A Kinetic Isotope Effect in the Formation of Lanthanide Phosphate Nanocrystals

Abstract

Mechanisms of nucleation and growth of crystals are still attracting a great deal of interest, in particular with recent advances in experimental techniques aimed at studying such phenomena. Studies of kinetic isotope effects in various reactions have been useful for elucidating reaction mechanisms, and it is believed that the same may apply for crystal formation kinetics. In this work, we present a kinetic study of the formation of europium-doped terbium phosphate nanocrystals under acidic conditions, including a strong H/D isotope effect. The nanocrystal growth process could be quantitatively followed through monitoring of the europium luminescence intensity. Hence, such lanthanide-based nanocrystals may serve as unique model systems for studying crystal nucleation and growth mechanisms. By combining the luminescence and NMR kinetics data, we conclude that the observed delayed nucleation occurs due to initial formation of pre-nucleation clusters or polymers of the lanthanide and phosphate ions, which undergo a phase transformation to crystal nuclei and further grow by cluster attachment. A scaling behavior observed on comparison of the H₂O and D₂O-based pre-nucleation and nanocrystal growth kinetics led us to conclude that both pre-nucleation and nanocrystal growth processes are of similar chemical nature.

Figure 1

(a) Luminescence spectrum (excitation wavelength = 365 nm) of a colloidal suspension of Eu³⁺-doped TbPO₄·D₂O NCs with 2:1 PO₄^3–:Ln³⁺ precursor ratio. Emission intensity was normalized to the strongest peak. (b) Luminescence of Eu³⁺ in D₂O and H₂O-based NC synthesis solutions over time at 50 °C, measured at 704 nm. The 704 nm emission line is attributed to the ⁵D₀ → ⁷F₄ transitions of the Eu³⁺ ion. (c) Luminescence of Eu³⁺ in D₂O and H₂O-based NC synthesis solutions over time at 40 °C, measured at the same wavelength. The insets of panels (b) and (c) show the expanded initial stages of the reaction. All other reaction conditions were identical, and the final concentrations of the NCs were similar, as verified by their luminescence intensities. The curves in panels (b) and (c) were normalized to the emission intensity at the end of the measurement.

Figure 2

(a,b) Scaling behavior of the normalized emission intensity of Eu³⁺ of D₂O and H₂O solutions with a 2:1 PO₄^3–:Ln³⁺ precursor ratio, at 50 and 40 °C, where the H₂O time axis is multiplied by factors of 4.5 and 10.5 for the curves taken at 50 and 40 °C, respectively.

Figure 3

Induction period as a function of volume fraction of H₂O in D₂O solution with a 2:1 PO₄^3–:Ln³⁺ precursor ratio at 50 °C. The inset shows the lower water concentration regime expanded. The lines are hand-drawn as guides for the eye. Induction times were measured twice for some of the concentrations, to provide an estimate of the uncertainty level, which seems to be roughly of the order of ∼30% of the induction time value.

Figure 4

³¹P NMR results for the NC growth process with a 1:1 PO₄^3–:Ln³⁺ precursor ratio performed in D₂O solution at 50 °C. (a) Eu³⁺ luminescence vs time curve with points indicating sampling times for the NMR experiments. (b) ³¹P peak chemical shift and peak area vs time for the experiment shown in panel (a). The peak integral was normalized to the peak area at t = 0.

Figure 5

Luminescence vs time measured at 50 °C in D₂O, following addition of two different quantities of NC seed particles (20 and 100 μL of seed solution), compared with an unseeded process. Inset: expanded view of the first 7 min of the 100 μL seeded growth process (dots) together with the fitted Finke–Watzky model (red line).