Optics and Photonics Journal, 2013, 3, 284-287
doi:10.4236/opj.2013.32B067 Published Online June 2013 (http://www.scirp.org/journal/opj)
Effect of the Cladding Layer Cavity on the Efficiency of
650 nm Resonant Cavity Light Emitting Diodes*
Jianjun Li#, Tao Liu, Jiachun Li, Xuan Ya
Key Laboratory of Opto-electronics Technology, Beijing University of Technology, Beijing, China
Received 2013
ABSTRACT
High efficiency 650 nm resonant cavity light emitting diodes (RCLEDs) with a cladding layer cavity are reported. The
epitaxial structure is grown with a metal-organic chemical vapor deposition (MOCVD) system. Al 0.5Ga 0.5 As/Al As
is used for the distributed Bragg reflectors (DBRs), and GaInP/AlGaInP multiple-quantum wells for the active region.
Two RCLED samples have been fabricated, one with a cladding layer cavity and the other without. Experimental results
show that the cladding layer cavity can improve the internal quantum efficiency effectively, so that an external quantum
efficiency of 7.4% at 20 mA is reached. Meanwhile, the sample with cladding layer cavity also shows a spectral stabil-
ity as the injected current changing from 20 mA to 100 mA.
Keywords: 650 nm; RCLED; MOCVD; DBR
1. Introduction
Because of the substrate absorption and the phenomenon
of total internal reflection, AlGaInP LEDs grown on
GaAs substrate suffer from a lower extraction efficiency
of 2% [1]. The resonant cavity light-emitting diodes
(RCLEDs) provide an effective way to improve the ex-
ternal efficiency by enhancing spontaneous emission and
directing much light normal to the semiconductor surface
[2]. Meanwhile, RCLEDs also offer many advantages
over conventional LED’s, such as the high temperature
stability [3-5], the improved spectral purity [6] and the
narrow far field beam [7,8]. However, experimental re-
sults of the external efficiency for RCLEDs around 650
nm wavelength are still not satisfied [9-12]. By now,
most efforts such as special mirrors [13], photonic crystal
[14] and 2-D grating [15] are made to improve RCLEDs
quantum extraction efficiency, while less attention is
paid to the internal quantum efficiency which is very
important for RCLEDs with a narrow active region.
In this paper, we have designed and fabricated two
RCLED samples with different cavity structure at 650
nm wavelength. For the sample with a cladding layer
cavity, the internal quantum efficiency nearly reaches
100%, so that an external efficiency nearly reaching the
theory limits is obtained.
2. Result and Discussion
Figure 1 shows the RCLED structure diagram. The basic
structure consists of a 34-pair Al0.5Ga0.5As/AlAs bot-
tom N-DBR, a 1 -cavity containing 3 quantum wells,
and a 6-pair Al0.5Ga0.5As/AlAs top P-DBR. In order to
form ohmic contact, there is a 5-nm-thick p+-GaAs on the
top of P-DBR. Two samples with different cavity struc-
ture are designed. For sample A, the cavity was formed
by three 5nm-Ga0.5In0.5P/5nm -(Al0.5Ga0.5)0.5In0.5P
QWs centered between (Al0.5Ga0.5)0.5In0.5P barrier
layers, the total optical thickness of the cavity is 1. For
sample B, the cavity is the same as sample A except that
the 1/4-(Al0.5Ga0.5)0.5In0.5P-thick barrier layers near
both top and bottom DBR are replaced by the 1/4-
(Al0.7Ga0.3)0.5In0.5P-thick cladding layers.
Three considerations have been made during the cavity
design of sample B. Firstly, the thickness of each clad-
ding layer is 1/4 optical wavelength, so that the standing
wave node within the cavity coincides with the
(Al0.5Ga0.5)0.5In0.5P/ (Al0.7Ga0.3)0.5In0.5P interface
to lower the optical loss. Secondly, because of the
nar-row active region and the limited conduction band
offset between GaInP and (Al0.5Ga0.5)0.5In0.5P mate-
rials, the carrier concentration within (Al0.5Ga0.5) 0.5
In0.5P bar-rier layers is not negligible. The introducing
of (Al0.7Ga0.3)0.5In0.5P cladding layer will raise the
car-rier concentration within (Al0.5Ga0.5)0.5In0.5P and
QW layers by confining carriers in a narrower region,
which is benefit for increasing the radiative recombine-
tion rate. Thirdly, the exchange of As and P at the Al-
*Supported by the Beijing Education Committee Science and Technolog
y
Plan Surface Projects under Grant No. KM200810005002.
#Corresponding author.

J. J. LI ET AL. 285
GaInP/AlAs interface will introduce many interface
states which serve as the non radiative recombination
centers. By using the cladding layer cavity to confine the
carriers, the minority carrier concentration at the Al-
GaInP/AlAs interface is lowered effectively, so that the
non radiative recombination which needs two types of
carriers is decreased. Shown in Figure 2 is the refractive
index and the calculated longitudinal optical field inten-
sity by the transfer matrix method [16] within sample B,
also shown in the inset is the enlarged profile near the
active region. The coupling between the active QW re-
gion and the cav-ity mode, and the optical field node at
the (Al0.5Ga0.5)0.5In0.5P/ (Al0.7Ga0.3)0.5In0.5P in-
terface can be seen clearly.
The entire RCLED structure is grown by an EMCORE
D125 MOCVD system on 15o off (100) oriented n+-
GaAs substrate. The metal-alkyl sources used are trime-
thylindium (TMIn), trimethylgallium (TMGa) and trime-
thylaluminium (TMAl), pure AsH3 and PH3 are used as
the hydride precursor gases, SiH4 and CCl4 are used as
the n- and p-type doping sources, respectively, the carrier
gas is H2 purified by Pd cell. V/III ratio for AlGaInP and
AlGaAs is 230 and 150, respectively. The samples are
grown at a temperature of 700oC, and the wafer carrier is
rotated at a speed of 1000 rpm through-out the process.
After growth, a Ti-Au layer is deposited on the wafer
top, then standard photolithograph and etching are em-
ployed to form the net pelectrode, the diameter of the
bonding contact pad is 100 μm. The n-electrode is
formed at a temperature of 400℃ by the evaporation of
AuGeNi- Au onto the backside of the substrate, which is
thinned to 100 μm. Finally, the wafer is diced into 300 ×
300 μm2 chips without encapsulating to evaluate the
photoelectric performance.
Figure 3 shows the measured output light power and
forward voltage versus DC current for both samples. The
highly doped DBRs and the high-quality net shape top
ohmic contacts result in a low voltage of 1.87 V at 20
mA for sample A, and 1.91 V for sample B. An output
power of 2.84 mW is achieved for sample B at 20 mA,
corresponding to an external quantum efficiency of 7.4%,
while the output power is only 0.85 mW for sample A at
the same current.
x
of (Al
x
Ga
1-x
)
0.5
In
0.5
P
0 0.5
1
Sample A
0 0.5
0.7
1/4
1/4
1/2
Sample B
x of (Al
x
Ga
1-x
)
0.5
In
0.5
P
GaAs Sub
N-DBR
P-DBR
Figure 1. RCLED structure diagram.
01000 2000 3000 4000 5000
0.0
0.8
1.6
2.4
3.2
4.0
0.0
0.2
0.4
0.6
0.8
1.0
500 600 700 800 90010001100
0. 0
0. 2
0. 4
0. 6
0. 8
1. 0
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
Refractive Index
position(nm)
Field Intensity(a.u.)
Posi t i on( nm)
Field Intensity(a.u.)
Refractive Index
1
Figure 2. Refractive index and calculated longitudi nal optical
field intensity profile for RCLED with a cladding layer
cavity.
10 30 50 70 90110
0.0
0.5
1.0
1.5
2.0
2.5
0
2
4
6
8
10
Foward Voltage(V)
Sample A
Sample B
DC current(mA)
Light Power(mW)
Figure 3. Light power and forward voltage versus dc current
for sample A and sample B.
The LED’s external quantum efficiency ηext is given
by,
int

ext extra

 (1)
where 
int is the internal quantum efficiency and

extra is
the extraction efficiency. For RCLEDs with DBRs as top
and bottom mirrors, the extraction efficiency can be ex-
pressed as [17],
1

extra
c
m
 (2)
and
1
0
1
1
2

 
h
c
h
nn
mm nn
(3)
where mc is the effective cavity order including penetra-
tion length into DBRs, m0 is the bare cavity order, nh and
nl is the high and low refractive index in DBRs, respec-
tively. According to the optical design, both sample A
and sample B should have the same extraction efficiency

J. J. LI ET AL.
286
of 8.5% by substituting m0 = 2 and refractive index value
from [18] into Equation. (2) and (3). So the large differ-
ence of the external quantum efficiency between sample
A and sample B is due to the different internal quantum
efficiency

int. That is to say, the cavity structure of sam-
ple B with cladding layer benefits a higher

int by con-
fining carriers effectively and reducing the AlGaInP/
AlAs interface recombination simultaneo-ussly. If we
have considered the opaque top electrode, which amounts
to 10% of the total top emitting surface, a

int of nearly
100% can be obtained for sample B.
For both samples, the optical power increases first
with increasing current before it rolls-over and decreas-
ing again due to the thermal effect. For sample B, the
light power reaches its maximum point, 7.72 mW at a
current of 90 mA. As author’s knowledge, that is the
highest output power for 650 nm RCLEDs with the same
structure. But for sample A, the maximum light power is
only 1.77 mW, meanwhile, because of the sample A’s
lower photoelectric efficiency and severer thermal effect,
it reaches its maximum power at a lower current of 70
mA. So we can say, because of the lower conduction
band offset for AlGaInP materials and the thinner active
region defined by the cavity thickness, a deliberate de-
sign of the cavity is very important for 650 nm RCLEDs.
Shown in Figure 4 is the electroluminescence (EL)
620 630 640650 660 670 680
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Sample A
Light Intensity( x10-4 W/nm)
Wavelength(nm)
current increased
from 20mA to 100mA
620 630 640 650 660 670 680
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Light Intensity( x10-4 W/nm)
Wavelength(nm)
Sample B
current increased
from 20mA to 100mA
Figure 4. Emission spectrum of sample A (a) and sample B
(b) at varying current between 20 mA and 100 mA with 10
mA increments.
spectrum of both samples at room temperature with DC
current varying between 20 mA and 100 mA (with 10
mA increments). Because of the high internal quantum
efficiency of sample B, the peak wavelength of sample B
is more stable than that of sample A. As DC current in-
creases from 20 mA to 100 mA, the peak wavelength of
sample A changes 4.4 nm from 647.2 nm to 651.6 nm,
while that of sample B is only 1.2 nm from 647.4 nm to
648.6 nm. Both samples show a same variability of the
spectral purity with the current increment, that is the
typical characteristics of RCLEDs [2]. When the injected
current increases from 20 mA to 100 mA, the full width
at half maximum (FWHM) of the spectrum increases
from 13 nm to 14.5 nm for sample B, and for sample A
that is from 12.5 nm to 13.8 nm.
3. Summary
In summary, both 650 nm RCLED samples with different
cavity structures have been fabricated. Experimental re-
sults show that for the sample with cladding layer cavity,
an external quantum efficiency as high as 7.4% has been
reached at 20 mA due to the high internal quantum effi-
ciency, meanwhile, the sample also has a high stability of
the peak wavelength and the spectral purity with the cur-
rent increment. The cavity design, especially for 650 nm
RCLEDs with AlGaInP and AlGaAs materials, is very
important to get high external quantum efficiency.
REFERENCES
[1] C. P. Kuo, R. M. Fletcher, T. D. Osentowski, M. C.
Lardizabal and M. G. Craford, “High Performance Al-
GalnP Visible Light-Emitting Diodes,” Applied Physics
Letters. Vol. 57, No. 27, 1990, 2937.
doi:10.1063/1.103736
[2] E. F. Schubert, Y. H. Wang, A. Y. Cho, L. W. Tu and G.
J. Zydzik, “Resonant Cavity Light-Emitting Diode,” Ap-
plied Physics Letters, 1992, Vol. 60, No. 8, 106489.
doi:10.1063/1.106489
[3] Y. A. Chang, C. L. Yu, I. T. Wu, H. C. Kuo, T. C. Lu, F.
I. Lai, L. W. Laih, L. H. Laih and S. C. Wang, “Design
and Fabrication of Temperature-Insensitive InGaP-InGaA
IP Resonant-Cavity Light-Emitting Diodes,” 2006 IEEE
Photonics Technology Letters, Vol. 18, No. 16, 1690.
doi:10.1109/LPT.2006.879931
[4] J. R. Chen, T. S. Ko, T. C. Lu, Y. A. Chang, H. C. Kuo,
Y. K. Kuo, J. Y. Tsai, L. W. Laih and S. C. Wang, “Fab-
rication and Characterization of Temperature Insensitive
660 nm Resonant-Cavity LEDs,” IEEE Journal of Light-
Wave Technology, Vol. 26, No. 13, 2008, 1891.
doi:10.1109/JLT.2008.920639
[5] H. Fardi, M. Jan, B. Van Zeghbroeck, 2012, Photovoltaic
Specialists Conference. 16 785
[6] E. F. Schubert, N. E. J. Hunt, R. J. Malik, M. Micovic and
D. L. Miller, “Temperature and Modulation Characteris-

J. J. LI ET AL.
287
tics of Resonant-Cavity Light-Emitting Diodes,” IEEE
Journal of Lightwave Technology, Vol. 14, No. 7, 1721.
doi.org/10.1109/50.507950
[7] R. Joray, M. Ilegems, R. P. Stanley, W. Schmid, R. Bu-
tendeich, R. Wirth, A. Jaeger and K. Streubel, “Far-Fields
Radiation Pattern of Red Emitting Thin-Film Resonant
Cavity LEDs,” IEEE Photonics Technology Letters, Vol.
18, No. 9, 2006, 1052. doi.org/10.1109/LPT.2006.873553
[8] S. Morikura and Y. Okamura. “Angle-Dependent Spectral
Width of Resonant-Cavity Light-Emitting Diode,” IEEE
Photonics Technology Letters, Vol. 21, No. 10, 2009, 21
660. doi.org/10.1109/LPT.2009.2016218
[9] S. W. Chiou, Y. C. Lee, Y. C. Yang, C. S. Chang and T.
P. Chen, “High-Performance Resonant Cavity
Light-Emitting Diode for Plastic Optical Fiber Applica-
tion,” Proceeding of SPIE, Vol. 5366, 2004.
doi.org/10.1117/12.525285
[10] M. M. Dumitrescu, M. J. Saarinen, M. D. Guina and M.
V. Pessa, “High-Speed Resonant Cavity Light-Emitting
Diodes at 650 nm,” IEEE Journal on Selected Topics in
Quantum Electronics, Vol. 8, No. 2, 2002, pp. 219-230.
doi:10.1109/2944.999174
[11] J. J. Li, Z. Yang, J. Han, J. Deng, D. S. Zou, Y. Z. Kang,
L. Ding and G. D. Shen. Acta Physica Sinica. 2009, 58
6304.
[12] R. Wirth, C. Karnutsch, S. Kugler, S. Thaler and K.
Streubel. “Red and Orange Resonant-Cavity LEDs,”
Proceedings of SPIE, Vol. 4278, 2001.
doi:10.1117/12.426855
[13] P. H. Lei and C. D. Yang. Solid- State Electronics, 2008,
52 227.
[14] K. Berggenek, C. Wiesmann, H. Zull, R. Wirth, P.
Sundgren, N. Linder, K. Streubel, and T. F. Krauss. “Di-
rectional Light Extraction from Thin-Film Resonant Cav-
ity Light-Emitting Diodes With a Photonic Crystal,” Ap-
plied Physics Letters, Vol. 93, No. 23, 2008, 231109.
doi:10.1063/1.3046130
[15] J. Rao, R. Winfield and S.O’ Brien. IEEE Photonics
Technology Letters. 2009, “Enhancement of Light Ex-
traction From Resonant Cavity Light-Emitting Diodes
Using a 2-D Grating Embossed in TiO2 Sol-Gel,” Vol. 21,
No. 13, 941.
doi:10.1109/LPT.2009.2020808
[16] J. J. Li and G.D. Shen. Journal of Optoelectronics. Laser.
2006, 17 1457.
[17] H. Benisty, H. D. Neve and C. Weisbuch. “Imoact of
Planar Microcavity Effects on Light Extraction-Part
Ⅰ:Basic concepts and Analytical Trends,” IEEE Journal
Quantum Electron, Vol. 34, No. 9,1998, pp. 1612-1631.
doi.org/10.1109/3.709578
[18] S. Gehrsitz, F. K. Reinhart, C. Gourgon, N. Herres, A.
Vonlanthen and H. Sigg, “The Refractive Index of
AlxGa1-xAs Below the Band Gap: Accurate Determina-
tion and Empircal Modeling,” Journal of Applied Physics,
2000, Vol. 87. 7825.doi:10.1063/1.373462