Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Communication
Journal of Materials Chemistry A
decreased the solubility and induced precipitation of this
compound. Thus, the presence of moisture in the solvent cau-
ses no signicant effects on the electrochemical behavior,
which consequently eliminates requirements for using costly
anhydrous solvents.
Diffusion coefficients were estimated for the electrochemi-
cally stable substances M2–M4, M6, and M7 using the rotating
disk electrode (RDE) measurements.
The resulting linear sweep voltammetry curves have exhibi-
ted two types of behaviors: simple one-electron reduction for
compounds M2 (Fig. S5a and b, ESI†), M3 (Fig. S5c and d, ESI†)
and M4 (Fig. S5e and f, ESI†), and simple two-electron reduction
for M6 (Fig. S5g and h, ESI†) and M7 (Fig. S5i and j, ESI†). For
the purpose of comparison, we measured under the same
conditions, the diffusion coefficient of TEMPOL+ (D ¼ 5.93 ꢃ
10ꢀ6 cm2 sꢀ1) as a reference compound (Fig. S5k and l, ESI†). It
was found that the diffusion of M2, M3, and M4 is faster as
compared to that of the bulkier M6 and M7, whereas all studied
arylamines showed comparable or higher diffusion coefficients
than TEMPOL+ (Table 1).
Fig. 1 Continuous voltammetric cycling of 3 mM (a) M1 and (b) M3 in
0.1 M TBABF4/MeCN solution.
During the next stage of this study, we investigated triaryl-
amine-based catholytes in the assembled laboratory RFB cells
(Fig. 2a, see the ESI† for specication of the cell parameters).
Butylviologen diperchlorate (V1) was selected as the redox-active
anolyte component (Fig. S2, ESI†). The reduction potential of V1
was found to be ꢀ0.75 V vs. Ag/AgNO3 (Fig. S4h, ESI†), leading to
a potential difference of 0.89–1.3 V vs. triarylamines (Fig. 2b).
In order to reveal the optimal electrolyte composition, we
explored systematically, the series of RFB devices using the M3/
V1 redox couple and various supporting salts: tetrabuty-
lammonium tetrauoroborate (TBABF4), tetrabutylammonium
hexauorophosphate (TBAPF6), NaClO4, LiClO4, and LiPF6.
Fig. 2c summarizes the observed charge–discharge cycling
behaviour of all the studied systems, while all the cycling
proles are presented in Fig. S8 (ESI†). It is clear that the
background electrolyte salt has a strong impact on the opera-
tional stability of RFBs. Using LiPF6 leads to the fastest capacity
decay, probably due to the chemical activity of this salt.23–25 To
evaluate the processes responsible for capacity decay, we
measured the pH values of all electrolyte solutions (Table S2,
ESI†). A decrease of the pH value of the LiPF6-based electrolyte
was observed, whiꢀch is apparently associated with the hydro-
lysis of Li+ or PF6 ions induced by the presence of residual
amounts of water. The generated acids could interact with tri-
arylamines, yielding quaternary ammonium bases, leading to
the fast capacity decay of the RFBs. In contrast, the TBABF4 and
NaClO4 based electrolytes are more sustainable towards
hydrolysis and their utilization in RFBs leads to the stable
cycling behavior. Therefore, these two salts were selected for the
further systematic study of different arylamine-based
catholytes.
Fig. 2 (a) Schematic diagram of the assembled RFB device; (b) CV
curves for a mixture containing V1 (10 mM) as the anolyte and M3 (10
mM) as the catholyte in 0.1 M TBABF4/MeCN at a scan rate of 200 mV
s
ꢀ1; (c) dependence of the discharge capacity of the cycle number for
10 mM redox-flow cells containing M3 and V1 during 50 cycles for
various background electrolytes.
capacities were delivered by the compounds based on triphe-
nylamines M3 and M4 as compared to their tetraphe-
nylbenzene-1,4-diamine analogs M6 and M7. The relatively
low charge and discharge capacities of RFBs based on M6/V1
and M7/V1 redox couples (Fig. S10, ESI†) might be explained
by the hindered transfer of the second electron, resulting in
the limitation of the practical specic capacity to ꢁ50% of its
theoretical value. Thus, the compounds based on triphenyl-
amine (M3, M4) can be considered as more promising since
they exhibit higher capacities due to the simplicity of single-
electron transfer and good stability of the generated radical
cation provided by the appropriate functionalization of the
TPA core.
To assess the charge–discharge cycling stability of RFBs
using compounds M3, M4, M6, and M7 as catholytes, another
series of the experiments was carried out (Fig. S8–S10, ESI†).
Fig. 3 shows the maximal obtained discharge capacities,
discharge capacities aer 50 cycles, and maximal coulombic
efficiencies (CEs) observed in these experiments. Higher
Comparing the effects of the supporting electrolyte salts on
the RFB performance, it can be concluded that the application
of TBABF4 results in higher capacities but lower CE as compared
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. A, 2021, 9, 8303–8307 | 8305