The wear resistance and Janka hardness of five United States hardwood species were evaluated for potential use in bridge decking and truck flooring. The species tested include ash (Fraxinus sp.), hickory (Carya sp.), red oak (Quercus sp.), sweetgum (Liquidambar styraciflua), and white oak (Quercus sp.). The specimens were prepared with the sizes of 1 by 2 by 4 inches (2.54 by 5.08 by 10.16cm) for abrasion test and 1 by 2 by 6 inches (2.54 by 5.08 by 15.24 cm) for Janka hardness testing. The specimens were cut from 30 individual parent boards of random width with clear sections for each species. The abrasion and Janka hardness tests were performed according to the American Society of Testing and Materials standards. All wear and hardness data were statistically analyzed by 1-way analysis of variance. The results of this study demonstrated that sweetgum with the lowest density had the greatest amount of thicknesses loss and thus lowest wear resistance. White oak was found to have the least thicknesses loss, thus highest wear resistance among the hardwood species tested. Hickory, with the highest density, had the highest hardness among the hardwood species tested, but it had relatively lower wear resistance comparing to ash, red oak, and white oak.
In many applications, wear resistance and hardness are tremendously important properties of wood and wood products. Wear resistance plays a crucial role in applications where high-volume foot or vehicular traffic in structures is observed, such as in the transportation industry for decking and in structures for bridges. Wear resistance and hardness tests are two methods that have been used extensively for measuring the resistance of wood floorings and other wood-based panel materials. Wear refers to the loss of material from the surface of a material by the mechanical process of rubbing the surface with abrasives. Abrasion is one of the actions that can cause wear. Wear and abrasion are often considered to be the same. Hardness is useful to determine directly how well a wood species withstands dents and dings, and to predict the resistance of a wood species in nailing, screwing, sanding, and sawing. These features (wear resistance and hardness) are primarily affected by wood density, temperature, and moisture (Janka 1906, V. Lorenz 1909, Ncube 2008). To some degree, wood anatomy is important as well, particularly in ring-porous hardwood where the specific gravity (SG) differences between earlywood (springwood) and latewood (summerwood) can be very large.
The direct relationship of wear resistance to relative density was reported by Franz and Hinken (1954). In their work specifically related to machining wood with abrasives, the scientists found that the relationship of woody fiber removed abrasively appeared to be associated to a large extent with species' density, as resistance to indentation is dependent primarily on this factor. In their work, less dense species were abraded more quickly. According to their research, relative wood density influenced, to the greatest degree, the extent of the penetration of the grit (abrasive) particles. The research also showed that wood at 12 percent moisture content (MC) abraded more quickly than that at 6 percent. In other research, a series of 20 species was rated based on their tendency to create “fuzz” (short bits of wood fiber that are attached to the board at one end and are free at the other) during sanding (Davis 1962). White oak, red oak, ash, and hickory were among the top five performers among woods that are currently available commercially. Experimentally, it was discovered that hardness is approximately proportional to the SG of wood (Janka 1906, v. Lorenz 1909). Kollmann and Cote (1968) reported on a series of research related to SG, hardness, and abrasion resistance. Newlin and Wilson (1919) experimentally discovered and reported a relationship between Janka hardness and SG. Janka (1906) and Janka and Hadek (1908 and 1915) proposed a modified Brinell hardness test for wood. In those works, the force required to completely embed a 0.444-inch-diameter (11.3-mm-diameter) steel hemisphere (which corresponds to 2 cm2 of surface area) into the specimen was determined. Janka tests have been standardized wherein they are conducted on sides and end of the specimen, with no distinction made for radial or tangential orientation (ASTM D143-14 [ASTM 2017b]).
A simple abrasion test that indicates the wear resistance of wood species would be of great value. In the 1940s, the US Forest Products Laboratory developed and evaluated the Navy wear-test machine for its feasibility in accessing the wear resistance of wood (Youngquist and Munthe 1948). The purpose of that work was to evaluate teak (Tectona grandis) and other wood species for use as decking on naval ships. That work resulted in the wear-resistance test and wear-resistance data for several species, currently listed in ASTM D2394-17 (ASTM 2017a). The results obtained from this wear-test machine support comparison and evaluation between new flooring materials and the wood species commonly used for flooring.
In this study, the wear resistance and hardness of five hardwood species (ash, hickory, red oak, sweetgum, and white oak) were evaluated for potential use in high-wear environments such as bridge decking and trailer flooring. Trailer and truck decking and flooring need to have appropriate levels of abrasion resistance, compression strength, biological durability, and flexural strength. Apitong (Dipterocarpus sp.), which is an imported tropical hardwood, can meet these property requirements. It has appropriate strength, high resistance to abrasion and decay, heavy thicknesses, and particularly clear pieces (Gerry 1952). The US military has been using apitong as flooring for its tactical trailers for several decades. However, this wood has become increasingly rare and many of its subspecies are critically endangered, making it unavailable for future use. Therefore, finding sustainable alternative materials has become a critical need. Among currently available options, US hardwoods offer the greatest potential for a sustainable and cost-effective material that can perform well in a wide range of environmental conditions. Therefore, the objective of this study was to evaluate the wear and surface hardness characteristics of five US hardwood species and rank the species for their suitability for bridge decking and trailer flooring applications. Based on previous research (Carmona et al. 2020 a, 2020b; Franca et al. 2021; Shmulsky et al. 2021), the authors did not suspect that these properties had changed over time; however, not all species have been previously investigated and these particular properties are not frequently investigated or reported.
Materials and Methods
Five species of hardwoods were selected for abrasion (also known as “wear”) and hardness tests. These five species were ash, hickory, red oak, sweetgum, and white oak. For each species, approximately 350 board feet (0.83 m3) of random width, 1-inch-thick (2.5-cm-thick), variable grade rough lumber was procured. Both kiln-dried (ash, hickory, and sweetgum) and green lumber (red and white oak) were received. The green lumber was air-dried to approximately 12 to 15 percent MC prior to processing. From the parent packs of lumber, approximately 30 individual parent boards were then selected from each group. By selecting material in this manner, each test specimen came from a unique parent board thereby capturing as much variability as possible. A clear section approximately 30 inches (76 cm) long was then removed from each parent board. From this section, test specimens were prepared. For preparation, first the 30-inch-long (76-cm) sections were skim planed on two faces and jointed along one edge. The sections were then ripped to yield 2-inch-wide (5-cm) clear strips. After the strips were ripped, a 6-inch-long (15.24-cm) hardness specimen was cut from each. This action resulted in a hardness specimen approximately 1 by 2 by 6 inches3 (2.54 by 5 by 15.24 cm3). Next, the remaining 24-inch-long (61-cm) strips were replaned to 0.75-inch (1.90-cm) thickness. Clear 4-inch-long (10.16-cm) abrasion specimens were then cut from the 0.75-inch thick, 2-inch-wide (1.90 by 5-cm) strips. Abrasion specimens were then prepared per ASTM D2394-17 (ASTM 2017a) wherein 0.5-inch-wide (1.27-cm) rabbet cuts were made into the ends of the specimens to facilitate mounting on the abrasion tester (Fig. 1). This action left a 2 by 3-inch (5 by 7.62-cm) face to be abraded. Next, 4-inch-long (10.16-cm) MC and SG specimens were cut from the strips. Each MC/SG specimen was thus approximately 0.75 by 2 by 4 inches3 (1.90 by 5 by 10.16 cm3). All specimens (abrasion, hardness, MC, and SG) were then acclimated in a 12 percent MC environmental chamber at 70°F (21°C) and 65 percent relative humidity for a minimum of 2 weeks. Wear and hardness tests were done with random or noncontrolled orientation with respect to tangential and radial directions. In this manner, it was anticipated that the variation of both orientations, combined, was captured.
Following MC acclimation, abrasion specimens were tested on a navy-type wear tester according to ASTM D2394-17(ASTM 2017a; Fig. 2). Briefly, for this testing, each specimen was mounted on a plate that rotated at 32 ½ revolutions per minute (RPM) with a 10-pound (4.5-kg) weight mounted above and thereby applying downward pressure. This mounting plate was elevated off and then immediately returned to the abrading plate, via cam followers, 1/16 inch (1.6 mm) twice per rotation. The abrading plate rotated in the same direction as the specimen mounting plate at a rate of 23 ½ RPM. The abrading plate had a constant flow of new 80 grit aluminum oxide media applied for the duration of the test. Specimen thickness was measured at five locations (i.e., the four corners and the center) before testing and then at 100-rotation intervals for the duration of the test. This process was repeated until each specimen had undergone 500 rotations. In Figure 2, the machine has guards installed over its gear and chain works. One of the clamps used to affix the specimen to the mounting plate is shown in the forefront. The fixture that holds the 10-pound weight and mounts at the top of the shaft to which the specimen becomes affixed has been removed and is not shown.
Hardness specimens were tested according to ASTM D143-14 (ASTM 2017b). This test was performed by penetrating the surface of the specimen with a 0.444 in diameter (11.3-mm-diameter) steel ball to a depth of 0.222 inches (5.63 mm) at a rate of 0.25 inches per minute (6.35 mm per minute). The specimens in this study were tested once on each end and once on each wide face, for a total of 4 penetrations per specimen. The average force required for the two end penetrations was used for analysis of end hardness. The average force required for the two wide face penetrations was used for analysis of face hardness.
Moisture content, density, and SG measurements
The experimental design was completely randomized. All abrasion and hardness data were analyzed by 1-way analysis of variance (ANOVA) using the procedure for general linear mixed models (PROC GLIMMIX) of SAS 9.4 (SAS Institute, 2013). Differences were considered significant with a P value less than or equal to 0.05. Summary statistics for abrasion and hardness are all reported. Additionally, the statistical model represented by 1-way ANOVA is Yi = μ + Ti + Ei, where μ is the population mean, Ti is the effect of different species (T = 1 to 5), and Ei is the residual error.
Results and Discussion
The statistical summaries for MC as percent dry basis, density, and SG are shown in Table 1. According to the results, all specimens had a MC between 12 and 16 percent at the time of testing. Among all five hardwood species, hickory had the highest density (0.029 lb/in3 or 0.80 g/cm3) and SG (0.71), and sweet gum had the lowest density (0.022 lb/in3 or 0.602 g/cm3) and SG (0.54); white oak, red oak and ash fell between.
The abrasion summary statistics are illustrated in Table 2 as the thickness loss based on 500 revolutions of the abrading disk. Among all five species tested, sweetgum, with the lowest density, had the most amount of wear (0.0092 in or 0.234 mm in thickness loss), which is expected and consistent with previous reports (Youngquist and Munthe 1948, Franz and Hinken 1954). It means that pieces with lower density and higher MC are abraded more quickly. However, hickory, with the highest density, did not yield the least amount of wear as we would expect; instead, it had a thickness loss of (0.0084 in or 0.212 mm), second highest among the five species tested. White oak, on the other hand, was found to have the least amount of wear (0.0058 in or or 0.147 mm in thickness loss), indicating highest abrasion resistance. Table 3 shows the relative order of the abrasion resistance of five hardwood species, from the most amount of wear to least amount of wear: sweetgum, hickory, ash, red oak, and white oak. According to the statistical ANOVA results, there were statistically significant differences in thickness loss (P < 0.0001) among all five hardwood species (Table 3). The abrasion test data of this study indicated that no definite relationship exists between abrasion resistance and wood density within the hardwood species tested. The other observation we made in this study is that the abrasion test data of sweetgum exhibited a coefficient of variation of 42.6 percent, much higher than other species (23.2% to 26.2%). We are not clear about what caused this unusually high variation in wear resistance in sweetgum, but the growth characteristics or structure variation may be a contributing factor.
Summary statistics of Janka hardness test results (both face and end hardness) are also demonstrated in Table 4. The ANOVA results of Janka hardness showed that there were statistically significant differences in both face and end hardness (P < 0.0001) among all five hardwood species tested (Tables 5 and 6). The SG has the most effective impact on the hardness of wood species (Panshin and de Zeeuw 1980). It is well documented that there is a linear relationship between hardness and density (Ylinen 1943, Miyajima 1963, Kollmann and Cote 1968, Holmberg 2000). The results of Janka hardness and SG tests in the current research proved that there is a direct relation between hardness and SG. Hickory, with the highest SG, had the highest level of hardness in both face and end with the values of 2030 lbf and 2020 lbf (9,031 and 8,986 N) while sweet gum, with the lowest SG, had the lowest hardness in both face and end with the values of 976.72 lbf and 1323 lbf (4,345 and 5,886 N). Ash, red oak, and white oak had the moderate face and end Janka hardness results that were close to one another.
Investigation of the abrasion resistance and Janka hardness of five US hardwood species indicated that white oak had the least thicknesses loss, thus highest wear resistance among the five hardwood species tested. It is surmised that white oak's higher density, as compared to red oak, accounted for its superior performance. Sweetgum, with the lowest density, had the greatest thicknesses loss and thus lowest wear resistance. Though not compared statistically, the side/face hardness results for these species appear similar to or greater than those presented in the Wood Handbook (Forest Products Laboratory 2010). Hickory, with the highest density, had the highest hardness among the hardwood species tested, but it had relative higher thickness loss, thus lower wear resistance comparing to ash, red oak, and white oak. Hickory, with the highest SG and acceptable changes in thickness loss, had the best abrasion resistance. Hickory can be a potential candidate for use in the trailer flooring and truck decking.
This publication is a contribution of the Forest and Wildlife Research Center, Mississippi State University. This study was conducted through a cooperative research agreement (FS 20-JV-11111133-032) between Mississippi State University and USDA Forest Service, Forest Products Laboratory, and partially funded by USDA Forest Service, Forest Products Laboratory. The authors also acknowledge the US Endowment for Forests and Communities for its contributions to this research.
The authors are, respectively, Postdoctoral Associate (firstname.lastname@example.org), Department Head and Warren S. Thompson Professor (email@example.com), and Graduate Students (firstname.lastname@example.org [corresponding author], email@example.com), Department of Sustainable Bioproducts, Mississippi State University, Mississippi State; Research Forester, USDA Forest Service, Northern Research Station, Starkville, Mississippi(firstname.lastname@example.org); and Supervisory Research Gen. Engineer (Robert.email@example.com) and Research Forest Products Technologist (Xiping.firstname.lastname@example.org), USDA Forest Service Forest Products Laboratory, Madison, Wisconsin. This paper was received for publication in December 2021. Article no. 21-00074.
*This article is part of a series of selected articles addressing a theme of bringing academia, industry, and government entities together to work on innovation and applied technologies. The research reported in these articles was presented at the PTF BPI Conference, held on November 1–3, 2021, in St. Simons Island, Georgia. All articles are published in this issue of the Forest Products Journal.