2.2.1 Simulation for PL Uniformity
The polished acrylic extended surface of 30mm length and 8 mm diameter was considered for simulation. The TracePro simulation software was used for ray tracing of blue photons at the polished acrylic surface, Fig. 2(a). The specular transmission of the acrylic surface was 0.8 (80%) with 0.15 (15%) losses due to total material absorption or absorption/mm and back reflections. It was found that most of the blue components in the middle section of the small rod were waveguided to the other end, Fig. 2(b) and no scattering of blue photons at the center was observed. The simulation results of the polished acrylic extended surface justified the experimental results of PL non-uniformity, Fig. 1(e). Maximum blue photons were waveguided inside the acrylic rod resulting unexcited phosphor layer in the middle section. Therefore, it was decided to diffuse the polished surface of the rod uniformly throughout the length to extract out optimal blue component for phosphor excitation.
This resulted in reducing the specular transmission up to 40% and increasing the bidirectional transmission distribution function (BTDF) to 40%, Fig. 2(c). The simulation results in Fig. 2(d) shows that the blue light was extracted at the middle section of the diffuser significantly making the scattering nearly uniform from one end to the other end of the rod. But there is still higher blue irradiance at the initial section of the rod this is due to the fact that more number of blue photons were extracted out at the beginning. Therefore, the uniform diffusing method was not suitable for the initial section of rod because it reduced the waveguided photons that were transferred to the middle section, leading to higher initial scattering for phosphor excitation.
The maximum photons must waveguide at the initial section to propagate further into the middle section and the final end of the extended cylindrical diffuser. To resolve this problem, variable diffusing of surface was required which was termed as gradient diffusing technique. In the process of gradient diffusing, first the diffuser was virtually divided into ten sections (3mm each), Fig. 3(a), and the diffuser's initial sections were made with maximum specular transmission to reduce the waveguiding loss at a critical angle inside the diffuser.
Section 1 in Fig. 3 (a) was completely polished after that the rod was properly diffused in ascending order from section 2 to 5. From sections 6 to 10, the descending order diffusing was maintained. The main motive of this gradient diffusing was to use both waveguiding and scattering properties of the diffuser. The specular transmission (ST) and BTDF parameters taken in diffusing are shown in Fig. 3(b). The simulation showed the proper distribution of blue irradiance at the phosphor layer surface; Fig. 3(c). Fig. 3(d) showed the blue photons and fluorescence photon scattering on the cylindrical circumference detection recorded at a 10 mm perpendicular distance from the source. Through the simulation parameters of gradient diffusing, the experimental rod was prepared for further development.
2.2.2. Development of Extended Diffuser for PL Uniformity
In the development process, the three different small rods (polished, uniform diffusing, and gradient diffusing) were used in accordance with simulation for experimental verification. Fig. 4 (a, b, c) are the schematics of polished, uniform diffusing, and gradient diffusing rods, respectively. Fig. 4 (d) was the spectra of polished rod taken at 5 mm perpendicular distance from the rod which shows the high nonuniformity in PL of blue radiance along the length of the rod. Fig. 4(e) is the blue irradiance spectra along the length which showed the maximum scattering at initial section of the rod, and there is a large gap between 0 cm and 2.5 cm length.
To fill this spectral bridge, the gradient diffusing technique was adopted according to simulation parameters. The spectral gap was observed, Fig. 4(f), but it shows a better result than the previous design because the blue radiance distribution was observed from 0.5 cm to 2.5 cm, Fig. 4 (d & e). This was the maximum that can be achieved experimentally through the gradient diffusing method. Considering these improvements in blue photon extraction to the phosphor layer, the gradient phosphor coating technique was adopted using the screen-printing technique. In this phosphor coating process, the optimal concentration was used for particular blue irradiance to maintain the PL uniformity, i.e., for higher blue irradiance, the higher concentration of phosphor was adopted, and for lower blue radiance, the required amount of phosphor concentration was adopted for same PL. The same thickness of the phosphor layer was maintained throughout the rod which was 150µm, and the phosphor: adhesive ratio was changed along the length, as shown in Fig. 5(a). The total (phosphor + adhesive concentration) was 0.00611 gm/cm2 throughout the diffuser surface. Fig. 5 (a) shows that adhesive concentration was increased on decreasing the phosphor concentration. The UV adhesive was adopted for phosphor coating because its refractive index is nearer to glass.
As the phosphor concentration + adhesive concentration was managed constant which was 0.00611 gm/cm2, the layer thickness was 150 µm throughout the length which simplified the design. The variation of phosphor and adhesive concentration (7:2) also indicated that the phosphor layer's refractive index was continually changing from high value to lower value, resulting in higher extraction of blue source along the length of the diffuser to adjust the maximum illuminance uniformity, Fig. 5(b). The 3D design is shown in Fig. 6(a) of phosphor layer coating on diffusers' surface from higher to lower value. When the blue irradiance from a single LED source was properly distributed inside the extended diffuser, the approximate PL uniformity was observed, as shown in Fig. 6(b). The lighting device illuminated the same color along each section of the diffuser due to PL uniformity. This work can be internally involved in illumination systems such as automobile LED headlight structures. The intensity along the length of rod with gradient diffusing (GD) was improved in the term of illuminance uniformity than the uniform diffusing (UD) technique-based diffuser, Fig. 6(c). But the intensity at the diffuser's initial sections was still higher because of higher scattering from the blue LED.
In contrast, the dome-shaped structure in the diffuser for the LED system helped in the directionality, but it was not enough.
To normalize the higher luminance observed at the initial section, the diffuser was covered by a parabolic reflector to redirect the light towards the desired region which is shown in figure 7(a). Figure 7(b) shows the photograph of the source without the mirror at the end section.
The light was uniform and less hazardous in comparison to chip-on-board LED or surface-mounted LED. The spectrum was taken along the length of the rod without mirror coating. This was to understand the exact waveguiding of the blue component throughout the rod after the gradient diffusing technique. The light was observed approximately uniform. After mirror coating at the end of the rod, enhancement in light was improved for non-hazardous and efficient illumination.