Fractionation of thermomechanical pulp in pressure screening

An experimental study on the classification of fibres with slotted screen plates

Ari Ämmälä

Department of Process and Environmental Engineering, University of Oulu

Abstract

Pressure screening, nowadays the most widely used method for cleaning pulp, has been traditionally investigated as a debris removal process. The aim of this thesis, however, was to study it with a view to the fractionation of pulps, examining systematically and extensively the effects of screening parameters on fractionation under actual working conditions in order to provide an insight into its possibilities and limitations as a fractionation method. The experimental work was performed with a full-scale two-stage pressure screen connected to an industrial TMP process. Fractionation of the pulp was analysed in terms of consistency, freeness, optical fibre length distribution, coarseness index and Bauer-McNett fractions. Two sampling systems were used during the screening experiments, manual and semiautomatic. The latter was assessed to be more reliable, as reflected in lower stochastic variation and the absence of a systematic bias in the mass balance errors over the screen. The poorer reliability of the manual sampling system was offset by the large number of screening tests, however.

The results of the screening experiments showed that with a given design of the screen plate, the separation of each fraction was dependent almost exclusively on the mass and volumetric reject rates. The mass flow of fines, defined as the Bauer-McNett P200 fraction, was dependent mostly on the volumetric reject rate, while the mass flow of fibrous fractions (R200, R50, R30, R16) depended mostly on the mass reject rate. The mass reject rate obtained in pressure screening was a result of the choice of operating parameters, but fractionation efficiency could not be affected by using different combinations of these parameters (feed consistency, rotor tip speed and slot velocity) if the mass and volumetric reject rates were kept constant. The slot width together with the contouring of the screen plate affected the fractionation efficiency as compared with the situation at constant mass and volumetric reject rate. Increased fractionation was obtained by reducing the slot width and contouring. The pulp passage ratio, which combines the mass and volumetric reject rates into one parameter, was found to be a expedient way of expressing the fractionation of pulp, as it was possible to present fractionation uniformly as a function of this ratio. The change in freeness was found to correlate quite well with that in Bauer-McNett fractions, and it was a good indicator of fractionation efficiency in screening. Apart from fractionation according to length (or Bauer-McNett fractions), the slotted pressure screen was also found to classify the fibres according to their coarseness. The coarseness difference was partially dependent on the fibre length, but additionally the coarseness in the accept pulp for any given fibre length class was always lower than that in the reject pulp. The difference obtained seemed to depend on the passage ratio of the pulp.

This thesis provides new information for the modelling of pulp quality and the design of fractionation experiments, fractionation processes and screen room control strategies.

Table of Contents
Acknowledgements
1. Introduction
1.1. Background
1.2. Problems for investigation
1.3. Research assumptions
1.4. Hypotheses to be tested
1.5. Research environment
1.6. Outline of the thesis
2. Fractionation of pulp by pressure screening
2.1. General
2.2. Pressure screen design
2.3. Parameters affecting screening performance
2.3.1. Classes of parameters
2.3.2. Design parameters
2.3.3. Operating parameters
2.3.4. Furnish parameters
2.4. A mathematical approach to screening performance
2.4.1. Development of equations
2.4.2. Basic equations
2.4.3. Screening efficiency
2.4.4. Passage of pulp
2.4.5. Length-based passage of fibres
2.5. Fractionation ability of pressure screens
3. Numerical characterisation of fractionation
3.1. General
3.2. Fractionation yield
3.3. Passage ratio
3.4. The fractionation index of Karnis
3.5. The fractionation index of Olson et al.
3.6. Q-index
4. Experimental work
4.1. Outline of the experimental work
4.2. Analytical methods for characterising the fractionation of pulp
4.2.1. Storage of pulp samples
4.2.2. Consistency
4.2.3. Freeness
4.2.4. Bauer-McNett classification
4.2.5. Modelling of Bauer-McNett fractions
4.2.6. Optical fibre length and coarseness
4.2.7. Coarseness index by image analysis
4.2.8. Summary of laboratory analyses
4.3. Effects of operating parameters on fractionation
4.4. Effects of screen plate design on fractionation
4.5. Role of fibre coarseness in fractionation
4.6. Role of sampling in experimental work
4.6.1. Experimental arrangement
4.6.2. Statistical analysis
5. Results
5.1. Effects of operating parameters on fractionation
5.1.1. Presentation of the results
5.1.2. R16 fraction
5.1.3. R30, R50 and R200 fractions
5.1.4. P200 fraction
5.2. Effects of screen plate design on fractionation
5.2.1. Presentation of results
5.2.2. Bauer-McNett fractions
5.2.3. Freeness
5.3. Role of fibre coarseness in fractionation
5.3.1. Presentation of results
5.3.2. Coarseness as measured by different methods
5.3.3. Effect of fibre length on corrected coarseness
5.3.4. Coarseness index in different fibre length classes
5.3.5. Relative effects of fibre length and coarseness on screening
5.4. Role of sampling in experimental work
6. Discussion
6.1. Effects of operating parameters on fractionation
6.2. Effects of screen plate design on fractionation
6.2.1. Bauer-McNett fractions
6.2.2. Freeness
6.3. Role of fibre coarseness in fractionation
6.4. Role of sampling in experimental work
6.4.1. Background
6.4.2. Sampling methods and phenomena affecting sampling
6.4.3. Analytical error
6.4.4. Manual vs. semi-automatic sampling
7. Concluding remarks
8. Recommendations for future work
References
1.
B. The Valmet TAP03 modified laboratory screen
C. Example calculation of the fractionation index of Karnis
List of Tables
1. Experimental testing of hypotheses.
2. Laboratory analyses and standard methods used in the experimental work.
3. Key data on the fed pulp in the pressure screening experiments.
4. Operating ranges used in the 71 screening tests.
5. Typical Bauer-McNett distribution of TMP in the feed of the first screening stage at a freeness of 150 ml, and in that of the second stage at a freeness of 340 ml.
6. Screen plate designs.
7. Operating ranges used in the 89 screening tests.
8. General information on the screen plate designs and screening conditions used in the experiments.
9. Pipe diameters and the range of flow parameters in the test series with manual and semi-automatic sampling valves.
10. Fibre lengths and coarseness values determined by different methods, data from test series 1a and 1b. 5 experiments in each series.
11. Analysis results of the test series 2, 3 and 4.
12. Statistics on the mass balance errors.
13. Results of example calculations for different sampling systems at the 95% confidence level.
A3.1. Example of the calculation of the cumulative curve and fractionation index of fibre length.
List of Figures
1. A modern pressure screen (Hautala et al. 1999).
2. General constructions for the rotor: a. semi-open, b. closed, c. open, showing F = feed, A = accept, R = reject, f = hydrofoil, and M = motor (Niinimäki et al. 1996a).
3. Rotor types with different pulsation elements. A: foils, B: bumps, C: radial vanes, and D: tapered surface (Bliss 1990).
4. A wedge wire screen basket. W = wire width, S = slot width and P = contour (profile) height (Kleinhappel et al. 1995).
5. Screening efficiency as a function of the mass reject rate (based on Eq. 6).
6. Material balance around an annular element, shown in cross-section (Gooding & Kerekes 1989).
7. Schematic illustration of the definition of the fractionation index (Karnis 1997).
8. Effect of hot disintegration on freeness at 85°C (Ämmälä et al. 2000a).
9. Comparison of two Bauer-McNett apparatuses at three level of freeness.
10. Correlation of fractions between results obtained with two apparatuses.
11. Verification of modelled and measured Bauer-McNett fractions.
12. Correlation between corrected coarseness and the coarseness index. The data points are from the feed, accept and reject pulps of test series 1a, 1b and 3 (see section 4.5).
13. Flow diagram showing the connection of the two-stage pressure screen to the mill process.
14. Schematic diagram of the semi-automatic piston valve used in the experiments (Reisto 1990).
15. Schematic illustration of the two-stage screen used in the experiments.
16. Yield of the R16 fraction in the accept flow as a function of the mass reject rate. 1A denotes the accept flow in the first screening stage, and 2A that in the second screening stage. RRv denotes the average volumetric reject rate of the test series. The bold line denotes the mass accept ratio.
17. Yield of the R30 fraction in the accept flow as a function of the mass reject rate.
18. Yield of the R50 fraction in the accept flow as a function of the mass reject rate.
19. Yield of the R200 fraction in the accept flow as a function of the mass reject rate.
20. Yield of the P200 fraction in the accept flow as a function of the mass reject rate.
21. Yield of the P200 fraction in the accept flow as a function of the volumetric reject rate. RTF denotes the average reject thickening factor of the test series. The bold line denotes the volumetric accept ratio.
22. Passage ratios of the R16 and R30 fractions as functions of that of the pulp in the first screening stage. The bold line denotes the overall passage ratio of the pulp.
23. Passage ratios of the R16 and R30 fractions as functions of that of the pulp in the second screening stage.
24. Passage ratios of the R50, R200 and P200 fractions as functions of that of the pulp in the first screening stage.
25. Passage ratios of the R50, R200 and P200 fractions as functions of that of the pulp in the second screening stage.
26. Fractionation index of Karnis for fibre length in the first screening stage.
27. Fractionation index of Karnis for fibre length in the second screening stage.
28. Normalised accept freeness as a function of the passage ratio in the first screening stage for different screen plate designs.
29. Freeness of the accept and reject flows modelled with Eq. 32 vs. measured freeness.
30. Corrected coarseness as a function of length-weighted fibre length. The data points are from the feed, accept and reject pulps of test series 1a and 1b.
31. Coarseness index as a function of fibre length in pulp from the first screening stage at mill A (test series 2). The results are averages from 6 experiments.
32. Fig. 32. Coarseness index as a function of fibre length in pulp from the first screening stage at mill B (test series 3). The results are averages from 5 experiments.
33. Coarseness index as a function of fibre length in pulp from laboratory screening. Fed pulp from the first screening stage of mill A (test series 4). The results are averages from 11 experiments.
34. Relative differences in the coarseness index (total and independent of fibre length) between the accept and reject streams as functions of the respective differences in length-weighted fibre length.
35. Relative coarseness index differences (total and independent of fibre length) in the screening test series as functions of the passage ratio of the pulp. Notations next to the test points refer to the screen plate used.
36. Calculated mass balance error in the test series with manual ball valves and semi-automatic piston valves.
37. Relative effect of cross-sectional dimensions of fibres on flexibility.
38. Some typical arrangements for sampling pipes.
A1.1. Front view of the screening unit.
A1.2. Rear view of the screening unit.
A2.1. General view of the laboratory screen.
A3.1. Cumulative curves for fibre length in pressure screening.
A3.2. Fractionation index of fibre length.
  Next
  Acknowledgements

Homepage of this publication

Library Units | Collections | Databases | Library News | Library Services | Electronic Collection | Links elsewhere | Alphabetical Index

© 2000 Oulu University Library