One -Shot TEMPO -Periodate Oxidation of Native Cellulose [611769]

Carbohydrate Polymers
Manuscript Draft

Manuscript Number: CARBPOL -D-19-02975

Title: One -Shot TEMPO -Periodate Oxidation of Native Cellulose

Article Type: Research Paper

Keywords: Carboxylated Cellulose, TEMPO, Periodate, degree of
substitution (DS), solubility, Degree of polymerization (DP)

Abstract: TEMPO/NaClO/NaBr and sodium periodate were combined in a one –
shot reaction to oxidise cellulose from bleached pulp. Ox idation of
cellulose forms two fractions: a highly -carboxylated water -insoluble (up
to 1.9 mmol COO -/g) and a water -soluble fraction (up to 4 mmol COO -/g).
Results show that these regioselective oxidants work in synergy to
produce fully -oxidised 2,3,6 -tricarboxycellulose. Increasing the
periodate concentration results in fibrillation and extensive
depolymerisation of the pulp cellulose as more residual aldehyde groups
participate in the depolymerisation process. X -ray diffraction (XRD)
reveals that the add ition of periodate increases cellulose crystallinity,
retains crystal size but slightly alters the XRD pattern. The degree of
substitution (DS), which governs the solubility of carboxylated
cellulose, can be controlled by varying the periodate concentratio n.
Combining TEMPO/NaBr/NaClO with sodium periodate simultaneously in a one –
shot reaction produces low -cost cellulose with controlled level of
carboxylation and unique properties.

One-Shot TEMPO -Periodate Oxidation of Native Cellulose
David Mendoza1, Christine Browne1, Vikram Raghuwanshi1, George Simon2, Gil Garnier1*
1Bioresource Processing Research Institute of Australia (BioPRIA),
Department of Chemical Engineering, Monash University, Clayton, VIC 3800, Australia
2Department of Materials Science and Engineering, Monash University, Clayton, VIC 3800, Australia

Corresponding author: [anonimizat]

Highlights:
 Oxidation of native cellulose in a one -shot TEMPO -periodate oxidation reaction forms
highly -carboxylated water -insoluble and water -soluble fractions.

 Combining TEMPO/NaClO/NaBr simultaneously can produce a fully -oxidised 2,3,6 –
tricarboxycellulose.

 The concentratio n of periodate in the one -shot reaction plays a crucial role in the
chemical and physical properties of the oxidised cellulose derivatives .

 The degree of substitution (DS) can be manipulated by periodate concentration in the
one-shot reaction. DS govern s the solubility of carboxylated cellulose.
Highlights (for review)

1
One-Shot TEMPO -Periodate Oxidation of Native Cellulose 1
David Mendoza1, Christine Browne1, Vikram Raghuwanshi1, George Simon2, Gil Garnier1* 2
1Bioresource Processing Research Institute of Australia (BioPRIA), 3
Department of Chemical Engineering, Monash University, Clayton, VIC 3800, Australia 4
2Department of Materials Science and Engineering, Monash University, Clayton, VIC 3800, Australia 5
6
Corresponding author: [anonimizat] 7
8
ABSTRACT 9
TEMPO /NaClO/Na Br and sodium periodate were combined in a one -shot reaction to oxidise 10
cellulose from bleached pulp. Oxidation of cellulose forms two fractions: a highly – 11
carboxylated water -insoluble (up to 1.9 mmol COO -/g) and a water -soluble fraction (up to 4 12
mmol COO-/g). Results show that these regioselective oxidants work in synergy to produce 13
fully-oxidised 2,3,6 -tricarboxycellulose. Increasing the periodate concentration results in 14
fibrillation and extensive depolymerisation of the pulp cellulose as more residual aldehyde 15
groups participate in the depolymerisation process. X-ray diffraction (XRD) reveals that the 16
addition of periodate increases cellulose crystallinity , retains crystal size but slightly alters 17
the XRD pattern. The degree of substitution (DS), which governs the solubility of 18
carboxylated cellulose, can be controlled by varying the periodate concentration. Combining 19
TEMPO/NaBr/NaClO with sodium periodate simultaneously in a one -shot reaction produces 20
low-cost cellulose with controlled level of carboxyl ation and unique properties. 21
22
Keywords: Carboxylated Cellulose, TEMPO, Periodate, degree of substitution (DS), 23
solubility, Degree of polymerization (DP) 24
25
Graphical Abstract: 26
27
28
29
30
*Manuscript
Click here to view linked References

2
1. INTRODUCTION 31
Cellulose is an ideal building block to design functional materials, owing to its 32
abundance, renewability, excellent mechanical properties and tunable surface chemistry. 33
The three reactive hydroxyl groups of the cellulose anhydroglucose units are effective 34
reactive sites for a myriad of functional groups or molecules leading to a wide variety of 35
applications (Klemm, Heublein, Fink, & Bohn, 2005; Klemm, Philpp, Heinze, Heinze, & 36
Wagen knecht, 1998) . 37
Oxidation is a common reaction to prepare cellulose for m edicinal, pharmaceutical 38
and industrial applications. Selecting the type of cellulose, reagents, and condition enables 39
to produce different oxidation products varying in properties (Klemm et al., 1998) . The 40
resulting carboxyl moieties form intermediate sites for further functionalisation or grafting 41
biomolecules (Curvello, Raghuwanshi, & Garnier, 2019) . Oxidation of cellulose is also used 42
to impart surfa ce charge density to reduce the energy consumption required to delaminate 43
fibres into nanocellulose fibrils (Akira Isogai & Bergström, 2 018; A. Isogai, Saito, & 44
Fukuzumi, 2011) . There are many reports of functional products from cellulose oxidation 45
(Curvello, Mendoza, et al., 2019; A. Isogai et al., 2011; Mendoza et al., 2019; Uribe, 46
Chiromito, Carvalho, Arenal, & Tarpani, 2017; Yang et al., 2017) . Among all oxidation 47
protocols, TEMPO -mediated and periodate oxidation reactions are considered to be the 48
most practical, owing to their regioselectivity, efficiency and “green” conditions. 49
2,2,6,6 -Tetramethylpiperidine -1-oxyl-mediated (or TEMPO -mediated ) oxidation of 50
cellulose (Figure 1a) selectively converts cellulose primary hydroxymethyl groups into 51
carboxylic functionalities. TEMPO is a water -soluble and stable radical catalyst commonly 52
combined with NaClO and NaBr as oxidants. De Nooy (de Nooy, Besemer, & van Bekkum, 53
1995) first reported the TEMPO -mediated oxidation of water -soluble starch, pullulan, and 54
dextran to sodium C6 -carboxy late groups in water at pH 10 and room temperature. It is 55
however Isogai et al. that methodologically developed and characterized this oxidation 56
protocol for cellulose, preparing nanocellulose products (A. Isogai, 2018; A. Isogai & Kato, 57
1998; A. Isogai et al., 2011; A. Isogai & Shibata, 2 001; T. Isogai, 2017; T. Isogai, Saito, & 58
Isogai, 2010; T. Isogai, Yanagisawa, & Isogai, 2009; Mendoza, Batchelor, Tabor, & Garnier, 59
2018; Mendoza, Gunawardhana, Batchelor, & Gamier, 2018; Mendoza et al., 2019; Saito et 60
al., 2009) . 61
Periodic acid or periodate is commonly used to selectively oxidise vicinal hydroxyl 62
groups in cellulose into aldehyde groups, whilst simultaneously breaking the C2 –C3 linkage 63
(Figure 1b) (H. Liimatainen, Visanko, Sirv io, Hormi, & Niinimaki, 2012; Nevell, 1957; Sirvio, 64
Hyvakko, Liimatainen, Niinimaki, & Hormi, 2011) . The resulting dialdehydes can be oxidised 65

3
further into carboxylic groups to introduce imine bonds between cellulose and amine groups , 66
for example. Periodate oxidation was investigated for biodegradable and biocompatible 67
beads (Lindh, Carlsson, Str ømme, & Mihranyan, 2014; Ruan, Strømme, & Lindh, 2018) . 68
Recently, these two selective oxidation protocols were combined in a one -shot 69
reaction to oxidise microcrystalline cellulose (Figure 1c) (Coseri et al., 2015) . Water -soluble 70
2,3,6 -tricarboxycellulose with a carboxylate content of 3.11 mmol/g were thus prepared 71
(Coseri et al., 2015) . However, this protocol has never been explored to oxidise native 72
cellulose from bleached Eucalyptus kraft pulp which contains residual hemicellulose and 73
lignin. Further, the yields and properties of the polymers achieved have not been rigorously 74
quantified. In this study, the oxidised products are characterised in terms of mass yield, 75
surface charge density, c hemical structure and degree of polymerisation. Changes in 76
crystallinity, crystal size and fibre morphology are investigated. The synergy of 77
TEMPO/NaClO/NaBr and sodium periodate oxidation systems are studied. The effects of 78
periodate concentration on the properties of oxidised cellulose are assessed. By varying the 79
periodate concentration in the one -shot reaction, the degree of substitution (DS) can be 80
manipulated. The degree of substitution (DS) is defined as th e number of hydroxyl groups 81
on an anhydroglucose monomeric unit that has been chemically subsituted; it ranges from 0 82
(pure cellulose) to 3 (fully functionalized). We highlight in this study that DS governs the 83
solubility of carboxylated cellulose. 84
85
86
Figure 1 : Schematic representation of the s elective oxidation reactions of cellulose . a) 87
TEMPO/NaBr/NaClO oxidation (C6 only); b) periodate (C2 and C3) and c) 88
combination of these reaction s (C6, C2, C3) . 89

4
90
2. MATERIALS AND METHODS 91
2.1 Materials 92
Never -dried b leached Eucalyptus Kraft (BEK) pulp (~10 wt.% solids content, Australian 93
Paper, Maryvale, Australia ) was used as the native cellulose sample. This is a standard 94
grade used for fine paper production. All chemicals are analytical grade and used without 95
further purification. 2,2,6,6 -Tetramethylpiperidine -1-oxyl (TEMPO) , sodium bromide (NaBr) , 96
sodium periodate (NaIO 4) and hydroxylamine hydrochloride (NH 2OH.HCl) were purchased 97
from Sigma -Aldrich (NSW, Australia) . Hydrochloric acid (HCl) and sodium hydroxide (NaO H) 98
pellets were purchased from ACL Laboratories and Merck, respectively. 12 w/v% sodium 99
hypochlorite (NaClO) was purchased from Thermo Fisher Scientific and used as received. 100
2.2 Oxidation Protocols 101
All oxidation protocols are based on the TEMPO -periodate oxidation of microcrystalline 102
cellulose reported previously (Baron et al., 2019; Coseri et al., 2015) . Table I shows the 103
oxidation protocols and the corresponding amount of each component in the oxidation 104
system. In the one -shot oxidation reaction (TEMPO/NaClO/NaBr/NaIO 4), the concentration 105
of NaIO 4 was varied from 0 to 5.0 mmol/g cellulose. For comparison, control reactions 106
including the classical TEMPO and periodate reactions were also performed. A set of 107
sequential reactions (i.e. TEMPO then periodate and vice versa) was also performed. 108
Table I. Experimental protocols. Oxidation was perfo rmed using 10 g of cellulose from BEK 109
pulp at pH 10.5. 110
111
In a typical reaction, 10 g (dry weight basis) of BEK pulp was suspended in 1200 mL 112
distilled water with the designated amounts of TEMPO, NaBr, and/or NaIO 4. The reaction 113
vessel was covered with alumin ium foil to prevent any photoinduced decomposition of 114 Oxidation Protocol Component Concentration (mmol/g) Total
Reaction
Time (hr) Product
Code TEMPO NaBr NaClO NaIO 4
Control reactions
TEMPO/NaClO/NaBr 0.5 8.0 8.0 – 4 TOF
NaIO 4 – – – 2.5 4 P1OF
NaIO 4/NaClO/NaBr – 8.0 8.0 2.5 4 P2OF
One-shot reaction
TEMPO/NaClO/NaBr/NaIO 4 0.5 8.0 8.0 0 – 5.0 4 TPOF
Sequential reactions
TEMPO/NaClO/NaBr then NaIO 4 0.5 8.0 8.0 2.5 8 TPOFs
NaIO 4 then TEMPO/NaClO/NaBr 0.5 8.0 8.0 2.5 8 PTOFs

5
periodate. NaClO (12% v/v, pH adjusted to 10.5), if necessary, was then added dropwise 115
under constant stirring . The pH of the reaction was maintained at 10 .5 by adding 0.5 M 116
NaOH. After 4 hours, the oxidation reaction was stopped by quen ching with ca. 10 mL 117
ethanol. The water -insoluble fraction was vacuum filtered and washed several times with 118
distilled water until neutral pH. The fibres were then freeze -dried for 24 h. In addition, the 119
water -soluble fraction was recovered by precipitatio n with an excess amount of ethanol, 120
followed by centrifugation at 14,000 rpm for 10 mins. The water -soluble fibres were then 121
redissolved in water, dialyzed for at least 3 days and lyophilized for 24 h. 122
For the sequential reactions, 10 g of the water -insoluble fraction from the first 123
oxidation reaction was subjected to the next oxidation reaction, for a total oxidation time of 8 124
hours. The water -soluble fraction was obtained after the second oxidation reaction. 125
The mass yields of the reactions were measur ed by weighing the obtained freeze – 126
dried products on an analytical balance . 127
2.3 Characterization of the oxidised fibres 128
Carboxylate and aldehyde contents and degree of substitution (DS) 129
The carboxylate content of the oxidised fibres was determined by an electric 130
conductivity method (Mendoza, Batchelor, et al., 2018; Perez, Montanari, & Vignon, 2003) . 131
Freeze -dried oxidised fibres (ca. 0.1 g) were suspended in 40mL deionised water . Prior to 132
titration, 100 µL 0.1% (w/v) NaCl was added to increase sample conductivity and the pH was 133
adjusted to 2.5 -3 to protonat e all the carboxylate groups . Conductimetric titration was 134
performed using a 0.1 N NaOH as titrant operating at a rate of 0.1 mL/min (Mettler Toledo 135
T5 titrator ). The carboxylate content (mmol COO-/g fibre) was calculated using Equation 1: 136
137

where V 2 and V 1 are the volume of titrant required to neutralise the carboxylic groups 138
(plateau region in the titration curve), C is the NaOH concentration (mol /L), and w is the dry 139
sample weight. 140
141
Similarly, the degree of substitution (DS) was determined using the titration results. 142
DS was expressed as the ratio of the amount of the sodium carboxylate groups and the total 143
hydroxyl groups in the anhydroglucose unit (AGU). The degree of substitution was 144
calculated by the following equation: 145
146

6

where V 2 and V 1 are the volume of titrant required to neutr alise the carboxylic groups 147
(plateau region in the titration curve), C is the NaOH concentration (mol/L), w is the dry 148
sample weight and 111 is the molar mass difference of 2,3,6 -tricarboxycellulose and AGU. 149
150
The a ldehyde content was also determined using an oxime reaction, as previously 151
reported (Sirvio et al., 2011) . Freeze -dried oxidised fibres (ca. 0.1 g) were suspended i n 0.2 152
M NH 2OH.HCl buffered in 0.1 M acetate buffer (pH 4.5). The suspension was then stirred for 153
4 hours at room temperature and the resulting fibres washed several times with distilled 154
water and isolated by vacuum filtration followed by freeze -drying for 24 h. The nitrogen 155
content of the oxidised fibres, which is directly related to the aldehyde content, was 156
determined using a PerkinElmer CHNS/O 2400 Series II elemental analyser. 157
158
Degree of polymerisation (DP v) 159
The average viscometric degree of polymeris ation (DP v) was determined as reported 160
by Shinoda et al. (Shinoda, Saito, Okita, & Isogai, 2012) . A known amount of the oxidised 161
fibres (about 0.25 g) w as dissolved in 0.5 M copper ethylenediamine (CED, 50 mL) for 30 162
min ("ISO 5351:2010 Pulps – Determination of limiting viscosity number in cupri – 163
ethyle nediamine (CED) solution," 2010) . Intrinsic viscosities of the solutions were obtained 164
by using a Cannon−Fenske capillary viscometer, and these values were converted to DP v 165
values with the Mark−Houwink−Sakurada equation (Saito et al., 2009) where [ŋ] is the 166
average viscosity measured from three replicates: 167

FT-IR 168
Approximately 50 mg of freeze -dried fibres were analysed with a Fourier Transform Infrared 169
(FT-IR) spectrometer (Agilent Technologies Cary 630 FTIR) equipped with a diamond ATR 170
accessory. Eight scans at 4 cm-1
resolution were recorded in the range of 4000 -500 cm-1. 171
172
13C NMR 173
The water -insoluble fibres (TPOFa, T OF, BEK) were subject ed to cross -polarization/magic 174
angle spinning (CP -MAS) 13C NMR analysis (Bruker -Avance III 300) with a MAS rate of 10 175
kHz. Samples were packed uniformly in a 4 mm zirconium oxide rotor. However, the water – 176

7
soluble fraction was dissolved in D 2O and 13C-NMR spectra were recorded with a Bruker – 177
Avance DRX 600 MHz Spectrometer at 16,000 scans. 178
179
X-ray Diffraction 180
Freeze -dried water -insoluble fibres were subjected to powder X -ray diffraction analyses 181
using a Bruker D8 Advance. The samples were pressed at 20,000 psi for 30 seconds to form 182
a pellet. X -ray diffractograms were recorded from 10 to 30 ˚ of diffraction angle 2θ Ni-filtered 183
Cu K α radiation (λ = 0.1548 nm) at 40 kV and 40 mA. The crystallinity index (CI) was 184
calculated from the ratio of the area of the crystalline region and the total area of the curve 185
(Equation 3). Areas were estimated by curve -fitting using the Gaussian function. Crystal 186
sizes of the 200 direction of cellulose I were calculated from the full widths at half heights of 187
the diffraction peaks by Debye -Scherrer equation. 188

where A c and A T are the area of the crystalline region and the total area, respectively. 189
190
Dynamic Light Scattering and Zeta Potential 191
DLS and zeta -potential of the oxidised fibres were measured based on a previous report 192
(Okita, Saito, & Isogai, 2010) . Suspensions of the oxidised fibres (0.01 wt%) were sonicated 193
for 2 mins using an ultrasonic homogeniser at 19.5 kHz and 70% amplitude (ON/OFF, 5 s). 194
Unfibrillated fibres were separated by centrifuga tion at 12000g for 5 min and the 195
supernatants were analysed by DLS and zeta -potential using a particle size and zeta 196
potential analyser (Brook haven Nanobrook Omni). Measurements were performed five times 197
for each sample. 198
Field Emission -Scanning Electron Microscopy (FE -SEM) 199
Scanning electron microscopy (SEM) was performed using a n FEI Nova NanoSEM 450 200
equipped with a field -emission source. Freeze -dried fibres were mounted onto carbon stubs, 201
coated with iridium coating and examined at 5 kV. 202
203
204
205

8
206
3. RESULTS 207
3.1 Mass yield, carboxylic content and degree of substitution (DS) 208
Table II shows the mass yield of the water -insoluble and water -soluble fractions with 209
their corresponding carboxylate contents and degree of substitution (DS) after oxidation. The 210
highes t amount of water -insoluble fraction was recovered from cellulose oxidised with NaIO 4 211
and TEMPO/NaClO/NaBr (>92% yield). However, periodate oxidation did not significantly 212
introduce carboxylate groups in cellulose (0.06 mmol COO-/g). The typical 213
TEMPO/NaCl O/NaBr oxidation produced highly -oxidised fibres with COO-
content of 1.35 ± 214
0.11 mmol/g and DS of 0.37 . One-shot TEMPO -periodate oxidation and sequential oxidation 215
reactions, on the other hand, significantly reduced the yield of the water -insoluble fraction 216
(73%) of high carboxylate content (1.5 -1.9 mmol COO-/g) and DS (0.4 -0.52). Further, all 217
oxidation systems produced a water -soluble fraction except for the oxidation with periodate 218
alone. The one -shot TEMPO -periodate oxidation produced the highest amount of water – 219
soluble fraction (8%), as well as the highest carboxylate content (4.03 mmol COO-/g pulp ) 220
and DS (1.17) among all fractions. 221
Table II. Mass yield (%) and carboxylate content (mmol/g) of the water -insoluble and water – 222
soluble fractions for the different oxidation protocols. 223
*NaIO 4 concentration was set to 2.5 mmol/g cellulose ; oxidation reactions were performed at pH 10.5, room 224
temperature for 4h. 225
226
Based on these results, we investigated the effect periodate concentration has on the 227
recovery of the water -insoluble and water -soluble fractions and their carboxylate content 228
(Figure 2). In general, increasing the periodate concentration decreases the yield of the 229
water -insoluble fraction, whilst increasing the amount of water -soluble fraction (Figure 2a). 230
The decrease in the yield of the water -insoluble fraction coincides with a drastic increase in 231
its carboxylate content, especially at NaIO 4 concentrations greater than 2.5 mmol/g (Figure 232
2b). Increasing the periodate concentration also increased the carboxylate content of the 233
water -soluble products, which however, started to diminish over 1 mmol/g NaIO 4 (Figure 234
2c). By varying the periodate concentration in the one -shot reaction, the degree of 235 Oxidation Protocol* Water -Insoluble Fraction Water -Soluble Fraction
Mass yield
(%) COO-
Content
(mmol/g) DS Mass yield
%) COO-
Content
(mmol/g) DS
NaIO 4 92 0.06 0.01 – – –
NaIO 4/NaClO/NaBr 79 0.59 0.16 5.5 – –
TEMPO/NaClO/NaBr 93 1.35 0.37 1.3 2.72 0.63
TEMPO/NaClO/NaBr/NaIO 4 73 1.91 0.52 8.0 4.03 1.17
TEMPO/NaClO/NaBr then NaIO 4 73 1.47 0.40 5.6 2.92 0.63
NaIO 4 then TEMPO/NaClO/NaBr 73 1.61 0.44 1.3 2.60 0.65

9
substitution can be controlled from 0.2 to 1.1. Figure 2d shows a remarkable decrease in 236
the yield of the water -insoluble product as DS reaches 0.4. On the contrary, an abrupt 237
increase in the water -soluble fraction recovery occurs at DS = 1.1. There is a non -linear 238
correlation between the DS of the water -insoluble and water -soluble fractions. 239
240
Figure 2 : Effect of periodate concentration on the mass yield of the water -insoluble and 241
water -soluble fractions from TEMPO/NaClO/NaBr/periodate -oxidation of 242
native cellulose (a). Mass yield, carboxylate content, and DS of the water – 243
insoluble (b) and wate r-soluble fractions (c) from 244
TEMPO/NaClO/NaBr/periodate -oxidation of native cellulose at different 245
periodate concentrations. Effect of DS on the mass yield of water -insoluble 246
and water -soluble fractions [inset shows the relationship of the DS of the 247
water -insoluble and water -soluble fractions] (d). 248
3.2 Chemical Composition 249
The changes in chemical structure and composition of native cellulose were probed 250
spectroscopically. The FTIR spectra of the original native cellulose and the water -insoluble 251
and water -soluble fractions after the different oxidation reactions are shown in Figure 3. A 252
typical broad O -H stretching at 3300 cm-1 is present in all cellulose samples. In addition, 253
sp3-hybridized C -H stretching at 2900 cm-1 is also apparent, although the peak int ensity 254
decreased after oxidation (almost complete disappearance of the peak in the water -soluble 255
fractions), indicating that the C -H groups (in C6, C2 , and/or C3) originally present in cellulose 256
have disappeared due to oxidation. Moreover, the C -O-C stret ching vibrations at 1000 -1060 257
0 2 4 60246810DSCOO
– content (mmol/g)
Periodate added (mmol/g) Yield
Carboxylate content
DSYi
eld (%)
0246
0.751.001.251.50D C
0.2 0.6 0.8 1.0 1.2056065707580859095
0.25 0.30 0.35 0.400.50.60.70.80.91.01.11.21.3DSwater-soluble fraction
DSwater-insoluble fractionYield (%)
DSWater-insoluble fraction
Water-soluble fraction
0 2 4 660708090100
Periodate added (mmol/g) Yield
Carboxylate content
DS
1.21.41.61.82.0
0.200.250.300.350.400.450.50
DSYield (%)
COO- content (mmol/g)
0 1 2 3 4 5020406080100Yield (%)
Periodate added (mmol/g) Water-insoluble Fraction
Water-soluble Fraction
Overall YieldA B

10
cm-1, –OH in -plane bending at 1201 cm-1, C–H deformation stretching vibration, at 1112 cm-1 258
and asymmetry stretch vibration at 1165 cm became weaker after oxidation which suggest s 259
some degradation has occurred. Sharp peaks at 1615 cm-1 are present in most oxidised 260
samples, indicating the formation of carboxylate groups under the sodium salt form. Among 261
the water -insoluble fractions, TPOFa (one -shot oxidation reaction) exhibited th e highest 262
intensity of the latter peak and thus the highest degree of oxidation (Figure 3a). On the 263
contrary, the absence of the 1615 cm -1 peak in P1OFa (periodate -oxidised cellulose) 264
reveals that no carboxylate groups were formed. Instead, a very weak pea k at 1717 cm-1 is 265
present in P1OFa which corresponds to the C=O stretching vibration of the aldehyde groups. 266
Unreacted sodium periodate was even isolated in the water -soluble fractions P2OFb and 267
TPOFsb. The FTIR spectra of these fractions closely resemble those of sodium 268
paraperiodate (Na 3H2IO6), a product of the reaction of NaClO and IO 3- (Hen rikki Liimatainen, 269
Sirviö, Pajari, Hormi, & Niinimäki, 2013) . 270
271
Figure 3 : FT-IR spectra of the water -insoluble (a) and water -soluble fractions (b) from 272
different oxidation reactions of native cellulose. 273
274
The oxidised fibres were also analysed by NMR: i.e., the water -insoluble fractions by 275
13C CP -MAS solid -state NMR and the water -soluble fractions by solution 13C NMR . Figure 4a 276
shows the characteristic NMR signals between 65 and 105 ppm are present in all samples. 277
A peak around 105 ppm corresponds to the C1 carbon of cellulose while the two peaks at 278
around 88.6 ppm and 82.6 ppm are assigned to the C4 carbon in the crystalline and non – 279
crystalline regions, respectively. In addition, signals for the C6 carbon at the crystalline and 280
non-crystalline regions were located at 64.8 ppm and 62.7 ppm, respectively. Multiplet 281
signals between 76 -70 ppm are assigned to C2, C3, and C5 carbons (Foster et al., 2018) . 282
Significant spectral modifications were observed after oxidation of native cellulose. In 283
parallel with the FTIR results, successful carboxylat ion was evidenced by the presence of a 284
distinct peak at around 175 ppm in all oxidised samples except P1OFa . Moreover, the peak 285
4000 3500 3000 2500 2000 1500 1000 500P2OF
TPOFs
PTOFs
TPOF
Wavenumber (cm-1)TOFCOO-Na+
4000 3500 3000 2500 2000 1500 1000 500P1OF
TPOFs
PTOFs
TOF
BEK
Wavenumber (cm-1)COO-Na+
TPOFP2OFA B

11
intensity for the C6 non -crystalline signals also decreased as some CH 2-OH groups were 286
oxidised to COOH. In addition, the peak for the C4 non -crystalline region changed from 287
82.6 ppm to 83.9 ppm after oxidation with TEMPO or TEMPO -periodate. TPOFa exhibited a 288
lower intensity at signals arising from C2, C3, and C5 compared to the original sample, 289
implying the conversion of the OH groups in C2 and C3 to carbonyl groups (Coseri et al., 290
2015) . P1OF did not show any resonance in the carbonyl region which would arise from the 291
two carbonyl moieties formed during periodate oxidation. Similar to FTIR results, the addition 292
of NaClO and NaBr to periodate (P2OFa) promoted the oxidation of hydroxyl groups into 293
carboxylate groups . 294
The water -soluble fractions dissolved in D 2O were analysed by solution 13C NMR. 295
Figure 4b shows the 13C NMR spectrum of TPOFb in D 2O. Interestingly, the water -soluble 296
fraction isolated from the one -shot reaction exhibited three dominant peaks in the range of 297
174-177 ppm, indicating the presence of three different COOH groups from the 298
anhydroglucose unit of the cellulose. The water -soluble product closely resembles the 299
structure of 2,3,6 -tricarboxycellulose or sodium mesotartarate/monohydrated glyox ylate 300
alternating c o-polyacetal . However, the presence of small and unknown signals in the 13C 301
NMR spectrum also suggests that side reactions are occurring. In contrast, oxidation using 302
TEMPO/NaClO/NaBr only shows one distinct peak in the carbonyl region (around 175 ppm) 303
which corresponds to the C6 carboxylate group formed. Interestingly, performing the 304
oxidation reactions sequentially (in either order) did not result in the formation of 2,3,6 – 305
tricarboxycellulose. 306

12
307
Figure 4. 13C CP/MAS NMR spectra (a) of cellulose and water -insoluble fractions and 13C- 308
NMR spectrum of water -soluble fractions in D 2O (b) from different oxidation reactions of 309
native cellulose. 310
311
3.3 Degree of Polymerisation (DP v) 312
Table III compares the DP v of BEK and the water -insoluble fractions after the 313
different oxidations at pH 10.5. From the original native cellulose DP v of 1354, a dramatic 314
decrease in DP v of the water -insoluble fractions occurred. Oxidation of native cellulose with 315
TEMPO/NaClO/NaBr caused the DP V to drop to 262, similar to the results from TEMPO – 316
mediated oxidation of bleached hardwood kraft pulp and cotton linter (Kitaoka, Isogai, & 317
Onabe, 1999; Saito & Isogai, 2004) . The addition of 2.5 mmol/g of sodium periodate induces 318
extensive depolymerisation, causing a drop of DP v to 234. This is also the case when 319
TEMPO -mediated and periodate oxidations (and vice versa) are performed sequentially. 320
Moreover, performing periodate oxidation at pH 10.5 depolymerises c ellulose to DP v = 701 321
whilst the addition of NaClO and NaBr to this oxidation system further promoted 322
depolymerisation (DP v = 368). 323
200 180 160 140 120 100 80 60 40 20 0P2OFaP1OFa
TPOFa
TOFa
Chemical Shift (ppm)PTOFsa
TPOFsa
200 180 160 140 120 100 80 60
Chemical shift, ppmC6C6OONa
C6OONa
C5,3,2C4
onlyTEMPO/NaClO/NaBrreactionOne-shotOxidations
C6,3,2OONa
C6OONa C1C1C1C1
TPOFsb
PTOFsb
TPOFbCOO-
TOFbSequential
C5,3,2
C4C5,3,2C4C4C5,3,2
BEKCOO-A
B

13
We also investigated the effect increasing periodate concentration has on the DP v of 324
the water -insoluble fractions from the one -shot TEMPO -periodate oxidation. Increasing the 325
periodate amount in the one -shot reaction decreases the DP v of the water -insoluble fraction 326
(Figure 5a). This correlates with the increase in the carboxylate cont ent and degree of 327
substitution of the water -insoluble fraction as the periodate concentration increases. As 328
shown in Figure 5b, as DS increases, DP v decreases with a consequent decrease in mass 329
yield. 330
Table III: Average viscometric degree of polymerisat ion (DP v) of the water -insoluble 331
fractions from different oxidation reactions of native cellulose. Cellulose fibres 332
were dissolved in 0.1 M CED solution at 25 ˚C. 333
*NaIO 4 concentration was set to 2.5 mmol/g cellulose ; oxidation reactions were performed at pH 10.5, room 334
temperature for 4h. 335
336
337
Figure 5 : Effect of periodate concentration on the average viscometric degree of 338
polymerisation (DP v) and carboxylate content of the water -insoluble fractions 339
from TEMPO /NaClO/ NaBr/periodate -oxidation of native cellulose [DP v = 340
1354] (a). Effect of DS on the DP v and yield of the water -insoluble fractions 341
(b). 342
343
3.4 Crystallinity 344
Figure 6a shows the XRD patterns of native cellulose before and after the different 345
oxidation reactions. The typical diffraction peaks of cellulose I at 2θ = 14.8 ˚ ( ), 16.4˚ 346
0.24 0.26 0.28 0.30 0.32 0.34 0.36 0.38 0.40160180200220240260280300 DPv
Yield
DSDPv
50556065707580859095100105
Yield (%)
0 1 2 3 4 5200220240260280300 Degree of Polymerisation
Carboxylate Content
Periodate added (mmol/g)Degree of Polymerisation (DPv)
1.21.31.41.51.61.71.81.92.0
Carboxylate content (mmol/g)A BOxidation Protocol* Degree of Polymerisation (DP v)
Native Cellulose (BEK) 1354 ± 27
NaIO 4 705 ± 13
NaIO 4/NaClO/NaBr 368 ± 15
TEMPO/NaClO/NaBr 272 ± 13
TEMPO/NaClO/NaBr/NaIO 4 (one-shot reaction) 234 ± 12
TEMPO/NaClO/NaBr then NaIO 4 236 ± 13
NaIO 4 then TEMPO/NaClO/NaBr 247 ± 14

14
(110) and 22.6 ˚ (200) (Saito & Isogai, 2004) are observed. In most cases, the X -ray 347
diffrac tion patterns slightly change after the oxidation reactions, except for those oxidised in 348
the one -shot reaction (TPOFa). For TEMPO/NaOCl/NaBr/periodate -oxidised cellulose, the 349
diffraction peak at the 200 plane shifted to lower 2 θ. Oxidation of native cell ulose increases 350
the crystallinity index of the water -insoluble fraction (Figure 6b). Cellulose treated with the 351
one-shot oxidation reaction has the highest crystallinity index (CI 80%) . However, the 352
crystal size at the (200) plane is not significantly af fected by oxidation. 353
The effect of periodate concentration on the XRD pattern, crystallinity and crystal size 354
of native cellulose was determined for the one -shot reaction (Figure 7). Increasing the 355
periodate concentration also increases the shift in t he diffraction peak of the 200 plane 356
(Figure 7a – except for 5.0 mmol/g). Increasing the periodate concentration also increases 357
the crystallinity but retains the crystal size of cellulose (Figure 7B). 358
359
360
Figure 6: Changes in the crystallinity of native cellulose after different oxidation 361
reaction . X-ray diffraction patterns of oxidised native cellulose s (a) and their 362
corresponding crystallinity indices (%) and crystal size (200 plane) (b). Native 363
cellulose was oxidised at pH 10.5 and at room temperature for 4 h. 364

15
365
Figure 7: X-ray diffraction patterns of the oxidised native cellulose (a). Effect of 366
perioda te concentration on the crystallinity and crystal size (in the 200 367
direction) of the water -insoluble products from 368
TEMPO /NaClO/NaBr/periodate -oxidation of native cellulose (b). Effect of DS 369
on the crystallinity and yield of the water -insoluble product (c). 370
371
372
3.5 Fibre Morphology 373
FE-SEM images of the water -insoluble fibres from the different oxidation systems are 374
shown in Figure 8. Fibres after periodat e (Figure 8b) and TEMPO/NaClO/NaBr (Figure 8e) 375
oxidations at pH 10.5 are morphologically similar to those of the pristine cellulose (Figure 376
8a). However, the addition of periodate in the TEMPO/NaClO/NaBr system creates more 377
fibrillated and shorter fibres ( Figure 8f -j). The fibrillation of the long fibres can be readily 378
observed from concentrations of 0.1 mmol/g periodate and the formation of short fragments 379
or particles are noticeable after oxidation with 0.5 mmol/g periodate in the one -shot reaction. 380
Seque ntial oxidations also promote the fibrillation and scission of fibres (Figure 8c and d). 381
Dynamic light scattering analysis of the fibre suspensions reveals that increasing the 382
periodate concentration in the one -shot reaction decreases the effective fibril diameter. 383
C
0 1 2 3 4 5020406080100 Crystallinity Index
Crystal size
Periodate added (mmol/g)Crystallinity Index (%)
0246810
Crystal Size (nm)
10 15 20 25 302.5 mmol/g 5.0 mmol/g
1.0 mmol/g
0.5 mmol/g
0.1 mmol/g
0 mmol/g
Diffraction angle 2θ (˚)BEK
0.24 0.26 0.28 0.30 0.32 0.34 0.36 0.38 0.40406080100 Crystallinity Index
Yield
DSwater-insoluble fractionCrystallinity Index (%)
406080100
Yield (%)A
B

16
384
Figure 8: Changes in fibre morphology of oxidised native cellulose. FE -SEM images of 385
native cellulose (a) after periodate oxidation (b), periodate then 386
TEMPO/NaClO/NaBr oxidation (c), TEMPO/NaClO/NaBr then periodate 387
oxidation (d), TEMPO/ NaClO/NaBr oxidation (e), and one -shot 388
TEMPO/NaClO/NaBr/periodate oxidation at increasing periodate 389
concentrations (0.1 -5.0 mmol/g) (f-g). Effect of periodate concentration on the 390
effective diameter (as determined by DLS) of TEMPO/NaClO/NaBr/periodate – 391
oxidised cellulose fibres (k). 392
393
394
4. DISCUSSION 395
396
4.1 Mechanism of cellulose TEMPO -periodate oxidation 397
Based on results and literature, a mechanism of cellulose TEMPO -periodate 398
oxidation is proposed in Figure 9. The three hydroxyl groups of cellulose are converted to 399
carboxylic acids in the presence of TEMPO/NaClO/NaBr and sodium periodate at pH 10.5. 400
The reaction proceeds by converting the three hydroxyl groups to aldehydes and ultimately, 401
to carboxylic acids. In the first stage of the reaction, the C6 -OH groups are oxidised to 402

17
aldehydes by TEMPO and NaClO. The cationic TEMPO+ ions generated from the oxidation 403
of TEMPO with NaClO preferentially oxidises the primary alcohol group at C6, forming 404
aldehyde groups and N -hydroxy -TEMPO molecules (A. Isogai & Shibata, 2001; Kitaoka et 405
al., 1999; Saito & Isogai, 2004; Saito, Kimura, Nishiyama, & Isogai, 2007; Sait o, Nishiyama, 406
Putaux, Vignon, & Isogai, 2006) . At the same time, the C2 – and C3 -OH groups are converted 407
to 2,3 -dialdehydes following the periodate oxidation mechanism. The vicinal diols react with 408
periodate to form a cyclic periodate ester which readily undergoes rearrangement, forming 409
two aldehyde groups and IO 3- as a by -product (Coseri et al., 2015; Kristiansen, Potthast, & 410
Christensen, 2010; H. Liimatainen et al., 2012) . However, as revealed by FT -IR and 13C 411
NMR analyses, no or very few aldehydes were formed. Oxime reaction and subsequent 412
elemental analysis confirm the presence of only a few a ldehyde groups in P1OFa (0.01 413
mmol/g pulp). Periodate oxidation, usually performed at low pH and elevated temperatures 414
(Wei 2016; Kim et al. 2000; Kristiansen et al. 2010; Sirvio et al. 2011), is severely inhibited at 415
alkaline pH (Nevell 1957). Performing periodate oxidation at pH 10.5 severely hampers the 416
formation of aldehyde groups in cellulose , as the resulting complex does not readily 417
decompose into iodate and an aldehyde . However, the presence of other oxidants and co – 418
catalysts such as NaClO and NaBr in the system (product P2OFa) synergistically work in the 419
formation of dialdehydes and/or carboxylic acid. 420
In the second stage, the aldehyde groups are further oxidised to carboxylic acids in 421
the presence of TEMPO/NaBr/NaClO. Under alkaline conditions, a ldehydes undergo 422
hydration and successive dissociation which enable TEMPO+ to form new covalent bonds, 423
resulting in the formation of carboxy groups and N -hydroxy -TEMPO molecules . Aldehydes 424
might be converted to carboxylate groups due to the presence of NaBrO and/or NaClO from 425
the TEMPO/NaBr/NaClO system at pH 10 -11(T. Isogai et al., 2010) . 426
427
428

18
429
430
Figure 9 : Proposed chemical reaction for the oxida tion of cellulose in the presence of 431
TEMPO/NaBr/NaClO and sodium periodate. 432
433
4.2. Effect of periodate on cellulose one -shot TEMPO -periodate oxidation 434
The presence of periodate ions is crucial in the one -shot oxidation of cellulose. Thus, 435
the effect of periodate concentration on the properties of the oxidised products was 436
investigated. 437
4.2.1. Carboxylate content 438
Oxidation of bleached pulp in a typical TEMPO/NaBr/NaClO reaction form s a highly – 439
carboxylated water -insoluble fraction (1.35 mmol COO-/g). This coincides with previous 440
studies on TEMPO -oxidation of native cellulose (using ca. 8 mmol/g NaClO) (A. Isogai & 441
Kato, 1998; Mendoza, Batchelor, et al., 2018; Mendoza, Gunawardhana, et al., 2018; 442
Mendoza et al., 2019; Shinoda et al., 2012) . However, t he addition of per iodate in the 443
classical TEMPO/NaClO/NaBr system, either in a one -shot or sequential reaction, increases 444
the carboxylate contents and DS of the water -insoluble fractions and decreases their mass 445
recovery (Table II). This increase in carboxylate content is expected, as sodium periodate 446
introduces more sites (i.e. two aldehyde groups) for carboxylate formation. Interestingly, 447
there is a remarkable increase in the carboxylate content up to 1.9 mmol/g (corresponding to 448
the addition of 2.5 mmol/g IO 3-). However, Shinoda et al. reported the highest carboxylate 449
content for the classical TEMPO -oxidation of native cellulose to be only up to 1.7 mmol/g 450

19
(Shinoda et al., 2012) . In theory, the maximum carboxylate content TEMPO -oxidation of 451
native cellulose can only be 1.61 mmol/g (Hiraoki, Ono, Saito, & Isogai, 2015) . These results 452
suggest that in the one-shot oxidation system, the three hydroxyl groups (C6, C2 and C3) 453
exposed in the crystal surface may be oxidised to carboxylate groups. 454
The oxidation of the three hydroxyl groups was further evidenced by the formation of 455
2,3,6 -tricarboxycellulose ( TCC) in the water -soluble product of the one -shot reaction (Figure 456
4). The formation of this highly oxidised product (4.01 mmol/g) strongly supports the 457
mechanism proposed (Figure 9). Interestingly, TCC only forms when native cellulose is 458
oxidised with TEM PO/NaBr/NaClO and sodium periodate in a one -shot reaction. On the 459
contrary, performing TEMPO and periodate oxidations in a two -step process yielded water – 460
soluble products similar to 6 -celluronic acid. 461
4.2.2. Crystallinity and Crystal Size 462
Significant peak shift in the 200 plane was observed in TEMPO -periodate oxidised 463
cellulose (Figure 6 and 7). This observation is not commonly observed in TEMPO -oxidised 464
native celluloses as the crystal structure usually remains unchanged (Okita et al., 2010; 465
Saito & Isogai, 2004) . The shifts in the diffraction peak at the 200 plane may be attributed to 466
the addition of carboxylate and aldehyde groups on the crystal surface of cellulose I. 467
Increasing the periodate concentration also increases the crystallinity index of the water – 468
insoluble fraction whilst simultaneously decreasin g the mass yield (Figure 7). This is 469
correlated with the accessibility and crystalline state of cellulose . Oxidation of the highly 470
crystalline cellulose I is difficult; as only the accessible hydroxyl groups on the surface are 471
oxidised (Saito, Yanagisawa, & Isogai, 2005) On the other hand, the amorphous regions, 472
susceptible to oxidation, are dissolved during the washing process because of their 473
enhanced water -solubility (Saito & Isogai, 2004) with the crystals remaining . However, this 474
oxidation clearly did not cause significant changes to the crystal size (Figure 7B). 475
4.3. Degree of polymerisation (DP v) and fibre morphology 476
The influence of TEMPO -periodate oxidation on the average viscometric degree of 477
polymerization (DP v) was determined. Copper ethylenediamine (CED) was selected as 478
solvent as it can completely dissolve partially – and fully-oxidised cellulose 479
molecules (Shinoda et al., 2012) . It was assumed from a previous study (Smith, Bampton, & 480
Alexander, 1963) that the same Mark -Houwink -Sakura da equation (DP = 1.75 [ ŋ]) is valid for 481
both the oxidised and non -oxidised cellulose. 482
Table III shows a dramatic decrease in DP v of the water -insoluble fractions . These 483
DPv values are close to the levelling -off DP (LODP) reported for native celluloses hydrolysed 484

20
by dilute acid (Yachi et al. 1983). Several studies have reported important depolymerisation 485
of cellulose during TEMPO -oxidation (Coseri, 2017; Saito, Hirota, T amura, & Isogai, 2010; 486
Shinoda et al., 2012) . For instance, DPv of regenerated celluloses, ranging from 220 to 680, 487
decreased to 40 after oxidation (T. Isogai & Isogai, 2009; T. Isogai et al., 2009; Shibata, 488
Yanagisawa, Saito, & Isogai, 2006) and DP v of bleached wood pulps and cotton linters also 489
decreased to 200 –300(A. Isogai & Shibata, 2001; Kitaoka et al., 1999) . Depolymerisation in 490
TEMPO -mediated oxidation systems is attributed to the β-elimination of glycosid ic bonds as 491
C6 aldehydes are formed in the oxidation process. In addition, some radical species in the 492
oxidation system may also form and contribute to the depolymerisation. 493
Figure 5 demonstrates that the addition of periodate in the TEMPO/NaClO/NaBr 494
system further promotes depolymerisation. More aldehyde groups in the C2 and C3 495
positions participate in the depolymerisation process th rough a β -elimination mechanism. 496
Extensive depolymerisation was also observed as the oxidation time and TEMPO and 497
NaClO concentrations were increased (Saito et al., 2010) . Figure 5b i ndicates that the DS 498
negatively influences DP v. 499
Furthermore, increasing the periodate concentration in the one -shot reaction 500
promotes fibrillation of the fibres and formation of short fragments. Perpendicular scissions 501
along the fibre axis were clearly o bserved. Similar observations were reported by Saito et al. 502
for TEMPO – oxidations of different types of cellulose (Saito & Isogai, 2004) . As indicated by 503
XRD, the formation of short fragments and fine particles results from the depolymerisation of 504
the non -crystal line region, liberating water -insoluble cellulose crystals. 505
506
4.3. DS governing the solubility of carboxylated cellulose 507
The role of DS on the solubility of cellulose derivatives is well documented (Brydson, 508
1999; Samaranayake & Glasser, 1993; Wüstenberg, 2014) . At low DS (0.8 -1.3), ethyl 509
cellulose is soluble in water as replac ing some hydroxyl groups by ethoxy groups reduces 510
the hydrogen bonding across the cellulosic ch ains (Brydson, 1999) . Sodium carboxymethyl 511
cellulose, on the other hand, readily dissolves in water at DS greater than 0.7. Similarly, at 512
this DS, carboxymethyl groups are sufficient to disrupt the crystalline region of cellulose and 513
become water -soluble (Wüstenberg, 2014) . 514
In the one -shot reaction, varying the periodate concentration controls the degree of 515
substitution. DS correlates with the yield of the water -insoluble and water -soluble fractions 516
(Figure 2D). The DS of the water -insoluble frac tions also correlates with their water -soluble 517
fractions. Increasing the DS of the water -insoluble product to 0.39 results in a significant 518

21
increase in the yield of the water -soluble fraction(DS=1.1). This is expected since more 519
carboxylate groups increase the polarity of cellulose. The degree of polymerisation 520
decreases with the extent of reaction, represented by DS (Figure 5B). The decrease in the 521
degree of polymerisation leads to a decrease in the yield of the water -insoluble fraction, as 522
long molecules are depolymerised into short chains. Similarly, increasing DS also increases 523
the crystallinity of the water -insoluble fractions (Figure 7C). Indeed, at DS = 0.39, almost 524
100% crystallinity was achieved. These results are consistent with the decrease in the yield 525
of the water -insoluble fraction; i.e., the accessible amorphous regions become more soluble 526
as DS increases. However, 1.1 was the highest DS achieved. The varying degree of 527
accessibility of the hydroxyl groups results in a heterogenous oxidation r eaction. In addition, 528
the presence of residual lignin and hemicellulose in the sample is likely to interfere with the 529
oxidation reaction. 530
The schematic representation of the one -shot TEMPO -periodate oxidation of native 531
cellulose is represented in Figure1 0. Depending on the degree of substitution or carboxylate 532
content, the TEMPO -periodate oxidation of native cellulose results in the formation of 533
partially -oxidised water -insoluble and fully -oxidised water -soluble fractions. The water – 534
insoluble fraction is a crystalline -rich fraction where surface carboxylate groups are present. 535
In contrast, the water -soluble fraction is composed of a variety of fully -oxidised 536
polyglucoronans, including 2,3,6 -tricarboxycellulose. 537
538
Figure 10: Schematic mechanism of the cell ulose one -shot TEMPO -periodate oxidation: 539
formation of water -insoluble and water -soluble fractions. 540
541
5. CONCLUSION 542
Oxidation of native cellulose from Kraft bleached pulp by combining 543
TEMPO/NaClO/NaBr and sodium periodate in a one -shot reaction result s in the formation of 544
two fractions: a highly -carboxylated water -insoluble (up to 1.9 mmol COO-/g) and a water – 545
soluble fraction (up to 4 mmol COO-/g). The concentration of periodate in the one -shot 546

22
reaction plays a crucial role in the chemical and physical pr operties of the oxidised cellulose 547
derivatives . The a ddition of periodate in the classical TEMPO/NaClO/NaBr system 548
significantly increase s the carboxylate content of the water -insoluble fraction, whilst 549
decreas ing the mass recovery. FT-IR and 13C NMR analyses confirm the introduction of 550
carboxylate groups in native cellulose and that a highly -water -soluble 2,3,6 – 551
tricarboxycellulose was prepared in the water -soluble fraction. Viscosimetric analysis 552
reveal s that increasing the periodate concentrat ion promotes extensive depolymerisation of 553
the native cellulose due to the presence of more residual aldehyde groups participating in 554
the depolymerisation reaction . X-ray diffraction show s that the periodate increases the 555
crystallinity and retains the crys tal size of cellulose I in the 200 plane , but also causes slight 556
alterations in the XRD pattern due to the oxidation of the accessible C2, C3 and C6 hydroxyl 557
groups on the surface of the crystalline cellulose . FE-SEM and DLS analyses also reveal 558
that incre asing the periodate concentration in the one -shot reaction promotes the fibrillation 559
and reduction of fibres into small fragments. Furthermore, when periodate and TEMPO – 560
mediated oxidations are performed sequentially in two steps, different oxidation produc ts are 561
obtained, suggesting that the TEMPO/NaClO/NaBr and periodate oxidation systems work in 562
synergy to carboxylate the three reactive hydroxyl groups in cellulose. 563
Varying the periodate concentration in the one -shot reaction control s the degree of 564
substi tution (DS). DS governs the solubility of carboxylated cellulose in water. By increasing 565
the DS of the water -insoluble product to 0.39, a significant increase in the yield of the water – 566
soluble product (DS=1.1) occurs. Depending on DS or carboxylate conten t, the TEMPO – 567
periodate oxidation of native cellulose results in the formation of partially -oxidised water – 568
insoluble (crystalline -rich) and fully -oxidised water -soluble fractions (polyglucoronans from 569
the amorphous regions). The one -shot TEMPO -periodate oxi dation allows the synthesis of 570
low-cost cellulose with controlled level of carboxylation and unique properties. 571
572
6. ACKNOWLEDGMENT 573
Fundin g from the Australian Research Centre -Industry Transformation Research Hub; 574
Processing Advanced Lignocellulosics (PALS) [grant number IH130100016 ] is gratefully 575
acknowledged. The authors also acknowledge the use of facilities of the Monash X -ray 576
Platform, Monash Centre for Electron Microscopy, and the NMR Laboratory in the School of 577
Chemistry, Monash University Clayton. 578
7. REFERENCES 579

23
Baron, R. I., Bercea, M., Avadanei, M., Lisa, G., Biliuta, G., & Coseri, S. (2019). Green route for the 580
fabrication of self -healable hydrogels based on tricarboxy cellulose and poly(vinyl alcohol). 581
International Journal of Bi ological Macromolecules, 123, 744 -751. 582
Brydson, J. A. (1999). 22 – Cellulose Plastics . In J. A. Brydson (Ed.), Plastics Materials (Seventh Edition) 583
(pp. 613 -634). Oxford: Butterworth -Heinemann 584
Coseri, S. (2017). Cellulose: To depolymerize… or not to? Biotechnology Advances, 35(2), 251 -266. 585
Coseri, S., Biliuta, G., Zemljic, L. F., Srndovic, J. S., Larsson, P. T., Strnad, S., . . . Lindstrom, T. (2015). 586
One-shot carboxylation of microcrystalline cellulose in the presence of nitroxyl radicals and 587
sodium pe riodate. Rsc Advances, 5(104), 85889 -85897. 588
Curvello, R., Mendoza, L., McLiesh, H., Manolios, J., Tabor, R. F., & Garnier, G. (2019). Nanocellulose 589
Hydrogel for Blood Typing Tests. ACS Applied Bio Materials . 590
Curvello, R., Raghuwanshi, V. S., & Garnier, G . (2019). Engineering nanocellulose hydrogels for 591
biomedical applications. Adv Colloid Interface Sci, 267, 47 -61. 592
de Nooy, A. E. J., Besemer, A. C., & van Bekkum, H. (1995). Highly selective nitroxyl radical -mediated 593
oxidation of primary alcohol groups in water -soluble glucans. Carbohydrate Research, 594
269(1), 89 -98. 595
Foster, E. J., Moon, R. J., Agarwal, U. P., Bortner, M. J., Bras, J., Camarero -Espinosa, S., . . . 596
Youngblood, J. (2018). Current characterization methods for cellulose nanomaterials. Chem 597
Soc Rev, 47(8), 2609 -2679. 598
Hiraoki, R., Ono, Y., Saito, T., & Isogai, A. (2015). Molecular mass and molecular -mass distribution of 599
TEMPO -oxidized celluloses and TEMPO -oxidized cellulose nanofibrils. Biomacromolecules, 600
16(2), 675 -681. 601
ISO 5351:2010 Pulps – Determination of limiting viscosity number in cupri -ethylenediamine (CED) 602
solution. (2010). 603
Isogai, A. (2018). Advanced TEMPO -mediated oxidation of cellulose: Preparation of TEMPO -CNCs 604
and layer -by-layer peeling of surface cellulose molecules on cellulose microfibrils. Abstracts 605
of Papers of the American Chemical Society, 256. 606
Isogai, A., & Bergström, L. (2018). Preparation of cellulose nanofibers using green and sustainable 607
chemistry. Current Opinion in Green and Sustainable Chemis try, 12, 15 -21. 608
Isogai, A., & Kato, Y. (1998). Preparation of polyglucuronic acid from cellulose by TEMPO -mediated 609
oxidation. Cellulose, 5(3), 153 -164. 610
Isogai, A., Saito, T., & Fukuzumi, H. (2011). TEMPO -oxidized cellulose nanofibers. Nanoscale, 3(1), 71 – 611
85. 612
Isogai, A., & Shibata, I. (2001). Chemical modifications of cellulose by TEMPO -mediated oxidation. 613
Sen-I Gakkaishi, 57(6), P163 -P167. 614
Isogai, T. (2017). Overview of TEMPO -mediated oxidation of polysaccharides. Abstracts of Papers of 615
the American Che mical Society, 253. 616
Isogai, T., & Isogai, A. (2009). Degree of polymerization (DP) of cellouronic acid prepared from ball – 617
milled native celluloses by TEMPO -mediated oxidation. Abstracts of Papers of the American 618
Chemical Society, 237. 619
Isogai, T., Saito, T., & Isogai, A. (2010). TEMPO Electromediated Oxidation of Some Polysaccharides 620
Including Regenerated Cellulose Fiber. Biomacromolecules, 11(6), 1593 -1599. 621
Isogai, T., Yanagisawa, M., & Isogai, A. (2009). Degrees of polymerization (DP) and DP distributio n of 622
cellouronic acids prepared from alkali -treated celluloses and ball -milled native celluloses by 623
TEMPO -mediated oxidation. Cellulose, 16(1), 117 -127. 624
Kitaoka, T., Isogai, A., & Onabe, F. (1999). Chemical modification of pulp fibers by TEMPO -mediated 625
oxidation. Nordic Pulp & Paper Research Journal, 14(4), 279 -284. 626
Klemm, D., Heublein, B., Fink, H. P., & Bohn, A. (2005). Cellulose: Fascinating biopolymer and 627
sustainable raw material. Angewandte Chemie -International Edition, 44(22), 3358 -3393. 628

24
Klemm, D., Philpp, B., Heinze, T., Heinze, U., & Wagenknecht, W. (1998). Comprehensive cellulose 629
chemistry. Volume 1: Fundamentals and analytical methods . Weinheim: Wiley -VCH Verlag 630
GmbH. 631
Kristiansen, K. A., Potthast, A., & Christensen, B. E. (2010). Periodate oxida tion of polysaccharides for 632
modification of chemical and physical properties. Carbohydrate Research, 345(10), 1264 – 633
1271. 634
Liimatainen, H., Sirviö, J., Pajari, H., Hormi, O., & Niinimäki, J. (2013). Regeneration and Recycling of 635
Aqueous Periodate Solution i n Dialdehyde Cellulose Production. Journal of Wood Chemistry 636
and Technology, 33(4), 258 -266. 637
Liimatainen, H., Visanko, M., Sirvio, J. A., Hormi, O. E., & Niinimaki, J. (2012). Enhancement of the 638
nanofibrillation of wood cellulose through sequential period ate-chlorite oxidation. 639
Biomacromolecules, 13(5), 1592 -1597. 640
Lindh, J., Carlsson, D. O., Strømme, M., & Mihranyan, A. (2014). Convenient One -Pot Formation of 641
2,3-Dialdehyde Cellulose Beads via Periodate Oxidation of Cellulose in Water. 642
Biomacromolecules, 15(5), 1928 -1932. 643
Mendoza, L., Batchelor, W., Tabor, R. F., & Garnier, G. (2018). Gelation mechanism of cellulose 644
nanofibre gels: A colloids and interfacial perspective. J Colloid Interface Sci, 509, 39 -46. 645
Mendoza, L., Gunawardhana, T., Batchelor, W., & Gamier, G. (2018). Effects of fibre dimension and 646
charge density on nanocellulose gels. Journal of Colloid and Interface Science, 525, 119 -125. 647
Mendoza, L., Hossain, L., Downey, E., Scales, C., Batchelor, W., & Garnier, G. (2019). Carboxylated 648
nanocellul ose foams as superabsorbents. Journal of Colloid and Interface Science, 538, 433 – 649
439. 650
Nevell, T. P. (1957). 35 -The mechanism of the oxidation of cellulose by periodate. Journal of the 651
Textile Institute Transactions, 48(12), T484 -T494. 652
Okita, Y., Saito, T ., & Isogai, A. (2010). Entire Surface Oxidation of Various Cellulose Microfibrils by 653
TEMPO -Mediated Oxidation. Biomacromolecules, 11(6), 1696 -1700. 654
Perez, D. D., Montanari, S., & Vignon, M. R. (2003). TEMPO -mediated oxidation of cellulose III. 655
Biomacromo lecules, 4(5), 1417 -1425. 656
Ruan, C. -Q., Strømme, M., & Lindh, J. (2018). Preparation of porous 2,3 -dialdehyde cellulose beads 657
crosslinked with chitosan and their application in adsorption of Congo red dye. Carbohydrate 658
Polymers, 181, 200 -207. 659
Saito, T., H irota, M., Tamura, N., & Isogai, A. (2010). Oxidation of bleached wood pulp by 660
TEMPO/NaClO/NaClO2 system: effect of the oxidation conditions on carboxylate content 661
and degree of polymerization. Journal of Wood Science, 56(3), 227 -232. 662
Saito, T., Hirota, M ., Tamura, N., Kimura, S., Fukuzumi, H., Heux, L., & Isogai, A. (2009). 663
Individualization of Nano -Sized Plant Cellulose Fibrils by Direct Surface Carboxylation Using 664
TEMPO Catalyst under Neutral Conditions. Biomacromolecules, 10(7), 1992 -1996. 665
Saito, T., & Isogai, A. (2004). TEMPO -mediated oxidation of native cellulose. The effect of oxidation 666
conditions on chemical and crystal structures of the water -insoluble fractions. 667
Biomacromolecules, 5(5), 1983 -1989. 668
Saito, T., Kimura, S., Nishiyama, Y., & Isogai, A. (2007). Cellulose nanofibers prepared by TEMPO – 669
mediated oxidation of native cellulose. Biomacromolecules, 8(8), 2485 -2491. 670
Saito, T., Nishiyama, Y., Putaux, J. L., Vignon, M., & Isogai, A. (2006). Homogeneous suspensions of 671
individualized micr ofibrils from TEMPO -catalyzed oxidation of native cellulose. 672
Biomacromolecules, 7(6), 1687 -1691. 673
Saito, T., Yanagisawa, M., & Isogai, A. (2005). TEMPO -mediated oxidation of native cellulose: SEC – 674
MALLS analysis of water -soluble and -insoluble fractions in the oxidized products. Cellulose, 675
12(3), 305 -315. 676
Samaranayake, G., & Glasser, W. G. (1993). Cellulose derivatives with low DS. II. Analysis of 677
alkanoates. Carbohydrate Polymers, 22(2), 79 -86. 678

25
Shibata, I., Yanagisawa, M., Saito, T., & Isogai, A. (2006). SEC-MALS analysis of cellouronic acid 679
prepared from regenerated cellulose by TEMPO -mediated oxidation. Cellulose, 13(1), 73 -80. 680
Shinoda, R., Saito, T., Okita, Y., & Isogai, A. (2012). Relationship between length and degree of 681
polymerization of TEMPO -oxidi zed cellulose nanofibrils. Biomacromolecules, 13(3), 842 -849. 682
Sirvio, J., Hyvakko, U., Liimatainen, H., Niinimaki, J., & Hormi, O. (2011). Periodate oxidation of 683
cellulose at elevated temperatures using metal salts as cellulose activators. Carbohydrate 684
Polymers, 83(3), 1293 -1297. 685
Smith, D. K., Bampton, R. F., & Alexander, W. J. (1963). Use of New Solvents for Evaluating Chemical 686
Cellulose for the Viscose Process. Industrial & Engineering Chemistry Process Design and 687
Development, 2(1), 57 -62. 688
Uribe, B. E. B., Chiromito, E. M. S., Carvalho, A. J. F., Arenal, R., & Tarpani, J. R. (2017). TEMPO – 689
oxidized cellulose nanofibers as interfacial strengthener in continuous -fiber reinforced 690
polymer composites. Materials & Design, 133, 340 -348. 691
Wüstenberg, T. (2014). Sodium Carboxymethylcellulose. Cellulose and Cellulose Derivatives in the 692
Food Industry . 693
Yang, W. S., Bian, H. Y., Jiao, L., Wu, W. B., Deng, Y. L., & Dai, H. Q. (2017). High wet -strength, 694
thermally stable and transparent TEMPO -oxidized cellulose nanofibr il film via cross -linking 695
with poly -amide epichlorohydrin resin. Rsc Advances, 7(50), 31567 -31573. 696
697
698
699
700

Graphical Abstract
Click here to download high resolution image