Archives

  • 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-07
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • br Experimental Commercial Luxeon Rebel ES

    2018-11-02


    Experimental Commercial Luxeon Rebel ES LEDs produced by Philips Lumileds have been the subject of our investigation. The area of the emitting chip was 1340 µm × 1340 µm. This device is known as the thin-film flip-chip InGaN/GaN multiple-quantum-well LED (TFFC LED) [3]. To provide efficient heat removal from the active region, the LED chip is mounted on a ceramic submount with a p-layer and then the submount is soldered on a metal-core printed circuit board (MCPCB). The MCPCB is in turn mounted onto an outside bulk radiator. The general view of the LED and the photomicrograph of the operating LED chip are shown in Fig. 1.
    Simulation To interpret the experimental results, we carried out coupled hybrid 1D/3D simulations of the current spreading in the LED chip and heat transfer in the LED chip and the submount [8]. The active area was regarded as a thin layer with the Shockley dependence of the local current density on the local applied p–n junction bias voltage where m is the ideality factor, and the saturation current is assumed to increase exponentially with temperature
    Here ħω is a mean energy of emitted photons; j0 is a fitting parameter for the magnitude of the saturation current which was adjusted to fit experimental turn-on voltage value. The internal quantum efficiency () dependence on the current density was approximated by the ABC-model where the recombination constants A = 1 × 107 s–1, B = 2 × 10–11 cm3/s, and C = 3 × 10–30 cm6/s. Here n is a concentration of electron-hole pairs injected into the LED active region; d is an effective thickness of the total active region; q – cytokine receptor charge. The coupled problem of the current spreading and heat transfer in the LED chip was solved cytokine receptor numerically using the commercial SimuLED software package [9]. The sheet resistance of the n-GaN contact layer being equal to 22 Ω/□, and the specific resistance of the contact formed to p-GaN being equal to 10–3  Ω cm2, were chosen, respectively, as the input parameters. In carrying out the simulation, we assumed the heat to be released through the bottom surface of the submount where the boundary conditions of the third kind for the heat flux were fulfilled. The heat-transfer coefficient was chosen to be 5 × 104  W/(m2 K) fitting the overall thermal resistance to the measured value. The submount thermal conductivity was assumed to be of 50 W/(m K). The thermal resistance was calculated as a ratio of the average overheating of the active region to the total electric power supplied to LED, in order to keep consistence with our experimental data. It is worthy to note that the computations should include the ceramic submount to enable lateral heat transfer. It is important for the correct prediction of the lateral temperature distribution, as the bulk submount provides more intensive lateral heat transfer than the chip with removed sapphire substrate, the chip thickness being only ∼(3 – 5) μm after Pacinian corpuscles removal. The computed active region overheating and current density profiles in the cross-section S simulated for various driving currents are plotted in Figs. 6c and 7с, respectively. These figures show the lateral current density non-uniformity to increase dramatically with current. At the driving currents of 700 mA and higher, the maximum current density at the mesa edge is more than two times higher than the typical current density between the mesas. The current density next to the external chip boundary is even lower. As a rule, the heat generation rate in the active region is roughly proportional to the current density with some deviation from the linearity coming from the lateral variation of the internal quantum efficiency and . Nevertheless, the lateral distribution of the overheating is predicted to be much more uniform than that of the current density. Simulation shows that the lateral temperature variation does not exceed ∼20% of the mean overheating at the driving current as high as 2 A. This latter fact is attributed to intensive lateral heat transfer in the ceramic submount indicating that its thermal properties are essential for the overall thermal management of the LED chip.