Water-Processable P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ Cathode Material for Sodium Ion Batteries

Due to the voltage plateaus in the P2-Na_0.67Ni_0.33Mn_0.67O₂ cathode material, efforts have been made to incorporate electrochemically inactive elements (Li⁺, Mg²⁺, Zn²⁺, Ti⁴⁺) to mitigate the Na⁺/vacancy ordering and/or irreversible P2-O2 phase transformation at high voltages (e.g., 4.5 V).^12,22–25 However, incorporating inactive elements decreases the Ni concentration and ultimately reduces the capacity.^12,23,25 Multiple transition metals in one layered oxide are usually beneficial to battery performance.^26–28 Therefore, we incorporated Cu^16,17 to substitute Ni and Ti to substitute Mn to form the resulting material of P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂. The X-ray diffraction pattern (XRD, Figure 1a) shows the typical P2-type structure with minor impurities. The primary particle (layered plates stacking to form a step-like morphology) size was approximately 1–5 μm (Figure 1b). Furthermore, we discovered that the particles had smooth surfaces free of tiny particles. The well crystalline nature of the P2 material with a typical layered structure was demonstrated by the transmission electron microscopy (TEM) imaging (Figure 1c). The diffraction dots of the Fast Fourier Transform (FFT) presented two different periodic contrasts (Figure 1c inserted), which might be due to the cation ordering of two Na⁺ sites in P2 phase (Figure 1a inserted).^29,30 The Ti⁴⁺, Mn⁴⁺, Cu²⁺ and Ni²⁺ were verified using the soft X-ray absorption spectroscopy (XAS) in the TEY mode,³¹ which is in agreement with the charge neutrality of the material (Figure 1d). We performed XRD on the same cathode material after it was either stirred in water or stored in an ambient environment. Many sodium containing layered oxides cannot maintain their original crystal structure when storing in an environment where water is present, let alone being soaked in water with continuous stirring for a long period.^{11,13,16,18,32} However, there were no other impurity peaks for these water treated samples (Figure 2a). The supernatant solution remained colorless after 6 days of stirring in water, indicating that the Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ cathode material was resilient against dissolution. For the one stored in the ambient environment for 6 months, we observed that the (002) peak slightly shifted to a lower angle, indicating expanded c parameter due to the formation of carbonate species on the surfaces (i.e., loss of Na⁺ from the layered lattice). We did not observe any major morphology changes other than the formation of tiny particles roughening the surfaces for the latter three samples (Figures 2b–2d). The atomic ratios detected by inductively coupled plasma atomic emission spectroscopy (ICP-AES) are presented in Table I, showing that the atomic ratio in all of the samples stayed close to the nominal ratio of Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11. These results provided direct evidence for the bulk structural stability of the P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂. More characterizations are needed to explore the interfacial stability against water since the interfacial properties also influence the battery performance.^20,33 The water/air stability of layered cathode materials is related to the TM-TM and TM-O coordination, electronic configuration of various TMs, and chemical and physical properties of sodium containing layered oxide compounds.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Multiple voltage plateaus (labeled from I to III) for the P2-Na_0.67Ni_0.33Mn_0.67O₂ cathode material are regarded as a drawback due to the periodic Na⁺/vacancy ordering and disordering at the stages I and II (Figure 3a) and the undesired phase transformation from P2 to O2 at the stage III.^12,30,34 The cells containing the P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ cathode material presented much smoother charge/discharge curves compared to the P2-Na_0.67Ni_0.33Mn_0.67O₂ cathode material, suggesting the Cu/Ti substitution can effectively mitigate the Na⁺/vacancy ordered structure³⁵ and partially hinder the unwanted, irreversible structural change (Figure 3a).¹² In addition, the initial charge and discharge capacity of 130 and 180 mAh/g were achieved at C/5, respectively. Based on the capacity and charge/discharge profiles, the half-cell can deliver a discharge energy density of 544 Wh/kg.^6,36 The cells containing the water-stirred cathode material and the one with the directly water-processed cathode delivered a lower initial discharge capacity of 170 mAh/g (Figures 3c–3d). The water-stirring treatment was not detrimental to the cycling performance. The initial charge and discharge capacity of 130 and 162 mAh/g were obtained at C/5 in the voltage range of 1.0–4.5 V for the cell containing the water-processed cathode (Figure 3d). The charge/discharge profiles for the initial ten cycles of the water-processed cathode overlapped better than the ones in Figures 3b–3c. We discovered minor charge/discharge voltage fading for the water-processed cathode, while major voltage decay was observed in the initial 10 cycles for the former two electrodes. We performed cyclic voltammetry (CV) on the cells containing the pristine and the one containing water-processed cathode. For the cell containing the pristine cathode (Figure 3e), there were 5 distinct oxidation peaks and two reduction peaks in the voltage range of 2.8–4.5 V, and only one redox couple in the voltage range of 1.0–2.8 V for the first cycle. Two drastic changes were discovered in the following cycles. First, with increasing cycle number, the current decreased considerably for the redox peaks at in the 4.0–4.5 V and 1.5–2.3 V. Second, the oxidation peak around 2.1 V shifted to a higher voltage, while the corresponding reduction peak around 1.5 V migrated to a lower voltage. The CV features indicated the irreversible phase transformation that subsequently induced rapid voltage and capacity fading, which is consistent with the charge/discharge profiles (Figure 3b).³⁷ In contrast, there are only two main redox couples in the cell containing the water-processed cathode in the range of 2.8–4.5 V (Figure 3f). In addition, except for the minor changes to the first cycle, the shape of the curves and the peak positions overlapped well, indicating good structural reversibility during the electrochemical cycling. The results are indirect evidence that the water-processing for the P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ cathode material can be beneficial for the battery performance.

Zoom In Zoom Out Reset image size

We also compared the battery performance of the cells cycled in a narrow voltage range (2.5–4.5 V vs. Na⁺/Na) (Figure 4). For the pristine cathode, despite the well overlapped charge curves, the discharge voltage profiles faded continuously from a starting voltage of 4.25 V to one lower than 3.9 V in the first 20 cycles (Figure 4a). The voltage fading might be ascribed to the irreversible phase transformation and crack formation.^12,37 In the following cycles, the charge/discharge voltage profiles overlapped well and a capacity retention of 83% was achieved after 50 cycles. For the cell containing the water-processed cathode (Figure 4b), the irreversible capacity at about 4.0 V can only be found in the initial charge cycle and vanished in the subsequent cycles. The speed of the discharge voltage fading is much slower (Figure 4b). The capacity retention for the water-processed cathode was 84% after 50 cycles, which is close to that of one containing the pristine cathode (Figure 4c). However, significant differences were observed concerning the discharged average voltage profiles (Figure 4d). For the pristine cathode, the initial average voltage of 3.64 V quickly decayed to 3.35 V with a retention of only 91% after 50 cycles. For the water-processed cathode, the initial average voltage was slightly lower (3.6 V) but maintained a higher retention of 97% after 50 cycles. The results give direct evidence that the water-processed cathode was more stable.

Zoom In Zoom Out Reset image size

The bulk P2 structure did not change after being stirred in water for 6 days, even after being stored in the ambient environment for 6 months. However, at the surface, the crystal structure, electronic configuration, and valence states of transition metals are usually transformed by the solvent treatment, which eventually governs the performance of batteries.^19,20 To understand the surface change of the P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ material, the global surface sensitive soft X-ray absorption spectroscopy (XAS) was performed at the O K-edge and Mn L-edge in the total electron yield (TEY) mode³¹ (Figure 5). The TEY mode probes for the chemical information of a metal oxide material with a depth sensitivity of 5−10 nm.²¹ For the pristine P2- Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ material, the peak located at 534 eV is associated with the surface carbonate species (Figure 5a).^19,20,38 The peaks lower than 533 eV are ascribed to the hybridization of TM3d and O2p orbitals. The carbonate species disappeared after the samples were stirred in water or processed into electrodes (either with PVDF or sodium alginate binder). Negligible change was found for the powders after being stirred in water for 1 day and 6 days. The result indicates the superior stability of the cathode surface against water. However, a major difference was discovered for the electrode processed using sodium alginate (SA) in water. One possible reason is that the O K-edge information mainly comes from the oxygen in sodium alginate. The Mn oxidation state at the surface is indicative of the stability of the TM3d–O2p hybridization in Mn-containing sodium layered oxides.^20,38,39 Mn⁴⁺ remained on the surface of Mn for all the powders and electrodes (Figure 5b), providing another direct evidence for the surface stability. Therefore, the surface chemistry of the P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ cathode material is highly stable against water, thus allowing for a low-cost water manufacturing process.

Zoom In Zoom Out Reset image size

As shown in Figure 6, there was no visible difference between the as-made electrode and the cycled electrode (20 cycles) for both the NMP/PVDF-processed (Figures 6a–6b) and water/SA-processed electrodes (Figures 6d–6e). The water-processed cathode seemed to display a higher packing density of cathode particles. For the cross-section after 20 cycles (Figure 6c), we observed the electrode was slightly detached from the current collector in the NMP/PVDF-processed electrode. This could have been caused by the repeated expansion/contraction of the electrode during electrochemical cycling. The water-processed electrode remained in good contact with the current collector (Figure 6f). The difference could be partially responsible for the improved stability in the water-processed cathode. It has been reported that the SA is an effective binder to suppress the irreversible structural transformation and improv the capacity retention.⁸ In addition, the inhibited voltage fading in the water-processed cathode might be associated with the properties of the cathode−electrolyte interface that is not detectable by the SEM imaging. The rich carboxylic groups in the SA binder might improve the electrochemical stability of cathode particles with the electrolyte.¹⁰

Zoom In Zoom Out Reset image size

Finally, we measured the electrochemical impedance spectroscopy (EIS) of the cells after 1, 5, 20, and 60 cycles (Figure 7). For both cells, all of the spectra showed two semi-circles: one at the high frequency region (HF, left semi-circle) that is associated with the resistance (R_i) at the electrode–electrolyte interface (EEI, including CEI and SEI), and the other (R_ct) at the middle frequency region (MF, right semi-circle) that is ascribed to the charge transfer process.^40–42 For the cell containing the NMP/PVDF-processed cathode, the radius of the semi-circles gradually decreased with the increase of cycle number (Figure 7a), suggesting the reduction of both resistances originating from interfaces and charge transfer. For the cell containing the water-processed cathode, there was a significant reduction of the resistances from the 1^st cycle to the 5^th cycle and then a negligible change was noticed until 60 cycles (Figure 7b). The significant impedance decreasing of the cell containing the pristine cathode material most likely originated from the sodium anode. In addition, the improved interface of the water-processed cathode might also help deliver the overall less resistive and more stable interfaces. In order to make a direct comparison between the two cells, we used the same modeling circuit (Figure 7) to fit the R_i and R_ct values (Table II). After the initial cycle, the cell containing the water-processed cathode showed consistently smaller R_i and R_ct. The results indicated that the water-processed cathode allowed for a stable and less resistive interface, which provides partial explanation for the improved electrochemical performance observed in the cells containing the water-processed cathode.

Zoom In Zoom Out Reset image size

All the materials were synthesized through a solid-state reaction method. The stoichiometric starting materials of Na₂CO₃, CuO, NiO, MnO₂, TiO₂ were mixed by ball milling in a ZrO₂ jar at 400 rpm for 8 h. Then, the uniformly mixed materials were transferred into an alumina crucible and heated at 900°C for 12 h in a box furnace to obtain the final product (P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂). The P2-Na_0.67Ni_0.33Mn_0.67O₂ was synthesized without adding TiO₂ or CuO. The aging samples (samples soaked in water for 1 day and 6 days) were collected directly from the pristine powder and stirred in DI water for 1 day and 6 days at 1000 rpm. Then, the powders were centrifuged from water and dried in a vacuum oven at 100°C overnight.

The morphologies of the materials and electrodes were investigated using a scanning electron microscope (SEM) (LEO FESEM) at an accelerated voltage of 5 KV. The overall elemental composition of the material was evaluated using inductively coupled plasma atomic emission spectroscopy (ICP-AES). The ICP-AES instrument is a Spectro ARCOS SOP (side-on or radial view of the plasma interface) made by Spectro Analytical Instruments, Inc. For sequential multiple elemental measurement, the relative standard deviation and the coefficient of variation are less than 2% for concentration higher than the background equivalent concentrations. The X-ray powder diffraction (XRD) of the materials was obtained on a PANalytical X-ray diffractometer with the Cu K_α (λ = 1.54 Å) radiation. Transmission electron microscopy (TEM) was performed on a JOEL 2100 S/TEM operated at 200 keV. Soft XAS measurements were performed on the 31-pole wiggler beamline 10–1 at Stanford Synchrotron Radiation Lightsource (SSRL) using a ring current of 350 mA and a 1000 l·mm⁻¹ spherical grating monochromator with 20 μm entrance and exit slits, providing ∼10¹¹ ph·s⁻¹ at 0.2 eV resolution in a 1 mm² beam spot. Data were acquired under ultrahigh vacuum (10⁻⁹ Torr) in a single load at room temperature using total electron yield (TEY). All spectra were normalized by the current from freshly evaporated gold on a fine grid positioned upstream of the main chamber. XAS samples were mounted on an aluminum sample holder with double-sided carbon tape in a Ar-filled glove box and transferred to the load-lock chamber in a double-contained container, using a glove bag purged with argon for the transfer.

The composite cathodes were prepared by spreading the slurry (N-Methyl-2-pyrrolidone as the solvent, or water as the solvent) with active materials (90 wt%), acetylene carbon (5 wt%) and binder (5 wt%, polyvinylidene fluoride or sodium alginate) and casting them on carbon-coated aluminum foils. The electrodes were then dried overnight at 120°C in a vacuum oven and transferred into a Ar filled glove box for future use. The active mass loading for all the electrodes was ∼ 4 mg/cm². CR2032 coin cells (half cells) were assembled in a Ar-filled glove box (O₂<0.5 ppm, H₂O<0.5 ppm) using the composite cathode, sodium foil as the anode, Whatman glass fiber (1827-047 934-AH) as the separator, and saturated NaPF₆ dissolved in propylene carbonate (PC) as the electrolyte. All the coin cells were cycled with an electrochemical workstation (Wuhan Land Company) at 22°C. The average voltage and energy density were calculated using the LANDdt software. 1C was defined as fully charging a cathode in 1 h, corresponding to a specific current density of 120 mA/g. The electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV) measurements were performed on a Princeton Applied Research of VersaSTAT 4 at 22°C.

In this study, we proposed a quaternary P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ cathode material that achieved a high energy density (discharge capacity: 180 mAh/g, discharge energy: 544 Wh/kg) at 22°C. The Cu/Ti substitution mitigated the Na⁺/vacancy ordering. The aging experiments showed that the P2 material had a superior structural and chemical stability in water and in the ambient environment for 6 months. The aging process had a negligible effect on the battery performance. Furthermore, the cells containing the water-processed cathodes delivered improved cycle life, more stable voltage profiles, and decreased cell impedance. These results show the unique properties of the P2-Na_0.67Ni_0.22Cu_0.11Mn_0.56Ti_0.11O₂ cathode material that can further reduce the cost of material production, storage, and transportation, ultimately lowering the price of sodium ion batteries. Therefore, we anticipate that this work will enable more efforts in developing air-stable cathode materials and promoting the practical applications of sodium ion batteries. Finally, there are still some open questions regarding how water molecules interact with the cathode particles, thus we expect to perform more studies to further understand the interaction mechanism at the molecular level.

The work was supported by the Department of Chemistry Startup Funds and ICTAS Junior Faculty Award at Virginia Tech. The Stanford Synchrotron Radiation Light Source, a Directorate of SLAC National Accelerator Laboratory and an Office of Science User Facility is operated for the US Department of Energy Office of Science by Stanford University. Use of the Stanford Synchrotron Radiation Light source, SLAC National Accelerator Laboratory, is supported by the US Department of Energy, Office of Science (SC), Office of Basic Energy Sciences (BES) under Contract No.DE-AC02-76SF00515. Z. Y. and Y. D. acknowledge the support by the U.S. DOE SC Early Career Research Program. A portion of the work was performed using EMSL, a DOE SC User Facility sponsored by the Office of Biological and Environmental Research. The authors declare no competing financial interest.

Feng Lin 0000-0002-3729-3148