Effect of the Cladding Layer Cavity on the Efficiency of 650 nm Resonant Cavity Light Emitting Diodes

doi:10.4236/opj.2013.32B067

Paper Menu >>

Journal Menu >>

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

Email: #[email protected]

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

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.

of (Al

1-x

)

0.5

0 0.5

1

Sample A

0 0.5

0.7

1/4

1/2

Sample B

x of (Al

1-x

)

0.5

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

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],



extra



(2)

and



 

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