Magnetic material structures, devices and methods
Abstract
This record has no abstract on file.
Term
Term ended
Expired 29 August 2022, 4.1 years ago.
- Priority
- Filed
- Granted
- Expired
- Today
3 claims: 3 independent, 0 dependent
- 1基板と、一軸対称性 が破ら れた構造を形成する少なくとも1つの磁性bcc-d層と、前記基板と前記 少なくとも1つの磁性 bcc-d層とのあいだに配置された(111)テクスチャ を有する 六方晶原子テンプレートを構成する少なくとも1つの層とからなり、前記 少なくとも1つの 磁性bcc-d層 は、 (110 )テ クスチャを有し、かつ前記(111)テクスチャ の 六方晶原子テンプレート上にエピタキシャル され 、前記 少なくとも1つの 磁性bcc-d層 は 、 2以上で4 以下の優位なバリアント を有し 、前記 少なくとも1つの磁性 bcc-d層がFeまたはFeCoもしくはFeを含む材料を含んでなる磁性体構造。
- 2基板と、一軸対称性 が破ら れた構造を形成する少なくとも1つの磁性bcc-d層と、前記基板と前記 少なくとも1つの磁性 bcc-d層とのあいだに配置された(111)テクスチャ を有する 六方晶原子テンプレートを構成する少なくとも1つの層とからなり、前記 少なくとも1つの 磁性bcc-d層 は、 (110 )テ クスチャを有し、かつ前記(111)テクスチャ の 六方晶原子テンプレート上にエピタキシャル され 、前記 少なくとも1つの 磁性bcc-d層 は 、 2以上で4 以下の優位なバリアント を有し 、前記 少なくとも1つの磁性 bcc-d層がFeまたはFeCoもしくはFeを含む材料を含んでなる、磁性体構造を組み込んだ磁気デバイス。
- 3(111)テクスチャ の 六方晶原子テンプレートを提供すること、および、一軸対称性 が破ら れた構造を 有 する少なくとも1つの磁性bcc-d層を 形成 することからなり、前記 少なくとも1つの 磁性bcc-d層 は、 (110 )テ クスチャを有し、かつ前記(111)テクスチャ の 六方晶原子テンプレート上にエピタキシャルし、前記 少なくとも1つの 磁性bcc-d層 は 、 2以上で4 以下の優位なバリアント を有し 、前記 少なくとも1つの磁性 bcc-d層がFeまたはFeCoもしくはFeを含む材料を含んでなる、磁性体構造の製造方法。
Independent claims3
120 paragraphs, as filed
The present invention relates to a magnetic material structure, a method for forming the magnetic material structure, and an apparatus formed from the magnetic material structure.
As homes, offices, transportation systems, offices, and factories become more automated and electronically connected, more computers, communication equipment, wireless communication equipment, electronic game machines, entertainment systems, mobile information terminals, transportation vehicles High-performance and low-cost electronic circuits, sensors, transducers, data storage systems and magnetic thin film materials as electronic devices and electrical products such as manufacturing tools, store tools, and household appliances become more sophisticated. Demand for other magnetic devices that use the system has increased and is expected to continue to increase. In order for these devices to remain competitive in the market, each product type must be more sophisticated, unobtrusive and usually cheaper than its predecessor. Therefore, there is a greater demand for technical improvements in materials and structures for these devices.
For all these applications, the performance of magnetic materials is improved if the magnetic properties can be better controlled during construction. Two known properties that are often considered as inherent magnetic properties are the saturation magnetization Ms and the magnetocrystalline anisotropy energy density constant (usually represented by the K letter in the subscript). Magnetic anisotropy energy means that magnetization has a preferred direction, or orientation or orientation in several directions. That is, the energy of the system is minimal when the magnetization vector is oriented in a certain direction. These directions are called the magnetized easy axes, and conversely, the magnetized hard axes are the directions of magnetism where the energy is maximized. However, it should be noted that magnetic anisotropy is not really an intrinsic property in general in that the material is not perfectly formed. Nevertheless, good performance in device applications often depends on a single preferred magnetic orientation or anisotropy orientation, thus achieving the desired uniaxial anisotropy during the manufacturing process. Is important. An object of the present invention is to provide a new mechanism for controlling the magnetocrystalline anisotropy of a magnetic thin film. This is expected to improve the performance of almost all magnetic devices.
In general, anisotropic energy is a function of the orientation of the magnetization vector with respect to a given physical axis. Here, "uniaxial" anisotropy is defined as being present when the magnetization angle θ is rotated 180 degrees from the physical axis and the anisotropy energy density function contains only one maximum and one minimum. .. Similarly, the "ideal uniaxial" anisotropic energy has an energy equation of sin.<sup>2</sup>θ or cos<sup>2</sup>It is defined as present when it has only θ dependence. Materials and equipment processed to achieve the desired orientation or anisotropy are generally difficult and impossible, probably because the mechanism for achieving the orientation of anisotropy has not been well understood. There is even. Moreover, uniform control of the orientation of the magnetic anisotropy is often difficult to achieve and maintain during the manufacturing process when a number of different desired material properties must be obtained simultaneously.
[Background of oriented soft magnetic film] For example, in magnetic devices such as sensors, transducers, transformers, inductors, signal mixers, flux concentrators, recording medium keepers, data recording and playback transducers, the magnetic response to the driving magnetic field has high sensitivity and low coercive force (Hc). Is common. Simply put, the material inherently has non-hysteretic behavior. For such behavior, the device is configured such that the applied magnetic field is directed along the difficult axis of magnetization of the uniaxial magnetic material. This minimizes the coercive force and hysterical effects often associated with domain wall movement of multiaxial materials. For example, a material with biaxial anisotropy has two easy-to-magnetize axes and two difficult-to-magnetize axes, exhibiting hysteresis and loss. For many of these applications, a linear or substantially linear response is also advantageous, but for other applications such as signal mixers, a controlled non-linear response is desirable. To obtain a linear magnetic response, apply a magnetic field in the direction of the difficult axis of magnetization, and the anisotropic energy density function is not only uniaxial, but also a single sin.<sup>2</sup>(θ) or negative cos<sup>2</sup>It is necessary to have a (θ) dependence, where θ is the angle measured between the magnetization vector direction and the physically determined easy axis of magnetization. Mathematical identity sin<sup>2</sup>(θ) = 1-cos<sup>2</sup>By arbitrarily determining (θ) and the origin in the energy function, sin<sup>2</sup>(θ) or -cos<sup>2</sup>Equivalent physical behavior can be obtained by using (θ). [1] of FIG. 1 is a diagram showing the angle of the magnetization vector with respect to the easy axis of magnetization located at the square vs. zero degree of the sinusically anisotropic energy density. FIG. 2 is a diagram showing the response of the components of magnetization Mx and My as a function of the magnetic field Hx = Ha applied along the direction x of the magnetization difficulty axis. The straight line is refracted only at the position where the magnetization is saturated, or at the position where it is completely in line with the applied magnetic field [2]. For such a particular uniaxial anisotropy, the value Hk of the magnetic field applied along the x direction, known as the anisotropy, arises. These response curves do not show hysteresis, but are often referred to as hysteresis loops. sin<sup>2</sup>(θ) Due to the shape of the energy function, the response Mx in the direction of the difficult magnetization axis is linear and completely reversible. My is the response in the y direction to the magnetic field applied in the x direction. The curve shape shown is a quadratic curve when the magnitude of the applied magnetic field is less than Hk, and My is zero for larger magnetic fields. Secondary behavior is Ms<sup>2</sup>= Mx<sup>2</sup>+ My<sup>2</sup>Therefore, it is necessary for the straight line Mx, where Ms is the magnitude of the total constant saturation magnetization vector. Anisotropic energy is uniaxial but ideal sin<sup>2</sup>(θ) If not determined by the form of the function, the magnetic response is not linear. However, no practical material examples have been known that show both uniaxial energy curves and non-linear Mx vs. Hx behavior.
sin<sup>2</sup>The material showing the form of the (θ) energy density function is often referred to as the Stoner-Wohlfarth behavior by the well-known uniaxial single domain magnetization theory. However, the thin film is locally sin<sup>2</sup>(θ) Although it may show the form of a function, it is generally a multimagnetic domain. Domain wall movement can generally be observed as long as the magnetization difficulty axis direction is the same in all respects in the sample and the applied driving magnetic field is not reliably parallel to the magnetization difficulty axis. This movement results in a coercive force mechanism and hysteretic energy loss. The lossless behavior of the sample represented by FIG. 2 is due to the magnetization rotating in response to the applied magnetic field rather than the response via domain wall movement. The multi-axis anisotropic material is constantly switched via domain wall movement and therefore does not experience any loss.
For uniaxial soft thin films, when the hard-to-magnetize axis is driven by first applying a magnetic field in the direction of the easy-magnetization axis and then holding a constant bias magnetic field in the same direction to remove the 180-degree domain wall. It is well known that all materials can appear to be in a single domain. Therefore, due to this bias magnetic field Hb = Hy in the direction of the easy axis of magnetization, the application of all finite magnetic fields Hx in the direction of the difficult axis of magnetization cannot completely drive the magnetization vector to the maximum energy value [3], and the response is It is always reversible and therefore lossless. This is not the case for materials with polyanisotropic axes. For uniaxial materials, rotary response is important for many sensor devices, replacing the magnetic sensor material with a hard magnetic material oriented by applying a small magnetic field or to provide an effective bias magnetic field. By coupling, it is common in various forms of magnetic resistance sensors for applying a bias magnetic field in the direction of the easy axis of magnetization.
For some sensor applications such as security devices, and for certain electronic mixing circuit devices, soft, low-loss, magnetic properties are desirable at the same time as certain non-linear responses. In these applications, the driving magnetic field has traditionally and generally been oriented in the direction of the easy axis of magnetization or in the direction of the lowest magnetic anisotropy energy. Usually, the domain wall movement is remarkable in the same direction. Domain wall movement generally results in very non-linear responses or even strong hysteretic behavior.
Some security devices, archle monitoring devices, archle identification devices or inventory devices rely on detecting non-linear behavior or the tuning signal generated when the material is driven to saturation. An example of a number of these types of surveillance systems and tags is disclosed in US Pat. No. 3,747,086. This type of tag response is disclosed as providing a large number of bits of information to identify the object described in US Pat. No. 5,538,803. Other tag devices are based on the magnetic elastic effect and mechanical resonance, where there is a coupling between the magnetization and the mechanical strain of the material. An example of this type of tag is disclosed in US Pat. No. 4,510,489. In these devices, it is desirable to drive the magnetization in the direction of the difficult magnetization axis so that the rotation of the magnetization becomes dominant and the magnetic hysteretic loss is minimized. By using this mode and by using a magnetic field to drive the device at its mechanical resonance frequency, a significant amount of energy is stored in the device. Therefore, even after the driving magnetic field is removed, the mechanical vibration continues and the magnetic elastic properties reverse the process of transmitting the magnetic field because the mechanical stress oscillates the magnetization vector direction synchronously with the mechanical vibration. .. The dependent magnetic dipole then emits a magnetic field at a definite resonance frequency that can be detected to confirm the presence of the tag. Uniaxial anisotropy is required in such devices to achieve low loss. Due to the magnetic elastic coupling between the mechanical strain and the magnetic moment orientation, the fundamental wave of the mechanical resonance frequency is emitted as an oscillating magnetic dipole. However, if uniaxial anisotropy is non-ideal and results in a non-linear response, the tuning will be detectable in the presence of the drive signal. This is advantageous, but the available materials that form such devices have not been known so far.
Similarly, for analog mixing circuitry, a non-linear response is desirable. When two separate sinusoidal signals are simultaneously fed to the components of a non-linear circuit device, the multiplication process results in beating of the two signals. This causes further tuning at the sum and difference frequencies of the first signal. Therefore, the information contained in the modulated carrier frequency signal can be deviated to the beat frequency. Typically, this shifts the information-bearing bandwidth to a higher carrier frequency bandwidth (modulation), or the information-bearing bandwidth signal is close to zero frequency for demodulation, or by zero frequency. This is done to bring it back to a closer bandwidth. Frequency shift techniques are common in telecommunications and signal processing, as well as many other signal processing applications. Non-linear circuit responses, circuit components, and the execution of circuits used to perform these signal mixing processes are described in numerous electrical engineering circuit textbooks. For example, "Electronic of Communications Techinique" by PHYoung and SM There are two books, "Physics of Semiconductor Devices" by SMSze. Ideally, the non-linear device used to mix the signal should be efficient, low loss, and low noise. Usually, a non-linear silicon activator is used. When the magnetic device was used for the same application as before, the domain wall movement caused a loss to the signal and caused noise. A magnetic material with a non-linear response that operates with a low loss magnetization rotation rather than domain wall movement is highly desirable.
The use of cubic materials is common for almost all transducer and sensor applications that require low anisotropy values to provide high sensitivity. However, obtaining a single axis of anisotropic energy density, which is almost always essential to obtain the desired low loss, low noise, and magnetic properties, is difficult due to its trifold crystal symmetry of the cubic material. is there. Thin or thick film materials are widely used. For example, many devices, such as data storage and reproduction transducers or magnetic field transducers, generally use a thin film material of face-centered cubic fcc. When this material is prepared with a (001) crystal texture or texture, the anisotropic energy has multiple magnetic domain easy axes and magnetizations on the film surface because the angle function on the film surface is biaxial. It has a difficult axis and produces a non-linear and hysteretic magnetic response, resulting in a noisy signal. Therefore, a (111) crystal texture is preferred, where the (111) crystal texture is mathematically shown when the magnetization is limited to the (111) texture plane, and the projection of the crystal in the (111) crystal direction with respect to the film plane. Due to tri-fold symmetry, the cubic material has no net first-order anisotropic energy density. Even in the case of moderately saturated magnetized thin films, magnetization is essentially limited to the membrane surface due to the demagnetizing force associated with the flat membrane shape.
(111) In order to achieve a single uniaxial anisotropy in a cubic material with a textured surface, the material is usually heat treated or deposited directly in the presence of an applied magnetic field. The resulting single domain easy axis is aligned with the direction of the applied magnetic field, and the magnetization difficulty axis is perpendicular to the applied magnetic field. Furthermore, the Mx vs. Hx response function is a linear response. While the mechanism for the cause of this induced magnetic anisotropy is not well understood, it is often argued that an atomic pair rule mechanism arises to break the directional symmetry in the (111) plane with respect to each grain of material. ing. That is, on a local scale within each grain, the pair of atoms are aligned along the magnetic field applied during deposition or annealing, reducing magnetic field energy. Interestingly, annealing in differently oriented magnetic fields often redirects the induced anisotropy, indicating that the orientation-inducing mechanism is reversible. It is believed that this local rule allows fcc materials such as NiFe alloys and permalloys to have small induced uniaxial anisotropy. Body-centered cubic crystals bcc of magnetic thin films, or bcc derivatives of Fe, FeCo, FeAl and similar configurations are such applications because the required (111) texture is not formed during the deposition of bcc symmetric crystals. It is rarely used by people.
It will be understood that certain texture orientations tend to appear during the growth of the metal thin film. This is driven by surface energy and surface bond minimization and is modified by surface mobility items that can be somewhat controlled by the substrate and process conditions. One simple empirical rule is that membrane surface energy is minimized when the atomic surface structure is most tightly packed. For fcc crystals, the crystals on the surface are most densely packed in the (111) plane, so the texture is most likely to occur. (001) Textures are not very likely to occur, but can occur where high surface energy (110) textures are unlikely to occur. On the other hand, for bcc crystals, the most dense atomic surface is the (110) texture, which generally appears, and sometimes occurs because the (001) texture is then the low energy texture, and the high energy surface ( 111) Almost no texture is produced. Therefore, for fcc crystals, the (111) texture is defined as natural, and for bcc crystals, the (110) texture is defined as naturally occurring. The fcc (111) and bcc (110) textures occur naturally when low surface energy substrates such as amorphous metals that tend to wet or wet the deposited material are selected. On the other hand, when the substrate is not wet-treated, such as an oxidized surface where the deposited material tends to bond with oxygen atoms and limit the mobility of the atomic surface, the limited orientation in the deposited membrane is common. Seen, or at best a set of mixed textures. fcc crystals tend to form only weak (111) and (001) textures, and bcc crystals tend to form weak (110) and (001) textures. For grains of magnetic cubic crystal thin films with (110) or (001) textures, there are numerous easy and difficult magnetization axes on the grain surface. further, Since each grain has a random in-plane orientation with respect to the other grains, these polycrystalline magnetic materials result in a collection of a large number of random anisotropic axes. These materials provide both a non-linear response and a high coercive force associated with loss and noise. Therefore, bcc or often (110) or (001) textures and growing bcc derivatives are rarely used in equipment. For this reason, bcc materials with high saturation magnetization values are rarely used by device designers who do not want domain wall movement.
The very authoritative textbook "Physics of Ferromagnetism," by Soshin Chikazumi, on the thermal induction anisotropy of permalloy to see the poor understanding of the causes of uniaxial anisotropy in cubic materials. It is enough to refer to pages 299 to 309 of "2nd Edition". Professor Chikazumi attempts to explain "the phenomenon of" directional rules "or the anisotropic arrangement of pairs of different atoms such as Ni-Ni, Fe-Fe, or Ni-Fe". This is described in detail in the subject matter and commentary in the literature. The theory is that interatomic spacing between Ni-Fe pairs is narrower than other possible pairs, resulting in lattice distortion from the atomic pairs. Crystalline uniaxial magnetic anisotropy is proposed as resulting from the magnetic elastic energy associated with the resulting lattice strain. Professor Chikazumi further outlines the second theory, which presumes that "rules are caused by the clear volume development of the regular phase, and the shape anisotropy of the second phase. As a result, the induced anisotropy will be explained. " Geometric grain shapes other than spheres can generate significant shape anisotropic energy. However, it should be pointed out that no physical evidence is provided to support the theory.
Shape anisotropy is a conceptually reassuring explanation, and this phenomenon can be observed in elongated magnetic powder as used in magnetic particle data storage tapes and bar-shaped permanent magnets. Here, the magnetic domain easy axis is aligned with the longer dimension. However, Professor Chikazumi can use the relatively complex paired regular model of dipole-dipole interactions proposed by Neel to qualitatively explain magnetic field induction anisotropy. While showing that, there are no quantitative decisions, "the reason why no quantitative decisions may be obtained may be the failure to achieve various quantities of estimates and perfect thermal equilibrium." Stated. Nonetheless, Professor Chikazumi commented on the texture of the material: "Magnetic annealing is most effective for <111> annealing, then <110>, and so much for <100>. It's not the target. " In this article, Professor Chikazumi describes the parallel orientation of the applied magnetic field and the crystallographic orientation during the annealing process.
Most recently, "Soft High Saturation Magnetization (Fe)" published in September 2000 by NX Sun and SX Wang.<sub>0.7</sub>Co<sub>0.3</sub>)<sub>1-x</sub> N<sub>x</sub> Thin Films For Inductive Write Heads IEEE Transactions On Magnetics, Vol. 36, No. 5 states that bcc-like material orientation is achieved during deposition in an applied magnetic field. The publication further showed that the nitrogen content is required, strains the film, forms small grain structures, and shifts and widens the peak angle of (110) x-ray diffraction. "Significant amount" of second crystalline magnetic phase Fe<sub>4</sub>It was shown that N appears on the membrane. Chikazumi suggested that this anisotropic behavior may be due to the strain, shape, and pairing rules associated with the second phase. Note that the orientation at this time is obtained on a Si (100) oxidized texture substrate and the FeCoN (110) texture is weak according to the applicant's standards. Furthermore, it was shown that the coercive force of the difficult-to-magnetize axis could be reduced by sandwiching the FeCoN film between two permalloy films, but the orientation of the FeCoN film was improved by depositing on the permalloy. Is not shown or evidence is not presented. In practice, a comparison of the hysteresis loops of the magnetization difficulty axis with respect to the membrane with and without the permalloy membrane shows a similar anisotropic magnetic field required to saturate the magnetization. This indicates that the permalloy layer did not improve orientation in the structure.
[Background of orientation using rigid magnetic film] Unlike magnetic field sensing devices and energy conversion devices, which typically require soft magnets, devices such as magnetic storage media and permanent magnets have higher uniaxial anisotropic energy and higher residue to achieve higher coercive force Hc. A preferred orientation is required to achieve the magnetic value. In motors and actuators, it directly affects the amount of work the device can do, and in recording media, it directly affects the output signal level and signal pulse width or flux transition width, and thus the recording density. Even in these hard magnet applications, it is desirable to include a soft magnetic material to enhance overall performance. For example, vertical thin film recording media have long been considered to replace horizontal thin film media in the future. However, for the recording system to function properly, it is desirable that the soft magnetic keeper layer or lower layer be placed on the opposite side of the hard magnetic recording layer in the direction perpendicular to the recording head. This soft layer provides a flux return path, or flux concentrator, for the magnetic field of the recording head, and also flux to stabilize the recorded bits against demagnetizing energy after the head is removed. Provide a closure pass. The former allowed the recording head to function in a medium with a higher coercive force, and the latter was improved for patterns recorded by removing some of the self-demagnetizing energy associated with vertical recording. Provides a sense of stability. In horizontal recording media, the soft magnetic material also improves the stability of the recorded pattern by reducing the self-demagnetizing energy of the hard recording layer. However, due to the lack of good methods of controlled anisotropic orientation, a soft magnetic sublayer has been found that provides a satisfactory keeper layer for either vertical or horizontal media. Absent. The lower layer of the soft magnetic material, which does not have uniformly controlled anisotropic orientation, results in medium noise caused by the domain wall due to the domain wall moving bulk Heisen phenomenon. When the magnetic head passes over the media data, the scattered media bits flyn. The magnetic field is effectively shorted out and the magnetic pattern in the lower layer of the soft magnetic material is relaxed to a new position. When a considerable domain wall is involved in this process, the domain wall generally separates from local pinning defects and causes abrupt changes in the magnetic structure. This bulk Heizen phenomenon causes a noise signal to appear in the data reproduction head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. , Relaxes the magnetic pattern in the lower layer of the soft magnetic material to a new position. When a considerable domain wall is involved in this process, the domain wall generally separates from local pinning defects and causes abrupt changes in the magnetic structure. This bulk Heizen phenomenon causes a noise signal to appear in the data reproduction head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. , Relaxes the magnetic pattern in the lower layer of the soft magnetic material to a new position. When a considerable domain wall is involved in this process, the domain wall generally separates from local pinning defects and causes abrupt changes in the magnetic structure. This bulk Heizen phenomenon causes a noise signal to appear in the data reproduction head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. When a considerable domain wall is involved in the process, the domain wall generally separates from local pinning defects and causes abrupt changes in the magnetic structure. This bulk Heizen phenomenon causes a noise signal to appear in the data reproduction head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. When a considerable domain wall is involved in the process, the domain wall generally separates from local pinning defects and causes abrupt changes in the magnetic structure. This bulk Heizen phenomenon causes a noise signal to appear in the data reproduction head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. A noise signal appears on the playback head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. A noise signal appears on the playback head. For either vertical or horizontal hard disk recording, at a wider position on the track where the recorded bit or flux pattern was recorded than the bit length, the desired anisotropic structure for the soft layer has a magnetization difficulty axis. Orient the recorded track and place the easy magnetization axis in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. , The easy axis of magnetization is to be placed in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. , The easy axis of magnetization is to be placed in the track direction. Therefore, in a conventional hard disk system, the easy axis of magnetization of the soft film must be oriented in the radial direction, and the difficult axis of magnetization must be oriented in the circumferential direction. Since the easy axis of magnetization of the soft magnetic material of the same structure is 90 degrees with respect to the field generated by the flux transition of the hard recording layer, the magnetization vector of the soft magnetic material is rotated by spin rotation, and the domain wall movement that generates noise. Is avoided. The long-standing technical problem is solved by providing a uniaxial soft magnetic material lower layer having a radial orientation. Since the surface density of the data storage device is forwarded, the size of the magnetic particles is reduced to the extent that the magnetic recording state is close to the thermal stability limit value. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. The magnitude of is reduced to the extent that the magnetic recording state is close to the thermal stability limit. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium. The magnitude of is reduced to the extent that the magnetic recording state is close to the thermal stability limit. Therefore, even horizontal recording stability and transition length will benefit from the use of a soft magnetic material lower layer to reduce the bit transition demagnetization effect. Again, the domain wall movement produces noise, so the ideal orientation of the soft lower layer should be such that the easy axis of magnetization is radial to minimize the potential for domain wall noise. Similarly, by providing a keeper layer in which the magnetization vector is rotated by spin rotation and the domain wall movement is minimized, it is possible to obtain a magnetic tape and a data storage system in which the xy address can be specified by anisotropic orientation control. The present invention relating to an oriented soft magnetic material provides a significantly improved future magnetic medium.
[Background of hard magnetic film orientation] The horizontal medium of modern hard disks is formed from polycrystalline thin films that are essentially composed of uniaxial grains randomly oriented on the substrate surface. Regeneration by averaging the signals from randomly oriented grains results in an isotropic response around the disc radius. This orientation randomness avoids turning twice for modulation during disk rotation of the anisotropic response. By mechanically grooved the surface of the disc before depositing the thin film layer, some small amount of orientation in the circumferential direction is often observed. However, the orientation ratio of the hard remanent magnetism in the track direction OR the remanence magnetism in the radial direction is rarely 1.2, and more often less than 1.1, which makes the substrate smoother and the medium film. As the thickness becomes thinner, it becomes smaller, and a higher surface recording density becomes possible. Similarly, the ratio of coercive forces in two directions is often referred to as the orientation ratio OR, the maximum of which is similar to the maximum of the residual magnetic ratio. The cause of the orientation of the third hard magnetic material has been argued for years. It is believed that the orientation of the hard magnetic material was caused by the c-axis of the hexagonal close-packed cobalt alloy, preferably in the direction in which the grooves were formed, while mechanically before deposition on the medium. It has also been shown that it diminishes or disappears when the disc is thermally cycled after the groove is formed. Due to the latter phenomenon, the cause of orientation is the thermal stress between the substrate and the membrane layer as the substrate stress associated with the groove is relieved from the heat used or generated during deposition. There is also an opinion that it depends. U.S. Pat. No. 5,989, In 674, Marinero et al. Outlined several patents and publications that clarified the cause of media orientation, during the deposition of hcp cobalt alloys on mechanically grooved substrates. The causes of the orientation of are described, citing stress, shape, and even crystal orientation. Others have described that a small orientation ratio can be achieved by depositing the medium material at an inclination angle with respect to the disk surface. See U.S. Pat. No. 4,776,938. Although this approach has shown some effect, it does not give significantly improved OR for grooved substrates and the deposition method described is significantly inefficient with respect to material deposition. Furthermore, the crystallographic cause of the same orientation effect has never been clearly stated or proven, and it is impossible to attribute it to the particle shape effect. Despite being patented a few years ago, it is not the technical method currently used in manufacturing.
However, whatever the cause of the orientation, it has proven to be advantageous in magnetic recording to achieve higher resolution, shorter flux transition lengths, and better thermal stability. There is no doubt that this orientation will benefit most future hard disk data storage systems. Similarly, in particulate tape, the individual needle particles are generally physically oriented in their longitudinal axis during the wet coating process, thus orienting their easily magnetized axis in the track direction. This results in higher signal levels, shorter transition lengths, and higher recording densities. For optimum performance, the easily magnetized axial orientation of the hard magnetic medium layer is for horizontal or vertical media to benefit from magnetic moment rotation instead of domain wall movement to avoid the bulk Heizen noise phenomenon of the soft keeper layer. In either case, it should be oriented in the direction of the applied head magnetic field, while the easily magnetized axial orientation of the associated soft keeper layer is recorded perpendicular to the applied head magnetic field direction for the horizontal medium. An interesting observation was made that it had to be on the track.
[Background of epitaxial thin film growth] The most frequently used concept is that the thin film preferably develops with atoms arranged on the surface of the thin film to minimize the atomic binding energy. This indicates that the most stable crystal surface develops when the atoms on the surface are arranged in the most dense arrangement consistent with the crystal structure. That is, for the fcc lattice and fcc derivatives, the (111) texture develops in thin films because the atoms are most densely packed on the crystal plane. The next most commonly occurring texture is (001), and the (110) texture is rarely seen. On the other hand, for bcc and bcc derivatives, the (110) texture develops preferentially because the atoms are most densely packed on the crystal plane. The next most commonly occurring texture is (001) and the (111) texture is never observed. As described in U.S. Pat. No. 5,693,426, rarely, for some materials, if the bcc derivative of B2, the crystal structure, forms a (112) texture when deposited under the correct conditions. It turned out that there is. However, even B2 crystals easily form (110) textures on non-oxidized surfaces, and the first deposited atoms have high surface mobility during film growth. Similarly, for dense hexagonal (hcp) crystals, the (0002) texture is the most densely packed and most commonly developed.
Similarly, there are several publications explaining the epitaxial relationships that are effective when the second crystalline material is deposited directly on the lower layer. Some are described in US Pat. No. 6,248,416. Assuming that the atomic lattice spacing between the first and second materials fits well with the texture of some second layers, the following relationship to the texture of the first layer is obtained. However, if the atomic lattice matching is not sufficient, epitaxial growth does not occur. For example, in US Pat. No. 6,248,416, which is incorporated herein by reference, Lambeth et al. et al. Lambeth et al. Was described.
The textures developed on these substrates are very strong and exhibit higher order and single orientation. The patent and other publications show that two different fcc materials can grow epitaxially on top of each other with the same texture. Similarly, two different bcc materials can grow epitaxially on top of each other with the same texture. However, be careful when one crystalline material class is deposited on different crystalline material classes. Not only does epitaxial growth on the first layer give a crystalline texture to the second layer, but the texture and in-plane orientation of the second layer are determined by the texture and orientation of the first layer. For example, in US Pat. No. 6,248,416, a clean Si (001) single crystal produces an fccAg (001) texture, on which bccCr develops a (001) texture. The authors are hcpCo (11<u style="single">2</u>0) Shows that the texture grows on the Cr (001) texture and the possible c-axis orientations are pre-determined as two possible directions by the crystallographic orientation of the single crystal wafer. Similarly, it is shown that a very rare bccCr (112) texture develops when a Si (110) textured single crystal is used as the substrate. Co (10<u style="single">1</u>0) The texture is produced by this Cr surface, and the c-axis of a single Co in the plane is parallel to the Cr <110> direction. Another epitaxial texture relationship described in US Pat. No. 6,248,416 is described. If the first single crystal Si substrate is (111), the epitaxial fccAg is (111) textured and oriented in-plane in the same direction as the Si crystal. When another fcc such as Cu or Ni is deposited on it, it becomes epitaxial and has (111) texture and orientation. Each of these textures is strongly oriented because the underlying layer is strongly oriented, allowing the growth of one epitaxial layer to continue in the next. Furthermore, we show how this structure is used to develop a hard magnetism perpendicular to the (0002) texture film using the Ti layer. It does not show how Cr grows on the (111) fcc layer or the Co (0002) textured layer. Other publications, H.Gong et al. (H.Gong, W.Yang, M.Rao, EE Epitaxial growth of quad-crystal Co-alloy magnetic recording media by DE Laughlin and DN Lambeth IEEE Transactions onNagnetics, 35 (5), 1999, pp.2676-2663) and G. Zangari ) Et al. (G. Zangari, B. Lu, DE Laughlin and DN Lambeth) Structure and Magnetic Properties of Sm-Co Thin Films on Cr / Ag / Si Templates Journal of Applied Physics, Vol.85 (8), April 15, 1999, In pp.5759-5761), it was shown that a strong Cr (110) texture occurs when bccCr is deposited on the (111) single crystal Si substrate and the subsequent epitaxial (111) textured fcc layer. However, the atomic arrangement of the (110) bcc crystal plane is rectangular, and the fcc atomic arrangement of the (111) crystal plane is hexagonal symmetric. Gong et al., Like Zngari et al., Stated that there are three possible ways (variants) to orient the bccCr (110) plane on a surface in which atoms are arranged in a hexagonal shape. These crystal planes and crystal directions are shown in FIG. These three orientations are such that (110) the <001> direction of the textured Cr is definitely parallel to <110>, or (111) the direction of the hexagonal atomic lattice plane of the textured fcc is <112>. Corresponds to the vertical case. The relationship between these three Cr (110) oriented variants with the fcc lower layer is shown using the following notation.<maths num="1"><img file="JP4698142B2_D0001.tif" /></maths> The publication and US Pat. No. 6,248,416 focus on viable hard hcp and soft fcc magnetic structures and are not shown to work with anything other than non-magnetic bcc materials such as Cr. For each of these three Cr variants, four possible hcpCo (1011) quad-crystal variants are obtained. This Co layer contains grains with four possible easy-to-magnetize axial directions when grown on each Cr variant. As a result, the four crystal structures of Co consist of grains each having one of the twelve possible easy-to-magnetize axial directions. There are four possible Co orientations for each of the three possible Cr variants.
FIG. 3 shows the positions of the three bccCr (110) oriented variants [4] with respect to the fcc (111) epitaxial template [5]. Example <001> bcc direction and example <112> fcc direction are shown along with other related crystallographic directions. Therefore, general publications state that there are three possible Cr variants that can grow on (111) fcc textured substrates. (111) The same results apply to single crystals of all sizes, assuming that they are textured fcc and the lattice constants are well adapted to cause epitaxial growth. Therefore, since the polycrystalline fcc film grown on a non-single crystal substrate having a strong (111) texture contains a large number of single crystal grains, the same result can be obtained for each of the individual grains. However, if the grains are small enough, only one or two of the three possible bcc (110) textured orientation variants can grow in a given grain, further depending on the deposition process situation. It is also possible for all three variants to coexist in a single fcc (111) textured grain. Furthermore, because the growth potential in each variant is equal, a sufficiently large sample will have an equal volume of each variant and the magnetic material will have symmetrically oriented magnetic properties. This is seen when epitaxial growth occurs on a single crystal substrate, or when the final film is polycrystalline and grown on a non-single crystal substrate.
Here, these effects will be described in terms of anisotropic energy density.
[(110) Magnetic anisotropy energy of textured thin film] (111) Cubic magnetocrystalline anisotrophic energy when magnetization is limited to the single crystal thin film surface of the cubic material to understand the magnetic properties of the (110) textured magnetic bcc material grown on the textured fcc substrate. Consider the formula. With reference to Figure 3, consider the in-plane unit cell atomic surface [6] of one variant of the grown (110) textured film. Figure 4 shows the crystal magnetic energy plotted where the magnetization is limited to the (110) crystal plane.<sub>1</sub>> 0 and K<sub>2</sub>It is shown as a function of the angle of the one variant with respect to = 0. Crystal magnetic energy density is E<sub>110</sub>Expressed as (θ) (see Physics of Ferromagnetism, 2nd Edition by Soshin Chikazumi, p.259-256), at this time.<maths num="2"><img file="JP4698142B2_D0002.tif" /></maths>Or when grown for the tuning of θ<maths num="3"><img file="JP4698142B2_D0003.tif" /></maths>Therefore, θ is the magnetization vector direction measured from the <100> bcc direction on the film surface, and K<sub>1</sub>And K<sub>2</sub>Are the primary crystal magnetic anisotropy energy density constant and the secondary crystal magnetic anisotropy energy density constant, respectively. Higher-order energy terms are usually smaller and are ignored here.
For pure Fe, K<sub>1</sub>Is often about + 4.7 × 10<sup>5</sup>Although it is said to be erg / cc, it can be prepared by adding a chemical agent. On the other hand, the irregular bcc form of Co<sub>50</sub>Fe<sub>50</sub>Is about -1.5 x 10<sup>5</sup>Considered to be erg / cc, the rule bcc derivative B2 form of Co<sub>50</sub>Fe<sub>50</sub>Is about zero. Here, K of regular alloys and irregular alloys<sub>1</sub>The value is about the same as when the Co atom percentage is 35% or less. K<sub>1</sub>The value is Co<sub>35</sub>Fe<sub>65</sub>About +1 × 10<sup>5</sup>It is erg / cc and is close to the Fe value such that the Co content is less than 25%. The alloy is of particular interest due to its high Ms value and anisotropic properties as a function of the composition. For the anisotropy constant, refer to Modern Magnetic Materials Principles and Applications by RC O'Handley, p.190-192, and for the saturation magnetization value, refer to p.145. K<sub>2</sub>Higher-order anisotropy constants such as are not well known, but are usually small in magnitude and not so heavily weighted in energy equations. K<sub>1</sub>Only when the role of K is disappearing<sub>2</sub>The term plays an important role in the calculation. Note that the 2θ tuning plays a dominant role in the description of the variant sets described below, while the 4θ tuning cancels out. Furthermore, the 6θ tuning of the secondary anisotropic energy term is weighted significantly less than the 2θ tuning and 4θ tuning. Therefore, in many cases, the effect of this 6θ tuning can be ignored. Those skilled in the field of magnetism will understand that the role of higher order anisotropic energy densities is small in the present invention and will be omitted in the following description for the sake of brevity.
Function (K<sub>1</sub>The minimum energy minimum value for> 0) is the in-plane <100> direction, and there are two equal maximum values [7] in the in-plane <111> direction. Therefore, the local minimum value [8] is located between the two <111> directions and is oriented in the <110> direction. Therefore, this texture does not produce uniaxial behavior and undergoes domain wall movement, resulting in highly non-linear switching in all directions of the applied magnetic field when the sample is driven closer to saturation. When the magnetic structure is single domain wall orientation, monotonous behavior is seen over a fairly large region on the energy curve at positions where the response is potentially reversible and low loss. Theoretically, if the magnetization is first directed all in the <001> direction and then rotated by the applied magnetic field in one direction of the <111> hard-to-magnetize axis in the plane, until the magnetization approaches the hard-to-magnetize axis. The process is reversible and therefore lossless. However, when the magnetization crosses the in-plane <111> direction, it suddenly jumps toward the local minimum energy value in the in-plane <110> direction. This jump is not reversible and therefore results in a lossy or low sea process. Domain wall movement governs all possible rotational mechanisms, either because a 90 degree domain wall is almost always present, or because it is formed between in-plane <001> and <110> oriented magnetizations. .. Defects in grain boundaries or membrane structures result in local domain wall movement jumps, again resulting in a loss mechanism.
In the region from 0 to 180 degrees of the energy plot in Figure 4, K<sub>1</sub>When> 0, there are two <100> directions, two <111> directions, and one minimum value [9] corresponding to each <110> direction of the variant [6] in Figure 3. A maximum value [7] and a local minimum value [8] can be seen. These extrema are K<sub>1</sub>The opposite is true when <0. It is clear that this curve does not meet our definition for uniaxial magnetic materials.
Abstract of the invention The present invention relates to an oriented thin film magnetic material and devices such as a magnetic sensor, a transducer, an electronic circuit component, a recording head, a recording medium, and a data storage system having the oriented thin film magnetic material, and more specifically. Involves in crystallinely oriented thin film materials and oriented magnetic layer structures that utilize materials such as iron, nickel and cobalt, and their alloys. In particular, the present invention relates to a structure for achieving uniaxial crystal magnetic orientation using the (110) texture of a body-centered cube (bcc) or body-centered cube-derived crystal thin film structure. In general, bcc or bcc derivative materials have higher saturation magnetization than face-centered cubic crystals, and the present invention that controls the orientation of bcc and bcc derivative materials makes it possible to construct new devices. Has good orientation, high magnetization, high magnetic permeability and low loss. These devices are used to detect or determine magnetic fields, store and extract data, and also convert energy or perform electronic signal processing. The improved magnetic properties or orientation of the magnetic properties of the material when used in the above-mentioned devices can provide better technical performance or allow the construction of small, fast or inexpensive systems. In many existing magnetic devices that utilize conventional magnetic material structures, performance can be significantly improved by replacing the conventional magnetic material structure with the new structure described in the present application.
The present invention will be better understood by reference to the drawings.
The material structures of the present invention are included in both magnetically rigid and soft magnetic devices, including magnetic field sensors, transducers, electronic circuit components, magnetic solid-state memories, security devices, recording heads, recording media and data storage systems. Obtained and obtained an oriented magnetic layer structure utilizing the magnetic materials oriented to the thin films described above, more specifically, crystallinely oriented thin film materials and materials such as iron, nickel, cobalt, and alloys thereof. Including. These structures are particularly useful when embodiments of the device require a special magnetic orientation, a particular linear or non-linear response, or a low loss response.
By carefully controlling the epitaxial growth of textured bcc or bcc-derived thin films of (110) crystals on highly oriented (111) hexagonal atom templates, Applicants can apply for six new crystals with special orientation relationships. Invented a set of variants. The selection and growth of a specially exchange-coupled subset of these six orientation relationships results in a symmetric magnetic thin film with broken symmetry. (111) Textured fcc, (111) Textured fcc derivatives, or (0002) Textured hcp crystals are examples of (111) Textured hexagonal atom templates. In each case, the atoms in the template are close-packed. If a highly oriented (111) hexagonal template is available, the bcc or bcc derivative (110) texture structure can grow epitaxially. Six new variants can be constructed by combining the two, highly oriented and using special process techniques. Since bcc and bcc-derived materials generate a single class of behavior under many circumstances, the notation "bcc-d" is used to represent either bcc or bcc-derived crystal structure. Similarly, the notation "fcc-d" is used to represent either fcc or fcc-derived crystal structure. Since the closest hexagonal pattern of atoms is formed in the (111) plane, for example, a material having a NaCl prototype structure can be considered. The (111) plane of these fcc-d atoms and the (0002) plane of the hcp crystal are always close-packed, which is different from the (111) atomic crystal plane of many non-close-packed fcc Brave lattices. For example, a Si, diamond-structured substrate is one of the fcc Brave lattice F3dm space groups. Similarly, C15 (Str<u style="single">u</u>kturbericht display) is also one of the F3dm groups that do not have the closest (111) plane. From FIG. 3, it can be seen that there are three hexagonal template <110> directions similar to (10). By carefully controlling the epitaxial growth situation, it can be seen that a highly (110) textured bcc-d crystal can be grown in one of its <111> directions parallel to the hexagonal template <110> direction. This orientation differs from the orientation obtained by the Gong and Zangari processes. In the process of the present invention, the <110> direction of bcc-d is no longer parallel to the hexagonal template <112> direction.
Before discussing the new six crystal (110) textured bcc-d variant system, we will examine the three crystal (110) textured variant system in detail. E as an energy equation for one variant<sub>110</sub>(θ) was described above and explained in the context of FIG. If there are two dominant variants of this (110) textured bcc on the same (111) textured fcc crystal surface, and also by contacting each other or with another magnetism When magnetically coupled through the material, the magnetization vectors of each variant tend to move or rotate together. Similarly, as described in the Gong and Zangari literatures, all three variants are present and the three variants are magnetic, as illustrated in Figure 3 [4] and supported for Cr. When coupled with each other, each magnetization tends to move with the other magnetization. Not surprisingly, Cr has no magnetic moment and is not mentioned in either the Gong or Zangari literature. The anisotropic energy density equation for two equal amounts of crystal variants can be expressed as the sum of each variant along with the magnetization rotation.<maths num="4"><img file="JP4698142B2_D0004.tif" /></maths>
Here, θ is the angle of the magnetization vector with respect to the <100> direction of the original variant [6]. Since the in-plane fcc <112> direction of the lower layer is 60 degrees apart, the easily magnetized axes of the two other bcc (110) texture variants are located 60 degrees apart from the first variant. A simple trigonometry shows that the sum of the energies of these three variants has a constant of (7/32) K1 and is independent of θ. Therefore, there is no net anisotropy direction. However, in actual materials, domain walls can be formed, so the local anisotropy of each variant results in significant coercive force and lossy behavior, along with material defects.
Since the reference point in all energy calculations is chosen arbitrarily, constants can always be added to the energy function without changing the expected physical behavior. Furthermore, since the added energy in the case of the above three variants is constant, it is clear that the energy function is simply proportional to the negative of one variant if there are only two coupled variants. Mathematically speaking,<maths num="5"><img file="JP4698142B2_D0005.tif" /></maths>Will be. Therefore, for materials that have only two variants<maths num="6"><img file="JP4698142B2_D0006.tif" /></maths>Will be.
Therefore, after normalizing the energy density by the appropriate unit volume of the material, the energy curve of the two coupled variants is 1/2 of the inverse energy curve of one variant with the total unit volume of the same material. .. Since the curve shape and energy minimum and maximum values can be replaced in position, the difficult-to-magnetize axis and easy-to-magnetize axes are also replaced. All localized easy-to-magnetize axes or minimum energies, eg [8], are localized difficult-to-magnetize axes or maximum energies. Note that the constants in the energy equation are arbitrary and do not affect the physical response, so they can be considered zero in all analyses. Therefore, the response of the magnetically coupled pairs of these variants is K.<sub>1</sub>The sign of the value changes and it appears as one variant material with the amplitude of the difference at the extremum reduced by 1/2. Therefore, whether the material of the (110) textured cubic magnetic material is composed of one variant, two coupled variants or three coupled variants, it is uniaxial from the magnetocrystalline anisotropy energy density function. No behavior is obtained, but a form of biaxial behavior is obtained. Therefore, if there are three possible variants of the bcc magnetic material at 60 degree intervals, as suggested for non-magnetic bccCr by Gong et al. And Zangari et al., And (110) textured cubes. Regardless of whether the magnetic film is epitaxially grown on a polycrystalline or single crystal, it cannot have uniaxial behavior. Therefore, the M vs. H response function has coercive and lossy behavior. Furthermore, even if the volume of each of the three variants is arbitrarily selected, uniaxial behavior cannot be obtained. When the volumes of the three variants are not equal, the present application describes it as "broken symmetry".
If all three variants are present, but one or more have a larger volume than the others, it will be clear that the energy properties of the variant with the larger volume dominate. To clarify this, consider the case of Fig. 3. In the figure, six have only 1/2 the volume, and the other two equal volume variants each have only 1/4 the volume. At this time, 1/2 of the six volumes is equal to the volume of each of the other two. The energy terms from four-thirds of the total volume are equally divided and their sum is a constant. Therefore, the energy curve of the shape shown in FIG. 4 can be obtained from the remaining 1/2 of variant 6. In the present application, a variant having a larger volume is referred to as "dominant".
Note that no US Pat. No. 6,258,416 or the literature of Gong et al. Or Zangari et al. Describes the use of any magnetic bcc or bcc-derived thin films. Furthermore, no magnetic bcc or bcc derivative material studied and provided as a magnetic layer is described. Even if magnetic bcc or bcc derivative materials have been studied and shown to have the same three variant crystal orientation relationships found for bcc (110) Cr when placed on the (111) fcc layer, their The magnetic properties would have been lossy and would have had a significant coercive force. Furthermore, it has never suggested that samples were obtained in situations where the volumes of each variant were not equal. Therefore, there has never been a description of a variant system in which symmetry is broken or a dominant variant system.
Similarly, note that in the description of FeCoN thin films, Sun and Wang have not proven for any (110) textured variant. Furthermore, it does not explain the observed anisotropy and shows that it does not depend on any ordered variant structure because the anisotropy field was similar whether or not the permalloy was placed under CoFeN. Was done. The result is a distorted grid and prominent Fe<sub>4</sub>It showed the N second phase and showed a possible anisotropic mechanism associated with distortion or shape effects as described by Chikazumi.
It is well known that crystallographic twinning can be formed on the (112) plane of bcc-d. The (112) plane includes both the <110> and <111> directions. Thus, a pair of bcc-d twins can form a interface in the direction of the plane containing the surface normal of the thin film surface and one of the hexagonal template <110> directions. In this arrangement, the hexagonal template <110> direction coincides with one of the bcc-d <111> directions. Therefore, there are two possible variants of the bcc-d (110) textured film grown in the <111> direction perpendicular to the hexagonal template <110> direction. There are three possible pairs of twins because there are three possible hexagonal template <110> directions. Therefore, there are six possible bcc-d variants that can be grown. However, these variants do not have to be twinned in order to be present. For clarity, only two of the six possible variants are shown in Figure 5. The other four variants can be configured by rotating the positions of the two variants by 60 and 120 degrees. The figure is an illustration showing the orientation of the variants, and it will be understood that one variant spans a considerable amount of atomic distance.
The angle formed between the bcc-d <111> direction and the bcc-d <100> direction is represented by arctan (2) = 54.736 degrees. Similarly, the angle between the <11> and <110> directions is arctan (1 / 2) = 35.264 degrees. The sum of these two angles is, of course, 90 degrees, forming the angle between the <100> and <110> directions. It is important that this 54.736 degree value differs by 5.264 degrees from the 60 degree interval in the <110> direction of the hexagonal template. Conversely, in the variants observed by Gong and Zangari, the <100> orientation of the six proposed bcc-d variants is not parallel to any hexagonal template <110> orientation. In contrast to the symmetric arrangement of Gong and Zangari, where no <111> bcc direction is parallel to the hexagonal template <110> direction, in our new variant, bcc-d <111> One of the directions is parallel to one of the hexagonal template <110> directions, and the bcc-d <100> direction is not parallel to the <110> hexagonal template. Therefore, for a coordinate system where θ = 0 in the hexagonal template <110> direction, the bcc-d variant crystal <100> direction and the easy axis of magnetization (K).<sub>1</sub>> 0) is at βi = 0 ± δ degrees, 60 ± δ degrees and 120 ± δ degrees, where δ = 60-arctan (2) = 5.264 degrees. In Figure 5, note the orientation of each of the two bcc-d variants in the two <111> directions. The orientations of the two <111> directions coincide with the 60 and 120 degree hexagonal template <110> directions, but near the 0 degree axis differ by + 2δ and -2δ from the <110> axis. When exchange-coupled, this variant pair has the energy function E as described below.<sub>2-alc2</sub>It is expressed as (θ).
For (110) textured bcc-d crystal structures containing six equally occurring magnetically coupled variant volumes, the anisotropic energy density function is<maths num="7"><img file="JP4698142B2_D0007.tif" /></maths>In this case, the signs of + and-represent two functions for each angle in the same equation. As mentioned above, the total anisotropic energies of the six variants of this symmetrically balanced arrangement or the six Es.<sub>110</sub>It is mathematically shown that the energy term has no angle dependence, so the anisotropic energy density function is constant K.<sub>1</sub>It becomes (7/32). However, of particular interest is when certain odd or even variants are coupled. It is important to consider that any one or more of the six variants can grow on a hexagonal template and are presented as twin pairs or as a balanced equivalent material. Not required to be done. However, for clarity, in the discussion in the next few paragraphs, we consider that the amounts of material in each of the variants of the coupled group are balanced to be equal. So, for example, if only three variants are coupled, assume that the volumes of each of the three variants are equal.
If there is only one of the six variants, the energy density function is the energy density E shifted by a particular variant offset angle βi.<sub>110</sub>It is simply reduced to (θ). Similarly, if 5 of the 6 variants are present, as in the previous logic, and ignoring the insignificant energy constant offset or origin, the resulting normalized energy density function is missing. It is -1/5 of the peak-to-peak value of the energy function of the variant being used. As shown in Figure 4, E<sub>110</sub>The (θ) function has two maximums, one minimum, and a local minimum (K).<sub>1</sub>For> 0 or K<sub>1</sub>Since it has two minimum values, one maximum value, and a local maximum value with respect to <0), these two possible pairs of coupled variants provide an energy function that does not represent uniaxial magnetic material behavior. There are three possible pairs of coupled variants, namely two, three, and four variants per combination. However, since there are six variants to choose from, there is one or more ways to form each of these combinations. Here, we continue to consider that each variant in the set is balanced or has an equal amount of material.
As mentioned above, the single crystal <100> direction (and K) oriented to one of the six angles βi = + 0 ± δ degrees, +60 ± δ degrees and + 120 ± δ degrees.<sub>1</sub>There are six possible bcc-d variants, each with an easy axis of magnetization) in the plane when> 0. These six possible variants are a1 (60 + δ), a2 (60-δ), b1 (0 + δ), b2 (0-δ), c1 (120 + δ), and c2 (120-δ), respectively. Then, the corresponding corners in parentheses represent the corner positions in the bcc-d <100> direction with respect to the given variant in the <110> direction [10] of the hexagonal template. For example, the a1 and a2 variants are placed at 60 degrees in the <110> direction of the hexagonal template, or + δ and -δ away from the same position, respectively.
Special case: When δ = 7.5 degrees. Before discussing the energy density curves for various other combinations of variants, consider two of each of the three specific pairs of 15 possible pairs. However, consider here a special case where the deviation angle δ of the variant is not equal to 60-arctan (2) = 5.264 degrees but equal to 7.5 degrees. The three specific sets of energy density functions of interest are:<maths num="8"><img file="JP4698142B2_D0008.tif" /></maths>It is expressed as.
Subscripted energy function notation, for example E<sub>2-b1a2</sub> (Θ) represents two variants b1 and a2 coupled as a pair. The "" "symbol is used to indicate this special case of δ = 7.5 degrees. In each of these three cases, it can be seen that the coupled variants are phase-shifted exactly 45 degrees from each other. By using trigonometry for a particular δ = 7.5 degree angle, it is cos that depends on θ for each of these pairs separated by a 45 degree phase angle.<sup>2</sup>It is easy to see that it is (θ). Especially,<maths num="9"><img file="JP4698142B2_D0009.tif" /></maths>It is expressed as.
For example, E<sub>2-a1c2</sub> (Θ) represents a coupled pair of variants with Stoner-Wallfers' difficult-to-magnetize axes at zero degrees and easy-to-magnetize axes at 90 degrees, with individual variants representing their respective easy-to-magnetize axes (. K<sub>1</sub>For> 0, the <100> direction) is angled to +67.5 degrees and 112.5 degrees. That is, each of these paired variants has an easy axis of magnetization (K).<sub>1</sub>The angle of> 0) produces an energy function of the same functional form as the uniaxial Stoner-Wallfirth, which is located intermediate between the easy-to-magnetize axis positions of the two individual variants. This Stoner-Wallfers behavior occurs only due to the coupling of the two variants and their particular 60-2δ = 45 degree orientation with respect to each other. Furthermore, the anisotropic energy density function constant K<sub>1</sub>If <0, it will be clear that the directions of the easy-to-magnetize and hard-to-magnetize axes resulting from the coupled pairs of variants are reversed.
Each of these paired variants not only results in a uniaxial energy function, but also the Mx vs. Hx response function and the non-linear dual, driven by a zero-loss magnetizing difficulty axis, as explained earlier in Figure 2. The following My vs. Hy response function is also obtained. (According to a reasonable estimate, this is also the case for δ = 60-arctan (2) = 5.264 degrees.) Each of these three coupled pairs (δ = 7.5) is (110) textured bcc-d. It represents an ideal way to form an ideal uniaxial response from a magnetic material. As mentioned above, the sum of the energy functions of all six possible variants is the constant K.<sub>1</sub>For each of these three paired variants, the sum of the energy functions of the remaining four variants is equal to -1/4 of the sum of the paired variants because it is equal to (7/32). Therefore, there are three specific pairs of four-coupled variants whose sum is the Stoner-Wallfers-type energy function, but because of the negative sign, their hard-to-magnetize and easy-to-magnetize axes are the corresponding cups. Rotated by 90 degrees with respect to the ringed pair, the peak-to-peak energy amplitude swing is half that of the coupled pair. As such, these three sets of four coupled variants result in a Stoner-Wallfers linear and lossless M-to-H response function when driven in the direction of the net magnetization difficulty axis. From a notational point of view, the energy function E<sub>2-a1b2c1c2</sub> (Θ) and the corresponding E<sub>4-a1b2c1c2</sub>Note that the coupled variant represented by (θ) is the complement. in short,<maths num="10"><img file="JP4698142B2_D0010.tif" /></maths>Is.
Note that there are two other coupled variants of twelve possible combinations, none of which yields a uniaxial energy function. Some are inferior to others in the depth of local energy minimums. Similarly, the corresponding 12 other possible complementary four-coupling variants can also be written as non-uniaxially coupled pairs of crystallographic variants, thus exhibiting uniaxial behavior. Absent. This analysis only considers 20 possibilities for a set of 3 coupled variants. Based on the same theory as above, 10 out of 20 is the other 10's complement (negative). 14 of 20 have either a double minimum or maximum or a local extremum. The other six satisfy the definition of a uniaxial anisotropic energy function, but each have a wavy portion of the energy curve [11] located approximately halfway between the two values on one side of one minimum and one maximum. Including. The response curve when driven by the angle between this wavy portion and the difficult axis of magnetization is substantially lossless, but non-linear over most of the response curve. When driven in the direction of an angle constrained by the smooth portion of the curve near the magnetizing difficulty axis [12], the response is lossless and more linear only over that portion of the drive. These six energy curves E<sub>2-a1b2</sub>A plot of one of (θ) is shown in FIG.
In summary, excluding the six singular variants mentioned above and the absence of all six, there are 56 possible combinations of equally weighted variants. Therefore, when the phase angle of the variant is set to 7.5 degrees, six possible balanced variant combinations can be obtained that give an accurate Stoner-Wallfirth-like energy curve and corresponding M vs. H response function. In addition, there are six possible balanced variant combinations that can give a uniaxial but wavy energy density function. All others behave non-uniaxially.
Here, for the sake of perfection, a higher-order anisotropic energy term is briefly considered. Secondary anisotropic energy density constant K<sub>2</sub>An example of an energy equation for a coupled pair of variants when is non-zero<maths num="11"><img file="JP4698142B2_D0011.tif" /></maths>Indicated by.
As a result of examining the above function, the 6θ tuning is K.<sub>2</sub>The size of is K<sub>1</sub>It was found to be most important for uniaxial behavior when it is equal to or larger than the size of. However, this is not the case for many materials and is often K.<sub>2</sub>Is not known because it is not important. For example, in iron, | K<sub>1</sub>/ K<sub>2</sub>|> 5, but some literature states that it is not measurable. Nevertheless, if necessary, K<sub>2</sub>Similar analyzes may be performed, including.
Special case: δ = 5.264 degrees Now, let us return to the case where the phase angle δ of the variant is represented by 60-arctan (2) = 5.264 degrees. The 5.264 degree phase angle is not significantly different from the 7.5 degree value, but the resulting energy curve for the same combination of variants with a 5.264 degree phase shift is similar to the 7.5 degree special case described above. Uniaxial energy density curves are obtained for the same three pairs of variants described above, but with the smaller δ used.<maths num="12"><img file="JP4698142B2_D0012.tif" /></maths>Is obtained.
Each of these energy functions or variant pairs is the same if they are simply phase-shifted by 60 degrees on the hexagonal atom template.<maths num="13"><img file="JP4698142B2_D0013.tif" /></maths>Will be.
Similarly, there are three sets of corresponding complementary four-coupling variants, resulting in an energy curve that is similar but inverted and has a reduced amplitude difference. Moreover, each of these energy functions or four coupled variants is the same if they are simply phase-shifted by 60 degrees on the hexagonal atom template.
K as an example of the shape of the function<sub>1</sub>= 1, K<sub>2</sub>Using = 0, in FIG. 7, the anisotropic energy density function is<maths num="14"><img file="JP4698142B2_D0014.tif" /></maths>Is plotted with respect to.
In addition, FIG. 8 is a plot of the anisotropic energy density function for a complementary set of coupled variants, corresponding to four variant combinations each.<maths num="15"><img file="JP4698142B2_D0015.tif" /></maths>It is expressed as.
As mentioned above, ignoring the additional constants, the anisotropic energy density function shown in FIG. 8 has only -1/2 of the amplitude difference of the anisotropic energy function shown in FIG. These two functions are the sin of the Stoner-Wallfers model.<sup>2</sup>(θ) or cos<sup>2</sup>It is very similar to the (θ) dependency, but you can see that the curvatures around the maximum and minimum values are different. Although uniaxial in behavior, the magnetization difficulty axis n response function is somewhat non-linear and there is no low sea process. The curve shape at the extremum is inverted, so the easy-to-magnetize and hard-to-magnetize axis deviations from the Stoner-Wallfers model are also K.<sub>1</sub>Replaced when <0.
E<sub>2-a1c2</sub>The Mx [13] vs. Hx and My [14] vs. Hx response curves corresponding to (θ) are plotted in FIG. The Mx vs. Hx curve (solid line) shows the non-linearity caused by the fact that δ is not exactly equal to 7.5 degrees. However, it shows a lossless response. The My response curve resembles the quadratic behavior of the Stoner-Wallfers response. For perfection, Figure 10 is similar to Figure 6, which shows the case of δ = 7.5 degrees, E for δ = 5.264 degrees.<sub>2-ab1b2</sub>(θ) Shows the coupled variant energy curve. The curve is substantially flat to the right of the maximum value. However, it shows that it is a uniaxial material. These comparisons show that for δ = 5.264 degrees, there are 12 possible uniaxially coupled variants.
Part of the membrane is a coupled variant pair, eg E<sub>2-a1c2</sub>A four-by-four coupled variant consisting of (θ) with the rest of the membrane rotated 90 degrees, for example E<sub>4-b1a2c1b2</sub>When composed of (θ), both have a common easy-to-magnetize axis and difficult-to-magnetize axis. The two parts of the membrane must not be strongly coupled so that two separate coupled variants can set their functions individually. At this time, the two nonlinear response functions are averaged to obtain a substantially linear net response function. However, the response of the physical sample may contain some wavy parts. In order to rotate the four-coupled variants, it is important that the two parts of the material are rotated 90 degrees with respect to each other so that the difficult axes of magnetization of the two parts of the material are aligned. Not surprisingly, since the hexagonal template repeats rotations every 60 degrees, it turns out that the template only needs to be rotated 30 degrees to actually equal a 90 degree shift. This is because the mechanism for breaking symmetry is common for both variant pairs. Not surprisingly, other pairs of variants occur with respect to the orientation of the template between the 0 and 30 degree directions, even if they are not balanced, and can result in substantially linear behavior. Using a single sample with one or more groups of different couples of coupled variants, this concept is an epitaxially grown polycrystal on a randomly oriented (111) hexagonal atom template of a polycrystal. (110) It is a very influential concept because it allows a nearly linear magnetic response function even for textured bcc-d films. This makes it possible for a single crystal substrate to obtain optimal and most easily understandable performance, but having a single crystal is no longer a requirement for achieving a substantially linear, low loss magnetic response function. The only requirement is that each group of coupled variants be selected and placed in an appropriately rotated polycrystalline grain template to produce a substantially common axis of difficulty in magnetization for the entire sample. When a non-single crystal substrate is used for the deposition of highly (111) textured polycrystalline hexagonal atom templates, the membrane consists of individual grains with random in-plane orientation. All polycrystalline samples Techniques for obtaining the same easy-to-magnetize and hard-to-magnetize axes across the body yield a set of suitable (110) textured bcc-b coupled uniaxial variants for each randomly oriented hexagonal template. That is. A way to achieve this is to provide an effectively driven growth process that preferentially selects the appropriate coupled variant for each hexagonal template orientation used. The experimental results described below show the desired easy-to-magnetize axis during magnetic variant film deposition or during angular vacuum vapor deposition of bcc-d material to select the direction for the growth of crystallographic orientation. Both magnetic fields applied in the direction were used. Excellent uniaxial behavior was obtained by deposition in the applied magnetic field. However, a unique subset of many variant sets was obtained in both cases. These methods resulted in controlled uniaxial behavior with respect to the magnetic film. Other methods that effectively provide a preferred crystallographic growth direction are chemical plating in an applied magnetic field, chemical plating from a bath solution flowing along the substrate in a particular direction, onto a miscut single crystal substrate. And deposits on distorted or deformed hexagonal templates can be included. A unique subset of Liant pairs was obtained in both cases. These methods resulted in controlled uniaxial behavior with respect to the magnetic film. Other methods that effectively provide a preferred crystallographic growth direction are chemical plating in an applied magnetic field, chemical plating from a bath solution flowing along the substrate in a particular direction, onto a miscut single crystal substrate. And deposits on distorted or deformed hexagonal templates can be included. A unique subset of Liant pairs was obtained in both cases. These methods resulted in controlled uniaxial behavior with respect to the magnetic film. Other methods that effectively provide a preferred crystallographic growth direction are chemical plating in an applied magnetic field, chemical plating from a bath solution flowing along the substrate in a particular direction, onto a miscut single crystal substrate. And deposits on distorted or deformed hexagonal templates can be included.
It is believed that there is a similar mechanism in the symmetry-breaking mechanism that causes crystallographic orientation on the film surface, as well as preferring the formation of the lowest energy atomic arrangements that can produce a favorable crystal texture. Especially on the surface, the resulting texture is usually the closest atomic arrangement. Therefore, for the angled deposits of bcc-d material, we expect the <111> direction to tend to align with the direction of the depositing material. In addition, there are two <111> directions and a number of <110> hexagonal template directions in the plane of the bcc-d (110) texture, so the <110> template direction and the <111> direction are parallel to the deposition direction. One is exactly aligned or <110> at +60 or -60 degrees It turns out that there is some tolerance for alignment with one of the directions. The latter is usually the case for uniaxial materials. Moreover, the energy of film formation must be balanced with respect to the surface binding energy, otherwise the symmetry cannot be broken. Similarly, it was found that unless the process situation was optimal, not only was the symmetry of the variants broken, but the set of variants would be the three variants described by Gong and Zangari instead of the six new variants. The desired membrane structure can be obtained by adjusting the energy of these competitions. Adjustment parameters include deposition rate, substrate temperature, vacuum quality against film oxidation, and interatomic spacing of hexagonal templates and composition of materials. It is not necessary to achieve magnetic coupling between variants to achieve a structure with broken symmetry. However, in order to achieve uniaxial behavior, the appropriate set of variants must be exchange-coupled. The exchange coupling may be performed directly between the individual contacting variants if the exchange length has a size and spacing such that it is comparable to or smaller than the grain size of the variant. However, more preferred is to bring the magnetic hexagonal template material into direct contact with the magnetic bcc-d film to assist exchange coupling through the hexagonal phase. Further, the hexagonal magnetic exchange coupling layer may be placed either below or above the magnetic bcc-d layer, but may also be placed under the membrane so that it can also be used as a hexagonal atom template. Is most preferable. It is also feasible to construct a large number of alternating layers and bcc-d magnetic layers of the hexagonal template. Most preferably, it can also be used as a hexagonal atom template. It is also feasible to construct a large number of alternating layers and bcc-d magnetic layers of the hexagonal template. Most preferably, it can also be used as a hexagonal atom template. It is also feasible to construct a large number of alternating layers and bcc-d magnetic layers of the hexagonal template.
To show that the set of all variants does not result in a uniaxial energy function, a set of three variants, E<sub>3-b1b2c1</sub>Is plotted in Figure 11. It is clear that the behavior is biaxial rather than uniaxial.
It is clear that there are many possible ways to couple a set of variants to achieve a lossless magnetization response function. For epitaxial growth on a single crystal substrate, the resulting difficult and easy magnetization axes are related to the substrate orientation. However, in the case of polycrystalline, the epitaxially grown film contains a large number of variants corresponding to the crystal orientation of the grains of the individual hexagonal templates. The appropriate set of coupled variants is selected for a particular oriented polycrystalline hexagonal template using process conditions and symmetry-breaking mechanisms. The set of these coupled variants does not need to be equally weighted in terms of volume in the sample. In addition, for the use of both randomly in-plane oriented polycrystalline hexagonal template membranes and single crystal substrates, the selected coupled set variants are magnetically coupled sufficiently strongly that each variant of the set. The concept that the magnetization vectors of are rotated together is important. Coupling is likely to be done through a magnetic exchange coupling. Thereby, the crystallographic features of uniaxial and substantially uniaxial bcc-d materials with low coercive force and therefore low loss magnetic properties are described as "breached uniaxial symmetry". Crystallographically "broken symmetry" material is defined as being present when the individual set of variants does not contain all six (110) textured bcc-d variants of equal weight. The expression "mechanism of breaking symmetry" is used to describe the active process of obtaining a "broken symmetry" crystal structure. Due to these conceptual complexity and variety, only the cases where the coupled pairs of variants are equally weighted (balanced) in volume of material have been described in detail. This is not always the case, and other interesting energy functions and response curves may be obtained. One trivial but important case can be imagined by considering a membrane composed of six coupled variants. As mentioned above, the volume of each variant is equal If so, the energy density function is constant and has no θ dependence. However, uniaxial behavior can occur if two of the six variants have a different volume than the other four variants. Two variants with different volumes of material are three Stoner-Wallfers variant pairs (E)<sub>2</sub> (Θ) or E<sub>2</sub>Consider a simple case where it is selected as one of (θ)). If the amount of material for this pair is slightly larger than the other four variants, the symmetry is broken and uniaxial behavior results. However, the difference in energy density between the minimum and maximum values is reduced by the relative difference in the volume of the material. For example, of the six coupled variants, each of the paired variants, such as Stoner Walfers, accounts for 30% of the total product, and each of the other four variants accounts for 10% of the total product. Suppose. At this time, there is an energy difference between the minimum value and the maximum value equal to 40% as compared with the case where all the materials are obtained only with a pair such as Stoner-Wallfirth. As a result, the magnetic response function becomes more sensitive to the applied magnetic field as the energy difference between the minimum and maximum values decreases with respect to the total magnetic volume. This represents a method of adjusting the slope of the magnetic response function or the relative permeability .
In many magnetic devices, it is advantageous to have a large number of thin film layers. In many cases, the thin film layers need to interact magnetically or electronically, so specifically, an anisotropic orientation relationship must exist between them. For example, in order to minimize the noise signal, it is very useful to supply control force in the direction of the magnetoresistive axis of the magnetoresistive sensor to stabilize the device against the formation or movement of the domain wall in the sensor material. is there. This is often achieved by placing a rigid magnetic material at the end of the rectangular device so that it provides a small static magnetic field or is exchange-coupled to the soft membrane. Similarly, in these devices, the magnetization needs to be biased to a partially rotated state. This causes the magnetization to swing positively or negatively in the direction of the difficulty axis of magnetization, avoiding signal clipping or extreme non-linearity. In this anisotropic magnetoresistive device, the signal resistance changes with the orientation angle of the magnetization with respect to the current direction. Thereby, when the second layer whose hardness or anisotropy is controlled is placed in contact with the first layer or in the vicinity of the first layer, it exerts the stabilizing or orientation effect as described above. Can be provided. Similarly, in a spin valve or a magnetoresistive magnetoresistive sensor, electron spin transfer on the interface between two stacked magnetic layers is performed via the relative orientation of the two magnetic layers. RC O'Handle Another method of using a technique in which uniaxial symmetry is broken to orient the magnetic medium is described. This same approach may be used to form or orient a rigid or antiferromagnetic layer structure.
New integrated circuit technologies have emerged based on the spin transfer process and the orientation of the magnetic layer as described above. The technique is often referred to as spintronics. The simplest concept is a magnetic spin transistor whose conductivity is a function of magnetic spin injected into the current controlled junction region. The concept of coulometric control is similar to carrier control or magnetic field control generated in bipolar or field effect transistors. However, with respect to spintronics, the orientation of the magnetic moment in the junction region of the device determines the spin-dependent injection of electron carriers. Therefore, control of the orientation and magnetic state of these layers is important. It is envisioned that transistors as well as electronic logic and state devices can be constructed. In addition, these devices can be imagined to have memory functionality, as the magnetic state remains even after the electrical power is removed by the correct design. The design and functionality of these future electronic devices is unpredictable, but there is one thing that is certain. Anisotropic energy density and anisotropic orientation are important. The mechanism that breaks uniaxial symmetry provides a methodology for controlling angular and magnetic states.
Other magnetic circuit components that benefit from improved anisotropy control include inductive properties. It is believed to include conventional inductors with a small magnetic material in the vicinity of conductors that carry current. The magnetic conductivity of the magnetic material makes it possible to store more energy and can therefore be used to manipulate the signal waveform as in a filter, or to transfer or convert power to an alternative voltage or current level. Thin film structures are particularly useful on microcircuit scales, which usually exclude the use of high frequency magnetic materials. Both inductors and transformers in which magnetic devices are formed using meandering paths of conductors and layers of magnetic material are conceivable. The design must include a sufficiently high extrinsic permeability, and therefore a significant degree of flux path closure to achieve efficiency. A substantially linear magnetic response is desirable for linear signal transfer and not particularly important for power transfer. On the other hand, analog signal mixers, such as those commonly used in wireless communications, are obtained from uniaxial materials using an easy-to-magnetize axis response to a hard-to-magnetize axis drive magnetic field due to the presence of a bias magnetic field in the direction of the easy-to-magnetize axis. Requires a non-linear response. However, what is important in each device is that the magnetic loss is minimal. Otherwise, the device puts noise into the signal, the energy tends to be dispersed and overheated in each cycle, and has a limited frequency response. It is clear that the mechanism for breaking uniaxial symmetry provides a methodology for controlling magnetic anisotropy and orientation and enables these desired device attributes.
Another magnetic field sensor that uses a non-linear response function and greatly benefits from lossless processes and high frequency operation is known as a fluxgate. Fluxgates are commonly used to sense the magnetic field of the earth and are widely used in wells to monitor the direction of drilling. It is especially important for deep well drills such as the oil industry where the cost of drilling is significant. It is very important to have a sturdy magnetic field sensor to provide information to avoid hitting any material. There are several physical designs for these devices. However, they all operate under the simple concept of driving one magnetic axis and inductively measuring the time measure of the change in magnetization caused by an external magnetic field in the direction of the orthogonal axis. The non-linear response obtains a signal proportional to both the external magnetic field and the drive frequency and is the second tuning to the drive magnetic field frequency. Since the response is proportional to the drive frequency, the amplitude of the signal can be significantly increased by high frequency drive for small applied magnetic fields. However, if the material has a loss, the drive frequency is limited, and if the loss is due to the Bulk Heizen phenomenon associated with domain wall movement, the measurement sensitivity is also noise limited. If the magnetic material has a lossless uniaxial response as obtained by a structure with broken uniaxial symmetry, the device is operated up to higher frequencies and has better sensitivity to external magnetic fields.
Another family of magnetic devices is a device commonly referred to as a tag, which can be used for archicle monitoring and archicle identification. Some of these devices operate on the principle of non-linear response, others operate on the principle of linear response with energy storage for later radiation. Controlling the anisotropic energy density and direction is important in each. Non-linear response devices typically allow abrupt switching of magnetic states via domain wall movements that generate large amounts of tuning in the driving magnetic field, by tuning being captured using an electromagnetic sensing antenna. Detected. In its simplest manifestation, the non-linear response tag is simply a piece of very soft magnetic material with a uniaxial anisotropy direction. This is easily formed by utilizing a material structure in which uniaxial symmetry is broken.
The linear tag operates on the principle of coupling the magnetization orientation to the mechanical resonance of the acoustic vibration in the tag. This is done through the magnetoelastic properties, which are complementary properties to the magnetostriction of the material. It is desirable that these materials have a large magnetostrictive coefficient. When the magnetization is rotated away from the easy-to-magnetize axis by a driving magnetic field in the direction of the hard-to-magnetize axis, the material becomes elongated or shrinks depending on the sign of the coefficient. Such a shape change corresponds to a sound wave moving in the direction of the tag. When the sound wave reaches the end of the material, it tends to reflect in the opposite direction. Therefore, resonance occurs when the tag is driven by a magnetic field that oscillates at a period corresponding to the acoustic travel time between the ends of the tag. Similarly, if the driving magnetic field is fixed, the length of the tag can be adjusted to fit. In bulk material equipment without a substrate, the acoustic properties are determined by the acoustic properties of the magnetic material. However, when the magnetic material is laminated with another material or deposited on the substrate, the acoustic propagation properties are determined by the combined properties of the magnetic material and the laminate or substrate. However, the acoustic resonance state of the combined set of materials must still be compatible with the magnetic field drive frequency. Tag detection occurs after the driving magnetic field is interrupted. Due to acoustic resonance, the tag continues to be elongated and contracted at the mechanical resonance frequency. Due to magnetostrictive coupling to acoustic vibrations, the magnetization vector oscillates at a mechanical frequency, producing a magnetic field at a very specific mechanical resonance frequency that can be detected by the detection antenna. With respect to the magnetoresistive magnetic field sensor, the magnetoelastic tag works optimally when a DC bias magnetic field is used to tilt the magnetization from the easy axis of magnetization. In this way, the magnetization vector oscillates at the bias point, avoiding doubling the frequency and improving efficiency and sensitivity. For thin films deposited on Si substrates, the high mechanical Q of Si further improves the mechanical resonance process. When substrates of different lengths are used, a large number of resonance frequencies can be utilized. Microelectromechanical structure or MEMS technology Ideally, a large number of unique tags are suitable for incorporation into such a device when desired. Resonant beams of various lengths are micromachined on a single substrate for a single device. Magnetic elasticity properties are deposited for all beams. Similar to electronic circuit technology, many of the devices can be processed on Si wafers in parallel. When the device is driven by an oscillating magnetic field, only a beam of length corresponding to the driving magnetic field frequency oscillates and stores energy. A unique signal signature can be obtained by sweeping the drive frequency and sequentially electronically listening to the radiated magnetic field response of each beam through the antenna, or by applying an impulse magnetic field and observing various response frequencies. Can be obtained. Various acoustic MEMS structures are obtained to allow for generic MEMS structures that can be programmed after the device has been configured. For example, mechanical braking of a given beam of a given length of a given device can result in a unique signal signature. Therefore, a myriad of unique tag devices can be constructed from a single MEMS-treated Si substrate. Thereby, it would be conceivable to label each product with a unique tagging device with a unique signal signature. Important for all these devices in terms of operation is the control and orientation of the uniaxial magnetic anisotropy energy. Magnetic material structures with broken uniaxial symmetry are ideal for this application. A signal signature can be obtained. Various acoustic MEMS structures are obtained to allow for generic MEMS structures that can be programmed after the device has been configured. For example, mechanical braking of a given beam of a given length of a given device can result in a unique signal signature. Therefore, a myriad of unique tag devices can be constructed from a single MEMS-treated Si substrate. Thereby, it would be conceivable to label each product with a unique tagging device with a unique signal signature. Important for all these devices in terms of operation is the control and orientation of the uniaxial magnetic anisotropy energy. Magnetic material structures with broken uniaxial symmetry are ideal for this application. A signal signature can be obtained. Various acoustic MEMS structures are obtained to allow for generic MEMS structures that can be programmed after the device has been configured. For example, mechanical braking of a given beam of a given length of a given device can result in a unique signal signature. Therefore, a myriad of unique tag devices can be constructed from a single MEMS-treated Si substrate. Thereby, it would be conceivable to label each product with a unique tagging device with a unique signal signature. Important for all these devices in terms of operation is the control and orientation of the uniaxial magnetic anisotropy energy. Magnetic material structures with broken uniaxial symmetry are ideal for this application.
Further, as described above, in the hard disk medium, it is potentially important to place the soft keeper layer in the vicinity of the hard medium layer. Improved media noise SNR, PW50, thermal stability, and overwrite properties by using uniaxial materials and controlling the layer orientation with respect to the recording track and with respect to the preferred orientation of the hard memory layer. The surface density can be improved through. The use of orientation to determine the bcc-d variant as the lower layer controls some of the orientation of the hard magnetic layer.
By examining the epitaxial growth of various membrane laminates on a single crystal substrate, some show unique characteristics when combined with a newly invented, coupled variant structure that was previously invented or not utilized. It was found that there is a unique epitaxial relationship.
The three Si crystal substrate texture orientations (100), (110) and (111) were first investigated by depositing epitaxial Ag on these clean surfaces. Some epitaxial relationships are US Pat. No. 6,248,416, Heng's literature, Zangari's literature, and, for example, Cu (fcc) on Ag (fcc), Ag (fcc) or Permalloy on Cu (fcc). Co (fcc) on (fcc), Cu (fcc) or permalloy (fcc) Cr (bcc) on Ag (fcc), Ti (hco) on Ag (fcc), or Cr (bcc) or The growth of Co (hcp) on Ti (hco) is already known from similar publications that have been widely studied. However, depositing bcc-d magnetic material is not described anywhere and does not disclose any structure whose symmetry is broken. We observed symmetry-breaking structures by applying symmetry-breaking methods and controlling sedimentation conditions. For example, using both deposition in an applied magnetic field and angled deposition on a highly oriented hexagonal template to highly textured and broken symmetry iron, iron compounds, and iron alloys. Obtained.
Further, as described in the above literature, one bcc-d is epitaxially grown on another bcc-d and yet another bcc-d while having a strong texture and three Si crystal textures <100>, < The orientation relationship between bcc-d on each of 111> and <110> can be maintained. There are other relationships as well. Epitaxial fccAg, Au, or Al always forms the same texture as the Si substrate surface. On this fcc-d substrate, some epitaxial relationships are obtained for subsequent layers. Yet other epitaxial relationships are created in these layers. The same texture is obtained when a bcc-d is grown on another bcc-d. The same texture is obtained when fcc-d is grown on fcc-d. When the bcc-d material is epitaxially grown on the fcc-d material with a given texture, it is represented by the "/" symbol and the following texture relationship is obtained. bcc-d texture / fcc-d texture (100) / (100) (110) / (111) (112) / (110) The following texture relationship is obtained when the fcc-d material is epitaxially grown on bcc-d. fcc-d texture / bcc-d texture (100) / (100) (111) / (110) (110) / (112) When an hcp material such as a Co alloy is epitaxially grown on bcc-d, the following relationship is obtained. hcp texture / bcc-d texture (11<u style="single">2</u>0)/(100) (10<u style="single">1</u>1)/(110) (10<u style="single">1</u>0) / (112) When Ti is epitaxially grown on fcc, the resulting Ti is hcp and the following texture relationship is obtained. hcp Ti texture / fcc texture (0002) / (111) However, since the bulk phase diagram of the Co alloy shows both the hcp state at low temperatures and the fcc state at some non-extreme temperature, it is possible to grow either hcp or fccCo. It has been found that epitaxial Co most commonly forms the fcc state and the texture of the underlayer when grown on the fcc underlayer. Therefore, it behaved like other fcc-d materials deposited on the fcc-d template. However, when formed on bcc-d, the hcp state was formed, resulting in the texture relationship described above. When Co is epitaxially grown on another hcp, it becomes an hcp and has the same texture and orientation as the underlying hcp.
In all cases except for the six bcc-d variant structures, the in-plane orientation relationship is as described in US Pat. No. 6,248,416 and the literature by Heng and Zanagari. However, we have invented some new relationships with oxides that allow the construction of new equipment. For the three fcc metals, Cr, Fe, and Al, and for these selected alloys, the surface of the metal is formed as an oxide, and when the same metal is deposited on the oxide on the surface, the second metal Textures and in-plane orientations such that the layers sometimes grow epitaxially and the first metal layer is close to unoxidized are possible. This is true for all three Si substrate textures. However, Si (111) is the most stable to oxidation, then Si (100), and the least stable is Si (110). Therefore, when growing an epitaxial film, according to current technology, the film that develops on Si (111) is usually of high quality, so the resulting texture of the subsequent layers is also of high quality. It is the easiest to analyze. Due to the high texture and lack of large angle grain boundaries, when the metal is exposed to oxygen, it is uniformly passivated in the thin oxide layer and stopped being consumed for oxidation. Oxides can be as thin as 1-3 nanometers when exposed to moderate atmospheric conditions, but add heat and moisture during direct epitaxial oxide deposition or during subsequent oxygen exposure. It can be made thicker. While extremely thin and difficult to measure, each of these oxides is generally prototype α-Al.<sub>2</sub>O<sub>3</sub>It is possible to form a rhombohedral corundum crystal phase represented by. Very interestingly, α-Cr<sub>2</sub>O<sub>3</sub>And α-Fe<sub>2</sub>O<sub>3</sub>Form a continuous solid solution. When the first metal is a Cr alloy and is oxidized and the second metal is Fe, the second metal formed to a certain limited degree can have the same texture and orientation as the first metal. I understand. The same is true when Fe is the first metal and Cr is the second metal. However, when Fe is oxidized on its surface, many magnetic properties are degraded. In addition, there are many additive elements that can be added to Cr, Fe, or Al and oxidized to the same crystal structure. For example, α-V<sub>2</sub>O<sub>3</sub>Phase and α-Ti<sub>2</sub>O<sub>3</sub>Both phases are present. α-V<sub>2</sub>O<sub>3</sub>Is α-Fe<sub>2</sub>O<sub>3</sub>And form a solid solution. When a limited amount of Al is added to Fe and oxidized, α-Al<sub>2</sub>O<sub>3</sub>The crystal structure is still obtained. Similarly, FeTiO<sub>3</sub>, MnTiO<sub>3</sub>, CoTiO<sub>3</sub>, NiTiO<sub>3</sub>Is α-Fe<sub>2</sub>O<sub>3</sub>It is known that it can be added to. Addition to Cr, Fe, or Al in small amounts and still α-Al when oxidized<sub>2</sub>O<sub>3</sub>There are many other metals that give rise to the structure. However, we recall here that in our sample, the oxide was very thin and it was difficult to determine the exact crystal structure. One or more oxidized structures can transfer the texture and orientation of the first metal layer to the second metal layer. Both Fe and Cr have cubic spinel structure A<sub>1</sub>B<sub>2</sub>O<sub>4</sub>There are many transition metals, and metal combinations that can be added to these to form and form spinels. In addition, the oxides formed may have one or more crystal structures. CrO<sub>2</sub>Is an interesting possibility because its approximately half-filled conduction band is essentially half-filled with a single magnetic spin orientation, holding a significant amount of potential in a spin valve or tunnel structure. However, corundum-type structures are the best candidates for oxides due to their stability and are generally known to form on Fe, Cr, and Al metal surfaces. While amorphous oxides are also formed, it is difficult to assume that a strong texture is transmitted from one metal to the other via the amorphous oxide.
With respect to bcc-d textures, the textures of the six variants mentioned above are when one bcc-d is placed directly on another metal bcc-d or on an oxidized surface. , Propagate variant crystal orientation. As a result, the uniaxial bcc-d layer in which the first symmetry is broken is grown, and the bcc-d layer in which the second symmetry is broken is epitaxially grown, oxidized, and another bcc-d is grown. It is possible to grow the layer on the oxide with the same texture as the first bcc-d. An interesting magnetic arrangement can be constructed from such layer-to-layer texture relationships. The pair of two variants epitaxially grown on each other has anisotropic energy constants of the same sign, and their easy-to-magnetize axes are aligned. However, something interesting happens when one layer has a positive anisotropic energy density constant and the other has a negative anisotropic energy density constant. As a result, the two layers have easy-to-magnetize axes at 90 degrees to each other. That is, one orientation is caused by a symmetry-breaking mechanism, and by using a differently signed constant bcc-d, the process normally required to break the symmetry of the second bcc-d film is used. Instead, the easy axis of magnetization of the second layer is rotated 90 degrees with respect to the first bcc-d layer. In addition, the thickness of the non-magnetic bcc-d layer, which can be placed between the two magnetic layers, can be used to adjust the magnetic exchange coupling between the layers. This can be adjusted from strong to weak as the non-magnetic layer becomes thicker. By adjusting the thickness of the non-magnetic layer and the thickness of each of the two magnetic layers, the saturated magnetization product, both the angle between the magnetization vectors of the two magnetic layers and the in-plane orientation of the two magnetic layers. Can be adjusted. Thereby, the strength of the exchange coupling between the two magnetic bcc-d layers can be controlled by selecting the coupling material or by controlling the thickness of the coupling layer, so that the coupling is performed. It is possible to select the net magnetization easy axial orientation of the film structure. These features are associated with the corners between the two magnetic layers, as well as changes in the corners. It is useful in the design of various magnetic devices because it allows the stiffness to be varied without the use of additional rigid or antiferromagnetic layers. This type of behavior is used in equipment structures such as spin valves or spend-dependent tunneling equipment.
It was also found that a strong (111) fcc-d texture grows when grown on six variants of bcc-d. In addition, the in-plane orientation is expected to be in the same direction as the hexadecimal (hex) template below. Therefore, a layered structure, eg bcc-d (110) / fcc-d (111) / bcc-d (110) / hex (111) / Ag (111) / Si (111) Is feasible, and either or both of bcc-d breaks symmetry using a symmetry breaking mechanism. (Here, the film on the left side of the "/" is deposited after the film on the right side of the "/". It is easier to remember which layer is above the angle of inclination of the "/"). Not surprisingly, this approach is also applied via the grain orientation averaging process when the substrate is polycrystalline. Furthermore, when both layers are uniaxial, the easy axes of magnetization of the two magnetic bcc-d layers can be adjusted relative to each other by using a mechanism that breaks symmetry at different angles for each layer. In another example, an epitaxial uniaxial layer structure consisting of many layers, bcc-d / Cr / O / Cr / bcc-d / hex (111) / board Is formed as. Again, the "/" symbol indicates a layer change, and the "0" layer indicates that the previous membrane surface was exposed to oxygen or air. The bcc-d layer is magnetic (110) textured and has broken symmetry. This chromium example represents a non-magnetic bcc-d material that is epitaxially grown on a magnetic bcc-d layer and allows an oxide layer that does not come into contact with the magnetic layer. Due to this structure, the two magnetic layers are electrically isolated from each other via the oxide layer, but uniaxial behavior can still be obtained by utilizing the structure in which the symmetry is broken. What is not shown but suggested in the same structure is a variant exchange coupling material or layer. For example, the magnetic hex (111) layer. In fact, it is not uncommon for a hex (111) layer structure to consist of one or more layers. Similarly, the fcc-d magnetic variant coupling layer, eg NiFe, can be placed on top of the final bcc-d layer.
For the thin film oxide layer, a spin-dependent electric tunnel conduction device is configured as a large effect magnetic field sensor. These devices, like spin valve devices, rely on electron transfer, which is sensitive to the relative orientation of the two magnetic layers. However, in a tunneling device, the current is directed perpendicular to the membrane surface and the tunnel must be through the oxide. Traditionally, they have usually been composed of fccCo or permalloy magnetic layers using Al oxide insulators. The use of Al oxide as an insulator is widely disclosed. However, oxide deposition is difficult to control and oxidation of the deposited very thin polycrystalline Al oxide layer usually results in conductive pinholes. Occasionally, Al is deposited as a metal and later oxidized, but the texture uniformity of polycrystalline oxide remains a problem due to the limited quality of the metallic texture in it. The thickness of the oxide layer should be only about 1-2 nanometers so that the tunnel energy barrier is at an appropriate value. The highly oriented film used to replace the first Cr layer or to form the above-mentioned oxide structure in which Al is placed in addition to the first Cr layer is used in spin-dependent tunneling equipment. It is clear that it will be useful. This is especially true when configured on a single crystal Si substrate. The uniaxial properties of the broken symmetry structure make it possible to deploy a fairly simple device or configuration process.
Another example of a large number of epitaxial pairs of layers allows the oxide insulator O to separate a conductor such as fccAg or Cu from two magnetic layers. bcc-d / Cr / O / Cr / Ag / Cr / O / Cr / bcc-d / hex (111) / Substrate For simplicity, the texture and orientation relationships are not shown, but are believed to exist as described above. When an Ag conductor carries an electric current, the two magnetic bcc-d layers guide the magnetic flux generated in a closed path around the conductor (eg Ag). In this structure, each or both magnetic layers may be uniaxial with equal or different anisotropic directions. A lithographically defined current path with a higher relative permeability represents a small electron thin film inductor, and additional conductors and magnetic layers make up a large number of turn electron transformers. obtain. Inductance is proportional to relative permeability. It has been found that relative permeability greater than 10 is fairly feasible and exceeds hundreds in bcc-d film structures where uniaxial symmetry is broken in some structures. Further, for thin metal magnetic films, the maximum operating frequency is determined by the ferromagnetic resonance frequency or, if the structure is thick, the eddy current. Our model shows that it is appropriate to predict operating frequencies of several gigahertz with Fe or Fe alloys. For uniaxial materials, the resonance frequency is proportional to the root of the cross product of the saturated magnetization Ms and the anisotropic magnetic field Hk. H for linear uniaxial materials<sub>k</sub>= 2K<sub>u</sub>/ M<sub>s</sub>Is. Therefore, the ferromagnetic resonance frequency is proportional to the anisotropic energy density, and the magnetic permeability is vice versa. For thicker membranes whose operating frequency can be limited by vortex currents, the addition of a few percent of chemicals from secondary materials such as Si, N, C, or Al to Fe or FeCo significantly increases resistance. be able to. For example, Si by weight of Fe reduces conductivity by a factor of 5. The transmission depth of the eddy current, or skin depth, is proportional to the square root of the inverse transmittance-conductivity-frequency product. We believe that for the proposed geometry, the thickness of the magnetic layer must approach 1 micron before the eddy current limits the operating frequency. Therefore, inductors, transformers, and sensors formed using material structures with broken symmetry should function in the gigahertz range.
However, due to the set of coupled variants with various broken symmetries and the fact that epitaxial growth should be considered to occur in each individual grain of polycrystalline material, the texture relationship and device concept is like Si. It can be applied to both the polycrystalline film structure and the single crystal structure of a single crystal substrate. Similarly, other single crystal substrates such as Ge and GaAs that are generally available can be used to produce thin film structures with very strong texture orientation and broken uniaxial symmetry, in which texture. Is selected to show the hex (111) template and assumes that an epitaxially grown thin film is obtained.
Yet another example of a device that benefits from a broken symmetry structure is the development of magnetic recording media. At this time, the hard magnetic layer is usually made of an hcp Co alloy such as CoCrTa, CoCrPt, CoCrPtTa, CoCrPtB, CoCrPtCuB and the like. In all of these, the desired crystal structure is hcp to allow for highly anisotropic energy density constants. Traditionally, hcp Co has been grown on a lower layer of Cr or Cr alloy to control the crystal texture, so that the disk substrate has a preferred orientation greater than 1 when compared to the orientation ratio perpendicular to the recording direction. It is mechanically textured (scratched) in the recording direction. By preferentially orienting the hard disk recording medium in the recording direction, higher coercive force and better thermal stability can be obtained. On the Cr (110) textured lower layer, the hcp Co alloy is (10)<u style="single">1</u>1) It is well known to develop crystalline textures. The texture is referred to as a Coalloy 4 crystal because there are 4 possible c-axis orientations on a single crystal Cr grain with a (110) texture. This is not an ideal hcp texture because the hcpc-axis, which is the easy axis of magnetization, is tilted toward or about 28 degrees from the film surface. Nevertheless, this texture is widely used in media and is quite advantageous when oriented in the direction of the recording magnetic field. Due to the demagnetizing field of the thin film, the magnetization vector is placed in the direction of the c-axis towards the plane of the film. This reduces the possible number of in-plane projections of the c-axis on a single Cr (110) textured crystal to two. The two are + arctan (1 / 2) = 35.26 degrees and -arctan (1 / 2) = 35.26 degrees with respect to the bcc-d <100> direction. Considering the bcc-d coupled variant pair as the template underlayer for epitaxial Co alloy growth, the symmetry-broken film provides a template in the preferred direction for Co alloy growth. This technique of selecting a textured and oriented bottom layer provides a passage for achieving a hard disk alignment medium. Coupled Variant vs. E<sub>2-a1c2</sub>Consider (θ). This broken symmetry-broken coupled variant pair has <111> crystal lattice directions parallel to the <110> direction of the (111) hexagonal template in the directions θ = 60 and 120. From Fig. 5, you can easily see the <100> direction of the two bcc-d cells. When a Co alloy 4 crystal is grown, the projection of the c-axis is about +40.3 or -40.3 degrees with respect to the bcc-d <100> direction.
Table 1 shows all possible in-plane magnetization easy axial directions of a Co alloy with a four crystal structure when grown on all six possible bcc-d variants. An ideal hcp c / a = 2 (2/3) = 1.63 crystal unit cell is assumed for the angle calculation. As an example of understanding the calculations in Table 1, consider the first variant a1. The <100> direction is 65.26 degrees in Fig. 5. Therefore, the two in-plane Coc-axis projections are approximately 105.6 degrees and 25.0 degrees, corresponding to the + and-items in the first row of Table 1. E with broken symmetry in Figure 5<sub>2-a1c2</sub>Considering only the two variants of the bonded variants, the Co alloy crystal anisotropy direction selected from Table 1 can be predicted. These two bcc-d variants, whether coupled or not, give rise to eight possible Co alloy variants associated with obtaining two four crystals. There are four for each of the two a1 and c2 bcc-d variants. However, since the out-of-plane demagnetizing force drives the magnetization vector toward the thin film surface, the four c-axis projections on the film surface a.<sub>1</sub>105.6 and 25 degrees with respect to c<sub>2</sub>It becomes 155 degrees and 74.4 degrees with respect to. Since the former two are closer to the 90 degree direction than the latter two are closer to the 90 degree direction, the net coercive force in the 90 degree direction is higher than the net coercive force in the direction of the 0 degree position. Unlike soft magnetic films, for good signal-to-noise ratio, it is desirable to have four possible hcp cobalt orientations decoupled individual grains. For this reason, bcc-d, whose non-magnetic symmetry is broken, is best provided with a template for the growth of Co alloys. However, the template in which the magnetic bcc-d symmetry is broken may be grown first, on which the non-magnetic variant template may be epitaxially grown. As mentioned above, it may be desirable to have a soft magnetic keeper layer underneath the oriented hard magnetic layer, although not required for all medium structures. The ideal easy-to-magnetize axis direction for the keeper is perpendicular to the recording track, and the stray magnetic field from the recording bit drives the magnetization of the keeper layer in the direction of its hard-to-magnetize. However, E<sub>2-a1c2</sub>The easy magnetization axis of the example is also at the 90 degree position. However, as mentioned above, K<sub>1</sub>It can be easily reoriented to the 0 degree position by selecting a material <0. A method that breaks the preferred symmetry due to manufacturing constraints on circular discs is to deposit the bcc-d material on the surface of the circular disc in the radial direction at an angle to the normal of the substrate. The broken symmetry structure also applies to polycrystalline films, where the (111) hexagonal template is highly textured and the deposition angle to the normal of the substrate is limited to within 15 Ω 75 degrees from the normal. can get. The angle of incidence is shown in FIG. 12, and the incoming bcc-d material moves in the direction of -r, which forms an angle Ω with respect to the normal (z) with respect to the substrate surface (xy). When the material to be deposited arrives at a glazing angle of 75 degrees, the method of breaking the symmetry is very effective, although the efficiency of the deposition is reduced. When the deposition angle is 15 degrees, which is close to the normal line, the efficiency of deposition is excellent, but the objectivity tends not to be broken. Applying a magnetic field is also one of the most effective ways to break symmetry, but K<sub>1</sub>> 0 bcc-d layer and E<sub>2-a1c2</sub>A magnetic field oriented in the track direction is required to achieve this. Radial or circumferential magnetic fields are too conductive to be included in modern hard disk media manufacturing processes due to the standard manufacturing process of simultaneously depositing material on both sides of the disk. There is no sex.<tables num="1"><img file="JP4698142B2_D0016.tif" /></tables>
In addition, it was found that a strong (111) texture was first obtained by depositing a wet metal layer on the surface of a typical hard disk substrate. The increased atomic surface mobility of the wet layer allows the close-packed texture to be formed from hexagonal template materials such as fcc Ag, Cu, NiFe, Fe and even hcp Co alloys. twenty five<sub>at%</sub>> x> 50<sub>at%</sub>When is Co<sub>1-x</sub>Cr<sub>x</sub>Non-magnetic Co alloys are preferred when exchange coupling of bcc-d is not desirable. It has been found that the deposited amorphous metal layer provides an excellent wet layer.
In particular, in all applications where a wet or wet layer is desired, materials that form a C15 or C14 crystal structure under NiP or bulk material equilibrium conditions form an excellent amorphous wet layer when vacuum deposited. Examples of C15 and C14 structures are Cr<sub>1-x</sub>Ta<sub>x</sub>And Cr<sub>1-x</sub>Nb<sub>x</sub>, Or Fe<sub>1-x</sub>Ta<sub>x</sub>, And Fe<sub>1-x</sub>Nb<sub>x</sub>At this time, 55 <x <75 atomic percentages. Here, it is noted that in the Cr-Ta alloy, the equilibrium phase diagram shows the phase transition from high temperature to low temperature from the crystal structure of C14 to the crystal structure of C15. By pairing these complicated balanced crystal structures, the tendency of forming an amorphous film in an unbalanced thin film deposition process increases. Strong (111) or (0002) textures were obtained by using these amorphous layers in each hexagonal atom template material fcc-d or hcp, respectively. For these two C15 materials, it was found that moderately strong textures still grow when they are exposed to oxygen. Therefore, the grain size of the hexagonal template layer and subsequent layers can be made smaller by processing control.
Magnetic media that should not be used in disk formats do not require cylindrical geometry, but modern hard disk media do. A new vacuum sputtering cathode has been invented to achieve a uniform symmetrical arrangement with favorable magnetic orientation around the disk substrate structure. FIG. 13 is a cross-sectional view of a rod-shaped sputtering target [15] made of bcc-d material for depositing on a disk substrate. Shields [16] and magnets [17] are placed around the target to facilitate sputtering plasma at low Ar gas pressures. The disk substrate [18] is held axially with the sputtering target, but is separated by the distance that the sputtered material [27] arrives at the disk surface in the radial direction and at an incident angle. Ar gas is introduced into the vacuum chamber through a passage [26] between the target and the water cooling shield. As a result, the sputtering gas [25] is concentrated in the vicinity of the sputtering target and minimized in the vicinity of the disk substrate. The sputtering gas also provides a cooling mechanism that transfers heat from the target to the water cooling shield. A low gas pressure is desirable to allow a mean free path for the application of sputtered material, which is equal to or longer than the distance from the target to the disk substrate. This prevents randomization in the direction of the sputtered material by avoiding gas collisions. Sputtering wears the target into a predetermined somewhat conical shape, forming a deposition path from the target to the disk at an angle of incidence in the desired range of 15-75 degrees. However, it was found that a narrower angle distribution centered at 45 degrees was preferred. The rod diameter of the target is smaller than the inner diameter of the desired disc recording substrate, and the material leaving the target follows a radial path to the disc essentially at all positions on the disc surface. In combination with the contour of the target surface, the distance from the disc to the target is adjusted to set the deposition angle. Produced by the magnetic field of a sputtering cathode magnet To avoid the resulting non-uniformity, the magnet structure may be rotated around the target rod during the deposition process. Although not required here, the target wears smoothly to some conical points and magnetizes the target in the direction of the center of the disc along the normal path from the disc surface during the deposition of any one disc. You can imagine moving forward and centering on. This makes it possible to adjust the deposition angle and deposition rate at various positions in the radial direction of the disc. Similarly, advancing the overall position of the target with respect to the magnet makes it possible to constantly replace the worn target. This allows deposition on a large number of discs before the vacuum is opened to replace the consumed target. Spinning the sputtering magnet around the target provides an interesting advantage. The magnetic field from the magnet can be arranged to provide a small but non-negligible magnetic field on the disk surface oriented around the circumference of the disk. This orientation provides an angular deposition, a mechanism that breaks the second symmetry in addition to the energy mechanism of magnetic field deposition to facilitate bcc-d orientation of magnetic materials. K And you get an interesting advantage. The magnetic field from the magnet can be arranged to provide a small but non-negligible magnetic field on the disk surface oriented around the circumference of the disk. This orientation provides an angular deposition, a mechanism that breaks the second symmetry in addition to the energy mechanism of magnetic field deposition to facilitate bcc-d orientation of magnetic materials. K And you get an interesting advantage. The magnetic field from the magnet can be arranged to provide a small but non-negligible magnetic field on the disk surface oriented around the circumference of the disk. This orientation provides an angular deposition, a mechanism that breaks the second symmetry in addition to the energy mechanism of magnetic field deposition to facilitate bcc-d orientation of magnetic materials. K<sub>1</sub>For> 0 magnetic bcc-d materials, the facilitated magnetization axes from the mechanisms that break both symmetries match.
Methods and structures have been invented for obtaining oriented hard Coalloys using structures with broken symmetry. It is clear that they can be used to pin the soft magnetic layer via an exchange coupling in many magnetic devices that utilize the soft magnetic layer.
In order to further confirm the information already described in the description of the present invention, a selected set of physically constructed materials and an analysis of these materials will be described in more detail below.
A large number of thin film sample structures have been manufactured and analyzed. Four types of substrates were first utilized: Si (111), Si (110), Si (100) and glass. The substrate is carefully cleaned by intensively cleaning with a solvent such as acetone, toluene, and isopropyl alcohol, and then with detergent and hot water. It is then extensively rinsed and dried with distilled water, isopropyl alcohol and distilled water. This removes all organic residues, water, and the most foreign metals from the substrate. The Si substrate is etched and dried with 49% hydrofluoric acid for about 30 seconds. The substrate is mounted on a carrier and quickly introduced into the vacuum system via a load locking system. It is well known that HF etching of Si leaves hydrogen bonds remaining on the Si surface to temporarily prevent oxidation. The vacuum deposition system is turbomolecularly pumped at low temperature. 1x10<sup>-7</sup>A vacuum base pressure of torr or better is generally obtained. Using a vacuum manipulator, the substrate carrier is attached to one of the three heated carrier positions and awaits thin film deposition.
The mounting of the substrate carrier facilitated planetary motion with a radius of approximately 7 inches in a vertical plane. There are three possible locations where the board carrier can be mounted, two in the planetary circle and one in the center of the planetary axis. The four magnetron sputtering target positions allow up to four different materials to be deposited at different corners of the deposit, each of which can be deposited as many times as needed in a single deposition sequence. The three targets are placed along the edge of a circle with a radius of about 7 inches aligned with the planetary position of the substrate carrier. In this way, the three targets and the two coaxial substrate holders can be aligned to face each other for right angle deposition, or the substrate can be aligned at any position in the planetary circle for angular placement. Can be rotated to. For example, one target is placed at the same position as 3 o'clock AM on the surface of an imaginary clock, and the substrate carriers are opposed by an adjustable distance in the direction of the normal to the planetary surface. Can be rotated to any position on the clock face of. The substrate carrier can be moved from or towards the target with a runout of about 7 inches or more, which is less than 2 inches of minimum spacing. The fourth target is the center of the planetary axis of motion, located on the plane with the other targets. Similarly, the third substrate carrier position is located at the planetary center on the substrate carrier surface. The carrier can be moved from the planetary position back to the center using a mechanical manipulator within a time of less than 1 minute.
Targets with 1-inch and 2-inch diameters are small compared to the distance between the target and the substrate. As described in FIG. 13, the argon gas introduced into the target surface allows plasma, while increasing the average free path length in the remaining chambers and limiting the application of sedimentary material. .. Secondary gases such as N or H can also be introduced through the same or different paths, allowing interaction with deposited layers. Each substrate carrier position is individually heated and monitored, and the temperature can vary from room temperature to about 350 ° C. Room temperature is selected to optimize the treatment for the particular single crystal substrate texture selected, except when experiments are not performed on a large number of substrate textures at the same time. The substrate is heated before and during deposition, and the temperature can change to a limited temperature as each layer is deposited. The sputtering cathode is DC powered up to 500 watts available. The physical dimensions of bcc-d as it is deposited on the substrate in the planetary circle from the central cathode can be measured from the normal of the substrate and the deposition angle can range from 75 to 33 degrees. did. For smaller angles, the substrate is mounted at a 45 degree angle to the substrate carrier and the target-to-carrier angle is adjusted. Moreover, under some circumstances, deposition at smaller angles can be achieved using the target and substrate positions in the planetary circle, but the substrate is rotated relative to the target. Of course, the distance between the target and the substrate can also be adjusted, but it creates a potential difference in deposition rate.
While a number of deposition conditions have been investigated, it has been found that the use of layers with a thickness of 20-100 nm facilitates simpler optical, magnetic, and microstructural analysis. Therefore, many bcc-d films are prepared with a thickness of 20 nm to 200 nm. Most Ag layers on Si substrates are 40 nm thick, and sometimes a second Ag and / or Cr layer is deposited as the final capping layer to prevent corrosion of the magnetic layer. The thickness of the hexagonal template varies and can be as thin as 1 nm and as thick as 40 nm. Typically, the hexagonal template of the magnetic exchange coupling is maintained at a minimum thickness, i.e. 1-5 nm, so that the bulk material properties are not significant compared to the bcc-d material properties. This facilitated a simpler interpretation of magnetic data.
Commonly considered materials are the fcc Ag, Cu, Ni, NiFe, Al, Co, and CoCrTa, the bcc-d Cr, Fe, Nb, NiAl, and FeCo, and the hcp Ti and. Includes with CoCrTa.
Other bcc-d materials that may be used in combination with Fe or FeCo include W, Mo, C, Cr, Ti, Ta, Si, Al, N, Cu, and B. Other fcc-d materials that can be used include CoCr, Au, Pt, Rh, Pd, and Ir, and other hcp materials that can be used include Re, Ru, Gd, Ti, and Tb. The material selected is selected for its chemical, magnetic, or lattice matching properties. A considerable degree of freedom in the atomic lattice constant is sometimes desirable to achieve epitaxial growth.
The following notation is used to specify the sequence of membrane layers to be prepared. bcc / hcp / bcc-d / bcc-d <55, <xyz> / hex / fcc / Sub (tex) Again, in the above notation, the membrane to the left of the "/" is deposited after the membrane to the right of the "/". From the angle of inclination of the "/", it is easy to remember which layer is on top. The "tex" symbol represents the single crystal substrate texture used. <55 indicates the angle at which bcc-d is deposited, and the number identifies the angle from the normal of the membrane. In general, the <symmetry symbol indicates that a symmetry-breaking mechanism is used when used. Similarly, when the magnetic field Ha is used to break symmetry, the magnetic field is applied along the hexagonal crystal direction <xyz>. Similarly, when the angle of deposition is used to break symmetry, the in-plane direction of the sedimentary material is displayed to follow the <xyz> crystal direction.
Therefore, as described in the example above, fcc is first deposited at right angles on the substrate, then hexagonal templates are deposited at right angles, and bcc-d is hex at an angle of 55 degrees to the normal to the plane. Stacked parallel to the template <xyz> direction. Subsequently, a second bcc-d is deposited, hcp is deposited, and finally bcc is deposited. If the surface is oxidized or deposited by exposure, the layer is marked "0".
U.S. Pat. No. 6,248,416 and the literature by Gong and Zanagari have shown that the texture and epitaxial relationships described above grow in Si. The present invention does not have six bcc-d variants, a variant exchange coupling, and an equal variant volume weighting, rather than the three variants shown in US Pat. No. 6,248,416 and Gong and Zangari. Involved in breaking the symmetry of a pair of coupled variants. That is, in order to break symmetry, the three variants as described in the Gong and Zangari literature or the six variants described in this application are not balanced.
Although the surface of the first metal formed oxides when exposed to air, the epitaxial growth of the bcc-d texture was first investigated. What is important to ensure this aspect of the invention is that, although no physical evidence is presented in this application, the crystallinity underneath the film exhibits a very uniformly oriented texture. , Consider that the density of the underlying membrane is very high. When the film has a very high texture, for some textures the grain boundaries have only low corners or small gaps between grains. This ensures that the membrane surface is passivated instead of being oxidized to the grain boundaries, and the membrane is separated from the membrane by expansion. It is very important for polycrystalline materials such as Fe, which tend to oxidize badly.
First, we provide an idea of the degree of texture obtained by observing the θ-20 X-ray diffraction scan of a conventional polycrystalline film structure. Hcp-CoCr<sub>12</sub>Ta<sub>2</sub>/ Cr / glass
It has a conventional, simple structure and is sometimes sputtered onto a hard disk medium. Examination of the X-ray scan revealed that there were two peaks. Cr (110) has a peak height of about 150 counts, resulting in 4 Coalloy crystals (10).<u style="single">1</u>The peak height of 1) is 25 counts, and one crystal (10).<u style="single">1</u>The peak height of 0) was 40 counts. The baseline is relatively noisy and there are no other identifiable peaks. The same scanning time, X-ray anodic current, and voltage are used for all θ-20 X-ray diffraction scanning examples described below.
Four samples are formed at the same time, each having a different substrate, namely glass, Si (100), Si (110), Si (111), and Si is washed and HF treated as described above. It is then placed in a vacuum system and quickly heated to about 165 ° C before the first Ag deposition takes place. The composition of the layer is Hcp-CoCr<sub>12</sub>Ta<sub>2</sub>(58nm) / Fe (37nm) <45 / O / Fe (37nm) <45 / Ag (40nm) / sub Is.
The values in parentheses are the nominal film thickness in nanometers. There is a step of exposure to oxygen during the process. This requires the sample to be removed from the vacuum system for a short time after the first two membrane deposits. In the table below, the items are listed as (X-ray count) / (X-ray texture).<tables num="2"><img file="JP4698142B2_D0017.tif" /></tables>
There are no observed diffraction peaks other than the single crystal substrate or secondary diffraction peaks, indicating a strong single texture orientation for each layer. The texture behaved as expected as described above. This result can grow with the same texture as Cr when bcc-dFe is deposited on the hexagonal template or on each of the other single crystal textures. Since the second layer Fe has the same peaks as the first layer, the lack of all further peaks along with the intensity of the peaks gives the texture even if the Fe in the second layer grows on the oxidized surface. A good indicator that it is the same as the texture of the first layer. However, since the peak is on the same linearity as the peak of the first film, there is a problem as to whether the peak of the second layer is actually present. hcp It was found that the peak of Coalloy and its strength provide a template with excellent epitaxial growth of the second layer of Fe on Fe-O compared to glass, providing excellent growth of Coalloy film. It was. This confirms that the texture of the second Fe layer is the same as the texture of the first layer of Fe. Epitaxial growth on oxides is confirmed for all textures, including the use of glass as the substrate. The latter is largely due to the excellent texture of Ag (111) on glass. In another experiment, amorphous Cr<sub>35</sub>When the wet layer of Ta is deposited on the glass before Ag, the strength of the Ag (111) texture is improved several times over the value when deposited directly on the glass. It should be noted that the lattice matching between Cr and Fe, which have approximately the same atomic lattice spacing, and Ag is not particularly good. This was investigated because another hexagonal template material (fcc-d or hcp) can grow epitaxially on Ag better than bcc-d. Hexagonal templates of Cu, Ni and NiFe, and the combined layer fcc-Co on Cu were found to produce stronger Fe or Cr textures when placed on top of the Ag template.
To further explore the oxides located between the two different materials, the following membrane structures were constructed.<tables num="3"><img file="JP4698142B2_D0018.tif" /></tables>
From this table, Si (100) epitaxy on oxides is good, but oxide-mediated epitaxy is less common when Si (111) is used. Cr epitaxy on Fe-O was found to be very good for Si (110), but the Co texture was not particularly good. The time and temperature used to prepare this sample is the same as that used for the Fe-O-Fe sample (99C15-19-2) described above, within experimental error.
The roles of Cr and Fe were exchanged and the experiment was repeated.<tables num="4"><img file="JP4698142B2_D0019.tif" /></tables>
Here, highly textured Cr on Ag and Fe (100) was found to grow very well on Cr-O. However, since the peaks were wide, it was not known whether many Fe (112) were grown, and the peaks of Cr and Fe were so close that it was difficult to observe. Some increased epitaxial growth of Fe (110) was observed, but it was difficult to distinguish due to the large amount of Cr (110). Nevertheless, Co (10<u style="single">1</u>The peak of 1) is the only peak that exists. For Si (110), Co (10)<u style="single">1</u>The appearance of 0) textures is strong evidence that some Fe was (112) textures.
Overall, it can be seen that Fe / O / Fe grew epitaxially for all three Si textures for a particular deposition temperature, deposition rate, and generally for a particular treatment condition. Similar results were obtained for Cr / O / Cr. For Cr / O / Fe / Ag / Si, the Cr (100) texture grew very well and the Cr (112) texture grew slightly. The Cr (110) texture did not grow very well. Therefore, bcc-d grows very well on its oxide. In addition, the bcc-d (100) texture appears to grow very well on other components. The (110) and (111) textures grow, but not so much. We changed the treatment temperature in the deposition of Ag on Si and found that the ideal temperature changed considerably according to the orientation of Si. The processing tolerance is even more dramatic. The Si (111) substrate could be treated at a higher temperature before being oxidized, which improved some of the subsequent deposition. The Si (100) textured substrate has the next widest processing temperature range, and the Si (110) substrate has the narrowest and lowest allowable processing temperature range. This is a natural result because the surface of Si (110) is the most unstable with respect to the fcc Brave lattice or the fcc-d lattice. Each substrate texture has its own optimum temperature and temperature range, and the particular temperature depends on the deposition rate and moisture during the time interval immediately prior to placing the substrate in the vacuum system. The quality of the HF-etched Si surface is highly dependent on the environment before Si enters the vacuum system. Therefore, it is difficult to optimize all the conditions at the same time, but when it is realized, the result is that the epitaxial growth through oxides on other Si textures is significantly improved. Fortunately, the (111) surface is the most stable for hexagonal template materials, allowing a wider tolerance for temperature and deposition rate. Therefore, oxides are present for each material combination.
The six variant structures will be described below. In fact, E<sub>2-a1c2</sub>A set of variants is shown. The following film structure is formed by being deposited with an incident angle. Two samples are prepared at the same time, but each has a slightly different deposition angle, a different thickness, and above all, a different angle with respect to the grid plane direction of the template. This is achieved by mounting two boards on the same carrier with different hex template angles. The deposition directions are the <110> template direction for sample 0909-6 and the <112> template direction for sample 0909-5, respectively. Each structure is Sample name: 0909-6 Ag (200nm) / Cr (40nm) / Fe (37nm) <45 <110> / Cu (10nm) / Ag (40nm) / Si (111) Sample name: 0909-5 Ag (200nm) / Cr (40nm) / Fe (60nm) <55 <112> / Cu (10nm) / Ag (40nm) / Si (111) Is defined as.
Note that the Cu hexagonal template cannot provide the necessary exchange coupling for uniaxial MH curves due to its non-magnetic nature. Nevertheless, the following shows that the required variant pairs have been obtained. When Cu is replaced with fcc-Co, Ni, or NiFe, or when Ni or NiFe is deposited on Fe, the coupled pairs are exchange-coupled. In this example, the final thick Ag and Cr layers are applied to avoid the possibility of Fe corrosion after the sample is formed. It turns out that these layers, especially Cr, are very effective in preventing corrosion. Samples are stored for 1 year without any obvious corrosion of Fe. This is concluded because the magnetic properties of Fe did not change even though they were very sensitive to oxidation. The Ag protective layer was somewhat discolored.
X-ray pole map analysis was performed. This test makes it possible to rotate the sample while setting the X-ray diffraction angle with respect to a specific reflective surface. Fe is (110) textured and the (100) plane is tilted at 45 degrees with respect to the surface of the film. The sample is tilted at 45 degrees and θ-20 is set to be reflected from the (100) crystal plane. The sample is then rotated 360 degrees and the signal is monitored. According to Figure 3, if there are only three possible variants, as supported by Gong and Zangari, three peaks should be observed per 180 degree rotation, with equal intervals of 60 degrees. Should be separated by. This is repeated during the second 180 degree rotation. However, when there are 6 variants as proposed in the present invention, 6 peaks are observed per 180 degree rotation, 12 peaks during a full 360 degree rotation, as shown in FIG. Is observed. If the variants are equally weighted by the volume of material and the substrate is perfectly cut, then all variands in the assembly are visible. We observed both types of pairs mentioned above. Therefore, not all samples are necessarily represented by the energy equations associated with the set of six variants. The set of three variants by Gong and Zangari is most commonly observed because it is important to implement an energy mechanism that rotates the crystal plane to separate the diffraction peaks. FIG. 14 is a diagram showing pole figure analysis for sample 0909-6. Careful analysis of the data revealed that the peak heights of the a1 and c2 variants were more than twice as high as the diffracted peaks of any other variant. However, measurements of related quantities are made on the area below the peak curve, which is a very time consuming and difficult measurement method, but shows a significant portion of the material in the two variants. .. Theoretically, only the Fe (100) diffraction peak is set because the θ-20 angle is specially set for the crystal plane of Fe (100) by scanning. Should indicate. However, Ag (220) diffraction lines can be seen in the vicinity, and the peaks are so strong that the tails of the six peaks are also shown. Similarly, there are a few very strong Si (400) peaks. The angle between the pairs of the six bcc-d variants is experimentally determined to be 2δ = 2 (5.26) degrees, as described above. However, the angle between the peaks in the pole figure of FIG. 14 is smaller due to the coordinate frame rotation required to move from the (110) and (100) planes. In the figure, the spread of any pair of peaks must be -2 (3.74) = 7.5 degrees. In fact, this was observed. The data provide strong explanation or evidence for both the concept of the six variants and the mechanism that breaks symmetry. Results are not clearly shown for sample 0909-5, where the deposition direction projection is in the <112> direction along the hexagonal template. Again, it was shown that there are two of the six dominant variants, but E<sub>2-a1b2</sub>Represented by. For reference, the pole figure peaks for these variants a1 and b2 are shown in Figure 14 along with the other four variants, E.<sub>2-a1b2</sub>The corresponding dicrystal energy curve for is shown in FIG. It is clear that no uniaxial behavior can be obtained from the same variant pair, even when magnetically coupled. Therefore, deposits with an angle of incidence in the <112> template direction will give completely different results. In summary, when deposited on a single crystal substrate with (111) texture, one set of variants produces uniaxial behavior when the above-mentioned angle of deposition is used as the mechanism for breaking symmetry, while the other sets Does not occur. In order to achieve uniaxial behavior using the method of breaking symmetry, the template and the correct orientation of the deposition direction are required. However, the symmetry of the variant is broken in both orientations.
It has been found that even when hexagonal templates of non-magnetic materials are used, they have exchange-coupled variants and it is possible to obtain uniaxially symmetric broken behavior. As mentioned above, the structure of the sample produced different grain sizes when treated at different deposition rates and temperatures, allowing magnetic exchange between the two. In addition, angular deposition results in grain morphology with grain anisotropy. However, by studying membranes deposited in both the <112> and <110> hexagonal template directions, the process of breaking symmetry and the uniaxial behavior caused by crystal orientation are anisotropic related to the grain shape effect. It turns out that it still exists with sex. A hexagonal template that induces good epitaxial growth with the correct and strongly textured and adapted atomic lattice spacing significantly aids in minimizing the shape effect of the particles.
A similar set of experiments is performed with a magnetic hexagonal template and an applied magnetic field. That is, as a mechanism for breaking symmetry, a magnetic field is used instead of deposition with an angle of incidence. By using Ni for the hexagonal template and placing the permanent magnets on the substrate carrier near the substrate, an applied magnetic field in the 100 oersted (Oe) range is generated. The structure of the sample is Ag (200nm) / Cr (40nm) / Fe (50nm) <45 / Ni (5nm) / Ag (40nm) / Si (111) Will be.
When a magnetic field is applied approximately along the hexagonal template <112> direction, the resulting set of variants is E.<sub>2-a1c2</sub>And when a magnetic field is applied approximately along the <111> direction, one of a set of complementary four element variants with uniaxial behavior is selected. However, it is difficult to accurately observe which set of the four variants is dominant using X-ray pole measurement techniques. Therefore, unlike deposits with an angle of incidence, when symmetry is broken using a magnetic field mechanism, uniaxial behavior is that the magnetic field is oriented along either the crystal direction <110> or <112> of the hexagonal template. Obtained when attached. The magnetic permeability of the difficult-to-magnetize axis when the magnetic field breaking the symmetry, which causes the set of four variants to appear, should be higher than that of the set of two variants. It's hard to think about. The coercive force is higher and the uniaxial magnetic response function is less ideal. These magnetic effects indicate that it is more difficult to prepare a set of four variants when the processing tolerance is tighter. The magnitude of the applied magnetic field depends on some prepared samples of the sample-1. Very small segments of 5 to 2 mm are studied for their individual magnetic properties. Thereby, these segments represent individual samples prepared by the magnitude and direction of the various applied magnetic fields. From these measurements, the magnetic field-based symmetry-breaking mechanism yields a minimum of 10 Oe, which breaks the observable symmetry with respect to the magnetic field. It is stated here that the applicant was interested that the magnetic field of the earth or some other stray magnetic field might affect the results. The stray magnetic field is measured in the vicinity of the substrate carrier to be approximately 1 Oe or less than 1 Oe, which is comparable to the measured ground magnetic field strength. Nevertheless, the experiment is carried out, at which time the sample substrate is mounted on the sample carrier containing the applied field magnets. During the deposition process, the substrate carrier, and therefore the applied field magnet and sample, are rotated together. This provides a constant applied magnetic field, while rotation provides an averaging mechanism for all applied stray magnetic fields. Similarly, similar experiments were performed using deposition with an angle of incidence as a technique for breaking symmetry, instead of using no applied magnetic field. In a mechanism that breaks both symmetries, the magnetic results are the same as for samples prepared without rotation. Therefore, the applicant was convinced that the stray magnetic field did not play any important role in the experimental results.
Uniaxiality in the four variants is expected to be more easily achieved if there is a better vacuum system and the Si does not have to be passivated via the H bond from the HF etching technique. It is well known that silicon dioxide sublimates at about 850 ° C to give a clean Si surface. If the treatment is carried out in an ultra-high vacuum system, a cleaner substrate is available for the initial Ag deposition. This is expected to result in higher quality membranes and membrane structures and better control over the entire process.
The measurement of the magnetic response curve is performed on the above-mentioned sample. Since the magnetic field from the permanent magnet is not perfectly uniform across the sample, the symmetry-breaking mechanism is not perfectly aligned with the <112> direction of the hexagonal template. Nevertheless, the magnetic field direction across the measured sample did not vary by more than 10 degrees and was better aligned over most of the sample. Aligned well to show the mechanism of breaking symmetry through the magnetic response curve. Figure 15 shows Sample name: LS1425_2cx Ag (200nm) / Cr (40nm) / Fe (50nm) <H <112> / Ni (5nm) / Ag (40nm) / Si (111) It is a diagram showing the Mx and My vs. Hx response of one of the samples in which the exchange-coupled and uniaxial symmetry was broken for the two variants using the structure of.
The coercive force of only Ni was measured and found to be comparable to the 10 Oe observed for the difficult-to-magnetize axis loop of this sample. Therefore, it can be seen that the exchange coupling of membrane variants is not sufficient to achieve low coercive force. The replacement coupling layer must be magnetically soft. Permalloy, a well-known soft NiFe alloy, was replaced with Ni, at which time the coercive force dropped to less than 1.0 Oe. Interestingly, however, even with the Ni template, the sample behaves losslessly unless a high driving magnetic field is used to drive the sample until it is saturated. In another measurement in FIG. 15, the sample is first driven in the easy axial direction to remove most of the domain wall, and then on the hard magnetized axis to about 80% of the magnetic field required to saturate the sample. Driven in the direction. This minor loop response is the approximate decomposition limit of the instrument 1 Shows a coercive force less than Oe. FIG. 15 clearly shows the non-linear uniaxial response function when the sample is driven to saturation and the uniaxial result similar to the ideal diagram of FIG. This applies to both the narrow magnetized difficult axis loop Mx vs. Hx [19] and the somewhat quadratic My magnetized easy axis [20] response. The hard-to-magnetize axis loop Mx behaves potentially linearly, and the My response shows a non-linear response.
Layer-structured samples of other similar compositions formed by deposition with angles of incidence show no measurable coercive force or loss for minor magnetizing difficult axis loops. For these, magnetic permeability in the range of 80 to 1000 or more is observed.
Hard disk media samples are prepared using a manufacturing hard disk media vacuum system, but with a similar but less ideal technique than the deposition technique with incident angles described above. The source bcc-d material is transmitted in a radial pattern throughout the disc at different corners across the surface of the disc. This is achieved by masking most of the disc with a circular mask with a hole in the center and depositing from a standard sputtering target. Thereby, the sputtered material is essentially radially ejected from the hole at the angle of incidence. Some samples LS SDK-0505-1 Cr (40nm) / Fe (40nm) <~ 45 / Cu (20nm) / Cr<sub>35</sub>Ta (20nm) / Glass-Ceramic and LS SDK-0505-2 CoCrPt (20nm) / Cr (40nm) <~ 45 / Cu (20nm) / Cr<sub>35</sub>Ta (20nm) / Glass-Ceramic It is prepared with a layer structure formed by.
The magnetic properties of the first sample (Fe) show anisotropic behavior and the OR is greater than 1, providing confidence for performing sample 0505-2. Fe is converted to Cr so as not to obscure the magnetic results of the harder CoCrPt alloy with the properties of soft Fe. In the prepared sample, an OR in the range of 1.05 to 1.15 is obtained when the coercive force in the circumferential direction is divided by the coercive force in the radial direction. This is desirable and the overall coercive force is in the 2500 Oe range. No X-ray data can be obtained because the polycrystalline hexagonal template layer and the amorphous and ceramic mixed substrate have a very large set of interference of diffraction peaks.
Those skilled in the art will appreciate that some changes and changes can be made to specific aspects of the material structures, devices, methods, and equipment of the invention without departing from the scope of the invention. Such changes and changes shall be covered by the claims in the specification and attached.
<figref num="1">It is the figure which plotted the magnetic anisotropy energy density with respect to the uniaxial Stoner-Wallfirth model.</figref><figref num="2">Difficult to magnetize with respect to the magnetic field applied in the direction of the difficult-to-magnetize axis Hx for the uniaxial Stoner-Wallfers model Easy to magnetize with respect to the magnetic field applied in the direction of the difficult-to-magnetize axis Hx with respect to the axis response function Mx It is a figure which shows the axis response function My.</figref><figref num="3">It is a figure which shows the conventional three orientation variant arrangement of the (110) crystal plane of a bcc crystal when compared with the atomic arrangement of the (111) crystal plane of an fcc crystal.</figref><figref num="4">Primary anisotropic energy density constant K<sub>1</sub>It is a figure which plotted the magnetic anisotropy energy density vs. the magnetization direction in the (110) texture plane of a cubic crystal when is positive.</figref><figref num="5">It is a figure which shows two of six possible orientation variants of the (110) crystal plane of a bcc-d crystal compared with the atomic arrangement of the (111) crystal plane of a hexagonal lattice template crystal.</figref><figref num="6">K<sub>1</sub>For three (b1, b2 and c1) coupled (110) variants, which is a special case of the six possible bcc-d (110) textured variants at> 0, δ = 7.5 degrees. It is a figure which plots the magnetic anisotropy energy density in-face magnetization direction, and shows that there is one energy minimum value and one energy maximum value per 180 degree rotation.</figref><figref num="7">K<sub>1</sub>Magnetic anisotropy for two (a1 and c2) coupled (110) variants, which is a special case of the six possible bcc-d (110) textured variants at> 0, δ = 5.26 degrees. It is a figure which plots the anisotropic energy density in-face magnetization direction, and shows that there is one energy minimum value and one energy maximum value per 180 degree rotation.</figref><figref num="8">K<sub>1</sub>Four (a2, b1, b2 and c1) coupled (110), which is a special case of the six possible bcc-d (110) textured variants at> 0, δ = 5.26 degrees. It is a figure which plots the magnetic anisotropy energy density in-face magnetization direction with respect to a variant, and shows that there is one energy minimum value and one energy maximum value per 180 degree rotation.</figref><figref num="9">K<sub>1</sub>Shows the hard-to-magnetize axis response function Mx and the easy-to-magnetize axis response function My to the magnetic field applied in the direction of the hard-to-magnetize axis Hx with respect to the coupled variant pairs (a1 and c2) at> 0, δ = 5.26 degrees. It is a figure.</figref><figref num="10">K<sub>1</sub>For three (b1, b2 and c1) coupled (110) variants, which is a special case of the six possible bcc-d (110) textured variants at> 0, δ = 5.264 degrees. It is a figure which plots the magnetic anisotropy energy density in-face magnetization direction, and shows that there is one energy minimum value and one energy maximum value per 180 degree rotation.</figref><figref num="11">Magnetism for two (a1 and b2) coupled (110) variants, which is a special case of the six possible bcc-d (110) textured variants when K1> 0, δ = 5.264 degrees. It is a figure which plots the anisotropic energy density in-face magnetization direction, and shows that there is one energy minimum value and one energy maximum value per 180 degree rotation.</figref><figref num="12">Shows the coordinate system that determines the angle of atomic movement during deposition with an incident angle, the substrate surface is xy, the sedimentary material is in the direction of r, and the sedimentation angle ω is the angle between the z-axis and the r vector. It is a figure which shows that.</figref><figref num="13">FIG. 5 is a schematic representation of the geometry of a new sputtering cathode with respect to a disk surface for deposition with an angle of incidence.</figref><figref num="14">Six bcc with δ = 5.264, assuming that variants a1 and c2 dominate the other four variants to produce a structure with broken symmetry when the sample is produced in a deposit with an angle of incidence. It shows the -d (110) textured variant and also shows three Ag (220) peaks.</figref><figref num="15">Fe (K) is prepared with the easy axis of magnetization in the <112> direction of the hexagonal template, assuming that the mechanism that breaks the objectivity is the magnetic field applied in the <112> direction.<sub>1</sub>It is a figure which shows the data of the uniaxial magnetizing difficulty axis response Mx function and the magnetization easy axis response function My with respect to the magnetic field applied in the direction of the magnetization difficulty axis Hx about the sample of> 0).</figref>
Every citation, both waysCites: the store holds 1 of 2
| Document | Relation | Office |
|---|---|---|
| WO99024973A1 | Cites | World Intellectual Property Organization (WIPO) |
| XU J, HOWSON M A, HUCKNALL P, HICKEY B J, VENKATARAMAN R, HAMMOND C, WALKER M J, GREIG D (Leeds Univ., Leeds, GBR),Systematic study of molecular beam epitaxy growth and magnetic properties of Fe on Au(111),J App l Phys,米国,1997年 4月15日,Vol.81 No.8, Pt.2A,Page.3908-3910 | Non-patent | – |
12 members in 5 offices
Priority claims9
| Document | Office | Kind | Date |
|---|---|---|---|
| 31592001 | United States of America | P | |
| 31592001 | United States of America | P | |
| 60315920 | United States of America | – | |
| 0227327 | United States of America | W | |
| 0227327 | United States of America | W | |
| 2001315920 | – | – | – |
| 2002027327 | – | – | – |
| US20010315920P | – | – | – |
| WO2002US27327 | – | – | – |
Members12
| Document | Office | Kind | |
|---|---|---|---|
| WO03021579A1 | World Intellectual Property Organization (WIPO) | A1 | |
| US2004058196A1 | United States of America | A1 | |
| KR20040029111A | Republic of Korea | A | |
| EP1435091A1 | European Patent Office (EPO) | A1 | |
| JP2005502199A | Japan | A | |
| US7128988B2 | United States of America | B2 | |
| KR100763285B1 | Republic of Korea | B1 | |
| EP1435091A4 | European Patent Office (EPO) | A4 | |
| JP4698142B2This record | Japan | B2 | |
| EP1435091B1 | European Patent Office (EPO) | B1 | |
| EP2808869A2 | European Patent Office (EPO) | A2 | |
| EP2808869A3 | European Patent Office (EPO) | A3 |
31 legal events, as the office reported them to INPADOC
Over the term
Point at a mark for the eventEvents
| Event | Code | |
|---|---|---|
| Cancellation because of completion of termEXPY | EXPY | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Written notification of registration of transferJAPANESE INTERMEDIATE CODE: R350R350 | R350 | |
| Request for change of ownership or part of ownershipJAPANESE INTERMEDIATE CODE: R313113S111 | S111 | |
| Written request for registration of change of domicileJAPANESE INTERMEDIATE CODE: R313531S531 | S531 | |
| Receipt of annual feesJAPANESE INTERMEDIATE CODE: R250R250 | R250 | |
| Certificate of patent or registration of utility modelJAPANESE INTERMEDIATE CODE: R150R150 | R150 | |
| First payment of annual fees (during grant procedure)JAPANESE INTERMEDIATE CODE: A61A61 | A61 | |
| Request for written amendment filedJAPANESE INTERMEDIATE CODE: A523A521 | A521 | |
| Written permission of extension of timeJAPANESE INTERMEDIATE CODE: A602A602 | A602 | |
| Written request for extension of timeJAPANESE INTERMEDIATE CODE: A601A601 | A601 | |
| Notification of appointment of power of attorneyJAPANESE INTERMEDIATE CODE: A7423RD03 | RD03 | |
| Re-examination (zenchi) completed and case transferred to appeal boardAppealJAPANESE INTERMEDIATE CODE: A912A912 | A912 | |
| Transfer to examiner for re-examination before appeal (zenchi)AppealJAPANESE INTERMEDIATE CODE: A911A911 | A911 | |
| Request for written amendment filedJAPANESE INTERMEDIATE CODE: A523A521 | A521 | |
| Decision of refusalJAPANESE INTERMEDIATE CODE: A02A02 | A02 | |
| Written permission of extension of timeJAPANESE INTERMEDIATE CODE: A602A602 | A602 | |
| Written request for extension of timeJAPANESE INTERMEDIATE CODE: A601A601 | A601 | |
| Written permission of extension of timeJAPANESE INTERMEDIATE CODE: A602A602 | A602 | |
| Written request for extension of timeJAPANESE INTERMEDIATE CODE: A601A601 | A601 | |
| Written permission of extension of timeJAPANESE INTERMEDIATE CODE: A602A602 | A602 | |
| Written request for extension of timeJAPANESE INTERMEDIATE CODE: A601A601 | A601 | |
| Notification of reasons for refusalJAPANESE INTERMEDIATE CODE: A131A131 | A131 | |
| Report on retrievalJAPANESE INTERMEDIATE CODE: A971007A977 | A977 |
Numbers
- Publication
- 4698142
- Publication, DOCDB
- 4698142
- Publication, EPODOC
- JP4698142B
- Application
- 2003525841
- Application, DOCDB
- 2003525841
- Application, EPODOC
- JP20030525841
Titles2
- Japanese
- 磁性材料構造体、磁性材料構造体を用いる装置、および磁性材料の形成方法
- English
- Magnetic material structure, equipment using magnetic material structure, and method of forming magnetic material
Classification
- CPC, 21
- C23C14/225
- G11B5/7371
- B82Y10/00
- B82Y25/00
- B82Y40/00
- C23C14/228
- C23C14/35
- C30B23/02
- C30B29/52
- G11B5/3109
- G11B5/3903
- G11B5/3909
- G11B5/667
- G11B5/851
- H01F10/08
- H01F10/28
- H01F41/303
- G11B5/7373
- G11B5/674
- G11B5/678
- Y10S428/90
- IPC, 21
- H01F10 16
- G11B5 33
- G11B5 39
- G11B5 66
- G11B5 667
- G11B5 73
- G11B5 738
- H01L21 8246
- H01L27 105
- H01L43 08
- H01L43 10
- C23C14 22
- C23C14 35
- C30B23 02
- G11B5 31
- G11B5 673
- G11B5 851
- H01F10 08
- H01F10 28
- H01F10 32
- H01F41 30