- Email: [email protected]
Lignin cationization with glycidyltrimethylammonium chloride aiming at water purification applications
Industrial Crops & Products 104 (2017) 188–194
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Lignin cationization with glycidyltrimethylammonium chloride aiming at water purification applications
MARK
⁎
Ronny Wahlström , Anna Kalliola, Juha Heikkinen, Hanna Kyllönen, Tarja Tamminen VTT Technical Research Centre of Finland Ltd, Finland
A R T I C L E I N F O
A B S T R A C T
Keywords: Lignin Glycidyltrimethylammonium chloride Cationization Water purification Sulfate adsorption
Hardwood organosolv lignin (OSL) and enzymatic hydrolysis lignin (EHL) from softwood were cationized by glycidyltrimethylammonium chloride (GTAC) as new lignin starting materials for the reaction. The products were in detail characterized by 31P-NMR to elucidate the reactivity of different lignin functionalities in cationization with GTAC. For OSL, a high cationization level (degree of substitution 0.74) was reached leading to a water-soluble product. For EHL, low solubility and lower reactivity were observed, likely due to the high saccharide content. Further, lower reactivity of the guaiacyl type lignin present in softwood, compared to syringyl type lignin in hardwoods, which was shown to react efficiently during the derivatization of OSL, probably played a role. In parallel to cationization, an increase of carboxylic acids (from 0.03 up to 0.55 mmol/ g) in the lignin was observed as an unexpected side reaction, possibly as a result of alkaline oxygen oxidation. The applicability of the cationized lignins was tested for water purification applications in three model systems which have only sparingly been studied with cationized lignins. GTAC-OSL and GTAC-EHL were found promising for sulfate removal, with sulfate sorption capacities of up to 54 mg/g. GTAC-OSL was also found suitable for promoting kaolin settling and to some extent for humic acid coagulation.
1. Introduction Lignin is, after cellulose, the second most abundant biopolymer in the World and constitutes approximately 25% of the terrestrial plant biomass (Rastogi and Dwivedi, 2008). The pulping industry produces significant amounts of sulfite and kraft lignin, of which only a small part is sold as technical lignin whereas the majority is incinerated for energy. Future biorefineries will bring new lignin products, such as organosolv (OSL) and enzymatic hydrolysis lignins (EHL) on the market. Especially EHL, a by-product of the emerging cellulosic bioethanol industry, is expected to be produced in enormous quantities. OSLs are characterized by high purity, solubility in alkaline solutions and thus high reactivity and transformability (Rösch and Mülhaupt, 1994). EHL is fundamentally different from OSL, as it often contains high amounts of bound residual polysaccharides (mostly insoluble cellulose) after the enzymatic hydrolysis (Várnai et al., 2010), which decreases its purity, solubility and general reactivity. Apart from incineration, few industrial applications have so far been developed for EHL. Chemical lignin charge modifications by cationization (Kong et al., 2015; Matsushita and Yasuda, 2003), anionization (Cerrutti et al., 2012; Kalliola et al., 2015) or amination (Du et al., 2014; Matsushita
⁎
and Yasuda, 2003) have been reported in the literature. Many lignin derivatizations are nucleophilic substitution reactions performed in alkaline solution, in which the lignin is both soluble and activated as the phenolic groups become nucleophilic phenolates (Kong et al., 2015). Lignin derivatization is of particular interest as a means of tuning lignin solubility. Accordingly, introduction of charged groups onto lignin has been shown to increase its water solubility (Kong et al., 2015). Lignin cationization for water purification applications has been achieved via two alternative routes: amination or direct derivatization of the lignin with a quaternary ammonium group. Amination of lignin with dimethylamine via the Mannich reaction has been reported to produce a highly cationic lignin coagulant (Fang et al., 2010). The charge and coagulation efficiency of the aminated lignin are directly dependent on pH, as low pH leads to protonation of the amine groups. Cationization with e.g. glycidyltrimethylammonium chloride (GTAC) introduces quaternary ammonium groups to the lignin, which are charged irrespective of the pH. These have been reported to be useful as anion exchange resins (Matsushita and Yasuda, 2003) or flocculants for dye removal in waste waters (Kong et al., 2015). Coagulation/flocculation followed by solid/liquid separation, such as sedimentation, filtration or flotation, is one of the most common ways to remove suspended or dissolved solids, colloids, and organic
Corresponding author at: VTT Technical Research Centre of Finland Ltd, Tietotie 2, 02150 Espoo, Finland. E-mail address: ronny.wahlstrom@vtt.fi (R. Wahlström).
http://dx.doi.org/10.1016/j.indcrop.2017.04.026 Received 29 November 2016; Received in revised form 28 February 2017; Accepted 17 April 2017 0926-6690/ © 2017 Elsevier B.V. All rights reserved.
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
The charge of untreated and cationized lignins was determined by polyelectrolyte titration (SCAN-W 12:04) combined with streaming current detection (Mütek™ PCD-04 and titrator T3, BTG Instruments GmbH, Herrsching, Germany). A lignin solution of about 1 wt-% was prepared using 0.01 M sodium citrate buffer (pH 6). The solution was titrated using cationic and anionic polymers, diallyldimethylammonium chloride (polyDADMAC) or polyethene sodiumsulfonate (PESNa) as titrants. Lignin charge (meq/g) was computed based on the amount of polymer consumed at the equivalence point of the titration i.e. at the point where the measured potential passes zero. All the lignin solutions were titrated at pH 7.4–8.6.
matter present in water (Lee et al., 2014; Yang et al., 2016). Coagulants neutralize repulsive charges between particles and colloids and allow bigger aggregates to be settled. The most widely used flocculants are synthetic water soluble polymers based on polyacrylamide and its derivatives, but the use of synthetic flocculants suffers from relatively high production costs and environmental contamination caused by residual unreacted monomers, which typically are toxic and nonbiodegradable (Ahmed et al., 2016; Lee et al., 2014; Yang et al., 2016). Thus, there are activities to develop natural coagulants and flocculants based on renewable biopolymers with widespread availability, low price, non-toxicity and biodegradability. Cationic GTAC derivatives of chitin, starch, nanocrystalline cellulose and hemicelluloses have been studied as flocculants, as reviewed by Kong et al. (2015). Good results have been obtained in dye removal from aqueous solutions using quaternary ammonium salts of lignin in a number of studies (Fang et al., 2003; Kong et al., 2015; Yang et al., 2008). The flocculation efficiency of quaternary ammonium salts of lignin has further been increased by crosslinking it to water-soluble sodium alginate (Zhang et al., 2013). So far most studies on the use of cationized lignin in water purification applications have been concentrated on removing anionic dyes, the removal of which is easy to follow by common UV–vis spectrometry, but few other water purification applications have been reported with these materials. In this article we describe the synthesis of GTAC-derivatized OSL and EHL and chemical characterization of both the starting materials and products. The aim of the study was to produce GTAC-lignins with different solubility and cationization degree and pinpoint how the different solubility and reactivity characteristics of OSL and EHL could be used for optimizing GTAC-lignin performance in different water purification applications. The cationized lignins were screened for applicability in three different model water purification applications, i.e. sulfate removal, humic acid removal, and kaolin settling.
2.3. Synthesis of lignin glycidyltrimethylammonium chloride (GTAC) derivatives OSL or EHL was mixed in 0.5 M NaOH solution at 10 wt-% consistency until dissolved (OSL) or finely dispersed (EHL) and glycidyltrimethylammonium chloride (GTAC) was added to the mixture. The GTAC dosage was calculated as mol-% of the analyzed total amount of phenolic groups. In the case of EHL the phenolic groups could not be analyzed by 31P-NMR due to solubility issues and EHL was thus for GTAC dosage calculations assumed to contain as many phenolic groups as analyzed in OSL per weight unit. The synthesis was performed at 60 °C under gentle magnetic agitation for 20 h. The reaction completed, the mixture was cooled to room temperature and then poured into cellulose ester dialysis membranes (Spectra/Por™ cellulose ester dialysis membranes, Spectrum Laboratories, CA, USA) with a molecular weight cutoff of 500–1000 g/mol. The reaction mixture was dialyzed against ion-exchanged water until the conductivity and pH value of the permeate did not change after exchanging washing water. Finally, the lignin solution was freeze-dried to yield the product as a dark-brown fleecy solid. 2.4. Water purification testing
2. Materials and methods The synthesized GTAC-lignins were tested for sulfate removal and kaolin settling, and as a coagulant for humic acid. Kaolin was chosen as a model inorganic suspended solid and humic acid as a natural organic matter present in natural waters and wastewaters.
2.1. Materials Hardwood organosolv lignin (OSL) from an ethanol-based organosolv process was obtained from Fraunhofer Center for ChemicalBiotechnological Processes CBP (Leuna, Germany) and spruce enzymatic hydrolysis lignin (EHL) after steam explosion pretreatment from SP Technical Research Institute of Sweden (Skellefteå, Sweden) as kind gifts. The EHL was freeze-dried and milled with a Thomas Scientific 2279-K38 Variable Speed Digital ED-5 Wiley mill equipped with a 1 mm sieve.
2.4.1. Sulfate removal The untreated lignins (OSL, EHL) and selected cationized lignins (GTAC-OSL produced with 100 and 200 mol-%, and GTAC-EHL produced with estimated 200 mol-% GTAC dosage) were tested for their capability to remove SO42− anions in aqueous solution at pH 3, 6, and 9 at room temperature. The experimental solutions of sulfate (100 mg/L) were prepared by dissolving analytical grade Na2SO4 in ion-exchanged water. Solution pH was adjusted using 0.1 M HCl (pH 3) or 0.1 M NaOH (pH 9). The solution was stirred and GTAC-lignin was added at a dosage of 100 mg/10 mg SO42−. Sampling was done at 10, 30, and 60 min after the lignin addition. For sulfate analysis, the sample solution was filtered through a 0.2 μm membrane (Whatman FP 30/0,2 CA-S, GE Healthcare, Little Chalfont, UK) to separate the lignin material, i.e. the adsorbent. OSL, EHL, and cationized EHL could be separated by the membrane. Cationized OSLs (200 mol-% GTAC and partly 100 mol-% GTAC depending on pH) were soluble and passed the membrane, which was seen as brown color in the filtrate. In a parallel protocol, a 60 min sample was withdrawn and frozen to −18 °C. Freezing caused aggregation and sedimentation of the lignin material, which was separated by decanting from the supernatant after melting the solution. All the solutions were measured for residual sulfate anions by using a P/ACE MDQ capillary electrophoresis (CE) instrument equipped with a photodiode array UV–vis detector working via indirect detection at 230 nm (Beckman-Coulter Inc., Brea, CA, USA) and quantified based on sulfate standards. Electrolyte solutions were commercial from Analis (Ceofix Anions 5, Ordior/Analis, part number: 10-0004650). A bare
2.2. Chemical characterization Elemental analysis (carbon, hydrogen, oxygen, nitrogen and sulfur measured) was done with a Thermo Scientific Flash 2000 Organic Elemental Analyser (Rockford, IL, USA). Composition analysis was performed according to the NREL procedure (Sluiter et al., 2008) with the carbohydrates analyzed from the acid hydrolysate by HPLC as described by Tenkanen and Siika-aho (2000). Zeta potential was measured with a Zetasizer nano ZS (Malvern Instruments, Malvern, UK) in either milli-Q water or 0.1 M NaOH solution. The instrument measures the electrophoretic mobility of particles, which is converted into zeta potential values using the Smoluchowski model. The amount (mmol/g) of aliphatic –OH, carboxylic –OH and phenolic –OH (para-OH phenyl, condensed and syringyl –OH, guaiacyl –OH and catechol –OH groups) was quantified by 31P-NMR after phosphitylation as previously described (Granata and Argyropoulos, 1995). Conductivity measurements were performed with a Hach (Loveland, CO, USA) HQ14d conductivity meter. 189
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
fused silica capillary with inner diameter 75 μm, length to the UV detector 50 cm and total length 60 cm was used. The separation voltage was −30 kV with positive to negative polarity. Samples were injected using 0.5 psi pressure for 8 s. 2.4.2. Kaolin settling A kaolin suspension of 100 mg/L concentration (turbidity ∼100 nephelometric turbidity units, NTUs) was prepared to represent the turbidity level of surface waters. A kaolin product, Capim DG, of small particle size, D50 = 0.56 μm and 92% below 2-μm equivalent diameter (Imerys, Paris, France) was used in the study. Charge density measurements were done by jar tests after the addition of GTAC-OSL (200 mol% GTAC) at doses of 1, 5, 10, 20 and 30 mg/L to the kaolin suspension for the pre-screening of the required GTAC-OSL dosage. The jar tests for coagulation studies were performed using a Kemira Flocculator 2000 device (Kemira OYj, Helsinki, Finland), applying 10 s fast mixing, 10 min slow mixing, and settling overnight, after which a sample was taken from the supernatant for analysis. GTAC-OSL dosages of 1, 2, 3, 4, and 5 mg/L were studied. Turbidity expressed in NTUs was measured using a HACH 2100AN Turbidimeter (Loveland, CO, USA). pH and conductivity were measured using standard hand held meters, VWR pH 100 and VWR EC 300 respectively (Radnor, PA, USA). Charge density was measured by polyelectrolyte titration method using Mütek PCD 03 pH Particle Charge Detector and PCD-Titrator 2 (BTG Instruments GmbH, Herrsching, Germany). The metal salt ferric sulfate PIX-105 from Kemira OYj (Helsinki, Finland) was used as commercial coagulant reference. PIX-105 contained 11.29% of ferric iron (Fe3+), and the density was 1500 g/L. 2.4.3. Humic acid coagulation Humic acid test water was made with technical grade humic acid (Sigma-Aldrich, Inc., St. Louis, MO, USA) by initially preparing a stock solution of 1000 mg/L in 1 M NaOH. The test water representing a typical humic acid concentration in natural surface waters (TOC average in Finnish lakes 7.6 mg/L (Henriksen et al., 1998)) was prepared for this study with a total organic carbon (TOC) amount of 8 mg/L at pH 7. TOC was measured using a TOC-VCPH Total Organic Carbon Analyzer (Shimadzu, Kyoto, Japan). Similar jar tests of coagulation of humic acid as for kaolin suspension were performed.
Scheme 1. Reaction scheme for lignin derivatization with glycidyltrimethylammonium chloride (GTAC) under alkaline conditions and side reaction of GTAC with hydroxide anions.
the solution pH. The reaction is performed in alkaline solution, in which the base, typically NaOH, has the dual role of assisting in lignin dissolution and of activating the phenolic –OH groups by deprotonating them (Kong et al., 2015). The chemical selectivity of the nucleophilic attack on the epoxide reagent should clearly favor the phenolate anion, whereas the aliphatic –OH groups of the lignin, and in the case of the EHL, of the attached residual saccharides, are less nucleophilic. The epoxide is also consumed in a side reaction with the OH− ions, and the reaction is therefore sensitive to the applied conditions and especially reagent concentrations (Kong et al., 2015). In theory, the nucleophilic attack of the phenolate may take place either directly through the deprotonated oxygen of the phenolate, or through the conjugated, unsubstituted C3 and C5 positions. The possibility for a chain reaction, in which the GTAC would react with the newly formed aliphatic –OH groups of the GTAC substituents in the lignin cannot be excluded, even if aliphatic OH-groups are not expected to be sufficiently nucleophilic under the used conditions. In case such reactions take place, the charge of the lignin would be further increased, which could be beneficial if the target is to introduce maximum charge. The cationization of OSL was done with GTAC dosages of 25, 50, 75, 100 and 200 mol-% relative to the analyzed total amount of phenolic groups. EHL was reacted with a GTAC dosage corresponding to 200 mol-% for OSL on sample weight basis. In preliminary experiments, the removal of unreacted or degraded GTAC and NaOH by dialysis was monitored by elemental analysis of freeze-dried GTAC- lignin, whereby the nitrogen content was confirmed to stay constant when the permeate’s conductivity and pH did not increase after changing dialysis water, indicating complete removal of the reagents. The used dialysis membrane with 500–1000 Da molecular weight cutoff was suitable for easily passing through the low molecular weight reagents but no leakage of lignin was noticed based on following the color of the
3. Results and discussion 3.1. Lignin characterization and cationization with GTAC In this work, the cationization of two different lignin types was studied. The organosolv lignin (OSL) was soluble in NaOH solutions whereas the enzymatic hydrolysis lignin (EHL) was insoluble in a wide range of NaOH concentrations (up to 5 M NaOH tested). This difference was explained by the composition analysis, which showed the OSL to be practically pure lignin whereas EHL contained high amounts of carbohydrates, likely cellulose as high glucose amounts (45.1%) were found (Supporting information SI 1). The cellulose residues in EHL inhibited lignin solubility. 31P-NMR was performed to evaluate the amount of reactive phenolic groups in OSL, which was calculated as the sum of condensed and syringyl (i.e. all phenolic units with both the 3 and 5 positions substituted, 2.17 mmol/g), guaiacyl (0.63 mmol/g) and catechol (0.08 mmol/g) groups. OSL thus contained a total 2.88 mmol/ g of phenolic –OH groups. Similar analysis of EHL was not possible due to the sample’s insolubility. For calculating GTAC dosage in lignin derivatization, EHL was assumed to contain the same amount of phenolic groups per weight unit as OSL. Reacting lignin with trimethylammonium-carrying epoxide reagents is an easy and straight-forward way to obtain cationized lignin (Scheme 1). The lignin becomes permanently charged by the attachment of trimethylammonium groups, contrary to the case with amine or carboxyl groups, which are either charged or neutral depending on 190
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
levels as for OSL. The nitrogen content of EHL was only 1.6 wt-% with the 200 mol-% GTAC dosage that produced a nitrogen content of 2.6 wt-% (DS 0.74) for OSL. Because the EHL contained a high proportion of carbohydrates the DS of the lignin fraction in EHL could be fairly high based on the nitrogen content, but it is unknown to which extent the carbohydrates had been cationized during the reaction which complicates any DS calculations for EHL. The cationization under these conditions did not change the solubility of EHL markedly, as had been the original objective. Possibly, employing higher GTAC dosages or reaction conditions in which the aliphatic –OH groups would be more nucleophilic towards reaction with GTAC, could yield cationized EHL with higher DS and improved solubility characteristics. The charge of the lignin changed drastically as a result of the cationization (Table 1). The untreated OSL was practically chargeneutral, but performing the cationization reaction at low GTAC dosages introduced a negative charge on the material (−0.35 meq/g), which was in clear contradiction with original expectations. The negative charge decreased with increasing degree of cationization, and for the GTAC-OSL produced with 200 mol-% reagent dosage, the charge was positive with a value of 0.51 meq/g. Analysis of the 31P-NMR results explained the observed initial introduction of negative charge to the lignin: the amount of carboxylic acid groups in the lignin drastically increased by the applied treatments to a level of 0.40–0.55 mmol/g (Fig. 1). The level of introduced carboxylic acids was constant in all cationized OSL samples and did not further grow when increasing the GTAC dosage. The introduction of carboxylic acid groups to lignin under the mild reaction conditions was surprising and though interesting, the mechanism of the carboxylic acid formation was not further studied in this work. Likely, some chemical functionality in the OSL lignin structure was easily transformed into carboxylic acid independently of the GTAC dosage. In a reference treatment of OSL under alkaline cationization conditions but in absence of GTAC, carboxylic acid groups were similarly introduced, confirming that it was indeed the alkaline treatment that led to carboxyl formation and that the GTAC reagent did not play any notable role in this reaction. Possibly some of the syringyl or condensed groups were easily converted into carboxylic acid, but this hypothesis could not be pursued with the applied analytical methods because the syringyl and condensed signals are overlapping in the 31P-NMR spectra. Alkaline conditions in the presence of oxygen are known to induce anionic charge to lignin (Kalliola et al., 2015). Typically more harsh conditions are needed, but the structures easiest to oxidize appear to react already under the mild conditions applied during the cationization. The presence of both negative and positive charge in the cationized lignin might well influence its solution behavior, as well as performance in water purification applications, and thus should be further studied. With a growing number of ammonium groups attached to the lignin its overall charge grew to be positive. For
Table 1 Nitrogen content (wt-% of dry sample), amount of phenolic –OH groups (mmol/g of dry cationized lignin), degree of substitution (DS) of cationized lignin calculated based on the analyzed change of phenolic –OH groups, charge of cationized lignin (meq/g, milliequivalents per gram) as determined by polyelectrolyte titration (under pH conditions 7.4–8.6) and the assayed solubility of cationized organosolv (OSL) or enzymatic hydrolysis lignin (EHL) lignin samples. NA = Not Applicable. Material
GTAC dose (mol-%)
N (wt%)
Phenolic −OH (mmol/ g)
DS
Charge (meq/g)
OSL
0 25 50 75 100 200
0.2 0.8 1.4 1.8 2.1 2.6
2.88 2.24 1.83 1.58 1.31 0.75
0.00 0.22 0.36 0.45 0.55 0.74
−0.010 −0.354 −0.031 −0.029 −0.007 0.513
EHL
0 (200)
0.2 1.6
NA NA
NA NA
NA NA
Solubility
NaOH (aq) NaOH (aq) NaOH (aq) NaOH (aq) NaOH (aq) NaOH (aq) and water Insoluble Insoluble
solution. Cationized EHL could easily be washed as it was insoluble in water and alkaline solution. The reaction conditions were not further optimized in this work, as they have been studied in detail elsewhere (Kong et al., 2015). The amount of elemental nitrogen in the washed GTAC-lignin samples clearly responded to the GTAC dosage (Table 1). The amount of free phenolic groups as analyzed by 31P-NMR correlated to this result by decreasing in relation to GTAC dosage, suggesting that the primary regioselectivity of the cationization was on the deprotonated phenolate oxygen groups (C4 position). The degree of substitution (DS) was approximated as the ratio of substituted phenolic –OH groups (calculated as the difference between the original amount of phenolic groups in OSL and that analyzed in the derivatized samples) divided by the original number of –OH groups as given by the 31P-NMR results. This approximation assumed that the whole decrease in the amount of phenolic groups was due to derivatization with GTAC and did not take into account the weight increase caused by the attached GTAC substituents. The nitrogen amount in the samples as well as the estimated DS increased with GTAC dosage and possibly higher DS values could be achieved by further increasing the GTAC dosage. The water solubility was tested for all the derivatized samples and it was noticed that only OSL reacted with 200 mol-% of GTAC had become water soluble. This finding indicated that there was a clear threshold DS for GTAC-lignin water solubility, which in the case of the used OSL was between DS 0.55 and 0.74. The derivatization of EHL with a GTAC dosage corresponding to 200 mol-% for OSL clearly increased the nitrogen content of the EHL suggesting that this lignin was also cationized, but the cationization did not proceed to as high
Fig. 1. Abundance of different –OH and carboxylic acid groups in untreated and GTAC-derivatized lignin as analyzed by
191
31
P-NMR.
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
3.2. Water purification testing 3.2.1. Sulfate removal Sulfate is a common constituent of many natural waters and wastewaters, and industrial wastewaters may contain several thousand mg/L of dissolved sulfate (Lens et al., 2010). Several processes can be applied to promote the removal of sulfate, e.g. chemical precipitation by addition of barium or calcium salts, membrane filtration, reverse osmosis, and biological sulfate removal. In this work, sulfate removal using GTAC-derivatized OSL and EHL as sulfate adsorbents was studied. OSL and EHL and their GTAC-derivatives (OSL 100%, OSL 200%, and EHL 200%) were tested for sulfate removal at pH 3, 6, and 9 for 60 min. Thereafter, the solutions were filtered in order to separate the adsorbent material. However, cationized OSLs were mostly soluble and passed the membrane. Therefore, sulfate adsorption capacities of different lignin materials were compared in those samples where lignin material was separated by filtration followed by a freezing-melting procedure (SI 3). The insolubility of GTAC-EHL could thus be a process benefit in water purification applications as it can be easily filtered off the processed liquids as opposed to completely soluble adsorbents. Cross-linking of soluble GTAC-OSL could be studied as an alternative method to have an easier by filtration separable adsorbent. Sulfate adsorption by GTAC-OSL was more efficient than that of untreated OSL in acidic (pH 3) and near neutral (pH 6) conditions (Fig. 2). The highest sulfate adsorption capacity was 54 mg/g, at pH 3, in the case of GTAC-OSL (100 mol-%). In general, the GTAC-OSL (100 mol-%) was slightly more efficient for sulfate removal than GTAC-OSL (200 mol-%), which suggests that the cationicity of the GTAC-lignin might not be the only parameter to determine sulfate removal efficiency. EHL and GTAC-EHL could be separated from the sample solution by membrane filtration due to the insolubility of the samples. For EHL samples, the differences in sulfate concentration between the filtrated and decanted solutions were small (SI 3), thus validating the freezing-melting and decanting procedures. GTAC-EHL adsorbed sulfate in higher amounts than EHL at pH 3 (44 vs. 21 mg/g), but at pH 6 or 9 there was no difference in the adsorption capacity of untreated and cationized EHL, suggesting a strong pH dependency for the sulfate removal capacity of GTAC-EHL (Fig. 2). Ion exchange is likely the major mechanism governing the uptake of sulfate. Under acidic conditions the carboxylic groups, which in this work were found to be introduced to the lignin under cationization conditions, are protonated, which leads to the maximum positive charge in the adsorbent. Under pH 9 conditions, the carboxylic acids are deprotonated causing a lower positive charge in the adsorbent. This phenomenon is likely the reason for the highest sulfate removal capacity of the cationized OSL and EHL being under acidic conditions. In addition to direct sulfate-GTAC-lignin interactions, the charged state of the lignin particles can also be expected to govern the lignin particles behavior in solution which has an indirect effect on the sulfate adsorption capacity. It should also be noted that the charge introduced by GTAC derivatization is permanent and not pH dependent, like for the carboxyl groups. In a previous study, in which coconut coir carbonized with ZnCl2 was studied for sulfate removal in the pH range 2–11, it was found that the optimum pH for sulfate removal was achieved at a linear level between pH 3 and 9, but sulfate removal was lower at < pH 3 or > pH 9, due to adsorption competition between sulfate and the acid or base counter ions (Cl− or OH−) (Namasivayam and Sangeetha, 2008). Adsorption competition is not likely to have played a major role in this study, as the highest sulfate adsorption levels should then have been observed at pH 6, where the counter ion concentration is the lowest. The achieved levels of sulfate adsorption varies by several orders of magnitude (e.g. from 0.06 mg/g of coconut coir pit carbon to 156 mg/g for shrimp shells) for different adsorbent materials as compared by Namasivayam and Sangeetha (2008). Variations between different used experimental conditions may, however, make direct comparisons between literature results difficult.
Fig. 2. Sulfate adsorption capacity of untreated and GTAC-derivatized OSL and EHL with 60 min adsorption time.
EHL, reliable charge titration results were not obtained due to sample insolubility. The 31P-NMR characterization of cationized OSL indicated that the amount of aliphatic –OH groups had increased with increasing DS, which means that the –OH groups formed at the epoxide opening of GTAC are not particularly reactive for further chain reaction with other GTAC molecules (Fig. 1). The amount of condensed and syringyl –OH groups decreased quite linearly as the GTAC dosage was increased, but the guaiacyl –OH groups were less reactive and only decreased slightly with growing GTAC dosage. This result rules out the alternative reaction mechanism according to which the derivatization could take place on the 3 or 5 positions through conjugation of the negative phenolate charge, as this reaction is not possible for the 3,5-disubstituted syringyl units and the guaiacyls with one vacant 3 or 5 position showed only low reactivity. Also, if the guaiacyl units were reactive through their 3 or 5 position, this would be seen as an increase in the amount of condensed groups in 31P-NMR. In this work, the linearly decreasing trend of phenolic –OH groups was almost completely explained by the substitution of the syringyl and condensed –OH groups. This suggests that hardwood lignin, rich in syringyl units, is more amenable to cationization than softwood lignin containing a higher proportion of guaiacyl units. Results by Matsushita and Yasuda (2003) suggest that para-hydroxyl –OH groups would have a slightly higher reactivity than guaiacyl –OH groups towards GTAC when using lignin monomer model compounds, and the results in this work further establish that the condensed and syringyl –OH groups were more reactive than guaiacyl-OH groups towards GTAC, using polymeric lignin samples. The DS of GTAC-OSL (200 mol-% GTAC) was calculated to be 0.74 or in terms of cationicity 2.13 mmol/g (approximation in which the amount of available phenolic –OH groups in lignin has not been corrected with the mass growth caused by the grafted GTAC units), which corresponds to the cationicity previously reported for lignin cationized via the Mannich reaction (cationicity in range of 1.8–2.6 mmol/g) (Fang et al., 2010). The zeta potential was measured for the cationized OSL samples in 0.1 M NaOH solution, and in addition the water-soluble GTAC-lignin was analyzed in water solution (SI 2). In NaOH solution, the zeta potential was clearly negative for the untreated OSL, but the increasing degree of cationization led to rising zeta potential values and for the most cationic lignin sample with DS 0.74, the zeta potential was practically neutral in 0.1 M NaOH, but positive (+24 mV) in water. Du et al. (2014) measured zeta potential values of 27–32 mV in dilute hydrochloric acid solutions for lignin cationized by amination via the Mannich reaction.
192
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
important functional groups and binding sites (Greathouse et al., 2014; Kalinichev et al., 2011). In this work, the coagulation of commercial humic acid was studied as a model system for coagulating organic constituents from natural waters. When GTAC-OSL was introduced into the humic acid solution (8 mg/L, corresponding to the concentration of humic acid in natural surface waters), precipitates were formed with GTAC-OSL doses of 50 and 100 mg/L. They contained about half of the total organic matter (TOC) introduced in the test, i.e. the sum of TOC from humic acid and GTAC-OSL (Fig. 4). Possibly, the –OH and carboxyl groups in the GTAC-OSL could form hydrogen bonds with the humic acid and consequently enhance its removal efficiency as proposed in the literature (Zaki and El-Gendy, 2014). Although the formation of carboxylic acid groups under GTAC derivatization conditions to lignin caused its overall charge to decrease significantly especially at lower GTAC dosages, the introduction of the carboxylic groups might have improved the coagulation of humic acid. The commercial ferric sulfate coagulant PIX-105 efficiently reduced TOC at doses ≥ 45 mg/L (corresponding to 5 mg/L Fe3+). The TOC decrease with GTAC-OSL was not as large as with PIX-105, but taking into account that the GTAC-OSL introduced a large amount of organic carbon itself, part of the result may be due to the TOC introduced with the GTAC-OSL. The TOC of humic acid-free solution increased linearly with the GTAC-OSL (200 mol-%) dose due to the coagulants organic nature. To study the coagulation of humic acid with GTAC-OSL in detail, other methods than TOC analysis would be beneficial to use that can discriminate between humic acid and GTAC-lignin.
Fig. 3. Settling of kaolin particles using GTAC-derivatized organosolv lignin (OSL, 200 mol-% GTAC) as coagulant. Left vertical axis displays the turbidity of the kaolin dispersion (in nephelometric turbidity units, NTUs), right vertical axis displays the amount of total organic carbon (TOC) and the dashed line displays the TOC introduced by the addition of coagulant.
4. Conclusions Organosolv (OSL) and enzymatic hydrolysis lignin (EHL) were successfully cationized with (GTAC). Cationized lignins with different solubility characteristics were obtained, GTAC-OSL being water soluble and GTAC-EHL insoluble. GTAC-OSL adsorbed sulfate efficiently, promoted settling of kaolin and precipitated humic acid from solution. GTAC-EHL was efficient for sulfate removal under acidic conditions and was easy to remove in down-stream processing due to its insolubility. Carboxylic acids were introduced to the OSL during the mild conditions of the cationization reaction as a side reaction, which may explain part of the pH-dependency of sulfate removal with the GTAC-lignins.
Fig. 4. Measured amounts of total organic carbon (TOC, mg/L) in humic acid solutions (8 mg/mL) treated with GTAC-derivatized organosolv lignin (200 mol-% GTAC in derivatization) or with commercial PIX-105 coagulant.
3.2.2. Kaolin settling The GTAC-lignins were tested as kaolin settling agents. Particles of kaolin, a hydrous aluminum silicate, carry a negative charge under normal conditions in aqueous suspension (Tombácz and Szekeres, 2006). Electrostatic repulsion between particles can be reduced, e.g. by double-layer compression or ion shielding of the surface charges using electrolytes or polymers such as polyacrylamide, causing the particles to adhere to one another forming aggregates and flocs (Kim and Palomino, 2009; Nasser and James, 2009). Charge density determination revealed that the charge of the particles in kaolin/lignin suspensions remained negative (−3.4 μeq/ L) when 1 mg/L GTAC-OSL (200 mol-%) was added to 100 mg/L kaolin in water suspension, but changed to positive (+2.6 μeq/L) when 5 mg/ L of this product was added. Hence the dosage range from 1 to 5 mg/L was selected for jar tests. A low dose of 2 mg/L of GTAC-OSL (200 mol%), with a narrow optimum performance window from 1.5 to 2.5 mg/L, improved the settling of small kaolin particles (Fig. 3). Cationic GTACOSL (200 mol-%) appeared to neutralize the anionic charge density of kaolin. This made the floc formation possible and increased the settling of microscale particles under the mild settling forces. When the coagulant was overdosed, the charge exceeded the isoelectric point, and the repulsion accordingly increased, leading to increased turbidity. In previous work it has been shown that coagulation efficiency in waste water purification is highly dependent on the dose with clearly defined optima, which are specific for the type of impurity and coagulant used (Kong et al., 2015).
Acknowledgements Tarja Wikström and Vuokko Liukkonen are acknowledged for their skillful technical assistance, Atte Mikkelson for consulting the residual sulfate determination, and Pia Willberg-Keyriläinen for performing the 31 P- NMR analysis. Thanks are due to SP Technical Research Institute of Sweden and Fraunhofer Center for Chemical-Biotechnological Processes CBP for kindly providing the lignin samples. This work has been carried out within the AERTOs Bioeconomy Project 2014–2016 performed by 9 European RTOs. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.indcrop.2017.04.026. References Ahmed, T., Bhatti, Z.A., Maqbool, F., Mahmood, Q., Faridullah, S. Qayyum, Mushtaq, N., 2016. A comparative study of synthetic and natural coagulants for silver nanoparticles removal from wastewater. Desalin. Water Treat. 57, 18718–18723. Cerrutti, B.M., de Souza, C.S., Castellan, A., Ruggiero, R., Frollini, E., 2012. Carboxymethyl lignin as stabilizing agent in aqueous ceramic suspensions. Ind. Crops Prod. 36, 108–115. Du, X., Li, J., Lindström, M.E., 2014. Modification of industrial softwood kraft lignin using Mannich reaction with and without phenolation pretreatment. Ind. Crops Prod. 52, 729–735. Fang, G.Z., He, W.H., Song, Z.Q., 2003. Synthesis and performance of quaternary
3.2.3. Humic acid coagulation In humic acid, carboxyl, hydroxyl and phenolic groups are the 193
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al. ammonium salt of lignin as cationic flocculant. Linchan Huaxue Yu Gongye/Chem. Ind. For. Prod. 23, 37–41. Fang, R., Cheng, X., Xu, X., 2010. Synthesis of lignin-base cationic flocculant and its application in removing anionic azo-dyes from simulated wastewater. Bioresour. Technol. 101, 7323–7329. Granata, A., Argyropoulos, D.S., 1995. 2-Chloro-4,4,5,5-tetramethyl-1,3,2dioxaphospholane, a reagent for the accurate determination of the uncondensed and condensed phenolic moieties in lignins. J. Agric. Food Chem. 43, 1538–1544. Greathouse, J., Johnson, K., Greenwell, H., 2014. Interaction of natural organic matter with layered minerals: recent developments in computational methods at the nanoscale. Minerals 4, 519–540. Henriksen, A., Skjelvåle, B.L., Mannio, J., Wilander, A., Harriman, R., Curtus, C., Jensen, J.P., Fjeld, E., Moiseenko, T., 1998. Northern European Lake Survey, 1995: Finland, Norway, Sweden, Denmark, Russian Kola, Russian Karelia, Scotland and Wales. Ambio 27, 80–91. Kalinichev, A.G., Iskrenova-Tchoukova, E., Ahn, W.Y., Clark, M.M., Kirkpatrick, R.J., 2011. Effects of Ca2+ on supramolecular aggregation of natural organic matter in aqueous solutions: a comparison of molecular modeling approaches. Geoderma 169, 27–32. Kalliola, A., Vehmas, T., Liitiä, T., Tamminen, T., 2015. Alkali-O2 oxidized lignin – a biobased concrete plasticizer. Ind. Crops Prod. 74, 150–157. Kim, S., Palomino, A.M., 2009. Polyacrylamide-treated kaolin: a fabric study. Appl. Clay Sci. 45, 270–279. Kong, F., Parhiala, K., Wang, S., Fatehi, P., 2015. Preparation of cationic softwood kraft lignin and its application in dye removal. Eur. Polym. J. 67, 335–345. Lee, C.S., Robinson, J., Chong, M.F., 2014. A review on application of flocculants in wastewater treatment. Process Saf. Environ. Prot. 92, 489–508. Lens, P.N.L., Visser, A., Janssen, A.J.H., Pol, L.W.H., Lettinga, G., 2010. Biotechnological treatment of sulfate-rich wastewaters. Crit. Rev. Environ. Sci. Technol. 28, 41–88. Matsushita, Y., Yasuda, S., 2003. Preparation of anion-exchange resins from pine sulfuric acid lignin, one of the acid hydrolysis lignins. J. Wood Sci. 49, 423–429. Namasivayam, C., Sangeetha, D., 2008. Application of coconut coir pith for the removal of sulfate and other anions from water. Desalination 219, 1–13.
Nasser, M.S., James, A.E., 2009. The effect of electrolyte concentration and pH on the flocculation and rheological behaviour of kaolinite suspensions. J. Eng. Sci. Technol. 4, 430–446. Rösch, J., Mülhaupt, R., 1994. Mechanical properties of organosolv-lignin-filled thermoplastics. Polym. Bull. 32, 361–365. Rastogi, S., Dwivedi, U.N., 2008. Manipulation of lignin in plants with special reference to O-methyltransferase. Plant Sci. 174, 264–277. SCAN-W 12:04 (Scandinavian Pulp, Paper and Board testing committee method), 2004. Process Waters, Cationic Demand, Polyelectrolyte Titration with a Streaming Current Detector . Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2008. Determination of Structural Carbohydrates and Lignin in Biomass. Laboratory Analytical Procedure (LAP). National Renewable Energy Laboratory, Golden, Colorado. Tenkanen, M., Siika-aho, M., 2000. An alpha-glucuronidase of Schizophyllum commune acting on polymeric xylan. J. Biotechnol. 78, 149–161. Tombácz, E., Szekeres, M., 2006. Surface charge heterogeneity of kaolinite in aqueous suspension in comparison with montmorillonite. Appl. Clay Sci. 34, 105–124. Várnai, A., Siika-aho, M., Viikari, L., 2010. Restriction of the enzymatic hydrolysis of steam-pretreated spruce by lignin and hemicellulose. Enzyme Microb. Technol. 46, 185–193. Yang, A.L., Gao, W., Wei, W.Y., Jiang, W.J., 2008. Synthesis and flocculation performance of a new quaternary ammonium salt of lignin. Zhongguo Zaozhi Xuebao/Trans. China Pulp Pap. 23, 60–63. Yang, R., Li, H., Huang, M., Li, A., Yang, H., 2016. Chitosan-based flocculants and their applications in water treatment. Water Res. 95, 59–89. Zaki, A.A., El-Gendy, N.A., 2014. Removal of metal ions from wastewater using EB irradiation in combination with HA/TiO2/UV treatment. J. Hazard. Mater. 271, 275–282. Zhang, Q., Wang, D., Bei, Y., Ren, S., Fang, G., 2013. Flocculation performance of trimethyl quaternary ammonium salt of lignin-sodium alginate polyampholyte. BioResources 8, 3544–3555.
194
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Lignin cationization with glycidyltrimethylammonium chloride aiming at water purification applications
MARK
⁎
Ronny Wahlström , Anna Kalliola, Juha Heikkinen, Hanna Kyllönen, Tarja Tamminen VTT Technical Research Centre of Finland Ltd, Finland
A R T I C L E I N F O
A B S T R A C T
Keywords: Lignin Glycidyltrimethylammonium chloride Cationization Water purification Sulfate adsorption
Hardwood organosolv lignin (OSL) and enzymatic hydrolysis lignin (EHL) from softwood were cationized by glycidyltrimethylammonium chloride (GTAC) as new lignin starting materials for the reaction. The products were in detail characterized by 31P-NMR to elucidate the reactivity of different lignin functionalities in cationization with GTAC. For OSL, a high cationization level (degree of substitution 0.74) was reached leading to a water-soluble product. For EHL, low solubility and lower reactivity were observed, likely due to the high saccharide content. Further, lower reactivity of the guaiacyl type lignin present in softwood, compared to syringyl type lignin in hardwoods, which was shown to react efficiently during the derivatization of OSL, probably played a role. In parallel to cationization, an increase of carboxylic acids (from 0.03 up to 0.55 mmol/ g) in the lignin was observed as an unexpected side reaction, possibly as a result of alkaline oxygen oxidation. The applicability of the cationized lignins was tested for water purification applications in three model systems which have only sparingly been studied with cationized lignins. GTAC-OSL and GTAC-EHL were found promising for sulfate removal, with sulfate sorption capacities of up to 54 mg/g. GTAC-OSL was also found suitable for promoting kaolin settling and to some extent for humic acid coagulation.
1. Introduction Lignin is, after cellulose, the second most abundant biopolymer in the World and constitutes approximately 25% of the terrestrial plant biomass (Rastogi and Dwivedi, 2008). The pulping industry produces significant amounts of sulfite and kraft lignin, of which only a small part is sold as technical lignin whereas the majority is incinerated for energy. Future biorefineries will bring new lignin products, such as organosolv (OSL) and enzymatic hydrolysis lignins (EHL) on the market. Especially EHL, a by-product of the emerging cellulosic bioethanol industry, is expected to be produced in enormous quantities. OSLs are characterized by high purity, solubility in alkaline solutions and thus high reactivity and transformability (Rösch and Mülhaupt, 1994). EHL is fundamentally different from OSL, as it often contains high amounts of bound residual polysaccharides (mostly insoluble cellulose) after the enzymatic hydrolysis (Várnai et al., 2010), which decreases its purity, solubility and general reactivity. Apart from incineration, few industrial applications have so far been developed for EHL. Chemical lignin charge modifications by cationization (Kong et al., 2015; Matsushita and Yasuda, 2003), anionization (Cerrutti et al., 2012; Kalliola et al., 2015) or amination (Du et al., 2014; Matsushita
⁎
and Yasuda, 2003) have been reported in the literature. Many lignin derivatizations are nucleophilic substitution reactions performed in alkaline solution, in which the lignin is both soluble and activated as the phenolic groups become nucleophilic phenolates (Kong et al., 2015). Lignin derivatization is of particular interest as a means of tuning lignin solubility. Accordingly, introduction of charged groups onto lignin has been shown to increase its water solubility (Kong et al., 2015). Lignin cationization for water purification applications has been achieved via two alternative routes: amination or direct derivatization of the lignin with a quaternary ammonium group. Amination of lignin with dimethylamine via the Mannich reaction has been reported to produce a highly cationic lignin coagulant (Fang et al., 2010). The charge and coagulation efficiency of the aminated lignin are directly dependent on pH, as low pH leads to protonation of the amine groups. Cationization with e.g. glycidyltrimethylammonium chloride (GTAC) introduces quaternary ammonium groups to the lignin, which are charged irrespective of the pH. These have been reported to be useful as anion exchange resins (Matsushita and Yasuda, 2003) or flocculants for dye removal in waste waters (Kong et al., 2015). Coagulation/flocculation followed by solid/liquid separation, such as sedimentation, filtration or flotation, is one of the most common ways to remove suspended or dissolved solids, colloids, and organic
Corresponding author at: VTT Technical Research Centre of Finland Ltd, Tietotie 2, 02150 Espoo, Finland. E-mail address: ronny.wahlstrom@vtt.fi (R. Wahlström).
http://dx.doi.org/10.1016/j.indcrop.2017.04.026 Received 29 November 2016; Received in revised form 28 February 2017; Accepted 17 April 2017 0926-6690/ © 2017 Elsevier B.V. All rights reserved.
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
The charge of untreated and cationized lignins was determined by polyelectrolyte titration (SCAN-W 12:04) combined with streaming current detection (Mütek™ PCD-04 and titrator T3, BTG Instruments GmbH, Herrsching, Germany). A lignin solution of about 1 wt-% was prepared using 0.01 M sodium citrate buffer (pH 6). The solution was titrated using cationic and anionic polymers, diallyldimethylammonium chloride (polyDADMAC) or polyethene sodiumsulfonate (PESNa) as titrants. Lignin charge (meq/g) was computed based on the amount of polymer consumed at the equivalence point of the titration i.e. at the point where the measured potential passes zero. All the lignin solutions were titrated at pH 7.4–8.6.
matter present in water (Lee et al., 2014; Yang et al., 2016). Coagulants neutralize repulsive charges between particles and colloids and allow bigger aggregates to be settled. The most widely used flocculants are synthetic water soluble polymers based on polyacrylamide and its derivatives, but the use of synthetic flocculants suffers from relatively high production costs and environmental contamination caused by residual unreacted monomers, which typically are toxic and nonbiodegradable (Ahmed et al., 2016; Lee et al., 2014; Yang et al., 2016). Thus, there are activities to develop natural coagulants and flocculants based on renewable biopolymers with widespread availability, low price, non-toxicity and biodegradability. Cationic GTAC derivatives of chitin, starch, nanocrystalline cellulose and hemicelluloses have been studied as flocculants, as reviewed by Kong et al. (2015). Good results have been obtained in dye removal from aqueous solutions using quaternary ammonium salts of lignin in a number of studies (Fang et al., 2003; Kong et al., 2015; Yang et al., 2008). The flocculation efficiency of quaternary ammonium salts of lignin has further been increased by crosslinking it to water-soluble sodium alginate (Zhang et al., 2013). So far most studies on the use of cationized lignin in water purification applications have been concentrated on removing anionic dyes, the removal of which is easy to follow by common UV–vis spectrometry, but few other water purification applications have been reported with these materials. In this article we describe the synthesis of GTAC-derivatized OSL and EHL and chemical characterization of both the starting materials and products. The aim of the study was to produce GTAC-lignins with different solubility and cationization degree and pinpoint how the different solubility and reactivity characteristics of OSL and EHL could be used for optimizing GTAC-lignin performance in different water purification applications. The cationized lignins were screened for applicability in three different model water purification applications, i.e. sulfate removal, humic acid removal, and kaolin settling.
2.3. Synthesis of lignin glycidyltrimethylammonium chloride (GTAC) derivatives OSL or EHL was mixed in 0.5 M NaOH solution at 10 wt-% consistency until dissolved (OSL) or finely dispersed (EHL) and glycidyltrimethylammonium chloride (GTAC) was added to the mixture. The GTAC dosage was calculated as mol-% of the analyzed total amount of phenolic groups. In the case of EHL the phenolic groups could not be analyzed by 31P-NMR due to solubility issues and EHL was thus for GTAC dosage calculations assumed to contain as many phenolic groups as analyzed in OSL per weight unit. The synthesis was performed at 60 °C under gentle magnetic agitation for 20 h. The reaction completed, the mixture was cooled to room temperature and then poured into cellulose ester dialysis membranes (Spectra/Por™ cellulose ester dialysis membranes, Spectrum Laboratories, CA, USA) with a molecular weight cutoff of 500–1000 g/mol. The reaction mixture was dialyzed against ion-exchanged water until the conductivity and pH value of the permeate did not change after exchanging washing water. Finally, the lignin solution was freeze-dried to yield the product as a dark-brown fleecy solid. 2.4. Water purification testing
2. Materials and methods The synthesized GTAC-lignins were tested for sulfate removal and kaolin settling, and as a coagulant for humic acid. Kaolin was chosen as a model inorganic suspended solid and humic acid as a natural organic matter present in natural waters and wastewaters.
2.1. Materials Hardwood organosolv lignin (OSL) from an ethanol-based organosolv process was obtained from Fraunhofer Center for ChemicalBiotechnological Processes CBP (Leuna, Germany) and spruce enzymatic hydrolysis lignin (EHL) after steam explosion pretreatment from SP Technical Research Institute of Sweden (Skellefteå, Sweden) as kind gifts. The EHL was freeze-dried and milled with a Thomas Scientific 2279-K38 Variable Speed Digital ED-5 Wiley mill equipped with a 1 mm sieve.
2.4.1. Sulfate removal The untreated lignins (OSL, EHL) and selected cationized lignins (GTAC-OSL produced with 100 and 200 mol-%, and GTAC-EHL produced with estimated 200 mol-% GTAC dosage) were tested for their capability to remove SO42− anions in aqueous solution at pH 3, 6, and 9 at room temperature. The experimental solutions of sulfate (100 mg/L) were prepared by dissolving analytical grade Na2SO4 in ion-exchanged water. Solution pH was adjusted using 0.1 M HCl (pH 3) or 0.1 M NaOH (pH 9). The solution was stirred and GTAC-lignin was added at a dosage of 100 mg/10 mg SO42−. Sampling was done at 10, 30, and 60 min after the lignin addition. For sulfate analysis, the sample solution was filtered through a 0.2 μm membrane (Whatman FP 30/0,2 CA-S, GE Healthcare, Little Chalfont, UK) to separate the lignin material, i.e. the adsorbent. OSL, EHL, and cationized EHL could be separated by the membrane. Cationized OSLs (200 mol-% GTAC and partly 100 mol-% GTAC depending on pH) were soluble and passed the membrane, which was seen as brown color in the filtrate. In a parallel protocol, a 60 min sample was withdrawn and frozen to −18 °C. Freezing caused aggregation and sedimentation of the lignin material, which was separated by decanting from the supernatant after melting the solution. All the solutions were measured for residual sulfate anions by using a P/ACE MDQ capillary electrophoresis (CE) instrument equipped with a photodiode array UV–vis detector working via indirect detection at 230 nm (Beckman-Coulter Inc., Brea, CA, USA) and quantified based on sulfate standards. Electrolyte solutions were commercial from Analis (Ceofix Anions 5, Ordior/Analis, part number: 10-0004650). A bare
2.2. Chemical characterization Elemental analysis (carbon, hydrogen, oxygen, nitrogen and sulfur measured) was done with a Thermo Scientific Flash 2000 Organic Elemental Analyser (Rockford, IL, USA). Composition analysis was performed according to the NREL procedure (Sluiter et al., 2008) with the carbohydrates analyzed from the acid hydrolysate by HPLC as described by Tenkanen and Siika-aho (2000). Zeta potential was measured with a Zetasizer nano ZS (Malvern Instruments, Malvern, UK) in either milli-Q water or 0.1 M NaOH solution. The instrument measures the electrophoretic mobility of particles, which is converted into zeta potential values using the Smoluchowski model. The amount (mmol/g) of aliphatic –OH, carboxylic –OH and phenolic –OH (para-OH phenyl, condensed and syringyl –OH, guaiacyl –OH and catechol –OH groups) was quantified by 31P-NMR after phosphitylation as previously described (Granata and Argyropoulos, 1995). Conductivity measurements were performed with a Hach (Loveland, CO, USA) HQ14d conductivity meter. 189
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
fused silica capillary with inner diameter 75 μm, length to the UV detector 50 cm and total length 60 cm was used. The separation voltage was −30 kV with positive to negative polarity. Samples were injected using 0.5 psi pressure for 8 s. 2.4.2. Kaolin settling A kaolin suspension of 100 mg/L concentration (turbidity ∼100 nephelometric turbidity units, NTUs) was prepared to represent the turbidity level of surface waters. A kaolin product, Capim DG, of small particle size, D50 = 0.56 μm and 92% below 2-μm equivalent diameter (Imerys, Paris, France) was used in the study. Charge density measurements were done by jar tests after the addition of GTAC-OSL (200 mol% GTAC) at doses of 1, 5, 10, 20 and 30 mg/L to the kaolin suspension for the pre-screening of the required GTAC-OSL dosage. The jar tests for coagulation studies were performed using a Kemira Flocculator 2000 device (Kemira OYj, Helsinki, Finland), applying 10 s fast mixing, 10 min slow mixing, and settling overnight, after which a sample was taken from the supernatant for analysis. GTAC-OSL dosages of 1, 2, 3, 4, and 5 mg/L were studied. Turbidity expressed in NTUs was measured using a HACH 2100AN Turbidimeter (Loveland, CO, USA). pH and conductivity were measured using standard hand held meters, VWR pH 100 and VWR EC 300 respectively (Radnor, PA, USA). Charge density was measured by polyelectrolyte titration method using Mütek PCD 03 pH Particle Charge Detector and PCD-Titrator 2 (BTG Instruments GmbH, Herrsching, Germany). The metal salt ferric sulfate PIX-105 from Kemira OYj (Helsinki, Finland) was used as commercial coagulant reference. PIX-105 contained 11.29% of ferric iron (Fe3+), and the density was 1500 g/L. 2.4.3. Humic acid coagulation Humic acid test water was made with technical grade humic acid (Sigma-Aldrich, Inc., St. Louis, MO, USA) by initially preparing a stock solution of 1000 mg/L in 1 M NaOH. The test water representing a typical humic acid concentration in natural surface waters (TOC average in Finnish lakes 7.6 mg/L (Henriksen et al., 1998)) was prepared for this study with a total organic carbon (TOC) amount of 8 mg/L at pH 7. TOC was measured using a TOC-VCPH Total Organic Carbon Analyzer (Shimadzu, Kyoto, Japan). Similar jar tests of coagulation of humic acid as for kaolin suspension were performed.
Scheme 1. Reaction scheme for lignin derivatization with glycidyltrimethylammonium chloride (GTAC) under alkaline conditions and side reaction of GTAC with hydroxide anions.
the solution pH. The reaction is performed in alkaline solution, in which the base, typically NaOH, has the dual role of assisting in lignin dissolution and of activating the phenolic –OH groups by deprotonating them (Kong et al., 2015). The chemical selectivity of the nucleophilic attack on the epoxide reagent should clearly favor the phenolate anion, whereas the aliphatic –OH groups of the lignin, and in the case of the EHL, of the attached residual saccharides, are less nucleophilic. The epoxide is also consumed in a side reaction with the OH− ions, and the reaction is therefore sensitive to the applied conditions and especially reagent concentrations (Kong et al., 2015). In theory, the nucleophilic attack of the phenolate may take place either directly through the deprotonated oxygen of the phenolate, or through the conjugated, unsubstituted C3 and C5 positions. The possibility for a chain reaction, in which the GTAC would react with the newly formed aliphatic –OH groups of the GTAC substituents in the lignin cannot be excluded, even if aliphatic OH-groups are not expected to be sufficiently nucleophilic under the used conditions. In case such reactions take place, the charge of the lignin would be further increased, which could be beneficial if the target is to introduce maximum charge. The cationization of OSL was done with GTAC dosages of 25, 50, 75, 100 and 200 mol-% relative to the analyzed total amount of phenolic groups. EHL was reacted with a GTAC dosage corresponding to 200 mol-% for OSL on sample weight basis. In preliminary experiments, the removal of unreacted or degraded GTAC and NaOH by dialysis was monitored by elemental analysis of freeze-dried GTAC- lignin, whereby the nitrogen content was confirmed to stay constant when the permeate’s conductivity and pH did not increase after changing dialysis water, indicating complete removal of the reagents. The used dialysis membrane with 500–1000 Da molecular weight cutoff was suitable for easily passing through the low molecular weight reagents but no leakage of lignin was noticed based on following the color of the
3. Results and discussion 3.1. Lignin characterization and cationization with GTAC In this work, the cationization of two different lignin types was studied. The organosolv lignin (OSL) was soluble in NaOH solutions whereas the enzymatic hydrolysis lignin (EHL) was insoluble in a wide range of NaOH concentrations (up to 5 M NaOH tested). This difference was explained by the composition analysis, which showed the OSL to be practically pure lignin whereas EHL contained high amounts of carbohydrates, likely cellulose as high glucose amounts (45.1%) were found (Supporting information SI 1). The cellulose residues in EHL inhibited lignin solubility. 31P-NMR was performed to evaluate the amount of reactive phenolic groups in OSL, which was calculated as the sum of condensed and syringyl (i.e. all phenolic units with both the 3 and 5 positions substituted, 2.17 mmol/g), guaiacyl (0.63 mmol/g) and catechol (0.08 mmol/g) groups. OSL thus contained a total 2.88 mmol/ g of phenolic –OH groups. Similar analysis of EHL was not possible due to the sample’s insolubility. For calculating GTAC dosage in lignin derivatization, EHL was assumed to contain the same amount of phenolic groups per weight unit as OSL. Reacting lignin with trimethylammonium-carrying epoxide reagents is an easy and straight-forward way to obtain cationized lignin (Scheme 1). The lignin becomes permanently charged by the attachment of trimethylammonium groups, contrary to the case with amine or carboxyl groups, which are either charged or neutral depending on 190
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
levels as for OSL. The nitrogen content of EHL was only 1.6 wt-% with the 200 mol-% GTAC dosage that produced a nitrogen content of 2.6 wt-% (DS 0.74) for OSL. Because the EHL contained a high proportion of carbohydrates the DS of the lignin fraction in EHL could be fairly high based on the nitrogen content, but it is unknown to which extent the carbohydrates had been cationized during the reaction which complicates any DS calculations for EHL. The cationization under these conditions did not change the solubility of EHL markedly, as had been the original objective. Possibly, employing higher GTAC dosages or reaction conditions in which the aliphatic –OH groups would be more nucleophilic towards reaction with GTAC, could yield cationized EHL with higher DS and improved solubility characteristics. The charge of the lignin changed drastically as a result of the cationization (Table 1). The untreated OSL was practically chargeneutral, but performing the cationization reaction at low GTAC dosages introduced a negative charge on the material (−0.35 meq/g), which was in clear contradiction with original expectations. The negative charge decreased with increasing degree of cationization, and for the GTAC-OSL produced with 200 mol-% reagent dosage, the charge was positive with a value of 0.51 meq/g. Analysis of the 31P-NMR results explained the observed initial introduction of negative charge to the lignin: the amount of carboxylic acid groups in the lignin drastically increased by the applied treatments to a level of 0.40–0.55 mmol/g (Fig. 1). The level of introduced carboxylic acids was constant in all cationized OSL samples and did not further grow when increasing the GTAC dosage. The introduction of carboxylic acid groups to lignin under the mild reaction conditions was surprising and though interesting, the mechanism of the carboxylic acid formation was not further studied in this work. Likely, some chemical functionality in the OSL lignin structure was easily transformed into carboxylic acid independently of the GTAC dosage. In a reference treatment of OSL under alkaline cationization conditions but in absence of GTAC, carboxylic acid groups were similarly introduced, confirming that it was indeed the alkaline treatment that led to carboxyl formation and that the GTAC reagent did not play any notable role in this reaction. Possibly some of the syringyl or condensed groups were easily converted into carboxylic acid, but this hypothesis could not be pursued with the applied analytical methods because the syringyl and condensed signals are overlapping in the 31P-NMR spectra. Alkaline conditions in the presence of oxygen are known to induce anionic charge to lignin (Kalliola et al., 2015). Typically more harsh conditions are needed, but the structures easiest to oxidize appear to react already under the mild conditions applied during the cationization. The presence of both negative and positive charge in the cationized lignin might well influence its solution behavior, as well as performance in water purification applications, and thus should be further studied. With a growing number of ammonium groups attached to the lignin its overall charge grew to be positive. For
Table 1 Nitrogen content (wt-% of dry sample), amount of phenolic –OH groups (mmol/g of dry cationized lignin), degree of substitution (DS) of cationized lignin calculated based on the analyzed change of phenolic –OH groups, charge of cationized lignin (meq/g, milliequivalents per gram) as determined by polyelectrolyte titration (under pH conditions 7.4–8.6) and the assayed solubility of cationized organosolv (OSL) or enzymatic hydrolysis lignin (EHL) lignin samples. NA = Not Applicable. Material
GTAC dose (mol-%)
N (wt%)
Phenolic −OH (mmol/ g)
DS
Charge (meq/g)
OSL
0 25 50 75 100 200
0.2 0.8 1.4 1.8 2.1 2.6
2.88 2.24 1.83 1.58 1.31 0.75
0.00 0.22 0.36 0.45 0.55 0.74
−0.010 −0.354 −0.031 −0.029 −0.007 0.513
EHL
0 (200)
0.2 1.6
NA NA
NA NA
NA NA
Solubility
NaOH (aq) NaOH (aq) NaOH (aq) NaOH (aq) NaOH (aq) NaOH (aq) and water Insoluble Insoluble
solution. Cationized EHL could easily be washed as it was insoluble in water and alkaline solution. The reaction conditions were not further optimized in this work, as they have been studied in detail elsewhere (Kong et al., 2015). The amount of elemental nitrogen in the washed GTAC-lignin samples clearly responded to the GTAC dosage (Table 1). The amount of free phenolic groups as analyzed by 31P-NMR correlated to this result by decreasing in relation to GTAC dosage, suggesting that the primary regioselectivity of the cationization was on the deprotonated phenolate oxygen groups (C4 position). The degree of substitution (DS) was approximated as the ratio of substituted phenolic –OH groups (calculated as the difference between the original amount of phenolic groups in OSL and that analyzed in the derivatized samples) divided by the original number of –OH groups as given by the 31P-NMR results. This approximation assumed that the whole decrease in the amount of phenolic groups was due to derivatization with GTAC and did not take into account the weight increase caused by the attached GTAC substituents. The nitrogen amount in the samples as well as the estimated DS increased with GTAC dosage and possibly higher DS values could be achieved by further increasing the GTAC dosage. The water solubility was tested for all the derivatized samples and it was noticed that only OSL reacted with 200 mol-% of GTAC had become water soluble. This finding indicated that there was a clear threshold DS for GTAC-lignin water solubility, which in the case of the used OSL was between DS 0.55 and 0.74. The derivatization of EHL with a GTAC dosage corresponding to 200 mol-% for OSL clearly increased the nitrogen content of the EHL suggesting that this lignin was also cationized, but the cationization did not proceed to as high
Fig. 1. Abundance of different –OH and carboxylic acid groups in untreated and GTAC-derivatized lignin as analyzed by
191
31
P-NMR.
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
3.2. Water purification testing 3.2.1. Sulfate removal Sulfate is a common constituent of many natural waters and wastewaters, and industrial wastewaters may contain several thousand mg/L of dissolved sulfate (Lens et al., 2010). Several processes can be applied to promote the removal of sulfate, e.g. chemical precipitation by addition of barium or calcium salts, membrane filtration, reverse osmosis, and biological sulfate removal. In this work, sulfate removal using GTAC-derivatized OSL and EHL as sulfate adsorbents was studied. OSL and EHL and their GTAC-derivatives (OSL 100%, OSL 200%, and EHL 200%) were tested for sulfate removal at pH 3, 6, and 9 for 60 min. Thereafter, the solutions were filtered in order to separate the adsorbent material. However, cationized OSLs were mostly soluble and passed the membrane. Therefore, sulfate adsorption capacities of different lignin materials were compared in those samples where lignin material was separated by filtration followed by a freezing-melting procedure (SI 3). The insolubility of GTAC-EHL could thus be a process benefit in water purification applications as it can be easily filtered off the processed liquids as opposed to completely soluble adsorbents. Cross-linking of soluble GTAC-OSL could be studied as an alternative method to have an easier by filtration separable adsorbent. Sulfate adsorption by GTAC-OSL was more efficient than that of untreated OSL in acidic (pH 3) and near neutral (pH 6) conditions (Fig. 2). The highest sulfate adsorption capacity was 54 mg/g, at pH 3, in the case of GTAC-OSL (100 mol-%). In general, the GTAC-OSL (100 mol-%) was slightly more efficient for sulfate removal than GTAC-OSL (200 mol-%), which suggests that the cationicity of the GTAC-lignin might not be the only parameter to determine sulfate removal efficiency. EHL and GTAC-EHL could be separated from the sample solution by membrane filtration due to the insolubility of the samples. For EHL samples, the differences in sulfate concentration between the filtrated and decanted solutions were small (SI 3), thus validating the freezing-melting and decanting procedures. GTAC-EHL adsorbed sulfate in higher amounts than EHL at pH 3 (44 vs. 21 mg/g), but at pH 6 or 9 there was no difference in the adsorption capacity of untreated and cationized EHL, suggesting a strong pH dependency for the sulfate removal capacity of GTAC-EHL (Fig. 2). Ion exchange is likely the major mechanism governing the uptake of sulfate. Under acidic conditions the carboxylic groups, which in this work were found to be introduced to the lignin under cationization conditions, are protonated, which leads to the maximum positive charge in the adsorbent. Under pH 9 conditions, the carboxylic acids are deprotonated causing a lower positive charge in the adsorbent. This phenomenon is likely the reason for the highest sulfate removal capacity of the cationized OSL and EHL being under acidic conditions. In addition to direct sulfate-GTAC-lignin interactions, the charged state of the lignin particles can also be expected to govern the lignin particles behavior in solution which has an indirect effect on the sulfate adsorption capacity. It should also be noted that the charge introduced by GTAC derivatization is permanent and not pH dependent, like for the carboxyl groups. In a previous study, in which coconut coir carbonized with ZnCl2 was studied for sulfate removal in the pH range 2–11, it was found that the optimum pH for sulfate removal was achieved at a linear level between pH 3 and 9, but sulfate removal was lower at < pH 3 or > pH 9, due to adsorption competition between sulfate and the acid or base counter ions (Cl− or OH−) (Namasivayam and Sangeetha, 2008). Adsorption competition is not likely to have played a major role in this study, as the highest sulfate adsorption levels should then have been observed at pH 6, where the counter ion concentration is the lowest. The achieved levels of sulfate adsorption varies by several orders of magnitude (e.g. from 0.06 mg/g of coconut coir pit carbon to 156 mg/g for shrimp shells) for different adsorbent materials as compared by Namasivayam and Sangeetha (2008). Variations between different used experimental conditions may, however, make direct comparisons between literature results difficult.
Fig. 2. Sulfate adsorption capacity of untreated and GTAC-derivatized OSL and EHL with 60 min adsorption time.
EHL, reliable charge titration results were not obtained due to sample insolubility. The 31P-NMR characterization of cationized OSL indicated that the amount of aliphatic –OH groups had increased with increasing DS, which means that the –OH groups formed at the epoxide opening of GTAC are not particularly reactive for further chain reaction with other GTAC molecules (Fig. 1). The amount of condensed and syringyl –OH groups decreased quite linearly as the GTAC dosage was increased, but the guaiacyl –OH groups were less reactive and only decreased slightly with growing GTAC dosage. This result rules out the alternative reaction mechanism according to which the derivatization could take place on the 3 or 5 positions through conjugation of the negative phenolate charge, as this reaction is not possible for the 3,5-disubstituted syringyl units and the guaiacyls with one vacant 3 or 5 position showed only low reactivity. Also, if the guaiacyl units were reactive through their 3 or 5 position, this would be seen as an increase in the amount of condensed groups in 31P-NMR. In this work, the linearly decreasing trend of phenolic –OH groups was almost completely explained by the substitution of the syringyl and condensed –OH groups. This suggests that hardwood lignin, rich in syringyl units, is more amenable to cationization than softwood lignin containing a higher proportion of guaiacyl units. Results by Matsushita and Yasuda (2003) suggest that para-hydroxyl –OH groups would have a slightly higher reactivity than guaiacyl –OH groups towards GTAC when using lignin monomer model compounds, and the results in this work further establish that the condensed and syringyl –OH groups were more reactive than guaiacyl-OH groups towards GTAC, using polymeric lignin samples. The DS of GTAC-OSL (200 mol-% GTAC) was calculated to be 0.74 or in terms of cationicity 2.13 mmol/g (approximation in which the amount of available phenolic –OH groups in lignin has not been corrected with the mass growth caused by the grafted GTAC units), which corresponds to the cationicity previously reported for lignin cationized via the Mannich reaction (cationicity in range of 1.8–2.6 mmol/g) (Fang et al., 2010). The zeta potential was measured for the cationized OSL samples in 0.1 M NaOH solution, and in addition the water-soluble GTAC-lignin was analyzed in water solution (SI 2). In NaOH solution, the zeta potential was clearly negative for the untreated OSL, but the increasing degree of cationization led to rising zeta potential values and for the most cationic lignin sample with DS 0.74, the zeta potential was practically neutral in 0.1 M NaOH, but positive (+24 mV) in water. Du et al. (2014) measured zeta potential values of 27–32 mV in dilute hydrochloric acid solutions for lignin cationized by amination via the Mannich reaction.
192
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al.
important functional groups and binding sites (Greathouse et al., 2014; Kalinichev et al., 2011). In this work, the coagulation of commercial humic acid was studied as a model system for coagulating organic constituents from natural waters. When GTAC-OSL was introduced into the humic acid solution (8 mg/L, corresponding to the concentration of humic acid in natural surface waters), precipitates were formed with GTAC-OSL doses of 50 and 100 mg/L. They contained about half of the total organic matter (TOC) introduced in the test, i.e. the sum of TOC from humic acid and GTAC-OSL (Fig. 4). Possibly, the –OH and carboxyl groups in the GTAC-OSL could form hydrogen bonds with the humic acid and consequently enhance its removal efficiency as proposed in the literature (Zaki and El-Gendy, 2014). Although the formation of carboxylic acid groups under GTAC derivatization conditions to lignin caused its overall charge to decrease significantly especially at lower GTAC dosages, the introduction of the carboxylic groups might have improved the coagulation of humic acid. The commercial ferric sulfate coagulant PIX-105 efficiently reduced TOC at doses ≥ 45 mg/L (corresponding to 5 mg/L Fe3+). The TOC decrease with GTAC-OSL was not as large as with PIX-105, but taking into account that the GTAC-OSL introduced a large amount of organic carbon itself, part of the result may be due to the TOC introduced with the GTAC-OSL. The TOC of humic acid-free solution increased linearly with the GTAC-OSL (200 mol-%) dose due to the coagulants organic nature. To study the coagulation of humic acid with GTAC-OSL in detail, other methods than TOC analysis would be beneficial to use that can discriminate between humic acid and GTAC-lignin.
Fig. 3. Settling of kaolin particles using GTAC-derivatized organosolv lignin (OSL, 200 mol-% GTAC) as coagulant. Left vertical axis displays the turbidity of the kaolin dispersion (in nephelometric turbidity units, NTUs), right vertical axis displays the amount of total organic carbon (TOC) and the dashed line displays the TOC introduced by the addition of coagulant.
4. Conclusions Organosolv (OSL) and enzymatic hydrolysis lignin (EHL) were successfully cationized with (GTAC). Cationized lignins with different solubility characteristics were obtained, GTAC-OSL being water soluble and GTAC-EHL insoluble. GTAC-OSL adsorbed sulfate efficiently, promoted settling of kaolin and precipitated humic acid from solution. GTAC-EHL was efficient for sulfate removal under acidic conditions and was easy to remove in down-stream processing due to its insolubility. Carboxylic acids were introduced to the OSL during the mild conditions of the cationization reaction as a side reaction, which may explain part of the pH-dependency of sulfate removal with the GTAC-lignins.
Fig. 4. Measured amounts of total organic carbon (TOC, mg/L) in humic acid solutions (8 mg/mL) treated with GTAC-derivatized organosolv lignin (200 mol-% GTAC in derivatization) or with commercial PIX-105 coagulant.
3.2.2. Kaolin settling The GTAC-lignins were tested as kaolin settling agents. Particles of kaolin, a hydrous aluminum silicate, carry a negative charge under normal conditions in aqueous suspension (Tombácz and Szekeres, 2006). Electrostatic repulsion between particles can be reduced, e.g. by double-layer compression or ion shielding of the surface charges using electrolytes or polymers such as polyacrylamide, causing the particles to adhere to one another forming aggregates and flocs (Kim and Palomino, 2009; Nasser and James, 2009). Charge density determination revealed that the charge of the particles in kaolin/lignin suspensions remained negative (−3.4 μeq/ L) when 1 mg/L GTAC-OSL (200 mol-%) was added to 100 mg/L kaolin in water suspension, but changed to positive (+2.6 μeq/L) when 5 mg/ L of this product was added. Hence the dosage range from 1 to 5 mg/L was selected for jar tests. A low dose of 2 mg/L of GTAC-OSL (200 mol%), with a narrow optimum performance window from 1.5 to 2.5 mg/L, improved the settling of small kaolin particles (Fig. 3). Cationic GTACOSL (200 mol-%) appeared to neutralize the anionic charge density of kaolin. This made the floc formation possible and increased the settling of microscale particles under the mild settling forces. When the coagulant was overdosed, the charge exceeded the isoelectric point, and the repulsion accordingly increased, leading to increased turbidity. In previous work it has been shown that coagulation efficiency in waste water purification is highly dependent on the dose with clearly defined optima, which are specific for the type of impurity and coagulant used (Kong et al., 2015).
Acknowledgements Tarja Wikström and Vuokko Liukkonen are acknowledged for their skillful technical assistance, Atte Mikkelson for consulting the residual sulfate determination, and Pia Willberg-Keyriläinen for performing the 31 P- NMR analysis. Thanks are due to SP Technical Research Institute of Sweden and Fraunhofer Center for Chemical-Biotechnological Processes CBP for kindly providing the lignin samples. This work has been carried out within the AERTOs Bioeconomy Project 2014–2016 performed by 9 European RTOs. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.indcrop.2017.04.026. References Ahmed, T., Bhatti, Z.A., Maqbool, F., Mahmood, Q., Faridullah, S. Qayyum, Mushtaq, N., 2016. A comparative study of synthetic and natural coagulants for silver nanoparticles removal from wastewater. Desalin. Water Treat. 57, 18718–18723. Cerrutti, B.M., de Souza, C.S., Castellan, A., Ruggiero, R., Frollini, E., 2012. Carboxymethyl lignin as stabilizing agent in aqueous ceramic suspensions. Ind. Crops Prod. 36, 108–115. Du, X., Li, J., Lindström, M.E., 2014. Modification of industrial softwood kraft lignin using Mannich reaction with and without phenolation pretreatment. Ind. Crops Prod. 52, 729–735. Fang, G.Z., He, W.H., Song, Z.Q., 2003. Synthesis and performance of quaternary
3.2.3. Humic acid coagulation In humic acid, carboxyl, hydroxyl and phenolic groups are the 193
Industrial Crops & Products 104 (2017) 188–194
R. Wahlström et al. ammonium salt of lignin as cationic flocculant. Linchan Huaxue Yu Gongye/Chem. Ind. For. Prod. 23, 37–41. Fang, R., Cheng, X., Xu, X., 2010. Synthesis of lignin-base cationic flocculant and its application in removing anionic azo-dyes from simulated wastewater. Bioresour. Technol. 101, 7323–7329. Granata, A., Argyropoulos, D.S., 1995. 2-Chloro-4,4,5,5-tetramethyl-1,3,2dioxaphospholane, a reagent for the accurate determination of the uncondensed and condensed phenolic moieties in lignins. J. Agric. Food Chem. 43, 1538–1544. Greathouse, J., Johnson, K., Greenwell, H., 2014. Interaction of natural organic matter with layered minerals: recent developments in computational methods at the nanoscale. Minerals 4, 519–540. Henriksen, A., Skjelvåle, B.L., Mannio, J., Wilander, A., Harriman, R., Curtus, C., Jensen, J.P., Fjeld, E., Moiseenko, T., 1998. Northern European Lake Survey, 1995: Finland, Norway, Sweden, Denmark, Russian Kola, Russian Karelia, Scotland and Wales. Ambio 27, 80–91. Kalinichev, A.G., Iskrenova-Tchoukova, E., Ahn, W.Y., Clark, M.M., Kirkpatrick, R.J., 2011. Effects of Ca2+ on supramolecular aggregation of natural organic matter in aqueous solutions: a comparison of molecular modeling approaches. Geoderma 169, 27–32. Kalliola, A., Vehmas, T., Liitiä, T., Tamminen, T., 2015. Alkali-O2 oxidized lignin – a biobased concrete plasticizer. Ind. Crops Prod. 74, 150–157. Kim, S., Palomino, A.M., 2009. Polyacrylamide-treated kaolin: a fabric study. Appl. Clay Sci. 45, 270–279. Kong, F., Parhiala, K., Wang, S., Fatehi, P., 2015. Preparation of cationic softwood kraft lignin and its application in dye removal. Eur. Polym. J. 67, 335–345. Lee, C.S., Robinson, J., Chong, M.F., 2014. A review on application of flocculants in wastewater treatment. Process Saf. Environ. Prot. 92, 489–508. Lens, P.N.L., Visser, A., Janssen, A.J.H., Pol, L.W.H., Lettinga, G., 2010. Biotechnological treatment of sulfate-rich wastewaters. Crit. Rev. Environ. Sci. Technol. 28, 41–88. Matsushita, Y., Yasuda, S., 2003. Preparation of anion-exchange resins from pine sulfuric acid lignin, one of the acid hydrolysis lignins. J. Wood Sci. 49, 423–429. Namasivayam, C., Sangeetha, D., 2008. Application of coconut coir pith for the removal of sulfate and other anions from water. Desalination 219, 1–13.
Nasser, M.S., James, A.E., 2009. The effect of electrolyte concentration and pH on the flocculation and rheological behaviour of kaolinite suspensions. J. Eng. Sci. Technol. 4, 430–446. Rösch, J., Mülhaupt, R., 1994. Mechanical properties of organosolv-lignin-filled thermoplastics. Polym. Bull. 32, 361–365. Rastogi, S., Dwivedi, U.N., 2008. Manipulation of lignin in plants with special reference to O-methyltransferase. Plant Sci. 174, 264–277. SCAN-W 12:04 (Scandinavian Pulp, Paper and Board testing committee method), 2004. Process Waters, Cationic Demand, Polyelectrolyte Titration with a Streaming Current Detector . Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2008. Determination of Structural Carbohydrates and Lignin in Biomass. Laboratory Analytical Procedure (LAP). National Renewable Energy Laboratory, Golden, Colorado. Tenkanen, M., Siika-aho, M., 2000. An alpha-glucuronidase of Schizophyllum commune acting on polymeric xylan. J. Biotechnol. 78, 149–161. Tombácz, E., Szekeres, M., 2006. Surface charge heterogeneity of kaolinite in aqueous suspension in comparison with montmorillonite. Appl. Clay Sci. 34, 105–124. Várnai, A., Siika-aho, M., Viikari, L., 2010. Restriction of the enzymatic hydrolysis of steam-pretreated spruce by lignin and hemicellulose. Enzyme Microb. Technol. 46, 185–193. Yang, A.L., Gao, W., Wei, W.Y., Jiang, W.J., 2008. Synthesis and flocculation performance of a new quaternary ammonium salt of lignin. Zhongguo Zaozhi Xuebao/Trans. China Pulp Pap. 23, 60–63. Yang, R., Li, H., Huang, M., Li, A., Yang, H., 2016. Chitosan-based flocculants and their applications in water treatment. Water Res. 95, 59–89. Zaki, A.A., El-Gendy, N.A., 2014. Removal of metal ions from wastewater using EB irradiation in combination with HA/TiO2/UV treatment. J. Hazard. Mater. 271, 275–282. Zhang, Q., Wang, D., Bei, Y., Ren, S., Fang, G., 2013. Flocculation performance of trimethyl quaternary ammonium salt of lignin-sodium alginate polyampholyte. BioResources 8, 3544–3555.
194