197

DOI: 10.1557/adv.2016.134

Recombination Dynamics

of InGaN/GaN

Quantum

Wells

With Different Well Thickness

X. C. Wei, L. Zhang, N. Zhang, J. X. Wang and J. M. Li

Research and Development Center for Solid State Lighting, Institute of Semiconductors, Chinese

Academy of Sciences,

State Key Laboratory of Solid State Lighting,

Beijing Engineering Research Center for the 3rd Generation Semiconductor Materials and

Application,

Beijing 100083, People's Republic of China

ABSTRACT

Recombination dynamics of InGaN/GaN multiple quantum wells (MQWs) with different

well thickness have been studied. From the behaviour of temperature dependent

photoluminescence, we find that the activation energy decreases with the well thickness

increasing. In addition, with temperature changing from 10K to room temperature, the “W”

shape of full width of half maximum is also thickness related, and it becomes more obvious with

the well thickness increasing. These results indicate that the dominant recombination dynamics

change from exciton localization to quantum confined stark effect with well thickness increasing.

From our measurement, the InGaN/GaN

MQWs

with 3nm thickness seems a turning point,

which shows the best optimized optical and structural properties.

INTRODUCTION

GaN-based multiple quantum wells have been attracting lots of attention as the key material

for the fabrication of high brightness light emitting diodes (LEDs) and continuous wave (cw)

laser diodes because of the advantage of tuning ability of the full spectrum. [1, 2] The 254

lm/W LED and InGaN/GaN MQW LDs with a life time of more than 10,000 hours for cw

operation at room temperature have already been reported. But in contrast to the striking

technology development, the underlying emission mechanism of these devices is still not fully

understood. Two different radiative recombination mechanisms are generally accepted for

InGaN/GaN MQWs grown on the lattice-mismatched substrates. One is attributed to the

exciton localization effect. Experiments showed [3] that emission from InGaN may be due to

recombination of excitons localized at potential minima in the quantum well, and these

exciton localizations at deep traps is mainly originated from In-rich regions acting as quantum

dots. It has also been shown that the localized excitons within indium-rich regions resulting from

partial phase segregation in InGaN alloys are considered to prevent them from reaching

nonradiative recombination sites and play an important role for spontaneous emission. The

other is related to the internal piezoelectric field induced quantum confined stark effect (QCSE)

[4], which is mainly related to the strain caused by the lattice mismatch between GaN and

198

InGaN. The emission peak demonstrates an anomalous blue shift with temperature increasing,

and all these mechanisms and accordingly the electrical property and polarized field will be

strongly influenced by the well thickness.

In order to clarify the underlying physics of light emission of Nitride-based multiple

quantum wells, in this study, we prepared InGaN/GaN multiple quantum wells with different

well thickness from 1nm to 5nm and measured the temperature dependent photoluminescence

(PL). We found that for different thickness, the dominate recombination mechanism will be

changed, and sample with 3nm QWs is the break point. This has also been confirmed by time

resolved photoluminescence.

EXPERIMENTAL DETAILS

The InGaN/GaN MQWs were grown on a (0001)-oriented Patterned Sapphire Substrate

using metalorganic chemical vapor deposition (MOCVD). The precursors of Ga, In, N, and Si

were trimethylgallium (TMGa), trimethylindium (TMIn), ammonia (NH

), and silane (SiH

respectively. The QWs were grown under N

ambient

after a 1.5-μm-thick undoped GaN

buffer layer and a 2.5-μm-thick Si-doped GaN layer. In order to improve the crystal quality, the

active region of our samples include three parts: 1) First, 2 pre-wells with 2-nm-thick InGaN

wells and 40-nm-thick GaN barriers in order to decrease the strain between quantum well and

n-type GaN, improve the MQW crystal quality and reduce the Indium fraction localization; 2)

Then, 5 MQWs with 2-nm-thick InGaN wells and 14-nm-thick GaN barriers in order to reduce

the electron overflow; 3) Finally, 6 MQWs with different width InGaN wells (from 1 to 5nm)

and 14-nm-thick GaN barriers. During the sample growth, we kept the trimethylindium flow

and temperature fixed, and the nominal In composition in the active region of these samples

were roughly 13%.

During the measurements of the temperature dependent photoluminescence, the samples were

mounted in a closed-cycle He cryostat and the temperature was controlled from 10K to 295 K.

A 325 nm cw laser was used as an excitation light source with the spot size of ~50 μm, and the

excitation power is 30 mW. The photoluminescence signal from the sample was dispersed by a

PI sp500i monochromator and detected by a PI CCD detector. Time resolved

photoluminescence (TRPL) was performed using FLS920 system with a 375nm pulse laser.

DISCUSSION

Figure 1 shows the photoluminescence spectra of the samples with different quantum well

thickness at 295K. The consistent peaks appeared at ~430nm for these samples are due to the 5

MQWs with 2-nm-thick InGaN wells. We can see that the relative intensity of the MQWs firstly

increased with the thickness increasing from 1nm to 3nm, and afterwards decreased with the

thickness further increasing to 5nm. In addition, the main peak positions, which are due to 6

MQWs, show a red shift of 0.1eV/nm from 3.069eV to 2.616eV with increasing the well

thickness. It has been reported that the strain in MQWs increases with increasing the well

199

thickness [5]. Since the In content for all the samples were kept the same, the observed red shift

may be due to the increase of the strain effect and the decrease of bandgap with well thickness

increasing.

Figure 1 the photoluminescence of the samples at the room temperature (295K) excited by

325nm laser,

The red shifts of the photoluminescence emission peaks mentioned above also change with

the temperature, as shown in Figure 2. For MQWs with 1nm well thickness, the dependence of

peak energy on the temperature doesn’t follow the Vashini equation[6], and the peak positions

are almost constant within the whole temperature range, except for a small blue shift ~5meV

observed for MQWs with 1nm well thickness at the high temperature range. This indicates that

the exciton localization effects are dominant recombination dynamics for the thin MQW samples.

With temperature increasing, the redistribution of the excitons to the lower energy states will

shift the emission of exciton to a higher energy. But with the QW thickness increasing, a red

shift of peak energy started to appear and became larger with temperature increasing, especially

for 5nm QWs sample in our measurement, which showed a red shift ~ 80meV at 295K. This not

only indicates that the influence of the bandgap shrinkage of InGaN becomes more obvious, but

with the thickness increasing, the strain effect becomes bigger and quantum confined stark effect

(QCSE) starts to play an important role in the recombination dynamics. When the quantum well

thickness is very thin, the exciton localization effect is more likely to be dominant in the

quantum wells and in this case, the peak energy would be more stable. With the well thickness

increasing, exciton binding energy reduces and exciton localization effect becomes less effective

and QCSE start to dominate, and as a result, the peak energy becomes more changeable.

The FWHMs of these samples with changing the temperature have also been compared and the

results show a similar trend as the peak energy, as seen in Figure 3. Different mechanism was

dominant within the different temperature range. For example, MQWs with 4nm QWs had more

obvious “W” shaped temperature dependences of the FWHM from 10K to 300K, as normally

observed in InGaN MQWs [7]. While for those with well thickness 2nm, 3 nm, because of a

stronger localization effect, these W-shape temperature dependences of FWHMs were decreased.

The sample with well thickness 3nm is like a turning point. When the thickness is less than 3nm,

200

the exciton localization is the dominant effect, while when the thickness is more than 3nm, the

QCSE becomes more effective.

Figure 2 the degree of the peak shift with different thickness

Figure 3 the FWHM as a function of temperature for MQWs with well thickness 2nm, 3nm, 4nm

In addition, with temperature increasing, the thermal quenching of the integrated

photoluminescence intensity in InGaN MQWs is dominated by the activation of nonradiative

recombination processes. If there are several kinds of nonradiative centers, the following

Arrhenius plot expression is generally used to fit the integrated photoluminescence intensity I (T)

[8, 9, 10, and 11]:

Figure 4 shows the experimental and simulated normalized integrated photoluminescence

intensity as a function of Arrhenius equation for MQWs with different thickness. The points are

experimental data, and lines are corresponding to the simulation curve. The obtained activation

energy of all the samples is shown in the inset. Consistent with previous hypothesis, the InGaN

MQWs with 1nm well thickness show the strongest localization energy. While except for the

2nm well thickness, the activation energy decreased with increasing the well thickness. It has

been reported that the dominant quenching mechanism of the InGaN related photoluminescence

has been attributed to the thermionic emission of the photocarriers over the effective potential

barriers caused by potential fluctuations. While for InGaN MQWs with the well thickness

increasing, the thermal activation energy decreased and the thermal quenching of

201

photoluminescence intensity becomes easier. In other words, there is a dramatic decrease in the

thermal activation energy for the sample with 5nm well thickness having highest In composition.

But for sample with 2nm well thickness, it have a lower activation energy than sample with 3nm

QWs, the reason may be due to the influence of crystal quality, as stated in previous section.

Figure 4 Normalized Integrated photoluminescence intensity as a function of 1/T

(The inset is the plot of activation energy as a function of well thickness)





BAii

TkEC

exp1

(1)

Where I

is the integrated PL intensity at low temperature, C

are the constants related to the

density of nonradiative recombination centers, E

are the activation energies of the

corresponding nonradiative centers, and k

is the Boltzmann’s constant. Here, assuming one

dominate process, Equation (1) was modified as follows:



)/exp(1

TkEC



(2)

Figure 5 TRPL spectra of 5 samples at room temperature

202

In order to further investigate the influence of transient behavior of the recombination, we

also measured the time resolved photoluminescence. Figure 5 shows the results. The

fluorescence lifetime increases for MQWs with thickness increasing from 1nm to 3nm and

decreases afterwards. Non-radiative recombination becomes more active for these samples at

room temperature. Since the sample with 3nm QWs present a better crystal quality than the other

samples, less non-radiative centers would be involved in the carrier recombination processes,

and accordingly a longer lifetime would be observed. This result is also consistent with previous

structural and integrated PL measurement, showing that 3nm QWs are the optimized well

thickness to improve internal quantum efficiency.

CONCLUSIONS

In summary, we have investigated the carrier transports and recombination mechanism of

the MOCVD-grown InGaN/GaN MQWs with different thicknesses within the temperature

range from 10 to 295 K. The dependences of the emission energy and FWHMs as a function of

the photoluminescence temperature show that the carrier transports and recombination

mechanism change with the well thickness increasing. When the thickness is less than 3nm,

the exciton localization is the dominant effect. While with the thickness increasing and strain

effect becomes larger and the QCSE start to be dominant for carrier recombination. This

hypothesis is also consistent with the time resolved photoluminescence results. The well

thickness of 3nm is preferable. This study provides a useful guidance to fabricate a

high-performance LED with high-quantum efficiency.

REFERENCES

[1] Nakamura S, Science 281, 956 (1998)

[2] Watson-Parris D, Godfrey M J and Dawson P, Phys. Rev. B 83, 115321(2011)

[3] Zhao H P, Liu G Y and Tansu N, Appl. Phys. Lett. 97, 131114 (2010)

[4] Noh Y K, Kim M D and Oh J E, J. Appl. Phys. 110, 123108 (2011)

[5] Wang T, Bai J, Sakai S, and Ho J K, Appl. Phys. Lett. 78, 2617 (2001)

[6] Schomig H, Halm S, Forchel A, Bacher G, Off J and Scholz F, Phys. Rev. Lett. 92, 106802

(2004)

[7] Wang H L, Ji Z W, Qu S, Wang G, Jiang Y Z, Liu B L, Xu X G, and Mino H, Optics Express

20, 3932 (2012)

[8] Sun Y, Cho Y H, Kim H M, Kang T W, Kwon S Y and Yoon E, J. Appl. Phys. 100, 043520

(2006)

[9] Hugues M, Damilano B, Duboz J Y and Massies, J Phys. Rev. B 75, 115337 (2007)

[10] Hao M, Zhang J, Zhang X H and Chua S, Appl. Phys. Lett. 81, 5129 (2002)

[11] Han S H, Cho CY, Lee S J, Park T Y, Kim T H, Park S H, Kang S W, Kim J W, Kim Y C

and Park S J, Appl. Phys. Lett. 96, 051113 (2010)