Evaluation of Accuracy and Uncertainty of ISO 13503-2 Proppant Testing

Abstract

Proppant is one of the key aspects of a successful fracturing treatment and subsequent well production, and its quality is strictly controlled by ISO 13503-2, an international standard for proppant testing. Proppant suppliers, oilfield companies and their clients all around the world rely on the standard measurement procedures and limit values to ensure the quality of a product they produce or use for fracturing operations. ISO 13503-2 prescribes testing procedures; however, the standard does not contain information about uncertainty of final measurements. This information is essential for consistent quality assurance, for resolving inconsistencies between different laboratories, and for comparing properties of different products. An interlaboratory study was organized to evaluate the precision of ISO 13503-2 proppant testing. Four proppant samples were distributed among proppant quality control laboratories all around the world; 17 laboratories participated in the study. The obtained test data were used to calculate repeatability and reproducibility standard deviations (in accordance with ISO 5725-2) and uncertainty of ISO 13503-2 test data (in accordance with ISO 21748). The study showed that most proppant parameters measured using ISO 13503-2 methods yielded high uncertainty. For example, for turbidity and acid solubility values, the uncertainty was up to ±50%. Uncertainty of roundness and sphericity values was ±0.1 since the values are estimated by operator, making the test quite subjective. For crush resistance, the highest uncertainty among tested samples, ±40%, was observed for an HSP 30/50 sample measured at 15,000 psi stress. For absolute density, a systematic difference between values obtained using gas pycnometers from different manufactures was observed; also, the equipment differs by repeatability characteristics. The results of the interlaboratory study allowed estimating the uncertainty of ISO 13503-2 test methods. It was shown that high measurement uncertainty for some critical proppant parameters should be considered for correct interpretation of the obtained test results.