Plasmonic enhancement of the upconversion fluorescence in YVO₄:Yb³⁺, Er³⁺ nanocrystals based on the porous Ag film

The porous Ag film was prepared by the PMMA opal photonic crystals (OPCs) template removal method, and the OPCs were prepared according to our previous works [34]. The specific experiments are as follows. First, the OPCs were penetrated with a solution of AgNO₃ in a mixture of water and ethanol. After infiltration, the product was dried in air at room temperature for 30 min. The annealing was carried out by slowly increasing the temperature (50 °C h⁻¹) up to 500 °C for 3 h, then a porous Ag film formed. In the preparation process of the YVO₄: (15 mol%)Yb³⁺, xEr³⁺ (x = 1, 2, 3, 5, 8 mol%) precursor sol, the stoichiometric amounts of Y(NO₃)₃ · 6H₂O, Yb(NO₃)₃ · 6H₂O, Er(NO₃)₃ · 6H₂O and NH₃VO₄ was first dissolved in ethanol solution. The ethanol solution contained citric acid as the chelating agent. Then, appropriate amount of nitric acid was dissolved in ethanol. The mixture was stirred for 4 h, forming a transparent solution. The mixture solution was infiltrated in the voids of the porous Ag film. The further annealing was carried out with slowly elevated temperature (40 °C h⁻¹) up to 500 °C for 3 h. Finally, the porous Ag/YVO₄:Yb³⁺, Er³⁺ NCs hybrids were formed. The preparation process is shown in scheme 1.

Zoom In Zoom Out Reset image size

The samples were characterized with a JEOL JSM-7500 field emission scanning electron microscope at an accelerating voltage of 15 kV, and a Hitachi H-800 transmission electron microscope (TEM) operated at an acceleration voltage of 200 Kv. The crystalline structure of the samples was characterized by x-ray diffraction (XRD) (RigakuD/max-rA power diffractometer using Cu–KR radiation (λ)1.54178 Å). (UV–vis) absorption spectra were measured with a Shimadzu UV-3101PC UV–vis scanning spectrophotometer ranging of 200–1100 nm. A continuous 980 nm diode laser was used to pump the samples when investigating the steady-state spectra. The luminescent dynamics of Er³⁺ ions were investigated by a laser-system consisting of a Nd:YAG pumping laser (1064 nm), the third-order Harmonic-Generator (355 nm) and a tunable optical parameter oscillator (OPO, Continuum Precision II 8000). The laser has pulse duration of 10 ns, repetition frequency of 10 Hz and line width of 4–7 cm⁻¹.

Figure 1(a) shows the SEM image of the porous Ag film, which indicates that the Ag film is composed of randomly distributed and regular Ag spheres, ranging in size from 100 to 2000 nm, with many pores among the Ag spheres formed during the annealing. Figure 1(b) records the SEM image of the porous Ag/YVO₄:Yb³⁺, Er³⁺ film. It can be seen that the YVO₄:Yb³⁺, Er³⁺ UCNCs were mainly filled in the pores between the Ag particles and covered on the surface of the Ag NPs, indicating the formation of Ag/YVO₄:Yb³⁺, Er³⁺ hybrids. As a comparison, the SEM image of YVO₄:Yb³⁺, Er³⁺ film on the glass substrate (as the reference sample) was shown in figure 1(c), indicating that very thin but continuous YVO₄:Yb³⁺, Er³⁺ films were formed on the glass substrate, consisting of closely packed UCNCs, with the size of YVO₄:Yb³⁺, Er³⁺ NCs being ∼15 nm. Figure 1(d) represents the XRD patterns of the porous Ag film and the porous Ag/YVO₄:Yb³⁺, Er³⁺ composite film on the glass substrate and the corresponding standard cards, JCPDS 87–0717 for cubic silver and 17–0341 for tetragonal phase YVO₄. In the Ag/ YVO₄:Yb³⁺, Er³⁺ composite film, the (111) face of cubic silver and the characteristic diffraction peaks at (200), (112) and (312) of tetragonal phase YVO₄ can be clearly distinguished, and no impurity peaks appeared, indicating the formation of the porous Ag/YVO₄:Yb³⁺, Er³⁺ composite film. In most of the previous works, the metal/luminescence centers hybrids were prepared according to the chemical control methods, which introduced some species, such as surfactants and solvents, would significantly affect the effective of UCL and the stable of noble metal NPs [35, 36]. As a comparison, in the porous Ag film/YVO₄:Yb³⁺, Er³⁺ hybrids prepared by the PMMA template removal method not only can rule out the effect of the large phonon groups on the UCL, but also improving the stability of the Ag particles. Furthermore, the SPR peak of metal (the excitation wavelength) and the coupling distance between UCNCs and metal NPs is easy to control in our system, which leads to stable and reproducible UCL enhancement.

Zoom In Zoom Out Reset image size

In order to further determine the structure of the porous Ag/YVO₄:Yb³⁺, Er³⁺ hybrid film, the hybrid was scraped from the glass substrate and dispersed in ethanol, then dropped onto a copper grid for high-resolution TEM (HR-TEM) measurements. Figures 2(a) and (b) respectively show low and high magnification TEM images of Ag/YVO₄:Yb³⁺, Er³⁺ hybrids. It can be clearly seen that the YVO₄:Yb³⁺, Er³⁺ UCNCs were adsorbed on the surface of the Ag particle and the thickness of the surface YVO₄:Yb³⁺, Er³⁺ layer on the Ag particle was in the range of 20–100 nm. The obvious lattice fringes confirm the high crystallization of the samples, as shown in the HR-TEM image (figure 2(c)). A distance of about 0.36 nm between adjacent lattice fringes could be well assigned as the d spacing value of the (200) plane of YVO₄. The energy-dispersive x-ray (EDX) analysis in panel d of figure 2(d) shows the presence of Y V and O with an approximate stoichiometry of YVO₄, furthermore, the spectrum confirms the presence of Ag in the sample. Note that the peaks of C and Cu come from the surface of the copper grid used for TEM measurement. The EDX results indicate that the Ag/YVO₄:Yb³⁺, Er³⁺ hybrids were indeed formed.

Zoom In Zoom Out Reset image size

The absorption and reflectance spectra of the Ag nano-film were measured, as shown in figure 3. It's interesting to observe that the absorption of the porous Ag film is a broad band extending from 350 to 1100 nm, which largely differs from the absorption of Ag NPs and can be attributed to the strong plasmon coupling among the neighboring Ag NPs. From the reflection spectrum of the Ag nano-film, the reflectance is found to be about 30% in the range of 520–550 nm, and it gradually increases with increasing wavelength. In the NIR region (980 nm), the reflectance approaches about 40%. The UCL spectra in YVO₄:Yb³⁺, Er³⁺ samples were also recorded under the excitation of a 980 nm laser diode in air, as shown in figure 3. The UCL of Er³⁺ ions was clearly observed, mainly determined to green (²H_11/2, ⁴S_3/2–⁴I_15/2) and red (⁴F_9/2–⁴I_15/2) transitions, respectively. In contrast to YVO₄:Yb³⁺, Er³⁺ film, the overall fluorescence intensity of Er³⁺ in the Ag/YVO₄: Yb³⁺, Er³⁺ composite film sample increases up to 36 and 30 fold for ²H_11/2–⁴I_15/2 and ⁴S_3/2–⁴I_15/2 transitions. It should be noted that the SPR absorption of the porous Ag film covers the entire excitation (∼980 nm)/emission region (400–700 nm) of YVO₄: Yb³⁺, Er³⁺ samples.

Zoom In Zoom Out Reset image size

In order to investigate the modulation on the UCL of YVO₄: Yb³⁺, Er³⁺ UCNCs based on the porous Ag film. Figure 4(a) records the enhancement factor (EF) versus the 980 nm excitation power (EF is defined as the ratio of integral intensity between the porous Ag/YVO₄: Yb³⁺, Er³⁺ film and YVO₄: Yb³⁺, Er³⁺ film). It can be seen that the EF depends strongly on excitation power as well as the emission wavelength. It's exciting to observe that the maximum EF of the ²H_11/2–⁴I_15/2 and ⁴S_3/2–⁴I_15/2 transitions are about 36 fold and 30 fold, respectively, corresponding to a low excitation power of 120 mW. Based on our previous work [33], the EF in porous Ag film is obviously higher than in dense Ag film, because the UCNCs is convenient to fill in the porous Ag film and on the surface of Ag particles, which is useful for decreasing the coupling distance between UCNCs and Ag NPs, while in the dense Ag film, the UCNCs can only locate at the top of the Ag film layer. On the other hand, the UCL enhancement has been influenced by the scattering–absorption ratio, it is usually expected that the absorption term to cause quenching and the scattering term to cause enhancement. The absorption term increases as r³ (r is the radius of Ag sphere) where the scattering term increases as r⁶ [19]. So the large size (100–2000 nm) of Ag NPs is benefit for the UCL enhancement. And as the excitation power increases, the EF of all the transitions gradually decreases, which should be due to the local thermal effect induced by the laser exposure. In order to better understand the fact above, the UC emission intensity as a function of excitation power for the ²H_11/2, ⁴S_3/2–²H_15/2 transitions were recorded, as shown in figures 4(b) and (c). It can be seen that the values of slope n (I_UCLIⁿ) in Ag/YVO₄: Yb³⁺, Er³⁺ (1.03 for ²H_11/2–⁴I_15/2, 0.54 for ⁴S_3/2–⁴I_15/2) are smaller than those in YVO₄: Yb³⁺, Er³⁺ (1.76 for ²H_11/2–⁴I_15/2, 1.62 for ⁴S_3/2–⁴I_15/2), and the slope n for the emissions of Er³⁺ ions are all smaller than the required photon number (n = 2) for generating the corresponding transitions, even at low excitation power. All the facts above can be attributed to the saturation effect as well as the local thermal effect induced by the laser exposure [37]. Figure 4(d) displays the deduced local temperature (based on R_HS) of the sample as a function of 980 nm excitation power. It can be seen that the porous Ag/YVO₄: Yb³⁺, Er³⁺ composite film has a relatively higher temperature than the YVO₄: Yb³⁺, Er³⁺ nano-film, which is due to the absorption by the silver nano-film of the 980 nm excitation light and the further photo-thermal transfer, leading to the increase of nonradiative relaxation of Er³⁺ ions and the decrease of EF. This result is different to our previous reports, which would be due to the different Ag density and the different preparation process [32, 33]. It can be concluded that in the porous Ag/YVO₄: Yb³⁺, Er³⁺ composite film, the coupling distance between Ag particles and YVO₄:Yb³⁺, Er³⁺ NCs have been effectively reduced. It is well known that the maximum electric field intensity enhancement occurs mostly in the regions close to the surface of Ag NPs (within a few nanometers). So the EF has been improved considerably. Then on the other hand, as to the high Ag density and Yb³⁺ doped concentration in YVO₄ (the phonon energy ∼870 cm⁻¹), all of these can absorb 980 nm excitation light and effectively photo-thermal transfer, and the local thermal effect in Ag/YVO₄:Yb³⁺, Er³⁺ will become serous relative to the YVO₄:Yb³⁺, Er³⁺ thin film sample at the high 980 nm excitation power, so the EF decrease with the increase of the excitation power.

Zoom In Zoom Out Reset image size

In order to better understand the origin of UCL enhancement, the concentration-dependent UCL EF behavior in porous Ag/YVO₄:Yb³⁺, Er³⁺ films, the UCL dynamic processes of the excited states ⁴S_3/2 for different doped concentration of Er³⁺ ions were investigated. Figure 5(a) records the decay process of the ⁴S_3/2–⁴I_15/2 transitions of Er³⁺ ions (1 mol%) under the excitation of a 980 nm pulsed laser, which can be well fitted to the single exponential function. The decay time constants for Er³⁺ ions in YVO₄:Yb³⁺, Er³⁺ and Ag/YVO₄:Yb³⁺, Er³⁺ film samples are 11 μs and 9 μs, respectively. The reduction (∼20%) of decay time constant in Ag/YVO₄:Yb³⁺, Er³⁺ composite film can be mainly attributed to the nonradiative relaxation induced by the metal thermal effect. Considering that the porous Ag film had a broad SPR ranging from 320 nm to 1100 nm overlapping with all the emissions of YVO₄:Yb³⁺, Er³⁺ in this range (including ²F_5/2–²F_7/2 of Yb³⁺ at 980 nm) and the decay time constants of the green emission (⁴S_3/2–⁴I_15/2) for Er³⁺ ions are nearly unchanged. This indicates that the remarkable UCL enhancement was mainly induced by the coupling of the 980 nm excitation light with the SPR, rather than an increase in radiative transition rates of Er³⁺/Yb³⁺ ions due to the coupling of surface plasmon absorption of silver with emitted light. Combined the discussions above, the remarkable UCL enhancement originates from the effective coupling between SPR and the excitation light by adjusting the SPR peak to the excitation wavelength, controlling the effective coupling distance and improving the scattering–absorption ratio.

Zoom In Zoom Out Reset image size

Figure 5(b) shows the EF of the ²H_11/2, ⁴S_3/2–⁴I_15/2 transitions for Er³⁺ on doping concentration of Er³⁺ in different sample series, in YVO₄:Yb³⁺, Er³⁺ and Ag/YVO₄:Yb³⁺, Er³⁺ film samples. It can be seen that as the concentration increased from 1 mol% to 8 mol%, EF gradually decreased from ∼36-fold to 17-fold for ²H_11/2–⁴I_15/2 transition and ∼30-fold to 14-fold for ⁴S_3/2–⁴I_15/2 transition compared to the YVO₄:Yb³⁺, Er³⁺ film. EF was proportional to the inverse of Er³⁺ ions concentration, which is due to the increase of the cross relaxation processes among Er³⁺ ions with the increasing of Er³⁺ ions concentration. To clarify the origin of the decrease of EF with increasing of Er³⁺ ions concentration, the dependence of exponential spontaneous decay rates (SDR) of ⁴S_3/2–⁴I_15/2 for Er³⁺ on doping concentration of Er³⁺ in different sample series were presented in figure 5(c). It can be seen that in the Ag/YVO₄:Yb³⁺, Er³⁺ film samples the SDR of ⁴S_3/2–⁴I_15/2 increases rapidly, while in the YVO₄:Yb³⁺, Er³⁺ the SDR of ⁴S_3/2–⁴I_15/2 increases slowly with the increasing Er³⁺ concentration. Based on previous literatures, the SDR of ⁴S_3/2–⁴I_15/2 transition can be written as,

$$\begin{eqnarray*}&&k\propto {{k}_{{\rm R}}}+{{k}_{{\rm NR}}}+{{k}_{{\rm CR}}}[{\rm E}{{{\rm r}}^{3+}}],\end{eqnarray*}$$

where, k, k_R and k_NR represent the total SDR, the radiative decay rate and the nonradiative relaxation rate of ⁴S_3/2, respectively, and k_CR represents the cross relaxation rate from ⁴S_3/2. Based on the above equation, the values of (k_R + k_NR) in YVO₄:Yb³⁺, Er³⁺ and Ag/YVO₄:Yb³⁺, Er³⁺ film samples were deduced to be 75.5 ms⁻¹ and 64.4 ms⁻¹, respectively, while the values of k_CR in YVO₄:Yb³⁺, Er³⁺ and Ag/YVO₄:Yb³⁺, Er³⁺ film samples were deduced to be 13.6 ms⁻¹ mol⁻¹ and 27.8 ms⁻¹ mol⁻¹. In Ag/YVO₄:Yb³⁺, Er³⁺ samples, the local temperature is higher than that in YVO₄:Yb³⁺, Er³⁺ samples due to the absorption by the silver nano-film of the 980 nm excitation light and effective photo-thermal transfer, the cross relaxation processes happened largely in the high temperature. So the EF decreases with the increasing in the concentration of Er³⁺ ions. The simple schematic of UC populating and emission processes in Er³⁺, Yb³⁺-doped YVO₄ was drawn in figure 5(d).

To clarify the origin of luminescent enhancement, first, let us consider the influence of the porous Ag film on the radiative and nonradiative transition rates of Er³⁺ ions. It is apparent that the radiative decay rate in the Ag/YVO₄:Yb³⁺, Er³⁺ film sample almost unchanged relative to YVO₄:Yb³⁺, Er³⁺ film sample, while the value of nonradiative relaxation rate increases about 2 times. So the decrease of decay time constants (20%) could be attributed to the increase of the nonradiative rate, instead of the increase of the radiative rate induced by the emission coupling. Second, let us consider the optical reflection effect of the Ag film on the UCL. The transmittance of the YVO₄:Yb³⁺, Er³⁺ nano-film to the 980 nm light is 68% (measured by absorption spectroscopy), and the reflection of the Ag film to the 980 nm light is 40% and to the visible emissions is about 30%. Thus it can be roughly estimated that due to the reflection of the Ag film, the excitation light increases about 27%, while the emission light increases about 30%. Totally, the visible UCL should increase up to 1.57 times due to the reflection of the Ag film. This fact indicates that the UCL enhancement can be mainly attributed to the excitation filed enhancement induced by the coupling of SPR with the excitation light, instead of the emission coupling.