Carbon 196 (2022) 589–595

Contents lists available at ScienceDirect

Carbon
journal homepage: www.elsevier.com/locate/carbon

Achieving SEI preformed graphite in flow cell to mitigate initial lithium loss
Yikang Yu a, b, Zhenzhen Yang c, Yuzi Liu d, Jian Xie a, *
a
Department of Mechanical and Energy Engineering, Purdue School of Engineering and Technology, Indiana University – Purdue University Indianapolis, Indianapolis,
IN, 46202, United States
b
School of Mechanical Engineering, Purdue University, West Lafayette, IN, 47907, United States
c
Chemical Sciences and Engineering Division, Argonne National Laboratory, Lemont, IL, 60439, United States
d
Center for Nanoscale Materials, Argonne National Laboratory, Lemont, IL, 60439, United States

A R T I C L E I N F O A B S T R A C T

Keywords: The irreversible lithium loss due to the formation of solid electrolyte interphase (SEI) in the initial cycle on the
Graphite powder graphite anode greatly reduces the overall cell energy density of lithium ion batteries, that is, the lost Li ions from
Lithium ion batteries forming SEI lead to the decrease of Li ions for the intercalation. The method of cathode prelithiation has been
Solid electrolyte interphase
widely explored to compensate this lithium loss. However, these cathode additives with high lithium contents
Initial capacity loss
inevitably lower the loading of the cathode active materials. Here we report a novel approach to solve this
challenge, a facile graphite prelithiation method by preforming SEI layers on the surface of graphite powders
(Pre-SEI graphite) utilizing a specially designed flow cell. The Li accommodation in the graphite anode can be
controlled by the operating time and current density in the flow cell for the electrochemical SEI formation. As a
result, we demonstrate a 10% initial Columbic efficiency increase of the LiFePO4 electrode in a full cell
configuration using the Pre-SEI graphite, compared with the pristine graphite anode. The electrochemical pre­
formation of SEI on the graphite powders offers a complete solution to offset initial lithium loss without a
sacrifice of active cathode material loading.

1. Introduction reagents could greatly compensate the lithium loss and thus increase the
full cell energy density. Nevertheless, the additives of these reagents
The high energy density lithium ion batteries (LIBs) are continuously inevitably lower the active cathode material loading and leave inactive
pursued to fulfill the growing demands of the energy storage markets [1, residues which make it hard to fully unlock the potential of the full cell
2]. Typically, a commercial LIB cell consists of a lithium transition metal energy density. Moreover, those residues generated after the lithium
oxide cathode (i.e., LiFePO4, LiCoO2, and LiMn2O4, etc.) and a graphite compensation are typically not electrical conductive (e.g., S, metal
anode [3]. Their energy density is dictated by the reversible number of oxide, Table 1) and consequently, destroy the spatial continuity of the
the Li ions shuttled between the anode and cathode within per cell mass whole electrode for electron transportation. The use of these additives
and cell voltage. However, during the initial charge process, there are may also be accompanied with the evolution of undesired gas products
5–20% of the Li ions irreversibly consumed due to the formation of SEI such as CO2, O2 and N2, raising the safety concerns of battery operation
layer on graphite surface [4–7]. In the following cycles, there will be [13–15].
5–20% of Li ions shortage for shuttling between the anode and cathode On the other hand, direct anode prelithiations using chemical
in the LIB cell operation. As the consequence, this initial lithium loss method or stabilized lithium metal powder (SLMP), to name a few, are
(ICL) greatly reduces the overall battery cell energy density. Some effective approaches to elevate ICE but easily give rise to a further Li
higher capacity anode materials, such as Sn and Si, make this situation intercalation into these anode materials (e.g., graphite, SiOx) [9,16–18].
even worse considering their much lower initial Coulombic efficiency In this regard, an overcharged status is created once coupling them
(ICE) [8,9]. directly with most of the commercial Li-containing cathode materials (e.
To address this issue, the cathode prelithiation with high Li content g. LiCoO2, LiFePO4). Tedious procedures such as redundant Li extrac­
sacrificial reagents has been widely explored as a common strategy to tion, cell disassembly, electrode clean, and further full cell re-assembly
compensate this lithium loss [10–12]. The high lithium content of those are required for these prelithiated anode materials to ensure battery

* Corresponding author.
E-mail address: [email protected] (J. Xie).

https://doi.org/10.1016/j.carbon.2022.05.024
Received 22 February 2022; Received in revised form 9 May 2022; Accepted 11 May 2022
Available online 14 May 2022
0008-6223/© 2022 Elsevier Ltd. All rights reserved.
Y. Yu et al. Carbon 196 (2022) 589–595

Table 1 semi-flow cell, in which 50 mL 2 M LiTFSI/0.4 M LiNO3 in DOL was used

Representatives of cathode prelithiation additives to offset ICL. as electrolyte. It is worth noting that the typical ethylene carbonate (EC)
No. Additives Specific Capacity Lithiation Reference and ethyl methyl carbonate (EMC) based electrolyte (e.g., 1.2 M LiPF6 in
(mAh/g) Residues EC and EMC) leads to a severe graphite agglomeration and hinders the
1 Li2C3O5 430 CO2, C [15] electrolyte flow. 150 mg graphite powers were dispersed into this
2 Li2C2O4 545 CO2, C [15] electrolyte before the operation of the flow cell system. In this flow cell
3 LiN3 567 N2 [15] (Fig. 1a), one piece of lithium foil was placed on the steel anode side as
4 Cu/Li2O (1/ 574 CuO [4] reference and counter electrode, followed by a Celgard 2400 separator.
1)
5 Ni/Li2O (1/ 605 NiO [4]
A polytetrafluoroethylene (PTFE) interlayer with a thickness of 1 mm
1) was applied between the anode chamber and cathode chamber. After the
6 LiS2/Co 711 CoS2 [27] assembly of this flow cell in a Ar-filled glove box, it was connected with a
7 Fe/Li2O (3/ 747 Fe3O4 [4] peristaltic pump (MasterFlex, Cole-Parmer) and one electrolyte
4)
container with tubes (Fig. S1). Moreover, the electrolyte container was
8 Fe/Li2O (2/ 799 Fe2O3 [4]
3) kept stirring during the operation of the flow system to ensure a uniform
9 Li2S 1166 S [28] dispersion of graphite particles in the electrolyte and avoid any settle­
10 Li3N 1761 N2 [14] ment. With the connection of the whole flow system outside of the glove
box, the system was purged by Ar flow for 10 min to remove any O2 gas.
Then, the electrolyte flow was cycled for 30 min first (flow rate: 10 mL/
safety. In contrast, an artificial SEI has been perceived as another way to
min) without a connection to a potentiostat to obtain a superior wetting
protect anode material from potential side reaction. Certainly, the
of both Li foil and separator. During the finial operation, the voltage
reduced SEI propagation with an artificial SEI layer typically leads to an
window of the flow cell was set between 0.005 V and 2.000 V (vs. Li+/Li)
elevated ICE and electrochemical performance [19–22]. However, it is
to mimic the SEI formation process of the graphite in a normal coin cell
hard to control the loading of these artificial SEI to a precise level, which
with a current of 93 μA (1/600 C for 150 mg graphite) using a poten­
is crucial to maximize the energy density of a battery cell with a mini­
tiostat. After one charge/discharge cycle of SEI formation, these Pre-SEI
mum inactive material addition. Therefore, a complete solution to
graphite powders were washed with DOL using a centrifuge and were
minimize the Li loss remains a persistent challenge.
dried for further use. The main products of our preformed SEI are Li2CO3
Graphite is the main anode material choice for the commercial LIBs
and LiF, which is both stable at air condition. Thus, the contamination of
due to its superior reversibility and low cost [23–26]. Typically, the
air and moisture to our Pre-SEI graphite could be reduced with a limited
initial lithium loss caused by the SEI formation on the graphite anode
battery performance influence.
equals to 20–80 mAh g− 1 of lithium specific capacity. Thus, it is
meaningful to directly preform a SEI layer on graphite with certain
degree (e.g., >40 mAh g− 1) before its application on LIBs. Here, we 2.3. Electrolyte and electrode preparation
report a novel method to prepare SEI preformed graphite (Pre-SEI
graphite) to mitigate the initial lithium loss by reducing SEI formation The electrolyte used in coin cells was prepared by dissolving 1.2 M
process in battery cell, which is intrinsically different from those cath­ LiPF6 in the EC-EMC mixture solvent (3:7 by mass) with 2 wt% additives
ode or anode prelithiations. This Pre-SEI graphite is fabricated in flow of FEC. The graphite and Pre-SEI graphite electrodes were prepared by
cells to electrochemically mimic the formation of SEI within battery slurry coating on Cu foil, which comprised 80 wt% active materials
cells. This new concept of Pre-SEI graphite enables a direct Li compen­ (graphite or Pre-SEI graphite), 10 wt% Super P, and 10 wt% poly­
sation on freestanding graphite powders level rather than sophisticated vinylidene difluoride (PVDF) as the binder. Similarly, LFP electrodes
fabricated electrode. This feature is also beneficial for the further scale were followed the same mass ratio of 8:1:1 for LFP, Super P, and PVDF
up in industrial applications. Owing to the preformation SEI on graphite on Al foil. The areal capacity loading was about 1.6 mg cm− 2, and 4.5
surface, ~25 mAh g− 1 of the initial lithium loss has been compensated mg cm− 2 of anode materials and cathode materials, respectively.
based on graphite half cells. Such compensation is still slightly low to
fully address this issue, but future hope of Pre-SEI graphite as a complete 2.4. Electrochemical and materials characterization
solution to solve the ICL issue is associated with the development of the
electrolyte modification and flow cell design to a higher degree of SEI Electrochemical tests were performed using CR2032 type coin cell.
preformation. The electrochemical preformation of SEI on graphite For half-cells, Li foils were used as the counter/reference electrode, and
opens a new avenue to improve LIBs energy density by mitigating initial as-prepared electrodes (graphite, Pre-SEI graphite, LFP) were used as
lithium loss. working electrodes (d = 15 mm) with one piece of polypropylene
separator (d = 19 mm) (Celgard 2400) (USA). 75 μL of as-prepared EC-
2. Experimental section based electrolytes were used in each cell. The cut-off voltage was
controlled at 0.005–2.000 V and 2.400–4.000 V (vs. Li+/Li) for graphite
2.1. Materials and LFP, respectively. For full cells, the mass ratio of the LFP and

Graphite powders (MescoCarbon MicrBeads, MCMB), Li foil

(99.9%), LiFePO4 (LFP) and Super P conductive carbon black were ob­
tained from MTI Corporation (USA). Battery-grade LiPF6, ethylene car­
bonate (EC), and ethyl carbonate (EMC) were supplied by Novolyte
Technologies (China). 1,3-dioxolane (DOL, 99%), LiNO3 (99%), and
Fluoroethylene carbonate (FEC, 99%) were purchased from Sigma-
Aldrich (USA). lithium bis(trifluoromethanesulfonyl)imide (LiTFSI,
>98%) was supplied by TCI chemicals (Japan).

2.2. SEI preformation in flow system Fig. 1. Calculated cathode specific capacities and energy densities of NCM-
graphite cells after initial cycle using different cathode additives and Pre-SEI
The preforming SEI processes were conducted in a specially designed graphite. (A colour version of this figure can be viewed online.)

590
Y. Yu et al. Carbon 196 (2022) 589–595

graphite (or Pre-SEI graphite) was controlled to 2.8:1, which corre­ (0.005–2.0 V vs. Li+/Li) in the flow cell to mimic the formation of SEI in
sponds to a N/P ratio of ~1.02 according to the half-cell data. The cut- normal coin cells (Fig. 2, Fig. S1) while a Li foil was used as a reference
off voltage was controlled at 2.400–3.800 V (vs. Li+/Li) for these full and counter electrode. The Li accommodation in the graphite powder
cells. The cyclic voltammetry (CV) analyses of graphite and Pre-SEI anode can be simply controlled by the operating conditions such as the
graphite were conducted using a Solartron potentiostat (Ametek, UK) cycling time and current density in the flow cell for the electrochemical
with a range of 0.005–2.000 V (Li+/Li) at a scan rate of 0.5 mV s− 1. The SEI formation. Detailed set up information of this flow system can be
electrochemical impedance spectra (EIS) were measured using a Solar­ found in the experimental section. Once the graphite particles, dispersed
tron potentiostat with a AC voltage of 5 mV in the frequency range from in the electrolyte flow directly, and contact with the electrode of cathode
1000 kHz to 0.01 Hz (Model SI 1287). Galvanostatic charge-discharge of chamber, they gain electrons and render the formation of SEI layer with
all of the half-cells and full cells were collected on a LAND-CT2001 subsequent reduction of electrolyte at a desired potential range (e.g.,
battery testing instrument (Land Corp, China) at a determined C rate. <1.0 V vs. Li+/Li) as demonstrated in Fig. 2b (cell discharge). As a result,
The current density and specific capacity were based on the mass of the a SEI layer is formed on the graphite particle surface. It has been widely
active materials loading. The scanning electron microscopy (SEM) and accepted that the small current is favorable for the SEI formation pro­
transmission electron microscopy (TEM) images were acquired from a cess. Different current densities have been applied to explore the SEI
JEOL 7800F and JEOL 2100F, respectively (JEOL, Japan). XPS data of formation process in our flow cell. At a quite large current density
surface chemistry was acquired using PHI 5000 VersaProbe II system (1/100 C), there is limited SEI formed (<2 mAh g− 1, Fig. S2). Such low
(Physical Electronics, USA) that attached to an argon-atmosphere glo­ Li compensation makes it hard to be transformed to practical lithium ion
vebox to avoid any contamination of moisture and air. The high- batteries. However, a further reduction of formation current to 1/600 C
resolution spectra were obtained under the following conditions: 100 leads to a considerable SEI formation (~27.6 mAh g− 1). It is worth
μm beam (25 W) with Al Kα radiation (hυ = 1486.6 eV), Ar+-ion and noting that the contact between current collector and dispersed graphite
electron beam sample neutralization, fixed analyzer transmission mode, particles is vital to ensure the preformed SEI propagation. A greatly
and pass energy of 23.50 eV. The Shirly background data were sub­ reduced SEI preformation is demonstrated if elevating the flow rate from
tracted from all spectra. The spectra were fitted to multiple Gaussian 10 mL/min to 20 mL/min (~5.5 mAh g− 1, Fig. S3). Moreover, there is
peaks by using the software package (XPSPEAK). The reported atomic very limited lithium intercalation because the formation of insulating
concentrations were calculated from survey scans. The structure of SEI layer stops the further reaction between Li ions and graphite parti­
graphite and Pre-SEI graphite were examined by X-ray powder diffrac­ cles. Such conclusion is further verified using X-ray powder diffraction
tion (XRD) tests (Bruker D8 Discover, Bruker USA). (XRD), which presents a typical graphite layered structure pattern
(Fig. S4). After the formation of SEI on graphite, there is a 0.07◦ shift of
3. Results and discussion the 2θ value (002 peak) to a higher degree. Typically, the graphite (002)
peak shifts to a lower degree value along with the intercalation of
3.1. Design of Pre-SEI graphite lithium ions [30]. The slight difference may come from the XRD mea­
surement process (i.e., sample height difference). Accordingly, a SEI
To evaluate the concepts of the Pre-SEI graphite to mitigate the capacity as high as 27.6 mAh g− 1 (7.4% of theoretical capacity of
initial lithium loss, a full cell energy density calculations based on graphite) was formed in our Pre-SEI graphite at a low current density
LiNi0.8Co0.1Mn0.1O2 (NCM)-graphite full cell was conducted and condition.
compared with those typical cathode prelithiation additives (Table S1).
After the initial charge process, the high lithium content of those addi­ 3.2. SEI structures of Pre-SEI graphite
tives could greatly compensate the lithium loss and thus increase the
cathode specific capacity and full cell energy density in the following The SEI structure was investigated using electron microscope (SEM &
cycles (Fig. 1, Table S2). However, the inactive residues were generated TEM) and X-ray photoelectron spectroscopy (XPS). SEM images of SEI
after the lithium compensation. Unfortunately, those residues such as S layer formed on the surface of Pre-SEI graphite show a porous and rough
and metal oxides (Table 1), are typically electrical insulating, which morphology (Fig. 3d–f) compared with those pristine graphite particles
leads to a negative effect on electron transportation within the cathode (Fig. 3a–c). Such rough morphology has been further confirmed by TEM
electrode. Moreover, undesired gaseous N2 and CO2 may be produced imaging. Pristine graphite shows a clear surface with a layer structure
using some high lithium content additives (i.e., Li3N, Li2C3O5), posting a (Fig. 3g–h). In contrast, a rough surface of Pre-SEI graphite is formed by
battery safety issue as well. That is, a novel and complete lithium the decoration of SEI components (Fig. 3i). In particular, a SEI layer with
compensation strategy without inactive residues is highly desired. a thickness of 3–4 nm has been observed by the high-resolution TEM
Given this situation, we developed herein a new concept of “Pre-SEI (Fig. 3j). XPS analysis was further performed to study the chemical
graphite”, that is, a preformed SEI layer on graphite powder without compositions of the SEI layer formed in the flow cell. Elements of C, O, S,
using a battery cell, is proposed. In this configuration, a SEI layer is Li, N, and F are presented in our Pre-SEI graphite (Fig. S5) while ele­
directly formed over the surface of graphite particles in the anode and no ments of S, Li, N, and F have not been detected in the pristine graphite
inactive residues are produced, which means the full cell energy density sample (Fig. S6). A selected elements of Li 1s, C 1s, O 1s, N 1s, F 1s and S
could be elevated to a maximum extent compared those cathode addi­ 2p spectra of Pre-SEI graphite are presented in Fig. 4. The C 1s spectra
tives (Fig. 1, Table S2). According to our calculations, a fully Pre-SEI and O 1s spectra of Pre-SEI graphite are similar to graphite sample,
graphite (395 Wh kg− 1) could further increase up to 7% reversible en­ which both demonstrates a C–C and C–O bond located at ~285.0 and
ergy density based on a NCM-graphite full cell compared with those 286.5 eV, respectively. In contrast, new peaks located at ~287.5 and
cathode prelithiation methods (i.e. Li2C3O5, 368 Wh kg− 1). It is worth ~531.5 eV are found in Pre-SEI graphite, which can be ascribed to C=O/
noting that our Pre-SEI concept is different from those reported anode O–C–O and ROLi in the C 1s and O 1s spectra, respectively. Most
prelithiation methods. There is no further lithium intercalation process importantly, Li2CO3 and LiF have been revealed from Li 1s (Fig. 4e,
involved to avoid an overcharged status when our Pre-SEI graphite is ~55.5 eV for Li2CO3 and ~56.2 eV for LiF) and F 1s spectra (Fig. 4e,
coupled with those commercial Li-containing cathode materials (i.e. ~685.0 eV for LiF), which are typical inorganic components of SEI
LiFePO4, LiCoO2, LiNi0.8Co0.1Mn0.1O2, etc) for a whole battery. Inspired structure [31]. The S 2p3/2 peak at ~169.0 eV could be attributed to the
by Semi-solid flow batteries [29], a flow cell was designed to prepare our LiTFSI salts, which is in agreement with the N–S peak in the N 1s spectra
Pre-SEI graphite powder. For practical operation, the Pre-SEI graphite in Fig. 4d [32]. The lower energy binding peak (~167.0 eV) can be
was prepared by mixing graphite powder with electrolyte of lithium ion traced to the S=O bonding from the decomposition products of the
batteries and followed by an electrochemical discharge/charge process LiTFSI salts. These results strongly indicate that a SEI layer has been

591
Y. Yu et al. Carbon 196 (2022) 589–595

Fig. 2. Schematics of the flow cells design (a) and mechanism (b) for preformation of SEI layer on graphite surface. (A colour version of this figure can be
viewed online.)

Fig. 3. Morphologies of pristine graphite and Pre-SEI graphite: SEM images of pristine graphite (a–c) and Pre-SEI graphite (d-f), TEM images of pristine graphite
(g–h) and Pre-SEI graphite (i–j). (A colour version of this figure can be viewed online.)

formed over the graphite surface after the treatment in our flow cell 3.3. Electrochemical performance of Pre-SEI graphite
system. Moreover, It has been widely accepted that the SEI structures
dictate the electrochemical performance of battery materials [22,33]. The above analysis on the Pre-SEI graphite as a promising concept to
However, the SEI formation process is typically limited within the bat­ offset initial lithium loss has motivated us to experimentally confirm
tery cell, which means that the electrolyte additives for a stable SEI such possibility. Fig. 5a reveals the comparison of initial cyclic vol­
formation have to be compatible with anode and cathode materials. In tammetry (CV) curves between graphite and Pre-SEI graphite. The
this regard, our flow cell system may also be an intriguing strategy to cathodic peak of Pre-SEI graphite located around 0.5 V (vs. Li+/Li) is
artificially pre-design SEI structures outside the battery cell by simply greatly reduced compared with pristine graphite anode, which has been
modifying electrolyte components without the limitation of battery considered as the formation of SEI with the decomposition of electrolyte
materials compatibility. [34]. The reduced double layer charging in the CV curve from pre-SEI
graphite suggests the formation of the SEI layer over the graphite sur­
face because the preformed SEI layer reduces the surface area of
graphite powder. Such reduction indicates that the preformed SEI layer

592
Y. Yu et al. Carbon 196 (2022) 589–595

Fig. 4. XPS spectra acquired from Pre-SEI surface: (a) Li 1s, (b) O 1s, (c) C 1s, (d) N 1s, (e) F 1s and (f) S 2p. (A colour version of this figure can be viewed online.)

Fig. 5. Electrochemical performance of Pre-SEI graphite. (a) Initial CV curves of graphite and Pre-SEI graphite anode. (b) Initial charge and discharge curves of
graphite and Pre-SEI graphite at 0.1 C-rate, which indicate the initial capacity loss after initial cycle. Charge and discharge curves of graphite (c) and Pre-SEI graphite
(d) at different C-rates. (e) C-rate performance of Graphite and Pre-SEI graphite. (A colour version of this figure can be viewed online.)

is able to mitigate the further propagation of SEI structure. This finding loss has been compensated, thanks to the preformation of SEI layer
is well aligned with the initial charge/discharge curves of graphite and (Fig. 5b). The ICE of a pristine graphite has been improved from ~76.7%
Pre-SEI graphite half-cells, in which ~25 mAh g− 1 of the initial capacity to ~80.4% of our Pre-SEI graphite. The structure integrity of Pre-SEI

593
Y. Yu et al. Carbon 196 (2022) 589–595

graphite was further examined in half-cells with long term cycling and Declaration of competing interest
C-rate testing. These Pre-SEI graphite illustrates a stable cycling per­
formance (Figure S7, 0.1C for 100 cycles) and C-rate capability The authors declare that they have no known competing financial
(Fig. 5c–e), which is comparable with pristine graphite, suggesting a interests or personal relationships that could have appeared to influence
structure integrity after flow cell treatment. This finding is consistent the work reported in this paper.
with our XRD results (Fig. S4). Surprisingly, the Pre-SEI graphite even
demonstrates a higher capacity recovery after higher C-rate testing (2 Acknowledgements
C). It is possible that such a better capacity recovery comes from the
reduced surface damage induced by solvent co-intercalation. In tradi­ This project was partially supported by the Office of Vice Chancellor
tional EC-based electrolyte, a SEI consists of organic species and LiF, for Research, Indiana University Purdue University Indianapolis
which tends to be formed [34]. However, two inorganic components (IUPUI). The authors gratefully acknowledge the help from Post Test
include LiF and Li2CO3 have been observed as the main components Facility at Argonne National Laboratory, supported by the U.S. DOE
from the XPS results (Fig. 4a). The SEI component of Li2CO3 is capable of Vehicle Technologies office and use of Center for Nanoscale Materials, a
reducing graphite surface damage by preventing solvent U.S. Department of Energy Office of Science User Facility, as supported
co-intercalation [35]. As a result, a better capacity recovery is presented by the U.S. DOE, Office of Basic Energy Sciences, both under contract
in our Pre-SEI graphite because of a smaller surface damage at high rate number DE-AC02-06CH11357.
charge and discharge. Such interesting findings illustrate a great po­
tential of our Pre-SEI technique for an improved electrochemical per­ Appendix A. Supplementary data
formance with rational design of SEI structure. An EIS experiment was
carried out to study the interfacial properties of Pre-SEI graphite. The Supplementary data to this article can be found online at https://doi.
wetting degree of the electrolyte with electrode could be measured by org/10.1016/j.carbon.2022.05.024.
the high frequency resistance (HFR) of the cell [36,37]. It has been
found that our Pre-SEI graphite electrode demonstrates a slightly References
smaller HFR (Fig. S8, ~2.9 Ω) compared with pristine graphite (~3.0
Ω), suggesting a higher wetting degree due to the preformation of SEI [1] J.W. Choi, D. Aurbach, Promise and reality of post-lithium-ion batteries with high
energy densities, Nat. Rev. Mater. 1 (4) (2016) 16013.
layer. [2] W. Zuo, M. Luo, X. Liu, J. Wu, H. Liu, J. Li, M. Winter, R. Fu, W. Yang, Y. Yang, Li-
Next, we evaluated the Pre-SEI graphite electrochemical perfor­ rich cathodes for rechargeable Li-based batteries: reaction mechanisms and
mance in full cells by incorporating it with LiFePO4 (LFP) cathode ma­ advanced characterization techniques, Energy Environ. Sci. 13 (12) (2020)
4450–4497.
terials. The mass ratio between LFP and Pre-SEI graphite (or graphite) [3] M. Li, J. Lu, Z. Chen, K. Amine, 30 years of lithium-ion batteries, Adv. Mater. 30
was carefully controlled to 2.8:1, which was corresponding to a N/P (33) (2018) 1800561.
ratio of ~1.02 (Fig. S9). The full cell composed of LFP and Pre-SEI [4] Y. Sun, H.-W. Lee, Z.W. Seh, N. Liu, J. Sun, Y. Li, Y. Cui, High-capacity battery
cathode prelithiation to offset initial lithium loss, Nat. Energy 1 (1) (2016).
graphite cell exhibits an ICE value of 50.5% (Fig. S10), which is [5] S. Heng, L. Lv, Y. Zhu, J. Shao, W. Huang, F. Long, Q. Qu, H. Zheng, Organic salts
10.1% higher than those pristine graphite anode cells (40.4%). with unsaturated bond and diverse anions as substrates for solid electrolyte
Accordingly, the reversible cathode specific capacity with Pre-SEI interphase on graphite anodes, Carbon 183 (2021) 108–118.
[6] C. Sun, X. Zhang, C. Li, K. Wang, X. Sun, Y. Ma, Recent advances in prelithiation
graphite is 23.7% higher compared with the graphite anode cells
materials and approaches for lithium-ion batteries and capacitors, Energy Storage
(18.6 mAh g− 1) after 100 cycles at 0.1 C rate (Fig. S11). It is worth Mater. 32 (2020) 497–516.
noting that the increased capacity retention is benefited from the [7] X. Li, X. Sun, X. Hu, F. Fan, S. Cai, C. Zheng, G.D. Stucky, Review on
reduced initial lithium loss with preformed SEI on graphite anode. comprehending and enhancing the initial Coulombic efficiency of anode materials
in lithium-ion/sodium-ion batteries, Nano Energy 77 (2020) 105143.
[8] H. Xu, S. Li, C. Zhang, X. Chen, W. Liu, Y. Zheng, Y. Xie, Y. Huang, J. Li, Roll-to-roll
4. Conclusions prelithiation of Sn foil anode suppresses gassing and enables stable full-cell cycling
of lithium ion batteries, Energy Environ. Sci. 12 (10) (2019) 2991–3000.
[9] J. Jang, I. Kang, J. Choi, H. Jeong, K.W. Yi, J. Hong, M. Lee, Molecularly tailored
In summary, a novel Pre-SEI graphite concept has been reported to lithium–Arene complex enables chemical prelithiation of high-capacity lithium-ion
compensate the ICL of the lithium ion batteries, which is intrinsically battery anodes, Angew. Chem. 132 (34) (2020) 14581–14588.
different from those cathode or anode prelithiation strategies. Owing to [10] M. Diaz-Lopez, P.A. Chater, P. Bordet, M. Freire, C. Jordy, O.I. Lebedev, V. Pralong,
Li2O: Li–Mn–O Disordered rock-salt nanocomposites as cathode prelithiation
the preformation of SEI on graphite, the improvement of the ICE has additives for high-energy density Li-ion batteries, Adv. Energy Mater. 10 (7) (2020)
been confirmed in both half-cells and LFP based full cells. A higher ICE 1902788.
of Pre-SEI graphite could be further obtained by tuning electrolyte [11] Y. Sun, H.-W. Lee, Z.W. Seh, G. Zheng, J. Sun, Y. Li, Y. Cui, Lithium sulfide/metal
nanocomposite as a high-capacity cathode prelithiation material, Adv. Energy
components for SEI formation in flow cells, thus more explorations Mater. 6 (12) (2016) 1600154.
based on our Pre-SEI concept such as flow cell design and current control [12] Y. Qiao, H. Yang, Z. Chang, H. Deng, X. Li, H. Zhou, A high-energy-density and
may further enable a complete solution to fully address the ICL issue. long-life initial-anode-free lithium battery enabled by a Li2O sacrificial agent, Nat.
Energy (2021) 1–10.
The direct Li compensation of freestanding graphite powders make it
[13] Y. Bie, J. Yang, J. Wang, J. Zhou, Y. Nuli, Li2O2 as a cathode additive for the initial
easy to meet the industrial scale-up requirements. Moreover, this new anode irreversibility compensation in lithium-ion batteries, Chem Commun
concept may further allow us the pre-design of SEI structure outside the (Camb) 53 (59) (2017) 8324–8327.
battery cells with the modification of the electrolyte components, and [14] Y. Sun, Y. Li, J. Sun, Y. Li, A. Pei, Y. Cui, Stabilized Li3N for efficient battery
cathode prelithiation, Energy Storage Mater. 6 (2017) 119–124.
will be a transformative strategy to next-generation high capacity anode [15] D. Shanmukaraj, S. Grugeon, S. Laruelle, G. Douglade, J.-M. Tarascon, M. Armand,
materials with low ICE such as Si and Sn, which opens a new avenue to Sacrificial salts: compensating the initial charge irreversibility in lithium batteries,
fully unlock the energy density potential of practical Li-ion batteries. Electrochem. Commun. 12 (10) (2010) 1344–1347.
[16] J. Choi, H. Jeong, J. Jang, A.-R. Jeon, I. Kang, M. Kwon, J. Hong, M. Lee, Weakly
solvating solution enables chemical prelithiation of graphite–SiO x anodes for high-
CRediT authorship contribution statement energy Li-ion batteries, J. Am. Chem. Soc. (2021).
[17] M.G. Scott, Chemical formation of a solid electrolyte interface on the carbon
electrode of a Li-ion cell, J. Electrochem. Soc. 145 (5) (1998) 1506.
Yikang Yu: conceived the idea, prepared the materials and con­ [18] Y. Shen, X. Shen, M. Yang, J. Qian, Y. Cao, H. Yang, Y. Luo, X. Ai, Achieving
ducted electrochemical testing. Zhenzhen Yang: performed XPS mea­ desirable initial coulombic efficiencies and full capacity utilization of Li-ion
surements. Yuzi Liu: contributed to the TEM characterization. Jian Xie: batteries by chemical prelithiation of graphite anode, Adv. Funct. Mater. (2021)
2101181.
conceived the idea. All authors discussed the results and assisted during [19] H.-Y. Wang, F.-M. Wang, Electrochemical investigation of an artificial solid
manuscript preparation. electrolyte interface for improving the cycle-ability of lithium ion batteries using

594
Y. Yu et al. Carbon 196 (2022) 589–595

an atomic layer deposition on a graphite electrode, J. Power Sources 233 (2013) [29] E. Ventosa, O. Amedu, W. Schuhmann, Aqueous mixed-cation semi-solid hybrid-
1–5. flow batteries, ACS Appl. Energy Mater. 1 (10) (2018) 5158–5162.
[20] Q. Yang, W. Li, C. Dong, Y. Ma, Y. Yin, Q. Wu, Z. Xu, W. Ma, C. Fan, K. Sun, PIM-1 [30] N. Li, D. Su, In-situ structural characterizations of electrochemical intercalation of
as an artificial solid electrolyte interphase for stable lithium metal anode in high- graphite compounds, Carbon Energy 1 (2) (2019) 200–218.
performance batteries, J. Energy Chem. 42 (2020) 83–90. [31] X.Q. Zhang, X. Chen, X.B. Cheng, B.Q. Li, X. Shen, C. Yan, J.Q. Huang, Q. Zhang,
[21] Y.K. Kim, Y. Kim, J. Bae, H. Ahn, Y. Noh, H. Han, W.B. Kim, Implanting a Highly stable lithium metal batteries enabled by regulating the solvation of lithium
preferential solid electrolyte interphase layer over anode electrode of lithium ion ions in nonaqueous electrolytes, Angew Chem. Int. Ed. Engl. 57 (19) (2018)
batteries for highly enhanced Li+ diffusion properties, J. Energy Chem. 48 (2020) 5301–5305.
285–292. [32] V. Sharova, A. Moretti, T. Diemant, A. Varzi, R.J. Behm, S. Passerini, Comparative
[22] Z. Hou, J. Zhang, W. Wang, Q. Chen, B. Li, C. Li, Towards high-performance study of imide-based Li salts as electrolyte additives for Li-ion batteries, J. Power
lithium metal anodes via the modification of solid electrolyte interphases, Sources 375 (2018) 43–52.
J. Energy Chem. 45 (2020) 7–17. [33] Y. Yu, Y. Liu, J. Xie, Building better Li metal anodes in liquid electrolyte: challenges
[23] Y. Li, Y. Lu, P. Adelhelm, M.M. Titirici, Y.S. Hu, Intercalation chemistry of graphite: and progress, ACS Appl. Mater. Interfaces 13 (1) (2020) 18–33.
alkali metal ions and beyond, Chem. Soc. Rev. 48 (17) (2019) 4655–4687. [34] S.J. An, J. Li, C. Daniel, D. Mohanty, S. Nagpure, D.L. Wood, The state of
[24] S. Heng, X. Shan, W. Wang, Y. Wang, G. Zhu, Q. Qu, H. Zheng, Controllable solid understanding of the lithium-ion-battery graphite solid electrolyte interphase (SEI)
electrolyte interphase precursor for stabilizing natural graphite anode in lithium and its relationship to formation cycling, Carbon 105 (2016) 52–76.
ion batteries, Carbon 159 (2020) 390–400. [35] S. Bhattacharya, A.R. Riahi, A.T. Alpas, Electrochemical cycling behaviour of
[25] S. Wu, H. Wu, M. Zou, X. Shi, Y. Yuan, W. Bai, A. Cao, Short-range ordered lithium carbonate (Li2CO3) pre-treated graphite anodes – SEI formation and
graphitized-carbon nanotubes with large cavity as high-performance lithium-ion graphite damage mechanisms, Carbon 77 (2014) 99–112.
battery anodes, Carbon 158 (2020) 642–650. [36] D.S. Kim, Y.E. Kim, H. Kim, Improved fast charging capability of graphite anodes
[26] S. Flandrois, B. Simon, Carbon materials for lithium-ion rechargeable batteries, via amorphous Al2O3 coating for high power lithium ion batteries, J. Power
Carbon 37 (2) (1999) 165–180. Sources 422 (2019) 18–24.
[27] Y. Sun, H.W. Lee, Z.W. Seh, G. Zheng, J. Sun, Y. Li, Y. Cui, Lithium sulfide/metal [37] F.J. Günter, J.B. Habedank, D. Schreiner, T. Neuwirth, R. Gilles, G. Reinhart,
nanocomposite as a high-capacity cathode prelithiation material, Adv. Energy Introduction to electrochemical impedance spectroscopy as a measurement method
Mater. 6 (12) (2016) 1600154. for the wetting degree of lithium-ion cells, J. Electrochem. Soc. 165 (14) (2018)
[28] Y. Zhan, H. Yu, L. Ben, Y. Chen, X. Huang, Using Li2S to compensate for the loss of A3249.
active lithium in Li-ion batteries, Electrochim. Acta 255 (2017) 212–219.

595