A facile synthesis of 2,4-diaza-1-borines from anilines

Article abstract of DOI:10.1016/S0040-4039(02)00365-9

A general synthesis of diazaborines is described with the in situ preparation of amidines from anilines using a nitrile and AlCl₃, followed by the addition of BCl₃ and heating. The diazaborines were isolated as hydrochlorides in pure

Full text of DOI:10.1016/S0040-4039(02)00365-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3255–3257
A facile synthesis of 2,4-diaza-1-borines from anilines
George T. Lee, Kapa Prasad* and Oljan Repicˇ
Process Research and Development, Novartis Institute for Biomedical Research, One Health Plaza, East Hanover,
NJ 07936, USA
Received 31 January 2002; accepted 20 February 2002
Abstract—A general synthesis of diazaborines is described with the in situ preparation of amidines from anilines using a nitrile
and AlCl3, followed by the addition of BCl3 and heating. The diazaborines were isolated as hydrochlorides in pure form and, in
the case of 5, the structure was confirmed by an X-ray analysis. © 2002 Elsevier Science Ltd. All rights reserved.
The use of boron trichloride/aluminum chloride to
transform anilines and nitriles to 2-aminophenyl
ketones (Scheme 1) is commonly referred to as the
Sugasawa reaction.1 Under these conditions, amidine 3
is a competing by-product especially when anilines with
a deactivating substituent like bromine were used. With
the intention of extending the ‘Sugasawa ortho
acylation’ to amidines we subjected 3 to BCl3/AlCl3
conditions and found a product resulting from ortho
borylation and not ortho acylation. The present com-
munication describes the results of these investigations.
and AlCl3. In the present study (Scheme 2) we accom-
plished this in situ in toluene solution. Thus treatment
of 1 with propionitrile and AlCl3 in toluene followed by
heating to 90°C gave a quantitative yield of 3 as
monitored by HPLC. At this stage the reaction mixture
was cooled to 0°C, BCl3 was bubbled in, and refluxed
for 2 h. Upon quenching with ice water the precipitated
product (55% yield) was collected and characterized
further. Elemental analysis indicated a molecular for-
mula of C9H11BBrClN2O suggesting incorporation of
boron atom into the molecule. 11B NMR showed the
chemical shift at 20.71 ppm (DMSO-d6) which indicates
an sp2 B in an aromatic ring.2 1H NMR indicated the
absence of one of the ortho protons in the aromatic
system. As the original solid was not suitable for X-ray
study, attempts were made to find a solvent for crystal-
The standard methods for making amidines is to heat
the neat stoichiometric mixture of aniline and nitrile
CH3
CH3CH2CN
Cl
Cl
Br
Br
N+H
BCl3/AlCl3
Br
B
Br
B−Cl2
N
NH2
N
H
CH3
2
4
1
N
H
NH2
1
Heat;
H2O
CH3
O
CH3CH2CN
AlCl3
Toluene
BCl3
Br
Br
NH
OH
CH3
Cl−
N+H
Br
N
H
Br
B
NH2
NH
3
CH3
CH3
N
H
N
H
Scheme 1.
5
3
Keywords: amidines; ortho-borylation; 2,4-diaza-1-borines.
* Corresponding author.
Scheme 2.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00365-9

3256
G. T. Lee et al. / Tetrahedron Letters 43 (2002) 3255–3257
chloride salt is very stable and is resistant to H2O2
oxidation. Extension of this reaction to p-toluidine and
1-aminonaphthalene under the same conditions as with
4-bromoaniline resulted in the isolation of diazaborines
6 and 7 in 70% and 78% yield, respectively. Spectro-
scopic data and elemental analysis are in complete
agreement with the proposed structures.4
lization. Finally, water was found to be a suitable
solvent, although under these conditions the product
retained one equivalent of water as indicated by ele-
mental analysis (C9H13BBrClN2O2). Based on X-ray
analysis the diazaborine structure (see Fig. 1) was
assigned.
OH
The main feature in the crystal packing of 5 is that the
chloride ion and water molecules are linked to diaza-
borine through hydrogen bonding (Fig. 2). The forma-
OH
Cl−
N+H
Cl−
N+H
B
H3C
B
CH3
tion of
5
could be rationalized through the
CH3
N
H
intermediacy of amidine 3 as shown in Scheme 2.
N
H
It is interesting to note that the above diazaborine does
not add methanol or isopropanol on recrystallization
from these solvents. This is in contradiction to the
ready but reversible addition of weakly nucleophilic
species such as water and methanol to 2,4-diaza-1-
borine systems studied in the literature.3 The hydro-
6
7
The reaction of N,N%-diphenylacetamidine 8 with BCl3
in boiling benzene was reported earlier by Mikhailov
and co-workers3 and the product obtained after
quenching with water was proposed to have structure 9.
Based on analogy with our compounds it is tempting to
reconsider this compound as a chloride salt, although
more evidence is needed to confirm this one way or
other.
OH
N
Cl
Ph
B
Ph
BCl3
N
benzene
CH3
CH3
N
H
N
H
8
9
In conclusion we have demonstrated an efficient synthe-
sis of 2,4-diaza-1-borines from anilines with the in situ
intermediacy of amidines.
Acknowledgements
Figure 1. X-Ray structure of 5.
We thank Dr. Peter Karpinski (Novartis) for crystal
selection and Dr. Phillip Fanwick (Purdue University,
Indiana) for the X-ray structure of compound 5.
References
1. Sasakura, K.; Terui, Y.; Sugasawa, T. Chem. Pharm. Bull.
1985, 33, 1836 and references cited therein.
2. Groziak, M. P.; Ganguly, A. D.; Robinson, P. D. J. Am.
Chem. Soc. 1994, 116, 7597–7605.
3. Boldyrev, O. G.; Dorokhov, V. A.; Mikhailov, B. M. Izv.
Akad. Nauk USSR, Ser. Khim. 1985, 2, 428–430.
4. Representative procedure: 7-Bromo-3-ethyl-1,2-dihydro-1-
hydroxy-2,4-benzodiaza-1-borine mono hydrochloride (5):
A solution of 13.4 g of 4-bromoaniline (0.075 mol) in 21.1
g of propionitrile (0.375 mol) was added to a suspension of
31.4 g of anhydrous AlCl3 (0.16 mol) in 75 mL of toluene
at 5–50°C (exothermic). The contents were heated up to
107°C and maintained refluxing for 1.5 h. The reaction
mixture was cooled to −5°C; 18 g of boron trichloride gas
Figure 2. Crystal packing of 5.

G. T. Lee et al. / Tetrahedron Letters 43 (2002) 3255–3257
3257
(0.15 mol) was slowly bubbled into the mixture at −5°C
over 15 min (exothermic). The suspension was heated
again and refluxed for 4 h (HCl gas evolution). The
reaction mixture was cooled to 22°C and poured into a
mixture of 400 g of crushed ice and 100 mL of toluene
with efficient stirring. The slurry was warmed to 15°C and
maintained at this temperature for 0.5 h. The beige solid
was filtered, washed with 1000 mL of cold water (5°C) and
dried at 40–45°C (15 mbar) overnight to give 11.8 g (55%
yield) of 5; mp 223–224°C; IR (KBr): 3308, 3211, 3070,
2918, 2790, 1651, 1577, 1491, 1455, 1407, 1244, 1168, 1127,
1069, 892, 867, 830, 522 cm−1; UV: umax (nm) (‡×104)
C9H12BBrClN2O2: C, 35.17; H, 4.26; B, 3.52; Br, 27.61; Cl,
11.53; N, 9.11; O, 10.41. Found: C, 35.31; H, 4.13; B, 3.50;
Cl, 11.43; N, 9.02.
Compound 6: mp 187–188°C; IR (KBr): 3567, 3295, 2922,
2855, 1650, 1582, 1475, 1404, 1291, 1256, 1220, 1200, 881,
863, 828, 761, 528, 481 cm−1; UV. umax (nm) (‡×104)
1
(CH3OH) 263 (7.16), 204 (1.16); H NMR (DMSO-d6): l
7.78 (s, 1H), 7.54 (1H, dd, J=8.5 Hz), 7.44 (1H, d, J=8.5
Hz), 2.93 (2H, q, J=7.7 Hz), 2.44 (s, 1H, OH), 1.44 (3H,
t, J=7.7 Hz); 13C NMR (DMSO-d6): l 168.09, 141.27,
138.49, 135.46, 133.46, 119.37, 28.95, 21.67, 13.26; MS m/z
378, 376 (2MH+), 359 (2M−H2O+H+), 189 (MH+). Anal.
calcd for C10H14BClN2O: C, 53.50; H, 6.29; B, 4.82; Cl,
15.79; N, 12.48; O, 7.13. Found: C, 53.41; H, 6.21; B, 4.11;
Cl, 16.61; N, 12.48; O, 7.18.
1
(CH3OH) 264 (1.04), 206 (1.47); H NMR (DMSO-d6): l
7.85 (1H, d, J=2.3 Hz), 7.63 (1H, dd, J=2.3 and 8.6 Hz),
7.25 (1H, d, J=8.6 Hz), 4.93 (s, 1H, OH), 2.79 (2H, q,
J=7.7 Hz), 1.42 (3H, t, J=7.7 Hz); 13C NMR (DMSO-
d6): l 168.16, 141.27, 140.89, 135.80, 134.86, 120.79,
120.41, 29.16, 13.12; MS m/z 491, 489, 487 (2M−H2O+H+
), 255, 253 (MH+). Anal. calcd for C9H11BBrClN2O: C,
37.36; H, 3.83; B, 3.74; Br, 27.61; Cl, 12.25; N, 9.68; O,
5.53. Found: C, 37.37; H, 3.65; B, 3.70; Cl, 12.19; N, 9.59;
O, 5.62.
Sample for X-ray analysis: 1 g of 5 was dissolved in 25 mL
of water at 65°C, filtered, and the filtrate was left
overnight. The resulting suspension was filtered, and the
filter cake was dried at 22°C (15 mbar) for 16 h to give 0.5
g of a crystalline solid, which was found to be suitable for
X-ray analysis, mp 233–234°C. Anal. calcd for
Compound 7 (78% yield): mp 188–190°C; IR (KBr) 3405,
3181, 2980, 1644, 1603, 1551, 1520, 1492, 1461, 1392, 823,
751 cm−1; UV: umax (nm) (‡×104) (CH3OH) 303 (6.63), 234
(2.99); 1H NMR (DMSO-d6): l 8.72 (1H, m), 8.01 (1H,
m), 7.88 (2H, m), 7.73 (2H, m), 5.15 (s, 1H, OH), 3.16 (2H,
q, J=7.7 Hz), 1.52 (3H, t, J=7.7 Hz); 13C NMR (DMSO-
d6): l 169.62, 139.88, 137.25, 130.44, 129.81, 128.91,
128.49, 128.15, 124.58, 122.52, 29.05, 13.92; MS m/z 431
(2M−H2O+H+),
225
(MH+).
Anal.
calcd
for
C13H14BClN2O: C, 59.93; H, 5.42; B, 4.15; Cl, 13.61; N,
10.75; O, 6.14. Found: C, 60.06; H, 5.49; B, 4.11; Cl,
13.65; N, 10.79; O, 6.44.