D)l
D
Anders Rosenkilde, Ove Söderström
Measurements of Moisture Transport
Coefficients in Wood During Drying
Paper presented at the 5^^ International
lUFRO Wood Drying Conference^
Quebec City^ Canada^ August 13-17^ 1997
Trätek
Anders Rosenkilde, Ove Söderström
M E A S U R E M E N T S O F M O I S T U R E T R A N S P O R T C O E F F I C I E N T S
IN W O O D D U R I N G D R Y I N G
Paper presented at the 5^^ International l U F R O Wood Drying Conference,
Quebec City, Canada, August 13-17, 1997
Trätek, Rapport I 9702007
ISSN 1102- 1071
ISRN TRÄTEK - R - - 97/007 - - S E
Nyckelord
diffusion coefficients
moisture distribution
moisture gradients
surface emission
wood drying
Stockholm februari 1997
Rapporter från Trätek — Institutet för träteknisk forskning — är kompletta sammanställningar av forskningsresultat eller översikter, utvecklingar och studier. Publicerade rapporter betecknas med I eller P och numreras tillsammans med alla ut-gåvor från Trätek i löpande följd.
Citat tillåtes om källan anges.
Reports issued by the Swedish Institute for Wood Technology Research comprise complete accounts for research results, or summaries, surveys and
studies. Published reports bear the designation I or P and are numbered in consecutive order together with all the other publications from the Institute. Extracts from the te.xt may be reproduced provided the source is acknowledged.
Trätek — Institutet för träteknisk forskning — be-tjänar de fem industrigrenarna sågverk, trämanu-faktur (snickeri-, trähus-, möbel- och övrig träför-ädlande industri), träfiberskivor, spånskivor och ply-wood. Ett avtal om forskning och utveckling mellan industrin och Nutek utgör grunden för verksamheten som utförs med egna. samverkande och externa re-surser. Trätek har forskningsenheter i Stockholm. Jönköping och Skellefteå.
The Swedish Institute for Wood Technology Re-search .serves the five branches of the industry: .sawmills, manufacturing (joinery, wooden hous-es, furniture and other woodworking plants), fibre board, particle board and plywood. A research and development agreement between the industry and the Swedish National Board for Industrial and Technical Development forms the basis for the Institute's activities. The Institute utilises its own resources as well as those of its collaborators and other outside bodies. Our research units are located in Stockholm, Jönköping and Skellefteå.
Sammanfattning
När nya torkningsmodeller skall utvecklas är det oftast problem med att få tag i bra materialdata
såsom diffusionskoefficienter. I litteraturen saknas oftast kompletta materialdata, endast enstaka
värden eller värden inom snäva fukt- och temperaturområden kan erhållas. Det bör påpekas att
torkningsmodellema aldrig är bättre än de materialdata som ligger till grund för beräkningarna.
Föreliggande arbete har koncentrerats på att utföra bra experiment med avseende på fukttransport.
Resultaten från mätningarna kan användas för att verifiera redan existerande torkningsmodeller
eller för utvärdering av nya fukttransportkoefficienter.
I föreliggande arbete har mätningar av fuktfördelningen i furu {Pinus sihestris) under torkning
utförts. Mätningarna har utförts i provkroppar där fukttransporten skett i en av de tre
huvudrikt-ningarna radiell, tangentiell och longitudinell i förhållande till fiberriktningen och årsringarna.
Fukttransportkoefficienter har ur dessa mätningar utvärderats i alla tre huvudriktningarna i både
splint och kärna. Arbetet inkluderar även en mindre diskussion om möjliga förklaringar till varför
det är en så stor skillnad mellan uppmätt och beräknat ytövergångsmotstånd invid en torkande
träyta.
Measurements of Moisture Transport Coefficients in Wood
During Drying
A. Rosenkilde Trätek, Swedish Institute for Wood Technology Research
Box 5609, S-114 86 Stockholm, Sweden
O. Söderström Royal Institute of Technology,
Dept. of Building Sciences, Building Materials,
S-100 44 Stockholm, Sweden
ABSTRACT
When making new drying models there is always a problem with getting good moisture transport properties such as diffusion coefficients. Verj' often there are problems with finding accurate values in the literature, and to make new measurements can be both difficult and expensive. The drying models are never better than the physical property values that are used. This work has concentrated on performing good experiments that can be evaluated with different methods. The measurements can be used for verifying existing drying models or for the evaluation of moisture transport coefficients.
In this present work moisture gradients measurements have been performed in Scots pine (Pinus silvesiris). These measurements were done with the moisture flow in three different directions to grain, radial, tangential and longitudinal. Further, the moisture transport coefficients were evaluated in all three directions to grain. At the end we discuss some possible explanations to the difference between our measurements and calculations of tlie surface emission coefficient.
INTRODUCTION
When developing wood dr>'ing models for practical use in the saw milling industry it is of great importance to have accurate physical property values for the actual wood species. This present work has dealt with the problem to measure and evaluate moisture transport properties inside a wood sample during drying. Further, we have looked into the understanding of the boundar\ conditions during dr>ing, here called surface phenomena.
This present work continues some work that has been published earlier, Söderström and Salin (1993). Samuelsson and Arfvidsson (1994). Samuelsson and Söderström (1994) and Rosenkilde and Söderström (1996). An extended literature suney on surface emission factors were presented in the work by Söderström and Salin (1993). Further. Tong (1986) and Salin (1990) presented a literature sur\cy on moisture
transport in wood. Our interest when reading those literature surveys were focused on moisture transport properties for softwood, or more precise; Scots pine
(Pinus silvestris). In that literature we did not find any
complete set of moisture transport properties such as diffusion coefficients. Therefore, we decided to develop and evaluate some methods for determination of diffusion coefficients in a wide moisture content range. Those methods included both the measuring technique and the evaluation technique. Further, we needed more knowledge in the field of surface phenomena during drying of wood. Surface phenomena here includes surfaces emission coefficients and boundary conditions. To be able to study the surface phenomena, we built a experimental kiln, specially designed for detailed studies of a wooden surface during drying.
The aim with the present work was to develop and evaluate a method for determination of moislurc transport properties such as diffusion coefficients This
method was aimed to be used for the determination of dilTiision coefficients for Scots pine in both sap\\ood and heartwood, and in all three directions to grain separately. Further, the aim was to study the surface phenomena during dr>ing and to evaluate a new method to determine the surface emission factor.
METHODS FOR D E T E R M I N A T I O N DIFFUSION COEFFICIENTS
OF To be able to determine the moisture transport properties, measurements of moisture content distribution within specimens of Scots pine (Pinus
sihestris) were perfonned during drying. The
measurements were perfonned in sapwood and heartwood and in all three directions to grain separately. The specimen had the following dimensions: 29 x 29 .\ 29 mm. During the dr>'ing phase the climate was constant with a dry-bulb temperature of 60 °C and a wet-bulb temperature of 50 °C and an air velocity of 3±1 mis. The experiments are described in detail by Rosenkilde and Arfvidsson (1996).
Measuring methods
The methods of measurement that have been used in the present work were both destructive and non-destructive. The destructive method used a slicing technique where the specimens were cut with a knife into ten slices, each 3 mm thick. This was done at certain time intervals during the dr\'ing period. The moisture content for each slice was determined with the dry weight method. The slicing technique is described by Samuelsson and Arfvidsson (1994) and Rosenkilde and Arfvidsson (1996). An advantage with the slicing technique is that it is easy to perform and it requires no sophisticated equipments. A well performed measurement gives good accuracy. The results showed an accurac) between ± 0.3% and ± 1.4% moisture content, depending on the amount of specimens in each measurement and the resolution in the used balance. A disadvantage with the slicing technique is that the different measurements during the drying period are performed on different specimens since they are destroyed after each measurement. This presumes that all specimens have almost identical moisture transport properties, which could be ver> hard to achieve even i f all specimens origin from a narrow /one in the same log. The slicing technique could only be used when the specimens were dried in tangential or radial direction to grain where they were cut along the grain. It is almost impossible to cut the specimen across the grain with a knife. Further details can be found in
Samuelsson and Arfv idsson (1994) and Rosenkilde and Arfvidsson (1996).
The non-destructive method used computer tomography scanning technology, CT-scanning. With the CT-scanner the specimens were X-rayed at certain time intervals. The CT-scanner is described in detail in Herman (1980). Lindgren (1988). The experimental equipment is described by Rosenkilde and Arfvidsson (1996) and Wiberg (1996). With the CT-scanner the time intervals between the measurements can be ver>' short. This is an advantage in the first part of the drying period where the moisture transport flow is greatest. The resulting images from the CT-scanner had a ver\' fine resolution with 512 x 512 pi.xels. Each pixcel represents a measured value for area of 0,24 \ 0,24 mm. This fine resolution provides a more detailed analysis of the measurements. The greatest advantage with the CT-scanner was that the same specimen could be used for all measurements during a whole drying period. During a drying period the number of measurements could be 150. With the slicing technique this is nearly impossible to perfonn. With a good choice of resolution when analysing tlie CT-images the accuracy in moisture content can be very good. In the present measurements the accuracy were ± 0,4% moisture content below fibre saturation and ± 0,8% moisture content above fibre saturation. This accuracy was achieved in the analy sis w hen the measured area in the images were divided into ten smaller areas. Each of those areas consisted of 10 x 80 = 800 pi.xels which is equal to 2,4 x 19 = 45.6 mm". In the results a mean value for all 800 pixels were presented for each of these ten areas. Higher resolution in the evaluation provides a more detailed analysis but with less accuracy. Therefore, there are a balance between accuracy and resolution in the analysed result, A disadvantage wiili CT-scanning is the accuracy problems that arise when the measurements in tangential or radial direction should be analysed. The analysis uses a reference image with known moisture content. This reference image is subtracted from the other images that are taken at certain time intervals during the drying period in the same specimen. The resulting images show the density distribution at these time intervals caused by the distribution in moisture content. Since there is a great variation in density within a annual ring which is much greater than the density variation caused by the distribution in moisture content, it is \cry important that every point in the reference image is subtracted from the same point in the other images This is not easy to perform due to the fact that the images are not of the same size due to difTerence in shrinkage at difTcrcnl moisture content levels. In longitudinal
direction to grain, the shrinkage is very small and there is no variation in density caused by the annual rings. This makes this analysis easier lo perform and much more accurate.
Measured results
A complete presentation of all measured results are found in Rosenkilde and Arf\'idsson (1996) and in Rosenkilde (1996). We will here show some of the most interesting results. A l l measurements of moisture content distribution in Scots pine heartwood showed a profile in moisture content that decreased towards the surface. This can be observed in Figure 1.
Further, if the gradient dlijdx is equal to zero, the moisture content profiles are flat and horizontal. Figure 2 shows moisture content profiles for Scots pine sapwood dried in the longitudinal direction to grain. Above a moisture content of appro.ximately 30% the profiles are nearly flat and horizontal. According to equation (1) the moisture flux g should be nearly zero, but quite opposite it is ver>' high. An explanation to this is partly given by Spoiek and Plumb (1981). According to them, there is a certain saturation called the irreducible saturation where the liquid phase continuity is disrupted and the moisture transport caused by capillary pressure no longer exists. Instead, the moisture transport below this saturation is mainly a diffusion process which is much slower.
28,5 h 39,0 h 90,5 h 0 2 4 6 8 10 12
Distance from centre |mm|
- X - 6 . 0 h
I0,0h 0 2 4 6 8 10 12
Distance from centre [mm] FIGURE 1. Measured moisture content distribution in
Scots pine heartwood dried in the tangential direction to grain.
These results are expected when considering Fick s law which can be written as:
(1) where g is the moisture flux (kg/nrs). u is the moisture content (kg/kg), x is distance (m) and p is the wood density (kg/m^). According to equation ( I ) the moisture flux g increases with a increasing gradient in moisture content, dufdx. A increased gradient gives a increased inclination in the moisture content profiles.
FIGURE 2. Measured moisture content distribution in Scots pine sapwood dried in the longitudinal direction to grain.
Both Cloutier et al (1992) and Wiberg (1995) perfonned measurements of moisture content distribution during drying. They also reported results that are similar to the present results. Wiberg (1995) also concludes that free water in sapwood at high moisture contents migrates due to capillary forces and the drv'ing rate is controlled by bound water diffusion in a thin layer of wood near the surface. I f this is the case then there nmst be a great gradient in moisture content near the surface. Due to an edge effect in the measurements with the CT-scanncr it is not possible to measure the moisture content closer than approximately 2 mm from the surface. Further, the slicing technique
used here gave mean values for a 3 mm thick zone close to the wooden surface. This concludes that we need to improve the measuring methods i f a better resolution near the surface should be achieved.
Evaluation method
The evaluation method is developed by Claesson and Arfvidsson (1996) and it is also described by Rosenkilde and Arfvidsson (1996). This method uses the difference between moisture content profiles that are measured at different times during drying for the evaluation of Kirchhoffs potential y/ (kg/ms) and its dependence of the moisture content w (kg/m^). This potential describes the moisture transport properties for the measured material. The connection between the diffusion coefficient (mVs) and the Kirchhoff potential y/ is given by equation (2).
dw (2)
A detailed description of the evaluation method is found in Claesson and Arfvidsson (1996) and all results are found in Rosenkilde and Arfvidsson (1996). In Table 1 the evaluated diffusion coefficients for the measured moisture content profiles in Figure 1 and 2 are presented.
Table 1. The diffusion coefficient at 60 °C for Scots pine sapwood and heartwood
Sapwood longitudinal MC D-IO'" Heartwood tangential. MC D I O ' % mVs % m-/s 8.3 30 10.8 6.2 10.1 35 12.8 6.3 12.0 35 14.8 7.5 13.8 35 16.8 8.7 15.7 35 18.7 8.8 17.5 35 20.7 10 19.4 40 22.7 8.7 21.2 35 24.7 8.8 23.1 40 26.6 13 24.9 40 28.6 15 26.8 60 30.1 25 28.6 100 30.0 120
As expected the diffusion coefficient is much higher in longitudinal direction in sapwood. Note the increase in diffusion coefficient at moisture contents near 30%. This is due to the fact that the moisture transport process at this moisture content level changes from a diffusion process to a capillary pressure driven process. Spolek and Plumb (1981) called this moisture content level the irreducible saturation as mentioned above. SURFACE PHENOMENA DURING DRYING
The interaction of the surrounding air and a drying surface of a hygroscopic material has been the subject for a continuing discussion as in Babiak (1995) and Söderström (1996). Therefore, the interest for adequate measurements has increased.
By performing sorption experiments under conditions similar to those in a industrial kiln we evaluated the surface emission coefficient. S, for medium densit> fibre board (MDF). The experiments were performed in a specially designed laboratory kiln. A l l details about the experiments and the kiln are described in Samuelsson and Söderström (1994) and Rosenkilde and Söderström (1996). A detailed description of the evaluation method is given in Söderström and Sal in (1993). The results showed that the measured S was much lower (1/14 to 1/3) than the value calculated with boundary layer theory using Lewis' relation. The calculations are described in Rosenkilde and Söderström (1996). In that work we concluded that the
measured surface resistance. I/S. consisted of two parts: external surface resistance, described by boundary layer theory as mentioned above, and a greater internal resistance in the surface layers. A possible explanation to that internal resistance is given by Salin (1994, 1996). He suggested that there is a transfer resistance between the vapour in the void space and the bound water in the cell wall. This internal transfer resistance gives a non-equilibrium state inside the wood. Salin (1996) concluded that the moisture transport calculations in practice can be performed by using a smaller apparent mass transfer coefficient for external surfaces and that the apparent coefficient represents the value that is found experimentally i f normal measurement procedures are followed.
CONCLUSIONS
The diffusion coefficient has been determined for Scots pine sapwood and heartwood dried at 60 °C. The analysis showed that the used methods for measurement and evaluation ga\e good and accurate results in a moisture content range up to about 30%. The results shows very good correspondence with results reported by Morén (1987) and Malmquist (1990,
1991). For moisture content levels above 30% the method of evaluation needs improvement. Future work will focus on measurements of moisture content distribution above the fibre saturation point with a better resolution near the wooden surface.
The surface emission coefficient has been determined by performing sorption experiments with medium density fibre board. The experiments showed that measured surface emission coefficient is much lower than the value calculated with boundar>' layer theor>' using Lewis' relation. Salin (1994, 1996) suggested that the apparent lower surface emission coefficient is caused by a transfer resistance between the vapour in the void space and the bound water in the cell wall. REFERENCES
Babiak, M . 1995: Is Fick s law valid for the adsorption of water by wood ? Wood Sci Technol. 29:227-229.
Claesson, J.; Arfvidsson. J. 1996: A new method to evaluate flow coefficients from moisture distributions. 5"' International lUFRO Wood Drying Conference. Quebec City. Canada. August
13-17.
Cloutier, A.; Fortin. Y.; Dhatt, G. 1992: A wood drying finite element model based on the water potential concept. Drying Technology 10(5): 1151-1181.
Herman, G.T. 1980: Image reconstruction from projections - the ftindamentals of computerized tomography. New York, Academic Press. Lindgren, O. 1992: Medical CT-scanners for
non-destructive wood density and moisture content measurements. PliD Thesis. Luleå University of Technology, Sweden.
Malmquist, L. 1990: En förbättrad diffusionsmodell for kammartorkning, (In Swedish). Swedish Institute for Wood Technology Research, Report 1 9008043.
Malmquist, L. 1991: Lumber drying as a diffusion process. Holz als Roh- und Werkstoff 49:
161-167.
Morcn, T. 1987: Vidaretorkning och konditionering av furuvirke. (In Swedish). Forskningsrapport TULEA 1987:39 T. Luleå University of Technology, Sweden.
Morcn. T.; Salin, J-G.. Söderström, O. 1992: Determination of the surface emission factors in wood sorption experiments. 3"^ lUFRO
International Wood Dr\'ing Conference, Vienna, August 18-21. 1992:69-73.
Rosenkilde, A. 1996: Measurements of moisture content gradients and surface phenomena during drying of wood. Licentiate Thesis. Royal Institute of Technology, Sweden
Rosenkilde, A.; Arfvidsson. J. 1996: Measurement and evaluation of moisture transport coefficients during drying of wood. Submitted for publication in Holzforschung.
Rosenkilde, A.; Söderström, O. 1996: Surface phenomena during drying of MDF. Accepted for publication in Holzforschung.
Salin. J-G. 1990: Simulation of the timber dr\ ing process. Prediction of moisture and quality changes. PhD thesis. Åbo akademi University, Finland.
Salin, J-G. 1994: Mass transfer from wooden surfaces and internal non-equilibrium. 9'*' International Drying Symposium, Gold Coast. Australia, August
Salin, J-G. 1996: Mass transfer from wooden surfaces. 10*^ International Dr\'ing Symposium, Krakow, Poland, July 30-August 2.
Samuelsson, A.; Arfvidsson, J. 1994: Measurement and calculation of moisture content distribution during drying. 4* lUFRO International Wood Drying Conference, Rotorua, New Zealand, August 9-13, 79-86.
Samuelsson, A.; Söderström, O. 1994: Measurements of surface emission factors in wood drying. 4* lUFRO International Wood Drying Conference, Rotorua, New Zealand, August 9-13, 107-113.
Spoiek. G. A.; Plumb, O A. 1981: Capillary pressure in softwoods. Wood Sci. Technol. 15: 189-199 Söderström, O.; Salin, J-G. 1993: On determination of
surface emission factors in wood drying. Holzforschung 47:391-397.
Söderström, O. 1996: Surface conditions and Fick s law. Wood Sci Technol. 30: 149-151.
Tong, L. 1986: Moisture transport in wood and wood-based panels - A literature survey. Swedish Institute for Wood Technology Research, Report P 8609056.
Wiberg, P. 1995: Moisture distribution changes during drying. Holz als Roh und Werkstoff 53:402. Wiberg, P. 1996: CT-scanning during drying. Moisture
distribution in Pinus silvesths. 5^ International lUFRO Wood Drying Conference, Quebec City, C^ada, August 13-17.
Detta digitala dokument skapades med anslag från
Stiftelsen Nils och Dorthi
Troédssons forskningsfond
Trätek
I N S T I T U T E T FÖR T R Ä T E K N I S K F O R S K N I N
