The present invention relates to a method for forming a layer comprising an organic lanthanide complex on a substrate. The present invention further relates to a layered substrate comprising a substrate having a layer disposed thereon, wherein the layer comprises an organic lanthanide complex and less than 10 ²⁰ molecules/cm ³ water and an optoelectronic device comprising the same.

BACKGROUND TO THE INVENTION

Infrared optical gain media that could be simply integrated onto substrates, e.g. silicon photonic substrates, would provide a means of producing low cost silicon lasers. In this respect, lanthanide ions have long been of interest for their unique optical properties. For example, materials containing erbium ions have been studied for this purpose as erbium can be made to lase in the range 1530 to 1550 nm. However, it is thought that low absorption cross-section and concentration quenching restrict the materials’ applicability.

Organic sensitised lanthanides may provide one possible route to this goal as they are thought to address these issues. The sensitisation afforded by the organic ligands can increase the effective cross-section by several orders of magnitude, whilst placing the ions in an organic cage allows fine control of ion separation. However, even though it is now over 70 years since sensitisation of lanthanide ions was first observed, the quantum yield of the best organic erbium composites are still limited to around 8%. This is despite efforts to remove the main quenching centres in such materials, C-H and O-H vibrations, by producing fully halogenated organic materials. Indeed in a review article in 2015 (see Bunzli J-CG; On the design of highly luminescent lanthanide complexes; Coordination Chemistry Reviews; 2015;293-294:19-47.), it was stated“Minimization of non-radiative deactivation processes is well understood but not yet mastered at the experimental level and only a handful of compounds have quantum yields larger than 3-5% and none more than 10%, despite smart designs. Therefore this aspect remains an open field much in need of clever contributions.”

It is one object of the present invention to overcome at least some of the disadvantages of the prior art or to provide a commercially useful alternative thereto. It is a further object of the present invention to improve the quantum efficiency of lanthanide complexes on substrates.

SUMMARY OF THE INVENTION

In a first aspect the present invention provides a method for forming a layer comprising an organic lanthanide complex on a substrate, the method comprising:

(i) providing a substrate;

(ii) providing a vapour comprising an organic lanthanide complex; and

(iii) contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate, wherein the layer comprises less than 1 x10 ²⁰ molecules/cm ³ water.

Each aspect or embodiment as defined herein may be combined with any other aspect(s) or embodiment(s) unless clearly indicated to the contrary. In particular any feature indicated as being preferred or advantageous may be combined with any other feature or features indicated as being preferred or advantageous.

In a further aspect the present invention provides a layered substrate obtainable by the method described herein.

In a further aspect the present invention provides a layered substrate comprising a substrate having a layer disposed thereon, wherein the layer comprises an organic lanthanide complex and less than 1 x10 ²⁰ molecules/cm ³ water.

In a further aspect the present invention provides an optoelectronic device comprising the layered substrate as described herein. Said optoelectronic devices may include signal translating devices, including sources, optical amplifiers or optical splitters for

telecommunications applications.

In a further aspect the present invention provides a method for forming a layer comprising an organic lanthanide complex on a substrate, the method comprising:

(i) providing a substrate;

(ii) providing a vapour comprising an organic lanthanide complex; and

(iii) contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate, wherein step (iii) is carried out under conditions effective to minimise moisture incorporation into the layer.

In a further aspect the present invention provides a method for forming a layer comprising an organic lanthanide complex on a substrate, the method comprising:

(i) providing a substrate;

(ii) providing a vapour comprising an organic lanthanide complex; and

(iii) contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate, wherein the step of contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate:

(a) comprises heating the substrate to a temperature of from 30 to 200 °C; and/or

(b) comprises irradiating the vapour with a source of infrared energy; and/or

(c) is performed under ultra-high vacuum.

Other preferred embodiments of the compounds according to the invention appear throughout the specification and in particular in the examples.

The present inventors have surprisingly found that the quantum efficiency of lanthanide complexes on substrates can be significantly improved. Without wishing to be bound by theory it is thought that this improvement can be realised by reducing amounts of contaminants (such as water) being incorporated into the lanthanide complex layer during growth.

DETAILED DESCRIPTION OF THE INVENTION

Unless otherwise defined herein, scientific and technical terms used in connection with the present invention shall have the meanings that are commonly understood by those of ordinary skill in the art. The meaning and scope of the terms should be clear, however, in the event of any latent ambiguity, definitions provided herein take precedent over any dictionary or extrinsic definition.

As noted above, lanthanides are of particular interest for production of low cost silicon lasers. The photoluminescent properties of lanthanides have properties that may be favourable in a number of applications. In view of their wavelength of emission, lanthanide ions (and specifically erbium) may be particularly useful for telecommunication applications. Modern telecommunications rely on a network of silica optical fibre cables to carry data.

The optical fibres transmit certain wavelengths with lower loss. The emission wavelength of erbium (being in the range 1530 to 1550 nm) has particular advantages for use in telecommunication systems, as it falls within the low loss window of silica.

Despite this, direct excitation of lanthanide ions has been found to be difficult, due to the ions’ poor ability to directly absorb light. Sensitisation of the lanthanide luminescence can be used to enhance the photoluminescence of lanthanide materials. Sensitisation requires the incorporation of chromophores into the material containing the lanthanide ions. The chromophores may more effectively absorb incident light, hv, than compared to the lanthanide ions directly. Energy absorbed by the chromophore may then be transferred to a nearby (or coupled) lanthanide ion. The excited lanthanide ion will eventually relax, thereby emitting luminescence at its characteristic wavelength. In this way, the efficiency of energy absorption by the lanthanide is improved. Figure 7 shows a schematic representation of sensitisation of a lanthanide ion, in which energy, E, is transferred to a lanthanide complex (lanthanide ion 710 and ligand 720), from a coupled chromophore 730. The resultant luminescence 740 is emitted upon relaxation of an electron from the excited state, Ln ^(excited> , at the lanthanide ion to the ground state, i_n ^(9round> , of the lanthanide ion.

Quantum Efficiency as described herein is a well-known term in the art, and may be considered as a measure of the proportion of energy absorbed by the ion that contributes to emission at a desired wavelength. Quantum Efficiency values are provided herein as percentages. It is understood that these values may also be expressed as fractions, i.e. a 100% quantum efficiency is equivalent to a quantum efficiency of 1 ; an 80% quantum efficiency is equivalent to a quantum efficiency of 0.8. Quantum efficiency is measured by comparing the ratio of the measured lifetime of the lanthanide to the measured radiative lifetime calculated from the direct absorption of the same transition or from a full Judd-Ofelt calculation of the relevant transitions in the same lanthanide ion.

The water concentration in the layer is preferably determined from the measured lifetime of the lanthanide ions using a Fbrster Resonant Energy Transfer (FRET) model of the energy transfer from the lanthanide to a water molecule coupled with a 3 dimensional nearest neighbour probability calculation, as described in the experimental section below. Alternatively, the water concentration in the layer can be determined by performing x-ray photoelectron spectroscopy measurements on the layer and measuring the relative intensities of the oxygen 1 s signal due to oxygen atoms within the organic lanthanide complex (e.g. due to the P=0 bonds in imidodiphosphinate ligands) and the oxygen 1 s signal due to oxygen atoms present in water molecules. While this latter method is less preferred, it can be used as a comparison with the preferred FRET model method as a benchmark.

As used herein, the term“on” with respect to the layer and the substrate means“in direct contact”, in other words, without one or more additional layers therebetween.

Although the below description refers to an organic lanthanide complex, it will be understood that the below described examples could be used in conjunction with any organic complex comprising a rare earth element (such as a lanthanide, or chemically similar elements scandium and yttrium).

The present invention provides a method for forming a layer comprising an organic lanthanide complex on a substrate, the method comprising:

(i) providing a substrate;

(ii) providing a vapour comprising an organic lanthanide complex; and

(iii) contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate, wherein the layer comprises less than 1 x10 ²⁰ molecules/cm ³ water.

Preferably the layer comprises less than 5x10 ¹⁹ molecules/cm ³ water, more preferably less than 3x10 ¹⁹ , still more preferably less than 2x10 ¹⁹ and most preferably less than 1 x10 ¹⁹ molecules/cm ³ water. Preferably the layer comprises at least 5x10 ¹⁸ molecules/cm ³ water.

Preferably step (iii) is carried out under conditions effective to minimise moisture or water incorporation into the layer.

Preferably the step of contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate comprises heating the substrate to a temperature of from 30 to 200 °C, more preferably from 30 to 100 °C, more preferably from 30 to 70 °C, more preferably from 35 to 65 °C, still more preferably from 40 to 60 °C, and most preferably from 40 to 50 °C. Without wishing to be bound by theory, it is thought that contacting the at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate at a slightly elevated substrate temperature has the effect of reducing the sticking coefficient of the water molecules, whilst not greatly affecting that of the organic lanthanide complex, thereby reducing the water content in the resultant layer.

Preferably the step of contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate comprises irradiating the vapour with a source of infrared energy. Sources of infrared energy are known in the art. For example, preferably, the source of infrared energy comprises a tungsten halogen bulb. Without wishing to be bound by theory, it is thought that irradiating the vapour with a source of infrared energy raises the contaminant (e.g. water) molecules into an excited state, reducing their sticking coefficient in the layer being deposited.

Preferably the step of contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate comprises heating the substrate to a temperature of from 30 to 200 °C, more preferably from 30 to 100 °C, more preferably from 30 to 70 °C, more preferably from 35 to 65 °C, still more preferably from 40 to 60 °C, and most preferably from 40 to 50 °C; and the step of contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate comprises irradiating the vapour with a source of infrared energy.

Preferably the step of contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate is performed under ultra-high vacuum. Preferably the vacuum pressure is less than 10 ^~7 mbar, more preferably from 10 ^~10 to 10 ^~8 mbar.

Preferably the step of providing a vapour comprising an organic lanthanide complex comprises subliming or evaporating at least one organic lanthanide complex under a vacuum, preferably by heating the at least one organic lanthanide complex to a

temperature of from 100 to 650 °C, more preferably from 150 to 600 °C, more preferably from 150 to 550 °C, more preferably from 150 to 500 °C, more preferably from 150 to 450 °C, more preferably from 150 to 400 °C, more preferably from 150 to 350 °C, more preferably from 150 to 300 °C, more preferably from 150 to 250 °C, most preferably from 150 to 200 °C. Preferably the organic lanthanide complex has the general formula I_nX ₃ M(LnX ₄ ) or N(LnX ₄ ) ₂ ,

wherein Ln is a lanthanide selected from the group consisting of Er, Tm, Ho, Nd, Yb and mixtures of two or more thereof,

wherein M is a singly charged ion, optionally selected from the group consisting of Li ⁺ , Na ⁺ , K ⁺ , Rb ⁺ and Cs ⁺ ,

wherein N is a doubly charged cation, optionally selected from the group consisting of Be ²⁺ , Mg ²⁺ , Ca ²⁺ , Sr ²⁺ , Ba ²⁺ , Zn ²⁺ , and Cu ²⁺ , and

wherein X is a fully-halogenated aromatic ligand.

More preferably, the organic lanthanide complex has the general formula LnX ₃ ,

wherein Ln is a lanthanide selected from the group consisting of Er, Tm, Ho, Nd, Yb and mixtures of two or more thereof, and

wherein X is a fully-halogenated aromatic ligand.

Preferably Ln is Er and/or the fully-halogenated aromatic ligand is an imidodiphosphinate.

Preferably Ln is Er and/or X is represented by one or more of the following formulae:

Preferably the vapour and the layer further comprise a chromophore adapted to sensitise the lanthanide in the layer. Methods of adapting chromophores to sensitise lanthanides are known in the art and are disclosed in WO 2015/001338 A1 , the contents of which are incorporated herein by reference.

Preferably the chromophore is a halogenated aromatic chromophore, and more preferably the chromophore is a fully halogenated aromatic chromophore. Preferably, the

chromophore is a separate molecule from the lanthanide or rare earth complex (in other words, the chromophore is radiatively coupled to the organic lanthanide complex, but is not chemically bonded to the organic lanthanide complex). More preferably the chromophore is represented by one or more of the following formulae:

Preferably the lanthanide complex is erbium(lll) tetrakis(pentafluorophenyl)- imidodiphosphinate, Er(F-TPIP) ₃ .

Preferably the lanthanide complex is erbium(lll) tetrakis(pentafluorophenyl)- imidodiphosphinate, Er(F-TPIP) ₃ , and/or the chromophore is the zinc(ll) salt of 2-(3,4,5,6- tetrafluoro-2-hydroxyphenyl)-4,5,6,7-tetrafluorobenzothiazol e, Zn(F-BTZ) ₂ .

Preferably the vapour and the layer further comprise an organic yttrium complex.

Alternatively, an additional layer comprising an organic yttrium complex is formed on the layer comprising the organic lanthanide complex. Preferably the organic yttrium complex has the general formula YX ₃ , M(YX ₄ ) or N(YX ₄ ) ₂ , with X, M and N as defined above. More preferably the organic yttrium complex comprises Y(F-TPIP) ₃ . Preferably the organic yttrium complex is optically inert. Without wishing to be bound by theory, it is thought that Y(F-TPIP) ₃ is optically inert, but (a) may act as a dilutant such that Ln-Ln (e.g. Er-Er) interactions may be reduced, for instance when the organic yttrium complex is comprised within the layer comprising the organic lanthanide complex, and/or (b) may act as a capping layer to protect the layered substrate from the environment and may act as the cladding in a waveguide, for instance when the organic yttrium complex is formed as an additional layer on the layer comprising the organic lanthanide complex. Preferably the lanthanide has a quantum efficiency in the layer of at least 40%. More preferably the lanthanide has a quantum efficiency in the layer of at least 50%. More preferably the lanthanide has a quantum efficiency in the layer of at least 60%. More preferably the lanthanide has a quantum efficiency in the layer of at least 75%. More preferably the lanthanide has a quantum efficiency in the layer of at least 80%. More preferably the lanthanide has a quantum efficiency in the layer of at least 85%. More preferably the lanthanide has a quantum efficiency in the layer of at least 90%. More preferably the lanthanide has a quantum efficiency in the layer of at least 95%. Most preferably the lanthanide has a quantum efficiency in the layer of at least 96%, or at least

97%, or at least 98%, or at least 99%.

In one preferable embodiment, the layer consists of the organic lanthanide complex, the less than 1x10 ²⁰ molecules/cm ³ water, the optional chromophore, the optional organic yttrium complex, and any other unavoidable impurities.

In one preferable embodiment, the layer consists essentially of the organic lanthanide complex, the less than 1 x10 ²⁰ molecules/cm ³ water, the optional chromophore and the optional organic yttrium complex.

Preferably the substrate is a waveguide formed of silica, silicon, silicon nitrides, oxynitrides, doped nitrides, alumina and/or a tantalum oxide. More preferably the substrate is a waveguide formed of silica and/or silicon

Preferably, the method comprises:

(i) providing a substrate;

(ii) providing a vapour comprising an organic lanthanide complex; and

(iii) contacting at least a portion of the substrate with the vapour under a vacuum to form a layer comprising the organic lanthanide complex on the substrate, wherein the layer comprises less than 1 x10 ²⁰ molecules/cm ³ water;

wherein the organic lanthanide complex has the general formula LnX ₃ ,

wherein Ln is Er and/or X is represented by one or more of the following formulae:

wherein the step of providing a vapour comprising an organic lanthanide complex comprises subliming at least one organic lanthanide complex under a vacuum, preferably by heating the at least one organic lanthanide complex to a temperature of from 150 to 200 °C.

There is further provided a layered substrate obtainable by the method described herein.

There is further provided a layered substrate comprising a substrate having a layer disposed thereon, wherein the layer comprises an organic lanthanide complex and less than 10 ²⁰ molecules/cm ³ water.

There is further provided an optoelectronic device comprising a layered substrate as described herein.

In a further aspect there is a method for depositing a layer comprising an organic lanthanide complex on a substrate, the method comprising;

(a) providing a substrate;

(b) providing a vapour comprising the organic lanthanide complex;

(c) exposing the substrate to the vapour, to deposit a layer comprising the organic lanthanide complex on the substrate, wherein the layer comprises less than 1 x10 ²⁰ molecules/cm ³ water.

The substrate may be any wafer or underlying layer. For example, the substrate may be a silica, glass, silicon nitride or silicon wafer. The vapour may be the organic lanthanide complex in the gaseous phase, optionally together with a carrier gas. The vapour may be provided by subliming or evaporating at least one organic lanthanide complex under a vacuum. The vapour may further comprise a chromophore adapted to sensitise the lanthanide in the layer, which may be sublimed or evaporated together with the organic lanthanide complex. The vapour may include unintended impurities, for instance water.

The layer may be deposited or formed on the substrate as a result of physical deposition.

The step of exposing the substrate to the vapour to deposit a layer comprising the organic lanthanide complex on the substrate, comprises directing the vapour to make contact with a surface of the substrate. The step of exposing the substrate to the vapour is carried out under conditions effective to minimise moisture incorporation into the layer. Said conditions may comprise one or more of:

(i) heating the substrate to a temperature of from 30 to 200 °C; and/or

(ii) irradiating the vapour with a source of infrared energy; and/or

(iii) performing the step under ultra-high vacuum.

It is noted that, in addition to heating the substrate during the deposition of the layer comprising the organic lanthanide complex, the substrate optionally may be heated prior to deposition to a temperature of from 30 to 200 °C, or to a higher temperature. Examples of an ultra-high vacuum comprise a vacuum of less than 10 ^~7 mbar, or more preferably from 10 ^~10 to ^ 0 mbar.

The method results in a layered substrate, comprising at least the substrate and the layer comprising the organic lanthanide complex. The method may comprise application of additional layers, such that the resultant layered substrate comprises a plurality of layers on the substrate. For instance, a reflective layer may be applied on to the layer comprising the organic lanthanide complex. In one example, the reflective layer may be aluminium. One or more additional layers may be applied on to or above (with respect to the substrate) the layer comprising the organic lanthanide complex in order to incorporate the layer comprising the organic lanthanide complex into an optoelectronic device.

The layer formed according to the described method results in a layer comprising the organic lanthanide complex (including the lanthanide ion and ligand). Formation of the layer according to the method is by physical deposition of thermally stable material, rather than deposition via chemical reaction or decomposition of an unstable, gaseous material.

As a consequence, the resultant layer comprises the original organic lanthanide complex deposited on the substrate, rather than a chemically different material (such as a metal layer of the lanthanide, that may be obtained as a result of chemical decomposition for instance).

The layer may further comprise a chromophore adapted to sensitise the lanthanide in the layer. The layer may further comprise unintended impurities, including water. Most significantly, the described method results in a reduction of the water content (number of water molecules per cm ³ ) in the layer. This in turn increases the average spacing between a lanthanide ion and its nearest water molecule within the crystal structure of the layer. Reducing the proximity of water molecules to the lanthanide ions appears to reduce the energy transfer between the lanthanide ions and the water, and consequently increases the quantum efficiency of the material layer. The increased quantum efficiency opens doors to a number of advantageous applications for the material, in particular in optoelectronic devices.

These and other aspects of the invention will now be described with reference to the accompanying Figures:

Figure 1: displays the luminescence decay for the ⁴ li _{3/2— >} ⁴ li _5/2 transition for Er ³⁺ ions in Er(F-TPIP) ₃ under different conditions. The data is presented on a double exponential axis to clearly demonstrate the different lifetime components. Panel (a) is for a sample of pure Er(F-TPIP) ₃ powder and is mono-exponential with a decay time of -220 ps. Panel (b) is for a 500 nm thick film of Er _0.2 Y _0.8 (F-TPIP) ₃ grown at a partial pressure of H ₂ 0 of -4 x 10 ^~8 mbar. Panel (c) is for a 350 nm thick film of Er _0.2 Yo. ₈ (F-TPIP) ₃ grown at a partial pressure of H ₂ 0 of -3 x 10 ¹ ° mbar.

Figure 2: The squares are the average luminescence lifetime (left hand scale) for the ⁴ l _13/2 4I _{Ί 5/2} transition from Er ³⁺ ions in 100nm films of Er _0.2 Y _0.8 (F-TPIP) ₃ as a function of the partial pressure of water in the vacuum chamber. The triangles (right hand scale) are the water concentration in each film calculated from the percentage of the long lifetime component. The solid lines are a guide to the eye. Despite changing the partial pressure of water by nearly three orders of magnitude the water concentration is only reduced by a factor of 2.

Figure 3: provides a luminescence decay curve for the ⁴ l _13/2 ⁴ l _15/2 transition from Er ³⁺ ions in a film of Er _0.2 Yo. ₈ (F-TPIP) ₃ that was grown (a) at a substrate temperature of ~45°C. This film has an average lifetime of 8.1 ms. The partial pressure of water in the vacuum system during the growth of this film was -5 x 10 ⁸ mbar. An equivalent film grown at room temperature is shown in Figure 1 and has an average lifetime of only 2.5 ms. Panel (b) is a film grown at an equivalent temperature but with the incoming flux of organic molecules illuminated by the light from a tungsten halogen bulb. The average lifetime is 10 ms. This film was grown with a partial pressure of water in the vacuum system of only -2 x 10 ⁶ mbar. Figure 4: the calculated erbium emission lifetime as a function of distance of the erbium ion from a water molecule.

Figure 5: provides an example absorption spectrum for the ⁴ li _{5/2— <} - ⁴ li _3/2 transition in erbium in a sample of Er(F-TPIP) ₃ . Six samples were prepared with different amounts of Er(F- TPIP) ₃ and these were globally fitted to obtain an average radiative lifetime for the erbium ions of 2.8 ± 0.7 ms.

Figure 6: shows time resolved emission from erbium ions (at a wavelength of 1532 nm) in two nominally identical samples of [Er ₀ .4Yo.6(F-TPIP) ₃ ]o.75[Zn(F-BTZ) ₂ ]o.25 grown at either 25 °C or 55 °C. Both samples were grown in a vacuum with a base pressure of -10 ⁸ torr and with a partial pressure of water ~2 x 10 ⁹ torr. The 55 °C growth shows an average lifetime of ~2.4 ms compared with ~1.1 ms for the sample grown at 25 °C. Y(F-TPIP) ₃ was used in this layer to adjust the refractive index and control the erbium concentration for a particular application.

Figure 7: provides a schematic diagram of the sensitisation of lanthanide luminescence.

When introducing elements of the present disclosure or the preferred embodiments(s) thereof, the articles“a”,“an”,“the” and“said” are intended to mean that there are one or more of the elements. The terms“comprising”,“including” and“having” are intended to be inclusive and mean that there may be additional elements other than the listed elements.

The foregoing detailed description has been provided by way of explanation and illustration, and is not intended to limit the scope of the appended claims. Many variations in the presently preferred embodiments illustrated herein will be apparent to one of ordinary skill in the art, and remain within the scope of the appended claims and their equivalents.

The following non-limiting examples further illustrate the present invention.

EXPERIMENTAL

An erbium luminescence decay curve was measured for powders of the purified materials and there was always found to be monoexponential decay with a lifetime of -220 ps. Typical luminescence decay curves for a series of Er(F-TPIP) ₃ films grown under different vacuum conditions are presented in Figure 1 . As has been recognised by the inventors, it can be seen that as the vacuum improves a long lifetime component becomes increasingly dominant. This result indicates that it is likely to be contaminants in the vacuum which are being incorporated into the layer during growth that are controlling the quenching and using a residual gas analyser it was determined that the major hydrogen containing species in the vacuum was water vapour.

The luminescence decay curves clearly show two major components, one with a decay time of the order of 100’s of microseconds and a second of the order of ~10 ms. Whilst fitting of the long lifetime component is straightforward it was found that for the shorter lifetimes there is not a single lifetime present and the approach was taken of fitting the decay with a triple exponential. Analysis of many samples prepared under different conditions show that they can all be fitted with three distinct lifetimes of 0.19 ms ± 0.04 ms, 0.52 ms ± 0.07 ms and 10.6±2.4 ms and it is the ratio of these lifetimes that varies as the partial pressure of water during deposition changes. Whilst the two shorter lifetime components provide a good fit to all measured decay curves there is probably some distribution of lifetimes within the range of -200 ps to -1 ms depending of the average distance between the erbium ions and water molecules in the film.

Figure 2 shows the average measured erbium recombination lifetime as a function of the partial pressure of water in the vacuum during deposition. It can be seen that there is a clear correlation with the lifetime increasing as the partial pressure of water is reduced. At the simplest level one can calculate that at a partial pressure of water of 10 ^~7 mbar there is approximately a monolayer of water striking the sample surface every 10 s during growth which is comparable to the rate at which the film is being grown.

To calculate the effect of a single water molecule on the quenching of erbium in the films it was assumed that the energy transfer from erbium to water is by FRET and a Forster radius was calculated for the quenching of 0.994 nm. This allowed the emission lifetime of an Er ³⁺ ion for a given distance from a single water molecule to be calculated, as shown in Figure 4. Using this approach, it can be determined that for a water molecule at a distance of -0.5 nm from an Er ³⁺ ion the measured lifetime should be of the order of -220 ps, the typical lifetime measured in Er(F-TPIP) ₃ powder. This distance corresponds to the minimum distance a water molecule may be expected to get to an Er ³⁺ ion in an Er(F- TPIP) ₃ molecule, outside the imidodiphosphinate core but within the perfluorophenyl groups. It also suggests that during growth of an Er(F-TPIP) ₃ film under a -10 ⁶ mbar vacuum conditions, for example using a turbomolecular pump where most of the gas in the vacuum would be expected to be low molecular weight species such as water, this radiative lifetime is obtained. As the partial pressure of water during deposition is reduced the incorporation of water in the films is lowered and hence one begins to see longer lifetime components.

In order to model this effect, a 3D nearest neighbour approach was taken to calculate the probability, for a given concentration of water in the film, of an Er ³⁺ ion having a water molecule within a certain distance from the ion. Using the measured lifetimes for the Er ³⁺ ions and their associated uncertainties it can be estimated that the range of distances of water molecules from each Er ³⁺ ion are ~0.5 nm, -0.7 nm and -3.2 nm for the three lifetime components respectively. For a given density of water in the film, the expected distribution in the measured lifetimes can then be calculated. Given that it is known that there is some uncertainty in the absolute distribution of lifetimes in the two shorter lifetime components (for example, an increase in impurities at a surface or interface between the substrate and the film would predominantly affect the ratio of the two shorter components) only the percentage contribution of the longest lifetime component was used, which is always well defined, to estimate the water concentration.

Figure 2 shows the calculated concentration of water in each film as a function of the partial pressure of water in the vacuum system during growth. Whilst the calculated water concentration in the films falls (from -3 x 10 ² ° cm ³ to -1 x 10 ² ° cm ³ ) as the partial pressure of water in the vacuum system is decreased, the reduction in water concentration is not as dramatic as would be expected if this were the only source of water molecules present. Therefore, even under ultra-high vacuum (UHV) conditions there appears to be a significant flux of additional water molecules on the growing surface and this may be coming from the evaporation source. This could either be water outgassing from the hot source itself or water that is still contaminating the Er(F-TPIP) ₃ in the source. This hypothesis is supported by the fact that, even after the loaded source is baked under UHV conditions at 100 °C for 24 hours prior to growth, an increase in the measured partial pressure of water in the vacuum chamber during deposition is still seen. This increase is small (e.g. -1 x 10 ⁹ mtorr) but the substrate is placed in a direct line of sight of the source so it will see the full flux of outgassing water whereas the residual gas analyser is placed out of the molecular beam.

In view of the above, in order to overcome this limitation, the incorporation of residual water contamination in the films during growth was reduced by the present inventors. In one example, this was achieved by growing films at a slightly elevated substrate temperature (-45 °C). These films are dominated (68%) by a long lifetime component with a

recombination lifetime of 1 1.5±1 ms. The water incorporation was reduced further by combining the mild substrate heating with the illumination of the incoming flux of molecules that were forming the film with Infrared light from a tungsten halogen bulb.

Figure 3b shows the measured Er ³⁺ ion lifetime for a film with a similar substrate

temperature but where the incoming flux was illuminated. In this film -88% of the ions have the long lifetime component. This lifetime is comparable to the measured radiative lifetime of 12.6 ms for a single crystal of Er(F-TPIP) ₃ and is indicative that these ions have a Quantum Efficiency (QE) of approximately 100%. These very high QE films were grown in a vacuum with a partial pressure of water of only -2 x 10 ^~6 mbar where standard growth conditions would give an average lifetime of -1 ms. Without wishing to be bound by theory, it is believed that the cause of these exceptionally long lifetimes may be a further reduction in the sticking coefficient of water that has been illuminated, perhaps due to the irradiated water being in an excited state when it impinges upon the growing surface, which further reduces its sticking coefficient.

Methods

Film growth

All films (nominally 100 nm) were grown by vacuum sublimation onto clean glass substrates followed by the evaporation of a thin (100 nm) layer of aluminium which served to reflect the incident laser light back through the film to increase signal intensity. For the deposition two vacuum systems were used. For“low vacuum” evaporation (10 ^~6 mbar to 5 x 10- ⁷ mbar) a chamber with a turbomolecular pump and liquid nitrogen cold finger was used. For films grown under ultra-high vacuum conditions (down to 10 ¹⁰ mbar) a Chi-Vac Research and Development Co. Ltd. Deposition system was used. Elevated temperature deposition was performed in the home-made system with a sample heater consisting of a tungsten halogen bulb placed behind the glass substrate with a piece of silicon in contact with the substrate to improve heating. Films of Ero. ₂ Yo. ₈ (F-TPIP) ₃ were grown both by co- evaporation of Er(F-TPIP) ₃ and Y(F-TPIP) ₃ using different sources and by evaporating a Er _0.2 Yo. ₈ (F-TPIP) ₃ powder which was formed by reacting a solution of 20% ErCI ₃ and 80% YCI ₃ with FIFTPIP. The resulting Er _0.2 Yo. ₈ (F-TPIP) ₃ was then purified by train sublimation before use. Samples of all films were characterised for stoichiometry by Energy Dispersive X-ray Spectroscopy (EDS) and Rutherford Backscattering Spectroscopy (RBS) and found to have the correct composition. 20% erbium doped films were used to minimise the possibility of Er-Er interactions quenching the luminescence decay but at the low excitation intensities used we found identical results on comparable films of pure Er(F-TPIP) ₃ .

For substrate heating a tungsten halogen bulb was placed behind the sample holder. A 4 cm ² piece of silicon was placed directly on top of the glass substrate in order to absorb the infrared light from the lamp and heat the glass. To calibrate the temperature of the glass as a function of lamp power a series of thermochromic labels were placed on the glass and the maximum temperature calibrated against lamp power. The sticking coefficient of the Er _0.2 Y _0.8 (F-TPIP) ₃ is temperature sensitive at elevated temperatures and we found that for substrate temperatures greater than -60 °C the films were -10% thinner than the film thickness monitor indicated and for temperatures in excess of -70 °C the films were -90% thinner than indicated. At the temperature used in this work (-45 °C) there was no measurable change in thickness.

Initial measurements of the effect of IR illumination of the incoming Er _0.2 Y _0.8 (F-TPIP) ₃ plume was made by removing the silicon from the glass and allowing the light to pass directly through the substrate and illuminate the incoming molecular flux. Measurements of the real thickness of these films indicate that the substrate temperature was less than -50 °C.

Luminescence characterisation

Samples were characterised for their luminescence lifetime by exciting them with an -6 ns pulse at 520 nm from a Continuum Panther Optical Parametric Oscillator. The excitation wavelength was chosen to provide optimal absorption into the ⁴ h _5/2 ² H _11/2 transition of the Er ³⁺ ion. The decay from the photoexcited state to the emissive state was faster than the system response used (-30 ps) which was determined by the load resistance on the photomultiplier and the input resistance of the oscilloscope. These were adjusted for maximum gain rather than time response. Luminescence was recorded at room

temperature and dispersed in a Triax 550 spectrometer and detected using a Hamamastu R5509-73 cooled photomultiplier. Each decay was an average of 2500 pulses to reduce the background noise in the signals.

Analysis

The number density of H ₂ 0 molecules in the film was estimated using a combination of a FRET model and with a Nearest Neighbour probability calculation and comparing this to the experimental values.

The FRET method predicts a distance R _j at which 50% of excited donor (D) molecules transfer energy into the acceptor molecule (A). The rate of energy transfer is given by equation 1 :

Where k _nr is the non-radiative energy transfer between the donor and the acceptor, k _r is the radiative transition rate, R _j is the FRET distance, and h is the real distance between the donor and the acceptor.

To find R _j between the donor, an erbium ion, and the acceptor, an H ₂ 0 molecule, equation 2 was used:

Where F _D (A) is normalised emission spectra of the donor Er ion, e _A (l) is the molar attenuation coefficient of H ₂ 0 in units of M ¹ cm ¹ , l is wavelength in nm, KJ is the geometric factor, taken as 2/3, n is the refractive index, Q _D is the quantum yield of the donor in the absence of the acceptor, taken as 1.

The emission spectra of the Er(F-TPIP) ₃ was measured experimentally and the H ₂ 0 molar attenuation was calculated from water absorption cross-section data taken from FIITRAN04 database. The overlap integral F _Er3+ (l)e _Hz0 (l)l ⁴ άl was calculated geometrically as the area under the product curve. The limits of integration (1420 nm - 1580 nm) were selected to include only ⁴ li _3/2 ⁴ li _5/2 transition.

FRET radius was found to be R _j = 9.65 ± 0.35 A. This corresponds to a radial distance between an erbium ion and its nearest H ₂ 0 of -4.8A for a -200 pm lifetime, and -31.9 A for a -12.7 ms lifetime. A Nearest neighbour probability calculation was used to determine the percentage contribution of each distinct decay lifetime component with respect to radial separation between an erbium ion and H ₂ 0 molecules as a function of the density of water molecules in the film.

The 3D NN probability formula is given in equation 3:

Where r is the radial distance between an erbium ion and the nearest H ₂ 0 molecule, and p is the H ₂ 0 molecule density in the film.

These calculations assume that there is no restriction as to where a water molecule can be found in relation to a given erbium ion. However, in a real film the majority of the space is filled with the atoms of the ligands that surround the erbium ion. We have therefore use x- ray diffraction data for the Er(FTPIP) ₃ molecule to calculate the amount of empty space in a crystal of Er(FTPIP) ₃ as a percentage of the total volume of the crystal. Using this approach, we find that at most 30% of the total volume is free to accommodate water molecules within a solid film. Our calculated water concentrations are therefore scaled by 0.3 to give the effective H ₂ 0 density in the film.

From the experimental data three distinct radiative decay processes were measured with lifetimes of 0.19 ± 0.04 ms; 0.52 ± 0.07 ms; 10.6 ± 2.4 ms. According to these lifetimes and uncertainties FRET calculations were used to estimate integration limits for the probability function. This gave integration limits of (0 - 5.2 A), (5.2 A - 7.8 A) and (7.8 A - _° °), and the probability contribution for each range was calculated for water densities between 10 ¹⁹ cm ³ and 10 ²¹ cm ³ . The percentage contributions to the longest lifetime were then used in each film to determine the approximate water concentration present.

Further results

The above described results are compared to measurements on a single crystal of Er(FTPIP) ₃ for the radiative lifetime. Further measurements were performed on a series of bulk samples of Er(FTPIP) ₃ . The bulk samples each contained a different number of erbium ions. Figure 5 shows an absorption spectra for the ⁴ li _5/2 ⁴ li3/2 transition in erbium in a sample of Er(F-TPIP) ₃ . Six samples were prepared with different amounts of Er(F- TPIP) ₃ and these were globally fitted to obtain an average radiative lifetime for the erbium ions of 2.8 ± 0.7 ms, assuming no impurities are present. It is noted that where impurities (e.g. water) are present this can have a quenching effect to give a value for the average radiative lifetime of 2.4 ± 0.7 ms.

Further measurements of a layer of Er(FTPIP) ₃ deposited according to the described method show an improvement in the quantum yield when compared to material deposited via commonly known methods. Figure 6 shows measurements of the time resolved emission from two layers each comprising an organic erbium complex and a chromophore and deposited at either 25 °C or 55 °C respectively. The layers comprise nominally identical samples of Er(F-TPIP) ₃ , grown in a vacuum with a base pressure of -10 ⁸ torr and with a partial pressure of water ~2 x 10 ^~9 torr. The time resolved emission from erbium ions (at a wavelength of 1532 nm) in the two layers allows for quantification of the average lifetimes. The 55 °C growth shows an average lifetime of ~2.4 ± 0.26 ms compared with ~1 .1 ± 0.12 ms for the sample grown at 25 °C. As such, a clear increase in the average lifetime is apparent as a result of heating of the substrate during formation of the organic erbium complex layer.

The quantum yield of the deposited material can be found by comparison of the measured lifetimes for the two layers compared to the average radiative lifetime of erbium ions in the measured bulk samples, described above. An initial comparison of the measured lifetimes with the average radiative lifetime of the bulk erbium ions (2.8 ms) appears to suggest a quantum yield of -85% for the erbium complex layer grown at 55 °C, compared to a quantum yield of -39% for the layer grown at 25 °C.

However, this appears to represent too simplistic an approach. Further consideration of Figure 6 shows at least two components in the fitting line of the time resolved emission. A longer component with a lifetime approaching 9 ms is visible in the time resolved emission spectra of the layer grown at 55 °C. This component is attributed to the presence of the chromophore in the deposited layer, which transfers energy to the erbium atoms long after they would have decayed naturally. Nevertheless, removing the long component from the fitting to the time resolved emission allows for estimation of a quantum yield of -50% for the erbium in the layer grown at 55 °C. Similar analysis for the layer grown at 25 °C results in a quantum yield of -35%. As such, the claimed method for forming the layers comprising organic erbium complexes (in particular, layers formed whilst heating the substrate to a temperature of between 30 °C and 200 °C) is shown to significantly increase the quantum yield.

The above-described measurements were performed under a high vacuum with a very low partial pressure of water. As such, the overall effect on the quantum yield is less pronounced that earlier described measurements on layers formed in a vacuum where more water molecules may be present.

It should be noted that the inventors have observed that the described heating of the substrate during forming of a layer (comprising the organic erbium complex and the chromophore) appears to improve the performance of the chromophore as well as the quantum yield of the erbium. In other words, the described method appears to provide advantages for improving the lifetime and quantum yield of the chromophore in addition to that of the erbium ion.

We note that the above described experimental examples relate to organic erbium complexes. Nevertheless, it is understood that the advantages (and in particular, the improved quantum yield for the metal ion) is apparent for any layer comprising an organic complex comprising a rare earth ion. In particular, the rare earth ions may be a lanthanide or other chemically similar element. Thus, the described method may be applied for forming a layer comprising any organic lanthanide complex or organic rare earth ion complex on a substrate, and is not limited to layers comprising an organic erbium complex.