Nuclear Engineering and Design 288 (2015) 175–182 Contents lists available at ScienceDirect Nuclear Engineering and Design jou rn al hom ep age: www.elsevier.com/locate/nucengdes A computer-controlled experimental facility for krypton and xenon adsorption coefficient measurements on activated carbons Daniele Del Serra, Donato Aquaro, Dahmane Mazed, Fabio Pazzagli, Riccardo Ciolini ∗ Department of Civil and Industrial Engineering (DICI), University of Pisa, Largo Lucio Lazzarino 1, I-56122 Pisa, Italy h i g h l i g h t s • An experimental test facility for qualification of the krypton and xenon adsorption properties of activated carbons. • The measurement of the adsorption coefficient by using the elution curve method. • The simultaneous on-line control of the main physical parameters influencing the adsorption property of activated carbon. a r t i c l e i n f o Article history: Received 10 December 2014 Received in revised form 26 March 2015 Accepted 27 March 2015 C. Material properties a b s t r a c t An automated experimental test facility, intended specifically for qualification of the krypton and xenon adsorption properties of activated carbon samples, was designed and constructed. The experimental apparatus was designed to allow an on-line control of the main physical parameters influencing greatly the adsorption property of activated carbon. The measurement of the adsorption coefficient, based upon the elution curve method, can be performed with a precision better than 5% at gas pressure values ranging from atmospheric pressure up to 9 bar and bed temperature from 0 up to 80 ◦ C. The carrier gas flow rate can be varied from 40 up to 4000 N cm 3 min -1 allowing measurement of dynamic adsorption coefficient with face velocities from 0.3 up to 923 cm min -1 depending on the gas pressure and the test cell being used. The moisture content of the activated carbon can be precisely controlled during measurement, through the relative humidity of the carrier gas. © 2015 Elsevier B.V. All rights reserved. 1. Introduction In the operation of nuclear power plants, to prevent contami- nation of the atmosphere special provision must be made for the disposal of radioactive noble gases (krypton and xenon) generated during the fission process and escaped from damaged fuel rods (Foerster, 1971). Appropriate trap systems with activated carbon by delay beds, allowing a selective dynamic adsorption of the released radioactive gases, are broadly used in these plants (Underhill and Moeller, 1980a). In the process of dynamic adsorption, krypton and xenon gases are physically adsorbed from a moving carrier gas, commonly nitro- gen as in the present case, onto a porous material such as activated carbon, in a manner similar to that used in gas chromatogra- phy (Suzuki, 1990). Although the adsorbate radioactive gas is not permanently bound to the adsorber, its exit from the adsorber column is delayed, compared to the carrier gas. Non-permanent ∗ Corresponding author. Tel.: +39 050 2218026; fax: +39 050 2210604. E-mail address: r.ciolini@ing.unipi.it (R. Ciolini). bounding is basically caused by small-ranged and reversible van der Waals forces (dispersion forces) acting between the gas and adsorber molecules, and described by the Lennard–Jones potential. This selective dynamic adsorption process increases the reten- tion times of the noble gases inside the delay bed units such that the radioactive gases are allowed to decay, before their release, hence reducing consistently the activity released to the environ- ment. The gas adsorption can be performed at different pressure and gas flow rates, depending on the required volume of waste gas to deal with. Up to now, two main approaches were considered to describe the adsorption behavior of noble gases on activated carbon. The first one describes the delay bed as a continuous homo- geneous medium (continuous column theory), while the second one involves a discontinuous medium modeled by a series of theo- retical adsorption chambers to which the diffusion model is applied (theoretical plate theory). For linear systems, i.e. when very small gas tracer concentrations (partial pressures) are considered, both theories lead to the conclusion that the shape of the gas tracer pulse at the bed outlet, so-called elution curve, can be represented by a Gaussian distribution. http://dx.doi.org/10.1016/j.nucengdes.2015.03.019 0029-5493/© 2015 Elsevier B.V. All rights reserved.