A SUBSTRATE HAVING A FILM GROWN THEREON & A METHOD OF FORMING A FILM ON A SUBSTRATE
This invention relates to a substrate having a film grown thereon, and a method of forming a film on a substrate. A particularly preferred embodiment relates to a method for forming a film on a substrate, the film including highly orientated arrays of metal oxide nanorods, in particular tin (IV) oxide. Nanorods, in the context of the present application, are rodlike structures which have a thickness (i.e. side to side, as opposed to end to end) of something in the order ofless than several hundred nanometres. Preferred embodiments of the invention pertain to substrates with films comprised of nanorods grown on them, and to the use of such substrates in diverse applications such as the photo electrochemical conversion of water to hydrogen and oxygen, and gas sensing amongst others. Tin (IV) oxide (Sn02 - also called stannic oxide) is an insulator and an important colourless, relatively low-cost, large bandgap n-type semiconductor when doped, for example, by oxygen vacancies or by antimony or fluorine ions. It has been widely utilised as a transparent conducting oxide substrate, as a gas sensor, as well as an electrode material for energy conversion and storage applications. Many different manufacturing processes have previously been proposed for the manufacture of thin tin oxide epitaxial films, i.e. films having a thickness of less than a few microns. One previously proposed method, described in articles by Kolmakov et al (in Adv. Mater. 2003, 15(12), 997-1000) and Zheng et al (in Chem. Mater. 2001, 13(11), 3859-3861), provided for the production of a film comprised of tin oxide orientated nanorod arrays on a substrate. This method, whilst relatively promising, necessarily involved a number of relatively complicated processing steps utilising relatively specialised equipment, and hence would be relatively expensive to implement on a production scale. It is also the case that the nanorods produced by the method, whilst much improved, still had a rather ill defined morphology. As is well appreciated in the art, to fully exploit the potential of such films it is preferable for the morphology of the film, and in particular the shape of the nanorods, to be well defined and in one aspect the present invention seeks to improve upon the morphology of films produced thus far.
The thermodynamically stable crystal structure of SnO2 is the same as that of rutile TiO2 (tetragonal crystal system), and occurs in nature as the native mineral cassiterite, the principal ore of tin. Cassiterite is also isostructural to argutite (GeO2), paratellurite (TeO2), plattnerite (PbO2), stishovite (SiO2) and pyrolusite (MnO2) as well as VO2, CrO2, RuO2, NbO2, TaO2, OsO2 and IrO2. Each unit cell of the rutile structure is, as is well known in the art, tetragonal, and SnO2 crystallises in point group symmetry 4/mmm and space group P42/mnm ) with tin and oxygen atoms in 2a and in 2f positions, respectively. The unit cell consists of two tin atoms and four oxygen atoms. Each tin atom is situated amidst six oxygen atoms which approximately form the comers of a regular octahedron. Oxygen atoms are surrounded by three tin atoms which approximate the comers of an equilateral triangle. Two of each of the twelve octahedral edges are shared with other octahedra, and in the rutile structure edge sharing is symmetrical and opposed. The octahedra share edges to form linear chains along the c-axis, and these octahedral chains run parallel to the c-axis and combine by sharing two opposed edges per octahedron. Each such chain is surrounded and cross-linked to four identical octahedral chains twisted at 90° to the first chain, and adjacent chains are all staggered c/2 by the 42 screw axes that are oriented along tunnels parallel to the c-axis. Crystalline tin (IV) oxide with this structure exhibits a high electron density parallel to the c-axis as compared to parallel to the a-axis (which is perpendicular to the c-axis), as well as a high polarisability and birefringence. Typical lattice parameters are a = 4.737 A, and c = 3.186A yielding an axial ratio of a:c which is equal to 1 :0.672. The low index (110) face of the crystal is the thermodynamically most stable bulk termination and possesses the lowest surface energy, that is the surface excess free energy per unit area. The sequence of surface energy per crystal face is (110) < (100) < (101) « (001). The (110) stoichiometric surface in which one half of the surface cations are five-fold coordinated with oxygen ions and the other half are sixfold coordinated because of the presence of a row of bridging oxygen ions, yields tin atoms with the formal +IV oxidation state. Thus, surface and bulk have similar resistivities. The (110) surface has no net dipole moment in the [110] direction and is therefore a non-polar surface.
However, this centrosymmetric structure of low axial ratios reduces substantially the probability of anisotropic growth of the crystal along the [001] direction and therefore, the generation of c-elongated prismatic crystals. For these reasons, SnO2 nanorods and nanowires seldom grow along the c-axis but instead tend to grow along the [101], [301], [200] or [11-2] directions. Even though the production of c-axis elongated nanorods has not hitherto been practicable, it has been appreciated in the art that a c-axis elongated nanorod with (110) side faces grown from a substrate would exhibit a highly advantageous and optimised structure. For example, such nanorods would present principal (i.e. largest) (110) faces, and as it is the most thermodynamically stable termination of the crystal, the likelihood of surface reconstruction of these faces is at its lowest. Also of note, particularly in the context of applications such as gas sensing, is that the (110) surface is the most efficient for disassociating incident gas molecules due to the fact that the distance between neighbouring tin atoms is at its shortest at this point in the crystal. Furthermore, given that a tin oxide crystal exhibits its greatest electron density along the c-axis, a nanorod grown along the c-axis and arranged so that the c-axis is upstanding from the substrate would be most conductive along the length of the rod - thereby providing the highest conductive route for electron transport through the crystal from an end face thereof to the substrate. To this end, a principal aim of the invention is to provide a substrate having a crystalline tin (IV) oxide film grown thereon, the film comprising a plurality of c-axis elongated nanorods. A further aim of the present invention is to provide a practicable solution to the problem of how to grow a film comprised of a plurality of c-axis elongated nanorods, preferably tin oxide nanorods, from the surface of a substrate. In pursuit of this first mentioned aim, a preferred embodiment of the invention provides a substrate substantially as defined in Claim 1. Other aspects of the present invention provide a photoelectrode, a photoelectrochemical system for the cleavage of water into hydrogen and oxygen which includes such an electrode, and a gas sensor. Another preferred embodiment of the present invention provides a method of forming a crystalline tin (TV) oxide film on a substrate (for example to form a substrate as claimed in Claim 1), the film comprising a plurality of nanorods each of which includes crystals grown along the c-axis of the tin (IV) oxide crystal, the method
comprising: at least partly immersing a substrate in an aqueous solution of a tin salt in an acid medium with a second substance, the second substance being chosen to provide a slow-release source of a basic medium; said slow-release source decomposing and reacting with said tin salt to thereby effect growth of tin oxide crystals along said crystal c-axis to form said crystalline tin (IV) oxide nanorods on said substrate. This method is advantageous in that it provides a simple process which can produce, relatively inexpensively, three-dimensional arrays of highly orientated crystalline tin oxide nanorods on a substrate. Preferred features of embodiments of the invention are set out in the accompanying claims and elsewhere in this description, and further advantages of the present invention will no doubt become apparent following a reading of that description. Particularly preferred embodiments of the invention will now be described by way of illustrative example with particular reference to the accompanying drawings, in which: Fig. 1 is a representative illustration of a method in accordance with a preferred embodiment of the invention; Figs. 2a and 2b are SEM micrographs, at differing magnifications, of a film of tin (IV) oxide nanorods grown on a substrate in accordance with the teachings of the present invention; Fig. 3 is an x-ray diffraction plot of a tin (IN) oxide film grown on a substrate in accordance with the teachings of the present invention; Fig. 4 is a representation of HRTEM images of a film produced in accordance with the teachings of the present invention; Fig. 5 is a schematic representation of one illustrative application; and Fig. 6 is a schematic representation of a second illustrative application. Referring firstly to Fig. la, the method of the preferred embodiment - in general terms - involves the preparation of an appropriate solution 1 in a suitable container 2. As depicted in Fig. lb, a suitable substrate 3 is then at least partly immersed in the solution 1 by, in the preferred arrangement, standing the substrate against a wall 4 of the container so that the substrate is at least roughly perpendicular to a base 5 of the container.
This orientation of the substrate in the container is preferred because it helps to mitigate inhomogeneity of the film that could be caused by powder precipitating from the solution and falling onto the substrate film were the substrate to be placed, for example, along the bottom of the container. It should be noted however that the scope of the present invention is not limited solely to a process where the substrate is orientated substantially perpendicularly to the base of the container. As depicted in Fig. lc, once the substrate 3 is placed within the container 2, the container is closed by means of a screw cap 6 for example, and placed in an oven 7 at a predetermined temperature for a predetermined length of time. Once that predetermined time period has elapsed, the container is removed from the oven and the substrate is washed to remove any surplus solution. As will be demonstrated and explained hereafter in detail, the resulting washed (and optionally dried) substrate will be found to include a film comprised of a highly regular and orientated array of crystalline nanorods which has grown on the surface of the substrate in such a way that the rods are elongated along the crystal c-axis. As explained above in detail, this is unusual - but highly advantageous - given that reported tin (IV) oxide anisotropic nanocrystals grew along other axes. In certain prior art film deposition methods it has been necessary to subsequently heat treat (anneal) the substrate on which a film has been formed at a high temperature (typically in excess of several hundred degrees centigrade) to convert the film as deposited into a crystalline structure. The ultimate effect of this is to severely limit suitable substrates to only those which can withstand such high temperatures. An advantage of the method disclosed herein is that as the as-grown film is crystalline in structure, no such heat treatment is required. This means that the method disclosed herein can utilise any substrate. In the preferred embodiment, the solution 1 comprises an aqueous solution of a tin salt in an acid medium that also includes a substance, the like of which is well known in the art, which is a slow release source of a basic medium (for example upon heating of the solution) - the slowly released basic medium (hydroxyl ions) reacting with the tin salt to condense it into tin oxide onto the substrate. In general terms, the solution provides for the hydrolysis-condensation of hydrated metal cations via aqueous thermal decomposition (thermohydrolysis) of SnIV in an acid medium with reagent grade chemicals.
In a highly preferred embodiment, a 100 ml aqueous solution (MilliQ+, 18.2 MΩcm) consisted of 0.034g of SnCl4.5H20 and 0.920g of (NH2)2CO (known variously, but not exclusively, as Urea, carbamide or carbonyl diamine) in the presence of 5ml of fuming HC1 (37%). Urea was chosen because it is a stable, non-ionic, non-toxic, inexpensive, crystalline and water soluble compound; and because it affords simultaneously the hydrolysis-condensation by olation of the tetravalent post-transition metal ion Snw, most probably from the complex of zero charge [Sn(OH) (H2O)2]°, and the nucleation and growth of the stable tin oxide form rutile SnO2 by oxolation from the slow release of hydroxyl ions owing to its well-known thermal decomposition in aqueous solutions. It will be appreciated, however, by those persons of ordinary skill in the art that a variety of alternative compounds with similar slow-release of base (hydroxyl ions) properties may instead be used without departing from the spirit and scope of the present invention. Tin chloride was chosen because it is relatively stable, non-toxic, readily available as a reagent chemical (and is hence relatively inexpensive), and is soluble in water. It will be appreciated, however, by those persons of ordinary skill in the art that any tin salt may instead be used without departing from the spirit and scope of the present invention. Similar considerations apply to the choice of the acid: fuming hydrochloric acid. Once the solution has been prepared a substrate 3 is placed substantially upright against a wall 4 of the container 2, and the container is placed in an oven maintained at a relatively mild temperature (such as between 25 to 125 degrees, preferably 90 to 100 degrees, and most preferably 95 degrees centigrade) for a predetermined period of time - such as one to three, preferably two, days or thereabout. In circumstances where other slow-release compounds are used in place of urea it may not be necessary to heat the solution and as such the scope of the present invention should not be considered to be limited to a process where the solution must be heated. It is also the case that the time for which the solution is heated, if indeed it is heated, may be varied as required from a few hours to several days in order to produce a substrate with the required film morphology and thickness. In the preferred arrangement, the substrate 3 comprises a polycrystalline F- SnO2 glass substrate (i.e. a glass substrate coated with a layer of polycrystalline
fluorine doped tin (N) oxide - F-SnO2). One such suitable substrate is known as TEC- 8115 and is manufactured by Hartford Glass Company Inc. of Hartford City, Indiana, USA. Other commercially available transparent conducting oxide (TCO) substrates such as, for instance, indium tin oxide (ITO) can also be used. As an alternative, the substrate 3 could comprise a silicon or silicon oxide wafer; a bare piece of glass (such as a microscope slide) cleaned with diluted acid, ethanol and acetone and subsequently rinsed with MilliQ-water; plastics (such as fluorinated ethylene-propylene copolymer (FEP) and polypropylene (PP), amongst others); ceramics; metals or indeed any of a number of other suitable substrates. In general terms, the substrate can be any substrate which is coated with other materials or uncoated, of any thickness, composition or crystallinity, and can even be of relatively low surface quality (any imperfections merely serving to increase the growth effect). For photoelectochemical conversion of water to hydrogen and oxygen, it is particularly preferred for the substrate to comprise a transparent plate (e.g. a glass plate) to which a conducting layer has been applied, the nanorod tin oxide film being grown on that conducting layer. Once the aforementioned predetermined period of time has elapsed, the substrate can be removed from the container and washed, for example with MilliQ water, to mitigate potential contamination from residual salts. The substrate produced by this method includes a film that is comprised of well-aligned crystalline nanorods of generally square cross-section (as one might expect given the crystal structure and face stability of rutile and other iso-structural compounds) elongated along the aforementioned crystal c-axis and arranged in large, relatively uniform arrays. Typically, the nanorods grown by the method of the preferred embodiment are in the order of roughly 50 nm in width and about 500 nm in length. In the preferred embodiment the c-axis elongated nanorods are orientated such that each said rod is comprised of crystals grown along a c-axis which is generally perpendicular to the plane of the substrate. Advantageously, the orientation of the c-axis of the crystals and hence each nanorod with respect to the substrate can be varied by varying certain processing conditions, such as for example the metal salt concentration, pH and/or salt concentration of the solution, of the method disclosed. In general terms it is possible
by varying the processing conditions to grow nanorods with crystal c-axes orientated at any angle between 0 and 90 degrees to the substrate. It is even possible, again by varying the processing conditions, to grow an array of nanorods where the nanorods have different orientations with respect to the substrate. In cross-sectional elevation such an array would look similar to an opened fan and would be particularly advantageous in the field of photo electrochemical conversion of water to hydrogen and oxygen as the "fan" shape would ensure that at least some of the nanorods are pointing towards the sun as it moves during the course of a day. Figs. 2a and 2b are scanning electron micrographs (SEMs), at various magnifications, of a film of tin (IV) oxide nanorods grown on a substrate in accordance with the method above described. As shown in Figs. 2a and 2b, the film grown on the substrate clearly comprises a plurality of generally square cross-section nanorods, each rod in this instance being in the order of 50 nm in width and roughly 500 nm in length. To verify that the nanorods grown are indeed of rutile crystal structure and elongated along the crystal c-axis, an x-ray diffraction plot of a film produced in accordance with the method of the preferred embodiment was obtained and the results are depicted graphically in Fig. 3. Fig. 3 provides experimental confirmation that a film grown in accordance with the method of the preferred embodiment includes consists of rutile SnOz (cassiterite) with a texture effect that is in agreement with [001] orientated nanorods and a plurality of (110) faces. As is evident from Fig. 3, the relative maximum intensity sequence for a film produced in accordance with the teachings of the present invention is no longer (110) > (101) > (211) > (200) > (301) ~ (310) ~ (220) » (111) » (210) (as one would expect for a conventional rutile tin oxide crystal), but (101) > (110) > (211) » (002) ~ (200) ~ (112) > (111) ~ (220) ~ (301). Fig. 4 shows HRTEM (High Resolution Transmission Electron Microscopy) and fast Fourier transform (FFT) images of a prepared polycrystalline thin film (rutile). As is clearly visible in Fig. 4, each 50 nm nanorod grown in accordance with the teachings of the present invention consists of bundles of finer nanorods of about 2- 4 nm in diameter. The spacing of the lattice fringes was found to be 0.22, 0.235, and 0.16 nm, and these planes can best be indexed as (110), (111), and (002) of rutile SnO2, respectively. As a result, it is clear that the direction of growth of the SnO2
nanorods is along the c-axis and with side and top faces consisting of (110) and (001) planes, respectively. These nanofibres, being of a diameter which falls in the quantum size regime, are of significant relevance for the development of nanodevices. Specifically, the fact that each said nanorod is comprised of a plurality of much smaller nanofibres is highly advantageous as the size of those fibres is such that the likelihood of electron / hole recombination is significantly reduced, even for those minority carriers (holes for n-type semiconductor) with relatively short diffusion lengths. A further advantage is that the fibres are of such a size as to exhibit the aforementioned, and well known, quantum size effect. As has been explained above in detail, the (110) faces provide, in addition to improved stability, an improved surface and atomic configuration for an efficient chemisorption and dissociation of (for example) oxygenated compounds at the SnO2 interface due largely to the fact that at this point the interatomic distances between neighbouring tin atoms are at their lowest as compared to the (101) and (111) faces. Such particularity is of importance for improving the efficiency of current SnO2 devices, such as sensing and catalytic devices. Furthermore, as the exposed prismatic faces of the SnO2 nanorods described herein are the most stable and non-polar (110) faces, such unique morphology (square cross-section with well-defined faces) confers to these arrays great capabilities to develop innovative, functional and efficient SnO2 nanosensors. It is apparent from the theoretical and confirmatory experimental evidence presented above, that by virtue of the teachings of the present invention it is now possible to grow a tin oxide film on a substrate that includes a plurality of nanorods which are elongated along the crystal c-axis. It is also possible, by controlling the processing conditions of the method to orientate those nanorods with respect to the substrate. As mentioned above, this is a highly advantageous architecture for many applications. One such illustrative application is depicted in Fig. 5 and described hereafter. This application provides a cell for the well-known phenomenon of water cleavage into hydrogen and oxygen by incident light. Fig. 5 is an illustrative representation of a photoelectrochemical cell 20 for water cleavage which employs a photovoltaic cell 22 (such as a silicon photovoltaic cell). As shown, the part of the cell on the left (as depicted) comprises a compartment 24 which contains an aqueous electrolyte that is subjected to water photolysis. In the
preferred arrangement the electrolyte comprises water to which an electrolyte has been added for ionic conduction, or seawater. Light enters from the left side of the cell through a glass window 26. The light then crosses the electrolyte and impinges upon the front face of an electrode (28, 30, 32) which has been produced in accordance with the process described above (the electrode comprising a glass plate 32, a conducting layer 30 and a tin (IV) oxide film 28). The tin (IV) oxide film 28 absorbs part of the solar spectrum, and transmits the remainder to the photovoltaic cell 22 which in this instance is provided behind the back face of the tin (IV) oxide electrode. The photovoltaic cell 22 functions as a light driven electric bias which is operable to increase the electrochemical potential of the electrons that emerge from the tin (IN) oxide film. Behind the second cell there is provided a chamber 34 bounded by a glass plate 36 in which an electrolyte (of the same composition as that provided in the first cell) is provided, the two electrolytes being in fluid communication with one another by means of a glass frit 38 or ion conducting membrane. As depicted in Fig. 5, incident light cleaves water so that oxygen is evolved from the first compartment 24, and hydrogen is evolved at a cathode 40 immersed in the chamber 34. Another illustrative application is shown schematically in Fig. 6 of the accompanying drawings. This application provides a gas sensor having, as a core constituent, a substrate and crystalline film as hereinbefore described, as mentioned above, semiconducting tin (IV) oxide is well known in the art for its utility as a core component of a gas sensor. Referring now to Fig. 6, the gas sensor 50 comprises a substrate 3 which has had a tin (IN) oxide film grown on it by means of the method described herein. The substrate is coupled to a sensing device 52 which comprises, inter alia, a voltage source 54, an ammeter 56 (in series with the voltage source) and control logic 58. The voltage source 54 is configured to apply a known voltage to the substrate 3 and hence to the film provided thereon. Application of a voltage to the film causes a current to flow, and this current is measured by the ammeter 56. On exposure to certain gases (such as hydrogen, carbon monoxide, and many other gases the identity of which are well known in the art), the resistivity of the film changes, and this change in resistivity is manifested as a change in the current measured by the ammeter 56.
The control logic 58 is coupled to the ammeter and, in the preferred arrangement, is calibrated such that a current detected when the film is not exposed to a flow of a gas to be detected is taken as a nominal "zero point". Subsequent fluctuations in measured current on exposure to the gas or gases to be detected cause a derivation from this nominal "zero point", which can be reported to the user of the sensor by the control logic, for example by sounding an alarm. The control logic may also be configured to include a threshold, below which a change in resistivity that is indicative of only a relatively small amount of gas is not notified to the user. Previous prior art studies, the like of which will doubtless be well known to persons skilled in the art, have shown that systems employing such gas sensing films can be tailored not only to detect the presence of a given gas, but also to detect (at least approximately) the amount of that gas. Some previously proposed systems have been configured to detect multiple gases and to report, by virtue of the electrical properties of the film on exposure to the gas, not only the likely concentration of the gas but also the likely identity of that gas. Any of these prior art teachings can be enhanced by employing a film as herein described. In particular, by utilising a film manufactured in accordance with the teachings provided herein, either as a film on a single substrate or as an array of substrates and associates films, it is possible to provide a sensor that is more sensitive and reactive (i.e. more quickly responsive) than those sensors previously proposed. It is also the case that with a film produced in accordance with the teachings provided herein the (110) crystal faces of each nanorod (i.e. the most efficient crystal faces for the adsorption and dissociation of oxygenated compounds (CO, CO2, H2O etc.) as well as other gases (e.g. methane)) are exposed to the incident gas. Exposure of these crystal faces provides the sensor with both an increased yield and an increased selectivity as compared with existing SnO2 gas sensors. Another advantage of the preferred embodiment in particular is that the perpendicular orientation of the nanorods along with their c-axis crystal growth provides a good path (both physically and electronically) for the collection of electrons (signal) generated on exposure of the sensor to a gas to be sensed. This advantage also improves the performance of the film disclosed herein when used in a photocatalytic or photovoltaic system. Further advantages of the film disclosed herein, particularly in the case of a gas sensor, are that the penetration (exposure) of the film to the gas is increased (in
comparison with existing sensing films) as a result of the channels between adjacent nanorods. In other words, arranging the film as a series of nanorods, preferably upstanding nanorods, grown along the c-axis of the tin (IN) oxide crystal, allows the surface area of film exposed to the gas to be greatly increased. Whilst preferred embodiments of the present invention have been described above in detail, it will be understood and should be noted that this description is provided merely by way of illustrative example. As such, it should be noted that the scope of the invention is not limited to the particular embodiments described, and that modifications may be made without departing from the spirit and scope of the invention. For example, in a related aspect of the present invention it is possible to transform a substrate having a film formed in accordance with the teachings of the invention into a conducting substrate simply by doping and/or firing the substrate (and film) to render it doped or non-stoichiometric. In this modification it is preferred for the nanorods to be grown to a length of something in the order of less than hundreds of nanometres, preferably less than 200 nm. As a further example, it will also be appreciated by those persons skilled in the art that the process described above utilises quantities of chemicals that are appropriate for laboratory experimentation. Scaling up the teachings of the invention for commercial production will require these quantities to be increased. In a similar vein, it will also be appreciated by a person skilled in the art that references herein to an "aqueous solution" do not necessarily imply that the solution includes pure water, but could instead refer to a mixture of water and any other substance. It should also be noted that whilst certain combinations of features herein described have been explicitly enumerated in the accompanying claims, the invention is not limited to those particular combinations and instead extends to any combination of features described herein irrespective of whether that particular combination has explicitly been claimed at this time.