Upconversion Quantum Yield and Luminescence
of -NaYF4:Yb3+,Er3+ Nanoparticles:
Influence of Environment and Dopant Concentration
vorgelegt von
Diplom-Physiker
Martin Kaiser
von der Fakultät II – Mathematik und Naturwissenschaften
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Naturwissenschaften
– Dr. rer. nat. –
genehmigte Dissertation
Promotionsausschuss:
Vorsitzende: Prof. Dr. Ulrike Woggon
Gutachter: Prof. Dr. Axel Hoffmann
Gutachter: Prof. Dr. Oliver Benson
Gutachterin: Dr. Ute Resch-Genger
Tag der wissenschaftlichen Aussprache: 16.12.2020
Berlin 2021
ii
This thesis is dedicated to my daughter, Sophie Kaiser.
“Science is the process that takes us from confusion to understanding.”
Brian Greene
i
Abstract
Lanthanide-based upconversion (UC) nanometer (nm)-sized particles (UCNPs) exhibit
the unique ability to emit one higher-energy photon after the sequential absorption of two
or more near-infrared (NIR) photons. This UC luminescence (UCL) makes UCNPs very
attractive for the use as optical probes in the biomedical area. In this respect, the main
advantages of UCNPs are the suppression of environment-based background fluorescence
as well as their excitability with NIR-light allowing imaging applications in deep tissue layers
of several centimeters (cm).
The future progress in the development of brightly-luminescent UCNPs requires reliable
quantification methods of their optical properties to ensure comparability on an
international level. For this purpose, in the first study of this thesis an integrating sphere
setup (ISS), as well as corresponding relevant guidelines for the measurement conditions,
has been developed enabling the absolute determination of the key parameter upconversion
quantum yield
UC with minimum uncertainty (
UC := ratio of emitted high-energy photons
to absorbed low-energy photons). The main features of the developed ISS setup present the
wide tunability of P of over four orders of magnitude and a high linear detection range of
over 10 orders of magnitude allowing the comprehensive characterization of the nonlinear
luminescence behavior of UC materials. As the main subject of the investigations of this
thesis served a UC system based on a transparent host lattice -NaYF4, which was doped with
trivalent ytterbium ions (Yb3+), acting as absorber/antenna, and trivalent erbium ions (Er3+),
acting as emitter. The suppression of radiationless deactivation processes via the particle
surface presents a major challenge in the development of brightly-luminescent UCNPs. In
order to quantify the luminescence quenching of UCNPs, commercial µm-sized
-NaYF4:Yb3+(21%),Er3+(2%) particles (UCµP), with particle diameters large enough to neglect
surface effects, were comprehensively optically characterized. In this respect, a maximal
absolute
UC of 10.5% at P = 30 Wcm-2, considering the spectral region from 360 - 900 nm,
was measured for these UCµP.
New detailed insights of the UC processes of -NaYF4:Yb3+,Er3+ UCNPs were gained in
the second and third study by the investigation of high-quality UCNPs sample series with
systematical varied parameters. These UCNPs sample series were optically characterized
regarding their P-dependent
UC and UCL of the different Er3+ emission bands as well as of
the decay behavior of the intensity of the Er3+ and Yb3+ emission bands. The computer-
assisted simulation of these experimental data by utilizing a coupled rate equation system
turned out to be a powerful tool to underpin the photophysical interpretations. A special
ii
highlight of this advanced analysis represents the clarification of the conditions for the
controversially debated bi- and triphotonic population pathways of the red-emitting Er3+ 4F9/2
energy level. In the following, the main results of these studies are briefly presented.
The second study deals with 23 nm-sized UCNPs with different surrounding
environments, its interaction with the particle surface, and the accompanied influence on
the UC processes. A main conclusion of this study points out that for UCNPs dispersed in
water (H2O), at low P, the Er3+ 4F9/2 energy level is populated biphotonically by a nonresonant
Yb3+-Er3+ energy transfer (ET) from the low energetic Er3+ 4I13/2 energy level. Thereby, the high
population density of the low Er3+ energy levels is favored due to the coupling of excited Er3+
ions near the UCNPs particle surface with the O-H vibrational modes of the H2O molecules.
However, the use of very high P of ca. 1 kWcm-2 compensates H2O-induced increase of
nonradiative relaxation rates, and consequently, leads to moderately high
UC of about 0.5%
with a mainly triphotonic activation of the Er3+ 4F9/2 energy level for the UCNPs dispersed in
H2O.
In the third study, the optical properties of 33 nm-sized UCNPs dispersed in toluene as
function of the ion-ion distances were investigated. These distances can be directly tuned by
variation of the Yb3+ and the Er3+ dopant concentrations. In this respect, it was demonstrated
that the increase of the Yb3+ dopant concentration causes an enhancement of the triphotonic
activation of the red-emitting Er3+ 4F9/2 energy level. This effect is induced by an increased
back energy transfer from Er3+ to Yb3+ due to the reduction of the Er3+-Yb3+-distance.
Although the
UC is reduced by a faster energy migration to the UCNP surface, the overall
UCL intensity was overcompensated due to the higher number of absorbing Yb3+ ions.
However, the increase of the Er3+ concentration resulted in different trends of the relative
Er3+ red emission intensity. The intensity increase, observed at low P, was attributed to
increased biphotonic activation by enhanced nonradiative relaxation of the green-emitting
Er3+ 2H11/2/4S3/2. Contrary to that, at high P, the Er3+ red emission intensity diminishes when
the Er3+ concentration increases. Consequently, a rate equation analysis revealed that this
UCL behavior is a sign of a yet unknown depopulation rate.
iii
Zusammenfassung
Lanthanid-basierte upconversion (UC) (oder aufkonvertierende) Nanopartikel (UCNPs)
besitzen die spezielle Eigenschaft nach der sequentiellen Absorption von zwei oder mehreren
nahinfraroten (NIR) Photonen ein höherenergetisches Photon zu emittieren. Diese UC
Lumineszenz (UCL) macht UCNPs sehr attraktiv für den Einsatz als optische Nanosensoren im
biomedizinischen Bereich. Hierbei sind die Hauptvorteile von UCNPs die Vermeidung von
umgebungsbedingter Hintergrundfloureszenz sowie deren Anregbarkeit mit NIR-Licht, die
bildgebende Verfahren in tieferen Gewebeschichten von mehreren Zentimetern (cm)
erlaubt.
Der weitere Fortschritt in der Entwicklung von hell-lumineszierenden UCNPs erfordert
die verlässliche Quantifizierung ihrer optischen Eigenschaften um die Vergleichbarkeit auf
internationaler Ebene zu gewährleisten. Zu diesem Zweck wurde in der ersten Studie dieser
Doktorarbeit ein Ulbrichtkugelaufbau (ISS) sowie relevante Richtlinien für die
Messbedingungen entwickelt, die die absolute Bestimmung des Schlüsselparameters
upconversion Quantenausbeute (
UC) mit minimaler Unsicherheit ermöglichen
(
UC := Verhältnis von emittierten hochenergetischen Photonen zu absorbierten
niederenergetischen Photonen). Die Hauptbesonderheiten dieses ISS Aufbaus sind die weite
Durchstimmbarkeit der P über vier Größenordnungen und der große lineare
Detektionsbereich von über 10 Größenordnungen, die eine umfassende Charakterisierung
des nichtlinearen Lumineszenzverhaltens von UC Materialien gewährleistet. Als
Hauptgegenstand dieser Arbeit diente ein UC System basierend auf einen transparenten
Wirtskristall -NaYF4, der mit trivalenten Ytterbium-Ionen (Yb3+), fungieren als
Antenne/Absorber, und trivalenten Erbium-Ionen (Er3+), fungieren als Emitter, dotiert ist. Die
Vermeidung strahlungsloser Deaktivierungsprozesse über die Partikeloberfläche gilt als einer
der größten Herausforderungen in der Entwicklung von hell-leuchtenden UCNPs. Zur
Quantifizierung solcher Lumineszenzlöscheffekte von UCNPs wurden kommerzielle µm-
große -NaYF4:Yb3+(21%),Er3+(2%) Partikel (UCµP), deren Partikeldurchmesser hinreichend
groß ist um Oberflächeneffekte zu vernachlässigen, umfassend optisch charakterisiert. In
dieser Hinsicht, wurde für diese UCµP eine maximale
UC von 10.5% bei P = 30 Wcm-2 im
spektralen Bereich von 360 nm - 900 nm absolut gemessen.
Neue detaillierte Erkenntnisse über die UC Prozesse von −NaYF4:Yb3+,Er3+ UCNPs
wurden in der zweiten und dritten Studie anhand von hochqualitativen Probenserien mit
systematischen variierten Parametern erlangt. Diese UCNPs Probenserien wurden
iv
hinsichtlich der P-abhängigen
UC und UCL der verschiedenen Er3+ Emissionsbanden sowie
des Abklingverhaltens der Intensität von den Er3+ und Yb3+ Emissionsbanden untersucht. Die
computerunterstützte Simulation der gewonnenen experimentellen Daten mittels eines
gekoppelten Ratengleichungssystems erwies sich als ein mächtiges Werkzeug für die
Untermauerung der photophysikalischen Interpretationen. Ein besonderes Highlight dieser
fortschrittlichen Analyse präsentiert die Klärung der Bedingungen für die kontrovers
diskutierten bi- und triphotonischen Populationswege für das rot emittierende Er3+ 4F9/2
Energielevel. Im Folgenden sind die Hauptergebnisse dieser Studien kurz dargestellt.
Die zweite Studie beschäftigt sich mit 23 nm-großen UCNPs in verschiedenen
Partikelumgebungen, sowie dessen Interaktion mit der Partikeloberfläche und den hiermit
verbundenen Einfluss auf die UC Prozesse. Eine Hauptergebnis dieser Studie zeigt auf, dass
für UCNPs dispergiert in Wasser (H2O) das rot emittierenden Er3+ 4F9/2 Energielevel bei
niedriger P hauptsächlich biphotonisch über einen nichtresonanten Yb3+ - Er3+
Energietransfer (ET) vom niederenergetischen Er3+ 4I13/2 Energielevel bevölkert wird. Hierbei
wird die hohe Population der niederenergetischen Er3+ Energielevel durch die Kopplung von
oberflächennahen angeregten Er3+-Ionen mit den O-H Vibrationsmoden der H2O Moleküle
begünstigt. Indessen kompensierte die Verwendung von hohen P von ca. 1 KWcm-2 die H2O-
induzierten Löschraten, wodurch für die UCNPs dispergiert in H2O eine mäßig hohe
UC von
0.5% sowie hauptsächlich triphotonischen Aktivierung des Er3+ 4F9/2 Energielevels erreicht
wird.
In der dritten Studie wurden die optischen Eigenschaften von 33 nm-großen UCNPs
dispergiert in Toluol in Abhängigkeit der mittleren Ionenabstände untersucht. Diese
Abstände wurden über die Variation der Yb3+ und der Er3+ Dotierkonzentration
durchgestimmt. In dieser Hinsicht wurde demonstriert, dass die Erhöhung der Yb3+
Konzentrationen zu einer Verstärkung der triphotonischen Aktivierung des rot emittierenden
Er3+ 4F9/2 Energielevel führt. Dieser Effekt ist auf einen verstärkten Energierücktransfer vom
Er3+ zum Yb3+, aufgrund der Verringerung des Er3+-Yb3+ Abstandes, zurückzuführen. Indessen
führte die Erhöhung der Er3+ Konzentration zu verschiedenen Trends für die relative
Intensität der roten Er3+ Emissionsbande. Die Intensitätserhöhung bei Anregung mit geringer
P konnte auf die Verstärkung der nichtstrahlenden Rate des grün emittierenden Er3+
2H11/2/4S3/2 Energielevels zurückgeführt werden. Konträr dazu verringerte sich bei Anregung
mit hoher P die relative rote Er3+ Emissionsintensität mit steigender Er3+ Konzentration. Die
Ratengleichungsanalyse zeigt auf, dass dieses UCL-Verhalten ein Indiz für eine bisher
unbekannte Depopulationsrate ist.
v
List of Publications for this Cumulative Thesis
This cumulative thesis is based on the following publications, which are referred by capital
roman numerals:
I. M. Kaiser, C. Würth, M. Kraft, I. Hyppänen, T. Soukka, U. Resch-Genger
„Power-dependent upconversion quantum yield of NaYF4: Yb3+, Er3+ nano-and
micrometer-sized particles – measurements and simulations“
Nanoscale, 2017, 9, 10051-10058
DOI: 10.1039/C7NR02449E
II. C. Würth*, M. Kaiser*, S. Wilhelm, B. Grauel, T. Hirsch, U. Resch-Genger
„Excitation power dependent population pathways and absolute quantum yields
of upconversion nanoparticles in different solvents“
Nanoscale, 2017, 9 , 4283-4294
DOI: 10.1039/C7NR00092H
*equally contributed
III. M. Kaiser, C. Würth, M. Kraft, T. Soukka, U. Resch-Genger
“Explaining the influence of dopant concentration and excitation power density
on the luminescence and brightness of -NaYF4:Yb3+,Er3+ nanoparticles:
Measurements and simulations”
Nanoresearch, 2019, 12, 1871-1879
DOI: 10.1007/s12274-019-2450-4
Content of these publications have been reproduced with the permission of the
copyright holder
vi
Contents
Abstract .......................................................................................................................... i
Zusammenfassung .......................................................................................................... iii
List of Publications for this Cumulative Thesis .................................................................. v
Contents ......................................................................................................................... vi
1 Introduction ................................................................................................................. 1
1.1 Aims and outline of this cumulative thesis ............................................................... 4
2 Fundamentals .............................................................................................................. 7
2.1 Upconversion (UC) processes .................................................................................... 7
2.2 Energy transfer (ET) processes between Ln3+ ions .................................................... 9
2.3 Energy transfer upconversion (ETU)-based materials ............................................... 12
2.3.1 Luminescent centers – sensitizer and activator ions ...................................... 12
2.3.2 Host lattice ..................................................................................................... 13
2.4 Upconversion luminescence (UCL) of -NaYF4:Yb3+,Er3+ crystals............................... 15
2.5 Upconversion luminescence quantum yield (
UC) .................................................... 18
3 Reviews of Literature and Open Questions ................................................................... 21
3.1 Absolute
UC measurements .................................................................................... 21
3.2 Solvent-dependent luminescence quenching ........................................................... 22
3.3 Dopant concentration-dependent color tuning ....................................................... 23
4. Samples and Experimental Methods ........................................................................... 25
4.1 Overview of investigated -NaYF4:Yb3+,Er3+ samples ................................................ 25
4.2 Optical characterization methods ............................................................................ 27
4.2.1 Integrating sphere setup (ISS) ......................................................................... 28
4.2.2 Time-resolved photoluminescence spectroscopy ........................................... 36
vii
5 Measurement Strategies for the P-dependent
UC (
UC(P))........................................... 39
5.1 Measurement conditions for
UC(P) ......................................................................... 39
5.2 Validation of measured
UC(P) by comparison with literature data ......................... 46
5.2.1 Validation by comparison with measured results from Page et al. ................. 47
5.2.1 Validation by comparison with measured results from the van Veggel group 48
5.2.3 Validation by comparison with simulated results from the Berry group ......... 48
5.3 Conclusions of chapter 5 ........................................................................................... 50
6 Solvent-Dependent Optical Properties of -NaYF4:Yb3+,Er3+ UCNPs .............................. 51
6.1 Solvent-dependent upconversion quantum yield (
UC) ......................................... 51
6.2 Emission color of UCNPs dispersed in D2O and H2O............................................... 52
6.3 Conclusions of chapter 6 ..................................................................................... 56
7 Dopant Concentration-Dependent Optical Properties of -NaYF4:Yb3+,Er3+ UCNPs ....... 57
7.1 Dopant concentration-dependent upconversion luminescence (UCL) .................... 57
7.2 Influence of the dopant concentrations on rate equation constants ....................... 60
7.3 Conclusions of chapter 7 .......................................................................................... 63
8 General Conclusions for Publications ........................................................................... 64
Appendix ........................................................................................................................ I
List of Abbreviations ....................................................................................................... I
Bibliography ................................................................................................................... III
List of Publications .......................................................................................................... XII
Conference Talks and Posters ........................................................................................ XVI
Acknowledgement .......................................................................................................... XVIII
Contributions to the Manuscripts .................................................................................. XX
viii
1
1 Introduction
Optical probes are increasingly used in biomedical research to visualize, characterize and
quantify processes on cellular und subcellular levels. Such probes enable localized
measurements of pH, polarity, viscosity, and the detection of disease-related biomarkers.1-4
The most extensively investigated and applied chromophores for the design of optical probes
are fluorescent organic dyes and colloidal semiconductor nanocrystals called quantum dots
(QDs).5-7 Organic dyes and QDs stand out for their excellent brightness resulting from a) a
high photon absorption cross-section [cm²], which is a measure of the probability of an
incoming photon being absorbed, and b) a high photoluminescence quantum yield (
),
defined as ratio of emitted photons to absorbed photons.5 Organic dyes exhibit
unsymmetrical-shaped emission bands and a mirrored-shifted absorption band, see
Figure 1.1. The emission process stems from optical transitions, which either are localized
over the whole molecule or from an intramolecular charge transfer, i.e. that electron is
transferred from one part to another part of the dye molecule.5, 8 QDs have a Gaussian-
shaped emission band resulting from the recombination of electron-hole pairs called
excitons, which are produced by the absorption of photons with energies larger than the QDs
bandgap energy. The absorptivity of QDs increases with decreasing wavelength offering a
variable choice of the excitation wavelength, in contrast to organic dyes, see Figure 1.1.5 The
emission and absorption properties of QDs are mainly characterized by their material
composition, the particle size, and size distribution. Thereby, the decrease of the QDs
emission wavelength with shrinking diameter, known as quantum size effect, arises when
the QDs diameter (typically a few nanometers) is comparable or smaller than the material-
specific exciton Bohr radius equaling the distance of its electron-hole pairs.9, 10
Figure 1.1 Overview of a typical structure, emission band, and absorption band of
a) Organic dyes, b) QDs, and c) UCNPs. Reproduced from Würth et al.11 with
permission from Springer Nature.
2
Despite the great success of these emitters in biomedical applications, their potential is still
hampered due to the lack of stable, efficient near-infrared (NIR) emissive organic dyes and
QDs.12 Therefore, the common impairments of visible (vis) light emitting optical probes,
including scattering, absorption, and background fluorescence of surrounding biological
matrices, are restricting their scope of applications, e.g. the use in deep tissue imaging. A
further critical issue of organic dyes concern their low photostability, i.e. that they bleach
after long-term irradiation, which is very problematic for single particle tracking or long-term
studies in a medium or in living cells. Although QDs exhibit a high photostability, their blinking
behavior, i.e. that they can spontaneously switch from being emissive to a dark state, is
particularly undesirable for single particle tracking applications. Furthermore, the
assessment of the toxicity of these materials for in vitro and in vivo studies is an ongoing
controversial topic. In this respect, dyes have been widely characterized showing cytotoxicity
levels varying from low to very high.5 QDs, typically consisting of toxic heavy metals (e.g. Cd2+,
Hg2+, Zn2+, Pb2+), require a very stable surface chemistry as surface cover.7, 13, 14 However,
their nanotoxicological effects, i.e. that they may accumulate inside cells and damage them
as well as a potentially high surface reactivity, are still not properly understood.
Trivalent lanthanide-(Ln3+)-based upconversion (UC) nanoparticles (UCNPs) are an
emerging class of optical probes with a high potential of outperforming fluorescent dyes and
QDs in many application areas.15-31 UCNPs already demonstrated promising results as
contrast agent for deep tissue super-resolution imaging and spectroscopy,32, 33 as an optical
switch to control mouse brain activity34 in the optogenetics research area, and as active
media for micro-cavity lasers 35 with ultra-low threshold down to 30 Wcm-2. The most
efficient UCNPs materials are based on a combination of energy harvesting Ln3+ ions called
sensitizer ions (e.g. Yb3+, Nd3+) and emitting Ln3+ ions called activator ions (e.g.
Er3+,Tm3+,Ho3+) embedded in an inorganic transparent host material. 17, 21, 30 The parity
forbidden optical 4f-4f transitions of the Ln3+ exhibit long luminescence lifetimes > 100 µs,
and hence, UCNPs require rather low excitation power densities (P) in the 1 Wcm-2 - 103 Wcm-
2. Such low P can be easily provided by a low-cost continuous-wave laser diode. Moreover,
the UC luminescence (UCL) of UCNPs promise a very high signal-to-noise ratio due to the
circumvention of the detection of background fluorescence from surrounding biological
matrices, a large wavelength shift between emission and excitation (ranging from 130 nm to
500 nm) called anti-Stokes shift and several sharp tunable emission bands with line width <
20 nm (see Figure 1.1).17, 25 Additionally, their long luminescence lifetimes allow canceling
out sources of interference by temporal discrimination called time-gated emission
spectroscopy.17, 21, 30 The NIR-light excitation of UCNPs offers a deep penetration depth in
3
tissue connected with low photodamage on cells compared to the vis-light excitation used
for dyes and QDs. However, UCNPs exhibit rather low photon absorption cross-section
compared to organic dye and QDs, which is a major challenge to overcome (strategies are
addressed in the next paragraph). Further advantages of UCNPs compared to organic dyes
and QDs presents the high photostability and the absence of blinking, which are important
features for single particle tracking applications.36 Moreover, results of toxic studies of
UCNPs indicate a rather low nanotoxicology.17, 37, 38 However, these risks have to be more
intensely investigated to allow their use inside the human body.
Nowadays, the state-of-the-art synthetic protocols allow producing high-quality UCNPs
with defined crystal morphology, narrow size distributions, controlled dopant concentrations
and surface chemistries as well as different particle architecture. However, due to high
surface-to-volume ratio of UCNPs their maximal
UC is still one to two orders of magnitude
smaller than for bulk. A very promising strategy to avoid surface-related luminescence
quenching processes consists in the passivation of the surface with an undoped optically
inactive shell. Recently, a report showed that 23 nm-core particles -
NaYF4:Yb3+(18%),Er3+(2%) covered with a 22 nm-thick undoped -NaYF4 shell reaching a
similar
UC compared to µm-sized UC particles.39, 40 Nonetheless, the
UC of UCNPs with
diameters of less than 10 nm, desired for biomedical applications, was yet not fully restored
with this core-shell approach.41 Additionally, the strong absorption of water molecules (H2O)
at 980 nm, resonant to the Yb3+ absorption, and thus, drastically enhancing luminescence
quenching via the UCNPs surface,42 is limiting the use of Yb3+-based UCNPs in aqueous media.
In this respect, co-doping UCNPs with Nd3+, excitable at ca. 800 nm, offers an alternative way
to avoid absorption of H2O, and consequently, reduced heating and luminescence quenching
effects.43, 44 However, Nd3+-based UCNPs require more complicated synthetic routes and
particle design as well as exhibiting a much narrower absorption band compared to
conventional Yb3+-based UCNPs. Moreover, as mentioned previously, UCNPs have a rather
weak photon absorption cross-section of Ln3+ ions, typically 5 - 8 orders of magnitude lower
than for organic dyes and QDs. Most promising strategies to enhance the absorption of
UCNPs comprise a) enhancing local electromagnetic field via plasmonics effects based on
metal nanostructures or photonic crystals and b) sensitization of UCNPs with organic dyes or
QDs. UC luminescence enhancement in the order of three orders of magnitude have already
been reported with these strategies.45-49
The further rational design of brightly-luminescent UCNPs with industry-relevant
structural properties and functionalities requires understanding of (de-)population processes
and deactivation channels. This can be assessed with upconversion quantum yield (
UC)
4
measurements, particle absorption and decay behavior combined with theoretical
investigations. In particularly, the measurement of
UC, equaling the ratio of high energy
photons emitted to low energy photons absorbed, is known to be very challenging. Due to
its strong dependency on the P, absolute
UC values can be exclusively determined reliably
with a customized integrating sphere setup (ISS). This initiated an increasing number of
reports on absolute
UC measurements utilizing an ISS.50-55 Nonetheless, the comparison of
the
UC obtained of different laboratories remains a challenging task since most reports are
missing a comprehensive description of the instrument design, calibration and
characterization as well as measurement procedure. Furthermore, the evaluation of
reported
UC values complicates due to differences in the excitation wavelength, missing a
characterization of the excitation beam profile and the use of an insufficiently low P-range
of less than two orders of magnitude. Thus, there is an urgent need for the development of
standardized absolute
UC measurements with an ISS.
1.1 Aims and outline of this cumulative thesis
The main aims of this cumulative thesis:
(i) Designing, building-up and characterizing an ISS for absolute UC quantum yield
measurements for UC materials excitable at 980 nm.
(ii) Identifying UC population and depopulation processes for -NaYF4:Yb3+,Er3+ UCNPs
dispersed in H2O.
(iii) Understanding the influence of Yb3+ and Er3+ dopant concentration on emission
color of -NaYF4:Yb3+,Er3+ UCNPs.
In contrast to the three publications this cumulative thesis is based on, a more detailed
description of the custom-built ISS is provided (see Chapter 4). Furthermore, for each
publication an abbreviated version has been written (see Chapter 5,6 and 7). These parts
may serve as introduction for a broader readership and describe some aspects in more detail.
In the following, a brief outline of this thesis is provided:
5
• Chapter 2 provides fundamentals of the UC processes. In particular, the UC
processes of the investigated -NaYF4:Yb3+,Er3+ UC system are described in detail.
• Chapter 3 presents an overview of the state of the art for each publication and
introduces questions addressed by this thesis.
• Chapter 4 describes the new custom-built ISS and its characterization as well as
the analysis of the obtained data. An overview of the investigated UC samples is
also provided.
• Chapter 5 provides prerequisites for the measurement procedure and geometry
and for the optical properties of the UC material as dispersion or powder.
Furthermore, the results of
UC measurement for a µm-sized UC material were
validated by comparison with literature data.
• Chapter 6 presents studies of UCNPs in different solvents with a focus on the
optical properties of UCNPs in H2O and D2O. A model for the (de-)population
pathways of UCNPs in H2O was developed.
• Chapter 7 deals with the influence of the Yb3+ and Er3+ dopant concentrations on
the Er3+ green and red emission intensities of UCNPs. The interpretations were
supported with a comprehensive rate equation analysis.
6
7
2 Fundamentals
This chapter introduces the physical basics for the understanding of different UC
processes. Thereafter, conditions for the composition of an efficient Ln3+-based UC material
including luminescent centers and host lattices are discussed. Subsequently, a focused
overview of the properties of -NaYF4:Yb3+,Er3+ is given. This material is known to be the
most efficient NIR to green light converter based on energy transfer upconversion (ETU),
and hence, was comprehensively studied in this thesis. At the end of this chapter, a brief
theoretical overview of the key parameter
UC is provided.
2.1 Upconversion (UC) processes
PL is a process in which a material absorbs incident photons, exciting electrons to a
higher electronic excited state, and then emits photons after the relaxation of the electrons
to a lower energy state.56 In case of a linear absorption-emission PL process, the energy of
the emitted photons do not exceed the energy of the absorbed ones (Stokes shifted
emission). However, a UC process permits the combination of multiple absorbed photons
resulting in the emission of higher energy photon (anti-Stokes shifted emission). The era of
the UC field started in 1959, when Bloembergen proposed the idea of an infrared(IR)-
detector based on the sequential absorption of two NIR photons within energy levels of an
ion in a solid, called excited state absorption (ESA).57 Unfortunately, the lack of intense
coherent laser sources at this time made it impossible to observe any effects.30 In 1966, Auzel
extended the approach of Bloembergen by utilizing energy transfer between different ions
in a solid matrix, called ETU, which could even be detected with incoherent light sources.30,
58 Nowadays, ETU-based upconverters still belong to the most efficient UC materials.
Figure 2.1 shows a collection of different UC processes for the special case that exactly
two absorbed photons are summed-up resulting into the emission of a higher energy
photon.30 ESA is the simplest UC mechanism and involves a sequential absorption of two
photons via a real intermediate electronic energy state of a Ln3+-ion, see Figure 2.1 a).
However, this process is not very efficient, because of the rather low absorption coefficients
of suitable Ln3+ ions. Compared to ESA, ETU is about one hundred times more efficient.30 ETU
based upconverters are typically materials doped with two different Ln3+ called sensitizer and
activator. The sensitizer ions harvest the energy, i.e. they effectively absorb incoming
photons, and transfer it in sequential steps via nonradiative ET to the activator ion, see Figure
2.1 b). The efficiency of ETU is commonly boosted by high sensitizer concentrations allowing
8
fast energy migration between sensitizer ions inside the crystal lattice.59 In some special
cases, chemically identical ions can act simultaneously as activator and sensitizer ion, e.g. for
singly Er3+-doped 1520 nm excitable upconverters used for solar cell applications.60 The
efficiency of ESA and ETU processes both rely on metastable energy levels required for the
sequential absorption of photons to avoid early depopulation. This is satisfyingly fulfilled by
most Ln3+ ions, due to their parity-forbidden f-f transitions and manifold of energy levels (see
2.3.1 Luminescent centers).
Figure 2.1 c) - e) present UC processes based on partly missing or even without any
energy level called cooperative energy transfer (CET), two photon absorption (TPA) and
second harmonic generation (SHG). CET is similar to ETU, but misses a real intermediate
energy level for the activator ion. At the CET process, two sensitizer ions are simultaneously
transferring the energy via a virtual energy state to the emitting activator ion. Studies on
CET-based UC materials primarily focused on bulk and glasses due to their low UC
efficiencies.61-63 TPA and SHG typically occur in non-lanthanide materials and have to be
pumped with rather expensive lasers sources with pulse lengths in the pico- to femtosecond
time scale as they require very high P. TPA, which is based on the simultaneous absorption
of two photons without a real intermediate energy state, of fluorescent dyes molecules or
QDs is frequently used in high-resolution microscopy, bioassays, and even skin cancer
detection.64-66 SHG is a UC process even without occupying any real energy level utilizing the
nonlinear susceptibility of a medium. SHG is amongst other applications a popular choice for
the frequency doubling of lasers with maximal energy conversion efficiency of up to 70% for
optimal conditions.67
Figure 2.1 Schematics of different UC mechanism generating a high energy photon by combining
two low energy photons: a) ESA; b) ETU); c) CET; d) TPA; e) SHG.
9
At last, a brief description of the triplet-triplet annihilation (TTA) UC process is provided.
TTA-based systems have sizes of only few nanometers due to its molecular structure, and
thus, are very promising candidates for biomedical applications.68 At the TTA process the
energy of two triplet-excited dye molecules (annihilators) is combined by ET from one
molecule to the other molecule, followed by the emission of a high energy photon. To
enhance the absorptivity TTA molecules are typically bound to strong absorbing dye
molecules (sensitizers). Thereby, the sensitizers effectively transfer the absorbed energy to
the annihilators after an intersystem crossing, i.e. a sensitizer changes from an excited singlet
state to a lower energetic excited triplet state. The
UC values of such TTA-systems found to
be in the order of one percentage at rather low P of several hundred mWcm-2.68, 69 The
limiting factor of TTA-systems is the high sensitivity to oxygen of the involved triplet states
reducing the
UC.70-72
2.2 Energy transfer processes between Ln3+ ions
In this subsection, a theoretical overview of energy transfer (ET) processes between Ln3+
is provided. Energy transfer is present when absorption and emission of a photon do not
take place at the same luminescent center.30 ET processes can be divided into radiative and
nonradiative ET processes. Radiative ET usually plays a minor role for UC processes due to
the small absorption coefficients of Ln3+ ions. In contrast, nonradiative ET is a key feature for
ETU leading to effective migration and diffusion processes in the UC crystal lattice. Therefore
in this thesis, the term ET implies nonradiative ET. Figure 2.2 depicts radiative and
nonradiative ET processes for the simple case of only two ions interacting with each other.
These two ions are named sensitizer and activator, where the sensitizer transfers the
absorbed energy to the activator. Although donor and acceptor would be clearer terms,
sensitizer and activator are intentionally used by the UC community to avoid
misunderstanding with the nomenclature used for semiconductors.
Radiative ET takes place via an exchange of a real photon. The probability WSA to
transfer energy from the sensitizer to the activator for a radiative ET between two ions can
be calculated in dependence of the distance R with Equation 2.1.73
𝑊SA(𝑅) = 1
𝜏s 𝜎A
4𝜋𝑅2∫𝑔S(𝜈)𝑔A(𝜈)d𝜈 Equation 2.1
10
Here,
s is the PL lifetime of the sensitizer,
A the integrated absorption cross section of
the activator, and the integral represents the spectral overlap between the sensitizer
emission gs() and activator absorption gA(). Radiative ET does not affect the PL lifetime of
the donor, so that it can be experimentally distinguished from nonradiative ET processes.
For more complex systems the probability of a radiative ET WSA(R) depends not solely on R,
but also on the shape and scattering properties of the sample.
Nonradiative ET becomes relevant for short distances in the nanometer range requiring
high dopant concentrations of Ln3+. In this case the excitation energy is transferred from the
sensitizer ion to the activator without the exchange of a real photon, see Figure 2.2 b). Such
nonradiative ET via dipole-dipole interaction was firstly described by Förster in 1946.74 Later
this theory was extended by Dexter for electric multipolar interactions leading to Equation
2.2 providing the probability for this nonradiative ET process.75
𝑊SA(𝑅) = 1
𝜏s (𝑅0/𝑅)𝑘 Equation 2.2
Here, R0 is called the Förster radius, the distance of the two ions at which the probability
of an ET equals 50%, and k is a positive integer with k = 6 for dipole-dipole (Förster case), k
= 8 for dipole-quadrupole and k = 10 for quadrupole-quadrupole interactions. The Förster
radius depends on the overlap of the sensitizer emission and activator absorption, the
integrated absorption cross-section of the activator ion, and the photon energy. However,
for very short distances less than 1 nm the orbital overlap or Dexter exchange may become
the dominant nonradiative ET process. At this process the excited electrons are “hopping”
through the network. The probability WET for this process has an exponential relationship to
R.76
𝑊ET(𝑅)= 𝐾2e−2𝑅/𝐿 Equation 2.3
Here, K is a scaling parameter with a dimension of energy and the parameter L is named
“effective average Bohr radius”, which accounts for the distance decay of the interaction.
11
A phonon-assisted nonradiative ET can take place between different ions with an
energy mismatch between the energy levels of the sensitizer and activator, see Figure 2.2 c).
This energy mismatch can be overpassed via production or annihilation of phonons. Small
energy mismatches < 100 cm-1 can be bridged by thermal phonons, and larger energy
mismatches, even as high as several thousand cm-1 can be bridged by optical phonons of the
host lattice. In accordance to the Miyakawa-Dexter theory the probability for a phonon
assisted ET WPAT(E) can be described with Equation 2.4.77, 78
𝑊PAT(Δ𝐸) = 𝑊PAT(0)e−𝛽Δ𝐸 Equation 2.4
Here E is the energy difference between the sensitizer and the activator level, WPAT(0)
is the energy transfer probability for E = 0,
is a parameter characterized by the strength
of electron-lattice coupling and the nature of the phonon involved.
Cross-relaxation (CR) is a special case of nonradiative ET and refers to all types of ETs
between identical ions, where one ion transfers a portion of energy to another, see
Figure 2.2 d). CR becomes usually a dominant process for high dopant concentration of the
activator ions leading the undesired depopulation of the higher energy levels. This is also
called concentration quenching.
Figure 2.2 Schematics of radiative and nonradiative ET processes between two ions:
a) Resonant radiative ET; b) Resonant nonradiative ET; c) Phonon-assisted nonradiative
ET; d) Cross-relaxation. Based on a figure from Auzel.30
12
2.3 Energy transfer upconversion (ETU)-based materials
UC materials, based on ETU, consist of three components: i) host lattice, ii) sensitizer
ions, and iii) activator ions. The design of efficient UC materials requires careful selection of
these components.
2.3.1 Luminescent centers – sensitizer and activator ions
Ln3+ ions have excellent properties for the design of efficient UC materials due to their
multiple long-lived optical f-f transitions (with the exception of Yb3+ and Ce3+ with only one
optical transition). Figure 2.3 shows the energy level scheme for free Ln3+ ions with their 4f-
4f transitions. For free Ln3+ ions the optical transitions between the 4f-4f energy levels are
parity forbidden, i.e. that the probability for the emission of a photon is zero.79. This rule is
broken for Ln3+ ions embedded in a host lattice. Due to disturbances of the wavefunctions by
the strength and symmetry of the crystal field of the host lattice, these parity forbidden
transitions become weakly allowed. Because these 4f-4f optical transitions are shielded by
the outer 5s25p6 subshells,80 these effects are small. This implies that the optical transition
of the Ln3+ ions are characterized by long PL lifetimes in the microsecond to millisecond range
and narrow linewidths (full width at half maximum (FWHM) typically between 10-20 nm) in
the solid state.
In the case of ETU-based UC materials, the standard ingredients are Yb3+ as sensitizer
and Er3+, Tm3+ or Ho3+ as activator ions. The reason for Yb3+ being the best choice as sensitizer
is based on three facts: i) Only one excited state (2F5/2), ii) the high absorption cross section
compared to other lanthanide ions (1.2·10-20 cm² 81), and iii) the 2F5/2 energy level of Yb3+
(980 nm) matches energetically well with energy levels of the typical activator ions. For
example the Er3+ 980 nm 4I11/2 energy level is resonant to this Yb3+ energy level allowing
efficient Yb3+-Er3+ ET. In order to increase the absorption and energy migration usually high
dopant concentrations of Yb3+ in the range of 20% up to 98%.82 For activator ions a ladder
like energy level structure with equally-spaced intermediate energy levels is required. This is
fulfilled by Er3+, Tm3+ and Ho3+, see Figure 2.3. The dopant concentration of the activator ion
is commonly in the range of 1% to avoid concentration quenching induced by cross
relaxation. However, recent studies indicate that for very high P in the region of 1 MWcm-2
much higher activator concentrations of up to 20% for Er3+ are advantageous.83
The energy levels of Ln3+ are denoted in the Russell-Saunders notation using 4fN states
(N = 1 to 13) of 2S+1LJ multiplets, with S, L and J representing the total spin, orbital, and
angular momenta of the N 4f electrons.84 The splitting of the energy levels by electrostatic
13
interaction 2S+1L is in the order of 104 cm-1.85 In addition, the spin-orbit interaction leads to
further splitting of the levels into 2S+1LJ with a separation of the J states by 103 cm-1.
Moreover, the crystal field of the lattice leads to an additional separation of the energy states
in the order of 102 cm-1.85
Figure 2.3 Energy level diagram for the 4f levels of Ln3+ ions. Positions of energy levels are
calculated by using free ion parameters described by Walrand et al.86 Reprinted from
Mahata et al.80 with permission from Intech.
2.3.2 Host lattice
The choice of the host lattice is essential for an efficient UC emitter. An ideal host
material should meet the following four criteria:
i) large band gap > 6 eV to be transparent for excitation and UC emission light
ii) high tolerance for dopant concentrations of Ln3+ to avoid lattice defects
iii) low phonon energies of the host lattice to minimize nonradiative deactivation
iv) chemically and thermally inert
14
These criteria are met by inorganic host materials based on certain rare earth ions (Re3+):
Sc3+, Y3+, Gd3+, La3+ or Lu3+. These materials allow high dopant concentrations of the
luminescent centers that substitute these Re3+ ions in the host lattice. Moreover the
luminescent centers have similar ionic radii like all Re3+ ions (Ln3+ ions are a subgroup of the
group of Re3+ ions). The most common host matrices used for UC materials are oxides (e.g.
Y2O3),87 oxysulfides (e.g. Y2O2S),88 oxyphosphates (e.g. LuPO4 and YbPO4)89 and fluorides (e.g.
NaYF4)90. In addition, other alkaline earth ion and transition metal ion based compounds are
frequently used as host materials (e.g., Ca2+,Zr4+).91, 92
Among these materials, the fluoride-based NaReF4 have been used for decades as nm-
sized host lattices due to their low phonon energy (quantized energy portions of lattice
vibrations), excellent chemical stability, and thermal stability.93-95 NaReF4 has two different
crystal phase structures, the cubic (−NaReF4) and the hexagonal phase structure
(−NaReF4), see Figure 2.4.
Figure 2.4 Schematic presentation of the crystal structure of NaReF4 in the cubic and hexagonal-
phase. a) The cubic phase is characterized by an equal number of F- cubes containing
cations and vacancies; b) the hexagonal phase shows an ordered array of F- ions with
two kind of cation sites, where one is occupied by Na+ and the other one is randomly
occupied by either Na+ and Re3+. Reprinted from Wang et al.96 with permission from
Nature Publishing Group.
It is well established that the family of −NaReF4 is much more efficient as host lattice
for UC materials than −NaReF4, due to their lower phonon energies of about 350 cm-1 and
a lower symmetry of the crystal field.97, 98 These lower phonon energies result in a smaller
probability of phonon-based nonradiative deactivation. The probability of such a
15
deactivation process, also called multiphonon relaxation (MPR), depends exponentially on
the number of phonons with an energy
phonon needed to bridge the energy gap
between
the final and the initial energy level and can be calculated with Equation 2.5.99, 100
𝑊MPR(𝑝) = 𝛽𝑒−𝛼𝑝 Equation 2.5
Here,
and
depend on the specific properties of the host lattice and p =
/
phonon
equals the number of phonons required for the nonradiative deactivation. MPR is the
dominant nonradiative process for perfectly crystalline materials. However, for nm-sized
particles the nonradiative deactivation is mostly governed by lattice defects or molecules
near the surface or inside the particle volume.59, 101 In particular, this is critical for ET-based
UC systems, where the ET can be effectively transferred via energy migration to quenching
centers. A recent study showed that surface-related quenching and the number of defect
centers can be strongly minimized with a thick inactive shell and slow particle growth rates.39
2.4 Upconversion luminescence (UCL) of -NaYF4:Yb3+,Er3+ crystals
The UC system -NaYF4:Yb3+,Er3+ combines the previously discussed criteria for host
lattice and luminescent centers for an efficient ETU-based UC material. This material is
intensively studied since the mid-1960s and well known for its high UC efficiency. Typical
dopant concentrations range from 17-20% for the Yb3+-ion and 1-2% for the Er3+-ion dopant
concentrations aiming on an intense Er3+ green UCL.21, 97 ESA processes involving the Er3+
energy levels can be neglected, due to the 10-fold higher absorbance of Yb3+ compared to
Er3+. Figure 2.5 shows a typical UCL spectrum of −NaYF4:Yb3+,Er3+, which is dominated by
the Er3+ emission bands ranging from 370-870 nm. The strongest UC emission bands are
centered at 380 nm, 410 nm, 520 nm and 540 nm, 655 nm, 810 nm, and 850 nm. All Er3+
emission bands are characterized by a substructure of several sharp emission peaks, caused
by a splitting from the crystal field of the host lattice. The associated emissive Er3+ energy
levels can be populated via two or more energy transfers from Yb3+ to Er3+ and in
combination with internal nonradiative relaxation processes, see Figure 2.6.
16
Figure 2.5 UCL emission spectrum of µm-sized -NaYF4:Yb3+(21%),Er3+(2%) UC particles (UCµPBAM)
excited at 976.4 nm at a P of 20 Wcm-2. The typical Er3+ emission bands are centered at
380 nm (4G11/2 → 4I15/2), 410 nm (2H9/2 → 4I15/2), 520 nm/540 nm (2H11/2,4S3/2 → 4I15/2), 655
nm (4F9/2 → 4I15/2), 810 nm(4I9/2 → 4I15/2), and 850 nm (2H11/2,4S3/2 → 4I15/2). For a better
visualization the UV and NIR bands were multiplied by a factor of 4. The Er3+ green and
red emission bands account for > 80% of the overall UCL intensity.
The Er3+ green emission bands (2H11/2, 4S3/2 → 4I15/2) centered at 520 nm and 540 nm and
Er3+ NIR 850 nm emission band (2H11/2, 4S3/2 → 4I13/2) centered at 850 nm are activated via
two Yb3+-Er3+ ETs. The first ET populates the Er3+ 4I11/2 energy level, followed by a second ET
populating the Er3+ 4F7/2 energy level. Due to the high nonradiative rate for Er3+ 4F7/2 energy
level, the excited electron relaxes mainly to the directly lower lying 2H11/2,4S3/2 energy levels.
Subsequently, an optical transition can take place either to the ground state 4I15/2 or the first
excited state 4I13/2 of Er3+ resulting in the green 520/540 nm or NIR 850 nm emission,
respectively.
The Er3+ NIR 810 nm emission band (4I9/2 → 4I15/2) centered at 810 nm can be populated
via two different biphotonic pathways. The first population pathway is via CR from the
biphotonically activated 2H11/2,4S3/2 to 4I9/2 and the second via Er3+-Er3+ ET in case of a high
population of the Er3+ 4I13/2 energy level (4I13/2 +4I13/2→ 4I9/2).
17
The Er3+ ultraviolet (UV) emission band (4G11/2 → 4I15/2) centered at 380 nm and the Er3+
purple emission band (2H9/2 → 4I15/2) centered at 410 nm are populated by a triphotonic UC
process, i.e. that three sequential Yb3+-Er3+ ETs are involved. The first two ET steps are
identical as described above for the biphotonically activated green-emitting 2H11/2,4S3/2
energy levels. The third ET excites the electron to the 2G7/2 energy level, followed by
nonradiative relaxation to the 4G11/2 and 2H9/2 energy levels. The optical transition to the
ground state 4I15/2 then results in the UV or purple emission, respectively.
The Er3+ red emission band (4F9/2 → 4I15/2) centered at 655 nm has the most complex
nature of all emission bands as it can be activated via different biphotonic and triphotonic
pathways. A comprehensive overview of the conditions for bi- or triphotonic activation,
which was intensively debated within the last years, is given in the introduction of
Publication III. In the following a brief version of this overview is presented. The photonic
nature of the Er3+ red emission band depends on the crystal phase, size, particle architecture,
environment and excitation power density (P).23, 26, 29, 40, 42, 55, 96, 102-113 The triphotonic
population pathway suggested from the Berry group in 2014102 includes an Er3+-Yb3+ back
energy transfer (BET) from the triphotonically activated UV emissive 4G11/2 energy state to
the red emissive 4F9/2 energy state. The biphotonic activation of the Er3+ 4F9/2 energy level
can occur via two different population routes. The first biphotonic population pathway
directly feeds the 4F9/2 energy level via nonradiative relaxation from the biphotonically
activated 4S3/2 energy levels. Therefore, an energy gap in the order of 3000 cm-1 to 3200 cm-
1 has to be overpassed. In the case of UCNPs capped with oleic acid molecules the stretching
vibrational modes of the CH2-groups (2800 cm−1 to 2950 cm−1) are perfectly capable to bridge
this energy gap.114 The second biphotonic pathway occurs via the 1520 nm 4I13/2 energy level
of Er3+, which can be populated by nonradiative relaxation from one ET activated Er3+ 980
nm 4I11/2 energy level. Due to an energy mismatch of 1000 cm-1 to the Er3+ red emissive 4F9/2
, this ET needs assistance by phonons and requires population of the 4I13/2 energy level.
18
Figure 2.6 Term scheme illustrating the relevant population pathways for the different Er3+
emission bands for Yb3, Er3+ based UC materials. Solid arrows represent excitation or
de-excitation of an energy state by absorption or emission of a photon; the dashed
arrows represents ET processes including Yb3+,Er3+ ET , Er3+-Yb3+ BET, CR between two
Er3+ ions and Er3+-Er3+ ET; curved arrows presents nonradiative relaxation processes by
MPR. It should be noted that for simplification the splitting of energy levels by the crystal
field of the host lattice is not included in the term scheme.
19
2.5 Upconversion photoluminescence quantum yield (
UC)
The
(also often referred as “internal quantum yield"), presents the ratio of the number
of emitted (Nem) to the number of absorbed absorbed photons (Nabs), see Equation 2.6.11, 21,
115, 116 This key parameter characterizes the photon conversion performance of
photoluminescent materials.
𝛷 = 𝑁em/𝑁abs Equation 2.6
In the case of the
UC, only emitted photons with a higher energy than the energy of the
absorbed ones are considered (
em >
abs), see Equation 2.7.11, 53, 102, 115, 117
𝛷UC = 𝑁em/𝑁abs , for
em >
abs Equation 2.7
As the UC process is a nonlinear process, a higher excitation power density (P) generally
leads to an increase of
UC. In the case of a biphotonic UC process, the UCL intensity (IUC) is
proportional to P², and consequently,
UC increases linearly with P. However, due to the
energy conservation law, the P-dependent increase of
UC has to level off at higher P, and
thus
UC is limited by 100%/n, where n is the photonic order of the UC process.28, 118 The
saturation behavior of the P-dependent
UC (
UC(P)) for an ideal biphotonic emitter was
studied by the Andersson-Engels group. They assumed a two-level emitter system and
subsequently derived a formula for
UC(P) for this simple UC system, see Figure 2.7 a) and
Equation 2.8.119
𝜙UC(𝑃) = 𝜙UC,sat /(𝑃balance
𝑃+ 1) Equation 2.8
Here,
UC,sat is the saturated upconversion quantum yield, where
UC(P) converges at
high P, and Pbalance is the balancing excitation power density, where
UC(Pbalance) equals
UC,sat/2.
20
Figure 2.7 b) shows the typical shape of the
UC curve for such an ideal biphotonic
emitter, calculated with Equation 2.8. This formula is only valid for purely biphotonic UC
processes like the NIR 800 nm emission band of Tm3+,Yb3+ and the green 520/540 nm of
Yb3+,Er3+ based UC systems.40, 119, 120 In contrast, this formula is not suitable to describe the
shape of
UC(P) for the red emission of Yb3+,Er3+ or Yb3+,Ho3+ as here the population
processes are more complicated.40, 55, 102, 121
Figure 2.7 P-dependent
UC behavior of a two-level emitter ETU system: a) Term scheme of the
two-level emitter system; b)
UC(P) curve calculated for
UC,sat = 50% and
Pbalance = 1 Wcm-2 with Equation 2.8.119
The simulation of
UC(P) for more complex UC systems can be achieved by using a rate
equation system consisting of a set of coupled differential equations, see Equation 2.9.122
𝑑𝑁𝑖
𝑑𝑡 =∑populating rates − ∑depopulating rates Equation 2.9
Thereby, the population densities (Ni) of the most significant 4f energy levels of the Ln3+
ions and their interactions have to be considered. The interactions are represented by
populating and depopulating rates of the energy states including radiative rates and
nonradiative rates like ET, CR, and MPR.A very decent approach was demonstrated by the
Berry group for -NaYF4:Yb3+,Er3+ µm-sized and nm-sized UC systems with a rate equation
system including a set of nine energy levels for Er3+.102, 123 This rate equation system was later
also used in Publication II and III to draw and underpin interpretations of the influence on UC
mechanisms for -NaYF4:Yb3+,Er3+ UCNPs by modeling the P-dependent intensities of the Er3+
emission bands.
21
3 Reviews of Literature and Open Questions
In this chapter a review of the literature for each publication is presented. Furthermore,
open questions are formulated, which are addressed in the chapters 4, 5, 6 and 7.
3.1. Absolute upconversion photoluminescence quantum yield (
UC)
Recently, absolute
UC measurements have raised great interest in the UC community
as a quantitative comparison tool to identify strategies to improve the brightness of
UCNPs.124 Attempts to measure the
UC relative against a
UC standard or to determine
enhancement factors with a spectrofluorometer125-127 have shown limited reliability. This is
related to the fact that there is no reliable
UC standard available, and additionally, due to
often different scattering properties and different P-dependencies of
UC for UC materials.40
Therefore, the
UC can be only measured absolutely with a spectrally calibrated ISS done.51,
52, 54, 55, 128-132 Although, the increasing number of reports on ISS the comparability of the
UC
results obtained with these setups is still limited by the often missing description of the
instrument design and instrument characterization. For example, there is no report how the
influence of the excitation beam profile (BP), i.e. the spatial distribution of P, is considered.
Additionally, often reports provide only a single
UC value or the
UC has been measured
only in a small P-range of less than two orders of magnitude, which is insufficient for
monitoring saturation dynamics of the UC process. Up to the publication date of Publication
I (June 2017) the highest measured
UC value for Yb3+,Er3+-based systems was 7.8 % for a
-NaYF4:Yb3+(20%),Er3+(2%) µm-sized UC powder at P of 22 Wcm-2.54
Open questions
• What are the prerequisites for the design and characterization of an ISS for
reliable
UC measurements? (see 4.2.1 Integrating sphere )
• What are the prerequisites on the optical properties of the sample, measurement
procedure, and measurement geometry to obtain
UC values with minimum
uncertainty? (see 5.1. Measurement conditions for
UC(P))
22
3.2. Solvent-dependent luminescence quenching
UCNPs are generally more prone to surface quenching due to their much higher surface-
to-volume ratio and shorter distance between particle center and surface compared to
larger-sized particles. Thus, Ln3+ ions of UCNPs can transfer energy effectively to the surface
region, where the excited ions can be deactivated nonradiatively by impurities or crystal
defects on the surface, as well as vibrational modes of O-H and C-H bands of solvent and
ligand molecules.110, 112, 113, 133-136 Particularly, the interest in UCNPs for aqueous
environments for life sciences applications has triggered an increasing number of reports
studying solvent quenching mechanism in water (H2O).42, 107, 133, 137-139 Arppe et al. showed in
a joint publication in 2015 that Yb3+ ions play a major role for the migration of energy to the
surface. Therefore, they studied bare rod-shaped ca. 35 nm-sized -
NaYF4:Yb3+(17%),Er3+(2%) UCNPs dispersed in H2O and deuterated water (D2O), showing a
decrease of the Er3+ green (2H11/2,4S3/2 → 4I15/2) and Er3+ red (4F9/2 → 4I15/2) UCL intensity of
99.9% for UCNPs in H2O compared to D2O for the UCNPs in H2O.140 Wilhelm et al. reported
in a joint publication green-to-red ratios (Igreen/Ired) of NaYF4:Yb3+(17%),Er3+(2%) 23 nm-sized
UCNPs in dependence of different solvents and different surface ligands.138 This report did
not include a model for the change of population and depopulation processes in H2O.
Open questions
• How to measure
UC in strongly absorbent solvents like H2O? (see 5.1.
Measurement conditions for
UC(P), Heating of the UCNP sample and solvent)
• What is the influence on the
UC of surface protected UCNPs dispersed in
different solvents? (see 6.1 Solvent-dependent quantum yield (
UC))
• How are the UCL bands and de- and population processes of these UCNPs
affected by different solvents? (see 6.2 Emission color of UCNPs dispersed in D2O
and H2O)
23
3.3. Dopant concentration-dependent color tuning
The dopant ion concentration of Yb3+,Er3+-based UCNPs has direct influence on their
optical properties like
UC and emission color. An increase of the Yb3+ concentration leads to
an increase of the intensity of the Er3+ red emission band (4F9/2 → 4I15/2) at the expenses of
the Er3+ green emission band (2H11/2,4S3/2→ 4I15/2) as frequently reported.82, 104-106, 141-143
However, different explanations are given for this observation. For the variation of Er3+
concentration, ambiguous or even opposite trends of the Er3+ green-to-red intensity ratio
(Igreen/Ired) have been observed.141, 142, 144-146. Additionally, the evaluation of these data is
difficult as the dopant concentration-dependent Igreen/Ired is also affected by P, and most
publications provide the Igreen/Ired only for single P values. To clarify the mechanisms
influenced by the dopant concentration, a comprehensive rate equation analysis using P-
dependent spectral intensities for a broad P-range and lifetime data of the different emission
bands for -NaYF4:Yb3+,Er3+ is necessary. This has not been reported for varying the Yb3+ and
Er3+ concentration.
Open questions
• How do the Yb3+ and the Er3+ dopant concentration influence the P-dependent
UCL? (see 7.1 Dopant concentration-dependent upconversion luminescence
(UCL))
• How do the rate constants change by variations of the Yb3+ and the Er3+ dopant
concentration? (see 7.2 Influence of the dopant concentrations on rate equation
constants)
24
25
4. Samples and Experimental Methods
This chapter provides an overview of the investigated UC samples and a description of
the custom-built ISS with its optical and opto-electronic components and ratiometric
characterization, as well as the analysis of the data. At the end of the chapter, the commercial
setup for lifetime measurements is described.
4.1. Overview of investigated -NaYF4:Yb3+,Er3+ samples
In Table 4.1 all investigated -NaYF4:Yb3+,Er3+ UC particles are listed. For Publication I
commercial 3 µm-sized -NaYF4:Yb3+(21%),Er3+(2%) UC particles (PTIR 550) obtained from
Phosphor Technology Ltd. were studied.147 The UCNPs were synthesized by project partners
using different chemicals for the building blocks of
-NaYF4:Yb3+,Er3+ crystals. These reagents
were mixed and heated up to initialize formation of the UCNPs. To produce high quality
UCPNs with pure phase structure well-developed protocols with exact timing for each
heating step, high purity reagents, and optical growth control are required.138
UCNPs samples synthesized by Emilia Palo from the Prof. Soukka group, University
of Turku, Finland
Emilia Palo from the group of Tero Soukka from Finland synthesized the UCNPs, for
Publication I and Publication III, using a protocol from Ylihärsilä et. al.148 with small
modifications (see Table 4.1). For Publication I, oleic acid (OA)-capped 25 nm-sized -NaYF4:
Yb3+(17%) ,Er3+(3%) UCNPs were synthesized. For Publication III, two dopant concentration
series with different Yb3+ and Er3+ concentrations were synthesized (see Table 4.1). The
structural analysis was mainly performed by the Soukka group. This included X-ray diffraction
(XRD) measurements to verify the hexagonal crystal phase and transmission electron
microscopy (TEM) measurements to determine the size and size distribution of the UCNPs.
Moreover, measurements with a inductively coupled plasma optical emission spectrometer
(ICP-OES) were performed by M.Sc. Melissa Monks from the Federal Institute for Material
Research and Testing (BAM) in order to determine the dopant Er3+ and Yb3+ content of the
studied UCNP samples of Publication III.
26
UCNPs samples synthesized by Stefan Wilhelm from the Dr. Hirsch group,
University of Regensbur g, Germany
For Publication II, OA-capped 23 nm-sized -NaYF4:Yb3+(19%),Er3+(2%) UCNPs were
synthesized by Stefan Wilhelm, from the group of Dr. Hirsch from the University of
Regensburg, using a newly developed synthesis protocol 138 based on a protocol from Li. et
al in 2008 149. This new protocol yields high quality UCNPs with a small size distribution in
one large batch of about 2 grams. From this batch, surface modified UCNPs samples were
produced by functionalizing these UCNPs with different surface passivating and stabilizing
molecules to ensure dispersibility in different solvents.138 Two different surface modification
strategies were used to enable the transfer of the UCNPs to polar solvents like H2O, D20 and
DMF. The first strategy involves a complete exchange of the original OA molecules with
citrate and BF4-. In the second strategy OA-capped UCNPs are shelled with a second layer of
ligands molecules using DSPE, which are molecules soluble in polar and nonpolar solvents,
see Table 4.1. The Hirsch group also performed the structural analysis with TEM and XRD
measurements. Additionally, they did ICP-OES and dynamic light scattering (DLS)
measurements to determine dopant concentrations and solvodynamic particle diameters,
respectively.
27
Table 4.1 Investigated -NaYF4:Yb3+,Er3+ UC samples with respective average particle diameter, the
Yb3+ and the Er3+ dopant content, passivating ligands and solvent.
-NaYF4:Yb3+,Er3+
Samples
Particle
diameter
Yb3+
concentrationa
Er3+
concentrationa
Solvent
Ligand
Publication I
UCµPBAM powder
3 µm
21.4%
2.2%
-
-
UCNPs powder
25 nm
17%
3%
-
Oleic acid
UCNPs dispersion
25 nm
17%
3%
Toluene
Oleic acid
Publication II
UCNPs dispersion
23 nm
19.3%
2.3%
Cyclohexane
Oleic acid
UCNPs dispersion
23 nm
19.3%
2.3%
DMFb
BF4-
UCNPs dispersionc
23 nm
19.3%
2.3%
D2O
Citrate
UCNPs dispersionc
23 nm
19.3%
2.3%
H2O
Citrate
UCNPs dispersion
23 nm
19.3%
2.3%
D2O
DSPEd
UCNPs dispersion
23 nm
19.3%
2.3%
H2O
DSPEd
Publication III
UCNPs dispersion
33 nm
14.4%
0.9%
Toluene
Oleic acid
UCNPs dispersion
33 nm
14.4%
2.1%
Toluene
Oleic acid
UCNPs dispersion
33 nm
14.4%
3.1%
Toluene
Oleic acid
UCNPs dispersion
33 nm
14.4%
3.8%
Toluene
Oleic acid
UCNPs dispersion
33 nm
10.6%
3.1%
Toluene
Oleic acid
UCNPs dispersion
33 nm
17.1%
3.1%
Toluene
Oleic acid
UCNPs dispersion
33 nm
20.5%
3.1%
Toluene
Oleic acid
a Er3+ and Yb3+ substitute for the Y3+ ions of the -NaYF4 matrix; sum of dopant concentration of Er3+,Yb3+
and Y3+ equals 100%.
b N,N-dimethylformamide
c these samples are not discussed in this thesis
d 1,2-distearoyl-sn-glycero-3-phospho-ethanolamine-N-[methoxy-(poly-ethylene glycol)-2000]
(ammonium salt) (DSPE)
4.2 Optical characterization methods
In this section the custom-built new integration sphere setup for absolute measurement
of
UC and UCL is presented, as well as, the commercial setup used for time resolved PL
measurements.
28
4.2.1 Integrating sphere setup (ISS)
This subsection comprehensively describes the components and characterization of the
custom-built ISS. Moreover, the analysis of P-dependent UCL leading to information about
the photonic order and (de)population processes of the UC process is presented.
4.2.1.1 Components and characterization of the ISS
Figure 4.1 shows the ISS for absolute P-dependent
UC measurements, which I extended
from an existing ISS built by Dr. Christian Würth from the BAM.116, 150 A detailed description
of the ISS is also provided in the Supporting Information (SI) of Publication I. This ISS consists
of an excitation channel, where the excitation light is generated and guided to the sample,
and a detection channel to record the emitted UCL and excitation light in absolute spectral
photon fluxes (s-1m-2nm-1). The heart of the ISS is the integrating sphere. Its interior is
covered with a diffusive white reflective coating guarantying that after multiple reflections
the incident light is evenly distributed at any point of the inner surface. Consequently, the
measurements of photon fluxes with an ISS do not require any consideration of the direction
of propagation, the beam geometry and polarization of the excitation and emitted light.
Figure 4.1 Schematic presentation of the ISS for the absolute determination of the
UC(P) for UC
powder and UC dispersions. The setup is divided into an excitation channel including
the 8W 976 nm laser diode, filter wheels with attenuators with known transmission for
controlled attenuation of the excitation light and focusing optics, and a detection
channel including the integration sphere coupled into a filter wheel (equipped with band
pass filters), monochromator, and a Si-CCD detector. The sample was mounted in the
center of the integrating sphere. Reprinted from Kaiser et al.40 with permission from The
Royal Society of Chemistry.
29
Excitation channel
The excitation source is a special customized optical fiber-coupled laser diode (from
Roithner Lasertechnik GmbH) fulfilling the requirements for the absolute measurement of
UC(P) for NaYF4:Yb3+,Er3+ UC crystals. These requirements include a high power of 8 W for
characterizing the nonlinearity of the UC process, an adjusted emission wavelength of
976.4 nm that is resonant with the ground state absorption of Yb3+ (2F7/2 → 2F5/2 ) and a high
power stability < 0.1 % (see Figure 4.2) as prerequisite for minimum fluctuations of the
absolute
UC measurements.
Figure 4.2 Characterization results of the 976 nm laser diode used for
UC measurements:
a) Wavelength-dependent absorbance of Yb3+ (976.4 nm, 2F7/2 → 2F5/2 ) (black line) and
wavelength profile of the 976 nm laser diode with a FWHM of about 5 nm (red line),
which was adjusted to be resonant to the Yb3+ ground state absorption; b) Intensity
averaged wavelength of the laser beam wavelength with a stability < 0.4 nm over a
period of over 1 year; c) Laser power with a stability < 0.1 % for over 120 minutes.
Reprinted from Kaiser et al.40 with permission from The Royal Society of Chemistry.
The laser light from the 200 µm-sized output slit of the optical multimode-fiber was
collimated and then focused on the sample. In order to investigate excitation beam profile
(BP)-dependent effects on
UC, two different excitation geometries were realized by using
focal lenses with focal lengths of 500 mm and 125 mm to obtain a nearly homogenous Top
Hat BP (THexp) and an inhomogeneous Gaussian BP (Gaussexp), respectively (see Figure 4.3).
The P tuning range of the THexp is from 0.25 Wcm-2 to 410 Wcm-2 and for the Gaussexp from
2.5 Wcm-2 to 3400 Wcm-2, thus, allowing to tune P over four orders of magnitude. To
guarantee maximal stability of P, the laser power was attenuated by two automated filter
wheels equipped with reflective neutral density filters, which were placed between the
collimating and focusing optics. To avoid damage of the laser diode by directly back-reflected
30
laser light, these filter wheels were tilted by a small angle. Furthermore, for safety reasons
the reflected laser light was guided into a beam dump. In order to control the filter wheels,
the monochromator, and to readout the detector, a Labview program was written with the
help of the former bachelor student Nils Handelmann from the BAM.
Beam profile characterization
In order to verify BP-dependent effects on the
UC(P), the two realized excitation
geometries THexp and Gaussexp (see Figure 4.3 a)) were systematically characterized with a
beam profiler (Newport LBP2; pixel size: 8 x 9.3 µm). Therefore, cross-sections perpendicular
to the beam propagation were recorded in 1 mm steps for a path length of 10 mm illustrated
in Figure 4.3 b). The average beam diameter was defined at a 4 % intensity drop of the peak
intensity of the in direction of the laser beam propagation averaged intensity (see Figure 4.3
c)) to ensure comparable
UC for a biphotonic emitter at the same averaged P values. This
resulted in beam diameters of 1.4 mm and 0.55 mm for THexp and Gaussexp, respectively.
Figure 4.3 Experimentally realized illumination geometries used for the investigation of the
influence of the beam shape on the
UC(P) measurements. a) Schematic of the
realization of THexp BP (Top) and Gaussexp BP (Bottom) by the use of focal lenses with 500
mm and 125 mm focal length, respectively; b) Cross-sections of the P-distribution of
THexp (Top) and Gaussexp (Bottom) determined experimentally in 1 mm steps for a path
length of 10 mm; c) Averaged P-distribution in the propagation direction of the laser
beam for THexp and Gaussexp for a laser power of 8 W. Reprinted from Kaiser et al.40 with
permission from The Royal Society of Chemistry.
31
Detection channel
The first part of the detection channel is a BaSO4 inner-coated integrating sphere (from
Labsphere) with a diameter of 15 cm, reflection over 97% in the vis to NIR spectral region),
in which the sample was center-mounted with a BaSO4 coated sample holder. At the bottom
of the integrating sphere an optical fiber was attached guiding the light into a
monochromator (Shamrock 303i, grating with blaze angle of 500 nm, spectral resolution
0.5 nm, Andor Technology PLC) equipped with a filter wheel and a silicon-based charged-
coupled-device (Si-CCD) detector (Andor iDus CCD DU420-BRDD, 1024 x 256 Pixel with a pixel
width of 26 µm, Andor Technology PLC). A BaSO4-coated baffle was placed inside the
integrating sphere above the optical fiber to ensure that only diffusely reflected light was
detected. The filter wheel was equipped with five different bandpass filters to suppress
overexposure by stray light of the intense laser light and to cover the complete wavelength
region from 370 nm - 900 nm for the UCL detection. The laser light was recorded with a
calibrated absorptive neutral density filter with an attenuation factor of 5,600.
Calibration of the detection channel
The wavelength calibration and spectral calibration of the detection channel was
performed according to previously reported protocols.12, 150 Therefore, a correction function
for the wavelength scale was created by using a Hg-Ar discharge lamp from Ocean Optics
(HR4000CG-UV-NIR). This included the recording of the spectral atomic lines with the ISS and
comparison with the atomic spectra database from the National Institute of Standard and
Technology (NIST). This correction function was then used for all further measurements. The
optical response of the detection system was calibrated with a spectral radiance transfer
standard, which wavelength dependence (L
(
)) was calibrated against a black body radiator
by The National Metrology Institute of Germany (PTB). This spectral radiance transfer
standard consists of a halogen lamp mounted inside an integrating sphere, to guarantee a
diffuse spectral radiance. In order to determine the spectral instrument response functions,
the spectrum of the spectral radiance standard was recorded from 350 nm - 1050 nm for all
different bandpass filters used in the detection filter wheel. The obtained instrument
response functions were then multiplied with
/hc0 to obtain correction functions in units of
spectral photon flux s-1m-2nm-1. Additionally, an intensity correction function for different
illumination times of the Si-CCD detector has been recorded The validation of the instrument
response functions was performed by recording and comparing the intensity corrected
spectra of the certified spectral emission standards F003 - F005 from the BAM.151 In order to
control the filter wheels, the monochromator, and to readout the detector, a Labview
32
program was written with the help of the former bachelor student Nils Handelmann from
the BAM. Moreover, this program performed an automatic correction of the wavelength
scale and the spectral sensitivity of the readout data of the detector. Furthermore, the
complete analysis of the data, see 4.2.1.2 Data analysis, was automatized with an
additionally self-written Labview program.
Sample cells
All sample cells consist of high-quality quartz suprasil (QS) glass with an optical
transmission > 80% from 200 nm – 2500 nm. Small sample volumes have been used to
suppress indirect excitation and reabsorption effects (see 5.1. Measurement conditions for
UC(P)). The UCNP dispersion samples were filled into 10 mm (inner length) x 4 mm (inner
thickness) quartz cuvettes purchased from Hellma GmbH. As a blank sample, i.e. reference,
a cuvette filled only with the pure solvent was used. The powder samples were pressed into
5 mm (inner diameter) x 0.1 mm (inner thickness) custom-designed round quartz cuvettes
produced by Hellma GmbH. In this case, an empty round cuvette was used as a blank sample.
Identical measurement positions were ensured using the reflection with a Helium-Neon (He-
Ne)-laser.
4.2.1.2 Data analysis
In this subsection the analysis of the data obtained with the ISS is described. First, the
calculation of the P-dependent
UC is presented. Second, the further analysis of the P-
dependent UC emission spectra is detailed, which reveal additional information in respect
photonic order and the (de-)population pathways regarding the different UC emission bands.
Calculation of the P-dependent upconversion quantum yield (
UC(P))
The P-dependent
UC curve can be calculated as the quotient of the emitted and the
absorbed photon flux, see Equation 4.1. This calculation requires the measurement of a) the
P-dependent spectrally-corrected sample spectrum (Isample(,P)) and b) the P-dependent
spectrally-corrected blank sample spectrum (Iblank(,P)).
𝛷𝑈𝐶(𝑃)=emitted UC photon flux
absorbed photon flux =∫𝐼(𝜆,𝑃)𝑑𝜆
900 nm
350 nm
∫𝐼abs
982 nm
968 nm (𝜆,𝑃)𝑑𝜆 =∫𝐼sample(𝜆,𝑃) − 𝑇 𝐼blank(𝜆,𝑃)𝑑𝜆
900 nm
350 nm
∫𝐼blank
982 nm
968 nm (𝜆,𝑃) − 𝐼sample(𝜆,𝑃)𝑑𝜆
with 𝑇 = ∫ 𝐼sample(𝜆,𝑃)𝑑𝜆
982 nm
968 nm
∫ 𝐼blank(𝜆,𝑃)
982 nm
968 nm 𝑑𝜆 Equation 4.1
33
The P-dependent spectral photon flux (I(
,P)), needed for the calculation of the emitted
UC photon flux, was determined with a stray light correction from the excitation light of
Isample(
,P). Therefore, Isample(
,P) was subtracted by Iblank(
,P) multiplied with a factor T to
resulting to I(
,P). Thereby, the factor T considers scattering of the narrow band intense
excitation light inside the detection monochromator. This can result in artefacts in the
emission spectra, and subsequently, affects the calculated
UC. Particularly, for weakly
emissive UC samples this stray light correction has to be considered. The emitted absolute
photon flux, I(
,P) was then calculated by integration from 350 nm - 900 nm, thereby
including all UC emission bands of Er3+ and omitting the sensitizer Yb3+ luminescence above
900 nm. The absorbed absolute photon flux was obtained by integrating the absorbed
photon flux Iabs(
,P), equaling the difference of Iblank(
,P) and Isample(
,P), over the wavelength
range of the excitation peak, here 968 nm - 982 nm (see Figure 4.2 a)).
Analysis of the P-dependent spectral photon flux I(
,P)
Figure 4.4 a) presents a set of I(
,P) for a respective UC sample for a broad P-range. This
set of I(
,P) can be used to obtain the following parameters providing information on of the
UC processes for the different Er3+ emission bands: i) P-dependent integrated spectral
photon fluxes (I
(P)), ii) P-dependent relative spectral photon fluxes (Irel,
(P)), iii) P-
dependent green-to-red ratios (Igreen/Ired(P)), and iv) P-dependent slope factors (n(P)) (see
Figure 4.4 b), c), d) and e)). In the following these parameters are defined.
P-dependent integral spectral photon flux (I
(P))
The P-dependent integral spectral photon flux (I
(P)) was determined for each Er3+
emission band by integrating I(
,P) over the integration intervals presented in Table 4.2. The
I
(P) of the Er3+ green, red and purple emission bands are presented for a respective UC
sample in Figure 4.4 b).
34
Figure 4.4 Parameters derived from the P-dependent spectral photon flux (I(
,P)): a) I(
,P) of Er3+
purple (2H9/2 → 4I15/2), Er3+ green (2H11/2,4S3/2 → 4I15/2) and Er3+ red (4F9/2 → 4I15/2) emission
bands with P varying from 0.6 Wcm-2 to 380 Wcm-2 for a representative UCNPs sample;
b) P-dependent integral spectral photon flux (I
(P)); c) P-dependent relative spectral
photon flux (Irel,
(P)) presenting changes in the population density of the respective
emissive states; d) P-dependent green-to red ratio (Igreen/Ired(P)); e) P-dependent slope
factors (n(P)) revealing the photonic order and saturation behavior of the UC process.
Table 4.2 Intervals for integration of the intensity of the Er3+ UC emission bands.
color/center wavelength
Electronic transition
Integration interval
UV/380 nm
4G11/2 → 4I15/2
372 nm - 394 nm
Purple/410 nm
2H9/2 → 4I15/2
394 nm - 430 nm
Green/520nm,540 nm
2H11/2, 4S3/2 → 4I15/2
510 nm - 570 nm
Red/655 nm
4F9/2 → 4I15/2
630 nm - 685 nm
NIR 1/810 nm
4I9/2 → 4I15/2
783 nm - 833 nm
NIR 2/850 nm
2H11/2, 4S3/2 → 4I13/2
833 nm - 880 nm
Overall UCL intensity
370 nm - 900 nm
35
P-dependent relative spectral emission intensity (Irel,(P))
The P-dependent relative spectral emission intensity (Irel,
(P)) is defined by the quotient
of I
(P) to the overall UC photon flux (I
all(P)), see Equation 4.2.
𝐼rel,𝛥𝜆(𝑃)=𝐼𝛥𝜆(𝑃)
𝐼𝜆all(𝑃)=∫𝐼(𝜆,𝑃) 𝑑𝜆
𝜆em,up
𝜆em,low
∫𝐼
900 nm
350 nm (𝜆,𝑃)𝑑𝜆 Equation 4.2
Here,
em,up and em,low present the wavelengths of the upper and lower bounds used for the
integration of the UC emission bands, see Table 4.2. The Irel,
(P) of the Er3+ green, red and
purple emission bands are presented in Figure 4.4 c).
Irel,
(P) provides information about the population density of the respective energy
levels in dependence of P. The shape of Irel,
(P) was used to optimize rate equation constants
in Publication II and III. Moreover, the crossing point of the Er3+ green and red intensity, see
Figure 4.4 c), was recently used as indication point for the UC efficiency for -NaYF4:Yb3+,Er3+
UC systems with similar Yb3+ and Er3+ dopant concentrations.39
It should be noted, to avoid misunderstandings, that in Publications I and II Irel,(P)
referred to I
(P). The nomenclature defined in this thesis is in accordance to the
nomenclature used in Publication III.
Green-to-red ratio (Igreen/Ired)
Igreen/Ired(P) is the P-dependent ratio of the integral photon flux of the Er3+ green
(520,540 nm,2H11/2 ,4S3/2 → 4I15/2) to Er3+ red (655 nm,4F7/2 → 4I15/2) emission intensity, see
Equation 4.3 and Figure 4.4 d).
𝐼green 𝐼red
⁄(𝑃) = ∫𝐼(𝜆,𝑃) 𝑑𝜆
570 nm
510 nm
∫𝐼(𝜆,𝑃)
685 nm
630 nm 𝑑𝜆 Equation 4.3
This quantity is commonly used by the UC community to compare the properties of UC
samples with similar material composition and size. A high Igreen/Ired(P) for Yb3+,Er3+-based
upconverters can account for low nonradiative rates as the red emissive energy level can be
directly fed from the Er3+ 2H11/2 ,4S3/2 energy level via nonradiative relaxation (see Figure 2.7).
36
Slope factor (n(P))
The P-dependent slope factor (n(P)) represents the nonlinear increase of the emission
intensity with P (see Equation 4.4 and Equation 4.5). At low P, n(P) typically presents the
photonic order of the UC process. For example, in case of a biphotonic process n(P) equals
two at low P. Saturation of the UC process at higher P results in a decrease of n(P). For the
calculation of n(P) energetically neighboring data points I(P1),I(P2) were used, see
Figure 4.4 b) and Figure 4.4 e) and Equation 4.6.
𝐼(𝑃) ∝ 𝑃𝑛(𝑃) Equation 4.4
𝑛(𝑃)= 𝑑 log (𝐼(𝑃))
𝑑 log (𝑃) Equation 4.5
𝑛(𝑃)=log (𝐼(𝑃2)) − log (𝐼(𝑃1))
log (𝑃2) − log (𝑃1) Equation 4.6
An energy level can also be populated by a mixture of processes with different photonic
orders like the Er3+ red emissive 4F9/2 energy level, which can be populated via bi- and
triphotonic processes, see Figure 2.6. Such mixtures of different photonic population
processes can be characterized via the shape of n(P) of the Er3+ red emission band, see Figure
4.4 e).
4.2.2 Time-resolved photoluminescence spectroscopy
The lifetime curves of the emissive energy levels, characterized by their rise and decay
behaviors (see Figure 4.5), provide information about the temporal evolution of the
population and depopulation processes. Measurements of the lifetime curves were
performed with a commercial Edinburgh Instruments spectrofluorometer FSP-920. This
setup was equipped with an electrically-pulsed 1 W 978 nm laser diode (950 µs long square
pulse) to excite the sensitizer Yb3+ ions for the determination of the UCL lifetime curves for
the Er3+ 380 nm, 410 nm, 520/540 nm, 655 nm, 810 nm and the downshifted (DS) lifetime
curve of the 1520 nm emission bands. Additionally, DS luminescence lifetimes were
determined by directly exciting the Er3+ ions with an electrically pulsed xenon lamp (pulse
width ca. 1 µs) combined with an excitation double monochromator to select the respective
excitation wavelength. This provided the DS luminescence curves of the Er3+ 410, nm,
520/540 nm, 655 nm emission bands. Moreover, the DS lifetime of Yb3+ 976 nm emission
37
band was determined with excitation wavelength of 940 nm. The detection channel was also
equipped with a double monochromator to select the respective emission wavelength for
detection. The luminescence lifetime curves were then recorded with a red extended
photomultiplier tube (PMT) (R2658P) for the UV-NIR region, and a nitrogen cooled NIR PMT
(R5509P) for the Er3+ 1520 nm emission band. Distortion by stray light or undesired intensity
of other emission bands were canceled out by using different band pass filters.
In order to estimate the influence of the long-pulse excitation of the above described
setup, I also performed short-pulse measurements with a setup of the work group of Prof.
Dr. Hoffmann from the Technical University of Berlin for the UC samples investigated in
Publication I. Here, luminescence lifetime curves were obtained with short pulse excitation
with pulse lengths of ca. 10 ns and an excitation wavelength of 940 nm using a dye (IR140)
laser pumped with an excimer laser, see Figure 4.6. The deviations of the derived decay times
for short- and long-pulse excitation using a differential method, see Supporting Information
of Publication I, were < 10%.
It should be noted that the determined UC lifetimes are always results of combinations
of decay and ET processes.152 Furthermore, DS decay lifetimes can be affected by CR
processes.152 This underlines the fact that only with aid of a comprehensive rate equation
analysis, taking into account the significant interaction between the Yb3+ and Er3+ energy
levels, the population and depopulation dynamics can be described accurately as done in
Publications II and III.
Figure 4.5 Temporal rise and decay behavior of µm-sized -NaYF4:Yb3+(21%),Er3+(2%) UC particles
for the Er3+ purple at 410 nm (2H9/2 → 4I15/2), Er3+ green 520 nm/540 nm
(2H11/2,4S3/2 → 4I15/2) and 655 nm (4F9/2 → 4I15/2) emission bands. The lifetime curves were
obtained using a dye (IR140) laser pumped with an excimer laser with short pulse
excitation of 10 ns with an excitation wavelength of 940 nm with a setup from the work
group of Prof. Dr. Hoffmann from the Technical University of Berlin (TU-Berlin).
38
39
5 Measurement Strategies for the P-dependent
UC (
UC(P))
In this chapter prerequisites and challenges for absolute P-dependent
UC
measurements with the custom-built ISS (see 4.2.1 Integrating sphere setup (ISS)) are
presented for -NaYF4:Yb3+,Er3+ UC crystals. The first part of this chapter addresses stringent
requirements for the optical properties of the sample, the measurement geometry, and
measurement procedure. In the second part, the
UC(P) values of µm-sized
-NaYF4:Yb3+(21%),Er3+(2%) UC particles (UCµP) are validated by comparison to experimental
and theoretical literature data.
5.1. Measurement conditions for
UC(P)
In this section, the prerequisites and measurement strategies of the
UC measurement
are detailed. Disregarding these recommendations can lead to a high fluctuation, systematic
underestimation or overestimation of the measured
UC values. The following instructions
present a guideline for accurate P-dependent
UC measurements with minimum uncertainty.
Absorption of the UC sample
An accurate
UC measurement can be only guaranteed for a UC sample with a suitable
absorption. This parameter can be tuned by either varying the particle concentration or the
optical path length. The lower measuring limit of the UC sample absorption sensitively
depends on the stability of the excitation source and on the reproducibility of measurement
conditions of the sample and the blank. Whereas, the upper measuring limit depends on the
optical path length and nonlinearity of the UC material.
Figure 5.1 shows that with the use of a high-stability excitation source of 0.1%, as utilized
in this work, see Figure 4.2 c), a reasonable
UC determination requires UC samples with an
absorption larger than 1% to minimize fluctuation of a single
UC measurement down to 10%.
However, systematic errors regarding the positioning, the preparation and the optical
properties of the sample and the blank must also be considered. Although identical
positioning can be ensured with the reflection of a He-Ne laser, small slight deviations of the
reflectance of the quartz cuvettes from about 0.1% - 0.4% showed to be a crucial uncertainty
source at low UC sample absorptions. Therefore, a lower limit of 5% for the absorption of
the UC sample is recommended to minimize these systematic errors. In case of UCNPs
40
dispersions with very low particle concentrations, this recommendation may not be fulfilled.
However, the use of the same quartz cuvette for UC sample and blank has still proven to be
moderately precise for absorption measurements down to 2%.
A strong absorptive UC sample can lead attenuation of the laser beam inside the sample
volume resulting to inhomogeneous spatial emission for these nonlinear emitters. To
minimize this effect, an upper limit of about 20%, for the absorption of the UC sample with
an optical path length of 10 mm is recommended.
Figure 5.1 Relative fluctuations for a single
UC measurement in dependence of the fraction of
absorbed light for different laser power stabilities. The 976 nm laser diode used for the
absolute
UC measurements in this thesis has a power stability < 0.1 %. The relative
fluctuations of the
UC measurements were calculated from the quotient of the relative
laser power stability to the relative absorption of the UC sample.
Scattering of the UC sample
Another common issue, for
UC measurements, results from the scattering of the
excitation light by the UC sample. Particularly, scattering can critically influence the
UC
measurements for i) dispersions with UCNPs sizes larger than 50 nm or agglomerated UCNPs
and ii) for powder samples. Additionally, scattering of the excitation light by UCNPs
dispersion can lead to a lowered P inside the sample volume, and hence, results in
underestimated
UC values. Therefore, it was ensured for all investigated UCNPs dispersions
that no detectable scattering occurs at the excitation wavelength, see Figure 5.2.
Light scattering by powder samples can lead either to a decrease of P, due to a diffusion
of the incident light, or even to an increase of P by multiple reflections, dependent on the
preparation of the powder sample. In order to minimize internal scattering effects, the
powder samples were prepared in thin cell cuvettes having an inner thickness of 100 µm. In
addition, the sample holder, placed in the center of the integrating sphere, was tilted by 30°
to avoid that diffuse back-scattered excitation light leave the integrating sphere via the
41
entrance port. This excitation geometry has the advantage that there is no need for a non-
absorbing reference sample with similar scattering properties compared to the powder
sample, since most of the reflected incident light is collected with the integrating sphere.
Figure 5.2 Absorbance spectra of 25 nm-sized UCNPs dispersed in toluene taken from
Publication I. This shows the absence of scattering at the excitation wavelength of
976 nm for this UCNPs dispersion, ensuring no effect on the measured
UC. Reprinted
from Kaiser et al.40 with permission from The Royal Society of Chemistry.
Yb3+ 980 nm emission (4F5/2 → 4F7/2)
Figure 5.3 a) schematically shows the distortion of the measured absorption, due to the
overlap of emitted light from the sensitizing Yb3+ ion with excitation light. This issue was
addressed by subtracting the emitted intensity of Yb3+ from the recorded intensity.132
Therefore, the real shape of the Yb3+ 980 nm emission band was measured separately with
an excitation wavelength < 950 nm and then normalized to the distorted recorded sample
spectrum. Fischer et al. reported deviations of up to 40% for
UC of core-shell UCNPs
between Yb3+ emission-corrected and -uncorrected values.132 For the UCNPs studied in this
work, no Yb3+ 980 nm emission-correction was necessary, as the Yb3+ emission intensity is
weak for these unshelled systems.40 In case of the µm-sized -NaYF4:Yb3+,Er3+ particles
(UCµP), investigated in Publication I, a Yb3+ 980 nm emission-correction was required. The
relative deviations of Yb3+ emission-corrected to -uncorrected absorption as well as
associated
UC values additionally show a P dependence with 14% at a P of 0.2 Wcm-2 down
to 3% for a P of 100 Wcm-2, see Figure 5.3 b) and c).
42
Figure 5.3 Emission-correction method of the
UC(P) for the intensity of the Yb3+ 980 nm emission
band (4F5/2 → 4F7/2). a) Schematic of the Yb3+ emission-correction method including the
Yb3+ 980 nm emission band (black line, area under curve in shaded green), blank
spectrum (black line), distorted sample spectrum (dashed red line) and Yb3+ 980 nm
emission-corrected sample spectrum (blue line); b) Yb3+ emission-corrected absorption
values and c) Yb3+ emission-corrected
UC(P) values; of the UCµP sample studied in
Publication I. Reprinted from Kaiser et al.40 with permission from The Royal Society of
Chemistry.
Reabsorption and indirect excitation
Multiple diffuse reflections inside the integrating sphere can lead to reabsorption of the
emitted light and indirect absorption of the excitation light by the UC sample. Thereby,
reabsorption can reduce the emitted photon flux, whereas indirect excitation is associated
with a very low P, both leading to underestimated
UC values. These values can be corrected
with a method developed by MacDougall et al.,153 who extended the widely known
reabsorption-correction for linear emitters developed from de Mello.154 However, these
effects can be easily circumvented by minimizing the sample volume compared to the
integrating sphere volume, making the events of reabsorption and indirect excitation
negligible, and thus, a correction obsolete. Suitable sample cell sizes can be determined by
measuring
UC for a series of sample cells with varied volumes.
Excitation beam profile (BP)
Linear emitters like fluorescent dyes and QDs do not need any consideration of the P-
inhomogeneity of the BP. However, UCL depends strongly on P. Consequently, an
inhomogeneous BP results in a distribution of
UC(P) for these nonlinear emitters.
These BP-dependent effects on the
UC(P) were quantified by a calculation using the
experimentally realized BPs (see Figure 4.3). For this pupose, the formula of the Anderson
43
Engels group for the
UC(P) of an ideal biphotonic emitter (see Equation 2.8) was extended
by including the spatial P-distribution of the BP of the illuminated sample volume (detailed
in Publication I). Figure 5.4 shows the P-dependent deviation of the
UC(P) for the
experimentally realized THexp and Gaussexp shaped BPs (see Figure 4.3) compared to an ideal
Top Hat (THideal) shaped BP, i.e. a perfectly homogenous BP. At low P, The deviations of
UC(P)
for Gaussexp and THexp account to about 45% and 10%, respectively. This underlines the need
for a homogeneous Top Hat BP for accurate
UC(P) measurement.
Figure 5.4 Calculated deviation of
UC(P) for experimentally realized beam profiles. Here, the
experimentally realized THexp and Gaussexp BPs (see Figure 4.3) were compared to a
THideal BP. At low P, the deviation of the simulated
UC(P) values obtained with the THexp
and Gaussexp BPs accounts for up to ca. 10% and 45%, respectively. Reproduced from
Kaiser et al.40 with permission from The Royal Society of Chemistry.
Heating effects at high laser power
The excitation of the UC sample with high laser powers is accompanied by an increase of
its temperature. This results in a decrease of the measured
UC(P) due to thermal
deactivation of the emissive Er3+ energy levels. The heating of the UC sample depends on the
measurement time, laser power, and beam diameter, as well as on the UC sample specific
properties like size, absorptivity, crystalline quality, and the optical and thermal properties
of its environment (e.g. absorption and heat capacity of the solvent). By using a short
measurement time of less than 30 s, the thermally induced intensity loss was limited to
maximally 30 % at the highest P used for the investigated UC samples studied. In particular,
powder samples found to be quite prone to degradation for P > 130 Wcm-2 for the used beam
diameters of sizes in the mm-range. Instead, smaller beam diameters in the µm-range used
typically in microscopic studies allows excitation of UC powders with very high P of more
than 106 Wcm-2.155 UCNPs dispersions generally have a better heat transfer than powders
44
allowing excitation with laser powers of more than 8 W (used for the here investigated
UCNPs dispersions) without harming the samples.
The property of UCNPs to act as nanothermometers allows monitoring the temperature
increase for different P. Therefore, the temperature-dependent intensity-ratio
I520nm/I540nm(T) of the thermally coupled Er3+ 520 nm (2H11/2→ 4I15/2) and 540 nm (4S3/2→ 4I15/2)
emission bands is utilized, see Figure 5.5 a).156, 157 Note that the I520nm/I540nm(T) depends on
many parameters, like e.g. size, solvent and dopant concentration of the UCNPs and has to
be calibrated separately. Exemplarily, Figure 5.5 b) represent the temperature of UCNPs
dispersed in water (H2O) and heavy water (D2O) for different P, excited with a Gaussexp BP
for under 30 s, with an optical path length of 10 mm and a sample volume of about 1 ml. The
drastic increase of the temperature from 25 °C to 60 °C of UCNPs dispersed in H2O, at
P = 1000 Wcm-2, mainly results from the high absorption of H2O of 40% at 976 nm for the
chosen 10 mm optical path length. This underlines the critical influence of absorbing solvents
on the actual temperature of the sample for
UC measurements. In contrast, for UCNPs
dispersed in D2O the temperature increase is much less pronounced with 10° C at
P = 1000 Wcm-2. This increase results solely from nonradiative deactivation processes of the
UCNPs as D2O does not absorb at the excitation light.
Figure 5.5 Monitoring of the temperature increase of UCNPs by laser excitation during the
UC(P) measurement: a) Temperature-dependent Er3+ 520 nm to Er3+ 540 nm emission
intensity ratio I520nm/I540nm(T) from 20° C - 80° C separately recorded and used for
calibration; b) P-dependent temperature increase for DSPE-capped UCNPs dispersed
in H2O (blue) and D2O (black), for standard
UC measurement conditions described in
the text, calculated with the I520nm/I540nm(T). Reproduced from Würth, Kaiser et al.55
with permission from The Royal Society of Chemistry.
45
UC measurement in H2O at high P
Accurate
UC of measurements for UCNPs dispersed in H2O are very challenging, since
H2O has a high absorption coefficient at the excitation wavelength of 976 nm. This results in
heating up of the UCNPs sample for high P (see Figure 5.5 b)) resulting in a decrease in
intensity. Moreover, the absorption coefficient of H2O increases with increasing
temperature,158 which requires more specific measurement conditions for high P to avoid
overestimation of the measured
UC. This implies that blank and sample spectrum must be
recorded with the same illumination times to match their temperatures. Additionally, the
absorption of the UCNPs dispersion should be kept low at ca. 5-10%, so that nonradiative
deactivation processes minimally contribute to the increase of the temperature. In addition,
the high absorption of H2O leads to a lowered average P and to an increase of the
inhomogeneity of the BP. As discussed earlier in this section, an inhomogeneous BP leads to
overestimated
UC values.
46
5.2 Validation of measured
UC(P)
The aim of this section is to validate the measured
UC(P) data by comparing with
literature data obtained from different laboratories. For this purpose, high-quality bulk
systems with similar dopant concentrations are the best choice due to their low nonradiative
rates, and hence, well-comparable optical properties. As a representative bulk material,
commercially available 3 µm-sized −NaYF4:Yb3+(21%),Er3+(2%) UC particles (UCµPBAM) (from
Phosphor Technologies)147 were chosen. Reliable reports of
UC measurements of bulk
-NaYF4:Yb3+,Er3+ are rare since these measurements are challenging, as intensively
discussed in the previous sections of this thesis. Reported
UC data usually includes the
intensity data of a limited number or even only for a single UC emission band. Furthermore,
most works provide
UC data for small P-range < 2 orders of magnitude or even only a single
P, which is not enough to fully characterize the saturation behavior of the UC processes.
Therefore, this work provided a comprehensive characterization of the
UC of UCµPBAM
regarding the full spectral region from 360 - 900 nm, considering all UC emission bands with
a significant contribution to the overall UCL, recorded for a wide P range of over three orders
of magnitude with a well-characterized TH-shaped excitation beam profile.
Three reports were used for the comparison of the measured
UC(P) of UCµPBAM. The first
report from Page et al. (1997) provides the
UC(P) of the Er3+ green emission band
(
UC,green(P)) for µm-sized -NaYF4:Yb3+,Er3+ powder (UCµPPage).52 However, the dopant
concentration for UCµPPage was not specified. In the second report from the Van Veggel
group, the maximal
UC,green for -NaYF4:Yb3+(20%),Er3+(2%) particles with sizes >> 100 nm
(UCµPVanVeggel) was determined. Both of these reports used an ISS for the absolute
UC
determination. The third report from the Berry group in 2014 provides simulated values of
the spectral
UC(P)’s of different Er3+ emission bands of µm-sized rod-shaped
-NaYF4:Yb3+(18%),Er3+(2%) crystallites (UCµPBerry) (from Lorad Chemical Corp.).102
47
5.2.1 Validation by comparison with measured results from Page et al.
In 1997, Page et al. reported on the absolute determination of UC conversion efficiencies
for single emission bands for different green-, red- and blue-emitting UC phosphors.52
Thereby, they provided a detailed description of the spectral calibration of their ISS and
measurement procedure. For UCµPPage they determined the
UC,green(P) for a broad P-range
of over three orders of magnitude.
Figure 5.7 a) displays the
UC,green(P) curves for UCµPBAM and UCµPPage for a P-range from
0.2 Wcm-2 to 130 Wcm-2. The excellent agreement of these
UC,green(P) curves with similar
increase and identical maximal
UC,green of 2.4% at P = 20 Wcm-2 implies comparability of
these materials, i.e. crystal phase, crystal quality and Yb3+ and the Er3+ dopant
concentrations. Both
UC,green(P) curves starts to decrease for P > 20 Wcm-2, which may be
attributed to a competition of triphotonic to biphotonic processes at high P. In this respect,
thermal effects as reason for this decrease were excluded, see Publication I.
Figure 5.7 b) presents the normalized spectrally-corrected emission spectra of the Er3+
green and Er3+ red emission bands for UCµPBAM to UCµPPage at a P of ca. 20 Wcm-2. The good
match of these normalized spectra further supports the assumption that these materials are
comparable. Moreover, the domination of the Er3+ red emission intensity, known for its
triphotonic activation for µm-sized crystals,102 strengthen the hypothesis of triphotonic
processes being responsible for the fall of the
UC,green(P) at high P.
Figure 5.7 Validation of absolutely measured
UC(P) of the Er3+ green emission band (
UC,green(P))
for UCµPBAM (red color) to UCµPPage (black color) from Page et al. 52: a)
UC,green(P) curves
for UCµPBAM and UCµPPage displayed from 0.2 Wcm-2 to 130 Wcm-2 b) Normalized
spectrally-corrected emission spectra for UCµPBAM excited with P = 20 Wcm-2 and
UCµPPage excited with a P in the Wcm-2-range. The data from Page was digitized and
values were transformed from energy to photon numbers. Reprinted from Kaiser et al.40
with permission from The Royal Society of Chemistry.
48
5.2.2 Validation by comparison with measured results from the Van Veggel
group
The Van Veggel Group published in 2010, a strategy for the absolute measurement of
the
UC for UCNPs using a commercially available spectrophotometer (FLS 920 from
Edinburgh Instruments) combined with a 980 nm laser diode and an integrating sphere.51
This work is an important milestone in the UCNPs research area as the performance of UCNPs
was previously only compared relatively. In order to validate their
UC results, they
compared the maximal
UC,green value of 3 % for UCµPVanVeggel with the
UC,green value of 2.4%
from UCµPPage. This value also matches well with the
UC,green value of 2.4% measured for
UCµPBAM.
5.2.3 Validation by comparison with simulated results from the Berry group
In 2014, the Berry group simulated the spectral
UC(P)’s of the Er3+ green (
UC,green(P)),
Er3+ red (
UC,red(P)) and Er3+ purple (
UC,purple(P)) emission bands.102 These simulation results
were based on a comprehensive rate equation analysis utilizing measured P-dependent
intensity and UCL lifetime data of the Er3+ purple, green and red emission bands. This analysis
was based on a new rate equation model revisiting the back then outdated model of
population processes for the different Er3+ energy levels of -NaYF4:Yb3+,Er3+.
Figure 5.6 a) - d) shows the
UC,green(P),
UC,red(P),
UC,purple(P), and their sum (
UC,vis(P))
for UCµPBAM and UCµPBerry. The
UC,vis(P) and
UC,green(P) curves have similar shape for both
samples with relative deviation < 40%. Taking into account the complexity of the population
processes for the multitude Er3+ levels, these deviations between experimental and
simulated data can be considered as relatively small. However, the
UC,red(P) and
UC,purple(P)
show high deviations by factors of two and five, respectively. In particular for
UC,purple(P),
the shape also strongly differs between UCµPBerry and UCµPBAM. Moreover, the Igreen/Ired ratio
strongly deviates by a factor of ca. 3 between UCµPBerry and UCµPBAM. However, the Igreen/Ired
ratio determined for UCµPBAM are supported by the values of UCµPPage, see 5.2.1 Validation
by comparison with measured results from Page et al.. Further, the comparability of UCµPBerry
and UCµPBAM is underlined by similar luminescent decay kinetic of the different UC emission
bands, see Figure 5.6 e).
49
Figure 5.6 Validation of
UC(P)’s of experimental results for UCµPBAM (dark colored lines) by the
simulation results for UCµPBerry (light colored lines) from the Berry group:
a)-d) P-dependent spectral
UC of the Er3+ green (
UC,green(P)), red (
UC,red(P)) and purple
Er3+ emission bands (
UC,purple(P)), as well as the vis
UC(
UC,vis(P), equaling the sum of
UC,green(P),
UC,red(P) and
UC,purple(P)) for UCµPBAM and UCµPBerry; e) Comparison of
lifetime curves for UCµPBAM (exc = 940 nm, pulse energy of ca. 1 mJcm-2) compared to
UCµPBerry (exc = 943 nm, pulse energy of 66 mJcm-2) revealing similar (de-)population
dynamics for UCµPBAM and UCµPBerry. Reproduced from Kaiser et al.40 with permission
from The Royal Society of Chemistry.
50
5.3. Conclusions of chapter 5
A guideline for P-dependent
UC measurement with the newly custom-built ISS has been
developed. Special emphasis was given to the challenges and requirements on the optical
properties of the sample and measurement geometry for accurate
UC(P) measurements.
Overall, this underlines the need for careful consideration of these conditions to obtain P-
dependent
UC values with high precision. In particular, the choice of the BP showed to be
crucial factor for an accurate determination of the
UC(P) underlining the need for a
homogenous Top Hat BP.
The obtained
UC(P) data for a commercial µm-sized -NaYF4:Yb3+,Er3+ (UCµPBAM) was
validated with literature data from Page et al.52, the Van Veggel group51, and the Berry
group103. The independently measured maximal
UC,green(P) of UCµPBAM and UCµPPage with
2.4% as well as UCµPVanVeggel with 3% underlines the quality of spectral calibration and
suitability of the measurement strategies of the newly-developed ISS. The theoretical results
for UCµPBerry of the
UC,vis(P) and
UC,green(P) showed deviations up to 40% compared to the
measured values for UCµPBAM. Taken into account the complexity of the UC processes these
deviations can be considered as relatively small. However, the high deviations of UCµPBerry
and UCµPBAM for the
UC,red(P) and
UC,purple(P) by factors two and five, respectively, suggest
that a further optimization of their rate equation model for -NaYF4:Yb3+,Er3+ µm-sized UC
systems is needed.
These studies laid the ground for the following quantitative optical characterization of
-NaYF4:Yb3+,Er3+ UCNPs in dependence of the solvent and dopant concentration.
51
6 Solvent-Dependent Optical Properties of
−NaYF4:Yb3+,Er3+ UCNPs
This chapter deals with the influence of particle microenvironment and thereby, also
surface functionalization on the UCL and
UC of 23 nm-sized −NaYF4:Yb3+(19%),Er3+(2%)
UCNPs. In the first part, the impact of vibrational modes of different solvents on the
UC is
discussed. In the second part, a special emphasis was dedicated to the study of the
population mechanisms of the Yb3+ and Er3+ energy levels of UCNPs dispersed in water (H2O)
and heavy water (D2O). In contrast to H2O, the vibrational modes of D2O are shifted to lower
frequencies, and therefore, lead less likely to the depopulation of the Yb3+ and Er3+ energy
levels. This resulted to a model of the (de-)population processes for the participating Er3+
energy levels for UCNPs dispersed in H2O and UCNPs dispersed in solvents without UCL
quenching OH-bonds.
6.1 Solvent-dependent upconversion quantum yield (
UC)
In this section, the influence of different solvents on the
UC(P) of identical UCNPs
passivated with different surface passivating molecules was studied. Different surface
chemistry, i.e. different ligands, were required to render the UCNPs dispersible in different
solvents. The UCL and
UC(P)’s of the following four samples were recorded: i) oleate-capped
UCNPs dispersed in cyclohexane, ii) BF4-capped UCNPs dispersed in DMF and iii) DSPE-capped
UCNPs dispersed in H2O and iv) DSPE-capped UCNPs dispersed D2O (see 4.1 Overview of the
investigated
-NaYF4:Yb3+,Er3+ samples). The passivating surface molecules have two tasks: i)
to control particle colloidal stability by electrostatic or steric effects to prevent
agglomeration of the UCNPs and ii) to minimize the particle surface area accessible to the
solvent molecules aiming to reduce solvent-induced UCL quenching. The complete
suppression of solvent-induced UCL quenching requires a tight ligand shell preventing any
penetration of solvent molecules, which was not completely achieved here.
Figure 6.1 a) presents the P-dependent
UC of the investigated UCNPs. The
DSPE-capped UCNPs dispersed in D2O showed the best performance with a
UC of about
1.1% at P = 800 Wcm-2. In comparison, the UCNPs dispersed in DMF, cyclohexane and H2O
perform with 90%, 75% and 40% of this
UC value at 800 Wcm-2, respectively. Figure 6.1 b)
displays the absorbance spectra of the solvents near the spectral region of the Yb3+ 980 nm
emission band (2F5/2 → 2F7/2) This demonstrates a correlation for the reduction of the
UC by
52
the different solvents with the overlap of the absorbance spectra of the solvent and the Yb3+
980 nm emission band. This overlap provides a measure for the probability of the
nonradiative deactivation of near-surface excited Yb3+ ions via ET to the solvent molecules.
The strong absorption of the vibrational overtone O-H-mode at 980 nm of H2O can be
identified as the main factor for the diminishing of the
UC by a factor of 2.5 – 3 for the DSPE-
capped UCNPs in H2O compared to the UCNPs in D2O. This is also in accordance with the
reduction of the Yb3+ 980 nm emission lifetime from 160 µs to 40 µs, see Publication II. The
fact that Arppe et al. reported for 38 nm-sized bare UCNPs dispersed in H2O a UCL quenching
of 99.9% of the Er3+ green and red UCL compared to the same UCNPs dispersed in D2O
underlines the partial repulsion of H2O molecules from the surface of the 23 nm-sized DSPE-
capped UCNPs.
Figure 6.1 Surface quenching of 23 nm-sized -NaYF4:Yb3+(19%),Er3+(2%) UCNPs by solvent
molecules. a) Measured
UC(P) curves for UCNPs in different solvents: Oleate-capped
UCNPs in cyclohexane (C6H12) (orange squares), BF4-capped UCNPs in DMF (red circles)
as well as DSPE-capped UCNPs in heavy water (D2O) (black triangles) and water (H2O)
(blue triangles); b) Absorbance spectra of the solvents - same color code as in a) - and
the Yb3+ 980 nm emission band (2F5/2 → 2F7/2); Reproduced from Würth, Kaiser et al.55
with permission from The Royal Society of Chemistry.
6.2 Emission color of UCNPs dispersed in D2O and H2O
Subsequently, the differences in population processes at different P of the emissive Er3+
energy levels of the DSPE-capped UCNPs dispersed in H2O and D2O were investigated. Figure
6.2 a) shows the normalized UCL spectra of these UCNPs for a low P value of 16 Wcm-2 and a
high P value of 1000 Wcm-2. The spectra display the wavelength region from 500 nm –
53
900 nm including the Er3+ green at 520/540 nm (2H11/2,4S3/2 → 4I15/2), Er3+ red at 655nm (4F9/2
→ 4I15/2), Er3+ 810 nm (4I9/2 → 4I15/2), and Er3+ 850 nm (2H11/2,4S3/2 → 4I13/2) emission bands. At
P = 16 Wcm-2, an enhanced relative intensity of the Er3+ red and 810 nm emission bands can
be observed for the UCNPs dispersed in H2O compared to D2O. Thereby, the relative
intensities of the Er3+ green and Er3+ 850 nm emission bands, both originating from the
4S3/2,2H11/2 level are strongly reduced. This reveals that the population dynamics, at low P,
differ strongly for UCNPs dispersed in H2O and D2O. Contrarily, at a high P of 1000 Wcm-2,
the normalized UCL spectra are nearly identical in both solvents. Thus, the population
dynamics for the emissive Er3+ energy levels are assumed to be comparable. These
observations are supported by the P-dependent behavior of Igreen/Ired (Igreen/Ired(P)), see
Figure 6.2 b). For UCNPs in D2O, Igreen/Ired(P) decreases for increasing P and converge to a
constant value of about 0.4. In case of the UCNPs in H2O, the Igreen/Ired(P) starts at a 5 times
lower value, increases to a maximal value of about 0.7 and converges to the same Igreen/Ired(P)
value of about 0.4 as observed for the UCPNs in D2O. This difference of relative spectral
intensities of the UCL emission bands originate from the strong vibrational modes of H2O
located between 3300 cm−1 to 3700 cm−1 and their respective energy levels of Er3+ involved
in the population of the emissive states.159 In the case of D2O, these vibrational modes are
shifted to lower energies due to the higher mass of deuterium compared to the hydrogen
atoms, and thus, are not resonant anymore with the energy gaps between the Er3+ energy
levels, see Figure 6.3.
Figure 6.2 a) Normalized emission spectra of 23 nm-sized -NaYF4:Yb3+(19%),Er3+(2%) UCNPs
dispersed in D2O (black line) and H2O (blue line) for low P value of 16 Wcm-2 (top) and
a high P of 1000 Wcm-2 (bottom); b) P-dependent Igreen/Ired for UCNPs in D2O (black
symbols) and H2O (blue symbols). Reproduced from Würth, Kaiser et al. 55 with
permission from The Royal Society of Chemistry.
54
The energy scheme displayed in Figure 6.3 highlights the proposed dominant population
pathways of the different UCL bands of the UCNPs in H2O. This model accounts for the
efficient coupling of O-H vibrations to the Er3+ 4S3/2/2H11/2 and 4I11/2 energy levels, enhancing
nonradiative decay rates to the next lower energy level. At low P, the nonradiative decay of
the Er3+ 980 nm 4I11/2 energy level results in a high population of the Er3+ 1520 nm 4I13/2 energy
level for UCNPs dispersed in H2O. This favors the population from the Er3+ 4I13/2 energy level
to the Er3+ red emissive 4F9/2 energy level via Yb3+ to Er3+ ET (4I13/2 → 4F9/2) at low P, which
agrees with the observed increase of Igreen/Ired(P) in the low P-region. Additionally, the high
population of the Er3+ 4I13/2 energy levels is also responsible for the high intensity of the 810
nm emission band induced by the Er3+-Er3+ energy transfer process 4I13/2 + 4I13/2 → 4I9/2.160 In
contrast, for the UCNPs dispersed in D2O and organic solvents, at low P, the Er3+ red emissive
4F9/2 level is mainly populated via direct nonradiative relaxation from the Er3+ green emissive
4S3/2/2H11/2 level, see Figure 2.6. At high P, UCNPs showed solvent-independent population
of the Er3+ red emissive 4F9/2 energy level, by a triphotonic process via the 4G11/2 energy level
(see Figure 6.3, Right Panel), due to the compensation of solvent-induced quenching rates
by the high P.
These conclusions of the (de-)population of UCNPs in H2O and D2O were underpinned
with results from a rate equation analysis using the P-dependent emission spectra and
luminescence lifetimes of Er3+ 520/540 nm and 655 nm, as well as the Yb3+ 980 nm emission
bands, detailed in Publication II.
55
Figure 6.3 Energy level diagram for Yb3+-Er3+ interactions for UCNPs in H2O at low P (left) and
high P (right); red arrows show dominant population pathways; for high P the
dominant population pathways of UCNPs in H2O closely match with the photophysics
of UCNPs in organic solvents and D2O; Blue arrows: indicate nonradiative deactivation
by O–H vibrations of H2O. The arrow length represents the energy of the vibrational
mode. ET: energy transfer, ETU: energy transfer upconversion,and BET: back energy
transfer. Reprinted from Würth, Kaiser et al. 55 with permission from The Royal Society
of Chemistry.
56
6.3 Conclusions of chapter 6
In summary, the influence of vibrational modes of solvent molecules on the
UC(P) and
UCL was assessed for -NaYF4:Yb3+,Er3+ UCPNs with different surface modifications produced
from the same UCNPs batch. The optical absorption of the solvent vibrational modes at 980
nm shows to be directly connected to nonradiative deactivation of near-surface excited Yb3+
ions. In this respect, H2O with its highly absorptive vibrational modes, is acting as a strong
luminescent quencher for the UCL, as also previously reported.42, 138 The protection of the
surface of DSPE-capped UCNPs from H2O molecules is underlined by a reduction of UCL by
only about 60% in H2O compared to D2O. This is a strong improvement compared to the UCL
intensity loss of 99.9% of bare UCNPs dispersed in H2O reported by Arppe et al. et al.42, which
have a about five times higher particle volume compared to the here investigated UCNPs.
With the aid of UCL emission spectra and Igreen/Ired(P) ratios, the (de-)population
dynamics for the Er3+ energy levels of UCNPs in H2O were identified. At low P, the
fundamental vibrational modes of H2O from 3300-3700 cm-1 lead to high population of the
Er3+ 4I13/2 energy level, favoring the population pathway from the 4I13/2 energy level to the red
emissive 4F9/2 (655 nm) by an Yb3+ energy transfer. However, at high P, the quenching rates
are compensated resulting in identical (de-)population dynamics, and thus, identical
emission color for UCNPs dispersed in H2O to D2O. Interpretations for the (de-)population
dynamics in different solvents were supported by results of a rate equation analysis
performed with a model from the Berry group102, which was detailed in Publication II. These
methods for the refined understanding of the UC processes of UCNPs in H2O were a valuable
starting point for the investigations of the change of UC mechanisms by the variation of the
dopant concentrations, which is discussed in the next chapter.
57
7 Dopant Concentration-Dependent Optical
Properties of -NaYF4:Yb3+,Er3+ UCNPs
This chapter presents a study on the effect of varying the Yb3+ and Er3+ dopant
concentrations on the emission colors of -NaYF4:Yb3+,Er3+ UCNPs with similar size. The
P-dependent UCL data were measured with the ISS, and used as input data for a rate
equation model containing all relevant Er3+ levels to identify dopant concentration-
dependent parameters. This combination of experimental data and theoretical analysis
provided deep insights into the underlying UC mechanisms.
7.1 Dopant concentration-dependent upconversion luminescence
(UCL)
In this section, the experimental results of the P-dependent Er3+ green and red emission
intensity as a function of the Yb3+ and Er3+ dopant concentration of -NaYF4:Yb3+,Er3+ UCNPs
are discussed. The UCL was recorded over a wide P-range (> two orders of magnitude) for
two dopant concentration series, namely the Yb3+ and the Er3+ series. The Yb3+ series includes
four samples with Yb3+ sensitizer varied concentration from 11% - 21% and a constant Er3+
concentration of 3%. The Er3+ series consisted of four samples, with a Er3+ activator
concentration varied from 1% - 4% and a constant Yb3+ concentration of 14%. The sample
containing 14% Yb3+ and 3% Er3+ is part of both series. All UCNPs samples were similarly sized
to about 33 nm, and were capped with oleic-acid molecules to provide dispersibility in
toluene. This ensures that the observed effects are solely caused by the variation of dopant
concentrations. A list including both concentration series is provided in Table 4.1.
Two different P values were chosen as corner points for the comparison of the UCL of the
Yb3+ and Er3+ series. The first P value of Punsat = 1.8 Wcm-2 represents the unsaturated P-region
for these UCNPs: The population of the Er3+ activator ion energy levels induced by the Yb3+
sensitizer ion via ET is low, and thus, the events of BET from Er3+ to Yb3+ is less pronounced.
In this case, the biphotonic Er3+ green emission intensity increases with P². For higher P, the
increase of the Er3+ green emission intensity levels off due to saturation. The second value
Psat = 380 Wcm-2 marks the saturated P-region, where the intensity of the Er3+ green emission
band increases linearly with P. In this case, the Er3+ energy levels are highly populated so that
the contribution of BET from Er3+ to Yb3+ is relevant.
58
Figure 7.1 a) and c) presents the Igreen/Ired(P) for the Yb3+ and Er3+ series. The Igreen/Ired(P)
converges to constant maximum and minimum values for decreasing and increasing P.
Coincidentally, the Igreen/Ired(P) ratios reach minimum and maximum values at Psat and Punsat
for the investigated UCNPs. Analyzing the dopant concentration-dependent trend of the
Igreen/Ired(P) shows that with increasing Yb3+ concentration, the maximum value at low P is
unaffected, while the minimum value at high P decreases with increasing Yb3+ concentration.
Therefore the P-dependent color tuning range increases for increasing Yb3+, see Figure 7.1
a). In contrast, for increasing Er3+ concentration the maximum value at low P decreases, while
the minimum value at high P only shows a slight increase. Subsequently, the P-dependent
color tuning range can be increased with decreasing Er3+ concentration, see Figure 7.1 c).
Figure 7.1 b) and d) show the dopant concentration-dependent normalized UCL spectra
at Punsat and Psat for the Yb3+ and Er3+ series. In order to highlight changes of the population
of the respective energy levels, the intensity values of the UCL spectra were normalized to
the overall UCL intensity (Iall(P)) (integration interval from 370 nm to 900 nm). Subsequently,
the P-dependent intensities of the UCL spectra are relative contributions Irel(
,P). Further,
the P-dependent relative intensities of the Er3+ green and red emission bands are referred to
as Irel,green(P) and Irel,red(P) , respectively. The UCL spectra of the Yb3+ series at Punsat are
identical, underlining that the population dynamics of the respective Er3+ energy levels are
independent of Yb3+ concentration for this low P, see Figure 7.1 b). This reveals very good
comparability of crystallinity (low defect density), crystal size, surface chemistry, and spatial
Er3+ dopant ion distribution for this dopant concentration series, as Irel,green(P) and Irel,red(P)
are known to be sensitive to these parameters by affecting the rate constants of the Er3+
ions.40, 41, 55, 138, 161-163 Contrary, at Psat, Irel,red(P) increases by a factor of 1.5 and Irel,green(P)
drops by a factor of 2.3 for the Yb3+ concentration varied from 11% to 21%. An increase of
Irel,red for increasing Yb3+ concentration has been already reported in the literature, but the
concrete mechanism was still under discussion.82, 104-106, 141-143
The rate equation analysis, more detailed in the following section, showed that this
enhancement originates from an enhancement of the triphotonic population pathway to the
red emissive Er3+ 4F9/2 energy level fed by the triphotonically activated Er3+ 4G11/2 energy level
via a BET to Yb3+, see Figure 2.6.
59
For the Er3+ series, at Punsat, Irel,red(P) enhances and Irel,green(P) decreases by factors of 1.3
and 1.2 for Er3+ concentration varied from 1% to 4%, see Figure 7.2 d). This can be ascribed
to a higher number of surface or near-surface Er3+ ions, which enhance the probability of a
nonradiative relaxation from the Er3+ green emissive level 2H11/2,4S3/2 to the Er3+ red emissive
level 4F9/2 (see Figure 2.6), which was found to be the main process for the activation of the
Er3+ red emission band for these 33 nm-sized UCNPs dispersed in toluene at low P. Contrarily,
at Psat, Irel,red(P) decreases and Irel,green(P) increases each by a factor of only 1.1 for increasing
Er3+ concentration. These opposite trends for the Er3+ series at Psat and Punsat explain the
ambiguous spectral trends reported in the literature and underline the need of considering
a broad P-range.
Figure 7.1 Dopant concentration and P-dependent spectral properties of 33 nm-sized oleate-
capped
-NaYF4:Yb3+,Er3+ UCNPs in toluene. a), b): Yb3+ series with a Yb3+
concentration varied between 11% to 21% at a constant Er3+ concentration of 3%;
c), d): Er3+ series with a varied Er3+ concentration of 1% to 4% and a constant Yb3+
concentration of 14%; a), c) P-dependent measured (symbols) and simulated (lines)
Igreen/Ired values; b),d) UCL spectra from 515 nm to 675 nm for Punsat of 1.8 Wcm-2 and
Psat of 380 Wcm-2; all spectra were normalized to the total UCL intensity integrated
from 370 nm to 900 nm. Reprinted from Kaiser et al.164 with permission from Tsinghua
University Press.
60
7.2 Influence of the dopant concentrations on rate equation constants
To support the interpretation of the dopant concentration effects, a comprehensive rate
equation analysis of the UCNPs samples of the investigated Yb3+ and Er3+ series was
performed. The experimentally determined P-dependent intensity values and slope factors
(n(P)) of the Er3+ red and green emission bands as well as the luminescence decay curves of
the Er3+ green, Er3+ red, and Yb3+ 980 nm emission bands were used as fitting parameters.
The rate equation model, including all relevant Er3+ energy levels and interactions, was
adapted from the Berry group.102, 123 Despite the success of this rate equation model from
the Berry group in describing P-dependent UCL of UCNPs,123 a couple of rate constants had
to be changed drastically (by up to two orders of magnitude) for this work. To model the data
of the Yb3+ and Er3+ series simplifications were made, e.g. the obtained rate constants were
averaged over the complete particle volume. In contrast, Hossan et al. assumed an extra dark
layer for such core-only particles without an inactive shell layer.123 Comprehensive
descriptions of the optimization of the rate constants, simplifications, and fitting procedures
are provided in Publication III.
After optimizing the rate constants for the UCNPs sample containing 14% Yb3+ and
3% Er3+, only three parameters for the Yb3+ series and four parameters for the Er3+ series had
to varied, including the respective density of dopant ions. The good match of the fitted data
with the experimental data, including intensity and lifetime data of the Er3+ green and Er3+
red emission bands, see Figure 7.1 a) and c) and Publication III, supports the correct
identification of the affected relevant parameters responsible for the dopant concentration-
dependent color change of the UCNPs.
Figure 7.2 provides an overview of the rate constants affected by the variation of the
Yb3+ and Er3+ concentration. For both series, an increase of the nonradiative Yb3+ 980 nm rate
constant (kYb_NR) with increasing dopant concentration was observed. Thereby, kYb_NR was
directly assessed from the measured decay behavior of the Yb3+ 980 nm luminescence. The
increase of kYb_NR can be explained by a higher amount of near-surface Yb3+ and Er3+ ions that
can be deactivated by oleic-acid ligand molecules (C-H vibrational modes between 2700 cm-
1 and 2950 cm-1)114 or by the surrounding toluene solvent molecules (vibrational modes at
8000 cm-1)40. kYb_NR was found to be more sensitive to changes of the Er3+ concentration,
since the energy gap to the next lower lying level for the Er3+ 980 nm 4I11/2 is three times
smaller compared to the Yb3+ 980 nm 4F5/2 energy level. In addition, energy migration to the
particle surface is enhanced for higher dopant concentrations due to the shorter ion-ion
distances. The rate equation analysis showed that the change of kYb_NR does only affect the
61
saturation behavior of the population of the Er3+ activator energy levels, but has no effect on
the emission color of the UCNPs. Although the increase of kYb_NR also had a negative influence
on the
UC, it was shown that the increase of absorbing Yb3+ concentration overcompensates
this effect with respect to UC particle brightness, which is defined as product of the
UC,
cross section [cm²] of an Yb3+ ion in a NaYF4 matrix, and the number of absorbing Yb3+ ion
inside the UCNP, see Publication III.
For the Yb3+ series, all ET and BET rates were found to be enhanced simultaneously by
220% for a Yb3+ concentration from 11% to 21%, equaling a decrease of the average Yb3+-
Yb3+ and Yb3+-Er3+ distances by ca. 20% from 0.9 nm to 0.7 nm. The trend of the ion distance-
dependent ET and BET rates shows rather an exponential behavior, which may indicate that
the energy transfer between Yb3+ and Er3+ originates from electron hopping (Dexter ET), see
2.2 Energy transfer processes between Ln3+. This underlines the possibility to investigate the
physical nature of nonradiative ET for Yb3+,Er3+-based UC systems with a rate equation
analysis. Moreover, the rate equation analysis revealed that for increasing Yb3+
concentration, a pronounced BET from triphotonically activated 4G7/2 to the red emissive 4F9/2
is responsible for the increase of the Er3+ red emission intensity at Psat, see Figure 7.1 b). In
the case of the Er3+ series, ET and BET are not affected. This is in accordance with a self-
developed Monte Carlo simulation showing that for these relatively small Er3+ concentrations
the average distance of Er3+ ions to the nearest Yb3+ ion changes only minimally (see
Supporting Information of Publication III).
The observed color change for increasing Er3+ concentration results mainly from two
processes. As discussed previously, at low P, the enhanced nonradiative rate from the
2H11/2,4S3/2 to the red emissive 4F9/2 energy level (kNR6) by higher surface coupling results in an
enhancement of the Er3+ red emission intensity, see middle Panel of Figure 7.3 b). At high P,
the decrease of Irel,red(P) is assigned to a yet unknown Er3+-Er3+ ET leading to a depopulation
of the Er3+ red emissive 4F9/2 energy level. This was simulated with a simplified approach by
using an additional factor for the Yb3+-Er3+ ET rate kET5-8, which also depopulates the Er3+ 4F9/2
energy level. Therefore, the trend for the simulated change of Er3+-Er3+ ET, see Figure 2.7 b)
lower Panel, may not correspond to the real change of the Er3+-Er3+ ET rate. In order to
further optimize the rate equation system, four new possible candidates for the missing Er3+-
Er3+ ET rate constant have been suggested, which were detailed and discussed in Publication
III.
62
Figure 7.2 Dependence of the rate constants on the dopant concentration for varied
parameters of the rate equation analysis: a) Yb3+ series and b) Er3+ series; In the case
of the variation of the Yb3+ concentration, the nonradiative rate of the Yb3+ 980 nm 2F5/2
energy level and the ET and BET rates are influenced. For the Er3+ concentration, the
nonradiative rate of the Yb3+ 980 nm 2F5/2 energy level, the nonradiative relaxation from
the 2H11/2,4S3/2 to the red emissive level 4F9/2, and a not yet identified Er3+-Er3+ ET rate is
influenced. For the Er3+-Er3+ ET a simplified approach was used by varying an Yb3+-Er3+
ET depopulating the 4F9/2 energy level to simulate the enhancement of the red emission
band. In Publication III, four possible candidates for the missing Er3+-Er3+ were
discussed. Reproduced from Kaiser et al.164 with permission from Tsinghua University
Press.
63
7.3 Conclusions of chapter 7
In summary, the tuning of the UCL color of 33 nm-sized
-NaYF4:Yb3+,Er3+ UCNPs by the
variation of the Yb3+ and Er3+ dopant concentrations was studied for a P-range of over two
orders of magnitude. Combining experimental results and theoretical investigations with a
nine-level rate equation model from the Berry group, including the most significant energy
levels of Er3+,102, 123 allowed to make reliable statements about the influence of the dopant
concentrations on interactions between the ions and the particle surface.
The variation of the Yb3+ concentration from 11% to 21%, equaling a reduction in
Er3+-Yb3+ distance of 20%, resulted in an increase of 220 % for all Yb3+ to Er3+ ET rates and Er3+
to Yb3+ BET rates. This indirect measure of the change of ET processes opens the possibility
to study the physical nature of nonradiative ET processes. The corresponding increase of the
relative intensity of the Er3+ red emission band (Irel,red(P)) at high P was attributed to an
enhanced triphotonic activation of the red emissive Er3+ 4F9/2 energy level the Er3+ 4G11/2 to
the red emissive 4F9/2 energy level.
For the Er3+ series, different trends for Irel,red(P) at low P and high P were determined. At
low P, the increase of the number of Er3+ ions leads to an enhanced biphotonic activation of
the Er3+ red emission band via direct relaxation from the green emissive 2H11/2,4S3/2 to 4F9/2
energy levels, induced by an increased surface coupling of the Er3+ ions. At high P, a not yet
identified Er3+-Er3+ rate was found to be responsible for the decrease of Irel,red(P). Moreover,
these results reveal that the modulation range of the emission color by P can be extended
by increasing the Yb3+ and/or decreasing the Er3+ concentration. Overall, this highlights the
possibility of predicting the emission color for -NaYF4:Yb3+,Er3+ UCNPs by rate equation
modelling.
64
8 General Conclusions for Publications
In summary, an ISS has been designed, validated and utilized for the accurate absolute
determination of P-dependent
UC and UCL spectra for different −NaYF4:Yb3+,Er3+ UC
systems. The obtained
UC(P) and UCL results for -NaYF4:Yb3+, Er3+ UCNPs in different
solvents and with varied Yb3+ and Er3+ dopant concentrations, in combination with
theoretical modelling, allowed substantial extensions of the understanding of surface
quenching and color tuning.
Publication I summarized the challenges and requirements of the custom-built ISS for
UC(P) measurements for 976 nm-excitable UC materials. This includes prerequisites on
setup components and new calibration strategies. Moreover, measurement conditions for
powders and dispersed UC samples were derived with respect to measurement geometry,
experimental procedures, and optical properties of the UC material. This report can be used
as a protocol to perform
UC(P) measurements with minimum uncertainty in the order of
about 10 %. The maximum
UC of 10.5% at P = 30 Wcm-2 for a commercial
-NaYF4:Yb3+(21%),Er3+(2%) 3 µm-sized UC particles (UCµP) is nowadays being frequently
referenced as highest value for Yb3+,Er3+-based systems, underlining the overall accepted
accuracy of this new ISS.
In Publication II, identical UCNPs dispersed in organic solvents, water (H2O) and heavy
water (D2O) functionalized with different surface ligands were systematically studied.
Thereby the
UC was found to be mainly determined by the interaction of near-surface Yb3+
ions with the vibrational modes of the solvent and ligand molecules. In this respect, H2O leads
to the highest quenching rate from the UCNPs surface due to its high frequency O-H
vibrational modes resonant to the Yb3+ 980 nm energy level. For organic solvents and D2O,
the probability of energy transfer is less likely, since their C-H and O-D vibrational modes are
non-resonant to the Yb3+ 980 nm energy level as their frequencies are much lower. Instead,
the lower-frequency vibrational modes of these solvents in combination with the low
frequency vibrational modes of ligand molecules at ca. 3000 cm-1 clearly affected the
emission color of the UCNPs. Thereby, these vibrational modes can effectively bridge the
energy gaps between the multitudes of Er3+ energy levels ranging from 1000 cm-1 to 3500 cm-
1. Furthermore, the absolute UCL measurements, in combination with a rate equation
analysis, enabled the development of a new (de-)population model explaining the change in
emission color of UCNPs in H2O.
65
In Publication III, the influence of the dopant concentration on emission color and particle
brightness for a set of 33 nm-sized oleate-capped UCNPs dispersed in toluene was assessed
for Yb3+ and Er3+ dopant concentrations varied from 11 - 21% and 1 - 4%, respectively. The
experimental results, including emission intensities and slope factors of the Er3+ green and
red emission bands as well as luminescence lifetimes, were used for the optimization of the
rate constants of a rate equation model considering all significant Er3+ energy levels. As a
result an increase of the Yb3+ concentration was found to lead to enhanced triphotonic
activation and an increase in Er3+ concentration to an enhanced biphotonic activation of the
red emissive 4F9/2 energy level. Moreover, the results indicate that the particle brightness of
these UCNPs can be further improved with higher Yb3+ and lower Er3+ concentrations. This
enabled an estimate of trends for the ultimate goal to find optimized dopant concentrations
for maximal particle brightness in dependence of UCNPs size and particle architecture.
Overall, a refined understanding of UC processes of -NaYF4:Yb3+,Er3+ UCNPs was gained by
using quantitative measurements with the new custom-built ISS in combination with a
comprehensive rate equation analysis.
66
I
Appendix
List of Abbreviations
BET back energy transfer
BP beam profile
CCD charge-coupled device
CET cooperative energy transfer
CR cross-relaxation
DMF N,N-dimethylformamide, solvent for UCNPs
DS downshifted
DSPE 1,2-distearoyl-sn-glycero-3-phospho-ethanolamine-N-[methoxy-
(poly-ethylene glycol)-2000] (ammonium salt)
Eem energy of emitted photons
Eabs energy of absorbed photons
Er3+ trivalent erbium ions
ESA excited state absorption
ET energy transfer
ETU energy transfer upconversion
FWHM full width at half maximum
Gaussexp experimentally realized Gaussian beam profile
Ho3+ trivalent holmium ions
Isample(
,P) P-dependent spectrally-corrected sample spectrum
Iblank(
,P) P-dependent spectrally-corrected blank spectrum
I(
,P) P-dependent spectral photon flux
I
(P) I(
,P) for certain UC emission band or color
I
all(P) I(
,P) including all UC emission bands
Irel(
P) I(
,P) relative to I
all(P)
Irel,
(P) I(
,P) for certain UC emission band or color relative to I
all(P)
Igreen/Ired(P) green-to-red intensity ratio
ISS integrating sphere setup
Ln3+ trivalent lanthanide ions
Nd3+ trivalent neodymium ions
NIR near-infrared spectral region
II
n(P) P-dependent slope factor
Nem number of emitted photons
Nabs number of absorbed photons
OA Oleic acid, ligand molecules for UCNPs
P excitation power density
Pbalance balancing excitation power density
PBP excitation power for certain beam profile
PL photoluminescence
Re3+ trivalent rare earth ions
Si-CCD silicon-based charged-coupled-device
SHG second harmonic generation
TEM transmission electron microscopy
THexp experimentally realized Top Hat beam profile
THideal ideal Top Hat beam profile
Tm3+ trivalent thulium ions
TPA two-photon absorption
TTA triplet-triplet annihilation
UC upconversion
UCL upconversion luminescence
UCNPs upconversion nm-sized particles
UCµP upconversion µm-sized particles
UV ultraviolet spectral region
vis visible spectral region
Yb3+ trivalent ytterbium ion
-NaYF4 sodium yttrium tetraflouride in the cubic crystal phase
-NaYF4 sodium yttrium tetraflouride in the hexagonal crystal phase
photoluminescence quantum yield
UC upconversion photoluminescence quantum yield
UC(P) P-dependent
UC
UC,sat saturated
UC
UC,max maximum
UC
UC,green(P) P-dependent
UC of Er3+ green emission band
UC,red(P) P-dependent
UC of Er3+ red emission band
UC,purple(P) P-dependent
UC of Er3+ purple emission band
UC,vis(P) P-dependent
UC for visible spectral region
III
Bibliography
1. Joshi, B. P.; Wang, T. D., Targeted Optical Imaging Agents in Cancer: Focus on Clinical
Applications. Contrast Media Mol Imaging 2018, 2018, 2015237-2015237.
2. Wilhelm, S.; Tavares, A. J.; Dai, Q.; Ohta, S.; Audet, J.; Dvorak, H. F.; Chan, W. C. W., Analysis
of nanoparticle delivery to tumours. Nature Reviews Materials 2016, 1, 16014.
3. Dang, X.; Bardhan, N. M.; Qi, J.; Gu, L.; Eze, N. A.; Lin, C.-W.; Kataria, S.; Hammond, P. T.;
Belcher, A. M., Deep-tissue optical imaging of near cellular-sized features. Scientific Reports 2019, 9
(1), 3873.
4. Gerwing, M.; Herrmann, K.; Helfen, A.; Schliemann, C.; Berdel, W. E.; Eisenblätter, M.;
Wildgruber, M., The beginning of the end for conventional RECIST — novel therapies require novel
imaging approaches. Nature Reviews Clinical Oncology 2019, 16 (7), 442-458.
5. Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke, R.; Nann, T., Quantum dots
versus organic dyes as fluorescent labels. Nature Methods 2008, 5 (9), 763-775.
6. Medintz, I. L.; Uyeda, H. T.; Goldman, E. R.; Mattoussi, H., Quantum dot bioconjugates for
imaging, labelling and sensing. Nature Materials 2005, 4 (6), 435-446.
7. Petryayeva, E.; Algar, W. R.; Medintz, I. L., Quantum Dots in Bioanalysis: A Review of
Applications across Various Platforms for Fluorescence Spectroscopy and Imaging. Applied
Spectroscopy 2013, 67 (3), 215-252.
8. Mason, W. T., Fluorescent and luminescent probes for biological activity : a practical guide to
technology for quantitative real-time analysis. 2nd ed. ed.; San Diego (Calif.) : Academic press: 1999.
9. Ekimov, A.; Onushchehko, A., Quantum Size Effect in Three-Dimensional Microscopic
Semiconductor Crystals. ZhETF Pis ma Redaktsiiu 1981, 34, 363.
10. Yoffe, A. D., Low-dimensional systems: Quantum size effects and electronic properties of
semiconductor microcrystallites (zero-dimensional systems) and some quasi-two-dimensional
systems. Advances in Physics 2002, 51 (2), 799-890.
11. Würth, C.; Geißler, D.; Behnke, T.; Kaiser, M.; Resch-Genger, U., Critical review of the
determination of photoluminescence quantum yields of luminescent reporters. Analytical and
Bioanalytical Chemistry 2014, 1-20.
12. Hatami, S.; Würth, C.; Kaiser, M.; Leubner, S.; Gabriel, S.; Bahrig, L.; Lesnyak, V.; Pauli, J.;
Gaponik, N.; Eychmuller, A.; Resch-Genger, U., Absolute photoluminescence quantum yields of IR26
and IR-emissive Cd1-xHgxTe and PbS quantum dots - method- and material-inherent challenges.
Nanoscale 2015, 7 (1), 133-143.
13. Hardman, R., A toxicologic review of quantum dots: toxicity depends on physicochemical and
environmental factors. Environ Health Perspect 2006, 114 (2), 165-172.
14. Nikazar, S.; Sivasankarapillai, V. S.; Rahdar, A.; Gasmi, S.; Anumol, P. S.; Shanavas, M. S.,
Revisiting the cytotoxicity of quantum dots: an in-depth overview. Biophys Rev 2020, 12 (3), 703-718.
15. Mader, H. S.; Kele, P.; Saleh, S. M.; Wolfbeis, O. S., Upconverting luminescent nanoparticles
for use in bioconjugation and bioimaging. Current Opinion in Chemical Biology 2010, 14 (5), 582-596.
16. Chen, G. Y.; Agren, H.; Ohulchanskyy, T. Y.; Prasad, P. N., Light upconverting core-shell
nanostructures: nanophotonic control for emerging applications. Chemical Society Reviews 2015, 44
(6), 1680-1713.
17. Resch-Genger, U.; Gorris, H. H., Perspectives and challenges of photon-upconversion
nanoparticles - Part I: routes to brighter particles and quantitative spectroscopic studies. Analytical
and Bioanalytical Chemistry 2017, 409 (25), 5855-5874
18. Jaque, D.; Vetrone, F., Luminescence nanothermometry. Nanoscale 2012, 4 (15), 4301-4326.
19. Zhou, B.; Shi, B. Y.; Jin, D. Y.; Liu, X. G., Controlling upconversion nanocrystals for emerging
applications. Nature Nanotechnology 2015, 10 (11), 924-936.
IV
20. Chan, E. M., Combinatorial approaches for developing upconverting n anomaterials: high-
throughput screening, modeling, and applications. Chemical Society Reviews 2015, 44 (6), 1653-1679.
21. Haase, M.; Schafer, H., Upconverting Nanoparticles. Angewandte Chemie-International
Edition 2011, 50 (26), 5808-5829.
22. Wu, X.; Chen, G. Y.; Shen, J.; Li, Z. J.; Zhang, Y. W.; Han, G., Upconversion Nanoparticles: A
Versatile Solution to Multiscale Biological Imaging. Bioconjugate Chemistry 2015, 26 (2), 166-175.
23. Xu, C. T.; Zhan, Q. Q.; Liu, H. C.; Somesfalean, G.; Qian, J.; He, S. L.; Andersson-Engels, S.,
Upconverting nanoparticles for pre-clinical diffuse optical imaging, microscopy and sensing: Current
trends and future challenges. Laser & Photonics Reviews 2013, 7 (5), 663-697.
24. Wang, H. Q.; Batentschuk, M.; Osvet, A.; Pinna, L.; Brabec, C. J., Rare-Earth Ion Doped Up-
Conversion Materials for Photovoltaic Applications. Advanced Materials 2011, 23 (22-23), 2675-2680.
25. Gnach, A.; Bednarkiewicz, A., Lanthanide-doped up-converting nanoparticles: Merits and
challenges. Nano Today 2012, 7 (6), 532-563.
26. Wang, F.; Liu, X. G., Recent advances in the chemistry of lanthanide-doped upconversion
nanocrystals. Chemical Society Reviews 2009, 38 (4), 976-989.
27. Gorris, H. H.; Wolfbeis, O. S., Photon-Upconverting Nanoparticles for Optical Encoding and
Multiplexing of Cells, Biomolecules, and Microspheres. Angewandte Chemie-International Edition
2013, 52 (13), 3584-3600.
28. Zheng, W.; Huang, P.; Tu, D.; Ma, E.; Zhu, H.; Chen, X., Lanthanide-doped upconversion
nano-bioprobes: electronic structures, optical properties, and biodetection. Chem Soc Rev 2015, 44
(6), 1379-415.
29. Zhou, J.; Liu, Q.; Feng, W.; Sun, Y.; Li, F., Upconversion Luminescent Materials: Advances
and Applications. Chemical Reviews 2015, 115 (1), 395-465.
30. Auzel, F., Upconversion and anti-stokes processes with f and d ions in solids. Chemical
Reviews 2004, 104 (1), 139-173.
31. Han, S.; Deng, R.; Xie, X.; Liu, X., Enhancing Luminescence in Lanthanide-Doped
Upconversion Nanoparticles. Angewandte Chemie International Edition 2014, 53 (44), 11702-11715.
32. Peng, X.; Huang, B.; Pu, R.; Liu, H.; Zhang, T.; Widengren, J.; Zhan, Q.; Ågren, H., Fast
upconversion super-resolution microscopy with 10 μs per pixel dwell times. Nanoscale 2019, 11 (4),
1563-1569.
33. Chen, C.; Wang, F.; Wen, S.; Su, Q. P.; Wu, M. C. L.; Liu, Y.; Wang, B.; Li, D.; Shan, X.;
Kianinia, M.; Aharonovich, I.; Toth, M.; Jackson, S. P.; Xi, P.; Jin, D., Multi-photon near-infrared
emission saturation nanoscopy using upconversion nanoparticles. Nature Communications 2018, 9
(1), 3290.
34. Chen, S.; Weitemier, A. Z.; Zeng, X.; He, L.; Wang, X.; Tao, Y.; Huang, A. J. Y.; Hashimotodani,
Y.; Kano, M.; Iwasaki, H.; Parajuli, L. K.; Okabe, S.; Teh, D. B. L.; All, A. H.; Tsutsui-Kimura, I.; Tanaka,
K. F.; Liu, X.; McHugh, T. J., Near-infrared deep brain stimulation via upconversion nanoparticle–
mediated optogenetics. Science (New York, N.Y.) 2018, 359 (6376), 679-684.
35. Fernandez-Bravo, A.; Wang, D.; Barnard, E. S.; Teitelboim, A.; Tajon, C.; Guan, J.; Schatz,
G. C.; Cohen, B. E.; Chan, E. M.; Schuck, P. J.; Odom, T. W., Ultralow-threshold, continuous-wave
upconverting lasing from subwavelength plasmons. Nature Materials 2019, 18 (11), 1172-1176.
36. Wang, F.; Wen, S.; He, H.; Wang, B.; Zhou, Z.; Shimoni, O.; Jin, D., Microscopic inspection
and tracking of single upconversion nanoparticles in living cells. Light: Science &Amp; Applications
2018, 7.
37. Zhou, M.; Ge, X.; Ke, D.-M.; Tang, H.; Zhang, J.-Z.; Calvaresi, M.; Gao, B.; Sun, L.; Su, Q.;
Wang, H., The Bioavailability, Biodistribution, and Toxic Effects of Silica-Coated Upconversion
Nanoparticles in vivo. Front Chem 2019, 7, 218-218.
38. Gnach, A.; Lipinski, T.; Bednarkiewicz, A.; Rybka, J.; Capobianco, J. A., Upconverting
nanoparticles: assessing the toxicity. Chemical Society Reviews 2015, 44 (6), 1561-1584.
V
39. Homann, C.; Krukewitt, L.; Frenzel, F.; Grauel, B.; Würth, C.; Resch-Genger, U.; Haase, M.,
NaYF4:Yb,Er/NaYF4 Core/Shell Nanocrystals with High Upconversion Luminescence Quantum Yield.
Angewandte Chemie International Edition 2018, 57 (28), 8765-8769.
40. Kaiser, M.; Würth, C.; Kraft, M.; Hyppanen, I.; Soukka, T.; Resch-Genger, U., Power-
dependent upconversion quantum yield of NaYF4:Yb3+,Er3+ nano- and micrometer-sized particles -
measurements and simulations. Nanoscale 2017, 9 (28), 10051-10058.
41. Würth, C.; Fischer, S.; Grauel, B.; Alivisatos, A. P.; Resch-Genger, U., Quantum Yields, Surface
Quenching, and Passivation Efficiency for Ultrasmall Core/Shell Upconverting Nanoparticles. Journal
of the American Chemical Society 2018, 140 (14), 4922-4928.
42. Arppe, R.; Hyppänen, I.; Perala, N.; Peltomaa, R.; Kaiser, M.; Würth, C.; Christ, S.; Resch-
Genger, U.; Schäferling, M.; Soukka, T., Quenching of the upconversion luminescence of
NaYF4:Yb3+,Er3+ and NaYF4:Yb3+,Tm3+ nanophosphors by water: the role of the sensitizer Yb3+ in non-
radiative relaxation. Nanoscale 2015, 7 (27), 11746-11757.
43. Lyu, L.; Cheong, H.; Ai, X.; Zhang, W.; Li, J.; Yang, H.; Lin, J.; Xing, B., Near-infrared light-
mediated rare-earth nanocrystals: recent advances in improving photon conversion and alleviating
the thermal effect. NPG Asia Materials 2018, 10 (8), 685-702.
44. Wiesholler, L. M.; Frenzel, F.; Grauel, B.; Wurth, C.; Resch-Genger, U.; Hirsch, T., Yb,Nd,Er-
doped upconversion nanoparticles: 980 nm versus 808 nm excitation. Nanoscale 2019, 11 (28),
13440-13449.
45. Wang, X.; Valiev, R. R.; Ohulchanskyy, T. Y.; Ågren, H.; Yang, C.; Chen, G., Dye-sensitized
lanthanide-doped upconversion nanoparticles. Chemical Society Reviews 2017, 46 (14), 4150-4167.
46. Wen, S.; Zhou, J.; Schuck, P. J.; Suh, Y. D.; Schmidt, T. W.; Jin, D., Future and challenges for
hybrid upconversion nanosystems. Nature Photonics 2019, 13 (12), 828-838.
47. Das, A.; Mao, C.; Cho, S.; Kim, K.; Park, W., Over 1000-fold enhancement of upconversion
luminescence using water-dispersible metal-insulator-metal nanostructures. Nature
Communications 2018, 9 (1), 4828.
48. Dong, J.; Gao, W.; Han, Q.; Wang, Y.; Qi, J.; Yan, X.; Sun, M., Plasmon-enhanced
upconversion photoluminescence: Mechanism and application. Reviews in Physics 2019, 4, 100026.
49. Park, W.; Lu, D.; Ahn, S., Plasmon enhancement of luminescence upconversion. Chemical
Society Reviews 2015, 44 (10), 2940-2962.
50. Fischer, S.; Johnson, N. J. J.; Pichaandi, J.; Goldschmidt, J. C.; van Veggel, F. C. J. M.,
Upconverting core-shell nanocrystals with high quantum yield under low irradiance: On the role of
isotropic and thick shells. Journal of Applied Physics 2015, 118 (19), 193105.
51. Boyer, J. C.; van Veggel, F., Absolute quantum yield measurements of colloidal NaYF4: Er3+,
Yb3+ upconverting nanoparticles. Nanoscale 2010, 2 (8), 1417-1419.
52. Page, R. H.; Schaffers, K. I.; Waide, P. A.; Tassano, J. B.; Payne, S. A.; Krupke, W. F.; Bischel,
W. K., Upconversion-pumped luminescence efficiency of rare-earth-doped hosts sensitized with
trivalent ytterbium. Journal of the Optical Society of America B-Optical Physics 1998, 15 (3), 996-1008.
53. Fischer, S.; Martin-Rodriguez, R.; Frohlich, B.; Kramer, K. W.; Meijerink, A.; Goldschmidt, J.
C., Upconversion quantum yield of Er3+-doped -NaYF4 and Gd2O2S: The effects of host lattice, Er3+
doping, and excitation spectrum bandwidth. Journal of Luminescence 2014, 153, 281-287.
54. Pokhrel, M.; Kumar, G. A.; Sardar, D. K., Highly efficient NIR to NIR and VIS upconversion in
Er3+ and Yb3+ doped in M2O2S (M = Gd, La, Y). Journal of Materials Chemistry A 2013, 1 (38), 11595-
11606.
55. Würth, C.; Kaiser, M.; Wilhelm, S.; Grauel, B.; Hirsch, T.; Resch-Genger, U., Excitation power
dependent population pathways and absolute quantum yields of upconversion nanoparticles in
different solvents. Nanoscale 2017, 9 (12), 4283-4294.
56. Lakowicz, J., Principles of Fluorescence Spectroscopy. 2006; 3rd Edition.
VI
57. Bloembergen, N., Solid State Infrared Quantum Counters. Physical Review Letters 1959, 2 (3),
84-85.
58. Auzel, F. E., Materials and devices using double-pumped-phosphors with energy transfer.
Proceedings of the IEEE 1973, 61 (6), 758-786.
59. Nadort, A.; Zhao, J. B.; Goldys, E. M., Lanthanide upconversion luminescence at the
nanoscale: fundamentals and optical properties. Nanoscale 2016, 8 (27), 13099-13130.
60. Goldschmidt, J. C.; Fischer, S., Upconversion for Photovoltaics - a Review of Materials, Devices
and Concepts for Performance Enhancement. Advanced Optical Materials 2015, 3 (4), 510-535.
61. Strek, W.; Deren, P.; Bednarkiewicz, A., Cooperative processes in KYb(WO4)2 crystal doped
with Eu3+ and Tb3+ ions. Journal of Luminescence 2000, 87-89, 999-1001.
62. Arai, Y.; Yamashidta, T.; Suzuki, T.; Ohishi, Y., Upconversion properties of Tb3+–Yb3+ codoped
fluorophosphate glasses. Journal of Applied Physics 2009, 105 (8), 083105.
63. Hernández, I.; Pathumakanthar, N.; Wyatt, P. B.; Gillin, W. P., Cooperative Infrared to Visible
Up Conversion in Tb3+, Eu3+, and Yb3+ Containing Polymers. Advanced Materials 2010, 22 (47), 5356-
5360.
64. Teuchner, K.; Ehlert, J.; Freyer, W.; Leupold, D.; Altmeyer, P.; Stücker, M.; Hoffmann, K.,
Fluorescence Studies of Melanin by Stepwise Two-Photon Femtosecond Laser Excitation. Journal of
Fluorescence 2000, 10 (3), 275-275.
65. Helmchen, F.; Denk, W., Deep tissue two-photon microscopy. Nature Methods 2005, 2 (12),
932-940.
66. Meiling, T. T.; Cywiński, P. J.; Löhmannsröben, H.-G., Two-Photon Excitation Fluorescence
Spectroscopy of Quantum Dots: Photophysical Properties and Application in Bioassays. The Journal
of Physical Chemistry C 2018, 122 (17), 9641-9647.
67. Masada, G.; Suzudo, T.; Satoh, Y.; Ishizuki, H.; Taira, T.; Furusawa, A., Efficient generation
of highly squeezed light with periodically poled MgO:LiNbO3. Opt. Express 2010, 18 (12), 13114-
13121.
68. Gharaati, S.; Wang, C.; Förster, C.; Weigert, F.; Resch-Genger, U.; Heinze, K., Triplet–Triplet
Annihilation Upconversion in a MOF with Acceptor-Filled Channels. Chemistry – A European Journal
2020, 26 (5), 960-960.
69. Park, J.; Xu, M.; Li, F.; Zhou, H.-C., 3D Long-Range Triplet Migration in a Water-Stable Metal–
Organic Framework for Upconversion-Based Ultralow-Power in Vivo Imaging. Journal of the American
Chemical Society 2018, 140 (16), 5493-5499.
70. Kouno, H.; Sasaki, Y.; Yanai, N.; Kimizuka, N., Supramolecular Crowding Can Avoid Oxygen
Quenching of Photon Upconversion in Water. Chemistry – A European Journal 2019, 25 (24), 6038-
6038.
71. Baluschev, S.; Katta, K.; Avlasevich, Y.; Landfester, K., Annihilation upconversion in
nanoconfinement: solving the oxygen quenching problem. Materials Horizons 2016, 3 (6), 478-486.
72. Filatov, M. A.; Baluschev, S.; Landfester, K., Protection of densely populated excited triplet
state ensembles against deactivation by molecular oxygen. Chemical Society Reviews 2016, 45 (17),
4668-4689.
73. Auzel, F., Multiphonon Processes, Cross-Relaxation and Up-Conversion in Ion-Activated
Solids, Exemplified By Minilaser Materials. In Radiationless Processes, DiBartolo, B.; Goldberg, V., Eds.
Springer US: Boston, MA, 1980; pp 213-286.
74. Förster, T., Energiewanderung und Fluoreszenz. Naturwissenschaften 1946, 33 (6), 166-175.
75. Dexter, D. L., Theory of Sensitized Luminescence in Solids. Journal of Chemical Physics 1953,
21, 836-850.
76. Inokuti, M.; Hirayama, F., Influence of Energy Transfer by the Exchange Mechanism on Donor
Luminescence. The Journal of Chemical Physics 1965, 43 (6), 1978-1989.
77. Yamada, N.; Shionoya, S.; Kushida, T., Phonon-Assisted Energy Transfer between Trivalent
Rare Earth Ions. Journal of the Physical Society of Japan 1972, 32 (6), 1577-1586.
VII
78. Miyakawa, T.; Dexter, D. L., Phonon Sidebands, Multiphonon Relaxation of Excited States, and
Phonon-Assisted Energy Transfer between Ions in Solids. Physical Review B 1970, 1 (7), 2961-2969.
79. A. M. Dirac, P., The Quantum Theory of Emission and Absorption of Radiation. 1927; 114, p
243-265.
80. Mahata, M. K.; Hofsäss, H.; Vetter, U., Photon-Upconverting Materials: Advances and
Prospects for Various Emerging Applications. 2016; pp 109-131.
81. Strohhöfer, C.; Polman, A., Absorption and emission spectroscopy in Er3+–Yb3+ doped
aluminum oxide waveguides. Optical Materials 2003, 21 (4), 705-712.
82. Shen, B.; Cheng, S.; Gu, Y.; Ni, D.; Gao, Y.; Su, Q.; Feng, W.; Li, F., Revisiting the optimized
doping ratio in core/shell nanostructured upconversion particles. Nanoscale 2017, 9 (5), 1964-1971.
83. Gargas, D. J.; Chan, E. M.; Ostrowski, A. D.; Aloni, S.; Altoe, M. V. P.; Barnard, E. S.; Sanii,
B.; Urban, J. J.; Milliron, D. J.; Cohen, B. E.; Schuck, P. J., Engineering bright sub-10-nm upconverting
nanocrystals for single-molecule imaging. Nature Nanotechnology 2014, 9 (4), 300-305.
84. N. Russell, H.; A. Saunders, F., New Regularities in the Spectra of the Alkaline Earths. The
Astrophysical Journal 1924, 61, 38.
85. Walsh, B., Judd-Ofelt theory: Principles and practices. 2005; pp 403-433.
86. Görller-Walrand, C.; Binnemans, K., Chapter 155 Rationalization of crystal-field
parametrization. In Handbook on the Physics and Chemistry of Rare Earths, Elsevier: 1996; Vol. 23, pp
121-283.
87. Capobianco, J. A.; Vetrone, F.; Boyer, J. C.; Speghini, A.; Bettinelli, M., Enhancement of Red
Emission (4F9/2 → 4I15/2) via Upconversion in Bulk and Nanocrystalline Cubic Y2O3:Er3+. The Journal of
Physical Chemistry B 2002, 106 (6), 1181-1187.
88. van de Rijke, F.; Zijlmans, H.; Li, S.; Vail, T.; Raap, A. K.; Niedbala, R. S.; Tanke, H. J., Up-
converting phosphor reporters for nucleic acid microarrays. Nature Biotechnology 2001, 19 (3), 273-
276.
89. Heer, S.; Lehmann, O.; Haase, M.; Güdel, H.-U., Blue, Green, and Red Upconversion Emission
from Lanthanide-Doped LuPO4 and YbPO4 Nanocrystals in a Transparent Colloidal Solution.
Angewandte Chemie International Edition 2003, 42 (27), 3179-3182.
90. Heer, S.; Kömpe, K.; Güdel, H.-U.; Haase, M., Highly Efficient Multicolour Upconversion
Emission in Transparent Colloids of Lanthanide-Doped NaYF4 Nanocrystals. Advanced Materials 2004,
16 (23‐24), 2102-2105.
91. Wang, G.; Peng, Q.; Li, Y., Upconversion Luminescence of Monodisperse CaF2:Yb3+/Er3+
Nanocrystals. Journal of the American Chemical Society 2009, 131 (40), 14200-14201.
92. Chen, D.; Lei, L.; Zhang, R.; Yang, A.; Xu, J.; Wang, Y., Intrinsic single-band upconversion
emission in colloidal Yb/Er(Tm):Na3Zr(Hf)F7 nanocrystals. Chemical Communications 2012, 48 (86),
10630-10632.
93. Menyuk, N.; Dwight, K.; Pierce, J. W., NaYF4 : Yb,Er—an efficient upconversion phosphor.
Applied Physics Letters 1972, 21 (4), 159-161.
94. Suyver, J. F.; Grimm, J.; Krämer, K. W.; Güdel, H. U., Highly efficient near-infrared to visible
up-conversion process in NaYF4:Er3+,Yb3+. Journal of Luminescence 2005, 114 (1), 53-59.
95. Lay, A.; H. Sheppard, O.; Siefe, C.; A. McLellan, C.; D. Mehlenbacher, R.; Fischer, S.;
Goodman, M.; A. Dionne, J., Optically Robust and Biocompatible Mechanosensitive Upconverting
Nanoparticles. ACS Central Science 2019.
96. Wang, F.; Han, Y.; Lim, C. S.; Lu, Y.; Wang, J.; Xu, J.; Chen, H.; Zhang, C.; Hong, M.; Liu, X.,
Simultaneous phase and size control of upconversion nanocrystals through lanthanide doping. Nature
2010, 463, 1061.
97. Renero-Lecuna, C.; Martín-Rodríguez, R.; Valiente, R.; González, J.; Rodríguez, F.; Krämer,
K. W.; Güdel, H. U., Origin of the High Upconversion Green Luminescence Efficiency in β-
NaYF4:2%Er3+,20%Yb3+. Chemistry of Materials 2011, 23 (15), 3442-3448.
VIII
98. Klier, D. T.; Kumke, M. U., Analysing the effect of the crystal structure on upconversion
luminescence in Yb3+,Er3+-co-doped NaYF4 nanomaterials. Journal of Materials Chemistry C 2015, 3
(42), 11228-11238.
99. Riseberg, L. A.; Moos, H. W., Multiphonon Orbit-Lattice Relaxation of Excited States of Rare-
Earth Ions in Crystals. Physical Review 1968, 174 (2), 429-438.
100. Wermuth, M.; Riedener, T.; Güdel, H. U., Spectroscopy and upconversion mechanisms of
CsCdBr3Dy3+. Physical Review B 1998, 57 (8), 4369-4376.
101. Rabouw, F. T.; Prins, P. T.; Villanueva-Delgado, P.; Castelijns, M.; Geitenbeek, R. G.;
Meijerink, A., Quenching Pathways in NaYF4:Er3+,Yb3+ Upconversion Nanocrystals. ACS Nano 2018, 12
(5), 4812-4823.
102. Anderson, R. B.; Smith, S. J.; May, P. S.; Berry, M. T., Revisiting the NIR-to-Visible
Upconversion Mechanism in beta-NaYF4:Yb3+,Er3+. J Phys Chem Lett 2014, 5 (1), 36-42.
103. Berry, M. T.; May, P. S., Disputed Mechanism for NIR-to-Red Upconversion Luminescence in
NaYF4:Yb3+,Er3+. Journal of Physical Chemistry A 2015, 119 (38), 9805-9811.
104. Wang, F.; Liu, X., Upconversion Multicolor Fine-Tuning: Visible to Near-Infrared Emission from
Lanthanide-Doped NaYF4 Nanoparticles. Journal of the American Chemical Society 2008, 130 (17),
5642-5643.
105. Wang, J.; Deng, R.; MacDonald, M. A.; Chen, B.; Yuan, J.; Wang, F.; Chi, D.; Andy Hor, T.
S.; Zhang, P.; Liu, G.; Han, Y.; Liu, X., Enhancing multiphoton upconversion through energy clustering
at sublattice level. Nature Materials 2013, 13, 157.
106. Gao, D.; Zhang, X.; Chong, B.; Xiao, G.; Tian, D., Simultaneous spectra and dynamics
processes tuning of a single upconversion microtube through Yb3+ doping concentration and
excitation power. Physical Chemistry Chemical Physics 2017, 19 (6), 4288-4296.
107. Xue, X. J.; Uechi, S.; Tiwari, R. N.; Duan, Z. C.; Liao, M. S.; Yoshimura, M.; Suzuki, T.; Ohishi,
Y., Size-dependent upconversion luminescence and quenching mechanism of LiYF4: Er3+/Yb3+
nanocrystals with oleate ligand adsorbed. Opt Mater Express 2013, 3 (7).
108. Wei, W.; Zhang, Y.; Chen, R.; Goggi, J.; Ren, N.; Huang, L.; Bhakoo, K. K.; Sun, H. D.; Tan,
T. T. Y., Cross Relaxation Induced Pure Red Upconversion in Activator- and Sensitizer-Rich Lanthanide
Nanoparticles. Chemistry of Materials 2014, 26 (18), 5183-5186.
109. Wang, Y.; Liu, K.; Liu, X. M.; Dohnalova, K.; Gregorkiewicz, T.; Kong, X. G.; Aalders, M. C.
G.; Buma, W. J.; Zhang, H., Critical Shell Thickness of Core/Shell Upconversion Luminescence
Nanoplatform for FRET Application. J Phys Chem Lett 2011, 2 (17), 2083-2088.
110. Vetrone, F.; Naccache, R.; Mahalingam, V.; Morgan, C. G.; Capobianco, J. A., The Active-
Core/Active-Shell Approach: A Strategy to Enhance the Upconversion Luminescence in Lanthanide-
Doped Nanoparticles. Advanced Functional Materials 2009, 19 (18), 2924-2929.
111. Liu, G. K., Advances in the theoretical understanding of photon upconversion in rare-earth
activated nanophosphors. Chemical Society Reviews 2015, 44 (6), 1635-1652.
112. Bogdan, N.; Vetrone, F.; Ozin, G. A.; Capobianco, J. A., Synthesis of Ligand-Free Colloidally
Stable Water Dispersible Brightly Luminescent Lanthanide-Doped Upconverting Nanoparticles. Nano
Letters 2011, 11 (2), 835-840.
113. Liu, L.; Jiang, H. L.; Chen, Y. J.; Zhang, X. L.; Zhang, Z. G.; Wang, Y. X., Power dependence of
upconversion luminescence of Er3+ doped Yttria nanocrystals and their bulk counterpart. Journal of
Luminescence 2013, 143, 423-431.
114. Wu, S.; Ning, Y.; Chang, J.; Zhang, S., Upconversion photoluminescence enhancement and
modulation of NaYF4:Yb, Er through using different ligands. Journal of Luminescence 2013, 143 (0),
492-497.
115. Würth, C.; Grabolle, M.; Pauli, J.; Spieles, M.; Resch-Genger, U., Relative and absolute
determination of fluorescence quantum yields of transparent samples. Nat Protoc 2013, 8 (8), 1535-
1550.
IX
116. Würth, C.; Grabolle, M.; Pauli, J.; Spieles, M.; Resch-Genger, U., Comparison of Methods
and Achievable Uncertainties for the Relative and Absolute Measurement of Photoluminescence
Quantum Yields. Analytical Chemistry 2011, 83, 3431–3439.
117. Tikhomirov, V. K.; Adamo, G.; Nikolaenko, A. E.; Rodriguez, V. D.; Gredin, P.; Mortier, M.;
Zheludev, N. I.; Moshchalkov, V. V., Cathodo- and photoluminescence in Yb3+-Er3+ co-doped PbF2
nanoparticles. Opt. Express 2010, 18 (9), 8836-8846.
118. Pollnau, M.; Gamelin, D. R.; Luthi, S. R.; Gudel, H. U.; Hehlen, M. P., Power dependence of
upconversion luminescence in lanthanide and transition-metal-ion systems. Physical Review B 2000,
61 (5), 3337-3346.
119. Liu, H. C.; Xu, C. T.; Lindgren, D.; Xie, H. Y.; Thomas, D.; Gundlach, C.; Andersson-Engels, S.,
Balancing power density based quantum yield characterization of upconverting nanoparticles for
arbitrary excitation intensities. Nanoscale 2013, 5 (11), 4770-4775.
120. Kraft, M.; Würth, C.; Muhr, V.; Hirsch, T.; Resch-Genger, U., Particle-size-dependent
upconversion luminescence of NaYF4: Yb, Er nanoparticles in organic solvents and water at different
excitation power densities. Nano Research 2018, 11 (12), 6360-6374.
121. Pilch, A.; Würth, C.; Kaiser, M.; Wawrzyńczyk, D.; Kurnatowska, M.; Arabasz, S.; Prorok, K.;
Samoć, M.; Strek, W.; Resch-Genger, U.; Bednarkiewicz, A., Shaping Luminescent Properties of Yb3+
and Ho3+ Co-Doped Upconverting Core–Shell β-NaYF4 Nanoparticles by Dopant Distribution and
Spacing. Small 2017, 13 (47), 1701635.
122. Liu, H.; Huang, K.; Valiev, R. R.; Zhan, Q.; Zhang, Y.; Ågren, H., Photon Upconversion Kinetic
Nanosystems and Their Optical Response. Laser & Photonics Reviews 2018, 12 (1), 1700144.
123. Hossan, M. Y.; Hor, A.; Luu, Q.; Smith, S.; May, S.; Mary, B., Explaining the Nanoscale Effect
in the Upconversion Dynamics of β-NaYF4 :Yb3+ ,Er3+ Core and Core-Shell Nanocrystals. 2017; 121.
124. Gai, S. L.; Li, C. X.; Yang, P. P.; Lin, J., Recent Progress in Rare Earth Micro/Nanocrystals: Soft
Chemical Synthesis, Luminescent Properties, and Biomedical Applications. Chemical Reviews 2014,
114 (4), 2343-2389.
125. Shen, J.; Chen, G.; Ohulchanskyy, T. Y.; Kesseli, S. J.; Buchholz, S.; Li, Z.; Prasad, P. N.; Han,
G., Tunable Near Infrared to Ultraviolet Upconversion Luminescence Enhancement in (α-
NaYF4:Yb,Tm)/CaF2 Core/Shell Nanoparticles for In situ Real-time Recorded Biocompatible
Photoactivation. Small 2013, 9 (19), 3213-3217.
126. Chen, G.; Shen, J.; Ohulchanskyy, T. Y.; Patel, N. J.; Kutikov, A.; Li, Z.; Song, J.; Pandey, R.
K.; Ågren, H.; Prasad, P. N.; Han, G., (α-NaYbF4:Tm3+)/CaF2 Core/Shell Nanoparticles with Efficient
Near-Infrared to Near-Infrared Upconversion for High-Contrast Deep Tissue Bioimaging. ACS Nano
2012, 6 (9), 8280-8287.
127. Chen, G.; Ohulchanskyy, T. Y.; Kachynski, A.; Ågren, H.; Prasad, P. N., Intense Visible and
Near-Infrared Upconversion Photoluminescence in Colloidal LiYF4:Er3+ Nanocrystals under Excitation
at 1490 nm. ACS Nano 2011, 5 (6), 4981-4986.
128. D.S. Yasyrkina, S. V. K., A.V. Ryabova, D.V. Pominova,V.V. Voronov, R.P. Ermakov, P.P.
Fedorov, Dependence of quantum yield of up-conversion luminecence on the composition of flourite
- typesolid solution NaY1-x-yYbxEryF4. NANOSYSTEMS: PHYSICS, CHEMISTRY, MATHEMATICS 2013, 4
(5), 648–656.
129. Liu, Q.; Sun, Y.; Yang, T.; Feng, W.; Li, C.; Li, F., Sub-10 nm Hexagonal Lanthanide-Doped
NaLuF4 Upconversion Nanocrystals for Sensitive Bioimaging in Vivo. Journal of the American Chemical
Society 2011, 133 (43), 17122-17125.
130. Xu, C. T.; Svenmarker, P.; Liu, H.; Wu, X.; Messing, M. E.; Wallenberg, L. R.; Andersson-
Engels, S., High-Resolution Fluorescence Diffuse Optical Tomography Developed with Nonlinear
Upconverting Nanoparticles. ACS Nano 2012, 6 (6), 4788-4795.
131. Li, X.; Wang, R.; Zhang, F.; Zhao, D., Engineering Homogeneous Doping in Single Nanoparticle
To Enhance Upconversion Efficiency. Nano Letters 2014, 14 (6), 3634-3639.
X
132. Fischer, S.; Bronstein, N. D.; Swabeck, J. K.; Chan, E. M.; Alivisatos, A. P., Precise Tuning of
Surface Quenching for Luminescence Enhancement in Core–Shell Lanthanide-Doped Nanocrystals.
Nano Letters 2016, 16 (11), 7241-7247.
133. Wang, F.; Wang, J. A.; Liu, X. G., Direct Evidence of a Surface Quenching Effect on Size-
Dependent Luminescence of Upconversion Nanoparticles. Angewandte Chemie-International Edition
2010, 49 (41), 7456-7460.
134. Naccache, R.; Vetrone, F.; Mahalingam, V.; Cuccia, L. A.; Capobianco, J. A., Controlled
Synthesis and Water Dispersibility of Hexagonal Phase NaGdF4:Ho3+/Yb3+ Nanoparticles. Chemistry of
Materials 2009, 21 (4), 717-723.
135. Sudarsan, V.; van Veggel, F.; Herring, R. A.; Raudsepp, M., Surface Eu3+ ions are different
than "bulk" Eu3+ ions in crystalline doped LaF3 nanoparticles. Journal of Materials Chemistry 2005, 15
(13), 1332-1342.
136. Wang, Y.; Smolarek, S.; Kong, X. G.; Buma, W. J.; Brouwer, A. M.; Zhang, H., Effect of Surface
Related Organic Vibrational Modes in Luminescent Upconversion Dynamics of Rare Earth Ions Doped
Nanoparticles. J Nanosci Nanotechno 2010, 10 (11), 7149-7153.
137. Boyer, J. C.; Manseau, M. P.; Murray, J. I.; van Veggel, F., Surface Modification of
Upconverting NaYF4 Nanoparticles with PEG-Phosphate Ligands for NIR (800 nm) Biolabeling within
the Biological Window. Langmuir 2010, 26 (2), 1157-1164.
138. Wilhelm, S.; Kaiser, M.; Würth, C.; Heiland, J.; Carrillo-Carrion, C.; Muhr, V.; Wolfbeis, O.
S.; Parak, W. J.; Resch-Genger, U.; Hirsch, T., Water dispersible upconverting nanoparticles: effects
of surface modification on their luminescence and colloidal stability. Nanoscale 2015, 7 (4), 1403-
1410.
139. Quintanilla, M.; Cantarelli, I. X.; Pedroni, M.; Speghini, A.; Vetrone, F., Intense ultraviolet
upconversion in water dispersible SrF2: Tm3+, Yb3+ nanoparticles: the effect of the environment on
light emissions. Journal of Materials Chemistry C 2015, 3 (13), 3108-3113.
140. Arppe, R.; Nareoja, T.; Nylund, S.; Mattsson, L.; Koho, S.; Rosenholm, J. M.; Soukka, T.;
Schaferling, M., Photon upconversion sensitized nanoprobes for sensing and imaging of pH.
Nanoscale 2014, 6 (12), 6837-6843.
141. Mai, H. X.; Zhang, Y. W.; Sun, L. D.; Yan, C. H., Size- and phase-controlled synthesis of
monodisperse NaYF4 : Yb,Er nanocrystals from a unique delayed nucleation pathway monitored with
upconversion spectroscopy. J Phys Chem C 2007, 111 (37), 13730-13739.
142. Xu, D.; Liu, C.; Yan, J.; Yang, S.; Zhang, Y., Understanding Energy Transfer Mechanisms for
Tunable Emission of Yb3+-Er3+ Codoped GdF3 Nanoparticles: Concentration-Dependent Luminescence
by Near-Infrared and Violet Excitation. The Journal of Physical Chemistry C 2015, 119 (12), 6852-
6860.
143. Liao, J. S.; Nie, L. L.; Liu, S. H.; Liu, B.; Wen, H. R., Yb3+ concentration dependence of
upconversion luminescence in Y2Sn2O7:Yb3+/Er3+ nanophosphors. Journal of Materials Science 2014,
49 (17), 6081-6086.
144. Zhao, J.; Sun, Y.; Kong, X.; Tian, L.; Wang, Y.; Tu, L.; Zhao, J.; Zhang, H., Controlled Synthesis,
Formation Mechanism, and Great Enhancement of Red Upconversion Luminescence of NaYF4:Yb3+,
Er3+ Nanocrystals/Submicroplates at Low Doping Level. The Journal of Physical Chemistry B 2008, 112
(49), 15666-15672.
145. Zhu, H.; Lin, M.; Jin, G.; Lu, T. J.; Xu, F., A modified energy transfer model for determination
of upconversion emission of β-NaYF4:Yb,Er: Role of self-quenching effect. Journal of Luminescence
2017, 185, 292-297.
146. Li, D.; Qin, W.; Zhao, D.; Aidilibike, T.; Chen, H.; Liu, S.; Zhang, P.; Wang, L., Tunable green
to red upconversion fluorescence of water-soluble hexagonal-phase core-shell CaF2@NaYF4
nanocrystals. Opt Mater Express 2016, 6 (1), 270-278.
XI
147. Soukka, T.; Kuningas, K.; Rantanen, T.; Haaslahti, V.; Lovgren, T., Photochemical
characterization of up-converting inorganic lanthanide phosphors as potential labels. Journal of
Fluorescence 2005, 15 (4), 513-528.
148. Ylihärsilä, M.; Harju, E.; Arppe, R.; Hattara, L.; Hölsä, J.; Saviranta, P.; Soukka, T.; Waris,
M., Genotyping of clinically relevant human adenoviruses by array-in-well hybridization assay. Clinical
Microbiology and Infection 2013, 19 (6), 551-557.
149. Li, Z.; Zhang, Y.; Jiang, S., Multicolor Core/Shell-Structured Upconversion Fluorescent
Nanoparticles. Advanced Materials 2008, 20 (24), 4765-4769.
150. Würth, C.; Pauli, J.; Lochmann, C.; Spieles, M.; Resch-Genger, U., Integrating Sphere Setup
for the Traceable Measurement of Absolute Photoluminescence Quantum Yields in the Near Infrared.
Analytical Chemistry 2012, 84 (3), 1345-1352.
151. Resch-Genger, U.; Bremser, W.; Pfeifer, D.; Spieles, M.; Hoffmann, A.; DeRose, P. C.;
Zwinkels, J. C.; Gauthier, F.; Ebert, B.; Taubert, R. D.; Voigt, J.; Hollandt, J.; Macdonald, R., State-of-
the Art Comparability of Corrected Emission Spectra. 2. Field Laboratory Assessment of Calibration
Performance Using Spectral Fluorescence Standards. Analytical Chemistry 2012, 84 (9), 3899-3907.
152. Bergstrand, J.; Liu, Q.; Huang, B.; Peng, X.; Würth, C.; Resch-Genger, U.; Zhan, Q.;
Widengren, J.; Ågren, H.; Liu, H., On the decay time of upconversion luminescence. Nanoscale 2019,
11 (11), 4959-4969.
153. MacDougall, S. K. W.; Ivaturi, A.; Marques-Hueso, J.; Richards, B. S., Measurement procedure
for absolute broadband infrared up-conversion photoluminescent quantum yields: correcting for
absorption/re-emission. Rev Sci Instrum 2014, 85 (6), 063109.
154. C., d. M. J.; Felix, W. H.; H., F. R., An improved experimental determination of external
photoluminescence quantum efficiency. Advanced Materials 1997, 9 (3), 230-232.
155. Tian, B.; Fernandez-Bravo, A.; Najafiaghdam, H.; Torquato, N. A.; Altoe, M. V. P.; Teitelboim,
A.; Tajon, C. A.; Tian, Y.; Borys, N. J.; Barnard, E. S.; Anwar, M.; Chan, E. M.; Schuck, P. J.; Cohen,
B. E., Low irradiance multiphoton imaging with alloyed lanthanide nanocrystals. Nature
Communications 2018, 9 (1), 3082.
156. Marciniak, L.; Waszniewska, K.; Bednarkiewicz, A.; Hreniak, D.; Strek, W., Sensitivity of a
Nanocrystalline Luminescent Thermometer in High and Low Excitation Density Regimes. The Journal
of Physical Chemistry C 2016, 120 (16), 8877-8882.
157. Brites, C. D. S.; Balabhadra, S.; Carlos, L. D., Lanthanide-Based Thermometers: At the Cutting-
Edge of Luminescence Thermometry. Advanced Optical Materials 2019, 7 (5), 1801239.
158. Fafara, A. Near-Infrared Spectroscopic Measurement of Tissue Temperature, ln-Vivo.
Dissertation, Mc Gill University, 2000.
159. Lappi, S. E.; Smith, B.; Franzen, S., Infrared spectra of H(2)16O, H(2)18O and D(2)O in the
liquid phase by single-pass attenuated total internal reflection spectroscopy. Spectrochimica Acta
Part A: Molecular and Biomolecular Spectroscopy 2004, 60 (11), 2611-2619.
160. Fischer, S.; Ivaturi, A.; Fröhlich, B.; Rüdiger, M.; Richter, A.; Kramer, K. W.; Richards, B. S.;
Goldschmidt, J. C., Upconverter Silicon Solar Cell Devices for Efficient Utilization of Sub-Band-Gap
Photons Under Concentrated Solar Radiation. IEEE Journal of Photovoltaics 2014, 4 (1), 183-189.
161. Zhao, J.; Lu, Z.; Yin, Y.; McRae, C.; Piper, J. A.; Dawes, J. M.; Jin, D.; Goldys, E. M.,
Upconversion luminescence with tunable lifetime in NaYF4:Yb,Er nanocrystals: role of nanocrystal
size. Nanoscale 2013, 5 (3), 944-52.
162. Hudry, D.; Busko, D.; Popescu, R.; Gerthsen, D.; Abeykoon, A. M. M.; Kubel, C.; Bergfeldt,
T.; Richards, B. S., Direct Evidence of Significant Cation Intermixing in Upconverting Core@Shell
Nanocrystals: Toward a New Crystallochemical Model. Chemistry of Materials 2017, 29 (21), 9238-
9246.
163. Duhnen, S.; Haase, M., Study on the Intermixing of Core and Shell in NaEuF4/NaGdF4
Core/Shell Nanocrystals. Chemistry of Materials 2015, 27 (24), 8375-8386.
XII
164. Kaiser, M.; Würth, C.; Kraft, M.; Soukka, T.; Resch-Genger, U., Explaining the influence of
dopant concentration and excitation power density on the luminescence and brightness of β-
NaYF4:Yb3+,Er3+ nanoparticles: Measurements and simulations. Nano Research 2019, 12 (8), 1871-
1879.
XIII
List of Publications
First-author publication related to the upconversion research field referred in the
text with capital roman numerals
P1. M. Kaiser, C. Würth, M. Kraft, I. Hyppänen, T. Soukka, U. Resch-Genger
„Power-dependent upconversion quantum yield of NaYF4: Yb3+, Er3+ nano-and
micrometer-sized particles – measurements and simulations“
Nanoscale, 2017, 9, 10051-10058
DOI: 10.1039/C7NR02449E
P2. C. Würth*, M. Kaiser*, S. Wilhelm, B. Grauel, T. Hirsch, U. Resch-Genger
„Excitation power dependent population pathways and absolute quantum
yields of upconversion nanoparticles in different solvents“
Nanoscale, 2017, 9 , 4283-4294
DOI: 10.1039/C7NR00092H
*equally contributed
P3. M. Kaiser, C. Würth, M. Kraft, T. Soukka, U. Resch-Genger
“Explaining the influence of dopant concentration and excitation power density
on the luminescence and brightness of
-NaYF4:Yb3+,Er3+ nanoparticles:
Measurements and simulations”
Nanoresearch, 2019, 12, 1871-1879
First-author publication unrelated to the upconversion research field
P4. M. Müller*, M. Kaiser*, G.M. Stachowski, U. Resch-Genger, N. Gaponik, A.
Eychmüller
”Photoluminescence Quantum Yield and Matrix-Induced Luminescence
Enhancement of Colloidal Quantum Dots Embedded in Ionic Crystals”
Chemistry of Materials, 2014, 26 (10), 3231-3237
DOI: 10.1021/cm5009043
*equally contributed
Co-authored publications related to the upconversion research field
P5. S. Wilhelm, M. Kaiser, C. Würth, J. Heiland, C. Carrillo-Carrion, V. Muhr, O.S.
Wolfbeis, W.J. Parak, U. Resch-Genger, T. Hirsch
“Water dispersible upconverting nanoparticles: effects of surface modification
on their luminescence and colloidal stability”
Nanoscale, 2015, 7 , 1403-1410
DOI: 10.1039/c4nr05954a
XIV
P6. R. Arppe, I. Hyppänen, N. Perälä, R.Peltomaa, M. Kaiser, C. Würth, S. Christ, U.
Resch-Genger, M. Schäferling, T. Soukka
“Quenching of the upconversion luminescence of NaYF4: Yb3+, Er3+ and NaYF4:
Yb3+, Tm3+ nanophosphors by water: the role of the sensitizer Yb3+ in non-
radiative relaxation”
Nanoscale, 2015, 7 , 133-143
DOI: 10.1039/C5NR02100F
P7. A. Pilch, C. Würth, M. Kaiser, D. Wawrzyńczyk, M. Kurnatowska, S. Arabasz, K.
Prorok, M. Samoć, W. Strek, U. Resch‐Genger, A. Bednarkiewicz
“Shaping Luminescent Properties of Yb3+ and Ho3+ Co‐Doped Upconverting
Core–Shell β‐NaYF4 Nanoparticles by Dopant Distribution and Spacing”
Small, 2017, 13, 1701635
DOI:10.1002/smll.201770246
Co-authored publications unrelated to upconversion research field
P8. C. Würth, D. Geißler, T. Behnke, M. Kaiser, U. Resch-Genger
”Critical review of the determination of photoluminescence quantum yields of
luminescent reporters”
Analytical and bioanalytical chemistry, 2015 , 407, 59-78
DOI: 10.1007/s00216-014-8130-z
P9. S. Hatami, C. Würth, M. Kaiser, S. Leubner, S. Gabriel, L. Bahrig, V. Lesnyak, J.
Pauli, N. Gaponik, A. Eychmüller, U. Resch-Genger
“Absolute photoluminescence quantum yields of IR26 and IR-emissive Cd1− x
HgxTe and PbS quantum dots–method-and material-inherent challenges”
Nanoscale, 2015, 7, 133-143
DOI: 10.1039/C4NR04608K
P10. M.M. Lezhnina, H. Kätker, M. Kaiser, L. Stegemann, E. Voss, U. Resch-Genger, C.
Strassert, U. Kynast
”Absolute Quantum Efficiencies and Chemical Behaviour of Lanthanide Borate
Phosphors and Glazes”
Journal of Luminescence, 2016, 70, 387-394
DOI: 10.1016/j.jlumin.2015.05.005
P11. M.R. Wagner, G. Callsen, J.S. Reparaz, J.-H. Schulze, R. Kirste, M. Cobet, I.A.
Ostapenko, S. Rodt, C. Nenstiel, M. Kaiser, A. Hoffmann, A.V. Rodina, M.R.Phillips,
S.Lautenschläger, S. Eisermann, B.K. Meyer
“Bound excitons in ZnO: Structural defect complexes versus shallow impurity
centers”
Physical Review B , 2011, 84 (3), 035313
DOI: 10.1103/PhysRevB.84.035313
XV
P12. B.K. Meyer, J. Sann, S. Eisermann, S. Lautenschlaeger, M.R. Wagner, M Kaiser, G.
Callsen, J.S. Reparaz, A. Hoffmann
”Excited state properties of donor bound excitons in ZnO”
Physical Review B, 2010 , 82 (11), 115207
DOI: 10.1103/PhysRevB.82.115207
P13. G. Durkaya, M. Bügler, R. Atalay, I. Senevirathna, M. Alevli, O. Hitzemann,
M. Kaiser, R. Kirste, A. Hoffmann, N. Dietz
“The influence of the group V/III molar precursor ratio on the structural
properties of InGaN layers grown by HPCVD”
physica status solidi (a) , 2010, 207 (6), 1379-1382
DOI: 10.1002/pssa.200983622
P14. G.Durkaya, M.Alevli, M.Bügler, R.Atalay, S Gamage, M.Kaiser, R.Kirste,
A.Hoffmann, M Jamil, I Ferguson, N.Dietz
“Growth temperature-phase stability relation in In1-xGaxN epilayers grown by
high-pressure CVD”
MRS Proceedings, 2009, 1202, 1202-I05-21
DOI: 10.1557/PROC-1202-I05-21
XVI
Conference Talks and Posters
This list concludes all first-author conference talks and poster presentations.
C1. M. Kaiser, C. Würth, I. Hyppänen, E. Palo, T. Soukka, U. Resch-Genger “Optical
conversion efficiency of up-conversion nanoparticles as new class of
luminescent reporters” ICL 2014, Wroclaw, Poland, 2014
C2. M. Kaiser, C. Würth, I. Hypännen, E. Palo, T. Soukka, U. Resch-Genger
“Upconversion quantum yields of rare earth doped nanoparticles dependent on
dopant concentration” 78th DPG Spring Meeting 2014, Dresden, Germany, 2014
C3. M Kaiser, C. Würth, U. Resch-Genger, I. Hyppänen, T. Soukka “Integration Sphere
Setup for the Absolute Determination of Upconversion Quantum Yields of
Lanthanide Doped Nanoparticles” iNOW2013 International Nano
Optoelectronics Workshop, Cargese, France, 2013
C4. M. Kaiser, S. Wilhelm, C. Würth, J. Heiland, O.S. Wolfbeis, T.Hirsch, U.Resch-
Genger “Luminescence Properties of Upconverting NaYF4(Yb,Er) Nanoparticles
in water and heavy water” 13th Conference on Methods and Applications of
Fluorescence (MAF 13), Genoa, Italy, 2013
C5. M. Kaiser, C.Würth, U.Resch-Genger, I.Hyppänen, T.Soukka “Upconversion
nanoparticles as new class of fluorescent reporters – Tools to characterize their
signal-relevant optical properties” Biosensor Symposium, Berlin, Germany, 2013
C6. M. Kaiser, C. Würth, U. Resch-Genger, I. Hyppänen, T. Soukka “Tools for the
characterization of the signal-relevant properties of upconversion nanoparticles
as new class of fluorescent reporters” ANAKON, Essen, Germany, 2013
C7. M. Kaiser, J. Heiland, M. Kraft, C. Würth, S. Wilhelm,V. Muhr, N. Leibl, O. S.
Wolfbeis, T. Hirsch, U. Resch-Genger “Optical characterization of lanthanide
doped up-converting nanoparticles” BAM Wissensbörse, Berlin, Germany, 2013
C8. M. Kaiser, C. Würth, U. Resch-Genger, I. Hyppänen, T. Soukka “Absolute
Photoluminescence Quantum Yield of Hexagonal NaYF4:Er3+,Yb3+ Upconversion
Nanoparticles” 77th DPG Spring Meeting 2013, Regensburg, Germany, 2013
C9. M. Kaiser, C. Würth, M. Vorsthove, T. Felbeck, U. Kynast, U. Resch-Genger
“Luminescence Properties of Cer-dotierten Yttrium-Aluminium-Garnet (YAG:Ce)
Nanoparticles - Absolute Quantum Yields and Influence of Particle Size” Nano-
Additive, Berlin, Germany, 2012
XVII
C10. M. Kaiser, C. Würth, T. Felbeck, M. Vorsthove, U. Kynast, U. Resch-Genger
“Influence of the Particle Size on the Optical Properties of YAG:Ce”
Photochemie Tagung, Postdam, Germany, 2012
C11. M. Kaiser, U. Resch-Genger, C. Würth, U. Kynast, T. Felbeck, M. Vorsthove
“Luminescence Properties of Cer-dotierten Yttrium-Aluminium-Garnet (YAG:Ce)
Nanoparticles” Upcore Joint Seminar 2012, Regensburg, Germany, 2012
C12. M. Kaiser, U. Resch-Genger, C. Würth, U. Kynast, T. Felbeck, M. Vorsthove
“Luminescence Properties of Cer-dotierten Yttrium-Aluminium-Garnet (YAG:Ce)
Nanoparticles - Absolute Quantum Yields and Influence of Particle Size” 76th
DPG Spring Meeting 2012 ,Berlin, Germany, 2012
C13. M. Kaiser, M. R. Wagner, G. Callsen, A. Hoffmann, S. Lautenschläger, S.
Eisermann, B. K. Meyer “Excitons and their excitation channels in a-plane and c-
plane ZnO” 74th DPG Spring Meeting 2010, Regensburg, Germany, 2010
XVIII
Acknowledgement
First of all I want to express my gratitude to Prof. Dr. Axel Hoffmann from the Technical
University of Berlin (TU - Berlin) for supervising this work. He sharpened my scientific view
on the essential scientific findings. I am very thankful for him providing me a workplace and
allowing me to regularly present my results in his workgroup meetings.
Many thanks to Dr. Ute Resch-Genger from the Federal Institute for Materials
Research and Testing (BAM) for supervising this work. I am very thankful for the great trust
in my ideas of the design and development of the UC integrating sphere setup
measurement technique. Thanks for her permanent support and motivating words, which
helped me finishing this work.
Special thanks to Dr. Christian Würth from the BAM, who contributed by discussing
the physical processes behind the UC emitters in a critical and constructive manner, which
helped to select the best fruits from a manifold of ideas. Unforgettable for me is the time
at the first upconversion meeting in Regensburg, where we also later enjoyed the beer tents
and fun fair attractions.
Great thanks to Prof. Dr. Soukka from the University of Turku for fruitful physical
discussions and a friendly cooperation. Many thanks to Dr. Emilia Palo for the synthesis of
high-quality upconversion nanoparticles with different dopant concentrations, Dr. Iko
Hypännen for fruitful discussions as well as to Dr. Rikka Arppe and Dr. Satu Lithannen for
a great time in Poland at ICL 2014 in Wroclaw.
A lot of thanks to Dr. Thomas Hirsch from the University of Regensburg and his former
workgroup members: Prof. Dr. Stefan Wilhelm for the synthesis of bright upconversion
particles in different solvents and Dr. Josef Heiland for his assistance for the
characterization of the excitation beam profiles.
Great thanks to all former and current work group members from the Biphotonics
group (BAM) of Dr. Ute Resch Genger for the good working atmosphere:
Thanks to Dr. Jutta Pauli for introducing me to the spectrometers, Dr. Katrin
Hoffmann for friendly chats, Arne Güttler for technical help, Nils Handelmann for
programming, Dr. Markus Grabolle for fruitful physical discussions and always having an
XIX
open ear and Dr. Soheil Hatami for lunch times, sharing office lab work and sharing
frustrations. A lot of thanks to M.Sc. Florian Frenzel and M.Sc. Florian Weigert for nice
conversations. Thanks to Dr. Marco Kraft for finding the reason for the seeming avalanche
effect measurement artefact, for the
UC measurements of the Yb3+, Er3+ concentration
series, and a great time in Poland at the ICL 2014. Thanks to M.Sc. Bettina Grauel for
providing me with the Matlab program for the rate equation analysis and thanks to
M.Sc. Melissa Monks for doing ICP-OES measurements. Many thanks to Dr. Daniel Geißler,
Dr. Christian Würth and M.Sc. Bettina Grauel for careful proof reading of this thesis.
Also many thanks to all former and current work group members from the group of
Prof. Dr. Hoffmann for helpful feedbacks and a nice atmosphere:
Thanks to Dr. Christian Nenstiel, who was a great help with UC lifetime measurements
with the dye laser as well as to Dipl. Phys. Dipl. Kfm. Thomas Kure and Dr. Felix Nippert for
nice chats and good advice. Great thanks to Dr. Maximilian Ries being a great help with the
last corrections of my PhD presentation. Special thanks to Dr. Max Bügler for his good
friendship and the exciting billiard matches.
A lot of thanks to my close friends David, Einar, Fabian, Jan, Loic, Max, Murat and
Rachel for always having an open ear for me, helping me in difficult situations and having
good times together.
Special thanks to my mother Petra for her great support. Huge thanks to my daughter
Sophie giving me inner strength and a high motivation to achieve my goals.
XX
Contributions to the Manuscripts
Publication I
Manuscript Title:
Power-dependent upconversion quantum yield of NaYF4:Yb3+,Er3+ nano- and micrometer-
sized particles – Measurements and simulations
List of Authors:
Martin Kaiser, Christian Würth, Marco Kraft, Iko Hyppänen, Tero Soukka, Ute Resch-Genger
Substantial contributions:
A newly custom-built integrating sphere setup was designed, built up and calibrated. I
performed all measurements including excitation power density-dependent absolute
upconversion luminescence (UCL), upconversion quantum yield (
UC), and lifetime
measurements, did the analysis of the data and created all graphics. New approaches for the
analysis of the data as P-dependent slope factor and relative contribution were developed. I
experimentally realized and characterized the Top Hat and Gaussian beam profiles. I
validated the experimental results with respect to the influence of the beam profile on
UC
with theoretical simulations by using the formula of the Andersson-Engels group for ideal
biphotonic converters. I wrote the manuscript in close corporation with Dr. Ute Resch-
Genger and Dr. Christian Würth.
XXI
Publication II
Manuscript Title:
Excitation power dependent population pathways and absolute quantum yields of
upconversion nanoparticles in different solvents
List of Authors:
Christian Würth*, Martin Kaiser*,, Stefan Wilhelm, Bettina Grauel, Thomas Hirsch, Ute
Resch-Genger
*Equally contributed
Substantial contributions:
I performed all measurements including excitation power density-dependent absolute
upconversion luminescence (UCL), upconversion quantum yield (
UC), and lifetime
measurements. I developed a
UC measurement strategy for upconversion nanoparticles
(UCNPs) dispersed in water and did the analysis of all
UC and lifetime data and created many
of the graphics of the manuscript. Interpretations of the luminescence quenching and change
in the luminescence emission band ratios were concluded with Dr. Christian Würth. I
proposed that the increasing green-to-red intensity ratio at low P is an argument for the
population of the red emissive 4F9/2 energy level via 4I13/2 level. I drafted the paper in close
corporation with Dr. Christian Würth and Dr. Ute Resch-Genger.
XXII
Publication III
Manuscript Title:
Explaining the influence of dopant concentration and excitation power density on the
luminescence and brightness of
-NaYF4:Yb3+,Er3+ nanoparticles: Measurements and
simulations
List of Authors:
Martin Kaiser, Christian Würth, Marco Kraft, Tero Soukka, Ute Resch-Genger
Substantial contributions:
I guided Marco Kraft at measuring the excitation power density-dependent absolute
upconversion luminescence (UCL) and upconversion quantum yield (
UC) for the differently
doped
-NaYF4:Yb3+,Er3+ nanoparticles. I performed all luminescence lifetime measurements,
did the analysis of all data, and performed a rate equation analysis with a 9-level Er3+ rate
equation model. I did the photophysical interpretation, which was discussed with
Dr. Christian Würth. I wrote the manuscript in close corporation with Dr. Ute Resch-Genger.
