RESUMEN
The aim of the study was to compare the two reference methods for the determination of boron in water samples and further assess the impact of the method of preparation of samples for analysis on the results obtained. Samples were collected during different desalination processes, ultrafiltration and the double reverse osmosis system, connected in series. From each point, samples were prepared in four different ways: the first was filtered (through a membrane filter of 0.45 µm) and acidified (using 1 mL ultrapure nitric acid for each 100 mL of samples) (FA), the second was unfiltered and not acidified (UFNA), the third was filtered but not acidified (FNA), and finally, the fourth was unfiltered but acidified (UFA). All samples were analysed using two analytical methods: inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma optical emission spectrometry (ICP-OES). The results obtained were compared and correlated, and the differences between them were studied. The results show that there are statistically significant differences between the concentrations obtained using the ICP-MS and ICP-OES techniques regardless of the methods of sampling preparation (sample filtration and preservation). Finally, both the ICP-MS and ICP-OES methods can be used for determination of the boron concentration in water. The differences in the boron concentrations obtained using these two methods can be caused by several high-level concentrations in selected whole-water digestates and some matrix effects. Higher concentrations of iron (from 1 to 20 mg/L) than chromium (0.02-1 mg/L) in the samples analysed can influence boron determination. When iron concentrations are high, we can observe the emission spectrum as a double joined and overlapping peak.