Factors Affecting Permeability and Pit Aspiration in Coniferous

by GL COMSTOCK · 1968 · Cited by 157 — Longitudinal permeability of coniferous woods is controlled almost exclusively by the bordered pits. In green sapwood these pits are quite permeable and permit

132 KB – 13 Pages

PAGE – 1 ============
Wood Science and Technology Vol. 2 (1968) p. 279 -291 FOREST PRODUCTS LABORATORY (Madison, Wis., 53705) Forest Service, U.S. DEPARTMENT OF AGRICULTURE Approve d Technical Article Factors Affecting Permeability and Pit Aspiration in Coniferous Sapwood By G. L. COMSTOCK, Madison, Wis., and W. A. CÔTÉ JR., Syracuse, N. Y. Summary The influence of drying methods on the permeability of red pine and eastern hemlock sapwood was investigated. Permeability was found to be reduced by normal drying procedures to only a small percentage of the green permeability. The reduction was more severe at higher drying temperatures; less severe but still very large at -18° C. Pit aspiration was shown to be responsible for the reduction. Replacing the sap with surfactant solutions and organic liquids and evaporating them revealed that pit aspiration occurred with surfactant solutions having surface tension values of less than 20 dynes/cm and did not occur with organic liquids having surface tension values as high as 44 dynes/cm. It is suggested that a critical factor in pit aspiration is the adhesion of the torus to the pit border, and the failure of the organic liquids to cause pit aspiration is due to their inability to promote adhesion between the torus and pit border. Zusammenfassung Der Einfluß verschiedener Trocknungsverfahren auf die Durchlässigkeit des Splintholzes bei Kiefer und Eastern Hemlock wurde untersucht. Es zeigte sich, daß bei Anwendung üblicher Trocknungsverfahren die Durchlässigkeit gegenüber jener des frischen Holzes sehr deutlich vermindert war. Höhere Trocknungstemperaturen setzten die Durchlässigkeit weit stärker herab; bei -18° C war die Verminderung etwas weniger stark, jedoch noch immer beträchtlich. Als Ursache für die Durchlassigkeitsverminderung wurde der Tüpfelverschluß gefunden. Beim Austausch des Zellsaftes gegen oberflächenaktive Lösungen bzw. gegen organische Flüssigkeiten und bei nachfolgender Verdampfung derselben ergab sich, daß die oberflächenaktiven Lösungen mit Oberflächenspannungswerten von höchstens 20 dyn/cm einen Tüpfelverschluß erzeugten, die organischen Lösungen mit Oberflächenspannungswerten bis zu 44 dyn/cm jedoch nicht. Es wird postuliert, daß beim Tüpfelverschluß das Haften des Torus am Tüpfelrand das Kriterium bildet; die Unfähigkeit bestimmter organischer Lösungen, einen Tüpfelverschluß zu bilden, ist darin zu sehen, daß sie die Adhäsion zwischen dem Torus und dem Tüpfelrand nicht zu fördern vermögen. Introduction The permeability of wood is a measure of the ease with which fluids flow through it. Longitudinal permeability of coniferous woods is controlled almost exclusively by the bordered pits. In green sapwood these pits are quite permeable and permit easy passage of fluids and small suspended particles. During drying, however, these pits frequently become aspirated, resulting in a marked reduction in permeability. The research reported here is aimed at determining quantitatively the extent to which permeability is reduced by drying as influenced by drying conditions and the properties of the liquid evaporated from the wood. The influence of surface tension of the evaporating liquid is given particular attention, because it has previously been regarded as the most important factor in pit aspiration. Pit Aspiration Most important conifers, in particular those belonging to the Pinaceae family, possess a bordered pit structure with pit membrane characterized by a centralized

PAGE – 2 ============
280 G. L. C OMSTOCK and W. A. C ÔTÉ, J R. thickened disk, the torus, and a supporting membrane, or margo, consisting of strands of cellulose microfibrils [LIESE, 1965]. In the green condition, most of the sapwood pit membranes are centrally located and quite permeable. Drying the wood generally causes the torus to be displaced and come into contact with one of the pit borders. This phenomenon is referred to as pit aspiration and the forces which cause it to occur were discussed in detail by HART and THOMAS in 1967. In 1933,PHILLIPS made what appears to be the first comprehensive study of pit aspiration. He found that drying sapwood caused a gradual increase in the number of aspirated pits with loss of moisture down to the vicinity of the fiber saturation point. At this point, virtually all the springwood pits became aspirated; whereas about one-third of the summerwood pits remained unaspirated. He ascribed the greater tendency of summerwood pits to resist aspiration to the greater rigidity of the summerwood pit membrane. LIESE and BAUCH in 1967 reported observing the same phenomena. BRAMHALL™S 1967 permeability studies indicate that there is little change in summerwood permeability upon drying, but spring-wood permeability can vary as much as 30 times, depending upon the drying method. GRIFFIN in 1919 and 1924, PHILLIPS in 1933, ERICKSON and C RAWFORD in 1959, and LIESE and BAUCH in 1967 observed that pit aspiration does not occur in coniferous sapwood when the water is replaced by alcohol prior to drying. The prevention of aspiration is invariably ascribed to the lower surface tension of alcohol compared to water. ERICKSON and CRAWFORD observed that when sapwood was air dried the permeability was reduced by drying to only 1 . . . 3 percent of its original value in the green condition; whereas drying by solvent exchange with alcohol, acetone, or alcohol-benzene followed by evaporation of the solvent prevented pit aspiration and maintained the permeability at its original level. LIESE and BAUCH dried several conifers by solvent exchange, using alcohol-water and acetone-water mixtures. They observed that for solvent concentrations in excess of 75 percent ethanol or 80 percent acetone, pit aspiration was incomplete. These concentrations correspond to a surface tension of about 26 dynes/cm in the pure liquids, and LIESE and BAUCH concluded that a surface tension of 26 dynes/ cm is required to cause pit aspiration in springwood. The concept of the authors regarding pit aspiration is that three factors are involved, any one or combination of which may control aspiration. These are: (1) Surface tension forces tending to pull the torus into contact with the pit border. (2) Rigidity or stiffness of the pit membrane which results in a force opposing the surface tension forces exerted by the evaporating liquid. (3)Adhesion of the torus to the pit border when they are brought into contact. The first factor, surface tension forces of the evaporating liquid, is primarily dependent on the surface tension of the liquid and the sizes and shapes of the pit aperture, chamber, torus, and pores in the margo. In 1967, HART and THOMAS discussed in detail the forces which may develop due to the presence of air-liquid menisci in a pit. They concluded that the forces associated with a meniscus between the torus and pit border increase with loss of liquid as the torus is drawn closer to the pit border.

PAGE – 3 ============
Coniferous Sapwood Permeability and Pit Aspiration 281 The relative rigidity of the pit membrane compared to the surface tension forces will determine whether the torus is displaced sufficiently from the central position to come into contact with the pit, border. The stiffer and more rigid the membrane, the greater will be the force required to aspirate the pit. Summerwood pits tend to aspirate less than springwood pits, evidently because summerwood pits are usually smaller and have a thicker, more rigid membrane. Another factor which may affect, the rigidity of the pit membranes, when drying from non aqueous liquids, is the degree of swelling of the liquid. It is well known that most strength properties of wood, including the modulus of elasticity, increase with loss of moisture below the fiber saturation point). Strength of wood swollen with liquids other than water would be expected to change in similar fashion with the degree of swelling of the wood. For example, E RICKSON and R EES observed in 1940 that the crushing strength of wood decreased with increasing degree of swelling caused by the organic liquid in which the wood was soaked. It is antici pated, therefore, that the stiffness of the pit membrane may be increased by re placing water with liquids of lower swelling power. The third factor involves adhesive forces between the torus and the pit border. Assuming that the surface tension forces are sufficient to overcome the stiffness of the pit membrane and thus bring the torus into contact with the pit border, the torus will stay in this position only if there are adhesive forces between it and the pit border. This particular aspect has apparently been ignored in the past and is worthy of consideration. Whether an intermediate substance, such as a resin, is involved as an adhesive or whether there is direct bonding of the surface of the torus to the pit border, is unknown. The type of forces involved is likewise un known. Permeability Measurements The purpose of these experiments was to determine the extent to which the permeability of coniferous sapwood is changed by various drying methods and to determine whether the changes observed are related to aspiration of the bordered pits. Permeability measurements prior to drying were made with water. Measure ments after drying were made with nitrogen gas. That comparisons of the measure ments obtained by these methods are valid has been demonstrated by C OMSTOCK in 1967 and 1968. He showed that the permeability values obtained with gases and liquids are essentially identical if viscosity is included in the permeability equation and a correction is made for slip flow of the gas. That gas permeability increases with loss of moisture in the hygroscopic range because shrinkage of the wood causes the intertracheid pores to become larger was also shown. This effect was found to be very small in solvent -dried wood and somewhat larger in air -dried wood. Water permeability was determined using the technique described earlier [COMSTOCK 1965]. Permeability values were calculated using D ARCY™S law. (1) where k = permeability (Darcys) = viscosity (centipoise) Q = flow rate (cm 3/sec) A = flow area (cm 2) L = specimen length (cm) P = pressure drop (atm) 23 Wood Science and Technology, Vol. 2

PAGE – 4 ============
282 G. L. C OMSTOCK and W. A. C ÔTÉ JR. The superficial gas permeability, kg, was calculated according to D ARCY™S law for gases, which takes into account the compressibility of the gas. where Pis the absolute pressure at which Q is measure and P is the mean absolute pressure in the specimen. Slip flow was then corrected for by use of the following expression [C OMSTOCK 1967]: (3) where = mean free path of the gas = radius of the intertracheid pores The procedure used for gas permeability measurements was to measure k, at mean pressures of one and three atmospheres and to extrapolate the plot of k, vs. 1/ P . . . 1/ P equal to zero. The value of k obtained in this way is the slip -corrected permeability. The apparatus used for gas permeability measurements is similar to that described by C OMSTOCK in 1968. Influence of Temperature and Rate of Drying on Permeability This experiment was conducted to determine the extent to which the reduction in permeability on drying depends on the drying conditions employed. Five drying temperatures were used: -18°,20°, 60°, 100°, and 140° C. At the three intermediate temperatures, two rates of drying were used. Only slow drying was carried out at -18°C and only rapid drying at 140° C. Drying at -18°C was accomplished by placing the specimens in a desiccator over phosphorus pentoxide and placing the desiccator in a room controlled at -18° C. Drying at 140° C was accomplished by putting the specimens between two platens heated to 140° C and bringing the platens into contact with the wood. All other drying was done in a humidity -and temperature -controlled cabinet. Rapid drying was achieved by placing the specimens directly into the air stream in the cabinet. Slow drying was achieved by placing the specimens in a desiccator over P 2O5 and placing the desiccator in the cabinet at the desired temperature. Since no circulation was provided in the desiccator, drying was much slower than in the open air stream. Specimens were taken from three eastern hemlock trees, and two specimens from each tree were dried at each set of conditions. The specimens used in this experiment were approximately 1.3 cm in diameter by 2 cm along the grain. Specimens from a given tree were all taken from the same growth rings to minimize the influence of within -tree variation on the results. Water permeability of the green sapwood of several specimens was determined prior to drying so that the reduction due to drying could be estimated. The average permeability before drying was found to be 2.78 Darcys. The average permeability values after drying are plotted graphically as a function of drying temperature in Fig. 1. Specimens dried at -18°C had perme –

PAGE – 5 ============
Coniferous Sapwood Permeability and Pit Aspiration 283 ability values considerably higher than those dried at the higher temperatures, but they were still only about 5 percent as permeable as they were before drying. Above 27° C there is a grad ual trend toward decreasing permeability with increasing drying temperature. Rate of drying had no effect on per meability. This is contrary to the findings of E RICKSON and C RAWFORD in 1959, which showed that slow drying at room temperature resulted in higher permeability, whereas drying at 100° C had the same effect as drying rapidly at room temperature. The results of this experi ment indicate that pit aspira tion is most severe in specimens dried at high temperatures, but it appears to be reasonably complete in specimens dried at -18°C, since the permeability of specimens dried at this tem perature is only about one -twentieth as large as it was be fore drying. It was expected Fig. 1. Slip -corrected nitrogen permeability of dry specimens as a function of drying temperature and rate of drying. that pit aspiration would not Permeability before drying averaged 2.78 Darcys. occur in specimens dried at -18° C, because freezing of the free water in the wood should eliminate the surface tension forces which cause aspiration. In 1967, T HOMAS observed that vacuum freeze drying did prevent aspiration from occurring in pine, but the drying temperature was probably lower. It is speculated that sufficient liquid phase was present at -18°C to exert the attractive force to cause pit aspiration. The more severe aspiration at higher temperatures is apparently due to the greater plasticity of the wood at this temperature rather than surface tension, since surface tension decreases with increasing temperature. Influence of Surface Tension and Other Properties on Pit Aspiration To isolate some of the factors involved in pit aspiration, the sap in a series of sapwood permeability specimens was exchanged for liquids having different surface tension and swelling properties. Aqueous surfactant solutions and organic liquids were used to vary these properties. Two experiments were conducted using this exchange procedure. The first was done using eastern hemlock samples from the three trees used in the drying condition experiment. Two specimens from each tree were assigned randomly to each of the exchange treatments. The treatments are given in Table 1. The 23*

PAGE – 6 ============
284 G. L. C OMSTOCK and W. A. C ÔTÉ, J R. Table 1. Exchange Treatments given Samples Prior to Drying 0.1 % fluorocarbon surfactant (cationic) 0.1 % fluorocarbon surfactant (nonionic) code letter Treatment 0.1 % organo -silicone surfactant (nonionic) 0.1 % organo -silicone surfactant (cationic) Ethanol water 1 Ethanol Methanol acetone pentane 1 Methanol acetone toluene 1 Controls -air-dried immediately Arrow indicates succession of treatments on the same specimens. surfactants used are powerful surface tension depressors, which lower the surface tension well below the value of 26 dynes/cm suggested by LIESE and BAUCH in 1967 as a lower limit which causes aspiration. The other treatments were aimed at determining the effect on permeability of drying from nonaqueous liquids of varying surface tension. Table 2. Surface Tension of Several Exchange Solutions and the Permeability of Eastern Hemlock Sapwood after Drying from these Solutions1 –19.4 23.2 23.5 28.2 —-2.78 .053 .071 .114 .098 .094 .088 3.01 4.44 5.10 Treatment None -R 2 -air-dried A -FC-134 B -FC-170 C -L-77 C -L-79 K1 -Water K2 -Ethanol N -Toluene L -Pentane (water permeability) Initial2 dynes per cm –18.7 21.8 22.8 24.3 72.0 24.0 29.6 18.4 Surface tension Final3 Permeability4 Darcys Measurements of permeability were made with nitrogen gas after drying wood over P2O5. Permeability values are corrected for gas slip. Value obtained from tensiometer measurements on liquids before specimens were introduced. Value obtained from tensiometer measurements on liquids after specimens were removed. Average of six or more specimens. The results of this experiment are shown in Table 2. The surface tension values listed were determined on the solutions in which the specimens were immersed using a ring tensiometer. Air drying is seen to reduce permeability to about one-fiftieth of the permeability before drying. The effect of drying after

PAGE – 8 ============
286 G. L. COMSTOCK and W. A. CÔTÉ, JR. The swelling by the different liquids was determined experimentally using specimens about 2 cm square by 5 mm along the grain. These were ovendried, measured, and immersed in the desired organic liquid and the change in dimension after 6 clays of immersion was taken as the swelling. Radial and tangential swelling were measured on two specimens of each species for each liquid. The average of the radial and tangential values is expressed as the percentage of the swelling in water in Table 3. Some swelling occurred in organic liquids which were cited by STAMM in 1964 as nonswelling, such as benzene. This could have been due to trace amounts of water in the liquids or adsorption of atmospheric moisture during the measurements. The values in general, however, agree well with S TAMM™S values, with the exception of dioxane, for which S TAMM reports a value of 62 percent as opposed to 10 and 7 percent found here. Results of permeability measurements described earlier are also shown in Table 3. It is interesting that in no case involving solvent drying is there evidence of pit aspiration even though surface tension values as high as 44 dynes/cm were observed. The only cases where permeability was reduced were those of drying directly from the green condition and drying after going from sap to acetone and back to water. The latter drying procedure gave somewhat higher permeability than air drying, but still reduced permeability well below the values for the green wood. Red pine was reduced to about 15 percent of its original permeability and eastern hemlock was reduced to less than 1 percent of its original permeability. Permeability values after drying from any of the organic liquids were similar and slightly higher than the values measured with water before drying. The slightly higher values after solvent drying may be due to removal of extractives or perhaps deaspiration of some pits by the solvents. There does not appear to be any trend in permeability with either surface tension or swelling of the liquids, so it seems unlikely that the higher permeability is produced by evaporation of the solvent from the wood. The increase produced by solvent drying is small compared to the decrease caused by evaporation of water from the wood and mill not be considered further here. Electron Microscopic Examination of Pits in Hemlock Specimens of eastern hemlock dried from several of the different liquids were examined to see whether changes in the permeability were consistent with visual changes in the bordered pits. Replicas were prepared from split radial surfaces using the direct carbon replica technique described by C ÔTÉ KORAN, and D AY in 1964. The results of these investigations agreed with those from the permeability studies. Pits in air -dried specimens were almost invariably tightly aspirated as shown in Fig, 2. Tight aspiration is inferred when the imprint of the pit aperture on the torus is clearly visible, indicating that the torus is sealed tightly against the pit border. Examination of samples dried from surfactant solution C revealed that the pits are aspirated, but not tightly, since the aperture imprint was missing. Fig. 3 is typical of the pits found in specimens examined. The less severe aspiration is manifested also in many supporting strands in the margo being elevated above the back border. The less severe aspiration is evidently responsible for the slightly higher permeability compared to the air -dried controls.

PAGE – 9 ============
Coniferous Sapwood Permeability and Pit Aspiration 287 Fig 2. Electron micrograph of a tightly aspirated pit in air-dried eastern hemlock sapwood. Magn.: 5 800:1. Fig. 3. Electron micrograph of an eastern hemlock sapwood pit dried after the water had been replaced by surfactant solution C. Magn.: 5 600: 1.

PAGE – 10 ============
Fig 4. Electron micrograph of an unaspirated eastern hemlock sapwood pit dried by acetone exchange. Back border was torn away during replication, revealing the fibrillar structure of the margo. Magn.: 4 200:1 Fig. 5. Electron micriograph of two unaspirated pits in eastern hemlock sapwood dried by ethanol exchange. Torus extensions which are charateristic of hemlock are clearly visible. The dark band around the torus is the outline of the replica of part of the pit border behind the torus. Magn.: 3 150:1.

PAGE – 11 ============
Coniferous Sapwood Permeability and Pit Aspiration 289 Fig. 6. Electron micrograph of an unaspirated pit in eastern hemlock sapwood dried by solvent exchange using benzene as the evaporating liquid. The light circular area on the torus is the area of the pit aperture behind the torus. The dark band occurred because the entire pit border behind the torus was replicated and remained intact. Magn.: 4 700:1. Specimens dried after solvent -exchanging water for ethanol, acetone, benzene, and ethyl ether were examined and the invariable result was that the majority of the pits were unaspirated, which corresponds to the permeability findings. Typi cal electron micrographs are shown in Figs. 4, 5, and 6. Notice first that the tex ture of the margo is clearly visible in these unaspirated pits and the margo is seen to have a very high porosity. This corresponds to the 1966/67 findings of T HOMAS and N ICHOLAS on solvent-dried southern pines. Although different pits often exhibited greatly different margo densities, no consistent differences were observed between samples dried from the different organic liquids. This also agrees with the permeability results. Direct carbon replicas of unaspirated pits exhihit some peculiar traits not found in replicas of aspirated pits. For example, the dark band which appears around the edge of the torus in Figs. 5 and 6 is due to the deposition of a layer of carbon and some metal on the pit border behind the torus. Where the replica of this border occurs behind the torus there is a darker appearance because two layers of carbon must be penetrated by the electron beam, one being the replica of the torus, the other the replica of the back border. In Fig. 6 the entire back border is intact and the lighter area in the center of the torus corresponds to the pit aperture. Frequently, in replicating unaspirated pits, the back border is torn away during the process of replication because of its loose attachment to the annulus. When this occurs, a replica similar to that shown in Fig. 4 results, showing the

132 KB – 13 Pages