Since its introduction in the 1970s, oxygen delignification has become a standard process step for the production of bleached pulp because it minimises emission of chlorinated organic compounds when combined with elemental chlorine free (ECF) bleaching. The advantages are that the oxygen stage effluent becomes part of the kraft recovery cycle, while the operating costs are lowered because oxygen is the lowest cost bleaching chemical (van Heiningen et al. 2003).
Oxygen is reactive towards lignin at moderate temperatures in an alkaline environment. The negative aspect of oxygen delignification is its poor selectivity (i.e., delignification/cellulose degradation), especially at a high degree of delignification (Genco et al. 2012) and high initial kappa numbers (Kn) as recently reported for Southern pine kraft pulps (KP) (Tao et al. 2011). The selectivity of oxygen delignification may be increased by following Lachenal’s recommendation to use an activating agent X in an OXO sequence, where X is an electrophile such as chlorine, ozone, or a peracid placed in between two oxygen stages (Suess 2010; Lachenal et al. 1997). This can be achieved by distributing the alkali charge over several oxygen stages (Violette 2003) and/or by changing operating variables such as temperature and alkali dosage (van Heiningen and Ji 2012). The degree of delignification in oxygen/alkali process is influenced by the Kn of the incoming pulp with a higher reactivity at higher lignin content (Agarwal et al. 1999). Also, in contrast to its poorer selectivity in terms of the ratio delignification/cellulose degradation, the selectivity in terms of pulp yield/delignification of oxygen delignification is better than that of KP at the residual stage (Kleppe et al. 1972; Parsad et al. 1996; Hart and Connel 2006). Thus, a higher yield of bleached KP could be obtained if a high-Kn pulp (Kn45-90) is oxygen delignified to a standard incoming Kn for ECF bleaching. However, an improved ratio of delignification/cellulose degradation is needed to achieve the required high Kn reduction. A recent study on a conventional KP (Kn24) has shown that a high degree of delignification may be achieved by maintaining a constant and relatively low alkali concentration during oxygen bleaching (van Heiningen and Ji 2012). Nevertheless, for high Kn pulp this would require distributing the increased total alkali charge over multiple oxygen stages, and perhaps an interstage reactivation of lignin to maintain a high overall delignification rate and thus reasonable overall reaction time. This approach has been reported for a pine KP with Kn68 using peracetic acid (Pa) in an OPaO sequence. This results in an overall improved yield gain (2.5% based on wood) compared to conventional kraft-oxygen alkali process at the same final Kn level (Kn14). Nevertheless, the final viscosity values are not reported (Danielewicz and Surma-Slusarska 2006). Most recently, Vehmaa et al. (2012) reported the application of Px in interstages on a Kn79 Scandinavian softwood KP. By means of an OPxO sequence, the authors produced a pulp with Kn15 and an intrinsic viscosity of 791 ml g-1; no comparison with conventional kraft and oxygen delignified pulp was reported in terms of final pulp yield.
Thus far, different compounds have been studied as activators and catalysts for oxygen delignification to achieve greater delignification without negative effects on pulp yield and pulp strength properties (Suchy and Argyropoulos 2002). Px has been applied as an oxidizing agent to improve the reactivity of oxygen towards lignin (Allison and McGrouther 1995; Bouchard et al. 2001). The relatively low reactivity between oxygen and organic molecules is caused by the unusual electron configuration of oxygen (as triplet oxygen) that hinders molecular oxygen to react directly with organic substances, which are mostly in the singlet state. It is known that Px is a selective and efficient delignifying chemical under alkaline conditions as it produces a singlet oxygen molecule that easily attacks double bonds in lignin (Suchy and Argyropoulos 2002). The low lignin reactivity at extended oxygen delignification is frequently explained by the disappearance of non-condensed lignin units containing free phenolic OH groups (Kalliola et al. 2011). The hypothesis is that Px could compensate the decreased amount of reactive free phenolic groups obtained after the initial oxygen delignification.
In the present study it is attempted to overcome the low selectivity concerning the delignification/cellulose degradation ratio of oxygen delignification at Kn reductions >60% applied to high Kn softwood KP, while a pulp with a higher yield should be obtained than in the case of a conventional KP in combination with oxygen delignification from the same wood furnish. A multistage oxygen delignification will be in focus, in which interstage peroxymonosulfate treatment and distributed alkali charges will be applied. To evaluate cellulose degradation by radicals created by metal catalysed decomposition of peroxides in oxidative bleaching (Lachenal et al. 1997), EDTA treated and untreated pulps will be investigated.
Materials and methods
A high-yield pine kraft pulp (KP with Kn65) and a conventional pine KP with Kn24 derived from the same wood supply were obtained from a Finnish mill and cooked in a super-batch process. The pulp yields were estimated by the mill as 51% and 46%, respectively.
All O-stages were conducted in a rotating air bath digester each with four bombs of 2.5 l capacity. Lids were constructed with two fast-connect clamps for precise and rapid initial pressurisation with oxygen and rapid release of oxygen at the end of the process. In a typical experiment performed at 10% pulp consistency, a known weight of pulp (30 g o.d.) and MgSO4 (0.05% on o.d. pulp) and varying amounts of NaOH were introduced into the reactor. Then oxygen was injected twice to purge the reactor followed by an increase in oxygen pressure to the target pressure. The bombs are heated to the desired temperature, and when the required time at the target temperature was reached, the reaction was terminated by releasing the pressure and cooling the bomb by submersion in cold water. The pulp was then washed, centrifuged, and pressed to high consistency (above 30%) for the determination of pulp properties.
Chelation (Q stage)
Pulp chelation with EDTA was performed at a charge of 0.2% based on pulp, 50oC and consistency of 5% for 30 min. The pH was maintained at 5.5 by addition of sulphuric acid to the stirred pulp suspension. Then the pulp was centrifuged and homogenised. The content of the relevant metals in the present high Kn pulp before and after chelation is reported in Table 1.
Treatment with Px
The source of Px was oxone (potassium monopersulfate triple salt chemical, composition: KHSO5·0.5 KHSO4 0.5 K2SO4) obtained from Sigma-Aldrich. The active oxygen content of oxone was determined following the ASTM D2180 standard method. The pulp suspension was treated with the desired percentage of oxone in term of active oxygen. Px was added to the pulp suspension at 10% consistency at the desired active oxygen charge in a mixer (Kenwood Chef model, A701A) for 10 min at about 50oC. The pH inside the pulp suspension was maintained at 11–12 by adding Px in powder form and a caustic solution on opposing sides of the mixer, as described earlier, to prevent polysaccharide degradation due to the high acidity of Px (Bouchard et al. 2001)
Analytical measurements and procedures
The Kn, viscosity value and pulp dry matter were determined according to the standards Scan-CM 15:99, Scan-C 1:100 and Scan-C 3, respectively. Pulps with Kn numbers above 35 were delignified with chlorite (5 g pulp in 200 ml water + 5 g NaClO2 + 2 ml acetic acid at 70oC for 5 min) prior to pulp viscosity measurement. Neutral sugars and lignin content of the pulps were determined following the NREL/TP-510-42618 method (anion exchange chromatography, Dionex HPAEC). The total organic carbon (TOC), total inorganic carbon (TIC) and total carbon (TC) content of the spent liquors were measured in a Shimadzu analyzer based on the EN 1484 standard. The yield of the oxygen-delignified pulps was determined by weighing the pulp after each stage (dry matter content 29–33%) and the yield was correlated to the TC content of the corresponding effluent. The hexenuronic acid (HexA) content of the pulp before and after oxygen bleaching was determined according to Vuorinen et al. (1999). The Kn reported in the present paper has not been corrected for HexA content. The standard development of the Kn measurement is <3% and for the intrinsic viscosity measurements it is 2%. The cellulose DP is calculated from the intrinsic viscosity [η] of the pulp in ml g-1, and the weight fractions of hemicelluloses (H) and cellulose (G) in the pulp according to (van Heiningen et al. 2004):
The ratio delignification/cellulose degradation (selectivity) of the nth O or OPx stage, Sn, is defined as the Kn reduction during this stage, Kn-1-Kn, divided by the number of cellulose chain scissions during the nth stage, 1/DPn-1/DPn-1, the average DP of cellulose in the pulp after stage n and n-1, i.e.:
The lignin-free yield on wood after oxygen delignification is obtained by multiplying the cooking pulp yield with the lignin-free yield of the oxygen stage, i.e,.: 0.51 * yield in oxygen stage * ((100-Kn * 0.15)/100).
Yield determination from the total carbon (TC) content of the filtrate
The yield of oxygen delignification is defined as the weight of the washed pulp (o.d. basis) after delignification divided by the weight of the original o.d. pulp. Because of pulp handling losses and the relatively small yield loss during oxygen delignification (<10%), the determination of the pulp yield by weighing is not very exact. During oxygen delignification, the dissolved organic pulp weight ends up in solution as dissolved organics and carbonate; the latter due to reaction of generated CO2 with NaOH. Thus the loss of pulp weight may be related to the total carbon-sum of total inorganic carbon (TIC) and total organic carbon (TOC) in the solution. To verify this, a number of oxygen delignification experiments were done, whereby the pulp yield loss obtained by weighing was correlated to the increase in TC in the filtrate.
Were carried out in the air bath digester at 10% consistency with pulp weight of 10, 20, 30 and 50 g, 7 bar oxygen pressure at 90oC and 3% NaOH and 0.05% MgSO4 (based on o.d. pulp) after 10, 20, 30, and 60 min of reaction. Figure 1 shows that the yield loss obtained by weighing is well correlated with the TC content of the filtrate independently of the initial pulp weight and reaction time. This finding indicates that the percentage yield loss during oxygen delignification of pine KP can be derived from the TC content of the filtrate in ml g-1 pulp divided by 5.25. This value compares well with the value of 5.4 for softwood KP found by Salmela (2007).
Validation of air-bath digester system for oxygen delignification
It is expected that the oxygen delignification results obtained with the air bath digester will be affected by mass transfer above a certain pulp weight addition because oxygen will not be able to fully penetrate/diffuse into the 10% consistency pulp suspension despite the mixing induced by the slow rotation of the digesters. Also at high degree of filling of the digesters, the charge of oxygen will be insufficient to prevent a significant decrease in oxygen pressure following oxygen consumption. Therefore, the effect of pulp weight at 10% consistency on the delignification was investigated to determine the maximum pulp weight charge to the air bath digester, although the kinetics were not affected by either of these limitations. As Figure 2 illustrates, the delignification kinetics were unaffected up to a pulp weight of 50 g. The validation trials of the air bath reactor were the same as described above for the yield determination from the filtrate TC. Based on these results, the present oxygen delignification experiments were performed with 30 g pulp.
Results and discussion
EDTA treated and untreated high Kn65 pulps were oxygen delignified in a single stage, two-stage and three-stage process, with and without peroxymonosulfate (Px) treatment, i.e., OPx, OPxO, OPxOPxO, and O, OO, OOO respectively. No interstage washing was applied between the stages. The total NaOH charge in each process was 6% (based on o.d. pulp), with a distribution of 6, 4 + 2, and 3 + 2 + 1% on initial pulp weight in the oxygen stages, respectively. The total Px charge in each process is 2% as active oxygen, with a distribution of 2, 2, and 1 + 1% in the OPx, OPxO and OPxOPxO sequences, respectively. The conditions of these treatments and also reaction conditions are summarised in Table 2.
Figure 3 shows the effect of NaOH distribution and Px addition on the selectivity for unchelated and chelated high Kn pulp. For the two or three stage oxygen delignification trials, the first stage was performed at 85°C and the following at 95°C in accordance with the finding of Iribarne and Schroeder (1997), in the sense that the degree of delignification at the same pulp viscosity loss may be increased at higher temperature in the second oxygen stage. This has been explained by lower activation energy of high-reactive lignin compared to carbohydrates and the need to oxidize the residual low-reactive lignin at higher temperatures (Asgari and Argyropoulos 1998).
The results show that the degree of delignification of the high Kn pulp is much increased by the addition of Px in all sequences, and the viscosity/Kn ratio is maintained at the high degree of selectivity. Chelation improves the pulp viscosity, but it has no significant effect on the degree of delignification. Figure 4b shows that the OPxOPxO sequence for the chelated high Kn brown-stock produces a Kn15.4, 910 ml g-1 viscosity pulp, i.e., an acceptable quality for subsequent bleaching. Also included in Figure 4b are the results of an OO treatment of a Kn24 pulp. The resulting pulp has only slightly improved properties (Kn13, 950 ml g-1) compared to the chelated OPxOPxO treated pulp with high Kn. This shows that it is possible to produce a standard incoming Kn pulp for ECF bleaching without a large loss in cellulose DP starting from a Kn65 pulp and slightly higher yield by using multistage oxygen delignification with distributed alkali charge and interstage Px treatment.
The intrinsic viscosity ([η]), Kn, percentage delignification (100ΔKn/Kn0), cellulose DP and selectivity (S) after each sequence for the unchelated and chelated pulp (12 trials in total) are listed in Table 3. The data in Table 3 shows that distribution of the 6% alkali charge over three oxygen stages increases the degree of delignification to 69% and 76% without and with Px interstage treatment, respectively, independently of whether the pulp is chelated or not. It is also visible that Px treatment increases delignification by 7% together with a significant viscosity improvement compared to the corresponding OOO series. However, when all the alkali is charged in a single stage, the degree of delignification reaches only 44% and 60% in the O and OPx treatments, respectively, both of which are too low for producing a pulp of appropriate properties before the bleaching treatment. This is in agreement with data of Tao et al. (2011) who also found that the degree of delignification in a single oxygen stage is insufficient for high Kn pulp. The selectivity Sn listed in Table 3 decreases with increasing delignification, with larger decreases occurring for sequences without Px treatment. This is more clearly illustrated in Figure 4a,b. This figure also shows that the effect of Px treatment is more significant than the effect of chelation. It is interesting to see that the selectivity, Sn, approaches zero for the third O stage, while it is still 3×10-5 for the 3rd OPx stage irrespective of chelation pretreatment. This suggests that without Px treatment more stages beyond 3 will be highly unselective while with Px treatment further stages may be possible.
The reduction in selectivity with increasing degree of oxygen delignification may be interpreted that the reactivity of residual lignin decreases. It has been reported that residual lignin is less reactive due to increased retention of low reactive LCC (lignin-carbohydrate-complexes) – namely: xylan-lignin, glucomannan-lignin, and glucan-lignin complexes. Moreover, the relative amount of β-O-4 structures becomes lower and that of non-reactive condensed phenols higher (Lawoko et al. 2004). The comparison of the different treatments in terms of selectivity versus Kn shows that Px reactivates the residual lignin towards oxygen in the second and third stages as was also observed by Bouchard et al. (2001). The impact of the different oxygen delignification sequences on final pulp yield (based on original wood) can be seen in Table 4 and Figure 4c.
Figure 4c shows that the OPxOPxO sequence applied to the chelated high Kn pulp leads to a small improvement (0.3%) in lignin-free yield (on wood basis) compared to the unchelated pulp, which is about 0.5% higher than that after a double O stage applied to the conventional Kn24 pulp at the same Kn level. Figure 4c also shows that without application of the interstage Px treatment, i.e., after the OOO sequence, the Kn only decreases to about 20 irrespective of whether the high Kn pulp was chelated or not.
In addition, the lignin-free pulp yield at Kn20 after OOO is still about 1.5% lower than that after OPxOPxO at Kn15. Therefore the lignin-free yield of the conventionally Kn24 pulp, derived from the same wood supply as the Kn65 considered in the present study, lies in between that of the multistage oxygen treatment with or without interstage Px application. This result confirms the recent finding of Tao et al. (2011) that extended oxygen delignification of high-Kn pulp to the same final Kn as that of conventional single stage oxygen delignification of conventional KP derived from the same wood furnish leads to a lower pulp yield and a lower selectivity. However, the present study shows that by means of an interstage Px treatment in a OPxOPxO sequence the final lignin-free yield is about 0.5% higher than the conventional pulp at Kn15, while the intrinsic viscosity is only slightly smaller (about 900 vs. 950 mg g-1 at about Kn15, respectively). The lignin-free yield during oxygen delignification may be less affected by the cellulose chain cleavage because of oxidation of the newly created reducing group while this group is subject to alkaline peeling during pulping (Fengel and Wegener 1984). Evidence for this can be seen in Table 4, which shows multi-stage oxygen bleaching with and without Px significantly increase delignification with only a minor decrease in cellulose yield which is in agreement with the finding of Leroy et al. (2004). Comparison of the carbohydrate content of bleached pulp after OPxOPxO series with that of conventional oxygen delignified in Table 4 shows higher cellulose content for high Kn pulp even after three stages of oxygen delignification at the same Kn level, while the opposite is true for the hemicellulose content. This suggests that the higher hemicellulose content might be a factor for the slightly higher viscosity for OO-conventional pulp at the same Kn in comparison with OPxOPxO-pulp, as they serve to protect cellulose from harmful free hydroxyl radicals that are continuously generated during oxygen delignification. (Guay et al. 2002; Violette and van Heiningen 2002; van Heiningen and Violette 2003).
A Kn number reduction of 76% was obtained for a pine kraft pulp (KP) of initial Kn65 after a three-stage oxygen delignification with interstage peroxymonosulfate acid (Px) treatment (i.e., OPxOPxO) and 6% total alkali charge and 2% (as active oxygen) total Px charge. By distributing the alkali over the three stages and applying an initial chelation stage, a pulp with Kn15.4 and adequate viscosity level of 910 ml g-1 was produced. Without Px interstage treatment (i.e., OOO), a less delignified and lower viscosity pulp (Kn19.9 and viscosity 840 ml g-1) was obtained. The lignin-free yield of the OPxOPxO pulp derived from the high Kn65 pulp is about 0.5% higher than that of a conventional KP with Kn24 and then oxygen delignified pulp derived from the same wood supply at a final Kn15. The Px treatment leads to increased delignification and delignification/cellulose degradation ratios owing to higher reactivity of residual lignin. Removal of heavy metal ions from pulps prior to the O-stage by means of a chelating agent treatment improves the selectivity for the sequences with Px treatment.
Financial support from FIBIC (Finish Bioeconomy Cluster) is gratefully acknowledged. The authors wish to give special thanks to Heikki Tulokas for his technical assistance.
Agarwal, S.B., Genco, J.M., Cole, B.J.W., Miller, W. (1999) Kinetics of oxygen delignification. J. Pulp Pap. Sci. 25:361–366.Google Scholar
Allison, R.W., McGrouther, K.G. (1995) Improved oxygen delignification with interstage peroxymonosulfuric acid treatment. Tappi J. 78:134–142.Google Scholar
Asgari, F., Argyropoulos, D.S. (1998) Fundamentals of oxygen delignification. Part II. Functional group formation/elimination in residual kraft lignin. Can. J. Chem. 76:1606–1615.Google Scholar
Bouchard, J., Magnotta, V., Berry R.V. (2001) Improving oxygen delignification with peroxymonosulphate: The (OPx) process. J. Pulp Pap. Sci. 27:397–401.Google Scholar
Danielewicz, D., Surma-Slusarska, B. (2006) Oxygen delignification of high-kappa number pine kraft pulp. Fibres Text. East. Eur. 14:89–93.Google Scholar
Fengel, D., Wegener, G. Wood chemistry, ultrastructure,reactions. Walter de Gruyter, Berlin/New York, 1984.Google Scholar
Genco, J., van Heiningen. A., Miller W. (2012) Oxygen delignification. In: The bleaching Of Pulp-P. Eds. Hartand, A.Rudie, Chapter 2, Tappi Press, Atlanta, GA, USA.Google Scholar
Guay, D.F., Cole, B.J.W., Fort, R.C., Hausman, M.C., Genco, J.M., Elder, T.J. (2002) Mechanisms of oxidative degradation of carbohydrates during oxygen delignification. Part III: Reaction of photochemically generated hydroxyl radicals with 1,5-anhydrocellobitol and cellulose. J. Pulp Pap. Sci. 28:217–221.Google Scholar
Hart, P., Connel, D. (2006) The effect of digester kappa number on the bleachability and yield of EMCCTM softwood pulp. Tappi J. 67:127–137.Google Scholar
Iribarne, J., Schroeder, L. (1997) High-pressure oxygen delignification of kraft pulps,1: Kinetics. Tappi J. 80:134–250.Google Scholar
Kalliola, A., Kuitunen, S., Liitia, T., Rovio, S., Ohra-aho, T., Vuorinen, T.,Tamminen, T. (2011) Lignin oxidation mechanisms under oxygen delignification conditions. Part 1. Holzforschung 65:567–574.Google Scholar
Kleppe, P.J., Chang, H.-M., Eckert, R.C. (1972) Delignification of high yield pulp with oxygen and alkali. I. Preliminary studies on southern pines. Pulp Paper Canada 73:400–404.Google Scholar
Lachenal, D., Thi, N.B.N., Chirat, C., Soria, L. (1997) Optimum use of hydrogen peroxide in kraft pulp delignification. Pap. Puu 79:252–256.Google Scholar
Lawoko, M., Berggren, R., Berthold, F., Henriksson, G., Gellerstedt, G. (2004) Changes in the lignin-carbohydrate complex in softwood kraft pulp during kraft and oxygen delignification. Holzforschung 58:603–610.Google Scholar
Leroy, N., Chirat, C., Lachenal, D., Robert, D., Allison, R.W. (2004) Extended oxygen delignification. part 2: multi-stage oxygen bleaching with intermediate hypochlorous acid stages. Appita J. 57:224–227.Google Scholar
Parsad, B., Kirkman, A., Jameel, H.,Tratzl, J., Magnotta, V. (1996) Mill closure with high-kappa pulping and extended oxygen delignification. Tappi J. 79:144–152.Google Scholar
Salmela, M. (2007) Description of oxygen-alkali delignification of Kraft pulp using analysis of dissolved material. Dissertation/Thesis, University of Jyväskylä.Google Scholar
Suchy, M., Argyropoulos, D.S. (2002) Catalysis and activation of oxygen and peroxide delignification of chemical pulps:Review. Tappi J. 1:1–18.Google Scholar
Suess, H.U. Pulp bleaching today. Walter de Gruyter GmbH & Co. KG, Berlin/New York, 2010.Google Scholar
Tao, L.M., Joseph, G., Barbara, J.W.R.C., Fort, J.R. (2011) Selectivity of oxygen delignification for southern softwood kraft pulps with high lignin content. Tappi J. 10:29–39.Google Scholar
van Heiningen, A., Ji, Y. (2012) Southern pine oxygen delignified pulps produced in a Berty throughflow reactor: How to obtain the highest degree of delignification while maintaining pulp yield and quality. Tappi J. 11:9–18.Google Scholar
van Heiningen, A., Krothapalli, D.,Genco, J., Justason, A. (2003) A chemical reactor analysis of industrial oxygen delignification. Pulp Pap. Can. 104:96–101.Google Scholar
van Heiningen, A., Tunc, M.S., Gao, Y.D., Silva Perez, D. (2004) Relationship between alkaline pulp yield and the mass fraction and degree of polymerization of cellulose in the pulp. J. Pulp Pap. Sci. 30:211–217.Google Scholar
van Heiningen, A.,Violette, S. (2003) Selectivity improvement during oxygen delignification by adsorption of a sugar-based polymer. J. Pulp Pap. Sci. 29:48–53.Google Scholar
Vehmaa, J., Engstroem J., Piira, J. (2012) Method and arrangement for treating filtrate after oxygen delignification of chemical pulp cooked to a high kappa number. Finland- Patent number 2012-FI50272 2012127111, Andritz Oy, Finland.Google Scholar
Violette, S. (2003) Oxygen delignification kinetics and selectivity improvement PhD Theis, University of Maine.Google Scholar
Violette, S., van Heiningen, A. (2002). Selectivity improvement during oxygen delignification by adsorption of polymeric additives. International Pulp Bleaching Conference, Portland, OR, United States, May 19–23.Google Scholar
Vuorinen, T., Fagerstrom, P., Buchert, J., Tenkanen, M., Teleman, A. (1999) Selective hydrolysis of hexenuronic acid groups and its application in ECF and TCF bleaching of kraft pulps. J. Pulp Pap. Sci. 25:155–162.Google Scholar