Thin-foil SEM analysis of soil and groundwater colloids: Reducing instrument and operator bias

Authors
Citation
Jc. Seaman, Thin-foil SEM analysis of soil and groundwater colloids: Reducing instrument and operator bias, ENV SCI TEC, 34(1), 2000, pp. 187-191
Citations number
31
Categorie Soggetti
Environment/Ecology,"Environmental Engineering & Energy
Journal title
ENVIRONMENTAL SCIENCE & TECHNOLOGY
ISSN journal
0013936X → ACNP
Volume
34
Issue
1
Year of publication
2000
Pages
187 - 191
Database
ISI
SICI code
0013-936X(20000101)34:1<187:TSAOSA>2.0.ZU;2-0
Abstract
Scanning electron microscopy (SEM) combined with energy-dispersive X-ray sp ectrometry (EDS) has been widely used to characterize environmental colloid s with little regard for instrumental limitations. Details associated with electron emission source, excitation voltage, and even sample-coating proce dures are often omitted from publications because SEM techniques are genera lly considered routine. In the current study, the common analytical techniq ue of depositing colloidal samples on polycarbonate filters (i.e., conventi onal analysis) and carbon coating for microanalysis was compared to a thin- foil mounting technique that enhances particle recognition and reduces the nonspecific background (i.e., Bremsstrahlung X-rays) that interferes with c ompositional analysis of submicron particles. Carbon-coated particle resolu tion was degraded compared to metal-coated samples, making it difficult to identify smaller particles or discern fine-grained aggregates during EDS an alysis. The signal-to-noise ratio for the resulting EDS patterns was highly related to particle size, with smaller particles generating patterns domin ated by Bremsstrahlung X-rays. For the thin-foil method, EDS patterns for s maller particles produced better signal-to-noise ratios that were independe nt of particle size. The ability to easily discriminate particles from the background and collect detailed EDS spectra with minimal analysis time make s this technique ideal for instrumental automation.