1-Bromo-1,2-dicyanoethylene: A dicyanoacetylene synthon useful in tetraazaporphyrin synthesis

Article abstract of DOI:10.1142/S1088424619501165

Dicyanoacetylene, a highly reactive dienophile, readily reacts to form substituted maleonitriles which serve as precursors to soluble tetraazaporphyrins. Unfortunately, dicyanoacetylene is difficult to synthesize in large quantities. We report here a new dicyanoacetylene synthon, 1-bromo-1,2-dicyanoethylene, which is easily made in high yield by reaction of elemental bromine with fumaronitrile and easily purified by vacuum distillation. The resulting 5:1 mixture of geometric isomers reacts with dienes via a Diels-Alder reaction followed by elimination of HBr to give substituted maleonitriles. We demonstrate the utility of these reactions by reporting the synthesis of a novel, soluble tetraazaporphyrin bearing four 1,4-fused cyclohexyl groups on the macrocycle periphery.

Full text of DOI:10.1142/S1088424619501165

Journal of Porphyrins and Phthalocyanines
J. Porphyrins Phthalocyanines 2019; 23: 1092–1099
DOI: 10.1142/S1088424619501165
Published at http://www.worldscinet.com/jpp/
1
-Bromo-1,2-dicyanoethylene: A dicyanoacetylene
synthon useful in tetraazaporphyrin synthesis
◊
Jeffrey P. Fitzgerald* , John P. Korenak, Delano A. Steinaker,
Logan M. Swogger and Jessica A. Lois
Department of Chemistry, United States Naval Academy, 572 Holloway Road, Annapolis, MD, 21402, USA.
Received 13 June 2019
Accepted 8 August 2019
ABSTRACT: Dicyanoacetylene, a highly reactive dienophile, readily reacts to form substituted
maleonitriles which serve as precursors to soluble tetraazaporphyrins. Unfortunately, dicyanoacetylene is
difficult to synthesize in large quantities. We report here a new dicyanoacetylene synthon, 1-bromo-1,2-
dicyanoethylene, which is easily made in high yield by reaction of elemental bromine with fumaronitrile
and easily purified by vacuum distillation. The resulting 5:1 mixture of geometric isomers reacts with
dienes via a Diels–Alder reaction followed by elimination of HBr to give substituted maleonitriles. We
demonstrate the utility of these reactions by reporting the synthesis of a novel, soluble tetraazaporphyrin
bearing four 1,4-fused cyclohexyl groups on the macrocycle periphery.
KEYWORDS: 1-bromo-1,2-dicyanoethylene, 1-bromo-1,2-dicyanoethene, bromofumaronitrile,
bromomaleonitrile, dicyanoacetylene, synthon, equivalent, synthesis, tetraazaporphyrin, porphyrazine,
1
,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-hexadecahydro-1,4:8,11:15,18:22,25-tetraethano-29H,31H-
phthalocyanine.
R
R
R
R
INTRODUCTION
N
M
N
R
R
R
R
N
N
Tetraazaporphyrins (TAPs) are synthetic, planar,
aromatic ligands composed of four pyrrole rings bridged
by aza-nitrogen atoms and capable of binding a wide
variety of metal cations. TAPs are of interest due to
their structural similarity to ubiquitous and well-studied
porphyrins and phthalocyanines [1].
Tetraazaporphyrins are easily made by cyclizing
four equivalents of substituted maleonitrile (i.e. cis-1,2-
dicyanoethene) around a metal ion template, as shown in
Fig. 1 [2]. It should be noted that fumaronitriles (i.e. trans-
R
R
N
N
N
N
4
N
N
Fig. 1. Cyclocondensation of four equivalents of substituted
maleonitrile to a soluble tetrazaporphyrin
There are three general methods reported in the
literature for making the substituted 1,2-dicyanoethylene
precursor to soluble TAPs, as shown in Fig. 2. Oxidative
dimerization of phenylacetonitrile was first reported by
Linstead in his original paper on tetraazaporphyrins. This
method has been extended to other benzylnitriles but is
limited in that the solubilizing group must be aromatic [3].
Further, it produces a mixture of cis and trans products.
While the cis and trans products may be separated, it is
often impossible to photoisomerize the trans product to
1,2-dicyanoethene), which are capable of isomerizing
under the cyclization conditions, can be used to make
TAPs, but yields are generally lower. As is the case with
porphyrins and phthalocyanines, peripheral substituents
are necessary to solubilize the resulting TAP macrocycle.
◊
SPP full member in good standing.
*
6
Correspondence to: Jeffrey P. Fitzgerald, tel.: 1-410-293-
343, email: fitzgera@usna.edu.
Copyright © 2019 World Scientific Publishing Company

1-BROMO-1,2-DICYANOETHYLENE 1093
CN
CN
A third dicyanoacetylene synthon is 1-chloro-1,2-
dicyanoethylene which Luk’yanets et al. [12] reacted
with anthracene in their synthesis of dibenzobarr-
elenoporphyrazine. Presumably an initial Diels–Alder
product undergoes dehydrochlorination to give the
desired maleonitrile. There are two reported syntheses
of 1-chloro-1,2-dicyanoethylene in the literature
and both involve chlorination of succinonitrile. Cass
used gaseous chlorine as the chlorinating agent [13]
while Shevchenko and Kuhkar used phosphorous
pentachloride [14]. Neither of these chlorinating agents
are easy to work with. As noted by other authors [15],
it is difficult to control the reaction stoichiometry with
gaseous chlorine and the atom efficiency of phosphorus
pentachloride is poor. Exhaustive efforts were made to
reproduce these syntheses but, in our hands, yields were
consistently low and the selectivity was poor. Further,
we saw no improvements in either yield or selectivity
when using sulfuryl chloride, another common
chlorinating agent.
We report here the synthesis, characterization and
reactivity of 1-bromo-1,2-dicyanoethylene, a new and
convenient dicyanoacetylene synthon. As shown in
Fig. 3, replacing succinonitrile with fumaronitrile, a
more reactive carbon precursor, facilitated formation of
the desired product. Further, using elemental bromine as
the oxidizing agent made it easier to control the reaction
stoichiometry and improved the atom efficiency. The
resulting product, a 5:1 mixture of bromofumaronitrile
and bromomaleonitrile, can be used as a dicyanoactylene
equivalent without further purification.
CN
R
-
CH O , I
3
2
2
Ar +
CN
Ar
Ar
+
Ar
∆
CN
CN
Ar
CN
Br2 CuCN hv
DMF, ∆
R
CN
R
R
R
CN
R
CN
CN
CN
∆
+
CN
Fig. 2. Three published synthetic routes to substituted
maleonitriles
produce more cis isomer due to a competing ring closure
to the dihydrophenathrene [4].
A second, more general method, reported by
Fitzgerald et al. [5], involves the reaction of alkynes
with bromine to produce trans-dibromoalkenes. These
products are reacted in a Rosenmund-von-Braun-
like reaction to produce substituted fumaronitriles
which are photoisomerized to equilibrium mixtures
of fumaronitriles and maleonitriles. This three-step
process has the benefit of being able to make the
desired product with almost any R group (Ar, R, H),
but it also produces a mixture of dinitrile geometric
isomers which must be separated. Typically, the
separated fumaronitrile may be photoisomerized to
produce more maleonitrile.
The third general route to substituted maleonitriles is
the Diels–Alder reaction between dienes and dicyano-
acetylene [6]. This method produces only maleonitriles,
which is a major advantage over the previous two
syntheses. Unfortunately, dicyanoacetylene, which is
made by the solid-state dehydration of acetylene dicar-
boxamide [7], is difficult to make in quantities greater
than 1 gram. This difficulty severely limits the utility of
this approach.
Other authors have noted the difficulty of synthesizing
dicyanoacetylene and have sought synthetic alternatives.
Zavitsanos patented a direct synthesis by passing nitrogen
gas over graphite heated at 3000 K but was also limited to
small scale production [8]. Ciganek and Krespan reported
the use of phosgene to synthesize 4,5-dicyano-1,3-dithiol-
Br
CN
CHCl3
∆
CN
+
Br2
NC
NC
Br
not
isolated
HBr
NC
Br
H
NC
Br
CN
H
+
+
CN
Fig. 3. Synthesis of 1-bromo-1,2-dicyanoethylene, a novel
dicyanoacetylene synthon
2-one which is then pyrolyzed to dicyanoacetylene in the
gas phase [9]. Another synthetic alternative is dimethyl
acetylenedicarboxylatewhichwasusedbyOliverandSmith
We demonstrate the utility of the 1-bromo-1,2-
dicyanoethene mixture as a dicyanoacetylene synthon by
reporting its Diels–Alder reactions with three different
dienes to produce substituted maleonitriles. Finally,
we report the conversion of one of these maleonitriles
into a novel tetraazaporphyrin bearing four 1,4-fused
cyclohexyl units. The fused cyclohexyl groups prevent
p–p stacking of the macrocycle in solution and thereby
enhance its solubility.
[
[
10] to make magnesium dibenzobarrelenoporphyrazine
11]. Dimethyl acetylenedicarboxylate is reacted with
a diene and the resulting Diels–Alder adduct diester is
converted into a diamide and then dehydrated to a dinitrile.
While the overall yield and scalability are better than when
using dicyanoacetylene, this method adds at least two
synthetic steps after the Diels–Alder reaction.
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1093–1099

1
094 J. P. FITZGERALD ET AL.
1
3
0.57; H, 0.72; N, 17.74. H NMR (400 MHz, CDCl ):
EXPERIMENTAL
3
1
3
δ, ppm 6.80 (s). C NMR (400 MHz, CDCl ): δ, ppm
3
General
-1
1
(
1
17.91, 113.68, 113.27, 108.73. IR (ATR): n, cm 3165
m), 3027 (s), 2845 (m), 2238 (m), 1713 (m), 1588 (s),
All reagents were purchased from commercial sources
and used without further purification. UV-vis spectra
were recorded on a Hewlett-Packard 8452A diode array
spectrometer. IR spectra were recorded on a Nicolet iS5
spectrometer with an iD5 ATR attachment from Thermo
Scientific Corporation. NMR spectra were recorded
on a JEOL ECX-400 NMR spectrometer and chemical
shift values are reported relative to residual protons in
the solvent as an internal standard. Mass spectra were
recorded either on an Agilent 6890N gas chromatograph
with a 5975 mass selective detector or a ShimadzuAxima
MALDI-TOF mass spectrometer (without using any
matrix). Elemental analyses were conducted by Midwest
Microlabs of Indianapolis, IN.
404 (m), 1271 (s), 1100 (s), 1062 (s), 1030 (s), 578 (s).
GC-MS: m/z 158 [M Br], 156 [M Br], 77 [M-Br], 76
8
1
79
[M-HBr], 51 [M-BrCN].
Cis-1-bromo-1,2-dicyanoethylene. Anal. calcd. for
C HBrN : C, 30.61; H, 0.64; N, 17.84%. Found: C,
4
2
1
3
0.50; H, 0.98; N, 17.63. H NMR (400 MHz, CDCl ):
3
1
3
δ, ppm 6.53 (s). C NMR (400 MHz, CDCl ): δ, ppm
3
-1
1
(
1
(
17.79, 113.13, 112.47, 109.35. IR (ATR): n, cm 3165
m), 3027 (s), 2845 (m), 2238 (m), 1713 (m), 1588 (s),
404 (m), 1271 (s), 1100 (s), 1062 (s), 1030 (s), 578
8
1
79
s). GC-MS: m/z 158 [M Br], 156 [M Br], 77 [M-Br],
6 [M–HBr], 51 [M-BrCN].
,3-Dicyanobicyclo[2,2,1]hepta-2,5-diene. A 10 mL
7
2
round-bottom flask was charged with 1-bromo-1,2-
dicyanoethene (5:1 trans:cis mixture, 2.16 g, 13.8 mmol),
benzene (2 mL), and a magnetic stir bar. The reaction
mixture was cooled in an ice-water bath and then a cooled
solution of 1,3-cyclopentadiene (1.00 g, 15.1 mmol),
produced by the method of Williamson, Minard, and
Masters [17], in benzene (2 mL), was added. The homo-
genous mixture was stirred for 24 h, over which time it
slowly warmed to room temperature. A small sample of
the reaction mixture, examined by GC-MS, showed the
absence of dienophile starting material. The GC-MS also
showed the presence of three new products all with the
same mass spectrum (m/z = 143), consistent with the
desired Diels–Alder diastereomers (M-Br). To promote
the dehydrobromination reaction, triethylamine (1.45 g)
was added to the crude reaction mixture and the flask was
fitted with a heating mantle, a water-cooled condenser
and a gas inlet tube. The mixture was refluxed gently for
24 h under argon. The cooled mixture was transferred
to a separatory funnel using 50 mL of diethyl ether to
rinse the flask contents. This ether solution was washed
Synthesis
1
-Bromo-1,2-dicyanoethylene. A 500 mL round-
bottom flask, equipped with a magnetic stir bar, was
charged with fumaronitrile (10.0 g, 0.128 moles) and
chloroform (150 mLs). Over 5 min, Br (20.5 g, 0.128
2
moles) was added to the mixture and the flask was then
fitted with a heating mantle, a water-cooled condenser
and an argon inlet tube. The reaction mixture was heated
at gentle reflux under an argon atmosphere for two days,
after which time the red-brown color of the bromine was
gone and GC-MS analysis of a small aliquot indicated
that the reaction was complete. The reaction mixture
was vacuum filtered to remove a small amount of fine
off-white solid. The solid was washed with a small
amount of chloroform. The washings were combined
with the original filtrate and reduced to an oil on a rotary
evaporator. The oil was vacuum distilled at a pressure of
3
mm Hg, yielding one major fraction collected over a
temperature range of 40–45°C. The major fraction began
to solidify in the condenser and a heat gun was used to
melt it and allow it to collect in the receiver. The yellow
semicrystalline solid proved to be the desired products
twice with 50 mL of 5% aq. NaHCO
50 mL of aqueous brine solution, dried over anhydrous
MgSO , filtered and reduced on a rotary evaporator to
, washed once with
3
4
1
in a 5:1 trans:cis ratio by GC-MS and H-NMR. Yield =
produce a dark orange-brown oil. The oil was purified
by flash chromatography on silica gel using a hexanes
to 1/1 hexanes/dichloromethane solvent gradient. The
solvent was removed from the column eluent on a rotary
evaporator and the resulting light brown oil was dried
under vacuum. The oil crystallized over the course of
1
3.7 g (68%). This material is similar to that reported by
Moureu and Bongrand, who prepared it by addition of
HBr to dicyanoacetylene [16].
Typically, the above mixture was used directly in
subsequent Diels–Alder reactions. However, it could be
separated into the pure cis and pure trans components
1
several days. Yield = 1.54 g (78.8%), mp 45°C [18]. H
by flash chromatography (SiO column, 4 cm diameter ×
NMR (400 MHz, CDCl ): δ, ppm 6.91 (2H, s), 4.03 (2H,
3
2
13
1
0 cm length) using a solvent gradient of hexanes à 4/1
s), 2.40–2.32 (dd, 2 H). C NMR (400 MHz, CDCl
): δ,
3
hexanes/ether. The first component eluted, assigned to be
trans based on its higher R value, formed a white solid
ppm 141.72, 141.47, 113.24, 75.07, 55.59. IR (ATR): n,
cm 3065 (w), 2997 (w), 2955 (w), 2987 (w), 2216 (m),
1298 (s), 1281 (m), 824 (m), 730 (s). GC-MS: m/z 142
-1
f
(
mp = 49°C) on removal of solvent while the second
component, the cis isomer, was isolated as a pale yellow
oil.
[M], 141 [M-H]), 66 [M-C
4
N ].
2
2,3-Dicyanobicyclo[2,2,2]octa-2,5-diene. A 10 mL
round bottom flask, outfitted with a magnetic stir bar, was
charged with 1-bromo-1,2-dicyanoethylene (5:1 trans:cis
Trans-1-bromo-1,2-dicyanoethylene. Anal. calcd. for
C HBrN : C, 30.61; H, 0.64; N, 17.84%. Found: C,
4
2
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1094–1099

1-BROMO-1,2-DICYANOETHYLENE 1095
1
mixture, 2.00 g, 12.7 mmol) and 1,3-cyclohexadiene
6 mL, 63 mmol). The flask was fitted with a water-
7.40–7.42 (4H, dd), 7.13–7.10 (4H, dd), 5.41 (2H, s). H
(
NMR (400 MHz, Acetone-d ): δ, ppm 7.44–7.39 (4H,
dd), 7.14–7.09 (4H, dd), 5.41 (2H, s). C NMR (400
MHz, CDCl ): δ, ppm 141.08, 137.12, 126.71, 124.52,
113.84, 54.09. IR (ATR): n, cm 3061 (w), 3020 (w),
2219 (m), 1474 (m), 1461 (s), 1454 (s), 1185 (m), 1143
(m), 783 (s), 718 (s), 633 (s).
6
1
3
cooled condenser and then the stirred reaction mixture
was heated in an 80°C oil bath for 36 h. The clear
homogeneous initial mixture turned slightly cloudy
and darkened during heating. After heating, the excess
unreacted diene was removed under vacuum and the
resulting oil was dissolved in 10 mL of chloroform.
Triethylamine (2 mL) was added and the mixture was
stirred in an 80°C oil bath for 2 h. The cooled reaction
mixture was dissolved in 60 mL of dichloromethane.
This solution was washed twice with 50 mL of 5%
aq. NaHCO , dried over anhydrous CaCl , filtered and
3
-1
2,3-Dicyanobicyclo[2,2,2]oct-2-ene. A 100 mL 3-
neck, round-bottom flask, equipped with a magnetic stir
bar, was charged with HPLC grade acetone (40 mL) and
10% Pd/C (100 mg). An addition funnel was placed on
the middle neck and septa were used to cap the remaining
two necks. A solution of 2,3-dicyanobicyclo-[2.2.2]octa-
2,5-diene (1.00 g, 6.4 mmol) in HPLC grade acetone
(20 mL) was placed in the addition funnel. Both the
acetone solution in the flask and in the addition funnel
were saturated by bubbling with a slow steady stream of
hydrogen gas for 10 min. The solution in the addition
funnel was slowly added to the solution in the flask.
A balloon of hydrogen gas was filled and connected
to a syringe and passed through the septum into the
flask. Over the course of five hours, approximately one
equivalent of hydrogen was absorbed as shown by the
balloon deflating. The reaction mixture was filtered and
reduced on a rotary evaporator to produce a white solid
which was recrystallized from 3/1 hexanes/benzene to
3
2
reduced on a rotary evaporator to a brown oil that was
further dried under vacuum. The oil was stirred over 10
mL of CH Cl which did not fully dissolve the brown
2
2
oil. Ten mL of hexanes was added to the mixture which
caused the brown oil to congeal slightly. The supernatant
was poured from the brown oil and collected in a separate
flask. The brown oil was treated with CH Cl followed
2
2
by hexanes twice more, saving the supernatant each
time. The combined supernatants were loaded onto a
silica gel flash chromatography column (4 cm diameter ×
8
cm long) prepared from hexanes slurry and eluted
with 1/1 CH Cl /hexanes. The second compound eluted
2
2
was the desired product. Removal of the solvent gave
white crystals which could be purified by crystallization
from methanol or 3/1 hexanes/benzene. Yield = 1.26
give long white needles.Yield = 0.63 g (61%), mp 115°C
1
[6]. H NMR (400 MHz, CDCl ): δ, ppm 3.01 (2H, s),
3
1
13
g (63%), mp 101–102°C [6a]. H NMR (400 MHz,
1.73 (4H, d), 1.45 (4H, d). C NMR (400 MHz, CDCl ):
3
-1
CDCl ): δ, ppm 6.38 (2H, m), 4.02 (2H, s), 1.53 (4H,
δ, ppm 131.80, 114.48, 34.87, 24.71. IR (ATR): n, cm
3
3
1
m). C NMR (400 MHz, CDCl ): δ, ppm) 132.46,
2965 (s), 2947 (m), 2875 (m), 2217 (s), 1595 (w), 1475
(m), 1454 (m), 1324 (m), 1243 (s), 1162 (s), 1124 (s),
1018 (m), 853 (s), 807 (s). GC-MS: m/z 158 [M], 130
[M-C N ], 103 [M-HCN-C N ].
3
-
1
1
(
1
31.99, 114.09, 41.27, 24.22. IR (ATR): n, cm 2998
w), 2959 (w), 2889 (w), 2221 (m), 1587 (w), 1344 (m),
219 (w), 1154 (m), 1122 (m), 1105 (w), 846 (m) 828
m), 727 (s), 687 (s). GC-MS: m/z 128 [M-C N ], 101
4
2
4
2
(
[
2,3-Dicyanobicyclo[2,2,1]hept-2-ene. The above
procedure was repeated using 2,3-dicyanobicyclo[2.2.1]
hepta-2,5-diene (1.00 g, 7.0 mmol). Removal of solvent
from the filtered reaction mixture left a clear oil, which
4
2
M-HCN-C N ].
4
2
7
,8-Dicyano[2,3:5,6-dibenzo]bicyclo[2,2,2]octa-
2
1
1
1
,5,7-triene. A large diameter test tube (25 mm ×
80 mm) was charged with anthracene (1.78 g, 10 mmol),
-bromo-1,2-dicyanoethene (5:1 trans:cis mixture,
.57 g, 10 mmol) and a magnetic stir bar. While under
formed waxy crystals on standing. Yield = 0.71 g (70%),
1
mp 32°C [18] H NMR (400 MHz, CDCl ): δ, ppm 3.33
3
(2H, s), 1.96 (2H, d), 1.75 (1H, d), 1.44 (1H, d), 1.32
1
3
an argon atmosphere, this mixture was slowly heated
in an oil bath (160°C) for 2 h. As it heated, the solids
in the test tube liquefied. In some trials, the liquid
remained with continued heating and, in other trials, the
mixture resolidified. After heating, the dark black solid
product was extracted into chloroform, triethylamine
(2H, d). C NMR (400 MHz, CDCl ): δ, ppm 132.52,
-1
3
112.84, 48.79, 47.03, 24.20. IR (ATR): n, cm 3010 (w),
2982 (m), 2947 (m), 2878 (m), 2220 (s), 1282 (s), 1251
(m), 1105 (s), 948 (m), 868 (m), 812 (m). GC-MS: m/z =
144 [M], 116 [M-C N ], 89 [M-HCN-C N ].
4
2
4
2
Magnesium 1,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-
hexadecahydro-1,4:8,11:15,18:22,25-tetraethanoph-
thalocyanine. In a glove box, magnesium metal (1.00 g,
41 mmol) was added to 1-butanol (80 mL) in a 250 mL
round-bottom flask. The mixture was removed from the
glove box, iodine (0.20 g) was added and the mixture was
heated at reflux under argon until all the magnesium was
consumed. 2,3-Dicyanobicyclo[2,2,2]oct-2-ene (3.16 g,
20 mmol) was added to the mixture which was heated
at reflux under argon for another 18 h. Over this time,
the reaction mixture changed color from olive green to a
(
3 mL) was added, and the resulting mixture heated at
reflux for 2 more hours. On cooling, all the solvent and
unreacted triethylamine were removed under vacuum
and the resulting dark brown solid was dissolved in hot
acetonitrile, treated with decolorizing charcoal, filtered
and slowly allowed to cool. White crystals were deposited
over the course of two days. These were collected
by vacuum filtration, washed with cold acetonitrile
and dried under vacuum. Yield = 1.24 g (48.8%), mp
1
2
67–268°C [6a]. H NMR (400 MHz, CDCl ): δ, ppm
3
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1095–1099

1
096 J. P. FITZGERALD ET AL.
dark royal blue. When the reaction was complete, half of
the 1-butanol was removed by distillation. Then ethanol
RESULTS AND DISCUSSION
(
200 mL) was added followed by water (800 mL). The
As shown in Fig. 3, heating a 1:1 mixture of fumaro-
nitrile and bromine in chloroform at reflux produces a 5:1
mixture of bromofumaronitrile and bromomaleonitrile
isolated as a waxy yellow solid. We speculate that
the reaction proceeds via addition of bromine to
fumaronitrile to produce the 2,3-dibromosuccinonitrile
intermediate shown. We have no direct evidence for this
intermediate, nor have we made any attempt to isolate it.
The vapor above the refluxing mixture is highly acidic to
moist pH paper, consistent with the formation of gaseous
HBr as a reaction byproduct in the second step. The
product we report here is consistent with that reported by
Moreau and Bongrand who isolated it from the reaction
of dicyanoacetylene with aqueous HBr [16]. Moreau
and Bongrand report limited experimental details: a
yellowish, crystalline, slightly lachrymatory substance,
with a melting point of 48.5–49°C, which is readily
soluble in the usual organic solvents with the exception
of petroleum ether.
resulting precipitate was collected by vacuum filtration
and dried under vacuum. The desired compound was
extracted from the precipitate in a Soxhlet extractor
using chloroform. The resulting royal blue solution
was reduced on a rotary evaporator and then loaded
onto a flash silica gel column (4 cm diameter ×
1
0 cm length) using chloroform as the solvent. Elution
with chloroform containing increasing amounts of
ethyl ether caused the Mg macrocycle to elute from
the column. The solvent was removed using a rotary
evaporator and the product was dried under vacuum.
To get absolutely pure material, a CHCl solution of the
product was loaded onto a neutral alumina column and
3
eluted with a CHCl à 1/1 CHCl /ether solvent gradient.
3
3
A purple impurity eluted quickly followed by a highly
fluorescent red material which is the desired product.
The pure product could be isolated as microcrystals
by slow evaporation of ether and chloroform from
the column eluent. The crystals were collected by
vacuum filtration, washed with cold toluene and dried
under vacuum. Yield = 0.58 g (18%). Anal. calcd for
C H N Mg·½(CHCl ): C, 67.25; H, 5.64; N, 15.49%.
GC-MS of the vacuum-distilled product shows
two closely spaced but fully resolved peaks in the
chromatogram in a ~5:1 ratio. We tentatively assign the
first peak to the trans isomer, which we expect to be
more volatile due to its smaller dipole. We also expect
the trans product to be favored due to its smaller steric
strain. Both isomeric products give the same mass
spectrum. Parent peaks of equal intensity are found
at m/z = 156 and 158, corresponding to molecules
4
0
40
8
3
Found: C, 67.43; H, 5.30; N, 15.46. UV-vis (CHCl ):
lmax, nm (log e) 334 (sh), 348 (5.011), 542 (4.192), 568
(
3
1
sh), 592 (5.082). H NMR (400 MHz, CDCl /C D N):
3
5
5
δ, ppm 4.84 (8H, s), 2.28 (16H, d), 1.82 (16H, d).
1
3
C NMR (400 MHz, CDCl /C D N): δ, ppm 154.73,
3
5
5
-
1
79
81
1
(
1
(
51.38, 29.91, 27.77. IR (ATR): n, cm 3613 (w), 3101
w), 2943 (m), 2860 (m), 1465 (m), 1454 (m), 1159 (s),
087 (s), 1061 (m), 1022 (m), 880 (m), 861 (m), 744
containing
Br and
Br, respectively. Identical
base peaks at m/z = 77 are seen in both mass spectra
consistent with loss of bromine.
2
4
1
s). MALDI-TOF MS: m/z 656 [ Mg complex], 628,
Similarly, the H NMR spectrum of the yellow
6
00, 572 [losses of 28 = C H fragments via retro-Diels–
solid reveals two singlets at δ = 6.80 (trans) and δ =
2
4
Alder reactions].
6.53 (cis) ppm (in CDCl ), again in a 5:1 ratio. These
3
1
,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-Hexa
signals are due to the one vinylic hydrogen atom in each
product isomer. The higher chemical shift assigned to
the trans isomer is consistent with the higher reported
decahydro-1,4:8,11:15,18:22,25tetraethano-29H,31H-
phthalocyanine. The magnesium macrocycle (0.50 g,
.76mmol)washeatedatrefluxina5:1chloroform:acetic
1
0
H NMR chemical shifts for fumaronitrile (δ = 6.27)
acid solution (10 mL) for 30 min. The cooled solution
yields purple-black microcrystals of the desired product.
These are collected by vacuum filtration, washed with
cold toluene and dried under vacuum. Yield = 0.46 g
vs. maleonitrile (δ = 6.18) [19] and chlorofumaronitrile
(δ = 6.40) vs. chloromaleonitrile (δ = 6.31) [20].
The synthesis described here is highly solvent
dependent. Identical trials were conducted in diethyl
ether, dichloromethane, and a combination of these two
solvents and were monitored by GC-MS. These attempts
resulted in a slow rate of reaction eventually producing
polychlorinated byproducts and a minor amount of the
desired product. Benzene was also investigated as a
potential solvent, however, it resulted in no reaction.
Chloroform was ultimately determined to be the best
solvent, in which the reaction can easily be scaled up
to produce 10 or more grams of bromofumaronitrile/
bromomaleonitrile mixture.
(
95%). Anal. calcd for C H N ·(C H O )·2(H O): C,
40 42 8 2 4 2 2
6
9.02; H, 6.90; N, 15.33%. Found: C, 69.09; H, 6.52;
N, 15.38. UV-vis (CHCl ): l , nm (log e) 342 (4.901),
3
max
1
5
14 (sh), 552 (4.591), 600 (3.827), 623 (4.816). H
NMR (400 MHz, CDCl ): δ, ppm 4.90 (8H, s), 2.37
3
1
3
(
16H, d), 1.90 (16H, d), -3.12 (2H, s). C NMR (400
MHz, CDCl ): δ, ppm 151.19, 30.01, 27.93 (unable to
3
-
1
detect one of the ring C atoms). IR (ATR): n, cm 3298
(
(
w), 2943 (m), 2932 (m), 2859 (m), 1480 (m), 1449
m), 1152 (m), 1097 (s), 1081 (m), 975 (s), 857 (m),
7
5
48 (s), 683 (s). MALDI-TOF MS: m/z 634 [M+], 606,
78, 550, 522 [losses of 28 = C H fragments via retro-
There are two reports of 1-bromo-1,2-dicyanoethylene
in the literature, but synthetic details are lacking in both
reports [21].
2
4
Diels–Alder reactions].
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1096–1099

1-BROMO-1,2-DICYANOETHYLENE 1097
1
Interestingly, our attempts to make 1-chloro-1,2-
dicyanoethylene by reacting fumaronitrile (instead of
succinonitrile) with the aforementioned chlorinating
agents (Cl , PCl and SO Cl ) were also not productive.
While small amounts of the desired chloromaleonitrile/
chlorofumaronitrile were made, they were contaminated
by unreacted starting materials and polychlorinated
byproducts. We speculate that the relative rates of the
halogenation vs. elimination steps and/or the relative
reactivity of the starting material vs. intermediate vs.
product were not favorable in the case of chlorination but
were favorable in the case of bromination.
The H NMR of the intermediate Diels–Alder adduct
mixture, taken in d -acetone, shows four clear groupings
6
of peaks in an approximate 8:1:1:1 integral ratio (low
field to high field). These correspond to the eight aromatic
protons, the two bridgehead protons and a lone proton
on carbon bearing one of the nitrile groups. Due to the
mixture of cis and trans dinitriles and the asymmetry
caused by the bromine atom, each group of peaks is a
superposition of similar but inequivalent hydrogen atoms.
Rather than isolating the intermediate, the chloroform
extract of the Diels–Alder adducts was more typically
treated with triethylamine and heated at reflux to
promote the elimination reaction. This product, identical
to that reported by Weis [6a], can be isolated from the
reaction mixture on a gram scale by crystallization from
acetonitrile. The product gives a simplified H NMR
spectrum compared to the intermediate because of the
increase in symmetry on HBr elimination.
The bromofumaronitrile/bromomaleonitrile mixture
was also reacted with 1.3-cyclopentadiene to produce a
mixture of Diels–Alder adducts which could be treated
with triethylamine to yield the substituted maleonitrile,
as illustrated in Fig. 5 below.
2
5
2
2
Diels–Alder and dehydrobromination reactions
1
We have demonstrated the utility of 1-bromo-1,2-
dicyanoethylene as a dicyanoacetylene synthon by its
reaction with three different cyclic dienes of varying
degrees of reactivity. These include highly reactive
1
,3-cyclopentadiene, moderately reactive 1,3-cyclohex-
adiene and relatively unreactive anthracene. Both
bromofumaronitrile and bromomaleonitrile are active
dienophiles and readily add to dienes in a Diels–
Alder reaction. Although they produce diastereomeric
Diels–Alder adducts, elimination of HBr results in
identical substituted maleonitriles. Thus, the mixture
of bromofumaronitrile and bromomaleonitrile does
not need to be separated and can be used directly as a
dicyanoacetylene equivalent, as discussed below.
Br
CN
∆
CN
H
+
NC
Br
CN
Heating an equimolar mixture of anthracene and the
bromofumaronitrile/maleonitrile at high temperature
resulted in the expected cycloaddition reaction as shown
in Fig. 4. As it was heated, the solid reactants initially
liquefied. In some trials, the mixture resolidified on
continued heating but, in others, the reaction mixture
remained liquid. Once cooled, the crude product could
be extracted into chloroform and the Diels–Alder
adducts (trans and cis) could be isolated by column
chromatography from unreacted anthracene and a small
amount of the elimination product.
Et3N, ∆
CN
+
HBr
CN
Fig. 5. Reaction of 1-bromo-1,2-dicyanoethylene with
1
,3-cyclopentadiene
Equimolar amounts of diene and dienophile were
stirred in benzene for 24 h to yield a mixture of
diastereomeric intermediates as shown by GC-MS. Up
to four different diastereomeric Diels–Alder products are
possible due to exo and endo additions of both the cis
and trans dienophiles; three of these are evident in the
GC trace. The mass spectra of these three products are
identical and consistent with the proposed intermediate,
m/z = 143 (M-Br). No attempt was made to isolate or
separate these intermediates; the mixture was simply
carried on to the next step. Again, the elimination of HBr
was promoted by heating in the presence of triethylamine.
The desired maleonitrile product, isolated by extraction,
column chromatography and crystallization, was iden-
tical to that previously reported [18].
CN
Br
CN
∆
+
NC Br
CN
Et N, ∆
3
CN + HBr
CN
The utility of 1-bromo-1,2-dicyanoethylene as a
dicyanoacetylene equivalent useful in the preparation
of novel tetraazaporphyrin structures was demonstrated
as shown in Fig. 6. An initial Diels–Alder reaction with
Fig. 4. Reaction of 1-bromo-1,2-dicyanoethylene with
anthracene
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1097–1099

1
098 J. P. FITZGERALD ET AL.
1
2
,3-cyclohexadiene, followed by elimination, gave
,3-dicyanobicyclo[2.2.2]oct-2,5-diene identical to that
and octaethylporphyrin: specifically (a) a ~10 nm blue
shift in the Q band of the visible spectrum and (b) a high-
field shift in the δ of the internal NH absorbance in the
reported by Weis [6a]. This was then selectively hydro-
genated as described by Cookson et al. to give 2,3-
dicyanobicyclo[2.2.2]oct-2-ene [6b]. This symmetrical
maleonitrile was then cyclized into a novel TAP bear-
ing four 1,4-fused cyclohexyl units on each of the TAP
pyrrole subunits. The IUPAC name for this macrocycle is
1
H NMR spectrum. Ono and co-workers attribute these
differences to the peripheral 1,4-fused cyclohexyl groups
“locking” the porphyrin macrocycle into a more planar
structure when compared to the octaethyl-substituted
analog.
1
1
,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25 hexadecahy-
We have similarly compared the UV-vis and H
dro-1,4:8,11:15,18:22,25-tetraethanophthalocyanine,
but TCTAP (i.e. tetracyclohexyltetraazaporphyrin) is a
convenient shorthand abbreviation.
NMR spectra of the above reported TCTAP complexes
with that of the analogous octaethyltetraazaporphyrin,
OETAP [24]. See Table 1 for a comparison of the
UV-vis data. Like Ono et al., we see slight blue shifts
(
3–6 nm) in the lmax of the Q-band visible absorbances
CN
Br
CN
CN
1
) ∆
H2
for the TAP bearing four 1,4-fused cyclohexyl rings
compared to the OETAP analogs. This holds for both the
magnesium (II) and the free-base complex. However, the
+
2
) Et3N, ∆ Pd / C
NC
+2
-
B-band absorbances for Mg(TCTAP) and H (TCTAP)
1
2
) Mg , BuO , ∆
) H
2
+
are slightly red shifted relative to those of OETAP. We
also observe an upfield shift in the internal NH protons
of the ring-appended macrocycle (-3.12 ppm vs. -2.11
N
ppm for H (OETAP)). However, we have also observed
2
variability in the chemical shift of the internal H atoms
that is likely caused by either a concentration effect and/
or variations in the water content of the NMR solvent.
N
N
N
N
M = Mg+2
2
3
N
M
N
N
M = 2H+
Table 1. UV-vis data of TCTAP complexes in CHCl
3
-1
-1
compared to OETAP complexes (e in M ·cm )
Compound
lmax, nm (log e)
Q1
Fig. 6. Reaction of 1-bromo-1,2-dicyanoethylene with
1
,3-cyclohexadiene, precursor to the synthesis of TCTAP
B
Q2
Mg(TCTAP)
348 (4.8)
342 (4.9)
344 (4.9)
340 (4.9)
595 (4.9)
552 (4.6)
598 (4.9)
558 (4.6)
The hydrogenation step is important. Attempts to
cyclize the unhydrogenated precursor, 2,3-dicyanobicy-
clo[2.2.2]octa-2,5-diene, under standard Linstead
conditions produced only the unsubstituted phthalo-
cyanine. We attribute this to initial formation of the
bicyclohexyl-appended macrocycle followed by
elimination of four ethylene molecules via retro-Diels–
Alder reactions. However, we cannot rule out the
possibility of elimination prior to cyclization. Ono and
co-workers have made extensive and deliberate use of
this retro-Diels–Alder strategy in their work on organic
thin film semiconductor devices [22].
This highly symmetrical TCTAP, shown above, is
related to tetraanthraceno-TAP (a.k.a. dibenzobarreleno-
TAP) reported previously in that it has rigid groups on
both sides of the macrocycle plane. While the 1,4-fused
cyclohexyl groups do not extend as far above and below
the macrocycle plane as fused anthraceno groups do,
they are sufficient to solubilize the complex, presumably
by disrupting the p–p stacking of the aromatic core.
Ono and co-workers have synthesized a porphyrin
that is isostructural to TCTAP reported above [23].
They report several spectral differences between the
porphyrin bearing the four 1,4-fused cyclohexyl groups
H (TCTAP)
623 (4.8)
627 (4.8)
2
Mg(OETAP)
H (OETAP)
2
It is possible that the 1,4-fused cyclohexyl groups
prevent or reduce the macrocycle from ruffling or doming
(i.e. distortions from planarity) as proposed by Ono and
co-workers. If so, the effect seems smaller for the TAP
compared to the porphyrin. Alternatively, changes in the
bond angles on the macrocycle periphery may be altering
the energies of the p MOs, leading to the observed
spectral shifts. In reported OETAP crystal structures,
the average C -C -C bond angle is 129.23° (where Cb
b
b
ethyl
refers to the b-pyrrolic carbon atoms of the TAP) [24, 25].
The analogous average bond angle in Fe(TATAP)Cl
[11] (which has the four 1,4-fused cyclohexyl groups) is
114.83°. Thus, the ethano bridges in TCTAP are pulling
together the two carbon atoms bonded at the b-pyrrolic
positions of the macrocycle. These changes in bond
angles will certainly alter the energy and contribution of
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1098–1099

1-BROMO-1,2-DICYANOETHYLENE 1099
the p orbital on the b-pyrrolic carbons to the macrocycle
p MO’s which could be the cause of the slight shifts in
the visible absorbances.
Khim. 1972; 42: 2586. (d) Kopranenkov VN and
Rumyantseva GI. Zh. Obshch. Khim. 1975; 45:
1555–1559.
7
. (a) Moureu C and Bongrand J. Compt. Rend. 1910;
1
50: 225. (b) Khlifi M, Paillous P, Bruston P, Guil-
CONCLUSION
lemin J, BenilanY, Daoudi A and Raulin F. Spectro-
chim. Acta, Part A 1997; 53: 707–712.
. Zavitsanos PD, U.S. Patent 3,336,359, 1967.
. Ciganek E and Krespan CG. J. Org. Chem. 1968;
Dicyanoacetylene, although a useful precursor to
soluble substituted tetraazaporphyrins, is difficult to
prepare on a gram scale. We report here the synthesis
of a new dicyanoacetylene equivalent, 1-bromo-1,2-
dicyanoethene, from commercially available fumaro-
nitrile and bromine. The product was made and purified
via a vacuum distillation on a multi-gram scale in high
yield as a ~5:1 mixture of trans/cis isomers, as determined
8
9
3
3: 541–544.
1
1
0. Oliver SW and Smith TD. J. Chem. Soc., Perkin
Trans. 2 1987; 1579–1582.
1. Dibenzobarrelenoporphyrazine has also been
reported as tetraanthraeneotetraazaporphyrin. See
Fitzgerald JP, Lebenson JR, Wang G, Yee GT,
Noll BC and Sommer R. Inorg. Chem. 2008; 47:
1
by H-NMR and GC-MS. The utility of the 1-bromo-
1
,2-dicyanoethene mixture as a dicyanoacetylene syn-
thon was demonstrated by reactions with anthracene,
,3-cyclohexadiene and 1,3-cyclopentadiene. In each case,
4
520–4530.
1
2. Kopranenkov VN, Goncharova LS and Luk’yanets
EA. Zh. Obshch. Khim. 1988; 58: 1186–1187.
3. Cass OW. U. S. Patent 2,433,494, 1948.
1
an initial Diels–Alder reaction, followed by elimination of
HBr, produced a substituted maleonitrile. The synthetic
utility was further demonstrated by conversion of the
1
1
4. Shevchenko VI and Kuhkar VP. Zh. Obshch. Khim.
1
966; 36: 735–738.
1,3-cyclohexadiene adduct to a novel tetraazaporphyrin
1
5. Kondratenko NV, Movchun VN, Kopranenkov VN,
Luk’yanets EA and Yagupol’skii LM. Zh. Org.
Khim. 1992; 28: 2149–2155.
bearing four 1,4-fused cyclohexyl units. The fused
cyclohexyl groups prevent p–p stacking of the macrocycle
in solution and thereby enhance its solubility.
1
1
6. Moureu C and Bongrand JC. Compt. Rend. 1920;
1
70: 1025–1028.
7. Williamson KL, Minard RD and Masters KM. Mac-
roscale and Microscale Organic Experiments (5th
ed.), Houghton Mifflin: Boston, 2007; 665–667.
18. Blomquist AT and Winslow EC. J. Org. Chem.
1945; 10: 149–158.
REFERENCES
1
. Kadish KM, Smith KM and Guilard R. (Eds). The
Porphyrin Handbook, Vols. 1–20. Academic Press:
San Diego, 2000.
2
. (a) Cook A and Linstead R. J. Chem. Soc. 1937;
19. (a) Halter RJ, Fimmen RL, McMahon RJ, Peebles
SA, Kuczkowski RL and Stanton JF. J. Am. Chem.
Soc. 2001; 123: 12353–12363. (b) Bai C-X, Zhang
W-Z, He R, Lu X-B and Zhang Z-Q. Tetrahedron
Lett. 2005; 46: 7225–7228.
9
2
29–933. (b) Linstead RP. J. Chem. Soc. 1953;
873–2884. (c) Kopranenkov VN, Goncharova
LS, Marina LE and Luk’yanets EA. Chem. Hetero-
cycl. Compd. 1982; 18: 1269–1271. (d) Khelevina
OG, Chizhova NV and Stuzhin PA. J. Porphyrins
Phthalocyanines 2000; 4: 555–563. (e) Rodriguez-
Morgade MS and Stuzhin PA. J. Porphyrins Phtha-
locyanines 2004; 8: 1129–1165.
20. Morgan AR, Petousis NH and van Lier JE. Eur.
J. Med. Chem. 1997; 32: 21–26.
21. (a) Gobo NRS, Brocksom TJ, Zukerman-Schpector
J and de Oliveira KT. Eur. J. Org. Chem. 2013;
5028–5031. (b) Brown DG, Siddens JK, Diehl RE
and Wright DP Jr. Preparation of Arylpyrrole Pesti-
cides, BR 8803788A, 1989.
3
. (a) Piechucki C and Michalski J. Bull. Acad. Pol.
Sci., Ser. Sci. Chim. 1970; 18: 605–607. (b) Fitzger-
ald JP, Kaul BB and Yee G. Chem. Commun. 2000:
4
9–50. (c) Donzello MP, De Mori G, Viola E, Ercol-
ani C, Ricciardi G and Rosa A. Inorg. Chem. 2014;
3: 8009–8019. (d) Joslin EE, Zaragoza JPT, Baglia
RA, Siegler MA and Goldberg D. Inorg. Chem.
016; 55: 8646–8660 and references therein.
. Sargent M and Timmons C. J. Am. Chem. Soc.
963; 85: 2186–2187.
. Fitzgerald J, Taylor W and Owen H. Synthesis 1991;
86–688.
. (a) Weis C. J. Org. Chem. 1963; 28: 74–78.
b) Cookson RC, Dance J and Godfrey M. Tetrahe-
22. (a) Aramaki S, Sakai Y and Ono N. Appl. Phys.
Lett. 2004; 84: 2085–2087. (b) Akiyama T, Hirao A,
Okujima T, Yamada H, Uno H and Ono N. Hetero-
cycles 2007; 74: 835–842. (c) Yamada H, Okujima
T and Ono N, Chem. Commun. 2008; 2957–2974.
23. Ito S, Uno H, Murashima T and Ono N. Chem.
Comm. 1999; 2275–2276.
24. Fitzgerald JP, Haggerty BS, Rheingold AL,
May L and Brewer GA. Inorg. Chem. 1992; 31:
2006–2013.
5
2
4
5
6
1
6
(
25. Fitzgerald JP, Yap GPA, Rheingold AL, Brewer,
CT, May L and Brewer GA. J. Chem. Soc., Dalton
Trans. 1996; 1249–1253.
dron 1968; 24: 1529–1535. (c) Kopranenkov VN,
Rumyantseva GI and Luk’yanets EA. Zh. Obshch.
Copyright © 2019 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2019; 23: 1099–1099