Frost Damage of Roof Tiles in Relatively Warm Areas in Japan Influence of Surface Finish on Water Penetration Chiemi IBA Hokkaido Research Organization, Japan Shuichi HOKOI Kyoto University, Japan
INTRODUCTION Moisture in building materials significantly influences the durability of building envelopes. Accumulated moisture occasionally freezes in the building materials, severely damaging them. Frost damage has been observed not only in cold regions but also in relatively warm areas in Japan. mainly in roof tiles due to night-sky radiation.
From Previous Field surveys INTRODUCTION (continued) Minor spallings and flakings are the typical examples of frost damage in roof tiles in warm areas. flaking crack Minor circular spallings From Standard freezing thawing test Test specimens are fully saturated with water before the freezing. The method is not suitable for evaluating the actual damage to the roof tiles. larger and more serious cracks
INTRODUCTION (continued) The purpose of our research is to clarify the mechanisms and causes of frost damage in roof tiles in relatively warm area. In this paper Review of the previous experiments Water penetration experiment to examine moisture distribution in tiles A new freezing thawing test to propose more suitable methods Numerical analysis of the freezing thawing process The analysis does not correspond to the above experiment s conditions. Focus on the influence of surface finish and moisture permeability
REVIEW OF PREVIOUS EXPERIMENTS
REVIEW OF PREVIOUS EXPERIMENTS Water Penetration Experiments In Outdoor Environment Weather: cloudy with intermittent rain. In Laboratory Water replenishment Wet towel after 26 h after 9 h The area that absorbed water is clearly distinguished. The water content distribution is not uniform even in a single tile.
REVIEW OF PREVIOUS EXPERIMENTS Water Penetration Experiments Summary of water penetration experiments Japanese roof tiles generally have a surface finish to prevent rain water penetration. Although the water resistance of a surface finish is much higher than that of the tile body, a small amount of water could penetrate through the finish. The water distribution does not necessarily correspond to small visible cracks or scratches on the surface finish. The water may penetrate through invisible pinholes on the surface. The water content increases in particular small areas and non-uniform moisture distribution is formed.
REVIEW OF PREVIOUS EXPERIMENTS New Freezing-Thawing Test: Four types of water supply Simulate actual water supply: Partially immersed in water Small droplets Surface condensation Try to cause damages similar to those found in actual environment Test apparatus based on RILEM TC 176, Test Methods of Frost Resistance of Concrete +20 4h 3h 4h 1h -20 1 cycle = 12 h Temperature change in antifreeze solution Time The amount of water supplied increases successively from Pattern 1 to 4.
REVIEW OF PREVIOUS EXPERIMENTS New Freezing-Thawing Test: Results (Pattern 3&4) Appearance of the specimens after 56 cycles Spalling near the edge Pattern 3: Absorbed water (lower side) Pattern 4: Dropped and Absorbed water Spalling near the edge Upper surface Upper surface Lower surface Summary of freezing-thawing test Lower surface Many fine flakings Similar damage to that found under actual conditions was observed. If droplets continuously fall on the specimens, even small droplets can penetrate the tile and can cause damage.
REVIEW OF PREVIOUS EXPERIMENTS Supposition from Experimental Results 1. Invisible pinholes on the surface finish cause high water content in particular small areas (maybe near the surface). 2. When the temperature of the tile decreases below the freezing point, freezing occurs only in high water content area. 3. Small spallings or fine flakings are caused only in the area where the ice content is sufficiently high. To examine supposition 1 & 2, following analysis is conducted.
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Basic Equations The equations for simultaneous heat and moisture transfer, which consider freezing and thawing (Matsumoto et al. 1993) Moisture balance: Energy balance: Freezing condition: ρ lψ l t = ( λ T ) + {( λ + λ ) µ } Tg µ g µ l ρ iψ i t ( λ T ) + H { ( λ T ) + ( λ µ )} T cρψ = gl Tg µ g t µ = H log Unsaturated water permeability: li e T T o + H li ρiψ i t The amount of ice formation is determined according to the heat and moisture transfer balance. λ μl =A ψ l + (λ μl,sat A ψ l,sat ) (ψ l /ψ l,sat ) N where T = absolute temperature, K, T 0 = freezing temperature of free water (=273.16), K, t = time, s, λ = thermal conductivity, W/(m K), ρ = density, kg/m 3, c = specific heat, J/(kg K), λ μ = moisture conductivity by water chemical potential difference, kg/(m s J/kg) λ T = moisture conductivity by temperature difference, kg/(m s K) ψ = moisture content, m 3 /m 3, μ = water chemical potential (free water standard), J/kg λ μl = Unsaturated water permeability, kg/m s (J/kg), A = Linear factor,n = Exponential factor Subscript w = water; s = solid; g = gas; l = liquid; i = ice, sat = saturated
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Calculation Model Typical roof installed with roof tiles The undersurface of the roof tile is exposed to an air layer. A part of a tile Outdoor Environment 0.1mm Pinhole on finish (1/400 of the output area) Tile specimen 16 mm Output area (40 mm) Air layer = Outdoor air Y=0 X axis Calculation area 20 mm Upper side Y axis Lower side surface finish
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Relation between Water content and Freezing temperature Absorption Isotherm Nearly saturated, it can freeze at a temperature slightly below 0 C.
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Environmental Conditions Colder conditions of a warm area in Japan are assumed. Solar radiation and night-sky radiation is not considered. Temperature [ C] Humidity ratio [g/kg'] 11 10 9 8 7 6 5 4 3 2 1 0-1 -2-3 Temperature Precipitation 0.1[mm/h] Small precipitation for 3 hours During rain, the surface is assumed to be saturated with water Relative humidity Humidity ratio 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Elapsed time [h] 100 90 80 70 60 50 40 30 20 10 0 Relative humidity [%RH] Temperature drop after rain
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Calculation Patterns To examine the influence of Surface finish Moisture permeability on moisture the distribution Pattern No. Surface finish Parameter A Note a-1 Both surfaces 5.0 10-10 a-2 Both surfaces 2.5 10-9 A: 5 of a-1 b-1 Only upper surface 5.0 10-10 b-2 Only upper surface 2.5 10-9 A: 5 of b-1 a-1,a-2 Surface finish b-1,b-2 Unsaturated water permeability: λ μl =A ψ l + (λ μl,sat A ψ l,sat ) (ψ l /ψ l,sat ) N Surface finish Unfinished surface Moisture diffusion in the tile body becomes faster as a parameter A increases.
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Results: Time profile for Pattern a-1 Temperature [ C] Ice content [vol.%] 11 10 9 8 7 6 5 4 3 2 1 0-1 -2-3 30 25 20 15 10 5 0 Outdoor air Freezing process Upper surface 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Elapsed time [h] Temperature Distance from upper surface 0 mm 2 mm 4 mm 5.2 mm 5.6 mm 6 mm 7 mm 8 mm 16 mm 2 mm 4 mm 5.2 mm Ice content Thawing process 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Elapsed time [h] Freezing thawing process 5.6 mm 6 mm 7 mm Liquid water content [vol.%] Liquid water content 30 25 20 15 10 5 0 Upper surface 5.6 mm 2 mm 4 mm 5.2 mm 6 mm 7 mm Distance from upper surface 0 mm 2 mm 4 mm 5.2 mm 5.6 mm 6 mm 7 mm 8 mm 16 mm 8 mm Freezing thawing process Undersurface 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Elapsed time [h] Water content rapidly increases in the area near the surface during rain. When the temperature drops below the freezing point, water begins to freeze only in high water content area and ice content increases.
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Results: Change in moisture distribution (before freezing) a-1 Y=0 Line Moisture diffusion in the tile body slower faster a-2 Y=0 Line
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Liquid water content distribution before freezing (Y=0) Liquid water content [vol.%] Liquid water content [vol.%] 26 24 22 20 18 16 14 12 10 8 6 4 2 0 26 24 22 20 18 16 14 12 10 8 6 4 2 0 Soon after precipitation Just before freezing 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Distance from upper surface [mm] Pattern a-1 Soon after precipitation Just before freezing Pattern a-2 0h 1h 3h 4h 13h 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Distance from upper surface [mm] 0h 1h 3h 4h 13h Ice content distribution during the freezing process (Y=0) Pattern a-1 Pattern a-2 Liquid water content [vol.%] Liquid water content [vol.%] 26 24 22 20 18 16 14 12 10 8 6 4 2 0 26 24 22 20 18 16 14 12 10 8 6 4 2 0 13h 13h20m 13h40m 14h 14h20m 14h40m 15h 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Distance from upper surface [mm] 13h 13h20m 13h40m 14h 14h20m 14h40m 15h 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Distance from upper surface [mm]
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Summary of Numerical Analysis Even a small pinhole can allow significant amounts of water to penetrate into the tile body. If such pinholes are randomly distributed on the surface finish, the moisture content will show non-uniform distribution. If freezing occurs in the nearly saturated zone, the expansion pressure could damage the tile body. Not only expansion pressure of ice but also unfrozen water pressure is supposed to cause the destructive effect. Random moisture content distribution could be one reason for small spallings or flaking in an actual situation.
CONCLUSIONS In this study, the characteristics and causes of frost damage in roof tiles were investigated through water penetration experiments, a freezing thawing test, and a numerical analysis. When the water content inside a tile increases in particularly small area and the temperature decreases below the freezing point, freezing can occur only in the areas having high water content, causing minor spalling. When the water permeability is small, high local water content tended to occur, that increases the risk of frost damage. This information about hygrothermal properties of the surface finish and the tile body will help in material development and advancement in the future.
THANK YOU FOR YOUR ATTENTION!
CONTENTS 1. INTRODUCTION 2. REVIEW OF PREVIOUS EXPERIMENTS 2.1 Water Penetration Experiments 2.2 The New Freezing Thawing Test 3. NUMERICAL ANALYSIS OF THE FREEZING THAWING PROCESS 3.1 Basic Equations 3.2 Calculation Model and Material Properties 3.3 Calculation Conditions 3.4. Results 4. CONCLUSIONS 3.4. Results
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Material Properties Tile body Thermal conductivity (dry condition) 0.937 W/m K Vapor permeability 4.139 10-12 kg/m s Pa Water permeability (saturated condition) 3.66 m/s Specific heat 920 J/kg K Density 2100 kg/m 3 Porosity (= Maximum moisture content) 26.2 % Absorption isotherm Surface finish Thermal resistance Not considered Vapor resistance 5.35 10 9 m 2 s Pa/kg Water resistance 9.92 10 9 m 2 s Pa/kg Surface finish Pinhole Pinhole or Unfinished surface Vapor resistance (pinhole or unfinished surface) 2.42 10 6 m 2 s Pa/kg Water resistance (pinhole) 37015.0 m 2 s Pa/kg Water resistance (unfinished surface) 2677.8 m 2 s Pa/kg Unfinished surface
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Calculation Conditions Environmental conditions Colder conditions of a warm area in Japan is assumed. Solar radiation and night-sky radiation is not considered. Boundary conditions During rain Saturation Upper side (μ = 0.1) Pinhole on finish Temperature [ C] Humidity ratio [g/kg'] 11 10 9 8 7 6 5 4 3 2 1 0-1 -2-3 20 Precipitation 0.1[mm/h] 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Elapsed time [h] Upper side Lower side Temperature Relative humidity Humidity ratio Heat transfer coefficients Convective heat transfer coefficient 18.60 W/m 2 K Radiative heat transfer coefficient 4.65* W/m 2 K Convective heat transfer coefficient 2.30 W/m 2 K Radiative heat transfer coefficient 100 90 80 70 60 50 40 30 Relative humidity [%RH] 4.65* W/m 2 K Tile specimen Lower side *Design value of indoor heat transfer, assuming emissivity of material to be 0.9
NUMERICAL ANALYSIS OF FREEZING THAWING PROCESS Calculation Patterns To examine the influence of Surface finish Moisture permeability on moisture the distribution Pattern No. Surface finish Parameter A Note a-1 Both surfaces 5.0 10-10 a-2 Both surfaces 2.5 10-9 A: 5 of a-1 b-1 Only upper surface 5.0 10-10 b-2 Only upper surface 2.5 10-9 A: 5 of b-1 a-1,a-2 Surface finish b-1,b-2 Unsaturated water permeability: λ μl =A ψ l + (λ μl,sat A ψ l,sat ) (ψ l /ψ l,sat ) N Surface finish Unfinished surface Moisture diffusion in the tile body becomes faster as a parameter A increases.