Statement on Funding This work was performed within the Norwegian Research Centre for Solar Cell Technology as project number 193829. The Norwegian Research Centre for Solar Cell Technology is a centre for environment-friendly energy research co-sponsored by the Research Council of Norway and research and industry partners in Norway. Field of the invention

The present invention relates to luminescent materials comprising multivalent donor and acceptor cations, in which quenching via inter-metal charge transfer is suppressed by the provision of a thin layer of barrier material between the donor and acceptor. The present invention may find particular application in fluorescent lights which normally use Tb ³⁺ and Ce ³⁺ for the green and blue part of the white spectrum respectively, lasers and other optical systems, in addition to more advanced optical systems such as down conversion systems for solar cells.

Background

Luminescent compositions (such as so-called "phosphors", which - despite their name - may be phosphorescent or fluorescent) based on donor-acceptor interactions are well-known systems which find application in many fields including, but not limited to, CRT display screens, plasma displays, fluorescent lighting, sensors (e.g. optical sensing applications), LEDs (particularly white LEDs), and other applications where "down-shifting" or "down-conversion" of electromagnetic radiation of a first wavelength to electromagnetic radiation of a second, longer wavelength (e.g. conversion of UV radiation to visible or infrared radiation) is desirable. For example, phosphors which downconvert UV radiation into near-IR radiation find particular applicability in solar cells, where they help to maximise the amount of energy captured from solar radiation. "Down-conversion" relates to the conversion of a higher-energy photon into two or more lower-energy photons. "Down-shifting" refers to the conversion of a higher-energy photon into a lower-energy photon, i.e. in a one-to-one ratio. These terms are often (erroneously) used interchangeably in the literature.

A simplified model of the photophysics of such systems is as follows. The donor (denoted "D") is a species which absorbs a photon of a first wavelength, leading to the formation of an excited electronic state (denoted "D*"). The acceptor (denoted "A") is a species having sufficient absorption spectrum overlap with the emission spectrum of the donor, and being located within a sufficiently close distance to the donor, that nonradiative energy transfer can take place from the donor to the acceptor via Forster resonance energy transfer (FRET). The donor returns to its ground state, transferring energy to the acceptor and thus placing the acceptor into an excited electronic state (denoted "A*"). Subsequently, the acceptor returns to its own ground state, emitting the excitation energy as a photon of longer wavelength:

D + hv→ D*

D* + A→ D + A*

A*→ A + hv

Lanthanide ions, in particular Ce ³⁺ , Tb ³⁺ and Pr ³⁺ , are important luminescent ions which can be used as acceptors, for instance in lighting phosphors for fluorescent lights and LEDs. Lanthanide ions, in particular Ce ³⁺ , Tb ³⁺ , Yb ³⁺ and Pr ³⁺ , are also the most promising candidates for down-conversion of UV light in solar cells. However, strong UV absorption is crucial for both lighting phosphors and down-conversion. The UV absorption of Pr ³⁺ and Tb ³⁺ is very weak, due to the forbidden nature of their f-f transitions, and thus a "sensitizing" (i.e. donor) host material is usually used to provide proper optical absorption. The luminescence of Ce ³⁺ comes from an allowed f-d transition. This has an absorption which is strong enough by itself in many situations without requiring a sensitizer, but it can nevertheless also be beneficial to use a sensitizing host with stronger absorption so that less cerium is needed.

Many common sensitizing host materials for luminescent systems are based around very strong charge transfer absorption in d° transition metals, for instance (TiO ₆ ) ^8" , (VO ₄ ) ^3" and (Mo0 ₄ ) ^4" , in which charge transfer between the metal cation and its negatively-charged oxygen atom neighbours takes place. However, it has been found that these types of sensitizing host materials such as the common sensitizers YV0 ₄ and others which consist of at least one part capable of accepting an electron are poorly suited to lanthanides such as Ce ³⁺ , Pr ³⁺ and Tb ³⁺ . This is due to a low-lying inter-metal charge transfer (IMCT) in which a charge transfer takes place between the sensitizer metal ion and the lanthanide ion. In luminescent materials based on donor and acceptor ions, an electron charge transfer is possible between these ions if they are close enough to one another. If the charge transfer state also lies below the emissive state of the acceptor ion in energy, the emission of electromagnetic radiation from the acceptor may be fully quenched by this process.

In order to avoid quenching via IMCT it has often been necessary to rely on other types of donors and acceptors. However, this limits the choice of materials available and often forces the use of expensive, toxic or unstable materials. In addition, some systems which are desirable in theory simply do not work. For example, using Ce ³⁺ as a donor and Yb ³⁺ as an acceptor has been proposed as an efficient down-conversion system for solar cells, but the down conversion process is fully prevented by charge transfer from Yb ³⁺ to Ce ³⁺ (Ce ³⁺ + Yb ³⁺ → Ce ⁴⁺ + Yb ²⁺ ). It is therefore desirable to develop a donor-acceptor system which allows preferred donor and acceptor combinations to be employed but in which the IMCT quenching route is suppressed.

Description of the invention

The present inventors have now surprisingly found that compositions according to the invention are effective at suppressing charge-transfer quenching. The present invention therefore opens up a wider possible range of choices of donor/acceptor combinations and therefore allows access to better, cheaper and more robust systems. In addition, the present invention allows the development of new, advanced donor/acceptor systems, such as the Ce ³⁺ /Yb ³⁺ combination discussed above, which have previously been unachievable. The present invention exploits the differing distance dependencies of the (desired) FRET and

(undesired) IMCT energy transfer pathways to permit forward energy transfer from the donor to the acceptor via FRET to occur without obstruction but which blocks the undesirable IMCT de-activation pathway. FRET is a dipole-dipole interaction which can be efficient over several nanometers (a distance denoted herein as c ergy), typically up to about 10 nm, due to the strength of the dipole-dipole interaction varying as 1/r ⁶ where r is the distance between the two dipoles. By contrast, electron charge transfer requires the two cations to be next-cationic-neighbors in order to be efficient. The effective range for IMCT is generally a few tenths of a nanometer (a distance denoted herein as deiectron). This difference can be utilized to prevent the short range charge transfer while allowing the energy transfer, by providing a thin barrier layer of material between the donor and acceptor ions. If the barrier has a thickness "dba er" such that deiectnm < dbamer < d^gy, electron transfers between the donor and acceptor ions can be prevented while energy transfer via dipole-dipole interactions can take place. The barrier may therefore be referred to as a "charge transfer blocking layer". When such a charge transfer blocking layer is used for separation of the donors from the acceptors, undesired quenching may be substantially avoided while the desired energy transfer between the donors and acceptors may remain efficient. Charge transfer blocking layers may suitably be formed by atomic layer deposition (ALD), which in turn allows production of robust, uniform and versatile multilayer coatings with good adherence to a variety of surfaces.

Viewed from a first aspect the present invention therefore provides a luminescent composition comprising a donor and an acceptor, wherein: the donor comprises a first multivalent cation and the acceptor comprises a second multivalent cation;

each of the first multivalent cation and the second multivalent cation is independently a transition metal cation, a lanthanide cation, a post-transition metal cation or a metalloid cation with the proviso that the first multivalent cation and the second multivalent cation are different from one another;

and wherein the donor and acceptor are separated from one another by a charge transfer blocking layer. Due to its presence in the donor, the first multivalent cation may be referred to herein as a "donor ion" or "donor cation". Similarly the second multivalent cation may be referred to as an "acceptor ion" or "acceptor cation".

Although the background discussion above has centred around particular combinations of lanthanides and specific transition metal complexes as donors and/or acceptors, in principle the charge transfer quenching route can take place in any donor/acceptor system where both the donor and acceptor each have two or more accessible positive (i.e. charge > 0) oxidation states, i.e. where both the donor and acceptor cations are multivalent cations, provided that the donor-acceptor charge transfer state (or IMCT) lies below the emissive state of the acceptor. Thus, the donor may in principle comprise a transition metal cation, lanthanide cation, post-transition metal cation or metalloid cation provided that the cation is a multivalent cation. A multivalent cation is a cation having more than one accessible positive charge state, i.e. two, three or more states having charges greater than zero (such as +3 and +4, or +2, +3 and 6, for example). Such states should be accessible under conditions which are compatible with the use of the cationic species in the luminescent compositions of the invention. Elements which are theoretically capable of existing in more than one positive charge state, but in which such additional states are unstable and/or only accessible under extreme conditions, are not "multivalent" for the purposes of this invention. For example, zinc compounds predominantly exist as zinc (II) compounds, with zinc (I) being extremely rare and only normally accessible in

organometallic compounds with a Zn-Zn bond which undergoes rapid disproportionation, therefore zinc cations are not "multivalent" for the purposes of the present invention. Similarly, yttrium compounds predominantly exist as yttrium (III), with yttrium (I) and yttrium (II) only accessible under exotic conditions such as plasmas, gaseous oxide clusters, or molten chlorides, thus yttrium cations are not "multivalent" for the purposes of the present invention. Among the transition metals, those having a d° electronic configuration are preferred. In certain embodiments of the invention the first multivalent cation is thus a lanthanide cation or a transition metal cation having a d° electron count. Preferred lanthanide cations are Ce ³⁺ , Pr ³⁺ , Tb ³⁺ , and Eu ²⁺ . Transition metals of groups 4, 5 and 6 are particularly preferred for use as the first multivalent cation, and especially the cations of Ti, V, Cr, Zr, Nb, Mo, Hf, Ta and W having a d° electronic

configuration. In certain embodiments of the invention the first multivalent cation is selected from Ti ⁴⁺ , V ⁵⁺ , Cr ⁶ *, Cr ³⁺ , Zr ⁴⁺ , Nb ⁵⁺ , Mo ⁶⁺ , Mo ³⁺ , Hf ⁴ *, Ta ⁵⁺ , W** and W ³⁺ , and preferably from among Ti ⁴⁺ , V ⁵⁺ , Cr^ , Zr ⁴⁺ , Nb ⁵⁺ , Mo ⁶⁺ , Hf ⁴ *, Ta ⁵⁺ , and W ⁶ *. The first multivalent cation in certain embodiments of the invention may also be Mn ²⁺ or Mn ⁴⁺ .

Multivalent post-transition metal cations may also be employed as the first or second multivalent cation. As used herein, "post-transition metals" are Ga, In, Tl, Sn, Pb and Bi. Of the post-transition metal cations, Bi ³⁺ is especially preferred as a first multivalent cation.

Multivalent metalloid cations may also be employed as the first or second multivalent cation. As used herein, "metalloids" are Ge, As, Sb and Te. Of the metalloid cations, Sb ³⁺ is of particular interest.

The undesirable IMCT quenching pathway may in principle occur in either direction, i.e. with an electron transfer from acceptor to donor, or with an electron transfer from donor to acceptor. The direction of electron transfer will depend on the relative oxidising and reducing ability of the first and second multivalent cations. Thus in certain embodiments of the invention the first multivalent cation is an oxidising agent relative to the second multivalent cation. In other embodiments of the invention the first multivalent cation is a reducing agent relative to the second multivalent cation.

In certain embodiments of the invention the second multivalent cation may be selected from Ce ³⁺ , Pr ³⁺ , Tb ³⁺ , Eu ²⁺ , Mn ²⁺ , Sm ³⁺ , Eu ³⁺ ,Yb ³⁺ , W ⁶ *, Cr ³⁺ , Mn ⁴⁺ , Bi ³⁺ and Sb ³⁺ .

Relationships between suitable donor/acceptor pairs having complementary oxidizing/reducing behavior are shown in Table 1 below. In principle any combination of an oxidizing donor and reducing acceptor, or of a reducing donor and oxidizing acceptor, may be employed with the proviso that the donor and acceptor are not identical. This does not preclude using donors and acceptors which are different oxidation states of the same element, e.g. Eu ³⁺ and Eu ²⁺ .

Table 1

Thus it can be seen that certain ions can be oxidizing or reducing, and may function as either donors or acceptors, depending on the counterpart which is chosen. The skilled person will be able to choose appropriate combinations according to the desired emission characteristics and or end use of the composition.

As outlined in table 2, several common donor/acceptor ions can be grouped according to their expected solid-state red-ox potential:

Table 2

Accordingly, the present invention provides a solution to quenching problems which are often encountered if attempting to use combinations of oxidizing ions with reducing ions, and in particular when combining donor and acceptor ions from Table 2 which differ by three or more categories from one another. Such systems are particularly prone to charge transfer occurring due to the high thermodynamic favourability of a redox reaction arising from the difference in oxidizing/reducing potential and therefore the inclusion of a charge transfer blocking layer in accordance with the invention is particularly beneficial in such systems.

For example, a charge transfer blocking layer would be beneficial in the following situations, all of which constitute embodiments of the invention: the first multivalent cation is chosen from category 1 and the second multivalent cation is chosen from category 4, 5 or 6, or vice versa; the first multivalent cation is chosen from category 2 and the second multivalent cation is chosen from category 5 or 6, or vice versa; the first multivalent cation is chosen from category 3 and the second multivalent cation is chosen from category 6, or vice versa; the first multivalent cation is chosen from category 6 and the second multivalent cation is chosen from category 1, 2 or 3, or vice versa; the first multivalent cation is chosen from category 5 and the second multivalent cation is chosen from category 1 or 2, or vice versa; the first multivalent cation is chosen from category 4 and the second multivalent cation is chosen from category 1, or vice versa. In all such embodiments of the invention the charge transfer blocking layer is preferably an inorganic charge transfer blocking layer with a thickness in the range of 0.1 to 3.0 nm, such as 0.2 to 2.5 nm, 0.3 to 2.0 nm, 0.4 to 1.5 nm or 0.5 to 1.0 nm. Thus in certain embodiments the invention provides a luminescent composition comprising at least one donor and at least one acceptor separated from one another by an inorganic charge transfer blocking layer, wherein the first multivalent cation is an oxidizing agent relative to the second multivalent cation, and wherein the first and second multivalent cations differ by three or more categories from one another according to Table 2 as defined above, and wherein the thickness of the inorganic charge transfer blocking layer is in the range of 0.1 to 3.0 nm, such as 0.2 to 2.5 nm, 0.3 to 2.0 nm, 0.4 to 1.5 nm or 0.5 to 1.0 nm. The present inventors have found that layers having sub-nanometer thickness, i.e. 0.1 to 1 nm, e.g. 0.2 to 0.5 nm, are in many cases sufficient to act as charge transfer blocking layers for the purposes of the present invention.

Surprisingly, it has been found that even a charge transfer blocking layer which is only 1 atomic layer thick can be suitable for this purpose and therefore in preferred embodiments of the invention the charge transfer blocking layer may be 1 to 10 atomic layers thick, preferably 1 to 5 atomic layers thick or 1 to 4 atomic layers thick. Examples of combinations of donor and acceptor cation pairs which may particularly benefit from the present invention are:

Donor cations selected from among Ce and Eu in combination with acceptor cations selected from among Sm ³⁺ , Eu ³⁺ and Yb ³⁺

Donor cations selected from among Ti ⁴⁺ , V ⁵⁺ , Ta ⁵⁺ , Cr ⁶⁺ , Mo ⁶⁺ , W ⁶ * in combination with acceptor cations selected from among Ce ³⁺ , Pr ³⁺ , Tb ³⁺ , Eu ²⁺ , Mn ²⁺ , Bi ³⁺ .

Donor cations selected from among Ti ⁴⁺ , V ⁵⁺ , Ta ⁵⁺ , Cr ⁶⁺ , Mo ⁶⁺ , W ⁶ * in combination with acceptor cations selected from among Ce ³⁺ , Pr ³⁺ , Tb ³⁺ .

Donor cations selected from among the d° transition metals in combination with acceptor cations selected from among the lanthanides, in particular Ce ³⁺ , Pr ³⁺ , and Tb ³⁺ .

Donor cations selected from among the lanthanides, in particular Ce , Pr , Tb , Eu , or Bi ³⁺ , in combination with acceptor cations selected from among the lanthanides, in particular

Eu ^J+ and Yb

A donor cation which is Ce in combination with an acceptor cation which is Yb .

• Donor cations selected from Ti ⁴⁺ , V ⁵⁺ , Cr^ ,Zr ⁴⁺ , Nb ⁵⁺ , Mo ⁶⁺ , Hf ⁺ , Ta ⁵⁺ and W ⁶

combination with acceptors selected from among Ce ³⁺ , Pr ³⁺ and Tb ³⁺ .

All of the above-listed exemplary combinations form embodiments of the invention.

Other preferred embodiments of the invention include those wherein the donor cation is selected from Ti ⁴⁺ , V ⁵⁺ , Cr ⁶ ^, Mn ⁷⁺ , Zr ⁴⁺ , Nb ⁵⁺ and Mo6+, and wherein the acceptor cation is selected from Ce ³⁺ , Pr ³⁺ , Tb ³⁺ , Yb ²⁺ and Mn ²⁺ , preferably wherein the charge transfer blocking layer has a thickness in the range of 0.2 to 1.0 nm and is an inorganic charge transfer blocking layer; and those wherein the donor cation is selected from Hf ⁺ Ta ⁵⁺ , W ^6"1" , Eu ³⁺ , Yb ³⁺ , Cr ³⁺ , Bi ³⁺ , Sb ³⁺ and Mn ⁴⁺ , and wherein the acceptor cation is Ce ³⁺ , preferably wherein the thickness of the charge transfer blocking layer is in the range of 0.2 to 1.0 nm and the charge transfer blocking layer is an inorganic charge transfer blocking layer. In one embodiment the present invention provides a luminescent composition wherein the donor cation is selected from Ti ⁴⁺ , V ⁵⁺ , Cr ⁶⁺ , Mn ⁷⁺ , Zr ⁴⁺ , Nb ⁵⁺ and Mo ⁶⁺ , and wherein the acceptor cation is selected from Ce ³⁺ , Pr ³⁺ , Tb ³⁺ , Yb ²⁺ and Mn ²⁺ , and wherein the thickness of the inorganic charge transfer blocking layer is in the range of 0.2 to 1.0 nm.

In one embodiment the present invention provides a luminescent composition wherein the donor cation is selected from Hf ^t+ Ta ⁵⁺ , W* ^" , Eu ³⁺ , Yb ³⁺ , Cr ³⁺ , Bi ³⁺ , Sb ³⁺ and Mn ⁴⁺ , and wherein the acceptor cation is Ce ³⁺ , and wherein the thickness of the inorganic charge transfer blocking layer is in the range of 0.2 to 1.0 nm. The IMCT quenching mechanism discussed above is particularly prominent in systems that rely on a donor having a large charge transfer absorption (e.g. T1O ₂ , YVO ₄ ) and an acceptor which is a lanthanide or other acceptor species that can donate an electron to the absorbing species (e.g. Ce ³⁺ , Pr ³⁺ , Eu ²⁺ , Tb ³⁺ ). Thus, donor-acceptor pairs in which the donor comprises the first multivalent cation in the form of an oxide or organometallic compound are particularly preferred as the metal-ligand charge transfer interaction within such donor species enhances their efficiency of absorption of incident light. In particular, such donor-acceptor pairs are preferred in which the acceptor is a lanthanide, especially Ce ³⁺ , Pr ³⁺ , Eu ²⁺ or Tb ³⁺ .

Particularly preferred oxides include titanates, vanadates, chromates, tantalates, molybdates, and tungstates. Suitable organometallic compounds include ruthenium metal complexes, particularly those having two bipyridyl coordinating moieties. Particularly preferred ruthenium complexes include those of formula (I):

wherein each RI group is independently a straight or branched chain alkyl or oligoalkoxy chain such as C-H^ ₊ i where n is 1 to 20, preferably 5 to 15, most preferably 9, 10 or 11, or such as

C-(-XC _n H2n-)m-XCpH2pH, where n is 1, 2, 3 or 4, preferably 2, m is 0 to 10, preferably 2, 3 or 4, p is an integer from 1 to 15, preferably 1 to 10, most preferably 1 or 7, and each X is independently O, S or ΝΉ, preferably O; and wherein each R2 group is independently a carboxylic acid or alkyl carboxylic acid, or the salt of any such acid (e.g. the sodium, potassium salt etc) such as a

C _n Eb _n COOY group, where n is 0, 1, 2 or 3, preferably 0 and Y is H or a suitable metal such as Na, K, or Li, preferably Na; and wherein each R3 group is single or double bonded to the attached N

(preferably double bonded) and is of formula CHa-Z or C=Z, where a is 0, 1 or 2 as appropriate, Z is a hetero atom or group such as S, O, SH or OH, or is an alkyl group (e.g. methylene, ethylene etc) bonded to any such a hetero atom or group as appropriate; R3 is preferably =C=S. A preferred ruthenium-based donor is of the above formula (I), wherein each RI is nonyl, each R2 is a carboxylic acid or sodium salt thereof and each R3 is double-bonded to the attached N and of formula =C=S.

Ruthenium dyes are discussed in many published documents including, for example, Kuang et al. Nano Letters 6 769-773 (2006), Snaith et al. Angew. Chem. Int. Ed. 44 6413-6417 (2005), Wang et al. Nature Materials 2, 402-498 (2003), Kuang et al. Inorganica Chemica Acta 361 699-706 (2008), and Snaith et al. J Phys, Chem. Lett. 112 7562-7566 (2008), the disclosures of which are hereby incorporated herein by reference, as are the disclosures of all material cited therein. Other sensitizers which will be known to those of skill in the art include metal-phthalocyanine complexes such as zinc phthalocyanine PCHOOl, the synthesis and structure of which is described by Reddy et al. (Angew. Chem. Int. Ed. 46 373-376 (2007)), the complete disclosure of which

(particularly with reference to Scheme 1), is hereby incorporated by reference, Metal-Porphyrin complexes, Squaraine dyes, Thiophene based dyes, fluorene based dyes, molecular dyes and polymer dyes. Examples of Squaraine dyes may be found, for example in Burke et al., Chem. Commun. 2007, 234, and examples of polyfluorene and polythiophene polymers in McNeill et al., Appl. Phys. Lett. 2007, 90, both of which are incorporated herein by reference. The present invention exploits the fact that forward energy transfer from the donor to the acceptor is non-radiative and takes place via dipole-dipole interactions which transfer energy in a through-space interaction having a distance dependency proportional to 1/r ⁶ without any emission of radiation or transfer of electrons; in contrast, the charge transfer mechanism, which leads to quenching, relies on the physical transfer of an electron (not just the energy) and can only be efficient between next- neighbour cations. The charge transfer blocking layer is a physical barrier of material between the donor and acceptor cations which therefore separates the cations such that charge transfer is not possible but which still permits non-radiative energy transfer to occur from the donor to the acceptor. The charge transfer blocking layer may in certain embodiments be up to about 10 nm in thickness, which in many cases corresponds to the maximum distance over which effective non-radiative energy transfer can take place. However, the present inventors have found that thinner layers, such as those having thicknesses of up to about 5 nm, up to about 3 nm, such as 0.1 to 2 nm, or even sub-nanometer thickness, i.e. 0.1 to 1 nm, or 0.2 to 0.5 nm, are in many cases sufficient to suppress the IMCT quenching pathway. Surprisingly, it has been found that even a charge transfer blocking layer which is only 1 atomic layer thick can be suitable for this purpose and therefore in preferred embodiments of the invention the charge transfer blocking layer may be 1 to 10 atomic layers thick, preferably 1 to 5 atomic layers thick or 1 to 4 atomic layers thick. It is generally preferred to keep the charge transfer blocking layer as thin as possible, in order to avoid adversely affecting the efficiency of the forward energy transfer from donor to acceptor.

As the name suggests, the "charge transfer blocking layer" has the function of preventing the formation of undesirable, emission-quenching charge transfer states between the donor and acceptor cations as described elsewhere herein. If donor and acceptor cations were to be chosen which were not capable of forming emission-quenching charge transfer states (e.g. if the donor was not sufficiently oxidising relative to the acceptor, or vice versa), then a barrier between those cations would not count as a "charge transfer blocking layer" as the presence or absence of that barrier would not, in fact, have any effect on the formation or non-formation of a charge transfer state. Thus, for example, if a Zn ²⁺ / Mn ²⁺ (e.g. ZnS/MnS) or Y ³⁺ / Eu ³⁺ (e.g. Y2O ₃ EU2O ₃ ) donor/acceptor pair were to be separated by a barrier layer such as S1O2, that barrier layer would not be a "charge transfer blocking layer" because the Zn ²⁺ /Mn ²⁺ and Y ³⁺ /Eu ³⁺ systems are not capable of forming emission-quenching charge transfer states even in the absence of the barrier. (As noted elsewhere above, such systems are also outside the scope of the present invention due to neither Zn nor Y being "multivalent"). The charge transfer layer should preferably be of uniform thickness. In general this may be understood as meaning that the thickness of the layer at any two points where it separates the donor and acceptor should not deviate by more than ±0.5 nm, preferably not more than ±0.25 nm and particularly preferably not more than ±0.1 nm. However, "holes" or discontinuities in the charge transfer blocking layer can be tolerated provided that these represent only a small fraction of the overall surface area, such that overall the charge transfer blocking layer is still effective at eliminating the vast majority of charge-transfer mediated quenching. Thus, while the charge transfer layer is preferably continuous (i.e. contains no discontinuities where charge transfer could occur), it may in certain embodiments cover all or substantially all (e.g. at least 85%, preferably at least 90%, particularly preferably at least 95%, and especially at least 98% of) the surface area common to the donor and acceptor.

As used herein, a charge transfer blocking layer which is 1 atomic layer thick is therefore a monolayer covering all or substantially all (e.g. at least 85%, preferably at least 90%, particularly preferably at least 95%, and especially at least 98% of) the surface on which it is coated.

A monolayer may be produced by one cycle of ALD deposition or may require multiple cycles of ALD deposition, depending on the material deposited, the precursors employed, and the reaction conditions used. The skilled person will be able to select appropriate numbers of ALD cycles to achieve a monolayer having regard to the particular circumstances. The actual thickness, in nm, of an individual atomic layer will depend on the precise compound being used; however, for practical purposes, an atomic layer of any suitable charge transfer blocking material will be at least 0.1 nm in thickness.

Thus in certain preferred embodiments of the invention the donor and acceptor are separated from one another by a charge transfer blocking layer which is a monolayer.

The material of the charge transfer blocking layer may be any suitable material provided that it does not form a charge transfer state with either the donor cation or the acceptor cation which lies below the emissive state of either the donor or the acceptor, as this would re-open a further charge transfer quenching pathway. The charge transfer blocking layer may be an inorganic or organic material. Suitable materials for use as charge transfer blocking layers in all embodiments of the invention include Ti0 ₂ , Hf0 ₂ , A1 ₂ 0 ₃ , Si0 ₂ , Y ₂ 0 ₃ , YF ₃ , LaF ₃ , MgF ₂ , CaF ₂ , NaYF ₄ , YP0 ₄ , A1N, Si ₃ N ₄ (which are self-evidently inorganic materials) and polymers (which are self-evidently organic materials).

Yb ₂ 0 ₃ and YbF ₃ are also appropriate materials for use as charge transfer blocking layers in many embodiments of the invention, but should not be employed if the first multivalent cation is Ce ³⁺ and the second multivalent cation is Tb ³⁺ .

In certain embodiments of the invention the charge transfer blocking layer is passive, which means that it does not interact with either the donor cation or the acceptor cation to form a charge transfer state at all.

In certain embodiments of the invention the charge transfer blocking layer is active, which means that it may form a charge transfer state with the acceptor cation provided that such charge transfer state does not lie below the emissive state of the acceptor cation in energy. In such embodiments the charge transfer blocking layer should not form a charge transfer state with the donor cation where the charge transfer state lies below the emissive states of either the donor or acceptor in energy. In these embodiments, the charge transfer blocking layer still prevents the formation of a charge transfer state between the donor and acceptor, and therefore prevents the undesirable IMCT quenching pathway. The use of an active layer may be desirable, for example, in "down-conversion" applications where it is desired to convert a first photon into two lower-energy photons, for instance when using Ce ³⁺ as donor cation and Yb ³⁺ as acceptor cation. In certain embodiments the charge transfer blocking layer may be Yb ₂ 0 ₃ or YbF ₃ , the first multivalent cation may be Ti ⁴⁺ (e.g. the donor is Ti0 ₂ ) and the second multivalent cation is Pr ³⁺ . In such embodiments the charge transfer blocking layer prevents IMCT between Ti ⁴⁺ and Pr ³⁺ but the Yb ³⁺ in the charge transfer blocking layer can accept energy from Pr ³⁺ without forming a charge transfer state with either Ti ⁴⁺ or Pr ³⁺ that lies below the emissive states of these two cations.. In this manner down-conversion of one photon absorbed by Ti ⁴⁺ into two photons (both emitted by Yb ³⁺ after energy transfer from Pr ³⁺ to Yb ³⁺ ) may be achieved. A Yb ₂ 0 ₃ or YbF ₃ layer may also be employed as a charge transfer blocking layer where the first multivalent cation is Ti ⁴⁺ (e.g. the donor is Ti0 ₂ ) and the second multivalent cation is Tb ³⁺ . A charge transfer state may be formed in which Yb ³⁺ is reduced to Yb ²⁺ and Tb ³⁺ is oxidised to Tb ⁴⁺ , but this lies at too high an energy relative to the emissive state of Tb ³⁺ to cause any quenching.

In certain embodiments of the invention the charge transfer blocking layer is optically transparent. The donor and acceptor may preferably be in the form of layers (e.g. thin films) which are separated by the layer of charge transfer blocking material. The layers may be flat, i.e. planar, or may be curved, and may for example be coated onto a planar substrate such as glass (e.g. the glass of a solar cell). Alternatively the layers may have three-dimensional structure. In another embodiment the donor and acceptor may respectively form the core and cladding of a cable with the charge transfer blocking material layer interposed between the core and cladding, or the acceptor and donor may respectively form the core and cladding of a cable with the charge transfer blocking material layer interposed between the core and cladding. Preferably such layers are homogenous or substantially homogenous in composition.

Accordingly, in certain embodiments, the present invention provides a composition comprising a donor and an acceptor, wherein:

the donor is a layer, preferably a homogenous layer, which comprises a first multivalent cation and the acceptor is a layer, preferably a homogenous layer, which comprises a second multivalent cation; each of the first multivalent cation and the second multivalent cation is independently a transition metal cation, a lanthanide cation, a post-transition metal cation or a metalloid cation, with the proviso that the first multivalent cation and the second multivalent cation are different from one another; and wherein the donor and acceptor layers are separated from one another by a charge transfer blocking layer.

Preferably the donor and acceptor layers are separated from one another only by the charge transfer blocking layer.

Thus in certain embodiments the present invention also provides a luminescent multilayer structure comprising a homogenous layer comprising donor cations,

a homogenous layer comprising acceptor cations,

wherein said layers are only separated by a charge transfer blocking layer,

wherein the donor cation is an oxidizing agent or reducing agent relative to the acceptor cation. Preferably the compositions and structures of the invention are not in the form of quantum dots or nanoparticles.

As ALD is particularly suitable for preparation of the compositions of the invention, and as ALD may conveniently be used for doping a matrix with a dopant, the claimed luminescent compositions may be produced as a single matrix comprising the donor and the acceptor. For example, the donor may be in the form of a layer of matrix material doped with the first multivalent cation, the acceptor may be in the form of a layer of matrix material doped with the second multivalent cation, and the charge transfer blocking layer may be a layer of undoped matrix material. Thus in such embodiments the matrix material employed for the donor layer and for the acceptor layer may be the same as the material which is employed as the material of the charge transfer blocking layer. In certain embodiments of the invention the donor layer may for example be alumina doped with a transition metal ion, the barrier layer may be alumina, and the acceptor layer may be alumina doped with a lanthanide. In other embodiments the matrix materials of the donor and acceptor layers may be different from one another. The material of the barrier layer may be the same as the donor matrix material or the same as the acceptor material or it may be different from both.

The luminescent materials of the invention may comprise a plurality of donor layers and/or a plurality of acceptor layers, provided that any donor and acceptor layers which are present are separated from one another by a layer of charge transfer blocking material. Where a plurality of donor layers are present, the donor ion present in each of the plurality of donor layers may be the same or different. Where a plurality of acceptor layers are present, the acceptor ion present in each of the plurality of acceptor layers may be the same or different.

The compositions of the invention may be manufactured via atomic layer deposition (ALD) techniques known to those skilled in the art. Thus in a further aspect the invention provides a method of manufacturing a luminescent composition, comprising the steps of:

(i) providing a substrate;

(ii) depositing a layer of a first material onto the substrate via ALD;

(iii) depositing a charge transfer blocking layer onto the layer of the first material using ALD; and

(iv) depositing a layer of a second material onto the charge transfer blocking layer via ALD, wherein the first material is a layer of a donor and the second material is a layer of an acceptor, or alternatively wherein the first material is a layer of an acceptor and the second material is a layer of a donor.

Steps (ii), (iii) and (iv) may be independently repeated a desired number of times, for example in order to fabricate a multilayer composition comprising a plurality of layers of donor and/or a plurality of layers of acceptor. When steps (ii), (iii) and (iv) are repeated the material deposited in each repetition of any given step may be the same or different, provided that adjacent layers of donor and acceptor, of donor and charge transfer blocking layer, or of acceptor and charge transfer blocking layer, do not comprise the same material as one another. Thus it is possible to build up a multilayer composition comprising a plurality of donors and/or a plurality of acceptors, for example (e.g. in order to fabricate a multicolour phosphor comprising a plurality of acceptor cations).

The substrate may for example be glass, silicon, or silicon dioxide. The donor and acceptor and the first and second multivalent cations may self-evidently be any of the embodiments of such materials described elsewhere herein. Similarly the charge transfer blocking layer may be any of the embodiments described herein. Each of the steps of depositing the donor layer, acceptor layer and/or charge transfer blocking layer may be independently repeated as often as desired. In certain embodiments of the invention each of such steps may be independently repeated as often as necessary to provide a multilayer structure comprising a plurality of donor layers and/or a plurality of acceptor layers. Optionally and preferably, an annealing step may be performed after completing the desired number of ALD deposition steps. It has been found that annealing can, in many cases, enhance the emission intensity obtained from the compositions by at least an order of magnitude. Annealing may be performed at a temperature of 500 °C or above, preferably 600 °C or above, 700 °C or above, 750 °C or above, or 800 °C or above, preferably at no more than 1500 °C, no more than 1200 °C or no more than 1000 °C. Annealing may be performed in air, other atmospheres or in vacuum.

Potential applications of materials according to the invention include the following:

• Phosphor thermometry: Good sensitization of Tb ³⁺ and Pr ³⁺ is difficult, apart from through organic ligands which will decompose at high temperatures. The present invention allows easier sensitization of these ions (and also Ce ³⁴ ), allowing a combination of several phosphor materials at once which increases the temperature range and sensitivity of the measurement. Reliable surface temperatures of internal engine parts like gas turbine blades can hardly be measured by conventional thermometry during operation. However, phosphor thermometry may instead be used to measure the temperature when contact thermometers cannot be used, e.g. in relation to internal engine parts, moving engine parts, catalysts in reaction vessels etc. Such methods may rely on luminescent phosphor powders which have to be attached to the surface by a binder and they are prone to flaking due to temperature cycling. Thus, a luminescent composition as claimed herein may provide several benefits when used for phosphor thermometry. For example, ALD may be used to coat any engine part, completely or partly, with the luminescent composition as claimed herein. Such coated engine part may be utilized in a phosphor thermometry system with several advantages including increased robustness, uniformity and versatility. The luminescent composition may be applied as a multilayer coating with improved adherence on a variety of surface materials and surface geometries, even turbine blades. Reduced coating thickness and reduced friction of the coating may also be advantages connected to claimed composition. Finally, a protective coating may also be applied over the claimed luminescent composition by ALD or any other suitable coating method. Thus in certain embodiments the present invention provides an internal engine part or an internal reaction vessel part comprising a coating covering the part (partly or completely), wherein the coating comprises the luminescent composition as described herein, and particularly wherein the luminescent composition is a luminescent composition wherein the donor cation is selected from Ti ⁴⁺ , V ⁵⁺ , Cr ⁶⁺ , Mn ⁷⁺ , Zr ⁴⁺ , Nb ⁵⁺ and Mo^, and wherein the acceptor cation is selected from Ce ³⁺ , Pr ³⁺ , Tb ³⁺ , Yb ²⁺ and Mn ²⁺ , and wherein the thickness of the inorganic charge transfer blocking layer is in the range of 0.2 to 1.0 nm, or wherein the donor cation is selected from Hf ⁺ Ta ⁵⁺ , W ^6"1" , Eu ³⁺ , Yb ³⁺ , Cr ³⁺ , Bi ³⁺ , Sb ³⁺ and Mn ⁴⁺ , and wherein the acceptor cation is Ce ³⁺ , and wherein the thickness of the inorganic charge transfer blocking layer is in the range of 0.2 to 1.0 nm. Particularly preferably, the coating in such internal engine parts or reaction vessel parts comprises a plurality of donor layers and acceptor layers.

Multicolor phosphors, for example white LEDs employing a combination of a blue LED with green and red phosphors. Blue (Ce ³⁺ ), green (Tb ³⁺ ) and red (Eu ³⁺ ) phosphors may be employed together, for example as separate acceptor layers of a multilayer embodiment of the invention as described herein. This reduces the need to mix different phosphor powders that have different thermal, mechanical and chemical properties. YV0 ₄ doped with Eu ³⁺ is an effective red phosphor, but YVO ₄ cannot be combined with Ce ³⁺ and Tb ³⁺ to form blue or green phosphors, because V ⁵⁺ can be reduced to V ^44" by Ce ³⁺ or Tb ³⁺ , leading to IMCT quenching. Thus in certain preferred embodiments of the present invention the first multivalent cation is V ⁵⁺ and the second multivalent cation is Ce ³⁺ or Tb ³⁺ . Preferably the donor is YVO ₄ and in such embodiments the charge transfer blocking layer is preferably YPO ₄ as this has the same crystal structure as YVO ₄ but P ⁵⁺ cannot participate in IMCT quenching. The charge transfer blocking layer thus enables energy transfer from V ⁵⁺ to Tb ³⁺ without IMCT quenching.

The present invention further opens up the possibility of choosing host materials that can withstand strong electric fields, intense light sources like lasers/laser diodes, chemical environments etc. This could be useful for optical sensing in harsh environments, such as monitoring chemical reactions/combustion and high intensity light sources.

Down conversion for solar cells. Combining Yb ³⁺ with either Tb ³⁺ or Pr ³⁺ has shown the possibility to split one UV photon into two NIR photons. However, it is very hard to sensitize Tb ³⁺ and Pr ³⁺ with something that absorbs strongly, due to charge transfer quenching as described above. The present invention circumvents this problem, opening up a new class of materials for down-conversion in solar cells. Thus in certain embodiments the acceptor cation may be Tb ³⁺ or Pr ³⁺ and the charge transfer blocking layer may be an active layer, preferably a layer comprising Yb ³⁺ , which (as described above) may form a charge transfer state with Tb ³⁺ or Pr ³⁺ while still blocking charge transfer to the donor cation (preferably Ti ⁴⁺ ), thus enabling down-conversion.

The invention will now be illustrated further by means of the following non-limiting Examples and Figures.

Figure 1 shows how the differences in distance dependence of FRET and IMCT results in a minimum separation, d _min , needed to avoid IMCT and a maximum separation which still allows FRET.

Figure 2 shows absorption spectra and luminescence from 260 nm excitation of materials described in the Example. This shows a clear increase in emission intensity with increasing barrier thickness and also shows the low efficiency in the absence of a barrier layer. Figure 3 shows a decay curve ofTb ³⁺ 545 nm emission showing a drastic increase when a 4 cycle AI2O3 barrier is introduced, as detailed in the Example. The quenching rate is much slower when the Ti ⁴⁺ and Tb ³⁺ are separated from each other.

Figure 4 shows transmission spectra of x6\ samples as detailed in the Example, showing both a stronger absorption due to more (TiO ₆ ) ^8" and a redshift due to more bulk-like absorption in thicker Ti0 ₂ layers.

Figure 5 shows PLE of "as-deposited" and annealed samples of 100- 61 as detailed in the Example, showing that the emission intensity increases with increasing T1O2 cycles, and a redshift going from 1 to 3 T1O2 cycles. Note that 100-061 does not contain any T1O2 cycles.

Figure 6: Monitoring the emission of the system described in the Example while changing the excitation wavelength (photoluminescence excitation measurement) shows that with increasing thickness of the barrier layer, the absorption of the excitation light is blue-shifted.

Example

An example of the invention is described for a system where T1O2 is the donor (and the absorbing species) and Tb ³⁺ is the acceptor (and emitting species). Without a barrier separating the Ti ⁴⁺ from the Tb , this system will be highly quenched due to the charge transfer reaction Ti + Tb → Ti + Tb ⁴⁺ . A1 ₂ 0 ₃ was employed as a barrier material (i.e. as a charge transfer blocking layer; in this instance the charge transfer blocking layer is a passive layer). The effect of a sub-nanometer separation between UV absorbing (TiO ₆ ) ^8" and green emitting Tb ³⁺ was studied to see how this would affect the (wanted) FRET transfer between the two cations (Ti ⁴⁺ donor and Tb ³⁺ acceptor) and the strongly quenching inter metal charge transfer (IMCT) state.

By introducing a few cycles of AI2O3 as a barrier material to separate Ti ⁺ and Tb ³⁺ , quenching due to IMCT was effectively prevented while still allowing FRET from (TiO ₆ ) ^8" to Tb ³⁺ .

The barrier also allowed the T1O2 layer thickness to be increased, and therefore allowed an increase in absorption of incident light, both by increasing the absolute amount of T1O2 in the film and by allowing a 2D-network of (TiOe) ^8" to form which gives rise to redshifted and more bulk-like absorption. This shows how sub-nanometer barriers can be used to control the transfer and quenching mechanisms in such d° transition metal materials with luminescent Tb ³⁺ , and that ALD is ideally suited for designing these types of nanostructures. This technique should easily be generalizable to other d° transition metal systems like YVO4 and the other lanthanides that are prone to IMCT in these host materials (Ce ³⁺ , Pr ³⁺ and Tb ³⁴ ).

Samples were grown by ALD and consisted of 500 supercycles of repeating 1 Ti0 ₂ / x A1 ₂ 0 ₃ / 1 TbA10 ₃ / x A1 ₂ 0 ₃ cycles, where x = 0, 2 and 4. In addition, a sample consisting of just alternating Ti0 ₂ / TbO _x cycles was prepared. It is clear that the emission intensity (shown in Figure 2) increases up to two orders of magnitude by separating the T1O2 from Tb ³⁺ with just 4 cycles of AI2O ₃ .

In addition, the lifetime of the emission was drastically increased when the barrier was increased in thickness, as shown in Figure 3. This showed that the barrier prevented the charge transfer quenching while still enabling efficient donor-to-acceptor energy transfer (via FRET). This example employed ALD as a deposition technique, but the invention is not exclusive to ALD and can be made by other routes as well.

Atomic layer deposition is a technique where the film is normally built up one sub-monolayer at a time and thus is well suited for such a barrier structure. Since most of the lanthanides grow very similarly, the ALD method described in this Example can easily be generalised to other lanthanides, host materials and with other barrier materials. The thin films were deposited in a F-120 Sat-type ALD-reactor (ASM Microchemistry Ltd) using a combination of conventional processes for deposition of TbO _x and Ti0 ₂ . Terbium oxide often forms mixed Tb ³⁺ /Tb ⁴⁺ oxides. Since this depends on the oxidizing conditions and the overall film chemistry, this cycle is denoted herein as TbCv TbO _x was deposited using Tb(thd)3 (where thd is 2,2,6,6-tetramethyl-3,5-heptanedionate, Strem > 99.9 %) and ozone, while Ti0 ₂ was deposited using TiCL _t (Aldrich > 99.9 %) and water. The ozone was produced from 99.8 % 0 ₂ (AGA) in a BMT 803N Ozone Generator. AI2O ₃ for the barrier layer was deposited with TMA (where TMA is

trimethylaluminium) and H ₂ 0 with deposition parameters identical to Ti0 ₂ . Nitrogen was used as carrier and purge gas, and came from 2 sources due to ongoing lab upgrades: Separated from air in two different nitrogen generators (Schmidlin UHPN3001 N ₂ purifier, > 99.999% N ₂ + Ar purity) and 5.4N N ₂ bottles from Praxair. Depositions were done to verify that changing nitrogen source did not affect the depositions. The pulse durations for the deposition of TbO _x layers were 1.5 / 1.5 / 4 / 1.5 s for Tb(thd) ₃ / purge / 0 ₃ / purge. For Ti0 ₂ , pulse durations were 0.8 / 1 / 1 / 1 s for TiCL _t / purge / H ₂ 0 / purge. These pulse and purge times have proven to be sufficient for proper surface saturation and purging for the reactor used. All films were deposited at 300 °C.

Films were grown on p-type Si(100) and silica. Two silicon strips of 0.5 x 4 cm ² were positioned perpendicular to the gas flow and about 8 cm apart by the inlet and the exhaust of the deposition chamber to monitor the gradients of the growth. All substrates were wiped and dust removed using pressurized air. The native oxide on the silicon was not removed, but its thickness on each substrate was measured by ellipsometry before the deposition. The substrates were further pre-cleaned by an in- situ 30 minute pulsing sequence of ozone and nitrogen at the deposition temperature to remove any organic residues prior to the deposition. Multilayer films were grown by changing the sequences of Ti0 ₂ , TbO _x and A1 ₂ C>3 cycles. The general film structure is n repeating supercycles of

(Ti0 ₂ ) _x (Al ₂ 0 ₃ ) TbO _x ) _z (Al ₂ 0 ₃ ) _j ,, so that each Tb ³⁺ layer is separated from Ti0 ₂ by y cycles of AL2O ₃ on both sides. Samples will be named n-xyz to easily identify the cycle sequences, i.e. 50-141. and z are always single digits, so a sample named 50-1101 will have x = z = \ and y = 10.

The crystallinity and phase content of the deposited and annealed films were evaluated by means of a Bruker D8 Discover diffractometer using CuST _a i radiation and a Ge(l 11) monochromator. The chemical stoichiometry was measured by X-ray fluorescence on a Phillips PW2400 Spectrometer and interpreted with the Uniquant analysis software. Annealing was conducted in a preheated tube oven in air for 1 hour at a 600, 800 and 1000 °C. Film thickness was determined with a spectroscopic ellipsometer (J. A. Woollam alpha-SE) in the 390 - 1000 nm range, and modelled with a Cauchy function. Luminescence (PL) and excitation (PLE) measurements was done using an Edinburgh Instruments FLS920 fluorescence spectrometer with a 450 W Xe lamp as excitation source and a Hamamatsu R928 PMT for detection. PL decay measurements were performed with an optical parametric oscillator (OPO) system (Opotek HE 355 Π) pumped by the third harmonic of a Nd:YAG laser as excitation source. The OPO system was set to 355 run and a repetition rate of 20 Hz. The decay was recorded with the same equipment used for the excitation measurement. Fig 2 shows that when adding a AI2O ₃ barrier layer, both a blueshift of the absorption and a drastic increase in emission intensity is observed. A broad emission is also observed for the 141 sample, probably coming from (TiOe) ^8" due to incomplete FRET to Tb ³⁺ .

Fig 3 shows that a barrier of 4 cycles AI2O ₃ between each T1O2 and TbO _x cycle drastically increases the lifetime of the Tb ³⁺ emission.

Fig 4 shows the transmission of 100 supercycle films of x6\ on silica substrates. For low*, the film is partially transparent for UV, while for x = 6 the transmission reaches close to zero. Figure 5 shows the PLE spectra for the same samples. It is seen that both annealing and increasing x increases the intensity and redshifts the position of the excitation peak. The emission of the as deposited samples actually decreases with x, but it is seen that for annealed samples the intensity increases drastically. Preventing IMCT quenching:

Fig 3 clearly shows that a barrier thickness of 4 AI2O ₃ cycles drastically increases the lifetime of the emitting Tb ³⁺ state. This means that major quenching mechanism, which is the low lying IMCT state, is reduced since Ti ⁺ and Tb ³⁺ are to a much lesser extent closest cationic neighbours. Fig 2 shows that the absorption from T1O2 blueshifts when adding a barrier. This indicates that the absorption is transformed from bulk-like to cluster-like since the barrier-Tb-barrier layers break up the 3D network of linking (TiO ₆ ) ⁸ \

Increasing transfer efficiency:

From Fig 5 it is seen that annealing at 800 °C drastically increased the emission intensity. In addition, increasing x in the 100-J 61 samples actually decreased the emission intensity for as-deposited samples and samples annealed at 600 °C. The reason for this is likely due to quenching of the (TiO ₆ ) ^8" excited state. It is well known that CT states are often only luminescent at low temperatures and in well crystalline samples. In addition, no form of T1O2 is known to have particularly high luminescence efficiency. Thus it makes sense that these amorphous films show high quenching rates for the (TiO ₆ ) ^8" excited states which competes with FRET to Tb ³⁺ , and that the thicker the T1O2 layer is, the lower the FRET rate is due to increase in average Ti ⁺ -Tb ³⁺ distance. When annealing the samples, even though the samples are still macroscopically amorphous, the (TiOe) ^8" probably start to form more well defined nucleation clusters. This would reduce the quenching of the (TiO ₆ ) ^8- state, which means that it is easier for FRET to dominate more over the quenching and the increase in absorption due to more T1O2 cycles will be more important than the slight reduction in FRET rate. This way it is possible to obtain more bulk-like absorption in the TiC layer and still transfer this energy through FRET to Tb ³⁺ .

Fig 6 shows that with increasing thickness of the barrier layer, the absorption of the excitation light was blue-shifted. Without being bound by theory, this is thought to be due to interaction between close lying Ti ⁴⁺ which gives a more bulk -like absorption when the layer is thin, while the Ti ⁴⁺ becomes more separated with increasing layer thickness so that the absorption becomes quantum- confined and thus blue-shifted.

Overall it was therefore concluded that, by introducing a few cycles of AI2O ₃ to separate Ti from Tb ³⁺ , quenching due to IMCT was effectively prevented while still allowing FRET from (TiO ₆ ) ^8- to Tb ³⁺ . This barrier also allowed an increase in the T1O2 layer thickness to increase absorption, both by increasing the absolute amount of T1O2 in the film and by allowing a 2D-network of (TiO ₆ ) ^8- to form which gives rise to redshifted and more bulk-like absorption. This shows how sub-nanometer barriers can be used to control the transfer and quenching mechanisms in such d° transition metal materials with luminescent Tb ³⁺ , and that ALD is ideally suited for designing these types of nanostructures. This technique should easily be generalizable to other d° transition metal systems like YVO4 and the other lanthanides that are prone to IMCT in these host materials (Ce ³⁺ , Pr ³⁺ and Tb ³⁺ ).