Infrared Quantum Dots

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

PROGRESS REPORT

Infrared Quantum Dots**


By Edward H. Sargent*

Colloidal nanocrystals are quantum-size-effect tunable; offer an abundance of


available surface area for electronic and chemical interactions; and are processible
from organic or aqueous solution onto substrates rigid or flexible, smooth or rough,
flat or curved, inorganic or organic (including biological), crystalline or amorphous,
conducting, semiconducting, or insulating. With the benefit of over a decade's progress
in visible-light-emitting colloidal-quantum-dot synthesis, physical chemistry, and
devices, significant progress has recently been made in infrared-active colloidal quantum dots and devices. This
progress report summarizes the state-of-the-art in infrared colloidal quantum dots, with an emphasis on applications
and devices. The applications of interest surveyed include monolithic integration of fiber-optic and free-space-
communications photonic components with electronic substrates such as silicon and glass; in-vivo biological tagging
in infrared spectral bands in which living tissue is optically penetrable to a depth of 5±10 cm; solar and thermal
photovoltaics for energy conversion; and infrared sensing and imaging based on non-visible, including thermal,
signatures. The synthesis and properties of quantum dots are first reviewed: photoluminescence quantum efficiencies
greater than 50 % are achievable in solution, and stable luminescent dots are available in organic and aqueous
solvents. Electroluminescent devices based on solution processing have been reported with external quantum
efficiencies approaching 1 %. Photoconductive devices have been realized with 3 % internal quantum efficiencies,
and a photovoltaic effect was recently observed. Electro-optic modulation achieved by either field- or charge-induced
modification of the rate of optical absorption has been demonstrated based both on interband and intersubband
(intraband) transitions. Optical gain from these processible materials with a threshold of 1 mJ cm±2 and an optical net
modal gain coefficient of 260 ± 20 cm±1 have been reported.

1. The Need for Processible, Tunable Infrared-


Active Optoelectronic Materials

While humans see only in the visible, our ability to communi-


± cate information, detect and treat disease, harness new forms
[*] Prof. E. H. Sargent[+] of energy, and visualize threats to our safety and environment
Nanotechnology and Photonics
MIT Microphotonics Laboratory
depend increasingly on mastery of the infrared spectral region.
Massachusetts Institute of Technology Fiber-optic communications systems rely on the low-loss
Cambridge, MA 02139-4307 (USA) propagation of infrared optical signals inside pure glass fibers.
E-mail: [email protected]
An extended definition of the telecommunications wave-
[+] Permanent address: Nortel NetworksÐCanada Research Chair in
Emerging Technologies, Electrical & Computer Engineering, Univer-
length band spans 1200±1700 nm.[1] It is of interest to unite, in
sity of Toronto, Toronto, ON, M5S 1A1, Canada. a single monolithic fabrication process, the technologies which
[**] The author acknowledges the contributions of Ahmed Maria, underlie optical communication, wireless, and wireline com-
Ethan Klem, Fred Chang, Gerasimos Konstantatos, Ian Howard, munications. Lattice-matched compound semiconductor epi-
Joe Salfi, Larissa Levina, Paul Cyr, Sam Cauchi, Scott Kuntze, Shi-
guo Zhang, Sjoerd Hoogland, Steve McDonald, and Vlad Sukho-
taxy used in optoelectronics conflicts with convenient mono-
vatkin. Research presented herein was supported by Nortel Net- lithic integration. In contrast, solution-processible active
works; the Ontario Centres of Excellence through Materials and optoelectronic materials which provide a comprehensive set
Manufacturing Ontario, Photonics Research Ontario, Communica-
of canonical optoelectronic functionsÐlight emission, detec-
tions and Information Technology Ontario; the Ontario Research
and Development Challenge Fund and the Ontario Innovation Trust; tion, and modulation with customizable spectral responseÐ
the Natural Sciences and Engineering Research Council of Canada facilitate the physical convergence of a diverse array of com-
through its Discovery Grant program, Collaborative Research and munications and computing technologies. Integration of ac-
Development program, and Nano Innovation Platform; the Canada
Foundation for Innovation; and the Canada Research Chairs Pro- tive optoelectronics on a silicon chip is relevant to on-chip
gramme. and chip-to-chip interconnection and line-of-sight free-space

Adv. Mater. 2005, 17, No. 5, March 8 DOI: 10.1002/adma.200401552  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 515
E. H. Sargent/Infrared Quantum Dots
PROGRESS REPORT

communications, both for data communications and reliable photovoltaic cells that respond to infrared radiation from a
friend/foe identification.[2] fuel-fired emitter. Cogenerators of electricity and heat are
Infrared optical imaging of living tissue is an area of grow- proposed to be quiet, reliable, clean, and efficient, meeting
ing interest. In the visible spectral region, functionally specific the needs for remote and mobile applications.[7] A 1 cm2 sili-
in-vivo cancer targeting and imaging have recently been dem- con cell in direct sunlight will generate about 0.01 W, but an
onstrated using semiconductor quantum dots. Systemic injec- efficient infrared photovoltaic cell of equal size could produce
tion of multifunctional quantum-dot probes enabled sensitive 1 W in a fuel-fired system.
and multicolored fluorescence imaging of cancer cells in Sensing and imaging already rely on the infrared spectral
vivo.[3] Using visible fluorescent tags, however, deep organs region. Were large-area, low-cost infrared sensitization of ex-
such as the liver and spleen could not be detected in mice isting sensing and focal-plane-array technologies possible, the
because of the limited penetration depth of visible light. infrared could go from niche to ubiquity. The infrared spectral
Deep-tissue imaging requires the use of infrared light within a region manifests the thermal signatures of warm objects
spectral window separated from the major absorption peaks against a cold background and enables night vision based
of hemoglobin and water. The use of near-infrared-emitting on infrared illumination. It can reveal spectral signatures of
quantum dots is estimated to improve the tumor-imaging sen- chemical species of interest in environmental monitoring and
sitivity at least tenfold, allowing sensitive detection of 10±100 antiterrorism (biological and chemical threats): mid-infrared
cancer cells. High-transparency spectral bands centered at light (2.5±50 lm, 4000±200 cm±1) excites characteristic molec-
840 nm, 1110 nm, 1320 nm, and 1680 nm[4] enable depths of ular vibrations of particular utility that identify organic and
detection of 5±10 cm,[5] a capability which recently provided organometallic molecules. Collision-avoidance sensors in the
surgeons with direct infrared visual guidance throughout a infrared exploit atmospheric-condition independence of select
sentinel-lymph-node mapping procedure, minimizing incision infrared wavelengths. Light curtains emit infrared light beams
and dissection inaccuracies and permitting real-time confir- in front of a hazardous area being protected: a stop signal
mation of complete resection. is sent to the guarded machine when any of the beams are
Large-area, physically flexible photovoltaic sheets have re- blocked. Applications include perimeter guarding for indus-
cently received increasing attention as a low-cost, manufactur- trial robots and machinery and point-of-access guarding for
able, readily installed means of harnessing solar energy. While automated-machine assemblies.
efforts have been primarily focused on the visible spectral This broad set of infrared-based applications point to the
region, infrared photovoltaics are also attracting attention. If following set of requirements:
the sun's spectrum is divided into its visible and infrared parts, d A materials system that, rather than requiring lattice-
the sun's intensity of approximately 970 W m±2 includes ap- matched epitaxial growth on perfect crystalline substrates,
proximately 490 W m±2 of visible intensity (wavelengths is compatible with solution processing and large-area
< 0.7 lm) and 480 W m±2 of infrared intensity. The visible por- devices, allowing the realization of devices on physically
tion of the spectrum could be used for direct lighting, and the flexible substrates;
infrared portion harnessed for electrical power, arguing for in- d A materials system that, through straightforward variation
frared photovoltaics.[6] Proponents of thermal photovoltaics of synthetic parameters, allows controllable access to dif-
propose generating electricity from combustion energy using ferent portions of the broad infrared spectrum.

Prof. Sargent is the 2004±2005 Visiting Professor of Nanotechnology and Photonics in the
Microphotonics Center at the Massachusetts Institute of Technology. He holds the Nortel
Networks±Canada Research Chair in Emerging Technologies in the Edward S. Rogers Sr.
Department of Electrical & Computer Engineering at the University of Toronto.
Prof. Edward (Ted) H. Sargent was in 2003 named one of the world's top young innovators by
MIT's Technology Review magazine. He is author of well over one hundred papers published in
international refereed journals and presented at international conferences. Since 2003 he has given
invited lectures at Harvard, MIT, Caltech, Stanford, UCLA, UCSB, Oxford, Cambridge, and
University College London. He has addressed scientific researchers at the leading technical
conferences in the U.S., Japan, and Europe in the areas of nanotechnology, photonics, and optical
networking. His research has been widely reported in the popular press including recent coverage in
The Guardian, Wired, BusinessWeek, MSNBC, and National Geographic. In 2002, Prof. Sargent
was celebrated by the Institute of Electrical and Electronics Engineers (IEEE), ªFor
groundbreaking research in applying new phenomena and materials from nanotechnology
towards transforming fibre-optic communications systems into agile optical networks.º In 2004 he
was named to Canada's Top 40 Under 40, a list of achievers across business, medicine, and science.
Prof. Sargent received the B.Sc.Eng. (Engineering Physics) from Queen's University in 1995 and the
Ph.D. in Electrical and Computer Engineering (Photonics) from the University of Toronto in 1998.

516  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim http://www.advmat.de Adv. Mater. 2005, 17, No. 5, March 8
E. H. Sargent/Infrared Quantum Dots

PROGRESS REPORT
Colloidal quantum dots satisfy these requirements: they may
a b
be processed from organic or aqueous solution using spin-
casting, spray-coating, and blade- or drop-casting; their syn-
thetic protocols may be extended to large-scale production;
and their tailoring through the quantum-size effect has been
widely explored.
Prior to the progress reported below, one key question re-
mained: Could infrared devices be created which meet the
demanding requirements of the above-mentioned applica-
tionsÐluminescence and photocurrent efficiencies and depths c d
of modulationÐwithin a colloidal-quantum-dot materials

Photoluminescence intensity (arb. units)


platform?

Absorbance (arb. units)


2. Colloidal Quantum Dots: Synthesis
and Properties

We illustrate progress in infrared colloidal-quantum-dot


synthesis by summarizing results for PbS, PbSe, and InAs.
PbS colloidal nanocrystals may be prepared using a solu-
tion-based organometallic route[8] using suitable precursors.
PbO is dissolved in oleic acid and heated to 150 C to produce
a lead oleate precursor. The sulfur source bis(trimethylsilyl)- 600 1000 1400 1800 900 1100 1300 1500
sulfide (TMS) in octadecene is added to the lead oleate solu- Wavelength (nm)
tion, which is then removed from heat. The temperature is al-
lowed to drop to below the nucleation-threshold temperature Figure 1. Physical properties of PbS quantum dots synthesized by an or-
and maintained at the desired growth temperature. Oleic acid ganometallic route. a,b) High-resolution TEM images of quantum dots
with peak luminescence at 1440 nm. c) Absorption spectra of PbS dots
attaches to the surface of the nanocrystals, preventing them with exciton peaks size-effect tuned from 800±1800 nm. d) Absorption
from aggregating and passivating the surface, thus reducing and photoluminescence features of 6.5 nm diameter dots. Reprinted
non-radiative recombination. The size of the nanocrystals is with permission from [8].
controlled by varying the concentration of the capping ligand;
the injection temperature and injection time; the growth tem-
perature and growth time; and the ratio of oleic acid to ton radius for PbS is 20 nm, for InAs, 34 nm, and for PbSe,
lead to sulfur. When the nanocrystals reach the desired size, 46 nm.[9] Syntheses of PbS, PbSe, and core±shell InAs/ZnSe
they are precipitated by adding a polar solvent such as metha- typically result in dilute-solution photoluminescence quantum
nol and redispersed in a nonpolar solvent such as toluene. Fig- efficiencies in the range of 30±80 %.
ures 1a,b present transmission electron microscopy (TEM) In contrast with the organometallic route, synthesis in aque-
images of dots synthesized by the organometallic route. ous solution is highly desirable for compatibility with biological
The size-dependent properties of quantum-confined states assays. It also enables multilayer polymer/nanocrystal device
are illustrated in Figures 1c,d. The minimum transition energy fabrication for heterostructure engineering through the use
in the limit of large dots is given by the bulk bandgapÐ of alternating aqueous/organic solvents in sequential layers. A
0.41 eV in the case of PbS. The quantum-confined states with- one-stage water-based synthesis of stable and monodisperse
in each band may be conceptualized in an envelope-function PbS nanocrystals was recently reported to result in photo- and
picture of a particle in a box of size and shape corresponding electroluminescence in the spectral range of 1000±1400 nm.[10]
to the boundaries of the nanocrystals. Bulk PbS has reason- PbS nanocrystals were prepared in aqueous solutions using a
ably symmetric bands: the effective mass of a hole is compar- mixture of thiols as a stabilizing agent. The combination of cap-
able to the effective mass of an electron, and the quantum- ping agents thioglycerol (TGL) and dithioglycerol (DTG) re-
confined states are similarly spaced in the two bands. For- sulted in high-quality nanocrystals. Pb(CH3COO)2 and sodium
mally, interband electronic transitions can only occur between sulfide were used as lead and sulfur sources. Size-effect tun-
electron and hole states with the same quantum numbers in ability was demonstrated in the range of 1000±1400 nm. The
the case of perfectly symmetric bands and confining-envelope combination of TGL and DTG in the appropriate ratio was
potentials. For nanocrystals that are small relative to the Bohr necessary to obtain stable nanocrystals. The absolute photolu-
radius of the bulk exciton, the quantization energies of elec- minescence quantum yield was determined to be 17 %.
trons and holes are large compared to the Coulomb energy. In The need for fluorophores stable in biological media is par-
this regime, the Coulomb binding energy reduces the transi- ticularly urgent for in-vivo assays based on quantum dots
tion energy by a comparatively small amount. The Bohr exci- emitting in the transparent tissue bands at 1050±1200 nm.

Adv. Mater. 2005, 17, No. 5, March 8 http://www.advmat.de  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 517
E. H. Sargent/Infrared Quantum Dots
PROGRESS REPORT

The growth of PbS nanoparticles on a DNA template has re- this dark state until the ejected carrier returns to the nano-
cently been demonstrated to result in efficient infrared fluoro- crystal thermally or by tunneling, resulting in neutralization.
phores which are stable in blood plasma at 37 C for over one It has similarly been proposed that photoionization results
week.[11] Quantum dots were grown on single-stranded DNA, when a carrier is trapped on the surface of the nanocrystal
as revealed in elementally resolved high-resolution TEM. The and a further photon excites this carrier within its trap-state
DNA phosphate backbone and imino protons provided sites lifetime.
for passivation and solubilization of the nanocrystals in aque- The influence of energy transfer among a population of
ous solution. Photoluminescence quantum efficiencies as high closely spaced nanocrystals is one of importance in solid-state
as 11.5 % were obtained. devices. It has been shown[17,18] that in pure nanocrystal films
In the case of InAs, reports have included the synthesis of electron±hole pairs generated in smaller nanocrystals transfer
core±shell type I heterostructure dots with an InAs core, to the larger ones. Photoluminescence measurements of mixed
shape control over rods and tetrapods in InAs, detailed prob- films have revealed quenching of the luminescence of the small
ing of intersubband transitions in InAs,[12,13] and evidence of dots and enhancement of the luminescence of the large dots.
long-sustained charging in InAs nanocrystals.[14]
Extensive time-resolved spectroscopy of visible and infrared
nanocrystals has revealed some common trends, but also some 3. Electroluminescent Devices
quantitative variability. In the visible spectrum, holes are ob-
served to relax more rapidly within the bands. In PbSe, nano- Electroluminescent sources in the infrared link electronics-
crystals capped with oleate ligands and having high quantum based communications and signal-processing systems with the
efficiencies revealed, at low pump powers, a first interband carriers used in optical communications. Measures of practical
transition with a multi-exponential decay consisting of a interest include external quantum efficiencyÐthe ratio of
200 ps decay component and a very slow decay component photons extracted to electrons injectedÐand brightness. One
which was attributed to slow radiative relaxation. Time-re- inferred quantity often estimated and reported is the internal
solved photoluminescence studies have revealed a slow single- quantum efficiencyÐthe number of photons generated per in-
exponential decay with typical time constants in the vicinity of jected electron.
1 ls. Carrier dynamics depend strongly on the nanocrystal sur- Infrared-light-emitting diodes based on colloidal nanocrys-
face, a fact which explains both variability amongst nominally tals have been fabricated[19] based on core±shell InAs/ZnSe
similar experiments and differences in observed quantum effi- quantum dots in a poly[2-methoxy-5-(2¢-ethylhexoxy)-1,4-
ciencies. phenylenevinylene] (MEH±PPV) polymer blend, resulting in
Auger recombination, a non-radiative process which de- a reported external quantum efficiency of 0.5 % and an esti-
pends on the preexistence of carriers in a band and thus grows mated internal quantum efficiency of 1 %. In these devices,
strongly with both excitation and doping, has been studied ex- the turn-on voltage exceeded 15 V. Size-tunable electrolumi-
tensively. Klimov et al. measured decay times of 10 ps for the nescence with internal efficiencies of 1.2 % was reported
4-pair relaxation, 21 ps for the 3-pair relaxation, and 45 ps based on the core-only PbS system;[20] the efficiency has re-
for the 2-pair relaxation.[15] In this vein, photoluminescence cently been improved to 3 %.[21] PbSe nanocrystal-monolayer
degradation has been observed in samples optically pumped devices[22] within a solution-processed and evaporated organic
over the course of hours. The effect has been ascribed to Au- heterostructure were reported, with electroluminescence tun-
ger ionization of the nanocrystal, giving rise to a long-lived abilities of 1.3±1.55 lm demonstrated (Fig. 2a) and peak ex-
charging and thus a dark state.[16] The nanocrystal remains in ternal efficiencies of 0.001 % obtained.

a b

Figure 2. a) Electroluminescence spectra of a series of quantum-dot light-emitting devices tuned through the near infrared. Reprinted with permission
from [21]. b) Absolute transfer efficiency of excitations from MEH±PPV to nanocrystals for four ligand choices: oleate (C18), octadecylamine (A18),
dodecylamine (A12), and octylamine (A8). Reprinted with permission from [23]. Copyright 2004 American Institute of Physics.

518  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim http://www.advmat.de Adv. Mater. 2005, 17, No. 5, March 8
E. H. Sargent/Infrared Quantum Dots

PROGRESS REPORT
The exchange of surface-capping organic ligands on as- 4. Photoconductive and Photovoltaic Devices
synthesized nanocrystals in favor of shorter insulating ligands
appears to play a significant role in achieving high electro- It is desired to harvest light efficiently in the infrared region
luminescence quantum efficiency devices. For example, preci- in devices possessing characteristics such as ease of process-
pitated PbS-nanocrystal powder was dissolved in a small ing, low cost, physical flexibility, and large-area coverage. Sen-
amount of octylamine, the solution was heated at 50 C for sitizing conjugated polymers with infrared-active nanocrystal
24±48 h, and the resulting octylamine-capped nanocrystals quantum dots provides a spectrally tunable means of acces-
were isolated by precipitation with a polar solvent.[20] The sing the infrared while maintaining the advantageous proper-
procedure was repeated to ensure uniform ligand exchange, ties of polymers.
and the final precipitate was dispersed into the desired organ- A nanocomposite approach has recently been reported[27]
ic solvent for polymer-composite fabrication. Whereas photo- in which quantum-size-effect-tuned PbS nanocrystals provide
luminescence intensities for unexchanged and exchanged infrared sensitivity for MEH±PPV. This first report demon-
nanocrystal samples were comparable, the ~ 2 nm long oleate- strated the principle, but showed a very low efficiency of 10±5,
capped nanocrystal devices exhibited a much lower electrolu- necessitating the use of modulated illumination and a lock-in
minescence intensity in the infrared than those capped with amplifier to observe the photocurrent.
~ 1 nm long octylamine ligands. It was noted that for electro- Recently demonstrated,[28] also using solution-processed
luminescence, excitations generated within the polymer ma- materials, has been a three-orders-of-magnitude improvement
trix must be transferred to the nanocrystals, and that the rate in infrared photoconductive internal quantum efficiency com-
of energy transfer is presumably sensitive to both the choice pared to the previous result,[27] allowing observation of the
of capping ligand and the proximity between exciton donor photocurrent under continuous-wave illumination without re-
and acceptor. liance on lock-in techniques; an infrared photovoltaic effect
This observation prompted subsequent investigations using was also observed for the first time in such materials. Under
photoluminescence excitation spectra of the influence of li- ±5 V bias and illumination from a 975 nm laser, these detec-
gand chemistry and ligand length on energy transfer from a tors showed an internal quantum efficiency of 3 %, a ratio of
polymer matrix to nanocrystals.[23] PbS nanocrystals capped photocurrent to dark current of 630, and a maximum respon-
with as-synthesized oleate (~ 2 nm), octylamine (~ 1 nm), do- sivity of 3.1 ” 10±3 AW±1 (Fig. 3a). The photovoltaic response
decylamine (~ 1.5 nm), and octadecylamine (~ 2 nm) ligands under 975 nm excitation resulted in a maximum open-circuit
were combined with the host matrix MEH±PPV and formed voltage of ~ 0.36 V, a short-circuit current of 350 nA, and
into films. Composites based on as-synthesized PbS nanocrys- a short-circuit internal quantum efficiency of ~ 0.006 %. By
tals capped by oleate ligands provided transfer efficienciesÐ varying the size of the nanocrystals during processing, photo-
the number of excitons transferred to the nanocrystals divided current spectra were demonstrated with peaks tailored to
by total number of excitons generated in the polymerÐof 980 nm, 1.200 lm, and 1.355 lm (Fig. 3b).
20 %. Replacing these ligands with the shortest ligands gave This first report of a photovoltaic effect in PbS/MEH±PPV
a threefold improvement in excitation-transfer efficiency nanocomposites suggested that the system may form a type II
(Fig. 2b). heterostructure in which, following optical absorption result-
The other pressing issue in achieving high electrolumines- ing in the generation of an exciton, electrons remain confined
cence quantum efficiency devices is the maximization of the to the nanocrystal while holes separate into the polymer. This
internal efficiency of the nanocrystals in the solid state. A suggests that the mechanism of excitation in forward-biased
number of authors have remarked on an order-of-magnitude electroluminescence devices must indeed be energy-transfer
loss in photoluminescence quantum efficiency when process- based, given that separate electron and hole capture is not an
ing PbS and PbSe quantum dots from solution into nanocrys- energetically favored event. That both electroluminescent and
tal or nanocrystal/polymer composite films. Whereas thin-film photovoltaic devices may be realized in this one bulk
photoluminescence quantum efficiencies in the vicinity of 1 % structure suggests the need for nanocrystal/ligand/polymer
have typically been obtained, it has recently been shown that, heterojunction systems more specifically optimized to the ap-
by controlling the chemistry and packing fraction of quantum plication in question.
dots in polymers, photoluminescence quantum efficiencies as
high as 12 % may be obtained in the solid state.[24]
The design of efficient electroluminescent devices has 5. Optical Modulation: Interband and
recently be considered from a theoretical standpoint.[25,26] Intersubband
These investigations have emphasized the importance of
symmetry in the efficiency of electron and hole injection; of a Electric-field-induced modulation of absorption in PbS
sufficient areal cross-section of nanocrystals in the path of nanocrystal quantum dots across the spectral region 600±
bipolar current; and of finding a concentration regime in 2000 nm, which more than encompasses the entire telecommu-
which both energy-transfer efficiency and internal quantum nications band, has recently been demonstrated. The device
efficiency can be maximized. architecture consisted of PbS nanocrystals synthesized via the

Adv. Mater. 2005, 17, No. 5, March 8 http://www.advmat.de  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 519
E. H. Sargent/Infrared Quantum Dots
PROGRESS REPORT

a b

Figure 3. a) Photocurrent for a sample with ~90 wt.-% nanocrystals in the polymer/nanocrystal blend. The pump powers are shown in the legend.
Inset: Proposed simplified band diagram after the Mg electrode has been deposited and the sample reaches equilibrium. b) Photocurrent spectral
response (symbols) and the corresponding absorption spectra (solid lines) for three different samples. In each sample, the photocurrent response
closely follows the absorption of the nanocrystals. The sample absorption peaks are tuned to 955 nm, 1200 nm, and 1355 nm. Reproduced from [28].
Copyright 2004 Nature Publishing Group.

organometallic route and sandwiched between SiOx films. The lock-in detection, and from the signal the change in absorp-
bottom 360 nm thick dielectric layer was thermally evaporated tion per unit length was determined. We show in Figure 4 for
onto a glass substrate overcoated with indium tin oxide (ITO), 5 nm and 7 nm diameter nanocrystals the unmodulated ab-
which served as the bottom electrical contact. Nanocrystal sorption (Figs. 4a,b) and the measured change in absorption
films ~ 700 nm thick were spin-coated onto the SiOx-covered due to the modulated applied electric field (Figs. 4c,d). The
substrate and then overcoated with a top layer of 150 nm thick insets of Figures 4a,b show the set of optical transitions re-
SiO, followed by a 5 nm layer of Au as a top electrical contact. vealed in the quantum dots as a result of the sensitive electro-
A sine-wave voltage was applied with amplitudes between absorption spectroscopy provided by these measurements.
25 V and 65 V. Monochromatic light was incident perpendicu- These transitions correspond to the negative lobes of the elec-
lar to the sample surface and thus parallel to the direction of troabsorption spectrum. The dependence of the amplitude of
the applied field. The electroabsorption signal reported was the measured electroabsorption signal on applied peak volt-
acquired at twice the frequency of the applied voltage via age at each of the first three optical transitions revealed a

a b
e

c d

Figure 4. a,b) Unmodulated absorption spectra and c,d) electroabsorption spectra for devices based on 5 nm (a,c) and 7 nm (b,d) PbS nanocrystals.
The insets show the transitions corresponding to the first three peaks in the measured electroabsorption spectrum. e) The change in absorption coeffi-
cient in PbSe quantum dots in an electrochemical cell under bias. Reproduced with permission from [29]. Copyright 2003 American Chemical Society.

520  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim http://www.advmat.de Adv. Mater. 2005, 17, No. 5, March 8
E. H. Sargent/Infrared Quantum Dots

PROGRESS REPORT
quadratic dependence on field. The maxima in the electroab- sional superlattice effect which red-shifts emission relative to
sorption spectra closely correspond to the positions of the first absorption.
excitonic peak, confirming the predominance of excitonic
broadening as the basis for the observed effect.
Modulation based on charge injection, lowering the rate of 7. Challenges and Future Prospects
interband transitions by occupying final states and enabling
intersubband transitions by configuring carriers in the required The applications presented in the introduction to this
initial states, has been demonstrated in a variety of electro- Progress Report demand a variety of different performance
chemical-cell configurations. Hole injection into quantum-con- specifications from processible infrared optoelectronic ma-
fined states has been explored in PbSe,[29] with the correspond- terials.
ing spectroscopic changes measured (Fig. 4e). The exploitation Optoelectronic integration for the unification of communi-
of intersubband transitions, already widely exploited in epitax- cations and computing platforms demands the merging of
ial semiconductors, give promise for 2±10 lm modulators, de- sources, detectors, and modulators onto a single chip. As
tectors, and light emitters. While the electrochemical approach shown herein, each of these device types has been demonstrat-
gives a very strong absorption change, its timescale of operation ed individually, but the set of processes and a prototype which
tends to be measured in seconds, whereas a number of applica- unify them remain to be elaborated. Ultimately, the challenges
tions would benefit from at least kilohertz response times. associated specifically with integration on a silicon substrateÐ
the canonical platform for electronicsÐneed to be addressed.
In all electrically injected polymer-based devices, the problem
6. Optical Gain of charge-transport lengths and traversing inorganic/organic
interfaces is of urgent importance. It would be of tremendous
Room-temperature amplified spontaneous emission and interest to exploit the outstanding electrical properties of crys-
spectral narrowing at infrared wavelengths have been report- talline silicon, and from it, to transfer energy,[32] without the
ed in solution-processed films made up of both PbSe[30] and requirement of separately transferring electrons and holes,
PbS[31] quantum-dot nanocrystals. The results are relevant to across silicon/polymer/quantum dot interfaces. The use of sili-
optical amplification and, ultimately, the possibility of realiz- con nanocrystals[33] may provide a first means of addressing
ing lasers integrated on a variety of substrates. In the case of the paucity of excitons in bulk silicon at room temperature.
PbSe, spectral narrowing and optical gain (Fig. 5a) were re- Significant narrowing of the spectral shape of emission, de-
ported in an inorganic sol±gel titania matrix. In the case of tection, and modulation to the scale of one coarse wave-
PbS, the active optical medium provided amplified sponta- length-division multiplexed channelÐ20 nmÐwould further
neous emission at 1300 nm (Fig. 5c) and operated at room enhance the value of quantum-dot-based approaches. This
temperature without any additional matrix material, provid- hinges on creating dots with room-temperature homogeneous
ing an optical gain of 260 cm±1 and a pump threshold of linewidths in the vicinity of 10 meV and on reducing inhomo-
1 mJ cm±2 (Fig. 5b). Nanocrystals synthesized in aqueous solu- geneous broadening to this order.
tion and stabilized using short ligands resulted in a high quan- Electroluminescence efficiencies stand to be improved,
tum-dot volume fraction in solid films and a three-dimen- principally through further enhancements to infrared quan-

a b c
N

Figure 5. a) Bleaching in PbSe nanocrystals in hexane. Decay, measured at 1570 nm, increases with the pumping level; the gain lifetime near threshold
is approximately 10 ps. Reproduced with permission from [30]. Copyright 2003 American Chemical Society. b) Growth, in PbS nanocrystal-based films,
of amplified spontaneous emission in comparison with growth of spontaneous photoluminescence intensity with a threshold of ~1 mJ cm±2. c) Evolu-
tion of the emission spectra as a function of pump intensity normalized at the point of the photoluminescence peak position for 150 lm excitation
wavelength. Inset: Normalized emission spectra for low (PL) and high (ASE) pump intensities. (b,c) reproduced from [31]. Copyright 2005 Optical
Society of America.

Adv. Mater. 2005, 17, No. 5, March 8 http://www.advmat.de  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 521
E. H. Sargent/Infrared Quantum Dots
PROGRESS REPORT

tum-dot internal efficiencies: ligand, polymer, and device fab- [6] G. O. Schlengel, F. W. Burkholder, S. A. Kleim, W. A. Beckman,
rication strategies which preserve in the solid state the high B. D. Woods, J. D. Muhs, Sol. Energy 2004, 76, 359.
[7] L. M. Fraas, J. E. Avery, H. X. Huang, R. U. Martinelli, Semicond.
quantum yields that are already available in solution will sig-
Sci. Technol. 2003, 18, S165.
nificantly aid this cause. [8] M. A. Hines, G. D. Scholes, Adv. Mater. 2003, 15, 1844.
Photodetector quantum efficiencies and photovoltaic power [9] F. W. Wise, Acc. Chem. Res. 2000, 33, 773.
efficiencies will benefit from a stronger internal impetus to- [10] L. Bakueva, I. Gorelikov, S. Musikhin, X. S. Zhao, E. H. Sargent,
wards rapid charge separation prior to recombination. Shorter E. Kumacheva, Adv. Mater. 2004, 16, 926.
ligands, conducting ligands, and the removal of insulating li- [11] L. Levina, V. Sukhovatkin, S. Musikhin, S. Cauchi, E. H. Sargent,
unpublished.
gands in favor of direct polymer±nanocrystal contacts will aid
[12] D. Steiner, D. Katz, O. Millo, A. Aharoni, S. H. Kan, T. Mokari,
in this process, as will the use of smaller ionization-potential U. Banin, Nano Lett. 2004, 4, 1073.
polymers. External quantum efficiencies will be enhanced by [13] D. Krapf, S.-H. Kan, U. Banin, O. Millo, A. Sa'ar, Phys. Rev. B: Con-
increasing the effective optical density at the exciton peak dens. Matter Mater. Phys. 2004, 69, 733 011.
through increased nanocrystal loadings and increased interac- [14] E. Nahum, Y. Ebenstein, A. Aharoni, T. Mokari, U. Banin, N. Shi-
moni, O. Millo, Nano Lett. 2004, 4, 103.
tion lengths via a combination of thicker nanotextured active
[15] V. I. Klimov, A. A. Mikhailovsky, D. W. McBranch, C. A. Leather-
layers and the use of resonant cavity structures. dale, M. G. Bawendi, Science 2000, 287, 1011.
Electro-optic modulationÐeither in the form of electro-ab- [16] M. Nirmal, B. O. Dabbousi, M. G. Bawendi, J. J. Macklin, J. K.
sorption or electric-field-induced refractive index changesÐ Trautman, T. D. Harris, L. E. Brus, Nature 1996, 383, 802.
will require the combination of a significant depth of modula- [17] C. R. Kagan, C. B. Murray, M. G. Bawendi, Phys. Rev. B: Condens.
tion with bandwidths in the kilohertz range (friend/foe identi- Matter Mater. Phys. 1996, 54, 8633.
[18] S. A. Crooker, J. A. Hollingsworth, S. Tretia, V. I. Klimov, Phys.
fication) and well beyond for communications applications.
Rev. Lett. 2002, 89, 18 602/1.
Communications-wavelength lasing on a silicon substrate [19] N. Tessler, V. Medvedev, M. Kazes, S. Kan, U. Banin, Science 2002,
using gain provided by processible materials is impatiently 295, 1506.
awaited. First demonstrations will rely on a substantial optical [20] L. Bakueva, S. Musikhin, M. A. Hines, T.-W. F. Chang, M. Tzolov,
gain per unit length, combined with sufficiently high-quality- G. D. Scholes, E. H. Sargent, Appl. Phys. Lett. 2003, 82, 2895.
factor cavities to enable a predominance of stimulated emis- [21] L. Bakueva, G. Konstantatos, L. Levina, S. Musikhin, E. H. Sargent,
Appl. Phys. Lett. 2004, 84, 3459.
sion within the timeframe on which gain is availableÐeither
[22] J. S. Steckel, S. Coe-Sullivan, V. Bulovic, M. G. Bawendi, Adv. Mater.
through the gain lifetime itself, or through a temporally 2003, 15, 1862.
stretched excitation pulse. [23] T.-W. F. Chang, S. Musikhin, L. Bakueva, L. Levina, M. A. Hines,
Biological applications of infrared quantum dots are en- P. W. Cyr, E. H. Sargent, Appl. Phys. Lett. 2004, 84, 4295.
abled by already-demonstrated efficient, biocompatible, stable [24] T.-W. F. Chang, A. Maria, P. W. Cyr, V. Sukhovatkin, L. Levina,
E. H. Sargent, Synth. Met., Published Online November 21, 2004,
fluorophores emitting in the second tissue window. These
DOI: 10.1016/j.synthmet.2004.10.003.
await full exploitation in this promising application area. [25] A. Shik, G. Konstantatos, E. H. Sargent, H. E. Ruda, J. Appl. Phys.
Received: September 21, 2004 2003, 94, 4066.
Final version: November 2, 2004 [26] A. Shik, S. Yu, E. Johnson, H. Ruda, E. H. Sargent, Solid-State Elec-
tron. 2002, 46, 61.
[27] S. A. McDonald, P. W. Cyr, L. Levina, E. H. Sargent, Appl. Phys.
Lett. 2004, 85, 2089.
± [28] S. A. McDonald, G. Konstantatos, S. Zhang, P. W. Cyr, E. J. D.
[1] "Electronic & Photonic Integrated Circuit (EPIC)", Announcement Klem, L. Levina, E. H. Sargent, Nat. Mater., Published Online Janu-
BAA04-15, Defense Advanced Research Projects Agency, Arling- ary 9, 2005, DOI: 10.1038/nmat1299.
ton, VA 2004. [29] B. L. Wehrenberg, P. Guyot-Sionnest, J. Am. Chem. Soc. 2003, 125,
[2] K. Rosenbach, J. Schiller, in IEEE 2000 Radar Conf., IEEE, Piscat- 7806.
away, NJ 2000, p. 305. [30] R. D. Schaller, M. A. Petruska, V. I. Klimov, J. Phys. Chem. B 2003,
[3] X. Gao, Y. Cui, R. M. Levenson, L. W. K. Chung, S. Nie, Nat. Bio- 107, 13 765.
technol. 2004, 22, 969. [31] V. Sukhovatkin, S. Musikhin, I. Gorelikov, S. Cauchi, L. Bakueva,
[4] Y. T. Lim, S. Kim, A. Nakayama, N. E. Stott, M. G. Bawendi, J. V. E. Kumacheva, E. H. Sargent, Opt. Lett. 2005, 30, 171.
Frangioni, Mol. Imaging 2003, 2, 50. [32] M. Achermann, M. A. Petruska, S. Kos, D. L. Smith, D. D. Koleske,
[5] S. Kim, Y. T. Lim, E. G. Soltesz, A. M. De Grand, J. Lee, A. Nakaya- V. I. Klimov, Nature 2004, 429, 642.
ma, J. A. Parker, T. Mihaljevic, R. G. Laurence, D. M. Dor, L. H. [33] L. Pavesi, L. Dal Negro, C. Mazzoleni, G. Franzo, F. Priolo, Nature
Cohn, M. G. Bawendi, J. V. Frangioni, Nat. Biotechnol. 2004, 22, 93. 2000, 408, 400.

______________________

522  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim http://www.advmat.de Adv. Mater. 2005, 17, No. 5, March 8

You might also like