0 évaluation0% ont trouvé ce document utile (0 vote)
87 vues0 page
If applied under appropriate conditions, hydrogen peroxide is a very effective bleaching chemical. Hydrogen peroxyde decreases the demand for chlorine dioxide in the final bleaching stages of an ECF sequence. In 1995 the worldwide production of hydrogen peroxide was estimated to have reached more than 1,6 million metric tons.
If applied under appropriate conditions, hydrogen peroxide is a very effective bleaching chemical. Hydrogen peroxyde decreases the demand for chlorine dioxide in the final bleaching stages of an ECF sequence. In 1995 the worldwide production of hydrogen peroxide was estimated to have reached more than 1,6 million metric tons.
Droits d'auteur :
Attribution Non-Commercial (BY-NC)
Formats disponibles
Téléchargez comme PDF, TXT ou lisez en ligne sur Scribd
If applied under appropriate conditions, hydrogen peroxide is a very effective bleaching chemical. Hydrogen peroxyde decreases the demand for chlorine dioxide in the final bleaching stages of an ECF sequence. In 1995 the worldwide production of hydrogen peroxide was estimated to have reached more than 1,6 million metric tons.
Droits d'auteur :
Attribution Non-Commercial (BY-NC)
Formats disponibles
Téléchargez comme PDF, TXT ou lisez en ligne sur Scribd
Degussa AG HYDROGEN PEROXI DE I N HYDROGEN PEROXI DE I N CHEMI CAL PULP BLEACHI NG CHEMI CAL PULP BLEACHI NG -an overview Paper for the ABTCP meeting on pulp bleaching at Vitoria, Espirito Santo, Brasil, June 25. - 28, 1996 1. Abstract If applied under appropriate conditions, hydrogen peroxide is a very effective bleaching chemical. The very wide application range will be described, starting with boosting the brightness in an existing bleach plant in any extraction stage, as well as in an additional final treatment in the high density storage tower. In addition hydrogen peroxide decreases the demand for chlorine dioxide in the final bleaching stages of an ECF sequence. Significant savings are the result of peroxide addition in the second extrac- tion stage. The replacement of one chlorine dioxide stage in D 1 -D 2 final bleaching with peroxide and the application of a D 1 -P treatment also cuts the active chlorine demand. With hydrogen peroxide the kappa number in the first extraction stage can be lowered and the substitution of chlorine with chlorine dioxide made more efficient. In TCF sequences hydrogen peroxide is the chemical of choice to gain the required brightness. The conditions of peroxide bleaching are described and the benefit of magnesium salt addition and buffering with very small amounts of sodium silicate is dem- onstrated. In addition the effect of alkaline hydrogen peroxide on the corrosion of titanium is reviewed and underlined with laboratory data. Basically, corrosion does not take place under the typical reaction conditions for peroxide bleaching in an ECF sequence. 2. Introduction In 1995 the worldwide production of hydrogen per- oxide was estimated to have reached more than 1,6 million metric tons. The dominating amount of hy- drogen peroxide is used in bleaching processes. Hydrogen peroxide is generally supplied as an aqueous solution, typically in concentrations between 35% by weight and up to 70% by weight. These acidic solutions of hydrogen peroxide in water are very stable. Hydrogen peroxide can be stored for months in stainless steel tanks without significant changes of the content. Compared to water, the energy content of hydrogen peroxide is much higher. For water the heat of for- mation (H, eq. 1) from the elements is as low as - 286 kJ mol -1 , for hydrogen peroxide (eq. 2) the co- responding value is only -188 kJ mol -1 [1]. In conse- quence hydrogen peroxide is less stable and can disproportionate into water and oxygen. (1) H 2 + O 2 H 2 O H = -286 kJ mol -1 (2) H 2 + O 2 H 2 O 2 H = -188 kJ mol -1 Since the activation energy for the cleavage of the oxygen-oxygen bond is rather low (H = -71 kJ mol -1 ) [2], traces of contaminants can start this reaction. Basically the decomposition is a redox process. Hydrogen peroxide either supplies electrons and yields oxygen or accepts electrons and yields water. Metal salts of each state of oxidation can start the decomposition reaction. The first step can be the reduction of iron according to equation (3): (3) 2 Me 2+ + H 2 O 2 2 Me + + O 2 + 2 H + The alternative is the oxidation of a metal according to equation (4): (4) 2 Me + + H 2 O 2 + 2 H + 2 Me 2+ + 2 H 2 O The reaction certainly can also start with the reduced form of metal. The overall reaction is identical, it is the formation of water and oxygen from hydrogen peroxide and the redox system of the metal is acting as the catalyst [1]. The decomposition of hydrogen peroxide is in addition catalysed by alkali. These are the reaction steps: (5) H 2 O 2 + OH - H 2 O + HOO - (6) HOO - + H 2 O 2 H 2 O + O 2 + OH - Since bleaching with hydrogen peroxide requires alkaline conditions, this decomposition reaction is very important for the practice. Single electron transfer reactions of hydrogen per- oxide with catalysts yield radicals. These decom- position reactions take place with metals or enzymes like catalase. Radicals formation can also be the result of a thermal cleavage of the oxygen-oxygen bond. (7) H 2 O 2 + Me + OH - + . OH + Me 2+ (8) H 2 O 2 + . OH H 2 O + . OOH (9) H 2 O + . OOH . OO - + H 3 O + 2 The hydroxyl radical, the hydroperoxy radical and the superoxide anion radical are important intermediates. They cause side reactions in bleaching processes, with delignification as positive and depolymerisation of the cellulose as negative result. Generally radicals produce more negative effects than positive results on delignification. Therefore, transition metal ions are normally eliminated before a peroxide treatment. In pulp bleaching the formation of the perhydroxyl anion (eq. 5), a nucleophile intermediate, is re- sponsible for the oxidation of chromophores in lignin through the cleavage of side chains [3]. The effect of a hydrogen peroxide treatment is dominantly an increase of the brightness. Delignification with hydrogen peroxide is to a large extent the result of the action of the radicals produced in equations 7, 8 and 9 [3, 4]. At moderate temperature, under buffered conditions and in the absence of transition metals, the delignifying effect of hydrogen peroxide is limited. The degradation of polymerized lignin, which can be the result of high intensity pulping conditions, with hydrogen peroxide is limited. The perhydroxyl anion, being a nucleophile, cannot attack the electron rich aromatic rings of the residual lignin. 3. Application of hydrogen peroxide on mill scale Bleaching reactions with hydrogen peroxide are conducted in stainless steel reactors or in rubber coated carbon steel towers, to avoid decomposition of the peroxide and corrosion of the reactors. Alkaline peroxide solutions are not corroding stainless steel. Peroxide solutions are diluted and premixed with caustic soda before they are added to the pulp. The alternative is the addition of the technical grade solution of hydrogen peroxide (50% to 70%) to the pulp without dilution, which requires an efficient mixer in order to avoid local overconcentration. The advantage is the addition of a lower volume of diluting liquid which permits a higher consistency. Generally the consistency in a peroxide bleaching stage should be as high as possible. The increase of the relative concentration of the chemicals to the pulp and the elimination of dissolved organic compounds improves the efficency of the bleaching. Very alkaline peroxide solutions are agressive against another material which is widely used to in the construction of bleach plants: titanium. The corrosion of titanium through alkaline peroxide solutions is strongly depending on the peroxide concentration and the pH value. Table 1 and figure 1 have the results of laboratory tests with titanium metal treated with alkaline hydrogen peroxide at 80C. Table 1: Corrosion of titanium with alkaline hydrogen peroxide. Effect of peroxide concentration and pH value. A continuous flow of peroxide and alkali was applied to maintain the concentration of the chemicals, the amount of chemicals is calculated on liquid, it has to be multiplied by ten to correspond (at 10% consistency) to pulp. H 2 O 2 (%) NaOH (%) Na 2 CO 3 (%) resultin g pH corrosion rate (mm/year) 0,2 - 0,1 10,3 0 0,2 - 0,2 10,6 0 0,2 0,1 0,1 10,8 0 0,2 0,2 - 11,6 0,06 0,2 0,4 - 12,0 0,22 0,5 0,2 - 11,6 0,28 0,5 0,4 - 11,9 1,62 Na2CO3 %: 0,1 0,2 0,1 - - 0,2 0,2 0,2 0,2 0,2 10 10,5 11 11,5 12 12,5 pH value 0 0,5 1 1,5 2 2,5 corrosion (mm/year) resulting pH corrosion rate H2O2 %: NaOH %: - - 0,1 0,2 0,4 Figure 1: Corrosion of titanium with alkaline peroxide, effect of pH level The results indicate a significant increase of the corrosion level at high pH-values. In addition, the losses of material increase with increasing hydrogen peroxide concentration. On the other hand, at a pH below pH 11, no corrosion takes place. Hydrogen peroxide starts the corrosion only at higher concen- tration levels. The highest corrosion rate in the table coresponds to a hydrogen peroxide charge of 5 % H 2 O 2 on pulp at 10% consistency, an addition value, which is far away from typical peroxide charges. The same is valid for the charges of caustic soda. These are typically in the range of 2% to 3% NaOH on pulp in an extraction stage and are significantly lower in a final peroxide bleaching stage. In addition, in peroxide bleaching the pH level and the peroxide concentration decrease during the reaction. Thus, in the upper part of a bleaching tower the pH and the peroxide concentration are signifi cantly lower. The optimum parameters for hydrogen peroxide bleaching and, with restrictions, also for hydrogen peroxide delignification are not within the range of strong corrosion of titanium 3 metal. The corrosion of titanium is inhibited with calcium ions (water hardness) or magnesium salts [5, 6, 7]. Already 1 ppm of calcium has a pronounced effect [5]. Sodium silicate also inhibits corrosion [5]. Since magnesium is normally added to peroxide bleaching processes, a corrosion of titanium under typical mill conditions becomes very unlikely, if the amount of chemicals added is properly controlled. 4. Prerequisites of hydrogen peroxide application Transition metals are decomposing hydrogen perox- ide. A controlled decomposition with a well defined generation of radicals would be desirable from the point of an improvement of the delignification. Un- fortunately untill today no selective generation has been described. Typically the radicals produced through metal catalysed decomposition are inselec- tive and fiber damage dominates, as a result of cellulose depolymerisation. In consequence, metal impurities have to be removed from the pulp before a subsequent peroxide treatment [8, 9, 10]. The amount of transition metals being present in pulp differs in a wide range and is depending on the soil where the wood has been grown. Normally manga- nese and iron are dominating, other metals like copper and cobalt are present only in traces around 1 ppm. In sulphite pulping the removal of metal is very easy. During the acidic and reducing conditions of the pulping process, metals become water soluble and are easily removed during brown stock washing. In kraft pulping the transition metal ions become insoluble during the pulping process. They are re- duced to a low state of oxidation and precipitated as sulphides. These sulphides are very insoluble under alkaline and neutral conditions, so normal washing does not remove them. During oxygen delignification they might be oxidized to a higher state of oxidation, but the resulting hydroxides are still insoluble under the conditions of oxygen stage washing. They become water soluble under mild to strong acidic conditions. In conventional bleaching processes transition metals are removed during the acidic bleaching stages. Since hydrogen peroxide typically is applied in ECF bleaching only after the first D stage, the metal profile normally is already low enough and no specific measures are required. Figure 2 demonstrates the effect of the pH value on the elimination of iron and manganese from kraft pulp. initial 6 5 4 3 2 1,4 pH value 0 20 40 60 80 100 metals (ppm) Fe Mn Figure 2: Removal of iron and manganese from softwood kraft pulp with increasing acidity. All trials at 3% cons., 60C, 0.5h with H 2 SO 4 for acidification. In comparison to iron the removal of manganese obviously is much easier. Strong acidic conditions are required to lower the amount of iron. In TCF bleaching the removal of the metals becomes more important, because hydrogen peroxide is applied early in the sequence at much higher charges. As demonstrated in figure 3, the elimination of the metals is improved with the addition of chelating agents. Since compounds like EDTA are good chelating agents for double-charged metal ions, the addition of a small amount of sulphite improves the chelation. DTPA can chelate also higher charged metal ions. The metal ions present in the pulp are very likely bound to lignin or cellulose. EDTA and DTPA are competing with these compounds for the metal ions. In addition, alkali earth metals are also chelated by DTPA or EDTA, depending on their chelation constants. Metals like iron do have chela- tion constants, which are about two tenth powers higher, but this is compensated by the fact that the calcium concentration may be more than one hun- dred times higher compared with the concentration of iron. In kraft pulps calcium concentrations of about 5000 ppm are found. If the concentration of calcium is so much higher compared to the concentration of iron, then despite the higher chelation constant, the alkali earth metals are the preferred compounds for chelation. The demand for DTPA will therefore be higher than the stoichiometric equivalent calculated for iron and manganese. The results shown in figure 2 and 3 demonstrate a much stronger binding of iron to the residual lignin or to the cellulose compared to manganese. Even very high charges of DTPA do not affect the iron content at pH 6. In contrast, manganese values are significantly lowered. In consequence, an efficient removal of transition metals needs higher acidic conditions, even in the presence of a chelating agent. The necessity of efficient washing goes without saying. Table 2 gives an example for the metals removal from eucalyptus kraft pulp using acidic conditions with and without chelant addition. 4 DTPA (%): 0,25 0,5 1 all 0,25 % DTPA initial pH 6 pH 6 pH 6 pH 8 pH 7 pH 5 pH2 0 20 40 60 80 100 metals (ppm) Mn Fe Figure 3: Removal of iron and manganese from softwood kraft pulp with DTPA and increasing acidity. All trials at 3% cons., 50C, 0.5 h with H 2 SO 4 for acidification. Table 2: Removal of transition metals from oxygen delignified eucalyptus kraft pulp. Conditions: 3% consistency, 60C, 0,5 h, 1% H 2 SO 4 , 0,5% NaHSO 3, pH 2,3 pH value Fe (ppm) Mn (ppm) oxygen delignified 25 5 A (W) 2,3 11 <1 A (Q) with 0,1% EDTA 2,3 10 <1 Bleaching experiments demonstrated, that it is not necessary to decrease the metals content to ex- tremely low levels. The addition of a small amount of metal ions to a superchelated pulp has a more pronounced negative effect compared with the same amount of metal as residual. I.e. pulp chelated down to 20 ppm iron bleaches better than pulp superchelated down to 5 ppm iron and enriched artificially with another 5 ppm. Very obviously the normal residual of iron or other metals is tightly bound to the pulp and is not available for the decomposition reaction. In consequence there is no advantage of a further drop of the iron level. It is very difficult to describe a threshold value for a "no effect level for iron or manganese, because of the chelation of these metals to the pulp or the lignin. Typically the residual remaining after an acid wash/chelation treatment for iron differs between 10 and 30 ppm and for manganese between 1 and 5 ppm. If washing was efficient, the bleaching per- formance is all right within this range of residual metal ions. 5. Hydrogen peroxide in pulp DELIGNIFICATION 5.1 Hydrogen peroxide in ECF sequences As mentioned above (chapter 2), alkaline hydrogen peroxide has a limited delignification ability. In ECF bleaching it is applied to improve the lignin removal in the extraction stage. In the extraction stage the oxidized lignin is solubilized. The solubility of these compounds in water becomes much better under alkaline conditions, because salts are formed. The oxidation with chlorine dioxide produces less free phenolic groups and carboxylic acid groups on the high molecular residual lignin compared with chlorine. In consequence the combined effect of a chlorine dioxide treatment and a subsequent extraction stage is inferior compared with the result of chlorine plus extraction. This becomes visible in figure 4. 0 2 4 6 8 10 12 14 kappa D-E resp. C-E k-factor 0,25 k-factor 0,175 k-factor 0,1 Figure 4: Delignification of eucalyptus kraft pulp with sequence C-E resp. D-E and mixtures of Cl 2 and ClO 2 The addition of small amounts of oxygen to the extraction stage is a technical possibility since high shear mixers permit a fluidization of the pulp at medium consistency. Oxygen gas is added to the pulp in the form of micro bubbles, and a homogenous mixture of pulp, gas and water is generated. In 1979 the combined addition of oxygen and hydrogen peroxide was first introduced in a pulp mill in southern Germany. Both chemicals oxidize lignin into better soluble, lower molecular weight fragments. Oxygen reacts primarily as an electrophile with the phenolic structures, but produces also peroxo intermediates, which cleave side chains on the residual lignin in a nucleophile reaction. The amount of oxygen is normally set between 5 and 10 kg per ton of pulp. It is generally accepted to overdose oxygen, because oxygen is inexpensive. The consumption of oxygen is a function of the reaction temperature and the mixing efficiency. Increasing the pressure from 2 to 3 bars to higher levels, normally is only of benefit, if the mixer is inefficient. If the oxygen bubbles are really small and properly distributed, a pressure increase cannot produce much effect. With injectors or spargers gas bubbles of a larger diameter may result. Therefore 5 under these conditions a compression can be effec- tive. Because of the limited effect of hydrogen peroxide on deligification it does not make sense to apply larger amounts. Normally the amount of hydrogen peroxide is kept strictly well below 10 kg per ton of pulp. Typically about 3 to 5 kg/t are applied in an EOP stage. Figure 5 compares the effect of an E, an EO and an EOP treatment after chlorination and after chlorine dioxide treatment. C-E C-EO C-EOP D-E D-EO D-EOP 0 2 4 6 8 kappa 5 Figure 5: Effect of oxidative reinforcement of the extraction stage with oxygen and oxygen plus hydrogen peroxide. Eucalyptus kraft pulp kappa 16.2; active chlorine multiple 0.15; ~0.5 % O2, 0.5 % H 2 O 2 . The positive effect of the addition is clearly visible. It is smaller, if chlorine is applied and more pronounced after the less effective delignification with chlorine dioxide. The effect of an increasing hydrogen peroxide charge is given in Figure 6. In the temperature range of 70C to 80C, the effect on the kappa number reduction is pronounced within the addition range from zero to 0.5 % H 2 O 2 . Higher hydrogen peroxide charges lift the brightness, but do not alter the lignin content very much. Under these reaction conditions, a residual of peroxide at the end of the treatment demonstrates the absence of lignin structures which are easily oxidized. However, this effect can be improved in terms of brightness increase and delignification under more drastic conditions, e.g. higher temperature and higher peroxide dosage. Table 3 gives a comparison of the results achieved with increased temperature and pressure on a D 0 pretreated pulp. 0 0,25 0,5 0,75 1 % H2O2 3 4 5 kappa 65 70 75 80 85 brightness % ISO brightness kappa Figure 6: Lignin degradation and brightness increase of eucalyptus kraft pulp with increasing hydrogen peroxide charge in EOP subsequent to the D0 treatment with 0.15 active chlorine multiple: EOP conditions: 10 % cons., 75C, 30 min., 0.3 MPa O2 pressure, 60 min. without pressure. Table 3: Effect of temperature and pressure in an EOP treatment of eucalyptus kraft pulp on kappa, brightness and viscosity. D 0 with 0.15 % active chlorine multiple. Constant conditions in EOP: 1,8% NaOH; 10 % cons., 0.5 h pressure, 1.0 h without pressure (simulation of upflow tube and downflow tower). temp. (C) H2O2 (%) oxygen (MPa) kappa brightness (% ISO) viscosity (mPa . s) 80 0.5 0.3 4.0 81.4 11.2 80 1.0 0.3 3.8 82.5 11.1 100 0.5 0.3 3.2 81.6 10.3 100 0.5 0,5 3.3 81.7 10.3 100 1.0 0.5 3.1 82.8 9.2 The more drastic conditions yield lower residual lignin levels. While the effects on the Kappa number reduction remain moderate, the increase of the temperature basically lowers the pulp quality. At the higher temperature level, the viscosity of the pulp becomes lower. The improved delignification and the possible savings of chlorine dioxide are traded for a lower quality. In addition the higher temperature does have a negative impact on the temperature profile of the sequence. Normally the temperature changes from stage to stage should not be too high, because this might cause difficulties with the temperature level of the washing water circuits. The increase of the pressure does not produce a better peroxide performance. On mill scale certainly pressure can be important, especially if poor oxygen mixing has to be compensated, but this has no impact on the performance of hydrogen peroxide. 6 80 C 80 C 95 C 95 C 2 2,5 3 3,5 4 kappa 17 18 19 20 21 viscosity (mPa.s) kappa viscosity MgSO4: 0,1% -- 0,1% -- Figure 7: Effect of temperature increase and MgSO 4 addition in EOP treatment of softwood kraft pulp. Sequence O-D 0 with kappa-factor 0,175 in D 0 . Reaction in EOP with 1,8% NaOH, 0,3% H 2 O 2 The addition of magnesium salts improves the viscosity levels especially at higher temperature. Figure 7 illustrates an example for softwood kraft pulp. The increase of the EOP stage temperature only yields a moderate drop of the lignin level from kappa 3,4 to 3,1 (which is about 10%), and the viscosity losses can kept moderate, if magesium salts are added. With the addition of oxygen and peroxide in the extraction stage a decrease of the residual lignin level down to kappa numbers between 3 and 4 can be achieved with an active chlorine multiple in the D 0 stage as low as 0.15 to 0.2. Temperatures above the usual level of 75C to 85C in an EOP stage cannot be recommended since the benefit in terms of de- lignification is moderate and the pulp quality suffers. In ECF bleaching the addition of hydrogen peroxide to the first extraction stage has these advantages: Y H 2 O 2 improves the delignification and increases the brightness Y small amounts of H 2 O 2 already yield a pronounced effect Y H 2 O 2 addition can be combined with oxygen Y standard E stage temperature produces good results. 5.2 Hydrogen peroxide in TCF sequences In TCF bleaching sequences delignification is the biggest problem. Basically only three bleaching chemicals are available to decrease and bleach the lignin residual. Oxygen cannot oxidize lignin as efficiently as chlorine. Typically the oxygen stage is run at maximum effect, but the residual lignin level remains high. Ozone is very effective in lignin oxi- dation, but also reacts with cellulose. In addition it is expensive and difficult to mix in higher amounts in the medium consistency range prefered in pulp mills. Thus high ozone charges cannot be applied. This leaves a part of the delignification work for hydrogen peroxide. As mentioned above, hydrogen peroxide is not an ideal delignification agent. It needs a con- siderable amount of pushing, if more than just a small margin of delignification has to be achieved. TCF pulp bleaching technology started with the application of only two bleaching chemicals. This restricted the brightness gain to the semibleached level [8]. Aspa Bruk mills almost classical approach for TCF bleaching is basically a very simple one: 1. Push oxygen delignification to its limits, 2. intensify the oxidative extraction stage with a high peroxide charge, 3. use high temperature in combination with a lot of peroxide and extent the reaction time as long as the tower capacity permits. As any simple concept it has the advantage to work. On the other hand it does not yield miracles. Bright- ness is a function of the kappa number and the delignification with this concept is limited. The first TCF pulps reached only 72 to 75 % ISO brightness. It is difficult to push delignification with oxygen by just repeating the addition of oxygen. Two stages of oxygen treatment indeed improve the lignin removal, but the effect remains moderate [11]. Unfortunately a conventional oxygen stage is run with a very high carry-over of organic compounds, due to the countercurrent water flow into the recovery system. This sets limits to the performance of the oxygen treatment, because the organic material as well as the metal ions promote side reactions and are prohibitive for the application of a higher temperature. For softwood kraft pulp a total delignifi cation of about 60% is the highest level that can be reached without severe cellulose degradation. The remaining amount of lignin is still too high to allow very high brightness levels in a final treatment with peroxide only. An alternative to increase the delignification is the addition of hydrogen peroxide. The prerequisite for the application of hydrogen peroxide is the elimination of the transition metals which remain in the pulp after conventional oxygen delignification. An acidic washing stage removes the metal ions. This washing stage has another positive side effect, the pulp enters the oxygen/peroxide stage without a high load of organic material. The temperature in the oxygen treatment can be increased without severe viscosity losses. This is an advantage also for the consumption of hydrogen peroxide, which would otherwise not react completely. The effect of the addition of 7 hydrogen peroxide to a second oxygen stage following an acidic washing stage is described in figure 8. The dominant amount of additional de- lignification is achieved already with the second addition of oxygen at the higher temperature. The effect on pulp viscosity remains moderate. The addition of 1% or 2% H 2 O 2 increases the brightness of the pulp significantly and allows an additional small drop of the lignin level. first Ostage H2O2 charge (%): -- 1 2 -- 2 15,8 10,5 8,3 7,7 9,5 6,4 100C 100C 100C 110C 110C 5 7 9 11 13 15 17 kappa Figure 8: Delignification of softwood kraft pulp with a second oxygen treatment with and without hydrogen peroxide addition. Effect of temperature and hydrogen peroxide addition on Kappa number. First oxygen treatment to kappa 15,8 and 26,8 mPa . s The drawback of the addition of high peroxide charges is not only cost increase. Side reactions are taking place, which become visible as viscosity losses and a higher COD load, which corresponds to a yield drop. Figure 9 shows the resulting data for viscosity and the effluent load. The moderate drop of the viscosity with the second oxygen treatment is intensified by the addition of peroxide. Very obviously side reactions, like the thermal cleavage of peroxide into hydroxyl radicals, result in an attack of the cellulose chains. Considering the small additional kappa number drop, the steep increase of the COD load can be attributed more to the degradation of cellulose than to lignin removal. first Ostage H2O2 charge (%): -- 1 2 -- 2 100C 100C 100C 110C 110C 10 15 20 25 30 viscosity (mPa.s) 15 20 25 30 35 COD (kg/t) viscosity COD Figure 9: Delignification of softwood kraft pulp with a second oxygen treatment with and without hydrogen peroxide addition. Effect of temperature and hydrogen peroxide addition on viscosity and effluent load. First oxygen treatment to kappa 15,8 and 26,8 mPa . s For hardwood kraft pulp the results are very similar. Table 4 has the data for the treatment of an oxygen predelignified kraft pulp. Following the metals elimination, the second oxygen treatment decreases the residual lignin level from kappa number 12 depending on the temperature to kappa values of 8,4 resp. 6,5. The addition of hydrogen peroxide yields a significant increase in brightness, but only a moderate effect on the lignin. The lower lignin residual is accompanied by a loss of viscosity. Whether this brightness increase in the EOP stage is of benefit for the final brightness level, will be answered in chapter 6.2, together with the question of the total demand for hydrogen peroxide to reach high brightness levels. Table 4: Activation of a second oxygen stage with temperature and hydrogen peroxide addition. Eucalyptus kraft pulp, kappa 12 after first oxygen stage, acid wash with 0,1% EDTA; constant: 2% NaOH, 0,1% MgSO 4 , 10% cons., 1,5 h, 0,3 MPa O 2 tem- perature (C) kappa bright- ness (% ISO) viscosity (mPa.s) O 90 8,4 58,2 22,1 O 100 7,4 62,3 21,8 O 110 6,5 64,8 21,7 OP 110 6,0 82,2 17,1 peroxide addition in OP: 2% H 2 O 2 In TCF bleaching sequences hydrogen peroxide can be applied to decrease the residual of lignin in a second oxygen stage, which is normally a modified extraction stage. It is important to remove transition metals before hydrogen peroxide is added. If the oxygen addition is efficient, i.e. a high shear mixer is applied, oxygen itself is already very effective. This decreases the potential benefit of the peroxide addition: Y Hydrogen peroxide addition to a second oxygen stage (EOP stage) yields a high brightness in- crease - Y the effect on delignification remains moderate - Y at high temperature viscosity is negatively affected and the effluent load increases. 8 6. Hydrogen peroxide in pulp BRIGHTEN- ING 6.1.Hydrogen peroxide in ECF sequences Hydrogen peroxide is applied in the final stages of an ECF bleaching sequence with the target to optimize the requirement for bleaching chemicals. Basically in each extraction stage of a bleaching sequence the addition of hydrogen peroxide is possible. The question is, whether such an addition offers an economical advantage. For northern softwood and the bleaching sequence CD-E 1 -D 1 -E 2 -D 2 this question was positively answered already in 1981[12]. The addition of a small amount of hydrogen peroxide in the E 2 stage was economical, because it lowered significantly the chlorine dioxide requirement for final bleaching. Another variation of the addition of hydrogen peroxide became very typical in the United States: high density storage addition of hydrogen peroxide. It allowed pulp mills to boost the brightness in an ad- ditional bleaching stage without spending capital. In the race for top market pulp brightness and the resulting top selling price advantage, the additional two to three brightness points which are easily achieved with as little as 2 kg to 3 kg of H 2 O 2 do pay off. Certainly high density storage application of hydrogen peroxide is only a compromise. Neverthe- less, it makes sense, because the capital require- ment for an additional bleaching stage is high, and despite the always changing conditions in a storage tower, the results are rather satisfying. The good results of high density storage bleaching open an- other alternative: a final bleaching stage run with hydrogen peroxide. The application of hydrogen peroxide in a final bleaching stage is an answer to the concept of shortening the bleaching sequences and also the answer to the elimination of the second extraction stage. Instead of a D 1 -E 2 -D 2 sequence the D 1 -D 2 alternative was created to cut investment costs. It certainly needs more active chlorine compared with the longer sequence, but the savings in capital for one tower and one washer were believed to pay for. On the other hand, in some mills the higher chlorine dioxide requirement causes problems because of the limitation in the installed chlorine dioxide capacity. An increasing pulp production and the implementation of ECF bleaching can also tighten the availability of chlorine dioxide. Bleaching an eucalyptus kraft pulp with the sequence O-D 0 -EOP-D 1 -D 2 requires about 3,3 kg of active chlorine, to produce the market pulp target brightness of 90 + ISO. The lowest demand results with an active chlorine multiple of 0,15 to 0,175 in D 0 . The combined active chlorine multiple for the D 1 -D 2 stages is relatively high, a kappa factor of 0,5 is needed to achieve the target brightness. The typical increase of the brightness during the D 1 and D 2 stages is demonstrated in figure 10. 0 0,25 0,5 0,75 1 1,25 1,5 1,75 2 active chlorine (%) 87 88 89 90 91 brightness (%ISO) D1 D2 P Figure 10: Brightness increase of an O-D 0 -EOP prebleached eucalyptus kraft pulp (kappa number 3.5 and 82.1 % ISO brightness) with ClO 2 or H 2 O 2 applied in final bleaching at 10 % cons. D stages 75C, 3 h, P stage: 0.25 % H 2 O 2 , O.4 % NaOH, 80C, 3 h. In the D 1 stage increasing amounts of chlorine dioxide yield a significant brightening effect. Nevertheless, this effect levels of at higher chlorine dioxide charges. After an inital steep increase of the brightness, a doubling of the ClO2 charge does not significantly further lift the brightness. With only 0.5 % active chlorine the brightness is lifted from 82,1% ISO to 87.8 %, with twice the amount of chlorine dioxide (1 % active chlorine), only 88.4 % ISO result. The curve for the brightness development in D1 becomes very flat, thus even very high charges of ClO2 do not produce brightness levels above 89 % ISO. This is the reason why indeed an additional stage (D 2 ) is required and in the past a D 1 -E 2 -D 2 treatment was applied. The desired final brightness above 90 % ISO is achieved only after washing and the addition of more chlorine dioxide in the second D stage. Under these conditions a total input of 1.5 % active chlorine yields 90 % ISO brightness. Changing the bleaching conditions, for example the pH environment from acidic to alkaline, or applying a different bleaching agent, normally results in an advantage compared with the repeated addition of the same chemical. So, instead of two subsequent chlorine dioxide treatments, a chlorine dioxide stage followed by an alkaline peroxide stage should be more efficient. This becomes obvious in the left part of figure 10. Very low levels of chlorine dioxide in D 1 yield a sufficient activation to achieve the required brightness increase in the peroxide stage. In consequence, a significantly lower demand for chlorine dioxide is needed to achieve the brightness target. For another pulp sample the results are very similar, as is demonstrated in figure 11. Mill scale trials have reproduced these results and demon- strated its general applicability [13]. This effect can 9 be very attractive to mills with a limited chlorine dioxide capacity, or for mills which want to increase pulp production but not their chlorine dioxide gen- erator capacity. 90,7 91,3 91,3 90,4 91 90,9 88,7 88,7 89,2 89,2 89,3 89,3 0,1 0,1 0,15 0,15 0,2 0,2 active chlorine multiple in D1 88 89 90 91 92 brightness (% ISO) brightness D1 0,1% H2O2 0,2% H2O2 Figure 11: Brightness increase for different kappa-factors in D 1 and peroxide charges. Pre-bleaching with sequence O-D 0 -EOP-D 1 with oxygen stage kappa 11,1; kappa- factor 0,15 in D 0 ; P stage with 0,1 % MgSO 4 , 75C, 1,5 h 10% consistency The amount of hydrogen peroxide needed to boost the brightness is rather low. Figure 12 demonstrates the brightness increase achieved with an increasing hydrogen peroxide charge. Very small amounts of hydrogen peroxide already produce a brightness increase of about two points. In addition the bright- ness stability is improved. A comparison of the re- sults achieved with a final D 1 -D 2 and a D 1 -P treatment demonstrated a decrease from the average 4 to 5 points brightness loss after D 1 -D 2 to only 2 to 3 points after D 1 -P. In consequence, the target brightness on mill scale can be decreased. A calulation of the savings in chlorine dioxide against the demand for hydrogen peroxide shows a replacement ratio of up to 9 kg of chlorine dioxide (as active chlorine) with only 1 kg of hydrogen peroxide. The active chlorine multiple for the D 1 -D 2 stages can be lowered from 0,5 to a level as low as 0,1 for the D 1 -P combination. 0 0,05 0,1 0,15 0,2 0,25 H2O2 charge (%) 87 88 89 90 91 brightness (%ISO) Figure 12: Effect of an increased hydrogen peroxide charge in postbleaching. Bleaching of an O-D 0 -EOP-D 1 prebleached pulp. Kappa-factor in D 0 0.17, in D 1 0.2. The viscosity level in the final peroxide stage can be controlled with the addition of magnesium salts. As expected, the temperature of the P stage has an effect on pulp viscosity. A high peroxide residual also has a negative impact on the viscosity, especially at higher temperature and neutral pH. Table 5 gives the results for a variation of the bleaching conditions in the final peroxide stage. The much better thermal stability of the peroxide bleached pulps becomes obvious, if heat aging brightness losses are compared with standard D 1 -D 2 bleached pulp. In a final peroxide stage the application of pressure was recommended [14]. It is very difficult to understand why this recommendation was made. A small amount of hydrogen peroxide reacts already at 80C within two hours without the application of pressure. Figure 11 illustrates this. At higher temperature the viscosity values suffer and the brightness is not tremendously improved. Table 5: Improvement of the final viscosity after peroxide bleaching of eucalyptus kraft pulp. Pulp prebleached with O-D 0 -EOP-D 1 to 87.7 % ISO brightness. Constant: 10 % cons., 2 h. chemicals addition (%) temp. (C) residual H 2 O 2 bright- ness viscosity (mPa.s) heat ageing H2O2 NaOH MgSO4 sodium silicate (%) (% ISO) (points) 0.25 0.25 0.50 - 0.4 0.4 - - - - - - 70 70 70 0.02 0.01 0.10 87.6 90.2 90.8 8.5 10.8 10.1 7.2 2.0 2.8 0.25 0.25 0.25 0.4 0.3 0.3 0.1 0.1 0.1 - 0.25 0.50 70 70 70 0.04 0.04 0.05 90.5 90.7 90.8 11.2 13.4 13.8 2.6 2.6 2.4 0.1 0.3 0.1 0.50 70 0.03 90.5 13.9 2.4 0.1 0.2 0.3 0.3 0.1 0.1 0.50 0.50 90 90 0.01 0.01 90.7 90.9 12.8 12.5 2.5 2.6 control (D1D2 treatment): 75 - 90.4 13.2 4.6 10 In consequence for the application of hydrogen peroxide in the final stage of ECF sequences our recommendation is not to decide for drastic conditions, but for a smooth reaction at moderate conditions. A final P stage offers these advantages: Y Small amounts of hydrogen peroxide yield high brightness gains already at conventional temperature levels - Y the brightness stability is significantly inproved. The selection of the best conditions in the P stage has to be made with following facts: Y magnesium salt addition increases viscosity - Y buffering yields maximum brightness at highest viscosity - Y high temperature increases the brightness but yields lower viscosity - Y viscosity is negatively affected by residual peroxide at neutral pH. 6.2 Hydrogen peroxide in TCF sequences The basic approach to TCF bleaching with hydrogen peroxide is the simple translation of the available experience with mechanical pulp bleaching to chemical pulp. Metals elimination under mild acidic conditions in the presence of chelating agents or just washing under more acidic conditions generates the basis for peroxide bleaching. High consistency and buffered conditions improve the performance of hy- drogen peroxide and push the reaction to develop brightness and not dominantly lignin degradation. Certainly, under these conditions delignification takes place too, but not with the same intensity. This is demonstrated in figure 13, using a spruce sulphite pulp as an example. Under buffered conditions more brightness is generated and the decrease of the kappa number is smaller. Na- silicate (%) none 4 none 4 unblea. 1% 1% 2% 2% 0 5 10 15 20 kappa 55 60 65 70 75 brightness (%ISO) kappa brightness H2O2 Figure 13: Simultaneous brightening and delignification of spruce sulphite pulp with hydrogen peroxide with caustic soda or caustic soda and sodium silicate. Conditions: 70C, 1,5 h, 10% consistency Semibleaching of any type of pulp with hydrogen peroxide yields very good results at high consistency. The higher relative concentration of the bleaching chemicals and the elimination of dissolved organic compounds is favorable for the performance of the bleaching process [15]. European sulphite mills have modified their bleach plants to a combination of a medium consistency delignification stage and a high consistency bleaching stage [16]. Under high consistency conditions the activation of the peroxide reaction with alkali is achieved with very small amounts of caustic soda, see Table 6. A partial substitution of caustic soda with soda ash (Na 2 CO 3 ) is possible. This produces a very moderate effluent load (COD) which can be translated into a very low yield loss of the bleaching process [15]. Table 6: Decrease of effluent load with high consistency hy- drogen peroxide bleaching. Spruce sulphite pulp predelignified with EOP to kappa 7,6; brightness 75,3% ISO, constant peroxide charge 2%. consis- tency (%) NaOH (%) Na 2 CO 3 (%) Na- silicate (%) bright- ness (% ISO) COD (kg/t) 10 1,6 - - 85,0 29 30 0,5 - 0,75 86,7 17 30 0,25 0,25 0,75 87,1 15 Similarly kraft pulps can be bleached with hydrogen peroxide. Because of the more condensated residual lignin, the prerequisite for the bleaching of kraft pulps is a sufficient amount of delignification in the first bleaching stages [8]. The bleachability of a kraft pulp is depending on the pulping conditions and the intensity of the delignification conditions in the first bleaching stages. Typically hardwood pulps bleach much easier compared with softwoods. The degree of delignification is an important factor for the final brightness level. The basic delignification in TCF sequences is typically achieved with one or two oxygen stages. To improve the effect, a treatment with ozone or peracid can be conducted between the oxygen stages. An alternative is an ozone stage following the delignifi cation with oxygen. The second oxygen stage can be activated with hydrogen peroxide as described in chapter 5.2. Three main problems exist in TCF bleaching: 1. to achieve a kappa number low enough to produce the required brightness reliable, 11 2. to achieve this brightness with an acceptable input of bleaching chemicals, 3. to achieve an acceptable pulp strength. The first TCF softwood kraft pulps available on the market were bleached with a combination of oxygen delignifcation, metals removal, intensified oxidative extraction and a subsequent peroxide treatment. This was achieved with rather high hydrogen peroxide charges. As mentioned above, this procedure is the translation of the existing experience with mechanical pulp to kraft pulp and consequently the European Patent Office denied to grant a patent. Because the resulting kappa number was still high, only brightness levels of 70 % ISO to 75 % ISO were gained. Modified pulping methods, intensified oxygen delignification and increasing temperature levels in the peroxide stage resulted in a continuous improvement of the brightness to about 80 % ISO [17, 18]. The introduction of ozone into the bleaching sequences allowed a further increase of the brightness level. Still brightness 85 % ISO and more require a rather high input of hydrogen peroxide. Published peroxide requirements are about 3 % H 2 O 2 [19], but in actual mill practice 4 % and even 5% of hydrogen peroxide are applied. This is certainly also the result of inefficient mixing of the applied ozone charge and of not optimum conditions in bleaching. It looks as if medium consistency conditions only allow the addition of very small amounts of ozone. Peracids are an alternative to ozone for the activation of pulp prior to a second oxygen stage. If ozone is applied under high consistency conditions, very little ozone is required to yield a pronounced effect. In comparison to ozone the effect on delignification with peracids is limited. The advantages of peracids are the possibility of using standard medium consistency equipement, the option of maintaining the temperature level and the potentially simpler on-site generation. A comparison of the results of the activation with peracid or ozone shows a clear advantage for ozone in terms of delignification (Table 7). Very small amounts of ozone give a significant drop of the kappa number after the subsequent oxygen stage [20]. This kappa number decrease is accompanied by a drop of the viscosity values. Peracids do produce less delignification, but do have only a very moderate effect on the viscosity. Table 7: Activation of a second oxygen treatment with peracid or ozone. Pulp pretreated with sequence O-A (Q) - to kappa 10; ozonation at 30% cons., pH 4 to 5, ambient temp.; peracid stage at 10% consistency, 70C, 1 h, pH end 4,2 to 3,5; second oxygen stage with 2% NaOH, 0,1% MgSO 4 , 0,3 MPa O 2 at 110C, 1,5 h, 10% consistency sequence per- acetic Caro s acid ozone oxygen stage results acid (%) (%) (%) kappa # vis- cosity (mPas) O-A (Q) -O - - - 6,5 21,7 O-A (Q) -Paa-O 0,75 - - 5,5 21,6 O-A (Q) -Paa-O 1 - - 4,7 21,4 O-A (Q) -Paa-O 1,5 - - 4,1 20,6 O-A (Q) -Ca-O - 1,5 4,4 19,1 O-A (Q) -Z-O - - 0,2 5,8 18,8 O-A (Q) -Z-O - - 0,3 4,9 17,1 O-A (Q) -Z-O - - 0,5 2,3 12,5 In order to select the most appropriate pretreatment, the bleachability of these pulps has to be compared. If it would be unnecessary to delignify to very low kappa numbers, if pulps with a higher residual lignin level could be bleached to a high final brightness. The advantage of the still high viscosity at a higher kappa level after the second oxygen stage could be utilized. Thus the question is, whether kappa numbers around 4 already allow to achieve brightness levels above 90 % ISO with an acceptable peroxide input. In bleaching experiments with a pulp with a relative high residual of lignin ( kappa 6) the benefit of the addition of magnesium sulphate and sodium silicate became visible (Table 8). Brightness and viscosity were highest in the presence of both chemicals. In the absence of the stabilizers, the consumption of hydrogen peroxide was higher, but brightness and viscosity were low. Thus, a high turnover of peroxide not necessarily is an advantage. Despite the high reaction temperature of 90C and the long retention time of 8 hours in all experiments the amount of unconsumed hydrogen peroxide was high. The op- timization of the peroxide consumption with increased temperature and extended retention time is sumarized in Table 9. Table 8: Stabilization of hydrogen peroxide bleaching with sodium silicate and magnesium sulphate. Bleaching of kappa 6 eucalyptus kraft pulp with 4% H 2 O 2 at 90C, 10% cons., 8 h. trial NaOH (%) Na- silicate (%) MgSO 4 (%) H 2 O 2 residual (%) bright- ness (%ISO) visco- sity mPas 1 1,5 - - 1,22 85,8 11,2 2 1,4 0,5 - 1,71 86,2 14,4 3 1,4 1 - 1,72 86,4 14,6 4 1,5 - 0,1 1,68 86,8 12,3 5 1,5 - 0,2 1,72 86,9 12,5 12 6 1,4 0,5 0,1 1,88 87,5 14,7 Table 9: Peroxide bleaching at 75C, 90C and 110C; with 4% H 2 O 2 , 1,5% NaOH, 0,5% sodium silicate, 0,1% MgSO 4 , Pulp predelignifed with the sequence O-A (Q) - Ca-O-A (Q) to kappa 4,0. trial tempe- rature (C) consis tency (%) time (h) H 2 O 2 residual (%) bright- ness (%ISO) 1 75 25 24 0,91 89,3 48 0,37 90,5 2 90 25 4 1,07 89,1 6 0,45 90,2 3 90 10 8 1,67 89,3 24 0,74 90,8 4 110 10 2 0,68 88,1 3 0,24 88,5 Chemical reactions are accelerated with a higher concentration of the reaction partners and by a higher reaction temperature. The application of high consistency increases the relative concentration of bleaching chemicals to pulp. At 75C this produced the desired top brightness of more than 90% ISO, but required two days of reaction. An increase of the temperature to 90C did cut the reaction time to only 6 hours. Nevertheless, high consistency is not prefered in kraft pulp bleach plants, therefore the other alternative, temperature increase was also tested. The highest brightness was produced after 24 hours reaction time at 90C, the increase of the temperature above the boiling point of water to 110C yielded a very fast consumption of the hydrogen peroxide, but not an equivalent brightness increase. Obviously the high reaction temperature favors peroxide consumption, but not a consumption which yields bleaching reaction, but consumption based on side reactions. Therefore the comparison for the bleachability following ozone or peracid activation was conducted with only 90C as reaction temperature and one day of reaction time. The results in table 10 demonstrate a better bleachability with peroxide following the ozone/oxygen treatment. Very little ozone is required to achieve a brightness above 90% ISO with only 2% of H 2 O 2 . The advantage of the higher viscosity after the peracid treatment is more than compensated by the better increase of the brightness at a higher residual of lignin. With ozone less chemical yield better delignification and on top of this, brightness increase is higher with less peroxide input at the higher lignin residual. Because high charges of ozone are not required, the final viscosity values become similar to those with the peracid pretreatment. Table 10: Comparison of the sequences O-A (Q) -Z-O-A-P and O- A (Q) -Paa-O-A-P. Peroxide bleaching at 90C, 10% consistency, 24h per- acetic acid (%) ozone (%) kappa after oxygen H 2 O 2 (%) bright- ness (%ISO) visco- sity (mPa s) 0,3 4,9 2 90,7 14,8 0,3 4,9 2,5 91,8 14,7 0,5 2,3 2 90,9 13,2 0,5 2,3 3 91,8 11,0 0,75 5,5 3 88,4 15,5 1,5 4,1 2 88,8 19,1 1,5 4,1 3 90,5 15,0 These results are satisfying in terms of viscosity, chemicals requirement and brightness. Brightness increase is best after ozone activation and very little ozone is needed to produce a brightness above 90 % ISO in the final peroxide stage. There is one open question from chapter 5.2 remaining to be answered, the question whether a prebleaching with hydrogen peroxide in the second oxygen stage to improve delignification also produces a good bleachability in the final bleaching stage. Table 11 gives an answer. The significant brightness increase achieved with the addition of hydrogen peroxide to the second extraction does not have a pronounced effect on the demand for hydrogen peroxide in the final bleaching stage. Despite the high brightness after the OP treatment (82,2 % ISO, see table 4), the final peroxide stage still requires a high peroxide amount to produce 90 % ISO. Compared with the addition of peroxide to the second O stage, ozone activation is much more effective. Table 11: Comparison of the sequence O-A (Q) -Z-O-A-P and O A (Q) -OP-A-P. Bleaching in P at 90C, 10 % consistency, 24h sequence ozon e (%) H 2 O 2 in 2 nd O (%) kappa after oxyge n H 2 O 2 in P (%) bright- ness %ISO viscosit y (mPas) O-A (Q) -Z-O- A-P 0,3 -- 4,9 2 90,7 14,8 O A (Q) -OP-A- P -- 2 6,0 2 89,8 13,2 13 Still the time required to achieve the good brightness levels of table 9 is extremely long. A retention time of 24 hours would require an enormous volume for the bleaching tower. The alternatives are either a high consistency treatment as described in table 9, or a medium consistency treatment at a temperature somewhere between 90C and 110C. 2 4 6 8 12 16 20 24 time (h) 88 89 90 91 92 brightness (% ISO) 1,25% NaOH 1,50% 1,75% 2,00% 2,25% Figure 14: Effect of increasing caustic soda charge on brightness increase with time, bleaching with 3% H 2 O 2 , 0,5 % sodium silicate, 0,1% MgSO 4 , at 15% cons., 95C. Another option is the acceleration of the reaction with higher charges of caustic soda. This is demonstrated in figure 14 for a peroxide charge of 3% at 95C. The brightness development with time at different caustic soda charges indeed is faster with higher alkali amounts. Unfortunately, the faster increase does not yield a higher brightness, the reaction only levels off earlier. While a caustic soda charge of 1,25% does not produce sufficient activation to produce very high brightness even with very long retention time, alkali charges of 1,75% or 2% NaOH yield a rapid comsumption of the hydrogen peroxide, but only a limited brightness increase. In consequence, it does not make sense to apply very high charges of caustic. The reason is the faster consumption of the peroxide in side reactions. The decrease of the H 2 O 2 amount with time is shown in figure 15. A comparison of brightness achived and peroxide consumed clearly teaches the advantage of the lower caustic soda charges. The development of the highest brightness still needs a full day of retention time, but with the addition of 1,5% of caustic soda, after 12 to 16 hours, the brightness is very close to the optimum, only about one point short of the highest value. The negative side effect of high caustic soda charges is the increasing extraction effect. The addition of more caustic soda therefore also results in a lower yield. Especially at higher temperature this effect becomes very pronounced and puts some question marks on the recomendation of such harsh conditions in the literature [21]. A reaction temperature above 100C requires pressurized conditions. Recently it was recommended to pressurize the peroxide treatment in order to achieve a better brightness increase in TCF bleaching [22]. Unfortunately no mill data with pressurized conditions are available, the conditions cited in [22] were not pressurized. In figure 16 the COD values produced with different caustic soda charges at 95C and 110C are compared. The higher COD load which results from the high temperature treatment is clearly visible. H 2 O 2 residual (%) 2 4 6 8 12 16 20 24 time (h) 0,4 0,6 0,8 1 1,2 1,4 1,6 1,25% NaOH 1,50% 1,75% 2,00% 2,25% 1,5 % NaOH, 110C Figure 15: Effect of increasing caustic soda charge on peroxide consumption with time, bleaching with 3% H 2 O 2 , 0,5 % sodium silicate, 0,1% MgSO 4 , at 15% cons., 95C. 1,25 1,5 1,75 2 2,25 NaOH (%) 6 8 10 12 14 16 18 COD (kg/t) 95C 110C Figure 16: Effect of increasing caustic soda charge on the COD load of hydrogen peroxide bleaching at different temperature, conditions see figure 14 According to the rules of thermodynamics, the appli- cation of a very high pressure very likely hinders the decomposition of hydrogen peroxide into oxygen and water (equ. 2). The pressure needed is very high, it cannot be expected that only some atmospheres are sufficient. Experiments in a laboratory scale high shear mixer which permitted the mixing of hydrogen peroxide in the presence of oxygen at temperature with the preheated pulp, did not show any benefit of pressure in the range of one to four atmospheres [20]. Within the range of reproducability, the results shown in table 12 were identical. It is difficult to answer the 14 question, why different results with positive effects are reported in the literature [23], it looks as if the results are depending on the laboratory scale conditions and the equipement used for the trials. Table 12: Effect of pressurizing a peroxide bleaching stage with oxygen at 110C, pre- treatment with O-A (Q) -EOP- to kappa 6,0; bleaching with 4% H 2 O 2 , 1,5% NaOH, 0,5% sodium silicate, 0,1% MgSO 4 , at 10% cons., 2h in a high-shear mixer. trial oxygen pressure MPa H 2 O 2 residual % brightness %ISO viscosity mPas 1 - 0,76 86,4 15,8 2 0,2 0,81 86,6 15,4 3 0,3 0,79 86,3 15,1 4 0,4 0,82 86,7 15,3 Table 13 summarizes the compromise for the best results in terms of retention time and brightness increase. A total time demand of 8 hours retention time at 100C seems to be acceptable, it results in a brightness above 90% ISO, which is higher compared with the very fast, but less efficient reaction at 110C. Table 13: Optimization of the retention time/temperature relationship in peroxide bleaching. Pulp predelignified to kappa 4,3 with the sequence O-A (Q) -Z-O-A; const. 2,5% H 2 O 2 , 0,5% sodium silicate, 0,1% MgSO 4 tempe- rature (C) consi- stency (%) NaOH (%) time (h) bright- ness % ISO visco- sity mPas 95 15 1,5 8 90,0 13,9 16 91,4 13,7 100 15 1,75 8 90,3 13,8 12 90,5 13,6 110 10 1,5 2 88,5 13,3 The recommendation for the application of hydrogen peroxide in TCF sequences therefore is to concentrate the hydrogen peroxide input on one or two final bleaching stages. The prerequisite is a sufficient delignification with oxygen and ozone is the earlier bleaching stages. The performance of high hydrogen peroxide charges is best at temperatures lower than 100C because decomposition and intensified ex- traction are the negative side effects of higher tem- perature. The conditions for the peroxide stage are very flexible, but in order to have an acceptable re- tention time, the temperature and the consistency have to be high. Other parameters of importance are: Y magnesium salt addition increases viscosity - Y buffering yields maximum brightness at highest viscosity - Y high temperature increases the brightness, but with very high temperature bleaching becomes inefficient. 7. Summary The final conclusions in this look over peroxide ap- plication and performance in chemical pulp bleaching can be made as follows: Hydrogen peroxide is a very versatile bleaching agent. It can be applied in a wide range of conditions: Y Temperature range: (starting at about 20C) normally: 60C to 75C, in special cases up to 100C. Y Retention time range: (starting with about 30 min.) normally: 1,5 h to 2 h and up to 16 h, in special cases 1 day to 3 days. Y Consistency range: not below 8 % consistency, normally: 10% to 15% consistency, in special cases up to 30% consistency. Y In ECF bleaching sequences the application of hydrogen peroxide offers significant economical benefits. The addition of hydrogen peroxide in the extraction stage improves the delignification and allows lower kappa factors in D 0 . Y Final bleaching with hydrogen peroxide cuts the requirement for chlorine dioxide in D 1 . A very low input of hydrogen peroxide allows kappa factors in D 1 as low as 0,1 to 0,15. Y The decreased demand for chlorine dioxide yields lower cost, lower AOX of the effluent and a lower OX in the pulp. Y The moderate conditions (pH and concentration) required for bleaching with hydrogen peroxide do not result in a corrosion hazard for titanium equi- pement. Y In TCF bleaching sequences hydrogen peroxide is the chemical of choice to produce the required brightness level. Y Peroxide bleaching yields highest brightness in- crease at temperature levels between 70C and 90C with long and very long retention time. Y For kraft pulps the best compromise is an increase of the temperature close to the boiling point. YAt high temperature the addition of magnesium salts and buffering with very small amounts of soduim silicate is the best strategy to obtain highest viscosity values. 15 8. References [1] N. N. Greenwood and A. Earnshaw; Chemistry of the Elements; Pergamon Press, Oxford (1984) [2] J. Takagi and K. Ishigure; Thermal Decompo- sition of Hydrogen Peroxide and its Effect on Reactor Water Monitoring of Boiling Water Re- actors; Nuclear Science and Engineering: 89, 177 - 186 (1985) [3] J. Gierer; The Chemistry of Delignification; Holzforschung 36, (2) 55 - 64 (1982) [4] G. Gellersted and I. Petterson; Chemical As- pects of Hydrogen Peroxide Bleaching. Part II: The Bleaching of Kraft Pulp; J. Wood Chem. and Techn. 2 (3) 231 - 250 (1982) [5] R. W. Schutz and M. Xiao; Development of practical guidelines for titanium in alkaline per- oxide bleach solutions; Intn. Pulp Bleaching Conf. Poster Proceedings 153 - 157 (1994) [6] S. J. Clarke and D. L. Singbeil; Corrosion of titanium in alkaline peroxide bleaching media; Pulp Paper Canada 95 (10) T417 - T421 (1994) [7] J. MacDiarmid, R. Charlton and D. L. Reichert, Corrosion and material engineering considera- tions in hydrogen peroxide bleaching; TAPPI 7th Intn. Symposium on Corrosion Proceedings, 97 - 104 (1992) [8] H. U. Sss, W. Eul, O. Helmling; Semibleaching of kraft pulp using oxygen and hydrogen peroxide; Papier 43 (7) 318 - 323 (1989) [9] J. Bouchard, H. M. Nugent and R. M. Berry; A comparison between acid treatment and chela- tion prior to hydrogen peroxide bleaching of kraft pulps, J. Pulp and Paper Science; 21, (6), J203 - J208 (1995) [10] L. Lapierre, J. Bouchard, R. M. Berry and B. van Lierop; Chelation Prior to Hydrogen Peroxide Bleaching of Kraft Pulps: An Overview; Journal Pulp and Paper Science: 21, (8) J268 - J273 (1995) [11] A. P. Johnson; Worldwide Survey of Oxygen Bleach Plants; Non-Chlorine Bleaching Conf. Proceedings, Hilton Head, 1992 [12] H. Loutfi; The use of hydrogen peroxide in bleaching of kraft softwood pulp; CPPA annual meeting 1981, B71 - B77 [13] C. A. dos Santos, H. U. Sss, O. Mambrim Filho; Flexibilizao da sequncia de bran- queamento ECF da Bahia Sul Celulose s. a.; 28 Congresso anual de celulose e papel, So Paulo 1995 [14] R. Reeves, R. Boman and S. Nordn; Tappi 1995 Pulping Conf. Proceedings 263 - 267 [15] N. Nimmerfroh, H. U. Sss, J. D. Kronis; Two- stage, high brightness bleaching of sulfite pulp; 1992 Tappi Pulping Conf. Proceedings 791 - 801 [16] R. Hock, W. Czirnich; Chlorfreie Bleiche von Zellstoff mit Einbindung von Abwasser in den Kreislauf der Kochereiabwsser - Konzept und Ergebnisse einer neuen Bleichsequenz; Papier 47, (10A) V24 - V29 (1993) [17] J. Basta, L. Holtinger, L. Lundgren,and P. Fas- ten; Tappi 1991 Intn. Pulp Bleaching Conf. Proceedings Vol 3 , 23 - 33 [18] J. Basta, L. Holtinger, K. Gutke; Reducing Levels of AOX, Part 4: Lignox on Hardwood Pulps; Tappi 1991 Pulping Conf. Proceedings Vol 1; 153 - 158 [19] J. Basta, L. Andersson and W. Hermansson; The Potential of Lignox and Complementary Combinations; Proceedings Non-Chlorine Bleaching Conf., Hilton Head 1992 [20] H. U. Sss, N. F. Nimmerfroh and O. Mambrim Filho; TCF bleaching of eucalyptus kraft pulp: The selection of the right sequence and the best conditions; Tappi 1996 Intn. Pulp Bleaching Conf. Proceedings 253 - 260 [21] J. Basta, L. Holtinger, P. Lundgren and C. Persson; Emerging Technologies In TCF Bleaching; Tappi 1995 Pulping Conf. Proceed- ings, 53 - 57 [22] L. Sjdin, S. Norden and R. Boman; Extended Delignification with Oxygen and Hydrogen Peroxide in ECF and TCF Sequences, Tappi 1994 Pulping Conf. Proceedings, 21 - 27 [23] J. Devenyns, R. Detroz, A. Renders; Enhanced P stages for ECF and TCF bleaching; Tappi 1996 Intern. Pulp Bleaching Conf. Proceedings 295 - 302