14
1 Introduction
oxides [49, 50]. A better understanding of these layered structures, as determined in
Chap. 8, will help explain these discrepancies VII.
The final source of confusion in the literature relating to the lithium-rich structures
involves the use of excess lithium during synthesis. As discussed here, the samples
lose some lithium during synthesis and so it has become common practice to make
lithium-rich layered oxides with a small amount of excess lithium (e.g. 5 %) and to
assume the excess burns off during heating [51, 52]. It will be shown here that this is
not always the case and some of this excess lithium can be tolerated in the structures
such that they may lie below line II in Fig. 1.8. Lattice parameter contour plots prove
invaluable in determining the actual compositions of such published samples and
this will help explain why materials seemingly synthesized at the same composition
have very different electrochemistry (e.g., LiNi 0.5 Mn 0.5 O 2 [2, 52, 53]) VIII.
1.4.3 The Coexistence Region Between the Spinel and Layered
Structures
There are a few particularly strange results in the literature for samples made at the
composition of LiNi 0.5 Mn 0.5 O 2 . The first is that the XRD peaks of LiNi 0.5 Mn 0.5 O 2
are very broad when synthesized in oxygen at 800
◦ C. This broadening of the peaks
was interpreted by Jo et al. [54] as being due to small crystallite sizes of about 14 nm.
However, this is a small value for a sample made at these temperatures and so it is
important to consider the possibility that this broadening is due to multiple phases
with slightly different lattice parameters. By looking at samples over wide composition ranges and various synthesis conditions it is possible to distinguish between
these two possibilities and so one objective of this thesis is to better understand the
observations of Jo et al. [54] IX. Figure 1.12 shows an even more peculiar observation made in a sample of LiNi 0.5 Mn 0.5 O 2 by Hinuma et al. [55]. This sample,
made by ion exchange and heating to 1000
◦ C (pristine), showed a strange phase
transformation when annealed at 600
◦ C. The extra peaks found after annealing were
not identified and the sample returned to its single-phase layered structure when
annealed at 1000
◦ C and quenched back to room temperature. This dramatic change
between high and medium temperature behaviour was seen repeatedly during the
current study when comparing quenched samples to those obtained by slow cooling.
Thus, another objective of this study is to explain the results seen by Hinuma et al.
[55] X.
Finally, another significant motivation for this research is the recent interest in
composite electrodes combining spinel and layered structures in the Li–Mn–Ni–O
system [13, 14, 56, 57]. In such studies, the samples are assumed to be made up of
phases along lines I and II in Fig. 1.8, with the occasional presence of rocksalt contamination (this turns out to be the same rocksalt contaminant as seen in the spinel
samples and will be identified in Chap. 5). Figure 1.13 shows the XRD patterns obtained by Cabana et al. [13]. The results clearly show spinel and layered coexistence.
However, without clearly knowing the boundaries of the single-phase regions, nor the
1 Introduction
oxides [49, 50]. A better understanding of these layered structures, as determined in
Chap. 8, will help explain these discrepancies VII.
The final source of confusion in the literature relating to the lithium-rich structures
involves the use of excess lithium during synthesis. As discussed here, the samples
lose some lithium during synthesis and so it has become common practice to make
lithium-rich layered oxides with a small amount of excess lithium (e.g. 5 %) and to
assume the excess burns off during heating [51, 52]. It will be shown here that this is
not always the case and some of this excess lithium can be tolerated in the structures
such that they may lie below line II in Fig. 1.8. Lattice parameter contour plots prove
invaluable in determining the actual compositions of such published samples and
this will help explain why materials seemingly synthesized at the same composition
have very different electrochemistry (e.g., LiNi 0.5 Mn 0.5 O 2 [2, 52, 53]) VIII.
1.4.3 The Coexistence Region Between the Spinel and Layered
Structures
There are a few particularly strange results in the literature for samples made at the
composition of LiNi 0.5 Mn 0.5 O 2 . The first is that the XRD peaks of LiNi 0.5 Mn 0.5 O 2
are very broad when synthesized in oxygen at 800
◦ C. This broadening of the peaks
was interpreted by Jo et al. [54] as being due to small crystallite sizes of about 14 nm.
However, this is a small value for a sample made at these temperatures and so it is
important to consider the possibility that this broadening is due to multiple phases
with slightly different lattice parameters. By looking at samples over wide composition ranges and various synthesis conditions it is possible to distinguish between
these two possibilities and so one objective of this thesis is to better understand the
observations of Jo et al. [54] IX. Figure 1.12 shows an even more peculiar observation made in a sample of LiNi 0.5 Mn 0.5 O 2 by Hinuma et al. [55]. This sample,
made by ion exchange and heating to 1000
◦ C (pristine), showed a strange phase
transformation when annealed at 600
◦ C. The extra peaks found after annealing were
not identified and the sample returned to its single-phase layered structure when
annealed at 1000
◦ C and quenched back to room temperature. This dramatic change
between high and medium temperature behaviour was seen repeatedly during the
current study when comparing quenched samples to those obtained by slow cooling.
Thus, another objective of this study is to explain the results seen by Hinuma et al.
[55] X.
Finally, another significant motivation for this research is the recent interest in
composite electrodes combining spinel and layered structures in the Li–Mn–Ni–O
system [13, 14, 56, 57]. In such studies, the samples are assumed to be made up of
phases along lines I and II in Fig. 1.8, with the occasional presence of rocksalt contamination (this turns out to be the same rocksalt contaminant as seen in the spinel
samples and will be identified in Chap. 5). Figure 1.13 shows the XRD patterns obtained by Cabana et al. [13]. The results clearly show spinel and layered coexistence.
However, without clearly knowing the boundaries of the single-phase regions, nor the