DE102009057697A1 - Chemical media sensor, process for its production and uses - Google Patents

Abstract

The present invention relates to a method for producing porous nanogranular electrode layers for chemical media sensors, in which a porous coating of noble metals is produced on a semiconductor substrate. Likewise, the present invention relates to a chemical media sensor having a previously described electrode coating on a substrate. According to the invention also uses of the media sensor are given.

Description

The present invention relates to a method for producing porous nanogranular electrode layers for chemical media sensors, in which a porous coating of noble metals is produced on a semiconductor substrate. Likewise, the present invention relates to a chemical media sensor having a previously described electrode coating on a substrate. According to the invention also uses of the media sensor are given.

Gas-sensitive field-effect transistors based on semiconductors (ChemFET) are increasingly being used in sensor technology. In this case, the admission of the gate with the test species to be detected (for example a gas or a liquid or a mixture of gas or liquid) usually leads to a change of the potential at the gate electrode, whereby the channel impedance changes. This changes the current (channel current) flowing through the transistor from source to drain contact (eg. US 5,698,771 ). For low-temperature applications (up to approx. 200 ° C) silicon is particularly suitable as a semiconductor material. If one uses semiconductor materials with a large band gap (> 3 eV), such as. For example, gallium nitride (GaN) or silicon carbide (SiC), in principle, allows the use of these ChemFETs for sensor applications at temperatures up to 800 ° C in engine exhaust as well.

The semiconductor is usually used in the form of thin slices (wafers), which are initially provided with the required areas of defined conductivity and the electrodes. After application of the active electrode, the disks are usually divided into individual chips and mounted in housing.

For the sensitivity of the ChemFET sensors to different gases, the structure of the metallization of the gate electrode is crucial. While closed metal films with platinum or palladium are only sensitive to hydrogen, porous gate electrodes show high sensitivities to ammonia, hydrocarbons, hydrogen and nitrogen oxides (NO and NO ₂ ).

According to the state of the art, the production of the gate electrodes of FETs is carried out by semiconductor-compatible processes known to the person skilled in the art, such as the sputtering or vapor deposition of metals, eg. Aluminum, platinum, nickel, etc. The room temperature deposited gate layers are nearly closed, nonporous and thermally unstable metallic films. The achieved layer thicknesses vary between a few nanometers and a few micrometers. These closed metal layers are thermally unstable especially at higher temperatures (T> 200 ° C) and lose their closed structure. The coarsening and cracking in these layers can not be specifically controlled and are too coarse for the target application described here.

In the DE 10 2006 046 225 For example, a gas sensor based on a gas-sensitive field-effect transistor is disclosed. Here, a coated, gas-sensitive gate electrode is described. The gas-sensitive material can be applied by thick film technology or thin film technology. The sputtering is called. In contrast to the present invention, however, the invention relates only to the application in the low temperature range to a maximum of 90 ° C indoors. In addition, the production methods described here for a wafer-based coating are too cost-intensive.

A field-effect sensor with a gate electrode arrangement is described, for example, in US Pat DE 35 19 576 described. The gate electrode assembly also describes a layered structure including a titanium layer and a noble metal layer.

To achieve a high sensitivity of the gas sensor, the active electrode must have the largest possible surface area. In order to limit the size and minimize interference, usually layers are used with an open-porous structure, which often consist of metals for reasons of electrical conductivity.

So was already in DE 10 2007 015 874 a wet-chemical coating method for producing a highly porous electrode structure is described. In this case, nanoscale particles are applied to the gate region from colloidal solutions and converted by means of thermal aftertreatment into a metallic or ceramic conductive layer.

A very high sensitivity can be achieved if the inner surface of the porous structure and thus the total length of the so-called three-phase boundaries (electrode-dielectric-gas environment) is very large. This applies in particular to structures with nanometer dimensions.

The selectivity based on catalytic activity is also determined by the feature sizes. In general, the catalytic efficiency increases with decreasing radius of the catalyst particles and with increasing proportion of exposed phase boundaries.

The object of the present invention is, starting from the state of the art, a simplified and more cost-effective method for the production of porous electrode layers on substrates (for example as part of a launcher). Technology), whereby an improved homogeneity of the porosity of the structures can be achieved. It is likewise an object of the present invention to provide a gas sensor which, due to the improved homogeneity of the porous structuring of the electrode layer, has a higher sensitivity and a more homogeneous measuring accuracy. It is another object of the present invention to provide appropriate uses of such a sensor.

This object is achieved with respect to the method for producing porous, nanogranular electrode layers having the features of patent claim 1 with respect to the chemical media sensor having the features of patent claim 13 and with respect to the intended use of the media sensor having the features of patent claim 16. The respective dependent claims represent advantageous developments.

According to the invention, a method for the production of porous, nanogranular electrode layers for chemical media sensors for gases and / or liquids is proposed, in which a substrate by means of physical vapor deposition (PVD) at a working pressure in the range of 0.001 to 5 mbar with nanogranular particles of a catalytically active substance is coated.

Decisive in the method according to the invention is that the electrode layer deposited in the PVD process is structured in a porous or nanogranular manner. This is achieved by separating several areas in the PVD process. In a first region, the ablation of atoms of a target substrate takes place, which are accelerated in the direction of the substrate to be coated. Between the target substrate and the substrate to be coated, the individual atoms detached from the target substrate are agglomerated in an area of nucleation / clustering into nanometer-sized particles. These agglomerated particles whose mean diameter are in the nanometer range, finally strike the surface of the substrate to be coated and are deposited on the surface thereof. Due to the fact that not individual atoms, but already clusters of a certain size impinge on the surface of the substrate to be coated, resulting in a quasi-inhomogeneous deposition of the material from which the electrode layer is to be produced, which eventually leads to the porous formation of the electrode layer.

It is particularly advantageous in the aforementioned PVD method, for example compared to wet-chemical methods known from the prior art, that uniform coatings can be produced in a simple manner both with regard to the porosity and the layer thickness. Likewise, it is possible to dispense with further process steps, as provided for in the wet-chemical process. Furthermore, the layer coverings to be achieved over the duration of the coating or the rate at which the nanoparticles are deposited on the surface of the substrate to be coated can specifically influence the electrical properties of the electrode layer by means of the PVD coating method, which in the case of the wet-chemical methods mandatory tempering step was left to chance.

In a preferred embodiment of the method according to the invention, it is provided that the mean particle diameter d _{50 of} the nanogranular particles is set to a size between 1 and 250 nm, preferably between 5 and 100 nm, particularly preferably between 10 and 50 nm. By means of particle sizes of this type, optimal deposition rates on the one hand and particularly advantageous porosities of the electrode layers on the other hand can be achieved.

• a) Hauptgruppenmetalle, insbesondere Indium,
• b) Übergangsmetalle, insbesondere Platin, Palladium, Gold, Iridium, Rhodium, Ruthenium, Hafnium, Ytterbium, Zirkon, Titan, Scandium, Niob, und/oder Rhenium,
• c) Lanthanide, insbesondere Lanthan und/oder Cer,
• d) sowie deren Legierungen und/oder Mischungen.

Advantageous catalytically active substances that can be deposited on the substrate are selected from the group of

• a) main group metals, in particular indium,
• b) transition metals, in particular platinum, palladium, gold, iridium, rhodium, ruthenium, hafnium, ytterbium, zirconium, titanium, scandium, niobium, and / or rhenium,
• c) lanthanides, in particular lanthanum and / or cerium,
• d) and their alloys and / or mixtures.

Preference is given to using noble metals according to the invention in order to deposit them on the substrates.

PVD deposition is preferably followed by sputtering techniques, such as gas flow sputtering or magnetron sputtering.

In particular, the coating is carried out at a pressure of 0.003 mbar to 2 mbar, in the case of magnetron sputtering at 0.003 to 0.03 mbar, in the case of gas flow sputtering at 0.1 to 2 mbar. In addition, a distance of at least 80 mm, preferably 80 to 400 mm, in particular 100 to 200 mm, is preferably set between the sputtering source and the wafer. In this way, metal clusters in the size between 1 to 250 nm, preferably between 10 to 50 nm are generated by sputtering and directed onto the wafer. The resulting pores have a pore diameter of 0.1 to 250 nm, preferably between 1 to 50 nm. Thus, a degree of coverage of the surface to be coated by 50 to 90%, preferably from 60 to 75% can be achieved. The porosity is necessary to form a sufficiently large total length of three-phase boundaries that are most responsible for the gas reaction leading to the signal forming sensor.

In this case, layer thicknesses in the range between 10 to 500 nm, preferably between 20 to 200 nm, can be achieved. These layer thicknesses are sufficient for a necessary electrical conductivity of the electrode layer for the function of the sensor and allow for the sensors required rapid gas passage and gas exchange with the environment.

The substrate temperature during the coating process is between room temperature (i.e., 20 ° C) and 250 ° C, preferably between 80 and 140 ° C.

Due to the requirement according to the invention for the nanoporous electrode layer, in particular noble metals, such as platinum, palladium, gold, iridium, rhodium, ruthenium, indium, subgroup metals, in particular hafnium, lanthanum, ytterbium, zirconium, titanium, scandium, niobium, rhenium and lanthanides, such as lanthanum and cerium, as well as alloys and mixtures thereof. For the purpose of different catalytic activity, it is also possible to sequentially deposit multiple coatings with different materials (= laminates) for the purpose of forming specific selectivities. On the one hand, pre-reactions / catalytic conversions of undesired gases to upper material layers can take place so that they do not later contribute to signal formation. On the other hand, materials can additionally protect the layers with the three-phase boundaries (= contact layer, directly on the substrate surface) mechanically, thermally and against corrosive attacks.

For the purpose of a spatially resolved electrode coating (electrode is located only on a portion of the sensor chip), the nanoporous layer can be patterned on the substrate by means of different structuring options. For this purpose, lift-off processes, the use of a hard mask and / or so-called back sputtering methods are suitable.

The principle procedure is described as an example at the lift-off process:
The substrate to be coated before the electrode coating with a photoresist, with z. B. a spinning process, coated. The paint is exposed and developed at the points where the nanoporous electrode is to be located later. Ie. at these points, the substrate (the sensor) is no longer coated with photoresist. Thereafter, as described, the deposition of the porous, nanogranular electrode layer takes place over the entire surface of the substrate (for example, a wafer). After coating, the photoresist is dissolved in an organic solvent. In this case, all parts of the electrode layer, which are located on the photoresist, dissolved from the substrate and gives a localized electrode layer.

The application-specific structure sizes are in the range of 200 nm to 1 mm, preferably between 1 and 300 @m.

The activation and conditioning of the open-pored, nanoporous layer (s) is preferably carried out by one or more subsequent annealing or sintering steps at temperatures between 200 ° C and 700 ° C for a few minutes to hours. As a result, a thermodynamically stable nanoscale structure is achieved by reducing stresses or mechanical stress in the layer through the deposition and / or the material and achieving sintering to the substrate surface.

A further preferred embodiment of the method provides that between the substrate and the sputtering source, a potential difference is applied, which is preferably between 15 to 200 V, more preferably between 20 and 50 V.

Inert gases, in particular argon, krypton and / or xenon, are advantageously used as the sputtering gas.

Likewise, the possibility is provided that in addition to the inert gas, which serves as a sputtering gas, other gases such. As reactive gases or other inert gases, in the gas atmosphere, which is used in the process can be used. In particular, gases which are selected from the group consisting of oxygen, nitrogen, a hydrocarbon or mixtures thereof are suitable here.

It is possible to coat the substrate vertically, d. The nanoclusters formed during the process impinge on the surface of the substrate to be coated at an angle of incidence of approximately 90 °. However, it is also possible to coat the substrate by oblique vapor deposition.

Suitable substrate materials particularly preferred semiconductor in question, especially semiconductor selected from the group consisting of Si, Ge, GaN, SiC, GaP, GaAs, InP, InSb, InAs, GaSb, GaN, AlN, InN, Al _x Ga _1-x As, In _x Ga _1-x N, ZnO, ZnS, ZnSe, ZnTe, CdS, CdSe, CdTe, Hg _(1-x) Cd _x Te, BeSe, BeTe and / or HgS. Likewise, there is the possibility that the aforementioned semiconductor materials by Foreign atoms can be doped, for example N-doped or P-doped.

The coating can be carried out in one or more coating cycles, each with its own or common Temperschritten.

According to the invention, a chemical media sensor for gases and / or liquids is likewise provided, which comprises a porous, nanogranular electrode layer which is applied at least in regions to a substrate, characterized in that the average pore diameter d _{50 is} between 0.1 and 250 nm, preferably between 1 and 50 nm. The above-mentioned media sensor can be produced particularly advantageously on the aforementioned procedure. The media sensor according to the invention is distinguished by a large overall length of three-phase boundaries, which are mainly responsible for the gas or liquid reactions leading to signal formation at the sensor. Due to the large total length of the aforementioned three-phase boundaries and the resulting large surface of the electrode layer of the sensor, this has a high sensitivity for analytes to be examined. Likewise, this open-pore, nanoporous layer has a high catalytic activity.

By choosing the special parameters in the production of the electrode layer, the electrical conductivity of this layer can be largely influenced and thus tailor-made to the particular application. In particular, the electrical conductivity σ of the electrode layer at room temperature is between 1 Ω ^-1 cm ^-1 , to 1 MΩ ^-1 cm ^-1 .

According to the invention, sensors for gaseous and / or liquid analytes are thus provided in particular.

Gas sensors are understood below to mean electronic components whose electrical output signal is dependent on the concentration of specific gases in the vicinity of the active electrode of the gas sensor. Analogously, liquid sensors should be understood.

Gas sensors are used in safety technology to warn of excessive concentrations of dangerous gases or trigger automatic protection reactions.

Furthermore, gas sensors find application in the control of chemical conversion processes.

An important area here is the control of combustion processes for power generation, d. H. Conversion of chemical into mechanical or thermal energy, for example in power plants, stationary gas turbines, aircraft engines and piston engines, especially for vehicles.

The aim is to optimize the combustion of fuels and to minimize the resulting harmful exhaust gases, such as carbon monoxide, nitrogen oxides and hydrocarbons.

Important characteristics of gas sensors are their sensitivity, their selectivity, their response time and their lifetime.

According to the invention mentioned in claim 16 uses of the media sensor are also given.

These sensors described here are thus used in particular for measuring an analyte with a low concentration in the range of ppb to low percentage ranges, in particular for measuring in the ppm range. In particular, nitrogen oxides (NO, NO ₂ , N ₂ O), ammonia, carbon monoxide and hydrocarbons in a carrier gas of nitrogen, oxygen and optionally water can be measured as gas analytes. Liquid analytes are in particular ions of dissolved salts and organic components, but also reactive molecules, such as. B. hydrogen peroxide (H ₂ O ₂ ).

The present invention will be explained in more detail with reference to the following examples and the accompanying figures, without limiting the invention to the parameters specifically shown.

Example 1:

Coating of a pre-structured silicon wafer with a platinum active electrode by means of gas flow sputtering

A structured silicon wafer is placed in a vacuum chamber, which is subsequently evacuated. The wafer is heated to 120 ° C. Thereafter, the vacuum chamber is filled with argon, so that the pressure in the coating chamber is 0.5 mbar.

The gas flow sputtering source (with platinum as a sputtering target) is burned in with the source shutter closed for 3 minutes. Thereafter, the source shutter is opened and the wafer is coated for 20 minutes. The wafer is thereby moved in a linearly oscillating manner in front of the source, so that the coating of the wafer, which is as homogeneous and as flat as possible, is achieved. After completion of the layer, the separation of the electrical potentials and blocking the gas supply takes place. The vacuum chamber is subsequently vented and the wafer removed. After structuring the layer in a subsequent tempering step, the separation of the finished chips, which can be mounted and used directly afterwards.

Example 2:

Coating of a pre-structured wafer with a platinum-palladium alloy active electrode by means of magnetron sputtering

A structured silicon wafer is placed in a vacuum chamber, which is subsequently evacuated. Thereafter, the vacuum chamber is filled with argon, so that the pressure in the coating chamber is 0.08 mbar. The magnetron sputtering source (with a platinum-palladium alloy sputtering target) is baked with the source shutter closed for 5 minutes. Thereafter, the source shutter is opened and the wafer is coated for 40 minutes. The wafer is thereby moved in a linearly oscillating manner in front of the source, so that the coating of the wafer, which is as homogeneous and as flat as possible, is achieved. After completion of the layer, the separation of the electrical potentials and blocking the gas supply takes place. The vacuum chamber is subsequently vented and the wafer removed. After structuring of the layer in a subsequent tempering step, the singulation of the finished chips takes place, which can then be directly mounted and used.

The coatings obtained by the above coating methods are shown in the accompanying drawings 1 and 2 shown. 1 shows a scanning electron micrograph of an open-pore nanoporous metal layer, which was prepared according to Example 1 in a gas flow sputtering method. The open-pored structures of the electrode coating are clearly recognizable.

In 2 a scanning electron micrograph of an open-pore nanoporous metal layer is shown, which was prepared according to Example 2 by a Magnetronsputterverfahren.

By the two different procedures, different coatings can be achieved. The coating produced according to the gas flow sputtering process has a coarser structuring than the coating produced according to Example 2, which has a very homogeneous porous structure.

Furthermore, the metal clusters produced in accordance with the gas flow sputtering method in the porous layer have an average particle diameter d ₅₀ of 10 to 100 nm, which causes the coarse-pored structure. On the other hand, the metal clusters prepared according to the magnetron sputtering method show an average particle diameter d ₅₀ of <10 to 20 nm, thereby enabling a finer-scaled nanogranular layered structure.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 5698771 [0002]
• DE 102006046225 [0006]
• DE 3519576 [0007]
• DE 102007015874 [0009]

Claims (16)

Process for the production of porous, nanogranular electrode layers for chemical media sensors for gases and / or liquids, in which a substrate is coated by means of physical vapor deposition (PVD) at a working pressure in the range of 0.001 to 5 mbar with nanogranular particles of a catalytically active substance. A method according to claim 1, characterized in that the mean particle diameter d _{50 of} the nanogranular particles is between 1 and 250 nm, preferably between 5 and 100 nm, particularly preferably between 10 and 50 nm. Method according to one of the preceding claims, characterized in that the catalytically active substance is selected from the group of a) main group metals, in particular indium, b) transition metals, in particular platinum, palladium, gold, iridium, rhodium, ruthenium, hafnium, ytterbium, zirconium, titanium, scandium, niobium, and / or rhenium, c) lanthanides, in particular lanthanum and / or cerium d) and their alloys and / or mixtures. Method according to one of the preceding claims, characterized in that the coating is carried out by means of sputtering, in particular gas flow sputtering or magnetron sputtering. Method according to one of the preceding claims, characterized in that the substrate from the sputtering source at least 8 cm, preferably 8 to 40 cm, in particular 10 to 20 cm is spaced apart. Method according to one of the preceding claims, characterized in that a bias voltage is applied to the substrate, which is preferably between 15 to 200 V, more preferably between 20 and 50 V. Method according to one of the preceding claims, characterized in that argon, krypton and / or xenon is used as the sputtering gas. Method according to the preceding claim, characterized in that the sputtering gas in addition to argon at least one further gas, in particular selected from the group consisting of oxygen, nitrogen, a hydrocarbon or mixtures thereof. Method according to one of the preceding claims, characterized in that the substrate is coated by Schrägbedampfung. Method according to one of the preceding claims, characterized in that the substrate temperature during the coating between 20 and 250 ° C, preferably between 80 and 140 ° C. Method according to one of the preceding claims, characterized in that the substrate is selected from the group consisting of semiconductors, in particular Si, Ge, GaN, SiC, GaP, GaAs, InP, InSb, InAs, GaSb, GaN, AlN, InN, Al _x Ga _1-x As, In _x Ga _1-x N, ZnO, ZnS, ZnSe, ZnTe, CdS, CdSe, CdTe, Hg _(1-x) Cd _x Te, BeSe, BeTe and / or HgS. Method according to one of the preceding claims, characterized in that, following the coating step, the coated substrate is subjected to a tempering step at temperatures between 200 ° C and 700 ° C. Chemical media sensor for gases and / or liquids, comprising a porous, nanogranular electrode layer which is applied at least in regions on a substrate, characterized in that the average pore diameter d ₅₀ between 0.1 and 250 nm, preferably between 1 and 50 nm. Media sensor according to the preceding claim, characterized in that the layer thickness of the electrode layer is between 10 and 500 nm, preferably between 20 and 200 nm, particularly preferably between 50 and 100 nm. Media sensor according to one of the two preceding claims, characterized in that the electrode layer at room temperature an electrical conductivity σ in the range between 1 μΩ ^-1 cm ^-1 to 1 MΩ ^-1 cm ^-1 . Use of the media sensor according to one of claims 13 to 15 as a capacitive media sensor and / or field effect-resistive media sensor, in particular as a field effect transistor, as a resistive media sensor, as a semiconductor gas sensor, as a CMOS gas sensor, for the detection of nitrogen oxides, ammonia, hydrocarbons, oxygen and / or for the exhaust aftertreatment in internal combustion engines, in particular piston engines or aircraft engines, as a sensor for gaseous and / or liquid analytes.

Images