Beam-induced oxidation of mixed-valent Fe (oxyhydr)oxides (green rust) monitored by STEM-EELS. (July 2019)

• Record Type: Journal Article
• Authors: Freeman, H.M.; Perez, J.P.H.; Hondow, N.; Benning, L.G.; Brown, A.P.
• Is Part Of: Micron. Volume 122(2019)
• Page Start: 46
• Page End: 52
• Publication Date: 2019-07
• Subjects: Monochromated EELS -- Fe oxidation -- Green rust -- Cryo-TEM -- Low dose TEM; Microscopy -- Periodicals; Electron Probe Microanalysis -- Periodicals; Microscopy -- Periodicals; Microscopie -- Périodiques; Microscopy; Periodicals; 502.82
• DOI: 10.1016/j.micron.2019.02.002
• Languages: English
• ISSNs: 0968-4328

Abstract

Highlights: Energy resolved STEM-EELS to quantify Fe oxidation states in green rust. Demonstration of electron beam induced in-situ oxidation. Methodology for quantifying the interaction between green rust and toxic metals. Abstract: Analytical transmission electron microscopy (TEM) is often used to investigate morphologies, crystal structures, chemical compositions and oxidation states of highly reactive mixed-valent mineral phases. Of prime interest, due to its potential role in toxic metal remediation, is green rust sulphate (GRSO4 ) an Fe II -Fe III layered double hydroxide. In this study, we quantified the effects that TEM analysis has on GRSO4 in order to ensure the measured material properties are a result of synthesis and reaction kinetics, and not due to sample preparation and analysis technique. To do this, we compared two sample preparation techniques (anoxic drop-cast with drying, and frozen-hydrated cryogenic) and exposed samples to the electron beam for several minutes, acquiring fluence series between ca. 40 e − Å −2 and 10, 000 e − Å −2 . TEM imaging and electron diffraction showed that the hexagonal plate-like morphology and crystal structure of GRSO4 were largely unaffected by sample preparation and analysis technique. However, quantitative analysis of a series of monochromated Fe L 3, 2 -edge electron energy loss spectra (EELS) showed that electron irradiation induces oxidation. We measured an Fe(II)/Fe(III) ratio of 1.94 (as expected for GRSO4 ) at 50 e − Å −2 . However, above this fluence, the ratio logarithmically decreased and dropped to ca. 0.5 after 1000 e − Å −2 . This trend was approximately the same for both sample preparation techniques implying that it is the beam alone which causes valence state changes, and not exposure to oxygen during transfer into the TEM or the vacuum of the TEM column. Ultimately this work demonstrates that GR valence can be quantified by EELS provided that the sample is not over exposed to electrons. This also opens the possibility of quantifying the effect of redox-sensitive toxic metals (e.g., As, Cr, Se) on Fe oxidation state in GR phases (relevant to the treatment of contaminated soils and water) with a higher spatial resolution than other techniques (e.g., Mössbauer spectroscopy).