1. Introduction

According to the U.S. Department of Energy commercial buildings are responsible for approximately one-fifth of the U.S. energy consumption. The top three energy-intensive end-uses in the commercial building sector are lighting, space heating, and space cooling, which when added together constitute up to half of the total commercial site energy consumption [1]. The 2011 Buildings Energy Data Book reports that 41% of U.S. primary energy was consumed by the building sector. Of the 39 quads consumed in the building sector, commercial buildings accounted for 46%. Almost 70% of this energy consumption was used to cover space heating, space cooling, water heating, and lighting—with lighting comprising 13% of this energy consumption [1].

One significant energy saving strategy is daylight implementation, or “daylighting.” Daylighting is the controlled admission of natural light (direct sunlight and diffuse skylight) into a building to provide natural illumination, and potentially offset electric lighting. Natural light differs from artificial light in term of intensity, spectral content, and, more importantly, its natural alignment with human cycles. It contributes to human biological cycles such as the circadian cycle, a pattern that is essential to humans due to its effects on melatonin release and on physiological cycles such as the sleep-wake cycle [2]. Studies performed in school environments have indicated that students with the most daylighting in their classrooms progressed 20% faster on math tests and 26% faster on reading tests when compared to students in the classrooms with less daylighting [3]. The same results are believed to be achieved in the case of adult employees.

Good daylighting fenestration practice dictates that the window should ideally be composed of two discrete components: a daylight window (top portion) and a view window (lower portion) [4]. The daylight window allows for light admission in the room, whereas the view window permits the occupants to connect visually with the outside world. Daylight systems are installed on the daylight window portion and are based on optical devices that use reflection, refraction, and/or total internal reflection of sunlight and skylight to redirect the light to areas where it is needed, e.g., deeper in the room. Although daylighting has excellent color rendering and is believed to provide health and psychological benefits to the occupants, there have been some limitations. These limitations include high luminance and glare.

It is a common practice to use shades and/or louvers to “manage” sunlight penetration in buildings. However, most louver designs, such as venetian blinds, block the incoming sunlight more than redirect it. Hence, conventional louver systems may reduce significantly the daylight so that additional electric lighting is often needed to reach the minimum required illumination for office spaces [5]. In order to work around this problem, several teams have proposed [6, 7], and glass companies have commercialized [8, 9], transparent louvers that can redirect the sunlight to the ceiling and diffuse it for enhanced indoor illumination. These transparent louvers are made of inexpensive polymer materials and are installed between the glass panes, on the sun-facing side, or on the inward side of the window, and have a fixed rotation around their axes. However, these types of louvers can only perform light redirection, i.e., they are unable to follow the sun’s path and dynamically respond to occupant needs and outdoor environmental changes (e.g., overcast versus clear sky days). Advanced glazings belong to a group of so-called “switchable smart glazings” that allow dynamic control of solar and visible energy. These glazings are divided into two categories: (a) electrically activated glazings [10] (dispersed liquid crystals, dispersed particles, electrochromics) and (b) non-electrically activated glazings [11] (photochromics, thermochromics, thermotropics).

Other researchers have focused on energy harvesting rather than sunlight redirection. A European team has introduced two integrated façade systems that perform thermal energy harvesting. The first system is a Transparent Solar Thermal Collector (TSTC) [12]. It consists of a window in which thin, parallel aluminum strips are integrated into the window’s glazing. The strips block the incoming light from high solar angles, absorb the sun’s thermal energy, and direct it to vertical riser pipes at the side of the window which lead to a heat exchanger. The working fluid is a water-based solution of glycol and can vary depending on local climate conditions. As a result, the thermal energy from the sun is captured by the façade system and used for secondary thermally-driven applications, such as radiant space heating and solar cooling. The second façade component proposed by the same team is a Vacuum Tube Solar Air Collector System. The system is composed of horizontally placed absorber tubes [13]. Air from the building is ducted through the tubes, heated up by the sun, and then redirected back into the building. These types of projects perform only thermal energy harvesting and moreover their opaque components block the admission of daylight and/or outside view.

All the approaches discussed above perform either sunlight redirection or thermal energy harvesting. Our proposed liquid filled prismatic louver (LFPL) façade system is an integrated system that can perform both daylight and thermal energy harvesting. It actively manages the incoming solar radiation depending on outdoor weather conditions, occupant needs, and building energy requirements. Table 1 summarizes some of the advantages and limitations of the aforementioned technologies, as well as the proposed LFPL system.

Table 1. Technologies for the implementation of façade based energy efficiency solutions.

View Table | View all tables in this article

The goal of this study is to present the proposed LFPL system [21, 22], describe its basic principle of daylighting operation, and—based on a forward ray-tracing simulation study—evaluate its performance in increasing direct light penetration in a South facing office space. We compare the daylight penetration results that the LFPL system provides with a room in which no daylighting system is installed. The paper is structured as follows: Section 2 describes the LFPL façade system and its principle of operation. Section 3 describes the daylight simulation configuration and the illumination results. Finally, Section 4 discusses the engineering challenges of the LFPL system.

2. The principle of operation of the proposed liquid filled prismatic louver (LFPL) façade system

The spectrum of solar radiation covers a very wide range from ultraviolet (UV), to visible light (VIS) and infrared radiation (IR) as shown in Fig. 1(a). The VIS spectrum is used for daylighting (i.e., natural illumination). Hence, one goal of the proposed LFPL design is to optimize natural light inside an office without causing glare effects. IR, on the other hand, provides thermal heat, which may or may not be desirable depending on the climate and season. Hence, another goal of the LFPL system is to harvest and/or store solar IR radiation and use it for either heat load reduction or enhancement. Figure 1(a) also shows the high absorption of water in the various IR bands. This absorption spectrum is detailed in Fig. 1(b). Note that the absorption reaches high values for several broad bands at the IR (e.g., absorption coefficient k >10⁻³ cm⁻¹, around 2000 nm), whereas the absorption for the VIS is several orders of magnitude lower (e.g., <10⁻⁸ cm⁻¹). This shows that pure water a few inches deep is transparent to VIS (i.e., no visible effect on natural illumination), but can be used to absorb strongly in several IR bands and hence capture or harvest this solar IR energy in its content. Note also that water is inexpensive and noncorrosive, which can make the operation and maintenance of the system easier.

 

Fig. 1 (a) Spectral irradiance of solar radiation at the top of the atmosphere (black) and at sea level (red), significant absorption bands due to water absorption in the atmosphere are shown; (b) Absorption of water in the visible and IR spectral bands. The shaded area represents the visible spectrum.

Download Full Size | PDF

Figure 2(a) shows the proposed LFPL façade system, which consists of an array of hollow prismatic louver elements filled with liquid (e.g., water). The prismatic louvers are 33 inches long, and have an equilateral triangular cross section, with each side 7.62 cm (or 3 inches). The glass slats of the louvers have a thickness of 0.3175 cm (or 1/8 inch) [Fig. 2(b)]. This size was chosen because it is a standard size for glass slats in the US market. It also provides compatibility with the associated brass tubing and support system that allows us to rotate the prism and also circulate the water. Prisms have been used extensively in a variety of optical systems to redirect optical beams. This equilateral cross section prismatic configuration was chosen because of its potential to redirect light through refraction and/or reflection and its relative simplicity in construction.

 

Fig. 2 (a) The liquid filled prismatic louver system; the prismatic louver elements are interconnected to allow the liquid circulation; (b) typical optical paths for an incoming optical ray; (c) the prism configuration for rotation.

Download Full Size | PDF

To increase the thermal harvesting power of the LFPL, we propose that each louver element be interconnected with the element below, to allow for circulation of the fluid through all the louvers of the window. The circulation of the water increases the interaction length between solar radiation and water volume and hence improves the absorption capability of the water volume. Before the liquid reaches a temperature that is higher than the surrounding environment and therefore starts re-radiating its thermal energy, it can be flushed out of the prismatic louvers and delivered to a heat exchanger, where its thermal energy is removed and used for secondary applications, such as Solar Hot Water (SHW) systems. There are already available thermally-driven air conditioning technologies which are integrated with solar thermal systems. Hence, the LFPL façade system also has the potential to be integrated with an absorption or adsorption chiller system [23] that will dissipate cool air to the interior of the building. The heated liquid from the LFPL system can be used to assist in absorption or adsorption systems. Although the liquid in the prismatic louvers may not reach temperatures required in such systems (e.g., 85°-110° C for absorption technology, and 55°-90° C for adsorption technology [23]) the elevated temperature can assist the system by providing an initial higher temperature for the chiller.

The LFPL façade system principle of operation is based on (a) choice of materials (e.g., glass, liquid), (b) the geometry of the prism element, and (c) the relative orientation of the prism to the incoming radiation. Incoming direct solar radiation strikes the prismatic façade. The prism—due to its equilateral geometry—redirects the sunlight through a combination of refractions and/or reflections. Figure 2(b) shows a general configuration of such a series of refractions and reflections for a single beam. The beams are redirected towards the ceiling, where a diffusive reflective surface scatters and spreads (redistributes) the light through the entire room, resulting in deeper natural daylight penetration. When IR strikes the surface of the prism, it also undergoes refraction and reflection, but it experiences stronger absorption by the liquid inside the prismatic elements. The part of the solar radiation that is absorbed by the liquid, and hence increases its temperature can be redirected to secondary thermal applications for auxiliary heating/cooling systems.

The above features show that the prismatic louver façade has the potential to dynamically respond to weather condition changes, e.g., on different days and/or seasons. This flexible operation is based not only on the LFPL’s capability to integrate different flat glass plates, but also the ability to rotate the prism around its axis. This rotation is achieved by having a brass pipe act as both the inflow (or outflow) point and the rotation axis of the prism. A gear is connected to each of the brass pipes [Fig. 2(c)], and a chain links all the gears together. A low power stepper motor can move the chain and control the rotation of the LPFL elements. As a result, depending on the seasonal conditions and occupant needs, the elements of the LFPL façade system can rotate via an automation system and expose a different surface of the prism to the sun and thus achieve different functionality [22].

3. Illumination simulations

3.1 Simulation configuration

In this paper, we focus on modeling the LFPL effect on the redirection of the direct component of the solar radiation deeper into the room. The office under study is a South facing perimeter office in the College of Engineering at the City College of New York, NY, located at latitude 40.82° and longitude −73.95°. The room dimensions are: height (H) 2.6 m; length (L) 5 m; and width (W) 3.5 m [Fig. 3(a)]. The façade consists of three windows with two portions each. The bottom portion is the viewing region and no louvers are installed. The top portion is the daylighting portion and is where the louver system is installed. The dimensions of each of the three bottom (viewing) portion windows are 0.9 m × 0.4 m (L × H), with a sill height (from the floor to the bottom portion window) of 1 m. The dimensions of each of the three top portion windows are 0.84 m × 0.84 m (L × H). The entire window area (top and bottom portion) described above produces a 35% window-to-wall ratio (WWR). The louver system consists of three sets of prismatic louver arrays with 10 prism elements in each array [Fig. 3(a)]. The windows are assumed to be single pane clear glass.

 

Fig. 3 (a) The South facing room under study. The prismatic louvers are installed at the top portion of the windows, whereas the bottom portion is the viewing part of the window; (b) Simulated incoming solar radiation when no liquid filled prismatic louvers are installed and (c) when the LFPL façade system is installed.

Download Full Size | PDF

The office space illumination analysis was performed using LightTools 8.1, an illumination analysis software by Synopsys® [24]. LightTools’ Monte Carlo ray-tracing facilitates accurate spectral modeling of the illuminance distributions of the opto-mechanical model of the room. LightTools combines full optical accuracy, powerful optical and illumination analysis features, and an interactive graphical user interface in a 3D solid modeling environment. It can simulate polarization, scattering off surfaces and within materials, surface reflection and refraction, and the performance of thin film coatings and color filters. It allows interaction of optical and mechanical models with rays, simulating the full light propagation and has been used in illumination [25–28], displays [29, 30], solar concentrators [31], and PV [32] studies among other. LightTools’ Monte Carlo ray-tracing facilitates accurate spectral modeling of the illuminance, luminance, and intensity distributions anywhere in the opto-mechanical model. In our modeling analysis it was very important to use a ray-tracing tool of that detail in order to understand how the sun rays refract, reflect, and undergo total internal reflection in our LFPL system. LightTools can in addition be coupled with scripting commands in order to perform parametric studies and optimize the configuration. We used climatic data from NASA’s website [33] and we modeled the position of the sun for the date, time, and location of our study. Specifically, the daily averaged clear sky insolation values incident on a horizontal surface for direct normal radiation in New York City are 445 W (or 53,568 lux) for March equinox, 523 W (or 62,208 lux) for June solstice, 470 W (or 56,448 lux) for September equinox, and 305 W (or 36,864 lux) for December solstice. The lux values are calculated based on the “Standard Test Conditions” described by a spectrum of air mass 1.5 (i.e., AM1.5) [34].

We performed simulations for four representative days of the year—namely the spring and fall equinoxes and the summer and winter solstices—for office hours ranging from 8:00 AM to 4:00 PM. In our simulation and analysis, we focus on the redirection of direct sunlight only, and we assume all prism sides to be made of the same plain glass material (e.g., N-BK7 Schott). The simulation takes into account Fresnel reflection at all interfaces and accounts for multiple reflections. We run the simulation for 1,000,000 beams for each sun position making sure that for each run all beams are incident at the office façade. We also assume a receiver surface covering the entire work plane area, which is sectioned in a grid of 53 × 83 bins that correspond to an area of 4 cm² of a typical lux meter receiver. This ensures that we have enough beams falling at each receiver bin and obtain accurate results with an error of < 3.5%.

As discussed earlier, the combined effects of refraction and reflection of the incoming visible radiation through the prismatic louver façade for different prism orientations can redirect sunlight deeper into the room after diffused reflection from the ceiling. Figure 3(c) shows a typical example of how the prismatic louver façade redirects the visible part of the spectrum of the incoming solar radiation. Note that the scattered light from the ceiling comes onto the working space from above, which is expected to limit glare effects. We compare the illumination results between two scenarios. In Scenario 1, no prismatic louvers are used at the daylighting (top) portion of window. The light passes through the window unobstructed/unmanaged as shown in Fig. 3(b). In contrast, in Scenario 2 the louver system is installed on the top portion of the windows [Fig. 3(c)]. Depending on the prism orientation, a different degree of light redirection is achieved. Most of the redirected light strikes the ceiling, which is simulated as a Lambertian reflective diffusive surface (e.g., white paint) with reflectance of ρ = 90%, from which it scatters throughout the room. The remaining surfaces (i.e., surrounding walls and the floor) are simulated as mechanical absorbers (e.g., ρ = 0%). By running the simulations for different prismatic louver orientations, we can identify the optimum prism rotation angle for each time of the day that results in:

We simulated the South facing room from 8:00 AM to 4:00 PM, at 1-hour time steps, for each of the four representative days in the year (equinoxes and solstices) for both scenarios. We also included the Solar Noon (SN), which is the time of the day when the sun is at its highest elevation and has an azimuth angle of 180°. In our study, we have considered louvers with equilateral prismatic cross sections, with all sides made of the same materials and offering the same functionality. Hence, with every 120° of rotation the prismatic louver has the same effective position with respect to solar radiation, and the illumination results repeat. Therefore, we simulate the prism rotation at a range from 0° to 120° angle at a step of 1°. The illuminance is measured at the work plane, which is defined as the surface at a height of 0.762 m (or 30 inches) Above Finished Floor (AFF), where most office tasks are performed. In the next section, we present a collection of results obtained with the aforementioned simulation conditions. We present the most representative ones that show both the potential as well as limitations of the LFPL system.

3.2 Illumination simulation results and discussion

The effect of the LFPL façade system on the light redirection and the enhanced light penetration at the room is shown in Fig. 4, where we plot the horizontal illuminance (E_v) at the work plane versus the distance from the window for three different axes along the room. More specifically, we plot the illuminance along the axis connecting the center point of the South facing façade to the center point at the backside of the room. We also plot the E_v along an axis offset by a distance “d” equal to 1 m from each side of the center of the room in order to examine the uniformity of illumination at different areas of the room. Figure 4 depicts the results for SN on Fig. 4(a) June solstice, Fig. 4(b) September equinox, and Fig. 4(c) December solstice, for Scenario 1 (e.g., “without” prismatic louvers) and Scenario 2 (e.g., “with” prismatic louvers). The results for the March equinox are almost identical to the September equinox due to the very similar solar path on these two dates. Figure 4 shows that installing the liquid filled prismatic louvers and properly positioning them (i.e., rotation along their axes) for the different times of day allows natural light to:

 

Fig. 4 Illuminance (E_v) versus distance from façade at SN for (a) June solstice, (b) September equinox, and (c) December solstice for SN; (d) September equinox under three different transmittances for the viewing window (e.g., clear glass Tvis = 90%, Tvis = 50%, and Tvis = 25%). (Minimum required illumination is set at 300 lux).

Download Full Size | PDF

For the December solstice, we see that even though the prismatic louvers redistribute the incoming light throughout the room they also increase the illuminance at the front of the window. This is due to the lower elevation angle of the sun and the rotation of the prism to achieve a uniform deep penetration. In general the illumination at the front portion of the room is dominated by the light that is admitted into the room by the lower portion (viewing part) of the window. In order to reduce this excess illumination, we can change the tint of the viewing part of the window; and thus reduce the illumination at the front of the room. Figure 4(d) shows this effect in the case of September equinox, where different tinted glass has been used to cut down the direct sunlight by more than 80%. Therefore, it is useful to report the efficiency of the LFPL system in redirecting incoming light. We evaluated the available incident illuminance at SN for the four representative days at four locations: (a) just in front of the prisms; (b) just behind them; (c) on the work plane; and (d) on the entire ceiling area. The efficiency of the LFPL system in redirecting admitted daylight onto the ceiling was 51%, 60%, and 44% for June, March/September, and December, respectively. Our simulation demonstrates the maximum available illumination level that the LFPL system can achieve. However, if such high illumination levels are not needed, then illumination sensors and the controls system can reorient the prisms in order to achieve a less bright indoor environment.

This redistribution of daylight in the indoor space is also observed for all the other time periods under study. During the other simulated time periods (i.e., 8:00 AM to 4:00 PM) we observe similar light redirection, better uniformity, and increased illuminance level all the way to the innermost part of the room, while at the same time a reduction of the illuminance is seen at the front portion of the room. We should mention that in the early morning hours (e.g., before 10:00 AM), as well as in the late afternoon hours (e.g., after 4:00 PM), the redirected sunlight illumination may not reach the level of >300 lux for the entire depth of the room, due to the lower solar elevation and the steeper incident angle of the incoming direct solar radiation with respect to the window due to the azimuth angles. However, the levels of illumination still reach levels of more than 200 lux for the majority of cases, hence relaxing the artificial lighting requirements. Furthermore, in office environments with complementary task lighting, illumination levels from direct sunlight of 100-200 lux are acceptable [35].

To better illustrate the enhanced natural illumination (uniform, deep penetration at or above required illuminance levels) in the room, we also present color map plots of the illuminance (E_v) of the entire room at the work plane level for both scenarios (i.e., without and with prisms), for the September equinox and June solstice cases and for three representative hours of the day (i.e., 10:00 AM, SN, and 3:00 PM). As depicted in Fig. 5, a general trend appears at all the simulations. We observe that when no louvers are installed, there is an undesired, high intensity illumination level (e.g., >3,000 lux) at the front of the room (e.g., ~1-1.5 m from the window) and in the rest of the room (e.g., >1.5 m from the window), illumination that is so sparse as to be insufficient for office tasks. In contrast, when the prismatic louver façade is installed, and for an appropriate prism rotation, a big portion of the high intensity illumination is redirected to the interior of the room and most of the work plane is served with natural light. We observe similar results for the other dates in our simulations. For example, the March equinox shows a very similar performance as September [Fig. 5(a)] due to the very similar sun paths. In both equinox days, the simulations show that the prismatic louver façade efficiently redirects, redistributes, and enhances natural illumination indoors. On the summer solstice, the sun is at a higher altitude and steeper angle. Hence, as we see from Fig. 5(b), left column, when no prisms are installed, the sunlight reaches only a small part of the room closer to the window, whereas the rest of the room remains dark. This inefficiency is addressed by the LFPL façade system. As we see from Fig. 5(b), right column, the prisms redirect and redistribute the incoming natural light through refraction and reflection deeper in the room. On the winter solstice, when no prisms are installed, sunlight penetrates and reaches more than half the depth of the room. This is because during the winter solstice the sun is at a low altitude. However, since light enters unmanaged through the window, it reaches the work plane at an angle which may cause thermal discomfort due to the direct radiation and visual discomfort due to glare. Hence, in order to reduce glare, the occupants pull down the shades/blinds to block the sunlight and turn on the artificial lighting because there is not enough illumination [36]. In contrast, the prismatic louver façade redirects light to the ceiling, after which it is diffused from the ceiling and spread to the work plane from above—thereby alleviating thermal and glare discomfort.

 

Fig. 5 Color maps for illuminance at 10:00 AM, Solar Noon, 3:00 PM for (a) September equinox, and (b) June solstice.

Download Full Size | PDF

Table 2 shows the percentage of the work area illuminated to the given illumination levels for both scenarios. We see that in the case where no LFPL is used, only a small portion of the room achieves illumination levels between 300 lux and 3,000 lux. However, when the LFPL is used the work area coverage increases significantly (e.g., more than 80% for June, March and September). In the December solstice, the improvement is smaller due to the lower elevation angle of the sun. Another significant finding is that in Scenario 1, the areas with illumination in excess of 3,000 lux are more than double when compared with the case of LFPL. We should also note that all of this excess illumination (e.g., 100%), for June, September, and March, is within the first 1.5 m from the façade in both scenarios. Hence, this illumination is due to the admittance of direct sunlight from the viewing window. At the December solstice, one third of this excess illumination is in the first 1.5 m when the LFPL is not used, whereas it is more than 98% when the LFPL is used. This shows that the high illuminance zones observed in Scenario 1 have been redistributed in the room with the use of the LFPL. Installing tinted glass at the lower portion of the window, as described earlier, can reduce this excessively high illuminance performance at the front 1.5 m of the room. The general pattern in all color map plots of Fig. 5, as well as the additional results for the other times and dates (not shown due to page limitation), demonstrates that the LFPL façade system can effectively manage and control the incoming solar radiation for the benefit of the occupants.

Table 2. Comparison of a room without (Scenario 1) and with (Scenario 2) LFPL installed for the case of solar noon.

View Table | View all tables in this article

In Fig. 6, we plot the percentage of the work plane for which an illumination level of more than the minimum required illumination of 300 lux is obtained. This illumination optimization is achieved by properly rotating the prismatic louvers. Our analysis shows that during peak hours (e.g., 11:00 AM to 1:00 PM) when electricity is most needed and is most expensive, the prismatic louver façade system outperforms significantly Scenario 1 and shows the potential to use less electric lighting. In addition, during peak hours the LFPL façade system covers more than 65% of the work plane with natural light, reaching up to 85% coverage in some cases, hence significantly reducing the artificial light requirements.

 

Fig. 6 Percentages of the work plane that achieve an illumination level more than the minimum required illumination of 300 lux from 8:00 AM until 4:00 PM, achieved by proper rotation of the prismatic louvers. (a) March equinox, prism rotation, (b) June solstice, and (c) December solstice.

Download Full Size | PDF

We also observe very high natural light percentage coverage differences during the June solstice between the two simulated scenarios (see Table 2). This is because—as expected when no prisms are installed—the sun’s rays are concentrated at the window opening area due to the high altitude/steep angle sun position and hence the natural illumination levels deep in the room are very low. When prismatic louvers are installed and optimally rotated, the concentrated rays are redirected and redistributed to the room, contributing to much higher illumination levels. Lastly, in the case of the winter solstice simulation, the work space coverage percentage without the LFPL system is considerably higher than in the other simulated days. This is because of the low altitude angle of the sun during winter, which allows the sun to penetrate deeper into the room. However, even in the winter solstice case, when the LFPL façade system is installed on the windows, the natural illumination levels are increased and are consistently 20% higher than before. This shows the potential of the prismatic system to offset artificial lighting. There is an additional important difference between the two scenarios in the winter solstice case. In Scenario 1, the light admitted through the window is directed towards the work area and the occupant at an angle and causes a glare problem. In contrast, the light that is redirected by the LFPL reaches the work space from above, and thus limits glare effects.

The previous results were focused on the effects of the LFPL system for the scenario of a typical South facing perimeter office in NYC. This is just a representative case, and we showed that throughout the year and for various azimuth and elevation angles of the sun the LFPL system increases natural illumination in the room, reaches the 300 lux threshold, and consequently reduces the need for electric lighting. However, the LFPL system operation is not limited to the NYC climate and/or geographic location. NYC’s azimuth and elevation range of angles throughout the year is similar to other locations in the US at different times of the year. Hence, the results can be extended beyond NYC’s particular climatic condition. To confirm our assumption, we performed simulations for additional major metropolitan areas in the US (e.g., Charlotte, NC; Miami, FL, Houston, TX, Chicago, IL; Los Angeles, CA and Seattle, WA). We selected these cities so as to cover the entire US continental landscape and represent some of the major US metropolitan centers. For most days during the year, the cities under study have sun paths that fall within the two NYC extreme cases (i.e., winter and summer solstice). The exceptions are Miami, FL from April to August, and Seattle, WA from November to January. In these cases the sun follows a path that has elevation angles either larger or smaller than that of NYC.

Figure 7(a) shows the sun path for NYC during the two solstices and for Seattle and Miami for several months. For the case of Miami, we show the month of December, which falls within the NYC elevation angles, and the months of June, May and April, which extend beyond the NYC sun path. The sun path for July and August is similar to that of May and April, respectively. For the case of Seattle, we show the month of June, which falls within the NYC elevation angles, and the months of November and December, which extend beyond the NYC sun path (at lower elevation angles). The sun path for January is similar to that of November and is omitted. All other months, for Miami and Seattle, fall within the bounds set by the two solstice sun paths of NYC. Furthermore, all other major metropolitan cities under our study have sun paths that are within or slightly different than NYC’s azimuth and elevation range of angles and are not shown in Fig. 7(a), since the simulation analysis showed that the LFPL exhibits similar daylight performance as in the case of New York City.

 

Fig. 7 (a) Sun path diagram (i.e., elevation and azimuth) for NYC (solstices), Miami (April, May, June, and December) and Seattle (June, November, and December). Data are shown from 8:00 AM to 4:00 PM, and each marker location represents the sun position (i.e., elevation and azimuth) on the hour. Percentage of office area with illumination levels > 300 lux for (b) Miami in May and (c) Seattle in December.

Download Full Size | PDF

Figure 7(b) and Fig. 7(c) shows the percentage of the work plane for which illumination levels exceed the minimum required illumination of 300 lux for the more challenging cases of Miami (in May, Fig. 7(b)) and Seattle (in December solstice, Fig. 7(c)). As expected, the sun path at Miami follows an almost straight overhead path in June. Therefore direct sun rays do not always reach the prismatic louvers and hence they are not redirected towards the ceiling and then to the rest of the office space. Thus, due to the very steep elevation angle in the summer solstice for the case of Miami, the daylight enhancement effect is not as significant as in the other case studies. In particular, the LFPL system provides a 20-25% daylight enhancement at the work plane for only 2-3 hours during daytime. The results are significantly better for the month of April (and August), where the LFPL system provides more than 50% illumination coverage at the work plane level for 6 hours during daytime when compared to the low 10% illumination coverage when no prismatic louvers are installed.

For the case of Seattle during the winter months, with the sun at low elevation angles, the illuminance level is consistently better by >12% (from 10:00 AM to 2:00 PM) and by >20% (for the rest of the daytime), when the LFPL system is used compared to the case of a regular window [Fig. 7(c)].

4. Challenges of the LFPL system

We have described the illumination results of our simulations and the benefits of improving daylight admittance deeper into the room, while delivering more uniform illumination. We should also mention some of the challenges of the LFPL system, and of daylighting systems in general. As we noted before, our study focused on the redirection and redistribution of direct solar radiation. Although the diffuse solar radiation was not included, we do not anticipate that the prismatic louvers will significantly affect diffuse radiation, which comes from a wide range of different angles, apart from a relative reduction in the level of diffused illuminance in the room. We removed any contribution of scattering or diffusion from walls and/or furniture to make our results independent of these factors—which in general vary more among different office spaces than does the ceiling.

As a prototype the LFPL system will need a capital investment from the building owner. However, once the LFPL is accepted as a technology and is mass produced then the upfront cost will be significantly reduced due to economies of scale. In addition, the operation of the system is not energy intensive since it will use low power stepper motors. Lastly, maintenance costs will be limited too since the system is installed from the inside of the window façade and therefore it is not exposed to the dirt and dust of the outdoors. The economic benefits of the technology will be better described when the combined effects of better daylighting (e.g., deeper penetration and uniform illumination) that offsets artificial lighting as well as thermal harvesting benefits are measured in our future work. However, the economic analysis of the system is not the scope of this paper.

A question often raised is whether or not the material dispersion of the glass and the water causes wavelength (color) splitting of the redirected beams. In general, the prismatic geometry may cause dispersion of the colors. The degree of the dispersion depends on the angle of rotation and the angle of incidence of the incoming light. However, this happens when the prism is oriented as shown in Fig. 2, with the apex angle facing downwards, and hence the redirection is dominated by refraction. In contrast, when the apex angle faces upwards, the redirection of the light is dominated by total internal reflection (TIR). In the TIR case, the refraction angles of the light at the entry air-glass interface and at the exit glass-air interface cancel out and no color splitting is observed. In our analysis we found that the majority of optimum rotation angles were TIR dominated. In addition, even in the case where the redirection is dominated by refraction, only a small color splitting (a few mm) may be observed at the edges of the illumination patterns on the ceiling. However, there is no color splitting observed at the work space as the light is scattered by the ceiling in a diffused and not specular fashion.

Lastly, the LFPL is a light redirecting device and therefore as all beam radiation redirecting devices it works most favorable under clear sky weather conditions. The beam (direct) radiation is the one that contributes to glare and to overheating and that is why it is important to manage the direct component of solar radiation. Under overcast sky, glare or undesired heating doesn’t occur and as a result the potential light redirecting benefits are less significant.

5. Conclusion

We have proposed a liquid filled prismatic louver (LFPL) façade system as a solution to efficiently manage the incoming solar radiation for both daylighting (illumination) and thermal (heating/cooling) benefits. This would allow the façade of high-rise buildings to be transformed from a passive element to an energy-gaining component. In this paper, we focused on the daylight benefits of the LFPL façade system, and showed that with an appropriate selection of prism louver orientation, we can achieve not only significant daylight penetration of the direct component of the solar radiation deeper into the office space but also illumination uniformity and potential for reduced glare effects. In particular, we have presented detailed results for New York City and have extended our analysis for other metropolitan cities in the continental US of different climatic conditions. The illumination simulations of direct sunlight show that the LFPL considerably enhances indoor natural illumination in all cases with elevation angles and sun path within the range of that of NYC.

In addition, we have described how the heated liquid in the LFPLs has the potential to be used for secondary thermal applications. Potential applications include solar assisted air conditioning, solar assisted hot water, heat pumps, and energy extraction using heat exchangers. In terms of occupant comfort, employees experience the benefits of natural illumination and therefore may enhance their performance/productivity and regulate their circadian rhythms, which lead to improved psychology and quality of life. Future work will investigate the thermal harvesting potential of the system, as well as the experimental verification and demonstration of the system.