FIELD OF THE INVENTION

The present invention relates to a thin-film device, to a method for producing a thin-film device, to a protective layer for shielding an oxygen, moisture and/or carbon monoxide sensitive surface, to a method for shielding such a surface, to a method for forming a metal framework material, to a hydrogen sensor, to an apparatus for detecting hydrogen and to an elec- tro-magnetic transformer comprising said sensor.

BACKGROUND OF THE INVENTION

In a more generic perspective in an economy with hydrogen as a major energy carrier, the development of affordable, reliable, sensitive and selective hydrogen sensors is indispensa- ble. Several types of hydrogen sensors are currently available, which exploit the following detection mechanisms: catalytic, electrochemical, mechanical, optical, acoustic, thermal conductivity, resistance and work function. In principle, Pd-based optical fibre sensors could meet requirements if cross- contamination effect of a Pd surface by oxygen, moisture or carbon monoxide, for example, can be prevented. Such sensors can also be used for detecting hydrogen in other environments. However, since hydrogen detection often takes place in an explosive environment, cf. for leak detection or hydrogen- concentration measurements in gas streams, use of optical hydrogen sensors has a major advantage of being intrinsically safe due to the lack of electric leads in a sensing area. In addition known, fibre-optic, Pd-based thin-film hydrogen sensors represent a relatively cheap and reliable solution to this problem since they also allow for continuous sensing via remote hydrogen-gas detection, a key for personal and material safety. However, it is well known that Pd-based sensors may be poisoned or their sensing properties may be seriously influenced by oth ^¬ er gases (for example O2, CO) due to palladium's inherent cross-sensitivity towards contaminants. Contamination or degra ^¬ dation of a Pd sensing layer inhibits e.g. dissociation of hydrogen on a catalyst surface, severely limiting sensitivity of a final device. Thus a key issue is to prevent contaminants from reacting with the Pd surface. In other words, to provide a controllable and reliable sensor, applicable under a broad range of conditions.

The present thin-film device comprises a substrate, an active sensing layer whose optical properties change depending on hydrogen content and a protective layer on the active sensing layer.

Such thin-film devices are known from the prior art. As an example _/ WO2007/126313 discloses a switchable mirror device comprising an active layer, wherein said active layer changes its optical properties upon addition or removal of hydrogen and comprises a hydrogen and oxygen permeable and water impermeable layer, wherein said layer is liquid water impermeable and water vapour permeable and has hydrophobic surface properties.

In Bordiga et al . (Phys. Chem. Chem. Physics, Royal Soc. Chem., Cambridge, GB, vol. 9, nr. 21, May 14 2007, pp. 2676- 2685) recites amongst others HKUST-1. However, the HKUST-1 relates to a bulk material and is further not suited for the pur ^¬ pose indicated, as it lacks essential properties thereto, e.g. in terms of a multitude of small crystallites being formed. Bordiga relates to detection of (gaseous) species, such as hy- drogen, with IR-spectroscopy . It is noted that such detection has virtually nothing to with the present thin film device and present sensor, where changes in optical properties of an alloy are measured.

Thin-film devices of the prior art suffer a number of dis- advantages in particular with regards to achieving controlled and reliable absorption of a species such as hydrogen, such as over a broad range of hydrogen pressures, and/or in the presence of "poisoning" chemical species i.e. species that adverse ^¬ ly affect the performance of the thin-film device and/or the properties of the optical sensing layer over-time.

Several materials have been investigated for the development of a protective coating of a (Pd) film: amorphous carbon, titanium dioxide, and various polymers, such as polyvinyl ace ^¬ tate, polyvinyl chloride, cellulose acetate, and polystyrene were tested against the negative effects of oxygen, while hex- anethiolate was tested against ozone. In addition, the effect of a metal-free phtalocyanine coating against dry air (0 ₂ ) and that of a polyethylene- phtalocyanine double layer against moist air (H ₂ 0, 0 ₂ ) was studied. Such a double coating may in- crease a sensor's selectivity not only toward the moist air but towards CO and N0 ₂ also, however, only at prohibitively high hydrogen concentration. So far, there has been no viable way developed for a protective coating of Pd-based hydrogen sensors that would not seriously compromise its favourable properties such as low detection limits and response time, relatively easy and affordable fabrication, low weight and portability, etc.

The present invention therefore relates to a thin-film device and further aspects thereof, which overcomes one or more of the above disadvantages, without compromising functionality and advantages.

SUMMARY OF THE INVENTION

It is an object of the invention to overcome one or more limitations of the thin-film devices of the prior art and at the very least to provide an alternative thereto.

In a first aspect, the invention relates to a thin-film device according to claim 1, comprising the present MOF-layer.

In an example the present thin thin-film device comprises a substrate, an active sensing layer whose optical properties change continuously as a function of hydrogen content, a Pd caplayer to dissociate hydrogen which acts simultaneously as a protective coating for the sensing layer, and a protective layer coating the Pd which protects this caplayer.

It is noted that on the one hand the present invention provides controlled and reliable absorption. As the mechanism of absorption is in principle reversible, also controlled and reliable desorption is provided. Therewith a means for monitoring of e.g. a (varying) hydrogen concentration over a large range of pressures is provided. It is noted that the present optical system is much safer to use and to handle compared to e.g. electrical (conducting) sensors, especially in environments where a large electro-magnetic field may be present.

A person of skill in the art is able to identify many suitable substrate materials upon which a thin-film device such ^' as the thin-film device of the invention can be constructed. Examples of suitable substrate materials include glass, guartz, indium-tin oxide, etc. The substrate material is preferably optically transparent (more than 95%), at least over a proportion of the visible, UV and/or IR regions of the electromagnetic spectrum (200 ran- 3000 nm) . Such provides for use of white light, IR-light, UV-light, a laser with a specific wavelength, and combinations thereof.

As mentioned above, optical sensing layers with variable optical properties depending on e.g. a hydrogen content of the layer are known in the prior art, comprising an alloy. In an example the sensing layer is the present alloy. The alloy com ^¬ prises at least a first metal and a second metal, such as Mg and Ti. In principle the first and the second metal may be se ^¬ lected from the periodic system of elements. Typically characteristics of a metal are that they may oxidize, may form a cat- ion, may be an electrical conductor, etc. In the present alloy the first and second metal have a hydrogen eguilibrium pressure (Pa) . Such an equilibrium pressure per se can be measured with means known to a person skilled in the art. Typically it is measured at ambient temperature (273 °C) .

A catalyst such as in a layer is provided on top of the optical sensing layer, such as coating the optical sensing layer. Examples of such layers include for example Pd-layers. The Pd-layers may comprise pure Pd or mixtures comprising Pd. For example, Ag can be added in a quantity of for example 20-30 mole%. The catalyst may also relate to a complex layer, suited for the present purpose. Such layers serve to facilitate hydro ^¬ gen absorption by the optical sensing layer.

It is noted that a term as "on top" may relate to a sequence of e.g. layers, a first layer coating a second layer, a layer provided on an intermediate layer, the intermediate pro ^¬ vided on e.g. the sensing layer, etc. The layer may also partly be on top. In view of the present application such terminology is mainly functional of nature.

On top of the optical sensing layer, or where present, on top of the catalyst (layer) , a protective layer is provided, the protective layer not limiting functionality of the optical sensing layer, e.g. being permeable to relevant species, optical transparent, etc. and protecting the optical layer. Both the catalyst layer (where present) and the protecting layer are permeable to a species to be measured, such as hydrogen, and are optically transparent, at least over a range of the visi ^¬ ble, UV and/or IR regions of the electromagnetic spectrum. An example of a protecting layer from the aforementioned

WO2007/126313 is to provide a layer of Teflon. The protective layer is provided to improve the longevity of the thin-film de- vice through preventing deterioration of the catalyst and/or optical sensing layers and improves the handleability of the device through preventing a user from coming into contact with the optical sensing and/or catalyst layers. It is noted that the nature of Teflon and more specific sputtered PTFE makes it in principle difficult to process.

Control and reliability of e.g. hydrogen absorption is achieved with the thin-film device of the invention by providing a protective layer comprising a metal-organic framework ma- terial, or an optical sensing layer according to the invention, e.g. comprising an alloy, or preferably both.

Metal-organic framework materials (MOFs) relate to materials in which metal-to-organic ligand interactions yield porous coordination networks. Possible applications of MOFs include storage and separations of gases, sensors and catalysis. The present MOF' s may be applied in optical fibre sensors, switcha- ble mirrors, hydrogen sensors, protective coatings, etc. In particular, inorganic and metal-organic membranes provide high thermal, mechanical, and chemical stability. It has been found that MOFs typically display regular crystalline lattices with well-defined pore structures, high porosity, and a combination of organic and inorganic building blocks offers a virtually in ^¬ finite number of combinations. Furthermore, it has been found that chemical functionalization of organic linkers in the MOF structures affords facile control over the pore size, pore shape and over the chemical as well as physical properties, being suited for protective membranes. Such membranes provide a fast flux, sufficient robustness for reproducibility and high selectivity. Highly tuneable MOF structures in combination with an anchoring monolayer (in a present example an amino acid such as glycine) between a MOF and catalyst (Pd) layer improve sen ^¬ sor's properties. The present metal-organic framework determines diffusion of gaseous species and serves as a membrane layer intended for selected (hydrogen) species to reach the catalyst sensing layer; an optional glycine anchoring layer de ^¬ termines and improves kinetics.

It is noted that successful fabrication of metal-organic framework thin films of adequate quality for gas separation has been a challenging task, considered mainly due to unfavourable heterogeneous nucleation and relative weakness of coordination bonds between substrate and films applied. As a result the prior art MOF layers are of insufficient quality. Further, the use of a continuous MOF thin film as a gas selective membrane, to protect the sensor from contamination, has not been reported before. Using several techniques, such as epitaxial growth on a self-assembled monolayer (SAM) by solvothermal synthesis or layer-by-layer deposition, microwave and electrochemical syntheses, etc. sufficient quality continuous thin film MOF-layers have been provided by inventors. The present MOF-films have a thickness of 5 nm- 10 pm, preferably form 10 nm -1 μιη, more preferably form 50 nm -500 nm, such as from 100 nm - 250 nm. These present MOF-layers are used as protective coating for e.g. Pd-based optical sensors: the synthesis of a metal-organic framework thin film with high hydrogen permeability and recy- clability is provided.

Typically MOFs are prepared under solvothermal conditions in pure N, -diethylformamide or N, N-dimethylformamide, which slowly decompose upon heating and generate bases capable of deprotonating organic linker molecules. The latter may react with metal salts and produce 3D metal-organic networks.

As an alternative route, some MOFs may be prepared via a sequential route wherein primary building blocks of the MOFs, i.e. organic ligands and metal salts, are mixed and treated under solvothermal conditions. By using an appropriately func- tionalised organic surface (as a 2D nucleation site), MOF structures are grown in a layer-by-layer fashion.

Surprisingly, it has been found that the present MOFs are also suitable for use as protective layers to shield a surface such as the optical sensing layer of the invention from an en- vironment, whilst allowing molecules such as hydrogen to diffuse through the layer. The present invention of using a con ^¬ tinuous MOF thin film for e.g. sensor protection is considered novel. The present thin film forms a closed structure. Further the present layer is typically free of pinholes (some pinholes may occur, due to thermodynamics) .

As mentioned above, MOFs are porous-network materials.

There porosities and pore properties, such as approximate dimensions, can be tuned through varying the metals and linkers making up the MOF. They can be synthesized inexpensively, rela- tively easily, in high purity and in a highly crystalline form. Desirable performance of the thin-film device of the in ^¬ vention in terms of control and reliability of hydrogen absorp ^¬ tion can be achieved through either improvement separately or through the combination of improvements. Reliability relates particularly to reliability over time, such as tens of years, and with repeated use.

The invention also relates to a protective layer for shielding an oxygen, moisture, hydrogen sulphide and/or carbon monoxide sensitive surface, to a method for shielding such a surface, to a method for forming a metal framework material, to a hydrogen sensor and to an electro-magnetic transformer comprising said hydrogen sensor.

The present invention provides a solution to one or more of the above mentioned problems and overcomes drawbacks of the prior art.

Advantages of the present description are detailed

throughout the description.

DETAILED DESCRIPTION OF THE INVENTION

In an exemplary embodiment, the metal-organic framework material is selected from a group consisting: HKUST-1,

Cu/Zn (py/bipy/bpe) 0.5 [ (X-) FMA] , or combinations thereof.

In an exemplary embodiment, the device comprises a protective layer provided on the optical sensing layer through an adhesive layer wherein the adhesive layer comprises a self- assembled monolayer of an amino acid or of a carboxylic acid, the acid preferably having 2-30 carbon atoms, such as 5-20 carbon atoms, such as glycine, or of a silane.

In an exemplary embodiment, the optical sensing layer has a thickness in the range of 1.5-500 run, such as 10-100 nm.

In an exemplary embodiment, the protective layer has a thickness in the range of 0.02-200 μπι, such as 1 um.

Also a catalyst layer to enhance absorption is present, typically on top of the optical layer, either directly or with one or more intermediate layers .

In a second aspect, the invention relates to a method of for producing a thin-film device comprising providing a sub ^¬ strate, depositing an optical sensing layer on the substrate, the optical sensing layer comprising the present alloy, depositing a catalyst layer on the optical sensing layer, and providing a protective layer on the catalyst layer. Despite availability of MOF thin layers per se these have not been used yet to protect e.g. sensors before. It has been found that a catalyst layer, such as a Pd based layer, on top of sensor material is preferably protected against environmen- tal influences, such as poisoning. The present invention pro ^¬ vides a thin film of MOF that can be used and can be tuned as an optional protective coating, allowing absorption and desorp- tion of e.g. hydrogen in a controlled and reliable manner, having excellent stability and well-characterised physical proper- ties.

An interesting aspect is the use of a self-assembled monolayer of an intermittent adhesive type layer, such as glycine, as a template to grow the MOF (such as HKUST-1, Cu ₃ (BTC) ₂ - BTC=1 , 3 , 5-benzenetricarboxylate ) . A functional advantage there- of is that surface adsorption at a catalyst layer (such as Pd) is optionally separated from a selective membrane activity of the MOF. As such one can now address an optional interface problem separate from an optional membrane problem. Surprisingly a self-assembled monolayer of an intermittent layer such as glycine can be regarded as a template to grow a MOF (such as

HKUST-1) thin film on top of e.g. a Pd thin film. The provided stability in various liquids as well as its protective nature are important characteristics.

In a third aspect, the invention relates to a protective layer for shielding a sensitive surface comprising a metal- organic framework material "transparent to ¾" . Applications of the present layer relate e.g. to sensor, such as optical fibre sensors, such as a sensor for hydrogen, mirrors, such as switchable mirrors, glass, quartz, protecting of an underlying layer from e.g. a transformer oil, and as a protective coating in general. In an example a surface of a Pd-based catalyst layer on top of the present sensor material preferably is protected against poisoning. A prior art example thereof is use of PTFE against the detrimental adsorption of water. It appeared that PTFE allows also the measurement of ¾ in oil. It is noted that in general an organic based protection layer material as PTFE does not adhere particularly well to a layer to be protected; in other words growth of an organic based layer on a sensor layer, such as a Pd-layer, is not at all straightfor- ward. The present invention shows the proof of principle that a thin film of MOF can be used as an optional protective coating allowing absorption and desorption of hydrogen. As an advantage provided by use of a Self-assembled monolayer of glycine as a template to grow the MOF (in this case HKUST-1, Cu ₃ (BTC) ₂ - BTC=1 , 3 , 5-benzenetricarboxylate ) is that surface adsorption at the catalyst (Pd) can be (functionally and design wise) separated from a selective membrane activity of the MOF. A further advantage is that the present protective layer substantially protects a sensor against a large range of contaminations.

It has been found that HKUST-1 as a MOF material is easy to use in a synthesis as a thin film, using various techniques, for its stability and well-characterised physical properties, and for its remarkable selectivity towards hydrogen gas adsorption when processed as a thin film. The selectivity of the MOF structure can be highly improved with respect to the HKUST- lstructure by tailoring the pore shape and size by chemical functionalization and interpenetration . Also the self- assembling monolayer, such as glycine, can be functionalized, similar to the MOF material. It is noted that the term "adsorp- tion" does not necessarily mean that a given material can no longer be permeable to e.g. gasses.

In a fourth aspect the invention relates to a method for shielding a surface from one or more of oxygen, moisture, liquids in general (oil, water, alcohol) , carbon monoxide compris- ing providing a protective layer of a metal organic framework material on said surface.

In an fifth aspect, the invention relates to a method for forming a metal-organic framework material on a surface comprising a self-assembled monolayer of an amino acid or of a carboxylic acid, the acid preferably having 2-30 carbon atoms, such as 5-20 carbon atoms, such as glycine, and of a silane, as a SAM. The present method provides MOF growth on e.g. SAM- functionalised Pd. It is noted that use of glycine as SAM is considered novel. Inventors established that the present film relates to homogeneity and amorphousness of the HKUST-1 thin- film, which shows a striking contrast with previous results that always related to epitaxial growth (or at least films with high crystallinit ) . The present layer is of an amorphous material and yet its permeability to hydrogen is as good as that of a comparable crystalline material. In a sixth aspect, the invention relates to a sensor comprising the thin-film device of the invention. In a preferred example the sensor is a hydrogen sensor. The sensor may be provided with an optical transmitter, such as an optical fiber. Such provides e.g. as advantage that a measurement can take place at a spatial distance of detection. Even further the invention may relate to a combination of optical sensing layers, such as a stack of layers. Each layer or stack of layers may be optimised to sense a species, such as hydrogen, oxygen, nitro- gen, carbon monoxide, carbon dioxide, etc. Also, a layer or stack of layers may be optimised to determine a species in a first concentration range, and a further layer or stack of layers for determining a species in a second concentration range. Likewise and preferred a combination of various 2-d and 3-D do- mains may be used. Thereby an enlarged range of concentrations can be determined. Even further the sensor may comprise one or more of the above, e.g. layers for various species and layers for various concentrations of one or more species. Even further, other materials may be used in combination with the pre- sent optical layer to extend e.g. a pressure range and to in ^¬ corporate further species being measurable. An advantage is that the present invention allows for a combination of various optical layers without much extra measures to be taken in order to obtain a functional device.

In a seventh aspect, the invention relates to an electromagnetic transformer comprising the hydrogen sensor of the invention.

In an eighth aspect the invention relates to an apparatus for detecting hydrogen comprising a sensor, the sensor being locat- ed at a longitudinal side of an optical transmitter, the opti ^¬ cal transmitter comprising a central transmitting element, such as a guartz core, a transducer layer, preferably having a sur ^¬ face plasmon resonance frequency, an alloy according to the invention, and optionally a protection layer, preferably accord- ing to the invention, and a frequency shift detector.

With the optical re-resonator in combination with the frequency shift detector a resolution in the order of pm is obtained .

The above apparatus relates to a new design of a fiber op- tic Surface Plasmon Resonance (SPR) sensor using e.g. Palladium as a sensitive layer for hydrogen detection, or likewise an alloy according to the present invention. In an example, a transducer layer is deposited on the outside of a multimode fiber, after removing the optical cladding thereof. In an example the transducer layer is a multilayer stack made of a Silver, a Silica and a Palladium layer. Spectral modulation of light transmitted by the fiber allows to detect presence of hydrogen in the environment. The sensor is only sensitive to a Transverse Magnetic polarized light and Traverse Electric polarized light can be used therefore as a reference signal. A more reliable response is expected for the fiber SPR hydrogen sensor based on spectral modulation instead of on intensity modulation. The multilayer thickness defines the sensor performance. The silica thickness tunes the resonant wavelength, whereas the Silver and Palladium thickness determine the sensor sensitivity. In an optimal configuration (NA = 0.22, 100 yna core radius and transducer length = 1 cm) , a resonant wavelength is shifted over 17.6 nm at a concentration of 4% Hydrogen in Argon for the case of the 35 nm Silver/100 nm Silica/3 nm palladium multilayer. Amongst others the above results are published in two articles of one of the present inventors (Opt. Soc. America, 7 November 2011, Vol. 19, No. S6, ppA1175-1183 and Proc. SPIE, Vol. 8368, pp. 836804-1-12) .

The invention will hereafter be further elucidated through the following examples which are exemplary and explanatory of nature and are not intended to be considered limiting of the invention. To the person skilled in the art it may be clear that many variants, being obvious or not, may be conceivable falling within the scope of protection, defined by the present claims.

EXAMPLES

Experimental

Substrate preparation

Pd thin films with a thickness of 60 nm were deposited at room temperature on quartz substrates in an ultrahigh-vacuum (UHV) DC/RF magnetron sputtering system (base pressure 10 ^~8 mbar, Ar deposition pressure 0.003 mbar). By using an additional 3 nm Ti interlayer, adhesion of the sensing layer (Pd) to the quartz substrate is improved. The Pd thin film thickness was deter- mined by a profilometer (DekTak3) . HKUST-1 powder sample

All chemicals were purchased from Sigma Aldrich and were used without further purification. The synthesis of the powder sample was carried out in the same fashion as published by Chui and co-workers . The material was characterised without activation .

Growth of HKUST-1 on Pd thin films

Metal-organic framework growth on Pd surface was carried out using a layer-by-layer (LbL) deposition method on a self- assembled monolayer (SAM) . Inventors used glycine as the SAM layer. SAM deposition was carried out by soaking the Pd thin film overnight (ca. 16 h) , at ambient temperature in saturated solutions of glycine in 1:1 ethanol: de-ionised water mixture. The substrate was subsequently washed in the same solvent mix- ture thoroughly. The MOF thin film was deposited on the SAM using a layer-by-layer technique. HKUST-1 was grown in cycles, where one cycle consisted of subsequent immersions in 1 mM eth- anolic solutions of cupric acetate and benzenetricarboxylic acid. This procedure was repeated with rinsing and drying in ni- trogen flow after each immersion.

The number of MOF layers deposited ranged from 1-50. After deposition, the samples were immersed in ethanol for 24 hours and were kept under vacuum (10-4 mbar) until hydrogenation experi ^¬ ments. Other characterisation techniques (FTIR and XRD) were used after their exposure to air so results would be comparable with those on a non-activated powder sample.

Hydrogen sorption

Activation of HKUST-1 thin films was carried out in-situ in a hydrogenography setup by evacuation of ethanol-containing sam- pies at 100 °C, under vacuum (10 ^-4 mbar) for ca . 3 hours. Sub ^¬ sequently, hydrogenography was used to characterise an optical response induced in the MOF-Pd multilayer upon hydrogenation. This technique allows for an efficient and accurate study of the hydrogenation behaviour (thermodynamics and kinetics) of a MOF-Pd thin film. The logarithm of the optical transmission, ln(T/T ₀ ), is found to be a function of the hydrogen concentration in the Pd thin film and yields so-called pressure- transmission isotherms.

Using identical Pd thin films as substrates with different MOF coatings (1, 2, 4, 8, 10 and 50 cycles of HKUST-1), inventors were able to analyse the influence of a coating thickness simultaneously and under the same conditions. This approach pro ^¬ vided an easy selection of a most optimal MOF coating.

Results and Discussions

Self-assembled monolayers are functional and tuneable organic thin films that can be supported on a wide variety of substrates. Furthermore, by cho.osing the right synthetic conditions, highly ordered, 2D-like crystalline interfaces can be obtained. Their relative ease of preparation, spontaneous as- sembly, diverse functionality and ability to be patterned by microcontact-printing methods make SAMs a desirable interface for MOF growth.

The present choice to use bulky glycine as the SAM layer was an unusual one, since typically long and flexible molecules are applied, such as functionalised undecanoic acids. However, it was intended to keep the Pd surface in close proximity to the MOF layer. As a consequence sideward hydrogen diffusion was limited. In addition, glycine is a cheap, non-toxic and easily available material, which offers the possibility to binding with both, the amino- and carboxylate ends to the Pd surface. FTIR spectra recorded at various parts of the samples suggest that MOF coverage was achieved at all parts of the substrates. SEM images show the formation of rather smooth and homogeneous thin films of HKUST-1 on the Pd substrates. Prior art where HKUST-1 coatings were rather heterogeneous and did not uniform ^¬ ly cover the substrate. Such films consist of individual parti ^¬ cles of variable orientation and size, which are not specifi ^¬ cally bound to the substrate. However, for applications, in sensor technology _, in which, the diffusion into a thin film or membrane layers plays a decisive role, granular and heterogeneous coatings exhibit a number of major drawbacks: due to poly- crystallinity and/or heterogeneity of the thin films, a homogeneous diffusion of the molecules into a MOF is impossible or highly limited. Furthermore, inhomogeneous films of strongly varying thickness, containing a number of pin-holes are not very attractive, as these defects severely limit gas-separation efficiency for instance, and, in our case, could actually account for Pd-surface contaminations.

A low XRD-diffraction intensity suggests that the thin film is largely in an amorphous phase. The amorphous nature of the MOF film is also reflected in the fact that the crystallinity of the HKUST-1 thin film did not increase significantly upon in ^¬ creasing film thickness. On the contrary, the very broad [222] diffraction peak was not much better resolved for the thin film with 50 cycles of HKUST-1 deposited on it than for that with only 10 cycles. It is ^' known that the choice of a substrate as well as SAM influences greatly the resultant film morphology and orientation. It is surprising that use of a novel combina ^¬ tion of an e.g. Pd substrate and glycine ^' SAM layer (short and capable of bonding at different termini) leads to very different results, compared to prior art. The present HKUST-1 structure, is packed having no pores in it, that is all pores are parallel with a (Pd) substrate. Undoubtedly, it is a minimum requirement for a protective coating of a (Pd-based) hydrogen sensor to be permeable to hydrogen gas, or permeable to a species of interest in general.

To test the diffusion of the hydrogen through the MOF layer, inventors determined an effect of hydrogen-pressure steps on optical transmission of the MOF-Pd thin film by hydrogenogra- phy. The results were more than satisfactory.

The hydrogen diffusion through the MOF layer is not considered a limiting factor. Upon decreasing the temperature (to ambient temperature) one may observe that the hydrogen-desorption process becomes slower. Furthermore, the Pd and SAM-Pd thin film have the same desorption times whereas the MOF coated SAM-Pd thin films are a bit slower. Thus, at lower temperatures diffu ^¬ sion through a MOF structure plays a more important role. Fur ^¬ thermore, a small difference in final transmission values of distinct samples was observed, which is deemed to relate to variation of a MOF-coating thickness. The transmission change between 100 °C and ambient temperature for any given hydrogen pressure is explained by pressure-composition isotherms of a Pd thin film: a plateau-pressure during absorption and desorption at RT is typically 50 mbar and 5 mbar, respectively. However, at 100 °C the plateau-pressure shifts to 700 mbar for hydrogen absorption and 150 mbar for desorption.

In addition, upon cycling the MOF-SAM-Pd thin film proves to be mechanically stable and does not show any cracking, buckling or delamination . As an example, cycles 7-17 shows no signs of lay- er deterioration under the following cycling conditions: hydro- gen pressure was of 300 mbar at 100 °C for absorption, while desorption was performed in vacuum. Although a decrease of optical response of a loaded state upon cycling can be observed, it is noted this phenomenon also holds for a pure Pd thin film and thus, it is regarded due to intrinsic properties of the Pd sensing layer rather than to that of the MOF coating.

The presented hydrogenation behaviour of the MOF-Pd thin films shows that these types of protective MOF multilayers fulfil first requirements for a fibre-optic hydrogen sensor, such as a good stability, good kinetic performance, and being transparent for H ₂ . In addition to the diffusion through the MOF, the binding of the SAM layer, in this case glycine, to the Pd surface is found to determine to a large extent the hydrogenation kinetics: it is well known that C _x H _y -compounds can have a delay- ing effect on the hydrogenation behaviour due to the formation of palladium carbides and inhibit H2-dissociation and recombination reactions on the surface.

The present findings show that the growth of the SAM layer on the Pd surface does not influence the hydrogenation kinetics, whereas the growth of the MOF structure has a minor effect on the hydrogen desorption only at lower temperatures.

Conclusions

A novel concept of developing protective metal-organic framework membranes for coating Pd-based fibre-optic hydrogen sen- sors is provided. To test the feasibility of this idea, inventors have designed and manufactured continuous and homogeneous thin films of HKUST-1 on SAM-functionalised Pd substrates using a layer-by-layer technique. Contrarily to previously reported HKUST-1 thin films, it is found that the thin film obtained was amorphous and displayed a much more even surface coverage. Most importantly, it is demonstrated that the thin films synthesised did not change the hydrogen-sensing properties like optical re ^¬ sponse and response time significantly, a prerequisite for application. It is observed that the present glycine layer influ- ences the Pd surface properties, while the MOF controls the diffusion of the gaseous species. Ultimately, both of these properties determine an effectiveness of a coating. In addition the prepared multilayer structures appeared to be quite robust upon repeated hydrogen cycling.