Issue 5 2015 193 TOS forum wcsb7 proceedings www.impublications.com/wcsb7 Estimating granite roughness using systematic random sampling for the evaluation of radon gas emanation from ornamental granite rocks T.M. El Hajj, a I. Tertuliano, b T. Vieira, c A.C. Chieregati a and H. Delboni Jr. a a Dept. of Mining and Petroleum Engineering, University of Sao Paulo, Av. Prof. Mello Moraes 2373, 05508-030, Sao Paulo, Brazil. E-mail: thammiris.poli@usp.br , ana.chieregati@usp.br , hdelboni@usp.br b Dept. of Mechanical Engineering, University of Sao Paulo, Av. Prof. Mello Moraes 2231, 05508-030, Sao Paulo, Brazil. E-mail: iramar@usp.br c Dept. of Civil Engineering, University of Sao Paulo, Av. Prof. Almeida Prado 83, 05508-070, Sao Paulo, Brazil. E-mail: tiagovr@gmail.com There are three natural radioactive families according to their decay, which are: the uranium series ( 238 U decreasing to stable 206 Pb), the actinium series ( 235 U decreasing to stable 207 Pb) and the thorium series ( 236 U ® 232 Th decreasing to stable 208 Pb). The three series all have radon gas as an intermediary element, but each with a different atomic mass ( 222 Rn, 219 Rn and 220 Rn). The three isotopes are inert gases at ambient conditions and all are alpha particles emitters. Soils naturally emanate these radioactive gases in variable concentrations depending on composition and location. The radon radioactive emanation is a mass fow composed of radionuclides emitted to the atmosphere from the surface of the material, or transported to it. Emanation depends on the amount of radon atoms formed from the decay of radium and on the surface roughness of the material. Treatment such as polishing can be used to decrease radon gas emanation by closing open surface pores and reducing the specifc surface area. This study aims at evaluating granite roughness of experimental plates of ornamental rocks using a systematic random sampling approach in order to minimise analysis time. To validate the systematic minimum area sampling results these were compared to measurements made over the whole reference area. It is concluded that measurements can be conducted in just a few locations using systematic random sampling, signifcantly reducing the time for obtaining estimates of the granite’s roughness by factors 150–200. Introduction T his study addresses development of a fast method to obtain a granite plate’s roughness using systematic ran- dom sampling as a tool to minimize the measurement time without quality loss. Granite roughness is an important parameter for a correct evaluation of radon gas emanation. Esti- mating granite roughness is part of test regimens for characterizing gas emanation. This test is by far the slowest of all characterisation tests. It takes 16 days to obtain relevant data from a 20 × 20 cm plate. This motivated the authors to use a sampling method to reduce the time needed for estimating granite roughness. Among the available surface treatments aiming to decrease radon gas emanation, there is granite polishing which is a cheap and efficient method since it reduces the specific area and closes open surface pores. Figure 1 shows the difference of emanation between polished and rough surfaces measured by a radon meter, which has a scintillation cell as operating principle. The EU (European Union) published a Council Directive in December of 2013 (2013/59 EURATOM) that compels all member states to present a national action plan to address long-term risks from radon exposures by February of 2018. Guidance on methods and tools for measurements, identification of building materials with significant radon emanation are on the list of items to be consid- ered in preparing this action plan. In view of this EU document, the authors feel that methodological studies in this field are well motivated. Radon emanation There are three radon isotopes ( 222 Rn, 219 Rn and 220 Rn) which are alpha particles emitters, all of which are inert gases at ambient conditions. The radioactive emanation is a mass flow composed of radionuclides transported to the atmosphere from the material, depending on the amount of radon atoms formed from decay from radium and the physical characteristics e.g. surface roughness 1 . The amount of gas that reaches the surface is directly proportional to the specific area of the material. Inhalation of radon gas and its decay products is a health risk to humans. Alpha particles from the radioactive decay may reach lung tissue and cause damage that can lead to lung cancer. Most of radon gas exits the human body by exhalation before the decay process however, so most of the radioactive dose comes from the decay products that are inhaled as dust and become lodged in the lung tissue. These radionuclides decay quickly which results in fur- ther damage of the lung tissue 2 . Surface metrology – Interferometry Surface metrology is a branch of engineering related to measure- ment of roughness, sharpness, waviness and other surface param- eters, which are dependent on a given engineering application. Methods available to characterize surface texture can be classi- fied as contacting and non-contacting 3 . While contacting meth- ods demand physical contact to assess the surface topography, non-contacting methods, as the name implies, do not require any such. Surface interferometry is a non-contacting technique, based on a superposition principle. Two waves with no phase shift, identi- cal w.r.t. amplitude and frequency, when combined, will result in a wave with the same frequency but the amplitude will be doubled. This effect is known as constructive interference. Two waves with a phase shift of 180° will result in a wave with zero amplitude. This effect is known as destructive interference. The interaction between different waves in general results in patterns, known as fringes, showing constructive and destructive interference. Figure 2 shows doi: 10.1255/tosf.52