Research Article

pubs.acs.org/journal/ascecg

Properties of Cellulosic Material after Cationization in Different

Solvents
Nora Odabas,† Hassan Amer,†,‡ Markus Bacher,† Ute Henniges,*,† Antje Potthast,†
and Thomas Rosenau*,†
†
Division of Chemistry of Renewable Resources, Department of Chemistry, University of Natural Resources and Life Sciences
Vienna, Konrad-Lorenz-Straße 24, 3430 Tulln, Austria
‡
Department of Natural and Microbial Products Chemistry, National Research Centre, 33 Al Behous St., Dokki, Giza, Egypt
*
S Supporting Information
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: Fibers resulting from pulping of wood are negatively

Downloaded via THAMMASET UNIV LIB on March 2, 2023 at 03:57:54 (UTC).

charged by nature. Both the cellulose and hemicelluloses contribute to the

anionicity. The negative charge of cellulosic material can be reversed with
the same reagent commonly used for the cationization of starch. In this
study, a bleached Kraft pulp was cationized in systems with different
water-miscible organic solvents. Replacing 90% of the water with
isopropanol and particularly with tetrahydrofuran yielded higher degrees
of substitution and increased reaction efficiency. The degree of
substitution depends on the concentration of the cationization reagent
in water; partially replacing water with tetrahydrofuran can simulate a higher concentration while maintaining supramolecular
properties, such as crystallinity and polymer chain integrity.
KEYWORDS: Cellulose, Pulp properties, Hydrophilization, Lignocelluloses, Gel permeation chromatography, Degree of substitution,
Infrared spectroscopy

■ INTRODUCTION
Fibers resulting from pulping of wood are negatively charged by
yield issue, de la Motte et al.13 developed a spray technique that
resulted in up to 8% reaction efficiency for softwood Kraft
nature. The main reasons are uronic acids in hemicelluloses and pulp,11 but it involves solvent exchange and is therefore of less
pectins1 and, especially after chemical pulping, carboxyl practical relevance.
functionalities in the cellulose.2 To overcome the resulting The present study shows how replacing up to 90% of the
electrostatic repulsion among the fibers, papermakers usually water with different organic solvents can improve the reaction
add cationic auxiliaries to pulp. Cationic starch in particular is efficiency of cationization and how the reaction conditions
widely used;3 it is the most important cationized poly- affect the integrity of the cellulose. On the basis of one protocol
saccharide,4 though there are other examples for industrial for treating a bleached Kraft pulp, under otherwise identical
applications: Compared to the respective unmodified materials, conditions (liquid-to-solid ratio, time, temperature profile,
suspensions of cationized cellulose nanocrystals are more amount of alkali and amount of reagent), the water was
stable,5 cationized cotton allows salt-free dying,6 the absorption partially replaced by either dimethyl sulfoxide (DMSO),
capacity of cationized cellulose nanofibrils for anionic dyes tetrahydrofuran (THF), or isopropanol. All three solvents are
helps in wastewater treatment,7 and cationization of pulp fibers water-miscible; possible phase separation under the actual
can improve thermoplastic composites8 or serve as wet-end reaction conditions was studied. Isopropanol is a popular
additives for papermaking.9 solvent used in cellulose etherification, for the synthesis of
A common reagent for the cationization of both starch10 and carboxymethyl cellulose or hydroxyethyl cellulose,14 and in one
cellulose7,9 is 2,3-epoxypropyltrimethylammonium chloride study also for cellulose cationization.15 For the production of
(EPTMAC), used either directly, or formed in situ from the highly cationized starch, different water-miscible organic
more stable 3-chloro-2-hydroxypropyltrimthylammonium solvents including dimethyl sulfoxide,16 isopropanol,17 and
chloride.4 The modification is usually done in the presence of tetrahydrofuran18 have been used in attempts to limit side
varying amounts of water and sodium hydroxide. For the reactions as well as excessive swelling.18 Also hemicelluloses
cationization of cellulose, the latter is important as a have been cationized in DMSO−water19 and alcohol−water20
nucleophile that activates the hydroxyl groups of cellulose mixtures. Cellulose has been cationized in anhydrous DMSO21
toward etherification. Water, however, reduces the reaction
efficiency11 by causing side reactions with the caustic.12 Received: December 21, 2015
Literature values for reaction efficiencies for common methods Revised: February 15, 2016
range from below 1%5 to 3%.11 In an attempt to deal with the Published: February 29, 2016

© 2016 American Chemical Society 2295 DOI: 10.1021/acssuschemeng.5b01752

ACS Sustainable Chem. Eng. 2016, 4, 2295−2301
ACS Sustainable Chemistry & Engineering Research Article

to a very low degree of substitution. However, to our best The workup of TC was done differently: The material was filtered
knowledge, no studies on pulp cationization that assess off and mixed with water. After several repetitions of decanting and
different water−solvent mixtures have been published so far. refilling, the material was precipitated in acetone, filtered off, washed
Previous studies on pulp cationization have focused either on with some more acetone, and air-dried.
Determination of Degree of Substitution. The degree of
molecular properties, such as the substitution pattern,11,13 or on
substitution (DS) was determined by conductometric titration5 using a
the mechanical and optical properties of paper made from this system by Metrohm (Herisau, Switzerland) consisting of an 856
cationized pulp.9,21,22 Depending on the intended use of fibers, Conductivity Module, an 801 Stirrer, an 800 Dosino Dosing Device
the content of highly ordered areas23 (commonly referred to as and 807 Dosing unit, and a Pt1000/B/s conductivity electrode. Of
“crystallinity”)24 and molar mass distribution25 are of major each sample, 1 g was lyophilized and ground in a Retsch CryoMill
importance concerning the resulting material properties. This (Haan, Germany) for 9 min at a frequency of 25 s−1; of that, 100 mg
series of experiments also shows how the reaction conditions aliquots were suspended in 50 mL of deionized water and titrated with
influence these two quantities. To be able to detect whether a 10 mM aqueous silver nitrate solution. For IC and TC, 50 mg
observed effects are caused directly by the cationization or aliquots were used. 50 μL of silver nitrate solution were added every
rather by the reaction conditions, blank reactions were done 10 s. Triplicate measurements were done for each sample. Analysis and
with sodium chloride. explanation of these data are available as the Supporting Information.

■
Infrared Spectroscopy. For attenuated total reflectance infrared
spectroscopy (ATR-FTIR), each sample was measured in quad-
EXPERIMENTAL SECTION ruplicate on a PerkinElmer FT-IR Spectrometer Frontier (Waltham,
Material. Bleached softwood Kraft pulp in sheet-dried form was MA, USA). Spectra from 4000 to 650 cm−1 were recorded with a
kindly provided by Zellstoff Pöls AG, Austria. It contained 89% resolution of 4 cm−1, baseline corrected using the instrument software
cellulose, 11% hemicelluloses, and traces of lignin. (PerkinElmer Spectrum), and imported to Microsoft Excel 2010
Dimethylacetamide (DMAc; from Promochem, Teddington, United where they were normalized to 0 at the minimum absorption around
Kingdom), dimethyl sulfoxide (DMSO, from Roth, Karlsruhe, 1520 cm−1 and 1 to the maximum absorption around 1030 cm−1.
Germany), isopropanol (from Roth), tetrahydrofuran (THF, from Near infrared (NIR) spectra were recorded with an MPA Multi-
VWR), and hydrochloric acid (HCl; from VWR, Radnor, PA, USA) Purpose Analyzer from Bruker (Billerica, MA, USA) equipped with a
were purchased at analytical grade and were used without further fiber optic probe and a Te-InGaAs detector. The air-dried samples
purification. Acetone was purchased at technical grade at VWR. were air equilibrated for 1 week. Each sample was measured in five
EPTMAC (technical grade, 21.6% water, 90% purity of dry substance), spots with 32 scans from 12 500 to 4000 cm−1 with a resolution of 4
sodium chloride (NaCl), silver nitrate, and sodium hydroxide (NaOH) cm−1. OPUS 7.0 software from Bruker was used to do a rubber band
were purchased from Sigma-Aldrich (St. Louis, MO, USA), lithium baseline correction and normalization from 0 for the lowest and 2 for
chloride (LiCl) from VWR. Water was deionized to a conductivity of the highest value.
0.06 μS/cm for the reaction and for analyses, and to 0.4 μS/cm for the Determination of Zeta Potential. The zeta potential was
workup. determined on a SZP-04 device from Mütek (Filderstadt, Germany).
Reaction Protocol. The pulp was soaked in deionized water for 4 A 5 g portion of each sample was suspended in 500 mL deionized
h and dispersed using a kitchen blender with no blades (rotating disk water; the conductivity was set to 300 μS/cm; and the pH was set to 7
only). HCl was added to achieve pH 1. After 1 h, the pulp was filtered with NaCl, HCl, and NaOH. The zeta potential was measured 3−5
off and washed with water until the water pressed out from the filter times for each sample.
cake was neutral. The moisture content was determined by spreading Determination of Molar Mass Distribution. The molar mass
out about 1 g of the material on the dish of a Sartorius Moisture distributions of the starting material and the products of the blank
Analyzer MA35 (Göttingen, Germany) and then heating to 105 °C reactions were determined by gel permeation chromatography (GPC)
until a constant weight was reached. after dissolution in DMAc with 9% LiCl.2 20 mg aliquots of air-dried
For each assay, 5 g of acid-washed pulp (consistency 30%) was samples were mixed with 150 mL of deionized water in a kitchen
mixed with 225 mL of deionized water, DMSO, isopropanol, or THF. blender, filtered, washed with ethanol and DMAc, shaken on a GFL
After 1 h stirring at room temperature, 2.5 mL of 10 M NaOH were 3005 orbital shaker (GFL Gesellschaft für Labortechnik, Burgwedel,
added. After further 30 min, the reagent was added drop by drop: 17.8 Germany) in 4 mL of DMAc overnight, filtered off again, and mixed
mL of EPTMAC. This corresponds to about 3 mol/mol anhydrosugar with 2 mL of DMAc/LiCl 9% (w/v). After a week on the shaker, 300
unit (ASU). Blank reactions were done with 3.7 g of NaCl being added μL of the dissolved sample were diluted with 900 μL of DMAc, and
instead of the epoxide reagent, i.e., 3 mol/mol ASU. Deionized water filtered through 0.45 μm PTFE filters. The injection volume was 100
was added to make up a total solvent volume of 250 mL. After 15 h at μL. The GPC system had four serial columns (PL gel mixed A LS, 20
40 °C, the reaction was stopped by adding 12.5 mL of 4 M HCl. This μm), a multiangle laser light scattering detector (Dawn DSP with
amount decreased the pH to about 1 if water was the only solvent. The argon ion laser at λ0 = 488 nm, Wyatt Technology, Santa Barbara, CA,
reaction supernatant was filtered off through a fine cloth in a Büchner USA), and a refractive index detector (Shodex RI-71, Showa Denko
funnel and the solid material washed with water until neutral. Table 1 K.K., Tokyo, Japan) and was run with DMAc/LiCl 0.9% (w/v) as the
shows the full assay matrix with the subsequently used abbreviations. eluent.

■
The full set was prepared independently twice. The second batch also
included two assays where 187.5 mL of THF plus 61.5 mL of water
and 125 mL of THF plus 125 mL of water were used as solvent, RESULTS AND DISCUSSION
referred to as “75% THF” and “50% THF”, respectively. Though the organic solvents are fully miscible with water,
isopropanol and THF phase-separate with 4% aqueous NaOH:
Table 1. Abbreviations for Materials Resulting from Mixing 1 mL of 1 mol/L NaOH with 9 mL of THF resulted in
Cationization and Blank Assays two clear liquid phases, the volume of the lower (aqueous)
phase being about 200 μL. In the case of isopropanol, an even
solvent cationized material material after blank reaction smaller aqueous phase occurred. The actual cationization and
water WC WB blank assays also contained 2% pulp. The swollen fibers
DMSO + water DC DB impeded the observation of possible phase separationsthe
isopropanol + water IC IB observation of Klemm et al.26 that a cellulose/NaOH/water
THF + water TC TB phase is formed suggests that this may not have been the case at
2296 DOI: 10.1021/acssuschemeng.5b01752
ACS Sustainable Chem. Eng. 2016, 4, 2295−2301
ACS Sustainable Chemistry & Engineering Research Article

least for isopropanol. The desired cationization reaction and the

main side product are shown in Figure 1.

Figure 3. Cationization DS (from titration) for the first (open

diamond) and second (red diamond) sample set. Because the same
ratio of reagent to dry sample was applied in all assays, the DS directly
correlates to the reaction efficiency.
Figure 1. Schematic of pulp cationization. Besides the desired reaction,
the etherification of cellulose, 2,3-epoxypropyltrimethylammonium
chloride also forms a diol (highlighted structure) upon reaction with reactions was done twice. The results were quite similar, only
water. If the actual reactionas in this studyis followed by an acid
the DS for TC showed some deviation, probably because of
wash with hydrochloric acid, the counterion is changed to the shown
chloride. workup issues.
Preliminary tests had been conducted where bleached
For the reactions done in isopropanol and THF, the success chemical pulps were cationized in water only by protocols
of the cationization was already evident during the workup. similar to the one used for WC, applying 0.03−0.32 mmol
Additional charged groups should increase the hydrophilicity, epoxide reagent per milliliter of water. The results from these
and the washing of these materials with water was indeed more previous cationization assays (data not shown) led to the
difficult than washing the respective blank material: their high assumption that the reagent concentration might be critical for
swelling obviously impeded the filtration process. TC had a gel- the resulting DS. In this study, this concentration was 0.37
like texture when swollen in water (see Figure 2) and was even mmol epoxide reagent per milliliter water for WC and 3.7
mmol mL−1 for DC, IC, and TC. When using the data of both
current and previous water-only experiments for a linear fit, the
predicted DS for removing 90% of the water is close to the
result we got for TC (see Figure 4 A), suggesting that THF
does not cause any noticeable interfering reactions in this setup.
The DS of IC and especially DC are considerably lower,
additional side reactions seem to occur. The extrapolated
concentration span was, however, a multiple of the range used
for the regression, so two more points were added: 75% THF
and 50% THF (see the Experimental Section) corresponding to
1.5 and 0.8 mmol reagent per milliliter water, respectively (see
Figure 4B). The DS values resulting from those two assays were
as close to the extrapolated values as the DS of TC of set one.
We concluded that THF behaved as an “inert” solvent in this
reaction, while this was not the case for isopropanol and
Figure 2. Beaker with pulp cationized in THF and swollen in water, especially DMSO.
during workup. Image converted to grayscale and cropped with Infrared Spectroscopy. The ATR-FTIR spectra showed
IrfanView 4.35, Irfan Skiljan, Wiener Neustadt, Austria. the effect of cationization both directly and indirectly (see
Figure 5). The substitution as such can be tracked: the
stretching vibration of the C−N bonds18 of the substituent7
impossible to filter; therefore, the material was precipitated with (see Figure 1) is found at 1480 cm−1. Cationization is a means
acetone for filtration. For all four cationized materials, their of hydrophilization, which is visible in the absorption band of
tendency to stick to glassware was another clue for successful adsorbed water27 at 1640 cm−1 that increases roughly with the
modification. The effect of the cationization was also evident degree of substitution.
from the zeta potential which was negative for the starting Content of Highly Ordered Areas. Native celluloses, like
material as well as for blank materials, and positive for the those in the present pulp, are ordered in microfibrils to some
cationized materials. The starting material had a zeta potential extent. Though there is evidence that calling these ordered
of −30 mV, the blank materials between −22 and −28 mV, and regions “crystals” is not necessarily correct,28 the established
the cationized materials between +10 and +20 mV. term “crystallinity”24 may be understood as the fraction of
Degree of Substitution. Both WC and DC had a DS of anhydroglucose units that are situated in highly ordered regions
about 0.05, IC about three times more, and TC five to eight as opposed to those with little order that are part of truly
times more (see Figure 3). The fraction of the EPTMAC that amorphous material for a given cellulosic sample. Independent
formed a covalent bond to the starting material (see Figure 1), of the scholarly question whether the highly ordered regions
referred to here as reaction efficiency, increased almost 10-fold can be considered crystals or not, their content influences
when replacing 90% of the water with THF. The set of important properties such as the swelling of a cellulosic
2297 DOI: 10.1021/acssuschemeng.5b01752
ACS Sustainable Chem. Eng. 2016, 4, 2295−2301
ACS Sustainable Chemistry & Engineering Research Article

Figure 4. DS versus amount of epoxide reagent (mmol/mL of water). (A) Data from first set of experiments combined with results from pretests. A
linear fit from data of water-only experiments (0.03−0.37 mmol mL−1). (B) Both data sets plus earlier data, same linear fit.

while that at 1370 cm−1 has some variations. In our case, for
those two bands, the variations were larger for the 2900 cm−1
value, thus causing a large standard uncertainty of this index;
the standard deviations reported in this section are derived
from values of 4−5 individual spots per material. Therefore, the
absorptions at 1370 cm−1 were also directly compared (Figure
6B). The total crystallinity seemed to be unaffected by all
treatments.
Another CI definition was introduced by O’Connor,30
sometimes called the “lateral order index” (LOI), see the
Supporting Information for details on its determination. The
idea behind it is that the band at 1430 cm−1 is a measure for
cellulose I (typical crystal lattice of native cellulose, chains have
Figure 5. Cationization seen in ATR-FTIR: (black solid) starting
parallel orientation), while the band at 900 cm−1 is stronger for
material; (blue short dash) WC; (olive short dash dot) DC; (magenta
dash) IC; (orange short dot) TC. The numbers denote bands that cellulose II (cellulose crystal lattice after mercerization or
report cationization. 1640 cm−1 (1): adsorbed water, 1480 cm−1 (2): regeneration from dissolved state, chains have antiparallel
C−N stretching vibration. orientation) and amorphous cellulose.
LOI = H1430/H900
material. Bearing that in mind, ATR-FTIR spectra were used to
look into the content of highly ordered areas based on different For this index, larger differences were observed (see Figure
crystallinity indices (CI) from literature. 7). For IB, the ratio was clearly lower than for the other
A measure of “total crystallinity” can be defined as the ratio materials. In a purely aqueous system, NaOH with 0.1 mol/L
between the absorption band maxima at 1370 and 2900 cm−1 as or any concentration of sodium hydroxide not exceeding 1 mol
introduced by Nelson and O’Connor,29 for details please refer per liter water would not affect the crystallinity of cellulose.
to the Supporting Information available online. However, NaOH in isopropanol−water mixtures is known to
CIATR = H1370/H2900 cause an alkalization effect similar to a much more concentrated
NaOH in water alone,26 causing decrystallization and
Figure 6 A shows that the averages of the CIATR values for facilitating change of the crystal allomorph.31 The resulting
the cationized samples did not differ significantly. The idea of material contains more amorphous material and possibly also
this ratio is that the absorption at 2900 cm−1 is fairly constant some cellulose II. This is consistent with the observation that

Figure 6. Total crystallinity from ATR-FTIR. (A) Calculation of the CI as suggested by Nelson and O’Connor26 and (B) normalized absorption at
1370 cm−1 (H1370). The total crystallinity seems to be unaffected by the used treatments. Data from materials of two sets of experiments: (black)
first; (red) second set, (error bars) 2× standard deviation (SD).

2298 DOI: 10.1021/acssuschemeng.5b01752

ACS Sustainable Chem. Eng. 2016, 4, 2295−2301
ACS Sustainable Chemistry & Engineering Research Article

cellulose II as mentioned above. Solid-state NMR spectra (see

the Supporting Information) confirmed that IB contained more
amorphous material and indicated the presence of some
cellulose II. For TB, both the CINIR and the LOI were lower
than for Start, WB, and DB but closer to those than to IB; this
also matches the data from solid-state NMR. Therefore, TB
also contained more amorphous material than the original pulp
and a bit of cellulose II but is closer to the original pulp than to
IB.
All cationized materials had a lower CINIR than the materials
after the blank reaction; the higher the DS, the lower CINIR.
This is likely due to the larger amount of adsorbed water.33
Figure 7. Lateral order index (LOI). Data from materials of two sets of This index may thus be unsuitable for estimating crystallinity
experiments: (black) first; (red) second; (error bars) 2× SD. changes that accompany reactions that chemically alter the
hydrophilicity of cellulosic materials. The CINIR values for the
IB had a different texture when dry and did not resuspend as cationized materials were therefore omitted from Figure 8, but
readily in water. For the cationized materials, the measurement may be found in the Supporting Information.
variations are larger; the values may be falsified by the C−N Molar Mass Distribution. When determining molar mass
absorption band of the substituent at 1480 cm−1, being rather distributions by GPC with multiangle laser light scattering
close to the band at 1430 cm−1; see Figure 5. They were (MALLS) detection, it is necessary to know the analyte’s
therefore omitted in Figure 7 but may be found in the refractive index increment (dn/dc)μ in the given system.34 For
Supporting Information. cellulose, the refractive index increment in the used system is
Crystallinity may also be estimated from NIR spectra,32 see 0.136 mL g−1.35 It is highly probable that this value is affected
the Supporting Information for details: by cationization, at least to some minor degree. This is why
CINIR = H4760/(H4760 + H5187) only noncationized materials were analyzed with GPC as the
results for the cationized samples were dubious because of the
Among the unmodified blank materials (Figure 8), the only unknown refractive index change. The derivatization in water,
one with a significantly lower CINIR was again IB. So, two out of isopropanol/water and tetrahydrofuran/water did not affect
cellulose integrity. The only system that caused cellulose
degradation was the DMSO-containing one. Here, the shoulder
at around 30 kg mol−1 disappeared and the whole peak was
significantly shifted toward lower molar mass values. Thus, the
hemicelluloses may have been degraded or solubilized, and the
main cellulose peak is shifted from about 400 to 160 kg mol−1
(see Figure 9). This disagrees with the results of Ren et al.19,20
who found less degradation when using DMSO as a solvent for
cationizing hemicelluloses than when using a system with
alcohol, although the setups were not the same (different
reaction temperature, alkalinity, etc.). Our molar mass
Figure 8. CI of the sample specimens determined from NIR.32 Values determinations do, however, match the observation that
from two sets of experiments: (black) first, (red) second, (error bars) compared to the starting material, both DC and DB dissolved
2× SD. much faster and gave a much less viscous solution, while the
viscosity of all other samples remained high and undistinguish-
three CI show a decrease for treatment with isopropanol. Two able from the starting samples.
possible reasons, that are not mutually exclusive, are a real Comparison of Different Organic Media. As discussed
increase of amorphous material and the formation of some above, the choice of organic solvent affected the DS (see Figure

Figure 9. Gel permeation chromatography of noncationized materials. (A) Molar mass distributions and (B) elution profile: (black solid) starting
material; (blue short dash) WB; (olive short dash dot) DB; (magenta dash) IB; (orange short dot) TB. Only data for the first samples set shown,
results were similar for the second batch.

2299 DOI: 10.1021/acssuschemeng.5b01752

ACS Sustainable Chem. Eng. 2016, 4, 2295−2301
ACS Sustainable Chemistry & Engineering Research Article

3), but also other material properties. Concerning the DS, any Conductometric titration data analysis; content of highly
competitive reaction of the solvent with the reagent is relevant ordered areas by infrared spectroscopy and solid-state
as well as its behavior toward NaOH and water. In contrast to NMR (PDF)

■
starch or hemicelluloses for which homogeneous modification
occurs with some solvents, a homogeneous cationization of
cellulose, although reported,36 is not relevant for pulp when AUTHOR INFORMATION
going beyond small lab scale. Corresponding Authors
Isopropanol was chosen as cosolvent as it was expected for *E-mail: [email protected] (U.H.).
sterical reasons to be less reactive toward the epoxide than *E-mail: [email protected] (T.R.).
ethanol or methanol, and certainly less than water. In addition,
Notes
the rate of mercerization reaction in isopropanol is higher than
in ethanol or methanol.31 It was expected to improve the The authors declare no competing financial interest.
reaction efficiency also because of the above-discussed stronger
alkylation effect in derivatizations of cellulose.26 This effect was
confirmed and it was stronger than in the case of THF although
■ ACKNOWLEDGMENTS
The authors would like to thank the Flippr° project and all
the latter resulted in a higher DS. This matches the results of financial and scientific contributors to this project, including the
starch cationization in different solvents where THF proved Austrian Research Promotion Agency (F.F.G.). Special thanks
more beneficial than ethanol.18 go to Melanie Mayr, TU Graz, for performing the zeta potential
For both DC and DB, the reaction mixtures developed a measurements.

■
distinct unpleasant cabbage odor. This indicated that some
DMSO was reduced to dimethyl sulfide. The corresponding ABBREVIATIONS
reaction is a Swern-type process that might occur as soon as
DMSO is in contact with alkylating agents, in particular in ASU, anhydrosugar unit; ATR-FTIR, attenuated total reflec-
alkaline media. The outcome is degradation (reduction) of tance infrared spectroscopy; CI, crystallinity index; DMAc,
DMSO, mainly to dimethyl sulfide, which goes hand in hand N,N-dimethylacetamide; DMSO, dimethyl sulfoxide; DS,
with the oxidation of a coreactant, in this case cellulose.37 If this degree of substitution; EPTMAC, 2,3-epoxypropyltrimethyl-
oxidation affects cellulosic hydroxyl groups in accessible ammonium chloride; GPC, gel permeation chromatography;
regions, it is immediately followed by chain cleavage through H1470, absorption maximum of the band around 1470 cm−1;
beta elimination. If this process randomly happened for LOI, lateral order index; NIR, near-infrared; SD, standard
hemicelluloses and, for cellulose, in amorphous regions and deviation
along the surface of highly ordered regions, it did not
necessarily change the content of highly ordered areas much
(see Figures 6−8) while still considerably reducing the average
■ REFERENCES
(1) Holmbom, B.; Pranovich, A. V.; Sundberg, A.; Buchert, J.
polysaccharide chain length (see Figure 9). Charged groups in wood and mechanical pulps. International Cellucon

■
Conference [on] Pulp for Papermaking, Turku, Finland, Kennedy, J. F.,
CONCLUSION Phillips, G. O., Williams, P. A., Eds., Woodhead Publishing: Turku,
Finland, 1998; pp 109−119.
Pulp can be cationized with the same quaternary ammonium (2) Bohrn, R.; Potthast, A.; Schiehser, S.; Rosenau, T.; Sixta, H.;
epoxide reagent that is commonly used for starch. The success Kosma, P. The FDAM method: Determination of carboxyl profiles in
of the reaction is reflected in the reversion of the zeta potential cellulosic materials by combining group-selective fluorescence labelling
which turned positive for all samples treated with the epoxide with GPC. Biomacromolecules 2006, 7, 1743−1750.
reagent. A partial replacement of water with organic solvents (3) Schrijver, J. In Handbook of Paper and Board; Holik, H., Ed.,
gave much higher degrees of substitution, and higher reaction Wiley-VCH Verlag: Weinheim, 2013; Vol. 1, pp 146−160.
efficiencies. The resulting materials are visibly more hydro- (4) Prado, H. J.; Matulewicz, M. C. Cationization of polysaccharides:
philic. THF proved to be particularly suitable in this regard. A path to greener derivatives with many industrial applications. Eur.
Polym. J. 2014, 52, 53−75.
There was no noticeable effect on the highly ordered regions (5) Hasani, M.; Cranston, E. D.; Westman, G.; Gray, D. G. Cationic
of the cellulose except for material treated in isopropanol surface functionalization of cellulose nanocrystals. Soft Matter 2008, 4,
here some decrystallization and a partial conversion to cellulose 2238−2244.
II is supposed. There was no notable change in molar mass (6) Hashem, M.; Hauser, P.; Smith, B. Reaction efficiency for
distribution apart from the case of DMSO which causes Swern- cellulose cationization using 3-chloro-2-hydroxypropyl trimethyl
type side reactions and thus (hemi)cellulose degradation. ammonium chloride. Text. Res. J. 2003, 73, 1017−1023.
Under otherwise constant conditions, the degree of (7) Pei, A.; Butchosa, N.; Berglund, L. A.; Zhou, Q. Surface
substitution was linearly dependent on the reagent concen- quaternized cellulose nanofibrils with high water absorbency and
tration. A higher concentration of reagent in water was adsorption capacity for anionic dyes. Soft Matter 2013, 9, 2047−2055.
simulated by partially replacing water with THF. The (8) Cai, X.; Riedl, B.; Ait-Kadi, A. Effect of surface-grafted ionic
groups on the performance of cellulose-fiber-reinforced thermoplastic
combination of those two observations is helpful to plan and composites. J. Polym. Sci., Part B: Polym. Phys. 2003, 41, 2022−2032.
control laboratory-scale cationization assays.

■
(9) Law, K.-N.; Mao, L.; Brouillette, F.; Daneault, C. Novel use of
mechanical pulp. Appita J. 2009, 62, 118−122.
ASSOCIATED CONTENT (10) Hellwig, G.; Bischoff, D.; Rubo, A. Production of Cationic
Starch Ethers Using an Improved Dry Process. Starch/Staerke 1992,
*
S Supporting Information
44, 69−74.
The Supporting Information is available free of charge on the (11) de la Motte, H.; Westman, G. Regioselective cationization of
ACS Publications website at DOI: 10.1021/acssusche- cellulosic materials using an efficient solvent-minimizing spray-
meng.5b01752. technique. Cellulose 2012, 19, 1677−1688.

2300 DOI: 10.1021/acssuschemeng.5b01752

ACS Sustainable Chem. Eng. 2016, 4, 2295−2301
ACS Sustainable Chemistry & Engineering Research Article

(12) Kavaliauskaite, R.; Klimaviciute, R.; Zemaitaitis, A. Factors Modified Cottons: Part I: Physical and Crystalline Modifications and
influencing production of cationic starches. Carbohydr. Polym. 2008, Oxidation. Text. Res. J. 1958, 28, 382−392.
73, 665−675. (31) Olaru, N.; Olaru, L. Influence of Organic Diluents on Cellulose
(13) de la Motte, H.; Hasani, M.; Brelid, H.; Westman, G. Molecular Carboxymethylation. Macromol. Chem. Phys. 2001, 202, 207−211.
characterization of hydrolyzed cationized nanocrystalline cellulose, (32) Basch, A.; Wasserman, T.; Lewin, M. Near-infrared spectrum of
cotton cellulose and softwood kraft pulp using high resolution 1D and cellulose: A new method for obtaining crystallinity ratios. J. Polym. Sci.,
2D NMR. Carbohydr. Polym. 2011, 85, 738−746. Polym. Chem. Ed. 1974, 12, 1143−1150.
(14) Thielking, H.; Schmidt, M. In Ullmann’s Encyclopedia of (33) Schwanninger, M.; Rodrigues, J.; Fackler, K. A review of band
Industrial Chemistry; Wiley-VCH Verlag: Weinheim, Germany, 2000; assignments in near infrared spectra of wood and wood components. J.
Vol. 7, pp 381−397. Near Infrared Spectrosc. 2011, 19, 287−308.
(15) Olszewska, A.; Eronen, P.; Johansson, L.-S.; Malho, J.-M.; (34) Oberlerchner, J.; Rosenau, T.; Potthast, A. Overview of Methods
Ankerfors, M.; Lindström, T.; Ruokolainen, J.; Laine, J.; Ö sterberg, M. for the Direct Molar Mass Determination of Cellulose. Molecules 2015,
The behaviour of cationic NanoFibrillar Cellulose in aqueous media. 20, 10313.
Cellulose 2011, 18, 1213−1226. (35) Potthast, A.; Radosta, S.; Saake, B.; Lebioda, S.; Heinze, T.;
(16) Heinze, T.; Haack, V.; Rensing, S. Starch Derivatives of High Henniges, U.; Isogai, A.; Koschella, A.; Kosma, P.; Rosenau, T.;
Degree of Functionalization. 7. Preparation of Cationic 2-Hydrox- Schiehser, S.; Sixta, H.; Strlič, M.; Strobin, G.; Vorwerg, W.; Wetzel, H.
ypropyltrimethylammonium Chloride Starches. Starch/Starke 2004, Comparison testing of methods for gel permeation chromatography of
56, 288−296. cellulose: coming closer to a standard protocol. Cellulose (Dordrecht,
(17) (a) Kweon, M. R.; Bhirud, P. R.; Saskatoon, F. W. S. An Neth.) 2015, 22, 1591−1613.
Aqueous Alcoholic-Alkaline Process for Cationization of Corn and Pea (36) Yan, L.; Tao, H.; Bangal, P. R. Synthesis and Flocculation
Starches. Starch/Stärke 1996, 48, 214−220. (b) Kweon, M. R.; Behavior of Cationic Cellulose Prepared in a NaOH/Urea Aqueous
Sosulski, F. W.; Bhirud, P. R. Cationization of Waxy and Normal Corn Solution. Clean: Soil, Air, Water 2009, 37, 39−44.
and Barley Starches by an Aqueous Alcohol Process. Starch/Stärke (37) Henniges, U.; Kloser, E.; Patel, A.; Potthast, A.; Kosma, P.;
1997, 49, 59−66. Fischer, M.; Fischer, K.; Rosenau, T. Studies on DMSO-containing
(18) Wang, P.-x.; Wu, X.-l.; Xue, D.-h.; Kun, X.; Ying, T.; Du, X.-b.; carbanilation mixtures: chemistry, oxidations and cellulose integrity.
Li, W.-b. Preparation and characterization of cationic corn starch with Cellulose 2007, 14, 497−511.
a high degree of substitution in dioxane−THF−water media.
Carbohydr. Res. 2009, 344, 851−855.
(19) Ren, J. L.; Peng, F.; Sun, R. C.; Liu, C. F.; Cao, Z. N.; Luo, W.;
Tang, J. N. Synthesis of cationic hemicellulosic derivatives with a low
degree of substitution in dimethyl sulfoxide media. J. Appl. Polym. Sci.
2008, 109, 2711−2717.
(20) Ren, J. L.; Sun, R. C.; Liu, C. F.; Lin, L.; He, B. H. Synthesis and
characterization of novel cationic SCB hemicelluloses with a low
degree of substitution. Carbohydr. Polym. 2007, 67, 347−357.
(21) Ho, T. T. T.; Zimmermann, T.; Hauert, R.; Caseri, W.
Preparation and characterization of cationic nanofibrillated cellulose
from etherification and high-shear disintegration processes. Cellulose
2011, 18, 1391−1406.
(22) Montplaisir, D.; Chabot, B.; Daneault, C. Cationisation of
thermomechanical pulp fibres. Part 2: Influence on strength and
retention. Pulp Paper Can. 2006, 107, 39−42.
(23) Evans, R.; Newman, R. H.; Roick, U. C.; Suckling, I. D.; Wallis,
A. F. A. Changes in cellulose crystallinity during kraft pulping.
Comparison of infrared, X-ray diffraction and solid state NMR results.
Holzforschung 1995, 49, 498−504.
(24) Agarwal, U. P.; Ralph, S. A.; Reiner, R. S.; Baez, C. Probing
crystallinity of never-dried wood cellulose with Raman spectroscopy.
Cellulose 2016, 23, 125−144.
(25) Zou, X.; Gurnagul, N.; Uesaka, T.; Bouchard, J. Accelerated
aging of papers of pure cellulose: mechanism of cellulose degradation
and paper embrittlement. Polym. Degrad. Stab. 1994, 43, 393−402.
(26) Klemm, D.; Philipp, B.; Heinze, T.; Heinze, U.; Wagenknecht,
W. Functionalization of Cellulose; Wiley-VCH Verlag: Weinheim,
Germany, 1998; Vol. 2, p 40.
(27) Fengel, D.; Ludwig, M. Möglichkeiten und Grenzen der FTIR-
Spektroskopie bei der Charakterisierung von Cellulose. Teil 1.
Vergleich von verschiedenen Cellulosefasern und Bakterien-Cellulose.
Papier (Bingen, Ger.) 1991, 45, 45−51.
(28) Newman, R. H.; Hill, S. J.; Harris, P. J. Wide-Angle X-Ray
Scattering and Solid-State Nuclear Magnetic Resonance Data
Combined to Test Models for Cellulose Microfibrils in Mung Bean
Cell Walls. Plant Physiol. 2013, 163, 1558−1567.
(29) Nelson, M. L.; O’Connor, R. T. Relation of certain infrared
bands to cellulose crystallinity and crystal lattice type. Part II. A new
infrared ratio for estimation of crystallinity in celluloses I and II. J.
Appl. Polym. Sci. 1964, 8, 1325−1341.
(30) O’Connor, R. T.; DuPré, E. F.; Mitcham, D. Applications of
Infrared Absorption Spectroscopy to Investigations of Cotton and

2301 DOI: 10.1021/acssuschemeng.5b01752

ACS Sustainable Chem. Eng. 2016, 4, 2295−2301