Discussion on the stability of natural circulation loops with supercritical pressure fluids Seth Kofi Debrah a , Walter Ambrosini b,⇑ , Yuzhou Chen c a University of Ghana, School of Nuclear and Allied Sciences, Department of Nuclear Engineering, P.O. Box AE 1, Atomic Energy, Kwabenya, Accra, Ghana b Dipartimento di Ingegneria Civile e Industriale, Università di Pisa, Largo Lucio Lazzarino 2, Pisa 56126, Italy c China Institute of Atomic Energy, Department of Reactor Engineering Design, P.O. Box 275, Beijing 102413, PR China article info Article history: Received 10 August 2012 Received in revised form 24 October 2012 Accepted 25 October 2012 Available online 10 December 2012 Keywords: Natural circulation Supercritical pressure fluids Stability abstract The paper presents a methodology for the analysis of the flow stability in natural circulation loops con- taining fluids at supercritical pressure. The work was made possible by the availability of experimental data collected on an experimental apparatus at the China Institute for Atomic Energy, showing the onset of unstable behaviour. The RELAP5 code is firstly used for predicting the observed phenomena, obtaining a qualitative prediction of instabil- ities with quantitative discrepancies in relation to the conditions observed for the onset of unstable behaviour. In the aim to allow for a systematic discussion of stability phenomena, dimensionless parameters pre- viously proposed to analyse the stability of single heated channels with fluids at supercritical pressures are now adapted and applied to natural circulation loops. The proposed dimensionless numbers address purely thermal–hydraulic phenomena, disregarding the presence of heating structures. The combined analyses with the system code and an in-house code written in dimensionless form and based on the mentioned parameters allowed highlighting the importance of heat transfer to heating structures as a phenomenon contributing to stabilise natural circulation loops. Ó 2012 Published by Elsevier Ltd. 1. Introduction Flow stability of natural circulation loops has received attention in the last decade also as a consequence of the new interest for the Supercritical Water Reactor design in the frame of Generation IV studies (see the recently published textbooks by Pioro and Duffey, 2007; Oka et al., 2010; Schulenberg and Starflinger, 2012). Interesting previous work on the subject of natural circulation with supercritical pressure fluids has been performed by different research groups throughout the world (see e.g., Chatoorgoon, 2001; Chatoorgoon et al., 2005a,b; Jain and Corradini, 2006; Jain and Rizwan-uddin, 2008; Lomperski et al., 2004; Marcel et al., 2008, 2009; Rohde et al., 2011; Sharma et al., 2010a,b), highlight- ing some of its main features and also reporting on dimensionless parameters to be used in scaling experimental facilities in relation to the selection of the operating fluids. In addition to theoretical and model developments, aiming at anticipating the behaviour to be expected in experimental facilities and industrial scale systems, experiments are being performed and were recently published as a consequence of the need to validate the obtained predictions and to identify weak assumptions adopted in modelling. In particular, experiments with different flu- ids are reported, with an obvious preference for water, being the target fluid for SCWR applications (e.g., Swapnalee et al., 2012; Chen et al., 2012). In previous work, dimensionless parameters for analysing sta- bility of heated channels with supercritical fluids were proposed and applied to different channel configurations with different flu- ids and different numerical tools, namely 1D models and CFD codes (see, e.g., Ambrosini and Sharabi, 2008, 2007; Ambrosini, 2007, 2011; Sharabi et al., 2008). The absence of immediately applicable experimental data to be used for comparison with model predic- tions delayed the verification of applicability of these dimension- less parameters on an experimental ground up to recent times. Recent papers (Swapnalee et al., 2012; Xiong et al., 2012) provided a first discussion of these parameters on the basis of data related to H 2 O and CO 2 , though a systematic validation has still to come. On one hand, the derivation of these dimensionless parameters on the basis of 1D balance equations and their assessment made using calculated data for different fluids provides a reasonable basis for assigning confidence to their capabilities in establishing similari- ties; nevertheless, the comparison with experimental data only will decide about their effective usefulness. 0306-4549/$ - see front matter Ó 2012 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.anucene.2012.10.015 ⇑ Corresponding author. Tel.: +39 050 836673; fax: +39 050 2218065. E-mail addresses: s.debrah@gaecgh.org (S.K. Debrah), walter.ambrosini@ing. unipi.it (W. Ambrosini), chenyz@ciae.ac.cn (Y. Chen). Annals of Nuclear Energy 54 (2013) 47–57 Contents lists available at SciVerse ScienceDirect Annals of Nuclear Energy journal homepage: www.elsevier.com/locate/anucene