Accepted Manuscript

Advantages of InGaN/GaN multiple quantum wells with two-step grown low

temperature GaN cap layers

Yadan Zhu, Taiping Lu, Xiaorun Zhou, Guangzhou Zhao, Hailiang Dong, Zhigang
Jia, Xuguang Liu, Bingshe Xu

PII: S0749-6036(17)31161-8

DOI: 10.1016/j.spmi.2017.07.054

Reference: YSPMI 5165

To appear in: Superlattices and Microstructures

Received Date: 10 May 2017

Revised Date: 25 July 2017

Accepted Date: 26 July 2017

Please cite this article as: Yadan Zhu, Taiping Lu, Xiaorun Zhou, Guangzhou Zhao, Hailiang Dong,
Zhigang Jia, Xuguang Liu, Bingshe Xu, Advantages of InGaN/GaN multiple quantum wells with two-
step grown low temperature GaN cap layers, Superlattices and Microstructures (2017), doi: 10.1016
/j.spmi.2017.07.054

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Advantages of InGaN/GaN multiple quantum wells with two-

step grown low temperature GaN cap layers

Yadan Zhu1,2 Taiping Lu1,2,*, Xiaorun Zhou1,2, Guangzhou Zhao1,2, Hailiang Dong1,2,
Zhigang Jia1,2, Xuguang Liu1,3, and Bingshe Xu1,2,4,*

1Key Laboratory of Interface Science and Engineering in Advanced Materials,

Taiyuan University of Technology, Ministry of Education, Taiyuan 030024, China
2Research Center of Advanced Materials Science and Technology, Taiyuan
University of Technology, Taiyuan 030024, China
3College of Chemistry and Chemical Engineering, Taiyuan University of Technology,
Taiyuan 030024, China
4Institute of Atomic and Molecular Science, Shaanxi University of Science and
Technology, Xi'an 710021, China
E-mail: lutaiping@tyut.edu.cn and xubs@tyut.edu.cn

Abstract
Two-step grown low temperature GaN cap layers (LT-cap) are employed to improve
the optical and structural properties of InGaN/GaN multiple quantum wells (MQWs).
The first LT-cap layer is grown in nitrogen atmosphere, while a small hydrogen flow
is added to the carrier gas during the growth of the second LT-cap layer. High-
resolution X-ray diffraction results indicate that the two-step growth method can
improve the interface quality of MQWs. Room temperature photoluminescence (PL)
tests show about two-fold enhancement in integrated PL intensity, only 25 meV blue-
shift in peak energy and almost unchanged line width. On the basis of temperature-
dependent PL characteristics analysis, it is concluded that the first and the second LT-
cap layer play a different role during the growth of MQWs. The first LT-cap layer
acts as a protective layer, which protects quantum well from serious indium loss and
interface roughening resulting from the hydrogen over-etching. The hydrogen gas
employed in the second LT-cap layer is in favor of reducing defect density and
 ACCEPTED MANUSCRIPT

indium segregation. Consequently, interface/surface and optical properties are

improved by adopting the two-step growth method.

Key words: GaN cap layer, quantum wells, hydrogen, InGaN

1. Introduction
Indium gallium nitride (InGaN) has proven to be a very important material for
light emitting devices, such as blue/green light emitting diodes and laser diodes [1,2].
However, the growth of both high indium content and high quality InGaN films still
faces great challenges owning to the large lattice mismatch and low miscibility of
GaN and InN [3,4]. Furthermore, because of the weak bond strength and high
volatility of InN, the growth temperature of InGaN layers is usually reduced to obtain
high indium content [5-7]. In order to improve the quality of the multiple quantum
wells (MQWs), the growth temperature of GaN quantum barrier (QB) is usually
(~100℃) higher than that of InGaN QW [8,9]. However, indium atoms can easily
desorb from the QW surface during the temperature ramp-up process, resulting in the
reduction of indium incorporation efficiency [10,11]. Generally, depositing a cap
layer after the QW is an effective way to alleviate indium desorption [12,13]. The
GaN cap layer (LT-cap) grown at low temperature also brings some drawbacks, such
as a large number of defects and serious indium segregation, which deteriorate the
optical quality of MQWs [14,15]. It was reported that introducing hydrogen into
interruption process or QB growth can remove indium-rich clusters and reduce defect
density, thus enhance room temperature photoluminescence (PL) intensity [15-17]. In
our previous study, a small hydrogen flow was added to the carrier gas during the
growth of GaN LT-cap layer, which was effective to reducing non-radiative
recombination centers and indium segregation in QWs [18]. Nevertheless, hydrogen
would quickly react with indium atoms at QW surface, and then lead to the sharp
decrease of indium content, the large blue-shift and broadening of PL spectra [18,19].
In this paper, a two-step growth method of the LT-cap layer is employed during
the growth of InGaN/GaN MQWs. The first LT-cap layer grown in nitrogen
atmosphere is suggested to act as a protective layer, which protects the MQWs from
serious indium loss and interface roughening as a result of hydrogen over-etching.
 ACCEPTED MANUSCRIPT

The second LT-cap layer grown under a small hydrogen flow is suggested to reduce
non-radiative recombination centers and alleviate indium segregation in MQWs.
2. Experimental
The InGaN/GaN MQW structures were grown on c-plane (0001) sapphire
substrates by metal-organic chemical vapor deposition system (MOCVD). The
structure consisted of a 3.2-μm-thick nominally undoped GaN layer, and six pairs of
InGaN/GaN MQWs with nominal 2.3-nm-thick InGaN wells separated by 10.5-nm-
thick, lightly Si-doped (n-doping=3×1017 cm-3) GaN barriers. InGaN wells and GaN
barriers were grown at 730 and 850˚C, respectively. A 1.0-nm-thick GaN cap layer
was inserted between QW and QB layer, grown at the same temperature as that of
QW layer. Two samples with cap layers grown at different conditions were prepared.
The cap layer of sample A was grown in pure N2 carrier gas, while a two-step cap
layer growth method was employed in sample B. The growth parameters of the first
0.5-nm-thick LT-cap layer were the same as those of sample A, and the second 0.5-
nm-thick LT-cap layer was grown in a mixture N2/H2 carrier gas, of which hydrogen
flow was 200 sccm.
High-resolution X-ray diffraction (HRXRD) measurement was performed using
PANalytical Empyrean instrument to obtain the structure parameters. The PL
properties were characterized by 325 nm He-Cd continuous wave laser, and the
excitation power density was about 4.0 W/cm2. The temperature dependence of the
luminescence spectra was measured from 10 to 300 K in a closed loop He cryostat.
The luminescence was dispersed by a triple grating 50 cm monochromator and
detected by a GaAs photomultiplier tube using conventional lock-in technique. The
surface morphology was studied by atomic force microscopy (AFM) (SPA-300HV)
using tapping mode.
3. Results and discussion
The HRXRD ω/2θ scan curves for the (0002) plane reflection from the two
samples are shown in Fig. 1(a). The high order satellite peaks and Pendellösung
fringes are clearly observed for the two samples, indicating the sharp interface
between InGaN well and GaN barrier. The average indium content and well thickness
 ACCEPTED MANUSCRIPT

obtained by fitting the measured curves are 10.9% and 2.30 nm for sample A, and
8.1% and 2.28 nm for sample B, respectively. The period thickness of the two MQWs
samples keeps around 13.8 nm. Because the thickness of the first LT-cap layer is just
0.5 nm, the InGaN well surface may be partially covered by GaN. Hence, the indium
atoms at the uncovered sites may react with hydrogen during the growth of the second
LT-cap layer [18]. The presence of hydrogen can also reduce indium tails in the LT-
cap layer region, and therefore contributes to the decrease of average indium content
in MQWs [18,19]. The full widths at half maximum (FWHMs) of the InGaN “-1st”
diffraction peak of sample B is 159.6 arcsec, which is smaller than that of sample A
(166.5 arcsec), indicating the improved uniformity of indium distribution in MQWs.
The interface roughness can be calculated by fitting FWHMs of the XRD satellite
peaks using the following equation [20]:


n  0  (ln 2)1/2   m  n
0 (1)

where ωn is the FWHM of the nth satellite peak, ω0 represents an intrinsic width of
satellite peaks, Δθm is the angle spacing between the adjacent satellite peaks, and Λ0 is
the period thickness of the InGaN/GaN MQWs. The slope of the fitting line is related
to the interface roughness δ. The FWHM value as a function of satellite peak order is
illustrated in Fig. 1(b). It is obvious that sample B has a smaller absolute value of the
slope, as compared with its counterpart. The improved interface properties of sample
B indicate that the first 0.5-nm-thick LT-cap layer grown in pure nitrogen atmosphere
can effectively protect QW layer from hydrogen over-etching, avoiding the reduction
of QW thickness and interface roughening [18,19].
 ACCEPTED MANUSCRIPT

Fig. 1. (a) HRXRD ω/2θ scanning curves (black line) and simulations (red line) of the
two samples (b) The FWHM value as a function of satellite peak order and its linear
fitting for the samples.

The room temperature PL spectra are shown in Fig. 2. The multiple peak
shoulders originate from the Fabry-Perot interference between sapphire/GaN/air
interfaces. The integrated PL intensity of sample B shows about 2-times
enhancement, as compared with that of sample A. The PL peak energy and FWHM
obtained by Gaussian fitting for samples A and B are 2.697 eV and 158.1 meV, 2.722
eV and 159.6 meV, respectively. The PL peak energy is lower than the band gap
derived from the average indium composition of the two samples, which may relate to
the inhomogeneity of the indium distribution in QWs. For an inhomogeneous indium
distribution structure, the carriers would migrate into the regions with higher indium
concentration, and then recombine radiatively to light emission [21]. In other words,
the PL emission occurs mainly from the indium rich regions [22]. The almost
unchanged peak energy and FWHM further confirm that the indium-rich
luminescence centers located inside the InGaN QW are well protected by the first step
LT-cap layer [15,18].
 ACCEPTED MANUSCRIPT

Fig. 2. The room temperature PL spectra of the two samples.

In order to elucidate the physical origin of the improvement in optical quality,

the temperature-dependent PL characteristics of the two samples, namely, (a)
integrated intensity, and (b) peak energy, are shown in Fig. 3. The normalized
integrated PL intensity shows a monotonous decrease as the temperature increases
from 10 K to 300 K. The relationship between the PL intensity and temperature can
be well fitted by the following Arrhenius formula [20,23]:

I (T )  [1  C1 exp( E A1 / k BT )  C2 exp( E A 2 / k BT )]1 (2)

where I(T) represents the normalized integrated PL intensity, the parameters C1 and
C2 are two constants related with the density of non-radiative recombination centers in
the samples, EA1 and EA2 are the activation energies corresponding to the non-
radiative recombination process, and kB is Boltzmann’s constant. All the fitting results
are listed in Table 1. It can be seen that as hydrogen is introduced into the growth of
the second LT-cap layer, the energy related parameters EA1 and EA2 are barely
changed, indicating the same types of non-radiative recombination centers in MQWs
region [24]. In contrast, the density related parameters C1 and C2 dramatically
decrease from 2.35 to 0.82, and from 112.77 to 38.50, respectively. The reduced C1
and C2 of sample B indicate that adding a small hydrogen flow during the growth of
the second LT-cap layer can significantly reduce the number of non-radiative
recombination centers in MQWs region. Li et al. reported that increasing the thickness
 ACCEPTED MANUSCRIPT

of LT-cap layer can lead to the dramatical decrease of PL intensity [13], which means
that the quality of GaN cap layer grown at low temperature is inferior. It is reported
that impurities (carbon and oxygen) and V-defects, widely regarded as non-radiative
recombination centers, can be reduced by introducing hydrogen into GaN growth [25-
28]. Hence, it is reasonable to deduce that the density of non-radiative recombination
centers is diminished in the second LT-cap layer, which contributes to the alleviation
of PL quenching as the temperature increases. However, the integrated PL intensity of
sample A drops fastly even at low temperature region (10-70 K), which may correlate
with serious indium segregation at the upper well/barrier interface. In this work, the
LT-cap layer is deposited immediately after the growth of InGaN QW. Therefore, the
indium atoms formed as a result of indium segregation may lose chance to desorb or
evaporate from the InGaN surface, forming indium tails and indium-rich regions at
the upper QW/QB interface [13]. The presence of hydrogen during the growth of the
second LT-cap layer can dramatically reduce indium incorporation efficiency,
diminishing indium tails at upper well/barrier interface. Moreover, the indium-rich
regions at upper well/barrier interface formed by indium segregation commonly have
a larger size, as compared to indium rich dot-like luminescence centers located at the
center of QW [15]. Therefore, the indium rich regions formed by indium segregation
may be incompletely covered by GaN after the growth of the first LT-cap layer, as
aforementioned. The indium atoms at these uncovered indium-rich sites may react
with hydrogen to form volatile indium-hydride species at the initial stage of hydrogen
addition [29]. Therefore, the segregated indium atoms near the upper QW/QB
interface are significantly removed after the growth of the second LT-cap layer.
Meanwhile, the first LT-cap layer as well as indium-rich regions formed by indium
segregation acts as a protective layer, which prevents hydrogen from etching InGaN
well layer. As a result, indium-rich regions and indium tails at the upper well/barrier
interface formed by indium segregation are etched away, while indium-rich
luminescence centers located at the center of the QWs are almost unaffected, as
evidenced by only 25 meV blue-shift in PL peak energy at room temperature.
 ACCEPTED MANUSCRIPT

Fig. 3. Temperature-dependent PL characteristics of the two samples (a) integrated

intensity and (b) peak energy. The solid lines are the fitting results.

Table 1. Obtained fitting parameters: activation energies (EA1 and EA2), constants (C1
and C2) and carrier localization energy (σ).

Parameters C1 EA1 (meV) C2 EA2 (meV) σ (meV)

Sample A 2.35 5.4 112.77 53.2 12.6

Sample B 0.82 6.0 38.50 55.3 12.4

The temperature-dependent peak energy of the two samples is shown in Fig.

3(b). Two samples both exhibit S-shaped curves, which is a typical feature of carrier
localization in InGaN/GaN MQWs. The carrier localization degree can be obtained by
fitting the temperature-dependent peak energy curve by Band-tail model [30]:

T 2  2
E (T )  E g (0)  
T   k BT (3)

where E (T) is the peak energy at T, Eg (0) is the energy gap at 0 K, α and β are
Varshni coefficients. The third term comes from the localization effect, in which σ
indicates the degree of localization effect, and kB is the Boltzmann constant. The
values of σ are shown in Table 1. It can be seen that the peak energy of the two
samples shows obvious red-shift in the low temperature region (10 to 50 K), which is
commonly caused by carriers transporting to indium-rich deep localization centers
[31,32]. The red-shift value of peak energy for samples A and B is 8.7 and 2.3 meV,
respectively. The larger red-shift means that sample A has deeper localization centers
 ACCEPTED MANUSCRIPT

[31]. In InGaN based materials, deep localization centers originate from indium-rich
regions, which are mainly composed of self-assembled dot-like regions and indium
segregation regions [15,33,34]. The dot-like regions usually have high luminous
efficiency, and act as the luminescence centers in GaN based devices [33,34]. The
indium-rich regions formed by indium segregation near the QW/QB upper interface
are usually larger in size and abundant with defects [13,15]. As the temperature
increases from 10 to 50 K, carriers can get more energy and transport to the above
mentioned two types of indium-rich regions. Supposing that carriers mainly transport
to the self-assembled dot-like deep localization centers, there would be a slow decay
in integrated PL intensity and red-shift in peak energy. However, if carriers mainly
transport to the indium-rich deep localization centers as a result of indium
segregation, whose larger size would lower the ground state energy and induce larger
red-shift in peak energy. The most important thing is that defects commonly exist in
these large and deep localization centers [15,18]. As a result, a part of the free carriers
are captured by these defects and recombine non-radiatively, which leads to the fast
drop of the integrated PL intensity even at low temperature region. According to the
faster drop of integrated intensity and larger red-shift in peak energy, it is reasonable
to deduce that more serious indium segregation exists at upper QW/QB interface for
sample A.
As temperature increases from 70 to 150 K, thermally activated carriers will
jump out from deep localization centers and migrate to shallow localization centers,
which results in blue-shift in peak energy [31,32]. The blue-shift value of peak energy
for samples A and B is 3.4 and 1.0 meV, respectively. The smaller blue-shift in peak
energy for sample B can be explained by the following two aspects. First, hydrogen
can reduce the large-sized indium-rich regions formed by indium segregation, which
could enhance the quantum confinement of carriers in sample B. Second, indium tails
formed as a result of indium segregation may be etched by hydrogen, which would
partly change the surrounding matrix of the self-assembled dot-like luminescence
centers from low indium regions to GaN. In other words, the density of states of the
shallow localization centers is reduced, and the energy difference between deep
 ACCEPTED MANUSCRIPT

localization centers and surroundings is magnified [35]. Hence, carriers can be better
confined in these deep localization luminescence centers, contributing to the radiative
recombination, the smaller blue-shift and stronger localization effect. However, the
obtained localization energy for the two samples is almost equal, as shown in Table 1.
It is widely reported that thickness or composition fluctuation of the InGaN QW can
enhance carrier localization [19,35,36]. As presented in HRXRD results, sample B
has more uniform distribution of indium composition and smaller interface roughness,
thus the carrier localization effect is weakened [37]. Hence, the combined effect of the
two opposite factors makes the localization energy of sample B almost unchanged. As
the temperature further increases up to 300 K, temperature induced band-gap
shrinkage starts to dominate, which gives rise to the red-shift of the peak energy
[31,32].

Fig. 4. AFM images (5×5 μm) of the MQWs (a) sample A and (b) sample B.
Fig. 4 shows AFM images of the two samples with a scanning area of
5 μm×5 μm. It can be seen MQWs surface presents clear steps and terraces. The small
dark spots observed on the surface mostly correspond to V-pits related threading
dislocations (TDs) [38,39]. As shown in Fig. 4, sample A has a vaguer and narrower
terrace, accompanied by V-pits with higher density and larger size, as compared with
sample B. Shiojiri et al. reported that indium atoms can be easily trapped and
segregated around the core of TDs, to form a small mask that hinders Ga atom
migration [40]. The V-pits can also penetrate and propagate through MQW region,
deteriorating QW/QB interface quality [41]. Moreover, carriers are preferentially
captured at V-defects and recombine non-radiatively, resulting in a low emission
 ACCEPTED MANUSCRIPT

efficiency [42,43]. As hydrogen is applied during the growth of the second LT-cap
layer, indium-rich regions as a result of indium segregation can be significantly
decreased, which is in favor of atom migration and reducing pit density. In addition,
the presence of hydrogen can also enhance the surface diffusion of Ga atoms [44,45].
Therefore, 2-dimensional growth mode is enhanced in sample B, which is beneficial
to obtaining flat surface and sharp well/barrier interface. Consequently, the calculated
root mean square surface roughness (RMS) and pit density decrease from 0.619 nm
and 1.96×108 cm-2, to 0.402 nm and 1.52×108 cm-2, respectively. The smaller pit
density for sample B means more carriers can participate in radiative recombination,
and then contribute to the light emission. In future work, atom probe tomography
(APT) [46-49] or transmission electron microscopy (TEM) [50-52] will be applied to
directly observe the thickness and composition fluctuations, and also the indium
segregation within InGaN QWs.

4. Conclusion
The optical and structural characteristics of InGaN/GaN MQWs with two-step
grown GaN LT-cap layers are investigated. The very small blue-shift in peak energy
and almost unchanged line width verify that LT-cap layer at the first growth step can
effectively protect MQWs from sharp decrease of indium content and interface
roughening caused by hydrogen over-etching. The improvement in PL intensity
indicates that adding a small hydrogen flow during the growth of the second LT-cap
layer can effectively reduce the number of non-radiative combination centers and
alleviate indium segregation at QW/QB upper interface. Consequently, about 2-times
enhancement in PL intensity at a given emission energy is achieved, without
sacrificing interface/surface properties.
Acknowledgments
This work was supported by National Natural Science Foundation of China
(Grant Nos. 61504090 and 21471111), the Applied Basic Research Projects of Shanxi
Province (Grant No. 2016021028), the Scientific and Technological Innovation
Programs of Higher Education Institutions in Shanxi (Grant No. 2017124), the
 ACCEPTED MANUSCRIPT

National Key R&D Program of China (Grant No. 2016YFB0401803), the Shanxi
Provincial Key Innovative Research Team in Science and Technology (Grant No.
201605D131045-10).
References
[1] Y. Sun, K. Zhou, Q. Sun, J. Liu, M. Feng, Z. Li, Y. Zhou, L. Zhang, D. Li, S.
Zhang, M. Ikeda, S. Liu, H. Yang, Nat. Photonics 10 (2016) 595-599.
[2] T. Lu, S. Li, C. Liu, K. Zhang, Y. Xu, J. Tong, L. Wu, H. Wang, X. Yang, Y.
Yin, G. Xiao, Y. Zhou, Appl. Phys. Lett. 100 (2012) 141106.
[3] K.S. Ramaiah, Y.K. Su, S.J. Chang, C.H. Chen, F.S. Juang, H.P. Liu, I.G. Chen,
Appl. Phys. Lett. 85 (2004) 401-403.
[4] C. Du, Z. Ma, J. Zhou, T. Lu, Y. Jiang, P. Zuo, H. Jia, H. Chen, Appl. Phys. Lett.
105 (2014) 071108.
[5] C. Du, Z. Ma, J. Zhou, T. Lu, Y. Jiang, H, Jia, W. Liu, H. Chen, Appl. Phys. Lett.
104 (2014) 151102.
[6] T.H. Ngo, B. Gil, B. Damilano, K. Lekhal, P.D. Mierry, Superlattice Microstruct.
103 (2017) 245-251.
[7] D. Yang, L. Wang, W.B. Lv, Z.B. Hao, Y. Luo, Superlattice Microstruct. 82
(2015) 26-32.
[8] W. Liu, D.G. Zhao, D.S. Jiang, P. Chen, Z.S. Liu, J.J. Zhu, X. Li, F. Liang, J.P.
Liu, S.M. Zhang, H. Yang, Y.T. Zhang, G.T. Du, Superlattice Microstruct. 88
(2015) 50-55.
[9] X.H. Zheng, H. Chen, Z.B. Yan, D.S. Li, H.B. Yu, Q. Huang, J.M. Zhou, J. Appl.
Phys. 96 (2004) 1899-1903.
[10] S. Kim, K. Lee, K. Park, C.S. Kim, J. Cryst. Growth 247 (2003) 62-68.
[11] S.T. Pendlebury, P.J. Parbrook, D.J. Mowbray, D.A. Wood, K.B. Lee, J. Cryst.
Growth 307 (2007) 363-366.
[12] J.W. Ju, H.S. Kim, L.W. Jang, J. H. Baek, D.C. Shin, I.H. Lee, Nanotechnology
18 (2007) 295402.
[13] Z. Li, J. Liu, M. Feng, K. Zhou, S. Zhang, H. Wang, D. Li, L. Zhang, D. Zhao,
D. Jiang, H. Wang, H. Yang, Appl. Phys. Lett. 103 (2013) 152109.
 ACCEPTED MANUSCRIPT

[14] Y.L. Hu, R.M. Farrell, C.J. Neufeld, M. Iza, S.C. Cruz, N. Pfaff, D. Simeonov,
S. Keller, S. Nakamura, S.P. DenBaars, U.K. Mishra, J.S. Speck, Appl. Phys.
Lett. 100 (2012) 161101.
[15] Y.T. Moon, D.J. Kim, K.M. Song, C.J. Choi, S.H. Han, T.Y. Seong, S.J. Park, J.
Appl. Phys. 89 (2001) 6514-6518.
[16] S.M. Ting, J.C. Ramer, D.I. Florescu, V.N. Merai, B.E. Albert, A. Parekh, D.S.
Lee, D. Lu, D.V. Christini, L. Liu, E.A. Armour, J. Appl. Phys. 94 (2003)1461-
1467.
[17] Y.T. Moon, D.J. Kim, K.M. Song, D.W. Kim, M.S. Yi, D.Y. Noh, S.J. Park, J.
Vac. Sci. Technol. B 18 (2000) 2631-2634.
[18] Y. Zhu, T. Lu, X. Zhou, G. Zhao, H. Dong, Z. Jia, X. Liu, B. Xu, Appl. Phys.
Express 10 (2017) 061004.
[19] X. Ren, J.R. Riley, D.D. Koleske, L.J. Lauhon, Appl. Phys. Lett. 107
(2015) 022107.
[20] Y. Zhu, T. Lu, X. Zhou, G. Zhao, H. Dong, Z. Jia, X. Liu, B. Xu, Nanoscale Res.
Lett. 12 (2017) 321.
[21] D. Gerthsen, B. Neubauer, A. Rosenauer, T. Stepahn, H. Kalt, Q. Schön, M.
Heuken, Appl. Phys. Lett. 79 (2001) 2552-2554.
[22] S. Keller, S.F. Chichibu, M.S. Minsky, E. Hu, U.K. Mishra, S.P. DenBaars, J.
Cryst. Growth 195 (1998) 258-264.
[23] Y. Fang, L. Wang, Q. Sun, T. Lu, Z. Deng, Z. Ma, Y. Jiang, H. Jia, W. Wang, J.
Zhou, H Chen, Sci. Rep. 5 (2015) 12718.
[24] Y. Wang, X.J. Pei, Z.G. Xing, L.W. Guo, H.Q. Jia, H. Chen, J.M. Zhou, J. Appl.
Phys. 101 (2007) 033509.
[25] Y. Zhu, T. Lu, X. Zhou, G. Zhao, H. Dong, Z. Jia, X. Liu, B. Xu, Superlattice
Microstruct. 107 (2017) 293-298.
[26] E.L. Piner, M.K. Behbehani, N.A. El-Masry, J.C. Roberts, F.G. McIntosh, S.M.
Bedair, Appl. Phys. Lett. 71 (1997) 2023-2025.
[27] D.I. Florescu, S.M. Ting, J.C. Ramer, D.S. Lee, V.N Merai, A. Parkeh, D. Lu,
 ACCEPTED MANUSCRIPT

E.A. Armour, L. Chernyak, Appl. Phys. Lett. 83 (2003) 33-35.

[28] Z.D. Zhao, B. Wang, Y.P. Sui, W. Xu, X.L. Li, G.H. Yu, J. Electron. Mater. 43
(2014) 786-790.
[29] D.D. Koleske, J.J. Wierer, A.J. Fischer, S.R. Lee, J. Cryst. Growth 390 (2014)
38-45.
[30] P.G. Eliseev, P. Perlin, J. Lee, M. Osiński, Appl. Phys. Lett. 71 (1997) 569-571.
[31] T. Lu, Z. Ma, C. Du, Y. Fang, H. Wu, Y. Jiang, L. Wang, L. Dai, H. Jia, W. Liu,
H. Chen, Sci. Rep. 4 (2014) 6131.
[32] X. Li, D.G. Zhao, J. Yang, D.S. Jiang, Z.S. Liu, P. Chen, J.J. Zhu, W. Liu, X.G.
He, X.J. Li, F. Liang, L.Q. Zhang, J.P. Liu, H. Yang, Y.T. Zhang, G.T. Du,
Superlattice Microstruct. 97 (2016) 186-192.
[33] Y. Jiang, Y. Li, Y. Li, Z. Deng, T. Lu, Z. Ma, P. Zuo, L. Dai, L. Wang, H. Jia,
W. Wang, J. Zhou, W. Liu, H. Chen, Sci. Rep. 5 (2015) 10883.
[34] S.F. Chichibu, A. Uedono, T. Onuma, B.A. Haskell, A. Chakraborty, T.
Koyama, P.T. Fini, S. Keller, S.P. Denbaars, J.S. Speck, U.K. Mishra, S.
Nakamura, S. Yamaguchi, S. Kamiyama, H. Amano, I. Akasaki, J. Han, T. Sota,
Nat. Mater. 5 (2006) 810-816.
[35] R.A. Oliver, F.C.P. Massabuau, M.J. Kappers, W.A. Phillips, E.J. Thrush, C.C.
Tartan, W.E. Blenkhorn, T.J. Badcock, P. Dawson, M.A. Hopkins, D.W.E.
Allsopp, C.J. Humphreys, Appl. Phys. Lett. 103 (2013) 141114.
[36] O. Marquardt, T. Hickel, J. Neugebauer, C.G. Van de Walle, Appl. Phys. Lett.
103 (2013) 073115.
[37] T. Akasaka, H. Gotoh, Y. Kobayashi, H. Nakano, T. Makimoto, Appl. Phys.
Lett. 89 (2006) 101110.
[38] W. Lee, M.H. Kim, D. Zhu, A.N. Noemaun, J.K. Kim, E.F. Schubert, J. Appl.
Phys. 107 (2010) 063102.
[39] L. Lu, Z.Y. Gao, B. Shen, F.J. Xu, S. Huang, Z.L. Miao, Y. Hao, Z.J. Yang,
G.Y. Zhang, X.P. Zhang, J. Xu, D.P. Yu, J. Appl. Phys. 104 (2008) 123525.
[40] M. Shiojiri, C.C. Chuo, J.T. Hsu, J.R. Yang, H. Saijo, J. Appl. Phys. 99 (2006)
073505.
 ACCEPTED MANUSCRIPT

[41] N. Duxbury, U. Bangert, P. Dawson, E.J. Thrush, W. Van der Stricht, K. Jacobs,
I. Moerman, Appl. Phys. Lett. 76 (2000) 1600-1602.
[42] L.C. Le, D.G. Zhao, D.S. Jiang, L. Li, L.L. Wu, P. Chen, Z.S. Liu, Z.C. Li, Y.M.
Fan, J.J. Zhu, H. Wang, S.M. Zhang, H. Yang, Appl. Phys. Lett. 101 (2012)
252110.
[43] L.C. Le, D.G. Zhao, D.S. Jiang, L. Li, L.L. Wu, P. Chen, Z.S. Liu, J. Yang, X.J.
Li, X.G. He, J.J. Zhu, H. Wang, S.M. Zhang, H. Yang, J. Appl. Phys. 114 (2013)
143706.
[44] X. Zhou, T. Lu, Y. Zhu, G. Zhao, H. Dong, Z. Jia, Y. Yang, Y. Chen, B. Xu,
Nanoscale Res. Lett. 12 (2017) 354.
[45] R. Czernecki, S. Kret, P. Kempisty, E. Grzanka, J. Plesiewicz, G. Targowski, S.
Grzanka, M. Bilska, J. Smalc-Koziorowska, S. Krukowski, T. Suski, P. Perlin, M.
Leszczynski, J. Cryst. Growth 402 (2014) 330-336.
[46] M.J. Galtrey, R.A. Oliver, M.J. Kappers, C.J. Humphreys, D.J. Stokes, P.H.
Clifton, A. Cerezo, Appl. Phys. Lett. 90 (2007) 061903.
[47] L. Rigutti, B. Bonef, J. Speck, F. Tang, R.A. Oliver, Scr. Mater. (in press).
[48] G.H. Gu, D.H. Jang, K.B. Nam, C.G. Park, Microsc. Microanal. 19(S5) (2013)
99-104.
[49] L. Rigutti, I. Blum, D. Shinde, D. Hernandez Maldonado, W. Lefebvre, J.
Houard, F. Vurpillot, A. Vella, M. Tchernycheva, C. Durand, J. Eymery, B.
Deconihout, Nano Lett. 14 (2014) 107-114.
[50] L. Hoffmann, H. Bremers, H. Jönen, U. Rossow, M. Schowalter, T. Mehrtens, A.
Rosenauer, A. Hangleiter, Appl. Phys. Lett. 102 (2013) 102110.
[51] T. Mehrtens, M. Schowalter, D. Tytko, P. Choi, D. Raabe, L. Hoffmann, H.
Jönen, U. Rossow, A. Hangleiter, A. Rosenauer, Appl. Phys. Lett. 102 (2013)
132112.
[52] S. Kret, F. Ivaldi, K. Sobczak, R. Czernecki, M. Leszczynski, Phys. Status Solidi
A 207 (2010) 1101-1104.
 ACCEPTED MANUSCRIPT

Two-step grown low temperature cap layers are employed to improve the quality of active
region.

The first LT-cap layer grown in nitrogen atmosphere acts as a protective layer, which protects
quantum well from hydrogen over-etching.

The hydrogen gas employed in the second LT-cap layer is in favor of reducing defect density
and indium segregation.

Surface and interface properties are improved when the two-step growth method is adopted.

Room temperature photoluminescence intensity enhances by about 2-times, while peak

energy and line width are almost unchanged by using the two-step growth method.