Superbase ionic liquids for effective cellulose processing from dissolution to carbonisation

Article abstract of DOI:10.1039/c7gc02671d

A range of superbase derived ionic liquids (SILs) was synthesised and characterised. Their ability to dissolve cellulose and the characteristics of the produced fibres were correlated to their specific structural and solvent properties. 17 ionic liquids (ILs) (including 9 novel) were analysed and six ILs were selected to produce fibres: 1-ethyl-3-methylimidazolium acetate [C₂C₁im][OAc], 1-ethyl-3-methylimidazolium diethyl phosphate [C₂C₁im][DEP] and the SILs 1-ethyl-1,8-diazabicyclo[5.4.0]undec-7-enium diethylphosphate [DBUEt][DEP], 1,8-diazabicyclo[5.4.0]undec-7-enium acetate [DBUH][OAc], 1,5-diazabicyclo[4.3.0]non-5-enium acetate [DBNH][OAc] and 1-ethyl-1,5-diazabicyclo[4.3.0]non-5-enium diethylphsophate [DBNEt][DEP]. The mechanical properties of these fibres were investigated. The obtained fibres were then carbonised to explore possible application as carbon fibre precursors. The fibres obtained using a mixture of 1,5-diazabicyclo[4.3.0]non-5-enium based SILs with acetate and hexanoate anions (9 : 1), [DBNH][OAc][Hex], showed a promising combination of strength, stiffness and strain at failure values for applications in textiles and fibre reinforcement in renewable composites. Using Raman spectroscopy it is demonstrated that these fibres exhibit a relatively high degree of structural order, with fewer defects than the other materials. On the other hand, analogous fibres based on imidazolium cation with acetate and hexanoate anions (9 : 1), [C₂C₁im][OAc][Hex] showed a decline in the quality of the produced fibres compared to the fibres produced from [C₂C₁im][OAc], [C₂C₁im][DEP] or [DBNH][OAc][Hex].

Full text of DOI:10.1039/c7gc02671d

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: O. Kuzmina, J.
Bhardwaj, S. Rizal Vincent, N. D. Wanasekara, L. M. Kalossaka, J. Griffith, A. Potthast, S. Rahatekar, S. J.
Eichhorn and T. Welton, Green Chem., 2017, DOI: 10.1039/C7GC02671D.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Please do not adjust margins
Green Chemistry
Page 1 of 10
View Article Online
DOI: 10.1039/C7GC02671D
Green Chemistry
ARTICLE
Superbase ionic liquids for effective cellulose processing from
dissolution to carbonisation.
Received 00th January 20xx,
Accepted 00th January 20xx
Olga Kuzmina,a Jyoti Bhardwaj,b Sheril Rizal Vincent,b Nandula Dasitha Wanasekara,c Livia Mariadaria Kalossaka,a Jeraime
Griffith,a Antje Potthast,d Sameer Rahatekar,e Stephen James Eichhorn b,c and Tom Welton a
DOI: 10.1039/x0xx00000x
A range of superbase derived ionic liquids (SILs) was synthesised and characterised. Their ability to dissolve cellulose and
the characteristics of the produced fibres were correlated to their specific structural and solvent properties. 17 ionic
liquids (ILs) (including 9 novel) were analysed and six ILs were selected to produce fibres: 1-ethyl-3-methylimidazolium
acetate [C2C1im][OAc], 1-ethyl-3-methylimidazolium diethyl phosphate [C2C1im][DEP] and the SILs 1-ethyl-1,8-
diazabicyclo[5.4.0]undec-7-enium diethylphosphate [DBUEt][DEP], 1,8-diazabicyclo[5.4.0]undec-7-enium acetate
[DBUH][OAc], 1,5-diazabicyclo[4.3.0]non-5-enium acetate [DBNH][OAc] and 1-ethyl-1,5-diazabicyclo[4.3.0]non-5-enium
diethylphsophate [DBNEt][DEP]. The mechanical properties of these fibres were investigated. The obtained fibres were
then carbonised to explore possible application as carbon fibre precursors. The fibres obtained using a mixture of 1,5-
diazabicyclo[4.3.0]non-5-enium based SILs with acetate and hexanoate anions (9:1), [DBNH][OAc][Hex], showed a
promising combination of strength, stiffness and strain at failure values for applications in textiles and fibre reinforcement
in renewable composites. Using Raman spectroscopy it is demonstrated that these fibres exhibit a relatively high degree of
structural order, with fewer defects than the other materials. On the other hand, analogous fibres based on imidazolium
cation with acetate and hexanoate anions (9:1), [C2C1im][OAc][Hex] showed a decline in the quality of the produced fibres
www.rsc.org/
compared
to
the
fibres
produced
from
[C2C1im][OAc],
[C2C1im][DEP]
or
[DBNH][OAc][Hex].
development and commercialisation of polyester fibres in the
1950s4 defined the beginning of the development and
commercialisation of major man-made fibres.5 In the early
1970s, a single stage spinning and drawing process for nylon
fibres was commercialised,4 petroleum based synthetic
polymer fibres were proven to be more economically feasible
than cellulose fibres produced via the viscose process. These
approaches resulted in polyester fibre becoming the world’s
dominant fibre since 2007,6 scaling down of cellulose fibre
production and growing production of synthetic fibres, which
accounted for 82% of worldwide fibre production by 2014.7
This trend has not changed with the novel solvents for
cellulose that have been developed since the 1970s to replace
the environmentally harmful CS2 solvent in the viscose
process. These solvents included LiCl/N,N-dimethylacetamide
Introduction
Today's society is highly dependent on petroleum based, non-
renewable, polymers. To decrease the environmental impact
of these materials large-scale utilisation of naturally occurring,
abundantly available, polymers such as cellulose, chitin, starch
and keratin needs to be developed. The viscose process is the
most common way to manufacture cellulose fibres. Viscose
rayon – the first commercially available synthetic fibre was
developed around 1910 and used as a reinforcement material
for tyres and, subsequently by the 1920s, other mechanically
important rubber goods such as drive belts, conveyors and
hoses.1 However, the manufacture of viscose fibres is a
complex, multistep process that involves the use of very
aggressive chemicals, which produces a great deal of waste
and requires large volumes of fresh water.2 The viscose
process is also known to depolymerise cellulose.3 Nylon fibres,
first commercialised in the late 1930s, and consequent
systems,8,9
DMSO/paraformaldehyde10
and
N-
methylmorpholine-N-oxide (NNMO). The Lyocell (NNMO)
process offered some advantages in comparison with other
common processes for cellulose manufacture such as a low
number of technological steps, no derivatization, good solvent
recovery and significantly lower environmental hazard.
However, disadvantages and limitations of the Lyocell process
are the poor thermal stability of NNMO and its predisposition
to side reactions and by-product formation.11
At the same time synthetic polymers such as poly(acrylonitrile)
(PAN) supplanted cellulose as the main precursor material for
controlled pyrolysis for the production of commercial carbon
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 2 of 10
View Article Online
DOI: 10.1039/C7GC02671D
ARTICLE
Journal Name
fibres.12 However, the use of toxic solvents in the production
of PAN and the emission of waste gases in the manufacturing
process of carbon fibres made from it have raised
environmental concerns.13
Ionic liquid
Tmelt
Tonse Maximum
Cellulose
DPw (±5)
Regenerated cellulose has long been considered to be a
potentially renewable precursor material to manufacture
carbon fibres. Some of the first carbon fibres to be
manufactured from cellulose were obtained in the 1950s.14,15
Despite comparatively poor yield on carbonization, cellulose
fibres are competitive as a carbon fibre precursors with
respect to thermal conductivity, high purity and mechanical
flexibility.16 Improvement of the carbon yield and mechanical
properties of obtained carbon fibres through controlling the
cellulose fibres’ morphology and optimising processing
conditions have offered the main scope for research in this
area over the last decade.16–19
cellulose
solubility,
wt%
t
°C
[C2C1im][OAc]
[C2C1im][DEP]
[DBUEt][DEP]*
[DBUH][OAc]
[DBNH][OAc]
[DBNEt][DEP]*
[C2C1im][Hex]
RT liquid 215
RT liquid 260
RT liquid 246
35.2
23.0
23.0
22.5
22.0
21.9
16.1
5.5
231
240
246
56 °C
61°C
178
182
245
230
To overcome the economic and environmental issues of the
processing of cellulose, an environmentally benign and low
energy method to manufacture high performance natural
polymer fibres using solvents known as ionic liquids (ILs) has
been developed.20 However, most ILs are quite expensive and
have some other drawbacks limiting their application as
cellulose solvents, such as the fact that they can react with
cellulose, forming a side product.21–23 Some work has been
undertaken to optimise the existing IL technology and the
possibility of IL recovery and recycling by reducing the rate of
adduct accumulation when dissolving cellulose in carboxylate
ILs by incorporation of small mole fractions of a protic co-
RT liquid 207
RT liquid 223
244
-
-
-
-
-
-
-
-
[DBNMe][DMP] RT liquid 266
[DBNH][But]*
[TMGH][OAc]
[DBUH][Prop]
RT liquid
98°C
-
5.3
150
5.1 (110 °C)
4.3
RT liquid 166
[DBNEt][OAc]* RT liquid 183
2.1
solvent.24 Another promising work included
a study of
[DBNH][Hex]*
[DBUH][Hex]*
RT liquid 171
RT liquid 200
0.8
“greener” and “cheaper” morpholinium-based ILs capable of
producing highly concentrated cellulose solutions at 80-120 °C
and films; the production of fibres from these ILs has not yet
been possible.25–27 Recently, distillable SILs have been
developed and used for cellulose dissolution and the
production of high performance cellulose fibres. A new
process employing a superbase derived acetate IL, the fibres
from which are called IONCELL-F, has been developed.28–30 The
ionic liquid 1,5-diazabicyclo[4.3.0]non-5-enium acetate
[DBNH][OAc] allowed the production of highly oriented
cellulose fibres with superior tenacity, exceeding that of
commercial viscose and NMMO-based Lyocell fibers.28 While
[DBNH][OAc] can be recovered from the water by means of
distillation,31 its low thermal stability can affect the cellulose
dissolution and spinning processes, which are conducted at
elevated temperatures of 60-90 oC.29
0.7
solid until
157 °C
[TBDH][OAc]*
157 °C
235
-
* newly reported SILs; - not determined
use as textiles and potential carbon fibre precursors.
We studied the thermal stabilities of our SILs by TGA and
compared these to imidazolium based ILs with corresponding
anions. Fibres were spun from the solutions of SILs and
imidazolium based analogues under similar conditions and
their properties were compared. The obtained information
helped to
identify the most promising ILs for the production of cellulose
fibres on an industrial scale. We identified the potential of
these ILs to produce stable spinning solutions, and
characterised their mechanical properties. Following this we
processed them into carbon fibres. Post-carbonisation we
Our research is focused on exploring further the potential of
SILs, synthesis of novel cellulose dissolving ILs and the
development of sustainable and high performance fibres for
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 3 of 10
View Article Online
DOI: 10.1039/C7GC02671D
Journal Name
ARTICLE
investigated their microstructures using Raman spectroscopy. behaviour of cellulose dissolution (without quantification of
Progress has been made on the possibility to improve cellulose solubility) in 1,5-diazabicyclo[4.3.0]non-5-enium
currently applied industrial processes by substituting common propionate ([DBNH][Prop])
solvents with SILs for cellulose fibre processing, and diazabicyclo[4.3.0]non-5-enium
and
N-methyl-1,5-
phosphate
dimethyl
subsequent carbonisation. The advantages and possible ([DBNMe][DMP]) with [C2C1im][OAc]/water mixtures was
drawbacks of our materials choices are also discussed.
reported.40 Similarly the potential of the pretreatment of
cellulose for hydrolysis in [DBNH][OAC] and [TMGH][OAc] in
comparison to [C2C1im][OAc] was investigated, concluding no
difference in enzyme compatibility.41 To the authors’
knowledge the mechanical properties of fibres obtained using
[DBNH][OAc] have not been considered to date in relation to
their applicability for carbon fibres production.
Results and discussion
Cellulose solubility
Among the studied SILs, preference was given to ones that are
liquid at room temperature, exhibit a good thermal stability
and a high cellulose solubility. Cations of the studied SILs are
presented in Fig. 1. Additionally, 1-ethyl-3-methylimidazolium
acetate ([C2C1im][OAc]) and 1-ethyl-3-methylimidazolium
diethylphosphate ([C2C1im][DEP]) were studied to compare the
performance of SILs against imidazolium based ILs.
We decided to extend the research into the promising SILs
based on DBN, DBU, TMG superbases and also to synthesise
novel SIL based on the 1,5,7-triazabicyclo [4.4.0] dec-5-ene
(TBD) superbase.
We hypothesised that the hydrophilic cellulose fibrils could be
aligned with long chain carboxylic anions to produce better
structured highly oriented fibres. This hypothesis was explored
by synthesising and application of propanoate ([Prop]-), and
novel butanoate ([But]-) and hexanoate ([Hex]-) SILs with DBU
and DBN-based cations.
[C2C1im][OAc] is a known cellulose solvent which is reported to
be able to produce a spinnable cellulose solution with up to
15.8% cellulose at 85 °C.32 Promising results on low energy
solubilisation of microcrystalline cellulose (up to 10 wt% in
under 1 h in the temperature range 45 – 65 °C) were obtained
for [C2C1im]+-based ILs with dimethyl phosphate, methyl
methylphosphonate and methyl phosphonate anions.33 Recent
studies showed that [C2C1im][DEP] is a promising cellulose
solvent34 and allows the production of solutions with
concentrations of cellulose in the range 12-14% (at 100 °C)34
and fibres with good mechanical properties.35 Therefore it is of
interest to study the replacement of the acetate anion with
phosphate-based ones and its influence on cellulose reactivity
and formation of the side products.
Out of six studied acetate SILs studied here, only [DBNH][OAc]
and [DBUH][OAc] were able to dissolve cellulose in amounts
suitable for producing cellulose solutions of concentrations
substantially higher than 5 wt% at 80 °C (Table 1). Therefore,
the SILs with [DBNH]+ and [DBUH]+ cations were chosen for
further study.
Data in Table 1 also show that the solubility of cellulose
decreases with the elongation of the SIL’s anion chain length.
Thus, [C2C1im][Hex] was much less efficient in cellulose
dissolution than [C2C1im][OAc]. The replacement of the
acetate anion with propanoate, butanoate or hexanoate
anions resulted in SILs which were unable to dissolve cellulose
at concentrations that allow the spinning of fibres.
It is known that cellulose solubility is mainly determined by the
structure of the anion of the IL, with some influence of its
cation.36 Some acetate SILs have been already synthesised and
proven to be effective and easily recycled cellulose solvents for
manufacturing of high-performance fibres.29,30 15 and 17 %
cellulose solutions in [DBNH][OAc] were spun (Ioncell-F
process) and the obtained fibres showed high tenacities
(above 50 cN/tex) and high initial modulus of up to 34 GPa
which is comparable to the highest reported values for NMMO
Lyocell fibres of 61 cN/tex and 29.3 GPa, respectively.29 It was
also reported, without quantification however, that acetates
and propionates of the bases studied in this work are capable
to dissolve cellulose: 1-ethyl-3-methyl imidazolium (DBN), 1,8-
However, when these anions were replaced with
diethylphosphate, the cellulose dissolution ability of the SILs
was even better than for the corresponding acetate SILs. This
resulted in the following trend:
[DBNEt][DEP]>[DBNH][OAc]>>[DBNEt][OAc], indicating that
the structures of both the anion and the cation are important
for cellulose dissolution in SILs. The acetate SIL with the
[DBNEt]+ cation with an ethyl side chain showed a very low
cellulose solubility in comparison to the acetate SIL with the
protonated [DBNH]+ cation. The same [DBNEt]+ cation with the
[DEP]- anion outperformed the protonated [DBNH][OAc].
However, this was not the case with the dimethylphosphate
anion and the methyl substituted [DBNMe]+ cation, which
diazabicyclo[5.4.0]undec-7-ene
(DBU),
and
1,1,3,3-
tetramethylguanidine (TMG).33 Furthermore, high cellulose
solubility was reported in other work for propionate SILS with
[DBNH]-, [DBUH]- and [TMGH]- cations in range of 15-18 wt%.37
Additionally a few TMG-based SILs were studied, with only
carboxylic TMG-based SILs being found to be cellulose
dissolving and able to rapidly dissolve microcrystalline
cellulose (5 %, 18h, 100 °C) as well as being “distillable” to over
99% of purity and over 99% yield.38 [TMGH][OAc] requires
higher processing temperatures compared to NMMO-H2O,
[C2C1im][OAc] or [DBNH][OAc] solutions and the obtained
fibres were reported to be of poor quality.39 Separately similar
despite possessing
a very high thermal decomposition
temperature (Tonset), showed relatively poor cellulose
solubility; as low as 5.5 wt%. Six ILs, including two imidazolium-
,
two DBN- and two DBU-derived with acetate and
diethylphosphate anions with the highest maximum cellulose
solubility were chosen for further study.
We compared Kamlet-Taft parameters of chosen ILs. These
parameters are commonly used to describe the solvation
properties of a solvent.42,43 It is known that the ability of a
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 4 of 10
View Article Online
DOI: 10.1039/C7GC02671D
ARTICLE
Journal Name
Tonset values collected in Table 1 give the first approximation of
the thermal stability of ILs, but tend to overestimate their
solvent to dissolve cellulose can be correlated to the solvent
basicity (β)44 and its relation to the solvent acidity – “net
Ionic liquid
π*
α
β
β-α
[C2C1im][OAc]
[C2C1im][DEP]
1.01
1.37
0.49
0.19
1.09
1.03
0.60
0.85
[DBUEt][DEP]
[DBUH][OAc]
[DBNH][OAc]
1.91
1.77
1.01
0.54
0.56
0.62
1.17
1.05
1.17
0.64
0.48
0.55
[DBNEt][DEP]
[DBNEt][OAc]
1.85
1.35
0.65
0.42
1.11
1.19
0.45
0.77
basicity” (β-α). Table
2
summarises the Kamlet-Taft
parameters for the selected ILs.
We compared these parameters with those of [DBNEt][OAc], a
poor cellulose solvent. Experimentally it has been found that
to dissolve cellulose, ILs require β-values above 0.80 and “net
basicity” in the range 0.35-0.90.43 As can be seen from Table 2,
that all of our ILs chosen for further study satisfy these
requirements. However, [DBNEt][OAc], analysed here for
comparison showed β = 1.19. This could misleadingly suggest
that there may be an upper limit to the basicity required for a
good cellulose solvent, which is lower than 1.19.
long-term thermal stability. Therefore, the T0.01/10 already
reported for [C2C1im][OAc], and found to be 102 °C,45 was also
calculated for [C2C1im][DEP] and the SILs. The T0.01/10
parameter corresponds to the temperature at which 1%
weight loss will occur in 10h. From the data illustrated in
Supplementary Information Fig. S1 it was estimated that
T
0.01/10 for [DBNH][OAc] is 38 °C.
Bearing in mind that the melting temperature of [DBNH][OAc]
of 61 °C one could conclude that this SIL degrades even when
solid at a temperature slightly higher than room temperature.
The t0.99 vs temperature plots for other ILs are presented in
Supplementary Information (Fig. S2) and corresponding T0.01/10
values tabulated in Table 3. Our TGA analyses conducted at 80
°C showed that both [DBUEt][DEP] and [DBNEt][DEP] showed
good thermal stability, comparable to those of imidazolium
based ILs (Fig. S3a). The addition of cellulose generally
improves the thermal stability of all studied ILs, apart from
[DBUH][OAc] (Fig. S3b). This is possibly due to a specific
interactions between cellulose and [DBUH][OAc] resulting in
the production of non-stable side products at lower
temperatures. Overall, these results mean that the solutions of
[DBNH][OAc] and [DBUH][OAc] should not be stored at
elevated temperatures or heated for prolonged times when
large batches are processed. With regards to cellulose
degradation upon dissolution and regeneration our analysis
showed that these ILs degraded cellulose to approximately the
same level. The original cellulose polymerisation degree (DPw)
was measured as 263 and slightly decreased upon dissolution
and precipitation of the cellulose (Table 1, far right column).
However, another good cellulose solvent [C4C1im][OAc] has a β
= 1.2043 and only solvents with higher basicities such as 1.28
and 1.34 as for tributylphosphonium and trihexilphosphonium
dimethylphosphates
([P444][DMP]
and
[P666][DMP]),
respectively, are considered as explicitly non-solvents for
cellulose.43
Parviainen et al. have used Kamlet-Taft parameters to predict
cellulose solubility in a range of other SILs. Their work
concluded that [DBNH][Prop] and [TMGH][Prop] which have
Kamlet-Taft parameters close to those of [DBNH][OAc]
reported here, to be the most promising SIL for cellulose
dissolution.37
Viscosity of ILs and cellulose solutions
Among the six selected ILs with high cellulose solubility all,
apart from [DBNH][OAc] and [DBUH][OAc] are liquids at room
temperature; however their viscosities are very different. The
viscosity of the spinning solution is one of the most important
factors for fibre processing. The viscosity of the cellulose
solutions increases with an increased cellulose concentration.
However the viscosity of an IL has no correlation to its
dissolution capability, but rather to the speed of cellulose
dissolution.37
Table 3. Temperature of 1% weight loss of IL occurs within 10h (T0.01/10).
Fig. 2. shows the dependence of solution viscosity on the
o
cellulose content at the temperature of spinning, 80 C. At a
Ionic liquid
T0.01/10, oC
10245
high cellulose content the [C2C1im][OAc] solution remained
less viscous, while the [DBNH][OAc] solution showed the
highest viscosity. Diethyl phosphate ILs solutions up to 10 wt%
of cellulose possessed viscosities values very close to those for
[C2C1im][OAc] solutions, but their viscosities increase at higher
concentrations.
[C2C1im][OAc]
[C2C1im][DEP]
[DBUEt][DEP]
[DBUH][OAc]
[DBNH][OAc]
[DBNEt][DEP]
96
130
67
38
56
Thermal stability of ILs and cellulose solutions
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 5 of 10
View Article Online
DOI: 10.1039/C7GC02671D
Journal Name
ARTICLE
(from 78.9 to 209.9 MPa) of the obtained fibres. The obtained
tensile modulus is comparable to 15 GPa previously reported
for viscose fibres.46
Table 4. Mechanical properties of the produced fibres. Values in brackets are
standard deviations from the mean.
IL
Young’s
Modulus
(GPa)
Tensile
Strength
(MPa)
Diameter
(µm)
14.9 (±5.3) 250 (±50.3)
51.3 (±2.8)
[C2C1im][OAc]
20 (±2.3)
7.5 (±1.4)
5.3 (±0.8)
4.6 (±1.1)
281.5 (±30.5) 24.4 (±1.2)
[C2C1im][DEP]
[DBNEt][DEP]
97.3 (±21.6)
81 (±8.5)
45.8 (±2.1)
39.9 (±1.5)
38.3 (±1.2)
[DBUEt][DEP]
[DBNH][OAc]
78.9 (±9.0)
[DBNH][OAc]
[Hex]
14.4 (±1.8) 209.9 (±26.1) 28.8 (±2.3)
[C2C1im][OAc][ 5.6 (±0.6)
Hex]
80.2 (±7.2)
38.1(±3.0)
Perhaps in this IL mixture, the [OAc]- anion is responsible for
the dissolution of cellulose, while long chain [Hex]- affects the
alignment of the cellulose chains. Fibres produced from the
mixture of SILs are the most promising for textile applications
among all SIL-based fibres studied in this work. To investigate
this effect of elongation of the anion chain length on
imidazolium based ILs, we blended [C2C1im][Hex] and
[C2C1im][OAc] in the same ratio of 1:9 to produce
[C2C1im][Hex][OAc]. Surprisingly, the mechanical properties of
the cellulose fibres produced from [C2C1im][Hex][OAc] were
poorer when compared to the ones produced from pure
[C2C1im][OAc] solution and the blended fibre structure was the
roughest among all studied fibres (Fig. 3). These blended fibres
showed the smallest Young’s modulus and tensile strength
among fibres obtained from imidazolium-based ILs (Table 4).
While the obtained values for [C2C1im][OAc][Hex] were
comparable to those for other SILs, [DBNH][OAc][Hex] has
clearly outperformed the imidazolium-based blend. Perhaps
imidazolium cation, known to affect cellulose dissolution via
hydrogen bonding interactions,47–49 interfere with the [Hex]-
anion alignment of the cellulose chains.
Cellulose fibres characterisation
These six ILs ([C2C1im][OAc], [C2C1im][DEP], [DBUEt][DEP],
[DBUH][OAc] and [DBNH][OAc]) were used to produce 4 wt%
cellulose spinning solutions, from which fibres were spun.
Additionally, to study the influence of the length of the anion
chain of ILs on properties of the produced cellulose fibres,
fibres were spun from a mixture of 10 wt% (12.65 mol%)
[DBNH][Hex] in [DBNH][OAc] (further abbreviated as
[DBNH][Hex][OAc]). The 1:9 ratio between these SILs, one poor
and the other a good cellulose solvent, allowed the production
of a stable spinning solution.
The fibres’ diameters were measured using SEM. Five samples
from each fibre were tested at five different sections and the
mean measured value was taken to calculate the mean cross-
sectional area, assuming a circular cross section. Fig. 3 shows
SEM images of the fibres’ cross sections. The mean values of
the diameters measured are given in Table 4.
From Fig. 4 it can be seen that the fibres spun from
[C2C1im][DEP], [C2C1im][OAc] and [DBNH][OAc][Hex] exhibit a
good combination of strength, stiffness and relatively high
A slight variation in the diameters of the fibres was observed.
It was found that the cellulose fibres spun from [DBUH][OAc]
were not completely consolidated after coagulation and were
merged/fused with each other during winding. This made it
difficult to separate
a single filament and measure an
individual fibre diameter. Hence, mechanical testing was not
performed on these fibres. It is clear from Table 4 and Fig. 4
that fibres spun from [C2C1im][DEP] possess the highest
Young’s modulus of 20 (±2.3) GPa and a tensile strength of
281.5 (±30.5) MPa.
The replacement of an acetate anion with a diethyl phosphate
anion for the DBN derived SIL resulted in a slight increase in
both Young’s modulus and tensile strength of the produced
fibres.
Interestingly, the addition of [DBNH][Hex] to [DBNH][OAc],
resulting in [DBNH][OAc][Hex], significantly improved the
tensile modulus (from 4.6 to 14.4 GPa) and the tensile strength
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 6 of 10
View Article Online
DOI: 10.1039/C7GC02671D
ARTICLE
Journal Name
strain to failure. These properties are together attractive for on crystallinity and ordering. The carbon fibres derived from
applications in textiles and fibre reinforcement in renewable [C2C1im][OAc][Hex] exhibit slightly sharper and narrower D and
composites. However, the known reactivity of [C2C1im][OAc] G peaks indicating higher crystallinity and structural order than
with cellulose has prevented this ionic liquid from giving the rest of the fibres.
commercially spinnable fibres.24 Our studies confirmed a The low ID/IG ratio, indicates a higher G peak intensity
reactivity similar to [C2C1im][OAc] for [C2C1im][Hex] and corresponding to the formation of more structurally ordered
[C2C1im][DEP] (but not DBN- or DBU-based ILs) towards fibres; this decreases according to the trend:
cellulose (Supplementary Information, Fig. S3). Therefore, the [C2C1im][OAc] > [DBNH][OAc] ≈ [DBNEt][DEP] > [C2C1im][DEP] >
fibres produced from [DBNH][OAc][Hex] are of most interest [DBUEt][DEP]
>
[DBUH][OAc]
>
[C2C1im][OAc][Hex]
>
for future exploration in cellulose fibre processing.
[DBNH][OAc][Hex].
Carbonisation of cellulose fibres
Further work would be required to obtain fibres at higher
temperatures (>1200 °C) to follow the complete graphitization
of these materials.
All studied fibres were subjected to pyrolysis to produce
carbon filaments. We employed Raman spectroscopy as a
useful tool to characterise the microstructure of carbon fibres.
As shown in Fig. 5, two characteristic bands were observed,
which are the D-band located at ~1340 cm-1 and the G-band
located at ~1590 cm-1.
Conclusions
In this work we determined that SILs derived from DBU and
DBN with acetate and diethyl phosphate anions were
promising cellulose solvents, reaching cellulose solubilities as
high as their imidazolium analogues. Based on the maximum
cellulose solubility and thermal properties of these ILs, we
have determined six ILs as promising solvents for cellulose
processing: [C2C1im][OAc], [C2C1im][DEP], [DBUEt][DEP],
[DBUH][OAc] and [DBNH][OAc]. Their performance was
explained by Kamlet-Taft values. The relatively high basicity of
[DBNEt][OAc] (β = 1.19) could be a reason for its poor cellulose
dissolution ability in comparison to other acetate SILs based on
this superbase ([DBNH][OAc] and [DBNEt][DEP]). At a high
cellulose content the [C2C1im][OAc] solution remained less
viscous in comparison to the other ILs. [DBUEt][DEP] showed
viscosity of cellulose solutions similar to [C2C1im][OAc] at
moderate concentrations, and possessed higher thermal
stability; however the cellulose fibres produced from
[DBUEt][DEP] were of low quality and merged with each other.
It was found that the elongation of the carboxylic chain of the
anion led to a decreased cellulose dissolution ability. However,
the introduction of 10 wt% of the hexenoate SIL to
[DBNH][OAc]
The intensity ratio between the D and G bands (ID/IG) is widely
used to give information on the structure and crystallinity of
different forms of carbon materials.50 The G band arises from
an in-plane bond stretching motion of sp2-hybridized carbon
atoms and the D band is related to the breathing mode of the
aromatic ring near the basal edge.51 The third main
characteristic band at ~2700 cm−1
(G’ or 2D band) was not
observed for any of the fibres due to the absence of a highly
graphitic structure, which typically starts to evolve at
temperatures above 1200 °C.18 The D band can be correlated
to structural disorder in the fibres. This D band only appears in
the presence of disorder in carbon fibres.52
As shown in Table 5, the ID/IG ratios for all carbon fibres are
consistent with the values observed for other carbon fibres
carbonised from regenerated cellulose filaments.18,19 Carbon
fibres produced from cellulose solutions in [C2C1im][OAc][Hex]
and [DBNH][OAc][Hex] exhibit slightly lower ID/IG ratios than
other fibres that may correspond to the formation of more
crystalline structures.18,19
The full width half maximum (FWHM) values of both peaks are
also tabulated in Table 5 and these values provide information
Table 5. ID/IG ratios and Full Width Half Maxima (FWHM) values of the D and G
IL
Raman bands.
ID/IG
FWHM-D
band
FWHM-G
band
[C2C1im][OAc]
1.17
1.10
1.13
231
210
187
76
69
61
[C2C1im][DEP]
[DBNH][OAc]
[DBUH][OAc]
1.05
1.08
1.13
191
229
206
71
75
77
[DBUEt][DEP]
[DBNEt][DEP]
[DBNH][OAc][Hex]
[C2C1im][OAc][Hex]
1.02
1.04
188
172
73
67
resulted in stronger fibres and carbon fibres with a higher
order. Overall, fibres spun from the [DBNH][OAc][Hex] mixture
have mechanical properties similar to those obtained from
[C2C1im][OAc] and produce higher structurally ordered fibres
when carbonised. We hypothesise that the introduction of
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 7 of 10
View Article Online
DOI: 10.1039/C7GC02671D
Journal Name
ARTICLE
hexanoate anion assists the alignment of the cellulose chains were calculated from the absorption maxima as described in
and therefore produced fibres possessing higher mechanical the literature.54
properties. This principle did not work with the imidazolium Viscosity measurements were performed on a TA instruments
analogue [C2C1im][OAc][Hex]. Cellulose fibres produced from ‘AR2000ex’ rheometer fitted with a Peltier plate at 25 °C, using
the solution in [C2C1im][OAc][Hex] had a rough surface and a 40 mm, 2° steel cone. Measurements were performed at an
moderate mechanical properties. These fibres when angular velocity between 0.1 and 10 rad s-1 under a nitrogen
carbonised however gave highly ordered carbon fibers with atmosphere.
the second highest ID/IG value for the fibres studied in this Spinning solutions were prepared upon dissolution of 4 wt% of
work.
cellulose pulp in the ILs for 4h at 80 °C with a consequent
Therefore here we determined certain similarities and degassing of the obtained solutions for 2h in vacuum oven at
differences in the performance of imidazolium and superbase 80 °C.
based ILs for cellulose processing into textile fibres and carbon Cellulose fibres were spun at 200 rpm using a 340 mm
fibres and defined [DBNH][OAc][Hex] as the most promising diameter needle with a 2 mL min-1 extrusion rate. Wet
cellulose solvent among those studied here.
extrusion was performed with a Nexus 6000 syringe Pump,
with a draw ratio of 6. Water was used as coagulation bath and
an air gap of 3 mm was used between the water bath and the
needle. The fibres were kept in water to remove the residual
ionic liquid for two days and were left to dry at room
temperature for another two days.
Experimental section
Materials. Cellulose was purchased from Sigma-Aldrich. Ionic
liquids were synthesised by acid-base neutralisation from
corresponding bases and acids. The procedure is described in
detail in the Supplementary Information.
The fibres’ diameters were measured using a SEM-TM3030
Plus Table Microscope with a Deben Sprite Multi Axis Stage
Controller. Five samples from each fibre were tested from five
different circular cross-sections to measure the diameter, and
the mean value was taken to calculate the cross-sectional area
of the fibres.
Thermogravimetric Analysis (TGA) experiments were
performed on a PerkinElmer ‘Pyris 1 TGA’ analyser in the range
o
o
30 – 500 C with a ramping rate of 10 C min-1 and a nitrogen
flow of 20 ml min-1.
The tensile strength of the fibres was tested using a single fibre
testing machine from Dia-Stron Ltd. A 20N load cell was used
with a stretch rate of 2 mm min-1. The gauge length was set as
2 cm. Stress-strain curves were obtained from this single fibre
test for 10 samples from each fibre type. The mean cross-
sectional areas for each type of cellulose fibre calculated from
the mean diameter readings obtained from SEM testing were
used to calculate the tensile strength of the fibres. The Young’s
modulus were taken from the most linear portion of the
stress-strain curves.
Maximum cellulose solubility was determined by combining
the IL and an excess of pulp, stirring the mixture with an
overhead stirrer at 80 °C for
4 hours, followed by
centrifugation of undissolved cellulose pulp. The
homogeneous top phase obtained after centrifugation was
weighed and the cellulose extracted with ethanol and dried.
The ratio of the mass of the obtained dry cellulose to the mass
of its homogeneous solution gives the maximum cellulose
solubility in this IL. Obtained solutions however, were in a form
of a very high viscous gel, and therefore much lower
concentrations were prepared for further analysis and
spinning.
As the first step of the carbonisation procedure, stabilisation of
the precursor is required for decomposition and
rearrangement of the polymer in an oxidative atmosphere.
Tang and Bacon14,15 have shown that thermogravimetric
analysis (TGA) of cellulose fibres performed in air exhibited
significant pyrolytic degradation initiating at 240 °C. Therefore,
stabilization of our fibres was carried out at this temperature.
An Elite horizontal tube furnace (TSH16/50/180) was used for
both stabilisation and carbonisation. The cellulose fibres were
placed in a ceramic boat and inserted into the tube. These
fibres were initially stabilized by heating to 240 °C in air at a
rate of 5 °C min-1, followed by a 30 min isotherm at the final
maximum temperature. As the next step of the process, these
stabilised fibres were then carbonised by heating to 1000 °C in
an Argon atmosphere at a rate of 10 °C min-1, followed by a 60
min isotherm at the final maximum temperature.
Cellulose was dissolved at 80 °C using a hotplate and an oil
bath. Stirring was performed using a mechanical overhead
stirrer. Homogeneity of the solutions was confirmed using
optical microscopy.
Polymerisation degree (DPw) of cellulose was measured for
neat cellulose and cellulose recovered from its 10 wt% solution
in ionic liquid by Gel Permeation Chromatography (GPC) in a
lithium chloride/dimethylacetamide (LiCl/DMA) organic mobile
phase, as described by Potthast et al.53
Kamlet–Taft parameters (α, β and π*) were measured with the
application of the solvatochromic dyes: (2,6-dichloro-4-(2,4,6-
triphenyl-1-pyridinio)phenolate (Reichardt’s dye 33, from
Fluka), 4-nitroaniline (from Sigma-Aldrich), and N,N-diethyl-4-
nitroaniline (from Sigma-Aldrich). Stock solutions of each dye
in dichloromethane (DCM) were prepared. Equal volumes of
the stock solutions were combined with ionic liquid, then the
DCM was evaporated. The dye concentration was chosen to
produce spectra with an absorbance between 0.10 and 0.25.
UV-Vis measurements were carried out in quartz cuvettes with
a 10 mm path length at 25 °C. The Kamlet–Taft parameters
Raman spectra of carbonised fibres were obtained using a
Renishaw 1000 Raman imaging microscope equipped with a
thermoelectrically cooled CCD detector.
A laser with a
wavelength of 532 nm was used to record spectra from fibres
using an exposure time of 10 s and two accumulations. The
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 8 of 10
View Article Online
DOI: 10.1039/C7GC02671D
ARTICLE
Journal Name
laser intensity was reduced to 10 % and the beam was focused 24
M. T. Clough, J. A. Griffith, O. Kuzmina and T. Welton,
Green Chem., 2016, 18, 3758.
using a x50 objective lens onto one or a few carbon fibres.
25
R. Liu, Z. Chen, H. Ren and E. Duan, BioResources, 2017, 12
,
5407.
Conflicts of interest
26
H. wei Ren, Q. hong Wang, S. hui Guo, D. shun Zhao and C.
mao Chen, Eur. Polym. J., 2017, 92, 204.
There are no conflicts of interest to declare.
27
D. G. Raut, O. Sundman, W. Su, P. Virtanen, Y. Sugano, K.
Kordas and J. Mikkola, Carbohydr. Polym., 2015, 130, 18.
A. Michud, M. Tanttu, S. Asaadi, Y. Ma, E. Netti, P.
Kaariainen, A. Persson, A. Berntsson, M. Hummel and H.
Sixta, Text. Res. J., 2016, 86, 543.
28
Acknowledgements
We would like to thank the EPSRC grant number
EP/L017679/01 for financial support (AERO RB1717).
29
H. Sixta, A. Michud, L. Hauru, S. Asaadi, Y. Ma, A. W. T.
King, I. Kilpeläinen, M. Hummel, Y. Ma, C. Cross, E. J. Bevan
and C. Beadle, Nord. Pulp Pap. Res. Journa, 2015, 30, 43.
A. Parviainen, R. M. Wahlström, U. Liimatainen, T. Liitiä, S.
Notes and references
30
1
A. R. Horrocks and S. C. Anand, Handbook of technical
textiles, Woodhead Publishing Limited, Cambridge, 2000.
Kirk-Othmer., Kirk-Othmer Encyclopedia of Chemical
Technology, John Wiley and Sons, 5th edn., 2007.
F. E. Bartell and H. Cowling, Ind. Eng. Chem., 1942, 34, 607.
A. R. Bunsell, Handbook of Tensile Properties of Textile and
Technical Fibres, Woodhead Publishing Limited, New York,
2009.
Rovio and J. K. J. Helminen, RSC Adv., 2015, 5, 69728.
31
W. Ahmad, A. Ostonen, K. Jakobsson, P. Uusi-Kyyny, V.
Alopaeus, U. Hyväkkö and A. W. T. King, Chem. Eng. Res.
Des., 2016, 114, 287.
2
3
4
32
33
B. Kosan, C. Michels and F. Meister, Cellulose, 2008, 15, 59.
Y. Fukaya, K. Hayashi, M. Wada and H. Ohno, Green Chem.,
2008, 10, 44.
34
35
J. Vitz, T. Erdmenger, C. Haensch and U. S. Schubert, Green
Chem., 2009, 11, 417.
5
6
7
S. P. Mishra, 2000.
A. Carmichael, Text. World, 2015, 1, 1.
C. Zhu, R. M. Richardson, K. Potter, A. Koutsomitopoulou, J.
S. Van Duijneveldt, S. R. Vincent, N. D. Wanasekara, S. J.
Eichhorn and S. S. Rahatekar, ACS Sustain. Chem. Eng.,
Fiber Economics Bureau, Worldwide Manufactured Fiber
Production Synthetic vs Cellulosic, Arlington, 2014.
M. Terbojevich, A. Cosani, G. Conio, A. Ciferri and E.
Bianchii, Macromolecules, 1985, 646, 640.
C. L. Mccormick, P. A. Callais and B. H. J. Hutchinson,
Macromolecues, 1985, 18, 2394.
8
2016, 4, 454.
36
37
A. Brandt, J. Gräsvik, J. P. Hallett and T. Welton, Green
Chem., 2013, 15, 550.
9
A. Parviainen, A. W. T. King, I. Mutikainen, M. Hummel, C.
Selg, L. K. J. Hauru, H. Sixta and I. Kilpeläinen,
10
11
12
13
J. Massont and R. S. J. Manley, Macromolecules, 1991, 24
5914.
,
ChemSusChem, 2013, 6, 2161.
T. Rosenau, A. Potthast, H. Sixta and P. Kosma, Prog.
Polym. Sci., 2001, 26, 1763.
38
39
40
A. W. T. King, J. Asikkala, I. Mutikainen, P. Järvi and I.
Kilpeläinen, Angew. Chem. Int. Ed. Engl., 2011, 50, 6301.
L. K. J. Hauru, M. Hummel, K. Nieminen, A. Michud and H.
Sixta, Soft Matter, 2016, 12, 1487.
S.-J. Park and G.-Y. Heo, in Carbon Fibers, Springer, New Yo,
2015, pp. 31–66.
E. Frank, L. M. Steudle, D. Ingildeev, J. M. Sp??rl and M. R.
Buchmeiser, Angew. Chemie - Int. Ed., 2014, 53, 5262.
H. Parviainen, A. Parviainen, T. Virtanen, I. Kilpeläinen, P.
Ahvenainen, R. Serimaa, S. Grönqvist, T. Maloney and S. L.
Maunu, Carbohydr. Polym., 2014, 113, 67.
14
15
16
R. Bacon and M. M. Tang, Carbon N. Y., 1964,
2, 221.
M. Tang and R. Bacon, Carbon N. Y., 1964, , 211.
2
41
42
43
R. Wahlstrom, A. King, A. Parviainen, K. Kruus and A.
A. G. Dumanli and A. H. Windle, J. Mater. Sci., 2012, 47
4236.
,
Suurnakki, RSC Adv., 2013, 3, 20001.
L. Crowhurst, R. Falcone, N. L. Lancaster, V. Llopis-Mestre
and T. Welton, J. Org. Chem., 2006, 71, 8847.
17
18
19
20
21
Y. Kaburagi, K. Hosoya, A. Yoshida and Y. Hishiyama,
Carbon N. Y., 2005, 43, 2817.
L. K. J. Hauru, M. Hummel, A. W. T. King, I. Kilpeläinen and
H. Sixta, Biomacromolecules, 2012, 13, 2896.
K. Kong, L. Deng, I. a. Kinloch, R. J. Young and S. J. Eichhorn,
J. Mater. Sci., 2012, 47, 5402.
44
45
A. C. O’Sullivan, Cellulose, 1997, 4, 173.
A. E. Lewandowska, C. Soutis, L. Savage and S. J. Eichhorn,
Compos. Sci. Technol., 2015, 116, 50.
M. T. Clough, K. Geyer, P. A. Hunt, J. Mertes and T. Welton,
Phys. Chem. Chem. Phys., 2013, 15, 20480.
R. P. Swatloski, S. K. Spear, J. D. Holbrey and R. D. Rogers, J.
Am. Chem. Soc., 2002, 124, 4974.
46
47
48
M. G. Northolt, H. Boerstoel, H. Maatman, R. Huisman, J.
Veurink and H. Elzerman, Polymer (Guildf)., 2001, 42, 8249.
J. Zhang, H. Zhang, J. Wu, J. Zhang, J. Hea and J. Xiang,
Phys. Chem. Chem. Phys., 2010, 12, 1941.
O. Kuzmina and J. P. Hallett, Application, Purification and
Recovery of Ionic Liquids, Elsevier, Amsterdam, 1st edn.,
2016.
J. Zhang, H. Zhang, J. Wu, J. Zhang, J. Hea and J. Xiang,
Phys. Chem. Chem. Phys., 2010, 12, 14829.
22
23
T. Liebert and T. Heinze, BioResources, 2008, 3, 576.
G. Ebner, S. Schiehser, A. Potthast and T. Rosenau,
49
50
B. D. Rabideau, A. Agarwal and A. E. Ismail, 2014.
Tetrahedron Lett., 2008, 49, 7322.
N. Larouche and B. L. Stansfield, Carbon N. Y., 2010, 48
,
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 9 of 10
View Article Online
DOI: 10.1039/C7GC02671D
Journal Name
ARTICLE
620.
51
A. C. Ferrari and J. Robertson, Phys. Rev. B, 2001, 64
,
75414.
52
53
A. C. Ferrari and J. Robertson, Phys. Rev. B, 2000, 61
14095.
,
A. Potthast, S. Radosta, B. Saake, S. Lebioda, T. Heinze, U.
Henniges, A. Isogai, A. Koschella, P. Kosma, T. Rosenau, S.
Schiehser, H. Sixta, M. Strlič, G. Strobin, W. Vorwerg and H.
Wetzel, Cellulose, 2015, 22, 1591.
54
M. A. A. Rani, A. Brant, L. Crowhurst, A. Dolan, M. Lui, N. H.
Hassan and J. P, Phys. Chem. Chem. Phys., 2011, 13, 16831.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Green Chemistry
Page 10 of 10
View Article Online
DOI: 10.1039/C7GC02671D
Superbase ionic liquids for effective cellulose processing from dissolution
to carbonisation.
Olga Kuzmina,* Jyoti Bhardwaj, Sheril Rizal Vincent, Nandula Dasitha Wanasekara, Livia Mariadaria Kalossaka, Jeraime Griffith, Antje
Potthast, Sameer Rahatekar, Stephen James Eichhorn and Tom Welton
Strong cellulose fibres and highly ordered carbon fibres can be produced from solutions in superbase
ionic liquids.