Organic soluble indigoids derived from 3-hydroxybenzaldehyde for N-type organic field-effect transistor (OFET) applications

Article abstract of DOI:10.1016/j.orgel.2016.02.035

Two new series of organic soluble indigoids 7-7′-dialkoxyindigoids (2a, 2b) and 4,4′-dibromo-7,7′-dialkoxyindigoids (3a, 3b) (alkoxy = n-butoxy and n-octyloxy) were synthesized starting from the inexpensive 3-hydroxybenzaldehyde. The indigoids were solubl

Full text of DOI:10.1016/j.orgel.2016.02.035

Organic Electronics 32 (2016) 258e266
Contents lists available at ScienceDirect
Organic Electronics
journal homepage: www.elsevier.com/locate/orgel
Organic soluble indigoids derived from 3-hydroxybenzaldehyde for N-
type organic field-effect transistor (OFET) applications
,
Jenner H.L. Ngai a, Louis M. Leung a *, S.K. So b, Harrison K.H. Lee b
a Department of Chemistry, Hong Kong Baptist University, Kowloon, Hong Kong SAR, China
b Department of Physics, Hong Kong Baptist University, Kowloon, Hong Kong SAR, China
a r t i c l e i n f o
a b s t r a c t
Two new series of organic soluble indigoids 7-70-dialkoxyindigoids (2a, 2b) and 4,40-dibromo-7,70-dia-
lkoxyindigoids (3a, 3b) (alkoxy ¼ n-butoxy and n-octyloxy) were synthesized starting from the inex-
pensive 3-hydroxybenzaldehyde. The indigoids were soluble in common organic solvents including
chloroform, dichloromethane, toluene, ethyl acetate and ethers. The enhanced solubility was suggested
to be a lack of intermolecular hydrogen-bonds as confirmed by single crystal X-ray diffraction analyses. It
was found that intramolecular hydrogen-bonds in indigoids are crucial to the exhibition of field-effect in
OFETs, while intermolecular hydrogen-bonds only caused insolubility of the indigoids. Compared to the
pristine insoluble indigo (LUMO ¼ ꢀ3.55 eV and Eg ¼ 1.91 eV), the soluble indigoids containing electron
donating alkoxy side chains at the indigoid 7 and 70 positions were shown to have their LUMO decreased
by ꢀ0.13 to ꢀ0.26 eV. Among the indigoid studied, the soluble indigoid 3a containing electron donating
alkoxy side chains at the indigoid 7 and 70 positions and bromine groups at the indigoid 4 and 40 po-
sitions exihibited a narrowest bandgap energy with Eg ¼ 1.66 eV. Employing the same fabrication
technique and a bottom-gate-top-contact OFET configuration, the soluble indigoids were found to have
electron mobility similar to and within an order of magnitude of the pristine indigo.
Article history:
Received 2 December 2015
Received in revised form
23 February 2016
Accepted 23 February 2016
Available online xxx
Keywords:
OFET
Baeyer Drewson indigo synthesis
3-Hydroxybenzaldehyde
Organic semiconductors
Organic soluble indigoids
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
thus making it an excellent electron acceptor for n-type organic
electronic devices. The compound itself, however, is not soluble in
Indigo is a natural occurring blue pigment that has found ap-
plications since ancient times including in the wrappings for
Egyptian mummies [1].In order to prepare the dye , the leaves and
stalks of the harvested plants were submerged in water for
fermentation. The bio-matters eventually produced glucose con-
taining indoxyl. Followed by hydrolysis and oxidation, the indoxyl
compounds finally converted to indigo. Indigo is a durable, air
stable, non-toxic and biodegradable blue dye.
organic solvents and thus has limited processability. The poor sol-
ubility was attributed to the extensive intermolecular hydrogen-
bonds between the indigo molecules. Several attempts to
improve its solubility [4e6], however, were failed in creating an
organic soluble indigo derivative while preserving the conductive
indigoid (or indigotin) chromophore. Reactions on indigo usually
took place at the most reactive analogous amide 1 and 10 positions
(see Fig. 1), and thus destroying the inter- and intramolecular
hydrogen-bonds within the indigo molecules [7e9]. Direct halo-
genation of indigo has also been reported under harsh conditions
[10,11], but the products remained insoluble in organic solvents.
Thus mading it impossible to further functionalize the indigo
moiety. Sulfonation was by far the most facile reaction in producing
a water soluble indigo derivative [12]. The sulfonated indigo known
as “indigo carmine” has been applied as electrode materials for
rechargeable batteries. Indigo carmine, however, is not soluble in
organic solvents and the moisture sensitive hydrophilic protic
sulfonic acid groups are detrimental for charge transport in n-type
organic semiconductor applications.
Recently, it was recognized that indigo was an effective ambi-
polar and biodegradable semiconductor with hole and electron
mobility reported to be 0.02 cm2 Vꢀ1 ꢀ1. It has been successfully
s
applied as an active layer for n-type organic field-effect transistors
(OFET) [2,3]. The field-effect observed in its OFET is a result of high
electron or hole mobility modulated by an external electric field.
Indigo is an electron deficient compound that has low-lying LUMO
* Corresponding author.
E-mail addresses: jennerngai@gmail.com (J.H.L. Ngai), s20974@hkbu.edu.hk
(L.M. Leung).
http://dx.doi.org/10.1016/j.orgel.2016.02.035
1566-1199/© 2016 Elsevier B.V. All rights reserved.

J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
259
of the syntheses was suspected due to a mixed isomer were given
out in the non-regioselective nitration reaction. Separation of the
nitrated isomers before the final step, however, did not improve the
yield significantly. Alternative cause would be the poor solubility of
the compounds bearing the alkoxy side chains in the polar prop-
anone/NaOH environment thus leading to a low yield. Aldol
condensation of propanone itself could also lead to a formation of
side products rather than the desired indigoids as well.
Fig. 1. Compound 1 (indigo) with intramolecular hydrogen-bonds and the indigoid
chromophore under IUPAC fused ring numbering system.
2.2. Electrochemical properties
Moreover, p-type organic semiconductors such as poly-
thiophenes, tetracene, pentacene, rubrene and CuPc are readily
available while there are only few n-type organic semiconductors
including PCBM, ICBA, ICMA have been reported. In this report, the
synthesis of two series of organic soluble indigoids (see Fig. 2)
including 7,70-dialkoxyindigoids (2a, 2b) and 4,40-dibromo-7,70-
dialkoxyindigoids (3a, 3b) is being reported. The electron mobility
of the novel electron accepting n-type materials was measured by
the fabrication of n-type OFET.
Cyclic voltammetry (CV) measurement on compound 1 (indigo),
2a, 2b, 3a and 3b in DCM were effected using 0.1 M tetrabuty-
lammonium hexafluorophosphate (Bu4NPF6) as the electrolyte. The
sensitivity for the redox measurements were set initially at
1.0 ꢁ 10ꢀ5A/V. Ferrocene was used as the internal standard and its
HOMO was assigned to be ꢀ4.80 eV. The HOMO and LUMO energy
levels were calculated according to the following equations:
HOMO (eV) ¼ (Eox, onset e Eox, onset, ferrocene) þ 4.8
LUMO (eV) ¼ (Ered, onset e Eox, onset, ferrocene) þ 4.8
2. Results and discussion
2.1. Synthesis
Compound 1 (indigo), together with the other four soluble
indigoids (2a-b, 3a-b) all showed a characteristic double reduction
signals (see Fig. 3). All these compounds consist of two analogous
amide groups capable of forming mono- or di-amide anions when
deprotonated (or reduced). The oxidative potentials, however, was
insignificant to observe within the full scanning range from ꢀ1.8 V
to 1.8 V using a sensitivity at 1.0 ꢁ 10ꢀ5A/V partically for the highly
insoluble indigo. The oxidative potentials of indigo, however, was
only made observable when the sensitivity was increased to
1.0 ꢁ 10ꢀ6A/V. The low oxidative intensity was due to the electron
deficient nature of indigo and its derivatives. As these compounds
are highly electron deficient, the removal of an extra electron
(oxidation) from the molecules would be comparatively more
difficult than the addition of electrons (reduction) into the mole-
cules. As a result, the electron deficient nature make indigo and its
derivatives an excellent n-type organic semicondutor.
The onset value of Eox and Ered were estimated by extrapolating
the linear regions from the curves in the cyclic voltammograms.
The CV measured HOMO/LUMO (eV) for 1 (indigo), 2a-b, 3a-b and
their corresponding bandgap (Eg, in eV) are listed in Table 1.
As seen in Table 1, the 7,70-dialkoxyindigoids (2a, 2b) and 4,40-
Dibromo-7,70-dialkoxyindigoids (3a, 3b) all shown a decreased
LUMO compared to its parent reference compound 1 (indigo). The
HOMO on the other hand, remained more of less unchanged for the
two series of indigoids. The reduced LUMO level led to decreased
bandgap energies for compounds 2a-b and 3a-b. Hydrocarbon
substituents such as alkoxy and alkyl chains have been reported to
reduce the bandgap energy in semiconductive materials such as
polythiophenes and polythiazoles [17]. The narrowing of bandgap
energies in indigoids is in agreement with the aforementioned
examples. In both cases, the introduction of alkoxy groups lowered
the bandgap of 2a-b by 0.17e0.20 eV, while the bandgap of 3a-b
were further lowered by 0.22e0.25 eV.
An inexpensive 3-hydroxybenzaldehyde (3-HBA) was used as
the starting material for the synthesis of 7,70-dialkoxyindigoids (2a-
b). An Ullmann type etherification [13] was performed on 3-HBA
using DMF as the solvent to give the corresponding n-butoxy and
n-octyloxy benzaldehyde (5a-b) with 90e95% yield. Nitration on
5a-b was carried out with a nitrating reagent [14] to give the
alkoxynitrobenzaldehydes (6a-b) with 70e80% yield. The Baeyer
Drewson Indigo Synthesis [15] was then employed to produce the
7,70-dialkoxyindigoid (2a-b) with merely 12e18% yield.
The synthesis of 4,40-dibromo-7,70-dialkoxyindigoids (3a-b)
were similar to the procedures described above with 3-
alkoxybenzaldehyde (5a-b) employed as the starting material.
Bromination was performed by adding liquid Br2 to a chloroform
solution of 5a-b to give 2-bromo-5-alkoxybenzaldehyde (7a-b).
The bromination gave out a single ortho-directing regioselective
product 7a-b with 87e92% yield [16]. Nitration was then carried
out on 7a-b to give 6-bromo-3-alkoxy-2-nitrobenzaldehyde (8a-b)
with 72e80% yield. The Baeyer Drewson Indigo Synthesis was
employed to synthesize 4,40-dibromo-7,70-dialkoxyindigoid (3a-b)
with only 17e20% yield. The synthetic pathway is shown in Scheme
1.
The low yield for the Baeyer Drewson Synthesis in the last steps
4,40-Dibromo-7,70-dialkoxyindigoids (3a, 3b) have the lowest
bandgap energies reported. The additional electron-withdrawing
bromine atoms on 3a-b could have further lower the LUMO [18]
by making the indigoid core even more electron deficient. This
type of indigoid was considered to be the most electron deficient
among the different indigo species discussed in this article.
Fig. 2. Chemical structures of 7,70-dialkoxyindigoids (2a, 2b) and 4,40-dibromo-7,70-
dialkoxyindigoids (3a, 3b).

260
J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
Scheme 1. Synthetic routes for the organic soluble indigoids.
2.3. Optical properties
used as a reference compound for the study of hydrogen bond ef-
fects on the stacking of indigoids.
The UV-vis absorptions for all compounds were measured in
10e50 ppm chloroform solutions at room temperature. The UV-vis
spectra for the soluble indigoids (2a, 2b, 3a and 3b) were shown in
comparison to 1 (indigo) in Fig. 4. All of the samples showed intense
absorptions in the orange-red domain of the visible light region.
It was observed that compound 2a-b had a 28e30 nm red-
shifted in the absorption maximum from indigo. A red-shift in
absorption was mainly due to the auxochromic effect of the alkoxy
groups present on the indigoid core. However, the peak broadening
of indigoids comparative to indigo suggested that the material is
The absorptions were attributed to the
p
/
p
* transition [19,20]. All
either having an improved
stacking [22]. The observation is similar to the side-chain effects
in solid state polyhexylthiophenes [23]in which stacking
enhancement was a result of phase segregation between the hy-
drocarbon moieties and the aromatic thiophenes. The introduction
of alkoxy groups onto the indigo aromatic rings also lowered the
optical bandgap. Taking compound 2a-b as an example, their
bandgap energy reduced by approximately 0.13e0.14 eV in
p-bonds conjugation [21] or probably p-
soluble indigoids exhibited a broadened and red-shifted absorption
with respect to 1 (indigo). A higher full-width-half-maxima
(FWHM) value indicating the presence of auxochromic effect in
the soluble indigoids exerted by either the alkoxy or bromo moi-
eties on the compounds. The lmax and other optical data of each
sample were summarized in Table 2. Compound 4 will be further
discussed in the following X-Ray crystallography section as it was
p
p-p

J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
261
Fig. 4. Normalized UV-vis absorption spectra of 1 (indigo), 2a, 2b, 3a and 3b. Com-
pound 4 would be introduced in the next sections.
Table 2
UV-vis data of indigo and indigoids.a
FWHMc (nm)
Eg (eV)
b
Compound
lmax (nm)
ledge (nm)
1
605
635
633
640
649
651
702
706
707
707
72
99
99
95
96
1.90
1.77
1.76
1.75
1.75
2a
2b
3a
3b
a
Measurements were performed in chloroform solutions using a 1 cm quartz
cuvette at room temperature.
b
Eg was the optical bandgaps calculated from the absorption edges.
FWHM were estimated by the curve fitting function in the software OriginPro
c
9.0 by assuming the spectra followed the Gaussian system.
b which have the lowest bandgap energy among the samples
studied.
Fig. 3. Cyclic voltammograms of 1 (Indigo), 2a, 2b, 3a and 3b with ferrocene as an
internal standard. The arrows indicating the oxidation and reduction transition
regions.
2.4. Solubility
comparison to 1 (indigo) (see Table 2). The results were consistent
with the electrochemical bandgap energy estimated from cyclic
voltammetry.
Compounds 3a-b have the largest red-shift in their absorption
maximum. Their absorption maxima were red-shifted by
35e44 nm from indigo (1). A higher red-shift was conceivably due
to the synergistic effect of electron deficiency alkoxy and bromo
Solubility of the indigoids was validated by visual inspection on
the color of a solution containing 5 mg of the samples in 1 mL of the
solvent. The solution was first transferred to the top of a ~1 cm silica
gel column and was flushed down under an applied pressure. The
indigoids that are considered as soluble would pass through the
silica gel with their color retained, while insoluble indigoids would
have the color reside on top of the silica gel and only a colorless
filtrate was collected. Slightly soluble samples would produce a
pale blue filtrate even after successive washing with the solvent.
The solubility of the indigoids using different solvents is summa-
rized in Table 3.
substituents. The extension of
p system caused by the overlapping
p-orbital on the bromine substituents may also play a role in the
observed red-shift. The bandgap energy of 3a-b was further low-
ered by approximately 0.15 eV from 1 (indigo) (see Table 2). The
results were also in agreement with the CV data for compounds 3a-
Table 1
Electrochemical data of indigo and the indigoids.a
b
Compound
Eox, onset (V)
Ered, onset (V)
Eox, onset, ferrocene (V)
HOMO (eV)
LUMO (eV)
Eg (eV)
1c
1.106
1.140
1.027
1.116
1.162
ꢀ0.803
ꢀ0.599
ꢀ0.678
ꢀ0.540
ꢀ0.527
0.449
0.518
0.480
0.462
0.464
ꢀ5.46
ꢀ5.42
ꢀ5.35
ꢀ5.45
ꢀ5.50
ꢀ3.55
ꢀ3.68
ꢀ3.64
ꢀ3.80
ꢀ3.81
1.91
1.74
1.71
1.66
1.69
2a
2b
3a
3b
a
Measurements were performed in degassed DCM solutions with 0.1 M tetrabutylammonium hexafluorophosphate as the electrolyte and ferrocene as the internal
standard at room temperature.
b
Eg was the electrochemical bandgaps calculated from the difference of the HOMO LUMO.
As compound 1 (indigo) was only slightly soluble in DCM, the solution for 1 was slightly warmed and sonicated before the CV measurements in DCM.
c

262
J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
It had been reported that the poor solubility of indigo was
caused by strong intermolecular hydrogen-bonds. The enhanced
solubility of the indigoids (2a-b, 3a-b) was suggested to be a result
of the removal of the intermolecular hydrogen-bonds. An effective
intramolecular hydrogen-bond, however, should be preserved in
order for the stacking of the indigoids molecules to show the OFET
field-effect. The preservation and removal of inter- and intra-
molecular hydrogen-bonds were confirmed using single crystal x-
ray crystallography to be discussed in the following section.
Scheme 2. Synthesis of N-N0-dihexanoylindigoid (4).
2.5. X-ray crystallography analysis
In order to examine the effects of inter- and intramolecular
hydrogen-bonds on solubility and field-effect of the indigoids,
single crystal X-Ray analyses were performed on a selected indigoid
(2a) and a reference compound (4). The reference compound N-N0-
dihexanoylindigoid (4) was synthesized by acylation of compound
1 (indigo) with hexanoyl chloride (see Scheme 2). The hexanoyl
moiety allows indigoid (4) to be soluble yet at the same time
eliminated the possibility of inter- and intramolecular hydrogen-
bonds.
Single crystals of compound 2a and 4 were grown in chloroform
using hexane as the anti-solvent. The crystal growth took place
within 1e2 days under dark. The crystal structure of 2a and 4 were
shown in Fig. 5. Their images were resolved by an X-ray analysis
software Olex2 [24].
As shown in Fig. 5, the presence of intramolecular hydrogen-
bonds in 2a showed a completely different geometry from 4, as
the later compound had all of its hydrogen-bonds removed by
acylating of the analogous amides on its indigo core. For compound
2a, the atom-to-atom distance between the analogous amide
hydrogen and carbonyl oxygen was estimated to be 2.386 Å. The
distance was categorized as a “strong, mostly covalent” donor-
acceptor distance according to George A. Jeffrey [25]. The strong
intramolecular hydrogen-bonds in 2a provided a rigid and planar
molecular geometry between the two indoxyl rings as shown by
the single crystal X-Ray crystallography. The bulky butoxy groups
on the 7 and 70 position of 2a provided a steric effect to eliminate
intermolecular hydrogen-bonds between individual indigoid mol-
ecules, and thus afforded the soluble indigoid 2a. Moreover, a new
pair of “moderate” [25] intramolecular hydrogen-bonds with atom-
to-atom distance of 2.674 Å was also observed between the butoxy
oxygen atom and the analogous amide hydrogen atom on com-
pound 2a.
Fig. 5. Molecular structure of 2a and 4.
summary on the hydrogen-bonds on solubility and exhibition of
field-effects were given in Table 4. Interestingly, it was found that
intramolecular hydrogen-bonds within indigo or indigoids were
crucial for the exhibition of field-effect in OFET. As discussed earlier,
the intramolecular hydrogen-bonds preserved planarity of the in-
digo core and is one of the major factors for a working indigoid-
based OFET. Intermolecular hydrogen-bonds, however, only made
the indigoid insoluble. The natural occurring compound 1 (indigo)
would be of such case.
As seen in Fig. 4 previously, the UV-vis absorption for 4 was
found to have a significant blue-shift from 1 (indigo). The blue-shift
was probably attributed to the loss of the intra-molecular
hydrogen-bonds as the two indoxyl rings within 4 became no
longer co-planar, and therefore shortened the p-conjugation within
the compound. This is in consistent with its x-ray structure. The
In compound 4, the two amide hydrogen atoms on indigo were
substituted by two hexanoyl group, rendering it soluble in most
organic solvents. The central double bond on 4, however, was
greatly twisted. The two hydrophobic hexanoyl chains also tended
to align to one another, making the two indoxyl rings bent into a
“butterfly” geometry. As a result, the planarity of the indigo core
was completely destroyed.
Table 4
Effect of hydrogen bonds on solubility and OFET field-effect of indigo and their
derivatives.
Compound
Inter. H-bonds
Intra. H-bonds
Solubility
Field-effect
1
2a
4
þ
e
e
þ
þ
e
e
þ
þ
þ
þ
e
At the same time, OFET devices were fabricated to correlate the
influence of hydrogen-bonds on their field-effect properties. A
Notes: þ and ꢀ indicate the presence and absence of H-bonds, field-effect in OFET
and soluble or insoluble in common organic solvents.
Table 3
Solubility of indigo and indigoids in various solvents.
Compound
CHCl3
DCM
THF
EA
DMF
Acetone
Hexane
MeOH
H2O
1
INSa
INSa
INSa
INSa
SS
S
S
S
S
INS
SS
SS
S
INS
INS
INS
SS
INS
INS
INS
INS
INS
INS
INS
INS
INS
INS
2a
2b
3a
3b
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
SS
Notes: “S” indicates completely soluble; “SS” indicates slightly soluble; “INS” indicates insoluble.
a
Solution becomes slightly soluble upon slightly warmed and sonicated.

J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
263
CCDC numbers and other crystal packing images of compound 2a
and 4 can be found in the electronic supplementary materials of
this paper.
2.6. OFET measurements
The suitability of the soluble indigoids applied as n-type organic
semiconductor was demonstrated using a bottom-gate-top-contact
(BGTC) configuration OFETs (see Fig. 6). Polystyrene (PS) (MW
~35,000) was selected as the dielectric layer. A 30 mg/mL PS solu-
tion in chlorobenzene was spin-coated on a pre-cleansed p-doped
SiO2 substrate under nitrogen atmosphere and then vacuum dried.
The active layer, either compound 1 (indigo), 2a, 3a, 3b or 4, was
vacuum deposited on top of the dried polystyrene layer using an
Edwards (Auto 306A) evaporator. The same fabrication conditions
were employed in order to compare the results with the bench-
mark compound 1 (indigo). Al/LiF was used as the source/drain
electrodes and the channel width for all OFET devices was 50 mm.
Both indigoid series exhibited characteristic n-channel field-
effect behaviour (see Fig. 7). Under an applied gate voltage (VG),
the charge carriers, in this case electron, were accumulated at the
indigoid/PS interface. An effective resistance was created as elec-
trons travelled through the semiconducting active layer. As a result,
the current (IDS) flowing from source to drain was regulated by the
source/drain voltage (VDS) and the overall electron flows can be
modulated by the gate voltage (VG). The carrier mobility was
calculated at the saturated regime from the output curves (hori-
zontal regions from the VDS vs IDS curves in Fig. 7). The electron
mobility and current on/off ratio measured for the different indi-
goids are summarized in Table 5. As expected, compound 4 did not
exhibit any field-effect behaviour and its results were not
presented.
Fig. 7. Output curves of OFET devices based on 1 (indigo), 2a, 3a and 3b measured at
ambient temperature. Compound 3a showed a weaker transistor field-effect.
Table 5
Electron mobility of indigo and indigoids.
Compound S/D electrode mSat (cm [2]Vꢀ1 sꢀ1
)
Ion/off
1 (Indigo)
Al/LiF
Al/LiF
Al/LiF
Al/LiF
Al/LiF
2.40 ꢁ 10ꢀ5
1.66 ꢁ 10ꢀ5
4.80 ꢁ 10ꢀ6
2.20 ꢁ 10ꢀ5
6.1 ꢁ 102
1.8 ꢁ 10
N/A
2a
3a
3b
4
1.6 ꢁ 102
No field-effect observed No field-effect observed
The electron mobilities for the soluble indigoids are within the
same order of magnitude compared to 1 (indigo). The mobility
measured for all compounds (including indigo), however, are much
lower than the literature reported value (mSat ¼ 0.02 cm2Vꢀ1 sꢀ1 for
indigo [3]). It has been suggested that the performance for OFETcan
be greatly affected by the selection of the dielectric layer [26]. A
range of dielectric materials including vinyl polymers, crosslinkable
polymer precursor benzocyclobutene (BCB), and paraffin waxes
were studied for their effects on the performance of indigo-based
OFETs. As a result, electron mobility ranged from <10ꢀ4 to
10ꢀ3 cm2 Vꢀ1 sꢀ1 were detected. Recently, dielectric layer based on
a long aliphatic hydrocarbon chain, such as tetratetracontane (TTC),
was used as a template to induce the formation of a crystalline
structure for the indigo layer. An improvement in charge transport
properties resulted in OFETs [26] with highest msat at
9.1 cm2 Vꢀ1 sꢀ1 was reported. Therefore, the selection of a suitable
and compatible dielectric layer is crucial for a high performance
indigoid-based OFET. It is suggested either a photo-cured dielectric
layer or an inverted gate configuration would be used in future for
the optimization of the solution processed indigoid-based OFET
devices. The transistor transfer characteristic curves of OFET de-
vices based on 1 (indigo), 2a, 3a and 3b can be found in the sup-
plementary materials of this article.
3. Experimantal
3.1. Characterization
1H and 13C NMR spectra were recorded with a Bruker-AF301 AT
400 MHz spectrometer using CDCl3 or acetone-d6 as solvent and
tetramethylsilane as the internal standard. High resolution mass
spectrometry (HRMS) was carried out using a Bruker autoflex
MALDI-TOF mass spectrometer. Cyclic voltammetry (CV) mea-
surements were performed on a BAS CV-50W electrochemical
analyzer adapted with a conventional three-electrode configura-
tion consisting platinum working and auxiliary electrodes and a Ag/
AgCl reference electrode. The measurements were carried out in
0.1
M
DCM solution with tetrabutylammonium hexa-
fluorophosphate as the electrolyte and ferrocene as the internal
standard. UV-vis absorption spectra were measured in chloroform a
Cary UV-100 (Double Beam) spectrophotometer. Single-crystal
structural determination was performed using a Single-Crystal X-
Fig. 6. Configuration of a bottom-gate-top-contact (BGTC) OFET device.

264
J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
Ray Diffractometer (Smart 1000 with CCD Area Detector).
to remove any water residue. The desiccation salt was removed by
filtration and the filtrate was collected. Chloroform was used to
rinse any remaining product adhered on the salt and flasks. The
solvent was then removed by rotary evaporation, and dried in
vacuo to obtain a yellow brown viscous liquid of 2-bromo-5-
butoxybenzaldehyde (7a) (5.01 g, 87%). The product was used
directly in the next step and NMR analysis without further purifi-
cation by column chromatography. dH (Acetone-d6) 0.97 (3H, t, J 7.4,
CH3), 1.46e1.55 (2H, m, butoxy CH2), 1.74e1.81 (2H, m, butoxy CH2),
4.07 (2H, t, J 6.4, OCH2), 7.18 (1H, dd, J 8.8 and J 3.2, ArH), 7.36 (1H, d,
J 3.2, ArH), 7.63 (1H, d, J 8.8, ArH) and 10.25 (1H, s, CHO); dC
(Acetone-d6) 13.19,18.91, 30.94, 68.10,113.84,116.44,122.85,134.14,
134.75, 159.02, 190.81.
The synthesis of 2-bromo-5-(octyloxy)benzaldehyde (7b) was
done by the same procedure described to synthesize 2-bromo-5-
butoxybenzaldehyde (7a) except the 3-alkoxybenzaldehyde used
was 5.25 g of 3-(octyloxy)benzaldehyde (5b) instead of 5a. The
product given out was a yellow liquid of 2-bromo-5-(octyloxy)
benzaldehyde (7b) (6.46 g, 92%). dH (Acetone-d6) 0.88 (3H, t, J 6.8,
CH3), 1.26e1.52 (10H, m, octyloxy CH2), 1.76e1.83 (2H, m, octyloxy
CH2), 4.07 (2H, t, J 6.5, OCH2), 7.19 (1H, dd, J 8.8 and J 3.2, ArH), 7.37
(1H, d, J 3.2, ArH), 7.64 (1H, d, J 8.8, ArH) and 10.26 (1H, s, CHO); dC
(Acetone-d6) 14.37, 23.32, 26.67, 29.79, 30.00, 32.57, 69.32, 114.81,
117.32, 123.76, 135.08, 135.67, 159.96, 191.71.
3.2. Materials
Indigo,
3-hydroxybenzaldehyde,
1-bromobutane,
1-
bromooctane, bromine, potassium carbonate, sodium hydroxide
were purchased from Sigma Aldrich. Other solvents were analytical
grade and used as received.
3.3. Synthesis of indigoids
3-alkoxybenzaldehyde (5a-b): To a 500 mL round bottom flask,
10 g (0.0819 mol) of 3-hydroxybenzaldehyde (3-HBA) was dis-
solved in 150 mL dimethylformamide. 22.6 g (0.164 mol) potassium
carbonate was added to the solution of 3-HBA. 9.73 mL
(0.0901 mol) of 1-bromobutane was added in one portion to the
reaction mixture. The mixture was fitted with a condenser and was
heated to 160 ꢂC with a sand bath. The reaction was stirred vigor-
ously with a magnetic stirring bar with heating for 30 min. The
reaction temperature was then lowered to room temperature.
200 mL water was added to the mixture followed by 100 mL of
hexane. The solution was then stirred at room temperature until
the excess potassium carbonate powder was dissolved in the
aqueous layer. The solution was then transferred to a separating
funnel, and 100 mL of hexane was added. The aqueous layer was
discarded and the organic layer was washed with 10% HCl solution,
water and finally with saturated NaCl solution. The organic layer
was collected and anhydrous Na2SO4 or anhydrous MgSO4 was
added to remove residue of water in the organic solution. The
desiccation salt was removed by filtration and the filtrate was
collected. Chloroform was used to rinse any remaining product
adhered on the salt and flask. The solvent was then removed by
rotary evaporation, and dried in vacuo to obtain a pale yellow oily
liquid of 3-butoxybenzaldehyde (5a) (13.9 g, 95%). dH (Acetone-d6)
0.97 (3H, t, J 7.4, CH3), 1.46e1.56 (2H, m, butoxy CH2), 1.74e1.81 (2H,
m, butoxy CH2), 4.07 (2H, t, J 6.4, OCH2), 7.23e7.26 (1H, m, ArH),
7.43 (1H, d, J 2.6, ArH), 7.49e7.51 (2H, m, ArH) and 10.0 (1H, s, CHO);
dC (Acetone-d6) 13.26, 19.01, 31.09, 67.69, 113.27, 121.18, 122.31,
130.16, 138.22, 159.80, 191.89.
3-alkoxynitrobenzaldehyde
alkoxynitrobenzaldehyde (8a-b): To a 100 mL conical flask,
1523 L (0.0218 mol) 65% nitric acid was slowly added to 20 mL
(6a-b)
and
2-bromo-5-
m
(0.364 mol) 96% concentrated sulfuric acid at 0 ꢂC to make up the
nitrating reagent. The solution was stirred in an ice bath for 10 min.
It was added 3 g (0.0168 mol) of 3-butoxybenzaldehyde (5a)
dropwise to the nitrating reagent in 30 min via a dropping funnel.
The solution was kept stirring at 0 ꢂC for 30 min. Crushed ice was
added directly to the reaction mixture to precipitate the nitration
product. The crude product was collected by suction filtration and
washed with cold water. The crude product was dried in a vacuum
oven at room temperature to obtain a yellow powder of 3-
butoxynitrobenzaldehyde (6a) (3 g, 80%) and was used directly in
the next step without further purification.
The syntheses of 3-(octyloxy)nitrobenzaldehyde (6b), 6-bromo-
3-butoxynitrobenzaldehyde (8a) and 6-bromo-3-(octyloxy)nitro-
benzaldehyde (8b) were done by the same procedure described to
synthesize 3-butoxynitrobenzaldehyde (6a) except the benzalde-
hyde used was 3.94 g (0.0168 mol) of 3-(octyloxy)benzaldehyde
(5b), 5.5 g (0.0168 mol) of 2-bromo-5-butoxybenzaldehyde (7a) or
5.26 g (0.0168 mol) of 2-bromo-5-(octyloxy)benzaldehyde (7b)
instead of 5a. The products 6b (3.28 g 70%), 8a (4.06 g, 80%) and 8b
(4.33 g, 72%) obtained were all yellow powder and were used
directly in the next step without further purification or NMR
analyses.
The synthesis of 3-(octyloxy)benzaldehyde (5b) was done by the
same procedure described to synthesize 3-butoxybenzaldehyde
(5a) except the alkyl halide used was 13.7 mL (0.0901 mol) of 1-
bromooctane instead of 1-bromobutane. The product obtained
was a pale yellow oily liquid of 5b (17.3 g, 90%). dH (CDCl3) 0.89 (3H,
t, J 7.0, CH3), 1.29e1.48 (10H, m, octyloxy CH2), 1.76e1.83 (2H, m,
octyloxy CH2), 4.00 (2H, t, J 6.6, OCH2), 7.15e7.18 (1H, m, ArH), 7.38
(1H, d, J 2.1, ArH), 7.42e7.43 (2H, m, ArH) and 9.96 (1H, s, CHO); dC
(CDCl3) 14.10, 22.66, 26.01, 29.13, 29.24, 29.33, 31.81, 68.26, 112.72,
121.91, 123.25, 129.96, 137.76, 159.71, 192.16.
2-bromo-5-alkoxybenzaldehyde (7a-b): To a 500 mL round
bottom flask, 4 g (0.0224 mol) of 3-butoxybenzaldehyde (5a) was
dissolved in 200 mL chloroform. 1377 mL (0.0269 mol) of liquid
7,7′-dialkoxyindigoids (2a-b) and 4,4′-dibromo-7,7′-dialkox-
yindigoids (3a-b): To a 100 mL round bottom flask, 2 g of the crude
product of 3-butoxynitrobenzaldehyde (6a) was dissolved in 30 mL
propanone. The mixture was cooled to -5 ꢂC with an ice/CaCl2 bath
for 15 min. A solution of 4 mL 0.2 N sodium hydroxide solution was
added slowly to the propanone mixture with a dropping funnel
under moderate stirring. Upon the solution turned deep red in
color, 30 mL 0.4 N sodium hydroxide solution was added slowly
using a dropping funnel. The reaction mixture was stirred for
another 15 min. The temperature of the reaction mixture was then
raised to room temperature. When the mixture turned deep green
in color, the reaction mixture was further increased to refluxing
temperature and stirred for an additional 1 h. The temperature was
lowered to room temperature and 30 mL of methanol was added to
quench the reaction. The reaction mixture was cooled to 0 ꢂC with
an ice bath and stirred for 15 min. The crude product was filtered
bromine was dissolved in 10 mL chloroform in a dropping funnel
connected to the round bottom flask. The whole system was pro-
tected in nitrogen atomosphere. The bromine solution was slowly
added to the alkoxybenzaldehyde solution via a dropping funnel
over 30 min. The solution was kept at room temperature with
stirring. After the complete addition of bromine, the reaction was
stirred continuously for another 30 min 200 mL of cold water was
added to stop the reaction. The mixture was transferred to a
separating funnel. The crude product was extracted with chloro-
form and the aqueous layer was discarded. The organic layer was
further washed with a 20% sodium thiosulfate solution to remove
the remaining bromine. The product solution was then washed
with concentrated NaCl solution and dried with anhydrous MgSO4

J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
265
with suction filtration, washed with cold water and methanol, and
allowed to dry in a vacuum oven to obtain blue solids of 7,7′-
dibutoxyindigoid (2a). The blue solids were purified by flash col-
umn chromatography using toluene as eluent. The final purified
product of 2a (656 mg, 18%) were blue shiny crystalline needles. dH
(CDCl3) 1.02 (6H, t, J 7.4, CH3), 1.51e1.60 (4H, m, butoxy CH2),
1.82e1.89 (4H, m, butoxy CH2), 4.10 (4H, t, J 6.4, OCH2), 6.90 (2H, t, J
7.8, ArH), 6.99 (2H, dd, J 7.8 and J 0.7, ArH), 7.31 (2H, dd, J 7.6 and J
0.7, ArH) and 8.92 (2H, s, NH); dC (CDCl3) 13.88, 19.32, 31.18, 68.51,
115.71, 116.82, 120.80, 121.10, 121.75, 142.69, 145.50, 188.90; MALDI-
TOF-MS Calculated [M þ Hþ]: 407.1965; Found: 407.1963.
then further purified with silica gel column chromatography with
eluent of hexane: ethyl acetate ¼ 4:1. The final product of com-
pound 4 (1.47 g, 28%) was a reddish purple powder. dH (CDCl3) 0.85
(6H, t, J 6.8, CH3), 1.23e1.30 (8H, m, hexanoyl CH2), 1.76e1.82 (4H,
m, hexanoyl CH2), 2.84 (4H, br, hexanoyl CH2), 7.26 (2H, t, J 7.5,
ArH), 7.63e7.67 (2H, m, ArH), 7.76 (2H, d, J 7.6, ArH), 8.24 (2H, d, J
8.1, ArH); dC (CDCl3) 13.85, 22.41, 25.30, 31.41, 36.19, 117.00, 121.84,
124.35, 124.96, 125.99, 138.86, 149.37, 173.71, 184.15; MALDI-TOF-
MS Calculated [M þ Naþ]: 481.2103; Found: 481.2090.
4. Conclusions
The syntheses of 7,7'dioctyloxyindigoid (2b) was done by the
same procedure described to synthesize 7,7′-dibutoxyindigoid (2a)
except the nitrobenzaldehyde used were 2 g 6b. The product 2b
(446 mg, 12%) obtained were blue shiny crystalline needles. dH
(CDCl3) 0.90 (6H, t, J 6.8, CH3), 1.25e1.60 (20H, m, octyloxy CH2),
1.83e1.90 (4H, m, octyloxy CH2), 4.08 (4H, t, J 6.6, OCH2), 6.90 (2H, t,
J 7.7, ArH), 6.97 (2H, d, J 7.7, ArH), 7.31 (2H, d, J 7.6, ArH) and 8.92
(2H, s, NH); dC (CDCl3) 14.14, 22.69, 26.07, 29.15, 29.26, 29.38, 31.86,
68.80, 115.68, 116.79, 120.78, 121.08, 121.72, 142.64, 145.49, 188.89;
MALDI-TOF-MS Calculated [Mþ]: 518.3145; Found: 518.3133.
The synthesis of 4,4'dibromo-7,7′-dibutoxyindigoid (3a) and
4,4'dibromo-7,7′-dioctyloxyindigoid (3b) were done by the same
procedure described to synthesize 7,7′-dialkoxyindigoids (2a-b)
except the nitrobenzaldehyde used were 2 g of 2-bromo-5-
We have reported the synthesis of two series of novel organic
soluble indigoids namely 7,7′-dialkoxyindigoids (2a, 2b) and 4,4′-
dibromo-7,7′-dialkoxyindigoids (3a, 3b) both started from the low-
cost 3-hydroxybenzaldehyde. The indigoids are soluble in common
organic solvents including chloroform, DCM, THF, EA and DMF. The
HOMO/LUMO and bandgap energy (Eg) of the alkoxy and bromine
substituted indogoids were all found to be lower than their parent
indigo compound. The lowest HOMO/LUMO at ꢀ5.50 eV/ꢀ3.81 eV
and the lowest reported bandgap energy (Eg) at 1.69 eV was found
for compound 3a.
By analyzing single-crystal x-ray crystallography for the indi-
goids, it was found that the presence of intramolecular hydrogen-
bonds is crucial for the exhibition of field-effect behaviour, while
intermolecular hydrogen-bonds lead to insolubility. Long alkoxy
chain substituents at the 7 and 70 positions on the indigo core
effectively eliminated the intermolecular hydrogen-bonds for the
purpose of improve solubility, and at the same time preserving
intramolecular hydrogen-bonds necessary for field-effect
behaviour.
butoxybenzaldehyde
(8a)
or
2
g
of
2-bromo-5-
octyloxynitrobenzaldehyde (8b). The final product obtained from
8a was a bluish purple powder of 3a (747 mg, 20%). dH (CDCl3) 1.00
(6H, t, J 7.3, CH3), 1.49e1.58 (4H, m, butoxy CH2), 1.80e1.87 (4H, m,
butoxy CH2), 4.05 (4H, t, J 6.5, OCH2), 6.78 (2H, d, J 8.4, ArH), 6.97
(2H, d, J 8.4, ArH) and 8.98 (1H, s, NH); dC (CDCl3) 13.87, 19.30, 31.11,
68.80, 109.22, 117.42, 118.19, 121.59, 124.86, 134.95, 144.89, 186.60;
MALDI-TOF-MS Calculated [M þ Hþ]: 565.0157; Found: 565.0166.
The Baeyer-Drewson product obtained from 8b was a bluish
purple powder of 3b (642 mg, 17%). dH (CDCl3) 0.90 (6H, t, J 6.6,
CH3), 1.31e1.51 (20H, m, octyloxy CH2), 1.80e1.87 (4H, m, octyloxy
CH2), 4.03 (4H, t, J 6.6, OCH2), 6.77 (2H, d, J 8.4, ArH), 6.96 (2H, d, J
8.3, ArH) and 8.97 (2H, s, NH); dC (CDCl3) 14.15, 22.69, 26.02, 29.10,
29.23, 29.36, 31.86, 69.14, 109.17, 117.37, 118.15, 121.50, 124.82,
143.88, 144.86, 186.43; MALDI-TOF-MS Calculated [M þ Naþ]:
699.1229; Found: 699.1268.
The employment of the soluble indigoids as an active layer for n-
type
OFET
have
electron
mobility
ranged
from
1.66 ꢁ 10ꢀ5 cm2Vꢀ1 sꢀ1 to 2.20 ꢁ 10ꢀ5 cm2Vꢀ1 sꢀ1 and is within the
same order of magnitude for the pristine indigo fabricated under
the same conditions and configuration. The values, however, are
lower than those reported in literature and is suggested relating to
the selection of the dielectric layer. Further device improvement
included the use of a hydrophobic dielectric layer or employing a
different device configuration would be the subject for further
study. The bromine substituted 3a and 3b can be further func-
tionalized and their chemistry and properties will also be reported
in a future communication.
N-N′-dihexanoylindigoid (4): To a 250 mL round bottom flask,
3 g (0.0114 mol) of compound 1 (indigo) was dissolved in 100 mL
dry THF and was degassed for 10 min using a sonicator. The solution
was then protected in nitrogen atmosphere, followed by the addi-
tion of 4.38 g (0.0456 mol) of sodium tert-butoxide in a single
portion. The reaction mixture was stirred vigorously at room
Acknowledgements
The authors gratefully acknowledge the support by the Uni-
versity Faculty Research Grant (FRG 1/14-15/046). The authors also
thank Dr. Zhu Nianyong from Prof. Wong Wai-yeung's research
group for the X-ray crystallography measurements.
temperature for 15 min. A solution of 4780
mL (0.0342 mol) of
hexanoyl chloride was added slowly to the solution. The reaction
mixture was heated at 60 ꢂC for 30 min with an oil bath in nitrogen
atmosphere. Once the solution turned deep purple in color, the
reaction mixture was cooled to room temperature. 50 mL chloro-
form was added to the mixture and the purple slurry was filtered
and washed with chloroform to remove any organic insoluble side
products. The filtrate was collected and transferred to a separating
funnel. The organic portion was washed with 20% sodium bicar-
bonate solution to remove any unreacted acid. The organic layer
was washed 3 times with water to remove any water soluble im-
purities. The aqueous layer was discarded and the organic layer was
collected. Anhydrous MgSO4 was added to the product solution to
remove residue of water. The desiccation salt was removed by
simple filtration and the filtrate was collected. Chloroform was used
to rinse any remaining product adhered on the MgSO4 salt and
flask. The solvent was then removed by rotary evaporation to
obtain a reddish pink powder. N-N’-dihexanoylindigoid (4) was
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.orgel.2016.02.035.
References
[1] Bernard T. Golding, Colin Pierpoint, Education in Chemistry, Adv. Mater (1986)
71e73.
[2] I.V. Klimovich, L.I. Leshanskaya, S.I. Troyanov, D.V. Anokhin, D.V. Novikov,
A.A. Piryazev, D.A. Ivanov, N.N. Dremova, P.A. Troshin, J. Mater. Chem. C 2
(2014) 7621.
[3] Eric Daniel Gowacki, Gundula Voss, Lucia Leonat, Mihai Irimia Vladu,
Siegfried Bauer, Niyazi Serdar Sariciftci, Isr. J. Chem. 52 (2012) 540.
[4] E.D. Glowacki, N.S. Sariciftci, G. Voss, K. Demirak, M. Havlicek, N. Sünger,
A.C. Okur, U. Monkowius, J. Gasiorowski, L. Leonat, Chem. Commun. Camb.

266
J.H.L. Ngai et al. / Organic Electronics 32 (2016) 258e266
Engl. 49 (2013) 6063.
[5] P.H.H. Gray, Proc. R. Soc. B Biol. Sci. 102 (1928) 263e280.
[6] R. Rondao, J.S. Seixas de Melo, V.D.B. Bonifacio, M.J. Melo, J. Phys. Chem. A 114
(2010) 1699.
[7] Yukihiro Matsumoto, Hitoshi Tanaka, Heterocycles 60 (2003) 1805e1810.
[8] Bradley D. Smith, David E. Alonso, Jeffrey T. Bien, John D. Zielinski, Sheridan
L. Smith, Kenneth J. Haller, J. Org. Chem. 58 (1993) 6493e6496.
[9] Chang Guo, Bin Sun, Jesse Quinn, Zhuangqing Yan, Yuning Li, J. Mater. Chem. C
2 (2014) 4289e4296.
Keisuke Tajima, Synth. Met. 196 (2014) 139e144.
[18] Raghunath R. Dasari, Amir Dindar, Chi Kin Lo, Cheng-Yin Wang,
Cassandre Quinton, Sanjeev Singh, Stephen Barlow, Canek Fuentes-Hernan-
dez, John R. Reynolds, Bernard Kippelen, Seth R. Marder, Phys. Chem. Chem.
Phys. 16 (2014) 19345e19350.
[19] J. Seixas de Melo, R. Rondao, H.D. Burrows, M.J. Melo, S. Navaratnam, R. Edge,
G. Voss, J. Phys, Chem. 110 (2006) 13653e13661.
[20] S.R. Oakley, G. Nawn, K.M. Waldie, T.D. MacInnis, B.O. Patrick, R.G. Hicks,
Chem. Commun. Camb. Engl. 46 (2010) 6753.
~
ꢀ
[10] Albrecht Schmidt, Process of Brominating Indigo, United States Patent Office,
1905.
[21] D. Fichou, Handbook of Oligo- and Polythiophenes, Wiley-VCH, Weinheim,
1999.
[11] L.L. Lloyd, The Chemistry of Dyestuffs: A Manual for Students of Chemistry
and Dyeing, 1917, pp. 270e272.
[22] N.K. Al-Rasbi, C. Sabatini, F. Barigelletti, M.D. Ward, Dalton Trans. Camb. Engl.
2003) (2006) 4769.
[12] M. Yao, K. Kuratani, T. Kojima, N. Takeichi, H. Senoh, T. Kiyobayashi, Sci. Rep. 4
(2014) 3650.
[23] Wen Y. Huang, C.C. Lee, S.G. Wang, Y.K. Han, M.Y. Chang, J. Electrochem. Soc.
157 (9) (2010) B1336eB1342.
[13] Gondola Voss, Michael Gradzielski, Jiirgen Heinz, Helmut Reinked,
Carlo Unverzag, Helvetica Chim. Acta 86 (2003) 1982e2004.
[14] L. Rubenstein, J. Chem. Soc. 127 (1925) 1998.
[15] A. von Baeyer, Ber Dtsch, Chem. Ges. 12 (1879) 1309.
[16] Willem A.L. van Otterlo, Joseph P. Michael, Manuel A. Fernandes, Charles B. de
Koning, Tetrahedron Lett. 45 (2004) 5091e5094.
[24] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann, J. Appl.
Cryst. 42 (2009) 339e341.
[25] George A. Jeffrey, An Introduction to Hydrogen Bonding, Oxford University
Press, 1997.
[26] D.V. Anokhin, L.I. Leshanskaya, A.A. Piryazev, D.K. Susarova, N.N. Dremova,
E.V. Shcheglov, D.A. Ivanov, V.F. Razumov, P.A. Troshin, Chem. Commun.
Camb. Engl. 50 (2014) 7639.
[17] Qingshuo Wei, Shoji Miyanishi, Erjun Zhou, Kazuhito Hashimoto,