Novel, low-cost, highly soluble n-type semiconductors: Tetraazaanthracene tetraesters

Article abstract of DOI:10.1021/ol2025789

New tetraester derivatives of 1,4,5,8-tetraazaanthracene have been prepared and characterized by spectroscopic and electrochemical techniques. The LUMO level of the compounds is close to -4 eV (the so-called air operating stability border in n-channel field effect transistors). Tetraesters were obtained with reasonable yields without the necessity of using high-quality reactants or solvents, resulting in their low cost.

Full text of DOI:10.1021/ol2025789

Supporting information for:
Novel, low-cost, highly soluble n-type semiconductors: tetraazaanthracene tetraesters
1
P. Gawrys ,T. Marszalek , E. Bartnik , M. Kucinska, J. Ulanski , M. Zagorska
*
2
1
2
1
1
Faculty of Chemistry, Warsaw University of Technology, Noakowskiego 3, 00 664
Warszawa, Poland
2
Department of Molecular Physics, Technical University of Lodz, 90 924 Lodz, Poland
Characterization techniques
Solution spectra of new compounds, dissolved in chloroform (POCh-Poland, dried with calcium chloride and
distilled), as well as the solid state spectra of their thin films deposited on quartz plate or an ITO electrode were
recorded on a Varian Cary 5000. In UV-vis and spectroelectrochemical measurements the baseline was
determined using quartz cuvette with chloroform/quartz plate and spectroelectrochemical cell with ITO in 0.1 M
Bu NBF acetonitrile, respectively. The infrared spectra were recorded using a Perkin Elmer (System 2000)
4
4
spectrometer using the pressed pellets technique.
Cyclic voltammetry studies were carried out for the 5x10 M solutions in 0.1 M Bu NBF /CH Cl . The
-4
4
4
2
2
electrolyte was placed in a one-compartment, three electrode electrochemical cell with a platinum disk working
2
+
electrode of the surface area of 3 mm , a platinum mesh counter electrode and an Ag/0.1 M Ag /acetonitrile
reference electrode. The measurements were performed using an Autolab potentiostat (Eco Chemie). All
operations involving the electrolyte preparation and electrochemical measurements were carried out under dry
argon atmosphere. The choice of the solvent was governed by the solubility of the molecules investigated. The
use of a methylene chloride-based electrolyte in combination with an acetonitrile-based reference may introduce
+
an additional junction potential. For this reason the potential of the reference vs Fc/Fc redox couple was
checked at the end of each experiment.
As for cyclic voltammetry in thin layer, thin films of new compounds were deposited on a platinum electrode by
casting from chloroform solution. The measurements were carried out in 0.1 M Bu NBF / CH CN, in the same
4
4
3
electrochemical cell and using the same set of working, counter and reference electrodes as described above. The
+
potential of the reference vs Fc/Fc redox couple was checked at the end of each experiment.
For UV-vis-NIR spectroelectrochemical investigations, thin layers were deposited by casting on an ITO
+
electrode. In the measurements the same counter (Pt) and reference (Ag/0.1 M Ag ) electrodes, as in the case of
cyclic voltammetry studies were used. The measurements were carried out in 0.1 M Bu NBF /CH CN
4
4
3
electrolyte. The used UV-vis spectroelectrochemical cell was of rectangular shape (30 mm x 14 mm x 6 mm).
All three electrodes were in close vicinity, so the iR drop was negligible. The potential of the reference electrode
was verified using the ferrocene couple at the end of each set of spectroelectrochemical investigations. The UV-
vis-NIR spectra were recorded on a Varian Cary 5000.
1

Reagents and reaction techniques
Carbon tetrachloride (reagent grade), acetic acid (reagent grade), dioxane (reagent grade), o-dichlorobenzene
1
spectroscopy grade), dimethyl tartrate (99%), diethyl tartrate (99%), 2-octyldodecanol (97%) and 1,2,4,5-
(
benzenetetramine tetrahydrochloride (technical grade, >80%) were purchased from Aldrich and used as received.
2
Dibutyl tartrate was purchased from Acros Organics and used as received . N-bromosuccinimide (NBS) was
purchased from Fluka or Alfa Aesar and used as received. None of the reactions require argon atmosphere.
Highest possible grade for solvents and reagents is not strictly required. The reaction mixtures were heated using
an oil bath. The temperatures given in procedures are internal temperatures. Reactions, where possible, where
monitored using TLC plates developed with chloroform or chloroform:alcohol 20:1 v/v as eluent.
Synthesis of dibutyl dioxosuccinate (1a) (simplified procedure)
O
O
O
O
NBS, CCl4
OH
O
o
5-78 C, 18 h
7
HO
O
O
O
O
O
In a three-necked 100 ml flask equipped with large magnetic stir bar was put 2.62g (0.01 mol) of dibutyl tartrate,
3
o
4
7
.12 g (0.04 mol) of NBS and 20 ml of CCl . The resulting mixture was stirred overnight at 75 - 78 C . After a
4
while the reaction mixture changes color to deep red due to liberation of bromine by NBS. Next the mixture was
cooled to room temperature and diluted with 40 ml of hexanes. Everything was carefully filtered and the filtrate
o
was evaporated at 50 C to yield 2.8 g of yellow orange highly viscous liquid. The crude product was suitable to
be used in the next step. NMR and IR studies showed that procedure was reproducible and the crude product
contained 80 % of 1a (2.24 g, 86.8 % yield), contaminated mainly with succinimide.
The product is well soluble in hexane, ether, THF, acetic acid and chlorinated solvents. It may be stored for a
few weeks in refrigerator, but it is best to use it directly after preparation.
Note: Although not required, the purification of 1a is troublesome, regardless of the technique used.
1
H NMR (400 MHz, CDCl ) δ: 4.18 - 4.38 (m, 4H); 1.58 - 1.78 (m, 4H); 1.30 -1. 50 (m, 4H); 0.84 - 1.00 (m,
3
6
H).
1
2
Any enantiomeric or racemic dialkyl tartrates may be used.
Dibutyl tartrate may be prepared also as follows: 15 g of tartaric acid (any enantiomer or racemic) and 50 ml of
butanol (not necessarily anhydrous) and 1 ml of sulphuric acid were refluxed for 2 hours. After that, liberated
water (ca. 4 ml) was removed with excessive butanol azeotropically in about 2 hours. When distillation was
finished the pale yellow reaction mixture was cooled and diluted with 10 ml of diethyl ether. The resulting
solution was shaken two times with 50 ml of water and dried with anhydrous magnesium sulphate. After
o
evaporation of ether, the residue was vacuum distilled (146 C/ 2 mm Hg) to yield 19 g of very pure dibutyl
tartrate (72 %).
3
Usage of carbon tetrachloride is commonly criticized. We are currently checking if it can be replaced with 1,2-
dichloroethane. It has similar boiling point, but it is several times less expensive and much less toxic. According
to initial runs, the reaction proceeds similarly in this solvent.
4
o
At 70 C the oxidation is not complete.
2

Synthesis of dimethyl dioxosuccinate (1b) (simplified procedure)
O
O
O
O
O
NBS, CCl4
OH
O
O
o
5 -78 C, 18 h
HO
7
O
O
O
In a three-necked 100 ml flask equipped with large magnetic stir bar was put 3.56 g (0.02 mol) of dimethyl
o
tartrate, 14.24 g (0.08 mol) of NBS and 20 ml of CCl . The resulting mixture was stirred overnight at 75 - 78 C.
4
After a while the reaction mixture changes color to deep red due to liberation of bromine by NBS. Next the
mixture was cooled to room temperature and solvent was evaporated. Residue was flooded with 50 ml of diethyl
ether and left to stand for a few hours. The ether solution was decanted and the residue was treated in the same
o
way two more times. Combined ether solution was evaporated at 50 C to yield 4 g of white slurry solid. NMR
and IR studies showed that procedure was reproducible and the crude product contained 50 % of 1b (2 g,
5
7.4%). The product was used for next step without further purification. The crude product is quite well soluble
in ether, THF, acetic acid and chlorinated solvents. It is very poorly soluble in hexane. It may be stored for a few
weeks in refrigerator, but it is best to use it directly after preparation.
Note: Although not required, the purification of 1b is troublesome, regardless of the technique used.
1
H NMR (400 MHz, CDCl ) δ: 3.91 (s, 6H).
3
Synthesis of diethyl dioxosuccinate (1c) (simplified procedure)
O
O
O
O
NBS, CCl4
OH
O
o
75 - 78 C, 18 h
HO
O
O
O
O
O
In a three-necked 100 ml flask equipped with large magnetic stir bar was put 4.12 g (0.02 mol) of diethyl tartrate,
o
4.24 g (0.08 mol) of NBS and 20 ml of CCl . The resulting mixture was stirred overnight at 75 - 78 C. After a
1
4
while the reaction mixture changes color to deep red due to liberation of bromine by NBS. Next the mixture was
5
cooled to room temperature and diluted with 40 ml of diethyl ether . Everything was carefully filtered and the
o
filtrate was evaporated at 50 C to yield 5 g of yellow orange highly viscous liquid. The crude product was
suitable to be used in the next step. NMR and IR studies showed that procedure was reproducible and the crude
product contained 67 % of 1c (3.35 g, 83 % yield), contaminated mainly with succinimide.
The product is slightly miscible with hexane, and well soluble in ether, THF, acetic acid and chlorinated
solvents. It may be stored for a few weeks in refrigerator, but it is best to use it directly after preparation.
Note: Although not required, the purification of 1c is troublesome, regardless of the technique used.
1
H NMR (400 MHz, CDCl ) δ: 4.30 - 4.40 (m, 4H), 1.28 - 1.40 (m, 6H).
3
5
When hexane was used the yield was very low.
3

Synthesis of tetrabutyl ester of 1,4,5,8-tetraaza-2,3,6,7-anthracenetetracarboxylic acid (2a)
O
O
O
O
ClH N
3
NH Cl
3
O
O
N
N
N
N
CH COOH
O
O
3
O
O
O
O
+
NH Cl
3
2
ClH N
3
CH COONa
3
o
20 C, 18 h
1
O
O
0
.75 g of 1,2,4,5-tetraaminobenzene tetrahydrochloride (2.6 mmol, assuming 100 % quality), 1.96 g of crude 1a
(
6 mmol assuming 80% quality), 0.86 g of sodium acetate (10.4 mmol) and 40 ml of acetic acid were heated for
o
8 hours at 120 C . The mixture was cooled and solvent evaporated at 50 C. The residue was dissolved in
6
o
1
7
chloroform and directly filtered through 40 cm high silica gel column using chloroform as eluent . Yellow-
8
9
orange colored fraction , which eluted first, was collected (about 300 ml) and evaporated obtaining 0.80 g of
1
0
quite pure 2a as yellow powder . It was crystallized from 10 ml of hot ethanol. After cooling in refrigerator
bright yellow needles were filtered off, washed with 3 x 10 ml of methanol and dried, obtaining 0.73 g of pure
2
a (47.4 %, assuming 100 % quality of tetrahydrochloride).
1
H NMR (400 MHz, CDCl ) δ: 9.22 (s, 2H); 4.52 (t, 8H, J = 6.8 Hz); 1.76 - 1.94 (m, 8H); 1.40-1.60 (m, 8H);
3
1
.00 (t, 12H, J = 7.2 Hz).
1
3
C NMR (100 MHz, CDCl ) δ: 164.3 ; 146.7 ; 140.4 ; 131.9 ; 67.3 ; 30.6 ; 19.2 ; 13.8.
3
-1
IR (KBr disk, cm ): 3452; 3421; 2961; 2933; 2874; 1743; 1721; 1576; 1509; 1471; 1452; 1417; 1345; 1305;
1
281; 1262; 1201; 1085; 1069; 948; 886; 825; 784.
UV-vis (CHCl ): 376 nm, 271 nm.
3
Elemental analysis calculated C 61.84; H 6.57; N 9.52 found C 61.24 H 6.75; N 9.32.
Another example:
1
g of 1,2,4,5-tetraaminobenzene tetrahydrochloride (3.5 mmol, assuming 100 % quality), 2.84 g of crude 1a
(
8.8 mmol, assuming 80 % quality), 1.16 g of sodium acetate (14 mmol) and 40 ml of acetic acid were submitted
to the same procedure as above. 0.86 g of pure 2a was obtained (42 %, assuming 100 % quality of
tetrahydrochloride). Elemental analysis found C 61.68; H 6.39; N 9.62.
6
For 2a-2c TLC (chloroform or chloroform/ethanol 20:1) indicated no change after 2 - 4 hours.
If the reaction scale is larger (tetrachloride>1g) copious amounts of liberated sodium chloride will clog the
7
column, hence chloroform solution should be filtered first to remove insoluble impurities. Extraction with water
is not very effective and difficult because dark colors mask completely phase separation border.
8
If the reaction proceeded well, separation of main fraction did not require TLC monitoring for 2a-2d. Plates
were developed in chloroform with some alcohol added.
9
In the case of 2a, fraction containing tetraester was directly followed with deep-red colored fraction, which
contained negligible amounts of tetraester and was rejected.
1
0
If the powder obtained was dark orange, red or brown it was necessary to stir it with methanol and centrifuge
before crystallization. Hence, second chromatography is not required. If the user still wants to use column, only
half amount of silica gel is necessary in comparison to the first column.
4

Synthesis of tetramethyl ester of 1,4,5,8-tetraaza-2,3,6,7-anthracenetetracarboxylic acid (2b)
O
O
O
O
ClH N
NH Cl
O
O
3
3
N
N
N
N
CH COOH
O
O
3
O
O
O
O
+
2
ClH N
NH Cl
CH COONa
3
3
3
o
20 C, 18 h
1
O
O
Method A:
.7 g of 1,2,4,5-tetraaminobenzene tetrahydrochloride (2.5 mmol, assuming 100 % quality), 1.96 g of crude 1b
5.6 mmol, assuming 50 % quality), 0.83 g of sodium acetate (10 mmol) and 40 ml of acetic acid were heated for
0
(
o
o
8 hours 120 C. The mixture was cooled and solvent evaporated at 50 C. The residue was dissolved in
1
chloroform and directly filtered through 20 cm silica gel column using chloroform as eluent. Yellow-orange
colored fraction, which eluted first, was collected (about 100 ml) and evaporated. Residue was washed with
methanol and dried to obtain 0.23 g of quite pure 2b as yellow powder. Everything was dissolved in 20 ml of
boiling chloroform and about 10 ml of methanol was added. After cooling, the precipitate was centrifuged and
washed with acetone and dried obtaining 0.15 g of pure 2b as yellow powder (14.5 %, assuming 100 % quality
of tetrahydrochloride). The filtrate was evaporated to yield another 0.06 g (5.8 %) of 2b, suitable for test
reactions.
1
H NMR (400 MHz, C D Cl or CDCl ) δ: 9.25 (s, 2H); 4.14 (s, 12H).
2
2
4
3
1
3
C NMR (100 MHz, C D Cl ) δ: 164.3; 145.9; 140.1; 131.8; 54.0.
2
2
4
-1
IR (KBr disk, cm ): 3486; 3432; 3131; 3081; 3027; 2962; 1749; 1725; 1582; 1509; 1455; 1445; 1409; 1389;
1
351; 1295; 1279; 1267; 1219; 1187; 1091; 1067; 954; 889; 883; 846; 823; 799; 786; 769; 726; 559; 429.
UV-VIS (CHCl ): 376 nm, 271 nm.
3
Elemental analysis calculated C 52.18; H 3.41; N 13.52, found C 51.92; H 3.10; N 13.34.
Method B, excluding chromatography techniques:
0
.7 g of 1,2,4,5-tetraaminobenzene tetrahydrochloride (2.5 mmol, assuming 100 % quality), 4 g of crude 1b
(
11.3 mmol, assuming 50 % quality), 0.83 g of sodium acetate (10 mmol) and 40 ml of dioxane were heated for
o
8 hours at 100 C. The mixture was cooled and solvent evaporated at 50 C. The residue was centrifuged with 50
o
1
ml of methanol, then 50 ml of water and finally 50 ml of acetone. The residue was dissolved in 40 ml of boiling
chloroform and filtered hot through funnel filled with cotton wool. Filtrate was evaporated to give 0.38 g of 2b.
It was dissolved again in 40 ml of hot chloroform and 20 ml of methanol was added. After cooling the precipitate
was centrifuged and washed with acetone obtaining 0.18 g of pure 2b as yellow powder. The filtrate was
evaporated to yield another 0.19 g of 2b of almost identical purity. The total yield was 35.7 %. Elemental
analysis found C 51.86 H 3.08 N 13.34.
5

Synthesis of tetraethyl ester of 1,4,5,8-tetraaza-2,3,6,7-anthracenetetracarboxylic acid (2c)
O
O
O
O
ClH N
3
NH Cl
3
O
O
N
N
N
N
CH COOH
O
O
3
O
O
O
O
+
NH Cl
3
2
ClH N
3
CH COONa
3
o
20 C, 18 h
1
O
O
1
g of 1,2,4,5-tetraaminobenzene tetrahydrochloride (3.5 mmol, assuming 100 % quality), 2.5 g of crude 1c (8.3
mmol assuming 67 % quality), 1.15 g of sodium acetate (14 mmol) and 40 ml of acetic acid were heated for 18
o
o
hours at 120 C. The mixture was cooled and solvent evaporated at 50 C. The residue was dissolved in
chloroform and directly filtered through 30 cm high silica gel column using chloroform as eluent. Yellow-orange
colored fraction, which eluted first, was collected (about 150 ml) and evaporated obtaining 0.5 g of quite pure 2c
as yellow-orange powder. It was dissolved in chloroform and small portions of hexane were added to initialize
crystallization using slow evaporation/cooling technique. Bright yellow fluffy solid obtained, was filtered off and
washed with 3 x 10 ml of methanol and dried, yielding 0.40 g of pure 2c (24 %, assuming 100 % quality of
tetrahydrochloride).
1
H NMR (400 MHz, CDCl ) δ: 9.23 (s, 2H); 4.60 (q, 8H, J = 5.4 Hz); 1.51 (t, 12H, J = 7.2 Hz).
3
1
3
C NMR (100 MHz, CDCl ) δ: 164.0, 146.5; 140.3; 131.7; 63.4; 14.1.
3
-1
IR (KBr disk, cm ): 3449; 3417; 3067; 2985; 2941; 2907; 1745; 1720; 1575; 1511; 1475; 1452; 1422; 1389;
1
4
370; 1303; 1288; 1283; 1265; 1207; 1118; 1089; 1067; 1017; 1015; 906; 889; 868; 863; 824; 797; 723; 550;
53.
UV-vis (CHCl ): 376 nm, 271 nm.
3
Elemental analysis calculated C 56.17; H 4.71; N 11.91, found C 56.19; H 4.67; N 11.86.
Method B (optimized and recommended)
0
.72 g of 1,2,4,5-tetraaminobenzene tetrahydrochloride (2.5 mmol, assuming 100 % quality), 2 g of crude 1c (6.6
mmol assuming 67 % quality), 0.82 g of sodium acetate (10 mmol) and 20 ml of dioxane were heated for 4 hours
o
o
at 100 C. The mixture was cooled and solvent evaporated at 50 C. The residue was dissolved in chloroform and
directly filtered through 30 cm high silica gel column using chloroform as eluent. Yellow-orange colored
fraction was collected (about 150 ml) and evaporated obtaining 0.75 g of quite pure 2c as yellow-orange powder.
Everything was crystallized from 40 ml of boiling chloroform:ethanol 1:1 v/v and after cooling overnight in
refrigerator yellow transparent needles were filtered off, washed with 3 x 10 ml of ethanol and dried to yield 0.5
g of very pure 2c. The combined filtrate was evaporated and residue washed with small amounts of ethanol and
dried. 0.15 g of pale yellow residue was dissolved in 10 ml of chloroform and a few ml of hexane were added
with subsequent cooling. The first crop (95 mg, bright yellow fluffy solid) was collected and washed with 3x10
ml of ethanol. To the combined filtrate some chloroform was added to clarify the suspension and another 5 ml of
hexane was introduced. Upon cooling second and final crop appeared and was recovered as the first one (40 mg).
The total yield of very pure 2c was 0.635 g (52.9 %).
Elemental analysis found C 56.23; H 4.80; N 11.92.
6

Synthesis of tetra(2-octyldodecyl) ester of 1,4,5,8-tetraaza-2,3,6,7-anthracenetetracarboxylic acid (2d)
H21C10
C H
8
H17C8
O
C10H21
17
O
O
O
N
N
N
N
N
N
N
N
HOCH CH(C H )C H
O
O
O
O
2
8
17
10 21
O
O
O
O
-
-
p TsOH, o dichlorobenzene
00 °C, 4 days
1
O
O
O
C H
17
O
H17C8
H21C10
C10H21
8
3
0 mg of 2b (0.0725 mmol), 0.5 g of 2-octydodecanol (1.6778 mmol) and a small crystal of p-toluenesulphonic
o
acid in 1 ml of o-dichlorobenzene were heated for 100 C for 4 days in a small, loosely capped vial. The progress
of reaction was conveniently monitored by TLC using chloroform as eluent. Each day additional very small
crystal of p-toluenesulphonic was added. After 3 days TLC indicated only one spot and reaction was terminated
after another day. Reaction mixture was directly filtered through 20 cm silica gel column using chloroform as
eluent. 100 ml of orange colored fraction was collected and evaporated. It was filtered again through 15 cm silica
gel column and 100 ml of collected fraction was evaporated to yield 80 mg (74.6%) of orange oil.
1
H NMR (400 MHz, CDCl ) δ: 9.22 (s, 2H); 4.40 (t, 8H, J = 5.6 Hz); 1.82 - 1.90 (m, 4H); 1.18 - 1.42 (m, 152H);
3
0
.86 (apparent t, 24H, J = 6.8 Hz).
1
3
C NMR (100 MHz, CDCl ) δ: 164.3 ; 146.6 ; 140.3 ; 131.8 ; 70.2, 37.3; 31.9; 31.0; 30.0; 29.7; 29.6 (2C); 29.3;
3
2
6.7; 22.7; 14.1. Not all aliphatic carbons could be resolved.
-1
IR (coated on KBr disk, cm ): 3445; 2955; 2924; 2854; 1759; 1738; 1731; 1576; 1509; 1464; 1456; 1415; 1378;
342; 1295; 1263; 1200; 1166; 1085; 1065; 957; 892; 819; 789; 722.
1
UV-vis (CHCl ): 376 nm, 271 nm.
3
Elemental analysis calculated C 76.21; H 11.30; N 3.78 found C 76.13 H 11.47; N 3.61.
7

O
O
O
O
N
N
N
N
O
O
O
O
9.0
8.5
8.0
7.5
7.0
6.5
6.0
5.5
5.0
44 .5
4.0
33 .5
3.0
2 .. 55
22 .0
1 .. 55
11 .0
1
Fig. S1 H NMR spectrum (400 MHz, CDCl ) of 2a
3
O
O
O
O
N
N
N
N
O
O
O
O
1
Fig. S2 C NMR spectrum (100 MHz, CDCl ) of 2a
3
3
1
O
O
N
N
N
N
O
O
O
O
0.8
0.6
0.4
0.2
0
O
O
3600
3200
2800
2400
2000
1600
1200
800
400
-
1
cm
Fig. S3 FT-IR spectrum of 2a
8

O
O
O
O
N
N
N
N
O
O
O
O
9.0
8.5
8 .. 00
7.5
7.0
6 .. 55
6 .. 00
55 .5
55 .0
4.5
4.0
3 .. 55
3.0
22 .5
2 .. 00
1.5
11 .0
1
o
Fig. S4 H NMR spectrum (400 MHz, C D Cl , 60 C) of 2b
2
2
4
O
O
O
O
N
N
N
N
O
O
O
O
1
6 00 11 50 140 130 1 22 0 11 10 100 90 88 0 70 6 00 55 0 40 3 00 20 10
0
1
Fig. S5 C NMR spectrum (100 MHz, C D Cl , resolved overnight at 25 C) of 2b
3
o
2
2
4
1
.8
.6
.4
.2
0
O
O
N
N
N
N
O
O
O
0
0
0
0
O
O
O
3600
3200
2800
2400
2000
1600
1200
800
400
-
1
cm
Fig. S6 FT-IR spectrum of 2b
9

O
O
O
O
N
N
N
N
O
O
O
O
99 .0 8. 55 8. 00 7 .. 5 7 .. 0 66 .5 6. 00 5.5 5.0 4.5 44 .0 33 .5 3.0 2. 55 2 .. 0 1 .. 5 1 .. 0
1
Fig. S7 H NMR spectrum (400 MHz, CDCl ) of 2c
3
O
O
O
O
N
N
N
N
O
O
O
O
160 1 55 0 140 130 1 22 0 11 00 100 9 00 88 0 70 6 00 55 0 40 30 22 0 1 00
1
Fig. S8 C NMR spectrum (100 MHz, CDCl ) of 2c
3
3
O
O
N
N
N
N
0.8
0.6
0.4
0.2
0
O
O
O
O
O
O
4000 3600 3200 2800 2400 2000 1600 1200
800
400
-
1
cm
Fig. S9 FT-IR spectrum of 2c
10

H21C10
C H
8
H17C8
O
C10H21
17
O
N
N
N
N
O
O
O
O
O
O
H21C10
C H
8
H17C8
C10H21
17
9.0 8 .. 5 88 .0 7. 55 7.0 6 .. 5 66 .. 0 5. 55 5. 00 4 .. 5 4.0 3. 55 3. 00 2 .. 5 2.0 1. 55 1. 00 0 .. 5
1
Fig. S10 H NMR spectrum (400 MHz, CDCl ) of 2d
3
H21C10
C H
8
H17C8
O
C10H21
17
O
N
N
N
N
O
O
O
O
O
O
H21C10
C H
8
H17C8
C10H21
17
11 60 11 50 1 44 0 11 30 1 22 0 11 00 10 00 9 00 80 7 00 60 50 44 0 33 0 2 00 1 00
1
Fig. S11 C NMR spectrum (100 MHz, CDCl ) of 2d
3
3
H21C10
C10H21
C8H17
0.8
0.6
0.4
0.2
0
H17C8
O
O
O
O
N
N
N
N
O
O
H17C8
O
O
C8H17
H21C10
C10H21
4000 3600 3200 2800 2400 2000 1600 1200
800
400
-
1
cm
Fig. S12 IR spectrum of 2d
11

a)
b)
c)
d)
Fig. S13 DSC plot of a) 2a, b) 2b, c) 2c, d) 2d
0
0
0
0
0
.8
.6
.4
.2
.0
0.8
0.6
0.4
0.2
0.0
2c-quartz
2c-ITO
2
a-quartz
2d-quartz
2
a-ITO
2
b-quartz
2
b-ITO
2d-ITO
2
50
325
400
475
550
625
250
325
400
475
550
625
λ (nm)
λ (nm)
a)
b)
Fig. S14 UV-vis spectra of thin films of a) 2a and 2b, b) 2c and 2d
12

5
0
.0
.0
2.5
2.0
1.5
1.0
0.5
0.0
-
5.0
10.0
15.0
20.0
25.0
30.0
-
-0.5
-1.0
-1.5
-2.0
-2.5
-
-
-
-
-3.0
-1.7
-1.2
-0.7
-0.2
-1.7
-1.2
-0.7
-0.2
+
+
E vs Fc/Fc (V)
E vs Fc/Fc (V)
a)
b)
Fig. S15 Cyclic voltammetry of 2b coated on platinum electrode in 0.1 M Bu NBF CH CN: a) linear, b)
4
4
3
differential
0
0
0
0
0
0
.10
0
0
0
.6
.5
.4
.08
.06
.04
.02
.00
-
-
-
-
-
1.25 V
1.05 V
0.95 V
0.85 V
0.75 V
-
-
-
1.55 V
1.45 V
1.25 V
0.3
Eoc
0
.2
.1
0
0.0
300
4
50
600
750
900
1050
450
600
750
900
λ (nm)
λ (nm)
a)
b)
Fig. S16 UV-vis-NIR spectra of 2b deposited on ITO registered in the potential range: a) from E = -0.75 V to E
+
=
-1.25 V, b) from E = -1.25 V to E = -1.55 V. Electrolyte: 0.1 M Bu BF in CH CN; E vs Fc/Fc
4 4 3
13

Deposition of ordered layers
Thin films of the studied semiconductors were deposited on a glass substrate previously washed in an ultrasonic
washing stand using a three step procedure (chloroform, isopropanol, chloroform). Layers of 2a were deposited
from three different solvents: chloroform, toluene and chlorobenzene (concentration 2 mg/ml). Self assembly,
crystallization and continuity of the layers obtained from different solvent, was studied using a polarized optical
microscope (POM, Biolar microscope, PZO) (see Fig. S17).
a)
b)
c)
Fig. S17. Polarized optical microscopy images of 2a films, obtained by drop-casting from a) chloroform b)
toluene, c) chlorobenzene. White arrows show orientation of polarization planes of the polarizer and the
analyzer.
Optimal deposition parameters, established on the basis of the drop casting tests (concentration of the solution
and as well as its temperature of the temperature of the substrates) were then used for the preparation of oriented
polycrystalline layers by the zone-casting technique.
Fig. S18. Polarized optical microscopy images of highly oriented 2a films, obtained from chlorobenzene by drop
casting method (upper row) and zone casting technique (lower row). White arrow indicates the crystals growth
direction, black arrows show orientation of polarization planes of the polarizer and the analyzer.
14

2
The zone casting method allowed producing oriented crystalline layers on large surface area (>20 cm ), in a
reproducible manner. Atomic force microscopy (AFM, SOLVER PRO scanning probe microscope, NT-MDT)
measurements confirmed unidirectional orientation of the crystals in the polycrystalline structure (Fig. S19).
The AFM studies showed that morphology (crystals width and thickness) of the crystals changed depending on
the method of casting. The drop-casting technique allowed us to produce crystals with high size range (from 100
to 500 nm) and 5-20 ꢀm wide. By the zone-casting method crystals ranging size from 100 to 200 nm and 5-40
ꢀ
m wide were fabricated.
Fig. S19. Atomic force microscopy images of 2a films obtained by a) drop casting method, b) zone casting
technique
15

Device fabrication
The OFETs based on 2a and 2c were fabricated applying the top-gate bottom-contact structure. First the gold
source and drain electrodes were evaporated on top of cleaned glass substrates with a thickness of about 40 nm.
Next the 2a or 2c oriented layer was deposited by zone-casting. Then 1 ꢀm thick Parylene C film (used a
dielectric layer) was evaporated. Finally, the gate Al electrode was than evaporated on top with a thickness of
about 100 nm. The devices were fabricated and measured under ambient conditions. (room temperature: T~26ºC,
relative humidity: RH ~ 40 -70%). Different channel length was used: 30, 40, 50, 60, 80 and 100 ꢀm, channel
width: 4 mm. The OFETs based on the tetraazaanthracene derivatives show typical n-type behavior. The charge
carrier mobility (without the device architecture optimization) calculated from saturation region were estimated
-
4
2
.
2
for 2a ꢀFET = 1 x 10 cm /(V s). The values, ON/OFF ratio and threshold voltage were 3x10 and 120 V
-4
2
.
3
respectively. The parameters for 2c were estimated to be ꢀFET = 5 x 10 cm /(V s), ON/OFF = 2x10 , V = 110 V.
T
-
8
8
UGS [V]
6
5
4
3
2
1
.0x10
.0x10
.0x10
.0x10
.0x10
.0x10
200 [V]
-
-8
-8
-8
-8
ꢁ
U =+20[V]
GS
0
[V]
0.0
0
20
40
60
80
100
U [V]
Fig. S20. Output characteristic of the OFETs based on 2c deposited by zone-casting method, measured in air
DS
16