Table 1 Optimization results regarding color point deviation \(d^\alpha\), and power \(P^\alpha\) of algorithm 1, and TS-EMO for \(\alpha = 25^\circ\), and after local refinement.

From: Directional emission of white light via selective amplification of photon recycling and Bayesian optimization of multi-layer thin films

 

\(\alpha \,(^\circ )\)

25

45

90

\(P^\alpha \text { (W)}\)

Reference

0.201

0.559

1.052

TS-SOO

\(\mathbf {0.259}\)

0.640

0.985

TS-EMO

0.210

0.569

0.978

\(d^\alpha\) (1)

Reference

0.0015

0.0071

0.0146

TS-SOO

\(\mathbf {0.0007}\)

0.0230

0.0275

TS-EMO

0.0025

0.0262

0.0281

  1. The designs featuring a multi-layer thin film are compared with the reference design, where only a color point optimization is conducted. In addition to the direct optimization objectives regarding \(\pm 25^\circ\), we also report the color point deviations and energies for \(\pm 45^\circ\) and \(\pm 90^\circ\) observed for the respective optimized MTLFs. For \(\pm 25^\circ\), highest power and lowest color point deviations are given in bold.
Back to article page