Luminescent Rare-earth-based Nanoparticles: A Summarized Overview of their Synthesis, Functionalization, and Applications
- First Online:
- Cite this article as:
- Escudero, A., Carrillo-Carrión, C., Zyuzin, M.V. et al. Top Curr Chem (Z) (2016) 374: 48. doi:10.1007/s41061-016-0049-8
Rare-earth-based nanoparticles are currently attracting wide research interest in material science, physics, chemistry, medicine, and biology due to their optical properties, their stability, and novel applications. We present in this review a summarized overview of the general and recent developments in their synthesis and functionalization. Their luminescent properties are also discussed, including the latest advances in the enhancement of their emission luminescence. Some of their more relevant and novel biomedical, analytical, and optoelectronic applications are also commented on.
KeywordsLuminescence Nanoparticles Rare earths Synthesis Bioimaging Biosensing
1 Introduction, Chemical Composition, Luminescent Properties
Rare-earth (RE)-based nanoparticles (NPs) constitute one type of luminescent materials available in the literature. RE-based nanophosphors exhibit important advantages compared with the other available luminescent materials due to their lower toxicity, photostability, high thermal and chemical stability, high luminescence quantum yield, and sharp emission bands . These nanophosphors usually consist of a host inorganic matrix doped with luminescent lanthanide (Ln) cations. The final characteristic properties of the nanophosphors are highly influenced by both the inorganic matrix and the dopant. Fluoride matrices are used due to their low vibrational energies, which minimize the quenching of the exited state of the Ln cations and result in a higher quantum efficiency of luminescence [2, 3, 4]. Phosphate-based matrices attract interest for their high biocompatibility and good biodegradability . Other matrices such as vanadates, molybdates, and wolframates are used to enhance the global luminescent emission of the materials [6, 7], and some silicate-based matrices are appropriate for the production of persistent luminescent NPs [8, 9]. The election of the Ln cation or cations determines the final luminescent properties of the material. Luminescence is expected for most of the Ln3+ cations, but in practice most of the studies are focused on Eu3+, Tb3+/Ce3+, Dy3+, and Nd3+ cations, which produce red, green, yellow/orange luminescence, and near-infrared luminescence, respectively [10, 11, 12, 13, 14]. These cations are examples of the so-called downconversion (DC) luminescence (i.e. conventional Stokes type), in which higher energy photons are converted into lower energy photons. High research attention is attracted by upconverting nanoparticles (UCNPs), in which the sequential absorption of two or more photons leads to the emission of light at shorter wavelengths than the excitation wavelength (i.e. anti-Stokes type emission), which means that near infrared long-wavelength excitation radiation is converted into shorter visible wavelengths [15, 16]. Different mechanisms of upconversion luminescence have been reported for lanthanide-based materials. Excited-state absorption occurs for singly doped upconversion materials, and is based on a successive two-photon absorption of appropriate energy. Such absorption takes place via an excited metastable level, and the relaxation from the excited to the original ground state produces the upconversion luminescence. Such processes are highly inefficient, and low concentration of lanthanide dopants are normally required . Much higher efficiencies (but still quite low) are observed in energy-transfer processes, in which an absorbing sensitizer cation and an emitting activator cation are required. Er3+, Tm3+, and Ho3+ codoped with Yb3+ are commonly used as upconverting luminescent cation pairs. Yb3+ normally acts as a sensitizer cation, given its large absorption cross-section and the resonance of its 2F7/2 → 2F5/2 transition with many f–f transition of other lanthanide cations . After the absorption and energy transfer processes, the upconversion luminescence appears in the active cations by different mechanisms, including excited-state absorption, successive energy transfer, cross-relaxation upconversion, cooperative sensitization, and cooperative luminescence .
The main disadvantage of Ln-doped NPs is their relatively low global intensity luminescence, caused by the low absorptions of the parity forbidden Ln3+ 4f–4f transitions, and constituting a serious limitation for their use for different applications . Different energy transfer schemes from the host materials to the Ln cations are employed to enhance the global luminescence of downconverting lanthanide-doped NPs, which include the use of vanadate or oxyfluoride matrices [21, 22, 23, 24, 25]. Despite their highly scientific interest, extremely low upconversion efficiencies (in the range of 10−5 to 1 %) are normally observed for UCNPs, which depend also on the excitation laser power density, and on the material, size, and surface structure of the NPs [26, 27, 28]. Core/shell nanostructures, which minimize the surface quenching effects [29, 30, 31, 32], as well as the association with organic near-infrared (NIR) dyes, which can alleviate the inherently weak and narrow near-infrared absorption of the Ln ions , are used to enhance the luminescence of such materials. However, laser excitation sources are still required to study these particles.
In this article the more recent and common methods of synthesis of luminescent NPs based on RE will be briefly summarized, as well as the different existing functionalization strategies. Some of their imaging, sensing, and optoelectronic applications based on their fluorescent properties will also be mentioned. For a deeper and more detailed description, some excellent reviews can be found in the recent literature [18, 19, 34, 35, 36, 37, 38, 39].
2 Synthesis of Uniform Luminescent Nanoparticles
Recently, laser ablation of micrometric-sized powder Ln-doped particles has been used to produce Ln-doped NPs with a great control of their size and monodispersity .
3 Functionalization and Colloidal Stability
A functionalization process is especially required for the biomedical use of NPs, and it is mandatory for non-water-dispersible NPs. Functionalization not only increases the colloidal stability of the NPs by introducing electrostatic and/or steric repulsions , but also provides anchors for adding functional ligands of biomedical interest such as antibodies, peptides, proteins, and some anticancer drugs . Ligand exchange, polymer encapsulation, and silica encapsulation are common strategies used for the stabilization in water of native hydrophobic NPs. In the ligand exchange method, the original hydrophobic ligands are completely displaced by hydrophilic ligands (i.e. PEG-type and polymeric ligands, and anions such as citrate and BF4−) on the NP surface [60, 61]. Ligand exchange methods typically offer NPs with smaller hydrodynamic diameters but suffer (for most NP materials) from limited colloidal stability. Polymer coating yields NPs that are very colloidally stable, but that normally show larger hydrodynamic radii . In this strategy, the hydrophobically capped NPs are overcoated with amphiphilic polymers such as poly(isobutylene-alt-maleic anhydride) modified with dodecylamine (PMA), and its modifications with 4-(aminomethyl)pyridine (Py-PMA), and polyethylene glycol (PEG-PMA) [60, 63, 64, 65, 66]. The hydrophobic portion of the polymer intercalates with the hydrophobic ligands on the NP surface, leaving the hydrophilic portion of the polymer exposed to solution . Treatments with acids  or with excess of ethanol under ultrasonication  have also been used to remove the hydrophobic organic coating of the NPs, and the oleic acid ligands on the NPs can be oxidized with the Lemieux-von Rudloff reagent, yielding water-dispersible carboxylic acid-functionalized NPs .
4 Bioimaging Applications
Both downconverting and upconverting Ln-doped NPs can be used for fluorescent bioimaging applications. The features of such particles are highly influenced by their optical and luminescent properties, as previously reported . Ln-doped NPs show some advantages when compared with other available luminescent nanomaterials, such as their low photobleaching, nonblinking (i.e. they do not show intermittent emission), high efficiencies (for downconversion of NPs), possibility of multifunctional and multimodal imaging, and sharp emission lines, which prevents possible interferences . Moreover, they can even be used for long-term imaging due to their photostability . The sharp and specific bands of the direct excitation of the Ln-doped NPs constitute, however, a limitation, since specific equipment and appropriate laser wavelengths for the excitation are normally required. The main disadvantage of the use of Ln-doped NPs for fluorescent bioimaging applications relies on their low global luminescent emission. For the DC nanophosphors, the direct excitation of the Ln3+ cations (which normally consists of narrow and low absorbance bands, the more intense occurring at 393 nm for Eu3+, 349 and 366 nm for Tb3+, and 349 nm for Dy3+) is normally not enough to produce intense luminescence. As mentioned above, this can be overcome though an indirect excitation through the matrix, but ultraviolet excitations radiations are still required, which in some ways can be harmful for the cells.
5 Sensing and Analytical Applications
Application of Ln-doped NPs for sensing can be roughly divided into two classes: one is the directly observed luminescence from the Ln-doped NPs, and the other is based on energy transfer, and/or reabsorption processes, including fluorescence resonance energy transfer (FRET) and lanthanide-based resonance energy transfer (LRET).
An important feature of Ln-doped NPs for using their direct intrinsic fluorescence is that they present multiple emission lines, which allows ratiometric measurements because normally some of them are analytically sensitive, while others are insensitive and serve as reference signals . In other cases, the emission corresponding to the Ln3+ can be considered as insensitive, and the functionalization of the NPs with fluorophores whose intensity is affected by the presence of the analyte allows the design of other ratiometric sensors . FRET is an energy-transfer process between an excited donor fluorophore and a ground-state acceptor fluorophore in close proximity (approximately 1–20 nm). Due to their small size, lanthanide complexes (mostly based on Eu3+ and Tb3+) are often used in FRET-based assays [86, 87]. Small Ln-doped NPs (generally with sizes below 30 nm) can also be coupled with organic fluorophores, metallic NPs, or quantum dots for FRET/LRET-based sensing approaches, where Ln-doped NPs are typically the donor unit [88, 89, 90, 91]. For providing selective detection towards a specific analyte (e.g. biomolecules, ions, gas molecules), the NPs have to be functionalized with suitable groups/motifs that have a recognition capability of the target analyte. For example, a single-stranded DNA has been used as Hg2+-capturing element in the development of a method for determining Hg2+ ions based on a FRET mechanism between Tm3+, Yb3+-doped NaYF4 UCNPs as energy donor and a DNA intercalating dye (SYBR Green I) as energy acceptor . As the SYBR has a strong absorbance overlapping with the blue emission of the UCNPs, in the presence of Hg2+ ions there is a simultaneous decrease of the blue emission of the UCNPs and an increase of SYBR green emission. By monitoring the ratio of the acceptor emission to the donor emission, the Hg2+ ion can be detected at levels as low as 0.06 nM. This system allows for not only determining the concentration of Hg2+ but also monitoring changes in the distribution of Hg2+ in living cells by upconversion of luminescence bioimaging.
In-vivo pH sensing by an energy transfer mechanism has been carried out with Er3+, Yb3+-doped β-NaYF4 UCNPs functionalized with polyglutamic porphyrin–dendrimers. In this case, the green and red emission bands of the UCNPs are attenuated differently depending on the protonation state of the porphyrin .
The sensitive and specific interaction between avidin and biotin has been used for the detection and quantification of fluorescein isothiocyanate (FITC)-labeled avidin with biotinylated NPs. In this case, the increase in the emission of the FITC at the expense of the Ln3+ was correlated to the amount of FITC-labeled avidin, with a detection limit up to 3.0 nM . Dissolution-enhanced luminescent bioassays (DELBA) with ultra-small Ln-based NPs have been recently proposed for the detection of tumor biomarkers due to their relatively high stability and greater flexibility for bioconjugation. In this case, the NP size is a key factor, since ultra-small NPs minimize the interference with the antigen–antibody binding processes , while larger nanoprobes based on lanthanide-chelate-embedded polystyrene or silica NPs have a tendency to agglomerate and swell in aqueous solution . After an appropriate functionalization, often based on the well-known specific interaction between avidin and biotin, the NPs are able to recognize the analyte. The addition of an enhancer solution containing high chelating and sensitizing additives dissolves the NPs, releases the Ln3+ cations, and finally, highly luminescent lanthanide complexes (also called lanthanide luminescent micelles) are formed. This strategy has been used for the ultrasensitive in-vitro detection of carcinoembryonic antigen (CEA) , and prostate-specific antigens .
By using Au NPs as an acceptor instead of a fluorophore, the detection of trace amounts of avidin has been reported. In this system the selective and sensitive avidin–biotin interaction is responsible for bringing together the avidin-modified Er3+, Yb3+-doped NaYF4 NPs used as donor and the biotinylated-Au NPs, whose strong absorption at ~541 nm matches well with the green emission of the Ln-doped NPs, and therefore an effective FRET process occurs . An important advantage of this approach is its potential to be extended to wherever the avidin–biotin system functions, for example to study protein-protein interactions, ligand-receptor interactions, the formation of DNA duplexes, and so on.
Semiconductor QDs have also been combined with UCNPs in FRET configurations. The superiority of QDs as acceptors owes to the fact that they have broad excitation bands and size-tuneable emission wavelength, and thus the upconversion wavelength of the UCNP–QD couple may be continuously adjusted. Combining Tm3+, Yb3+-doped NaYF4 UCNPs as the energy donor and the CdTe QDs as the energy acceptor, the determination of lead ions in human serum with a detection limit of 80 nM has been achieved . Such a low detection limit is possible thanks to the use of the NIR-laser as excitation source, which is capable of overcoming self-luminescence from serum excitation with visible light.
A FRET process is not only possible between two NPs, but also between the emission bands of RE NPs and an enzyme absorbance band. Tm3+, Yb3+-doped Gd4O2S UCNPs have been used to monitor the redox state of a flavoenzyme (PETNR, pentaerythritol tetranitrate reductase) . Due to a variation in the absorbance profile of the flavin core of the enzyme upon reduction/oxidation, the FRET between the donor/acceptor units can effectively be turned ‘on’ or ‘off’ by changing the redox state of PETNR. The presence of two bands separated by over 300 nm allowed the ratiometric signaling of this process.
The multiplexing capabilities of Ln-doped NPs have been also demonstrated by using different UCNPs excited with the same IR laser. The simultaneous detection of two types of pathogenic bacteria (Salmonella Typhimurium and Staphylococcus aureus) was carried out by means of aptamer-conjugated Er3+, Yb3+, and Tm3+, Yb3+-doped NaYF4 UCNPs .
Interestingly, Ln-doped UCNPs can also be used as nanothermometers based on the strong temperature dependence of the fluorescence intensities from two emitting levels of lanthanides [104, 105]. This principle has been exploited for monitoring temperature changes in living cells, which is of particular interest for the investigation of enzyme reactions and sub-cellular processes . Wolfbeis et al. studied temperature sensing using UCNPs of varying size and RE dopants recently . They found that the core–shell structured hexagonal 2 % Er3+, 20 % Yb3+-doped NaYF4/NaYF4 UCNPs were more suitable for temperature sensing because their higher brightness allowed resolving temperature differences of less than 0.5 °C in the physiological range between 20 and 45 °C .
6 Optoelectronic Applications
Because of their unique optical properties, Ln-doped materials are also widely used for optoelectronic applications, which include laser sources , fiber-optic communication , light-emitting diodes and solid-state lightening [110, 111], and color display devices . These properties have been extensively studied in bulk materials since the last century, and nowadays the design and the study of the properties and applications of the nanostructured materials attract wide research interest. In particular, Ln-doped semiconductor nanostructures open up the possibilities for photonic/electronic integration and cheap CMOS-compatible optical sources, while efficient solid-state lighting based on Ln-doped nanostructures are increasingly playing a role in new green technologies .
Some Ln-doped NPs are used as color and white light-emitting materials. For the latter, both the selection of appropriate host matrices, which should be able to excite the Ln cations after one single wavelength absorption, and the optimization of the Ln contents, which can also absorb and transfer energy between them, are demanded. Examples of Ln-doped white light-emitting NPs are Dy3+, Eu3+-doped ZrO2 , Dy3+-doped yttria stabilized zirconia (YSZ) , and Dy3+, Tb3+, Eu3+-doped GdPO4 .
Ln-doped NPs are also used to improve the energy conversion efficiency in solar cells by overcoming both of the two primary loss mechanisms in solar cells. Such mechanisms are related to the absorption of photons with larger or lower energy than the bandgap of the solar cell, reducing their efficiency in practice. On one hand, DC NPs can absorb UV photons and re-emit them at longer wavelengths, where the solar cell exhibits a significantly better response . For example, Eu3+, Bi3+-doped YVO4 NPs have been used in Si-based solar cells to reduce the thermalization of charge carriers caused by the absorption of high-energy photons . On the other hand, UCNPs are used to transform low-energy photons into higher energy photons that can be used by the solar cells, and thus significantly enhance the efficiency of the photovoltaic device . Er3+, Yb3+-doped NaYF4 NPs, one of the most studied fluorides, has been used with this objective . In some cases, UCNPs are also associated with organic dyes .
7 Concluding Remarks and Future Outlook
Current and widely used strategies for synthesis and functionalization of RE-based NPs have been discussed, and some recent advances in their imaging, sensing, and optoelectronic applications have been mentioned. Even when the synthesis in organic media, normally in the presence of additives such as oleic acid, provides a powerful tool for the production of highly monodispersed NPs, the hydrophobic character of the resulting NPs requires the development of new synthesis routes yielding homogeneous, uniform, and water-dispersible NPs. The use of polyol-based solvents in homogeneous precipitation reactions at moderate temperatures has partially overcome this disadvantage, although reported NP sizes are larger (>40 nm). New synthesis methods yielding Ln-doped NPs with different shapes and sizes for many systems are still in demand, as well as universal functionalization strategies for hydrophilic NPs, as the layer-by-layer approach. Apart from the synthesis and functionalization perspectives, the main disadvantage of Ln-based NPs continues to be their relatively low emission intensity. Even when the indirect excitation with codoped inorganic matrices for downconverting NPs and core/shell structures for upconverting NP have demonstrated notable improvements, new strategies to enhance their luminescence are still demanded.
This work was supported by a Junta de Andalucía (Spain) Talentia Postdoc Fellowship, co-financed by the European Union’s Seventh Framework Programme, Grant agreement No. 267226, and by the European Commission (Grant FutureNanoNeeds to WJP). CCC acknowledges the Spanish Ministry of Economy and Competitiveness for a Juan de la Cierva—Incorporacion contract.