Wood Science and Technology

, Volume 47, Issue 3, pp 499–513

Distribution of preservatives in thermally modified Scots pine and Norway spruce sapwood


    • Department of Engineering Sciences and Mathematics, Division of Wood Science and Engineering, Wood PhysicsLuleå University of Technology
  • Lars Hansson
    • Department of Engineering Sciences and Mathematics, Division of Wood Science and Engineering, Wood PhysicsLuleå University of Technology
  • Tom Morén
    • Department of Engineering Sciences and Mathematics, Division of Wood Science and Engineering, Wood PhysicsLuleå University of Technology

DOI: 10.1007/s00226-012-0509-4

Cite this article as:
Ahmed, S.A., Hansson, L. & Morén, T. Wood Sci Technol (2013) 47: 499. doi:10.1007/s00226-012-0509-4


Studying the impregnation and distribution of oil-based preservative in dried wood is complicated as wood is a nonhomogeneous, hygroscopic and porous material, and especially of anisotropic nature. However, this study is important since it has influence on the durability of wood. To enhance the durability of thermally modified wood, a new method for preservative impregnation is introduced, avoiding the need for external pressure or vacuum. This article presents a study on preservative distribution in thermally treated Scots pine (Pinus sylvestris L.) and Norway spruce (Picea abies (L.) Karst.) sapwood using computed tomography scanning, light microscopy, and scanning electron microscopy. Secondary treatment of thermally modified wood was performed on a laboratory scale by impregnation with two types of preservatives, viz. Elit Träskydd (Beckers) and pine tar (tar), to evaluate their distribution in the wood cells. Preservative solutions were impregnated in the wood using a simple and effective method. Samples were preheated to 170 °C in a drying oven and immediately submerged in preservative solutions for simultaneous impregnation and cooling. Tar penetration was found higher than Beckers, and their distribution decreased with increasing sample length. Owing to some anatomical properties, uptake of preservatives was low in spruce. Besides, dry-induced interstitial spaces, which are proven important flow paths for seasoned wood, were not observed in this species.


Wood modification with heat is a widely investigated method to reduce equilibrium moisture content, water absorption, and to increase dimensional stability. Hydrophilic wood materials become more hydrophobic due to a large number of chemical changes including the esterification of hydroxyl groups (–OH) and a reduction in hemicelluloses and the number of accessible –OH groups within the wood (Alén et al. 2002; Bekhta and Niemz 2003). The reason for decay resistivity is the reduction in hemicelluloses content, moisture, and other wood components such as starch, fatty acids, and lipids which are essential for mold and fungal growth (Tjeerdsma et al. 1998). However, wood modification with heat is also reported to be vulnerable to different biodegrading agents, like marine borers (Westin et al. 2006), termites (Doi et al. 1999), and fungi (Jämsä and Viitaniemi 2001; Kamdem et al. 2002). Thermal modification could degrade wood components through the oxidation and/or hydrolysis process, and consequently, wood becomes easily attacked by fungi and/or insects after changing their composition (Doi et al. 1999). The use of thermally modified wood in ground contact is thus not recommended without further protection (Kamdem et al. 2002).

Thermal modification resulted in some anatomical changes like destruction of tracheid walls, ray tissues, pit deaspiration (Awoyemi and Jones 2011), and thus increase in the wood cell wall porosity (Andersson et al. 2005). Terziev and Daniel (2002) reported that some of the pit apertures are torn, and the appearance of micro (1–2 μm)- and nano-checks (10–20 nm) in the warty and S3 layers is noted, which provide direct penetration of the impregnated liquid into the S2 layer. They also suggested that thermal degradation of hemicelluloses makes the entire polymer complex more permeable to liquids. In this connection, increased wettability and water absorption of thermally modified pine sapwood are observed (Metsä-Kortelainen et al. 2006). Conventional kiln drying is also reported to have a distinct effect on the wood microstructure properties and hence an influence on liquid impregnation. In relation, Booker (1990) and Ahmed et al. (2012) studied the penetration paths of preservatives in dried wood. In addition to the important paths of penetration, they discussed the existence of secondary flow paths resulting from the collapse of thin cell wall layers of parenchymatous cells. Another report by Booker and Evans (1994) on the effect of air, kiln, and high temperature drying on radial gas permeability revealed that high temperature leads to increased gas permeability through the resin canal network, which is the predominant route for gas movement in radiata pine dried at high temperature. Furthermore, thermal modification does not limit the absorption of liquid in wood (Kamdem et al. 2002). Thus, it is postulated that thermally modified wood can be impregnated with preservative more easily. A previous study shows that thermally modified Scots pine and Norway spruce can uptake a considerable amount of oil-borne wood preservatives (Ahmed and Morén 2012).

Wood deterioration primarily results from the uptake of liquid water and moisture sorption, causing mechanical stresses and leading to colonization of wood decay fungi. The use of water repellents can be an effective tool to reduce wood deterioration (Sailer et al. 2000; Karlsson et al. 2011). These water-repellent compounds cause macro-pore blocking by depositing hydrophobic compounds in the lumens of vessel, tracheid, and ray cells (Weigenand et al. 2007). Thus, thermally modified wood in addition to water-repellent wood preservatives can provide a suitable method of protection against different biodegrading agents. However, the success of chemical wood preservation is dependent on the ability of the preservative to disperse throughout the wood cells, which is mainly related to the permeability of the different wood tissues, and the capability of the preservative to penetrate deeply into the wood structure (Rak 1976). Preservative penetration is more pivotal in the performance of treated wood than its retention in the wood tissue. Deep and uniform penetration of preservative is necessary rather than the least amount of retention. A lesser amount of retention and a shallow distribution of preservative, even at high concentration in the outer zone, do not provide adequate protection against wood-deteriorating organisms (Kollmann and Côté 1984; De Groot 1994). Successful impregnation with water-repellent wood preservative will protect thermally modified wood from attack of different decay fungi in outdoor application.

The literature discussed above draws attention to the susceptibility of thermally modified wood to different biodegrading agents. Since thermally modified wood is permeable, it can easily be impregnated with preservatives to ensure resistance against different biodegrading agents for use in outdoor applications. Thus, a very simple and effective method is used for secondary treatment of thermally modified wood (Ahmed and Morén 2012). As a part of the experiment, this paper provides information pertaining to preservative distribution in different cells of thermally modified Scots pine and Norway spruce sapwood. Observations were made mainly using computed tomography (CT) scanning, light microscopy (LM), and scanning electron microscopy (SEM).

Materials and methods

Wood species and thermal modification

Commercially treated thermally modified Scots pine (Pinus sylvestris L.) and Norway spruce (Picea abies (L.) Karst.) boards (ca. 50 × 125 mm²) were collected from Thermoplus, Arvidsjaur, Sweden. Moisture content of the boards was reduced to 18 % by kiln drying prior to thermal modification. Boards were thermally modified at 170 °C for 2.5 h. Saturated steam was used during drying and thermal modification as a protective vapor to prevent the wood from burning. More about this process introduced by the Danish company Wood Treatment Technology (WTT), can be found in Dagbro et al. (2010). Treated boards were brought to the laboratory, and sapwood was separated from the heartwood. Separated sapwood samples were used for the secondary treatment with preservatives. Additional green pine and spruce boards (ca. 50 × 125 mm²) were also collected before thermal modification. Sapwood portions were separated from the heartwood and left in the laboratory. After attaining equilibrium moisture content at room condition, they were used as source of control samples.

Preservatives used

Two different types of water-repellent preservatives were used viz. Elit Träskydd and pine tar. Elit Träskydd (Beckers, Stockholm, Sweden) is a water-miscible commercial product which contains additives such as propiconazole (0.6 %) and 3-iodo-2-propynyl butylcarbamate (IPBC) (0.3 %) in modified linseed oil as the binder and water as the solvent. In addition, supplement of 1.9 % (w/v) tert-butylhydroquinone (Sigma-Aldrich, St. Louis, USA) was added as antioxidant. On the contrary, commercially produced pine tar (Claessons Trätjära AB, Göteborg, Sweden) was mixed in boiled linseed oil (Claessons Trätjära AB). Turpentine (Claessons Trätjära AB) was added to thin the oil-based preservatives. The ratio of pine tar, linseed oil, and turpentine solution was 1:4:2. Preservative solutions were stirred properly before use. In the following, Elit Träskydd and pine tar will be referred to as Beckers and tar, respectively.

Preservatives treatment

Six end-matched sample blocks true in R/T orientation with dimensions of 20 × 20 × 150 mm³ (radial, R × tangential, T × longitudinal, L) from three separate boards were prepared (Fig. 1). Blocks free of knots, cracks, or any visible defects and with almost similar masses were selected, numbered, and impregnated with preservatives following the simple and effective method as mentioned in Ahmed and Morén (2012). In brief, wood samples were heated at 170 °C for 1 h in a drying oven and immediately submerged in the room-tempered preservative solution for simultaneous impregnation and cooling. After 2 h of soaking, additional preservative solution was gently wiped off the surface using paper towel and used for consecutive analyses. Preservative retention and distribution intensity were analyzed by computed tomography (CT) scanning technique. Three samples from each species (pine and spruce) and each of the two treatments (Beckers and tar) were used for preservatives impregnation. Thus, a total of 24 samples were used for the thermally modified and control groups.
Fig. 1

Preparation of sample blocks for secondary treatment of thermally modified wood with preservatives. Dimensions are not in scale

Computed tomography (CT)

A Siemens CT scanner (SOMATOM Emotion Duo, München, Germany) was used for measuring the density profiles in the treated wood. The 512 × 512 pixels CT images were obtained using scan settings of 130 kV, 250 mAs, and scan and delay time were 1.5 and 3.0 s, respectively. The intensity (gray scale) value of each pixel corresponds to the measured density (kg m−3) in that measuring volume. The volume or the voxel was 0.63 × 0.63 mm through the cross-section and 10 mm in the grain direction. A total of seven scans at 10-mm interval were performed in the grain direction for every sample. This covered almost half of the sample length. Scanning was performed before (oven dried) and after preservative treatment. Quantitative analysis of preservative distribution profile was calculated by subtracting the density profiles obtained from the preservative-treated wood from that of the oven-dried wood before impregnation using MatLab (The MathWorks, Inc., Natick, USA) software. Averaged data from the triplicate samples were used for each treatment profile.

Light microscopy (LM)

Sample blocks from preservative-treated wood with dimensions of ca. 5 (R) × 5 (T) × 5 (L) mm³ were prepared, and cross-, R, and T sections, ca. 15–20 μm thick, were obtained with a sliding microtome (Reichert, Vienna, Austria). The standard Sudan black procedure is proved to stain all available lipids (other than steroids) gray to blue-black (Chayen and Bitensky 1991). Thus, sections were treated with saturated Sudan black B (AppliChem GmbH, Darmstadt, Germany) in 70 % ethanol for 30 min, rinsed in 70 % ethanol to remove excess dye from the surface, washed well in water, and finally mounted in glycerin–gelatin jelly. The stained sections were examined under an optical microscope (Olympus SZH-ILLD, Tokyo, Japan) fitted with a USB uEye LE camera, UI-14xLE-C (IDS Imaging Development Systems GmbH, Obersulm, Germany). Digital images were captured and stored by means of uEye Demo, 3.70.0 software in a bitmap (BMP) file format.

Scanning electron microscopy (SEM)

Wood blocks measuring ca. 5 (R) × 5 (T) × 3 (L) mm³ were prepared from thermally modified wood. Blocks were soaked in water overnight for softening. Observed surfaces were finished on a microtome, dried, adhered to aluminum stubs with double-sided tape, and sputter-coated by gold using Desk II Cold Sputter (Denton Vacuum Inc., Moorestown, USA). At different magnifications, wood microstructures were then examined by using a JSM-5200 (JEOL, Akishima, Japan) electron microscope at an accelerated voltage of 15 kV. Six sample blocks from each of the thermally modified wood species were examined for the following quantified details: number of axial and radial resin canal per mm2, their tangential diameters delimited by epithelial cells, number of ray parenchyma per mm2 and cell wall thicknesses, and the number of bordered pits per longitudinal tracheid. For LM counting of number of bordered pits, the samples were split into matchstick-sized pieces and macerated in a solution of glacial acetic acid and hydrogen peroxide (1:1, v/v) at 60 °C for 18 h (Franklin 1945). A total of 35 measurements were carried out for each anatomical feature. Additional SEM observations were also carried out for the preservative impregnated samples.

Statistical analysis

To test whether there were any significant differences between the quantitative anatomical properties of pine and spruce, a Student’s two-tailed group t test (at α = 0.05, 0.01, 0.001) was performed using Excel 2003 program (Microsoft, Redmond, USA).

Results and discussion

Anatomical properties

It is apparent from Table 1 that significant differences of some of the anatomical features were observed between the two species. The number of axial and radial resin canals, cell wall thickness of ray parenchyma, and number of pits in the longitudinal tracheid in pine were found to be significantly different from spruce. Other differences such as the diameters of axial and radial resin canals and the number of ray parenchyma were observed to be statistically the same. The number and diameter of axial and radial resin canals of pine correspond well with the result by Ahmed et al. (2012). The thicknesses of ray parenchyma cell walls in spruce were thicker than pine, and this is in accord with Olsson et al. (2001). The average number of earlywood and latewood bordered pits was found to be significantly higher than those in spruce. In addition, the earlywood pit number was greater than in latewood which agrees with the findings reported by Jinxing (1989). Longitudinal tracheids with open bordered pits are considered the most important structure regulating the permeability of softwoods. However, the rays have also proven to be essential for the subsequent spreading of liquid. In the case of dried wood lumber, axial and radial resin canals are reported as the most important channels for liquid impregnation, as the conductivity of longitudinal tracheid decreases due to pit aspiration (Booker 1990). Number of resin canal is thus related to the performance of liquid permeability. Since the number of axial and radial resin canal was found to be higher in pine, it can be postulated that liquid impregnation and dispersion will be more efficient in this species rather than spruce.
Table 1

Mean (±standard deviation) of the selected anatomical features of thermally modified pine and spruce

Anatomical features



Level of significance

Number of axial resin canals per mm2

0.64 ± 0.11

0.27 ± 0.13


Number of radial resin canals per mm2

0.59 ± 0.22

0.40 ± 0.19


Diameter of axial resin canal (μm)

107.54 ± 15.91

106.73 ± 19.63


Diameter of radial resin canal (μm)

42.33 ± 5.70

41.75 ± 6.35


Number of ray parenchymas per mm2

23.00 ± 3.16

23.42 ± 6.92


Thickness of ray parenchyma cell wall (μm)

1.49 ± 0.29

2.67 ± 0.26


Number of pits per earlywood tracheid

90.37 ± 17.05

75.51 ± 15.24


Number of pits per latewood tracheid

39.97 ± 12.78

31.83 ± 10.48


** Significant difference at the 0.01 level (two tailed)

*** Significant difference at the 0.001 level (two tailed)

ns Nonsignificant

CT scanning

Since the wood is preheated, any air contained within the cell cavities and voids is also hot and therefore expands. Immersing the hot wood in preservative solution at room temperature gives rise to a rapid contraction of the air within the cell cavities and voids, resulting in the solution being sucked into and retained in the void spaces.

CT scan technique is frequently used for qualitative and quantitative analysis of wood density measurement (Lindgren et al. 1992; Scheepers et al. 2007; Sandberg and Salin 2010). Quantitative analysis of preservative retained was obtained from CT scan images. The retention of two different preservatives in different positions of the sample is shown in Fig. 2. But the average retention in this experiment measured by CT scanner was lower than the result measured by weight basis method reported elsewhere (Ahmed and Morén 2012). This type of variation could be due to using different sample boards and calculating the average of density from different positions of the sample. Instead, graphical representation of density profile in Fig. 3 completely visualizes the intensity of preservative distribution at different positions of wood samples. Species and preservative effects were also quite clear. It is certain that tar was more permeable than the Beckers. The divergence of Beckers and tar uptake is expected for their formulation differences. The reason for the higher permeability of tar could be explained by using turpentine as the thinner of the oil-based medium. Besides, it is reported that the polarity of the liquid could affect the permeability. For example, nonpolar liquids penetrate by bulk flow mainly through the cell lumens and pits, while polar compounds penetrate by both bulk flow and diffusion through the wood cell wall (Walters and Côté 1960; Bailey and Preston 1970). Formulated tar is mainly nonpolar, and some part of Beckers is comprised of polar liquid (see “Preservative used” section). So, theoretically, the Beckers uptake should be higher than the tar. But the results here demonstrated the opposite. It could be suggested that solutions in polar solvents are also able to move within the cell wall, sometimes resulting in the separation of the solute (like tert-butylhydroquinone) from the solvent. It thus raises an interesting uptake difference between tar and Beckers. A remarkable tar uptake by control pine sample was observed (Fig. 3). The reason is explained by SEM images in the next part.
Fig. 2

Preservative retention in different positions of pine and spruce. CPB control pine impregnated with Beckers, CSB control spruce impregnated with Beckers, CPT control pine impregnated with tar, CST control spruce impregnated with tar, TMPB thermally modified pine impregnated with Beckers, TMSB thermally modified spruce impregnated with Beckers, TMPT thermally modified pine impregnated with tar, and TMST thermally modified spruce impregnated with tar

Fig. 3

Preservative distribution in different positions of a control and b thermally modified pine and spruce. Abbreviations are the same as in Fig. 2

LM and SEM analyses

The distribution of preservative in different cells was observed by LM at two different positions viz. the end (Fig. 4) and the middle of samples (Fig. 5). Retained preservative in different wood cells was stained blue-black by the standard Sudan black B method. This dye solution was preferred for its ability to stain all available lipids other than steroids, and the distribution of preservatives was clearly observed in different cells. However, it is hard to explain whether the stained deposits in ray parenchyma, epithelial cell, or in resin canals were from the preservative liquid or from native wood resin. However, this staining procedure was proven to be a good technique to observe the oil-based preservative distribution in wood cells.
Fig. 4

Distribution of preservatives near the end grain position of Beckers impregnated pine (ac) and tar impregnated spruce (dg) stained blue-black by Sudan black B shown in their radial (a, c, d, f, and g) and tangential (b and e)–longitudinal surfaces. Arrows indicate the deposition of preservatives in cell lumen. Arrowheads show stained conductive bordered pit pairs. ARC axial resin canal. Scale bars 100 μm

Fig. 5

Distribution of preservatives in middle position of pine (a and b) and in spruce (c and d) stained blue-black by Sudan black B shown in radial–longitudinal surfaces. a Beckers impregnated pine. b Tar impregnated pine. c Beckers impregnated spruce. d Tar impregnated spruce. Scale bars 100 μm

Until, and unless, bordered pits are closed, longitudinal tracheids are considered the most important structure regulating the permeability of softwoods. Those longitudinal tracheids conducted preservative liquid through the permeable bordered pits that were open or partially open (Fig. 4c, f). In this regard, it is reported that all earlywood bordered pits are aspirated in air-dried pine and spruce sapwood (Liese and Bauch 1967). Bordered pit aspiration is considered to be caused by hydrogen bonding between the margo/torus and the adjacent cell wall (Thomas and Kringstad 1971). However, thermal modification can result in pit deaspiration (Awoyemi and Jones 2011) in which the hydrogen bonding in the aspirated pits may have been reduced. However, in dried wood, when bordered pits are aspirated, resin canals are reported the most important flow path (Booker 1990). It is also believed that axial and radial resin canals took active part in deep penetration of preservative liquid in thermally modified pine and spruce samples. Impregnated preservative through resin canal can diffuse to adjacent ray parenchyma and longitudinal tracheid (Fig. 4g). The overall preservative uptake was lessened through the length of the sample, and this trend is in agreement with Siau (1972). Thus, in middle of the sample, the deposition of preservative was found low (Fig. 5).

There were striking differences in tar uptake between the pine and spruce but that difference for Beckers was found low. Rhatigan et al. (2004) reported the Norway spruce as a lower permeable species when compared to Scots pine. In this case, anatomical structures are responsible for the lower permeability of the spruce. The lower frequency of axial and radial resin canals, thicker ray parenchyma, and a lower number of bordered pits per longitudinal tracheid (Table 1) could be responsible for the lower permeability. Moreover, thin cell wall layers of parenchyma cells in pine are prone to collapse even under kiln drying conditions, forming interstitial spaces (Bamber 1972; Ahmed et al. 2012). Initiation of thin-walled ray cell (1.49 μm) collapse was observed in control pine samples which were dried in an uncontrolled condition that is air drying (Fig. 6a). These kinds of dry-induced void spaces are considered as important liquid flow paths in dried lumber (Booker 1990). Thus, ray cell collapse in thermally modified wood is obvious for the pine samples (Fig. 6b). In thermally modified spruce, no such cell collapse was observed (Fig. 6c). It could be attributed to the presence of the thick-walled ray cell (2.67 μm). Drastic damages like the detachment of cell wall or cell wall layers due to thermal treatments were not observed in spruce. As a result, the spreading of preservative in different cells of spruce was restricted compared with pine.
Fig. 6

SEM micrographs of preservative impregnated pine and spruce shown in radial–longitudinal surfaces. Collapsed ray parenchyma in a control and b thermally modified pine. c Ray parenchyma in thermally modified spruce. d and e Control pine impregnated with tar. f Thermally modified pine impregnated with Beckers. g Thermally modified pine impregnated with tar. h Thermally modified spruce impregnated with Beckers. i Thermally modified spruce impregnated with tar. Arrow heads show preservative penetration through pit membranes. RP ray parenchyma, RT ray tracheid, HBPit half-bordered pit, ABPit aspirated bordered pit, DBPit disrupted bordered pit, EWTracheid earlywood tracheid, and LWTracheid latewood tracheid. Scale bars 50 μm

The number of pits per tracheid varied from earlywood to latewood. Even though latewood has fewer and smaller pits, they are reported more permeable than earlywood tracheid in dried material, while earlywood is more permeable than latewood in greenwood (Petty and Preston 1969). Distribution profiles of preservative obtained by CT scan also provide information concerning the higher permeability of latewood than the earlywood (Fig. 3). Gradual distribution of preservatives from latewood to earlywood was also apparent. When the wood is dried, the position of the pit membrane is changed by the process of aspiration in which the torus moves across the pit chamber to seal off either side of the pit aperture, thus preventing fluid flow through the pit pair. After thermal treatment, pit membrane disruption and open and/or partially open bordered pits were observed (Fig. 6f, h) through which preservative liquid can traverse to the neighboring tracheids (Fig. 6d, g, i). The highest preservative uptake was observed in control pine sample treated with tar. Thus, special attention was given to find out the reasons. Control wood samples were preheated at 170 °C for 1 h before immersing in preservative solution which resulted in some degree of pit deaspiration. It was revealed from the SEM observation (Fig. 6d, e) that higher frequency of open or partially open bordered pit and increased permeability of tar through half-bordered pit membrane are probably the explanation for the highest amount of preservative uptake.

Secondary treatment on thermally modified wood was performed successfully using a simple and effective procedure. But the performance of this kind of treated wood in field condition is yet to be conducted.


Secondary treatment of thermally modified pine and spruce was conducted with Beckers and tar preservative. Both species retained a considerable amount of preservative, especially tar, without applying any external vacuum or pressure. Preservative distribution in different cells of thermally modified wood was examined through CT scan and microscopy techniques. CT scan image analysis provided distribution profile of preservative solution in pine and spruce samples. On the other hand, LM and SEM techniques provided information about the reasons related to the variability of preservative uptake. Anatomical differences between pine and spruce were related to the variation in permeability. Disrupted and open bordered pit due to thermal modification took an active part in preservative uptake and distribution. The lower frequency of axial and radial resin canals and the inability to form secondary flow paths due to ray cell collapse restricted the preservative uptake and distribution in spruce. Besides, the compositions of preservative liquids were also responsible for the differences of penetration level. Using turpentine in the tar formulation increased the preservative impregnation in pine and spruce compared with Beckers. For both kinds of preservative, the uptake amount decreased with an increase in sample length. The investigation into preservative uptake and distribution using the CT scanning technique has been found to be a useful method. Further studies should preferably be extended to using different species with larger dimensions.


Financial support from the European Union and the European Regional Development Fund, the County Administration of Västerbotten, the municipality of Skellefteå, and TräCentrum Norr is gratefully acknowledged.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012