Wood-Based Nanocellulose In Lithium Ion Batteries and ...1393422/FULLTEXT02.pdf · Development of electrode-integrated cellulose separators was enabled by a non-aqueous drying method

Wood-Based

Nanocellulose In

Lithium Ion

Batteries and

Electrochemical

Coatings

Hyeyun Kim

Doctoral Thesis,2020

KTH Royal Institute of Technology

School of Engineering Sciences in Chemistry, Biotechnology and Health

Department of Chemical Engineering

Applied Electrochemistry

SE- 100 44 Stockholm, Sweden

© Hyeyun Kim 2020

TRITA-CBH-FOU-2020:12

ISBN: 978-91-7873-453-5

Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan

i Stockholm framlägges till offentlig granskning för avläggande av

teknologie doktorsexamen torsdag den 12 Mars 2020 kl. 10:00 i sal F3,

KTH, Lindstedtsvägen 26, Stockholm. Fakultetsopponent: Davide

Beneventi från CNRS

i

Abstract

Lithium ion batteries contain diverse functional polymeric materials, e.g.

binders and separators. Naturally self-assembled wood cellulose can be

disintegrated to nanosized particles with a diversity of morphology by top-

down processes, adjusting the manufacturing parameters. The

nanomaterials can then be reconstructed by bottom-up assembly to

structures similar to that of the polymeric materials in lithium ion batteries,

capable of replacing their functions and ensuring similar or improved

performance.

The aim of the thesis is to evaluate the feasibility of wood-based cellulose

nanofibers in lithium ion batteries and explore other possible applications.

The relationship between the characteristics of nanocellulose, treated by

different processes, and their performance as battery components were

investigated using electrochemical and in-operando measurements.

Development of electrode-integrated cellulose separators was enabled by a

non-aqueous drying method. This significantly improved the drying

efficiency and can be considered an eco-friendly process without using

hazardous chemicals. This study sheds the light on cellulose as a promising

separator material, satisfying the industrial needs without trade-off of

durability of the material and ion transport properties.

Other than lithium ion battery applications, cellulose nanofibrils are

introduced as a pH-responsive polymer and a precursor of hydrogel,

electrochemically coated on any conductive substrate. Not only hydrogel,

this electro-precipitation method also enables to fabricate single or multi-

layered composites. The hydrogel and the composites fabricated by this

technique can work as functional materials in the diverse electrochemical

applications.

In summary, the results indicate that using wood-based cellulose as a raw

material is beneficial to fabricate the functional materials by eco-friendly

manufacturing processes, available for a variety of electrochemical

applications, showing excellent performance.

Keywords

cellulose nanofibers, lithium ion battery, separator, electro-precipitation,

hydrogel

ii

Sammanfattning

Litiumjon-batterier innehåller komponenter, exempelvis bindemedel och

separatorer, som består av polymera material. Naturligt bildat cellulosa

från trä kan finfördelas till nanometerstora partiklar vars morfologi beror

på olika tillverkningsparameterar i sönderdelningsprocessen. Dessa

partiklar kan sedan användas för att återskapa nanomaterial med

strukturer liknande de i de polymera materialen i litiumjon-batterierna,

och som också kan ersätta de senare med likvärdiga eller bättre prestanda.

Det huvudsakliga målet med detta arbete är att utvärdera möjligheten att

använda nanofibrer av cellulosa från trä i litiumjon-batterier. Cellulosa

behandlat på olika sätt har undersökts som material i batterikomponenter,

genom elektrokemiska mätningar och in operando tekniker. Separatorer

av cellulosa integrerade med elektroder visade sig kunna tillverkas i ett

vattenfritt system, vilket avsevärt underlättar dess torkning, i en

miljövänlig process utan skadliga kemikalier. Denna studie lyfter fram

cellulosa som lovande material för separatorer, som motsvarar industriella

behov utan att försämra materialens livslängd och jonledande egenskaper.

Utöver applikationen litiumjon-batteri har nanofibrer av cellulosa

undersökts som pH-känslig polymer och utgångsmaterial för hydrogeler,

vilka har belagts på ledande ytor. Med denna metod kan man också

tillverka kompositmaterial innehållande nanopartiklar i enskilda eller i

flera skikt. Hydrogeler och kompositer syntetiserade genom denna teknik

kan fungera som funktionella material i olika elektrokemiska

tillämpningar.

Sammanfattningsvis, resultaten indikerar att användning av träbaserad

cellulosa som råmaterial är fördelaktigt vid tillverkning av funktionella

material genom miljövänliga processer, i olika elektrokemiska

användningar där de visar utmärkta prestanda.

Nyckelord

nanofibrer av cellulosa, litiumjonbatteri, separator, elektrokemisk

utfällning, hydrogel

iii

List of appended papers Paper I

Lithium ion battery separators based on carboxylated cellulose nanofibers

from wood.

Hyeyun Kim, Valentina Guccini, Huiran Lu, Germán Salazar-Álvarez,

Göran Lindbergh, Ann Cornell

ACS Applied Energy Materials 2018, 2, 1241–1250.

Paper II

Feasibility of chemically modified cellulose nanofiber membrane as

lithium ion battery separator.

Hyeyun Kim, Ulriika Mattinen, Valentina Guccini, Germán Salazar-

Álvarez, Rakel Wreland Lindström, Göran Lindbergh, Ann Cornell

Manuscript

Paper III

Spray-coated nanocellulose based separator/electrode assembly.

Hyeyun Kim, Carl Moser, Ulriika Mattinen, Gunnar Henriksson, Rakel

Wreland Lindström, Göran Lindbergh, Ann Cornell

Manuscript

Paper IV

Different manufacturing processes on tempo-oxidized carboxylated

cellulose nanofiber performance as binder for flexible lithium-ion batteries.

Huiran Lu, Valentina Guccini, Hyeyun Kim, German Salazar-Alvarez,

Goran Lindbergh, and Ann Cornell

ACS Applied Materials & Interfaces 2017, 9, 37712–37720.

Paper V

One-step electro-precipitation of nanocellulose hydrogels on conducting

substrates and its possible applications: coatings, composites, and energy

devices.

Hyeyun Kim, Balázs Endrődi, Germán Salazar-Alvarez, Ann Cornell.

ACS Sustainable Chemistry & Engineering 2019, 7, 24, 19415-19425.

iv

The contributions of the author to these papers are:

Paper I Performed the major experimental works. Oxidation of

the pulp was performed with Valentina Guccini.

Paper II Performed the major experimental works. Valentina

Guccini helped the oxidation of the pulp and the dialysis of the fibers. Gas

evolution measurements were conducted by Ulriika Mattinen.

Paper III Performed the major experimental works. Nanocellulose

sample was prepared by Carl Moser, and Ulriika Mattinen performed the

gas evolution analysis.

Paper IV Participated in the pulp oxidation and performed the

rheology measurements of TOCN gels.

Paper V Performed the major electrochemical works and physical

characterization. Hydrogel coating for HER selective layer was performed

with Balázs Endrődi.

v

List of abbreviations

CMC Carboxyl methyl cellulose

CNF Cellulose nanofibers

CV Cyclic voltammetry

DF Degree of fibrillation

ECSA Electrochemically active surface area

EDX Energy dispersive X-ray spectroscopy

FT-IR Fourier transform infrared

HER Hydrogen evolution reaction

IPA Isopropanol

LCO Lithium cobalt oxide (LiCoO2)

LFP Lithium iron phosphate (LiFePO4)

LIB Lithium ion batteries

LiPF6 Lithium hexafluorophosphate

LMO Lithium manganese oxide (LiMn2O4)

LTO Lithium titanium oxide (Li4Ti5O12)

NMC Lithium nickel manganese cobalt oxide (LiNi1/3Mn1/3Co1/3O2)

NMC622 LiNi0.6Mn0.2Co0.2O2

NMP N-Methyl-2-pyrrolidone

PVDF Polyvinylidene fluoride

PVDF-HFP Poly(vinylidene fluoride-co-hexafluoropropylene)

RDE Rotating disc electrode

SEA Separator-electrode assembly

SEI Solid electrolyte interface

SEM Scanning electron microscopy

TOCN TEMPO-mediated oxidized cellulose nanofibers

VC Vinylene carbonate

vi

Contents

1. Introduction ................................................................... 1

1.1. Scope of the thesis .................................................................. 2

2. Background ................................................................... 3

2.1. Lithium ion batteries ............................................................... 3

2.2. Lithium ion battery separators ............................................... 4

2.3. Wood based nanocellulose .................................................... 5

2.4. Nanocellulose in LIB ............................................................... 6

2.5. Separator-electrode assembly (SEA) .................................... 7

2.6. Gas evolution in LIB ................................................................ 9

2.7. Electrochemical coating of biopolymers and its

applications ........................................................................................ 10

3. Experimental ............................................................... 12

3.1. Preparation of cellulose nanofibers (CNF) .......................... 12

3.1.1. TEMPO-mediated oxidation of cellulose ..................... 12

3.1.2. Enzymatically treated CNF ............................................ 12

3.2. CNF-based LIB separators .................................................... 13

3.2.1. Fabrication of freestanding separators ....................... 13

3.2.2. Spray-coated cellulose-based SEA .............................. 13

3.2.3. Electrochemical characterization of LIB separators .. 14

3.2.4. Gas evolution analysis by mass spectrometry .......... 15

3.3. Rheology measurements of cellulose nanofibers.............. 16

3.4. Electro-precipitation of nanocellulose hydrogels and its

applications ........................................................................................ 16

3.4.1. Electro-precipitation of TOCN hydrogels .................... 16

3.4.2. Electro-precipitation of composites ............................ 16

3.4.3. Application I: Eco-friendly fabrication of electrode for

LIB 17

3.4.4. Application II: Selective layer for hydrogen production

17

3.5. Physical characterization of samples .................................. 18

4. Results and Discussions............................................ 19

4.1. CNF based separators with different charge density ........ 19

vii

4.1.1. Characterization of CNF and separators thereof ....... 19

4.1.2. Electrochemical performance of TOCN separators ... 22

4.1.3. CNF separators with different charge density ........... 25

4.2. Spray coated cellulose-based separator-electrode

assembly (SEA) .................................................................................. 27

4.2.1. Fabrication of SEA ........................................................ 27

4.2.2. Electrochemical performance of CNF-SEA ................ 31

4.2.3. Electrochemical performance of SiO2-CNF-SEA ........ 34

4.3. Gas evolution of LIBs with CNF separators ........................ 36

4.4. Discussion about CNF-based LIB separators ..................... 42

4.5. Rheological properties of TOCN ........................................... 44

4.6. Electro-precipitation of CNF hydrogel, composites thereof,

and its applications ............................................................................ 46

4.6.1. Electrochemical precipitation of TOCN Hydrogels .... 46

4.6.2. Electro-precipitation of composites ............................ 49

4.6.3. Application I: Fabrication of electrode for LIB ........... 51

4.6.4. Application II: Selective membrane for HER .............. 52

5. Conclusions and outlook ........................................... 55

5.1. CNF based separators with different charge density ......... 55


assembly ............................................................................................. 55

5.3. Electro-precipitation of TOCN hydrogel, composites

thereof, and their applications .......................................................... 56

5.4. Outlook .................................................................................... 56

6. Acknowledgements .................................................... 58

7. References .................................................................. 59

Introduction - 1

1. Introduction

Increase in energy consumption inevitably brings about global

environmental pollutions and climate change. Today, dramatic conversion

of energy sources from traditional fossil fuels to renewable energy has been

gradually achieved overcoming economical and technical barriers. Lithium

ion batteries (LIB) appeared in the 1990’s in market and have been rising

as the major power sources for small portable electronics to large scale in

transportation, electric vehicles and electric grid. They have been

highlighted as appealing energy storage systems, which can fulfill the

demands not only for low carbon emission but also for high energy density

and stable performance capable of long-term operation. The global LIB

market is forecasted to grow from 120 GWh of sales volume in 2017 to 215

GWh by 20251,2. However, the rapid growth of the automotive industry and

energy markets will entail the production of a giant amount of waste.

Namely, despite of endeavors to reduce the reliance on fossil fuels moving

toward energy production and storage systems with lower environmental

impacts, ironically, the alternative technologies are still dependent on high

consumption of petroleum-based materials and not environmentally

friendly processes. For instance, upstream material acquisition processes

including mining and extraction of inorganic compounds can result in

significant environmental pollution. Furthermore, the-state-of-the-art

manufacturing process of LIB uses plastics and organic components in

preparation of electrode slurry, binders, separators, electrolytes, and

packaging3. Therefore, urgent actions should be taken to ensure

sustainable consumption and production patterns, which is one of 17 goals

of United Nations to tackle climate change4.

A push to develop such sustainable manufacturing processes implies the

demands for applications of biodegradable materials emitting less carbon

dioxide and pollutants during the waste treatment processes5–8. However,

an important challenge of application of bio-based materials for large

scaled commercial batteries is their high risk of introducing trace water

into LIB, which influences on durability and safety. Nevertheless, it is

believed that a better understanding of the reactions of the materials in LIB

can make them more wide-spread in the market, applied for commercial

cells.

Apart from LIB, such biopolymers have obtained significant attention as

alternative to plastics, due to their similarities in terms of optical,

Introduction - 2

dimensional, chemical and mechanical characteristics. Their hygroscopic

properties make them more attractive in versatile applications, e.g.

biomedical engineering, drug delivery, tissue engineering and

adsorbents9,10. Not only replacing materials, but also developing facile and

efficient fabrication techniques to obtain materials with a desired property

by lower energy input is important. We believe that such research efforts

will be able to widen the spectrum of applications of biopolymers in a

variety of fields as functional materials.

1.1. Scope of the thesis

The main goal of this thesis was to examine the feasibility of wood-based

nanocellulose in LIB. Nanocellulose has been investigated as a promising

candidate replacing inactive components in LIB such as binders or

separators and reported good performance, comparable to conventional

materials. By tuning the manufacturing processes, by top-down treatments,

pulp can be disintegrated to nanosized materials with desirable

dimensions depending on its target applications. This study focuses on the

comparison of the characteristics of nanocellulose, produced by different

treatments, and their performance as functional materials in

electrochemical applications. Depending on the dimensional properties of

nanocellulose, the applications varied from separators to binders and their

electrochemical performance was evaluated in LIB. Microfibrillated

cellulose with nanofibril networks was highlighted as a promising

candidate for a lithium ion battery separator, particularly, for the fast

charging applications. The thesis includes failure and success cases of

nanocellulose separators. The cause of cell failure was elucidated by

electrochemical and in-operando gas measurements. Development of

facile and eco-friendly fabrication method for cellulose based separator-

electrode assembly (SEA) is introduced as well.

Other than LIB applications, individually disintegrated cellulose

nanofibrils were fabricated to hydrogel or composites by electrochemical

precipitation and further applications in electrochemical energy devices,

e.g. hydrogen production were investigated.

Background - 3

2. Background

2.1. Lithium ion batteries

The primary components of LIB are positive and negative electrodes,

separator, current collectors and electrolyte. A principle of LIB is that

transport of lithium ions between two electrodes generates free electrons

which can flow through an external circuit storing and providing electric

power during the charging and the discharging processes.

Three different types of positive electrodes mainly used are (1) layered

lithium transition metal layer oxides, e.g. lithium cobalt oxide (LiCoO2,

LCO), lithium nickel manganese cobalt oxide (LiNixMnyCo1-x-yO2, NMC), (2)

polyanions, e.g. lithium iron phosphate (LiFePO4, LFP), (3) spinel

materials, e.g. lithium manganese oxide (LiMn2O4, LMO)11. NMC as a

positive active material has taken the highest market share (29% in 2015)

followed by LCO (26%) and LFP (23%)2. Due to high nominal potential (≈

3.7 V vs Li+/Li), high capacity and energy density with good cycling stability,

it is expected that the global supply and demands of NMC will grow more

in the future. Graphite and carbonaceous materials have been used as

major negative electrodes in LIB, taking more than 90% of the market

share in 20152. Silicon based compounds and lithium titanium oxide

(Li4Ti5O12, LTO) are also used as negative active materials. Aluminum and

copper foils are used as the current collectors for the positive and the

negative electrodes, respectively.

An electrolyte is an ion conductor, facilitating transport of lithium ions

between the electrodes. The commonly used liquid electrolytes are

composed of lithium containing salts e.g. lithium hexafluorophosphate

(LiPF6), lithium bis(trifluoromethylsulfonyl)imide, or lithium perchlorate,

dissolved in carbonate solvents, such as ethyl carbonate or diethyl

carbonate. As binder, polymeric materials such as polyvinylidene fluoride

(PVDF), carboxyl methyl cellulose (CMC) and styrene butadiene have been

widely used12,13.

A separator is one of the major components in LIB and its primary role is

to prevent the direct contact between the positive and the negative

electrodes. The detailed process of manufacturing and characterization of

lithium ion battery separators and recent progress of developments will be

more explained in the following section.

Background - 4

2.2. Lithium ion battery separators

Lithium ion battery separator is a key safety system in LIB, preventing

short circuit and further incidents. It needs to be a thin porous membrane,

which can transport lithium ions between the positive and the negative

electrodes, while blocking their direct contact. To achieve low resistivity

and high volumetric/gravimetric capacity of LIB, commercial separators

have 7-25 µm of thickness preferably and ≈ 40% of porosity. The

development of thinner separators improves the performance of LIB but

also takes high risk of compromising the durability of and the safety of a

battery cell. For example, an exceptionally thin separator was blamed as

the contributory factor of fire or explosion incidents of electronic devices14.

The commercially available separators are predominantly made of

polyolefin materials, e.g. polyethylene (PE, melting point, tm, ≈ 135 °C),

polypropylene (PP, tm ≈ 160 °C), or PVDF (tm ≈ 177 °C) and polyethylene

terephthalate (PET, tm ≈ 260 °C) 3,15–18. The production of the porous

membranes is achieved by stretching the polymer sheets in wet and dry

status12,16. Since the separator is wound with the electrodes under high

tension during the cell assembly process, materials with high tensile

strength and Young’s modulus are required12. The polyolefin materials

with low melting temperature are advantageous for thermal shutdown

function, but susceptible to thermal runaway. Recent research efforts have

highlighted the importance of surface coatings of ceramic particles on the

commercial PE separators, which enhance their thermal stability and

dimensional integrity even at an elevated temperature, above 200 °C2,17,19.

As the demand for LIB rapidly increases, the total market for separators is

estimated to grow from 900 Mm2 in 2015 up to 2700 Mm2 in 202512. The

cost of the separators is expected to decrease considerably from 23% of a

cell cost in 200220 to ≈12% in 2030 (recalculated including the costs only

for materials) 1,21. It would be beneficial if separators could be produced

with lower costs, preferably < 2 $/m2 16,22,23. In addition, considering the

quantity of separators (that is, the high consumption of plastics) and the

manufacturing processes producing a large amount of waste solvents,

more efforts should be made to produce separators with environmentally

friendly materials and sustainable manufacturing processes2,12,24–26

Background - 5

2.3. Wood based nanocellulose

Cellulose is the most abundant biopolymer, primarily obtainable from

plant resources, e.g wood, which takes 30-40% of the cell wall components.

In the plant cell wall, polysaccharides are enzymatically synthesized to

cellulose microfibrils with ≈ 4 nm in width and assembled into micro-sized

fibers with highly oriented and hierarchical structure27,28. The cellulose

fibers are the main constituent of wood, making it strong and supporting

its structure.

The naturally assembled hierarchical structure of cellulose fibers is

advantageous to produce nanofibers, by delaminating the microfibrils.

However, strong interfibrillar bonding and the presence of crystalline

regions require high mechanical energy input for defibrillation. In order to

overcome such recalcitrance of biomass and obtain cellulose nanofibers

(CNF) with high efficiency, a variety of processes have been investigated

e.g. mechanical treatment 29, enzymatic treatment 30–32, chemical

modification such as (2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)-

mediated oxidation 33 and carboxymethylation 34.

Enzymatic hydrolysis is a common biological treatment to soften pulp,

which degrades cellulose chains using cellulases. Endoglucanase is one of

the widely used enzymes for pulp treatment, which can access less

crystalline regions in the cellulose chains and specifically hydrolyze

glycosidic bonds 30. Consequently, such enzymatic hydrolysis contributes

to loosening strong interfibrillar fiber networks and facilitates the

production of cellulose nanofibers by successive mechanical disintegration 35,36. It has been reported that enzymatically treated cellulose is mostly

micro-fibrillated cellulose, composed of partially delaminated nanofibril

networks.

TEMPO-mediated oxidation is one of the most commonly used chemical

modification methods to produce finely dispersed cellulose nanofibers.

During the oxidation process, hydroxyl groups at C6 positions of the

glucose units of cellulose chains are substituted to carboxylated groups

with sodium counterions. Introduction of the negatively charged groups

increases the electrostatic repulsion between fibrils and facilitates

consecutive mechanical disintegration of nanofibers. Combined with such

chemical modification and successive mechanical disintegration, this

method even enables production of individual TEMPO-oxidized cellulose

nanofibers (TOCN) with 3- 4 nm in width and a few microns in length. Such

Background - 6

nanofibrils can be dispersed in water with high colloidal stability. Due to

its high dispersibility and novel physicochemical properties, the TOCN has

been investigated in a variety of applications, e.g. energy devices, photonics,

biomedical engineering, drug delivery, absorbents6,9,37,38. It is very unique

and advantageous that such highly dispersed nanofibrils with low

dimension in width can be obtained from plant cellulose, which is not

available from other resources, such as bacterial (10-20 nm) or algae

cellulose (5-60 nm) 27,28,39.

Dimensions of cellulose nanofibers and their characteristics can be varied,

depending on the disintegration strategies and the parameters of the

production processes, e.g. the charge density, the degree of fibrillation, the

dose of enzymes. That is, nanocellulose with a desired dimension for the

target application can be acquired by adjusting the manufacturing process 40–42. The following sections will introduce the applications of wood-based

nanocellulose with different dimensions to electrochemical energy devices,

e.g. LIB and H2 production, and an electrochemical coating technique.

2.4. Nanocellulose in LIB

Cellulose has been highlighted as a promising candidate for future eco-

friendly material, which can replace some portion of petroleum-based

plastic materials. Even in LIB research, cellulose and its derivatives have

been investigated as binder, separator or precursor of carbon

materials36,38,40,43–49.

The composites containing cellulose nanofibers and active materials can

be fabricated to free-standing paper based electrodes by a paper-making

filtration process 38,44,48. Thanks to its good dispersibility in water, LIB

electrodes can be fabricated by aqueous based processes.

Not only as binders, cellulose has also been used as a promising separator

material due to its highly porous structure allowing transport of ions,

hydrophilicity absorbing liquid electrolyte efficiently, high thermal

stability, abundance and low material cost. In this respect, cellulose-based

separators have been already widely used in primary batteries for many

years. In LIB, Kuribayashi45 first demonstrated a cellulose separator with

submicron sized pores, capable of working in a LiCoO2 /carbon full cell in

the potential range between 2.7 to 4.2 V vs Li+/Li. The cellulose based

separators can be easily fabricated by a paper-making filtration

process36,40,47. Zolin et al.40 demonstrated the effect of the degree of beating

Background - 7

on the morphology and the electrochemical performance of the cellulose

separators in a bendable paper based flexible battery assembly. Chemically

modified cellulose or its derivatives have been investigated as separators

or gel-polymer electrolytes48,49. Leijonmarck et al. reported a single-paper

flexible LIB cell using TOCN both as binder and separator materials38.

Not only plant based cellulose, but also ones from other natural resources

have been examined as LIB separators, e.g. bacterial or algae cellulose 47,50,51. They have great advantages of naturally assembled mesoporous

structures and high crystallinity, which can result in high mechanical

strength and low risk of trace water in the cellulose chains47,50. However,

supply chain and extra energy input required to eliminate organism prior

to mechanical dispersion 51 would be considered as the key factor of the

commercialization of such materials.

2.5. Separator-electrode assembly (SEA)

A separator-electrode assembly is an integrated form of those two

compartments, which can be found in electrochemical devices, e.g.

microbial fuel cells52 and lithium ion batteries53–63, as shown in Figure 2.1.

Figure 2. 1. Schematic illustration of (A) a typical battery cell assembly

and (B) a SEA with a separator integrated with an electrode

In the state-of-the-art battery manufacturing process, the separately

produced separator and electrodes are assembled and wound to cylindrical

or prismatic cells. To achieve high production yield, a separator must have

sufficiently high tensile strength and Young’s modulus not to be broken

Background - 8

and deformed during the winding process. Such requirement limits the

commercialization of highly porous functional materials with relatively low

mechanical strength.

On the contrary, SEA is beneficial as it improves the mechanical strength

of the separators, reinforced by the metallic current collector, in the in-

plane direction. This means that, as long as the separator is attached to the

surface of the electrode, the tensile strength of the separator relies on that

of the current collector, rather than its own lesser strength. For practical

purpose, Toshiba, a Japanese lithium ion battery manufacturer, recently

announced the production plan of their new skin-coated electrode (ScdE)63.

The company highlights the advantages of the thin highly porous separator

made by electrospinning; low resistivity and highly improved capacity (≈

20%) of LIB.

SEA is an appealing approach, capable of widening the spectrum of

commercially available materials. Most of SEA studies have been

predominantly using inorganic nanoparticles e.g. Al2O3, SiO2, ZrO2,

Li7La3Zr2O12, with polymer binders such as PVDF, poly(vinylidene

fluoride-co-hexafluoropropylene) (PVDF-HFP), PVA (polyvinyl alcohol)

and poly(phenyleneoxide), as the primary materials 53–63. The inorganic

nanoparticles are advantageous to construct a separator with high surface

porosity, whereas there is a high risk of non-uniform deposition, which can

in the end affect the durability of the materials and the electrochemical

performance as well. In terms of polymer binders, PVDF requires to be

dissolved in hazardous solvents such as N-Methyl-2-pyrrolidone (NMP) or

N, N-dimethylformamide (DMF).

Another challenge with SEA is the cell configuration and the selection of

an electrode, on which to coat the separator. In a typical battery cell

configuration, the size of the negative electrode is similar or slightly larger

than that of the positive electrode64,65. In previous studies, coating the

separator part on the negative electrode (graphite or LTO) has been

commonly implemented54–56,60–62. However, volume expansion of graphite,

≈10%, can influence on the durability of the separator in SEA. Replacing

graphite with LTO, which has much lower volume change during battery

cycling, can improve its stability, 56,57,61,62 but it compromises the potential

and energy densities of the battery. Rather, coating the separator on the

positive electrode would be preferred, but only a few such cases have been

presented58. Considering the high market share of graphite and NMC,

SEAs suitable for a full cell consisting of a pair of those electrodes would

Background - 9

be the most favorable form, more likely where the separator is coated on

the positive electrode side.

2.6. Gas evolution in LIB

In-situ monitoring of LIB has provided better understanding about

fundamental mechanisms related to aging and promising next generation

energy storage materials26,66–69. In-operando mass spectrometry has been

highlighted as a powerful technique in LIB studies, in particular to measure

gases evolved and consumed during the LIB operation67,70–74. Gas

evolution in LIB during storage and in use is the major cause of cell failure

and further accidents75,76. Gas accumulated in the cell can result in poor

contact between LIB materials, and their delamination. It can further

increase the internal pressure and swelling of the cell and, in the end, likely

lead to explosion, risking serious damage of life and property.

The mechanism of gas formation in LIB has been explained related to the

effects of trace water67,70,71, electrolyte additives72,73, and the decomposition

of battery materials74,77, operating parameters, and prolonged cycle life of

the batteries71,78 For example, Metzger et al.70 elucidated the origin of H2

gas in LIB and effect of vinylene carbonate (VC) additive on suppressing

gas evolution using on-line electrochemical mass spectrometry.

Gases mainly detected in LIB are H2, C2H4, CO2, and CO. In the presence

of trace water, a significantly high amount of H2 gas can be measured by

mass spectrometry in the beginning of the cycling, along with the reduction

of trace water and the solid electrolyte interface (SEI) layer formation71.

Therefore, the H2 gas evolution can be an indicator to evaluate the

electrochemical stability of a hygroscopic material or LIB materials

produced by aqueous processes and explain the cause of cell failure or

other undesirable side reactions in LIB. In addition to reduction of trace

water, protic carbonate decomposed products at the cathode can diffuse to

the graphite and be reduced generating H2 gas during the cycling70. The

evolution of C2H4 is accompanied by the decomposition of electrolyte

solvent, thus the high amount is generated in the first charging due to SEI

layer formation and its evolution slows down during the cycling70,71. The

evolution reactions of CO2 and CO are dependent on the temperature and

the upper cut-off voltage79.

To meet an increasing interest in sustainability issues in LIB and reducing

the environmental impact in manufacturing processes, a variety of bio-

Background - 10

based materials have been applied and shown good electrochemical

performance7,26,47,50. Such bio-based materials are likely to contain surface

functional groups which can give materials better dispersible properties in

a solvent but also hygroscopic properties, unfavorable in LIB. Therefore,

prior to using bio-based materials, more comprehensive and careful

investigations about their chemistry and reaction mechanisms in LIB are

required. In this respect, an in-operando gas analysis system may provide

crucial information about the chemical or electrochemical stability a

material, especially, when it has a risk of introducing trace water into LIB.

2.7. Electrochemical coating of biopolymers and its

applications

Electrodeposition or electro-precipitation techniques have received

attention as a method to construct biopolymer-based hydrogels on desired

substrates in response to electrical signals. Chitosan was the first

biopolymer used for electrodeposition and its applications have been

extended to biosensors, enzyme immobilization, and drug delivery80, 81.

Other than chitosan, alginate82,83, gelatin84, CMC85, and silk fibroin86 have

been reported as stimuli responsive biomaterials capable of electro-

precipitation.

The most widely used mechanisms of electro-precipitation of such

biopolymer hydrogels are the reactions induced by i) pH dependent sol-gel

transition82,86–88 and ii) metal-ions coordination80,81,83,89.

For the pH dependent sol-gel transition, electrochemical water splitting

can be used to generate pH gradients at an electrode surface. In a water

splitting electrochemical cell, the O2 evolution reaction (OER) takes place

at the anode (2H2O ⟶ O2(g) + 4H+ + 4e-), where protons generated

decrease the local pH near the electrode. On the other hand, H2 gas and

hydroxide ions are produced (H2 evolution reaction, HER), leading to a

local pH increase at the cathode (2H2O + 2e- ⟶ H2 (g) + 2OH-). Alginate,

containing sodium carboxylate groups, can aggregate in acidic conditions

as sodium ions are substituted to protons and the surface charge is

neutralized (COO–Na+ + H+ → COOH + Na+). Taking another example,

chitosan chains are neutralized at alkaline conditions, thus chitosan can be

coated on the cathode surface in response to a local pH changes induced

by water splitting (chitosan – NH3+ + OH- → chitosan – NH2 + H2O) 87. Silk

Background - 11

fibroin is known to be electro-precipitated at a pH below 4.486. Wang et al. 88 reported on the fabrication of alginate/chitosan layer-by-layer

composites by alternately changing the direction of the electric field and

the coating medium. Since the electrochemical reaction-induced

precipitations of biopolymers occur in the diffusion layer of an electrode,

the growth of the hydrogel can be simply controlled by varying the

experimental parameters, such as the applied current or voltage.

Metal-ions coordinated gelation takes place at the anode surface, since the

it requires metal ions generated by the anodic dissolution. For instance,

chitosan hydrogel can be coated on the cathode surface in pH-responsive

gelation88, while it can be formed on the anode surface by a metal-ion

coordinated sol-gel transition reaction81.

It is known that most polysaccharides can be electro-precipitated to

hydrogel films on the anode surface, due to charge neutralization by a

localized pH gradient83. Considering the presence of functional groups of

TOCN, similar to alginate, TOCN can be also regarded as a pH-responsive

polymer, showing immediate aggregation at acidic conditions at pH < ≈

pKa 4.8 due to the protonation and charge neutralization of cellulose chains,

in the same way as alginate90. Such stimuli responsive characteristic makes

TOCN a more attractive material to be tuned and fabricated to hydrogel

materials for versatile applications, such as biomedical engineering, drug

delivery, tissue engineering and adsorbents9,10,85,91.

As a further benefit of electro-precipitation, nano-structured composite

materials can be easily fabricated80,82,89,92. Nanoscale particles dispersed in

the coating dispersion with stimuli responsive biopolymers can get

immobilized on the conductive substrates, simultaneously with electro-

precipitation of the biopolymers.

Experimental - 12

3. Experimental

3.1. Preparation of cellulose nanofibers (CNF)

3.1.1. TEMPO-mediated oxidation of cellulose

Never dried cellulose pulp (Domsjo Fabriker AB, Ornskoldsvik, Sweden)

was oxidized, following the earlier reported protocols by Saito et al.93.

Briefly, 40 g of pulp washed by a HCl solution (pH 2) was mixed with a 4

mmol of TEMPO and 40 mmol of NaBr in DI water. Pulps with different

charge densities (350, 650 and 1550 µmol COO-/g cellulose) were produced by

adjusting the amount of NaClO added. A 0.5 M NaOH solution was added

dropwise to keep the pH of the suspension at 10 during the reaction. 1

wt % TEMPO-mediated oxidized cellulose pulps suspended in water were

disintegrated by a micro-fluidizer (M-110EH, Micro fluidics Corp, United

States). The number of passages through the micro-fluidizer, degree of

fibrillation (DF), varied depending on the desired dimension of

nanocellulose. The TEMPO-mediated oxidized cellulose nanofibers were

named depending on their charge density. For instance, TOCN with 350

µmol/g was denoted TOCN350, while the ones with 650 and 1550 µmol/g

were TOCN650 and TOCN1550, respectively.

The TOCN suspensions treated once in the microfluidizer (DF 1) are used

to fabricate LIB separator materials (Paper I and II). Depending on the

types of functional groups, they were specifically denoted, e.g. TOCN350-

COO-Na+ and TOCN650- COO-Na+.

TOCN with 350, 650, and 1550 µmol/g treated by 6, 9, 12 times are

presented in Paper IV and V. Detailed protocols are addressed in Papers

I, II, IV and V.

3.1.2. Enzymatically treated CNF

Enzymatically treated fibers were manufactured following the protocols

reported by M. Henriksson et al.30. The pulp was dispersed in DI water,

passed through the 400 μm- and 200 μm-chambers once (DF 1) and a 1

wt % of aqueous suspension was prepared (Paper II).

The enzymatically treated cellulose dispersed in isopropanol (IPA) was

prepare by pretreating at 10% consistency in a phosphate buffer (11 mM

Na2HPO4 and 9mM NaH2PO4) using 25 ECU endoglucanase per gram of

Experimental - 13

dry fiber for 1 hour at 55°C. The pretreated fibers were washed and diluted

with IPA to 1 wt % and then processed in a microfluidizer (DF 3). An IPA

suspension of 0.64 wt % of CNF was obtained and used in Paper III.

3.2. CNF-based LIB separators

3.2.1. Fabrication of freestanding separators

In Paper I, an aqueous suspension containing TOCN350-COO-Na+ was

diluted to 0.1 wt % using a 10 mM HCl aqueous solution, hence Na+

counterions of COO-Na+ functional groups were substituted to proton. The

protonated form is denoted TOCN350-COOH. On the other hand, in Paper

II, the TOCN650-COO-Na+ was protonated by dialysis. A semi-permeable

dialysis bag containing 0.3 wt % of diluted TOCN650-COO-Na+ suspension

was immersed in 10 mM of HCl aqueous solutions stirred for 2 days. The

dialysis bag was transferred to DI water and washed until its conductivity

decreased below 1 mS/cm. The protonated fibers with 650 µmol/g were

named TOCN650-COOH.

Different types of cellulose nanofibers-based separators were prepared by

vacuum-filtration through a 0.22 µm pore sized membrane (Durapore,

Merck) and sequential solvent exchange was carried out by using 30 ml of

ethanol, acetone and pentane, successively. The filtered membranes were

dried overnight in a vacuum drier at 110 °C. The mass loading of the TOCN

membrane was ≈ 1 mg/cm2. The side of a membrane in contact with the

Durapore membrane during the vacuum-filtration process is denoted as

the bottom side, while the other side is named as the top side (located on

the top during the filtration). The TOCN based separators were denoted as

TOC followed by the quantity of surface charge density and the functional

groups, e.g. TOC350-COO-Na+ or TOC650-COOH (Paper I and II).

3.2.2. Spray-coated cellulose-based SEA

Spray-coating of a CNF-IPA suspension was implemented onto a heated

electrode (up to 145 oC), as shown in Figure 3.1A, until the desired

thickness of the CNF separator layer was achieved. A PVDF-HFP solution

dissolved in acetone was then spray-coated on the CNF coated on the

electrode and acetone was evaporated on the heating plate. The CNF

separator-electrode assembly was denoted as CNF-SEA. Different SEAs

Experimental - 14

were prepared varying the types of electrodes integrated with the separator

layer: (1) graphite SEA (separator thickness 30 µm), (2) graphite SEA (15

µm) – NMC (LiNi1/3Mn1/3Co1/3O2) SEA (15 µm) and (3) NMC SEA (22 µm)

with extra CNF-PVDF composite around the circular form of the electrode,

as a short-circuit preventive film for the full cell assembly (denoted as wing,

shown in Figure 3.1 B). The SEAs dried at different conditions (170 °C

overnight or 60 °C for 30 minutes) in a vacuum dryer were assembled to

batteries in an argon filled glove box.

A CNF-SiO2 NMC SEA was prepared by spray-coating 1 wt % of fumed

silica suspension dispersed in IPA on the top of a ready prepared CNF-

NMC SEA (30 µm thick in total). Afterward, PVDF-HFP/acetone mixture

was put on the SiO2 layer as shown in Figure 3.1C. A similar separator

layer was deposited on the surface of a high energy LiNi0.6Mn0.2Co0.2O2

(NMC622) electrode with a short-circuit preventive film was as well

around the edge of the electrode (named SiO2-CNF-NMC622 SEA, Figure

3.1 D).

Figure3. 1. Schematic illustration of fabrication of CNF-SEA (A) spray

coating of CNF and PVDF-HFP, successively, (B) fabrication of wing

around NMC SEA for full cell, (C) spray coating of SiO2 on the top of CNF-

SEA, (D) fabrication of wing around SiO2-CNF-NMC622 SEA for full cell

3.2.3. Electrochemical characterization of LIB separators

Electrochemical performance of CNF-based LIB separators were evaluated

by galvanostatic cycling tests and rate capability tests in the pouch cell

assembled in an argon-filled glove box (<1 ppm of O2 and H2O). Celgard

Experimental - 15

2325 was used as a reference separator and its electrochemical

performance was compared with that of CNF separators. Lithium nickel

manganese cobalt oxide positive electrodes (NMC, LiNi1/3Mn1/3Co1/3O2)

with 1.7 mAh/cm2 of capacity and graphite negative electrodes (2.2

mAh/cm2) were used as positive and negative electrodes, respectively.

High energy NMC622 has 3.5 mAh/cm2 of capacity and its matched

graphite has 4 mAh/cm2. As a liquid electrolyte, Selectilyte LP 40 from

BASF, composed of 1 M LiPF6 in ethylene carbonate (EC): diethyl

carbonate (DEC) 1:1 by weight, was used. The cells were cycled in the

potential window between 2.6 and 4.1 V versus Li+/Li. Ionic conductivity

of LP40 liquid electrolyte in a LIB separator was measured by

electrochemical impedance spectroscopy (EIS). In Paper I and II the

discharging rate capability tests and the long-term cycling tests were

carried out only using the protonated separators. In Paper III, liquid

electrolyte with VC was used for all electrochemical measurements. Details

are given in Paper I, II and III.

3.2.4. Gas evolution analysis by mass spectrometry

In-operando mass spectrometry measurements were conducted to analyze

the amount of gas evolved during the formation cycling (0.1C for 3 cycles).

The full cells with NMC and graphite but different separators were

assembled in a custom-designed stainless steel assembly-cell type. Briefly,

the gas analysis system consists of a battery cell setup with gas inlet

connected to a mass spectrometer. Argon gas is purged through a gas inlet

tubing and the gases evolved during the battery operation are pumped with

argon flow gas through a microflow capillary inlet (sample flow rate 12

μ/min) into an ion source. The ionized gas molecules are separated

depending on their mass/charge ratio (m/z) in the quadrupole mass filter

and analyzed in a detector. The types of gases analyzed were H2 (m/z 2),

C2H4 (m/z 26), H2O (m/z 18), CO2 (m/z 44), CO (m/z 28) and liquid

electrolyte. The electrochemical tests and the gas evolution measurements

by mass spectrometry were started simultaneously. The volume of the one-

compartment cell head space is approximately 10 ml, and no gas was added

to replace the sampled gas, which resulted in a decrease of the pressure and

absolute signal in the cell over time, whereas the ppm signal relates to gas

ratios in an accumulative way. Details are given in Paper II.

Experimental - 16

3.3. Rheology measurements of cellulose nanofibers

Rheological properties of 0.95 ± 0.05 wt % of concentration of TOCN with

different charge density and DF were measured. A rheometer with a

cylindrical cone-and-plate geometry (cone angle 2 °, diameter 25 mm, gap

height 50 μm) was used at a constant temperature of 23 °C. Oscillatory

frequency sweep tests were performed at angular frequency from 1 to 100

rad/s with constant 2 % strain. Details are given in Paper IV.

3.4. Electro-precipitation of nanocellulose hydrogels

and its applications

3.4.1. Electro-precipitation of TOCN hydrogels

TOCN with 650 µmol/g and DF 6 was diluted to 0.1 wt % aqueous

suspension using Ultra Turrax D125 Basic disperser at 10000 rpm for 5

min and named TOCN sol. Electro-precipitation of TOCN hydrogel was

conducted in a 2-electrode electrochemical cell setup on various substrates.

The substrates were immersed in the TOCN sol and connected as anodes

and a Pt mesh as a counter electrode. Anodic current was applied for a

certain deposition time to induce water electrolysis and gelation. TOCN sol

mixed with a pH indicator solution was prepared to visualize the pH profile

of TOCN sol during electro-precipitation.

3.4.2. Electro-precipitation of composites

The TOCN-graphite mixed sol was prepared by mixing 0.1 wt % of TOCN

and 2 wt % of graphite, followed by 20 min of ultrasonic treatment (Vibra-

cellTM, Sonics). Electro-precipitation of TOCN-graphite composite

hydrogel was conducted in the same cell setup as above. The thickness of

hydrogel composites on a copper foil were measured varying the deposition

conditions and the composition of the coating medium.

Multi-layered composites were fabricated by coating a hydrogel layer on

the top of the previous layer(s). A two-layered hydrogel was prepared by

switching the coating medium. A 1st coating layer was made by gelation of

TOCN-graphite mixed sol and a 2nd was made using TOCN sol (j = 2.5

mA/cm2 was applied for 30 s for each layer).

Experimental - 17

3.4.3. Application I: Eco-friendly fabrication of electrode for LIB

An aqueous suspension, composed of 0.1 wt % of TOCN with charge

density 1500 µmol and DF 6, 0.05 wt % of Super-P, and 2.34 wt % of

graphite, was prepared by ultrasonic treatment for 20 min. A circular

copper foil with 15 mm diameter was used as a coating substrate and later

as a current collector in LIB. The electrochemical coating was conducted in

the same 2-electrode cell setup, employing the copper foil as a working

electrode and a Pt cage as a counter electrode. 11.3 mA/cm2 of current

density was applied for 10 sec and subsequently the electrode was dried at

110 °C in a vacuum dryer overnight. The graphite coated copper electrode

was assembled with a Celgard 2325 separator and lithium foil as both

negative electrode and reference electrode. The cycling test of the graphite

electrode was performed by charging/discharging at the voltage range

from 2 mV to 1.5 V.

3.4.4. Application II: Selective layer for hydrogen production

This study was to investigate TOCN hydrogel as a possible oxygen

reduction reaction (ORR) protective layer during hydrogen evolution

production (HER). Pt is the most commonly used co-catalyst in

photocatalyst, where HER takes place. As a proof-of-concept, a HER model

study was carried out in electrochemical cell setups, see Figure 3.2. First,

a Pt rotating disc electrode (RDE) anode and Pt cage (cathode) were

immersed in the same container in Figure 3.2A, filled with TOCN sol, and

1 mA/cm2 of direct current was applied for 1 second. Photograph of the gel

coated Pt RDE is presented in Figure 3.2.A. Hydrogen production

capability of the Pt RDE in the presence/absence of the hydrogel protective

layer and the ORR preventive reaction were evaluated in a 3-eletrode cell

setup in Figure 3.2B, composed of a Pt RDE as a working electrode, a Pt

cage as counter electrode, and a Ag/AgCl reference electrode saturated by

KCl. The current density applied for HER is set in the same order of the

current density generated by photocatalyst.

To understand how hydrogel affects the HER, 3 different supplementary

electrochemical experiments were performed. (1) Electrochemically active

surface area (ECSA) of platinum in the presence/absence of the hydrogel

coating was measured following the protocol94,95. (2) To examine the effect

of hydrogel on permeation of H2, H2 oxidation was carried out96. (3) The

Experimental - 18

permeability of ferricyanide was measured by a CV measurement97. Details

can be found in Paper V.

Figure3. 2. Schematic illustrations of (A) the electrochemical coating and

(B) the measurement setups

3.5. Physical characterization of samples

The morphology of the samples was analyzed by Field emission scanning

electron microscopy (FE-SEM, Hitachi S-4800), at 1 kV. Energy dispersive

X-ray spectroscopy (EDX, X-Max 80 SDD, Oxford Instruments) was

operated with FE-SEM as an accessory, at 15 kV accelerating voltage (Paper

V). The microscopic analysis of TOCN single nanofibers used in Paper I

and II was carried out using a scanning electron microscope (JEOL JSM-

7401F, 20 kV) in transmission mode (STEM). In Papers I-III, the size

distribution of nanocellulose and the pore size distribution at the surface

of the separators were analyzed using ImageJ software v.151j898 and the

binarized microscopic images. In Paper I and II, overall porosity of the TOC

membranes was calculated from the weight change of a membrane before

and after immersing in butanol for 2 hours99. A Fourier transform infrared

(FT-IR) spectrometer was used to elucidate the mechanism of electro-

precipitation of TOCN (Paper V). Tensile strength of a specimens was

measured by an Instron 5944 single column tensile tester, with a rate of 10%

per minute.

Results and discussion - 19

4. Results and Discussions

4.1. CNF based separators with different charge density

4.1.1. Characterization of CNF and separators thereof

So far, the most common process of producing lab-made cellulose based

separators is vacuum filtration36,40,47,100. Using fibrous cellulose as raw

material for LIB separators is advantageous since the pores can be

naturally constructed between fibers. Nonetheless, the key factor of

producing highly porous separators aiming for fast charging application is

to maintain the porous structure upon drying, which can easily collapse

without sophisticated drying processes, due to high affinity between

hydrogen bonds and water.

The morphology of TOCN varied depending on the surface charge density.

Visuals of TOCN with different charge densities (350 and 650 µmol/g) and

size distribution of nanofibers are displayed in Figure 4.1 A. Both TOCN

are partially disintegrated, consisting of microfibers and numerous

nanofibrils exfoliated from the main body of microfibers. A TOCN650

single fiber looks more transparent in the STEM image with a lower

contrast (lighter color) than that of the TOCN350, which means that

TOCN650 has a less dense and more fibrillated structure due to the

increased interfibrillar repulsive forces.

The average width of TOCN350 fibers was reported to be 130 nm, which is

eight times higher than that of TOCN650 (16 nm). TOCN350 have

submicron sized fibers ranged 100 – 500 nm, whereas TOCN650 larger

than 100 nm were barely found. The results are in agreement with the

literature that TOCN with higher charge density has more dispersed

nanofibrillated networks compared to the lower charged one, when treated

under the same mechanical disintegration condition33.

The surface morphology of TOC350 membranes is addressed in Paper I.

The difference in physical properties and morphology between TOC350-

COO-Na+ and TOC350-COOH membranes is insignificant. However, their

electrochemical performance was distinctively different. The cell with

TOC350-COO-Na+ separator showed fast capacity fade, while the TOC350-

COOH counterpart showed good cycling stability. The electrochemical

performance of the TOC separators will be discussed in the following

section 4.1.2 and Figure 4.2.


Figure 4. 1. Comparison of characteristics of TOCN with different charge

densities (350 and 650 µmol/g) and membranes thereof, (A) microscopic

images of TOCN and size distribution, (B) the pore size distributions of the

top and the bottom sides of the TOC350-COOH and TOC650-COOH

separators

In this respect, the morphology of the membranes made of the protonated

TOCN with different carboxyl contents are compared in Figure 4.1.B. It

is noticeable that the nanofibrillated structure of the TOCN (STEM images

in Figure 4.1.A) and the surface morphology of the dried membranes

observed by SEM (Figure 4.1.B) are very similar. It indicates that the

mesoporous structure of the TOCN membranes originates from the

nanofibrillated network and there was no significant shrinkage of

nanofibrils during the membrane fabrication (protonation, filtration and


drying). The results emphasize that the sequential solvent exchange

effectively removes water from the cellulose network, consequently,

preventing significant pore collapse upon drying.

The separators fabricated by vacuum filtration had an asymmetrical

structure, where the top sides had a multi-scaled porous structure with

both macro- and nano-scaled pores, whereas the very flat bottom sides

only had mesopores. Porosity and the average pore width of the top sides

of the two separators were similar. On the contrary, for the bottom sides,

the TOC650 separator had lower porosity and smaller average pore width

than that of TOC350. The influence of such asymmetric structure on the

rate capability is seen in Figures 4.3 and 4.4.

Table 4. 1. Physical properties of Celgard 2325 and TOC separators

Porosity

(%)

Thickness

(μm)

Ionic

conductivity

(mS/cm)

Electrolyte

uptake (%)

Tensile

strength

at break

(MPa)

Celgard

2325 46 ± 2 23 ± 2 0.81 ± 0.02 85 ± 4 170

TOC350

- COOH 62 ± 3 34 ± 4 1.06 ± 0.07 ≥200 3.5

TOC650

- COOH 58 ± 2 33 ± 2 0.66 ± 0.03 ≥200 9

The low ionic conductivity of the TOC650 separator, seen in Table 4.1, is

likely due to the smaller average pore width and lower porosity in the

bottom side of TOC650. The highly dispersed fine nanofibrils were

vulnerable to tolerate the strong capillary forces between the fibers and

water. Thus, it is likely to more aggregate, compared to the microfibrillated

fibers, resulting in the pore closure during the drying process. On the other

hand, due to the compact structure with less porosity resulted from the

presence of a large amount of nanofibrils, TOC650 membrane has 2.6

times higher tensile strength than that of TOC350. The results demonstrate

that increase in charge density of CNF leads to decrease in the porosity and

the average pore width of the TOC650 membrane. The effect of charge

densities and the following morphological changes on the electrochemical

performance of LIB separators will be more discussed in the following

section 4.1.2.


4.1.2. Electrochemical performance of TOCN separators

Figure 4. 2. Electrochemical performance of the LiNMC|Graphite cells

with different separators at 0.1 C charging/discharging rate: (A) TOCN-

COO−Na+, (B) TOC350-COOH, (C) galvanostatic fast charging and

discharging cycling test at 1 C and (D) chronopotentiogram of the cells at

1st, 20th, 50th, and 100th cycles (discharging).

Figure 4.2 (A-B) shows the dramatic improvement in electrochemical

stability of a TOC350 separator, after the removal of sodium counter ions

in the functional group of TOCN. The cycling performance of the cell with

the TOC350-COOH separator at 1C charging/discharging rate was

comparable to that of Celgard 2325, with high stability (Figure 4.2C).

Relatively higher ohmic loss and lower specific capacity of the cell

containing the TOC350 separator shown in Figure 4.2 D were seemingly

attributed to its higher thickness and the low porosity at the bottom side,

compared to that of Celgard 2325.

Figure 4.3 A depicts the effect of the asymmetric structure of the TOC350

separator on their rate capability. In the full cell named TOC350-Top, the

top side of the separator is in contact with the negative electrode, whereas


the bottom side contacts the negative electrode in the cell named TOC350-

Bottom. Results in Figure 4.3 B demonstrate how the transport of lithium

ions varies depending on the surface porosity. When the porous top side

was faced towards the negative electrode, the transport of lithium ions was

facilitated during the discharging process, resulting in high discharging

capacity, even at 5 C (7.3 mA/cm2).

Figure 4. 3. (A) Schematic description of the cell configurations

membranes and lithium ion flux through the surface of separators

considering the asymmetric structure of TOC350-COOH, (B) the

discharging rate capability of the cells varying the orientation of the

separators, charged at 0.2 C and discharged at different rates

The cell with a TOC650-COO-Na+ separator in Figure 4.4A showed

failure during cycling test, similar to that of TOC350-COO-Na+ (Figure

4.2A). It can be speculated that such failure was attributed to (1) the

presence of strongly bound water or, predominantly, (2) the deposition of

sodium on the graphite surface. Berthold et al. 101 reported that the amount

of bound water of carboxymethyl cellulose (CMC) with sodium counterions

in functional groups becomes 2−4 times higher than that of the protonated


counterpart. Furthermore, once sodium is deposited on the surface of

graphite, it is likely to interrupt lithium intercalation into the graphite.

Figure 4. 4. Electrochemical performance of NMC|graphite full cell in

pouch cells with TOC650 separators with different counterions in

functional groups and electrolyte (A) TOC650-COO-Na+ without additive,

(B) TOC650-COOH without additive, (C) TOC650-COOH with 2wt% of VC

additive, (D) discharging rate capability test (charging rate 0.2 C), (E)

cycling stability test at 0.5 C of charging and discharging rate

Differently from the case of TOC350-COOH (Figure 4.2B), the cell

containing the protonated TOC650-COOH showed poor cycling stability,

despite of the removal of sodium ion (Figure 4.4B). Interestingly, the

electrochemical stability of the TOC650-COOH separator was considerably

improved by adding VC, as shown in Figure 4.4C. Thus, the cycling

stability test of TOC650 were performed with the electrolyte containing 2

wt % of VC (Figure 4.4 D-E). For further discussion, see section 4.3

about the cause of cell failure in the presence of sodium counterions in

sodium carboxylate surface groups and the effect of VC related to the gas

evolution in LIB.

The difference in the discharging specific rate capability of the TOC350 and

650 separators was greater at fast discharging rate, higher than 2C (Figure


4.3B and 4.4D). It is probably attributed to lower porosity of the bottom

side of the TOC650 separator. At 1C cycling test, the specific capacity of the

TOC650 cell was ≈80% of that of Celgard 2325 (Paper II), however at 0.5

C they showed similar capacities (Figure 4.4E). Accordingly, the results

of the rate capability tests imply that, by locating the porous side of the

separator in contact with the desired electrode and adjusting the testing

process, better discharging rate capability can be obtained.

Figure 4. 5. (A) Cycling stability test of the cell containing TOC650-

COOH separator, tested at 0.6 C of charging rate of constant current and

2.3 C of discharging rate as a function of cycle numbers, (B)

chronopotentiogram of the cell at the 20th cycles, as a function of time (x-

axis, bottom) and specific capacity (x-axis, top).

Figure 4.5 shows the cycling performance of the cell with TOC650-

COOH-Top containing VC additive, tested at 0.6 C (0.74 mA/cm2) of slow

charging and 2.3 C (2.95 mA/cm2) of fast discharging. The specific capacity

of the initial discharging was 105 mAh/cm2, which is 90% of the initial

capacity at 0.6 C. The results demonstrate that the TOC650 separator can

be applied for the battery with high power, applying a current > 2 mA/cm2.

4.1.3. CNF separators with different charge density

The relationship between the characteristics of nanocellulose, treated by

different processes, and their performance as separators is discussed in

this section. Figure 4.6 displays the electrochemical performance of the

different separators at 1C, applying a constant current of 1.5 mA/cm2. The

enzymatically treated cellulose has the same chemical structure as pristine

cellulose without carboxylic functional groups and its charge density is 44


µmol/g102. The enzymatically treated CNF membrane has poor mechanical

integrity due to its inhomogeneous dispersion. TOC separators show not

only higher mechanical strength but also better electrochemical

performance. Introducing 350 µmol/g of charge density in cellulose chain

is shown as the best condition to produce the separator with high

performance.

Figure 4. 6. Effect of the charge density of nanocellulose (degree of

fibrillation, 1 time passed) on electrochemical performance as LIB

separators (bar, specific capacity at 1C charging rate), and the mechanical

strength of the CNF membranes (dotted red line with diamond marker, the

same mass loading, produced by the same vacuum filtration)

Increasing the charge density contributed to improvement of the

mechanical strength of the cellulose membranes. The lower specific

capacity and higher mechanical strength of TOC650, compared to TOC350,

can be related to its morphological properties, e.g. lower surface porosity

and average pore width and ionic conductivity. Besides, TOC650 separator

requires VC additive to achieve the stable cycling performance.

Nevertheless, TOC650 can be regarded as a separator with good

mechanical integrity and electrochemical performance, capable of high

current density, ≈ 2.95 mA/cm2. To acquire a cellulose membrane with a

larger amount of finely dispersed cellulose nanofibers and higher tensile

strength, cellulose with higher charge density can be prepared and treated

to a higher degree of fibrillation. However, it is likely to increase the

amount of bound water in cellulose structure with the risk of failure during

cell cycling. Accordingly, it is noted that cellulose separators should be


produced with consideration of a trade-off between the desired properties,

e.g. mechanical strength, porosity and electrochemical performance.

To meet the industrial needs for high power or energy density battery

applications, highly porous separator materials which can apply >2.2

mA/cm2 are required103,104. In this respect, both TOC separators can be

considered as the promising separator materials fulfilling such

requirements. Particularly TOC350, is capable of sufficiently good

capacities; 1.5 mA/cm2 for 1C, 2.9 mA/cm2 for 2.3 C, and 7.3 mA/cm2 for

5C (Paper I). Therefore, it can be concluded that introduction of 350 - 650

µmol/g of carboxylate groups was beneficial to efficiently disintegrate

cellulose to nanofibers and to produce highly porous membrane for fast

discharging LIB applications, showing good electrochemical performance.

Further studies to improve its mechanical strength will be required, by

additional mechanical and chemical treatment38,60,105,106. In addition, such

aqueous-based processes require high energy input for the drying process,

e.g. overnight drying time at elevated temperature (>110 °C). A way to

overcome such drawback of the cellulose based porous separator is the

development of new types of integrated battery structures. In the following

section a CNF based separator-electrode assembly fabricated using a fast

and simple fabrication method, is addressed (Paper III).


assembly (SEA)

4.2.1. Fabrication of SEA

In the state-of-art battery manufacturing process, electrodes and

separators are separately produced and assembled later during a winding

process. A separator must have sufficiently high tensile strength and high

Young’s modulus not to be broken and deformed while wound with the

electrodes as a high pulling force is applied. Tensile strength of cellulose

separators has been reported in the range of 12-55 MPa, which is 10-30%

of that of commercial polyolefin separators (≈ 170 MPa) 36,49,60,107. Thus, it

is likely very challenging to obtain freestanding cellulose-based separators

with tensile strength comparable to that of the polyolefin ones without

compromising the rate capability and increase in resistivity.


Figure 4. 7. (A) Single CNF building block dispersed in IPA and (B) its

magnified view with nanofibril network, (C) porous CNF separator layer in

SEA, (D) histogram of the fiber width of nanofibrillated cellulose and (E)

histogram of pore size distribution of the CNF membrane.

Here we introduce a CNF separator/electrode assembly (SEA) produced by

spray-coating method, which is advantageous to supplement the low

mechanical strength of the separator. The spraying medium is

enzymatically treated cellulose mechanically dispersed in IPA. Figure

4.7A demonstrates a CNF building block, with 20 to 30 µm long and 10

nm to 2 µm wide. The difference in polarity between water and IPA has a

significant impact on the mechanical disintegration by the microfluidizer,

since the hydrogen bonds will be more stable in IPA than in water,

therefore the fibers are cut rather than delaminated.

Figure 4.7C exhibits the porous CNF separator in SEA, in which the

nanofibrillated structure of CNF was well-maintained after solvent

evaporation during the spray-coating made using a normal cosmetic spray.

The lower surface tension between the fibers and IPA enables fabrication

of a porous membrane without pore closure upon fast drying108,109. On the

contrary, if the CNFs were dispersed in water, it required a sophisticated

drying process, as above, e.g. solvent exchange, vacuum or freeze drying

for a long time (overnight or more)110. Size distributions of the width of the


fibers and the pore diameter are displayed in Figure 4.7D-E. The CNF

has similar dimensional characteristics as those of the TOC separators, but

slightly larger average pore width and higher porosity.

Figure 4.8. (A) Schematic illustrations of spray-coated CNF separator

/NMC electrode assembly, (B) CNF coated on the electrode surface and its

plain-view and cross-sectional images, thickness < 3 µm, (C) thickness ≈

22 µm, (D) PVDF-HFP spray-coated on the top of a 22 µm thick CNF layer

The CNF-SEA was fabricated by repeatedly spray-coating a CNF-IPA

suspension on the top of an electrode, until a desired thickness was

achieved, as shown in Figure 4.8. A layer of less than 3 µm was too thin

to be used as a separator, not fully covering the surface of the electrode,

which can result in short-circuit in LIB (Figure 4.8B). The electrode

surface is not observed anymore in Figure 4.8C-D, when the thickness

achieved was ≈ 22 µm. PVDF-HFP plays a role improving the dimensional


integrity of the CNF-SEA, binding the CNFs and increasing their flexibility.

Here the development of a separator layer of 22 µm is achieved, which is

slightly thinner than that of previously reported cellulose separators of ≈

30 µm 40,111 or other SEAs55,58,61. Woo et al.60 reported on a SEA with a 12

µm separator, which implies that SEA is a promising process to obtain a

thin separator with low resistivity. There was no significant difference in

the in-plane view of the CNF membrane observed before/after PVDF-HFP

coating, macro and mesopores remained without pore closure. Detailed

physical properties of the CNF-PVDF-HFP separator are addressed in

Table 4.2.

Table 4. 2 Physical properties of CNF-PVDF-HFP separator

CNF PVDF-HFP Ionic

conductivity Wettability

Thickness

(µm)

weight

(mg)

Thickness

(µm)

weight

(mg) mS/cm %

CNF/PVDF-

HFP 21 2 2±1

0.3

±0.1 1.08 ±0.2

256.7 ±

0.4

A membrane made of the CNF fibers with low aspect ratio and by the same

vacuum filtration had very weak mechanical strength, even in comparison

with the previously mentioned TOC separators. Nonetheless, it was

advantageous to use the fast evaporating organic solvent suspension, since

the spray-coated cellulose building block can adhere to the surface of the

electrode very well during solvent evaporation. It enables fast fabrication

of highly porous membranes, preventing adsorption of water into the

cellulose structure, which can cause detrimental side reactions in LIB. It

was also convenient to acquire a desired thickness, staking the CNF

building block by a bottom-up approach. The separator integrated with an

electrode can tolerate high tensile stress, when pulled in the in-plane

direction, since the mechanical strength of the SEA is more determined by

that of the current collector, to which the separator is attached. Tensile

strength of the NMC electrode used in this study was ≈30 MPa.

Accordingly, we can assume that the tensile strength of the SEA is

improved to that of NMC electrode composite.

Note that it is not easy to find suitable solvents and new materials for SEA,

other than NMP or dimethylformamide and inorganic nanoparticles54–

56,59,62. Presumably, there is less freedom when designing a separator for a

SEA if it requires extra pore forming processes such as phase separation,


using different polarity of solvents and solubility of materials. Besides, the

stability of the electrode materials to the solvents must be carefully

considered, not to deteriorate the cell performance. For instance, aqueous

suspensions are not preferred to be in contact with the normal electrodes,

since the active materials coated are likely detached from the current

collector as water pervades between the active materials and the binders.

Such limitation can be a major barrier for application of diverse materials

for SEAs. Casting a polymer solution without pore-forming agents, e.g.

inorganic nanoparticles, is likely to form a non-porous film, which is not

favorable as separators for high energy or power LIB applications. In this

respect, using CNF as a ready-porous building block and IPA breaks

through such obstacle, enabling the fabrication of SEAs capable of fast

charging (discussed in the following section 4.2.2), without degrading

the performance of the electrode. Here we can put more emphasis on the

advantages of this technique using cellulose as biodegradable material and

using non-hazardous solvents, which can be considered as an eco-friendly

process.

4.2.2. Electrochemical performance of CNF-SEA

The cycling performance of the cells depends on the configuration of the

SEAs. When the CNF separator was coated on the graphite electrode, the

cell showed poor cycling stability with a sudden drop in specific capacity

and low coulombic efficiency (or cell failure), due to high volume

expansion of graphite (Paper III). Volumetric change of graphite is known

as 10%112, while LTO is 0.2%113, LiMn1.5Ni0.5O4 (NMO)114 3-6.3%, LFP 7%114,

LCO 2%115, and NMC 2%11,115. To achieve successful cycling performance of

the SEAs with high volumetric change electrodes, such as carbonaceous

materials, tough and durable materials which can tolerate stress and strain

are necessitated, e.g. PVDF, PVDF-HFP, poly(phenylene oxide), or

inorganic nanoparticles 54,55,60,62. It seems that the CNF based composite

structure with low loading of PVDF-HFP is vulnerable to high stress and

strain from the volume expansion of the electrode.

Seeing the configuration of normal LIB, the sizes of the negative electrode

and the separator are in general slightly larger than that of the positive

electrode 64,65. Therefore, in SEA, it has been preferred to deposit a

separator on the negative electrode side54,55,60–62. LTO seems an

advantageous substrate, but because of its high operating potential ≈ 1.5 V


vs Li+/Li, it compromises the cell voltage and consequently power or

energy density of LIB56,57,61. In this respect, NMC seems a favorable

electrode material for CNF-SEA. To deposit a separator on the NMC side,

a cell design preventing direct contact between the edges of the electrodes

is required. Thus, extra CNF-PVDF-HFP was decorated along the edge of a

disc-type NMC electrode, as shown in Figure 4.9 A.

Figure 4. 9. (A) Photograph of CNF-NMC SEA and microscopic image of

the edge of the wing and the porous separator, (B-D) electrochemical

performance of lithium ion battery containing CNF-NMC SEA and the one

with Celgard 2325, (B) EIS of the CNF-NMC SEA after formation, (C)

cycling stability at 1C of charging rate, (D) rate capability test

After the SEI formation, the SEA had a lower resistivity than the cell

containing Celgard 2325 (Figure 4.9B), showing good electrochemical

performance as well (Figure 4.9C-D). Particularly, the CNF-SEA dried at

60 °C for 30 min in vacuum right after fabrication, showed excellent

electrochemical performance, similar to the one dried at 170 °C. It implies

that this manufacturing process can considerably reduce the energy and

time input for LIB production. After cycling tests, the separator layer


remained on the surface of the NMC electrode without significant

morphological changes and pore closure (Paper III).

Another advantage of the spray-coating method is that the CNF separator

can be coated on cells of any size and shape. A proof-of-concept of a small

sized SEA cell was made, inspired by a commercial prismatic cell, and

cycled under bent condition as shown in Figure 4.10. The CNF-NMC SEA

exhibited good cycling stability as 91.5% of the specific capacity remained

after 100 cycles, whereas the Celgard separator containing cell lost 18% of

the specific capacity during the same time. It is presumably attributed to

the good adhesion between the separator and the electrode in the SEA.

Figure 4. 10. Proof-of-concept of a small sized-prismatic CNF-NMC

SEA|graphite full cell cycled under bent condition (A) the image of the cell

configuration (the electrode located along the red line in the scheme) and

(B) cycling stability of the CNF-NMC SEA|graphite full cell at 0.25 C of

charging/discharging rate (marker open for charging, the filled for

discharging, x and + for coulombic efficiency)


4.2.3. Electrochemical performance of SiO2-CNF-SEA

Ceramic coated separators have been applied for commercial cells,

believed as a key component for high performance LIB 2,14,17,19. SiO2 can

tolerate external mechanical pressure and so enhance the rigidity of the

SiO2 coated separator.

Figure 4. 11. Schematic illustration of the effect of volume change of

electrodes on the separator in SEA during the cycling (A) CNF-SEA and (B)

SiO2-CNF-SEA and cycling performance of the full cell

Targeting the application for a high energy density battery, CNF-NMC622

SEAs were made and cycling tests were carried out. Since the NMC622

electrode was developed aiming for high energy density, the electrode and

the matched graphite also have ≈ 2 times higher mass loading and capacity

per unit area than that of NMC. It is assumed that the high volumetric

changes of NMC622 can increase the non-uniform pressure applied to the

CNF separator, which can lead to fracture and delamination of the

separator, as described in Figure 4.11A. Consequently, it caused poor

cycling performance with fluctuating capacity and internal short circuit

(Paper III). When a thin SiO2 layer was coated on the top of the CNF


separator, the layer could effectively tolerate the stress and the pressure

against the separator. Accordingly, it is believed that the SiO2 layer

contributes to the highly improved cycling stability of the SiO2-CNF

NMC622 SEA (Figure 4.11B). Furthermore, it seems that the SiO2 layer

not only enhances the mechanical stability of SEA but also facilitates

transport of lithium ions, displaying increased specific capacity.

Figure 4. 12. Schematic illustration of structure of SiO2-CNF NMC SEA

and morphology, electrochemical performance of SiO2-CNF-NMC SEA|Li-

metal battery (B) cycling stability test at 0.2 C of charging/discharging rate

(0.2 mA/cm2, marker open for charging, the filled for discharging, x and +

for Coulombic efficiency) and (C) the rate capability test

Lithium metal is ultimately the most favorable negative electrode, due to

its high specific capacity >3000 mAh/g and low redox potential, even

though its instability makes its commercialization challenging.

Nonetheless, it is still worth to put more efforts to develop stable and safe

lithium metal batteries. Unfortunately, the wood-based CNF separator

used in this study was not suitable for a lithium metal battery. This was

probably because of undesirable side reactions between lithium metal and


water residues, and highly porous and soft surface properties leading to

dendrite growth. The separator in contact with lithium metal got yellow

coloration on the surface, as shown in Figure 4.12A. On the contrary, a

thin SiO2 coating overlaid on the top of the CNF separator prevents direct

contact between CNF and lithium metal and consequently enables

successful cycling of a proof-of-concept lithium metal battery (Figure

4.12B-C). The cycling performance and the rate capability were

comparable to that of Celgard 2325.

4.3. Gas evolution of LIBs with CNF separators

Figure 4. 13. Gas evolution of NMC|Graphite full cells with different

separators and electrolyte components as a function of time and cycle

numbers (for 3 cycles at 0.1 C, C: charging, D: discharging), (A) H2 and (B)

C2H4, and the cells named after separator thereof

Results showing that protonation of TOC350 separator dramatically

improves the battery performance (Paper I) may indicate that the presence

of sodium ion causes failure of the cell. It was presumed to be due to bound

water (Paper I), according to the literature that the amount of bound water

varies depending on the type of counter ion in the carboxylate groups of


cellulose derivatives 101,116. Another possible explanation is that sodium

ions can be dissociated into the LIB electrolyte, and deposited on the

graphite electrode, which can interrupt lithium intercalation.

In-operando gas evolution measurements can help to better understand

the effect of sodium counterions in the surface functional groups of the

TOCN on the electrochemical performance of the LIB containing the TOC

separators. Note that the cycling behavior of the cells differed depending

on the cell design. Even though the cell with the TOC650-COO-Na+

separator showed a fast capacity fade in the pouch cell (Figure 4.4A), the

same cell assembly in the mass spectrometry measurement setup had

stable performance without fast capacity fading during the first 3 cycles at

0.1C of the cycling test (≈ 60 hours). It is most probably related to

continuous evacuation of gas during the mass spectrometry analysis.

Besides, the mass spectrometry cell setup required four times higher

amount of electrolyte, than that of the pouch cell, which can dilute the

sodium ion contents and their effects on the cell performance is attenuated.

On the contrary, the pouch cell is a closed system, where gas can

accumulate as the cycling test proceeds and it is likely to have more

aggressive side reactions, followed by cell failure.

Figure 4.13 demonstrates the H2 and C2H4 gas evolution measurements

with different separators and the addition of VC to the electrolyte. Since

there was no significant trend in the CO2 and CO gassing reactions at

potentials below 4.2 V, those results are not included. The highest amount

of H2 and C2H4 gases were released from the cell containing TOC650-

COOH separator, without VC, but the amount of gases evolved in the

TOC650-COO-Na+ cell was similar to that of Celgard 2325. The staircase-

like graph of H2 signal intensity indicates that the increase in the rate of

the gas evolution is related to the potential changes, which disagrees with

the literature that H2 gassing is a potential independent reaction79. See

Figure 4.13A, the rate of signal intensity rapidly increased in the time

ranges of 0-10, 20-30, and 40-50 hours during the charging (C), compared

to the rest of the time ranges, corresponding to the discharging (D).

The amount of gas evolved from the Celgard 2325 cell can be considered as

a reference where the “normal” gas evolution reactions take place without

significant side reactions between the electrode/electrolyte and the

separator. In the Celgard 2325 containing cell, a sudden increase of C2H4

are observed during the first charging (up to ≈ 10 hours in Figure 4.13B)

and a continuous gas evolution (H2 and C2H4) during the following cycling


process as well. We assume that, in the Celgard 2325 cell, gas evolution

reactions related to the decomposition of electrolyte only take place, as

proposed in the literature70 and it is schematically illustrated in Figure

4.14. At the positive electrode, the electrolyte solvent can decompose by

ring opening and the protic electrolyte decomposition compounds are

produced (Figure 4.14 (A)). These electrolyte decomposition products

can diffuse to the graphite electrode surface (Figure 4.14 (B)) and be

reduced and generate H2 gas (Figure 4.14 (C)), which causes the

continuous H2 gas evolution over the cycling. A sudden increase in the C2H4

signal in the beginning of the cycling test (0 ~ 10 hours) is because of the

decomposition of electrolyte solvents during the SEI layer formation

(Figure 4.14D)70,79. Note that the same gas evolution reactions take place

even in the cells containing the TOC650 separator.

Figure 4. 14, Schematic illustrations of electrolyte decomposition

reaction and gas evolution occurring in the cell containing Celgard 22325

separator as a reference cell (without VC)

That is, when a higher amount of gas was generated in the cells containing

TOC650 separators, it can be assumed that there were some additional side

reactions between the separator and the battery components. The possible

reactions taking place with cellulose components in LIB are the following

and described in Figures 4.15 and 16;

Reaction 1: reduction of H2O to H2 (H2O + e- → OH- + ½ H2 (g)) 70,71,117

Reaction 2: deposition of Na (Na++ e- → Na, 0.3 V vs Li+/Li)

Reaction 3: substitution of COO- and OH to COO-Li+118–120

Reaction 4: the reduction of H+ from COOH and OH to H2121.


More detailed reaction schemes are addressed in Paper II.

Figure 4. 15. Schematic illustrations of possible reactions in the cell

containing TOC650-COO-Na+ separator (without VC)

Figure 4.15 schematically describes the possible reactions taking place in

the cell containing a TOC650-COO-Na+ separator. The reduction of H2O

and protons at the graphite surface are likely to cause the high amount of

H2 evolution in the cell. However, seeing Figure 4.13 A, less H2 gas was

released from the TOC650-COO-Na+cell, compared to that of its

protonated counterpart, or similar to that of the Celgard 2325 cell. During

the first charging (~10 hours), the TOC650-COO-Na+ cell produced slightly

less C2H4 gas (red line, Figure 4.13B), compared to that of the Celgard

2325 reference cell (blue line, Figure 4.13B). Accordingly, it can be

speculated that a reduced active surface area on the graphite restricts the

formation of a stable SEI layer and other surface reactions as well. It

implies that probably not trace water but another reason, possibly sodium

deposition on the graphite surface, is a dominant cause for failure of the

pouch cell containing TOC650 separator with COO-Na+ functional groups.

Sodium ions in carboxylate groups can be replaced by lithium ions forming

(-COO-Li+) and dissociated to the electrolyte by ion exchange119,120.

Considering the electrochemical potential of sodium (-2.7 v vs SHE), the

sodium ions released would be deposited on the graphite surface at a

higher potential than lithium intercalation into graphite (0.3 V vs Li+/Li),

possibly causing sodium plating at the lower potential with little

intercalation 122,123. Thus, sodium deposition is likely to interrupt the

intercalation of lithium into graphite. A low concentration of sodium ions

(< 0.22 M) was found to enhance a stable SEI layer formation, and


consequently electrochemical stability of LIB thereof, whereas higher

concentrations of sodium added (> 0.44 M) deteriorated the performance

of the cell124. Thus, it is speculated that restricted lithium intercalation

possibly hinders the cell cycling, showing poor specific capacity and cycling

behavior as shown in the pouch cell, Figure 4.4A.

Figure 4. 16. Schematic illustrations of possible reactions in the cell

containing TOC650-COOH separator (without VC)

The highest amount of H2 gas was generated from the cell with a TOC650-

COOH separator and no VC addition. Figure 4.16 illustrates the possible

reactions taking place in the cell. A sudden increase of H2 in the initial

cycling (0-10 hours, black line in Figure 4.13A) can be possibly attributed

to Reactions 1, 3 and 4 at the graphite surface, and further continuous

evolution is due to the decomposition of electrolyte, the same as in the

Celgard 2325 reference cell (Figure 4.14). Due to the complicated

reactions of gas evolution, it was difficult to quantitatively elucidate the

origin of the H2 gas.

Interestingly, having 2 wt % VC in the electrolyte, overall gassing in the cell

with TOC-COOH separator was considerably suppressed, including H2 and

C2H4. A suggested mechanism of VC suppressing the gas evolution and

further side reactions in LIB is that a VC containing SEI layer prevents the

reactions described in Figure 4.16, and consequently much less H2 and

C2H4 gases are evolved 70,71.

Potential dependent H2 gas evolution shown in Figure 4.13A indicates

that CNF based separators likely have higher risk of cell failure without VC

when used in LIB with an electrode operated at high potential, e.g. NMC,

compared to low potential counterpart such as LFP or LTO. Nonetheless,


adding VC, the TOC separators are sufficiently stable to be applied for NMC

despite of such risk, and possibly capable of other high potential LIB, such

as lithium nickel cobalt aluminum oxide (LiNiCoAlO2) or high nickel NMC.

In this respect, the addition of VC is necessarily required when applying

nanocellulose to LIB. These reaction dynamics imply the need of careful

investigations about detailed chemistry of cellulose based material in LIB,

especially with high voltage positive electrode materials.

Figure 4. 17. H2 gas evolution of the CNF-NMC SEA|graphite LIBs dried

at 60 °C for 30 min, in the presence/absence of 2 wt % VC in the electrolyte.

As shown in Figure 4.17, the gas evolution from a CNF-SEA LIB was

analyzed in the same way as above. Here we considered H2 as the most

significant indicator to evaluate the electrochemical stability of a material

in a LIB. The enzymatically treated cellulose separator is expected to have

the similar reactions to that of TOC (Reactions 1,3 and 4). Even though

CNF was fabricated in a water-free environment, a high amount of gas was

released from the CNF-SEA cell without VC (Figure 4. 17). Nonetheless,

the amount of H2, even slightly higher than that of Celgard 2325 in Figure

4.13, is likely attributed to the reduction of the trace water and hydroxyl

surface groups, as previously mentioned (Reactions 1, 3 and 4). Water

could probably be introduced to the nanocellulose structure during the


enzymatic treatment conducted in aqueous environment and remain after

the short-term drying process. Nevertheless, the amount of H2 gas

generated from CNF-SEA (black line in Figure 4.17) was even less than

that of the aqueous based TOC separator, dried overnight (black line in

Figure 4.13A). The result greatly highlights the advantage of using CNF

treated in organic solvent which considerably improves drying efficiency.

By adding VC, the total amount of H2 gas evolved from the cell containing

CNF-SEA was significantly suppressed, ≈ 29% less than that of the Celgard

2325 reference cell (without VC). Thus, it can be concluded that the

combination of water-free conditions using CNF dispersed in IPA with the

addition of VC greatly increase the drying efficiency of the CNF based

separator manufacturing process.

4.4. Discussion about CNF-based LIB separators

To sum up, we have demonstrated wood based TOCN as LIB separator with

excellent electrochemical performance and high rate capability. Varying

the surface charge density of the cellulose, the effects of the carboxylate

surface group and its counterions on the morphology, physical properties

and electrochemical performance of the separators were studied.

Microfibrillated cellulose with nanofibril networks were obtained by

mechanically disintegrating the carboxylated pulps at mild conditions.

Increasing the charge densities of TOCN from 350 to 650 µmol/g, more

loose and fibrillated networks were acquired. Such a larger amount of

finely dispersed nanofibrils in TOCN650 contribute to the increase in the

tensile strength of the TOC650 separator, whereas it compromises its rate

capability and stability of LIB. Nevertheless, considering the asymmetric

structure of the membranes fabricated by vacuum filtration and adjusting

the cycling test conditions, better rate capability could be accomplished at

the charging rate > 2 mA/cm2. Electrochemical stability of the TOC350-

COO-Na+ separator was improved by substituting sodium ions to protons.

However, the TOC650-COOH separator necessarily required VC additive

for stable electrochemical cycling performance in LIB.

Even though the rate capability of TOC separators was sufficiently high, the

low mechanical strength remains a major drawback of the cellulose

separator. To overcome such disadvantage, a CNF based SEA was

developed. The CNF separator is integrated with the electrode and current

collectors and has highly enhanced dimensional stability. The CNF-SEA


was fabricated by spray-coating a CNF-IPA dispersion on a target electrode.

The method can be considered as an eco-friendly process, since it uses

biodegradable materials without using hazardous solvents. Additional

advantages of the CNF-SEA are (1) facile manufacturing of the separator

due to fast evaporation of IPA, (2) porous structure obtained without

additional pore-forming process, (3) less water adsorption, and (4) high

thermal stability. NMC SEA shows excellent electrochemical stability,

comparable to that of the cell with a Celgard 2325 separator, not only in a

normal disc-type small cell, but also in a proof-of-concept prismatic cell,

operated in bending. A SiO2 coating layer makes the CNF-SEA work not

only for high energy density batteries but also for lithium metal batteries.

We believe that the development of such SEA cell designs enables

applications of a variety of functional porous materials for separators,

which were considered as not attractive materials due to low mechanical

strength.

In-operando gas evolution measurements were carried out to investigate

the cause of failure of the cell containing cellulose based separators. It is

speculated that not trace water but sodium deposition on the graphite

surface is a dominant cause for failure of the cell containing TOC

separators with COO-Na+ functional groups. Noticeably high amounts of

H2 released from the TOC650-COOH cell seems to originate from the

reduction of (1) the trace water and (2) the degradation of electrolyte

decomposition products and (3) the reduction of protons, released from

COOH and OH, into H2 gas at the graphite surface. Such significant gas

generation measured by the mass spectrometry study can be related to

failure of the cell containing a TOC650-COOH separator in the closed

pouch cell system. The gas evolution was considerably suppressed by the

addition of VC, and the cells showed excellent electrochemical

performance with high cycling stability. Therefore, it is concluded that the

addition of VC is compulsory in the LIB likely to contain trace water (e.g.

produced by an aqueous process or containing hygroscopic materials,

biopolymers). CNF-SEA also demonstrated the same trend in the presence

/ absence of VC. Dispersing CNF in IPA, and adding of VC in CNF-SEA,

saves time and energy for drying process, without compromising the

electrochemical stability.


4.5. Rheological properties of TOCN

In the previous section, to obtain a desirable mesoporous structure as

separators, cellulose was treated to a low degree of fibrillation (DF), by

passing through the microfluidizer for 1-3 times. When increasing the DF

and the surface charge density, fibers become more dispersed to nanosized

individual fibrils. Such nanofibrils are not suitable for separators, since

during the drying process they are more likely to aggregate forming thin

and compact films with nanopores. However, due to highly enhanced

dispersibility, cellulose nanofibrils have unique optical, mechanical and

colloidal properties, different from that of microfibrillated TOCN. Then the

material is more likely suited as building block for hydrogels, available for

a variety of versatile soft matter applications, e.g. biomedical or tissue

engineering, or drug delivery80–83,88,89,125. In LIB applications it can be used

as a binder material, instead of PVDF.

Figure 4. 18. Effect of different manufacturing parameter of 0.95 wt % of

CNF suspension, the charge density (350, 650 and 1550 µmol/g) and the

degree of fibrillation (DF 6, 9, 12)

The strength of TOCN-nanofibril networks varies depending on the

pretreatment parameters (e.g. degree of fibrillation, the surface charge,

solid content) and the physical properties of the nanofibrils. It is well

known that high charge density, high aspect ratio and high concentration

can lead to higher storage modulus, with stronger interfibrillar interactions.

Oscillatory shear measurements were performed to study the viscoelastic

properties of the structures. The viscoelastic behavior of TOCN nanofibril

suspensions with different charge densities and different DF are shown in

Figure 4.18, expressed as storage modulus (G’). The high storage

modulus of the TOCN with 350 and 650 µmol/g of charge density and

treated at the mildest mechanical condition (DF 6) indicates that the TOCN


were less disintegrated compared to those with higher charge density and

higher DF.

The difference in G’ of TOCN nanofibrils with 350 µmol/g, when varying

DF, was negligible due to weak repulsive forces from the low charge density.

The storage modulus G’ of TOCN with high charge density, 650 and 1550

µmol/g, decreased as DF increased from 6 to 9. It is in agreement with the

literature that G’ decreases when increasing the degree of homogenization

due to reduced interfibrillar interactions between individually dispersed

nanofibrils126. The dispersion and the morphology of the individual

nanofibrils were analyzed by atomic force microscope and are presented in

Paper IV. The average widths of TOCN were 2.5 nm, which means that the

fibrillation of fibers to individual fibrils was already accomplished by

treating the pulp for 6 times in the microfluidizer. There was no difference

of G’ when DF increased from 9 to 12, regardless of the charge density. The

dispersibility of the nanofibrils was similarly good in DF 6 and 9, but fiber

lengths got shorter when increasing DF. It indicates that TOCN treated at

a too high degree of fibrillation will be mechanically cut, consequently have

low aspect ratio.

Such viscoelastic behavior of TOCN and their dispersibility in aqueous

suspension will influence the properties of a hydrogel or a composite

thereof. For example, TOCN1550 treated at DF6, which has high surface

charge density and low DF, showed the best performance as a flexible LIB

electrode binder, which seems attributed to its good dispersibility (Paper

IV).


4.6. Electro-precipitation of CNF hydrogel, composites

thereof, and its applications

4.6.1. Electrochemical precipitation of TOCN Hydrogels

Figure 4. 19. Schematic illustrations of the primary mechanism of pH

responsive sol-gel transition of TOCN induced by water oxidation at an

anode

The most common technique to make a cellulose based hydrogel is a pH

induced sol-gel transition process, typically performed by adding an acidic

solution to a TOCN suspension in a container with a desired shape. In this

study (Paper V), we introduce a pH induced sol-gel transition of TOCN but

triggered by an electrochemical reaction inducing a local pH change at the

surface of an electrode. Figure 4.19 schematically describes the

electrochemical precipitation process and changes in the colloidal stability

of TOCN at the surface of an anode.

The gelation of TOCN proceeds as the concentration of protons increases

in the vicinity of the anode. Basically, the electrochemical precipitation of

TOCN can occur at the surface of any conductive material which can

generate protons during an electrochemical reaction, as shown in Figure

4.20A. For example, carbon paper, copper and platinum can be used as

substrates, whereas aluminium and nickel with passive layers cannot. FT-

IR spectra confirm that the sol-gel transition takes place due to protonation

of TOCN. The peak at 1724 cm−1 corresponds to stretching of the

protonated carboxylate groups (Pt and carbon paper in Figure 4.20B). It

indicates that the gelation of TOCN was accomplished only by charge

neutralization due to protonation, without any other components inducing

crosslinking of TOCN. On the contrary, the pristine TOCN freeze-dried


hydrogel (aerogel) has a peak at 1607 cm−1, which corresponds to carbonyl

absorption of COO-Na+ functional groups. In terms of the gel on the copper

substrate, the peak at 1724 cm-1 is much less intense than that at 1607 cm-

1, implying that a smaller portion of the carboxyl groups are protonated in

this case. It is therefore assumed that the gel on copper was formed by two

processes occurring in parallel: 1) pH induced gelation, 2) copper ion

mediated cross linking.

Figure 4. 20. Anodically electro-precipitated TOCN hydrogels on

conductive substrates (A) copper wire, brass wire, platinum plate and

carbon paper (clockwise from the top left), (B) FT-IR spectra of freeze-

dried hydrogels of pristine TOCN (no template) as well as

electrochemically precipitated on different conductive substrates. In (B),

the spectra were shifted along the y-axis for better visibility.

The growth of a TOCN hydrogel at the anode was observed by using a

TOCN sol mixed with a universal pH indicator (red at pH < 3, yellow at pH

4, light green at pH 5-7, dark green at pH 8-9, pink at pH > 10), as shown

in Figure 4.21A. During the sol-gel transition of TOCN along with water

splitting, local pH changes at the electrode surfaces were indicated by color

changes. Red at the anode means the generation of protons forming acidic

conditions around the anode. Pink at the cathode means the production of

hydroxide ions forming alkaline conditions, while the bulk remained

neutral. The colored regions were propagated away from the surface to the

bulk, as the water splitting proceeded (at 15 sec in Figure 4.21A). It is

because of diffusion of protons and hydroxide ions generated from anode

and cathode, respectively. It implies that the gelation of TOCN can be

mainly controlled by diffusion of protons, adjusting the operating variables,

e.g. current, voltage or time. Figure 4.21B exhibits an almost linear growth


of the hydrogel, but the growth rate decreased when the applied charge

density was higher than 30 mC/cm2. Besides, due to flocculation of TOCN

at high charge density, the standard deviation also increased.

The advantage of this electrochemical precipitation is the freedom of the

experimental setup, which is not limited to the shape of the container,

rather directly immobilizing hydrogel on the surface of conductive

substrates with complex-shapes81,85,125.

Figure 4. 21. (A) Photographs of water splitting caused local pH change

induced TOCN sol-gel transition with universal pH indicator (red when pH

<3, pink when pH > 10), as a function of time (1, 5 and 15 sec) on a copper

foil anode. (B) Time dependent thickness growth as a function of current

density (1, 3, and 5 mA/cm2) and time (up to 20 sec) and photographs of

the gels made at the given deposition conditions


4.6.2. Electro-precipitation of composites

Electrochemical deposition of biopolymer-based composite coatings has

been studied using a variety of materials including carbon nanotubes,

metal/metal oxide nanoparticles and enzymes80–82,88,92,125. Such polymers

work as binders, adhering nanomaterials dispersed in the coating medium

on the target substrates. In this respect, TOCN can also be regarded as a

promising coating agent to fabricate hydrogel/aerogel composites,

including functional materials, by a one-pot electro-precipitation

technique.

Figure 4. 22. (A) Mass loading increase and thickness growth (measured

in dried status) of graphite/TOCN composite hydrogel with different

composition, as a function of current densities and time, (B) photographs

of the gels (Graphite 0.95%, TOCN 0.1% sol) made at the given deposition

conditions and (C) TGA results of graphite/TOCN composites varying the

compositions and the deposition conditions

The thickness and the mass loading of the composites (Figure 4.22A)

increase as a function of time and current applied, similarly to the TOCN

based hydrogel in Figure 4.22B. A poor coverage of graphite particles on

the substrate surface was observed when a too low current density (< 3


mA/cm2) was applied for very short time, compared to that of higher values

(>5 mA/cm2) for a longer deposition time. TGA results showed that the

composition of the composites can be controlled by the composition of the

coating medium.

Not only single layered hydrogels and composites, but also multiple

layered hydrogel structures can be built on a target substrate by the same

electro-precipitation methodology. As described in Figure 4.23, the

layers can be deposited on the top of each other, by switching coating

medium. The two-layered hydrogel on the bottom right side in Figure

4.23 was fabricated by applying a 1st layer of graphite/TOCN composite

(20:1) hydrogel on a copper substrate, and consecutively a 2nd layer in a

TOCN coating medium. The graphite hydrogel is clearly observed under

the TOCN 2nd layer. Likewise, a tri-layered composite, made in a similar

way, replacing coating medium and coating a new layer on the top of an

existing hydrogel layer, was demonstrated as well in Paper V.

Figure 4. 23. Schematic description of fabrication of multi-layer hydrogel

composite materials on conductive substrates by electrochemical

precipitation.

A limitation of the technique is a loss in colloidal stability of the coating

medium after each coating trial, due to aggregation of TOCN. Moreover,

coating efficiency of the micro/nanoparticles which can be

electrochemically oxidized, e.g. transition metals, is very restricted.

Nevertheless, it was suitable for coating of carbon materials and oxide


nanoparticles, e.g. silica or aluminum nanoparticles. Particularly, graphite

is an inert material at the given experimental conditions, not easily

oxidized. Another limitation is the necessity of an electronically conducting

substrate. In addition, the coating medium is preferably an aqueous

suspension, containing low solids content, diluted from a nanocellulose gel,

which has the limitation that it cannot be directly tuned from gel status and

requires large volume for storage as well as energy for preparation.

4.6.3. Application I: Fabrication of electrode for LIB

A fast and simple preparation of electrochemically deposited TOCN-

graphite composites gives an inspiration to its application as graphite

electrodes in LIB (Figure 4.24A). As shown in Figure 4.24B, the

composite has a microstructure suitable as LIB electrode - a porous

structure and active materials coated by binders. Chronopotentiogram of

the graphite negative electrode in LIB are displayed in Figure 4.24C,

which has very similar voltage profiles to that of typical graphite electrodes,

with the plateaus below 0.2 V. The specific capacity was slightly lower than

the theoretical value for graphite (372 mAh/g) and constantly faded

(Figure 4.24D). Further optimization could improve the electrochemical

performance of the electrode.

State-of-the-art electrode fabrication processes for lithium ion batteries

need hazardous solvents, e.g. NMP and fluorine containing plastic, e.g.

PVDF. Compared to that, this technique might be an appealing

environmentally friendly way of producing electrodes for different kinds of

batteries.

Figure 4. 24. Images of graphite electrode coated on the PEI-coated

copper electrode, (B) microscopic image of the graphite, (C)

chronopotentiogram of graphite electrode and (D) galvanostatic cycling

test of the cell at 0.1 C


4.6.4. Application II: Selective membrane for HER

With increasing demands for clean energy, hydrogen has been considered

as a promising energy carrier for renewable energy technologies, as a

possible replacement of fossil fuels127. Photocatalytic water splitting is one

promising future hydrogen production technique. To improve its efficiency,

the backward reaction of recombination of oxygen and hydrogen to water

on the cocatalyst surface, must be circumvented128. Here as a proof-of-

concept study, we investigated the hydrogen evolution reaction (HER)

selectivity in the absence/presence of oxygen on platinum (a typical co-

catalyst in photocatalysis) electrode coated with a selective hydrogel.

Figure 4. 25. Schematic illustrations of HER with/without ORR (oxygen

reduction reaction) protective layer

Figure 4.25 is a schematic description of HER with (left) and without

(right) a barrier layer against ORR, which is an unwanted side reaction in

this case. To prevent such backward reaction, HER selective protective

layers have been studied, mostly composed of an amorphous oxide layer

including chromium oxide129, titanium dioxide130, silicon dioxide96 and

vanadium oxide97. Other than those, biopolymer-based hydrogels may

possibly be used as ORR protective layers, if permeable to water and

smaller ions but blocking the transport of oxygen.

Electrochemically coated TOCN and TOCN-Triton X-100 hydrogels on the

surface of Pt rotating discs were prepared to evaluate their performance as

selective layers during HER. 0.01 wt % of Triton X-100 surfactant was

added to the 0.1% TOCN deposition solution, to enhance the stability of the

gels.


Figure 4. 26. Electrochemical measurements of gel coated Pt electrodes

in O2 saturated NaClO4 solution (pH=4) at rotating speed of 1000 rpm, ,

the magnified view of the green region (j=-2.5 – 0.5 mA/cm2) as shown in

the right side (scan rate 10 mV/s) (A) TOCN-Triton X-100 hydrogel, (B)

TOCN- hydrogel, at 2000 rpm (C) TOCN-Triton X-100 hydrogel and (D)

HER measurement of gel coated Pt

Figures 4.26A and B show the electrochemical measurements carried

out at a rotation rate of 1000 rpm. The current plateaus of the gel coated

electrodes observed between -0.2 V and -1 V are related to the diffusion

limited ORR. In comparison with bare platinum, both TOCN and TOCN-

Triton X-100 hydrogel coated Pt electrodes showed ≈60% reduced ORR

current. While ORR was being prevented, HER could take place on the gel

coated electrodes at potentials below -0.8 V. The Triton X-100 containing

hydrogel had slightly better ORR blocking (67%), which indicates that the

surfactant contributes to better coating of the Pt electrode and/or further

blocking the O2 transport. An oxidation peak appeared on the backward

anodic scan at -0.8 V and seems attributed to the back-oxidation of H2

trapped in the hydrogel. This indicates that HER takes place at the

interface between the Pt-RDE and the TOCN gel, hence water can

penetrate the gel, whereas O2 cannot and some H2 is also being trapped.


The short term stability of the layers was evaluated by cyclic

voltammograms recorded before and after 10 min of polarization at -5

mA/cm2 (Paper V and Figure 4.26A-B).

At the increased rotation rate of 2000 rpm or at higher current density (-

10 mA/cm2), Triton X-100 was a necessary additive to the TOCN gel to

obtain a stable and protective layer (Figure 4.26C and D). On the

contrary, the TOCN hydrogel without Triton X-100 was destroyed. TOCN

Triton X-100 hydrogels blocked 80% of the ORR reaction at 2000 rpm (at

potential -0.74 V), and still 75% after galvanostatic HER (Figure 4.26C

and D). We believe that the Triton X-100 contributes to improved

plasticity of the hydrogel, with higher tolerance against shear forces and

the through transport of hydrogen.

Figure 4. 27. Electrochemical measurements of performance of Pt

electrode in the presence/absence of the TOCN hydrogel surface coating in

the range of (A) hydrogen adsorption evaluating ECSA, (B) hydrogen

oxidation reaction and (C) K3Fe(CN)6 reduction reaction

To better understand the effect of hydrogels on HER, which selectively

block the transport of gas, supplementary experiments were carried out, as

shown in Figures 4.27A-C. The difference in ECSA between a Pt electrode

in the presence/absence of the TOCN hydrogel surface coating (Figure

4.27 A) was negligible. It indicates that the hydrogel is physically bound

to the electrode surface, without blocking active sites. The hydrogel can

suppress the hydrogen oxidation reaction (HOR) as well as ferricyanide

reduction. During the hydrogen oxidation (Figure 4.27B), 82% of HOR

was suppressed. The limiting current density of ferricyanide reduction on

bare Pt was j ≈ -4.4 mA/cm2, which is 3 times higher than for gel coated Pt,

j ≈ -1.5 mA/cm2 (67% suppressed, Figure 4.27C). It is assumed that the

hydrogel restricts the accessibility of molecules from the bulk electrolyte,

working as a barrier.

Conclusions and outlook- 55

5. Conclusions and outlook

5.1. CNF based separators with different charge density

The primary goal of this work is to evaluate the feasibility of wood-based

cellulose in LIB and look for further applications in electrochemical energy

devices. The chemically modified wood-based TOCN-materials have a

good potential as an eco-friendly alternative to conventional fossil fuel-

derived separators. Raw materials for separators were produced by

chemical treatment and consecutive mechanical treatment of cellulose

pulp at relatively mild condition compared to previous studies. Results

from electrochemical measurements and gas analyses using in-operando

mass spectrometry ensure the electrochemical stability of nanocellulose

based separators. The separators show good performance with high

stability and rate capability, even at fast charging conditions. Vinylene

carbonate significantly reduces the gas evolved in LIB, hence it is

necessitated when cellulose or other hygroscopic materials produced by

aqueous processes are applied for LIB.


assembly

The major advantage of a separator-electrode assembly is that the current

collector reinforces the mechanical strength of the separator, so it

alleviates the concern about low tensile strength of porous materials. In

this work, we demonstrate SEAs which consist of wood-based cellulose

nanofibers, fabricated by a facile and eco-friendly process. Spray-coating

of CNF dispersed in IPA enables fast fabrication of SEA without additional

pore-forming processes in a water-free condition. Such fabrication of SEA

can be considered as a highly promising technique to acquire thinner

separator, achieving low resistivity and better electrochemical

performance of LIB. NMC SEA shows excellent electrochemical stability,

especially at fast charging/discharging rate, not only in a normal pouch cell,

but also in a proof-of-concept prismatic cell, high density batteries and

lithium metal battery as well.


5.3. Electro-precipitation of TOCN hydrogel,

composites thereof, and their applications

TOCN as a pH-responsive biopolymer undergoes sol−gel transition at

acidic conditions (pH < 4), when an electrochemical reaction generates a

local pH decrease at the electrode surface. The hydrogel can be coated on

any electronically conducting material, even with a complex 3D shape,

during aqueous electrolysis. Furthermore, micro- or nanosized particles

containing TOCN hydrogel can be coated on the electrode surface to a

single or multiple layered composite. This simple and facile electrocoating

technique can be subject to various applications, such as coatings making

electrodes selective for the hydrogen evolution reaction, as well as a new

eco-friendly aqueous-based synthesis of Li-ion battery electrodes.

5.4. Outlook

In this thesis, we aim to fabricate wood-based lithium ion battery

separators capable of fast charging and with high electrochemical stability.

Despite successful development of the separators, it is still challenging to

fulfill other industrial needs, e.g. mechanical strength. Therefore, in order

to challenge such barrier, further mechanical treatment or chemical

modification can be implemented. The key is that the mechanical strength

of the porous membrane should be enhanced with little compromising the

electrochemical performance and the rate capability as separator in LIB.

Moreover, chemical modification of cellulose can open the possibility of

development of functionalized separators. In addition, fabrication of CNF-

composite separators or gel polymer electrolytes, blended with

reinforcement materials, can be tested.

We have speculated on the cause of cell failure related to the presence of

sodium counterions. In this respect, a more direct analysis can be carried

out for a better understanding of effects of sodium and lithium ions co-

dissolved in the full cell. Possibly, further surface characterization using

XPS or other analytical techniques can be conducted. Since it is assumed

that sodium deposition on the graphite surface is the major cause for

failure, replacing the negative electrode to LTO or a lithium metal electrode

is likely to attenuate the effect of sodium ions in LIB. The wood-based

separators presented in this thesis cannot be used with lithium metal

batteries, due to some deleterious side reactions, possibly between water


and lithium. Nonetheless, this study also found out that hierarchical

structure of cellulose is beneficial to acquire cellulose with different

dimensions, capable of diverse applications. For instance, combined with

characteristics of microfibrillated fibers as a fast charging separator and

TOCN nanofibrils as binder, multi layered separators with functional

layers e.g. ceramic or conductive carbon coatings, can be possibly

fabricated.

As a next step, compatibility of the cellulosed based separator for the next

generation of batteries, e.g. lithium metal battery and sodium ion battery

(SIB) can be evaluated. Besides, applications of sodium carboxylate groups

containing materials for a sodium ion battery is expected to be beneficial

since it can increase the number of mobile sodium in SIB, facilitating

transport of sodium ions.

Other than LIB and H2 production, TOCN hydrogels can be applied in

diverse electrochemical energy devices applications. As the follow-up of

our H2 production study, actual implementation with photocatalyst coated

by TOCN hydrogels can be carefully suggested. In addition to this, we can

find possible applications in other types of electrochemical energy devices,

e.g. fuel cells or electrode fabrications, or carrying out the similar studies

to other polysaccharides, e.g. hemicellulose. Cellulose will have a wider

spectrum of applications in the fields of bioengineering, including tissue

engineering, drug deliveries, scaffolds, wound healings. Here, we have

studied CNF-only hydrogels and CNF-nanoparticle composites,

immobilized on the substrates. In order to move toward hydrogels

available in a diversity of applications, it needs to investigate more about

fabrication of free-standing hydrogels. Moreover, further strategies to

make hydrogels firmer can be conducted, accompanied with rheological

studies, e.g. in-situ or post polymerization electrolytes.

References- 59

7. References (1) Tsiropoulos, I.; Tarvydas, D.; Lebedeva, N. Li-Ion Batteries for Mobility and

Stationary Storage Applications – Scenarios for Costs and Market Growth, EUR 29440 EN; 2018.

(2) Lebedeva, N.; Di Persio, F.; Boon-Brett, L. Lithium Ion Battery Value Chain and Related Opportunities for Europe, EUR 28534 EN; 2017.

(3) Emilsson, E.; Dahllöf, L. Lithium-Ion Vehicle Battery Production, C444, IVL Swedish Environmental Research Institute; 2019.

(4) UN General Assembly. Transforming Our World : The 2030 Agenda for Sustainable Development; 2015.

(5) Chen, W.; Yu, H.; Lee, S. Y.; Wei, T.; Li, J.; Fan, Z. Nanocellulose: A Promising Nanomaterial for Advanced Electrochemical Energy Storage. Chemical Society Reviews. Royal Society of Chemistry April 21, 2018, pp 2837–2872.

(6) Nirmale, T. C.; Kale, B. B.; Varma, A. J. A Review on Cellulose and Lignin Based Binders and Electrodes: Small Steps towards a Sustainable Lithium Ion Battery. International Journal of Biological Macromolecules. Elsevier B.V. October 1, 2017, pp 1032–1043.

(7) Lawton, J. A.; Lawton, S. L.; Rubin, M. K.; Fornes, V.; Pergher, S. B.; Maesen, T. L. M.; Buglass, J. G.; Tsapatsis, M.; Cai, R.; Yan, Y. S.; et al. A Major Constituent of Brown Algae for Use in High-Capacity Li-Ion Batteries. 2011, No. October, 75–80.

(8) Bresser, D.; Buchholz, D.; Moretti, A.; Varzi, A.; Passerini, S. Alternative Binders for Sustainable Electrochemical Energy Storage-the Transition to Aqueous Electrode Processing and Bio-Derived Polymers. Energy Environ. Sci. 2018, 11, 3096–3127.

(9) De France, K. J.; Hoare, T.; Cranston, E. D. Review of Hydrogels and Aerogels Containing Nanocellulose. Chem. Mater. 2017, 29, 4609–4631.

(10) Way, A. E.; Hsu, L.; Shanmuganathan, K.; Weder, C.; Rowan, S. J. PH-Responsive Cellulose Nanocrystal Gels and Nanocomposites. ACS Macro Lett. 2012, 1, 1001–1006.

(11) Whittingham, M. S. Lithium Batteries and Cathode Materials. Chem. Rev. 2004, 104, 4271–4302.

(12) Yoshio, M.; Brodd, R. J.; Akiya, K. Lithium-Ion Batteries; Kozawa, M. Y. • R. J. B. • A., Ed.; Springer Science+Business Media, LLC 2009, 2009.

(13) Buqa, H.; Holzapfel, M.; Krumeich, F.; Veit, C.; Novák, P. Study of Styrene Butadiene Rubber and Sodium Methyl Cellulose as Binder for Negative Electrodes in Lithium-Ion Batteries. J. Power Sources 2006, 161, 617–622.

(14) Loveridge, M. J.; Remy, G.; Kourra, N.; Genieser, R.; Barai, A.; Lain, M. J.; Guo, Y.; Amor-Segan, M.; Williams, M. A.; Amietszajew, T.; et al. Looking Deeper into the Galaxy (Note 7). Batteries 2018, 4, 1–11.

(15) Orendorff, C. J. The Role of Separators in Lithium-Ion Cell Safety. Electrochem. Soc. Interface 2012, 21, 61–65.

(16) Spotnitz, R. Separators for Lithium-Ion Batteries. In Handbook of Battery Materials; Claus Danie and Jurgen O. Besenhard, Ed.; Wiley-VCH verlag GmbH&Co. KGaA., 2011; pp 693–717.

(17) Waqas, M.; Ali, S.; Feng, C.; Chen, D.; Han, J.; He, W. Recent Development in Separators for High‐Temperature Lithium‐Ion Batteries. Small 2019, 1901689, 1901689.

(18) Solef® PVDF for Li-Ion Batteries | Solvay https://www.solvay.com/en/brands/solef-pvdf/solef-pvdf-li-ion-batteries (accessed Jan 2, 2020).

(19) Nestler, T.; Schmid, R.; Münchgesang, W.; Bazhenov, V.; Schilm, J.; Leisegang, T.; Meyer, D. C. Separators - Technology Review: Ceramic Based Separators for Secondary Batteries; 2014; Vol. 1597.

(20) Henriksen, G. L.; Amine, K.; Liu, J.; Nelson, P. A. Materials Cost Evaluation Report for High-Power Li-Ion HEV Batteries, Report Number ANL-03/5. ARGONNE Natl.

References- 60

Lab. online 2002, 1, 1–70. (21) Berckmans, G.; Messagie, M.; Smekens, J.; Omar, N.; Vanhaverbeke, L.; Mierlo, J.

Van. Cost Projection of State of the Art Lithium-Ion Batteries for Electric Vehicles up to 2030. Energies 2017, 10, 1314.

(22) Patry, G.; Romagny, A.; Martinet, S.; Froelich, D. Cost Modeling of Lithium-Ion Battery Cells for Automotive Applications. Energy Sci. Eng. 2015, 3, 71–82.

(23) Nelson, P. A.; Bloom, K. G.; I Dees, D. W. Modeling the Performance and Cost of Lithium-Ion Batteries for Electric-Drive Vehicles, Report Number ANL-11/32. ARGONNE Natl. Lab. online 2011.

(24) Gaines, L. The Future of Automotive Lithium-Ion Battery Recycling: Charting a Sustainable Course. Sustain. Mater. Technol. 2014, 1, 2–7.

(25) Larcher, D.; Tarascon, J. M. Towards Greener and More Sustainable Batteries for Electrical Energy Storage. Nat. Chem. 2015, 7, 19–29.

(26) Grey, C. P.; Tarascon, J. M. Sustainability and in Situ Monitoring in Battery Development. Nat. Mater. 2016, 16, 45–56.

(27) Cosgrove, D. J. Growth of the Plant Cell Wall. Nature Reviews Molecular Cell Biology. November 2005, pp 850–861.

(28) Dufresne, A. Nanocellulose: From Nature to High Performance Tailored Materials; 2017.

(29) Turbak, A. F.; Snyder, F. W.; Sandberg, K. R. Microfibrillated Cellulose, a New Cellulose Product: Properties, Uses, and Commercial Potential. In J. Appl. Polym. Sci.: Appl. Polym. Symp.;(United States); 1983; Vol. 37.

(30) Henriksson, M.; Henriksson, G.; Berglund, L. A.; Lindström, T. An Environmentally Friendly Method for Enzyme-Assisted Preparation of Microfibrillated Cellulose (MFC) Nanofibers. Eur. Polym. J. 2007, 43, 3434–3441.

(31) Moser, C.; Lindström, M. E.; Henriksson, G. Toward Industrially Feasible Methods for Following the Process of Manufacturing Cellulose Nanofibers. BioResources 2015, 10, 2360–2375.

(32) Moser, C.; Henriksson, G.; Lindström, M. E. Specific Surface Area Increase during Cellulose Nanofiber Manufacturing Related to Energy Input. BioResources 2016, 11, 7124–7132.

(33) Isogai, A.; Saito, T.; Fukuzumi, H. TEMPO-Oxidized Cellulose Nanofibers. Nanoscale 2011, 3, 71–85.

(34) Wågberg, L.; Winter, L.; Ödberg, L.; Lindström, T. On the Charge Stoichiometry upon Adsorption of a Cationic Polyelectrolyte on Cellulosic Materials. Colloids and Surfaces 1987, 27, 163–173.

(35) Lekha, P.; Mtibe, A.; Motaung, T. E.; Andrew, J. E.; Gibril, M. Effect of Mechanical Treatment on Properties of Cellulose Nanofibrils Produced from Bleached Hardwood and Softwood Pulps. 2016, 18, 457–466.

(36) Chun, S.-J.; Choi, E.-S.; Lee, E.-H.; Kim, J. H.; Lee, S.-Y.; Lee, S.-Y. Eco-Friendly Cellulose Nanofiber Paper-Derived Separator Membranes Featuring Tunable Nanoporous Network Channels for Lithium-Ion Batteries. J. Mater. Chem. 2012, 22, 16618–16626.

(37) Zhu, H.; Luo, W.; Ciesielski, P. N.; Fang, Z.; Zhu, J. Y.; Henriksson, G.; Himmel, M. E.; Hu, L. Wood-Derived Materials for Green Electronics, Biological Devices, and Energy Applications. Chem. Rev. 2016, 116, 9305–9374.

(38) Leijonmarck, S.; Cornell, A.; Lindbergh, G.; Wågberg, L. Single-Paper Flexible Li-Ion Battery Cells through a Paper-Making Process Based on Nano-Fibrillated Cellulose. J. Mater. Chem. A 2013, 1, 4671–4677.

(39) Saito, T.; Nishiyama, Y.; Putaux, J. L.; Vignon, M.; Isogai, A. Homogeneous Suspensions of Individualized Microfibrils from TEMPO-Catalyzed Oxidation of Native Cellulose. Biomacromolecules 2006, 7, 1687–1691.

(40) Zolin, L.; Destro, M.; Chaussy, D.; Penazzi, N.; Gerbaldi, C.; Beneventi, D. Aqueous Processing of Paper Separators by Filtration Dewatering: Towards Li-Ion Paper Batteries. J. Mater. Chem. A 2015, 3, 14894–14901.

References- 61

(41) Henriksson, M.; Berglund, L. a.; Isaksson, P.; Lindström, T.; Nishino, T. Cellulose Nanopaper Structures of High Toughness. Biomacromolecules 2008, 9, 1579–1585.

(42) Taheri, H.; Samyn, P. Effect of Homogenization (Microfluidization) Process Parameters in Mechanical Production of Micro- and Nanofibrillated Cellulose on Its Rheological and Morphological Properties. Cellulose 2016, 23.

(43) Lee, J. H.; Lee, S.; Paik, U.; Choi, Y. M. Aqueous Processing of Natural Graphite Particulates for Lithium-Ion Battery Anodes and Their Electrochemical Performance. J. Power Sources 2005, 147, 249–255.

(44) Jabbour, L.; Destro, M.; Gerbaldi, C.; Chaussy, D.; Penazzi, N.; Beneventi, D. Aqueous Processing of Cellulose Based Paper-Anodes for Flexible Li-Ion Batteries. J. Mater. Chem. 2012, 22, 3227.

(45) Kuribayashi, I. Characterization of Composite Cellulosic Separators for Rechargeable Lithium-Ion Batteries. J. Power Sources 1996, 63, 87–91.

(46) Li, M. X.; Wang, X. W.; Yang, Y. Q.; Chang, Z.; Wu, Y. P.; Holze, R. A Dense Cellulose-Based Membrane as a Renewable Host for Gel Polymer Electrolyte of Lithium Ion Batteries. J. Memb. Sci. 2015, 476, 112–118.

(47) Pan, R.; Cheung, O.; Wang, Z.; Tammela, P.; Huo, J.; Lindh, J.; Edström, K.; Strømme, M.; Nyholm, L. Mesoporous Cladophora Cellulose Separators for Lithium-Ion Batteries. J. Power Sources 2016, 321, 185–192.

(48) Leijonmarck, S.; Cornell, A.; Lindbergh, G.; Wågberg, L. Flexible Nano-Paper-Based Positive Electrodes for Li-Ion Batteries-Preparation Process and Properties. Nano Energy 2013, 2, 794–800.

(49) Zhu, Y. S.; Xiao, S. Y.; Li, M. X.; Chang, Z.; Wang, F. X.; Gao, J.; Wu, Y. P. Natural Macromolecule Based Carboxymethyl Cellulose as a Gel Polymer Electrolyte with Adjustable Porosity for Lithium Ion Batteries. J. Power Sources 2015, 288, 368–375.

(50) Xu, D.; Wang, B.; Wang, Q.; Gu, S.; Li, W.; Jin, J.; Chen, C.; Wen, Z. High-Strength Internal Cross-Linking Bacterial Cellulose-Network- Based Gel Polymer Electrolyte for Dendrite-Suppressing and High- Rate Lithium Batteries. ACS Appl. Mater. Interfaces 2018, 10, 17809–17819.

(51) Gwon, H.; Park, K.; Chung, S.-C.; Kim, R.-H.; Kang, J. K.; Ji, S. M.; Kim, N.-J.; Lee, S.; Ku, J.-H.; Do, E. C.; et al. A Safe and Sustainable Bacterial Cellulose Nanofiber Separator for Lithium Rechargeable Batteries. Proc. Natl. Acad. Sci. 2019, 116, 19288–19293.

(52) Li, W.-W.; Sheng, G.-P.; Liu, X.-W.; Yu, H.-Q. Recent Advances in the Separators for Microbial Fuel Cells. Bioresour. Technol. 2011, 102, 244–252.

(53) Hoerpel, G.; Hennige, V.; Hying, C.; Augustin, S. Inorganic Separator-Electrode-Unit for Lithium-Ion Batteries, Method for the Production Thereof and Use Thereof in Lithium Batteries, US 8,163,441 B2,24, Apr, 2012.

(54) Huang, X. Development and Characterization of a Bilayer Separator for Lithium Ion Batteries. J. Power Sources 2011, 196, 8125–8128.

(55) Jung, Y. C.; Kim, S. K.; Kim, M. S.; Lee, J. H.; Han, M. S.; Kim, D. H.; Shin, W. C.; Ue, M.; Kim, D. W. Ceramic Separators Based on Li+-Conducting Inorganic Electrolyte for High-Performance Lithium-Ion Batteries with Enhanced Safety. J. Power Sources 2015, 293, 675–683.

(56) Singh, N.; Galande, C.; Miranda, A.; Mathkar, A.; Gao, W.; Reddy, A. L. M.; Vlad, A.; Ajayan, P. M. Paintable Battery. Sci. Rep. 2012, 2, 6–10.

(57) Choi, K.; Cho, S.; Chun, S.; Yoo, J. T.; Lee, C. K.; Kim, W.; Wu, Q.; Park, S.; Choi, D.; Lee, S.; et al. Heterolayered, One-Dimensional Nanobuilding Block Mat Batteries. 2014, No. 2.

(58) Xiao, W.; Zhao, L.; Gong, Y.; Wang, S.; Liu, J.; Yan, C. Preparation of High Performance Lithium-Ion Batteries with a Separator-Cathode Assembly. RSC Adv. 2015, 5, 34184–34190.

(59) Gong, S.; Jeon, H.; Lee, H.; Ryou, M. H.; Lee, Y. M. Effects of an Integrated Separator/Electrode Assembly on Enhanced Thermal Stability and Rate Capability

References- 62

of Lithium-Ion Batteries. ACS Appl. Mater. Interfaces 2017, 9, 17814–17821. (60) Woo, J. J.; Zhang, Z.; Amine, K. Separator/Electrode Assembly Based on Thermally

Stable Polymer for Safe Lithium-Ion Batteries. Adv. Energy Mater. 2014, 4, 1–4. (61) Mi, W.; Sharma, G.; Dong, X.; Jin, Y.; Lin, Y. S. Electrode-Supported Thin α-

Alumina Separators for Lithium-Ion Batteries. J. Power Sources 2016, 305, 209–216.

(62) Han, Y.; Ye, L.; Boateng, B.; Sun, Q.; Zhen, C.; Chen, N.; Shi, X.; Dickerson, J. H.; Li, X.; He, W. Direct Electrophoretic Deposition of an Ultra-Strong Separator on an Anode in a Surfactant-Free Colloidal System for Lithium Ion Batteries. J. Mater. Chem. A 2019, 7, 1410–1417.

(63) TOSHIBA. New Separator-free Lithium-ion Rechargeable Battery Developed by Toshiba https://www.toshiba.co.jp/about/press/2018_06/pr0401.htm (accessed Jun 4, 2018).

(64) Murray, V.; Hall, D. S.; Dahn, J. R. A Guide to Full Coin Cell Making for Academic Researchers. J. Electrochem. Soc. 2019, 166, A329–A333.

(65) Ramadass, P.; Haran, B.; White, R.; Popov, B. N. Capacity Fade of Sony 18650 Cells Cycled at Elevated Temperatures: Part I. Cycling Performance. J. Power Sources 2002, 112, 606–613.

(66) Harks, P. P. R. M. L.; Mulder, F. M.; Notten, P. H. L. In Situ Methods for Li-Ion Battery Research : A Review of Recent Developments. J. Power Sources 2015, 288, 92–105.

(67) Bernhard, R.; Meini, S.; Gasteiger, H. A. On-Line Electrochemical Mass Spectrometry Investigations on the Gassing Behavior of Li4Ti5O12 Electrodes and Its Origins. J. Electrochem. Soc. 2014, 161.

(68) Nam, K.-W.; Bak, S.-M.; Hu, E.; Yu, X.; Zhou, Y.; Wang, X.; Wu, L.; Zhu, Y.; Chung, K.-Y.; Yang, X.-Q. Combining In Situ Synchrotron X-Ray Diffraction and Absorption Techniques with Transmission Electron Microscopy to Study the Origin of Thermal Instability in Overcharged Cathode Materials for Lithium-Ion Batteries. Adv. Funct. Mater. 2013, 23, 1047–1063.

(69) Novák, P.; Goers, D.; Hardwick, L.; Holzapfel, M.; Scheifele, W.; Ufheil, J.; Würsig, A. DEMS_ Advanced in Situ Characterization Methods Applied to Carbonaceous Materials. J. Power Sources 2005, 146, 15–20.

(70) Metzger, M.; Strehle, B.; Solchenbach, S.; Gasteiger, H. A. Origin of H2 Evolution in LIBs: H2O Reduction vs. Electrolyte Oxidation. J. Electrochem. Soc. 2016, 163, A798–A809.

(71) Bernhard, R.; Metzger, M.; Gasteiger, H. A. Gas Evolution at Graphite Anodes Depending on Electrolyte Water Content and SEI Quality Studied by On-Line Electrochemical Mass Spectrometry. J. Electrochem. Soc. 2015, 162, A1984–A1989.

(72) Pritzl, D.; Solchenbach, S.; Wetjen, M.; Gasteiger, H. A. Analysis of Vinylene Carbonate (VC) as Additive in Graphite/LiNi 0.5 Mn 1.5 O 4 Cells. J. Electrochem. Soc. 2017, 164, A2625–A2635.

(73) Petibon, R.; Xia, J.; Burns, J. C.; Dahn, J. R. Study of the Consumption of Vinylene Carbonate in Li[Ni0.33Mn0.33Co0.33]O2/Graphite Pouch Cells. J. Electrochem. Soc. 2014, 161, A1618–A1624.

(74) Vetter, J.; Holzapfel, M.; Wuersig, A.; Scheifele, W.; Ufheil, J.; Novák, P. DEMS_In Situ Study on CO2 Evolution at Lithium-Ion Battery Cathodes. J. Power Sources 2006, 159, 277–281.

(75) Arora, P.; White, R. E.; Doyle, M. Capacity Fade Mechanisms and Side Reactions in Lithium-Ion Batteries. J. Electrochem. Soc. 1998, 145, 3647–3667.

(76) Sullivan, J. P.; Fenton, K. R.; El, F.; Marquez, G.; Harris, C. T.; Carl, C.; Hudak, N. S.; Jungjohann, K. L.; Kliewer, C. J.; Mccarty, K.; et al. The Science of Battery Degradation,Sandia Report, SAND2015-0132; 2015.

(77) Jung, R.; Metzger, M.; Maglia, F.; Stinner, C.; Gasteiger, H. A. Oxygen Release and Its Effect on the Cycling Stability of LiNi x Mn y Co z O 2 (NMC) Cathode Materials for Li-Ion Batteries. J. Electrochem. Soc. 2017, 164, A1361–A1377.

References- 63

(78) Liu, K.; Liu, Y.; Lin, D.; Pei, A.; Cui, Y. Materials for Lithium-Ion Battery Safety. Sci. Adv. 2018, 4.

(79) Galushkin, N. Е.; Yazvinskaya, N. N.; Galushkin, D. N. Mechanism of Gases Generation during Lithium-Ion Batteries Cycling. J. Electrochem. Soc. 2019, 166, A897–A908.

(80) Suginta, W.; Khunkaewla, P.; Schulte, A. Electrochemical Biosensor Applications of Polysaccharides Chitin and Chitosan. Chem. Rev. 2013, 113, 5458–5479.

(81) Geng, Z.; Wang, X.; Guo, X.; Zhang, Z.; Chen, Y.; Wang, Y. Electrodeposition of Chitosan Based on Coordination with Metal Ions in Situ-Generated by Electrochemical Oxidation. J. Mater. Chem. B 2016, 4, 3331–3338.

(82) Cheong, M.; Zhitomirsky, I. Electrodeposition of Alginic Acid and Composite Films. Colloids Surfaces A Physicochem. Eng. Asp. 2008, 328, 73–78.

(83) Cheng, Y.; Luo, X.; Betz, J.; Payne, G. F.; Bentley, W. E.; Rubloff, G. W. Mechanism of Anodic Electrodeposition of Calcium Alginate. Soft Matter 2011, 7, 5677–5684.

(84) Peng, X.; Liu, Y.; Bentley, W. E.; Payne, G. F. Electrochemical Fabrication of Functional Gelatin-Based Bioelectronic Interface. Biomacromolecules 2016, 17, 558–563.

(85) Wang, Y.; Zhang, Z.; Wang, M.; Guo, C.; Liu, H.; Zeng, H.; Duan, X.; Zhou, Y.; Tang, Z. Direct Electrodeposition of Carboxymethyl Cellulose Based on Coordination Deposition Method. Cellulose 2018, 25, 105–115.

(86) Bressner, J. E.; Marelli, B.; Qin, G.; Klinker, L. E.; Zhang, Y.; Kaplan, D. L.; Omenetto, F. G. Rapid Fabrication of Silk Films with Controlled Architectures via Electrogelation. J. Mater. Chem. B 2014, 2, 4983–4987.

(87) Cheng, Y.; Luo, X.; Betz, J.; Buckhout-White, S.; Bekdash, O.; Payne, G. F.; Bentley, W. E.; Rubloff, G. W. In Situ Quantitative Visualization and Characterization of Chitosan Electrodeposition with Paired Sidewall Electrodes. Soft Matter 2010, 6, 3177–3183.

(88) Wang, Z.; Zhang, X.; Gu, J.; Yang, H.; Nie, J.; Ma, G. Electrodeposition of Alginate/Chitosan Layer-by-Layer Composite Coatings on Titanium Substrates. Carbohydr. Polym. 2014, 103, 38–45.

(89) Taira, N.; Ino, K.; Kunikata, R.; Shiku, H.; Matsue, T.; Terauchi, M.; Gakumasawa, M.; Suda, A. Local Hydrogel Fabrication Based on Electrodeposition with a Large-Scale Integration (LSI)-Based Amperometric Device. Sensors Actuators B Chem. 2018, 277, 95–101.

(90) Sato, Y.; Kusaka, Y.; Kobayashi, M. Charging and Aggregation Behavior of Cellulose Nanofiber in Aqueous Solution. Langmuir 2017, 33, 12660–12669.

(91) Rizwan, M.; Yahya, R.; Hassan, A.; Yar, M.; Azzahari, A. D.; Selvanathan, V.; Sonsudin, F.; Abouloula, C. N. PH Sensitive Hydrogels in Drug Delivery: Brief History, Properties, Swelling, and Release Mechanism, Material Selection and Applications. Polymers (Basel). 2017, 9, 1–37.

(92) Wu, L. Q.; Lee, K.; Wang, X.; English, D. S.; Losert, W.; Payne, G. F. Chitosan-Mediated and Spatially Selective Electrodeposition of Nanoscale Particles. Langmuir 2005, 21, 3641–3646.

(93) Saito, T.; Kimura, S.; Nishiyama, Y.; Isogai, A. Cellulose Nanofibers Prepared by TEMPO-Mediated Oxidation of Native Cellulose. Biomacromolecules 2007, 8, 2485–2491.

(94) Zamburlini, E.; Jensen, K. D.; Stephens, I. E. L.; Chorkendorff, I.; Escudero-Escribano, M. Benchmarking Pt and Pt-Lanthanide Sputtered Thin Films for Oxygen Electroreduction: Fabrication and Rotating Disk Electrode Measurements. Electrochim. Acta 2017, 247, 708–721.

(95) Mayrhofer, K. J. J.; Strmcnik, D.; Blizanac, B. B.; Stamenkovic, V.; Arenz, M.; Markovic, N. M. Measurement of Oxygen Reduction Activities via the Rotating Disc Electrode Method: From Pt Model Surfaces to Carbon-Supported High Surface Area Catalysts. Electrochim. Acta 2008, 53, 3181–3188.

(96) Bau, J. A.; Takanabe, K. Ultrathin Microporous SiO2 Membranes Photodeposited

References- 64

on Hydrogen Evolving Catalysts Enabling Overall Water Splitting. ACS Catal. 2017, 7, 7931–7940.

(97) Endrődi, B.; Smulders, V.; Simic, N.; Wildlock, M.; Mul, G.; Mei, B.; Cornell, A. In Situ Formed Vanadium-Oxide Cathode Coatings for Selective Hydrogen Production. Appl. Catal. B Environ. 2018, 244, 233–239.

(98) Schneider, C. A.; Rasband, W. S.; Eliceiri, K. W. NIH Image to ImageJ: 25 Years of Image Analysis. Nat. Methods 2012, 9, 671–675.

(99) Ding, G.; Qin, B.; Liu, Z.; Zhang, J.; Zhang, B.; Hu, P.; Zhang, C.; Xu, G.; Yao, J.; Cui, G. A Polyborate Coated Cellulose Composite Separator for High Performance Lithium Ion Batteries. J. Electrochem. Soc. 2015, 162, 834–838.

(100) Jabbour, L.; Bongiovanni, R.; Chaussy, D.; Gerbaldi, C.; Beneventi, D. Cellulose-Based Li-Ion Batteries: A Review. Cellulose 2013, 20, 1523–1545.

(101) Berthold, J.; Desbrières, J.; Rinaudo, M.; Salmén, L. Types of Adsorbed Water in Relation to the Ionic Groups and Their Counter-Ions for Some Cellulose Derivatives. Polymer (Guildf). 1994, 35, 5729–5736.

(102) Fall, A. B.; Lindström, S. B.; Sundman, O.; Ödberg, L.; Wågberg, L. Colloidal Stability of Aqueous Nanofibrillated Cellulose Dispersions. Langmuir 2011, 27, 11332–11338.

(103) Zheng, H.; Li, J.; Song, X.; Liu, G.; Battaglia, V. S. A Comprehensive Understanding of Electrode Thickness Effects on the Electrochemical Performances of Li-Ion Battery Cathodes. Electrochim. Acta 2012, 71, 258–265.

(104) Gallagher, K. G.; Trask, S. E.; Bauer, C.; Woehrle, T.; Lux, S. F.; Tschech, M.; Lamp, P.; Polzin, B. J.; Ha, S.; Long, B.; et al. Optimizing Areal Capacities through Understanding the Limitations of Lithium-Ion Electrodes. J. Electrochem. Soc. 2016, 163, A138–A149.

(105) Xu, Q.; Kong, Q.; Liu, Z.; Zhang, J.; Wang, X.; Liu, R.; Yue, Li.; Cui, G. Polydopamine-Coated Cellulose Microfibrillated Membrane as High Performance Lithium-Ion Battery Separator. RSC Adv. 2014, 4, 7845–7850.

(106) Šauperl, O.; Stana-Kleinschek, K. Differences Between Cotton and Viscose Fibers Crosslinked with BTCA. Text. Res. J. 2010, 80, 383–392.

(107) Qiu, L.; Shao, Z.; Yang, M.; Wang, W.; Wang, F.; Xie, L.; Lv, S.; Zhang, Y. Electrospun Carboxymethyl Cellulose Acetate Butyrate (CMCAB) Nanofiber for High Rate Lithium-Ion Battery. Carbohydr. Polym. 2013.

(108) Salas, C.; Nypelö, T.; Rodriguez-Abreu, C.; Carrillo, C.; Rojas, O. J. Nanocellulose Properties and Applications in Colloids and Interfaces. Curr. Opin. Colloid Interface Sci. 2014, 19, 383–396.

(109) Lundahl, M. J.; Cunha, A. G.; Rojo, E.; Papageorgiou, A. C.; Rautkari, L.; Arboleda, J. C.; Rojas, O. J. Strength and Water Interactions of Cellulose i Filaments Wet-Spun from Cellulose Nanofibril Hydrogels. Sci. Rep. 2016, 6, 1–13.

(110) Toivonen, M. S.; Kaskela, A.; Rojas, O. J.; Kauppinen, E. I.; Ikkala, O. Ambient-Dried Cellulose Nanofibril Aerogel Membranes with High Tensile Strength and Their Use for Aerosol Collection and Templates for Transparent, Flexible Devices. Adv. Funct. Mater. 2015, 25, 6618–6626.

(111) Krol, L. F.; Beneventi, D.; Alloin, F.; Chaussy, D. Microfibrillated Cellulose-SiO2 Composite Nanopapers Produced by Spray Deposition. J. Mater. Sci. 2015, 50, 4095–4103.

(112) Schweidler, S.; De Biasi, L.; Schiele, A.; Hartmann, P.; Brezesinski, T.; Janek, J. Volume Changes of Graphite Anodes Revisited: A Combined Operando X-Ray Diffraction and in Situ Pressure Analysis Study. J. Phys. Chem. C 2018, 122, 8829–8835.

(113) Belharouak, I.; Koenig, G. M.; Amine, K. Electrochemistry and Safety of Li4Ti5O12 and Graphite Anodes Paired with LiMn2O4 for Hybrid Electric Vehicle Li-Ion Battery Applications. J. Power Sources 2011, 196, 10344–10350.

(114) Kuppan, S.; Xu, Y.; Liu, Y.; Chen, G. Phase Transformation Mechanism in Lithium Manganese Nickel Oxide Revealed by Single-Crystal Hard X-Ray Microscopy. Nat.

References- 65

Commun. 2017, 8, 1–10. (115) Koerver, R.; Zhang, W.; De Biasi, L.; Schweidler, S.; Kondrakov, A. O.; Kolling, S.;

Brezesinski, T.; Hartmann, P.; Zeier, W. G.; Janek, J. Chemo-Mechanical Expansion of Lithium Electrode Materials-on the Route to Mechanically Optimized All-Solid-State Batteries. Energy Environ. Sci. 2018, 11, 2142–2158.

(116) Berthold, J.; Olsson, R. J. O.; Salmén, L. Water Sorption to Hydroxyl and Carboxylic Acid Groups in Carboxymethylcellulose (CMC) Studied with NIR-Spectroscopy. Cellulose 1998, 5, 281–298.

(117) Ota, H.; Sakata, Y.; Inoue, A.; Yamaguchi, S. Analysis of Vinylene Carbonate Derived SEI Layers on Graphite Anode. J. Electrochem. Soc. 2004, 151, A1659.

(118) Xie, L.; Zhao, L.; Wan, J.; Shao, Z.; Wang, F.; Lv, S. The Electrochemical Performance of Carboxymethyl Cellulose Lithium as a Binding Material for Anthraquinone Cathodes in Lithium Batteries. J. Electrochem. Soc. 2012, 159, 499–505.

(119) Komaba, S.; Yabuuchi, N.; Ozeki, T.; Okushi, K.; Yui, H.; Konno, K.; Katayama, Y.; Miura, T. Functional Binders for Reversible Lithium Intercalation into Graphite in Propylene Carbonate and Ionic Liquid Media. J. Power Sources 2010, 195, 6069–6074.

(120) Jeschull, F.; Maibach, J.; Memm, M.; Brandell, D. Solid Electrolyte Interphase (SEI) of Water-Processed Graphite Electrodes Examined in a 65 MAh Full Cell Con Fi Guration. Appl. energy Mater. 2018, 1, 5176–5188.

(121) Strmcnik, D.; Castelli, I. E.; Connell, J. G.; Haering, D.; Zorko, M.; Martins, P.; Lopes, P. P.; Genorio, B.; Østergaard, T.; Gasteiger, H. A.; et al. Electrocatalytic Transformation of HF Impurity to H2 and LiF in Lithium-Ion Batteries. Nat. Catal. 2018, 1, 255–262.

(122) Moriwake, H. Why Is Sodium-Intercalated Graphite Unstable? 2017, 36550–36554. (123) Kondo, Y.; Fukutsuka, T.; Miyazaki, K.; Miyahara, Y.; Abe, T. Investigation of

Electrochemical Sodium-Ion Intercalation Behavior into Graphite-Based Electrodes. J. Electrochem. Soc. 2019, 166, A5323–A5327.

(124) Komaba, S.; Itabashi, T.; Watanabe, M.; Groult, H.; Kumagai, N. Electrochemistry of Graphite in Li and Na Salt Codissolving Electrolyte for Rechargeable Batteries. J. Electrochem. Soc. 2007, 154, A322.

(125) Gray, K. M.; Liba, B. D.; Wang, Y.; Cheng, Y.; Rubloff, G. W.; Bentley, W. E.; Montembault, A.; Royaud, I.; David, L.; Payne, G. F. Electrodeposition of a Biopolymeric Hydrogel: Potential for One-Step Protein Electroaddressing. Biomacromolecules 2012, 13, 1181–1189.

(126) Nechyporchuk, O.; Belgacem, M. N.; Pignon, F. Current Progress in Rheology of Cellulose Nanofibril Suspensions. Biomacromolecules 2016, 17, 2311–2320.

(127) Crabtree, G. W.; Dresselhaus, M. S.; Buchanan, M. V. The Hydrogen Economy. Phys. Today 2004, No. December, 39–45.

(128) Qureshi, M.; Shinagawa, T.; Tsiapis, N.; Takanabe, K. Exclusive Hydrogen Generation by Electrocatalysts Coated with an Amorphous Chromium-Based Layer Achieving Efficient Overall Water Splitting. ACS Sustain. Chem. Eng. 2017, 5, 8079–8088.

(129) Maeda, K.; Teramura, K.; Lu, D.; Saito, N.; Inoue, Y.; Domen, K. Noble-Metal/Cr2O3 Core/Shell Nanoparticles as a Cocatalyst for Photocatalytic Overall Water Splitting. Angew. Chemie - Int. Ed. 2006, 45, 7806–7809.

(130) Eisenberg, D.; Ahn, H. S.; Bard, A. J. Enhanced Photoelectrochemical Water Oxidation on Bismuth Vanadate by Electrodeposition of Amorphous Titanium Dioxide. J. Am. Chem. Soc. 2014, 136, 14011–14014.

Documents

Wood-Based Nanocellulose In Lithium Ion Batteries and ...1393422/FULLTEXT02.pdf · Development of electrode-integrated cellulose separators was enabled by a non-aqueous drying method