Hydride generation atomic absorption spectrometry (HGAAS) with different types of atomizers (flame heated quartz T-tube and W-coil) was evaluated for comparison of interference effects of some hydride forming elements for arsenic determination. W-coil was used as an on-line atomizer or as a trapping surface. Mutual interferences of Sb(III), Se (IV), Sn (II) and Te(IV) on arsenic signal were compared with flame heated quartz T-tube atomizer and W-coil; on-line atomization and in-situ trapping modes. When W-coil was operated at on-line atomization mode the interference magnitudes could not be improved as much as expected with respect to quartz T-tube atomizer. However, decent improvements were achieved for interference magnitudes of each interferent species with in-situ trapping on iridium coated W-coil. Under optimized conditions and using 60 s trapping period (sample volume 6 ml), the limit of detection was found as 0.03 ng ml(-1) and the calibration was linear in between 0.25-4.0 ng ml(-1) As. The precision of the analytical method was less than 2.5% in terms of RSD (n = 11) for 1.0 ng ml(-1) arsenic concentration. The accuracy of the method was tested with certified reference materials; fortified water TMDA 61 (NWRI) and waste water EU-L-1 (SCP). The results were in a good agreement with the certified values at 95% confidence level. (C) 2016 Elsevier B.V. All rights reserved.