Substituted triphenylamines as building blocks for star shaped organic electronic materials

Article abstract of DOI:10.1039/c4nj01695e

A versatile synthetic protocol toward a series of various substituted triphenylamine derivatives serving as building blocks for organic electronic materials was developed. Key steps during synthesis were either Ullmann condensations or nucleophilic aromatic substitutions giving rise to structural modification of triphenylamines and their electronic nature. In turn, these scaffolds were exemplarily attached to a dendritic tris(2-thienyl)benzene core affording star shaped organic semiconducting materials which were characterized regarding their photo-physical, electro-chemical and thermal properties. A strong influence of the substituent's nature on both photo-physical and morphological thin film characteristic of star shaped target compounds was observed. The applicability of these materials in organic electronic devices was demonstrated in an organic field effect transistor configuration yielding a hole mobility of nearly 10^-3 cm² V^-1 s^-1. The performance of the materials can be correlated to the substituents applied. This journal is

Full text of DOI:10.1039/c4nj01695e

NJC
View Article Online
View Journal
PAPER
Substituted triphenylamines as building blocks for
star shaped organic electronic materials†
Cite this:DOI: 10.1039/c4nj01695e
Daniel Lumpi,‡a Brigitte Holzer,‡a Johannes Bintinger,ab Ernst Horkel,*a
Simon Waid,c Heinz D. Wanzenbock,c Martina Marchetti-Deschmann,d
¨
Christian Hametner,a Emmerich Bertagnolli,c Ioannis Kymissisb and
Johannes Frohlicha
¨
A versatile synthetic protocol toward a series of various substituted triphenylamine derivatives serving
as building blocks for organic electronic materials was developed. Key steps during synthesis were
either Ullmann condensations or nucleophilic aromatic substitutions giving rise to structural modification
of triphenylamines and their electronic nature. In turn, these scaffolds were exemplarily attached to
a dendritic tris(2-thienyl)benzene core affording star shaped organic semiconducting materials which
were characterized regarding their photo-physical, electro-chemical and thermal properties. A strong
influence of the substituent’s nature on both photo-physical and morphological thin film characteristic
of star shaped target compounds was observed. The applicability of these materials in organic electronic
devices was demonstrated in an organic field effect transistor configuration yielding a hole mobility of
Received (in Montpellier, France)
30th September 2014,
Accepted 15th December 2014
DOI: 10.1039/c4nj01695e
nearly 10ꢀ3 cm2 Vꢀ1
s
ꢀ1. The performance of the materials can be correlated to the substituents applied.
www.rsc.org/njc
on the molecular structure and the morphology of the semicon-
ducting material.2,7 Therefore, a broadly applicable material with
Introduction
Organic electronic (OE) thin film devices have gained raising high charge carrier mobility independent of processing techniques
interest from academia and industry due to the variety of is desirable.8,9 The quest for novel compounds is an ongoing
possible applications with enormous commercial potential.1 process in material science.
This field covers the application of conducting and semi-
conducting organic materials within electronic devices such as history in the field of OE.10 While high hole mobilities (up to
organic field effect transistors (OFETs),2 organic light emitting 1.3 cm2 Vꢀ1 ꢀ1) have been observed for thin films of poly-
Poly- and oligothiophene based compounds have a long
s
devices (OLEDs)3 and organic photovoltaics (OPVs).4,5 The appli- (3-hexylthiophene) (P3HT),11 the material cannot be used in the
cation of these materials in the field of organic electronics allows light emitting layers of an OLED due to fluorescent self-
for solution processability, thus rendering OE-technology com- quenching caused by strong p–p stacking.12 Another useful
patible with established high-throughput printing techniques class of OE materials are modified triphenylamine (TPA) struc-
and with the potential to realize thin, flexible and light-weight tures, which have excellent electron donor and hole transport
devices with low manufacturing costs.6
properties.13–17 The combination of these structural scaffolds
In general, a high charge carrier mobility is mandatory for results in a material class incorporating good charge carrier
the performance of different devices and strongly depends both transport as a result of the oligothiophene unit and enhanced
luminescent properties by circumventing self-quenching (e.g.
a Institute of Applied Synthetic Chemistry, Vienna University of Technology,
Getreidemarkt 9/163OC, A-1060 Vienna, Austria.
BMA-1T, Fig. 1).18 Moreover, star-shaped compounds are found
to exhibit better solubility and film-forming properties than
their linear counterparts.8,19,20
E-mail: ernst.horkel@tuwien.ac.at
b Department of Electrical Engineering, Columbia University, 520W 120th street,
Combining TPA and thiophenes in C3 symmetric configura-
tions leads to enhanced electronic and luminescent properties
with the benefit of solution processability.21 To realize C3
symmetry, one can think of two possible molecular designs.
One possibility is to use the TPA moiety as a central node with
thiophene moieties as dendrons.19,22,23 Alternatively, a core
(e.g. tris(2-thienyl)benzene) can be used to ensure C3 symmetry
bearing TPA units as side arms.24 Based on these two possible
Suite 1300, 10027 New York, NY, USA
c Institute of Solid State Electronics, Vienna University of Technology, Floragasse 7,
A-1040 Vienna, Austria
d Institute of Chemical Technologies and Analytics, Vienna University of Technology,
Getreidemarkt 9/164IAC, A-1060 Vienna, Austria
† Electronic supplementary information (ESI) available: 1H and 13C NMR spectra
of compounds 6c, 6e–l, 8c, 8e–g, 2, S2–3, absorption and emission spectra of
compounds S1–3 and HOMO–LUMO plots of S1–3. See DOI: 10.1039/c4nj01695e
‡ Daniel Lumpi and Brigitte Holzer contributed equally to this article.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
gain access to TPAs bearing electron withdrawing substituents.
Reaction conditions for the consecutive transformation to the
above-mentioned boronic acid esters also depend on the sub-
stituents’ nature. While lithiation protocols can be applied for
most TPAs under investigation, Miyaura borylation is required
for TPAs bearing ꢀM substituents (Scheme 2).
With these TPA boronic acid esters in hand, a set of three
star shaped compounds S1–3 was realized using a Suzuki
reaction. This nontoxic procedure proofed superior in yield to
a previously reported procedure incorporating a Stille reaction
for the same compound S1.24
Measurements of electro-chemical and optical properties (e.g.
fluorescence spectra and quantum yields) of the star shaped
compounds S1–3 reveal the strong influence of the substituents’
nature on the TPA moiety on the above-mentioned properties. The
materials were characterized with respect to thermal stability,
film-forming upon spin coating and semiconducting behaviour
(charge carrier mobility in OFET configuration) to probe device
applicability. Mobilities of almost 10ꢀ3 cm2 Vꢀ1 sꢀ1 and low
threshold voltages of approx. 0 V were recorded using methyl-
TPA substituted derivative S2.
Experimental section
Synthesis and characterization
Substances purchased from commercial sources were used
as received. Anhydrous N,N-dimethylformamide (DMF), n-butyl-
lithium solution (2.5 M in hexanes) and 1,3,5-tribromobenzene
were purchased from Aldrich Chemical Co. Isopropyl pinacol
borate27 (CAS 61676-62-8), (IPr)Pd(allyl)Cl28 (CAS 478980-03-9)
were synthesized according to literature. Anhydrous tetrahydro-
furan (THF), diethyl ether, toluene were prepared immediately
prior to use by a PURESOLV-plant (it-innovative technology inc.).
Technical grade solvents were distilled prior to use. Analytical
Fig. 1 Linear motifs (top) presented by Noda;18 star shaped molecules
described in this work (bottom).
structural connectivities, this paper focuses on the latter possi- TLC was performed on Merck silica gel 60 F254 plates. Chromato-
bility, since substituent alteration on the three TPA moieties graphic separations at preparative scale were carried out on silica
constituting the outer sphere of the molecule strongly influences gel (Merck silica gel 60, 40–63 mm), alox or RP-gel. Nuclear
glass- and film-forming behaviour8 of the resulting materials, in magnetic resonance (NMR) spectra were obtained using a Bruker
turn affecting device performance.
DPX-200 or Avance DRX-400 Fourier transform spectrometer
Due to the widespread application of TPA-based compounds operating at the following frequencies: DPX-200: 200.1 MHz (1H)
as building blocks in the field of organic electronics,25 the and 50.3 MHz (13C); DRX-400: 400.1 MHz (1H) and 100.6 MHz
synthetic protocol presented in this work is of considerable (13C). The chemical shifts are reported in delta (d) units, parts
interest for the molecular design and potential modifications per million (ppm) downfield from tetramethylsilane using
(e.g. in terms of photo-physical properties) toward a broad range solvent residual signals for calibration. Coupling constants
of novel materials. Fine tuning of material properties can be are reported in Hertz; multiplicity of signals is indicated by
realized by applying structurally and electronically diverse TPA using following abbreviations: s = singlet, d = doublet, t = triplet,
motifs. However, due to the broad spectrum of substituents q = quartet. The multiplicity of 13C signals were obtained by
(electron donating or withdrawing) attached to the TPA, no measuring JMOD spectra. UV/VIS absorption and fluorescence
general synthetic protocol is available. Access to TPAs bearing emission spectra were recorded in THF solutions (1 mg mLꢀ1
)
electron donating substituents is often described in literature, with a Perkin Elmer Lambda 750 spectrometer and an Edinburgh
mainly applying Ullmann condensation.26 On the other hand, FLS920, respectively. Cyclic voltammetry was performed using a
this synthetic technique often fails when applied towards the three electrode configuration consisting of a Pt working electrode,
synthesis of electron withdrawing groups. As a result of our a Pt counter electrode and an Ag/AgCl reference electrode and a
investigations to overcome this drawback, we present a general PGSTAT128N, ADC164, DAC164, external, DI048 potentiostat
protocol using nucleophilic aromatic substitution in order to provided by Metrohm Autolab B. V. Measurements were carried
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
out in a 0.5 mM solution in anhydrous DCM (oxidation scan) Several scans were performed from different parts of the
with Bu4NBF4 (0.1 M) as the supporting electrolyte. The solu- samples to check the uniformity of the surface. Final images
tions were purged with nitrogen for 15 minutes prior to were measured from a scanning area of 10 ꢁ 10 mm2 with a tip
measurement. HOMO energy levels were calculated from the velocity in the range from 16 to 18 mm sꢀ1. No image processing
onset of oxidation. The onset potential was determined by the except flattening was done. Roughness values were calculated
intersection of two tangents drawn at the background and as root-mean-square (rms) values.
the rising of oxidation peaks. High-resolution mass spectra
All computations were performed using the Gaussian 09
(HRMS) were acquired as radical cations using a SYNAPT HDMS package, revision A.02.29 For the calculation of HOMO/LUMO
instrument (Waters, Manchester, UK) equipped with a matrix- levels of compounds S1–3, ground state (S0) geometries were
assisted laser desorption/ionization (MALDI) source (MALDI- optimized in gas phase within C3 symmetry using the Becke
HRMS). Samples were applied at 1 mg mLꢀ1 in THF on stainless three parameters hybrid functional with Lee–Yang–Perdew
steel using nitroanthracene (3 mg mLꢀ1 in THF) as MALDI correlation (B3LYP)30,31 in combination with Pople basis set
matrix. All MS spectra were recorded as accurate mass data with 6-311+G*.32 To obtain vertical absorption and emission of
Angiotensin II (m/z 1046.542) as internal lock mass achieving a model compounds Sm1–3, ground state S0 and first excited
mass accuracy of 15–40 ppm (i.e. Dm/z 0.01–0.04 amu). GC-MS singlet state S1 were optimized applying DFT and time-
measurements were conducted on a GC-MS hyphenation from dependent (TD) DFT level of theory using M06-2X33,34 func-
Thermo Finnigan: focus GC with a BGB5 column (l = 30 m, + = tional in combination with the polarized double zeta SVP basis
0.25 mm, 0.25 mm film); DSQ II Quadrupole (EI+ mode). set.35 This parameterization was shown to be superior in terms
The thermal behavior of substances S1–3 was studied with of accuracy for the calculation of vertical transitions.36 Geometry
differential scanning calorimetry (DSC) and thermogravimetric optimizations were performed without symmetry constraints
analysis (TGA) using a Netzsch simultaneous thermal analyzer and solvent effects were included through the polarizable con-
(STA 449 F1 Jupiter). Powder samples with a mass of approx. tinuum model (PCM)37 in its linear response (LR-PCM)38 and
10 mg were lightly pressed into the bottom of open aluminum state specific (SS-PCM)39 formulations, always considering the
pans and heated at 10 1C minꢀ1 from 25 1C to 500 1C under equilibrium time regime (eq.) for the excited state.
N2 gas at a flow rate of 40 mL minꢀ1. The STA 449 type-K
thermocouples were calibrated using indium, tin, bismuth and
zinc metals.
General procedure for the synthesis of 6a–f
Bottom-gate, top-contact OFETs were fabricated on ITO Synthesis was performed according to Goodbrand.26 4-Bromo-
substrates with a 250 nm layer of vacuum deposited parylene-C aniline 4a (1.0 eq.), substituted iodobenzene 5a–f (2.2 eq.), KOH
which serves as a common gate electrode/gate dielectric struc- (7.8 eq.), CuCl (0.04 eq) and 1,10-phenanthroline monohydrate
ture. The substrates were ultrasonically cleaned with water, (0.04 eq.) were suspended in anhydrous toluene. After purging the
acetone and isopropanol followed by a 20 min treatment of apparatus with argon, the reaction mixture was refluxed on a
UV/ozone. Compounds S1–3 (25 nm) were vacuum deposited Dean Stark trap for an appropriate time maintaining the argon
through a shadow mask ( p = 10ꢀ7 Torr, rate 0.1 Å sꢀ1). The atmosphere. Reaction progress was monitored by GC/MS. The
devices were completed by evaporation of gold source and drain reaction mixture was cooled to room temperature and water
electrodes (50 nm) on top. All characterizations were performed was added to solve the potassium hydroxide. Phases were
in ambient air. Current–voltage characteristics of the fabricated separated and the hydrous phase was extracted with toluene
devices were recorded using a Keithley model 4200 semiconduc- three times. The combined organic layer was washed with brine,
tor characterization system. Output and transfer characterization dried over anhydrous sodium sulfate and evaporated to dryness
were recorded for at least four samples of the same channel under reduced pressure. The obtained crude product was purified
width (W = 2000 mm) and channel length (L = 100 mm).
by Kugelrohr distillation and subsequent recrystallisation from
The spin-coated samples were prepared by the following methanol or acetonitrile.
procedure: the ITO substrates were ultrasonically cleaned with
4-Bromo-N,N-bis(4-fluorophenyl)benzeneamine (6c). Accord-
water, acetone and isopropanol followed by blow-drying with ing to the general procedure; 4a (7.04 g, 41 mmol), 5c (20.0 g,
N2. 400 mL of a solution of the organic substance in chloroform 90 mmol), KOH (17.94 g, 320 mmol), CuCl (162 mg, 1.6 mmol)
(20 mg mLꢀ1) was prepared and filtered through a syringe filter and phenanthroline monohydrate (325 mg, 1.6 mmol) were
(Teflon + 25 mm, 0.20 mm). Consecutively 300 mL of the refluxed for 16 h using 250 mL anhydrous toluene. Kugelrohr
prepared organic solution was spin-coated (2000 rpm sꢀ1; 30 s) distillation (130 1C, 3.8 ꢁ 10ꢀ1 mbar) and recrystallization from
directly on ITO.
methanol gave colourless crystals of 6c (10.33 g, 70% of theory).
The typical layer thickness was in the range of 30–50 nm as TLC (silica gel, hexanes): rf = 0.36. FP = 56–58 1C. 1H NMR
revealed from atomic force microscopy (AFM)-measurements. (200 MHz, CD2Cl2): d = 7.38–7.25 (m, 2H), 7.11–6.91 (m, 4H),
For AFM studies, a Multimode V scanner in conjunction with a 6.89–6.72 (m, 8H), 6.91–6.80 (m, 2H) ppm. 13C NMR (50 MHz,
Nanoscope V controller (Veeco Instruments) was used. Image CD2Cl2): d = 159.7 (s, JCF = 242.6 Hz), 147.8 (s), 144.0 (s, JCF
=
processing and data analysis was performed using Gwyddion 2.8 Hz), 132.7 (d), 126.9 (d, JCF = 8.1 Hz), 124.3 (d), 116.7
software version 2.40. Samples were measured in tapping mode (d, JCF = 22.6 Hz), 114.7 (s) ppm. MS (EI): m/z 359 (M+, 100%),
using PPP-NCHR probes obtained from Nanoandmore in air. 279 (22), 184 (14).
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
4-Bromo-N,N-bis[4-(trimethylsilyl)phenyl]benzeneamine (6e). 7b (192 mg, 1.1 mmol) and KOtBu (118 mg, 1.05 eq.) were stirred
According to the general procedure; 4a (7.82 g, 45.5 mmol), 5e in 2 mL of dry DMSO at 120 1C for 12 h. After cooling to r.t., the
(27.62 g, 100 mmol), KOH (19.89 g, 355 mmol), CuCl (180 mg, solvent was removed under reduced pressure. The residue was
1.8 mmol) and phenanthroline monohydrate (360 mg, 1.8 mmol) taken up with chloroform and filtered over a pad of celite. The
were refluxed for 24 h using 250 mL anhydrous toluene. Kugelrohr solvent was removed under reduced pressure and the crude
distillation (130 1C, 6.5 ꢁ 10ꢀ2 mbar) and recrystallization from product purified by column chromatography (40 g silica gel,
acetonitrile gave colourless crystals of 6e (12.9 g, 60% of theory). hexanes/ethyl acetate gradient 20 - 50%) to give 6h as light
1
TLC (silica gel, hexanes): rf = 0.39. FP = 143–145 1C. H NMR brown powder (184 mg, 77% of theory). TLC (silica gel, hexanes/
1
(200 MHz, CDCl3): d = 7.42–7.27 (m, 6H), 7.08–6.90 (m, 6H), ethyl acetate = 1/1): rf = 0.20. FP = 240 1C. H NMR (200 MHz,
0.24 (s, 18H) ppm. 13C NMR (100 MHz, CDCl3): d = 147.9 (s), CDCl3): d = 7.86–7.73 (m, 4H), 7.55–7.44 (m, 2H), 7.22–7.11
146.9 (s), 134.64 (s), 134.61 (d), 132.4 (d), 126.1 (d), 123.5 (d), (m, 4H), 7.06–6.96 (m, 2H), 3.05 (s, 6H) ppm. 13C NMR (50 MHz,
115.5 (s), 112.4 (s), ꢀ0.8 (q) ppm. MS (EI): m/z 469 (M+, 100%), CDCl3): d = 151.0 (s), 144.5 (s), 134.8 (s), 133.6 (d), 129.4 (d), 128.4
454 (61), 452 (58), 220 (65).
(d), 123.2 (d), 119.7 (s), 44.8 (q) ppm.
4-Bromo-N,N-bis[4-(1,1-dimethylethyl)phenyl]benzeneamine
4,40-[(4-Bromophenyl)imino]bisbenzonitrile (6i). Under an
(6f). According to the general procedure; 4a (3.7 g, 21.5 mmol), argon atmosphere, 4-bromoaniline (8.1 g, 47 mmol) and CsF
5f (12.3 g, 47.3 mmol), KOH (9.4 g, 168 mmol), CuCl (85 mg, (14.3 g, 94 mmol) were suspended in 120 mL of dry DMSO.
0.86 mmol) and phenanthroline monohydrate (170 mg, A solution of 7c (12.5 g, 103 mmol) in 50 mL of dry DMSO was
0.86 mmol) were refluxed for 36 h using 130 mL anhydrous added drop wise and the reaction mixture was heated to 110 1C
toluene. Kugelrohr distillation (160 1C, 2.2 ꢁ 10ꢀ1 mbar) and for 50 h. After cooling to r.t., the reaction mixture was poured
recrystallization from acetonitrile gave colourless crystals of 6f on 800 mL of water and extracted three times with 200 mL
(5.8 g, 61% of theory). TLC (silica gel, hexanes): rf = 0.33. FP
=
chloroform. The combined organic layer was washed with
165–168 1C. 1H NMR (200 MHz, CD2Cl2): d = 7.36–7.22 (m, 6H), brine, dried over anhydrous sodium sulfate and the solvent
7.07–6.95 (m, 4H), 6.95–6.82 (m, 2H), 1.32 (s, 18H) ppm. 13C was removed under reduced pressure. The crude product was
NMR (50 MHz, CD2Cl2): d = 148.1 (s), 146.9 (s), 145.3 (s), 132.4 crystallized from ethanol to give 6i (4.05 g, 23% of theory) as
(d), 126.8 (d), 124.8 (d), 124.5 (d), 114.0 (s), 34.8 (s), 31.8 (q) ppm. light brown solid. TLC (silica gel, hexanes/ethyl acetate = 4/1):
MS (EI): m/z 435 (M+, 73%), 422 (96), 420 (100), 204 (35), 176 (47). rf = 0.41. FP = 261–262 1C. 1H NMR (200 MHz, CDCl3): d = 7.58–
7.42 (m, 6H), 7.14–7.03 (m, 4H), 7.03–6.93 (m, 2H) ppm. 13C
General procedure for the synthesis of 6g–6l (nucleophilic
substitution)
Synthesis was performed according to Davey40 and Gorvin.40,41 (EI): m/z 373 (M+, 100%), 293 (29), 192 (38), 147 (80).
NMR (50 MHz, CDCl3): d = 149.9 (s), 144.3 (s), 133.8 (d), 133.6
(d), 128.3 (d), 123.3 (d), 119.7 (s), 118.9 (s), 106.5 (s) ppm. MS
Under an argon atmosphere, 4-haloaniline 4a,b (1 eq.), fluoro-
4-Iodo-N,N-bis(4-nitrophenyl)benzeneamine (6j). Under an
benzene 8a–c (2.2 eq.) and base (CsF or KOtBu, 2.0–2.2 eq.) were argon atmosphere, 4-iodoaniline (5.6 g, 26 mmol) and KOtBu
stirred in DMSO (stored over molecular sieve, 3 Å) at 120 1C (6.1 g, 54 mmol) were suspended in 150 mL of dry DMSO. To
overnight. The solvent was removed under reduced pressure the stirred suspension 7a (8.00 g, 57 mmol) was added drop
and the remaining crude product dissolved in chloroform. After wise and the reaction was heated to 120 1C for 40 h. After
washing with water, the organic layer was dried over anhydrous cooling to r.t., the solvent was removed under reduced pres-
sodium sulfate and evaporated to dryness under reduced sure. The residue was taken up with chloroform and filtered
pressure. Further purification was achieved by column chromato- over a pad of celite. The solvent was removed under reduced
graphy or recrystallization.
pressure and the crude product crystallized from pyridine to
4-Bromo-N,N-bis(4-nitrophenyl)benzeneamine (6g). Under give 6j (6.96 g, 58% of theory) as orange solid. TLC (silica gel,
an argon atmosphere, 4-bromoaniline (6.88 g, 40 mmol) and hexanes/ethyl acetate = 9/1): rf = 0.26. FP = 295–297 1C. 1H NMR
CsF (12.15 g, 80 mmol) were suspended in 100 mL of dry (400 MHz, CDCl3): d = 8.19–8.10 (m, 4H), 7.76–7.68 (m, 2H),
DMSO. To the stirred suspension 7a (12.42 g, 88 mmol) was 7.17–7.09 (m, 4H), 6.94–6.87 (m, 2H) ppm. 13C NMR (100 MHz,
added drop wise and the reaction was heated to 110 1C for 40 h. CDCl3): d = 151.6 (s), 144.9 (s), 143.4 (s), 139.8 (d), 128.8 (d),
After cooling to r.t., the reaction mixture was poured on 800 mL 125.9 (d), 122.9 (d), 91.4 (s) ppm. MS (EI): m/z 461 (M+, 100%),
of water. The formed brown precipitate was filtered over a glass 369 (11), 241 (83).
sinter funnel and crystallized from pyridine to give 6g (9.00 g,
4-Iodo-N,N-bis[4-(methylsulfonyl)phenyl]benzeneamine (6k).
54% of theory) as yellow crystals. TLC (silica gel, hexanes/ethyl Under an argon atmosphere, 4-iodoaniline (4.38 g, 20 mmol), 7b
1
acetate = 9/1): rf = 0.76. FP = 312–315 1C. H NMR (400 MHz, (7.6 g, 44 mmol) and KOtBu (4.94 g, 44 mmol) were stirred in
CDCl3): d = 8.19–8.10 (m, 4H), 7.57–7.49 (m, 2H), 7.17–7.09 150 mL of dry DMSO at 120 1C for 18 h. After cooling to r.t., the
(m, 4H), 7.07–7.00 (m, 2H) ppm. 13C NMR (100 MHz, CDCl3): solvent was removed under reduced pressure. The residue was
d = 151.7 (s), 144.2 (s), 143.3 (s), 133.9 (d), 128.7 (d), 125.9 (d), taken up with 100 mL of chloroform and washed with 50 mL
122.8 (d), 120.5 (s) ppm. MS (EI): m/z 413 (M+, 100%), 323 (20), of water. After drying over anhydrous sodium sulfate, the
321 (20), 241 (96).
solvent was removed under reduced pressure and the crude
4-Bromo-N,N-bis[4-(methylsulfonyl)phenyl]benzeneamine (6h). product was purified by recrystallization from acetic acid/water
Under an argon atmosphere, 4-bromoaniline (86 mg, 0.5 mmol,), (v/v = 15/1) to give 6k as a brown solid (6.74 g, 64% of theory).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
TLC (silica gel, hexanes/ethyl acetate = 1/1): rf = 0.19. FP = 232– procedure; 6e (7.03 g, 15 mmol) was lithiated with n-BuLi
235 1C. 1H NMR (200 MHz, CDCl3): d = 7.83–7.73 (m, 4H), 7.70– (7.2 mL, 18 mmol) in 75 mL of anhydrous THF for 2 h before
7.62 (m, 2H), 7.23–7.11 (m, 4H), 6.92–6.83 (m, 2H), 3.04 (s, 6H) isopropyl pinacol borate (3.35 g, 18 mmol) was added. After
ppm. 13C NMR (50 MHz, CDCl3): d = 151.0 (s), 145.2 (s), 139.5 (d), general workup the crude product was purified by recrystalliza-
134.8 (s), 129.4 (d), 128.5 (d), 123.3 (d), 90.5 (s), 44.8 (q) ppm.
tion from acetonitrile to give 8e as white crystalline powder
4,40-[(4-Iodophenyl)imino]bisbenzonitrile (6l). Under an (6.60 g, 85% of theory). TLC (silica gel, hexanes/ethyl acetate =
argon atmosphere, 4-iodoaniline (4.48 g, 20.5 mmol) and CsF 8/1): rf = 0.72. FP = 206–209 1C. 1H NMR (200 MHz, CDCl3):
(6.52 g, 43 mmol) were suspended in 40 mL of dry DMSO. d = 7.73–7.64 (m, 2H), 7.44–7.34 (m, 4H), 7.13–7.02 (m, 6H), 1.34
A solution of 7c (5.45 g, 45 mmol) in 20 mL of dry DMSO was (s, 12H), 0.26 (s, 18H) ppm. 13C NMR (50 MHz, CDCl3): d = 150.4
added drop wise and the reaction mixture was heated to 140 1C for (s), 147.9 (s), 136.1 (d), 134.7 (s), 134.5 (d), 124.0 (d), 122.8 (d),
18 h. After cooling to r.t., the solvent was removed under reduced 83.8 (s), 25.1 (q), ꢀ0.8 (q) ppm (C–B not detected). MS (EI): m/z
pressure. The residue was taken up with chloroform and filtered 515 (M+, 100%), 500 (22), 400 (11), 243 (11).
over a pad of celite. The solvent was removed under reduced
N,N-Bis[4-(1,1-dimethylethyl)phenyl]-4-(4,4,5,5-tetramethyl-
pressure and the crude product purified by column chromato- 1,3,2-dioxaborolan-2-yl)benzeneamine (8f). According to the
graphy (90 g silica gel, dry loading on 10 g silica, hexanes/ethyl general procedure; 6f (10.91 g, 25 mmol) was lithiated with
acetate gradient 5 - 20%) to give 6l (2.10 g, 24% of theory) as light n-BuLi (12 mL, 30 mmol) in 150 mL of anhydrous THF for 2 h
yellow solid. TLC (silica gel, hexanes/ethyl acetate = 4/1): rf = 0.42. before isopropyl pinacol borate (5.58 g, 30 mmol) was added.
FP = 228–230 1C. 1H NMR (200 MHz, CDCl3): d = 7.73–7.61 (m, 2H), After general workup the crude product was triturated with
7.59–7.45 (m, 4H), 7.15–7.02 (m, 4H), 6.92–6.81 (m, 2H) ppm. 13
C
methanol to give 8f as white crystalline powder (9.05 g, 79% of
NMR (50 MHz, CDCl3): d = 149.9 (s), 145.0 (s), 139.5 (d), 133.8 (d), theory). TLC (silica gel, hexanes/ethyl acetate = 8/1): rf = 0.71.
128.5 (d), 123.4 (d), 118.9 (s), 106.5 (s), 90.5 (s) ppm. MS (EI): m/z FP = 212–214 1C. 1H NMR (200 MHz, CDCl3): d = 7.64–7.53
421 (M+, 100%), 294 (12), 192 (12).
(m, 2H), 7.37–7.26 (m, 4H), 7.10–7.00 (m, 4H), 7.00–6.91
(m, 2H), 1.37–1.29 (m, 30H) ppm. 13C NMR (50 MHz, CDCl3):
d = 151.5 (s), 147.2 (s), 145.2 (s), 136.2 (d), 126.8 (d), 125.4 (d),
General procedure for the synthesis of 8a–f
Synthesis was performed according to Anemian:42 all operations 120.9 (d), 84.0 (s), 34.8 (s), 31.8 (q), 25.3 (q) ppm (C–B not
were performed under argon atmosphere. Triphenylamine (6a–f) detected). MS (EI): m/z 483 (M+, 16%), 468 (18), 368 (12), 168
(1.0 eq.) was dissolved in anhydrous THF. The solution was (11), 57 (100).
cooled to ꢀ80 1C and a solution of n-butyllithium (2.5 M in
N,N-Bis(4-nitrophenyl)-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-
hexanes) (1.2 eq.) was added via a syringe, keeping the tempera- 2-yl)benzeneamine (8g). Synthesis was performed according to
ture below ꢀ75 1C. The reaction mixture was stirred at ꢀ80 1C Murata.43 In a 100 mL round-bottom flask, 6j (2.08 g, 4.5 mmol,
for appropriate time before isopropyl pinacol borate (1.2 eq.) was 1.0 eq.) and PdCl2(dppf) (98.8 mg, 0.135 mmol, 0.03 eq.) were
added drop-wise at the same temperature. After slowly warming suspended in 20 mL of dry dioxane. After flushing the apparatus
to room temperature (approx. 2 h), stirring was continued over- with argon, triethylamine (1.37 g, 13.5 mmol, 3.0 eq.) and pina-
night. The solvent was distilled off under reduced pressure and colborane (749 mg, 5.85 mmol, 1.3 eq.) were added drop wise. After
the residue was distributed between water and chloroform. The stirring for 15 h at room temperature the solvent was removed
phases were separated and the aqueous phase was extracted twice under reduced pressure. The crude product was purified by column
with chloroform. The combined organic layer was washed with chromatography (100 g basic alumina, dry loading on 6 g neutral
brine and dried over sodium sulfate. The solvent was removed alumina, hexanes/DCM gradient 25 - 100%) to give 8g as orange
under reduced pressure to give the crude product, which was powder (390 mg, 19% of theory). TLC (silica gel, hexanes/ethyl
further purified by recrystallization or Kugelrohr distillation.
acetate = 8/1): rf = 0.52. FP = 175–178 1C. 1H NMR (200 MHz, CDCl3):
N,N-Bis(4-fluorophenyl)-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan- d = 8.18–8.06 (m, 4H), 7.89–7.77 (m, 2H), 7.20–7.07 (m, 6H), 1.33
2-yl)benzeneamine (8c). According to the general procedure; 6c (s, 12H) ppm. 13C NMR (50 MHz, CDCl3): d = 151.9 (s), 147.6 (s),
(4.8 g, 13.3 mmol) was lithiated with n-BuLi (6.4 mL, 16 mmol) 143.2 (s), 137.1 (d), 126.0 (d), 125.7 (d), 123.0 (d), 84.3 (s),
in 50 mL anhydrous THF for 2 h before isopropyl pinacol borate 25.1 (q) ppm (C–B not detected). MS (EI): m/z 461 (M+, 100%),
(3.0 g, 16 mmol) was added. After general workup the crude 361 (21), 268 (17).
product was purified by Kugelrohr distillation (130 1C, 8.0 ꢁ
10ꢀ2 mbar) to give 8c as white crystalline powder (4.9 g, 91%
General procedure for the Suzuki coupling towards 2, S1–3
of theory). TLC (silica gel, hexanes/ethyl acetate = 8/1): rf = 0.67. The synthesis was performed according to Marion.44 Under an
FP = 92–95 1C. 1H NMR (200 MHz, CD2Cl2): d = 7.65–7.53 argon atmosphere, (hetero)aromatic halide (1.0 eq.), boronic
(m, 2H), 7.16–6.84 (m, 10H), 1.31 (s, 12H) ppm. 13C NMR ester (3.0–4.5 eq.) and KOtBu (3.0–4.5 eq.) were suspended in a
(50 MHz, CD2Cl2): d = 159.8 (s, JCF = 242.8 Hz), 151.2 (s), sufficient amount of solvent (IPA/H2O = 3/1; degassed by bub-
143.9 (s, JCF = 3.2 Hz), 136.3 (d), 127.5 (d, JCF = 8.2 Hz), 120.7 bling with argon). A solution of (IPr)Pd(allyl)Cl (0.02–0.05 eq.) in
(d), 116.7 (d, JCF = 22.6 Hz), 84.1 (s), 25.2 (q) ppm (C–B not degassed IPA was added and the reaction mixture was refluxed
detected). MS (EI): m/z 407 (M+, 100%), 349 (10), 307 (18).
for an appropriate time, monitoring the conversion by TLC. After
N,N-Bis[4-(trimethylsilyl)phenyl]-4-(4,4,5,5-tetramethyl-1,3,2- completion, the reaction mixture was distributed between water
dioxaborolan-2-yl)benzeneamine (8e). According to the general and chloroform; the phases were separated and the aqueous
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
layer was extracted with chloroform three times. The combined purified by column chromatography (90 g silica gel, hexanes/
organic layer was dried over anhydrous sodium sulfate and the DCM 25 - 30%) to give S3 as yellow powder (323.3 mg, 56% of
solvent removed under reduced pressure to give the crude theory). TLC (silica gel, hexanes/DCM = 5/1): rf = 0.32. 1H NMR
product. Purification was achieved by column chromatography. (200 MHz, CD2Cl2): d = 7.72 (s, 3H), 7.56–7.45 (m, 6H), 7.39
2,20,200-(1,3,5-Benzenetriyl)tristhiophene (2). According to the (d, J = 3.76 Hz, 3H), 7.24 (d, J = 3.78 Hz, 3H), 7.14–6.92 (m, 30H)
general procedure; 1,3,5-tribromobenzene (0.787 g, 2.5 mmol), ppm. 13C NMR (50 MHz, CD2Cl2): d = 159.7 (s, JCF = 242.5 Hz),
2-thiopheneboronic acid pinacol ester (2.364 g, 11.25 mmol) and 148.2 (s), 144.6 (s), 144.1 (s, JCF = 3.0 Hz), 142.1 (s), 136.1 (s),
KOtBu (1.262 g, 11.25 mmol) were suspended in 50 mL solvent. 128.2 (s), 126.98 (d), 126.97 (d, JCF = 8.1 Hz), 125.4 (d), 123.7 (d),
(IPr)Pd(allyl)Cl (71.4 mg, 125 mmol; dissolved in 1 mL IPA) was 122.8 (d), 121.8 (d), 116.7 (d, JCF = 22.7 Hz) ppm. MS (MALDI-TOF):
added before refluxing for 1 h. After general workup the crude calcd for C72H45F6N3S3: 1161.2680; found: 1161.3152.
product was purified by column chromatography (90 g silica gel,
hexanes/DCM 0 - 3%) to give 2 as white powder (0.697 g, 86%
of theory). TLC (silica gel, hexanes/ethyl acetate = 20/1): rf = 0.35.
Results and discussion
Synthesis
1H NMR (200 MHz, CDCl3): d = 7.74 (s, 3H), 7.40 (dd, J = 1.1, 3.6
Hz, 3H), 7.33 (dd, J = 1.1, 5.1 Hz, 3H), 7.11 (dd, J = 3.6, 5.1 Hz, 3H)
ppm. 13C-NMR (50 MHz, CDCl3): d = 143.7 (s), 135.9 (s), 128.3 (d), The synthetic linkage of TPA structures 8 bearing either electron
125.6 (d), 124.1 (d), 122.9 (d) ppm.
withdrawing (ꢀI, ꢀM-effect) or donating (+I, +M) substituents
4,40,400-(1,3,5-Benzenetriyltri-5,2-thiophenediyl)tris[N,N-diphenyl- with 3 is realized via Suzuki cross-coupling reactions.45,46 There-
benzenamine] (S1). According to the general Suzuki procedure; fore, a general cross-coupling procedure was developed tolerat-
compound 3 (280.1 mg, 0.5 mmol), boronic acid pinacol ester 6a ing a broad spectrum of functionalized TPA boronic acid esters
(835.0 mg, 2.25 mmol) and KOtBu (252.5 mg, 2.25 mmol) were (Scheme 1).
suspended in 12 mL solvent. (IPr)Pd(allyl)Cl (14.3 mg, 25 mmol;
Stille cross-coupling reactions are commonly applied in
dissolved in 1 mL IPA) was added before refluxing for 2 h. After literature,47 however they require toxic stannyl intermediates.
general workup the crude product was purified by column In order to avoid these undesirable precursors, highly stable
chromatography (90 g silica gel, hexanes/DCM = 6/1); after pinacol boronic acid esters are used in Suzuki cross-coupling
removing the solvent under reduced pressure, the residue was reactions ensuring high conversion, selectivity, and tolerance of
dissolved in 40 mL of boiling 2-butanone. The product was functional groups representing a more convenient synthetic
precipitated by adding 50 mL of 2-propanol to give S1 as yellow approach.48 Herein we report the synthetic pathway towards
powder (512.8 mg, 97% of theory). TLC (silica gel, cyclohexane/ diversely substituted triphenylamine structures (R = –H, –Me,
1
DCM = 3/1): rf = 0.30. H NMR and 13C NMR data according to –OMe, –TMS, –tBu, –F, –NO2, –CN, –SO2Me), functionalization
literature.21 MS (MALDI-TOF): calcd for C72H51N3S3: 1053.3245; towards boronic acid esters and subsequent cross-coupling
found: 1053.3020.
with brominated tris(2-thienyl)benzene 3 via Suzuki reaction.
4,40,400-(1,3,5-Benzenetriyltri-5,2-thiophenediyl)tris[N,N-bis-
In order to synthesize the desired TPA building blocks two
(4-methylphenyl)benzenamine] (S2). According to the general synthetic pathways towards halide species 6a–l were developed
Suzuki procedure; compound 3 (280.1 mg, 0.5 mmol), boronic depending on the nature of substituent R (Scheme 2).
acid pinacol ester 6b (898.5 mg, 2.25 mmol) and KOtBu (252.5 mg,
The conversion of suitably substituted iodobenzenes 5a–f
2.25 mmol) were suspended in 12 mL solvent. (IPr)Pd(allyl)Cl and 4-bromoaniline 4a to the corresponding bromo substituted
(14.3 mg, 25 mmol; dissolved in 1 mL IPA) was added before triphenylamines 6a–f was realized in good to excellent yields
refluxing for 2 h. After general workup the crude product was via Ullmann condensation applying copper(I)chloride and 1,10-
purified by column chromatography (90 g silica gel, hexanes/DCM phenanthroline as catalytic system.26 When applying the same
25 - 27%) to give S2 as yellow powder (473.3 mg, 83% of theory). strategy on substrates bearing ꢀM substituents relatively low
TLC (silica gel, cyclohexane/DCM = 3/1): rf = 0.30. 1H NMR conversions have been observed. Consequently, nucleophilic
(200 MHz, CD2Cl2): d = 7.74 (s, 3H), 7.53–7.44 (m, 6H), 7.40 aromatic substitution of properly substituted fluorobenzenes
(d, J = 3.78 Hz, 3H), 7.23 (d, J = 3.80 Hz, 3H), 7.15–7.05 (m, 12H), 7a–c with anilines 4a,b was chosen as an alternative synthetic
7.05–6.94 (m, 18H), 2.32 (s, 18H) ppm. 13C NMR (50 MHz, approach leading to bromo (6g–i) and iodo (6j–l) substituted
CD2Cl2): d = 148.5 (s), 145.5 (s), 144.9 (s), 141.8 (s), 136.2 (s), triphenylamine derivatives in good yields (Table 1).
133.7 (s), 130.50 (d), 127.5 (s), 126.8 (d), 125.45 (d), 125.39 (d),
123.4 (d), 122.6 (d) 121.7 (d), 20.1 (s) ppm. MS (MALDI-TOF): achieved via metal halogen exchange using n-butyllithium and
calcd for C78H63N3S3: 1137.4184; found: 1137.4006. further conversion with isopropyl pinacol borate in good yields
The synthesis of the boronic acid pinacolates 8a–f was
4,40,400-(1,3,5-Benzenetriyltri-5,2-thiophenediyl)tris[N,N-bis- (Table 1). Again, substrates bearing ꢀM substituent turned out
(4-fluorophenyl)benzenamine] (S3). According to the general to be troublesome, since neither bromo 6g–i nor iodo derivatives
procedure; compound 3 (280.6 mg, 0.5 mmol), boronic acid 6k,l showed proper conversion. Nevertheless, starting from iodo
pinacol ester 6c (916.3 mg, 2.25 mmol) and KOtBu (252.5 mg, derivative 6j the synthesis of boronic acid esters 8g could be
2.25 mmol) were suspended in 12 mL solvent. (IPr)Pd(allyl)Cl achieved (despite the rather low yields of 18%) when switching
(14.3 mg, 25 mmol; dissolved in 1 mL IPA) was added before to the palladium catalyzed Miyaura borylation using pinacol
refluxing for 2 h. After general workup the crude product was borane as boron source (Scheme 2).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
Scheme 1 Synthetic pathway towards S1–3. Reaction conditions: (i) thiophene-2-boronic acid pinacol ester, KOtBu, (IPr)Pd(allyl)Cl, isopropanol/water,
reflux; (ii) NBS, chloroform/glacial acetic acid, r.t.; (iii) 8a–c, KOtBu, (IPr)Pd(allyl)Cl, isopropanol/water, reflux.
Scheme 2 Synthesis of intermediates 6 and boronic acid esters 8. Reaction conditions: (i) KOH, CuCl, phenanthroline in toluene, reflux; (ii) n-BuLi,
ꢀ78 1C in THF, then isopropyl pinacol borate; (iii) CsF or KOtBu in DMSO, DT; (iv) pinacolborane, triethylamine, PdCl2(dppf) in dioxane, r.t. For a definition
of R and X please refer to Table 1.
For the consecutive Suzuki cross-coupling reaction, palla-
dium NHC-complex (IPr)Pd(allyl)Cl showed superior efficiency
over tetrakis(triphenyl)palladium(0) in terms of reactivity
Table 1 Experimental data for compounds 6a–l and boronic acid esters
8a–f
Ullmann
condensation
Nucleophilic aromatic
substitution
Synthesis of boronic
esters
and selectivity. Additionally, high conversion and high toler-
ance of TPA boronic acid esters 8a–f bearing either electron
donating (e.g. –Me) or electron withdrawing substituents
(e.g. –F) were observed using isopropanol–water as solvent
and KtOBu as base.
Applying these reaction conditions, compounds 2 and S1–3
were obtained. The synthesis of 2 was achieved in good yield of
86% via cross-coupling of thiophene-2-boronic acid pinacol
R
Yield (%)
X
R
Yield (%)
R
Yield (%)
6a
H
74
6g Br NO2
54
8a
8b Me
8c
H
77
74
91
6b Me 89
6c 70
6h Br SO2Me 77
F
6i Br CN
23
58
F
6d OMe 86
6e TMS 60
6f tBu 61
6j
I
I
I
NO2
SO2Me 64
CN 24
8d OMe 76
8e TMS 85
6k
6l
8f tBu
79
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
Table 2 Photophysical characterization of compounds S1–3 and computational data for compounds Sm1–3
a
b
d
e
f
g
h
Comp.
lmax (emax
)
(nm)
lmax (nm)
lem (ff)c (nm)
lem (nm)
lem (nm)
Comp.
lmax (nm)
lem (nm)
lem (nm)
S1
S2
S3
302 (61.2), 387 (112.9)
304 (61.7), 394 (104.2)
297 (51.0), 384 (104.7)
474
482
472
461 (62)
474 (71)
459 (62)
471
477
466
486
498
482
Sm1
Sm2
Sm3
344
349
343
450
454
449
445
459
445
a
b
c
Measured in THF solution, emax ꢁ 10ꢀ3 in Mꢀ1 cmꢀ1 in parentheses. Absorption of solid samples. Measured in a THF solution;
d
e
f
g
ff fluorescence quantum yield. Film samples. Solid samples. Calculated absorption of model compounds Sm1–3. Calculated emission
(LR-PCM) of model compounds Sm1–3. Calculated emission (SS-PCM) of model compounds Sm1–3.
h
ester and 1,3,5-tribromobenzene 1 according to the above n–p* transition of the triphenylamine moiety, the longer wave-
protocol. Consecutive bromination with NBS in chloroform/glacial length absorption originates from the p–p* transitions of the
acetic acid gave the tribromo derivative 3, the precursor for the electron-donating triphenylamine moiety to the electron-
final Suzuki coupling with boronic acid esters 8a–c. Star-shaped accepting thiophene moiety.49 A clear trend can be observed,
compounds S1–3 bearing either electron donating or withdrawing as the electron withdrawing fluoro derivative S3 shows the
substituents were obtained in satisfactory to excellent yields absorption maximum with the highest energy, whereas the
(Table 3). These solid materials appear yellow in color and are electron donating methyl derivative S2 has the absorption
soluble in common organic solvents. The characterization of maximum at the longest wavelength; the absorption maximum
1
S1–3 was performed by H/13C-NMR spectroscopy and HR-MS of the electron neutral S1 lies in between.
analysis. The data are consistent with the proposed structural
formulations.
These findings are supported by quantum chemical calcula-
tions of model substances Sm1–3 (Fig. 2). In order to mimic C3
Photo-physical properties of S1–3 were determined by symmetric substances S1–3, suitable model substances were
UV-VIS and fluorescence spectrometry in THF solution, for chosen, comprising one dendritic arm and the benzene core.
amorphous films and in the solid state (Table 2 and Fig. 3). This approach ensures a realistic representation of S1–3 while
Two absorption maxima were located in relatively narrow keeping the computational costs at a reasonable level. Absorp-
ranges of 297–304 and 384–394 nm for substances S1–3 when tion and emission wavelengths in THF of Sm1–3 were estimated
measured in THF solution. While the former is assigned to the by applying density functional (DFT) and time dependent
density functional (TDDFT) level of theory according to a
protocol presented recently.36 The calculated absorption and
emission wavelengths of Sm1–3 are in agreement with experi-
mental data of S1–3 showing the lowest shift in emission for
fluoro-substituted S3/Sm3 and highest emission shifts for
methyl-substituted compounds S2/Sm2 (Table 2).
Again this behavior can be correlated to the increased
electron density on the aromatic ring (Fig. 3) in the case of S2
(lcm = 474 nm) causing a significant bathochromic shift and a
decreased electron density in the case of S3 (lcm = 459 nm)
causing a hypsochromic shift in reference to S1 (lcm = 461 nm)
(Table 2). Stokes shifts for all compounds S1–3 are comparable
Fig. 2 Model substances Sm1–3.
and range from 0.51–0.53 eV. Quantum yields in solution are
Fig. 3 Absorption (left) and emission (right) spectra of S1–3 in THF solution.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
the quantum yield is considerably higher (ff = 0.71) for the methyl
substituted derivative S2.
Emission spectra of solid samples S1–3 are in the narrow
range of 482–498 nm, whereas spin coated samples are in the
range of 466–477 nm (for spectra please refer to the ESI†).
Electro-chemical characteristics of the star-shaped mole-
cules S1–3 were investigated by cyclic voltammetric methods
(Fig. 4). The first oxidation potentials were used to determine
the HOMO energy levels. Ferrocene served as an external
standard for calibrating the potential and calculating the
HOMO levels (ꢀ4.8 eV). All compounds S1–3 undergo reversible
oxidation indicating the formation of stable cation radicals.
Only small variations of HOMO levels (ꢀ5.19 eV to ꢀ5.24 eV)
were observed. The LUMO levels were determined from the
optical bandgap and the HOMO energy level obtained from CV
measurements and vary in the range from ꢀ1.98 to ꢀ2.10 eV.
Calculated HOMO/LUMO levels are in good accordance with
the experimental data for compounds S1 and S3, whereas the
deviation is somewhat higher for substance S2. The pertinent
data are listed in Table 3.
Fig. 4 Cyclic voltammograms of S1–3 in DCM.
shifted to higher values correlating with enhanced electron density
(Table 2). The values of S1 and S3 are identical (ff = 0.62), whereas
Thermal properties were investigated by DSC and TGA
measurements. Glass-forming properties and thermal stability
of the compounds were examined by simultaneous thermal
analysis (STA). The glass transition temperature (Tg) was esti-
mated from DSC curves (Fig. 5). The determined Tg values are
in the range from 121 to 135 1C, the methyl substituted S2
showing the highest value while fluoro substituted S3 exhibits
the lowest Tg. All materials exhibit high thermal stability as
Table 3 Experimental data, physical characterization and computational
data of compounds S1–3
a
b
Yield Tg
Comp. (%)
Td
Band gapc HOMO/
HOMO/
(1C) (1C) (nm, eV)
LUMOd (eV) LUMOe (eV)
S1
S2
S3
97
83
50
128 541 387, 3.20
135 528 394, 3.14
121 538 384, 3.23
ꢀ5.19/ꢀ1.99 ꢀ5.17/ꢀ1.76
ꢀ5.24/ꢀ2.10 ꢀ5.01/ꢀ1.66
ꢀ5.21/ꢀ1.98 ꢀ5.36/ꢀ1.91
a
b
Obtained from DSC measurements. Thermal decomposition tem- evidenced by decomposition temperatures corresponding to
perature determined from 5% mass loss. Bandgaps were determined
5% mass loss between 528 and 541 1C (Table 3 and Fig. 5).
d
This high thermal stability allows for fabrication of homo-
relative to ferrocene (Fc/Fc+, Eox = 446 mV). The concentration of the geneous and stable amorphous thin films by vacuum deposition.
c
from the onsets of the absorption in THF solution. HOMO-levels
determined from solutions measured in DCM. All Eox data are reported
compounds used in this experiment was 1 mg mLꢀ1 and the scan rate was
The semiconducting properties of the materials under inves-
50 mV sꢀ1. LUMO levels were determined from the optical band gap and
tigation were tested by using a top-contact OFET architecture as
the HOMO energy level according to the following equation: ELUMO
=
e
EHOMO + Ebandgap
.
Calculated HOMO/LUMO levels (B3LYP/6-311+G(d)). shown in Fig. 6. A hole mobility up to 7.7 ꢁ 10ꢀ4 cm2 Vꢀ1 sꢀ1
Fig. 5 TGA (left) and DSC (right) measurements of S1–3.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
Fig. 6 Output characteristics (left) and corresponding transfer curve (right) of S2.
and average threshold voltages of approx. 0 V which can be thickness (30–50 nm) and low roughness of the surface (root
attributed to a low background charge density in the material mean square = 0.63, 0.32 and 0.34 nm for S1, S2 and S3
were recorded using S2 as semiconducting layer (Fig. 6), while respectively). However, when examining the surface of S1
no working OFET device could be fabricated using S1 or S3. sub-mm sized pinholes were observed. Potential reasons for
As the performance of an OFET highly depends on the quality pinholes are trapped air, solvent gas-bubble formation or
of the interface between dielectric and organic semiconductor, contaminations on the substrate or in the organic solution.
these results might be correlated to the tendency of S2 to from Since numerous spin coating samples were prepared applying
nearly defect free films, which was observed in spin coating the same procedure yielding reproducible results, the origin of
experiments.
these defects is unlikely to come from the fabrication process.
All compounds S1–3 readily form stable glasses. Spin Therefore, the formation of these pinholes is probably sub-
coated thin films of S1–3 were produced on ITO substrates stance specific and might be attributed to higher propensity to
using chloroform as solvent. Their surface morphology and crystallization of S1 compared to S2 and S3. Thus, in contrast
thickness were further investigated by atomic force microscopy to S1, solution-processed films of S2 and S3 can readily be
(AFM, Fig. 7). In principle, all films showed very little average prepared.
Fig. 7 Topography AFM images of S1–3 (a–c).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
NJC
Paper
14 J. Zhang, D. Deng, C. He, Y. He, M. Zhang, Z.-G. Zhang,
Z. Zhang and Y. Li, Chem. Mater., 2011, 23, 817.
15 Z. Ning and H. Tian, Chem. Commun., 2009, 5483.
16 A. Iwan, D. Sek, D. Pociecha, A. Sikora, M. Palewicz and
H. Janeczek, J. Mol. Struct., 2010, 981, 120.
17 A. Iwan and D. Sek, Prog. Polym. Sci., 2011, 36, 1277.
18 T. Noda, H. Ogawa, N. Noma and Y. Shirota, Adv. Mater.,
1997, 9, 720.
19 J. Roncali, P. Leriche and A. Cravino, Adv. Mater., 2007,
19, 2045.
20 C. He, Q. He, Y. Yi, G. Wu, F. Bai, Z. Shuai and Y. Li, J. Mater.
Chem., 2008, 18, 4085.
Conclusion
This contribution outlines an optimized synthetic pathway towards
a broad spectrum of p-substituted triphenylamines, suitable as
building blocks for organic semiconducting materials.
We employed these triphenylamine moieties in C3 symmetric
dendritic tris(2-thienyl)benzene structures and elucidated the
influence of the molecular design on photo-physical properties.
The electronic characteristics of the substituents show a strong
influence on the fluorescence spectra thus enabling color tuning.
These star shaped materials exhibit good solubility and thermal
stability allowing for both solution processing and physical vapor
deposition, respectively. The novel materials were tested as semi-
conductors in organic field effect transistor devices, exhibiting
´ ˆ
`
21 A. Cravino, S. Roquet, P. Leriche, O. Aleveque, P. Frere and
J. Roncali, Chem. Commun., 2006, 1416.
22 T.-T. Bui, L. Beouch, X. Sallenave and F. Goubard, Tetrahedron
Lett., 2013, 54, 4277.
23 N. Metri, X. Sallenave, L. Beouch, C. Plesse, F. Goubard and
C. Chevrot, Tetrahedron Lett., 2010, 51, 6673.
24 K. R. Justin Thomas, J. T. Lin, Y.-T. Tao and C.-W. Ko, Chem.
Mater., 2002, 14, 1354.
25 A. Iwan, H. Janeczek, B. Kaczmarczyk, B. Jarzabek,
M. Sobota and P. Rannou, Spectrochim. Acta, Part A, 2010,
75, 891.
hole mobilities of almost 10ꢀ3 cm2 Vꢀ1 sꢀ1
.
Spin coating experiments showed a strong correlation between
the nature of the substituents and the morphology of the resulting
thin film; similar results were observed for transistor perfor-
mance. The structural versatility and capability of our molecular
design will contribute to the further development of functional
organic materials.
Acknowledgements
26 H. B. Goodbrand and N.-X. Hu, J. Org. Chem., 1999, 64,
670.
J. Bintinger gratefully acknowledges funding from the Marshall
¨
Plan Foundation. The authors thank Dr K. Fottinger for supporting
¨
27 M. W. Andersen, B. Hildebrandt, G. Koster and R. W.
the photophysical characterization, D. Bomze, A. Aster, M. Lunzer
and K. Kandra for contributing to the synthetic experiments.
M. Eberl is acknowledged for assisting in cyclic voltammetric
measurements and K. Fohringer and P. Kautny for STA
measurements.
Hoffmann, Chem. Ber., 1989, 122, 1777.
28 O. Navarro and S. P. Nolan, Synthesis, 2006, 366.
29 J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima and Y. Honda, Gaussian 09,
Revision A.02, Gaussian, Inc., Wallingford, CT, 2009.
30 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785.
References
1 S. R. Forrest, Nature, 2004, 428, 911.
2 J. Mei, Y. Diao, A. L. Appleton, L. Fang and Z. Bao, J. Am.
Chem. Soc., 2013, 135, 6724.
31 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
32 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem.
Phys., 1980, 72, 650.
3 Y.-S. Tyan, J. Photonics Energy, 2011, 1, 011009.
4 M. C. Scharber and N. S. Sariciftci, Prog. Polym. Sci., 2013,
38, 1929.
33 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2007, 120, 215.
34 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 119,
525.
´
5 B. Kippelen and J.-L. Bredas, Energy Environ. Sci., 2009,
2, 251.
6 H. Klauk, Organic electronics II: more materials and applications,
Wiley-VCH, Weinheim, 2012.
¨
35 A. Schafer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992,
97, 2571.
7 Y. Yamashita, Sci. Technol. Adv. Mater., 2009, 10, 024313.
8 Y. Shirota and H. Kageyama, Chem. Rev., 2007, 107, 953.
9 J. Kwon, M. K. Kim, J.-P. Hong, W. Lee, S. Lee and J.-I. Hong,
Bull. Korean Chem. Soc., 2013, 34, 1355.
¨
36 D. Lumpi, E. Horkel, F. Plasser, H. Lischka and J. Frohlich,
ChemPhysChem, 2013, 14, 1016.
37 M. Cossi, V. Barone, R. Cammi and J. Tomasi, Chem. Phys.
Lett., 1996, 255, 327.
¨
10 F. Geiger, M. Stoldt, H. Schweizer, P. Bauerle and E. Umbach,
38 M. Cossi and V. Barone, J. Chem. Phys., 2001, 115, 4708.
39 R. Improta, G. Scalmani, M. J. Frisch and V. Barone, J. Chem.
Phys., 2007, 127, 074504.
Adv. Mater., 1993, 5, 922.
11 S. H. Kim, K. Hong, K. H. Lee and C. D. Frisbie, ACS Appl.
Mater. Interfaces, 2013, 5, 6580.
40 M. H. Davey, V. Y. Lee, L.-M. Wu, C. R. Moylan, W. Volksen,
A. Knoesen, R. D. Miller and T. J. Marks, Chem. Mater., 2000,
12, 1679.
12 Y. Shirota, J. Mater. Chem., 2000, 10, 1.
13 N. Metri, X. Sallenave, C. Plesse, L. Beouch, P.-H. Aubert,
F. Goubard, C. Chevrot and G. Sini, J. Phys. Chem. C, 2012,
116, 3765.
41 J. H. Gorvin, J. Chem. Soc., Perkin Trans. 1, 1988, 1331.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem.

View Article Online
Paper
NJC
42 R. Anemian, D. C. Cupertino, P. R. Mackie and S. G. Yeates, 46 J. K. Choi, K. Cho and T.-H. Yoon, Synth. Met., 2010,
Tetrahedron Lett., 2005, 46, 6717. 160, 1938.
43 M. Murata, T. Oyama, S. Watanabe and Y. Masuda, J. Org. 47 R. A. Rossi, J. Organomet. Chem., 2014, 751, 201.
Chem., 2000, 65, 164.
44 N. Marion, O. Navarro, J. Mei, E. D. Stevens, N. M. Scott and
S. P. Nolan, J. Am. Chem. Soc., 2006, 128, 4101.
48 R. Martin and S. L. Buchwald, Acc. Chem. Res., 2008,
41, 1461.
49 Z. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike,
H. Miyamoto, T. Kajita and M. Kakimoto, Adv. Funct. Mater.,
2008, 18, 584.
`
45 A. Leliege, J. Grolleau, M. Allain, P. Blanchard, D. Demeter,
T. Rousseau and J. Roncali, Chem. – Eur. J., 2013, 19, 9948.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015