Ding et al. J Wood Sci
(2020) 66:55
Page 2 of 9
the anisotropic structure of wood. In this case, SA alone can barely present the morphological characteristics of wood dust. Besides, it is hard for the elongated or fibrous wood particles to fall through the sieves. The sieving efficiency, the percentage of the particles that can prop- erly fall through the sieves according to their width, was reported around 70% [21]. In some studies, IA was sug- gested as an alternative method or a combination of SA and IA was applied to get a more comprehensive under- standing of particle morphology [24, 25]. Once consid- ered time consuming, IA systems are now capable of handling a large quantity of particles and presenting the statistical results instantaneously. The major advantage of IA is that, besides size distribution, it can give quan- titative particle shape distribution. In this study, 2 IA technologies, i.e., scanning electron microscopy (SEM) and flatbed scanning image analysis, were applied. The former was used as a qualitative description method and the latter provided quantitative analysis. The results were compared with those of the SA to evaluate the robustness of the technologies. Experimental Materials MDF sanding dust was taken from a MDF mill in Jiangsu Province, China. The main panel constituents were hybrid poplar ( Populus sp.) fibers and UF resin. The panel sanding line was composed of 3 wide belt sanding machines. Five types of sanding belts were mounted and the grit sizes were P36, P80, P120, P150 and P180 from the entry to the outlet of the sanding line. MDF panels were fed at a speed of 55 m/min and sanded at a speed of 1460 rpm. The dust emitted during the sanding process was collected by a dust collecting system and stored in a silo where the dust was sampled for the experiments. The moisture content of the sample dust was 6.5%. Sieve analysis (SA) In the sieve analysis, 85 g sample particles were sieved by a sieve shaker (A3, Fritsch GmbH, Idar-Oberstein, Ger- many) for 10 min with 3 mm amplitude. The sieve stack was composed of 5 sieves, and their mesh sizes were 1000, 500, 250, 100 and 40 μm from the top to the bot- tom, respectively. Wood dust retained on each sieve and the collecting pan was then weighted for size distribution by an electronic balance (BS2202S, Sartorius AG, Goet- tingen, Germany). The analysis was performed twice and the average values were considered the results. Scanning electron microscopy (SEM) The particles used for SEM analysis were taken from sub- samples left on each sieve and the collecting pan. They were dried to the oven dry state, and then coated with
the wood processing industry [7], and are more likely to penetrate into human respiratory system. The amount of respirable dust generated during the sanding of MDF is more than other wood working processes, including solid wood sanding [8, 9]. The fineness of MDF dust also increases both the feasibility and violence of dust explo- sion [10, 11]. Proper handling of MDF sanding dust requires a full understanding of its properties. Morphological charac- teristics including particle size distribution (PSD) and shape distribution are the fundamental factors influenc- ing dust handling behaviors such as flowability, bulk den- sity and compressibility, etc. [12–14]. Large particles with spherical shape generally have good flowability, which deteriorates with decreased particle size as the inter-par- ticle cohesive force increases [15]. For irregularly shaped particles, the relative motion becomes difficult due to the presence of more contacting points between them. If elongated and hook-shaped particles are involved, it will be more complicated because they tend to form bridges by particle interlocking [16]. Limited studies have been performed on size and shape characteristics of MDF sanding dust. Mazumder sug- gested that a significant portion of MDF sanding dust was respirable particles with aerodynamic diameters smaller than 10 μm, and that the particles were of irregu- lar shapes with sharp edges [5]. Chung et al. investigated the MDF sanding dust emitted from handheld sander and found the portion of respirable dust was less than 10%, but a portion as high as 30% was also cited in his paper [9]. Očkajová et al. studied the size distribution of MDF sanding dust by sieve analysis and found that 96.16% of the sample particles were smaller than 100 μm, and that the most common particles were in the range between 32 and 63 μm [17]. No quantitative study on shape distribu- tion of MDF sanding dust has been found by the authors of this paper. In this study, morphological characteristics of MDF sanding dust were investigated by sieve analysis (SA) and image analyses (IA). SA has been widely used to determine the PSD of bio-based particles. Its popular- ity derives from low cost, simple procedure, straightfor- ward results and the similarity to the particle separating practice in wood-based panel industry. SA is a standard method to determine the PSD of some bio-based par- ticles [18–20], and has been applied in many scientific studies [21–23]. But in recent years, questions arose on the competency of SA for bio-based particles. The size that SA measures is the second smallest dimension, i.e., the width of a par- ticle [21, 24]. For spherical particles, the PSD obtained by SA is quite reasonable. But most wood dust generated from mechanical processes is irregularly shaped due to
Made with FlippingBook - Online catalogs