JPH0441517B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a method of forming an inductor in a Josephson circuit.

<Prior art and its problems> The important active electronic circuit element in the Josephson circuit is the Josephson switching gate, and the main circuit elements that make up these gates are the Josephson switching element or the Josephson junction element. In addition to such active circuit elements, inductors as passive elements are also important circuit elements.

This is because the magnetic coupling portion 4 (expressed as mutual inductance in circuit terms) between the input line or control line 2 of the magnetic flux quantum interference Josephson switching gate 1 and the switching loop 3 shown in FIG. ) and the Josephson model shown in Figure 1B.
It forms the self-inductance 10 of the memory loop 7 of the memory 5, and the magnetic coupling part (mutual inductance) 4 between the bit line 6 and the memory loop 7.
This is because it is necessary to form.

Of course, such inductors are needed in a variety of circuit elements, not just the two examples shown, not only in active circuit elements, but also to provide intentional time delays to signals or to provide isolation between circuits. They may also be incorporated as passive elements between active circuit stages.

In addition, in FIGS. 1A and 1B, the magnetic coupling part 4, that is, the primary of the transformer that constitutes the transformer,
Each inductor as a secondary winding is designated 9, 10, and the Josephson junction elements in switching loop 3 and memory loop 7 are designated 8.
It is shown. However, since these operations are well known, a description thereof will be omitted.

However, actually manufacturing such an inductor while ensuring an inductance value above a certain value generally requires a considerable area on a circuit forming board.

For example, considering the circuit example shown in FIGS. 1A and 1B, there is a sufficiently strong magnetic coupling (a sufficiently large mutual inductance) between the external signal lines 2 to 6 and the switching loop 3 to the memory loop 7. ) 4 and to obtain a sufficiently large inductance in the inductor 10 on the switching loop 3 or memory loop 7 side, the width and width originally required for the loop line of the switching loop 3 and memory loop 7 must be It is not possible to obtain sufficient inductance by measuring the length and area alone.

Therefore, in the inductor 10 on the switching loop 3 and memory loop 7 side,
The area of this part must be quite large.

However, this is a significant drawback for Josephson circuits, which had the potential to achieve extremely high levels of integration in the first place.

Therefore, as a means to solve this problem, an inductor forming method as shown in FIG. 1C was first considered.

This example shows a cross-sectional structure when the equivalent circuit of the Josefson memory shown in FIG. 1B is actually formed on the substrate 11.
an insulating block portion 13 in which an inductor 10 extending between a pair of Josephson junction elements 8 is disposed on a connection line portion 12 extending between each lower electrode of the Josephson junction elements 8; It was formed so as to pass over the top of the .

Certainly, if you do this, this insulating lump part 13
A portion 14 extending in the height direction is formed along the side surface of the line, and therefore, if this height is made higher than a certain level, the effective line length will increase accordingly, ensuring the necessary inductance value. , and the plane area can be kept relatively small.

However, this method of simply increasing the line area of the inductor in the height direction has another major drawback.

First, when the height of the insulating block portion 13 becomes considerably high, the planarity of the upper surface structure portion where the signal line 2 and the like are to be formed is impaired, and as a result, the unevenness becomes severe. etc., wire breaks called so-called stage breaks are likely to occur, and the yield of the overall integrated circuit concept is significantly reduced.

Secondly, even if it is not a problem compared to the planar area, it is still undesirable to increase the height in the height direction. In particular, in view of the fact that so-called three-dimensional structures are being proposed in the future, it is better that each functional layer or level should not only have good flatness but also be as thin as possible.

On the other hand, as a method to obtain an inductor with a significant value in the Josefson circuit using still another method,
As exemplified in Japanese Patent Application Laid-Open No. 57-114294, an opening is made in a ground plane made of a superconducting material formed on an insulator or semiconductor base layer, and the ground plane is passed over the ground plane. Another method is to add inductance to the superconducting line.

However, this method takes into consideration the fact that, conversely, a superconducting line placed close to a ground plane cannot exhibit significant inductance as a result of being subjected to a kind of magnetic shielding effect by the ground plane. By removing a predetermined area of the ground plane, which is the shield structure, the influence of this ground plane can be removed, and equivalently the distance between the superconducting line and the ground plane, where an inductor should be obtained, can be reduced. The aim is to obtain a result that is extremely large and equivalent to that of Totsuta.

Therefore, when using this method, the larger the area of the opening in the ground plane, the longer the line that crosses this opening increases, and the equivalent effective inductance increases accordingly. However, this also means that since there is a limit to the opening area, it is difficult to obtain such a large value of inductance.

Furthermore, this approach often presents various problems and contradictions.

For example, as specified in the publication, magnetic flux may be accidentally tramped to the edge of the opening in the ground plane. This is a factor that leads to circuit malfunction, which is by no means desirable, and to prevent this, a special structure such as a moat structure is required.

Furthermore, since an edge is formed in the opening and this becomes the magnetic pole, the effect of this must also be considered.

Furthermore, if we look at the structure exemplified in the publication, we find that the part that goes over the opening in the ground plane is slightly convex toward the opening. It is formed.

However, this depression is the result of opening an opening in the ground plane, and when an insulating layer is grown on top of it, the part of the insulating material that fills the opening becomes more depressed than the other part, and therefore, it becomes even more superfluous. When a conductive line is formed, this superconducting line is also formed due to a corresponding dent in this part, and in reality, it is a slight dent corresponding to the thickness of the ground plane. , it was never intentionally formed.

Let's consider a case where this is intentionally formed and the groove depth is sufficient. In this case, such an inductor forming method ends up with extremely disadvantageous results as described below.

If you think about it a little bit, if you intentionally make a recess in the superconducting line by sandwiching an insulating layer toward the opening in the ground plane, the line length will increase by the amount of the recess, and the inductance will correspondingly increase. It seems like it will increase.

However, as long as this technique of creating an opening in the ground plane is used, this often works in the opposite direction, lowering the inductance.

This is because, as mentioned earlier, this method removes the influence of the ground plane, which is equivalent to avoiding the ground plane at a distance from the inductor formation line. Therefore, the line section where the inductor is to be formed is concave and downwards in the direction of the recess of the aperture.This means that the line section where the inductor is to be formed is recessed in the direction of the recess of the aperture. The result is that you go.

Therefore, when forming an inductor using such a method, instead of simply growing the insulating layer between the ground plane and the superconducting line, the insulating layer between the ground plane and the superconducting line is grown, as shown in the above publication. In order to prevent the formation of dents, it is preferable to form the surface of the insulating layer as flat as possible by mounding the ground plane opening, and furthermore, as described in connection with Fig. 1C. Incidentally, it may be better to also employ an insulating lump portion 13 that rises in the height direction.

However, it is clear that doing so does not eliminate the problem of flux trapping, which is an inherent drawback of opening an opening in the ground plane; The disadvantages mentioned regarding the configuration of Figure C are also added.

The present invention has been made in view of such conventional circumstances, and in order to keep the plane area small when forming an inductor in a Josefson circuit, the thickness of the circuit layer does not need to be increased to a large extent. The flatness of the superstructure can be maintained well, and
The present invention aims to provide an inductor formation method with a new configuration that does not cause unexpected magnetic flux traps, etc., as seen in the conventional method of creating an opening in the ground plane and providing inductance to the line passing over it. It is.

<Means for Solving the Problems> In order to achieve all of the above objects, the present invention intentionally forms a deep groove in the portion of the non-conductor base layer where the inductor is to be formed in the Josefson circuit. The inductor-forming material is attached along the side and bottom surfaces, and an inductor whose inductance is increased at least by the presence of the line portion that is continuous in the depth direction along the inner surface of the groove is formed. A method for forming an inductor for a Josephson circuit specified by the following configuration requirements is provided.

<Operations and Effects> In the present invention, a groove is dug in the non-conductor base layer, and an inductor forming material is adhered along the inner surface and bottom surface of the groove, so that the depth increases along the inner surface of the groove. Only for the part of the track that continues in the direction,
An inductor can be formed that has at least an increased inductance than the existing one.

In other words, even though the inductor forming material is deposited along the inner surface of the groove, the inductance does not increase or, conversely, decrease. In manufacturing, if the groove depth is increased within a range that does not cause breakage, etc., the inductance can be increased accordingly.

After all, according to the present invention, the inductor in the Josefson circuit can be formed in a small area and with the protruding portion in the height direction kept small, so that it is possible to achieve high integration and improve product yield regarding the upper wiring layer. You can get a big effect.

<Example> FIG. 2 shows a first example of the present invention,
The equivalent circuit of the constructed circuit is shown in FIG. 1A. Therefore, corresponding components are designated by the same reference numerals as those shown in FIG. 1A.

Generally, in this type of Josefson circuit,
The substrate material is silicon, etc., but its thickness is considerably thicker than the thickness of the circuit layer in order to physically support the entire circuit, and even with some mechanical processing, it There is no problem with strength. One of the features of the present invention is that it focuses on this point.

That is, in this embodiment of the present invention, as a result, the inductor 10 magnetically coupled to the inductor 9 of the signal line 2 is changed to a portion extending between the lower electrodes of each Josefson junction element 8 (the first
In FIG. A groove 16 is formed at the location, the sides of which are preferably tapered.

The thickness of the substrate 11 is so thick as to be inconsequential compared to the depth of this groove, so that its mechanical strength is not sacrificed in any way. However, as mentioned above, as seen in the conventional example of forming an inductor by forming an opening in the ground plane, it is difficult to intentionally create a groove that is sufficiently deep compared to the resulting depression. It may be formed into

In any case, once the groove portion 16 is intentionally formed in this way, the groove portion 1
When a material such as a suitable metal for forming the lower electrodes of both Josephson junction elements 8 and the inductor 10 described above is vapor-deposited on the plane area including the groove 6 using a conventional technique, the inner surface and bottom surface of the groove 16 are deposited. The material is continuously deposited on the other hand, and the required inductor 10 is formed.

The connection portions 15 to the lower electrode of each Josefson junction element 8 on both sides of the inductor 10 thus formed have a length in the depth direction along the inner surface of the groove 16 intentionally formed according to the present invention.

Therefore, if this depth is set in design, an arbitrary required inductance value can be set for the inductor 10.
will be obtained.

The reason why the side surface of the groove 16 is tapered is to prevent the step from being broken in the portion 15 when forming the inductor 10 portion by vapor deposition.

However, if the required inductance value can be secured even if the depth is not very deep, this may not be necessary, and the side surfaces of the groove may simply be steep sides. The material for forming the inductor 10 is of course a matter of choice, but it may be a superconducting material, or it may be a normal conducting material if only the inductor needs to be formed. .

Once the desired inductor 10 is formed in this way, the groove-shaped portion formed on the surface of the inductor 10 is filled with an insulating member made of an appropriate material by an appropriate known method. A lump portion 13 is formed.

On the other hand, a barrier layer as a Josephson junction is similarly formed on the lower electrode of each Josephson junction element 8 using a known technique, and then the upper electrode of each Josephson junction element 8 and the connection line 12 between them are formed. do.

By filling the entire area with an insulating layer 19 and finally forming the signal line 2 on top of the insulating layer 19, the structure of the equivalent circuit shown in FIG. 1A can be obtained.

Note that since the Josephson barrier layer is generally extremely thin, it is simply represented by a somewhat thick line in the figure.

In any case, with the above configuration, the conventional disadvantage of having to provide the part 14 to increase the inductance value in the direction of increasing the thickness of the entire circuit device with respect to the board can be solved, and therefore the flat surface Not only physically, but also in volume and three-dimensional space.
Since the factor can be improved and the upper structure can be flattened, it is possible to reliably prevent breakage and the like when forming the upper wiring portion such as the signal line 2.

However, if the inductance value of the inductor 10 portion is simply determined in the depth direction, the depth of the groove 16 becomes quite deep, and as mentioned above, there is a step in the material such as metal for forming the inductor 10 portion. It is conceivable that there may be cases where this occurs.
If there is such a possibility, it is sufficient to change it as shown in the third figure.

That is, according to the present invention, the trenches 16 in the depth direction are intentionally formed in the substrate 11 and the inductor 10 portion is formed along the inner surface of the grooves 11, but the circuit 1 as a whole is not required. A line 12 connecting the electrodes is connected to the upper part of the inductance value.
In order to make the insulating lump portion 13 fill in the groove in the depth direction as described with reference to FIG.
As seen in the conventional example, a raised insulating lump portion 13' portion can be formed at the same time, and a line portion 14 extending in the height direction can also be formed in the upper interelectrode connection line 12.

Conversely, as mentioned above, compared to the requirement for the planar occupied area, the height direction is allowed to a certain extent, so even when applying the present invention in accordance with FIG. In order to further reduce the plane area, the groove depth may be kept shallow by allowing a slight upward bulge.

It is obvious that the embodiments shown in FIGS. 2 and 3 can also be applied to the inductors in the circuit elements of the equivalent circuit shown in FIG. It can also be applied to inductors in passive elements that do not need to be magnetically coupled to various signal lines such as lines 2 and 6 or other inductor parts;
It is also clear.

Furthermore, as so-called three-dimensional structures are developed in the future, the invention can also be applied between each layer or level and the levels above and below it.

In this case, for example, the connecting line portion 12, which is flat in FIG. It is sufficient if the lower part of the layer part is facing. In other words, when looking at the inductor portion of each layer in the vertical direction, it is possible to create a slightly nested shape, which not only reduces the plane area but also
The thickness in the height direction does not need to increase that much.

In the illustrated embodiment, the portion on which the inductor is formed is a substrate 11 made of silicon or the like, but considering the above, it can be generalized to the non-conductor base layer 11. A non-conductor is a ground or
This is a comprehensive concept for excluding conductive layer members such as planes and limiting the invention to semiconductor layers such as silicon, insulator layers, etc. in the above embodiments.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of an inductor in a conventional Josephson circuit element, and FIGS. 2 and 3 are explanatory diagrams of each embodiment of the present invention. In the figure, 2 and 6 are signal lines, and 3 is Josephson
switching loop; 7 is Josephson memory loop; 8 is Josephson junction element; 9,1
0 is an inductor, 11 is a substrate, 13 is an insulating lump portion, 14 is a line portion extending in the height direction, 15 is a line portion extending in the groove depth direction, and 16 is a groove.

Claims (1)

[Scope of Claims] 1. A method for forming an inductor on the surface of a non-conductor base layer in a Josefson circuit, comprising: intentionally forming a deep groove in a portion of the non-conductor base layer where the inductor is to be formed; The inductor-forming material is deposited along the inner surface and bottom surface of the groove, and at least the inductance of the line portion that is continuous in the depth direction along the inner surface of the groove is reduced. 1. A method for forming an inductor in a Josephson circuit, characterized by: forming an inductor with increased .