9.2 Experimental Design
119
However, the XRD patterns in the Lin paper are not sufficiently detailed to determine
whether or not peak broadening was present in the regular cooled sample. Part of
the current study therefore deals with reproducing these data in order to determine
whether peak broadening is associated with this decrease in first cycle capacity.
It was also necessary to identify the features in the XRD patterns which are most
sensitive to phase separation and then search for them carefully. In Chap. 4, in the
Li–Co–Mn–O system, peak broadening at high angles (>60
◦ ) in the XRD patterns
was found to first indicate the formation of layered–layered composites and this
corresponded to synthesis conditions where Wen et al. [22] found nano-composites.
Upon slower cooling of the combinatorial samples described in Chap. 4, the broad
peaks eventually resolved into two separate peaks corresponding to each layered
phase such that the formation of nano-composites can be viewed as an incomplete
phase separation into a layered–layered two-phase structure. To our knowledge, this
was the first time that features in XRD patterns were correlated to nano-scale phase
separation. Similar careful analysis of XRD patterns will be applied to samples in
the Li–Mn–Ni–O system here. Therefore, the condition used to identify the presence
of nano-composites in the XRD pattern was the broadening of the high angle peaks.
However, to distinguish this from microstrain peak broadening, a necessary condition
to identify the formation of a nano-composite is that clear phase separation must be
seen in the XRD if conditions are exaggerated (e.g., if cooling rates are slowed down
further as done in Chap. 4 or if oxygen partial pressures are increased as used here).
9.2 Experimental Design
In order to obtain a sample showing these first signs of phase separation, the boundaries of the single-phase region were varied by changing the oxygen partial pressure
during heating. Figure 9.1 shows the upper layered boundary under various conditions. The boundaries for samples heated in air were presented in previous chapters
while the approximate boundaries for 2 % oxygen are based on the current study and
so should only be considered accurate near compositions A 9 and B 9 . The boundary
moves upward as oxygen partial pressure decreases. Therefore, a sample made in
pure oxygen at composition A 9 will be multiphase if quenched while the same composition in 2 % oxygen will be single phase. As will be shown in the results section,
the regular cooled sample made at composition A 9 in 2 % oxygen showed the signs of
being a layered–layered nano-composite. All samples made by the one-pot synthesis
method at composition A 9 were heated at the rate of 5
◦ C/min up to 800
◦ C and held
there for 5 h. The atmospheres used during heating were either air, or a flow of gas
with either 2, 5.5, or 100 % oxygen content with the remainder being either nitrogen
or argon gas. The samples were then either quenched or regular cooled. Samples
at the composition B 9 were made by Aaron Rowe and Eric McCalla using the tank
reactor method. Samples were heated at 900
◦ C for 12 h before either quenching or
regular cooling.
119
However, the XRD patterns in the Lin paper are not sufficiently detailed to determine
whether or not peak broadening was present in the regular cooled sample. Part of
the current study therefore deals with reproducing these data in order to determine
whether peak broadening is associated with this decrease in first cycle capacity.
It was also necessary to identify the features in the XRD patterns which are most
sensitive to phase separation and then search for them carefully. In Chap. 4, in the
Li–Co–Mn–O system, peak broadening at high angles (>60
◦ ) in the XRD patterns
was found to first indicate the formation of layered–layered composites and this
corresponded to synthesis conditions where Wen et al. [22] found nano-composites.
Upon slower cooling of the combinatorial samples described in Chap. 4, the broad
peaks eventually resolved into two separate peaks corresponding to each layered
phase such that the formation of nano-composites can be viewed as an incomplete
phase separation into a layered–layered two-phase structure. To our knowledge, this
was the first time that features in XRD patterns were correlated to nano-scale phase
separation. Similar careful analysis of XRD patterns will be applied to samples in
the Li–Mn–Ni–O system here. Therefore, the condition used to identify the presence
of nano-composites in the XRD pattern was the broadening of the high angle peaks.
However, to distinguish this from microstrain peak broadening, a necessary condition
to identify the formation of a nano-composite is that clear phase separation must be
seen in the XRD if conditions are exaggerated (e.g., if cooling rates are slowed down
further as done in Chap. 4 or if oxygen partial pressures are increased as used here).
9.2 Experimental Design
In order to obtain a sample showing these first signs of phase separation, the boundaries of the single-phase region were varied by changing the oxygen partial pressure
during heating. Figure 9.1 shows the upper layered boundary under various conditions. The boundaries for samples heated in air were presented in previous chapters
while the approximate boundaries for 2 % oxygen are based on the current study and
so should only be considered accurate near compositions A 9 and B 9 . The boundary
moves upward as oxygen partial pressure decreases. Therefore, a sample made in
pure oxygen at composition A 9 will be multiphase if quenched while the same composition in 2 % oxygen will be single phase. As will be shown in the results section,
the regular cooled sample made at composition A 9 in 2 % oxygen showed the signs of
being a layered–layered nano-composite. All samples made by the one-pot synthesis
method at composition A 9 were heated at the rate of 5
◦ C/min up to 800
◦ C and held
there for 5 h. The atmospheres used during heating were either air, or a flow of gas
with either 2, 5.5, or 100 % oxygen content with the remainder being either nitrogen
or argon gas. The samples were then either quenched or regular cooled. Samples
at the composition B 9 were made by Aaron Rowe and Eric McCalla using the tank
reactor method. Samples were heated at 900
◦ C for 12 h before either quenching or
regular cooling.