You are on page 1of 7

Journal of Energy Storage 18 (2018) 414–420

Contents lists available at ScienceDirect

Journal of Energy Storage


journal homepage: www.elsevier.com/locate/est

Effect of graphene oxide with different oxygenated groups on the high-rate T


partial-state-of-charge performance of lead-acid batteries

Weijia Caia, Kai Qia, Zhenyu Chena,c, Xingpeng Guoa,c,d, Yubing Qiua,b,
a
School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan, 430074, PR China
b
Key Laboratory of Material Chemistry for Energy Conversion and Storage (Huazhong University of Science and Technology), Ministry of Education, PR China
c
Hubei Key Laboratory of Materials Chemistry and Service Failure, Wuhan, 430074, PR China
d
School of Chemistry and Chemical Engineering, Guangzhou University, Guangzhou, 510006, PR China

A R T I C LE I N FO A B S T R A C T

Keywords: Four graphene oxide samples with different oxygenated groups are prepared and characterized, and then em-
Lead-acid battery ployed as the additives in the negative active materials to investigate the effect of different oxygenated groups of
High-rate partial-state-of-charge graphene oxide on the H2 evolution performance of negative plates and the high-rate partial-state-of-charge
Graphene oxide cycle life of simulated lead-acid batteries. The results indicate that the H2 evolution and the reduction of PbSO4
Oxygenated groups
processes are largely accelerated by the graphene oxide additives. An increase in CeO groups (CeOH and
Hydrogen evolution performance
CeOeC) of graphene oxide can largely promote the H2 evolution process. The addition of the graphene oxide
additives largely increases the surface area and total pore volume of the negative plates, meanwhile obviously
increases the hydrophilicity of negative active materials to facilitate the diffusion of acid into the inner of the
plate. The high-rate partial-state-of-charge cycle life of the simulated test cells containing the graphene oxide
samples is prolonged significantly, and especially those containing graphene oxide with less CeO groups and
more carbonyl and carboxyl groups have the longest cycle life, which seems to be more appropriate as the
additive of negative plates.

1. Introduction PbO/CB composites displayed the similar effect [24]. Different surface
functional groups of carbon materials may play an important role in
Working under conditions of high-rate partial-state-of-charge their H2 inhibition performance.
(HRPSoC) continuously, lead-acid batteries will suffer quick battery Based on our previous work [25], adding graphene oxide (GO) in
failures owing to the irreversible sulfation of negative plates [1,2]. In negative plates can significantly prolong the HRPSoC cycle life of lead-
recent years, various carbon materials, including carbon black (CB) acid batteries, however, also promote the H2 evolution in the negative
[3–5], activated carbon (AC) [6–10], graphite [11,12], carbon nano- plate obviously. According to the effect of functional groups of AC and
tube [13–16], graphene [17–19], etc., were directly introduced into the CB, different oxygenated groups of GO may produce a different effect
negative active materials (NAM) to prevent the progressive sulfation of on the H2 evolution process. But few studies have focused on this. In
negative plates and significantly prolong the HRPSoC cycle life of lead- this paper, GO with different oxygenated groups were prepared and
acid batteries [20]. However, these lead-carbon batteries suffer serious characterized, and then their influence on the H2 evolution ability of
hydrogen evolution to dry out their electrolyte quickly [21]. negative plates and on the HRPSoC cycle life of simulated test cells was
In order to overcome this H2 evolution problem, the carbon mate- investigated to identify the detailed effect of different oxygenated
rials were modified with various methods [6,7,22], e.g. the modifica- groups of GO. The results of this work may benefit the selection of
tion of their surface functional groups [8], the development of metals appropriate GO for the further modification of its functional groups or
(or metal oxides)/carbon materials composites [9,10,23] and the the preparation of metals (metal oxides)/GO composites for lead-acid
combination of these two methods [24]. It was reported that acidic batteries.
surface functional groups of AC could promote the H2 evolution in
negative plates, while those alkaline surface functional groups could
inhibit it [8]. We also found that the surface functional groups of CB in


Corresponding author at: School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan, 430074, PR China.
E-mail addresses: caiwj_yuke@163.com (W. Cai), qikai@hust.edu.cn (K. Qi), chenzhenyu@mail.hust.edu.cn (Z. Chen), guoxp@mail.hust.edu.cn (X. Guo),
qiuyubin@mail.hust.edu.cn (Y. Qiu).

https://doi.org/10.1016/j.est.2018.06.004
Received 23 March 2018; Received in revised form 7 June 2018; Accepted 7 June 2018
2352-152X/ © 2018 Elsevier Ltd. All rights reserved.
W. Cai et al. Journal of Energy Storage 18 (2018) 414–420

Table 1 respectively, and those without GO are blank plates. A prepared ne-
Content of the materials used in negative pastes (wt. %, versus the lead oxide). gative plate (1.0 × 1.5 cm2) and two commercial positive plates
Acetylene Short BaSO4 Humic H2SO4* H2O GO (2.8 × 2.8 cm2) were separated with a glass mat (AGM, thick-
black fibers Acids samples ness = 2 mm) separator, and then 65 mL H2SO4 (1.28 g cm−3) was
added to form a flooded simulated lead-acid test cell [25].
Blank 0.2 0.05 0.4 0.7 12.5 10 0
GO added 0.2 0.05 0.4 0.7 12.5 13 - 15 1.0

* Density = 1.25 g cm−3. 2.3. Electrochemical measurements and HRPSoC cycle tests

2. Experimental The cathodic polarization curves of the prepared negative plates


(Blank, GO-1, GO-2, GO-3 and GO-4) were measured in H2SO4
2.1. Preparation of GO (1.28 g cm−3) at room temperature (25 ℃) using a GAMRY Reference
3000 electrochemical workstation. Two platinum electrodes and a Hg/
A water-enhanced oxidation method, which belongs to a modified Hg2SO4 electrode were used as the counter electrode and the reference
Hummer’s method, was employed to prepare GO samples with different electrode, respectively. The polarization potential range is from the
species of oxygenated groups [26,27]. In an ice-bath (< 10 ℃), 1.0 g open circuit potential (EOCP) to −1.5 V vs. Hg/Hg2SO4 and the scan rate
graphite powder (99.95%, supplied by Qingdao Huatai, China) and is 1.0 mV s−1.
3.0 g KMnO4 were added in 46 mL concentrated H2SO4 with the dif- Before the HRPSoC cycle test, the 2C current was determined from
0 the measured Peukert plots for the different kinds of simulated test cells
ferent initial addition of distilled water (V H 2O
= 0, 4 and 12 mL) and
were stirred continuously, then maintained at 40 ℃ for 2 h. According and the state-of-charge (SoC) of the test cells was discharged to 50%.
0
to the different V H 2O
(0, 4 and 12 mL), the obtained GO is denoted as The schedule of the HRPSoC cycle test was: charge (2C, 30 s); rest
GO1-M, GO2-M, GO3-M, respectively. The fourth GO sample was syn- (10 s); discharge (2C, 30 s), rest (10 s). The limit of end-of-discharge
thesized in the system without the initial addition of distilled water voltage (Vdischarge) and end-of-charge voltage (Vcharge) was 1.60 V and
0
(V H 2O
= 0). But after the oxidation at 40 ℃ for 2 h, 100 mL distilled 3.0 V, respectively. After the first cycle-set test (2C-1 set), the second
water was slowly added into the reaction system and maintained at (2C-2set) and the third cycle-set (2C-3set) test were conducted ac-
90 ℃ for 15 min. The obtained GO is denoted as GO4-H [28]. After cording to the same procedures. All these tests were performed at room
centrifuging, washing by distilled water, and vacuum drying, the final temperature (25 ℃) and repeated at least three times.
GO samples were prepared for the further test in this work. Except for
graphite powder, other reagents and chemicals used in this preparation
step were analytical grade (AR). 2.4. Physical characterization

The as-prepared samples of GO1-M, GO2-M, GO3-M, and GO4-H


2.2. Preparation of negative plates and simulated test cells were characterized using X-ray diffraction (XRD, XRD-7000S, Japan)
with Cu Kα radiation, Fourier transform infrared spectroscopy (FTIR,
Lead oxide power (oxidation degree = 82.6%) was supplied by a Vertex 70, Germany) in reflectance mode, and X-ray photoelectron
commercial battery company. The content (versus the lead oxide) of spectroscopy (XPS, VG Multilab 2000) analyses (the binding energies
other materials in negative pastes is listed in Table 1, in which humic were calibrated relative to the C 1 s peak from GO sample at 284.6 eV).
acids (≥90%, Aladdin) and BaSO4 (AR) are reagents, and acetylene The morphology of the negative plates was observed using scanning
black and short fibers are supplied from commercial battery companies. electron microscope (SEM, Nova NanoSEM 450, America). The specific
The apparent density of the negative pastes was about 4.2 g cm−3, surface and porosity analysis for the negative plates, which had been
which was adjusted by distilled water. The negative plate was prepared dried in vacuum at 120 ℃ for 2 h, was conducted using an ASAP2020 M
using a Pb-Sn-Ca grid (1.0 × 1.5 cm2) and 1.0 g negative paste, and analyzer (America). The water contact angle of the prepared negative
then cured carefully and formed under a commercial formation pro- plate was measured with a JC2000 contact-angle meter (Shanghai
cedure (H2SO4 density: 1.05 g cm−3; constant formation current: Zhongchen Digital Technic Apparatus, China).
8.5 mA g−1 and 17 mA·g−1; total charge delivered to NAM:
535.5 mAh g−1) [25]. The negative plates containing GO1-M, GO2-M,
GO3-M, and GO4-H are nominated as GO-1, GO-2, GO-3, and GO-4,

Fig. 1. (a) XRD patterns and (b) FTIR spectra of different GO samples.

415
W. Cai et al. Journal of Energy Storage 18 (2018) 414–420

Fig. 2. XPS high-resolution spectra of C 1 s for (a) GO1-M, (b) GO2-M, (c) GO3-M, (d) GO4-H.

Table 2 increase in the contents of −OH and CeOeC on GO sheets. The in-
Relative contents of C–O, carbonyl, and carboxyl groups as well as the oxidation tensity ratio of the C]O (1725 cm−1)/H2O (1620 cm−1) peaks in GO1-
degree of the GO samples. M, GO2-M, and GO3-M is approximate, but much smaller than that in
Samples C 1s GO4-H. These results imply that GO4-H may have a higher content of
carbonyl groups (C]O) [27]. Apparently, the oxidation degree and
IC–O / ICC IC=O / ICC IO–C=O / ICC SCO / SC-Total (%) functional groups of different GO samples change with different
synthesis conditions.
GO1-M 1.42 0.17 0.06 57.14
GO2-M 1.89 0.19 0.05 63.81 Fig. 2 gives the XPS high-resolution spectra of C 1 s for the GO
GO3-M 2.18 0.22 0.05 65.16 samples, which consist of four types of carbon bonds: CeC/C = C
GO4-H 1.33 0.20 0.08 55.38 (284.6 eV, Peak 1), CeO (286.6 ± 0.1 eV, Peak 2), C]O (287.8 eV,
Peak 3), and OeC = O (289.0 eV, Peak 4) [26,34,35]. Table 2 lists the
peak intensity ratios of oxygenated carbon atoms (CeO, C]O, and
3. Results OeC = O) to intact carbons (CeC and C]C), i.e. IC–O/ICC, IC=O/ICC,
and IO–C=O/ICC, which generally demonstrate the relative contents of
3.1. Characterization of GO samples oxygenated carbon atoms. Also, the peak area ratio of oxygenated
carbon atoms (SCO) to the total C 1 s (SC–Total) in Table 2 reflects the
Fig. 1a shows the XRD patterns of different GO samples, in which oxidation degree of GO [24,25]. Among GO1-M, GO2-M and GO3-M,
the diffraction peaks of GO1-M, GO2-M, GO3-M, and GO4-H are mea- the content of CeO groups increases obviously (IC–O/ICC rises from 1.42
sured at 10.08°, 10.38°, 10.43° and 11.03°, respectively, while that of to 2.18) to improve the oxidation degree of GO (SCO/SC–Total rises from
graphite (2θ = 26.6°) is completely disappeared [29–31]. These results 57.14% to 65.16%), indicating that the hydroxyl and epoxy groups
indicate that the graphite structure is disassembled and GO sheets are (−OH and CeOeC) increase obviously with an increase in the initial
formed. The FTIR spectra in Fig. 1b show typical absorption peaks of addition of water content. For GO4-H, the content of CeO groups and
GO, including 1725 cm−1 (C]O), 1620 cm–1 (−OH bending modes in the oxidation degree are all the lowest (IC–O/ICC = 1.33 and SCO/
water molecules), 1395 cm−1 (−OH bending vibration in CeOH), SC–Total = 55.38%), but it has relatively higher carbonyl (C]O) and
1226 cm−1 (CeO) and 1055 cm−1 (CeOeC) [27,32,33]. The OeH carboxyl (OeC = O) contents [27]. These results further prove the
stretching vibrations in the region of 3600–3300 cm−1 correspond to above speculation from the FTIR analysis in Fig. 1b.
the hydroxyl and carboxyl groups of GO and residual water between GO
sheets [26]. Among GO1-M, GO2-M, and GO3-M, the absorption peaks
of −OH and CeOeC increase with an increase in V H 0
2O
, suggesting the

416
W. Cai et al. Journal of Energy Storage 18 (2018) 414–420

were obtained from the results in Fig. 3 and the parallel experiments (at
least three times) to observe the effect of the GO additives [24,25],
including EH (H2 evolution potential), IEH (the current density of the
reduction of PbSO4 at EH ) and IH,-1.4V (the H2 evolution current density
at −1.4 V vs. Hg/Hg2SO4, which equals I-1.4V −IEH ). Table 3 lists the
average values and standard deviations of these characteristic para-
meters.
Compared with the characteristic parameters of the blank plate in
Table 3, all the EH values of the negative plates containing different GO
samples (GO-1∼ GO-4) become more positive, while their IEH and
IH,-1.4V values increase significantly. These results indicate that both the
H2 evolution and the reduction of PbSO4 processes in the negative
plates have been promoted obviously with the addition of the GO
samples [25]. From GO-1 to GO-3, their EH values become slightly more
positive while IEH increases to some extent, but IH,-1.4V increase largely.
Compared with GO-1, GO-4 has more positive EH and larger IEH and
IH,-1.4V values. Among the studied negative plates, the GO-4 plate con-
Fig. 3. Cathodic polarization curves of the negative plates (Blank, GO-1, GO-2, taining GO4-H has relatively larger IEH and smaller IH, −1.4V values,
GO-3, and GO-4) in H2SO4 solution (1.28 g cm−3). suggesting that GO4-H may be a more appropriate additive for the
negative plate. Apparently, the promotion effect of the four GO samples
on the hydrogen evolution and the reduction of PbSO4 is different.
Table 3
Characteristic parameters obtained from the results in Fig. 3 and the parallel Because the GO samples have different oxygenated groups (see
experiments. Table 2), the above results imply that different oxygenated groups may
play a different role in these promotion effect, which is discussed later.
Samples EH (V, vs. Hg/Hg2SO4) IEH (mA cm−2) IH,-1.4V (mA cm−2)

Blank −1.298 ± 0.005 0.67 ± 0.16 0.31 ± 0.21


GO-1 −1.269 ± 0.001 1.74 ± 0.29 1.79 ± 0.08 3.3. HRPSoC cycle life of the simulated test cells
GO-2 −1.249 ± 0.002 1.85 ± 0.29 3.57 ± 0.51
GO-3 −1.244 ± 0.005 2.27 ± 0.30 6.10 ± 0.72 The Vdischarge - cycle number curves of the simulated test cells with
GO-4 −1.253 ± 0.005 1.99 ± 0.21 2.77 ± 0.15
the negative plates containing different GO samples are shown in Fig. 4,
which were measured under the HRPSoC condition within 2C-3set.
According to the results in Fig. 4 and the parallel tests (at least three
times), the average HRPSoC cycle life (CL) of the test cells and its
standard deviation are obtained, as shown in Table 4 [24,25]. From
Fig. 4 and Table 4, it can be seen that the addition of the GO samples in
the negative plates can significantly increase the CL of the simulated
test cells and those containing GO4-H have the longest CL value. These
results are consistent with our previous work [24] and suggest that
GO4-H may be a more appropriate additive for the negative plate.

3.4. Microstructure of the negative plates

Fig. 5 shows the SEM morphologies of the NAM in the negative


plates (Blank, GO-1, GO-2, GO-3 and GO-4) freshly prepared and tested
after 2C-3set. For the freshly prepared negative plates, those containing
various GO samples display more porous surface than the blank plate,
especially the GO-4 plate has large irregular holes and lamellar struc-
ture, as shown in Fig. 5(AI–EI). The BET surface area (SBET) and total
Fig. 4. Vdischarge - cycle number curves under the HRPSoC condition for the pore volume (Vpore) of these freshly prepared negative plates are pre-
simulated test cells with different negative plates (Blank, GO-1, GO-2, GO-3, sented in Fig. 6. Before the cycle test, the negative plates containing
and GO-4) at the 2C discharge rate within the third cycle-set. various GO samples have much larger SBET and Vpore than those of the
blank plate, which demonstrate their change in microstructure with the
3.2. H2 evolution performance of the negative plate addition of GO additives. After the 2C-3set cycle test, the PbSO4 crystal
size on the surface of GO-1, GO-2, GO-3 and GO-4 plates becomes much
Fig. 3 shows the cathodic polarization curves of the negative plates smaller than that on the blank plate, as shown in Fig. 5(AII–EII), im-
(Blank, GO-1, GO-2, GO-3, and GO-4) in H2SO4 solution (1.28 g cm−3), plying that the GO additives prevent the growth of PbSO4 crystals on
which display the same features that have been discussed in our pre- the negative plates.
vious paper [24,25]. In this case, the same characteristic parameters

Table 4
HRPSoC cycle life of the test cells within the 2C-3set cycle test.
Discharge rate Blank GO-1 GO-2 GO-3 GO-4

2C 374 ± 110 13527 ± 1402 15954 ± 1877 11580 ± 4340 19672 ± 1387

417
W. Cai et al. Journal of Energy Storage 18 (2018) 414–420

Fig. 5. SEM images of the NAM in the negative plates before (I) and after the HRPSoC cycle test (II): (A) Blank; (B) GO-1; (C) GO-2; (D) GO-3, (E) GO-4. (The negative
plates were fully recharged after the 2C-3set test).

418
W. Cai et al. Journal of Energy Storage 18 (2018) 414–420

and GO (SGO) in the negative plates [36,37] and therefore, their surface
features and real reaction area (Sreal = SPb + SGO) determine its re-
action rate [24,25]. The SBET and Vpore of the negative plates containing
the GO samples increase obviously than those of the blank plate (see
Fig. 6), which can largely increase their Sreal [25]. In this case, the H2
evolution in these negative plates is accelerated (see Table 3) [24,25].
For the plates of GO-1, GO-2, and GO-3, there is no obvious dif-
ference among their SBET and Vpore values (see Fig. 6), suggesting that
their Sreal should be approximated with each other, but their IH, −1.4V
values increase obviously from GO-1 to GO-3 (see Table 3). Thus, it can
be speculated that the promotion of the H2 evolution process in these
plates should be mainly ascribed to the GO surface features, i.e., the
different oxygenated groups on GO. Because only the content of CeO
groups (CeOH and CeOeC) increases obviously from GO1-M to GO3-M
(see Table 2), it can be deduced that an increase in the CeO groups on
the GO samples can promote the H2 evolution.
The GO-4 plate has the largest SBET and Vpore among the four plates
Fig. 6. Specific surface and pore analysis results for the negative plates (Blank, containing the GO additives, but its IH, −1.4V increase slightly than that
GO-1, GO-2, GO-3, and GO-4) before the HRPSoC cycle test. of the GO-1 plate (see Table 3), which may be related to the relatively
higher contents of C]O and OeC]O groups on GO4-H. Even though
the C]O and OeC]O groups on GO can promote H2 evolution, they
are also the active sites for the electrodeposition of Pb [8] to decrease
SGO and weaken their promotion effect to some extent. In this case, the
GO-4 plate does not show the strongest H2 evolution performance.
According to the above discussion, all the oxygenated groups on GO
can promote the H2 evolution in the negative plate, especially the C–O
groups. In this case, other groups, such as reported pyrrole nitrogen
[38], or metal or metal oxide with higher EH [6,7,39] need to be
modified on GO to inhibit the H2 evolution, which will be performed in
our future research.

4.2. Effect of different GO oxygenated groups on the HRPSoC cycle life of


the simulated test cells

Generally, the CL of the simulated test cells is controlled by the


growth rate of PbSO4 crystals in the negative plates [1,24,25]. Ac-
Fig. 7. Water contact angles for the NAM in the negative plates (Blank, GO-1,
cording to Pavlov et al. [36], under the HRPSoC condition of this work,
GO-2, GO-3, and GO-4).
the dominating elementary processes in the negative plates should be
the electrochemical reactions of Pb oxidation and PbSO4 reduction, and
4. Discussion the chemical reactions of formation and dissolution of PbSO4. In this
case, the effect of the capacitive charging and discharging processes on
4.1. Effect of different GO oxygenated groups on the H2 evolution process the growth of PbSO4 crystals in the negative plates should not be the
main issue. With an increase in the SBET and Vpore of the GO-1, GO-2,
The H2 evolution can take place on the surface of Pb particles (SPb) GO-3 and GO-4 plates (see Fig. 6), their largely increased Sreal provides

Fig. 8. Real negative plate (a) and schematic diagrams for (b) GO1-M, GO3-M and GO4-H; and for the cross section of (c) the Blank, (d) GO-1 and (e) GO-4 plates.

419
W. Cai et al. Journal of Energy Storage 18 (2018) 414–420

extra nucleation sites for PbSO4 crystals to facilitate the formation of containing GO4-H has moderate hydrophilicity, the largest SBET and
the small-sized PbSO4 crystals with higher solubility [24,25,40]. Thus, Vpore.
the reduction of PbSO4 crystals is accelerated for the increase in the (4) GO with relatively lower contents of CeO groups and higher con-
dissolution of PbSO4 crystals. The obvious increase in IEH of these plates tents of carbonyl and carboxyl groups (e.g. GO4-H) seems to be
(see Fig.3 and Table 3) can prove this acceleration effect [24,25]. more appropriate as the additive of negative plates.
Moreover, smaller PbSO4 crystals on the surface of these negative
plates, as shown in Fig. 5(AII–EII), further verify that the addition of the Acknowledgments
GO additives suppresses the growth of PbSO4 crystals. As a result, the
CL of the simulated test cells is significantly prolong (see Fig. 4 and The authors appreciate the analysis support of State Key Laboratory
Table 4). This effect of the GO additives should be ascribed to their of Material Processing and Die & Mould Technology and Center of
change on the microstructure of the negative plates [4]. Forecasting and Analysis, Huazhong University of Science and
Another effect of GO may result from its obvious water absorb- Technology.
ability, which can influence its dispersion state in the NAM to change
the microstructure of the negative plate, meanwhile, increase the hy- References
drophilicity of the NAM to facilitate acid diffusion into their inner vo-
lume at high rates of charge and discharge [34,41]. Therefore, the Sreal [1] L.T. Lam, N.P. Haigh, C.G. Phyland, A.J. Urban, J. Power Sources 133 (2004)
126–134.
of the negative plate can be further enlarged to impede the growth of
[2] P.T. Moseley, D.A.J. Rand, B. Monahov, J. Power Sources 219 (2012) 75–79.
PbSO4 crystals. In order to confirm the above speculation, water contact [3] E. Ebner, D. Burow, A. Börger, M. Wark, P. Atanassova, J. Valenciano, J. Power
angle tests were performed for the NAM in the Blank, GO-1, GO-2, GO-3 Sources 239 (2013) 483–489.
[4] D. Pavlov, P. Nikolov, T. Rogachev, J. Power Sources 196 (2011) 5155–5167.
and GO-4 plates to verify their difference [42]. The prepared negative [5] D.G. Enos, S.R. Ferreira, H.M. Barkholtz, W. Baca, S. Fenstermacher, J.
plate was dried in vacuum and ground into powder, and then the Electrochem. Soc. 164 (2017) A3276–A3284.
powder was pressed to a smooth sheet under a pressure of 1960 N cm−2 [6] B. Hong, L. Jiang, H. Xue, F. Liu, M. Jia, J. Li, Y. Liu, J. Power Sources 270 (2014)
332–341.
for the contact angle test, as shown in Fig. 7. The NAM in the blank [7] L. Zhao, B. Chen, J. Wu, D. Wang, J. Power Sources 248 (2014) 1–5.
plate shows hydrophobicity, while those containing the GO samples [8] L. Wang, H. Zhang, G. Cao, W. Zhang, H. Zhao, Y. Yang, Electrochim. Acta 186
(2015) 654–663.
show hydrophilicity, which seems to increase with the oxidation degree [9] J. Xiang, C. Hu, L. Chen, D. Zhang, P. Ding, D. Chen, H. Liu, J. Chen, X. Wu, X. Lai,
of the GO samples (except GO4-H, see Table 2). The increase in the J. Power Sources 328 (2016) 8–14.
hydrophilicity of the NAM containing GO4-H may be related to its re- [10] J. Lian, W. Li, F. Wang, J. Yan, K. Wang, S. Cheng, K. Jiang, J. Electrochem. Soc.
164 (2017) A1726–A1730.
latively higher contents of C]O and OeC]O groups. The GO samples [11] P. Bača, K. Micka, P. Křivík, K. Tonar, P. Tošer, J. Power Sources 196 (2011)
in the NAM indeed increase its water absorbability. 3988–3992.
Fig. 8 presents the schematics to show the comprehensive effects of [12] M. Fernández, J. Valenciano, F. Trinidad, N. Muñoz, J. Power Sources 195 (2010)
4458–4469.
different GO samples on the negative plates. The GO-1 and GO-4 plates [13] N. Sugumaran, P. Everill, S.W. Swogger, D.P. Dubey, J. Power Sources 279 (2015)
are selected for the comparison because GO1-M and GO4-H have si- 281–293.
[14] M. Saravanan, P. Sennu, M. Ganesan, S. Ambalavanan, J. Electrochem. Soc. 160
milar oxidation degree but different contents of oxygenated groups (see (2012) A70–A76.
Table 2). The NAM in GO-4 has moderate hydrophilicity to facilitate [15] R. Marom, B. Ziv, A. Banerjee, B. Cahana, S. Luski, D. Aurbach, J. Power Sources
acid diffusion into the inner of the plate, while the GO-4 plate has the 296 (2015) 78–85.
[16] A. Banerjee, B. Ziv, E. Levi, Y. Shilina, S. Luski, D. Aurbach, J. Electrochem. Soc.
largest SBET and Vpore (see Fig. 6). These two effects may result in the 163 (2016) A1518–A1526.
largest Sreal to prevent the growth of PbSO4 crystals (as shown in [17] S.M. Kumar, S. Ambalavanan, S. Mayavan, RSC Adv. 4 (2014) 36517–36521.
[18] K.Y. Kan, X. Zhang, S.C.T. Kwok, F. Ciucci, M.M.F. Yuen, RSC Adv. 5 (2015)
Fig. 8c– e). Therefore, the simulated test cells containing GO-4 have the
71314–71321.
longest CL (see Table 4). These results suggest that higher hydro- [19] Q. Long, G. Ma, Q. Xu, C. Ma, J. Nan, A. Li, H. Chen, J. Power Sources 343 (2017)
philicity of GO (such as GO3-M, see Fig.7) may be harmful to its dis- 188–196.
[20] P.T. Moseley, J. Power Sources 191 (2009) 134–138.
persion in NAM resulting in a decrease in the SBET and Vpore of the [21] M. Blecua, E. Fatas, P. Ocon, B. Gonzalo, C. Merino, F. de la Fuente, J. Valenciano,
negative plate (such as GO-3, see Fig. 6). Contrarily, the lower hydro- F. Trinidad, Electrochim. Acta 257 (2017) 109–117.
philicity of GO (such as GO1-M, see Fig.7) may weaken the acid dif- [22] D. Pavlov, P. Nikolov, J. Electrochem. Soc. 159 (2012) A1215–A1225.
[23] P. Tong, R. Zhao, R. Zhang, F. Yi, G. Shi, A. Li, H. Chen, J. Power Sources 286
fusion into the inner of the plate so as to reduce the inhibition effect on (2015) 91–102.
the growth of PbSO4 crystals (as shown in Fig. 8d) and therefore, de- [24] H. Yang, Y. Qiu, X. Guo, Electrochim. Acta 235 (2017) 409–421.
[25] H. Yang, Y. Qiu, X. Guo, Electrochim. Acta 215 (2016) 346–356.
crease the CL of the test cells (see Table 4). The hydrophilicity of GO is [26] J. Chen, B. Yao, C. Li, G. Shi, Carbon 64 (2013) 225–229.
closely related to its oxidation degree, which is mainly determined by [27] J. Chen, Y. Zhang, M. Zhang, B. Yao, Y. Li, L. Huang, C. Li, G. Shi, Chem. Sci. 7
the contents of CeO groups. In this case, the GO samples with relatively (2016) 1874–1881.
[28] C.K. Chua, M. Pumera, Chem. Soc. Rev. 43 (2014) 291–312.
lower CeO groups and higher contents of C]O and OeC = O groups [29] Y. Tang, Y. Liu, W. Guo, T. Chen, H. Wang, S. Yu, F. Gao, J. Phys. Chem. C 118
(e.g. GO4-H) to maintain a moderate hydrophilicity may be more ap- (2014) 24866–24876.
propriate as the additives of the negative plates. [30] J. Chen, F. Chi, L. Huang, M. Zhang, B. Yao, Y. Li, C. Li, G. Shi, Carbon 110 (2016)
34–40.
[31] H. Wang, X. Yuan, Y. Wu, H. Huang, G. Zeng, Y. Liu, X. Wang, N. Lin, Y. Qi, Appl.
5. Conclusions Surf. Sci. 279 (2013) 432–440.
[32] S. Stankovich, R.D. Piner, S.T. Nguyen, R.S. Ruoff, Carbon 44 (2006) 3342–3347.
[33] T. Szabó, O. Berkesi, I. Dékány, Carbon 43 (2005) 3186–3189.
(1) The addition of the GO samples (GO1-M, GO2-M, GO3-M, GO4-H) [34] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu,
with different oxygenated groups can largely promote the H2 evo- S.T. Nguyen, R.S. Ruoff, Carbon 45 (2007) 1558–1565.
[35] J. Zhang, H. Yang, G. Shen, P. Cheng, J. Zhang, S. Guo, Chem. Commun. 46 (2010)
lution and reduction of PbSO4 processes in the negative plate. An 1112–1114.
increase in the content of CeO (CeOH and CeOeC) groups of GO [36] D. Pavlov, P. Nikolov, J. Power Sources 242 (2013) 380–399.
[37] W.L. Zhang, J. Yin, Z.Q. Lin, J. Shi, C. Wang, D.B. Liu, Y. Wang, J.P. Bao, H.B. Lin, J.
has strong promotion effect on the H2 evolution process.
Power Sources 342 (2017) 183–191.
(2) The addition of the GO samples can increase the HRPSoC cycle life [38] H.M. Jeong, J.W. Lee, W.H. Shin, Y.J. Choi, H.J. Shin, J.K. Kang, J.W. Choi, Nano
of the simulated test cells significantly and those containing GO4-H Lett. 11 (2011) 2472–2477.
[39] L. Zhao, B. Chen, D. Wang, J. Power Sources 231 (2013) 34–38.
have the longest cycle life. [40] D. Pavlov, P. Nikolov, T. Rogachev, J. Power Sources 195 (2010) 4435–4443.
(3) The addition of the GO samples can largely increase the SBET and [41] J. Xiang, P. Ding, H. Zhang, X. Wu, J. Chen, Y. Yang, J. Power Sources 241 (2013)
Vpore of the negative plate. In addition, the GO samples obviously 150–158.
[42] A. Yan, X. Xiao, I. Külaots, B.W. Sheldon, R.H. Hurt, Carbon 44 (2006) 3116–3120.
increase the hydrophilicity of the NAM. The negative plate

420