High-sensitivity avalanche photodiodes (APDs) are used to amplify weak optical signals in a wide range of applications, including telecommunications, data centers, spectroscopy, imaging, light detection and ranging, medical diagnostics, and quantum applications. This paper reports antimony-based separate absorption, charge, and multiplication structure APDs on InP substrates. Al0.7In0.3As0.79Sb0.21 is used for the multiplier region, and InGaAs is used as the absorber. The excess noise is comparable to that of silicon APDs; the k-value is more than one order of magnitude lower than that of APDs that use InP or InAlAs for the gain region. The external quantum efficiency without an anti-reflection coating at 1550 nm is 57%. The gradient of the temperature coefficient of avalanche breakdown voltage is 6.7 mV/K/μm, which is less than one-sixth that of InP APDs, presenting the potential to reduce the cost and complexity of receiver circuits. Semi-insulating InP substrates make high-speed operation practical for widely reported AlxIn1−xAsySb1−y-based APDs.

Highly sensitive photodetectors are key components in applications ranging from thermal imaging, optical communication, biological and chemical agent detection, and light detection and ranging (LiDAR).1–4 Avalanche photodiodes (APDs) have widely been deployed in these applications because the low-noise internal gain achieved through impact ionization provides the potential for a higher signal-to-noise performance than p–i–n photodiodes. However, the maximum useful multiplication gain is limited by two factors. First, the impact ionization process’s stochastic nature is a noise source that increases with increasing gain.5 The figure of merit for this noise is the excess noise factor, F(M), which is treated as a multiplicative of the shot noise. The excess noise factor can be approximated by
FM=kM+1k21M,
(1)
where ⟨M⟩ is the average gain and k is the ratio of impact ionization coefficients of holes (β) to those of electrons (α) (if βα). Second, the avalanche buildup time affects the gain–bandwidth product, which can limit the response time at high gain.6 Due to these two limiting mechanisms, a lower k-value is always desirable to provide lower excess noise (i.e., higher signal-to-noise ratio) and a higher response speed.

A lower k-value can be realized by band structure engineering7,8 or by selecting semiconductor materials with favorable impact ionization coefficients.9–11 For most applications, selecting semiconductor materials with preferable parameters wins due to their simpler device structures. There have been several widely investigated semiconductor materials with low excess noise, such as Si,12,13 InAs,14 and Hg0.7Cd0.3Te.15 However, internal limitations, such as the limited infrared spectral region for Si, the requirement for cryogenic temperatures for Hg0.7Cd0.3Te, and expensive and small-size substrates and tunneling dark current for InAs, are concerns for these materials.11 Recently, antimony (Sb)-based APDs have attracted interest due to their low excess noise.2 One of these materials is AlxIn1−xAsySb1−y grown lattice-matched to GaSb. It has achieved excess noise comparable to Si16–18 with the added advantage that the bandgap energy can be adjusted to operate from the near-infrared to the mid-wavelength infrared.19 

APD receivers typically exhibit significant changes in sensitivity with changing temperature due to the strong temperature dependence of the breakdown voltage. An active variable bias circuit or an embedded thermoelectric cooler is required to tune the reverse bias or control the temperature for stable gain, increasing the detection system complexity. Meanwhile, the Sb-based material systems have a weak temperature dependence of avalanche breakdown20–25 since the addition of Sb gives rise to high alloy disorder potentials, making the temperature-independent alloy scattering dominate over the temperature-dependent phonon scattering.

Another critical parameter is the APD bandwidth. The bandwidth of APDs based on AlxIn1−xAsySb1−y lattice-matched to GaSb is limited by the highly doped GaSb substrates, leading to a low RC-limited bandwidth.26 This has been addressed using Al0.79In0.21As0.74Sb0.26 lattice-matched to semi-insulating InP substrates.27 The advantages of InP-substrate-based APDs for short-wavelength infrared (SWIR) applications have widely been reported.28–32 The most commonly used telecommunication APD structure consists of separate absorption, charge, and multiplication (SACM) regions. InGaAs is the material of choice for the absorption region,32–34 although InGaAs/GaAsSb type-II superlattices have proved useful for operation at longer wavelengths, e.g., 2 µm.35 Either InAlAs (k ∼ 0.2) or InP (k ∼ 0.45) is used as the multiplication layer in the SACM structure. However, their high excess noise and weak temperature stability impair the performance of APDs. This paper reports the first InGaAs/Al0.7In0.3As0.79Sb0.21 (hereafter AlInAsSb) SACM APD structure on an InP substrate. AlInAsSb, grown as a digital alloy,19 is used for the multiplication layer to achieve low excess noise and strong temperature stability. InGaAs is used as the absorber for SWIR photodetection.

The structure details are shown in Fig. 1(a). The APD consists of a 1-µm AlInAsSb multiplier (with a bandgap at ∼1.29 eV36) and a 1-µm InGaAs absorber. Since phase separation remains a challenge for the growth of random alloy AlInAsSb on InP37 with a predicted miscibility gap spanning nearly all lattice-matched compositions,38 AlInAsSb was grown as a digital alloy19 with a 10 monolayer period comprised of AlAs, AlSb, and InAs layers. Furthermore, several SACM APDs were grown with different charge layer parameters to optimize the electric field distribution at the absorber and multiplier. The charge layer parameters shown in Fig. 1(a) were determined to provide a high electric field at the AlInAsSb multiplier for the impact ionization process while keeping a low electric field at the InGaAs absorber to suppress the tunneling effect.

FIG. 1.

(a) Epitaxial structure of InGaAs/AlInAsSb SACM APDs. The AlInAsSb was grown as a digital alloy. (b) Band diagram at 0 V and (c) HR-XRD omega–2theta scans of InGaAs/AlInAsSb APDs.

FIG. 1.

(a) Epitaxial structure of InGaAs/AlInAsSb SACM APDs. The AlInAsSb was grown as a digital alloy. (b) Band diagram at 0 V and (c) HR-XRD omega–2theta scans of InGaAs/AlInAsSb APDs.

Close modal

Figure 1(b) shows the band diagram of InGaAs/AlInAsSb APDs at zero bias. The significant conduction band offset between InGaAs and AlInAsSb makes the design of the bandgap grading scheme important. The scheme used In0.53(AlxGa1−x)0.47As and AlxIn1−xAsySb1−y lattice-matched to InP. A previous report of an InGaAs/Al0.85Ga0.15As0.56Sb0.44 SACM APD used a single In0.53(AlxGa1−x)0.47As step bandgap grading scheme, leaving a significant conduction band offset between the grading layer and the Al0.85Ga0.15As0.56Sb0.44 charge layer.34 This work used a combination of In0.53(AlxGa1−x)0.47As and AlxIn1−xAsySb1−y bandgap grading layers to assist carrier injection from the absorber to the multiplier. Figure 1(c) shows the high-resolution X-ray diffraction (HR-XRD) pattern of the investigated APDs with the labeled fringe peaks of InP substrate, InGaAs absorber, and AlInAsSb multiplier/charge layer.

Figure 2(a) shows the current–voltage (I–V) characteristics under dark and 1550-nm illumination conditions of a 150-μm-diameter device at room temperature. The measured punch-through voltage is ∼48 V. SACM APDs exhibit a step in the photocurrent when the depletion edge extends to the absorber. These InGaAs/AlInAsSb APDs show a second step at 55–57.5 V in the photocurrent curve, which might result from the traps created during the growth of the bandgap grading layers. With increasing temperature, the second step gradually disappears. The following analysis of multiplication gain and excess noise is based on measurements above 57.5 V. The reverse bias at 57.5 V cannot be regarded as the unity gain point. Instead, the multiplication gain was determined by fitting the measured excess noise to the measured photocurrent at the reverse bias above 57.5 V.39,40 The maximum gain is ∼17. A higher gain can be achieved with a higher charge layer doping to suppress the onset of tunneling in the InGaAs absorber.

FIG. 2.

(a) Current–voltage (I–V) characteristics under dark and 1550-nm illumination conditions for 150-μm-diameter InGaAs/AlInAsSb SACM APDs at room temperature. (b) Excess noise factor vs gain. (c) External quantum efficiency vs wavelength. (d) Simulated −3 dB bandwidth and gain–bandwidth product vs gain for 40-μm-diameter InGaAs/AlInAsSb SACM APDs.

FIG. 2.

(a) Current–voltage (I–V) characteristics under dark and 1550-nm illumination conditions for 150-μm-diameter InGaAs/AlInAsSb SACM APDs at room temperature. (b) Excess noise factor vs gain. (c) External quantum efficiency vs wavelength. (d) Simulated −3 dB bandwidth and gain–bandwidth product vs gain for 40-μm-diameter InGaAs/AlInAsSb SACM APDs.

Close modal

Figure 2(b) shows the excess noise characteristics under 1550-nm illumination. The dashed lines are plots using Eq. (1) for k-values ranging from 0 to 0.1. The measured k-value of the InGaAs/AlInAsSb APD is 0.02 to 0.04, consistent with previous excess noise measurements of Al0.79In0.21As0.74Sb0.26 p+–i–n+ APDs,27 comparable to Ge/Si SACM APDs39,41 and an order of magnitude lower than InGaAs/InP42,43 or InGaAs/InAlAs44,45 SACM APDs. The low excess noise in these APDs originates from the addition of the heavy Sb atom, introducing a high valence band spin–orbit splitting energy that restricts hole impact ionization,46 similar to GaAsBi.11 

Figure 2(c) shows the external quantum efficiency at unity gain from 800 to 1900 nm. Specifically, at 1064, 1310, and 1550 nm, the efficiencies are 73%, 61%, and 57%, respectively. Higher responsivity at designated wavelengths could be realized with an anti-reflection coating and a photon-trapping structure.47 

Another important parameter of InGaAs/AlInAsSb APDs is the −3 dB bandwidth consisting of transit-time-limited and RC-limited bandwidths. The transit-time-limited bandwidth has been calculated via the random path length (RPL) model,48 and the RC-limited bandwidth has been calculated for 40-μm-diameter devices. Figure 2(d) shows the simulated −3 dB bandwidth and the corresponding gain–bandwidth product.

The temperature coefficient of avalanche breakdown voltage, Cbd, characterizes the shift in breakdown voltage with varying temperature,
Cbd=ΔVbdΔT,
(2)
where ΔVbd is the breakdown voltage change and ΔT is the temperature change. The I–V characteristics of 150-μm-diameter SACM APDs were measured from 200 to 340 K under dark and 1550-nm illumination conditions. The gain curves at various temperatures are shown in Fig. 3(a). The breakdown voltages were determined as the intercepts at 1/M = 0, as shown in Fig. 3(b). The Cbd was determined to be (14.58 ± 0.63) mV/K via linear fitting regression. The gradient of Cbd, P, was used to compare the temperature sensitivity between different material-based APDs with different depletion layer thicknesses,21 and the P of the InGaAs/AlInAsSb SACM APD is 6.7 mV/K/μm. The P values for InAlAs, InP, and Si APDs49,50 and AlAs0.56Sb0.44 APDs lattice-matched to InP21 are listed in Table I. The temperature-dependent dark current is shown in Fig. 3(c). Dark current from tunneling is observed at high reverse biases, indicating that a higher gain can be obtained with a higher charge layer doping to suppress the high electric field at the InGaAs absorber.
FIG. 3.

(a) Measured gain vs reverse bias, (b) inverse gain (symbols) and linear fitting (lines) under 1550-nm illumination, and (c) dark current as a function of reverse bias from 200 to 340 K with 20 K steps.

FIG. 3.

(a) Measured gain vs reverse bias, (b) inverse gain (symbols) and linear fitting (lines) under 1550-nm illumination, and (c) dark current as a function of reverse bias from 200 to 340 K with 20 K steps.

Close modal
TABLE I.

Comparison of the gradient of Cbd.

MaterialGradient of Cbd (mV/K/μm)
InGaAs/AlInAsSb APDs 6.7 
AlAs0.56Sb0.44 APDs21  8.5 
InAlAs APDs50  16.5 
Si APDs49  25 
InP APDs50  43 
MaterialGradient of Cbd (mV/K/μm)
InGaAs/AlInAsSb APDs 6.7 
AlAs0.56Sb0.44 APDs21  8.5 
InAlAs APDs50  16.5 
Si APDs49  25 
InP APDs50  43 

The room-temperature and cryogenic-temperature electrical characterizations of the investigated InGaAs/AlInAsSb APD exhibit its high performance in comparison with commercially available APDs, making such an APD a promising candidate in SWIR applications. For example, currently available LiDAR systems in autonomous vehicles primarily utilize Si APDs operating at 900 nm. However, high-sensitivity APDs operating at longer wavelengths are preferable due to the improved detection range and resolution, and the longer operation wavelength is more eye-safe. At the same time, the strong temperature dependence of Si APDs complicates the optical receiver design, and the increased receiver cost results in these LiDAR systems being less competitive in the market, making it necessary to employ an APD with strong temperature stability capable of operating under varying temperatures. Furthermore, InGaAs/InP or InGaAs/InAlAs APDs are usually fabricated on InP-substrate-based optical communications, but their noise performance is limited. A new multiplier lattice-matched to InP with lower excess noise is beneficial. Therefore, the InGaAs/AlInAsSb APDs have potential to be widely used in these applications as a promising replacement of APDs in the current market.

We report low-noise, weak-temperature-sensitivity 1550-nm InGaAs/AlInAsSb SACM APDs on InP substrates. For APDs on InP substrates under 1550-nm illumination, the InGaAs/AlInAsSb SACM APDs’ excess noise is significantly lower than that of InAlAs-based and InP-based APDs. At 1550 nm, the external quantum efficiency is 57% without an anti-reflection coating. The weak temperature dependence of avalanche breakdown has been demonstrated, making these APDs promising candidates for photodetection under varying ambient conditions. Furthermore, the use of semi-insulating InP substrates is beneficial for high-speed operation.

The epitaxial structures were grown on Fe-doped semi-insulating InP substrates using solid-source molecular beam epitaxy (MBE) in an EPI Mod Gen II MBE system. Dual filament effusion cells were used to supply Al, Ga, and In fluxes. Dimeric arsenic was supplied using a valved arsenic cracker source operated at 900 °C, and dimeric antimony was supplied using a valved corrosive antimony cracker source operated at 900 °C. Prior to the growth initiation, the InP native oxide was removed in the chamber by thermal desorption with an arsenic beam equivalent pressure of 6 × 10−6 Torr supplied to stabilize the wafer surface. The substrate temperature during growth was measured using real-time band-edge thermometry and emissivity correcting pyrometry, and it was maintained at 470 °C for the growth of all epitaxial layers. The InGaAs N-contact layer, P-contact layer, and UID absorber were grown as the conventional random alloy InGaAs. The AlInAsSb field control, multiplication, charge, and blocking layers and the InGaAs → InAlAs and InAlAs → AlInAsSb grading layers were grown as digital alloys (DAs). 

The DA AlInAsSb growth conditions on InP were adapted from the growth of DA AlInAsSb on GaSb.19 Namely, the DA was grown in a ten monolayer (2.93 nm) period with alternating layers of AlAs, AlSb, and InAs. The InSb layer used in the growth of DA AlInAsSb on GaSb was omitted, since the increase in lattice-mismatch strain between InSb and InP degraded the structural quality of the AlInAsSb on InP. Accordingly, the aluminum fraction of the ten monolayer DA AlInAsSb on InP was determined by the sum of the thicknesses of the AlAs and AlSb layers, and the antimony fraction is determined by the thickness of the AlSb layer. The indium and aluminum fluxes were calibrated through the growth of InGaAs and InAlAs random alloys of varying compositions, and the AlAs, AlSb, and InAs growth rates were fixed at 0.695 monolayer/s for the DA layers. During the growth of the DA AlInAsSb, the arsenic beam equivalent pressure was maintained at 2.3 × 10−6 Torr, and the antimony beam equivalent pressure was maintained at 3.2 × 10−6 Torr layers. The arsenic and antimony source shutters were used to transition from AlAs to AlSb and vice versa.

The InAlAs → AlInAsSb grading layer was grown in the same fashion as the uniform composition DA layers utilizing 10 monolayer periods. The composition was linearly graded from AlInAsSb to InAlAs over 17 periods (50 nm) by adjusting the relative thicknesses of the AlAs, AlSb, and InAs layers. The InGaAs → InAlAs digital grading layer consisted of alternating layers of AlAs, GaAs, and InAs. The period of the InGaAs → InAlAs digital grading layer was reduced from 10 monolayers to 5 monolayers to accommodate the increase in lattice-mismatch strain from the increased indium fraction in DA InAlGaAs. The Ga flux for the GaAs DA layers was constrained by the flux needed to lattice-match the random alloy InGaAs layers, resulting in a GaAs growth rate of 0.648 monolayer/s. The composition was linearly graded from InAlAs to InGaAs by adjusting the relative thickness of the AlAs, GaAs, and InAs layers over 34 periods (50 nm). Prior to the device growth, the uniform composition and graded DA materials were grown as isolated calibration structures to verify lattice-matching conditions to the InP substrate.

The HR-XRD pattern of InGaAs/AlInAsSb APDs was acquired using a Bruker D8 X-ray diffractometer. This instrument was equipped with a Cu Ka radiation source with a wavelength of 0.1542 nm, operating at 40 kV and 40 mA. The angle, denoted as ω, was systematically varied in the range of 29.5°–34.5°, encompassing a broad spectrum to capture detailed insights into digital-alloy grown layers in the sample.

The fabrication flow is shown in Fig. 4. The circular mesas were defined by conventional photolithography via an AZ 5214 photoresist, and the structures were formed via chemical etching. The solution of H3PO4, H2O2, and H2O was used to etch InGaAs layers, and the solution of citric acid, H3PO4, H2O2, and H2O was used to etch the other Al-containing layers. The chemical etching was used to obtain a smooth sidewall in comparison with the dry etch method, and different chemical solutions were used for different layers to obtain a straighter sidewall of APDs. Then, the p-type and n-type contacts were defined by conventional photolithography via an AZ 5214 photoresist. The Ti/Au was then deposited on top and bottom InGaAs contact layers via electron-beam evaporation, and the contacts were created after the lift-off process. Finally, the SU-8 was spun on the sidewall to suppress the surface leakage current.

FIG. 4.

Fabrication of InGaAs/AlInAsSb SACM APDs.

FIG. 4.

Fabrication of InGaAs/AlInAsSb SACM APDs.

Close modal

A stabilized tungsten light source and a monochromator were used to provide the normal illumination on the devices in a wide wavelength range. The photocurrent was measured using the lock-in amplifier, which can remove the dark current. A calibrated strained-layer InGaAs photodiode and the studied APDs were illuminated, respectively, and the corresponding photocurrents were obtained. Since the responsivity/external quantum efficiency of the calibrated strained-layer InGaAs photodiode is known, the external quantum efficiency of the studied devices can be determined.

The room-temperature current–voltage characteristics were determined with a Keithley 2400 source meter, and a 1550-nm temperature-stabilized laser coupled into a lensed fiber was used to illuminate the devices. The cryogenic-temperature current–voltage characteristics were determined with an HP 4145 semiconductor parameter analyzer, and the same laser was used to provide the 1550-nm illumination. The temperature-dependent measurements were performed in a liquid-nitrogen-cooled cryogenic chamber.

Before the excess noise measurements on the APDs, system calibration of the Agilent N8973A noise figure analyzer was performed by using a calibrated Agilent 346A noise source. The APDs were then DC-biased using a Keithley 2400 source meter via a bias tee, and a 1550-nm temperature-stabilized laser coupled into a lensed fiber was used to illuminate the devices. Finally, the AC component of the output current through the bias tee was read via the noise figure analyzer at different reverse biases. The excess noise measurements at a higher gain were terminated due to the current tolerance of the noise figure analyzer.

As for the transit-time-limited bandwidth, the RPL model calculated the time it took for the avalanche buildup in APDs by considering the random ionization times of carriers in the multiplier. The probability distribution functions (PDFs) for carrier ionization path length, determined by the electron and hole ionization coefficients,51 were employed to calculate the impact ionization probability within the AlInAsSb multiplier. With a total of 500,000 trials, we can achieve statistically reliable results of multiplication gain and transit time. To obtain the impulse current response for each trial, Ramo’s theorem52 was applied. The avalanche current was calculated by i = qvwd, where v is the carrier drift velocity, q is the electric charge, and wd is the depletion thickness. The frequency response was then determined by performing a Fourier transform on the mean current impulse response, and the transit-time-limited bandwidth, ftr, can be determined. As for the RC-limited bandwidth, it can be calculated via fRC=12πCAPDRl+Rs, where CAPD is the APD’s capacitance after full depletion, Rl is the load resistance, and Rs is the series resistance. Finally, the −3 dB bandwidth can be calculated via f3dB=11ftr2+1fRC2.

This work was supported by the Directed Energy-Joint Technology Office (DE-JTO), Award No. N00014-17-1-2440.

The authors have no conflicts to disclose.

B.G. and M.S. contributed equally to this work.

Bingtian Guo: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Writing – original draft (lead); Writing – review & editing (equal). Mariah Schwartz: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Writing – review & editing (equal). Sri H. Kodati: Methodology (supporting). Kyle M. McNicholas: Methodology (supporting). Hyemin Jung: Methodology (supporting). Seunghyun Lee: Investigation (supporting). Jason Konowitch: Methodology (supporting). Dekang Chen: Methodology (supporting). Junwu Bai: Methodology (supporting). Xiangwen Guo: Methodology (supporting). Theodore J. Ronningen: Methodology (supporting); Project administration (supporting). Christoph H. Grein: Supervision (supporting). Joe C. Campbell: Supervision (supporting); Writing – review & editing (supporting). Sanjay Krishna: Funding acquisition (equal); Project administration (equal); Supervision (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
J. C.
Campbell
,
J. Lightwave Technol.
34
,
278
(
2016
).
2.
J. C.
Campbell
,
IEEE J. Sel. Top. Quantum Electron.
28
,
3800911
(
2021
).
3.
B.
Wang
and
J.
Mu
,
PhotoniX
3
,
8
(
2022
).
4.
S.
Krishna
,
S.
Lee
,
S. H.
Kodati
,
M.
Schwartz
,
H.
Jung
,
T. J.
Ronningen
,
B.
Guo
,
A. H.
Jones
,
M.
Winslow
,
J. C.
Campbell
, and
C. H.
Grein
,
IEEE J. Quantum Electron.
58
,
4500207
(
2022
).
5.
R. J.
McIntyre
,
IEEE Trans. Electron Dev.
ED-13
,
164
(
1966
).
6.
R. B.
Emmons
,
J. Appl. Phys.
38
,
3705
(
1967
).
7.
S. D.
March
,
A. H.
Jones
,
J. C.
Campbell
, and
S. R.
Bank
,
Nat. Photonics
15
,
468
(
2021
).
8.
F.
Capasso
,
W.-T.
Tsang
, and
G. F.
Williams
,
IEEE Trans. Electron Devices
30
,
381
(
1983
).
9.
S. H.
Kodati
,
S.
Lee
,
B.
Guo
,
A. H.
Jones
,
M.
Schwartz
,
M.
Winslow
,
N. A.
Pfiester
,
C. H.
Grein
,
T. J.
Ronningen
,
J. C.
Campbell
, and
S.
Krishna
,
Appl. Phys. Lett.
118
,
081106
(
2021
)
10.
S.
Lee
,
B.
Guo
,
S. H.
Kodati
,
H.
Jung
,
M.
Schwartz
,
A. H.
Jones
,
M.
Winslow
,
C. H.
Grein
,
T. J.
Ronningen
,
J. C.
Campbell
, and
S.
Krishna
,
Appl. Phys. Lett.
120
,
071101
(
2022
).
11.
Y.
Liu
,
X.
Yi
,
N. J.
Bailey
,
Z.
Zhou
,
T. B. O.
Rockett
,
L. W.
Lim
,
C. H.
Tan
,
R. D.
Richards
, and
J. P. R.
David
,
Nat. Commun.
12
,
4784
(
2021
).
12.
C. A.
Lee
,
R. A.
Logan
,
R. L.
Batdorf
,
J. J.
Kleimack
, and
W.
Wiegmann
,
Phys. Rev.
134
,
A761
(
1964
).
13.
V. M.
Robbins
,
T.
Wang
,
K. F.
Brennan
,
K.
Hess
, and
G. E.
Stillman
,
J. Appl. Phys.
58
,
4614
(
1985
).
14.
P. J.
Ker
,
J. P. R.
David
, and
C. H.
Tan
,
Opt. Express
20
,
29568
(
2012
).
15.
X.
Sun
,
J. B.
Abshire
,
J. D.
Beck
,
P.
Mitra
,
K.
Reiff
, and
G.
Yang
,
Opt. Express
25
,
16589
(
2017
).
16.
M. E.
Woodson
,
M.
Ren
,
S. J.
Maddox
,
Y.
Chen
,
S. R.
Bank
, and
J. C.
Campbell
,
Appl. Phys. Lett.
108
,
081102
(
2016
).
17.
A. H.
Jones
,
A. K.
Rockwell
,
S. D.
March
,
Y.
Yuan
,
S. R.
Bank
, and
J. C.
Campbell
,
IEEE Photonics Technol. Lett.
31
,
1948
(
2019
).
18.
Y.
Yuan
,
A. K.
Rockwell
,
Y.
Peng
,
J.
Zheng
,
S. D.
March
,
A. H.
Jones
,
M.
Ren
,
S. R.
Bank
, and
J. C.
Campbell
,
J. Lightwave Technol.
37
,
3647
(
2019
).
19.
S. J.
Maddox
,
S. D.
March
, and
S. R.
Bank
,
Cryst. Growth Des.
16
,
3582
(
2016
).
20.
A. H.
Jones
,
Y.
Yuan
,
M.
Ren
,
S. J.
Maddox
,
S. R.
Bank
, and
J. C.
Campbell
,
Opt. Express
25
,
24340
(
2017
).
21.
X.
Jin
,
S.
Xie
,
B.
Liang
,
X.
Yi
,
H.
Lewis
,
L. W.
Lim
,
Y.
Liu
,
B. K.
Ng
,
D. L.
Huffaker
,
C. H.
Tan
,
D. S.
Ong
, and
J. P. R.
David
,
IEEE J. Sel. Top. Quantum Electron.
28
,
3801208
(
2022
).
22.
B.
Guo
,
S. Z.
Ahmed
,
X.
Xue
,
A.-K.
Rockwell
,
J.
Ha
,
S.
Lee
,
B.
Liang
,
A. H.
Jones
,
J. A.
McArthur
,
S. H.
Kodati
,
T. J.
Ronningen
,
S.
Krishna
,
J. S.
Kim
,
S. R.
Bank
,
A. W.
Ghosh
, and
J. C.
Campbell
,
J. Lightwave Technol.
40
,
5934
(
2022
).
23.
Y.
Cao
,
T.
Osman
,
E.
Clarke
,
P. K.
Patil
,
J. S.
Ng
, and
C. H.
Tan
,
J. Lightwave Technol.
40
,
4709
(
2022
).
24.
X.
Zhou
,
C. H.
Tan
,
S.
Zhang
,
M.
Moreno
,
S.
Xie
,
S.
Abdullah
, and
J. S.
Ng
,
R. Soc. Open Sci.
4
,
170071
(
2017
).
25.
S.
Abdullah
,
C. H.
Tan
,
X.
Zhou
,
S.
Zhang
,
L.
Pinel
, and
J. S.
Ng
,
Opt. Express
25
,
33610
(
2017
).
26.
A. H.
Jones
,
Y.
Shen
,
K.
Sun
,
D.
Chen
,
S. D.
March
,
S. R.
Bank
, and
J. C.
Campbell
,
Opt. Express
29
,
38939
(
2021
).
27.
S. H.
Kodati
,
S.
Lee
,
B.
Guo
,
A. H.
Jones
,
M.
Schwartz
,
M.
Winslow
,
N. A.
Pfiester
,
C. H.
Grein
,
T. J.
Ronningen
,
J. C.
Campbell
, and
S.
Krishna
,
Appl. Phys. Lett.
118
,
091101
(
2021
).
28.
M.
Nada
,
Y.
Muramoto
,
H.
Yokoyama
,
T.
Ishibashi
, and
H.
Matsuzaki
,
J. Lightwave Technol.
32
,
1543
(
2014
).
29.
M.
Nada
,
H.
Yokoyama
,
Y.
Muramoto
,
T.
Ishibashi
, and
H.
Matsuzaki
,
Opt. Express
22
,
14681
(
2014
).
30.
M.
Nada
,
T.
Yoshimatsu
,
Y.
Muramoto
,
H.
Yokoyama
, and
H.
Matsuzaki
,
J. Lightwave Technol.
33
,
984
(
2015
).
31.
M.
Nada
,
T.
Yoshimatsu
,
F.
Nakajima
,
K.
Sano
, and
H.
Matsuzaki
,
J. Lightwave Technol.
37
,
260
(
2019
).
32.
A.
Beling
,
X.
Xie
, and
J. C.
Campbell
,
Optica
3
,
328
(
2016
).
33.
F.
Yu
,
T.-C.
Tzu
,
J.
Gao
,
T.
Fatema
,
K.
Sun
,
P.
Singaraju
,
S. M.
Bowers
,
C.
Reyes
, and
A.
Beling
,
IEEE J. Sel. Top. Quantum Electron.
29
,
3800106
(
2022
).
34.
S.
Xie
,
X.
Zhou
,
S.
Zhang
,
D. J.
Thomson
,
X.
Chen
,
G. T.
Reed
,
J. S.
Ng
, and
C. H.
Tan
,
Opt. Express
24
,
24242
(
2016
).
35.
D. S. G.
Ong
,
J. S.
Ng
,
Y. L.
Goh
,
C. H.
Tan
,
S.
Zhang
, and
J. P. R.
David
,
IEEE Trans. Electron Dev.
58
,
486
(
2011
).
36.
B.
Guo
,
A. H.
Jones
,
S.
Lee
,
S. H.
Kodati
,
B.
Liang
,
X.
Xue
,
N. A.
Pfiester
,
M.
Schwartz
,
M.
Winslow
,
C. H.
Grein
,
T. J.
Ronningen
,
S.
Krishna
, and
J. C.
Campbell
,
Appl. Phys. Lett.
119
,
171109
(
2021
).
37.
S.
Tomasulo
,
M.
Gonzalez
,
M. P.
Lumb
,
C. R.
Brown
,
A. H.
Dicarlo
,
I. R.
Sellers
,
I.
Vurgaftman
,
J. R.
Meyer
,
R. J.
Walters
, and
M. K.
Yakes
,
J. Cryst. Growth
548
,
125826
(
2020
).
38.
A. N.
Semenov
,
V. A.
Solov’Ev
,
B. Y.
Meltser
,
Y. V.
Terent’ev
,
L. G.
Prokopova
, and
S. V.
Ivanov
,
J. Cryst. Growth
278
,
203
(
2005
).
39.
H.-D.
Liu
,
H.
Pan
,
C.
Hu
,
D.
McIntosh
,
Z.
Lu
,
J.
Campbell
,
Y.
Kang
, and
M.
Morse
,
J. Appl. Phys.
106
,
064507
(
2009
).
40.
A. H.
Jones
,
S. D.
March
,
S. R.
Bank
, and
J. C.
Campbell
,
Nat. Photonics
14
,
559
(
2020
).
41.
Y.
Kang
,
H.-D.
Liu
,
M.
Morse
,
M. J.
Paniccia
,
M.
Zadka
,
S.
Litski
,
G.
Sarid
,
A.
Pauchard
,
Y.-H.
Kuo
,
H.-W.
Chen
,
W. S.
Zaoui
,
J. E.
Bowers
,
A.
Beling
,
D. C.
McIntosh
,
X.
Zheng
, and
J. C.
Campbell
,
Nat. Photonics
3
,
59
(
2009
).
42.
N.
Susa
,
H.
Nakagome
,
O.
Mikami
,
H.
Ando
, and
H.
Kanbe
,
IEEE J. Quantum Electron.
16
,
864
(
1980
).
43.
L. J. J.
Tan
,
J. S.
Ng
,
C. H.
Tan
, and
J. P. R.
David
,
IEEE J. Quantum Electron.
44
,
378
(
2008
).
44.
C.
Lenox
,
H.
Nie
,
P.
Yuan
,
G.
Kinsey
,
A. L.
Homles
,
B. G.
Streetman
, and
J. C.
Campbell
,
IEEE Photonics Technol. Lett.
11
,
1162
(
1999
).
45.
Y.
Yuan
,
D.
Jung
,
K.
Sun
,
J.
Zheng
,
A. H.
Jones
,
J. E.
Bowers
, and
J. C.
Campbell
,
Opt. Lett.
44
,
3538
(
2019
).
46.
B.
Guo
,
X.
Jin
,
S.
Lee
,
S. Z.
Ahmed
,
A.
Jones
,
X.
Xue
,
B.
Liang
,
H.
Lewis
,
S. H.
Kodati
,
D.
Chen
,
T. J.
Ronningen
,
C. H.
Grein
,
A. W.
Ghosh
,
S.
Krishna
,
J. P. R.
David
, and
J. C.
Campbell
,
J. Lightwave Technol.
40
,
4758
(
2022
).
47.
D.
Chen
,
S. D.
March
,
A. H.
Jones
,
Y.
Shen
,
A. A.
Dadey
,
K.
Sun
,
J. A.
McArthur
,
A. M.
Skipper
,
X.
Xue
,
B.
Guo
et al
,
Nat. Photonics
17
,
594
(
2023
).
48.
D. S.
Ong
,
K.
Li
,
G. J.
Rees
,
J. P. R.
David
, and
P. N.
Robson
,
J. Appl. Phys.
83
,
3426
(
1998
).
49.
D. J.
Massey
,
J. P. R.
David
, and
G. J.
Rees
,
IEEE Trans. Electron Dev.
53
,
2328
(
2006
).
50.
L. J. J.
Tan
,
D. S. G.
Ong
,
J. S.
Ng
,
C. H.
Tan
,
S. K.
Jones
,
Y.
Qian
, and
J. P. R.
David
,
IEEE J. Quantum Electron.
46
,
1153
(
2010
).
51.
T.
Ronningen
,
S.
Kodati
,
X.
Jin
,
S.
Lee
,
H.
Jung
,
X.
Tao
,
H.
Lewis
,
M.
Schwartz
,
N.
Gajowski
,
P.
Martyniuk
,
B.
Guo
,
A. H.
Jones
,
J. C.
Campbell
,
C. H.
Grein
,
J. P. R.
David
, and
S.
Krishna
,
Appl. Phys. Lett.
123
,
131110
(
2023
).