenge, since there is no general method for the arylation of triaryl amines. Contrary to other quaternary ammonium salts, tetraaryl ammonium salts should be very chemically stable. The ipso carbons are not very

Full text of DOI:10.1055/s-0036-1591848

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2018, 29, A–E
letter
A
en
S. Aharonovich et al.
Letter
Synlett
An Effective Synthesis of N,N-Diphenyl Carbazolium Salts
Sinai Aharonovicha
Nansi Gjinecia
Dario R. Dekelb
Charles E. Diesendruck*a
0
0
0
0
0-
0
0
1
5-
5
7
6
1-
3
6
6
a Schulich Faculty of Chemistry, Technion – Israel Institute
of Technology, Technion City, Haifa, 32000, Israel
charles@technion.ac.il
b Wolfson Department of Chemical Engineering, Technion
– Israel Institute of Technology, Technion City, Haifa,
32000, Israel
Published as part of the Special Section 9th EuCheMS
Organic Division Young Investigator Workshop
Received: 29.09.2017
Accepted after revision: 07.11.2017
Published online: 20.12.2017
DOI: 10.1055/s-0036-1591848; Art ID: st-2017-b0722-l
Abstract Tetraaryl ammonium salts are a synthetic challenge, since
there is no general method for the arylation of triaryl amines. Contrary
to other quaternary ammonium salts, tetraaryl ammonium salts should
be very chemically stable. The ipso carbons are not very electrophilic,
since the positive charge is distributed throughout the pi systems and
they have no acidic β hydrogens. Here we demonstrate a simple ap-
proach to N,N-diphenyl carbazolium salts using only three synthetic
steps, allowing for an easy production of these salts in large amounts
and in a relatively short time. In addition, we study the Cu(I) catalyzed
multi-arylation of 2,2’-diaminobiphenyl, focusing on the regioselectivi-
ty of each step. Finally, we characterize, for the first time, the solid state
structure of a tetraaryl ammonium salt.
Charles E. Diesendruck studied chemistry at the Ben-Gurion Univer-
sity of the Negev (Israel). After working for a few years in industry, he
returned to Ben-Gurion University to get a M.Sc. and a Ph.D. in chemis-
try under the supervision of Prof. N. Gabriel Lemcoff, working on the
synthesis and study of NHC–metal complexes. He then moved to the
Beckman Institute for Advanced Science and Technology at the Univer-
sity of Illinois at Urbana-Champaign as a postdoctoral research fellow,
working with Prof. Jeffrey S. Moore on the study of mechanochemistry
and development of self-healing materials. In 2014, he moved back to
Israel to start his independent career in the Schulich Faculty of Chemis-
try at the Technion – Israel Institute of Technology. His research focuses
on the study of the relation between chemistry and mechanics in poly-
meric materials. In addition, his group develops new bond-forming re-
actions that are relevant in the preparation of functional polymers,
including mechanoresponsive materials and membranes for fuel cells.
Key words Quaternary ammonium salt, diazonium salt, selective ary-
lation, copper catalysis, Ullmann reaction
Quaternary ammonium salts are important industrial
chemicals with many uses such as antimicrobials, surfac-
tants, antistatic agents, and more.1,2 In addition, these oni-
um salts play a significant role in phase-transfer catalysis3
and anion-exchange membranes.4–6 Notably, in most cases,
the ammonium substituents are acyclic aliphatic chains.
The addition of phenyl substituents, given their electron-
withdrawing effects, leads to a faster decomposition of
these salts, typically through a Hofmann mechanism.7 Yet,
tetraaryl ammonium salts could present interesting struc-
tural features and improved chemical stability. Contrary to
other quaternary ammonium salts, all carbons connected to
the nitrogen are sp2, and therefore, cannot be attacked by
nucleophiles in SN2 reactions.8,9 In addition, while
tetraarylammonium salts may have β-hydrogens (ortho),
‘Hofmann’ would produce a benzyne, a decomposition
mechanism significantly higher in energy compared to a
regular E2 mechanism.7 Therefore, while these compounds
should be thermodynamically unstable, they could present
high kinetic stability.
Nonetheless, tetraaryl ammonium salts are simple mol-
ecules which are very challenging to make.10 Classical ary-
lation reactions like Buchwald–Hartwig11,12 and Ullmann13
do not work with tertiary amines. Chan–Lam14 coupling
was shown to work with sp2 nitrogens in heterocycles to
make cyclic sp2 ammonium salts such as imidazolium and
pyridinium salts, but not with simple acyclic tertiary
amines.15 We have recently demonstrated the arylation of
tertiary amines to quaternary ammonium salts using ben-
zyne chemistry,16 yet, our methodology does not work for
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

B
S. Aharonovich et al.
Letter
Synlett
O2N
H2N
Ph2N
PF6
Ph
Ph
[H]
PhI
1. NaNO2
N
2. urea
Cu(I)
3. NH4PF6
NO2
NH2
NH2
A
B
C
Scheme 1 Synthesis of N,N-diphenylcarbazolium hexafluorophosphate
triaryl amines, as these are completely planar and not nu-
cleophilic.17,18 Still, two examples of syntheses of tetraaryl
ammonium salts can be found from the 1960s19 and 1970s,
based on the intramolecular cyclization of diazonium
salts.20 However, these are multistep and tedious syntheses,
and lacking sufficient physical characterization of the final
compounds. In another more recent example, the peculiar
decomposition of hexatritium benzene was employed to
form highly electrophilic singlet phenyl cations, which re-
acted in situ with triphenylamine.21 Yet, all these methods
did not allow for proper characterization of the tetraaryl
ammonium products.
Here, we describe a new synthesis of N,N-diphenylcar-
bazolium salts carried out in only three steps based on the
intramolecular cyclization of diazonium salts with tertiary
amines, without need for chromatographic purification,
and with an overall yield of 12% (Scheme 1), compared to
less than 1.8% (seven steps) in previous methods.20 While
our synthesis still does not address the challenge of devel-
oping a general method for the arylation of triaryl amines10
(i.e., some of the steps present low yields), the simplicity of
the method allows for the synthesis of N,N-diarylcarbazoli-
um salts in large amounts and in only a short period of time.
In our synthesis design, we focused on Cu(I) Ullmann-
type catalysis, a well-established amine arylation method,
which utilizes low cost and available reagents. While main-
ly described for arylation of secondary amines, there are
also examples of arylation of primary amines in the pres-
ence of a strong base and a chelating ligand.22 We envi-
sioned that after the first arylation, the monoaryl D
(Scheme 2), having a primary and a secondary amine
group, may undergo subsequent arylation of either the ary-
lated nitrogen or the –NH2. Since the secondary amine is
more acidic, and assuming the deprotonation is a pre-rate-
determining step, it is conceivable that arylation of the
same nitrogen would be kinetically preferred, leading to
the diaryl B, which is the precursor of the carbazolium C.
The synthesis starts from 2,2′-dinitrobiphenyl, which is
hydrogenated under acidic conditions to the 2,2′-diamino-
biphenyl (A) using zinc. While A is commercially available,
it is significantly more expensive than the dinitro precur-
sor. Compound A can be purified by extractions with acidic
water and neutralization back to the organic soluble prod-
uct with ammonium hydroxide, which also complexes the
zinc salts in the aqueous phase, providing A in 94% isolated
yield in only a few hours, without the need of chromatogra-
phy purification.
When A is arylated by 2.1 equivalents of iodobenzene in
the presence of potassium t-butoxide and CuI/o-phenanth-
roline (20% mol), the main products are the desired com-
pound B (60% yield) and the triarylated compound F (25%),
whereas the 2,2′-di-N-phenylaminobiphenyl (E) is not ob-
served at all under our conditions. This result corroborates
our hypothesis: The secondary amine is indeed more reac-
tive, and the triarylated byproduct F is obtained by subse-
quent arylation of the desired product B.
Ph2N
If the arylation is performed with higher amounts of the
ligand, and with only 1 equivalent of iodobenzene (Scheme
2), the main product is the monoarylated compound D,
which is isolated in 14% yield, mainly due to low conver-
sion. Compound B is also seen by NMR analysis, but in low
amounts. Interestingly, subsequent arylation of the same
amine group dramatically reduced the basicity of the un-
substituted, adjacent, amine. Thus, while A is fully proton-
ated at pH = 1 (10 wt% H3PO4), D is not, and can be fully pro-
tonanted only at pH < 0 (25 wt% HCl), whereas B and F are
not protonated even by 32 wt% HCl. We took advantage of
this phenomenon to separate A from the mixture of B and
F, and from D, avoiding separation by chromatography.
To increase the yield of B, we attempted to block one of
the amine groups toward arylation by double silylation
(Scheme 3) with trimethylsilyl (TMS), a labile protecting
group. Interestingly, the precipitated LiCl, formed after the
first silylation of A in xylenes, redissolved during the sec-
B
NH2
PhHN
H2N
H2N
PhI (2.1 equiv)
E
F
Cu(I) (0.2 equiv)
ligand (0.2 equiv)
NHPh
Ph2N
NHPh
NH2
D
A
PhI (1 equiv)
Cu(I) (0.2 equiv)
ligand (0.3 equiv)
NHPh
H2N
D
NHPh
Scheme 2 Ullmann-type sequential arylation of 2,2′-diaminobiphenyl
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

C
S. Aharonovich et al.
Letter
Synlett
ond lithiation, presumably due to formation of soluble lithi-
um anilide LiCl aggregates. After the reaction workup, all of
the TMS moieties are hydrolyzed to silanol. However, unex-
pectedly, the major product in high conversion was the
monoaryl D, even when 3 equivalents of PhI were used.
The identification of the mono (D), di (B vs. E) and tri-
arylated (F) products is challenging by H NMR spectros-
1
copy, since all the signals are simply ‘aromatic’, but very
simple using 13C NMR spectroscopy (Figure 1). The number
of aromatic carbon signals (16 in D and B, 10 for E, 20 for F),
in addition to their chemical shift and intensity clearly
pointed to each of the different arylation products.
H2N
H2N
BuLi (1.1 equiv) BuLi (1 equiv)
TMSCl (1 equiv) TMSCl (1 equiv) PhI (3 equiv)
For the carbazolium synthesis, we used the mixture of B
and F without separation, since F is expected to convert
into a nitrosamine, while B produces the desired quaterna-
ry ammonium salt, which is easily separated from the other
products by extraction with water. This final step involves
two reactions which are done in one pot. First, diazotiza-
tion with sodium nitrite in acetic acid at low temperatures
produced the diazonium compound. Subsequent addition
of urea and heating lead to fast decomposition of the diazo-
nium compound to provide the final compound C as an ace-
tate salt. The acetate salt was isolated, but was extremely
hygroscopic, complicating characterization. The yellowish
solid would become a liquid in a matter of minutes when
exposed to air. Aqueous salt metathesis of the acetate with
ammonium hexafluorophosphate provided, after filtration
and drying, pure compound C (hexafluorophosphate) as a
light-yellow powder in 22% isolated yield, again, without
need for chromatographic purification.
Both carbazolium salts (acetate and hexafluorophos-
phate) were characterized in solution by 1H NMR spectros-
copy and showed a doublet of two protons above 8 ppm,
which belongs to the carbazole part. For the bis-2,2′-bi-
phenylylene ammonium cation it was claimed that this sig-
nal represents the 6,6′-carbazole hydrogens;20 however, no
similar phenyl protons were that deshielded in our spectra.
In order to shed more light on this, additional structural
characterization was required.
xylenes
Cu(I) (0.2 equiv)
ligand (0.2 equiv)
NHPh
NH2
D
A
Scheme 3 Arylation of the silylated 2,2′-diaminobiphenyl to the
monoaryl D
This result, along with the lack of the tetraaryl product,
can be rationalized by evaluation of the effects of the N-sub-
stituents on the arylation reaction. Both Ph and TMS groups
have comparable σp substituent constants (0.0923 and
–
0.07,24 respectively), indicating similar ability to stabilize a
negative charge conjugated to the aromatic ring. Their
Charton steric constants (3.00 and 4.20, respectively)25
show that the TMS substituent is bulkier than the Ph.
Hence, the formation of the copper(I) amido complex and
its interaction with the aryl iodide,26,27 both likely pre-rate
determining steps,28 would be more hindered in the TMS
case. In addition, the introduction of three electron-with-
drawing substituents would reduce the nucleophilicity of
the monosubstituted nitrogen, which would hamper the
copper–nitrogen bond-formation step.29 This hypothesis is
supported experimentally by the lack of formation of the
tetraaryl from the triaryl, and by the observed reduction of
basicity as function of substitution (F < B < D < A, vide
supra). Therefore, it may be advantageous in this case to use
a less bulky and more electron-donating protecting group.
Figure 1 13C NMR spectra of the mono- (D), di- (B) and tri- (F) arylated
compounds
Figure 2 X-ray structure of N,N-diphenylcarbazolium hexafluorophos-
phate C (hydrogen atoms and disorder at the anion omitted for clarity)
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

D
S. Aharonovich et al.
Letter
Synlett
Single crystals of the N,N-diphenylcarbazolium hexa-
fluorophosphate (C) were grown by slow evaporation of
chloroform–acetonitrile (2:1 v/v) solutions. The X-ray
structure is shown in Figure 2. To the best of our knowl-
edge, this is the first ever X-ray structure of a tetraaryl
ammonium salt. The carbazolium presents unusual,
extremely long C–N bonds. Whereas these bonds in ammo-
nium salts are typically around 1.465 Å,30 the C–N bonds in
this structure are 1.508, 1.514, and 1.517 Å; some of the
longest C–N bonds reported without adding any steric
hindrance. To put this result into perspective, the C–N bond
distances in N-phenylcarbazole31 are 1.394–1.427 Å,
whereas the bond distances in C are comparable to those
observed by our group in the Ph3NMe+ cation.16 Chloroform
co-crystallizes with the salt, and there is an intermolecular
hydrogen bond between its hydrogen and the hexafluoro-
phosphate counteranion, as suggested by the fluorine–
hydrogen distances which are smaller than the sum of their
van der Waals radii.32 The supramolecular structure is gov-
erned by a complex network of F···H, F···Cl, and Cl···C inter-
actions,33 which connect each molecule to some of its
neighbors. This structural feat certainly points to the lower
thermodynamic stability of the tetraaryl ammonium salt,
which impedes the planarization of the nitrogen, leading to
conjugation between the aromatic systems. Interestingly,
despite these weaker C–N bonds, the carbazolium cation
seems to be fairly thermally robust, as indicated by the ab-
perimental section and copies of the 1H NMR, 13C NMR, and 19F NMR
spectra for all compounds as well as a cif file for compound C (CCDC
1577018).
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
ortioInfgrmoaitn
References and Notes
(1) Jennings, M. C.; Minbiole, K. P. C.; Wuest, W. M. ACS Infect. Dis.
2015, 1, 288.
(2) Kuang, M.; Shiyi, Z.; Lei, J.; Li, Q. J. Appl. Polym. Sci. 2008, 109,
3887.
(3) Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc. 1997, 119, 12414.
(4) Dekel, D. R.; Amar, M.; Willdorf, S.; Kosa, M.; Dhara, S.;
Diesendruck, C. E. Chem. Mater. 2017, 29, 4425.
(5) Varcoe, J. R.; Atanassov, P.; Dekel, D. R.; Herring, A. M.; Hickner,
M. A.; Kohl, P. A.; Kucernak, A. R.; Mustain, W. E.; Nijmeijer, K.;
Scott, K.; Xu, T.; Zhuang, L. Energy Environ. Sci. 2014, 7, 3135.
(6) Gottesfeld, S.; Dekel, D. R.; Page, M.; Bae, C.; Yan, Y.; Zelenay, P.;
Kim, Y. S. J. Power Sources 2018, 375, 170.
(7) von Hofmann, A. W. Ann. Chem. Pharm. 1851, 78, 253.
(8) Marino, M. G.; Kreuer, K. D. ChemSusChem 2015, 8, 513.
(9) Dekel, D. In Encyclopedia of Applied Electrochemistry; Kreysa,
G.; Ota, K.; Savinell, R. F., Eds.; Springer: New York, NY, 2014, 26.
(10) Arava, S.; Diesendruck, C. E. Synthesis 2017, 49, 3535.
(11) Hartwig, J. F. Acc. Chem. Res. 2008, 41, 1534.
(12) Ruiz-Castillo, P.; Buchwald, S. L. Chem. Rev. 2016, 116, 12564.
(13) Ullmann, F.; Bielecki, J. Eur. J. Inorg. Chem. 1901, 34, 2174.
(14) Lam, P. Y.; Clark, C. G.; Saubern, S.; Adams, J.; Winters, M. P.;
Chan, D. M.; Combs, A. Tetrahedron Lett. 1998, 39, 2941.
(15) Li, S.; Yang, F.; Lv, T.; Lan, J.; Gao, G.; You, J. Chem. Commun.
2014, 50, 3941.
1
(16) Hirsch, M.; Dhara, S.; Diesendruck, C. E. Org. Lett. 2016, 18, 980.
(17) Cui, C. M.; Zhou, H. P.; Zhu, C. H.; Wu, J. Y. Acta Crystallogr., Sect.
E: Struct. Rep. Online 2006, 62, o1762.
sence of decomposition products in the H NMR spectrum
of the acetate after 30 minutes at 200 °C.
To conclude, we described a simple, three-step synthesis
of a N,N-diphenylcarbazolium salt using an intramolecular
diazonium-coupling strategy.34 The final product was fully
characterized, and we present, for the first time, an X-ray
structure of a tetraaryl ammonium salt. In addition, we de-
scribe the multiarylation of diaminobiphenyls using an
Ullmann strategy. We demonstrated that an aniline, after
the first arylation, is more reactive towards a second aryla-
tion compared to other aromatic amines present in the
molecules. In the near future, we will further study the
chemical stability of this tetraarylammonium salt and syn-
thesize additional ones using similar synthetic strategies.
(18) Sobolev, A. N.; Belsky, V. K.; Romm, I. P.; Chernikova, N. Y.;
Guryanova, E. N. Acta Crystallogr., Sect. C: Cryst. Struct. Commun.
1985, 41, 967.
(19) Nesmeyanov, A. N.; Tolstaia, T. P.; Grib, A. V. Dokl. Akad. Nauk
SSSR 1963, 153, 608.
(20) Hellwinkel, D.; Seifert, H. Eur. J. Inorg. Chem. 1972, 105, 880.
(21) Shchepina, N. E.; Nefedov, V. D.; Toropova, M. A.; Avrorin, V. V.;
Lewis, S. B.; Mattson, B. Tetrahedron Lett. 2000, 41, 5303.
(22) Patil, N. M.; Kelkar, A. A.; Chaudhari, R. V. J. Mol. Catal. A: Chem.
2004, 223, 45.
(23) Tseng, K.-C. Acta Chim. Sin. 1966, 32, 107.
(24) Buschman, O.; Grosjean, M.; Nasielski, J.; Martin, R. H. Helv.
Chim. Acta 1964, 47, 2037.
(25) (a) Hansch, C.; Leo, A.; Hoekman, D. Exploring QSAR: Hydropho-
bic, Electronic, and Steric Constants; American Chemical Society:
Washington, DC, 1995, 254. (b) Hansch, C.; Leo, A.; Hoekman, D.
Exploring QSAR: Hydrophobic, Electronic, and Steric Constants;
American Chemical Society: Washington, DC, 1995, 273.
(26) Ran, C.; Dai, Q.; Harvey, R. G. J. Org. Chem. 2005, 70, 3724.
(27) Alcalde, E.; Dinarés, I.; Rodríguez, S.; Garcia de Miguel, C. Eur.
J. Org. Chem. 2005, 1637.
Funding Information
This work was funded by the Ministry of National Infrastructure, En-
ergy and Water Resources of Israel (grant 216-11-048); the Ministry
of Science, Technology & Space of Israel through the M.era-NET Trans-
national Call 2015, NEXTGAME project grant 3-12948; and the Euro-
pean Union's Horizon 2020 research and innovation program under
(28) Shaughnessy, K. H.; Ciganek, E.; DeVasher, R. B.; Denmark, S. E.
Copper-Catalyzed Amination of Aryl and Alkenyl Electrophiles;
John Wiley and Sons: Hoboken: NJ, 2014, 6.
grant agreement No 721065.
H
2
0
2
0
E
n
erg
y
7(
2
1
0
6
5)
(29) Cristau, H.-J.; Cellier, P. P.; Spindler, J.-F.; Taillefer, M. Chem. Eur.
J. 2004, 10, 5607.
(30) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A.
G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1.
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0036-1591848. Included are a complete ex-
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

E
S. Aharonovich et al.
Letter
Synlett
(31) Avendaño, C.; Espada, M.; Ocaña, B.; García-Granda, S.; Díaz, M.
del. R.; Tejerina, B.; Gómez-Beltrán, F.; Martínez, A.; Elguero, J.
J. Chem. Soc., Perkin Trans. 2 1993, 1547.
(32) Mantina, M.; Chamberlin, A. C.; Valero, R.; Cramer, C. J.; Truhlar,
D. G. J. Phys. Chem. A 2009, 113, 5806.
(33) Bissantz, C.; Kuhn, B.; Stahl, M. J. Med. Chem. 2010, 53, 5061.
(34) Representative Experimental Procedures and Characteriza-
tion Data 2,2′-Diaminobiphenyl (A)
Analytical Data of Compound B
Red-brown solid, 60%; Rf = 0.57 (CHCl3). 1H NMR (400 MHz,
CDCl3): δ = 7.66, (d, J = 8.0 Hz, 1 H), 7.59, (d, J = 8.0 Hz, 1 H), 7.53
(dt, J = 8.0, 1.3 Hz, 1 H), 7.43 (t, J = 8.0 Hz, 1 H), 7.25 (t, J = 8.0 Hz,
4 H, Ph-m), 7.12 (t, J = 8.0 Hz, 2 H, Ph-p), 7.07–7.02 (m, 6 H),
6.75 (t, J = 8.0 Hz, 1 H), 6.58 (d, J = 8.0 Hz, 1 H), 3.38 (br s, 2 H,
NH2). 13C NMR (75 MHz, CDCl3): δ = 147.34 (C(Ph)–N, 2 C),
146.04 (C–N, 1 C), 143.35 (C–N, 1 C), 136.82 (C–C, 1 C), 132.46
(C–H(biphenyl), 1 C), 130.10 (C–H(biphenyl), 1 C), 128.86 (C–H
(biphenyl), 1 C), 128.62 (C–H(biphenyl), 1 C), 128.50 (C–H(Ph),
4 C), 127.89 (C–H(biphenyl), 1 C), 125.25 (C–H(biphenyl), 1 C),
125.10 (C–C, 1 C), 122.02 (C–H(Ph), 4 C), 121.36 (C–H(Ph), 2 C),
118.03 (C–H(biphenyl), 1 C), 115.26 (C–H(biphenyl), 1 C). APCI-
HRMS: m/z calcd for C24H21N2 [M + H]+: 337.1699; found:
337.1719 (–5.8 ppm).
In a 3 L beaker flask, glacial acetic acid (190 mL) is added to a
stirred solution of 2,2′-dinitro biphenyl (13.00 g, 53.23 mmol)
in absolute ethanol (1.9 L), followed by portionwise addition of
zinc powder (75.00 g, 1147 mmol) during 2 min. Once the exo-
thermic reaction subsides (10 min) the beaker is left stirring for
additional 5 min after which it is allowed to reach RT. The
yellow-green reaction mixture is filtered, the unreacted zinc
powder is washed with small amounts of ethanol, and the fil-
trate is concentrated to a thick syrup, which is dissolved in 16%
hydrochloric acid (80 mL). The aqueous phase is extracted with
CHCl3 (5 × 20 mL), until no color is observed in the organic
phase, and then added slowly to 25% NH4OH (1 L), for liberation
of the amine and complexation of zinc(II). The free amine is
extracted with CHCl3 (4 × 100 mL), and the organic phase is
washed with water (40 mL) and dried over Na2SO4. After the
solvent is removed in vacuo, A is obtained as a light brown oil
which solidifies to an off-white solid (9.23 g, 94%).
N,N-Diphenylcarbazolium Hexafluorophosphate (C)
In a 100 mL Erlenmeyer beaker, a mixture of B and F (4.36 g) is
dissolved in glacial acetic acid (53 mL). The solution is cooled to
0–5 °C using an ice bath, and the solidified acetic acid solution
was crushed by a steel spatula. A solution of NaNO2 (4.15 g, 60.2
mmol) in water (5.5 mL) is added, and the resulting dark slurry
stirred mechanically (using a spatula) for 20 min. Urea prills
(3.32 g, 55.3 mmol) are then added, and the mixture stirred
mechanically (using a spatula) for 1 h at 40 °C, during which a
black tar appears. The reaction mixture is concentrated under
reduced pressure and the residue dissolved in CHCl3 (50 mL), fil-
tered, and the filtration cake washed with small amounts of
CHCl3. The solvent is removed in vacuo, and the dark residue
partitioned between diethyl ether (40 mL) and water (10 mL).
The organic phase is further extracted with water (5 mL), and
the volume of the united aqueous phase reduced to ca. 6 mL in
vacuo, and extracted with small amounts of diethyl ether until
no color is seen in the organic phase. Removal of water and
acetic acid from the yellow aqueous phase by reduced pressure
azeotropic distillation with toluene and freeze-drying affords
the acetate salt as a highly hygroscopic yellow solid, which still
contains acetic acid and water. Pure C is obtained by addition of
75% (w/v) NH4PF6 solution (5 mL) to the concentrated solution
of acetate. The resulted light yellow precipitate is collected,
washed with small amounts of water, and dried under reduced
pressure.
2-Amino,2′-(N,N-diphenylamino)biphenyl (B)
In a 250 mL flame-dried Schlenk flask, A (9.00 g, 48.9 mmol) is
deoxygenated (see Supporting Information). Xylenes (135 mL)
are transferred via a syringe, and to the resulting solution KOtBu
(12.09 g, 107.7 mmol) is added under argon. The mixture is
stirred for 10 min at RT, during which a brown-yellow color
develops. PhI (20.93 g, 102.6 mmol) is injected to the Schlenk
flask under argon, followed by addition of CuI (1.86 g, 9.77
mmol) and o-phenanthroline (1.76 g, 9.77 mmol). The mixture
is stirred at 125 °C for 3.5 h, after which it is allowed to reach RT.
The mixture is filtered, and the filtration cake washed with
small amounts of CHCl3 (ca. 70 mL). The solids are dissolved in
25% NH4OH (200 mL), and the dark green-blue aqueous phase is
extracted with CHCl3 (2 × 30 mL) which is added to the filtrate.
The filtrate is concentrated to a dark paste, which is dissolved in
CHCl3 (200 mL) and extracted with small amounts of 25%
NH4OH, until the aqueous phase is free from the copper com-
plexes color. The organic phase is then extracted with 0.1 M HCl
(3 × 250 mL) for separation of the unreacted A and phenanthro-
line, which can be recycled. The organic phase is washed with
aq sat. NaHCO3, water, dried over Na2SO4, filtered, and the
solvent is removed in vacuo to give a dark paste, consisting of a
mixture of B and F (ca. 7:3 mol ratio), which is used for the next
step without further purification. Samples of pure B and F are
obtained by column chromatography of this mixture.
Analytical Data of Compound C
Yellow solid, 857 mg, 22% from B; H NMR (400 MHz, CD3CN):
1
δ = 8.21 (d, J = 8.0 Hz, 2 H, biphenyl), 7.81 (t, J = 8.0 Hz, 2 H), 7.73
(d, J = 8.0 Hz, 2 H), 7.67–7.62 (m, 4 H), 7.56 (t, J = 8.0 Hz, 4 H, m-
Ph), 7.47 (d, J = 8.0 Hz, 4 H, o-Ph). 13C NMR (75 MHz, CD3CN): δ
= 151.31 (2 C), 147.74 (2 C), 133.40 (C–H(biphenyl), 2 C), 132.53
(C–H(biphenyl), 2 C), 132.19 (C–H(biphenyl), 2 C), 131.88 (C–H
(Ph), 4 C), 131.52 (C–C(biphenyl), 2 C), 124.53 (C–H(biphenyl),
2 C), 123.55 (C–H(Ph), 4 C). 19F NMR (377 MHz, CD3CN): δ = –72.82
(d, J1= 707 Hz, P–F). ESI-HRMS: m/z calcd for C24H18N [M – PF6]+:
320.1439; found: 320.1420 (–5.9 ppm).
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E