Epitaxial lateral overgrowth can be implemented using patterned sapphire substrates (PSS) and SiO<sub>2</sub> nanorod arrays (NR). Both PSS and SiO<sub>2</sub> arrays are fabricated using nanoimprint lithography. In this paper, we study patterned sapphire substrate width and period and SiO<sub>2</sub> nanorod array z position to optimize GaN LED light extraction and improve the device efficiency. First, we compare our simulation with pervious experimental data from other group. The simulation results match the experimental results in the trend. Second, we investigate PSS design and optimization, and find that by setting the period and width of PSS to 2μm and 1.3μm respectively, the light extraction can be increased by 47.9%. We also optimize the z-position of the SiO<sub>2</sub> nanorod array to 7.1μm increases the light extraction by 51.8%, which is much better improvement prediction compared to the published experimental data. Finally, we find that the appearance of the reflection layer has major effects on light extraction. Ag layer can increase or decrease the light extraction efficiency. From these simulations we find a maximum increase in light extraction of 128% for a l LED with an Ag reflection layer compared to a conventional LED.
Today’s advanced technology allows engineers to fabricate GaN LEDs with various heights, widths, shapes, and materials. Total internal reflection is a key factor in GaN LED design, because all light that is created inside the LED is lost unless it approaches the chip to air interface at an angle less than 23.58° with respect to the normal. The narrow range of angles at which light can successfully escape the chip is a result of the large difference in refractive indices between GaN and air. Adding a layer of ITO to the GaN reduces the difference in refractive indices between steps and increases the critical angle to 28.4°. Transmitting from ITO into epoxy reduces this difference in refractive indices again, bringing the critical angle to 47.9°. Because a higher critical angle should allow more light to escape the LED, we focus on enhancing light extraction efficiency of GaN LED's that utilize an ITO to epoxy interface using FDTD simulations. The simulation results show us that increasing the critical angle to 47.9° improves light extraction by 40%, proving that the critical angle does play a significant role in light extraction. From this initial result, we then compare light extraction efficiencies of ITO and GaN gratings over varied grating periods, and show that adding an Ag reflection layer improves overall efficiency. Finally, we show that the light extraction for LED's utilizing an Ag reflection layer is highly dependent on the sapphire substrate thickness.
This study concentrates on solar light absorption power in a silicon solar cell using a double diffraction triangular
nano-grating. The first grating is located on top of the solar cell and the second grating is located on bottom of the
solar cell above a reflective metallic substrate of Ar (Si<sub>3</sub> N<sub>4</sub> ) (Argon gas mixed with Silicon Nitride). We simulate the
solar cell behavior over varying grating parameters as it absorbs sunlight and compare the average power output
absorbed at the center of the solar cell. Each case simulates a period (A<sub>t</sub> ) that varies from 100nm to 800nm in 100nm
interval for the top lattice, while maintaining the bottom lattice at a constant period (A<sub>b</sub> ). We repeat this procedure
for the bottom lattice, changing the lattice period from 100nm to 800nm in 100nm interval in order to find the
optimized case. We also consider solar spectrum irradiation under wavelengths ranging from 300nm to 1100nm in
50nm intervals. The total power absorption improvement is about 170% compared to the non-grating case, occurring
in the weighted solar cell simulation with top grating period greater than 300nm and bottom grating period of
This study analyzes optical confinement factor and light emitting mode order for three different GaN LEDs:
a conventional LED, thin Film LED, and thin Film LED with a photonic crystal (PhC) grating. For the first
structure, we increase the thickness of AlxGa1-xN from 0 to 600nm, alter the x composition in AlxGa1-xN
from 0.05 to 0.2 in steps of 0.05, and adjust the p-GaN and n-GaN thicknesses each from 0 to 200nm. For
the second structure, we alter the n-GaN substrate thickness from 300-1000nm in steps of 100nm and 1000-
4000nm in steps of 1000nm. These simulations show that increasing the substrate thickness causes the light
emitting mode order to increase. The higher the mode, the more current is needed to make the device emit
light. Higher current leads to shorter device lifetime. The last structure contains a photonic crystal grating
with a period T = 100nm, 230nm, 460nm, 690nm, 920nm, 1500nm, 2000nm, 3000nm and 50% duty cycle.
For each grating period, we display the effects on optical confinement factor and optical field intensity. The
results show that changing the grating period does not affect the mode order, but does affect the optical
field intensity. A larger grating period corresponds to lower optical field intensity. Maximizing optical field
intensity increases the brightness of the device. The simulation method above can be used to improve the
efficiency, brightness, and lifetime of GaN LEDs by reducing the effects of transverse mode coupling and
maximizing the optical field intensity.
In this paper, we use a Finite-Difference Time-Domain GaN LED model to study constant wave (CW)
average power of extracted light. The structure simulated comprises of a 200nm-thick p-GaN substrate,
50nm-thick MQW, 400nm-thick n-GaN substrate, and a 200nm n-GaN two-dimensional Photonic
Crystal(2PhC) grating. We focus on optimizing three design parameters: grating period (A), grating height
(d), and fill factor (FF). In the primary set of simulations, we fix the fill factor at 50% and simulate ten
different grating periods (100 to 1000nm in steps of 100nm) and four different grating heights (50 to
200nm in steps of 50nm), and calculate the average power output of the device. The results from these
simulations show that for both conical and cylindrical gratings, the maxmium light extraction improvement
occurs when A =100nm. In the second set of simulations, we maintain a constant grating period A = 100nm
and sweep the fill factor from 25 to 75%. The results of these simulations show that the fill factor affects
clyindrical and conical gratings differently. As a whole, we see that the nano-structure grating mostly
depends on period, but also depends on height and fill factor. The grating structure improves light
extraction in some cases, but not all.
We study solar-cell designs using nano-grating on both top (transmission) and bottom (reflection) of the
solar cell. First, we perform simulations based on rigorous coupled wave analysis (RCWA) to evaluate
the diffraction top gratins. In RCWA method, we calculate up to 20 harmonics, and sweep the launch
angle of incident light from 0 to 90 degree. The incident light varies from100nm to 1200nm wavelength.
Triangular grating can achieve higher light absorption compared to the rectangular grating. The best top
grating is around 200nm grating period, 100nm grating height, and 50% filling factor, which responses to
37% improvement for triangular grating and 23% for rectangular grating compared to non-grating case.
Then, we use Finite-Difference Time-Domain (FDTD) to simulate transmission/reflection double grating
cases. We simulated triangular-triangular (top-bottom) grating cases and triangular-rectangular (top-bottom)
grating case. We realize solar cell efficiency improvement about 42.4%. For the triangular-triangular
(top-bottom) grating case, the 20% efficiency improvement is achieved. Finally, we present
weighted-light simulation for the double grating for the first time and show the best grating can achieve
104% light improvement, which is quite different from traditional non-weighted calculation.
Writing optimization code for commercially available ray-tracing software, we explore variations of concentrator
geometry where sunlight is first incident onto a stationary primary mirror of circular cross section. The reflected light is
incident onto a smaller, secondary moveable mirror which focuses the light onto a target. Simulations show
concentrations on the order of 30 solar equivalents are possible.