Drying process for white Beech Allegretti Ottaviano, Cividini Rodolfo, Travan Livio Abstract According to periodical trend, the market of semi-finished wood products in Europe repeatedly demands dry beech with a typical natural pale colour turning to grey- white for different end uses going from furniture to interior design. One of the problem is that nobody really knows what white beech means, and the absence of standards and methods for grading creates confusion and misunderstanding between buyers and sellers. The general opinion is to considered white beech, the sawn timber from 5 to 100 mm thick with a light and homogeneous colour within all the thickness from the surface to the core. Those characteristics are not easy to reach, above all for large thickness timbers and during the summer. Many factors influence the colour of the dried wood but the drying process, as well as the operations before the drying such as the log storage and the sawn process are fundamental. The paper condensates the experiences and knowledge of the authors and it reviews the main rules to apply during drying to attempt to reach good results. Trend of colour in European wood industry in the last century Main requirement for all wood product manufacturers is the uniformity and precise definition of characteristics of the material. The maximum value of the wood material is probably reached in the production of furniture and interior decoration. In these products, superficial colour is, of course, the principal factor which undergoes the vagaries of fashion. Compared to species with more definite colours, such as mahogany, walnut and oak, beech wood, due to its heterogeneous characteristics is not an ideal material for industrial production. Before the First World War, in Europe darker colours were appreciated and beech, steamed to dark reddish-brown, was used in substituting walnut, especially in curved furniture elements (Tonet). After the Second World War, starting from the 80 s and 90 s, market started to demand light colour furniture; natural beech, without false heartwood (red heart) was now much appreciated. The white colour was usually obtained following traditional air drying, down to 0% MC, sometimes under cover in heated chambers to 10% MC. Only rarely was beech dried in kilns. Industrial production now requires that wood is dried directly in kilns decreasing drying time but increasing dramatically the risk of undesired discolourations. In the kilns, it is difficult to preserve the light colour of wet beech. At the same time, white beech is not clearly defined, ranging from white to pale reddish, with subjective and indistinct interpretations by the operators. Characteristics of beech wood Beech heartwood is undifferentiated from sapwood, has a uniform white-yellowish colour, often containing false heartwood. It has diffuse porous with a fine texture, rings are distinguishable and late wood, in thicker rings, takes up most of the space giving it a reddish-brownish hue with evident veining. Moisture content of wet from saw wood is high (85%-95%), fibre saturation point is approximately 30-3% at the temperature of 0-5 C, basic density is 560 kg/m 3 but very variable. Total shrinkage in radial direction ranges from % to 9% and in tangential direction from 9% to 0%, volumetric shrinkage ranges from 11% to 9%; beech wood has therefore a high deformability and due to the presence of numerous and conspicuous parenchymatic rays, is prone to checking. Beech wood often has growth stresses, which together with the presence of conspicuous rays, cause checking and splitting in wood, even before kiln drying. Normal wood has a high permeability to water and water vapour while false heartwood is impermeable, because of the occurrence of tylosis. Immediately after felling, the colour of wood starts turning to reddish, the faster the higher the ambient temperature and relative humidity of the air. It is perishable, in
warm and humid conditions, especially if felled at the beginning of the growing season when sugar and starch are present in wood in high concentration. When wet, colour changes rapidly; it slows down approaching hygroscopic MC and becomes stable when dry < 0%. The problem of white beech The darkening of the beech material at high temperature and humidity seem to be related to the enzymatic oxidization of accessory compounds, like high condensed phenolic extractives formed during drying (Koch, Bauch, 000). Many experiments have been made to assess the climatic parameters influencing the discolouration of light coloured wood species and the ways to prevent it, not only in beech (Aleon 00, Trübswetter 1995) but also ash (Straze A., Oven P., Zupancic M., Gorisek Z., 003), maple (Yeo H., Smith W.B. 003) and birch (Kärki T., Möttönen V. 004). The results of these tests seem to coincide in giving the major role to the combination of temperature and moisture content (MC) which triggers the enzymatic oxidization. Furthermore, time seems to be another important factor because it a long period of warm and humid conditions increases the oxidative action. These phenomena are not a problem in the normal drying and steaming of beech where the temperature ranges between 55 75 C and the colour of the dried wood is a homogeneous red-yellow. In fact, steaming of beech is now a common practice also to allow the commercialisation of beech material containing false heartwood to reduce the conspicuous colour differences; in the past, beech containing false heartwood was used as fuelwood only. Problems arise when the target colour of kiln-dried lumber is the so called white beech. In this case, the most evident and unwanted discolouration is a darkening of the inner core while the surface portions remain light in colour, producing what is called the sandwich effect. This phenomenon can also occur in wood, air dried in summer, in humid conditions with no wind. In general, because of the dynamic of transport of water in wood during drying this problem is proportional to the thickness of the sawn timber. Normally, thin boards dried at low temperatures will not discolour. Also lateral boards which dry faster will discolour less than radial boards. On the contrary, in large thicknesses the core stays humid for a period longer than the surface, causing a darkening of the core. On the other hand, thick beech boards are prone to casehardening due to a too high drying rate during the II drying stage. Casehardening produces several problems, one of these is an inhomogeneous colour with a thin light dry shell and a dark and humid core. Such correspondence of effect due to opposite causes makes the experimental tuning of the drying process an hard task, above all during summer and for thicknesses higher than 40-50 mm. In fact, in case of beech material thicker than 40-50 mm the tendency is to pre-cut it into elements before drying, reducing drying time and therefore the risk of discolourations. Something similar to the casehardening occurs in winter, especially when wood is frozen. It consists in a very thin white shell, with the rest of the core still moist. Figure 1- moderated sandwich effect in 50 mm thick sawn timber
There are infinite gradation of results between the perfect homogeneous white beech and the worst sandwich effect; the absence of standards and methods for grading the quality of colour creates confusion and mis-understanding between buyers and sellers. Not only, technological and aesthetic requirements, in the manufacturing of end-products demand different colour hues that are often all called white beech. Difficulties arise, for example, when having to match different lots of dried material. Because the technical difficulty to manage the production of white beech discolorations are a major cause of losses at sawmills processing this species (Koch, 004). Some colorimetric observations In the present work, a proposal for the characterization of the term discolouration, through spectrophotometric colour analysis is discussed. This technique overcomes the problem of the subjectivity of the perception of the colour and is it widely used at scientific level for the colour characterisation of wood, even if with some well documented problems, mainly due to the local variability of the wood. Colour measurement was performed on beech specimens dried under different conditions. In particular we examined the colour of : Specimen A, 30 mm, T drying 30 C (initial temperature) white beech schedule; Specimen B, 30 mm, T drying 50 C (initial temperature) normal beech schedule Specimen C, 50 mm, air dried; Specimen D, 85 mm, T drying 6 7 C (initial temperature) drying schedule for white beech (sandwich effect); Specimen E, 67 mm, T drying 6 7 C (initial temperature) drying schedule for white beech (sandwich effect); Specimen F, 55 mm, T drying 50 C (initial temperature) drying schedule for normal colour; Specimen 7, 55 mm, T drying 50 C (initial temperature) drying schedule for normal colour + steaming treatment. The colour was measured by a portable spectrophotometer, providing a d/8 (hemispheric light, 8 ) specular component included measurement, being the area measured window 5 mm large. The results are expressed in terms of colour coordinates L*, a*, b* of colour space CIELAB with light D65 (daylight UV included) for a standard eye at 10 where: L* is the brightness ranging from 100 (black) to +100 (white); a* ranges from 60 (blue) to + 60 (yellow) and in terms of absolute values, it indicates the colour saturation (0 = gray); b* ranges from 60 (green) to + 60 (red). C = a + b ranges from 0 to 84,8 and it gives the value of the saturation of colour. Our attention was mainly focused on the colour differences, trying to understand if it is possible to graduate the intensity of colour difference in the discolouration called sandwich effect. The colour difference can be evaluated in terms of L* c* and b* and C. The global colour difference between colour 1 and is given by E = ( L b L1 ) + ( a a1) + ( b 1) The magnitude of E can be classified according in table below coming from a non- standard rule for grading the colour performances of printer an PC monitor: 0,< E Not visible difference 0,< E< Small difference < E<3 Colour difference visible with high quality screen 3< E<6 Colour difference visible with medium quality screen 6< E<1 High colour difference E>1 Different colours
Table 1 The colour is influenced by many factors such as the anatomical direction of the surface and the ring density. Because of such variability and the limited number of measurements performed, the results must be considered only as a base for a discussion and for possible future developments. The main results show that: 1. The average coordinates of white beech are L= 7,6; a= 7; b= 19,1, C= 0,4. The variability is quite low but the data are not sufficient to assess differences among different white gradations.. The coordinates for normal beech show a high variability, corresponding to the high variability of possible combination of drying temperature and post-processing treatments and conditions. We measured the range of value of: L= 56 68; a= 9-13; b= 1-7, C=6-30. 3. Compared to white beech, the colour of the normal beech is darker and more saturated. The average E between white and normal beech = 1 (high colour difference). Average C=5. 4. The difference of colour in the sandwich effect mainly concerns L* which become darker going from the surface (av. L= 75,5) to the core (av. L =69,8) while the saturation shows a slight increment (av. C =1,6). The global colour variation E measured is in the range 5,7 6,5 (sensible colour difference). Drying techniques to obtain white beech The main rules for the drying process in conventional kilns are: very short time between sawing operations and start of drying; the drying time for the I and II drying stage should be as short as possible; temperature of I and II drying stage must not exceed 30-35 C. In order to combine the right elements, a low EMC of air is needed in the kiln together with a very accurate control of the drying parameters because: too high temperature produces a reddish discolouration; too low temperature produces moulds and decaying; diminishing EMC of air improves colour but causes drying stresses (casehardening). This requires special kilns with oversized vents, better if equipped with fans, very efficient air circulation system and possibly, more than 6 MC probes. Such probes as well the probes for the measurement of temperature and humidity of the circulating air must be always efficient and calibrated. If EMC wafers are used to assess the air humidity, the hysteresis (up to 5% during sorption and desorption) has to be taken into account. Moreover, the control of the material before and during the drying process is very important and the use of wood samples is highly recommended. However, short drying times and low temperatures are not always easy to combine in a conventional kiln especially during summer when it is often impossible. Furthermore Pre-dried material is usually already affected by discolouration (white shell or sandwich) which could be improved only by steaming, reaching a uniform reddish colour. The colour of dried beech wood (<5% MC) could not be evidently altered. A series of different drying schedules are presented in the appendix to illustrate the effort to obtain different colour hues or overcoming technological limitations due to unfavourable climatic conditions. They are the results of a long practical experience of the authors in this field coming from the observation of hundreds of drying cycles performed on wet-from-saw, pre-dried (45-55% MC) or dried (<5%) material in industrial kilns in different countries of Europe. Some accessory operations have to be considered: in winter, for example, if frozen material is present, this must undergo a de-icing phase to avoid the further occurrence of casehardening and discolouration. Of course, equalization and conditioning phases have to be performed too (the latter not only at the end of drying), to relax the stresses generated inside wood, in the kiln. All this can be considered the state of the art of the topic and it must be applied taking into account all the rules for a correct management of the process.
However, it is very important to point out that, despite all the good practices and correct management, these recommendations are not sufficient to control the colour perfectly, so that significative differences of colour can occur among different pieces of the same lot or among different lots dried in the same conditions. Conclusion Despite many efforts at scientific and industrial level, the white colour for beech dried in conventional kilns is still a problem far from having found a definite solution. The technical problems are in the heterogeneity of the material and of the variables involved in the colour change phenomena as well as in the physics of the process itself. The latter concerns the difficulty to set given air characteristics in the kiln under some circumstance (i.e. low temperature and EMC during summer) but also the mechanism of internal movement of water within wood which generates MC gradients and consequents quality loss due to internal stresses, casehardening and colour difference between core and surface (sandwich effect). Because of this, the conduction of the drying process is a complicate equilibrium between fast drying rate to avoid colour change and slow drying rate to avoid mechanical degradation. The problem is proportional to the thickness of the sawn boards so that is practically impossible to obtain white beech when it is thicker than 40-50 mm. In this case the best solution is to load the kiln with semifinished pieces (typically 5 x 5 cm pieces for furniture) in order to avoid, mitigate or to conceal the sandwich effect. Beside the technical issues, there are often some commercial and even legal problems due to the subjectivity of the perception of the colour and the total absence of rules or grading standards. They would be useful to define quality criteria and to improve the conduction of the process. Colorimetric analysis at the industrial level could be developed for the control of the process and for the in-line after process selection of wood pieces. This paper suggests some simple colorimetric rules that could be used in practice to define what can be considered white beech and the related discolourations, first of all the so called sandwich effect. In fact a significant relation between this form of discolouration and the colorimetric difference E (higher than 5-6 when sandwich is present.) was observed. Many more samples must be still analysed to validate statistically these rules but the method seems promising in its practical applicability, suggesting for example the definition of some grading classes for colour difference and a standard method for the measure. Literature Cited Aleon D. 00, Beech Drying with Light Colour, without Stain and with Low Residual Stresses, Proceedings of COST Action E15 4 th Workshop, Santiago de Compostela, Spain Allegretti O, 004, Non Symmetrical drying test experimental results for free and constrained samples of beech wood, 5th European COST E15 workshop,athen. Cividini R. 000, Conventional Kiln-Drying of Lumber, Compendium, NARDI Srl, Italy Cividini R., Valenti L., Allegretti O., 003, Investigation on MC gradients in vacuum-press drying process of beech elements, proceedings of 8 th International IUFRO Wood Drying Conference, Brasov, Romania. Cividini R. 007, Faggio: segati. Guida all essiccazione artificiale, NARDI Srl, Italy, in print Kärki T., Möttönen V. 004, Preventing Discolouration of Birch (Betula pendula, B. pubescens) in HT Drying, 5th European COST E15 workshop,athen. Koch G. 004, Discoloration of Wood during Storage and Drying, 5th European COST E15 workshop,athen. Koch G., Bauch J. 000, Discolouration in European Beechwood (Fagus sylvatica L.) during storage and drying. Proceedings of COST Action E15 nd Workshop on Quality Drying of Hardwood, Sopron, Hungary Koch G., Welling G. 006, Verfärbungen nicht immer ist die Trocknung schuld (Colour changes not always the fault of drying) EDG Drying Seminar, Hamburg
Negri M., Tessadri B., Gaeti N., Valenti.L., 004, physical properties in wood during press-vacuum drying: colour changes and other phenomena 5th European COST E15 workshop,athen. Straze A., Oven P., Zupancic M., Gorisek Z., 003, Colour Changes of Ash-Wood (Fraxinus excelsior L.) during Conventional Drying, in Proceedings of 8 th IUFRO International Wood Drying Conference Trübswetter T. 1995, Die Trocknung heller Laubhölzer und Ihre Problematik, Holz-Zentralblatt, Stuttgart, Nr 133 Yeo H., Smith W.B. 003, Effect of Temperature and Moisture Content on the Discoloration of Hard Maple Lumber in Proceedings of 8 th IUFRO International Wood Drying Conference
Appendix: recommended Drying schedules (MC measured at ½ thickness) 1. - Normal schedule Wet 40 35 30 5 0 17 15 55 16,5 5 17,5 50 18 40 40 35 30 5 0 18 15 55 13,6 4 5 15 4 50 16,7 3 35 35 35 30 5 0 18 15 55 11,4 6,5 5 1,5 5 50 13,9 4 30 30 30 30 5 0 18 15 57 10 8 54 11 7 5 1, 5 5 5 5 5 5 0 18 15 60 8,8 10 58 9,8 8 55 10,8 7 0 18 15 64 7,4 13 6 8,6 10 58 9,5 9 0 0 0 0 0 0 0 18 15 67 6,4 15 65 7 14 6 7,7 1 18 18 18 18 18 18 18 18 15 70 5 0 68 5,5 17 65 6 16,5 15 15 15 15 15 15 15 15 15 75 4 4 7 4,3 19 70 4,7 1 1 1 1 1 1 1 1 1 1 75 4 4 7 4,3 19 70 4,7 1. - Schedule to obtain a slightly reddish hue in warm summer climate Wet 40 35 30 5 3 0 17 15 45 14 3 43 15 3 40 16,5 40 40 35 30 5 0 18 15 45 1 4,5 43 13 4 40 14 3,5 35 35 35 30 5 0 18 15 45 11,5 6 43 1,5 5 40 13,6 4 30 30 30 30 5 0 18 15 45 9,6 8 43 10,4 6 40 11,4 5,5 5 5 5 5 5 0 18 15 47 8,4 7,5 45 9, 6,5 4 10 7 0 18 15 50 7,6 10 47 8,3 10 45 9,1 9 0 0 0 0 0 0 0 18 15 55 6,8 13 51 7,3 13 49 8 11 18 18 18 18 18 18 18 18 15 60 5,6 17 56 6 17 53 6,5 135,5 15 15 15 15 15 15 15 15 15 65 4,4 1,5 60 4,7 0 57 5,1 19 1 1 1 1 1 1 1 1 1 65 3,6 5 6 3,8 5 60 4, 3. - Schedule to preserve the white-yellowish colour in warm summer climate Wet 40 35 30 5 3 0 17 15 37 13 4,5 37 13,5 4 37 14 3,5 40 40 35 30 5 0 18 15 37 10 7 37 10,7 6 37 11,5 5,5 35 35 35 30 5 0 18 15 37 8,3 9,5 37 8,9 9 37 9,6 7,5 30 30 30 30 5 0 18 15 40 7,3 8 39 7,9 11 39 8,5 9 5 5 5 5 5 0 18 15 43 6,7 1,5 4 7,1 1 41 7,6 10,5 0 18 15 46 5,9 13,5 45 6,3 13,5 43 6,8 1,5 0 0 0 0 0 0 0 18 15 50 4,8 19 50 5, 17 46 5,6 13,5 18 18 18 18 18 18 18 18 15 55 3,8 3 55 4 5 0 4,4 0 15 15 15 15 15 15 15 15 15 57 3,1 5 57 3,3 5 57 3,6 4 1 1 1 1 1 1 1 1 1 60 3 7 60 3 7 60 3 7 4. -Winter schedule to preserve the natural colour in cold and fresh seasons Wet 40 35 30 5 3 0 17 15 37 13 4,5 37 13,5 4 37 14 3,5 40 35 35 30 5 0 18 15 30 10 7 9 10,7 6 8 11,5 5, 35 30 35 30 5 0 18 15 30 8 8,5 9 8.5 7,5 8 9,5 6,5 30 30 30 30 5 0 18 15 3 6,3 11,5 30 6,7 11 9 7,1 10,5 5 5 5 5 5 0 18 15 35 5,7 13 3 6,1 1 31 6,5 11 0 18 15 40 5,1 15,5 36 5,5 14 35 5,8 13 0 0 0 0 0 0 0 18 15 45 4,3 17 4 4,5 17 40 4,8 17 18 18 18 18 18 18 18 18 15 50 3 4 48 3,6 1 45 4 19 15 15 15 15 15 15 15 15 15 55,9 6 5 3 4 48 3, 1 1 1 1 1 1 1 1 1 55 3 5 55 3 5 50 3 5 Note: at 5% MC condition if necessary for a time (hours) = to thickness (cm) at: 45 10 7,5 4 10,7 6 40 11,5 5,5