We report single-photon emission from electrically driven site-controlled InGaN/GaN quantum dots. The device is fabricated from a planar light-emitting diode structure containing a single InGaN quantum well, using a top-down approach. The location, dimension, and height of each single-photon-emitting diode are controlled lithographically, providing great flexibility for chip-scale integration.

Single-photon source (SPS) is a critical resource for a wide range of applications, including quantum communication,1 optical quantum computing,2,3 and precision measurement.4,5 Quantum dot (QD) based on III-N semiconductor has become a promising candidate for practical, on-demand SPS for its distinct advantages. As a wide-bandgap semiconductor with large exciton binding energy (28 meV in bulk GaN), III-N QD allows single-photon emission up to room temperature.6 Compared to other materials that can be used for SPSs,7–12 chip-scale classical devices have been well developed for III-N, since III-N is an important material system for optoelectronics and power electronics. For III-N based SPSs, electrical-injection13 and site control6,14—two critical requirements for chip-scale SPS—have also been demonstrated, although not simultaneously. Integrating electrical injection in site-controlled QDs remains challenging for QDs fabricated by bottom-up approaches or in non-planar structures, such as micro-pyramids,15 inversed micro-pyramids,16 and nano-pillars.6 In these structures, the QDs are typically sandwiched between non-planar wetting layer and barrier layers. Therefore, they are susceptible to high series resistance and large leaky current. In this work, we use QDs fabricated by a top-down approach,14,17 where the p-i-n junction was formed via standard thin-film epitaxy as in a typical III-N light-emitting diode (LED), and the current-pathway naturally overlaps with the QD active region. With this structure, we demonstrate electrically driven single-photon emission from site-controlled InGaN/GaN QDs.

The fabrication procedure of the InGaN single-photon-emitting diode (SPED) is illustrated in Fig. 1. To fabricate the site-controlled QDs, we use a planar epitaxial structure (Fig. 1(a)) similar to a standard InGaN based LED. It consists of a single InGaN/GaN quantum-well (QW) with a nominal indium composition of 15% and thickness of 3 nm, sandwiched between p-GaN and n-GaN. The planar structure is provided by NOVAGAN and is grown by metalorganic chemical vapor deposition (MOCVD) on a (0001)-oriented planar sapphire substrate. For simplicity, no electron or hole blocking layer is employed but can be added in future work. The room temperature carrier concentrations in p-GaN and n-GaN are measured using the Hall effect to be p=1×1017 cm−3 and n=1×1018 cm−3, respectively.

FIG. 1.

Schematic of sample fabrication procedure: (a) planar wafer grown by MOCVD on (0001) sapphire, (b) patterning of InGaN nano-pillars using EBL and ICP-RIE, (c) selective chemical etching to form vertical nano-pillar sidewalls, (d) planarization with SOG, (e) SOG etchback to expose p-GaN, and (f) formation of metal contacts.

FIG. 1.

Schematic of sample fabrication procedure: (a) planar wafer grown by MOCVD on (0001) sapphire, (b) patterning of InGaN nano-pillars using EBL and ICP-RIE, (c) selective chemical etching to form vertical nano-pillar sidewalls, (d) planarization with SOG, (e) SOG etchback to expose p-GaN, and (f) formation of metal contacts.

Close modal

We create 90 × 90 nano-pillar arrays with 5 μm pitch using electron-beam lithography and inductively coupled plasma-reactive-ion etching (ICP-RIE) (Fig. 1(b)). More details of the fabrication procedure have been reported elsewhere.17 The slanted nano-pillar sidewalls is made vertical using a 2% buffered KOH solution (AZ400) (Fig. 1(c)). Etching with KOH has been reported to help remove the plasma damaged sidewall and reduce the density of surface defect states.18 Fig. 2 shows the scanning-electron-microscope (SEM) image of a nano-pillar of 30 nm in diameter and 230 nm in height. Due to strain-relaxation at the sidewall, an effective confinement potential for the exciton is formed transverse to the pillar axis, which tightly confines the exciton to the center of the QD and shields it from the sidewall.19 

FIG. 2.

A 45°-view SEM image of a nano-pillar containing a site-controlled single InGaN QD in a GaN PIN junction. Two 12-nm thick unintentionally doped (uid) GaN layers are used as the quantum barrier.

FIG. 2.

A 45°-view SEM image of a nano-pillar containing a site-controlled single InGaN QD in a GaN PIN junction. Two 12-nm thick unintentionally doped (uid) GaN layers are used as the quantum barrier.

Close modal

To integrate electrical contacts, we planarize the sample with a 580 nm thick spin-on-glass (SOG) layer cured at 400 °C for an hour (Fig. 1(d)). The SOG is then etched back with CF4-CHF3 plasma to expose p-GaN for the p-contact (Fig. 1(e)). An indium-tin-oxide (ITO) conductive film is deposited by DC sputtering, patterned using HCl, and annealed at 400 °C for 5 min in the forming gas. Finally, both n- and p-contacts are metalized using Ti/Au (40 nm/500 nm) pads (Fig. 1(f)). All 90 × 90 nano-pillars in an array are connected to the same ITO contact.

We perform micro-electroluminescence (μ-EL) and micro-photoluminescence (μ-PL) on individual nano-pillars using the same optical setup as reported earlier.14,20 The sample is mounted in a temperature-stabilized continuous-flow cryostat. For EL measurements, DC is supplied to the sample by a source-meter unit. For PL measurements, excitation is created by a pulsed 370 nm laser obtained from frequency doubling a 740 nm mode-locked Ti:Sapphire laser. The excitation wavelength is chosen such that only the QD and not GaN is excited. The EL and PL emission is collected by an objective lens with a numerical aperture (NA) of 0.5 and a focal length of 4 mm. The excitation laser is removed using a spectral filter. The emission is then projected onto the confocal plane of a pair of identical lenses with a 75 mm focal length. Emission from a single QD is isolated by a pinhole on the confocal plane with a 25 μm diameter.14 The emission spectrum is recorded by a liquid nitrogen-cooled CCD camera attached to a monochromator with 0.1 nm wavelength resolution at 400 nm. The second-order correlation (g(2)(t)) function is measured using a Hanbury Brown-Twiss (HBT) interferometer consisting of a beamsplitter, two avalanche photo-detectors (APDs), and a time correlator (TC). Each of the APD-TC arm has a time resolution of 200 ps.

We first study the EL properties of a single nano-pillar. Fig. 3(a) shows EL intensity as a function of applied voltage. At around 4 V forward bias, we observe a rapid turn-on of EL, indicating the presence of both electrons and holes in the InGaN region. The EL spectra at different forward voltages above 4 V are shown in Fig. 3(b). Each spectrum is composed of a dominant zero-phonon line (ZPL) at 3.06 eV with a weak shoulder ∼20 meV on the higher energy side and an optical-phonon peak at 2.97 eV. The 90 meV optical-phonon energy is consistent with our previous results.14 The ZPL and its higher energy shoulder are most likely due to emission from single exciton (X) and one of its many negatively charged states (Xn, n = 1, 2, …).21 This is because compared to the holes in the p-GaN region, electrons in the n-GaN region have a higher concentration and mobility (∼100 cm2 V−1 s−1 for electrons22 and ∼10 cm2 V−1 s−1 for holes23). Therefore, electrons arrive at the InGaN QD at a bias Vbias < 4 V, before holes arrive at a higher voltage. The broad linewidth of the ZPL is likely due to spectral diffusion, because there was large local charge fluctuations caused by the current flow as well as a relatively large permanent dipole moment of the exciton due to a strong piezoelectric field along the growth direction.

FIG. 3.

EL properties at 10 K. (a) Integrated EL intensity vs. applied bias. (b) EL spectra of a single QD at various bias voltages. All spectra are normalized to their maximum intensity and shifted vertically for easy comparison. (c) EL intensity vs. angle of linear polarization at 5.7 V forward bias. The solid line is a fit. (d) g(2)(t) of EL at 5.7 V forward bias without background subtraction. The solid line is a fit.

FIG. 3.

EL properties at 10 K. (a) Integrated EL intensity vs. applied bias. (b) EL spectra of a single QD at various bias voltages. All spectra are normalized to their maximum intensity and shifted vertically for easy comparison. (c) EL intensity vs. angle of linear polarization at 5.7 V forward bias. The solid line is a fit. (d) g(2)(t) of EL at 5.7 V forward bias without background subtraction. The solid line is a fit.

Close modal

The EL features a high degree of linear polarization as shown in Fig. 3(c), which is taken by rotating a half-wave plate in front of a linear polarizer. The EL intensity vs. polarization angle (θ) data are fitted using the equation (I1I2)/(I1+I2)×cos2(θ+θ0)+(I2)/(I1+I2). The degree of linear polarization is obtained from (I1I2)/(I1+I2)=0.70±0.04. The polarization angle θ0 is random among different QDs and does not correspond to any specific crystal orientation. Linearly polarized emission has been observed in InGaN QDs including electrically driven ones.14,24–26 It has been attributed to the anisotropy in the InGaN lateral dimension.27 In our QDs, anisotropy can be caused by both the EBL and RIE processes. An anisotropic lateral shape leads to an asymmetric strain profile which mixes A and B exciton states, resulting in the linearly polarized emission.24,27

We confirm the single-photon nature of the EL by measuring the second-order correlation (g(2)) of the ZPL. The ZPL is selected by a spectral filter with a bandwidth of 50 meV and centered at 3.065 eV. The result is shown in Fig. 3(d) which shows clear antibunching. We fit the data with g(2)(t)=g(2)(0)+(1g(2)(0))(1exp(|t|/τ)), which yields g(2)(0)=0.42±0.05 (or g(2)(0)=0.38±0.05 after adjusting for the APD dark counts28) and τ = 4 ns. The non-zero g(2)(0) is due to higher order multi-exciton emission whose quantum efficiency is largely suppressed by the non-radiative recombination at the nano-pillar sidewall.14 The parameter τ represents the decay time of the ZPL. The slow decay time is due to the strong polarization field in the InGaN region which reduces the overlap between the electron and hole wavefunctions.29 

The current-voltage (I-V) characteristics from the nano-pillar array are shown in Fig. 4. At a small forward bias of Vbias < 2 V, the current increases exponentially (inset of Fig. 4) with increasing bias, as expected for an ideal p-i-n junction. However, as Vbias ramps above 2 V the increase in the current becomes slow and nearly linear. This suggests a non-negligible series resistance due to the non-ohmic ITO/p-GaN contact.30 Using the slope of the I-V curve at Vbias > 2 V and the diameter of the nano-pillar, we estimate the specific contact resistivity at the ITO/p-GaN contact to be 102 Ω cm2. This is consistent with reported values which range from 103 to 101 Ω cm2.31–33 

FIG. 4.

Current-voltage (I-V) characteristic of the DIN at 10 K. The upper-left inset is a semi-log plot of the I-V curve. The lower-right inset illustrates the effective circuit in which the PIN-junction and the ITO/p-GaN Schottky contact are connected in series.

FIG. 4.

Current-voltage (I-V) characteristic of the DIN at 10 K. The upper-left inset is a semi-log plot of the I-V curve. The lower-right inset illustrates the effective circuit in which the PIN-junction and the ITO/p-GaN Schottky contact are connected in series.

Close modal

Finally, we confirm the above electrical characteristics of the device with the bias-dependence of PL of the QDs. We measure the PL for Vbias=05 V with a fixed laser excitation intensity of 100 W/cm2. As shown in Fig. 5, the PL spectra red shifts linearly with increasing bias voltage for Vbias=02 V, at a rate of 10 meV/V. At Vbias > 2 V, the PL energy stabilizes at ∼3.06 eV.

FIG. 5.

PL properties at 10 K. (a) PL spectra of the DIN at different bias voltages Vbias. (b) The PL peak energy vs. Vbias.

FIG. 5.

PL properties at 10 K. (a) PL spectra of the DIN at different bias voltages Vbias. (b) The PL peak energy vs. Vbias.

Close modal

Such voltage dependence of the PL energy can be understood as follows. The PL energy is influenced by both the built-in polarization field Epol and the external electric field Epin due to the donor (acceptor) depletion in the n-GaN (p-GaN) region as illustrated in Fig. 6. Epol is determined by the spontaneous and piezoelectric polarizations in the InGaN region34 and is independent of Vbias. Epin is proportional to the depletion width in the n-GaN and p-GaN regions35 and therefore decreases as the voltage across the junction, Vpin, increases. At Vbias < 2 V, VbiasVpin. As a result, the increase in Vbias corresponds to a rapid shrinkage of the depletion width, leading to a decrease in |Epin|. Due to the quantum confined Stark effect,36,37 a stronger total electric field Et=Epol+Epin leads to a lower PL energy. Since Epol and Epin have opposite directions, the redshift observed in Fig. 5 suggests that |Epin|<|Epol|. As Vbias increases to above 2 V, the resistance across the pin junction becomes small compared to the p-contact resistance discussed earlier, and the change of Vpin versus Vbias becomes negligible. Therefore, Epin stabilizes and so does the PL energy.

FIG. 6.

Schematic energy band diagrams of a nano-pillar at Vbias = 0 V (upper panel), 0<Vbias<2V (middle panel), and Vbias > 2 V (lower panel).

FIG. 6.

Schematic energy band diagrams of a nano-pillar at Vbias = 0 V (upper panel), 0<Vbias<2V (middle panel), and Vbias > 2 V (lower panel).

Close modal

In summary, we demonstrated electrically pumped, site- and structure-controlled SPED based on III-N QDs. The QDs were fabricated via a scalable process starting from a standard blue-green InGaN LED structure. The high degree of control over the position and dimension of the QD, together with the compact device footprint, makes the demonstrated SPED suitable for on-chip integration with other electrical and optical components. Future work to reduce the contact resistance, minimize the background emission by optimizing the doping profile, and increase quantum confinement with AlGaN barrier may enabled lower g(2) and higher operating temperature.6,26

We acknowledge financial supports from the National Science Foundation (NSF) under Award No. DMR 1409529 for work related to materials properties and device design and DMR 1120923 (MRSEC) for work related to the measurements.

1.
C. H.
Bennett
and
G.
Brassard
, in
International Conference on Computers, Systems & Signal Processing
,
1984
.
2.
M. A.
Pooley
,
D. J. P.
Ellis
,
R. B.
Patel
,
A. J.
Bennett
,
K. H. A.
Chan
,
I.
Farrer
,
D. A.
Ritchie
, and
A. J.
Shields
,
Appl. Phys. Lett.
100
,
211103
(
2012
).
3.
J. L.
O'Brien
,
G. J.
Pryde
,
A. G.
White
,
T. C.
Ralph
, and
D.
Branning
,
Nature
426
,
264
(
2003
).
4.
M.
Xiao
,
L. A.
Wu
, and
H. J.
Kimble
,
Phys. Rev. Lett.
59
,
278
(
1987
).
5.
E. S.
Polzik
,
J.
Carri
, and
H. J.
Kimble
,
Phys. Rev. Lett.
68
,
3020
(
1992
).
6.
M. J.
Holmes
,
K.
Choi
,
S.
Kako
,
M.
Arita
, and
Y.
Arakawa
,
Nano Lett.
14
,
982
(
2014
).
7.
P.
Michler
,
A.
Imamoglu
,
M. D.
Mason
,
P. J.
Carson
,
G. F.
Strouse
, and
S. K.
Buratto
,
Nature
406
,
968
970
(
2000
).
8.
B.
Lounis
and
W. E.
Moerner
,
Nature
407
,
491
(
2000
).
9.
S.
Bounouar
,
M.
Elouneg-Jamroz
,
M. I.
Den Hertog
,
C.
Morchutt
,
E.
Bellet-Amalric
,
R.
André
,
C.
Bougerol
,
Y.
Genuist
,
J.-P.
Poizat
,
S.
Tatarenko
,
K.
Kheng
, and
M. D.
Hertog
,
Nano Lett.
12
,
2977
(
2012
).
10.
A. J.
Morfa
,
B. C.
Gibson
,
M.
Karg
,
T. J.
Karle
,
A. D.
Greentree
,
P.
Mulvaney
, and
S.
Tomljenovic-Hanic
,
Nano Lett.
12
,
949
(
2012
).
11.
N.
Mizuochi
,
T.
Makino
,
H.
Kato
,
D.
Takeuchi
,
M.
Ogura
,
H.
Okushi
,
M.
Nothaft
,
P.
Neumann
,
A.
Gali
,
F.
Jelezko
,
J.
Wrachtrup
, and
S.
Yamasaki
,
Nat. Photonics
6
,
299
303
(
2012
).
12.
S.
Castelletto
,
B. C.
Johnson
,
V.
Ivády
,
N.
Stavrias
,
T.
Umeda
,
A.
Gali
, and
T.
Ohshima
,
Nat. Mater.
13
,
151
156
(
2014
).
13.
S.
Deshpande
,
J.
Heo
,
A.
Das
, and
P.
Bhattacharya
,
Nat. Commun.
4
,
1675
(
2013
).
14.
L.
Zhang
,
C.-H.
Teng
,
T. A.
Hill
,
L.-K.
Lee
,
P.-C.
Ku
, and
H.
Deng
,
Appl. Phys. Lett.
103
,
192114
(
2013
).
15.
P. R.
Edwards
,
R. W.
Martin
,
I. M.
Watson
,
C.
Liu
,
R. A.
Taylor
,
J. H.
Rice
,
J. H.
Na
,
J. W.
Robinson
, and
J. D.
Smith
,
Appl. Phys. Lett.
85
,
4281
(
2004
).
16.
M. H.
Baier
,
C.
Constantin
,
E.
Pelucchi
, and
E.
Kapon
,
Appl. Phys. Lett.
84
,
1967
(
2004
).
17.
L. K.
Lee
and
P.-C.
Ku
,
Phys. Status Solidi C
9
,
609
(
2012
).
18.
Q.
Li
,
K. R.
Westlake
,
M. H.
Crawford
,
S. R.
Lee
,
D. D.
Koleske
,
J. J.
Figiel
,
K. C.
Cross
,
S.
Fathololoumi
,
Z.
Mi
, and
G. T.
Wang
,
Opt. Express
19
,
25528
(
2011
).
19.
L.
Zhang
,
T. A.
Hill
,
C.-H.
Teng
,
B.
Demory
,
P.-C.
Ku
, and
H.
Deng
,
Phys. Rev. B
90
,
245311
(
2014
).
20.
L.
Zhang
,
L.-K.
Lee
,
C.-H.
Teng
,
T. A.
Hill
,
P.-C.
Ku
, and
H.
Deng
,
Appl. Phys. Lett.
104
,
051116
(
2014
).
21.
L.
Zhang
,
C.-H.
Teng
,
P.-C.
Ku
, and
H.
Deng
,
Phys. Rev. B
93
,
085301
(
2016
).
22.
W.
Gotz
,
N. M.
Johnson
,
C.
Chen
,
H.
Liu
,
C.
Kuo
, and
W.
Imler
,
Appl. Phys. Lett.
68
,
3144
(
1996
).
23.
H.
Nakayama
,
P.
Hacke
,
M. R. H.
Khan
,
T.
Detchprohm
,
K.
Hiramatsu
, and
N.
Sawaki
,
Jpn. J. Appl. Phys., Part 2
35
,
L282
(
1996
).
24.
M.
Winkelnkemper
,
R.
Seguin
,
S.
Rodt
,
A.
Schliwa
,
L.
Reiß mann
,
A.
Strittmatter
,
A.
Hoffmann
, and
D.
Bimberg
,
J. Appl. Phys.
101
,
113708
(
2007
).
25.
C.-W.
Hsu
,
A.
Lundskog
,
K. F.
Karlsson
,
U.
Forsberg
,
E.
Janzén
, and
P. O.
Holtz
,
Nano Lett.
11
,
2415
(
2011
).
26.
S.
Deshpande
and
P.
Bhattacharya
,
Appl. Phys. Lett.
103
,
241117
(
2013
).
27.
C.-H.
Teng
,
L.
Zhang
,
T. A.
Hill
,
B.
Demory
,
H.
Deng
, and
P.-C.
Ku
,
Appl. Phys. Lett.
107
,
191105
(
2015
).
28.
R.
Brouri
,
A.
Beveratos
,
J. P.
Poizat
, and
P.
Grangier
,
Opt. Lett.
25
,
1294
(
2000
).
29.
A.
Jarjour
,
R.
Oliver
,
A.
Tahraoui
,
M.
Kappers
,
C.
Humphreys
, and
R.
Taylor
,
Phys. Rev. Lett.
99
,
197403
(
2007
).
30.
T.
Margalith
,
O.
Buchinsky
,
D. A.
Cohen
,
A. C.
Abare
,
M.
Hansen
,
S. P.
DenBaars
, and
L. A.
Coldren
,
Appl. Phys. Lett.
74
,
3930
(
1999
).
31.
D. W.
Kim
,
Y. J.
Sung
,
J. W.
Park
, and
G. Y.
Yeom
,
Thin Solid Films
398–399
,
87
(
2001
).
32.
Y.
Ding
,
W.
Guo
,
Y.
Zhu
,
J.
Liu
, and
W.
Yan
,
J. Semicond.
33
,
066004
(
2012
).
33.
J.-H.
Choi
,
S.-H.
Jang
, and
J.-S.
Jang
,
Electron. Mater. Lett.
9
,
425
(
2013
).
34.
F.
Bernardini
and
V.
Fiorentini
,
Appl. Surf. Sci.
166
,
23
(
2000
).
35.
S. M.
Sze
and
K. K.
Ng
,
Physics of Semiconductor Devices
, 3rd ed. (
John Wiley & Sons
,
Hoboken, New Jersey
,
2007
).
36.
D. A. B.
Miller
,
D. S.
Chemla
,
T. C.
Damen
,
A. C.
Gossard
,
W.
Wiegmann
,
T. H.
Wood
, and
C. A.
Burrus
,
Phys. Rev. Lett.
53
,
2173
(
1984
).
37.
S. F.
Chichibu
,
A. C.
Abare
,
M. P.
Mack
,
M. S.
Minsky
,
T.
Deguchi
,
D.
Cohen
,
P.
Kozodoy
,
S. B.
Fleischer
,
S.
Keller
,
J. S.
Speck
,
J. E.
Bowers
,
E.
Hu
,
U. K.
Mishra
,
L. A.
Coldren
,
S. P.
DenBaars
,
K.
Wada
,
T.
Sota
, and
S.
Nakamura
,
Mater. Sci. Eng., B
59
,
298
(
1999
).