Organic melt, electride, and CVD induced in situ deposition of luminescent lanthanide imidazolate MOFs on nanostructured alumina

Abstract

The highly luminescent MOFs ³_∞[Ce(Im)₃ImH]·ImH, ³_∞[Tb(Im)₃], and the dense framework ³_∞[Sr_0.95Eu_0.05(Im)₂] (Im⁻ = imidazolate-anion, C₃N₂H₃⁻) were grown on nanostructured macroporous aluminium oxide (AAO) membranes. Thereby, luminescent coatings on the membranes were achieved. Three different growth processes unusual for MOFs were investigated and compared: a film growth process in situ to MOF formation from the linker melt, electride induction with solvated electrons, and chemical vapour deposition (CVD) to additionally utilize the gas phase. Deposition from the organic melt has proved to be a fast approach to achieve various film thicknesses of the luminescent frameworks. The electride-based approach offers excellent homogenization at an atomic level for the highest quantum yields of QY > 90% for ³_∞[Sr_0.95Eu_0.05(Im)₂] including the formation of barite rose analogous crystals prior to growth of a complete film on AAO membranes. For ³_∞[Tb(Im)₃] and ³_∞[Ce(Im)₃ImH]·ImH, deposition of bundles of crystals by CVD on AAO is possible while also maintaining the luminescence of the original MOFs but without complete layers. In order to elaborate the divalent character of europium, being the basis of the high efficiency of the luminescence, EPR studies were carried out on ³_∞[Sr_1−xEu_x(Im)₂], x = 0.01, 0.05 and 1.

---

Introduction

Metal organic frameworks (MOFs) exhibit a large variety of properties or even a combination of them, which have given inspiration in the field of materials science for almost 25 years.^1–8 By variation of the linker molecules, metal centres, and template molecules, MOFs allow the modification as compared e.g. to zeolites. The use of lanthanides as metal centres leads to hybrid materials with intense photoluminescence properties,^9–12 and thin luminescent films of MOFs and COPs¹³ (coordination polymers) can also be generated. The development and processing of MOF films is a recent and rapidly growing research area,^14–16 as these films are of interest for sensing and membrane technologies.^17–21 Both established and newly developed coating techniques are investigated: inspired by the generation of coordination polymers and zeolitic films, seeded-growth, Langmuir–Blodgett and liquid phase epitaxy as well as layer-by-layer methods were applied with great success.^22–24 The main approach to form MOF films is deposition from liquid phases like solutions, whereas CVD and PVD are exceptions due to the intrinsic low vapor pressure and limited thermal stability of the frameworks.^{14–16,23,24} Almost all individual frameworks exhibit different chemical behavior as well as stability and therefore an individual film growth is necessary. Due to the lack of common applicability the controlled SBU (secondary building unit) approach (CSA) is a recent effort towards generalisation.²⁵ A multitude of substrates such as metals, metal oxides, and graphite have already been successfully used to grow MOFs and coordination polymers upon them.^26–28 We recently demonstrated that anodic aluminium oxide (AAO) as a patterned surface can be used to generate thin luminescent films of luminescent dense frameworks.¹³ Luminescent thin MOF-films represents a new topic that combines the photoluminescence of a MOF with a structured surface that can potentially increase the energy efficiency of the luminescent framework.²⁹ This is useful for potential applications in lighting techniques as well as for sensing.

Recently, we have shown that in situ framework/film formation from an imidazole melt and elemental metals like strontium and europium yields thin and highly luminescent MOF films of ³_∞[Sr_0.95Eu_0.05(Im)₂] on macroporous, self-ordered AAO.¹³

Herein, we elaborate the in situ coating approach from the linker melt for the luminescent MOFs ³_∞[Ce(Im)₃ImH]·ImH and ³_∞[Tb(Im)₃] maintaining the MOF luminescence. We provide insights into homogenization of Sr and Eu at an atomic level by an electride approach in liquid ammonia to increase the luminescence efficiency, and elaborate the paramagnetic properties of the bulk material by EPR spectroscopy. Finally, we show options on the deposition of ³_∞[Ce(Im)₃ImH]·ImH and ³_∞[Tb(Im)₃] by CVD from the gas phase on AAO.

Results and discussion

In situ deposition of ³_∞[Ce(Im)₃ImH]·ImH (1), ³_∞[Tb(Im)₃] (2) and ³_∞[Sr_0.95Eu_0.05(Im)₂] (3) on AAO membranes from an imidazole melt and by CVD

The starting point for in situ coating investigations of nanostructured macroporous anodic aluminium oxide (AAO)‡ with the luminescent MOFs ³_∞[Ce(Im)₃ImH]·ImH³⁰ (1) and ³_∞[Tb(Im)₃]³¹ (2) was the initial growth of ³_∞[Sr_0.95Eu_0.05Im₂] (3) on AAO substrates from an imidazole melt according to ref. 13 in an approach that combines the framework and film formation. Both MOFs the cerium imidazolate 1 and the terbium imidazolate 2 can be synthesized from the metals and a melt of imidazole, which renders coating with them accessible through this procedure, in which the organic melt is in contact with the AAO substrate during framework formation. The MOFs are microporous, with 1 showing a preferred uptake of CO₂ of 160 cm³ g⁻¹ and a moderate surface area of S_BET = 480 m² g⁻¹ at −75° C,³⁰ and 2 preferring adsorption of NH₃ (92 cm³ g⁻¹ = 4.5 mass%) at 4.5 bar NH₃ and room temperature (see Fig. S1, ESI†).

More intriguing, the MOFs show interesting luminescence behaviour upon excitation with UV-light. ³_∞[Ce(Im)₃ImH]·ImH³⁰ (1) is a blue emitting MOF with prominent, parity allowed and therefore strong 5d→4f emission of Ce³⁺, whereas ³_∞[Tb(Im)₃]³¹ (2) shows the typical green emission of 4f–4f transitions of Tb³⁺ fed by a strong antenna effect of imidazole, which functions as a sensitizer for Tb³⁺ circumventing the low light uptake due to the parity forbidden character of intra-4f transitions.

AAO consists of arrays of aligned cylindrical macropores with narrow diameter distribution and uniform macropore depth, which are characterized by local (“polycrystalline”) hexagonal ordering.^32,33 Topographically patterned AAO substrates with defined macropore size as well as geometry were chosen to check the possible structuring effects of the MOF films. In order to further develop and promote the concept of the in situ coating from an organic melt, the lanthanide-based frameworks 1 and 2 were chosen to generate thin luminescent films on self-ordered AAO substrates while retaining their original bulk properties like crystal structure and photoluminescence. As shown for the deposition of ³_∞[Sr_0.95Eu_0.05Im₂] films on AAO membranes, the pore walls exhibit a suitable number of coordination sites – mostly OH – for the highly oxophilic terbium and cerium ions and thus the corresponding MOFs ³_∞[Tb(Im)₃] and ³_∞[Ce(Im)₃ImH]·ImH. In order to utilize and investigate possible layer thicknesses and layer formation mechanisms, AAO membranes with a pore diameter ranging from 40 to 400 nm and a pore depth of 100 μm were investigated.

The use of the MOFs ³_∞[Tb(Im)₃] as well as ³_∞[Ce(Im)₃ImH]·ImH yields luminescent layers with variable thickness. The initial growth of 1 and 2 starts at the macroporous mouths of the substrate channels (Fig. 1a). As previously envisaged,¹³ the deposition of the MOFs depends on substrate parameters, such as the surface potential of the pore walls, the number of anchoring sites, and the size of the substrate apertures. In addition, MOF parameters like the crystal cell extensions, surface groups, etc. will also influence the growth on the substrate.

Fig. 1 (a) Initial growth of ³_∞[Ce(Im)₃ImH]·ImH (1) at the pore mouths of a 300/100 AAO substrate from the melt; (b) crystalline closed film layer of 1 on a 300/100 AAO substrate; (c) initial growth of ³_∞[Tb(Im)₃] (2) at the pore mouths of a 400/100 AAO substrate; (d) growth of a thick layer of ³_∞[Tb(Im)₃] (140 μm) on a 40/100 AAO substrate.

Structured thin films can be grown on substrates with large nanostructuring and apertures of 300 and 400 nm in diameter, which are subsequently forming closed layers upon further growth, whereas small apertures of 40 to 100 nm directly give closed films that show some crystal displacement if grown to thick films on the micrometer scale (for both, see Fig. 1 and Scheme 1).

Scheme 1 (a) Initial growth of ³_∞[Ce(Im)₃ImH]·ImH (1) at the pore walls inside the channels during the melt approach; (b) subsequent overgrowth of the AAO pore mouths with the MOF.

In order to elaborate the possibility of a CVD deposition of both MOFs from the melt approach a setup was built that prevented direct contact between the melt and the AAO substrate (see Fig. 2). SEM investigations show that isolated crystalline particles are grown on the AAO surface without formation of films (Fig. S9–S11, ESI†). The size of the MOF crystals grown by CVD ranges from 1 to 3 μm for ³_∞[Ce(Im)₃ImH]·ImH (Fig. S9, ESI†) and 1–20 μm for ³_∞[Tb(Im)₃] (see Fig. S11, ESI†).

Fig. 2 Experimental setup (left) for the CVD deposition of ³_∞[Tb(Im)₃] (top) and ³_∞[Ce(Im)₃ImH]·ImH (bottom) on AAO.

EDX element mapping analysis was carried out to prove the composition of the crystalline depositions and show that all elements of the original frameworks ³_∞[Tb(Im)₃] as well as ³_∞[Ce(Im)₃ImH]·ImH were present (Fig. 3 and S12†). In order to check that the films retained their bulk properties, powder diffraction and photoluminescence investigations were also carried out on the coated AAO substrates. Comparison of the CVD deposition with the films generated by melt contact shows the chemical identity of both depositions.

Fig. 3 (a) EDX spectrum of CVD-deposited crystals of ³_∞[Tb(Im)₃] (2) on AAO; (b–f) EDX element mappings for Tb, N, C, Al and O.

The diffraction patterns of the AAO substrates coated with ³_∞[Ce(Im)₃ImH]·ImH as well as with ³_∞[Tb(Im)₃] and the corresponding bulk material show identical reflections (see Fig. S3 and S4, ESI†). The powder patterns also match with single crystal X-ray data.^30,31 It can be concluded that the coating of AAO substrates was successful and that deposition is possible by both, the in situ melt approach and CVD. However, only the contact between the melt and the substrate can be used to generate films.

The photoluminescence spectra of AAO substrates coated with 1 as well as of the bulk material are shown in Fig. 4 (top). SEM investigations show that the deposition of ³_∞[Ce(Im)₃ImH]·ImH (1) on AAO substrates leads to a lower amount of deposited material compared to 2 providing the same deposition parameters and therefore thinner layers in the in situ melt approach. As the low amount of deposited substance complicates analytics that require a certain amount of material like X-ray powder diffraction, a long time X-ray investigation was applied to generate suitable X-ray data. It proves that growth maintains the crystal structure (see Fig. S4, ESI†). Photoluminescence spectroscopy was also utilized as further proof of the successful coating of the AAO substrates with ³_∞[Ce(Im)₃ImH]·ImH.

Fig. 4 Excitation (left) and emission spectra (right) of AAO membranes coated with ³_∞[Ce(Im)₃ImH]·ImH (top), ³_∞[Tb(Im)₃] (bottom), and the corresponding bulk material.

The luminescence of Ce³⁺ originates from parity allowed 4f–5d transitions for excitation and vice versa 5d–4f transitions for emission. The 5d–4f transitions with energy gaps <33 000 cm⁻¹ by crystal and ligand field splitting are an exception observed for Eu²⁺ and Ce³⁺.³⁴ Due to the participation of d-states they are directly influenced by chemical surroundings of the Ce³⁺ ions. This prominent feature is demonstrated by the Y₃Al₅O₁₂ : Ce³⁺ laser (YAG : Ce), which emits in the yellow region.³⁵ As shown for Eu²⁺ above and ref. 13 this type of luminescence can be used as a fingerprint for the chemical surroundings. Almost identical excitation and emission spectra can be observed for the ³_∞[Ce(Im)₃ImH]·ImH coated AAO substrate and the bulk material, with no remarkable shift in excitation and emission as well as no modification of band half-width observed. It is therefore concluded that the coating of AAO membranes with ³_∞[Ce(Im)₃ImH]·ImH was successful.

For the MOF ³_∞[Tb(Im)₃] (2), the coated substrates as well as the bulk materials exhibit the previously described broadband excitation spectra based on the linking imidazolate anion.³¹ The observed emission spectra for coated substrates as well as for the bulk material exhibit sharp 4f–4f-transitions typical for the Tb³⁺ transitions between the excited ⁵D₄ state and the ⁷F₆₋₀ levels of Tb³⁺ ground states (Fig. 4, bottom). The quantum yield of 2 was also determined in this study with QY = 23% (investigated vs. BAM : Eu²⁺ as the standard).

Alike MOF 1, the photoluminescence properties of deposited 2 match with the bulk properties. This is also in accordance with the results of X-ray powder diffraction, which also proves the identity of bulk and deposited framework 2. ³_∞[Tb(Im)₃] was thereby successfully grown on AAO substrates retaining both crystal structure and material properties.

In order to also investigate time-dependent changes of grown layers by the in situ coating process, ³_∞[Sr_0.95Eu_0.05(Im)₂] (3) films were grown on 400/100-AAO membranes (pore diameter: 400 nm, pore depth: 100 μm) according to ref. 13 with three different annealing times for the imidazole melt contact (1, 12 and 48 hours). After the annealing, film growth was stopped. It can be considered that a longer annealing time results in formation of thicker layers. However, the time dependence has not previously been elaborated. SEM investigations show that the initial growth on the pore mouths and inside the pore apertures can be assigned to an annealing for a short time like 1 h. 12 h annealing leads to a thin continuous layer of 3, and longer annealing times lead to further growth of crystalline layers of several micrometers (Fig. 5). Accordingly, we can now point out that annealing and melt contact for about one hour can be used to gain the initial growth stage; 12–15 hours annealing gives continuous thin films on the nanometer scale, whereas longer growth is not useful to achieve a thin film but gives further crystalline growth.

Fig. 5 Time dependent deposition of ³_∞[Sr_0.95Eu_0.05(Im)₂] on AAO, from initial growth inside the substrate channels via formation of continuous thin layers to thick crystalline growth (time scale: annealing times 1 h, 12 h and 48 h, from the left to the right).

In order to check if bulk properties are maintained, powder diffraction and photoluminescence investigations were carried out on ³_∞[Sr_0.95Eu_0.05(Im)₂] coated AAO substrates (see Fig. 6 and S5, ESI†). Comparison of powder patterns of the bulk material with the coated AAO membrane points out that all reflections of ³_∞[Sr_0.95Eu_0.05(Im)₂] can be found with different intensity ratios without a preferred crystal orientation (see S2, ESI†).

Fig. 6 Excitation and emission spectra of ³_∞[Sr_0.95Eu_0.05(Im)₂] coated 400/100-AAO membranes and the bulk material after an annealing time of 12 hours, respectively.

As observed for the luminescent MOFs, ³_∞[Ce(Im)₃ImH]·ImH³⁰ (1) and ³_∞[Tb(Im)₃]³¹ (2) films of 3 retain the original framework luminescence. A broadband excitation which ranges from λ = 250–460 nm is the basis of the intense photoluminescence of the framework ³_∞[Sr_0.95Eu_0.05(Im)₂] and is caused by excitation of both the ligand 1H-imidazole and europium(II). The resulting emission exhibits an exceptionally high quantum yield for MOFs, previously reported with 80%.³⁶

Broadband emission is typical for divalent europium and is based on transitions between 5d and the 4f⁸S_7/2 energy levels. This luminescence can be utilized as a fingerprint for the chemical surrounding due to the participation of the 5d energy levels. Comparison between the excitation and the emission spectra of ³_∞[Sr_0.95Eu_0.05(Im)₂] films on AAO substrates and the spectra of the corresponding bulk material shows almost identity except for a slight shift of the excitation band. Emission maxima, profiles and half-width are identical (Fig. 6 and Fig. S5, ESI†). Both analytical methods demonstrate that the dense framework ³_∞[Sr_0.95Eu_0.05(Im)₂] has been successfully grown on 400/100-AAO substrates maintaining the crystal structure as well as its photoluminescence properties.

In situ deposition of ³_∞[Sr_0.95Eu_0.05(Im)₂] (3) on AAO membranes by electride induction from liquid NH₃

We have recently shown that thin films of the dense framework series ³_∞[Sr_1−xEu_x(Im)₂] (x = 0–1, Im⁻ = imidazolate anion) can be achieved by the new in situ coating approach from organic melts on nanostructured macroporous AAO‡ substrates.¹³ Thereby, a one-pot-synthesis of non-noble alkaline earth and lanthanide metals with 1H-imidazole and AAO substrates (pore diameter of 40–400 nm, pore depth 100 μm) was utilized. The nanostructured AAO^32,33 substrates were chosen to check for possible structuring effects of the MOF films. Additionally, an electride induced in situ growth was shown once again from liquid ammonia exhibiting a different growth behaviour from the melt approach that excluded the 400 nm wide channels from growth without explanation.¹³ We have now elaborated the initial growth process of the electride induced procedure and can now explain this behaviour.

The synthesis of the dense co-doped framework ³_∞[Sr_0.95Eu_0.05(Im)₂] was carried out in liquid ammonia according to ref. 36. The ratio of strontium/europium of 95 : 5 was chosen because it is known to give the highest quantum yield of >80%, which is an exception among coordination polymers.³⁶ The use of liquid ammonia aims at mixing of the metals strontium and europium at the atomic level, as both can be dissolved in liquid ammonia.^37–40 Electride induced in situ coating approaches with the dense framework ³_∞[Sr_0.95Eu_0.05(Im)₂] (3) and 400/100-AAO membranes (pore diameter: 400 nm, pore depth: 100 μm) have been repeatedly carried out followed by SEM investigations on the obtained products. Initial growth of the dense framework 3 on AAO is accompanied by extraordinarily shaped crystallites that look like barite roses in shape (see Fig. 7).

Fig. 7 Barite rose analogous crystals of ³_∞[Sr_0.95Eu_0.05(Im)₂] (3, left) and the growth of a continuous layer by the electride coating approach (right).

This remarkable crystal shape can only be found on the AAO surface and is never observed in the bulk material. Natural occurrences of these bizarre crystals are known as barite roses (desert roses, sand roses) that can be found in deserts (North America, North Africa, and Mexico), or generally in regions with arid climate conditions.⁴¹ The formation of sand roses is the result of a very rapid evaporation of surface water which requires the promotion of groundwater via capillary forces. Salts dissolved in the water crystallise and interact with the grains of sand forming the characteristic foliaceous structures.⁴¹ It is therefore likely that in the electride induced system used here, the formation of the barite analogous crystals can be occurring because of a rapid evaporation of liquid ammonia during the in situ coating process. Initial crystallisation processes of the framework on the channel wall intersections of the AAO membranes together with the fast evaporation of liquid ammonia can thereby cause formation of our barite rose analogous crystals (see Scheme 2).

Scheme 2 Initial growth of the barite rose analogous crystals of ³_∞[Sr_0.95Eu_0.05(Im)₂] at the intersection of the channel walls of the AAO substrates.

SEM investigations allow the conclusion that the formation of the barite rose analogues of ³_∞[Sr_0.95Eu_0.05(Im)₂] is the growth following seeding (Fig. 7). Formation of a continuous layer of 3 consumes the barite rose analogues, which cannot be seen thereafter anymore. The concave macropore walls of the macropore apertures act as anchoring sites for the metal ions of 3 giving the barite roses from a low number of crystalline seeds, whereas the growth of the MOF resulting from the melt approach extracts into the apertures of AAO (see Fig. S6, ESI†).

EDX element mapping analyses were also carried out and show that all elements of the original framework 3 can be found (Fig. 8). In a previous report¹³ we identified a difference in the ratio of Sr : Eu between bulk and deposited layers from the imidazole melt, on the one hand, and a deposited layer, if an electride is used, on the other hand, without being able to offer an explanation. We can now explain this difference, as point-ID measurements of the barite rose analogous crystals indicate exactly this difference. The ratio of strontium/europium is lower than the original ratio of the reactant metals, with the barite rose analogous crystals exhibiting a higher europium content than the original ratio of strontium/europium (see Fig. S7 and S8, Table 1, ESI†). Accordingly, the formation of the barite roses is the source for the change in the metal ratio. This can originate from different crystallization conditions that favour europium in the formation of this extraordinary crystal shape.

Fig. 8 EDX spectrum of barite rose analogous crystals of ³_∞[Sr_0.95Eu_0.05(Im)₂] (3) deposited on AAO substrates (top); (a–f) EDX element mappings for Eu, N, C, Sr, Al and O.

Additional photoluminescence studies and EPR spectroscopy of ³_∞[Sr_0.95Eu_0.05(Im)₂] (3)

The electride induced approach to mix the dissolved metals at an atomic level can have beneficial effects on the luminescence properties as it can avoid clustering of luminescence centers and thereby reduce quenching by concentration. This effect was successfully shown for a hydride based synthesis of the yellow emitting barium imidazolate co-doped with Eu^II,⁴² which exceeds the mixing of metal particles, but is still beyond the atomic level. We now went one step further to dissolution in liquid ammonia and thereby re-investigated the efficiency of the luminescence process accompanied by this ideal mixing. It is intriguing that the efficiency is indeed increased. For a content of 95% strontium and 5% europium, the quantum yield (determined vs. BAM : Eu²⁺ as the standard and absolutely) was previously determined as QY = 78% (absolute) to 83% (relative QY).³⁶ For an isoconstitutional framework from the electride approach with the same Sr : Eu ratio of 95 : 5, determination of the quantum yield gives an increase up to QY = 90% (absolute) to 92% (relative QY) at an absorption of 0.7, which marks this co-doped dense framework 3 as one of the most efficient luminescent coordination polymers today, the rise in efficiency being combined with an efficient distribution of the luminescence centers. As this efficiency is combined with the parity allowed photoluminescence of Eu^II, which is less stable than Eu^III, we also wanted to prove the divalence of europium. Moreover, a non-luminescent phase with additional coordinating imidazole molecules is known to be present in the system Eu/ImH. Due to the lack of luminescence, the presence of this compound, even as a minority phase, reduces the efficiency of the PL of a product from europium and imidazole and should be avoided.

We therefore employed electron paramagnetic resonance (EPR) spectroscopy to investigate the incorporation of paramagnetic S-state Eu^II ions with a total electron spin S = 7/2 into ³_∞[Sr_1−xEu_x(Im)₂]. We have investigated three samples with different ratios of Sr : Eu, x = 0.01, 0.05, and 1. Their room temperature Q-band EPR powder spectra are displayed in Fig. 9. (For further X-band and low temperature spectra see Fig. S13–S21, ESI.†) The spectrum of the sample with x = 0.01 having the lowest Eu concentration (Fig. 9a) exhibits the best spectral resolution with a number of well resolved lines. They are caused by the zero field splitting (zfs) of Eu^II in a distorted octahedral crystal field,⁴³ which leads to line splitting of about 300–500 G and a considerable smaller splitting of about 40 G due to the hyperfine interaction (hfi) with the ¹⁵¹Eu and ¹⁵²Eu nuclei both with a nuclear spin I = 5/2. Already at Sr : Eu ratios of 95 : 5 significant line broadening effects are observed in the Eu^II EPR spectra due to magnetic dipole–dipole interactions among the Eu^II ions (Fig. 9b) and consequently the ¹⁵¹Eu and ¹⁵²Eu hfi cannot be resolved anymore. For the magnetically non-diluted sample with x = 1 even the Eu^II zfs is not observable anymore and the spectrum consists of a single symmetric line (Fig. 9c). It has an almost Lorentzian line shape which is indicative of the presence of substantial magnetic exchange interactions among the Eu^II ions in that sample. Spectral simulations of the EPR spectra of the samples with x = 0.01 and 0.05 reveal that only a single Eu^II species A exists in the x = 0.01 sample with orthorhombic symmetry (for details of the spectral simulation and spin Hamiltonian parameters see ESI†) in accordance with the distorted octahedral metal site in ³_∞[Sr_1−xEu_x(Im)₂]. Otherwise in the case of the sample with x = 0.05 the simulations indicate a superposition of the spectrum of species A with an additional spectrum of a second minor Eu^II species B, again with orthorhombic symmetry. Both Eu^II species A and B differ mainly in their zfs parameters B⁰₂ (310 MHz for A versus 480 MHz for B) which become slightly smaller at low temperatures.

Fig. 9 Eu^II Q-band EPR spectra of ³_∞[Sr_1−xEu_x(Im)₂] with (a) x = 0.01, (b) x = 0.05, and (c) x = 1, recorded at T = 300 K.

In order to elucidate the nature of the new species B observed at higher Eu content in ³_∞[Sr_1−xEu_x(Im)₂] we further investigated a ²_∞[Sr_0.95Eu_0.05(Im)₂(ImH)₂] sample that might correspond to a low temperature impurity phase of the ³_∞[Sr_1−xEu_x(Im)₂] high temperature phase. However, neither species B nor species A are detected in the EPR spectrum of ²_∞[Sr_0.95Eu_0.05(Im)₂(ImH)₂], which display rather the existence of a third Eu^II species C (see ESI†). Therefore, we can rule out a contamination of ³_∞[Sr_1−xEu_x(Im)₂] by a small amount of the low temperature phase ²_∞[Sr_0.95Eu_0.05(Im)₂(ImH)₂]. Furthermore, the EPR spectra of ³_∞[Sr_1−xEu_x(Im)₂] with x = 0.01 and 0.05 do not give any indication of the formation of Eu^II pairs.

Conclusions

To summarize, we were able to show that the creation of luminescent MOF films on self-ordered AAO substrates can be achieved by the concept of an in situ coating approach from an organic melt and metal reagents. The concept covers several luminescent lanthanide imidazolate MOFs, as shown for the two highly luminescent MOFs ³_∞[Ce(Im)₃ImH]·ImH and ³_∞[Tb(Im)₃]. In both cases, thin as well as thick luminescent films, retaining of the original crystal structures and bulk properties was achieved. The concept can be further widened to additional deposition techniques: deposition of the same MOFs on AAO was successfully combined with CVD deposition of highly luminescent crystalline MOF particles. SEM/EDX investigations support these findings and confirm that the concave walls of the macropores at the macropore apertures act as anchoring sites, where the MOF growth starts.

Finally, the use of electride induced coating from liquid ammonia was elaborated based on the formation of thin films of ³_∞[Sr_0.95Eu_0.05(Im)₂] regarding the initial growth phase. Independent from the melt contact, the initial formation of extraordinary barite rose analogous crystals was identified for this in situ coating process prior to the formation of highly luminescent thin films of ³_∞[Sr_0.95Eu_0.05(Im)₂] on AAO membranes. Time-dependent formation of the films and the control of layer thickness were also shown. Furthermore, the electride approach offers an improved homogenization of the co-doped metal centres at the atomic level, which even leads to improved bulk properties like an extraordinary quantum yield of QY = 92%. EPR spectroscopy of ³_∞[Sr_0.99Eu_0.01(Im)₂] and ³_∞[Sr_0.95Eu_0.05(Im)₂] reveals the presence of a major Eu^II species with distorted octahedral symmetry which we assign to the Eu^II incorporated at regular metal ion lattice sites being sufficiently magnetically diluted. On the basis of our EPR experiments, the formation of Eu^II pairs and clusters can be ruled out just as the presence of the non-luminescent network ²_∞[Sr_1−xEu_x(Im)₂(ImH)₂]. Altogether, this report adds three unusual in situ coating approaches to the formation and generation of luminescent MOF thin films. These materials can be potentially used for lighting techniques or sensing owing to the advantage that a low amount of material is required to achieve a film/coating. Due to the moisture sensitivity of the lanthanide imidazolates constituting the films (including Eu^II), additional water resistant coating can be recommended to allow applications under a normal atmosphere.

Experimental

X-ray powder diffraction

Powder samples of the bulk materials were prepared in Lindemann glass capillaries (Ø = 0.3 mm) under an inert gas. The powder diffractograms were recorded on a Bruker D8 Discover with Da Vinci design and a LynxEye detector in transmission geometry with a focusing Göbel mirror with CuKα-radiation (λ = 1.54056 Å). The diffractograms of the coated AAO substrates were recorded on a Si-waver, sealed with a plastic dome, in reflection geometry with CuKα-radiation (λ = 1.54056 Å).

Photoluminescence and vibrational spectroscopy

Excitation and emission spectra were recorded with a Horiba Jobin Yvon Fluorolog 3 photoluminescence spectrometer with a 450 W Xe lamp, an integration sphere, Czerny–Turner double grating (1200 grooves per mm) excitation and emission monochromators, and an FL-1073 PMT detector.

Excitation spectra were recorded from 250 to 450 nm and corrected for the spectral distribution of the lamp intensity using a photodiode reference detector. Emission spectra were recorded from 380 to 650 nm and corrected for the spherical response of the monochromators and the detector using typical correction spectra provided by the manufacturer.

The quantum yield was also recorded using a photoluminescence spectrometer, Horiba Jobin Yvon Spex Fluorolog 3, equipped with a 450 W xenon lamp, double monochromators for excitation and emission, an integrating sphere (Ulbricht sphere), and a photomultiplier as the detector. For excitation wavelengths >330 nm, the absolute quantum yield was determined according to Friend.⁴⁴ First the diffuse reflection of the sample was determined at a certain excitation wavelength. Second, the emission was measured under excitation at the same wavelength. Integration of the reflected and emitted photons by the use of the Ulbricht sphere resulted in the absolute quantum yield. Corrections were made accounting for the spectral power of the excitation source, the reflection behaviour of the Ulbricht sphere, and the sensitivity of the detector. For excitation wavelengths <330 nm, the reflectivity of the Ulbricht sphere starts to be perceptibly below 100%, resulting in a less precise determination of the absolute quantum yield. Therefore, the relative quantum yield was determined by comparison to a reference phosphor with a defined quantum yield. For that, BaMgAl₁₀O₁₇ : Eu²⁺ (BAM : Eu²⁺, Philips) was used as a reference⁴⁵.

FTIR spectra in the MIR area were recorded using a Thermo Nicolet FTIR-380 spectrometer; samples were measured with an ATR crystal.

EPR spectroscopy

Continuous wave EPR experiments have been performed on Bruker EMX 10–40 Q-band (34 GHz) and ELEXYS E5080 X-band (9.5 GHz) spectrometers equipped with Oxford Instruments He cryostats CF935 and ESR910, respectively.

SEM, EDX and elemental analyses

SEM, EDX: film thicknesses and morphological investigations were performed on a field emission scanning electron microscope (FE-SEM) ULTRA plus (Zeiss) with a GEMINI e-beam column. Element analysis, mapping, and point-ID measurements were carried out by electron dispersive X-ray spectroscopy (EDX) using a silicon drift detector (SDD) X-Max 50 mm² (Oxford Instruments) at 15 keV for the ratio of Sr : Eu in the barite rose analogous crystals of ³_∞[Sr_0.95Eu_0.05(Im)₂] and for the corresponding bulk material as well as for the identification of the corresponding elements in the deposited layers of ³_∞[Ce(Im)₃ImH]·ImH and ³_∞[Tb(Im)₃]. The instruments were calibrated using calibration files of Oxford Instruments prior to the analysis.

Elemental analysis for C, H and N was performed with a Euro Ea CHNSO Analyzer (HEKAtech GmbH).

Acknowledgements

We acknowledge the support of the Deutsche Forschungsgemeinschaft within the priority program SPP-1362 “MOFs”.

Notes and references

1. (a) B. F. Hoskins and R. Robson, J. Am. Chem. Soc., 1989, 111, 5962; (b) B. F. Hoskins and R. Robson, J. Am. Chem. Soc., 1990, 112, 1546.
2. (a) G. Ferey, Chem. Soc. Rev., 2008, 37, 191; (b) J. R. Long and O. M. Yaghi, Chem. Soc. Rev., 2009, 38, 1213; (c) L. J. Murray, M. Dincă and J. R. Long, Chem. Soc. Rev., 2009, 38, 1294; (d) C. Janiak and J. K. Vieth, New J. Chem., 2010, 34, 2366.
3. (a) O. M. Yaghi, H. Li, C. Davis, T. Richardson and T. L. Groy, Acc. Chem. Res., 1998, 31, 474; (b) S. Kitagawa and M. Kondo, Bull. Chem. Soc. Jpn., 1998, 71, 1739; (c) H. Le, M. Eddaoudi, M. O'Keefe and O. M. Yaghi, Nature, 1999, 402, 276; (d) G. Ferey, Chem. Mater., 2001, 13, 3084; (e) C. Janiak, Dalton Trans., 2003, 2781; (f) S. Kitagawa and K. Uemura, Chem. Soc. Rev., 2005, 34, 109; (g) S. Kaskel, Nachr. Chem., 2005, 53, 394; (h) A. K. Cheetham and C. N. R. Rao, Science, 2007, 318, 58.
4. (a) T. M. Reinecke, M. Eddaoudi, M. Fehr, D. Kelley and O. M. Yaghi, J. Am. Chem. Soc., 1999, 121, 1651; (b) J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous Mater., 2004, 73, 3; (c) P. Horcajada, C. Serre, M. Vallet-Regi, M. Sebban, F. Taulelle and G. Férey, Angew. Chem., Int. Ed., 2006, 45, 5974.
5. (a) S. Kitagawa, R. Kitaura and S. Noro, Angew. Chem., Int. Ed., 2004, 43, 2334; (b) G. Fèrey, Chem. Soc. Rev., 2008, 191, 191; (c) N. Stock and S. Biswas, Chem. Rev., 2012, 112, 933; (d) S. Kaskel, Chem.-Ing.-Tech., 2010, 82, 1019.
6. (a) M. D. Allendorf, C. A. Bauer, R. K. Bhakta and R. J. T. Houk, Chem. Soc. Rev., 2009, 38, 1330; (b) Y. Cui, Y. Yue, G. Qian and B. Chen, Chem. Rev., 2012, 112, 1126.
7. (a) C. Gomes Silva, A. Corma and H. Gracia, J. Mater. Chem., 2010, 20, 3141; (b) H. Khajavi, J. Gascon, J. M. Schins, L. D. A. Siebbeles and F. Kapteijn, J. Phys. Chem. C, 2011, 115, 12487; (c) R. E. Morris and X. Bu, Nat. Chem., 2010, 2, 353.
8. (a) A. K. Cheetham, C. N. R. Rao and R. K. Feller, Chem. Commun., 2006, 4780; (b) J. Rocha, L. D. Carlos, F. A. Almeida Paz and D. Ananias, Chem. Soc. Rev., 2011, 40, 926.
9. (a) D. F. Sava, L. E. S. Rohwer, M. A. Rodriguez and T. M. Nenoff, J. Am. Chem. Soc., 2012, 134, 3983; (b) A. Ablet, S.-M. Li, W. Cao, X.-J. Zheng, W.-T. Wong and L.-P. Jin, Chem. – Asian J., 2013, 8, 95; (c) P. Falcaro and S. Furukawa, Angew. Chem., Int. Ed., 2012, 51, 8431; (d) M. D. Allendorf, C. A. Bauer, R. K. Bhakta and R. J. T. Houk, Chem. Soc. Rev., 2009, 38, 1330.
10. P. R. Matthes, C. J. Höller, M. Mai, J. Heck, S. J. Sedlmaier, S. Schmiechen, C. Feldmann, W. Schnick and K. Müller-Buschbaum, J. Mater. Chem., 2012, 22, 10179.
11. J.-C. Rybak, L. V. Meyer, J. Wagenhöfer, G. Sextl and K. Müller-Buschbaum, Inorg. Chem., 2012, 51, 13204.
12. J. Heine, T. Wehner, R. Bertermann, A. Steffen and K. Müller-Buschbaum, Inorg. Chem., 2014, 53, 7197.
13. L. V. Meyer, J. Vogt, F. A. Brede, H. Schäfer, M. Steinhart and K. Müller-Buschbaum, CrystEngComm, 2013, 15, 9382.
14. D. Zacher, R. Schmid, C. Wöll and R. A. Fischer, Angew. Chem., Int. Ed., 2011, 50, 176.
15. O. Shekhah, J. Liu, R. A. Fischer and C. Wöll, Chem. Soc. Rev., 2011, 40, 1081.
16. A. Bétard and R. A. Fischer, Chem. Rev., 2012, 112, 1055.
17. (a) J. Liu, F. Sun, F. Zhang, Z. Wang, R. Zhang, C. Wang and S. Qiu, J. Mater. Chem., 2011, 21, 3775; (b) M. D. Allendorf, A. Bétard and R. A. Fischer, in Metal–Organic Frameworks: Applications from Catalysis to Gas Storage, ed. D. Farrusseng, Wiley-VCH, Weinheim, 2011.
18. J. Gascon and F. Kapteijn, Angew. Chem., Int. Ed., 2010, 49, 1530.
19. H. Liu, H. Wang, T. Chu, M. Yua and Y. Yang, J. Mater. Chem. C, 2014, 2, 8683.
20. N. Campagnol, E. R. Souza, D. E. De Vos, K. Binnemans and J. Fransaer, Chem. Commun., 2014, 50, 12545.
21. V. Stavila, A. A. Talin and M. D. Allendorf, Chem. Soc. Rev., 2014, 43, 5994.
22. (a) R. Ranjan and M. Tsapatsis, Chem. Mater., 2009, 21, 4920; (b) C. M. Lew, R. Cai and Y. Yan, Acc. Chem. Res., 2010, 43, 210; (c) V. Varela-Guerrero, Y. Yoo, M. C. McCarthy and H.-K. Jeong, J. Mater. Chem., 2010, 20, 3938; (d) X. Zou, G. Zhu, F. Zhang, M. Guo and S. Qiu, CrystEngComm, 2010, 12, 352.
23. (a) R. Makiura, S. Motoyama, Y. Umemura, H. Yamanaka, O. Sakata and H. Kitagawa, Nat. Mater., 2010, 9, 565; (b) S. Motoyama, R. Makiura, O. Sakata and H. Kitagawa, J. Am. Chem. Soc., 2011, 133, 5640.
24. (a) C. M. Bell, M. F. Arendt, L. Gomez, R. H. Schmehl and T. E. Mallouk, J. Am. Chem. Soc., 1994, 116, 8374; (b) S. Cobo, G. Molnar, J. A. Real and A. Bousseksou, Angew. Chem., Int. Ed., 2006, 45, 5786.
25. (a) S. Hausdorf, F. Baitalow, T. Böhle, D. Rafaja and F. O. Mertens, J. Am. Chem. Soc., 2010, 132, 10978; (b) A. Bétard, S. Wannapaiboon and R. A. Fischer, Chem. Commun., 2012, 48, 10493.
26. (a) S. Hermes, F. Schröder, R. Chelmowski, C. Wöll and R. A. Fischer, J. Am. Chem. Soc., 2005, 127, 13744; (b) H. Bux, F. Liang, Y. Li, J. Cravillon, M. Wiebcke and J. Caro, J. Am. Chem. Soc., 2009, 131, 6000; (c) S. R. Venna and M. A. Carreon, J. Am. Chem. Soc., 2010, 132, 76.
27. (a) J. Gascon, S. Aguado and F. Kapteijn, Microporous Mesoporous Mater., 2008, 113, 132; (b) E. Biemmi, C. Scherb and T. Bein, J. Am. Chem. Soc., 2007, 129, 8054; (c) E. Biemmi, A. Darga, N. Stock and T. Bein, Microporous Mesoporous Mater., 2008, 114, 380.
28. (a) M. Arnold, P. Kortunov, D. J. Jones, Y. Nedellec, J. Kärger and J. Caro, Eur. J. Inorg. Chem., 2007, 60; (b) J. Yao, D. Dong, D. Li, L. He, G. Xu and H. Wang, Chem. Commun., 2011, 47, 2559; (c) A. Centrone, Y. Yang, S. Speakman, L. Bromberg, G. C. Rutledge and T. A. Hatton, J. Am. Chem. Soc., 2010, 132, 15687.
29. (a) K. Binnemans, Chem. Rev., 2009, 109, 4283; (b) L.-W. Han, J. Lü, T.-F. Liu, S.-Y. Gaoa and R. Cao, Dalton Trans., 2010, 39, 10967; (c) H. Guo, Y. Zhu, S. Qiu, J. A. Lercher and H. Zh, Adv. Mater., 2010, 22, 4190; (d) A. Feinle, F. Lavoie-Cardinal, J. Akbarzadeh, H. Peterlik, M. Adlung, C. Wickleder and N. Huesing, Chem. Mater., 2012, 24, 3674.
30. L. V. Meyer, F. Schönfeld, A. Zurawski, M. Mai, C. Feldmann and K. Müller-Buschbaum, Dalton Trans. , submitted.
31. K. Müller-Buschbaum, S. G. Torres, P. Larsen and C. Wickleder, Chem. Mater., 2007, 19, 655.
32. (a) H. Masuda and K. Fukuda, Science, 1995, 268, 1466; (b) H. Masuda, K. Yada and A. Osaka, Jpn. J. Appl. Phys., 1998, 37, L1340.
33. (a) W. Lee, K. Nielsch and U. Gösele, Nanotechnology, 2007, 18, 475713; (b) K. Schwirn, W. Lee, R. Hillenbrand, M. Steinhart, K. Nielsch and U. Gösele, ACS Nano, 2008, 2, 302; (c) W. Lee, K. Schwirn, M. Steinhart, E. Pippl, R. Scholz and U. Gösele, Nat. Nanotechnol., 2008, 3, 234; (d) M. Steinhart, Adv. Polym. Sci., 2008, 220, 123.
34. P. Dorenbos, J. Lumin., 2000, 91, 91.
35. S. P. Nakamura, Proc.Soc. Photo-Opt. Instrum. Eng., 1997, 3002, 26.
36. A. Zurawski, M. Mai, D. Baumann, C. Feldmann and K. Müller-Buschbaum, Chem. Commun., 2011, 47, 496.
37. C. Hadenfeldt, H. Jacobs and R. Juza, Z. Anorg. Allg. Chem., 1970, 379, 144.
38. C. C. Quitmann and K. Müller-Buschbaum, Z. Anorg. Allg. Chem., 2004, 630, 2422.
39. K. Müller-Buschbaum, Z. Anorg. Allg. Chem., 2004, 630, 895.
40. U. Schindewolf, Chem. unserer Zeit, 1970, 4, 37.
41. W. A. Tarr, Am. Mineral., 1933, 18, 260.
42. J.-C. Rybak, M. Hailmann, P. R. Matthes, A. Zurawski, J. Nitsch, A. Steffen, J. G. Heck, C. Feldmann, S. Götzendörfer, J. Meinhardt, G. Sextl, H. Kohlmann, S. J. Sedlmaier, W. Schnick and K. Müller-Buschbaum, J. Am. Chem. Soc., 2013, 135, 6896.
43. S. Scheizer, G. Corradi, A. Edgar and J.-M. Spaeth, J. Phys.: Condens. Matter, 2001, 13, 2331.
44. J. V. de Mello, H. F. Wittmann and R. H. Friend, Adv. Mater., 1997, 9, 230.
45. (a) A. L. I. Stevels and A. J. M. Schrama de Pauw, J. Electrochem. Soc., 1974, 123, 691; (b) S. Marks, J. Heck, P. O. Burgos, C. Feldmann and P. W. Roesky, J. Am. Chem. Soc., 2012, 134, 16983.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental details, XRPD data, photoluminescence spectra, SEM/EDX, EPR spectra and analysis. See DOI: 10.1039/c4qi00184b

‡ Self-ordered AAO with a pore diameter of 400 nm and a pore depth of 100 μm was prepared following the two-step anodization of aluminium introduced by Masuda and coworkers.³² The porous alumina layer formed by the first anodization was selectively removed by etching with a solution consisting of 1.8 g chromium(VI) oxide and 7.1 g 85% phosphoric acid filled up to 100 ml with deionized water. Al substrates patterned with hexagonal arrays of indentations (replicas of the pore bottoms of the first Al layer) obtained in this way were subjected to a second anodization step carried out as described above (anodization time 20 h). The second anodization yielded AAO with pores ∼100 μm in depth and ∼180 nm in diameter. The pores were isotropically widened to 400 nm for 2 h at 30° C by treatment with an aqueous solution of 10 wt% phosphoric acid (13.3 g H₃PO₄ (85%) + 100 g H₂O) under continuous stirring.

---