on heat and paper: from hot pressing to impulse technology9246/fulltext01.pdf · on heat and paper:...

On Heat and Paper:

From Hot Pressing ToImpulse Technology

Marco F. C. Lucisano

Doctoral Dissertation

Stockholm 2002Royal Institute of TechnologyDepartment of Fibre and Polymer TechnologyDivision of Paper TechnologySE-100 44 Stockholm, Sweden

mailto:[email protected]

Typeset in LATEX.

TRITA-PMT REPORT 2002:10ISSN 1104-7003ISRN KTH/PMT/R- -2002/10- -SE

Akademisk avhandling som med tillstand av Kungliga Tekniska Hogskolan iStockholm framlagges till offentlig granskning for avlaggande av teknologiedoktorsexamen torsdagen den 12 december 2002 klockan 10:00 i STFI-salen,Drottning Kristinasvag 61, Stockholm. Avhandlingen forsvaras pa engelska.

Copyright © Marco Lucisano 2002Stockholm 2002, Kista Snabbtryck AB

O buono Appollo, a l’ultimo lavorofammi del tuo valor sı fatto vaso,

come dimandi a dar l’amato alloro.Infino a qui l’un giogo di Parnaso

assai mi fu; ma or con amenduem’e uopo intrar ne l’aringo rimaso.

Entra nel petto mio, e spira tuesı come quando Marsıa traestide la vagina de le membra sue.O divina virtu, se mi ti presti

tanto che l’ombra del beato regnosegnata nel mio capo io manifesti,

vedra’mi al pie del tuo diletto legnovenire, e coronarmi de le foglie

che la materia e tu mi farai degno.Sı rade volte, Padre, se ne coglie

per trıunfare o cesare o poeta,colpa e vergogna de l’umane voglie,

che parturir letizia in su la lietadelfica deıta dovria la fronda

peneia, quando alcun di se asseta.Poca favilla gran fiamma seconda:

forse di retro a me con miglior vocisi preghera perche Cirra risponda.

(Dante Alighieri, Paradiso, I, 13–36)

Lucisano, Marco F. C. (2002)On Heat and Paper: From Hot Pressing to Impulse Technology

STFI, Swedish Pulp and Paper Research InstituteBox 5604, SE-11486 Stockholm, Sweden

Abstract

Impulse technology is a process in which water is removed from a wet paperweb by the combined action of mechanical pressure and intense heat. Thisresults in increased dewatering rates, increased smoothness on the roll sideof the sheet, and increased density. Although the potential benefits of im-pulse pressing have been debated over the past thirty years, its industrialacceptance has been prevented by web delamination, which is defined as areduction in the z-directional strength of paper.

This thesis deals with the mechanism of heat transfer with phase changeduring impulse pressing of wet paper. The results of four complementaryexperimental studies suggest that little or no steam is formed in an impulsenip prior to the point of maximum applied load. As the nip is unloadedand the hydraulic pressure decreases, hot liquid water flashes to steam. Weadvance the argument that the force expressed upon flashing can be usedto displace liquid water, in a mechanism similar to that originally proposedby Wahren. Additionally, model experiments performed in a novel exper-imental facility suggest that the strength of flashing-assisted displacementdewatering can be maximized by controlling the direction of steam venting.If this solution could be exploited in a commercially viable impulse press,delamination would cease to be an issue of concern.

The thesis includes a study of the web structure of delaminated pa-per. Here, we characterized delaminated paper by the changes in transversepermeability and cross-sectional solidity profiles measured as a function ofpressing temperature. We found no evidence that wet pressing and impulsepressing induced stratification in non-delaminated sheets and concluded thatthe parabolic solidity profiles observed were due to capillary forces presentduring drying. Further, the permeability of mechanically compressed never-dried samples was found to be essentially constant for pressing temperatureslower than the atmospheric boiling point of water and to increase signifi-cantly at higher pressing temperatures. We propose this to be a result ofdamage to the cell wall material due to flashing of hot liquid water in thefiber walls and lumina.

Finally, we present a method and an apparatus for measurement of thethermal properties of water-saturated paper webs at temperatures and pres-sures of interest for commercial high-intensity processes. After validation,the method was successfully applied to measure the thermal conductivity,

http://www.stfi.se

thermal diffusivity and volumetric heat capacity as functions of temperatureand solids content for water-saturated blotter paper. Here, we found thatthe thermal conductivity increased with solids content in the range from30% to 55%, which is in conflict with the commonly stated assumptionsof a decreasing trend. We propose that this discrepancy could be due tothe thermal conductivity of air-free fibers wetted by unpressable water only,being significantly different from that of dry cellulose.

KEYWORDS: Paper, High-Intensity Presses, Impulse Drying, Heat Trans-fer, Phase Change, Heat-Pipe Effect, Steam, Delamination, Flashing, Per-meability, Z-Direction, Thermal Conductivity, Thermal Diffusivity, HeatCapacity, Resistivity Probes, Visualization.

vi

Preface

Impulse technology is a novel web consolidation technique in which wateris removed from a wet paper web by the combined action of mechanicalpressure and intense heat. Although its potential benefits in terms of highpost-press dryness and better paper quality have been discussed over thepast thirty years, no commercial installation has ever been attempted.

One central issue in any investigation of the physics of impulse technol-ogy is the understanding of the complex interactions of heat with water inthe porous structure of the paper sheet. Yet, liquid-vapor phase change phe-nomena in a wet web undergoing impulse pressing are still largely unknown.Without a complete understanding of the physics of the process, the taskof applying impulse technology to real papermachine operation has requiredthe engagement of three of the largest research institute in the pulp and pa-per field (IPC/IPST, Paprican and STFI) and of numerous other researchgroups.

This thesis presents an investigation of the physics of impulse drying,focusing on the mechanisms of heat transfer with phase change in a water-saturated paper web. The issue of web delamination and the structure ofdelaminated paper have been studied in a combined laboratory and pilotstudy, the results of which are presented in Paper A. Phase change phe-nomena in impulse technology are described in four articles (Papers B–E)presenting a series of experimental studies of heat transfer and flashing evap-oration during an impulse event. Additionally, a new method and apparatusfor the measurement of the thermal properties of water-saturated paper websis presented in Paper F.

The printed version of this thesis is accompanied by a CD-ROM, whichcontains additional material that cannot be transferred on paper with anyavailable printing technique. Computer simulations, animations and filmed

segments are reached by clicking on the icon, in the outer margin

of the page. Mathematical details, computer code and other backgroundinformation that could not be fitted into the main text, can be reached by

clicking on the icon. All direct links to the world wide web have beentested on October 28th, 2002.

http://www.ipst.edu

http://www.paprican.ca

http://www.stfi.se

List of Publications

This thesis is based on the following papers:

A. On the Characterization of the Delamination Process During ImpulsePressingLucisano, M. F. C. and Martinez, D. M. (2001), Nordic Pulp andPaper Research Journal, 16(4):362–368.

B. On the Mechanism of Steam Forming During Impulse Pressing of WetPaper WebsLucisano, M. F. C., Mazzatorta, P. and Martinez, D. M. (2001), NordicPulp and Paper Research Journal, 16(4):355–361.

C. The Role of Evaporative Dewatering in Impulse PressingLucisano, M. F. C., Petrini, J. B. and Martin, A. R. (2003), TAPPIJournal, 2(2).

D. An Experimental Investigation of Liquid-Vapor Phase Change Phe-nomena in Impulse Technology. Part I: Resistivity Probe Measure-ments.Martin, A. R. and Lucisano, M. F. C. (2002), submitted for publicationto Journal of Pulp and Paper Science.

E. An Experimental Investigation of Liquid-Vapor Phase Change Phe-nomena in Impulse Technology. Part II: Visualization Trials.Lucisano, M. F. C. and Martin, A. R. (2002), submitted for publicationto Journal of Pulp and Paper Science.

F. Characterizing the Thermal Properties of Wet Paper Webs with aHeat-Pulse MethodLucisano, M. F. C. and Land, C. (2002).

ix

Other relevant publications not included in the thesis:

1. Characterizing the Changes in Transverse Permeability and Cross-Sectional Solidity Profiles During Impulse Pressing of PaperLucisano, M. F. C. and Martinez, D. M. (2000), Progress in PaperPhysics Seminar, ENPG, Grenoble.

2. On the Mechanism of Heat Transfer with Phase Change During Im-pulse Pressing of PaperLucisano, M. F. C. (2000), Licentiate Thesis, Department of Pulpand Paper Chemistry and Technology, Royal Institute of Technology,Stockholm.

3. The Role of Evaporative Dewatering in Impulse PressingLucisano, M. F. C., Petrini, J. B. and Martin, A. R. (2002), TAPPIPaper Summit, Atlanta, GA.

4. Characterizing the Thermal Properties of Wet Paper Webs with a HeatPulse MethodLucisano, M. F. C. and Land, C. (2001), Progress in Paper PhysicsSeminar, ESPRI, Syracuse, NY.

5. An Experimental Investigation of Flashing in Water Saturated PorousMedia with Application to Impulse TechnologyLucisano, M. F. C. and Martin, A. R. (2002), Progress in Paper PhysicsSeminar, ESPRI, Syracuse, NY.

x

Contents

1 Background 1

1.1 Basics of Wet Pressing . . . . . . . . . . . . . . . . . . . . . . 2

1.2 High-Intensity Pressing and Drying Processes . . . . . . . . . 4

2 Impulse Technology 6

2.1 Dryness and Energy Efficiency . . . . . . . . . . . . . . . . . 7

2.2 Density and Strength Properties . . . . . . . . . . . . . . . . 9

2.3 Surface and Printing Properties . . . . . . . . . . . . . . . . . 10

3 The Physics of an Impulse Event 12

3.1 Hot and Superhot Pressing . . . . . . . . . . . . . . . . . . . 13

3.2 Evaporative Dewatering . . . . . . . . . . . . . . . . . . . . . 13

3.3 Steam-Assisted Displacement Dewatering . . . . . . . . . . . 16

4 Runnability Issues and Technical Solutions 18

4.1 Delamination . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

4.2 Delamination and Hot Nip Physics . . . . . . . . . . . . . . . 20

4.3 Nip Unloading and Delamination . . . . . . . . . . . . . . . . 22

4.4 Technical Solutions for Delamination Control . . . . . . . . . 23

5 Summary of the Literature 26

6 Objectives 26

7 Delamination During Impulse Pressing 27

7.1 Changes in Web Solidity . . . . . . . . . . . . . . . . . . . . . 28

7.2 Permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

xi

8 The Role of Phase Change Phenomena in the Physics ofImpulse Technology 33

8.1 Dewatering Performance . . . . . . . . . . . . . . . . . . . . . 33

8.2 Interfacial Contact Phenomena . . . . . . . . . . . . . . . . . 37

8.3 Thickness-Direction Profiles . . . . . . . . . . . . . . . . . . . 39

8.3.1 Temperature Profiles . . . . . . . . . . . . . . . . . . . 42

8.3.2 Resistivity Probes - Local Phasic Composition . . . . 44

8.3.3 Z-Directional WRV Gradients . . . . . . . . . . . . . . 46

8.4 Flashing-Assisted Displacement Dewatering . . . . . . . . . . 48

8.4.1 STFI’s FlashLab . . . . . . . . . . . . . . . . . . . . . 49

8.4.2 Visualization Experiments . . . . . . . . . . . . . . . . 51

8.4.3 Displacement Dewatering as a Heat Transfer Mechanism 53

9 Thermal Properties of Water-Saturated Paper 55

9.1 The Transient Line Source Technique . . . . . . . . . . . . . . 55

9.2 Thermal Properties of Water-Saturated Paper . . . . . . . . . 57

10 Summary and Conclusions 61

11 Recommendations for Further Work 63

12 Acknowledgments 65

13 Notation 66

xii

A whole is that which has beginning, middleand end.

Aristotle (384 BC–322 BC)

1 Background

The earliest recording of paper manufacture has been credited to the Chineseminister Ts’ai L Lun, who filtered a suspension of silk and mulberry bark ina bamboo mold in 105 B.C. Yet, the first industrial papermachine was notdeveloped before the 19th century. Since this time, papermachines have beencontinually under development but the individual unit operations withinthe machine have remained, quite remarkably, the same. A papermachinecomprises three sections: the wire section, the press section and the dryingsection (Figure 1). In the first step, a jet of stock, i.e. a fibre-water The Principles of

Papermakingsuspension, emerging from the headbox impinges on a permeable movingwire. Water is forced through the wire by the hydrodynamic pressure createdby the momentum of the jet, by gravity or by the application of vacuum.Thereupon, the wet paper web is mechanically pressed or squeezed, removingmost of the unbound excess water. In the third step, the paper web ispassed over steam-heated cylinders where most of the remaining water isevaporated.

The drying process is a massive energy consumer as two thirds of the to-tal energy consumption for the paper machine is connected to the operationof the dryers (McConnell, 1980). Moreover, Figure 1 illustrates that thetotal mass of water removed in the dryer section is very moderate comparedto the dewatering capacity of an entire papermachine. As investment costsare proportional to the number of drying cylinders (sometimes more than120 cylinders), a reduction of their number would mean substantial savingsin capital costs. Consequently, great effort has been put into optimizing the

Marco Lucisano

Figure 1: Water removal in a Fourdrinier papermachine (Krook et al.,1996).

dryer section, since even a small improvement of the process can lead tolower capital and operational costs.

Another area experiencing extensive development is the press section: infact, as a rule of thumb, a 1% increase in solids content after the pressesreduces the drying load by 4 to 5% (TAPPI, 2000). Here, a number of pro-The Press Sectioncess alternatives, ranging from extended press nips to high-temperature pro-cesses, have been introduced throughout the years. This thesis concentrateson impulse technology1, a consolidation process in which the wet pressingand drying operations can be combined into a single event by pressing thesheet in a nip formed between a felt-covered shoe and a solid roll, externallyheated to temperatures exceeding 200 ◦C.

1.1 Basics of Wet Pressing

Wet pressing is a unit operation in which the wet paper web is dewateredthrough volume reduction. This occurs in the press section, where the paperweb is compressed between two permeable felts or between a felt and solidsurface. Here, water is expressed from the interfibre pores and the fiberwalls and pressed into the felt (Carlsson et al., 1977). Pressing technology

1“Impulse Technology” has been given many names since its inception in the early1970’s. Besides “Impulse Technology” (STFI) and “Impulse Drying” (IPC), Luiten (2001)reports “Impulse Pressing” (Albany International), “High Temperature Pressing” (Beloit)and “High-Efficiency Drying” (Beloit). In spite of the variety of the terminology adopted,an analysis of the literature reveals virtually no difference among the different technical so-lutions or the underlying physics. Here, we will use “Impulse Drying”, “Impulse Pressing”and “Impulse Technology” without further distinction for any web consolidation processultimately originating from the initial idea by Wahren (1978).

2

http://www.stfi.se

http://www.ipst.edu

On Heat and Paper: From Hot Pressing to Impulse Technology

has undergone a rapid evolution since its inception and the main advancesin both machinery and scientific understanding have been reviewed in theliterature, e.g. Wrist (1964), Wahlstrom (1969), Ceckler and Thompson(1982), MacGregor (1989), Wahlstrom (1991), Danielsson (1998), Paulapuro(2001) and Wahlstrom (2001).

Comprehensive descriptions of the physics of wet pressing have beengiven by Campbell (1947), Wahlstrom (1960a,b, 1969), Nilsson and Larsson(1968), MacGregor (1989) and Paulapuro (2001). They described a pictureof pressing in which the externally applied pressure is counterbalanced by theadditive sum of the stress in the fiber network and the hydraulic pressure ofwater in the web. This idea, originating from Terzaghi (1944), was applied tothe compression behavior of paper webs by Campbell (1947). Although theapplicability of Terzaghi’s principle as a tool for quantitative predictionshas been questioned by Kataja et al. (1995), it still constitutes the basisfor our understanding of pressing. As a result, press nips are traditionallydivided in two categories: (i) in the first case, the press nip is thought to be Compression-Limited

Nips“compression-controlled”. Here, the mechanical stress in the fiber networkis the dominating factor and the maximum web dryness is determined bythe applied pressure, with no effect of the pressing time. On the other hand,(ii) the nip can be considered “flow-controlled” when the viscous resistancebetween the water and fibers controls the amount of dewatering. Here, webdryness increases with the residence time in the nip.

As a consequence of the applicability of Terzaghi’s principle to flow-limited nips, the web layers closer to the felt are compacted first, with themaximum density at the sheet-felt interface. MacGregor (1983, 1989) de- Flow-Limited Nipsscribed this phenomenon as stratification and its existence has been observedin laboratory experiments (Burns et al., 1990, 1993). Moreover, the result-ing density profile in the thickness direction of the sheet has been measuredin dry paper (Szikla and Paulapuro, 1989; Szikla, 1992). On the other hand,densification in a compression-limited nip is essentially uniform across thethickness of the paper web, giving rise to a uniform density profile.

As the speed of modern papermachines increased throughout the years,a number of innovative web consolidation techniques has been presented Extended Nip

Presses (ENP)and discussed in the literature as complements or alternatives to the tradi-tional press configuration (Sprague, 1985, 1988; Cutshall, 1994; Wahlstrom,1991; Baker, 1999). One example of this is the extended press nip, in whichone of the press rolls is replaced with a concave press shoe. The shape ofthe pressure pulse can be controlled and the residence time in the pressnip is increased, resulting in significant dryness increases especially for flow-controlled nips (Wahlstrom, 1991). The introduction of extended nip presseshas proven so beneficial in terms of papermachine operation and productproperties, that they have become standard elements of modern press sec-

3

Marco Lucisano

tions.

1.2 High-Intensity Pressing and Drying Processes

High-intensity processes constitute an important class of novel consolida-tion techniques based on the interactions between water, fibers and thermalenergy. The common feature to all high-intensity processes is the use ofthe positive effects of increased web temperature on both dewatering rateand web consolidation: (1) the viscous resistance of water is reduced withincreasing temperature (Harper, 1955); (2) the compressibility of the fibermaterial is increased (Norberg and Back, 1967; Wiseman, 1976; Back, 1979;Andersson and Back, 1981; Lobosco and Kaul, 2001). The fundamentalHigh Temperature

Effects onDewatering

ideas behind the use of high-intensity presses originate from Mason’s pro-cess for the production of hardboard and to successive applications to verycoarse thermomechanical pulps (Mason, 1931; Mason et al., 1937; Mason,1938a,b; Mason et al., 1939). The first applications to continuous paper-making processes came in the 1950’s, when it was observed that post-pressdryness increases with increasing web temperature (Harper, 1955) and adevice was invented to heat up the wet web just before it enters the presssection (Mazer, 1956, 1957). Since then, four main process alternatives havebeen developed (Cutshall, 1993): (i) steam showers in the press section(McDiarmid and Cutshall, 1982; Cutshall, 1994; Pattersson and Iwamasa,1999), (ii) hot presses (Walker, 1990; Fougler, 1991; Langdon, 1991), (iii)press drying (Back, 1984) and (iv) impulse drying (Wahren, 1978).

The term “press drying” refers to processes in which paper is dried underrestraint due to the application of pressure in the z-direction during part orPress Dryingall of the drying process while, at the same time, the paper is in contactwith a very hot drying surface, so that the temperature of the wet web isconsiderably in excess of 100 ◦C (Back, 1984).

The physics of press drying has been thoroughly described in the liter-ature and several reviews are available (e.g. Back, 1984; Michell, 1984; We-ston, 1985; White, 1986). The most intensive period of research and develop-ment of press drying took place in the 1970’s with the parallel programmesof the Swedish Forest Products Research Laboratory (STFI), the ForestProducts Laboratory in Madison, WI and at St. Anne’s Board Mill. Ad-ditionally, Lehtinen followed the first direct attempts to apply press-dryingto paper production and proposed a process in which the web is dried un-der mechanical pressure between two impermeable steel bands, one of whichCondebeltis steam-heated while the other is cooled by water (Lehtinen, 1978, 1984,1995a,b). The process has now been commercialized as Condebelt, and thereare two industrial installations (Retulainen and Hamalainen, 2000; Houvila,2001).

4

http://www.stfi.se


Four main independent factors appear to control the final strength char-acteristics of press dried products as well as the overall process performance(Lehtinen, 1986a,b; Back and Swenson, 1981; Back, 1986; Paulapuro, 1988):(i) the z-directional mechanical pressure on the wet web, (ii) the moisturecontent in the web, (iii) the while-wet dwell time and (iv) the temperatureof the web. An independent control of these parameters would allow com- Press Drying:

Four ControlParameters

plete fine tuning of the operation of all high-intensity pressing processes.In particular, the dwell time of the web within a certain moisture rangeappears to have a decisive influence on the strength characteristics of theproduct (Lehtinen, 1986a). Yet, design considerations create links betweenthe four governing factors in all laboratory test equipments and productionmachines.

The similarities between impulse drying and press drying have been dis-cussed in the literature since the inception of the technique (Sprague, 1987; Press Drying and

Impulse DryingBack, 1990; Sparkes and Poirier, 1990a; Back, 1991a,c; Pikulik et al., 1995),since both processes are based on the simultaneous use of intense heat andmechanical pressures to enhance sheet consolidation and dewatering. Yet,Sprague (1987) argues that the differences in operating conditions, mecha-nisms of dewatering and densification and technical challenges on the waytowards a commercial installation are of such central character that impulsedrying and press drying are to be treated as distinct processes. In spite ofthese differences, it seems clear that most phenomena discussed in connec-tion to press drying have some relevance in the investigation of the physicsof impulse technology.

5

Marco Lucisano

Get your facts first, and then you can distortthem as much as you please.

Mark Twain (1835–1910)

2 Impulse Technology

Impulse technology is a web consolidation technique in which water is re-moved by the combined action of mechanical pressure and intense heat. InImpulse technologyits present form, the process has been developed from the ideas proposedindependently and for different constructive solutions by Gottwald et al.(1967) and Wahren (1978) and later named “Impulse Drying” (Arenanderand Wahren, 1983; Wahren, 1983). In impulse technology, wet pressing anddrying are combined into a single event by pressing the sheet in a nip formedbetween a felt-covered shoe and a heated solid roll.

In spite of the potential benefits of impulse pressing and after thirty yearsfrom its conception, no commercial installation has ever been attempted. Onthe other hand, impulse technology has enjoyed a widespread exposure withsome 300 publications in the scientific and technical literature. Generalreviews of the impulse technology literature have been published since thelate 1980’s: researchers at Lund University presented an all-around analysisof impulse technology in a series of three reports (Stenstrom, 1989; Persson,1994; Nilsson, 1998). Timofeev and Mujumdar (1993) published a somewhataged bibliography of the literature concerning press drying, Condebelt andimpulse drying. Additionally, the history, geography and economical aspectsof the process development have been summarized by van Lieshout (1999)and Luiten (2001).

Impulse technology combines wet pressing and drying into a single oper-ation by pressing the sheet in a nip formed between a felt-covered shoe anda solid roll, externally heated to temperatures exceeding 200 ◦C. One key

6


Table 1: Comparison between a traditional multicylinder dryer and animpulse unit (Stenstrom, 1990b).

Unit Multicylinder Impulse

Dryer temperature ◦C 100− 150 200− 450Mechanical pressure kPa 2− 5 1000− 5000Sheet temperature ◦C < 100 100− 200Residence time s 0.2− 0.5 0.005− 0.05Drying speed kg/m2 ·h 10− 30 2000− 8000Specific energy use kJ/kgH2O 2800− 3500 500− 2200

difference between traditional multicylinder drying and impulse technologyis that the latter uses energy of higher grade (electricity, gas or high-pressuresteam) to dewater the sheet (see Table 1 for a comparison of some basicengineering variables). Yet, impulse drying has generally been postulated A New Dimension

for Papermakingto be economically advantageous since:

(1) It uses less energy than conventional drying because all water may notneed to be evaporated. This might result in a strong reduction of cap- Dryness and Energyital costs for large drier sections and associated peripheral equipment.

(2) It allows the production of paper with improved mechanical properties. MechanicalProperties

(3) It results in paper with better surface properties. Printing Surface

A wide variety of paper products has been thoroughly investigated in thepast thirty years, in order to find an optimal candidate for a full-scale com-mercial installation. In fact, the literature presents studies focusing on prod-ucts ranging from tissue (Hollmark et al., 1999, 2000) to liner (Horn et al.,1986; Lavery, 1987b; Orloff, 1992a; Orloff and Crouse, 1998; Orloff et al.,2000b, 1999; Orloff, 1998) and from newsprint (Lavery, 1988a; Sprague andLavery, 1988; Sparkes and Poirier, 1990b; Poirier and Sparkes, 1991) to finepaper (Rodden, 1987; Svanholm, 2002). Indeed, if the effects of impulsepressing could be fine-tuned at will, the process could add the “new dimen-sion” to papermaking that was originally envisioned.

2.1 Dryness and Energy Efficiency

The most obvious goal driving the development of high-intensity pressingand drying techniques is the quest for higher drying rates and more energyefficient mechanisms of water removal. The basic effects of temperature onwater removal are well known since the 1950’s: most literature studies show

7

Marco Lucisano

that water removal due to elevated temperatures increases linearly withincreasing heat transfer to the web, which in its turn presents a linear cor-relation with the pressing temperature (e.g. Macklem and Pulkowski, 1988;Karrila et al., 1991; Larsson, 1999). High post-press dryness is desirablesince it results in significant reductions of the drying load, which is of greatPost-Nip

Dryness value for drier-limited machines. Indeed, impulse pressing is characterizedby exceptional drying rates, when compared with the 15− 30 kgH2O/h ·m2

that are typical for cylinder dryers. Sprague (1987) reported water removalrates in the range 1500− 8000 kg/h ·m2 for a virgin kraft linerboard furnishand 2500− 8000 kg/h ·m2 for newsprint, and these values are representa-tive of the ranges commonly reported in the literature. These exceptionaldrying rates are mirrored in high dryness increases after the impulse nip,which Luiten (2001) summarized. Impulse technology results in post-nipdryness increases of 2–20% over the reference pressing technology. Themagnitude of the dryness increase depends largely on the basis weight ofthe specific product, with lighter paper qualities experiencing the highestincreases (Macklem and Pulkowski, 1988).

Although the extent of water removal is central for the industrial applica-tion of impulse drying, energy efficiency is more important for the economicEnergy

Efficiency feasibility of the process. A relevant parameter in this respect is the specificenergy use2 defined as the ratio of the energy requirement to the mass ofwater removed from the sheet (Lavery, 1987a). The specific energy use forimpulse pressing varies from 115 kJ/kgH2O (Orloff and Sobczynski, 1993) to3000 kJ/kgH2O (Larsson, 1999), as compared with 3000− 4500 kJ/kgH2Owhich is representative for a multicylinder dryer (Larsson, 1999; van Lieshout,1999). The specific energy use is a function of the ingoing solids, furnish andoperational parameters such as the peak pressure and pressing temperature.

In order to investigate the impact of the installation of an impulse uniton the total energetic balance of a papermachine, Stenstrom (2000) com-Energy

and Economy pared the total specific energy use for two cases: (i) a traditional presssection (composed of three nips) followed by a cylinder dryer and (ii) twotraditional press nips followed by an impulse unit and a cylinder dryer. A pa-permachine equipped with an impulse unit would need to have a post-pressdryness of approximately 73% in order to match the energy consumptionof a traditional system pressing from 35% to 42%. Whereas this is feasiblewith low basis weight products (e.g. Hollmark et al., 1999, 2000), the eco-nomic disadvantage due to the use of energy of higher grade (and price) hasto be compensated with increased product properties for an economically

2Larsson (1999) and Larsson and Stenstrom (2001) refined the definition of specific en-ergy use by introducing two additional concepts: (i) the total specific energy use (TSEU)defined as the ratio of the energy input to the impulse unit to the total dewatering by allmechanisms and (ii) the impulse specific energy use (ISEU) calculated as the heat trans-ferred to the paper sample divided by the water removal caused by elevated temperatures.

8


feasible application to high grammage grades.

2.2 Density and Strength Properties

The effects of impulse pressing on consolidation and on the development ofsheet density have been thoroughly investigated in the literature ever sincethe process was conceived in the 1970’s. Nonetheless, no exact descriptionof the physics of sheet consolidation in impulse drying is available in theliterature. Moreover, the effect of impulse technology on paper propertiesappears to depend both on the furnish and on the process variables and nounified quantitative description is available in the literature.

It is generally agreed that impulse technology provides additional densi-fication, which can translate in enhanced mechanical properties (e.g. Lavery,1987c,b, 1988a; Sparkes and Poirier, 1990b; Karrila et al., 1991; Orloff et al.,1995, 2000b; Krook et al., 1996; Rigdahl et al., 2000; Martinez et al., 2001;Mendes et al., 2002a). Yet, as the pressing temperature increases, a majorrunnability problem known as delamination occurs, which ultimately resultsin strength loss3.

The dynamic density development during an impulse event has beeninvestigated with laboratory experiments in platen press simulators (Burton, Dynamic

Densification Studies1985; Burton et al., 1986; Burton and Sprague, 1987; Burton, 1987; Burnset al., 1990, 1993; Rudman and Orloff, 2001). They showed that impulsedrying generates a unique density profile in the thickness direction of the webas the well-known stratification phenomena on the felt side of the sheet wasaccompanied by the development of a high-density region at the surface incontact with the hot pressing platen. This resulted in a “U”-shaped densityprofile with local maxima at both paper surfaces.

It is still unclear whether the improved mechanical properties are exclu-sively due to the increased density or if some additional effects are present. Strength:

Press Drying andImpulse Drying

The same question has been discussed for press drying where it has beenargued that there is no difference in the density-strength relation betweenconventional wet pressing and press drying (e.g. Wedel, 1983). Nevertheless,long exposures of wet paper samples to high temperature cause the flow ofboth lignin (> 15 s) and hemicellulose (> 1 s, Setterholm and Benson, 1977;Horn, 1979; von Byrd, 1979). Hemicellulose appears to be the primary factorresponsible for the development of bond strength in press drying, whereaslignin, which by itself has little effect on bonding and strength, protectsfibre bonds from water, thereby enhancing wet strength in high yield pulps.Unfortunately, the literature presents no direct investigation of the effect of

3The physics of delamination and the technical solutions adopted for delaminationcontrol are reviewed in Chapter 4.1

9

Marco Lucisano

impulse pressing on the mode of fiber bonding in the sheet, but nip residencetime is expected to be a central issue in this respect. In fact, laboratory ex-periments show that the use of long nips for impulse pressing of paper resultsin a significant increase in dryness, density and paper strength (Ahrens andAstrom, 1986; Back, 1991c; Orloff and Phelan, 1993; Larsson and Orloff,2000)

Improved mechanical properties are of central importance for the eco-nomic assessment of impulse technology. In spite of the potential energy sav-Paper Strength

and Economy ings, impulse pressing needs to compensate for the economic disadvantageof high grade energy. If the enhanced mechanical strength could be trans-lated into a basis weight reduction, additional economical benefits could beascribed to impulse technology. Although the envisioned reduction of rawmaterial costs has been discussed in the literature, only one study reports apilot scale experiment during which the basis weight could be reduced with-out any loss of functional properties, while production velocity was increased(Orloff et al., 2000a).

2.3 Surface and Printing Properties

1 mm

A

1 mm

B

Figure 2: Micrographs of the surface of (A) liner produced on EuroFEXwith conventional pressing technology and (B) liner pressed at 180 ◦C(Pictures courtesy of Erik Blohm, STFI).

During an impulse event, one surface of the wet paper sheet comes incontact with a hot smooth metal surface. Irrespectively of the effects on theSmoothnessbulk properties of the sheet, the paper surface in contact with the pressingsurface experiences a physical environment which has favorable effects onits smoothness and printability. The literature is unanimous in stating thatthe surface of impulse pressed paper is smoother than that of a control pro-duced by wet pressing (Figure 2, Steen, 1986; Rigdahl et al., 2000; Mendeset al., 2002b). Surface roughness decreases with increasing temperature and

10


pressing impulse, whereas parameters such as pulse shape, platen materialand furnish type have limited influence (Martinez et al., 2001). Addition-ally, Poirier and Sparkes (1991) report that even the roughness on the feltside of the sheet can be reduced by appropriate choice of the felt. Thisis in contrast with most other studies, where an unchanged (or somewhatincreased) roughness of the felt side of the paper is reported (e.g. Rigdahlet al., 2000).

The idea of reducing the surface roughness by pressing wet paper againsta smooth surface of high temperature is not new and can be dated back tothe 1930’s (e.g. Williams, 1932; Schorger, 1934). At that time, it was re- Calendering and

Impulse Technologyalized that calendering could be reduced or eliminated by pressing the wetweb against a hot surface for a sufficient number of times. The perspec-tive to produce paper with little or no calendering is interesting both fromthe economic point of view and because of its implications on the strengthproperties of the product (Rodden, 1987; Poirier and Sparkes, 1991; Orloffet al., 2000b). It is important to note that unlike calendering, which reducessurface roughness and bulk at the expense of sheet strength, impulse dryingcan reduce roughness and bulk with a positive effect on strength.

Impulse pressing is an intrinsically asymmetric process, resulting in avery smooth surface (the one in contact with the hot pressing surface) anda rougher surface (the one in contact with the felt). Indeed, impulse tech- Impulse Pressing:

Asymmetric Processnology shares this asymmetry with all existing and envisioned press dryingprocesses. In view of the asymmetric nature of the process, impulse technol-ogy could be extremely useful for qualities that require one smooth surfaceonly. Additionally, impulse technology could be used to correct undesiredtwo-sidedness, for example to reduce the intrinsic asymmetry of Fourdrinierpaper, thereby giving better printability. Further, the installation of two im-pulse units inverted with respect to each other would provide a possibility totreat both sides of the paper. This configuration is installed on EuroFEX,STFI’s research papermachine, and it was envisioned for the pilot machineat Paprican (Sparkes and Poirier, 1990b,a; Poirier and Sparkes, 1991).

11

http://www.stfi.se/documents/facts/tour.htm#EuroFEX

http://www.stfi.se

http://www.paprican.ca

Marco Lucisano

How often have I said to you that when you have eliminatedthe impossible, whatever remains, however improbable, must be thetruth?

Sir Arthur Conan Doyle (1859–1930, The Sign of Four)

3 The Physics of an Impulse Event

Impulse pressing is similar in nature and mode of operation to traditionalwet pressing technologies. Yet, its characteristic feature is the simultaneousaction of intense thermal energy and mechanical pressure. This results inincreased smoothness on the roll-side of the sheet (Burton et al., 1986; Lav-ery, 1987a, 1988b) and decreased bulk (Sprague and Burton, 1986; Lavery,1987a). Moreover, the specific energy use for water removal in excess oftraditional wet pressing operation has been reported to be lower for impulsepressing than for classic drying operations (e.g. Devlin, 1986; Larsson, 1999;Nilsson, 2001a).

The high heat flux and water removal rates experienced with impulsedrying suggest that the mechanisms that control dewatering differ signifi-cantly from those of conventional pressing and drying operations. We referThe Impulse Effectto any non-traditional dewatering mechanisms occurring in connection withan impulse event as “impulse effect”.

An analysis of the literature published in the field suggests that threedistinct modes of water removal can be active during an impulse event:

(i) Hot and superhot pressing, i.e. dewatering by volume reduction, en-hanced by temperature effects on network compressibility and waterviscosity.

12


(ii) Evaporative dewatering including flashing and the heat-pipe effect, inwhich thermal energy is used to evaporate water.

(iii) Steam-assisted displacement dewatering in which liquid water is dis-placed by the action of a vapor phase.

These three main dewatering mechanisms are reviewed briefly in the follow-ing sections.

3.1 Hot and Superhot Pressing

Hot pressing techniques can be defined as a family of consolidation strategieswhich enhance dewatering though volume reduction by temperature effectson compressibility and water viscosity. Hot pressing units are run at tem- Two-Phase Systemsperatures below 100 ◦C. Here, the web in the nip can be considered as atwo-phase system (liquid water-solid fibers), at least after the initial com-pression stages during which air is expressed from the sheet. We will term“superhot pressing” all the applications in which the wet web is essentiallya two-phase system, even if hot-roll temperatures exceed 100 ◦C.

The role of heat in hot and superhot pressing has been reviewed inChapter 1.2. The same phenomena are generally considered to be active inimpulse pressing (e.g. Wahren, 1982; Arenander and Wahren, 1983; Back,1991c; Krook et al., 1996). Indeed, Nilsson (2001b) presented a superhotone-dimensional model of impulse pressing and Martinez (1999), Riepen(2000) and Gustafsson et al. (2001) extended a traditional wet pressingmodel to account for the effects of heat transfer by conduction and forcedconvection.

3.2 Evaporative Dewatering

Although hot and superhot pressing should have a central role in impulsetechnology, the temperatures used in an impulse nip are so high that liquid-vapor phase change phenomena are likely to occur under favorable condi-tions. This would result in a three-phase system: liquid water, vapor andfibers. Here we will consider two modes of liquid-vapor phase change: Three-Phase

Phenomena

(i) traditional evaporation or drying and

(ii) flashing.

13

Marco Lucisano

TH

Ts

P=Ps(T

s)

Vapor

Liquid

h(t)x

∞ T0

Figure 3: Evaporation of a liquid in a half-space.

Additionally, more complex mechanisms of heat transfer, such as the heat-pipe effect could be active.

Drying of solids refers to the use of evaporation to remove a liquid froma solid matrix. This is usually accomplished by supplying enough thermalenergy to overcome the latent heat of evaporation of the liquid phase. Mod-elling phase change phenomena with a moving interface of phase transition(also known as “moving boundary problems” or “Stefan problems”) is acomplex task. Insight into these phenomena can be gained from Stefan’sDryingoriginal studies on the heat flux in a medium that undergoes phase changeat some fixed temperature (Ts), accompanied by the absorption or libera-tion of latent heat (Rubinstein, 1971). In our case, we consider a column ofwater heated from the top where a layer of steam is forming4 (Figure 3).If heat is transferred by conduction alone, the solution to Stefan’s problemin one dimension is given by Neumann’s solution (Hill, 1987):

The NeumannSolution

Tv(x, t) = TH +Ts − TH

erf( ε2√

αv)erf(

x

2√

αvt), (1)

Tl(x, t) = T0 +Ts − T0

erfc( ε2√

αl)erfc(

x

2√

αlt), (2)

where Tv and Tl are the temperatures in the vapor and liquid phase, respec-tively, T0 is the original temperature of the body, TH is the temperatureof the heated source, α is the thermal diffusivity of each phase, and ε is aconstant related to the rate of growth of the moving boundary between theliquid and vapor phases. The velocity of the propagating interface (h(t)),

4Here we assume that the two phases, liquid and vapor, have the same (constant)density.

14


i.e. the rate of drying, is found to be proportional to the square root of theelapsed time, i.e. h ∝ t−

12 . This analysis has been extended to a variety

of problems (Meirmanov, 1992) and has been applied to porous media (e.g.Gupta, 1974; Cho, 1975).

Flashing, or flash evaporation, is a second mode of evaporation, whichhas a central importance in impulse technology. The process of flash evapo- Flashingration occurs when a liquid is exposed to a pressure lower than the saturationpressure at its temperature. In contrast with other evaporation processesfor which heat is supplied to the liquid from the outside, sensible heat storedin the liquid as superheat is converted into latent heat of evaporation duringflashing. In the case of impulse pressing, the thermal energy required forflashing, is transferred to the sheet during the initial phases of the press-ing event. Indeed, Macklem and Pulkowski (1988) proposed that impulsepressing can be interpreted as a combination of wet pressing (taking placein the nip) and rapid flashing evaporation, occurring when the mechanicalload is reduced. According to this interpretation, all or most of the watertrapped in the web remains in liquid phase in the press nip, in spite of thehigh temperatures recorded in direct experiments. This is possible becauseof the high hydrodynamical pressures generated during the rapid mechanicalloading of the wet web (Campbell, 1947; Wahlstrom, 1960a,b, 1969; Nilssonand Larsson, 1968).

The heat-pipe effect is a non-conventional convective mechanisms ofhigh efficiency. Heat-pipes have been considered in previous works asso- The Heat-Pipe Effectciated with high-intensity pressing (Lehtinen, 1992; Riepen, 2000; Spragueand Burton, 1986), and several authors have speculated that they mightbe the reason for the high heat fluxes measured during impulse pressing(Lindsay and Sprague, 1989; Lindsay, 1989; Ramaswamy, 1990; Ramaswamyand Lindsay, 1998). Indeed, Dreshfield and Han (1956) showed that cyclicevaporation-condensation phenomena are active during paper drying. His-torically, the concept of a heat-pipe was introduced by Grover et al. (1964),who showed that heat may be transferred in unsaturated porous media byinternal fluid recirculation. For this to occur, the temperatures at the bound-aries of the porous media must be set such that: the liquid phase boils nearthe hotter end of the porous media and the vapor phase condenses near thecolder end (Figure 4). Capillary forces, gravity and gradients in saturationpressure drive the motion of the fluids. This results in the latent heat ofevaporation being transferred between the ends of the porous media via themotion of the fluid. This cyclic convective mechanism is extremely efficientas it can be used to transfer large quantities of heat with minimal temper-ature differences. The heat-pipe effect has been studied thoroughly in wetporous media by Udell (1983), Udell and Jennings (1984), Udell (1985a),Udell (1985b) and Udell and Fitch (1985) who investigated the thermo-dynamics of phase change in both homogeneous and heterogeneous solid

15

Marco Lucisano

CondensationEvaporation

Q

Capillary Wick

Vapour Transport Q

Figure 4: Schematic of an ideal heat-pipe.

matrices.

Drying, flashing and heat-pipe phenomena are equivalent as to their en-ergy efficiency as dewatering processes. In fact, they all require an energyEvaporative

Dewatering andEconomy

quantity corresponding (at least) to the latent heat of evaporation of liquidwater. Indeed, as far as the economy of impulse drying is concerned, evapo-rative drying is a highly undesirable mode of dewatering in an impulse nip.In impulse technology, the abundantly available low-pressure steam of a tra-ditional dryer is replaced with high-temperature, high-grade energy (gas orelectricity). The introduction of impulse technology would be economicallyunviable if both processes had the same specific energy use. However, it hasbeen suggested the non-traditional mechanisms could be active in impulsetechnology and that they would result in high energy efficiency.

3.3 Steam-Assisted Displacement Dewatering

The physics of an impulse event is more complex than just the superposi-tion of wet pressing and evaporative drying and the existence of additionaldewatering mechanisms has been assumed.Displacement

DewateringAccording to the original description of the impulse event, steam is

formed in the first part of the press nip as soon as the sheet comes incontact with the heated roll (Wahren, 1978). The resulting steam pressureis expected to act as an extra dewatering force, displacing free liquid waterfrom the wet sheet into the felt (Figure 5). Additionally, the heat trans-fer between the steam, the fibers and the water remaining in liquid phaseis assumed to be so effective that an efficient heat-up of the sheet is ob-tained with little energy losses. In the literature, this theory is referred toas “steam-assisted displacement dewatering” (Wahren, 1982; Arenander andWahren, 1983; Orloff et al., 1998b; Lindsay, 1989). Arenander and Wahren(1983), Devlin (1986) and Rudemiller (1989) report indications that sucha vapor-phase displacement mechanism drives water from the sheet due tovolume expansion upon phase change.

16


HEAT

STEAM FRONT

DISPLACEMENT

OF WATER

Figure 5: Steam-assisted water displacement (Rigdahl et al., 2000): anidealized steam front expels part of the water in the compressed paperweb during the impulse process. This additional mechanism is assumedto remove water in the liquid phase, therefore accounting for the highenergy efficiency of impulse technology.

In most descriptions of the mechanism of impulse technology, steamformation is assumed to take place in the initial portion of the nip, whereits effects on dewatering would be most beneficial. Nevertheless, in the onlydirect attempt to study the existence of vapor fronts, steam layers appearedin the latter portions of the impulse event and their penetration into thethickness direction of the sheet was limited (Zavaglia and Lindsay, 1989).Indeed, laboratory experiments on a press simulator showed a rapid increasein local sheet temperature followed by a quick temperature decrease as thenip opened and and the sheet was unloaded (Orloff et al., 1998b). Althoughthe presence of a steam front in the earlier stages of the impulse event iscontroversial (Back, 1991c; Lindsay, 1991; Back, 1991b), steam formationwhen unloading the hot nip, i.e. flashing, is generally accepted (e.g. vanTilburg (1975a), Back (1991c), Orloff et al. (1998b)).

17

Marco Lucisano

He who laughs has not yet heard the bad news.

Bertolt Brecht (1898–1956)

4 Runnability Issues and Technical Solutions

In spite of the positive effects reported for both sheet properties and en-ergy economy, impulse technology has never passed the pilot trial stage. Infact, the industrial acceptance of the process has been prevented by seriousrunnability problems encountered both in laboratory scale simulations andin pilot scale applications.

The issue of adhesion of the web to the surface of the hot roll has been dis-cussed to some extent in the technical literature (e.g. Crouse, 1997; Orloff,1992a). When the sheet sticks to the roll, this results in surface pickingAdhesion

to the Hot Roll which affects the surface characteristics of the product. Additionally, anyresidues left on the hot roll have to be doctored away to guarantee smoothoperation of the press nip. Adhesion to the press roller is typically lim-ited to a temperature interval extending from approximately 80− 90 ◦C to150− 180 ◦C. Although no complete mechanism of sticking has ever beenoffered and no thorough investigation is available, the low frequency of pub-lished studies suggests that adhesion is not a major issue. Indeed, sinceadhesion appears at temperatures that are too low for effective impulsepressing, a production-scale papermachine might never encounter it.

4.1 Delamination

The most serious hinder to the development of impulse technology is delam-ination, which is defined as a drastic reduction in the z-directional strengthDelamination:

The Stumbling Block of paper (Figure 6). It is generally accepted that delamination occurs when

18


the force dissipated by steam flow is greater than the z-directional strengthof the wet web and that the steam responsible for delamination is generatedby flashing upon unloading of the nip (van Tilburg, 1975b; Crouse and Woo,1989).

Figure 6: Left: Surface of an impulse pressed sheet with a delaminationblister in the center (courtesy of Narcis Mesic). Right: Cross-section of adelaminated sheet, showing a porous domain that divides the sheet in twolayers. Fibres in the upper left corner of the picture are oriented verticallyinstead of horizontally. The orientation of these fibers, together with thefact that the sheet is delaminated, indicates that the force dissipated bythe steam flow exceeded the z-directional strength of the web.

Delamination is not unique to impulse technology and it is often men-tioned in the patent literature concerning heat treatment of wet porousmaterials. Indeed, in patents disclosing a process for manufacturing hard-board, Mason et al. (1939) and Wilcox (1953) describe blistering and explo-sion of the product upon opening of a fully heated press nip, if the pressingstrategy does not provide for venting of the moisture originally present inthe sheet. Additionally, as the first patent correlated to impulse drying(Gottwald et al., 1967) appears to be be a direct evolution of a method forcast coating by contacting paper with a highly-polished roller at high tem-perature (Gottwald and Haigh, 1963), it is interesting to notice that the hotcast-coating process, as originally proposed by Bradner (1929), suffers fromblistering defects, which originate from a violent evolution of steam fromthe aqueous coating color (Hart, 1959).

Delamination is usually characterized by studying the evolution of oneor more strength properties with increasing temperature. A critical delam- Tc,

The CriticalDelaminationTemperature

ination temperature is then defined as the hot roll temperature at whichthe measured strength property starts to decrease upon further increase ofthe pressing temperature. Traditionally, Scott Bond and STFI compressionstrength are chosen to pinpoint the onset of delamination (e.g. Poirier and

19

Marco Lucisano

Sparkes, 1991; Orloff, 1992a; Orloff et al., 1998b; Larsson and Stenstrom,1998). A decrease of sheet density can also be interpreted as a sign of the on-set of delamination. Alternatively, an increase of the standard deviation ofthe measured properties is used to identify incipient delamination (Orloff,1991a, 1992b). Unfortunately, the identification of the critical delamina-tion temperature by investigating the changes in strength properties of thesamples appears to be a highly inconsistent method, as the general trendsobserved are dependent on the specific strength property chosen (Persson,2000).

4.2 Delamination and Hot Nip Physics

The onset of delamination is determined by the physical properties of boththe fibres and the paper web, as well as by the constructive and operativeparameters of the impulse nip. Here, we introduce a classification betweenfibre/web-related parameters and machine parameters. Additionally, phe-nomena of interest in the loading phase of the nip are separated from thoserelevant during nip unloading.

As to the phenomena taking place prior to the point of maximum ap-plied load, the thermal properties of the paper web are likely to be themost important governing parameter related to the web and fibre structure.Loading Phase

Web/FiberParameters

Fibre type, initial temperature and water content should have some influ-ence on the thermal properties of the web. Additionally, the initial webtemperature influences delamination: the critical delamination temperaturedecreases significantly if the web is heated prior to the impulse nip (Poirierand Sparkes, 1991).

The effects of machine-related parameters on phenomena influencing thecritical delamination temperature during nip loading have been thoroughlyLoading Phase

Machine Parameters described in the literature. The influence of pressing temperature on sheetdelamination is self-evident, with more severe delamination at higher tem-peratures. Additionally, the thermal properties of the hot roll have a largeinfluence on the delamination tendency. Hot roll covers with low thermaldiffusivity increase the operating window of impulse technology since theytransfer less heat to the paper sample (< 1 · 10−6 m2/s, Orloff et al., 1992;Orloff, 1994b). In contrast to that, Wedel (1995) argues that the hot rollcover such be such that its high thermal diffusivity allows large heat fluxesinto the sheet.

The quantity:Press RollThermal Mass K =

√ρcpk, (3)

which is known as “thermal mass”, has a large influence on the criticaldelamination temperature since it influences the mode of heat transfer to

20


the wet web. The use of non-porous materials with low thermal mass(K ≤ 3000 W

√s/m2 ·K) has been suggested for delamination control (Orloff,

1991b; Orloff and Lenling, 1993; Orloff, 1994c). Suitable materials for thesurface layer of the hot roll should then be chosen from the group consist-ing of ceramics, polymers, inorganic plastics, glass, composite materials andcermets.

Although the inception and extension of delamination appears to be

Simplified ContactModel

correlated to the total energy transferred to the sheet, no unique value ofa universal critical energy for delamination appears to exist (Phelan et al.,1995). This is not surprising considering that the total energy transferred tothe sheet is not an unique descriptor of temperature profile inside the paperweb.

The applied load impacts the onset of delamination both through theshape of the pressure profile (Crouse et al., 1991; Liang et al., 1999; Orloffand Phelan, 1993; Betz, 1996) and through the absolute level of mechanicalloading. The effect of the press load on the onset of delamination seems Applied Loadintercorrelated both with the material properties of the hot surface and withthe permeability of the sheet (Orloff, 1992a,c; Orloff and Phelan, 1993).

Figure 7: Temperature in the center layer of the web in single-nip pressdrying at different pressing temperatures and pressures (Back and Swen-son, 1981).

Larsson and Stenstrom (1998) report that the delamination tendency isdrastically reduced by maintaining a low linear load and defined a critical Critical Pressurepressure for a given hot surface temperature as the peak pressure where thecreated vapor pressure is low enough not to destroy the paper sheet. Mostlikely, this is correlated with the fact that, for any given mechanical load, amaximum temperature exists, above which water in the wet web evaporatesto steam. With higher hydraulic pressure in the web, the phase changetemperature increases, a phenomenon well-known for press drying of paper(Figure 7).

21

Marco Lucisano

Krook and Stenstrom (1998) divide impulse pressing events in two cat-egories according to the relative magnitude of the temperature of the hotplate and of the saturation temperature at the applied pressure. If (i) the hotsurface temperature is lower than the saturation temperature correspond-ing to the resulting pressure in any part of the web, then the internal sheettemperature is controlled by the temperature of the hot plate. On the otherhand, if (ii) the hot surface temperature is higher than the saturation tem-perature corresponding to the hydraulic pressure in any part of the sheet,then the internal web temperature is pressure controlled. This characteriza-tion is in agreement with the results presented by Betz (1996), who reportedthat delamination can be induced by increasing the mechanical load at thepresses.

4.3 Nip Unloading and Delamination

Phenomena taking place before pressure release are responsible for the pre-requisites needed for delamination to take place, but delamination itselfUnloading Phase

Web/FiberParameters

takes place only upon unloading of the nip (Larsson and Stenstrom, 1998).In order for delamination to take place, at least the layers of the wet webcloser to the hot surface have to reach temperatures exceeding 100 ◦C be-fore pressure is released. If water above 100 ◦C is in the liquid state, i.e.subcooled water, flashing to steam takes place upon unloading of the nip,with a dramatic increase in the specific volume of water. On the other hand,if steam is present inside the web, the volume expansion upon pressure re-duction would not be as violent. Orloff et al. (1998b,c,a) and Orloff (1998)showed that delamination is due to drag forces experienced by the top layersof the paper sample when subcooled water flashing to steam is vented to theexternal atmosphere. Therefore, the flow resistance of the paper sample islikely to have a major influence on the delamination tendency.

The resistance to flow of a porous material is usually characterized byits permeability, an empirical parameter describing a linear relationship be-Delamination and

Permeability tween the flow through a porous bed and the pressure drop driving the flow(Scheidegger, 1974). Many attempts have been made to relate the delam-ination tendency to the permeability of the web (e.g. Orloff and Lindsay,1992a; Boerner, 1992; Orloff and Lindsay, 1992b; Boerner and Orloff, 1994;Orloff, 1994a). Delamination is more severe for sheets of low permeability.As the mode and extend of delignification, refining and pressing control thesheet permeability, it is expected that maximum benefit from impulse dryingwould be obtained for high-yield furnishes that have been minimally refinedand pressed to high dryness levels. Increased beating results in decreaseddelamination temperature (Orloff, 1994a; Mesic, 2002). This suggests thatrefining should be limited to the minimum required for product aesthetics.

22


The basis weight of the web is a further parameter of importance for thedelamination tendency of the sheet (e.g. Crouse et al., 1991; Larsson and Basis WeightStenstrom, 1998), with thicker paper grades being more prone to delami-nation. Given that both the total energy transferred and the temperatureprofile inside the sheet as measured away from the hot surface appear to beindependent of sheet grammage (Krook and Stenstrom, 1998), Larsson andStenstrom (1998) concluded that thicker sheets delaminate more promptlybecause the vapor pressure inside the web at the instant of depressurizationis higher than for thinner webs. This is a result of the lower vapor flowthrough the sheet taking place inside the nip.

4.4 Technical Solutions for Delamination Control

Sheet delamination has proven to be the major stumbling block for thedevelopment of impulse technology as a commercially feasible technique.Therefore, a variety of technical solutions has been developed to preventor control delamination and its negative effects on paper quality and pro-cess runnability (Table 2). Besides the process modifications reviewed inthe previous two sections, the patent literature reports a number of anti-delamination techniques, some of which are summarized here.

The use of chemistry for delamination prevention is mainly connectedto the phenomena taking place during nip unloading, since an appropriate Chemical Additiveschoice of the chemical environment influences both the porous structure ofpaper and its strength in the wet state. Westman et al. (2002) suggestedthat adding a chemical system comprising a polymer component and micro-or nanoparticles to the papermaking stock (before sheet forming) improvesthe runnability of the process by increasing the critical delamination tem-perature and Scott Bond values. Mesic (2002) applied the idea but did notfind any significant effects on the critical delamination temperature. On theother hand, carboxymethylcellulose and chitosan were found to increase thecritical delamination temperature, their effect being primarily attributed tohydrogen bonding interactions (Mesic, 2002). The positive effects were par-ticularly evident on lightly beaten chemical pulps, which would open the wayto a new strategy to widen the operational window of impulse technology:replacing beating with the addition of an appropriate strength agent.

Orloff et al. (1997a, 1998a,b) proposed an interesting strategy to pre-vent delamination, an excellent summary of which is provided by Orloff(1998). Recognizing that the ultimate cause of delamination is flashing ofsubcooled water upon unloading of the impulse nip, Orloff and co-workers

23

Marco Lucisano

Table 2: A summary of technical solutions to prevent delamination. (ex-tended from Larsson and Stenstrom, 1998).

Method References Description

Low basis weight Crouse et al. (1991) Samples of higher basis weight have lowercritical delamination temperature.

High permeability Boerner and Orloff Lower flow resistance for vapor venting to the(1994) atmosphere results in weaker drag forces onOrloff (1994a) the wet web.

Low heat flux Phelan et al. (1996) Ceramic materials transfer less energy toOrloff (1991b) the wet web, reducing the risk of delamination.

Achieved by choosing a material with lowthermal diffusivity and thermal mass.

Metal band Nilsson and Norman Heated metal band increases dewatering and(2002a; 2002b) retards delamination.

Low pressure Larsson and Stenstrom Applied pressure below the critical pressure(1998) prevents delamination.

Optimized pressure Betz (1996) Optimization of the pressure profile andprofile Crouse et al. (1991) adjustments of the point of maximum pressure.

reduce the delamination tendency.

Porous hot surface Stenstrom (1990b) Allows steam ventilation upon unloadingPulkowski et al. (1992) of the press nip.

Multiple pulses Bluhm et al. (1991) Repeated pressure application reducesVomhoff (2000) delamination damage and gives higher

dryness.

High ambient Orloff et al. (1997a) Pressurization to levels exceeding the criticalpressure Orloff et al. (1998b) ambient pressure inhibits delamination by

Orloff et al. (1998b) reducing drag forces on the web.

Chemical additives Westman et al. (2002) The use of chemical additives in the form ofMesic (2002) a polymer component and a particle system

prevents delamination.

increased the ambient pressure after the nip5, thereby preventing delamina-tion. They defined a “critical ambient pressure” associated to a hot platenCritical Ambient

Pressure temperature, as the minimum ambient pressure that has to be applied toprevent delamination, as measured by both direct visual inspection and out-of-plane ultrasonic testing. The fact that the critical ambient pressure didnot need to be as high as the saturation pressure for water evaporationat temperatures close to the setpoint of the hot platen, showed that themethod worked not by preventing flashing of water into steam, but ratherby limiting the drag forces experienced by the top layers of the paper sam-ple. Subsequently, laboratory experiments showed that delamination couldbe inhibited by properly controlling the load applied to the sheet as thenip opens (Orloff et al., 1998c). IPST and Beloit designed an adjustable

5A similar solution was proposed a few years earlier by Babinsky and Mumford (1995).Yet, this older patent failed to recognize the mechanism by which delamination is pre-vented.

24

http://www.ipst.edu


ramp shoe that allowed adjustment of the nip decompression on a runningmachine and managed to eliminate picking and to control sticking (Orloffand Crouse, 1998; Orloff et al., 2000b).

Another constructive solution proposed to limit the negative effects ofdelamination is the replacement of the solid surface of the hot pressing rollwith an open structure. The idea, originally proposed for the drying section Porous Pressing

Surfaces(e.g. Krikorian, 1967; Ely, 1974), was adapted to impulse technology byPulkowski et al. (1989), Stenstrom (1990b,a) and Pulkowski et al. (1992),who suggested that the high-pressure vapor developed in the hot nip coulddiffuse into the porous layer, without disrupting the paper. Additionally,the porous layer tends to reduce thermal transfer between the press rolland the web, which further inhibits delamination of the pressed web. Yet,porous platens do not appear to operate by an impulse mechanism but by acombination of steam forming and venting at the interface between the hotroll and the sheet, augmented by hot pressing (Orloff, 1994b). Moreover, themanufacture of heatable porous rolls is difficult and highly impractical sincelarge stresses are generated between the individual particles, which resultsin rather poor mechanical properties. Therefore a grooved roller or metalband is suggested as the means of dewatering (Keller, 1992). Additionally,markings on the paper surface and uneven pressing could result from such aconstruction (Stenstrom, 1990a). An alternative constructive solution wouldbe the use of a heatable felt passing through the nip with the paper web; thefelt could be of metal fibers (Pulkowski et al., 1992) or of other heat-resistentfibres (Wiberg, 1987).

Nilsson (2001a) and Nilsson and Norman (2002a,b) used a thin heatedmetal band to replace the heat press roll in a series of experiments with alaboratory shoe press. They reported increased dewatering without exten-sive delamination problems. Additionally, this technical solution would giveample space for heating with energy sources such as gas burners, which areinherently more attractive that electricity.

Another technical solution for delamination prevention is the use of mul-tiple hot nips. The idea, originally developed to enhance traditional cylinder Multiple Nipsdrying, has evolved through a large number of patents (e.g. Haigh, 1967;Gersbeck et al., 1975; Miller, 1987, 1988a,b, 1989a,b; Busker et al., 1988;Miller, 1991; Bluhm et al., 1991; Vomhoff, 2000). Although both the techni-cal and constructive solutions and the profiles of pressure application expe-rienced by the web are rather different in the different cases, all multi-pulsepatents claim advantages in terms of higher post-unit dryness and reducedrisks for delamination, when comparing with a classic impulse unit. Unfor-tunately, no independent investigation of the effects of multiple press pulsehas ever been reported in the literature.

25

Marco Lucisano

5 Summary of the Literature

(i) Impulse technology is a web consolidation process in which water isremoved from wet paper by the combined action of mechanical pressureand intense heat.

(ii) There is a large body of information available in the literature on thechanges in paper property and increases in dewatering rates producedby impulse pressing. From this, the mechanism by which steam isformed has been inferred. Few of these works specifically address thecentral issue of this work, that being, to experimentally characterizephase change phenomena in paper webs undergoing an impulse event.

(iii) The industrial application of impulse technology has been limited byseveral runnability problems, among which delamination is the mostsevere. Although much is known about how fiber type and processingconditions affect the process runnability, delamination remains themajor obstacle on the way towards developing impulse technology asa commercially feasible technique.

6 Objectives

The specific objectives of this thesis are:

(i) To characterize the changes in web structures during the delaminationprocess (Paper A).

(ii) To experimentally characterize phase change processes during impulsepressing of wet paper webs (Paper B — Paper E).

(iii) To experimentally determine the thermal properties of water saturatedpaper webs, under conditions of interest for high-intensity processes(Paper F).

26


Measure what is measurable, and make mea-surable what is not so.

Galileo Galilei (1564–1642)

7 Characterizing the Delamination ProcessDuring Impulse Pressing

Delamination, i.e. a strength loss in the thickness direction of the sheet, isthe most serious runnability issue preventing the industrial application ofimpulse technology. Paper A presents two complementary studies charac-terizing the changes in web structure upon impulse pressing of paper. In Paper Athe first study, we measured the change in solidity of wet paper webs as afunction of pressing temperature. The paper samples were cross-sectionedand imaged, allowing the solidity profile to be determined. In the secondstudy we measured the changes in permeability of both dried and never-driedsamples.

The majority of the experimental work presented in this study was car-ried out on the research papermachine EuroFEX at STFI that has beendescribed previously (Rigdahl et al., 2000). The papermachine was config- Experimental Detailsured with the roll-blade forming unit, operated at a constant machine speed(600 m/min) and basis weight (60 g/m2), with linear loads of 60 kN/m inthe first roll press, and 600 kN/m in the second and third extended pressnips. The web dryness prior to the impulse nip was approximately 30%.The only machine variable that was adjusted was the temperature of thethird press nip, which was increased stepwise from 40 ◦C to 290 ◦C (or tothe point of complete delamination).

27


http://www.stfi.se

Marco Lucisano

0 20 40 60 80 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

AHot Roll Side

Felt Side

Solidity [%]

Pos

ition

[Dim

ensi

onle

ss]

0 20 40 60 80 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

CHot Roll Side

Felt Side

Solidity [%]

Pos

ition

[Dim

ensi

onle

ss]

0 20 40 60 80 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

BHot Roll Side

Felt Side

Solidity [%]

Pos

ition

[Dim

ensi

onle

ss]

0 20 40 60 80 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

DHot Roll Side

Felt Side

Solidity [%]

Pos

ition

[Dim

ensi

onle

ss]

Figure 8: Solidity profiles of oven- and freeze-dried sheets sampled beforethe impulse nip (A and B) and after the nip (C and D). The pressingtemperature was set to 290 ◦C: 4 = oven-dried samples; ◦ = freeze-dried samples. The error bars report the standard deviation of the localdensity profile.

7.1 Changes in Web Solidity

Cross-sectional solidity profiles were evaluated from micrographs of papercross-sections using an image-analysis technique (Ratto, 2001). A compar-Solidity Profiles by

Image Analysis ison of some typical profiles of oven- and freeze-dried samples reported inFigure 8 shows that the drying history affected the measured solidity pro-files. In fact, the thickness of non-delaminated freeze-dried samples wasapproximately 40% larger than the thickness of corresponding oven-driedsamples indicating a substantial action of capillary forces during drying.Moreover, a parabolic profile was evident in all non-delaminated oven-driedsamples (Figure 8) whereas an essentially uniform profile was found forfreeze-dried sheets. On the basis of these results, we advance the argument⇒ No Heat-Induced

Stratification that for non-delaminated impulse pressed sheets, the resulting parabolicsolidity variation is a phenomenon resulting from capillary forces present

28


during oven-drying and not a result of hydrodynamic forces induced ontothe fibers during the pressing event. No evidence was found to justify thehypothesis of heat-induced stratification (Burton and Sprague, 1987) for theconditions tested.

Figure 9: Cross-section of a freeze-dried SBK sheet pressed with the hotroll temperature set to 290 ◦C. The top surface was in contact with thehot roll. The picture represents a 0.5 mm wide section.

Figure 9 shows the cross-section of a freeze-dried SBK sheet impulsepressed at 290 ◦C on EuroFEX. The arrangement of the fiber material inthe top left corner of the sheet illustrates clearly the damage caused bythe venting of steam during decompression. As the effects of delamina-tion became noticeable, the standard deviation of the local solidity valuesincreased reaching values up to two-three times the standard deviation ofnon-delaminated paper (compare Figure 8A with C and B with D). As a ⇒ A Tool for

Delamination Studiesresult we propose that the magnitude of the standard deviation can be usedas a measure the extent of delamination.

7.2 Permeability

The ease with which water flows through a paper web is characterized byits permeability. Darcy proposed an empirical law, which correlates the

29


Marco Lucisano

superficial velocity, u, of a fluid flowing through a porous medium of thick-ness L directly to the pressure drop, ∆P , across a porous material and toa factor called permeability, κ and inversely to the viscosity of the fluid, µ(Scheidegger, 1974):

u = −κ

µ

∆P

L. (4)

We measured the permeability of oven- and freeze-dried samples with airas the permeating fluid according to the Bendtsen standard procedure.Furthermore, the permeability of never-dried sheets was measured in acompression-permeability cell with water at 45 ◦C as the permeating fluid,following the method outlined by Vomhoff (1998). Tests were performed atthree different mechanical loadings, namely 1.4, 2.4 and 3.4 MPa in order tobracket the typical loading conditions experienced in a press nip. For bothdry and never-dried samples, we report modified permeabilities, κm, definedas:

κm = κ · ρweb, (5)

where ρweb is the bulk density of the paper web.

0 50 100 150 200 250 3000

1

2

3

4

5

6

7

8

Temperature [°C]

Mod

ified

Per

mea

bilit

y [1

0−11

kg/

m]

A B

Figure 10: Modified permeability of dry SBK handsheets (60 g/m2) asa function of pressing temperature. Values calculated from Bendtsen airpermeance. 4 = oven-dried samples; ◦ = freeze-dried samples.

The results of the permeability measurements on dried SBK handsheetsare shown in Figure 10. The permeability to air of both oven- and freeze-Permeability of Dry

Paper dried SBK handsheets was found to decrease nearly linearly up to 200 ◦Cand then become essentially constant. These results are in agreement withsimilar trends reported in the literature (Orloff et al., 1998b; Orloff andCrouse, 1999; Larsson and Stenstrom, 1998) and reflect the fact that paper

30


densified with increasing pressing temperature by reduction of the interfibre ⇒ Densificationspacing.

0 50 100 150 200 250 30010

0

101

102

Temperature [°C]

Mod

ified

Per

mea

bilit

y [1

0−14

kg/

m]

A B

Figure 11: Modified permeability of never-dried SBK handsheets pressedon a heated platen press. Measurements performed with 45 ◦C wateras permeating fluid. Mechanical pressure applied during measurements:◦ = 1.4 MPa; 4 = 2.4 MPa; 2 = 3.4 MPa. Filled symbols indicate adhe-sion to the hot press surface.

Figure 11 reports the results of the permeability measurements per-formed on never-dried SBK handsheets. Here, the permeability to water of Permeability of Wet

Papernever-dried sheets remained essentially constant at pressing temperaturesbelow 90 ◦C and increased linearly at temperature above 100 ◦C. We as-cribe the difference between the trends observed in Figure 10 and Fig-ure 11 to the different loading conditions used for the two types of mea-surements. During air permeability measurements, the air was likely topermeate through the sample through the open pore space between thefibers. Yet, under the mechanical loads used for the water permeabilitytests, the interfibre porosity would probably disappear, leaving the waterflow to permeate through the cell wall material to a larger extent. We assertthat the observed increase in water permeability with pressing temperaturecan be related to damage occurring in the cell wall material during the de-lamination event. Indeed, the study of the micrographs obtained during the ⇒ A Measure of

DelaminationDamage?

measurements of solidity profiles showed the presence of radial fractures inthe fiber wall material and we speculate that this might be due to flashingof water located in the fiber wall and lumen (Figure 12).

31

Marco Lucisano

Figure 12: Cross-section of SBK paper pressed on EuroFEX at 290 ◦C.Radial fractures in the fiber walls are evident under microscopic analysis.

32


It isn’t that they can’t see the solution. It is that they can’t seethe problem.

Gilbert K. Chesterton (1874–1936, The Scandal of Father Brown)

8 The Role of Phase Change Phenomena in thePhysics of Impulse Technology

Delamination prevention is central to the industrial application of impulsepressing and the literature has dedicated great attention to this issue. Itis generally agreed that delamination occurs when the flow of steam gen-erated upon flashing is greater that the z-directional strength of the wetweb. Therefore, phase change phenomena are not only the core of impulsepressing as an energy efficient dewatering and consolidation technique, butalso the ultimate reason for the major obstacle to implementation.

This section summarizes four experimental studies focusing on the roleof phase change phenomena in the physics of impulse technology. From theexperimental results we propose an alternative view on impulse technologyin which the path of events centers around the concept of “flashing-assisteddisplacement dewatering”.

8.1 Dewatering Performance

The objective of this investigation was to link heat transfer to dewatering,specifically evaporative losses, in impulse pressing experiments. Two inde- Paper Cpendent laboratory devices — a platen press and a shoe press — were usedto impulse press softwood bleached kraft (SBK) webs of a range of basisweights (w = 20− 200 g/m2).

33

Marco Lucisano

hh qT ′′&,

control volume steam

web

felt

(1)initial loading

(2)completion

of wet pressing

(3)post-nip

evaporation

Figure 13: Schematic of idealized in-nip dewatering and post-nip evapo-ration in impulse pressing.

In order to evaluate the possible water removal mechanisms in impulsepressing, it is instructive to consider a general energy balance of the process.As shown in Figure 13, the heated surface with temperature Th suppliesheat to the web while inside the nip. Part of this heat serves to raise thetotal enthalpy of the web (fibers, water, air, and possibly steam), whilea portion of this energy is lost to the felt by conduction and convection.Simplified Energy

Balance When the nip opens, the hot, wet web releases energy via flashing, thussupplying additional drying potential. We define the control volume as theweb plus any evaporated steam, and if thickness-direction gradients andthe contribution of air are ignored, the total change in enthalpy ∆h′′1−3 (inkJ/m2) is

∆h′′1−3 = w

(hf,3 − hf,1) + (MR2hw,2 −MR1hw,1)︸︷︷︸(a)

+hv(MR2 −MR3)︸︷︷︸(b)

,

(6)where hf is the enthalpy of the fibers, hw is the enthalpy of water, hv isthe heat of vaporization, MR is the moisture ratio at a given state, and thesubscripts 1, 2 and 3 refer to the system conditions illustrated in Figure 13.Term (a) represents the change in sensible heat, while term (b) representsthe change in latent heat. If ∆h′′1−3 is proportional to q′′h, the energy suppliedby the heated surface, Equation 6 suggests that

q′′h ∝ a∆h′′th,sen + b∆h′′th,lat. (7)

where a and b are two coefficients of proportionality. In this expression,∆′′

th,sen is the theoretical sensible heat change and is ∆′′th,lat the theoretical

34


latent heat change, defined via

∆h′′th,lat = whv(MR2 −MR3). (8)

According to this line of reasoning, any dewatering in addition to the wetpressing contribution is completely embodied in ∆h′′th,lat. A plot of q′′h versus Two Regimes for

Impulse Pressing∆h′′th,lat thus gives an indication of the importance of evaporative dewater-ing. Two different regimes may be identified via slope b:

b ≥ 1 felt heat losses and/or evaporative dewatering important;b < 1 evaporative dewatering and other mechanisms important.

A softwood bleached kraft (SBK) pulp refined to 25 ◦SR was used as thefurnish in all experiments. Handsheets with basis weight w ranging from Experimental Details20 to 200 g/m2 were produced with a Finnish former and the solids contentwas approximately 30% after forming and light pressing. Commercial finepaper felts were used in all tests. The platen press was equipped witha load cell, heated steel block and a control system. An ultra-thin, fastresponse thermocouple allows for the concurrent measurement of Th andthe instantaneous heat flux. The total supplied energy is evaluated from

Heat Flow

surface temperature data according to the numerical procedure presented byTaler (1996). A shoe press developed by the Department of Pulp and PaperChemistry and Technology at the Royal Institute of Technology (KTH) inStockholm was also employed. As described in Nilsson (2001a), the KTHShoe Press features a heated metal band that is passed through the nipformed between a 300 mm diameter by 200 mm wide roll and a 42 mm longconcave shoe (the web, felt, and a lubricated plastic sheet follow the band).In these experiments, the band was sufficiently thick (0.5 mm, steel) so thatits thermal performance was fairly close to that of a more massive heatedroll (Nilsson, 2001a).

The relationship between the energy supplied by the heated surface,q′′h, and the theoretical latent heat of drying, ∆h′′th,lat, is illustrated in Fig-ure 14. A straight line is fitted to the data in each case, and a reference linewith a slope of one is drawn from the common y-intercept for comparison.For the given process conditions, it appears that evaporative dewatering Dewatering

Performanceand/or felt heat losses dominate for w ≤ 60 g/m2. With w = 60 g/m2, shoepress data obtained in an independent study (Nilsson, 2001a) are shownfor comparison, and indicate that the platen press and shoe press performsimilarly. At higher basis weights, b drops well below one, suggesting thatother mechanisms contribute in part to the enhanced dewatering. Sensibleheating effects are reflected in the y-intercept values, which rise with anincrease in basis weight (see Equation 6).

With these results, it is possible to estimate the relative importance ofdewatering via paths other than evaporation for the high basis weight data

35

Marco Lucisano

y = 1.33x - 0.70r 2 = 0.68

0

20

40

60

80

100

120

0 20 40 60 80 100 120∆h" th,lat (kJ/m2)

q"h (k

J/m

2 )

reference

Ay = 1.07x + 18.8

r 2 = 0.85

Shoe press:y = 0.94x - 3.06

r 2 = 0.87

0

20

40

60

80

100

120

0 20 40 60 80 100 120∆h" th,lat (kJ/m2)

q"h (k

J/m

2 )

reference

reference (shoe press)

B

y = 0.68x + 28.2r 2 = 0.79

0

25

50

75

100

125

150

0 25 50 75 100 125 150∆h" th,lat (kJ/m2)

q"h (k

J/m

2 )

referenceC

y = 0.56x + 29.4r 2 = 0.73

0

25

50

75

100

125

150

0 25 50 75 100 125 150

∆h" th,lat (kJ/m2)

q"h (k

J/m

2 ) reference

D

Figure 14: Supplied energy versus theoretical energy, platen press ex-periments with basis weight w = (A) 20 g/m2; (B) 60 g/m2; (C) 120 g/m2

and (D) 200 g/m2. Shoe press results shown in part (B) obtained fromNilsson (2001a).

(w ≥ 120 g/m2). If the felt heat losses are negligible, slope b from Equation 7may be rewritten as

b =∆MRevap

∆MRtot −∆MRref, (9)

where ∆MRevap is the actual moisture ratio change due to evaporation,∆MRtot is the total moisture ratio change, and ∆MRref is the reference(unheated) moisture ratio change. An impulse pressing efficiency, εimp, maybe defined asImpulse Pressing

Efficiencyεimp =

∆MRtot −∆MRref −∆MRevap

∆MRevap=

1− b

b(10)

and represents the ratio of enhanced non-evaporative dewatering to evapora-tive dewatering. For w = 120 g/m2, εimp = 0.47, implying that evaporativedewatering and other mechanisms have a similar effect. At a higher basisweight (200 g/m2), εimp = 0.78, so most of the enhanced dewatering in thiscase is largely due to mechanisms other than evaporation.

36


The nature of the water removal mechanisms has a number of impor-tant consequences. From an energy point of view, evaporative dewatering ⇒ Energy Economyresulting from impulse pressing is less desirable than displacement dewater-ing, since low-cost steam in the dryer section could just as well be used tovaporize water from the web. The present findings show that lightweightsheets are most susceptible to evaporative effects, so extra attention may berequired for these grades.

It is not surprising that evaporative effects are most apparent for lightersheets. However, the fact that other mechanisms are at play at higher basisweights is intriguing. Some authors have concluded that dewatering in im- ⇒ Grammage

Dependent Effectspulse pressing is characterized mainly by hot pressing with no in-nip phasechange contribution, accompanied by evaporation (Macklem and Pulkowski,1988; Larsson and Stenstrom, 2001). Hot pressing effects such as reducedviscosity and increased web compressibility may play a role, but they shouldhave a diminishing effect at higher basis weights, owing to the sharp tem-perature gradients within the web.

The findings of Paper C suggest that some mechanisms (“impulse ef-fect”), different from the simple superposition of hot pressing and evapora-tion, might be active during an impulse event. In the literature, a steam- Impulse Effectassisted displacement mechanism has been suggested, in which liquid wateris displaced by the action of an expanding steam layer. Nevertheless, thereis little or no direct experimental evidence supporting this occurrence. Inthe following sections we investigate some aspects of the physics of impulsetechnology, with particular interest in the physics of phase change in a wetpaper web.

8.2 Interfacial Contact Phenomena

Contact phenomena during an impulse event and the presence of steam atthe interface between the wet sheet and the hot pressing plate are issuesthat received scarce attention in the literature and no direct investigationsare available. Photographic studies of contact phenomena in traditional wet Paper Epressing have been performed (Heller and Bliesner, 1977; Wahren, 1989) andthese approaches can also be extended to hot and impulse pressing.

We investigated contact phenomena at the interface between the wet weband a hot press surface with a modified laboratory platen press (Figure 15).Here, the heated press platen was replaced by a preheated removable 10 cmdiameter by 1 cm thick borosilicate safety sight glass mounted from below,and high-speed photography was employed to visualize the interfacial inter-actions. Paper samples (60 g/m2 basis weight) and felts were mounted inthe press with a ring/clamp arrangement. Commercial fine paper press felts Experimental

Procedure37

Marco Lucisano

high-speed monochrome camera

web

hot glass platen

felt

appliedloadpressing head

IR laser illumination

mirror

connection to load cell and base

high-speed monochrome camera

web

hot glass platen

felt



mirror


web

hot glass platen

felt



mirror


Figure 15: Schematic of visualization apparatus and high-speed cameraset-up.

were used in all tests. A mirror allowed for a maximum viewing area of about4 cm2. In these trials, several glass platens were heated in a furnace with asetpoint temperature of around 250 ◦C. Once uniformly heated, one glassplaten was removed from the oven and placed on the stand, and the surfacetemperature was then checked with a thermocouple. The pressing head withattached sample was lowered into place, a high-speed camera was activated,and the pressing event was initiated. The maximum applied pressure wasapproximately 8 MPa and the pulse shape simulated an extended press nipas in Paper C. Afterwards, the glass plate was removed and placed backinto the oven. Images were transferred to a PC equipped with a video cap-turing card. A fresh glass platen was taken from the oven for the next trial.Each video sequence was edited on the PC to eliminate superfluous framesand an imaging software package (Scion Image, Scion Corp.) was used toidentify bubble/liquid interfaces.

A significant difference between conventional laboratory pressing exper-iments on a platen press and the visualization trials was the use of a glassPossible Limitations?plate instead of a metal platen (i.e. poor conductor vs. good conductor ofheat). Transient thermal characteristics, or “thermal mass” of the heatedmaterial (Phelan et al., 1995) can be expressed via k√

α, where α is the ther-

mal diffusivity and k is the thermal conductivity. This ratio is roughly anorder of magnitude higher for steel than for borosilicate glass at 200 ◦C,which implies that heat transfer and hence vaporization potential is not asgreat in the visualization trials.

Figure 16 shows images from separate unheated and heated trials witha residence time of 25 ms and a maximum applied pressure of 8 MPa. AnShort Press Pulsesimage filtering process was applied to the left hand portion of each image;

38


total width of the image was 0.5 mm. Initially, the presence of what seemsto be pin-sized surface air bubbles was apparent as bright spots on the rawimages (right side). These bubbles could be more clearly seen after filtering(left side). In the unheated trials, these bubbles disappeared once the loadwas actually applied. This can be interpreted as a complete saturation ofthe web once the hydraulic pressure reached a sufficient level. Heated trialsproduced slightly different results, as some bright spots were present eveninside the nip. This could indicate that air/steam bubbles persisted at theinterface between the web and the hot glass platen. From viewing this ⇒ Steam

upon Flashingsequence, it appeared that these bubbles were relatively static and did notexhibit erratic behavior one might expect from rapid steam generation. Thesituation was quite different at unloading, as shown at +27 ms (Figure 16).The increased reflectance suggests flashing at the end of the pulse.

An increase in nip residence time to 250 ms revealed a more dramaticpattern of events for the impulse trials. Images obtained from one of these Long Press Pulsesheated trials are shown in Figure 17. With this long residence time, brightspots (possibly bubbles) appeared soon after loading and increased in densityacross each image. Moreover, fluctuating vapor regions could be clearly seenin the videos, which are traced in the unedited images from +120 ms to theend of the pulse. These regions were essentially superimposed over a largearea where spots of high reflection could still be seen.

To summarize, videos obtained from trials with short residence times didnot indicate any significant in-nip steam generation. This type of phenom-ena could only be seen for pulse lengths well beyond those encountered inindustrially relevant impulse press nips.

8.3 Thickness-Direction Profiles

The study of phase change phenomena at the interface between the hot pressplaten and the paper web has been complemented by a series of laboratoryinvestigations of the local phasic composition across the thickness directionof the paper web. Here, we embedded micro-thermocouples (Paper B) atdifferent elevations in a sample undergoing compression in a heated platenpress. Phase change phenomena inside the sheet were studied by comparisonof our experimental results with well-established descriptions of drying ofwet porous media. Yet, local sheet temperatures only offer indirect clueson phase change phenomena of interest in impulse pressing. Therefore,we implemented the concept of the resistivity probe as a direct sensor ofthe phasic composition inside a wet paper sheet (Paper D). Finally, weused a heated laboratory-scale shoe press to investigate the differences inWRV (Water Retention Value) between different layers within a 150 g/m2

composite sheet (Paper C).

39

Marco Lucisano

unheated platen heated platen

-3 ms

0.5 mm

+7 ms

+27 ms (pulse end)

Figure 16: Images obtained for wet pressing and impulse pressing of a60 g/m2 web with a nip residence time of 25 ms and a maximum appliedload of 8 MPa. Glass platen surface temperature was about 220 ◦C priorto the pressing event. Video sampling rate was 1000 fps and a 4x zoomlens was employed.

40


0.5 mm

+0 ms(pulse start)

+100 ms

+180 ms +240 ms(pulse end)

Figure 17: Images obtained for impulse pressing of a 60 g/m2 web witha nip residence time of 250 ms and a maximum applied load of 8 MPa.Glass platen surface temperature was about 220 ◦C prior to the pressingevent. Video sampling rate was 1000 frames per second and a 4x zoomlens was employed.

41

Marco Lucisano

8.3.1 Temperature Profiles

In this study, we measured the transient temperature profiles of wet paperPaper Bwebs undergoing compression at constant rate in a heated platen press.Thin, micro-thermocouples (diameter 25 µm) were embedded into wet papersheets at different elevations allowing the local temperature to be recordedas a function of time. The experimental conditions were such that the pulsePressing

Experiments length was varied between approximately 100 ms and 4− 5 s and the initialtemperature of the platen press was set from 150 ◦C to 300 ◦C.

0 1 2 3 4 5 60

50

100

150

200

250

300

Hot Plate

Time [s]

Tem

pera

ture

[°C

]

8 7 6 5432 1

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5−3

−2.25

−1.5

−0.75

0

Time [s]

Hea

t Flu

x [M

W/m

2 ]

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.50

2.5

5

7.5

10

Pre

ssur

e [M

Pa]

0 20 40 60 80 100 120 140 160 1800

50

100

150

200

250

300

Hot Plate

Time [ms]

Tem

pera

ture

[°C

]

1

25−4−3

67

8

0 45 90 135 180−3

−2.25

−1.5

−0.75

0

Time [ms]

Hea

t Flu

x [M

W/m

2 ]

0 45 90 135 1800

2.5

5

7.5

Pre

ssur

e [M

Pa]

Figure 18: Internal web temperatures (lower graphs) during impulsepressing of SBK paper with a laboratory platen press. The top graphsreport the heat flux out of the hot pressing plate and the applied pressurepulse. The duration of the press pulse was (A) 5 s and (B) 120 ms. Thehot plate temperature was set to 300 ◦C the peak pressure was 8.4 MPa,and the total grammage was 500 g/m2. Thermocouple signal numbersincrease towards the hot platen.

Approximately six hundred experiments were performed using the platenpress. The forms of the temperature profiles shown in Figure 18 are repre-sentative of all cases tested and we found that the compression rate was themain variable influencing the overall shape of the profile observed. Beforediscussing these results, it must be noted that we are uncertain whether

42


or not heat is being transferred in the axial direction only. This has beentacitly assumed in previous works (Devlin, 1986; Burton et al., 1986; Krookand Stenstrom, 1998; Orloff et al., 1998b). Our confidence in interpret-ing the data using a one-dimensional framework diminishes with increasingplaten temperature and compression rate as we observed a physical processwe define as a “temperature inversion”, i.e. the lower portions of the pa-per samples were warmer than those closer to the heating element. This Temperature

Inversioncould be caused by fingering, due to uneven pressure application (Ellis,1981; Vomhoff, 1998) or by variations in the local grammage causing freefluid channels. Further, accurately decoupling these effects from the realtemperature measurements is difficult, if not impossible.

As shown in Figure 18A, with a pulse length of approximately 5 s, fourdistinct regions are evident in the measured temperature profile: (i) initiallythe sheet was heated to 100 ◦C, whereupon (ii) the temperature remainedconstant at the atmospheric boiling point of water until the felt side of thesheet reached 100 ◦C. After that, (iii) the temperature increased again until(iv) an almost stationary condition was reached just before the nip opened. Long Press PulsesIndeed, this response is very similar to findings in related literature (Luikov,1975; von Byrd, 1982; Rosen, 1987; Christensen et al., 1982; Hasatani et al.,1991). Our findings, however, do not qualitatively resemble the temperatureprofiles given from the solution of Stefan problem. We found qualitative dif-ferences between our results and the Stefan problem in the region 1 < t < 3 sof Figure 18A. Here, the upper thermocouple remained constant, while thebottom of the sheet heated up to a uniform temperature of 100 ◦C. We es-timate that during this period heat was added to the paper sample at aconstant rate of 0.25 MW/m2. With the Stefan problem, there must be atemperature gradient at all points over the domain, since heat is transferredby conduction alone. As we measured a heat flux with essentially no tem-perature gradient, we must conclude that conduction was not the primarymechanism by which heat was transported in this region and the dominantmode of heat transfer is likely to be convective, that is a heat-pipe.

With faster compression rates, we saw no evidence of a heat-pipe as allof the thermocouple signals crossed the 100 ◦C line with no discontinuity intheir first derivative. We speculate that this occurred as the increasing com- Short Press Pulsespression rate would cause an increase in the hydraulic pressure due to eitherweb stratification or viscoelastic effects. The increase in hydraulic pressurewould tend to shift the boiling point of water and prevent steam generation.This behavior suggests that for short pulses the hydraulic pressure in mostof the sheet was high enough to prevent evaporation and water is present inthe liquid phase until the pressure is released.

Finally, we observed a sudden increase in temperature when the mechan-ical load was released from the paper sample, see Figure 18B at t = 120 ms.

43

Marco Lucisano

This was observed to occur with platen temperatures greater than 200 ◦Cand with compression times smaller than 500 ms. We measured temperatureincreases as high as 70 ◦C in 5 ms. This occurrence was accompanied by awhistling noise during the experiments. This phenomenon is also consistentTemperature

Increaseupon Unloading

with the temperature profiles published by (Orloff et al., 1998b). Exper-iments where later repeated with more realistic pressing conditions (presstimes reaching dow to 25 ms, pressure pulse similar to the one used in Pa-per C) with layered sheets structures of grammage ranging from 40 g/m2 to300 g/m2 (Results unpublished). The qualitative trends observed in PaperB were confirmed even in those experiments.

8.3.2 Resistivity Probes — Local Phasic Composition

Investigating the physics of impulse technology through the study of thelocal temperature field provides some interesting clues on the mechanismsof phase change active in an impulse nip. Nevertheless, thermocouples giveno direct information on the local phasic composition and only indirectindications of phase change.Paper D

The (electro)resistivity probe is a simple device that allows for detectionof phase composition of a sample via local measurement of conductivity,Resistivity Probes:

Basic Concepts capacitance, inductance, or impedance. This technique has been used ex-tensively in a variety of applications outside the paper science field, includingboiling heat transfer, droplet flows, and fluidized beds (Cheremisinoff, 1986).One of the simplest resistivity probes consists of two insulated wires sepa-rated by a gap. If a liquid bridges the gap, the electrical resistivity at the tipis relatively low, whereas the resistance is many orders of magnitude higherin the presence of a gas.

In this study, probes were constructed by gluing together two poly-amide-insulated copper wires with a diameter of 50 µm. The probes wereconnected to a computer equipped with an AD converter via a signal pream-plifier. An output of 0 V indicates the presence of a nonconducting mediumPressing

Experiments (gas or vapor), while an output of 3− 4.5 V indicates a conducting medium(liquid). Resistivity probes were embedded alongside micro-thermocouplesin a layered structure, similar to that used in Paper B. Pressing experi-ments were performed with a heated press. Impulse pressing experimentswere conducted under realistic operating conditions, i.e. with surface tem-peratures up to 300 ◦C, nip residence times of 25 ms, applied pressure around4 MPa and pulse shape corresponding to that of Paper C.

The probes were initially tested to rule out any possible interference ofthe fiber network on the probe signal. Additionally, simple model experi-ments in a water bath showed that phase change could be detected from themeasured signals.

44


Hot Platen

20 g/m2T

RP

T

T

T

20 g/m2

20 g/m2

RP

60 g/m2

80 g/m2

Felt

0 0.5 1 1.5 2 2.5 3 3.50

100

200

300

Time [s]

Tem

pera

ture

[ºC

]

20 g/m2 40 g/m2

60 g/m2

120 g/m2

Hot Plate

0 0.5 1 1.5 2 2.5 3 3.50

1

2

3

4

5

Time [s]

Pro

be S

igna

l [V

]

0 0.5 1 1.5 2 2.5 3 3.50

50

100

150

200

250

Pre

ssur

e [k

Pa]20 g/m2

40 g/m2

Figure 19: Resistivity probe signal and internal web temperature duringa long low-pressure pulse. Two probes (RP) and four thermocouples (T)were embedded in a wet SBK sample of total grammage 200 g/m2.

The temperature profile inside a paper sheet during a very long, lowpressure pulse is well known and presents clear indications of the phasetransition from liquid water to steam (Luikov, 1975). Figure 19 shows the Validation:

Detection ofEvaporation

results of measurements performed with embedding two resistivity probesand four thermocouples into a 200 g/m2 sheet. The onset of dry-out at eachlayer can be identified by the local temperature increasing above 100 ◦C.Indeed, comparison with the probe signals confirms that resistivity probescan be used to identify the presence of steam inside the web, since the voltageacross the probe started to decrease (i.e. the resistivity of the mediumincreased) approximately at the same instant in which the correspondinglocal temperature raised above 100 ◦C.

Impulse pressing trials on samples of total web basis weight 40 g/m2 Hot Pressing Trialsshowed phase change after decompression, and no vapor fronts could beseen before the point of maximum applied load. When similar pressurepulses and temperatures were applied to webs of grammage higher than120 g/m2, a sudden increase in temperature was recorded just before theend of the pressure pulse (Figure 20). This occurrence is in agreementwith previous experimental studies (Paper B). Resistivity probes embeddedalongside with thermocouples in thick samples detected a phase transitionin the warmer layers of the sheet, where steam is expected to form uponunloading of the nip. Additionally, no evidence of phase transition wasseen for the probes closer to the felt, were the temperature remained under100 ◦C.

A number of other test conditions were considered to see if there wereany significant deviations from the trends discussed above. For platen tem-peratures of around 180 ◦C and above, thin sheets exhibited phase change ⇒ Steam

upon Flashing45

Marco Lucisano

Hot Platen

20 g/m2

40 g/m2T

RP

T

T40 g/m2

RP 100 g/m2

Felt

0 10 20 30 40 50 60 700

100

200

300

Time [ms]

Tem

pera

ture

[ºC

]

20 g/m2

60 g/m2

100 g/m2

Hot Plate

0 10 20 30 40 50 60 700

1

2

3

4

5

Time [ms]

Pro

be S

igna

l [V

]

0 10 20 30 40 50 60 700

2

4

6

8

10

Pre

ssur

e [M

Pa]20 g/m2

100 g/m2

Figure 20: Resistivity probe signal and internal web temperature duringa 25 ms impulse pressing simulation. Two probes (RP) and three ther-mocouples (T) were embedded in a wet SBK sample of total grammage200 g/m2.

tendencies only after the pressure pulse had ended, while flashing was seenduring the decompression phase in thicker sheets (> 120 g/m2). For pulsesshorter than 1 s, nip residence time did not have much influence on thisbehavior, nor did the magnitude of applied pressure have any effect. Thesefindings point to the existence of steam only in conjunction with flashingat unloading, confirming the theory originally proposed by Macklem andPulkowski (1988) as well as the findings of Zavaglia and Lindsay (1989).

8.3.3 Z-Directional WRV Gradients

The study of the local temperature profiles during an impulse event (PaperB) shows that the paper layers closest to the hot press surface reach highertemperatures than the average sheet temperature in the nip. Additionally,Paper D shows that, for high grammage webs, flash evaporation upon nipunloading takes place in the hotter layers only, with no phase change inthe proximity of the felt. Here, we wanted to investigate whether the non-Paper Cuniform physical environment in the thickness direction of the sheet has anyinfluence on the fibers in the sheet. Therefore, we measured a layerwiseWRV profile for composite 150 g/m2 sheets pressed on the KTH laboratoryshoe press.

Five 30 g/m2 sheets were stacked together to form a composite web.Experimental DetailsAfter pressing, each layer was carefully removed, placed in a plastic bag, andsubsequently analyzed. The five layers obtained from one test were weighedin order to determine moisture profiles, while water retention value (WRV )measurements were performed for a separate test with identical processing

46


1,20

1,30

1,40

1,50

1,60

0,35 0,40 0,45 0,50 0,55

Final Dryness

WR

V (g

/g)

25°C100°C200°C300°C

roll-side layers

Figure 21: Water retention value (WRV ) versus outgoing dryness foreach layer within a 150 g/m2 composite sheet, shoe press experiments.Data points represent the average of two experiments at a given positionand temperature and error bars show the estimated uncertainty in thedata. For the unheated case, trials were performed with sheets pressed upto four times in succession, and a straight line is fitted to the data.

conditions. The WRV , defined as

WRV =mw

mbd, (11)

was determined according to SCAN–C 63:00. In this method, the wet sampleis centrifuged at 3000 g for 15 min, and the mass of water after centrifuging(mw) and the bone-dry mass (mbd) are determined gravimetrically. A slightdeviation from the standard’s procedure was necessary: since the sampleshad a low mass, each sample was wetted before centrifuging.

The findings are summarized in Figure 21, which shows WRV for eachlayer as a function of dryness. A reference curve was obtained from wetpressing experiments where the composite sheet was pressed one, two, orfour times in succession. Most of the subsequent data fall on this curve,with exception of the layers adjacent to the heated band. Here, elevatedtemperatures cause a departure from the expected WRV -dryness relation-ship. Figure 21 also reveals that the roll-side layer reached a much higher WRV Profilesdryness level as compared to the remaining layers with Th > 100 ◦C. Thisis more clearly seen in Figure 22, which illustrates the variation of WRVin the thickness direction for the different temperatures.

47

Marco Lucisano

1,20

1,30

1,40

1,50

1,60

1 2 3 4 5

Transverse Position

WRV

(g/g

)25°C100°C200°C300°C

roll side felt side

Figure 22: Water retention value (WRV ) for the five layers within a150 g/m2 composite sheet, shoe press experiments. Data points representthe average of two experiments at a given position and temperature andthe error bar show the estimated uncertainty in the data.

Unfortunately, the experimental uncertainty makes it difficult to attaina definitive conclusion on the processes generating a deviation from thedensity-WRV relationship at elevated temperatures. Several mechanismscould be active simultaneously and cause or contribute to both densificationand hornification. In fact, both wet pressing and evaporative drying causehornification, and WRV versus dryness data typically fall on a single curveif the conditions for water removal are the same (Weise, 1998). Hornifica-tion may be enhanced at higher temperatures (Weise, 1998), so the presentfindings may simply reflect the sharp temperature gradients in the thicknessdirection. Another possibility is that evaporation, which is highest in thelayers closest to the heated surface, leads to a reduction in WRV .

8.4 Flashing-Assisted Displacement Dewatering

In spite of almost 30 years of active research, direct experimental evidenceof steam-assisted displacement dewatering in impulse pressing is scarce andinconsistent. A general result of our experimental studies (Paper B, Pa-per D and Paper E) is that little or no steam is observed prior to pressurerelease. This would rule out traditional steam-assisted displacement as adewatering mechanism. Yet, Paper C reports indications of some dewater-ing mechanism other than hot pressing and evaporative drying, being active

48


during impulse pressing of paper samples with grammage > 60 g/m2.

Whereas no steam is observed before pressure release, we found clearindications of flashing evaporation upon unloading of the nip. Additionally,Paper B reports evidence of a sudden increase in temperature when themechanical load was released from the paper sample (Figure 18B) resultingin temperature increases as high as 70 ◦C in 5 ms. This phenomenon is alsoconsistent with the temperature profiles published by Orloff et al. (1998b).

8.4.1 STFI’s FlashLab

Circulating heat-transfer oil

P

Cooling water

Wire

Expansiontank

Escape valve

Guard heaters

insulationThermal

Tem

pera

ture

mea

sure

men

ts

Spring

Figure 23: Schematic of STFI’s FlashLab in the configuration used forflashing experiments.

We investigated flashing phenomena upon unloading of an impulse nipin STFI’s FlashLab, a novel experimental facility for the study of phase Paper B

Paper Echange-induced phenomena in water-saturated porous media. A detaileddescription of FlashLab’s characteristics was presented by Lucisano (2000).FlashLab is composed of four subsystems: the flashing cell, a heating sys-tem with circulating oil, a cooling and pressurization system with circulatingwater and a process control and data acquisition system (Figure 23). Themeasurement cell is approximately 45 cm long with a square cross-sectionof 0.01 m2. Its front face is equipped with two transparent windows to allowdirect visual inspection. A series of couplings located on the back of thecell in planes parallel and perpendicular to the main axis of the cell body

49

http://www.stfi.se

Marco Lucisano

allow local measurements with thermocouples, resistivity probes and othersensors. Cellulose fibers are packed in the measuring cell and compressedby means of a permeable piston, thereby guaranteeing hydraulic continuitybetween the cooling water and the water trapped in the fiber mat. Theconsistency of the fiber mat is set by adjusting the position of the pressingpiston before the system was closed. Consequently, the structural networkpressure and the hydraulic pressures in the sample cell can be set indepen-dently of each other, allowing a wider range of operating conditions thanthose achievable in press situations where the two pressures are coupled toeach other through the externally applied load. The maximum oil temper-ature in the heating loop is 180 ◦C and the cooling water reservoir can bepressurized to 0.6 MPa (absolute). Two solenoid valves were mounted on thepressurization vessel allowing pressure release in time intervals ranging fromca 50 ms to 10 s. A LabVIEW-based control system (National Instruments)provided data acquisition, heat-up control and system pressure control.

Configuration A Configuration B

Wire-covereddrilled heater

Fiber Cake

Spring

Solid heater

Fiber Cake

Spring

V2

P

V1

Expansion Tank

Figure 24: Schematic of STFI’s FlashLab with (A) a wire-covered drilledheater and (B) a solid heater. The arrows indicate the direction of dewa-tering upon pressure release. In Configuration A, the relative magnitudeof drainage through the heater and through the backing wire can be setby adjusting valves V1 and V2.

In the study of Paper E, we equipped FlashLab with two differentheaters: in the first case (A), we allowed bi-directional drainage through thebacking wire (upwards, Figure 24A), whereas in the second configuration(B) drainage took place through the backing wire only, with a solid copperheater (Figure 24B).

50


8.4.2 Visualization Experiments

The absence of steam generation before pressure is released implies that allphase change phenomena of interest for impulse pressing are connected toflashing of liquid water to steam, which takes place on nip decompression. Paper EIn Paper E, we performed model experiments with STFI’s FlashLab inorder to shed some light on the role of flashing in impulse pressing. Wesimulated the physical conditions immediately prior to flashing by heating athick water-saturated fiber cake to temperatures exceeding 100 ◦C, while thehydraulic pressure of the system was increased well above the atmosphericlevel to prevent evaporation. Additionally, a structural pressure was gen-erated by compressing the fiber cake between the heater and a permeablewire (Figure 23).

The experiments showed that the phenomena induced by flashing uponpressure release were strongly dependent on the path of dewatering allowedin the sample cell. Figures 25A1 and 25A2 illustrate delamination occur-ring as a result of flashing. Here a wire-covered drilled heater was used, Delaminationwhich allowed drainage both through the backing wire and through theheater. When the hydraulic pressure war released, the bidirectional flowof vapor disrupted (delaminated) the fiber cake in the sample holder. Theposition of the damaged zone changed when the flow through the heater wasincreased, which could be done by adjusting valves V1 and V2 in Figure 24(Configuration A). The situation experienced by the fiber cake in Config-uration A is analogous to that experienced by a paper sheet at the exit ofan impulse nip, where steam and water are vented both towards the pressroll and towards the felt (e.g. Larsson and Stenstrom, 1998; Orloff et al.,1998b,c). Displacement

DewateringDelamination disappeared completely when a solid heater was used (Fig-

ures 25B1 and 25B2). In this case, steam generated by flashing could bevented through the backing wire only (Figure 24, Configuration B). Thiscaused three phenomena to occur: (i) initially, the expanding steam cham-ber on top of the fiber cake compressed the spring supporting the wire; (ii)additionally, the fiber cake was consolidated by volume reduction due to thevapor pressure. Yet, the most interesting phenomenon was (iii) the displace-ment of liquid water from the fiber cake by action of the expanding steamlayer.

We propose that this mechanism should be termed “flashing-assisted dis-placement dewatering” since it differs from the steam-assisted displacement ⇒ Flashing-Assisted

DisplacementDewatering

of liquid water originally proposed by Wahren (1982) because of the dif-ferent driving force. As the hydraulic pressure was released, a “dry line”moved through the entire fiber cake displacing water in the intrafiber poros-ity. Although the steam front was not perfectly straight, water displacement

51

http://www.stfi.se

Marco Lucisano

Initial conditionsA0

Initial conditionsB0

DelaminationA1

ConsolidationB1

Delamination + water displacement

A2

Consolidation + water displacement

B2

Figure 25: Path of events on pressure release in STFI’s FlashLab forConfiguration A (A0 − A1 − A2) and Configuration B (B0 −B1 −B2) inFigure 24. Prior to pressure release, fiber cakes with initial solids contentapproximately 20% were heated from above (2 h, 160 ◦C) under 0.7 MPa(absolute) hydraulic pressure. The lines in subfigures A1, A2 and B2

indicate the position of the expanding steam front (dry line).

52


0 1 2 3 4 5 6 7 8 90

20

40

60

80

100

120

140

160

180

Time [s]

Tem

pera

ture

[°C

]

T15

T33

T51

T71

A0 5 10 15

0

20

40

60

80

100

120

140

160

180

Time [s]

Tem

pera

ture

[°C

] Tsat

(P)

T15

T33

T51

T71

B

Figure 26: Temperature profiles measured in STFI’s FlashLab: A: ther-mal profile when flashing water in a SBK web with ca 40% dryness. Thetemperature of the heating fluid was 170 ◦C and the initial water pressure0.5 MPa (absolute). B: thermal profile when flashing an open pool of wa-ter heated from above. The temperature of the heating fluid was 140 ◦Cand the initial pressure 0.5 MPa (absolute). Subscripts indicate the dis-tance in mm from the hot surface to the point at which the temperatureis measured.

resulted in a reasonably uniform dewatering of the fiber cake. Experimentsrepeated with SBK fibers pressed to different initial solids contents (10-40%)showed flashing-assisted displacement dewatering for all tested conditions.Indeed, steam fronts appeared even during experiments performed with Con-figuration A, but they were somewhat weaker and resulted in less uniformdewatering.

In summary, the results of these flashing experiments show that dis-placement of liquid water by action of an expanding steam layer can play asignificant role in the physics of dewatering in an impulse nip. Additionally,our model experiments showed that delamination can be prevented by elim-inating its ultimate causes. However, care should be taken when attemptingto apply the principles discussed above to actual papermaking operations.In fact, non-uniformities in sheet structure and formation could result in asignificant reduction of the efficiency of flashing-assisted displacement as adewatering mechanism, by favoring viscous fingering. This would be partic-ularly problematic with webs of low grammage.

8.4.3 Displacement Dewatering as a Heat Transfer Mechanism

In Paper B, we observed an increase in the recorded temperature at theend of the pressure pulse, when wet paper webs were pressed using a heatedplaten press. In order to investigate this occurrence, we used STFI’s Flash-Lab in Configuration B (Figure 24). Here, a 10 cm bed of SBK fibres, Experimentalwith an initial dryness of 40%, was pressurized to approximately 0.5 MPa

53

http://www.stfi.se

Marco Lucisano

(absolute), and subjected to a steady state linear temperature gradient of1.4 ◦C/mm by maintaining the upper boundary at 170 ◦C while cooling thelower boundary. Thermocouples were placed at four different elevationsalong the length of the apparatus, namely 15, 33, 51, and 71 mm away fromthe upper surface and are referred to as Ti, where i is the measurementposition. The boiling point of water at 0.5 MPa is 150 ◦C. When releasingthe hydraulic pressure, the following behavior was observed (Figure 26A):(i) the temperature near the hot plate (T15) decreased due to the suddenincrease in volume; (ii) in the central region of the fibre bed (T51) the tem-perature increased rapidly, followed by a slow temperature decrease; and (iii)the temperature near the bottom of the bed (T71) was essentially constantby during the experiment. This is shown in Figure 26A. These experimentswere repeated at temperatures lower than the boiling temperature of waterat 0.5 MPa (absolute), and the thermal history upon pressure release wasqualitatively similar to that of Figure 18A.

In a complementary study, the experiment was repeated with only water(Figure 26B). Here, the temperature of the hot oil was set to be 140 ◦Cand the initial pressure was ca 0.5 MPa (absolute). Even in this case a rel-evant temperature increase was recorded when the hydraulic pressure wasreleased in the water tank. When comparing video recordings with the mea-Water Poolsured temperature profiles, it was concluded that three zones were formed:(i) a layer of dry steam, the temperature of which essentially followed theequilibrium curve of water as pressure was reduced; (ii) a well mixed layerof evaporating liquid water; and (C) a region with water being forced down-wards under the action of the expanding steam layer.

On the basis of these model experiments, we suggest that flashing of hotwater, together with steam expansion in the opening press nip, could explainthe large temperature peaks measured during our platen press simulations.⇒ Displacement

Dewatering as a HeatTransfer Mechanism

As the sheets are unloaded, the specific volume of water on the hot side ofthe sheet increases either by flashing or by expansion of steam formed inprevious stages of the impulse event. Consequently, hot water is displacedthrough the wet web towards the felt, thereby heating up the sheet. Theprocess continues until one of the following three situations terminates it:(i) The press nip opens completely and vapor escapes towards the hot sideof the sheet (this is generally considered to be the mechanism for sheet de-lamination). (ii) The expanding vapor layer reaches the sheet-felt interfaceand it is vented to the atmosphere. A channelling phenomenon would en-hance this mechanism especially when working with low basis weights. (iii)A steady state is reached requiring no more expansion of the steam layer. Itshould be noticed that this mechanism is not dissimilar from that originallyproposed for impulse drying (Wahren, 1978, 1982).

54


The important thing in science is not so muchto obtain new facts as to discover new ways ofthinking about them.

Sir William Bragg (1890–1971)

9 Characterizing the Thermal Properties of WetPaper Webs with a Heat-Pulse Method

The results of Paper B — Paper E support the view of impulse press-ing as superhot pressing (as defined in Chapter 3.1) followed by flashing,a theory originally proposed by Macklem and Pulkowski (1988). In light Backgroundof the increasing importance of modelling for both scientific investigationsand process simulation and control, we proposed a new method and appara-tus for the simultaneous measurement of the thermal conductivity, thermaldiffusivity and volumetric heat capacity of water-saturated paper samples. Paper FIn fact, very few investigations of the thermal properties of wet fiber matshave been reported in the literature and the dryness range of interest wasthat typical for drying, calendering and electro-photographic printing (e.g.Kerekes, 1980; Kartovaara et al., 1985; Ratto and Rigdahl, 1998).

The method is based on a solution of Fourier’s equation of heat conduc-tion around a linear one-dimensional source. The equipment is designed formeasurements on water-saturated and unsaturated samples preconditionedto temperatures reaching 180 ◦C and pressed to dryness levels between 20%and 60%.

9.1 The Transient Line Source Technique

In its simplest form, the line-source method is based on the theory of heattransfer from an infinite one-dimensional line source, embedded in an infi-

55

Marco Lucisano

Figure 27: Sample curve for a measurement on water saturated blot-ter paper at 23 ◦C. The measurement was performed with probe dis-tance r = 2.5 mm at room temperature, with specific supplied heatq′ = 10.0 W/m, t0 = 22 s.

nite homogeneous, isotropic medium. Two thin parallel wires are embeddedMeasuring Principlein the sample: (i) a linear heater of small diameter and (ii) a thin thermo-couple. The temperature field within the sample is perturbed by an heatpulse generated when imposing a known voltage difference over the heater.The thermal response of the sample is then measured by the thermocouple(Figure 27).

In single point measurements (SPM), the thermal properties of the sam-

Exact Solution

ple are calculated using data measured at a single instant in time, most con-veniently at the point of maximum temperature, tm, corresponding to tem-perature increase ∆Tm (Figure 27). Yet, since single point measurementsrely completely on the exact determination of the maximum (tm,∆Tm), theiraccuracy is low when the sampled data are noisy or when the shape of thethermal response is such the (tm,∆Tm) cannot be determined with preci-sion. A better approach is to use a non-linear least-squares method (NLM)Method of Solutionto fit the analytic solution to the entire sampled data set (Bristow et al.,1995). We used a combination of both methods, where single points calcu-lations are used to provide approximated values of the thermal parameters,which are then used as initial guesses for the non-linear fitting routine. Thethermal properties of the sample were determined by using Gauss-Newtonmethod with Levenberg-Marquardt modifications as implemented in MAT-LAB Statistics Toolbox (The Mathworks, Inc.).

The experimental work was performed with a new laboratory apparatus

56


Figure 28: STFI’s two-wire cell: a novel experimental facility for mea-surement of thermal properties at high temperature and pressure.

specifically designed to measure the thermal properties of paper at hightemperature and pressure (Figure 28). The sample and the probe were A New Laboratory

Facilityenclosed in a vessel that can be pressurized to 0.7 MPa (absolute) and heatedto 180 ◦C.

9.2 Thermal Properties of Water-Saturated Paper

The experimental method was initially validated by measuring on water atroom temperature, after immobilization as a 2.5% agar gel, as suggested by Validationvan Haneghem et al. (1983) and Ren et al. (1999). Table 3 shows that theheat-pulse technique we selected could give measurement data of satisfactoryprecision.

We applied the heat pulse method to the measurement of the ther-mal conductivity, diffusivity and heat capacity of water-saturated blotterpaper at 23 ◦C and 83 ◦C and at solids content between 30% and 55%(Figure 29). The first issue to be considered is that the experimental re-sults present a considerable spread. Nevertheless, the overall trends appearsufficiently well-defined to be discussed here.

57

Marco Lucisano

Table 3: Comparison between measured and literature value for water atroom temperature immobilized as 2.5% agar gel.

Diffusivity Conductivity Heat Capacity[m2/s] [W/m · K] [J/m3

· K]

Measured 1.4 · 10−7 ± 0.1 · 10−7 0.61± 0.03 4.3 · 106 ± 0.2 · 106

Literature 1.43 · 10−7 0.60 4.17 · 106

0.3 0.35 0.4 0.45 0.5 0.55

0.6

0.8

1

1.2

1.4

1.6

Solids Content [−]

k [W

/m,K

]

Figure 29: Thermal conductivity, k, of water-saturated blotter paper at23 ◦C (•) and 83 ◦C (�).

Increasing the sample temperature resulted in no significant response onthe thermal conductivity of water-saturated blotters. This is in contrastwith results previously reported for the same temperature interval (Ned-erveen and Finken, 1992), where a strong effect of temperature on the ther-mal conductivity was observed. This suggests that the measurements wereinfected by mechanisms other than pure conduction. Indeed, the low sen-⇒ Temperature

Effects sitivity of thermal conductivity on temperature, as measured in this study,agrees with the general behavior of most common materials (Perry et al.,1984). Increasing temperature from 23 ◦C to 83 ◦C increased the volumet-ric heat capacity of the sample by approximately 50% (Figure 30), whichresulted in a decrease of the corresponding thermal diffusivity (Figure 31).

The effect of solids content on the thermal conductivity, diffusivity andheat capacity of water-saturated blotter paper is clearly illustrated in Fig-ure 29 through to Figure 31: both the thermal conductivity and the ther-⇒ Effects of solids

content mal diffusivity increase with solids, whereas the volumetric heat capacitydecreases with increasing solids.

The behavior of the thermal conductivity with solids content presents aninteresting anomaly, when compared with the normal practice to calculateAn Anomaly?

58


0.3 0.35 0.4 0.45 0.5 0.550

0.5

1

1.5

2

2.5

3

3.5x 10

6


ρ c

[J/m

3 ,K]

Figure 30: Volumetric heat capacity, ρcp, of water-saturated blotter pa-per at 23 ◦C (•) and 83 ◦C (�).

the thermal properties of wet paper by linear combination of the thermalproperties of water and of dry cellulose. In fact, the thermal conductivi-ties of dry cellulose and water are 0.157 W/m ·K (Kartovaara et al., 1985)and 0.60 W/m ·K, respectively. This would imply a decreasing trend withincreasing solids. Although at the present time we are unable to offer anexplanation for this discrepancy, we suggest that the thermal conductivityof fibers wetted by unpressable water only might be significantly differentfrom that of dry cellulose. Additionally, we are not aware of any previousmeasurements of the thermal properties of cellulose in an air-free environ-ment. The presence of air might seriously influence the measurements, thusresulting in a significantly lower thermal conductivity than that of a com-pletely air-free sample. If this was the case, a better approach to calculatethe thermal conductivity of water-saturated paper would be an appropriatecombination of the properties of water and wetted air-free fibers. Still, wewould need more experimental work to substantiate or reject this specula-tion.

59

Marco Lucisano

0.3 0.35 0.4 0.45 0.5 0.550

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8x 10

−6


α [m

2 /s]

Figure 31: Thermal diffusivity, α, of water-saturated blotter paper at23 ◦C (•) and 83 ◦C (�).

60


Now this is not the end. It is not even thebeginning of the end. But it is, perhaps, the endof the beginning.

Sir Winston Churchill (1874–1965)

10 Summary and Conclusions

The mechanisms of impulse pressing has been widely discussed in the lit-erature since the technique was conceived in the early 1970’s; yet no clearpicture has emerged. The first conceptual model was put forth by Wahrenwho speculated that a steam front would develop at the hot roll surfaceand, in theory, be able to displace any unbound water from the wet web.In contrast to this, it has also been suggested that impulse events couldbe interpreted as a combination of high-temperature (superhot) pressingand rapid flash evaporation. Despite this longstanding debate, there is verylittle experimental evidence to confirm these theories. Yet, it is clear thatevaporative phenomena play a central role in the physics of impulse technol-ogy, being connected both to its innovative mechanism and to delamination,the major hinder on the path towards commercial implementation. In thisthesis, we discuss one major aspect of impulse pressing, that being, themechanism of heat transfer with phase change in an impulse nip.

In this work, we propose an alternative path of events for impulse tech-nology and introduce the concept of “flashing-assisted displacement dewa-tering”, thereby unifying the two main theoretical approaches put forwardin the literature. Here we suggest that little or no steam is formed in animpulse nip prior to the point of maximum applied load. Because of theexternally applied pressure, the water in the wet web reaches temperatureshigher than the atmospheric boiling point without evaporating. As the nipopens, the hydraulic pressure decreases and the hot (subcooled) water vi-olently flashes to steam. It is widely agreed that flashing is the ultimate

61

Marco Lucisano

cause of sheet delamination, which takes place if the drag forces dissipatedby the steam flow is greater than the z-directional strength of the wet web.

We advance the argument that the force expressed on flashing can beused to drive displacement of liquid water, in a mechanism similar to thatoriginally proposed by Wahren. Additionally, model experiments performedin a novel experimental facility suggest that nip unloading could be carriedout to maximize flashing-assisted displacement dewatering, by controllingthe direction of steam venting. If this solution could be exploited in acommercially viable impulse press, the physics of this new process alternativewould be such that delamination would cease to be an issue of concern.

A study of the structural changes induced by the impulse process re-vealed no evidence that wet pressing and impulse pressing induced stratifica-tion in non-delaminated sheets (Paper A). We concluded that the paraboliccross-sectional solidity profiles observed were due to capillary forces presentduring drying. Further, the permeability of mechanically compressed never-dried samples was found to be essentially constant for pressing temperatureslower than the atmospheric boiling point of water and to increase signifi-cantly at higher pressing temperatures. We propose this to be a result ofdamage to the cell wall material due to flashing of hot liquid water in thefiber walls and lumina.

Finally, we present a method and an apparatus for measurement of thethermal properties of water-saturated paper webs at temperatures and pres-sures of interest for commercial high-intensity processes (Paper F). Aftervalidation, the method was applied to the measurement of the thermal con-ductivity, diffusivity and volumetric heat capacity of water-saturated blotterpaper, which were studied as functions of temperature and solids content.An interesting result was the observed behavior of the thermal conductivityas a function of solids content. Here, we measured an increasing trend, whichis in conflict with the traditionally accepted assumption which describes theconductivity of wet paper as a decreasing function of solids. Although nowell-substantiated explanation could be developed at this stage, the newapparatus provides the opportunity of new and interesting insights into thethermal properties of wet paper.

62


We used to think that if we knew one, weknew two, because one and one are two. Weare finding that we must learn a great deal moreabout ‘and’.

Sir Arthur Eddington (1882–1944)

11 Recommendations for Further Work

On the base of the model experiments performed in this work, we have sug-gested a novel path of events to describe the physics of phase change inimpulse pressing. In particular, we have suggested that delamination canbe eliminated using a novel dewatering mechanism. Here, the direction ofsteam venting upon nip unloading was controlled. The force of expandingsteam generated by flashing at nip unloading is the core of flashing-assisteddisplacement as a dewatering mechanism. Here, we suggest that a pressur-ized chamber, sealed as in Miller (1988c), should be arranged immediatelyfollowing the press shoe (Figure 32). Provided that the tension of the pressfelt can be controlled at will, the force of the expanding steam would resultin displacement of the unbound water present in the web and felt.

We suggest that the position of the turning roll with respect to the endof the press shoe should be such that sufficient time is provided for thesteam front to pass the interface between the web and the felt. This wouldminimize any effect of after-press rewetting since liquid water would not bepresent at the point of separation between the web and the felt.

Additionally, we propose that an impulse unit built to maximize theeffects of flashing-assisted water displacement should be used to replace thefirst press nip. In fact, since flashing can displace unbound water only(Persson and Stenstrom, 1999), the energy efficiency of the process is boundto increase with increasing initial moisture ratio. The roll temperature and

63

Marco Lucisano

Web

Felt

Heated Press Roll

Press Shoe

Flashing-AssistedDisplacement Dewatering

Figure 32: Schematic of a suggested configuration for an impulse pressdesigned for flashing-assisted displacement dewatering. The lift-off of theweb and felt in the flashing zone is exaggerated for the purpose of illus-tration.

length of the press shoe should be determined by the total energy needed toevaporate enough water to drive displacement dewatering. In particular, itis advantageous to select high roll temperatures, short contact times and rollcovers of high thermal mass to maximize energy efficiency (see Chapter 4.2).

Finally, it should be pointed out that flashing-assisted displacement de-watering is intrinsically asymmetric and that the installation of a second in-verted unit would be completely ineffective. On the contrary, we suggestedthat the last unit of an impulse technology based press section should consistof an impulse nip of more traditional construction, inverted with respect tothe first press and having the main purpose of correcting two-sidedness.

64


12 Acknowledgments

I would like to express my gratitude to my supervisors Ass. Prof. MarkMartinez, Prof. Mikael Rigdahl, Dr. Andrew Martin, Dr. Hannes Vomhoffand Prof. Bo Norman for their continuous encouragement, support andadvice. The many discussions with Marie Backstrom, Joanna Hornatowskaand Daniel Soderberg are highly appreciated.

Sune Karlsson’s invaluable assistance in planning and performing ex-periments rendered life in the laboratory much easier. Albin Andersson,Annika Bergman, Annette Lofstrand, Jimmy Moller and Alessandro Stranoare thanked for their patience and precision in carrying out many of theexperiments on which this work is based.

Odd Karlsen’s continuous support and ingenious solutions during thenever-ending design and construction of STFI’s FlashLab are acknowledgedwith gratitude. The efforts of Odd Karlsen and Jarmo Tolunen resulted inthe construction of the experimental cell for measurement of the thermalproperties of paper.

My colleagues and friends at STFI have always been a source of inspi-ration and I wish to thank everyone for an interesting and fruitful workingenvironment. Dario Cuva, Cecilia Land, Paolo Mazzatorta, Joachim Petriniand Marıa Pilar Ruiz Roman kept me alert by making sure that no twoworking days were alike.

Financial support from the Foundation for Strategic Research, ForestProduct Industry Research College (FPIRC) program, the Swedish ResearchCouncil and STFI’s Impulse Technology program is gratefully acknowledged.

Stockholm, October 28th, 2002

Marco Lucisano

65

http://www.stfi.se

http://www.stfi.se

http://www.pmt.kth.se/fpirc/INDEX_E.HTM

http://www.pmt.kth.se/fpirc/INDEX_E.HTM

http://www.stfi.se

Marco Lucisano

13 Notation

a Coefficient in Equation 7 dimensionlessb Coefficient in Equation 7 dimensionlessc Heat capacity W/m3 ·Kh Thickness of the steam layer mhf Enthalpy of fibers J/kghv Latent heat of vaporization J/kghw Enthalpy of water J/kgHT-CTMP High-temperature chemical

thermomechanical pulpk Thermal conductivity W/m ·KK Thermal mass W

√s/m2 ·K

L Sheet caliper mm Mass kgMR Moisture ratio kg/kgNLM Non-linear modelq′′h Heat flux W/m2

q′′h Supplied energy J/m2

P Pressure N/m2

SBK Softwood bleached kraftSPM Single point measurementSR Schopper-Rieglert Time st0 Duration of the heat pulse sT Temperature ◦CTi Temperature at i mm from the hot surface ◦Cu Superficial fluid velocity m/sw Basis weight g/m2

W !RV Water retention value kg/kgx Horizontal coordinate my Vertical coordinate m

Greek Symbolsα Thermal diffusivity m2/sε Constant in Equations 1 and 2 m/s0.5

εimp Impulse pressing efficiency dimensionlessκ Darcian permeability m2

κm Modified permeability kg/mλ Latent heat of evaporation J/kgµ Viscosity Pa · sρweb Paper density kg/m3

66


Subscripts0 Initial statebd Bone-dryevap Evaporatedf FiberH Hot surfacei Initiall Liquid phasem Midnipref References saturation conditionsth, lat Theoretical, latentth, sen Theoretical, sensibletot Totalv Vapor phasew Water

67

Marco Lucisano

References

Ahrens, F. W. and Astrom, A. 1986 High-intensity drying of paper.Drying Technology 4 (2), 145–270.

Andersson, L. and Back, E. L. 1981 The effect of temperature up to90°C on dewatering of wet paper webs, evaluated in a press simulator. InTAPPI Engineering Conference, Atlanta, GA, pp. 311–323. TAPPI Press.

Arenander, S. and Wahren, D. 1983 Impulse drying adds new dimen-sion to water removal. Tappi Journal 66 (9), 123–126.

Babinsky, V. A. and Mumford, W. G. 1995 Chambered nip drying ofpaperboard webs. U.S. Patent no. 5,404,654.

Back, E. L. 1979 Consolidation — A look into the future (including hotpress nips). Paper Technology and Industry 20 (9), 304–309.

Back, E. L. 1984 A review of press drying. Tech. Rep. D-224. STFI, Stock-holm.

Back, E. L. 1986 Repeated hot extended press nips as compared to pressdrying. In New Technologies in Web Consolidation and Drying , vol. 2.Brighton, Sussex: PIRA.

Back, E. L. 1990 Presstorkning/impulstorkning — Begransademojligheter. Svensk Papperstidning 93 (7), 131–132, 134, 136, 138(In Swedish).

Back, E. L. 1991a Developments in drying technologies — A literaturereview. Tech. Rep. Pira Information Services, Leatherhead, UK.

Back, E. L. 1991b Press drying/impulse drying: The author’s answer toLindsay. Tappi Journal 74 (9), 239–240.

Back, E. L. 1991c Why is press drying/impulse drying delayed? TappiJournal 74 (3), 135–147.

Back, E. L. and Swenson, R. 1981 The present state of press-drying ofpaper. In The Role of Fundamental Research in Papermaking , vol. 1, pp.343–384. Cambridge: Mechanical Engineering Publishing Ltd.

Baker, C. F. 1999 Overview of pressing and drying techniques for improvedpaper mill productivity. In Scientific and Technical Advances in Pressingand Drying . Edinburgh: PIRA International.

68


Betz, A. I. 1996 Impulse drying of paper — An experimental study ofdelamination. Master’s thesis, Department of Chemical Engineering I —Lund University, Sweden.

Bluhm, R., Holik, H. and Mirsberger, P. 1991 Method for themechanical-thermal dewatering of a fibre stock web. U.S. Patent no.5,071,513.

Boerner, J. 1992 The effect of basis weight on sheet permeability andcritical impulse-drying temperature. Master’s thesis, Institute of PaperScience and Technology, Atlanta, GA.

Boerner, J. and Orloff, D. I. 1994 Effects of basis weight and free-ness on sheet permeability and critical impulse-drying temperature. TappiJournal 77 (2), 163–168.

Bradner, D. B. 1929 Process of calendering mineral-coated paper andproduct. USA Patent no. 1,719,166.

Bristow, K. L., Bilskie, J. R., Kluitenberg, G. J. and Horton, R.1995 Comparison of techniques for extracting soil thermal properties fromdual-probe heat-pulse data. Soil Science 160 (1), 1–7.

Burns, J. R., Conners, T. E. and Dean, L. J. 1990 Dynamic measure-ments of density gradient development during wet pressing. Tappi Journal73 (4), 107–113.

Burns, J. R., Lindsay, J. D. and Conners, T. E. 1993 Dynamic mea-surements of stratified consolidation in a press nip. Tappi Journal 76 (8),87–96.

Burton, S. W. 1985 An investigation of dynamic densification under im-pulse drying conditions. Progress Report Six. Tech. Rep. The Institute ofPaper Chemistry.

Burton, S. W. 1987 An investigation of the z-directional profile develop-ment during impulse drying. PhD thesis, The Institute of Paper Chem-istry, Appleton, WI.

Burton, S. W. and Sprague, C. H. 1987 The instantaneous measure-ment of density profile development during web consolidation. Journal ofPulp and Paper Science 13 (5), J145–J150.

Burton, S. W., Sprague, C. H. and Ahrens, F. W. 1986 Densitydevelopment during impulse drying of paper. In 5th International DryingSymposium, vol. 1, pp. 364–370. Cambridge, MA.

Busker, L. H., Cronin, D. C., Macklem, E. A. and Pulkowsli, J. H.1988 Heated extended nip press apparatus. U.S. Patent no. 4,738,752.

69

Marco Lucisano

von Byrd, L. 1979 Press drying — Flow and adhesion of hemicellulose andlignin. Tappi Journal 62 (7), 81–84.

von Byrd, L. 1982 Drying and Heat Transfer Characteristics duringBench-Scale Press Drying of Linerboard , pp. 83–90. Washington: Hemi-sphere.

Campbell, W. B. 1947 The physics of water removal. Pulp and PaperMagazine of Canada 48 (3), 103–109, 122.

Carlsson, G., Linstrom, T. and Soremark, C. 1977 Expression ofwater from cellulosic fibers under compressive load. In Fibre-Water Inter-actions in Papermaking , vol. 1, pp. 389–409. Oxford: British Paper andBoard Industry Federation.

Carslaw, H. S. and Jaeger, J. C. 1959 Conduction of Heat in Solids,2nd edn. Oxford University Press.

Ceckler, W. T. and Thompson, E. V. 1982 The university of maine atorono wet pressing project. Tech. Rep. DOE/CS/40064-3 (DE83009342).United States Department of Energy, Orono.

Cheremisinoff, N. P. 1986 Instrumentation for Complex Fluid Flows.Lancaster, PA: Technomic Publ.

Cho, S. H. 1975 An exact solution of the coupled phase change problem ina porous medium. International Journal of Heat and Mass Transfer 18,1139–1142.

Christensen, A., Skjoldebrand and Hallstrom, B. 1982 SurfaceDrying in Heat Processes of Solid Foods, pp. 6–11. Washington: Hemi-sphere.

Crouse, J. W. 1997 Alloy coating for wet and high temperature pressingroll. U.S. Patent no. 5,632,861.

Crouse, J. W., Pulkowski, J. H. and Porter, R. J. 1991 Method forheated extended nip pressing. U.S. Patent no. 5,047,122.

Crouse, J. W. and Woo, Y. D. 1989 Delamination — A stumbling blockto implementing impulse drying technology for linerboard. Tappi Journal72 (10), 211–215.

Cutshall, K. A. 1993 Sheet consolidation and hot pressing. In PracticalAspects of Pressing and Drying , pp. 219–227. Atlanta, GA.

Cutshall, K. A. 1994 Sheet consolidation and hot pressing. In PracticalAspects of Pressing and Drying Short Course, pp. 221–229. TAPPI Press.

70


Danielsson, K. 1998 Identification of a wet pressing model for a researchpaper machine. Licentiate Thesis, Department of Signals, Sensors andSystems, Royal Institute of Technology, Stockholm.

Devlin, C. P. 1986 An investigation of the mechanism of high-intensitypaper drying. PhD thesis, Institute of Paper Chemistry, Appleton, WI.

Dreshfield, A. C. and Han, S. T. 1956 The drying of paper. TappiJournal 39 (7), 449–455.

Ellis, E. R. 1981 Compressibility and permeability of never dried bleachedsoftwood kraft pulp and its application to the prediction of wet pressingbehaviour. PhD thesis, University of Maine, Orono.

Ely, D. A. 1974 Grooved dryer. U.S. Patent no. 3,815,256.

Fougler, M. 1991 Hot pressing. In TAPPI Engineering Conference, vol. 1,pp. 305–311. Nashville, TN: TAPPI Press.

Gersbeck, R., Reschke, R. and Posielsky, R. 1975 Apparatus forthe continuous production of thin wood-wool boards. U.S. Patent no.3,874,962.

Gottwald, B. C. and Haigh, J. M. 1963 Method and apparatus for castcoating paper. U.S. Patent no. 3,110,612.

Gottwald, B. C., Halsey, B. S., Williams, R. C. and Haigh, J. M.1967 Continuous process of drying uncoated fibrous webs. U.S. Patent no.3,354,035.

Grover, G. M., Cotter, T. P. and Erickson, G. F. 1964 Structuresof very high conductance. Journal of Applied Physics 35, 1990–1991.

Gupta, L. N. 1974 Approximate solution of the generalized Stefan’s prob-lem in a porous medium. International Journal of Heat and Mass Transfer17 (2), 313–321.

Gustafsson, J.-E., Kaul, V. and Lobosco, V. 2001 Densification anddewatering in high temperature wet pressing. In The Science of Paper-making , vol. 1, pp. 679–708. FRC.

Haigh, J. M. 1967 Apparatus and method for drying a fibrous web. U.S.Patent no. 3,319,352.

van Haneghem, m. A., Schenk, J. and Boshoven, H. P. A. 1983An improved nonsetady-state probe method for measurements in granu-lar materials. Part II: Experimental results. High Temperatures — HighPressures 15, 367–374.

71

Marco Lucisano

Harper, J. C. 1955 The application of a suction hot press to the driersection of a paper machine. Tappi Journal 38, 229–233.

Hart, R. T. 1959 Process for finishing coated paper. U.S. Patent no.2,919,205.

Hasatani, M., Arai, N., Katsuyama, H., Harui, H., Itaya, Y.,Fushida, N. and Tatsukawa, N. 1991 Heat and Mass Transfer in Breadduring Baking in an Electric Oven, pp. 385–393. Amsterdam: Elsevier.

Heller, H. H. and Bliesner, W. C. 1977 A photographic stucy of wetpressing phenomena. In Engineeing Conference, pp. 1–7. Atlanta, GA:TAPPI Press.

Hill, J. M. 1987 One-Dimensional Stefan Problems: an Introduction, Pit-man Monographs on Pure and Applied Mathematics, vol. 31. Harlow:Longman.

Hollmark, H., Reiner, L., Lamb, H.-m. and Wallenius, H. 1999 Pa-per having a three-dimensional pattern. International Patent Applicationno. WO 99/36617.

Hollmark, H., Reiner, L., Wallenius, H., Billgren, T. and Klere-lid, I. 2000 Method of producing an extensible paper having a three-dimensional pattern and a paper produced by the method. InternationalPatent no. WO 00/20685.

Horn, R. A. 1979 Bonding in press-dried sheets from high-yield pulps —The role of lignin and hemicellulose. Tappi Journal 62 (7), 77–80.

Horn, R. A., Bormett, D. W. and Setterholm, V. C. 1986 Pressdrying recycled multi-ply boxboard. Tappi Journal 69 (11), 80–84.

Houvila, V. 2001 Experience from condebelt sturt-up at Dong Il. Interna-tional Paperworld (4), 545–549.

Karrila, S., Salmi, J., Paulapuro, H., Laapotti, J. and Palokan-gas, A. 1991 Laboratory simulation of impulse drying — Recent studiesat the FPPRI. In Helsinki Symposium on Alternate Methods of Pulp andPaper Drying , pp. 79–104. Finnish Pulp and Paper Research Institute.

Kartovaara, I., Rajala, R. et al. 1985 Conduction of heat in paper. InTransactions of the 8th Fundamental Research Symposium — Papermak-ing Raw Materials, vol. 1, pp. 381–412. London: Mechanical EngineeringPublications Ltd.

Kataja, M., Kirmanen, J. and J., T. 1995 Hydrostatic and strucuralpressure in compressed paper webs and press felts. Nordic Pulp and PaperResearch Journal 10 (3), 162–166.

72


Keller, S. F. 1992 Apparatus and method for removing fluid from a fibrousweb. U.S. Patent no. 5,152,874.

Kerekes, R. J. 1980 A simple method for determining the thermal con-ductivity and contact resistence of paper. Tappi Journal 63 (9), 137–140.

Krikorian, G. 1967 Absorbent dryer. U.S. Patent no. 3,296,710.

Krook, R., Hollmark, H. and Stenstrom, S. 1996 Some effects ofpress drying on paper properties. Nordic Pulp and Paper Research Journal11 (3), 141–151.

Krook, R. and Stenstrom, S. 1998 Temperature gradients and heatflux measurements in hot pressing of paper. Experimental Heat Transfer11 (3), 221–240.

Langdon, E. 1991 Advances in hot pressing technology. Appita Journal44 (4), 245–246.

Larsson, H. 1999 Heat transfer and dewatering phenomena in impulsedrying. Licentiate Thesis, Department of Chemical Engineering 1, LundUniversity, Lund, Sweden.

Larsson, H. and Stenstrom, S. 1998 Critical pressure control of delam-ination in impulse drying. Tappi Journal 81 (7), 117–122.

Larsson, H. and Stenstrom, S. 2001 Experiments in a pilot-scale im-pulse unit — Energy efficiencies. Nordic Pulp and Paper Research Journal16 (2), 162–167.

Larsson, H. and Orloff, D. I. 2000 Lab-scale impulse drying exper-iments with extension to higher impulse. In 12th International DryingSymposium. Noordwijkerhout, The Netherlands.

Lavery, H. P. 1987a High-intensity drying processes — Impulse drying —Report Two. Tech. Rep. 81. United States Department of Energy.

Lavery, H. P. 1987b Impulse drying of linerboard. In Papermakers Con-ference, pp. 43–49. Atlanta, Ga.

Lavery, H. P. 1987c Impulse drying of newsprint. Journal of Pulp andPaper Science 13 (6), J178–J184.

Lavery, H. P. 1988a Impulse drying of newsprint. In Annual Meeting —Technical Section, Canadian Pulp and Paper Association, vol. B, pp. 121–126. Montreal, Que.

Lavery, Hugh, P. 1988b High-intensity drying processes — Impulse drying— Report Three. Tech. Rep. 80. United States Department of Energy.

73

Marco Lucisano

Lehtinen, J. A. 1978 Method of drying a cardboard of a paper web anddrying device for applying this method. USA Patent no. 4,112,586.

Lehtinen, J. A. 1984 Method of continuous drying of a paper or otherporous web and a drying device for applying this method. USA Patentno. 4,461,095.

Lehtinen, J. A. 1986a A critical review of the conventional hot platenpressing method of laboratory press drying. Drying Technology 4 (2),183–215.

Lehtinen, J. A. 1986b Press consoidation — The condensing belt system.In New Technologies in Web Consolidation and Drying , vol. 2. Brighton,Sussex: PIRA.

Lehtinen, J. A. 1992 The heat pipe process in intraweb heat transfer inhot surface paper drying. Paper ja Puu — Paper and Timber 74 (7),560–571.

Lehtinen, J. A. 1995a Condebelt drying of paper and paperboard foroptimizing quality and production of many grades. Drying Technology13 (8), 2049–2067.

Lehtinen, J. A. 1995b Condebelt drying: Quality results and process de-velopment. Paper Technology 36 (10), 67–71.

Liang, S., McCarville, D. and Lange, D. 1999 Extended nip pressapparatus. U.S. Patent no. 5,980,693.

van Lieshout, M. 1999 Impulse drying: The state of the technology. PaperTechnology 40 (6), 35–40.

Lindsay, J. D. 1989 The physics of impulse drying: New insights fromnumerical modelling. In 9th Fundamental Research Symposium — Fun-damentals of Papermaking , vol. 2, pp. 679–729. Mechanical EngineeringPublications Ltd.

Lindsay, J. D. 1991 Discussion of impulse drying physics: Comments on E.L. Back’s “Why is press drying/impulse drying delayed?” Tappi Journal74 (9), 238–242.

Lindsay, J. D. and Sprague, C. H. 1989 MIPPS: a numerical moving-boundary model for impulse drying. Journal of Pulp and Paper Science15 (4), J135–J141.

Lobosco, V. and Kaul, V. 2001 An elastic/viscoplastic model of the fibrenetwork stress in wet pressing. Part 2: Accounting for pulp properties andweb temperature. Nordic Pulp and Paper Research Journal 16 (4), 313–318.

74


Lucisano, M. F. C. 2000 On the mechanism of heat transfer with phasechange during impulse pressing of paper. Licentiate Thesis, Department ofPulp and Paper Chemistry and Technology, Royal Institute of Technology,Stockholm.

Luikov, A. V. 1975 Systems of differential equations of heat and masstransfer in capillary porous media. International Journal of Heat and MassTransfer 18 (1), 1–14.

Luiten, E. E. M. 2001 Beyond energy efficiency — Actors, networks andgovernment intervention in the development of industrial process technolo-gies. PhD thesis, University of Utrecht, Leidschendam, The Netherlands.

MacGregor, M. A. 1983 A description of sheet stratification caused bywet pressing. Tappi Journal 66 (6), 53–57.

MacGregor, M. A. 1989 Wet pressing in 1989 — A historic perspective,analysis and commentary. In Transactions of the 9th Fundamental Re-search Symposium, vol. 2, pp. 511–586. London: Mechanical EngineeringPublications Ltd.

Macklem, E. A. and Pulkowski, J. H. 1988 Impulse drying — A press-ing/flashing drying phenomenon. In TAPPI Engineering Conference, pp.447–453. Atlanta, GA.

Martinez, D. M. 1999 An estimate of the in-plane velocity difference be-tween the fibres and water during wet pressing: The saturated case. Jour-nal of Pulp and Paper Science 25 (12), 403–409.

Martinez, D. M., Drotz, M., Lai, R. and Martin, A. R. 2001 Dry-ness and physical properties of paper webs in laboratory-scale impulseprocessing. Drying Technology 19 (10), 2435–2450.

Mason, W. H. 1931 Process and apparatus for disintegration of fibrousmaterial. U.S. Patent no. 1,824,221.

Mason, W. H. 1938a Momentary-pressure process of making ligno-celluloseproducts. U.S. Patent no. 2,140,189.

Mason, W. H. 1938b Process for making ligno-cellulose fiber products.U.S. Patent no. 2,120,137.

Mason, W. H., Boehm, R. M. and Koonce, W. E. 1937 Molding com-position and process of making the same. U.S. Patent no. 2,080,078.

Mason, W. H., Boehm, R. M. and Simpson, G. G. 1939 Process formaking hardboard. U.S. Patent no. 2,317,394.

75

Marco Lucisano

Mazer, J. 1956 Aquavac — New method for increasing dryiness. PaperTrade Journal 140 (14), 34.

Mazer, J. 1957 Aquavac: a new drying method. Tappi Journal 40 (9),233A.

McConnell, R. R. 1980 A review of the state-of-the-art in paper drying— Report One. Tech. Rep.. Institute of Paper Chemistry, Appleton, WI.

McDiarmid, R. and Cutshall, K. A. 1982 Steam showers increasnewsprint machine press-section efficiency. Pulp and Paper 56 (8), 110–112.

Meirmanov, A. M. 1992 The Stefan Problem, de Gruyter Expositions inMathematics, vol. 3. Berlin: Walter de Gruyter.

Mendes, N., Philippi, P. C. and Lamberts, R. 2002a A new mathemat-ical method to solve highly coupled equations of heat and mass transfer inporous media. International Journal of Heat and Mass Transfer 2002 (4),509–518.

Mendes, P., Belgacem, M. N., Aguilar, H., Ranita, J., Costa,A. P., Silvy, J. and Costa, C. A. V. 2002b Impulse drying effect onsome properties of paper made from hardwood and softwood portoguesepulps. Paperi ja Puu — Paper and Timber 84 (4), 263–273.

Mesic, N. 2002 Effects of chemicals on the delamination and physical prop-erties of impulse-pressed paper sheets. Licentiate Thesis, Department ofPulp and Paper Chemistry and Technology, Royal Institute of Technology,Stockholm.

Michell, A. J. 1984 Press drying of paper — An overview. APPITA 37 (4),325–329.

Miller, R. R. 1987 Belt and drum-type press. U.S. Patent no. 4,710,271.

Miller, R. R. 1988a Belt and drum-type pressing apparatus. U.S. Patentno. 4,758,310.

Miller, R. R. 1988b Heated drum having high thermal flux and belt pressusing the same. U.S. Patent no. 4,781,795.

Miller, R. R. 1988c Method and apparatus for treatment of a permeableweb with a fluid. U.S. Patent no. 4,740,305.

Miller, R. R. 1989a Belt and drum-type press with supplemental niploading means. U.S. Patent no. 4,877,487.

76


Miller, R. R. 1989b High heat flux roll and press utilizing same. U.S.Patent no. 4,889,048.

Miller, R. R. 1991 Multi-pulse units boost drying rate, reduce sheet de-lamination problems. Pulp and Paper 65 (5), 145–149.

Nederveen, C. J. and Finken, J. G. M. 1992 Thermal conductivity mea-surements on wet paper samples at high temperatures. Drying Technology10, 189–198.

Nilsson, F. 2001a Impulse drying of paper using a thin heated metal band.Licentiate Thesis, Department of Pulp and Paper Chemistry and Tech-nology, Royal Institute of Technology, Stockholm.

Nilsson, F. and Norman, B. G. 2002a Impulse drying using a thin heatedmetal band with backing insulation — Part 1: Dewatering, delaminationand temperature events. Nordic Pulp and Paper Research Journal 17 (3),326–332.

Nilsson, F. and Norman, B. G. 2002b Impulse drying using a thin heatedmetal band with backing insulation — Part 2: The influence of thermalenergy on dewatering and delamination. Nordic Pulp and Paper ResearchJournal 17 (3), 333–339.

Nilsson, J. 1998 A literature survey of impulse drying, 1994–1998. Tech.Rep. LUTKDH/(TKKA-7004)/1-36/1998. Lund University.

Nilsson, J. 2001b Heat transfer mechanisms in high-temperature pressingof paper. Licentiate Thesis, Department of Chemical Engineering I, LundUniversity, Sweden.

Nilsson, P. and Larsson, K. O. 1968 Paper web performance in a pressnip. Pulp and Paper Magazine of Canada 69 (24), 66–73.

Norberg, K. G. and Back, E. L. 1967 Pressbetingelseras inverkanpa lufttorkade arks tathet och styrkeegenskaper. Svensk Papperstidning70 (20), 668–672, (In Swedish).

Orloff, D. I. 1991a High-intensity drying processes — Impulse drying —Report Four. Tech. Rep. DOE/CE/40738-T4. United States Departmentof Energy.

Orloff, D. I. 1991b Impulse drying: Heterogeneous press roll surfaces offernew product properties. In Nonwovens Conference, pp. 393–397. MarcoIsland, FL: TAPPI Press.

Orloff, D. I. 1992a A comparison of impulse drying to doublefelted pressing on pilot-scale shoe presses and roll presses. Tech. Rep.DOE/CE/40738-T6. United States Department of Energy.

77

Marco Lucisano

Orloff, D. I. 1992b Impulse drying of linerboard: Control of delamination.Journal of Pulp and Paper Science 18 (1), J23–J32.

Orloff, D. I. 1992c Impulse drying of paper: A review of recent reserach.Tech. Rep. IPST Technical Paper Series 427. Institute of Paper Scienceand Technology.

Orloff, D. I. 1994a Impulse drying of recycled multi-ply linerboard: Lab-oratory scale studies. Tappi Journal 77 (2), 169–179.

Orloff, D. I. 1994b Method and apparatus for drying web. U.S. Patentno. 5,353,521.

Orloff, D. I. 1994c Method and apparatus for drying web. U.S. Patentno. 5,327,661.

Orloff, D. I. 1998 Impulse drying of board grades: An emerging technol-ogy. Paper Age 114 (12), 22–23.

Orloff, D. I. and Crouse, J. W. 1998 Impulse drying: Status of thepilot-scale research program. In Engineering Conference, vol. 3, pp. 879–890. Miami, FL: TAPPI.

Orloff, D. I. and Crouse, J. W. 1999 Impulse drying: Status of thepilot scale research program. Tappi Journal 82 (9), 143–149.

Orloff, D. I., Kloth, G. R. and Rudemiller, G. R. 1992 Methodand apparatus for drying web. U.S. Patent no. 5,101,574.

Orloff, D. I. and Lenling, W. J. 1993 Method and apparatus for dryingweb. U.S. Patent no. 5,272,821.

Orloff, D. I. and Lindsay, J. D. 1992a The influence of yield, refiningand ingoing solids on the impulse drying performance of a ceramic coatedpress roll. In TAPPI Papermakers Conference, vol. 1, pp. 85–93. Atlanta,GA: TAPPI Press.

Orloff, D. I., Patterson, T. F. and Krause, A. M. 1998a Open-ing the operating window of impulse drying — I: The effect of ambientpressure at the nip opening. Tappi Journal 81 (7), 113–116.

Orloff, D. I., Patterson, T. F. and M., K. A. 1997a Method and ap-paratus for drying a fiber web at elevated ambient pressures. U.S. Patentno. 5,598,642.

Orloff, D. I., Patterson, T. F. and Parviainen, P. M. 1998b Open-ing the operating window of impulse drying — II: Pressure differential asa source of delamination. Tappi Journal 81 (8), 195–203.

78


Orloff, D. I., Patterson, T. F., Phelan, P. M. and Rudman, I.1998c Opening the operating window of impulse drying — III: Controlleddecompression experiments. Tappi Journal 81 (9), 158–162.

Orloff, D. I. and Phelan, P. M. 1993 Heat transfer during impulse dry-ing: The influence of press impulse and pressure profile. In 29th NationalHeat Transfer Conference, vol. ASME HTD238, pp. 17–23. Atlanta, GA:ASME.

Orloff, D. I., Phelan, P. M. and Asensio, C. 2000a Impulse drying ofa south african furnish. In Engineering Conference. Atlanta, GA: TAPPIPress.

Orloff, D. I., Phelan, P. M. and Challas, J. 1999 Impulse dryingof board grades: Converting trials. In Engineering/Process and ProductQuality Conference, vol. 1, pp. 51–67. Anaheim, CA: TAPPI.

Orloff, D. I., Phelan, P. M. and Crouse, J. W. 1995 Linerboard dry-ing on a sheet-fed pilot impulse drying shoe press. Tappi Journal 78 (1),129–141.

Orloff, D. I., Phelan, P. M. and Crouse, J. W. 2000b Impulse dryingof board grades: Pilot production trials. Tappi Journal 83 (9).

Orloff, D. I. and Sobczynski, S. F. 1993 Impulse drying pilot pressdemonstration: Ceramic surfaces inhibit delamination. Paper Technology34 (10), 24–32.

Orloff, D. I. and Lindsay, J. D. 1992b The influence of yield, refiningand ingoing solids on the impulse drying performance of a ceramic coatedpress roll. In Papermakers Conference, vol. 1, pp. 85–93. TAPPI Press.

Pattersson, T. F. and Iwamasa, J. M. 1999 Review of web heating andwet pressing literature. In TAPPI Papermakers Conference, vol. 3, pp.1255–1278. Atlanta, GA: TAPPI Press.

Paulapuro, H. 1988 Pressing and drying — A revolution ahead? In XXIIIEUCEPA Conference, vol. 1, pp. 245–267. Harrogate: EUCEPA.

Paulapuro, H. 2001 Wet pressing — Present understanding and futurechallenges. In The Science of Papermaking , vol. 1, pp. 639–678. Oxford:FRC.

Perry, R. H., Green, D. W. and Maloney, J. O., ed. 1984 ChemicalEngineers’ Handbook , 6th edn. New York: McGraw-Hill.

Persson, H. 1994 A literature review of impulse drying, 1988–1994. Tech.Rep. LUTKDH/(TKKA-7005)/1-42/1994. Lund University.

79

Marco Lucisano

Persson, H. and Stenstrom, S. 1999 NIPSIM simulator for the experi-mental investigation of delamination and dewatering phenomena relatedto heated press nips. In TAPPI Engineering Conference, vol. 1, pp. 85–94.Anaheim, CA: TAPPI Press.

Persson, J. 2000 En kritisk granskning av metoder for att detektera de-laminering i papper. Master’s thesis, Abo Akademi, Kemisk TekniskaFakulteten, Abo, Finland, (In Swedish).

Phelan, P. M., Kerschner, C. M. and Orloff, D. I. 1995 The effectof thermal mass and peak pressure on impulse drying energy transfer. InTAPPI Engineering Conference, vol. 2, pp. 651–658. Dallas, TX: TAPPIPress.

Phelan, P. M., Orloff, D. I. and Kerschner, C. M. 1996 The effectof thermal mass and peak pressure on energy transfer in impulse drying.Tappi Journal 79 (11), 131–137.

Pikulik, I. I., Sparkes, D. G., Poirer, N. A. and Crotogino, R. H.1995 Dewatering and consolidation of wet webs. In Annual Meeting Tech-nical Section, vol. A. Montreal, Canada: CPPA.

Poirier, D. J. and Sparkes, D. G. 1991 Impulse drying on a pilot papermachine. In Helsinki Symposium on Alternate Methods of Pulp and PaperDrying , pp. 279–298. Finnish Pulp and Paper Research Institute.

Pulkowski, J. H., Macklem, E. A. and Busker, L. H. 1989 Heatedextended nip press with porous roll layers. U.S. Patent no. 4,874,469.

Pulkowski, J. H., Macklem, E. A. and Busker, L. H. 1992 Heatedextended nip press with porous roll layer. U.S. Patent no. 5,082,533.

Ramaswamy, S. 1990 Analysis of heat and mass transfer during dryingof paper/board under conventional and high-intensity conditions. PhDthesis, State University of New York, Syracuse, NY.

Ramaswamy, S. and Lindsay, J. D. 1998 The role of vapor formationin high-intensity drying — Description and comparison of two models.Nordic Pulp and Paper Research Journal 13 (4), 299–309.

Ren, T., Noborio, K. and Horton, R. 1999 Measuring soil water con-tent, electrical conductivity, and thermal properties with a thermo-timedomain reflectometry probe. Soil Sci. Soc. Am. J. 63 (3), 450–457.

Retulainen, E. and Hamalainen, A. 2000 Three years of condebeltdrying at stora enso’s pankoski mill. Tappi Journal 83 (5), 84.

80


Riepen, M. 2000 An inside view on impulse drying phenomena by mod-elling. In 12th International Drying Symposium. Noordwijkerhout, TheNetherlands.

Rigdahl, M., Backstrom, M., Hedstrom, C.-G., Norman, B., Kil-ian, M. and Talja, R. 2000 Impulse technology on the EuroFEX ma-chine. TAPPI J. 83 (8), 90.

Rodden, G. 1987 Impulse drying and heat recovery head the list of thelatest developments. Pulp and Paper Canada 88 (9), 56–58.

Rosen, H. N. 1987 Recent Advances in Drying of Solid Wood , vol. 4, pp.99–146. New York: Hemisphere.

Rubinstein, L. I. 1971 The Stefan Problem, Translations of MathematicalMonographs, vol. 27. Providence, RI: American Mathematical Society.

Rudemiller, G. R. 1989 A fundamental study of boiling heat transfermechanisms related to impulse drying. PhD thesis, The Institute of PaperScience and Technology, Atlanta, GA.

Rudman, I. and Orloff, D. I. 2001 Dynamic thickness and temperaturemeasurements during wet pressing and impulse drying. In EngineeringConference. San Antonio, TX: TAPPI Press.

Ratto, P. and Rigdahl, M. 1998 Temperature distributions in a papersheet subjected to a short pressure pulse from a heated plate. Nordic Pulpand Paper Research Journal 13 (2), 101–106.

Ratto, P. 2001 On the compression properties of paper — Implicationsfor calendering. PhD thesis, Royal Institute of Technology, Stockholm.

Scheidegger, A. E. 1974 The Physics of Flow through Porous Media.Toronto: University of Toronto Press.

Schorger, A. W. 1934 Process and apparatus for drying fibrous sheet.U.S. Patent no. 1,951,710.

Setterholm, V. and Benson, R. E. 1977 Variables in press drying pulpsfrom sweetgum and red oak. U.S. Department of Agriculture, Forest Ser-vice.

Sparkes, D. G. and Poirier, D. 1990a Dynamic pilot scale press dryingequipment. In Modern Technologies in Pressing and Drying , vol. 2. Pira,paper 28.

Sparkes, D. G. and Poirier, D. 1990b Impulse drying at paprican. Pulpand Paper Canada 91 (4), T145–T150.

81

Marco Lucisano

Sprague, C. H. 1985 High-intensity drying processes. Tech. Rep. UnitedStates Department of Energy.

Sprague, C. H. 1987 Impulse drying and press drying: A critical compar-ison. In Papermakers Conference, pp. 35–41. Atlanta, Ga.

Sprague, C. H. 1988 New developments in web consolidation for paper-making. In Ekman Days, pp. 1–7. Stockholm: SPCI.

Sprague, C. H. and Burton, S. W. 1986 New mechanisms in web consol-idation. In New Technologies in Web Consolidation and Drying . Brighton:PIRA.

Sprague, C. H. and Lavery, H. P. 1988 Impulse drying: How mills cantake advantage of it. In Annual Meeting – Technical Section, CanadianPulp and Paper Association, , vol. B, pp. 215–221. Montreal, Que.

Steen, T. 1986 Press drying for surface smoothness improvement. In NewTechnologies in Web Consolidation and Drying , vol. 2. Brighton, Sussex:PIRA.

Stenstrom, S. 1989 Impulstorkning av papper. STU projekt 742-88-02837.Lunds Tekniska Hogskolan, (In Swedish).

Stenstrom, S. 1990a Delamination in impulse drying: Some comments toimprove dryer design. Tappi Journal 73 (9), 259–260.

Stenstrom, S. 1990b Drying and energy. In 24th EUCEPA Conference —Paper Technology , pp. 220–235. Stockholm.

Stenstrom, S. 2000 Energirelaterad forskning avseende impulstorkning —Slutrapport 1997–1999. Tech. Rep. EFI 00/6. Statens Energimyndighet,(In Swedish).

Svanholm, E. 2002 Coating of impulse pressed fine paper — Influenceof the coating color composition. Master’s thesis, Department of Pulpand Paper Chemistry and Technology, The Royal Institute of Technology,Stockholm, Sweden, (In Swedish).

Szikla, Z. 1992 On the basic mechanisms of wet pressing. PhD thesis,Helsinki University of Technology, Helsinki.

Szikla, Z. and Paulapuro, H. 1989 Changes in z-direction density dis-tribution of paper in wet pressing. Journal of Pulp and Paper Science15 (1), J11–J17.

Taler, J. 1996 Theory of transient experimental techniques for surfaceheat transfer. International Journal of Heat and Mass Transfer 39 (17),3733–3748.

82


TAPPI 2000 Practical papermaking — The pressing operation. CD-ROM— ISBN: 1-930657-74-9.

Terzaghi, K. 1944 Theoretical Soil Mechanics. New York: John Wiley andSons.

van Tilburg, J. 1975a Expanding fibrous or plastic material by addingpuffing agent under pressure and subsequent pressure release. USA Patentno. 3,880,705.

van Tilburg, J. 1975b Expanding fibrous or plastic material by addingpuffing agent under pressure and subsequent pressure release. USA Patentno. 3,880,705.

Timofeev, O. and Mujumdar, A. S. 1993 Press drying, condebelt dryingand impulse drying of paper — A selective bibliography. Drying Technol-ogy 11 (7), 1885–1902.

Udell, K. S. 1983 Heat transfer in porous media heated from above withevaporation, condensation and capillary effects. Journal of Heat Transfer105 (3), 485–492.

Udell, K. S. 1985a An exact solution to high-temperature drying of highpermeability capillary porous media. In ASME National Heat TransferConference. Denver: ASME, paper no. 85-HT-66.

Udell, K. S. 1985b Heat transfer in porous media considering phase changeand capillarity — The heat pipe effect. International Journal of Heat andMass Transfer 28 (2), 485–495.

Udell, K. S. and Fitch, J. S. 1985 Heat and mass transfer in capillaryporous media considering evaporation, condensation and non-condensablegas effects. In 23rd National Heat Transfer Conference, vol. 45, pp. 103–110.

Udell, K. S. and Jennings, J. D. 1984 A composite porous heat pipe.In 5th International Heat Pipe Conference.

Vomhoff, H. 1998 Dynamic compressibility of water-saturated fibre net-works and influence of local stress variations in wet pressing. PhD thesis,Royal Institute of Technology, Stockholm.

Vomhoff, H. 2000 An apparatus and process for drying a paper web.International Patent no. PCT/SE99/01806.

Wahlstrom, B. 1960a A long term study of water removal and moisturedistribution on a newsprint machine press section — Part I. Pulp andPaper Magazine of Canada 61 (8), T379–T401.

83

Marco Lucisano

Wahlstrom, B. 1960b A long term study of water removal and moisturedistribution on a newsprint machine press section — Part II. Pulp andPaper Magazine of Canada 61 (9), T418–T451.

Wahlstrom, B. 1969 Our present understanding of the fundamentals ofpressing. Pulp and Paper Magazine of Canada 70 (10), 76–96.

Wahlstrom, B. 1991 Pressing — The state of the art and future possibil-ities. Paper Technology 32 (2), 18–27.

Wahlstrom, B. 2001 Wet pressing in the 20th century: Evolution, under-standing and future. Pulp and Paper Canada 102 (12), T333–T340.

Wahren, D. 1978 Forfarande och anordning for konsolidering och torkningav en fuktig poros bana. Swedish Patent no. 7803672-0.

Wahren, D. 1982 Method and apparatus for the rapid consolidation ofmoist porous web. USA Patent no. 4,324,613.

Wahren, D. 1983 Impulse drying. In Executives’ Conference, pp. 54–59.Appleton, WI: IPC.

Wahren, D. 1989 Rewetting inside the wet press nip. In Fundamentals ofPapermaking — Transactions of the 9th Fundamental Research Sympo-sium, vol. 3, pp. 228–230. London: Mechanical Engineering Publications.

Walker, K. 1990 Advances in hot pressing technology. Tappi Journal73 (8), 99–101.

Wedel, G. L. 1983 Press drying or dry pressing. In Press Dry Conference.Madison, WI: Forest Products Laboratory.

Wedel, G. L. 1995 Impulse dryer roll with shell of high thermal diffusivity.International Patent Application no. WO 95/10659.

Weise, U. 1998 Hornification — Mechanisms and therminology. Paperi jaPuu — Paper and Timber 80 (2), 110–115.

Westman, C., Lindstrom, T., Laine, J. and Norell, M. 2002 Processfor the production of paper. U.S. Patent Application no. 2002/0062938.

Weston, J. D. W. 1985 Press drying — A literature review. part of thepress consolidation programme. Tech. Rep.. Pira Information Services,Leatherhead, UK.

White, C. F. 1986 Press consolidation and drying. In New Technologies inWeb Consolidation and Drying , vol. 2. Brighton, Sussex: PIRA.

Wiberg, P.-A. 1987 Method and apparatus for drying a moist fibre web.U.S. Patent no. 4,691,449.

84


Wilcox, H. 1953 Interrelationships of temperature, pressure and pressingtime in the production of hardboard from douglas-fir fiber. Tappi Journal36 (2), 89–94.

Williams, H. R. 1932 Paper drying means. U.S. Patent no. 1,873,949.

Wiseman, N. 1976 Wet pressing in papermaking. Trend 26, 11–14.

Wrist, P. E. 1964 The present state of our knowledge of the fundamentalsof wet pressing. Pulp and Paper Magazine of Canada 65 (7), T284–T296.

Zavaglia, J. C. and Lindsay, J. D. 1989 Flash x-ray visualization ofmultiphase flow during impulse drying. Tappi Journal 72 (9), 79–85.

85


TH

Ts

P=Ps(T

s)

Vapor

Liquid

h(t)x

∞ T0

Figure 33: Evaporation of a liquid in a half-space.

Stefan Problem: The Neumann Solution

Understanding the mechanism of steam forming in the impulse nip is dif-ficult. Insight into this phenomenon can be gained by first examining asomewhat simpler, yet related, class of heat transfer problems traditionallycalled “moving boundary problems”, also referred to as “Stefan problems”.Stefan’s original work considered the heat flux in a medium that undergoesphase change at some fixed temperature (Ts), accompanied by the absorp-tion or liberation of latent heat. In our case, we consider a column of waterbeing heated from the top where a layer of steam is forming (see Figure 33).Evaporation takes place at a liquid-vapor front located at h = h(t). We willset up a solution of the problem under the following assumptions:

(i) One-dimensional semi-infinite liquid domain, originally at a tempera-ture T0 lower than the boiling point of the liquid Ts = Ts(Ps).

(ii) Heat transferred by conduction only.

(iii) The pressure in the vapor phase is kept constant, which implies thatTs = Ts(Ps) = const., throughout the process.

(iv) The density, thermal conductivity and heat capacity of both the liquidand the gas phase can are constants.

A time-dependent heat balance for the vapor phase gives:

DE1 :∂Tv

∂t= αv

∂2Tv

∂x2, for t > 0, and 0 ≤ x ≤ h(t), (12)

87

Marco Lucisano

and for the liquid phase:

DE2 :∂Tl

∂t= αl

∂2Tl

∂x2, for t > 0, and h(t) < x < ∞,

(13)

to be solved with initial condition:

IC : Tl = T0, for t = 0, in x > 0, (14)

and boundary conditions:

BC1 : Tv = TH at x = 0, for t > 0, (15)BC2 : Tl = T0 for x →∞ and t > 0. (16)

The coupling conditions at the evaporation front x = h(t) are given by:

BC3 : Tv = Tl = T0 at x = h(t), for t > 0, (17)

BC4 : − kv∂Tv

∂z+ kl

∂Tl

∂z= λρl

dh

dtat x = h(t), for t > 0. (18)

A solution that satisfies the differential equations and the initial andboundary conditions is given by:

Tv(x, t) = TH + A erf(

x

2√

αvt

), in 0 ≤ x ≤ h(t), (19)

Tl(x, t) = T0 + B erfc(

x

2√

αlt

), in h(t) ≤ x ≤ ∞. (20)

In order to guarantee a constant evaporation temperature, the positionof the liquid-vapor interface has to assume the form:

h(t) = εt12 , (21)

where ε is a constant yet to be determined. Constants A and B in Equa-tions (19) and (20) can be determined by considering BC3, Equation (17),thus:

A =Ts − TH

erf(

ε2√

αv

) , (22)

B =Ts − T0

erfc(

ε2√

αl

) , (23)

88


The temperature profiles in the two phases, Equations (19) and (20),can then be written as:

Tv(x, t) = TH +Ts − TH

erf(

ε2√

αv

) erf(

x

2√

αvt

), (24)

Tl(x, t) = T0 +Ts − T0

erfc(

ε2√

αl

) erfc(

x

2√

αlt

), (25)

BC4, Equation (18) can be used to determine the value of ε in Equa-tion (21). Equations (24) and (25) give:

∂Tv

∂x=

Ts − TH√παvt

e−14

x2

αvt

erf(

12

ε√αv

) , (26)

∂Tl

∂x= −Ts − T0√

παlt

e− 1

4x2

αlt

erf(

12

ε√αl

) . (27)

Derivation of Equation (21) with respect to time gives:

dh

dt=

ε

2√

t, (28)

(26), (27) and (28) into (18) give:

λρlε +2kv√παv

Ts − TH

erf(

ε2√

αv

)e−14

ε2

αv +2k2√παl

Ts − T0

erfc(

ε2√

αl

)e− 1

4ε2

αl = 0 (29)

which can be solved numerically for ε.

Using the physical properties of Table 4, Equations 24 and Equations 25can be used to calculate the heat-up history of a water pool initially atT0 = 25 ◦C heated from above by a source at TH = 300 ◦C with Ts = 100 ◦Cand Ps = 1 MPa (Figure 34).

89

Marco Lucisano

0 50 100 150 2000

50

100

150

200

250

300

Time [s]

Tem

pera

ture

[°C

]

A

0 5 10 15 200

20

40

60

80

100

120

Time [s]

Tem

pera

ture

[°C

]

B

Figure 34: Temperature profiles during the evaporation of a liquid halfspace. A = complete thermal history, B = heat-up of the liquid phase. Thetemperature curves report model predictions at 8 equally spaced points ina layer of total thickness 1.5 mm.

Table 4: Physical properties of water and steam.

Property Value Unitskl(T = 60◦C) 0.652 W/m ·Kρl(T = 60◦C) 983.2 kg/m3

cp,l(T = 60◦C) 4148 J/kg ·Kkv(T = 200◦C) 0.04 W/m ·Kρv(T = 200◦C) 7.84 kg/m3

cp,v(T = 200◦C) 3500 J/kg ·Kλ(T = 100◦C) 2.257 MJ/kg

Notation

h Thickness of the steam layer mt Time sT Temperature ◦CTi Temperature at i mm from the hot surface ◦Cx Spatial coordinate mGreek Symbolsα Thermal diffusivity m2/sε Constant in equations 1 and 2 m/s0.5

ρ Density kg/m3

λ Latent heat of evaporation J/kgSubscripts0 Initial stateH Hot surfacel liquid phases saturation conditionsv vapor phase

90


A Simplified Thermal Contact Model

Orloff (1991b); Orloff and Lenling (1993); Orloff (1994c) suggested the useof non-porous materials with low thermal mass (K ≤ 3000 W

√s/m2 ·K) for

delamination control. The choice of a hot surface with low thermal mass iseasily explained in the light of the mathematics describing heat conductionupon contact of two bodies originally at different temperatures.

x

02

2222 ,,,Tck p αρ

01

1111 ,,,Tck p αρ

∞−

∞+

Figure 35: Schematic of an infinite composite body.

Let us consider an infinite composite one-dimensional body, in whichthe region x > 0 is of one substance with constant thermal properties k1,ρ1, cp1, α1 and the region x < 0 is of another substance with constantthermal properties k2, ρ2, cp2, α2 (Figure 35). Assuming perfect thermalcontact between the two half-bodies, the temperature field in the systemcan is obtained by solving the Fourier equation:

∂T1

∂t= α1

∂2T1

∂x2for x > 0, t > 0, (30)

∂T2

∂t= α2

∂2T2

∂x2for x < 0, t > 0, (31)

with initial and boundary conditions:

T1(x, t) = T01 for x > 0, t = 0, (32)T2(x, t) = T02 for x < 0, t = 0, (33)T1(x, t) = T2(x, t) at x = 0, t > 0, (34)

k1∂T1

∂x= k2

∂T2

∂xat x = 0, t > 0. (35)

91

Marco Lucisano

Solving for the temperature profile, we get (Carslaw and Jaeger, 1959):

T1(x, t) =k1α

− 12

1 (T01 − T02)

k1α− 1

21 + k2α

− 12

2

1 +k2α

− 12

2

k1α− 1

21

erfx

2√

α1t

, (36)

T2(x, t) =k1α

− 12

1 (T01 − T02)

k1α− 1

21 + k2α

− 12

2

erfc|x|

2√

α2t, (37)

whereThermal mass

kiα− 1

2i = ki

√ρicpi

ki=

√kiρicpi ≡ Ki, (38)

which is defined as the thermal mass of material i.

0 5 10 15 20 250

50

100

150

200

250

Time [ms]

Tem

pera

ture

[°C

] A

B

C AB

C

SS1914

PTFE

Figure 36: Simplified thermal contact model between a wet paper sheet(modelled as water and initially at T01 = 25 ◦C) and a hot press surface(SS1914 or PTFE, initially at T02 = 300 ◦C). Temperatures A, B and Care given at 20, 40 and 60 µm from the interface surface (see Figure 35).

Equations 36 and 37 can be used to calculate the temperature profileat contact of a cold half-space (thermally described as water) with with ahot half-space of low thermal mass (PTFE) and high thermal mass (SS1914,Figure 36). This shows that the maximum temperature attained in the caseof PTFE is much lower than that for the metal platen. In fact, the maximumtemperature achievable in the paper web, T2,max, can be calculated usingthis simplified model to:

T2,max =K1

K1 + K2, (39)

where K1 is the thermal mass of the hot pressing plate and K2 is the ther-mal mass of the paper sample. Equation 39 implies that the maximumtemperature in the paper web can be reduced to any arbitrary value by anappropriate choice of the thermal mass of the hot platen material.

92


0 5 10 15 20 250

1

2

3

4

5

6

7

8x 10

4

Time [s]

Ene

rgy

[J/m

2 ]

tSS

tCer

SS1914

Ceramic

A

0 50 100 150 2000

50

100

150

200

250

300

Position [µ m]

Tem

pera

ture

[°C

]

SS1914

Ceramic

B

Figure 37: (A) Energy transfer from a hot press platen into a wet papersample, calculated with the simplified model of Figure 36. The initialweb temperature was 25 ◦C and the temperature of the hot platen is setto 300 ◦C. (B) Temperature profiles inside the paper web plotted when40000 J/m2 have been transferred to the sheet.

Although the inception and extension of delamination appears to becorrelated to the total energy transferred to the sheet, no unique value of A Critical Heat of

Delamination?a universal critical energy for delamination appears to exist (Phelan et al.,1995). This is not surprising considering that the total energy transferredto the sheet is not an unique descriptor of temperature profile inside thesample. Figure 37 shows that a metal surface gives a temperature profilethat is much steeper and reaches higher temperatures than that of a ceramicmaterial transferring the same amount of energy.

Notation

cp Heat capacity W/kg ·KK Thermal mass W

√s/m2 ·K

t Time sT Temperature ◦Cx Spatial coordinate mGreek Symbolsα Thermal diffusivity m2/sρ Density kg/m3

Subscripts0 Initial state1 Hot half-space2 Cold half-paceH Hot surface

93

Marco Lucisano

Line-Source Method: Exact Solution

An exact solution of Fourier’s equation of heat conduction

∂T

∂t= α ·

(∂2T

∂r2+

1r·∂T

∂r

), (40)

with initial and boundary conditions:

T = T0, for t = 0, and for r = ∞ (t finite), (41)

− limr→0

2πkr∂T

∂r= q′ for 0 < t ≤ t0, (42)

− limr→0

2πkr∂T

∂r= 0 for t > t0, (43)

can be obtained by application of a points source method (Carslaw andJaeger, 1959). The temperature rise at any point of a homogeneous infinitesolid due to a heat quantity Qρc instantaneously generated at t = 0 at apoint (x′, y′, z′) can be described by:

∂2T

∂x2+

∂2T

∂y2+

∂2T

∂z2=

1α

∂T

∂t, (44)

which is satisfied by:

T (x, y, z, t) =Q

8 (παt)32

e−{(x−x′)2+(y−y′)2+(z−z′)2}

4αt , (45)

where Q is the strength of the source, defined as the temperature rise of theunit volume of the substance due to the heat liberated by the source.

The temperature distribution around an instantaneous line source ofstrength Q′, liberated at time t = 0, parallel to the x-axis and passingthrough the point (x′, y′, 0) can be obtained by integrating Equation 45along the line, considered as a distribution of instantaneous point sources ofstrength Qdz:

∆T (x, y, z, t) =Q′

8 (παt)32

∫ ∞

−∞e−{(x−x′)2+(y−y′)2+(z−z′)2}

4αt dz′, (46)∣∣∣=

Q′

4παte−{(x−x′)2+(y−y′)2}

4αt dz′, (47)

where the quantity of heat liberated per unit length of line is Q′ρc.

94


The temperature response of the material due to a heat pulse of finiteduration in time can be calculated by integrating Equation 47. If heat isreleased with rate ρcφ(t), per unit time per unit length of a one-dimensionalheater parallel to the x-axis and passing through the point (x′, y′, 0), thetemperature field around the heater is expressed by:

∆T (x, y, z, t) =∫ t

0φ(t′)e−

{(x−x′)2+(y−y′)2}4αt

dt′

t− t′. (48)

or:

∆T (x, y, z, t) =∫ t

0φ(t′)e−

r2

4αtdt′

t− t′. (49)

where r2 ≡ (x− x′)2 + (y − y′)2.

A square heat pulse of length t0 and strength Q is represented by:

φ(t) =

{Q′ for 0 < t ≤ t0,0 otherwise,

(50)

which gives the following temperature field around the heater:

∆T (x, y, z, t) =

1

4πα

∫ t0 Q′e−

r2

4αt′ dt′

t−t′ for 0 < t ≤ t0,

14πα

∫ t00 Q′e−

r2

4αt′ dt′

t−t′ for t > t0,

(51)

and rewriting:

∆T (x, y, z, t) =

1

4πα

∫ t0 Q′e−

r2

4αt′ dt′

t−t′ for 0 < t ≤ t0,

14πα

∫ t0 Q′e−

r2

4αt′ dt′

t−t′ −1

4πα

∫ tt0

Qe−r2

4αt′ dt′

t−t′ for t > t0.(52)

Introducing the exponential integral Ei:

Ei(x) ≡∫ x

−∞

et

tdt, (53)

Equation 52 becomes:

∆T (r, t) =

Q′

4πα

[−Ei

(−r2

4αt

)], for 0 < t ≤ t0

Q′

4πα

[Ei

(−r2

4α(t−t0)

)− Ei

(−r2

4αt

)], for t > t0

(54)

95

Marco Lucisano

Thermal Diffusivity

The thermal diffusivity of the sample can be obtained from the coordinates(tm, Tm) of the point of maximum temperature increase, which is determinedby differentiation of Equation 54:

∂T

∂t=

Q′

4πα

[− 1

t− t0e− r2

4α(t−t0) +1te−

r2

4αt

]= 0, (55)

Rearranging:

er2

4αt− r2

4α(t−t0) =t− t0

t, (56)

and solving for α:

α =r2

4

1tm−t0

− 1tm

log(

tmtm−t0

) . (57)

Notation

cp Heat capacity W/kg ·KQ Strength of a point source K ·m3

Q′ Line source strength per unit time m2/K · sr Radial coordinate mt Time sT Temperature ◦Cx Spatial coordinate my Spatial coordinate mz Spatial coordinate mGreek Symbolsα Thermal diffusivity m2/sφ Time-dependent strength of a line source m2 ·K/sρ Density kg/m3

Subscripts0 Initial statem Point of maximum temperature increase

96

on heat and paper: from hot pressing to impulse technology9246/fulltext01.pdf · on heat and paper:...

Documents