Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (19.64 MB, 415 trang )
material and decrease cycling stability. The stability of the delithiated phase (FePO4)
contributes to the extraordinary cycling stability of lithium iron phosphate. The diffusion-driven proliferation of the phases determines the characteristics only under
extremely high currents (> 20 C)  but, at high temperatures (> 50 °C), LFP exhibits better charging and discharging characteristics than the materials discussed above.
However, the poor electrical and lithium-ion conductivity is the biggest problem
of Li1–xFePO4. Smaller crystallites should reduce the distance for lithium-ion diffusion and also decrease the distance the electrons need to travel through the material.
This is why many technologies have been developed to manufacture LFP particles
in the submicron or even nanometer range. Usually, lithium iron phosphates mostly
consist of primary particles with a diameter of ~200 nm that agglomerate into secondary particles with a diameter of 5 to 10 μm. These nanoparticles make capacities
of > 100 mAh/g at high C rates (> 20 C) possible [40, 41].
Scaling down the particles compensates for the lithium-ion conductivity. However,
it does not improve electrical conductivity. Poor electrical conductivity of a cathode
material leads to poor cell performance. This is the reason why improving electrical
conductivity has become one of the key goals of research into LFP. Carbon coating,
for example, can significantly improve LFP conductivity. This technology enables
capacities of up to 165 mAh/g (97 % of the theoretical capacity) and improves the
charging and discharging properties at high currents . The dependence of electrical conductivity on electrochemical characteristics makes the quality of carbon
coating very important and a co-determinant for the cathode material’s qualities.
During production, the coating also prevents the aggregation of particles and protects the cathode material from inadvertent oxidation and degradation .
There is no evidence that oxygen evolves in LFP compared to oxidic materials.
This is due to the strong covalent bonds within the phosphate molecule . Thus,
LFP has an excellent thermal stability and is compatible with all common electrolytes .
In 1997, in the first publication , it was already shown that the olivine structure is also stable for other lithium transition metal phosphates. Incorporation of
manganese, cobalt, or nickel enables higher working voltages and higher energy
densities. Lithium manganese phosphate (LMP) Li1–xMnPO4 offers a working
voltage of 4.1 V vs. Li/Li+ . This is equal to an energy density of 656 Wh/kg
with the same theoretical capacity as LFP. LMP has a flat plateau, similar to LFP. In
contrast to the spinels (LiMn2O4), the manganiferous olivines do not exhibit JahnTeller instability of the delithiated material and are extremely well cyclable .
But Li1–xMnPO4 possesses a weaker electrical conductivity than the iron analog,
thus requiring considerably smaller particle sizes (< 80 nm) . Current developments of phosphate-based cathode materials favor mixed iron-manganese phosphates with > 60 % manganese (LFMP) .
For olivines, the voltage plateaus of Co3+/Co2+ and Ni3+/Ni2+ are around 4.8 V vs.
Li/Li+  and 5.1 V vs. Li/Li+, respectively . Most commercially available
electrolytes disintegrate from around 4.3 V, as mentioned for high-voltage spinels.
Therefore, new electrolytes need to be developed to allow the usage of 5-V olivines
in cells with graphite anodes. But, in contrast to LFP, heavy metal ions pose a health
4 Cathode materials for lithium-ion batteries39
Comparison of cathode materials
Cathode materials are suitable for different applications with different requirements.
This is due to their specific characteristics. But the decision as to which cathode
material to use does not only depend on the electrochemical data (Table 4.1). It also
depends on the costs, the service life, and the safety characteristics (Fig. 4.7).
Table 4.1 Capacity, working voltage, and energy density of the discussed cathode materials
Energy density/Wh kg–1
Fig. 4.7 Comparison of cathode materials pertaining to safety, specific energy, specific power,
service life, costs, and fast chargeability (based on )
The layered oxides are advantageous in applications where a high energy density
is of importance. Their high working voltages and resulting energy densities are
explained by the redox properties of the nickel, cobalt, and manganese ions. In
the lithosphere, these metals are very scarce, resulting in high prices on the global
market (Co: USD 26 kg–1, Ni: USD 10 kg–1, Mn: USD 1.7 kg–1, as of August 2016).
Iron is the second most common element on Earth and, even when compared with
manganese, extremely inexpensive (USD 0.057/kg–1). (Source: www.InfoMine.
com) Thus, LFP is the cheapest possibility for producing 1 kg of cathode material.
Due to its lower energy density in comparison with NCM and NCA, this advantage
is less pronounced.
Oxides pose a safety risk, as they can trigger the evolution of oxygen, thus causing
a fire or even an explosion of the cell. There is virtually no risk of this occurring
for LFP due to its intrinsic characteristics. LFP shows no thermal effects up to
300 °C as opposed to other cathode materials. Almost all oxidic cathode materials,
however, exhibit strong exothermal effects. Furthermore, LFP and LFMP are nontoxic and non-hazardous to the environment, as is LMO. This is not the case for
heavy metal-containing oxidic compounds.
Lithium diffusion is possible in one direction with LFP, while the other materials
have two-dimensional, if not even three-dimensional, lithium diffusion paths. Nevertheless, LFP is the only cathode material that can be quickly charged and discharged.
It has not yet reached the high energy (Wh/kg) and power densities (W/kg) of oxidic
cathode materials. These values can be increased by 20 % in relation to LFP by using
manganese (LFMP). Nevertheless, at this stage, LFP is a safe and resilient cathode
material for usage in stationary and (plug-in) HEV applications.
Oxides have been a focal point of research and development since the Eighties.
They have been modified many times, resulting in materials that can be produced
on a large scale and are commercially successful. The development of the relatively
“young” phosphate-based cathode materials is still in its infancy. It is possible to
achieve higher working voltages with higher energy densities by substituting manganese, cobalt, or nickel for iron. This makes phosphate-based compounds very
promising cathode materials for future high-voltage applications.
Akimoto J, Gotoh Y, Oosawa Y (1998) J Solid State Chem 141:298
Wang HF, Yang YI, Huang BY, Sadoway DR, Chiang YT (1999) J Electrochem Soc 146:473
Ohzuku T, Brodd RJ (2007) J Power Sources 174:449
Ohzuku T, Ueda A (1994) Solid State Ionics 69:201
Yuan LX et al (2011) Goodenough. Energy Environ Sci 4:269
Wang GX et al (2001) J Power Sources 97−98:298
Molenda J, Marzec J (2009) Funct Mater Lett 3:1
Amatucci GG, Tarascon JM, Klein LC (1996) Solid State Ionics 83:167
Breuer H (2000) dtv-Atlas Chemie, Vol. 1, 9th edition. dtv, München
Dahn JR, Vonsacken U, Michal CA (1990) Solid State Ionics 44:87
Molenda J, Marzec J (2003) Funct Mater Lett 115:115
Rougier A, Gravereau P, Delmas C (1996) J Electrochem Soc 143:1168
4 Cathode materials for lithium-ion batteries41
13. Pouilliere C, Croguennec L, Biensan P, Willmann P, Delmas C (2000) J Electrochem Soc
14. Z. Lu, Macneil DD, Dahn JR (2004) Electrochem Solid-State Lett 14:A191
15. Naghash AR, Lee JY (2001) Electrochim Acta 45:2293
16. Park SH, Sun YK, Park KS, Nahm KS, Lee YS, Yoshio M (2002) Electrochim Acta 41:1721
17. Mishra SK, Ceder G (1999) Phys Rev B 59:6120
18. Armstrong AR, Bruce PG (1996) Nature 381:499
19. Ceder G, Van der Ven A (1999) Electrochim Acta 45:131
20. Makimura Y, Ohzuku T (2003) J Power Sources 119:156
21. Koyama Y, Tanaka I, Adahi H, Makimura Y, Ohzuku T (2003) J Power Sources 119:644
22. Hwang BJ, Tsai YW, Carlier D, Ceder G (2003) Chem Mater 15:3676
23. Wang L, Li J, He X, Pu W, Wan C, Jiang C (2009) J Solid State Electrochem 13:1157
24. Yoon WS, Paik Y, Yang XQ, Balasubramanian M, McBreen J, Grey CP (2002) Elektrochem
Solid-State Lett 5:A263
25. Park OK, Cho Y, Lee S, Yoo H-C, Song H-K, Cho J (2011) Energy Environ Sci 4:1621
26. Ellis BL, Lee KT, Nazar LF (2010) Chem Mater 22:691
27. Gnanaraj JS, Pol VG, Gedanken A, Aurbach D (2003) Electrochem Commun 5:940
28. Thackeray MM, Dekock A, Rossouw MH, Liles D, Bittuhn R, Hoge D (1992) J Electrochem
29. Benedek R, Thackeray MM (2006) Eectrochem Solid-State Lett 9:A265
30. Cho J, Thackeray MM (1999) J Electrochem Soc 146:3577
31. Cho J (2008) J Mater Chem 18:2257
32. Thackeray MM, Shao-Horn Y, Kahaian AJ, Kepler KD, Vaughey JT, Hackney SA (1998)
Electrochem Solid-State Lett 1:7
33. Shin YJ, Manthiram A (2994) J Electrochem Soc 151: A208
34. Deng BH, Nakamura H, Yoshio M (2008) J Power Sources 180:864
35. Xia YG, Zhang Q, Wang HY, Nakamura H, Noguchi H, Yoshio M (2007) Electrochim Acta
36. Kim DK, Muralidharan P, Lee HW, Ruffo R, Yang Y, Chan CK, Peng H, Huggins RA, Cui Y
(2008) Nano Lett 8:3948
37. Liu GQ, Wen L, Liu YM (2010) J Solid-State Electrochem 14:2191
38. Padhi AK, Nanjundaswamy KS, Goddenough JB (1997) J Electrochem Soc 144:1188
39. Morgan D, Van der Ven A, Ceder G (2004) Electrochem Solid-State Lett 7:A30
40. Wang YG, Wang YR, Hosono EJ, Wang KX, Zhou HS (2008) Angew Chem Int Ed 47:7461
41. Kang B, Ceder G (2009) Nature 458:190
42. Ravet N, Chouinard Y, Magnan JF, Besner S, Gauthier M, Armand M (2001) J Power Sources
43. Koltypin M, Aurbach D, Nazar L, Ellis B (2007) Electrochem Solid-State Lett 10:A40
44. MacNeil DD, Lu ZH, Chen ZH, Dahn JR (2002) J Power Sources 108:8
45. Wang DY et al (2009) J Power Sources 189:624
46. Drezen T, Kwon NH, Miners JH, Poletto L, Graetzel M (2007) J Power Sources 174:949
47. Yamada A, Takei Y, Koizumi H, Sonoyama N, Kanno R (2006) Chem Mater 18:804
48. Amine K, Yasuda H, Yamachi M (2000) Electrochem Solid State Lett 3:178
49. Zhou F, Cococcioni M, Kang K, Ceder G (2004) 6:1144
50. Geoffroy D (2012) Phosphates
Anode materials for lithium-ion batteries
Călin Wurm, Oswin Oettinger, Stephan Wittkaemper,
Robert Zauter, and Kai Vuorilehto
5.1 Anode active materials – introduction������������������������������������������������������������������������������ 44
5.2Production and structure of amorphous carbons and graphite ���������������������������������������� 45
5.3 Lithium intercalation in graphite and amorphous carbons���������������������������������������������� 47
5.4Production and electrochemical characteristics of C/Si or C/Sn components ���������������� 52
5.5 Lithium titanate as anode material ���������������������������������������������������������������������������������� 53
5.6 Anode active materials – outlook ������������������������������������������������������������������������������������ 54
5.7 Copper as conductor at the negative electrode ���������������������������������������������������������������� 55
5.7.1 Requirements for the copper foil ������������������������������������������������������������������������ 55
5.7.2Comparison of RA copper foil with electrolytically produced copper foil �������� 56
5.7.3 Substitution of aluminum for copper? ���������������������������������������������������������������� 57
C. Wurm (*)
Robert Bosch Battery Systems GmbH, Heilbronner Strasse 358-360,
70469 Stuttgart, Germany
SGL Carbon GmbH, Werner-von-Siemens-Strasse 18, 86405 Meitingen, Germany
GTS Flexible Materials GmbH, Hagener Strasse 113, 57072 Siegen, Germany
Wieland-Werke AG, Graf-Arco-Strasse 36, 89079 Ulm, Germany
Aalto University Helsinki, Kemistintie 1, 02150 Espoo, Finland
© Springer-Verlag GmbH Germany, part of Springer Nature 2018
R. Korthauer (ed.), Lithium-Ion Batteries: Basics and Applications,
C. Wurm et al.
Anode active materials – introduction
Secondary lithium cells initially had a metallic lithium foil as an anode (negative
electrode) . Pure lithium has a very high specific capacity (3,860 mAh/g) and a
very negative potential, resulting in very high cell voltage. However, cycling efficiency decreases as lithium dissolves repeatedly while the cell is discharging and
lithium is deposited as it is charging. This means that two or three times the normal
amount of lithium must be used. In addition, lithium can be deposited as foam and
as dendrites. The latter might grow through the separator [2, 3]. These dendrites can
cause local short circuits, which might result in the cell completely self-discharging
or, in the worst case, lead to an internal thermal chain reaction, fire, or explosion.
Today, only small cells (especially button cells) are mass-produced. They meet
low requirements in regard to cycling stability and fast chargeability with lithium
metal anodes. Recently, however, trials with high-capacity cells have been resumed.
Special separators (cf. Bolloré ) are used in these trials. The most important issues
for these LMP (lithium metal polymer) cells are still safety and cycling stability.
To produce safe cells with a good cycling efficiency, lithium metal is replaced by
a so-called lithium intercalation material [5–8]. Generally, the intercalation process,
e.g., in carbon, is a reversible and loss-free process, and lithium plating does not
occur [9, 10]. The lithium-ion cell of the typical 3C market (portable consumer
applications) is equipped with anodes made of graphite. There is an increasing
focus on amorphous carbons (hard carbons and soft carbons) in the new applications with higher power and energy density as well as improved safety. In part, they
Fig. 5.1 Specific capacity and potential vs. Li/Li+ of the most important anode materials
[according to 11]
5 Anode materials for lithium-ion batteries45
exhibit a better current capability and are safer and more stable in combination with
novel electrolytes and cathode materials.
Their lithium storage capacity is much higher than that of graphite thanks to the
usage of metals and alloys (intermetallic compounds) that can reversibly react with
lithium. Metal-based systems are not yet being mass-produced in spite of intensive
R&D efforts. So there is still a substantial need for research. Even if adding carbons
(e.g., C/Si composite, C/Sn composite) enables improvement, cycling stability
remains insufficient. Lithium titanate and titanium oxide are promising as anode
active materials for improving cycling stability and meet extraordinarily high power
and safety requirements. However, the specific capacity of these materials is very
low and the potential vs. lithium is very high (Fig. 5.1).
The electrochemical characteristics of these materials play an important role. In
addition, good processability (rheology characteristics, adhesion on metal foils,
etc.) is a must during electrode production. Fig. 5.1 provides an overview of the
specific capacity and the potential for the most important anode materials .
The following sections will provide information on the different anode materials.
5.2Production and structure of amorphous
carbons and graphite
Amorphous carbon (hard and soft carbons) and graphite both occur naturally, and
synthetic synthesis of both is possible. A typical representative of naturally occurring amorphous carbon is anthracite. Naturally occurring graphite is called natural
graphite. The largest natural graphite deposits are in Asia, mainly in China, but also
in India and North Korea. In the Western hemisphere, natural graphite is mainly
mined in Brazil and Canada. It is assumed that ca. 70 to 80 % of all natural graphite
is deposited in China. For this reason, the European Union Raw Materials Initiative
included natural graphite in its list of 14 critical raw materials in 2010. Natural
graphite needs to be separated from the lode matter to make it utilizable for battery
applications. It also needs to be purified chemically, thermally, and/or both.
The starting materials for synthesized amorphous carbon and graphite are usually
byproducts of the coal and petroleum industry. In the latter, the most common ma
terials are petroleum coke and resins with a high content of aromatic compounds,
e.g., phenolic resins. In the coal industry, coal tar pitch and high-isotropic coke are
the most important raw materials. The starting materials are carbonized or calcined
at low temperatures (800 °C to 1,200 °C). The first product is amorphous carbon. If
the carbonized starting material (e.g., furan resin or phenolic resin) remains amorphous during the subsequent heat treatment up to 3,000 °C, it is called hard or
non-graphitizable carbon. If the carbonized starting material (e.g., coke or pitch)
transforms into graphite above 2,500 °C, it is called soft or graphitizable carbon.
Fig. 5.2 provides an overview of the typical production process of synthetic
graphite. The main raw materials are calcined coke as solid material and coal tar
pitch as binder. The solid materials are preprocessed (ground, sieved, screened, and
classified). The material is then mixed with molten, liquid binder pitch to produce
C. Wurm et al.
Fig. 5.2 Schematic diagram of synthetic graphite production
a plasticized “green” mass, which subsequently is shaped by extrusion or compression molding, for example.
The solidified green body is then fired at 800 °C to 1,200 °C in the absence
of oxygen. The pitch carbonizes, creating amorphous carbon as binder phase. To
transform amorphous carbon into synthetic graphite, the body is introduced into the
so-called packing material and treated at > 2,500 °C. Contemporary graphitization
technologies date back to Acheson or Castner [12, 13] and were invented at the
end of the 19th century. The typical amount of electricity needed to produce 1 kg
graphite in the Acheson furnace is 3 to 4 kWh and the required time in the furnace is
around three weeks. The high temperatures cause the contamination to escape from
the graphite, and the typical graphite layer structure develops (Figs. 5.3a and b).
Graphite is one of the best-known carbon allotropes. It is made of parallelly stacked graphene layers. Graphene is a hexagonal lattice of sp2-hybridized
carbon atoms. The widespread hexagonal form of graphite has the stack sequence
ABABAB (Fig. 5.3b). The distance between the graphene layers is 0.3354 nm.
The rhombohedral graphite modification with the stack sequence ABCABC is of
lesser importance; its percentage can reach 20 % during forming processes such
as graphite grinding. It is possible to reduce this rhombohedral ratio again with
a high-temperature treatment . The crystallographic density of both graphite
forms is 2.26 g/cm3.
Many times, additional finishing processes are employed depending on the application to use graphite optimally as an active material (Fig. 5.4). The final particle
size and shape are modified after graphite grinding (Fig. 5.4a). This ensures that the
specific surface area is as small and the surface morphology as smooth as possible