Field of the Invention

The invention relates to an electron multiplier assembly for multiplication of electrons in an avalanche between electrode plates and to an avalanche particle detector comprising such an electron multiplier, as well as to a method for multiplying electrons in an amplification gap.

Background and State of the Art Avalanche particle detectors are widely used in particle physics for the detection, tracking and identification of radiation or particles. These detectors have also found wide applications in biology and medical technology.

Avalanche detectors typically collect primary electrons released by particle impact or radia- tion and guide them to an amplification gap, which is a region with a strong electric field. In the amplification gap, the primary electrons initiate an electron avalanche by impact ionization. The avalanche generates enough electrons to create a pulse current large enough to be collected on a readout device and analyzed by readout electronics. The collected electron charge may indicate the presence and physical properties of the incident particles or radiation, such as charge, energy, momentum, direction of travel or other attributes.

In some conventional detectors, the strong amplification field which is necessary to initiate a Townsend avalanche comes from a thin wire at a positive high voltage potential. This same thin wire also collects the electrons from the avalanche and guides them towards the readout electronics. More recently, attention has focused on so-called micro pattern gas detectors (MPGDs), which can be manufactured by employing semiconductor fabrication techniques and hence can be mass-produced at low costs and in an impressive variety of geometries.

One such detector that has found wide applications is the gas electron multiplier (GEM) de- tector described in EP 0 948 803 Bl and in Fabio Sauli, "Gas Electron Multiplier (GEM) De- tectors: Principles of Operation and Applications", Cem Report RD51 -NOTE-2012-007. Fig. 1 is a schematic diagram taken from Sauli's paper that shows the general structure and functioning of a GEM 100. The detector 100 comprises a detection chamber filled with a gas, for instance a mixture of argon and methane. In the detection chamber, a GEM plate 1 10 is lo- cated between a drift electrode 112 and a collection electrode 114 comprising a plurality of readout pads 116. The GEM plate 110 consists of a dielectric insulator sheet 118 sandwiched between a metallic cathode layer 120 and a metallic anode layer 122. In the GEM plate 110, a plurality of through-holes 124 are arranged in a matrix or array pattern. The through-holes 124 extend through the entire GEM plate 110, i.e., through the cathode layer 120, dielectric insulator sheet 118 and anode layer 122.

In the original GEM design described in EP '803 and by Sauli, the through-holes 124 typically have a diameter of 20 to 100 μπι and are arranged with a pitch of typically 50 to 300 μιη. The thickness of the insulator sheet 118 could be about 50 μπι, and the cathode layer 120 and the anode layer 122 typically have a thickness of about 5 μηι. A method of manufacturing a gas electron multiplier by means of photolithography and etching techniques is described in International Patent Publication WO 2009/127220 Al.

More recently, so-called Thick-GEM detectors (TGEMs) were introduced as a simpler and more robust design in which holes may be drilled in standard printed circuit boards. The TGEMs can be thought of as scaled-up versions of the original GEM detectors, with a hole diameter typically in the range of 0.3 to 1 mm, a pitch between neighboring holes of typically between 0.7 and 2 mm, and a thickness in the range of 0.4 to 3 mm. A more detailed description of TGEMs can be found in V. Peskov et al, Nucl. Instrum. Methods A 478 (2002) 377.

So-called resistive TGEMs (RETGEMs) employ resistive electrodes instead of metallic electrodes, which provides superior protection against sparks and discharges. RETGEMs are described in further detail in V. Peskov et al, Nucl. Instrum. Methods A 576 (2007) 362. Operation of the original GEM detector, the Thick-GEM (TGEM) and the resistive TGEM (RETGEM) are basically alike, and will be summarized briefly with further reference to Fig. 1. A first voltage is applied between the drift electrode 112 and the collection electrode 114. A second voltage is applied between the cathode layer 120 and the anode layer 122. Each of the through-holes 124 then behaves like an electric dipole. The electric dipole field is super- posed with the electric field between the drift electrode 112 and the GEM plate 110 and the electric field between the GEM plate 110 and the collection electrode 114. The superposition of these three electric fields leads to a resulting electric field with the field line structure schematically indicated in Fig. 1. The through-holes 124 lead to a strong local concentration of the electric field, and hence to an enhancement of the electric field amplitude in a so-called field tube.

If a primary electron is generated somewhere between the drift electrode 112 and the GEM plate 110, the electron will drift towards the GEM plate 110 due to the electric field E, and will be drawn into one of the through-holes 124 by the dipole field. In the through-hole 124, the electric field amplitude is locally enhanced, and the primary electrons are accelerated to sufficiently large speeds to induce an avalanche multiplication by ionizing further gas molecules within the through-hole 124. The through-holes 124 hence serve as an amplification gap of the detector assembly.

Part of the positive ions created by the impact ionization process are drawn towards the drift electrode 112, while part of the electron cloud resulting from the avalanche process is accelerated in the opposite direction and towards the collection electrode 114, where it will induce a signal on the readout pads 116.

GEM-type detectors have been found to work highly reliably with a number of detector gases and when employed in high-energy physics laboratories, where environmental conditions can be carefully controlled. However, the inventors have found that when operated with some photosensitive vapours, such as tetrakis-dimethylamine-ethylene (TMAE), or when operated in ambient air, strong leakage currents may appear across the dielectric insulator sheet 118 along the walls of the through-holes 124. The leakage currents may lead to spurious pulses and a severe degradation of the detector performance, and they limit the voltage that can be applied across the amplification gap between the cathode layer 120 and the anode layer 122. The leakage currents are particularly pronounced when the detector is operated in ambient air at humidity levels beyond 30 %. These problems have so far limited the applications of avalanche detectors under atmospheric conditions, in photosensitive gases, or in rough environments. What is needed is a more robust detector assembly that can be reliably operated with a large variety of detector gases, including ambient air, and in rough operating environments.

Overview of the Invention

This objective is achieved with an electron multiplier assembly according to independent claim 1. The dependent claims relate to preferred embodiments.

An electron multiplier assembly according to the present invention comprises a first electrode plate, said first electrode plate comprising a first electrode layer and a first plurality of through-holes extending through said first electrode plate, and a second electrode plate, said second electrode plate comprising a second electrode layer and a second plurality of through- holes extending through said second electrode plate, wherein said second plurality of through- holes are aligned with said first plurality of through-holes. According to the invention, a gap extends between said first electrode plate and said second electrode plate, and in a region of said gap said first electrode plate is supported on said second electrode plate only by means of a plurality of spacer elements.

The inventors found that leakage currents in the detector can be sufficiently suppressed by replacing the dielectric insulator sheet 118 of a conventional GEM-type detector as shown in Fig. 1 with a detector structure in which a gap or space extends between the cathode layer and the anode layer, and in which the cathode layer is supported on the anode layer only by means of a plurality of spacer elements. Compared to the conventional detector design^ the through- holes in the electron multiplier assembly of the present invention are wall-less, avoiding the leakage currents that have formed along the walls of the through-holes in conventional detector designs.

Denoting a space between a first through-hole formed in said first electrode plate and a second through-hole aligned with said first through-hole in said second electrode plate an ampli- fication channel in said gap between said first electrode plate and said second electrode plate, according to the invention a plurality of neighboring amplification channels in said gap are not separated by walls. In the absence of walls between neighboring amplification channels, leakage currents between the first electrode layer and the second electrode layer are suppressed. Said spacer elements may comprise individual or isolated spacer elements formed on said first electrode plate and/or said second electrode plate in the pitch between neighboring through- holes. Said spacer elements may also comprise wall portions that extend on said first elec- trode plate and/or said second electrode plate along a distance that spans a plurality of pitch lengths between neighboring through-holes.

Optionally, said spacer elements may also comprise boundary walls or a boundary support extending along part of or the entire length and/or width of the first electrode plate and/or second electrode plate along a boundary of the first electrode plate or second electrode plate, respectively. Such boundary walls may provide additional support of the first electrode plate on the second electrode plate and vice-versa, and may also serve to seal the detection chamber against the environment. However, according to the invention, said first electrode plate may not be continuously or everywhere supported on said second electrode plate.

The number of spacer elements may be at least 10 times smaller, preferably at least 20 times smaller and particularly preferably at least 50 times smaller than the number of through-holes in said first electrode plate or the number of through-holes in said second electrode plate.

The inventors found that said first electrode plate may be supported on said second electrode plate only by relatively few isolated spacer elements without comprising the stability and integrity of the detector assembly. By reducing the number of spacer elements, the leakage cur- rents can be further suppressed.

In a preferred embodiment, at least part of said spacer elements are non-conductive spacer elements. Preferably, said spacer elements comprise a plastic material, a glass material, or a printed circuit board material.

The inventors found that these materials are well suitable for effectively insulating the first electrode layer against the second electrode layer. In a preferred embodiment, at least part of said spacer elements comprise at least one inwardly directed groove formed in a side surface of said spacer element.

A plurality of grooves may be formed in said side surface of said spacer element.

The inventors found that inwardly directed grooves along the side surfaces of the spacer elements provide an effective singularity for the current path in the amplification gap, and may greatly assist in blocking the formation of leakage currents along the sidewalls of the spacer elements.

Preferably, said groove extends into said spacer element in a direction perpendicular to a normal direction of said first electrode layer and/or said second electrode layer, or perpendicular to an axis of the spacer element, or perpendicular to the shortest connection between the first electrode layer and the second electrode layer.

In a preferred embodiment, a depth of said groove amounts to at least 10 % of a height of said spacer element, preferably at least 20 % of a height of said spacer element.

By providing grooves of sufficient depth and appropriated shape, the formation of leakage currents along the spacer walls may be greatly reduced.

In a preferred embodiment, said groove may be a rectangular groove. A rectangular groove, in the context of the present invention, may be a groove with a rectangular or square cross- section.

A width of said groove may likewise amount to at least 10 % of a height of said spacer element, and particularly at least 20 % of a height of said spacer element.

In the context of the present invention, a width of said groove may be a dimension of said groove along a normal direction of said first electrode layer and/or said second electrode layer, or along an axis of said spacer element, or along a shortest direction connecting said first electrode layer to said second electrode layer.

Said groove may circumvent the entire spacer element. In a preferred embodiment, at least part of said spacer elements may comprise a protective coating.

Said protective coating may be applied along the side walls of said spacer elements. The protective coating may protect the spacer elements against gases or humidity that may diffuse into the spacer material. The inventors found a protective coating particularly suitable for spacers formed of fiber glass material The spacers may be coated with Parylene vacuum deposited or by any other hydrophobic coating.

In a preferred embodiment, said first electrode layer and/or said second electrode layer comprises a resistive material.

The inventors found that a resistive first electrode layer and/or second electrode layer allows to better protect the electron multiplier assembly against sparks and discharges, thereby making it more robust and suitable for use in harsh conditions where sparks and discharges are unavoidable.

Preferably, said first electrode plate and/or said second electrode plate comprises a plurality of pad elements, wherein said pad elements each comprise a dielectric material and underlie a spacer element.

The pad elements may serve as dielectric protective pads that electrically insulate the spacer elements against the electrode layers, and hence further reduce the formation of leakage currents along the spacer walls.

Preferably, a diameter of said pad elements is no larger than a pitch between neighboring through-holes in said first electrode plate or second electrode plate, respectively.

Each of the pad elements may underlie a single spacer element, i.e., only one spacer element.

In a preferred embodiment, said first electrode plate comprises a dielectric layer facing said second electrode plate. Alternatively or additionally, said second electrode plate may comprise a dielectric layer facing said first electrode plate. Said dielectric layer may be formed on and in direct contact with said first electrode layer and second electrode layer, respectively. In a preferred embodiment, said dielectric layer extends over the entire length and/or width of said first electrode layer and second electrode layer, respectively.

The inventors found that a dielectric layer allows the reduction of the electric field formed between neighboring through-holes formed in one and the same plate, and hence contributes to a better focusing of the electrical field in the amplification gap between the first electrode plate and the second electrode plate. At the same time, the dielectric layer provides additional mechanical stability and rigidity to the electron multiplier assembly.

In an embodiment of the invention, said first electrode plate may comprise a third electrode layer, wherein said first electrode layer and said third electrode layer are formed on opposite sides of said first dielectric layer. In other words, said first electrode layer and said third electrode layer may sandwich said first dielectric layer.

Alternatively or additionally, said second electrode plate may comprise a fourth electrode layer wherein said second electrode layer and said fourth electrode layer are formed on opposite sides of said second dielectric layer. Hence, said second electrode layer and said fourth electrode layer may sandwich said second dielectric layer.

The third electrode layer and the fourth electrode layer may be layers with some or all of the properties described above for the first electrode layer and second electrode layer, respectively. In particular, the third electrode layer and/or the fourth electrode layer may be resistive layers.

In a preferred embodiment, said first electrode plate and/or said second electrode plate com- prises a plurality of separations or separation elements electrically dividing or partitioning said first electrode plate and second electrode plate, respectively, into a plurality of sub-plates.

A sub-division of the electrode plates allows the variation of the electric field strength over separate portions of the electron multiplier assembly. At the same time, the sub-division al- lows to spatially resolve a detector signal with high accuracy in a configuration in which readout means are integrated into the electron multiplier assembly.

In a preferred embodiment, said first electrode layer and/or second electrode layer may be formed from a plurality of sectoral electrode layers, wherein neighboring sectoral electrode layers are spaced apart so that a separation is formed between them

This structure may be formed using semiconductor fabrication techniques, e.g., by selectively depositing material on an underlying dielectric layer.

In another embodiment, grooves may be formed in said first electrode layer and/or second electrode layer after forming the respective electrode layer. In a further embodiment, said separation elements comprise a plurality of dielectric strips.

Said separations or separation elements may comprise a resistive material whose resistivity is greater than a resistivity of said first electrode layer or second electrode layer, respectively. In a preferred embodiment, said gap between said first electrode plate and said second electrode plate has a constant width.

Preferably, said first electrode plate and said second electrode plate are spaced by a distance that amounts to no more than two times a diameter of said through-holes in said first electrode plate or said second electrode plate, and in particular to no more than 1.5 times said diameter.

Preferably, said first electrode plate and said second electrode plate are spaced by a distance that amounts to at least 0.5 times a diameter of said through-holes in said first electrode plate or said second electrode plate, and in particular to at least 0.7 times said diameter.

The inventors found these parameter ranges particularly suitable for the formation of electron avalanches with high yield. Preferably, a diameter of said through-holes in said first electrode plate and/or said second electrode plate is at least 0.1 mm, and particularly at least 0.3 mm.

In a preferred embodiment, a diameter of said through-holes in said first electrode plate and/or said second electrode plate amounts to no more than 2 mm, and preferably no more than 1 mm.

In a preferred embodiment, a pitch between neighboring through-holes in said first electrode plate and/or said second electrode plate amounts to at least 0.7 mm, and preferably at least 1 mm.

In a preferred embodiment, a pitch between neighboring through-holes in said first electrode plate and or said second electrode plate amounts to at least 1.5 times the diameter of said through-holes.

In a further preferred embodiment, a pitch between neighboring through-holes in said first electrode plate and/or said second electrode plate amounts to no more than 3 times a diameter of said through-holes. In a preferred embodiment, a pitch between neighboring through-holes in said first electrode plate and/or said second electrode plate amounts to no more than 3 mm, and preferably no more than 2 mm.

In a preferred embodiment, said electron multiplier assembly comprises first polarization means coupled to said first electrode layer and adapted to raise said first electrode layer to a first potential, and second polarization means coupled to said second electrode layer and adapted to raise said second electrode layer to a second potential higher than said first potential. In a preferred embodiment, said first electrode plate and/or said second electrode plate comprises a readout layer coupled to said first electrode layer or said second electrode layer, respectively.

Preferably, said readout layer may comprise a plurality of readout strips or readout pads. By incorporating the detector readout into the electron multiplier assembly, a particularly small and compact detector configuration can be achieved. In a preferred embodiment, said readout layer may be integrated into a dielectric layer connected to said first electrode layer or said second electrode layer, respectively.

The invention also relates to a detector comprising a detection chamber, a drift electrode placed in said detection chamber, a collection electrode placed in said detection chamber, wherein said collection electrode is spaced apart from said drift electrode, and at least one electron multiplier assembly with some or all of the features described above, wherein said electron multiplier assembly is placed in said detection chamber between said drift electrode and said collection electrode. In a preferred embodiment, said electron multiplier assembly is an assembly comprising a readout layer connected to said first electrode layer or said second electrode layer, respectively, as described above. Said collection electrode may incorporate said readout layer of said electron multiplier assembly. The collection electrode may be integrated into or may coincide with one of the electrode layers or readout layers of the electron multiplier assembly. In this way, readout of the charges resulting from the electron multiplication may be conveniently incorporated into the electron multiplier assembly. This configuration results in a very compact detector design, which is particularly suitable for applications across industries that require small and lightweight de- tectors.

In a preferred embodiment, the detector comprises a plurality of electron multiplier assemblies stacked in said detection chamber between said drift electrode and said collection electrode, wherein each said electron multiplier assembly is an assembly with some or all of the features described above. Neighboring electron multiplier assemblies may be positioned in said detection chamber such that their through-holes are mutually aligned.

The electron multiplier assembly according to the invention can be provided as a module that allows for easy stacking. By lining up the through-holes and suitably choosing the voltages that are applied to the first and second electrode layers of the stacked modules, a cascaded detector operation can be achieved. The number of electron multiplier assemblies in the stack can be selected in accordance with the requirements of the application. The electron multiplier assembly according to the invention is hence very versatile, can be mass-produced, and can be employed in a variety of applications.

In a preferred embodiment, the detector comprises drift polarization means adapted to raise said drift electrode to a drift potential, first polarization means coupled to said first electrode layer and adapted to raise said first electrode layer to a first potential higher than said drift potential, second polarization means coupled to said second electrode layer and adapted to raise said second electrode layer for a second potential higher than said first potential, and collection polarization means adapted to raise said collection electrode to a collection potential higher than said second potential. In an embodiment of the invention, said detection chamber may comprise a chamber wall adapted to seal said detection chamber against an outer environment.

Said detection chamber may be adapted to be filled with a detection gas, such as a gas adapted to convert incident ultraviolet irradiation or other types of incident irradiation into electrons.

In a preferred embodiment, said detection wall comprises a window which is at least partially transparent for ultraviolet light.

This configuration is particularly advantageous for applications to the detection of ultraviolet light, or to the detection of variations in incident ultraviolet light, such as for fire detectors or smoke detectors.

In another embodiment, said detection chamber is in fluid communication with an outer environment and is filled with ambient air.

It is a particular advantage of the present invention that it allows the avalanche particle detector to be operated in ambient air, and even in humid air. The possibility of operating the particle detector in ambient air permits to make use of the invention in applications relating to the detection of dangerous gases or radioactivity.

In a preferred embodiment, said drift electrode is removable or detachably mounted in said detection chamber.

In particular, said particle detector may comprise mounting means which allow the drift electrode to be removed from said detection chamber, and to be replaced with a different drift electrode.

This configuration is particularly advantageous in the detection of radioactive particles, where radioactive decay products may gather at the drift electrode and may cause persistent detector signals that could interfere with new measurements. By interchanging the drift electrode, these remnants may be removed from the detector, and hence the detector can be quickly re- initialized for new measurements.

As a particular advantage, the detector according to the present invention is insensitive to sunlight, and hence may be operated for the detection of ultraviolet light without an additional filter.

The avalanche particle detector according to the present invention allows various applications ranging from the detection of gases and/or radioactivity to the detection of smoke and/or fire.

In a preferred embodiment, the invention relates to a detector comprising an avalanche parti- cle detector with some or all of the features described above, as well as at least one ultraviolet light source, wherein said ultraviolet light source is adapted to shine ultraviolet light into said detection chamber. The gas detector further comprises signal analysis means coupled to said collection electrode, wherein said analysis means is adapted to read out the collected charge signal from said collection electrode and to infer from said charge signal the presence and/or quantity of said ultraviolet light incident in said detection chamber.

The detector may comprise focusing means for focusing said ultraviolet light into said detection chamber. The focusing means may comprise at least one optical lens. By means of the focusing means, UV light from predefined spatial directions may be selectively collected. The invention also relates to a gas detector comprising an avalanche particle detector with some or all of the features described above, as well as at least one ultraviolet light source, wherein said ultraviolet light source may be positioned externally to said detection chamber and is adapted to shine ultraviolet light into said detection chamber. The gas detector further comprises signal analysis means coupled to said collection electrode, wherein said signal analysis means is adapted to read out a collected charge signal from said collection electrode and to infer from said charge signal a presence and/or a concentration of a predetermined gas present in said detection chamber.

In particular, said detection chamber may be in fluid contact with an outer environment, and said gas may be a gas present in an ambient air for which safety thresholds exist.

Said gas may be a gas that generates electrons when irradiated with ultraviolet light. These electrons may serve as primary electrons that trigger a signal in the avalanche particle detector and allow the detection of a presence and/or a concentration of said gas in said detection chamber.

In another aspect, the invention relates to a smoke detector comprising an avalanche particle detector with some or all of the features described above, and with at least one ultraviolet light source being adapted to shine ultraviolet light into said detection chamber. The smoke detector further comprises signal analysis means coupled to said collection electrode, wherein said signal analysis means are adapted to read out a collected charge signal from said collection electrode and to detect from said charge signal an attenuation of said ultraviolet light.

In this instance, the detection chamber may be filled with a detection gas that generates electrons upon irradiation with ultraviolet light. These electrons may serve as primary electrons for the avalanche particle detector, and the collected charge signal may hence be indicative of variations in the amount of incident ultraviolet light.

Said ultraviolet light source may be positioned externally to said detection chamber. Smoke in an area between an ultraviolet light source and the detection chamber will usually lead to an attenuation of said ultraviolet light, and may hence be detected based on the collected charge signal. The invention further relates to a method for multiplying electrons in an amplification gap, said method comprising the steps of providing an electron multiplier assembly in a detection chamber, said electron multiplier assembly being an assembly with some or all of the features described above, wherein a space between a first through-hole formed in said first electrode plate and a second through-hole aligned with said first through-hole in said second electrode plate defines an amplification channel in said gap. The method further comprises a step of raising said first electrode layer to a first potential, and raising said second electrode layer to a second potential such that an electric field is formed in said amplification channel.

As described above, the electron multiplier assembly according to the present invention may be used both in the detection of gases as well as in the detection of variations of ultraviolet light, such as for detecting flames or smoke.

Hence, in one aspect the invention relates to a method for detecting a gas, comprising the steps of providing an avalanche particle detector with some or all of the features described above, and providing at least one ultraviolet light source and shining ultraviolet light into said detection chamber. The method further comprises the step of collecting a charge signal from said collection electrode, and inferring from said charge signal a presence and/or concentration of a gas in said detection chamber. Said ultraviolet light source may be located externally to said detection chamber.

In another aspect, the invention relates to a method for detecting ultraviolet light, comprising the steps of providing an avalanche particle detector with some or all of the features described above, wherein said detection chamber is filled with a gas sensitive to ultraviolet light. The method further comprises a step of collecting a charge signal from said collection electrode and inferring from said charge signal a presence and/or quantity of said ultraviolet light incident in said detection chamber.

Applications of the detection of ultraviolet light may be in the detection of flames or smoke. In particular, in one aspect the invention relates to a method for detecting smoke, comprising the steps of providing an avalanche particle detector with some or all of the features described above, and providing at least one ultraviolet light source and shining ultraviolet light into said detection chamber. The method further comprises the step of collecting a charge signal from said collection electrode, and inferring from said charge signal an attenuation of said ultraviolet light incident on said detection chamber.

Said light source may be located externally to said detection chamber.

In a preferred embodiment, said light source may be a pulsed light source.

In a further embodiment, the method may comprise a step of inferring from said charge signal a presence and/or quantity of ultraviolet light incident on said detection chamber (52) from a source different from said one or several ultraviolet light sources, in particular different from a pulsed light source.

This configuration allows the detector to detect smoke from a signal attenuation and flames or fire from the emergence of signals from another source. Hence, smoke and flame can be de- tected simultaneously.

The invention provides a compact and robust electron multiplier assembly or avalanche particle detector, respectively, which can be operated reliably with a large number of gases, including ambient air. The device can be made lightweight and small, and is very versatile and particularly suited for applications in harsh conditions. Possible applications outside the field of high-energy physics are numerous, and comprise the detection of radioactivity, the detection of radon for earthquake prediction, the detection of dangerous gases, or the detection of fire or smoke. Detailed description of preferred embodiments

The features and numerous advantages of the present invention will be best understood from a detailed description of preferred embodiments with reference to the accompanying drawings, in which: shows a schematic cross-sectional view of a conventional GEM detector as described above;

is a schematic perspective view of an electron multiplier assembly according to an embodiment of the present invention;

is a schematic cross-sectional view of a spacer element with grooves, as it may be employed in an electron multiplier assembly according to the present invention;

is a schematic perspective view of an electron multiplier assembly according to another embodiment of the present invention, in which the electrode plates comprise resistive layers and dielectric layers;

is a perspective schematic view of an electron multiplier assembly according to another embodiment of the present invention, in which resistive strips are formed in one of the electrode plates;

is a perspective schematic view of an electron multiplier assembly according to another embodiment of the present invention, in which a readout layer is integrated in a dielectric multilayer sheet assembly;

is a schematic cross-sectional view of an avalanche particle detector comprising an electron multiplier assembly according to an embodiment of the present invention;

is a cross-section showing the structure of Fig. 7a together with an electric field configuration that may arise when operating the device according to an embodiment of the present invention;

is a schematic cross-sectional view of an avalanche particle detector according to the present invention with a stacked array of two electron multiplier assemblies;

is a schematic cross-section of the structure of Figure 8a with the electrical field configuration in addition;

is a schematic cross-section of a detector for UV light according to an embodiment of the present invention;

is a schematic drawing illustrating a method for smoke detection that employs an avalanche particle detector according to an embodiment of the present invention; Fig. 10b shows signals that may be collected when operating a combined smoke and flame detector according to an embodiment of the present invention;

Fig. 1 1 is a schematic diagram illustrating a method for detecting dangerous gases according to an embodiment of the present invention;

Fig. 12 is a schematic cross-section of an avalanche particle detector that may be used in a method for detecting dangerous gases according to another embodiment of the present invention; and

Fig. 13 is a schematic cross-section of an avalanche particle detector employed in a method for detecting radioactive gases according to an embodiment of the pre- sent invention.

Structure of an electron multiplier assembly according to an embodiment of the present invention Fig. 2 is a perspective schematic drawing that shows a section of an electron multiplier assembly 10 that may be placed in the detection chamber between a drift electrode (not shown) and a collection electrode (not shown) of a gas electron multiplier detector. For instance, the electron multiplier assembly 10 may replace the conventional GEM plate 110 in the detector configuration shown in Fig. 1, or any other GEM-type detector, such as a Thick-GEM (TGEM) or a Resistive TGEM (RETGEM). As regards the overall detector design and detector readout, reference is hence made to Fig. 1 and the description of the prior art in the patent and research references cited in the Introduction. For simplicity and conciseness, the following description will focus only on the design and operation of the electron multiplier assembly 10 according to embodiments of the present invention.

The electron multiplier assembly 10 comprises a first electrode layer or cathode layer 12 and further comprises a second electrode layer or anode layer 14 spaced apart from said cathode layer 12 such that a gap G is formed between the cathode layer 12 and the anode layer 14. In the configuration shown in Fig. 2, the cathode layer 12 and the anode layer 14 extend in parallel, such that the distance t between these layers is constant. The distance t between the cathode layer 12 and the anode layer 14 may amount to between 0.4 mm and 3 mm in a TGEM or RETGEM design, but may be chosen smaller or larger, depending on the specific application. In the configuration shown in Fig. 2, the cathode layer 12 and the anode layer 14 may be plane layers. However, for other applications it may be more desirable to have a curved or bended cathode layer 12 and/or anode layer 14, as described in further detail in EP 0 948 803 Bl . Fig. 2 only shows a small part of the electrode layers, and hence the depicted structure should be imagined as being extended both along the width direction and the length direction. A plurality of circular through-holes 16 are formed in an array in said cathode layer 12. The through-holes 16 may have a diameter of between 0.3 mm and 1 mm, with a pitch of between 0.7 mm and 7 mm between any two neighboring through-holes 16. The diameter of the through-holes 16 can be chosen approximately equal to the spacing t between the cathode layer 12 and the anode layer 14. The through-holes 16 may be formed in the cathode layer 12 by precision drilling as in conventional TGEM or RETGEM detectors. However, the holes 16 may also be formed in the cathode layer by 12 employing photolithography and etching, depending on the desired size of the holes 16 and the required precision.

As can be taken from Fig. 2, the anode layer 14 underlying the cathode layer 12 comprises a corresponding plurality of through-holes 18 of an identical size and spacing. The cathode layer 12 and the anode layer 14 are positioned in the detection chamber relative to one another such that the through-holes 16 in the first electrode layer 12 are aligned with the through- holes 18 in the second electrode layer 14. Alignment of the through-holes 16, 18 creates a focusing effect for the electric field lines, as described in the introductory part of the specifi- cation with reference to the conventional GEM detectors.

While Fig. 1 shows a configuration with circular through-holes 16, 18, through-holes of different shape such as rectangular or square through-holes may likewise be employed in a detector according to the present invention.

The cathode layer 12 and the anode layer 14 may be metallic layers, such as layers formed from copper or some other conductive material.

However, high-resistivity layers, such as a high-resistive polyimide, electra polymers or car- bonite Kapton, may also be employed as a cathode layer 12 and/or an anode layer 14. The sheet resistivity of the high-resistivity layers may be in the range of between 100 kn and 1 ΜΩ per square. The thickness of the resistive layers 12, 14 may amount to 0,1 mm to 0,3 mm. The inventors found that high-resistivity layers are particularly suitable to protect the detector and the readout electronics from possible sparks and discharges. The resistive layers also limit the metal evaporation that could occur in the vicinity of the holes due to powerful sparks, and might result in leak currents and detector degradation.

The cathode layer 12 is electrically connected to first polarization means 20 adapted to raise the cathode layer 12 to a first potential V,. The anode layer 14 is electrically connected to second polarization means 22 adapted to raise the anode layer 14 to a second potential V ₂ higher than the first potential Vi, so that electrons are drawn into the amplification gap G and from the cathode layer 12 towards the anode layer 14 when the respective voltages V„ V ₂ are applied.

But contrary to a conventional GEM electrode, such as the GEM plate 110 shown in Fig. 1, the electron multiplier assembly 10 according to the present invention does not comprise an insulator sheet or insulator layer between the cathode layer 12 and the anode layer 14. Rather, the cathode layer 12 is supported on the anode layer 14 merely by a plurality of spacer elements 24 that may be formed on the anode layer 14 and the cathode layer 12 at regular intervals. The space between the first through-hole 16 formed in the cathode layer 12 and the second through-hole 18 that is aligned with the first through-hole 16 in the anode layer 14 de- fines an amplification channel in the gap G between the first electrode plate and the second electrode plate. Contrary to conventional GEM-type detectors, the amplification channels are wall-less in the amplification gap G, in the sense that any two neighboring amplification channels are not separated by walls, but merely separated by the detector gas. The spacer elements 24 may be formed from a plastic material or a printed circuit board material, but may also be formed from glass. The number of the spacer elements 24 could be twenty times or even fifty times smaller than the number of through-holes 16, 18, and they could be arranged at regular intervals along the width direction and length direction of the cathode layer 12 and anode layer 14. The spacer elements 24 and protective pads 26 are pref- erably positioned as far away as possible from the through-holes 16, 18 on the cathode 12 and anode layer 14, so to avoid any inference of the spacer elements 24 with the electrical dipole fields formed in the amplification gap G between the cathode layer 12 and the anode layer 14. When fabricating the electron multiplier assembly 10, the cathode layer 12 and the anode layer 14 may be first formed separately, and the through-holes 16, 18 may be drilled or etched. The spacer elements 24 may then be glued to the anode layer 14, and the cathode layer 12 may be glued to the opposite ends of the spacer elements 24 and aligned such that the through-holes 16 in the cathode layer 12 and the through-holes 18 in the anode layer 14 are in one-to-one correspondence and perfect alignment.

The spacers 24 may be cylindrical in shape or rectangular in shape, but any other shape may likewise be employed and may be selected depending on the application and the demands of the fabrication process.

In addition to the spacer elements 24, boundary walls or boundary spacers (not shown) may extend along the entirely or part of the outer circumference of the cathode layer 12 and the anode layer 14 to support the cathode layer 12 on the anode layer 14.

As shown in Fig. 2, dielectric protective pads 26 are further provided on the anode layer 14 so that they underlie the spacer elements 24 at one end thereof. Similar protective pads 26 may be formed on the underside of the cathode layer 12, and may separate the spacer elements 24 from the cathode layer 12 at the opposite end of the spacer elements 24. They are not visible in the perspective view of Fig. 2. The protective pads 26 may be formed from Kapton or Teflon. Any other suitable highly resistive material may likewise be employed, preferably with a sheet resistivity of at least 10 ¹² Ω/square.

The protective pads 26 may extend over an area of the cathode layer 12 and anode layer 14, respectively, that is only slightly larger than a base area of the spacer elements 24. Hence, in the configuration shown in Fig. 2 each pad element 26 underlies one and only one spacer element 24. The protective pads 26 electrically insulate the spacer elements 24 from the cathode layer 12 and anode layer 14, respectively, and hence prevent the formation of leakage currents along the sidewalls of the spacer elements 24.

Fig. 3 shows a cross-section of a circular spacer element 24 according to an embodiment of the present invention. As can be taken from Fig. 3, the spacer element 24 comprises a rectangular groove 28 that extends around the entire circumference of the spacer element approximately halfway along the height of the spacer element 24, i.e., halfway between an upper end 30 of the spacer element 24 (connected to the cathode layer 12) and a lower end 32 of the spacer element 24 (opposite to the upper end 30 and connected to the anode layer 14). The depth of the groove (in a direction perpendicular to the surface of the spacer element 24) may amount to approximately 1/5 or 1/4 of the height of the spacer element, and the width of the groove 28 (along the cylinder axis, i.e., along a direction that provides the shortest connection between the cathode layer 12 and anode layer 14) may have approximately the same dimensions. For instance, assuming a spacer 24 with a height of 1 mm, the groove 28 could be chosen to have a depth and a width of approximately 200 μπι.

The inventors found that the groove 28 provides a singularity that effectively impedes the formation of electric fields and leakage currents along the surface of the spacer elements 24. Fig. 3 shows a rectangular groove 28, but the invention is not so limited and other grooves with other geometrical shapes may be employed as well. The inventors found that rectangular or angular grooves 28 are preferable over grooves with rounded edges, since the former provide a more pronounced singularity for the leakage currents and electric fields that may form along the walls of the spacers 24.

Some materials that may be employed to form the spacer elements 24, such as printed circuit board or G-10, are known to absorb water or vapor. Their resistivity may hence change with time, and this may affect the stability of the detector operation. These effects can be countered by providing a protective coating 34 covering the outer surface of the spacer elements 24, and also extending into the rectangular groove 28. Vacuum-deposited parylene is a suitable coating, but other hydrophobic coatings may likewise be employed.

Fig. 4 is a perspective schematic view of an electron multiplier assembly 10 according to another embodiment of the present invention. The electron multiplier assembly 10 shown in Fig. 4 is generally very similar to the electron multiplier assembly described in detail with reference to Figs. 2 and 3 above, and corresponding parts are denoted with the same reference numerals. A detailed description can hence be omitted, and in the sequel only the distinctions over the embodiment of Fig. 2 will be explained.

The electron multiplier assembly 10 as shown in Fig. 4 differs from the structure shown in Fig. 2 in that the first electrode plate comprises an additional dielectric layer 36 formed on and connected to the underside of the cathode layer 12. Similarly, the second electrode plate comprises a dielectric layer 38 formed on and connected to an upper side of the anode layer 14. The first dielectric layer 36 and the second dielectric layer 38 hence face each other and delimit the amplification gap G between the first electrode plate and the second electrode plate. The spacer elements 24 then extend between the first dielectric layer 36 and the second dielectric layer 38, which also serve to insulate the spacer elements 24 from the underlying cathode layer 12 and anode layer 14, respectively. Additional dielectric protective pads 26 are hence no longer required.

The dielectric layers may be formed from FR4 (flame retardant 4), and may have a thickness in the range of 0,4 mm to 1 mm.

The through-holes 16 in the first electrode plate are formed to extend through both the cathode layer 12 and the first dielectric layer 36. Similarly, the through-holes 18 extend through both the anode layer 14 and the second dielectric layer 38.

The dielectric layers 36, 38 provide additional rigidity and stability to the electron multiplier assembly. In addition, the dielectric layers 36, 38 reduce the electric stray fields that may form between neighboring amplification channels, and hence contribute to a better focusing of the field lines in the amplification gap G.

Fig. 5 shows a perspective schematic view of an electron multiplier assembly 10 according to another embodiment of the invention. The embodiment shown in Fig. 5 is generally similar to the embodiment of Fig. 4, and corresponding components are denoted with like reference numerals. The electron multiplier assembly 10 shown in Fig. 5 only differs from the structure shown in Fig. 4 in that the cathode layer 12 is subdivided or partitioned into a plurality of sub- plates 40a, 40b, 40c by means of separation grooves 42a, 42b. The separation grooves 42a, 42b extend through the first electrode layer 12 down to the underlying first dielectric layer 36. For instance, the separation grooves 42a, 42b might be etched into the resistive cathode layer 12. The resistive grooves 42a, 42b hence electrically insulate the different cathode layer sub- plates 40a, 40b, 40c. Depending on the application, the separation may be used to apply electric fields of different strength to different parts of the electron multiplier assembly 10, or to provide a spatial resolution, such as when a readout structure may be incorporated into the electron multiplier assembly 10. A cathode plate incorporating inner readout strips 44 is schematically shown in the perspective drawing of Fig. 6. The plate of Fig. 6 generally corresponds to one of the plates of the electron multiplier assembly 10 described above with reference to Fig. 5, but readout strips 44 in x-direction and in y-direction (perpendicular to the x-direction) are integrated into the di- electric layer 36. This plate may be employed as a cathode plate or an anode plate of an electron multiplier assembly 10.

Readout electrodes (not shown) are connected to the readout strips 44, and direct the electric charges collected on the readout strips 44 to readout circuitry (not shown) for subsequent analysis. The readout strips 44 may be arranged in a variety of geometries, depending on the application. Semiconductor fabrication techniques such as multi-layer printed circuit technology may be employed to form the readout strips 44.

As will be evident from Figs. 2 to 7, the electron multiplier assembly 10 according to the present invention combines a simple and robust detector design with a highly reliable detection. The electron multiplier assembly 10 according to the invention is small, robust and lightweight.

Operation of the Multiplier Assembly Detector operation of an electron multiplier assembly according to an embodiment of the invention will now be described with reference to Figures 7a and 7b.

The electron multiplier assembly 10 may be operated in the same way as a conventional GEM electrode. Starting from the conventional GEM detector shown in Fig. 1, the GEM plate 1 10 with the dielectric insulator sheet 1 18 may be replaced by the multiplier assembly 10. Hence, the electron multiplier assembly 10 is placed between a drift electrode and a collection electrode.

Figure 7a shows an avalanche particle detector 48 according to the invention with an electron multiplier assembly 10 and a drift electrode 50 spaced apart in a detection chamber 52. In the configuration shown in Figure 7a, the electron multiplier assembly 10 comprises a cathode plate with a cathode layer 12 and a first dielectric layer 36 formed underneath. Through-holes 16 extend through the cathode layer 12 and first dielectric layer 36. The anode plate comprises an anode layer 14 with a second dielectric layer 38 formed on top, so that it faces the first dielectric layer 36. Through-holes 18 extend through anode layer 14 and second dielectric layer 38. First polarization means 20 are provided to raise the cathode layer 12 to a first potential V), and second polarization means 22 are provided to raise the anode layer 14 to a second potential V ₂ higher than the first potential Vi. The voltage V ₂ -Vi may amount to 1 kV in sealed applications, and typically 3-4 kV in ambient air applications.

In the avalanche particle detector 48 shown in Figure 7a, the anode plate may comprise readout strips, as described above with reference to Figure 6. The readout strips may be connected to readout electronics 54 for readout and analysis. In this configuration, a separate collection electrode 114 is not required, which allows to achieve a particularly compact detector design.

However, instead of the electron multiplier assembly according to the embodiment of Figure 6 assemblies according to the embodiments described previously with reference to Figures 2 to 5 may likewise be employed. In this case, the avalanche particle detector 48 comprises an additional collection electrode 114 provided with readout pads 116, which can be positioned in the detection chamber 52 below the electron multiplier assembly 10, i.e., opposite from the drift electrode 50 with respect to the electron multiplier assembly 10. The configuration would then correspond to the configuration described in Figure 1 with reference to the prior art.

The drift electrode 50 corresponds to the drift electrode 112 described with reference to Figure 1 above. Drift polarization means 56 are connected to the drift electrode 50 to raise the drift electrode 50 to a drift potential V _dr lower than the first potential Vi and the second potential V ₂ .

Figure 7b shows the same avalanche particle detector 48 described above with reference to Figure 7a, but in addition illustrates the electric field configuration that arises when electrical potentials Vjr < Vi < V ₂ are applied to the drift electrode 50, cathode layer 12 and anode layer 14, respectively. As can be taken from a comparison of Figure 7b with Figure 1, the field con- figuration largely corresponds to the field configuration in a conventional GEM detector, and again strong dipole fields will form in the through-holes 16, 18.

The detection region 52 may be filled with a detector gas such as a mixture of neon and methane, or with a photosensitive vapor for the detection of incident photons. These vapors con- vert photons into primary electrons. Suitable photosensitive vapors comprise ethyl-ferrocene (EF) and tetrakis-dimethylamine-ethylene (TMAE). However, depending on the application the detection chamber 52 may also be filled with ambient air. Primary electrons generated in the drift zone between the drift electrode and the electron multiplier assembly 10 are drawn towards the cathode layer 12 and into the amplification channels formed between the through- holes 16, 18. Due to the strong electric field gradient in the amplification channels, the primary electrons are accelerated to sufficiently large speeds to induce an avalanche multiplication within the amplification channels. The positive charges created by the impact ionization process in the amplification channels are drawn towards the drift electrode 50, whereas the electron avalanche cloud is accelerated in the opposite direction and collected on the collection electrode for subsequent readout. The readout electronics 54 and data analysis may fully correspond to those in an conventional GEM-type detector.

Since the amplification channels are wall-less, and the spacer elements 24 are located remotely from the through-holes 16, 18 and are well-insulated, the risk of leakage currents forming between the cathode layer 12 and the anode layer 14 is significantly reduced. The inventors found that leakage currents are efficiently suppressed even after days of continuous operation of the detector device, and even when the detector is operated in ambient air at significant humidity levels. The efficient suppression of leakage currents allows to achieve much higher gas gains than conventional GEM detectors and permits a reliable and reproducible detection even in harsh operating environments. Cascaded Mode

Depending on the application, several of the electron multiplier assemblies 10 as shown in Figs. 2, 4, and 5 may be stacked so that the through-holes 16, 18 of neighboring electron multiplier assemblies 10 are all aligned. This allows to operate an electron detector in a cascaded mode and to achieve particularly high amplification gains.

Figure 8a illustrates an avalanche particle detector 48' comprising two electron multiplier assemblies 10, 10' in an aligned configuration and spaced apart in the detection chamber 52. The lower electron multiplier assembly 10 may correspond to the assembly described above with reference to Figures 7a and 7b, and the anode layer 14 may again comprise readout strips coupled to readout electronics 54.

The upper electron multiplier assembly 10' is largely identical to the electron multiplier as- sembly 10 and comprises a first electrode plate or cathode plate with a first cathode layer 12' and a first dielectric layer 36' formed underneath the cathode layer 12'. The second electrode plate or anode plate comprises an anode layer 14' and a second dielectric layer 38' formed on top of the anode layer 14' so to face the first dielectric layer 36'. Through-holes 16', 18' are formed in the first and second electrode plates of electron multiplier assembly 10' so to be aligned with the through-holes 16, 18 formed in the electron multiplier assembly 10 underneath.

Figure 8a shows a configuration with two stacked electron multiplier assemblies 10, 10'. However, as will be understood by one skilled in the art, any number of electron multiplier assemblies may be stacked in the detection chamber 52, depending on the desired amplification gain.

Different from the lowermost electron multiplier assembly 10, the electron multiplier assembly 10' does not comprise readout means, but may be chosen in accordance with the embodi- ment described above with reference to Figure 4. However, in an alternative configuration the electron multiplier assembly 10' may likewise be provided with readout means, either in addition to or instead of the readout means provided at the electron multiplier assembly 10.

During operation, a drift potential V _dr will be applied to the drift electrode 50, and voltages Vi, V ₂ will be applied to the cathode layer 12 and anode layer 14 of the electron multiplier assembly 10, whereas voltages V and V ₂ will be applied to the cathode layer 12' and anode layer 14' of the electron multiplier assembly 10' such that V _dr < V j < V < Vi < V ₂ . The resulting field configuration is schematically illustrated in Figure 8b. Again, strong dipole fields will form in the through-holes 16, 18 and 16' that form the amplification gap.

The distance between neighboring electron multiplier assemblies 10, 10' may amount to 2 mm to 3 mm, preferably measured between an underside of the anode layer 14' and a top side of the cathode layer 12. The voltage difference V ₂ - Vi may amount to 200 V in sealed applications, and 1 kV in ambient air. Various Exemplary Applications

Various applications of the electron multiplier assembly 10 according to the present invention will now be described with reference to Figures 9 to 13.

Figure 9 illustrates an avalanche particle detector 48 employing a two-stage cascade of electron multiplier assemblies 10, 10' as described with reference to Figures 7 and 8 above for the detection of flames or fire. The detector 48 comprises a chamber wall 58 that defines the de- tection chamber 52 in which the drift electrode 50 and electron multiplier assemblies 10, 10' are placed, and seals the detection chamber 52 against the outer environment. As explained above with reference to Figures 7 and 8, the detection chamber 52 may be filled with a UV- sensitive gas such as TMAE, and a UV-transparent window 60 may be provided on one side of the chamber wall 58 so to allow incident ultraviolet irradiation to pass through the UV- transparent window 60 and into the detection chamber 52.

Flames occurring in air are known to emit UV light, and this can be exploited for their detection. A flame in the vicinity of the avalanche particle detector 48 will send UV light through the window 60, and will generate primary electrons in the photosensitive vapor TMAE. The primary electrons will trigger an avalanche of secondary electrons in the cascaded electron multiplier assemblies 10, 10', which will then be detected in the readout electronics 54 and will generate a readout signal indicative of the incident ultraviolet light. The avalanche particle detector 48 hence serves as a flame or fire detector. Due to the high quantum efficiency of TMAE vapors (generally above 30%), flames can be detected with very high sensitivity. In addition, the photo-sensitive vapor may be provided at atmospheric pressure, and hence there is no mechanical constraint on the size of the UV- transparent window 60. Sensitivity increases with the square of the window diameter. The inventors found that even with a window diameter of only about 4 cm, the sensitivity of the detector configuration shown in Figure 9 is up to three orders of magnitude higher than conventional UV flame sensors.

As an additional advantage, the detector configuration according to the present invention is not sensitive to ultraviolet sunlight. Hence, it can be operated without additional filters. To avoid temperature dependence of the sensitivity and the applied voltages, the TMAE vapor may be cooled down to 5°C to 10°C. The inventors found this range to guarantee stability in the temperature interval of between 3°C and 60°C, which is sufficient for all indoor applica- tions.

The same detector structure 48 can also be applied for the detection of smoke when operated in conjunction with one or several pulsed UV sources positioned in the vicinity of the detector device. A method for detecting smoke is schematically illustrated in Figure 10a, and is based on the realization that smoke attenuates UV light so that smoke may be detected based on a decrease in the number of incident UV photons.

As shown in Figure 10a, the detector 48 may be placed in the center of a monitoring area 62, such as an office space or an assembly hall. A plurality of pulsed UV sources 64a, 64b, 64c may be arranged along the boundaries of the smoke monitoring area 62, and may direct pulsed UV light in the direction of the detector 48. The UV light from the pulsed sources 64a to 64c will hence be detected in the avalanche particle detector 48 in the same way as described above in conjunction with the detection of flames, and will provide pulse signals in the readout electronics 54.

Assume now that a pocket of smoke 66 forms in some part of the monitoring area 62, as indicated by the shaded area in Figure 10a. The smoke 66 will attenuate the UV signal from at least one of the pulsed UV sources, such as UV source 64a, and this attenuation will be detected in the detector 48. If this happens, the detector 48 may send a signal, such as a wireless alarm signal 68, to a smoke surveillance unit, and countermeasures may be initiated.

Usually a fire involves both smoke and flames, and the detector 48 can detect both of them due to the different nature of signals associated with these effects. This is illustrated in the schematic diagram of Figure 10b, which shows UV pulses 70 generated by a pulsed UV source 64a in comparison with attenuated UV pulses 72 and signals 74 as they may be produced by open fire. As can be taken from Figure 10b, UV pulses 70 occur at regular intervals, and are typically much larger in amplitude than signals 74 generated by flames. An attenuation due to smoke will usually lead to a decrease in amplitude of the pulsed signals as illustrated for signals 72, but their periodicity will not change. This allows to clearly attribute the regular pulses to the UV sources 64a to 64c, even if the pulses are attenuated due to smoke. A smoke alarm may be triggered if the amplitude of the UV pulses 70 falls below a predetermined threshold. On the other hand, a fire alarm may be triggered if additional UV signals 74 are detected that lack the periodicity of the UV pulses 70. Hence, the detector 48 according to the present invention can detect both smoke and fire with a high degree of reliability.

A similar principle can also be applied for the detection of dangerous gases in air, as will now be described with reference to Figure 11. Again, this may involve the detector structure 48 described above with reference to Figures 7 and 8. The gas detector further comprises a UV source 76, which may be a pulsed UV source as described above with reference to Figure 10a in the context of smoke detection, but may also be a continuous low power UV lamp. The appearance of dangerous gases (as indicated by the arrows in Fig. 11) will change the trans- parency of the air en route from the UV source 76 to the UV-transparent window 60 of the detector 48, which can then again be detected as a decrease in the number of incident photons. The inventors found that UV absorption or transmission measurements can provide a reliable indicator of the appearance of dangerous gases in air, and an alarm may be triggered if the amount of incident photons falls below a predetermined threshold.

Gases that absorb strongly in the UV range and can hence be reliably detected include gasoline, acetone, alcohol, and a large range of other toxic gases.

The applications described above with reference to Figures 9 to 11 rely on sealed detectors. However, as explained above, it is a decisive advantage of the present invention that the electron multiplier assembly may also be operated in air, and this configuration may likewise be employed both in the detection of dangerous gases and in the detection of radioactivity, as will now be described with reference to Figures 12 and 13. Figure 12 illustrates a configuration of the invention for the detection of dangerous gases in air by means of photoionization. The configuration of the detector structure 48 largely corresponds to the detector structure explained above with reference to Figure 9, and comprises a drift electrode 50 and a cascaded set of electron multiplier assemblies 10, 10' in a detection chamber 52. Reference is made to the description of Figures 7a to 9 for a detailed description of the detector structure. However, contrary to the embodiments described above with reference to Figures 9 to 1 1, the detector 48 shown in Figure 12 does not comprise a sealed detection chamber 52 defined by chamber walls 58. Rather, the detection chamber 52 is bounded only by a shielding mesh 78 with a plurality of openings, and is hence in fluid communication with the outer environment. The detection chamber 52 is hence filled with ambient air. Light from an external UV source 76 permeates through the shielding mesh 78 and into the detection chamber 52.

Many dangerous gases, such as benzene vapors, are known to be easily ionized by UV light. When such gases are present in the ambient air, primary electrons will be generated in the detection chamber 52, and will be amplified by the electron multiplier assembly 10, 10' and generate a detectable current in the readout electronics 54. An alarm may be triggered if the detected current exceeds a predetermined threshold. The inventors found that a detector 48 employing an electron multiplier assembly 10 according to the present invention may allow increasing the sensitivity by one to two orders of magnitude over what can be achieved with conventional gas detectors, such as those relying on ionization chambers. Another important application of an ambient air detector configuration according to the present invention lies in the field of detecting radioactive particles, such as radon (Rn). Figure 13 shows a detector 48 that basically corresponds to the detector explained above with reference to Figure 12, with the exception that the distance between the drift electrode 50 and the electron multiplier assembly 10' is increased from about 2 cm to 4 cm or more.

Again, the detection chamber 52 is in fluid contact with the outer environment through a shielding mesh 78 (not shown in Figure 13), and is hence filled with ambient air. The presence of radon in the ambient air may be detected via the appearance of negative ions associated with the radon decay as well as its progenies Po214 and Po218. Alpha particles from the radon decay chain ionize water or oxygen molecules on their path through the detection chamber 52. The negatively charged particles are drawn towards the strong dipole field in the electron multiplier assembly 10'. In the strong electric field, the outer shell electrons are detached, and serve as primary electrons that start an avalanche in the amplification gap of the electron multiplier assemblies 10', 10. The positively charged decay products will accumulate at the drift electrode 50, as schematically indicated in Figure 13. They may in turn produce further decay products, which may generate further signals in the detector device. The inventors found it useful to provide the detector with an exchangeable, removable drift electrode 50, which allows to get rid of the progenies quickly so that a new run of measurements can be started immediately after the drift electrode 50 has been exchanged.

By tuning the drift field between the drift electrode 50 and the electron multiplier assemblies 10, 10', one can selectively detect either Rn only or the Rn ⁺ progeny. In a weak electric field only Rn alpha particles are detected, whereas in stronger fields both Rn and their progeny alpha particles are detected.

The present invention allows to provide a robust and inexpensive Rn detector. Radon is known as a possible earthquake precursor, and hence the invention may be employed in a network of detectors that may be installed in wells and operate in on-line mode for a more reliable earthquake prediction.

Further applications of the invention, both the closed detector configuration and the open air configuration, lie in the detection of charged particles via their UV scintillation spectra. Charged particles, including alpha particles are known to produce UV scintillation light in a number of different media, including air, noble gases, and noble liquids. Since the detector of the present invention is insensitive to visible light, but is very sensitive in the UV region, it can be employed in the detection of charged particles, including alpha particles. Particular advantages which distinguish the invention over conventional designs are a high efficiency in the UV region, and insensitivity to visible light as well as in insensitivity to magnetic fields.

The embodiments described above and the accompanying figures merely served to illustrate the invention and the numerous advantages it entails, but should not be understood to imply any limitation. The scope of the invention is to be determined solely by the appended claims. Reference Signs

10, 10' electron multiplier assembly

12, 12' first electrode layer/ cathode layer

14, 14' second electrode layer/ anode layer

16, 16' through-holes in first electrode layer 12

18, 18' through-holes in second electrode layer 14

20, 20' first polarization means

22, 22' second polarization means

24 spacer elements

26 dielectric protective pads

28 rectangular groove in spacer element 24

30 upper end of spacer element 24

32 lower end of spacer element 24

34 protective coating of spacer element 24

36, 36' first dielectric layer

38, 38' second dielectric layer

40a, 40b, 40c cathode layer sub-plates

42a, 42b resistive strips

44 readout strips

46 multilayer printed circuit board sheet

48, 48' avalanche particle detector

50 drift electrode

52 detection chamber

54 readout electronics

56 drift polarization means

58 chamber wall

60 UV -transparent window

62 smoke monitoring area

64a, 64b, 64c pulsed UV sources

66 pocket of smoke alarm signal

UV pulses

attenuated UV pulses signals generated by flames

UV source

shielding mesh

conventional GEM detector

GEM electrode

drift electrode

collection electrode readout pad

dielectric insulator sheet cathode layer

anode layer

through-hole