Scalable synthesis of pure and stable hexaaminobenzene trihydrochloride

Article abstract of DOI:10.1055/s-0032-1317696

Synthesis of hexaaminobenzene (HAB) in pure and stable form has remained as an important challenge for a long time, since it is a fascinating synthon for the synthesis of aromatic nitrogenous compounds having many interesting applications. Here, we report an improved synthesis of pure and stable HAB form using modified catalytic hydrogenation in aqueous acidic medium. The structure of needle-shaped HAB crystals was confirmed by single-crystal X-ray diffraction study. The synthetic protocol could thus be a simple, but efficient for the large-scale synthesis of highly pure and stable HAB. Georg Thieme Verlag Stuttgart New York.

Full text of DOI:10.1055/s-0032-1317696

246▌
LETTER
letter
Scalable Synthesis of Pure and Stable Hexaaminobenzene Trihydrochloride
Pure and Stable Hexaaminobenzene Trihydrochloride
Javeed Mahmood, Dongwook Kim, In-Yup Jeon, Myoung Soo Lah,* Jong-Beom Baek*
Interdisciplinary School of Green Energy/Low-Dimensional Carbon Materials Center, Ulsan National Institute of Science and Technology
(UNIST), 100 Banyeon, Ulsan, 689-798, South Korea
Fax +82(52)2172019; E-mail: jbbaek@unist.ac.kr; E-mail: mslah@unist.ac.kr
Received: 04.10.2012; Accepted: 02.11.2012
TATB (Figure 1, a). Although some quantity of HAB was
Abstract: Synthesis of hexaaminobenzene (HAB) in pure and sta-
ble form has remained as an important challenge for a long time,
since it is a fascinating synthon for the synthesis of aromatic nitrog-
enous compounds having many interesting applications. Here, we
produced by using literature-reported methods with poor
quality in very low yield (Figure 1, b), all cases have failed
during the catalyst removal and purification. The extreme-
report an improved synthesis of pure and stable HAB form using ly poor stability of free HAB was the major contributing
modified catalytic hydrogenation in aqueous acidic medium. The
structure of needle-shaped HAB crystals was confirmed by single-
crystal X-ray diffraction study. The synthetic protocol could thus be
a simple, but efficient for the large-scale synthesis of highly pure
and stable HAB.
factor by judging its white color rapidly turning into dark
brown. Kohne et al. have reported the synthesis of HAB
13
trihydrochloride crystals. However, the procedure is te-
dious, time-consuming, and requires extreme care. Any
minor mistake can ruin the whole effort, and thus obtain-
ing pure HAB in large scale has been almost impossible.
Nevertheless, the quality of HAB trihydrochloride crys-
tals is the best among reported methods, although it is still
unsatisfactory (Figure 1, c). After repeated failures to syn-
thesize highly pure HAB crystals, we have now optimized
a new and convenient procedure for the synthesis of this
fragile compound in a stable form by using a modified cat-
alytic reduction.
Key words: hexaaminobenzene, palladium, reduction, TATB,
TNA
The usefulness of the hexaaminobenzene (HAB) can be
imagined on the basis of the new type of polynitrogeneous
heterocyclic systems including derivatives of hexaazatri-
phenylene,1 hexaazacoronene, fused benzotriazoles,
,2
3
4
5
,6
7
benzotris(imidazole), benzotris[1,2,5]thiadiazole, ben-
zotris[1,2,5]selenodiazole, aromatic hexaamides with the
7
4
properties of lyotropic liquid crystals, and many other in-
teresting materials.8 This interesting symmetric com-
–10
pound has been known to synthetic community for many
1
1
decades. There have been many attempts to synthesize
HAB by reducing 1,3,5-triamino-2,4,6-trinitrobenzene
2
3
(
TATB) with phenylhydrazine, catalytic hydrogenation,
1
and sodium in liquid ammonia, which are commonly
used methods. In all these cases, the HAB obtained is in
the form of free amine, which is highly sensitive to oxy-
gen and light, decomposed into dark brown amorphous
Figure 1 (a) TATB; (b) HAB crystals as one of the best obtained
samples from conventional catalytic reduction of TATB in the pres-
3
ence of H and palladium (10 wt%) on activated carbon (Pd/C); (c)
2
HAB trihydrochloride crystals obtained from the reduction of TATB
product both in diffuse daylight and in the dark within a
4
in the presence of SnCl /HCl system.
2
few hours even under the argon atmosphere.3
,12,13
Al-
though HAB trihydrochloride has been envisaged in one
step in an aqueous HCl solution in the presence of SnCl2,
The overall synthesis started with the well-known nitra-
tion of p-nitroaniline into 2,4,6-trinitroaniline in concen-
trated sulfuric acid in the presence of potassium nitrate
4
its purity and stability is still not satisfactory. Hitherto,
there have thus been no reasonably feasible reports for the
large-scale synthesis of highly pure HAB trihydrochloride
crystals in an efficient way, due mainly to the instability
of HAB. As a result, the most reported methods have di-
rectly used HAB for subsequent reactions without purifi-
cation to remove reduction catalyst and possible
impurities, if any.1
1
4
(
Scheme 1). The purified 2,4,6-trinitroaniline was treat-
ed with 4-amino-1,2,4-triazole (ATA) and sodium
methoxide in dimethyl sulfoxide (DMSO) under vicarious
nucleophilic reaction conditions, and subsequent acid
quenching with 0.4 M HCl produced 1,3,5-triamino-
15
2
,4,6-trinitrobenzene (TATB) in high yield. Interesting-
,3
ly, the TATB in DMSO displays a mesomeric form to re-
lief some of its strain by delocalizing its lone-pair electron
We surveyed different literature procedures to obtain
highly pure HAB by using a variety of conventional meth-
ods involving the reduction of aromatic nitro groups on
13
into the aromatic ring (Scheme 2). C NMR confirms the
proposed mesomerism in TATB.
Finally, the TATB was reduced to form HAB trihydro-
chloride by a modified procedure to give highly pure
creamy white needlelike crystals (Figure 2, a), highly
SYNLETT 2013, 24, 0246–0248
Advanced online publication: 04.12.2012
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
DOI: 10.1055/s-0032-1317696; Art ID: ST-2012-U0852-L
Georg Thieme Verlag Stuttgart · New York
©

LETTER
Pure and Stable Hexaaminobenzene Trihydrochloride
247
NH2
NH2
NH2
NH2
O2N
NO2
1) ATA, MeONa, DMSO
O2N
H2N
NO2
NH2
H N
NH2
NH2
2
1
) KNO3, H2SO4
) aq NaCl
1) H2, Pd/C
⋅3HCl
2
2) aq HCl
2) concd HCl
2
H N
NO2
NO2
NO2
NH2
1
2
3
Scheme 1 The synthesis of HAB trihydrochloride
The density of electrons on the HAB is extremely high,
because of the presence of six strong electron-donating
+
NH2
NH2
O2N
H2N
NO2
NH2
O2N
H2N
NO2
NH groups around the aromatic benzene ring, which
2
DMSO
makes this compound highly reactive and unstable in its
free form (base). To reduce the electronic density by
push–pull mechanism, we converted it into its hydrochlo-
ride salt form, which reduces the electronic density to
some extend but it is still very reactive when it is in the
dissolved state. The HAB could be a fascinating building
NH2
N
N
O
+
O–
–O
+
O–
Scheme 2 Mesomerism of TATB in DMSO
1
6–19
transparent and reflective (Figure 2, b), indicating the high block for the synthesis of discotic liquid crystals,
quality of crystals. Crystal-length dimensions are longer compounds with low barrier to both electron and hole
than 1 mm (Figure 2, c) and some are even longer than 5 transport for potential applications in organic electron-
2
0
21
mm from the optical microscope image. The crystals are ics, synthesis of macrocyclic compounds, hexaazatri-
8
,22
stable in a closed vial since five months (Figure 2, d). The phenylene-based donor–acceptor molecules, electron-
absolute crystal structure of HAB trihydrochloride has deficient expanded-electron systems, ferromagnetic or-
2
3
3
,24
25
been solved for the first time by single-crystal X-ray dif- ganic salts, and fluorescence sensor for cadmium(II).
fraction analysis (Figure 2, e and f, see also items S1–S3
In summary, we have the strong hunch that our method
would be equally useful for the synthesis of other sensi-
tive aromatic amines from their parent polynitroarenes.
Albeit the modified reduction method reported here has
not been tested on other polynitroarene compounds, it
would be of great interest for the reduction of other sensi-
tive compounds with this newly developed method. Fur-
thermore, pure HAB could be an important building block
for the synthesis of many aesthetic and application-specif-
ic compounds.
in the Supporting Information). Hence, the large-scale
synthesis of pure HAB trihydrochloride crystals was
achieved by an optimized synthetic procedure (see details
in the experimental section). All substituted aromatic nitro
compounds should be handled with care because of their
explosive nature, but we did not face any such accident in
our extended experimental work.
Solvents, chemicals, and reagents were purchased from Aldrich
Chemical Inc., unless otherwise stated. Reactions were performed
under nitrogen or argon atmosphere using oven-dried glassware.
1
4
15
2
,4,6-Trinitroaniline and 1,3,5-triamino-2,4,6-trinitrobenzene
were synthesized according to the literature. 4-Nitroaniline
185310) 99%, DMSO (276855) 99.9%, Pd/C 10% (75990), HCl
7% (435570), and EtOAc (anhyd) were purchased from Sigma-Al-
(
3
drich Chemical Inc. Nitric acid 60% (010223) was purchased from
DC Chemical Co. Ltd. Celite 545 (C0340) was provided by Samc-
hun Pure Chemical Co. Ltd.
Infrared (FT-IR) spectra were recorded on a Perkin-Elmer Spec-
trum 100 FT-IR spectrophotometer. Single-crystal X-ray diffrac-
1
tion data were taken on R-AXIS RAPID II (Rigaku, Japan). H
1
3
NMR and C NMR spectra were taken by using a FT-NMR 600
MHz VNMRS 600 (Varian, USA) spectrometer. Mass spectra were
recorded on a 320-MS Varian, USA. Elemental analysis was carried
out on Flash 2000 Thermo Scientific, Netherland. Hydrogenation
was carried out using shaker-type Parr Hydrogenation Apparatus
Model-3911. Melting points were calculated on KSP1N automatic
melting point meter (A. Krüss Optronic GmbH, Germany).
Figure 2 (a) Digital photograph of HAB trihydrochloride crystals on
PTFE filter membrane; (b), (c) optical microscopic images of the
HAB trihydrochloride crystals; (d) digital photograph of HAB in vial
after five months; (e) a ball-and-stick diagram of formula unit of the
HAB trihydrochloride; (f) hydrogen-bonding-driven molecular pack-
ing viewed along the crystallographic c axis, where the hydrogen
bonds were represented using dotted lines. Color codes: carbon, gray;
hydrogen, dark gray; nitrogen, blue; chloride, green.
Synthesis of TNA (1)
4-Nitroaniline (20 g, 0.088 mol) and H SO (100 mL) were charged
2
4
in a round-bottom flask. A solution of KNO (70 g, 0.70 mol) in
3
H SO (100 mL) was then added dropwise at 50 °C. The mixture
2
4
was heated to 80 °C for 3 h and 110 °C for 3 h. After completion of
©
Georg Thieme Verlag Stuttgart · New York
Synlett 2013, 24, 246–248

248
J. Mahmood et al.
LETTER
the reaction, the reaction mixture was cooled to ambient tempera- Acknowledgment
ture and poured into ice water. The precipitates were collected by
suction filtration, air-dried, and recrystallized from very dilute aq
This work was supported by World Class University (WCU), US-
Korea NBIT, Mid-Career Researcher (MCR) and Basic Research
Laboratory (BRL) programs through the National Research Foun-
dation (NRF) of Korea and US Air Force Office of Scientific Re-
search (AFOSR).
HCl solution to give 16.5 g (50% yield) of 2,4,6-trinitroaniline
1
4
(
glassy light yellow crystal); mp 186–188 °C (lit. 188 °C). MS (EI,
+
DIP): m/z (%) = 229 (100) [M + H] . Anal. Calcd (%) for C H N O
6
4
4
6
(
1
228.12): C, 31.59; H, 1.77; N, 24.56; O, 42.08. Found: C, 30.51; H,
1
.30; N, 20.70; O, 43.86. H NMR (600 MHz, DMSO): δ = 8.59 (2
1
3
H), 5.60 (2H) ppm. C NMR (600 MHz, DMSO, S6): δ = 141.86
Supporting Information for this article is available online at
(C-4), 125.27 (C-2), 124.55 (C-3), 160.62 (C-1) ppm. See also items
http://www.thieme-connect.com/ejournals/toc/synlett. SnuIpofoipmngr irtSatnoIuipfog
r
t
iornat
S4–S6 in the Supporting Information.
Synthesis of TATB (2)
NaOMe (23.8 g, 0.44 mol) was added to a solution of TNA (4.56 g, References
0
.02 mol) and ATA (16.8 g, 0.2 mol) in DMSO (300 mL). The red-
(
(
(
1) Rogers, D. Z. J. Org. Chem. 1986, 51, 3904.
2) Kohne, B.; Praefcke, K. Liebigs Ann. Chem. 1985, 522.
3) Breslow, R.; Maslak, P.; Thomaides, J. S. J. Am. Chem. Soc.
dish-orange suspension was stirred at ambient temperature for 3 h.
The reaction mixture was then poured into cold aq HCl (0.4 M) so-
lution. The resulting precipitates were collected by suction filtra-
tion, washed with distilled H O and dried. The solid was dissolved
in DMSO with small amount of NaOH with gentle heating up to 70
1
984, 106, 6453.
2
(
4) Khisamutdinov, G. K.; Korolev, V. L.; Kondyukov, I. Z.;
Abdrakhmanov, I. S.; Smirnov, S. P.; Fainzilberg, A. A.;
Dorokhov, V. G. Russ. Chem. Bull. 1993, 42, 136.
5) Kohne, B.; Praefcke, K.; Derz, T.; Gondro, T.; Frolow, F.
Angew. Chem., Int. Ed. Engl. 1986, 25, 650.
°C, when the compound was completely dissolved the solution was
poured into ice-cold aq HNO (0.4 M) solution. The product precip-
itated was again collected by filtration to give 4.62 g (89% yield) of
TATB (deep bright yellow powder); mp >300 °C (dec.) [lit. >300
3
(
2
+
°C (dec.)]. MS (EI, DIP, S7): m/z (%) = 259.1 (100) [M + H] . Anal.
(6) Wolff, J. J.; Limbach, H.-H. Liebigs Ann. Chem. 1991, 691.
(7) Praefcke, K.; Kohne, B.; Korinth, F. Liebigs Ann. Chem.
1990, 203.
(8) Juárez, R.; Ramos, M.; Segura, J. L.; Orduna, J.; Villacampa,
B.; Alicante, R. J. Org. Chem. 2010, 75, 7542.
(9) Ishi-i, T.; Murakami, K.-I.; Imai, Y.; Mataka, S. J. Org.
Chem. 2006, 71, 5752.
(10) Ishi-i, T.; Murakami, K.-I.; Imai, Y.; Mataka, S. Org. Lett.
2005, 7, 3175.
(11) Flurscheim, B.; Holmes, E. L. J. Chem. Soc. 1929, 330.
(12) Thomaides, J.; Maslak, P.; Breslow, R. J. Am. Chem. Soc.
1988, 110, 3970.
Calcd (%) for C H O N (258.14): C, 27.9; H, 2.32; N, 32.55; O,
6
6
6
6
1
3
3
7.2. Found: C, 28.03; H, 1.86; N, 26.57; O, 44.79. C NMR (600
MHz, DMSO, S8): 113.52 (C-4), 117.48 (C-2), 147.01 (C-3),
57.14 (C-1). The FT-IR spectrum shows an asymmetrical N–H
1
–
1
stretching absorption peak at 3363 cm and a symmetrical N–H
stretching vibration at 3225 cm , which are characteristic peaks for
TATB. Other peaks are N–O asymmetrical stretching at 1560 cm ,
symmetrical N–O stretching at 1201 cm , C–N stretching (amino
–
1
–1
–
1
–
1
group) 1613 and 798 cm , and skeletal stretching of the ring at
–
1
1475, 1180, 1053 cm . See also items S7 and S8 in the Supporting
Information.
(
(
13) Kohne, B.; Praefcke, K. Liebigs Ann. Chem. 1987, 265.
14) You, B. W.; Bo, Z. W.; Zhi, H. Y.; Yan, S.; Hui, Q. Chin. J.
Energ. Mater. 2011, 19, 142.
Synthesis of HAB (3)
TATB (3.0 g, 0.012 mol) was taken in a high-pressure hydrogena-
tion bottle with 10% Pd/C (500 mg) and pure EtOAc (150 mL) as a
solvent. The reaction flask was placed and fixed on hydrogenation
(
(
(
(
(
15) Mitchell, A. R.; Pagoria, P. F.; Schmidt, R. D. US 5633406,
1
997.
apparatus and agitated under H (4.2 bar) until all the yellowish col-
or of the reactant disappeared completely within 3 d. Then, concen-
trated HCl (90 mL) was added, and the reaction was continued
2
16) Ong, C. W.; Liao, S.-C.; Chang, T. H.; Hsu, H.-F.
Tetrahedron Lett. 2003, 44, 1477.
17) Yeh, M.-C.; Liao, S.-C.; Chao, S.-H.; Ong, C. W.
Tetrahedron 2010, 66, 8888.
18) Ong, C. W.; Liao, S.-C.; Chang, T. H.; Hsu, H.-F. J. Org.
Chem. 2004, 69, 3181.
19) Ishi-i, T.; Hirayama, T.; Murakami, K.-i.; Tashiro, H.;
Thiemann, T.; Kubo, K.; Mori, A.; Yamasaki, S.; Akao, T.;
Tsuboyama, A.; Mukaide, T.; Ueno, K.; Mataka, S.
Langmuir 2005, 21, 1261.
under H for an additional 5 h. The reaction mixture was filtered un-
2
der reduced pressure over Celite to remove catalyst. The HAB tri-
hydrochloride crystallized out very nicely (see item S9 in the
Supporting Information) in high yield. The precipitates were col-
lected by suction filtration by using polytetrafluoroethylene (PTFE)
membrane (0.5 μm pore) and dried in an oven at 70 °C for 4 h under
–
4
reduced pressure (6.6·10 bar). The white small crystals were redis-
solved in deionized H O and filtered through PTFE membrane to re-
2
(
20) Barlow, S.; Zhang, Q.; Kaafarani, B. R.; Risko, C.; Amy, F.;
Chan, C. K.; Domercq, B.; Starikova, Z. A.; Antipin, M. Y.;
Timofeeva, T. V.; Kippelen, B.; Brédas, J.-L.; Kahn, A.;
Marder, S. R. Chem.–Eur. J. 2007, 13, 3537.
move solid impurities, if any, and added 80 mL of concentrated
HCl. The flask was tightly sealed and placed in the freezer until very
nice big crystals developed. The crystals were collected on PTFE
membrane filter in the room environment and washed thoroughly
with EtOAc and dried in the vacuum oven (Figure 2, a) to afford 3.0
g (92% yield). HAB trichloride shows no melting point but becomes
(21) Secondo, P.; Fages, F. Org. Lett. 2006, 8, 1311.
(22) Juárez, R.; Ramos, M. M.; Segura, J. L. Tetrahedron Lett.
2007, 48, 8829.
(23) Ishi-i, T.; Hirashima, R.; Tsutsumi, N.; Amemori, S.;
Matsuki, S.; Teshima, Y.; Kuwahara, R.; Mataka, S. J. Org.
Chem. 2010, 75, 6858.
(24) Breslow, R.; Jaun, B.; Kluttz, R. Q.; Xia, C.-Z. Tetrahedron
1982, 38, 863.
(25) Zhao, Q.; Li, R.-F.; Xing, S.-K.; Liu, X.-M.; Hu, T.-L.; Bu,
X.-H. Inorg. Chem. 2011, 50, 10041.
1
1
dark at above 250 °C. (lit. >240 °C darken). MS (EI, DIP): m/z (%)
+
+
=
169.1 (78.5) [M + H] , 167.1 (100) [M – H] . Anal. Calcd (%) for
C H N Cl (277.58): C, 25.96; Cl, 38.32; H, 5.45; N, 30.28. Found:
6
15
6
3
C, 26.54; Cl, 36.54; H, 4.96; N, 25.05; O, 6.89. FT-IR: 3384.41,
3249.28, 2977.06, 2564.72, 1671.16, 1636, 1582.02, 1554,
–
1
1479.74, 1279, 1202.9, 1166.34, 1096 cm . See also item S10 in
the Supporting Information.
Synlett 2013, 24, 246–248
© Georg Thieme Verlag Stuttgart · New York