- Email: [email protected]
Emerging non-traditional Förster resonance energy transfer configurations with semiconductor quantum dots: Investigations and applications
Accepted Manuscript Title: Emerging Non-Traditional F¨orster Resonance Energy Transfer Configurations with, Semiconductor Quantum Dots: Investigations and Applications Author: W. Russ Algar Hyungki Kim Igor L. Medintz Niko Hildebrandt PII: DOI: Reference:
S0010-8545(13)00150-1 http://dx.doi.org/doi:10.1016/j.ccr.2013.07.015 CCR 111750
To appear in:
Coordination Chemistry Reviews
Received date: Revised date: Accepted date:
19-4-2013 21-7-2013 22-7-2013
Please cite this article as: W.R. Algar, H. Kim, I.L. Medintz, N. Hildebrandt, Emerging Non-Traditional F¨orster Resonance Energy Transfer Configurations with, Semiconductor Quantum Dots: Investigations and Applications, Coordination Chemistry Reviews (2013), http://dx.doi.org/10.1016/j.ccr.2013.07.015 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Manuscript
1
Emerging Non-Traditional Förster Resonance Energy Transfer Configurations with
2
Semiconductor Quantum Dots: Investigations and Applications
3 4 5
ip t
W. Russ Algar1,*, Hyungki Kim1, Igor L. Medintz2, Niko Hildebrandt3
6 1
9
Department of Chemistry
2036 Main Mall
11
Vancouver BC V6T 1Z1
12
Canada
13
Phone: 604-822-2464
14
Fax: 604-822-2847
15
[email protected]
an
10
us
University of British Columbia
M
8
cr
7
17
2
d
16
Center for Bio/Molecular Science and Engineering U.S. Naval Research Laboratory Code 6900
19
4555 Overlook Avenue SW
20
Washington DC 20375
21
USA
23
3
Ac ce p
22
te
18
Institut d'Electronique Fondamentale
24
Université Paris-Sud
25
91405 Orsay Cedex
26
France
27 28
*To whom correspondence should be addressed: [email protected] (W.R.A.)
1 Page 1 of 63
Abstract
30
Förster resonance energy transfer (FRET) configurations incorporating colloidal semiconductor
31
quantum dots (QDs) have proven to be a valuable tool for bioanalysis and bioimaging. Mirroring
32
well established techniques with only fluorescent dyes, ―traditional‖ FRET configurations with
33
QDs have involved single-step energy transfer to organic dye acceptors mediated by
34
biomolecular interactions. Here, we review recent progress in characterizing non-traditional
35
FRET configurations incorporating QDs and their application to challenges in biosensing, energy
36
conversion, and fabrication of optoelectronic devices. Such non-traditional FRET configurations
37
with QDs include substitution of organic dyes with lanthanide complexes, polypyridyl transition
38
metal complexes, azamacrocyclic metal complexes, graphene (oxide), carbon nanotubes, gold
39
nanoparticles, and dyes exhibiting photochromism. Other non-traditional configurations of
40
interest include FRET relays (with or without organic dyes) that feature multiple sequential
41
energy transfer steps, and thin films of QDs where discrete FRET pairs cannot be defined,
42
including those where QDs are layered in a size-sequential or ―rainbow‖ structure. The
43
calculation of FRET efficiencies and donor-acceptor distances in the above configurations are
44
reviewed, as are distance scaling relationships for non-zero dimensional acceptors, and the
45
related dipolar energy transfer mechanism, nanosurface energy transfer (NSET). To illustrate the
46
utility of non-traditional QD-FRET configurations, we highlight examples of optically
47
switchable probes, photonic wires, time-gated and multiplexed probes for biosensing, enhanced
48
light harvesting in QD and dye sensitized solar cells (DSSC), and colour conversion in light
49
emitting diodes (LEDs). We close by providing a perspective on how the combined utility of
50
these non-traditional QD-FRET configurations may be useful for engineering complex nanoscale
51
devices in the future.
53 54
cr
us
an
M
d
te
Ac ce p
52
ip t
29
Keywords: quantum dot, Förster resonance energy transfer (FRET), metal complex, biosensing, thin film, solar cell
55
2 Page 2 of 63
ip t
cr
us
an
M
d te
79
Table of Contents Abbreviations .................................................................................................................................. 4 1. Introduction ................................................................................................................................. 6 2. Quantum Dots and FRET............................................................................................................ 8 3. Photochromic FRET ................................................................................................................. 11 4. FRET with Metal Complexes ................................................................................................... 14 4.1. Lanthanide complexes ....................................................................................................... 14 4.2. Polypyridyl transition metal complexes............................................................................. 17 4.3. Azamacrocyclic metal complexes...................................................................................... 18 4.4. Distinguishing FRET from charge transfer........................................................................ 21 5. Energy Transfer with Other Nanomaterials .............................................................................. 23 5.1. Carbon nanomaterials ........................................................................................................ 24 5.2. Gold nanoparticles ............................................................................................................. 25 6. Biomolecule Supported FRET Relays ..................................................................................... 28 6.1. Long distance relays .......................................................................................................... 28 6.2. Concentric relays ............................................................................................................... 29 7. FRET in Thin Films of Quantum Dots ..................................................................................... 31 7.1. Characterization of energy transfer .................................................................................... 32 7.2. FRET in quantum dot sensitized solar cells ....................................................................... 34 7.3. Luminescent films and composite materials ...................................................................... 38 8. Summary and Perspective ......................................................................................................... 40 Acknowledgements. ...................................................................................................................... 41 References ..................................................................................................................................... 42
Ac ce p
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
3 Page 3 of 63
Abbreviations
81
η
Power conversion efficiency
82
0D
Zero-dimensional
83
1D
One-dimensional
84
2D
Two-dimensional
85
2PE
Two-photon excitation
86
AFP
Alpha-fetoprotein
87
APC
Allophycocyanin
88
AOT
Dioctyl sulfosuccinate
89
Au NP
Gold nanoparticle
90
BIPS
1'-3-dihydro-1'-(2-carboxyethyl)-3,3-dimethyl-6-nitrospiro-[2H-1-benzopyran-
cr us
an
91
ip t
80
2,2'-(2H)-indoline] BRET
Bioluminescence resonance energy transfer
93
CNT
Carbon nanotube
94
CRET
Chemiluminescence resonance energy transfer
95
CrONO
trans-Cr(cyclam)(ONO)2+
96
CT
Charge transfer
97
DMPET
Dipole-to-metal particle energy transfer
98
dsDNA
Double-stranded DNA
99
DSSC
100
EYFP
101
FLIM
102
FRET
103
FWHM
104
GO
105
IPCE
106
ITO
Indium-doped tin oxide
107
LbL
Layer-by-layer
108
LDH
layered double hydroxide
109
LED
Light emitting diode
110
LLnC
Luminescent lanthanide complex
Ac ce p
te
d
M
92
Dye sensitized solar cell Enhanced yellow fluorescent protein Fluorescence lifetime imaging microscopy Förster resonance energy transfer Full-width-at-half-maximum Graphene oxide
Incident-photon-to-current conversion efficiency
4 Page 4 of 63
LSPR
Localized surface plasmon resonance
112
LTbC
Luminescent Tb3+ complex
113
MAA
Mercaptoacetic acid
114
MBP
Maltose binding protein
115
MOF
Metal organic framework
116
MPA
3-Mercaptopropionic acid
117
M-PC
Metallophthalocyanines
118
NO
Nitric oxide
119
NSET
Nanosurface energy transfer
120
PALM
Photoactivation localization microscopy
121
PC
Phthalocyanine
122
pcFRET
Photochromic FRET
123
PDDA
Poly(diallyldimethylammonium chloride)
124
PDT
Photodynamic therapy
125
PL
Photoluminescence
126
PP
Polypyridine
127
PSS
Poly(sodium-4-styrene sulfonate)
128
QD
Quantum dot
129
QDSSC
Quantum dot sensitized solar cell
130
SOFI
131
SPR
132
ssDNA
133
STORM
Stochastic optical reconstruction microscopy
134
TA
Transient absorption
135
WSPO
cr us
an
M
d
te
Ac ce p
136
ip t
111
Super-resolution optical fluctuation imaging Surface plasmon resonance Single-stranded DNA
Spironaphthoxazine dye
137
5 Page 5 of 63
1. Introduction
139
Colloidal semiconductor nanocrystals, better known as quantum dots (QDs), are prospective
140
materials for a wide variety of applications. QDs are characterized by a unique set of optical
141
properties that primarily arise from quantum confinement effects [1-4]. Foremost among these
142
properties is their bright photoluminescence (PL), the colour of which can be tuned on the basis
143
of nanocrystal size and composition. Other favourable optical properties include broad
144
absorption spectra with large one-photon (ε = 104–107 M–1cm–1) and two-photon (σ2PE = 103–104
145
GM) absorption cross-sections, large effective Stokes shifts (up to hundreds of nanometers),
146
spectrally narrow and symmetric PL emission (full-width-at-half-maximum, FWHM = 25–35
147
nm), and good resistance to photobleaching [1]. These properties have made QDs very attractive
148
probes for bioimaging and bioanalysis, where they are often touted for their multiplexing
149
capability and suitability for single particle visualization and tracking [1, 5-10]. Equally
150
important in such applications is the interface of the QD, which represents a nanoscale scaffold
151
for chemistry and functionalization. The physical properties of the QD can be tailored through
152
application of ligand or polymer coatings [11], and biological activity can be obtained through
153
bioconjugation [12]. The latter can include binding or other reactions with biomarkers, targeting
154
of cells and tissues, and delivery of therapeutics. In abiotic roles, QDs also represent building
155
blocks for optically active thin films, superlattice structures [13], and various composite
156
materials [14] for optoelectronic applications such as light emitting diodes (LEDs) and displays,
157
lasers, photodetectors, and solar cells [15].
cr
us
an
M
d
te
Ac ce p
158
ip t
138
159
QDs become even more powerful tools when their above-mentioned attributes are combined
160
with electron transfer or Förster resonance energy transfer (FRET) processes that can, for
161
example, modulate QD PL to generate active signaling or sensitize a secondary process. FRET is
162
perhaps best known as a spectroscopic tool for biophysical studies, including vesicle fusion and
163
membrane dynamics [16], protein folding [17], DNA detection [18], enzyme assays [19], ligand-
164
receptor and protein-protein interactions [20], both in the ensemble and at the single molecule
165
level [21-23]. In typical FRET experiments, a biomolecule is co-labeled with a donor
166
fluorophore and an acceptor chromophore, or, alternatively, two interacting biomolecules are
167
individually labeled with donor and acceptor. In the former case, conformational changes affect
6 Page 6 of 63
the fluorescence from the donor (and acceptor) due to the nanometric distance-dependence of
169
FRET; in the latter case, changes in fluorescence are due to association or dissociation. To a
170
large extent, these highly successful biological applications of FRET have inspired similar
171
developments that incorporate QDs and their unique optical properties as participants in FRET,
172
typically as donors. There is now a myriad of studies where QDs are used in either of the two
173
general FRET configurations noted above, and several comprehensive reviews have been written
174
on the utility of these configurations in bioimaging and bioanalysis [24-26].
cr
ip t
168
175
This review examines some of the recent work in the literature involving non-traditional FRET
177
configurations based on QDs. We use the term ―traditional‖ to refer to energy transfer in a single
178
step between a QD and an organic dye molecule, usually as a discrete pairing of a QD donor
179
with one or more equivalent organic dye acceptors in bulk solution. Here, we discuss QD-FRET
180
configurations that depart from this norm in one or more important ways. Deviations may
181
include the substitution of conventional organic dyes with other chromophores; for example,
182
lanthanide and other metal complexes, gold nanoparticles, carbon allotropes, or even organic
183
dyes with unusual properties, such as photochromism. Special attention is paid to the
184
characterization of energy transfer in these systems, including other mechanisms that may
185
supersede or compete with FRET. Moreover, by our definition, a departure from traditional QD-
186
FRET configurations is not limited to the choice of acceptor. Non-traditional configurations may
187
also include architectures where multiple energy transfer pathways are incorporated; for
188
example, as in the case of multistep sequential energy transfer, otherwise known as a FRET
189
―relay‖ or ―cascade.‖ Configurations of this nature have been realized with QD-bioconjugates in
190
solution and with thin films of QDs at an interface. The latter are also non-traditional FRET
191
configurations in that discrete donor-acceptor pairs cannot be defined. While the success of
192
traditional QD-FRET methods has been impressive, even greater capability and innovations are
193
expected from non-traditional QD-FRET configurations. As will be discussed, new capabilities
194
in bioanalysis and therapeutics, enhancements in solar energy conversion, and fabrication of
195
improved optoelectronic devices have already been demonstrated.
Ac ce p
te
d
M
an
us
176
196
7 Page 7 of 63
2. Quantum Dots and FRET
198
Prior to discussing non-traditional QD-FRET configurations, it is worthwhile to review the
199
fundamentals of FRET theory, including its ―traditional‖ application with QDs. FRET is a
200
resonant dipole-dipole coupling interaction that occurs through-space to transfer excitation
201
energy from a donor fluorophore to an acceptor chromophore. The energy transfer is
202
radiationless, occurring without the involvement of photons. Theodor Förster, after whom the
203
process is named, first elucidated the mechanism between 1946–1948 [27-30]. The genius of
204
Förster was that he was able to describe this process in terms of spectroscopic donor and
205
acceptor properties that could be measured through relatively simple experiments.
us
cr
ip t
197
206
The rate of energy transfer in FRET is given by eqn. 1, where τD–1 = k0 = kr + knr is the inverse
208
native lifetime of the donor excited state, kr is the rate of radiative relaxation, knr is the rate of
209
non-radiative relaxation (i.e., internal conversion and non-FRET quenching mechanisms), r is
210
the distance separating the donor and acceptor, and R0 is the Förster distance for the donor-
211
acceptor pair.
M
an
207
6
(1)
te
d
k FRET
1 æR ö = ç 0÷ tD è r ø
The Förster distance is defined by eqn. 2, where ΦD is the PL quantum yield of the donor, NA is
213
Avogadro‘s number, n is the refractive index of the medium, J(λ) is the spectral overlap integral,
214
and κ2 is the orientation factor [31].
Ac ce p
212
9(ln10)FDk 2 J(l ) R = = (8.79 ´10-28 mol) n-4FDk 2 J(l ) 5 4 128p N A n 6 0
(2)
215
The spectral overlap integral measures the degree of resonance between the donor emission
216
spectrum and acceptor absorption spectrum. It is defined by eqn. 3, where FD(λ) is the
217
wavelength-dependent fluorescence intensity of the donor, ε(λ) is the wavelength-dependent
218
molar absorption coefficient for the acceptor, and λ is the wavelength.
ò F (l )e (l )l J(l ) = ò F (l )d l D
A
4
dl
(3)
D
8 Page 8 of 63
219
To utilize eqn. 2 directly, the units for λ and ε(λ) are cm and cm2 mol–1 (=103 M–1 cm–1),
220
respectively. Its often more convenient to calculate the spectral overlap in units of nm and M–1
221
cm–1 for λ and ε(λ), in which case R0 = 0.02108 (2Dn–4J)1/6 in units of nm.
222 The orientation factor, 0 ≤ κ2 ≤ 4, accounts for the dependence of the dipole-dipole interaction on
224
their relative orientation. The orientation factor is calculated from eqn. 4, where θT is the angle
225
between the donor emission and acceptor absorption transition dipoles, and θD and θA are the
226
angles those transition dipoles make to the vector connecting them. While occasionally a
227
contentious point, κ2 = 2/3 tends to be assumed as a first approximation in most FRET studies.
228
This value is strictly true for isotropic and dynamically random orientations of the donor and
229
acceptor transition dipole moments during the donor excited state lifetime (in the presence of
230
acceptor). When these conditions cannot be reasonably assumed, determination of κ2 is
231
considerably more complex. Although such circumstances are beyond our scope here, more
232
detailed discussion can be found in van der Meer‘s text [32].
M
an
us
cr
ip t
223
233
2
(4)
d
k 2 = ( cosq T - 3cosq D cosq A )
The FRET efficiency, 0 ≤ EFRET ≤ 1, is given by eqn. 5, which simplifies to the terms on the
235
right-hand side for a discrete donor-acceptor pair. It can be seen from this equation that R0
236
represents the donor-acceptor separation at which the FRET efficiency is 50%.
Ac ce p
te
234
kFRET R06 EFRET = = kr + knr + kFRET R06 + r 6
(5)
237
Experimentally, the FRET efficiency can be measured from quenching of donor emission,
238
eqn. 6, where the subscript D denotes an isolated donor quantity, the subscript DA denotes a
239
donor quantity measured in the presence of acceptor, I is a fluorescence intensity, and <τ> is an
240
amplitude-weighted average lifetime.
EFRET = 1-
t I DA = 1- DA ID tD
(6)
241 242
As discussed elsewhere, the applicability of the FRET formalism with QDs has been validated
243
by many studies [25, 26]. Donor-acceptor separations are measured from the center of the QD,
9 Page 9 of 63
such that its radius imposes a minimum separation. It is usually assumed that κ2 = 2/3 for QD
245
donors due to their two-fold degenerate (circular) transition dipole, the rotational freedom of
246
proximal dyes, and the random assembly of those dyes to the QD surface. As donors, QDs are
247
advantageous in that their narrow, continuously tunable emission (by size or composition)
248
permits optimization of the spectral overlap integral while also minimizing crosstalk in
249
measurements of donor and (fluorescent) acceptor PL. Their broad absorption spectra also
250
provide flexibility in choice of excitation wavelength so that direct excitation of an acceptor dye
251
can be minimized. The other general advantages of QDs—their brightness, resistance to
252
photobleaching, and more facile multiplexing—are also relevant to FRET configurations.
253
Moreover, the large surface area of QDs permits arraying of multiple acceptors per QD, thereby
254
providing another mechanism to optimize FRET efficiencies. For the case of N equivalent
255
acceptors around a central QD donor, eqn. 5 can be modified to eqn. 7, where it is seen that
256
FRET efficiency increases as the number of acceptors increases. Given the approximately
257
spherical shape of QDs, such arrangements are easily accessed in real experiments.
M
an
us
cr
ip t
244
(7)
d
NR06 EFRET = NR06 + r 6
QDs are also excellent acceptors for certain special classes of donors (e.g., lanthanides [33],
259
chemiluminophores [34], and bioluminescent proteins [35]). Here, the primary advantage of QDs
260
is their absorption spectra, which lead to very large spectral overlap integrals and thus Förster
261
distances. Arraying multiple donors per QD also increases the probability of energy transfer. The
262
probability of FRET-sensitization of a central QD acceptor by M donors is given by eqn. 8. Note
263
that this expression does not count the number of energy transfers to a QD from M donors, but
264
rather the probability of at least one such event occurring.
Ac ce p
te
258
æ r6 ö PA = 1- ç 6 6 ÷ è R0 + r ø
M
(8)
265
More detailed information on traditional QD-FRET can be found in recent review articles [25,
266
26, 33, 36].
267
10 Page 10 of 63
3. Photochromic FRET
269
Photochromic dyes exhibit reversible colour changes with optical stimulation. The change in
270
absorption spectra is accompanied by changes in molecular and electronic structure, including
271
photoinduced ring opening or closing, cis/trans isomerism, cycloaddition, and proton or electron
272
transfer reactions [37-39]. The photogenerated states typically undergo thermal reversion back to
273
the initial state. Photochromic dyes are candidate materials for applications requiring or
274
benefiting from optical switching; for example, data storage, sensing, and super-resolution
275
imaging [39-41]. Research on photochromism has included both photochromic FRET (pcFRET),
276
where energy transfer is modulated ‗on‘ and ‗off‘ through photoinduced changes in the spectral
277
overlap integral [42], and the design of photochromic nanoparticle constructs [43]. This section
278
reviews studies on pcFRET with QD donors, which have largely made use of
279
spiroindolinobenzopyran and diheteroarylethene dyes as photochromic acceptors. In traditional
280
QD-dye FRET pairs, the spectral overlap integral is generally constant. In contrast, when using
281
photochromic acceptor dyes, the spectral overlap integral is dynamic and optically controllable.
M
an
us
cr
ip t
268
282
The first instance of pcFRET with QD donors was reported by Medintz et al. [44], who labeled
284
maltose binding protein (MBP) with BIPS dye (1',3-dihydro-1'-(2-carboxyethyl)-3,3-dimethyl-6-
285
nitrospiro-[2H-1-benzopyran-2,2'-(2H)-indoline]). In its spiropyran form, BIPS is nearly
286
colourless, exhibiting absorption bands at wavelengths < 400 nm. Upon UV irradiation, BIPS
287
photoconverts to its coloured merocyanine form with an absorption band at ca. 550 nm. When
288
polyhistidine-tagged MBP was labeled with BIPS (5 dyes/MBP) and self-assembled to
289
CdSe/ZnS QDs (20 MBP/QD; see Fig. 1A), FRET from the QD to the BIPS could be reversibly
290
modulated with ca. 60% FRET efficiency. The QD PL was centered at 555 nm and resonant with
291
the absorption peak of the merocyanine form. The broad absorption spectrum of the QD
292
permitted its excitation and interrogation of the pcFRET process at a wavelength minimum
293
common to both the spiropyran and merocyanine forms of BIPS, thereby avoiding
294
photoconversion of the BIPS to one state while measuring the effect of the other state.
Ac ce p
te
d
283
295 296
Other groups have also studied pcFRET between QDs and spiroindolinobenzopyran dyes. For
297
example, Zhu et al. directly conjugated a thiol-linker modified BIPS derivative, 5-(1,3-dihydro-
298
3,3-dimethyl-6-nitrospiro[2H-1-benzopyran-2,2'-(2H)-in-dole])hexane-1-thiol with CdSe/ZnS
11 Page 11 of 63
QDs and demonstrated photoswitching of FRET [45]. Similarly, Tomasulo et al. investigated
300
pcFRET between CdSe/ZnS and CdS QD donors and a BIPS derivative with a dithiolane linker
301
[46]. With CdSe QDs, 45% PL quenching efficiency was observed upon photoswitching from
302
the spiropyran to the merocyanine form [46]. In the case of the CdS QDs, the method of
303
preparation of the CdS QDs was found to impact photochromism [47]. Two CdS QD samples,
304
both coated with 1-decanethiol, but one prepared via a solvothermal synthesis with TOPO
305
ligands and the other prepared in dioctyl sulfosuccinate (AOT) inverse micelles, were found to
306
induce positive and negative photochromism, respectively, when BIPS was conjugated. These
307
results were consistent with the known sensitivity of BIPS photochromism to the polarity of the
308
local microenvironment, and the negative photochromism was attributed to electrostatic
309
interactions with surface defects on the QDs synthesized from AOT inverse micelles. In a recent
310
study, Lee et al. adsorbed a spironaphthoxazine dye (WSPO) on aqueous CdTe QDs coated with
311
3-mercaptopropionic acid (MPA) and observed six cycles of pcFRET with efficiencies >80%
312
upon conversion to the merocyanine form [48]. The difference in spectral overlap with the QD
313
between the spiropyran and merocyanine forms is shown in Fig. 1B. These authors have also
314
prepared photochromic composite materials with CdTe QDs, WSPO, and poly(N-
315
isopropylacrylamide) [49].
cr
us
an
M
d te
316
ip t
299
Jares-Erijman et al. demonstrated pcFRET from a CdSe/ZnS QD donor to a photochromic
318
diheteroarylethene dye acceptor of the dithienylethene family [50]. The dye was synthesized as a
319
biotin derivative and assembled to QDs modified with streptavidin. Similar to BIPS, the closed
320
form of the dithienylethene dye had a visible absorption band in resonance with the QD (peak PL
321
at 565 nm) to generate FRET. QD PL quenching could be reversibly modulated over 14
322
complete cycles with ~30% FRET efficiency. A heterodiarylethene dye was chosen because this
323
family generally exhibits more robust photochromic properties (i.e., less switching fatigue) than
324
spiropyran dyes. Jares-Erijman and coworkers also recently synthesized amphiphilic polymer
325
coatings for CdSe/ZnS and CdSe/CdS/ZnS QDs with pendant dithienylethene dyes sequestered
326
within the hydrophobic layer to provide an optimal microenvironment for photoconversion (Fig.
327
1C) [51, 52]. In one of these designs, there was an average of ~35 dithienylethene dyes per QD
328
and approximately one non-photochromic red fluorescent dye per QD as an internal standard.
Ac ce p
317
12 Page 12 of 63
329
Photoswitching was possible over 15 complete cycles with alternating illumination at 340 nm
330
and 545 nm and two-fold changes in QD PL intensity.
331 Similar to the examples above, Erno et al. also found that a dithienylethene could engage in
333
pcFRET with CdSe/ZnS QDs; however, these authors observed switching fatigue after just three
334
cycles [53]. This behaviour was attributed to electron transfer competing with FRET and causing
335
the irreversible reduction of the dye. It is important to note that the chemistry in this system was
336
quite different from that used by Jares-Erijman. Hydrophobic QDs were directly modified with a
337
large excess of the dithienylethene dye, such that strong quenching of QD PL was observed with
338
both the open (~86%) and closed forms (~96%) of the dyes. Efficient and robust photoswitching
339
of QD PL via pcFRET should be possible with carefully designed interfaces between the QD and
340
photochromic dye.
an
us
cr
ip t
332
341
Several super-resolution imaging techniques, e.g., stochastic optical reconstruction microscopy
343
(STORM) and photoactivation localization microscopy (PALM), require that fluorescent dyes or
344
proteins switch between bright and dark states [54, 55]. Another technique, super-resolution
345
optical fluctuation imaging (SOFI), developed by Dertinger et al., requires reversibly switchable
346
emitters with optically distinguishable states but not necessarily bright and dark states [56]. QDs
347
that can be reversibly quenched through pcFRET are potential probes for these and similar super-
348
resolution methods. The optical properties of QDs offer several benefits in diffraction-limited
349
imaging and may be similarly amenable to super-resolution imaging. In this context, remaining
350
challenges include enhancing the efficiency of energy transfer, optimizing photoconversion in
351
QD-photochromic dye conjugates, minimizing thermal reversion, and evaluating the stochastic
352
optical switching of individual QDs assembled with multiple photochromic acceptor dyes. While
353
the blinking or PL intermittency of QDs has already been used to enable super-resolution
354
imaging [56, 57], pcFRET may be able to offer a greater degree of control. Other potential
355
applications for QD-pcFRET pairs include use as probes for fluorescence lifetime imaging
356
microscopy (FLIM) or sensitive lock-in detection due to the ability to controllably modulate QD
357
PL lifetime and intensity [52]. QD-pcFRET pairs may also prove useful as switchable elements
358
for molecular photonic logic devices [43, 58].
Ac ce p
te
d
M
342
359
13 Page 13 of 63
4. FRET with Metal Complexes
361
The photochromic acceptor dyes discussed in the previous section are examples of organic dyes
362
with special properties, but are organic dyes nonetheless. This section describes FRET between
363
QDs and lanthanide or other metal complexes. Lanthanide complexes are interesting due to their
364
very unique optical properties, which, when paired with QDs, comprise a fascinating non-
365
traditional FRET system. Other metal complexes, and particularly those with transition metal
366
centres, are interesting due their ability to engage in both charge transfer (CT) and FRET with
367
QDs. The CT mechanism is often more intuitive and distinguishing between FRET and CT is not
368
always straightforward (see Section 4.4). The pairing of a QD with a metal complex can thus be
369
a non-traditional FRET configuration due to the observation of FRET in preference to CT. In
370
addition, these configurations can be non-traditional due to the ability of that FRET process to
371
sensitize subsequent chemical reactions at the metal centre.
372
4.1. Lanthanide complexes
373
Two unique optical properties of trivalent lanthanide ions are their narrow, line-like emission
374
spectra and their remarkably long excited state lifetimes, which are on the order of microseconds
375
to milliseconds [59, 60]. Luminescence arises from parity and spin-multiplicity forbidden 4f-4f
376
transitions that are shielded from the surrounding environment by the 5s2 and 5p6 sub-shells.
377
Since the 4f-4f transitions are very weak (ε < 3 M–1 cm–1 [60]), luminescence is not practically
378
observed from isolated lanthanide ions. Rather, lanthanide ions are complexed with organic
379
ligands that serve as antennae to sensitize their luminescence. The ligands, which have non-
380
trivial molar absorption coefficients (~104 M–1 cm–1), are generally considered to sensitize
381
lanthanide luminescence through a multi-step process: (i) light absorption and excitation to the
382
singlet excited state of the ligand; (ii) intersystem crossing to the triplet state of the ligand; (iii)
383
energy transfer from the ligand to the lanthanide ion; and (iv) emission from the lanthanide ion
384
(in competition with non-radiative relaxation processes) [59, 60]. The long excited state lifetimes
385
have made lanthanide ions particularly attractive for time-resolved fluorescent assays with ultra-
386
low background, especially with FRET to allophycocyanin (APC) [61-64], and, more
387
pertinently, good donors for QD acceptors in FRET [33].
Ac ce p
te
d
M
an
us
cr
ip t
360
388 389
The first work pairing luminescent lanthanide complexes (LLnCs) with QDs was reported by
390
Hildebrandt and coworkers [65, 66]. Streptavidin conjugates of Eu3+ and Tb3+ complexes were 14 Page 14 of 63
titrated with biotinylated CdSe/ZnS QDs (Fig. 2A) and increasing amounts of time-gated QD PL
392
were observed. Excitation was at 308 nm to be resonant with the absorption band of the
393
lanthanide complexes and PL was collected with a 250 µs delay after pulsed excitation and
394
integrated over 750 µs. What was particularly noteworthy about this configuration was that the
395
QDs served as energy acceptors for the LLnC donors. In traditional FRET configurations with
396
organic dyes, QDs are rather poor acceptors. Due to their broad and intense absorption spectra,
397
QDs are always directly excited with the dye donor; however, only the ground state of the QD is
398
a good acceptor. The comparatively longer excited state lifetime of a QD (>10 ns vs. < 5 ns for
399
most dyes) is such that only a trivial fraction of the population is able to return to its ground state
400
and effectively serve as an acceptor for a proximal excited state dye [67]. In contrast, the 340–
401
680 µs and 1.48 ms luminescence lifetimes of the Eu3+ and Tb3+ complexes, respectively,
402
provided the opportunity for the QDs to relax to their ground state (τQD < 100 ns) while a
403
majority of the lanthanides remained in their excited state [66]. These conditions permit efficient
404
energy transfer to a QD acceptor. A hallmark of LLnC-to-QD FRET is that the QD acquires an
405
apparent PL lifetime on the same order of magnitude as the LLnC since the rate of energy
406
transfer is much slower than the decay rate of the QD [66, 68] .
d
M
an
us
cr
ip t
391
407
One outcome of the use of QDs as acceptors for LLnC donors is access to very large Förster
409
distances (R0 ≈ 7–11 nm) due to the large molar absorption coefficients and broad absorption
410
spectra of QDs [65, 66, 68-70]. Larger Förster radii lead to more efficient FRET, which
411
translates into more sensitive analyses and/or compatibility with nanoscale architectures of larger
412
dimensions. Importantly, Hildebrandt and coworkers have shown that energy transfer from
413
luminescent Tb3+ complexes (LTbCs) to QDs can be utilized as a ―molecular ruler‖ analogous to
414
other FRET pairs; the PL decays of the LTbC and QD can be analyzed to extract distance
415
information such as the size and shape of non-spherical QDs [68, 69]. This analysis is done on
416
the basis of PL decays. Ideally, the LTbC donor intensity decays (denoted D) is fit to a single
417
exponential function, eqn. 9, where τD is the unquenched native lifetime of the LTbC donor, AD
418
is the total amplitude, and t is time.
Ac ce p
te
408
I D (t) = AD exp(-t / t D )
(9)
419
A contribution from the native LTbC decay (i.e., unquenched donor) is usually observed even in
420
the presence of QD acceptors (Fig. 2B). The PL intensity decay for LTbC-QD FRET pairs 15 Page 15 of 63
421
(denoted DA) is thus fit with a multiexponential function with N ≥ 2 components, one of which
422
is the unquenched LTbC decay, as shown in eqn. 10. N -1 é ù I DA (t) = ADA êa D exp(-t / t D ) + åa DA,i exp(-t / t DA,i ) ú i=1 ë û
(10)
The average FRET efficiency can be calculated from eqn. 11, where <τDA> is the average decay
424
time for the LTbC in the presence of QD acceptor (without the native LTbC decay), and related
425
to an average donor-acceptor distance.
cr
t DA 1 é åa DA,it DA,i ù = 1- ê ú tD t D êë åa DA,i úû
(11)
us
E = 1-
ip t
423
Alternatively, the individual FRET-quenched LTbC lifetime components may be used to
427
calculate distance components. Hildebrandt et al. have used this approach to assess the effective
428
size and shape of functionalized QDs in aqueous environments (cf. the artificial environment
429
used in electron microcopy) [69]. In some experiments, a strictly monoexponential decay will
430
not be observed for LTbC donor alone, and corrections are necessary to account for additional
431
short lifetime components [69]. The QD acceptor PL decays can also be analyzed to extract
432
FRET efficiencies and donor-acceptor distances. The analysis of acceptor PL decay, while more
433
complex and beyond our scope here (see refs. [68, 69, 71]), often provides better signal-to-
434
background ratios since it largely avoids contributions from unquenched LTbC. The above
435
methodology should be equally applicable with other LLnCs as well (e.g., Dy3+, Eu3+, Tm3+).
M
d
te
Ac ce p
436
an
426
437
Another important outcome of FRET between LLnCs and QDs is the combined utility of FRET
438
and background rejection (e.g., autofluorescence, scattering) associated with time-gating. This
439
capability is highly advantageous for assays in complex biological matrices prone to scattering or
440
autofluorescent interferences. Time-gated homogeneous immunoassays for estradiol and alpha-
441
fetoprotein (AFP), a cancer biomarker, have been developed on the basis of FRET between
442
LLnCs and QDs. In the former, CdTe QDs (peak PL at 728 nm) served as acceptors for Eu3+ and
443
Tb3+ chelate derivatives of estradiol in a competitive assay format [72]. In the AFP sandwich
444
assay, hydrophobic CdSe/ZnS QDs (peak PL at 605 nm) were doped into polystyrene
445
microparticles that were conjugated with anti-AFP antibodies; a second set of anti-AFP
446
antibodies were labeled with a luminescent Tb3+ complex. When AFP was present in a sample, it
447
bridged the QDs and Tb3+ complexes to generate FRET [73]. A further advantage of LLnC-QD 16 Page 16 of 63
FRET pairs is that relatively high levels of multiplexing are possible due to the broad absorption
449
of QDs, which permits the use of one type of LLnC as a common donor for multiple QD
450
acceptors. For example, Geißler et al. have demonstrated the multiplexed detection of FRET
451
from a luminescent Tb3+ complex to CdSe/ZnS QDs with peak PL at 529, 565, 604, and 653 nm,
452
and CdSeTe/ZnS QDs with peak PL at 712 nm (Fig. 2C) [70]. This system was assembled using
453
botinylated QDs and LLnC-labeled streptavidin as a model for five-plex bioaffinity assays. The
454
Tb3+ complex had sharp emission lines at ca. 490, 545, 585, and 620 nm, permitting sensitization
455
of all the QD acceptors and resolution of their emission.
cr
ip t
448
us
456
Beyond assays, the time-gating process essentially represents dynamic switching between
458
―FRET off‖ conditions in the prompt time regime (typically < 102 ns) and ―FRET on‖ conditions
459
in a gated time regime (> 102 µs), which might be expected to have implications for developing
460
photonic logic devices.
461
4.2. Polypyridyl transition metal complexes
462
Polypyridines (PPs) are a well-known class of coordination complex that have numerous
463
applications due to their (electro)chemical and optical properties [74, 75]. The redox activity of
464
polypyridine complexes has arguably lead to a general expectation that these compounds should
465
engage in photoinduced charge transfer reactions with QDs. Commensurate with these
466
expectations, rhenium bipyridine [76], ruthenium bipyridine [77], ruthenium phenanthroline
467
[78], and oxo-centered triruthenium clusters (with pyridine ligands) [79] have been shown to
468
function as electron or hole acceptors for photoexcited QDs. Nonetheless, recent work has
469
demonstrated that FRET can occur between QDs and certain osmium PP complexes.
M
d
te
Ac ce p
470
an
457
471
In one study, McLaurin et al. developed a fluororatiometric oxygen sensor by pairing a CdxZn1–
472
xSe/CdyZn1–yS
473
1,10-phenanthroline)(N-(6-aminohexyl)-4'-methyl-2,2'-bipyridine-4-carboxamide)osmium(II)
474
bis(hexafluorophosphate) ([OsII(Ph2phen)2(Nbpy)](PF6)2) [80]. The Os(II)PP complex was
475
directly conjugated to a carboxylated polymer coating on the QD through a short alkyl amine
476
linker via carbodiimide chemistry. Under two-photon excitation (2PE), the QD served as an
477
antenna to absorb light and transfer the excitation energy to the Os(II) complex. The quantum
478
yield of the resulting 2PE-FRET-sensitized Os(II)PP phosphorescence was sensitive to oxygen,
core/shell QD donor with a phosphorescent Os(II)PP acceptor, bis(4,7-diphenyl-
17 Page 17 of 63
with progressive quenching observed, relative to the QD PL intensity, for O2 pressures between
480
0–760 torr [80]. In another study, Stewart et al. thoroughly characterized energy transfer between
481
an aqueous CdSe/ZnS QD and an Os(II)PP complex, [OsII(bpy)2(phen-NCS)](PF6)2 (bpy, 2,2'-
482
bipyridine; phen, phenanthroline) [81]. The Os(II)PP complex was attached to the distal terminus
483
of a short peptide sequence that self-assembled to the QDs via a hexahistidine tag (Fig. 3A-i).
484
The QD PL was progressively quenched as increasing numbers of Os(II)PP-peptide were
485
assembled per QD. Despite the weak light absorption by the Os(II) complex (εmax, vis = 12 500 M–
486
1
487
625 nm), measurements of steady-state PL (Fig. 3A-ii), time-resolved PL, and femtosecond
488
transient absorption were consistent with a FRET mechanism of energy transfer. Inhomogeneous
489
quenching of the QD PL that paralleled the spectral overlap (Fig. 3A-iii) was also indicative of
490
FRET rather than CT. Nonetheless, there were some nuances that reflected photophysical
491
differences between the QD-Os(II)PP complex pair and more conventional QD-dye FRET pairs,
492
suggesting the possibility of non-FRET contributions to energy transfer [81].
493
4.3. Azamacrocyclic metal complexes
494
Azamacrocycles are ubiquitous in coordination chemistry. These ligands and their metal
495
complexes have found widespread use as catalysts, dyes, receptors for ion sensing, and
496
therapeutics, among other applications. For example, porphyrin and phthalocyanine (PC)
497
complexes are characterized by high absorption coefficients, fast electron transfer reactions, and
498
redox and optical properties that can be tuned via organic substituents or selection of metal
499
centers [82]. A prominent biological application of these azamacrocycles is their use as
500
sensitizers for photodynamic therapy (PDT) [83, 84]. PDT involves the photoinduced production
501
of reactive oxygen species in situ to kill cancerous or other harmful tissue. Several porphyrins
502
have been approved for clinical use in North America and Europe, including porfimer
503
(Photofrin®), temoporfin (Foscan®), talaporfin (Laserphyrin®), and verteporfin (Visudyne®).
504
To date, most studies of FRET between QDs and porphyrins and related compounds, such as
505
phthalocyanines and chlorins, have been in the context of potential applications in PDT.
cr
ip t
479
Ac ce p
te
d
M
an
us
cm–1 at 490 nm) and small spectral overlap with various QD donors (peak PL at 520, 530, 550,
506 507
Burda‘s group was among the first to investigate energy transfer between hydrophobic QDs and
508
metallophthalocyanines (M-PCs) [85-88]. Silicon PCs with different axial ligands were adsorbed
509
to hydrophobic CdSe QDs and photophysical characterization revealed that there were non18 Page 18 of 63
FRET contributions to energy transfer [85, 86]. The non-FRET contributions were attributed to
511
the involvement of surface states in energy transfer. Growth of a thin ZnS shell (i.e., CdSe/ZnS
512
QD) passivated these states and reduced the overall energy transfer efficiency by a factor of two
513
[86], suggesting that FRET may have been in competition with one or more charge transfer
514
pathways. A study by Wen et al. has also suggested that Dexter energy transfer is a possible
515
competing mechanism with FRET, at least between CdTe QDs and meso-tetra(4-
516
sulfonatophenyl)porphine dihydrochloride [89]. However, Nyokong‘s group has surveyed energy
517
transfer between CdTe QDs and many different M-PCs, attributing quenching of QD PL and
518
sensitization of M-PC emission to FRET in each case [90-97]. These M-PCs incorporated Zn(II),
519
Al(III), In(III), Ge(IV), Si(IV), Sn(IV), and Ti(IV) centers with a diverse array of ligands, and
520
were associated with the QDs through either adsorption or covalent attachment to their ligands.
us
cr
ip t
510
an
521
Much of the motivation for the abovementioned and other similar studies is the possibility of
523
2PE FRET as a possible modality for PDT. NIR radiation is ideal for tissue penetration;
524
however, PCs tend to have modest 2PE cross-sections. For example, the Si-PCs used by Burda‘s
525
group had a two-photon absorption cross-section of only 750 GM for an excitation band between
526
1300–1400 nm [88]. This spectral region is shared by the vibrational overtones of water and thus
527
is not well-suited for penetrating biological tissue. In contrast, the QD antennae had much larger
528
two-photon absorption cross-sections and permitted excitation within the biological tissue
529
window (ca. 650–950 nm [98]). Burda and coworkers were able to show that 2PE energy
530
transfer between CdSe QDs and Si-PCs was possible with 38% efficiency, although singlet
531
oxygen generation was not investigated [88]. Fortunately, subsequent studies have addressed this
532
latter point.
d
te
Ac ce p
533
M
522
534
In one of the first studies on singlet oxygen generation via QD-FRET, Tsay et al. labeled
535
phytochelatin-related peptides with chlorin e6 (no metal centre) as both an acceptor for
536
CdSe/CdS/ZnS QDs (peak PL at 620 nm) and a photosensitizer [99]. An average of 26 chlorin e6
537
molecules per QD were required to reach 50% FRET efficiency. The QDs were exploited as light
538
harvesting antennae, having a 10-fold larger absorption coefficient at 610 nm than the chlorin
539
had at its peak absorption (654 nm). The QD-chlorin conjugates had singlet oxygen quantum
540
yields between 10–31% depending on the excitation wavelength. Although these values were
19 Page 19 of 63
lower than for chlorin alone [99], this drawback could potentially be offset by the stronger
542
absorption of the QDs. In contrast, Tekdaş et al. studied putative FRET from QD donors to tetra-
543
triethyleneoxythia substituted Al(III)- (Fig. 3B), Ga(III)- and In(III)-PC acceptors and found that
544
the QD enhanced singlet oxygen quantum yields, suggesting potential use of QD-M-PC
545
conjugates as photosensitizers in PDT [100]. Furthermore, Qi et al. were able to show that 2PE
546
FRET between CdSe/ZnS QDs (peak PL at 595 nm) and a porphyrin (TrisMPyP-COOH; no
547
metal centre) produced twice the amount of singlet oxygen as 2PE of TrisMPyP-COOH alone
548
[101]. This study very clearly illustrates the power of using QDs as both antennae and scaffolds
549
for photosensitizers: despite a meager Förster distance of 2.7 nm and very low pairwise FRET
550
efficiency, enhanced singlet oxygen production was possible due to the large two-photon
551
absorption cross-section of the QD (ca. 800–2400 GM units between 800–1100 nm) and the
552
capacity for loading ~100 porphyrin molecules per QD. Note that the QDs were coated with an
553
amphiphilic polymer such that the Förster distance was shorter than the actual donor-acceptor
554
separation. In the foregoing examples, it is probable that FRET was the dominant, if not
555
exclusive, mechanism of energy transfer due to the core/shell/shell [99] and core/shell/polymer
556
[101] structures of the QDs.
cr
us
an
M
d
557
ip t
541
Beyond singlet oxygen generation, QDs are also potential FRET antennae for other
559
photosensitized reactions of metal complexes. For example, nitric oxide (NO) is an important
560
biological signaling molecule that is involved in thrombosis, vasodilation, neural activity, and
561
immune response, while also having potential as a cancer therapeutic [102-105]. It can be
562
photogenerated in situ from metal nitrosyl and nitrito complexes [106]. Ford‘s group found that
563
one such complex, a chromium(III) tetraazacyclotetradecane complex, trans-Cr(cyclam)(ONO)2+
564
(abbreviated CrONO), was able to serve as a FRET acceptor for aqueous CdSe/ZnS QDs [107-
565
109]. Although CrONO is an efficient photogenerator of NO (ΦNO = 25%), its practical use is
566
limited by weak light absorption (εCrONO < 150 M–1 cm–1) [109]. To address this limitation,
567
CrONO was adsorbed on QDs coated with anionic DHLA ligands through electrostatic
568
interactions (Fig. 3C-i). The QDs had molar absorption coefficients that were >1000-fold larger
569
than that for CrONO, and were thus able to serve as light harvesting antennae. UV light was
570
absorbed by the QDs and the energy transferred, via FRET, to the quartet excited state of the
571
CrONO, which subsequently produced NO as a result of a transition to its reactive doublet state
Ac ce p
te
558
20 Page 20 of 63
(Fig. 3-ii) [109]. Enhancements in NO production were between 3-fold and 15-fold [107-109].
573
Given the large two-photon absorption cross-sections of QDs, this type of configuration is also
574
likely to be suitable for harvesting NIR excitation light, which is most suitable for use in tissues
575
and other biological matrices. In addition to CrONO, FRET was observed with a dichloro
576
analog, trans-Cr(cyclam)Cl2+ (without NO production), but not a dicyano analog, trans-
577
Cr(cyclam)(CN)2+, that lacked spectral overlap with the QDs [108, 109]. In addition to QD-
578
FRET sensitized production of NO, these results show that FRET between QDs and transition
579
metal complexes can be tuned by selection of axial ligands.
us
580
cr
ip t
572
Finally, in addition to their role as sensitizers for PDT, porphyrins and phthalocyanines are also
582
potential sensitizers for dye sensitized solar cells (DSSCs) [82, 110]. DSSCs are a promising
583
technology that may provide a low-cost and sustainable source of energy in the future; however,
584
their practical use is currently hindered by lower efficiencies than competing technologies [111].
585
Improvements in sensitizers are critical for improving DSSC efficiencies, and porphyrins and
586
phthalocyanines are being investigated for this purpose [112-114]. Some of these investigations
587
are now also incorporating QDs and FRET. Recently, Jin et al. prepared porphyrin derived metal
588
organic frameworks (MOFs) that were surface-modified with CdSe/ZnS QDs [115]. The MOF
589
comprised 1,2,4,5-tetrakis(4-carboxyphenyl)-benzene connected by zinc centers; the QDs were
590
coated with PEG-amine, which bound to the zinc centers to immobilize the QDs. Very efficient
591
energy transfer (>80%, ~1 ns) from the QDs to the MOF was observed and attributed to FRET.
592
Electron transfer from the QDs to the MOF was not energetically allowed and the PL decay for
593
the MOF did not show any indication of hole transfer. Applying a monolayer of red-emitting
594
QDs to the MOF increased its light harvesting capability by 51% under one sun conditions. In
595
devices with a 10 porphyrin layer MOF, which corresponds to the estimated exciton migration
596
distance in the material, an enhancement of 5–10% is expected [115]. Section 7.2 describes
597
further applications for QD-FRET in DSSC development.
598
4.4. Distinguishing FRET from charge transfer
599
While it is relatively straightforward to establish that an energy transfer mechanism exists
600
between a QD and metal complex, it is not always straightforward to unambiguously determine
601
that quenching of QD PL is due to a Förster-type dipole-dipole interaction (FRET) rather than
602
photoinduced charge transfer (CT). Even QD-fluorescent dye systems are not always
Ac ce p
te
d
M
an
581
21 Page 21 of 63
straightforward. For example, it would generally be assumed, at first glance, that Rhodamine B
604
would engage in FRET with QDs; however, Lian‘s group recently found that FRET was only
605
responsible for ~16% of the energy transfer from CdS QDs to adsorbed Rhodamine B dye
606
molecules [116, 117]. The remaining ~84% was attributed to electron transfer. Similar results
607
were found for methylene blue adsorbed on CdSe QDs (~6% FRET; ~94% charge transfer)
608
[118], albeit that this dye is better known for its propensity for charge transfer.
ip t
603
cr
609
Experimentally, both FRET and photoinduced CT quench steady-state QD PL and decrease the
611
QD PL lifetime. Using these parameters to track changes in energy transfer efficiency, E, with
612
increases in the number of acceptors per QD will not provide any direct insight into the
613
quenching mechanism. All mechanisms will yield an analogous trend according to eqn. 12,
614
where kET, i is the rate of an arbitrary energy transfer pathway. N
ET, i
i
M
E=
åk
an
us
610
N
k0 + å kET, i
(12)
i
In contrast, experiments that vary donor-acceptor parameters such as spectral overlap and
616
separation distance can provide insight into the mechanism of energy transfer. For example,
617
FRET would be characterized by a linear dependence on the spectral overlap integral and
618
orientation factor, and an inverse sixth power dependence on the donor-acceptor separation.
619
Unfortunately, systematically and reproducibly varying these parameters is not always practical
620
or even feasible, and therefore other observations must provide insight. When the acceptor is
621
fluorescent, the observation of sensitized acceptor emission is often a good indication of FRET.
622
Similarly, when QD donor quenching is observed without significant spectral overlap, CT is a
623
likely mechanism. This was the case with quenching of QD PL by dopamine, which has been
624
confirmed to be due to CT [119, 120]. However, as demonstrated by the aforementioned studies
625
with Os(II)PP complexes, even weak absorption can be sufficient for non-trivial spectral overlap
626
[81]. Changes in QD PL quenching with reduction or oxidation of a putative CT acceptor
627
(electron or hole) can also provide insight. For example, Burks et al. did not observe NO
628
production upon oxidation of CrONO (Section 4.3), which supported a FRET mechanism for
629
energy transfer [109], while Medintz et al. exploited the poor CT between QDs and the reduced
Ac ce p
te
d
615
22 Page 22 of 63
630
hydroquinone form of dopamine for pH sensing (the oxidized quinone form of dopamine was a
631
good electron acceptor) [119].
632 One of the most important techniques for assessing possible CT interactions is transient
634
absorption (TA) spectroscopy, which permits tracking of ultra-fast processes and identification
635
of transient intermediates. Observation of spectroscopic signatures for oxidized or reduced forms
636
(as appropriate) of metal complexes or dyes permits assignment of CT as the energy transfer
637
mechanism. Luminescence upconversion and other pump-probe experiments can also be useful
638
in this regard. Nonetheless, definitive evidence for CT can be challenging to obtain when the
639
acceptor has weak or absent transient absorption features.
640
us
cr
ip t
633
Certainly, the majority of the literature would suggest that, at short distances (e.g., adsorbed
642
metal complexes or dyes) and with core-only QDs, CT is the preferred mechanism of energy
643
transfer. However, Medintz and coworkers [78] and Benson‘s group [121] have demonstrated
644
that ruthenium(II) phenanthroline complexes can engage in charge transfer with core/shell QDs
645
when bridged by peptides or proteins. Conversely, as described in Section 4.3, chromium(III)
646
cyclam complexes and PCs engage in FRET when adsorbed to QDs. CT is clearly not limited to
647
core-only QDs and near contact distances between donor and acceptor, nor is FRET limited to
648
core/shell QDs and large donor-acceptor separations. The assumptions that (i) short-range energy
649
transfer with hydrophobic core-only QDs occurs through CT and (ii) long-range energy with
650
aqueous core/shell QDs occurs via FRET are good starting points for directing the design of
651
photophysical experiments to characterize energy transfer, but are indeed assumptions that need
652
to be tested.
653
5. Energy Transfer with Other Nanomaterials
654
Although the metal complexes in the previous section represent non-traditional partners for QDs
655
in FRET, a feature they have in common with organic dyes is that they can be approximated as
656
zero-dimensional (0D) or point dipoles. Similarly, while QDs are much larger than both organic
657
dyes and metal complexes, they too have been shown to reduce to approximate point dipoles
658
[122]. Combinations of these materials thus comprise 0D-0D pairs and energy transfer is
659
described by the Förster formalism. However, there are increasing reports of energy transfer
Ac ce p
te
d
M
an
641
23 Page 23 of 63
between QDs and other nanomaterials that are not necessarily 0D. Dipole-dipole coupling is still
661
expected to occur in many of these instances, but the dimensionality of the other nanomaterial
662
impacts the formalism and, more pertinently, the distance scaling of the process. This section
663
reviews some of what is currently known about dipole-dipole energy transfer between QDs,
664
carbon nanomaterials, and gold nanoparticles.
665
5.1. Carbon nanomaterials
666
There have been several recent reports of pairing QDs with, for example, graphene [123],
667
graphene oxide (GO) [124], and carbon nanotubes (CNT) [125] for putative FRET assays. The
668
cited examples are nucleic acid assays where the proximity between the carbon nanomaterials
669
and QDs is mediated by the stronger affinity of single-stranded oligonucleotides (ssDNA) toward
670
carbon nanomaterials than their double-stranded counterparts (dsDNA). The difference in
671
affinity arises from the relative inability of the purine and pyrimidine bases in dsDNA to interact
672
with the carbon nanomaterials. When the ssDNA probes are conjugated to QDs, the absence of
673
target DNA results in adsorption of the conjugates and efficient quenching of QD PL.
674
Hybridization with target DNA disrupts the interactions of the ssDNA probe with the carbon
675
surface and the proximity required for energy transfer is lost. In addition to DNA, Liu et al. have
676
reported a homogeneous sandwich immunoassay where GO-reporter antibody conjugates and
677
QD-capture antibody conjugates were brought together in the presence of target antigen,
678
resulting in quenching of QD PL [126]. While such applications establish that energy transfer
679
occurs, they do not clearly establish that the mechanism of energy transfer is FRET.
cr
us
an
M
d
te
Ac ce p
680
ip t
660
681
Morales-Narváez et al. found that the effective quenching of QD PL by carbon nanomaterials
682
followed the order graphite < carbon nanofiber ≈ multiwall CNT < GO (Fig. 4A) [127]. This
683
order of quenching efficiency followed the trend in surface area (m2 g–1) for the carbon
684
nanomaterials. It was also suggested that the planar nature of the GO provided a more favourable
685
κ2 for energy transfer. The overall trend was not unexpected given the elegant results from
686
Shafran et al. [128], who showed that a CNT behaved as a one-dimensional (1D) acceptor for
687
CdSe/ZnS QDs, and Jander et al. [129], who showed that a 9 nm thick amorphous carbon film
688
behaved as a two-dimensional (2D) acceptor for CdSe/CdS nanorods. These dimensionalities are
689
in contrast to the conventional Förster formalism, which models the donor and acceptor as 0D
690
dipoles. The distance (r) dependent scaling for the rate of energy transfer with a 0D donor is r–6 24 Page 24 of 63
691
for a 0D acceptor, r–5 for a 1D acceptor, r–4 for 2D acceptor, and r–3 for a 3D acceptor. A r–4
692
dependence was confirmed with the thin film of amorphous carbon [129] and is theoretically
693
expected for graphene [130].
694 The evidence to date suggests that QDs can engage in Förster-like energy transfer through
696
dipolar interactions with carbon nanomaterials. However, the rate of energy transfer in such
697
configurations is not expected to have the inverse sixth-power distance scaling of conventional
698
FRET. Moreover, charge transfer has been reported between various QD materials (e.g., CdSe,
699
CdTe) and graphene, GO [131-133], and CNTs [134, 135], such that there is a possible
700
competition between dipolar energy transfer and charge transfer. Again, distinguishing these
701
mechanisms may not be straightforward, and careful design of the interface between the QD and
702
carbon nanomaterial may be able to tip the scales in favour of one mechanism or the other.
703
5.2. Gold nanoparticles
704
Gold nanoparticles (Au NPs) have been paired with QDs as energy acceptors in more studies
705
than can be reviewed here (for example, see refs. [136-139] and many others). In most of these
706
studies, the Au NPs have served a function analogous to a non-fluorescent molecular dye
707
acceptor (i.e., dark quencher). Their primary advantage was more efficient quenching over
708
longer distances. Here, we briefly review what is known about the mechanism of energy transfer
709
between QDs and Au NPs.
cr
us
an
M
d
te
Ac ce p
710
ip t
695
711
To appreciate energy transfer between QDs and Au NPs, it is worthwhile to briefly review
712
energy transfer between organic dyes and Au NPs. Dulkeith et al. studied the quenching of dye
713
fluorescence by Au NPs ranging from 1–30 nm in size [140, 141]. The quenching was found to
714
be due to changes in both the radiative and non-radiative rates of the dye in close proximity to
715
the Au NP; however, the Gersten-Nitzan model [142] for calculating radiative and non-radiative
716
rates near dielectric nanoparticles overestimated both quantities. Strouse and coworkers similarly
717
studied the quenching of fluorescent dyes by 1–2 nm Au NPs and found excellent agreement
718
with a nanosurface energy transfer (NSET) model [143-145], which was developed from the
719
work of Chance, Prock, and Silbey [146], as well as Persson and Lang [147, 148], on energy
720
transfer from a dipole to a metal surface. In NSET, the coupling is between the donor dipole and
721
the nearly free electrons in the conduction band of the NP, which is considered to be an 25 Page 25 of 63
approximate surface of localized dipoles. The rate of energy transfer in this system has an
723
inverse fourth-power dependence, such that the efficiency of energy transfer is given by eqn. 13,
724
where d is the donor-nanosurface separation and d0 is the characteristic distance. The value of d0
725
is defined by eqn. 14, where c is the speed of light, ωD is the angular frequency of the donor, and
726
ωf and kf are the angular frequency and Fermi wavenumber, respectively, for the nanosurface
727
[143]. The values for gold are substituted in the right-hand side of eqn. 14.
c3F D
nw D2 w f k f
(
= 2.23 ´10-25 cm s
)
c3F D nw D2
us
d04 = 0.225
d04 d04 + d 4
cr
ENSET =
ip t
722
(13)
(14)
Experiments have shown that the efficiency of NSET improves as overlap between the donor
729
emission (520–780 nm) and the localized surface plasmon resonance (LSPR) of a 2 nm Au NP
730
(~525 nm) increases [144]. The most important consequence of the NSET mechanism is that
731
energy can be efficiently transferred to the Au NP over larger distances (up to ~15–20 nm) than
732
FRET between dyes (< 10 nm).
M
d
733
an
728
Considering QDs, an early study by Gueroui and Libchaber found that single-pair energy transfer
735
between CdSe/ZnS QDs and 1.4 nm Au NPs was consistent with FRET [149], although it was
736
later suggested that the narrow range of distance data also fit a NSET model [150]. A detailed
737
study by Pons et al. found that the energy transfer efficiency between CdSe/ZnS QDs and 1.4 nm
738
Au NPs had a distance dependence that was consistent with NSET [150]. The QD-Au NP
739
separation distance was controlled through relatively rigid ß-sheet peptide spacers that were
740
labeled at their distal terminus with the Au NPs and self-assembled to the QDs. Importantly, in a
741
direct comparison with FRET between QDs and fluorescent dyes, energy transfer to the Au NPs
742
was found to be more efficient over longer distances. This result is a particularly important
743
advantage with QDs since the donor-acceptor separation is taken from the center of the QD, such
744
that the radius of the inorganic nanocrystal and the thickness of any organic coating set a
745
minimum separation. In another study, Li et al. found that energy transfer from CdSe/ZnS QDs
746
to 3 nm Au NPs had a NSET-like d–4 dependence, whereas energy transfer to 80 nm Au NPs had
747
a FRET-like d–6 dependence [151]. Energy transfer to 15 nm Au NPs also fit better to a FRET-
748
like model, although the predicted differences between NSET and FRET were less significant for
Ac ce p
te
734
26 Page 26 of 63
this size of Au NP (Fig. 4B). Note that these were rather limited data sets, with data for only
750
three different donor-acceptor separations collected per size of Au NP. Interestingly, Han et al.
751
found that quenching of QD PL by slightly smaller 12 nm Au NPs was consistent with a NSET
752
model [152]. Analysis of steady state and time-resolved PL data indicated a QD-Au NP
753
separation of 7.9–8.1 nm, which was corroborated by TEM imaging. Zhang et al. similarly
754
demonstrated that the PL from thin films of CdTe QDs (five sizes/colours; peak PL between
755
534–660 nm) were quenched in accordance with NSET when controllably positioned at specific
756
distances above a thin film of 5.5 nm Au NPs [153]. Here, the NSET efficiency was given by
757
eqn. 15, where rAu is the radius of the Au NPs, cAu is the concentration of Au NPs per unit area,
758
and d is the distance between the QDs and Au NP layers.
ENSET = 1+
3( d - rAu )
3
cAup d04 ( 3d - rAu )
an
-1
us
cr
ip t
749
759
(15)
The possibility that the mechanism of energy transfer depends on the size of the Au NP has been
761
recognized with molecular dye donors. For example, Griffin et al. have shown that although
762
NSET is able to predict the quenching efficiency as a function of distance with up to 8 nm Au
763
NPs, the model appears to underestimate the quenching efficiency with 40 nm and 70 nm Au
764
NPs [154]. Notably, the NSET model does not require a resonant interaction and small (≤ 2 nm)
765
Au NPs do not display a strong LSPR, suggesting that, as Au NPs increase in size and the
766
magnitude of their LSPR increases, the NSET model becomes insufficient. Discrepancies may
767
also be due to the increased contribution of scattering relative to absorption in the extinction
768
spectra of the Au NPs [144]. Another model, dipole-to-metal particle energy transfer (DMPET),
769
has been proposed to explain the quenching of dyes by Au NPs [155]. This model has not been
770
widely applied with QDs and has been suggested to reduce to other models within the constraints
771
of real experimental systems [144, 150].
Ac ce p
te
d
M
760
772 773
Further studies are needed to clearly define the Au NP size regimes that accept energy from QDs
774
through NSET, FRET-like, or other dipolar and higher order (e.g., quadrupolar) interactions.
775
Although conceptually similar, the scaling differences between FRET and NSET are significant.
776
Detailed knowledge of electronic coupling interactions between QDs and Au NPs should
777
facilitate the rational design of nanosystems that range from highly sensitive biodiagnostic 27 Page 27 of 63
probes to hierarchically structured optically active materials where, for example, QDs are
779
arranged around large Au NPs, or small Au NPs are arranged around QDs.
780
6. Biomolecule Supported FRET Relays
781
As noted in the Introduction, we define traditional FRET configurations as those between QDs
782
and organic dyes and occurring in a single step. Discrete assemblies of QDs and dyes that
783
incorporate multiple energy transfer steps, i.e., a relay or cascade, have been much less
784
frequently reported than their single-step counterparts; however, these non-traditional
785
configurations have capabilities that cannot be accessed in traditional configurations.
786
6.1. Long distance relays
787
While the range of energy transfer between a QD donor and dye acceptor is effectively limited to
788
< 1.5R0, the net distance over which energy is transferred can be increased by utilizing multiple,
789
successive energy transfer steps. Such systems with cascading energy transfer are referred to as
790
FRET relays due to the function of at least one dye as both an acceptor and donor (i.e., a ―relay‖
791
point). The first QD-FRET relay was reported by Medintz et al. in their seminal work on the use
792
of QDs and FRET for biosensing [156]. The relay was used to increase the efficiency of energy
793
transfer between an initial QD donor and terminal dye acceptor [156]. The terminal dye, Cy3.5,
794
was a label on cyclodextrin, which itself was situated in the binding pocket of MBP. The MBP
795
was self-assembled to QDs (peak PL at 530 nm) and was site-specifically labeled with a Cy3
796
relay point that could achieve ~95% quenching of QD PL at a 10:1 ratio of MBP:QD and 20%
797
net energy transfer efficiency to the terminal Cy3.5. Configurations without the intermediate dye
798
and FRET relay did not function [156]. Five years later, Lu et al. assembled a three-fluorophore
799
system that comprised CdSe/ZnS QDs (peak PL at 517 nm), enhanced yellow fluorescent protein
800
(EYFP), and Atto647 dye [157]. The QD served as an initial energy donor, with preferential
801
energy transfer to the EYFP (R0 = 3.9 nm) but viable energy transfer to the Atto647 (R0 = 3.2
802
nm). EYFP and Atto647 also formed a good FRET pair (R0 = 4.8 nm). The EYFP was site-
803
specifically modified with a single-stranded oligonucleotide using a heterobifunctional
804
crosslinker, and ~14 equivalents of the modified EYFP were self-assembled to the QD through a
805
polyhistidine tag. The conjugate was completed by hybridization of the oligonucleotide with its
806
Atto647-labeled complement. The dominant energy transfer pathway was the relay from the QD
Ac ce p
te
d
M
an
us
cr
ip t
778
28 Page 28 of 63
807
to the EYFP to the Atto647, with the net FRET efficiency reaching up to 5–17% for 13–9 nm
808
distances versus < 1% for direct transfer from the QD to Atto647 [157].
809 The two foregoing examples clearly show that even a single FRET relay step can improve
811
energy transfer efficiency over distances > 1.5R0. Recently, Boeneman et al. used QDs (peak PL
812
at 530 nm) as a central scaffold for the assembly of DNA-based photonic wires incorporating
813
multiple FRET relay steps (Fig. 5A) [158]. Excitation energy was initially transferred from the
814
QD and three subsequent energy transfer steps occurred down the length of the DNA ―wire‖
815
[158]. The wire was made by hybridizing four single strands of oligonucleotide, labeled with
816
either Cy3, Cy5, Cy5.5, or Cy7, in sequence, along a longer template strand. The template strand
817
was modified with a polyhistidine tag for self-assembly to CdSe/ZnS QDs. The dyes were
818
spaced ~3.3 nm apart and R0 values for sequential energy transfer steps ranged from 3.8–5.3 nm.
819
Energy could be transferred over distances > 15 nm, although it was found that the net end-to-
820
end efficiency was limited by the properties of the fluorescent dyes. Re-emission at the Cy5 and
821
Cy5.5 was limited to 2.2% and 1%, respectively, and the Cy7 effectively functioned as a dark
822
quencher (Fig. 5B-D). Nonetheless, the QD served as a superior light absorber and sensitizer for
823
initiating the FRET cascade along the photonic wire. Design improvements, such as selection of
824
brighter fluorescent dyes and optimized spacing of those dyes, can be expected to improve
825
stepwise efficiencies and permit substantial energy transfer over photonic wires 10–20 nm in
826
length.
827
6.2. Concentric relays
828
FRET relays have also been developed where cascaded energy transfer steps do not move the
829
excitation energy progressively further away from a central QD. Such configurations are referred
830
to as ―concentric‖ FRET relays, and their design is greatly facilitated by the ability to array
831
multiple fluorophores around a central QD. Donor or acceptor partners can be co-assembled at
832
similar distances and at separate attachment points so that multiple energy transfer steps occur
833
within a compact spherical volume that is concentric to the QD. Algar et al. have developed two
834
systems of this type for biosensing applications [71, 159, 160].
Ac ce p
te
d
M
an
us
cr
ip t
810
835 836
The first concentric FRET relay reported was based on energy transfer from a LTbC donor to a
837
CdSe/ZnS QD (peak PL at 620 nm), then from the QD to a deep red fluorescent dye, Alexa Fluor 29 Page 29 of 63
647 (A647). The Förster distances for these FRET pairs were quite large (LTbC-QD, R0 = 10.1
839
nm; QD-A647, R0 = 7.5 nm), affording energy transfer efficiencies > 90% (LTbC-QD) and
840
> 75% (QD-A647) when LTbC- and A647-labeled peptides were self-assembled to the QDs.
841
What is particularly interesting about this system is that it operated in two modes (Fig. 6A):
842
prompt and time-gated. A flash of UV light excited both the LTbC and QD efficiently. Prompt
843
energy transfer from the QD (τ ≈ 50 ns) to ground state A647 (τ ≈ 1 ns) occurred within < 250
844
ns, such that both the QD and FRET-sensitized A647 had emitted and returned to their ground
845
state prior to appreciable decay of the LTbC excited state (τ ≈ 2.6 ms). Time-gated energy
846
transfer from the LTbC to the QD, with relayed energy transfer to the A647, could then be
847
observed over a period > 1 ms following a time-delay of 55 µs between excitation and
848
acquisition of PL signals.
us
cr
ip t
838
an
849
One important advantage of the time-gated FRET relay was that it took information carried by
851
the QD-A647 FRET pair in the prompt time domain (< 10–7 s) and transferred it to the timescale
852
of 10–4–10–3 s. FRET pairs analogous to the QD-A647 have been widely utilized in various assay
853
configurations [25, 26], and time-gating is well-known to afford superior signal-to-background
854
ratios when analyzing complex biological samples [161, 162]. The time-gated FRET relay was
855
able to combine these two capabilities in a single nanoparticle vector. Assays for protease
856
activity were done by choosing the peptide linker between the QD and A647 to be a substrate for
857
a proteolytic enzyme, and by choosing the peptide linker between the LTbC and QD to be a non-
858
substrate. Monitoring the time-gated A647/QD PL ratio permitted tracking of proteolysis. A
859
second advantage of the time-gated FRET relay was the possibility of multiplexed assays and
860
biosensing with a single colour of QD vector. This capability is in contrast to the current state-of-
861
the-art, which is to use N colours of QD to detect N analytes [163]. In one multiplexed assay, two
862
target DNA sequences were detected by modifying the QD with oligonucleotide probes and
863
labeling the two target sequences with either LTbC or A647 [71]. Hybridization provided the
864
proximity for FRET with the QD. One target could be detected through the prompt A647/QD PL
865
ratio, while the second target could be detected through the total time-gated PL from the QD and
866
A647. In a second multiplexed assay, the LTbC and A647 were again attached to the QD via
867
peptide linkers; however, in this case, the peptides were respective substrates for two different
868
proteolytic enzymes (Fig. 6B-i) [159]. With appropriate cross-calibration (Fig. 6B-ii), the prompt
Ac ce p
te
d
M
850
30 Page 30 of 63
869
A647/QD PL ratio and the time-gated QD/LTbC PL ratio afforded orthogonal and quantitative
870
tracking of the two proteases (Fig. 6C). Multiplexed assays of proteolytic activity, multiplexed
871
inhibition assays, and pro-protease activation assays were possible in this format.
872 A second time-gated FRET relay was developed without a time-gating component. In this relay,
874
a CdSe/ZnS QD (peak PL at 520 nm) served as an initial donor and sensitizer rather than an
875
energy transfer intermediary [160]. Alexa Fluor 555 (A555) and A647 dyes were assembled
876
concentrically around the QD using peptide linkers (Fig. 7A). Energy could be transferred
877
directly from the QD to the A555 (R0 = 4.7 nm), or to the A647 (R0 = 3.0 nm), and from the
878
A555 to the A647 (R0 = 5.7 nm). Different combinations of A555 and A647 per QD resulted in
879
unique combinations of A555/QD and A647/QD PL ratios. These two ratios were mapped
880
against one another to create a two-dimensional parameter space that ultimately correlated PL
881
intensities with the number of A555- and A647-labeled peptides per QD. Multiplexed assays of
882
proteolytic activity and pro-protease activation (Fig. 7B) were again possible. Additionally,
883
analysis and modeling of the energy transfer efficiencies between the A555 and A647 at different
884
assembly ratios was used to gain insight into the interfacial distribution of self-assembled
885
peptides. The data was more consistent with approximately equidistant spacing rather than
886
random spacing [160]. While single-step FRET configurations have proven invaluable for
887
characterization of the radial dimensions of QD-bioconjugates, concentric FRET configurations
888
may permit further characterization of interfacial structure. Such capability is, of course, in
889
addition to multiplexed sensing in a single vector, which should be especially amenable to
890
visualization of coupled biochemical processes in cells and tissues.
891
7. FRET in Thin Films of Quantum Dots
892
Traditional and non-traditional QD-FRET assemblies dispersed in solution are especially useful
893
for applications in bioimaging and bioanalysis. Applications in optoelectronics, however, tend
894
more toward thin films of QDs or composite materials. FRET is both intrinsic to thin films of
895
QDs and can be engineered to enhance certain optical properties. This section discusses the
896
characterization of energy transfer in thin films of QDs, as well as the application to DSSCs and
897
other optoelectronic materials and devices. An example of utilizing thin films of QDs and FRET
898
for bioanalysis is reviewed elsewhere in this issue [164].
Ac ce p
te
d
M
an
us
cr
ip t
873
31 Page 31 of 63
7.1. Characterization of energy transfer
900
FRET can occur in thin films, solids, and composites of QDs whenever the QDs are at
901
sufficiently high density. The effect is subtle and appears as a small bathochromic shift in the PL
902
spectra of approximately monodisperse QDs when compared to bulk solution. This effect was
903
first reported by Kagan et al. for close-packed solids of CdSe QDs and is a consequence of the
904
inhomogeneous distribution of sizes (< 5%) with energy transfer from slightly smaller QDs to
905
slightly larger ones [165]. Crooker et al. were able to spectrally resolve PL quenching across an
906
inhomogeneous distribution of QDs (i.e. the width of the QD PL spectrum) and show the
907
smallest QDs were quenched by the greatest degree, with progressively less quenching as size
908
increased (longer emission wavelength) [166]. Analogous energy transfer also occurs when
909
different size distributions of approximately monodisperse QDs are mixed at sufficiently high
910
density (e.g., QDs with green and red PL), albeit that quenching of smaller donor QDs and PL
911
enhancement of the larger acceptor QDs is more easily resolved.
an
us
cr
ip t
899
M
912
In the years since the initial reports noted above, Bradley and coworkers have extensively
914
studied FRET between QDs in different thin film architectures [167-171]. The thin films were
915
made up of mercaptoacetic acid (MAA)-coated CdTe QDs assembled through layer-by-layer
916
(LbL) assembly with poly(sodium-4-styrene sulfonate) (PSS) and poly(diallyldimethyl-
917
ammonium chloride) (PDDA) on quartz substrates. In one study, a mixed monolayer of green
918
QD donors (peak PL at 524 nm) and orange QD acceptors (617 nm) were deposited and studied
919
[169]. The FRET efficiency was dependent on the ratio of acceptors to donors, reaching values
920
as high as 90% for a 2:1 ratio (cf. 59% for a 1:4.5 ratio). Both nearest-neighbour and long-range
921
FRET interactions were identified. The trends in FRET efficiency were explained by a 2D
922
energy transfer model that accounted for the dimensions of the QDs, imposing a minimum
923
center-to-center donor-acceptor separation or ―exclusion zone.‖ The FRET efficiency is given by
924
eqn. 16, where cA is the concentration of acceptors per unit area, and α and β are numerical
925
constants (see ref. [172]) related to the ratio Rex/R0, where Rex = RD + RA is the radius of the
926
exclusion zone (equal to the sum of the radii of the donor QD, RD, and acceptor QD, RA).
Ac ce p
te
d
913
(
)
(
EFRET = 1- a1 exp b1 R02cA - a 2 exp b2 R02cA
)
(16)
927
Interestingly, the foregoing study also found that energy transfer within the ensemble of
928
monodisperse acceptor QDs was significant while energy transfer within the ensemble of 32 Page 32 of 63
monodisperse donor QDs was not. Subsequent work revealed that the extent of intra-ensemble
930
FRET depended on the degree of inhomogenous broadening in that particular ensemble, which
931
was reflected by the PL FHWM and Stokes shift [168]. There was also a clear correlation with
932
density: as the concentration of monodisperse green QDs increased in a (sub-)monolayer film,
933
the observed bathochromic shift in their PL spectrum increased, the PL lifetime on the short
934
wavelength side of the QD PL spectrum decreased, and the difference in PL lifetime between the
935
short and long wavelength sides of the PL spectrum increased [168]. The FRET efficiency within
936
an approximately monodisperse film of QDs is given by eqn. 17, where r ≥ Rex = 2RQD, and the
937
numerical constant is defined for r in units of nm. 6
æ r ö n4 = 1+ ç è 0.0211÷ø 6k 2F D J ( l )
(17)
an
EFRET
-1
us
cr
ip t
929
Interestingly, intra-ensemble energy transfer within a monolayer of green QDs (peak PL at 547
939
nm) was found to affect the inter-ensemble energy transfer to a monolayer of orange QDs (peak
940
PL at 610 nm) [170]. Monolayers of the two QDs were separated by PSS/PDDA multilayers of
941
variable thickness (ca. 0.5–12.0 nm). While the donor-acceptor FRET efficiency increased as the
942
surface concentration of acceptors increased, the efficiency decreased as the surface
943
concentration of donor QDs increased due to enhancements in the competing donor-donor FRET
944
pathway. The latter behaviour was accounted for by introducing a concentration dependent donor
945
QD quantum yield term in 2D FRET models [170]. The FRET efficiency for this thin film
946
architecture is given by eqn. 18, where d is the distance between the parallel donor and acceptor
947
monolayers.
Ac ce p
te
d
M
938
EFRET -1 = 1+
948
2d 4 cAp R06
(18)
949
Bradley and coworkers have also studied surface plasmon resonance (SPR) enhanced FRET in
950
thin films of CdTe QDs [167, 171]. In one study, three monolayers of Au NPs (7–8 nm) with a
951
resonance at 565 nm were deposited on a quartz substrate. A mixed monolayer of green QDs
952
(peak PL at 525 nm) and orange QDs (peak PL at 600 nm) was then deposited on the substrate
953
with variable thicknesses (ca. 5.6–22 nm) of polyelectrolyte spacer layers between the Au NPs
954
and QDs [167]. A distance of ca. 14 nm between the QDs and Au NPs was found to provide a
955
2.7-fold enhancement of FRET between the two colours of QD. In a second study, an intervening 33 Page 33 of 63
layer of Au NPs was positioned 2 nm from a monolayer of QD donors (peak PL at 560 nm) and
957
12 nm from a monolayer of QD acceptors (peak PL at 615 nm) using polyelectrolyte spacers
958
[171]. Even with a ca. 23.4 nm center-to-center separation between the QD donors and
959
acceptors, a FRET efficiency of 8% was observed, representing an 80-fold increase in the rate of
960
energy transfer and an effective increase in the Förster distance from 3.9 nm to 7.9 nm [171]. In
961
both cases, the FRET enhancement was attributed to the surface plasmon dipole field enhancing
962
the electronic coupling between the donor and acceptor QDs. Similar effects have also been
963
observed by other researchers. For example, Wang et al. observed enhanced FRET between
964
CdSe/ZnS QD donors and CdSe/ZnTe QD acceptors drop cast with Au NPs (20 nm) on silicon
965
substrates [173].
us
cr
ip t
956
966
Beyond FRET between two colours of QD, graded or ―rainbow‖ thin films of QDs have also
968
been of interest [174]. These films stack layers of QDs from green to yellow to orange to red PL,
969
or vice versa, to create directional energy transfer. Franzl et al. have assembled and characterized
970
a LbL structure with a hierarchical funnel of QD colours (peak PL): green (515 nm)/yellow (540
971
nm)/orange (570 nm)/red (620 nm)/orange/yellow/green (Fig. 8A) [175]. Nearly complete
972
quenching of the green, yellow, and orange QD PL was observed with a large concomitant
973
enhancement in red QD PL. One layer of red QDs in the graded film had more than 4-times more
974
PL emission than a monochrome film with 7 layers of red QDs (equivalent optical density). The
975
graded structure increased the exciton density in the red QD layer by a factor of 28. The
976
enhancement in the graded structure is largely attributed to exciton recycling [175]. In closely
977
packed films, defect rich QDs can trap excitons from defect poor QDs, which is exacerbated by
978
increases in carrier trapping with above band-gap excitation [176]. The graded structure is
979
thought to permit transfer of trapped excitons to the luminescent states of larger QDs, in addition
980
to transfer between the luminescent states of smaller and larger QDs [175, 176]. Multilayer thin
981
films of this type have tremendous potential for harvesting light and transporting that energy to a
982
functional interface (vide infra)—a process conceptually analogous to the role of energy transfer
983
in nature‘s photosynthetic machinery.
984
7.2. FRET in quantum dot sensitized solar cells
985
As noted above, one application for FRET within thin films of QDs is solar energy harvesting.
986
Without a change in our primary energy sources, there will be an estimated 14 TW gap between
Ac ce p
te
d
M
an
967
34 Page 34 of 63
global energy supply and demand within 35–40 years [111]. While radiation from the sun
988
provides more than enough energy to meet present and future energy demands, current solar cell
989
technology is not cost effective enough for widespread use in preference to fossil fuels and other
990
non-renewable sources of energy [177]. DSSCs are a promising low-cost alternative to silicon
991
solar cells but are currently much less efficient [111, 177-179]. Some of the research efforts to
992
address this limitation of DSSCs have focused on the combination of QDs and FRET.
ip t
987
cr
993
In a typical DSSC, sunlight is absorbed by a dye coated onto a thin layer of nanocrystalline TiO2
995
supported on a transparent conducting electrode. Upon photoexcitation, the dye transfers an
996
electron to the TiO2, the residual hole is scavenged by a redox mediator (i.e., electrolyte) for
997
transport to the cathode, and the electrons entering the external circuit can do electrical work
998
[177, 178]. While the ideal sensitizing dye has strong, broad light absorption across the visible
999
and near-infrared (NIR) regions of the spectrum and exhibits efficient charge separation, most
1000
real dyes do not fully meet these criteria. For example, ruthenium PP complexes have broad
1001
absorption spectra but relatively low absorption coefficients [111, 179], whereas organic dyes
1002
may have large absorption coefficients but only over a narrow wavelength range [177, 179].
1003
Increasing the efficiency with which the full solar spectrum is harvested is thus one of the current
1004
challenges facing DSSCs [180, 181]. QDs can provide rapid charge separation, can be prepared
1005
at low cost, and, importantly, exhibit light absorption that is strong, broad, and tunable [179,
1006
182]. As such, QD sensitized solar cells (QDSSCs) are an active area of research.
an
M
d
te
Ac ce p
1007
us
994
1008
One strategy for preparing QDSSCs is to deposit a multilayer ―rainbow‖ film of QDs onto TiO2
1009
instead of a sensitizing dye. FRET occurs between the layers and enhances QDSSC performance.
1010
For example, Ruland et al. have reported a ―rainbow‖ solar cell design where four colours of
1011
CdTe QDs were sequentially arranged through LbL assembly on an indium-doped tin oxide
1012
(ITO) electrode (Fig. 8B) [181]. The QDs had first exciton absorption peaks at 486, 522, 554,
1013
and 580 nm, respectively, and corresponding PL maxima at 522 (green), 553 (yellow), 585
1014
(orange), and 620 nm (red). Photovoltaic performance was evaluated in cells with a polysulfide
1015
electrolyte and Cu2S counter electrode, using four layers of red QD as a reference configuration.
1016
A forward configuration, with the QDs deposited in order of increasing size (ITO/greenred),
1017
was found to have both higher spectral incident-photon-to-current conversion efficiencies
35 Page 35 of 63
(IPCEs; Fig. 8B-iv) and a 25% higher power conversion efficiency (η) than the reference
1019
configuration. The enhancements were attributed to FRET rapidly funneling energy toward the
1020
electrolyte interface where charge separation occurred via reduction of the electrolyte [181]. The
1021
efficiency of the FRET process was even sufficient to compensate for an unfavourable energy
1022
gradient for hole transport up the QD gradient to the ITO. A reverse configuration
1023
(ITO/redgreen) performed less favourably than the reference cell. In a similar study, Santra
1024
and Kamat deposited alloyed CdSeS QDs in a sequential architecture on TiO2 (Fig. 8C-i) [183].
1025
The QD PL emission was compositionally tuned to green, orange, or red without changing
1026
nanocrystal size (4.5 ± 0.1 nm). Cells prepared with a single layer (one colour) of QD had
1027
maximum IPCE values between 35–42% (Fig. 8C-ii) and η increased from 1.97% to 2.81% with
1028
decreasing band gap due to different levels of light harvesting. In contrast, two-colour bilayer
1029
films (orange/red) or three-colour trilayer films (green/orange/red) had η > 3.0% and maximum
1030
IPCE values between 37–51%. Cells prepared with a non-sequential mixture of the three colours
1031
of QD had η = 2.34%. The η values for the real bilayers and trilayers were 30–60% larger than
1032
predicted for simple additive effects from the QDs, a result that was attributed to cascaded FRET
1033
and/or electron transfer from larger band gap QDs to smaller band gap QDs [183].
cr
us
an
M
d
1034
ip t
1018
In addition to sequential ordering of QD band gap energies (i.e., colour), Choi et al. have found
1036
that the order of core versus core/shell QDs is important in QDSSCs [184]. Cationic or anionic
1037
core-only CdSe QDs with short ligands and CdSe/CdS/ZnS QDs with long ligands were
1038
prepared with nearly identical band gap energies and were LbL-assembled on TiO2. As per the
1039
design of the experiment, the core-only QDs exhibited 3-fold faster electron transfer to the TiO2
1040
than the core/shell/shell QDs. Consequently, when the core-only QDs were sandwiched between
1041
the TiO2 and core/shell/shell QDs in a cell (polysulfide electrolyte and CoS counter electrode),
1042
the observed photocurrent and η were > 3.6-fold larger than a configuration with the
1043
core/shell/shell QDs sandwiched between TiO2 and core-only QDs. This enhancement was
1044
attributed to ~86% efficient FRET from the core/shell/shell QDs to the core-only QDs, and
1045
subsequent rapid electron transfer from the core-only QDs to the TiO2. Although the core-only
1046
QDs were 1.3-times more efficient than the core/shell/shell QDs in single layer cells, the addition
1047
of a layer of core/shell/shell QDs (and FRET) onto the core-only QDs resulted in 340% and
1048
420% enhancements in photocurrent and η [184].
Ac ce p
te
1035
36 Page 36 of 63
1049 A second FRET strategy for sensitized solar cells has been to use QDs as donors for acceptor
1051
dyes that efficiently transfer charge to TiO2 in DSSCs. For example, squaraine dyes, which
1052
exhibit strong light absorption between 600–800 nm [185], have been widely utilized in this type
1053
of configuration. The underlying idea is that the QD absorbs visible light and transfers the energy
1054
to the squaraine dye via FRET for cascaded sensitization of the TiO2. The squaraine dye is also
1055
able to directly absorb red/NIR light and sensitize the TiO2. Zaban‘s group developed the
1056
strategy of depositing CdSe/CdS/ZnS QDs on nanocrystalline TiO2, overcoating the QDs with a
1057
layer of amorphous TiO2, and then adsorbing an asymmetric squaraine dye with deep red light
1058
absorption (ε = 319 000 M–1 cm–1) (Fig. 9A) [179, 186, 187]. The CdS/ZnS shell around the
1059
CdSe core served as a barrier for direct charge injection into the nanocrystalline TiO2 and the
1060
amorphous TiO2 layer served as a barrier to protect the QDs from the corrosive effects of an I–
1061
/I3– electrolyte. FRET between the QDs and squaraine dye was ca. 40–70% efficient depending
1062
on the thickness of the amorphous TiO2 layer, increasing as the film thickness increased from
1063
2 nm to 4 nm, then decreasing as the thickness increased up to 11 nm [187]. These trends were
1064
attributed to changes in the effective surface concentration of the dye and the average donor-
1065
acceptor separation distance. With a 4 nm thick film, an internal quantum efficiency of 82% was
1066
achieved, with 70% due to FRET and 12% due to reabsorption of QD PL emission by the dye
1067
[187]. The IPCE for the cells was 6–10% in the range of ca. 400–675 nm and, compared to cells
1068
with squaraine dye alone, offered a substantial increase in IPCE between 400–550 nm (Fig. 9B)
1069
[186].
cr
us
an
M
d
te
Ac ce p
1070
ip t
1050
1071
An alternative method of incorporating QD antennae into DSSCs for improved efficiency has
1072
been to use squaraine dye derivatives as linkers between the TiO2 electrode and QDs. In one
1073
study, Grätzel‘s group synthesized a squaraine dye with two carboxyl groups to bind to
1074
nanocrystalline TiO2 and two decyl chains to help adsorb hydrophobic CdSe QDs capped with
1075
tri-n-octylphosphine (Fig. 10A-i) [180]. The TiO2 electrode was modified with the dye by dip
1076
casting and subsequently spin coated with QDs. The cells, operated with a Co2+/Co3+ electrolyte,
1077
exhibited a higher and broader IPCE spectrum than cells prepared with QDs or squaraine dyes
1078
alone (Fig. 10A-ii), and η increased from 0.79% to 1.48%. The FRET efficiency between the
1079
QDs and squaraine dye was ca. 69% [180]. In another study, Kamat‘s group synthesized an
37 Page 37 of 63
asymmetric squaraine dye with a carboxylate linker to bind to TiO2 and an alkyl thiol linker to
1081
bind to CdSe QDs (Fig. 10B-i) [188]. The QD PL was quenched 80% due to FRET with the
1082
squaraine dye. Cells prepared with QDs or dye alone had η values of 0.15% and 3.05%,
1083
respectively. In contrast, the QD/squaraine hybrid cells had η = 3.65% and exhibited higher
1084
IPCE at wavelengths between 400–530 nm (Fig. 10B-ii).
ip t
1080
1085
Clearly, QD-FRET has significant potential for enhancing DSSC technology. Both ―rainbow‖
1087
thin films with cascaded energy transfer and composite films with QD-dye FRET pairs have
1088
demonstrated superior light harvesting capability, as shown by improved IPCE and η values. One
1089
may speculate that some combination of these two strategies may yet provide the greatest degree
1090
of enhancement and help move DSSCs closer to practical use.
1091
7.3. Luminescent films and composite materials
1092
Thin films of QDs are also of interest for the fabrication of light-emitting diodes (LEDs) and
1093
display technologies. For example, QD-LEDs have been commercialized in Sony‘s new
1094
Triluminos display technology. This section briefly reviews some the utility of QD-FRET in
1095
constructing LEDs and developing optically active materials that may have similar applications.
us
an
M
d
te
1096
cr
1086
Demir and coworkers have been proponents of using FRET within thin films of QDs for
1098
improving colour conversion of near-UV/blue LEDs [189-193]. Currently, broadband emitting
1099
phosphors are the preferred method for colour conversion to white light; however, this approach
1100
suffers from low colour rendering indices and poor tuning ability, the latter of which can be
1101
important for optimizing lighting for different conditions and applications (e.g., indoor vs.
1102
outdoor, room illumination vs. greenhouses) [194]. The narrow, tunable emission of QDs can
1103
produce pure colours much better than phosphors. One mechanism of utilizing QD-FRET within
1104
LED technology is to prepare mixed films of two colours of QD, which absorb light from an
1105
underlying near-UV InGaN/GaN LED with FRET between the two populations of QDs. The
1106
energy transfer has the effect of fine tuning colour and increasing quantum efficiency ca. 10-
1107
15% by recycling trapped excitons [189-191]. Demir and coworkers have also shown that
1108
adjusting the spacing between donor and acceptor QDs (peak PL at 595 and 645 nm) in bilayer
1109
assemblies can systematically shift chromaticity coordinates for the overall emission from the
1110
thin film [195]. QD-FRET is also a possible means by which to access efficient green and yellow
Ac ce p
1097
38 Page 38 of 63
LEDs. While alloyed InGaN/GaN and AlGaInP are optimal materials for blue and red LEDs,
1112
respectively, they both suffer from significantly decreased internal quantum efficiency and
1113
luminous efficiency when tuned to intermediate visible wavelengths. Bright yellow LEDs could
1114
be accessed using a mixed film of QDs with peak PL at 490 and 548 nm on a InGaN/GaN LED
1115
[191]. Most recently, Demir‘s group has also shown that FRET from an InGaN/GaN quantum
1116
well can also contribute to sensitization of a bilayer of green and red QDs (peak PL at 550 and
1117
600 nm) provided that the GaN cap is sufficiently thin (~ 3 nm) [192]. The efficiency of exciton
1118
migration from the quantum well was 83%, and the green/red QD bilayer structure enhanced
1119
exciton transfer efficiency by 64% compared to a bilayer of only red QDs.
us
cr
ip t
1111
1120
QD-FRET has also been exploited in developing novel luminescent materials. For example,
1122
Liang et al. fabricated ultrathin films with uniform morphology and controllable thickness based
1123
on the alternating assembly of layered double hydroxide (LDH) nanosheets with CdTe QDs
1124
[196]. A thin film with 40 LDH/green QD (peak PL at 535 nm) layers assembled with 24
1125
LDH/red QD (peak PL at 635 nm) layers exhibited 2-fold brighter red PL than 60 layers of
1126
LDH/red QD. This enhancement was attributed to FRET, which could be manipulated by
1127
controlling the order of green and red QD layers in the thin films. The FRET efficiency was 56%
1128
in thin films with five [LDH/green QD/LDH/red QD] units, and an increase in FRET efficiency
1129
to 91% was possible with three [LDH/green QD/(LDH/red QD)2] units. Both configurations
1130
provided substantially more (~7-fold) red QD PL than equivalent systems with only red QDs.
1131
Bednarkiewicz et al. prepared mixed films of upconverting Er3+/Yb3+:NaYF4 nanoparticle donors
1132
and CdSe QD acceptors [197]. When excited at 976 nm, the Er3+/Yb3+:NaYF4 nanoparticles
1133
exhibited green luminescence at 540 nm, which was resonant with the CdSe QDs (peak PL at
1134
580 nm). The FRET efficiency was only 14.8% due to the low quantum yield of the donors,
1135
resulting in a Förster distance of only 1.5 nm. Nonetheless, configurations of this type are
1136
interesting due to their potential to convert IR light into specific colours of visible light by
1137
selection of specific QDs. These and other QD-FRET-based luminescent materials could have
1138
applications in LED or other solid-state lighting devices.
Ac ce p
te
d
M
an
1121
39 Page 39 of 63
8. Summary and Perspective
1140
This review has described a variety of non-traditional FRET configurations incorporating QDs.
1141
We defined traditional QD-FRET pairs as those with QD donors and organic dye acceptors,
1142
where energy transfer occurred in a single step. One set of non-traditional QD-FRET
1143
configurations included single-step energy transfer with substitution of conventional organic
1144
dyes by photochromic dyes as acceptors, luminescent lanthanide complexes as donors,
1145
polypyridyl or azamacrocyclic metal complexes as acceptors, and Au NPs, GO, or CNTs as
1146
multidimensional acceptors. Additional non-traditional QD-FRET configurations included linear
1147
or concentric multi-step energy transfer relays that incorporated organic dyes and lanthanide
1148
complexes, as well as multicolour thin films and composite materials of QDs. Studies to date
1149
have ranged from experimental and theoretical characterization of these FRET systems, to
1150
applications such as multiplexed biosensing, preliminary work toward PDT, improved solar
1151
energy conversion, and fabrication of optoelectronic devices.
an
us
cr
ip t
1139
M
1152
Taken altogether, the non-traditional QD-FRET configurations discussed here comprise a highly
1154
versatile toolkit for the design and engineering of nanoscale devices. pcFRET provides the
1155
opportunity for optical switching and FRET between lanthanide complexes and QDs is
1156
potentially useful for timing optical signals. Both of these capabilities are critical to developing
1157
photonic logic circuits. ―Rainbow‖ thin films of QDs can efficiently downconvert short
1158
wavelength or broadband light to long wavelengths, while composite films of QDs with
1159
upconverting materials can achieve the opposite. FRET relays are also a means of transferring
1160
energy over length scales of 10–20 nm and Au NP or carbon nanomaterials can be used to
1161
quench excitation energy at similar distances. Furthermore, energy transfer from QDs to metal
1162
complexes can convert optical signals into chemical reactivity, as seen with the generation of
1163
singlet oxygen and nitric oxide. The inverse, conversion of chemical reactivity into optical
1164
signals, can be achieved through bioluminescence and chemiluminescence resonance energy
1165
transfer (BRET/CRET). Although beyond our scope here, an array of such BRET/CRET
1166
configurations with QDs have been demonstrated by Rao‘s group [35] and Willner‘s group [34,
1167
198-201], respectively. Albeit ambitious, one can imagine the future integration of many of these
1168
capabilities through top-down and bottom-up assembly to create complex nano-machines.
Ac ce p
te
d
1153
1169
40 Page 40 of 63
Although there is a predominance of traditional QD-FRET configurations and emphasis on
1171
biological applications in the literature, it is important to recognize that QD-FRET research is
1172
multidisciplinary and diverse. Herein, we have sought to highlight the capabilities and evolving
1173
role of QD-FRET beyond the typical norms. New, non-traditional QD-FRET configurations with
1174
novel properties will be able to help address challenges in healthcare, energy, and other areas of
1175
need.
1176
Acknowledgements.
1177
W.R.A. and H.K. acknowledge the Natural Sciences and Engineering Research Council of
1178
Canada (NSERC), the Canada Foundation for Innovation (CFI), the Peter Wall Institute for
1179
Advanced Studies, and the University of British Columbia for support of their research. W.R.A.
1180
is also grateful for a Canada Research Chair (Tier 2). I.L.M. acknowledges the U.S. Naval
1181
Research Laboratory Nanosciences Institute and the Defense Threat Reduction Agency Joint
1182
Science and Technology Office Military Interdepartmental Purchase Request # B112582M.
M
an
us
cr
ip t
1170
1183
Ac ce p
te
d
1184
41 Page 41 of 63
1185
References
1186
te
d
M
an
us
cr
ip t
[1] W.R. Algar, K. Susumu, J.B. Delehanty, I.L. Medintz, Anal. Chem. 83 (2011) 8826-8837. [2] A.M. Smith, S.M. Nie, Acc. Chem. Res. 43 (2010) 190-200. [3] A.P. Alivisatos, J. Phys. Chem. 100 (1996) 13226-13239. [4] S.J. Rosenthal, J.C. Chang, O. Kovtun, J.R. McBride, I.D. Tomlinson, Chem. Biol. 18 (2011) 10-24. [5] H. Mattoussi, G. Palui, H.B. Na, Adv. Drug. Deliver. Rev. 64 (2012) 138-166. [6] E. Petryayeva, W.R. Algar, I.L. Medintz, Appl. Spectrosc. 67 (2013) 215-252. [7] P. Zrazhevskiy, M. Sena, X.H. Gao, Chem. Soc. Rev. 39 (2010) 4326-4354. [8] V. Biju, T. Itoh, M. Ishikawa, Chem. Soc. Rev. 39 (2010) 3031-3056. [9] B.Y.S. Kim, J.T. Rutka, W.C.W. Chan, New Engl. J. Med. 363 (2010) 2434-2443. [10] Z. Jin, N. Hildebrandt, Trends Biotechnol. 30 (2012) 394-403. [11] R.A. Sperling, W.J. Parak, Phil. Trans. R. Soc. A 368 (2010) 1333-1383. [12] W.R. Algar, D.E. Prasuhn, M.H. Stewart, T.L. Jennings, J.B. Blanco-Canosa, P.E. Dawson, I.L. Medintz, Bioconjugate Chem. 22 (2011) 825-858. [13] C.B. Murray, C.R. Kagan, M.G. Bawendi, Annu. Rev. Mater. Sci. 30 (2000) 545-610. [14] N. Tomczak, D. Jańczewski, M. Han, G.J. Vancso, Progress Polymer Sci. 34 (2009) 393430. [15] D.V. Talapin, J.S. Lee, M.V. Kovalenko, E.V. Shevchenko, Chem. Rev. 110 (2010) 389458. [16] R.F.M. DeAlmeida, L.M.S. Loura, M. Prieto, Chem. Phys. Lipids 157 (2009) 61-77. [17] B. Schuler, W.A. Eaton, Curr. Opin. Struct. Biol. 18 (2008) 16-26. [18] V.V. Didenko, Biotechniques 31 (2001) 1106-1121. [19] J.P. Goddard, J.L. Reymond, Trends Biotechnol. 22 (2004) 363-370. [20] F. Ciruela, Curr. Opin. Biotechnol. 19 (2008) 338-343. [21] R. Roy, S. Hohng, T. ha, Nat. Methods 5 (2008) 507-516. [22] D.W. Piston, G.J. Kremers, Trends Biochem. Sci. 32 (2007) 407-414. [23] E.A. Jares-Erijman, T.M. Jovin, Nat. Biotechnol. 21 (2003) 1387-1395. [24] R. Freeman, I. Willner, Chem. Soc. Rev. 41 (2012) 4067-4085. [25] I.L. Medintz, H. Mattoussi, Phys. Chem. Chem. Phys. 165 (2009) 17-45. [26] W.R. Algar, A.J. Tavares, U.J. Krull, Anal. Chim. Acta 673 (2010) 1-25. [27] T. Förster, Naturwissenschaften 33 (1946) 166-175. [28] T. Förster, Discuss. Faraday Soc. 27 (1959) 7-17. [29] T. Förster, Intermolecular energy migration and fluorescence {Ann. Phys. 2, 55 (1948, translation)}, in: E.V. Mielczarek, E. Greenbaum, R.S. Knox (Eds.) Biological Physics, American Institute of Physics, New York, 1993. [30] T. Förster, J. Biomed. Opt. 17 (2012) 011002. [31] H.C. Cheung, Resonance energy transfer, in: J.R. Lakowicz (Ed.) Topics in Fluorescence Spectroscopy, Plenum Press, New York, 1991, pp. 127-176. [32] B.W. VanderMeer, G. Coker, S.Y.S. Chen, Resonance Energy Transfer: Theory and Data, VCH Publishers, New York, 1994. [33] L.J. Charbonnière, N. Hildebrandt, Eur. J. Inorg. Chem. (2008) 3241-3251. [34] R. Freeman, X.Q. Liu, I. Willner, J. Am. Chem. Soc. 133 (2011) 11597-11604.
Ac ce p
1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228
42 Page 42 of 63
te
d
M
an
us
cr
ip t
[35] Z.Y. Xia, J.H. Rao, Curr. Opin. Biotechnol. 20 (2009) 37-44. [36] A.L. Rogach, T.A. Klar, J.M. Lupton, A. Meijerink, J. Feldmann, J. Mater. Chem. 19 (2009) 1208-1221. [37] M. Tomasulo, I. Yildiz, F.M. Raymo, Inorg. Chim. Acta 360 (2007) 938-944. [38] K. Matsuda, M. Irie, J. Photochem. Photobiol. C 5 (2004) 169-182. [39] F.M. Raymo, M. Tomasulo, Chem. Soc. Rev. 34 (2005) 327-336. [40] F.M. Raymo, M. Tomasulo, Chem. Eur. J. 12 (2006) 3186-3193. [41] M. Irie, Chem. Rev. 100 (2000) 1685-1716. [42] L. Giordano, T.M. Jovin, M. Irie, E.A. Jares-Erijman, J. Am. Chem. Soc. 124 (2002) 74817489. [43] R. Klajn, J.F. Stoddart, B.A. Grzybowksi, Chem. Soc. Rev. 39 (2010) 2203-2237. [44] I.L. Medintz, S.A. Trammell, H. Mattoussi, J.M. Mauro, J. Am. Chem. Soc. 126 (2003) 3031. [45] L. Zhu, M.Q. Zhu, J.K. Hurst, A.D.Q. Li, J. Am. Chem. Soc. 127 (2005) 8968-8970. [46] M. Tomasulo, I. Yildiz, F.M. Raymo, Aust. J. Chem. 59 (2006) 175-178. [47] M. Tomasulo, F.M. Raymo, Inorg. Chim. Acta 360 (2007) 938-944. [48] E.M. Lee, S.Y. Gwon, Y.A. Son, S.H. Kim, Spectrochim. Acta A 97 (2012) 699-702. [49] E.M. Lee, S.Y. Gwon, Y.A. Son, S.H. Kim, Spectrochim. Acta A 97 (2012) 806-810. [50] E. Jares-Erijman, L. Giordano, C. Spagnuolo, K. Lidke, T.M. Jovin, Mol. Cryst. Liq. Cryst. 430 (2005) 257-265. [51] S.A. Díaz, G.O. Menéndez, M.H. Etchehon, L. Giordano, T.M. Jovin, E.A. Jares-Erijman, ACS Nano 5 (2011) 2795-2805. [52] S.A. Díaz, L. Giordano, T.M. Jovin, E.A. Jares-Erijman, Nano Lett. 12 (2012) 3537-3544. [53] Z. Erno, I. Yildiz, B. Gorodetsky, F.M. Raymo, N.R. Branda, Photochem. Photobiol. Sci. 9 (2010) 249-253. [54] B. Huang, M. Bates, X. Zhuang, Annu. Rev. Biochem. 78 (2009) 993-1016. [55] M. Heilemann, P. Dedecker, J. Hofkens, M. Sauer, Laser Photonics Rev. 3 (2009) 180-202. [56] T. Dertinger, R. Colyer, G. Iyer, S. Weiss, J. Enderlein, Proc. Natl. Acad. Sci. USA 106 (2009) 22287-22292. [57] K.A. Lidke, B. Rieger, T.M. Jovin, R. Heintzmann, Optics Express 13 (2005) 7052-7062. [58] A.P. deSilva, N.D. McClenaghan, Chem. Eur. J. 10 (2004) 574-586. [59] N. Sabbatini, M. Guardigli, J.M. Lehn, Coord. Chem. Rev. 123 (1993) 201-228. [60] J.C.G. Bünzli, C. Piguet, Chem. Soc. Rev. 34 (2005) 1048-1077. [61] I. Hemmilä, V. Laitala, J. Fluoresc. 15 (2005) 529-542. [62] I. Hemmilä, J. Biomol. Screen. 4 (1999) 303-307. [63] G. Mathis, Clin. Chem. 41 (1995) 1391-1397. [64] G. Mathis, J. Biomol. Screen. 4 (1999) 309-313. [65] N. Hildebrandt, L.J. Charbonnière, M. Beck, R.F. Ziessel, H.-G. Löhmannsröben, Angew. Chem. Int. Ed. 44 (2005) 7612-7615. [66] L.J. Charbonnière, N. Hildebrandt, R.F. Ziessel, H.-G. Löhmannsröben, J. Am. Chem. Soc. 128 (2006) 12800-12809. [67] A.R. Clapp, I.L. Medintz, B.R. Fisher, G.P. Anderson, H. Mattoussi, J. Am. Chem. Soc. 127 (2005) 1242-1250. [68] F. Morgner, D. Geißler, S. Stufler, N.G. Butlin, H.-G. Löhmannsröben, N. Hildebrandt, Angew. Chem. Int. Ed. 49 (2010) 7570-7574.
Ac ce p
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273
43 Page 43 of 63
te
d
M
an
us
cr
ip t
[69] K.D. Wegner, P.T. Lanh, T. Jennings, E. Oh, V. Jain, S.M. Fairclough, J.M. Smith, E. Giovanelli, N. Lequeux, T. Pons, N. Hildebrandt, ACS Appl. Mater. Inter. (2013) DOI: 10.1021/am3030728. [70] D. Geißler, L.J. Charbonnière, R.F. Ziessel, N.G. Butlin, H.-G. Löhmannsröben, N. Hildebrandt, Angew. Chem. Int. Ed. 49 (2010) 1396-1401. [71] W.R. Algar, D. Wegner, A.L. Huston, J.B. Blanco-Canosa, M.H. Stewart, A. Armstrong, P.E. Dawson, N. Hildebrandt, I.L. Medintz, J. Am. Chem. Soc. 134 (2012) 1876-1891. [72] H. Härmä, T. Soukka, A. Shavel, N. Gaponik, H. Weller, Anal. Chim. Acta 604 (2007) 177183. [73] M.J. Chen, Y.S. Wu, G.F. Lin, J.Y. Hou, M. Li, T.C. Liu, Anal. Chim. Acta 741 (2012) 100-105. [74] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. Vonzelewsky, Coord. Chem. Rev. 84 (1988) 85-277. [75] K. Kalyanasundaram, M. Gratzel, Coord. Chem. Rev. 177 (1998) 347-414. [76] J. Huang, D. Stockwell, Z. Huang, D.L. Mohler, T. Lian, J. Am. Chem. Soc. 130 (2008) 5632-5633. [77] M. Sykora, M.A. Petruska, J. Alstrum-Acevedo, I. Bezel, T.J. Meyer, V.I. Klimov, J. Am. Chem. Soc. 128 (2006) 9984-9985. [78] I.L. Medintz, T. Pons, S.A. Trammell, A.F. Grimes, D.S. English, J.B. Blanco-Canosa, P.E. Dawson, H. Mattoussi, J. Am. Chem. Soc. 130 (2008) 16745-16756. [79] A.J. Morris-Cohen, K.O. Aruda, A.M. Rasmussen, G. Canzi, T. Seideman, C.P. Kubiak, E.A. Weiss, Phys. Chem. Chem. Phys. 14 (2012) 13794-13801. [80] E.J. McLaurin, A.B. Greytak, M.G. Bawendi, D.G. Nocera, J. Am. Chem. Soc. 131 (2009) 12994-13001. [81] M.H. Stewart, A.L. Huston, A.M. Scott, A.L. Efros, J.S. Melinger, K.B. Gemmill, S.A. Trammell, J.B. Blanco-Canosa, P.E. Dawson, I.L. Medintz, ACS Nano 6 (2012) 5330-5343. [82] A.W. Hains, Z. Liang, M.A. Woodhouse, B.A. Gregg, Chem. Rev. 110 (2010) 6689-6735. [83] R. Bonnett, Chem. Soc. Rev. 24 (1995) 19-33. [84] L.M. Moreira, F.V. DosSantos, J.P. Lyon, M. Maftoum-Costa, C. Pacheco-Soares, N.S. daSilva, Aust. J. Chem. 61 (2008) 741-754. [85] S. Dayal, Y. Lou, A.C.S. Samia, J.C. Berlin, M.E. Kenney, C. Burda, J. Am. Chem. Soc. 128 (2006) 13974-13975. [86] S. Dayal, C. Burda, J. Am. Chem. Soc. 129 (2007) 7977-7981. [87] S. Dayal, J. Li, Y.S. Li, H. Wu, A.C.S. Samia, M.E. Kenney, C. Burda, Photochem. Photobiol. 84 (2008) 243-249. [88] S. Dayal, C. Burda, J. Am. Chem. Soc. 130 (2008) 2890-2891. [89] Y. Wen, W. Song, L. An, Y. Liu, Y. Wang, Y. Yang, Appl. Phys. Lett. 95 (2009) 143702. [90] M. Idowu, T. Nyokong, J. Luminesc. 129 (2009) 356-362. [91] M. Idowu, T. Nyokong, Polyhedron 28 (2009) 891-896. [92] A. Erdoğmuş, S. Moeno, C. Litwinski, T. Nyokong, J. Photochem. Photobiol. A 210 (2010) 200-208. [93] S. Moeno, E. Antunes, S. Khene, C. Litwinski, T. Nyokong, Dalton Trans. 39 (2010) 34603471. [94] J. Britton, E. Antunes, T. Nyokong, J. Photochem. Photobiol. A 210 (2010) 1-7. [95] N. Masilela, T. Nyokong, J. Photochem. Photobiol. A 247 (2012) 82-92. [96] M. Idowu, J.Y. Chen, T. Nyokong, New J. Chem. 32 (2008) 290-296.
Ac ce p
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319
44 Page 44 of 63
te
d
M
an
us
cr
ip t
[97] S. Moeno, T. Nyokong, J. Photochem. Photobiol. A 201 (2009) 228-236. [98] A.M. Smith, M.C. Mancini, S. Nie, Nat. Nanotechnol. 4 (2009) 710-711. [99] J.M. Tsay, M. Trzoss, L. Shi, X. Kong, M. Selke, M.E. Jung, S. Weiss, J. Am. Chem. Soc. 129 (2007) 6865-6871. [100] D.A. Tekdaş, M. Durmuş, H. Yanik, V. Ahsen, Spectrochim. Acta A 93 (2012) 313-320. [101] Z.D. Qi, D.W. Li, P. Jiang, F.L. Jiang, Y.S. Li, Y. Liu, W.K. Wong, K.W. Cheah, J. Mater. Chem. 21 (2011) 2455-2458. [102] P. Pacher, J.S. Beckman, L. Liaudet, Physiol. Rev. 87 (2007) 315-424. [103] G. Ferrer-Sueta, R. Radi, ACS Chem. Biol. 4 (2009) 161-177. [104] M. DeRidder, D. Verellen, V. Verovski, G. Storme, Nitric Oxide Biol. Chem. 19 (2008) 164-169. [105] B. Fitzpatrick, M. Mehibel, R.L. Cowen, I.J. Stratford, Nitric Oxide Biol. Chem. 19 (2008) 217-224. [106] P.C. Ford, Acc. Chem. Res. 41 (2008) 190-200. [107] D. Neuman, A.D. Ostrowski, R.O. Absalonson, G.F. Strouse, P.C. Ford, J. Am. Chem. Soc. 129 (2007) 4146-4147. [108] D. Neuman, A.D. Ostrowski, A.A. Mikhailovsky, R.O. Absalonson, G.F. Strouse, P.C. Ford, J. Am. Chem. Soc. 130 (2008) 168-175. [109] P.T. Burks, A.D. Ostrowski, A.A. Mikhailovsky, E.M. Chang, P.S. Wagenknecht, P.C. Ford, J. Am. Chem. Soc. 134 (2012) 13266-13275. [110] G. de la Torre, C.G. Claessens, T. Torres, Chem. Commun. (2007) 2000-2015. [111] M. Grätzel, Acc. Chem. Res. 42 (2010) 1788-1798. [112] W.M. Campbell, A.K. Burrell, D.L. Officer, K.W. Jolley, Coord. Chem. Rev. 248 (2004) 1363-1379. [113] L.L. Li, E.W.G. Diau, Chem. Soc. Rev. 42 (2013) 291-304. [114] H. Imahori, T. Umeyama, S. Ito, Acc. Chem. Res. 42 (2009) 1809-1818. [115] S. Jin, H.J. Son, O.K. Farha, G.P. Wiederrecht, J.T. Hupp, J. Am. Chem. Soc. 135 (2013) 955-958. [116] A. Boulesbaa, Z. Huang, D. Wu, T. Lian, J. Phys. Chem. C 114 (2010) 962-969. [117] A. Boulesbaa, A. Issac, D. Stockwell, Z. Huang, J. Huang, J. Guo, T. Lian, J. Am. Chem. Soc. 129 (2007) 15132-15133. [118] J. Huang, Z. Huang, Y. Yang, H. Zhu, T. Lian, J. Am. Chem. Soc. 132 (2010) 4858-4864. [119] I.L. Medintz, M.H. Stewart, S.A. Trammell, K. Susumu, J.B. Delehanty, B.C. Mei, J.S. Melinger, J.B. Blanco-Canosa, P.E. Dawson, H. Mattoussi, Nat. Mater. 9 (2010) 676-684. [120] X. Ji, G. Palui, T. Avellini, H.B. Na, C.Y. Yi, K.L. Knappenberger, H. Mattoussi, J. Am. Chem. Soc. 134 (2012) 6006-6017. [121] S.R. Opperwall, A. Divakaran, E.G. Porter, J.A. Christians, A.J. DenHartigh, D.E. Benson, ACS Nano 6 (2012) 8078-8086. [122] C. Curutchet, A. Franceschetti, A. Zunger, G.D. Scholes, J. Phys. Chem. C 112 (2008) 13336-13341. [123] C. Zhang, J. Xu, S. Zhang, X. Ji, Z. He, Chem. Eur. J. 18 (2012) 8296-8300. [124] H. Dong, W. Gao, F. Yan, H. Ji, H. Ju, Anal. Chem. 82 (2010) 5511-5517. [125] J. Tian, H. Zhao, M. Liu, Y. Chen, X. Quan, Anal. Chim. Acta 723 (2012) 83-87. [126] M. Liu, H. Zhao, X. Quan, S. Chen, X. Fan, Chem. Commun. 46 (2010) 7909-7911. [127] E. Morales-Narváez, B. Pérez-López, L.B. Pires, A. Merkoçi, Carbon 50 (2012) 29872993.
Ac ce p
1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365
45 Page 45 of 63
te
d
M
an
us
cr
ip t
[128] E. Shafran, B.D. Mangum, J.M. Gerton, Nano Lett. 10 (2010) 4049-4054. [129] S. Jander, A. Kornowski, H. Weller, Nano Lett. 11 (2011) 5179-5183. [130] Z. Chen, S. Berciaud, C. Nuckolls, T.F. Heinz, L.E. Brus, ACS Nano 4 (2010) 2964-2968. [131] S. Kaniyankandy, S. Rawalekar, H.N. Ghosh, J. Phys. Chem. C 116 (2012) 16271-16275. [132] D. Zhang, L. Gan, Y. Cao, Q. Wang, L. Qi, X. Guo, Adv. Mater. 24 (2012) 2715-2720. [133] I.V. Lightcap, P.V. Kamat, J. Am. Chem. Soc. 134 (2012) 7109-7116. [134] G. Mountrichas, A.S.D. Sandanayaka, S.P. Economopoulos, S. Pispas, O. Ito, T. Hasobe, N. Tagmatarchis, J. Mater. Chem. 19 (2009) 8990-8998. [135] S. Leubner, G. Katsukis, D.M. Guldi, Faraday Discuss. 155 (2012) 253-265. [136] E. Oh, M.Y. Hong, D. Lee, S.H. Nam, H.C. Yoon, H.S. Kim, J. Am. Chem. Soc. 127 (2005) 3270-3271. [137] B. Tang, L.H. Cao, K.H. Xu, L.H. Zhuo, J.H. Ge, Q.F. Li, L.J. Yu, Chem. Eur. J. 14 (2008) 3637-3644. [138] Y.P. Kim, Y.H. Oh, E. Oh, S. Ko, M.K. Han, H.S. Kim, Anal. Chem. 80 (2008) 46344641. [139] J. Liu, J.H. Lee, Y. Lu, Anal. Chem. 79 (2007) 4120-4125. [140] E. Dulkeith, A.C. Morteani, T. Niedereichholz, T.A. Klar, J. Feldmann, S.A. Levi, F.C.J.M. vanVeggel, D.N. Reinhoudt, M. Möller, D.I. Gittins, Phys. Rev. Lett. 89 (2002) 203002. [141] E. Dulkeith, M. Ringler, T.A. Klar, J. Feldmann, A. Muñoz Javier, W.J. Parak, Nano Lett. 5 (2005) 585-589. [142] J. Gersten, A. Nitzan, J. Chem. Phys. 75 (1981) 1139-1152. [143] T.L. Jennings, M.P. Singh, G.F. Strouse, J. Am. Chem. Soc. 128 (2006) 5462-5467. [144] M.P. Singh, G.F. Strouse, J. Am. Chem. Soc. 132 (2010) 9383-9391. [145] C.S. Yun, A. Javier, T. Jennings, M. Fisher, S. Hira, S. Peterson, B. Hopkins, N.O. Reich, G.F. Strouse, J. Am. Chem. Soc. 127 (2004) 3115-3119. [146] R.R. Chance, A. Prock, R. Silbey, Adv. Chem. Phys. 37 (1978) 1-65. [147] B.N.J. Persson, J. Phys. C. 11 (1978) 4251-4269. [148] B.N.J. Persson, N.D. Lang, Phys. Rev. B 26 (1982) 5409-5415. [149] Z. Gueroui, A. Libchaber, Phys. Rev. Lett. 93 (2004) 166108. [150] T. Pons, I.L. Medintz, K.E. Sapsford, S. Higashiya, A.F. Grimes, D.S. English, H. Mattoussi, Nano Lett. 7 (2007) 3157-3164. [151] M. Li, S.K. Cushing, Q. Wang, X. Shi, L.A. Hornak, Z. Hong, N. Wu, J. Phys. Chem. Lett. 2 (2011) 2125-2129. [152] H. Han, V. Valle, M.M. Maye, J. Phys. Chem. C 116 (2012) 22996-23003. [153] X. Zhang, C.A. Marocico, M. Lunz, V.A. Gerard, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, A.S. Susha, A.L. Rogach, A.L. Bradley, ACS Nano 6 (2012) 9283-9290. [154] J. Griffin, A.K. Singh, D. Senapati, P. Rhodes, K. Mitchell, B. Robinson, E. Yu, P.C. Ray, Chem. Eur. J. 15 (2009) 342-351. [155] R. Carminati, J.J. Greffet, C. Henkel, J.M. Vigoureux, Opt. Commun. 261 (2006) 368-375. [156] I.L. Medintz, A.R. Clapp, H. Mattoussi, E.R. Goldman, B. Fisher, J.M. Mauro, Nat. Mater. 2 (2003) 630-638. [157] H. Lu, O. Schöps, U. Woggon, C.M. Niemeyer, J. Am. Chem. Soc. 130 (2008) 4815-4827. [158] K. Boeneman, D.E. Prasuhn, J.B. Blanco-Canosa, P.E. Dawson, J.S. Melinger, M. Ancona, M.H. Stewart, K. Susumu, A. Huston, I.L. Medintz, J. Am. Chem. Soc. 132 (2010) 1817718190.
Ac ce p
1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411
46 Page 46 of 63
te
d
M
an
us
cr
ip t
[159] W.R. Algar, A.P. Malanoski, K. Susumu, M.H. Stewart, N. Hildebrandt, I.L. Medintz, Anal. Chem. 84 (2012) 10136-10146. [160] W.R. Algar, M.G. Ancona, A.P. Malanoski, K. Susumu, I.L. Medintz, ACS Nano 6 (2012) 11044-11058. [161] A.K. Hagan, T. Zuchner, Anal. Bioanal. Chem. 400 (2011) 2847-2864. [162] K. Hanaoka, K. Kikuchi, S. Kobayashi, T. Nagano, J. Am. Chem. Soc. 129 (2007) 1350213509. [163] W.R. Algar, U.J. Krull, Anal. Bioanal. Chem. 398 (2010) 2439-2449. [164] M.O. Noor, E. Petryayeva, A.J. Tavares, U. Uddayasankar, W.R. Algar, U.J. Krull, Coord. Chem. Rev. this issue. (2013). [165] C.R. Kagan, C.B. Murray, M.G. Bawendi, Phys. Rev. B 54 (1996) 8633-8643. [166] S.A. Crooker, J.A. Hollingsworth, S. Tretiak, V.I. Klimov, Phys. Rev. Lett. 89 (2002) 186802. [167] V.K. Komarala, A.L. Bradley, Y.P. Rakovich, S.J. Byrne, Y.K. Gun'ko, A.L. Rogach, Appl. Phys. Lett. 93 (2008) 123102. [168] M. Lunz, A.L. Bradley, W.Y. Chen, V.A. Gerard, S.J. Byrne, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, Phys. Rev. B 81 (2010) 205316. [169] M. Lunz, A.L. Bradley, W.Y. Chen, Y.K. Gun'ko, J. Phys. Chem. C 113 (2009) 30843088. [170] M. Lunz, A.L. Bradley, V.A. Gerard, S.J. Byrne, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, Phys. Rev. B 83 (2011) 115423. [171] M. Lunz, V.A. Gerard, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, A.S. Susha, A.L. Rogach, A.L. Bradley, Nano Lett. 11 (2011) 3341-3345. [172] P.K. Wolber, B.S. Hudson, Biophys. J. 28 (1979) 197-210. [173] C.H. Wang, C.W. Chen, Y.T. Chen, C.M. Wei, Y.F. Chen, C.W. Lai, M.L. Ho, P.T. Chou, M. Hofmann, Appl. Phys. Lett. 96 (2010) 071906. [174] A.A. Mamedov, A. Belov, M. Giersig, N.N. Mamedova, N.A. Kotov, J. Am. Chem. Soc. 123 (2001) 7738-7739. [175] T. Franzl, T.A. Klar, S. Schietinger, A.L. Rogach, J. Feldmann, Nano Lett. 4 (2004) 15991603. [176] T.A. Klar, T. Franzl, A.L. Rogach, J. Feldmann, Adv. Mater. 17 (2005) 769-773. [177] Y. Ooyama, Y. Harima, Chem. Phys. Chem. 13 (2012) 4032-4080. [178] X. Chen, C. Li, M. Grätzel, R. Kostecki, S.S. Mao, Chem. Soc. Rev. 41 (2012) 7909-7937. [179] S. Buhbut, S. Itzhakov, D. Oron, A. Zaban, J. Phys. Chem. Lett. 2 (2011) 1917-1924. [180] L. Etgar, J. Park, C. Barolo, V. Lesnyak, S.K. Panada, P. Quagliotto, S.G. Hickey, M.K. Nazeeruddin, A. Eychmüller, G. Viscardi, M. Grätzel, RSC Adv. 2 (2012) 2748-2752. [181] A. Ruland, C. Schulz-Drost, V. Sgobba, D.M. Guldi, Adv. Mater. 23 (2011) 4573-4577. [182] S. Rühle, M. Shalom, A. Zaban, Chem. Phys. Chem. 11 (2010) 2290-2304. [183] P.K. Santra, P.V. Kamat, J. Am. Chem. Soc. 135 (2013) 877-885. [184] S. Choi, H. Jin, J. Bang, S. Kim, J. Phys. Chem. Lett. 3 (2012) 3442-3447. [185] L. Beverina, P. Salice, Eur. J. Org. Chem. (2010) 1207-1225. [186] S. Buhbut, S. Itzhakov, E. Tauber, M. Shalom, I. Hod, T. Geiger, Y. Garini, D. Oron, A. Zaban, ACS Nano 4 (2010) 1293-1298. [187] S. Itzhakov, S. Buhbut, E. Tauber, T. Geiger, A. Zaban, D. Oron, Adv. Energy Mater. 1 (2011) 626-633. [188] H. Choi, P.K. Santra, P.V. Kamat, ACS Nano 6 (2012) 5718-5726.
Ac ce p
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457
47 Page 47 of 63
te
d
M
an
us
cr
ip t
[189] S. Nizamoglu, H.V. Demir, Opt. Express 16 (2008) 13961-13968. [190] S. Nizamoglu, H.V. Demir, Appl. Phys. Lett. 95 (2009) 151111. [191] S. Nizamoglu, E. Sari, J.H. Baek, I.H. Lee, H.V. Demir, IEEE J. Sel. Top. Quant. Elect. 15 (2009) 1163-1170. [192] S. Nizamoglu, P.L. Hernández-Martínez, E. Mutlugun, D.U. Karatay, H.V. Demir, Appl. Phys. Lett. 100 (2012) 241109. [193] H.V. Demir, S. Nizamoglu, T. Erdem, E. Mutlugun, N. Gaponik, A. Eychmüller, Nano Today 6 (2011) 632-647. [194] S. Nizamoglu, H.V. Demir, J. Appl. Phys. 105 (2009) 083112. [195] N. Cicek, S. Nizamoglu, T. Ozel, E. Mutlugun, D.U. Karatay, V. Lesnyak, T. Otto, N. Gaponik, A. Eychmüller, H.V. Demir, Appl. Phys. Lett. 94 (2009) 061105. [196] R. Liang, S. Xu, D. Yan, W. Shi, R. Tian, H. Yan, M. Wei, D.G. Evans, X. Duan, Adv. Func. Mater. 22 (2012) 4940-4948. [197] A. Bednarkiewicz, M. Nyk, M. Samoic, W. Strek, J. Phys. Chem. C 114 (2010) 1753517541. [198] R. Freeman, J. Girsh, A.F.J. Jou, J.A.A. Ho, T. Hug, J. Dernedde, I. Willner, Anal. Chem. 84 (2012) 6192-6198. [199] R. Freeman, B. Willner, I. Willner, J. Phys. Chem. Lett. 2 (2011) 2667-2677. [200] E. Golub, A. Niazov, R. Freeman, M. Zatsepin, I. Willner, J. Phys. Chem. C 116 (2012) 13827-13834. [201] X.Q. Liu, R. Freeman, E. Golub, I. Willner, ACS Nano 5 (2011) 7648-7655.
Ac ce p
1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481
48 Page 48 of 63
Figure Captions
us
cr
ip t
Figure 1. Photochromic FRET with QD donors. (A) pcFRET between a CdSe/ZnS QD and BIPS dye. The BIPS dyes are labels on maltose binding protein (MBP) self-assembled to the QD through a pentahistidine tail (5HIS). UV light converts the dye to the coloured merocyanine form to activate FRET; visible light converts the dye to the colourless spiropyran form to inactivate FRET. Reprinted with permission from ref. [44]. Copyright 2003 American Chemical Society. (B) Example of photochromic changes in spectral overlap between a CdTe QD and WSPO: (left) strong overlap with the merocyanine form; (right) minimal overlap with the spiropyran form. Reprinted with permission from ref. [48]. Copyright 2012 Elsevier. (C) pcFRET between a CdSe/ZnS QD and a photochromic dithienylethene dye: (i) the dyes were pendant groups on an amphiphilic polymer coating; (ii) photoswitching of QD PL. Figure reprinted with permission from ref. [51]. Copyright 2011 American Chemical Society.
te
d
M
an
Figure 2. FRET from luminescent Tb3+ complexes (LTbC) to QDs. (A) Assembly of LTbClabeled streptavidin (LS) to biotinylated QDs. (B) LTbC PL decay curves for (i) biotinylated QDs assembled at different ratios with LTbC-streptavidin (SAv; ~4.2 LTbC per SAv) and (ii) LTbC-peptides self-assembled to QDs at different ratios. The quenched LTbC decay component (a) and native LTbC decay component (b) are indicated in both panels. The native contribution increases as the relative amount of QDs decreases. (C) Spectral multiplexing with LTbC donor and five different QD acceptors: (i) LTbC emission and composite PL spectra for LTbC with QDs (peak PL at 529, 565, 604, 653, 712 nm); (ii) emission contributions from the QDs are highlighted. Panels (A) and (C) reprinted with permission from ref. [70]. Copyright 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. Panel (B) reprinted with permission from refs. [69, 71]. Copyright 2012 and 2013 American Chemical Society. Figure 3. FRET from QD donors to metal complexes. (A) FRET from a CdSe/ZnS QD donor to an Os(II)PP complex: (i) illustration of the conjugate, where the Os(II)PP complex is a label on a self-assembled peptide; (ii) progressive quenching of QD PL (inset) and sensitization of Os(II)PP complex PL as the number of labeled peptides per QD increases; (iii) wavelength dependent quenching of QD PL (blue and green curves) and the correlation with the Os(II)PP absorbance spectrum as evidence of FRET. Reprinted with permission from ref. [81]. Copyright 2012 American Chemical Society. (B) PL spectra showing quenching of CdTe QD PL and sensitization of PC PL due to FRET. The structure of the PC is shown in the inset, where M = Al(III). Reprinted with permission from ref. [100]. Copyright 2012 Elsevier. (C) FRET from a CdSe/ZnS QD to a cyclam complex of Cr(III), CrONO: (i) illustration of the conjugate, where the CrONO is adsorbed; (ii) mechanism of FRET-induced NO release from the complex (left) and comparison of NO production with and without QDs (right; measured with a NO-specific electrode). Reprinted with permission from refs. [107, 109]. Copyright 2007 and 2012 American Chemical Society.
Ac ce p
1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527
Figure 4. (A) Quenching efficiencies for QDs deposited on different carbon materials. GO is the most efficient quencher. Reprinted with permission from ref. [127]. Copyright 2012 Elsevier. (B) Energy transfer between QDs and Au NPs of different size: (i) double-stranded DNA linkage to control QD-Au NP separation; (ii) QD PL spectrum and absorption spectra of 3 nm, 15 nm, and 80 nm Au NPs; (iii) quenching efficiency versus donor-acceptor separation for the three sizes of
49 Page 49 of 63
Au NP and the expected behaviours for FRET and NSET mechanisms. Reprinted with permission from ref. [151]. Copyright 2011 American Chemical Society.
cr
ip t
Figure 5. Multi-step FRET relay along a DNA photonic wire. (A) Illustration of CdSe/ZnS QDs self-assembled with Cy3, Cy5, Cy5.5 and Cy7-labeled DNA. (B) Composite PL spectra for QDs self-assembled with DNA photonic wires with a progressive increase in the number of energy transfer steps. (C) Deconvolved PL spectra for the each individual chromophore when all are present along the DNA wire. (D) Composite PL spectra for DNA labeled with Cy3, Cy5, Cy5.5 and Cy7 dyes at the indicated positions and without a central QD scaffold. Reprinted with permission from [158]. Copyright 2010 American Chemical Society.
M
an
us
Figure 6. Time-gated concentric FRET relay and application to biosensing. (A) Temporal evolution of energy transfer from the QD to A647 dye (FRET2) and from an LTbC (Tb) to the QD (FRET1) in the time-gated FRET relay. (B) (i) QD bioconjugate with the A647 and LTbC assembled to the QD through peptide substrates for chymotrypsin (ChT) and trypsin (TRP), respectively; (ii) changes in prompt A647/QD PL ratio and time-gated QD/LTbC PL ratio with changes in the number of A647, n, and LTbC, m, per QD. (C) Multiplexed tracking of protease activity by (i) measuring the A647/QD and QD/LTbC PL ratios over time, and (ii) conversion to the amount of each hydrolyzed peptide product using the data in (B). The solid arrows indicate increasing concentrations of trypsin; the dashed arrows indicate increasing concentrations of chymotrypsin. Figure reprinted from ref. [159]. Copyright 2012 American Chemical Society.
te
d
Figure 7. Spectral concentric FRET relay and application to biosensing. (A) Illustration of the QD bioconjugate and FRET pathways where the A555 and A647 dyes are labels on peptides self-assembled to the QD. The peptides are selected to be substrates for the enzymes chymotrypsin (ChT) and trypsin (TRP), respectively. (B) Schematic showing changes in the concentric FRET relay during activation of pro-chymotrypsin to chymotrypsin by trypsin. (C) Multiplexed tracking of pro-protease activity: (i) changes in the A555/QD and A647/QD PL ratios over time; (ii) mapping of PL ratios to the number of A555 and A647 per QD; and (iii) conversion of PL data to progress curves for the hydrolysis of each peptide substrate. The arrows indicate increasing concentrations of trypsin; the pro-chymotrypsin concentration was constant. Reprinted from ref. [160]. Copyright 2012 American Chemical Society.
Ac ce p
1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
Figure 8. ―Rainbow‖ thin films and cascaded energy transfer for QDSSCs. (A) Graded multilayer thin film of CdTe QDs for exciton recycling: (i) absorption and emission spectra for the CdTe QDs; (ii) reference configuration (REF) and cascaded energy transfer (CET) configuration; (iii) enhancement of red QD PL and near complete quenching of green, yellow, and orange QD PL. Reprinted with permission from ref. [175]. Copyright 2004 American Chemical Society. (B) Illustration of (i) a solar cell and forward (FWD), reference (REF), and reverse (REV) configurations for a multilayer film of QDs; (ii) corresponding energy level diagrams; (iii) absorption and PL spectra; (iv) IPCE spectra showing superior performance for the forward configuration. Reprinted with permission from ref. [181]. Copyright 2011 WILEYVCH Verlag GmbH & Co. KGaA, Weinheim. (C) Illustration of (i) gradient structure of alloyed CdSeS QDs on TiO2 and its energy level diagram; (ii) IPCE spectra for various combinations of QD colours. Reprinted with permission from ref. [183]. Copyright 2013 American Chemical Society.
50 Page 50 of 63
ip t
Figure 9. Energy transfer from QDs embedded in amorphous TiO2 to squaraine dyes for QDSSC applications. (A) TEM image of a modified electrode (top left) and schematic of the electrode sensitization process: (i) light absorption and exciton generation in the QD; (ii) excitation of the dye via FRET; (iii) charge separation with injection into the nanocrystalline TiO2. (B) IPCE spectra for electrodes modified with QD, dye, and both QD and dye. Figure reprinted with permission from refs. [179, 186], Copyright 2010 and 2011 American Chemical Society, and ref. [187], Copyright 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
te
d
M
an
us
cr
Figure 10. Use of heterobifunctional squaraine dyes as linkers between TiO2 and CdSe QDs, showing (i) schematics and energy level diagrams; and (ii) IPCE spectra for TiO2 modified with CdSe QDs, squaraine dye, and both. Energy is transferred from the QDs to the dye, which injects an electron in the TiO2. (A) Carboxyl and alkyl chain modifications. Reprinted with permission from ref. [180]. Copyright 2012 Royal Society of Chemistry. (B) Carboxyl and thiol group modifications. Reprinted with permission from ref. [188]. Copyright 2012 American Chemical Society.
Ac ce p
1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
51 Page 51 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract
Page 52 of 63
*Highlights (for review)
Highlights
te
d
M
an
us
cr
ip t
Energy transfer between quantum dots and metal complexes Energy transfer between quantum dots and other nanomaterials Multi-step Förster resonance energy transfer relays and cascades Energy transfer in monolayer or multilayer thin films of quantum dots Applications in biosensing, solar energy conversion, optoelectronic devices
Ac ce p
Page 53 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 1
Page 54 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 2
Page 55 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3
Page 56 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4
Page 57 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 5
Page 58 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 6
Page 59 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 7
Page 60 of 63
Ac ce p
te
d
M
an
us
cr
ip t
Figure 8
Page 61 of 63
Ac ce p
te
d
M
an
us
cr
ip t
Figure 9
Page 62 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 10
Page 63 of 63
S0010-8545(13)00150-1 http://dx.doi.org/doi:10.1016/j.ccr.2013.07.015 CCR 111750
To appear in:
Coordination Chemistry Reviews
Received date: Revised date: Accepted date:
19-4-2013 21-7-2013 22-7-2013
Please cite this article as: W.R. Algar, H. Kim, I.L. Medintz, N. Hildebrandt, Emerging Non-Traditional F¨orster Resonance Energy Transfer Configurations with, Semiconductor Quantum Dots: Investigations and Applications, Coordination Chemistry Reviews (2013), http://dx.doi.org/10.1016/j.ccr.2013.07.015 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Manuscript
1
Emerging Non-Traditional Förster Resonance Energy Transfer Configurations with
2
Semiconductor Quantum Dots: Investigations and Applications
3 4 5
ip t
W. Russ Algar1,*, Hyungki Kim1, Igor L. Medintz2, Niko Hildebrandt3
6 1
9
Department of Chemistry
2036 Main Mall
11
Vancouver BC V6T 1Z1
12
Canada
13
Phone: 604-822-2464
14
Fax: 604-822-2847
15
[email protected]
an
10
us
University of British Columbia
M
8
cr
7
17
2
d
16
Center for Bio/Molecular Science and Engineering U.S. Naval Research Laboratory Code 6900
19
4555 Overlook Avenue SW
20
Washington DC 20375
21
USA
23
3
Ac ce p
22
te
18
Institut d'Electronique Fondamentale
24
Université Paris-Sud
25
91405 Orsay Cedex
26
France
27 28
*To whom correspondence should be addressed: [email protected] (W.R.A.)
1 Page 1 of 63
Abstract
30
Förster resonance energy transfer (FRET) configurations incorporating colloidal semiconductor
31
quantum dots (QDs) have proven to be a valuable tool for bioanalysis and bioimaging. Mirroring
32
well established techniques with only fluorescent dyes, ―traditional‖ FRET configurations with
33
QDs have involved single-step energy transfer to organic dye acceptors mediated by
34
biomolecular interactions. Here, we review recent progress in characterizing non-traditional
35
FRET configurations incorporating QDs and their application to challenges in biosensing, energy
36
conversion, and fabrication of optoelectronic devices. Such non-traditional FRET configurations
37
with QDs include substitution of organic dyes with lanthanide complexes, polypyridyl transition
38
metal complexes, azamacrocyclic metal complexes, graphene (oxide), carbon nanotubes, gold
39
nanoparticles, and dyes exhibiting photochromism. Other non-traditional configurations of
40
interest include FRET relays (with or without organic dyes) that feature multiple sequential
41
energy transfer steps, and thin films of QDs where discrete FRET pairs cannot be defined,
42
including those where QDs are layered in a size-sequential or ―rainbow‖ structure. The
43
calculation of FRET efficiencies and donor-acceptor distances in the above configurations are
44
reviewed, as are distance scaling relationships for non-zero dimensional acceptors, and the
45
related dipolar energy transfer mechanism, nanosurface energy transfer (NSET). To illustrate the
46
utility of non-traditional QD-FRET configurations, we highlight examples of optically
47
switchable probes, photonic wires, time-gated and multiplexed probes for biosensing, enhanced
48
light harvesting in QD and dye sensitized solar cells (DSSC), and colour conversion in light
49
emitting diodes (LEDs). We close by providing a perspective on how the combined utility of
50
these non-traditional QD-FRET configurations may be useful for engineering complex nanoscale
51
devices in the future.
53 54
cr
us
an
M
d
te
Ac ce p
52
ip t
29
Keywords: quantum dot, Förster resonance energy transfer (FRET), metal complex, biosensing, thin film, solar cell
55
2 Page 2 of 63
ip t
cr
us
an
M
d te
79
Table of Contents Abbreviations .................................................................................................................................. 4 1. Introduction ................................................................................................................................. 6 2. Quantum Dots and FRET............................................................................................................ 8 3. Photochromic FRET ................................................................................................................. 11 4. FRET with Metal Complexes ................................................................................................... 14 4.1. Lanthanide complexes ....................................................................................................... 14 4.2. Polypyridyl transition metal complexes............................................................................. 17 4.3. Azamacrocyclic metal complexes...................................................................................... 18 4.4. Distinguishing FRET from charge transfer........................................................................ 21 5. Energy Transfer with Other Nanomaterials .............................................................................. 23 5.1. Carbon nanomaterials ........................................................................................................ 24 5.2. Gold nanoparticles ............................................................................................................. 25 6. Biomolecule Supported FRET Relays ..................................................................................... 28 6.1. Long distance relays .......................................................................................................... 28 6.2. Concentric relays ............................................................................................................... 29 7. FRET in Thin Films of Quantum Dots ..................................................................................... 31 7.1. Characterization of energy transfer .................................................................................... 32 7.2. FRET in quantum dot sensitized solar cells ....................................................................... 34 7.3. Luminescent films and composite materials ...................................................................... 38 8. Summary and Perspective ......................................................................................................... 40 Acknowledgements. ...................................................................................................................... 41 References ..................................................................................................................................... 42
Ac ce p
56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
3 Page 3 of 63
Abbreviations
81
η
Power conversion efficiency
82
0D
Zero-dimensional
83
1D
One-dimensional
84
2D
Two-dimensional
85
2PE
Two-photon excitation
86
AFP
Alpha-fetoprotein
87
APC
Allophycocyanin
88
AOT
Dioctyl sulfosuccinate
89
Au NP
Gold nanoparticle
90
BIPS
1'-3-dihydro-1'-(2-carboxyethyl)-3,3-dimethyl-6-nitrospiro-[2H-1-benzopyran-
cr us
an
91
ip t
80
2,2'-(2H)-indoline] BRET
Bioluminescence resonance energy transfer
93
CNT
Carbon nanotube
94
CRET
Chemiluminescence resonance energy transfer
95
CrONO
trans-Cr(cyclam)(ONO)2+
96
CT
Charge transfer
97
DMPET
Dipole-to-metal particle energy transfer
98
dsDNA
Double-stranded DNA
99
DSSC
100
EYFP
101
FLIM
102
FRET
103
FWHM
104
GO
105
IPCE
106
ITO
Indium-doped tin oxide
107
LbL
Layer-by-layer
108
LDH
layered double hydroxide
109
LED
Light emitting diode
110
LLnC
Luminescent lanthanide complex
Ac ce p
te
d
M
92
Dye sensitized solar cell Enhanced yellow fluorescent protein Fluorescence lifetime imaging microscopy Förster resonance energy transfer Full-width-at-half-maximum Graphene oxide
Incident-photon-to-current conversion efficiency
4 Page 4 of 63
LSPR
Localized surface plasmon resonance
112
LTbC
Luminescent Tb3+ complex
113
MAA
Mercaptoacetic acid
114
MBP
Maltose binding protein
115
MOF
Metal organic framework
116
MPA
3-Mercaptopropionic acid
117
M-PC
Metallophthalocyanines
118
NO
Nitric oxide
119
NSET
Nanosurface energy transfer
120
PALM
Photoactivation localization microscopy
121
PC
Phthalocyanine
122
pcFRET
Photochromic FRET
123
PDDA
Poly(diallyldimethylammonium chloride)
124
PDT
Photodynamic therapy
125
PL
Photoluminescence
126
PP
Polypyridine
127
PSS
Poly(sodium-4-styrene sulfonate)
128
QD
Quantum dot
129
QDSSC
Quantum dot sensitized solar cell
130
SOFI
131
SPR
132
ssDNA
133
STORM
Stochastic optical reconstruction microscopy
134
TA
Transient absorption
135
WSPO
cr us
an
M
d
te
Ac ce p
136
ip t
111
Super-resolution optical fluctuation imaging Surface plasmon resonance Single-stranded DNA
Spironaphthoxazine dye
137
5 Page 5 of 63
1. Introduction
139
Colloidal semiconductor nanocrystals, better known as quantum dots (QDs), are prospective
140
materials for a wide variety of applications. QDs are characterized by a unique set of optical
141
properties that primarily arise from quantum confinement effects [1-4]. Foremost among these
142
properties is their bright photoluminescence (PL), the colour of which can be tuned on the basis
143
of nanocrystal size and composition. Other favourable optical properties include broad
144
absorption spectra with large one-photon (ε = 104–107 M–1cm–1) and two-photon (σ2PE = 103–104
145
GM) absorption cross-sections, large effective Stokes shifts (up to hundreds of nanometers),
146
spectrally narrow and symmetric PL emission (full-width-at-half-maximum, FWHM = 25–35
147
nm), and good resistance to photobleaching [1]. These properties have made QDs very attractive
148
probes for bioimaging and bioanalysis, where they are often touted for their multiplexing
149
capability and suitability for single particle visualization and tracking [1, 5-10]. Equally
150
important in such applications is the interface of the QD, which represents a nanoscale scaffold
151
for chemistry and functionalization. The physical properties of the QD can be tailored through
152
application of ligand or polymer coatings [11], and biological activity can be obtained through
153
bioconjugation [12]. The latter can include binding or other reactions with biomarkers, targeting
154
of cells and tissues, and delivery of therapeutics. In abiotic roles, QDs also represent building
155
blocks for optically active thin films, superlattice structures [13], and various composite
156
materials [14] for optoelectronic applications such as light emitting diodes (LEDs) and displays,
157
lasers, photodetectors, and solar cells [15].
cr
us
an
M
d
te
Ac ce p
158
ip t
138
159
QDs become even more powerful tools when their above-mentioned attributes are combined
160
with electron transfer or Förster resonance energy transfer (FRET) processes that can, for
161
example, modulate QD PL to generate active signaling or sensitize a secondary process. FRET is
162
perhaps best known as a spectroscopic tool for biophysical studies, including vesicle fusion and
163
membrane dynamics [16], protein folding [17], DNA detection [18], enzyme assays [19], ligand-
164
receptor and protein-protein interactions [20], both in the ensemble and at the single molecule
165
level [21-23]. In typical FRET experiments, a biomolecule is co-labeled with a donor
166
fluorophore and an acceptor chromophore, or, alternatively, two interacting biomolecules are
167
individually labeled with donor and acceptor. In the former case, conformational changes affect
6 Page 6 of 63
the fluorescence from the donor (and acceptor) due to the nanometric distance-dependence of
169
FRET; in the latter case, changes in fluorescence are due to association or dissociation. To a
170
large extent, these highly successful biological applications of FRET have inspired similar
171
developments that incorporate QDs and their unique optical properties as participants in FRET,
172
typically as donors. There is now a myriad of studies where QDs are used in either of the two
173
general FRET configurations noted above, and several comprehensive reviews have been written
174
on the utility of these configurations in bioimaging and bioanalysis [24-26].
cr
ip t
168
175
This review examines some of the recent work in the literature involving non-traditional FRET
177
configurations based on QDs. We use the term ―traditional‖ to refer to energy transfer in a single
178
step between a QD and an organic dye molecule, usually as a discrete pairing of a QD donor
179
with one or more equivalent organic dye acceptors in bulk solution. Here, we discuss QD-FRET
180
configurations that depart from this norm in one or more important ways. Deviations may
181
include the substitution of conventional organic dyes with other chromophores; for example,
182
lanthanide and other metal complexes, gold nanoparticles, carbon allotropes, or even organic
183
dyes with unusual properties, such as photochromism. Special attention is paid to the
184
characterization of energy transfer in these systems, including other mechanisms that may
185
supersede or compete with FRET. Moreover, by our definition, a departure from traditional QD-
186
FRET configurations is not limited to the choice of acceptor. Non-traditional configurations may
187
also include architectures where multiple energy transfer pathways are incorporated; for
188
example, as in the case of multistep sequential energy transfer, otherwise known as a FRET
189
―relay‖ or ―cascade.‖ Configurations of this nature have been realized with QD-bioconjugates in
190
solution and with thin films of QDs at an interface. The latter are also non-traditional FRET
191
configurations in that discrete donor-acceptor pairs cannot be defined. While the success of
192
traditional QD-FRET methods has been impressive, even greater capability and innovations are
193
expected from non-traditional QD-FRET configurations. As will be discussed, new capabilities
194
in bioanalysis and therapeutics, enhancements in solar energy conversion, and fabrication of
195
improved optoelectronic devices have already been demonstrated.
Ac ce p
te
d
M
an
us
176
196
7 Page 7 of 63
2. Quantum Dots and FRET
198
Prior to discussing non-traditional QD-FRET configurations, it is worthwhile to review the
199
fundamentals of FRET theory, including its ―traditional‖ application with QDs. FRET is a
200
resonant dipole-dipole coupling interaction that occurs through-space to transfer excitation
201
energy from a donor fluorophore to an acceptor chromophore. The energy transfer is
202
radiationless, occurring without the involvement of photons. Theodor Förster, after whom the
203
process is named, first elucidated the mechanism between 1946–1948 [27-30]. The genius of
204
Förster was that he was able to describe this process in terms of spectroscopic donor and
205
acceptor properties that could be measured through relatively simple experiments.
us
cr
ip t
197
206
The rate of energy transfer in FRET is given by eqn. 1, where τD–1 = k0 = kr + knr is the inverse
208
native lifetime of the donor excited state, kr is the rate of radiative relaxation, knr is the rate of
209
non-radiative relaxation (i.e., internal conversion and non-FRET quenching mechanisms), r is
210
the distance separating the donor and acceptor, and R0 is the Förster distance for the donor-
211
acceptor pair.
M
an
207
6
(1)
te
d
k FRET
1 æR ö = ç 0÷ tD è r ø
The Förster distance is defined by eqn. 2, where ΦD is the PL quantum yield of the donor, NA is
213
Avogadro‘s number, n is the refractive index of the medium, J(λ) is the spectral overlap integral,
214
and κ2 is the orientation factor [31].
Ac ce p
212
9(ln10)FDk 2 J(l ) R = = (8.79 ´10-28 mol) n-4FDk 2 J(l ) 5 4 128p N A n 6 0
(2)
215
The spectral overlap integral measures the degree of resonance between the donor emission
216
spectrum and acceptor absorption spectrum. It is defined by eqn. 3, where FD(λ) is the
217
wavelength-dependent fluorescence intensity of the donor, ε(λ) is the wavelength-dependent
218
molar absorption coefficient for the acceptor, and λ is the wavelength.
ò F (l )e (l )l J(l ) = ò F (l )d l D
A
4
dl
(3)
D
8 Page 8 of 63
219
To utilize eqn. 2 directly, the units for λ and ε(λ) are cm and cm2 mol–1 (=103 M–1 cm–1),
220
respectively. Its often more convenient to calculate the spectral overlap in units of nm and M–1
221
cm–1 for λ and ε(λ), in which case R0 = 0.02108 (2Dn–4J)1/6 in units of nm.
222 The orientation factor, 0 ≤ κ2 ≤ 4, accounts for the dependence of the dipole-dipole interaction on
224
their relative orientation. The orientation factor is calculated from eqn. 4, where θT is the angle
225
between the donor emission and acceptor absorption transition dipoles, and θD and θA are the
226
angles those transition dipoles make to the vector connecting them. While occasionally a
227
contentious point, κ2 = 2/3 tends to be assumed as a first approximation in most FRET studies.
228
This value is strictly true for isotropic and dynamically random orientations of the donor and
229
acceptor transition dipole moments during the donor excited state lifetime (in the presence of
230
acceptor). When these conditions cannot be reasonably assumed, determination of κ2 is
231
considerably more complex. Although such circumstances are beyond our scope here, more
232
detailed discussion can be found in van der Meer‘s text [32].
M
an
us
cr
ip t
223
233
2
(4)
d
k 2 = ( cosq T - 3cosq D cosq A )
The FRET efficiency, 0 ≤ EFRET ≤ 1, is given by eqn. 5, which simplifies to the terms on the
235
right-hand side for a discrete donor-acceptor pair. It can be seen from this equation that R0
236
represents the donor-acceptor separation at which the FRET efficiency is 50%.
Ac ce p
te
234
kFRET R06 EFRET = = kr + knr + kFRET R06 + r 6
(5)
237
Experimentally, the FRET efficiency can be measured from quenching of donor emission,
238
eqn. 6, where the subscript D denotes an isolated donor quantity, the subscript DA denotes a
239
donor quantity measured in the presence of acceptor, I is a fluorescence intensity, and <τ> is an
240
amplitude-weighted average lifetime.
EFRET = 1-
t I DA = 1- DA ID tD
(6)
241 242
As discussed elsewhere, the applicability of the FRET formalism with QDs has been validated
243
by many studies [25, 26]. Donor-acceptor separations are measured from the center of the QD,
9 Page 9 of 63
such that its radius imposes a minimum separation. It is usually assumed that κ2 = 2/3 for QD
245
donors due to their two-fold degenerate (circular) transition dipole, the rotational freedom of
246
proximal dyes, and the random assembly of those dyes to the QD surface. As donors, QDs are
247
advantageous in that their narrow, continuously tunable emission (by size or composition)
248
permits optimization of the spectral overlap integral while also minimizing crosstalk in
249
measurements of donor and (fluorescent) acceptor PL. Their broad absorption spectra also
250
provide flexibility in choice of excitation wavelength so that direct excitation of an acceptor dye
251
can be minimized. The other general advantages of QDs—their brightness, resistance to
252
photobleaching, and more facile multiplexing—are also relevant to FRET configurations.
253
Moreover, the large surface area of QDs permits arraying of multiple acceptors per QD, thereby
254
providing another mechanism to optimize FRET efficiencies. For the case of N equivalent
255
acceptors around a central QD donor, eqn. 5 can be modified to eqn. 7, where it is seen that
256
FRET efficiency increases as the number of acceptors increases. Given the approximately
257
spherical shape of QDs, such arrangements are easily accessed in real experiments.
M
an
us
cr
ip t
244
(7)
d
NR06 EFRET = NR06 + r 6
QDs are also excellent acceptors for certain special classes of donors (e.g., lanthanides [33],
259
chemiluminophores [34], and bioluminescent proteins [35]). Here, the primary advantage of QDs
260
is their absorption spectra, which lead to very large spectral overlap integrals and thus Förster
261
distances. Arraying multiple donors per QD also increases the probability of energy transfer. The
262
probability of FRET-sensitization of a central QD acceptor by M donors is given by eqn. 8. Note
263
that this expression does not count the number of energy transfers to a QD from M donors, but
264
rather the probability of at least one such event occurring.
Ac ce p
te
258
æ r6 ö PA = 1- ç 6 6 ÷ è R0 + r ø
M
(8)
265
More detailed information on traditional QD-FRET can be found in recent review articles [25,
266
26, 33, 36].
267
10 Page 10 of 63
3. Photochromic FRET
269
Photochromic dyes exhibit reversible colour changes with optical stimulation. The change in
270
absorption spectra is accompanied by changes in molecular and electronic structure, including
271
photoinduced ring opening or closing, cis/trans isomerism, cycloaddition, and proton or electron
272
transfer reactions [37-39]. The photogenerated states typically undergo thermal reversion back to
273
the initial state. Photochromic dyes are candidate materials for applications requiring or
274
benefiting from optical switching; for example, data storage, sensing, and super-resolution
275
imaging [39-41]. Research on photochromism has included both photochromic FRET (pcFRET),
276
where energy transfer is modulated ‗on‘ and ‗off‘ through photoinduced changes in the spectral
277
overlap integral [42], and the design of photochromic nanoparticle constructs [43]. This section
278
reviews studies on pcFRET with QD donors, which have largely made use of
279
spiroindolinobenzopyran and diheteroarylethene dyes as photochromic acceptors. In traditional
280
QD-dye FRET pairs, the spectral overlap integral is generally constant. In contrast, when using
281
photochromic acceptor dyes, the spectral overlap integral is dynamic and optically controllable.
M
an
us
cr
ip t
268
282
The first instance of pcFRET with QD donors was reported by Medintz et al. [44], who labeled
284
maltose binding protein (MBP) with BIPS dye (1',3-dihydro-1'-(2-carboxyethyl)-3,3-dimethyl-6-
285
nitrospiro-[2H-1-benzopyran-2,2'-(2H)-indoline]). In its spiropyran form, BIPS is nearly
286
colourless, exhibiting absorption bands at wavelengths < 400 nm. Upon UV irradiation, BIPS
287
photoconverts to its coloured merocyanine form with an absorption band at ca. 550 nm. When
288
polyhistidine-tagged MBP was labeled with BIPS (5 dyes/MBP) and self-assembled to
289
CdSe/ZnS QDs (20 MBP/QD; see Fig. 1A), FRET from the QD to the BIPS could be reversibly
290
modulated with ca. 60% FRET efficiency. The QD PL was centered at 555 nm and resonant with
291
the absorption peak of the merocyanine form. The broad absorption spectrum of the QD
292
permitted its excitation and interrogation of the pcFRET process at a wavelength minimum
293
common to both the spiropyran and merocyanine forms of BIPS, thereby avoiding
294
photoconversion of the BIPS to one state while measuring the effect of the other state.
Ac ce p
te
d
283
295 296
Other groups have also studied pcFRET between QDs and spiroindolinobenzopyran dyes. For
297
example, Zhu et al. directly conjugated a thiol-linker modified BIPS derivative, 5-(1,3-dihydro-
298
3,3-dimethyl-6-nitrospiro[2H-1-benzopyran-2,2'-(2H)-in-dole])hexane-1-thiol with CdSe/ZnS
11 Page 11 of 63
QDs and demonstrated photoswitching of FRET [45]. Similarly, Tomasulo et al. investigated
300
pcFRET between CdSe/ZnS and CdS QD donors and a BIPS derivative with a dithiolane linker
301
[46]. With CdSe QDs, 45% PL quenching efficiency was observed upon photoswitching from
302
the spiropyran to the merocyanine form [46]. In the case of the CdS QDs, the method of
303
preparation of the CdS QDs was found to impact photochromism [47]. Two CdS QD samples,
304
both coated with 1-decanethiol, but one prepared via a solvothermal synthesis with TOPO
305
ligands and the other prepared in dioctyl sulfosuccinate (AOT) inverse micelles, were found to
306
induce positive and negative photochromism, respectively, when BIPS was conjugated. These
307
results were consistent with the known sensitivity of BIPS photochromism to the polarity of the
308
local microenvironment, and the negative photochromism was attributed to electrostatic
309
interactions with surface defects on the QDs synthesized from AOT inverse micelles. In a recent
310
study, Lee et al. adsorbed a spironaphthoxazine dye (WSPO) on aqueous CdTe QDs coated with
311
3-mercaptopropionic acid (MPA) and observed six cycles of pcFRET with efficiencies >80%
312
upon conversion to the merocyanine form [48]. The difference in spectral overlap with the QD
313
between the spiropyran and merocyanine forms is shown in Fig. 1B. These authors have also
314
prepared photochromic composite materials with CdTe QDs, WSPO, and poly(N-
315
isopropylacrylamide) [49].
cr
us
an
M
d te
316
ip t
299
Jares-Erijman et al. demonstrated pcFRET from a CdSe/ZnS QD donor to a photochromic
318
diheteroarylethene dye acceptor of the dithienylethene family [50]. The dye was synthesized as a
319
biotin derivative and assembled to QDs modified with streptavidin. Similar to BIPS, the closed
320
form of the dithienylethene dye had a visible absorption band in resonance with the QD (peak PL
321
at 565 nm) to generate FRET. QD PL quenching could be reversibly modulated over 14
322
complete cycles with ~30% FRET efficiency. A heterodiarylethene dye was chosen because this
323
family generally exhibits more robust photochromic properties (i.e., less switching fatigue) than
324
spiropyran dyes. Jares-Erijman and coworkers also recently synthesized amphiphilic polymer
325
coatings for CdSe/ZnS and CdSe/CdS/ZnS QDs with pendant dithienylethene dyes sequestered
326
within the hydrophobic layer to provide an optimal microenvironment for photoconversion (Fig.
327
1C) [51, 52]. In one of these designs, there was an average of ~35 dithienylethene dyes per QD
328
and approximately one non-photochromic red fluorescent dye per QD as an internal standard.
Ac ce p
317
12 Page 12 of 63
329
Photoswitching was possible over 15 complete cycles with alternating illumination at 340 nm
330
and 545 nm and two-fold changes in QD PL intensity.
331 Similar to the examples above, Erno et al. also found that a dithienylethene could engage in
333
pcFRET with CdSe/ZnS QDs; however, these authors observed switching fatigue after just three
334
cycles [53]. This behaviour was attributed to electron transfer competing with FRET and causing
335
the irreversible reduction of the dye. It is important to note that the chemistry in this system was
336
quite different from that used by Jares-Erijman. Hydrophobic QDs were directly modified with a
337
large excess of the dithienylethene dye, such that strong quenching of QD PL was observed with
338
both the open (~86%) and closed forms (~96%) of the dyes. Efficient and robust photoswitching
339
of QD PL via pcFRET should be possible with carefully designed interfaces between the QD and
340
photochromic dye.
an
us
cr
ip t
332
341
Several super-resolution imaging techniques, e.g., stochastic optical reconstruction microscopy
343
(STORM) and photoactivation localization microscopy (PALM), require that fluorescent dyes or
344
proteins switch between bright and dark states [54, 55]. Another technique, super-resolution
345
optical fluctuation imaging (SOFI), developed by Dertinger et al., requires reversibly switchable
346
emitters with optically distinguishable states but not necessarily bright and dark states [56]. QDs
347
that can be reversibly quenched through pcFRET are potential probes for these and similar super-
348
resolution methods. The optical properties of QDs offer several benefits in diffraction-limited
349
imaging and may be similarly amenable to super-resolution imaging. In this context, remaining
350
challenges include enhancing the efficiency of energy transfer, optimizing photoconversion in
351
QD-photochromic dye conjugates, minimizing thermal reversion, and evaluating the stochastic
352
optical switching of individual QDs assembled with multiple photochromic acceptor dyes. While
353
the blinking or PL intermittency of QDs has already been used to enable super-resolution
354
imaging [56, 57], pcFRET may be able to offer a greater degree of control. Other potential
355
applications for QD-pcFRET pairs include use as probes for fluorescence lifetime imaging
356
microscopy (FLIM) or sensitive lock-in detection due to the ability to controllably modulate QD
357
PL lifetime and intensity [52]. QD-pcFRET pairs may also prove useful as switchable elements
358
for molecular photonic logic devices [43, 58].
Ac ce p
te
d
M
342
359
13 Page 13 of 63
4. FRET with Metal Complexes
361
The photochromic acceptor dyes discussed in the previous section are examples of organic dyes
362
with special properties, but are organic dyes nonetheless. This section describes FRET between
363
QDs and lanthanide or other metal complexes. Lanthanide complexes are interesting due to their
364
very unique optical properties, which, when paired with QDs, comprise a fascinating non-
365
traditional FRET system. Other metal complexes, and particularly those with transition metal
366
centres, are interesting due their ability to engage in both charge transfer (CT) and FRET with
367
QDs. The CT mechanism is often more intuitive and distinguishing between FRET and CT is not
368
always straightforward (see Section 4.4). The pairing of a QD with a metal complex can thus be
369
a non-traditional FRET configuration due to the observation of FRET in preference to CT. In
370
addition, these configurations can be non-traditional due to the ability of that FRET process to
371
sensitize subsequent chemical reactions at the metal centre.
372
4.1. Lanthanide complexes
373
Two unique optical properties of trivalent lanthanide ions are their narrow, line-like emission
374
spectra and their remarkably long excited state lifetimes, which are on the order of microseconds
375
to milliseconds [59, 60]. Luminescence arises from parity and spin-multiplicity forbidden 4f-4f
376
transitions that are shielded from the surrounding environment by the 5s2 and 5p6 sub-shells.
377
Since the 4f-4f transitions are very weak (ε < 3 M–1 cm–1 [60]), luminescence is not practically
378
observed from isolated lanthanide ions. Rather, lanthanide ions are complexed with organic
379
ligands that serve as antennae to sensitize their luminescence. The ligands, which have non-
380
trivial molar absorption coefficients (~104 M–1 cm–1), are generally considered to sensitize
381
lanthanide luminescence through a multi-step process: (i) light absorption and excitation to the
382
singlet excited state of the ligand; (ii) intersystem crossing to the triplet state of the ligand; (iii)
383
energy transfer from the ligand to the lanthanide ion; and (iv) emission from the lanthanide ion
384
(in competition with non-radiative relaxation processes) [59, 60]. The long excited state lifetimes
385
have made lanthanide ions particularly attractive for time-resolved fluorescent assays with ultra-
386
low background, especially with FRET to allophycocyanin (APC) [61-64], and, more
387
pertinently, good donors for QD acceptors in FRET [33].
Ac ce p
te
d
M
an
us
cr
ip t
360
388 389
The first work pairing luminescent lanthanide complexes (LLnCs) with QDs was reported by
390
Hildebrandt and coworkers [65, 66]. Streptavidin conjugates of Eu3+ and Tb3+ complexes were 14 Page 14 of 63
titrated with biotinylated CdSe/ZnS QDs (Fig. 2A) and increasing amounts of time-gated QD PL
392
were observed. Excitation was at 308 nm to be resonant with the absorption band of the
393
lanthanide complexes and PL was collected with a 250 µs delay after pulsed excitation and
394
integrated over 750 µs. What was particularly noteworthy about this configuration was that the
395
QDs served as energy acceptors for the LLnC donors. In traditional FRET configurations with
396
organic dyes, QDs are rather poor acceptors. Due to their broad and intense absorption spectra,
397
QDs are always directly excited with the dye donor; however, only the ground state of the QD is
398
a good acceptor. The comparatively longer excited state lifetime of a QD (>10 ns vs. < 5 ns for
399
most dyes) is such that only a trivial fraction of the population is able to return to its ground state
400
and effectively serve as an acceptor for a proximal excited state dye [67]. In contrast, the 340–
401
680 µs and 1.48 ms luminescence lifetimes of the Eu3+ and Tb3+ complexes, respectively,
402
provided the opportunity for the QDs to relax to their ground state (τQD < 100 ns) while a
403
majority of the lanthanides remained in their excited state [66]. These conditions permit efficient
404
energy transfer to a QD acceptor. A hallmark of LLnC-to-QD FRET is that the QD acquires an
405
apparent PL lifetime on the same order of magnitude as the LLnC since the rate of energy
406
transfer is much slower than the decay rate of the QD [66, 68] .
d
M
an
us
cr
ip t
391
407
One outcome of the use of QDs as acceptors for LLnC donors is access to very large Förster
409
distances (R0 ≈ 7–11 nm) due to the large molar absorption coefficients and broad absorption
410
spectra of QDs [65, 66, 68-70]. Larger Förster radii lead to more efficient FRET, which
411
translates into more sensitive analyses and/or compatibility with nanoscale architectures of larger
412
dimensions. Importantly, Hildebrandt and coworkers have shown that energy transfer from
413
luminescent Tb3+ complexes (LTbCs) to QDs can be utilized as a ―molecular ruler‖ analogous to
414
other FRET pairs; the PL decays of the LTbC and QD can be analyzed to extract distance
415
information such as the size and shape of non-spherical QDs [68, 69]. This analysis is done on
416
the basis of PL decays. Ideally, the LTbC donor intensity decays (denoted D) is fit to a single
417
exponential function, eqn. 9, where τD is the unquenched native lifetime of the LTbC donor, AD
418
is the total amplitude, and t is time.
Ac ce p
te
408
I D (t) = AD exp(-t / t D )
(9)
419
A contribution from the native LTbC decay (i.e., unquenched donor) is usually observed even in
420
the presence of QD acceptors (Fig. 2B). The PL intensity decay for LTbC-QD FRET pairs 15 Page 15 of 63
421
(denoted DA) is thus fit with a multiexponential function with N ≥ 2 components, one of which
422
is the unquenched LTbC decay, as shown in eqn. 10. N -1 é ù I DA (t) = ADA êa D exp(-t / t D ) + åa DA,i exp(-t / t DA,i ) ú i=1 ë û
(10)
The average FRET efficiency can be calculated from eqn. 11, where <τDA> is the average decay
424
time for the LTbC in the presence of QD acceptor (without the native LTbC decay), and related
425
to an average donor-acceptor distance.
cr
t DA 1 é åa DA,it DA,i ù = 1- ê ú tD t D êë åa DA,i úû
(11)
us
E = 1-
ip t
423
Alternatively, the individual FRET-quenched LTbC lifetime components may be used to
427
calculate distance components. Hildebrandt et al. have used this approach to assess the effective
428
size and shape of functionalized QDs in aqueous environments (cf. the artificial environment
429
used in electron microcopy) [69]. In some experiments, a strictly monoexponential decay will
430
not be observed for LTbC donor alone, and corrections are necessary to account for additional
431
short lifetime components [69]. The QD acceptor PL decays can also be analyzed to extract
432
FRET efficiencies and donor-acceptor distances. The analysis of acceptor PL decay, while more
433
complex and beyond our scope here (see refs. [68, 69, 71]), often provides better signal-to-
434
background ratios since it largely avoids contributions from unquenched LTbC. The above
435
methodology should be equally applicable with other LLnCs as well (e.g., Dy3+, Eu3+, Tm3+).
M
d
te
Ac ce p
436
an
426
437
Another important outcome of FRET between LLnCs and QDs is the combined utility of FRET
438
and background rejection (e.g., autofluorescence, scattering) associated with time-gating. This
439
capability is highly advantageous for assays in complex biological matrices prone to scattering or
440
autofluorescent interferences. Time-gated homogeneous immunoassays for estradiol and alpha-
441
fetoprotein (AFP), a cancer biomarker, have been developed on the basis of FRET between
442
LLnCs and QDs. In the former, CdTe QDs (peak PL at 728 nm) served as acceptors for Eu3+ and
443
Tb3+ chelate derivatives of estradiol in a competitive assay format [72]. In the AFP sandwich
444
assay, hydrophobic CdSe/ZnS QDs (peak PL at 605 nm) were doped into polystyrene
445
microparticles that were conjugated with anti-AFP antibodies; a second set of anti-AFP
446
antibodies were labeled with a luminescent Tb3+ complex. When AFP was present in a sample, it
447
bridged the QDs and Tb3+ complexes to generate FRET [73]. A further advantage of LLnC-QD 16 Page 16 of 63
FRET pairs is that relatively high levels of multiplexing are possible due to the broad absorption
449
of QDs, which permits the use of one type of LLnC as a common donor for multiple QD
450
acceptors. For example, Geißler et al. have demonstrated the multiplexed detection of FRET
451
from a luminescent Tb3+ complex to CdSe/ZnS QDs with peak PL at 529, 565, 604, and 653 nm,
452
and CdSeTe/ZnS QDs with peak PL at 712 nm (Fig. 2C) [70]. This system was assembled using
453
botinylated QDs and LLnC-labeled streptavidin as a model for five-plex bioaffinity assays. The
454
Tb3+ complex had sharp emission lines at ca. 490, 545, 585, and 620 nm, permitting sensitization
455
of all the QD acceptors and resolution of their emission.
cr
ip t
448
us
456
Beyond assays, the time-gating process essentially represents dynamic switching between
458
―FRET off‖ conditions in the prompt time regime (typically < 102 ns) and ―FRET on‖ conditions
459
in a gated time regime (> 102 µs), which might be expected to have implications for developing
460
photonic logic devices.
461
4.2. Polypyridyl transition metal complexes
462
Polypyridines (PPs) are a well-known class of coordination complex that have numerous
463
applications due to their (electro)chemical and optical properties [74, 75]. The redox activity of
464
polypyridine complexes has arguably lead to a general expectation that these compounds should
465
engage in photoinduced charge transfer reactions with QDs. Commensurate with these
466
expectations, rhenium bipyridine [76], ruthenium bipyridine [77], ruthenium phenanthroline
467
[78], and oxo-centered triruthenium clusters (with pyridine ligands) [79] have been shown to
468
function as electron or hole acceptors for photoexcited QDs. Nonetheless, recent work has
469
demonstrated that FRET can occur between QDs and certain osmium PP complexes.
M
d
te
Ac ce p
470
an
457
471
In one study, McLaurin et al. developed a fluororatiometric oxygen sensor by pairing a CdxZn1–
472
xSe/CdyZn1–yS
473
1,10-phenanthroline)(N-(6-aminohexyl)-4'-methyl-2,2'-bipyridine-4-carboxamide)osmium(II)
474
bis(hexafluorophosphate) ([OsII(Ph2phen)2(Nbpy)](PF6)2) [80]. The Os(II)PP complex was
475
directly conjugated to a carboxylated polymer coating on the QD through a short alkyl amine
476
linker via carbodiimide chemistry. Under two-photon excitation (2PE), the QD served as an
477
antenna to absorb light and transfer the excitation energy to the Os(II) complex. The quantum
478
yield of the resulting 2PE-FRET-sensitized Os(II)PP phosphorescence was sensitive to oxygen,
core/shell QD donor with a phosphorescent Os(II)PP acceptor, bis(4,7-diphenyl-
17 Page 17 of 63
with progressive quenching observed, relative to the QD PL intensity, for O2 pressures between
480
0–760 torr [80]. In another study, Stewart et al. thoroughly characterized energy transfer between
481
an aqueous CdSe/ZnS QD and an Os(II)PP complex, [OsII(bpy)2(phen-NCS)](PF6)2 (bpy, 2,2'-
482
bipyridine; phen, phenanthroline) [81]. The Os(II)PP complex was attached to the distal terminus
483
of a short peptide sequence that self-assembled to the QDs via a hexahistidine tag (Fig. 3A-i).
484
The QD PL was progressively quenched as increasing numbers of Os(II)PP-peptide were
485
assembled per QD. Despite the weak light absorption by the Os(II) complex (εmax, vis = 12 500 M–
486
1
487
625 nm), measurements of steady-state PL (Fig. 3A-ii), time-resolved PL, and femtosecond
488
transient absorption were consistent with a FRET mechanism of energy transfer. Inhomogeneous
489
quenching of the QD PL that paralleled the spectral overlap (Fig. 3A-iii) was also indicative of
490
FRET rather than CT. Nonetheless, there were some nuances that reflected photophysical
491
differences between the QD-Os(II)PP complex pair and more conventional QD-dye FRET pairs,
492
suggesting the possibility of non-FRET contributions to energy transfer [81].
493
4.3. Azamacrocyclic metal complexes
494
Azamacrocycles are ubiquitous in coordination chemistry. These ligands and their metal
495
complexes have found widespread use as catalysts, dyes, receptors for ion sensing, and
496
therapeutics, among other applications. For example, porphyrin and phthalocyanine (PC)
497
complexes are characterized by high absorption coefficients, fast electron transfer reactions, and
498
redox and optical properties that can be tuned via organic substituents or selection of metal
499
centers [82]. A prominent biological application of these azamacrocycles is their use as
500
sensitizers for photodynamic therapy (PDT) [83, 84]. PDT involves the photoinduced production
501
of reactive oxygen species in situ to kill cancerous or other harmful tissue. Several porphyrins
502
have been approved for clinical use in North America and Europe, including porfimer
503
(Photofrin®), temoporfin (Foscan®), talaporfin (Laserphyrin®), and verteporfin (Visudyne®).
504
To date, most studies of FRET between QDs and porphyrins and related compounds, such as
505
phthalocyanines and chlorins, have been in the context of potential applications in PDT.
cr
ip t
479
Ac ce p
te
d
M
an
us
cm–1 at 490 nm) and small spectral overlap with various QD donors (peak PL at 520, 530, 550,
506 507
Burda‘s group was among the first to investigate energy transfer between hydrophobic QDs and
508
metallophthalocyanines (M-PCs) [85-88]. Silicon PCs with different axial ligands were adsorbed
509
to hydrophobic CdSe QDs and photophysical characterization revealed that there were non18 Page 18 of 63
FRET contributions to energy transfer [85, 86]. The non-FRET contributions were attributed to
511
the involvement of surface states in energy transfer. Growth of a thin ZnS shell (i.e., CdSe/ZnS
512
QD) passivated these states and reduced the overall energy transfer efficiency by a factor of two
513
[86], suggesting that FRET may have been in competition with one or more charge transfer
514
pathways. A study by Wen et al. has also suggested that Dexter energy transfer is a possible
515
competing mechanism with FRET, at least between CdTe QDs and meso-tetra(4-
516
sulfonatophenyl)porphine dihydrochloride [89]. However, Nyokong‘s group has surveyed energy
517
transfer between CdTe QDs and many different M-PCs, attributing quenching of QD PL and
518
sensitization of M-PC emission to FRET in each case [90-97]. These M-PCs incorporated Zn(II),
519
Al(III), In(III), Ge(IV), Si(IV), Sn(IV), and Ti(IV) centers with a diverse array of ligands, and
520
were associated with the QDs through either adsorption or covalent attachment to their ligands.
us
cr
ip t
510
an
521
Much of the motivation for the abovementioned and other similar studies is the possibility of
523
2PE FRET as a possible modality for PDT. NIR radiation is ideal for tissue penetration;
524
however, PCs tend to have modest 2PE cross-sections. For example, the Si-PCs used by Burda‘s
525
group had a two-photon absorption cross-section of only 750 GM for an excitation band between
526
1300–1400 nm [88]. This spectral region is shared by the vibrational overtones of water and thus
527
is not well-suited for penetrating biological tissue. In contrast, the QD antennae had much larger
528
two-photon absorption cross-sections and permitted excitation within the biological tissue
529
window (ca. 650–950 nm [98]). Burda and coworkers were able to show that 2PE energy
530
transfer between CdSe QDs and Si-PCs was possible with 38% efficiency, although singlet
531
oxygen generation was not investigated [88]. Fortunately, subsequent studies have addressed this
532
latter point.
d
te
Ac ce p
533
M
522
534
In one of the first studies on singlet oxygen generation via QD-FRET, Tsay et al. labeled
535
phytochelatin-related peptides with chlorin e6 (no metal centre) as both an acceptor for
536
CdSe/CdS/ZnS QDs (peak PL at 620 nm) and a photosensitizer [99]. An average of 26 chlorin e6
537
molecules per QD were required to reach 50% FRET efficiency. The QDs were exploited as light
538
harvesting antennae, having a 10-fold larger absorption coefficient at 610 nm than the chlorin
539
had at its peak absorption (654 nm). The QD-chlorin conjugates had singlet oxygen quantum
540
yields between 10–31% depending on the excitation wavelength. Although these values were
19 Page 19 of 63
lower than for chlorin alone [99], this drawback could potentially be offset by the stronger
542
absorption of the QDs. In contrast, Tekdaş et al. studied putative FRET from QD donors to tetra-
543
triethyleneoxythia substituted Al(III)- (Fig. 3B), Ga(III)- and In(III)-PC acceptors and found that
544
the QD enhanced singlet oxygen quantum yields, suggesting potential use of QD-M-PC
545
conjugates as photosensitizers in PDT [100]. Furthermore, Qi et al. were able to show that 2PE
546
FRET between CdSe/ZnS QDs (peak PL at 595 nm) and a porphyrin (TrisMPyP-COOH; no
547
metal centre) produced twice the amount of singlet oxygen as 2PE of TrisMPyP-COOH alone
548
[101]. This study very clearly illustrates the power of using QDs as both antennae and scaffolds
549
for photosensitizers: despite a meager Förster distance of 2.7 nm and very low pairwise FRET
550
efficiency, enhanced singlet oxygen production was possible due to the large two-photon
551
absorption cross-section of the QD (ca. 800–2400 GM units between 800–1100 nm) and the
552
capacity for loading ~100 porphyrin molecules per QD. Note that the QDs were coated with an
553
amphiphilic polymer such that the Förster distance was shorter than the actual donor-acceptor
554
separation. In the foregoing examples, it is probable that FRET was the dominant, if not
555
exclusive, mechanism of energy transfer due to the core/shell/shell [99] and core/shell/polymer
556
[101] structures of the QDs.
cr
us
an
M
d
557
ip t
541
Beyond singlet oxygen generation, QDs are also potential FRET antennae for other
559
photosensitized reactions of metal complexes. For example, nitric oxide (NO) is an important
560
biological signaling molecule that is involved in thrombosis, vasodilation, neural activity, and
561
immune response, while also having potential as a cancer therapeutic [102-105]. It can be
562
photogenerated in situ from metal nitrosyl and nitrito complexes [106]. Ford‘s group found that
563
one such complex, a chromium(III) tetraazacyclotetradecane complex, trans-Cr(cyclam)(ONO)2+
564
(abbreviated CrONO), was able to serve as a FRET acceptor for aqueous CdSe/ZnS QDs [107-
565
109]. Although CrONO is an efficient photogenerator of NO (ΦNO = 25%), its practical use is
566
limited by weak light absorption (εCrONO < 150 M–1 cm–1) [109]. To address this limitation,
567
CrONO was adsorbed on QDs coated with anionic DHLA ligands through electrostatic
568
interactions (Fig. 3C-i). The QDs had molar absorption coefficients that were >1000-fold larger
569
than that for CrONO, and were thus able to serve as light harvesting antennae. UV light was
570
absorbed by the QDs and the energy transferred, via FRET, to the quartet excited state of the
571
CrONO, which subsequently produced NO as a result of a transition to its reactive doublet state
Ac ce p
te
558
20 Page 20 of 63
(Fig. 3-ii) [109]. Enhancements in NO production were between 3-fold and 15-fold [107-109].
573
Given the large two-photon absorption cross-sections of QDs, this type of configuration is also
574
likely to be suitable for harvesting NIR excitation light, which is most suitable for use in tissues
575
and other biological matrices. In addition to CrONO, FRET was observed with a dichloro
576
analog, trans-Cr(cyclam)Cl2+ (without NO production), but not a dicyano analog, trans-
577
Cr(cyclam)(CN)2+, that lacked spectral overlap with the QDs [108, 109]. In addition to QD-
578
FRET sensitized production of NO, these results show that FRET between QDs and transition
579
metal complexes can be tuned by selection of axial ligands.
us
580
cr
ip t
572
Finally, in addition to their role as sensitizers for PDT, porphyrins and phthalocyanines are also
582
potential sensitizers for dye sensitized solar cells (DSSCs) [82, 110]. DSSCs are a promising
583
technology that may provide a low-cost and sustainable source of energy in the future; however,
584
their practical use is currently hindered by lower efficiencies than competing technologies [111].
585
Improvements in sensitizers are critical for improving DSSC efficiencies, and porphyrins and
586
phthalocyanines are being investigated for this purpose [112-114]. Some of these investigations
587
are now also incorporating QDs and FRET. Recently, Jin et al. prepared porphyrin derived metal
588
organic frameworks (MOFs) that were surface-modified with CdSe/ZnS QDs [115]. The MOF
589
comprised 1,2,4,5-tetrakis(4-carboxyphenyl)-benzene connected by zinc centers; the QDs were
590
coated with PEG-amine, which bound to the zinc centers to immobilize the QDs. Very efficient
591
energy transfer (>80%, ~1 ns) from the QDs to the MOF was observed and attributed to FRET.
592
Electron transfer from the QDs to the MOF was not energetically allowed and the PL decay for
593
the MOF did not show any indication of hole transfer. Applying a monolayer of red-emitting
594
QDs to the MOF increased its light harvesting capability by 51% under one sun conditions. In
595
devices with a 10 porphyrin layer MOF, which corresponds to the estimated exciton migration
596
distance in the material, an enhancement of 5–10% is expected [115]. Section 7.2 describes
597
further applications for QD-FRET in DSSC development.
598
4.4. Distinguishing FRET from charge transfer
599
While it is relatively straightforward to establish that an energy transfer mechanism exists
600
between a QD and metal complex, it is not always straightforward to unambiguously determine
601
that quenching of QD PL is due to a Förster-type dipole-dipole interaction (FRET) rather than
602
photoinduced charge transfer (CT). Even QD-fluorescent dye systems are not always
Ac ce p
te
d
M
an
581
21 Page 21 of 63
straightforward. For example, it would generally be assumed, at first glance, that Rhodamine B
604
would engage in FRET with QDs; however, Lian‘s group recently found that FRET was only
605
responsible for ~16% of the energy transfer from CdS QDs to adsorbed Rhodamine B dye
606
molecules [116, 117]. The remaining ~84% was attributed to electron transfer. Similar results
607
were found for methylene blue adsorbed on CdSe QDs (~6% FRET; ~94% charge transfer)
608
[118], albeit that this dye is better known for its propensity for charge transfer.
ip t
603
cr
609
Experimentally, both FRET and photoinduced CT quench steady-state QD PL and decrease the
611
QD PL lifetime. Using these parameters to track changes in energy transfer efficiency, E, with
612
increases in the number of acceptors per QD will not provide any direct insight into the
613
quenching mechanism. All mechanisms will yield an analogous trend according to eqn. 12,
614
where kET, i is the rate of an arbitrary energy transfer pathway. N
ET, i
i
M
E=
åk
an
us
610
N
k0 + å kET, i
(12)
i
In contrast, experiments that vary donor-acceptor parameters such as spectral overlap and
616
separation distance can provide insight into the mechanism of energy transfer. For example,
617
FRET would be characterized by a linear dependence on the spectral overlap integral and
618
orientation factor, and an inverse sixth power dependence on the donor-acceptor separation.
619
Unfortunately, systematically and reproducibly varying these parameters is not always practical
620
or even feasible, and therefore other observations must provide insight. When the acceptor is
621
fluorescent, the observation of sensitized acceptor emission is often a good indication of FRET.
622
Similarly, when QD donor quenching is observed without significant spectral overlap, CT is a
623
likely mechanism. This was the case with quenching of QD PL by dopamine, which has been
624
confirmed to be due to CT [119, 120]. However, as demonstrated by the aforementioned studies
625
with Os(II)PP complexes, even weak absorption can be sufficient for non-trivial spectral overlap
626
[81]. Changes in QD PL quenching with reduction or oxidation of a putative CT acceptor
627
(electron or hole) can also provide insight. For example, Burks et al. did not observe NO
628
production upon oxidation of CrONO (Section 4.3), which supported a FRET mechanism for
629
energy transfer [109], while Medintz et al. exploited the poor CT between QDs and the reduced
Ac ce p
te
d
615
22 Page 22 of 63
630
hydroquinone form of dopamine for pH sensing (the oxidized quinone form of dopamine was a
631
good electron acceptor) [119].
632 One of the most important techniques for assessing possible CT interactions is transient
634
absorption (TA) spectroscopy, which permits tracking of ultra-fast processes and identification
635
of transient intermediates. Observation of spectroscopic signatures for oxidized or reduced forms
636
(as appropriate) of metal complexes or dyes permits assignment of CT as the energy transfer
637
mechanism. Luminescence upconversion and other pump-probe experiments can also be useful
638
in this regard. Nonetheless, definitive evidence for CT can be challenging to obtain when the
639
acceptor has weak or absent transient absorption features.
640
us
cr
ip t
633
Certainly, the majority of the literature would suggest that, at short distances (e.g., adsorbed
642
metal complexes or dyes) and with core-only QDs, CT is the preferred mechanism of energy
643
transfer. However, Medintz and coworkers [78] and Benson‘s group [121] have demonstrated
644
that ruthenium(II) phenanthroline complexes can engage in charge transfer with core/shell QDs
645
when bridged by peptides or proteins. Conversely, as described in Section 4.3, chromium(III)
646
cyclam complexes and PCs engage in FRET when adsorbed to QDs. CT is clearly not limited to
647
core-only QDs and near contact distances between donor and acceptor, nor is FRET limited to
648
core/shell QDs and large donor-acceptor separations. The assumptions that (i) short-range energy
649
transfer with hydrophobic core-only QDs occurs through CT and (ii) long-range energy with
650
aqueous core/shell QDs occurs via FRET are good starting points for directing the design of
651
photophysical experiments to characterize energy transfer, but are indeed assumptions that need
652
to be tested.
653
5. Energy Transfer with Other Nanomaterials
654
Although the metal complexes in the previous section represent non-traditional partners for QDs
655
in FRET, a feature they have in common with organic dyes is that they can be approximated as
656
zero-dimensional (0D) or point dipoles. Similarly, while QDs are much larger than both organic
657
dyes and metal complexes, they too have been shown to reduce to approximate point dipoles
658
[122]. Combinations of these materials thus comprise 0D-0D pairs and energy transfer is
659
described by the Förster formalism. However, there are increasing reports of energy transfer
Ac ce p
te
d
M
an
641
23 Page 23 of 63
between QDs and other nanomaterials that are not necessarily 0D. Dipole-dipole coupling is still
661
expected to occur in many of these instances, but the dimensionality of the other nanomaterial
662
impacts the formalism and, more pertinently, the distance scaling of the process. This section
663
reviews some of what is currently known about dipole-dipole energy transfer between QDs,
664
carbon nanomaterials, and gold nanoparticles.
665
5.1. Carbon nanomaterials
666
There have been several recent reports of pairing QDs with, for example, graphene [123],
667
graphene oxide (GO) [124], and carbon nanotubes (CNT) [125] for putative FRET assays. The
668
cited examples are nucleic acid assays where the proximity between the carbon nanomaterials
669
and QDs is mediated by the stronger affinity of single-stranded oligonucleotides (ssDNA) toward
670
carbon nanomaterials than their double-stranded counterparts (dsDNA). The difference in
671
affinity arises from the relative inability of the purine and pyrimidine bases in dsDNA to interact
672
with the carbon nanomaterials. When the ssDNA probes are conjugated to QDs, the absence of
673
target DNA results in adsorption of the conjugates and efficient quenching of QD PL.
674
Hybridization with target DNA disrupts the interactions of the ssDNA probe with the carbon
675
surface and the proximity required for energy transfer is lost. In addition to DNA, Liu et al. have
676
reported a homogeneous sandwich immunoassay where GO-reporter antibody conjugates and
677
QD-capture antibody conjugates were brought together in the presence of target antigen,
678
resulting in quenching of QD PL [126]. While such applications establish that energy transfer
679
occurs, they do not clearly establish that the mechanism of energy transfer is FRET.
cr
us
an
M
d
te
Ac ce p
680
ip t
660
681
Morales-Narváez et al. found that the effective quenching of QD PL by carbon nanomaterials
682
followed the order graphite < carbon nanofiber ≈ multiwall CNT < GO (Fig. 4A) [127]. This
683
order of quenching efficiency followed the trend in surface area (m2 g–1) for the carbon
684
nanomaterials. It was also suggested that the planar nature of the GO provided a more favourable
685
κ2 for energy transfer. The overall trend was not unexpected given the elegant results from
686
Shafran et al. [128], who showed that a CNT behaved as a one-dimensional (1D) acceptor for
687
CdSe/ZnS QDs, and Jander et al. [129], who showed that a 9 nm thick amorphous carbon film
688
behaved as a two-dimensional (2D) acceptor for CdSe/CdS nanorods. These dimensionalities are
689
in contrast to the conventional Förster formalism, which models the donor and acceptor as 0D
690
dipoles. The distance (r) dependent scaling for the rate of energy transfer with a 0D donor is r–6 24 Page 24 of 63
691
for a 0D acceptor, r–5 for a 1D acceptor, r–4 for 2D acceptor, and r–3 for a 3D acceptor. A r–4
692
dependence was confirmed with the thin film of amorphous carbon [129] and is theoretically
693
expected for graphene [130].
694 The evidence to date suggests that QDs can engage in Förster-like energy transfer through
696
dipolar interactions with carbon nanomaterials. However, the rate of energy transfer in such
697
configurations is not expected to have the inverse sixth-power distance scaling of conventional
698
FRET. Moreover, charge transfer has been reported between various QD materials (e.g., CdSe,
699
CdTe) and graphene, GO [131-133], and CNTs [134, 135], such that there is a possible
700
competition between dipolar energy transfer and charge transfer. Again, distinguishing these
701
mechanisms may not be straightforward, and careful design of the interface between the QD and
702
carbon nanomaterial may be able to tip the scales in favour of one mechanism or the other.
703
5.2. Gold nanoparticles
704
Gold nanoparticles (Au NPs) have been paired with QDs as energy acceptors in more studies
705
than can be reviewed here (for example, see refs. [136-139] and many others). In most of these
706
studies, the Au NPs have served a function analogous to a non-fluorescent molecular dye
707
acceptor (i.e., dark quencher). Their primary advantage was more efficient quenching over
708
longer distances. Here, we briefly review what is known about the mechanism of energy transfer
709
between QDs and Au NPs.
cr
us
an
M
d
te
Ac ce p
710
ip t
695
711
To appreciate energy transfer between QDs and Au NPs, it is worthwhile to briefly review
712
energy transfer between organic dyes and Au NPs. Dulkeith et al. studied the quenching of dye
713
fluorescence by Au NPs ranging from 1–30 nm in size [140, 141]. The quenching was found to
714
be due to changes in both the radiative and non-radiative rates of the dye in close proximity to
715
the Au NP; however, the Gersten-Nitzan model [142] for calculating radiative and non-radiative
716
rates near dielectric nanoparticles overestimated both quantities. Strouse and coworkers similarly
717
studied the quenching of fluorescent dyes by 1–2 nm Au NPs and found excellent agreement
718
with a nanosurface energy transfer (NSET) model [143-145], which was developed from the
719
work of Chance, Prock, and Silbey [146], as well as Persson and Lang [147, 148], on energy
720
transfer from a dipole to a metal surface. In NSET, the coupling is between the donor dipole and
721
the nearly free electrons in the conduction band of the NP, which is considered to be an 25 Page 25 of 63
approximate surface of localized dipoles. The rate of energy transfer in this system has an
723
inverse fourth-power dependence, such that the efficiency of energy transfer is given by eqn. 13,
724
where d is the donor-nanosurface separation and d0 is the characteristic distance. The value of d0
725
is defined by eqn. 14, where c is the speed of light, ωD is the angular frequency of the donor, and
726
ωf and kf are the angular frequency and Fermi wavenumber, respectively, for the nanosurface
727
[143]. The values for gold are substituted in the right-hand side of eqn. 14.
c3F D
nw D2 w f k f
(
= 2.23 ´10-25 cm s
)
c3F D nw D2
us
d04 = 0.225
d04 d04 + d 4
cr
ENSET =
ip t
722
(13)
(14)
Experiments have shown that the efficiency of NSET improves as overlap between the donor
729
emission (520–780 nm) and the localized surface plasmon resonance (LSPR) of a 2 nm Au NP
730
(~525 nm) increases [144]. The most important consequence of the NSET mechanism is that
731
energy can be efficiently transferred to the Au NP over larger distances (up to ~15–20 nm) than
732
FRET between dyes (< 10 nm).
M
d
733
an
728
Considering QDs, an early study by Gueroui and Libchaber found that single-pair energy transfer
735
between CdSe/ZnS QDs and 1.4 nm Au NPs was consistent with FRET [149], although it was
736
later suggested that the narrow range of distance data also fit a NSET model [150]. A detailed
737
study by Pons et al. found that the energy transfer efficiency between CdSe/ZnS QDs and 1.4 nm
738
Au NPs had a distance dependence that was consistent with NSET [150]. The QD-Au NP
739
separation distance was controlled through relatively rigid ß-sheet peptide spacers that were
740
labeled at their distal terminus with the Au NPs and self-assembled to the QDs. Importantly, in a
741
direct comparison with FRET between QDs and fluorescent dyes, energy transfer to the Au NPs
742
was found to be more efficient over longer distances. This result is a particularly important
743
advantage with QDs since the donor-acceptor separation is taken from the center of the QD, such
744
that the radius of the inorganic nanocrystal and the thickness of any organic coating set a
745
minimum separation. In another study, Li et al. found that energy transfer from CdSe/ZnS QDs
746
to 3 nm Au NPs had a NSET-like d–4 dependence, whereas energy transfer to 80 nm Au NPs had
747
a FRET-like d–6 dependence [151]. Energy transfer to 15 nm Au NPs also fit better to a FRET-
748
like model, although the predicted differences between NSET and FRET were less significant for
Ac ce p
te
734
26 Page 26 of 63
this size of Au NP (Fig. 4B). Note that these were rather limited data sets, with data for only
750
three different donor-acceptor separations collected per size of Au NP. Interestingly, Han et al.
751
found that quenching of QD PL by slightly smaller 12 nm Au NPs was consistent with a NSET
752
model [152]. Analysis of steady state and time-resolved PL data indicated a QD-Au NP
753
separation of 7.9–8.1 nm, which was corroborated by TEM imaging. Zhang et al. similarly
754
demonstrated that the PL from thin films of CdTe QDs (five sizes/colours; peak PL between
755
534–660 nm) were quenched in accordance with NSET when controllably positioned at specific
756
distances above a thin film of 5.5 nm Au NPs [153]. Here, the NSET efficiency was given by
757
eqn. 15, where rAu is the radius of the Au NPs, cAu is the concentration of Au NPs per unit area,
758
and d is the distance between the QDs and Au NP layers.
ENSET = 1+
3( d - rAu )
3
cAup d04 ( 3d - rAu )
an
-1
us
cr
ip t
749
759
(15)
The possibility that the mechanism of energy transfer depends on the size of the Au NP has been
761
recognized with molecular dye donors. For example, Griffin et al. have shown that although
762
NSET is able to predict the quenching efficiency as a function of distance with up to 8 nm Au
763
NPs, the model appears to underestimate the quenching efficiency with 40 nm and 70 nm Au
764
NPs [154]. Notably, the NSET model does not require a resonant interaction and small (≤ 2 nm)
765
Au NPs do not display a strong LSPR, suggesting that, as Au NPs increase in size and the
766
magnitude of their LSPR increases, the NSET model becomes insufficient. Discrepancies may
767
also be due to the increased contribution of scattering relative to absorption in the extinction
768
spectra of the Au NPs [144]. Another model, dipole-to-metal particle energy transfer (DMPET),
769
has been proposed to explain the quenching of dyes by Au NPs [155]. This model has not been
770
widely applied with QDs and has been suggested to reduce to other models within the constraints
771
of real experimental systems [144, 150].
Ac ce p
te
d
M
760
772 773
Further studies are needed to clearly define the Au NP size regimes that accept energy from QDs
774
through NSET, FRET-like, or other dipolar and higher order (e.g., quadrupolar) interactions.
775
Although conceptually similar, the scaling differences between FRET and NSET are significant.
776
Detailed knowledge of electronic coupling interactions between QDs and Au NPs should
777
facilitate the rational design of nanosystems that range from highly sensitive biodiagnostic 27 Page 27 of 63
probes to hierarchically structured optically active materials where, for example, QDs are
779
arranged around large Au NPs, or small Au NPs are arranged around QDs.
780
6. Biomolecule Supported FRET Relays
781
As noted in the Introduction, we define traditional FRET configurations as those between QDs
782
and organic dyes and occurring in a single step. Discrete assemblies of QDs and dyes that
783
incorporate multiple energy transfer steps, i.e., a relay or cascade, have been much less
784
frequently reported than their single-step counterparts; however, these non-traditional
785
configurations have capabilities that cannot be accessed in traditional configurations.
786
6.1. Long distance relays
787
While the range of energy transfer between a QD donor and dye acceptor is effectively limited to
788
< 1.5R0, the net distance over which energy is transferred can be increased by utilizing multiple,
789
successive energy transfer steps. Such systems with cascading energy transfer are referred to as
790
FRET relays due to the function of at least one dye as both an acceptor and donor (i.e., a ―relay‖
791
point). The first QD-FRET relay was reported by Medintz et al. in their seminal work on the use
792
of QDs and FRET for biosensing [156]. The relay was used to increase the efficiency of energy
793
transfer between an initial QD donor and terminal dye acceptor [156]. The terminal dye, Cy3.5,
794
was a label on cyclodextrin, which itself was situated in the binding pocket of MBP. The MBP
795
was self-assembled to QDs (peak PL at 530 nm) and was site-specifically labeled with a Cy3
796
relay point that could achieve ~95% quenching of QD PL at a 10:1 ratio of MBP:QD and 20%
797
net energy transfer efficiency to the terminal Cy3.5. Configurations without the intermediate dye
798
and FRET relay did not function [156]. Five years later, Lu et al. assembled a three-fluorophore
799
system that comprised CdSe/ZnS QDs (peak PL at 517 nm), enhanced yellow fluorescent protein
800
(EYFP), and Atto647 dye [157]. The QD served as an initial energy donor, with preferential
801
energy transfer to the EYFP (R0 = 3.9 nm) but viable energy transfer to the Atto647 (R0 = 3.2
802
nm). EYFP and Atto647 also formed a good FRET pair (R0 = 4.8 nm). The EYFP was site-
803
specifically modified with a single-stranded oligonucleotide using a heterobifunctional
804
crosslinker, and ~14 equivalents of the modified EYFP were self-assembled to the QD through a
805
polyhistidine tag. The conjugate was completed by hybridization of the oligonucleotide with its
806
Atto647-labeled complement. The dominant energy transfer pathway was the relay from the QD
Ac ce p
te
d
M
an
us
cr
ip t
778
28 Page 28 of 63
807
to the EYFP to the Atto647, with the net FRET efficiency reaching up to 5–17% for 13–9 nm
808
distances versus < 1% for direct transfer from the QD to Atto647 [157].
809 The two foregoing examples clearly show that even a single FRET relay step can improve
811
energy transfer efficiency over distances > 1.5R0. Recently, Boeneman et al. used QDs (peak PL
812
at 530 nm) as a central scaffold for the assembly of DNA-based photonic wires incorporating
813
multiple FRET relay steps (Fig. 5A) [158]. Excitation energy was initially transferred from the
814
QD and three subsequent energy transfer steps occurred down the length of the DNA ―wire‖
815
[158]. The wire was made by hybridizing four single strands of oligonucleotide, labeled with
816
either Cy3, Cy5, Cy5.5, or Cy7, in sequence, along a longer template strand. The template strand
817
was modified with a polyhistidine tag for self-assembly to CdSe/ZnS QDs. The dyes were
818
spaced ~3.3 nm apart and R0 values for sequential energy transfer steps ranged from 3.8–5.3 nm.
819
Energy could be transferred over distances > 15 nm, although it was found that the net end-to-
820
end efficiency was limited by the properties of the fluorescent dyes. Re-emission at the Cy5 and
821
Cy5.5 was limited to 2.2% and 1%, respectively, and the Cy7 effectively functioned as a dark
822
quencher (Fig. 5B-D). Nonetheless, the QD served as a superior light absorber and sensitizer for
823
initiating the FRET cascade along the photonic wire. Design improvements, such as selection of
824
brighter fluorescent dyes and optimized spacing of those dyes, can be expected to improve
825
stepwise efficiencies and permit substantial energy transfer over photonic wires 10–20 nm in
826
length.
827
6.2. Concentric relays
828
FRET relays have also been developed where cascaded energy transfer steps do not move the
829
excitation energy progressively further away from a central QD. Such configurations are referred
830
to as ―concentric‖ FRET relays, and their design is greatly facilitated by the ability to array
831
multiple fluorophores around a central QD. Donor or acceptor partners can be co-assembled at
832
similar distances and at separate attachment points so that multiple energy transfer steps occur
833
within a compact spherical volume that is concentric to the QD. Algar et al. have developed two
834
systems of this type for biosensing applications [71, 159, 160].
Ac ce p
te
d
M
an
us
cr
ip t
810
835 836
The first concentric FRET relay reported was based on energy transfer from a LTbC donor to a
837
CdSe/ZnS QD (peak PL at 620 nm), then from the QD to a deep red fluorescent dye, Alexa Fluor 29 Page 29 of 63
647 (A647). The Förster distances for these FRET pairs were quite large (LTbC-QD, R0 = 10.1
839
nm; QD-A647, R0 = 7.5 nm), affording energy transfer efficiencies > 90% (LTbC-QD) and
840
> 75% (QD-A647) when LTbC- and A647-labeled peptides were self-assembled to the QDs.
841
What is particularly interesting about this system is that it operated in two modes (Fig. 6A):
842
prompt and time-gated. A flash of UV light excited both the LTbC and QD efficiently. Prompt
843
energy transfer from the QD (τ ≈ 50 ns) to ground state A647 (τ ≈ 1 ns) occurred within < 250
844
ns, such that both the QD and FRET-sensitized A647 had emitted and returned to their ground
845
state prior to appreciable decay of the LTbC excited state (τ ≈ 2.6 ms). Time-gated energy
846
transfer from the LTbC to the QD, with relayed energy transfer to the A647, could then be
847
observed over a period > 1 ms following a time-delay of 55 µs between excitation and
848
acquisition of PL signals.
us
cr
ip t
838
an
849
One important advantage of the time-gated FRET relay was that it took information carried by
851
the QD-A647 FRET pair in the prompt time domain (< 10–7 s) and transferred it to the timescale
852
of 10–4–10–3 s. FRET pairs analogous to the QD-A647 have been widely utilized in various assay
853
configurations [25, 26], and time-gating is well-known to afford superior signal-to-background
854
ratios when analyzing complex biological samples [161, 162]. The time-gated FRET relay was
855
able to combine these two capabilities in a single nanoparticle vector. Assays for protease
856
activity were done by choosing the peptide linker between the QD and A647 to be a substrate for
857
a proteolytic enzyme, and by choosing the peptide linker between the LTbC and QD to be a non-
858
substrate. Monitoring the time-gated A647/QD PL ratio permitted tracking of proteolysis. A
859
second advantage of the time-gated FRET relay was the possibility of multiplexed assays and
860
biosensing with a single colour of QD vector. This capability is in contrast to the current state-of-
861
the-art, which is to use N colours of QD to detect N analytes [163]. In one multiplexed assay, two
862
target DNA sequences were detected by modifying the QD with oligonucleotide probes and
863
labeling the two target sequences with either LTbC or A647 [71]. Hybridization provided the
864
proximity for FRET with the QD. One target could be detected through the prompt A647/QD PL
865
ratio, while the second target could be detected through the total time-gated PL from the QD and
866
A647. In a second multiplexed assay, the LTbC and A647 were again attached to the QD via
867
peptide linkers; however, in this case, the peptides were respective substrates for two different
868
proteolytic enzymes (Fig. 6B-i) [159]. With appropriate cross-calibration (Fig. 6B-ii), the prompt
Ac ce p
te
d
M
850
30 Page 30 of 63
869
A647/QD PL ratio and the time-gated QD/LTbC PL ratio afforded orthogonal and quantitative
870
tracking of the two proteases (Fig. 6C). Multiplexed assays of proteolytic activity, multiplexed
871
inhibition assays, and pro-protease activation assays were possible in this format.
872 A second time-gated FRET relay was developed without a time-gating component. In this relay,
874
a CdSe/ZnS QD (peak PL at 520 nm) served as an initial donor and sensitizer rather than an
875
energy transfer intermediary [160]. Alexa Fluor 555 (A555) and A647 dyes were assembled
876
concentrically around the QD using peptide linkers (Fig. 7A). Energy could be transferred
877
directly from the QD to the A555 (R0 = 4.7 nm), or to the A647 (R0 = 3.0 nm), and from the
878
A555 to the A647 (R0 = 5.7 nm). Different combinations of A555 and A647 per QD resulted in
879
unique combinations of A555/QD and A647/QD PL ratios. These two ratios were mapped
880
against one another to create a two-dimensional parameter space that ultimately correlated PL
881
intensities with the number of A555- and A647-labeled peptides per QD. Multiplexed assays of
882
proteolytic activity and pro-protease activation (Fig. 7B) were again possible. Additionally,
883
analysis and modeling of the energy transfer efficiencies between the A555 and A647 at different
884
assembly ratios was used to gain insight into the interfacial distribution of self-assembled
885
peptides. The data was more consistent with approximately equidistant spacing rather than
886
random spacing [160]. While single-step FRET configurations have proven invaluable for
887
characterization of the radial dimensions of QD-bioconjugates, concentric FRET configurations
888
may permit further characterization of interfacial structure. Such capability is, of course, in
889
addition to multiplexed sensing in a single vector, which should be especially amenable to
890
visualization of coupled biochemical processes in cells and tissues.
891
7. FRET in Thin Films of Quantum Dots
892
Traditional and non-traditional QD-FRET assemblies dispersed in solution are especially useful
893
for applications in bioimaging and bioanalysis. Applications in optoelectronics, however, tend
894
more toward thin films of QDs or composite materials. FRET is both intrinsic to thin films of
895
QDs and can be engineered to enhance certain optical properties. This section discusses the
896
characterization of energy transfer in thin films of QDs, as well as the application to DSSCs and
897
other optoelectronic materials and devices. An example of utilizing thin films of QDs and FRET
898
for bioanalysis is reviewed elsewhere in this issue [164].
Ac ce p
te
d
M
an
us
cr
ip t
873
31 Page 31 of 63
7.1. Characterization of energy transfer
900
FRET can occur in thin films, solids, and composites of QDs whenever the QDs are at
901
sufficiently high density. The effect is subtle and appears as a small bathochromic shift in the PL
902
spectra of approximately monodisperse QDs when compared to bulk solution. This effect was
903
first reported by Kagan et al. for close-packed solids of CdSe QDs and is a consequence of the
904
inhomogeneous distribution of sizes (< 5%) with energy transfer from slightly smaller QDs to
905
slightly larger ones [165]. Crooker et al. were able to spectrally resolve PL quenching across an
906
inhomogeneous distribution of QDs (i.e. the width of the QD PL spectrum) and show the
907
smallest QDs were quenched by the greatest degree, with progressively less quenching as size
908
increased (longer emission wavelength) [166]. Analogous energy transfer also occurs when
909
different size distributions of approximately monodisperse QDs are mixed at sufficiently high
910
density (e.g., QDs with green and red PL), albeit that quenching of smaller donor QDs and PL
911
enhancement of the larger acceptor QDs is more easily resolved.
an
us
cr
ip t
899
M
912
In the years since the initial reports noted above, Bradley and coworkers have extensively
914
studied FRET between QDs in different thin film architectures [167-171]. The thin films were
915
made up of mercaptoacetic acid (MAA)-coated CdTe QDs assembled through layer-by-layer
916
(LbL) assembly with poly(sodium-4-styrene sulfonate) (PSS) and poly(diallyldimethyl-
917
ammonium chloride) (PDDA) on quartz substrates. In one study, a mixed monolayer of green
918
QD donors (peak PL at 524 nm) and orange QD acceptors (617 nm) were deposited and studied
919
[169]. The FRET efficiency was dependent on the ratio of acceptors to donors, reaching values
920
as high as 90% for a 2:1 ratio (cf. 59% for a 1:4.5 ratio). Both nearest-neighbour and long-range
921
FRET interactions were identified. The trends in FRET efficiency were explained by a 2D
922
energy transfer model that accounted for the dimensions of the QDs, imposing a minimum
923
center-to-center donor-acceptor separation or ―exclusion zone.‖ The FRET efficiency is given by
924
eqn. 16, where cA is the concentration of acceptors per unit area, and α and β are numerical
925
constants (see ref. [172]) related to the ratio Rex/R0, where Rex = RD + RA is the radius of the
926
exclusion zone (equal to the sum of the radii of the donor QD, RD, and acceptor QD, RA).
Ac ce p
te
d
913
(
)
(
EFRET = 1- a1 exp b1 R02cA - a 2 exp b2 R02cA
)
(16)
927
Interestingly, the foregoing study also found that energy transfer within the ensemble of
928
monodisperse acceptor QDs was significant while energy transfer within the ensemble of 32 Page 32 of 63
monodisperse donor QDs was not. Subsequent work revealed that the extent of intra-ensemble
930
FRET depended on the degree of inhomogenous broadening in that particular ensemble, which
931
was reflected by the PL FHWM and Stokes shift [168]. There was also a clear correlation with
932
density: as the concentration of monodisperse green QDs increased in a (sub-)monolayer film,
933
the observed bathochromic shift in their PL spectrum increased, the PL lifetime on the short
934
wavelength side of the QD PL spectrum decreased, and the difference in PL lifetime between the
935
short and long wavelength sides of the PL spectrum increased [168]. The FRET efficiency within
936
an approximately monodisperse film of QDs is given by eqn. 17, where r ≥ Rex = 2RQD, and the
937
numerical constant is defined for r in units of nm. 6
æ r ö n4 = 1+ ç è 0.0211÷ø 6k 2F D J ( l )
(17)
an
EFRET
-1
us
cr
ip t
929
Interestingly, intra-ensemble energy transfer within a monolayer of green QDs (peak PL at 547
939
nm) was found to affect the inter-ensemble energy transfer to a monolayer of orange QDs (peak
940
PL at 610 nm) [170]. Monolayers of the two QDs were separated by PSS/PDDA multilayers of
941
variable thickness (ca. 0.5–12.0 nm). While the donor-acceptor FRET efficiency increased as the
942
surface concentration of acceptors increased, the efficiency decreased as the surface
943
concentration of donor QDs increased due to enhancements in the competing donor-donor FRET
944
pathway. The latter behaviour was accounted for by introducing a concentration dependent donor
945
QD quantum yield term in 2D FRET models [170]. The FRET efficiency for this thin film
946
architecture is given by eqn. 18, where d is the distance between the parallel donor and acceptor
947
monolayers.
Ac ce p
te
d
M
938
EFRET -1 = 1+
948
2d 4 cAp R06
(18)
949
Bradley and coworkers have also studied surface plasmon resonance (SPR) enhanced FRET in
950
thin films of CdTe QDs [167, 171]. In one study, three monolayers of Au NPs (7–8 nm) with a
951
resonance at 565 nm were deposited on a quartz substrate. A mixed monolayer of green QDs
952
(peak PL at 525 nm) and orange QDs (peak PL at 600 nm) was then deposited on the substrate
953
with variable thicknesses (ca. 5.6–22 nm) of polyelectrolyte spacer layers between the Au NPs
954
and QDs [167]. A distance of ca. 14 nm between the QDs and Au NPs was found to provide a
955
2.7-fold enhancement of FRET between the two colours of QD. In a second study, an intervening 33 Page 33 of 63
layer of Au NPs was positioned 2 nm from a monolayer of QD donors (peak PL at 560 nm) and
957
12 nm from a monolayer of QD acceptors (peak PL at 615 nm) using polyelectrolyte spacers
958
[171]. Even with a ca. 23.4 nm center-to-center separation between the QD donors and
959
acceptors, a FRET efficiency of 8% was observed, representing an 80-fold increase in the rate of
960
energy transfer and an effective increase in the Förster distance from 3.9 nm to 7.9 nm [171]. In
961
both cases, the FRET enhancement was attributed to the surface plasmon dipole field enhancing
962
the electronic coupling between the donor and acceptor QDs. Similar effects have also been
963
observed by other researchers. For example, Wang et al. observed enhanced FRET between
964
CdSe/ZnS QD donors and CdSe/ZnTe QD acceptors drop cast with Au NPs (20 nm) on silicon
965
substrates [173].
us
cr
ip t
956
966
Beyond FRET between two colours of QD, graded or ―rainbow‖ thin films of QDs have also
968
been of interest [174]. These films stack layers of QDs from green to yellow to orange to red PL,
969
or vice versa, to create directional energy transfer. Franzl et al. have assembled and characterized
970
a LbL structure with a hierarchical funnel of QD colours (peak PL): green (515 nm)/yellow (540
971
nm)/orange (570 nm)/red (620 nm)/orange/yellow/green (Fig. 8A) [175]. Nearly complete
972
quenching of the green, yellow, and orange QD PL was observed with a large concomitant
973
enhancement in red QD PL. One layer of red QDs in the graded film had more than 4-times more
974
PL emission than a monochrome film with 7 layers of red QDs (equivalent optical density). The
975
graded structure increased the exciton density in the red QD layer by a factor of 28. The
976
enhancement in the graded structure is largely attributed to exciton recycling [175]. In closely
977
packed films, defect rich QDs can trap excitons from defect poor QDs, which is exacerbated by
978
increases in carrier trapping with above band-gap excitation [176]. The graded structure is
979
thought to permit transfer of trapped excitons to the luminescent states of larger QDs, in addition
980
to transfer between the luminescent states of smaller and larger QDs [175, 176]. Multilayer thin
981
films of this type have tremendous potential for harvesting light and transporting that energy to a
982
functional interface (vide infra)—a process conceptually analogous to the role of energy transfer
983
in nature‘s photosynthetic machinery.
984
7.2. FRET in quantum dot sensitized solar cells
985
As noted above, one application for FRET within thin films of QDs is solar energy harvesting.
986
Without a change in our primary energy sources, there will be an estimated 14 TW gap between
Ac ce p
te
d
M
an
967
34 Page 34 of 63
global energy supply and demand within 35–40 years [111]. While radiation from the sun
988
provides more than enough energy to meet present and future energy demands, current solar cell
989
technology is not cost effective enough for widespread use in preference to fossil fuels and other
990
non-renewable sources of energy [177]. DSSCs are a promising low-cost alternative to silicon
991
solar cells but are currently much less efficient [111, 177-179]. Some of the research efforts to
992
address this limitation of DSSCs have focused on the combination of QDs and FRET.
ip t
987
cr
993
In a typical DSSC, sunlight is absorbed by a dye coated onto a thin layer of nanocrystalline TiO2
995
supported on a transparent conducting electrode. Upon photoexcitation, the dye transfers an
996
electron to the TiO2, the residual hole is scavenged by a redox mediator (i.e., electrolyte) for
997
transport to the cathode, and the electrons entering the external circuit can do electrical work
998
[177, 178]. While the ideal sensitizing dye has strong, broad light absorption across the visible
999
and near-infrared (NIR) regions of the spectrum and exhibits efficient charge separation, most
1000
real dyes do not fully meet these criteria. For example, ruthenium PP complexes have broad
1001
absorption spectra but relatively low absorption coefficients [111, 179], whereas organic dyes
1002
may have large absorption coefficients but only over a narrow wavelength range [177, 179].
1003
Increasing the efficiency with which the full solar spectrum is harvested is thus one of the current
1004
challenges facing DSSCs [180, 181]. QDs can provide rapid charge separation, can be prepared
1005
at low cost, and, importantly, exhibit light absorption that is strong, broad, and tunable [179,
1006
182]. As such, QD sensitized solar cells (QDSSCs) are an active area of research.
an
M
d
te
Ac ce p
1007
us
994
1008
One strategy for preparing QDSSCs is to deposit a multilayer ―rainbow‖ film of QDs onto TiO2
1009
instead of a sensitizing dye. FRET occurs between the layers and enhances QDSSC performance.
1010
For example, Ruland et al. have reported a ―rainbow‖ solar cell design where four colours of
1011
CdTe QDs were sequentially arranged through LbL assembly on an indium-doped tin oxide
1012
(ITO) electrode (Fig. 8B) [181]. The QDs had first exciton absorption peaks at 486, 522, 554,
1013
and 580 nm, respectively, and corresponding PL maxima at 522 (green), 553 (yellow), 585
1014
(orange), and 620 nm (red). Photovoltaic performance was evaluated in cells with a polysulfide
1015
electrolyte and Cu2S counter electrode, using four layers of red QD as a reference configuration.
1016
A forward configuration, with the QDs deposited in order of increasing size (ITO/greenred),
1017
was found to have both higher spectral incident-photon-to-current conversion efficiencies
35 Page 35 of 63
(IPCEs; Fig. 8B-iv) and a 25% higher power conversion efficiency (η) than the reference
1019
configuration. The enhancements were attributed to FRET rapidly funneling energy toward the
1020
electrolyte interface where charge separation occurred via reduction of the electrolyte [181]. The
1021
efficiency of the FRET process was even sufficient to compensate for an unfavourable energy
1022
gradient for hole transport up the QD gradient to the ITO. A reverse configuration
1023
(ITO/redgreen) performed less favourably than the reference cell. In a similar study, Santra
1024
and Kamat deposited alloyed CdSeS QDs in a sequential architecture on TiO2 (Fig. 8C-i) [183].
1025
The QD PL emission was compositionally tuned to green, orange, or red without changing
1026
nanocrystal size (4.5 ± 0.1 nm). Cells prepared with a single layer (one colour) of QD had
1027
maximum IPCE values between 35–42% (Fig. 8C-ii) and η increased from 1.97% to 2.81% with
1028
decreasing band gap due to different levels of light harvesting. In contrast, two-colour bilayer
1029
films (orange/red) or three-colour trilayer films (green/orange/red) had η > 3.0% and maximum
1030
IPCE values between 37–51%. Cells prepared with a non-sequential mixture of the three colours
1031
of QD had η = 2.34%. The η values for the real bilayers and trilayers were 30–60% larger than
1032
predicted for simple additive effects from the QDs, a result that was attributed to cascaded FRET
1033
and/or electron transfer from larger band gap QDs to smaller band gap QDs [183].
cr
us
an
M
d
1034
ip t
1018
In addition to sequential ordering of QD band gap energies (i.e., colour), Choi et al. have found
1036
that the order of core versus core/shell QDs is important in QDSSCs [184]. Cationic or anionic
1037
core-only CdSe QDs with short ligands and CdSe/CdS/ZnS QDs with long ligands were
1038
prepared with nearly identical band gap energies and were LbL-assembled on TiO2. As per the
1039
design of the experiment, the core-only QDs exhibited 3-fold faster electron transfer to the TiO2
1040
than the core/shell/shell QDs. Consequently, when the core-only QDs were sandwiched between
1041
the TiO2 and core/shell/shell QDs in a cell (polysulfide electrolyte and CoS counter electrode),
1042
the observed photocurrent and η were > 3.6-fold larger than a configuration with the
1043
core/shell/shell QDs sandwiched between TiO2 and core-only QDs. This enhancement was
1044
attributed to ~86% efficient FRET from the core/shell/shell QDs to the core-only QDs, and
1045
subsequent rapid electron transfer from the core-only QDs to the TiO2. Although the core-only
1046
QDs were 1.3-times more efficient than the core/shell/shell QDs in single layer cells, the addition
1047
of a layer of core/shell/shell QDs (and FRET) onto the core-only QDs resulted in 340% and
1048
420% enhancements in photocurrent and η [184].
Ac ce p
te
1035
36 Page 36 of 63
1049 A second FRET strategy for sensitized solar cells has been to use QDs as donors for acceptor
1051
dyes that efficiently transfer charge to TiO2 in DSSCs. For example, squaraine dyes, which
1052
exhibit strong light absorption between 600–800 nm [185], have been widely utilized in this type
1053
of configuration. The underlying idea is that the QD absorbs visible light and transfers the energy
1054
to the squaraine dye via FRET for cascaded sensitization of the TiO2. The squaraine dye is also
1055
able to directly absorb red/NIR light and sensitize the TiO2. Zaban‘s group developed the
1056
strategy of depositing CdSe/CdS/ZnS QDs on nanocrystalline TiO2, overcoating the QDs with a
1057
layer of amorphous TiO2, and then adsorbing an asymmetric squaraine dye with deep red light
1058
absorption (ε = 319 000 M–1 cm–1) (Fig. 9A) [179, 186, 187]. The CdS/ZnS shell around the
1059
CdSe core served as a barrier for direct charge injection into the nanocrystalline TiO2 and the
1060
amorphous TiO2 layer served as a barrier to protect the QDs from the corrosive effects of an I–
1061
/I3– electrolyte. FRET between the QDs and squaraine dye was ca. 40–70% efficient depending
1062
on the thickness of the amorphous TiO2 layer, increasing as the film thickness increased from
1063
2 nm to 4 nm, then decreasing as the thickness increased up to 11 nm [187]. These trends were
1064
attributed to changes in the effective surface concentration of the dye and the average donor-
1065
acceptor separation distance. With a 4 nm thick film, an internal quantum efficiency of 82% was
1066
achieved, with 70% due to FRET and 12% due to reabsorption of QD PL emission by the dye
1067
[187]. The IPCE for the cells was 6–10% in the range of ca. 400–675 nm and, compared to cells
1068
with squaraine dye alone, offered a substantial increase in IPCE between 400–550 nm (Fig. 9B)
1069
[186].
cr
us
an
M
d
te
Ac ce p
1070
ip t
1050
1071
An alternative method of incorporating QD antennae into DSSCs for improved efficiency has
1072
been to use squaraine dye derivatives as linkers between the TiO2 electrode and QDs. In one
1073
study, Grätzel‘s group synthesized a squaraine dye with two carboxyl groups to bind to
1074
nanocrystalline TiO2 and two decyl chains to help adsorb hydrophobic CdSe QDs capped with
1075
tri-n-octylphosphine (Fig. 10A-i) [180]. The TiO2 electrode was modified with the dye by dip
1076
casting and subsequently spin coated with QDs. The cells, operated with a Co2+/Co3+ electrolyte,
1077
exhibited a higher and broader IPCE spectrum than cells prepared with QDs or squaraine dyes
1078
alone (Fig. 10A-ii), and η increased from 0.79% to 1.48%. The FRET efficiency between the
1079
QDs and squaraine dye was ca. 69% [180]. In another study, Kamat‘s group synthesized an
37 Page 37 of 63
asymmetric squaraine dye with a carboxylate linker to bind to TiO2 and an alkyl thiol linker to
1081
bind to CdSe QDs (Fig. 10B-i) [188]. The QD PL was quenched 80% due to FRET with the
1082
squaraine dye. Cells prepared with QDs or dye alone had η values of 0.15% and 3.05%,
1083
respectively. In contrast, the QD/squaraine hybrid cells had η = 3.65% and exhibited higher
1084
IPCE at wavelengths between 400–530 nm (Fig. 10B-ii).
ip t
1080
1085
Clearly, QD-FRET has significant potential for enhancing DSSC technology. Both ―rainbow‖
1087
thin films with cascaded energy transfer and composite films with QD-dye FRET pairs have
1088
demonstrated superior light harvesting capability, as shown by improved IPCE and η values. One
1089
may speculate that some combination of these two strategies may yet provide the greatest degree
1090
of enhancement and help move DSSCs closer to practical use.
1091
7.3. Luminescent films and composite materials
1092
Thin films of QDs are also of interest for the fabrication of light-emitting diodes (LEDs) and
1093
display technologies. For example, QD-LEDs have been commercialized in Sony‘s new
1094
Triluminos display technology. This section briefly reviews some the utility of QD-FRET in
1095
constructing LEDs and developing optically active materials that may have similar applications.
us
an
M
d
te
1096
cr
1086
Demir and coworkers have been proponents of using FRET within thin films of QDs for
1098
improving colour conversion of near-UV/blue LEDs [189-193]. Currently, broadband emitting
1099
phosphors are the preferred method for colour conversion to white light; however, this approach
1100
suffers from low colour rendering indices and poor tuning ability, the latter of which can be
1101
important for optimizing lighting for different conditions and applications (e.g., indoor vs.
1102
outdoor, room illumination vs. greenhouses) [194]. The narrow, tunable emission of QDs can
1103
produce pure colours much better than phosphors. One mechanism of utilizing QD-FRET within
1104
LED technology is to prepare mixed films of two colours of QD, which absorb light from an
1105
underlying near-UV InGaN/GaN LED with FRET between the two populations of QDs. The
1106
energy transfer has the effect of fine tuning colour and increasing quantum efficiency ca. 10-
1107
15% by recycling trapped excitons [189-191]. Demir and coworkers have also shown that
1108
adjusting the spacing between donor and acceptor QDs (peak PL at 595 and 645 nm) in bilayer
1109
assemblies can systematically shift chromaticity coordinates for the overall emission from the
1110
thin film [195]. QD-FRET is also a possible means by which to access efficient green and yellow
Ac ce p
1097
38 Page 38 of 63
LEDs. While alloyed InGaN/GaN and AlGaInP are optimal materials for blue and red LEDs,
1112
respectively, they both suffer from significantly decreased internal quantum efficiency and
1113
luminous efficiency when tuned to intermediate visible wavelengths. Bright yellow LEDs could
1114
be accessed using a mixed film of QDs with peak PL at 490 and 548 nm on a InGaN/GaN LED
1115
[191]. Most recently, Demir‘s group has also shown that FRET from an InGaN/GaN quantum
1116
well can also contribute to sensitization of a bilayer of green and red QDs (peak PL at 550 and
1117
600 nm) provided that the GaN cap is sufficiently thin (~ 3 nm) [192]. The efficiency of exciton
1118
migration from the quantum well was 83%, and the green/red QD bilayer structure enhanced
1119
exciton transfer efficiency by 64% compared to a bilayer of only red QDs.
us
cr
ip t
1111
1120
QD-FRET has also been exploited in developing novel luminescent materials. For example,
1122
Liang et al. fabricated ultrathin films with uniform morphology and controllable thickness based
1123
on the alternating assembly of layered double hydroxide (LDH) nanosheets with CdTe QDs
1124
[196]. A thin film with 40 LDH/green QD (peak PL at 535 nm) layers assembled with 24
1125
LDH/red QD (peak PL at 635 nm) layers exhibited 2-fold brighter red PL than 60 layers of
1126
LDH/red QD. This enhancement was attributed to FRET, which could be manipulated by
1127
controlling the order of green and red QD layers in the thin films. The FRET efficiency was 56%
1128
in thin films with five [LDH/green QD/LDH/red QD] units, and an increase in FRET efficiency
1129
to 91% was possible with three [LDH/green QD/(LDH/red QD)2] units. Both configurations
1130
provided substantially more (~7-fold) red QD PL than equivalent systems with only red QDs.
1131
Bednarkiewicz et al. prepared mixed films of upconverting Er3+/Yb3+:NaYF4 nanoparticle donors
1132
and CdSe QD acceptors [197]. When excited at 976 nm, the Er3+/Yb3+:NaYF4 nanoparticles
1133
exhibited green luminescence at 540 nm, which was resonant with the CdSe QDs (peak PL at
1134
580 nm). The FRET efficiency was only 14.8% due to the low quantum yield of the donors,
1135
resulting in a Förster distance of only 1.5 nm. Nonetheless, configurations of this type are
1136
interesting due to their potential to convert IR light into specific colours of visible light by
1137
selection of specific QDs. These and other QD-FRET-based luminescent materials could have
1138
applications in LED or other solid-state lighting devices.
Ac ce p
te
d
M
an
1121
39 Page 39 of 63
8. Summary and Perspective
1140
This review has described a variety of non-traditional FRET configurations incorporating QDs.
1141
We defined traditional QD-FRET pairs as those with QD donors and organic dye acceptors,
1142
where energy transfer occurred in a single step. One set of non-traditional QD-FRET
1143
configurations included single-step energy transfer with substitution of conventional organic
1144
dyes by photochromic dyes as acceptors, luminescent lanthanide complexes as donors,
1145
polypyridyl or azamacrocyclic metal complexes as acceptors, and Au NPs, GO, or CNTs as
1146
multidimensional acceptors. Additional non-traditional QD-FRET configurations included linear
1147
or concentric multi-step energy transfer relays that incorporated organic dyes and lanthanide
1148
complexes, as well as multicolour thin films and composite materials of QDs. Studies to date
1149
have ranged from experimental and theoretical characterization of these FRET systems, to
1150
applications such as multiplexed biosensing, preliminary work toward PDT, improved solar
1151
energy conversion, and fabrication of optoelectronic devices.
an
us
cr
ip t
1139
M
1152
Taken altogether, the non-traditional QD-FRET configurations discussed here comprise a highly
1154
versatile toolkit for the design and engineering of nanoscale devices. pcFRET provides the
1155
opportunity for optical switching and FRET between lanthanide complexes and QDs is
1156
potentially useful for timing optical signals. Both of these capabilities are critical to developing
1157
photonic logic circuits. ―Rainbow‖ thin films of QDs can efficiently downconvert short
1158
wavelength or broadband light to long wavelengths, while composite films of QDs with
1159
upconverting materials can achieve the opposite. FRET relays are also a means of transferring
1160
energy over length scales of 10–20 nm and Au NP or carbon nanomaterials can be used to
1161
quench excitation energy at similar distances. Furthermore, energy transfer from QDs to metal
1162
complexes can convert optical signals into chemical reactivity, as seen with the generation of
1163
singlet oxygen and nitric oxide. The inverse, conversion of chemical reactivity into optical
1164
signals, can be achieved through bioluminescence and chemiluminescence resonance energy
1165
transfer (BRET/CRET). Although beyond our scope here, an array of such BRET/CRET
1166
configurations with QDs have been demonstrated by Rao‘s group [35] and Willner‘s group [34,
1167
198-201], respectively. Albeit ambitious, one can imagine the future integration of many of these
1168
capabilities through top-down and bottom-up assembly to create complex nano-machines.
Ac ce p
te
d
1153
1169
40 Page 40 of 63
Although there is a predominance of traditional QD-FRET configurations and emphasis on
1171
biological applications in the literature, it is important to recognize that QD-FRET research is
1172
multidisciplinary and diverse. Herein, we have sought to highlight the capabilities and evolving
1173
role of QD-FRET beyond the typical norms. New, non-traditional QD-FRET configurations with
1174
novel properties will be able to help address challenges in healthcare, energy, and other areas of
1175
need.
1176
Acknowledgements.
1177
W.R.A. and H.K. acknowledge the Natural Sciences and Engineering Research Council of
1178
Canada (NSERC), the Canada Foundation for Innovation (CFI), the Peter Wall Institute for
1179
Advanced Studies, and the University of British Columbia for support of their research. W.R.A.
1180
is also grateful for a Canada Research Chair (Tier 2). I.L.M. acknowledges the U.S. Naval
1181
Research Laboratory Nanosciences Institute and the Defense Threat Reduction Agency Joint
1182
Science and Technology Office Military Interdepartmental Purchase Request # B112582M.
M
an
us
cr
ip t
1170
1183
Ac ce p
te
d
1184
41 Page 41 of 63
1185
References
1186
te
d
M
an
us
cr
ip t
[1] W.R. Algar, K. Susumu, J.B. Delehanty, I.L. Medintz, Anal. Chem. 83 (2011) 8826-8837. [2] A.M. Smith, S.M. Nie, Acc. Chem. Res. 43 (2010) 190-200. [3] A.P. Alivisatos, J. Phys. Chem. 100 (1996) 13226-13239. [4] S.J. Rosenthal, J.C. Chang, O. Kovtun, J.R. McBride, I.D. Tomlinson, Chem. Biol. 18 (2011) 10-24. [5] H. Mattoussi, G. Palui, H.B. Na, Adv. Drug. Deliver. Rev. 64 (2012) 138-166. [6] E. Petryayeva, W.R. Algar, I.L. Medintz, Appl. Spectrosc. 67 (2013) 215-252. [7] P. Zrazhevskiy, M. Sena, X.H. Gao, Chem. Soc. Rev. 39 (2010) 4326-4354. [8] V. Biju, T. Itoh, M. Ishikawa, Chem. Soc. Rev. 39 (2010) 3031-3056. [9] B.Y.S. Kim, J.T. Rutka, W.C.W. Chan, New Engl. J. Med. 363 (2010) 2434-2443. [10] Z. Jin, N. Hildebrandt, Trends Biotechnol. 30 (2012) 394-403. [11] R.A. Sperling, W.J. Parak, Phil. Trans. R. Soc. A 368 (2010) 1333-1383. [12] W.R. Algar, D.E. Prasuhn, M.H. Stewart, T.L. Jennings, J.B. Blanco-Canosa, P.E. Dawson, I.L. Medintz, Bioconjugate Chem. 22 (2011) 825-858. [13] C.B. Murray, C.R. Kagan, M.G. Bawendi, Annu. Rev. Mater. Sci. 30 (2000) 545-610. [14] N. Tomczak, D. Jańczewski, M. Han, G.J. Vancso, Progress Polymer Sci. 34 (2009) 393430. [15] D.V. Talapin, J.S. Lee, M.V. Kovalenko, E.V. Shevchenko, Chem. Rev. 110 (2010) 389458. [16] R.F.M. DeAlmeida, L.M.S. Loura, M. Prieto, Chem. Phys. Lipids 157 (2009) 61-77. [17] B. Schuler, W.A. Eaton, Curr. Opin. Struct. Biol. 18 (2008) 16-26. [18] V.V. Didenko, Biotechniques 31 (2001) 1106-1121. [19] J.P. Goddard, J.L. Reymond, Trends Biotechnol. 22 (2004) 363-370. [20] F. Ciruela, Curr. Opin. Biotechnol. 19 (2008) 338-343. [21] R. Roy, S. Hohng, T. ha, Nat. Methods 5 (2008) 507-516. [22] D.W. Piston, G.J. Kremers, Trends Biochem. Sci. 32 (2007) 407-414. [23] E.A. Jares-Erijman, T.M. Jovin, Nat. Biotechnol. 21 (2003) 1387-1395. [24] R. Freeman, I. Willner, Chem. Soc. Rev. 41 (2012) 4067-4085. [25] I.L. Medintz, H. Mattoussi, Phys. Chem. Chem. Phys. 165 (2009) 17-45. [26] W.R. Algar, A.J. Tavares, U.J. Krull, Anal. Chim. Acta 673 (2010) 1-25. [27] T. Förster, Naturwissenschaften 33 (1946) 166-175. [28] T. Förster, Discuss. Faraday Soc. 27 (1959) 7-17. [29] T. Förster, Intermolecular energy migration and fluorescence {Ann. Phys. 2, 55 (1948, translation)}, in: E.V. Mielczarek, E. Greenbaum, R.S. Knox (Eds.) Biological Physics, American Institute of Physics, New York, 1993. [30] T. Förster, J. Biomed. Opt. 17 (2012) 011002. [31] H.C. Cheung, Resonance energy transfer, in: J.R. Lakowicz (Ed.) Topics in Fluorescence Spectroscopy, Plenum Press, New York, 1991, pp. 127-176. [32] B.W. VanderMeer, G. Coker, S.Y.S. Chen, Resonance Energy Transfer: Theory and Data, VCH Publishers, New York, 1994. [33] L.J. Charbonnière, N. Hildebrandt, Eur. J. Inorg. Chem. (2008) 3241-3251. [34] R. Freeman, X.Q. Liu, I. Willner, J. Am. Chem. Soc. 133 (2011) 11597-11604.
Ac ce p
1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228
42 Page 42 of 63
te
d
M
an
us
cr
ip t
[35] Z.Y. Xia, J.H. Rao, Curr. Opin. Biotechnol. 20 (2009) 37-44. [36] A.L. Rogach, T.A. Klar, J.M. Lupton, A. Meijerink, J. Feldmann, J. Mater. Chem. 19 (2009) 1208-1221. [37] M. Tomasulo, I. Yildiz, F.M. Raymo, Inorg. Chim. Acta 360 (2007) 938-944. [38] K. Matsuda, M. Irie, J. Photochem. Photobiol. C 5 (2004) 169-182. [39] F.M. Raymo, M. Tomasulo, Chem. Soc. Rev. 34 (2005) 327-336. [40] F.M. Raymo, M. Tomasulo, Chem. Eur. J. 12 (2006) 3186-3193. [41] M. Irie, Chem. Rev. 100 (2000) 1685-1716. [42] L. Giordano, T.M. Jovin, M. Irie, E.A. Jares-Erijman, J. Am. Chem. Soc. 124 (2002) 74817489. [43] R. Klajn, J.F. Stoddart, B.A. Grzybowksi, Chem. Soc. Rev. 39 (2010) 2203-2237. [44] I.L. Medintz, S.A. Trammell, H. Mattoussi, J.M. Mauro, J. Am. Chem. Soc. 126 (2003) 3031. [45] L. Zhu, M.Q. Zhu, J.K. Hurst, A.D.Q. Li, J. Am. Chem. Soc. 127 (2005) 8968-8970. [46] M. Tomasulo, I. Yildiz, F.M. Raymo, Aust. J. Chem. 59 (2006) 175-178. [47] M. Tomasulo, F.M. Raymo, Inorg. Chim. Acta 360 (2007) 938-944. [48] E.M. Lee, S.Y. Gwon, Y.A. Son, S.H. Kim, Spectrochim. Acta A 97 (2012) 699-702. [49] E.M. Lee, S.Y. Gwon, Y.A. Son, S.H. Kim, Spectrochim. Acta A 97 (2012) 806-810. [50] E. Jares-Erijman, L. Giordano, C. Spagnuolo, K. Lidke, T.M. Jovin, Mol. Cryst. Liq. Cryst. 430 (2005) 257-265. [51] S.A. Díaz, G.O. Menéndez, M.H. Etchehon, L. Giordano, T.M. Jovin, E.A. Jares-Erijman, ACS Nano 5 (2011) 2795-2805. [52] S.A. Díaz, L. Giordano, T.M. Jovin, E.A. Jares-Erijman, Nano Lett. 12 (2012) 3537-3544. [53] Z. Erno, I. Yildiz, B. Gorodetsky, F.M. Raymo, N.R. Branda, Photochem. Photobiol. Sci. 9 (2010) 249-253. [54] B. Huang, M. Bates, X. Zhuang, Annu. Rev. Biochem. 78 (2009) 993-1016. [55] M. Heilemann, P. Dedecker, J. Hofkens, M. Sauer, Laser Photonics Rev. 3 (2009) 180-202. [56] T. Dertinger, R. Colyer, G. Iyer, S. Weiss, J. Enderlein, Proc. Natl. Acad. Sci. USA 106 (2009) 22287-22292. [57] K.A. Lidke, B. Rieger, T.M. Jovin, R. Heintzmann, Optics Express 13 (2005) 7052-7062. [58] A.P. deSilva, N.D. McClenaghan, Chem. Eur. J. 10 (2004) 574-586. [59] N. Sabbatini, M. Guardigli, J.M. Lehn, Coord. Chem. Rev. 123 (1993) 201-228. [60] J.C.G. Bünzli, C. Piguet, Chem. Soc. Rev. 34 (2005) 1048-1077. [61] I. Hemmilä, V. Laitala, J. Fluoresc. 15 (2005) 529-542. [62] I. Hemmilä, J. Biomol. Screen. 4 (1999) 303-307. [63] G. Mathis, Clin. Chem. 41 (1995) 1391-1397. [64] G. Mathis, J. Biomol. Screen. 4 (1999) 309-313. [65] N. Hildebrandt, L.J. Charbonnière, M. Beck, R.F. Ziessel, H.-G. Löhmannsröben, Angew. Chem. Int. Ed. 44 (2005) 7612-7615. [66] L.J. Charbonnière, N. Hildebrandt, R.F. Ziessel, H.-G. Löhmannsröben, J. Am. Chem. Soc. 128 (2006) 12800-12809. [67] A.R. Clapp, I.L. Medintz, B.R. Fisher, G.P. Anderson, H. Mattoussi, J. Am. Chem. Soc. 127 (2005) 1242-1250. [68] F. Morgner, D. Geißler, S. Stufler, N.G. Butlin, H.-G. Löhmannsröben, N. Hildebrandt, Angew. Chem. Int. Ed. 49 (2010) 7570-7574.
Ac ce p
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273
43 Page 43 of 63
te
d
M
an
us
cr
ip t
[69] K.D. Wegner, P.T. Lanh, T. Jennings, E. Oh, V. Jain, S.M. Fairclough, J.M. Smith, E. Giovanelli, N. Lequeux, T. Pons, N. Hildebrandt, ACS Appl. Mater. Inter. (2013) DOI: 10.1021/am3030728. [70] D. Geißler, L.J. Charbonnière, R.F. Ziessel, N.G. Butlin, H.-G. Löhmannsröben, N. Hildebrandt, Angew. Chem. Int. Ed. 49 (2010) 1396-1401. [71] W.R. Algar, D. Wegner, A.L. Huston, J.B. Blanco-Canosa, M.H. Stewart, A. Armstrong, P.E. Dawson, N. Hildebrandt, I.L. Medintz, J. Am. Chem. Soc. 134 (2012) 1876-1891. [72] H. Härmä, T. Soukka, A. Shavel, N. Gaponik, H. Weller, Anal. Chim. Acta 604 (2007) 177183. [73] M.J. Chen, Y.S. Wu, G.F. Lin, J.Y. Hou, M. Li, T.C. Liu, Anal. Chim. Acta 741 (2012) 100-105. [74] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. Vonzelewsky, Coord. Chem. Rev. 84 (1988) 85-277. [75] K. Kalyanasundaram, M. Gratzel, Coord. Chem. Rev. 177 (1998) 347-414. [76] J. Huang, D. Stockwell, Z. Huang, D.L. Mohler, T. Lian, J. Am. Chem. Soc. 130 (2008) 5632-5633. [77] M. Sykora, M.A. Petruska, J. Alstrum-Acevedo, I. Bezel, T.J. Meyer, V.I. Klimov, J. Am. Chem. Soc. 128 (2006) 9984-9985. [78] I.L. Medintz, T. Pons, S.A. Trammell, A.F. Grimes, D.S. English, J.B. Blanco-Canosa, P.E. Dawson, H. Mattoussi, J. Am. Chem. Soc. 130 (2008) 16745-16756. [79] A.J. Morris-Cohen, K.O. Aruda, A.M. Rasmussen, G. Canzi, T. Seideman, C.P. Kubiak, E.A. Weiss, Phys. Chem. Chem. Phys. 14 (2012) 13794-13801. [80] E.J. McLaurin, A.B. Greytak, M.G. Bawendi, D.G. Nocera, J. Am. Chem. Soc. 131 (2009) 12994-13001. [81] M.H. Stewart, A.L. Huston, A.M. Scott, A.L. Efros, J.S. Melinger, K.B. Gemmill, S.A. Trammell, J.B. Blanco-Canosa, P.E. Dawson, I.L. Medintz, ACS Nano 6 (2012) 5330-5343. [82] A.W. Hains, Z. Liang, M.A. Woodhouse, B.A. Gregg, Chem. Rev. 110 (2010) 6689-6735. [83] R. Bonnett, Chem. Soc. Rev. 24 (1995) 19-33. [84] L.M. Moreira, F.V. DosSantos, J.P. Lyon, M. Maftoum-Costa, C. Pacheco-Soares, N.S. daSilva, Aust. J. Chem. 61 (2008) 741-754. [85] S. Dayal, Y. Lou, A.C.S. Samia, J.C. Berlin, M.E. Kenney, C. Burda, J. Am. Chem. Soc. 128 (2006) 13974-13975. [86] S. Dayal, C. Burda, J. Am. Chem. Soc. 129 (2007) 7977-7981. [87] S. Dayal, J. Li, Y.S. Li, H. Wu, A.C.S. Samia, M.E. Kenney, C. Burda, Photochem. Photobiol. 84 (2008) 243-249. [88] S. Dayal, C. Burda, J. Am. Chem. Soc. 130 (2008) 2890-2891. [89] Y. Wen, W. Song, L. An, Y. Liu, Y. Wang, Y. Yang, Appl. Phys. Lett. 95 (2009) 143702. [90] M. Idowu, T. Nyokong, J. Luminesc. 129 (2009) 356-362. [91] M. Idowu, T. Nyokong, Polyhedron 28 (2009) 891-896. [92] A. Erdoğmuş, S. Moeno, C. Litwinski, T. Nyokong, J. Photochem. Photobiol. A 210 (2010) 200-208. [93] S. Moeno, E. Antunes, S. Khene, C. Litwinski, T. Nyokong, Dalton Trans. 39 (2010) 34603471. [94] J. Britton, E. Antunes, T. Nyokong, J. Photochem. Photobiol. A 210 (2010) 1-7. [95] N. Masilela, T. Nyokong, J. Photochem. Photobiol. A 247 (2012) 82-92. [96] M. Idowu, J.Y. Chen, T. Nyokong, New J. Chem. 32 (2008) 290-296.
Ac ce p
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319
44 Page 44 of 63
te
d
M
an
us
cr
ip t
[97] S. Moeno, T. Nyokong, J. Photochem. Photobiol. A 201 (2009) 228-236. [98] A.M. Smith, M.C. Mancini, S. Nie, Nat. Nanotechnol. 4 (2009) 710-711. [99] J.M. Tsay, M. Trzoss, L. Shi, X. Kong, M. Selke, M.E. Jung, S. Weiss, J. Am. Chem. Soc. 129 (2007) 6865-6871. [100] D.A. Tekdaş, M. Durmuş, H. Yanik, V. Ahsen, Spectrochim. Acta A 93 (2012) 313-320. [101] Z.D. Qi, D.W. Li, P. Jiang, F.L. Jiang, Y.S. Li, Y. Liu, W.K. Wong, K.W. Cheah, J. Mater. Chem. 21 (2011) 2455-2458. [102] P. Pacher, J.S. Beckman, L. Liaudet, Physiol. Rev. 87 (2007) 315-424. [103] G. Ferrer-Sueta, R. Radi, ACS Chem. Biol. 4 (2009) 161-177. [104] M. DeRidder, D. Verellen, V. Verovski, G. Storme, Nitric Oxide Biol. Chem. 19 (2008) 164-169. [105] B. Fitzpatrick, M. Mehibel, R.L. Cowen, I.J. Stratford, Nitric Oxide Biol. Chem. 19 (2008) 217-224. [106] P.C. Ford, Acc. Chem. Res. 41 (2008) 190-200. [107] D. Neuman, A.D. Ostrowski, R.O. Absalonson, G.F. Strouse, P.C. Ford, J. Am. Chem. Soc. 129 (2007) 4146-4147. [108] D. Neuman, A.D. Ostrowski, A.A. Mikhailovsky, R.O. Absalonson, G.F. Strouse, P.C. Ford, J. Am. Chem. Soc. 130 (2008) 168-175. [109] P.T. Burks, A.D. Ostrowski, A.A. Mikhailovsky, E.M. Chang, P.S. Wagenknecht, P.C. Ford, J. Am. Chem. Soc. 134 (2012) 13266-13275. [110] G. de la Torre, C.G. Claessens, T. Torres, Chem. Commun. (2007) 2000-2015. [111] M. Grätzel, Acc. Chem. Res. 42 (2010) 1788-1798. [112] W.M. Campbell, A.K. Burrell, D.L. Officer, K.W. Jolley, Coord. Chem. Rev. 248 (2004) 1363-1379. [113] L.L. Li, E.W.G. Diau, Chem. Soc. Rev. 42 (2013) 291-304. [114] H. Imahori, T. Umeyama, S. Ito, Acc. Chem. Res. 42 (2009) 1809-1818. [115] S. Jin, H.J. Son, O.K. Farha, G.P. Wiederrecht, J.T. Hupp, J. Am. Chem. Soc. 135 (2013) 955-958. [116] A. Boulesbaa, Z. Huang, D. Wu, T. Lian, J. Phys. Chem. C 114 (2010) 962-969. [117] A. Boulesbaa, A. Issac, D. Stockwell, Z. Huang, J. Huang, J. Guo, T. Lian, J. Am. Chem. Soc. 129 (2007) 15132-15133. [118] J. Huang, Z. Huang, Y. Yang, H. Zhu, T. Lian, J. Am. Chem. Soc. 132 (2010) 4858-4864. [119] I.L. Medintz, M.H. Stewart, S.A. Trammell, K. Susumu, J.B. Delehanty, B.C. Mei, J.S. Melinger, J.B. Blanco-Canosa, P.E. Dawson, H. Mattoussi, Nat. Mater. 9 (2010) 676-684. [120] X. Ji, G. Palui, T. Avellini, H.B. Na, C.Y. Yi, K.L. Knappenberger, H. Mattoussi, J. Am. Chem. Soc. 134 (2012) 6006-6017. [121] S.R. Opperwall, A. Divakaran, E.G. Porter, J.A. Christians, A.J. DenHartigh, D.E. Benson, ACS Nano 6 (2012) 8078-8086. [122] C. Curutchet, A. Franceschetti, A. Zunger, G.D. Scholes, J. Phys. Chem. C 112 (2008) 13336-13341. [123] C. Zhang, J. Xu, S. Zhang, X. Ji, Z. He, Chem. Eur. J. 18 (2012) 8296-8300. [124] H. Dong, W. Gao, F. Yan, H. Ji, H. Ju, Anal. Chem. 82 (2010) 5511-5517. [125] J. Tian, H. Zhao, M. Liu, Y. Chen, X. Quan, Anal. Chim. Acta 723 (2012) 83-87. [126] M. Liu, H. Zhao, X. Quan, S. Chen, X. Fan, Chem. Commun. 46 (2010) 7909-7911. [127] E. Morales-Narváez, B. Pérez-López, L.B. Pires, A. Merkoçi, Carbon 50 (2012) 29872993.
Ac ce p
1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365
45 Page 45 of 63
te
d
M
an
us
cr
ip t
[128] E. Shafran, B.D. Mangum, J.M. Gerton, Nano Lett. 10 (2010) 4049-4054. [129] S. Jander, A. Kornowski, H. Weller, Nano Lett. 11 (2011) 5179-5183. [130] Z. Chen, S. Berciaud, C. Nuckolls, T.F. Heinz, L.E. Brus, ACS Nano 4 (2010) 2964-2968. [131] S. Kaniyankandy, S. Rawalekar, H.N. Ghosh, J. Phys. Chem. C 116 (2012) 16271-16275. [132] D. Zhang, L. Gan, Y. Cao, Q. Wang, L. Qi, X. Guo, Adv. Mater. 24 (2012) 2715-2720. [133] I.V. Lightcap, P.V. Kamat, J. Am. Chem. Soc. 134 (2012) 7109-7116. [134] G. Mountrichas, A.S.D. Sandanayaka, S.P. Economopoulos, S. Pispas, O. Ito, T. Hasobe, N. Tagmatarchis, J. Mater. Chem. 19 (2009) 8990-8998. [135] S. Leubner, G. Katsukis, D.M. Guldi, Faraday Discuss. 155 (2012) 253-265. [136] E. Oh, M.Y. Hong, D. Lee, S.H. Nam, H.C. Yoon, H.S. Kim, J. Am. Chem. Soc. 127 (2005) 3270-3271. [137] B. Tang, L.H. Cao, K.H. Xu, L.H. Zhuo, J.H. Ge, Q.F. Li, L.J. Yu, Chem. Eur. J. 14 (2008) 3637-3644. [138] Y.P. Kim, Y.H. Oh, E. Oh, S. Ko, M.K. Han, H.S. Kim, Anal. Chem. 80 (2008) 46344641. [139] J. Liu, J.H. Lee, Y. Lu, Anal. Chem. 79 (2007) 4120-4125. [140] E. Dulkeith, A.C. Morteani, T. Niedereichholz, T.A. Klar, J. Feldmann, S.A. Levi, F.C.J.M. vanVeggel, D.N. Reinhoudt, M. Möller, D.I. Gittins, Phys. Rev. Lett. 89 (2002) 203002. [141] E. Dulkeith, M. Ringler, T.A. Klar, J. Feldmann, A. Muñoz Javier, W.J. Parak, Nano Lett. 5 (2005) 585-589. [142] J. Gersten, A. Nitzan, J. Chem. Phys. 75 (1981) 1139-1152. [143] T.L. Jennings, M.P. Singh, G.F. Strouse, J. Am. Chem. Soc. 128 (2006) 5462-5467. [144] M.P. Singh, G.F. Strouse, J. Am. Chem. Soc. 132 (2010) 9383-9391. [145] C.S. Yun, A. Javier, T. Jennings, M. Fisher, S. Hira, S. Peterson, B. Hopkins, N.O. Reich, G.F. Strouse, J. Am. Chem. Soc. 127 (2004) 3115-3119. [146] R.R. Chance, A. Prock, R. Silbey, Adv. Chem. Phys. 37 (1978) 1-65. [147] B.N.J. Persson, J. Phys. C. 11 (1978) 4251-4269. [148] B.N.J. Persson, N.D. Lang, Phys. Rev. B 26 (1982) 5409-5415. [149] Z. Gueroui, A. Libchaber, Phys. Rev. Lett. 93 (2004) 166108. [150] T. Pons, I.L. Medintz, K.E. Sapsford, S. Higashiya, A.F. Grimes, D.S. English, H. Mattoussi, Nano Lett. 7 (2007) 3157-3164. [151] M. Li, S.K. Cushing, Q. Wang, X. Shi, L.A. Hornak, Z. Hong, N. Wu, J. Phys. Chem. Lett. 2 (2011) 2125-2129. [152] H. Han, V. Valle, M.M. Maye, J. Phys. Chem. C 116 (2012) 22996-23003. [153] X. Zhang, C.A. Marocico, M. Lunz, V.A. Gerard, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, A.S. Susha, A.L. Rogach, A.L. Bradley, ACS Nano 6 (2012) 9283-9290. [154] J. Griffin, A.K. Singh, D. Senapati, P. Rhodes, K. Mitchell, B. Robinson, E. Yu, P.C. Ray, Chem. Eur. J. 15 (2009) 342-351. [155] R. Carminati, J.J. Greffet, C. Henkel, J.M. Vigoureux, Opt. Commun. 261 (2006) 368-375. [156] I.L. Medintz, A.R. Clapp, H. Mattoussi, E.R. Goldman, B. Fisher, J.M. Mauro, Nat. Mater. 2 (2003) 630-638. [157] H. Lu, O. Schöps, U. Woggon, C.M. Niemeyer, J. Am. Chem. Soc. 130 (2008) 4815-4827. [158] K. Boeneman, D.E. Prasuhn, J.B. Blanco-Canosa, P.E. Dawson, J.S. Melinger, M. Ancona, M.H. Stewart, K. Susumu, A. Huston, I.L. Medintz, J. Am. Chem. Soc. 132 (2010) 1817718190.
Ac ce p
1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411
46 Page 46 of 63
te
d
M
an
us
cr
ip t
[159] W.R. Algar, A.P. Malanoski, K. Susumu, M.H. Stewart, N. Hildebrandt, I.L. Medintz, Anal. Chem. 84 (2012) 10136-10146. [160] W.R. Algar, M.G. Ancona, A.P. Malanoski, K. Susumu, I.L. Medintz, ACS Nano 6 (2012) 11044-11058. [161] A.K. Hagan, T. Zuchner, Anal. Bioanal. Chem. 400 (2011) 2847-2864. [162] K. Hanaoka, K. Kikuchi, S. Kobayashi, T. Nagano, J. Am. Chem. Soc. 129 (2007) 1350213509. [163] W.R. Algar, U.J. Krull, Anal. Bioanal. Chem. 398 (2010) 2439-2449. [164] M.O. Noor, E. Petryayeva, A.J. Tavares, U. Uddayasankar, W.R. Algar, U.J. Krull, Coord. Chem. Rev. this issue. (2013). [165] C.R. Kagan, C.B. Murray, M.G. Bawendi, Phys. Rev. B 54 (1996) 8633-8643. [166] S.A. Crooker, J.A. Hollingsworth, S. Tretiak, V.I. Klimov, Phys. Rev. Lett. 89 (2002) 186802. [167] V.K. Komarala, A.L. Bradley, Y.P. Rakovich, S.J. Byrne, Y.K. Gun'ko, A.L. Rogach, Appl. Phys. Lett. 93 (2008) 123102. [168] M. Lunz, A.L. Bradley, W.Y. Chen, V.A. Gerard, S.J. Byrne, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, Phys. Rev. B 81 (2010) 205316. [169] M. Lunz, A.L. Bradley, W.Y. Chen, Y.K. Gun'ko, J. Phys. Chem. C 113 (2009) 30843088. [170] M. Lunz, A.L. Bradley, V.A. Gerard, S.J. Byrne, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, Phys. Rev. B 83 (2011) 115423. [171] M. Lunz, V.A. Gerard, Y.K. Gun'ko, V. Lesnyak, N. Gaponik, A.S. Susha, A.L. Rogach, A.L. Bradley, Nano Lett. 11 (2011) 3341-3345. [172] P.K. Wolber, B.S. Hudson, Biophys. J. 28 (1979) 197-210. [173] C.H. Wang, C.W. Chen, Y.T. Chen, C.M. Wei, Y.F. Chen, C.W. Lai, M.L. Ho, P.T. Chou, M. Hofmann, Appl. Phys. Lett. 96 (2010) 071906. [174] A.A. Mamedov, A. Belov, M. Giersig, N.N. Mamedova, N.A. Kotov, J. Am. Chem. Soc. 123 (2001) 7738-7739. [175] T. Franzl, T.A. Klar, S. Schietinger, A.L. Rogach, J. Feldmann, Nano Lett. 4 (2004) 15991603. [176] T.A. Klar, T. Franzl, A.L. Rogach, J. Feldmann, Adv. Mater. 17 (2005) 769-773. [177] Y. Ooyama, Y. Harima, Chem. Phys. Chem. 13 (2012) 4032-4080. [178] X. Chen, C. Li, M. Grätzel, R. Kostecki, S.S. Mao, Chem. Soc. Rev. 41 (2012) 7909-7937. [179] S. Buhbut, S. Itzhakov, D. Oron, A. Zaban, J. Phys. Chem. Lett. 2 (2011) 1917-1924. [180] L. Etgar, J. Park, C. Barolo, V. Lesnyak, S.K. Panada, P. Quagliotto, S.G. Hickey, M.K. Nazeeruddin, A. Eychmüller, G. Viscardi, M. Grätzel, RSC Adv. 2 (2012) 2748-2752. [181] A. Ruland, C. Schulz-Drost, V. Sgobba, D.M. Guldi, Adv. Mater. 23 (2011) 4573-4577. [182] S. Rühle, M. Shalom, A. Zaban, Chem. Phys. Chem. 11 (2010) 2290-2304. [183] P.K. Santra, P.V. Kamat, J. Am. Chem. Soc. 135 (2013) 877-885. [184] S. Choi, H. Jin, J. Bang, S. Kim, J. Phys. Chem. Lett. 3 (2012) 3442-3447. [185] L. Beverina, P. Salice, Eur. J. Org. Chem. (2010) 1207-1225. [186] S. Buhbut, S. Itzhakov, E. Tauber, M. Shalom, I. Hod, T. Geiger, Y. Garini, D. Oron, A. Zaban, ACS Nano 4 (2010) 1293-1298. [187] S. Itzhakov, S. Buhbut, E. Tauber, T. Geiger, A. Zaban, D. Oron, Adv. Energy Mater. 1 (2011) 626-633. [188] H. Choi, P.K. Santra, P.V. Kamat, ACS Nano 6 (2012) 5718-5726.
Ac ce p
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457
47 Page 47 of 63
te
d
M
an
us
cr
ip t
[189] S. Nizamoglu, H.V. Demir, Opt. Express 16 (2008) 13961-13968. [190] S. Nizamoglu, H.V. Demir, Appl. Phys. Lett. 95 (2009) 151111. [191] S. Nizamoglu, E. Sari, J.H. Baek, I.H. Lee, H.V. Demir, IEEE J. Sel. Top. Quant. Elect. 15 (2009) 1163-1170. [192] S. Nizamoglu, P.L. Hernández-Martínez, E. Mutlugun, D.U. Karatay, H.V. Demir, Appl. Phys. Lett. 100 (2012) 241109. [193] H.V. Demir, S. Nizamoglu, T. Erdem, E. Mutlugun, N. Gaponik, A. Eychmüller, Nano Today 6 (2011) 632-647. [194] S. Nizamoglu, H.V. Demir, J. Appl. Phys. 105 (2009) 083112. [195] N. Cicek, S. Nizamoglu, T. Ozel, E. Mutlugun, D.U. Karatay, V. Lesnyak, T. Otto, N. Gaponik, A. Eychmüller, H.V. Demir, Appl. Phys. Lett. 94 (2009) 061105. [196] R. Liang, S. Xu, D. Yan, W. Shi, R. Tian, H. Yan, M. Wei, D.G. Evans, X. Duan, Adv. Func. Mater. 22 (2012) 4940-4948. [197] A. Bednarkiewicz, M. Nyk, M. Samoic, W. Strek, J. Phys. Chem. C 114 (2010) 1753517541. [198] R. Freeman, J. Girsh, A.F.J. Jou, J.A.A. Ho, T. Hug, J. Dernedde, I. Willner, Anal. Chem. 84 (2012) 6192-6198. [199] R. Freeman, B. Willner, I. Willner, J. Phys. Chem. Lett. 2 (2011) 2667-2677. [200] E. Golub, A. Niazov, R. Freeman, M. Zatsepin, I. Willner, J. Phys. Chem. C 116 (2012) 13827-13834. [201] X.Q. Liu, R. Freeman, E. Golub, I. Willner, ACS Nano 5 (2011) 7648-7655.
Ac ce p
1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481
48 Page 48 of 63
Figure Captions
us
cr
ip t
Figure 1. Photochromic FRET with QD donors. (A) pcFRET between a CdSe/ZnS QD and BIPS dye. The BIPS dyes are labels on maltose binding protein (MBP) self-assembled to the QD through a pentahistidine tail (5HIS). UV light converts the dye to the coloured merocyanine form to activate FRET; visible light converts the dye to the colourless spiropyran form to inactivate FRET. Reprinted with permission from ref. [44]. Copyright 2003 American Chemical Society. (B) Example of photochromic changes in spectral overlap between a CdTe QD and WSPO: (left) strong overlap with the merocyanine form; (right) minimal overlap with the spiropyran form. Reprinted with permission from ref. [48]. Copyright 2012 Elsevier. (C) pcFRET between a CdSe/ZnS QD and a photochromic dithienylethene dye: (i) the dyes were pendant groups on an amphiphilic polymer coating; (ii) photoswitching of QD PL. Figure reprinted with permission from ref. [51]. Copyright 2011 American Chemical Society.
te
d
M
an
Figure 2. FRET from luminescent Tb3+ complexes (LTbC) to QDs. (A) Assembly of LTbClabeled streptavidin (LS) to biotinylated QDs. (B) LTbC PL decay curves for (i) biotinylated QDs assembled at different ratios with LTbC-streptavidin (SAv; ~4.2 LTbC per SAv) and (ii) LTbC-peptides self-assembled to QDs at different ratios. The quenched LTbC decay component (a) and native LTbC decay component (b) are indicated in both panels. The native contribution increases as the relative amount of QDs decreases. (C) Spectral multiplexing with LTbC donor and five different QD acceptors: (i) LTbC emission and composite PL spectra for LTbC with QDs (peak PL at 529, 565, 604, 653, 712 nm); (ii) emission contributions from the QDs are highlighted. Panels (A) and (C) reprinted with permission from ref. [70]. Copyright 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. Panel (B) reprinted with permission from refs. [69, 71]. Copyright 2012 and 2013 American Chemical Society. Figure 3. FRET from QD donors to metal complexes. (A) FRET from a CdSe/ZnS QD donor to an Os(II)PP complex: (i) illustration of the conjugate, where the Os(II)PP complex is a label on a self-assembled peptide; (ii) progressive quenching of QD PL (inset) and sensitization of Os(II)PP complex PL as the number of labeled peptides per QD increases; (iii) wavelength dependent quenching of QD PL (blue and green curves) and the correlation with the Os(II)PP absorbance spectrum as evidence of FRET. Reprinted with permission from ref. [81]. Copyright 2012 American Chemical Society. (B) PL spectra showing quenching of CdTe QD PL and sensitization of PC PL due to FRET. The structure of the PC is shown in the inset, where M = Al(III). Reprinted with permission from ref. [100]. Copyright 2012 Elsevier. (C) FRET from a CdSe/ZnS QD to a cyclam complex of Cr(III), CrONO: (i) illustration of the conjugate, where the CrONO is adsorbed; (ii) mechanism of FRET-induced NO release from the complex (left) and comparison of NO production with and without QDs (right; measured with a NO-specific electrode). Reprinted with permission from refs. [107, 109]. Copyright 2007 and 2012 American Chemical Society.
Ac ce p
1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527
Figure 4. (A) Quenching efficiencies for QDs deposited on different carbon materials. GO is the most efficient quencher. Reprinted with permission from ref. [127]. Copyright 2012 Elsevier. (B) Energy transfer between QDs and Au NPs of different size: (i) double-stranded DNA linkage to control QD-Au NP separation; (ii) QD PL spectrum and absorption spectra of 3 nm, 15 nm, and 80 nm Au NPs; (iii) quenching efficiency versus donor-acceptor separation for the three sizes of
49 Page 49 of 63
Au NP and the expected behaviours for FRET and NSET mechanisms. Reprinted with permission from ref. [151]. Copyright 2011 American Chemical Society.
cr
ip t
Figure 5. Multi-step FRET relay along a DNA photonic wire. (A) Illustration of CdSe/ZnS QDs self-assembled with Cy3, Cy5, Cy5.5 and Cy7-labeled DNA. (B) Composite PL spectra for QDs self-assembled with DNA photonic wires with a progressive increase in the number of energy transfer steps. (C) Deconvolved PL spectra for the each individual chromophore when all are present along the DNA wire. (D) Composite PL spectra for DNA labeled with Cy3, Cy5, Cy5.5 and Cy7 dyes at the indicated positions and without a central QD scaffold. Reprinted with permission from [158]. Copyright 2010 American Chemical Society.
M
an
us
Figure 6. Time-gated concentric FRET relay and application to biosensing. (A) Temporal evolution of energy transfer from the QD to A647 dye (FRET2) and from an LTbC (Tb) to the QD (FRET1) in the time-gated FRET relay. (B) (i) QD bioconjugate with the A647 and LTbC assembled to the QD through peptide substrates for chymotrypsin (ChT) and trypsin (TRP), respectively; (ii) changes in prompt A647/QD PL ratio and time-gated QD/LTbC PL ratio with changes in the number of A647, n, and LTbC, m, per QD. (C) Multiplexed tracking of protease activity by (i) measuring the A647/QD and QD/LTbC PL ratios over time, and (ii) conversion to the amount of each hydrolyzed peptide product using the data in (B). The solid arrows indicate increasing concentrations of trypsin; the dashed arrows indicate increasing concentrations of chymotrypsin. Figure reprinted from ref. [159]. Copyright 2012 American Chemical Society.
te
d
Figure 7. Spectral concentric FRET relay and application to biosensing. (A) Illustration of the QD bioconjugate and FRET pathways where the A555 and A647 dyes are labels on peptides self-assembled to the QD. The peptides are selected to be substrates for the enzymes chymotrypsin (ChT) and trypsin (TRP), respectively. (B) Schematic showing changes in the concentric FRET relay during activation of pro-chymotrypsin to chymotrypsin by trypsin. (C) Multiplexed tracking of pro-protease activity: (i) changes in the A555/QD and A647/QD PL ratios over time; (ii) mapping of PL ratios to the number of A555 and A647 per QD; and (iii) conversion of PL data to progress curves for the hydrolysis of each peptide substrate. The arrows indicate increasing concentrations of trypsin; the pro-chymotrypsin concentration was constant. Reprinted from ref. [160]. Copyright 2012 American Chemical Society.
Ac ce p
1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
Figure 8. ―Rainbow‖ thin films and cascaded energy transfer for QDSSCs. (A) Graded multilayer thin film of CdTe QDs for exciton recycling: (i) absorption and emission spectra for the CdTe QDs; (ii) reference configuration (REF) and cascaded energy transfer (CET) configuration; (iii) enhancement of red QD PL and near complete quenching of green, yellow, and orange QD PL. Reprinted with permission from ref. [175]. Copyright 2004 American Chemical Society. (B) Illustration of (i) a solar cell and forward (FWD), reference (REF), and reverse (REV) configurations for a multilayer film of QDs; (ii) corresponding energy level diagrams; (iii) absorption and PL spectra; (iv) IPCE spectra showing superior performance for the forward configuration. Reprinted with permission from ref. [181]. Copyright 2011 WILEYVCH Verlag GmbH & Co. KGaA, Weinheim. (C) Illustration of (i) gradient structure of alloyed CdSeS QDs on TiO2 and its energy level diagram; (ii) IPCE spectra for various combinations of QD colours. Reprinted with permission from ref. [183]. Copyright 2013 American Chemical Society.
50 Page 50 of 63
ip t
Figure 9. Energy transfer from QDs embedded in amorphous TiO2 to squaraine dyes for QDSSC applications. (A) TEM image of a modified electrode (top left) and schematic of the electrode sensitization process: (i) light absorption and exciton generation in the QD; (ii) excitation of the dye via FRET; (iii) charge separation with injection into the nanocrystalline TiO2. (B) IPCE spectra for electrodes modified with QD, dye, and both QD and dye. Figure reprinted with permission from refs. [179, 186], Copyright 2010 and 2011 American Chemical Society, and ref. [187], Copyright 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
te
d
M
an
us
cr
Figure 10. Use of heterobifunctional squaraine dyes as linkers between TiO2 and CdSe QDs, showing (i) schematics and energy level diagrams; and (ii) IPCE spectra for TiO2 modified with CdSe QDs, squaraine dye, and both. Energy is transferred from the QDs to the dye, which injects an electron in the TiO2. (A) Carboxyl and alkyl chain modifications. Reprinted with permission from ref. [180]. Copyright 2012 Royal Society of Chemistry. (B) Carboxyl and thiol group modifications. Reprinted with permission from ref. [188]. Copyright 2012 American Chemical Society.
Ac ce p
1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
51 Page 51 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract
Page 52 of 63
*Highlights (for review)
Highlights
te
d
M
an
us
cr
ip t
Energy transfer between quantum dots and metal complexes Energy transfer between quantum dots and other nanomaterials Multi-step Förster resonance energy transfer relays and cascades Energy transfer in monolayer or multilayer thin films of quantum dots Applications in biosensing, solar energy conversion, optoelectronic devices
Ac ce p
Page 53 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 1
Page 54 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 2
Page 55 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3
Page 56 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4
Page 57 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 5
Page 58 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 6
Page 59 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 7
Page 60 of 63
Ac ce p
te
d
M
an
us
cr
ip t
Figure 8
Page 61 of 63
Ac ce p
te
d
M
an
us
cr
ip t
Figure 9
Page 62 of 63
Ac
ce
pt
ed
M
an
us
cr
i
Figure 10
Page 63 of 63