Towards novel thieno-fused subporphyrazines via functionalized thiophene precursors

Article abstract of DOI:10.1080/17415993.2020.1753190

Thieno-fused subporphyrazines (TPs) containing a central and axially substituted boron atom are a class of compounds with interesting optical and redox properties. Here we present our efforts towards expanding this class of compounds using various thiophene substrates that were prepared by cyanation or nitration of 3,4-dibromothiophene. Moreover, we show that one TP derivative forms a 2:1 complex in the solid state with C₇₀.

Full text of DOI:10.1080/17415993.2020.1753190

Journal of Sulfur Chemistry
ISSN: 1741-5993 (Print) 1741-6000 (Online) Journal homepage: https://www.tandfonline.com/loi/gsrp20
Towards novel thieno-fused subporphyrazines via
functionalized thiophene precursors
Mads Georg Rasmussen, Henrik Gotfredsen, Anders Kadziola & Mogens
Brøndsted Nielsen
To cite this article: Mads Georg Rasmussen, Henrik Gotfredsen, Anders Kadziola & Mogens
Brøndsted Nielsen (2020): Towards novel thieno-fused subporphyrazines via functionalized
thiophene precursors, Journal of Sulfur Chemistry, DOI: 10.1080/17415993.2020.1753190
To link to this article: https://doi.org/10.1080/17415993.2020.1753190
View supplementary material
Published online: 20 Apr 2020.
Submit your article to this journal
Article views: 6
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=gsrp20

JOURNAL OF SULFUR CHEMISTRY
https://doi.org/10.1080/17415993.2020.1753190
Towards novel thieno-fused subporphyrazines via
functionalized thiophene precursors
Mads Georg Rasmussen , Henrik Gotfredsen , Anders Kadziola and
Mogens Brøndsted Nielsen
Department of Chemistry, University of Copenhagen, Copenhagen, Denmark
ABSTRACT
ARTICLE HISTORY
Received 13 February 2020
Accepted 2 April 2020
Thieno-fused subporphyrazines (TPs) containing a central and axi-
ally substituted boron atom are a class of compounds with inter-
esting optical and redox properties. Here we present our efforts
towards expanding this class of compounds using various thio-
phene substrates that were prepared by cyanation or nitration of
3,4-dibromothiophene. Moreover, we show that one TP derivative
KEYWORDS
Cyanation; fullerene;
nitration;
subphthalocyanine;
thiophene
forms a 2:1 complex in the solid state with C70
.
Introduction
Boron subphthalocyanine with an axial chloride on the boron (SubPc-Cl; Figure 1) was
first synthesized in 1972 by Meller and Ossko [1]. Throughout the years, its derivatives
have gained attention due to promising applications in organic photovoltaics [2–5] as well
as organic light-emitting diodes (OLEDs) [6,7]. The 14-π aromatic SubPc chromophore
consists of three aza-linked isoindole moieties with an sp3-hybridized boron metalloid
center; giving rise to its unique concavo-convex shape [8–10].
As shown by Kobayashi and co-workers [11], extensive modifications of the
SubPc core have been done by exchanging one isoindole unit with aromatic dicar-
bonitriles (e.g. 1,8-naphthalenedicarbonitrile, 2,2ꢀ-biphenyldicarbobonitrile, and 4,5-
CONTACT Mogens Brøndsted Nielsen
mbn@chem.ku.dk
Department of Chemistry, University of
Copenhagen, Universitetsparken 5, DK-2100 Copenhagen, Denmark
This article makes reference to supplementary material available on the publisher’s website at https://doi.org/10.1080/
17415993.2020.1753190
© 2020 Informa UK Limited, trading as Taylor & Francis Group

2
M. G. RASMUSSEN ET AL.
Figure 1. Boron subphthalocyanine chloride (SubPc-Cl), boron subphthalocyanine with axial aryl
group (SubPc-Ar), and thieno-fused subporphyrazines TPCl2-Ar and (TPCl2)2-Ar.
phenanthrenedicarbonitrile), yielding less symmetrical subporphyrazine (SubPz) scaf-
folds. Such reduction in symmetry (C3v in SubPcs to Cs in SubPzs) leads to a characteristic
splitting of the Q-band absorption for SubPzs originating from the breakdown of the
degeneracy of the S0 →S1 electronic transition. Recently, Nielsen and co-workers [12]
introduced heterocyclic thiophene units onto the SubPz core, yielding thieno-fused sub-
porphyrazines (TPCl2-Ar and (TPCl2)2-Ar); (Figure 1). This new chromophore class
(which we here generally name TP) exhibits desired photophysical properties regarding
solar energy harvesting; namely redshift of Q-band absorption in addition to broaden-
ing of the Q-band without compromising the absorption intensity. Replacing benzene for
thiophene units in the SubPc scaffold additionally enhances the capability of TPs to act
as electron donors, reflected in the lowering of the oxidation potential [12]. Thiophene
presents a versatile synthetic building block, which we thought to employ towards func-
tionalizing the thiophene α-positions at the TP chromophore periphery. This modification
poses an interest for further elucidation of the photophysical properties of the TP series
and how these properties can be manipulated by substituent effects. For this reason, com-
bined with the intrinsic property of enhanced donor strength of the TP core, we decided
to work towards a TPX2-Cl push–pull system (Scheme 1), by the installation of highly
electron-withdrawing X = NO2 groups (EWG) at the TP thiophene α-positions. Such tar-
gets would require suitably functionalized 3,4-dicyanothiophene precursors, which we set
out to synthesize.
While push–pull systems of the TP class are still elusive, we herein report the synthe-
sis of three nitro-substituted thiophene derivatives and one new silyl-functionalized TP
derivative all en route towards the intriguing thieno-fused SubPz scaffolds with strongly
electron-withdrawing groups.

JOURNAL OF SULFUR CHEMISTRY
3
NC
CN
S
X
X
Cl
X
X
S
I
+
N
N
B
N
BCl3
N
N
CN
CN
N
TPX2-Cl
X = EWG or PG
II
Scheme 1. Synthetic strategy for formation of TPX2-Cl, X = EWG or PG. EWG = electron-withdrawing
group; PG = protecting group.
Results and discussion
The synthetic approach for obtaining unsymmetrically substituted TP-scaffolds relies on a
statistical cyclotrimerization reaction of suitable 1,2-dicyano derivatives (I and II) medi-
ated by a strong Lewis acid, commonly BCl3. During synthesis, chlorination poses a
problem, as observed for both SubPc [13] and TP derivatives [12], likely occurring via elec-
trophilic aromatic substitutions annotated to in situ generation of Cl2 [14]. Thus, from the
3,4-dicyanothiophene substrate, only chlorinated TP products were isolated. Protection
group (PG) functionalized TP derivatives, therefore, presented a potential way of avoiding
chlorination at the α-positions while also providing the opportunity for functionalization
at the same positions after the formation of the chromophore scaffold. The synthesis hereof
demanded the preparation of suitable 3,4-dicyanothiophene derivatives I to be mixed with
an excess of phthalonitrile II in the cyclotrimerization reaction (Scheme 1).
Synthesis of 3,4-dicyano derivatives
The initially targeted nitro-functionalized TP precursor 1(Scheme 2) was to be synthesized
by cyanation of thiophene derivative 2, available from a patented nitration protocol of 3,4-
dibromothiophene 3 [15]. However, in our hands, when those conditions were applied to
3, only formation of the mono-nitrated thiophene derivative 4 was observed; nevertheless,
in excellent yield (95%).
To obtain 2, more forcing conditions had to be applied. Indeed, by use of 20% oleum
instead of conc. sulfuric acid, the di-nitrated product 2 was obtained in 77% after recrystal-
lization (Scheme 2). The following cyanation of 2 and 4, respectively, under Rosenmund
von-Braun conditions (using CuCN in DMF at elevated temperature), however, proved
challenging. Only decomposition of the nitro-thiophene 2 was observed, while success-
ful cyanation of 4 was achieved, after comprehensive temperature and solvent screening,
giving the mono-nitrated TP precursor 5, albeit only in a yield of 10%. As synthesis and
purification of 5 proved tedious, a reverse functionalization protocol towards 5 was inves-
tigated via the dicyanothiophene 6 (easily obtained by cyanation of 3). Nitration of 6
successfully gave 5; however, in very poor yield upon full conversion of 6. Nevertheless,
a new potential precursor for TP formation was at hand.
As the formation of the TP scaffold ought to be performed under forcing conditions in
presence of BCl3 (vide infra), acid-resistant -SiR3 groups posed as interesting candidates

4
M. G. RASMUSSEN ET AL.
Br
Br
NO2
H2SO4 / HNO3
CuCN
DMF
120 °C, 22 h
10%
CHCl3
50 °C, 24 h
95%
S
4
NC
CN
Br
Br
NC
NC
CN
NO2
CuCN
H2SO4 / HNO3
DMF
CHCl3
50 °C, 17 h
3%
S
6
S
3
120 °C, 60 h
S
5
54%
Br
Br
NO2
CN
NO2
CuCN
Oleum / HNO3
CHCl3
50 °C, 24 h
77%
O2N
O2N
DMF
120 °C, 1.5 h
S
2
S
1
Scheme 2. Synthesis of thiophene derivatives.
Table 1. Synthesis of α-functionalized 3,4-dicyano derivatives.
Starting compound
R
Product
Yield/%
6
6
TIPS
TES
7a
7b
53
61
for protection groups to avoid in situ chlorination of the thiophene α-positions. Addition-
ally, a mild methodology for the removal of such silyl protection groups in the presence
of the SubPc scaffold has already been developed within our group [16]. The protected TP
precursors 7a,b (Table 1) were afforded via a lithiation-substitution protocol of dicyan-
othiophene 6, by treatment with lithium diisopropylamide (LDA) at −78°C, followed by
addition of either triisopropylsilyl-(TIPS) or triethylsilyl-(TES) chloride.
Formation of thieno-fused subporphyrazine scaffolds
With the TP precursors 5, 7a and 7b in hand, the formation of new TP derivatives was
investigated (Table 2). Cyclotrimerization of the nitro-functionalized thiophene 5 and
phthalonitrile in a 1:5 stoichiometry was initially attempted in a 1M BCl3 solution in p-
xylene. Unfortunately, upon heating towards reflux (137°C) only decomposition of the
starting material 5 was observed. Changing the solvent to o-dichlorobenzene (o-DCB)
and adding the BCl3 (1M) in a hexane solution allowed a lower initial concentration of
BCl3. By initially distilling off the hexane as the temperature was slowly elevated towards
the reflux point of o-DCB (180°C), a gradual concentration of the reaction mixture was

JOURNAL OF SULFUR CHEMISTRY
5
Table 2. Reaction conditions for formation of TPCl2-Ar and 10.
CN
S
S
Cl
Cl
TIPS
N
TIPS
Cl
Cl
CN
CN
II
+
BCl3
BCl3
N
N
B
N
N
B
N
NC
N
N
R1 = H, R2 = NO2
5 / II = 1:5
R1 = R2 = TIPS
7a / II = 1:5
N
N
N
N
R1
R2
tBuPhOH
Toluene
S
8
5, 7a, 7b
9
BCl3
R1 = R2 = TES
7b / II = 1:5
S
S
Cl
N
Cl
OPhtBu
X1
N
X2
Cl
TIPS =
TES =
Si
N
B
N
N
B
N
in situ chlorination
N
N
N
N
, > 5 min
Si
N
N
10: X1 = X2 = TES
11: X1 = TES; X2 = Cl
(racemic mixture)
TPCl2-Ar
Thiophene substrate
R1, R2
Solventa
Product
Yield/% (time/h)
5b
5
7a
7b
7b
H, NO2
H, NO2
TIPS, TIPS
TES, TES
TES, TES
p-xylene
o-DCB
o-DCB
o-DCB
o-DCB
–
−(0.5)
1 (0.5)
−(24)
4 (0.5)
3 (2.5)
8c
n.r.
10
8c
Note: n.r.: No reaction.
aHeated to reflux; bp: p-xylene (137°C), o-DCB (180°C).
bDecomposition of starting compound.
cYield determined from the p-tBuC6H4O-derivative TPCl2-Ar (2 steps).
achieved. This approach successfully yielded a cyclotrimerization product without any
apparent decomposition. TLC analysis only revealed one intensely purple-colored product
eluting above parent SubPc-Cl. Surprisingly, high-resolution mass spectrometry (HRMS)
revealed an intense peak at m/z = 504.97603 corresponding to the protonated species of
the bis-chlorinated TP derivative 8. This suggested two-fold α-chlorination of the thio-
phene moiety, implicating also that the nitro group in the 2-position of 5 had been replaced.
The structure of 8 was decisively confirmed by replacement of the axial chloride with
4-tert-butylphenol in refluxing toluene (this replacement simplifies purification by chro-
matography), which gave the dark purple TPCl2-Ar in 1% yield. Characterization of the
isolated product was in complete agreement with data previously reported by Gotfredsen
et al. [12].
The bulky TIPS-protected TP precursor 7a presented an immediate solution to avoid
the readily occurring chlorination of the thiophene α-position. However, upon treatment
of 7a even with large excess of BCl3 at 180°C, no conversion or apparent degradation of
7a was observed over the course of 24 h (neither in the presence nor in the absence of
phthalonitrile). As 7a proved stable and unreactive towards the given conditions, the less
hindered TES-protected 7b was investigated for possible TP formation. To our satisfaction,
successful cyclotrimerization with phthalonitrile gave the dark pink TES-functionalized

6
M. G. RASMUSSEN ET AL.
TP chromophore 10. From analysis by TLC of the reaction progress it was observed that
upon reflux at 180°C, initially only formation of 10 occurred. This was shortly followed by
formation of the light purple mono-chlorinated TES-TP 11 after only 5 min (supported
by HRMS data), which was followed by formation of the dark purple bis-chlorinated TP 8
(confirmed by isolation as TPCl2-Ar). This observation suggested a rapid stepwise chlo-
rination of 10, despite having TES-protection groups at the thiophene α-positions. The
optimal reaction time for obtaining 10 versus the competing chlorination was concluded
to be 30 min. By a parallel experiment with increased reaction time, full consumption
of 10 and 11 was observed within 2.5 h. The only TP product formed, being the bis-
chlorinated TP 8 (isolated as TPCl2-Ar), was obtained in a surprisingly high yield of 3%;
a 6-fold increase when compared to our previously reported synthesis of TPCl2-Ar, from
the non-chlorinated thiophene 3 [12]. As reflected in how readily the silyl-functionalized
thiophene moiety of 10 and 11 underwent chemical transformation, both 10 and 11 proved
to be unstable even when purified, stored at −18 ° under inert atmosphere. Nevertheless,
a sufficiently pure sample could be obtained by utilizing size exclusion chromatography
to allow confirmation of the structure. However, photo- and electrochemical properties
remain unknown. On account of the sp3-hybridized boron center and hence cone-shaped
structure, compound 11 with two different substituents on one thiophene ring is chi-
ral. As the starting materials are not chiral, this product has to be formed as a racemic
mixture.
Figure 2. ORTEP diagram of 2:1 complex of (TPCl2)2-Ar and C70. CCDC 1980090. Crystals were grown
from a toluene/heptane bi-layer system. Thermal ellipsoids are displayed at 50%. Hydrogen atoms are
omitted for clarity.

JOURNAL OF SULFUR CHEMISTRY
7
TP complex formation with c70 in the solid phase
The thieno-fused subporphyrazine (TPCl2)2-Ar was previously shown to co-crystalize
with Buckminsterfullerene C60 in a 1:1 complex, promoted by π-π interactions [12], while
SubPc-Ar formed 2:1 complexes with C60 as well as with C70 [10].1 Crystallization of
(TPCl2)2-Ar with the ellipsoid-shaped C70 yielded a 2:1 complex, which was determined
by X-ray crystallographic analysis (Figure 2). The average distance between the convex
(C70) and concave (TP) π-systems is approximately 3.4 Å (see Supporting Information for
details). The boron-boron distance between two (TPCl2)2-Ar molecules complexing the
same C70 is 14.5 Å.
Conclusion
In this work we have presented two new synthetic protocols for thiophene func-
tionalization, providing three new dicyanothiophene derivatives, designed to undergo
cyclotrimerization reaction. All were investigated for reactivity towards cyclotrimerization
with phthalonitrile, yielding one new silyl-functionalized thieno-fused subporphyrazine
in addition to revealing two new synthetic pathways for obtaining a di-chlorinated
thieno-fused subporphyrazine. Furthermore, a 2:1 packing motif of a chloro-thieno-
subporphyrazine and fullerene C70 was observed in the solid state by X-ray crystallo-
graphic analysis. Synthesis of functionalized thieno-fused subporphyrazines is clearly a
challenging task, and new strategies need to be explored in future work.
Experimental
General methods
All water sensitive reactions were carried out in flame-dried, and N2- or Ar-flushed glass-
ware. o-DCB was distilled prior to use from CaH2. Diisopropylamine was distilled from
NaH prior to use. Anhydrous DMF, toluene and THF were tapped from an Innovative
Technology (IT) plant model PS-MD-05. Thin layer chromatography (TLC) aluminum
sheets precoated with silica gel were used. Flash column chromatography was carried out
using SiO2 with particle size of 40–63 μm, whereas dry column vacuum chromatography
was carried out with SiO2 of 15–40 μm particle size. Size exclusion chromatography was
carried out using Bio-Beads® S-X 8. Chemicals and solvents were purchased from the sup-
pliers: Sigma Aldrich, Flourochem and VWR. A technical grade of petroleum ether was
1
used with boiling point 40–65°C. Both H and 13C NMR spectra were recorded on a
Bruker 500 MHz instrument with a non-inverse cryoprobe. As reference, the deuterated
solvents were used: CDCl3 (1H NMR: δ = 7.26 ppm, 13C NMR: δ = 77.16 ppm), CD2Cl2
(1H NMR: δ = 5.32 ppm, 13C NMR: δ = 54.00 ppm). HRMS was recorded on a Bruker
SolariX XR MALDI-FT-ICR instrument with dithranol as matrix. All melting points are
uncorrected. ABruker FT-IR instrument using attenuated total reflectance (ATR) sampling
technique was used for recording IR-spectra.
2,5-Dinitro-3,4-dibromothiophene (2)
A flask was charged with a solution of 3,4-dibromothiophene (3.00 g, 12.4 mmol) and
HNO3 (5 mL, 68%) in CHCl3 (5 mL). Fuming H2SO4 (5 mL, 20% SO3) was added dropwise

8
M. G. RASMUSSEN ET AL.
to the colorless reaction mixture over 5 min. The reaction mixture was heated to 50°C
and stirred for 24 h, cooled to RT, diluted by addition of H2O (20 mL) and extracted with
CH2Cl2 (75 mL). The organic phase was washed with H2O (3 × 50 mL), dried over MgSO4,
filtered and concentrated in vacuo. The crude product was recrystallized in a 1:2 boiling
solution of CH2Cl2/petroleum ether, which yielded the title compound 2 (3.17 g, 77%)
as bright yellow crystals. Rf (70% CH2Cl2/heptanes) = 0.69. M.p.: 128°C (Lit. [19]: 128°C
from EtOH). 13C NMR (126 MHz, CDCl3): δ = 148.28, 117.93 ppm. IR (ATR, neat): 1530,
1488, 1337, 1311, 1290, 1113, 907, 872, 834, 746, 615, 478 cm−1. The title compound was
not observed using the following methods: GC-MS, MALDI+ or ESI+.
2-Nitro-3,4-dibromothiophene (4)
A flask was charged with a solution of 3,4-dibromothiophene (3.00 g, 12.4 mmol) and
HNO3 (5 mL, 68%) in CHCl3 (10 mL). H2SO4 (5 mL, 95%) was added dropwise to the
colorless reaction mixture over 5 min. The reaction mixture was heated to 50°C and
stirred for 24 h, cooled to RT, diluted by addition of H2O (30 mL) and extracted with 1:4
CH2Cl2/petroleum ether (100 mL). The organic phase was washed with H2O (3 × 30 mL),
dried over MgSO4, filtered and concentrated in vacuo. The crude product was recrystal-
lized in a 3:1 boiling solution of CH2Cl2/petroleum ether, which yielded the title compound
4 (3.39 g, 95%) as yellow crystals. Rf (70% CH2Cl2/heptanes) = 0.66. M.p.: 110–112°C (Lit.
1
[19]: 115–116°C from EtOH). H NMR (500 MHz, CDCl3): δ = 7.62 (s, 1H) ppm. 13C
NMR (126 MHz, CDCl3): δ = 127.87, 117.93, 116.93, 116.82 ppm. IR (ATR, neat): 3094,
1529, 1489, 1386, 1311, 1291, 1113, 907, 833, 754, 746, 726 cm−1. The title compound was
not observed using the following methods: GC-MS, MALDI+ or ESI+.
2-Nitro-3,4-dicyanothiophene (5)
Method A: A flame-dried flask was charged with a mixture of 4 (10.0 g, 34.9 mmol),
CuCN (12.4 g, 139 mmol) in anhydrous DMF (100 mL), heated to 120°C and stirred for
22 h. The reaction mixture was cooled to RT and filtered through a short silica plug
(SiO2, 15–40 μm) eluted with CH2Cl2 (combined fractions 800 mL). The filtrate was
washed with sat. aqueous Na2S2O3 (500 mL), sat. aq. NaHCO3 (1000 mL) and water
(1500 mL), dried over MgSO4, filtered and concentrated in vacuo. The crude residue was
dissolved in CH2Cl2 and purified by dry column vacuum chromatography (SiO2, 0–80%
CH2Cl2/heptanes, 15% increment, 150 ml fractions), followed by flash column chromatog-
raphy (SiO2, 8% EtOAc/toluene) in addition to gravity column chromatography (SiO2, 80%
CH2Cl2/heptane) which afforded the title compound 5 (640 mg, 10%) as an off-white solid.
1
Rf (80% CH2Cl2/heptanes) = 0.22. M.p.: 195°C. H NMR (500 MHz, CDCl3): δ = 8.17
(s, 1H) ppm. 13C NMR (126 MHz, CDCl3): δ = 156.65, 138.62, 114.68, 111.79, 110.27,
108.64 ppm. IR (ATR, neat): 3113, 2246, 1546, 1519, 1451, 1339, 1174, 1060, 913, 871, 821,
780, 744, 536, 481, 452, 429, 409 cm−1. The title compound was not observed using the
following methods: GC-MS, MALDI+ or ESI+. Anal. Calcd (%) for C6HN3O2S: C, 40.23;
H, 0.56; N, 23.46; S,17.90. Found: C, 40.28; H, 0.69; N, 22.55; S, 17.76.
Method B: A flask was charged with a solution of 6 (200 mg, 1.49 mmol) and HNO3
(5 mL, 68%) in CHCl3 (10 mL). H2SO4 (5 mL, 95%) was added dropwise to the color-
less reaction mixture over 5 min. The reaction mixture was heated to 50°C and stirred
for 17 h, cooled to RT, diluted by addition of H2O (30 mL) and extracted with CH2Cl2

JOURNAL OF SULFUR CHEMISTRY
9
(150 mL). The organic phase was washed with H2O (3 × 30 mL), dried over MgSO4, fil-
tered and concentrated in vacuo. Purification by gravity column chromatography (SiO2,
80% CH2Cl2/heptane) afforded the title compound 5 (8 mg, 3%) as an off-white solid. M.p,
1H- and 13C NMR spectroscopic data in accordance with data reported for compound
obtained by method A.
3,4-Dicyanothiophene (6)
A flame-dried flask was charged with a mixture of 3,4-dibromothiophene (12.0 g,
49.6 mmol), CuCN (17.8 g, 198 mmol) and anhydrous DMF (100 mL), which was heated
to 120°C and stirred for 60 h. The reaction mixture was cooled to RT and filtered through a
short silica plug (SiO2, 15–40 μm) then eluted with CH2Cl2 (combined fractions 800 mL).
The filtrate was washed with sat. aq. Na2S2O3 (500 mL), sat. aq. NaHCO3 (500 mL)
and water (500 mL), dried over MgSO4, filtered and concentrated in vacuo. The crude
residue was dissolved in CH2Cl2 and purified by flash column chromatography (SiO2, 70%
CH2Cl2/heptanes), which afforded the title compound 6 (3.65 g, 54%) as a white solid.
1
Rf (70% CH2Cl2/heptanes) = 0.13. M.p.: 170–171°C (Lit. [20]: 170–171.5°C). H NMR
(500 MHz, CDCl3): δ = 8.06 (s, 2H) ppm. 13C NMR (126 MHz, CDCl3): δ = 137.07,
113.29, 111.86 ppm. HRMS (MALDI+, dithranol): m/z = 135.00223 [M + H+], calcd. for
[C6H3N2S+]: m/z = 135.00114. IR (ATR, neat): 3125, 3108, 2236, 1614, 1508 1450, 1169,
874, 832, 534, 476, 449, 432 cm−1. Anal. Calcd (%) for C6H2N2S: C, 53.72; H, 1.50; N,
20.88; S, 23.90. Found: C, 52.93; H, 1.45; N, 20.43; S, 23.93.
2,5-Bis(triisopropylsilyl)−3,4-dicyanothiophene (7a)
A flame-dried, argon-flushed flask was charged with a solution of freshly distilled diiso-
propylamine (1.26 mL, 8.94 mmol), anhydrous THF (20 mL) and cooled to −15°C. n-BuLi
(5.6 mL, 9.0 mmol, 1.6 M in hexane) was added dropwise over 10 min while the reaction
temperature was maintained at −15°C. The reaction mixture was stirred for an addi-
tional 30 min and cooled to −78°C. The prepared LDA solution was cannulated to a
flame-dried, argon-flushed flask charged with a solution of 6 (200 mg, 1.49 mmol) in
anhydrous THF (20 mL) at −78°C. The bronze reaction mixture was allowed to warm
up to –15°C and stirred for 10 min. The now light green colored reaction mixture was
cooled to −78°C, and chlorotriisopropylsilane (0.82 mL, 3.7 mmol, 97% w/v) was added.
The reaction mixture was allowed to warm to RT. The reaction was quenched with
H2O (25 mL) and the organic phase was isolated and concentrated in vacuo. Purifica-
tion by flash column chromatography (SiO2, 70% CH2Cl2/heptanes) afforded a crude
yellow crystalline compound. The crude product was recrystallized in boiling methanol
and charcoal, which yielded the title compound 7a (352 mg, 53%) as a white crystalline
solid. Rf (70% CH2Cl2/heptanes) = 0.47. M.p.: 114–115°C. 1H NMR (500 MHz, CDCl3):
δ = 1.62 (sep, J = 7.5 Hz, 6H), 1.14 (d, J = 7.5 Hz, 36H) ppm. 13C NMR (126 MHz,
CDCl3): δ = 154.57, 121.48, 114.45, 18.56, 11.97 ppm. HRMS (MALDI+, dithranol):
m/z = 447.26859 [M + H+], calcd. for [C25H43N2SSi2+]: m/z = 447.26800. Anal. Calcd
for C24H42N2SSi2: C, 64.51; H, 9.47; N, 6.27; S, 7.17. Found: C, 64.54; H, 9.62; N, 6.32; S,
6.75.

10
M. G. RASMUSSEN ET AL.
2,5-Bis(triethylsilyl)-3,4-dicyanothiophene (7b)
A flame-dried, argon-flushed flask was charged with a solution of freshly distilled diiso-
propylamine (6.31 mL, 44.7 mmol), anhydrous THF (25 mL) and cooled to −15°C. n-BuLi
(28 mL, 45 mmol, 1.6 M in hexane) was added dropwise over 15 min while the reaction
temperature was maintained at −15°C. The reaction mixture was stirred for an additional
30 min and cooled to −78°C. The prepared LDA solution was cannulated to a flame-dried,
argon-flushed flask charged with a solution of 6 (1.00 g, 7.45 mmol) in dry THF (50 mL)
at −78°C. The bronze reaction mixture was allowed to warm to −15°C and stirred for
10 min. The now light green colored reaction mixture was cooled to −78°C, and chlorotri-
ethylsilane (3.22 mL, 18.7 mmol, 97% w/v) was added. The reaction mixture was allowed
to warm up to −20°C. The reaction was quenched by dropwise addition of sat. aq. NH4Cl
(40 mL), and the organic phase was isolated and concentrated in vacuo. Purification by
flash column chromatography (SiO2, 70% CH2Cl2/heptanes) afforded the title compound
7b (1.65 g, 61%) as a bronze oil. Rf (70% CH2Cl2/heptanes) = 0.45. 1H NMR (500 MHz,
CDCl3): δ = 1.02–0.98 (m, 30H) ppm. 13C NMR (126 MHz, CDCl3): δ = 156.64, 120.62,
113.94, 7.31, 3.74 ppm. HRMS: The title compound was not observed using the following
methods: GC-MS, MALDI+ or ESI+. Anal. Calcd for C18H30N2SSi2: C, 59.61; H, 8.34; N,
7.72; S, 8.84. Found: C, 58.96; H, 8.39; N, 7.76; S, 8.03.
TP-TES2-Cl (10)
A flame-dried flask was charged with 7b (303 mg, 835 μmol) and phthalonitrile (535 mg,
4.18 mmol) in o-DCB (15 mL). A solution of BCl3 (14.0 mL, 14.0 mmol, 1M in hexanes)
was added to the reaction mixture, and it was heated to reflux (68°C) for 30 min. The hex-
anes were removed by distillation, and the reaction mixture was further refluxed (180°C)
for 30 min. The reaction mixture was cooled to RT while applied with a N2 stream and
continued stirring to remove excess BCl3. Following, the reaction mixture was concen-
trated to dryness in vacuo. The crude mixture was filtered through a short silica plug
(SiO2, 15–40 μm, 50% EtOAc/toluene) and concentrated to dryness in vacuo. Purifica-
tion by gravity column chromatography (SiO2, 1.5% EtOAc/toluene) yielded the crude 10
containing some 11 (while compound 8 has been removed; 8: HRMS analysis (MALDI+,
dithranol): m/z = 504.97603 [M + H+], calcd. For [C22H9BCl3N6S+]: m/z = 504.97625).
Extensive size exclusion column chromatography (S-X 8, toluene) afforded the title com-
pound 10 (19 mg, 4%) as a pink solid, by removal of 11 (11: HRMS (MALDI+, dithranol):
m/z = 585.10198 [M + H+], calcd. For [C28H24BCl2N6SSi+]: m/z = 585.10170).
10: Rf (1.5% EtOAc/toluene) = 0.51. 1H NMR (500 MHz, CD2Cl2): δ = 8.86–8.81 (m,
2H), 8.75–8.72 (m, 2H), 7.97–7.88 (m, 4H), 1.48–1.31 (m, 12H), 1.15–1.09 (m, 18H)
ppm. 13C NMR (126 MHz, CD2Cl2): δ = 150.06, 149.24, 147.20, 146.84, 140.81, 131.66,
130.54, 130.33, 129.85, 122.46, 122.32, 7.95, 4.73 ppm. HRMS: (MALDI+, dithranol):
m/z = 665.22677 [M + H+], calcd. For [C34H38BCl3N6SSi+]: m/z = 665.22715
TPCl2-Ar
Method A: A flame-dried flask was charged with 7b (255 mg, 703 μmol) and phthaloni-
trile (450 mg, 3.52 mmol) in o-DCB (15 mL). A solution of BCl3 (6.0 mL, 6.0 mmol, 1M in
hexanes) was added to the reaction mixture, and it was heated to reflux (68°C) for 30 min.
The hexanes were removed by distillation, and the reaction mixture was further refluxed

JOURNAL OF SULFUR CHEMISTRY
11
(180°C) for 2.5 h. The reaction mixture was cooled to RT while applied with a N2 stream
and continued stirring to remove excess BCl3. The reaction mixture was concentrated to
dryness in vacuo and transferred to a flask with 4-tert-butylphenol (430 mg, 2.86 mmol) in
toluene (15 mL). The reaction mixture was heated to reflux for 48 h and concentrated in
vacuo. The crude mixture was concentrated in vacuo and purified by gravity column chro-
matography (SiO2, 1.5% EtOAc/toluene), followed by flash column chromatography (SiO2,
12% EtOAc/heptanes), which afforded the title compound TPCl2-Ar (11 mg, 3%) as a pur-
ple solid. Rf (1.5% EtOAc/toluene) = 0.39. M.p.: > 230°C. 1H NMR (500 MHz, CDCl3):
δ = 8.84–8.75 (m, 4H), 7.91–7.85 (m, 4H), 6.78 (d, J = 8.7 Hz, 2H), 5.37 (d, J = 8.7 Hz,
2H), 1.09 (s, 9H) ppm. 13C NMR (126 MHz, CDCl3): δ = 151.25, 150.11, 148.98, 145.96,
143.89, 134.04, 131.30, 130.18, 129.92, 129.56, 125.96, 122.41, 122.11, 121.11, 117.85,
33.99, 31.46 ppm. HRMS (MALDI+, dithranol): m/z = 619.10386 [M + H+], calcd. for
[C32H22BCl2N6OS+]: m/z = 619.10404.
Method B: A flame-dried flask was charged with 5 (250 mg, 1.40 mmol) and phthaloni-
trile (894 mg, 6.98 mmol) in o-DCB (15 mL). A solution of BCl3 (24.0 mL, 24.0 mmol, 1M
in hexanes) was added to the reaction mixture, and it was heated to reflux (68°C) for 30 min.
The hexanes were removed by distillation, and the reaction mixture was further refluxed
(180°C) for 30 min. The reaction mixture was cooled to RT while applied with a N2 stream
and continued stirring to remove excess BCl3. The reaction mixture was concentrated to
dryness in vacuo and transferred to a flask with 4-tert-butylphenol (850 mg, 5.59 mmol)
and N,N-diisopropylethylamine (0.73 mL, 4.19 mmol) in toluene (15 mL). The reaction
mixture was heated to reflux for 24 h and concentrated in vacuo. The crude mixture was
filtered through a short silica plug (SiO2, 15–40 μm, 50% EtOAc/toluene) and concen-
trated to dryness in vacuo. Purification by gravity column chromatography (SiO2, 1.5%
EtOAc/toluene) followed by gravity column chromatography (SiO2, 25% EtOAc/heptane)
afforded the title compound TPCl2-Ar (8 mg, 1%).
Note
1. For other studies on subporphyrazine derivatives with fullerenes, see for example, [17,18].
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
The Independent Research Fund Denmark Natural Sciences [grant number 8021-00009B] is
acknowledged for financial support.
ORCID
Mads Georg Rasmussen http://orcid.org/0000-0003-4562-5832
Henrik Gotfredsen http://orcid.org/0000-0002-9559-7190
Mogens Brøndsted Nielsen http://orcid.org/0000-0001-8377-0788

12
M. G. RASMUSSEN ET AL.
References
[1] Meller A, Ossko A. Phthalocyaninartige Bor-Komplexe. Monatshefte für Chemie. 1972;103:
150–155.
[2] Cnops K, Rand BP, Cheyns D, et al. 8.4% efficient fullerene-free organic solar cells exploiting
long-range exciton energy transfer. Nat Commun. 2014;5:3406.
[3] Huang T, Chen H, Feng J, et al. Rylene Annulated subphthalocyanine: a promising cone-shaped
non-fullerene acceptor for organic solar cells. ACS Materials Lett. 2019;1:404–409.
[4] Gotfredsen H, Neumann T, Storm FE, et al. Donor–acceptor-functionalized subphthalocya-
nines for dye-sensitized solar cells. ChemPhotoChem. 2018;2:976–985.
[5] Copley G, Hwang D, Kim D, et al. First-generation subporphyrinatoboron(iii)sensitizers
surpass the 10% power conversion efficiency threshold. Angew Chem. 2016;128:10443–10447.
[6] Plint TG, Lessard BH, Bender TP. Doping chloro boron subnaphthalocyanines and chloro
boronsubphthalocyanine in simple OLED architectures yields warm white incandescent-like
emissions. Opt Mater (Amst). 2018;75:710–718.
[7] Man J, He S, Zhang T, et al. Black phase-changing cathodes for high-contrast organic light
emitting diodes. ACS Photonics. 2017;4:1316–1321.
[8] Claessens CG, González-Rodríguez D, Rodríguez-Morgade MS, et al. Subphthalocyanines,
subporphyrazines, and subporphyrins: singular nonplanar aromatic systems. Chem Rev.
2014;114:2192–2277.
[9] Fulford MV, Jaidka D, Paton AS, et al. Crystal structures, reaction rates, and selected physical
properties of halo-boronsubphthalocyanines (halo = fluoride, chloride, and bromide). J Chem
Eng Data. 2012;57:2756–2765.
[10] Gotfredsen H, Holmstrøm T, Muñoz AV, et al. Complexation of fullerenes by subphthalocya-
nine dimers. Org Lett. 2018;20:5821–5825.
[11] Shimizu S, Kobayashi N. Structurally-modified subphthalocyanines: molecular design towards
realization of expected properties from the electronic structure and structural features of
subphthalocyanine. Chem Commun. 2014;50:6949–6966.
[12] Gotfredsen H, Storm FE, Muñoz AV, et al. Thieno-fused subporphyrazines: a new class of light
harvesters. Chem Eur J. 2017;23:16194–16198.
[13] Zhu H, Shimizu S, Kobayashi N. Subazaphenalenephthalocyanine: a subphthalocyanine ana-
logue bearing a six-membered ring unit. Angew chem Int Ed. 2010;49:8000–8003.
[14] Claessens CG, González-Rodríguez D, McCallum CM, et al. On the mechanism of boron-
subphthalocyanine chloride formation. J Porphyr Phthalocyanines. 2007;11:181–188.
[15] Ting Q, Chunjing L.
[Method for func-
tionally substituted dicyanoethylene compounds]. People’s Republic of China patent CN
106966922 A, 2017, Jul. 21.
[16] Gotfredsen H, Broløs L, Holmstrøm T, et al. Acetylenic scaffolding with subphthalocyanines
– synthetic scope and elucidation of electronic interactions in dimeric structures. Org Biomol
Chem. 2017;15:9809–9823.
[17] Sánchez-Molina I, Claessens CG, Grimm B, et al. Trapping fullerenes with jellyfish-like
subphthalocyanines. Chem Sci 2013;4:1338–1344.
[18] Wang Y, Uchihara K, Mori S, et al. 1,3-dithiole-2-one-fused subphthalocyanine and sub-
porphyrazine: synthesis and properties arising from the 1,3-dithiole-2one units. Org Lett
2019;21:3103–3107.
[19] Steinkopf W, Jacob H, Penz H. Studien in der Thiophenreihe. XXVI. Isomere Bromthiophene
und die Konstitution der Thiophendisulfonsäuren [Studies in the thiophene series. XXVI.
Isomeric bromothiophenes and the constitution of thiophene disulfonic acids]. Liebigs Ann
Chem. 1934;512:136–164.
[20] Nielsen CB, Bjørnholm T. New regiosymmetrical dioxopyrrolo- and dihydropyrrolo-
functionalized polythiophenes. Org Lett 2004;6:3381–3384.