Documente Academic
Documente Profesional
Documente Cultură
PII: S0959-6526(19)34784-5
DOI: https://doi.org/10.1016/j.jclepro.2019.119914
Reference: JCLP 119914
Please cite this article as: Chun-Han Ko, Bing-Yuan Yang, Lang-Dong Lin, Fang-Chih Chang, Wen-
Hua Chen, Impact of pretreatment methods on production of bioethanol and nanocrystalline
cellulose, Journal of Cleaner Production (2019), https://doi.org/10.1016/j.jclepro.2019.119914
This is a PDF file of an article that has undergone enhancements after acceptance, such as the
addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive
version of record. This version will undergo additional copyediting, typesetting and review before it
is published in its final form, but we are providing this version to give early visibility of the article.
Please note that, during the production process, errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
nanocrystalline cellulose
Chen d
10617, Taiwan
b Department of Cultural Heritage Conservation, National Yunlin University of
*Corresponding author.
1
Journal Pre-proof
ABSTRACT
Cellulose, the most abundant terrestrial biopolymer, has immense potential for a range
alkaline and acid pretreatment methods have been developed to recover mono sugars
(NCC) from pretreated Formosan alder biomass was conducted in two stages.
Formosan alder biomass was treated using unbleached kraft (UEK), acid steam
explosion (SEP), and bleached acid steam explosion (BSEP) and hydrolyzed using
cellulase. Filter paper (FP) and alpha-cellulose were used as positive controls.
Bioethanol fermentation using E. coli Ko11 was conducted. Residual biomass was
further recovered to produce NCC. After 96 h enzyme hydrolysis, the highest sugar
yields were 383.4 and 329.84 mg glucose/ g biomass for the FP and UEK samples.
The highest bioethanol fermentation yields were 4.18 and 3.62 g bioethanol/ g
biomass for FP and UEK samples. During enzyme hydrolysis of the biomass,
crystallinity increased. NCC yields were 613.93 nm (68%) and 1,583 nm (73%) for
the FP and UEK samples. The negative impact of lignin content was reflected in the
particle length of the final NCC. Further lignin removal is required for optimal NCC
production. This study showed that Formosan alder biomass could be used to produce
Acid hydrolysis
2
Journal Pre-proof
List of abbreviations
and fermentation
1. Introduction
the use of renewable resources. Directing biomass for various needs (bulk/higher
biofuels, and viable and efficient biorefinery concepts are being ever increasingly
explored for their potentials (Langeveld et al., 2010). Many non-fuel bioproducts (e.g.,
development options for biorefineries that can be used to generate them alongside
biofuels, power, and by-products. Biobased products offer the most development
perspectives, combining small-scale, less endowed farmers and large market volumes
with medium to high price levels (Langeveld et al., 2010). A European Union (EU)
ecosystems, use natural inputs, expend minimum amounts of energy, and do not
3
Journal Pre-proof
produce waste as all materials discarded by one process are inputs for another process
and are reused in the ecosystem (Nebe, 2011). Concepts of the circular economy,
green economy and bioeconomy are joined to reconcile economic, environmental, and
(D'Amato et al., 2017). For a more sustainable economy, renewable biomass such as
crop residues and dedicated energy crops are being increasingly used for the
economy also composes one major challenge faced by the society nowadays.
petro-based resources as a consequence of the population growth, and the finite nature
of oil resources are serving reasons for the transition from the current fossil-based to a
biobased economy (Mussatto, 2017). In the context of circular bioeconomy, the global
bioenergy sector may face challenges and biomass conversion must retain a high
efficiency. Bioenergy can be integrated into the cascade biorefinery and/or offer waste
residual biomass potential in the EU-27 plus Switzerland was conducted estimating
biomass residues stemming from 4 main activities: i) agriculture; ii) forestry residues;
iii) urban greenery management; and iv) agri-industrial food process waste and
municipal biodegradable waste (Hamelin et al., 2019). Esen and Yuksel (2013) also
bio-based systems, such as bioenergy and biorefinery, and their prospects in its
ingredients, fine and specialty chemicals, polymers, alongside biofuel and bioenergy
4
Journal Pre-proof
Lignocelluloses are the most promising alternative energy resources that can be
improve the digestibility of lignocelluloses but also facilitate the recovery of lignin
products (Hassan et al., 2019). The recalcitrant nature of lignocellulosic materials that
include lignin, hemicelluloses and the high crystallinity index (CI) of cellulose are the
involves disrupting and removing the cross-linkages of lignin and hemicelluloses that
cover the cellulose fibers (Jahnavi et al., 2017). The potential for use of
receiving attention (De Corato et al., 2018), with a focus high-value-added products in
glycerin, proteins, peptides, and enzymes (De Corato et al., 2018). Fan et al. (2018)
5
Journal Pre-proof
process, lignocellulosic materials are fractionated into alkali soluble lignin and
hemicellulose and 50% of lignin from wood fiber leading to a more exposed cellulose
non-edible, is highly regarded in the sustainable production of fuels and has social,
economic, and environmental benefits (Liew et al., 2014). There are four main steps
distillation (Mohapatra et al., 2017). Bauli et al. (2019) found that the impact of
enzymatic hydrolysis was lower for selected environmental impact categories. Among
these, pretreatment is the most challenging step for increasing the accessibility to
well-established processes.
celluloses (NCC). Methods for isolating cellulose from wood or agriculture residues
have received a lot of attention (Brinchi et al., 2013). NCC is defined as cellulose in
terms of a nanometer in at least one dimension. There are several ways to isolate NCC
from raw plant material, such as grinding, acid hydrolysis, and steam explosion
(Brinchi et al., 2013). The renewable plant fibers can be used as a source of NCC to
6
Journal Pre-proof
enhance the sustainability of the final composites (Sarasini and Fiore, 2018). NCC
industrial application. When cellulose biomass is treated with an acid, the pulp will
start to degrade. The degradation will start with the most accessible parts of the fiber,
followed by the reducing end groups and the crystal surfaces. The properties of
nanocellulose are affected by time, temperature, acid concentration, and acid to pulp
ratio (Sun et al., 2016a). The types of acid also play a role in the final properties of
NCC, and sulfuric acid is the most commonly used (Liu et al., 2018). Cotton fibers
and bagasse were used to produce NCC through sulfuric acid hydrolysis, and a
decrease in the non-crystalline region was shown during acid hydrolysis (Sun et al.,
2016a). The enzymatic hydrolysis apparatus could process much greater volumes of
NC, optimizing the energy use (Bauli et al., 2019). Formosan alder (Alnus formosana)
land-slide or barren sites and has a wide altitude and latitude range (Liu et al., 1988).
A. formosana has nitrogen fixation ability with the assistance of Rhizobium flora.
Given the abundance of A. formosana, it has the potential for use as a raw material for
reductions in industrial energy, water and utility use (Klemeš et al., 2018). Dorotić et
al. (2019) proposed an integrated carbon neutral energy plan for an island with
optimal utilization of renewable energy sources for power, heating, cooling and
7
Journal Pre-proof
transport sectors. Tomić and Schneider (2018) combined the circular economy and
closing the loop material- and energy-wise and to reduce the embodied energy of
recycled materials. Walmsley et al. (2018) also found that the substantial benefits
could be arised from combining and extending the scope of methods. The orange
(Tsukamoto et al., 2013). These studies show the importance of integrated approaches
solving future problems of energy and material. There are few holistic studies
NCC from pretreated Formosan alder biomass was conducted in two stages.
Formosan alder biomass was pretreated using unbleached kraft (UEK), acid steam
explosion (SEP), and bleached acid steam explosion (BSEP) and then hydrolyzed
using cellulase. Filter paper and alpha-cellulose were used as positive controls. The
Process parameters for the bioethanol and NCC processes were examined. The life
The research framework for this study is shown in Fig. 1. In this study, Formosan
alder biomass (hardwoods) treated by unbleached kraft (UEK), acid steam explosion
(SEP), bleached acid steam explosion (BSEP), filter paper and alpha-cellulose were
hydrolyzed by cellulase in the first step. Hydrolysates were collected for bioethanol
8
Journal Pre-proof
fermentation. Leftover solids after fermentation were collected and subject to acid
Experimental Forest of central Taiwan. The wood chips were subjected to pulping
processes. A batch reactor was loaded with 700 g of oven-dried wood chips with a
liquor-to-wood ratio of 4 (w/v). The cooking liquor contained NaOH and Na2S in
quantities sufficient to provide 25% of sulfidity and 17% active alkali concentration
(w/w as Na2S on wood basis). The cooking temperature ranged from 25 C to 160 C
with a heating ratio of 1.5 C per min. The cooking temperature was maintained
isothermally for 180 mins. The unbleached kraft pulp was denoted UEK.
The SEP was produced from wood chips of A. formosana, dried for 24 h and then
soaked in 1.5% sulfuric acid for 96 h. The optimal conditions for steam explosion
were based on Ko et al. (2012), using a 1:7 solid/liquid ratio, treated at 180oC for 10
mins. A component of the SEP was bleached using glacial acetic acid and sodium
chlorite (NaClO2) to remove the remaining lignin. This treatment was repeated until
101777401), Saint Louis, USA. TOYO Qualitative Filter paper (FP), Grade NO. 1
(Toyo, Tokyo, Japan) and was denoted as FP. All reagents were purchased from
All of the respective samples were oven-dried and crushed using a grinder and
9
Journal Pre-proof
Holocellulose content was measured using TAPPI T9 standards. The ground sample
was treated with sodium chlorite and glacial acetic acid in a hot water bath. After the
sample became white, the suspension was then cooled in an ice bath, filtered, and
washed with distilled water. The cellulose content was determined by TAPPI T203
standards. The ash content of the pulp was determined using a muffle furnace by
Hydrolysis was conducted in a total volume of 200 mL liquid with 0.1 M Tris-buffer
(pH 7), 2.5% (w/v) samples in a 250 mL conical flask. The flasks were water bathed
at 50oC, shaken at 100 rpm for 96 h, and the samples were analyzed every 12 h.
Activities for all the cellulases were measured using the dinitrosalicylic acid.
E. coli KO11 was purchased from the American Type Culture Collection (ATCC).
Before simultaneous saccharification and fermentation (SSF), the E. coli KO11 was
pre-cultured in a liquid medium (with 10 g/L tryptone, 5 g/L yeast extract,10 g/L
(OD600, Jasco V-530, Japan) was raised to approx. 1.4 prior to fermentation.
The flask and samples were autoclaved at 121C for 20 mins. Each flask
contained 0.5% w/v yeast extract,1% w/v tryptone, and 5 mL of E. coli KO11
10
Journal Pre-proof
inoculums with the highest enzyme loading. The flasks were shaken slowly (100 rpm)
at 37C and samples were taken at every 12 h for 120 h and frozen until analysis.
identify the alcohols and mono-sugars at 70C; sulfuric acid (0.0085 N) was used as
Perspective biomass and fermentation residue were treated using sulfuric acid
(65%) to obtain NCC at different conditions. The acid hydrolysis was stopped
immediately using a 7-fold dilution with deionized water. The mixture was
centrifuged at 4C at a speed of 10,000 rpm for 10 mins. The centrifugation was
repeated three times, and the precipitate was then placed in regenerated cellulose
dialysis tubing with a molecular weight cutoff of 12,000 – 14,000 and dialyzed using
solution (around 1-3%) was prepared by dissolving the different kinds of cellulose
magnetic stirring at 110oC in an oil bath until the cellulose had been dissolved
oxidation and degradation during the dissolution of cellulose.The temperature did not
exceed 130oC to prevent degradation of the solution and generated byproducts. After
the cooking process, the samples were rinsed using deionized water owing to remove
the solvent, then freeze-dried and stored for the next experiment.
11
Journal Pre-proof
Particle size distributions of the biomass were measured using MorFi (TECHPAP,
Saint Martin d’Hères, France) at the Chung-Hwa Pulp Corporation in Hualien. More
detailed analysis of these methods is described in earlier work (Ko et al., 2019). The
samples were placed into a 1 L beaker then diluted with 1 L of RO water, to achieve a
uniform 0.3% solution. The beaker and fiber solution was then placed on the holder of
the MorFi analyzer. The particle size of samples after acid hydrolysis was measured
using dynamic light scattering (DLS, SZ-100, HORIBA Scientific, France) in the
range between 0.4 nm and 10 μm, under the following conditions: particle refractive
index 1.60, particle absorption coefficient 0.01, water refractive index 1.33,
temperature 25 C, and general calculation model for irregular particles. Each
voltage of 15 kV. Prior to SEM examination, the samples were freeze-dried and then a
fine layer of gold was deposited on the samples by means of a plasma sputtering
conducted using a powder X-ray diffractometer (D8, Bruker AXS Gmbh, Karlsruhe,
12
Journal Pre-proof
Germany) equipped with a radiation source operating at 40 kV and 40 mA. The XRD
The CI was calculated from the heights of the 200 peak (I200, 2θ= 22.6) and the
intensity minimum between the 200 and 110 peaks (Iam, 2θ= 18) based on Eq. (1).
I200 represents both crystalline and amorphous material, whereas Iam represents the
amorphous material.
bioethanol production (Chang et al., 2017). SSF waste utilization will be suitable for
regions that rely on imported energy resources. A life cycle impact assessment of
after 90 mins hydrolysis. This study used the IMPACT 2002+ model with midpoint
evaluation. This method analyzes four impacts: human health, ecosystem quality,
climate change, and resources. The energy inputs/outputs in terms of machinery usage,
chemicals, and water consumption were considered for different residues. This study
used the inventory data of the built-in database (Ecoinvent) of the SimaPro 8.2 Life
Cycle Assessment (LCA) software which were modified to reflect the actual situation
in Taiwan.
Chemical composition of the biomass is listed in Table 1. The results are based on
and UEK is higher than SEP. The lignin content of SEP is higher than the other four
13
Journal Pre-proof
materials. With respect to the extreme SEP pretreatment, the structures between
cellulose and lignin are degraded, but the lignin is not dissolved. As such, SEP is
slightly. The results for BSEP showed that the bleaching treatment removed the lignin
efficiently, and the total lignin content decreased from 36.35% to 0.26% (Table 1).
Bauli et al. (2019) and Shanmugarajah et al. (2019) also found that the washing, alkali,
bleaching, and acid treatments were effective in the removal of hemicellulose, lignin,
Three dosages of 1.5, 4.5, and 9.0 CMCase/mL were used in the hydrolysis
using the three dosages yielded 140.03, 293.85, and 383.40 mg glucose /g dried
biomass; equivalent to 3.56, 7.47, and 10.63 g glucose /L. After enzymatic hydrolysis,
the remaining solid biomass 68.22, 57.22, and 43.86 (%, w/w), was recovered by
centrifugation and used for NCC production. Hydrolysis of the α-cellulose sample
using the above three dosages yielded 84.25, 195.49, and 241.92 mg glucose /g dried
biomass; equivalent to 2.11, 4.89, and 6.05 g glucose /L. After enzymatic hydrolysis,
the remaining solid biomass 72.31, 62.18, and 51.29 (%, w/w), was recovered by
centrifugation and used for NCC production. Hydrolysis of the UEK sample using the
three dosages yielded 121.11, 255.18, and 329.84 mg glucose /g dried biomass;
equivalent to 3.03, 6.38, and 8.25 g glucose /L. After enzymatic hydrolysis, the
remaining solid biomass 73.42, 57.76, and 45.89 (%, w/w), was recovered by
centrifugation and used for NCC production. Hydrolysis of the SEP sample using the
14
Journal Pre-proof
three dosages yielded 74.91, 171.79, and 209.32 mg glucose/g dried biomass,;
equivalent to 1.87, 4.29, and 5.23 g glucose /L. After enzymatic hydrolysis, the
remaining solid biomass 72.71, 68.22, and 58.31 (%, w/w), was recovered by
centrifugation and used for NCC production. Hydrolysis of the BSEP sample using
the three dosages yielded 98.70, 223.86, and 277.44 mg glucose/g dried biomass;
equivalent to 2.47, 5.60, 6.94 g glucose /L. After enzymatic hydrolysis, remaining
76.10, 62.67, 49.69 (%, w/w) solid biomass, was recovered by centrifugation and
subjected for NCC production. The mass ratio of solid to liquid was 2.5%, (w/w). For
The maximum glucose yield (383.40 mg/g dried biomass) was obtained for FP
while UEK had a glucose yield of 329.84.40 mg/g dried biomass. Biomass with more
At the end of incubation, the samples were centrifuged for 9,000 rpm at 4oC for 20
mins, and then the substrates and supernatant were separated and used for NCC
production.
E. coli KO11 yielded 0.86, 2.10, and 2.52 g ethanol/L. The 96 h fermentation of UEK
hydrolysate by E. coli KO11 yielded 1.28, 2.67, and 3.62 g ethanol/L. The 96 h
fermentation of SEP hydrolysate by E. coli KO11 yielded 0.75, 1.82, and 2.14 g
15
Journal Pre-proof
reduced using low severity acid pretreatment (Davis et al., 2013). In this study, the
concentrations of 5-HMF and furfural inhibitors were below the detection limit. The
Optimal ethanol yields are from FP and UEK hydrolysates (4.18 and 3.62 g
glucose/L) using the highest CMCase dosages. The theoretical maximum ethanol
solution. The lower ethanol yields from α-cellulose and SEP hydrolysates than FP and
UEK ones are due to its high cellulose crystallinity and high lignin contents. Song et
al. (2019) also found that the production of bioethanol from hardwoods is difficult due
to the low level of enzymatic hydrolysis caused by the high lignin contents.
The CIs of five biomass materials were determined by X-ray diffraction; the XRD
pattern of (α-cellulose, FP, and SEP) is shown in Fig. 4. FP has the highest CI,
79.22%. The order of CI was FP > α-cellulose > UEK > BSEP > SEP. The bleaching
process results in more crystal regions in the cellulose and increases the CI from
61.19% for SEP to 65.58% for BSEP. The CI of SEP pulp was lower than UEK
owing to the pretreatments that leave more lignin in the SEP, resulting in lower
cellulose content. The CI values are higher from those obtained using alkaline
16
Journal Pre-proof
(2019). The SEP pretreatment only broke the bonds between the amorphous and
crystalline cellulose but did not remove the lignin. The CI and the presence of lignin
are the main factors which affect the reactivity and accessibility of cellulose (Lai,
2017).
Nanoparticles in water with a higher zeta potential can generate electrostatic repulsion,
then form a stable suspension. All materials had negatively charged surfaces due to
the esterification of hydroxyl groups during the process of acid hydrolysis. As shown
in Fig. 5, the zeta potential value increased with the increased acid hydrolysis duration,
which means that the stability of the cellulose suspensions increased with an increase
Fig. 6 (a) shows the fiber lengths of materials measured using MorFi. The average
fiber lengths of FP, α-cellulose, UEK, SEP, and BSEP were 1.004, 0.486, 0.712,
0.604, and 0.582, respectively. The average fiber length of FP was greater than the
other materials. Fig. 6 (b) shows the fiber widths of materials measured using MorFi.
The average fiber widths of FP, α-cellulose, UEK, SEP, and BSEP were 23.1, 16.8,
18.2, 43.3, and 34.7, respectively. SEP, UEK, and α-cellulose fibers are narrower than
the other materials. Fig. 7 shows the fiber fine lengths of materials measured using
MorFi. The average fiber lengths of FP, α-cellulose, UEK, SEP, and BSEP were 40,
17
Journal Pre-proof
The mass balance for each sample is shown in Table 2. All samples show a
decrease in yield with an increase in hydrolysis duration. On the contrary, the yield of
The yields of UEK and SEP are higher than the other four samples due to the
presence of lignin which cannot be dissolved by acid and thus impedes the hydrolysis
process. Lignin affects not only the yield but also the particle size distribution which
will be discussed in the next section. Among the other three substrates, FP,
α-cellulose, and BSEP, the yield of FP is higher than α-cellulose and BSEP, because
the acid tends to attack the amorphous region, and the higher CI of FP makes
hydrolysis of the cellulose more difficult. The CI and the presence of lignin are the
DLS as shown in Fig. 8(a). The results show that during FP hydrolysis, smaller
cellulosic particles could be obtained. After 4 h acid hydrolysis of FP, the results
show that particle sizes after FP hydrolysis ranged from 82 to 2,511 nm and the
using DLS as shown in Fig. 8(b). The first peak of the particle sizes of α-cellulose is
between 39-105 nm, the second peak is between 118-655 nm; the average is
208.4±16.8 nm. After 1 h of stirring, the reaction became more uniform and the
18
Journal Pre-proof
Changes in particle size distribution during BSEP hydrolysis were measured using
DLS as shown in Fig. 8(c). BSEP has sizes distributed from 72 to 118 nm and the
difference between 0.5-1 h, as the initial solution was viscous which impeded the
By comparing the CI of α-cellulose, FP, and BSEP, it is found that the lower of
the CI was, the higher the efficiency of hydrolysis and the smaller the particle size
was. There is no positive correlation between fiber length and particle size. The above
results show that the influence of the CI of the raw material on particle size is greater
than that of the fiber length. The CI of the raw material is a key factor affecting the
efficiency of hydrolysis.
hydrolyzed cellulose fibers. Upon drying, all materials tend to be self-assembled into
micron width and micrometer length fibers. This may be due to the strong
groups of cellulose during the drying process. The reduction in fiber diameter was
observed across all the fiber samples after acid hydrolysis due to the successful
Table 3 lists the assessment results for the endpoint impact indicators for NCC
from BSEP, FP, and α-cellulose. The results show that the production of NCC from
19
Journal Pre-proof
of dialysis water, acid, and drying. The production of NCC from FP has a lower
mitigate the impacts on human health, climate change, and resource depletion during
the production of NCC, there is a need for improvements to the hazardous processing
phases. As reported by Bauli et al. (2019), the most significant environmental impact
in NCC production relates to the energy expenditure during heating and drying
processes.
raw materials attributed different chemical constitution and cellulose structures which
influenced the effects of acid hydrolysis and varying pretreatments also play an
important role. Morais et al. (2013) reported that cotton linters provide higher purity
and crystallinity during acid hydrolysis and the properties of the NCC product are
similar to those in this study. The main differences in preparation were alkaline and
mechanical pre-treatment. The results of this study show that multiple pretreatments
could facilitate the production of NCC even when using complex raw material.
plant containing less lignin with higher accessibility to acid hydrolysis. The NCC
produced through 75 mins acid hydrolysis of Imperata brasiliensis was similar this
study, indicating that pretreatments could shorten the reaction time and enhance the
Tian et al. (2016) used a similar method for the production of NCC which
combined mechanical and chemical treatments and found that, as in this study,
20
Journal Pre-proof
mechanical pretreatment could shorten the acid hydrolysis time effectively and
reaction time was curtailed to below 40 mins. The final properties of the NCC were
Kargarzadeh et al. (2012) found that the native properties and chemical
composition of the raw materials influence the production of NCC. Kenaf bast is a
type of high-quality fiber with high cellulose content and high proportion of slender
fibers (Palle, 2008). Under similar acid hydrolysis conditions, NCC from kenaf bast
fibers showed higher recovery and more rod-like structures due to the fiber structure
and lower lignin content which act as a barrier during the NCC production. Yields and
sizes of NCC derived from corncob residue (Liu et al., 2016) were similar to that
4 Conclusions
In this study, five materials (α-cellulose, FP, SEP, UEK, and BSEP) were used to
produce NCC using different pretreatments and durations of acid hydrolysis, and the results
were analyzed and compared using DLS, XRD, and SEM techniques. Smaller cellulose
particles can be obtained with increasing hydrolysis time, and the average particle size after
4 h acid hydrolysis is SEP > UEK > FP > α-cellulose > BSEP. After 96 h enzyme
hydrolysis, the highest sugar yields were 383.4 and 329.84 mg glucose/ g biomass for FP
and UEK samples. The highest bioethanol fermentation yields were 4.18 and 3.62 g
bioethanol/ g biomass for FP and UEK samples. NCC yields were 68 and 73% for FP and
UEK samples. NCC yields were 613.93 and 1,583 nm for FP and UEK samples. The
results indicate that the influence of CI of raw materials on the particle size distribution is
greater than that of the fiber length of raw materials, so the CI of raw materials is a key
factor which affects the efficiency of acid hydrolysis. The zeta potential value for all
21
Journal Pre-proof
materials increased with increased acid hydrolysis time because the esterification of
hydroxyl groups during the process of acid hydrolysis would form the negative charges on
the cellulose surface. Lignin constituents in biomass, such as UEK and SEP, would affect
the efficiency of acid hydrolysis. Compared with α-cellulose, FP, SEP, UEK, and BSEP,
the particle size distribution results of UEK and SEP show a wide range. The zeta potential
value of UEK and SEP are lower than the other materials, as particles of UEK and SEP
could not disperse well in solution. The delignification process (to separate the lignin and
cellulose after acid hydrolysis) is an important step during the process of NCC preparation.
Acknowledgements
The financial supports by the Ministry of Science and Technology, Taiwan for the
References
Bauli, C.R., Rocha, D.B., de Oliveira, S.A., Rosa, D.S., 2019. Cellulose
nanostructures from wood waste with low input consumption. J. Clean. Prod.
Brinchi, L., Cotana, F., Fortunati, E., Kenny, J., 2013. Production of nanocrystalline
Chang, F.C., Lin, L.D., Ko, C.H., Hsieh, H.C., Yang, B.Y., Chen, W.H., Hwang,
feedstocks and two fermentation waste reutilization schemes. J. Clean. Prod. 143,
22
Journal Pre-proof
973-979. https://doi.org/10.1016/j.jclepro.2016.12.024.
Chundawat, S.P., Donohoe, B.S., da Costa Sousa, L., Elder, T., Agarwal, U.P., Lu,
F., Ralph, J., Himmel, M.E., Balan, V., Dale, B.E., 2011. Multi-scale
973-984. http://xlink.rsc.org/?DOI=c0ee00574f.
D'Amato, D., Droste, N., Allen, B., Kettunen, M., Lähtinen, K., Korhonen, J.,
Leskinen, P., Matthies, B.D., Toppinen, A., 2017. Green, circular, bioeconomy: a
https://doi.org/10.1016/j.jclepro.2017.09.053.
Davis, R., Tao, L., Tan, E.C.F., Biddy, M.J., Beckham, G., Scarlata, C., Jacobson, J.,
Cafferty, K., Ross, J., Lukas, J., Knorr, D., Schoen, P. 2013. Process design and
https://doi.org/10.2172/1107470.
https://doi.org/10.1016/j.carbpol.2018.03.055.
De Corato, U., De Bari, I., Viola, E., Pugliese, M., 2018. Assessing the main
https://doi.org/10.1016/j.rser.2018.02.041.
23
Journal Pre-proof
Dorotić, H., Doračić, B., Dobravec, V., Pukšec, T., Krajačić, G., Duić, N., 2019
intermittent renewable energy sources. Renew. Sust. Energ. Rev. 99, 109-124.
https://doi.org/10.1016/j.rser.2018.09.033.
Esen, M., Yuksel, T., 2013. Experimental evaluation of using various renewable
https://doi.org/10.1016/j.enbuild.2013.06.018.
Fan, J., Santomauro, F., Budarin, V. L., Whiffin, F., Abeln, F., Chantasuban, T.,
Gore-Lloyd, D., Henk, D., Scott, R.J., Clark, J., Chuck, C.J., 2018. The additive
https://doi.org/10.1016/j.jclepro.2018.07.088.
Hamelin, L., Borzęcka, M., Kozak, M., Pudełko, R., 2019. A spatial approach to
Hassan, S.S., Williams, G.A., Jaiswal, A.K., 2019. Moving towards the second
https://doi.org/10.1016/j.rser.2018.11.041.
Jahnavi, G., Prashanthi, G.S., Sravanthi, K., Rao, L.V., 2017. Status of availability
https://doi.org/10.1016/j.rser.2017.02.018.
Kargarzadeh, H., Ahmad, I., Abdullah, I., Dufresne, A., Zainudin, S.Y., Sheltami,
24
Journal Pre-proof
and thermal stability of cellulose nanocrystals extracted from kenaf bast fibers.
Klemeš, J.J., Varbanov, P.S., Wan Alwi, S.R., Abdul Manan, Z., 2018. Process
Ko, C.H., Wang, Y.N., Chang, F.C., Chen, J.J., Chen, W.H., Hwang, W.S., 2012.
Ko, C.H., Yang, C.Y., Chang, F.C., Lin, L.D., 2019. Effect of Paenibacillus
https://doi.org/10.1016/j.jenvman.2019.03.133.
https://doi.org/10.2135/cropsci2009.09.0529.
Liu, C., Li, B., Du, H., Lv, D., Zhang, Y., Yu, G., Mu, X., Peng, H., 2016.
716-724. https://doi.org/10.1016/j.carbpol.2016.06.025.
Liu, F., Huang, K., Zheng, A., Xiao, F.S., Dai, S., 2018. Hydrophobic solid acids
and their catalytic applications in green and sustainable chemistry. ACS Catal. 8,
372-391. https://doi.org/10.1021/acscatal.7b03369.
Liu, Y., Lu, F., Ou, C., 1988. Trees of Taiwan. Monographic publication 7, 91-92.
25
Journal Pre-proof
1131-1142. https://doi.org/10.1016/j.jclepro.2019.04.362.
Mohan, S.V., Nikhil, G.N., Chiranjeevi, P., Reddy, C.N., Rohit, M.V., Kumar, A.N.,
2-12. https://doi.org/10.1016/j.biortech.2016.03.130.
Pennisetum sp. using Taguchi and artificial neural networking for enhanced
https://doi.org/10.1016/j.jenvman.2016.09.060.
https://doi.org/10.1016/j.carbpol.2012.08.010.
Nebe, S., 2011. Bio-based economy in Europe: state of play and future
26
Journal Pre-proof
Liew, W.H., Hassim, M.H., Ng, D.K., 2014. Review of evolution, technology and
https://doi.org/10.1016/j.jclepro.2014.01.006.
Palle, I., 2008. Carboxymethylation of cellulose from kenaf (hibiscus cannabinus l.)
Sadhukhan, J., Martinez-Hernandez, E., Murphy, R.J., Ng, D.K.S., Hassim, M.H.,
Ng, K.S., Hassim, M.H., Ng, K.S., Kin, W.Y., Jaye, I.F.M., Leung Pah Hang
Sarasini, F., Fiore, V., 2018. A systematic literature review on less common natural
https://doi.org/10.1016/j.jclepro.2018.05.197.
Shanmugarajah, B., Chew, I.M., Mubarak, N.M., Choong, T.S., Yoo, C., Tan, K.,
2019. Valorization of palm oil agro-waste into cellulose biosorbents for highly
https://doi.org/10.1016/j.jclepro.2018.10.342.
Song, K., Ji, Y., Wang, L., Wei, Y., Yu, Z., 2018. A green and environmental
benign method to extract cellulose nanocrystal by ball mill assisted solid acid
https://doi.org/10.1016/j.jclepro.2018.06.128.
Song, Y., Cho, E.J., Park, C.S., Oh, C.H., Park, B.J., Bae, H.J., 2019. A strategy for
https://doi.org/10.1016/j.renene.2019.03.032.
27
Journal Pre-proof
Sun, B., Zhang, M., Hou, Q., Liu, R., Wu, T., Si, C., 2016a. Further characterization
Sun, S., Sun, S. Cao, X., Sun, R., 2016b. The role of pretreatment in improving the
49-58. https://doi.org/10.1016/j.biortech.2015.08.061.
Tian, C., Yi, J., Wu, Y., Wu, Q., Qing, Y., Wang, L., 2016. Preparation of highly
https://doi.org/10.1016/j.carbpol.2015.09.055.
Tomić, T., Schneider, D.R., 2018. The role of energy from waste in circular
economy and closing the loop concept - energy analysis approach. Renew. Sust.
Tsukamoto, J., Durán, N., Tasic, L., 2013. Nanocellulose and bioethanol production
from orange waste using isolated microorganisms. J. Brazil. Chem. Soc. 24,
1537-1543. http://dx.doi.org/10.5935/0103-5053.20130195.
Walmsley, T. G., Varbanov, P. S., Su, R., Ong, B., Lal, N., 2018. Frontiers in
process development, integration and intensification for circular life cycles and
28
Journal Pre-proof
Figure captions:
cellulase. Left panel: 1.5 CMCase/mL; middle panel: 4.5 CMCase/mL; right
Fig. 5. Effect of hydrolysis durations on the zeta potential of FP, α-cellulose, and
BSEP.
Fig. 8. Effect of hydrolysis times on the size distribution of (a) FP, (b) α-cellulose, (c)
BSEP.
Fig. 9. SEM images of raw materials and acid hydrolyzed cellulosic biomass obtained
from (A) (a) FP, (B) (b) BSEP, and (C) (c) α-cellulose.
29
Journal Pre-proof
Filter paper
Biomass
α-Cellulose
UEK
Pretreatment
SEP
Enzyme BSEP
Hydrolysis
Hydrolysate
Fermentation
Leftover
Solid Bioethanol
Acid
Hydrolysis
Nanocrystalline
Cellulose
Fig. 1.
30
Journal Pre-proof
Fig. 2.
31
Journal Pre-proof
Fig. 3.
32
Journal Pre-proof
Fig. 4.
33
Journal Pre-proof
Fig. 5.
34
Journal Pre-proof
(a) (b)
Fig. 6.
35
Journal Pre-proof
Fig. 7.
36
Journal Pre-proof
(a)
(b)
(c)
Fig. 8.
37
Journal Pre-proof
A a
B b
B
C c
Fig. 9.
38
Journal Pre-proof
Table 1
39
Journal Pre-proof
Table 2
Mass balance for acid hydrolysis of biomass. Numbers denote hydrolysis times.
40
Journal Pre-proof
Table 3
Assessment of the endpoint impact indicator for BSEP, FP, and α-cellulose.
41
1
3 Table 4
Raw Material Procedures Particle Size (nm) Zeta ( mV) Recovery (%) Reference
1.Delignification
Corncob residue 2.Hydrochloric acid 198 ± 51 nm −33.8 ± 1.7 34.5% Palle, 2008
hydrolysis
1.Delignification Kargarzadeh et al.,
Kenaf bast fibers 124 ± 42 nm -8.77 ~ -95.3 55%
2. Sulfuric acid hydrolysis 2012
Cotton linters Sulfuric acid hydrolysis 179.3 nm -45.3 ± 1.4 - Morais et al., 2013
1. Sulfuric acid hydrolysis Length:100-300
Bleached eucalyptus wood pulp -13.3 ~ -44.5 - Tian et al., 2016
2.High-pressure homogenizer Width: 5-10
1.Milling machine
Cellulose fiber 300-400 nm - 59% Song et al., 2018
2. Oxalic acid hydrolysis
de Carvalho
Imperata brasiliensis Sulfuric acid hydrolysis 149.5 −30.83 ± 1.3 37-45%
Benini et al., 2018
1.Alkaline pretreatment
Pinus taeda 200 nm - 11.43 Bauli et al., 2019
2.Enzymatic hydrolysis
5
42
Journal Pre-proof
nanocrystalline cellulose
Chen d
Science and Technology, Yunlin, Taiwan (No. 123, Sec. 3, University Rd, Douliou,
E-mail:
*Corresponding author.
Research Highlights:
processing.
e. Nanocrystalline cellulose yields were 73% for unbleached kraft pulp as the
feedstock.