A simple dynamic measurement technique for comparing thermal insulation performances of anisotropic building materials

Transcription

1 Energy and Buildings 39 (2007) A simple dynamic measurement technique for comparing thermal insulation performances of anisotropic building materials Bulent Yesilata a, *, Paki Turgut b a Harran University, Mechanical Engineering Department, Sanliurfa, Turkey b Harran University, Civil Engineering Department, Sanliurfa, Turkey Received 29 September 2006; received in revised form 28 November 2006; accepted 30 November 2006 Abstract Measuring or estimating thermal properties of anisotropic building materials can be key obtaining the optimum performance for a particular application. The intensive researches on development of new building materials have necessitated in situ thermal testing apparatuses in most research laboratories. Only few standardized techniques are available for accurate thermal testing of anisotropic materials, and they are generally expensive. In the present study, common thermal testing methods are reviewed in brief. A simple and inexpensive thermal testing technique is proposed. The measurement is based on analysis of transient data, which is suitable for comparing effective thermal transmittances of both isotropic and anisotropic building materials. Sample measurements with ordinary concrete and rubberized concretes are performed. The effective thermal transmittances of rubberized concretes are found to be considerably lower than that of the ordinary one. # 2006 Elsevier B.V. All rights reserved. Keywords: Measurement methods; Thermal properties; Adiabatic-box technique; Anisotropy; Building composites 1. Introduction The measurement of thermal transport properties for a solid material is a key issue to attain optimum performance for a particular application. Over the years a number of measurement techniques have been developed for this purpose [1]. The earliest group of measurement techniques is the steady-state techniques. The technique is based on establishing a temperature gradient over a known thickness of a sample and controlling the heat flow from one side to the other. Steady-state techniques are primarily suitable for analyzing materials with low or average thermal conductivities at moderate temperatures [2]. The transient (dynamic) techniques measure temperature time response of the sample when a signal is sent out to create heat in the body. These methods can be used for measuring thermal diffusivity, thermal conductivity, or both, for broader range of temperatures and thermal properties [3]. A well-known transient method for thermal diffusivity is the Laser Flash [4 7]. A group of new apparatus known as Contact Transient Methods has recently * Corresponding author. Tel.: x1094; fax: addresses: byesilata@yahoo.com, byesilata@harran.edu.tr (B. Yesilata). become very attractive and popular for all types of materials since they can be used to measure several thermal properties simultaneously or separately [8 11]. The quick development of new and advanced materials for broad ranges of applications has necessitated the introduction of completely new techniques for producing reliable data for the growing demand. Part of this interest is a search for innovative building systems that combine higher efficiency and quality in the building process with improved thermal resistance. The most important aspects of innovation might be in the development of integrated insulation products [12]; such as the insulated, reinforced concretes [13], two or threewythe precast sandwich wall panels [14], and rubberized concretes [15]. An issue that arises out of this activity is the need to establish the thermal properties of the alternative systems and products. Accurate thermal characteristics are required to guide product development and manufacturing. Methods and data exist for dealing with the common building walls and insulations, but new systems and products are generally lacking such data. The information, to be generally useful in the building industry, needs to be in a form that is accurate, easily applied, and versatile enough to span the typical variations in the building configuration and the properties of the materials [13] /$ see front matter # 2006 Elsevier B.V. All rights reserved. doi: /j.enbuild

2 1028 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) It is often not enough to get approximate data from standard tables since small variations in composition, processing parameters and utilization conditions of materials change the behavior and properties. Anisotropy due to crystal structure, material type and form and method of fabrication can cause large variations in property depending on the heat flow direction within the material [16]. Recent parametrical studies for insulated, reinforced concretes [13] and for threewythe precast concretes [14], have successively demonstrated that these structural variables can have significant impacts on the thermal performance. Measurements of real samples of anisotropic building structures are thus necessary to support thermal design. As a consequence of the wide ranges of thermal property, a measurement method has to be selected depending on the following criteria: possible sample size and shape, temperature range, and thermal conductivity range [2 4]. In this paper, we review common steady-state and transient measurement techniques and introduce a new apparatus designed for thermal performance estimation of anisotropic building assemblies. 2. Short review of common measurement techniques 2.1. Steady-state measurement techniques The steady-state measurement techniques are based on establishing a temperature gradient over a known thickness of a sample to control the heat flow from one side to the other. A one-dimensional flow approach has been employed most frequently, but also other geometrical arrangements are used. The thermal conductivity is simply determined by measuring the temperature gradient and the heat flow through the sample [17 19]. Most commonly used steady-state measurement techniques are summarized in Appendix A and schematically shown in Fig. 1. The guarded hot plate apparatus [20] shown in Fig. 1(a) consists of uniformly wound heaters in a central metered section and in a thermally isolated guard area separated by a small coplanar gap. A number of temperature sensors are fitted tightly in all surfaces at appropriate positions in the central and guard sections. Measured dc-power is applied to the hot plate and the various temperatures in the cold plate and guard sections adjusted and carefully controlled to produce uniform temperatures at the specimen surfaces. The zero temperature difference across the gap and the desired temperature difference across the specimen pieces are required to obtain thermal conductivity. For the heat-flow meter technique [21], a square sample with a well-defined thickness is inserted between two plates as indicated in Fig. 1(b). The heat flow through the sample is measured with calibrated heat flow sensors after a fixed temperature gradient is established. For larger samples and higher thermal conductivity ranges guarded heat-flow meters can be used. The measurement principle remains nearly the same, but the test section is surrounded by a guard heater, resulting in higher measurement temperatures. The hot-box technique [22 24] is normally used measuring the overall heat transfer through large, inhomogeneous structures. The overall thermal resistance (R-value), which includes air film resistances in the cold and warm sides along with heat conduction resistance of the specimen, is obtained from such measurement. A large specimen is placed between a hot and a cold chamber operating at fixed temperatures, humidity and air flow conditions. In the guarded hot-box version [22], shown in Fig. 1(c), a guarded metering box is attached to the central section of the specimen. Temperature sensors are placed at positions approximately opposite those in the specimen to obtain the corresponding air temperatures. Testing is performed Fig. 1. Common steady-state measurement techniques: (a) guarded hot-plate, (b) basic heat flow-meter and (c) guarded hot-box.

3 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) establishing and maintaining a desired steady temperature difference across a test panel for a period of time so that constant heat flow and steady temperature are ensured. When air temperatures across the metering box wall are maintained the same, the heat interchange between the metering box and the guard box is zero. At this time, the heat flow (dc-power) is measured. This is a measure of heat in the metering box through a known area of the panel. In the calibrated hot-box version [23,24], the outer walls of the hot chamber are made with very thick insulation to minimize conduction losses and the power flow through those walls is measured for a range of hot chamber and laboratory temperatures, using the calibration panels. In addition to experimental measurement techniques described above, steady-state calculation methods (i.e. ASH- RAE Handbook methods and the finite element method) are also used for thermal performance evaluation of complex building structures. The theoretical methods are not described here, but the brief descriptions of those can be found in [14,25] Transient (dynamic) measurement techniques Transient measuring methods have become established in the last few decades for studying materials with high thermal conductivities and for taking measurements at high temperatures. Besides their high precision and broad measuring range, transient methods feature a comparably simple sample preparation and the ability to measure up to C. Transient methods can be divided into two categories depending on the apparatus used for measurements: contact and optical techniques. Contact Transient Methods have become popular and widely used because of their simplicity in concept and realization [8 11]. An appropriately sized rectangular or cylindrical specimen containing an embedded simple geometric form of a low heat capacity heat source together with one or more combined or separate temperature sensors is allowed to equilibrate at a given temperature [26]. The components of a basic apparatus along with the structure of heat source are shown in Fig. 2(a). A heat pulse or heat flux in the form of a step-wise function is produced by an electrical current in the source to generate a dynamic temperature field within the specimen. The temperature change with time (temperature response) is measured by a sensor(s) which is either unified with the heat source or placed a fixed distance from the source (temperature functions). The response is then analyzed in accordance with a model and set of solutions developed for the representative set-up and designed for the specific geometry and assumed boundary conditions. Depending Fig. 2. Schematics of common transient measurement techniques: (a) contact transient and (b) laser flash.

4 1030 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) upon the geometries of the specimen and source, and the method of the temperature field generation, one or more thermophysical properties can be obtained separately or simultaneously [8]. The common geometries and modes of heat sources for contact transient methods are summarized in Appendix B. Optical transient techniques require relatively more expensive set-ups and are quiet advanced. They cover broad ranges of both temperature and thermal conductivity as listed in Appendix C. The measurement principle is based on the generation and detection of energy pulse (in Laser Flash technique) or thermal waves (in the others). The thermal diffusivity of a material is directly measured, and therefore one needs to obtain a value for heat capacity from another measurement to finally calculate thermal conductivity [6]. Naturally, error propagation from the two measurements can lead to lower accuracy in the final result [1]. The most frequently used optical transient technique is the laser flash method [4 7], which is schematically shown in Fig. 2(b). The specimen is placed in a furnace and heated to a uniform temperature. A short (1 ms or less) pulse coming from a laser or a flash lamp irradiates one surface of the specimen. The resulting temperature rise on the rear surface (and/or the front surface in some cases) is measured either with a fixed thermocouple or more usually by an IR detector. The thermal diffusivity is calculated from the temperature versus time curve and the thickness of the sample. 3. The proposed dynamical measurement technique Most of the techniques described in preceding section have significant drawbacks for measuring effective thermal properties of anisotropic materials. Anisotropy due to crystal structure, material type and form and method of fabrication can cause large variations in property depending on the heat flow direction within the material. The sample geometry displays thermal variations in two perpendicular directions, which must be measured simultaneously. The contact transient techniques, especially the Gustafsson Probe or the Hot Disk, have recently been adapted for such a measurement [16]. The anisotropic building materials have relatively low effective thermal conductivity values; thus, sample size tends to be large resulting in longer measurement time [27]. The location of thermocouples and the quality of contact resistance between the thermocouple and the sample surface are also serious concerns for obtaining accurate measurement. Finding solutions to these drawbacks are relatively expensive [21]. A dynamic adiabatic-box technique proposed here is intended to overcome some difficulties mentioned above. It is a secondary Fig. 3. (a) Schematic of dynamic adiabatic-box apparatus (T: temperature sensor, RH: relative humidity sensor) and (b) dimensions of the adiabatic-box (in cm).

5 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) but simple and inexpensive technique. The technique can be used for pre-estimation purpose for materials developed in a laboratory before taking absolute measurements with one of the standardized techniques, if necessary. The schematic of the apparatus is illustrated in Fig. 3(a). The main component of the apparatus is the adiabatic-box, whose outer and bottom walls are heavily insulated to minimize heat losses to its surroundings. The test specimen with much higher thermal conductivity and thinner in size forms the top wall of the box to provide one-dimensional axial heat flow. The dimensions of the box are shown in Fig. 3(b). A heater controlled by a thermostat is installed close to the bottom wall to heat small depth of water to a certain temperature (ranges between 35 and 55 8C). Using water in the box helps to obtain more homogenous temperature distribution in lateral direction during heating and experiments. The evaporation of water is prevented due to adjusted low operating temperatures. The box is placed in a cold chamber operating at controlled fixed temperatures, humidity and air flow conditions. Temperature and relative humidity sensors are positioned at several points to measure the corresponding air conditions. A highly sensitive immersible temperature sensor with internal data-logger is put into water to monitor its temperature. The specimen is tightly installed as soon as water is heated to a desired temperature. The heater is then turned off and transient data recording starts at this time (t = 0). Cooling rate of water can be considered a measure of specimen thermal transmittance since major part of the heat is transferred through the specimen surface. The analysis of time temperature curve indeed allows estimation of overall heat transfer coefficient (between water and cold air). Measurements can be extended a variety of hot and cold space conditions, including effects of convection which could be a significant component of heat transfer particularly for roof specimens. 4. Sample measurements and evaluation procedure The technique proposed here is found to be more utilizable and free of many concerns when different anisotropic specimens are tested at purely identical hot and cold space conditions [28,29]. Simple analysis of time temperature curves for the specimens allows fair comparison of their effective thermal transmittances under specified conditions. The specimens used for sample measurements are the ordinary concrete and the concretes with scrap rubber. Geometric specifications of the specimens are given in Fig. 4. The rubberized concrete specimens used here can in principle be considered as fibrous composites and they are anisotropic at macroscopic level. The main object in these sample measurements is to investigate effect of scrap rubber addition on thermal transmittance of ordinary concrete. The subject is of practical interest in civil engineering community and there is lack of an appropriate thermal testing method due to complex structure of rubberized concretes (see [15,30] for recent reviews on the subject). Ordinary Portland cement (Type I, with density of 305 kg/ m 3 ), natural sand and water were used for preparation of fresh concrete. The mixture proportions by weight were respectively 1:2.75:0.5, and the largest diameter of fine-aggregate was 5 mm, well within the corresponding ASTM grading standard. The density and thickness of scrap rubber settled into fresh concrete were 0.84 g/cm 3 and 2 mm respectively. The rubber came from a local company which recycles scrap automobile tires into a sheet form. Strip and circular rubbers at desired dimensions were obtained by using a cutting and punch machines, respectively. Rubbers are settled into and lined up center surface of fresh concrete before pouring the rest of it to form the specimens illustrated in Fig. 4. Due to anisotropic future of specimens, three samples for each specimen were prepared. All specimens were cured for 10 days in a controlled environment before thermal tests were done. The chemical and mechanical tests of fresh and hardened concretes with and without rubber were accordingly performed. These data are not presented here due to being beyond scope of the present work. Thermal test results of the specimens are given in Fig. 5. The instantaneous temperatures given in the graph represent temperature of water in the box, whose top surface is covered Fig. 4. Geometric specifications of the specimens (from left to right: standard concrete, strip-rubberized concrete, and circular-rubberized concrete).

6 1032 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) by the corresponding specimens. The cooling rate of water is directly related with thermal transmittance of the specimen in question since major heat loss pass through from the specimen surface. Some heat losses at insignificant level could be possible from the other surfaces of the box; however, this does not affect comparison since all specimens are subject to the same internal and external conditions. Temperature curve for each specimen is the average of temperatures obtained from the three identical samples. Reproducibility of the experiments with identical specimens was fairly good (deviations within 3%). The instant value of cold room temperature was obtained by averaging instant temperatures taken at three different points in the room. The local variations in cold-air temperatures were all remained within 2 8C. The time-averaged value of instant temperatures in cold room is observed to be constant as shown with a solid trend line in Fig. 5(a). The time temperature curves for specimens show similar behavior at the beginning of the experiments. The temperatures sharply decrease first and then slow down. Significant differences exist in cooling rates of the specimens at intermediate times. The largest heat loss rate occurs for the standard concrete (no rubber case). The addition of strip rubber lowers the thermal transmittance of the concrete, or improves its insulation property. The best insulation improvement can be obtained with circular-rubberized concrete since the instant temperatures remain higher during the experimental-time. The equilibrium temperature could not be obtained although experiments last about 20 h. The dimensionless temperatures (u * ) for the specimens are thus defined as, u ¼ TðtÞ T o ¼ u (1) Tðt ¼ 0Þ T o u i where T(t) and T(t =0)=T i indicate, respectively, water temperatures at the beginning and at any instant time of the experiments, and T o represents the time-averaged temperature of the cold space. This definition is useful even for the specimens tested at different internal and external temperatures, preserving the same heat transfer mechanism. The variations of u * with time are shown in Fig. 5(b) for the specimens tested. The differences in thermal behaviors can now be observed more clearly. The experimental time can be considered long enough to make fair comparison since the decrease rate of the curves get nearly constant at the end of experiments. The physical meaning of the dimensionless temperature parameter is that the ratio of hot water energy at any time to that at the beginning of the experiment (available energy). The corresponding meaning can be expressed for the hot water with known mass (m) and specific heat (c p ) as, u ¼ Q ¼ mc pðt T o Þ Q i mc p ðt i T o Þ ¼ u : (2) u i Sum of the differences in u * values allows comparing thermal transmittances of two different specimens. The percentagewise difference can be calculated by the following equation, X ¼ Pt¼te t¼0 u t Pt¼te (3) t¼0 u t 2 where t e indicates the total experimental time, and the subscripts 1 and 2 correspond, respectively, rubberized concrete (strip or circular) and standard concrete (no rubber). This selection is useful for determining effect of rubber addition on thermal transmittance of the standard concrete. The values of X for strip-rubberized concrete and circularrubberized concrete are found to be 8% and 14.7%, respectively. 5. Concluding remarks Fig. 5. The transient temperatures: (a) in dimensional form and (b) in dimensionless form. The dynamic adiabatic technique described here is found to be functional and robust for first estimation of thermal insulation performances of complex structured (i.e. inhomogeneous, anisotropic, layered) flat specimens. The thermal conditions in hot and cold spaces can easily be arranged and changed to examine various internal and external environment effects on thermal behavior. Although testing time is quiet long

7 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) due to low effective thermal conductivity of specimens, maintaining the required cold space conditions during this time are easy. The thermal insulation performances of rubberized concrete specimens are compared with that of the ordinary one to demonstrate the measurement and evaluation procedures of the technique. It is found that the insulation performance is improved as much as 14.7% by addition of circular rubber matrix into the ordinary concrete. More work and tests are necessary to increase the utilization and standardization of the proposed technique. Acknowledgements This work was elaborated with the support of the Research Project funded by Turkish Scientific and Technical Research Institute (under grant TUBITAK-MAG-105M021). Appendix A Short description of common steady-state measurement techniques Measurement technique Material type Temperature range (8C) Property range (W/m K) 1. Guarded hot-plate (flat or cylinder) Homogeneous composites; 180 to to 2 insulation materials 2. Heat-flow meter (basic or guarded) Insulation materials 100 to to Hot-box apparatus (guarded or calibrated) 20 to 40 Collected information from [2,20 22]. Building assemblies containing insulation, wood, masonry, glass and other materials Thermal conductance range of (m 2 K)/W Appendix B. Short description of common contact transient measurement techniques Appendix C. Short description of optical transient measurement techniques Measurement technique Material type Temperature range (8C) Property range (W/m K) 1. Laser flash method Metals; polymers; ceramics 2. Angstrom method Metals; alloys; (a) basic and (b) diamond; modified semiconductors; ceramics and polymers; multi-layered composites 3. Modulated beam Metals; polymers; technique ceramics 4. Photothermal methods Small specimens of most solid materials 100 to to to 1300 Above to to to to 200 Collected information from [4,7,21]. References Collected information from [3,21,26]. [1] M.J. Assael, M. Dix, K. Gialou, L. Vozar, W.A. Wakeham, Application of the transient hot-wire technique to the measurement of the thermal conductivity of solids, International Journal of Thermophysics 23 (2002) [2] E.R.G. Eckert, R.J. Goldstein, Measurements in Heat Transfer, Hemisphere Publishing, Washington, [3] L. Kubicar, Pulse Method of Measuring Basic Thermophysical Parameters, Elsevier Publishing, Amsterdam, [4] J.W. Parker, J.R. Jenkins, P.C. Butler, G.I. Abbott, Flash method of determining thermal diffusivity, heat capacity and thermal conductivity, Journal of Applied Physics 32 (1961) [5] G. Brauer, L. Dusza, B. Schulz, New laser flash equipment LFA 427, Interceramics 41 (1992)

8 1034 B. Yesilata, P. Turgut / Energy and Buildings 39 (2007) [6] J. Xue, R. Taylor, An evaluation of specific heat measurement methods using the laser flash technique, International Journal of Thermophysics 14 (1993) [7] L. Torgrim, J.T. Barrett, Simple and inexpensive flash technique for determining thermal diffusivity of ceramics, Journal of the American Ceramic Society 74 (1991) [8] S.E. Gustafsson, Transient plane source technique for thermal conductivity and thermal diffusivity measurements of solid materials, Review of Scientific Instruments 62 (1991) [9] S.E. Gustafsson, M.A. Chohan, K. Ahmed, A. Maqsood, Thermal properties of thin insulating layers using pulse transient hot strip measurements, Journal of Applied Physics 55 (1984) [10] S.E. Gustafsson, T. Long, Transient plane source (TPS) technique for measuring thermal properties of building materials, Fire and Material 19 (1995) [11] L. Kubicar, C. Bohac, A step-wise method for measuring thermophysical parameters of materials, Measurement Science and Technology 11 (2000) [12] A.M. Papadopoulos, State of the art in thermal insulation materials and aims for future developments, Energy and Buildings 37 (2005) [13] G.F. Jones, R.W. Jones, Steady-state heat transfer in an insulated, reinforced concrete wall: theory, numerical simulations, and experiments, Energy and Buildings 29 (1999) [14] B.J. Lee, S. Pessiki, Thermal performance evaluation of precast concrete three-wythe sandwich wall panels, Energy and Buildings 38 (2006) [15] M. Nehdi, A. Khan, Cementitious composites containing recycled tire rubber: an overview of engineering properties and potential applications, Cement, Concrete, and Aggregates 23 (2001) [16] D. Lundström, B. Karlsson, M. Gustavsson, Anisotropy in thermal transport properties of cast g-tial alloys, Zeitschrift fur MetaIlkunde 92 (2001) [17] J.R. Mumaw, Heat transmission measurements in thermal insulations, in: R.P. Tye (Ed.), ASTM STP 544, ASTM International, West Conshohocken, PA, 1974, pp [18] J.R. Mumaw, Thermal insulation performance, in: D.L. McElroy, R.P. Tye (Eds.), ASTM STP 718, ASTM International, West Conshohocken, PA, 1980, pp [19] R. Robert, Zarr, A history of testing heat insulators at the national institute of standards and technology, ASHRAE Transactions 107 (2001) 11. [20] American Society for Testing and Materials, ASTM C : Standard test method for steady-state heat flux measurements and thermal transmission properties by means of the guarded-hot-plate apparatus, Annual Book of ASTM Standards, vol , West Conshohocken, PA, [21] The Virtual Institute for Thermal Metrology (EVITHERM), Common thermal conductivity or thermal diffusivity measurement methods, accessed at on September 15, [22] American Society for Testing and Materials, ASTM C236-89: Standard test method for steady-state thermal performance of building assemblies by means of a guarded hot box, Annual Book of ASTM Standards, vol , West Conshohocken, PA, [23] P.R. Achenbach, Design of a calibrated hot-box for measuring the heat, air, and moisture transfer of composite building walls, in: Thermal performance of the exterior envelopes of buildings, Proceedings-1, 1981, pp [24] R.R. Zarr, D.M. Burch, T.K. Faison, C.E. Arnold, Thermal resistance measurements of well-insulated and superinsulated residential walls using a calibrated hot box, ASHRAE Transactions 92 (1986) [25] D.M. Burch, B.A. Licitra, D.F. Ebberts, R.R. Zarr, Thermal resistance measurements and calculations of an insulated concrete block wall, ASHRAE Transactions 95 (1989) [26] The UK s National Measurement Laboratory (NPL), Standards for Contact Transient Measurements of Thermal Properties, accessed at on September 27, [27] A.A. Abdou, I.M. Budaiwi, Comparison of thermal conductivity measurements of building insulation materials under various operating temperatures, Journal of Building Physics 29 (2005) [28] B. Yesilata, P. Turgut, Insulation property of rubberized mortar, Journal of Polytechnic 8 (2005) (in Turkish). [29] B. Yeşilata, P. Turgut, Y. Işiker, Experimental investigation on insulation properties of polymeric-waste-added concretes, Journal of Thermal Sciences and Technology 26 (2006) (in Turkish). [30] R. Siddique, T.R. Naik, Properties of concrete containing scrap-tire rubber an overview, Waste Management 24 (2004)