Carbohydrate Polymers 186 (2018) 358–366

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Cationic xylan- (2-methacryloyloxyethyl trimethyl ammonium chloride) T

polymer as a flocculant for pulping wastewater
⁎ ⁎
Xiaoqian Chena,b, Chuanling Sia, , Pedram Fatehib,
a
Tianjin Key Laboratory of Pulp and Paper, Tianjin University of Science & Technology, Tianjin, China
b
Chemical Engineering Department, Lakehead University, Thunder Bay, ON, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: In this work, pulp mill wastewater was treated with a novel polymer flocculant, which was synthesized through
Xylan polymerizing (2-methacryloyloxyethyl) trimethyl ammonium chloride (DMC) and xylan. Xylan-DMC polymer
Carbohydrates removed 94.5% of turbidity, 61.7% of chemical oxygen demand (COD), 45.0% of lignin, 65.7% of sugar and
Wastewater 73.5% of biological oxygen demand (BOD) from the wastewater at the polymer concentration of 500 mg/L. The
Flocculation
flocculation mechanism was fundamentally assessed with determining hydrodynamic size and chord length of
Hydrodynamic size
flocs in the wastewater at different dosages of xylan-DMC polymer in the system. By adding polymers, the
Chord length
number of flocs with a small chord length (10–50 μm) decreased, while that with a large chord length
(150–300 μm) increased, indicating the small particles agglomerated via bridging induced by the polymer. The
sedimentation of formed flocs was quantitatively investigated by a vertical scan analyzer and the results depicted
that the flocs could settle readily from the system.

1. Introduction grafted cassava starch (Razali & Ariffin, 2015), PDMC grafted starch
(PDMC-g-STC) (Wang, Yuan, Wang, & Yu, 2013) and PDMC grafted
Wastewater produced in the pulping industry contains fine sus- chitosan (PDMC-g-chitosan) (Wang, Chen, Ge, & Yu, 2007; Wang, Chen,
pended and dissolved solids, inorganic and organic particles and other Yuan, Sheng, & Yu, 2009) were produced and their flocculation per-
impurities, which are difficult to settle due to their small sizes and formance was assessed in kaolin suspensions or various pulping and
surface charges. The removal of these colloidal particles from the municipal wastewaters. The turbidity of a kaolin suspension decreased
wastewater becomes a serious challenge for industry (Lee, Robinson, & to 25 NTU from 600 NTU when 0.1 mg/L of PDMAC-g-STC was added
Chong, 2014; Nasser & James, 2006). Colloids in the wastewater are to the suspension (Wang et al., 2013). Wang et al. (2009) reported that
usually negatively charged, and hence flocculation process with ca- 99.4% of turbidity, 81.3% of lignin and 90.7% of COD could be re-
tionic flocculants is one of the most widely used techniques for treating moved from a pulp mill wastewater at 17.8 mg/L concentration of
wastewater (Teh, Budiman, Shak, & Wu, 2016). PDMC-g-chitosan in the effluent. In the study conducted by Pal, Ghorai,
However, commonly used synthetic cationic polymers, such as poly Dash, Ghosh, and Udayabhanu (2011), 0.5 mg/L of polyacrylamide
(2-methacryloyloxyethyl) trimethyl ammonium chloride (PDMC) and grafted carboxymethyl guar gum (PAM-g-CMG) was able to decrease
poly (diallyldimethyl ammonium chloride) (PDADMAC) are expensive the turbidity, suspend solids and COD from 64 to 9 NTU, 350 to 30 mg/
and resistant to biodegradation, which hampers their uses in industry L and 540 to 210 mg/L, respectively (Pal et al., 2011). In our previous
and introduces environmental problems (Bolto & Gregory, 2007). In study, a novel polymer was synthesized via polymerizing DMC and
this respect, there is an increasing demand for more environmentally xylan (xylan-DMC) and the copolymer was utilized as a flocculant in
friendly and effective flocculants in industry. removing anionic dyes from solution systems as well as kaolin and
Recently, the flocculants produced by grafting cationic synthetic bentonite from suspension systems (Wang, Hou, Kong, & Fatehi, 2015;
polymers onto biodegradable polysaccharide backbones have received Wang, Konduri, Hou, & Fatehi, 2016). The results indicated that xylan-
great attentions (Constantin, Mihalcea, Oanea, Harabagiu, & DMC removed 98% of azo-dye (Orange 16) at the concentration of
Fundueanu, 2011; Dax et al., 2014; Wang et al., 2014). In the past, 160 mg/L in the solution (S. Wang et al., 2015). The removals of kaolin
many flocculants, such as hydroxypropyl methyl cellulose grafted with and bentonite particles were 78.5% and 97.1%, respectively, at the
polyacrylamide (HPMC-g-PAM) (Das, Ghorai, & Pal, 2013), PDADMAC xylan-DMC concentration of 16 mg/L (Wang et al., 2016). However, the

⁎
Corresponding authors.
E-mail addresses: sichli@tust.edu.cn (C. Si), pfatehi@lakeheadu.ca (P. Fatehi).

https://doi.org/10.1016/j.carbpol.2018.01.068
Received 24 September 2017; Received in revised form 17 January 2018; Accepted 20 January 2018
Available online 03 February 2018
0144-8617/ © 2018 Elsevier Ltd. All rights reserved.
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

properties of pulping wastewater, dye solutions and clay suspensions ratio and the pH was adjusted to 7 with 0.1 mol/L NaOH. Then, 0.03 g
are very different. Pulping wastewater contains polymeric and mono- of potassium persulfate was added to the system as an initiator and the
meric organic materials with slightly negative zeta potential originating polymerization was performed at 80 °C under nitrogen environment for
from carboxylate group attached to the organic materials (Razali, 3 h. The solution was cooled to room temperature and mixed with
Ahmad, Ahmad, & Ariffin, 2011). Dye pigments are generally water ethanol to precipitate xylan-DMC polymer. The suspension was cen-
soluble and highly charged (Fang, Cheng, & Xu, 2010) and clay sus- trifuged at 3500 rpm for 10 min and the precipitated xylan-DMC
pensions contain inorganic particles that are substantially larger than polymer was dried at 105 °C and used as a flocculant for removing
the particles in pulping wastewaters (Wang et al., 2016). It is well organics from the wastewater.
known that the properties of particles significantly impact the process
employed for their separation from colloidal systems. As there was no 2.3. Flocculation
report regarding the application of this novel polymer (xylan-DMC) in
treating pulping wastewater, it was not clear if this polymer could act as In this set of experiments, 10 g/L of xylan-DMC and xylan solutions
a flocculant for pulping wastewater. The first objective of this study was was prepared via mixing the polymers with deionized water at room
to evaluate if xylan-DMC could be an efficient flocculant for pulping temperature. Different dosages (0–800 mg/L based on TMPL) of the
wastewater. polymer solutions were added to 50 mL of TMPL in 125 mL Erlenmeyer
Flocculation phenomenon occurs when small particles agglomer- flasks. Then, the flasks were sealed and shaken at room temperature
ated into larger flocs, which helps their settlement and separation from and 100 rpm for 30 min. After flocculation, the suspensions were cen-
solutions. Monitoring the growth of the aggregated particles, the be- trifuged at 3000 rpm for 5 min and the filtrates were collected for
havior of agglomerated constituents in solutions and the sedimentation analyses.
performance of the aggregated particles are crucial in assessing and The numbers and size of particles in the suspensions at different
understanding flocculation processes (Blanco, De La Fuente, Negro, polymer dosages were analyzed using a focused beam reflectance
Minte, & Tijero, 2002; Zhou & Franks, 2006). One method to identify measurement (FBRM) (Particle Track E25 probe Mettler-Toledo
the mechanism of flocculation is to evaluate the de-flocculation/re- AutoChem, USA). The methodology of FBRM is based on the applica-
flocculation of aggregated particles formed at different shear forces in tion of a highly-focused laser beam scanning across particles in a sus-
solutions (Blanco, Negro, & Tijero, 2002). In the present study, focused pension (Wang et al., 2014). This technique provides the chord length
beam reflectance measurement (FBRM) was used to monitor the size, of the particles in a real time (Negro et al., 2006). In this set of ex-
numbers and reversibility of formed flocs in flocculation process periments, the FBRM probe was inserted into a 500 mL glass beaker
(Hukkanen & Braatz, 2003; Kail, Marquardt, & Briesen, 2009; Marugán containing 200 mL of TMPL samples. Different dosages of xylan-DMC
et al., 2008; Negro, Sánchez, Fuente, Blanco, & Tijero, 2006), and a polymers were gradually added to the system, and the chord lengths of
vertical scan analyzer was employed to study the sedimentation per- flocs were monitored at an agitation intensity of 250 rpm as described
formance of the formed flocs in the wastewater system (Kaombe, Lenes, in the previous study (Fatehi, Gao, Sun, & Dashtban, 2016). In another
Toven, & Glomm, 2013). set of experiments, 500 mg/L of xylan-DMC polymer was maintained in
For the first time, the application of xylan-DMC polymer in the a 200 mL TMPL sample that was stirred at 250 rpm and then the
pulping wastewater was analyzed comprehensively, and the floccula- measurement was conducted. Afterward, the stirring intensity was in-
tion mechanism was investigated fundamentally in this work. creased to 750 rpm to break the formed flocs (i.e., de-flocculation
Furthermore, the changes in the particle size and the settlement of the process). After reaching a steady state condition, the agitation intensity
flocs were monitored, which are critical in developing an industrially was reduced to 250 rpm to evaluate the re-flocculation of particles in
attractive flocculation process for treating wastewater. the FBRM analysis.

2. Methods and materials 2.4. Sedimentation analysis

2.1. Materials The sedimentation of TMPL suspension in the presence and absence
of xylan-DMC polymer was investigated using a vertical scan analyzer,
Xylan from beechwood, [2-(methacryloyloxy) ethyl] trimethyl am- Turbiscan Lab Expert (Formulaction Inc., France). In this set of ex-
monium chloride solution (DMC), 80 wt.% in H2O, sodium chloride, periments, 20 mL of TMPL suspensions with desired flocculant dosages
potassium persulfate (≥99.0%), ethanol (99.5%) and sodium nitrate was contained in cylindrical glass cells. The cells were vertically
(> 99.0%) were all purchased from Sigma-Aldrich company. Anionic scanned with electro luminescent diode light source at 880 nm wave-
polyvinyl sulfate (PVSK) was provided from Wako Pure Chem. Ltd. length every 25 s for 1 h at 30 °C. Two synchronous optical sensors re-
Japan and diluted to 0.0055 mol/L prior to use. Polydiallyldimethyl ceive light transmitted through the sample (180° from the incident
ammonium chloride (PDADMAC) was obtained from Sigma Aldrich light, transmission sensor), and light backscattered by the sample (45°
Company and diluted to 0.0055 mol/L prior to use. Thermomechanical from the incident light, backscattering detector). The light passes
pulping spent liquor, which was generated via hot water treatment of through a transmission zone of the samples, while it cannot pass
wood chips prior to their refining, was collected from a pulp mill lo- through the sediment zone of the samples (He, Zhang, & Fatehi, 2016).
cated in Northern Ontario, Canada. This wastewater was denoted TMPL The higher transmission zone implies a clearer suspension due to the
in this work, and kept in a refrigerator at 4 °C in closed plastic barrels sedimentation of particles (Konduri & Fatehi, 2017). Based on the data
prior to use. Silicon oil was supllied by Formulaction company and used collected, the clarity in the transmission zone of the samples was as-
as received for calibrating the vertical scan analyzer. sessed with respect to the transition of silicon oil used for calibrating
the instrument and the results were analyzed by Turbisoft 2.1 software
2.2. Preparation of cationic xylan-DMC polymer (Kaombe et al., 2013; Qin, Yu et al., 2016).

Xylan-DMC was synthesized as described in a previous study (S. 2.5. Molecular weight analysis
Wang et al., 2015). 1 g of xylan was dissolved in 40 mL of water in a
250 mL three-neck glass flask to generate 25 g/L xylan solution. The Xylan and xylan-DMC polymer (at 5 g/L concentration) in 0.1 mol/L
flask was kept in a water bath at 30 °C and was deoxygenated using sodium nitrate solutions were prepared by stirring the solutions of the
nitrogen atmosphere for 30 min. Then, a certain amount of DMC was polymers at 300 rpm for 24 h, and then the solution was filtered with a
added drop-wisely to the solution to generate a 3 mol/mol DMC/xylose 0.2 μm nylon filter. The filtered solutions were used for molecular

359
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

weight analysis using a gel permeation chromatography (Malvern GPC oxygen concentration in the samples were measured using LBDO 101
max VE2001 Module + Viscotek TDA 305) with reflective index (RI), probe (HACH Company, USA) and the bottles were stored in an in-
viscometer detectors and Poly Analytic PAA206 and PAA203 columns. cubator at 20 °C for 5 days. After incubation, the amount of dissolved
0.1 mol/L sodium nitrate solution was used as a solvent and eluent at oxygen concentration was determined for the BOD analysis (Latif &
35 °C with a flow rate of 0.7 mL/min. Poly (ethylene oxide) was used as Dickert, 2015).
standard for the instrument calibration. For total sugar analysis, the samples were hydrolyzed with 4 wt.%
sulfuric acid at 121 °C in an oil bath (AC200, Thermo Fisher Scientific
2.6. Charge density analysis Inc., USA). Then, the concentration of monosaccharides was de-
termined using an ion chromatography (IC) unit that was equipped
Initially, 0.05 g of xylan or xylan-DMC polymer was dissolved in with a Thermo Scientific Electrochemical detector. The acid hydrolysis
50 g of water, and the solution was incubated in a water bath shaker was conducted to convert polysugars to monosugars as the IC instru-
(Innova 3100, Brunswick Scientific, Edison, USA) and shaken (150 rpm) ment cannot detect polysugars (Fatehi, Ryan, & Ni, 2013; Fatehi et al.,
at 30 °C for 2 h. Then, the charge density of the samples was measured 2016). Therefore, the total monosugars in hydrolyzed samples was re-
using a particle charge detector (Mutek PCD 04, Germany) with ported as the sugar content of the samples. CarboPacTM SA10 column
0.0055 mol/L PVSK or PDADMAC standard solutions. (Dionex Corporation, Canada) was set at 30 °C, and 1.0 mmol/L of KOH
solution was used as eluent in this analysis. The flow rate was set at
2.7. Elemental analysis 1.2 mL/min. The lignin content of the samples was determined using a
UV/Vis spectrophotometry (Genesys 10S, Thermo Scientific Inc., USA)
The xylan and xylan-DMC polymers were dried at 60 °C in an oven, at the wavelength of 205 nm. The turbidity of the samples was assessed
and approximately 0.02 g of dried samples was transferred into the by using a turbidimeter (2100AN, HACH Company, USA) at room
carousel chamber of the elemental analyzer (Elementar vario EL, temperature (Fatehi et al., 2013, 2016).
Germany). The combustion was carried out at 1200 °C, and the gener-
ated gasses were reduced to analyze carbon, hydrogen, and nitrogen 3. Results and discussion
compounds.
3.1. Characterization of xylan-DMC polymer
2.8. Fourier transform infrared (FTIR)
The properties of xylan and xylan-DMC polymer are listed in
The FTIR analysis of xylan and xylan-DMC polymers was conducted Table 1. The reaction scheme of xylan and DMC is presented in sup-
using a Fourier transform infrared spectrophotometer (Bruker Tensor plementary materials (Fig. S1). The charge density of xylan was
37, Germany). In this measurement, 0.1 g of dried samples was used for −0.09 meq/g and the charge density of xylan-DMC polymer was
analysis and each spectrum was recorded in 32 scans with a resolution +2.98 meq/g. The carbon and hydrogen contents of the xylan-DMC
of 4 cm−1 in the range of 600–4000 cm−1. polymer were 44.97% and 8.76%, respectively, which were higher than
those of xylan (40.13% and 6.41%). These results are attributed to the
2.9. Zeta potential and hydrodynamic size analyses higher carbon and hydrogen contents of DMC segment that induced the
overall carbon and hydrogen contents of the xylan-DMC polymer dif-
The zeta potential and hydrodynamic size of particles in TMPL be- ferent from those of xylan. Xylan contained no nitrogen. As can be seen
fore and after flocculation were measured. In this analysis, the TMPL in Table 1, the nitrogen content of xylan-DMC polymer was 4.12%,
containing different concentrations of xylan-DMC polymer and NaCl which was originated from the quaternary ammonium group of the
was prepared via shaking the mixtures at 100 rpm for 30 min, the mixed DMC in xylan-DMC polymer. Furthermore, xylan-DMC polymer had a
samples were diluted 20 times with a 1 mmol/L KCl solution without higher molecular weight (254,862 g/mol) than xylan (37,427 g/mol),
filtration. A NanoBrook PALS (Brookhaven Inc., USA) was used to as- which was consistent with the previous work and these changes in-
sess both zeta potential of the samples and hydrodynamic size of the dicated that the polymerization of xylan with DMC was successfully
flocs in the samples at 25 °C. In another set of experiment, the pH of achieved (S. Wang et al., 2015). Other properties, including hydro-
TMPL was adjusted to different values ranging from 1.88 to 10.5 via dynamic diameter, viscosity, grafting ratio of the polymer were re-
using 1 mol/L of NaOH or H2SO4 solution, and then 500 mg/L xylan- ported in our previous work (S. Wang et al., 2015).
DMC polymer was added. After 30 min of mixing at 100 rpm, the TMPL Functional groups in xylan-DMC polymer were characterized by
samples before and after the polymer addition were used in the zeta FTIR analysis in Fig. 1. The new characteristic bands at 1718 cm−1
potential and hydrodynamic size analyses. All experiments were per- assigned to the C]O stretching vibration and 1472 cm−1 assigned to
formed in triplicates and the average values were reported in the study. the CeN stretching vibration in the xylan-DMC polymer (Fang et al.,
2010; Wang et al., 2013), which confirmed the polymerization of xylan
2.10. Wastewater analysis and DMC.

The chemical oxygen demand (COD), biological oxygen demand 3.2. Properties of TMPL wastewater
(BOD) and turbidity of the wastewater samples before and after floc-
culation were determined according to the standard methods The characteristics of TMPL used in this experiment are presented in
(Federation & APHA, 2005). For the COD analysis, 2 mL of TMPL
samples were mixed with the premixed chemical COD kits (CHEMetrics Table 1
Inc., USA) and then incubated in a block digester at 150 °C for 2 h. After Properties of xylan and xylan-DMC copolymer.
cooling for 30 min, the absorbency of the sample was measured at
Sample Xyaln Xylan-DMC
620 nm (Oveissi, Sitter, & Fatehi, 2016). The BOD analysis was per-
formed using 300 mL incubation bottles with glass stoppers. Initially, Charge density, meq/g −0.09 +2.98
the nutrient enriched water was prepared by adding a BOD nutrient Mw, g/mol 37,427 254,862
buffer formulation to the 4 L dilution water and shaken vigorously for Mn, g/mol 21,834 100,494
Carbon, wt% 40.13 44.97
1 min to dissolve the nutrients. Then, 5 mL TMPL samples before and
Hydrogen, wt% 6.41 8.76
after flocculation were added to the BOD incubation bottles and the Nitrogen, wt% 0 4.12
bottles were filled with the nutrient enriched water. The dissolved

360
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

Fig. 1. FTIR spectra of xylan and xylan-DMC polymer.

Table 2. It was found that the TMPL contained 1.47 g/L of lignin, polymer, the zeta potential of TMPL was increased and reached neutral
1.42 g/L of sugars and 0.87 g/L of BOD. The concentration of COD was at 500 mg/L dosage and a higher dosage of the polymer resulted in a
high (4470 mg/L). A high turbidity (1257 NTU) and suspended solid charge reversion.
(356 mg/L) were also observed (Guibai & Gregory, 1991). The zeta The correlation among the zeta potential, turbidity and COD would
potential of TMPL was −30.7 mV, and the hydrodynamic size of its provide insights on the mechanism of flocculation. Fig. 3b shows the
particles was 423 nm. turbidity and COD of TMPL as a function of its zeta potential when
different dosages of xylan-DMC polymer were added to TMPL. The
3.3. Application of xylan-DMC polymer to TMPL turbidity and COD contents were reduced linearly when the zeta po-
tential of TMPL/xylan-DMC system approached zero implying that
The flocculation performance of xylan and xylan-DMC polymer as charge neutralization played a role in the removal of the formed flocs.
flocculants in TMPL was investigated. Fig. 2 shows the effect of dosage It was stated in the literature that the flocculation mechanism could
on the COD, BOD, lignin, sugar and turbidity of TMPL. It was ob- be studied via monitoring the variations in the zeta potential and hy-
servable that xylan had poor flocculation capability because of its drodynamic size of a system containing a flocculant at different pHs
slightly negative charge density and low molecular weight (Table 1). An (Qin, Yang et al., 2016). Fig. 4 shows the zeta potential and hydro-
efficient flocculation performance was observed for xylan-DMC dynamic sizes of particles in the TMPL before and after adding xylan-
polymer under the same experimental conditions. Similar BOD (73.5%) DMC polymer at 500 mg/L dosage (i.e., the optimum dosage) and dif-
and sugar (75%) removals were obtained (Fig. 2), which implied that ferent pHs. It is observed that TMPL had a negative zeta potential over
the removal of BOD and sugars was related, and this is attributed to the the entire pH range and it decreased with increasing pH due to the
biodegradability of sugars. However, the results depicted that sugar promoted ionization of colloidal substances in TMPL at a high pH (Z.
compounds were still present after flocculation in TMPL. These residual Wang et al., 2015). The increased ionization also induced the repulsion
sugars were most likely attached to lignin by the lignin-carbohydrate force between the particles and within the particle structure stretching
linkages and were probably reluctant to be removed (Fu & Lucia, 2003; the particles and thus enlarging the hydrodynamic size of particles
Tunc, Lawoko, & van Heiningen, 2010). It is also observable that lignin (Fig. 4) (Fatehi, Xiao, & van de Ven, 2011). After adding xylan-DMC
and COD removals were incomplete, but turbidity removal was sub- polymer, the neutralization of the particles in the TMPL by xylan-DMC
stantial. It can be stated that the residual organic compounds (in lignin, polymer raised the zeta potential of TMPL/xylan-DMC system at any pH
sugar or lignin-carbohydrate complex form) were probably small and (Fig. 4), but the zeta potential values generally decreased with in-
soluble in TMPL that they did not contribute to the turbidity (Fatehi creasing pH (Das & Somasundaran, 2001). In the pH range above the
et al., 2016). The removals reached the highest at the concentration of isoelectric point (pH 6.6), the zeta potential became negative in the
500 mg/L xylan-DMC polymer in the TMPL, and thus this dosage was TMPL/xylan-DMC system and the hydrodynamic size of the flocs in the
selected for further investigation. TMPL/xylan-DMC system increased significantly. At a high pH, the
presence of (OH−) would hamper the impact of the quaternary am-
monium group of xylan-DMC polymer for adsorption and thus for
3.4. Flocculation mechanism neutralization of the TMPL. As the hydrodynamic sizes of the particles
in the TMPL/xylan-DMC system increased with pH, and the size of
Generally, when polymers with an opposite charge to that of a particles was substantially larger than that in the TMPL/xylan-DMC
colloid are added to a colloidal system, the polymers will interact with system, polymer bridging must have played a primary role (Ho, Norli,
components of the colloidal system and neutralize the overall charges Alkarkhi, & Morad, 2010; Suopajärvi, Liimatainen, Hormi, & Niinimäki,
on the particles and thus change the zeta potential of the system. Fig. 3a 2013). The formation of large flocs in an alkaline environment was also
shows the zeta potential of the TMPL as a function of the flocculant reported to be led by polymer bridging in other studies (Lee et al., 2014;
dosage. Xylan did not impact the zeta potential of TMPL, as it had a Yang et al., 2012; Wang et al., 2007).
limited anionic charge density. However, by adding more of xylan-DMC

Table 2
Characterization of TMPL.

Parameter COD, mg/L BOD, mg/L Lignin, g/L Sugar, g/L Turbidity, NTU pH Zeta potential, mV Particle size, nm Suspend solid, mg/L

TMPL 4470 873 1.47 1.02 957 6.6 −30.7 423 356

361
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

3.5. Flocculation process

The square-weighted counts of flocs generated at different dosages

of the xylan-DMC polymer in the TMPL were presented as a function of
the chord length distribution of the flocs in Fig. 5a. For the blank TMPL
sample, the average particle size was 40 μm (Kail, Briesen, &
Marquardt, 2008). With adding xylan-DMC polymer, the average chord
length and total counts of particles increased, and the increase in the
concentration of xylan-DMC from 500 to 800 mg/L did not change the
chord length distributions of the flocs. These results are consistent with
that of another work in which the floc size increased when flocculants
were added, but no change was observed when the total dosage was
higher than the optimal (Blanco, Negro et al., 2002). The results also
showed that most of the particles had chord lengths between 10 and
300 μm after flocculation.
Fig. 5b shows the counts of flocs with three chord lengths. By in-
creasing the dosage of xylan-DMC polymer, an increase in the number
of counts for all particle sizes was observed. When the concentration
was increased from 400 to 500 mg/L, the number of particles in the size
range of 10 and 50 μm decreased and that in the ranges of 50 and
150 μm as well as of 150 and 300 μm increased, suggesting that smaller
particles collided into larger particles (Yao, Nan, & Chen, 2014). A
concentration of more than 500 mg/L did not increase the particle size
to a larger degree indicating the maximum size of flocs in the TMPL
(Zhou & Franks, 2006). Similar results were reported about the counts
of flocs in a papermaking effluent when cationic poly acryl amide
(CPAM) was used as a flocculant (Blanco, De La Fuente et al., 2002).
The number of flocs smaller than 30 μm decreased while the number of
flocs in the range of 200 and 500 μm increased at a higher dosage of
CPAM. The number of flocs larger than 500 μm increased and the
number of flocs in the range of 200–500 μm decreased when more
CPAM was added to the effluent (Blanco, De La Fuente et al., 2002). It
can also be seen that the counts of small particles (10–50 μm) increased
and the counts of large particles (150–300 μm) decreased slightly, when
the dosage of xylan-DMC increased from 500 mg/L to 800 mg/L. In the
xylan-DMC polymer is presented in excess in TMPL, small flocs with a
net charge would be created and these small flocs might repel each
other, preventing the creation of large flocs in the TMPL system (Sher,
Malik, & Liu, 2013). It also indicated that 500 mg/L xylan-DMC
polymer was the optimum dosage for bridging flocculation.
In order to study the fundamentals of floc formation, the floccula-
tion of TMPL with xylan-DMC polymer at 500 mg/L dosage was studied
at different shear intensities in Fig. 5c. When xylan-DMC was added to
TMPL at the minute of 5, the number of flocs in all three ranges in-
creased, which also increased the median size of flocs. When the shear
rate was increased from 250 rpm to 750 rpm, the number of flocs in the
small and medium ranges were increased but that in 150–300 μm de-
creased, implying that the bridges between the large particles broke at
this high shear rate. The stirring intensity was reduced to the initial
level after 6 min and re-flocculation took place. Interestingly, the
number of small and medium flocs in the range of 10–50 μm and
50–150 μm dropped, but the number of large flocs in 150–300 μm range
increased. These results show that the small particles joined together
forming larger aggregates increasing the mean chord size in the re-
flocculation step. It can be seen from Fig. 5c that the number of small
flocs were higher than those before the shear increase, but the medium
and large flocs numbers did not reach their original value of 2900 and
200 μm2/s after re-flocculation. As the number of small sized flocs in-
creased, and the number of medium and large flocs decreased after re-
flocculation; it may be concluded that polymer bridging was the main
Fig. 2. Effect of flocculant dosage on COD, lignin, sugar, BOD and turbidity of TMPL. mechanism of flocculation, and basically increasing the share rate
permanently broke large flocs to small ones (Biggs, Habgood, &
Jameson, 2000). The permanent change in the size of large flocs ren-
dered this flocculation process irreversible.

362
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

Fig. 3. a) Zeta potential of TMPL as function of the polymer dosage, b)

Effect of the zeta potential on the turbidity and COD concentration of
TMPL when different dosages of xylan-DMC were added to TMPL.

3.6. Sedimentation characteristics implying that it was highly stabilized at this dosage. The clarity of
suspensions increased as time elapsed, especially for TMPL suspension
The sedimentation performance of TMPL at different concentrations with 500 mg/L xylan-DMC dosage, indicating that a significant particle
of xylan-DMC polymer was investigated. The transmission zone of sedimentation occurred after particle flocculation. In this case, 91.72%
suspension samples over 60 min are presented in Fig. 6. The transmis- of TMPL suspension became transparent within 20 min with 500 mg/L
sion of TMPL was low (near zero) indicating a strong turbidity for the xylan-DMC polymer. The rapid sedimentation of TMPL at 500 mg/L
sample. 100 mg/L concentration of xylan-DMC polymer in the TMPL xylan-DMC concentration illustrated that the formed flocs could settle
had no significant effect on the clarity (i.e., transmission of light), readily from the suspension (Das et al., 2013; Fang et al., 2010; Lee

Fig. 4. Zeta potential and hydrodynamic sizes of TMPL as a function of pH

with or without 500 mg/L of xylan-DMC.

363
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

Fig. 5. a) Particle size distribution of flocs in TMPL at different dosages of

xylan-DMC polymer; b) The square weighted counts of particles for three
different chord length sizes as a function of xylan-DMC dosage; and c) The
size of chord length for three ranges of flocs as a function of treatment time
when the shear rate changed from 250 rpm to 750 rpm and vice versa
(xylan-DMC dosage of 500 mg/L).

et al., 2014). Furthermore, a lower light transmission was observed in hydrodynamic size and chord length analyses indicated that the charge
TMPL at 800 mg/L than at 500 mg/L xylan-DMC polymer dosage, neutralization and polymer bridging mechanisms played roles in the
which was due to the fact that more of small particles generated in the flocculation process. An increase in the number of counts for all the
system at this dosage (800 mg/L) (Fig. 5c). These particles were diffi- particle sizes was observed with increasing the dosage of xylan-DMC
cult to settle down. These results imply that the flocs formed at the polymer in TMPL. At a dosage higher than 400 mg/L, the number of
optimum concentration (500 mg/L) were able to settle much quicker flocs in the chord length range of 10 and 50 μm decreased and that in
than the flocs formed at other dosages. the ranges of 50–150 and 150–300 μm increased, suggesting that
smaller particles collided into larger particles. The de-flocculation/re-
flocculation analysis confirmed the flocculation of particles via bridging
4. Conclusions mechanism. The rapid sedimentation of TMPL with 500 mg/L of xylan-
DMC polymer confirmed that the formed flocs settled from the system
The flocculation efficiency of xylan-DMC polymer in TMPL waste- readily, which makes xylan-DMC polymer an attractive flocculant for
water was investigated. At 500 mg/L of xylan-DMC polymer, the re- industrial use.
moval rates for turbidity, COD, lignin, sugar and BOD were 94%,
61.7%, 45.0%, 65.7%, and 73.5%, respectively. The zeta potential,

364
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

Fig. 6. The transmission of TMPL/xylan-DMC system (as % of that of si-

licon oil as standard) as a function of time at different concentrations of
xylan-DMC in the TMPL suspension.

Acknowledgements Chemistry Research, 42(19), 4269–4276.

Guibai, L., & Gregory, J. (1991). Flocculation and sedimentation of high-turbidity waters.
Water Research, 25(9), 1137–1143.
This work was supported by NSERC, CRC, CRIBE, NOHFC programs He, W., Zhang, Y., & Fatehi, P. (2016). Sulfomethylated kraft lignin as a flocculant for
and the Chinese Academy of Forestry [grant number TGB2016002]. cationic dye. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 503,
This work was also supported by National Key Research and 19–27.
Ho, Y. C., Norli, I., Alkarkhi, A. F., & Morad, N. (2010). Characterization of biopolymeric
Development Program of China [Grant No. 2017YFB0307903], NSERC, flocculant (pectin) and organic synthetic flocculant (PAM): A comparative study on
CRC and CRIBE NOHFC programs. treatment and optimization in kaolin suspension. Bioresource Technology, 101(4),
1166–1174.
Hukkanen, E. J., & Braatz, R. D. (2003). Measurement of particle size distribution in
Appendix A. Supplementary data
suspension polymerization using in situ laser backscattering. Sensors and Actuators B:
Chemical, 96(1), 451–459.
Supplementary data associated with this article can be found, in the Kail, N., Briesen, H., & Marquardt, W. (2008). Analysis of FBRM measurements by means
of a 3D optical model. Powder Technology, 185(3), 211–222.
online version, at https://doi.org/10.1016/j.carbpol.2018.01.068.
Kail, N., Marquardt, W., & Briesen, H. (2009). Estimation of particle size distributions
from focused beam reflectance measurements based on an optical model. Chemical
References Engineering Science, 64(5), 984–1000.
Kaombe, D. D., Lenes, M., Toven, K., & Glomm, W. R. (2013). Turbiscan as a tool for
studying the phase separation tendency of pyrolysis oil. Energy & Fuels, 27(3),
Biggs, S., Habgood, M., & Jameson, G. J. (2000). Aggregate structures formed via a 1446–1452.
bridging flocculation mechanism. Chemical Engineering Journal, 80(1), 13–22. Konduri, M. K., & Fatehi, P. (2017). Dispersion of kaolin particles with carbox-
Blanco, A., De La Fuente, E., Negro, C., Monte, M. C., & Tijero, J. (2002). Focused beam ymethylated xylan. Applied Clay Science, 137, 183–191.
reflectant measurement as a tool to measure flocculation. Tappi Journal, 1(10), Latif, U., & Dickert, F. L. (2015). Biochemical oxygen demand (BOD). In L. M. Moretto, &
14–20. K. Kalcher (Eds.). Environmental analysis by electrochemical sensors and biosensors (pp.
Blanco, A., Negro, C., & Tijero, J. (2002). Flocculation monitoring: Focused beam re- 729–734). New York: Springer.
flectance measurement as a measurement tool. The Canadian Journal of Chemical Lee, C. S., Robinson, J., & Chong, M. F. (2014). A review on application of flocculants in
Engineering, 80(4), 1–7. wastewater treatment. Process Safety and Environmental Protection, 92(6), 489–508.
Bolto, B., & Gregory, J. (2007). Organic polyelectrolytes in water treatment. Water Marugán, J., Calles, J. A., Dufour, J., Giménez-Aguirre, R., Peña, J. L., & Merino-García,
Research, 41(11), 2301–2324. D. (2008). Characterization of the asphaltene onset region by focused-beam laser
Constantin, M., Mihalcea, I., Oanea, I., Harabagiu, V., & Fundueanu, G. (2011). Studies on reflectance: A tool for additives screening. Energy & Fuels, 23(3), 1155–1161.
graft copolymerization of 3-acrylamidopropyl trimethylammonium chloride on Nasser, M. S., & James, A. E. (2006). The effect of polyacrylamide charge density and
pullulan. Carbohydrate Polymers, 84(3), 926–932. molecular weight on the flocculation and sedimentation behaviour of kaolinite sus-
Das, K. K., & Somasundaran, P. (2001). Ultra-low dosage flocculation of alumina using pensions. Separation and Purification Technology, 52(2), 241–252.
polyacrylic acid. Colloids and Surfaces A: Physicochemical and Engineering Aspects, Negro, C., Sánchez, L. M., Fuente, E., Blanco, Á., & Tijero, J. (2006). Polyacrylamide
182(1), 25–33. induced flocculation of a cement suspension. Chemical Engineering Science, 61(8),
Das, R., Ghorai, S., & Pal, S. (2013). Flocculation characteristics of polyacrylamide grafted 2522–2532.
hydroxypropyl methyl cellulose: An efficient biodegradable flocculant. Chemical Oveissi, F., Sitter, T., & Fatehi, P. (2016). PDADMAC as a flocculant for lignosulfonate of
Engineering Journal, 229, 144–152. NSSC pulping process. Biotechnology Progress, 32(3), 686–691.
Dax, D., Chávez, M. S., Xu, C., Willför, S., Mendonça, R. T., & Sánchez, J. (2014). Cationic Pal, S., Ghorai, S., Dash, M. K., Ghosh, S., & Udayabhanu, G. (2011). Flocculation
hemicellulose-based hydrogels for arsenic and chromium removal from aqueous so- properties of polyacrylamide grafted carboxymethyl guar gum (CMG-g-PAM) syn-
lutions. Carbohydrate Polymers, 111, 797–805. thesised by conventional and microwave assisted method. Journal of Hazardous
Fang, R., Cheng, X., & Xu, X. (2010). Synthesis of lignin-base cationic flocculant and its Materials, 192(3), 1580–1588.
application in removing anionic azo-dyes from simulated wastewater. Bioresource Qin, Y., Yang, D., Gu, F., Li, X., Xiong, W., & Zhu, J. Y. (2016). Biorefinery lignosulfonates
Technology, 101(19), 7323–7329. as a dispersant for coal water slurry. Sustainable Chemical Processes, 4(1), 1–8.
Fatehi, P., Xiao, X., & van de Ven, T. (2011). Quantitative analysis of cationic poly (vinyl Qin, Y., Yu, L., Wu, R., Yang, D., Qiu, X., & Zhu, J. (2016). Biorefinery lignosulfonates
alcohol) diffusion into the hairy structure of cellulose fiber pores: Charge density from sulfite-pretreated softwoods as dispersant for graphite. ACS Sustainable
effect. Langmuir, 27(22), 13489–13496. Chemistry & Engineering, 4(4), 2200–2205.
Fatehi, P., Ryan, J., & Ni, Y. (2013). Adsorption of lignocelluloses of model pre-hydrolysis Razali, M. A. A., & Ariffin, A. (2015). Polymeric flocculant based on cassava starch grafted
liquor on activated carbon. Bioresource Technology, 131, 308–314. poly diallyldimethyl ammonium chloride: Flocculation behavior and mechanism.
Fatehi, P., Gao, W., Sun, Y., & Dashtban, M. (2016). Acidification of prehydrolysis liquor Applied Surface Science, 351, 89–94.
and spent liquor of neutral sulfite semichemical pulping process. Bioresource Razali, M. A. A., Ahmad, Z., Ahmad, M. S. B., & Ariffin, A. (2011). Treatment of pulp and
Technology, 218, 518–525. paper mill wastewater with various molecular weight of polyDADMAC induced
W.E. Federation, & American Public Health Association (2005). Standard methods for the flocculation. Chemical Engineering Journal, 166(2), 529–535.
examination of water and wastewater. Washington, DC, USA: American Public Health Sher, F., Malik, A., & Liu, H. (2013). Industrial polymer effluent treatment by chemical
Association (APHA). coagulation and flocculation. Journal of Environmental Chemical Engineering, 1(4),
Fu, S., & Lucia, L. A. (2003). Investigation of the chemical basis for inefficient lignin 684–689.
removal in softwood kraft pulp during oxygen delignification. Industrial & Engineering

365
X. Chen et al. Carbohydrate Polymers 186 (2018) 358–366

Suopajärvi, T., Liimatainen, H., Hormi, O., & Niinimäki, J. (2013). dosage on the mechanism of colloidal substances retention onto pulp. BioResources,
Coagulation–flocculation treatment of municipal wastewater based on anionized 9(2), 3225–3235.
nanocelluloses. Chemical Engineering Journal, 231, 59–67. Wang, S., Hou, Q., Kong, F., & Fatehi, P. (2015). Production of cationic xylan-METAC
Teh, C. Y., Budiman, P. M., Shak, K. P. Y., & Wu, T. Y. (2016). Recent advancement of copolymer as a flocculant for textile industry. Carbohydrate Polymers, 124, 229–236.
coagulation-flocculation and its application in wastewater treatment. Industrial & Wang, Z., Wang, X., Fu, Y., Li, Z., Zhang, F., & Qin, M. (2015). Colloidal behaviors of
Engineering Chemistry Research, 55(16), 4363–4389. lignin contaminants: Destabilization and elimination for oligosaccharides separation
Tunc, M. S., Lawoko, M., & van Heiningen, A. R. (2010). Understanding the limitations of from wood hydrolysate. Separation and Purification Technology, 145, 1–7.
removal of hemicelluloses during autohydrolysis of a mixture of southern hardwoods. Wang, S., Konduri, M. K., Hou, Q., & Fatehi, P. (2016). Cationic xylan-METAC copolymer
BioResources, 5(1), 356–371. as a flocculant for clay suspensions. RSC Advances, 6(46), 40258–40269.
Wang, J. P., Chen, Y. Z., Ge, X. W., & Yu, H. Q. (2007). Gamma radiation-induced grafting Yang, Z., Shang, Y., Huang, X., Chen, Y., Lu, Y., Chen, A., et al. (2012). Cationic content
of a cationic monomer onto chitosan as a flocculant. Chemosphere, 66(9), 1752–1757. effects of biodegradable amphoteric chitosan-based flocculants on the flocculation
Wang, J. P., Chen, Y. Z., Yuan, S. J., Sheng, G. P., & Yu, H. Q. (2009). Synthesis and properties. Journal of Environmental Sciences, 24(8), 1378–1385.
characterization of a novel cationic chitosan-based flocculant with a high water-so- Yao, M., Nan, J., & Chen, T. (2014). Effect of particle size distribution on turbidity under
lubility for pulp mill wastewater treatment. Water Research, 43(20), 5267–5275. various water quality levels during flocculation processes. Desalination, 354,
Wang, J. P., Yuan, S. J., Wang, Y., & Yu, H. Q. (2013). Synthesis, characterization and 116–124.
application of a novel starch-based flocculant with high flocculation and dewatering Zhou, Y., & Franks, G. V. (2006). Flocculation mechanism induced by cationic polymers
properties. Water Research, 47(8), 2643–2648. investigated by light scattering. Langmuir, 22(16), 6775–6786.
Wang, L., Zhang, Y., Chen, H., Xia, X., Liu, Z., & Hu, Z. (2014). Effect of fixing agent

366