JP2019009354A - Method for manufacturing magnetic resistance change element - Google Patents

Abstract

To provide a method for manufacturing a magnetic resistance change element, by which an MTJ element can be processed with good precision.SOLUTION: A method for manufacturing a magnetic resistance change element comprises the steps of: forming first to third hard mask films over an upper electrode of a magnetic tunnel junction layer; forming a first resist pattern with first parallel lines extending in a first direction on the first hard mask film; forming a first hard mask pattern by etching the first hard mask film using the first resist pattern as a mask; forming a second resist pattern with second parallel lines extending in a second direction perpendicular to the first direction over the first hard mask pattern and the second hard mask film; forming an island-like hard mask so that island-like portions remain just at intersection points of the first parallel lines and second parallel lines by etching while using the second resist pattern as a mask; and removing, by etching, the upper electrode and the magnetic tunnel junction layer in regions excluding the island-like portions in the hard mask.SELECTED DRAWING: Figure 1

Description

  The present disclosure relates to a method of manufacturing a magnetoresistance change element.

  A magnetic tunnel junction (hereinafter referred to as MTJ) element used in a magnetoresistive random access memory (hereinafter referred to as MRAM) includes a fixed magnetic layer, a free magnetic layer, and a tunnel insulating film disposed therebetween. Including. The magnetization direction of the fixed magnetization layer is fixed, and the magnetization direction of the free magnetization layer is variable. The resistance value of the MTJ element is low when the magnetization direction of the free magnetic layer and the magnetization direction of the fixed magnetic layer are in the same parallel direction, and high when the antiparallel state is in the opposite direction. . The parallel state and the antiparallel state may be associated with 0 and 1 of the stored data, respectively.

  MRAM is classified into a write wiring type and a spin injection type from the viewpoint of a writing method. In the write wiring type, the magnetization direction of the free magnetic layer is controlled by the magnetic field generated by the current flowing through the write word line. In the spin injection type, the magnetization direction of the free magnetic layer is controlled by a spin transfer effect generated when a current is passed through the MTJ element. Currently, the spin injection type is the mainstream. This is because a write wiring is unnecessary in the spin injection type, and this is advantageous for miniaturization which is essential for increasing the capacity.

  In the spin injection type, the amount of current required for magnetic field reversal is determined by the current density. The smaller the element area, the smaller the amount of necessary reversal current. By miniaturizing the MTJ element, the operating current can be reduced and the memory capacity can be increased. Therefore, research and development for miniaturization has been energetically performed.

  In order to form a fine MTJ element with high accuracy, it is necessary to form a fine resist pattern with high accuracy. However, since the MTJ element has a pillar shape in which each element is isolated, the resist pattern also has a pillar shape in which each point is isolated. In a pillar-shaped resist pattern in which each point is isolated, problems such as pattern collapse and variation in pattern dimensions due to a resolution limit become more noticeable as the size decreases. As a result, there are problems such as a decrease in yield and characteristic variation of the manufactured magnetoresistive change element.

JP 2008-91824 A JP 2010-45398 A

  In view of the above, a method for manufacturing a magnetoresistive change element capable of processing an MTJ element with high accuracy is desired.

  A magnetoresistive change element manufacturing method includes: a hard mask film having a three-layer structure including a first hard mask film, a second hard mask film, and a third hard mask film in order from the top on an upper electrode of a magnetic tunnel junction layer. Forming a first resist pattern having a plurality of first parallel lines extending in a first direction on the first hard mask film, and etching the first hard mask film using the first resist pattern as a mask; Thus, a first hard mask pattern having a plurality of parallel lines extending in the first direction is formed, and a second direction perpendicular to the first direction is formed on the first hard mask pattern and the second hard mask film. Forming a second resist pattern having a plurality of second parallel lines extending to the first hard mask pattern using the second resist pattern as a mask By etching the second and third hard mask films, the second and third hard mask films are formed by at least one of the first to third hard mask films remaining only at the intersections of the first parallel lines and the second parallel lines. Each step includes forming an island-shaped hard mask and removing the upper electrode and the magnetic tunnel junction layer by etching in a region other than the position of the island-shaped hard mask.

  According to at least one embodiment, there is provided a method for manufacturing a magnetoresistive change element capable of processing an MTJ element with high accuracy.

It is a flowchart which shows an example of embodiment of the manufacturing method of a magnetoresistive change element. It is a figure which shows an example of the laminated structure in which the hard mask film | membrane of a 3 layer structure was formed on the upper electrode of a magnetic tunnel junction layer. FIG. 3 is a diagram illustrating a state in which a first resist pattern is formed on the stacked structure illustrated in FIG. 2. It is a figure which shows the laminated structure of the state after selectively removing the 1st hard mask film | membrane by an etching. It is a figure which shows the laminated structure of the state after removing a 1st resist pattern. It is a figure which shows the state in which the 2nd resist pattern was formed on the laminated structure. It is a figure which shows the laminated structure of the state after selectively removing the 1st hard mask pattern and the 2nd and 3rd hard mask film | membrane by an etching. It is a figure which shows the laminated structure of the state after removing a 2nd resist pattern. It is a figure which shows the laminated structure of the state after selectively removing an upper electrode layer by an etching. It is a figure which shows the laminated structure of the state after selectively removing an MTJ layer by an etching. It is a figure which shows the state which formed the 2nd resist pattern on the bark layer formed as a base layer. FIG. 12 is a diagram illustrating an example of the configuration of the bark layer.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following elements, the same or corresponding components are referred to by the same or corresponding numbers, and the description thereof is omitted as appropriate.

  FIG. 1 is a flowchart showing an example of an embodiment of a method for manufacturing a magnetoresistance change element. In manufacturing a magnetoresistive change element by the manufacturing method shown in FIG. 1, first, a laminated structure in which layers constituting the magnetoresistive change element including the magnetic tunnel junction layer are laminated is prepared. Next, in step S10, a three-layer hard mask film having a first hard mask film, a second hard mask film, and a third hard mask film is formed on the upper electrode of the magnetic tunnel junction layer in order from the top.

  FIG. 2 is a view showing an example of a laminated structure in which a three-layer hard mask film is formed on the upper electrode of the magnetic tunnel junction layer. FIG. 2A is a plan view of the top surface of the multilayer structure as viewed from above, and FIG. 2B is a cross-sectional view of the multilayer structure taken along line 2A-2A in FIG.

  2B includes a silicon substrate 1, a silicon oxide film 2, a lower electrode 3, an MTJ layer 4, an upper electrode 5, a third hard mask film 6, a second hard mask film 7, and a first structure. A hard mask film 8 is included. The silicon substrate 1 is a plate-like member on which circuit elements such as MTJ elements and electrodes are formed.

A silicon oxide film 2 is formed on the upper surface of the silicon substrate 1. In order to form the silicon oxide film 2, the silicon oxide film 2 may be grown on the surface of the silicon substrate 1 by thermal oxidation, or a CVP (Chemical Vapor Deposition: CVD) method or the like may be formed on the surface of the silicon substrate 1. SiO ₂ may be deposited by:

A lower electrode 3 is formed on the silicon oxide film 2. As the lower electrode 3, a tantalum (Ta) layer, a ruthenium (Ru) layer, and a tantalum (Ta) layer may be formed in this order by sputtering. The tantalum (Ta) layer, ruthenium (Ru) layer, and tantalum (Ta) layer may be, for example, 5 nm, 20 nm, and 15 nm thick, respectively. The bottom of the tantalum layer, due to poor adhesion between the Ru and SiO _2, a contact layer provided between the Ru and SiO ₂ in order to adhere. The ruthenium layer is a layer provided to reduce the resistance value of the lower electrode 3. The uppermost tantalum layer is an etching stopper layer when processing the MTJ element provided on the upper surface of the lower electrode 3.

  An MTJ layer 4 is formed on the lower electrode 3. As the MTJ layer 4, a fixed magnetic layer, an insulating layer, and a free magnetic layer may be formed in this order. The fixed magnetic layer, the insulating layer, and the free magnetic layer that are each layer of the MTJ element may be formed by sputtering. The fixed magnetization layer may be a 1 nm thick layer formed of, for example, CoFeB (cobalt iron boron). The insulating layer may be a 0.8 nm thick layer made of, for example, MgO. The free magnetic layer may be a 1.5 nm thick layer made of, for example, CoFeB.

  An upper electrode layer 5 is formed on the MTJ layer 4. As the upper electrode layer 5, a tantalum layer, a ruthenium layer, and a tantalum layer may be formed in this order. The tantalum layer, ruthenium layer, and tantalum layer may have a film thickness of, for example, 1 nm, 5 nm, and 120 nm, respectively.

  As described above, in step S <b> 10 of FIG. 1, the third hard mask film 6, the second hard mask film 7, and the first hard mask film 8 are formed on the upper electrode layer 5 in this order. These hard mask films may be deposited by CVD or by sputtering.

The third hard mask film 6 may be silicon oxide (SiO ₂ ). However, the third hard mask film 6 is not limited to silicon oxide (SiO ₂ ), and may be, for example, silicon nitride (SiN). The second hard mask film 7 may be polysilicon. However, the second hard mask film 7 is not limited to polysilicon, and may be, for example, silicon carbide (SiC) or carbon (C). The first hard mask film 8 may be silicon oxide (SiO ₂ ). However, the first hard mask film 8 is not limited to silicon oxide (SiO ₂ ), and may be, for example, silicon nitride (SiN). By using these materials, a function as an etching stopper of the second hard mask film 7 and a function as an adhesion layer of the third hard mask film 6 described later can be appropriately realized.

  Returning to FIG. 1, in step S <b> 11, a first resist pattern having a plurality of first parallel lines extending in the first direction is formed on the first hard mask film 8.

  FIG. 3 is a view showing a state in which a first resist pattern is formed on the laminated structure shown in FIG. FIG. 3A is a plan view of the upper surface of the laminated structure as viewed from above, and FIG. 3B shows the laminated structure taken along line 3A-3A in FIG. FIG.

  As shown in FIGS. 3A and 3B, the first resist pattern 9 is exposed to the first hard mask film by exposing the photoresist layer formed on the upper surface of the first hard mask film 8 with a desired pattern. 8 is formed on the upper surface. The resist material may be an organic material including a novolak resin as a base resin and a compound as a photosensitive agent. The first hard mask film 8 includes a plurality of parallel lines extending in the first direction (lateral direction in the drawing), and the width of each line (first parallel line) corresponds to the diameter of the magnetoresistive change element to be formed. It may be.

  Since the first resist pattern 9 formed in the exposure process has a line shape, unlike a pillar-shaped resist pattern in which each point is isolated, there is less risk of pattern collapse, and it is affected by variations in pattern dimensions due to resolution limitations. Hateful. By using the first resist pattern 9 having a line shape with good controllability as described above, it is possible to stably and accurately perform pattern transfer in the following etching process.

  Returning to FIG. 1, in step S12, the first hard mask film 8 is etched using the first resist pattern 9 as a mask to form a first hard mask pattern having a plurality of parallel lines extending in the first direction. That is, the parallel line pattern of the first resist pattern 9 is transferred to the first hard mask film 8.

  In the step of etching the first hard mask film 8, the second hard mask film 7 is used as an etching stopper. By using the second hard mask film 7 as an etching stopper, the thickness (height) of the first hard mask pattern is controlled to an appropriate thickness (that is, a thickness substantially equal to the thickness of the first hard mask film 8). Easy to do. By controlling the first hard mask pattern to an appropriate thickness, a second resist pattern described later can be stably and accurately formed.

The etching at this time may be reactive ion etching. The etching gas is a fluorine-based gas as an etching gas that can efficiently remove the first hard mask film 8 and increase the selectivity between the first hard mask film 8 and the second hard mask film 7. It's okay. Examples of such a gas include CF ₄ , CHF ₃ , a mixed gas of CF ₄ and Ar, a mixed gas of CHF ₃ and Ar, and the like. The selection ratio at this time is preferably 5 or more.

  FIG. 4 is a diagram showing a stacked structure in a state after the first hard mask film 8 is selectively removed by etching. FIG. 4A is a plan view of the top surface of the multilayer structure as viewed from above, and FIG. 4B is a diagram of the multilayer structure taken along line 4A-4A in FIG. FIG.

  As shown in FIGS. 4A and 4B, the portion of the first hard mask film 8 that is not covered with the first resist pattern 9 is removed by etching, so that the portion below the first resist pattern 9 is removed. A first hard mask pattern 8L is formed. That is, the first hard mask film 8 remains only below the parallel line pattern of the first resist pattern 9, and the remaining first hard mask film 8 becomes the first hard mask pattern 8L. In a region other than the position where the first hard mask pattern 8L as the parallel line pattern is present, the second hard mask film 7 is the uppermost layer.

  The third hard mask film 6 provided below the second hard mask film 7 is formed of a material having a property of being in close contact with the upper electrode layer 5 and the second hard mask film 7. If the second hard mask film 7 is formed directly on the upper surface of the upper electrode layer 5 without providing the third hard mask film 6, the second hard mask film 7 can be easily peeled off from the upper electrode layer 5. There is sex. By providing the third hard mask film 6 between the upper electrode layer 5 and the second hard mask film 7, the adhesion of the second hard mask film 7 can be improved.

  Returning to FIG. 1, in step S13, the first resist pattern 9 is removed by ashing or an organic solvent. At this time, when the second hard mask film 7 is polysilicon, ashing using oxygen as a reaction gas may be performed. When the second hard mask film 7 is silicon carbide or carbon, a peeling process using an organic solvent as a peeling solution may be performed.

  FIG. 5 is a view showing the laminated structure in a state after the first resist pattern 9 is removed. FIG. 5A is a plan view of the top surface of the multilayer structure as viewed from above, and FIG. 5B is a diagram of the multilayer structure taken along line 5A-5A in FIG. FIG. As shown in FIGS. 5A and 5B, only the first hard mask pattern 8 </ b> L having a parallel line shape is formed on the upper surface of the second hard mask film 7 by removing the first resist pattern 9. It has become a state.

  Returning to FIG. 1, in step S14, a second resist pattern having a plurality of second parallel lines extending in a second direction perpendicular to the first direction is formed on the first hard mask pattern 8L and the second hard mask film 7. Form.

  FIG. 6 is a diagram illustrating a state in which the second resist pattern is formed on the laminated structure. 6A is a plan view of the upper surface of the laminated structure as viewed from above, and FIG. 6B shows the laminated structure taken along line 6A-6A in FIG. FIG.

  As shown in FIGS. 6A and 6B, the photoresist layer formed on the upper surfaces of the first hard mask pattern 8L and the second hard mask film 7 is exposed in a desired pattern, whereby a second resist pattern is obtained. 10 is formed. The resist material may be an organic material including a novolak resin as a base resin and a compound as a photosensitive agent. The second resist pattern 10 includes a plurality of parallel lines extending in the second direction (vertical direction in the drawing), and the width of each line (second parallel line) corresponds to the diameter of the magnetoresistive change element to be formed. It may be.

  Since the second resist pattern 10 formed in the exposure process has a line shape, unlike a pillar-shaped resist pattern in which each point is isolated, there is less risk of pattern collapse, and the second resist pattern 10 is affected by variations in pattern dimensions due to resolution limitations. Hateful. By using the line-shaped second resist pattern 10 having good controllability in this way, it is possible to stably and accurately perform pattern transfer in the subsequent etching process.

  In addition, in order to form the second resist pattern 10 stably and accurately, the height of the unevenness of the formation surface is negligible in the exposure process, that is, the height of the first hard mask pattern 8L is negligible. It is preferable that the height is. As described above, since the thickness of the first hard mask pattern 8L is appropriately controlled so as not to be too thick, the presence of the first hard mask pattern 8L is an obstacle when the second resist pattern 10 is formed. Never become. In particular, if a sufficiently thick park layer is provided below the photoresist layer as will be described later, the thickness of the first hard mask pattern 8L in the formation of the second resist pattern 10 is substantially ignored. It becomes possible.

  Returning to FIG. 1, in step S15, the first hard mask pattern 8L and the second and third hard mask films 7 and 6 are etched using the second resist pattern 10 as a mask. Thus, an island-shaped hard mask formed by at least one of the first to third hard mask films 8, 7, and 6 remaining only at the intersection of the first parallel line and the second parallel line is formed. To do.

In this etching process, the upper electrode layer 5 is used as an etching stopper. The etching may be reactive ion etching. The etching gas may be, for example, CF ₄ , CHF ₃ , a mixed gas of CF ₄ and Ar, a mixed gas of CHF ₃ and Ar, or the like. The selection ratio at this time is preferably 5 or more.

  FIG. 7 is a diagram showing a stacked structure after the first hard mask pattern 8L and the second and third hard mask films 7 and 6 are selectively removed by etching. FIG. 7A is a plan view of the upper surface of the laminated structure as viewed from above, and FIG. 7B shows the laminated structure taken along line 4A-4A in FIG. FIG.

  As shown in FIGS. 7A and 7B, at portions other than the intersection (see FIG. 6A) between the second resist pattern 10 and the first hard mask pattern 8L, the first hard mask pattern 8L, The second hard mask film 7 and the third hard mask film 6 are removed. As a result, the first hard mask 8P, which is the remaining portion of the first hard mask film 8, the second hard mask 7P, which is the remaining portion of the second hard mask film 7, and the remaining portion of the third hard mask film 6. An island-shaped hard mask 20P including the three hard masks 6P is formed. The hard mask 20P has an island-like shape with each point isolated, formed only at the intersections of the plurality of first parallel lines and the plurality of second parallel lines. In a region other than the position where the island-like hard mask 20P is present, the first hard mask film 8, the second hard mask film 7, and the third hard mask film 6 are removed, and the upper electrode layer 5 is the most layered structure. It is the upper layer.

  In the example shown in FIGS. 7A and 7B, the second resist pattern 10 remains on the hard mask 20P. However, at the stage where the etching in step S15 in FIG. The second resist pattern 10 may be completely removed. Since it is assumed that the second resist pattern 10 is removed at a position other than the intersection of the plurality of first parallel lines and the plurality of second parallel lines, the second resist pattern 10 is naturally also at the position of the intersection. It may have been removed.

  Further, in the second resist pattern 10 having the line shape shown in FIG. 6A, the portions other than the intersection are first to third hard masks located below the portions as shown in FIG. 7A. It is assumed that the films 8 to 6 are completely removed by the progress of etching. Accordingly, a part of the first to third hard mask films 8 to 6 may be removed even at the position of the intersection by the progress of such etching. For example, a part of the first hard mask film 8 is removed even at the position of the intersection, and the hard mask 20P includes a part of the first hard mask 8P, the second hard mask 7P, and the third hard mask 6P. May be included. Alternatively, for example, the first hard mask film 8 may be completely removed, and the hard mask 20P may include only the second hard mask 7P and the third hard mask 6P. Alternatively, for example, the entire first hard mask film 8 and a part of the second hard mask film 7 are removed, and the hard mask 20P includes only a part of the second hard mask 7P and the third hard mask 6P. Also good. Alternatively, for example, the first hard mask film 8 and the second hard mask film 7 may be completely removed, and the hard mask 20P may include only the third hard mask 6P.

  Returning to FIG. 1, in step S16, the second resist pattern 10 is removed by ashing or an organic solvent. This step may be performed only when the second resist pattern 10 remains after etching. Even if the second resist pattern 10 remains after etching, this step is not necessarily performed.

  FIG. 8 is a view showing the laminated structure in a state after the second resist pattern 10 is removed. FIG. 8A is a plan view of the top surface of the multilayer structure as viewed from above, and FIG. 8B is a diagram of the multilayer structure taken along line 8A-8A in FIG. FIG. As shown in FIGS. 8A and 8B, by removing the second resist pattern 10, only the island-like hard mask 20 </ b> P having an independent shape at each point is formed on the upper surface of the upper electrode layer 5. It is in a formed state.

  Returning to FIG. 1, in step S <b> 17, the upper electrode layer 5 is etched using the hard mask 20 </ b> P as a mask to form a metal material hard mask having island-like shapes with isolated points. That is, the island-like pattern of the hard mask 20P is transferred to the upper electrode layer 5 by removing the upper electrode layer 5 by etching in a region other than the position of the island-like hard mask 20P.

In the step of etching the upper electrode layer 5, the uppermost layer of the MTJ layer 4 is used as an etching stopper. The etching at this time may be reactive ion etching using, for example, CF ₄ , CHF ₃ , a mixed gas of CF ₄ and Ar, a mixed gas of CHF ₃ and Ar, or the like as an etching gas. The selection ratio at this time may be 10 or more, for example.

  FIG. 9 is a diagram showing a stacked structure after the upper electrode layer 5 is selectively removed by etching. FIG. 9A is a plan view of the top surface of the multilayer structure as viewed from above, and FIG. 9B is a diagram of the multilayer structure taken along line 9A-9A in FIG. FIG.

  As shown in FIGS. 9A and 9B, a portion of the upper electrode layer 5 not covered with the hard mask 20P is removed by etching, so that a metal hard mask 5P is formed under the hard mask 20P. It is formed. That is, the upper electrode layer 5 remains only below the island-like pattern of the hard mask 20P, and the remaining upper electrode layer 5 becomes the metal hard mask 5P. In a region other than the position where the metal hard mask 5P exists, the MTJ layer 4 is the uppermost layer of the laminated structure.

  In the example shown in FIGS. 9A and 9B, only the portion corresponding to the third hard mask 6P of the hard mask 20P remains at the end of etching. The portion of the hard mask 20P remaining at this time is not limited to this example.

  Returning to FIG. 1, in step S18, the MTJ layer 4 is etched using the metal hard mask 5P (and the remaining hard mask 20P) as a mask to form pillar-shaped MTJ elements. That is, the MTJ layer 4 is removed by etching in a region other than the position of the island-shaped hard mask 20P (and the metal hard mask 5P), and the island-shaped pattern of the metal hard mask 5P is transferred to the MTJ layer 4.

In the step of etching the MTJ layer 4, the lower electrode 3 is used as an etching stopper. The selection ratio at this time may be 5 or more, for example. The etching may be reactive ion etching. The etching gas may be any gas that reacts with CoFeB and MgO, which are the materials of the MTJ layer 4. For example, an ethanol-based gas may be used as the etching gas. Alternatively, a mixed gas of CO and NH ₃ or CH ₃ OH may be used as an etching gas.

  FIG. 10 is a view showing a stacked structure after the MTJ layer 4 is selectively removed by etching. FIG. 10A is a plan view of the upper surface of the laminated structure as viewed from above, and FIG. 10B shows the laminated structure cut along the line 10A-10A in FIG. FIG.

  As shown in FIGS. 10A and 10B, the portion of the MTJ layer 4 not covered with the metal hard mask 5P is removed by etching, so that the MTJ element 4P is formed below the metal hard mask 5P. It is formed. That is, the MTJ layer 4 remains only below the island-like pattern of the metal hard mask 5P, and the remaining MTJ layer 4 becomes the MTJ element 4P. In a region other than the position where the MTJ element 4P exists, the lower electrode 3 is the uppermost layer of the laminated structure. In the example shown in FIGS. 10A and 10B, only the portion corresponding to the metal hard mask 5P remains at the end of etching. For example, since the third hard mask 6P still remaining in the states of FIGS. 9A and 9B is not a component of the magnetoresistive element, it is completely removed in the states shown in FIGS. 10A and 10B. It is preferable that That is, in the state shown in FIGS. 10A and 10B, it is preferable that only the upper electrode (metal hard mask 5P) that is a component of the magnetoresistive change element remains on the MTJ element 4P. .

  As described above, a magnetoresistive change element in which the pillar-shaped MTJ element 4P and the upper electrode 5P are formed on the lower electrode 3 is obtained. In the example of FIG. 10A, for convenience of illustration, a total of four MTJ elements 4P are formed by two vertically and horizontally. Actually, a desired memory can be obtained by arranging a large number of MTJ elements 4P vertically and horizontally. A two-dimensional matrix of magnetoresistive elements having capacitance may be formed.

  In the above-described method of manufacturing a magnetoresistive change element, both the first resist pattern 9 and the second resist pattern 10 formed by exposure are line-shaped. Therefore, unlike the case of a pillar-shaped resist pattern in which each point is isolated, there is less risk of pattern collapse and it is less susceptible to variations in pattern dimensions due to resolution limitations. Thereby, pattern transfer in the etching process can be stably realized with high accuracy.

  As described above, in order to stably and accurately form the second resist pattern 10, it is preferable that the height of the formation surface to be negligible in the exposure process. As will be described below, if a sufficiently thick park layer is provided below the photoresist layer, the thickness of the first hard mask pattern 8L in the formation of the second resist pattern 10 is substantially ignored. It becomes possible to do.

  FIG. 11 is a diagram illustrating a state in which the second resist pattern 10 is formed on the bark layer formed as the base layer. FIG. 11A is a plan view of the upper surface of the laminated structure as viewed from above, and FIG. 11B shows the laminated structure taken along line 11A-11A in FIG. FIG.

  When forming a second resist pattern having a plurality of second parallel lines on the first hard mask pattern 8L and the second hard mask film 7 in step S14 of FIG. 1, the bark layer 30 is formed as shown in FIG. It may be formed as an underlayer. Specifically, in the step of forming the second resist pattern 10, a bark film 30 having a thickness equal to or greater than the thickness of the first hard mask film 8 is formed, and the second resist pattern 10 is formed on the bark film 30. You can do it. In FIG. 11, in order to make the relationship between the thicknesses of the first hard mask pattern 8L and the bark layer 30 easier to see, the arrangement direction of the laminated structure is different from that in FIG. The laminated structure in the arrangement state in which the lines extend in the horizontal direction of the drawing is shown.

  As shown in FIG. 11B, the thickness of the bark layer 30 is larger than the thickness (height) of the first hard mask pattern 8L, and the bark layer 30 covers the first hard mask pattern 8L. It is formed. For this reason, the upper surface of the bark layer 30 is not affected by the first hard mask pattern 8L, and becomes a substantially flat plane with no unevenness. Accordingly, when the second resist pattern 10 is formed on the upper surface of the bark layer 30, an appropriate exposure process can be performed, and the second resist pattern 10 with high accuracy can be formed.

  FIG. 12 is a diagram illustrating an example of the configuration of the bark layer 30. The bark layer 30 includes an SOG (Spin on Glass) film 31 as an upper layer and an SOC (Spin on Carbon) film 32 as a lower layer.

The main component of the SOG film 31 may be SiO _{2. As} an additive, methylsilsesquioxane for improving adhesion, a photoacid generator, a photosensitive resin composition for improving exposure performance, and antireflection A film agent may be included. The SOG film 31 has a function as an antireflection film as well as a function of reducing the resist film thickness. The main component of the SOC film 32 may be carbon. The SOC film 32 has a function as an antireflection film as well as a function of flattening the steps. The SOC film 32 may also serve as a hard mask.

  As described with reference to FIG. 2, in step S <b> 10 of FIG. 1, the third hard mask film 6, the second hard mask film 7, and the third hard mask film 6 are formed on the upper electrode layer 5. The film thickness of the first hard mask film 8 at this time is preferably set to such a thickness that the step of the first hard mask pattern 8L can be flattened by the bark layer 30 having the above structure.

  As mentioned above, although this invention was demonstrated based on the Example, this invention is not limited to the said Example, A various deformation | transformation is possible within the range as described in a claim.

DESCRIPTION OF SYMBOLS 1 Silicon substrate 2 Silicon oxide film 3 Lower electrode 4 MTJ layer 5 Upper electrode layer 6 3rd hard mask film 7 2nd hard mask film 8 1st hard mask film 9 1st resist pattern 10 2nd resist pattern

Claims (5)

A three-layer hard mask film having a first hard mask film, a second hard mask film, and a third hard mask film is formed on the upper electrode of the magnetic tunnel junction layer from the top,
Forming a first resist pattern having a plurality of first parallel lines extending in a first direction on the first hard mask film;
Etching the first hard mask film using the first resist pattern as a mask to form a first hard mask pattern having a plurality of parallel lines extending in the first direction;
Forming a second resist pattern having a plurality of second parallel lines extending in a second direction perpendicular to the first direction on the first hard mask pattern and the second hard mask film;
By etching the first hard mask pattern and the second and third hard mask films using the second resist pattern as a mask, the first hard line remaining only at the intersection of the first parallel line and the second parallel line. Forming an island-shaped hard mask formed of at least one of the first to third hard mask films;
A method of manufacturing a magnetoresistive change element, comprising: removing each of the upper electrode and the magnetic tunnel junction layer by etching in a region other than the position of the island-shaped hard mask.   2. The method of manufacturing a magnetoresistive element according to claim 1, wherein the second hard mask film is used as an etching stopper in the step of etching the first hard mask film.   3. The method of manufacturing a magnetoresistive element according to claim 1, wherein the third hard mask film is formed of a material having a property of being in close contact with the upper electrode and the second hard mask film.   4. The first hard mask film and the third hard mask film are silicon oxide films or silicon nitride films, and the second hard mask film is a polysilicon film, a silicon carbide film, or a carbon film. The manufacturing method of the magnetoresistive change element of description. The step of forming the second resist pattern includes:
Forming a bark film having a thickness equal to or greater than the thickness of the first hard mask film;
5. The method of manufacturing a magnetoresistive change element according to claim 1, further comprising forming the second resist pattern on the bark film.

Images