Excitation efficiency and limitations of the luminescence of Eu3+ ions in GaN

Brandon Mitchell; Dolf Timmerman; Jun Tatebayashi; Masaaki Ashida; Shuhei Ichikawa; Yasufumi Fujiwara

REVIEW 4 major objections 6 minor 22 references

Excitation efficiency and limitations of the luminescence of Eu3+ ions in GaN

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper identifies two europium sites in GaN whose very different excitation cross-sections set both the peak and the ceiling of red emission efficiency.

desk verdict Solid quantitative study of Eu:GaN excitation cross-sections and QE limits; the cross-section contrast is likely robust, but inconsistencies in trap identity and numbers need fixing. read the letter →

arxiv 1909.01538 v1 pith:DD6WTMSU submitted 2019-09-04 cond-mat.mtrl-sci

Dolf Timmerman , Brandon Mitchell , Shuhei Ichikawa , Jun Tatebayashi , Masaaki Ashida , Yasufumi Fujiwara This is my paper · ORCID

classification cond-mat.mtrl-sci

keywords Eu-dopedGaNredluminescencephotoluminescencequantumefficiencyexcitationcross-sectionrare-earthdopingcarriertrapmicro-LEDgalliumnitride

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to explain what controls the efficiency of red luminescence from europium-doped gallium nitride, a candidate red emitter on the GaN platform. It reports that the few europium ions sitting in a minority site are about two orders of magnitude easier to excite than the abundant majority-site ions, so they dominate at low pump fluence while the majority site only begins to contribute under strong pumping. The external quantum efficiency is limited at low fluence by an efficient carrier trap that steals excitation energy, and at high fluence by saturation of the scarce efficient sites combined with the tiny cross-section of the majority site. Under pulsed excitation at low temperature, the external quantum efficiency reaches 46%, which makes the material attractive for red micro-LEDs operated at low current density.

What carries the argument

The argument is carried by a two-level saturable-excitation kinetics model. For a short excitation pulse the fraction of excited Eu ions obeys $Eu^* = Eu_{tot}(1-e^{-\sigma_{ex}f})$, and for steady current injection the excited fraction saturates as $Eu^* = Eu_{tot}\sigma_{ex}\tau j/q / (1+\sigma_{ex}\tau j/q)$. Spectral deconvolution separates the minority OMVPE8 peak from the overlapping OMVPE4/7 peak; because the PL lifetimes are similar across sites, the paper assumes similar emission efficiencies and uses the OMVPE8 saturation curve to rescale and subtract the OMVPE7 contribution, isolating the majority OMVPE4 response. The model then attributes the low-fluence rise of quantum efficiency to filling of a competing carrier trap and the high-fluence fall to saturation of the efficient minority sites.

What would settle it

Perform site-selective resonant excitation or calibrated decay measurements to determine the radiative efficiency of the majority OMVPE4 site independently of the equal-lifetime assumption. If its true radiative efficiency is much lower than that of the minority site, the rescaled excited-fraction curves and the claimed two-order-of-magnitude cross-section contrast, and with them the high-fluence efficiency ceiling, would need to be revised.

Watch

Extended reading notes

Core claim

The central claim is a quantitative account of excitation in GaN:Eu. Using a saturating-exponential model, $Eu^* = Eu_{tot}(1-e^{-\sigma_{ex}f})$, for pulsed above-bandgap excitation, the paper derives effective excitation cross-sections of $\sigma_{ex} \approx 1.6\times10^{-15}$ cm$^2$ for the minority OMVPE8 site and $\sigma_{ex} \approx 1.2\times10^{-17}$ cm$^2$ for the majority OMVPE4 site; under current injection the corresponding values are $3.0\times10^{-15}$ cm$^2$ and $6.0\times10^{-18}$ cm$^2$. Because the minority sites, though comprising under ten percent of the europium ions, are so much easier to excite, they dominate the emission at low fluence. The external quantum efficiency is then governed by two competing limits: a carrier trap with an effective cross-section around $10^{-13}$ cm$^2$ suppresses the low-fluence efficiency until it is filled, and the scarcity of the efficient minority sites plus the small cross-section of the majority sites caps the high-fluence efficiency. With this behaviour the external quantum efficiency peaks near a photon fluence of $7\times10^{13}$ cm$^{-2}$, reaching about 0.29 at room temperature and 0.46 at 77 K.

Load-bearing premise

The whole two-order-of-magnitude contrast between the minority and majority europium sites rests on the assumption that all sites emit with the same intrinsic efficiency because their photoluminescence lifetimes look alike; if the rare bright sites actually emit light more efficiently than the abundant sites, the rescaled excited-fraction curves and the predicted efficiency ceiling would not be valid.

Editorial extensions

If this is right

Eu-doped GaN red LEDs should be designed to run at the pump fluence or current density where the carrier trap is filled but the minority sites are not yet saturated, rather than at the highest possible drive.
The measured 46% external quantum efficiency at low temperature is a benchmark for red emitters on GaN; operating in the pulsed, low-fluence regime used here should be part of any comparison.
Reducing the density of the large-cross-section carrier trap (the paper argues it is likely gallium-vacancy-related) should raise the low-fluence quantum efficiency directly.
Because the majority OMVPE4 site has an excitation cross-section near $10^{-17}$ cm$^2$, increasing the europium concentration alone will not raise high-fluence output unless the majority site's excitation pathway is also made more efficient.
Micro-LEDs, which typically run at low current densities, align with the regime where the efficient minority sites dominate, making GaN:Eu a promising red emitter for that application.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same saturable-trap logic implies that pulsed or modulated drive at constant average power should outperform continuous-wave drive in Eu:GaN, because cw excitation lets the competing trap recapture carriers many times during a europium decay; the paper notes this direction but does not measure a duty-cycle series.
If the limiting low-fluence trap is a gallium vacancy, then samples grown or annealed to vary the III/V ratio should show a directly proportional change in the low-fluence quantum efficiency; this is a testable prediction the paper does not perform.
The two-order-of-magnitude cross-section gap suggests that converting majority sites into minority-like configurations, through co-doping or growth conditions that change the local defect environment, could lift the high-fluence efficiency ceiling substantially, a route the authors leave unexplored.
A direct site-selective measurement of the majority site's radiative efficiency would settle whether the reported cross-section contrast survives without the equal-efficiency assumption, and would turn the apparent two-site model into a firmly measured one.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

The paper is worth a read. What's actually new: site-resolved effective excitation cross-sections for Eu3+ in GaN under both optical and electrical pumping, and a fluence-dependent external QE that peaks at ~0.29 at RT and ~0.46–0.48 at 77 K. The authors separate the two dominant Eu sites and show the minority site has a cross-section roughly two orders of magnitude larger than the majority site, and that a large-cross-section trap limits QE at low fluence while saturation of the high-efficiency sites limits it at high fluence. That story is coherent and the fits are internally consistent.

The paper does several things well: top-hat excitation profile, integrating sphere for absolute QE, spectral deconvolution, and the use of pulsed excitation to avoid cw trap dynamics. The saturating-trap model is a natural explanation for the low-fluence behavior, and the comparison to prior low QE values is fair.

Now the soft spots, in proportion. The most serious is the trap identity: the Discussion says the carbon concentration is too low (1.2e16 cm-3) and concludes it's likely a Ga vacancy, while the Conclusion says the limiting trap 'we have related to unintentionally doped carbon.' That's a direct contradiction. The numbers also don't line up: the Abstract says 46%, the Conclusion says 48%; the table lists the majority-site concentration as 5e19 cm-3 while the text says 9e19 cm-3. No error bars anywhere. The carbon concentration figure is given without a reference or measurement description.

I don't fully buy the stress-test's claim that the two-order-of-magnitude cross-section contrast rests on the equal-efficiency assumption. The cross-sections come from the exponential curvature of the saturation curves, so a constant efficiency offset scales the ordinate but not the fitted exponent. The assumption matters, though, for the absolute excited fractions used to estimate the total excited Eu concentration and the trap upper limit, and for the statement about the fraction of minority sites excited. That part should be flagged, but not as a fatal flaw.

All in all, the paper deserves a serious referee. It needs a revision that reconciles the trap conclusion, fixes the numerical inconsistencies, and adds error bars. A reader in nitride optoelectronics or rare-earth doping will get real value from the cross-section data. I'd engage with it.

Referee Report

4 major / 6 minor

Summary. The manuscript reports on the excitation efficiency and external luminescence quantum efficiency (QE) of Eu3+ ions in GaN under optical and electrical excitation. By fitting pulsed-photoluminescence saturation curves with Eq. (3), the authors extract an effective excitation cross-section of 1.6 x 10^-15 cm^2 for the minority OMVPE8 site and 1.2 x 10^-17 cm^2 for the majority OMVPE4 site, suggesting a two-order-of-magnitude contrast. Under electroluminescence, the corresponding values are 3.0 x 10^-15 cm^2 and 6.0 x 10^-18 cm^2. The external QE is measured versus fluence and temperature, reaching about 0.46 at low temperature. The measurements are interpreted with a two-regime model: at low fluence, an efficient carrier trap (the H1 trap) reduces the QE until it is filled; at high fluence, the limited number of high-efficiency minority sites and the small cross-section of the majority site limit the QE. The paper concludes that Eu:GaN is promising for low-current-density applications such as micro-LEDs.

Significance. If the reported cross-section contrast is correct, the work provides quantitative guidance for operating Eu-doped GaN as an efficient red emitter, particularly in micro-LEDs where low current densities are typical. The experimental approach, including absolute QE measurements in an integrating sphere and the decomposition into minority and majority site contributions, is useful and goes beyond earlier qualitative work. The two-regime QE model is plausible and connects the fluence-dependent QE to a saturable trap and to site-specific excitation cross-sections. However, the central quantitative claim rests on an unverified equal-efficiency assumption for the two sites, and the trap-concentration estimate contains internal inconsistencies. The significance is therefore conditional on resolving these issues.

major comments (4)

[Effective excitation cross-section of photoluminescence] The ordinate of Fig. 1c is calibrated by converting measured PL intensity into an absolute excited fraction under the assumption that the emission efficiency of OMVPE8 and OMVPE4/7 is similar because their PL lifetimes are similar. This is not a sufficient condition: the radiative efficiency is the ratio of the radiative rate to the total decay rate, and similar lifetimes do not imply equal radiative rates. The paper's own Fig. 3 shows that the majority site is thermally quenched much more strongly than OMVPE8, which is direct evidence that the nonradiative rates differ. Therefore the fitted OMVPE4 cross-section of 1.2 x 10^-17 cm^2 is not an unambiguous excitation cross-section, and the two-order-of-magnitude contrast between the two sites is not established. The authors should provide an independent calibration of the per-site emission efficiency, for example by resonant excitation or by per-site absolute quantum efficiency measurements, or they should explicitly quantify the systematic uncertainty this assumption introduces.
[Effective excitation cross-section of photoluminescence and Figure 1] The isolation of the OMVPE4 contribution from the OMVPE4/7 peak assumes that OMVPE7 has exactly the same fluence dependence as OMVPE8, differing only by a scaling factor, and that it saturates at the same fluence. Although the manuscript states this is observed, no evidence is shown for the relationship between OMVPE7 and OMVPE8 intensities, and no deconvolution details are provided. Given that OMVPE7 and OMVPE8 are distinct charge states of the same defect configuration (Ref. 8), their capture cross-sections could in principle differ. The authors should justify this scaling or estimate the resulting systematic error in the derived OMVPE4 cross-section.
[Discussion] The derivation of the trap-concentration upper limit NT = 2.0 x 10^17 cm^-3 is not transparent. Using the values quoted in the same paragraph (Eu* = 2.4 x 10^17 cm^-3 at 77 K and QE = 0.46), the nonradiative recombination density is about 2.8 x 10^17 cm^-3, not 2.0 x 10^17 cm^-3. In addition, the assertion that 'all of the efficient trapping centers will be occupied' at the optimum fluence is not supported by any independent measurement. The trap assignment is also internally inconsistent: the Discussion concludes that the competing trap is more likely related to the Ga vacancy, while the Conclusion states that the authors 'related' the trap to unintentionally doped carbon. These inconsistencies need to be resolved because they are load-bearing for the proposed low-fluence loss mechanism.
[Discussion and Table 1] Table 1 lists the majority-site concentration as 5 x 10^19 cm^-3, but the Discussion text in the same section states a concentration of 9 x 10^19 cm^-3, and the Introduction says OMVPE4 comprises about 90% of the total 1 x 10^20 cm^-3 Eu concentration. This factor-of-two inconsistency changes the estimated total excited Eu3+ concentration and propagates directly into the trap-concentration estimate. The text and table should be reconciled.

minor comments (6)

[Experimental] The phrase 'The growth temperature of the of the optical active layer' contains a duplicated article; it should read 'The growth temperature of the optical active layer.'
[Discussion] The capture cross-section of the H1 trap is written as '10^-13 cm^-2' in two places; the unit should be cm^2, not cm^-2.
[Abstract, main text, and Conclusion] The abstract and the temperature-dependence section report a maximum external QE of 46%, while the Conclusion states 48% at 77 K. These values should be made consistent.
[Conclusion] The Conclusion attributes the competing carrier trap to 'unintentionally doped carbon,' which contradicts the Discussion's conclusion favoring Ga vacancies; this needs to be aligned with the main text.
[Experimental] The sentence 'A laser repetition rate of 1 kHz ensured that all excited Eu3+ ions relaxed to the ground state before the next pulse arrived' is grammatically awkward; consider rewriting, for example, 'A laser repetition rate of 1 kHz ensures that...'.
[Discussion] The phrase 'of which a fraction of f * sigma_ex = 0.1 is excited' uses notation that could be confused with a product of two variables; clarify that this is the dimensionless product sigma_ex * f = 0.1.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the cross-sections are obtained by standard least-squares fits to measured saturation curves; the equal-efficiency assumption is an empirical premise, not a derivation that reduces to its own inputs.

full rationale

The paper's derivation chain is self-contained experimental characterization. The minority-site excitation cross-section (1.6e-15 cm2) is extracted by fitting Eq. (3) to the measured OMVPE8 fluence dependence, and the majority-site cross-section (1.2e-17 cm2) is likewise obtained by fitting Eq. (3) to the high-fluence OMVPE4/7 data after rescaling. The rescaling uses an explicitly stated assumption that similar PL lifetimes imply similar emission efficiencies, which allows the OMVPE8 saturation value to calibrate the vertical scale. This is an empirical calibration assumption and a potential source of systematic error, but it is not circular: the fitted OMVPE4 cross-section is not definitionally equal to any input parameter, and the paper does not rename a fitted quantity as a prediction. The trap interpretation borrows the H1 identity from cited literature, but the trap parameters are inferred from an independent QE-versus-fluence measurement, and the trap concentration estimate is a derived upper limit rather than an input. No uniqueness theorem, ansatz, or prior result by the same authors is used to force the central conclusion. The dominant concern in the manuscript is the validity of the equal-efficiency assumption, which belongs to correctness risk, not circularity. Therefore the circularity score is 0.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central quantitative results are four fitted cross-sections and an inferred trap cross-section/concentration. The rate-equation analysis is standard, but the site-separation and equal-efficiency assumptions are imported from prior work or asserted. No new physical entities are introduced; the H1 trap is borrowed from the literature.

free parameters (6)

OMVPE8 (minority site) PL excitation cross-section = 1.6 x 10^-15 cm^2
Fit of Eq. (3) to the pump-fluence dependence of the deconvoluted OMVPE8 peak (Fig. 1b).
OMVPE4 (majority site) PL excitation cross-section = 1.2 x 10^-17 cm^2
Fit of Eq. (3) to the high-fluence part of the OMVPE4/7 peak after assuming OMVPE7 saturates like OMVPE8.
OMVPE8 EL excitation cross-section = 3.0 x 10^-15 cm^2
Fit of Eq. (6) to the current-density dependence of OMVPE8 (Fig. 4b).
OMVPE4 EL excitation cross-section = 6.0 x 10^-18 cm^2
Fit of Eq. (6) after subtracting a rescaled OMVPE7 contribution from the OMVPE4/7 EL peak.
Competing trap effective cross-section = ~10^-13 cm^2
Inferred from the saturation photon fluence (~10^13 cm^-2) in the low-fluence QE rise; assigned to the H1 trap.
Competing trap concentration upper limit = NT = 2.0 x 10^17 cm^-3
Derived from the maximum QE (0.46 at 77 K) and the estimated concentration of excited Eu ions, assuming all traps are filled at the optimum fluence.

assumptions (5)

domain assumption Rate equation model Eq. (1)-(3) for pulsed excitation: relaxation neglected within the pulse, full ground-state recovery between 1 kHz pulses.
State in the effective excitation cross-section section; relies on Ref. 12. If traps or Eu excited states do not fully relax, the saturation fit is biased.
domain assumption All Eu sites have similar emission efficiencies because their PL lifetimes are similar.
Stated in Section 'Effective excitation cross-section of photoluminescence'; load-bearing for scaling OMVPE8 saturation value to OMVPE4/7.
domain assumption OMVPE7 saturates at the same fluence as OMVPE8 and its contribution to OMVPE4/7 can be scaled from OMVPE8 data.
Used to isolate OMVPE4 in both PL and EL; no direct measurement of OMVPE7 alone.
domain assumption The low-fluence QE rise is caused by filling of a single dominant hole trap (H1) with cross-section ~10^-13 cm^2.
Inference from the QE vs fluence shape; H1 identity taken from Refs. 13,14 rather than measured in this sample.
domain assumption The measured carbon concentration (1.2 x 10^16 cm^-3) is accurate and relevant to trap identity.
Stated without method in Discussion; used to rule out carbon and favor Ga vacancy, yet Conclusion attributes the trap to carbon.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Excitation efficiency and limitations of the luminescence of Eu3+ ions in GaN." pith.science (2026). https://pith.science/paper/DD6WTMSU

@misc{pith2026190901538,
  author       = {Pith},
  title        = {Pith review of: Excitation efficiency and limitations of the luminescence of Eu3+ ions in GaN},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DD6WTMSU}},
  note         = {Machine review of arXiv:1909.01538}
}

read the original abstract

The excitation efficiency and external luminescence quantum efficiency of trivalent Eu3+ ions doped into gallium nitride (GaN) was studied under optical and electrical excitation. For small pump fluences it was found that the excitation of Eu3+ ions is limited by an efficient carrier trap that competes in the energy transfer from the host material. For large pump fluences the limited number of high-efficiency Eu3+ sites, and the small excitation cross-section of the majority Eu3+ site, limit the quantum efficiency. At low temperatures under optimal excitation conditions, the external luminescence quantum efficiency reached a value of 46%. These results show the high potential for this material as an efficient light emitter, and demonstrates the importance of the excitation conditions on the light output efficiency.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

[1]

Zhu, W. et al. High-Power Eu-Doped GaN Red LED Based on a Multilayer Structure Grown at Lower Temperatures by Organometallic Vapor Phase Epitaxy. MRS Adv. 2, 159–164 (2017)

work page 2017
[2]

& Fujiwara, Y

Mitchell, B., Dierolf, V., Gregorkiewicz, T. & Fujiwara, Y. Perspective: Toward efficient GaN-based red light emitting diodes using europium doping. J. Appl. Phys. 123, 160901 (2018)

work page 2018
[3]

Fleischman, Z. et al. Excitation pathways and efficiency of Eu ions in GaN by site-selective spectroscopy. Appl. Phys. B 97, 607–618 (2009)

work page 2009
[4]

& Dierolf, V

Woodward, N., Nishikawa, A., Fujiwara, Y. & Dierolf, V. Site and sample dependent electron–phonon coupling of Eu ions in epitaxial-grown GaN layers. Opt. Mater. (Amst). 33, 1050–1054 (2011)

work page 2011
[5]

& Dierolf, V

Fujiwara, Y. & Dierolf, V. Present understanding of Eu luminescent centers in Eu-doped GaN grown by organometallic vapor phase epitaxy. Jpn. J. Appl. Phys. 53, (2014)

work page 2014
[6]

Roqan, I. S. et al. Identification of the prime optical center in GaN : Eu 3 +. Phys. Rev. B 81, 085209 (2010)

work page 2010
[7]

Wakamatsu, R. et al. Luminescence properties of Eu-doped GaN grown on GaN substrate. Jpn. J. Appl. Phys. 52, 08JM03 (2013)

work page 2013
[8]

Mitchell, B. et al. Charge state of vacancy defects in Eu-doped GaN. Phys. Rev. B 96, 064308 (2017)

work page 2017

Show all 22 references

[9]

Zhu, W. et al. Re-Excitation of Trivalent Europium Ions Doped into Gallium Nitride Revealed through Photoluminescence under Pulsed Laser Excitation. ACS Photonics 5, 875–880 (2018)

2018
[10]

Zhu, W. et al. Enhanced photo/electroluminescence properties of Eu-doped GaN through optimization of the growth temperature and Eu related defect environment. APL Mater. 4, 056103 (2016)

2016
[11]

& Fujiwara, Y

Nishikawa, A., Kawasaki, T., Furukawa, N., Terai, Y. & Fujiwara, Y. Room-Temperature Red Emission from a p-Type/Europium-Doped/n-Type Gallium Nitride Light-Emitting Diode under Current Injection. Appl. Phys. Express 2, 071004 (2009)

2009
[12]

Mitchell, B. et al. Color-Tunablility in GaN LEDs Based on Atomic Emission Manipulation under Current Injection. ACS Photonics 6, 1153–1161 (2019)

2019
[13]

& Shiojima, K

Honda, U., Yamada, Y., Tokuda, Y. & Shiojima, K. Deep levels in n-GaN Doped with Carbon Studied by Deep Level and Minority Carrier Transient Spectroscopies. Jpn. J. Appl. Phys. 51, 04DF04 (2012)

2012
[14]

Auret, F. D. et al. Electrical characterisation of hole traps in n-type GaN. Phys. status solidi 201, 2271–2276 (2004)

2004
[15]

Tokuda, Y. et al. Hole traps in n-GaN detected by minority carrier transient spectroscopy. Phys. Status Solidi Curr. Top. Solid State Phys. 8, 2239–2241 (2011)

2011
[16]

Armstrong, A. et al. Impact of deep levels on the electrical conductivity and luminescence of gallium nitride codoped with carbon and silicon. J. Appl. Phys. 98, 053704 (2005)

2005
[17]

de Boer, W. D. A. M. et al. Optical excitation and external photoluminescence quantum efficiency of Eu3+ in GaN. Sci. Rep. 4, 5235 (2014)

2014
[18]

4f–4f Luminescence of Rare-Earth Centers in II–VI Compounds

Boyn, R. 4f–4f Luminescence of Rare-Earth Centers in II–VI Compounds. Phys. status solidi 148, 11–47 (1988)

1988
[19]

& Fujiwara, Y

Wakamatsu, R., Lee, D., Koizumi, A., Dierolf, V. & Fujiwara, Y. Luminescence properties of Eu-doped GaN under resonant excitation and quantitative evaluation of luminescent sites. J. Appl. Phys. 114, 043501 (2013)

2013
[20]

& Boyn, R

Zimmermann, H. & Boyn, R. Donor-Type Tm Centres in ZnS Crystals. Phys. status solidi 139, 533–545 (1987)

1987
[21]

Mitchell, B. et al. The role of donor-acceptor pairs in the excitation of Eu-ions in GaN:Eu epitaxial layers. J. Appl. Phys. 115, 204501 (2014)

2014
[22]

Singh, A. K. et al. Eu–Mg defects and donor–acceptor pairs in GaN: photodissociation and the excitation transfer problem. J. Phys. D. Appl. Phys. 51, 065106 (2018). Figure 1 Top: PL spectra for an excitation wavelength of 350 nm for various pump fluences. Bottom: Intensity of ...

2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Zhu, W. et al. High-Power Eu-Doped GaN Red LED Based on a Multilayer Structure Grown at Lower Temperatures by Organometallic Vapor Phase Epitaxy. MRS Adv. 2, 159–164 (2017)

work page 2017

[2] [2]

& Fujiwara, Y

Mitchell, B., Dierolf, V., Gregorkiewicz, T. & Fujiwara, Y. Perspective: Toward efficient GaN-based red light emitting diodes using europium doping. J. Appl. Phys. 123, 160901 (2018)

work page 2018

[3] [3]

Fleischman, Z. et al. Excitation pathways and efficiency of Eu ions in GaN by site-selective spectroscopy. Appl. Phys. B 97, 607–618 (2009)

work page 2009

[4] [4]

& Dierolf, V

Woodward, N., Nishikawa, A., Fujiwara, Y. & Dierolf, V. Site and sample dependent electron–phonon coupling of Eu ions in epitaxial-grown GaN layers. Opt. Mater. (Amst). 33, 1050–1054 (2011)

work page 2011

[5] [5]

& Dierolf, V

Fujiwara, Y. & Dierolf, V. Present understanding of Eu luminescent centers in Eu-doped GaN grown by organometallic vapor phase epitaxy. Jpn. J. Appl. Phys. 53, (2014)

work page 2014

[6] [6]

Roqan, I. S. et al. Identification of the prime optical center in GaN : Eu 3 +. Phys. Rev. B 81, 085209 (2010)

work page 2010

[7] [7]

Wakamatsu, R. et al. Luminescence properties of Eu-doped GaN grown on GaN substrate. Jpn. J. Appl. Phys. 52, 08JM03 (2013)

work page 2013

[8] [8]

Mitchell, B. et al. Charge state of vacancy defects in Eu-doped GaN. Phys. Rev. B 96, 064308 (2017)

work page 2017