networks of lithium and cobalt and that the battery-related application
contained significant potential lithium stocks. Hao et al. (2017) ana-
lyzed the anthropogenic cycle of lithium in China in 2015 using a static
MFA model and showed that the growth of electric vehicles with a
critical dependence on LIB technology would greatly increase China's
lithium demand. Zeng and Li (2015) assessed the sustainability of co-
balt reserves in China through a top-down MFA model and found that
the rapid development of CEs and EVs resulting in the shortage supply
of cobalt. This overview shows the importance of LIBs in the anthro-
pogenic cycles of lithium and cobalt. However, these studies usually
focus on one kind of element from LIBs at the global or national level,
and rarely MFA models have been used to analyze the difference and
connections of multiple materials in an application (Weil et al., 2018).
In addition, despite the rapid development of the LIB industry in China,
many studies usually employ a static top-down MAF model
(Harper et al., 2011 ; Yan et al., 2017). Even though few MFA studies
use a dynamic model, dynamic parameters such as battery life, battery
capacity, and battery recycling rate are rarely used in the model
(Song et al., 2019).
Therefore, this study aims to fill this gap and analyze the evolution
of the industrial metabolism of lithium and cobalt for LIBs in China
from 2000 to 2018 using a dynamic bottom-up MFA model. Many dy-
namic parameters were used in the dynamic MFA model to simulate the
evolution of battery production and use and related resource metabo-
lism. Based on the flows and stocks of LIBs and their quantities of li-
thium and cobalt, the evolution of production, use, EOL, and recycling
of LIBs in China were analyzed, and the driving factors of the evolution
of anthropogenic cycles of lithium and cobalt were also discussed.
Furthermore, the implications for sustainable development of LIB in-
dustries in China were analyzed.
2. Methods and data
2.1. Material flow analysis framework
The target system of MFA is defined by the spatial and temporal
boundaries (Müller et al., 2014). The spatial boundary in this study was
the mainland of China, excluding Taiwan, Hong Kong, and Macau. The
temporal boundary was defined as the period from 2000 to 2018 be-
cause LIBs experienced a change in the development process from
scratch during this period.
The LIB system was the “cradle to grave” scope in this study, re-
ferring to the entire life cycle of LIBs, starting from raw material pro-
duction, production and use of LIBs, and recycling. Retrospective MFA
was used to calculate the flows and stocks of lithium and cobalt within
the LIB system of China. From the life cycle perspective, the life span of
the LIB system was divided into five stages, namely mining, refining,
manufacturing, use, and waste management (including echelon utili-
zation and material recovery), as shown in Fig. 1.
2.1.1. Mining
China has the most abundant lithium resources worldwide
(Song et al., 2019). There are two types of natural lithium resources,
namely ore and brine. Although brine accounts for 80.5% of the lithium
reserves in China, more than 90.0% of domestic lithium production
comes from ores (USGS, 2018). Owing to the high proportion of mag-
nesium and lithium in salt lakes in China, it is difficult to extract li-
thium. Cobalt resources are scarce in China, and cobalt is usually mined
as a by-product of nickel, copper, and other metals. Therefore, China's
lithium and cobalt resources are heavily dependent on imports. China
imports two types of raw cobalt materials, namely cobalt ore and in-
termediate products of the wet metallurgy process.
2.1.2. Refining
Metal minerals are converted to chemical compounds for LIB pro-
duction in this stage. There are numerous materials after refining, such
as metal carbonate, metal hydroxide, and metal chloride. However,
production information is not available for all of these materials. For
example, lithium carbonate is mainly used to produce LIB cathode
materials, including LiCoO
2
(LCO), LiMnO
2
(LMO), LiNi
x
Mn
y
Co
1-x-y
O
2
(NCM), and LiFePO
4
(LFP), among others. Therefore, only lithium
carbonate is considered as the output of the lithium mineral refining
stage.
2.1.3. Manufacturing
The manufacturing stage refers to the LIB production process and
the assembly process of LIB-driven products. Commercially, LIBs typi-
cally differ by the material of the cathode, with graphite used as the
anode. LIB materials are constantly updated to meet evolving demands.
From past to present, LIBs have changed from early LCO, LMO, and LFP
to NCM/NCA (LiNiCoAlO
2
) batteries. The metal compositions of the
four types of LIBs are shown in Table S5.
2.1.4. Use
The main uses of LIBs include CEs, EVs, and ESS. With the wide-
spread use of mobile phones and laptops, CEs require batteries with
higher energy density. LIBs have gradually replaced nickel-cadmium
and nickel-hydrogen batteries as the mainstream batteries in CEs be-
cause of the high specific energy and no memory effect (Espinoza et al.,
2014). Meanwhile, the fast-growing electric automobiles consume a
large number of LIBs. In addition, in the market of electric bicycles,
electric tricycles, renewable energy storage systems in centralized and
distributed wind power and photovoltaic power stations, and unin-
terruptable power supply (UPS) in communication base stations, banks,
and post offices, lead-acid batteries (LABs) are being replaced by LIBs
due to the lower specific energy density and the risk of lead pollution
(Liu et al., 2016). In this study, mobile phones, laptops, tablets, digital
cameras and power banks are selected to represent CEs. EVs are divided
into electric automobiles, electric bicycles, and electric tricycles. Elec-
tric automobiles are further divided into all-electric passenger cars,
buses, special vehicles, and plug-in hybrid passenger cars and buses.
ESS includes renewable energy storage systems and UPS. The actual
service life of LIBs in different end-use products is obtained through the
investigation, as shown in Table S6.
2.1.5. Waste management
There are three directions for EOL LIBs, namely echelon utilization,
material recovery, and storage or disposal. When the battery capacity of
a vehicle decays to below 80% of the original battery capacity, it is
difficult to meet the requirements of EVs, but can be used in areas with
low battery performance, such as ESS, which is the echelon utilization
of power batteries (Han et al., 2019). At present, the batteries reused by
echelon utilization in China are mainly LFP batteries retired from
electric automobiles (Wang and Zhao, 2015). These batteries can con-
tinue to be used as stationary batteries for 5 years (Ji and Zhou, 2019).
However, many LIBs are stored by consumers at the end of their service
life due to the lack of an effective collection system (Ichiro DAIGO,
2007). In addition, some EOL LIBs are disposed of through landfilling
and incineration as municipal solid waste (Song et al., 2017). This study
treated both as the loss because the amount of storage and disposal of
EOL LIBs could not be quantified due to lack of the relevant data.
Therefore, this study assumed that uncollected batteries were lost in
storage or disposal as municipal solid waste by incineration and land-
filling. More information about the recycling of LIBs can be found in the
Supporting Information (SI).
2.2. Calculation of stocks and flows
Based on the principles of MFA, calculations describing each process
should obey the mass conservation law (Müller et al., 2014). This study
uses a dynamic bottom-up MFA to calculate the stocks of LIBs and
metals in LIBs, which are described in Fig. 2.
W. Liu, et al.
Resources, Conservation & Recycling 164 (2021) 105122
2