3.5. Quality parameters and method trueness
Under the selected physical and chemical variables detailed in the
foregoing sections, the analytical performance of the FI-SPE-HG-AAS
method using pro-MWCNTs (substrate A) and NPC (substrate B) for
Sb(III) retention resulted as follows: the repeatability, expressed as relative
standard deviation (RSD) was 2.1 and 2.4% (for substrate A and B,
respectively) for ten replicatemeasurements for 0.1 μg L−1. The limitsof
detection (LODs) calculated as the concentration associated with three
times the standard deviation of 10 independent measurements of the
procedural blank (3σ criterion) were 1.47 ng L−1 (NPC) and0.52 ng L−1 (pro-MWCNTs). The limits of quantification (10σ) resulted
to be 4.85 and 1.72 ng L−1 for NPC and pro-MWCNTs, respectively. The
calibration curves were linear from levels close to the LOD up to
120 ng L−1. A good correlation coefficient has been obtained with
R2 = 0.998. According to Fillela et al. [2] typical concentrations of Sb
in natural waters are b1 μg L−1 while ocean surface concentrations
are ~0.2 μg L−1. The LODs obtained for both nanomaterials evidence
that the proposed method is fully able to determine Sb(III) in different
categories of natural waters.