Difficulties Encountered in Analytical Work of Used Oil
There were several problems encountered during the analytical procedures. First, it was discovered that the inhomogeneity of the used oil samples could cause inconsistent results unless the samples were prepared with special care at each step. The bulk sample had to be continuously stirred and the representative portion had to be carefully transferred to make emulsions or used as it is for the test runs. When water and oil are separated after the reaction, representative samples have to be taken from each phase for analysis. For analysis conducted by GC (e.g., PAHs, PCBs, BTEX, and chlorinated hydrocarbons), the amount of sample taken for preparation of the dilution is no more than one gram. For metal analysis by ICP-MS, only 20 mg or even 10 mg was taken from 114 each sample for preparation. It is obvious that if a sample taken for analysis is not from a homogeneous solution, replicated results cannot even be anticipated for the same sample matrix. Thus, it is critical to develop a secure sampling procedure for both water and oil phase samples in order to obtain reliable and reproducible results. Especially, homogeneity for oil phase samples needs more efforts to achieve due to the viscous property of oil. For the analyses that are performed in other facilities (BNL analytical lab, or other commercial labs), specific instruction should be given to the technicians involved to use same sampling techniques and procedures. Secondly, it should be noted that used oil samples may be aged through time and some of the constituents may change or be released. A blank sample, therefore, has to be prepared in each set of test runs for analysis; it will serve as the new baseline to compare the degradation efficiency. Lastly, the dirty and complex nature of the used oil samples makes the trace level analyses a little difficult. For example, a cleaning procedure to remove some interference factors is required for all samples; otherwise, PCBs peaks cannot be clearly identified and quantified. The detection limit for determining the organic contaminants (PCBs, PAHs, and chlorinated compounds) in the oil phase is very high since further concentration of these components in oil phase is very difficult and there is too much interference from the samples themselves.