We demonstrate a III-nitride nonpolar vertical-cavity surface-emitting laser (VCSEL) with a photoelectrochemically (PEC) etched aperture. The PEC lateral undercut etch is used to selectively remove the multi-quantum well (MQW) region outside the aperture area, defined by an opaque metal mask. This PEC aperture (PECA) creates an air-gap in the passive area of the device, allowing one to achieve efficient electrical confinement within the aperture, while simultaneously achieving a large index contrast between core of the device (the MQW within the aperture) and the lateral cladding of the device (the air-gap formed by the PEC etch), leading to strong lateral confinement. Scanning electron microscopy and focused ion-beam analysis is used to investigate the precision of the PEC etch technique in defining the aperture. The fabricated single mode PECA VCSEL shows a threshold current density of ∼22 kA/cm2 (25 mA), with a peak output power of ∼180 μW, at an emission wavelength of 417 nm. The near-field emission profile shows a clearly defined single linearly polarized (LP) mode profile (LP12,1), which is in contrast to the filamentary lasing that is often observed in III-nitride VCSELs. 2D mode profile simulations, carried out using COMSOL, give insight into the different mode profiles that one would expect to be displayed in such a device. The experimentally observed single mode operation is proposed to be predominantly a result of poor current spreading in the device. This non-uniform current spreading results in a higher injected current at the periphery of the aperture, which favors LP modes with high intensities near the edge of the aperture.

Vertical-cavity surface-emitting lasers (VCSELs) have a number of unique properties relevant for many applications, including a low threshold current, high power density, potential for single mode emission, low output beam divergence, and emission normal to the substrate, allowing one to fabricate high-density 2D arrays. VCSELs have been developed for a number of material systems, including GaAs-, InP-, and GaN-based (III-nitride) systems.1 In each system, the unique material properties necessitates different structural designs and distributed Bragg reflector (DBR) designs. One of the most notable differences between the systems is the way in which the aperture is defined. In GaAs-based VCSELs, the native-oxide aperture is the established technology most commonly used. This aperture is formed by hydrolyzing the sidewalls of AlGaAs or AlAs layers in a steam atmosphere furnace at ∼400–500 °C, to yield lateral oxidation in the form of AlxOy.1–3 In InP-based VCSELs, lateral oxidation is not easily achieved; thus, the aperture is often formed using a buried tunnel junction or a selective undercut etch close to the active region to form an air-gap aperture.1,4–6 This air-gap aperture is fabricated by selectively etching InAlAs or AlGaInAs in a solution of citric acid and hydrogen peroxide.5,6 For GaN-based VCSELs, a number of methods have been used to define the aperture, including a simple dielectric layer,7–16 a plasma passivated p-GaN layer,17 and an ion implanted layer.18 The dielectric aperture has been used with a large variation in the degree of success. Theoretical analysis suggests that using the standard dielectric aperture design can lead to significant amounts of diffraction loss,19,20 which correlates well with some reported experimental results.13,14,18 The ion implanted aperture has been shown to result in a reduction in threshold current density (Jth), compared with a standard SiNx aperture; however, the index contrast between the implanted region and the inner aperture is very small, restricting the ion implanted aperture to fairly large diameter devices.18 Developing GaN-based VCSELs with aperture designs similar to those used in GaAs-based and InP-based VCSELs would be beneficial; however, lateral oxidation and selective undercut wet etching are not easily achieved in the III-nitrides.21 III-nitrides can be selectively etched using photoelectrochemical (PEC) etching, which has been utilized for a number of devices and applications. Here, we report on a nonpolar GaN-based VCSEL with a PEC etched aperture (PECA) (i.e., an air-gap aperture). This flip-chip dual dielectric DBR VCSEL also uses PEC etching for substrate removal, as has been reported previously.13,14,18,22

The PECA method has been demonstrated in a number of devices in the literature. The initial demonstrations formed a PECA on optically pumped microdisk lasers.23–25 A PECA was also used to confine current in a III-nitride current aperture vertical electron transistor.26,27 These studies, and the majority of work investigating PEC etching, have been carried out on epitaxialy layers grown on c-plane GaN,28–32 which has significantly different etching behavior than m-plane GaN epitaxial layers.14,33,34 More recently, a III-nitride edge-emitting laser has also been fabricated using the PECA technique.35,36 Also, the basic mechanism of the PEC undercut etch has been used to form air-gap DBRs.37–39 

The device demonstrated here was processed in parallel with those described in Ref. 22 and used the same epitaxial structure, with an ITO intracavity contact.40 The metal-organic chemical vapor deposited epitaxial structure of the fully processed device consists of a 6.95 λ cavity with a 7× multi-quantum well (MQW), composed of 3 nm InGaN active QWs (A) (λ 405 nm) and 1 nm GaN barriers (B), all grown on intentionally miscut m-plane GaN substrates from Mitsubishi Chemical Corporation.41–43 Figure 1 shows a schematic of the device immediately after the PEC aperture (PECA) is defined and after the fabrication process is complete. The fabrication procedure was similar to that described in Refs. 13, 14, and 18, with modifications for incorporating the PECA (details are given below). Following epitaxial growth and p-GaN activation, a dry etch was performed to define a mesa with an etch depth below the active MQW and above the sacrificial MQW (Fig. 1(a)). Next, a Ti/Au mask was patterned, defining the PECA pattern. This Ti/Au layer also served as a protective layer for the structural support ring surround the core of the device, while simultaneously acting as the PEC cathode in the areas in contact with the n-GaN. The support ring, seen in Fig. 1, is necessary because one must make the mesa area large enough for probing the n-contact (Fig. 1(b)); however, it is also important to reduce the lateral etch distance for the PECA, in order to minimize the structural weakness created when the PECA air-gap is formed. It is of note that sonicating these devices causes catastrophic damage to the majority of devices on a chip. Following the Ti/Au deposition, the sample was submerged in 0.1M KOH and illuminated with a 405 nm LED array (FWHM = 16 nm, ∼12 W output power (∼65 mW/cm2)) for 30 min, yielding the PECA via undercut etching of the active MQW not protected by the opaque Ti/Au mask. M-plane MQWs are known to have anisotropic undercut etch rates, with the Ga-face etching faster than the N-face;14,33,34 thus, the etch on one side of the aperture reaches the Ti/Au mask before the other side. This implies that the total etch time is defined by the etch rate of the N-face. Once the etch reaches the edge of the opaque Ti/Au hardmask defining the aperture, photogenerated holes are no longer available to continue the PEC etch process, preventing further etching below the Ti/Au hardmask. It is important to note that using low KOH concentrations is necessary to reduce the degree of purely chemical etching, which causes roughening. After the PECA was defined, the Ti/Au mask was removed in aqua regia, and the remainder of the device was processed using methods described previously.13,14,18,22 A schematic cross-section of the completed device is shown in Fig. 1(b), where the n-side DBR (n-DBR) is composed of a 12 period (P) stack of ¼-wave SiO2/Ta2O5 layers, while the p-DBR is composed of a 16 P SiO2/Ta2O5 stack. The VCSEL has been flip-chip bonded to a Cu block using a Au-Au compression bond at 200 °C for 2 h. The substrate removal for the flip-chip bond, which is discussed elsewhere,14,18,22 also used PEC etching to undercut the sacrificial MQW seen in Fig. 1(a).

FIG. 1.

(a) Schematic of a partially processed VCSEL, immediately after the PEC aperture (PECA) is defined. (b) Schematic of a completed PECA VCSEL.

FIG. 1.

(a) Schematic of a partially processed VCSEL, immediately after the PEC aperture (PECA) is defined. (b) Schematic of a completed PECA VCSEL.

Close modal

Following fabrication, a focused-ion beam cross-sectional analysis was carried out on one of the VCSELs on the chip. Figure 2 shows scanning electron microscope (SEM) images taken after forming the cross-section. The PECA is clearly visible, with an air-gap thickness of ∼30 nm. This thickness is approximately equal to the active MQW total thickness, demonstrating the precise nature of this undercut etch technique. Observing the region where the lateral etch stops, we see a slight slope at the edge of the aperture (∼26°). This tapering suggests the Ti/Au mask does not yield a perfect etch selectivity between the areas illuminated during the PEC etch and the areas covered with the opaque Ti/Au mask. This is likely a result of scattered light at the Ti/Au mask edge. In general, tapered apertures can be beneficial, as demonstrated on GaAs-based VCSELs with tapered oxide apertures;44 however, the specific effects of such tapering depend on the position of the aperture relative to the longitudinal mode peaks and nulls in the cavity, the aperture diameter, and the slope of the tapering. More analysis is necessary to determine the degree to which the tapering observed here affects the VCSEL performance.

FIG. 2.

SEM micrograph of a PECA VCSEL cross-section made using a focused ion beam (FIB). (a) shows a zoomed-out view, giving perspective on the position of the DBRs, cavity, and air-gap PEC aperture. (b) shows a zoomed-in view of the PEC aperture, where the air-gap is seen to be ∼30 nm thick (roughly the thickness of the MQW) and the aperture appears to end in an angled etch.

FIG. 2.

SEM micrograph of a PECA VCSEL cross-section made using a focused ion beam (FIB). (a) shows a zoomed-out view, giving perspective on the position of the DBRs, cavity, and air-gap PEC aperture. (b) shows a zoomed-in view of the PEC aperture, where the air-gap is seen to be ∼30 nm thick (roughly the thickness of the MQW) and the aperture appears to end in an angled etch.

Close modal

Observing the lasing performance of a PECA VCSEL with a 12 μm aperture diameter (Fig. 3), we see a threshold current of ∼25 mA (∼22.1 kA/cm2), with a peak output power of 180 μW at ∼100 kA/cm2, and a top-side differential efficiency, ηd,top, of ∼0.07%. The differential resistance is ∼42.82 Ω. The device was measured under pulsed operation (0.3% duty cycle and 100 ns pulse width). It is important to note that we measure the current density in terms of the electrically injected area, instead of the observed lasing area, as III-nitride VCSELs often have significantly different near-field emission profiles (discussed in detail later). Figure 3(b) shows the emission spectrum vs. current density, where the lasing wavelength, λ, is observed to be ∼417 nm, with a spectrometer resolution limited FWHM of ∼2 nm. Assuming a group index, ng, of ∼3.3,45 the mode spacing, dλ, is calculated to be ∼21 nm (dλ=λ2/(2ngLeff), where Leff is the effective cavity length).46 Thus, this is a single longitudinal mode VCSEL. Observing the log-intensity plot in the inset, we see the spontaneous emission peak at ∼405 nm. This misalignment between the approximate peak gain wavelength and cavity resonance wavelength (i.e., the gain offset) was caused by an unintentional incorporation of a Ta2O5 layer at the start of the n-DBR deposition. Such gain offsets have been used advantageously in InP- and GaAs-based VCSELs; however, a comprehensive study on the proper gain offset for enhancing the performance of violet GaN-based VCSEL with a PECA design has not been reported. It is of note that simulations of violet c-plane VCSELs, with SiNx apertures and hybrid DBR designs, have been reported,47 which show that the optimal gain offset is dependent on the aperture diameter, lasing linearly polarized (LP) mode, as well as cavity length.

FIG. 3.

(a) Pulsed LIV characteristics of a PECA VCSEL with 12 μm aperture diameter, an ITO intracavity contact, and a 6.95λ cavity. The threshold current is ∼25 mA (22 kA/cm2) (b) Emission sprectrum vs. current density.

FIG. 3.

(a) Pulsed LIV characteristics of a PECA VCSEL with 12 μm aperture diameter, an ITO intracavity contact, and a 6.95λ cavity. The threshold current is ∼25 mA (22 kA/cm2) (b) Emission sprectrum vs. current density.

Close modal

Comparing this PECA VCSEL with the Al ion implanted aperture (IIA) VCSEL, also employing an ITO intracavity contact, reported in Ref. 22, we see that the IIA VCSEL had a lower Jth (8 kA/cm2), however, only reached a peak output power of ∼80 μW at ∼60 kA/cm2 (ηd,top 0.06%). It is possible that the difference in Jth, but similarity in ηd,top, is a result of a difference in the transparency current density, which would not be unexpected as the two devices have different lasing wavelengths (∼410 vs. ∼417 nm). However, more work is necessary to develop a complete understanding of the differences and similarities in device performance of IIA and PECA VCSELs, as both devices show significantly different near-field mode profiles, described in detail next.

Prior to this report, all VCSELs fabricated by our group have shown filamentary lasing in the aperture.13,14,18,22 This includes VCSELs with SiNx apertures13,14 and more recent VCSELs with IIA designs.18,22 Filamentation has also been observed by other groups researching III-nitride VCSELs.7,10,11,15,17,48–50 In some publications, the degree of filamentation is difficult to determine due to the imaging camera being over saturated when the image was taken.7–9 It is of note that the dielectric aperture VCSEL reported by Izumi et al. appears to have a well-defined mode profile.16 In comparison to the filamentary lasing observed in our dielectric aperture and IIA designs, the PECA VCSEL near-field emission profile, shown in Fig. 4, shows a clearly defined single LP mode, meaning that the PECA is a single longitudinal and lateral mode device. This suggests that a method for eliminating filamentation is to use a design with a high core-cladding refractive index contrast. Observing Fig. 4, we see no evidence of higher order modes turning on as the current is increased.

FIG. 4.

near-field emission intensity imaged using optical microscopy at various current densities. The emission profile shows 24 lobes located near the edge of the aperture. This mode profile corresponds to the LP12,1 mode.

FIG. 4.

near-field emission intensity imaged using optical microscopy at various current densities. The emission profile shows 24 lobes located near the edge of the aperture. This mode profile corresponds to the LP12,1 mode.

Close modal

To understand the origin of such a mode profile, we first calculated the effective mode index in the core (within the aperture) and cladding (outside the aperture) in the longitudinal direction, using the 1D transmission matrix method discussed in Refs. 18, 22, and 40. This yields a core-cladding effective index step, Δneff, of 0.049 (2.344–2.295). For the case of an IIA, Δneff is predicted to be ∼0.001.18,22 In InP-based air-gap aperture VCSELs, Δneff is ∼0.4, which is similar to the Δneff in GaAs-based VCSELs with oxide apertures.4,6 In general, increasing Δneff not only improves modal confinement but can also lead to increased scattering loss caused by the aperture.4 Furthermore, a larger core-cladding index contrast allows higher order modes to be more easily supported in a fiber or VCSEL, due to the increase in the normalized frequency, V.1,51 It is also possible that current induced index variations may play an important role in determining lasing performance, particular for the case of 7λ cavities, where the current spreading may be non-ideal.52,53 More sophisticated analytical simulations are necessary to determine the degree to which scattering loss, core-cladding index contrast, current induced index fluctuation, and the gain offset are contributing to the differences in lasing performance for weakly and strongly index guided VCSELs.

Following the determination of the core-cladding index contrast, we carried out a 2D axis-symmetric simulation in COMSOL's “Electromagnetic Waves, Frequency Domain” physics module to calculate the LP mode profiles. The boundaries of the core-cladding simulation were assumed to be perfect electrical conductors. Figure 5 shows all LP modes of a lower order than that observed in the PECA VCSEL. The modes profiles are organized according to their radial modal index, l, and azimuthal modal index, m (LPl,m). Comparing Fig. 4 with Fig. 5, we see that the PECA VCSEL shows LP12,1 lasing. Because this is a high-order mode, one would expect to see the low-order modes lasing as well. If we consider the fact that the n-GaN and ITO layers are fairly thin (∼760 nm and ∼47 nm, respectively), we can realize that this device would be expected to have a significant current spreading resistance on the n-side and p-side (ITO). Assuming an n-GaN mobility of 200 cm2/V s and a carrier concentration of 2.5 × 1018 cm−3, we estimate the resistivity to be 1.25 × 10−2 Ω cm, giving a sheet resistance of ∼160 Ω. For ITO, the resistivity is ∼5 × 10−4 Ω cm,18,40 giving a sheet resistance of ∼100 Ω. The large spreading resistance on the p- and n-side of the device suggests that the edge of the aperture may receive significantly more injected current than the center of the aperture. Because the high order LP12,1 mode has its peak intensity near aperture edge, while lower order modes have peak intensities nearer to the center of the aperture (Fig. 5), one would expect the high-order modes to reach threshold before the low-order modes, due to the non-uniform current spreading. This analysis is in agreement with the simulation reports from Ref. 47, where non-uniform current spreading in large aperture VCSELs results in higher order LP modes being favored. The researchers also show that longer cavity lengths generally result in a decrease in the order of the primary lasing LP mode and that the aperture diameter, as well as gain offset, can heavily influence mode selection.47 It is possible that modes higher than the LP12,1 mode are not observed because of a higher degree of scattering loss.4 Thus, it seems likely that the single lateral mode emission results from a balance of non-uniform current spreading, causing lower order modes to be suppressed, and aperture induced scattering loss, which suppresses higher order modes. A similar simulation was also performed to model the IIA VCSEL. Here, all supported modes showed an effective mode index with an imaginary component, implying that they are not completely confined to the core region. Also, only the LP0,1, LP1,1 and LP1,2, and LP0,2 modes were supported in the core of the device, highlighting the significant difference in confinement between the IIA and PECA design. Overall, these results suggest that the filamentary lasing observed in the IIA design is a result of gain guiding dominating the mode profile behavior, rather than index guiding, which dominates in the PECA case.

FIG. 5.

Simulated linearly polarized (LPl,m) mode profiles as a function of the radial modal index, m, and azimuthal modal index, l. The modes shown correspond to those with normalized propogation constants greater than the experimentall observed mode, LP12,1. The simulations were carried out using COMSOL's “Electromagnetic Waves, Frequency Domain” module.

FIG. 5.

Simulated linearly polarized (LPl,m) mode profiles as a function of the radial modal index, m, and azimuthal modal index, l. The modes shown correspond to those with normalized propogation constants greater than the experimentall observed mode, LP12,1. The simulations were carried out using COMSOL's “Electromagnetic Waves, Frequency Domain” module.

Close modal

In conclusion, we discussed the procedure for defining an air-gap aperture using PEC etching. The lasing characteristics of a PECA VCSEL were analyzed and compared with a previously reported IIA VCSEL.22 The PECA VCSEL shows a similar ηd,top, but a significant increase in peak output power. An increase in Jth is also observed for the PECA VCSEL. Images of the near-field profile showed that the PECA VCSEL has single lateral mode emission; however, the single mode is of a very high-order (LP12,1). Due to the large spreading resistance on the p-side and n-side of the device, we hypothesized that the high-order mode emission is a result of poor current spreading across the aperture, causing more current to be injected near the edge of the aperture, which then suppresses low-order modes. Modes higher than LP12,1 are also not observed, which we attribute to increased scattering losses expected for these higher order modes. This VCSEL used an ITO intracavity contact. Future PECA VCSELs could show significant improvement overall by using a III-nitride tunnel-junction intracavity contact.22 

The authors would like to thank Mitsubishi Chemical Corporation for providing high-quality free-standing m-plane GaN substrates. This work was funded in part by the King Abdulaziz City for Science and Technology (KACST) Technology Innovations Center (TIC) program and the Solid State Lighting and Energy Electronics Center (SSLEEC) at the University of California, Santa Barbara (UCSB). Partial funding for this work came from Professor Boon S. Ooi at King Abdullah University of Science and Technology (KAUST), through his participation in the KACST-TIC program. A portion of this work was done in the UCSB nanofabrication facility, with the support from the NSF NNIN network (ECS-03357650), the UCSB Materials Research Laboratory (MRL), which was supported by the NSF MRSEC program (DMR-1121053), and the California NanoSystem Institute's (CNSIs) Center for Scientific Computing at UCSB.

1.
R.
Michalzik
,
VCSELs: Fundamental, Technology and Applications of Vertical-Cavity Surface-Emitting Lasers
, 1st ed. (
Springer
,
Berlin, Heidelberg
,
2013
).
2.
D. L.
Huffaker
,
D. G.
Deppe
,
K.
Kumar
, and
T. J.
Rogers
,
Appl. Phys. Lett.
65
,
97
(
1994
).
3.
K. D.
Choquette
,
R. P.
Schneider
,
M. H.
Crawford
,
K. M.
Geib
, and
J. J.
Figiel
,
Electron. Lett.
31
,
1145
(
1995
).
4.
D.
Feezell
,
D. A.
Buell
,
D.
Lofgreen
,
M.
Mehta
, and
L. A.
Coldren
,
IEEE J. Quantum Electron.
42
,
494
(
2006
).
5.
M. H. M.
Reddy
,
D.
Feezell
,
T.
Asano
,
D. A.
Buell
,
A. S.
Huntington
,
R.
Koda
, and
L. A.
Coldren
,
J. Electron. Mater.
33
,
118
(
2004
).
6.
D.
Feezell
,
D. A.
Buell
, and
L. A.
Coldren
,
IEEE Photonics Technol. Lett.
17
,
2017
(
2005
).
7.
Y.
Higuchi
,
K.
Omae
,
H.
Matsumura
, and
T.
Mukai
,
Appl. Phys. Express
1
,
121102
(
2008
).
8.
K.
Omae
,
Y.
Higuchi
,
K.
Nakagawa
,
H.
Matsumura
, and
T.
Mukai
,
Appl. Phys. Express
2
,
052101
(
2009
).
9.
D.
Kasahara
,
D.
Morita
,
T.
Kosugi
,
K.
Nakagawa
,
J.
Kawamata
,
Y.
Higuchi
,
H.
Matsumura
, and
T.
Mukai
,
Appl. Phys. Express
4
,
072103
(
2011
).
10.
T.-C.
Lu
,
C.-C.
Kao
,
H.-C.
Kuo
,
G.-S.
Huang
, and
S.-C.
Wang
,
Appl. Phys. Lett.
92
,
141102
(
2008
).
11.
T.-C.
Lu
,
J.-R.
Chen
,
S.-W.
Chen
,
H.-C.
Kuo
,
C.-C.
Kuo
,
C.-C.
Lee
, and
S.-C.
Wang
,
IEEE J. Sel. Top. Quantum Electron.
15
,
850
(
2009
).
12.
T.-C.
Lu
,
T.-T.
Wu
,
S.-W.
Chen
,
P.-M.
Tu
,
Z.-Y.
Li
,
C.-K.
Chen
,
C.-H.
Chen
,
H.-C.
Kuo
,
S.-C.
Wang
,
H.-W.
Zan
, and
C.-Y.
Chang
,
IEEE J. Sel. Top. Quantum Electron.
17
,
1594
(
2011
).
13.
C. O.
Holder
,
J. S.
Speck
,
S. P.
DenBaars
,
S.
Nakamura
, and
D.
Feezell
,
Appl. Phys. Express
5
,
092104
(
2012
).
14.
C. O.
Holder
,
J. T.
Leonard
,
R. M.
Farrell
,
D. A.
Cohen
,
B.
Yonkee
,
J. S.
Speck
,
S. P.
DenBaars
,
S.
Nakamura
, and
D. F.
Feezell
,
Appl. Phys. Lett.
105
,
031111
(
2014
).
15.
T.
Onishi
,
O.
Imafuji
,
K.
Nagamatsu
,
M.
Kawaguchi
,
K.
Yamanaka
, and
S.
Takigawa
,
IEEE J. Quantum Electron.
48
,
1107
(
2012
).
16.
S.
Izumi
,
N.
Fuutagawa
,
T.
Hamaguchi
,
M.
Murayama
,
M.
Kuramoto
, and
H.
Narui
,
Appl. Phys. Express
8
,
062702
(
2015
).
17.
G.
Cosendey
,
A.
Castiglia
,
G.
Rossbach
,
J.-F.
Carlin
, and
N.
Grandjean
,
Appl. Phys. Lett.
101
,
151113
(
2012
).
18.
J. T.
Leonard
,
D. A.
Cohen
,
B. P.
Yonkee
,
R. M.
Farrell
,
T.
Margalith
,
S.
Lee
,
S. P.
DenBaars
,
J. S.
Speck
, and
S.
Nakamura
,
Appl. Phys. Lett.
107
,
011102
(
2015
).
19.
E.
Hashemi
,
J.
Gustavsson
,
J.
Bengtsson
,
M.
Stattin
,
G.
Cosendey
,
N.
Grandjean
, and
A.
Haglund
,
Jpn. J. Appl. Phys., Part 1
52
,
08JG04
(
2013
).
20.
E.
Hashemi
,
J.
Bengtsson
,
J.
Gustavsson
,
M.
Stattin
,
G.
Cosendey
,
N.
Grandjean
, and
A.
Haglund
,
Opt. Express
22
,
411
(
2014
).
21.
D.
Zhuang
and
J. H.
Edgar
,
Mater. Sci. Eng. R
48
,
1
(
2005
).
22.
J. T.
Leonard
,
E. C.
Young
,
B. P.
Yonkee
,
D. a.
Cohen
,
T.
Margalith
,
S. P.
DenBaars
,
J. S.
Speck
, and
S.
Nakamura
,
Appl. Phys. Lett.
107
,
091105
(
2015
).
23.
E. D.
Haberer
,
R.
Sharma
,
C.
Meier
,
a. R.
Stonas
,
S.
Nakamura
,
S. P.
DenBaars
, and
E. L.
Hu
,
Appl. Phys. Lett.
85
,
5179
(
2004
).
24.
E. D.
Haberer
,
R.
Sharma
,
a. R.
Stonas
,
S.
Nakamura
,
S. P.
DenBaars
, and
E. L.
Hu
,
Appl. Phys. Lett.
85
,
762
(
2004
).
25.
A. C.
Tamboli
,
E. D.
Haberer
,
R.
Sharma
,
K. H.
Lee
,
S.
Nakamura
, and
E. L.
Hu
,
Nat. Photonics
1
,
61
(
2007
).
26.
Y.
Gao
,
R. A.
Stonas
,
I.
Ben-Yaacov
,
U.
Mishra
,
S. P.
DenBaars
, and
E. L.
Hu
,
Electron. Lett.
39
,
148
(
2003
).
27.
Y.
Gao
,
I.
Ben-Yaacov
,
U. K.
Mishra
, and
E. L.
Hu
,
J. Appl. Phys.
96
,
6925
(
2004
).
28.
A. R.
Stonas
,
P.
Kozodoy
,
H.
Marchand
,
P.
Fini
,
S. P.
DenBaars
,
U. K.
Mishra
, and
E. L.
Hu
,
Appl. Phys. Lett.
77
,
2610
(
2000
).
29.
A. R.
Stonas
,
T.
Margalith
,
S. P.
DenBaars
,
L. A.
Coldren
, and
E. L.
Hu
,
Appl. Phys. Lett.
78
,
1945
(
2001
).
30.
E.
Trichas
,
C.
Xenogianni
,
M.
Kayambaki
,
P.
Tsotsis
,
E.
Iliopoulos
,
N. T.
Pelekanos
, and
P. G.
Savvidis
,
Phys. Status Solidi
205
,
2509
(
2008
).
31.
A. C.
Tamboli
,
A.
Hirai
,
S.
Nakamura
,
S. P.
DenBaars
, and
E. L.
Hu
,
Appl. Phys. Lett.
94
,
151113
(
2009
).
32.
E.
Trichas
,
M.
Kayambaki
,
E.
Iliopoulos
,
N. T.
Pelekanos
, and
P. G.
Savvidis
,
Appl. Phys. Lett.
94
,
173505
(
2009
).
33.
A. C.
Tamboli
,
M. C.
Schmidt
,
A.
Hirai
,
S. P.
DenBaars
, and
E. L.
Hu
,
J. Electrochem. Soc.
156
,
H767
(
2009
).
34.
A. C.
Tamboli
,
M. C.
Schmidt
,
S.
Rajan
,
J. S.
Speck
,
U. K.
Mishra
,
S. P.
DenBaars
, and
E. L.
Hu
,
J. Electrochem. Soc.
156
,
H47
(
2009
).
35.
L.
Megalini
,
D. L.
Becerra
,
R. M.
Farrell
,
A.
Pourhashemi
,
J. S.
Speck
,
S.
Nakamura
,
S. P.
Denbaars
, and
D. A.
Cohen
,
Appl. Phys. Express
8
,
042701
(
2015
).
36.
L.
Megalini
,
L. Y.
Kuritzky
,
J. T.
Leonard
,
R.
Shenoy
,
K.
Rose
,
S.
Nakamura
,
J. S.
Speck
,
D. A.
Cohen
, and
S. P.
Denbaars
,
Appl. Phys. Express
8
,
066502
(
2015
).
37.
R.
Sharma
,
E. D.
Haberer
,
C.
Meier
,
E. L.
Hu
, and
S.
Nakamura
,
Appl. Phys. Lett.
87
,
051107
(
2005
).
38.
R.
Tao
,
M.
Arita
,
S.
Kako
, and
Y.
Arakawa
,
Appl. Phys. Lett.
103
,
201118
(
2013
).
39.
R.
Tao
,
M.
Arita
,
S.
Kako
,
K.
Kamide
, and
Y.
Arakawa
,
Appl. Phys. Lett.
107
,
101102
(
2015
).
40.
J. T.
Leonard
,
D. A.
Cohen
,
B. P.
Yonkee
,
R. M.
Farrell
,
S. P.
Denbaars
,
J. S.
Speck
, and
S.
Nakamura
,
J. Appl. Phys.
118
,
145304
(
2015
).
41.
R. M.
Farrell
,
D. A.
Haeger
,
X.
Chen
,
M.
Iza
,
A.
Hirai
,
K. M.
Kelchner
,
K.
Fujito
,
A.
Chakraborty
,
S.
Keller
,
S. P.
DenBaars
,
J. S.
Speck
, and
S.
Nakamura
,
J. Cryst. Growth
313
,
1
(
2010
).
42.
R. M.
Farrell
,
D. A.
Haeger
,
X.
Chen
,
C. S.
Gallinat
,
R. W.
Davis
,
M.
Cornish
,
K.
Fujito
,
S.
Keller
,
S. P.
DenBaars
,
S.
Nakamura
, and
J. S.
Speck
,
Appl. Phys. Lett.
96
,
231907
(
2010
).
43.
A.
Hirai
,
Z.
Jia
,
M. C.
Schmidt
,
R. M.
Farrell
,
S. P.
DenBaars
,
S.
Nakamura
,
J. S.
Speck
, and
K.
Fujito
,
Appl. Phys. Lett.
91
,
191906
(
2007
).
44.
E. R.
Hegblom
,
D. I.
Babic
,
B. J.
Thibeault
, and
L. A.
Coldren
,
IEEE J. Sel. Top. Quantum Electron.
3
,
379
(
1997
).
45.
K. G.
Gan
and
J. E.
Bowers
,
IEEE Photonics Technol. Lett.
16
,
1256
(
2004
).
46.
L. A.
Coldren
,
S. W.
Corzine
, and
M. L.
Mašanović
,
Diode Lasers and Photonic Integrated Circuits
(
John Wiley & Sons, Inc.
,
Hoboken, NJ, USA
,
2012
).
47.
M.
Marciniak
,
P.
Śpiewak
,
M.
Wieckowska
, and
R. P.
Sarzala
,
Przegląd Elektrotechniczny
1
,
127
(
2015
).
48.
T.-C.
Lu
,
S.-W.
Chen
,
T.-T.
Wu
,
P.-M.
Tu
,
C.-K.
Chen
,
C.-H.
Chen
,
Z.-Y.
Li
,
H.-C.
Kuo
, and
S.-C.
Wang
,
Appl. Phys. Lett.
97
,
071114
(
2010
).
49.
M.
Kawaguchi
,
O.
Imafuji
,
K.
Nagamatsu
,
K.
Yamanaka
,
S.
Takigawa
, and
T.
Katayama
,
Proc. SPIE
8986
,
89861K
(
2014
).
50.
W.-J.
Liu
,
X.-L.
Hu
,
L.-Y.
Ying
,
J.-Y.
Zhang
, and
B.-P.
Zhang
,
Appl. Phys. Lett.
104
,
251116
(
2014
).
51.
G. P.
Agrawal
,
Fiber-Optic Communication Systems
, 4th ed. (
John Wiley & Sons, Inc.
,
2010
).
52.
W. G.
Scheibenzuber
,
U. T.
Schwarz
,
T.
Lermer
,
S.
Lutgen
, and
U.
Strauss
,
Appl. Phys. Lett.
97
,
021102
(
2010
).
53.
L.
Redaelli
,
H.
Wenzel
,
J.
Piprek
,
T.
Weig
,
S.
Einfeldt
,
M.
Martens
,
G.
Lükens
,
U. T.
Schwarz
, and
M.
Kneissl
,
IEEE J. Quantum Electron.
51
,
2000506
(
2015
).