The work leading to this invention has received funding from the European Research Council under the European Union's Seventh Framework Programme (FP7/2007-2013) / ERC grant agreement n° [247368].10

FIELD

The present invention relates to a magnetic storage device and method and, in particular but not exclusively, to a magnetic data storage device and method utilising techniques of spintronics to enable data storage by using forced disturbances in magnetic coupling between layers of magnetic material to store information.

BACKGROUND

In digital memory technologies, the main technological approach to non-volatile memory uses FLASH memory. FLASH memory performs data storage by using charges floating in an oxide layer to provide individual data bit records. FLASH memory is known to have a number of drawbacks in terms of limited lifecycle (maximum read/write cycles), high power requirements, and slow read/write times. These drawbacks are the focus of a number of approaches in the mass storage application field including the use of DRAM caches to mask write latency, use of compression to avoid write amplification, and using over-provisioning to provide clean blocks for write operations.

In addition, and in view of the drawbacks of FLASH and similar solid state electronic memory technologies, there have been proposed some techniques for solid state magnetic memory. In such technologies, the data storage would be provided by some form of magnetic state retention in contrast to the electrical state retention of electronic memory. One such approach is that of Magnetoresistive RAM (MRAM). A number of MRAM techniques have been described, including the following.

US Patent No 7,226,796 describes synthetic antiferromagnet structures for use in magnetic tunnel junctions in MRAM technology.

US Patent No 6,898,1 12 describes a synthetic antiferromagnetic structure for magnetoelectronic devices.

"Magnetic Domain-Wall Racetrack Memory", S.S.P.P.Parkin, M.Hayashi, L.Thomas, Science 320, 190 (2008); and US Patents Nos 6,834,005, 6,898, 132, 6,920,062, 7,031 ,178, and 7,236,386 describe different examples of a 3-dimensional non-volatile data storage device based on magnetic domain walls moving up shift registers comprising vertical tracks of magnetic material. All of the arrangements presented in these documents use spin transfer to propagate the domain walls. This technology does not use synthetic antiferromagnets at all, rather it uses a shift register arrangement for propagation of magnetic domains.

"Room temperature magnetic quantum cellular automata", R.P.Cowburn, M.E.Welland, Science 287, 1466-1468 (2000) and "Magnetic nanodots for device applications", R.P.Cowburn, Journal of Magnetism and Magnetic Materials. 242, 505-51 1 (2002) describe a soliton on a chain of magnetostatically parallel coupled ferromagnetic disks arranged adjacent one another in a single plane. This technology provided no synchronous mechanism for propagation, no possibility of putting a stream of bits in the same conduit, and no mechanism for unidirectional propagation of data.

PCT patent application publication no WO2002/041492 describes two systems: one is domain wall logic using magnetic nanowires, the other is the quantum cellular automata system described in Science 287, 1466-1468 (2000) mentioned above using magnetostatically coupled dots carrying a soliton.

Probing antiferromagnetic coupling between nanomagnets, R.P.Cowburn, Phys. Rev. B 65, 092409 (2002) describes chains of magnetostatically coupled ferromagnetic disks with an anisotropy. One of the conclusions of this work was that data could not be reliably propagated over any significant distance because the driving force decayed away. As with the other Cowburn works mentioned above, the disks were in the same plane.

US Patent 6,531 ,723 describes a magnetoresistance random access memory for improved scalability. This document introduces what is now known as "Toggle MRAM". The document describes an MRAM cell with a synthetic antiferromagnet of N layers to hold a single data bit. This arrangement provides increased switching volume leading to improved scalability.

US Patent 6,545,906 describes a method of writing to a scalable magnetoresistance random access memory element. This has the same inventors as US6,531 ,723 and describes how a localised rotating field for the above N-layered SAF stack can be produced by delaying the current pulses through orthogonal word and bit lines.

Parish MCB, Forshaw M , I EE Proceedings-Circuits Devices and Systems 151 (5) 480-485 showed in a figure six repeat layers of a synthetic antiferromagnet, including ellipticity to create anisotropy, in the context of quantum cellular automata, as in Science 287, 1466-1468 (2000) mentioned above. The focus of this part of the paper is on keeping the layers coupled antiferromagnetically without error. The disclosure is thus very similar to Phys. Rev. B 65, 092409 (2002) mentioned above.

Property variation with shape in magnetic nanoelements, R.P.Cowburn, Invited Topical Review, J. Phys. D 33, R1-R16 (2000); Lateral interface anisotropy in nanomagnets, R.P.Cowburn, D.K.Koltsov, A.O.Adeyeye, M.E.Welland,, J. Appl. Phys. 87, 7067-7069 (2000) and Superparamagnetism and the future of magnetic random access memory, R.P.Cowburn, J. Appl. Phys. 93, 9310 (2003), all provide examples of elliptical single layer magnetic structures to create shape anisotropy.

Additional work of R.P.Cowburn is set out in WO2010/055329. This document uses a column of antiferromagnetically coupled discs to produce a memory structure. The structure of this document has a column comprising a plurality of layers of magnetic material, each sized to adopt a single magnetic domain state, and a plurality of layers of non-magnetic material arranged as spacer layers between adjacent ones of the layers of magnetic material; such that successive magnetic layers in the column are magnetically antiparallel coupled. Thereby the column is operable to maintain therein a plurality of stable transitions of an order parameter of the magnetisations between the magnetic layers, the transitions having a chirality.

SUMMARY

The present invention has been conceived in the knowledge of limitations and drawbacks in conventional systems.

Accordingly, viewed from a first aspect, the present invention can provide a magnetic memory structure comprising a linear arrangement of alternating magnetic and non-magnetic layers and in which an easy axis of anisotropy of each layer is offset from the equivalent axis of an adjacent layer so as to impart embedded chirality to the arrangement. Thereby a transition in order parameter between different regions of the column has a chirality imparted to it by the embedded chirality of the column.

Viewed from another aspect, the present invention can provide a magnetic memory structure . The structure can comprise a plurality of layers of magnetic material and a plurality of layers of non-magnetic material arranged as spacer layers between adjacent ones of the layers of magnetic material. The magnetic layers can be arranged relative to one another such that the easy axis of anisotropy of each layer is rotationally offset from that of the next magnetic layer; such that the pattern of easy axes of anisotropy of the successive magnetic layers in the structure has chirality, and such that the structure is operable to maintain therein a plurality of stable transitions of an order parameter of the magnetisations between ones of the magnetic layers, the transitions having a chirality imparted by the structure. Thereby a memory structure can be provided that can hold abrupt, stable transitions which remain persistently chiral despite their abruptness.

In some embodiments, the chirality of the transitions follows the direction of the chirality of the easy axes of anisotropy of the successive magnetic layers in the structure and provides an inter-layer rotational offset greater than the offset provided by the chirality of the structure between two successive magnetic layers where a transition is present. Thus the chirality of the transition follows the inherent chirality of the structure but the angular shift in magnetisation alignment across the frustration is greater, in the direction of the chirality, than the angular offset between successive layers.

In some embodiments, the dominant magnetic coupling between successive magnetic layers is RKKY coupling. Thus abrupt transitions capable of being closely spaced can be utilised. In some embodiments, the transitions define a plurality of regions within the structure, each region having an order parameter opposite to the order parameter of an adjacent region. Thus the transitions can be used to define regions having different properties within the structure. In some embodiments, the rotational offset of anisotropy easy axis between successive magnetic layers is in the range of approximately 3 to 20 degrees. Thus a useful amount of inter-layer offset can be achieved.

In some embodiments, each stable transition is a soliton. In some embodiments, the soliton is a topological soliton. In some embodiments, the topological soliton is a kink soliton.

In some embodiments, the structure comprises between 100 and 100,000 magnetic layers. Thus large and small devices can be provided. In some embodiments, each magnetic layer is sized to adopt a single magnetic domain state.

In some embodiments, the transitions can be propagated along the structure. Thus transition positions can be varied over time to facilitate altering memory contents of the structure. In some embodiments, the transitions can be propagated along the structure under the influence of an externally applied rotating magnetic field aligned to the chirality of the transitions. Thus propagation can be controlled.

In some embodiments, the structure further comprises an arrangement operable to introduce a transition. Thus transitions can be introduced to facilitate altering memory contents of the structure. In some embodiments the arrangement comprises a charge pulse conduit arranged parallel and adjacent to an end magnetic layer in the structure operable to carry an electrical charge pulse therethrough. Thus efficient transition injection can be performed. In some embodiments, the structure further comprises a drive element operable to cause a charge pulse to travel through the charge pulse conduit whilst a rotating magnetic field is applied to the structure. Thus the writing can be combined with an external field to facilitate secure and controlled altering of memory contents of the structure.

In some embodiments, the structure further comprises an arrangement operable to read a transition. Thus stored memory contents can be retrieved. In some embodiments, the arrangement to read a transition uses at least one of a giant magneto resistance spin valve, a tunnel magneto resistance structure, and a magnetic tunnel junction stack. Thus a flexible approach to transition reading can be taken. In some embodiments, the arrangement to read a transition is arranged such that the transition is maintained in the structure after reading thereof. Thus a persistent memory can be achieved. In some embodiments, the placement of the arrangement to introduce a transition into a structure and the arrangement to read a transition from the structure is such as to enable operation of the structure as a first in first out shift register, or as a a first in last out shift register. Thus a variety of operation modes are possible. In some embodiments, each transition has a size governed by the ratio of the average interaction field in one magnetic layer due to the presence of the neighbouring magnetic layer to the anisotropy field strength in the magnetic layer. Thus the behaviour of the structure can be determined based upon the properties of the materials and/or dimensions of the structure. In some embodiments the ratio has a value less than or equal to 1. In some embodiments the ratio has a value in the range 0.1 to 0.9. In some embodiments the ratio has a value in the range 0.2 to 0.6. In some embodiments the ratio has a value in the range 0.2 to 0.4.

In some embodiments, the transition has a size equal to or less than two magnetic layers. In some embodiments the transition has a size equal to or less than one magnetic layers. Thus abrupt transitions can be used to enable a high data density.

In some embodiments, the transition spacing is less than or equal to 6 magnetic layers. In some embodiments, the transition spacing is less than or equal to 4 magnetic layers. In some embodiments, the transition spacing is less than or equal to 2 magnetic layers. Thus closely packed transitions can be used to enable a high data density.

Viewed from another aspect, the present invention can provide a magnetic memory circuit comprising at least one structure as set out above; and a signal supply conduit operable to carry a write signal to or read signal from the structure. Thus the memory structures can be incorporated into or formed as part of a memory device.

In some embodiments, a plurality of structures are provided and the signal supply conduit comprises an arrangement to address individual ones of the plurality of structures. Thus memory circuit can access individual ones or groups of a number of memory structures.

Viewed from a further aspect, the present invention can provide a magnetic memory device comprising the memory circuit described above and a magnetic field generator operable to generate a rotating magnetic field. Thus a memory device can include its own propagation field generator.

In some embodiments, the magnetic field generator comprises a configuration to supply a field inducing signal to the signal supply conduit. Thus write and read operations can be coordinated with transition propagation.

In some embodiments, the magnetic field generator comprises a pair of current carrying conductors oriented substantially orthogonally to one another. In some embodiments, the magnetic field generator is configured to controllably subject at least one of the plurality of the magnetic memory structures and data writing elements to a rotating magnetic field. Thus propagation of transitions can be achieved using the field.

In some embodiments, the memory device further comprises an electrical interface operable to receive data bits for writing into the device. In some embodiments, the memory device further comprises an electrical interface operable to transmit data bits for read from memory structures of the device. Thus the magnetic memory structure can be interfaced to an electronic circuit for delivery of data.

Viewed from another aspect, the present invention can provide a method of storing data within a memory structure comprising a plurality of layers of magnetic material and a plurality of layers of non-magnetic material arranged as spacer layers between adjacent ones of the layers of magnetic material, the magnetic layers being arranged relative to one another such that the easy axis of anisotropy of each layer is rotational ly offset from that of the next magnetic layer; such that the pattern of easy axes of anisotropy of the successive magnetic layers in the structure has chirality; the method comprising introducing into the structure a plurality of stable transitions of an order parameter of the magnetisations between ones of the magnetic layers, the transitions having a chirality maintained by the embedded chirality of the structure.

In some embodiments, the data is encoded using one of: an order parameter state, a presence or absence of an order parameter transition; a phase shift key encoding or a hard disk encoding schema to represent data values. Thus a variety of data encoding schemes are available.

In some embodiments, the structure comprises between 100 and 100,000 magnetic layers. Thus large and small memories can be provided.

In some embodiments, each transition is a soliton. In some embodiments, the soliton is a topological soliton. In some embodiments, the topological soliton is a kink soliton.

In some embodiments, the method further comprises introducing a transition into the structure by passing a charge pulse through a charge pulse conduit arranged parallel and adjacent to an end magnetic layer in the structure. In some embodiments, each magnetic layer is sized to adopt a single magnetic domain state.

In some embodiments, the method further comprises reading a transition from the structure using at least one of a giant magneto resistance spin valve, a tunnel magneto-resistance structure, and a magnetic tunnel junction stack. In some embodiments, the method further comprises maintaining the transition in the structure after reading. In some embodiments, the method further comprises operating the structure as a first in first out shift register or as a first in last out shift register.

In some embodiments, the method further comprises applying an externally generated rotating magnetic field to the structure to cause propagation of chiral transitions along the structure. In some embodiments, the method further comprises generating the rotating magnetic field using a pair of current carrying conductors oriented substantially orthogonally to one another.

In some embodiments, the introducing comprises introducing transitions having a size equal to or less than two magnetic layers. In some embodiments, the introducing comprises introducing transitions having a size equal to or less than one magnetic layers. Thus abrupt transitions can be used to enable a high data density.

In some embodiments, the introducing comprises introducing transitions having a spacing less than or equal to 6 magnetic layers. In some embodiments, the introducing comprises introducing transitions having a spacing less than or equal to 4 magnetic layers. In some embodiments, the introducing comprises introducing transitions having a spacing less than or equal to 2 magnetic layers. Thus closely packed transitions can be used to enable a high data density.

Viewed from another aspect, the invention can provide apparatus comprising a stack of magnetically coupled magnetic elements wherein successive ones of the magnetic elements are arranged such that the easy axis of anisotropy of each magnetic element is rotationally offset from that of neighbouring magnetic elements such that a lowest energy state of the stack is offset from antiparallel coupling between successive magnetic elements and wherein the rotational offset is approximately equal between successive layers and follows a spiral pattern through the stack, such that one or more stable frustrations in magnetisation alignment can be maintained within the stack, the frustration comprising a chiral soliton in magnetisation direction within the stack and having its chirality maintained by the spiral pattern of anisotropy easy axes within the stack.

Viewed from a further aspect, the invention can provide apparatus comprising a stack of alternate layers of magnetic material and non-magnetic material wherein the magnetic domain states of adjacent ones of the magnetic material layers are coupled via RKKY interactions and wherein successive layers of magnetic material within the stack are arranged such that the easy axis of anisotropy of each layer of magnetic material is rotationally offset from that of neighbouring layers of magnetic material such that a lowest energy state of the stack is offset from antiparallel magnetic coupling between successive layers of magnetic material so as to impart embedded chirality into the stack and such that one or more stable frustrations in magnetisation alignment can be maintained within the stack, the frustration having a chiral property imparted by the embedded chirality of the stack.

Further aspects and embodiments will be apparent from the following detailed description.

BRIEF DESCRIPTION OF THE FIGURES

Specific embodiments of the present invention will now be described by way of example only with reference to the accompanying figures in which:

Figure 1 shows schematically a stack made up of successive magnetic and non-magnetic layers;

Figure 2 shows a plot of soliton width against coupling strength;

Figure 3 shows a plot of propagation barrier against soliton spacing;

Figure 4 shows a plot of propagation barrier against coupling strength;

Figure 5 shows schematically offset coupling between layers;

Figure 6 shows schematically the easy axis of anisotropy of each of a plurality of successive layers in a layered structure made up of successive magnetic and non-magnetic layers;

Figure 7 shows schematically the layered structure of Figure 6 in which a single soliton divides areas having different order parameters;

Figures 8a, 8b and 8c shows schematically the propagation of a pair of solitons along the layered structure;

Figure 9a shows schematically the layered structure of Figure 6 in which a series of solitons encode a four bit data sequence;

Figure 9b shows schematically the encoded data sequence of Figure 9a after three cycles of propagation along the layered structure;

Figure 10 shows a plot comparing nucleation and propagation field strengths;

Figure 1 1 shows a plot of propagation barrier against separation;

Figure 12 shows schematically a data storage device having multiple data storage layered structures;

Figures 13a, 13b and 13c illustrate coding approaches for data storage; and

Figure 14 shows schematically a soliton injector;

Figure 15 shows schematically a spin valve based read-out detector;

Figure 16 shows schematically a TMR based soliton detector;

Figure 17 shows schematically another TMR based soliton detector;

Figure 18 shows schematically another TMR based soliton detector; and

Figure 19 shows schematically an MTJ bases soliton detector. While the invention is susceptible to various modifications and alternative forms, specific embodiments are shown by way of example in the drawings and are herein described in detail. It should be understood, however, that drawings and detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents and alternatives falling within the scope of the present invention as defined by the appended claims.

SPECIFIC DESCRIPTION

Particular examples to illustrate the operation of the present invention will now be described with reference to the Figures. The skilled reader will appreciate therefrom that the principles and concepts which underlie the present invention can be implemented in a variety of ways which may extend beyond the examples described herein. The present invention is therefore not limited by the disclosure of the specific examples, and rather is limited only by the spirit and scope of the appended claims.

Figure 1 shows schematically a stack made up of successive magnetic and non-magnetic layers. In such a structure, the relative sizes of the magnetic and non-magnetic layers can affect the manner in which the magnetic layers interact with one another. The two principle mechanisms via which the magnetic layers can interact are dipolar field coupling (magnetostatic interactions) and RKKY (Ruderman-Kittel-Kasuya-Yosida) exchange interactions.

Suitable techniques for creating a stack of respective magnetic and non-magnetic layers include thin film deposition techniques such as electro-deposition and physical vapour deposition. Examples of physical vapour deposition include sputter deposition. Taking the example of sputter deposition, it is possible to achieve a structure with a controlled easy axis of anisotropy by either depositing the magnetic layers in the presence of an externally applied magnetic field (a technique described in Gentils, A, Chapman, JN, Xiong, G, et al, Variation of domain-wall structures and magnetization ripple spectra in permalloy films with controlled uniaxial anisotropy, J APPL PHYS, 2005, Vol: 98, ISSN: 0021 -8979) or depositing the magnetic layers from an oblique deposition angle such that the deposition is performed from an angle offset from the longitudinal axis of the stack (a technique described in US 6,818,961).

As has been shown in WO2010/055329, it is possible to maintain within a stack of anti- ferromagnetically coupled alternating magnetic and non-magnetic layers a topological kink soliton having handedness (chirality). Such a chiral soliton is mobile, localised and persistent.

As will be appreciated from WO2010/055329, greatest data storage density is achieved by both minimising the soliton size and minimising the soliton spacing. In counterpoint, it will also be appreciated that the soliton must maintain its property of chirality. If the soliton is compressed to a size where it loses chirality it can no longer be propagated along a stack and if allowed to re-expand it may (approximately 50% chance) expand to have the opposite chirality to its original chirality. This size and spacing conundrum can be considered in terms of the physical properties of the magnetic and non-magnetic layers within the stack.

The relative strength of the exchange coupling field (which can be thought of as the coupling strength between adjacent magnetic layers or as the average interaction field in one magnetic layer due the presence of the neighbouring magnetic layer, in any case expressed as a magnetic field) J* and the anisotropy field (anisotropy field strength of each magnetic disk) Hu governs the size of the soliton. In the following discussions of ratios of J* to Hu these are based upon the magnitude of J*. In that adjacent layers are anti-parallel coupled, the actual value of J* will be negative. Where J*/Hu is large, the soliton size is large and where J*/Hu is small, the soliton size is small. As noted above, increased data storage capacity is achieved by reduced soliton size. Thus an increase in data storage capacity is achieved by reducing the value of J*/Hu. However, as noted above, once the soliton size is reduced too far the soliton loses chirality and becomes unstable. It has been found that this instability in soliton chirality occurs at J*/Hu~1.

Figure 2 shows a plot of soliton width against coupling strength. In this graph Hu is 80 Oe and J* is varied to shows the soliton size (width) achieved for that particular combination of J* and Hu. As the soliton behaves in many ways as a wave, the soliton width is plotted as a 10- 90% width (rather than an absolute end to end width) of the soliton in terms of number of magnetic layers. For ease of reference, the graph is marked with a line showing J*=Hu (i.e. J*/Hu=1) As can be seen from the graph of Figure 2, the soliton width drops from around 2.5 layers where J*/Hu is just greater than 1 (i.e. the soliton is still inherently chiral) to just over 1 layer where J*/Hu is just less than 1 , and indeed can drop down to less that 1 layer where J* drops far enough.

A further constraint is that in order for the soliton to be stable and persistent, it will not propagate along the stack (in either direction) without the application of a rotating magnetic field to the stack. This resistance to propagation can be considered as a propagation barrier. If the energy required to over come the propagation barrier, ΔΕρ, which is measured using the pseudo-unit of energy kT (where k is the Boltzmann constant and T is temperature), drops below a value of approximately 60kT then the risk of spontaneous movement of a soliton along the stack increases to a statistically significant level. For example, if the value of ΔΕρ drops to around 50kT, then there is a probability of around once per ten years for each individual soliton to move up or down the stack by one disk position. It will be appreciated that for a data device storing millions of bits of data using a corresponding number of solitons, this may translate to a significant risk of data errors. In counterpoint to the requirement to maintain the propagation barrier high enough to prevent spontaneous soliton shifting, the individual solitons can have a relatively long 0-100% width as compared to their 10-90% width discussed above. This extended yet finite soliton width creates a repulsive effect from one soliton to the next which can have the effect of reducing the external field strength ΔΕρ required to overcome the propagation barrier. If the inter- soliton repulsive effect approaches or exceeds the propagation barrier, then the mere presence of a second soliton too close to a first soliton could propagate one or both solitons along the stack in an uncontrolled and unexpected manner. If the solitons are being used to encode data then this would likely cause data error and thus render a data storage device unreliable.

This balance between propagation barrier and soliton spacing is illustrated by the plot in Figure 3 of propagation barrier against soliton spacing (taking the spacing as the gap between the 90% width tail of one soliton to the 90% width tail of the next soliton). This shows, for J*/Hu=1 , how resistant the soliton is to accidental propagation along the stack by thermal fluctuations in the absence of any applied field at each of a number of different soliton spacings. As can be seen from Figure 3, the value of ΔΕρ is maintained at an acceptable level while the soliton spacing is at least 6 layers, but once the spacing drops to 5 layers, the ΔΕρ is such that less stability of soliton position is provided and for high density spacings where the solitons are only 3 or 4 layers apart, the ΔΕρ value drops to unusably low levels. Thus it is clear that even for the situation of a very small soliton on the verge of losing inherent chirality (J*~Hu), the minimum obtainable spacing is limited by the inter-soliton repulsive effect.

In addition, the energy required to overcome the propagation barrier, ΔΕρ, also falls as soliton size increases. This is illustrated in Figure 4, which shows a plot of propagation barrier against coupling strength without the effect of neighbouring solitons. Here it can be seen that as the magnitude of J* increases for a fixed value of Hu, the energy required to overcome the propagation barrier decreases markedly. Thus, while for a ratio J*/Hu~0.5 the value of ΔΕρ rises to around 800kT, once the value of that ratio drops to much more than 1 the value of ΔΕρ drops to below the usable threshold of ~60kT.

By way of the arrangements described in the following examples and embodiments, it is shown that it is possible to reduce the magnitude of J*/Hu to less than 1 so as to provide reduced soliton width, reduced soliton spacing and increased ΔΕρ.

Figure 5 illustrates schematically a layered structure made up of alternating magnetic and non-magnetic layers and having angularly offset easy axes of anisotropy between adjacent magnetic layers. Each double-headed arrow represents a single magnetic layer (for clarity the non-magnetic spacer layers are not illustrated) in a layered structure. As can be seen, there is a rotational offset between the easy axis of anisotropy between adjacent magnetic layers such that the easy axis directions define a spiral pattern over the course of a number of layers. It should be appreciated that the schematic of Figure 5 illustrates the "empty" state of the layered structure and as such there are no introduced solitons and the layered structure is shown in terms of anisotropy axis direction, not magnetisation direction. Thus the spiral pattern as illustrated is entirely due to the properties of the magnetic layers making up the layered structure and are not related to any introduced frustration or soliton in magnetisation alignment.

In the illustrated example of Figure 5, the offset is approximately 10° per layer, such that over 18 layers the axis returns to its original orientation (as an axis under rotation returns to its original position after 180° of rotation). The number of layers taken for the easy axis to return to its original orientation can be described as the wavelength of the embedded chirality. In the present examples, where there is approximately anti-ferromagnetic coupling between adjacent layers, the value of the wavelength of chirality of the layered structure has no effect whatsoever on the number of layers that a soliton propagates per rotation of an externally applied propagation field.

In other examples, greater or lesser amounts of angular offset between layers can be used. In the present examples, the angular offset is in the range of around 5° to around 10°. In other examples, the range can be around 2° to around 20°. If the angular offset is too small, there is a risk that the offset is removed altogether for at least some layers due to inaccuracies in a manufacturing process. Some manufacturing processes for a layered structure having a spiral pattern of easy axis of anisotropy between successive layers according to these examples can naturally result in an error of up to ±2 or 3 degrees in anisotropy easy axis direction for each layer. Another risk of a too small angular offset is that it may be effectively removed by natural fluctuations in the magnetisation due to domain ripple. Domain ripple can occur where there is a relatively large area of magnetisation which is being held close to an anisotropy easy axis, whereby an instability sets in and the domain effectively breaks into regions either side of the axis having wave behaviour along the axis. Thus there is a tendency to experience natural dispersal in anisotropy direction. If the angular offset is too large, then the energy barrier to propagation starts to reduce such that the operating margin reduces. As will be appreciated from the following discussion, there is a considerable tolerance to imprecise offset steps from manufacturing. For example, for a nominal offset of 10° per layer, it is not necessary for every layer to have exactly the same offset and the layered structure will still perform adequately with the layers having varying offsets in the range of around 7° to 13°. If the variability of offsets between successive layers increases too far, it increases the risk of the zero point being crossed (such that there is no offset for a pair of layers) and creates a risk of soliton pinning whereby a soliton is unable to move in the presence of the normal propagation field.

Thus the presently described arrangements make use of a mobile, localised and persistent region of frustration in magnetisation alignments within a layered structure having alternating magnetic and non-magnetic layers. This region of frustration can be described as a topological kink soliton and has handedness and thus will be referred to herein as a soliton and as a chiral soliton. As will now be appreciated, in the present approaches, irrespective of any chirality that the soliton itself possesses at the time of its creation the chirality of the soliton is imparted by the properties of the layered structure as the layered structure has inbuilt handedness.

As will be appreciated, in the layered structure illustrated in Figure 5 and in other similar arrangements having a spiral pattern of easy axes of anisotropy between adjacent magnetic layers, the adjacent magnetic layers of the layered structure are not strictly 180° anti- ferromagnetically coupled but instead have alignment of 180° minus the angular offset in low energy magnetisation alignment positions (easy axis of anisotropy) from one layer to the next. As the alignment is tending toward the anti-parallel, the value of J* as between adjacent layers will be negative and all discussion of ration of J* to Hu are based upon the magnitude of J*

The arrangements of the present disclosure are configured to rely on RKKY coupling to govern the behaviour of interactions between the magnetic layers in a layered structure. To the extent that dipolar coupling occurs within the arrangements of the present disclosure, its effects are subsidiary to the effects of the RKKY coupling. This reliance upon RKKY coupling rather than magnetostatic coupling is due to RKKY coupling being localised (i.e. the effects of RKKY coupling between two magnetic layers do not affect more distant layer layers) whereas dipolar is longer range and thus can affect layers either side of the pair of layers. The reliance upon RKKY coupling in preference to magnetostatic coupling may lead to a more complex fabrication process but, as will be come apparent from the discussion below, enables the use of smaller, closer packed solitons. As mentioned above, the arrangements set out herein utilise rest state inter-layer magnetisation directions that are not anti-parallel aligned. As will become clear from the following disclosure, the RKKY coupling provides: (i) the correct strength to balance the chosen anisotropy strength to give a compact soliton; and (ii) reproducibility in manufacture.

Therefore, in the arrangements set forth in the present disclosure, the non-magnetic layer is sufficiently thin to allow RKKY coupling to dominantly contribute to the inter-layer interactions. In some examples, the non-magnetic layers have a thickness in the range of approximately 0.5 to 2.5 nm. As RKKY coupling is independent of the lateral size of the layer elements the layer area does not alter the strength of the inter-layer coupling. On the other hand, dipolar does depend on lateral size such that a larger layer area would be expected to have a smaller dipolar coupling field strength. However, as dipolar coupling acts in same direction as the RKKY coupling of the layers in the structure (i.e. generally tending towards antiferromagnetic alignment) it is not necessary to eliminate bipolar coupling. The present examples are set out on the basis that RKKY coupling is dominant over dipolar coupling in determining inter-layer behaviour. The relative layer thicknesses are arranged to balance the inherent anisotropy within each magnetic layer and the RKKY coupling between the layers so as to create a structure in which a chiral soliton can be injected, propagated, and read and in which such a chiral soliton is persistent and stable. In some examples, the non-magnetic layers are formed from a non-magnetic metal and the magnetic layers are formed from a magnetic metal. In some examples, the non-magnetic material can be copper or aluminium and the magnetic metal can be nickel-iron. Some further specific examples of suitable material combinations include iron/chromium (in this example it may be necessary to perform epitaxial growth of the iron layers in order to introduce uniaxial anisotropy in the magnetic layers); cobalt/copper (this is electrodepositable and involves no precious metals required); and cobalt-iron-boron/ruthenium; or cobalt-iron/ruthenium (these materials are used in conventional MRAM technologies so there is familiarity with these materials in manufacturing industry). In one example, the non-magnetic layers are around 2nm thick and the magnetic layers are around 5nm thick. In some examples, the non-magnetic layers are in the range of around 0.5 to 3.5nm thick and the magnetic layers are in the range of around 1 nm to around 9nm thick. Other specific thicknesses and materials can be selected without departing from the scope of the present invention.

Both of the manufacturing approaches mentioned above (depositing the magnetic layers in the presence of an externally applied magnetic field and depositing the magnetic layers from an oblique deposition angle) can be used to create a structure where the individual magnetic layers have an easy axis of anisotropy which is angularly offset from that of the adjacent magnetic layers. To achieve the angular offset in easy axis of anisotropy between different magnetic layers then, in the externally applied magnetic field approach, the magnetic field direction is altered relative to the layered structure from one layer deposit operation to the next. In the oblique deposition angle approach, the angular offset in easy axis of anisotropy between different magnetic layers is achieved by rotating the layered structure relative to the deposition source between each magnetic layer deposition. Thus in both of these techniques it is possible for the deposition target (the layered structure), the direction providing element or both to be rotated.

By utilising an arrangement as set out herein using a layered structure of alternating layers of magnetic and non-magnetic material with embedded chirality, if a wrong-handed soliton is inserted or created then it automatically flips to follow the embedded chirality once a propagation field is applied. Thus a danger of a wrong-handed soliton propagating the wrong way through the layered structure and colliding with and cancelling a neighbouring correct- handed soliton is avoided.

Thus there has now been described an arrangement for creating a layered structure of interleaved magnetic and non-magnetic layers in which a soliton can be formed, maintained and propagated and where a chiral property of the soliton is provided by embedded chirality within the layered structure. This embedded chirality enables a soliton to have a small size without a chiral property of the soliton being lost. Thus such an arrangement enables the ratio J*/Hu to be less than 1. By making J*/Hu<1 , a soliton contained in the stack: becomes abrupt or small (thus enabling increased data storage density); interacts less with neighbouring solitons (thus enabling reduced soliton spacing); and has a higher energy barrier to propagation (thus increasing data stability).

With reference to Figure 6, there is shown schematically a layered structure (for clarity of this illustration each magnetic layer is shown as having nominal thickness and the non-magnetic layers are omitted) in an "empty" state showing the magnetisation direction in each layer. As can be seen from the Figure, the magnetisation direction of each layer follows the easy axis of anisotropy for that layer. The magnetisation directions of adjacent layers are as near to opposite as the spiral pattern of the easy axes of the successive layers permits. In the present example, the nominal offset between layers is approximately 10° and the spiral is left-handed when viewed from the top of the layered structure in the orientation shown on the page of Figure 6. Thus the magnetisation direction of each layer is offset by 170° (or 190° depending on viewpoint) from each adjacent layer. It will be appreciated that a layered structure can be constructed with the opposite handedness to that illustrated in Figure 6. Such a right handed layered structure works as the left handed layered structure shown in the present examples with the exception that propagation directions are reversed for a given propagation field direction.

As Figure 6 illustrates the "empty" state of a layered structure, it also illustrates the concept of an order parameter. As all of the magnetisation directions are aligned to follow the spiral pattern of the easy axes of anisotropy of the magnetic layers through the layered structure with all magnetisation directions being aligned as far away as possible from the magnetisation directions of all adjacent layers, all of the layers have the same order parameter. The order parameter is arbitrarily defined with reference to this Figure that all of the layers in Figure 6 have order parameter value 1. If all of the magnetisation directions were reversed, then they would have order parameter value -1.

In the following the order parameter is defined according to the following relationship: order parameter for a given layer is equal to the cosine of the difference of the in-plane magnetisation direction of that layer minus the anisotropy easy axis direction in that layer. This relationship can be expressed by the following equation:

Where n is the layer number, θ _η is the in-plane magnetization direction of layer n, and ΔΦ _Κ is the angular change in anisotropy easy axis between two successive layers. The expression ((n(TT-A ⁽ t>K))Mod2TT) describes the anisotropy easy axis direction in layer n. This takes into account the effect on that easy axis by the antiferromagnetic component of the inter-layer coupling by the n Pi term. As will be appreciated, the order parameter value could also be defined as the negative of the outcome of the above equarion.

Thus there has now been described an arrangement whereby a structure having alternating magnetic and non-magnetic layers and having embedded chirality in the form of a spiral pattern of easy axes of anisotropy of the magnetic layers has the capability to have all magnetic layers having a magnetisation direction corresponding to one of two possible order parameters.

Figure 7 shows schematically the stack of Figure 6 in which a single soliton divides areas having different order parameters. To facilitate observation of the soliton within the layered structure, Figure 7 also includes a graph showing order parameter for each layer (labelled "spin number"). As can be seen from Figure 7, in this example a narrow soliton is present at layers 14-15. In this example the soliton is abrupt (i.e. narrow) and thus has a width of approximately 1 disk. As in other parts of the present description, the soliton widths are approximate and not necessarily integer numbers of disks due to the wave-like behaviour of the soliton. This soliton has chirality imparted to it by the embedded chirality of the layered structure such that even with this narrow size it maintains its handedness property and thus will propagate correctly within the structure. Thus all layers "above" the soliton as seen in the orientation of Figure 7 have order parameter 1 and all layers "below" the soliton have order parameter -1. As will be appreciated, the ability to divide the stack into regions of different order parameters using stable transitions enables the storage of data within the layered structure.

Figures 8a, 8b and 8c show schematically the propagation of a pair of solitons along the layered structure. In these figures, the layered structure contains two solitons, each having a width of approximately 1 layer, and being spaced by a relatively wide spacing of 10 layers. As there are two solitons present, the layered structure has three order parameter regions. The "top" and "bottom" regions as observed in the orientation of Figure 8a have order parameter 1 and the middle region has order parameter -1. Propagation of the solitons through the layered structure is achieved by subjecting the stack to a rotating magnetic field where the rotation is substantially coplanar with the plane of each of the magnetic layers. The strength of the field is sufficiently large to overcome the energy barrier to propagation. In some examples, the rotating magnetic field may rotate in a plane that is non-coplanar with the planes of the magnetic layers. In such an example, the propagation strength of that rotating field is the resolved component of the field acting in the planes of the magnetic layers.

As shown in Figure 8a, the solitons start at layer positions (spin numbers) 10 and 20. Then, following one complete 360° anti-clockwise cycle of an externally applied magnetic field the solitons are moved by two disk positions to arrive at the positions shown in Figure 8b. As the layered structure of this example has a left-handed (anticlockwise) spiral direction, the application of an anticlockwise external field moves the solitons down the structure in the orientation shown in Figure 8b. Thus the new soliton positions are at layer positions (spin numbers) 12 and 22. Following a further complete 360° anti-clockwise cycle of an externally applied magnetic field the solitons arrive at the positions shown in Figure 8c. Again the solitons move down the spiral in the orientation shown in Figure 8b such that the new soliton positions are at layers (spin numbers) 14 and 24.

It is noted that the soliton spacing does not appear completely equal when viewed in the order parameter graph. This is due to the interaction of a soliton with discrete nature of the system of the present examples being made up of separate layers. Thus when the propagation field rotates, a soliton that bounds a region of order parameter 1 from a region of order parameter -1 (considered in terms of the direction of propagation) moves at a different point in the rotation of the propagation field to a soliton that bounds a region of order parameter -1 from a region of order parameter 1 (considered in terms of the direction of propagation). Thus, in the order parameter graphs, some of the transitions catch this offset shifting phenomenon and therefore appear to show a slight variation in the soliton spacing.

Thus it has been shown that application of an external rotating magnetic field to the layered structure propagates solitons in the stack in the direction of rotation of that magnetic field. Where the layered structure has left-handed embedded chirality a clockwise field will move the solitons in a first direction and an anticlockwise field will move the solitons in a second direction. Whereas if the layered structure has right-handed embedded chirality a clockwise field will move the solitons in the second direction and an anticlockwise field will move the solitons in the first direction.

Figures 9a and 9b show schematically a layered structure in which a series of solitons encode a data sequence and the propagation of those solitons. As is shown in Figure 9a, four solitons are present and thus define five order parameter regions within the layered structure. The solitons are again narrow and this time the spacing is reduced to only three disks (leading edge to leading edge) in the closest case.

Figure 9b illustrates the soliton positions after three complete 360° clockwise cycles of an externally applied magnetic field. As the layered structure of this example has a left-handed (anticlockwise) spiral direction, the application of a clockwise external field moves the solitons up the layered structure in the orientation shown in Figure 9b. Thus the new soliton positions have each moved up the layered structure by six layers.

Thus there have been described a number of mechanisms by way of which a layered structure of alternating magnetic and non-magnetic layers can be configured to maintain and propagate solitons therein. By providing the layered structure with an inherent, embedded chirality it is possible to have solitons of a size too small to possess their own chirality without losing a chiral property for the solitons. Thus a small soliton size and small soliton spacing can be achieved. The properties of such an embedded chirality layered structure are illustrated in Figures 10 and 1 1.

Figure 10 shows a plot comparing nucleation and propagation field strengths for a column having embedded chirality at a nominal 10° offset per magnetic layer and Hu of 80 Oe. For this graph, each magnetic layer is a circular disk with diameter 1 μηι and thickness 5nm. The line at J*=80 represents the threshold J*/Hu=1 below which a soliton carries no intrinsic chirality as a property of itself. Both the soliton propagation field and the soliton nucleation field lines correspond to the single y-axis of filed strength (H) measured in Oersteds (Oe). The propagation field is the minimum external rotating magnetic field strength required in order to propagate a soliton along the layered structure (i.e. to overcome the energy barrier propagation ΔΕρ) to for a given value of J* (and hence soliton size). The nucleation field is the minimum external rotating magnetic field strength that will cause spontaneous nucleation of solitons within the layered structure. As will be appreciated, a memory device using a layered structure according to the present examples for data storage requires that solitons can be propagated through the layered structure without spontaneous creation of new solitons within the layered structure in an uncontrolled manner. As can be seen from the plot of Figure 10, even at a J*/Hu of around 0.2 there is still a separation of around 25 Oe between the nucleation field and propagation field. By referring back to Figure 2, is can be seen that with J*/Hu ratio of 0.2 results in a soliton size of around 0.8 layers (10%-90% width). Even if an operating point nearer J*/Hu=0.5 were chosen, which provides a propagation field to nucleation field separation of around 150 Oe, it can be seen from Figure 2 that soliton width remains of the order of 0.8 layers (10%-90% width). Thus it is clear that there is no stability problem in terms of collision of propagation and nucleation fields created by dropping below the J*=Hu intrinsically chiral soliton threshold.

Referring now to Figure 1 1 , this shows a plot of propagation barrier against separation. For this graph, each magnetic layer is a circular disk with diameter 1 μηι and thickness 5nm. The inter-layer offset in this case is 5°. Also, J* is set to 30 Oe and Hu is 80 Oe (giving J*/Hu~0.38). The graph of Figure 11 plots the propagation barrier in kT against the leading edge to leading edge soliton spacing. Thus the spacing encompasses the width of the soliton as well as the gap between solitons. As can be seen from Figure 1 1 , the propagation barrier (which corresponds directly to the energy required to overcome the propagation barrier, ΔΕρ) remains constant as the separation reduces to 4 layers and even rises slightly as the separation further reduces to 2 layers. Thus a maximum soliton density of around 50% of the number of disks in the layered structure is achieveable.

Although the detailed examples of Figures 10 and 11 are based upon magnetic layers in the form of circular disks with diameter 1 μηι and thickness 5nm, the disk sizes can be scaled as required for a particular formation technique or desired physical entity size. In some examples, such scaling would first scale the volume down to a point where the value of ΔΕρ is acceptably large to enable reliable and stable operation but not so large as to require unnecessarily big discs (assuming Hu is kept constant), and then scale the thickness and radius to keep the disc volume approximately constant (in order to maintain the value of ΔΕρ approximately constant for a given value of Hu). Thus in the second scaling, the thickness would scale inversely with the square of the radius.

Thus there have been described a number of arrangements that enable operation of a soliton containing layered structure having a J*/Hu value less than 1 and providing thereby a small soliton size and a high soliton density (low soliton separation). Thus a high density of data storage can be achieved without a loss of stability in the solitons stored in the layered structure and without a loss of chirality of the solitons due to small soliton size.

With reference to Figure 12, approaches for using such a layered structure within a data storage device will now be discussed.

Figure 12 shows schematically a data storage device 1 having multiple data storage layered structure 3. In this example, the layered structures 3 are depicted as columns or stacks. Most of the device can be formed, for example, using a Back End Of Line (BEOL) process on a CMOS chip 7. The number of repeat layers in each stack 3 governs the maximum capacity in terms of number of data bits per stack. As mentioned above, a soliton density of around one soliton per two magnetic layers can be achieved by arrangements of the present examples. Although small height stacks can be used, it will be understood that taller stacks provide greater storage capacity per chip area. Therefore, in the following examples, the stacks 3 may have as few as 100 magnetic layers (providing a maximum data density of approximately 50 stored bits per stack). In other examples, each stack may have 1 ,000, 10,000 or 100,000 magnetic layers (for data densities of 500, 5,000 or 50,000 bits per stack respectively). Where each stack has of the order of 10,000 to 100,000 magnetic layers, a suitable high aspect ratio fabrication process may be used.

At one end of each stack 3 there is a soliton injector device 5 which converts electrical signals from the CMOS logic of the chip 7 into a stream of solitons representing a serial data stream that is to be stored in the stack 3. At the same end (as shown) or the other end (not shown) of the stack there is a detector device which is used during data retrieval to convert the magnetic state at the end of the stack back into electrical pulses. An external applied field generator 9 can be operated to apply a magnetic field which rotates in the plane of the device and which can therefore drive solitons along each stack 3.

In some examples, the data blocks can be arranged to provide some stacks 3 in which the data therein is propagated clockwise by a clockwise field and some stacks in which the data therein is propagated anticlockwise by a clockwise field. This arrangement provides for a pair of stacks to be operated as a single memory element. This can be achieved by electronically linking the read/write circuitry for the two stacks such that as solitons are propagated out of the clockwise stack in order to read the data encoded therein, the same data is rewritten into the anticlockwise stack by insertion of solitons encoding the same data. Formation of a device having some clockwise and some anticlockwise stacks can be achieved by using some form of shadow mask during deposition steps so as to cause layer deposits on only the correct half of the columns at any one time. An alternative arrangement to achieve the same effect has all of the stacks having the same handedness, but using localised propagation fields in the device. Thus for every clockwise cycle of propagation field for one group of the stacks, a matching anticlockwise cycle of propagation field is applied to a corresponding second group of stacks so as to cause solitons in the first group to be propagated in the opposite direction to those ion the second group. A further arrangement to achieve the same effect is to provide the read/write element at or near the centrepoint of a single clockwise or anticlockwise stack. An alternative that is more space efficient would provide a single stack with a read element at one end and a write element at the other end. Thus as every bit is propagated out of and read from the read end, the same value is rewritten and propagated into the write end.

The data storage device can be implemented with either a global field rotator (as shown in Figure 12) or with local field generators for individual ones or groups of stacks 3. Where a field is generated which affects more than one stack, any use of the field to propagate solitons up or down those stacks will affect all of the driven stacks at once, leading to parallel propagation of solitons in the affected stacks.

For data read and write the data can be stored to each individual stack at a rate based upon the frequency of the rotating magnetic field. The total data rate of a data storage device incorporating multiple stacks can be further increased by reading/writing in parallel to multiple stacks.

For generation of the external propagation field, a number of options can be utilised. For a small stack on a convention bit-line/word-line matrix, the applied field can be applied using low magnitude dephased pulses on these lines. For such arrangements, and for arrangements where this is not technically possible due to the design of the bit-lines and word-lines or the size of the stack, other approaches can be considered.

One efficient way to generate the rotating magnetic field that propagates solitons through the layered structure is by a pair of current carrying conductors oriented at right angles to each other. Various arrangements can be used to implement such a system, including: a pair of coplanar waveguides, as described in P. Martin Pimentel et.al. "A new crossed coplanar waveguide design for ultrafast magnetization switching utilizing polymer insulation layers", Appl. Phys. Lett. 88, 122510 (2006) doi: 10.1063/1.2186947;

a pair of current-carrying strip lines fabricated either on the same chip as the data- storing layered structure, or on a separate chip that is flip-chip bonded to the data-storing layered structure chip;

a pair of planar coils fabricated by MEMS technology, either on the same chip as the data-storing layered structures, or on a separate chip that is flip-chip bonded to the data- storing layered structure chip;

· two pairs of external coils; or

two pairs of external coils wrapped around a ferrite ring

The two field generators may be powered by a cosine and sine wave of current, thus forming a rotating field. However, in order to reduce power dissipation, they may also be powered by orthogonal current pulses, since soliton propagation does not require a smooth rotation of applied field.

Thus there have now been described a number of examples for generation of a rotating magnetic field to drive the propagation of solitons through a layered structure.

Thus a number of examples of a field driven device for propagation of solitons through an inherently chiral layered structure has been provided. It is also possible to create a device based upon spin momentum transfer to create a current driven device. Such a current drive device would be arranged to pass an electrical current along the length of the layered structure. Each magnetic layer in the structure causes the conduction electrons to become spin polarised. As the spin polarised current leaves one layer and enters the neighbouring layer, a torque is created due to the difference in direction between the magnetisation direction of the layer being entered and the spin of the conduction electrons. Within regions of constant order parameter, although there is a large angle between the conduction electron spin and the magnetisation direction of the layer being entered (i.e. 180 degrees - anisotropy rotation offset angle per layer), the torque is small since torque is maximised for 90 degrees between these directions. Consequently, the spin transfer effect has little effect. Conversely, at a soliton (i.e. a region of transition of the order parameter), the angle between the spin of the conduction electrons and the magnetisation of the layer being entered is closer to 90 degrees and so a strong torque is exerted on the soliton. This has the effect of propagating it asynchronously along the stack. Within each stack, data can be encoded according to one of a number of possible schemes.

Two specific example encoding schemes are illustrated with reference to Figures 13a, 13b and 13c. The first scheme uses the order parameter to represent data values. As such all regions having a first order parameter are treated as carrying a data value of 1 and all regions having the opposite data parameter are treated as carrying a data value of 0. Each predetermined number of layers having the given order parameter corresponds to a single data bit. The second scheme uses the presence or absence of a soliton to represent data values. As such each predetermined possible soliton position can represent either a data value of 1 or 0 depending on whether a soliton is present or not. These encoding schemes are illustrated in Figures 13a, 13b and 13c.

Figure 13a shows schematically a layered structure in which a number of regions of different order parameter (value 1 or -1) are defined by the presence of a number of solitons.

Figure 13b shows the data content as present according to the first encoding scheme, where the value of the order parameter encodes the data. In this example an order parameter value of 1 corresponds to a data value of 1 and an order parameter value of -1 corresponds to a data value of -1 , although the reverse encoding (1→0, -1→1) is also possible. Thus, as is seen, the data carried by the stack according to this encoding scheme provides a data value of 1 for every inter-possible-soliton-position region that carries an order parameter value of 1 and a data value of 0 for every inter-possible-soliton-position region that carries an order parameter value of -1.

Figure 13c shows the data content as present according to the second encoding scheme, where the presence or absence of a soliton encodes the data. In this example a soliton presence at a possible-soliton-position corresponds to a data value of 1 and soliton absence at a possible-soliton-position corresponds to a data value of -1 , although the reverse encoding (soliton present→0, soliton absent→1) is also possible. Thus, as is seen, the data carried by the stack according to this encoding scheme provides a data value of 1 for every possible-soliton-position that carries a soliton and a data value of 0 for every possible-soliton- position that does not carry a soliton.

Another suitable encoding scheme would be to use an exact copy of existing hard disc coding schemes. Thus a 1 to -1 order parameter transition in the stack can be taken as equivalent to head to head in hard disk encoding and a -1 to 1 order parameter transition in the stack can be taken as equivalent to tail to tail in hard disk encoding (or vice versa). By taking such an encoding approach, existing hard disk drive turbo codes can be recycled for use in a solid state magnetic data store. Another suitable example of an encoding scheme is one utilising phase shift keying. This example has application in a situation where neighbouring stacks of a device are close enough that interactions between adjacent stacks are a possible source of erroneous behaviour. In this approach, the mark space ratio is controlled and the overall structure of soliton presence or absence between neighbouring stacks is approximately controlled. In one specific example, the encoding used could utilise a soliton at every fourth position (position 1 of every group of four positions) and an encode data values by inserting a soliton at position 2 for data value one or position 3 for the data value zero (or vice versa). In this example, position 4 would always be empty of a soliton Such an encoding example provides that half of the possible soliton positions have a soliton present and thus smoothes the presence or absence of solitons in the event of a large number of sequential data bits of the same value.

Thus there have now been described a possible data storage structure utilising soliton holding layered structures with embedded chirality and data encoding schemes by which solitons within the layered structure can encode data for storage and later retrieval.

Having considered possible techniques for data storage using a layered structure with a spiral pattern of easy axes of anisotropy over successive layers, there are now considered some properties that set out the individual magnetic layer sizes and the data storage densities that can be achieved.

The properties that govern the scaling of disk sizes can be described by the following set of equations. In these equations, it is assumed that each magnetic layer is a circular disk of diameter w and thickness t and that the layered structure contains N magnetic layers and the total thickness of the structure is thus Nt+(N-1)t _in suiatingiayer- However it is possible to ignore tinsuiatingiayer (the thickness of the interlayer spacer) as this can be assumed to be small compared to the magnetic layer thickness. The equations can be adjusted for alternative magnetic layer geometries.

The magnetic energy of each disk is K _u w ² t (where K _u is the anisotropy energy density of the material making up the magnetic disk) and so the energy barrier separating propagation states ΔΕρ is n _, K _u w ² t where η is an efficiency factor that links the anisotropy energy barrier to the soliton propagation barrier. The value of η is believed to approach 1 for abrupt (narrow) solitons and to fall as soliton width increases. For the purposes of the present examples, η can be taken to have a value of 0.75 or greater.

As has been discussed above, in order to provide long term stability of solitons within the layered structure, a minimum value of ΔΕρ is established to provide the desired level of stability. In the present discussion, a value of ΔΕρ that is believed to provide 10 year stability is selected, such that r|K _u w ² t =60k _B T.

For the following discussion, parameter D is defined to be the effective areal density and n to be the minimum soliton spacing (leading edge to leading edge or tail to tail). Therefore D=N/w ² n.

Combining these enables the expression of the effective areal density in terms of the disk thickness gives:

ΝίηΚ

60 k _B Tn with the disk width given by:

[2] can be rewritten as: _{D =} ^hr l ^K «

60k _B Tn ^[4] where h is the total thickness of the stack. In other words, the overall density doesn't depend on whether the structure uses lots of thin layers or a few thick layers. What does change with individual disk thickness is the required lateral dimensions of the disk.

Lithographic limits may impose a maximum aspect ratio r, such that h=r w. In some examples, it can be appropriate to assist the formation of a high aspect ratio (many layered) structure by using a lithography size (layer size) three or four scaling generations behind the smallest commercially available lithography size. Equation [4] can therefore be rewritten (ignoring the extra thickness of the interlayer spacer which is assumed to be small compared to the magnetic layer thickness) r ^{W r} l

60k _B Tn ^[5]

This means that the areal density grows linearly with the size of the disk. This is because wider elements can be thinner at constant energy and so it is possible to have more layers in a given height of layered structure and the structure itself can be higher if the disk is wider due to lithographic aspect ratio limitations.

The density can be expressed in terms of t instead of w by substituting [3] into [5]. In many cases the ultimate limit on D is set by the minimum possible value of t. Thus:

D = r ^■ [6]

60k _B Tnt _l

And

60k _R T

w max [7]

Alternatively, if the boundary of acceptable size is governed by T _min and total stack thickness rather than T _min and lithography constraints, then equation [4] gives the maximum density and equation [7] gives the disk size needed given the minimum layer thickness.

Some illustrative example numbers derived using these equations are now provided. Starting with the disk width and using equation [7], taking values of K _u =4.8x 104 erg-cm ^"3 (which equivalent to 80Oe anisotropy field Hu, assuming that the saturation magnetisation strength of the material making up the disc, Ms, =1.5T), t _min =5nm and η=0.75, then: max = 1.2 x 10 ⁵ cm = 12 nm

0.75 x 4.8 x l0 ⁴ x 5 x l0 ^"7

If it were desired to make the value of w _max larger to enable a coarser lithography or to provide a greater planar area to increase the density of power dissipated during field-pulse- induced soliton injection (discussion of soliton injection mechanisms is presented below), then in order to increase the disk area without a loss in density it is necessary to reduce t _min without losing the anisotropy properties of the layer. For example, reducing t _min by a factor of 10 would increase w _max by a factor of 3.2, such that in this example w _max would rise to 380nm.

Looking now to examples for the total density and keeping the values of K _u =4.8x 104 erg-cm ^"3 (equivalent to 80Oe anisotropy field assuming Ms=1.5T), η=0.75, and taking values of ϊι=10μηι and n=4, then using equation [4]:

_ 10 x lQ- ⁴ x 0.75 x 4.8 x l0 ⁴ _{12 2} , ₂

D = 7 = 3.6 x 10 bits I cm = 23 TBits I inch

60 x l .38 x l0 ^"16 x 300 x 4 Thus, for a device utilising layered structures made of 2000 magnetic layers and storing data bits at a density of 1 bit per four layers, where each stack of magnetic layers is 120nm in diameter and is 10μηι in total thickness, a data density of 23 TBits/inch ² is possible. For comparison, current generations of hard disks are at around 0.5 TBits/inch ² while solid state memory such as FLASH memory is around 0.1 TBits/inch ² .

As mentioned above, an even thicker (more layered) stack would provide for even greater storage capacities and, as increasing the stack thickness doesn't increase the footprint, the areal density scales with in increase in stack thickness. Thus a doubling of stack thickness to 20μηι would double the density and a ten-fold increase in stack thickness to 100μηι would result in a ten-fold increase in density.

Thus there have now been described some properties relating to the storage densities achievable within a layered structure having alternating magnetic and non-magnetic layers and where the easy axis of anisotropy of each magnetic layer is offset from that of its neighbouring magnetic disks such that a chiral property is imparted to the structure.

As mentioned above, a data storage device can store data bits by injecting one or more solitons into the layered structure and propagating them along the layered structure. The soliton(s) would then remain in the layered structure until required, at which point it would be either be propagated through to the other end of the layered structure and detected as it leaves (a First In First Out serial shift register) or the rotating field direction would be reversed and the data sequence would be propagated out of the same end that it was introduced, in reverse order (a First In Last Out serial shift register).

In either case, it is necessary to inject controllably at one end of the layered structure a sequence of solitons representing the data bits to be stored. Suitable coding mechanisms that could be used have been discussed above. In the present examples, it is most likely that the data sequence for storage will begin in electrical form, and that this must be converted to magnetic form by some injector device at one end of the layered structure. A suitable injector device will now be discussed with reference to Figure 14.

Figure 14 shows a very simple form of an injector. The layered structure 3 is fabricated on top of a short current-carrying strip line 30 on a CMOS integrated circuit. The two ends of the strip line are attached to metallic vias 32 which connect to transistors within the main body of the CMOS circuit. The transistors control current pulses 34 through the strip line, which in turn generate localised pulses of magnetic field 36 which predominantly affect the magnetisation of the nearest layers of the structure 3. The strength of the magnetic field from a current-carrying strip line falls off above it on a lengthscale approximately equal to the width of the strip line itself. For a layered structure having a 7nm repeat period for the stack, a 100nm wide strip line would therefore generate a magnetic field across the first 14 repeat layers or so, which is more than is necessary to inject a soliton from the end of the layered structure. However, the very end layer of the structure 3 is easier to switch magnetically than the others near it, since it only has one nearest neighbour disk providing a stabilising coupling field - all of the other disks (except the disk at the other end, which is too far away from the strip line to be influenced) have two stabilising nearest neighbours. By a correct choice of current amplitude to induce the pulse 34, it is possible for the current-carrying strip line to switch only the bottom magnetic disk even though the magnetic field from it extends higher up the layered structure.

As will be apparent from the above discussion, it is in fact not necessary to inject a soliton having any given handedness as the embedded chirality of the layered structure will cause the soliton to align to the embedded chirality once propagated. However, it is good practice to intend to align the handedness of an injected soliton to that of the embedded chirality. The injector device therefore also controls the chirality of the injected soliton. According to the present examples, this can be achieved by control of the timing of the strip line current pulse with respect to the externally applied rotating propagation field phase.

If these conditions are controlled such that the strip line current is fired at the instant that the propagation field is parallel to the direction of the strip line field, this maximises the strength of the field at the bottom disk since the propagation field and the local strip line field are reinforcing each other to the maximum degree. In this circumstance, the two relevant parameters are (i) magnitude of the current in the strip line and (ii) the phase angle of the externally applied rotating field at which the strip line is fired.

The power dissipated during the injection process by Joule heating of the stripline can be minimised by one or more of the following approaches: (i) using as short a current pulse as possible (typically of the order of 10ns duration); (ii) using as low a frequency as possible for the external rotating field (thereby reducing the duty cycle of the strip line current); and (iii) cladding the outside of the strip line with a magnetic material to concentrate the magnetic flux lines generated around it, as described in US 6,724,652. It will be appreciated that approach (ii) runs counter to achieving high data throughput rates and so this approach may be selected not to be implemented where a high data rate is required, with heating avoidance and/or mitigation being provided by either or both of approach (i) and (iii). Thus a mechanism and example arrangement for introducing solitons into a layered structure with embedded chirality in the easy axes of anisotropy between layers have now been described. As will be appreciated from the foregoing, one use of this technique is for data storage. In the following, a mechanism and example arrangement for reading solitons from such a structure will now be described.

As mentioned above, the options for using a layered structure having solitons introduced thereinto for data storage purposes are a First In First Out (FIFO) shift register (where reading occurs at the opposite end of the stack to writing) and a First In Last Out (FILO) shift register (where reading and writing occur at the same end of the stack). Other options include the pseudo-persistent storage approach mentioned above where two stacks are paired to provide the effect of a read/write element in the middle of the stack.

The present example assumes that a First In Last Out (FILO) shift register is provided such that the data are injected into the bottom of the soliton stack and shifted up the stack during writing; for read-back, the rotating field is reversed and the data are shifted back down the stack and out of the bottom in reverse order). In this arrangement it is necessary to measure the magnetisation state of the lowest magnetic layer. Efficient methods to achieve this include a magnetoelectronic method to convert the magnetisation direction of the lowest magnetic layer into either a change in electrical resistance or a change in voltage. Methods for doing this include a giant magneto resistance spin valve and a tunnel magneto resistance structure.

The Giant Magneto Resistance (GMR) spin valve approach uses a technique described in, for example, US 4,949,039. An example of a GMR spin valve structure as applied to the soliton storing layered structure as described above is shown in Figure 15.

This implementation uses the end magnetic layer 50 of the layered structure 3 closest to the reading/writing element 5 to form the free layer in a Giant Magneto Resistance spin valve 58. The GMR spin valve has four layers, a pinning non-magnetic layer 56, a pinned magnetic layer 54, a further magnetic layer 52 and the free layer 50 (provided by the end magnetic disk 50 of the stack 3). An electrical current is passed through the pinned magnetic layer 54 which is in electrical contact via the non-magnetic spacer 52 with the end magnetic disk 50 in the layered structure 3. Some of the electrical current leaks between the pinned magnetic layer 54 and the bottom magnetic disk 50, leading to spin-dependent scattering and hence an electrical resistance for the combined circuit that depends on the relative orientation of the magnetisation in the bottom magnetic disk in the stack and the magnetisation in the pinned magnetic layer. The bottom layer 56 of the spin-valve structure 58 can double up as the current-carrying strip-line used for writing (see element 30 in Figure 14).

The first non-magnetic spacer layer 51 within the stack may be made from an electrical insulator to prevent current from leaking into higher magnetic layers in the SAF stack.

Thus the use of a GMR spin valve to write solitons into and read solitons out of a layered structure with embedded chirality has now been described.

An alternative reader structure based upon a Tunnel Magneto Resistance (TMR) structure will now be described. Such a structure provides a high usable magnitude of the read-out signal, thus enabling a high density arrangement of stacks on the circuit which supplies the data to the layered structures (such as CMOS circuit 24). A TMR structure is more complex to fabricate than the GMR spin valve mentioned above, as it requires an electrical connection to be made above the bottom magnetic layer in the layered structure. In this example the bottom of the layered structure is isolated from a pinned ferromagnetic layer by a tunnel barrier (e.g. a thin magnesium oxide layer, see, for example, S.S. P. Parkin et.al. "Giant tunnelling magnetoresistance at room temperature with MgO (100) tunnel barriers" Nature Materials 3, 862-867 (2004) doi: 10.1038/nmat1256) and an electrical current is passed between the bottom magnetic layer of the layered structure and the pinned ferromagnetic layer via the tunnel barrier. The TMR effect gives a very strong dependence of resistance on the relative orientation of the magnetisation in the two magnetic layers, allowing the magnetic state of the bottom of the layered structure to be easily detected.

According to the present examples, the electrical current can either be localised to flow through only a small number of magnetic layers within the layered structure by applying side- contacts to the layered structure, or, it may flow through all of the magnetic layers within the layered structure by applying a top-contact to the layered structure (which can be easier to fabricate). If a top-contact approach is used, the magnetic detection can be localised to any point within the layered structure by using electrically conducting non-magnetic spacer layers throughout the layered structure, except at the point where the soliton is to be detected where a tunnel barrier layer is used instead. Even though many layers of the layered structure are involved in the passage of the electrical current, the electrical resistance is dominated by the tunnel barrier and the spin-dependence of the resistance of the layered structure is dominated by the relative orientation of the magnetisation on either side of the tunnel barrier.

Figure 16 shows an example implementation of this principle. The TMR structure 68 includes a pinning non-magnetic layer 66, a pinned ferromagnetic layer 64, a tunnel barrier layer 62 and the bottom magnetic disk 60 of the layered structure 3. In this example, the tunnel barrier 62 is placed between the bottom magnetic layer 60 of the SAF 5 and a pinned reference layer 64 outside of the layered structure. The inter-disk spacer layers 69 of the layered structure 3 are provided from electrically conductive material which causes the electrical resistance through the structure a a whole to be dominated by the tunnel barrier layer 62. This leads to an electrical resistance through the SAF which depends on the direction of magnetisation of the bottom magnetic layer 60 in the layered structure.

Figure 17 shows an example where the tunnel barrier 62 is placed between the bottom 60 and second from bottom 61 magnetic layers of the layered structure 3 and there is no pinned reference layer outside of the layered structure. Again the non-magnetic spacer layers of the layered structure (except the tunnel barrier layer 62) are of electrically conductive material. This leads to an electrical resistance through the device as a whole which depends on the relative orientation of the magnetisation in the bottom and second from bottom layers in the layered structure. This would change as a soliton passes through: when any soliton is far away these two layers have approximately 170° between their magnetisation (substantially anti-parallel coupling with the offset for inherent chirality of the layered structure); and when a soliton is across the tunnel barrier, these two layers have approximately 90° between their magnetisation. Thus this example provides for a transition detection rather than a state detector.

A further variation on this principle is shown in Figure 18 where the tunnel barrier layer 62 is placed at the top of the layered structure between the top 60a and second from top 61a magnetic layers in the layered structure 3. As in Figure 16, this allows transition detection of a soliton passing through the tunnel barrier. Because this detector is at the top of the layered structure stack, data can be read out without reversing the sense of the externally applied rotating field, making a First In First Out (FIFO) shift register. This also allows data to be written and read at the same time.

In the described arrangements using a TMR detection system, the soliton injection arrangement is omitted for simplicity of understanding. A separate (or partially integrated) injection arrangement can use a strip line type arrangement as discussed above.

A variation on the TMR detection system which does not require the fabrication of a top contact involves forming a conventional magnetic tunnel junction (MTJ) stack, as is currently used for MRAM, and placing the layered structure stack on top of the free layer, in full ferromagnetic contact as shown in Figure 19. The MTJ is formed from a magnetic tunnel junction free layer 70, a non-magnetic tunnel barrier layer 72, a pinned ferromagnetic layer 74 and a pinning antiferromagnet 76. The bottom magnetic layer 60 of the layered structure is in full ferromagnetic contact with the MTJ free layer 70.

In this arrangement, no electrical current passes through the layered structure 3 itself and the output is determined purely by the relative orientation of the magnetisation in the MTJ free layer 70 and the magnetisation in the MTJ pinned layer 74. However, the ferromagnetic contact between the MTJ free layer 70 and the magnetic layer 60 at the bottom of the SAF stack results in strong coupling between the magnetic state of the free layer 70 and the bottom layer 60 of the layered structure 3. Thus whenever a soliton switches the bottom layer 60 of the layered structure 3, the free layer 70 of the MTJ also switches. It is expected that improved performance is achieved when the free layer of the MTJ is thinner than the bottom magnetic layer of the layered structure.

It will be appreciated that although the layered structures shown in the examples set out above include only a small number of layers in the layered structure 3, this is for the purposes of making the figures clear and easily understandable, and, as discussed above, each stack may include a much larger number of layers, for example 100 to 100,000 or even more layers..

It will be appreciated that arrangements could be constructed with reading elements at both ends to facilitate use as a FIFO and a FILO shift register. Reading and/or writing elements could also be placed at a location other than the end of a stack. One example of such an arrangement could have reading and writing elements in the middle of the stack, so as to provide that data is written when applying a rotating magnetic field in a first direction to enable half of the stack to be used for storage. Data would then be read by reversing the magnetic field direction. As noted above, an arrangement with the read element in the middle of the stack could provide that, if a non-destructive read-method is utilised, the stored solitons pass into the second half of the stack and thus are maintained in memory after reading. In this way a persistent memory can be provided where reading the data does not cause deletion thereof. As mentioned above, this effect can also be achieved by use of a cooperatively controlled pair of stacks where one layered structure has written into it the data read from the other.

It will be appreciated that although the majority of layers of a layered structure as described above use dominant RKKY coupling to achieve negative coupling (i.e., tending toward anti- parallel coupling in so far as the rotational offset between layers allows), there may be specific layer pairs in which the RKKY coupling provides positive coupling (i.e. tending toward parallel coupling). Examples of such locations where positive coupling is used could include the interface to a GMR or TMR read-out element. This positive coupling could come about because insulator spacer layers (as mentioned above as a way of keeping the sense current from leaking into higher layers) have a positive (or zero) RKKY coupling. So a device might have negative coupling throughout the layer structure except at an end layer where positive coupling occurs.

Thus a number of examples of layered structures with embedded chirality enabling the writing of solitons thereinto and the reading of solitons therefrom have now been described. A large number of different data storage devices can be implemented using such structures and including any number of individual layered structures.

It will be appreciated that references to a stack or column refer to specific examples of a suitable layered structure and no particular orientation or aspect ratio of such a layered structure is implied by use of either the term stack or column. It will also be appreciated that reference herein to the "top" or "bottom" of elements such as the stack or column are references to the orientations shown in the Figures and that the devices and arrangements described herein may be inverted or tilted by any angle in any plane without affecting the operation thereof and thus the "top" and "bottom" can be considered as ends according to the particular orientation of the device at a given time.

The skilled reader will appreciate that the various described arrangements for a column of RKKY coupled magnetic discs with embedded chirality which can maintain an introduced soliton therein and have that soliton propagated therethrough by an externally applied rotating magnetic field and have solitons written thereinto and read therefrom are examples which illustrate the concepts underlying the present invention. Various modifications, alterations and equivalents may be employed without departing from the spirit and scope of the present invention.