Search for: All records

A high efficiency, high brightness, and robust micro or sub-microscale red light emitting diode (LED) is an essential, yet missing, component of the emerging virtual reality and future ultrahigh resolution mobile displays. We report, for the first time, to our knowledge, the demonstration of an N-polar InGaN/GaN nanowire sub-microscale LED emitting in the red spectrum that can overcome the efficiency cliff of conventional red-emitting micro-LEDs. We show that the emission wavelengths of N-polar InGaN/GaN nanowires can be progressively shifted from yellow to orange and red, which is difficult to achieve for conventional InGaN quantum wells or Ga-polar nanowires. Significantly, the optical emission intensity can be enhanced by more than one order of magnitude by employing anin situannealing process of the InGaN active region, suggesting significantly reduced defect formation. LEDs with lateral dimensions as small as$\sim 0.75\mathrm{\mu m}$, consisting of approximately five nanowires, were fabricated and characterized, which are the smallest red-emitting LEDs ever reported, to our knowledge. A maximum external quantum efficiency$\sim 1.2\%$was measured, which is comparable to previously reported conventional quantum well micro-LEDs operating in this wavelength range, while our device sizes are nearly three to five orders of magnitude smaller in surface area.

 

Compressibility and viscosity of polymer feedstock are critical to their volumetric flow rate, weld strength, and dimensional accuracy in material extrusion additive manufacturing. In this work, the compressibility and viscosity of an acrylonitrile butadiene styrene (ABS) material is characterized with an instrumented hot end design. Experiments are first performed with a blocked nozzle to characterize the compressibility behavior. The results closely emulate the pressure-volume-temperature (PVT) behavior of a characterized generic ABS. Experiments are then performed with an open nozzle over a range of volumetric flow rates and temperatures. The static pressure data is fit to power-law, Ellis, and Cross viscosity models and the dynamic melt pressure data is then used to jointly fit material constitutive models for compressibility and viscosity. The results suggest that the joint fitting substantially improves the fidelity relative to the separately characterized viscosity and compressibility. The implemented methods support material extrusion process simulation and control including real-time identification of process faults such as (1) limited melting capacity of the hot end, (2) skipping (grinding) of the extruder drive gears, (3) low initial nozzle temperature, (4) varying flow rates associated with the intermeshing gear tooth velocity profile, and (5) delays and reduced melt pressures due to drool prior to extrusion. The ability to monitor the printing process for faults in real time, such as that presented in this work, is critical to born qualified parts. Additionally, these approaches can be used to screen new materials and identify optimal processing conditions that avoid these process faults. 

The nozzle pressure was monitored in a fused filament fabrication process for the printing of high impact polystyrene. The contact pressure, defined as the pressure applied by the newly deposited layer onto the previous layer, is experimentally calculated as the difference between the pressure during printing and open discharge at the same volumetric flow rates. An analytical method for estimating the contact pressure, assuming one-dimensional steady isothermal flow, is derived for the Newtonian, power-law, and Cross model dependence of shear rates. A design of experiments was performed to characterize the contact pressure as a function of the road width, road height, and print speed. Statistical analysis of the results suggests that the contribution of the pressure driven flow is about twice that of the drag flow in determining contact pressure, which together describe about 60% of the variation in the observed contact pressure behavior. Modeling of the elastic and normal stresses at the nozzle orifice explains an additional 30% of the observed behavior, indicating that careful rheological modeling is required to successfully predict contact pressure. 

The weighted constraint satisfaction problem (WCSP) is a powerful mathematical framework for combinatorial optimization. The branch-and-bound search paradigm is very successful in solving the WCSP but critically depends on the ordering in which variables are instantiated. In this paper, we introduce a new framework for dynamic variable ordering for solving the WCSP. This framework is inspired by regression decision tree learning. Variables are ordered dynamically based on samples of random assignments of values to variables as well as their corresponding total weights. Within this framework, we propose four variable ordering heuristics (sdr, inv-sdr, rr and inv-rr). We compare them with many state-of-the-art dynamic variable ordering heuristics, and show that sdr and rr outperform them on many real-world and random benchmark instances.