Solvent-free thiophene-based electrolytes: synthesis of new liquid-crystalline ionic conductors for batteries: part I

Article abstract of DOI:10.1039/C8DT02968G

New liquid crystalline thiophene esters, thiophene-3-yl 3,4,5-tris(n-dodecan-1-yloxy)benzoate (ThBz1) and thiophene-3-yl 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoate (ThBz2), were synthesized for use in lithium-ion batteries as safe, new solvent-fre

Full text of DOI:10.1039/C8DT02968G

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Solvent-free thiophene-based electrolytes:
synthesis of new liquid-crystalline ionic
conductors for batteries: part I
Cite this: Dalton Trans., 2018, 47,
15714
a
b
b
Krzysztof Artur Bogdanowicz, * Pawel Gancarz,
Damian Pociecha,
Michal Filapek,
c
d
e
a
Monika Marzec,
Ida Chojnacka
and Agnieszka Iwan
New liquid crystalline thiophene esters, thiophene-3-yl 3,4,5-tris(n-dodecan-1-yloxy)benzoate (ThBz1)
and thiophene-3-yl 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoate (ThBz2), were synthesized for
use in lithium-ion batteries as safe, new solvent-free ionic conductors. Both compounds were syn-
1
13
thesized in multi-step reactions and were characterized via H, C NMR and FT-IR spectroscopy,
confirming the formation of the desired products. ThBz1 and ThBz2 showed good thermal stability and
liquid crystalline behaviour. Electrochemical tests demonstrated a wide electrochemical window of 4 V
Received 20th July 2018,
Accepted 5th October 2018
DOI: 10.1039/c8dt02968g
for both tested ThBz compounds. Moreover, preliminary battery tests for new electrolytes in the LiCoO
2
/
rsc.li/dalton
Li Ti 12 system confirmed their applicability in lithium-ion batteries.
4
5
O
cyclability and safety of the system. Among the targets are the
widely used liquid electrolytes with soft materials like gel elec-
Introduction
1
3
Currently, there is worldwide focus on increasing the renew- trolytes that contain conductive polymers.
One of the
able energy production in global markets by replacing conven- solvent-free methods includes the use of liquid crystals (LCs)
tional fuels like gas oil with, for example, solar and wind as the electrolyte. LCs combine peculiar properties like fluidity
1
,2
power. However, harvesting solar or wind energy has its dis- and the ordered structures of fluids and crystalline solids.
advantages due to the limited availability of its maximum When incorporated into structures with active elements (i.e.
potency. This issue is possibly the main driving force for the stilbene groups, dendrons), LCs enable the formation of self-
3
,4
development of energy storage systems like galvanic cells.
organized, nano- and sub-nanosized 1D, 2D and 3D structures
1
4
Reversible galvanic cells are of great interest to scientists capable of transporting particles like ions in a selective way.
and various industries, as reflected in the global industrial Kato and co-workers demonstrated the well-organized for-
interest and the exponential increase in the number of pub- mation of proper chemical structures. The building block
5
,6
lished articles on batteries.
molecules contain several components such as ionic liquids
The technology of rechargeable batteries, in particular, and ionophoric moieties, both ensuring ion transport, and
lithium-ion batteries, has broad applicability in diverse porta- mesogenic groups that ensure proper self-assembly
1
5,16
ble systems such as mobile phones, electric or hybrid cars and processes.
stationary systems as the electrical energy storage for units pro-
ducing electricity from renewable sources.
At present, researchers engaged in lithium-ion battery works.
Additionally, the self-assembled structure can be induced
and improved by thermal treatment as shown in our previous
7
–12
1
7–19
Their use for applications in lithium-ion batteries
studies are focused on many aspects like the durability, as electrolytes, in photovoltaic devices as semiconductors and
potentially in fuel cells as ion-conducting separators has been
proven but requires more attention to improve their
2
0
performance.
a
Military Institute of Engineer Technology, Obornicka 136 Str., 50-961 Wroclaw,
The main goal of our study is to design and synthesize
novel liquid crystalline molecules containing the thiophene
unit modified with dendritic wedge-shaped molecules. In
order to modify the properties of new polymers with thiophene
rings, molecular engineering was involved based on the use of
starting materials with different chemical structures. It was
expected that these new materials would form ion-conductive
Poland. E-mail: bogdanowicz@witi.wroc.pl
b
Institute of Chemistry, University of Silesia, 9 Szkolna Str., 40-006 Katowice, Poland
c
University of Warsaw, Department of Chemistry, Zwirki i Wigury 101, 02-089
Warsaw, Poland
d
Institute of Physics, Jagiellonian University, prof. S. Lojasiewicza 11, 30-
3
48 Krakow, Poland
e
Department of Analytical Chemistry and Chemical Metallurgy, Wroclaw University
of Technology, Wyb. St. Wyspianskiego 27, 50-370 Wroclaw, Poland
15714 | Dalton Trans., 2018, 47, 15714–15724
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
channels, depending on their chemical structure. Therefore, in
this study, we investigated two solvent-free thiophene-based
electrolytes. We report the influence of molecular interaction
on the thermal, structural and electrochemical behaviour of
thiophene based electrolytes. We also investigate the influence
of LiTFSI on the electrochemical properties of both com-
pounds towards applications in organic devices. To the best of
our knowledge, for the first time, solvent-free thiophene-based
electrolytes are analysed for use in lithium-ion batteries.
Results and discussion
Scheme 2 Scheme showing Bz1 and ThBz1 synthesis.
In this paper we propose for the first time two solvent-free
thiophene-based electrolytes, thiophene-3-yl 3,4,5-tris(n-
dodecan-1-yloxy)benzoate (ThBz1) and thiophene-3-yl 3,4,5-tris
benzoate. Next, the methyl ester was converted into the potass-
ium salt in a one step reaction with ethanolic KOH (Scheme 2).
It is known from the literature that both Bz1 and Bz2 have
been used to modify polymers in order to induce the liquid
crystalline hexagonal structure. Due to their particular mole-
cular structure, the aromatic core is responsible for the planar
form and rigidity, whereas the long alkyl chains give the mobi-
lity. The combination of those two features translates to liquid
crystalline (LC) properties that can be transposed onto another
[
4-(n-dodecan-1-yloxy)benzyloxy]benzoate (ThBz2), synthesized
by S 1 and S 2 substitutions as is schematically shown in
N
N
Scheme 1. In the first step of the reaction, two mesogens, pot-
assium 3,4,5-tris(n-dodecan-1-yloxy)benzoate (denoted as Bz1)
and potassium 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzo-
ate (Bz2), were synthesized. The first step of the synthesis of
both mesogenic groups and the third step in the synthesis of
Bz2 followed the S 1 mechanism, where the leaving group was
N
a halogen atom and the hydroxyl group at the phenyl ring was
the nucleophile. The last reaction between 3-chlorothiophene
2
1,22
molecule via chemical bonding.
The major difference
between both mesogenic groups lies in the influence of the
presence of additional phenyl rings; they increase the isotropi-
zation temperature of the LC, and the increased bulkiness
narrows the orientation of channels formed by the columnar
and potassium carboxylic salt obeyed the S
chlorine as the leaving group and the carboxyl anion as the
nucleophile. Obtained products were characterized by H,
NMR and FT-IR spectroscopy, which fully confirmed the pro-
posed structure of all synthesized compounds.
N
2 mechanism with
1
13
C
2
2
structure.
On the other hand, in the synthesis route of Bz2, the p-(n-
dodecan-1-yloxy)benzyl alcohol was obtained by refluxing a
solution of 4-hydroxybenzyl alcohol, 1-bromododecane, anhy-
In detail, the thiophene-3-yl 3,4,5-tris(n-dodecan-1-yloxy)
benzoate and thiophene-3-yl 3,4,5-tris[4-(n-dodecan-1-yloxy)
benzyloxy]benzoate were obtained via 3-chlorothiophene modi-
fication by a bulky tapered group, i.e., potassium 3,4,5-tris(n-
dodecan-1-yloxy)benzoate (abbreviated here Bz1) or potassium
drous potassium carbonate and
a catalytic amount of
18-crown-6 ether in acetone under argon for 24 hours. p-(n-
Dodecan-1-yloxy)benzyl alcohol was converted into a chloride
in the reaction with thionyl chloride at room temperature in
dichloromethane to obtain 4-(n-dodecan-1-yloxy)benzyl chlor-
ide, which was allowed to react with methyl 3,4,5-trihydroxy-
benzoate. Subsequently, potassium salt was prepared from the
methyl ester by reaction with ethanolic KOH (Scheme 3).
3
,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoate (abbreviated
here Bz2) in order to get the hexagonal columnar mesophases.
The synthesis pathway to the required bulky tapered groups
includes multi-step reactions.
For Bz1, 1-bromododecane reacted with methyl 3,4,5-
hydroxy benzoate to give methyl 3,4,5-tris(n-dodecan-1-yloxy)
Structural characterization of ThBz1 and ThBz2
FT-IR spectra were obtained in order to confirm the chemical
structure of the synthesized compounds. Fig. 1 presents the
whole spectral range between 500 to 4000 cm− for ThBz1 and
ThBz2.
1
In the spectra of both compounds, the most prominent
bands are assigned to the ν (CH ) and ν (CH ) stretching
s
2
as
2
modes of aliphatic long chains in the range of
−
1
2
1
750–3000 cm . Also, the (ν Ph) in the spectral range of
700–1500 cm− and the characteristic (γ CH ) vibrations of
1
r
aromatic rings were observed. Moreover, the C–S vibration at
20 cm− is a confirmation of the thiophene unit, and the
1
7
Scheme 1
S
N
1 (top) and S
N
2 (bottom) substitution, where R is the
−
organic group, L is the leaving group, and Nu is a nucleophile.
ν(C–C) modes are present in the low wavenumber range of
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15714–15724 | 15715

View Article Online
Paper
Dalton Transactions
The structures of the ThBz1 and ThBz2 were characterized
1
by NMR spectroscopy. Fig. 2a and 3a present the H NMR
1
spectra of ThBz1 and ThBz2 in CDCl
3
, respectively. All
H
NMR spectra are characterized by relatively sharp signals in
three regions. In the case of ThBz1, the aromatic region shows
1
partially overlapped signal at 7.34 (Fig. 2a). Considering the
relative areas, the signal at 7.34 ppm can be assigned to the
protons of the Bz1 unit ortho to the carbonyl group and to the
proton in position 5 in the thiophene ring. On analysing the
relative areas of the signals at 7.29 and 6.19 ppm, they can be
assigned to 2 thiophene protons in positions 2 and 4. The fol-
lowing signal at 4.00 ppm corresponds to the methylene
attached to the oxygen in the long alkyl chains, whereas the
characteristic signals of the protons that correspond to the
long alkyl chains in Bz1 can be observed in the upfield region
at 1.78, 1.44, 1.26 and 0.88 ppm.
In the case of ThBz2, the signals are similar to those
observed for ThBz1 with additional signals corresponding to
different structures of Bz2. The difference lies in the aromatic
and middle region: first, the protons coming from the four-
substituted phenyl ring (benzoate unit) appear at 7.53 ppm,
due to the presence of the benzyl units substituted in positions
1
and 4; characteristic doublet of doublets (dd) can be
Scheme 3 Scheme showing Bz2 and ThBz2 synthesis.
Fig. 1 FT-IR spectra of ThBz1 and ThBz2 recorded at 293 K. Arrows
indicate characteristic signals.
00–1000 cm− . The carbonyl ν(CvO), characteristic of ester
1
5
−
1
carbonyls, appears at 1725 and 1722 cm for ThBz1 and
ThBz2, respectively. Due to the presence of several aromatic
segments in ThBz2, the carbonyl signal overlaps with signals
assigned to phenyl ring modes at approximately 1690 cm−
1
.
Confirmation of ester bond formation can be also found in the
spectra as the signals assigned to ν (CvO)–O and ν COC at
242 and 1104 cm− , respectively. The FT-IR spectra of ThBz1
1
1
and ThBz2 fully confirm the proper compositions – all the
1
Fig. 2 (a) H NMR and (b) 13C NMR spectra of ThBz1. Arrows indicate
expected essential bands are present in the spectra.
characteristic peaks.
15716 | Dalton Trans., 2018, 47, 15714–15724
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Thermal and mesomorphic characterization of ThBz1 and
ThBz2
Both thiophene compounds, ThBz1 and ThBz2, were charac-
terized using POM, DSC, TGA and XRD techniques to investi-
gate their thermal stabilities and their liquid crystalline
properties.
The thermal stability of thiophene derivatives was investi-
gated using TGA under a nitrogen atmosphere (Fig. 4). In
general, the thermal stability of ThBz1 is greater compared to
ThBz2, with the temperatures corresponding to 5% weight
loss, T5%, being equal to 350 °C and 233 °C, respectively. T50%
for both are similar (405 °C for ThBz1 and 398 °C for ThBz2),
whereas the weight residue for ThBz1 was lower than in the
case of ThBz2 and equal to 7% and 13%, respectively. The
difference in the thermal behaviour can be explained by the
difference in the structures of both compounds. ThBz2 has
benzyl groups connected to the phenyl groups. It is known
2
3
that this kind of bond has a greater tendency to rupture. On
the other hand, the residual value is higher as a result of a
larger number of phenyl units in the molecule.
Textures registered by POM for both studied compounds
are presented in Fig. 5. In both cases, only one colour texture
was detected during heating, which indicates one anisotropic
phase. The texture slightly changed colour with increasing
temperature to become black at the transition to the isotropic
phase.
On cooling, the texture is dark and become a little brighter
with decreasing temperature. During the second heating, this
dark texture does not change much and remains dark for
heating and cooling.
1
13
Fig. 3 H NMR and C NMR spectra of ThBz2 (a and b, respectively).
Arrows indicate characteristic peaks.
The temperatures obtained by POM for both ThBz1 and
ThBz2 match well with the effects observed for the DSC curves.
The calorimetric data are presented in Table 1. The analysis of
the endothermic effects during the heating scan for ThBz1
reveals the presence of two maxima at 10.6 °C and 48.9 °C.
Judging from the shape of these signals it seems that approxi-
mately four overlapping effects occurred, like the melting of
different segments of the structure, for instance, aliphatic
observed at 7.20–7.35 ppm and 6.40–6.80 ppm. Other
additional signals present at 4.8–5.0 correspond to the methyl-
oxy linkage between aromatic rings in the structure. The
signals of the thiophene protons also shift and appear at 6.87
and 7.25 ppm. When compared with the starting materials Bz1
and Bz2, the signals from the aromatic region shift downfield
to 2 ppm.
1
3
Fig. 2b and 3b show the C NMR data of for both syn-
thesized compounds. The carbonyl carbon of the introduced
ThBz1 appeared at 165.9 ppm. The aromatic carbons appeared
between 108.8 and 158.6 ppm. The aliphatic carbons beta to
the oxygen and the rest of the aliphatic chain of the long alkyl
chains appear at 14.0–31.8 ppm, as expected, whereas the
carbon labelled as alpha to the oxygen in the alkyl chains
appeared at 67.8 ppm. In the case of ThBz2, additional signals
coming from the aromatic benzylic groups appeared at 108.8
and 158.6 ppm, and the benzylic methylenes appeared at
7
0.8 ppm for the 3 and 5 substituted groups, whereas the
group in position 4 appeared downfield at 74.6 ppm.
Regarding the thiophene unit, the signals appeared in the
region similar to the aromatic carbons from Bz moieties, as
broad, low-intensity signals sometimes occur together with
signals from the mesogenic unit.
Fig. 4 TGA curves for ThBz1 and ThBz2 during heating at a rate of 10 K
−1
min
.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15714–15724 | 15717

View Article Online
Paper
Dalton Transactions
from the isotropic liquid is still present as can be seen in
Fig. 6. Similar results were observed for ThBz2, which is in
good accordance with the abovementioned results from DSC
and POM. However, on analysing the Debye rings for both
fresh samples, an evident partial arch can be seen in both
cases (Fig. 6). It is known that the presence of narrow arches is
2
4
linked to the information about orientation; the narrower
Fig. 5 Optical textures registered for ThBz1 and ThBz2 during heating
−1
and cooling at a rate of 10 K min (cross polarizes).
Table 1 DSC data for ThBz1 and ThBz2
Heat (μV s− mg−1
)
1
Code
Mass (mg)
T
trans (°C)
Temperatures and heat of thermal effects determined from the heating
curves
ThBz1
4.5
5.8
10.6
258.42
44.00
359.49
163.06
107.96
312.12
2
3
4
5.4
7.4
8.9
ThBz2
−10.5
44.2
Temperatures and heat of thermal effects determined from the cooling
curves
ThBz1
a
4.5
5.8
6.5
0.5
13.8
457.77
114.63
280.49
1
b
ThBz2
a
b
g p g p
T = −27.5 °C, ΔC = −0.807 μV. T = −42.4 °C, ΔC = −1.059 μV.
chains. The glass transition was observed with the cooling
scan at −27.5 °C, together with two exotherms at 6.5 °C and
1
0.5 °C.
In the case of ThBz2, the heating scan depicts two maxima
at −10.5 °C and 44.2 °C, which under closer evaluation rep-
resent single thermal effects. Also, in this case, the glass tran-
sition was only observed during cooling at a lower temperature
than for ThBz1, −42.4 °C, as expected. During cooling, one
endothermic effect was observed at 13.8 °C.
XRD analysis for ThBz1 revealed the crystalline phase in the
fresh sample at T = 30 °C, an isotropic liquid at T = 50 °C, and
slow recrystallization after cooling back to T = 30 °C; the signal Fig. 6 XRD diffractograms and Debye rings for ThBz1 and ThBz2.
15718 | Dalton Trans., 2018, 47, 15714–15724
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
the arch, the more oriented the structure is. In the present to the next two oxidation stages. In the LiTFSI solution, the
work, despite the fact that the columnar structure could not be first oxidation step has a higher intensity than the second oxi-
clearly confirmed, it is evident that some self-orientation of dation, but the third step is the most apparent. In the case of
the fresh sample occurs.
the ThBz2, two oxidation stages are well presented in both sup-
porting electrolyte solutions, but the third oxidation process
can be seen between the first and the second oxidation peak.
The small peak appears before 0 V during oxidation of the
ThBz2 in the LiTFSI solution.
The onset potentials listed in Table 2 are in good agreement
with different supporting electrolyte solutions for the first oxi-
dation step. The first oxidation onsets are similar and equal to
Electrochemistry of ThBz1 and ThBz2
In Table 2, the onsets and peak potentials of the ThBz1 and
the ThBz2 compounds determined for the oxidation processes
are presented. Measurements were made with different sup-
porting electrolytes in propylene carbonate.
Firstly, DPV experiments were carried out in the propylene
0
.97 V (Bu
(LiTFSI) for ThBz1 and ThBz2, respectively.
Reduction of the ThBz1 and the ThBz2 in the Bu NPF solu-
4 6 4 6
NPF ), 1.06 V (LiTFSI) and 1.00 V (Bu NPF ), 0.98
carbonate towards Bu NPF as well as in the LiTFSI. Fig. 7 pre-
4
6
sents the oxidation processes of the substances ThBz1 and
ThBz2 revealed by differential pulse voltammetry. The ThBz1
undergoes three-stage oxidation; however, the first oxidation
4
6
tion starts at −2.83 V and −2.87 V, respectively (Fig. 8). The
potential during measurements was driven to lower values
than the start of the decomposition of the Bu NPF in the pro-
4 6
process has a small intensity in the Bu NPF solution, contrary
4
6
pylene carbonate. It allowed the observation of the possible
reduction processes of the examined substances in the range
of the lithium cation reduction in propylene carbonate,
and indicates that ThBz1 and ThBz2 undergo single step
reduction before the decomposition of the supporting electro-
Table 2 The onsets (on), the anodic (Epa) and the cathodic (Epc) peak
potentials of the ThBz1 and ThBz2 from the oxidation processes. The
potentials determined with less accuracy than ±0.01 V, due to the peak
broadening or the absence of the well-shaped maximum, are in italics.
25,26
The dash means that it was not possible to determine an onset or a peak lyte solution.
potential
Reduction of the ThBz1 and the ThBz2 starts at lower
potentials (−2.33 V and −2.50 V, respectively) in the LiTFSI
solution, which is shown in Fig. 9, but about 20 times lower
current was exchanged with the working electrode.
ThBz1
Bu NPF
ThBz2
Bu NPF
Parameters
4
6
LiTFSI
4
6
LiTFSI
Afterward, the oxidation and the reduction processes were
examined by use of the cyclic voltammetry. As can be seen in
Fig. 10, onsets of the oxidation processes are similar for
both substances and are in good agreement with the DPV
voltammogram. On the CV voltammogram, the mild slope of
the first oxidation process of the ThBz1 in the LiTFSI solution
was observed, thus the first onset potential was not accurate
but in the Bu NPF it is equal to 0.99 V. The second onset is
DPV ox 1 on
DPV ox 1 Epa
DPV ox 2 on
DPV ox 2 Epa
DPV ox 3 Epa
DPV red 1 on
DPV red 1 Epc
CV ox 1 on
CV ox 1 Epa
CV ox 2 on
CV ox 2 Epa
CV ox 3 on
0.97
—
1.06
1.15
1.44
1.55
2.20
−2.33
−2.74
0.25
—
0.89
—
1.37
−2.96
1.00
1.19
1.03
1.62
—
−2.87
−3.29
1.03
1.17
—
0.98
1.11
1.41
1.53
—
−2.50
−2.79
1.04
1.20
1.44
—
1.12
1.31
1.36
−2.83
−3.32
0.99
1.13
1.16
—
4
6
1.31
1.37
−2.91
well defined and equals 0.89 V. The first onset potential of the
—
−3.03
—
−3.02
ThBz2 equals 1.03 V in the Bu
LiTFSI solution.
4 6
NPF solution and 1.04 V in the
CV red on
Fig. 7 DPV voltammogram: oxidation processes of the ThBz1 (black)
and the ThBz2 (green) in a solution of Bu NPF (solid line) and LiTFSI
dashed line) in the propylene carbonate. The background is marked in
red.
4
6
Fig. 8 Differential pulse voltammograms of ThBz1 (black) and ThBz2
(
(green) in the Bu
4
NPF
6
solution in the propylene carbonate. The back-
ground is marked in red.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15714–15724 | 15719

View Article Online
Paper
Dalton Transactions
son with the background is essential. The flat oxidation peak
appears when the compounds are reduced in the LiTFSI
solution.
In order to evaluate the behaviour of the new thiophene-
based electrolyte, a preliminary electrochemical study was per-
formed for the battery system of a CR 2025 coin cell LiCoO
2
/
Li Ti 12 containing ThBz1 or ThBz2 as the electrolyte, with
4
5
O
+
equimolar amounts of lithium salt (LiTFSI) to ensure Li
transport.
First, electrochemical impedance spectroscopy experi-
ments were performed on the tested batteries. As can be
seen in Fig. 12a, the Nyquist representation revealed only
Fig. 9 Differential pulse voltammograms of the ThBz1 (black) and
ThBz2 (green) in the solution of LiTFSI in the propylene carbonate. The
background is marked in red.
Fig. 10 Cyclic voltammograms of the ThBz1 (black) and the ThBz2
(
green) in the solution of Bu
4 6
NPF (solid line) and LiTFSI (dashed line) in
the propylene carbonate.
Fig. 11 Cyclic voltammograms of ThBz1 (black) and ThBz2 (green) in
the solution of Bu NPF (solid line) and LiTFSI (dashed line) in the propy-
4
6
lene carbonate. The background is marked in red.
The cycling voltammetry also reveals the reduction process
of the ThBz1 as well as ThBz2, which is shown in Fig. 11.
Fig. 12 Nyquist plots for batteries containing ThBz1 and ThBz2 as elec-
trolytes (a), and the real Z’ and the imaginary −Z’’ of impedance versus
These reduction processes start at about −3 V, thus a compari- frequency for ThBz1 (b) and ThBz2 (c).
15720 | Dalton Trans., 2018, 47, 15714–15724
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
one half circuit, which does not allow us to draw any
specific conclusion regarding the specific elements of the
architecture. It is certain that the constructed system is con-
Experimental
Materials
ductive due to lithium diffusion across the ThBz1 and Inorganic and organic compounds were provided by Sigma-
ThBz2 structure. Moreover, the Z′ and −Z″ versus frequency Aldrich and Fisher Scientifics and used as received.
for ThBz1 and ThBz2 (Fig. 12b and c, respectively) demon-
strate the instability of the values for the frequency range of thesized according to the procedure reported by Percec et al.
Potassium 3,4,5-tris(n-dodecan-1-yloxy)benzoate was syn-
2
7
0
.1–1000 Hz.
(Bz1) and potassium 3,4,5-tris[4-(n-dodecan-1-yloxy)bezyloxy]
2
8
The electrochemical performances of batteries with benzoate (Bz2) were synthesized as previously reported.
ThBz1 and ThBz2 charged to 1.0 V were investigated. The full
performance cycle of charging–discharging at 0.1C and cut-
off 0.5–1.0 V was difficult to correctly register for both battery
types due to the fluctuation of voltage. Maintaining a con-
stant 0.1C of discharge without cut-off values, the fluctuation
appeared within the 0–5 V range for both ThBz1 and ThBz2
as is presented in Fig. 13a for the only linear part of the
curve in the range of 4.0–5.2 V. We intended to roughly esti-
mate the value of the capacity of the battery for each
maximum of voltage and the results are presented in
Fig. 13b. It is evident that the process of discharge is sud-
denly interrupted by some other process, i.e. the formation
of the solid electrolyte interphase (SEI) and the solid per-
meable interface (SPI).
Potassium 3,4,5-tris(n-dodecan-1-yloxy)benzoate (Bz1)
1
3
H NMR (400 MHz, CDCl ) δ ppm: 7.43 (s, 2H Ar–H–COOH),
7
.34 (d, 4H, –O–Ar–H–CH –O–Ar–COOH from the 2 and 6 posi-
2
tions of the external benzylic units), 7.26 (d, 2H, –O–Ar–H–
CH –O–Ar–COOH from the 2 and 6 positions of the internal
2
benzylic units), 6.89 (d, 4H, –O–Ar–H–CH –O–Ar–COOH from
2
the 3 and 5 positions of the external benzylic units), 6.76 (d,
2
2H, –O–Ar–H–CH –OAr–COOH from the 3 and 5 positions of
the central benzylic units), 5.06 (s, 4H, –CH –O–Ar–COOH
2
from the 3 and 5 positions of the benzoic group), 5.04 (s, 2H,
2
–CH –O–Ar–COOH from the para position of the benzoic
group), 3.94 (two overlapped t, 6H, –CH –O–Ar), 1.79 (m, 6H,
2
–
CH –CH –O–Ar), 1.45 (m, 6H, –CH (CH ) –O–Ar), 1.27 (m,
2 2 2 2 2
4
8H, –(CH
2
)
8
–), 0.88 (t, 9H, –CH
3
).
1
3
C NMR (100 MHz, CDCl ) δ ppm: 171.8 (–COOH), 159.1
3
(
ArC–O–(CH ) –CH in the lateral benzylic unit), 159.1 (ArC–
2
10
3
O–(CH
2
)
10–CH
3
in the central benzylic unit), 152.8 (ArC meta
to –COOH), 143.2 (ArC para to –COOH), 130.5 (ArC meta to –O–
CH ) –CH in the lateral benzylic units), 129.4 (ArC meta to
(
2
11
3
ArC–O–(CH
2 3
)11–CH in the central benzylic unit), 128.6 (ArC–
CH –O–Ar–COOH), 124.1 (ArC–COOH), 114.6 (ArC ortho to –O–
2
(
CH ) –CH in the lateral benzylic units), 114.3 (ArC ortho to
2
11
3
–O–(CH
2
3
)11–CH in the central benzylic units), 109.7 (ArC ortho
to COOH), 74.8 (–CH –O–Ar–COOH in the central benzylic
2
unit), 71.1 (–CH –O–Ar–COOH in the lateral benzylic unit),
2
2 2 3
68.2 (–CH –O–(CH )10–CH in the lateral benzylic unit), 68.1
(
–CH –O–(CH ) –CH in the central benzylic unit), 32.4
2
2 10
3
(
–CH –CH –CH ), 29.8––29.4 (–(CH ) –), 26.2 (–CH –(CH ) –
2 2 3 2 6 2 2 8
3 2 3 2 3
CH ), 22.8 (–CH –CH ), 14.3 (–CH CH ).
Potassium 3,4,5-tris[4-(n-dodecan-1-yloxy)bezyloxy]benzoate
Bz2)
(
1
H NMR (400 MHz, CDCl ) δ ppm: 7.26 (s, 2H, Ar–H–COOK),
3
2 3
7.23 (d, 4H, –O–Ar–H–CH –O–Ar–H–COOCH from the 2 and 6
positions of the external benzylic units), 7.15 (d, 2H, –O–Ar–H–
CH –O–Ar–COOCH from the 2 and 6 positions of the internal
2
3
benzylic units), 6.79 (d, 4H, –O–Ar–H–CH
from 3 and 5 positions of external benzylic units), 6.65 (d, 2H,
O–Ar–H–CH –O–Ar–COOCH from 3 and 5 positions of
2 3
–O–Ar–COOCH
–
2
3
central benzylic units), 4.94 (s, 4H, –CH
the 3 and 5 positions of the benzoic group), 4.90 (s, 2H, –CH
O–Ar–COOCH3 from para position of the benzoic group), δ
ppm) = 3.94 (two overlapped t, 6H, –CH –O–Ar), 3.78 (s, 3H,
COOCH ), 1.68 (m, 6H, –CH –CH –O–Ar), 1.36 (m, 6H, –CH
(CH ) –O–Ar), 1.16 (m, 48H, –(CH ) –), 0.78 (t, 9H, –CH ).
2 3
–O–Ar–COOCH from
2
–
(
–
2
Fig. 13 Correlation of capacity versus voltage for the linear range (a)
and the estimated capacity for the performed cycles (b).
3
2
2
2
–
2
2
2 8
3
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15714–15724 | 15721

View Article Online
Paper
Dalton Transactions
1
3
C NMR (100 MHz, CDCl
3
) δ ppm: 173.9 (–COOK), 159.2 129.2, 128.7, 114.6, 114.3, 9.2 (–(OvC)–C aromatic), 74.8, 68.1
(
ArC–O–(CH ) –CH in the lateral benzylic units), 159.1 (ArC– (Ar–O–CH – aliphatic), 58.7, 32.0, 29.57, 26.2, 24.0, 18.4, 14.0.
2
10
3
2
FT-IR shift cm− : 2920 (ν as CH
1
), 2851 (ν CH2), 1722
2 s
O–(CH
2
)
10–CH
3
in central benzylic unit), 152.8 (ArC meta to
3 3
–COOCH ), 142.5 (ArC para to –COOCH ), 130.4 (ArC meta to (ν CvO), 1683 (ν Ph), 1611 (ν Ph), 1585(ν Ph), 1512 (ν Ph),
–
O–(CH ) –CH in the lateral benzylic units), 129.4 (ArC meta 1466 (ν Ph), 1419 (ν Ph), 1382 (δ CH ), 1370 (δ CH ), 1300
2
11
3
2
2
to –O–(CH
CH
2
)
11–CH
3
in the central benzylic units), 128.7 (ArC– (ω CH
3 r r r r
), 125.1 (ArC–COOCH ), 114.6 (ArC ortho to 1029 (δ CH /ν CO), 986 (γ CH ), 929 (γ CH ), 820 (γ CH ), 720
r
), 1242 (ν (CvO)–O), 1173 (δ CH
r
), 1104(δ CH/ν COC),
2
–O–Ar–COOCH
3
(
–O–(CH ) –CH in the lateral benzylic units), 114.2 (ArC ortho (ν C–S), 588 (δ Ph); ν – stretching vibrations, δ – in-plane defor-
2
11
3
to (–O–(CH
ortho to –COOCH
benzylic unit), 71.2 (–CH –O–Ar–COOCH in the lateral
benzylic units), 68.2 (–CH
2
)
11–CH
3
in the central benzylic unit), 109.2 (ArC mation, γ – out-of-plane deformation, ω – wagging, ρ – rocking,
–O–Ar–COOCH in the central Ph – phenyl, CH –CH group in phenyl ring.
All compounds were characterized with H NMR and
in the lateral benzylic NMR spectroscopy, using deuterated chloroform (CDCl ) as a
in the central benzylic units), solvent with a Jeol ECZ-400 S spectrometer ( H – 400 MHz,
3
), 74.8 (–CH
2
3
r
1
13
C
2
3
2
–(CH
2
)
10–CH
3
3
1
units), 68.1 (–CH –(CH )10–CH
2
2 3
1
3
5
2
2.4 (–COOCH ), 32.1 (–CH –CH –CH ), 29.4–29.9 (–(CH ) –), tetramethylsilane; C – 100 MHz, tetramethylsilane) at room
3
2
2
3
2 6
1
3
6.2 (–CH
2
–(CH
2
)
8
–CH
3
), 22.9 (–CH
2
–CH
3
), 14.3 (–CH
3
).
temperature with proton noise decoupling for C NMR and
1
using a pulse delay time of 5 s for H NMR. Measurements
Synthesis of thiophene-3-yl esters
were carried out at room temperature on 10–15% (w/v) sample
In a round bottom flask, under an inert gas atmosphere, solutions. Heteronuclear multiple quantum correlation
-chlorothiophene (0.001 mol) was dissolved in 6 mL of anhy- (HMQC) experiments were performed according to standard
drous tetrahydrofuran. Suitable amounts of Bz1 or Bz2 procedures.
0.0012 mol) and tetrabutylammonium bromide (0.0012 mol) Texture observations were performed by using a Nikon
were added. The reactions were carried out at 60 °C for 8 days. Eclipse polarizing microscope equipped with Fine
3
(
a
The mixture was poured into a mixture of ice and water. The Instruments WTMS-14C heating stage. Measurements were
resulting precipitate was filtered and dried. The precipitate done with crossed polarizers (90°) during heating, starting
was then purified on a Soxhlet apparatus using water as the from room temperature up to a completely dark texture and
solvent.
subsequently cooling down to −10 °C, with a heating/cooling
rate equal to 10 K min− . The samples were smeared on a
microscope slide and covered with a second slide to form a
1
Thiophen-3-yl 3,4,5-tris(n-dodecan-1-yloxy)benzoate (ThBz1)
1
H NMR (400 MHz, CDCl ) δ ppm: 7.34 (3 H, thiophene and thin layer between two glass slides.
3
aromatic), 7.29 (1H, thiophene), 6.19 (1H, thiophene), 4.00
The differential scanning calorimetry measurements were
–), 1.44 made using the DSC 92 Setaram differential scanning calori-
6H, Ar–O–CH –CH –CH –), 1.26 (48 H, –CH – aliphatic), 0.88 meter. The samples were weighed successively in platinum
(
(
(
2 2 2 2 2
6H, Ar–O–CH –CH –), 1.78 (6H, Ar–O–CH –CH –CH
2
2
2
2
9 H, CH
3
aliphatic).
capsules before placing in a calorimeter. The measurements
) δ ppm: 165.3 (–O–(CvO)–Ar), were carried out in an air atmosphere. At the beginning the
1
3
C NMR (100 MHz, CDCl
3
1
1
7
2
53.0 and 152.7 (Caromatic–O–Aliph), 143.2, 141.8 130.11, samples were cooled to −100 °C, then heated to 200 °C, cooled
29.22, 128.12, 123.4, 108.5 and 108.4 (–(OvC)–C aromatic), again to −100 °C and heated to 300 °C. The heating and
1
3.6 and 73.5 (Ar–O–CH
2
– aliphatic), 69.3, 63.1, 32.0, 29.7, cooling rates were 10 K min− , and the cooling medium was
6.2, 22.8, 14.2.
liquid nitrogen.
FT-IR shift cm− : 2918 (ν as CH
1
), 2849 (ν
CH
), 1725
TGA was carried out with a Mettler TGA/SDTA 851e thermo-
2
s
2
(
(
(
ν CvO), 1683 (ν Ph), 1586(ν Ph), 1502 (ν Ph), 1466(ν Ph), 1429 balance. Cured samples with an approximate weight of 10 mg
ν Ph), 1378 (δ CH ), 1332 (ω CH ), 1223 (ν (CvO)–O), 1190 were degraded between 30 and 800 °C at a heating rate of 10 K
2
2
), 929 (γ min− in nitrogen (100 mL min ) and measured under
1
−1
δ CH
), 862 (ρ CH
δ Ph); ν– stretching vibrations, δ – in-plane deformation, γ –
out-of-plane deformation, ω – wagging, ρ – rocking, Ph – powders were obtained at room temperature with an FTIR
r
), 1151(δ CH/ν COC), 1117 (ν COC), 982 (γ CH
), 759 (γ CH ), 720 (ρ CH /skel. CH /ν C–S), 545 normal conditions.
Fourier transform infrared (FT-IR) spectra of polymer
r
CH
(
r
2
r
r
2
phenyl, CH
r
–CH group in phenyl ring.
spectrophotometer (Thermo Scientific Nicolet 6700) with a
resolution of 8 cm− and 64 scans per spectra, in transmit-
tance mode. An attenuated total reflection (ATR) accessory
with thermal control and a DTGS TEC detector was used to
1
Thiophene-3-yl 3,4,5-tris[4-(n-dodecan-1-yloxy)bezyloxy]
benzoate (ThBz2)
1
H NMR (400 MHz, CDCl ) δ ppm: 7.53 (2H, aromatic), obtain FT-IR spectra.
3
6
.82–7.30 (15 H, aromatic and thiophene), 5.00 (6H, ArO–CH
2
2
–
–
Electrochemical measurements were carried out in the pro-
pylene carbonate (Sigma-Aldrich, anhydrous, 99.7%) solutions
Ar), 4.00 (6H, Ar–O–CH –CH –), 1.75 (6H, Ar–O–CH –CH
2
2
2
CH –), 1.24 (54H, –CH – aliphatic), 0.87 (9H, CH aliphatic).
and
) δ ppm: 165.82 (–O–(CvO)–Ar), (Bu
sulfonimide (LiTFSI, Sigma-Aldrich, 99.95%) as the supporting
Ar, meta), 141.0 (Caromatic–O–CH –Ar, para), 130.2, 130.0, 129.4, electrolyte. The supporting electrolyte concentration equaled
with
tetrabutylammonium
hexafluorophosphate
2
2
3
1
3
C NMR (100 MHz, CDCl
3
4
NPF , TCI, min. 99.0%) or lithium bis(trifluoromethane)
6
1
2
59.2 and 159.1 (Caromatic–O–Aliph), 152.48 (Caromatic–O–CH –
2
15722 | Dalton Trans., 2018, 47, 15714–15724
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
to 0.1 mol L− and was 100 times greater than the analyte con- lytes can be used as components of lithium-ion batteries. The
centration. Before each experiment, argon 5.0 gas was passed electrochemical performance revealed some distortion during
through the solution for 30 seconds (time determined empiri- the discharge of the battery, probably related to SEI and/or SPI
cally) for oxygen removal. The three-electrode system and one- formation, which requires further investigation. Moreover,
compartment cell were applied. A commercially available further studies without solvent vs. lithium metal reference
1
2
mm diameter glassy carbon disc electrode was used as the electrodes are required to check the stability of the ThBz com-
working electrode, a coiled platinum wire with surface 20 pounds and possible multiple cycle experiments in the
times greater than the working electrode’s surface was the lithium-ion battery arrangement should be carried out.
counter electrode and a silver wire was used as a quasi-refer-
ence electrode. All potentials are related to the ferrocene refer-
0
+
ence redox couple Fc /Fc , which, in the case of cyclic voltam- Conflicts of interest
metry, means the average of its anodic and cathodic peak
1
Authors declare no conflicts of interests.
potential (Epa + Epc), and the maximum of the oxidation
2
potential in the case of differential pulse voltammetry. Onset
potentials were approximated from extrapolation of the almost
linear part of the peak’s slope and the baseline. Also, peak
potentials are given due to the non-ideal shape of some peaks
and difficulties with straight fitting. Cyclic voltammetry and
the differential pulse voltammetry were used to examine the
oxidation and reduction ability of ThBz compounds. Fixed
parameters were used in these methods: the scan rate was 0.1
Acknowledgements
This work is financially supported by the Polish Ministry of
National Defence (585/2016/DA) within the 1 edition of
Kościuszko’s program for Polish researchers returning to
Poland.
st
−1
−1
V s in the CV and 0.01 V s in DPV, and the step potential
was 0.005 V, modulation amplitude 0.025 V, modulation time
References
0
.05 s, interval time 0.5 s. All potentials are given with an accu-
racy of ±0.01 V if not stated otherwise. All measurements were
made with the Autolab PGSTAT 128 N series instrument pro-
duced by Eco Chemie B.V.
1 S. M. Ismael, S. H. E. Abdel Aleem, A. Y. Abdelaziz and
A. F. Zobaa, Renewable Energy, 2019, 130, 1002.
2 L. Reichenberg, F. Hedenus, M. Odenberger and
F. Johnsson, Renewable Energy, 2018, 129, 334.
3 P. Nabielek, H. Dumke and K. Weninger, Energy Sustainable
Soc., 2018, 8, 5.
The positive electrode consisted of LiCoO
2
(93.5 wt%),
carbon black (4 wt%) and poly(vinylidene fluoride)(PVDF)
(
2.25 wt%), and the negative electrode consisted of Li
4 5 12
Ti O
(
80 wt%), carbon black (10 wt%) and PVDF (10 wt%). The
4 D. Burmester, R. Rayudu, W. Seah and D. Akinyele,
Renewable Sustainable Energy Rev., 2017, 67, 760.
5 J. Goodenough and M. H. Braga, Dalton Trans., 2018, 47,
645.
battery testing was performed on a CR 2025 coin cell LiCoO2/
Li Ti 12. The separator and electrolyte were the synthesized
compounds ThBz1 or ThBz2 mixed with LiTFSI (1 : 1 mol : mol)
4
5
O
and dissolved in CHCl to form a 30% solution. The electrolyte
was deposited on both electrodes and chloroform was evapor-
6 A. M. Haregewoin, A. S. Wotango and B.-J. Hwang, Energy
Environ. Sci., 2016, 9, 1955.
3
ated in a glove box in an argon atmosphere (1 mL for
7 W. Luo, Y. Wang, L. Wang, W. Jiang, S.-L. Chou, S. X. Dou,
H. K. Liu and J. Yang, ACS Nano, 2016, 10(11), 10524.
8 B. Y. Guan, L. Yu, J. Li and X. W. Lou, Sci. Adv., 2016, 2,
e1501554.
2
1
2.5 cm ). Loading of the anode and cathode were 15 mg.
Battery charge and discharge were registered using a compu-
ter-controlled battery analyser BST8-3 (MTI Corp., USA).
9
S. Wang, B. Y. Guan, L. Yu and X. W. Lou, Adv. Mater.,
017, 29(37), 1702724.
2
1
0 M. Yu, J. Ma, H. Song, A. Wang, F. Tian, Y. Wang, H. Qiua
and R. Wang, Energy Environ. Sci., 2016, 9, 1495–1503.
Conclusions
Both ThBz1 and ThBz2 have similar structural units like the 11 E. Lim, H. Shim, S. Fleischmann and V. Presser, J. Mater.
thiophene ring and gallic acid core. These compounds have Chem. A, 2018, 6, 9480.
comparable oxidation and reduction potentials and thus prob- 12 Y. Zheng, J. Liu, Y.-P. Liao, G. Ungar and P. V. Wright,
ably undergo similar reactions, regardless of the substituents. Dalton Trans., 2004, 3053.
It is possible that both compounds will react with metallic 13 P. G. Bruce, Dalton Trans., 2006, 1365.
lithium when measurements vs. lithium reference are con- 14 T. Sakamoto, T. Ogawa, H. Nada, K. Nakatsuji, M. Mitani,
ducted, but the nature of these reactions is not yet known. The
examined compounds have electrochemical windows about 4
V wide.
B. Soberats, K. Kawata, M. Yoshio, H. Tomioka, T. Sasaki,
M. Kimura, M. Henmi and T. Kato, Adv. Sci., 2018, 5,
1700405.
Both investigated compounds are thermally stable in a 15 A. Yamashita, M. Yoshio, S. Shimizu, T. Ichikawa, H. Ohno
nitrogen atmosphere and exhibit liquid crystalline properties,
as expected. Preliminary studies show that synthesized electro-
and T. Kato, J. Polym. Sci., Part A: Polym. Chem., 2015, 53,
366.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15714–15724 | 15723

View Article Online
Paper
Dalton Transactions
1
1
6 B. Soberats, E. Uchida, M. Yoshio, J. Kagimoto, H. Ohno 22 J. C. Ronda, J. A. Reina and M. Giamberini, J. Polym. Sci.,
and T. Kato, J. Am. Chem. Soc., 2014, 137(27), 9552. Part A: Polym. Chem., 2004, 42, 326.
7 K. A. Bogdanowicz, S. V. Bhosale, Y. Li, I. F. J. Vankelecom, 23 J. M. Saveant, Acc. Chem. Res., 1993, 26(9), 455.
R. Garcia-Valls, J. A. Reina and M. Giamberini, J. Membr. 24 N. E. Widjonarko, Coatings, 2016, 6(4), 54.
Sci., 2016, 509, 10.
8 K. A. Bogdanowicz, D. Pirone, J. Prats-Reig, V. Ambrogi,
J. A. Reina and M. Giamberini, Polymers, 2018, 10, 416.
25 C. O. Laoire, E. Plichta, M. Hendrickson, S. Mukerjee and
K. M. Abraham, Electrochim. Acta, 2009, 54, 6560–
6564.
1
1
2
2
9 K. A. Bogdanowicz, P. Sistat, J. A. Reina and M. Giamberini, 26 N. Mozhzhukhina and E. J. Calvo, J. Electrochem. Soc., 2017,
Polymer, 2016, 92, 58. 164, A2295.
0 T. Onuma, E. Hosono, M. Takenouchi, J. Sakuda, S. Kajiyama, 27 V. Percec, J. A. Heck, D. s. Tomazos and G. Ungar, J. Chem.
M. Yoshio and T. i. Kato, ACS Omega, 2018, 3, 159. Soc., Perkin Trans. 2, 1993, 2381.
1 J. C. Ronda, J. A. Reina, V. Cadiz, M. Giamberini and 28 S. V. Bhosale, M. A. Rasool, J. A. Reina and M. Giamberini,
L. Nicolais, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 2918. Polym. Eng. Sci., 2012, 53, 159.
15724 | Dalton Trans., 2018, 47, 15714–15724
This journal is © The Royal Society of Chemistry 2018