Printer Friendly

Properties of hot-melt adhesives used in kitchen cabinets.

Abstract

Hot-melt adhesives are now used for attaching the back panel to the sides, top, and bottom parts of kitchen and bath cabinets. Despite the increasing use of hot-melt adhesives, little is known about their strength performance under the various conditions to which they are subjected during product transportation and service. This study evaluated the shear, tensile, and bending strength of different joint types with three hot-melt adhesives. The study also evaluated the effect of temperature on joints made with hot-melt adhesives, the contribution of staples to the strength of adhesive joints, and the effect of using excess adhesive on joint strength. There was no significant difference between hot-melt adhesives when tested under shear, tensile, and bending loading conditions with one exception. In the staple test, ethyl vinyl acetate performed the best. Dado joints performed best in bending but worst in tensile strength. Butt joints were best in tensile loading. The strength of hot-melts generally decreases as temperature increases. Plastic failure occurs at elevated temperatures while brittle failure occurs at lower temperatures. Joints with straight-driven staples had a better performance in bending than joints with staples driven at an angle. There was no significant difference between the joints with staples and those without staples. The use of excess adhesive provided no significant increase in strength. Statistical selection models showed that the optimal combination of factors based on the joint study was: dado joint, ethylene vinyl acetate adhesive, no staples, and room temperature conditions.

**********

Adhesives have been used to bond wood together since at least the time of the ancient Egyptians. Animal glues were the adhesives used until the development of casein glues for wood aircraft frames during World War II. Synthetic resins such as urea-formaldehyde (UF) and polyvinyl acetate (PVA) were developed subsequently and remain in common use. These adhesives all function by forming a very thin layer or glueline which bonds the wood surfaces together.

Hot-melt adhesives were first introduced to the furniture industry in the 1950s. Recently, they have found applications in the assembly process, particularly in attaching the back panel to the top, bottom, and side panels of kitchen cabinets. The adhesive is applied as a "bead" at temperatures of at least 300[degrees]F and upon cooling develops maximum strength. Typical open time is about 8 to 10 seconds. For all practical purposes, clamps and curing times are eliminated. This study focuses on the strength of hot-melt adhesive joints.

The Wood Handbook (USDA 1999) provides a good overview of the different types of adhesives, their uses, and how different wood characteristics and processing methods affect their performance with particular emphasis on wood-frame construction.

Bogner et al. (1997) carried out shear tests to determine temperature resistance and durability of hot-melt adhesives. The temperature used ranged from 14[degrees]F to 167[degrees]F, relative humidity (RH) ranged from 30 to 95 percent, and exposure time ranged from 1 to 5 hours. Bogner's tests do not conform to any particular standard. The polyolefin and polyurethane hot-melt adhesives that were tested had the greatest resistance to temperature change, prolonged temperature, and humidity changes. The ethylene vinyl acetate (EVA) and polyamide adhesives decreased in strength with prolonged temperature and humidity elevation.

Information from the kitchen cabinet industry indicates that EVA and polyamide are the two most commonly used hot-melt adhesives. Concern exists over their performance at reduced and elevated temperatures, however, because cold temperatures in the 0[degrees]F range and moderately elevated temperatures of 120[degrees]F are encountered during shipment, storage, and use.

Objectives

The overall objective of this study was to evaluate the strength and performance of hot-melt adhesives for joints in wood furniture and cabinet manufacturing. Because of the complexity of the issues, several smaller studies were conducted.

The first study evaluated the shear strength of three different hot-melt adhesives exposed to temperatures of -4[degrees]F through 120[degrees]F.

The second study looked at the tensile strength of three different hot-melt adhesives, three joint types, and the use of excess adhesive.

The third, and most extensive study, determined the bending strength of four joint types constructed with three different hot-melt adhesives along with the effect of staples and climatic cycling.

Materials and methods

The factors evaluated in this study include shear, tensile, and bending strength of different joint configurations made from particleboard. Using three different hot-melt adhesives, the effect of climate cycling, the use of excessive adhesive, and the use of staples were evaluated.

All joints were constructed from M-3 industrial-grade southern yellow pine particleboard. The stock was conditioned at 73[degrees]F and 35 percent RH for several weeks before testing and use. The panels were tested for density, modulus of elasticity (MOE), modulus of rupture (MOR), internal bond (IB), face and edge screw withdrawal, density, and moisture content (MC) in accordance with ASTM D1037 (ASTM 2001). The surface bond strength was determined in accordance with ASTM D5651 (ASTM 2002a). The results for density, MOE, IB, and screw withdrawal from the face and edge all exceeded the standard given for M-3 grade particleboard. MOR was 12 percent below the cited value for the M-3 grade. Surface bond was 277 psi.

Three different hot-melt adhesives were used: EVA, polyamide, and a proprietary product. Their properties, as provided by the supplier, are given in Table 1.

The proprietary product had a longer open time which is significant for the assembly process. The hot-melts were applied in accordance with supplier recommendations using a Reka Tr50.3.333 extrusion gun. The amount of adhesive did not vary, because the gun used (Reka Tr50.3.33) at high temperatures functions to transform the solid form of the adherent to a liquid before dispersing it under pressure. The application pressure used for this study was 80 psi with a maximum application of glue per minute of 330 mL.

Shear strength tests

ASTM D905 (ASTM 2002b) provides a detailed description of the shear test procedure and specimen size. This procedure was followed with the exception of specimen size. The standard procedure calls for a specimen with a shear area of 1-3/4 by 2 inches. The specimens used for this evaluation were constructed and tested according to the ASTM D905 standard for shear specimens except for the fact that the bonded area was 2 inches wide by 3/4 inch along the machine direction of the board. Thus, the total surface area loaded in shear was 1.5 in. (2) as compared to 3 in. (2) for the standard specimens.

Hot-melts react differently than other standard adhesives. There is a very limited open time, so clamping is not possible. The adhesives were applied to both surfaces to be bonded and held in place by hand pressure. A t-test analysis comparing results for standard size shear specimens to the size used in this study showed that the smaller sized specimens gave more consistent results at the 0.05 significance level. It simply took too long to uniformly apply the hot-melt to the larger sized standard specimen and then bring both surfaces into contact.

For the shear tests, 75 specimens were prepared representing three hot-melt adhesives, five temperature exposures, and five replications. The temperatures to which the specimens were exposed were -4[degrees]F, 73.4[degrees]F, 86[degrees]F, 104[degrees]F, and 120[degrees]F. The temperature conditions were an industrial recommendation based on temperatures to which the joints are subjected during transportation and when in service. Each sample was randomly subjected to one of the above temperature conditions for 48 hours and then left in laboratory ambient conditions of 73[degrees]F and 35 percent RH for 1 hour before testing using standard procedures. The interior of the specimens was still warm when tested. A 3 x 5 full factorial experimental model with five replications was used to evaluate the data.

Tensile strength tests

This test evaluated the performance of three different joint configurations and the three different adhesives. The three joint configurations included a face to edge dado joint, a face to edge butt joint, and an edge to edge butt joint as shown in Figure 1. The adhesive was applied uniformly to both mating surfaces and hand pressure was applied immediately.

[FIGURE 1 OMITTED]

Use of excess adhesives was also evaluated during this test. A second set of joints was prepared in the same manner as those discussed above but with additional adhesives applied as shown in Figure 2. The specimens were tested as shown in Figure 3. The loading rate was 0.05 in./min of crosshead travel.

[FIGURES 2-3 OMITTED]

The tensile strength for the joints tested was calculated as the force applied at failure per unit of cross-sectional area of the joint. The cross-sectional area was taken to be the summation of total area of contact for the joint members. For the face to edge and edge to edge butt joints, the cross-sectional area was calculated as the area of the end specimen, which was 1/2 inch wide by 4-1/2 inches long. For the face to edge dado joint, the area was calculated as the cross-sectional area of the side member (1/2 by 4-1/2 in.) plus the area of the side member which contacted the back member in the dado times two (2 1/4 by 4-1/2).

A 3 x 3 x 2 full factorial experimental model with five replications per cell was used for the analysis.

Bending strength tests

This test evaluated the bending strength of four joint types and the three hot-melt adhesives. In addition, the effect of staples and temperature and humidity cycling on joint strength was evaluated. The joint types were L-shaped and named rabbet 1, rabbet 2, dado, and butt joint as shown in Figure 4.

[FIGURE 4 OMITTED]

The adhesive was applied to the mating surfaces of both joint members followed by hand pressure.

To evaluate the effect of staples on the bending strength of the joints described above, a set of 240 specimens was prepared. Staples with a leg length of I inch and a crown width of 3/8 inch were used. The staples were driven into the specimen immediately after gluing with an air staple gun using a pressure of 60 to 70 psi. Two staples spaced 2-3/4 inches apart at equal distances from the edges of the specimen were used. Based on the industrial application and joint configuration, staples for dado and butt joints were applied at an angle while those for rabbet 1 and rabbet 2 were applied at 90[degrees] to the back member.

In addition, the effects of temperature and humidity cycling were also evaluated. Two tests with 180 specimens each were used. The first test was based on ASTM D1183 (ASTM 2002c) standard test methods for resistance of adhesives to cyclic laboratory aging conditions and the second was in response to an industry recommendation called the "extreme condition test." ASTM D1183 standard for interior joints was used in this study because it takes into consideration the effect of both temperature and RH. The test conditions are summarized in Table 2.

All bending tests were carried out on a universal testing machine as shown in Figure 5. The back member of the specimen was bolted to the jig, which was in turn fastened to the machine bed. Vertical loads were applied to the side member at a point 6 inches from the mid-point of the joint. The rate of loading was 0.1 in./min of load head travel.

[FIGURE 5 OMITTED]

A 4 x 3 x 2 (four joint types, three hot-melts, and two staples conditions, present or absent) full factorial experimental model with 10 replications per cell was used for evaluation with three hot-melt adhesives and the presence or absence of staples as the main factor effects. A Tukey's Studentized multiple comparison test (Neter et al. 1996) was carried out after the analysis of variance (ANOVA) to determine whether there was any significant difference within each of the main factor effects.

A 4 x 3 x 3 full factorial experimental model with 10 replications per cell was used to evaluate the main factor effects of four joint configurations, three hot-melt adhesives, and temperature and humidity cycling. A Dunnett's t-test (SAS 2001) was used for comparison of the two climatic factors to the control. For hot-melt adhesives and joint configuration, a Tukey's Studentized Range test (Neter et al. 1996) was used to determine whether there was any significant difference within each of the main factors after ANOVA.

Results

Shear strength test

The average shear forces at different temperature levels for the three adhesive types are shown in Table 3. The proprietary product had the highest shear values at temperatures of -4[degrees]F and 73.4[degrees]F while the EVA had the highest values at temperatures of 86[degrees]F and higher. In Table 3, it can be seen that joint strength generally decreased as temperatures increased from -4[degrees]F to 120[degrees]F. Combining all of the temperature levels, there was no statistically significant difference at the 0.05 level between the three hot-melt adhesives. ANOVA showed that hot-melt adhesive type and temperature factors were significant at the 0.05 level with a coefficient of correlation of 8.5 percent. Tukey's Studentized Range test showed that shear strength for temperatures of 86[degrees]F and 104[degrees]F, regardless of hot-melt type, were not significantly different.

Two types of joint failures were observed based on the different temperature conditions. Both types of failures were in the adhesive layer, therefore, the results probably represent the shear strength of the hot-melt adhesive itself.

Tensile strength test

The mean tensile strength in pounds per square inch of contact area for the three joint types and three hot-melt types is shown in Table 4. The results show that the face to edge dado joint had lower average values for all adhesive types than the face to edge and edge to edge butt joints.

An ANOVA test with joint types, hot-melt adhesives, and amount of adhesive as the main factors was carried out to determine the contribution of each factor to the tensile strength of the joints. Results showed that joint type, hot-melt adhesive, and amount of adhesive were all significant at the 0.05 level. Besides the difference in adhesive used for the one-level interaction, joint type was the only significant factor at the 0.05 level. The joint type factor accounted for 99.68 percent of the total sums of squares. These results indicated that although other factors were significant at the 0.05 level, the type of joint used was a major contributor to the tensile strength of the joint.

Since main factor effects were significant in the ANOVA, a pairwise comparison between joint type, hot-melt adhesive, and amount of adhesive by means of the Tukey procedure at a significance level of 0.05 was required to determine the individual influence on the strength of the joint. For the three joint configurations, the results showed that face to edge butt joints and edge to edge butt joints were not significantly different at the 0.05 level. The results also showed that the mean tensile strength value of face to edge dado joints was significantly lower than that of face to edge and edge to edge butt joints.

A comparison study of the three hot-melt adhesives showed that there was no significant difference between their tensile strength at the 0.05 level and that their performance under tensile loading was comparable.

The tensile force in pounds for the three types of joints and three hot-melt adhesives was also evaluated statistically. The results were identical to those obtained from the tensile strength in pounds per square inch.

For the amount of adhesive used in joint construction, a comparison study showed that there was no significant difference at the 0.05 level between the use of excess amounts of adhesive and the normal amount.

Three types of joint failure were observed (Fig. 6). For the face to edge butt joint, the hot-melt adhesive simply pulled particles loose from the face of the board. Thus, the surface bond strength of the particleboard was probably less than the tensile strength of the hot-melt adhesive. In the edge to edge butt joint, the adhesive pulled particles loose mostly from the core of the adjacent material where density was at a minimum. Near the surface of the particleboard where density increases, the adhesive simply pulled loose from the board. The face to edge dado joint failed due to the reduction in cross-sectional area in the back member. During testing of this joint, bending failure occurred because of the 1.5-inch-wide space between the supports (Fig. 3a). No bending was observed in the other two joint types.

[FIGURE 6 OMITTED]

Bending strength test

The average bending strength values of four joint types, three hot-melt adhesives, and presence or absence of staples are given in Table 5. Test results for joint type, hot-melt adhesive, presence or absence of staples, and one-level and two-level interaction factors with 240 observations are presented in Table 6. From the results of the F-test, the main effects of joint types, hot-melt adhesives, and staples were each significant at the 0.05 level. The two-level interaction of hot-melt and joint types and the interaction of staples and joint types also had a significant effect at the 0.05 level on the bending strength of the adhesive joints. Since main factor effects were significant in the ANOVA, pairwise comparison between hot-melt adhesives, joint types, and staples by means of the Tukey procedure, at a significance level of 0.05, was required to determine their individual influence on the strength of the joint. For the hot-melt adhesives, the results showed that EVA had the best bending strength performance of the three adhesives considering all joint types. The proprietary product and polyamide hot-melt adhesives were not significantly different at the 0.05 level.

Joint type comparison results at a significance level of 0.05 showed that the dado joint had the best performance followed by the butt joint. Rabbet 1 and rabbet 2 joints were the weakest but not significantly different from each other. The comparison between presence and absence of staples considering all joints showed that between the two there was no significant difference at the 0.05 level. To understand the effect of staple direction on the strength of the adhesive joint, a comparison study of all joints with staples driven straight and at an angle was carried out. Results showed that joints with staples driven straight had a better performance compared to the joints with staples driven at an angle. Another comparison study including staples driven straight and all joints without staples showed that joints with staples driven straight had a better performance than joints without staples. The results also showed that joints with staples driven at an angle were not significantly different from joints without staples at the 0.05 level. From these results, it can be noted that staple direction has a major effect on the strength of the joint, mainly because of the depth of penetration created by the angles to which the staples are driven, or because the force is not perpendicular to the length of the staple.

The ANOVA results for the effect of climatic cycling on adhesive joint strength are presented in Table 7. From the results, the main effects of climatic cycling, joint type, and hot-melt adhesive were each significant at the 0.05 level. The two-level and three-level interaction of climatic cycling tests, joint types, and hot-melt adhesives were also significant. A pairwise comparison of hot-melt adhesives and joint types by means of the Tukey procedure, using a significance level of 0.05, showed that the EVA adhesive had the best strength performance followed by the proprietary product. For both the extreme and ASTM D1183 climatic cycling tests, EVA had the best performance as compared to the proprietary and polyamide hot-melt adhesives. Under the control test, polyamide adhesive performed the best compared to the other two hot-melt adhesives.

Three types of joint failures were observed in the bending specimens. Figure 7(a) shows failure in the reduced cross section of the side member of the dado joint. Fifty-three percent of the dado joints failed in this manner. Figure 7(b) shows separation of the end of the side member for a dado joint. This type of failure occurred in those joints subjected to climatic cycling and was probably in response to the reduced strength of the hot-melt adhesives at elevated temperatures. Figure 7(c) shows failure of the adhesive after climatic cycling. This type of failure occurred in 96 percent of the rabbet 1, rabbet 2, and butt joints.

[FIGURE 7 OMITTED]

Conclusions

In shear, two of the adhesives performed differently depending on temperature exposure. The strength of hot-melt adhesives generally decreases as temperature increase. An elastic failure occurs in the adhesive at elevated temperatures, while a brittle failure occurs at lower temperatures.

In tensile loading, the joint configuration was significant, but the performance of none of the three adhesives tested was superior. In the face to edge and end to end butt joints, failure occurred by small particles being pulled loose from the face of the board. In the face to edge dado joint, failure was due to a reduced cross section in the back member.

In bending, the EVA adhesive performed the best. The dado joint out-performed the other three types. The use of staples did not improve joint strength. When staples are driven straight, the bending strength is improved compared to those driven at an angle. Elevated temperatures resulted in adhesive failure as compared to failure in the board itself.

Based on the results of shear, tensile, and bending strength tests, a selection of factors which will achieve optimal performance based on the test conditions was done using several selection models: Cp criterion, adjusted coefficient of determination ([r.sup.2]), and prediction sum of squares (Neter et al. 1996). All of the selection models showed that the best combination of factors were dado joints, EVA adhesive, no staples, and room temperature conditions.

Literature cited

American Society for Testing and Materials (ASTM). 2001. Standard test method for evaluating properties of wood based fiber and particle panel materials. D1037-93. Vol. 04.10. ASTM, West Conshohocken, PA.

--. 2002a. Standard test method for surface bond strength of wood based fiber and particle panel materials. D5651-95a. Vol. 04.10. ASTM, West Conshohocken, PA.

--. 2002b Strength properties of adhesive bonds in shear by compression loading. D905-98. Vol. 15.06. ASTM, West Conshohocken, PA.

--. 2002c Standard test methods for resistance of adhesives to cyclic laboratory aging conditions. D1183-96. Vol. 15.06. ASTM, West Conshohocken, PA.

Bogner, A., B. Ljulkjka, and I. Grbac. 1997. Method for testing the resistance of wood hot-melt adhesives to temperature changes and weathering. Drvna Ind. 47:108-113.

Neter, J., M.H. Kutner, C.J. Nachtsheim, and W. Wasserman. 1996. Applied Linear Statistical Models. 4th ed. Irwin, Chicago, IL. pp. 341345.

SAS. 2001. SAS/ETS[R] Software: Changes and Enhancements. Release 8.2, SAS Inst., Inc., Cary, NC.

USDA Forest Service, Forest Products Laboratory (USDA). 1999. Wood Handbook: Wood as an Engineering Material. Forest Products Society, Madison, WI. pp. 1-23.

Silas Tora *

Daniel Cassen *

Eva Haviarova *

The authors are, respectively, Graduate Research Assistant, Professor, and Assistant Professor (stora@purdue.edu; dcassens@purdue.edu; ehaviar@purdue.edu), Dept. of Forestry and Natural Resources, Purdue Univ., West Lafayette, IN. The authors wish to acknowledge the Seemac Graduate Fellowship Award for financial support. They also acknowledge Dr. Carl Eckelman, Professor, Dept. of Forestry and Natural Resources for his technical contribution. This paper was received for publication in March 2005. Article No. 10017.

* Forest Products Society Member.

[c] Forest Products Society 2006. Forest Prod. J. 56(11/12):43-50.
Table 1.--Properties and characteristics of the hot-melt adhesives
used in this study.

 Hot-melt adhesives

Properties and characteristics EVA Proprietary product

Color White White water
Application temperature 325 to 350 350 to 375
 ([degrees]F)
Ring and hall softening point 180 183
 (degreesF)
Brookfield viscosity at 6,000 to 8,000 2,700
 350[degrees]F (cPs) (a)
Storage Less than 1 year at T < 100[degrees]F
Set time(s) 1 to 2 1 to 2
Open time (b) (s) 7 to 8 10

 Hot-melt adhesives

Properties and characteristics Polyamide

Color Yellow
Application temperature 338 to 374
 ([degrees]F)
Ring and hall softening point 275 to 284
 (degreesF)
Brookfield viscosity at 13,000
 350[degrees]F (cPs) (a)
Storage 3 years at T < 58[degrees]
 to 76 [degrees]F
Set time(s) 1 to 2
Open time (b) (s) 6 to 8

(a) cPs = centipoises, 1 lb/ft s = 1488.2; cPs = 1.4882 Pa S

(b) ASTM D4497

Table 2.--Time, temperature, and RH for the two cycling
tests used in the study.

Climatic
cyclic test Equipment Time Temperature RH
 (h) ([degrees]F) (%)

ASTM D1183 EC (a) 24 73.40 85
 EC 24 120 25
 EC 72 73.40 85
 EC 48 120 25
Extreme condition (b) Freezer 24 -4 N/A
 Oven 24 120 N/A
 Freezer 24 -4 N/A
 Oven 24 120 N/A
 Freezer 24 -4 N/A
Conditioning (c) Room 1 74.30 50

(a) EC = Environmental chamber.

(b) Extreme condition not based on standard.

(c) Conditioning for both ASTM D1183 and extreme condition.

Table 3.--Average shear force of three different hot-melt adhesives
subjected to different temperature conditions.

 Hot-melt adhesive

Temperature Polyamide EVA Proprietary product

([degrees]F) lb

-4 674 (0.54) (a) 610 (1.08) 782 (0.77)
73 (b) 506 (0.83) 446 (l.21) 628 (0.54)
86 399 (0.80) 458 (0.41) 441 (0.66)
104 380 (0.38) 441 (0.12) 383 (0.40)
120 339 (0.38) 432 (1.30) 371 (2.64)

(a) Values in parentheses are coefficient of variation.

(b) Room temperature condition used as the control sample.

Table 4.--Mean tensile strength in pounds per square inch
of contact area for the three hot-melt adhesives and the
three joint configurations.

 Average tensile strength
Joint
configuration Polyamide EVA Proprietary product

 (lb/[in.sup.2])

Face to edge butt 319 (5.24) (a) 316 (7.31) 329 (5.77)
Edge to edge butt 339 (3.89) 330 (3.52) 330 (4.70)
Face to edge dado 64 (35.16) 84 (37.02) 77 (35.32)

(a) Values in parentheses are coefficient of variation.

Table 5.--Average bending strength in pound inches of four joint
types, three hot-melt adhesives, and presence or absence of staples.

 With staples

Joint Proprietary
type Polyamide EVA product

 (lb in)

Rabbet 1 915 (4.5) (a) 938 (4.5) 940 (3.7)
Rabbet 2 898 (2.1) 1,021 (6.3) 937 (4.8)
Dado 1,171 (4.8) 1,174 (4.2) 1,173 (4.3)
Butt 1,100 (6.3) 1,093 (3.9) 1,073 (2.6)

 Without staples

Joint Proprietary
type Polyamide EVA Product

 (lb in)

Rabbet 1 915 (6.2) 940 (4.0) 892 (3.7)
Rabbet 2 953 (1.9) 1,074 (2.0) 943 (3.6)
Dado 1,168 (2.7) 1,022 (4.3) 1,153 (1.7)
Butt 1,168 (5.2) 1,093 (3.7) 1,006 (4.9)

(a) Values in parentheses are coefficient of variation.

Table 6.--Reduced ANOVA model test results with joint type, hot-melt
adhesives, staples, and the one-level interaction effect of hot-melt
and joint type and staples and joint type.

Factor DF Mean square F-value p-value

Joint type 3 461,567,254 115.62 <0.0001
Hot-melt 2 117,545,478 29.44 <0.0001
Staples 1 100,025,334 25.06 <0.0001
H-J (a) 6 170,925,590 42.86 <0.0001
S-J (b) 3 19,726,753 4.94 0.0024

(a) H-J = hot-melt adhesives and joint type interaction term.

(b) S-J = staple effect and joint type interaction term.

Table 7.--A full ANOVA model test results for the climatic cycling
effect.

Factor DF Mean square F-value p-value

Hot-melt 2 228,552.00 110.5 <0.0001
Joint configuration 3 3,398,515.87 1643.07 <0.0001
Climate 2 422,272.90 204.15 <0.0001
H-J (a) 6 166,914.67 80.7 <0.0001
H-C (b) 4 360,784.30 174.43 <0.0001
C-J (c) 6 225,539.17 109.04 <0.0001
H-C-J (d) 12 109,287.37 52.84 <0.0001

(a) H-J = hot-melt adhesives and joints configuration interaction term.

(b) H-C- hot-melt adhesive and climatic condition interaction term.

(c) C-J = climatic condition and joint configuration interaction term.

(d) H-C-J = second level interaction term with the three main factors.
COPYRIGHT 2006 Forest Products Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

 
Article Details
Printer friendly Cite/link Email Feedback
Author:Tora, Silas; Cassen, Daniel; Haviarova, Eva
Publication:Forest Products Journal
Geographic Code:1USA
Date:Nov 1, 2006
Words:4788
Previous Article:Innovation in China's furniture industry.
Next Article:Performance of stock kitchen cabinets and laboratory-produced cabinets based on joint tests.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters