Efficient eco-clean upgrading of isolated cellulose fibers by polyoxometalate (POM) catalyzed ozonation boosted by enzymes

Anatoly A. Shatalov
Forest Research Centre (CEF), School of Agriculture, University of Lisbon, Tapada da Ajuda, 1349-017 Lisbon, Portugal. E-mail: anatoly@isa.ulisboa.pt; Fax: +351 213653338

Received 21st July 2017 , Accepted 8th September 2017

First published on 8th September 2017

An extremely selective and effective delignification (bleaching) approach for upgrading of isolated cellulose fibers by ozone in the presence of mixed-addenda α-Keggin-type polyoxometalates (POM), such as heteropolyanions of series [PMo(12−n)VnO40](3+n)−, in aqueous solvent solutions has been developed, which is substantially superior to conventional bleaching techniques. Additional lignin removal (by 38.3%), improvement in brightness (by 16.5%), and the simultaneous increase in intrinsic viscosity (by 5.4%) observed after POM/ozonation of commercial eucalypt kraft pulp led to significant improvement in process selectivity and efficiency (by 157% and 125%, respectively) compared to common ozonation in water (with no catalyst and solvent). The intensification of POM catalysis by enzymatic pre-treatment of kraft pulp with highly specific xylanase preparations allowed further improving the selectivity and efficiency of POM-ozonation by ca. 60% and 40%, respectively. The integration of the POM/O3 stage into short totally chlorine-free bio-bleaching sequences, with hydrogen peroxide or/and alkali as the only additional reagents, made it possible to achieve the target properties of high grade cellulose fibers for commercial applications in a sustainable and ecologically friendly way.

1. Introduction

The global concerns about energy security and climate change have sparked a surge of interest in the exploration of alternative (renewable) energy sources, such as biomass.1 The integrated biorefinery concept of biomass conversion into fuels and chemicals is now viewed as the most potential approach towards a sustainable bio-based economy.2

In particular, lignocellulosic biomass has been the focus of much current attention, as a widespread, abundant, diverse and low-cost renewable feedstock for biorefining technologies.3 Lignocellulosics are largely composed of three main chemical constituents (biopolymers): cellulose (35%–50%), hemicelluloses (25%–30%) and lignin (15%–30%).4 Development of cost-effective eco-clean methods for primary fractionation (separation) of these biopolymers is of critical importance for sustainable biomass valorization. The main and most important (industrial) approach to separate the carbohydrate portion from lignin is delignification, when biomass is treated by lignin-degrading chemicals (chemical pulping) to achieve lignin fragmentation and solubilization.5 The remaining solid residue (chemical pulp) consists essentially of cellulose, partially degraded hemicelluloses and residual (3–5% of total) lignin.5 Commercially, on a global scale, about 89% of chemical pulps in the world are produced by kraft pulping and only about 5% – by the sulfite process.6 Additional delignification (bleaching) is needed to upgrade the custom properties (such as the degree of purity, brightness, etc.) of isolated cellulosic fibers (chemical pulps). Chlorine-based bleaching using chlorine dioxide (ClO2) as a primary reagent for the oxidative degradation and removal of residual lignin (so-called ECF, or elemental chlorine-free bleaching) is still a predominant industrial process for the preparation of high grade fully-bleached pulps.7

In the last two decades, the growing environmental public pressure to eliminate the discharges of chlorinated toxic substances, particularly chlorinated organics, from bleach mill effluents has caused substantial interest in non-chlorine oxidative bleaching chemicals, such as molecular oxygen, hydrogen peroxide and ozone for totally chlorine-free (TCF) bleaching technology.8 Of all oxygen-centered chlorine-free bleaching reagents, ozone is the most powerful and particularly potential oxidizing agent.9 However, despite the extremely high oxidizing potential of ozone (+2.07 eV),10 the low ozone selectivity towards lignin, due to unwanted reactions with carbohydrates, restricts the delignification capacity of ozone bleaching and limits its further application. Particularly, the secondary reactive species of ozonation solution such as the free-radical by-products of ozonation (hydroxyl and perhydroxyl radicals) react more readily with carbohydrates, having a dramatic impact on the strength properties and yield of the resulting (bleached) cellulose fibers.9 The selectivity improvement of ozone bleaching still needs to be solved. The change in the redox properties of the oxidation bleaching system by application of highly specific chemical catalysts for the delignification reaction, such as polyoxometalates, can be a feasible way to increase the selectivity and efficiency of cellulose pulp ozonation.

The early transition metal oxygen anion clusters (polyoxometalates, POMs) find wide application as green catalysts for selective oxidation of different organic substances, including phenolics.11 Heteropolyoxometalates (free acids and salts of heteropolyanions, HPAs) possessing the properties of both strong acids with extremely high Brønsted and Lewis acidity and very powerful multi-electron oxidants hold great interest now as bi-functional catalysts for various homogeneous and heterogeneous reaction systems.12 The unique combination of HPA properties, such as the structural and functional mobility that can be easily controlled during their rather simple and low-cost synthesis, high solubility in water and various oxygen-containing organic solvents, high stability over a wide range of reaction conditions and, finally, easy regeneration (re-oxidation) after utilization made HPAs very attractive catalysts for delignification of lignocellulosics.13

Enzymatic catalysis also offers environmental alternatives to conventional separation techniques. The natural origin, non-toxicity and mild operation conditions of enzymes stimulated their applicability in the pulp and paper industry (so-called bio-pulping and bio-bleaching).14,15 The use of hemicellulolytic (such as xylanases) and ligninolytic (such as fungal laccases) enzymes can modify the cell-wall structure by partial carbohydrate and lignin degradation, thereby facilitating subsequent main chemical processing and acting as an auxiliary intermediate separation stage.

An extremely selective and effective bleaching approach for upgrading of chemical pulps by ozone in the presence of mixed-addenda α-Keggin-type HPAs of series [PMo(12−n)VnO40](3+n)− in solvent media has been developed, which is substantially superior to conventional bleaching techniques. With an aim to maximize the catalytic effect of POMs, the enzymatic pre-treatment of chemical (kraft) pulp with xylanase preparations before POM-catalyzed ozonation has been done, and the integration of the POM/O3 bleaching stage into short TCF bio-bleaching sequences together with other oxygen-centered (non-chlorine) oxidative reagents has been examined. The principal results of this study are discussed in the present paper.

2. Experimental section

2.1. Materials and chemicals

Industrial unbleached eucalypt (E. globulus L.) kraft pulp (Portucel Mill, Portugal) having 42.6% ISO brightness, 11.8 kappa number and 1291 cm3 g−1 CED intrinsic viscosity was used for this study. Before bleaching, the pulp was thoroughly washed with deionized water to remove all residual black liquor, air-dried and analyzed on chemical composition.

The commercial xylanase preparation (endo-1,4-β-xylanase activity, EC was purchased from AB Enzymes (Germany). The xylanase activity of the product was standardized by the supplier to 190[thin space (1/6-em)]000 XU g−1, where one xylanase unit (XU) was defined as the amount of enzymes needed to produce carbohydrates with a reducing power corresponding to one nanomole of xylose from birch xylan in one second under assay conditions. The DNS-xylanase assay was used to check the xylanase activity in citrate phosphate buffer at pH 7.16

Aqueous 0.2 M solution of heptamolybdo-pentavanado-phosphate heteropolyanion HPA-5 (sodium salt) was synthesized by stoichiometric reaction of MoO3, V2O5, NaH2PO4 and Na2CO3 according to a previously described procedure.17

All other chemicals used were of analytical grade purity and purchased by Sigma-Aldrich, Fluka and Riedel-de-Haën chemical companies.

2.2. Cellulose pulp treatment

Enzymatic pre-treatment of kraft pulp (X-stage) was performed in sealed double-layer plastic bags incubated in a water bath under conditions recommended by the supplier: 76 XU g−1 enzyme dosage; 10% consistency; pH 7.0; 65 °C; and 180 min. After the enzymatic stage, the pulp samples were carefully washed with deionized water and used for further bleaching experiments. The reference (control) pulp samples were treated exactly the same way, but without enzyme addition. Two replicated enzymatic as well as control treatments were performed.

Catalytic ozonation (ZPOM-stage) (3% consistency; 0.8% ozone charge; pH 2; 20 °C) was performed in a 1 L Fischer glass batch reactor equipped with a high-speed Teflon-covered stirrer and connected with a laboratory Fischer-502 ozone generator. Acidic pulp treatment (pH 2; 30 min) was performed before ozonation to increase ozone selectivity towards lignin. After ozonation, the pulp samples were thoroughly washed with deionized water. Ozone concentration was measured by the common iodometric method. Two replicated ozonation experiments were performed for each experimental condition set.

Hydrogen peroxide bleaching (P-stage) was performed under the following conditions: 1.5% NaOH; 2.5% H2O2; 90 °C; 240 min and 10% consistency. Some additional chemicals were used in this bleaching stage, such as Epsom salt (magnesium sulfate, 0.3%) and DTPA (diethylenetriaminepentaacetic acid, 0.2%), to prevent the radical degradation reactions of carbohydrates. Pulp chelating (Q-stage) with ethylenediaminetetraacetic acid (EDTA) was done before the hydrogen peroxide bleaching to remove the transition metals and therefore to prevent peroxide decomposition. The peroxide bleaching as well as the pulp chelating were performed in sealed plastic bags plunged into an agitated water bath with controlled heating. After the bleaching, the pulp samples were thoroughly washed by deionized water.

Alkaline extraction in the presence of a reducing agent (ER-stage: 1.0% NaOH; 0.1% NaBH4; 60 °C; 60 min and 10% consistency), as well as alkaline extraction in the presence of hydrogen peroxide (Ep-stage: 2.0% NaOH; 0.5% H2O2; 70 °C; 180 min and 10% consistency), was performed in sealed plastic bags plunged into an agitated water bath, in a procedure similar to that for hydrogen peroxide bleaching.

To assess the bleach boosting effect of xylanases on catalytic ozonation, the control (enzyme-free) pulp samples were processed under essentially identical chemical bleaching conditions applied to the pulp samples pre-treated with enzymes.

2.3. Analytical methods

Residual lignin in pulp samples was determined as a kappa number according to TAPPI T 236 cm-85. Pulp viscosity was measured in cupriethylenediamine (CED) solution according to SCAN-CM 15:88. Pulp optical properties (ISO brightness and DIN 6167 C/2 yellowness index) were measured by CM-3630 Spectrophotometer (Minolta) using Paper Control software.

Carbohydrate analysis was performed by HPLC using an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) operating at 50 °C, in combination with a cation H+-guard column (Bio-Rad, Hercules, CA, USA). The mobile phase was 10 mN sulfuric acid and the flow rate was 0.6 ml min−1.

Hexenuronic acid groups were quantified by selective hydrolysis in formic acid–sodium formate buffer followed by UV spectroscopy (Shimadzu, UV-160A) of the formed 2-furoic acid at 245 nm.

The surface morphology of the treated cellulose fibers was studied by scanning electron microscopy (SEM) using a JEOL FEG-SEM model JSM7001F with an accelerating voltage of 15 kV. The samples were deposited on a double-sided carbon tape and coated with a carbon film on a Quorum Technologies model Q150 T ES.

3. Oxidation catalysis by HPAs

POMs, as the metal oxygen anion clusters, are primarily composed of the d0 early transition metal cations in their high oxidation states (most commonly WIV, MoVI and VV) and oxide anions (oxoanions) with a variety of structures and sizes.11,18,19 Structurally, POMs are built from MOx (M – metal cation; x = 4–6) polyhedra blocks (most commonly MO6 octahedra) linked together by one (corner sharing), two (edge sharing) or three (face sharing) oxygen atoms (bridging oxo-groups) (Fig. 1). POMs are generally divided into two generic classes: the isopolyoxometalates [MmOy]n, containing only d0 metal cations and oxide anions and the heteropolyoxometalates, as soluble polyoxoion salts of anions of general formula [XxMmOy]q (xm), which contain one or more d or p heteroatoms X (P, Si, B, etc.) additionally to addenda atoms M (early transition metal in high oxidation state, such as MoVI and WVI) and oxide anions.11
image file: c7gc02218b-f1.tif
Fig. 1 The Keggin structure of α-[PM12O40]3− anion in polyhedral and ball-and-stick representation. The central PO4 tetrahedron is surrounded by 12 MO6 octahedra: the terminal (Ot), corner-sharing (Oc), edge-sharing (Oe) and phosphorous-linked (Op) oxygen atoms and the metal atoms (primarily W, Mo, and V) are depicted by small and big balls, respectively.

The α-Keggin-type mixed-addenda heteropolyanions [XxM′nM12−nO40]q and, particularly, molybdo-vanado-phosphate heteropolyanions (Mo-V-P)-HPAs of series [PMo12−nVnO40](3+n)− (HPA-n, where n denotes the number of vanadium atoms in HPA composition), were recognized as very efficient catalysts for homogeneous liquid-phase oxidation reactions.20–22

Oxidation catalysis by (Mo-V-P)-HPAs occurs due to the ability of VV to accept electrons from the substrate via a Mars-van Krevelen mechanism:

HPA-n + Sub + mH+ → Hm(HPA-n) + Subox(1)
where Hm(HPA-n) = Hm[PMo12−nVVnmVIVmO40](3+n)− is the partially reduced form of the catalyst with m atoms of vanadium(IV). To maintain the charge of the polyanion, the HPA-n reduction in solution is accompanied by its protonation.23

The reduced VIV can be oxidized back to VVvia reaction with an appropriate oxidant (such as dioxygen, ozone or hydrogen peroxide), closing thereby the redox cycle by catalyst:

Hm(HPA-n) + m/4O2 → HPA-n + m/2H2O(2)

The (Mo-V-P)-HPAs of series [PMo(12−n)VnO40](3+n)− with n > 1 have therefore the property of easily reversible multi-electron oxidants. Substitution of MoVI by VV in the polyanion molecule tends to increase the oxidation potential of HPAs as a result of extra-electron trapping by vanadium atoms.12 The effective HPAs re-oxidation (recuperation) in the separate reactor (two-stage oxidation) or in the same reactor (one-stage oxidation) provides selectivity of the overall catalytic process.

The HPA-n solutions are the complex systems containing multiple reactive forms of catalysts. Under acidic conditions, the parent (Mo-V-P)-HPAs undergo reversible dissociation with release of VO2+ ions and formation of so-called defect (lacunary) heteropoly species. The partial reduction of HPAs, with formation of heteropoly blues or mixed-valence complexes containing V(IV) [PMo12−nVVnmVIVmO40](3+n)−, causes an additional reversible release of VO2+ ions from the coordination sphere of the polyanion:12

[Hn+m−1PMo12−nVVnmVIVmO40]4− ↔ VO2+(3)

In reaction solution, the VO2+ and VO2+ ions stay in equilibrium with the heteropolyanion species and with one another:

VO2+ + H2O ↔ VO2+ + 2H+ + e(4)

The VO2+ ions, having higher oxidation potential than the parent HPA (0.87 V vs. 0.71 V, respectively), were shown to be the principal active species in the catalyzed aerobic (dioxygen) oxidation reactions of organics.23

The high redox reversibility of POMs (HPAs) and their readiness for the oxidation of phenolic (lignin-like) compounds24,25 suggested the use of POM catalysts in the delignification technology of lignocellulosic biomass for selective catalytic oxidation (degradation) of native and residual lignins, such as the dioxygen delignification of wood chemical pulps.13 Of all tested (Mo-V-P)-HPAs, the heteropolyanion [PMo7V5O40]8− or HPA-5 showed the best bleaching results in terms of delignification ability and re-oxidation in the presence of molecular oxygen.26 However, the high medium acidity (pH 1–2) required for the optimal function and structural stability of (Mo-V-P)-HPAs, and the poorly-controlled release of VO2+ (or VO2+ in the case of reduced HPA) ions from the parent HPA under acidic conditions, affected carbohydrate complex of the bleached pulp samples, causing undesirable hydrolytic and oxidative polysaccharide degradation with respective loss in pulp viscosity and strength,27 which restricted substantially the possibility for commercial realization of this technology.

4. HPA-catalyzed ozone delignification

The low pH of the reaction medium (pH 1–2) needed for oxidative catalysis by (Mo-V-P)-HPAs, being a main limitation for oxygen delignification, is a great advantage for ozone delignification, which operates under the same (acidic) pH range (pH 1–2). In HPA/O3 bleaching, the extremely high oxidation potential of ozone can reduce substantially the redox cycle of catalyst regeneration in comparison with oxygen or peroxide delignification, thereby providing an accelerated rate of bleaching reactions and a higher efficiency of the bleaching process as a whole. The process selectivity is provided, in this case, by the stoichiometric electron-transfer mechanism of HPA-catalyzed lignin oxidation, in contrast with nonselective and uncontrolled radical-induced ozone delignification, inevitably accompanied by substantial degradation of carbohydrates.

The development of a new potential POM-catalyzed ozone-based bleaching system has been reported.28–30 The cellulosic pulp ozonation catalyzed by the HPA-5 [PMo7V5O40]8− polyanion in organic solvent media was found to be an extremely effective and selective bleaching approach, providing a method for a substantial increase in pulp selectivity and delignification in comparison with other known ozone-based bleaching techniques.

According to eqn (1) and (2), the general mode of HPA action in ozone delignification can be expressed by the following reaction scheme:

HPA(ox) + lignin → HPA(red) + lignin(ox)(5)
HPA(red) + O3 → HPA(ox) + H2O(6)

Selective pulp bleaching (delignification) in HPA/O3 reaction system is possible due to fulfillment of the following thermodynamic conditions:

E° (lignin) < E° (HPA) < E° (O3)(7)
where E° (lignin), E° (HPA) and E° (O3) are the oxidation potentials, respectively, of lignin (0.4–0.6 eV vs. NHE at pH 1), HPA (0.68–0.71 eV vs. NHE at pH 1) and ozone (2.07 eV at pH 2).

HPAs having a lower kinetic barrier against lignin oxidation than that of ozone, oxidize the lignin, and then the reduced form of HPAs is re-oxidized by ozone at the same reaction stage. The selectivity is provided by the stoichiometric electron-transfer mechanism of POM catalyzed lignin oxidation, in contrast to unselective and uncontrolled radical-induced ozone delignification inevitably accompanied by substantial carbohydrate degradation. The addition of organic solvents, as radical scavengers, was expected to further improve ozonation performance in terms of process selectivity.

A number of common low-boiling polar aprotic and protic organic solvents were tested as a potential reaction media for POM-catalyzed ozone delignification of industrial eucalypt (E. globulus) kraft pulps.28 It has been shown that even a moderate proportion of the solvent (6% w/w) in the reaction solution can significantly improve the selectivity and efficiency of POM-catalyzed ozonation in comparison with the solvent-free (POM/water) and, especially, with the conventional (water) ozonation systems, as a result of the remarkable increase in pulp delignification and brightening, as well as the reduced degradation of cellulose fiber. Four reaction systems using methanol, ethanol, acetone and dioxane showed particularly significant results, substantially superior to those of solvent-free experiments (Fig. 2). The best ozonation selectivity (104% and 80% increase over the control (water) and HPA/water, respectively) was shown in the HPA/ethanol solution, followed by dioxane, acetone and methanol. The HPA/acetone solution showed the best ozonation effectiveness both in terms of pulp delignification (73% and 60% increase over the control and HPA/water, respectively) and brightening (32% and 30% increase), followed by the methanol, ethanol and dioxane solutions.

image file: c7gc02218b-f2.tif
Fig. 2 Effect of different organic solvents on selectivity (as lignin decrease per unit of intrinsic viscosity decrease) and efficiency (as lignin decreases and brightness increases per unit of ozone applied) improvement of POM-catalyzed ozonation of industrial eucalypt (E. globulus) kraft pulp in comparison with a conventional control (without catalyst and solvent) ozonation in water. Ozonation conditions: 6% (w/w) solvent; [O3] = 0.8%; [HPA-5] = 0.5 mM; pH = 2.

The strong positive effect of organic solvents on POM-catalyzed ozone delignification, besides the above mentioned suppression of radical-induced polysaccharide degradation, can be explained by better ozone solubility in organic solvents in comparison with water, as a result of reduced interfacial tension of the liquid phase,31 leading to facilitated ozone mass transfer to the solvent solution.32 Within one-stage POM/O3 bleaching (the case of this study), the higher concentration of dissolved (active) ozone in solution enhances the catalyst (HPA) regeneration (re-oxidation) during bleaching (eqn (6)), accelerating in its turn the rate (and efficiency) of delignification reaction as a whole (eqn (5)). The higher solubility of lignin (or lignin degradation products) in organic solvents can also contribute to better pulp delignification during solvent-based ozonation in comparison with water.33 The dissolution effect varies for different solvents and depends on their hydrogen-bonding capacity (δ-value), which normally increases in solvent mixtures with water.34

To confirm the effect of the solvent on POM oxidation catalysis, the re-oxidation capacity of the partially reduced HPA-5 polyanion (HmHPA-5 or heteropoly blue) by ozone (eqn (2)) in aqueous acetone solution (10 vol%) and water (control) has been examined under conditions simulating the pulp bleaching process, and the kinetics of POM re-oxidation by ozone has been described, using an original analytical approach developed for multi-component reaction systems.30 It has been shown that the bulk portion (ca. 90%) of the partially reduced catalyst can be readily re-oxidized back (recovered) within the first few minutes of ozonation (Fig. 3, top), thus confirming the closure of the redox catalyst cycle, as a pre-requisite for the practical feasibility of oxidation catalysis by POMs. As suggested, the addition of an organic solvent into the reaction mixture had favorable effect on HmHPA-5 re-oxidation by ozone, increasing the effective reaction rate constant of catalyst recovery by up to 50% compared with ozonation in water (k of 0.29 min−1vs. 0.43 min−1 for ozonation in water and 10% v/v acetone, respectively, Fig. 3, bottom), and was evidently one of the principal reasons for the high efficiency of solvent-assisted HPA/O3 bleaching. The residual portion of HmHPA-5 reacted very slowly (at a two orders lower rate) with ozone. The slow formation of active intermediate complexes between HmHPA-5 and O3 under low degrees of catalyst reduction (m < 1.5) at the end of the re-oxidation reaction, similar to that observed for (Mo-V-P)-heteropoly blues oxidation by dioxygen,35,36 was suggested to be a rate-determining reaction step responsible for this low-rate stage of catalyst re-oxidation.

image file: c7gc02218b-f3.tif
Fig. 3 (Top) Re-oxidation of partially reduced catalyst (HmHPA-5) by ozone in water (1) and 10 vol% acetone solution (2) at a 0.39 mmol min−1 O3 flow rate and pH 2. (Bottom) Kinetics of HmHPA-5 re-oxidation by ozone in 10 vol% acetone solution: (1) experimental kinetic curve ln[HmHPA-5] = ln([HmHPA-5]1 + [HmHPA-5]2) = f(t); (2) calculated kinetic curve ln[HmHPA-5]2 = f(t); (3) calculated kinetic curve ln([HmHPA-5] − [HmHPA-5]2) = ln[HmHPA-5]1 = f(t); [HmHPA-5]1 and [HmHPA-5]2 are two reactive forms of catalyst in solution.

The effect of solvent and catalyst concentration, pH and ionic strength on the ozone delignification (bleaching) of industrial eucalypt kraft pulp was examined using aqueous acetone solution as a model solvent-based reaction system.29 The solvent proportion was found to be the most influencing factor of HPA-catalyzed ozonation, having a significant effect on the efficiency and selectivity of the delignification process, followed by reaction medium acidity and HPA concentration. The elimination of the Donnan effect by an increase in the medium ionic strength in the pulp suspension was also very effective in further enhancing the delignification capacity of the tested catalytic bleaching system. Under the optimized ozonation conditions (40% w/w acetone, [O3] = 0.8%, [HPA] = 1.0 mM, pH 2), additional lignin removal by 38.3%, increase in brightness by 16.5% and a simultaneous increase in intrinsic viscosity by 5.4%, of the bleached kraft pulp were achieved compared to the control ozonation (without solvent and catalyst) in water. Impressive improvement in the ozonation selectivity and delignification efficiency (by 157% and 125% compared to conventional (water), and by 108% and 97% compared to HPA/water ozonation) has been demonstrated (Fig. 4), pointing to the high potential of the developed solvent-based catalytic oxidative system for quality upgrading of chemical cellulose pulps.

image file: c7gc02218b-f4.tif
Fig. 4 Effect of xylanase pre-treatment on selectivity and efficiency improvement of optimized POM-catalyzed ozonation of industrial eucalypt (E. globulus) kraft pulp in aqueous acetone solution in comparison with conventional control (with no catalyst and solvent) ozonation in water. Ozonation conditions: 40% (w/w) solvent; [O3] = 0.8%; [HPA-5] = 1.0 mM; pH = 2.

5. Enzymatic boosting of HPA-catalyzed ozonation

5.1. Effect of xylanase pre-treatment

Pre-activation (pre-treatment) of cellulose chemical pulp with highly specific hemicellulolytic enzymes before POM-catalyzed delignification has been examined with an aim of further intensification of POM catalysis. To assess the effect of enzymatic pre-treatment, the industrial eucalypt (E. globulus) kraft pulp was treated with commercial xylanase preparation (endo-1,4-β-xylanase activity; EC followed by ozonation in aqueous acetone solution in the presence of [PMo7V5O40]8− polyanion (HPA-5), using previously established optimal ozonation conditions.29 The bleaching results were compared with those for control pulp samples processed under essentially identical conditions, but without the use of enzymes.

The key role of xylanases in cellulose pulp bleaching is to enhance the bleaching effect of other chemicals in subsequent bleaching stages, i.e., bleach boosting.8 The limited hydrolysis of the xylan network improves fiber permeability thereby increasing the accessibility of lignin to bleaching reagents and facilitating the removal of lignin degradation products in solution. The positive effect is commonly attributed to the selective hydrolysis of re-deposited xylan from the fiber surfaces and pores and/or increased extractability of lignin–carbohydrate complexes (LCC).15,37 So, it was highly expected that xylanases will boost the following POM-catalyzed ozonation by providing easier access for large POM molecules to the active reaction centers of lignin in the fiber cell-walls.

It can be seen from Table 1 that even simple control (with no enzyme) hydrothermal treatment under gentle conditions caused some change in pulp properties compared with the initial (untreated) kraft pulp, as a result of the occurring auto-hydrolytic processes. The exclusive effect of xylanases on pulp bleaching was therefore assessed through comparison with the control samples. Similar to our previous observations,38,39 the bleaching action of xylanases, so-called direct brightening and delignification, was already noted immediately after the enzymatic X-stage. This caused the increase in brightness and decrease in lignin content (by 7.7% and 13.2%, respectively, Table 1) over the control, as a result of enzymatic attack on LCC with removal of some lignin components and lignin-associated chromophore groups. The increased proportion of cellulose in the xylanase-treated pulp samples (due to partial removal of xylan) was an obvious reason for the enhanced (by 13 points) intrinsic viscosity as compared with the control.

Table 1 Enzymatic pre-treatment of industrial eucalypt (E. globulus) kraft pulp with a commercial xylanase preparation (EC and its effect on subsequent POM-catalyzed ozonation in aqueous solvent solution
  Initial pulp Xa X-ZPOM[thin space (1/6-em)]b
Controlc Enzyme Control Enzyme
a Enzymatic pre-treatment (X-stage): pH 7.0; 65 °C; 180 min b POM-catalyzed ozonation (ZPOM-stage): 0.8% O3; 1.0 mM HPA-5; 40% (w/w) acetone; pH 2. c Enzyme-free treatment. d Corrected to contribution of HexA where each 10 μmol of HexA corresponded to kappa number increase by 0.86 units.46 e Monosaccharide content as % on initial (untreated) oven-dry pulp. f Hexenuronic acid content as μmol per gram of oven-dry pulp. g Trace quantities.
Brightness (% ISO) 42.60 43.45 46.81 64.30 71.84
Yellowness index (DIN 6167) 34.10 33.27 30.06 21.16 14.94
Kappa number 11.84 11.12 9.65 3.14 1.85
Kappa number actuald 7.55 6.85 6.03 2.21 1.21
Intrinsic viscosity (mL g−1) 1291 1293 1306 986 996
Glucose (% odp)e 82.53 82.40 82.33 81.78 81.66
Galactose (% odp) 0.44 0.43 0.36 0.36 0.24
Mannose (% odp) 0.16 0.15 0.15 Trg Tr
Xylose (% odp) 16.16 15.66 14.93 14.58 13.65
Arabinose (% odp) 0.37 0.35 0.34 0.27 0.26
HexA (μmol g−1)f 49.84 49.63 42.15 10.86 7.55

The branched heteroxylan of E. globulus wood, which represents ca. 20% of total wood carbohydrates, has a unique and quite unusual structure for wood xylans, being a (2-O-α-D-galactopyranosyl-4-O-methyl-α-D-glucurono)-D-xylan composed of galactosyl, 4-O-methyl-glucuronosyl and xylosyl residues with a molar ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]3[thin space (1/6-em)]:[thin space (1/6-em)]30, respectively.40 During kraft pulping, the structure of this heteroxylan is modified significantly due to intense scission of the xylan backbone and side-chain elimination, which reduces the average molecular weight of xylan in kraft pulp from 25[thin space (1/6-em)]600 to 13[thin space (1/6-em)]700.41 The 4-O-methyl-glucuronic acid (MeGlcA) groups are degraded by ca. 75–90% under alkaline pulping conditions, and the residual MeGlcA in the kraft pulps are almost completely (by 83–88%) converted to hexenuronic acid (4-deoxy-β-L-threo-hex-4-enopyranosyluronic acid or HexA) by β-elimination of methanol.42,43 As can be seen from Table 1, the xylanase pre-treatment affected solely the xylan-associated monosaccharides, leading to removal of the 4.7% xylose, 16.3% galactose and 15.2% hexenuronic acid groups. No cellulose degradation by xylanases was observed, pointing to the high purity (selectivity) of the used xylanase preparations and the absence of any cellulase activity. HexA, due to its unsaturated nature, can have a negative effect on the optical properties (brightness and brightness reversion) of the bleached pulps,44 acting as carbohydrate-derived chromophores during bleaching. It has been reported that the direct brightening effect of xylanase pre-treatment (also observed in this study) is primarily caused by HexA removal with solubilized HexA-carrying xylooligosaccharide fractions, such as aldohexa- and aldopentahexenuronic acids (Xyl5-HexA and Xyl4-HexA),45 underlining the important role of HexA in cellulose pulp bleaching with xylanases.

As shown in Table 1, xylanase pre-treatment has significantly improved (boosted) the following POM-catalyzed ozone bleaching of kraft pulp in solvent media. Considering the reported impact of HexA on the standard (permanganate) method of kappa number determination (commonly used for the estimation of residual lignin content in chemical pulp), the calculated values of the actual kappa number, corrected to the contribution of HexA,46 were used to evaluate the bleach boosting effects caused by xylanase pre-treatment. Additional lignin removal by 45.3%, with improvement in brightness by 11.7%, was observed after catalytic ozonation of enzymatically pre-treated pulp samples in comparison with control (with no enzyme) treatments. The intrinsic viscosity of the pre-treated pulp samples was improved after ozonation by 10 points over the control. Accounting for changes in the pulp properties caused by the control experiments, xylanase pre-treatment led to a significant increase in selectivity (by ca. 60%) and efficiency (by ca. 40% and 60% for the efficiency of delignification and brightening, respectively) of the POM-catalyzed ozonation (Fig. 4), resulting in impressive overall bleaching improvement (by 216%, 167% and 137%, respectively) in comparison with conventional ozonation technology (denoted as control (water) in Fig. 4).

Besides promoting delignification, xylanase pre-treatment also caused some additional loss (by 5.6%) of xylan polysaccharide during the following POM-ozonation, as well as significant debranching of residual xylan in ozonated pulps, with elimination of ca. 33% Gal and ca. 30% HexA relative to the control (Table 1), as a result of improved fiber accessibility. Evidently, the intensive removal of HexA was one of the main reasons, along with increased delignification, for the remarkable improvement in brightness after POM-ozonation of pulp samples treated with xylanases. No effect of xylanase pre-treatment on cellulose degradation (loss) during POM-catalyzed ozonation was revealed.

The SEM analysis of processed eucalypt kraft pulp samples (Fig. 5) showed substantial modification of cellulose fiber surfaces during xylanase treatment. In contrast to the smooth and homogeneous cellulose fibers of the initial (untreated) pulp samples (Fig. 5a), the fibers more heterogeneous, having peeled surfaces with small filaments and signs of fibrillation were observed in the pulp samples subjected to the action of xylanases (Fig. 5b). This change in surface properties made the cellulose fibers more permeable (open) to POM catalyst and ozone at a subsequent bleaching stage, resulting in significant intensification (acceleration) of the main chemical reactions taking place during POM-catalyzed ozone bleaching and described by eqn (5) and (6) (i.e., oxidative degradation of lignin by POM and POM recovery by ozone, respectively), thereby greatly improving (boosting) POM-catalyzed ozonation as a whole. The catalytic ozonation itself did not caused any essential change in fiber surface morphology in comparison with xylanase-treated fiber (Fig. 5c), thereby confirming the high protective effect of the developed ozonation system.

image file: c7gc02218b-f5.tif
Fig. 5 SEM images of cellulose fibers from initial (unbleached) industrial eucalypt (E. globulus) kraft pulp (a), from the same pulp after xylanase pre-treatment (b) and subsequent POM-catalyzed ozonation in aqueous solvent solution (c).

5.2. TCF bleaching with an integrated POM-catalyzed ozonation stage

The integration of a solvent-assisted POM/O3 stage into the short TCF bio-bleaching sequences has been studied in the attempt to maximize the bleaching effect of POMs and further develop the quality parameters of bleached cellulose fibers, such as brightness and purity. Three model sequences were tested: (1) X-ZPOM-ER-(Q)P, (2) X-EP-ZPOM and (3) X-(Q)P-ZPOM, where X – xylanase pre-treatment, ZPOM – solvent assisted POM/O3 bleaching, Ep – alkaline extraction in the presence of hydrogen peroxide, ER – alkaline extraction in the presence of reducing agent, and P – hydrogen peroxide bleaching with preliminary pulp chelating (Q). The progress of bleaching was monitored through the change in the principal pulp properties (such as residual lignin content (as kappa number), brightness and intrinsic viscosity) with each bleaching stage as compared to the initial (untreated) eucalypt kraft pulp. As can be seen from Fig. 6, the alkaline extraction (ER) followed by the hydrogen peroxide bleaching (QP) of POM-ozonated (xylanase pre-treated) pulp (Sequence 1) caused a substantial improvement in the quality of the cellulose fibers, decreasing by ca. 49% the residual lignin content (up to 0.95 kappa number) and increasing by ca. 7% the cellulose brightness (up to 76.8% ISO). The use of the reducing agent (sodium borohydride) during alkaline extraction (to convert the alkali-sensitive carbonyl groups formed during ozonation to stable hydroxyl groups) and pulp chelating (Q) with EDTA before hydrogen peroxide bleaching (to limit the effect of transition metals on alkaline decomposition of peroxide) allowed minimizing the cellulose degradation during bleaching while maintaining the fairly high level of pulp intrinsic viscosity (910 mL g−1) after complete four-stage bleaching. The introduction of alkaline extraction (in the presence of peroxide, EP) between xylanase pre-treatment and POM-ozonation stage made it possible to reduce the bleaching sequence up to three stages (Sequence 2), but still improved the bleaching results compared to Sequence 1. The content of residual lignin in bleached pulps was decreased up to 0.88 kappa number, and the brightness was increased up to 78.4% ISO. The higher intrinsic viscosity of the pulp samples bleached by Sequence 2 (926 mL g−1) permitted improving the overall selectivity of the bleaching process by 5.1% in comparison with Sequence 1. As illustrated in Fig. 4, the most significant improvement of the bleaching results was achieved by three-stage bleaching Sequence 3, where hydrogen peroxide bleaching (QP) was inserted between the xylanase pre-treatment and POM/O3 bleaching stages. The significantly reduced content of lignin (up to 0.67 kappa number) and increased pulp viscosity (up to 969 mL g−1) have provided remarkable improvement of the overall bleaching selectivity by 15.6% and 21.5% in comparison with Sequences 2 and 1, respectively. The brightness of the bleached pulp after Sequence 3 overcame the limit of 80% ISO, approaching closely the level of so-called fully bleached pulp.
image file: c7gc02218b-f6.tif
Fig. 6 TCF bio-bleaching of industrial eucalypt (E. globulus) kraft pulp with integrated POM-catalyzed ozone stage: change in residual lignin content (as kappa number), brightness (% ISO) and intrinsic viscosity (mL g−1) of bleached pulp during each stage of X-ZPOM-ER-QP; X-EP-ZPOM and X-QP-ZPOM bleaching sequences (where X – xylanase pre-treatment; ZPOM – POM-catalyzed ozonation in solvent medium; Er – alkaline extraction with reducing agent; QP – hydrogen peroxide bleaching with preliminary pulp chelating; EP – alkaline extraction with hydrogen peroxide).

The high selectivity of bio-bleaching sequences with incorporated POM/O3 stage was confirmed by carbohydrate analysis of the bleached pulp samples. As seen from Table 2, despite the enhanced intensity of the oxidative delignification reactions, only moderate carbohydrate degradation occurred during overall bleaching, leading to loss of ca. 18–24% of the total pulp xylan (ca. 21–27% Xyl) and 46–59% of the associated Gal. The loss of only 1.8% of total cellulose (ca. 2% Glc) was detected after complete bleaching, independently the sequence applied. The lower carbohydrate losses observed in bleaching Sequences 2 and 3 contributed to their better selectivity towards lignin degradation reactions as compared with Sequence 1.

Table 2 Change in carbohydrate composition (% on dry initial pulp) of xylanase pre-treated eucalypt kraft pulp during short TCF bleaching sequences with integrated POM-catalyzed ozonation stage
  Initial pulp Enz. Xa Sequence 1 Sequence 2 Sequence 3
ZPOMb ER[thin space (1/6-em)]c QPd Loss EP[thin space (1/6-em)]e ZPOM Lossf QP ZPOM Loss
a Enzymatic pre-treatment (X-stage): pH 7.0; 65 °C; 180 min. b POM-catalyzed ozonation (ZPOM-stage): 0.8% O3; 1.0 mM HPA-5; 40% (w/w) acetone; pH 2. c Alkaline extraction with reducing agent (ER-stage): 1% NaOH; 0.1% NaBH4; 60 °C; 60 min. d Hydrogen peroxide bleaching (P-stage: 3% H2O2; 1.5% NaOH; 0,3% MgSO4; 0,2% DTPA; 90 °C; 4.5 h) with preliminary pulp chelating (Q-stage: 0,3% EDTA; pH 4.5; 50 °C; 30 min). e Alkaline extraction with hydrogen peroxide (EP-stage): 2.0% NaOH; 0.5% H2O2; 70 °C; 180 min. f Total component loss (%) as compared with initial unbleached pulp.
Glc 82.53 82.33 81.66 81.01 80.70 2.2 81.97 80.92 2.0 81.90 80.98 1.9
Xyl 16.16 14.93 13.65 12.42 11.74 27.3 13.15 12.69 21.5 13.55 12.79 20.8
Gal 0.44 0.36 0.24 0.20 0.18 58.9 0.30 0.23 47.7 0.29 0.24 45.9

Thus, even a very short (in comparison with industrial analogs) eco-friendly TCF bio-bleaching sequence consisting only of the enzymatic pre-treatment stage, POM-catalyzed ozonation and one additional chemical bleaching stage, without any oxygen pre-delignification commonly used to remove the main portion of pulp lignin, can provide the quality parameters of the bleached cellulose fibers required for the high tech applications. It is important to note that, regardless of the bleaching sequence used, the POM-ozonation was responsible for removal of 72–80% of the total lignin pulp, pointing to the critical importance of this stage for the final development of the properties of bleached cellulose fibers. The proven effective POMs re-oxidation (recovery) by ozone during the delignification process30 opens the possibility of developing the totally effluent-free and zero-waste technology of POM-catalyzed pulp ozonation, with continuous recirculation of the liquid process streams (catalyst and solvent) in a close-loop mode scheme, after separation from cellulose fiber, and with carbon dioxide and water as the only byproducts after deep lignin oxidation. This will exclude (or minimize) any catalyst and solvent losses, making the POM/O3 bleaching stage, as well as the entire bleaching process with integrated POM-catalyzed ozonation, an economically sustainable and ecologically attractive technology.

6. Conclusions

The properties of crude cellulose fibers, such as unbleached chemical pulps, isolated from various plant sources, can be effectively upgraded by solvent-assisted ozonation in the presence of Keggin-type heteropolyoxometalates (POMs) as selective catalysts of the delignification reaction. POM oxidative catalysis can be substantially intensified (boosted) by fibers (pulps) pre-treatment with highly specific hemicellulolytic enzymes, such as xylanases, leading to further significant improvement in fiber qualities. By a combination of POM-catalyzed ozonation with other non-chlorine oxygen-centered oxidative reagents, such as hydrogen peroxide, within the short TCF bio-bleaching (with enzymatic pre-treatment) sequences, the standard properties of cellulose fibers required for high tech applications can be easily achieved in an effective and environmentally friendly way.

Conflicts of interest

There are no conflicts to declare.


CEF is a research unit funded by Fundação para a Ciência e a Tecnologia I.P. (FCT), Portugal (UID/AGR/00239/2013). The financial support from FCT, Portugal within research contract SFRH/BPD/112207/2015 is gratefully acknowledged.


  1. A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney, C. A. Eckert, W. J. Frederick Jr., J. P. Hallett, D. J. Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer and T. Tschaplinski, Science, 2006, 311, 484–489 CrossRef CAS PubMed.
  2. B. Kamm and M. Kamm, Adv. Biochem. Eng./Biotechnol., 2007, 105, 175–204 CrossRef CAS PubMed.
  3. M. FitzPatrick, P. Champagne, M. F. Cunningham and R. A. Whitney, Bioresour. Technol., 2010, 101, 8915–8922 CrossRef CAS PubMed.
  4. C. Somerville, H. Youngs, C. Taylor, S. C. Davis and S. P. Long, Science, 2010, 329, 790–791 CrossRef CAS PubMed.
  5. H. Sixta, Handbook of Pulp, Wiley-VCH Verlag, GmbH&Co. KGaA, Weinheim, 2006 Search PubMed.
  6. M. Ek, Pulping Chemistry and Technology, De Gruyter, Berlin, Germany, 2016 Search PubMed.
  7. H. U. Suess, Pulp Bleaching Today, De Gruyter, Berlin, Germany, 2010 Search PubMed.
  8. P. Bajpai, Environmentally benign approaches for pulp bleaching, Elsevier, 2nd edn, 2012 Search PubMed.
  9. B. Van Lierop, A. Skothos and N. Liebergott, in Pulp Bleaching. Principles and Practice, ed. C. W. Dence and D. W. Reeve, TAPPI Press, Atlanta, 1996, pp. 321–345 Search PubMed.
  10. R. C. Rice and A. Netzer, Handbook of Ozone Technology and Application, Ann Arbor Science Publishers, Ann Arbor, 1982 Search PubMed.
  11. M. T. Pope, in Handbook on the Physics and Chemistry of Rare Earths, ed. K. A. Gschneidner Jr., J. C. Bunzli and V. K. Pecharsky, Elsevier, 2007, vol. 38, ch. 240, pp. 337–382 Search PubMed.
  12. I. V. Kozhevnikov, Catalysis by polyoxometalates, Wiley and Sons, Chichester, England, 2002 Search PubMed.
  13. A. A. Shatalov, in Polyoxometalates: Properties, Structure and Synthesis, ed. A. P. Roberts, Nova Science Publishers, INC., New York, 2016, ch. 4, pp. 77–114 Search PubMed.
  14. A. I. Cañas and S. Camarero, Biotechnol. Adv., 2010, 28, 694–705 CrossRef PubMed.
  15. P. Bajpai, Xylanolytic Enzymes, Academic Press, 2014 Search PubMed.
  16. M. J. Bailey, P. Biely and K. Poutanen, J. Biotechnol., 1992, 23, 257–270 CrossRef CAS.
  17. K. I. Matveev, N. B. Shitova, Z. P. Pai, V. F. Odyakov, O. K. Akmalova, L. I. Kuznetsova, T. A. Basalaeva, A. B. Rumiantsev and L. N. Shadrin, SU Pat, 421226 Int.C1.B01J 37/04, 1972 Search PubMed.
  18. M. T. Pope and A. Muller, Polyoxometalate chemistry from topology via self-assembly to applications, Kluwer Academic Publishers, Dordrecht, 2001 Search PubMed.
  19. C. L. Hill, in Comprehensive Coordination Chemistry. Part II: From Biology to Nanotechnology, ed. A. G. Wedd, Elsevier, Oxford, UK, 2004, vol. 4, pp. 679–759 Search PubMed.
  20. R. Neumann and A. M. Khenkin, Chem. Commun., 2006, 24, 2529–2538 RSC.
  21. C. L. Hill, J. Mol. Catal. A: Chem., 2007, 262, 2–6 CrossRef CAS.
  22. I. V. Kozhevnikov, J. Mol. Catal., 1997, 117, 151–158 CrossRef CAS.
  23. I. V. Kozhevnikov, Chem. Rev., 1998, 98, 171–198 CrossRef CAS PubMed.
  24. O. A. Kholdeeva, A. V. Golovin, R. I. Maksimovskaya and I. V. Kozhevnikov, J. Mol. Catal., 1992, 75, 235–244 CrossRef CAS.
  25. M. Lissel, H. J. Wal and R. Neumann, Tetrahedron Lett., 1992, 33, 1795–1798 CrossRef CAS.
  26. A. R. Gaspar, J. A. F. Gamelas, D. V. Evtuguin and C. P. Neto, Green Chem., 2007, 9, 717–730 RSC.
  27. A. A. Shatalov, D. V. Evtuguin and C. Pascoal Neto, Carbohydr. Polym., 2000, 43, 23–32 CrossRef CAS.
  28. A. A. Shatalov and H. Pereira, Chem. Eng. J., 2009, 155, 380–387 CrossRef CAS.
  29. A. A. Shatalov and H. Pereira, Bioresour. Technol., 2010, 101, 4616–4621 CrossRef CAS PubMed.
  30. A. A. Shatalov and H. Pereira, Bioresour. Technol., 2010, 101, 9330–9334 CrossRef CAS PubMed.
  31. A. K. Bin, Ozone: Sci. Eng., 2006, 28, 67–75 CrossRef CAS.
  32. E. Cogo, J. Albert, G. Malmary, C. Coste and J. Molinier, Chem. Eng. J., 1999, 73, 23–28 CrossRef CAS.
  33. J. Quesada, M. Rubio and D. Gomes, J. Appl. Polym. Sci., 1998, 68, 1867–1876 CrossRef CAS.
  34. C. Schuerch, J. Am. Chem. Soc., 1952, 74, 5061–5067 CrossRef CAS.
  35. E. G. Zhizhina, V. F. Odyakov, M. V. Simonova and K. I. Matveev, React. Kinet. Catal. Lett., 2003, 78, 373–379 CrossRef CAS.
  36. I. V. Kozhevnikov, Yu. V. Burov and K. I. Matveev, Izv. Akad. Nauk SSSR, Ser. Khim., 1981, 11, 2428–2435 Search PubMed.
  37. A. Kantelinen, B. Hortling, J. Sundquist, M. Linko and L. Viikari, Holzforschung, 1993, 47, 318–324 CrossRef CAS.
  38. A. A. Shatalov and H. Pereira, Ind. Crops Prod., 2007, 25, 248–256 CrossRef CAS.
  39. A. A. Shatalov and H. Pereira, Biochem. Eng. J., 2008, 40, 19–26 CrossRef CAS.
  40. A. A. Shatalov, D. V. Evtuguin and C. P. Neto, Carbohydr. Res., 1999, 320, 93–99 CrossRef CAS PubMed.
  41. A. A. Shatalov and H. Pereira, in 20-th International Carbohydrate Symposium, Proceedings, Hamburg, Germany, 2000, p. 370.
  42. J. Buchert, A. Teleman, V. Harjunpää, M. Tenkanen, L. Viikari and T. Vuorinen, Tappi J., 1995, 78, 125–130 CAS.
  43. A. A. Shatalov and H. Pereira, Cellulose, 2004, 11, 109–117 CrossRef CAS.
  44. Z. H. Jiang, B. Van Lierop and R. Berry, Tappi J., 2000, 83, 167–175 CAS.
  45. A. A. Shatalov and H. Pereira, Bioresour. Technol., 2009, 100, 3069–3075 CrossRef CAS PubMed.
  46. G. Gellerstedt and J. Li, Carbohydr. Res., 1996, 294, 41–51 CrossRef CAS PubMed.

This journal is © The Royal Society of Chemistry 2017