Raw lead materials contain many residual elements. With respect to setting 'safe' levels for these elements, each country has its own standard, but the majority of the present specifications for the lead used to prepare battery oxide apply to flooded batteries that employ antimonial grids. In these batteries, the antimony in the positive and negative grids dominates gassing characteristics so that the influence of residual elements is of little importance. This is, however, not the case for valve-regulated lead-acid (VRLA) batteries, which use antimony-free grids and less sulfuric acid solution. Thus, it is necessary to specify 'acceptable' levels of residual elements for the production of VRLA batteries. In this study, 17 elements are examined, namely: antimony, arsenic, bismuth, cadmium, chromium, cobalt, copper, germanium, iron, manganese, nickel, selenium, silver, tellurium, thallium, tin, and zinc. The following strategy has been formulated to determine the acceptable levels: (i) selection of a control oxide; (ii) determination of critical float, hydrogen arid oxygen currents; (iii) establishment of a screening plan for the elements; (iv) development of a statistical method for analysis of the experimental results. The critical values of the float, hydrogen and oxygen currents are calculated from a field survey of battery failure data. The values serve as a base-line for comparison with the corresponding measured currents from cells using positive and negative plates produced either from the control oxide or from oxide doped with different levels of the 17 elements in combination. The latter levels are determined by means of a screening plan which is based on the Plackett-Burman experimental design. Following this systematic and thorough exercise, two specifications are proposed for the purity of the lead to be used in oxide production for VRLA technology. (C) 2009 Elsevier B.V. All rights reserved.