LETTERS
PUBLISHED ONLINE: 13 JULY 2015 | DOI: 10.1038/NNANO.2015.136
Room-temperature single-photon generation from
solitary dopants of carbon nanotubes
Xuedan Ma, Nicolai F. Hartmann, Jon K. S. Baldwin, Stephen K. Doorn* and Han Htoon*
On-demand single-photon sources capable of operating at
room temperature and the telecom wavelength range of
1,300–1,500 nm hold the key to the realization of novel technologies that span from sub-diffraction imaging to quantum key
distribution and photonic quantum information processing1–3.
Here, we show that incorporation of undoped (6,5) singlewalled carbon nanotubes into a SiO2 matrix can lead to
the creation of solitary oxygen dopant states capable of
fluctuation-free, room-temperature single-photon emission in
the 1,100–1,300 nm wavelength range. We investigated the
effects of temperature on photoluminescence emission
efficiencies, fluctuations and decay dynamics of the dopant
states and determined the conditions most suitable for the
observation of single-photon emission. This emission can in
principle be extended to 1,500 nm by doping of smallerbandgap single-walled carbon nanotubes4,5. This easy
tunability presents a distinct advantage over existing defect
centre single-photon emitters (for example, diamond defect
centres)1–3,6. Our SiO2-encapsulated sample also presents
exciting opportunities to apply Si/SiO2-based micro/nanodevice fabrication techniques in the development of electrically
driven single-photon sources and integration of these sources
into quantum photonic devices and networks.
The intentional incorporation of impurities and defects serves as
a powerful tool for modification of the electronic and optical properties of host materials and subsequent enabling of new functionalities7. Recent studies have shown that the introduction of such new
optical transitions is also possible in single-walled carbon nanotubes
(SWCNTs) through the incorporation of low-level oxygen or aryl
diazonium covalent dopants on their side walls4,5,8. By capturing
excitons that might otherwise have recombined at quenching
centres, thus allowing efficient radiative recombination instead,
the covalent dopant states of SWCNTs dramatically enhance the
photoluminescence emission efficiency (from ∼1% to 28%) and
shift the photoluminescence of SWCNTs deeper into the near-infrared spectral regime (1.1–1.3 µm)4,5,8,9. These new transitions not
only make envisioned applications of SWCNTs in optoelectronic,
sensing and imaging technologies more feasible, but also present
potential for room-temperature single-photon generation at
telecom wavelengths8,10,11.
Due to the one-dimensional nature of SWCNTs, which allows
free diffusion of tightly bound excitons along their length12–14, in
undoped SWCNTs, single-photon generation requires localization
of band-edge excitons (E11 excitons) to quantum dot-like states15–17
(Fig. 1a). Because the localization potential results from environmental fluctuations that are only a few meV deep, single-photon
generation in SWCNTs has been observed previously only at cryogenic temperatures15–17. In contrast, optical transitions of doped
SWCNTs created via attachment of either ether-d or epoxide-l
chemical functional groups are observed to be pinned to deep
trap states located ∼130 and 300 meV below the band-edge
*
*−
and E11
transitions, respectively)4,8,10, as illustrated
(denoted as E11
in Fig. 1b. They are therefore believed to have potential as single
quantum emitters at room temperature4,8,10. Beyond enabling a
new class of quantum light sources, these dopant-induced deep
trap states of SWCNTs could present promising opportunities to
couple a localized quantum two-level system with the unique mechanical and electrical transport properties of SWCNTs17. Specifically,
they may allow the realization of proposed schemes for coupling of
single spins to nanomechanical resonators18 and optomechanical
cooling19 in suspended nanotubes, as well as all-optical manipulation
of electron spin20, all of which rely on strong exciton localization.
Despite these opportunities, the first critical step required for their
realization, namely photon correlation experiments demonstrating a
vanishing probability of two-photon emission in one excitation
cycle (also known as photon antibunching), have not yet been
reported for emission from the dopant-induced deep trap states.
The realization of room-temperature single-photon emission
from such dopant sites requires the removal of several key experimental barriers. Currently, both oxygen and diazonium doping
are achieved through prolonged exposure of SWCNTs to doping
agents in aqueous suspension4,5. Although doped SWCNTs emit
reasonably stable photoluminescence in their native solution
environment4, the photoluminescence signal shows rapid fluctuations
both in intensity and wavelength when placed on solid-state substrates10. The photoluminescence also bleaches over a 10–30 min
timescale when the tubes are subjected to intense laser excitation10.
These fluctuations and bleaching, as well as a lack of low darkcount single-photon detectors for the near-infrared spectral
regime in which the dopant emission occurs, remain the major technical challenges in carrying out the second-order photon correlation
experiment necessary for direct proof of single-photon emission.
As a solution to this issue, here we demonstrate that electronbeam deposition of SiO2 on surfactant-wrapped, undoped (6,5)
SWCNTs (Fig. 2a) leads to the creation of isolated oxygen dopant
states capable of emitting redshifted photoluminescence free of fluctuations both in intensity and spectral position (Fig. 2b). Figure 2c,d
compares photoluminescence images of SWCNTs taken before and
after deposition of the SiO2 layer. Figure 2c (the ‘before’ image)
shows SWCNTs emitting only in a 972–1,028 nm wavelength
range, as is typical for undoped (6,5) E11 emission, but Fig. 2d
(the ‘after’ image) shows that more than 95% of the tubes now
emit photoluminescence at wavelengths longer than 1,045 nm. A
careful inspection of the low-temperature photoluminescence
spectra of individual SiO2-encapsulated tubes and their distribution
histogram further reveals that they exhibit all the key spectral
features observed in the low-temperature photoluminescence
spectra of individual oxygen-doped SWCNTs created using the
solution-based approach (Supplementary Fig. 1a–d), as recently
reported10. Specifically, the histogram for SiO2-encapsulated tubes
Center for Integrated Nanotechnologies, Materials Physics and Applications Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545,
USA. * e-mail: [email protected]; [email protected]
NATURE NANOTECHNOLOGY | VOL 10 | AUGUST 2015 | www.nature.com/naturenanotechnology
© 2015 Macmillan Publishers Limited. All rights reserved
671
LETTERS
NATURE NANOTECHNOLOGY
a
DOI: 10.1038/NNANO.2015.136
b
E11
+ −
kBT < ΔE
at 4 K
+ −
+ −
Ether-d
Epoxide-l
E11
130 meV
E11
+ −
+
−
E*11
300 meV
kBT > ΔE
at RT
E*11−
a
b
Au
Si
1,150
Intensity (a.u.)
SiO2
Wavelength (nm)
Figure 1 | Mechanism of photon antibunching from undoped tubes at cryogenic temperatures and exciton localization in oxygen-doped tubes at room
temperature. a, Upper panel: schematic representation of single-photon generation from undoped tubes at cryogenic temperatures. Photon antibunching at
these temperatures becomes possible due to localization of excitons in local environment-induced zero-dimensional trap states (red dotted line) that locate a
few meV (ΔE, green arrow) below the E11 state. Lower panel: photon antibunching from undoped tubes disappears at room temperature (RT) because
thermal energy kBT overcomes the trapping energy ΔE. b, Introduction of ether-d and epoxide-l groups (l/d: C–O–C bond aligned parallel/perpendicular to
the tube axis) to the side walls of (6,5) SWCNTs leads to the creation of deep trap states located 130 meV (E11* ) and 300 meV (E11*− ) below the E11 state,
respectively. These trap states strongly localize excitons to 3–4 nm regions of the dopant sites10. As these trapping energies are much larger than the kBT
value at room temperature, we expect the localization to be conserved up to room temperature.
1,100
1,050
150
300
450
600
Time (s)
c
975−1,025 nm
E11
15
> 1,045 nm
Undoped
1,500
Count
4,000
10
5
1,200
800
0
5 μm
1,050
1,200
1,350
5 μm
Wavelength (nm)
d
975−1,025 nm
E*11
Count
2,000
1,000
5 μm
20
E11
> 1,045 nm
Doped
6,000
E*11−
1,200
0
1,050
1,200
1,350
5 μm
Wavelength (nm)
Figure 2 | Schematic of SiO2 matrix-incorporated SWCNTs and their optical properties. a, Schematic representation of SWCNTs incorporated into a SiO2
matrix. b, Temporal evolution (false colour) of photoluminescence spectra from an individual SWCNT embedded in a SiO2 matrix at 4 K. Green curve:
temporal behaviour of the integrated photoluminescence intensity from the tube. c,d, Left and right panels: wide-field photoluminescence images of SWCNTs
before (c) and after (d) incorporation into the SiO2 matrix in the wavelength ranges of 975–1,025 nm (left panels) and >1,045 nm (right panels).
Photoluminescence images of all the undoped SWCNTs appear in the left panel of c, and only dark counts can be observed in the right panel of c.
Photoluminescence images of tubes incorporated into a SiO2 matrix appear in the right panel of d, but only a single spot (marked by a white dashed circle)
can be observed in the left panel of d. Middle panels: photoluminescence peak distribution histograms of undoped tubes (c) and tubes incorporated into a
SiO2 matrix (d), fitted with Gaussian functions. The yellow (975–1,025 nm) and blue (>1,045 nm) shading indicate the wavelength ranges over which the
wide-field images were taken.
672
NATURE NANOTECHNOLOGY | VOL 10 | AUGUST 2015 | www.nature.com/naturenanotechnology
© 2015 Macmillan Publishers Limited. All rights reserved
NATURE NANOTECHNOLOGY
Counts
30
e
g2(0) = 0.05
150 K
15
Counts
a
LETTERS
DOI: 10.1038/NNANO.2015.136
20
10
0
g2(0) = 0.32
298 K
10
5
0
−33
−22
−11
11
0
22
33
−33
−22
−11
f
1,108 nm
1,000
1,100
1,200
1,300
1,400
1,050
Q = 1.4
0
0
2,000
0
4,500
1
2
3
4
1,350
1,000 2,000 3,000 4,000 0
Time (s)
h
τ = 573 ps
0
33
Q = 11.7
3
2
1
0
Counts/bin
Normalized
intensity (a.u.)
Time (s)
d
Intensity (kHz)
2
1,000
1,200
5
τ = 243 ps
0
6
5,000
Counts/bin
Normalized
intensity (a.u.)
Intensity (kHz)
g
0
22
Wavelength (nm)
3
1
11
1,272 nm
Wavelength (nm)
c
0
Time delay (ns)
Intensity (a.u.)
b
Intensity (a.u.)
Time delay (ns)
1
Time (ns)
2
3
4
5
6
Time (ns)
Figure 3 | Photon antibunching and carrier dynamic properties of SWCNTs embedded in a SiO2 matrix. a–h, g (2) functions (a,e), photoluminescence
spectra (b,f), time traces with corresponding Mandel Q values24 (c,g) and decay curves (d,h) of two individual SWCNTs measured at 150 K (a–d) and
298 K (e–h), respectively. Count rate histograms with a binning time of 100 ms are plotted on the right of each corresponding time trace (c,g). The
histograms are fitted with Poisson functions (red curves). The yellow area in g represents a section of the time trace showing stable photoluminescence
emission. Photoluminescence decay curves (d,h, black) are fitted with single exponential functions (red) and the lifetimes are marked on the side.
The instrument response function is also shown in d (grey curve).
(Fig. 2d) shows two groups of peaks centred at ∼1,120 nm and
*
) and
1,250 nm—exactly the same positions as the ether-d (E11
*−
epoxide-l ((E11 ) emission bands of solution-processed oxygendoped tubes (Supplementary Fig. 1f )10. Based on these results, we
conclude that low concentrations of oxygen dopants are introduced
to the sidewalls of SWCNTs during electron-beam deposition of the
top SiO2 layer. This indirect doping probably results from reaction
of the tubes with oxygen radicals produced in the dissociation of
SiO2 into mixed oxides (that is, SiOx) during electron-beam
evaporation21. Surprisingly, a significant fraction of these matrixencapsulated dopant states are found to emit strong photoluminescence free of fluctuations both in intensity and energy for
long periods (100s of minutes) under continuous intense laser
irradiation (840 nm, 24.0 kW cm–2, Fig. 2b and Supplementary
Fig. 2). This remarkable photoluminescence stability decidedly
enables exploration of the quantum optical properties of individual
SWCNT dopant states, which was not possible previously.
We performed time-tagged time-correlated single-photon counting and Hanbury Brown–Twiss (HBT) experiments on over 180
individual dopant emission peaks at sample temperatures from
4 K to 298 K. Figure 3 presents second-order photon correlation
( g (2)) traces (Fig. 3a,e), photoluminescence spectra (Fig. 3b,f ),
photoluminescence time traces (Fig. 3c,g) and photoluminescence
decay curves (Fig. 3d,h) of two representative individual doped
SWCNTs acquired at 150 and 298 K, respectively. Additional data
sets representing measurements at 4, 50, 100, 200, 230 and 260 K
are provided in Supplementary Fig. 2. Low-temperature (4 K and
50 K) photoluminescence spectra of individual dopant states
(Supplementary Figs 1a,c and 2b,f ) show sharp asymmetric peaks
resulting from strong exciton localization10. At higher temperatures,
the photoluminescence peaks become symmetric due to thermal
broadening (Fig. 3b,f and Supplementary Fig. 2). In some cases
more than one deep trap emission peak appears due to the presence
of multiple dopants in the excitation spot (Supplementary Fig. 4b),
but the appearance of single isolated peaks in most photoluminescence
spectra indicates that dopants remain solitary for a majority of tubes.
The g (2) traces provide a measure of the probability of two successive photon detection events as a function of time delay between the
two detection events. For excitation with a pulse train, disappearance
of the peak of a g (2) trace at zero time delay indicates photon antibunching—a vanishing probability for the emission of two or more
photons per excitation cycle—and hence is regarded as the defining
characteristic of a single-photon emitter. The g (2) traces of Fig. 3a,e
and Supplementary Fig. 2 show clear photon antibunching, with
the areas of the centre peaks significantly smaller than those of the
side peaks. The g (2) traces plotted for timescales up to 330 ns, on
the other hand, show slight photon-bunching of the side peaks,
apparent as a relative increase in side-peak areas observed at short
time delays (<100 ns) in comparison to those appearing at longer
delays (Supplementary Fig. 3). This slight bunching indicates the
existence of submicrosecond-scale photoluminescence intensity fluc*
*−
and E11
dopant
tuations16,17. Despite this bunching, we can find E11
(2)
peaks with a degree of photon antibunching ( g (0), defined as the
area of the centre peak normalized to that of the side peaks at
delay >200 ns) of less than 0.1 for temperatures up to 230 K (Figs 3
and 4b (red data points) and Supplementary Fig. 2). A g (2)(0)
value as low as 0.32, a value comparable to reports of other singlephoton emission22,23, was also observed at room temperature (Fig. 3e).
NATURE NANOTECHNOLOGY | VOL 10 | AUGUST 2015 | www.nature.com/naturenanotechnology
© 2015 Macmillan Publishers Limited. All rights reserved
673
LETTERS
NATURE NANOTECHNOLOGY
a
b
Antibunching probability (%)
16
8
60
50
0.8
40
0.6
30
0.4
20
0.2
10
0.0
0
0
50
100
150
200
250
0
300
50
100
150
200 250 300
Temperature (K)
Temperature (K)
c
d 600
500
2,000
Lifetime (ps)
Photoluminescence
intensity (Hz)
g2 (0)
Mandel Q
24
DOI: 10.1038/NNANO.2015.136
400
300
200
1,000
100
0
50
100
150
200
250
300
0
50
100
150
200
250
300
Temperature (K)
Temperature (K)
Figure 4 | Effects of temperature on photoluminescence emission properties of individual dopant states and probability of observing single-photon
generation. a–d, Temperature dependences of averaged Mandel Q parameter24 values of all the investigated individual dopant states (black data points) and
their distributions (black error bars) and those of the dopant states showing photon antibunching (red data points and error bars), respectively (a), percentages
of sampled nanotubes showing photon antibunching (that is, g (2)(0) < 0.5) (b, black data squares), lowest g (2) observed at each temperature (b, red circles),
averaged photoluminescence intensity (c) and lifetime (d) of all individual dopant states investigated (black squares) and those showing photon antibunching
(red circles). The nanotubes that show photon antibunching consistently have higher photoluminescence intensity and longer photoluminescence lifetimes
than the average values.
The photon-bunching of side peaks indicates the existence of
submicrosecond-scale photoluminescence fluctuations, but photoluminescence time traces acquired simultaneously with the g (2)
traces (for example, Fig. 3 and Supplementary Fig. 2) show that
most of the dopant states (up to 230 K) exhibit photoluminescence
free of the long-timescale (>0.1 s) fluctuations and photobleaching
that could render single-photon emission unreliable. Above this
temperature, the population of dopant states showing long-timescale fluctuations and bleaching after prolonged exposure (>1 h)
to laser irradiation (for example, Fig. 3g) increases significantly.
To further quantify these long-timescale fluctuations, we calculate
the Mandel Q parameter, which provides a measure of the deviation
of the photoluminescence intensity distribution from the Poissonian
distribution of detector shot noise (that is, Q = (〈n 2〉 – 〈n〉2)/〈n〉,
where n represents the number of photoluminescence counts
measured within 0.1 s)24 for each photoluminescence time trace.
Photoluminescence time traces of the dopant states acquired at
temperatures below 230 K are characterized by Q values less than
5.0 (Figs 3c, 4a and Supplementary Fig. 2), indicating that photoluminescence emission is essentially free of long-timescale fluctuations
(Q = 1 is the shot noise limit). Q values for time traces acquired at
higher temperatures, however, increase to values larger than 5.0
(for example, Fig. 3g and Supplementary Fig. 2w) due to the intensity fluctuations and bleaching developed under prolonged laser
excitation. Even in such cases, sections of photoluminescence time
traces that are thousands of seconds long with a Q value less than
5.0 can be readily found, as marked in Fig. 3g (yellow region).
These data thus demonstrate single-photon generation paired with
good overall optical characteristics.
At 4 K, of the 33 doped tubes investigated, six were observed to
exhibit antibunching with g (2) values of 0.04–0.39. The existence of
multiple photoluminescence emission peaks (Supplementary Fig. 4b),
674
rapid photoluminescence fluctuation/bleaching (Supplementary
Fig. 4c,d) and relatively low photoluminescence intensities prevented
observation of photon antibunching in other tubes at such low
temperatures. The probability of observing states capable of
single-photon emission increases with temperature, reaching a
peak value of 55% at 200 K (Fig. 4b, black data points). A further
increase in temperature leads to a decrease in the probability to
10% at 298 K. Not surprisingly, this trend closely follows the temperature dependence of the average photoluminescence intensity of
the dopant states, which also peaks at 200 K (Fig. 4c, black
squares). A similar temperature-dependent intensity trend was
reported in the ensemble photoluminescence study of Miyauchi
and co-authors8. This behaviour is explained by a model considering exciton diffusional transport and reduction of the bright exciton
population of the dopant sites due to the existence of a dark state
also associated with the dopants that exists 15 meV below the
bright exciton. Because the sensitivity of our experiment is just sufficient to detect the emission of solitary dopant states, a reduction of
photoluminescence intensity leads directly to a decrease in the probability of observing the solitary dopant states shown in Fig. 4b.
Although the existence of the dark state is undesirable in this
respect, it may open a route for manipulating the spin and/or
orbital angular momentum of trapped excitons16,18,20. We note that
the decrease in probability as temperature is increased above 200 K
is a consequence of an increase in the occurrence of intensity fluctuations and bleaching, as mirrored by the averaged Q values of individual dopant states shown in Fig. 4a. Finally, the data in Fig. 3a–d and
the observation that the photoluminescence time traces show very
little fluctuation (that is, Q < 3.0, Fig. 4a) at 150–200 K demonstrate
that this temperature range may be a suitable point for on-demand
single-photon generation. This temperature can be readily reached
through the use of commercial thermoelectric coolers. On the other
NATURE NANOTECHNOLOGY | VOL 10 | AUGUST 2015 | www.nature.com/naturenanotechnology
© 2015 Macmillan Publishers Limited. All rights reserved
NATURE NANOTECHNOLOGY
DOI: 10.1038/NNANO.2015.136
LETTERS
hand, our observation of room-temperature photon antibunching
indicates that 200 K is not an intrinsic limit.
*
and
Photoluminescence decay curves measured at 4 K for both E11
*−
peaks (for example, Supplementary Fig. 2d) show very similar
E11
single exponential decays. Their lifetimes are distributed in the range
200–800 ps with an average of 431 ± 28 ps, and show no correlation
with emission wavelength (Supplementary Fig. 5). The distribution
of photoluminescence lifetimes shows little dependence on sample
temperature up to 150 K (Fig. 4d and Supplementary Fig. 5).
Photoluminescence lifetime, however, decreases with a further increase
in sample temperature. The transition point of the lifetime-temperature dependence at ∼200 K roughly coincides with the point where
the photoluminescence intensity reaches a maximum (Fig. 4c,d).
This trend may be explained as another consequence of the dark
state that is suggested to exist below the lowest emissive state.
Although the average dopant state lifetime decreases to 120 ± 13 ps
at room temperature, the photoluminescence decays of some dopant
states remain predominantly single-exponential at low pump
fluences and display lifetimes as long as 373 ps (Supplementary
Fig. 5). We also observe that the dopant states that show photon
antibunching consistently have longer photoluminescence lifetimes
than the average values (red circles in Fig. 4d and red crosses in
Supplementary Fig. 5). In contrast, the photoluminescence lifetimes
of E11 band-edge excitons of undoped tubes12,25,26 as well as those
measured at wavelengths shorter than 1,025 nm in this study
(Supplementary Figs 5 and 6) are in the range of 10–50 ps, an
order of magnitude smaller than those of the dopant states. Such
short lifetimes are attributed to encounters between freely diffusing
excitons and defect sites that allow efficient non-radiative recombination12,27. The dramatic (approximately an order of magnitude)
increase in the photoluminescence lifetime of the dopant states provides direct evidence that non-radiative recombination is inhibited
as a direct consequence of strong exciton pinning. The lifetime
thus may be approaching the intrinsic radiative recombination
time26. This system may therefore support the long, radiativedecay-limited lifetimes necessary for the generation of indistinguishable single photons and the operation of quantum devices16,17.
These results together establish solitary oxygen dopant states of
SWCNTs as a viable system for the development of single-photon
sources and integrated photonic networks. In particular, encapsulation of the doped SWCNTs in a SiO2 matrix opens a new route to
the application of well-developed Si/SiO2-based micro-/nanodevice
fabrication techniques in their integration into single-tube fieldeffect transistor structures28,29 or photonic/dielectric metamaterial
waveguides/cavities30,31. Beyond that, our results will motivate
more theoretical/experimental efforts in exploring new types of
covalent dopants with quantum-optical properties capable of
enabling other quantum information-processing functionalities.
6. Castelletto, S. et al. A silicon carbide room-temperature single-photon source.
Nature Mater. 13, 151–156 (2014).
7. Koenraad, P. M. & Flatte, M. E. Single dopants in semiconductors. Nature Mater.
10, 91–100 (2011).
8. Miyauchi, Y. et al. Brightening of excitons in carbon nanotubes on
dimensionality modification. Nature Photon. 7, 715–719 (2013).
9. Wang, Q. H. & Strano, M. S. Carbon nanotubes: a bright future for defects.
Nature Chem. 5, 812–813 (2013).
10. Ma, X. et al. Electronic structure and chemical nature of oxygen dopant states in
carbon nanotubes. ACS Nano 8, 10782–10789 (2014).
11. Iwamura, M. et al. Nonlinear photoluminescence spectroscopy of carbon
nanotubes with localized exciton states. ACS Nano 8, 11254–11260 (2014).
12. Crochet, J. J. et al. Disorder limited exciton transport in colloidal single-wall
carbon nanotubes. Nano Lett. 12, 5091–5096 (2012).
13. Cognet, L. et al. Stepwise quenching of exciton fluorescence in carbon nanotubes
by single-molecule reactions. Science 316, 1465–1468 (2007).
14. Wang, F., Dukovic, G., Knoesel, E., Brus, L. E. & Heinz, T. F. Observation of
rapid Auger recombination in optically excited semiconducting carbon
nanotubes. Phys. Rev. B 70, 241403(R) (2004).
15. Hoegele, A., Galland, C., Winger, M. & Imamoglu, A. Photon antibunching in
the photoluminescence spectra of a single carbon nanotube. Phys. Rev. Lett. 100,
217401 (2008).
16. Walden-Newman, W., Sarpkaya, I. & Strauf, S. Quantum light signatures and
nanosecond spectral diffusion from cavity-embedded carbon nanotubes.
Nano Lett. 12, 1934–1941 (2012).
17. Hofmann, M. S. et al. Bright, long-lived and coherent excitons in carbon
nanotube quantum dots. Nature Nanotech. 8, 502–505 (2013).
18. Pályi, A., Struck, P. R., Rudner, M., Flensberg, K. & Burkard, G. Spin–orbitinduced strong coupling of a single spin to a nanomechanical resonator.
Phys. Rev. Lett. 108, 206811 (2012).
19. Wilson-Rae, I., Galland, C., Zwerger, W. & Imamoglu, A. Exciton-assisted
optomechanics with suspended carbon nanotubes. New J. Phys. 14, 115003 (2012).
20. Galland, C. & Imamoğlu, A. All-optical manipulation of electron spins in
carbon-nanotube quantum dots. Phys. Rev. Lett. 101, 157404 (2008).
21. Ohring, M. Materials Science of Thin Films (Academic, 2001).
22. Hausmann, B. J. M. et al. Integrated diamond networks for quantum
nanophotonics. Nano Lett. 12, 1578–1582 (2012).
23. Mizuochi, N. et al. Electrically driven single-photon source at room temperature
in diamond. Nature Photon. 6, 299–303 (2012).
24. Margolin, G., Protasenko, V., Kuno, M. & Barkai, E. Photon counting statistics
for blinking CdSe–ZnS quantum dots: a Levy walk process. J. Phys. Chem. B 110,
19053–19060 (2006).
25. Berciaud, S., Cognet, L. & Lounis, B. Luminescence decay and the absorption
cross section of individual single-walled carbon nanotubes. Phys. Rev. Lett. 101,
077402 (2008).
26. Wang, F., Dukovic, G., Brus, L. E. & Heinz, T. F. Time-resolved fluorescence of
carbon nanotubes and its implication for radiative lifetimes. Phys. Rev. Lett. 92,
177401 (2004).
27. Gokus, T. et al. Mono- and biexponential luminescence decays of individual
single-walled carbon nanotubes. J. Phys. Chem. C 114, 14025 (2010).
28. Avouris, P. & Chen, J. Nanotube electronics and optoelectronics. Mater. Today 9,
46–54 (2006).
29. Mueller, T. et al. Efficient narrow-band light emission from a single carbon
nanotube p–n diode. Nature Nanotech. 5, 27–31 (2010).
30. Sun, S. et al. Gradient-index meta-surfaces as a bridge linking propagating waves
and surface waves. Naure. Mater. 11, 426–431 (2012).
31. Poddubny, A., Iorsh, I., Belov, P. & Kivshar, Y. Hyperbolic metamaterials. Nature
Photon. 7, 948–957 (2013).
Methods
Acknowledgements
Methods and any associated references are available in the online
version of the paper.
This work was conducted at the Center for Integrated Nanotechnologies, a US Department
of Energy, Office of Basic Energy Sciences user facility, and supported by Los Alamos
National Laboratory (LANL) Directed Research and Development Funds.
Received 28 January 2015; accepted 2 June 2015;
published online 13 July 2015
Author contributions
References
1. Acosta, V. & Hemmer, P. Nitrogen-vacancy centers: physics and applications.
MRS Bull. 38, 127–133 (2013).
2. Aharonovich, I. et al. Diamond-based single-photon emitters. Rep. Prog. Phys.
74, 076501 (2011).
3. Gordon, L. et al. Quantum computing with defects. MRS Bull. 38,
802–807 (2013).
4. Ghosh, S., Bachilo, S. M., Simonette, R. A., Beckingham, K. M. & Weisman, R. B.
Oxygen doping modifies near-infrared band gaps in fluorescent single-walled
carbon nanotubes. Science 330, 1656–1659 (2010).
5. Piao, Y. M. et al. Brightening of carbon nanotube photoluminescence through
the incorporation of sp 3 defects. Nature Chem. 5, 840–845 (2013).
H.H., S.K.D. and X.M. conceived and designed the experiment. X.M., under the supervision
of H.H., performed all spectroscopy studies and data analysis. N.F.H., under the supervision
of S.K.D., performed carbon nanotube separation chemistry. J.K.S.B and X.M. performed
electron-beam deposition of SiO2. N.F.H. and S.K.D. assisted in analysis and interpretation
of the results. X.M. and H.H. prepared the manuscript with assistance from all other
co-authors.
Additional information
Supplementary information is available in the online version of the paper. Reprints and
permissions information is available online at www.nature.com/reprints. Correspondence and
requests for materials should be addressed to S.K.D. and H.H.
Competing financial interests
The authors declare no competing financial interests.
NATURE NANOTECHNOLOGY | VOL 10 | AUGUST 2015 | www.nature.com/naturenanotechnology
© 2015 Macmillan Publishers Limited. All rights reserved
675
LETTERS
NATURE NANOTECHNOLOGY
Methods
Sample preparation. Chirality-enriched (6,5) SWCNTs wrapped with 1 wt%
sodium dodecylbenzene sulphonate were prepared in a two-step aqueous two-phase
extraction32–34 with non-sonicated SWCNT dispersions used as the starting
material35. SWCNTs were drop-cast on Si or quartz substrates over-coated with
100 nm Au and 100 nm electron-beam-deposited SiO2 layers. Au and SiO2 layers
were added to enhance the photon collection efficiency and to suppress
metal-induced photoluminescence quenching, respectively. A 10-nm-thick top layer
of SiO2 was then deposited on the SWCNTs by electron-beam deposition. Both top
and bottom SiO2 layers were deposited at a rate of 0.2 nm s–1 using 99.99% pure SiO2
target (CERAC). The base pressure of the chamber was 2.8 × 10−8 torr and the
substrate was maintained at room temperature.
Single SWCNT optical spectroscopy. A home-built micro-photoluminescence
system was used to perform all spectroscopy experiments. SWCNT samples were
loaded into a continuous-flow liquid He cryostat (Oxford Instruments) and
excited resonantly at their phonon side band of the E11 band-edge exciton at
840 nm with a pulsed laser (150 fs, 90 MHz) at 5–20 µW average power. An infrared
objective (Olympus) with NA = 0.65, ×50 magnification was used to confocally
excite and collect the photoluminescence signal. Photoluminescence images
and spectra were taken with a two-dimensional InGaAs array camera and
one-dimensional InGaAs liner array detector mounted on a 150 mm spectrograph,
DOI: 10.1038/NNANO.2015.136
respectively. For time-correlated photon counting and HBT experiments, the
photoluminescence signal was coupled into an optical fibre and a 1:1 optical
fibre beamsplitter was used to split the signal into the two channels of a
superconducting nanowire single-photon detector (Single Quantum Eos 210).
Photoluminescence time traces, photoluminescence decay curves and g (2)
functions were extracted from macro- and micro-times of photon detection events
recorded using HydraHarp (PicoQuant) time-correlated photoncounting electronics.
References
32. Khripin, C. Y., Fagan, J. A. & Zheng, M. Spontaneous partition of carbon
nanotubes in polymer-modified aqueous phases. J. Am. Chem. Soc. 135,
6822–6825 (2013).
33. Fagan, J. A. et al. Isolation of specific small-diameter single-wall carbon
nanotube species via aqueous two-phase extraction. Adv. Mater. 26,
2800–2804 (2014).
34. Subbaiyan, N. K., Cambré, S., Parra-Vasquez, A. N., Doorn, S. K. & Duque, J. G.
Role of surfactants and salt in aqueous two-phase separation of carbon
nanotubes toward simple chirality isolation. ACS Nano 25, 1619–1628 (2014).
35. Subbaiyan, N. K. et al. Benchtop aqueous two-phase extraction of isolated
individual single-walled carbon nanotubes. Nano Res. 8, 1755–1769 (2015).
NATURE NANOTECHNOLOGY | www.nature.com/naturenanotechnology
© 2015 Macmillan Publishers Limited. All rights reserved