EP0242902A2 - Protection device - Google Patents
Protection device Download PDFInfo
- Publication number
- EP0242902A2 EP0242902A2 EP87200556A EP87200556A EP0242902A2 EP 0242902 A2 EP0242902 A2 EP 0242902A2 EP 87200556 A EP87200556 A EP 87200556A EP 87200556 A EP87200556 A EP 87200556A EP 0242902 A2 EP0242902 A2 EP 0242902A2
- Authority
- EP
- European Patent Office
- Prior art keywords
- indium
- electrode
- electrodes
- layer
- switching element
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01C—RESISTORS
- H01C7/00—Non-adjustable resistors formed as one or more layers or coatings; Non-adjustable resistors made from powdered conducting material or powdered semi-conducting material with or without insulating material
- H01C7/10—Non-adjustable resistors formed as one or more layers or coatings; Non-adjustable resistors made from powdered conducting material or powdered semi-conducting material with or without insulating material voltage responsive, i.e. varistors
- H01C7/102—Varistor boundary, e.g. surface layers
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01C—RESISTORS
- H01C7/00—Non-adjustable resistors formed as one or more layers or coatings; Non-adjustable resistors made from powdered conducting material or powdered semi-conducting material with or without insulating material
- H01C7/10—Non-adjustable resistors formed as one or more layers or coatings; Non-adjustable resistors made from powdered conducting material or powdered semi-conducting material with or without insulating material voltage responsive, i.e. varistors
- H01C7/1013—Thin film varistors
Definitions
- This invention relates to circuit protection devices, and especially to devices for protecting electrical circuits against voltage transients that are caused by an electromagnetic pulse, e.g. lightning, and also the transients that are caused by electrostatic discharge.
- chalcogenide glasses by which is meant glasses formed from elements of group VIB of the periodic table (IUPAC l965 revision) together with other elements, especially those of groups IVB and VB, for example as described in U.S. Patent No. 3,27l,59l to Ovshinsky.
- Certain of these glasses can be used to form "threshold” devices by which is meant devices that will change from a high resistance state to a low resistance state on application of a high voltage (the lowest such voltage being referred to as the "threshold voltage”) but which will remain in their low resistance state only for as long as a small "holding" current is maintained.
- chalcogenide glasses can be used to form "memory" devices which will change from a high resistance state to a low resistance state on application of a high voltage and which will remain in the low resistance state, even when no voltage is applied, until an appropriate, different, voltage pulse is applied.
- threshold devices are appropriate for the production of circuit protection devices.
- the chalcogenide glass materials have the advantage that they exhibit very short switching times between their high and low resistance states where the voltage transient that causes switching is significantly higher (e.g. about 50 V or more) than the threshold voltage, typically less than l nanosecond, which is sufficiently fast for protecting circuits from the transient.
- circuit protection devices that are formed from a number of specific germanium/selenium/ arsenic amorphous compositions are described.
- the devices described therein display surprisingly high "energies to latch", that is to say, the devices can withstand surprisingly high electrical energies due to the voltage transients before they latch permanently in their low resistance state.
- one factor in determining the quality of the switch is the contact resistance between the chalcogenide glass material and the electrodes, and that a reduction in the contact resistance can increase the energy to latch of the switch even to the extent that the improvement in the energy to latch caused by a substantial reduction in electrode contact resistance may even effectively cause a composition to change from one that exhibits memory characteristics to one that exhibits threshold characteristics. It is accordingly conjectured that the memory characteristics previously observed by Pinto and Ovshinsky et al could, in fact, be caused by a high electrode contact resistance.
- a threshold switching device which comprises a switching element formed from an amorphous chalcogenide composition and a pair of electrodes that are in contact with the composition, the device containing an indium or indium containing layer in contact with the composition.
- the device exhibits an energy to latch of at least 40mJ, more preferably at least 60mJ and especially at least l00mJ for reasons given in our copending European application No. 0l9689l.
- the electrode contact resistance is sufficiently low that the overall electrode-to-electrode resistance of the device (in its low resistance state) is not more than l0 ohms and especially not more than l ohm, the most preferred resistance being less than 0.l ohm.
- indium or indium containing interlayers Another advantage of the use of indium or indium containing interlayers is that the adhesion of the electrode to the switching element can be considerably improved, to the extent that it is possible reliably to form electrical connections to the electrodes by wire bonding techniques.
- Previously proposed devices have failed on a number of occasions by separation of the electrode from the chalcogenide glass, even after very small electrical transients were applied. The separation, which was often associated with sparking from the electrode, and ultimately complete destruction of the device, can be eliminated or considerably reduced by the use of an interlayer according to the invention.
- the present invention is applicable to chalcogenide glass compositions in general, and especially oxygen-free glasses, especially those containing sulphur and/or selenium optionally and preferably together with group IVB and group VB elements such as germanium, silicon, phosphorous, arsenic and antimony.
- Preferred glasses are those containing germanium, selenium and arsenic, for example those disclosed in our copending British Application described above which contain at least l5 atomic % selenium but preferably not more than 75 atomic % selenium.
- the arsenic content is preferably at least l0 atomic % but preferably not more than 65 atomic %.
- the composition preferably contains at least 5 atomic % germanium but not more than 42 atomic % germanium.
- compositions contain not more than 35 atomic % germanium, more preferably not more than 30 atomic % germanium and especially not more than 25 atomic % germanium. Also, the compositions preferably contain at least 20 atomic % selenium and especially at least 30 atomic % selenium but preferably not more than 70 atomic % selenium and especially not more than 60 atomic % selenium. The compositions preferably contain at least 20 atomic % arsenic and especially at least 25 atomic % arsenic, but preferably not more than 60 atomic % arsenic and especially not more than 55 atomic % arsenic.
- compositions used in the devices according to the invention for example minor amounts of the elements antimony, bismuth, silicon, tin, lead, halogens and some transition metals provided that the presence of such materials does not deleteriously affect the properties, such as the energy to latch and/or off resistivity, to a significant degree. It is preferred, however, for the compositions to contain substantially no tellurium since the presence of tellurium has been found to reduce the off resistivity of the materials severely, although, in certain circumstances, small quantities of tellurium may be tolerated, e.g. up to l0 atomic %, but preferably less than 5 atomic %.
- the films of glass and metals are preferably deposited by vacuum techniques such as vacuum evaporation or d.c. (in the case of metals) or r.f. (in the case of chalcogenide glass) maagnetron sputtering.
- the vapour may be generated by heating an appropriate mixture of the components (not necessarily having the same composition as the intended glass) or the separate components may simultaneously be heated.
- a substrate of suitable electrode design and masked to the desired pattern is coated sequentially with indium (to a thickness of about 0.l to 0.5 micrometre) and then with chalcogenide glass to about l0 micrometres thickness in a vacuum chamber at l0 ⁇ 3 to l0 ⁇ 4 Pa pressure by evaporating the coating materials from separate resistance heated boats.
- a second indium interlayer and a top electrode are deposited onto the glass by a similar process, although in order to deposit a thick metal electrode film in a reasonable time, a higher evaporation rate may be achieved by using an electron beam heated source rather than a resistance heated boat.
- the layers should be deposited without releasing the vacuum except where necessary to change or realign masks and then only when the substrates are cold and by admitting an inert gas such as nitrogen or argon, in order to reduce the possibility of surface oxidation or contamination.
- an inert gas such as nitrogen or argon
- an indium containing chalcogenide glass interlayer of 0.l to 0.5 micrometres in thickness is formed between the chalcogenide glass switching element and the electrodes by a vacuum evaporation process.
- the indium containing glass has the same composition (except for the presence of indium) as that of the switching element, and preferably has an indium content of at least l%, preferably at least 5% but preferably not more than 30%, more preferably not more than 20% and especially not more than l5% indium, based on the total weight of the glass (including indium).
- an indium layer (e.g. 0.l micrometre thick) may be provided on each electrode and an indium containing glass interlayer (e.g. about 0.5 micrometre thick) may be provided between the glass forming the switching element and the indium layer.
- an indium containing glass interlayer e.g. about 0.5 micrometre thick
- this modification is one example of a method of forming a device in which there is a concentration gradient of indium as one penetrates the glass, which varies from about l00% indium adjacent to the electrode to about 0% indium at a depth of 0.3 to l micrometre into the glass.
- a number of glass composition having different indium contents ranging from pure indium down to low may be deposited sequentially onto the electrode and, after deposition of the switching material, may be deposited in reverse order.
- indium and the switching element glass composition may be deposited from different boats simultaneously at the beginning and end of the deposition process. In this case the electrical power provided to heat the two boats is continuously varied so that a high rate of deposition of indium and a low rate of deposition of glass is achieved adjacent to the electrodes, the deposition rate of indium falling and the glass deposition rate rising as the distance from the electrodes increases.
- the entire switching element it is possible for the entire switching element to contain indium.
- the device must include a layer between the switching element and at least one of the electrodes that contains a substantially higher (e.g. at least l0%, especially at least 20 atomic % higher) indium content.
- the indium may evaporate at a faster rate than the other components and so the element may have a higher indium content adjacent to one of the electrodes, in which case it may be appropriate to deposit a layer of indium on top of the deposited glass immediately before depositing the second electrode since that part of the glass layer will be depleted in indium.
- the dimensions of the switching element used in the device according to the invention will depend on the particular chalcogenide glass composition that is used to form it, although the thickness of the switching element will usually be not more than 40 micrometres, preferably not more than 20 micrometres, but usually at least 5 and preferably at least l0 micrometres.
- the cross-sectional area of the switching element, in a plane normal to the direction of current flow therethrough, will depend on the maximum current flow. It is preferably at least 0.5mm2, the preferred size being about lmm2 for discrete devices and for the maximum pulse level, at least 2mm2.
- the devices may be incorporated in an electrical circuit in any suitable position, normally being connected between a current carrying line and earth, (the term "earth” in this context including any structure having an appropriate shape and/or capacity so that it can absorb the charge generated by the transient, and includes for example connection to the chassis and the like in vehicles such as aircraft), and, of course, more than one such device may be employed in the electrical circuit.
- the devices are conveniently incorporated in other electrical components for example electrical connectors, in which case the device will usually be connected between a current carrying element of the device and a terminal or other part of the device to be earthed e.g. a conductive housing.
- the device will revert to its high resistance state as soon as the transient voltage has subsided, it is still possible for the device to be forced into a permanent low resistance state, for example if the current associated with the transient is unduly large or if a number of rapid transients are experienced.
- whether or not the device will become permanently conductive depends on the amount of energy absorbed by the device from the transient. In some applications it may be desirable for the protection device to fail in this way, that is to say, so that the equipment is still protected against transients but will not function until the protection device is replaced or reset.
- the device may be connected in series to means that will exhibit a high resistance to the intended electrical circuit current at least when the switching element has become permanently conductive.
- the switching element may be connected to the current carrying line or to earth via a fuse or switch that is capable of transmitting currents passed through it when the switching element is in its threshold mode but will change to a high resistance state when the switching device has become permanently conductive.
- the device may include a capacitor to ensure that the device exhibits a high resistance to all frequencies below the cut-off limit of the capacitor.
- a capacitor in series with the switching element has a number of advantages as mentioned in our copending British Patent Application No. 8508305 entitled “Overvoltage protection Device: filed on 29th March, l985, the disclosure of which is incorporated herein by reference.
- a capacitor of appropriate size for example from l0pF to 2 microfarads in series with the switching element will enable the transient current to be transmitted to earth, since most of the power of the transient current occurs at frequencies above l0 kHz, but will exhibit a high impedance to the intended currents in the circuit which will have significantly lower frequencies or will be a direct current.
- the use of a capacitor will prevent or significantly reduce the possibility of the switching element latching in its low resistance state after a transient has occurred. Such latching may occur in the absence of a capacitor due to direct current flow through the switching element keeping the switching element in its low resistance state.
- FIG. l of the drawings shows schematically a circuit protection device in accordance with the present invention.
- the device has been formed by depositing 0.l micrometre thick indium layer 20l onto a copper electrode 202 by a vacuum evaporation method, followed by a l0 micrometre thick germanium/arsenic/selenium glass layer 2l0, a further indium layer 203 and a further copper electrode 204, the further layers all being formed by vacuum evaporation from resistance heated boats with the exception of the copper electrode 204 which was formed by electron-beam evaporation.
- Adjacent to the indium layers 20l,203 are two further layers 205,206 which comprise the germanium/arsenic/selenium glass and a small quantity of indium.
- the layers 205,206 may be formed by diffusion of part of the indium from layers 20l,203 into the glass layer 202 or may be formed as a separate layer by any of the methods described above.
- a connection arrangement for connecting two coaxial cables comprises a connector shell l and a male connector 2.
- the male connector 2 comprises a pin 3, the central and rear portion of which is hollow for receiving the central conductor of a coaxial cable to be connected (not shown).
- the pin has a fluxed solder ring 4 and a number of apertures (not shown) beneath the solder ring which communicate between the solder ring 4 and the hollow interior of the pin 3.
- the rear end l0 of the pin is firmly located in a connector housing 5 by means of an electrically insulating plastics spacer 6.
- the housing 5, which provides the electrical connection between the shields of the cables to be connected, has a termination portion 7 on which is mounted a solder impregnated braid 8 and solder ring 9.
- the rear end l0 of the pin is provided on its outer surface with an electrode, e.g. a copper electrode followed by a l0 micrometre thick switch element having an indium/chalcogenide glass/indium construction that has been formed thereon by a vapour deposition method, and the outer surface of the glass element ll has been provided with a further thin (about l0 micrometres thick) electrode, e.g. formed from copper by an electron beam evaporation method.
- the electrode is electrically connected to the housing 5 by means of a column or wire l2 of solder or other suitable conductive material.
- the outer jacket, shield and dielectric are cut back by appropriate amounts and the cable inserted into the connector so that the exposed end of the internal conductor is located within the hollow interior of the pin 3, the dielectric abuts the rear end of the spacer 6 and the exposed shield is located within the solder impregnated braid 8.
- the connector is then briefly heated, for example by means of a hot-air gun, to fuse the solder rings 4 and 9 and to form solder connections between the pin 3 and central conductor and between the braid 8 and coax cable shield.
- the connector will function exactly as a standard coaxial connector until the connected cable experiences a voltage transient whereupon the potential difference across the thickness of the glass layer ll will cause the glass to become electrically conductive and form a closed circuit between the central conductor and the shield.
- the connector comprises a connector housing 2l and a pair of connector wafers 22 and 23 that can be inserted into the housing.
- Each wafer 22,23 has a number of (usually 20 or 40) metallic electrical conductors 24 extending therethrough which terminate at one end either in the form of pins 25 or complementary "tuning fork" female contacts and at the other end in the form of contacts 26 for connection to a flat cable or to a number of small single conductor wires.
- the particular means used for connecting the conductors 24 to the wires or flat cable is not shown but usually comprises one or more solder devices for example as described in U.S. Patent Specification No. 3,852,5l7.
- a stepped recess 27 is made that extends across the width of the entire wafer to expose each of the conductors.
- a copper electrode is deposited onto the individual conductors 24 followed by a 0.l micrometre thick indium layer, a l0 micrometre thick layer 28 of the selenium-germanium-arsenic glass described above, a further 0.l micrometre thick indium layer and a thin, e.g. about l0 micrometres thick, electrode 29 formed e.g. from copper, gold or aluminium on top of the indium layer.
- An additional conductive layer 30 or "ground plane” of gold or aluminium is located on the wafer material in the stepped recess 27, the ground plane being electrically earthed for example to the metallic housing of the connector or to an earth pin.
- Each electrode 29 is connected to the ground plane by means of a wire 3l formed from e.g. gold or aluminium and bonded to the electrode 29 and ground plane 30 by conventional wire bonding techniques.
- a single layer 28 of the glass, indium and electrode 29 may be deposited across the entire width of the wafer (except for a single conductor which is to act as the ground conductor) in which case only a single wire 3l is necessary for connection to the ground plane.
- the ground plane and wire can even be dispensed with if the chalcogenide glass layer 28, and indium and electrode 29 are deposited over all the conductors 24.
- the selenium-germanium-arsenic glass layer, indium layersand electrodes are deposited onto the common ground plane 30, and the wires 3l connect the conductors 24, after any appropriate surface preparation if necessary, with the electrode of the glass layer.
- FIG. 5 shows schematically a further modification of the wafer shown in figures 3 and 4.
- the glass layers 28, indium layers and electrodes 29 are deposited on the conductors 24 as described above and are electrically connected to a ground plane 30 by means of wires 3l.
- a l00 nano-farad capacitor 40 is located in the recess 27 and is connected between the ground plane and an earth terminal or housing of the connector.
- any transient current having a frequency spectrum above about l MHz is conducted directly to earth while any direct currents or alternating currents of frequencies significantly lower than about l MHz are blocked by the capacitor.
- This modification of the device has the advantage that it reduces or eliminates the possibility of the glass switching layers 28 being held in their low resistance state by the direct currents in the electrical system after the transient has been transmitted to earth.
- a Ge14 As34 Se52 glass was formed by mixing the components, which were of at least 99.99% purity, and melting the mixture in a silica ampoule under a vacuum or under reduced argon pressure. During melting, which was carried out at temperatures of up to l000°C and for periods of up to 48 hours, the ampoule was rocked to ensure that a homogeneous melt was obtained.
- a l0 micometre thick layer of the glass so prepared was formed by vapour deposition at a pressure of l0 ⁇ 3 to l0 ⁇ 4 Pa using a resistance heated source. Deposition rates of 0.3 to l.0 micrometres per minute were employed. The layer was provided with a number of electrodes of varying electrode contact resistance, and the energy to latch of the switch devices so formed was determined as described above.
- Example l a gold spring probe of l mm diameter was applied to the top surface of the glass layer with a force of lg.wt., the opposite surface being provided with a copper film electrode.
- Example 2 copper electrode was provided on each surface of the glass layer by a vacuum evaporation method. However the electrodes were allowed partially to oxidize by readmission of air into the vacuum chamber while the electrodes were hot.
- Example 3 a 0.5 micrometre thick layer of the same glass composition but containing l5% indium was provided on each surface of the glass layer and 0.l micrometre thick layer of indium was provided between each indium containing glass and (unoxidized) vacuum deposited copper electrodes. The results are shown in table I.
- Example 4 a 0.2 micrometre thick layer of indium was provided between the chalcogenide glass and each electrode by vacuum evaporation from a resistance heated boat.
- Example 5 the same procedure as Example 4 was used with the exception that the indium layers each had a thickness of 0.05 micrometres.
- Single shot pulses generated by means of a pulser l were passed to the switching element 2 connected in series with a current limiting resistor 3 having a resistance R2 of from 40 to l00 ohms and the voltage across the switching element was observed by means of oscilloscope 4.
- the voltage generated by the pulser l was gradually increased (about 5 to l0 pulses being passed for each voltage level from the pulser) until the switching element latched in its low resistance state (determined by subsequently measuring its resistance.
- E L The energy to latch the device, E L was determined by the equation: where J is the energy available from the pulser; V is the peak voltage from the pulser when latching occurred; R1 is the internal output impedance of the pulser (5 ohms); and R2 is the resistance of the current limiting resistor.
- the resistance across the device was determined by replacing the device with standard resistances and comparing the voltage/time curve with that of the device when subjected to standard pulses. It was observed that the peak current flowing through the device was approximately l ampere in all cases so that the resistance across the device (in ohms) is numerically equal to the voltage across the electrodes (in volts).
Landscapes
- Engineering & Computer Science (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Emergency Protection Circuit Devices (AREA)
- Semiconductor Integrated Circuits (AREA)
- Protection Of Static Devices (AREA)
- Details Of Connecting Devices For Male And Female Coupling (AREA)
Abstract
Description
- This invention relates to circuit protection devices, and especially to devices for protecting electrical circuits against voltage transients that are caused by an electromagnetic pulse, e.g. lightning, and also the transients that are caused by electrostatic discharge.
- One class of material that has been proposed for use in the manufacture of circuit protection devices in general are the chalcogenide glasses, by which is meant glasses formed from elements of group VIB of the periodic table (IUPAC l965 revision) together with other elements, especially those of groups IVB and VB, for example as described in U.S. Patent No. 3,27l,59l to Ovshinsky. Certain of these glasses can be used to form "threshold" devices by which is meant devices that will change from a high resistance state to a low resistance state on application of a high voltage (the lowest such voltage being referred to as the "threshold voltage") but which will remain in their low resistance state only for as long as a small "holding" current is maintained. Other chalcogenide glasses can be used to form "memory" devices which will change from a high resistance state to a low resistance state on application of a high voltage and which will remain in the low resistance state, even when no voltage is applied, until an appropriate, different, voltage pulse is applied. As will be appreciated, only threshold devices are appropriate for the production of circuit protection devices.
- The chalcogenide glass materials have the advantage that they exhibit very short switching times between their high and low resistance states where the voltage transient that causes switching is significantly higher (e.g. about 50 V or more) than the threshold voltage, typically less than l nanosecond, which is sufficiently fast for protecting circuits from the transient.
- In our copending British Application No. 8508304, the entire disclosure of which is incorporated herein by reference, circuit protection devices that are formed from a number of specific germanium/selenium/ arsenic amorphous compositions are described. The devices described therein display surprisingly high "energies to latch", that is to say, the devices can withstand surprisingly high electrical energies due to the voltage transients before they latch permanently in their low resistance state.
- Without being bound by any particular theory, it is believed that one factor in determining the quality of the switch is the contact resistance between the chalcogenide glass material and the electrodes, and that a reduction in the contact resistance can increase the energy to latch of the switch even to the extent that the improvement in the energy to latch caused by a substantial reduction in electrode contact resistance may even effectively cause a composition to change from one that exhibits memory characteristics to one that exhibits threshold characteristics. It is accordingly conjectured that the memory characteristics previously observed by Pinto and Ovshinsky et al could, in fact, be caused by a high electrode contact resistance.
- According to the present invention, there is provided a threshold switching device which comprises a switching element formed from an amorphous chalcogenide composition and a pair of electrodes that are in contact with the composition, the device containing an indium or indium containing layer in contact with the composition.
- By the use of an indium containing layer it is possible in many instances to reduce the electrical contact resistance and thereby to increase the energy to latch of the device to allow optimisation of other properties of the compositions. Preferably the device exhibits an energy to latch of at least 40mJ, more preferably at least 60mJ and especially at least l00mJ for reasons given in our copending European application No. 0l9689l.
- Preferably the electrode contact resistance is sufficiently low that the overall electrode-to-electrode resistance of the device (in its low resistance state) is not more than l0 ohms and especially not more than l ohm, the most preferred resistance being less than 0.l ohm.
- Another advantage of the use of indium or indium containing interlayers is that the adhesion of the electrode to the switching element can be considerably improved, to the extent that it is possible reliably to form electrical connections to the electrodes by wire bonding techniques. Previously proposed devices have failed on a number of occasions by separation of the electrode from the chalcogenide glass, even after very small electrical transients were applied. The separation, which was often associated with sparking from the electrode, and ultimately complete destruction of the device, can be eliminated or considerably reduced by the use of an interlayer according to the invention.
- The present invention is applicable to chalcogenide glass compositions in general, and especially oxygen-free glasses, especially those containing sulphur and/or selenium optionally and preferably together with group IVB and group VB elements such as germanium, silicon, phosphorous, arsenic and antimony. Preferred glasses are those containing germanium, selenium and arsenic, for example those disclosed in our copending British Application described above which contain at least l5 atomic % selenium but preferably not more than 75 atomic % selenium. The arsenic content is preferably at least l0 atomic % but preferably not more than 65 atomic %. In addition or alternatively, the composition preferably contains at least 5 atomic % germanium but not more than 42 atomic % germanium.
- The most preferred compositions contain not more than 35 atomic % germanium, more preferably not more than 30 atomic % germanium and especially not more than 25 atomic % germanium. Also, the compositions preferably contain at least 20 atomic % selenium and especially at least 30 atomic % selenium but preferably not more than 70 atomic % selenium and especially not more than 60 atomic % selenium. The compositions preferably contain at least 20 atomic % arsenic and especially at least 25 atomic % arsenic, but preferably not more than 60 atomic % arsenic and especially not more than 55 atomic % arsenic.
- It is possible for quantities e.g. up to l0% or sometimes more, of other materials to be present in the compositions used in the devices according to the invention, for example minor amounts of the elements antimony, bismuth, silicon, tin, lead, halogens and some transition metals provided that the presence of such materials does not deleteriously affect the properties, such as the energy to latch and/or off resistivity, to a significant degree. It is preferred, however, for the compositions to contain substantially no tellurium since the presence of tellurium has been found to reduce the off resistivity of the materials severely, although, in certain circumstances, small quantities of tellurium may be tolerated, e.g. up to l0 atomic %, but preferably less than 5 atomic %.
- The films of glass and metals are preferably deposited by vacuum techniques such as vacuum evaporation or d.c. (in the case of metals) or r.f. (in the case of chalcogenide glass) maagnetron sputtering. The vapour may be generated by heating an appropriate mixture of the components (not necessarily having the same composition as the intended glass) or the separate components may simultaneously be heated.
- In one preferred method of forming the device, a substrate of suitable electrode design and masked to the desired pattern is coated sequentially with indium (to a thickness of about 0.l to 0.5 micrometre) and then with chalcogenide glass to about l0 micrometres thickness in a vacuum chamber at l0⁻³ to l0⁻⁴ Pa pressure by evaporating the coating materials from separate resistance heated boats. A second indium interlayer and a top electrode are deposited onto the glass by a similar process, although in order to deposit a thick metal electrode film in a reasonable time, a higher evaporation rate may be achieved by using an electron beam heated source rather than a resistance heated boat. In order to achieve good adhesion and a low contact resistance, the layers should be deposited without releasing the vacuum except where necessary to change or realign masks and then only when the substrates are cold and by admitting an inert gas such as nitrogen or argon, in order to reduce the possibility of surface oxidation or contamination.
- In another method, an indium containing chalcogenide glass interlayer of 0.l to 0.5 micrometres in thickness is formed between the chalcogenide glass switching element and the electrodes by a vacuum evaporation process. Preferably the indium containing glass has the same composition (except for the presence of indium) as that of the switching element, and preferably has an indium content of at least l%, preferably at least 5% but preferably not more than 30%, more preferably not more than 20% and especially not more than l5% indium, based on the total weight of the glass (including indium).
- In a modification of this method an indium layer (e.g. 0.l micrometre thick) may be provided on each electrode and an indium containing glass interlayer (e.g. about 0.5 micrometre thick) may be provided between the glass forming the switching element and the indium layer. As will be appreciated, this modification is one example of a method of forming a device in which there is a concentration gradient of indium as one penetrates the glass, which varies from about l00% indium adjacent to the electrode to about 0% indium at a depth of 0.3 to l micrometre into the glass. This may be achieved in the most preferred method in a variety of ways: for example a number of glass composition having different indium contents ranging from pure indium down to low (e.g. 0.5% indium) may be deposited sequentially onto the electrode and, after deposition of the switching material, may be deposited in reverse order. Alternatively, indium and the switching element glass composition may be deposited from different boats simultaneously at the beginning and end of the deposition process. In this case the electrical power provided to heat the two boats is continuously varied so that a high rate of deposition of indium and a low rate of deposition of glass is achieved adjacent to the electrodes, the deposition rate of indium falling and the glass deposition rate rising as the distance from the electrodes increases.
- It is possible for the entire switching element to contain indium. However, in this case the device must include a layer between the switching element and at least one of the electrodes that contains a substantially higher (e.g. at least l0%, especially at least 20 atomic % higher) indium content. In some cases the indium may evaporate at a faster rate than the other components and so the element may have a higher indium content adjacent to one of the electrodes, in which case it may be appropriate to deposit a layer of indium on top of the deposited glass immediately before depositing the second electrode since that part of the glass layer will be depleted in indium.
- The dimensions of the switching element used in the device according to the invention will depend on the particular chalcogenide glass composition that is used to form it, although the thickness of the switching element will usually be not more than 40 micrometres, preferably not more than 20 micrometres, but usually at least 5 and preferably at least l0 micrometres. The cross-sectional area of the switching element, in a plane normal to the direction of current flow therethrough, will depend on the maximum current flow. It is preferably at least 0.5mm², the preferred size being about lmm² for discrete devices and for the maximum pulse level, at least 2mm².
- The devices may be incorporated in an electrical circuit in any suitable position, normally being connected between a current carrying line and earth, (the term "earth" in this context including any structure having an appropriate shape and/or capacity so that it can absorb the charge generated by the transient, and includes for example connection to the chassis and the like in vehicles such as aircraft), and, of course, more than one such device may be employed in the electrical circuit. The devices are conveniently incorporated in other electrical components for example electrical connectors, in which case the device will usually be connected between a current carrying element of the device and a terminal or other part of the device to be earthed e.g. a conductive housing.
- Although in most instances the device will revert to its high resistance state as soon as the transient voltage has subsided, it is still possible for the device to be forced into a permanent low resistance state, for example if the current associated with the transient is unduly large or if a number of rapid transients are experienced. As mentioned above, whether or not the device will become permanently conductive depends on the amount of energy absorbed by the device from the transient. In some applications it may be desirable for the protection device to fail in this way, that is to say, so that the equipment is still protected against transients but will not function until the protection device is replaced or reset. In other applications it may be desirable for the device to fail in a high resistance (open circuit) state so that the equipment will carry on functioning although unprotected from subsequent transients. Thus in some cases the device may be connected in series to means that will exhibit a high resistance to the intended electrical circuit current at least when the switching element has become permanently conductive. Thus, for example, the switching element may be connected to the current carrying line or to earth via a fuse or switch that is capable of transmitting currents passed through it when the switching element is in its threshold mode but will change to a high resistance state when the switching device has become permanently conductive.
- Alternatively or in addition, the device may include a capacitor to ensure that the device exhibits a high resistance to all frequencies below the cut-off limit of the capacitor. The use of a capacitor in series with the switching element has a number of advantages as mentioned in our copending British Patent Application No. 8508305 entitled "Overvoltage protection Device: filed on 29th March, l985, the disclosure of which is incorporated herein by reference. Briefly, the use of a capacitor of appropriate size, for example from l0pF to 2 microfarads in series with the switching element will enable the transient current to be transmitted to earth, since most of the power of the transient current occurs at frequencies above l0 kHz, but will exhibit a high impedance to the intended currents in the circuit which will have significantly lower frequencies or will be a direct current. Also, the use of a capacitor will prevent or significantly reduce the possibility of the switching element latching in its low resistance state after a transient has occurred. Such latching may occur in the absence of a capacitor due to direct current flow through the switching element keeping the switching element in its low resistance state.
- A device in accordance with the present invention and articles that incorporate the device will now be described by way of example with reference to the accompanying drawings, in which:
- Figure l is a sectional view of a device in accordance with the invention with the thickness grossly exaggerated for the sake of clarity;
- Figure 2 is a side view, partly in section, of a BNC coaxial connector that incorporates a circuit pro tection device according to the invention;
- Figure 3 is a side view, partly in section, of a flat cable mass termination connector and wafers that incorporate a circuit protection device according to the invention;
- Figure 4 is an enlarged view of part of the connector shown in figure 3; and
- Figure 5 is a perspective view of a modification of the wafers shown in figure 3;
- Figure l of the drawings shows schematically a circuit protection device in accordance with the present invention. The device has been formed by depositing 0.l micrometre thick indium layer 20l onto a
copper electrode 202 by a vacuum evaporation method, followed by a l0 micrometre thick germanium/arsenic/selenium glass layer 2l0, afurther indium layer 203 and afurther copper electrode 204, the further layers all being formed by vacuum evaporation from resistance heated boats with the exception of thecopper electrode 204 which was formed by electron-beam evaporation. Adjacent to theindium layers 20l,203 are two further layers 205,206 which comprise the germanium/arsenic/selenium glass and a small quantity of indium. The layers 205,206 may be formed by diffusion of part of the indium fromlayers 20l,203 into theglass layer 202 or may be formed as a separate layer by any of the methods described above. - Referring to figure 2 of the accompanying drawings, a connection arrangement for connecting two coaxial cables comprises a connector shell l and a
male connector 2. Themale connector 2 comprises apin 3, the central and rear portion of which is hollow for receiving the central conductor of a coaxial cable to be connected (not shown). The pin has a fluxedsolder ring 4 and a number of apertures (not shown) beneath the solder ring which communicate between thesolder ring 4 and the hollow interior of thepin 3. The rear end l0 of the pin is firmly located in aconnector housing 5 by means of an electrically insulatingplastics spacer 6. Thehousing 5, which provides the electrical connection between the shields of the cables to be connected, has atermination portion 7 on which is mounted a solder impregnated braid 8 and solder ring 9. - The rear end l0 of the pin is provided on its outer surface with an electrode, e.g. a copper electrode followed by a l0 micrometre thick switch element having an indium/chalcogenide glass/indium construction that has been formed thereon by a vapour deposition method, and the outer surface of the glass element ll has been provided with a further thin (about l0 micrometres thick) electrode, e.g. formed from copper by an electron beam evaporation method. The electrode is electrically connected to the
housing 5 by means of a column or wire l2 of solder or other suitable conductive material. - In order to connect a coaxial cable to the connector, the outer jacket, shield and dielectric are cut back by appropriate amounts and the cable inserted into the connector so that the exposed end of the internal conductor is located within the hollow interior of the
pin 3, the dielectric abuts the rear end of thespacer 6 and the exposed shield is located within the solder impregnated braid 8. The connector is then briefly heated, for example by means of a hot-air gun, to fuse the solder rings 4 and 9 and to form solder connections between thepin 3 and central conductor and between the braid 8 and coax cable shield. - The connector will function exactly as a standard coaxial connector until the connected cable experiences a voltage transient whereupon the potential difference across the thickness of the glass layer ll will cause the glass to become electrically conductive and form a closed circuit between the central conductor and the shield.
- Referring to figures 3 and 4, a mass termination connector such as that described in British Patent No. l,522,485 (the disclosure of which is incorporated herein by reference) is schematically shown.
- The connector comprises a connector housing 2l and a pair of
connector wafers wafer electrical conductors 24 extending therethrough which terminate at one end either in the form ofpins 25 or complementary "tuning fork" female contacts and at the other end in the form ofcontacts 26 for connection to a flat cable or to a number of small single conductor wires. The particular means used for connecting theconductors 24 to the wires or flat cable is not shown but usually comprises one or more solder devices for example as described in U.S. Patent Specification No. 3,852,5l7. - In each of the
wafers 22 and 23 a steppedrecess 27 is made that extends across the width of the entire wafer to expose each of the conductors. in one embodiment of this connector, a copper electrode is deposited onto theindividual conductors 24 followed by a 0.l micrometre thick indium layer, a l0 micrometrethick layer 28 of the selenium-germanium-arsenic glass described above, a further 0.l micrometre thick indium layer and a thin, e.g. about l0 micrometres thick,electrode 29 formed e.g. from copper, gold or aluminium on top of the indium layer. An additionalconductive layer 30 or "ground plane" of gold or aluminium is located on the wafer material in the steppedrecess 27, the ground plane being electrically earthed for example to the metallic housing of the connector or to an earth pin. Eachelectrode 29 is connected to the ground plane by means of a wire 3l formed from e.g. gold or aluminium and bonded to theelectrode 29 andground plane 30 by conventional wire bonding techniques. - Alternatively, a
single layer 28 of the glass, indium andelectrode 29 may be deposited across the entire width of the wafer (except for a single conductor which is to act as the ground conductor) in which case only a single wire 3l is necessary for connection to the ground plane. Alternatively the ground plane and wire can even be dispensed with if thechalcogenide glass layer 28, and indium andelectrode 29 are deposited over all theconductors 24. - In an alternative construction, the selenium-germanium-arsenic glass layer, indium layersand electrodes are deposited onto the
common ground plane 30, and the wires 3l connect theconductors 24, after any appropriate surface preparation if necessary, with the electrode of the glass layer. - Figure 5 shows schematically a further modification of the wafer shown in figures 3 and 4. In this form of wafer the glass layers 28, indium layers and
electrodes 29 are deposited on theconductors 24 as described above and are electrically connected to aground plane 30 by means of wires 3l. In addition, however, a l00 nano-farad capacitor 40 is located in therecess 27 and is connected between the ground plane and an earth terminal or housing of the connector. In this form of device any transient current having a frequency spectrum above about l MHz is conducted directly to earth while any direct currents or alternating currents of frequencies significantly lower than about l MHz are blocked by the capacitor. This modification of the device has the advantage that it reduces or eliminates the possibility of the glass switching layers 28 being held in their low resistance state by the direct currents in the electrical system after the transient has been transmitted to earth. - A Ge₁₄ As₃₄ Se₅₂ glass was formed by mixing the components, which were of at least 99.99% purity, and melting the mixture in a silica ampoule under a vacuum or under reduced argon pressure. During melting, which was carried out at temperatures of up to l000°C and for periods of up to 48 hours, the ampoule was rocked to ensure that a homogeneous melt was obtained.
- A l0 micometre thick layer of the glass so prepared was formed by vapour deposition at a pressure of l0⁻³ to l0⁻⁴ Pa using a resistance heated source. Deposition rates of 0.3 to l.0 micrometres per minute were employed. The layer was provided with a number of electrodes of varying electrode contact resistance, and the energy to latch of the switch devices so formed was determined as described above.
- In Example l a gold spring probe of l mm diameter was applied to the top surface of the glass layer with a force of lg.wt., the opposite surface being provided with a copper film electrode.
- In Example 2 copper electrode was provided on each surface of the glass layer by a vacuum evaporation method. However the electrodes were allowed partially to oxidize by readmission of air into the vacuum chamber while the electrodes were hot.
- In Example 3 a 0.5 micrometre thick layer of the same glass composition but containing l5% indium was provided on each surface of the glass layer and 0.l micrometre thick layer of indium was provided between each indium containing glass and (unoxidized) vacuum deposited copper electrodes. The results are shown in table I.
- In Example 4 a 0.2 micrometre thick layer of indium was provided between the chalcogenide glass and each electrode by vacuum evaporation from a resistance heated boat. In Example 5 the same procedure as Example 4 was used with the exception that the indium layers each had a thickness of 0.05 micrometres.
- The energy to latch of the device, and hence of the glass material, was measured by means of a circuit as shown in figure 6. Single shot pulses generated by means of a pulser l were passed to the
switching element 2 connected in series with a current limitingresistor 3 having a resistance R₂ of from 40 to l00 ohms and the voltage across the switching element was observed by means ofoscilloscope 4. The voltage generated by the pulser l was gradually increased (about 5 to l0 pulses being passed for each voltage level from the pulser) until the switching element latched in its low resistance state (determined by subsequently measuring its resistance. -
- This equation gives very good agreement with values obtained by integrating the current and voltage curves of the pulses.
- The resistance across the device was determined by replacing the device with standard resistances and comparing the voltage/time curve with that of the device when subjected to standard pulses. It was observed that the peak current flowing through the device was approximately l ampere in all cases so that the resistance across the device (in ohms) is numerically equal to the voltage across the electrodes (in volts).
Claims (9)
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
ES553479 | 1986-03-26 | ||
ES553479A ES8900238A1 (en) | 1985-03-29 | 1986-03-26 | Circuit protection device |
GB868607719A GB8607719D0 (en) | 1986-03-27 | 1986-03-27 | Circuit protection device |
GB8607719 | 1986-03-27 |
Publications (2)
Publication Number | Publication Date |
---|---|
EP0242902A2 true EP0242902A2 (en) | 1987-10-28 |
EP0242902A3 EP0242902A3 (en) | 1988-08-31 |
Family
ID=26156147
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP87200556A Withdrawn EP0242902A3 (en) | 1986-03-26 | 1987-03-24 | Protection device |
Country Status (1)
Country | Link |
---|---|
EP (1) | EP0242902A3 (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5663861A (en) * | 1995-06-07 | 1997-09-02 | Littelfuse, Inc. | Resettable automotive circuit protection device |
US5682130A (en) * | 1995-03-22 | 1997-10-28 | Styrna; Michael | Circuit protection device with female terminals and PTC element |
EP0864204A1 (en) * | 1995-11-28 | 1998-09-16 | Energy Conversion Devices, Inc. | Integrated drivers for flat panel displays employing chalcogenide logic elements |
US5945903A (en) * | 1995-06-07 | 1999-08-31 | Littelfuse, Inc. | Resettable automotive circuit protection device with female terminals and PTC element |
WO2004081617A3 (en) * | 2003-03-12 | 2005-09-29 | Micron Technology Inc | Chalcogenide glass constant current device, and its method of fabrication and operation |
CN102751319A (en) * | 2012-07-04 | 2012-10-24 | 中国科学院上海微系统与信息技术研究所 | Chalcogenide compound-based surge protection device and preparation method thereof |
CN109473411A (en) * | 2018-09-17 | 2019-03-15 | 上海音特电子有限公司 | A kind of thin-film material and application method for integrated circuit input output pin overvoltage protection |
CN110729399A (en) * | 2018-07-17 | 2020-01-24 | 旺宏电子股份有限公司 | Switching element and memory element |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3271591A (en) * | 1963-09-20 | 1966-09-06 | Energy Conversion Devices Inc | Symmetrical current controlling device |
DE1437387A1 (en) * | 1964-07-03 | 1968-11-28 | Danfoss As | Overvoltage protection device on telephone network lines |
DE1614272A1 (en) * | 1966-08-17 | 1970-02-26 | Philips Nv | A semiconductor device comprising at least one on an A?, B? attached contact and method of making such a device |
US3852517A (en) * | 1972-06-12 | 1974-12-03 | Raychem Corp | Conductive insert for heat recoverable electrical connector |
US3983076A (en) * | 1973-07-02 | 1976-09-28 | Energy Conversion Devices, Inc. | N-type amorphous semiconductor materials |
US3993394A (en) * | 1974-07-31 | 1976-11-23 | Raychem Corporation | Connector half having connector wafer retained therein |
EP0198624A1 (en) * | 1985-03-29 | 1986-10-22 | Raychem Limited | Overvoltage protection device |
-
1987
- 1987-03-24 EP EP87200556A patent/EP0242902A3/en not_active Withdrawn
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3271591A (en) * | 1963-09-20 | 1966-09-06 | Energy Conversion Devices Inc | Symmetrical current controlling device |
DE1437387A1 (en) * | 1964-07-03 | 1968-11-28 | Danfoss As | Overvoltage protection device on telephone network lines |
DE1614272A1 (en) * | 1966-08-17 | 1970-02-26 | Philips Nv | A semiconductor device comprising at least one on an A?, B? attached contact and method of making such a device |
US3852517A (en) * | 1972-06-12 | 1974-12-03 | Raychem Corp | Conductive insert for heat recoverable electrical connector |
US3983076A (en) * | 1973-07-02 | 1976-09-28 | Energy Conversion Devices, Inc. | N-type amorphous semiconductor materials |
US3993394A (en) * | 1974-07-31 | 1976-11-23 | Raychem Corporation | Connector half having connector wafer retained therein |
EP0198624A1 (en) * | 1985-03-29 | 1986-10-22 | Raychem Limited | Overvoltage protection device |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5682130A (en) * | 1995-03-22 | 1997-10-28 | Styrna; Michael | Circuit protection device with female terminals and PTC element |
US5663861A (en) * | 1995-06-07 | 1997-09-02 | Littelfuse, Inc. | Resettable automotive circuit protection device |
US5945903A (en) * | 1995-06-07 | 1999-08-31 | Littelfuse, Inc. | Resettable automotive circuit protection device with female terminals and PTC element |
EP0864204A1 (en) * | 1995-11-28 | 1998-09-16 | Energy Conversion Devices, Inc. | Integrated drivers for flat panel displays employing chalcogenide logic elements |
EP0864204B1 (en) * | 1995-11-28 | 2003-09-24 | Energy Conversion Devices, Inc. | Integrated drivers for flat panel displays employing chalcogenide logic elements |
KR100766678B1 (en) * | 2003-03-12 | 2007-10-15 | 마이크론 테크놀로지, 인크 | Chalcogenide glass constant current device, and its method of fabrication and operation |
WO2004081617A3 (en) * | 2003-03-12 | 2005-09-29 | Micron Technology Inc | Chalcogenide glass constant current device, and its method of fabrication and operation |
US7315465B2 (en) | 2003-03-12 | 2008-01-01 | Micro Technology, Inc. | Methods of operating and forming chalcogenide glass constant current devices |
US7542319B2 (en) | 2003-03-12 | 2009-06-02 | Micron Technology, Inc. | Chalcogenide glass constant current device, and its method of fabrication and operation |
CN101807666A (en) * | 2003-03-12 | 2010-08-18 | 微米技术有限公司 | Chalcogenide constant current device and manufacture method thereof and action |
CN102751319A (en) * | 2012-07-04 | 2012-10-24 | 中国科学院上海微系统与信息技术研究所 | Chalcogenide compound-based surge protection device and preparation method thereof |
CN102751319B (en) * | 2012-07-04 | 2015-04-15 | 中国科学院上海微系统与信息技术研究所 | Chalcogenide compound-based surge protection device and preparation method thereof |
CN110729399A (en) * | 2018-07-17 | 2020-01-24 | 旺宏电子股份有限公司 | Switching element and memory element |
CN110729399B (en) * | 2018-07-17 | 2023-08-25 | 旺宏电子股份有限公司 | Switching element and memory element |
CN109473411A (en) * | 2018-09-17 | 2019-03-15 | 上海音特电子有限公司 | A kind of thin-film material and application method for integrated circuit input output pin overvoltage protection |
Also Published As
Publication number | Publication date |
---|---|
EP0242902A3 (en) | 1988-08-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US4924340A (en) | Circuit protection device | |
US4928199A (en) | Circuit protection device | |
US4952173A (en) | Circuit protection device | |
EP0514454B1 (en) | Device for motor and short-circuit protection | |
US5213517A (en) | Separable electrodes with electric arc quenching means | |
JP4397972B2 (en) | Transient voltage protection device and manufacturing method thereof | |
EP0264315B1 (en) | Wave propagation structures for the suppression of over-voltages and the absorption of transitory waves | |
JPS63100702A (en) | Material and method for avoiding excessive electrical stress | |
US5151384A (en) | Amorphous silicon switch with forming current controlled by contact region | |
US4807081A (en) | Circuit protection arrangement | |
EP0198624A1 (en) | Overvoltage protection device | |
EP0242902A2 (en) | Protection device | |
DE102009017518A1 (en) | Circuit protection device including resistor and fuse element | |
US10134556B2 (en) | Composite fuse element and method of making | |
EP0196891B1 (en) | Circuit protection device | |
CN103081262A (en) | Overvoltage protection component, and overvoltage protection material for overvoltage protection component | |
US6838972B1 (en) | PTC circuit protection devices | |
US4887182A (en) | Circuit protection device | |
WO1990000826A1 (en) | Circuit protection arrangement | |
US4860155A (en) | Overvoltage protection device | |
US20060066435A1 (en) | Composite fuse element and methods of making same | |
US4890182A (en) | Circuit protection device | |
Zhang et al. | Characteristic voltage of programmed metal-to-metal antifuses | |
JPS62249403A (en) | Limit switching device | |
EP0231310B1 (en) | Prefabricated electric connection device to link on a set of bars |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 19870324 |
|
AK | Designated contracting states |
Kind code of ref document: A2 Designated state(s): AT BE CH DE ES FR GB IT LI NL SE |
|
PUAL | Search report despatched |
Free format text: ORIGINAL CODE: 0009013 |
|
RHK1 | Main classification (correction) |
Ipc: H01C 7/10 |
|
AK | Designated contracting states |
Kind code of ref document: A3 Designated state(s): AT BE CH DE ES FR GB IT LI NL SE |
|
17Q | First examination report despatched |
Effective date: 19900806 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 19901218 |
|
RIN1 | Information on inventor provided before grant (corrected) |
Inventor name: PORTER, CHRISTOPHER JOHN |