WO2020146630A1

WO2020146630A1 - Red phosphorus/carbon nanocomposite as high capacity and fast-charging battery anode material

Info

Publication number: WO2020146630A1
Application number: PCT/US2020/012931
Authority: WO
Inventors: Yi Cui; Li Wang; Yongming Sun
Original assignee: The Board Of Trustees Of The Leland Stanford Junior University
Priority date: 2019-01-10
Filing date: 2020-01-09
Publication date: 2020-07-16

Abstract

We provide a red phosphorus/carbon nanostructure, featuring amorphous red phosphorus nanodomains embedded in the nanopores of micrometer-scale conductive carbon with interior nanoscale void spaces, a conductive phosphorus-free carbon surface and a high tap density. A battery anode can be configured as a thin layer of such particles, Such anodes are suitable for use in high energy density and fast-charging lithium-ion batteries.

Description

Red phosphorus/carbon nanocomposite as high capacity and fast-charging battery anode material

FIELD OF THE INVENTION

This invention relates to electrodes for batteries.

BACKGROUND

Numerous materials are under investigation for use in improved batteries. One such material is red phosphorus, which has desirable theoretical properties for use in lithium batteries. However, there are substantial

practical difficulties in using red phosphorus in

batteries, such as its poor electrical conductivity, and its significant volume change during battery cycling.

Accordingly, it would be an advance in the art to provide improved phosphorus containing electrode structures for batteries .

SUMMARY

In this work, we provide a red phosphorus/carbon nanostructure, featuring amorphous red phosphorus

nanodomains embedded in the nanopores of micrometer-scale conductive carbon with interior nanoscale void spaces, a conductive phosphorus-free carbon surface and a high tap density. A battery anode can be configured as a thin layer of such particles.

Applications include but are not limited to high energy density and fast-charging lithium-ion batteries. Significant advantages are provided. At an industrial- level areal capacity loading (3.5 mAh cm^-2 ) , a red

phosphorus/carbon electrode shows considerably better fast charging capability than commercial graphite and Li4Ti50i2 electrodes, as well as much higher volumetric capacity and specific capacity based on the whole electrode. Meanwhile, at a high areal capacity loading of ~3.0 mAh cnr² at 0.86 mA cnr², our red phosphorus/carbon electrode shows excellent long-term cycling stability with Coulombic efficiency of 100.0 (± 0.1)% from the 5th to 500th cycle and delivers an areal capacity of 2.7 mAh cnr² at the 500th cycle.

Significant features include:

1) Good electronic conductivity at both the particle and electrode levels to facilitate fast electron transport.

2) Micrometer-scale particles and hierarchical particle size distribution to realize space-efficient packing.

3) Internal void space of particles to buffer the volume change and outer electrolyte blocking layer to stabilize the solid-electrolyte interface (SEI) . Also, one needs to distinguish the difference between the carbon/P hybrids and carbon/P composites. The simple mixture of carbon and P (hybrids) usually have inferior battery performance in terms of cycling stability and Coulombic efficiency.

4) The microscale size of carbon particles lead to the high density of electrodes, which enables high volumetric capacity at the electrode level.

5) The nano-voids help to buffer the volume change at the particle level, and then also at the electrode level, which help the electrode achieve long cycle life, even at high areal capacity loading. 6) The P-free surfaces provide the high electrical

conductivity of the materials, which is helpful in

achieving good rate capability of materials. Meanwhile, because P is trapped inside the pores of carbon and not on the outer surface of carbon particles, the side reaction between P and liquid electrolytes is avoided and the

Coulombic efficiency of materials is improved. Note that structures in some previous publications are more like P/Carbon hybrids, the structures are not well designed or characterized.

Many plant-derived porous carbon, including carbons from nut shells, coconut shells, fruit shells, et al, can be used as the hosts for P. Other active carbon or porous carbon can also be used as the hosts for P. The methods to achieve P/carbon composites can be using a vaporization, adsorption, and surface cleaning method as described below, and can also be other chemical and physical methods, e.g., ball milling. The electrically conductive surfaces of the P/C composite can also be realized by various routes, such as depositing a conducting coating on the P/carbon

composite .

BRIEF DESCRIPTION OF THE DRAWINGS

FIGs. 1A-B show theoretical battery performance parameters for several battery anode materials.

FIGs. 1C-D show a preferred embodiment of the

invention .

FIG. 2A schematically shows a fabrication process for P/C composite particles.

FIGs. 2B-H show characterization results of fabricated P/C composite particles. FIGs. 3A-B shows battery cycling results for a Li battery having an anode with P/C composite particles.

FIG. 3C is an image of three battery anode structures tested in these experiments.

FIGs. 3D-G show experimental results relating to the structures of FIG. 3C.

FIGs. 4A-D show cycling stability results for a Li battery having an anode with P/C composite particles.

FIGs. 4E-G are images comparing fresh P/C composite particle electrode structures to these structures after cycling .

DETAILED DESCRIPTION

A) Introduction

High energy density lithium-ion batteries (LIBs) with superior fast-charging capability are highly desirable for portable electronics and electric vehicles. However, the energy and power density of LIBs based on the conventional intercalation-type graphite anode and the L1C0O2 cathode are approaching their theoretical limit. Recently, extensive research has been conducted with a focus on increasing the energy density of lithium-based batteries by using new battery chemistries, such as Si and Li metal anodes, and S (Li-S batteries) and O2 (Li-air batteries) cathodes, while less attention has been paid to improving the charging rate capability of batteries.

Low ionic and electronic conductivity of materials lead to sluggish solid-state diffusion processes and have long been considered as the main reasons for the inferior rate capability of electrodes and the low power density of

LIBs. The most common strategies to address these challenges are to use nanoscale particles and conducting coatings, which reduce the ion/electron transport distance and improve the conductivity at the particle level.

Successful examples using such strategies have been shown for traditional intercalation type electrode materials (e.g., Li₄Ti₅0i₂ and LiFePCy) . Very recently, a three- dimensional holey-graphene/Nb2C>5 composite was designed and delivered excellent rate capability due to its good charge carrier transport properties. Moreover, carbon-coated L13VO4 spheres have been successfully synthesized and showed great advances as high-rate anode materials. However, when the areal capacity of these electrodes reaches substantially high levels (e.g., > 3 mAh cm^-2 ) , their rate capability is greatly reduced because of increasing charge carrier diffusion lengths in thicker electrodes. Also, the energy density of LIBs using these low-capacity materials is significantly limited due to the large volume and weight of electrodes .

To make a good electrode for LIBs with high energy and high power density, we believe that high capacity and suitable working potential of the electrode materials are two other important parameters in addition to good

electronic and ionic conductivity. FIG. 1A shows calculated thicknesses of the electrodes versus theoretical volumetric capacities of materials. The calculation is based on electrodes with an areal capacity of 3.5 mAh cnr², 20% porosity, and 90% active material volume percentage, and theoretical volumetric capacities of active materials.

For a given electrode areal capacity, porosity, and active material volume percentage, the higher the

volumetric capacity the material delivers, the smaller the thickness of the electrode is and the shorter the charge carrier transport paths are within the electrode (FIG. 1A) , leading to better rate performance. This reduction in volume and weight of the electrodes by using high-capacity battery materials also significantly increases the overall battery energy density. In parallel, suitable working potential is also of vital importance to realize LIBs that combine high energy and power density. In principle, the cathode potential should be high and the anode potential should be low to achieve a high cell output voltage and high energy density. However, to realize high capacity retention during fast charge, the intrinsic working

potential of the cathode should be moderately low to buffer the large overpotentials and to prevent overshooting the cycling voltage cutoffs or electrolyte stability windows.

On the other hand, if the intrinsic potential of the anode is too close to 0 V (vs. Li⁺/Li), large overpotentials involved during rapid charging may make the anode potential drop below 0 V (vs. Li⁺/Li) and prematurely terminate electrochemical lithiation reactions, leading to low capacity retention. But even worse, lithium metal plating takes place at such low potentials, giving rise to major safety concerns. In other words, an ideal anode material for fast-charging LIBs should possess both high capacity and relatively low yet safe lithiation potential.

Here, we show that red phosphorus (P) is an attractive anode material for fast-charging LIBs with high energy density due to the combined advantages of its high capacity and ideal lithiation potential. A red P/carbon (P/C) composite nanostructure, featuring micrometer-scale P/C composite grains with a conductive carbon superficial layer, a P/C composite core, internal nanometer-scale void space and red P nanodomain, and high density, has been successfully synthesized. The rationally-designed P anode material delivers excellent fast-charging capability and high capacity based on the volume and weight of the whole electrode, and presents the best cyclability and Coulombic efficiency [100.0 (±0.1)%] among all reported high-capacity anode materials at a commercial battery level areal

capacity .

B) Results and Discussion

Bl) Anode design for fast-charging LIBs

FIG. IB shows potential versus capacity plots for some typical anode materials. With its low lithiation potential (~0.1 V on average vs. Li⁺/Li) and limited capacity,

graphite, the currently predominant anode, has inherent materials shortcomings for fast-charging LIBs. A commercial graphite electrode exhibits a high overpotential of ~0.12 V at 1 C and loses almost all its capacity in a half cell using a Li metal counter electrode due to this

overpotential. Although Li4TisOi2 has been widely studied and is already used as an anode material for fast charging in electric buses, the energy density of LIBs with Li4TisOi2 anodes is severely limited by its high working potential ( ~1.5 V vs. Li⁺/Li) and low capacity (175 mAh g^_1, 611 mAh cnr³) . It is worthwhile to note that although Si displays the highest capacity among the electrode materials, its low lithiation potential ( ~0.2 V on average vs. Li⁺/Li) makes it non-ideal for fast-charging LIBs.

P reacts electrochemically with both lithium and sodium via an alloy-reaction mechanism (P + 3X⁺ + 3e X3P, X: Li, Na) at an attractive potential (e.g., ~0.7 V in average vs. Li⁺/Li for red P) and delivers a high

theoretical capacity (2,596 mAh g^-1, 6, 075 mAh cnr³) .

Phosphorus is an ideal anode material for fast-charging LIBs due to the combined advantages of high capacity and reasonable lithiation potential: relatively low to achieve high cell output voltage, and relatively high to realize high capacity retention at high charging current densities with large overpotentials and avoid lithium metal plating without safety concerns. Accordingly, P is an ideal anode material for fast-charging LIBs in consideration of both battery safety and energy density. Among the three main allotropes of P (white, red and black) , red P is the most attractive due to its abundance, low cost and good

stability in air, and is suitable for widespread industrial applications .

In addition to choosing the right materials,

structural design motivated by providing good electronic conductivity and space-efficient packing is also important for achieving good fast-charging capability. High

conductivity and nanoscale structural design can facilitate fast electrochemical reactions and high active material utilization at the materials level. Space-efficient packing of particles can reduce the thickness of electrodes and the electron/ion diffusion length at the electrode level, all while improving volumetric energy density of LIBs. In addition, in order to maintain a stable red P anode for fast-charging LIBs, close attention should be paid to addressing its tremendous volume change during the

lithiation/delithiation process, which is a key challenge for all high-capacity battery chemistries.

FIGs. 1C-D show an exemplary preferred anode

configuration based on the analysis above. Here FIG. 1C shows electrode 104 disposed on a current collector 102. Electrode 104 includes numerous particles, e.g., as shown. Each of these particles has a first region (e.g., 108, gray shading) encapsulated by a second region (e.g., 106, black outline) . Electrode 104 preferably has a hierarchical particle size distribution to realize space-efficient packing, as shown. One way to define a hierarchical particle size distribution is as a distribution having two or more sets of particles having substantially different sizes (i.e., particle sizes differing by a factor of 1.5 or more) . That way the smaller particles can fill in the gaps between the larger particles, improving the overall density of the electrode.

FIG. ID is an enlarged view of the first region of one of the particles. Here the first region is a composite having red phosphorus 114 disposed in pores of a porous carbon matrix 112. The red phosphorus partially but not completely fills the pores of the porous carbon matrix, resulting in voids (e.g., 116) .

The second regions of the particles (e.g., 106 on FIG. 1C) are electrically conductive and include no

phosphorus. In the example below, the second regions are formed by washing the phosphorus out of surface regions of composite P/C particles. An alternative is to coat P/C particles with an electrically conductive and phosphorus- free composition.

Preferably the particles have sizes between 0.5 jjm and 20 miii. Preferably the pores of the porous carbon matrix have pore sizes in a range from 2 nm to 300 nm. Preferably a hierarchical pore size distribution is used to realize space-efficient pore packing. One way to define a

hierarchical pore size distribution is as a distribution having two or more sets of pores having substantially different sizes (i.e., pore sizes differing by a factor of 1.5 or more) . That way the smaller pores can fill in the gaps between the larger pores, improving the overall density of the phosphorus/carbon composite. The surface area of the porous carbon matrix is preferably between 300 m²g^_1 and 2500 m²g^_1. The pore volume of the porous carbon matrix is preferably between

0.25 cm³g^_1 and 2 cm³g^_1. The phosphorus weight fraction of the micro-particle is preferably between 20% and 80%. The anode thickness (h on FIG. 1C) is preferably 100 mpi or less and is more preferably 50 jjm or less.

This preferred configuration provides good electronic conductivity via the porous carbon matrix 112 on FIG. ID (particle level) and via the second regions (e.g., 106) on FIG. 1C (electrode level), both schematically shown with white arrows, to facilitate fast electron transport. It also provides space efficient particle packing and good stability. The internal nanoscale void space and red P nanodomains of the particles buffer the volume change and the outer electrolyte blocking layer (e.g., 106 on FIG. 1C) stabilizes the solid-electrolyte interface (SEI) .

B2) Synthesis and characterization of the P/C nanocomposite

A simple, low-cost, large-scale (1 kg per batch) synthesis route was developed for the synthesis of a red P/C nanocomposite for fast-charging LIBs using an

adsorption and surface cleaning strategy shown in FIG. 2A. Micrometer-scale nanoporous carbon particles (204 on

FIG. 2A) and red P powder were sealed in a stainless steel vessel in argon atmosphere and annealed at 450 °C for 3 h. Above the sublimation temperature of red P (416 °C) , the produced P vapor (202 on FIG. 2A) penetrated into the nanoporous carbon particles via diffusion and easily reached the inner nanopores. During the cooling process, the P vapor in the nanopores turned into red P via

condensation that deposited onto the inner walls of nanopores. These nanopores may not be fully filled, forming internal nanoscale void space since solid red P has higher density than P vapor. After cooling to room temperature, the result is a P-loaded carbon particle 206 that includes superficial phosphorus 208 on it. The superficial P on the P/C composite particles was cleaned away with a surface treatment by using a solvent to provide particles each having a first region 210 encapsulated by a second region 212 where the first and second regions are as described above .

An inexpensive, plant derived-porous carbon was chosen as the host for P filling due to its reasonable particle size, high specific surface area and large pore volume (1445 m² g^_1 and 0.78 cm³ g^_1, respectively) . See the nitrogen sorption measurements of FIG. 2B. Transmission electron microscopy (TEM) images show that the carbon particles possess an interconnected nanoporous network with pore size less than 5 nm, in agreement with the nitrogen sorption results (FIG. 2C) . These numerous nanopores with high porosity allow a high P filling content (~50 wt%), which enables high capacity based on the total weight and volume of the materials. The as-achieved red P/C

nanocomposite is stable for lithium ion batteries. The red P in porous carbon sublimes after 390 °C under argon atmosphere. Under air atmosphere, red P/porous carbon composite starts to catch fire at 480 °C, which is much higher than the temperature for thermal runaway of cells using conventional organic electrolyte (~180 °C) .

FIG. 2D shows a typical scanning transmission electron microscopy (STEM) and the corresponding energy-dispersive X-ray (EDX) elemental mapping images of the red P/C

composite. The uniform elemental distribution of P and C in the whole investigated area is observed, suggesting that P is uniformly embedded into the carbon host. The specific surface area and pore volume of the P/C nanocomposite are only 9 m² g^_1 and 0.02 cm³ g^_1 (FIG. 2B) , respectively, indicating that almost all the nanopores of the particles are filled and/or blocked by red P.

FIG. 2E displays a representative high-resolution TEM (HRTEM) image for a P/C composite particle. In contrast to uniform greyscale of a pure porous carbon particle (FIG.

2C) , a dark/light contrast is clearly observed for a P/C composite particle, where the inner region is dark and the surface is light. The dark inner region of the particle includes amorphous species and the nanopores of the carbon matrix cannot be observed, suggesting the embedding of amorphous red P nanodomains into the internal nanopores of the carbon. The amorphous structure of P is also confirmed by the X-ray diffraction (XRD) pattern, which shows no obvious peaks (FIG. 2F) . The amorphous structure and small size of red P nanodomains inside the nanopores of the carbon helps to speed up the electrochemical reaction kinetics. A light surface layer with thickness of ~20 nm is observed for the same P/C composite particle, corresponding to a pure carbon surface (FIG. 2E) . Thus, the P/C composite particles possess a P-free thin carbon superficial layer and a P/C composite core with interior interconnected conductive network, which results in a high electrical conductivity of 2.3 S cnr¹ for the P/C composite powders.

The interior interconnected network of carbon creates a facile electron transport pathway within the P/C particles, and the pure carbon surface of such particles forms an electronically conductive network around and between neighboring particles to achieve good electrical

conductivity at the electrode level, both of which are important in achieving fast electron transport at the materials and electrode levels (FIGs. 1C-D) . Moreover, the pure carbon surface prevents the direct contact between the P and electrolytes and limits side reactions.

Scanning electron microscopy (SEM) images show that the P/C nanocomposites have micrometer-scale particles with irregular shapes and a wide size distribution mainly between 3 to 8 pm (FIG. 2G) , without obvious morphology change in comparison with the bare carbon. The size and shape of the P/C nanocomposite promote space-efficient packing with a high tap density of up to 1.0 g cnr³, which facilitates the preparation of densely compacted electrodes with small thickness. A cross-sectional SEM image verifies the existence of numerous nanometer-scale void spaces inside the P/C composite particles (FIG. 2H) . These

nanometer-scale void spaces can act as internal buffers for the local volume expansion of P and restrict the overall expansion of the whole P/C composite particles during the lithiation process, ensuring that the particles and

electrodes remain intact upon long-term cycling and the SEI on the particle surface is stable and thin.

B3) Fast-charging capability of the P/C electrode

The as-synthesized P/C composite nanostructure with high conductivity and short transport path for charge carriers affords remarkable fast-charging capability.

Electrochemical measurements were performed at increasing cycling rates at a moderate areal capacity shown in

FIGs. 3A-B. The initial 5 cycles were performed at slow rates of 0.1 and 0.2 C, allowing for the activation of materials and the formation of a stable SEI on the particle surface. The P/C electrode shows stable cycling over various current densities. It exhibits an average lithiation potential of 0.66 V vs. Li⁺/Li and a high

specific capacity of 2173 mAh g^_1 at 1 C, corresponding to an areal capacity of 1.0 mAh cnr² at 1.1 mA cm^-2. The

specific capacity of the P/C electrode is as high as 1829 mAh g^_1 at 6 C (6.6 mA cnr²) . In other words, a battery with such a P/C anode can fill up 84% of its capacity within 10 minutes, reflecting its excellent fast-charging capability. Although a large overpotential of 0.36 V is produced at 8 C (8.8 mA cm^-2 ) , the electrode shows a safe average lithiation potential of 0.30 V vs. Li⁺/Li and delivers a high specific capacity of 1523 mAh g^_1, benefiting from the intrinsic, relatively high lithiation potential of red P and optimized material structure design. To avoid the influence of Li metal counter electrode under fast charging rate in half cells during long-term cycling, LiFeP0₄ I I P/C full cells were constructed. Charge/discharge measurements of the full cell using various charging rate and prolonged cycling at 4 C were performed. The capacity at 0.5 C (the 2^nd cycle) and 4 C were 124.0 mAh g^_1 and 106.9 mAh g^_1 (the 36^th cycle), respectively. Stable battery cycling were achieved at 4 C from the 36^th to 100^th cycles with high capacity retention of 96%.

Good fast-charging capability of electrodes at a high areal capacity is of practical importance for industrial applications. To show the significant advantages of P/C electrodes for high energy density, fast-charging LIBs,

P/C, Li₄Ti₅0i2 and graphite electrodes were closely compared with the same content of materials ratio (90wt% materials, 5wt% binder and 5wt% carbon black) and an industrial-level areal capacity of ~3.5 mAh cnr² at 0.5 mA cm^-2. FIGs. 3C-D show the cross-section SEM images of P/C, Li₄Ti₅0i₂ and graphite electrodes and their average thicknesses. Due to the high capacity of the P/C material and its space- efficient packing, the average thickness of P/C electrodes is 21.5 pm, much thinner than the 76.3 and 124.5 pm for Li₄Ti₅0i₂ and graphite electrodes, respectively (FIG. 3D) . The much smaller thickness of the P/C electrodes provides a much shorter path to facilitate fast ion and electron transport at the electrode level. Meanwhile, a high

volumetric capacity of 1628 mAh cnr³ is achieved based on the whole P/C electrode, which is significantly higher than that of Li₄Ti₅0i₂ and graphite electrodes (459 and 281 mAh cnr³, respectively) (FIG. 3E) . Moreover, the P/C electrode delivers the highest capacity based on the total weight of the electrodes. The average total weight of the P/C

electrode is 4.3 mg cnr², which is much less than 12.5 mg cnr² for the graphite electrode and 23.6 mg cnr² for the Li₄Ti₅0i₂ electrode (FIG. 3D) . The capacities based on the average total weight of electrodes are 824, 148 and

279 mAh g^_1 for the P/C, Li₄Ti₅0i₂ and graphite electrodes, respectively (FIG. 3E) . To better evaluate the capacity of battery electrodes, one should normalize the capacity based on the total weight of electrode, current collector and electrolyte intake.

Such data was collected and analyzed. Due to the small electrode thickness and weight, the electrolyte intake for the P/C electrode (0.25 mg cnr²) is much less than that for Li₄Ti₅0i₂ (9.08 mg cnr²) and graphite (4.03 mg cnr²)

electrodes. The total weight of P/C electrode, including electrode, absorbed electrolyte and current collector, is 11.70 mg cnr², much less than 35.66 mg cnr² for Li₄Ti₅0i₂ electrode and 23.76 mg cnr² for graphite electrode, even when considering that the weight and density of the Cu current collector for P/C electrode is higher than that of the A1 current collector for Li₄Ti₅0i₂ electrode. Normalized for the total weight, the capacity of the P/C electrode is 299 mAh g ¹, which is much higher than 98 mAh g^_1 for

Li₄Ti₅0i₂ electrode and 147 mAh g^_1 for graphite electrode. Additionally, beyond just the much higher capacity of the P/C anode than the Li₄Ti₅0i₂ electrode, the P/C electrode exhibits much lower working potential (~0.7 V) than the Li₄Ti₅0i₂ electrode (~1.5 V) . Much higher energy density can thus be achieved for the cells using the P/C anode than the Li₄Ti₅0i₂ anode. Therefore, by using such high-capacity P/C electrodes with small thickness, high energy density LIBs with excellent fast-charging capability can be realized.

Charging capability of these P/C, graphite and Li₄Ti₅0i₂ electrodes was measured by changing the lithiation current density from 0.25 to 5 mA cm^-2. As shown in FIG. 3F, the areal capacity of the tested P/C electrode is 3.52 mAh cnr² at 0.5 mA cnr², corresponding to a high specific capacity of 2015 mAh g^_1. When the current density is increased from 0.5 to 3 mA cnr², only a slight decrease in capacity is

observed. The P/C electrode delivers an areal capacity of 2.94 mAh cnr² at 3 mA cnr², corresponding to a high specific capacity of 1684 mAh g^_1. Meanwhile, an areal capacity of 1.54 mAh cnr² is achieved at a high areal current density of 5 mA cnr².

The rate-dependent charging capability of graphite and Li₄Ti₅0i₂ electrodes were also measured. FIG. 3G compares the capacity retention for the P/C, Li₄Ti₅0i₂ and graphite electrodes at various areal current densities. The graphite electrode loses most of its capacity at 2 mA cm^-2. In contrast, the Li₄Ti₅0i₂ has 91% capacity retention at 2 mA cnr², showing much better rate capability. Impressively, the P/C electrode has comparable capacity retention (92%) to that of Li₄Ti₅0i₂ at 2 mA cnr² and even higher capacity retention at higher current densities. The P/C electrode retains 84 and 44% of its capacity at 3 and 5 mA cnr², respectively, much higher than the 71 and 36% for the

Li₄Ti50i2 electrode, demonstrating the superior fast-charging capability of the P/C electrode at an areal capacity level of commercial lithium-ion battery cell.

B4) Cycling stability of the P/C electrode

The P/C nanocomposite also affords remarkable cycle life at an areal capacity level of commercial lithium-ion battery cell (FIG. 4A) . For good comparison to previous reports, the content of P/C material in the electrode is 80wt%. A P/C electrode with 1.6 mg cnr² P loading delivers a discharge areal capacity of 3.0 mAh cnr² at the 10^th cycle at 0.86 mA cnr² (C/5) and 90% capacity retention at the 500^th cycle. The corresponding discharge specific capacity for the 500^th cycle is 1625 mAh g^_1, which is more than 4 and 9 times the theoretical capacities of graphite and Li4TisOi2, respectively. This is first time that both appropriately high areal capacity and stable long-term cycling have been shown for P-based anodes, and the cycle stability is among the best for high-capacity anodes. This remarkable cycling performance data should be distinguished from most commonly reported results, which are normalized by the weight of active materials at a low areal capacity or shown with limited cycles (e.g., 100 or less cycles) . The P/C

electrode exhibits stable sloping lithiation and

delithiation potential profiles with an average lithiation voltage of 0.69 V (vs. Li⁺/Li), delithiation voltage of 1.02 V (vs. Li⁺/Li) and potential hysteresis of 0.33 V during cycling (FIG. 4B) . Only a slight increase in overpotential and decrease in capacity are observed after 500 cycles, which may arise from the degradation of the Li metal counter electrode. In addition to capacity stability, Coulombic

efficiency is another important concern for high-capacity electrode materials. The initial-cycle Coulombic efficiency of the P/C electrode corresponding to FIG. 4A is 80.5%.

This initial lithium loss can be addressed by prelithiation strategies in full cell configuration. The Coulombic efficiency exhibits a rapid increase during cycling, reaching 96.1%, 99.7% and 99.8% for the 2^nd, 3^rd and 4^th cycle, respectively, and it is maintained at high values between 99.9% to 100.1% after the 4^th cycle, with an average Coulombic efficiency of 100.0% (FIG. 4C) , which is similar to, if not better than, the commercial graphite anode.

Stable and high Coulombic efficiency is a decisive factor for the practical application of high-capacity anode materials into full-cell systems and therefore, in this respect, our results indicate great promise for the

industrial application of P/C electrodes. The Nyquist plots of the as-prepared P/C electrodes show a suppressed

semicircle in the high-middle frequency region and an oblique straight line in the low frequency region for different cycles (FIG. 4D) . The semicircle diameter of the plot for the first cycle is much larger than that for the following cycles, mainly due to the reduced

contact/interfacial impedance of the electrolyte against the Li metal counter electrode after the first cycle. The similar semicircle diameters after first cycle verifies the good stability of the P/C electrodes.

Stable cycling of a high-areal-capacity electrode with excellent Coulombic efficiencies has strict requirements in terms of structural stability both at the particle and electrode levels. SEM imaging was performed to investigate and compare the morphology and structure of the P/C

nanocomposite and electrode before and after 100 cycles. The cycled electrode remains in good contact with the current collector and preserves its morphological integrity and electrical interconnectivity throughout the whole measured area without any cracks and contact losses

(FIG. 4E) . Meanwhile, no cracks are observed for individual P/C composite particles after cycling. To more closely investigate the structural stability of the P/C

nanocomposite at the particle level, an in situ

electrochemical cell was built inside the TEM. In contrast to the large volume expansion of pure P (~300%), only slight volume expansion ( ~12% in one dimension, ~40% for the whole particle) was observed during the lithiation process of one single P/C composite particle and the investigated particle preserved its structural integrity throughout the lithiation and delithiation processes without any cracking. The volume expansion at the particle level in the in-situ TEM experiment should be different from electrode-based volume change or thickness variation during cycling. The interspace between particles in the electrode may accommodate some volume change of the

particles during lithiation and delithiation processes. Thus, it is expected that the measured volume expansion at the electrode level will be smaller than that of a single particle observed using in-situ TEM.

A Vernier Caliper provides direct and accurate

measurement of the thickness of electrodes. The volume change of the overall electrode during charging and

discharging depending on time was investigated. The value of the thickness change of a P/C electrode is reversible after the first cycle. The thickness was 72 pm for the fully charged state and 58 pm for the fully discharged state for a P/C electrode with an areal capacity of ~3.5 mAh cnr² at 0.5 mAh cnr² after the first cycle. The corresponding reversible electrode expansion in thickness of the electrode is ~25%. Cross-section SEM images of a P/C composite particle shows that a uniform and thin SEI layer (< 50 nm) forms on the outer surface of a P/C composite particle (FIG. 4F) after 100 cycles and the interior nanoscale void spaces are well maintained for buffering volume change (FIG. 4G) . These results indicate that good structural stability of the P/C composite are maintained both at the electrode and particle level, which enables stable battery cycling and high Coulombic efficiency of 100.0%. Moreover, defects in the activated carbon produced from coconut shell may also have advantages in improving the cycling stability of the as-achieved P/C anode.

C) Conclusions

Our studies show that red P with high capacity and relatively low yet safe lithiation potential is an

attractive anode material for high energy density, fast charging LIBs. We demonstrate that electrodes utilizing a red P/C nanocomposite with an optimized structure display much better capacity retention and deliver a much higher capacity based on both the total weight and volume of the electrodes than that of the commercial graphite and Li4TisOi2 electrodes at the areal capacity level of commercial lithium-ion battery cell (~3.5 mAh cnr²) . Furthermore, the P/C electrode displays a combination of stable cycling and 100.0 (±0.1)% Coulombic efficiency during cycling at an appropriately high areal capacity of ~3 mAh cm^-2. Due to its superior electrochemical performance, easy preparation, and low cost, we believe that the P/C nanocomposite will have important applications in advanced fast-charging LIBs with high energy density. D) EXPERIMENTAL PROCEDURES

Dl) Materials Synthesis

Coconut shell-derived, hierarchically nanoporous C particles were prepared according to a previous report with minor modifications. Coconut shells were first cleaned, crushed and sieved. The coconut shells were carbonized under nitrogen atmosphere at 700 °C for 2 hours. The as- achieved carbon was then soaked in 50wt% potassium

hydroxide (KOH) solution with KOH/C mass ratio of 4. After drying at 120 °C for 5 h, the mixture was heated to 850 °C under nitrogen gas flow and followed by 2 hours activation under carbon dioxide (CO2) gas flow. Porous carbon was achieved after washing. The red P/C nanocomposite was prepared using a vaporization, adsorption, and surface cleaning method. Commercial red P and the as-achieved nanoporous C particles with micrometer size were used as the starting materials. Red P was milled and washed with deionized water before using. Porous carbon particles and excess red P powder were sealed in a stainless steel vessel in argon atmosphere and annealed at 450 °C for 3 h. The excess red P makes sure that there is enough P vapor and pressure for the conversion from white P to red P and long cooling time of 15 hours provides enough time for this conversion. Last, the superficial P on the P/C composite particles was washed away using CS2 solvent to form a pure carbon surface. The washing process was performed in the hood to avoid the intake of toxic CS2 vapor. For the future application in industrial battery materials synthesis and battery manufacturing, there are safety measures for the use of toxic solvents (e.g., CS2) . D2 ) Characterizations

The SEM images were taken with an FEI XL30 Sirion SEM. An FEI Helios NanoLab 600i DualBeam FIB/SEM system was used to acquire the cross-section SEM images of the P/C

particles. The XRD measurements were conducted on a

panalytical X'pert diffractometer with Ni-filtered Cu Ka radiation. The thicknesses of electrodes were measured using a Vernier Caliper. An FEI Titan 80-300 environmental TEM was used for TEM, HRTEM and STEM images, EDS mapping and electron energy loss spectroscopy (EELS) spectra collection, and in situ TEM measurements. The in situ electrochemical cell was built in a piezo-controlled, electrical biasing TEM-AEM holder (Nanofactory Instruments) with a P/C particle as the working electrode and Li metal as the counter electrode. By applying a voltage bias, lithium ions flowed through the native oxide layer of Li metal to react with P reversibly. A standard four-probe method was used for the electrical conductivity measurement of the red P/C nanocomposite. Before measurement, the sample powders were compressed into a pellet with 10 mm in diameter and 1 mm in thickness under a pressure of 20 MPa.

D3) Electrochemical measurements

A slurry method was used to prepare the electrodes by mixing the active materials, carbon black and

polyvinylidene fluoride (PVDF) binder in N-methyl-2- pyrrolidinone (NMP) solvent. The electrodes (P/C, Li₄Ti₅0i₂ and graphite) for rate capability measurements were

prepared with 90 wt% materials, 5 wt% carbon black and 5 wt% PVDF binder. P/C electrodes with two different total mass loadings of 0.9-1.1 mg cnr² and 4.2-4.4 mg cnr² are used in our experiment. The corresponding mass loadings of active P are 0.4-0.5 mg cnr² and 1.9-2.0 mg cnr², respectively. P/C electrodes for long-term cycling

stability test had 80 wt% of P/C composite, 10 wt% carbon black and 10 wt% PVDF with P mass loading of 1.6-1.7 mg cnr² and total mass loading of 4.0-4.2 mg cm^-2. 2032-type coin cell (MTI Corporation) cells were assembled in an Argon- filled glove box using Li metal as the counter electrode, a Celgard 2325 membrane as the separator and 1 M LiPF₆ in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 v/v) as the electrolyte. 120 pL of electrolyte was used for the coin cell configuration. The battery performances were performed on an Arbin 96-channel battery tester or a LAND 8-channel battery tester. The

galvanostatic charge/discharge measurement for P/C | | Li metal cells was carried out with the cut-off potential range of 0.01-2 V. Graphite | | Li metal cells were charged and discharged between 0.01 to 1 V. The cut-off potential range for Li4TisOi2/Li metal cells was 1-2.5 V. For the measurement of fast-charging capability of our P/C anode, half cells using lithium metal as the counter electrode were assembled. To minimize the influence from the lithium metal counter electrode, P/C | | Li metal cells were

discharged at increasing C rate but charged at constant C/5 rate. The values of the overpotential at fast charging rate were calculated by taking the potential difference between potential plateau at fast rates and the equilibrium

potentials calculated according to the charge-discharge curves at a low current rate of 0.2 C. The specific

capacity is calculated based on the mass of P, if not stated otherwise.

Claims

1. A micro-particle for use in a battery anode, the microparticle comprising:

a first region encapsulated by a second region;

wherein the first region is a composite having red phosphorus disposed in pores of a porous carbon matrix; wherein the red phosphorus partially but not

completely fills the pores of the porous carbon matrix in the first region;

wherein the second region comprises one or more electrically conductive materials and does not include phosphorus .

2. The micro-particle of claim 1, wherein the one or more electrically conductive materials include the porous carbon matrix having no phosphorus in its pores.

3. The micro-particle of claim 1, wherein a size of the micro-particle is between 0.5 jjm and 20 mpi.

4. The micro-particle of claim 1, wherein the pores of the porous carbon matrix have pore sizes in a range from 2 nm to 300 nm.

5. The micro-particle of claim 4, wherein the pores of the porous carbon matrix comprise two or more sets of pores having substantially different sizes.

6. The micro-particle of claim 1, wherein a surface area of the porous carbon matrix is between 300 m²g^_1 and 2500 m²g^_1.

7. The micro-particle of claim 1, wherein a pore volume of the porous carbon matrix is between 0.25 cm³g^_1 and 2 cm³g^_1.

8. The micro-particle of claim 1, wherein a phosphorus weight fraction of the micro-particle is between 20% and 80%.

9. A battery anode comprising a multiplicity of micro particles of claim 1 disposed on a current collector.

10. The battery anode of claim 9, wherein sizes of the micro-particles are between 0.5 jjm and 20 mpi.

11. The battery anode of claim 10, wherein the micro particles of the battery anode comprise two or more sets of micro-particles having substantially different sizes.

12. A method of making micro-particles for use in a battery anode, the method comprising:

providing micro-particles of a porous carbon matrix; diffusing phosphorus vapor into the porous carbon matrix; condensing the phosphorus vapor in pores of the porous carbon matrix; surface treating the micro-particles to provide a phosphorus-free second region of the micro-particles; whereby the micro-particles have a first region encapsulated by a second region;

wherein the first region is a composite having red phosphorus disposed in pores of a porous carbon matrix;

wherein the red phosphorus partially but not

completely fills the pores of the porous carbon matrix in the first region;

wherein the second region does not include phosphorus.

13. The method of claim 12, wherein the surface treating the micro-particles comprises washing the micro-particles with a solvent to remove phosphorus from pores of the porous carbon matrix in the second region.

14. The method of claim 12, wherein the surface treating the micro-particles comprises coating the micro-particles with a phosphorus-free, electrically conductive

composition .