Solar heat harvesting and transparent insulation in textile architecture inspired by polar bear fur

Solar heat harvesting and transparent insulation in textile architecture inspired by polar bear fur

Energy and Buildings 103 (2015) 96–106 Contents lists available at ScienceDirect Energy and Buildings journal homepage:

3MB Sizes 16 Downloads 40 Views

Energy and Buildings 103 (2015) 96–106

Contents lists available at ScienceDirect

Energy and Buildings journal homepage:

Solar heat harvesting and transparent insulation in textile architecture inspired by polar bear fur Sebastian Engelhardt a,b,∗ , Jamal Sarsour a a b

Institute of Textile Technology and Process Engineering Denkendorf, Koerschtalstrasse 26, 73770 Denkendorf, Germany Carinthia University of Applied Sciences, Europastrasse 4, 9524 Villach, Austria

a r t i c l e

i n f o

Article history: Received 5 December 2014 Received in revised form 16 June 2015 Accepted 18 June 2015 Available online 22 June 2015 Keywords: Biomimetics Textile architecture Polar bear fur Transparent insulation Passive solar energy

a b s t r a c t Solar thermal technology is a promising key strategy for future renewable energy production. Various concepts exist that use solar collectors and heat mirrors, built from rigid materials, to gather thermal energy from solar radiation. A new approach is the utilization of textile materials to build solar thermal collector systems with flexible material properties, lightweight design and improved material-efficiency. A solar collector, based on a multi-layer arrangement of technical textiles and foil membranes, has been realized by the ITV Denkendorf (Institute of Textile Technology and Process Engineering Denkendorf). The proposed collector system allows transparent insulation in textile-based buildings while gathering thermal energy simultaneously. The system is inspired by the transparent insulation and heat harvesting strategies of polar bear fur and can inform textile-based envelopes of future transparent buildings. In this study, different material arrangements and the influence of different parameters on the temperature distribution along the collector were tested. Air temperatures up to 150 ◦ C (302 ◦ F) could be generated inside the collector system. Furthermore, a closer look at the polar bear fur and other related principles in nature delivered additional concepts for energetic optimization. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Due to increasing energy demands and environmental concerns, new sustainable, efficient, environmental-friendly and first-of-all safe energy concepts are desirable. One promising strategy is the capture of solar energy via solar thermal collectors where solar radiation is gathered and converted into thermal energy to heat up a medium like water or air to high temperatures [1]. The ITV Denkendorf developed a new concept for energy capture in textilebased buildings by using solar thermal strategies inspired by the polar bear fur [2]. The polar bear (Ursus maritimus) is known for its efficient heat insulating properties by means of a thick fur of nearly transparent and hollow hair. The skin underneath the fur is black and absorbs light, which passes through the transparent fur, and conversion into thermal energy takes place. Several air cavities inside the dense underfur ensure an efficient insulation by trapping warm air close to the bear’s skin [2–4]. These principles have been used to develop a collector system that is based on a multi-layer

∗ Corresponding author. Present address: University of Akron, Department of Biology, 302 Buchtel Common, Akron, OH 44325, USA. E-mail addresses: [email protected] (S. Engelhardt), [email protected] (J. Sarsour). 0378-7788/© 2015 Elsevier B.V. All rights reserved.

arrangement of technical textiles and foil membranes, aiming at transparent heat insulation and efficient solar thermal energy capture in textile-based buildings. The different layers build a channel system consisting of two ethylene tetrafluoroethylene (ETFE) membranes as translucent, but heat insulating top layers and a bottom, black silicone layer for the absorption of light. Light that passes through the transparent ETFE-membranes gets absorbed at the black silicone layer. The emitted heat is trapped inside the system due to the insulating material properties of the ETFE-membranes. Additionally, the two ETFE-membranes are arranged at a distance of 1 cm to each other to provide an air buffer that minimizes the heat loss at the top of the system. Between the lower ETFE-membrane and silicone absorber, an air-permeable polyester spacer fabric textile provides a 1–1.5 cm thick layer for a second air buffer. For the purpose of energy transport, an airflow inside this spacer fabrics layer can be generated by a fan or gas pump. The air takes up thermal energy while passing through the system and can be piped to a thermo-chemical energy storage unit as illustrated in Fig. 1. Underneath the silicon layer a 2 cm thick layer of insulation foam helps minimize heat losses at the bottom of the system. This study focuses particularly on the optimization of the collector system in order to gain the highest possible temperature output. The aim is to increase the air temperature by testing different materials and modifying the system’s arrangement. Therefore, we analyzed the

S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106


Fig. 1. Schematic illustration of the absorption of solar radiation and heat transport inside the collector.

dependency of parameter changes such as irradiance or airflow velocity on the output temperature while varying between different system arrangements. Measurements were performed under controlled conditions on a test bench as well as under real conditions in nature. Subsequent to this study, a prototype building was set up by the ITV Denkendorf to consider the ideal parameters as investigated in this study. Fig. 2 shows a draft illustrating the functional principles of the proposed building: air from the outside gets sucked into the system by a vacuum gas pump. The air slowly flows through the collector, which forms the building’s envelope at the sunlit side of the building, while taking up thermal energy. The hot air can then either be piped directly into the building for heating purposes, exit the building if too much thermal energy has been generated, or be piped into a thermo-chemical energy storage unit. Prior to this study a simplified computational fluid dynamicssimulation (CFD-simulation) was modeled for making predictions about the possible temperature gain of the proposed system. For the model the finite volume method was used, which is based on Euler and Navier–Stokes equations. The system was modeled using the software Star-ccm+ by CD-adapco. Three-layer combinations of the proposed collector system were simulated, varying between different material arrangements. For the upper layer, material properties of ETFE-foil and PTFE (polytetrafluoroethylene) coated glass fiber fabric were simulated, while for the bottom layer of the system, material properties for high absorption (black surface) as well as for low emissivity/high reflectance were modeled. The air gap width

between the upper and bottom layer as well as the airflow velocity inside the system were varied for the simulation. The second layer of ETFE-foil on top of the system as well as different spacer fabric textiles inside the airflow channel were not considered in the model. The geometry of the airflow channel was set to 10 m in length and 1 m in width. The radiation intensity was kept constant at 800 W/m2 . Further assumptions were made in the model such as: solar radiation with a constant 90◦ angle of incidence, zero diffuse radiation from the environment, a constant ambient temperature of 25 ◦ C, still air conditions, a laminar airflow inside the system (Re = 1773) and a perfect insulated bottom of the system. The results of the CFD-simulation represent the system’s steady state after an infinite running time under constant conditions. The highest air temperatures could be observed when choosing an airflow velocity of 0.7 m/s and a system arrangement consisting of an upper ETFE-membrane, an air gap width of 0.02 m and a black bottom layer. The results of this combination are represented in Fig. 3 and show that the air temperature after an airflow distance of 10 m amounts to 100 ◦ C. Fig. 3a compares the temperature courses along the system’s upper ETFE-membrane, the airflow channel (working fluid) and the black bottom layer depending on the airflow distance. Fig. 3b shows the simulated temperature distribution along the three collector layers depending on the airflow distance in form of a 2-D heat transfer model. It is expected that the output air temperature under real conditions will be lower than the simulated temperature, which is based on several simplifications that

Fig. 2. Illustration of the prototype building’s architecture and functional principle. Source: Reproduced with permission [5]. 2010, Institute of Textile Technology and Process Engineering Denkendorf.


S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106

Fig. 3. (a) Results of the CFD-model showing the expected temperature profiles along different collector layers and (b) the same results represented in a two-dimensional heat transfer model. Source: Adapted from [6].

have been made to the model such as a constant 90◦ radiation angle of incidence, no wind and a perfect insulation of the bottom of the system [6]. 2. Materials and methods 2.1. Test bench measurements In order to determine the optimal parameters for our system we first measured the temperature distribution along a test bench

under controlled conditions. This allowed us to simulate parameter changes such as different sun energy levels and airflow velocities, as well as testing different materials and their arrangement inside the collector. The temperature distribution along a collector of 4 m in length and 0.4 m in width was recorded over 1 h of irradiation, with different radiation intensities, airflow velocities and material arrangements. Our goal was to identify the system arrangement and airflow velocity that results in the highest air temperature at the end of the system. The different system arrangements will later be compared to each other to discuss their implementation

S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106


Fig. 4. Test bench setup and temperature sensor positioning along the collector system. Source: Reproduced with permission [5]. 2010, Institute of Textile Technology and Process Engineering Denkendorf.

Fig. 5. (a) Comparison of the terrestrial solar spectrum, the spectral distribution of a 5800 K blackbody radiator and the spectrum of a 3200 K tungsten-halogen lamp [adapted from [7]] and (b) the optical properties of a transparent ETFE layer with a thickness of 0.25 mm [6].

potential into the final prototype under consideration of material costs, manufacturing expenses and complexity of the system in relation to the reached air temperature. The test bench setup is illustrated in Fig. 4. First, an airflow was generated by a fan. To

ensure a laminar flow, the air was piped through a 5 m long wind channel before entering the collector. Solar radiation was simulated by halogen floodlights that were spaced in two rows of 11 floodlights each (22 floodlights total) at a distance of 1 m above


S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106

the collector. Each floodlight contained one halogen light bulb. By varying between light bulbs with different wattage amounts, different radiation intensities could be simulated. A pyranometer (a device measuring a light source’s radiation intensity in W/m2 from a field of view of 180◦ ) was used to measure the global radiation intensity along the collector’s surface underneath each pair of floodlights. Changes of light bulbs were made accordingly to guarantee a homogeneous radiation along the collector’s surface based on the pyranometer measurements (a maximum deviation of 10 W/m2 from the intended radiation intensity was tolerated for each pyranometer measuring point). While passing through the 4 m long collector, the air took up thermal energy and after that left the system through an outlet. It has to be mentioned that the solar spectrum differs from the radiation spectrum of halogen lamps and therefore the results of the test bench measurements are not directly comparable to those performed under real conditions in nature. Fig. 5a compares the terrestrial radiation spectrum of the sun with the spectrum of a 3200 K tungsten-halogen lamp. As illustrated in Fig. 5a, the solar spectrum has its irradiance peak at around 500 nm while the irradiance peak of the halogen spectrum is located at around 800 nm. Fig. 5b shows the optical properties of a transparent ETFE-membrane with a thickness of 0.25 mm. The optical properties of ETFE were determined by UV/VIS and IR spectroscopy prior to this study at the ITV Denkendorf [6]. From Fig. 5b we can see that the transmittance of light through ETFE is slightly higher at around 800 nm (approximate peak of the tungsten-halogen spectrum) in comparison to light with a wavelength of around 500 nm (approximate peak of the terrestrial solar spectrum). Thus, we would expect that a slightly higher amount of radiation energy would transmit though the ETFE-membranes if halogen floodlights are used as source of radiation. The reflectance properties of ETFE show no significant differences between light of 500 and 800 nm in wavelength. Fig. 4 also shows the temperature sensors positioning along the test bench: Temperature sensor 1 measured the temperature before the air entered the collector. The air temperatures in different zones along the system were measured by temperature sensors 2–5, which were positioned at a distance of 1 m from each other inside the spacer fabrics layer of the collector. Temperature sensor 6 measured the temperature of the outgoing air at the outlet of the system. 2.1.1. Determination of the optimal airflow velocity To reach the highest possible air temperature inside the collector, we first had to determine the optimal airflow velocity. We expected that a slower airflow inside the system leads to a higher air temperature, since the air remains longer inside the collector and therefore takes up a higher amount of thermal energy. For the test bench experiments the airflow was generated by a fan with speed adjusted by a dial. The resulting airflow velocity inside the system was measured with an airflow velocity sensor at the inlet of the collector. Changes of the fan’s speed were made at the dial according to the airflow velocity measured externally by the sensor. To verify a constant airflow throughout the system a second sensor was used to measure the airflow velocity at the outlet of the collector prior to every experiment when airflow velocities were modified. A maximum deviation of 0.05 m/s between the measured airflow velocities of both sensors was tolerated. Only airflow velocities which ensured a laminar flow throughout the system were tested. For that reason, the fan was adjusted for mean airflow velocities (vm ) of 0.6, 1.2 and 3 m/s. A mean airflow velocity (vm ) of 0.6 m/s inside the collector was the lowest adjustable velocity. Mean airflow velocities below 0.6 m/s were not able to generate a consistent air transport through the system, since no discernible airflow was measurable at the outlet of the collector. For that reason a distinction between a generated airflow and random movement of the environment’s air could not be made for

mean airflow velocities below 0.6 m/s. All airflow velocities were applied for radiation intensities (IR ) of 400, 800 and 1000 W/m2 . We used an irradiation intensity of approximately 400 W/m2 to simulate moderate solar radiation, 800 W/m2 for simulating average sun intensities and 1000 W/m2 to simulate strong solar radiation. Radiation intensities, simulated by the halogen floodlights, were verified with a pyranometer prior to each experiment when light bulbs have been changed to vary between different radiation intensities. All measurements were performed for a period of 1 h and measurement values were recorded every 10 s. All measurements were performed three times to guarantee a reproducibility of the results and average temperatures of the three measurements were formed. 2.1.2. Testing different material arrangements Different materials and their sequencing inside the collector system were tested to identify the collector arrangement that generates the highest temperature at the end of the system. The available materials included black and transparent spacer fabrics, ETFE-layers, black silicon layers as absorbers and insulation foams. All measurements were performed under the same conditions with a consistent mean airflow velocity (vm ) of 0.6 m/s (based on the results discussed in 3.1) and under radiation intensities (IR ) of 400, 800 and 1000 W/m2 . Fig. 6 illustrates different layer combinations that were tested including five-layer arrangements (Fig. 6a and b) as well as six-layer arrangements (Fig. 6c and e). The six-layer arrangements included an additional layer of spacer fabrics between upper and lower ETFE-membrane. We expected that an additional layer of spacer fabrics between the ETFE-membranes would provide better insulation, since without this layer, the upper ETFE-layer would bend down slightly due to an insufficient spanning and therefore a 1 cm thick air buffer would not be guaranteed. Additionally, we took a closer look at the heat-harvesting mechanisms of the polar bear fur. We expected that a large amount of thermal energy is emitted into the bear’s body to increase its body temperature. This principle has not been taken into account for the previous collector arrangements, since a part of the heat, emitted at the black silicone absorber, was lost into the insulation foam at the bottom of the system. For this reason a second airflow channel underneath the black silicon layer was implemented into the system to collect also the heat at the undersurface of the silicone layer. Furthermore, the bottom of the system was insulated by using an additional layer of aluminum foil to minimize the heat loss into the insulation foam. Also an additional layer of transparent spacer fabrics between upper and lower ETFE-membrane was added to ensure an air buffer of 1 cm thickness on top of the system. The two airflow channels above and underneath the black silicone absorber included black spacer fabrics. The arrangement of this eight-layer collector system is illustrated in Fig. 7. 2.2. Measurements under real solar radiation To test the collector under real conditions in nature, a test panel of 1 m in length and 50 cm in width was constructed. The five-layer arrangement including one layer of black spacer fabrics was used. The experiments were performed from sunrise to sunset for five days in the end of June in Denkendorf, Germany (starting date: 06/25/2012, final date: 06/29/2012). Denkendorf is located at a latitude of 49◦ north. The panel was set up at a 45◦ angle of inclination and manually positioned toward the sun on the hour. No automated tracking mechanism was used and only the azimuth angle was adapted. The inclination angle of the panel was kept at 45◦ throughout the whole measurement campaign. A detailed protocol of the adapted azimuth angle after each re-positioning step is displayed in Table 1. An azimuth angle of 0◦ represents a panel orientation toward south while an azimuth angle of ±180◦

S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106


Fig. 6. Illustration of the different tested collector arrangements.

Fig. 7. Eight-layer collector arrangement including one airflow channel above and one underneath the black light absorbing silicone layer.

Table 1 Protocol of the adapted azimuth angles after each re-positioning step of the collector panel. Time of day

Azimut angle [◦ ]

8 am 9 am 10 am 11 am 12 pm 1 pm 2 pm 3 pm 4 pm 5 pm 6 pm 7 pm 8 pm

−90 −75 −60 −40 −15 +15 +45 +65 +80 +90 +100 +110 +120

represents an orientation toward north. A positioning toward westerly directions is represented by positive angles (e.g. southwest = +45◦ ) while easterly directions are represented by negative angles (e.g. northeast = −135◦ ). To generate an airflow between lower ETFE-layer and silicone absorber inside the collector, an air pump was connected to the panel. The radiation intensity of the sun was measured with a pyranometer, which was placed horizontally on the ground at a distance from the collector panel that ensured that the panel did not shadow the pyranometer as the sun moved throughout the day. A 45◦ inclination of the pyranometer, analogous to the inclination of the collector panel, did not result in any change of the measured sun intensity, thus, the pyranometer was kept horizontally on the ground throughout the whole experimental campaign. A shadowed temperature sensor was used to record the ambient temperature (T0 ), one temperature sensor was

positioned inside the airflow channel (TA ) and another sensor was placed between silicon layer and insulation (TB ). Measurement values were recorded every 30 s from sunrise until sunset. An image of the outdoor test panel setup is shown in Fig. 8.

3. Results and discussion 3.1. Determination of the optimal airflow velocity The results of our measurements were reproducible for all tested arrangements and radiation intensities. As expected, the highest temperatures for all performed measurements were recorded close to the outlet of the system at measuring point T5 , since air close to the system’s end took up a greater amount of thermal energy, due to the longer airflow distance. The temperature at measuring point T6 was significantly lower compared to measuring points inside the collector. This is based on the time the air is cooling down while traveling the 0.2 m distance from the end of the collector to sensor T6 . Furthermore, the insulation at the system’s outlet was not sufficient, thus, the air temperature dropped before the air reached sensor T6 . For that reason, the recorded data of sensor T6 was not considered in subsequent experiments. The measurements fulfilled our expectation that lower airflow velocities are favorable in order to reach higher air temperatures. This is based on the longer time the air needs to pass through the collector. A mean airflow velocity of 0.6 m/s, which was the lowest adjustable velocity that ensured a laminar and consistent airflow throughout the system, resulted in the highest temperature output compared to higher airflow velocities. For that reason an airflow velocity of 0.6 m/s was maintained for all following test bench measurements.


S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106

Fig. 8. Outdoor setup of a 1 m long collector panel including one layer of black spacer fabrics.

3.2. Testing different material arrangements All measurements yielded reproducible results for all tested material arrangements and radiation intensities. The highest temperatures were measured close to the outlet of the system (temperature sensor ST5 ) and did not increase significantly after 1 h of irradiation. For illustration, Fig. 9 shows the temperature distribution along the test bench for a measurement period of 1 h with 800 W/m2 using a six-layer collector arrangement, including two layers of transparent spacer fabrics. The figure shows a consistent temperature increase with increasing airflow distance inside the system from measuring point T1 to T5 . After around 30 min the temperature curves transition into a plateau phase and only a slight increase in temperature can be observed at all measuring points. At this point the collector approaches steady state, which is an expected behavior since radiative heat losses increase overproportionally with the temperature [8]. When reaching steady state, the heat losses by radiation emission to the environment equal the absorbed radiation plus the liberated heat inside the system [9]. The highest air temperature of 123 ◦ C could be measured at measuring point T5 after an irradiation time of 1 h. Preliminary experiments showed that no discernible increase of the air temperature could be achieved after 1 h of irradiation, therefore all test bench measurements in this study were only performed for an irradiation period of 1 h. For radiation intensities of 400 and 1000 W/m2 equivalent results could be achieved regarding a consistent temperature increase with increasing airflow distance inside the system. The maximum temperatures of all tested arrangements, after a measuring period of 1 h under different radiation intensities, are compared in Table 2. Only measuring point T5 is shown, since the highest temperatures were obtained close to the system’s end due to the longer airflow distance inside the collector. Table 2 demonstrates, that a five-layer arrangement (arrangement 2) including one layer of transparent spacer fabrics, as well as the six-layer arrangement, including two layers of black spacer fabrics (arrangement 3), are not favorable in respect to gain the highest possible temperatures inside the collector. The arrangements 4 and 5 yielded the highest temperatures inside the system. All in all, arrangement 1, including only one layer of black spacer fabrics inside the airflow channel, should be implemented and optimized for further implementation. Even if the maximum temperature at T5 is slightly lower compared to arrangement 4 and 5 for example, the system requires only one layer of spacer fabrics and is therefore the most material-efficient solution while the maximum

achievable temperatures with such an arrangement are still very high. Additionally, it is to be expected that due to the black color of the black spacer fabrics textile, light is absorbed more homogeneously inside the airflow channel since absorption does not only occur at the black silicon layer at the bottom of the system, which would result in a higher heat loss into the insulation foam. Arrangement 6 shows that a high amount of thermal energy can be collected underneath the silicon absorber, which would have been lost without using a second airflow channel. In consideration of the high material consumption an eight-layer system is not favorable in comparison to system 1. Surprisingly, if we compare the results in Table 2 to the simulated air temperatures of the CFD-model in Fig. 3a and b, we observe that the simulated air temperatures are significantly lower than those determined experimentally. For instance, after an airflow distance of 3.5 m (measuring point T5 ) the final air temperatures range from 102 ◦ C to 123 ◦ C, depending on the system arrangement (see Table 2). The results of the CFD-simulation in Fig. 3a and b show that the working fluid’s temperature after an airflow distance of 3.5 m amounts to 74 ◦ C. The significant difference between the simulated and experimental results are expected to be mainly due to the second ETFE-membrane on top of the system, which was not considered in the CFD-model. Also the halogen floodlights used for the test bench experiments may have led to a higher amount of radiation energy being absorbed inside the system based on the optical properties of the ETFE-layers as discussed above. Furthermore, the slightly lower airflow velocity adjusted during the test bench measurements might have had an additional accumulative effect on the output temperature. It is also conceivable that the spacer fabrics layer inside the airflow channel, which was not considered in the simulation model, had an effect on the laminar airflow by partially reducing the airflow velocity and causing local turbulences when streamlines are obstructed by connecting nodes of the textile. From Fig. 9 it is also seen that the temperature gain decreases along the collector as the air gets heated. From this, the temperature gain per meter of airflow was extrapolated for a five-layer collector system, including one layer of black spacer fabrics, under different radiation intensities as shown in Fig. 10a. Values are shown for a collector panel of 3.5 m in length based on the collector length of the test bench setup. The calculated temperature gains per meter are based on experimental values (measuring points T1 to T5 ) of three independent experiments. We fit a non-linear function to the data to show the trend in temperature gain with increasing distance. It can be seen from Fig. 10a that with advancing airflow distance the temperature gain of the air decreases. From the shape of the curves in Fig. 10a it can be anticipated that after an airflow distance of 3.5 m the temperature gain will more and more approach values close to zero. Fig. 10b plots the maximum service temperatures after 1 h of irradiation depending on different radiation intensities along a collector of 3.5 m in length. Data from three independent experiments were used and a non-linear function was fit through the data points to show the trend in temperature increase with increasing airflow distance. Fig. 10c plots the final temperatures after 1 h of irradiation for measuring points T1 to T5 depending on the radiation intensity. Data from three independent experiments were used and a linear function was fit to show the trend in temperature increase with increasing irradiance. Also, the collector efficiency () has been calculated for different radiation intensities by dividing the system’s heat energy output (Qout ) by the heat energy input (Qin ), where Qin is the radiation intensity multiplied by the collector’s surface area multiplied by the irradiation period (e.g. for the test bench setup: 800 W/m2 × 1.4 m2 × 1 h = 1120 Wh) and Qout is the product of the total mass of air (m) flowing through the system during the irradiation period, the specific heat of air (cp ) and the temperature gain of the system over the irradiation time (T). The results of the

S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106


Fig. 9. Temperature curves for measuring points T1 to T5 over a measuring period of 60 min with IR = 800 W/m2 and vm = 0.6 m/s using a six-layer collector arrangement, including two layers of transparent spacer fabrics.

Table 2 Maximum temperatures of measuring point T5 after 1 h of irradiation with different radiation intensities and a mean airflow velocity of 0.6 m/s using different collector arrangements. T5 400 W/m2


(1) 5 Layers (black spacer fabrics) (2) 5 Layers (transparent spacer fabrics) (3) 6 Layers (2× black spacer fabrics) (4) 6 Layers (2× transparent spacer fabrics) (5) 6 Layers (trans. and black spacer fabrics) (6) 8 Layers a b

76 C 70 ◦ C 70 ◦ C 78 ◦ C 81 ◦ C 76 ◦ Ca 70 ◦ Cb

T5 800 W/m2 ◦

119 C 102 ◦ C 111 ◦ C 123 ◦ C 121 ◦ C 119 ◦ Ca 109 ◦ Cb

T5 1000 W/m2 141 ◦ C 122 ◦ C 129 ◦ C 150 ◦ C 150 ◦ C 140 ◦ Ca 131 ◦ Cb

Temperature above silicone layer. Temperature below silicone layer.

Table 3 Calculated efficiencies of a five-layer collector system, including one layer of black spacer fabrics, for a radiation period of 1 h under radiation intensities of 400, 800 and 1000 W/m2 . Irradiance (W/m2 )

Efficiency (, %)

400 800 1000

44.24 39.99 39.25

efficiency calculation for a five-layer collector system, including one layer of black spacer fabrics, are shown in Table 3. 3.3. Measurements under real solar radiation During the five-day measuring period different weather conditions were observed. The most diverse weather conditions were recorded on measurement day 3 (06/27/2012), representing the biggest change in temperature and cloudiness degree compared to other days of the measurement campaign. Therefore, data from day 3, which best represents the impact of diverse weather phenomena on the test panel’s operation temperature, will be presented hereafter. Day 3 started overcast in the early morning but the cloudiness degree decreased continuously during midmorning and the sky became completely cloudless around noon. During afternoon, the cloudiness degree increased slightly and the sun was partially blocked by clouds from time to time. Around 7 pm the sun set down behind the surrounding mountains and direct

radiation was no longer available. Fig. 11 compares the temperature profiles at measuring points T0 , TA and TB with the light intensity variations during the day. It is shown that the ambient temperature (T0 ) rises during the early morning and remains constant, averaging between 25 and 30 ◦ C during the day until the evening. The temperature profile of temperature sensor TA inside the airflow channel of the collector panel is approximately the same compared with sensor TB underneath the black silicone absorber. If we compare the temperature curves of TA and TB with the curve shape of the sun’s irradiance, it is obvious that the irradiance of the sun and the temperature inside the collector are highly related to each other. An increase of the sun’s radiation intensity results in a temperature increase inside the collector. If the sun was blocked by clouds the temperature inside the collector decreased. The ambient temperature has no significant effect on the temperature inside the collector. The results also show that only direct radiation with an intensity of at least 700 W/m2 is high enough to heat up the air inside the collector to temperatures higher than 100 ◦ C, so diffuse radiation does not result in any temperature increase inside the system. The maximum temperature during the day, measured inside the airflow channel, amounts to 128 ◦ C at a radiation intensity of 1200 W/m2 . It has to be mentioned that the panel’s length only amounted to 1 m. For that reason the measurements in nature cannot be compared directly to the test bench measurements but give an approximation of the available thermal energy generated by solar radiation under natural conditions. It also has to be pointed out that the change of the sun’s angle of altitude throughout the


S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106

Fig. 10. Extrapolation of (a) the temperature gain per meter of airflow, (b) the maximum service temperatures depending on the airflow distance and (c) the final temperatures for measuring points T1 to T5 depending on the irradiance for irradiation periods of 1 h.

day and year has an effect on the collector’s output temperature. Since the panel was kept constant at an incident angle of 45◦ we expect that a lower amount of solar energy was captured than if the panel would have tracked the sun to keep the sun’s angle of incident 90◦ normal to the panel’s surface. Also the seasonal changes of the sun’s altitude angle have to be taken into account. When we performed the outdoor experiments in the end of June, very intense solar radiation was present at our latitude of 49◦ North. It is clear that we have to expect lower operational temperatures in months when the sun does not reach its zenith. However, the impact of different solar angles on the collector’s temperature output will partially be compensated by the architecture of the prototype building. The active side of the roof will form a 180◦ hemisphere that is oriented toward the south. The different collector tracks covering the roof will be active or inactive depending on the sun’s azimuth angle that changes throughout the day. Furthermore, the collector tracks will be curved to maximize the likelihood that some part of the collector tracks will face the sun in a 90◦ normal angle.

3.4. Further biomimetic optimization concepts The measurements under natural conditions show that the system is highly dependent on direct solar radiation and no heat is generated if the weather conditions do not allow direct sunlight. For that reason, principles in nature could yield suitable solutions for how to couple diffuse light into the system in order to generate energy even during cloudy days. A suitable heat trapping mechanism of the polar bear’s hair via light scattering cores inside the polar bear’s hollow guard hair has been suggested by different studies [4,10]. The mechanism was first described by Grojean et al. [4] in 1980, who suggested an optical model by analyses of polar bear guard hair. The transparent hair includes a hollow and translucent core, which occupies around a third of the hair’s diameter. The cores are free of pigments, but filled with a fine membrane structure, thus the outer surface of the hair is smooth, whereas the inner core surface is rough. The rough surface is responsible for scattering phenomena of light and therefore for the white experience of the fur. A part of the light seems to be scattered forward

S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106


Fig. 11. Comparison of the temperature distribution at measuring points T0 , TA and TB to the sun’s level of irradiance along the collector panel set up in nature throughout a day in late June in Denkendorf, Germany.

inside the hair and guided to the black skin where absorption and heat emission takes place [4]. This model has been confirmed and modified by Tributsch et al. [10] in 1990 by taking a closer look at the scattering of luminescence light. The study showed that a little amount of visible light undergoes luminescence conversion and re-emission inside the hair’s hollow core. This process repeats and leads to the guiding of light toward the polar bear’s black skin where it is absorbed and converted into heat [10]. It has to be mentioned that this proposed heat trapping mechanism is controversial and even disputed by more recent publications [11]. However, even if this mechanism is not present in nature, a technical realization via optical fibers on top of the outer layer of the system could be conceivable and should be taken into account for following studies. Another possible optimization is the utilization of roughened or structured ETFE-surfaces that could couple in diffuse light of different angles of incidence. Biological models for structured anti-reflective surfaces can be found among various species in nature. For instance, insects, especially night active moths, have compound eyes with structured surfaces that consist of nano-scaled periodic arrays of cuticular protuberances. Those protuberances are nipple shaped structures, arranged in a distance smaller than the wavelength of the incident light [12]. The refractive index at the tip of those protuberances equals the refractive index of air, whereas closer to the protuberance’s bottom the refractive index changes to that of the lens material, which is in this case chitin [13]. It is expected that this “moth eye-effect” improves the light sensitivity and therefore the moth can see better during twilight [14]. Also, camouflage properties have been discussed, since the reflection of light would lead to a higher visibility, which would allow an easier detection by predators [12]. This might also be an explanation for nano-structured anti-reflective wings, which exist among various moth species. Furthermore, anti-wetting properties of nano-structured wing surfaces of moths have been investigated [15]. The functions of such anti-reflective surfaces as well as the potential of mimicking those principles for technical applications are highly discussed [12,15–20]. If the ETFE-membranes on top of our system could be coated with such a nano-structured antireflective surface, more solar energy could be gathered. It might

also be possible that such a coating would allow us to couple diffuse light into our system when no direct solar radiation is available. A further optimization concept to increase the efficiency of the system could be an alternative mechanism of ventilation. The gas pump, which will be implemented into the final prototype building, consumes electrical energy to generate an airflow inside the system. It is conceivable that a ventilation of air could also be generated by the building’s architecture itself. Some aquatic plants in nature, for example the yellow water lily (Nuphar lutea), make use of a pressurized ventilation or thermo-diffusion. The temperature of the air inside the plant’s leaves is higher than the ambient temperature and with that the pressure inside the leaves is higher as well [21]. This pressure gradient allows a flow-through pattern through the plant, which is determined by the individual pore size of the plant’s leaves [21,22]. A two-way flow in separate air-channels has been observed in Nuphar lutea and other aquatic plants [22]. This flowthrough pattern allows the transport of oxygen down to the roots as well as the transport of CO2 and CH4 from the roots toward the atmosphere [23]. A closer look at the principles of thermo-diffusion in plants could generate concepts of how an architectonic ventilation system could be realized without the utilization of auxiliary energy devices. 4. Conclusion The main goal of this study was to find the optimal collector arrangement, to reach the highest possible air temperature inside the system. The test bench measurements showed that a mean airflow velocity of 0.6 m/s yielded the highest temperatures, since the slower moving air had more time to take up thermal energy inside the system. An airflow with a velocity of 0.6 m/s was expected to be laminar and was measurable also at the outlet of the system and with that ensured a consistent flow throughout the system. The comparison of the temperature distribution along the system, using different material arrangements, yielded maximum air temperatures of 150 ◦ C under radiation intensities of 1000 W/m2 close to the end of the collector, if using an arrangement that included either two layers of transparent spacer fabrics or an arrangement


S. Engelhardt, J. Sarsour / Energy and Buildings 103 (2015) 96–106

with one layer of transparent spacer fabrics between the ETFEmembranes and one layer of black spacer fabrics inside the airflow channel of the system. For further applications it is recommended to use a five-layer arrangement, including one layer of black spacer fabrics, since this setup would be more material-efficient while very high temperatures can still be generated. It could also be demonstrated that the utilization of a second airflow channel underneath the black silicone absorber is useful to collect additional thermal energy, which otherwise would have been lost. If a greater air volume is required to store more thermal energy, depending on the used storage system, a second airflow channel should be taken into consideration. The measurements under real solar radiation in nature yielded maximum temperatures of 128 ◦ C inside the collector panel after an airflow distance of only 1 m and under radiation intensities of around 1200 W/m2 . In general, the temperature inside the panel was higher than 100 ◦ C for most of the day when radiation intensities of at least 700 W/m2 were present. The measurements were important to give an estimation of the efficiency of future construction under natural solar radiation, since the solar spectrum differs from that of the halogen-floodlights used during the test bench measurements. It is expected that higher temperatures will be achieved if longer collector panels with a thicker insulation will be implemented into the final prototype building of the ITV Denkendorf. Compared to the heat-harvesting mechanisms of the polar bear fur, it has to be mentioned that only the principles of the fur’s transparent insulation were implemented into the collector system. The possible luminescence collection of diffuse light and light scattering along the hollow core’s inside the polar bear’s hair was not realized. The measurement results show that the collector is highly dependent on direct solar radiation and diffuse sunlight could not increase the air temperature inside the system significantly. The implementation of optical fibers could be a solution to couple in diffuse light and harvest energy even if direct solar radiation is not available. Furthermore, several biomimetic concepts were discussed that could optimize the system additionally (e.g. anti-reflective surface coatings). All in all, the results in this study verify that the utilization of a solar thermal systems, based on textile materials, can yield high operation temperatures that are sufficient to store thermal energy efficiently via thermo-chemical, long-term storage. For instance silica gel, zeolite and other materials have been proposed for usage in thermo-chemical energy storage systems [24,25]. Those materials allow the seasonal storage of thermal energy without significant energy losses [25]. Compared to phase change materials, thermo-chemical material also have the advantage of their high energy density allowing for more compact storage systems compared to sensible or latent thermal energy systems [24]. In temperate zones for example, thermo-chemical energy storage in combination with the proposed collector system could inform light-weight, self-sufficient and textile-based buildings that harvest and store all required energy from the sun during summer and utilize this energy during cooler seasons. Different optimization concepts were introduced and discussed and should be taken into account for future research. Acknowledgements We would like to thank Prof. Emeritus Dr. Helmut Tributsch for several critical discussions about the functional principles of the polar bear fur that provided crucial and inspiring information for this study. We also wish to acknowledge Dr. Thomas Stegmaier from the ITV Denkendorf for the supervision of this project. We would also like to thank Wolfgang Siegle and Konstandinos Rekcin

from the ITV Denkendorf for the practical support during the laboratory work. Furthermore, we would like to thank the Carinthia University of Applied Sciences, which has been the primary institution of the correspondingly author during the time of this study. In this regard we would like to acknowledge the master’s degree program in Biomimetics in Energy Systems and program director D. ID. Mag. Peter Piccottini. References [1] V. Wesselak, T. Schabbach, Regenerative Energietechnik, Springer-Verlag, Berlin/Heidelberg, 2009, pp. 173–177. [2] T. Stegmaier, M. Linke, H. Planck, Bionics in textiles: flexible and translucent thermal insulations for solar thermal applications, Philos. Trans. R. Soc. A 367 (2009) 1749–1758. [3] I. Mazzoleni, Architecture Follows Nature – Biomimetic Principles for Innovative Design, CRC Press, 2013, pp. 125–129. [4] R.E. Grojean, J.A. Sousa, M.C. Henry, Utilization of solar radiation by polar animals: an optical model for pelts, Appl. Opt. 19 (3) (1980) 339–346. [5] Institute of Textile Technology and Process Engineering Denkendorf, Energieeffizientes Textiles Bauen mit transparenter Wärmedämmung für die solarthermische Energiegewinnung nach dem Vorbild des Eisbärfells, http:// (accessed 11.11.14). [6] T. Stegmaier, J. Sarsour, R. Wagner, R. Blum, B.H. Kroeplin, A. Kneer, W. Arnold, Abschlussbericht Eisbärbauten (Förderkennzeichen: UT 50-55), (accessed 02.04.15). [7] M.W. Davidson, Tungsten-halogen incandescent lamps, http://zeiss-campus. (accessed 02.04.15). [8] M.N. Fisch, V. Huckemann, Nonfocusing solar heat collection (incl. seasonal heat storage), in: K. Heinloth (Ed.), Energy Technologies – Renewable Energy (Landolt-Börnstein: Numerical Data and Functional Relationships in Science and Technology – New series) VIII/3C, Springer, 2006, pp. 304–307. [9] E.A. Milne, Thernodynamics of the stars, in: G. Eberhard, A. Kohlschüüter, H. Ludendorff (Eds.), Handbuch der Astrophysik: Band III/Erste Hälfte Grundlagen der Astrophysik Dritter Teil, German edition, Springer, 2013, pp. 96–99. [10] H. Tributsch, H. Goslowsky, U. Küppers, H. Wetzel, Light collection and solar sensing through the polar bear pelt, Sol. Energy Mater. 21 (1990) 219–236. [11] D.W. Koon, Is polar bear hair fiber optic? Appl. Opt. 37 (15) (1998) 3198–3200. [12] D.G. Stavenga, S. Foletti, G. Palasantzas, K. Arikawa, Light on the moth-eye corneal nipple array of butterflies, Proc. Biol. Sci. 273 (2006) 661–667. [13] P.B. Clapham, M.C. Hutley, Reduction of lens reflection by moth eye principle, Nature 244 (1973) 281–282. [14] W.H. Miller, Intraocular filters, in: H. Autrum, R. Jung, W.R. Loewenstein, D.M. MacKay, H.-L. Teuber (Eds.), Handbook of Sensory Physiology VII/6A, Springer, Berlin/Heidelberg/New York, 1979, pp. 69–143. [15] G.S. Watson, S. Myhra, B.W. Cribb, J.A. Watson, Putative functions and functional efficiency of ordered cuticular nanoarrays on insect wings, Biophys. J. 94 (8) (2008) 3352–3360. [16] Q. Chen, G. Hubbard, P.A. Shields, C. Liu, D.W. Allsopp, W.N. Wang, S. Abbott, Broadband moth-eye antireflection coatings fabricated by low-cost nanoimprinting, Appl. Phys. Lett. 94 (2009) 263118. [17] W.L. Min, B. Jiang, P. Jiang, Bioinspired self-cleaning antireflection coatings, Adv. Mater 20 (20) (2008) 3914–3918. [18] C.H. Sun, P. Jiang, B. Jiang, Templated biomimetic multifunctional coatings, Appl. Phys. Lett. 92 (2008) 5. [19] C.J. Ting, C.F. Chen, C.P. Chou, Subwavelength structures for broadband antireflection application, Opt. Commun. 282 (3) (2009) 434–438. [20] Y.M. Song, S.J. Jang, J.S. Yu, Y.T. Lee, Bioinspired parabola subwavelength structures for improved broadband antireflection, Small 6 (9) (2010) 984–987. [21] J.W. Dacey, Knudsen-transitional flow and gas pressurization in leaves of Nelumbo, Plant Physiol. 85 (1) (1987) 199–203. [22] J. Mevi-Schutz, W. Grosse, A two-way gas transport system in Nelumbo nucifera, Plant Cell Environ. 11 (1) (2006) 27–34. [23] J.W. Dacey, Pressurized ventilation in the yellow waterlily, Ecology 62 (5) (1981) 1137–1147. [24] A.H. Abedin, M.A. Rosen, A critical review of thermochemical energy storage systems, Open Renew. Energy J. 4 (2011) 42–46. [25] International Energy Agency, Thermal properties of materials for thermo-chemical storage of solar heat, in: C. Bales (Ed.), A report of IEA Solar Heating and Cooling programme – Task 32, Advanced storage concepts for solar and low energy buildings, 2005, May, publications/downloads/task32-Thermal Properties of Materials.pdf (accessed 02.05.15).