Environmental Fate and Toxicity of III-V Engineered Nanoparticles in Semiconductor Manufacturing
Author
Nguyen, Chi HuynhIssue Date
2019Advisor
Sierra-Alvarez, ReyesField, James A.
Metadata
Show full item recordPublisher
The University of Arizona.Rights
Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.Embargo
Release after 12/03/2021Abstract
Engineered nanoparticles (NPs) have many unique electronic, chemical, and optical properties. Gallium arsenide (GaAs) NPs and indium arsenide (InAs) NPs are being considered in different semiconductors and electronic devices due to their favorable properties such as high electron mobility with wide and adjustable band gaps and their reduced power consumption. Polishing of thin films in a process known as chemical and mechanical planarization (CMP) could lead potentially to the release of byproducts from GaAs and InAs, such as soluble III-V materials (Ga, In, and arsenic (As) species) and particulate III-V materials (GaAs NPs, InAs NPs, gallium oxide (Ga2O3), indium oxide (In2O3)) into the wastewater stream. Furthermore, the CMP process utilizes cerium oxide (CeO2), silica (SiO2), and alumina (Al2O3) as abrasive particles in slurries to polish and create flat surfaces. As a result, CeO2, SiO2, and Al2O3 NPs will also be present in the waste streams of the semiconductor industry. Additionally, some NPs can promote the transport of toxic chemicals into cells through the “Trojan Horse” effect, which can potentially alter the toxicity of the NPs via adsorbed chemicals. Among the soluble III-V ions, arsenic is a concern for the Trojan Horse effect because it is a highly toxic and carcinogenic element with soluble species that can become adsorbed onto NP surfaces. Although the toxicity of soluble arsenic species is well established, little is known about the potential toxicity of other soluble III-V materials and III-V NPs. Therefore, it is important to study and understand the environmental fate and toxicity of these materials. The objectives of this work are to investigate the potential toxicity and environmental fate of III-V nanomaterials and byproducts that could be formed in CMP slurries of semiconductor manufacturing effluents as well as to study the impact of CeO2 NPs on the sorption and toxicity of As species. The acute toxicity of GaAs, InAs, Ga2O3, and In2O3 particulates was investigated using two microbial assays targeting methanogenic archaea and the marine bacterium, Allivibrio fischeri. The results showed that GaAs and InAs NPs were acutely toxic towards these microorganisms while Ga2O3 and In2O3 NPs were not. The release of soluble arsenic species from arsenide NPs was shown to play a key role in the toxic effect of the arsenide NPs. Their toxicity increased with decreasing particle size and with increasing time because of the progressive corrosion of the NPs in the aqueous bioassay medium. In summary, the toxicity exerted by the arsenide NPs under environmental conditions will vary depending upon the particle size, dissolution time, and aqueous chemistry. In addition to microbial toxicity, these materials could cause toxic effects to human health. The toxic effects of Ga-based and In-based NPs (GaAs, InAs, Ga2O3, and In2O3) as well as soluble III-V salts (AsIII, AsV, GaIII, and InIII) on human bronchial epithelial cells (16HBE14o-) were evaluated using an impedance-based real time cell analyzing (RTCA) system. The results showed that only dissolved arsenic (AsIII and AsV) and arsenide particulate compounds caused significant inhibition at low concentrations (IC50 values after 16 h of exposure (16 h-IC50): 2.4 mg AsIII/L, 4.5 mg AsV/L, 6.2 mg GaAs NPs/L, and 68 mg InAs NPs/L). Similar to the case with the microbial toxicity, the cytotoxicity of the arsenide NPs to 16HBE14o- cells was mainly caused by dissolution of toxic As species (mostly AsIII). On the other hand, the soluble salt, GaIII-citrate, and Ga2O3 NPs caused mild inhibition while InIII-citrate and In2O3 NPs were not toxic at the concentrations tested (16 h-IC50 value: 260 mg Ga2O3 NPs/L), while InIII-citrate and In2O3 NPs were not toxic at the concentration tested (16 h-IC50 value: > 300 mg In2O3 NPs/L). In conclusion, GaAs NPs and, to a lesser extent InAs NPs, display toxicity to human lung cells and the adverse effects are expected to increase with increasing NP dissolution. Besides Ga- and In-based NPs, abrasive NPs used in CMP slurries (i.e. CeO2, Al2O3, colloidal SiO2 (c-SiO2), and fumed SiO2 (f-SiO2)) will be present in semiconductor effluents and they could potentially be toxic to microbial communities and human health. The toxic effects of well-characterized model CMP slurries on the marine bacterium A. fischeri and human bronchial epithelial cells 16HBE14o- cells were investigated. The results showed that f-SiO2 and CeO2 slurries did not cause acute toxicity on A. fischeri at concentrations as high as 1136 and 909 mg/L, respectively. In contrast, c-SiO2 and Al2O3 caused about 30% inhibition on microbial activity after 30 min of exposure at relatively high concentration (1364 mg c-SiO2/L and 1364 mg Al2O3/L). Lung cells were more sensitive to exposure to some of these inorganic oxide NPs. High concentrations (250 and 500 mg/L) of c-SiO2 and f-SiO2 slurries led to lung cell death in the RTCA assay. On the other hand, CeO2 and Al2O3 slurries were either not inhibitory or only showed limited inhibitory effect on 16HBE14o- cells after 24 h of exposure. As a whole, the results indicate that the abrasive NPs used in CMP are not likely to cause acute environmental and health risks at the low concentrations expected in surface water (< 1 mg/L). The presence of CMP NPs together with III-V soluble species can lead to emerging environmental and health problems. This study showed that CeO2 NPs effectively decreased the concentration of available AsIII in the culture medium through adsorption; hence showing the effect completely opposite of the “Trojan Horse” effect. Additionally, this work showed that internalization of CeO2 NPs by human lung cells was observed in vesicles (most likely lysosomes). Taken as a whole, this dissertation demonstrates that GaAs NPs, and to a lesser extent InAs NPs, cause acute toxicity to the studied microbial targets and to human lung cells due to the corrosion of the nanomaterials in the aqueous environment, and the ensuing release of toxic AsIII. Other nanomaterials anticipated in CMP effluents (i.e. Ga2O3 NPs and In2O3 NPs) were relatively non-toxic. Abrasive NPs like CeO2, Al2O3, c-SiO2, and f-SiO2 used in CMP slurries caused certain toxic effects to human bronchial epithelial cells at relatively high concentrations. The existence of CeO2 NPs along with soluble AsIII in waste effluent led to adsorption of AsIII onto CeO2 NPs that, as a result, decreased the toxicity of AsIII dramatically.Type
textElectronic Dissertation
Degree Name
Ph.D.Degree Level
doctoralDegree Program
Graduate CollegeEnvironmental Engineering