Surface Plasmons, Fluorescence and Their Coupling
Hybrid metal-fluorescent emitters heterostructures have attracted much attention over the past decade because the interaction between the metal plasmons and the fluorescence excitons provides an opportunity to tailor fluorescence. So far, different metal-fluorescence emitters heterostructures have been investigated especially at the single fluorescent emitters level to fully understand the interaction between plasmons and excitonic fluorescence. In this chapter, the basic principles of surface plasmon resonance and fluorescence are introduced. Furthermore, and based on the state of the art, the interactions between plasmons and excitons are discussed.
Surface Plasmon Resonance
Introduction to Plasmon resonance
Plasmon resonance is an optical phenomenon. In metals, free electrons are not bound to a specific metal ion but circulating in the whole metal crystal; they migrate freely on the surface of the metal. Oscillating electric fields originating from incident light can delocalize the electrons and force them to move away from the metal framework. This situation is short lived: the free electrons will be pulled back because of the Coulombic attraction coming from the positively charged cations within the crystal. The resonance condition is established when the frequency of the incident light is coupled to the natural frequency of the electron oscillation in the metal.1 The natural frequency of the electronic oscillation for a bulk metal of infinite size only depends on the free electron density; most metals have ultraviolet plasmonic frequencies, except the plasmons of Au, Ag and Cu being located in the visible region.2 Two types of surface plasmons can be generated: Propagating Surface Plasmon Resonance (PSPR) and Localized Surface Plasmon Resonance (LSPR) (which occurs in small size metal particles).
Propagating Surface Plasmon Resonance (PSPR)
PSPR occurs on extended smooth metal surfaces. Positive and negative charges are generated as the electron density waves propagate along the metal surface under the excitation of incident light as shown in Figure 1.1 (a).3 The resulting plasmons decay evanescently in the direction perpendicular to the metal surface with 1/e decay lengths in the range of 200 nm. However, simple illumination with light on a smooth metal surface in air is not sufficient to induce plasmon resonance because the momentum of the light in the interfacial plane is always smaller than that of the SPs.
Figure 1.1. Schematic description of (a) propagating surface plasmon resonance (PSPR) and (b) localized surface plasmon resonance (LSPR) of a metal nanosphere. Reprinted from Ref .
Auxiliary optical devices, such as a glass prism with higher refractive index than the metal and optical grating are usually employed to generate PSPR.4 In the case of a glass prism, the light passes from the prism to the metal. Total internal reflectance will appear when the angle of the incident light is equivalent or beyond a critical angle. New evanescent light waves propagate from the interface into the metal in a very short distance and will couple to metal plasmons at a particular angle. The interaction between a molecular layer and the metalwill cause the change of refractive index of the media near the metal film and therefore the plasmon-coupled condition. This technique has been widely used in bio-analysis. When the incident light illuminates a metal grating, the wave vector will be enlarged as a result of light diffraction.5 The wave vector of diffracted light in a specific order parallel to the interface is in accordance with the surface plasmons; PSPR is then produced along the grating surface. It should be emphasized that PSPR can be excited by simple illumination with light on randomly roughened metal surfaces because diffraction effects derived from the roughness of the metal film provide the desired wave vector compensation for plasmon resonance.
Localized Surface Plasmon Resonance (LSPR)
LSPR occurs in metal particles with the size smaller than the wavelength of the incident light. Different from the previous PSPR, the oscillation of the free electrons induced by the incident electromagnetic field is strongly confined to the volume of the metal particles. The separation of the charges creates a dipole, whose direction can be switched with a change in the electric field, as displayed in Figure 1.1 (b).3 When the frequency of the natural dipole oscillation is matched to the incident light, a strong light absorption appears, which is referred as LSPR. Direct illumination of light on metal particles can produce a strong LSPR. The electromagnetic field resides very close to the particle surface due to the localization of the plasmon to the surface of the small metal particles, generating much higher local field enhancements than those of PSPR.1
LSPR strongly depends on the components, size and shape of the metal particles, and the dielectric constant of the local environment. For instance, dipole plasmons of Au nanospheres are excited when the size is smaller than 30 nm, thus displaying one single absorption peak, while higher order resonances can be obtained with increasing the size of Au nanospheres, displaying multiple absorption peaks.1 Thanks to the remarkable development of nanometer-scale fabrication technologies, a variety of metals (mainly Ag and Au because of their chemical stability and strong LSPR adsorption in the visible region) with various shapes and sizes have been successfully synthesized. Figure 1.2 shows TEM images of Au nanorods with various aspect ratios and corresponding absorption spectra (h).
Au nanorods splits into two bonds corresponding to the oscillation of the free electrons parallel and perpendicular to the long axis of the Au nanorods. The transverse mode has a similar resonance than the Au nanosphere at around 520 nm; while the resonance of the longitudinal mode can be tuned by changing the nanorods’ aspect ratio (length/diameter).
Nowadays, metal nanoparticles with different morphologies including nanospheres, nanorods,12,13 nanoprisms,14 bowtie nanoantennas,15 nanoshells,16,17 thin films,18,19 nanovoids,20 hole arrays,21 nanopatterns22 et al. have been used to investigate the interaction between plasmon and fluorescence.
Figure 1.2. (a-g) TEM images of Au nanorods with various aspect ratio and (h) corresponding absorption spectra. The insets of (c) and (g) show the photograph of the corresponding aqueous dispersion of gold nanorods. (h) UV-vis-NIR spectra of gold NRs shown in a and b (red curve), c and d (blue curve), e and f (green curve), and g (black curve), respectively. Each spectrum is normalized by its absorption at 400 nm. Scale bars: (a) 500 nm, (b) 50 nm, (c) 200 nm, (d) 50 nm, (e) 200 nm, (f) 50 nm, (g) 100 nm. Reprinted from Ref .
Optical properties of metal nanoshells
Metal nanoshells are a new class of nanoparticles composed of a single dielectric core nanoparticle (such as silica) coated with a metallic shell.23 The plasmon resonance of metal nanoshells can be tuned across the visible or infrared regions by varying the size of the core and the shell thickness. The capacity to tune plasmons resonance opens up opportunities for applications in the biomedical and surface-enhanced Raman scattering fields. Zhou et al. synthesized the first metal nanoshell which consisted of an Au2S core and an Au shell by mixing chloroauric acid and sodium sulfide in one step.24 The nanoshells plasmon resonance could be adjusted from 520 to 900 nm through variations in the ratio between the two precursors. However, this synthetic approach lacks a precise control over the core and shell dimensions.
Figure 1.3: (a) Theoretically calculated optical resonances of Au nanoshells as a function of the ratio between the silica core radius and the Au shell thickness. (b) A photograph of aqueous dispersions of Au nanoshells with different frequencies of plasmonic resonances. (c) Plasmon hybridization in metallic nanoshells arises from the interaction between cavity plasmons and sphere plasmons. (d) The strength of the interaction between the sphere and cavity plasmons is dependent on the thickness of the shell. Reprinted from Ref [23, 25].
In 1998, silica/Au core/nanoshell nanoparticles were successfully synthesized by Halas (Figure 1.3a and b).23 The separate synthesis of the silica core and the Au nanoshell enables a good control of their dimensions. Very tiny Au nanoparticles (seeds) are attached electronically to the silica surface functionalized with amino-propyltriethoxysilane. Subsequent chemical reduction of Au3+ to Au0 induces a further growth of the Au seeds, which results in the formation of continuous and polycrystalline Au nanoshells with a typical thickness between 5 – 30 nm. During this process, as the gold seeds grow, their plasmon peak becomes slightly red shifted; when the growing gold seeds begin to coalesce and form islands on the nanoparticle surface, the decreasing distance among these islands lead to increased inter-particle plasmon coupling and a red-shifted plasmon resonance.23 The plasmon-derived optical properties of metallic nanoshells can be understood in light of the plasmon hybridization model of metallic nanoshells.25 The outer nanoshell surface can be considered as a sphere, while the inner surface of the nanoshell can be considered as a cavity (Figure 1.3c). Similar to the hybridization of atomic orbitals, cavity plasmons and sphere plasmons mix and hybridize, resulting in two new resonances, the “bonding” plasmon and the “anti-bonding” plasmon. The strength of the hybridization is determined by the thickness of the nanoshell. Decreasing the thickness of the nanoshell leads to a stronger interaction, corresponding to a red-shifted resonance (Figure 1.3d), which is coincident with the theoretically calculated results. Silica/Au core/shell nanostructures display many unique optical properties, which, among others, include photothermal performance and production of solar steam nanobubbles.
Introduction to the fluorescence phenomenon
Nowadays, fluorescence is the most common labelling technique in biological and chemical sciences thanks to its high sensitivity. The excitation and subsequent emission processes of fluorophores are usually illustrated by the Jablonski diagram, as shown in Figure 1.4.28 Following light absorption, fluorophores are excited to a higher vibrational energy state, either S1 or S2. Then, these fluorophores will relax to the lowest vibrational level of S1 through a rapid process called internal conversion (within 10 ps), which is much faster than the fluorescence emission process. The process of returning from S1 to S0 is the fluorescence emission. Alternatively, fluorophores in the S1 state can also undergo a spin conversion (intersystem crossing) to the first triplet state, T1, resulting in phosphorescence emission.
Figure 1.4. Fluorescence absorption and emission processes illustrated by the Jablonski diagram. The singlet ground, first, second electronic states, triplet ground and first triplet state are depicted by S0, S1, S2, T0 and T1, respectively. Reprinted from Ref. .
Characteristics of Fluorescence Emission
Due to energy losses, the energy of the fluorescence photon is normally smaller than that of the exciting photons. This is displayed by a longer fluorescence wavelength than that of the exciting light (Stokes shift). In recent years, other nonlinear optical processes were discovered, in which the sequential absorption of two or more photons leads to the emission of light at shorter wavelength than the excitation wavelength; this up-conversion is an anti-Stokes emission.30 Lifetime and quantum yield are the two most important parameters of fluorescent substances. The fluorescence lifetime (τ) is defined as the time for which the fluorescence intensity decays to 1/e of the initial intensity after excitation. It determines the decay time of the first singlet state, in other words, an average value of the time spent in the excited state, and it is given by, where kf is the radiative decay rate and Σkn is the sum of the non-radiative decay rate. The fluorescence quantum yield (Q) is the ratio of the number of photons emitted to the number of photons absorbed. The decay of excited molecules includes radiative and non-radiative transition processes; hence the quantum yield is given by, Higher quantum yield represents stronger fluorescent intensity. The quantum yield can be close to 1 if the rate of radiative decay is much larger than the non-radiative decay rate. A variety of processes during which radiative and/or non-radiative rates are changed can decrease the fluorescence intensity. This is referred to as quenching. It should be noted that an increase of the radiative decay rate will result in higher quantum yields.
Classification of Fluorescent Substances
Fluorescent substances can be divided into several major classes.31 Organic fluorophores are the first observed fluorescent substances. Nowadays, commercial organic fluorophores with emission spectra ranging from ultraviolet to near-infrared (NIR) are widely utilized as bio-markers and in optoelectronic devices. However, organic fluorophores have broad absorption/emission spectra and low photostability due to their intrinsic photophysical properties. This limits the simultaneous detection of multiple signals and diminishes their effectiveness in long-term imaging. Fluorescent proteins are another type of fluorescent substances. Their fluorescence derives from aromatic amino acids. These fluorescent proteins have been used as fusion tags to study protein dynamics. Such proteins are usually genetically encoded. This characteristic, combined with their low fluorescent intensity, limits their practical application as injectable target-specific fluorescent probes.In recent years, inorganic nanoparticles, such as semiconductor nanocrystals (quantum dots, QDs) and up-conversion nanocrystals have been greatly developed. This was motivated by their unique optical properties,33,34 as for example, broad excitation ranges, narrow emission peaks, spectra spanning the UV to near-infrared, high fluorescence intensity and photostability. Furthermore, the QDs surface can be conjugated with a variety of bio-molecules to enable their application in bio-imaging.35 In spite of their intrinsic optical advantages, a major concern still remains around QDs, and that is their possible toxicity due to the heavy metals contained in QDs. In the next part, the general characteristics of QDs will be presented in detail.
Introduction to Colloidal Semiconductors
Colloidal semiconductors are light-emitting particles commonly synthesized in the solution phase. In the last several decades, they have attracted considerable attention as an important new class of materials in bio-imaging, sensitive detection and optoelectronic devices.36 When bulk semiconductors absorb energy, an electron in the valence band (VB) can be excited to the conduction band (CB) (the minimum energy required to excite an electron is the band gap energy (Eg)), and make the semiconductor conductive. An electron-hole pair, known as the exciton, is produced during this process. Relaxation of the excited electron back to the hole can give rise to the photon emission. The exciton has a finite size within the crystal defined by the Bohr exciton radius.37 When the radius of the semiconductor particle is close to or smaller than its Bohr exciton radius, the motion of electrons and holes is greatly spatially confined. This strong dependence on the size is called quantum confinement effect. From the molecular orbital theory, when the crystal size is reduced to a certain value, the electron energy levels of the semiconductor become discontinuous; as a result, the band gap energy continuously increases as the crystal size decreases. Colloidal semiconductors with typical dimensions in the range of 1-100 nm possess size-dependent absorption and fluorescence spectra with discrete electronic transitions. For instance, in the case of CdSe nanocrystals, their emission can span almost the entire visible spectrum from blue to red simply by varying the size of the nanocrystal from ~2.2 to ~5.5 nm, as shown in Figure 1.5.
Figure 1.5. Illustration of quantum confinement effects in CdSe colloidal nanocrystals. A red shift in emission is observed for CdSe nanocrystals of increasing size.
Table of contents :
Chapter 1 Surface Plasmons, Fluorescence and Their Coupling
1.1 Surface Plasmon Resonance
1.1.1 Introduction to Plasmon resonance
1.1.2 Propagating Surface Plasmon Resonance (PSPR)
1.1.3 Localized Surface Plasmon Resonance (LSPR)
1.1.4 Optical properties of metal nanoshells
1.2.1 Introduction to the fluorescence phenomenon
1.2.2 Characteristics of Fluorescence Emission
1.2.3 Classification of Fluorescent Substances
1.2.4 Colloidal Semiconductors
188.8.131.52 Introduction to Colloidal Semiconductors
184.108.40.206 Nanocrystal Structure
220.127.116.11 Optical properties
1.3 The interaction between plasmons and excitons
1.3.1 Weak coupling between surface plasmons and excitons
18.104.22.168 Enhancement of fluorescent intensity
22.214.171.124 New phenomena arising from the coupling between excitons and plasmons
1.3.2 Strong coupling between surface plasmons and excitons References
Chapter 2 Synthesis of Quantum Dots and Their Incorporation in Silica
2.1 Synthesis of QDs
2.1.2 Precursors preparation
2.1.3 Synthesis of 6 nm-in-radius-CdSe/CdS core/shell QDs
126.96.36.199 Synthesis of CdSe QDs
188.8.131.52 Formation of a CdS shell on CdSe cores
2.1.4 Synthesis of 15 nm-in-radius-CdSe/CdS core/shell QDs
184.108.40.206 Synthesis of CdSe cores
220.127.116.11 Formation of a CdS shell on CdSe cores
2.1.5 Synthesis of CdSe/CdS/ZnS
18.104.22.168 Synthesis of CdSe/CdS
22.214.171.124 ZnS shell growth
2.2 Synthesis of QD/SiO2
2.2.2 Selection of surfactant
126.96.36.199 Synthesis of QD/SiO2 NPs with Igepal CO-520
188.8.131.52 Synthesis of QD/SiO2 NPs with Triton X-100
184.108.40.206 The effect of the surfactant on the QD/SiO2 NPs size
2.2.3 Effect of the QDs amount
2.2.4 Growth of thicker SiO2 layer via Stöber method
220.127.116.11 Regrowth of silica on QD/SiO2 NPs (radius-62 nm)
18.104.22.168 Regrowth of silica on QD/SiO2 NPs (radius-25 nm)
22.214.171.124 The effect of the QD/SiO2 starting size on the silica regrowth
Chapter 3 Generalized Synthesis of QD/SiO2/Au Core/shell/shell Nanoparticles
3.2 Functionalization of QD/SiO2 NPs
3.2.1 Synthesis of poly(1-vinylimidazole-co-vinyltrimethoxysilane), PVIS
3.2.2 Functionalization of the surface of QD/SiO2 NPs with PVIS or APTMS
3.3 Gold seeds adsorption
3.3.1 Preparation of the gold seeds solution
3.3.2 Adsorption of the gold seeds onto QD/SiO2 NPs
3.3.3 Determination of the optimal amount of PVIS for the functionalization of QD/SiO2 NPs
3.3.4 Effect of the functionalizing agent on the amount of adsorbed gold seeds
3.3.5 Effect of the pH on the amount of adsorbed gold seeds
126.96.36.199 PVIS-functionalized QD/SiO2 NPs
188.8.131.52 APTMS-functionalized QD/SiO2 NPs
3.3.6 Fresh gold seeds adsorption
184.108.40.206 Effect of the functionalizing agent on the amount of adsorbed fresh gold seeds
220.127.116.11 Effect of the gold seeds aging on the amount of PVIS-adsorbed gold seeds
18.104.22.168 Effect of the pH on the amount of PVIS-adsorbed fresh gold seeds
3.4 Growth of the gold nanoshell from QD/SiO2/Auseeds NPs
3.4.1 Reduction methods tested
3.4.2 Results, interpretations and choice of the reduction method
3.4.3 PVP-K12-assisted gold nanoshell growth
3.4.4 Effect of the gold seeds coverage density
22.214.171.124 Gold seeds coverages obtained with APTMS or PVIS
126.96.36.199 Gold seeds coverages obtained with PVIS at different pH values
3.4.5 Effect of the monolayer/multilayer adsorption of the freshly-made gold seeds
3.4.6 Synthesis of golden QDs with tailored thicknesses of gold and silica
188.8.131.52 Adjustment of the gold nanoshell thickness on QD/SiO2 NPs with a fixed size
184.108.40.206 Gold nanoshell formation on QD/SiO2 with various silica thicknesses
Chapter 4 Optical Properties of Golden QDs
4.2 Optical study of golden 15 nm-in-radius-QDs
4.2.1 Mechanisms of the fluorescence emission for CdSe/CdS QDs and for golden QDs
4.2.2 Ensemble measurements
4.2.3 Single nanoparticle measurements
Chapter 5 Self-assembled Colloidal Superparticles
5.2 Synthesis of SPs
5.2.1 Self-assembly of hydrophobic nanocrystals: the mechanism
5.2.2 Synthesis of colloidal SPs via self-assembly of QDs
5.2.3 Adjustment of the size of SPs
5.3 The incorporation of SPs in silica
5.3.1 The effect of injection solvent on the silica shell growth
5.3.2 The effect of PVP molecular weight on the silica shell growth
5.3.3 Growth of thick silica shell on SPs/SiO2
5.4 Synthesis of SPs and SPs/SiO2 from different types of nanocrystals
5.5 Synthesis of SPs/SiO2/Au nanoshells (golden SPs)