SUMMARYDust is a key component of fusion power device accident source term. Understanding the amount of dust expected in fusion power devices and its physical and chemical characteristics is needed to verify assumptions currently used in safety analyses. An important part of this safety research and development work is to characterize dust from existing experimental tokamaks. In this report, we present the collection, data analysis methods used, and the characterization of dust particulate collected from various locations inside the General Atomics DIII-D vacuum vessel following the June 1998 vent. DIII-D is located in San Diego, California. The collected particulate was analyzed at the Idaho National Engineering and Environmental Laboratory (INEEL).Two methods were used to collect particulate with the goal of preserving the particle size distribution and physical characteristics of the particulate. Choice of collection technique is important because the sampling method used can bias the particle size distribution collected. Vacuum collection on substrates and adhesion removal with metallurgical replicating tape were chosen as non-intrusive sampling methods. Sampling was completed in four areas of the machine; 0 to 90° area, the 90° to 180° area, the 180° to 270° area, and the 270° to 360° area. The 0° direction designates the north side of the DIII-D machine. Seventeen samples were collected including plasma facing surfaces in lower, upper, and horizontal locations, surfaces behind floor tiles, surfaces behind divertor tiles, and surfaces behind ceiling tiles. The results of the analysis are listed below.• Our Brunauer, Emmett, and Teller (BET) specific surface area analysis results are based upon three measurements of a sample combined from approximately 75% of the filter housing samples. The samples were combined to yield a detectable surface area of dust. The combined sample had an average specific surface area of 3.18 m 2 /gram from a total sample weight of 0.11 grams. This value is very similar to the value obtained in the 1996 DIII-D dust campaign and is a factor of 2 greater than that for theoretically dense graphite spheres. This indicates that as previously observed, the particulate is not spherical and is composed of agglomerates of smaller particles.• The particle size distribution analysis results of dust taken from inside the vacuum vessel show a range of count median diameter (CMD) values ranging between 0.81 µm and 2.89 µm with a range of geometric standard deviation (GSD) values between 1.33 and 3.43. D MVS values ranged between 1.1 and 13.3 µm. The mass median diameter (MMD) of the particulate was determined from the cascade impactor data. The MMD's were 7.25 µm with a GSD of 1.4 and 5.44 µm with a GSD of 2.2. The observed particles appeared in the form of both flakes and spheres. Compositional analysis by scanning electron microscopy/energy dispersive spectroscopy (SEM/EDS) showed individual dust composed of molybdenum, iron, boron, and aluminum. Overall chemical analysis by inductively couple...