1 Case Study Frost Action Mechanisms of Clay Roofing Tiles: 2 Case Study 3 J. Ranogajec 1 ; P. Kojic ´ 2 ; O. Rudic ´ 3 ; V. Ducman 4 ; and M. Radeka 5 4 Abstract: Clay roofing tiles fired at five different temperatures (900°C, 920°C, 960°C, 1,000°C, 1,020°C) were subjected to the investi- 5 gations of textural characteristics and frost resistance prediction. The closed container and hydraulic pressure mechanisms proved to be 6 dominant at lower firing temperatures (900°C, 920°C), whereas micro-ice-lens formation mechanism have significant role at higher temper- 7 atures (960°C, 1,000°C). The highest resistance is noticed for the samples fired at 1,020°C where the frost action mechanisms are balanced 8 because the porous structure that compensates the local stresses developed during freezing. The prediction of frost action durability of clay 9 roofing tiles, the appearance of the first cracks, and the prediction of frost action mechanisms given by the statistical model, showed a high 10 level of agreement. The statistical model contains capillary pores (P 0.1 ), and the ratio of frozen and unfrozen water (I ) as significant para- 11 meters for describing susceptibility to closed container mechanism and micro-ice-lens formation mechanism, respectively. DOI: 10.1061/ 12 (ASCE)MT.1943-5533.0000500. © 2012 American Society of Civil Engineers. 13 CE Database subject headings: Materials engineering; Porous media; Frost; Optimization; Microstructure; Regression analysis. 14 Author keywords: Materials engineering; Porous media; Frost; Optimization; Microstructure; Regression analysis. 15 Introduction 16 Materials based on fired clay, such as bricks or clay roofing tiles, 17 have been used for building for hundreds of years because of their 18 good production flexibility, good mechanical behavior, and good 19 resistance to climatic conditions. The formation of a microstructure 20 with desirable characteristics is considered to influence the frost 21 resistance proprieties of masonry materials (Junge 2000; Vojnic 22 et al. 2002). Porous materials (fired clay materials, concrete, 23 etc.) are in contact with water and thus susceptible to frost action. 24 The durability of clay roofing tiles, as the ability to withstand se- 25 vere outside climatic conditions is the most important requirement 26 that has to be considered in a structural design of the buildings 27 (Lisø et al. 2007). Clay roofing tiles resistance toward frost action 28 is a very complicated process because of complex relationships 29 among raw materials characteristics, production parameters, and 30 properties of the final products. The total volume of pores and pore 31 size distribution were found to be the most important parameters to 32 obtain appropriate properties of clay roofing tiles, resistible to frost 33 action (Sveda 2001; Carretero et al. 2002). Considering their 34 susceptibility to frost action, Ravaglioli (1976) reported the 35 significance of pore size interval within 0.25–1.4 μm. The extent 36 of damages caused by frost action is strictly dependent on the ex- 37 posed surface area, pore size distribution (Sveda 1999; Nakamura 38 1988), and on the water saturation degree of the material (Hoffman 39 and Niesel 1988). 40 For prediction of frost resistance characteristics of construction 41 materials many standardized methods, which simulate natural 42 conditions, have been developed. There are also indirect methods 43 (Arnott 1990; Franke and Bentrup 1993a, 1993b; Koroth et al. 44 1988; Maage 1990a, 1990b; Robinson 1995) that connect the 45 frost resistance with technological properties (water absorption 46 in different experimental conditions, capillary coefficient, satura- 47 tion coefficient) and with some microstructural properties (cumu- 48 lative volume of pores, median pore size, size and internal specific 49 surface of pores) of building materials. 50 One of the indirect methods of wide recognition of frost 51 resistance prediction (Maage 1990a, 1990b), used in this work 52 is Maage’ s factor (DF), given by Eq. 1. The total volume of pores 53 (PV) and the share of pores at certain pore diameter greater than 54 3 μm(P3) are the basic parameters of this prediction method. DF ¼ 3.2 PV þ 2.4 · P3 (1) 55 On the basis of Maage’ s factor (DF), the following classification 56 was proposed: DF > 70, high probability for the material to 57 be frost resistant in severe climatic conditions; 55 < DF < 70, 58 uncertain zone of the frost resistance; and DF < 55, low proba- 59 bility for the material to be frost resistant. 60 The ASTM standard (ASTM C 1167-03 2009) is another 61 indirect method that is fast and simple for the execution and that 62 provides a prediction of frost resistance through the saturation 63 factor S. The classification of resistance to freezing-thawing proc- 64 esses is as follows: S < 0.74, high probability for the material to be 65 frost resistant in severe climatic conditions; 0.74 < S < 0.84, 66 uncertain zone of frost resistance; and S < 0.84, low probability 67 for the material to be frost resistant. 1 Full Professor, Univ. of Novi Sad, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia (Corresponding author). E-mail: janjar@uns.ac.rs 2 Research Assistant, Univ. of Novi Sad, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia. E-mail: kojicpredrag@gmail.com 3 Research Assistant, Univ. of Novi Sad, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia. E-mail: ognjenrudic@yahoo.com 4 Head of the Laboratory, Slovenian National Building and Civil Engineering Institute, Dimičeva 12, SI-1000 Ljubljana, Slovenia. E-mail: keramika@zag.si 5 Professor, Univ. of Novi Sad, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia. E-mail: mirka@uns.ac.rs Note. This manuscript was submitted on September 20, 2011; approved on February 12, 2012; published online on February 13, 2012. Discussion period open until February 1, 2013; separate discussions must be submitted for individual papers. This paper is part of the Journal of Materials in Civil Engineering, Vol. 24, No. 9, September 1, 2012. © ASCE, ISSN 1076- 0431/2012/9-0-0/$25.00. JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / SEPTEMBER 2012 / 1