“ II 'Hz; I ‘ “y . I“: H .w I H I r ‘I y; I l H I‘ I I > I ‘ ‘ '—I_._. i Ibo—t _cnoooo 2 THE EFFECTIVENESS OF POLYETHYLENE GLYCOL AS A BIRCH VENEER STABILIZER Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY LoweIl Dale Meflers 19530 IIIIIIHIIHHIIHIIIHHIIIIHIIHIIIHHIIlmuuIumm 31293 01103 2400 LIBRARY Abstract This research study was conducted primarily to investigate practical methods of stabilizing wood with polyethylene glycol. To investigate this, three methods of apolication, three concen- trations of solution and three moisture conditions of the wood at the time of aoolication, were tested. Each combination of variables was reolicated two times. The effectiveness of the stabilization was measured between oven dry and soaked condi- tion and the results were compared with the control soecimens that had no treatment. The retention of the polyethylene glycol in the wood after treating was calculated and comoared with the stabilization effectiveness. The results of the research are summarized in the following list: 1. Pressure treating and soaking gave the best birch veneer stabilizaticn effectiveness of the methods tested. The dipping method gave significantly less veneer stabilization effectiveness. 2. Birch veneer stabilization is not affected signifi- cantly by the moisture content of the specimens, for solutions tested in this study, except when pressure treating or dipuing veneer which contains 90 per cent moisture. 3. There is no significant difference between treating with 33.0% or 25.4% oolyethylene glycol solutions. lb.6% solution produced significantly less birch veneer stabilization. Using 1e.et, 25.5% or 33.0% polyethylene glycol solutions, a oressure treatment or a diooing treat- ment of 1/20" birch veneers can be comoleted in 72 hours and a soaking treatment in 80 hours. The oolyethylene glycol treatment does darken he wood. THE EFETCTIV:EESS CF RUJIETE LEHE GLZCTL n3 A BIRCH VEIZER STaBILIZER By LENELL Deli 313 TEES A THESIS Submitted to the College of agricu ture Michigan State University of Agriculture and Apnlied Science in partial fulfillment of the requirements for the degree of LmoTER OF SCIENCE Department of Forest Products 1960 hporoved L- 'r *' } ii AC KNO; 1' LEDGE {XE NT 8 The writer wishes to express his appreciation to Dr. A. E. Wylie, and Dr. alan Sliker for their guidance and criticism in the preparation of this manuscript. Grateful acknowledgement is also extended to Dr. N. D. Eaten for his advice and assistance on the statistical analysis of this study. Finally, the author is incebted to Henry A. Huber from the Dow Chemical Co. and Ray Birdsall from the United States Plywood Corporation for their wonderful cooperation and donation of materials. iii r—3 “’1 E I C‘ r23 0 59 F3 :4 :3 U‘ I, aCKNO.1” MIG IEPTS . . . . . . . . . . . . . _JST OF TaBLES . . . . . . . . . . . . . . . . . LIST OF FIGURES. . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . . . . Past Attempts to Solve the Problem. . . . . Previous dork with Polyethylene Glycol 1030 . Relation of Fiber Structure to Stabilization Purpose of this Study . . . . . . . . . . . PROCEDURE . . . . . . . . . . . . . . . . . . . . Preparation of Samples. . . . . . . . . . . . Treating by Dipping . . . . . . . . . . . . Treating by Soaking . . . . . . . . . . . Treating by Pressure. . . . . . . . . . . . . Conditioning Control Specimens . . . . . . . RESUETS nKD ANALYSIS . . . . . . . . . . . Stabilization Effectiveness . . . . . . . . . Treatment Concentration . . . . . . . . . . . Treatment Methods . . . . . . . . . . . . . Treatment Conditions . . . . . . . . . . . . Retention of Polyethylene Glycol . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . COIICLUS IOIJS o o o o o o o o o o o o o o o o o o o APPIC: IDH O O O O O O O O O O O O O O O O O O O O O BELIOGRK‘PHY O O O O O O O O O O O O O O O O O O 0 ii iv H \IWWH 03 . 27 37 Table VI. VII. VIII. iv LIST Q_ _§BLES Design of Experiment ... ..................... average Percent Stabilization Effectiveness ReS‘JltS. O O O o - Average Percent Stabilization Effectiveness Results ................... . ...... . ........ .... Analysis of Variance Results.. ................ Studentized Range Test Results.. ........... ... Dimensional Change from Oven Dry Condition.... Percent Stabilization for Each Specimen....... Percent Retention for Each Specimen....... 29 29 .30 kt.) \h w 0\ LIST @ mamas Figure Page 1. Diagram of the Cellulose Framework ................. 6 2. Diagram of a Veneer Testing Specimen ....... ....... . lO 3. Bar Graph of the average Percent Stabilization lO. Effectiveness for Different Solutions............... 15 Bar Graph of the average percent Stabilization Effectiveness for Different Treating Methods.. ...... 17 Bar Graph of the Average Percent Stabilization Effectiveness for all Conditions with Each Solution Using Each Method.... ..... . ............ .... ........ 20 Bar Graph of the average Percent Polyethylene Glycol Retained for Different Treating Nethods. ..... 21 Bar Graph of the Average Percent Polyethylene Glycol Retained for Different Solutions..... ...... .. 22 Graph Showing Interaction Between Condition and Solution... ............................. . ....... 31 Graph Showing Interaction Between Method and Solution..... ................................... 32 Graph Showing Interaction Between Condition and IslethOd ......OOOOOOOOOOO......OOOOOOOIOOOOOOOIOO 33 INTRODUCTION One of the most disturbing characteristics of wood used in its natural state is its tendency to shrink and swell perpendicu- lar to the grain, when the moisture content of the wood varies in the range from zero to about 30 percent. Loosed joints, finish failures and many other difficulties in wood products result from their dimensional instability. In addition, since wood is an orthotrOpic material, it does not shrink and swell equally in all directions. This characteristic causes stresses to build up in the wood, causing warping or checking. If wood is to maintain its present importance in the world, a solution to this problem must be found. Past attempts £2 Solve the Problem One of the first attempts was to close out all moisture from the wood cells with an external or internal moisture barrier (8). If wood could be surrounded by a perfectly moisture-proof cover, swelling could be prevented. These treatments have been effective in slowing down the rate of moisture absorption but after a long exposure to moisture the wood would reach its completely swelled condition. The internal coatings have been less effective in cutt- ing down the rate of swelling than surface coatings, mainly due to their reduced thickness. Cross Linking Another method was to form cross links between the cellulose chains which make up the fiber structure. Cross links are chemical links attached to the hydroxyl groups. The cross links hold the cellulose chains together and restrict the amount of water that can enter the wood, thus reducing the amount the wood can swell. Formaldehyde vapor in the presence of a strong mineral acid catalyst was shown to stabilize the dimensions of wood due to cross linking between cellulose molecules (9). The wood is embrittled to a great extent due to the required acid. Other cross linking agents must be found that do not require such acetic conditions before this type of treatment can find extensive use. Bulking Agents Another method of wood stabilization is one of depositing a bulking agent within the cell wall thus holding the wood in its swollen condition. Efforts were made to find a bulking agent that reacts with the hydroxyl groups of the ca lulose and does not em- brittle the fibers. The acetylation process was tried (8). The wood was dried to about 2 percent moisture, then vapors of acetic anhydride and pyridine were absorbed by the veneer. The results were 70 percent antishrink efficiency, treating at 900 C. for six hours. This efficiency rating was based on a control, under the same conditions, that had no treatment. The tests showed that acelyated wood has about the same strength properties as untreated wood. The highest dimensional stabilization of wood to the pre- sent time has been by acetylation. A second bulking method makes use of a wax which is insoluble in water (8). The water in the cell wall must be replaced by a solvent for the wc*. Then the solvent may be replaced by some type of water insoluble substance whose molecular dimensions are not too large to stop penetration of the fine cell wall capilla- ries. This process, however, is too involved to be practical. The most recent method and the one used in this study is that of depositing polyethylene glycol, a water soluble bulking agent, into the cell wall structure, thus holding the wood fibers in a swelled condition (10). Polyethylene glycols are high mo- lecular weight compounds, produced by reacting alkylene oxides with compounds having an active hydrogen ion. They are made with molecular weights from 200 through 600, are soluble in water in all proportions and with slow drying conditions can completely replace water in wood. Tests using this material in wood have shown almost complete dimensional stabilization. The solubility decreases with each increase in molecular weight. Past tests have shown that cross sections of wood treated with the lower molecular weight ma- terial tended to remain damp at lower humidity ranges. For this reason polyethylene glycol with a molecular weight of 1000 has been used extensively in tests for wood stabilization. Previous work With Polyethylene Glycol 1000 The only published work on polyethylene glycol as a wood stabilization bulking agent was done by Stamm (ll) of the Forest Products Laboratory at Madison, Wisconsin. He found that about 30 percent of the dry weight of the wood, of this chemical is re- quired to give 80 percent wood stabilization. His tests showed that polyethylene glycol is effective in stopping or reducing checking when drying tree cross sections. He found that face checking of plywood after cycling between 90 and 30 percent relative humidity was practically eliminated. Stamm conducted strength tests after treating with poly- ethylene glycol. The data showed practically no effect upon the toughness of the wood. The abrasive strength and modulus of rup- ture were not affected except for the normal strength reduction when wood is in a swelled condition. His decay resistance tests showed that all specimens containing at least 17.6 percent poly— ethylene glycol showed no decay. Specimens containing only 8.6 percent polyethylene glycol showed some decay. He also conducted gluing tests on veneer treated with poly- ethylene glycol. He used casein, polyvinyl emulsion, cold-setting urea resin, cold-setting resorcinol resin, hot-setting phenolic resin and a hot animal glue. The data indicated that good glue joints can be obtained for wood containing as much as 33 percent polyethylene glycol with casein, cold-setting urea and hot-setting phenolic resin giving best results. Polyvinyl emulsion and resor- cinol glues showed a decrease in tensile strength with each increase in polyethylene glycol content. The hot animal glue showed no de— crease in tensile strength with increasing polyethylene glycol content. He conducted finishing tests on ponderosa pine specimens, using seven different finishes. Three days after applying the second coat the specimens were tacky, especially those containing the largest amount of polyethylene glycol. These specimens later dried and r881 ted in a satisfactory finish. Stamm's last tests were to determine the leaching charac- teristics of polyethylene glycol. Iis tests showed that in spite of the fact that polyethylene glycol can be readily leached from the wood, a good surface coating practically stopped all leaching. Relation pf Fiber Structure t9 Stabilization Considerable research has been conducted to discover how wood cell walls are constructed and how and why wood swells when it is subjected to moisture. Of all the theories expounded over the past years the Fringe Miceller Theory has the support of most students of the subject (6). It is generally conceded that the fibrils which make up the cell walls are formed by a large number of long chain cellulose molecules. If the cellulose molecules were arranged in parallel throughout a fibril, there would be no space for water to enter and swell the fibril. However, they are aligned parallel only in some areas. These areas are called crystalloids. In other ‘ areas they are not aligned in any special way. These are called the amorphous regions and this is where water has a chance to enter between the disarranged long molecules where it is attracted and held by the hydroxyl groups. This theory of the minute structure of a fiber is entirely deductive and based on chemical and physical evidence. The struc— ture has never been seen, since present light and electron micro— scopes cannot bring it into view. It is from this theory that scientists reasoned that if one could deposit a bulking agent between the cellulose molecules in this amorphous region, the wood would be held in a swelled condition after drying. Polyethylene glycols were chosen since they are Figure l. I}. fi Diagram of the cellulose framework as conceived in the Fringe Miceller Theory: (A) amorphous region, with no regular arrangement of the cellulose chains; (B) crystalline region with regularly arranged cellulose chains; (0) intermicellar space. soluble in water and had previously been shown to completely displace water. Purpose 9f this Study Considerable basic research has been conducted by Stamm (ll) to determine the effectiveneSS of polyethylene glycol as a wood stabilizer. Th,se tests took many weeks to complete and the results were almost complete wood stabilization. This extensive a treatment would be completely impractical for modern industry. The purpose of this study was to investigate more practical methods of treatment and still obtain a satisfactory degree of stabilization. The specific objectives were: 1. To develop a practical method of application. 2. To discover the effect of different percent poly- ethylene glycol in the treating solution. .3. To determine the effect of different moisture contents in the specimens treated. 4. To determine the percent polyethylene glycol absorbed, based on the oven dry weight. 5. To observe the color change after treating. PROCEDURE Preparation 2f Samples Birch (Betula alleghaniensis) face veneer cut l/20th inch thick was selected for this study. Birch was selected because it shrinks and swells to a large degree and it is also of great commercial importance. This veneer had been rotary cut and dried with the common veneer roller dryer. No special plan was used in selecting the veneer specimens. Sixty-two specimens were cut to size, ll X 5 inches, on a shop band saw. They were cut with the fiber direction parallel to the five-inch side and all measurements were made perpendicular to the fiber direction. The specimens were then sanded lightly to remove any loose pieces of wood that could fall off later and in— validate the weight measurements. They were checked closely for checks, splits or any other damage that could change the shrinkage or swelling measurements and all were marked with a letter and number combination as indicated in Figure 2. at this time midpoints along each 5 inch side were marked with a sharp pencil. This pro- vided a definite location to measure after each change in moisture condition. One-fourth inch from the five-inch side at either marked midpoint a one-fourth inch knife mark was made parallel to the fiber direction. This knife mark was blackened with a sharp pencil and is called the first measuring point. \O *1 Table I. Design of the nxoeriment. Numbers in the » cells incicate the number of replications. moisture of Concentration Hethods ___ Specimens, % of solution, % Bio bOuk Pressure 33.0 2 2 2 O 75.4 2 2 2 16.6 2 2 2 33.0 2 2 2 20 25.4 2 2 2 16.6 2 2 2 33.0 2 2 2 90 25.4 2 2 2 Figure 2. 10 Grain Direction Tangential Direction //;;w First measuring point Midpoints Diagram of a veneer testing specimen. An example of the letter and number identification is: (A) 16.6% polyethylene glycol solution. (1) First replication. (D) % moisture content of the specimen treated. (P) Pressure treating method. 11 Six prOperly marked specimens were placed in the oven at 1050 C. for 12 hours. They were oven dry since they had ceased to lose weight after weighing at an hour inverval. One specimen was taken from the oven and the weight was recorded. Then ten inches was measured off along the midpoints from the first mea- suring point. A l/Lth inch knife mark was made at the ten—inch point. This knife mark was blackened with a sharp pencil. The preceeding process was continued until all six specimens had been weighed and marked. This process was completed as quickly as possible so the specimens would not absorb enough moisture from the air in the room to seriously affect the data. All measurements were made with a 15-inch steel ruler that was marked off in l/lOOth inch increments. The specimens were placed on a flat surface and the ruler was placed on edge at the marked midpoints. The ruler was then forced down and the readings were made between the blackened marks. It was felt that the error in measurement would not be more than l/lOOth inch. The six speci- mens were then conditioned in the humidity cabinet at 1000 F, and 95% relative humidity for three days. It was concluded that they had reached the equilibrium moisture content since their weight ceased to change when weighed at hour intervals. Six other specimens were oven dried, weighed, measured and marked just as those above. They were conditioned by submerging them in water at room temperature for ten hours. The last six of the 18 specimens needed for each treating method were oven dried, weighed, measured and marked as above. These were left in the oven dry condition for treating. 12 It was necessary to schedule the conditioning of the above Specimens so all 18 would be rea y for treating at the same time. The last step before treat ng was preparing the polyethylene glycol for treatment. This material is a wax-like solid at room temperature. Solutions of 33, 25.4 and 16.6 percent polyethylene glycol in water were measured and placed in separate beakers. These solutiins were heated to 150° F. and the polyethylene glycol was completely dissolved in the water. These solutions were then poured into separate 6 x 12 inch cake pans and kept at 1500 F. Treating, by Dipping All the dipping specimens were completely submerged in the correct solution one at a time. They were immediately removed and the excess solution Was allowed to drain away. After a 15 minute draining period all specimens were placed in the humidity cabinet at 100° F. and 953-3 relative humidity and allowed to condi- tion for three days. It was felt that they had reached equilibrium moisture content since their weight did not change at one-hour intervals. They were weighed and measured and these data were recorded. Next the specimens were oven dried, weighed and mea- sured, then placed in water to soak for 12 hours and a final measurement was recorded. 13 Treating py Seaking The 18 conditioned specimens were completely submerged in the proper solution and placed in the oven at 1500 F. for eight hours. After this treatment they were removed and the excess solution was wiped from their surfaces. After treating they were handled the same as the dipped specimens. Treating by Pressugg The last treatment method was that of forcing the polyethy- lene glycol solution into the veneer by pressure. The specimens were completely submerged in the correct solution, then the solu- tions were heated to 1800 F. It was felt that the temperature of the solution would cool to about 1500 F. when full pressure was obtained since the temperature in the pressure chamber was 85° F. It took 15 seconds for the pressure to reach 150 p.s.i. This pressure was maintained for 15 minutes, then it took 90 seconds to release the pressure. The specimens were then removed and conditioned, weighed and measured, as with the other methods. I Conditioning Control Specimens The final procedure was to condition the eight remaining control specimens. These were conditioned the same as the others except they received no treatment with polyethylene glycol. They were oven dried, measured and marked, then conditioned to equili- brium moisture content in the humidity cabinet at 1000 F, and 95% relative humidity. They were weighed and measured, soaked in water for 10 hours, then measured again. This gave a base with which to compare all other treatments. RESULTS AHD ANaLYSIS Stabilization Effectiveness Tne control specimens had all shrunk to their oven dry size after being conditi;ned at 90% relative humidity and 1000 F. then oven dried. Their average change in length from oven dry to soaked condition was .75 inch. The percent stabilization effectiveness which was calculated for each treated specimen was based on a maximum shrinkage of .75 inch for a 10 inch specimen. Treated veneer would not shrink to the original 10 inches when oven dried. The increase from the original 10 inches to the oven dry length of the treated veneer is a measure of the treatment's stabilization effectiveness and when the .75 inch basic swelling length was divided into this distance, a percent stabilization effectiveness figure was obtained for each specimen and was used to evaluate each treatment. Treatment Concentration The effect of treating birch veneer with 16.6%, 25.4% and 33.0% polyethylene glycol 1000, dissolved in water, was tested using an analysis of variance (Table IV). This analysis shows a significant difference in the percent stabilization averages of the three concentrations of solution. To determine exactly which concentration average was significantly different a studentized range test was conducted. This showed a signific nt difference between the percent stabilization average of the 16.6% solution and both 33.0% and 25.4% solutions (Table V). The effectiveness of 15 100% 90% m 80% U) o 5 . .3 70% +7 0 (3 u 60¢ “ 23 p s -g 50% N -H 1:1 4 ('8 40 ,0 33 13 30 % o o 53 o. 207; 10 % 0% 16.6; 25.4% 33.0; Percent Polyethylene Glycol in water Figure 3. Average percentaiabilization effectiveness for three solution concentrations. 16 these concentrations are also illustrated by the use of bar graphs (Figure 3). The Values represented by the bar graphs are the averages of all the percent stabilization effectiveness fi- gures for each treatment concentration. The graph shows there is an increase in stabilization effectiveness with each increase in treatment concentration. There was a large increase in effec- tiveness between the 16.6% and 25.4% and a small increase between 25.4% and 33.0%. This indicates the relation between concentra- tion of the solution and stabilization effectiveness is not linear and further investigation should be conducted to find the best concentration. Treatment methods The effects of treating the specimens using the dipping, soaking and pressure treatment methods were analyzed by an analy- sis of variance as shown in Table IV. There is a sign'ficant difference between the percent stabilizati n averages of the three methods used. The studentized range test showed a significant dif- ference between averages for dipping and both soaking and pressure treatments. There was no significant difference between soaking and pressure treatment averages. The bar graph (Figure 4) shows a large inc ease in stabili- ‘3 zation effectiveness from dipping to pressure or soaking methods. 100% 90% 33 a 80% (D p -H ‘53 70;; ‘H c... [:1 a w»: H .p S ...q 50;) :1 ,0 S on 40;, .p :1 E”: 30 55 5‘3 20% 10% 0% Figure 4. l7 Dip Soak Pressure Methods of Treatment Average percent stabilization effectiveness for three treating methods. 18 Treatment Conditions The last variable analyzed was the moisture content of the veneer treated. The analysis of variance showed no significant difference among the percent stabilization averages of the veneer treated at three moisture contents (Table IV). The analysis of variance for all variables showed large interaction effects. This means that the stabilization effec- tiveness did not change directly as one variable changed. The effectiveness changes depending on the combination of the varia- bles used. This is shown more clearly in the graphs in Figures 9, 10 and 11. The crossing of the lines shows that the effective- ness of the stabilization is not dependent on any one variable but on how the combinations react together. Retention 2: Polyethylene Glycol The retention of polyethylene glycol in the specimens was calculated by subtracting the weight of the specimens oven dry without treatment from their weight oven dry after treating. This gives the weight of the polyethylene glycol retained. Then the oven dry weight before treating was divided into the retained weight. This gives the percent polyethylene glycol retained, based on the oven dry weight. Figure 6 shows the average percents of polyethylene glycol retained with each treating method. This graph shows that there is an increase in retention from dipping to pressure or soaking method. There is a large increase in retention from dipping to l9 pressure but very little difference between pressure and soakin" methods. Figure 7 shows a steady increase in retention from 16.6% to 25.4% to 33.0% solutions of polyethylene glycol in water. 20 F O 82 80 8m 8e 80 new .zoapmnpcmosoo Soapsaom Use eczema pepm0H©efi on» we Hoohaw msefihnpozaoe a h! 8e 8m 8m 8H npws nmpmmnp eyes muoEHerm omega .mHSpwHoe gem pm Umpmehp N was epdpmflos RON pm Umpmmnp N .ehdpmwos mo pm empmenp N .mcoewoedm hemme> human 0 mo :oapwuwafinmpm peaches emmHebm may enummmpm xaom mo.mm mam mnemmeam vaom. “3.8. 93 mssmmmhm xwom mo.04 4 ES 9.. to .m gag 100% 9076 8023 70% 60% :1 O «4 +5 ,, m 5056 +3 0.) (1’. +5. 40% G) O :4 CD 9* 30% Figure 6. 21 Dip Soak Pressure Methods of Treatment The average percent polyethylene glycol retained after treatment, based on oven dry weight of untreated wood. 10Q3 90% 80% 70% 60% SL1 O *4 . -g 50» 0 +3 32 4033 .p a (D 2 g 3 O73 2 5 10% 0% Figure 7. 22 16.6% 25.4% 33.05 % Polyethylene Glycol in Water The average percent polyethylene glycol retained after treatment, based on oven dry weight of untreated wood. DISCUSSION The results of this study show that two of the factors that determine the effectiveness of stabiliz ng wood with polyethylene CHtLOH and the concentration of the H- glycol are methods of appl polyethylene glycol solution. This analysis indicated that between 30 and 40 percent sta- bilizati n of wood can be obtained by using at least 25 percent polyethylene glycol in the water solution and using an 8—hour soak or a 15-minute at 150 p.s.i. pressure treatment. The statistical analysis showed no significant difference between the average per— cent stabilization effectiveness between 33.0% and 25.43 solutions or between the 8-hour soak and the 15-minute pressure treatment. (Table V). There was significant interaction effect for methods and conditions. The soaking method gives much greater stabiliza- tion effectiveness than the pressure method when the veneer treated has 90 percent moisture. This indicates that the higher moisture in the veneer did not allow the polyethylene glycol to penetrate into the veneer in the 15-minute treating period. The stabilization effectiveness dropped off quickly when the dipping method or the 16.6% solution was used. Apparently the dipping method did not allow enough time for the polyethylene glycol to penetrate into the wood cells. The 16.6% solution appar— ently did not contain eno gh bulking agent to properly displace the water in the cell walls in the short treating period. There was no significant difference between averages of sta- bilization effectiveness for the moisture conditions tested. It is apparent fnam this study that the value of stabilizing wood with polyethylene glycol with these methods would depend on the value of 30 to 40 percent stabilization. Whether 30 to 40 percent stabilized wood will check and warp the same as untreated wood must be investigated. If this percent stabilization is not good enough, then the longer and more costly treatments may be used. It has been shown by Stamm (10) that 90 percent wood sta- bilization can be obtained with longer treating and drying schedules. The results also indicate that the effectiveness of the sta— bilization varies directly with the percent retention of the poly— ethylene glycol. This can be seen by comparing the percent retention graph (Figure 6 and 7) that corresponds with the percent effective- ness graph (Figure 3 and 4) for each treatment. This result could be expected since the more bulking agent retained in the wood cells the greater should be its effectiveness in holding the wood in its swelled condition. Color comparisons were made between specimens from the same sheet of veneer that had treatment and some that had no treatment. There is a slight darkening of the veneer as the chemical retention is increased. This is not a very distasteful color and should not be objectionable except for the very finest finish. Measured from oven dry to soaking wet the veneer that was treated swelled more than the veneer with no treatment (Table VI). This indicates that the more polyethylene glycol that is retained 25 in the veneer the greater the veneer will swell. This data also shows that at the same relative humidity and temperature wood treated with polyethylene glycol will obtain a higher equilibrium moisture content than wood with no treatment. As indicated in Table VI, wood at 90 percent relative humidity and 100° F. and retaining 20 percent polyethylene glycol had almost swelled to completely soaked dimensions. The seriousness of this change in the swelling characteristics of wood would depend on the moisture content range that the wood will be used. 26 CONCLUSIONS Several conclusions drawn from the results and analysis of this study are: l. 3. Pressure treating and soaking gave the best veneer stabilization effectiveness of the methods tested. The dipping method gave significantly less veneer stabilization effectiveness. Birch veneer stabilization is not affected signifi— cantly by the moisture content of the specimens for solutions tested in this study, except when pressure treating or dipping veneer which contains 90% moisture. There is no significant difference between treating with 33.0% or 25.4% Polyethylene glycol solutions. 16.6% solution produced significantly less birch veneer stabilization. Using 16.6%, 25.4% or 33.0% polyethylene glycol solutions, a pressure trea ment of l/20th inch birch veneers can be completed in 72 hours and a soaking treatment in 80 hours. The polyethylene glycol treatment does darken the wood. 27 APPENDIX dd .L'C) Table II. Two Way tables showing average percent stabilization effectiveness. ’ "“ "it‘t’c‘eatfsaists‘;Content of Specimens before Treatment Concentration of Solution, % 0 20 90 av. 33.0 37.7 42.9 23.5 38.0 25.4 39.9 35.1 28.6 34.6 16.6 15.3 14.2 26.0 18.5 average 31.0 30.7 29.4 Method Concentration . of Solution,§ Dip Soak Pressure Av. 33.0 12.2 46.9 55.1 38.1 25.4 7.3 52.6 43.4 34.6 16.6 4.6 31.3 19.5 18.5 Average 8.1 43.6 39.4 29 Table III. Two way table showing average percent stabilization effectiveness. Method Percent Moisture Content of Specimens Before Dip Soak Pressure AV. Ersamszlt O 8.9 39.9 44.2 31.0 20 15.3 30.2 46.6 30.7 90 O 60.6 27.5 29.4 Average 8.1 43.6 39.4 Table IV. Analysis of Variance Among Uethod, Condition and solution Degrees of Sum of flean Freedom Squares Square Total 53 26,927 method 2 13,592 6,796 Condition 2 27 13 Solution 2 3,914 1,957 M x C 4 4,877 1,219 M x S 4 1,681 420 C x s 4 1,128 282 H x C x S 8 1,189 148 Error 27 519 19 —-_ ..__..._.-.-—_- -.- Table V. Testing for significant difference of treatment averages with the studentized range test. Values with lines connecting them are not significantly different. Hethods Dip Soak Pressure Averages 8.1 43.6 39.4 Solutions 33.0» 25.4% 16.6% Averages 38.1 34.6 18.5 Conditions 0% 20% 90% Averages 31.0 30.7 __ m_ 29.4 Percent Stabilization Effectiveness 6O 5O 4O 3O 20 10 - 25 .473 ' 16.6% ' - 33 .035 0% 20% 90% Foisture Condition of Specinens When Treated 0. Interaction between concentrations and conditions. Percent Stabilization Effectiveness 60 5O 4O 3O 20 10 Figure 9. \\\\\\\\ Soak Pressure L Dip 33.0% 25.4% 16.6% Percent Polyethylene 1ycol in Cater Interaction between methods and concentrations. 60ft 50 401—» 30 T I 20 we L Dip Soak Jethods of Treating Figure 10. Interaction between moisture conditions and methods. 20;; 023 Pressure Table VI. The average distance a 10-inch, oven dry, non- treated specimen swelled when at the indicated condition. Heasurenent in inches. 90% Relative Humidity Water Soaked 1000 F. Temperature Condition Not Hot Hethod Treated Treated Treated Treated Dipoing .57 .3 .75 .75 SOE-il'ced .75 .38 07(9 .75 Pressure .75 .38 .80 .75 Average .69 .38 .78 .75 \) U1 Table VII. The percent stabilization for each specimen using the indicated condition, solution and method of treat me nt . ......-- .——. .——————- .———. . ———._-——. ...—fl--.“ . _ .. _ —-_.——..» - ...---_-_—..—- Moisture of Specimens when Concentratibn . ~~ - v- -— Treated, % of Solution,% Dip Soak Pressure 33.0 10.7 41.3 64.0 10.7 42.7 57.3 0 25.4 6.7 57.3 49.3 8.0 61.3 57.3 16.6 14.7 21.3 16.0 2.7 16.0 21.3 33.0 73.3 33.3 c8.0 {VUQ7 4103 62.7 16.0 45.3 44.0 16.6 0.7 6.7 29.3 4.0 10.7 28.0 33.0 0 65.3 40.0 0 57.3 33.7 90 25.4 0 50.7 32.0 0 57.3 32.0 16.6 0 56 0 18.7 0 77.3 12.0 Table VIII. he percent polyethylene y‘vcol reta'ted after ).A-'/ v' m l tre tin? with the indicated method, condition W —.. .. . .. ...-...— _—-——..-..—'-——_. Moisture of Specimens Concentration Uhen Treated,% of Solution, 5 D10 503k Pressure 33.0 l 0 1709 23.4.0 . 18.0 23.5 8.6 18.1 3 6 0 ’30 “...! g 53.0 12.1 15. 21.7 13.1 15. 21.4 To :75. 8.4 8. 17.9 9.5 18.5 18.3 18.6 5.2 5.3 10.5 509 5.6 005 33.0 5.2 24.1 19.1 3.8 22.9 17.5 90 25.4 7. 18.9 13.5 19.0 12.6 16.6 MN 0 \JQ 0 WC) l. "1 AL. 10. 11. 37 GRAPHY Brown, H. P., Panshin, n. J., and Forsaith, C. C. T3Ktbook of Hood Technology. Vol. I, New York: ncGra.v— —Hill Book Comyany, 1919. Di}:on, '“Iilfrid J o , 1368.1. CAUSE ', Fr: 11%. J 0 Introduction to Statistical analysis. Second Edition, New York; HcGraw-Hill Book Company, 1957. Forest Products Laboratory. Hood Handbook. agriculture H nriboo‘.f No. 72, mashington, D. 0.: United States Government Printing Office, 1955. 'h nsion 1 St.: .biliza tion Seminar No. 2145, January 21-23, 1959 --.—...- Hunt, George H., and Garrett, Georre L., ‘999 Preservati- New York: ”C6rzw-nrll Book CO‘banu’ 1053 Jane, F.W., The Structure of Wood. New York: The Hcdillan Company, 10 56. -, D. L., Sprou11,Reavis 0., Further Experiments 93 Din nsio n81 Sta b11150t1(n b‘ allylation. Vol. I, he. 1, ' b ——.....—.. gr, 1951. Forest Product sJournal. Stamm, alfred J., and Harris, Elwin B. Chemical Process of ..ood. New York: Chemical Publishing Company, Inc. 19 Stamm, Alfred J., Effect of Formaldehyde Treatment Upon the Dimensional Stabilization of Hood. Forest Products Journal. Vol. II, No. 2, Hay 1956. S amn, Alfred J. Dimensional Stabilization of hood with CarbOWaxes, Fores Products Journal, Vol. VI, Ho. 5, June 1953. Steam, Alfred J., The Effect of Polyethylene Glycol Tre eatnent Uoon the Dimensional Stability and 0 gher Properties of hood. K U. S.Forest Products Laboratory, 195' HICHIGRN STATE UNIV. LIBRGRIES 3129301 1032400