(Young Min Jin)
12*
(Joon Hyuk Lee)
3
(Sang Sun Choi)
1
Copyright © 2020 The Korean Institute of Metals and Materials
Key words(Korean)
crystallisation, x-ray diffraction, thermodynamic properties, fibre, chemical treatment
1. INTRODUCTION
Lyocell is a regenerated cellulosic biopolymer fundamental to the development of membranes,
films and fibres [1-3]. The fibre form comes from the wood pulp of eucalyptus dissolved in N-methyl-morpholine-N-oxide,
which adheres to the cellulose microfibrils [4]. Lyocell has long attracted attention, particularly for its high degree of nanoscale
modification. Alkali alters the physiochemical behaviour of lyocell, viz., the dimensional,
mechanical, and supramolecular properties of cellulose II fibres [5-8]. Different pretreatment approaches have been introduced to alter the structure of
biomass using NaOH and KOH (Figure 1). Some earlier studies employing optical microscopy and X-ray diffraction (XRD) have
provided insights in terms of crosslinking between the alkalis and fabric forms of
lyocell, and for characterizing the subset of physiochemical dynamics. Currently,
lyocell approaches are being integrated with precise engineering, with the result
that they possess a Brunauer-Emmett-Teller (BET) surface area of more than 1,000 m2 g-1. However, previous results based on samples with less than 1,000 m2 g-1 may hinder their adoption in applications. Accordingly, a distinct need exists to
determine how alkalis affect the BET surface areas of samples.
One of the drivers of nanocellulose technology is lyocell-based activated carbon fibres.
We designate these ACFs for convenience. ACFs are produced in activated fibrous form
from fibres and are ideally suited for microelectrodes, nanosheets, and as adsorbents
due to their outstanding chemical stability, high flexibility, and small thermal expansion
coefficient [9-13]. A theoretical and experimental cornerstone, to determine the supramolecular shift
of recently reported ACFs, is critical to understanding how crystallinity varies with
different alkali dosages. In a proof-of-principle experiment, we demonstrate the degree
of crystallinity of ACFs via NaOH and KOH based fibre modification. We take advantage
of the X-ray diffraction (XRD) technique to analyze the difference in structural composition
produced by various fibre modifications. A thermo-gravimetric analysis (TGA) was also
employed to cross-validate and assess the degree of crystallinity. The paper is organized
as follows. In Sec. II, we introduce the method of sample preparation and procedures.
After describing the experimental setup and techniques used to measure the full probability
distributions of the output ACFs, Sec. III is devoted to the analysis of various modifications.
Conclusions and further perspectives are given in Sec. IV. The as-prepared ACFs have
hierarchically interconnected macro-, meso-, and micropores with a decent surface
area. Taking the XRD and TGA results together, we are confident that ACFs de facto obey the general characteristics of lyocell-based fibres, regardless of their BET
surface area. Additionally, we also suggest the optimum proportion of impregnating
elements under the concept of precise engineering.
2. EXPERIMENTAL
2.1 Synthesis and structural characterization of samples
A 100% lyocell fibre was kindly supplied by Hyosung Corporation (South Korea) and
used in this work. The denier and filament were 1650d and 900f grade, respectively.
The NaOH and KOH alkalis were all analytical reagents and the solutions were all formulated
in deionized water. The alkali treatments were performed on fabric pieces of a given
dimension (3 cm × 3 cm) using a batch method. The pieces were padded through the alkali
solutions with a nip pressure of 1 bar. The first alkalis treatments employed 10~25
% of NaOH and KOH for 3 h at ambient room temperature, respectively. The designated
proportion of alkalis is shown in Table 1. ACF-W is a counterpart reference sample treated with deionized water. We limited
our dosage to 25 % to prevent the possible reduction of crystallinity. The secondary
treatment was conducting using 4 % each of KOH and H3PO4 for another 3 h [14]. Note that KOH was applied to design a porous surface on the sample, while H3PO4 was applied for better yield in the high temperature range during carbonisation and
oxidation [15-16]. Afterwards, the samples were neutralized in deionized water for 10 trials and linedried
overnight. An activated form of the samples was prepared using a three step procedure
which included oxidation (stabilisation), carbonisation, and activation.
The schematic diagram in Figure 2(a) shows the overall process. N2 gas is released from the N2 gas tank for carbonization, and H2O from the heating mantle was included once the activation started. The first step
was to oxidize the samples at a constant temperature of 300 °C with a heating rate
of 5 °C min-1 for 1 h (Figure 2(b)). Subsequently, the samples were heated at a constant rate of 10 °C min-1 in the presence of a N2 stream (0.5 L min-1) up to 800 °C for 1 h (Figure 2(c)). Finally, the samples were activated for 30 min at the same temperature via a mixture
of H2O + N2 flow. The BET method (N2-BET) revealed that all the modified samples reached more than 1,000 m2 g-1 (Micrometrics, ASAP2460) [17]. These numerical values suggest that a higher dosage of alkali solutions of either
NaOH or KOH can result in a higher BET surface area compared to ACF-W. In all circumstances,
a treatment of NaOH revealed higher BET surface characteristics than KOH. As depicted
in Figure 3, the SEM images also supported that the 25 % NaOH treatment produced the most porous
surface structure among all variations.
2.2 Measurements
The crystalline structure of the samples was determined using XRD (Bruker, D8 Advance)
with monochromatic CuKa radiation (λ = 0.1542 nm) in the Bregg-Brentano reflection geometry [18,19]. The analysis was conducted via a designated condition with a step size of (0.02°),
2θ range (which varied with the impregnating material), and step time (0.1 sec). The
degree of crystallinity (DC) was calculated as the ratio between the corresponding
area to the crystalline phase and the total area on the curve, written as Eq. 1;
where IC is the crystalline phase and IT is the total area under the XRD pattern. The thermal stability of the samples was
measured through TGA (Sinco, N-1000). The samples were scanned from ambient room temperature
to 300 °C using N2 gas with a heating rate of 5 °C min-1 for 100 min.
3. RESULTS AND DISCUSSION
Figure 4 provides the XRD spectra of samples, with the peaks labelled to indicate their crystal
lattice assignments. The spectra of the modified (ACF-Nn and ACF-Kn) samples are clearly different than the pristine sample (ACF-W). The peaks for the
shoulder (2θ = 20°) and the top (2θ = 22.5°) in the intensity profiles of the alkali-treated samples revealed a tendency
to become sharper than that of ACF-W. Interestingly, the intensity increased with
an increase in KOH content of 20 % or less, compared to ACF-W, but the opposite result
occurred in ACF-K25. Further, a general trend of anti-correlation between the KOH
dosage and intensity was observed. Here, the ACF-K10 showed the highest DC increment
with 2.1 % better crystallinity than ACF-W (Table 2). Meanwhile, the resulting intensity after NaOH treatment was found to decrease in
all cases compared to ACF-W. Such decrements are mainly due to the alkali treatment.
The alkali modification swells the lyocell fibres and weakens the dense structure
and intermolecular bonds of the fibrous polymer chain [18]. This results in a rearrangement between the cellulose molecules, and drives a transformation
from an antiparallel to a parallel conformation.
A characteristic band of 2θ = 12.3° was observed in ACF-W, which corresponds to the (1 0 1) lattice plane. However,
there was only a single broadband between 2θ = 20° ~ 22°, indicating that a premature to classify ACF-W as a cellulose II crystallite.
In the modified samples, the characteristic peaks of cellulose II were observed Figures 4(b) and 4(d). The observation of major peaks with diffraction angles assigned to the (1 0 1) and
(0 0 2) lattice planes suggest that ACF-Nn and ACF-Kn yield fair characteristics of the cellulose II crystallites, which is consistent
with previous reports [19]. Overall, ACF-Nn exhibited more preferable cellulose II crystallites than ACF-Kn, since Na+ has a greater hydration radius compared to K+ ions [20-22]. It appears that NaOH treatment may reduce the DC while such change in structure
may alter crystallites into a more distinct form of cellulose II. In sum, both NaOH
and KOH had a visible influence on the DC shift, and thus were vital to controlling
alkali concentrations to achieve the target crystallinity for further use.
After treatment in alkalis, the samples were assessed by TGA to further investigate
their conformation resulting from thermal gravimetric mass loss. In general, lyocell-based
fibres show two phases of weight loss, due to dehydration and decomposition of the
cellulose [23]. Figures 5(a) and 5(b) show that the samples obeyed general properties, with a major weight loss at temperatures
between 150 °C and 200 °C after dehydration. Comparing the alkali treatments performed
in this study, both NaOH and KOH demonstrated a strong upward linear relationship
compared to ACF-W. It is therefore suggested that the thermal stability of the treated
samples was improved via the fibre-reinforcing effect [24]. Under similar treatment dosages, ACF-Nn holds better thermal stability than ACF-Kn. It was also found that 15 % and 20 % dosages resulted in the best thermal stability
in ACF-Nn and ACF-Kn, respectively.
4. CONCLUDING REMARKS
In conclusion, we studied the effect of alkali treatment of ACFs with high BET surface
areas. By probing the domain orientations of samples using XRD and TGA, we were able
to map the findings as follows.
(1) A 25 % dosage of NaOH provided the highest BET surface area with the most porous
surface structure compared to other samples.
(2) ACF-K15 showed the highest DC, while the use of 15 ~ 25 % NaOH resulted in the
most distinct form of cellulose II.
(3) The flame retardancy of samples was improved significantly after alkali treatment.
A 15 % of NaOH was found to provide the best thermal stability among all samples.
(4) Combining all the presented strategies, these findings offer a fundamental route
to control the behaviour of ACFs that may aid in the growing field of highly porous
material research using lyocell-based fibres.
Acknowledgements
The authors are grateful to the Korea Testing Certification for the use of their facilities.
REFERENCES
Håkansson K. M. O., Fall A. B., Lundell F., Yu S., Krywka C., Roth S. V., Santoro
G., Kvick M., Wittberg L. P., Wågberg L., Söderberg L. D., Nat. Commun,5, 4018 (2014)
Eichhorn S. J., Dufresne A., Aranguren M., Marcovich N. E., Capadona J. R., Rowan
S. J., Weder C., Thielemans W., Roman M., Renneckar S., Gindl W., Veigel S., Keckes
J., Yano H., Abe K., Nogi M., Nakagaito A. N., Mangalam A., Simonsen J., Benight A.
S., Bismarck A., Berglund L. A., Peijs T., J. Mater. Sci,45, 1 (2010)
Lim K. Y., Seong Y. J, Kim B. C., Polym. J,35, 691 (2003)
Kim D. B., Lee W. S., Jo S. M., Lee Y. M., Kim B. C., Polym. J,33, 139 (2001)
Sukmawan R., Takagi H., Nakagaito A. N., Compos. Part. B-Eng,84, 9 (2016)
Sayyed A. J., Mohite L. V., Deshmukh N. A., Pinjari D. V., Carbohyd. Polym,206, 220
(2019)
Kim H. G., Lee Y. S., Carbon. Lett,29, 529 (2019)
Tan S. J., Supri A. G., Polym Bull,73, 539 (2016)
Wang J., He J., Ma L., Zhang Y., Shen L., Xiong S., Qu M., Chem Eng J,390, 124508
(2020)
Carraro P. S., Spessato L., Crespo L. H., Yokoyama J. T., Fonseca J. M., Bedin K.
C., Almeida V. C., J. Mol. Liq,289, 111150 (2019)
Lu H., Zhao X. S., Sustain. Energ. Fuels,1, 1265 (2017)
Choi S. S., Lee J. H., Jin Y. M., Lee S. H., Carbon. Lett,29, 633 (2019)
Frank E., Steudle L. M., Ingildeev D., Spörl J. M., Buchmeiser M. R., Angew. Chem.
Int. Ed,53, 5262 (2014)
Dobele G., Rossinskaja G., Telysheva G., Meier D., Faix O., J. Anal. Appl. Pyrol,49,
307 (1999)
Sevilla M., Ferrero G. A., Fuertes A. B., Carbon,114, 50 (2017)
Renders T., Schutyser W., Van den Bosch S., Koelewijn S. F., Vangeel T., Courtin C.
M., Sels B. F., ACS Catal,6, 2055 (2016)
Lee J. H., Lee S. H., Suh D. H., Carbon. Lett,30, 99 (2020)
Kim S. Y., Lee H. S., Met. Mater. Int,25, 528 (2019)
Oh H. G., Park D. K., Kim Y. D., Lee M. S., Korean J. Met. Mater,57, 808 (2019)
Aguado R., Lourenço A. F., Ferreira P. J., Moral A., Tijero A., Cellulose,26, 5925
(2019)
Satari B., Karimi K., Molaverdi M., Bioprocess. Biosyst. Eng,41, 249 (2018)
Kielland J., J. Am. Chem. Soc,59, 1675 (1937)
Voronova M. I., Petrova S. N., Lebedeva T. N., Ivanova O. N., Prusov A. N., Zakharov
A. G., Fibre. Chem,36, 408 (2004)
Jaturapiree A., Manian A. P., Lenninger M., Bechtold T., Carbohyd. Polym,86, 612 (2011)
Lizundia E., Vilas J. L., León L. M., Carbohyd. Polym,123, 256 (2015)
Park J. W., Lee T. H., Back J. H., Jang S. W., Kim H. J., Skrifvars M., Compos. Part.
B-Eng,167, 387 (2019)
Figures and Tables
Fig. 1.
Proposed mechanism of lyocell fibre by the impregnation of alkalis.
Fig. 2.
Schematic diagram for experimental procedures of samples is illustrated in (a). A
modulated temperature range of oxidation (stabilisation) is plotted in (b) and (c)
shows that of carbonisation and activation.
Fig. 3.
SEM data for analysis of surface structure before and after the application of alkalis
with a magnitude of x 5.0 K. Samples of ACF-Nn and Kn are listed in (a)-(d) and (e)-(h), respectively. A counterpart sample of ACF-W is
shown in (i). Herein, evenly developed phases of macro-, meso-, and micro-pores were
successively observed.
Fig. 4.
XRD patterns of (a)-(b) ACF-Nn and (c)-(d) ACF-Kn versus a counterpart sample of ACF-W.
Fig. 5.
TGA curves of (a) ACF-Nn and (b) ACF-Kn versus a counterpart sample of ACF-W. Herein, the bottom axis of Time (min) can be
converted to Temperature (°C) by multiplying three.
Table 1.
Samples modified under various proportion of alkali solutions.
|
Classification
|
First treatment
|
Secondary treatment
|
BET surface area (m2 g-1)
|
|
ACF-N10
|
10 % of NaOH
|
KOH (4 %) + H3PO4 (4 %)
|
1072
|
|
ACF-N15
|
15 % of NaOH
|
1,154
|
|
ACF-N20
|
20 % of NaOH
|
1,240
|
|
ACF-N25
|
25 % of NaOH
|
1,298
|
|
ACF-K10
|
10 % of KOH
|
1011
|
|
ACF-K15
|
15 % of KOH
|
1,046
|
|
ACF-K20
|
20 % of KOH
|
1,109
|
|
ACF-K25
|
25 % of KOH
|
1,121
|
|
ACF-W
|
Deionized water
|
824
|
Table 2.
DC of samples by various impregnation methods.
|
Classification
|
%DC
|
|
ACF-N10
|
90.2
|
|
ACF-N15
|
90.6
|
|
ACF-N20
|
88.8
|
|
ACF-N25
|
88.3
|
|
ACF-K10
|
93.6
|
|
ACF-K15
|
92.1
|
|
ACF-K20
|
91.2
|
|
ACF-K25
|
87.6
|
|
ACF-W
|
91.5
|