Abstract: High-purity quartz sand (with an SiO₂ content ≥99.998%) is a critical raw material in the semiconductor and photovoltaic industries. The content of impurity elements directly influences product performance, and their total concentration must be strictly controlled below 20μg/g. Inductively coupled plasma-mass spectrometry (ICP-MS) and inductively coupled plasma-optical emission spectroscopy (ICP-OES) are the primary techniques used for the determination of impurity elements in high-purity quartz sand. ICP-MS offers extremely low detection limits and high sensitivity. However, single-quadrupole ICP-MS suffers from significant mass spectral interferences caused by polyatomic ions, particularly due to its limited mass resolution. Common sample decomposition methods include single-acid digestion with hydrofluoric acid (HF) or mixed-acid digestion. Nevertheless, these approaches often require large amounts of reagents, may result in incomplete dissolution of refractory elements, and can lead to the loss of volatile components. In this study, high-purity quartz sand samples were decomposed using a closed-vessel acid digestion method with hydrofluoric and nitric acids. Boron was complexed with a mannitol solution to minimize its loss. A total of 16 impurity elements were accurately determined by high-resolution inductively coupled plasma-mass spectrometry (HR-ICP-MS). The closed-vessel acid digestion method employed a minimal amount of highly toxic reagents, achieved complete sample decomposition, and utilized mannitol complexation to prevent the loss of volatile elements. Appropriate isotope selection and medium-to-high resolution settings (R ≥ 4000) effectively separated the mass spectral peaks of target elements from those of interfering ions, thereby mitigating polyatomic ion interferences. This method was applied to the analysis of a high-purity quartz sand international standard reference material (IOTA-CG) and several real samples. The results obtained for the standard sample were in good agreement with the certified values, with relative error (RE) for all elements ≤10.0%. For the real samples, the standard addition recovery rates ranged from 92.0% to 108.0%, the relative standard deviation (RSD, n=12) for all elements were ≤9.78%, and the limits of detection (LOD) ranged from 0.000089μg/g to 0.33μg/g. This method addresses the challenge of accurately and simultaneously determining multiple elements in high-purity quartz sand, providing technical support for its quality evaluation and high-end applications.