PatentDe  


Dokumentenidentifikation EP0610278 07.05.1998
EP-Veröffentlichungsnummer 0610278
Titel RESISTIVES SPEICHERELEMENT
Anmelder The University Court of the University of Edinburgh, Edinburgh, GB
Erfinder ROSE, Mervyn, John Kingsbridge Eskill, Angus DD8 3TP, GB;
HAJTO, Janos, Edinburgh EH16 6JS, GB;
OWEN, Alan, Ernest, Edinburgh, EH10 5ER, GB;
OSBORNE, Ian, Stewart 49 Burnside Street, Dundee DD8 3TP, GB;
SNELL, Anthony, James 10A Eskmill Road, Midlothian EH26 8PA, GB;
LECOMBER, Peter, George, +, GB
Vertreter derzeit kein Vertreter bestellt
DE-Aktenzeichen 69224993
Vertragsstaaten DE, FR, GB, IT, NL, SE
Sprache des Dokument En
EP-Anmeldetag 20.10.1992
EP-Aktenzeichen 929216802
WO-Anmeldetag 20.10.1992
PCT-Aktenzeichen GB9201929
WO-Veröffentlichungsnummer 9308575
WO-Veröffentlichungsdatum 29.04.1993
EP-Offenlegungsdatum 17.08.1994
EP date of grant 01.04.1998
Veröffentlichungstag im Patentblatt 07.05.1998
IPC-Hauptklasse G11C 27/00
IPC-Nebenklasse G06G 7/60   

Beschreibung[en]

This invention relates to resistive memory elements suitable for use as synaptic elements in neural networks.

Resistive memory elements usually take the form of a resistive layer which is located between two electrodes, e.g. metallic electrodes, which give access to the element. The resistive layer is usually formed of a host material, e.g. amorphous silicon, which contains a dopant to improve its host properties. In addition, the memory element includes an active species which plays a substantial part in the conduction of an electric current and adjusts the resistance of the resistive layer. The active species is usually a metal dispersed in the host as small particles or as atoms.

(Synaptic elements are usually configured as arrays in which many elements, e.g. 103 - 108, are supported on a single substrate. The substrate provides mechanical support for all of the elements, but the substrate has little or no effect upon the performance of the elements.)

Each element may have its own separate electrodes, so that arrays can be connected into networks as desired. It is also possible that any one element may share one or both of its electrodes with other elements. This means that the elements are permanently connected into particular network patterns. Arrays of elements are conveniently manufactured by depositing the various layers on a suitable substrate. The layers may be patterned by using well established techniques such as masking and etching.

It has been mentioned above that the resistive layer is usually formed of a host material which includes an active species. The elements V, Co, Ni, Pd, Fe and Mn are particularly suitable as the active species for use with amorphous silicon as the host material. The active species is conveniently introduced into a memory element by making one electrode out cf the active species and causing it to migrate into the host. The process of migration is "forming" and it comprises applying a high voltage, e.g. of the order 8 - 35 volts, for a short time, e.g. 200 - 1,000 ns. The forming process reduces the resistance of the element from a very high value (in practice an infinite value) to a value which is suitable for use in a neural network, e.g. to a value within the range 103 - 106 Ω.

It is emphasised that forming introduces the active species into the host material to give an initial value to the resistance. After forming, the resistance can be adjusted to a target value, and it is possible for the resistance to be adjusted and readjusted many times during the life of an elememt The resistance of the element is used as the memory parameter. Resistive elements are used as the programmable memories in neural networks.The circuitry of a neural network addresses a particular element by applying an address voltage to the element, whereby the resistance of the element produces a particular current.

It is important to recognise that each element has two threshold voltages. The lower threshold voltage represents the upper limit for the address voltage. In other words the element is addressed by applying a voltage less than the lower threshold voltage to it. The upper threshold voltage constitutes the value at which the forming process may occur. Thus a voltage in excess of the upper threshold voltage is normally applied to each element in order to form it and, thereafter it is undesirable to apply voltages above the upper threshold value. Voltages which lie between the upper and lower threshold values are conveniently called "programming" voltages because voltages in this range are utilised to adjust the resistance of the element. Since the resistance of the element constitutes the memory parameter, voltages between the upper and lower threshold values are utilised for a process equivalent to the storage of data. (The data stored in its synaptic elements is an important part of the program of a neural network.)

ELECTRONIC LETTERS (Vol. 24, No. 19 15th September 1988, at pages 1231 and 1232) which is used to delimit claims 1 and 2 discloses MOSFET devices with a floating polysilicon gate. The programming of the devices with a series of pulses each having a duration of 1µs and 15V is also disclosed. The MOSFET devices can be used as memory elements in neural-type networks.

European patent specification 0 382 230 82 relates to information processing systems comprising a plurality of input terminals and a plurality of output terminals. The systems also include data transfer members and optical control means. The data transfer members may be variable resistance elements such as field effect transistors.

It has been proposed to adjust the resistance of a synaptic element by applying thereto a voltage the value of which is selected so as to produce the desired result. It has been found that this method is difficult to operate because it is difficult to ascertain the value required to achieve the desired result. It is an object of this invention to facilitate the adjustment of the resistance of a memory element to a target value.

This invention is based on the unexpected discovery that the resistance of an element can be adjusted to a target value by varying the time for which the voltage is applied (instead of varying the magnitude). Thus, the resistance of an element can be adjusted to a target value by selecting a suitable voltage and applying a pulse of an appropriate duration. Preferably, the duration of the pulse is between 10 - 1,000 ns. The polarity of the pulse is important in that one polarity causes a decrease in the value of the resistance, and the opposite polarity causes an increase in the value of the resistance.

The invention is defined in the claims.

It is possible to establish correlations between pulse duration and resistance values. Using such a correlation, a target value of resistance can be achieved by applying a pulse of the appropriate duration and polarity. However, the variability of the forming step makes it difficult to establish an accurate correlation which has wide, general applicability. This invention includes a preferred embodiment which allows pulse widths to be used for adjusting the resistance of a memory element, even when no accurate correlation is available.

The preferred embodiment of the invention is based upon the discovery that, when several pulses of different duration are applied to a memory element, the result is determined only by the longest pulse. Thus, when shorter pulses are applied after a longer pulse, the subsequent, shorter pulses produce no noticeable effect.

This observation gives rise to a preferred method.

  • (a) Depending on whether the resistance of the element is too high or too low, the appropriate polarity is selected.
  • (b) A series of pulses is applied. It is convenient to measure the resistance after each pulse. The series includes later pulses which have longer durations than earlier pulses. Preferably the duration of each pulse is greater than the duration of its predecessor.
  • (c) When the resistance of the element approximates or passes the target value, the series is terminated.

At the completion of step (c) the resistance may, fortuitously, be close enough to the target value to regard the adjustment as complete. Alternatively, the process will overshoot.

The overshoot cannot be corrected by continuing with the same polarity, but the following procedure enables a further adjustment to the value of the resistance to be made. The further procedure comprises repeating the (a), (b) and (c) given above, but selecting the reverse polarity in step (a). In addition smaller increments in pulse duration are preferably used during step (b) whereby a smaller overshoot, if any, results. This procedure can be repeated as often as desired using reverse polarity and smaller increments in pulse duration at each repetition. For example, in a first series of pulses, the pulse duration may be increased by a constant increment of 100ns, e.g. pulses of 100ns, 200ns, 300ns and so on are applied until the series overshoots the target value.

In the next series the reverse polarity is used and the pulse durations are increased by increments of 10ns, e.g. pulse durations of 10ns, 20ns, 30ns and so on would be appropriate. It will be appreciated that further sequences may be utilised, each being of opposite polarity to its predecessor and having a shorter pulse increment than its predecessor. It is suggested that, for most cases, increments of 10ns will produce the target value as closely as desired.

The invention will now be described by way of example with reference to the accompanying drawing which shows a single synaptic element which is assumed to be comprised in an array of similar elements (now shown in the drawing).

The element shown in the Figure is mechanically supported on a substrate 10. Since many, e.g. 104, other similar elements (not shown) are supported on the same substrate 10 it is important that the substrate has a high electrical resistance. A glass sheet is suitable but a sheet of silicon the surface of which has been oxidised to SiO2 is preferred. The thickness of the substrate 10 is not of major importance provided it is capable of supporting the device.

The synaptic element comprises a metallic layer, generally indicated by the numeral 12. Layer 12 is conveniently 50 to 100 nm thick and Cr is a suitable metal. The host material 11, e. g. amorphous silicon doped with a small amount of boron, is also supported on the substrate 10 but the host material partially overlaps and is in electrical contact with part of the metallic layer 12. Thus the layer 12 comprises two functional regions; namely a first electrode 13 in contact with host material 11 and first contact 14 to which external electrical leads may be attached.

The host material 11 is surrounded by an insulating matrix 15. A second metallic region 16 overlays the insulating matrix 15. Preferably the second metallic region 16 is vanadium and it is 40 - 70 nm thick. As an alternative to the preferred metal (V) other metals, e.g. Co, Ni, Pd, Fe and Mn also give very good results. The second metallic region 16 includes a cap electrode 17 and a second contact region 18, which is in mechanical contact with the substrate 10. The cap electrode 17 passes through an aperture in the insulating matrix 15 to make electrical contact with the host material 11. The size and location of this aperture defines the electrode 17. There is a region 19 of the host material 11 which is located between the cap electrode 17 and the first electrode 13. The region 19 predominates in the passage of an electric current through the synaptic element. Forming and programming probably affect the properties of the region 19.

In use, e.g. when the element is connected into a neural network, external connectors are attached to the contacts 14 and 18 whereby the host material, and preferably its region 19, are incorporated into an external circuit.

One method of producing an array of synaptic elements, each as shown in the Figure will now be described by way of example.

First, a uniform metallic layer was deposited onto the substrate 10 using an evaporative technique. The uniform layer was etched, using conventional mask techniques, into the metallic sheets 12. Host material was deposited, using plasma enhanced chemical vapour deposition at 220 - 300°, to give a complete layer preferably 60 - 150 nm thick. Using conventional masks and etching the complete layer was divided into the individual synaptic elements 11.

Next, an insulating layer was applied over the whole substrate, including the host regions 11 and the first electrodes 13. Unwanted insulating material was removed using conventional masking and etching techniques. Thus the insulating material between the elements was removed and the first contacts 14 were exposed. The substrate between the individual elements was aiso exposed.

A significant feature of the etching is that an aperture (in the insulating material) was created on top of each host region 11. This aperture is important because it defined accurately the position and size of the cap electrode 17. The etching of the insulating layer is a convenient way of providing accurate apertures.

Finally, an overall second metallic layer was deposited and etched to give the cap electrode 17 and the second contact regions 18. The second metallic layer filled the apertures in the insulating layer so that size and location of the cap electrode 17 was accurately defined by the apertures. Etching the second metallic layer completes this part of the process and gives elements as shown in the Figure.

After the depositions and shapings were complete, the preparation of the synaptic element was completed by "forming". Forming comprised: applying a voltage of +16 volts for 300 ns to cap electrode 17. The application of this voltage caused the resistance of the element to drop from a very high, effectively infinite value, to a lower value within the range 1 - 2,000 kΩ. The forming process renders the element usable as a resistive synaptic element, but the resistance thereof needs to be adjusted to a target value, and this can be done by applying programming voltages to control the resistance as necessary. It is interesting to note that the forming pulse had positive polarity, and it caused a reduction in the resistance. After the forming pulse has been applied, a positive voltage will tend to increase the resistance of the element and a negative voltage will tend to cause a decrease. That is, after forming, the effect of polarity is reversed.

It should be noted that forming leaves the resistance at an arbitrary value. There are three ways of dealing with this circumstance.

  • (a) Apply a long negative pulse, e.g. -3v for 10 µs, to adjust the resistance to a low value. Thus first programming will always need to increase a low initial value to a higher target value.
  • (b) Apply a long positive pulse, e.g. +3V for 10 µs, to adjust the resistance to a high value. Thus first programming will always need to reduce a high initial value to a lower target value.
  • (c) Take no preliminary action to adjust the initial value of resistance which occurs after forming. This means that the resistance value at first programming is arbitrary and it is necessary to measure the resistance value in order to decide which polarity to use for the first programming.

Although it has been experimentally demonstrated that the forming and programming processes create the resistance as appropriate, these processes are not fully understood. It is believed that forming transfers some material from the electrode 17 into the region 19, whereby the transferred material acts as an active species for the conduction of an electric current through the element. According to this hypothesis, the programming voltages affect the distribution and/or the concentration of the active species whereby the resistance of the element is affected. Whatever the explanation, it has been experimentally observed that the application of pulses in accordance with the invention adjusts the resistance of the element to a target value.

In this specification voltages are specified as applied to electrode 17 with electrode 13 as the zero reference. The forming was achieved with +16 volts.

Tables 1, 2, 3 and 4 below illustrate the effect produced by pulses of varying duration on three synaptic elements made as described above. In all cases procedure (a) was adopted, i.e. after forming the resistance was adjusted to a low value whereby programming was required to increase the value of the resistance. Pulse Height +2.6V PULSE NUMBER PULSE DURATION ns RESISTANCE kΩ 1. 10 4. 57 2. 20 10 3. 50 26.3 4. 100 56. 2 5. 200 121 6. 600 398 7. 1000 660 8. 2000 1380
Pulse height +2.4V PULSE NUMBER PULSE DURATION ns RESISTANCE kΩ 1. 10 2.75 2. 20 4. 75 3. 50 13.1 4. 100 27. 5 5. 200 56. 2 6. 600 178 7. 1000 316 8. 2000 630 9. 5000 1780
Pulse Height +2.13V PULSE NO. PULSE DURATION ns RESISTANCE kΩ 1. 10 2.5 2. 20 2.5 3. 30 3.23 4. 50 4.78 5. 100 9.55 6. 200 21.37 7. 400 46.6 8. 600 72.4 9. 1000 120 10. 2000 263 11. 4000 562 12. 10000 1320

It will be noted that pulse 4 of Table 1 had a duration of 100 ns and it gave a resistance of 56.2 kΩ. Pulse 2 preceded pulse 4, the duration of pulse 2 was 20 ns and it gave a resistance of 10 kΩ. If a pulse of 20 ns were applied after pulse 4, the resistance would remain at 56.2 kΩ, i.e. the shorter pulse would have no effect. It will also be noted that pulse 7 of Table 1 had a duration of 1,000 ns and it gave a result of 660 kΩ. If a pulse of 1,000 ns had been applied immediately after forming, i.e. if pulses 1 - 6 had been omitted, the result would still have been 660 kΩ. This emphasises that it is only the duration of the longest pulse which is significant. The shorter pulses, whether applied before or after, have little or no effect.

It is emphasised that if the figures given in Tables 1, 2 and constituted a reliable correlation, the pulse duration needed to produce any target value could be selected and applied. However, it is very difficult to establish reliable correlations, so an alternative technique, illustrated in Figure 4 is preferred. Target Value 200 kΩ PULSE NUMBER PULSE VOLTAGE PULSE DURATION ns RESISTANCE kΩ 1. +2.6 100 54 2. +2.6 200 115 3. +2.6 300 185 4. +2.6 400 230 5. -2.6 10 221 6. -2.6 20 217 7. -2.6 30 210 8. -2.6 40 206 9. -2.6 50 198

Table 4 illustrates a procedure for increasing an initial low resistance to a target value of 200 kΩ.

Pulses 1 to 4 use the positive polarity to increase the resistance. After pulse 3 the resistance is too low at 185 kΩ so pulse 4 is applied which makes the resistance too high at 230 kΩ. Therefore the polarity is reversed and shorter pulses are applied to reduce the resistance. After pulse 9 the resistance is 198 kΩ and an error of 2 kΩ is regarded as acceptable. If greater accuracy were required the polarity would be reversed (i.e. back to positive) and even shorter increments used.


Anspruch[de]
  1. Verfahren zum Einstellen des Widerstands eines Widerstands-Speicherelements auf einen Zielwert, wobei das besagte Widerstands-Speicherelement eine zwischen zwei metallischen Elektroden (12, 13, 14 und 16, 17, 18) befindliche Widerstandsschicht (11) umfaßt und die besagte Widerstandsschicht (11) aus einem Grundmaterial (11) gebildet ist, das eine aktive Sorte enthält, die einen wesentlichen Anteil an der Leitung eines elektrischen Stromes hat und den Widerstand der Widerstandsschicht (11) einstellt; dadurch gekennzeichnet, daß das Verfahren folgendes umfaßt:
    • (a) Herstellen einer Korrelation zwischen der Dauer eines entweder an ähnliche Elemente oder an dasselbe Element angelegten Spannungsimpulses und dem Widerstand des Elements nach Anlegen eines Impulses mit der besagten Dauer,
    • (b) Verwenden der besagten Korrelation zur Bestimmung der Dauer des zur Erzeugung des Zielwertes geeigneten Programmierungsimpulses, und
    • (c) Anlegen eines Programmierungsimpulses mit der durch die besagte Korrelation angegebenen Dauer zur Erzeugung des besagten Zielwertes.
  2. Verfahren zur Einstellung des Widerstandes eines Widerstands-Speicherelements auf einen Zielwert, wobei das besagte Widerstands-Speicherelement eine zwischen zwei metallischen Elektroden (12, 13, 14 und 16, 17, 18) befindliche Widerstandsschicht (11) umfaßt und die besagte Widerstandsschicht (11) aus einem Grundmaterial (11) gebildet ist, das eine aktive Sorte enthält, die einen wesentlichen Anteil an der Leitung eines elektrischen Stromes hat und den Widerstand der Widerstandsschicht einstellt; dadurch gekennzeichnet, daß das besagte Verfahren das Auswählen einer Polarität, die dazu geeignet ist, das Verschieben des Widerstandes auf den gewünschten Zielwert hin zu bewirken, und das Anlegen einer Mehrzahl von Impulsen mit derselben besagten Polarität und im wesentlichen derselben Spannung aber unterschiedlichen Dauern umfaßt, wobei die besagte Spannung innerhalb des Programmierungsbereichs liegt, wobei ein späterer Impuls eine längere Dauer als ein früherer Impuls aufweist, wodurch bewirkt wird, daß sich der Widerstand des Elements auf den besagten Zielwert hin verschiebt und wobei die Anlegung der besagten Impulse abgeschlossen wird, wenn der Widerstand des Elements dem Zielwert gleich wird oder ihn übertrifft.
  3. Verfahren nach Anspruch 2, wobei die Mehrzahl von Impulsen eine Abfolge darstellt, in der die Dauer jedes Impulses länger als die Dauer seines unmittelbaren Vorgängers ist.
  4. Verfahren nach einem der Ansprüche 2 oder 3, das eine erste Behandlung mit einer ersten Polarität gefolgt von einer zweiten Behandlung mit der umgekehrten Polarität umfaßt; wobei bei der zweiten Behandlung kleinere Schritte der Dauer verwendet werden, wodurch der Widerstand bei der zweiten Behandlung langsamer als bei der ersten Behandlung verschoben wird, so daß bei der zweiten Behandlung eine bessere Annäherung an den Zielwert erreicht wird als bei der ersten Behandlung.
  5. Verfahren nach einem beliebigen der vorhergehenden Ansprüche, wobei die aktive Sorte ein im Grundmaterial (11) entweder als kleine Teilchen oder Atome verstreutes Metall ist.
  6. Verfahren nach Anspruch 5, wobei die aktive Sorte aus demselben Metall wie eine Elektrode (13, 17) gebildet ist.
  7. Verfahren nach Anspruch 6, wobei das Grundmaterial (11) amorphes Silicium oder dotiertes amorphes Silicium ist und eine Elektrode (17) und die aktive Sorte unter V, Co, Ni, Pd, Fe und Mn ausgewählt sind.
Anspruch[en]
  1. A method of adjusting the resistance of a resistive memory element to a target value wherein said resistive memory element comprises a resistive layer (11) located between two metallic electrodes (12, 13, 14 and 16, 17, 18) and said resistive layer (11) is formed of a host material (11) which contains an active species which plays a substantial part in the conduction of an electric current and adjusts the resistance of the resistive layer (11); characterised in that the method comprises:
    • (a) establishing a correlation between the duration of a voltage pulse applied either to similar elements or to the same element and the resistance of the element after a pulse of said duration has been applied,
    • (b) utilising said correlation to determine the duration of the programming pulse appropriate for producing the target value, and
    • (c) applying a programming pulse of the duration specified by said correlation to produce said target value.
  2. A method of adjusting the resistance of a resistive memory element to a target value wherein said resistive memory element comprises a resistive layer (11) located between two metallic electrodes (12, 13, 14 and 16, 17, 18) and said resistive layer (11) is formed of a host material (11) which contains an active species which plays a substantial part in the conduction of an electric current and adjusts the resistance of the resistive layer; characterised in that said method comprises selecting a polarity adapted to cause the resistance to move to ards the desired target value and applying a plurality of pulses having the same said polarity and substantially the same voltage but differing durations wherein said voltage is within the programming range wherein a later pulse has a greater duration than an earlier pulse, whereby the resistance of the element is caused to move towards said target value and wherein the application of said pulses is terminated when the resistance of the element becomes equal to or overshoots the target value.
  3. A method according to claim 2, wherein the plurality of pulses constitutes a sequence in which the duration of each pulse is greater the duration of its immediate predecessor.
  4. A method according to either of claims 2 or 3, which comprises a first treatment at a first polarity followed by a second treatment at the reverse polarity; said second treatment utilising smaller increments of duration whereby the resistance is moved more slowly in the second treatment than in the first treatment so that a better approximation to the target value is achieved in the second treatment than in the first treatment.
  5. A method according to any one of the proceedings claims, wherein the active species is a metal dispersed in the host material (11) either as small particles or atoms.
  6. A method according to Claim 5, wherein the active species is formed of the same metal as one electrode (13, 17).
  7. A method according to Claim 6, wherein the host material (11) is amorphous silicon or doped amorphous silicon and one electrode (17) and the active species are selected from V, Co, Ni, Pd, Fe and Mn.
Anspruch[fr]
  1. Procédé de réglage de la résistance d'un élément de mémoire résistif à une valeur cible dans lequel ledit élément de mémoire résistif comprend une couche résistive (11) située entre deux électrodes métalliques (12, 13, 14 et 16, 17, 18) et ladite couche résistive (11) est formée d'une matière hôte (11) qui contient une espèce active qui joue un rôle considérable dans la conduction d'un courant électrique et règle la résistance de la couche résistive (11); caractérisé en ce que le procédé comprend:
    • (a) l'établissement d'une corrélation entre la durée d'une impulsion de tension appliquée soit à des éléments semblables, soit au même élément et la résistance de l'élément après qu'une impulsion de ladite durée ait été appliquée,
    • (b) l'utilisation de ladite corrélation pour déterminer la durée de l'impulsion de programmation appropriée pour produire la valeur cible, et
    • (c) l'application d'une impulsion de programmation de la durée spécifiée par ladite corrélation pour produire ladite valeur cible.
  2. Procédé de réglage de la résistance d'un élément de mémoire résistif à une valeur cible dans lequel ledit élément de mémoire résistif comprend une couche résistive (11) située entre deux électrodes métalliques (12, 13, 14 et 16, 17, 18) et ladite couche résistive (11) est formée d'une matière hôte (11) qui contient une espèce active qui joue un rôle considérable dans la conduction d'un courant électrique et règle la résistance de la couche résistive; caractérisé en ce que ledit procédé comprend la sélection d'une polarité adaptée pour forcer la résistance à se déplacer vers la valeur cible désirée et l'application d'une pluralité d'impulsions ayant la même dite polarité et substantiellement la même tension mais des durées différentes dans lequel ladite tension est comprise dans les limites de la plage de programmation dans lequel une impulsion ultérieure a une plus longue durée qu'une impulsion antérieure, si bien que la résistance de l'élément est forcée de se déplacer vers ladite valeur cible et dans lequel l'application desdites impulsions est terminée quand la résistance de l'élément devient égale ou supérieure à la valeur cible.
  3. Procédé selon la revendication 2, dans lequel la pluralité d'impulsions constitue une séquence dans laquelle la durée de chaque impulsion est supérieure à la durée de celle qui la précède immédiatement.
  4. Procédé selon l'une ou l'autre des revendications 2 et 3, lequel comprend un premier traitement à une première polarité suivi d'un deuxième traitement à la polarité inverse; ledit deuxième traitement utilisant des incréments de durée plus petits si bien que la résistance se déplace plus lentement dans le deuxième traitement que dans le premier traitement de telle sorte que dans le deuxième traitement l'approximation vers la valeur cible réalisée est meilleure que dans le premier traitement.
  5. Procédé selon l'une quelconque des revendications précédentes, dans lequel l'espèce active est un métal dispersé dans la matière hôte (11) sous forme soit de petites particules, soit d'atomes.
  6. Procédé selon la revendication 5, dans lequel l'espèce active est formée du même métal qu'une électrode (13, 17).
  7. Procédé selon la revendication 6, dans lequel la matière hôte (11) est du silicium amorphe ou du silicium amorphe dopé et une électrode (17) et l'espèce active sont sélectionnées à partir de V, Co, Ni, Pd, Fe et Mn.






IPC
A Täglicher Lebensbedarf
B Arbeitsverfahren; Transportieren
C Chemie; Hüttenwesen
D Textilien; Papier
E Bauwesen; Erdbohren; Bergbau
F Maschinenbau; Beleuchtung; Heizung; Waffen; Sprengen
G Physik
H Elektrotechnik

Anmelder
Datum

Patentrecherche

Patent Zeichnungen (PDF)

Copyright © 2008 Patent-De Alle Rechte vorbehalten. eMail: info@patent-de.com