ECF Bleaching with Molybdenum Activated Acid Peroxide and Its Impact on Eucalyptus Pulp Properties and Effluent Quality

Acid hydrogen peroxide catalyzed with molybdenum (PMo stage) is effective to decrease pulp kappa number and potentially minimize chlorine dioxide demand in subsequent ECF bleaching. This study aimed at developing cost-effective methods for applying PMo as the first stage bleaching for eucalyptus kraft pulp and evaluating their impact on pulp properties and effluent quality in relation to current ECF bleaching techniques. The PMo(EP)DP and A/PMo(EP)DP sequences proved to be most cost-effective for PMo stage application. The pulp properties and effluent quality derived from bleaching with these two sequences were compared with those of the conventional bleaching with the DHT(EP)DP sequence. The PMo(EP)DP sequence produced pulp of refinability and physical properties similar to that of the reference but the A/PMo(EP)DP one resulted in pulp of 20% lower refinability and 10% lower tearing strength. Untreated effluents of the PMo(EP)DP and A/PMo(EP)DP sequences presented similar COD, but lower AOX and color than the DHT(EP)DP effluents. None of the effluents exhibited acute toxicity. The effluents from the PMo(EP)DP sequence showed much lower BOD and BOD/COD ratio than that of the A/PMo(EP)DP one. All effluents were readily treated in bench-scale reactors, and achieved COD removal efficiencies of greater than 70%.


Introduction
There has been a growing demand for light Elemental Chlorine Free (light-ECF) pulps, particularly in Europe, in order to replace the very expensive Totally Chlorine Free (TCF) ones.There is no clear definition for the so-called light-ECF pulps, but the market tends to accept under this denomination the bleached pulps that contain less than 50-60 g Cl -/adt pulp of organically bound chlorine substance (OX), as measured by a standard test method such as SCAN-CM 52:94, PTS-RH 012/90, ISO 11480, etc [1].Such a low bound chlorine target requires that ECF bleaching be implemented under unusual operational conditions, i.e., with very low chlorine dioxide charges [2].Many bleaching alternatives have been evaluated to minimize chlorine dioxide use for eucalyptus kraft pulps, one of them being the so-called molybdenum activated hydrogen peroxide stage (PMo).
The use of siliconmolybdate activated peroxide blea-ching has been very thoroughly investigated for bleaching or Scandinavian kraft pulp [3].The great advantage of this stage in relation to other stages run under acidic conditions is the minimum formation of oxalic acid that can cause scaling in thee bleach plant.Oxalic acid is produced through acid hydrolysis that occurs when an acid stage (e.g.ozone and chlorine dioxide) is used to remove pulp hexenuronic acids [4].In relation to potential molybdenum residuals in the bleaching filtrate [3] have shown that the amount is sufficiently low in order to cause significant damages to the effluent treatment system.On the other hand, no effect in the chemical recovery system was seen due to the low amount of molybdenum used in the reaction.In addition, part of the filtrate can be cycled back to the PMo stage thus recovering significant amounts of the molybdenum catalyst [5].
The acid peroxide bleaching activated with heavy metal salts was investigated firstly by [6] and subsequently by [7] who evaluated its reactivity towards several Properties and Effluent Quality lignin models compounds.The main conclusions were that molybdenum salts are more effective than the tungsten ones in the pH range of 2-5.Furthermore, the lignin models containing a free phenolic hydroxyl group were more reactive than those without these groups.The delignification action during the molybdenum catalyzed acid peroxide bleaching (PMo stage) was ascribed to the ability of the molybdate to form together with the hydrogen peroxide the diperoxi-molybdenum complex.Several different molybdenum-containing species can serve as catalysts, including sodium and ammonium molybdate, molybdenum trioxide, siliconmolybdate, etc.
Reference [1] has studied the use sodium and ammonium molybdate activated peroxide bleaching for eucalyptus pulps and determined that optimal conditions for such treatment include temperature of 90-95 o C for 2 hours, pH 3.5 and molybdenum and hydrogen peroxide charges of 0.1 kg/t and 5 kg/t, respectively.They also concluded that the kappa number reduction in the PMo stage can reach 7 units and occurs mainly through hexenuronic acids removal (HexA's) from the pulp.[8] Evaluated the sodium molybdate activated peroxide stage for bleaching of an oxygen delignified eucalyptus kraft pulp (high HexA content) to 90% ISO brightness in short ECF (Elemental Chlorine Free) and TCF (Totally Chlorine Free) sequences: PMoDP, PMoD/P, PMoD(PO), PmoQ (PO) and XPMoQ (PO.Among the ECF alternatives, the three-stage PMoDP sequence proved attractive particularly in regard to effluent quality.However, a four stage sequence-XPMoQ(PO)-was required to achieve full brightness using the TCF technology.
It is apparent that the PMo stage is an alternative for bleaching eucalyptus pulps, particularly with the focus of shortening the current bleach plants and recycling part of the bleach plant filtrate.However, many questions still remain in regard to PMo stage application for eucalyptus pulps.For example, what types of sequences are more suited for such stage?Would the PMo stage work well in any type of eucalyptus pulp or only on those rich in hexenuronic acids?What is the impact of such stage on eucalyptus pulp properties and effluent quality and treatability?Some of these questions are addressed in this study that embraces the following objectives: (1) evaluating PMo stage in various sequences for bleaching eucalyptus pulp of two different origins (high/low HexA); (2) determining the impact of using the PMo stage on bleaching effluent load and treatability and on pulp refinability and physical-mechanical properties.

Bleaching Experiments
Two eucalyptus pulp samples were collected in the last washer after the oxygen delignification stage of two different bleach plants.The main specifications of these two pulp samples are presented in Table 1.These samples were designated as easy-and difficult-to-bleach pulps on the basis of their chemical requirements to achieve 90% ISO brightness with the D(EP)DP sequence.Except when otherwise stated, pro analysis reagents were used throughout the study.The (NH 4 ) 6 Mo 7 O 24 .4H 2 O salt was used as the source of molybdenum catalyst and was purchased from the Aldrich Chemical company.
The molybdenum activated acid peroxide stage was optimized taking into account the temperature (85 and 95 o C), the end pH (3.0 and 4.5) and the molybdenum dose (0.10, 0.15 and 0.25 kg/odt pulp).The reaction time (2 h), reaction consistency (10%) and hydrogen peroxide dose (5 kg/odt pulp) were kept constant throughout the study.
The PMo stage optimized according to the conditions above was applied to the PMo(EP)DP, A/PMo(EP)DP, PMo/Q(PO) and PMoZ/ED bleaching sequences.The D(EP)DP, D HT (EP)DP, A/D(EP)DP, D/Q(PO) and Z/ED-P sequences were used as references.Oxidant charges were optimized in the D 1 and (PO) stages in order to achieve 90% ISO final brightness.
Hot chlorine dioxide (D HT ), pressurize peroxide (PO), hot acid hydrolysis (A) and ozone (Z) bleaching stages were carried out in a model Mark V mixer/reactor (Quantum Technologies Inc.).Conventional alkaline extraction (E), extraction with peroxide (EP), peroxide (P), chlorine dioxide (D) and molybdenum activated acid peroxide bleaching (PMo) and chelation (Q) stages were carried out in polyethylene bags.The main reaction conditions used are presented in Table 2.After each bleaching stage, run in duplicate, the samples were washed with excess distilled water, except in those cases where the washing step was deliberately omitted.Reagent doses are expressed in kg per oven dried ton of pulp (odt).Concentrations of oxidants and residuals were measured according to [9].
Pulp kappa number, viscosity, brightness and brightness stability values were measured according to TAPPI standard procedures.Hexenuronic acids were measured by UV after their conversion into furoic acid using acid hydrolysis with formic acid [10].Pulp OX and filtrate AOX values were measured in an absorbable organic halogen analyzer (ECS 1600-Euroglas), according to SCAN procedure.Filtrate total organic carbon (TOC) values were measured directly in a Shimadizu model 5000A TOC analyzer.Bleached pulp samples were refined in a PFI mill at 0, 1000, 2000 and 3000 revolutions and tested for physical-mechanical properties according to Tappi standard procedures.Bleaching chemical costs After each bleaching stage of selected sequences filtrates were collected.The filtrates were mixed together and made into the combined effluent of the sequence.The mixed effluent was characterized for the chemical oxygen demand (COD), biochemical oxygen demand (BOD), adsorbable organic halogen (AOX) and color, according to standard procedures [11].Effluent toxicity was determined using the short-term Daphnia similis acute toxicity assay, as described by the [12].

Optimization of the Molybdenum Activated
Peroxide Stage (PMo)

Difficult-to-Bleach Pulp
The kappa number results (Figure 1) show that both PMo and A/PMo stages delignification efficiency are slightly better at pH 3 than at pH 4.5.But, increasing molybdenum dose from 0.15 to 0.25 kg/odt has little effect on efficiency.In fact, increasing the molybdenum dose at pH 4.5 resulted in decreased delignification efficiency.Raising the PMo stage temperature form 85 to 95 o C positively affected the efficiency.The A/PMo stage showed higher impact on pulp kappa number than the PMo one likely due to significant HexA removal in the A/part of the A/PMo stage.The kappa number decrease with lowering pH and increasing temperature is explained by more efficient removal of pulp HexA under Properties and Effluent Quality these conditions.It is well known that a significant fraction of the kappa removal in the PMo stage is due to HexA elimination [7].Since molybdenum functions as catalysts for the acid peroxide reaction, increasing dose above a certain limit has seemingly no effect on kappa number.However, the negative effect of molybdenum dose on delignification efficiency is of difficult explanation.Pulp viscosity (Figure 2) behaved similarly to kappa number, i.e., the lower the kappa number after the PMo or A/PMo stages the lower the viscosity.Highest viscosities were achieved with the PMo stage run at 85 o C whereas lowest values were seen for the A/PMo stage.Brightness (Figure 3), on the other hand, showed a peculiar behavior.The highest values were measured on the pulps treated with the PMo stage run at 95 o C. In spite of its high delignification efficiency the A/PMo stage did not bleach the pulp accordingly.In fact, the brightness  The results previously discussed do not allow for a clear definition of the most appropriate condition to run the PMo and A/PMo stages.As far as delignification efficiency and brightness gains, the dose of 0.15 kg Mo/odt pulp is certainly sufficient and the reaction at pH 3 at 95 o C is more efficient.However, the opposite is observed for pulp viscosity.On the other hand, running the PMo stage at 95 o C is more costly than at 85 o C, with the gains due to the high temperature being questionable.

Easy-to-Bleach Pulp
The results presented in Figure 4 corroborate those previously found for the difficult-to-bleach pulp (Figure 1) that better delignification efficiency is achieved at pH 3 than at pH 4.5.On the other hand, increasing molybdenum catalyst dose from 0.10 to 0.15 kg/odt pulp has no significant impact on delignification efficiency but negatively affect viscosity, particularly for the A/PMo treatment (Figure 5).The values of brightness tended to be higher for the A/PMo than for the PMo treated pulp (Figure 6).The catalyst dose did not affect brightness substantially.Therefore, for the easy-to-bleach pulp sample, a molybdenum catalyst dose of 0.10 kg Mo/odt pulp should be sufficient.This result is in agreement with those of [7].

Comparison of PMo and A/PMo Stages Performances on Difficult-and Easy-to-Bleach Pulps
A comparison of the PMo and A/PMo stage perform-Properties and Effluent Quality   ances for the so-called difficult-and easy-to-bleach pulps, under similar operational conditions is shown in Table 3.
Contrarily to the anticipated, it was observed that both PMo and A/PMo stage delignification efficiencies, as measured by the kappa drop and brightness gain across the stages, are higher for the difficult-than for the easy-to-bleach pulp sample.These results indicate that molybdenum activated peroxide stages are efficient to deal with lignin structures that are normally difficult to remove from pulps using conventional ECF bleaching.Note that the definition of difficult-and easy-to-bleach was based on the chemical requirements to bleach the pulp to 90% ISO with the D(EP)DP sequence.

Difficult-to-Bleach Pulp
The

Easy-to-Bleach Pulp
For the easy-to-bleach pulp, the following sequences were evaluated: D(EP)DP, D HT (EP)DP, A/D(EP)DP, Z/EDP, PMo(EP)DP, A/PMo(EP)DP and PMoZ/ED (Table 5).The sequences containing the PMo and A/PMo stages showed higher total active chlorine demand and relative chemical costs in relation to their reference counterparts.Overall the viscosities of the pulps treated with sequences containing PMo and A/PMo stages were lower.The sequences PMo(EP)DP and A/ PMo(EP)DP produced pulp of lower brightness reversion in relation to their references, but reversion was high when the PMo stage was applied in the PMoZ/ED sequence.relative bleaching chemical costs for selected sequences applied to the difficult and easy-to-bleach pulps.Reference sequences as well as those containing PMo and A/PMo stages are presented.The values clearly indicate that the easy-to-bleach pulp respond significantly better to the selected bleaching sequences than the difficult-to-bleach one.Since the two pulps behaved inversely than expected in the PMo and A/PMo stages alone (Table 3), it is concluded that the significant difference in bleachability between the difficult-and easy-Properties and Effluent Quality to-bleach pulps is in fact due to differences in their reactivity towards chlorine dioxide bleaching.

Refinability and Physical Properties of Easy-to-Bleach Pulp Sample
Considering the results discussed in the previous session, the A/PMo(EP)DP, PMo(EP)DP and D HT (EP)DP (reference) sequences were selected to produce pulp for refinability and physical property evaluations.Only the easy-to-bleach pulp sample was investigated.For the drainage degrees of interest (20 and 35°SR), no significant differences in refinability between the pulps bleached with the D HT (EP)DP and PMo(EP)DP sequences were seen.On the other hand, the pulp bleached with the A/PMo(EP)DP sequence was more difficult to refine since it required more energy (~20%) to achieve a given drainage degree (Figure 7).The latter sequence contains a hot acid hydrolysis stage which may have dissolved some of the pulp hemicelluloses during bleaching and that negatively affected refinability.Eucalyptus pulps are usually commercialized for printing & writing (P & W) and tissue paper grades.These two types of products require different levels of refining, with the tissue grade pulps requiring very low beating (20-25°SR).On the other hand, the drainage degree required for manufacturing P & W paper grades is usually in the range of 30-40°SR.Thus the 20 and 35°SR values were selected for property evaluations aimed at tissue and P&W paper grades.
The results presented in Table 7 show that in regard to tensile related properties, including tensile index, tensile energy adsorption, modulus of elasticity and stretch there were no significant differences among the three bleached pulps both for tissue and printing and writing paper grades, indicating no effect of the PMo stage.At lower beating degree (tissue grade pulps) the tear index tended to be about 10% lower for the pulps bleached with the sequences containing the PMo stage.With more beating (P&W grade pulp) the PMo(EP)DP pulp recovered the tear strength showing values similar to the that of the reference pulp but the pulp bleached with the A/PMo-(EP)DP sequence still presented a 10% lower tear index at 35°SR.The lower tear strength of the pulp bleached with the A/PMo(EP)DP sequence is in line with its low viscosity at the end of bleaching.The A/PMo(EP)DP sequence produced pulp of significantly higher air resistance in relation to the reference at the high 35°SR but significantly lower at 20°SR.This result is explained by the potential of this pulp to form fines during beating due to its lower integrity (low viscosity).The PMo(EP)DP bleached pulp did not show this trend.Although its air resistance was lower than that of the reference at 20 o SR, it did not increase so much with prolonged beating, indicating better fiber integrity.No significant differences were observed in bulk among the various pulps both at 20 and 35°SR.

Effluent Load and Treatability of Easy-to-Bleach Pulp
On the basis of the results discussed in the previous ses-  sion, the A/PMo(EP)DP, PMo(EP)DP and D HT (EP)DP (reference) sequences were selected to produce effluent for load and treatability evaluations.Only the easyto-bleach pulp sample was investigated.Combined effluent samples were prepared by mixing equal amounts of filtrates from each bleaching stage of the sequence.Characterization of the raw effluents, expressed as both concentration (mg/L) and load (kg/odt) is presented in Table 8.In general, incorporation of the molybdenum catalyzed acid peroxide stage improved raw effluent quality and resulted in lower effluent pollutant loads.Although COD loads were similar for all three effluents, the BOD load of the PMo(EP)DP sequence was much lower than that of the D HT (EP)DP and A/ PMo(EP)DP sequences.Therefore, the BOD/COD relationship, typically used to express effluent biodegradability and applicability of biological treatment processes, was lowest for the PMo(EP)DP effluent (0.29), indicating that the organic matter in this effluent was in a less readily biodegradable form than that dissolved in the other bleaching effluents evaluated.
Not surprisingly, the highest effluent AOX values were found in the D HT (EP)DP effluent, because of the use of chlorine dioxide in the first bleaching stage.Incorporation of the molybdenum catalyzed acid peroxide stage eliminates the possibility of formation of organochlorine compounds during the initial bleaching stages.It should be noted, however, that AOX formation is very low, even when the D HT stage was used.
Effluent color was greatly reduced by the use of the molybdenum catalyzed acid peroxide stage instead of the hot chlorine dioxide stage.The beneficial effect on effluent color was greatest when the peroxide stage was preceded by a hot acid stage (A/PMo).
None of the effluents evaluated exhibited acute toxicity to Daphnia similis, indicating that molybdenum used as catalyst was not dissolved in the effluents in toxic levels.Given the absence of toxicity in the raw effluents, toxicity was not evaluated in treated effluents, since it is generally accepted that biological treatment is capable of eliminating acute toxicity from bleaching effluents [13].
Biological treatment stability is presented in Figure 8. Treated effluent COD values rapidly stabilized and remained relatively constant after ten treatment cycles in the SBRs.Based on these results, more detailed effluent characterization was initiated on day 19, and continued until day 24.Results of the final effluent characterization are presented in Table 9.Average removal efficiencies over the course of the effluent characterization are presented in Table 10.
Final effluent quality measured as BOD and COD was not statistically different for the three effluents, while lower final effluent color and AOX were obtained in the effluents that incorporated the molybdenum catalyzed acid peroxide stage.The low biodegradability (BOD/C-OD ratio) in the raw PMo(EP)DP effluent was confirmed

Figure 1 .
Figure 1.Influence of pH and molybdenum charge on difficult-to-bleach pulp kappa number after PMo (85 o C), A/PMo (85 o C) and PMo (95 o C) stages carried out at the following constant conditions: 5 kg/t H 2 O 2 , 2 h and 10% consistency.Kappa in = 10.9.In the X axis, 0.15 pH 3 indicates reaction at pH 3 with 0.15 kgMo/odt pulp.The same apply to the other acronyms.

Figure 2 .
Figure 2. Influence of pH and molybdenum charge on difficult-to-bleach pulp viscosity after PMo (85 o C), A/PMo (85 o C) and PMo (95 o C) stages carried out at the following constant conditions: 5 kg/t H 2 O 2 , 2 h and 10% consistency.Kappa in = 10.9.

Figure 3 .
Figure 3. Influence of pH and molybdenum charge on difficult-to-bleach pulp brightness after PMo (85 o C), A/PMo (85 o C) and PMo (95 o C) stages carried out at the following constant conditions: 5 kg/t H 2 O 2 , 2 h and 10% consistency.Kappa in = 10.9. after A/PMo stage was in several cases lower than that of the PMo stage run at 85 o C. Pulp darkening during the A/ part of the A/PMo stage may explain the low brightness values.The results previously discussed do not allow for a clear definition of the most appropriate condition to run the PMo and A/PMo stages.As far as delignification efficiency and brightness gains, the dose of 0.15 kg Mo/odt pulp is certainly sufficient and the reaction at pH 3 at 95 o C is more efficient.However, the opposite is observed for pulp viscosity.On the other hand, running the PMo stage at 95 o C is more costly than at 85 o C, with the gains due to the high temperature being questionable.

Figure 4 .
Figure 4. Influence of pH and molybdenum charge on easyto-bleach pulp kappa number after PMo, and A/PMo stages carried out at the following constant conditions: 5 kg/t H 2 O 2 , 85 o C, 2 h and 10% consistency.Kappa in = 11.9.In the X axis, 0.1pH3 indicates reaction at pH 3 with 0.10 kgMo/odt pulp.The same apply to the other acronyms.

Figure 5 .
Figure 5. Influence of pH and molybdenum charge on easyto-bleach pulp viscosity after PMo, and A/PMo stages carried out at the following constant conditions: 5 kg/t H 2 O 2 , 85 o C, 2 h and 10% consistency.Kappa in = 11.9.

Figure 6 .
Figure 6.Influence of pH and molybdenum charge on easyto-bleach pulp brightness after PMo, and A/PMo stages carried out at the following constant conditions: 5 kg/t H 2 O 2 , 85 o C, 2 h and 10% consistency.Kappa in = 11.9.
sequences evaluated for the difficult-to-bleach pulp included PMo(EP)DP, A/PMo(EP)DP, PMo/Q(PO) and PMo(PO)D.References to these sequences included the D(EP)DP, D HT (EP)DP, A/D(EP)DP and D/Q(PO) ones.The results presented in Table4indicate that the target brightness of 90% ISO was achieved only with the sequences PMo(EP)DP and A/PMo(EP)DP.The total active chlorine demand and bleaching cost of the sequences containing the PMo and A/PMo stage are higher than their reference counterparts.On the other hand, their pulp brightness reversion and viscosity are lower than those of the references.The sequences PMo/Q(PO) and Pmo-(PO)D produced brightness in the range of 86-88% ISO at very high chemical doses.The 90% ISO brightness target could potentially have been achieved with these sequences, albeit at a very high cost.Among the reference bleaching processes, lowest bleaching costs were achieved with the D HT (EP)DP sequence.

Figure 7 .
Figure 7. Refinability curves of pulp samples bleached by different procedures.
were obtained by multiplying chemical doses (kg/odt pulp) by chemical prices (US$/kg chemical).Relative bleaching chemical costs were derived by considering.the lowest chemical costs as 100 and whatever higher or lower number as a percentage over this value.The following chemical prices (US$/kg of product at 100%) were used to calculate bleaching costs: