Mr Shem, thank You for answer to my question. Howewer, by my opinion, such methods are approximate, because observed characteristics strongly depend of surrounding conditions-- temperature, humidity, degree of air motion, content of different reactive substances in surroundings, etc. These various conditions are practically impossible to take precisely into account. I think, it's necessary to consider phenomena that not depend on such conditions. Unfortunately, I can't figure out something appropriate, exept nuclear processes (something like isotopic method). But it's very difficult, if possible at all, to work out something like this in adaptation to above mentioned question.