WO2019243614A1 - Electrode active materials and method for their manufacture - Google Patents

Abstract

An electrode active material is disclosed for an electrochemical device. The electrode active material may be an anode active material. The electrode active material comprises granules of nanoparticles, the nanoparticles being at least partially sintered together. The nanoparticles have an average diameter in the range 2-1000 nm. The granules have an average diameter in the range 1-25 µm and larger than the diameter of the nanoparticles. The granules are porous and comprise a coating of carbon formed at least at the surfaces of pores in the granules. The BET surface area of the granules is 1- 50 m2/g. The nanoparticles comprise one or more of FexOy, Fe203, Fe304, FeO, MnxOy, MnO2, Mn203, Mn304, MnO, ZnxFeyOz, ZnFe204, ZnxCoyOz, ZnCo204, ZnxMnyOz, ZnMn204, CoxSnyOz, Co2SnO4, MnxSnyOz, Mn2SnO4, NixSnyOz, Ni2SnO4, CoxSnyOz, CoSnO3, NixSnyOz, NixSnO3, CoxSiyOz, Co2Si04, FexSiyOz, Fe2SiO4, MnxSiyOz, Mn2SiO4, SnxOy, SnO2, GexOy, GeO2, SbxOy, Sb2O5, SixOy, S1O2, SiOx, CoxOy, CoO, Co3O4, CuxOy, CuO, CU2O, NixOy, NiO, ZnxOy, ZnO, CrxOy, Cr2O3, MoxOy, MoO3, MoO2, RuxOy, RUO2, Sn, Si, FexSi, CuxSi, LixTiyOz Li2TiO3, NbxOy, Nb2G5, NbxMyOz (M = Fe, Ti, V, Mo, Zr, Mg, W, Ga, K), LiNiCoMnO2, LiNiCoAIO2, LiMn204, LiNi0.5Mn1.5O4, LiC0O2, LiFeP04, and their combinations.

Description

ELECTRODE ACTIVE MATERIALS AND METHOD FOR THEIR MANUFACTURE

The project leading to this application has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 796648.

BACKGROUND TO THE INVENTION

Field of the invention

The present invention relates to electrode active materials and to methods for the manufacture of electrode active materials. Such materials are of interest for example as electrode active materials in lithium ion batteries.

Related art

WO2018/087215 (PCT/EP2017/078732) and WO2019/052670 (PCT/EP2017/073515) are disclosures of nanostructured materials and methods for their manufacture, the nanostructured materials being of interest as electrode active materials for lithium ion batteries. These disclosures originate from the inventors’ research group.

Lithium ion (Li-ion) batteries are a commonly used type of rechargeable battery with a global market estimated at $1 1 bn in 2010 and predicted to grow to $50bn by 2020. This large market is divided between various applications, ranging from transport and utility- scale energy storage to consumer electronics. Accordingly, rechargeable (secondary) Li- ion batteries are currently under intense research and development to improve their performance to reach industrial demands of the technology [Goodenough and Park (2013)]

A typical lithium-ion battery is composed of multiple cells connected in series or in parallel. Each individual cell is usually composed of an anode (negative polarity electrode) and a cathode (positive polarity electrode), separated by a porous, electrically insulating membrane (called a separator), immersed into a liquid (called an electrolyte) enabling lithium ions transport.

In most systems, the electrodes are composed of an electrochemically active material - meaning that it is able to chemically react with lithium ions to store and release them reversibly in a controlled manner - mixed if necessary with an electrically conductive additive (such as graphitic carbon) and a polymeric binder. A slurry of these components is coated as a thin film on a current collector (typically a thin foil of copper or aluminium), thus forming the electrode. In the known Li ion battery technology, the low theoretical capacity (about 370 mA g^·1) of graphite anodes is a serious impediment to its application in high-power electronics, automotive and industry. Among a wide range of potential alternatives proposed recently, Si, M_xSy, Fe_xO_y, and niobium-based materials are the main contenders to replace graphite as the active material of choice. Si has about 10 times more theoretical capacity than graphite but its dramatic volume expansion (up to about 400%) severely limits high- power applications. Although this problem can be partially tackled by carbon coating [Liu et al (2014)], implementation of these in large scale is still problematic. Similarly, metal sulphide (M_xS_y) electrodes, despite their high theoretical capacity not only suffer from volume expansion but dissolution of polysulfides that form during charge/discharge

[Liang et al (2015)] in battery electrolytes. On the other hand, Fe_xO_y-nanocarbon [Tuek et al (2014)] has now emerged as a promising anode material platform because of its higher (600-1000 mAh/g, or 600-800 mAh/g sustained) capacity than graphite, good capacity retention at high rates, environmental-benignity, high corrosion resistance, low- cost, non-flammability and high-safety. Fe_xO_y based anodes operate via conversion or conversion alloying, as explained in Loeffler et al (2015).

Tuek et al (2014) and Ren et al (2015) are two examples of conversion batteries, meaning that the chemical mechanisms leading to lithium ions storage and release is a conversion reaction. The conversion mechanism can be generally described as follows:

TM_xO_y + z e- + z Li⁺ -> x TM⁽⁰⁾ + Li_zO_y

where TM is a transition metal and TM<°> refers to is elemental form. Upon battery charging, lithium ions diffuse and react into these materials, and nanoscale metallic domains of TM<°> are formed, embedded in an amorphous matrix of Li_zO_y. The reaction is reversed during battery discharge.

Conversion anodes have recently been referred to as the next generation anodes

[Loeffler et al (2015)]. As explained in Loeffler et al (2015), an appealing feature of conversion materials is their ability to store more equivalents of lithium (two to eight per unit formula of the starting material) than any insertion compound (up to two), resulting in substantially higher specific capacities. However, conversion materials exhibit a series of severe drawbacks which necessarily need to be overcome before they can be seriously considered for commercial applications [Cabana et al (2010)]. These drawbacks are also explained in Loeffler et al (2015). The conversion reaction inherently causes a massive structural reorganisation, which potentially leads to a loss of electrical contact and electrode pulverisation. Moreover, conversion materials can suffer from a very high reactivity towards commonly used electrolytes and a marked (dis-)charge voltage hysteresis, considerably affecting the energy storage efficiency of such electrodes. The elevated operational potentials of many conversion materials also limit the achievable energy density and the large first-cycle irreversible capacity is considered to be unacceptable for practical applications.

Conversion or conversion/alloy materials nevertheless present an attractive development route to enable high energy and power dense batteries, that can charge and discharge extremely fast with a long lifetime [Poizot et al (2000); Cabana et al (2010); Reddy et al (2013)]. It is therefore of interest to consider the issues that such materials typically encounter upon battery cycling, due to their poor conductivity, high voltages versus Li, irreversible capacity losses due to side reactions, and expansion/contraction of the electrode upon cycling leading to failure. With respect to this last point, it is noted here that“electrode pulverisation” refers to the loss of electrode mechanical integrity after charge and discharge cycling. Upon active material lithiation and delithiation, the active material swells and contracts, creating internal stresses that can ultimately lead to structural damage.

Additionally, there are other materials which also present attractive routes to high energy and power dense batteries. For example, 3D intercalation materials such as lithium titanate or niobium oxide composite materials have the potential to be used in

development of the next generation of anodes.

SUMMARY OF THE INVENTION

Previous work by the inventors’ research group has identified Fe_xOy-nanocarbon structures as providing an advantageous basis for the development of new anode materials for lithium ion batteries. That work is set out in WO2018/087215

(PCT/EP2017/078732) and WO2019/052670 (PCT/EP2017/073515). The present disclosure is based on insight gained while researching those Fe_xOy-nanocarbon structures.

However, the present inventors have realised that the insights developed in their initial work on conversion materials and conversion-alloying materials are applicable to a wider range of electrode active materials. For example, the present disclosure applies to conversion materials and conversion alloy materials and also to 3D intercalation materials (for example), all being of interest for the development of new anode materials for lithium ion batteries.

The present inventors consider that hierarchically structured particles of 3D intercalaction materials are capable of providing significant benefits in terms of their electrochemical performance. This is due to increased material availability through nano-sizing, increased tapped density allowing improved subsequent electrode densities, and improved electronic conductivity, provided that an appropriate conductive coating is applied. Such insight is supported, for example, by Zhang et al (2013).

Accordingly, the present disclosure seeks to provide one or more further electrode active materials and one or more methods for its manufacture, optionally further improved with respect to the inventors’ research group previous work.

Accordingly, in a first preferred aspect, the present invention provides an electrode active material for an electrochemical device, the electrode active material comprising granules of nanoparticles, the nanoparticles being at least partially sintered together, the nanoparticles having an average diameter in the range 2-1000 nm, the granules having an average diameter in the range 1 -25 pm and larger than the diameter of the

nanoparticles, the granules being porous and comprising a coating of carbon formed at least at the surfaces of pores in the granules, wherein the BET surface area of the granules is 1-50 m²/g, wherein the nanoparticles comprise one or more of Fe_xO_y, Fe₂<D₃, Fe30₄, FeO, Mn_xO_y, MnC>2, Mn₂C>₃, Mn₃0₄, MnO, Zn_xFe_yO_z, ZnFe2<D₄, Zn_xCo_yO_z, ZnCo₂0₄, Zn_xMn_yO_z, ZnMn₂0₄, Co_xSn_yO_z, Co₂Sn0₄, Mn_xSn_yO_z, Mn₂Sn0₄, Ni_xSn_yO_z, NhSnCU, Co_xSn_yO_z, CoSnCb, Ni_xSn_yO_z, NiSnCb, Co_xSi_yO_z, Co₂Si0₄, Fe_xSi_yO_z, Fe₂Si0₄, Mn_xSi_yO_z, Mn2Si0₄, Sn_xO_y, Sn02, Ge_xO_y, Ge02, Sb_xO_y, Sb20s, Si_xO_y, S1O2, SiO_x, Co_xO_y, CoO,

Co₃0₄, Cu_xO_y, CuO, CU₂O, Ni_xO_y, NiO, Zn_xO_y, ZnO, Cr_xO_y, Cr₂0₃, Mo_xO_y, M0O₃, M0O₂, Ru_xO_y, RUO2, Sn, Si, Fe_xSi, Cu_xSi, Li_xTi_yO_z, Li₂Ti03, Nb_xO_y, Nb20s, Nb_xM_yO_z (M = Fe, Ti,

V, Mo, Zr, Mg, W, Ga, K), LiNiCoMn02, LiNiCoAI02, LiMn20₄, LiNio.5Mni.50₄, UC0O2, LiFeP0₄, and their combinations.

Where compositions are expressed in terms such as Zn_xFe_yO_z and the like, it is to be understood that x, y and z are selectable for that combination of elements independently of other compositions listed.

In a second preferred aspect, the present invention provides a method for the

manufacture of an electrode active material for an electrochemical device, the method comprising:

providing a dispersion of nanoparticles in a carrier liquid, the nanoparticles having an average diameter in the range 2-1000 nm;

including in the dispersion a carbon precursor material that is soluble and/or dispersible in the carrier liquid and optionally a dispersant to aid in the effective dispersion of nanoparticles;

spray drying the dispersion to form precursor granules of nanoparticles with the carbon precursor material distributed in the precursor granules;

subjecting the precursor granules to a heat treatment to degrade the carbon precursor material, the heat treatment thereby providing granules of nanoparticles, wherein the nanoparticles are at least partially sintered together, the granules having an average diameter in the range 1-25 pm, the granules being porous at least in part due to the degradation of the carbon precursor material, the granules comprising a coating of carbon formed at least at the surfaces of pores in the granules from the degradation of the carbon precursor material,

wherein the nanoparticles comprise one or more of Fe_xO_y, Fe2C>3, Fe30₄, FeO, Mn_xO_y, MnC>2, Mn₂0₃, Mn30₄, MnO, Zn_xFe_yO_z, ZnFe2<D₄, Zn_xCo_yO_z, ZnCo20₄, Zn_xMn_yO_z,

ZnMn₂0₄, Co_xSn_yO_z, Co₂Sn0₄, Mn_xSn_yO_z, Mn₂Sn0₄, Ni_xSn_yO_z, Ni₂Sn0₄, Co_xSn_yO_z, CoSnCb, Ni_xSn_yO_z, NiSnOs, Co_xSi_yO_z, Co2Si0₄, Fe_xSi_yO_z, Fe2Si0₄, Mn_xSi_yO_z, Mn2Si0₄, Sn_xO_y, SnC>2, Ge_xO_y, Ge0₂, Sb_xO_y, Sb₂0s, Si_xO_y, S1O2, SiO_x, Co_xO_y, CoO, Co30₄, Cu_xO_y, CuO, CU2O, Ni_xO_y, NiO, Zn_xO_y, ZnO, Cr_xO_y, Cr203, Mo_xO_y, M0O3, M0O2, Ru_xO_y, RUO2,

Sn, Si, Fe_xSi, Cu_xSi, Li_xTi_yO_z Li₂Ti03, Nb_xO_y, Nb₂0s, Nb_xM_yO_z (M = Fe, Ti, V, Mo, Zr, Mg, W, Ga, K), LiNiCoMn02, LiNiCoAI02, LiMn20₄, LiNio.5Mni.50₄, UC0O2, LiFeP0₄ and their combinations.

In a third preferred aspect, the present invention provides an electrochemical device comprising an anode, a cathode and an electrolyte disposed between the anode and the cathode, wherein the anode comprises an electrode active material according to the first aspect.

In a fourth preferred aspect, the present invention provides a use of an electrode active material according to the first aspect as an anode active material, or a component of an anode active material, in an anode in conjunction with a cathode and an electrolyte in a lithium ion battery for charging and discharging of the lithium ion battery.

It is to be understood that the electrode active material may be a component in the anode. Further components may also be present, such as known materials for anodes such as graphite.

In a fifth preferred aspect, the present invention provides a method for processing an electrode active material according to the first aspect as or in an anode active material for a lithium ion battery, the method including diffusing lithium ions into the anode active material.

The first, second, third, fourth and/or fifth aspect of the invention may have any one or, to the extent that they are compatible, any combination of the following optional features.

As will be understood, the partial sintering of the nanoparticles together helps to provide the granule with some structural integrity. However, avoiding complete sintering of the nanoparticles provides the granule with open porosity that is useful for the electrode active material. The partial sintering is considered to provide a useful balance between continuity of a network of the active material within the granule and a suitable open porosity to permit transport of Li ions towards active material in the interior of the granules and not only at the exterior of the granules.

In the present disclosure, unless stated otherwise, the average particle size is to be understood as the d50 particle size. Preferably, the particle size distribution (e.g. for the nanoparticles) is relatively narrow. For example, the particle size distribution (e.g. for the nanoparticles) may be such that particle diameter d10 is at least 0.75 times d50 and d90 is at most 1 .75 times d50.

It will further be understood, in the light of the present disclosure, that the granule surface area is dependent on the nanoparticle particle size. In the case where the nanoparticles are perfectly spherical, the surface area to volume ratio of a sphere is 6/d (where d is the diameter of the sphere). Accounting for the partial sintering of the nanoparticles, their practical non-sphericity and typical particle size distribution, the surface area per unit volume of the granules is preferably at least 1/d, more preferably at least 2/d, still more preferably at least 3/d. d may be expressed as the d50 average nanoparticle size. Here, the volume of the granules is intended to be taken as the volume excluding porosity. It may be determined based on knowledge of the mass of the granules and the

composition of the granules, the composition of the granules being used to express the material density of the granules (i.e. the density of a fully dense material with zero porosity). The particle size of the nanoparticles in the granules can be determined by electron microscopy.

The lower limit for the average diameter of the nanoparticles may be 5 nm or 10 nm. The upper limit for the average diameter of the nanoparticles may be 500 nm, 400 nm, 300 nm, 200 nm or 100 nm.

The lower limit for the average diameter of the granules may be 0.2 pm, 1 pm, 2 pm, 4 pm or 5 pm. The upper limit for the average diameter of the granules may be 50 pm, 40 pm, 30 pm, 20 pm or 10 pm.

The BET surface area of the granules may be 1 -30 m²/g. Within this range, the BET surface area may be at least 2 m²/g, at least 5 m²/g, at least 10 m²/g, at least 15 m²/g, or at least 20 m²/g. Alternatively, the inventors find that in some circumstances the BET surface area should be kept in the lower part of this range. Accordingly, in some embodiments, the BET surface area of the granules may be at most 25 m²/g, at most 20 m²/g, at most 15 m²/g, or at most 10 m²/g.

The coating of carbon may comprise graphitic carbon. This can be assessed using Raman spectroscopy. The graphitic carbon may comprise graphene. For example, the coating may comprise a single layer of graphene or few layer graphene (FLG). Non- graphitic carbon may also be present, but it is preferred that at least a proportion of the coating of carbon comprises graphitic carbon.

The pores in the granules may have an average diameter in the range 5-500 nm.

Typically, this value will scale with the diameter of the nanoparticles. The granules may comprise a pore volume of at least 0.01 cm³/g, or at least 0.02 cm³/g. The pores in the granules may provide continuous void channels extending continuously across the whole granule. The pores in the granules may include at least one pore with a pore size of at least 1 pm. Alternatively (or additionally) the pores in the granules may include at least one pore with a maximum pore size of at least 30 times the average nanoparticle diameter.

The surface of the granules may have an average fractal dimension of at least 1.7. The surface of the granules may have a local radius of curvature that is greater than the nanoparticle radius at all points on the surface of the granule.

The nanoparticles may form a continuous but only partially sintered network throughout each granule.

The granule size distribution may satisfy one or both of the relationships:

d10 > d50/2

d90 < 3 times d50

where d 10, d50 and d90 apply to the granule particle size distribution.

The carbon coating may be formed in an amount of 0.1-20 wt% or more preferably in an amount of 0.1-15 wt% based on the weight of the granules. The proportion of carbon may be measured by TGA which may be performed in air. The lower limit of the amount of carbon may be 1 wt%. The carbon coating may be formed in an amount of 1-10 wt% based on the weight of the granules. Preferably, the carbon coating conforms to the pore surfaces. In this way, the carbon coating forms a protective yet conductive layer over the available surface of the conversion material of the electrode active material. The thickness of the carbon coating may be in the range 0.3-100 nm. This can be measured, for example, using TEM.

The material of the nanoparticles may be an electrochemical conversion material, electrochemical alloying material or electrochemical conversion/alloying material.

Alternatively, the material of the nanoparticles may be a 3D intercalation material.

In some embodiments, the nanoparticles comprise a metal oxide, such as a transition metal oxide. The BET surface area of the granules may be at least 1.2 times the BET surface area of the precursor granules. For example, the BET surface area of the precursor granules may be at most 15 m²/g. For example, the BET surface area of the granules may be at least 20 m²/g. The BET surface area of the granules may be 1-30 m²/g.

The precursor granules may comprise a pore volume of at most 0.05 cm³/g. On the other hand, the granules may comprise a pore volume of at least 0.01 cm³/g, or at least 0.02 cm³/g. However, when considering granules manufactured from specific precursor granules, the granules preferably have a larger pore volume than the precursor granule.

In the method, dispersion of the nanoparticles may be achieved by providing a force to make a homogeneous dispersion by, but not limited to: homogenization, ball-milling, ultrasonication, probe sonication, mechanical stirring, micronisation, high-shear milling, planetary mixing.

In the method, the heat treatment may comprise heating to a temperature in the range 300-1200°C in an inert or reducing atmosphere, in which case the lower limit of temperature in this step may be at least 350°C, at least 400°C, or at least 450°C.

Alternatively, the heat treatment may comprise heating to a temperature in the range 450-1350°C in an inert or reducing atmosphere.

In the method, a further carbon coating may be applied so as to reduce the BET surface area of the granules. The further carbon coating may include milling carbon in the presence of the granules. The further carbon coating may be based on pitch carbon.

Further optional features of the invention are set out below.

Other aspects and embodiments of the invention provide the aspects and embodiments described above with the term“comprising” replaced by the term“consisting of” and the aspects and embodiments described above with the term“comprising” replaced by the term’’consisting essentially of.

“and/or” where used herein is to be taken as specific disclosure of each of the two specified features or components with or without the other. For example“A and/or B” is to be taken as specific disclosure of each of (i) A, (ii) B and (iii) A and B, just as if each is set out individually herein. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described by way of example with reference to the accompanying drawings in which:

Fig. 1 shows an SEM image of spray dried precursor granules comprising 30nm Fe₂0₃ nanoparticles and maltodextrin.

Fig. 2 shows an SEM image of granules comprising 30nm Fe₂0₃ nanoparticles and maltodextrin after heat treatment in N₂ at 550°C, thus carbonising the maltodextrin and reducing the Fe₂C>3 to Fe30₄.

Fig. 3 shows a higher magnification image of a granule according to Fig. 2.

Fig. 4 shows volume-based particle size distribution and cumulative distribution by laser diffraction sizing for precursor granules formed from 30 nm Fe₂03 nanoparticles and maltodextrin.

Fig. 5 shows volume-based particle size distribution and cumulative distribution by laser diffraction sizing for heat treated granules of Fe₃0₄-carbon granules formed by heat treatment of the precursor granules of Fig. 4.

Fig. 6 shows an XRD pattern for a heat-treated sample containing nanocrystallites of Fe3<D₄.

Fig. 7 shows cycling data of an active material according to an embodiment based on iron oxide particles (using maltodextrin as the carbon precursor material) in a half cell configuration, the graph showing the specific capacity versus the number of cycles at different charging rates.

Fig. 8 shows the galvanostatic charge/discharge profiles of the material reported in Fig. 7. Fig. 9 shows an SEM image of granules comprising 30nm Fe₂0₃ nanoparticles and sucrose after heat treatment in N₂ at 550°C, thus carbonising the sucrose and reducing the Fe₂C>3 to Fe30₄.

Fig. 10 shows a higher magnification images of granules as described with respect to Fig. 9

Fig. 11 shows a further SEM image of granules comprising 30nm Fe₂0₃ nanoparticles and sucrose after heat treatment in N₂ at 550°C, thus carbonising the sucrose and reducing the Fe₂C>3 to Fe30₄.

Fig. 12 shows cycling data of an active material according to an embodiment based on iron oxide particles (using sucrose as the carbon precursor material) in a half cell configuration, the graph showing the specific capacity versus the number of cycles at different charging rates.

Fig. 13 shows the galvanostatic charge/discharge profiles of the material reported in Fig. 12.

Fig. 14 shows TGA results for Fe3<D₄-carbon granules manufactured using maltodextrin as the carbon precursor material. TGA was carried out in air, resulting in oxidation and combustion taking place. Fig. 15 shows TGA results for Fe₃0₄-carbon granules manufactured using sucrose as the carbon precursor material. TGA was carried out in air, resulting in oxidation and combustion taking place.

Fig. 16 shows a Raman spectrum for commercial graphite, for comparison.

Fig. 17 shows a Raman spectrum for maltodextrin 17-19.9 DE, used as a carbon precursor material in some embodiments.

Fig. 18 shows a Raman spectrum for Fe₃0₄ nanoparticles of average diameter 8 nm.

Fig. 19 shows a Raman spectrum for Fe₂0₃ nanoparticles of average diameter 30 nm.

Fig. 20 shows a Raman spectrum for spray dried precursor granules of 30 nm Fe₂0₃ nanoparticles with maltodextrin.

Fig. 21 shows a Raman spectrum for spray dried precursor granules of 30 nm Fe₂0₃ nanoparticles with sucrose.

Fig. 22 shows a Raman spectrum for granules formed by low temperature heat treatment of spray dried granules formed of 30 nm Fe₂0₃ nanoparticles with maltodextrin.

Fig. 23 shows a Raman spectrum for granules formed by high temperature heat treatment of spray dried granules formed of 30 nm Fe₂0₃ nanoparticles with maltodextrin. Fig. 24 shows a Raman spectrum for granules formed by high temperature heat treatment of spray dried granules formed of 30 nm Fe₂0₃ nanoparticles with sucrose.

Fig. 25 shows cycling data of an active material for comparison based on iron oxide particles (using maltodextrin as the carbon precursor material) subjected to a low temperature heat treatment (350°C) in a half cell configuration, the graph showing the specific capacity versus the number of cycles at different charging rates.

Fig. 26 shows N₂ physisorption isotherms according to BET analysis of different materials: 30 nm Fe₂0₃ nanoparticles; spray dried precursor granules of 30 nm Fe₂0₃ nanoparticles with maltodextrin; Fe₃0₄-C granules formed by high temperature heat treatment of spray dried granules formed of 30 nm Fe₂0₃ nanoparticles with maltodextrin. Fig. 27 shows pore size distribution results based on the N₂ physisorption isotherms of Fig. 26.

Fig. 28a shows an SEM image of a transversal section of granules of S1 obtained by focused ion-beam milling (“FIB”).

Fig. 28b shows an SEM image of a transversal section of granules of S2 obtained by focused ion-beam milling (“FIB”).

Fig. 29a shows cycling data of S1 in a half coin cell vs Li/Li+ at a constant charging rate.

Fig. 29b shows cycling data of S1 in a half coin cell vs Li/Li+ at varying charging rates.

Fig. 30 shows cycling data of S2 in a half coin cell vs Li/Li+ at a constant charging rate. Fig. 31 a shows TEM images of S1.

Fig. 31 b shows TEM images of S3.

Fig. 32a shows a schematic representation of a granule of S1.

Fig. 32b shows a schematic representation of a granule of S3.

Fig. 33 shows cycling data of S3 in a half coin cell vs Li/Li+ at a constant charging rate. Fig. 34a shows an SEM image of S1. Fig. 34b shows an SEM image of S4.

Fig. 35a shows a schematic representation of a network of particles in S1.

Fig. 35b shows a schematic representation of a network of particles in S4.

Fig. 36a shows cycling data of S4 in a half coin cell vs Li/Li+ at a constant charging rate. Fig. 36b shows cycling data of S4 in a half coin cell vs Li/Li+ at varying charging rates.

Fig. 37a shows the granule particle size distributions, measured by laser diffraction, of S1. Fig. 37b shows the granule particle size distributions, measured by laser diffraction, of S5. Fig. 38 shows cycling data of S5 in a half coin cell vs Li/Li+ at a constant charging rate. Fig. 39a shows a TEM image of S1.

Fig. 39b shows a TEM image of S1.

Fig. 40 shows an SEM image of S6, which was heat treated at 700 °C.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS. AND FURTHER

OPTIONAL FEATURES OF THE INVENTION

In summary of the present disclosure, granules primarily composed of metal oxides or other materials as listed above and a carbon coating are manufactured by industrially scalable spray-drying and heat treatment processes. The granules are then used as electrode materials in rechargeable Li-ion batteries. Due to their nanostructured architecture, the granules address many of the typical difficulties associated with metal oxides as Li-ion battery materials. This enables operation of Li-ion cells at high capacities and high charging rates, with long cycle lifetimes, which is highly desirable in various industries.

Before setting out the preferred embodiments of the present disclosure in detail, further published work is first discussed, since it provides a useful background with which to understand the present disclosure.

Atchudan et al (2016) discloses the synthesis of 9 nm Fe₃0₄ nanoparticles, and use these as individual templates from which to grow graphitic carbon layers. The metal oxide template is then removed by acid treatment to result in hollow carbon nanospheres. To grow the graphitic layers, the nanoparticles are simply ground together with sucrose, and heat treated in a tube furnace at 450°C for 2 hours under a 5% Fh in Ar atmosphere. The nanoparticles act as catalysts for this low temperature growth of graphite layers.

Zhang et al (2015) discloses an anode material based on a composite of a-Fe2C>3 and graphitic carbon. This material is manufactured in a 3-step process. First, a mixture of iron citrate salt and carbon precursor sucrose is spray dried to obtain particles of both, homogeneously mixed. These particles are then carbonised by heat treatment at 700°C in N₂ for 5 hours to decompose the sucrose into graphitic carbon and the iron citrate into a mixture of FesC and Fe₃0₄, followed by an additional heat treatment in air at 350°C for 5 hours to mostly oxidise the metal back to a-Fe203.

Jia et al (2015) discloses that a slurry of nanoparticles, conductive carbon material (carbon nanotubes, graphene), sucrose, and a polymer surfactant was atomised into an aerosol. The aerosol was then dried in a tube furnace under N₂ at 450°C to result in mesoporous composite particles. This overall process is termed spray pyrolysis. In some cases, a further heat treatment step was carried out at 560°C to enhance particle sintering. This paper aims to achieve a homogeneous dispersion of the CNTs and Ti0₂ nanoparticles within the resultant particles, and aims to demonstrate fast-charging capabilities of the materials in Li-ion batteries.

CN103435105 discloses an iron oxide/carbon composite lithium ion battery anode material. A soluble iron salt is used in combination with ammonia water prior to spray- drying precursor particles, followed by heat treatment in a furnace.

According to an embodiment, granules are manufactured as follows, using a two-step, industrially scalable process.

First, nanoparticles of diameter 2-1000 nm (i.e. transition metal oxide materials such as Fe_xO_y, Fe₂C>3, Fe30₄, FeO, Mn_xO_y, Mn0₂, Mn₂C>3, Mn30₄, MnO, Zn_xFe_yO_z, ZnFe₂0₄, Zn_xCo_yOz, ZnCo₂0₄, Zn_xMn_yO_z, ZnMn₂0₄, Co_xSn_yO_z, Co₂Sn0₄, Mn_xSn_yO_z, Mn₂Sn0₄, Ni_xSn_yO_z, Ni₂Sn0₄, Co_xSn_yO_z, CoSnCb, Ni_xSn_yO_z, NiSnCb, Co_xSi_yO_z, Co₂Si0₄, Fe_xSi_yO_z, Fe₂Si0₄, Mn_xSiyO_z, Mn₂Si0₄, Sn_xO_y, Sn0₂, Ge_xO_y, Ge0₂, Sb_xO_y, Sb₂Os, Si_xO_y, Si0₂, SiO_x, Co_xO_y, CoO, Co₃0₄, Cu_xO_y, CuO, Cu₂0, Ni_xO_y, NiO, Zn_xO_y, ZnO, Cr_xO_y, Cr₂0₃, Mo_xO_y, M0O3, MO0₂, Ru_xO_y, RU0₂, Sn, Si, Fe_xSi, Cu_xSi, Li_xTi_yO_z Li₂Ti03, Nb_xO_y, Nb₂Os, Nb_xM_yO_z (M = Fe, Ti, V, Mo, Zr, Mg, W, Ga, K), LiNiCoMn0₂, LiNiCoAI0₂, LiMn₂0₄, LiNi_0.5Mni_.5O₄, LiCo0₂, LiFeP0₄ and their combinations) and a water-soluble and/or dispersible carbon- based additive, as the carbon precursor material (e.g. a carbohydrate such as

maltodextrin, dextrin, sucrose, lactose, hydroxyethylcellulose ; or a dispersible carbon such as carbon black, Super P, graphene, graphene oxide) were mixed into a slurry at total concentrations ranging from 0-50 wt%. Preferably the slurry is water based, but mixtures with water-miscible solvents can be used to aid dispersion such as ethanol.

A surfactant (stabiliser, or dispersant) may be included in the slurry to help dispersion of the nanoparticles. Preferably, the surfactant is composed only of C, H, N, O, and optionally Li so it can be removed by the subsequent heat treatment. Optionally, other conductive carbon materials may be included in the slurry to increase conductivity in the resultant granules where required. A suitable additive in this case is carbon black or reduced graphene oxide. In order to make a homogeneous slurry, energy typically must be input by an appropriate method to effectively disperse the nanoparticles. This can be by, but not limited to: homogenization, ball-milling, ultrasonication, probe sonication, mechanical stirring, micronisation, high-shear milling, and planetary mixing.

This slurry was then spray dried in water in a commercial spray dryer, with typical outlet temperatures of 90-1 10°C. Depending on the concentration of the slurry and other routine operating conditions of the spray dryer, the spray drying results in mostly dry precursor granules of diameter in the range 0.05 - 500 pm. After suitable control, the particle size and distribution can be controlled to the desired range.

After spray drying, the precursor granules were heat treated (450-1350°C) in a commercial furnace. Suitable furnaces include rotary, static, or fluidised bed types. The atmosphere for the heat treatment was varied, using air, inert gases such as N₂/He/Ar, and optionally with a proportion of H₂ (0-50 %). The heat treatment was carried out at atmospheric pressure. With heat treatment in air, this results in the removal of the organic carbon additive through combustion if above the thermal decomposition temperature. With heat treatment in an inert or reducing atmosphere, carbonisation (pyrolysis) occurs, by which the carbon additive is degraded to leave a carbon film. The granules become more porous as a result. The structure of the granules is based on a framework of the nanoparticles, which are at least partially sintered as a result of the heat treatment. As discussed in more detail later, the porosity of the granules is relevant for their use in Li-ion batteries due to enhanced ion diffusion, and provides granules that are resistant to degradation such as by volume expansion effects. The carbon coating additionally provides a high electrical and thermal conductivity, mechanical stability, and reduces the formation of the SEI. In some embodiments a further heat treatment step at lower temperatures (100-600°C, optionally in the presence of 0₂, e.g. air) was carried out to adjust the oxidation state or crystallite microstructure of the metal oxide (e.g. from Fe30₄ to Fe₂0₃, or from small crystallites to large crystallites).

The properties of the granules were tested in the context of their use as an electrode active material in Li ion batteries. Battery testing has been carried out in research scale coin cells in the half-cell configuration versus Li metal, as is standard practice. The materials tested were mostly targeted for use as negative electrode materials (anode material) for full cells (in combination with suitable electrolytes, separators, and positive electrodes). However, it is to be noted that the material may be used as a positive electrode material (cathode material) in some embodiments.

Upon spray drying the slurry of metal oxide nanoparticles and organic carbon materials, dry precursor granules in the nanometre-micrometre (10 ⁸ m to 10^-4 m) diameter size range were obtained. Spray-drying is a process by which a fine aerosol of liquid droplets are rapidly dried in a flow of hot gas, widely used in industry and well-respected for its scalability. See Vehring et al (2007) and Vehring (2008). When appropriate conditions are employed in the spray drying process, narrow distributions of precursor granule sizes can be obtained as desired (e.g. 2-5 pm range). The granules are a homogeneous mixture of the metal oxide nanoparticles, interspersed with the organic carbon additives. More advanced mixtures are obtainable through controlling conditions and precursors appropriately [Ogi et al (2014)]. This is exemplified with 30 nm Fe₂0₃ nanoparticles and maltodextrin in Fig. 1.

Upon heat treatment of the precursor granules, the organic material is removed either completely by combustion in air or converted to a conductive amorphous and/or graphitic carbon thin film on the surface of the metal oxide particles (carbonisation). Carbonisation to achieve graphite would typically require temperatures over 2000°C and extended time scales, if the precursor granules were purely organic precursors. However, in the presence of catalytic (e.g. metal oxide) nanomaterials, this temperature can be as low as 450°C and the heat treatment completed within minutes. This is confirmed in the academic literature (see Atchudan et al (2016) and Zhang et al (2015)).

In either case, the particles become highly porous due to the removal of the majority of organic material. This is exemplified in Figs. 2 and 3, where the same sample as in Fig. 1 has been heat treated in inert gas. The increased porosity is thought to allow rapid Li ion diffusion throughout the particle, thus enabling fast charge and discharging of a full Li ion battery. This is supported by the battery cycling data at different charge/discharge rates, reported below, where a very low drop in capacity is observed with increasing

charge/discharge rates. This is even more clearly seen in the results in which sucrose is used as the carbon material precursor.

The increased porosity in the heat-treated granule also improves the cycle life as the architecture allows for volume expansion to occur without degradation of the electrode (such degradation may otherwise occur for example through electrode pulverisation, or irreversible capacity losses due to excess SEI formation). The presence of a carbon film increases the conductivity and further prevents degradation through providing

reinforcement and an unreactive substrate for the SEI to form upon.

Depending on the heat treatment conditions, it is possible to control the oxidation state of the metal, where the granules comprise metal oxide. This is confirmed by simple measurement of powder X-Ray Diffraction (XRD) patterns as in Fig. 6. Organic carbon materials and metal oxides can catalytically affect each other’s properties as previously mentioned for carbonisation, allowing control over the oxide present, which is beneficial for their different performance as Li ion battery electrode materials (e.g. Fe₂0₃ versus Fe30₄). The size of individual crystallite size and distribution can be controlled similarly with heat treatment, which has a direct effect on Li ion battery performance. Once heat treatment is complete, the granules can be combined with standard Li ion battery materials to form a cell. The active electrode material can simply be a replacement to current powdered electrode materials such as graphite. This is a key advantage of the present disclosure: streamlined integration to current manufacturing processes.

The electrode active material is coated on a current collector to form an electrode. This is carried out in combination with a solvent (e.g. water, NMP, etc.), a binder (e.g. PVDF, CMC-SBR, sodium alginate, etc.), and where desired a conductive additive (e.g. carbon black, carbon nanotubes, etc.). It is preferred that a relatively small amount of binder is used, e.g. not more than 5 wt% binder based on the weight of the granules. However, in alternate embodiments, it is possible to use up to 10 wt% binder or 10-20 wt% binder. After mixing/grinding as required, the viscous mixture is coated onto the current collector (e.g. Cu foil for a negative electrode, Al foil for a positive electrode). On a laboratory scale this is carried out for example by doctor blade coating and calendaring, and on large scales this can be carried out by processes such as roll-to-roll coating. The coated electrode is then typically dried in an oven under vacuum to enhance adherence to the current collector and to remove the solvent.

The electrode containing the active electrode material is then combined with the opposite electrode (i.e. a positive and negative electrode). Between the electrodes there is a separator to prevent a short circuit (e.g. PP/PE). The separator is soaked in a suitable electrolyte in an inorganic solvent (e.g. 1 M LiPF₆ in EC/DEC). Finally, the whole system is encapsulated in a cell casing.

The battery testing reported here is for the half-cell configuration, in coin cells. This means the electrode of the active material is paired with Li metal as opposed to its opposite electrode, thus enabling full assessment of its fundamental properties.

Results for one of these cells is shown in the half-cell configuration in Figs. 7 and 8, at different charge/discharging rates. The slurry for coating was prepared in water with the active material, CMC-SBR and Super P in an 8:1 :1 ratio. The electrolyte used was 1 M LiPF₆ in EC/DEC + 10 wt% FEC.

The Initial Coulombic Efficiency (ICE) was 78%, which implies the amount of unwanted side reactions leading to a loss of capacity after the first cycle (besides typical SEI formation) are significantly minimised, a rare achievement for metal oxide conversion materials. The specific capacity is high for a mass-producible material, but it is at the time of writing considered that the most significant advantage is the excellent cycle stability and limited capacity drop at high charge/discharge rates. This indicates that these materials are particularly advantageous as electrode active materials for Li-ion batteries.

Fig. 9 shows an SEM image of granules comprising 30nm Fe₂0₃ nanoparticles and sucrose after heat treatment in N₂ at 550°C, thus carbonising the sucrose and reducing the Fe₂0₃ to Fe₃0₄. Fig. 10 shows a higher magnification images of granules as described with respect to Fig. 9. Fig. 1 1 shows a further SEM image of granules comprising 30nm Fe₂0₃ nanoparticles and sucrose after heat treatment in N₂ at 550°C, thus carbonising the sucrose and reducing the Fe₂0₃ to Fe₃0₄.

Fig. 12 shows cycling data of an active material according to an embodiment based on iron oxide particles (using sucrose as the carbon precursor material) in a half cell configuration, the graph showing the specific capacity versus the number of cycles at different charging rates. The use of sucrose rather than maltodextrin gives a more graphitic coating and, it is considered, a more homogeneous conformal coating. Fig. 13 shows the galvanostatic charge/discharge profiles of the material reported in Fig. 12.

The present inventors have undertaken additional work to investigate the nature of the carbon coating formed on the granules and also to investigate further details of the structure of the granules.

Fig. 14 shows TGA results for Fe₃0₄-carbon granules manufactured using maltodextrin as the carbon precursor material. TGA was carried out in air, resulting in oxidation and combustion taking place. The increase in mass starting around 150°C relates to Fe₃0₄ oxidising to Fe₂0₃, and the decrease in mass relates in the combustion of the carbon component present. Fig. 15 shows TGA results for Fe₃0₄-carbon granules manufactured using sucrose as the carbon precursor material. TGA was carried out in air, resulting in oxidation and combustion taking place. The increase in mass starting around 150°C relates to Fe₃0₄ oxidising to Fe₂0₃, and the decrease in mass relates in the combustion of the carbon component present.

Fig. 16 shows a Raman spectrum for commercial graphite, for comparison. Fig. 17 shows a Raman spectrum for maltodextrin 17-19.9 DE, used as a carbon precursor material in some embodiments. Fig. 18 shows a Raman spectrum for Fe₃0₄

nanoparticles of average diameter 8 nm. Fig. 19 shows a Raman spectrum for Fe₂0₃ nanoparticles of average diameter 30 nm.

Fig. 20 shows a Raman spectrum for spray dried precursor granules of 30 nm Fe₂0₃ nanoparticles with maltodextrin. Fig. 21 shows a Raman spectrum for spray dried precursor granules of 30 nm Fe₂0₃ nanoparticles with sucrose. Fig. 22 shows a Raman spectrum for granules formed by low temperature heat treatment of spray dried granules formed of 30 nm Fe₂03 nanoparticles with maltodextrin. This lower temperature sample appears to be more amorphous than the sample reported in Fig. 23, with mixed Fe₂C>3 and Fe30₄ species present.

Fig. 23 shows a Raman spectrum for granules formed by high temperature heat treatment of spray dried granules formed of 30 nm Fe₂0₃ nanoparticles with maltodextrin. This indicates that the carbon coating is a mixed graphitic/amorphous carbon film. Fig.

24 shows a Raman spectrum for granules formed by high temperature heat treatment of spray dried granules formed of 30 nm Fe₂03 nanoparticles with sucrose. The carbon coating is shown to be more graphitic than for granules formed using maltodextrin.

Zhang et al (2013) is an example literature paper with Raman data Fe3<D₄ and graphene mixtures. Ferrari (2007) is an example literature paper with Raman data of carbon nanotubes versus graphite versus amorphous carbon.

Based on the Raman analysis provided here and compared with the example literature papers, it can be seen that the present embodiments achieve a carbon coating that is at least partially graphitic. This is significant in view of electrical conductivity advantages provided by graphitic (as opposed to amorphous) carbon, and in terms of providing a protective layer between the electroactive material (e.g. conversion material) and the electrolyte, and for increasing the mechanical integrity.

Fig. 25 shows cycling data of an active material for comparison based on iron oxide particles (using maltodextrin as the carbon precursor material) subjected to a low temperature heat treatment (350°C) in a half cell configuration, the graph showing the specific capacity versus the number of cycles at different charging rates. This indicates the effect of low temperature heat treatment on performance.

Fig. 26 shows N₂ physisorption isotherms according to BET analysis of different materials. These materials are: 30 nm Fe₂C>₃ nanoparticles; spray dried precursor granules of 30 nm Fe₂0₃ nanoparticles with maltodextrin; Fe3<D₄-C granules formed by high temperature heat treatment of spray dried granules formed of 30 nm Fe₂0₃ nanoparticles with maltodextrin. Fig. 27 shows pore size distribution results based on the N₂ physisorption isotherms of Fig. 26. The data is also summarised in Table 1. Table 1 - BET analysis

Surface Area Pore Volume Main pore size

(m² g-¹) (cm³ g-¹) (nm)

Fe2C>3 NPs 36 0.1 1 50

Fe2C>3 Spray dried 9 0.02 43

Fe30₄ - C 42 0.09 51

In Figs. 26 and 27, and in Table 1 ,“Fe2<D3 NPs” refers to the commercial alpha-Fe₂<D₃ 30 nm feedstock used, and the data is here as a comparison to a nanoparticle powder. “Fe2C>3 Spray dried” refers to the precursor granules formed after spray drying, composed of a fine mixture of the nanoparticles and a carbohydrate (maltodextrin in this case). “Fe₃0₄ - C” refers to the granules after heat treatment in inert gas, resulting in sintering and carbonisation.

As will be understood, assessment of BET surface area of nanoparticles is subject to the packing arrangement of the nanoparticles. Therefore it does not necessarily reflect the total cumulative geometric surface area of the nanoparticles. In part, this helps to explain why the BET surface are of the nanoparticles is shown as a value (expressed in terms of m² g^·1) that is lower for the nanoparticles than for the granules after heat treatment. Additionally, the proportion of C present will also affect the surface area per unit mass, in view of the density of C compared with the density of the nanoparticles.

A decrease in specific surface area and porosity is seen after spray drying, which is a result of taking the nanoparticles and making them into larger precursor granules with a carbohydrate matrix. After heat treatment, the majority of the carbohydrate is removed and a carbon film remains around the sintered nanoparticles. As the granules are now highly porous, we see a large increase in the specific surface area.

Fig. 4 shows volume-based particle size distribution and cumulative distribution by laser diffraction sizing for precursor granules formed from 30 nm Fe₂<D₃ nanoparticles and maltodextrin. Fig. 5 shows volume-based particle size distribution and cumulative distribution by laser diffraction sizing for heat treated granules of Fe3<D₄-carbon granules formed by heat treatment of the precursor granules of Fig. 4.

Table 2 provides particle size distribution data for precursor granules formed from 30 nm Fe2<D3 nanoparticles and maltodextrin and also for heat treated granules of Fe3<D₄-carbon granules formed by heat treatment of the precursor granules. This data is taken from Figs. 4 and 5. Table 2 - particle size distribution for precursor granules and heat treated granules

Sample Dio /pm D₅o /pm D₉o /pm

Spray dried 1.52 6.15 34.89

Heat treated 3.68 11.75 37.16

With the benefit of the discussion set out above of the embodiments of the disclosure, it is possible to consider how the disclosure compares with some of the prior art references mentioned above.

Considering Atchudan et al (2016), this document simply provides disclosure of taking Fe30₄, mixing it with organic carbon materials, and heat treating to achieve a carbon coating. Atchudan et al (2016) is not concerned with battery materials. In the present embodiments, in contrast, spray-dried precursor granules are typically a homogeneous mixture of many metal oxide nanoparticles and the carbon component. The heat treatment step in the embodiments here results in highly porous granules where the nanoparticles are no longer distinct structures and they have been partially sintered, upon which there is also provided a graphitic carbon coating.

Considering Zhang et al (2015), iron citrate is used a precursor. In contrast, the present embodiments use a slurry of metal oxide nanoparticles which is spray dried directly. This makes the present embodiments much more widely applicable to other materials (i.e. any metal oxide nanoparticle that is suitably electrochemically active). It is noted that iron citrate is also limited by its very poor solubility in water (0.1 M used here), whereas slurries can be spray dried up to 50 wt.%. It is also noted that the granules formed in the present embodiments, in view of the partial sintering of neighbouring particles, are significantly different in terms of morphology and porosity to the particles formed in Zhang et al (2015).

Considering Jia et al (2015), the present embodiments have various differences to the disclosure of Jia et al (2015). The present embodiments use a spray-drying process as opposed to spray pyrolysis. These are significantly different techniques. Spray-drying allows extremely high throughput (lab scale kg/day, industrial scale kiloton/year), and the drying occurs through interaction of a fine liquid aerosol with fast-moving hot gas. The solid granules are then collected in either a cyclone or a bag filter in large quantities. Spray pyrolysis, on the other hand, relies on slow atomisation, and drying in a furnace rather through interaction with hot gas. Collection then takes place on a filter, typically providing milligram quantities per day (at the laboratory scale). Furthermore, Jia et al (2015) aims to provide the inclusion and homogeneous dispersion of conductive carbon materials inside of the particles as synthesised, thereby increasing conductivity and mechanical strength for Li-ion batteries. In the present embodiments, there is instead a focus on the production of porous granules, where the heat treatment step forms a conductive carbon coating. This is a key advantage, as dispersion and control of the conductive carbon materials can often be a challenge, and so direct formation of conductive carbon on the granules in-situ is an easier and more widely applicable process.

Considering CN 103435105, in the present embodiments a slurry of metal oxide nanoparticles is used, rather than the soluble iron salt precursor of CN103435105.

Further work carried out by the present inventors is presented below. Specifically, the work relates to a comparative study demonstrating a relationship between granule characteristics and electrochemical properties of the electrode active material. The granule characteristics studied are: the size distribution of the granules, the

characterisation of the pores in the granules; the continuity of the primary particle network within the granules; and the geometry of the particle surface. The above characteristics and morphologies depend to varying extents on the degree of partial sintering between primary particles in the material.

For the study, six types of sample of an electrode active material (S1-S6) were created using different methodologies.

Common to all the methodologies, was the spray drying of a carbon precursor and active material feedstock as described previously, followed by a heat treatment in a rotary batch furnace for two hours at 550°C in a nitrogen atmosphere. The differences between the sample types are the composition of the feedstock and the manner in which the feedstock dispersion was prepared. These differences are illustrated in Table 3.

Table 3

Sample Feedstock preparation details Comments

Fe₂0₃ 30 nm nanoparticles 10 w.%+

multilayer graphene oxide 0.2 w.%+ sucrose

51 4 w.% + ammonium oleate 0.1 w.% Optimum reference sample dispersed in water with a T25 homogeniser

(25000 rpm for 30 min)

Fe₂0₃ 30 nm nanoparticles 10 w.%+ sucrose

52 4 w.% + ammonium oleate 0.1 w.% Compact granules, used in dispersed in water with a planetary ball mill study of pores

operated at 400 rpm for 30 min

Fe₂0₃ 30 nm nanoparticles 10 w.%+ sucrose

4 w.% + ammonium oleate 0.1 w.% Highly curved, non-spherical

53

dispersed in water with low energy granules, used in study of mechanical mixing (magnetic stirrer bar, granule surface geometry

-120 rpm)

Fe₂03 30 nm nanoparticles 10 w.%+

multilayer graphene oxide 1 w.%+ sucrose 4 Non-continuous network of ^ w.% + ammonium oleate 0.1 w.% dispersed active material dispersed in in water with a T25 homogeniser (25000 rpm carbon matrix, used for study for 30 min) of network continuity

Fe₂0₃ 30 nm nanoparticles 50 w.%+ sucrose

Very polydisperse sample used 50 w.% ammonium oleate 0.1 w.% dispersed

S5 for study of granule size

in water with low energy mechanical mixing

distribution

(magnetic stirrer bar, -120 rpm)

Complete sintering of the 30

Granules of spray-dried Fe₂0₃ 30 nm

nm nanoparticles results in

S6 nanoparticles and maltodextrin heat treated

formation of bulk crystals with at 700°C for 1 h

no electrochemical activity

The samples S1-S6 had different granule morphologies. These differences are briefly discussed in the“Comments” column of Table 3.

Pore characterisation

To study the effects of pore characteristics on electrochemical performance, samples S1 and S2 are compared. SEM imaging of transversal sections of granules of S1 obtained by focused ion-beam milling (“FIB”) show that S1 has a network of pores that includes continuous void channels extending continuously across the whole granule, as shown in Figure 28a, whereas S2 displays only short range channels, as shown in Figure 28b.

Considering the same figures, large pores (> 1 pm/30x primary particle diameter) are present in S1 , and absent in S2; the largest pore that could be imaged in S2 is about 380 nm in size.

Figure 29a, Figure 29b and Figure 30 illustrate the respective cycling data of S1 and S2 in a half coin cell vs Li/Li+. In these Figures, the black points relate to lithiation

(discharging) and the white points relate to de-lithiation (charging). All other cycling data presented in this disclosure follows this convention. Specifically, the Figures illustrate graphs of specific capacity against number of cycles. Figure 29a relates to cycling data of S1 under a constant rate of charging/discharging (1 C). Figure 29b relates to cycling data of S1 under varying charging rates (0.1 C, 0.25C, 0.5C, 1 C, and 0.5C). Figure 30 shows cycling data of S2 under a constant charging rate (1 C).

The results show that S1 performs better than S2 in terms of specific capacity at low rate (about 800 mAh/g vs. about 600 mAh/g at 0.1 C), capacity retention at high rate (78% capacity retention vs. 70% capacity retention going from 0.1 C to 1 C rate), and cycle life (about 100% capacity retention vs. 8% capacity retention after 50 cycles).

Without wishing to be bound by theory, the inventors speculate that these differences in performance can be attributed to the different pore characteristics of the respective samples S1 and S2. The larger and continuous pores of S1 , compared to those of S2, allow for easier access to the whole volume of the granule for lithium ions upon lithiation, thus providing S1 with higher specific capacities than S2. Moreover, the

homogeneous/simultaneous lithiation of the whole granule in S1 , in contrast to the preferential lithiation of the shell of the granule in S2, results in fewer internal stresses within the granule of S1 compared to S2. Therefore, there is less pulverisation in S1 compared to S2, resulting in a higher cycle life for S1.

Granule surface geometry

Images and schematic illustrations of S1 and S3 are shown in Figures 31a, 31 b, 39a and 39b. Specifically, Figure 31a shows TEM images of S1 and Figure 31 b shows TEM images of S3. Figure 32a shows a schematic representation of a granule of S1 , while Figure 32b shows a schematic representation of a granule of S3. Figures 39a and 39b show respective TEM images of S1.

From these images, it can be seen that S1 has spherical granules with a smooth surface, whereas the granules in S3 are non-spherical with a fractal-like morphology and an irregular surface. Fractal dimensions of both samples was estimated using the TEM pictures shown on Figures 31a and 31 b, post-processed step using software FracLac and ImageJ. S1 has a fractal dimension of 1.76, approaching that of a sphere, and S3 has a fractal dimension of 1.59. Moreover, Figure 31 b shows that the surface of the granules in S3 have predominantly radii of curvature much smaller than the radius of a primary particle, which is not the case for S1 in Figure 31 a. This radius of curvature difference is illustrated schematically in Figures 32a and 32b.

These differences in morphology can be attributed to: differences in primary particle packing (S1 has a higher packing efficiency compared to S3); and differences in the distribution of the carbon coating (the carbon coating is a homogeneous surface distribution for S1 ).

Figure 33 illustrates the respective cycling data of S3 in a half coin cell vs Li/Li+, cycled at 1 C. The graph shown in Figure 33 corresponds to those of Figures 29a and 29b, and specifically to Figure 29a relating to S1 cycling data. These results show that S1 performs better than S3 in terms of specific capacity at low rate (about 800 mAh/g vs. about 700 mAh/g at 0.1 C), capacity retention at high rate (78% capacity retention vs.

72% capacity retention going from 0.1 C to 1 C rate), and cycle life (about 100% capacity retention vs. 28% capacity retention after 50 cycles).

The difference in granule surface geometry between the two samples can be attributed to the difference in electrochemical performance. The smooth surface and spherical morphology of S1 allows for higher granule robustness upon volume expansion upon cycling, thus providing a longer cycle life for S1. The higher capacity and capacity retention at high charging rates for S1 is explained by the better packing of the active material in this sample, which allows for better electrical contacts between the active material across the electrode. Moreover, the higher tapped density allowed by the morphology of S1 is beneficial to the overall cell performance in terms of the energy density obtainable with the electrodes.

Continuity of the primary particle network

Figures 34a and 34b show respective SEM images of S1 and S4. From these images, it can be seen that S4 exhibits a discontinuous network of primary particles. The primary particles are well dispersed in the carbon matrix, with gaps between primary particle aggregates. In contrast, S1 has a continuous network of primary particles throughout the whole granule, as can also be seen in Figure 28b. The differences between the networks of S1 and S4 are illustrated schematically in Figures 35a and 35b, which correspond to networks of particles of S1 and S4 respectively.

Figures 36a and 36b illustrate respective cycling data of S4 in a half coin cell vs Li/Li+. Specifically, the Figures illustrate graphs of specific capacity against number of cycles. Figure 36a relates to cycling data of S4 under a constant rate of charging/discharging (1 C). Figure 36b relates to cycling data of S4 under varying charging rates (0.1 C, 0.25C, 1 C, and 0.5C). Comparing Figures 36a and 36b with Figures 29a and 29b shows that S1 performs better than S4 in terms of specific cycle life (about 100% capacity retention vs. 80% capacity retention after 50 cycles).

The difference in the continuity of the primary particle networks between the two samples can explain the difference in electrochemical performance between the samples. A continuous network of primary particles, as in S1 , allows for a more robust granule structure, which enables better mechanical resistance to volume expansion and therefore a longer cycle life. The surface of carbon matrix-active material interfaces is likely to be higher in S4 than in S1 , resulting in higher intra-granule resistance to electron transfer, especially at high-rate, which is a possible explanation for the higher specific capacities observed with S1. Moreover the higher tapped density allowed by the morphology of S1 is beneficial to the overall cell performance in terms of energy density.

Figure 40 illustrates an SEM image of sample S6, which was heat treated at 700°C. The image demonstrates that heat treatment in harsh, high temperature conditions leads to uncontrolled sintering of the granules, and the formation of a bulk crystalline material with no electrochemical activity. This demonstrates that the control necessary to be exerted over the material architecture, in order to retain the optimal morphology, is not straightforward.

Granule size distribution

Figures 37a and 37b show respectively the granule particle size distributions, measured by laser diffraction, of (a) S1 and (b) S5. S5 is more polydisperse than S1 , and S1 satisfies the both the relationships D10 > D50/2 and D90 < 3 ^* D50, which is not the case of S5.

Figure 38 illustrates the respective cycling data of S5 in a half coin cell vs Li/Li+ cycled at 1 C. A comparison between Figure 38 and Figures 29a and 29b shows that S1 performs better than S5 in terms of specific capacity at low rate (about 800 mAh/g vs. about 600 mAh/g at 0.1 C).

The difference in granule size distribution between the two samples can explain the difference in electrochemical performance between the samples. Polydispersity, and in particular the presence of very large (>75 urn) aggregates, which disturb electrode production upon coating and result in bad adhesion and conductivity at the electrode level, explains the poor performance of S5 in terms of capacity compared with S1. In addition to the further work based on Fe₂0₃, the present inventors have carried out similar work based on ZnFe₂0₄. This work is presented below. ZnFe₂0₄ is considered to be a conversion-alloying material. As with the work based on Fe₂0₃, this work relates to a comparative study demonstrating relationships between granule characteristics and electrochemical properties of the electrode active material.

For the study, two types of sample of an electrode active material (S7 and S8) were created using different methodologies which are described in Table 4.

Table 4

Name Feedstock preparation for spray drying Comments

A homogenised ZnFe₂0₄-

ZnFe₂0₄ 30 nm nanoparticles 4 w.% + sucrose

based granule. Heat 0.5 w.% + graphene oxide 0.5 w.% + ammonium

S7 treated in inert

oleate 0.7 w.% were dispersed in water with a

atmosphere as for iron high shear homogeniser at 25000 rpm for 30 min.

oxide.

A ball-milled ZnFe₂0₄-

ZnFe₂0₄ 30 nm nanoparticles 16 w.% + sucrose

based granule. Heat 3.8 w.% + ammonium oleate 0.2 w.% were

S8 treated in inert

dispersed in water with a planetary ball mill

atmosphere as for iron operated at 400 rpm for 30 min.

oxide.

Figure 41 illustrates SEM images of S7, and Figure 42 illustrates SEM images of S8. These images show that the particles have different morphologies in terms of size distribution, porosity, and characteristics of the primary particle network. Figures 43 and 44 illustrate respective cycling data of S7 and S8 in half cells vs Li/Li+. Both samples were cycled at 0.1 C for 2 cycles, and at 0.5C thereafter. As with the work relating to Fe₂03, the inventors consider that the morphological differences between S7 and S8 result in the different electrochemical performances shown in Figures 43 and 44. Sample S7 shows about 100% capacity retention after 50 cycles, whereas sample S8 shows a 42% capacity retention. Samples S7 and S8 thus demonstrate that a granule architecture can be achieved with a ZnFe₂0₄-based active material which demonstrates good electrochemical performance. The results of this work also demonstrate that homogenizing and ball milling samples results in a different electrochemical performance, as is the case for the iron oxide-based granules.

This is of interest as it expands the compositions tested to reflect the list in claim 1 , and also demonstrates a conversion-alloying mechanism as opposed to pure conversion.

Without wishing to be bound by theory, the present inventors consider that the presence of small quantities of graphene oxide aids the creation of the particle morphology presented in this disclosure. For example, a graphene oxide content of at least at least 0.1 w.% or at least 0.2 w.% (based on the total solid materials plus carrier liquid) can be included in the starting materials to aid the creation of a desirable particle morphology.

While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

All references referred to above and/or listed below are hereby incorporated by reference. Non-patent references

Liu, N.; Lu, Z.; Zhao, J.; McDowell, M. T.; Lee, H.-W.; Zhao, W.; Cui, Y. Nat. Nanotechnol. 2014, 9 (3), 187-192.

Liang, X.; Hart, C.; Pang, Q.; Garsuch, A.; Weiss, T.; Nazar, L. F. Nat. Commun. 2015, 6, 5682.

Tuek, J.; Kemp, K. C.; Kim, K. S.; Zbo il, R. ACS Nano, 2014, 8, 7571-7612.

N. Loeffler, D. Bresser, S. Passerini, Secondary Lithium-Ion Battery Anodes: From First Commercial Batteries to Recent Research Activities, Johnson Matthey Technol. Rev. 59 (2015) 34-44. doi:10.1595/205651314X685824.

Ren, S., Zhao, X., Chen, R., Fichtner, M., Carbon-Nanofibers Encapsulated Metal Oxide Nanocomposite and Its Application as Conversion Anode Material for Lithium Ion

Batteries, ECS Transactions, 64 (22) 155-164 (2015)

Cabana, J.; Monconduit, L.; Larcher, D.; Palacin, M. R. Adv. Mater. 2010, 22 (35).

J. B. Goodenough, K. Park, J. Am. Chem. Soc. 135, 1167-1176 (2013). P. Poizot, S. Laruelle, S. Grugeon, L. Dupont, J. Tarascon, Nature. 407, 496-499 (2000).

M. V Reddy, G. V. S. Rao, B. V. R. Chowdari, Chem. Rev. 1 13, 5364-5457 (2013).

R. Vehring, W. R. Foss, D. Lechuga-Ballesteros, J. Aerosol Sci. 38, 728-746 (2007).

R. Vehring, Pharm. Res. 25, 999-1022 (2008).

T. Ogi, A. B. D. Nandiyanto, K. Okuyama, Adv. Powder Technol. 25, 3-17 (2014).

R. Atchudan, S. Perumal, T. N. Jebakumar Immanuel Edison, Y. R. Lee, Mater. Lett. 166, 145-149 (2016).

H. Zhang, X. Sun, X. Huang, L. Zhou, Nanoscale. 7, 3270-3275 (2015).

X. Jia et al., Aerosol-Assisted Heteroassembly of Oxide Nanocrystals and Carbon

Nanotubes into 3D Mesoporous Composites for High-Rate Electrochemical Energy Storage; Small 11 , 3135-3142 (2015).

Fan Zhang, Tengfei Zhang, Xi Yang, Long Zhang, Kai Leng, Yi Huang and Yongsheng Chen, A high-performance supercapacitor-battery hybrid energy storage device based on graphene-enhanced electrode materials with ultrahigh energy density, Energy Environ. Sci., 2013, 6, 1623

Andrea C. Ferrari, Raman spectroscopy of graphene and graphite: Disorder, electron- phonon coupling, doping and nonadiabatic effects, Solid State Communications 143 (2007) 47-57

Zhonghua Zhang, Guicun Li, Hongrui Peng and Kezheng Chen, Hierarchical hollow microspheres assembled from N-doped carbon coated LUTisO·^ nanosheets with enhanced lithium storage properties, J. Mater. Chem. A, 2013, 1 , 15429

Claims

1. An electrode active material for an electrochemical device, the electrode active material comprising granules of nanoparticles, the nanoparticles being at least partially sintered together, the nanoparticles having an average diameter in the range 2-1000 nm, the granules having an average diameter in the range 1-25 pm and larger than the diameter of the nanoparticles, the granules being porous and comprising a coating of carbon formed at least at the surfaces of pores in the granules, wherein the BET surface area of the granules is 1 -50 m²/g, wherein the nanoparticles comprise one or more of Fe_xO_y, Fe₂C>₃, Fe30₄, FeO, Mn_xO_y, MnC>2, Mn₂C>₃, Mn₃0₄, MnO, Zn_xFe_yO_z, ZnFe2<D₄, Zn_xCo_yOz, ZnCo20₄, Zn_xMn_yO_z, ZnMn20₄, Co_xSn_yO_z, Co2Sn0₄, Mn_xSn_yO_z, Mn2Sn0₄, Ni_xSn_yO_z, Ni₂Sn0₄, Co_xSn_yO_z, CoSnC>3, Ni_xSn_yO_z, NiSnCh, Co_xSi_yO_z, Co₂Si0₄, Fe_xSi_yO_z, Fe2Si0₄, Mn_xSiyO_z, Mn2Si0₄, Sn_xO_y, Sn02, Ge_xO_y, Ge02, Sb_xO_y, Sb20s, Si_xO_y, S1O2, SiO_x, Co_xO_y, CoO, Co₃0 , Cu_xO_y, CuO, Cu₂0, Ni_xO_y, NiO, Zn_xO_y, ZnO, Cr_xO_y, Cr₂0₃, Mo_xO_y, M0O3, M0O2, Ru_xO_y, RUO2, Sn, Si, Fe_xSi, Cu_xSi, Li_xTi_yO_z Li₂Ti03, Nb_xO_y, Nb20s, Nb_xM_yO_z (M = Fe, Ti, V, Mo, Zr, Mg, W, Ga, K), LiNiCoMn0₂, LiNiCoAI0₂, LiMn₂0₄, LiNi_0.5Mni_.5O₄, UC0O2, LiFeP0₄, and their combinations.

2. An electrode active material according to claim 1 wherein the coating of carbon comprises graphitic carbon.

3. An electrode active material according to claim 1 or claim 2 wherein the carbon coating is formed in an amount of 0.1-15 wt% based on the weight of the granules, measurable by TGA in air.

4. An electrode active material according to any one of claims 1 to 3 wherein the carbon coating conforms to the pore surfaces.

5. An electrode active material according to any one of claims 1 to 4 wherein the thickness of the carbon coating is in the range 0.3-100 nm.

6. An electrode active material according to any one of claims 1 to 5 wherein the nanoparticles have an average diameter in the range 5-200 nm.

7. An electrode active material according to any one of claims 1 to 6 wherein the granules have an average particle diameter in the range 5-20 pm.

8. An electrode active material according to any one of claims 1 to 7 wherein the BET surface area of the granules is in the range 1-30 m²/g.

9. An electrode active material according to any one of claims 1 to 7 wherein the pores in the granules provide continuous void channels extending continuously across the whole granule.

10. An electrode active material according to any one of claims 1 to 9 wherein the pores in the granules have a maximum pore size of at least 1 pm or at least 30 times the average nanoparticle diameter.

1 1. An electrode active material according to any one of claims 1 to 10 wherein the surface of the granules have an average fractal dimension of at least 1 .7 and/or the surface of the granules has a local radius of curvature that is greater than the nanoparticle radius at all points on the surface of the granule.

12. An electrode active material according to any one of claims 1 to 1 1 wherein the nanoparticles form a continuous sintered network throughout each granule.

13. An electrode active material according to any one of claims 1 to 12 wherein the granule size distribution satisfies the relationships: d 10 > d50/2 and d90 < 3 times d50.

14. An electrode active material according to any one of claims 1 to 13 wherein the nanoparticles comprise a metal oxide.

15. A method for the manufacture of an electrode active material for an

electrochemical device, the method comprising:

providing a dispersion of nanoparticles in a carrier liquid, the nanoparticles having an average diameter in the range 2-1000 nm;

including in the dispersion a carbon precursor material that is soluble and/or dispersible in the carrier liquid and optionally a dispersant to aid in the effective dispersion of nanoparticles;

spray drying the dispersion to form precursor granules of nanoparticles with the carbon precursor material distributed in the precursor granules;

subjecting the precursor granules to a heat treatment to degrade the carbon precursor material, the heat treatment thereby providing granules of nanoparticles,

wherein the nanoparticles are at least partially sintered together, the granules having an average diameter in the range 1-25 pm, the granules being porous at least in part due to the degradation of the carbon precursor material, the granules comprising a coating of carbon formed at least at the surfaces of pores in the granules from the degradation of the carbon precursor material,

wherein the nanoparticles comprise one or more of

Fe_xO_y, Fe2C>3, Fe30₄, FeO, Mn_xO_y, MnC>2, M^h2q3, M^h3q₄, MnO, Zn_xFe_yO_z, ZnFe2<D₄, Zn_xCo_yO_z, ZnCo₂0₄, Zn_xMn_yO_z, ZnMn₂0₄, Co_xSn_yO_z, Co₂Sn0₄, Mn_xSn_yO_z, Mn₂Sn0₄, NixSriyOz, NhSnCU, Co_xSn_yO_z, CoSnCb, Ni_xSn_yO_z, NiSnOs, Co_xSi_yO_z, C02S1O4, Fe_xSi_yO_z, Fe₂Si0₄, Mn_xSiyO_z, Mn₂Si0₄, Sn_xO_y, Sn0₂, Ge_xO_y, Ge0₂, Sb_xO_y, Sb₂0s, Si_xO_y, S1O2, SiO_x, Co_xO_y, CoO, Co₃0₄, Cu_xO_y, CuO, CU2O, Ni_xO_y, NiO, Zn_xO_y, ZnO, Cr_xO_y, Cr₂0₃, Mo_xO_y, M0O3, M0O2, Ru_xOy, RU0₂, Sn, Si, Fe_xSi, Cu_xSi, Li_xTi_yO_z Li₂Ti0₃, Nb_xO_y, Nb₂0₅, Nb_xM_yO_z (M = Fe, Ti, V, Mo, Zr, Mg, W, Ga, K), LiNiCoMn0₂, LiNiCoAI0₂, LiMn₂0₄, LiNio.5Mn1.5O4, L1C0O2, LiFeP0₄ and their combinations.

16. A method according to claim 15 wherein a further carbon coating is applied so as to reduce the BET surface area of the granules.

17. A method according to claim 16 wherein the further carbon coating includes milling carbon in the presence of the granules.

18. A method according to any one of claims 15 to 17 wherein the BET surface area of the precursor granules is at most 15 m²/g.

19. A method according to any one of claims 15 to 18 wherein the BET surface area of the granules is 1-30 m²/g.

20. A method according to any one of claims 15 to 19 wherein the heat treatment comprises heating to a temperature in the range 400-1350°C in an inert or reducing atmosphere.

21. An electrochemical device comprising an anode, a cathode and an electrolyte disposed between the anode and the cathode, wherein the anode comprises an electrode active material according to any one of claims 1 to 14.

22. A use of an electrode active material according to any one of claims 1 to 14 as an anode active material, or a component of an anode active material, in an anode in conjunction with a cathode and an electrolyte in a lithium ion battery for charging and discharging of the lithium ion battery.

23. A method for processing an electrode active material according to any one of claims 1 to 14 as or in an anode active material for a lithium ion battery, the method including diffusing lithium ions into the anode active material.