Synthesis of perimidine and 1,5-benzodiazepine derivatives using tamed Br?nsted acid, BF3-H2O

Article abstract of DOI:10.1016/j.jfluchem.2013.03.023

The one pot synthesis of perimidine and 1,5-benzodiazepine derivatives via condensation of aryldiamines with ketones in the presence of BF ₃-H₂O is described. BF₃-H₂O plays a dual role, as an efficient non-oxidizing Br?nsted acid catalyst as well as an effective solvent making the reaction more economic and environmentally safer. The reaction is simple, fast and convenient.

Full text of DOI:10.1016/j.jfluchem.2013.03.023

Accepted Manuscript
Title: Synthesis of Perimidine and 1,5-Benzodiazepine
Derivatives Using Tamed Brønsted Acid, BF -H O
3
2
Author: G.K. Surya Prakash Farzaneh Paknia Arjun Narayan
Thomas Mathew George A. Olah
PII:
S0022-1139(13)00148-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.jfluchem.2013.03.023
FLUOR 8091
To appear in:
FLUOR
Received date:
Revised date:
Accepted date:
27-2-2013
21-3-2013
25-3-2013
Please cite this article as: G.K.S. Prakash, F. Paknia, A. Narayan, T.
Mathew, G.A. Olah, Synthesis of Perimidine and 1,5-Benzodiazepine Derivatives
Using Tamed Brønsted Acid, BF -H O, Journal of Fluorine Chemistry (2013),
3
2
http://dx.doi.org/10.1016/j.jfluchem.2013.03.023
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Synthesis of Perimidine and 1,5-Benzodiazepine Derivatives Using
Tamed Brønsted Acid, BF -H O
3
2
G. K. Surya Prakash,* Farzaneh Paknia, Arjun Narayan, Thomas Mathew,* and
George A. Olah
Loker Hydrocarbon Research Institute and Department of Chemistry, University of
Southern California, Los Angeles, California, 90089-1661
Page 1 of 24

Abstract: The one pot synthesis of perimidine and 1,5-benzodiazepine derivatives via
condensation of aryldiamines with ketones in the presence of BF -H O is described. BF -
3
2
3
H O plays a dual role, as an efficient non-oxidizing Brønsted acid catalyst as well as an
2
effective solvent making the reaction more economic and environmentally safer. The
reaction is simple, fast and convenient.
Key words: Brønsted acid · Diamines · Benzodiazepines · Perimidines · Condensation-
cyclization
2
Page 2 of 24

1
. Introduction
J. B. Conant [1] in 1927 gave the name superacids to acids, capable of protonating
bases that are very weak to form the conjugate acid. Until Olah’s studies focused on the
preparation of stable electron deficient species mainly carbocations using highly acidic
(superacidic) non aqueous solutions, superacidity and related characteristics were not
closely followed [2]. Later, Gillespie has proposed a widely accepted definition for
superacids as acids with acidity greater than 100 percent sulfuric acid [3]. Hammet acidity
of 100% sulfuric acid is -12, and all Brønsted acids having an H ≤ -12 are considered as
0
superacids. Trifluoromethanesulfonic acid (perfluoromethanesulfonic acid, triflic acid,
CF SO H) with Hammet acidity value (H ) of -14.1 is over 100 times stronger than 100%
3
3
0
sulfuric acid. The distinction between an acidic and superacidic medium is more of a
variation in ability to protonate increasingly weak bases by various acidic systems.
Anhydrous hydrogen fluoride (HF), fluorosulfonic acid (FSO H), and teflic acid (TeF OH)
3
5
are examples of fluorine containing Brønsted acids that exceed the acidity of pure sulfuric
acid with H values ranging from -14 to -15. However, many of these acids are very
0
corrossive, difficult to handle and HF, in particular, makes toxic areosols with moisture.
Being strong electron acceptors, Lewis acids cover a wide range of acidities
extending beyond the strength of most frequently used systems as AlCl and BF . In
3
3
extending the concept of superacidity, Olah, and coworkers suggested the use of anhydrous
AlCl , the most widely used Friedel-Crafts catalyst, as the arbitrary unit to define Lewis
3
superacids. Lewis acids stronger than anhydrous AlCl are thus categorized as Lewis
3
superacids [2]. They include among others, the fluorine containing BF , SbF , AsF , TaF ,
3
5
5
5
NbF , B(OSO CF ) , Sb(OTeF ) [4]. Boron trifluoride [5,6] is a well known highly
5
2
3 3
5 5
3
Page 3 of 24

o
o
electron deficient gaseous Lewis acid (m.p. -127 C and b.p. -100 C/ 35 Torr). Since
boron in BF has an empty p orbital and capable of complexation with Lewis n-donor
3
bases, it is highly oxophilic and complexation with protic oxygenated compounds such as
water and alcohols can give rise to strong conjugate Brønsted acid systems [7]. Meerwein
in 1933 found [8] that water can form stable complexes with BF and the system can act as
3
a Brønsted acid. Two types of water complexes are known with BF ; monohydrate (BF -
3
3
H O) and the dihydrate (BF -2H O) complexes. Unlike other boron halides, hydrolysis of
2
3
2
BF is very slow and the water complexation do not lead to hydrolysis easily. The
3
dihydrate complex is very stable, less fuming and its acidity is close to that of 100% nitric
acid. On the other hand, BF -H O 1:1 complex is a strong acidic system with the acidity
3
2
close to that of 100% sulfuric acid (H ~ 12) [9]. BF monohydrate complex is a colorless
o
3
o
fuming liquid with a density of 1.8 g/mL and melting point of 6.2 C. It is less expensive
than trifluoromethanesulfonic acid, easier to prepare and more importantly non-oxidizing
in nature. All these properties of the monohydrate complex, permit acid induced synthetic
transformations, which were previously carried out using expensive superacids. Over the
years, in our continued search for better catalyst system to improve efficacy, experimental
simplicity and economic viability, we have found that BF -H O acts as an excellent
3
2
Brønsted acid catalyst system for a wide range of reactions, such as preparation of alkyl
nitrates, sulfides from carbonyl compounds, thioacetalization of ketones, nitration of
aromatics using potassium nitrate, halogenation of deactivated aromatics using N-
halosuccinimide, Fries rearrangement, hydroxyalkylation of aromatics using benzaldehyde,
dialdehydes, pyridinecarboxaldehydes and quinolinecarboxaldehydes for the preparation of
triarylmethanes,
triarylmethane
aldehydes
and
anthracenes
as
well
as
4
Page 4 of 24

diarylmethylpyridines and diarylmethylquinolines, respectively, under mild conditions
10]. As BF -H O is a highly economic and non-oxidizing acid catalyst and can act as
[
3
2
good solvent for aromatics, we decided to examine the preparation of perimidines and
benzodiazepines using respectively 1,8-naphthalenediamine and 1,2-phenylene diamine
using BF -H O as the catalyst as well as the solvent.
3
2
Perimidines (peri-naphtho-fused pyrimidine ring systems) are of great interest
because they constitute an important class of natural and synthetic products, many of
which exhibit useful biological activity [11]. The perimidine ring system is the core
structure of many biologically active compounds such as spiro indole-perimidines 1 [12],
benzoperimidines 2 [13] and perimidine-1-propanamine 3 [14] used as antitumor agents
(Figure 1). 2,3-Dihydroperimidine derivatives have been used as dyes [15], antioxidant
stabilizers [16], photochromic compounds [17] and catalysts [18].
N
N
2 3 2
(CH ) NEt
HN
N
N
NH
O
N
HO
H
O
NH(CH ) N(CH )
2 2 2 5
1
2
3
Figure 1. Selected perimidines of medicinal importance
Recently, there has been an ongoing interest in the synthesis of perimidine ring
system as well [19]. However, the synthesis of perimidine derivatives using 1,8-
naphthalenediamine is not well explored due to limitations of available reagents, low yields
and the prevalence of side reactions [20a-c].
5
Page 5 of 24

Benzodiazepines (BZD)s and their polycyclic derivatives show a wide range of
biological activities [21]. Benzodiazepines are the most frequently used psychotropic drugs
because of their efficacy, rapid effect and low toxicity. Since the introduction of the first
BZD (Chlordiazepoxide, 4) in 1960, more than 50 BZD drugs have been synthesized for
therapeutic applications [22]. Benzodiazepines have also been extensively used as
anticonvulsant, antianxiety, analgesic, sedative, antidepressant, hypnotic and anti-
inflammatory agents [23a,b,c]. Furthermore, benzodiazepines have commercial use as dyes
for acrylic fibers in photography [23d]. Many bioactive compounds have been synthesized
based on a 1,5-benzodiazepine scaffold such as clozapine 5, the treatment of schizophrenia
since 1982 [24], isoTIBO 6 as HIV-1 non-nucleoside reverse transcriptase inhibitor [25],
imidazo (1,5)-benzodiazepine 7 as an anti HIV agent [26], and benzodiazepine 8 as a
calcium channel blocking agent (Figure 2). Furthermore, 1,5-benzodiazepines are valuable
synthons for the development of various fused ring benzodiazepine systems [27]. Due to
the wide range of clinical and pharmaceutical applications of benzodiazepines, synthesis of
this heterocyclic nucleus has received significant attention.
H CHN
3
CH3
S
N
O
N
H
N
N
N
N Me
HN
N
N
Cl
Cl
6, "isoTIBO"
Cl
4
, Chlordiazepoxide
5, Clozapine
Ph
NO2
N
H
N
N
N
CH3
CO Et
2
N
H
O
Me
CH3
7
8
Figure 2. Selected benzodiazepines of medicinal importance
6
Page 6 of 24

Several methods have been reported for the preparation of 1,5-benzodiazepines [28-
3
0]. Among these methods, the acid catalyzed condensation of o-phenylenediamines with
ketones is one of the simplest and most applicable approach for the synthesis of 1,5-
benzodiazepines. A variety of catalyst system such as BF -Et O [31a], MgO/POCl [31b],
3
2
3
polyphosphoric acid, SiO [31c], InBr [31d], PVP-FeCl [31e], Fe(ClO ) [31f], zeolite
2
3
3
4 3
[31g], p-toluenesulfonic acid [31h], NBS [31i], Sc(OTf)3 [31j], AcOH [31k],
heteropolyacids [31l] and ionic liquids [31m,n] have been used for these reactions. Many
of these processes suffer from various limitations, which includes harsh reaction
conditions, relatively long reaction times, the use of expensive reagents, often low yields,
the occurrence of several side reactions and difficult experimental/work-up procedures.
Herein, we report the application of BF -H O as an effective Brønsted acid catalyst
3
2
and a medium for the synthesis of the heterocycles of high pharmaceutical importanvce
such as dihydroperimidines (12a-f) and 1,5-benzodiazepines through condensation-
cyclization under relatively mild conditions (Figure 3).
R1
NH2 NH2
R
HN
NH
R1
10
O
R1
R
CH2
CH2
1
R
2a-g: R = CH
R
a-g
or
3
1
= CH , C H ,
=
3
2
5
2
9
CH2,
CH2
CH
2
F, CF
H
BF -H O
3
2
R1
R
CH3
H
N
N
N
O
R2
R2
CH
2
F
CH3
N
H
CH3
CH3
NH2
NH2
R2
H
1
14: R1 = CH2F
1
3a-e: R, R = CH
3
11a-e
Figure 3. Synthesis of dihydroperimidines 12, benzodiazepines 13, and
benzimidazoline 14
7
Page 7 of 24

2
. Results and Discussion
2
.1. Synthesis of 2,3-dihydroperimidine in the presence of BF -H O
3
2
Reaction of 1,8-naphthalenediamine with various aliphatic and alicyclic ketones
under mild conditions in the presence of BF -H O afforded 2,3-dihydroperimidines in
3
2
good yields (Figure 4).
R
R1
O
NH2 NH2
HN
NH
R1
BF
3 2
-H O
R
+
RT
9a-g
10
12a-g
R
R = CH
3
R = CH , C H ,
1
3
2
5
R1
2 2
CH F, CF H
R
CH
2
CH2
=
R1
CH2
CH2
Figure 4. BF -H O catalyzed synthesis of 2,3-dihydroperimidines 12a-g
3
2
Thus, the condensation of 1,8-naphthalenediamine with acetone, methyl ethyl ketone
and 2-butanone at room temperature afforded the corresponding perimidine derivatives in
good yields. However, in the case of cyclopentanone and cyclohexanone traces of the
corresponding aldol condensation products were also observed along with the expected
perimidine products (identified by GC-MS). The progress of all reactions was carefully
monitored by TLC and GC-MS and conditions were optimized for high selectivity and
yield. Fluoromethylated (monofluoro and difluoro) dihydroperimidine derivatives were
also prepared in good yields under similar reaction conditions using fluoroacetone
(monofluoro and difluoro). However, reaction with trifluoroacetone was not clean.
8
Page 8 of 24

Table 1. BF -H O catalyzed formation of 2,3-dihydroperimidine derivatives
3
2
Entry
1
Ketone (9a-g)
Product (12a-g)
Time (h) Yield (%)
NH
O
Me
Me
2
98
90
91
NH
Me
Me
Me
Et
9
a
12a
NH
O
Me
Et
2
3
2
NH
9b
12b
NH
O
Et
Et
2
Et
Et
NH
9c
12c
NH
O
Me
4
CH2F
2
66
Me
CH2F
NH
9d
12d
NH
O
Me
2
2
2
82
5
6
CF H
2
Me
CF H
NH
2
9
e
12e
O
NH
93
90
NH
9
f
12f
O
NH
7
NH
9g
12g
2
.2. Synthesis of 1,5-benzodiazepines in presence of BF -H O
3 2
Condensation of various o-phenylenediamines with acetone having α-hydrogens,
under mild conditions in the presence of BF -H O, affords 1,5-benzodiazepines in good to
3
2
excellent yields (Figure 5). As shown in our previous experiments [10], BF -H O acts as
3
2
an efficient catalyst as well as a solvent. In our attempt to make mono, di and
trifluoromethylated diazepines using fluorinated acteones, monofluoroacetone gave
®
fluoromethylated benzimidazoline 14 in 78% yield as observed when Nafion-H was used
9
Page 9 of 24

as a catalyst in our previous studies [32]. Reactions with difluoro and trifluroroacetone
were not clean. The results are summarized in (Table 2).
CH3
N
H N
3 2
BF -H O
R
2
R2
R2
RT - 50 o
O
+
C
R1
H N
N
2
CH3
CH3
H
9
a,d
11a-e
13a-e
1
1
R = CH ; R = CH , CH F
R, R = CH
3
3
2
3
H
N
CH3
CH2F
N
H
14
1
2
R = CH ; R = CH F; R = H
3
2
Figure 5. BF -H O catalyzed synthesis of 1,5-benzodiazepines 13 and benzimidazoline 14
3
2
Reaction of acetone with o-phenylenediamines was carried out at room temperature
Table 2, entries 1-5). Condensation of electron rich o-phenylenediamines, such as 4-
(
methyl-1,2-phenylenediamine
smoothly over a short period of time, 45 min and 1 h, respectively (Table 2, entries 2 and
). However, reaction of deactivated o-phenylenediamines, such as 4-fluoro-1,2-
and
4,5-dimethyl-1,2-phenylenediamine,
proceeded
3
phenylenediamine and 4-(trifluoromethyl)-1,2-phenylenediamine, required a relatively
longer period of time of 3 h and 6 h, respectively (Table 2, entries 4 and 5). The reaction of
4
-(trifluoromethyl)-1,2-phenylenediamine was not as clean as in the case of other o-
phenylenediamines. Reaction of unsymmetrical monosubstituted 1,2-phenylenediamine
with acetone gave rise to two isomeric products in a 1:1 ratio (Table 2, entries 2,4,5).
However, 3,4-diaminopyridine being highly deactivated by protonation of ring nitrogen,
did not react with acetone even at higher temperatures.
1
0
Page 10 of 24

Table 2. BF -H O catalyzed formation of 1,5-benzodiazepines and benzimidazoline
3
2
Entry Ketone (9)
Diamine (11)
Product (13, 14)
Time (h) Yield (%)
CH
3
O
NH2
N
2
1
99
1
H
3
C
CH
3
NH
2
CH
3
N
H
1
9a
11a
CH3
3a
CH3
CH3
N
N
H C
NH2
NH2
3
H C
3
99a
1:1
2
3
CH3
CH3
+
CH3
CH3
H C
N
H
3
N
H
11b
13b
13b'
CH3
H C
NH2
H C
N
3
3
0.75
99
CH3
CH3
H C
NH2
H C
3
3
N
H
11c
13c
CH3
CH
3
N
F
N
F
NH2
NH2
90a,b
1:1
3
4
5
+
CH3
CH
CH
3
F
N
H
N
H
CH3
3
11d
13d
13d'
CH3
CH
3
N
F3C
N
F3C
NH2
NH2
50b
1:1
+
6
CH3
CH3
CH
CH3
3
F3C
N
H
N
H
1
1e
1f
13e
1
3e'
NH2
NH2
N
NO REACTION
H
6
7
7
4
0
1
O
NH2
N
CH2F
78
H C
CH2F
CH3
3
N
H
NH2
9
d
1
1a
14
Reactions with acetone were carried out at room temperature. Yields were not optimized. b Mixture of regioisomers with some
a
unknown byproducts. c Reaction was carried out at 50 C.
o
2
.3. The plausible mechanism for the formation of 1,5-benzodiazepines and 2,3-
dihydroperimidines
The mechanism for the formation of 1,5-benzodiazepine is well known [31f, 31j] and
it involves formation of the intermediate diimine A, followed by its tautomerization and a
,3 proton shift to form the isomeric enamine B, and subsequent intramolecular imine-
1
1
1
Page 11 of 24

enamine electrocyclic cyclization followed by 1,5-proton transfer which leads to the seven
member benzodiazepine product (Figure 6).
Similar to the formation of 1,5-benzodiazepine, probable mechanism [33] for the
formation of perimidine and the benzimidazoline ring involves the classic aminal
formation by the nucleophilic attack of the amine group to the carbonyl group of the
ketone, producing the mono-imine intermediate, followed by cyclization through
nucleophilic attack of the second amine group to the electrophilic carbon center of the
imine group, followed by 1,3-proton transfer which leads to the corresponding perimidine
ring (Figure 6).
R1
O
H
H
NH2
NH2
R
NH2
N
N
N
NH
NH
R
1,3-proton
transfer
R
9a-g
R
R1
R1
3
BF -H
2
O
R1
-
2
H O
1
2 a-g
CH
CH3
1
0
O
O
3
CH3
H C CH3
N
N
NH2
3
N
H C
CH3
9a
3
R2
R2
R2
CH3
CH3
BF
3
-H
2
O
BF
3
-H
2
O
NH
2
NH2
-
H
2
O
- H O
2
11a-e
A
CH3
1
,3-H
shift
CH3
CH3
CH
3
CH3
H
N
CH
N
3
N
R2
R2
R2
CH
3
.
.
1
,5-proton
N
N
H
N
transfer
CH3
CH3
H
1
3 a-e
CH3
B
Figure 6. Plausible mechanism for the formation of 2,3-dihydroperimidines 12a-g and
,5-benzodiazepines 13a-e
1
1
2
Page 12 of 24

3
. Experimental
3
.1. General
Unless otherwise mentioned, all chemicals were purchased from commercial sources.
1
13
19
BF -H O was prepared by the method described below. H, C, and F NMR spectra
3
2
1
were recorded on 400 MHz and 500 NMR spectrometer. H NMR chemical shifts were
1
3
determined relative to internal tetramethylsilane at 0.0 ppm, C NMR chemical shifts were
13
19
determined relative to the C signal of CDCl at δ 77.0 ppm. F NMR chemical shifts
3
were determined relative to the signal of internal standard CFCl at δ 0.0 ppm. GC-MS
3
spectra were recorded on a ThermoFinnigan Mass (EI) spectrometer. HRMS data were
obtained from a high resolution Micromass GCT (GC-MS TOF) spectrometer.
3
.2. Typical procedure for the preparation of BF -H O (1:1) complex
3 2
Caution: BF forms toxic areosol clouds. Therefore, the whole procedure was carried
3
out in a well ventilated hood. In a Nalgene bottle (500 mL wide mouthed polypropylene
bottle from VWR), distilled water (36 g, 2 mol) was taken and fitted with a cap equipped
with an inlet reaching the bottom. BF was passed carefully into the bottle with frequent
3
o
‡
cooling intermittently (at -78 C) and shaking until a desired weight increase was
observed. The amount of BF needed is 134 g, which corresponds to a ratio of 1:1.
3
‡
Since the formation of BF
3
-H
2
O was highly exothermic, Nalgene bottle was placed in a dry-ice acetone bath intermittently when BF
became difficult, the bottle was taken out of the bath
intermittently and was placed back in the dry-ice acetone bath after the melting took place. The bottle cap was occasionally opened to
release the pressure of excess BF
3
was bubbled into water. When the complex solidified and bubbling of BF
3
3
.
1
3
Page 13 of 24

3
.3. General procedures for the BF -H O catalyzed reactions of 1,8-naphthalenediamine
3 2
and o-phenylenediamines and with ketones for the synthesis of 2,3-dihydroperimidines and
1
,5-benzodiazepines
The aryldiamines (2.0 mmol) were dissolved in excess of ketone (12 mmol) in a
o
pressure tube. The mixture was cooled to 0 C and BF -H O (5 g, 58 mmol) was added
3
2
slowly. The mixture was stirred for the required period of time (Table 1, 2) at room
temperature. Progress of the reaction was monitored by TLC (5:3 hexane/ethyl acetate) and
GC-MS. After the reaction was complete, the mixture was poured over ice (25 g),
neutralized with sodium bicarbonate and extracted with CH Cl (3 x 15 mL). The organic
2
2
extracts were combined, washed with water and dried over anhydrous Na SO . The solvent
2
4
was removed by vacuum evaporation and the crude products were purified with column
chromatography on silica gel (70-230 mesh) using hexane/ethyl acetate as eluent. All
products were characterized by spectral analysis (NMR and GC-MS) and the spectral data
were compared with those of the authentic samples in the case of known compounds 12a-
c, 12f, 12g, 13a-e and 14 [20a, 31, 32] as well as by HRMS analysis in the case of new
products 12d and 12e.
3
.4. Spectral data of products
2
,2-Dimethyl-2,3-dihydro-1H-perimidine (12a)
1
H NMR (400 MHz, CDCl ):  1.33 (s, 6H), 4.10 (brs, 2H), 6.37 (dd, J = 7.32 Hz, J = 1.10
3
Hz, 2H), 7.11 (dd , J = 8.42 Hz, J = 0.92 Hz, 2H), 7.19 (dd, J = 8.24 Hz, J = 7.32 Hz, 2H);
13
C NMR (100.5 MHz, CDCl ):  28.42, 64.25, 105.57, 112.63, 116.62, 126.87, 134.39,
3
+
1
40.26; GC-MS (EI), m/z, 197.5 (M ).
1
4
Page 14 of 24

2
-Ethyl-2’-methyl-2,3-dihydro-1H-perimidine (12b)
1
H NMR (400 MHz, CDCl ):  0.95 (t, J = 7.51 Hz, 3H), 1.40 (s, 3H), 1.72 (q, J = 7.51
3
Hz, 2H), 4.12 (brs, 2H), 6.44 (dd, J = 7.32 Hz, J = 0.733 Hz, 2H), 7.11 (dd , J = 8.06 Hz, J
1
3
=
0.75 Hz, 2H), 7.20 (t, J = 7.32 Hz, 2H); C NMR (100.5 MHz, CDCl ):  8.12, 26.11,
3
3
3.25, 66.73, 105.69, 112.95, 116.82, 127.00, 134.58, 140.23; GC-MS (EI), m/z, 212.3
+
(
M ).
2
,2-Diethyl-2,3-dihydro-1H-perimidine (12c)
1
H NMR (400 MHz, CDCl ):  0.89 (t, J = 7.51 Hz, 6H), 1.65 (q, J = 7.51 Hz, 4H), 4.13
3
(brs, 2H), 6.40 (dd, J = 7.32 Hz, J = 0.92 Hz, 2H), 7.08 (dd , J = 8.42 Hz, J = 0.92 Hz, 2H),
1
3
7
.19 (dd, J = 8.24 Hz, J = 7.32 Hz, 2H); C NMR (100.5 MHz, CDCl ):  7.67, 30.39,
3
+
6
8.75, 105.33, 112.73, 116.40, 127.01, 134.54, 140.36; GC-MS (EI), m/z, 226.3 (M ).
2
-(Difluoromethyl)-2-methyl-2,3-dihydro-1H-perimidine (12d)
1
Dark green oil. H NMR (500 MHz, CDCl ) δ = 1.43 (t, J = 1.7 Hz, 3H), 4.44 (s, 2H), 5.61
3
(t, J = 56.8 Hz, 1H), 6.44 (dd, J = 7.2, 1.1 Hz, 2H), 7.10 (dd, J = 8.3, 1.1 Hz, 2H), 7.14 –
13
7
.18 (m, 2H); C NMR (100.5 MHz, CDCl ) δ 19.54 (t, J = 1.4 Hz), 66.37 (t, J = 24.9
3
1
9
Hz), 106.02, 112.14, 113.69 (t, J = 250.8 Hz), 118.06, 127.31, 134.50, 138.02; F NMR
(
470 MHz, CDCl ) δ = -133.65 (d, J = 56.5 Hz); HRMS (EI) m/z calcd for C H F N
3
13 12
2
2
2
34.0969, observed 234.0978.
2
-(Fluoromethyl)-2-methyl-2,3-dihydro-1H-perimidine (12e)
1
Pink needles. H NMR (500 MHz, CDCl ) δ = 1.59 (d, J = 1.8 Hz, 3H), 4.27 (d, J = 47.5
3
Hz, 2H), 4.52 (s, 2H), 6.54 (dd, J = 7.3, 1.0 Hz, 2H), 7.20 (dd, J = 8.3, 1.0 Hz, 2H), 7.26
13
(
dd, J = 8.3, 7.2 Hz, 2H); C NMR (126 MHz, CDCl ) δ = 24.85, 64.98 (d, J = 21.1 Hz),
3
1
9
8
4.76 (d, J = 179.4 Hz), 106.30, 112.50, 117.76, 127.07, 134.51, 138.72; F NMR (376
1
5
Page 15 of 24

MHz, CDCl ) δ = -224.06 (td, J = 47.3, 2.2 Hz); HRMS (EI) m/z calcd for C H FN ,
3
13 13
2
2
16.1063, observed 216.1060.
Spiro[cyclopentane-1,2'(3'H)-[1H]perimidine] (12f)
1
H NMR (400 MHz, CDCl ):  1.7-1.8 (m, 8H), 4.22 (brs, 2H), 6.45 (dd, J = 7.15 Hz, J =
3
1
3
0
.80 Hz, 2H), 7.14 (dd , J = 8.20 Hz, J = 0.80 Hz, 2H), 7.20 (t, J = 7.70 Hz, 2H); C NMR
(
100.5 MHz, CDCl ):  22.88, 39.14, 75.94, 106.22, 113.69, 117.18, 126.78, 134.62,
3
+
1
40.91; GC-MS (EI), m/z, 224.4 (M ).
Spiro[cyclohexane-1,2'(3'H)-[1H]perimidine] (12g)
1
H NMR (400 MHz, CDCl ):  1.43-1.49 (m, 2H), 1.54-1.60 (m, 2H), 1.69-1.71 (m, 2H),
3
4
7
3
.35 (brs, 2H), 6.46 (dd, J = 7.3 Hz, J = 1.0 Hz, 2H), 7.11 (dd , J = 8.4 Hz, J = 1.0 Hz, 2H),
1
3
.21 (dd, J = 8.2 Hz, J =7.3 Hz, 2H); C NMR (100.5 MHz, CDCl ):  22.06, 24.94,
3
6.45, 65.14, 105.78, 113.17, 116.72, 126.86, 134.46, 139.79; GC-MS (EI), m/z, 239.5
+
(
M ).
2
,2,4-Trimethyl-2,3-dihydro-1H-1,5-benzodiazepine (13a)
1
H NMR (400 MHz, CDCl ):  1.31 (s, 6H), 2.21 (s, 2H), 2.35 (s, 3H), 3.42 (brs, 1H, NH),
3
1
3
6
.69-6.72 (m, 1H), 6.95-6.98 (m, 2H), 7.12-7.16 (m, 1H); C NMR (100.5 MHz, CDCl ):
3

29.55, 30.24, 44.92, 68.52, 121.61, 121.81, 125.51, 126.48, 137.60, 140.44, 172.33; GC-
+
MS (EI), m/z, 188.5 (M ).
2
,2,4,7-Tetramethyl-2,3-dihydro-1H-1,5-benzodiazepine (13b) & 2,2,4,8-tetramethyl-2,3-
dihydro-1H-1,5-benzodiazepine(13b’)
1
H NMR (400 MHz, CDCl ):  1.27 (s, 6H), 1.28 (s, 6H), 2.15 (s, 2H), 2.18 (s, 2H), 2.24
3
(s, 3H), 2.26 (s, 3H), 2.32 (s, 3H), 2.33 (s, 3H), 6.50 (d, J = 0.92 Hz, 1H), 6.62 (d, J = 7.87
1
6
Page 16 of 24

1
3
Hz, 1H), 6.74-6.78 (m, 2H), 6.94 (d, J = 1.65 Hz, 1H), 7.02 (d, J = 7.87 Hz, 1H),
C
NMR (100.5 MHz, CDCl ):  20.24, 20.50, 29.31, 29.83, 30.05, 44.66, 44.81, 67.44,
3
6
1
8.16, 121.48, 121.71, 122.22, 122.78, 125.77, 126.37, 126.50, 131.142, 134.88, 135.07,
+
37.43, 140.47, 150.08, 171.36, 172.39; GC-MS (EI), m/z, 202.4 (M ).
2
,2,4,7,8-Pentamethyl--2,3-dihydro-1H-1,5-benzodiazepine (13c)
1
H NMR (400 MHz, CDCl ):  1.31 (s, 6H), 2.17 (s, 3H), 2.18 (s, 3H), 2.19 (s, 2H), 2.33
3
1
3
(
s, 3H), 6.52 (s, 1H), 6.91 (s, 1H); C NMR (100.5 MHz, CDCl ):  19.13, 19.48, 29.98,
3
3
0.60, 45.46, 68.34, 123.10, 127.96, 130.29, 134.01, 135.84, 138.66, 172.10; GC-MS (EI),
+
m/z, 216.6 (M ).
7
-Fluoro-2,2,4-trimethyl-2,3-dihydro-1H-1,5-benzodiazepine (13d) and 8-fluoro-2,2,4-
trimethyl-2,3-dihydro-1H-1,5-benzodiazepine (13d’)
1
H NMR (400 MHz, CDCl ):  1.28 (s, 6H), 1.30 (s, 6H), 2.14 (s, 2H), 2.22 (s, 2H), 2.30
3
(s, 3H), 2.33 (s, 3H), 2.92 (brs, 2H, NH),6.39 (dd, J = 9.80 Hz, J = 2.65 Hz, 1H), 6.59-6.67
1
9
(
m, 3H), 6.79-6.82 (m, 1H), 7.03 (dd, J = 8.79 Hz, J = 6.23 Hz, 1H), F NMR (376 MHz,
+
CDCl ):  -118.27 (m), -122.17 (m); GC-MS (EI), m/z, 206.4 (M ).
3
2
-Fluoromethyl-2-methylbenzimidazoline (14)
1
H NMR (400 MHz, CDCl ):  1.54 (d, J = 2.1 Hz, 3H), 3.87 (brs, 2H, NH), 4.24 (d, J =
3
1
3
4
7.43 Hz, 2H), 6.57-6.71 (dm, 4H); C NMR (100.5 MHz, CDCl ):  24.22, (d, J = 19.9
3
1
9
Hz), 78.5 (d, J = 20.6 Hz), 87.23 (d, J = 177.4 Hz), 110.20, 120.64, 139.51; F NMR (376
MHz, CDCl ):  -223.47 (t, J = 47.8)
3
1
7
Page 17 of 24

4
. Conclusion
In summary, we have described a simple, efficient and environmentally friendly
method for the synthesis of 2,3-dihydro-1,5-benzodiazepines and 2,3-dihydroperimidines
in a single step with high yields and purity in most cases using BF -H O. The dual role of
3
2
BF -H O as the catalyst as well as solvent make the system more economical and reduces
3
2
amount of organic toxic waste and avoids the need of other expensive strong acids. Non-
oxidizing character of BF -H O helps to avoid many other side reactions involving
3
2
oxidation and makes the reaction relatively clean.
References
[1] N.F. Hall, J.B. Conant, J. Am. Chem. Soc. 49 (1927) 3047-3061.
[
2] G.A. Olah, G.K.S. Prakash, Á. Molnár, J. Sommer, Superacid Chemistry, Second
Ed.; John Wiley & Sons, Inc., Hoboken, New Jersey, 2009.
[
[
[
3] (a) R.J. Gillespie, T.E. Peel, E.A. Robinson, J. Am. Chem. Soc. 93 (1971) 5083-
5
087; (b) R.J. Gillespie, Peel, T. E. J. Am. Chem. Soc. 95 (1973) 5173-5178.
4] K.O. Christe, D.A. Dixon, D. McLemore, W.W. Wilson, J.A.Sheehy, J.A. Boatz, J.
Fluorine Chem. 101 (2000) 151-153.
rd
5] (a) D. R. Martin, in Kirk-Othmer Encyclopedia of Chemical Technology, 3 Ed.;
Grayson, M. Ed.; Wiley, New York, 1980, Vol. 10, pp 685-693; (b) N. Greenwood,
B.S. Thomas, in Comprehensive Inorganic Chemistry; A.F. Trotman-Dickenson, Ed.;
Pergamon, New York, 1973, Vol. 1, pp 956-973.
[6] H. Heaney, in Encyclopedia of Reagents for Organic Synthesis, Vol. 1, pp 651-664,
L.A. Paquette, Ed.; Wiley: Chichester and references therein.
1
8
Page 18 of 24

[
7] (a) H.D. Hinton, J.A. Nieuwland, J. Am. Chem. Soc. 54 (1932) 2017-2018; (b) G.F.
Hennion, H.D. Hinton, J.A. Nieuwland, J. Am. Chem. Soc. 55 (1933) 2857-2860; (c)
F.J. Sowa, J.A. Nieuwland, J. Am. Chem. Soc. 55 (1933), 5052-5053; (d) T.B.
Dorris, F.J. Sowa, J.A. Nieuwland, J. Am. Chem. Soc. 56 (1934) 2689-2690.
8] H. Meerwein, Chem. Ber. 66 (1933) 411-414.
[
[
[
9] D. Faracasiu, A. Ghenciu, J. Catal. 134 (1992) 126-133.
10] (a) G.A. Olah, Q. Wang, X.Y. Li, I. Bucsi, Synthesis (1992) 1085-1086; (b) G.A.
Olah, Q. Wang, N. Trivedi, G.K.S. Prakash, Synthesis (1992) 465-466; (c) G.A.
Olah, Q. Wang, X-Y. Li, G.K.S. Prakash, Synlett (1993) 32-34; (d) G.A. Olah, Q.
Wang, X.Y. Li, G.K.S. Prakash, Synthesis (1993) 207-208; (e) G.K.S. Prakash, T.
Mathew, D. Hoole, P.M. Esteves, Q. Wang, G. Rasul, G.A. Olah, J. Am. Chem. Soc.
1
26 (2004) 15770-15776; (f) G.K.S. Prakash, C. Panja, T. Mathew, G.A. Olah, Catal.
Lett. 114 (2007) 24-29; (g) G.K.S. Prakash, F. Paknia, S. Chacko, T. Mathew, G.A.
Olah, Heterocycles 76 (2008) 783-799; (h) G.K.S. Prakash, C. Panja, A. Shakhmin,
E. Shah, T. Mathew, G.A. Olah, J. Org. Chem. 74 (2009) 8659-8668.
[11] (a) K. Undheim, T. Benneche, Comprehensive Heterocyclic Chemistry, A.R.
Katritzky, C.W. Rees, E.F.V. Scriven, Eds.; Pergamon, Oxford, 1984, Vol. 6, p. 93.
(
c) I.A.S. Smellie, A. Fromm, S.A. Moggach, R.M. Paton, Carbohydrate Res. 346
2011) 43-49.
(
[12] K. Arya, A. Dandia, Lett. Org. Chem. 4 (2007) 378-383.
[
13] M. Dzieduszycka, S. Martelli, M. Arciemiuk, M.M. Bontemps-Gracz, A. Kupiec, E.
Borowski, Bioorg. Med. Chem. 10 (2002)1025-1035.
[14] J.M. Herbert, P.D. Woodgate, W.A. Denny, J. Med. Chem. 30 (1987) 2081-2086.
1
9
Page 19 of 24

[15] (a) W.D. Ramsden, L.F. Valente, L.S. Bernard, US Patent (2002) 6348592; (b) M.
Akram, Eur. Pat. Appl. (1996) 703223.
[16] R.F. Malherbe, US Patent (1983) 4389321.
[
17] (a) R. Davis, N. Tamaoki, Org. Lett. 7 (2005)1461-1464; (b) R. Davis, N. Tamaoki,
Chem. Eur. J. 13 (2007) 626-631.
[
18] (a) P. Bazinet, T.-G. Ong, J.S. O’Brien, N. Lavoie, E. Bell, G.P.A. Yap, I. Korobkov,
D.S. Richeson, Organometallics 26 (2007) 2885-2895; (b) B. Alici, I. Ozdemir, K.
Karaaslan, E. Cetinkaya, B. Cetinkaya, J. Mol. Catal. A: Chem 231 (2005) 261-264.
19] I. Yavari, M. Adib, F. Jahani-Moghaddam, H.R. Bijanzadeh, Tetrahedron 58 (2002)
[
6
901-6906.
[
20] (a) J. Zhang, S. Zhang, Synth. Commun. 37 (2007) 2615-2624; (b) S.-L. Zhang, J.-M.
Zhang, Chinese J. Chem. 26 (2008) 185-189; (c) R.R. Reddy, C.V.C. Rao, Ind J
Heterocycl. Chem. 2 (1993) 199-202.
[
21] (a) H. Schutz Benzodiazepines, Springer, Heidelberg, 1982. (b) J.K. Landquist In
Comprehensive Heterocyclic Chemistry, A.R. Katritzky, C.W. Rees, Eds.;
Pergamon: Oxford, 1984, Vol. 1, pp. 166. (c) R.I. Fryer, Bicyclic Diazepines, in: The
Chemistry of Heterocyclic Compounds, Taylor EC, Ed.; Wiley: New York, Vol. 50,
1
991, Chap. II.
[
22] G. Hancu, A. Gaspar, A. Gyeresi, J. Biochem. Biophys. Methods 69 (2007) 251-259.
23] (a) S. Ha, D. Shobha, E. Moon, M. A. Chari, K. Mukkanti, S.-H. Kim, K.-H. Ahn, S.
Kim, Bioorg. Med. Chem. Lett. 20 (2010) 3969-3971; (b) L.O. Randall, B. Kappel
Benzodiazepines, S. Garattini, E. Mussini, L.O Randall, eds.; Raven Press, New
York, 1973, p. 27; (c) J.R. De Baun, F.M. Pallos, D.R. Baker, US Patent (1976)
[
2
0
Page 20 of 24

3
978227; Chem Abstr 86 (1977) 5498d; (d) R.C. Haris, J.M. Straley, US Patent
(1968) 1537757; Chem Abstr 73 (1970) 100054w.
[
24] J. Su, H. Tang, B.A. McKittrick, D.A. Burnett, H. Zhang, A. Smith-Torhan, A.
Fawzi, J. Lachowicz, Bioorg. Med. Chem. Lett. 16 (2006) 4548-4553.
25] K.A. Parker, A. Dermatakis, J. Org. Chem. 62 (1997) 4164-4167.
26] L. Kosychova, L. Pleckaitiene, Z. Staniulyte, R. Janciene, A. Palaima, B.D.
Puodziunaite, ARKIVOC xiii (2006) 158-164.
[
[
[
[
[
[
[
27] M.C. Aversa, A. Ferlazzo, P. Giannetto, F.H. Kohnke, Synthesis (1986) 230-231.
28] M. Kodomari, T. Noguchi, T. Aoyama, Synth. Commun. 34 (2004) 1783-1790.
29] W. Zhong, Y. Zhang, X. Chen, Tetrahedron Lett. 42 (2001) 73-75.
30] M. Kidwai, P. Mothsra, Synth. Commun. 36 (2006) 817-824.
31] (a) J.A.L. Herbert, H.J. Suschitzky Chem. Soc. Perkin Trans. 1 (1974) 2657-2661;
(b) M.S. Balakrishnaa, B. Kaboudinb, Tetrahedron Lett. 42 (2001) 1127-1129; (c)
D.I. Jung, T.W. Choi, Y.Y. Kim, I.S. Kim, Y.M. Park, Y.G. Lee, D.H. Jung, Synth.
Commun. 29 (1999) 1941-1951; (d) J.S. Yadav, B.V.S. Reddy, S. Praveenkumar, K.
Nagaiah, Synthesis (2005) 480-484; (e) M.A. Chari, K. Syamasundar, Catal.
Commun. 6 (2005) 67-70; (f) M.M. Heravi, V. Zadsirjan, F.K. Behbahani, H.A.
Oskooie, J. Mol. Catal. A: Chem. 259 (2006) 201-206; (g) A. Hegedus, Z. Hell, A.
Potor, Catal. Lett. 105 (2005) 229-232; (h) K.P. Guzen, R. Cella, H.A. Stefani,
Tetrahedron Lett. 47 (2006) 8133-8136; (i) C.-W. Kuo, S.V. More, C.-F. Yao,
Tetrahedron Lett. 47 (2006) 8523-8523; (j) S.K. De, R.A. Gibbs, Tetrahedron Lett.
4
6 (2005) 181101813; (k) M. Pozarentzi, J. Stephanidou-Stephanatou, C.A.
Tsoleridis, Tetrahedron Lett. 43 (2002) 1755-1778; (l) M.M. Heravi, F. Derikvand,
2
1
Page 21 of 24

L. Ranjbar, F.F. Bamoharram, J. Mol. Catal. A: Chem. 261 (2007) 156-159; (m) D.V.
Jarikote, S.A. Siddiqui, R. Rajagopal, T. Daniel, R.J. Lahoti, K.V. Srinivassan
Tetrahedron Lett. 44 (2003) 1835-1838; (n) R. Jain, T. Yadav, M. Kumar, A.K.
Yadav, J. Heterocyclic Chem. 47 (2010) 603-610.
[32] G.K.S. Prakash, H. Vaghoo, C. Panja, A. Molnar, T. Mathew, G.A. Olah, Synthesis
(2008) 897-902.
[
33] G.K.S. Prakash, T. Mathew, C. Panja, H. Vaghoo, K. Venkataraman, G.A. Olah, Org.
Lett. 9 (2007) 179-182.
2
2
Page 22 of 24

1
Graphical Abstract
Synthesis of dihydroperimidines and their fluorinated derivatives has been achieved from the correspondiing
naphthalene diamines and acetone, butanone and mono and difluoroacetones using BF -H O, an efficient
3
2
Brønsted acid catalyst as well as the solvent. 1,5-Benzodiazepines were also prepared by similar strategy
from phenylene diamines. Non-oxidising conjugate Brønsted acid is asystem BF -H O is a good substitute
3
2
for other expensive acid catalysts and make the method efficient and more economical.
R1
NH
2
NH
2
R
HN
NH
R1
R
10
O
O
R1
R
CH2
CH2
CH2
1
R
2a-f: R = CH3
1
= CH
CH
3
, C
F, CF
2
H
5
,
=
9
a-f
2
CH ,
BF -H O
2
2
H
3
2
or
CH
3
R1
H
N
N
R2
R2
CH
CH3
2
F
R
NH2
N
H
N
CH3
R2
CH3
H
NH2
1a-e
13a-e: R, R1
= CH
14: R1 = CH
F
2
3
1
2
3
Page 23 of 24

Highlights
BF -H O, a non-oxidizing conjugate Brønsted acid system plays a dual role, as an efficient
3
2
catalyst as well as a good solvent for acid catalyzed reactions.
Many reactions which require superacidic conditions occur smoothly in BF -H O.
3
2
Application of BF -H O is extended towards the synthesis of perimidines and 1,5-
3
2
benzodiazepines under mild conditions.
2
4
Page 24 of 24