99

FEBRUARY 2005

99

Heat and Moisture Transfer with Sorption and Phase Change Through Clothing Assemblies Part I: Experimental Investigation JINTU FAN1

AND

XIAO-YIN CHENG

Institute of Textiles and Clothing, Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong ABSTRACT Part I of this two-part series reports on an experimental investigation of heat and moisture transfer through clothing assemblies consisting of porous fibrous battings sandwiched by inner and outer layers of a thin covering fabric. The experiments are conducted on a novel sweating, guarded hot plate at ⫺20 °C. The temperature and water content distribution within the porous fibrous battings are obtained for four combinations of two kinds of fibrous battings (polyester and viscose) and two kinds of covering fabrics (one highly permeable nylon and the other a less permeable laminate. Most of the changes in temperature distribution take place within 30 minutes of the tests, and moisture absorption by the hygroscopic viscose fibers affects the temperature distribution. The water content accumulates with time, and water content is higher at the outer regions than at the inner regions of the battings. The accumulation and distribution of water content is a combined result of moisture absorption, condensation, and liquid water movement. The experimental findings form a basis for the development of a theoretical model to be reported in Part II of the series.

In a cold environment, perspiration from a human body may condense within clothing, and the re-evaporation of condensates can cause serious “after-chill” discomfort [8, 15]. To prevent the detrimental effects of condensation, heat and moisture transfer within clothing assemblies should be better understood. Coupled heat and moisture transfer with phase changes in fibrous insulation was first considered by Henry in his theoretical model in 1940s [9]. However, little further progress was made until 1980s, when a number of theoretical models were proposed [1, 2, 4 – 8, 11, 13, 14, 16, 17, 19, 20] due to growing interest in textile science and other engineering fields such as civil engineering and energy conservation. These recent theoretical models have not been fully validated, however, due to the lack of experimental data. In this series, we conduct experiments of heat and moisture transfer through clothing assemblies consisting of porous fibrous battings sandwiched by inner and outer layers of thin covering fabrics on a novel sweating, guarded hot plate in freezing conditions (⫺20 °C). Based on the experimental findings, we develop a new theoret-

1 Corresponding author: tel. ⫹852-2766 6472, fax ⫹852-2773 1432, email: [email protected]

Textile Res. J. 75(2), 99 –105 (2005)

ical model. For the first time, our new model considers moisture movement induced by partial water vapor pressure, a super saturation state in the condensing region, as well as the dynamic moisture absorption of fibrous materials and the movement of liquid condensates. We compare the results of the new model and find good agreement with experimental results. Our work is now presented in this two-part series. Part I reports the experimental investigation, and Part II discusses the theoretical model, a comparison with experimental results, and the results of a computational simulation of the new model.

Review of Past Experimental Work Past experimental work on coupled heat and moisture transfer with phase change involved some very specific conditions. Thomas et al. [18] monitored the temperature change and moisture migration in an initially wet glassfiber insulation slab with impermeable boundaries subjected to one-dimensional temperature gradients on a guarded hot plate. Farnworth [8] reported the use of a sweating hot plate, where water was fed to the hot side of the fibrous insulation with a syringe pump. Shapiro and Motakef [14] reported an experiment in which liquid water was introduced close to the “hot” side of the 0040-5175/$15.00

100

TEXTILE RESEARCH JOURNAL

sample, evaporated there, and re-condensed toward the “cold” impermeable side of the slab. Wijeysundera et al. [21] reported two series of experiments. In the first series, water was sprayed on the hot face of the slab and, in the second series, the hot face of the insulation was directly exposed to a moist airflow. Transient temperature changes were monitored, and the total amount of moisture absorption/condensation after a period of time was measured. An experiment similar to Wijeysundera’s second series was conducted by Tao et al. [17], except that the cold side was subjected to the temperature below the triple point of water. Murata [12] conducted an experiment in which the hot side of a fibrous insulation slab was exposed to hot, moist air with a temperature close to 100°C and the cold side was in contact with an impermeable glass plate. The falling of condensate under gravity and convective heat transfer was an important phenomenon. In these past experimental investigations, moisture was introduced to the system by either spraying liquid water [14, 17, 18] onto the fibrous insulation or exposing the hot side of the insulation directly to moist airflow [8, 12, 17]. This, however, does not resemble many practical

situations where the fibrous insulation is sandwiched by two layers of moisture retarders, such as the case with clothing and building insulation. Recently, we (Fan et al. [3]) used a sweating, guarded hot plate to investigate coupled heat and moisture transfer through fibrous insulation sandwiched by two covering fabrics under low temperature conditions. However, we did not compare the effects of the fibrous insulation and covering fabrics or analyze them in view of clothing comfort.

Experimental INSTRUMENTATION The device for the experiments is shown schematically in Figure 1, modified from the one specified in ISO 11092:1993(E) to be used in freezing conditions. It has a shallow water container (1) with a porous plate (3) at the top. The container is covered by a manmade skin (2) made of a waterproof but moisture permeable breathable fabric. The edge of the breathable fabric is sealed within the container so that there is no water leakage. Water is supplied to the container from a water tank (16) through

FIGURE 1. Schematic design of the instrument.

FEBRUARY 2005

101

an insulated pipe (14). The water in the tank is pre-heated to 35°C, and the water level is checked frequently to ensure it is higher than the manmade skin, so that water is in full contact with the manmade skin at the top of the container. The water temperature in the container (1) is controlled at 33°C, simulating human skin temperature. To prevent heat loss from directions other than the upper right, the water container is surrounded by a guard with a heating element (13). The temperature of the guard is controlled so that its difference from that of the bottom of the container is less than 0.2°C. In order to prevent the air trapped underneath the manmade skin and the inflation of the manmade skin due to the water pressure underneath, the device is fitted with a ventilation tube (22) to ensure there is no air underneath the manmade skin, and an upper porous plate (23) above the manmade skin to keep it flat. The whole device is further covered by thick insulation foam. SAMPLES The samples tested in the experiments consisted of several thin layers of fibrous battings sandwiched by top and under layers of a covering fabric, simulating the construction of a “down” jacket. During testing, the samples were placed on top of the manmade skin, as shown in Figure 1. Two kinds of covering fabrics, varying in permeability and construction, were used in the investigation, and their specifications and properties are listed in Table I. Two fibrous battings, one hygroscopic viscose and one nonhygroscopic polyester, were used, the details of which are given in Table II. The resistance to air penetration was tested on a KES -F8-AP1 air permeability tester [10]. The thermal resistances of the covering fabrics were determined first by measuring the total thermal resistances when placing one, two, three, four, and five layers of a covering fabric on the guarded hot plate containing no water in an air-conditioned room, second by plotting the total thermal resistance against the number of layers, and finally by obtaining the slope of the linear regression line. The slope was then the thermal TABLE I. Properties of the covering fabric. Nylon fabric Construction

0.108 2.73E-04 3.15E-02 64.99 0.524

woven ⫹ membrane ⫹ warp knit 0.22 5.15E-04 3.16E-02 143.79 145.1

5.21E-07

3.55E-9

woven 2

Weight, kg/m Thickness, m Thermal resistance, Km2/W Water vapor resistance, s/m Resistance to air penetration, kPa.s/m Air permeability, m2/Pa.s

Three-layer laminated fabric

TABLE II. Properties of inner fibrous batting. Composition 2

Weight, kg/m Thickness, m Fiber density, kg/m3 Porosity Resistance to air penetration, kPa.s/m Air permeability, m2/Pa.s

Viscose

Polyester

0.145 1.94E-03 1.53E⫹03 0.951 0.062 3.11E-05

0.051 4.92E-03 1.39E⫹03 0.993 0.0061 8.06E-04

resistance of a single layer of the covering fabric. Similarly, the water vapor resistance was determined by measuring the total water vapor resistance by placing one, two, three, four, and five layers of a covering fabric on the guarded hot plate containing water in an airconditioned room, then plotting the total water vapor resistance against the number of layers of the covering fabric, and obtaining the slope of the linear regression line of the plot, which was the water vapor resistance of a single layer of the covering fabric. Since the thickness of a single layer of polyester batting is very different from a single layer of viscose batting, in order to make the total thickness of the plies of polyester battings close to the total thickness of the viscose batting plies for comparative reasons, six plies of polyester batting or fifteen plies of viscose batting (making a total thickness equal to 29 mm) were plied together for testing.

Experimental Procedure When measuring moisture absorption and condensation under cold conditions, we used the following procedure: 1. Condition the top and bottom layer of the covering fabric and the fibrous battings in an air-conditioned room with the temperature at 20.0 ⫾ 0.5°C and humidity at 65 ⫾ 5% for at least 24 hours. 2. Place the sweating, guarded hot plate in the airconditioned room with the temperature at 20.0 ⫾ 0.5°C and humidity at 65 ⫾ 5%. Start the temperature control and measurement system and wait until the temperature and power supply are stabilized. 3. Weigh and record the weights of each layer of fibrous batting. 4. Sandwich multiple layers of fibrous battings with the top and bottom layers of covering fabric and place the ensemble on top of the manmade skin of the instrument. 5. Place the sweating, guarded hot plate in a cold chamber with the temperature controlled at ⫺20 ⫾ 1°C. 6. Record the temperatures and power supply with time.

102

TEXTILE RESEARCH JOURNAL

7. After a pre-set time (i.e., 8, 16, or 24 hours), take out each layer of fibrous batting and weigh immediately on an electronic balance. 8. Calculate the percentage of moisture or water accumulation due to absorption or condensation on each layer of fibrous battings by Wc i ⫽

W ai ⫺ W oi ⫻ 100% W oi

,

(1)

where Wai is the weight of the ith layer of fibrous batting after testing on the sweating, guarded hot plate for a pre-set period of time, Woi is the weight of the ith layer of fibrous batting before testing on the sweating, guarded hot plate, and Wci is the increased percent water content.

Results and Discussion In Figures 2 to 5, 8, and 9, x⬘ is the dimensionless distance from the inner layer of the covering fabric and is calculated by x⬘ ⫽

distance total thickness of batting

.

FIGURE 3. Temperature distribution for six plies of polyester batting sandwiched by two layers of a laminated fabric.

(2)

TEMPERATURE DISTRIBUTION The temperature distributions within the fibrous battings are plotted against the distance from the bottom layer of the covering fabric in Figures 2 to 5 for two kinds of fibrous battings and two covering fabrics. In general, the temperature of the inner battings next to the warm “skin” increases quickly in the first few minutes and may even exceed the skin temperature of 35°C before it drops to a stabilized value. The outer battings

FIGURE 4. Temperature distribution for fifteen plies of viscose batting sandwiched by two layers of a nylon fabric.

FIGURE 2. Temperature distribution for six plies of polyester batting sandwiched by two layers of a nylon fabric.

close to the cold environment, on the other hand, decrease gradually. Most of the changes in temperature distribution occurred within about 0.5 hour, regardless of the batting and covering fabric types. Comparing Figures 2 and 3, which are for the same non-hygroscopic polyester batting but with differing covering fabrics, there is no significant difference in the stabilized temperature distribution, but the one with the more permeable nylon covering fabric reaches stabilization faster. As for the hygroscopic viscose battings covered with the highly permeable nylon fabric (see Figure 4) and the less permeable laminated fabric (see Figure 5), there is a significant temperature difference in the middle of the battings. From the beginning to 0.5 hour, the

FEBRUARY 2005

FIGURE 5. Temperature distribution for fifteen plies of viscose batting sandwiched by two layers of a laminated fabric.

temperature in the middle of the viscose battings covered with the highly permeable nylon fabric is significantly higher than that covered with the less permeable laminated fabric. At x⬘ ⫽ 0.6, the former is 5.0°C higher than the latter at 0.1 hour and 2.0 °C higher than the latter at 0.5 hour. The temperature difference decreases with time. At x⬘ ⫽ 0.6, there is virtually no difference in temperature between the former and the latter after 4 hours. This is due to the differences in moisture absorption within the fibrous battings. When the covering is the highly permeable nylon fabric, more moisture is transmitted into the viscose battings within the same period, and more moisture absorption takes place in the initial period, thus releasing a greater amount of the heat of moisture absorption and consequently causing a higher temperature. After about 4 hours, the viscose battings covered with either the nylon fabric or the laminated fabric are almost saturated, resulting in a smaller temperature difference. HEAT LOSS During “sweating” as simulated in our experiment, greater heat loss is preferable for thermal comfort. Figures 6 and 7 show the changes of power supply or heat loss with time for the two battings and covering fabrics, respectively. The initial fluctuation of the curves is understandably due to the PID algorithm used for controlling the temperature of the water within the shallow container. It is clear from Figure 6 that heat loss after stabilization through the polyester battings covered with the more permeable nylon fabric is about 5% greater than that through the battings covered with the less permeable laminated fabric, which may be attributed to the greater

103

FIGURE 6. Power supply for six plies of polyester batting sandwiched by two nylon and two laminated fabrics.

FIGURE 7. Power supply for fifteen plies of viscose batting sandwiched by two nylon and two laminated fabrics.

loss in latent heat of moisture transmission. Figure 7 shows that heat losses through the viscose battings are similar regardless of the covering nylon or laminated fabric. Comparing Figures 6 and 7, we see that the clothing assemblies with hygroscopic viscose batting have about 9% greater heat loss than those with nonhygroscopic polyester batting after stabilization. When condensation takes place, hygroscopic batting may not be as warm as non-hygroscopic batting. WATER CONTENT DISTRIBUTION To prevent “after-chill” discomfort, the accumulation of water due to moisture absorption and conden-

104 sation within fibrous batting should be as low as possible. Figures 8 and 9 show the distribution of water content within the fibrous battings after 8 and 24 hours for the two battings and covering fabrics, respectively. Water content within the fibrous battings is a combination of moisture absorption and condensation.

FIGURE 8. Water content in the fibrous batting after 8 hours at ⫺20 °C.

FIGURE 9. Water content in the fibrous batting after 24 hours at ⫺20 °C.

TEXTILE RESEARCH JOURNAL As we can see, water content in the batting next to the “skin” is nearly zero for the non-hygroscopic polyester and about 18% for the hygroscopic viscose. It remains almost unchanged from after 8 hours to after 24 hours for the polyester batting and only increases slightly for viscose batting. It is therefore reasonable to believe that there is no condensation in the batting next to the skin. Water accumulation is the result of moisture absorption. Polyester batting absorbs little moisture, and hence its water content remains zero. Viscose batting absorbs most of the moisture within 8 hours. The water content increases from the inner region to the outer region of the batting due to the increased condensation, and water content also accumulates with time. At the outer region, water content after 24 hours is about four times that after 8 hours for the polyester batting, and about two times that after 8 hours for viscose batting. The difference in water content in the outer region between the two batting types may be due to differences in the density of the battings or to the nature of the fibers or a combination of the two. The polyester batting is more porous and permeable, thus allowing more moist air to transmit or diffuse from its inner region to the outer region, where condensation takes place. Another possible reason is that viscose is hydrophilic and polyester is hydrophobic. The condensed water on the hydrophilic fiber surface tends to wick to regions where water content is lower. Comparing the two covering fabrics, we see that the water content at the outer region of the polyester battings covered by the permeable nylon fabric is about 10 –15% higher than that covered by the laminated fabric, and there is little difference in water content between the viscose batting covered with nylon or the laminated fabric. Considering the great differences in the air permeability of the two covering fabrics, the effect of the covering fabrics permeability on condensation within the battings is small. We can also see from the graphs that although the water content at the outer region is greater, the greatest water content may not occur at the outer most layer of the battings (i.e. x⬘ ⫽ 1.0). This may be due to the complex interaction of heat and moisture transfer in the battings. Condensation within the fibrous battings is a result of the moisture content within the air space exceeding the saturated moisture content, which is a function of temperature. As the temperature decreases from the inner region to the outer regions, the saturated moisture content initially reduces sharply and then more gradually from the inner region to the outer region (because the effect of temperature on the saturated moisture content is less pronounced at lower temperatures). On the other hand, from the inner region to the outer region, the moisture

FEBRUARY 2005

105

content decreases in an increasing rate due to increased condensation in the outer region. The greatest difference between the moisture content and the saturated one may not be at the outermost boundary (i.e., x⬘ ⫽ 1.0).

Conclusions By conducting experiments at ⫺20°C on a novel sweating, guarded hot plate, we have investigated coupled heat and moisture transfer through clothing assemblies with moisture absorption and phase change. We have found that most of the changes in temperature distribution within the fibrous battings sandwiched by inner and outer layers of thin covering fabrics take place within 30 minutes of exposure to cold conditions. Distributions of temperature and water content within the battings are affected by the moisture absorption properties of the fibers and the density or porosity of the battings. Moisture absorption of the fibers increases the temperature in the battings, and greater permeability of the battings results in a greater accumulation of condensed water in their outer regions. The effect of the permeability of the covering fabric on heat loss and water content distribution is relatively small, especially when the batting is a dense and less permeable type. Coupled heat and moisture transfer within clothing assemblies in freezing conditions is a complex combination of moisture absorption, condensation, and liquid water movement. The experimental investigation reported in this paper provides a basis for the theoretical modeling and numerical simulation to be reported in Part II.

5.

6.

7. 8.

9. 10. 11.

12.

13.

14.

15.

ACKNOWLEDGEMENT

16.

We would like to thank the Research Grant Committee of the Hong Kong University Grant Council for funding the project (PolyU 5142/00E).

17.

Literature Cited 1. Bouddour, A., Auriault, J. L., and Mhamdi-Alaoui, M., Heat and Mass Transfer in Wet Porous Media in the Presence of Evaporation-Condensation, Int. J. Heat Mass Trans. 41 (15), 2263–2277 (1998). 2. De Vries, D. A., The Theory of Heat and Moisture Transfer in Porous Media Revisited, Int. J. Heat Mass Trans. 30 (7), 1343–1350 (1987). 3. Fan, J., Cheng, X., and Chen, Y. S., An Experimental Investigation of Moisture Absorption and Condensation in Fibrous Insulations at Low Temperatures, Exper. Therm. Fluid Sci. 27 (6), 723–729 (2003). 4. Fan, J., Luo, Z., and Li, Y., Heat and Moisture Transfer with Sorption and Condensation in Porous Clothing As-

18.

19.

20.

21.

semblies and Numerical Simulation, Int. J. Heat Mass Trans. 43 (12), 2989 –3000 (2000). Fan, J., and Keighley, J. H., A Theoretical and Experimental Study of the Thermal Insulation of Clothing in Windy Conditions, Int. J. Cloth. Sci. Technol. 1 (1), 21–29 (1989). Fan, J., and Wen, X., Modeling Heat and Moisture Transfer through Fibrous Insulation with Phase Change and Mobile Condensates, Int. J. Heat Mass Trans. 45 4045– 4055 (2002). Farnworth, B., Mechanics of Heat Flow Through Clothing Insulation, Textile Res. J. 53 (12), 717–725 (1983). Farnworth, B., A Numerical Model of the Combined Diffusion of Heat and Water Vapor Through Clothing, Textile Res. J. 56 (11), 653– 665 (1986). Henry, P. S. H., Diffusion of Moisture and Heat through Textiles, Discuss. Faraday Soc. 3, 243–257 (1948). KES-F8-AP1 Air Permeability Tester, Instruction Manual, Kato Tech Co. Ltd, Kyoto, Japan. Motakef, S. M., and El-Masri, A., Simultaneous Heat and Mass Transfer with Phase Change in a Porous Slab, J. Heat Mass Trans. 29 (10), 1503–1512 (1986). Murata, K., Heat and Mass Transfer with Condensation in a Fibrous Insulation Slab Bounded on One Side by a Cold Surface, Int. J. Heat Mass Trans. 38 (17), 3253–3262 (1995). Ogniewicz, Y., and Tien, C. L., Analysis of Condensation in Porous Insulation, J. Heat Mass Trans. 24 (4), 421– 429 (1981). Shapiro, A. P., and Motakef, S., Unsteady Heat and Mass Transfer with Phase Change in Porous Slab: Analytical Solutions and Experimental Results, J. Heat Mass Trans. 33 (1), 163–173 (1990). Spencer-Smith, J. L., The Physical Basis of Clothing Comfort, Part 6: Application of the Principles of the Design of Clothing for Special Conditions, Clothing Res. J. 6 (1), 61– 67 (1978). Tao, Y. X., Besant, R. W., and Rezkallah, K. S., Unsteady Heat and Mass Transfer with Phase Changes in an Insulation Slab: Frosting Effects, Int. J. Heat Mass Trans. 34 (7), 1593–1603 (1991). Tao, Y. X., Besant, R. W., and Rezkallah, K. S., The Transient Thermal Response of a Glass-fiber Insulation Slab with Hygroscopic Effects, Int. J. Heat Mass Trans. 35 (5), 1155–1167 (1992). Thomas, W. C., Bal, G. P., and Onega, R. J., Heat and Moisture Transfer in a Glass Roof-insulating Material, ASTM STP 789, 1983, pp.582– 601. Vafai, K., and Sarkar, S., Condensation Effects in a Fibrous Insulation Slab, J. Heat Trans. 108 (8), 667– 675 (1986). Vafai, K., and Tien, H. C., A Numerical Investigation of Phase Change Effects in Porous Materials, Int. J. Heat Mass Trans. 32 (7), 1261–1277 (1989). Wijeysundera, N. E., Hawlader, M. N. A., and Tan, Y. T., Water Vapor Diffusion and Condensation in Fibrous Insulation, Int. J. Heat Mass Trans. 32 (10), 1865–1878 (1989). Manuscript received June 6, 2003; accepted February 19, 2004.