Photo‐Electrochemical Device Enabling Luminescence Switching of LaPO4:Ce,Tb Nanoparticle Layers

The luminescence properties of LaPO4:Ce,Tb nanoparticles are known to depend on the oxidation state of the cerium ions. However, their assembly into thin films exhibiting reasonable fast Ce3+/4+ electrochemistry is not trivial. Herein, the electrochemical luminescence switching of LaPO4:Ce,Tb nanoparticles, assembled as nonconducting thin films, using two electrocatalytic processes, is demonstrated. Due to the insulating nature of these nanoporous films, redox shuttles are used to access the redox active Ce3+/4+ species for electrochemical reactions. A series of redox shuttles with various redox potentials are employed to investigate their capability to electrochemically oxidize Ce3+ within the nanoparticles. Thereby the formal redox potential of the Ce3+/4+ couple in LaPO4:Ce,Tb nanoparticles is determined to lie within 0.89 and 1.15 V versus Ag/AgCl. In situ observation of repetitive luminescence switching is realized by assembling a device that allows UV light to enter the nanoparticle layer. With two redox shuttles present in the electrolyte, one for the oxidation of Ce3+ and the other for reduction of Ce4+, quenching and restoration of the luminescence is monitored. The resulting device represents the first down‐sizable logical AND gate with UV light and voltage inputs and a vis light output based on a solid state LaPO4:Ce,Tb layer.

The luminescence properties of LaPO 4 :Ce,Tb nanoparticles are known to depend on the oxidation state of the cerium ions. However, their assembly into thin films exhibiting reasonable fast Ce 3+/4+ electrochemistry is not trivial. Herein, the electrochemical luminescence switching of LaPO 4 :Ce,Tb nanoparticles, assembled as nonconducting thin films, using two electrocatalytic processes, is demonstrated. Due to the insulating nature of these nanoporous films, redox shuttles are used to access the redox active Ce 3+/4+ species for electrochemical reactions. A series of redox shuttles with various redox potentials are employed to investigate their capability to electrochemically oxidize Ce 3+ within the nanoparticles. Thereby the formal redox potential of the Ce 3+/4+ couple in LaPO 4 :Ce,Tb nanoparticles is determined to lie within 0.89 and 1.15 V versus Ag/AgCl. In situ observation of repetitive luminescence switching is realized by assembling a device that allows UV light to enter the nanoparticle layer. With two redox shuttles present in the electrolyte, one for the oxidation of Ce 3+ and the other for reduction of Ce 4+ , quenching and restoration of the luminescence is monitored. The resulting device represents the first down-sizable logical AND gate with UV light and voltage inputs and a vis light output based on a solid state LaPO 4 :Ce,Tb layer.
ensures adhesion of the film during electrochemical measurements, the temperature of 200 °C is not enough to remove the polymer from the film as shown by thermogravimetric analysis (TGA) of the composite ( Figure S2, Supporting Information). Due to the nonconducting properties of the nanoparticle films, electrochemical oxidation of Ce 3+ species through polarization of the electrodes at up to 2 V versus Ag/AgCl in 0.1 m LiClO 4 in acetonitrile could not be achieved. To enable redox reactions of the cerium atoms, redox shuttles were added to the electrolyte to transfer electrons between the current collector and the nanoparticles. As shown in Scheme 1a the films are nanoporous in the sense that the redox shuttles are able to percolate the film composed of the nanoparticles embedded in a polymer matrix. For the electrochemical oxidation of Ce 3+ species, redox shuttles are oxidized at the current collector followed by migration into the nanoparticle film. If the redox potential of the redox shuttle (E S 0 ) is more positive than the redox potential of the Ce 3+/4+ couple (E NP 0 ), electrons are transferred from Ce 3+ ions of the LaPO 4 :Ce,Tb nanoparticles to the oxidized redox shuttles. Thus, Ce 4+ species are generated while the previously oxidized redox shuttles are reduced to their initial state and therefore can again undergo oxidation at the current collector, yielding a socalled catalytic current. This mechanism is schematically shown in Scheme 1b. In cyclic voltammetry (CV) measurements (three electrode setup), this electron transfer mechanism results in an increased oxidation current in the first cycle as shown in Figure 1a. Accordingly, the luminescence of the nanoparticle film is fully quenched. The electrochemical reduction of Ce 4+ species follows the same principle when an electrochemically reducible redox shuttle is used (Scheme 1c). Here E S 0 has to be more negative than E NP 0 . In this case, an increased reduction current is observed in the first cycle of CV measurements (Figure 1b), leading to complete restoration of the luminescence. Further CVs where such kind of catalytic currents were observed are shown in Figures S3-S10, Supporting Information. The excess charge between the first and second cycle along with the amount of nanoparticles present in the film allows calculation of the quantity of oxidizable/reducible Ce 3+/4+ to ≈2.2%. Another important quantity is the formal redox potential of cerium species at the surface of LaPO 4 :Ce,Tb nanoparticles, especially since the redox potential of cerium is known to be strongly influenced by its chemical environment. Therefore, the capability to electrochemically oxidize Ce 3+ was investigated for redox shuttles with different E S 0 (molecular structures are shown in Figure S11, Supporting Information). This variation of E S 0 shows that E NP 0 has a value between 0.89 V and 1.15 V versus Ag/AgCl ( Figure 1c and Table S1, Supporting Information). Prior to these measurements, the redox potentials E S 0 of the redox shuttles were determined from CV measurements using a blank ITO working electrode. Repetitive electrochemical oxidation and reduction can now be achieved with two redox shuttles present in the electrolyte, one that oxidizes Ce 3+ and one that reduces Ce 4+ . In the experiments mentioned so far, the nanoparticle films had to be removed from the electrochemical cell after the measurements to check the luminescence under  the UV light. To realize an in situ observation of the repetitive luminescence switching of the nanoparticles, we built a full thin-film electrochemical device consisting of the nanoparticle/polymer composite layer on ITO coated glass, a separator, the counter electrode and a quartz glass window (Figure 2a). For UV excitation of the composite layer, a UV transmitting electrolyte/redox shuttle combination had to be used. A 0.15 m solution of LiClO 4 in acetonitrile containing 1,4-di-t-butyl-2,5-dimethoxybenzene (2 mm) as oxidizing redox shuttle and 7,8-dihydro-6H-dipyrido[1,2-a:2′,1′-c]- [1,4]diazepinium (0.25 mm) as reducing redox shuttle fulfills this requirement, at least for short path ways. Potentials of 2 V for oxidation and −1 V for reduction were repeatedly applied in intervals of 1 s while UV light was cast through the quartz glass onto the nanoparticle film. Images of the device in the on/off state are displayed in Figure 2b,c. A movie showing the switching was recorded (available as Supporting Information) from which the intensity change of the nanoparticle film was monitored as a function of time using ImageJ software. Repetitive quenching and restoration of the luminescence was achieved for multiple switching cycles (Figure 2e). From the time resolved intensity change over one switching cycle, displayed in Figure 2d, the switching times to achieve a 90% change in intensity can be estimated as ≈550 ms for quenching and ≈650 ms for restoration of the luminescence. The fast switching times and the complete quenching of the luminescence indicate that the mediators are able to rapidly access most LaPO 4 :Ce,Tb nanocrystals of the layer. In fact, switching remains possible even when the layer of LaPO 4 :Ce,Tb particles is diluted with an inert solid nanomaterial. Here, we used LaPO 4 :Eu nanocrystals, since this material cannot be oxidized or reduced within our potential range. In our diluted device, the nanoparticle layer consisted of a 1:19 mixture of LaPO 4 :Ce,Tb and LaPO 4 :Eu particles of which the latter alone shows a red color under UV excitation. The device therefore displays a yellow color in the reduced state (−1 V) and red color when cerium species are oxidized (+2 V) and shows that not only on/off-switching but also switching between two different emission colors is possible (see Figure 3).  Furthermore, this indicates that even in this diluted system, with respect to LaPO 4 :Ce;Tb, the redox shuttles reach all nanoparticles within the layer and electrons are not transferred via adjacent LaPO 4 :Ce;Tb particles. With the applied voltage and UV light as inputs and visible light as output the device based on LaPO 4 :Ce,Tb represents a logical AND gate ( Figure S12, Supporting Information). Our results also show that neither high temperature sintering processes nor the addition of solid conductive additives are required to utilize the redox-dependent optical properties of nonconductive nanoparticles in an electrochemical device. The redox shuttle-based working principle of our device may therefore enable the use of a large variety of different functional materials in opto-electrochemical devices.
In summary, the electrochemical luminescence switching of LaPO 4 :Ce,Tb nanoparticle thin films is demonstrated by a mediated electron transfer mechanism via small organic molecules (redox shuttles) present in the electrolyte. The formal redox potential of the Ce 3+/4+ species within the nanoparticles was determined within a narrow range by using redox shuttles with different redox potentials. Construction of a device containing an oxidizing and a reducing redox shuttle allowed in situ observation of repetitive luminescence on/off switching.
Film Preparation: Nanoparticle films were prepared on ITO coated glass, which was cleaned by subsequent ultrasonication in mucasol, water and acetone. Pluronic F127 (12 600 g mol −1 , molar nanoparticle/polymer ratio 1/0.005) was added to 1 mL of the methanolic nanoparticle solution (0.16 m). 15 µL of this mixture was drop casted onto the ITO substrate (1 cm 2 coated area) followed by heating the electrodes to 200 °C for 30 min. TGA of the composite and its individual components were performed on a NETZSCH STA 449 system with helium atmosphere.
Electrochemistry: Electrochemical measurements were performed in a three electrode system with an Ag/AgCl reference electrode and a platinum wire counter electrode. A 0.1 m solution of LiClO 4 in acetonitrile served as electrolyte. Redox shuttles (structures shown in Figure S11, Supporting Information) were added to the electrolyte to a concentration of 1 mm (separate solutions). To perform electrochemical reduction of Ce 4+ species, the nanoparticle films were chemically oxidized by immersion of the electrodes into an aqueous solution of KMnO 4 (1 mm) prior to electrochemical measurements. Cyclic voltammograms of the nanoparticle films in the electrolytes containing different redox shuttles were recorded with an Autolab PGSTAT 20 potentiostat. After the measurements, the electrodes were removed from the electrochemical cell and the success of the electrochemical luminescence switching was qualitatively checked under a UV light (254 nm).
Device Fabrication: On a 2.5 × 3 cm ITO glass, a nanoparticle film (1 × 1 cm) was prepared based on the above described method with a fourfold diluted solution. A 2.5 × 3 cm carbon plate with a 0.5 × 0.5 cm cavity (overlapping with the nanoparticle film) was used as counter electrode and separated from the ITO glass by a 50 µm gasket. A 2.5 × 2.5 cm quartz glass was placed on top of the carbon plate to close the device and allow entrance of UV light into the cell. Via vacuum backfilling, the device was filled with a 0.15 m solution of LiClO 4 in acetonitrile containing the redox shuttles 1,4-di-t-butyl-2,5dimethoxybenzene (2 mm) and 7,8-dihydro-6H-dipyrido[1,2-a:2′,1′-c]- [1,4] diazepinium (0.25 mm). The device was switched between 2 V and −1 V in 1 s intervals. A second device was built identically but the nanoparticle layer was prepared from a combined solution consisting of 95% of a methanolic solution (0.16 m) of 4 nm LaPO 4 :Eu particles and 5% of a methanolic solution (0.16 m) of 4 nm LaPO 4 :Ce,Tb particles. Pluronic F127 was added as described above. 30 µL of this undiluted mixture were drop casted onto the ITO (1 cm 2 coated area). Here too, potentials of −1 V and +2 V were applied but not in 1 s intervals.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.