close

Вход

Забыли?

вход по аккаунту

?

slct.201702367

код для вставкиСкачать
DOI: 10.1002/slct.201702367
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Full Papers
z Energy Technology & Environmental Science
Impact of Carbon Nanostructures as Additives with Spinel
Li4Ti5O12/LiMn2O4 Electrodes for Lithium Ion Battery
Technology
Binitha Gangaja, Kasireddy Siva Reddy, Shantikumar Nair, and
Dhamodaran Santhanagopalan*[a]
Spinel structured nanomaterials have shown good stability for
lithium ion storage applications. Among all, Li4Ti5O12 (LTO)
anode and LiMn2O4 (LMO) cathode is a potential combination
for high energy and high power applications. In the present
work, we utilize this specific combination to fabricate full-cells
in combination with carbon nanostructures as additives.
Typically, 20–25 nm sized LTO and 200–500 nm sized LMO
nanoparticle electrodes are composited with carbon nanostructures including, carbon nanotube (CNT), carbon black (CB)
and graphene nanoplatelets (GNP). High rate performance of
respective half-cells (lithium metal as counter electrode) of LTO
Introduction
Lithium ion batteries (LIBs) have attracted ample interest due
to its potential attributes like high gravimetric and volumetric
energy densities. Even though the technology is adequate for
portable electronic devices, the increased demand for high
energy storage applications prompts researchers to develop
electrode materials that can deliver high capacity with better
rate and cycling stability.[1] Considerable works have been
devoted to design and improve electrodes performance by
tuning synthesis condition to obtain materials with different
architectures, high surface area, improving particle-particle
contact etc.[2,3] Nevertheless, the quest to further advance
electrode performance is yet a blazing area of interest.
Spinel based electrodes are excellent class of LIB electrode
materials which has AB2O4 general structure and has the
advantage of reversible interaction of lithium into and out of
the structure leading to high cycling stability.[4,5] In consideration of the environmental benefits and cost effectiveness,
cubic spinel Li4Ti5O12 (LTO) and LiMn2O4 (LMO) emerged as
exciting anode and cathode materials respectively. LTO being
capable of delivering 175 mAh/g theoretical capacity when
discharged to 1.0 V thus minimizing the safety concern, makes
it a commercially accepted anode for lithium ion battery
application.[6–9] Nevertheless, its poor electronic conductivity is
[a] B. Gangaja, K. S. Reddy, Dr. S. Nair, Dr. D. Santhanagopalan
Centre for Nanosciences and Molecular Medicine,
Amrita University, AIMS (P.O), Kochi, India 682 041
E-mail: dsgopalan20710@aims.amrita.edu
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/slct.201702367
ChemistrySelect 2017, 2, 9772 – 9776
Wiley VCH
1730 / 101655
and LMO are tested up to 50C. It was found that half-cells with
CNT additive retained almost 80% of its 1C rate capacity at 50C
rate. Also both the electrodes exhibited 1000 cycles stability
with retention of about 80% at 10C rate cycling. Using these
CNT additive based electrodes, a full-cell fabricated and tested
exhibited high capacity and stable cycling over 500 cycles at
1000 mA/g specific current. The full-cell delivered power
density of about 2310 W/kg and energy density of about
140 Wh/kg that can be further improved for high power Li-ion
battery technology.
still a challenge.[10] Approaches to improve LTO conductivity
through techniques like in-situ and ex-situ carbon coating,
particle size reduction, doping etc are extensively being
explored.[6,7,11,12] On the other hand, spinel LMO cathode has
also engrossed intense attention due to its high voltage, three
dimensional diffusion pathways, non-toxicity and abundance.[13]
Yet the dissolution of Mn2 + leading to rigorous capacity loss is
a key hindrance.[14,15] Surface modifications via coating with
oxides and non-oxides, doping with metal and metal oxide
nanoparticles etc were reported to circumvent the dissolution
issues of LMO.[16,17]
In this report, we have focused on two prime aspects; (i) to
explore the effect of carbon additives on improving the
electrochemical performance of the synthesized spinel based
electrode materials. Carbon black is one of the conventionally
used carbon additives for electrode fabrication process. This is
probably due to the advantage of having high packing density
due to its nanosized and spherical morphology. Nevertheless,
the material fails to provide good particle to particle electrical
contact. Reports on other carbon additives like graphene
sheets, carbon nanotubes on battery cathodes and anode
materials are also investigated.[18,19] Yet details on the effect of
different conductive additives on both cathode and anode are
very minimal. Herein, we investigate the effect of three different commercially available conductive carbon additives that
are 0D, 1D and 2D nature. Specifically carbon black (CB), multiwalled carbon nanotubes (CNT) and graphene nanoplatelets
(GNP) have been used as additives to assess the electrochemical performance of both spinel LTO and LMO. Our result
direct to the conclusion that, no matter how fine is the active
electrode, conductive additive have crucial effect in improving
9772
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Montag, 23.10.2017
[S. 9772/9776]
1
Full Papers
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
the electrochemical performance especially at high rates.
Among the 3 conductive agents, LTO and LMO composited
with CNT additive were found to be excellent for high rate
performances. (ii) To extend the half-cell performance to a
viable device application, a full cell with best performing
spinel/carbon additive combination (LMO-CNT and LTO-CNT)
has been demonstrated.
Results and Discussion
Structural and morphological characterization of spinel
samples
Both spinel LTO and LMO were structurally characterized using
X-ray diffraction as depicted in Figure 1a & b. XRD pattern
Figure 2. Low magnification TEM images of (a) LTO (b) LMO and HRTEM of
LTO and LMO with lattice fringes marked.
corresponding to (111) plane of cubic spinel LTO/LMO as
represented in Figure 2c & d.
Structural and morphological characterization of conductive
carbon samples
Figure 3a, b & c shows the high resolution TEM images of three
carbon nanostructures with their corresponding low magnifica-
Figure 1. Structural characterization of synthesized spinel samples (a & c) XRay diffraction pattern of LTO and LMO (c & d) Raman spectra of spinel LTO
and LMO respectively.
correlates well with the cubic spinel structure with maximum
intensity peaking for (111) indices. Nano-morphology of LTO
can be confirmed from broad peak in XRD (Figure 1a) while
intense, sharp peaks of LMO correspond to relatively larger
particle size (Figure 1b). Absence of any other parasitic peaks
validates the purity of both the samples. Raman spectroscopy
was conducted to additionally confirm the sample purity.
Figure 1c shows the Raman spectra of LTO wherein peaks
centred at 235, 335, 430, 675 and 740 cm 1corresponds to F2g,
F2g, Eg and A1g bands of cubic spinel LTO.[20] LMO exhibit only
one Raman active vibration at 639 cm 1 (Figure 1d) corresponding to A1g mode resulting from the vibration of oxygen atom in
Mn O linkage.[21] TEM confirms the nano nature of LTO with
average particle size between 20–25 nm while LMO was found
to be large in size with 300 nm in average (refer Figure 2a & b).
HRTEM further confirm the lattice fringes spacing 0.48 nm
ChemistrySelect 2017, 2, 9772 – 9776
Wiley VCH
1730 / 101655
Figure 3. (a, b, c) represents the high magnification TEM images of CNT, CB
and GNP samples respectively with corresponding low magnification image
as inset, (d, e, f) shows the Raman spectra of CNT, CB and GNP samples
respectively.
tion images as inset. As can be seen, low magnification image
of CNT sample shows multi-walled tubes with length in microns
range and 5 to 20 nm outer diameters. High resolution TEM
9773
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Montag, 23.10.2017
[S. 9773/9776]
1
Full Papers
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
images show the crystalline graphitized nature of CNT. Carbon
black consists of spherical particles with average particle size
around 50 nm. Graphene nanoplatelets are observed to be
agglomerated with platelet morphology (inset) and lattice
fringes can be clearly seen from the high magnification images
confirming the crystalline nature. The carbon nanostructures
were structurally characterized by Raman spectroscopy to
obtain a brief understanding about the graphitized nature of
the samples. From the relative peak intensities of D and G
bands, it is clearly seen that CNT, GNP and CB samples has
graphitized nature signifying the presence of sp2 hybridized
carbon. The D and G bands of all 3 samples are at ~ 1349 cm 1
and ~ 1573 cm 1 respectively.
Electrochemical Characterization
Figure 4a shows the charge-discharge profiles of LTO with three
carbon samples at 1C rate (for LTO 1C = 175 mA/g). As can be
Figure 5. (a) Rate performance of LTO/C samples at different rates ranging
from 1C to 50C with discharge rate fixed at 1C (b) LMO/C sample discharge
rate performance from 1C to 50 C with charging rate fixed at 1C rate and (c)
long cycling performance of LTO/CNT and LMO/CNT at respective 10C rate
(both charge and discharge).
Figure 4. 1st cycle charge-discharge profiles of (a) LTO/C and (b) LMO/C
samples as indicated in legend.
seen, all the samples shows typical discharge plateau at 1.5 V
and slightly polarized charge plateau at 1.6 V, typical of
nanoscale effect of LTO electrode. This plateau corresponds to
the two phase reaction, transformation of Li4Ti5O12 phase to
Li7Ti5O12 and reverse upon charging. This further confirms that
the three carbon samples do not contribute electrochemically
in the voltage window between 1 V– 2.5 V. All three samples
show comparable capacity values at 1C rate with LTO-CB
rendering a best charge capacity of 148 mAh/g. Cycling profiles
of LMO-carbon electrodes at 1C rate (for LMO 1C = 148 mA/g)
within 3–4.5 V potential window are shown in Figure 4b.
LMO C electrodes show two charge plateaus at 4.0 V and 4.1 V
and two discharge plateaus at 4.1 V and 3.8 V in the potential
window 3 V to 4.5 V. Unlike LTO C electrodes, all 3 forms of
carbon additives with LMO shows dissimilar capacity values at
1C rate and among the three, LMO-CB displayed the best
capacity value at 1C rate.
To evaluate the electrochemical performance of LTO C and
LMO C electrodes at high rates, galvanostatic rate test were
conducted in respective potential windows from 1C to 50C as
depicted in Figure 5a & b. From rate performance of three
LTO C samples, LTO-CNT & LTO-CB electrode was observed to
have similar performance. Even at 50 C the electrodes were
ChemistrySelect 2017, 2, 9772 – 9776
Wiley VCH
1730 / 101655
able to deliver capacity as high as 124 mAh/g whereas the
same LTO material with GNP electrode could deliver barely 25
mAh/g at 5C. Since the CB and CNT electrodes showed similar
performance, we conducted rate tests for LTO/C samples with
both charge and discharge capacity at same rates as shown in
Figure S1. As observed, LTO-CNT exhibited excellent electrochemical performance particularly at high rates in comparison
to other two carbon additives. This excellent performance of
LTO-CNT electrode can be correlated to the formation of
uniform percolated network of highly conducting CNT with
nanosized LTO particles leading to enhanced electron transport
at high rates. Additionally, high aspect ratio and surface area of
CNT enables multi-point contact and better filling space with
active material leading to rapid charge transport and thus high
rate performance. Seggregation and improper distribution of
CB and GNP significantly possibly affects the electrochemical
performance at high rates irrespective of their conductivity
variations. Based on the rate performance of LTO-CB sample
with and without discharge capacity limitation, we can
substantiate that charge transfer during lithium intercalation is
the key hindrance for rate performance while de-intercalation is
relatively facile. This validates that incorporation of highly
conductive additive can significantly improve the rate performance at ultra-high rates. Considering the excellent performance
of LTO-CNT sample in both rate tests (with and without
discharge rate limitation), we have chosen LTO-CNT as the best
sample and were used for further analysis. Our model cathode
material LMO/C rate performances were depicted in Figure 5b.
Out of the three additives for LMO, LMO-CNT showed the best
performance delivering a discharge capacity as high as 82
mAh/g at 50C. Akin to LTO-GNP, LMO-GNP has extremely poor
performance (~ 5 mAh/g) at high rates. This can be related to
9774
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Montag, 23.10.2017
[S. 9774/9776]
1
Full Papers
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
the large particle size of LMO which prevent GNP from properly
blending with LMO.
Further to estimate the cycling stability of both LTO-CNT
and LMO-CNT electrodes, galvanostatic charge discharge tests
were conducted at respective 10C rates. Figure 5c represents
the cycling data for both the samples for 1000 cycles. Both
samples showed excellent cycling stability with Coulombic
efficiency > 99% (data displayed in Figure S2). LTO-CNT electrode delivered a discharge capacity of 155.8 mAh/g and
capacity retention of 80% even at the end of 1000 cycles while
discharge capacity of 104 mAh/g and retention of 82% was
offered by LMO-CNT electrode. This long cycling performance is
better than many reports available in literature for both LTO[6–9]
and LMO[14,15] whereas the same were even comparable with
many literatures with coating and doping.[22–25]
Figure S3 illustrates the electrochemical impedance spectra
of LTO/C samples at OCV condition. By gradually decreasing
the frequency from 100 kHz to 10 mHz, spectra originates with
three distinct regions. High frequency region, where real axis
intercept represents the equivalent series resistance (Rs), middle
frequency denotes charge transfer resistance (Rct) and low
frequency indicate diffusion or Warburg impedance. Among
the three LTO/C and LMO/C electrodes, Rs was a bit larger for
the CNT based electrodes which could be attributed to the
contact issues with the current collector however the Rct values
were minimal with 43.4 W and 49.4 W for LTO-CNT & LMO-CNT
respectively. The semicircle in the high frequency region and
the low frequency Warburg line meet at a lower Z’’ value for
the CNT composite electrode than other two carbon additives.
Such a complete semicircle in impedance for electrodes with
CNT additive indicates the formation of an effective percolated
conductive pathway compared to the electrodes with carbon
black and graphene platelet additives. This enables improved
performance rendered by spinel electrodes with CNT additives
especially at high rates. Cyclic voltammetry profiles presented
in Figure S4 displays relatively sharp redox peaks and low
polarization for CNT composite electrode compared to the
other two carbon additives indicating better kinetics.
To evaluate the benefit of these electrodes to be used for
commercial applications, full cell was fabricated with LTO-CNT
as anode and LMO-CNT as cathode. Figure 6a shows the rate
performance of the full-cell at different current densities with 5
cycles at each rate. About 75% of the discharge capacity value
is retained for 10 times increase in current density from 50 mA/
g to 500 mA/g. Figure 6b shows the charge discharge profiles
of full cell electrode at 1000 mA/g rate in the potential window
1.5 to 3 V. The high performances of half cells were reflected
even in full cell configuration wherein the cell was able to
deliver first cycle capacity as high as 86 mAh/g at 1000 mA/g
current rate. Upon long cycling, the full-cell showed good
performance in terms of Coulombic efficiency and cycling
stability as displayed in Fig. 6c. Capacity loss of merely 0.057%
per cycle was observed over 500 cycles. Full-cell rendered
power density of 2310 W/kg and energy density of 141 Wh/kg
at the end of 500th cycles.
ChemistrySelect 2017, 2, 9772 – 9776
Wiley VCH
1730 / 101655
Figure 6. Electrochemical performance of LTO-LMO full cell (a) Rate performance from 50 mA/g to 500 mA/g current density (b) 1st cycle charge
discharge profile at 1000 mA/g and (c) long cycling performance at a current
rate of 1000 mA/g for 500 cycles.
Conclusions
In summary we have conducted a study to evaluate the effect
of 3 different conductive carbon additives on electrochemical
performance of spinel electrodes using LMO and LTO as model
cathode and anode respectively. Even though the electrodes
were found to have comparable electrochemical performance
at low rate, significant different was observed at high rate for
both LTO and LMO. This can be related to the effectiveness of
conductive carbon in facilitating rapid electron transport. For
both LTO and LMO samples, CNT was found to be proficient in
terms of both capacity and cyclability specifically at high rates.
Further, full-cell was demonstrated with LTO-CNT and LMO-CNT
combination was capable of rendering excellent capacity and
cycling stability. Cycling for over 500 cycles was verified with
very minimal capacity loss of 0.057% per cycle at 1000 mA/g
specific current. Regardless of the active material’s performance, our results signify the necessity of employing highly
conducting, homogenous dispersion of conductive carbon
additive so as to provide excellent electrochemical performance
in full-cell configuration, even at high rates.
Supporting Information Summary
Detailed experimental procedure, rate performance of LTO/C
samples without limiting capacity, 10C Coulombic efficiency
plot of both LMO CNT & LTO-CNT, electrochemical impedance
spectra for LTO/C & LMO/C combination and cyclic voltammetric profiles at 0.5 mV/s scan rate can be obtained via supporting
info file from the article webpage.
9775
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Montag, 23.10.2017
[S. 9775/9776]
1
Full Papers
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Acknowledgements
Research on the LTO part was supported by Indian Space
Research Organization (ISRO), Government of India (Ref: ISRO/
RES/3/60/2015-16). DS is thankful to Science and Engineering
Research Board, Government of India for Ramanujan Fellowship
(Ref: SB/S2/RJN-100/2014). Authors thank Amrita University for
infrastructural support.
Conflict of Interest
The authors declare no conflict of interest.
Keywords: Carbon additives · Energy storage · Lithium ion
battery · Lithium manganese oxide · Lithium titanate
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
J. Banerjee, K. Dutta, Crit. Rev. Solid State Mater. Sci. 2016, 8436, 1–21.
N. Nitta, F. Wu, J. T. Lee, G. Yushin, Mater. Today 2015, 18, 252–264.
N. Mahmood, Y. Hou, Adv. Sci. 2014, 1, 1–20.
S. Patoux, L. Daniel, C. Bourbon, H. Lignier, C. Pagano, F. Le Cras, S.
Jouanneau, S. Martinet, J. Power Sources 2009, 189, 344–352.
E. Ferg, R. J. Gummow, A. de Kock, J. Electrochem. Soc. 1994, 141, L147.
L. Zhao, Y. S. Hu, H. Li, Z. Wang, L. Chen, Adv. Mater. 2011, 23, 1385–
1388.
C. Cheng, H. Liu, X. Xue, S. Cao, H. Cao, L. Shi, RSC Adv. 2014, 4, 63105–
63109.
L. Shen, C. Yuan, H. Luo, X. Zhang, K. Xu, Y. Xia, J. Mater. Chem. 2010, 20,
6998.
C. Kim, N. S. Norberg, C. T. Alexander, R. Kostecki, J. Cabana, Adv. Funct.
Mater. 2013, 23, 1214–1222.
E. Zhao, C. Qin, H. R. Jung, G. Berdichevsky, A. Nese, S. Marder, G. Yushin,
ACS Nano 2016, 10, 3977–3984.
ChemistrySelect 2017, 2, 9772 – 9776
Wiley VCH
1730 / 101655
[11] L. Shen, X. Zhang, E. Uchaker, C. Yuan, G. Cao, Adv. Energy Mater. 2012, 2,
691–698.
[12] J. Wang, H. Zhao, Z. Li, Y. Wen, Q. Xia, Y. Zhang, G. Yushin, Adv. Mater.
Interfaces 2016, 3, 1600003.
[13] O. K. Park, Y. Cho, S. Lee, H.-C. Yoo, H.-K. Song, J. Cho, Energy Environ. Sci.
2011, 4, 1621.
[14] Y. L. Ding, J. Xie, G. S. Cao, T. J. Zhu, H. M. Yu, X. B. Zhao, Adv. Funct.
Mater. 2011, 21, 348–355.
[15] D. Guan, C. Cai, Y. Wang, 2011 IEEE Green Technol. Conf. Green 2011 2011,
1465–1469.
[16] T. F. Yi, Y. R. Zhu, X. D. Zhu, J. Shu, C. B. Yue, A. N. Zhou, Ionics (Kiel).
2009, 15, 779–784.
[17] Z. Zhang, Z. Chen, G. b Wang, H. Ren, M. Pan, L. Xiao, K. Wu, L. Zhao, J.
Yang, Q. Wu, et al., Phys. Chem. Chem. Phys. 2016, 18, 6893–6900.
[18] a) R. Jiang, C. Cui, H. Ma, Electrochim. Acta 2013, 104, 198–200. b) N.
Takami, H. Inagaki, Y. Tatebayashi, H. Saruwatari, K. Honda, S. Egusa, J.
Power Sources 2013, 244, 469–475.
[19] a) Y.-H. Chen, C.-W. Wang, G. Liu, X.-Y. Song, V. S. Battaglia, A. M. Sastry, J.
Electrochem. Soc. 2007, 154, A978. b) I. Belharouak, M. K. Jr, T. Tan, H.
Yumoto, N. Ota, K. Amine, Chin. Chem. Lett. 2012, 159, 1165–1170.
[20] J. Y. Liao, V. Chabot, M. Gu, C. Wang, X. Xiao, Z. Chen, Nano Energy 2014,
9, 383–391.
[21] H. Choi, K. Lee, G. Kim, J. Lee, 2001, 44, 242–244.
[22] S. Nie, C. Li, H. Peng, G. Li, K. Chen, RSC Adv. 2015, 5, 23278–23282.
[23] F. Wang, L. Luo, J. Du, L. Guo, B. Li, Y. Ding, RSC Adv. 2015, 5, 46359–
46365.
[24] P. Ram, A. Gçren, S. Ferdov, M. M. Silva, R. Singhal, C. M. Costa, R. K.
Sharma, S. Lanceros-Mndez, New J. Chem. 2016, 40, 6244–6252.
[25] A. K. Haridas, C. S. Sharma, T. N. Rao, Electroanalysis 2014, 26, 2315–2319.
[26] K. S. Reddy, B. Gangaja, S. Nair, D. Santhanagopalan, Electrochim. Acta
2017,250, 359–367.
Submitted: October 9, 2017
Revised: October 11, 2017
Accepted: October 13, 2017
9776
2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Montag, 23.10.2017
[S. 9776/9776]
1
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 076 Кб
Теги
201702367, slct
1/--страниц
Пожаловаться на содержимое документа