Exploration of methods to remove implanted ²¹⁰Pb and ²¹⁰Po contamination from silicon surfaces

Abstract

Radioactive contaminants on the surfaces of detector components can be a problematic source of background events for physics experiments searching for rare processes. Exposure to radon is a specific concern because it can result in the relatively long-lived ²¹⁰Pb (and progeny) being implanted to significant subsurface depths such that removal is challenging. In this article we present results from a broad exploration of cleaning treatments to remove implanted ²¹⁰Pb and ²¹⁰Po contamination from silicon, which is a material frequently used in rare-events searches. We demonstrate for the first time that heat treatments (“baking”) can effectively mitigate such surface contamination, with a 1200 °C bake resulting in a 97${}_{-3}^{+2}$% reduction. We also report results using wet-chemistry and plasma-based methods, which show that etching can be highly effective provided the etch depth is sufficiently aggressive. Our survey of cleaning methods suggests consideration of multiple approaches during the different phases of detector construction to enable greater flexibility for efficient removal of ²¹⁰Pb and ²¹⁰Po surface contamination.

Introduction

Radon contamination requires a dedicated consideration for detectors that require low backgrounds from radioactivity to search for rare physics processes such as interactions from dark matter and neutrinoless double-beta decay (see, e.g., Refs. [1], [2], [3], [4], [5], [6]). Radon can manifest as a background in several ways. In this article, we focus on the long-lived surface contamination present on detector components exposed to radon, e.g., during fabrication and assembly. Specifically, when ²²²Rn decays in air, its progeny plate onto surfaces where the relatively prompt alpha decays of ²¹⁸Po and ²¹⁴Po can implant ²¹⁰Pb contamination to subsurface depths of many tens of nanometers [7], [8], [9]. The long 22 year half-life of ²¹⁰Pb [10] and its decay chain induce multiple types of background radiation – x-rays, betas, alphas, recoiling ions – for the full duration of detector operations.

There are two primary mitigation approaches: use of radon-reduced environments to prevent radon exposure and surface cleaning to remove ²¹⁰Pb and its progeny. To achieve sufficiently low radon levels, the former can require significant infrastructure such as a dedicated low-radon cleanroom (as in Refs. [11], [12], [13], [14], [15], [16]). When practical, simple cleaning methods can be inexpensive but may be only partially effective because a significant fraction of the contamination is too deeply implanted to be easily removed. Exploration of more effective cleaning treatments is motivated by a need for cost-effective options for efficient removal of surface contaminants.

Methods for removal of radon-progeny surface contamination have been explored for a variety of materials and reported in the literature. Cleaning of copper surfaces has been studied extensively, with methods ranging from relatively simple etching [8], [17], [18], [19], [20] to more complicated multi-step procedures [21]. Other investigations have considered materials such as polymers [7], [22], [23], stainless steel [8], [18], [24], semiconductor crystals [18], [25], [26], [27], and glass [28], using a variety of cleaning techniques such as leaching, etching, and electropolishing. Notably, silicon is a detector material often used in rare-event experiments [1], [29], [30], [31], [32], [33], [34]. There are many reports on surface treatments to advance fabrication of silicon-based devices (e.g., as in Refs. [35], [36], [37], [38], [39]). However, there appear to be few studies focused on removal of radioactive contaminants from silicon surfaces, such as the one in Ref. [27] which specifically targeted crystal sidewalls.

In this article, we report results to remove implanted ²¹⁰Pb and ²¹⁰Po from silicon surfaces using a variety of methods. Recognizing that there may be opportunities for surface treatments during different stages of detector fabrication (for other materials as well as silicon), we performed a general exploration using a broad range of processes, including heat treatments, wet chemistry, and plasma etching. The objective is to identify a suite of cleaning options to enable greater flexibility to effectively mitigate radon-progeny surface contamination. The paper is organized as follows. Section 2 describes our overall approach, including our preparation of ²¹⁰Pb-implanted samples in Section 2.1 and our surface-contamination measurement and analysis methods in Section 2.2. Section 3 motivates the different surface treatments, describes the specific procedures that we used, and presents our ²¹⁰Pb and ²¹⁰Po removal results. Finally, we summarize and discuss our results in Section 4.