Based on the knowledge gained from published studies, a new analytical method has been developed for the quantification of mercury (Hg) in the gas-vapor phase of mainstream cigarette smoke and in heated tobacco aerosol generated by a tobacco heating system (THS) using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). From a preliminary test, the mercury concentration in the particulate matter of mainstream smoke from Kentucky reference cigarettes 3R4F generated under the International Organization for Standardization (ISO) smoking regimen was compared with the mercury concentration in the gasvapor phase to assure that mercury is only measurable in the gas-vapor phase, as reported in an earlier published study. The particulate matter was collected using an electrostatic precipitation trap and was analyzed by ICP-MS after a mineralization step. The gas-vapor phase was trapped in the same smoking run as for the particulate matter using two impingers containing a nitric acid-hydrochloric acid-gold solution. The impingers were connected in series behind the electrostatic precipitation trap and the combined impinger solution was analyzed by ICP-MS after sample dilution without further sample treatment. The addition of gold has shown to be efficient for maintaining mercury in an ionized form in the impinger solution and to minimize the mercury memory effect in the sample introduction system of the ICP-MS. Only mercury in the gas-vapor phase could be quantified whereas the signal for mercury in the particulate matter was found close to those of blank solutions and was not measurable, as already mentioned in an earlier study. Following this preliminary test, the electrostatic precipitation trap was replaced by a Cambridge filter pad for the separation of the gas-vapor phase from the particulate matter where only mercury in the gas-vapor phase was quantified.
The method for the quantification of mercury in the gas-vapor phase of aerosols obtained under Health Canada (HC) and ISO smoking regimens was validated according to International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) and Association of Official Analytical Chemists (AOAC) guidelines. Accuracy profiles were evaluated as described in Association Française de Normalisation (AFNOR). The regression curve was shown to be linear within the evaluated concentration range from 25 pg/mL to 1000 pg/mL with a weighting factor 1/x. The coefficients of variation for repeatability (r) were 3.6% for 3R4F and 4.8% for THS under HC smoking regimen and 3.6% for 3R4F and 4.6% for THS under ISO smoking regimen. The coefficients of variation for intermediate precision (IP) were 7.7% for 3R4F and 7.7% for THS under HC smoking regimen and 4.7% for 3R4F and 4.6% for THS under ISO smoking regimen. The nominal mercury concentrations for 3R4F obtained during the validation under both HC and ISO smoking regimens were found to be in line with results reported in a previously published CORESTA study.