JP2008285690A - Method for making one-dimensional structure array or crossbar structure on substrate - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an easy fabrication method for dense topographically smooth or coplanar nano-line arrays or nano-crossbar structures. <P>SOLUTION: A dense topographically smooth or coplanar alternating line arrays is fabricated on a substrate, following the steps: (i) setting a substrate; (ii) setting a slits-having shadow mask 5 over the substrate in close proximity; (iii) feeding a raw material to deposit a first layer on the substrate; (iv) moving the shadow mask relative to the substrate by a defined distance rectangular to the length direction of the slits, keeping parallel to the substrate; (v) feeding another raw material to deposit a second layer on the substrate; (vi) moving the shadow mask relative to the substrate by a defined distance rectangular to the length direction of the slits, keeping parallel to the substrate; and (vii) repeating the above steps (iii)-(vi) until the total moved distance of the shadow mask or the substrate reaches the slits' pitch. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a method in which one-dimensional structures are arranged densely and at a small pitch on a substrate, and the surface after the production of the one-dimensional structure arrangement is kept smooth (or planar). The present invention also relates to a method for producing a crossbar structure, which is important in applying the above one-dimensional structure array to a device.

  Several methods for patterning or producing a dense one-dimensional structure array on a substrate are conventionally known, and examples thereof include photolithography, electron beam lithography, and nano printing.

The crossbar structure has been proposed as a basic structure for future computers (Non-Patent Document 1). In the simplest case, the crossbar structure is composed of two intersecting wirings (a kind of one-dimensional structure) and a functional material sandwiched between the intersections of these wirings. Here, the functional substance functions as a switch whose state is changed (on / off state: different resistance) by applying, for example, a potential difference pulse to two wirings (Non-Patent Document 2). This can be used, for example, for information storage or information processing.
Furthermore, in order to fabricate nanodevices arranged in parallel, the use of a moving shadow mask has been performed before (Non-Patent Documents 3 to 6).

Heath, J. R. et al: Science, 1998, 280, (5370), 1716-1721. Collier, C. P. et al: Science, 1999, 285, (5426), 391-394. Ono, K. et al: Jpn. J. Appl. Phys. Part 1-Regul. Pap. Short Notes Rev. Pap., 1996, 35, (4A), 2369-2371. Deshmukh, M. M. et al: Appl. Phys. Lett., 1999, 75, (11), 1631-1633. Luthi, R. et al: Appl. Phys. Lett., 1999, 75, (9), 1314-1316. Egger, S. et al: Nano Lett., 2005, 5, (1), 15-20. Egger, S. et al: J. Comb. Chem., 2006, 8, (3), 275-279.

  However, according to the conventional patterning method, it is very difficult or impossible to produce a planar and nano-sized (small pitch) one-dimensional structure array and a crossbar structure. An object of the present invention is to provide a method capable of easily producing a one-dimensional structure array or crossbar structure having a smooth (or planar) surface and closely arranged.

[Summary of the Invention]
In order to achieve the above object, the present inventors have come up with a new strategy for creating a one-dimensional structure array of small pitches using a movable shadow mask close to a substrate. Successfully completed the invention.
That is, the first (first invention) of the present invention is a method for producing a dense one-dimensional structure array on a substrate while keeping the surface in a smooth or planar state, and includes the following steps: This is a manufacturing method (see FIGS. 3 to 6).
(i) Set the substrate in the reaction chamber;
(ii) A shadow mask having a slit (the slit width is d and the slit pitch is f) is set close to the substrate;
(iii) supplying a raw material from a raw material supply source to deposit a first layer having a predetermined thickness on the substrate in correspondence with the slit of the shadow mask;
(iv) The shadow mask is perpendicular to the longitudinal direction of the slit by a predetermined distance u / u ₁ (u is smaller than d) while maintaining parallelism with the substrate. (Inversely, the substrate may be moved with respect to the shadow mask); the shadow mask is moved not only in the direction perpendicular to the slit length direction but also in the slit length direction. This can be done with some movement (hereinafter the same).
(v) supplying another source material from a source material source to deposit a second layer of a predetermined thickness on the substrate, corresponding to the slits of the shadow mask;
(vi) The shadow mask is moved in a direction perpendicular to the length direction of the slit by a predetermined distance u / u ₂ while maintaining parallelity with the substrate (with respect to the shadow mask). The substrate may be moved).
(vii) The above steps (iii) to (vi) are repeated until the total movement distance of the shadow mask or the substrate reaches the slit pitch (f).

The second (second invention) of the present invention is a method for producing a dense one-dimensional structure array on a substrate having a smooth or flat surface (three or more raw material supply sources). In order to form the components of FIG. 1, it is composed of three or more different raw materials (see FIG. 1), which is a manufacturing method including the following steps.
(i) Set the substrate in the reaction chamber;
(ii) A shadow mask having a slit (the slit width is d and the slit pitch is f) is set close to the substrate;
(iii) A raw material is supplied from a raw material supply source, and a first layer having a predetermined thickness is deposited on the substrate so as to correspond to the slit of the shadow mask;
(iv) The shadow mask is perpendicular to the longitudinal direction of the slit by a predetermined distance u / u ₁ (u is smaller than d) while maintaining parallelism with the substrate. Move in any direction (the substrate may be moved relative to the shadow mask).
(v) supplying another raw material from the raw material supply source, and depositing a second layer having a predetermined thickness corresponding to the slit of the shadow mask on the substrate;
(vi) The shadow mask is moved in a direction perpendicular to the length direction of the slit by a predetermined distance u / u ₂ while maintaining parallelism with the substrate (shadow mask). The substrate may be moved with respect to the substrate).
(vii) An n-th (n) -th source material is supplied from a source material supply source for a predetermined time, and an n-th layer corresponding to the shadow mask slit is deposited on the substrate;
The (viii) the shadow mask, while maintaining the concurrency of the substrate by a predetermined distance u / u n to the substrate _(u / u _n is less than d), in the longitudinal direction of the slit The substrate moves in a direction perpendicular to the substrate (the substrate may be moved with respect to the shadow mask).
(ix) Repeat steps (iii) to (viii) until the total movement distance of the shadow mask or substrate reaches the slit pitch (f).

The third (third invention) of the present invention is a method for producing a crossbar structure (see FIGS. 9-11) including the following steps.
(1) A dense one-dimensional structure array whose surface is maintained in a smooth or planar state is produced on the substrate by any of the above methods;
(2) depositing a functional layer on the one-dimensional structure array using a shadow mask having openings of a predetermined shape;
(3) Except for using a shadow mask whose slit length direction is perpendicular to the slit length direction used in the step (1), the function is the same as in the step (1). Creating a dense one-dimensional structural array on the layer with the surface kept smooth or planar;

The first invention method has the following advantages.
a) The line width (w) of the final structure is very small. If the manufacturing conditions are optimized, a structure of 5 nm or less can be formed. The line width (w) can be selected smaller than the width (D) of the opening (slit).
b) The surface of the produced one-dimensional array structure is finally smooth.
c) If alignment is performed correctly, the surface of the final structure is planar even when many one-dimensional structures are considered (see FIG. 5).
d) The method of the present invention can be applied to the production of structures made of various materials (metals, alloys, semiconductors and insulators). A single material may be used as long as a gaseous beam can be formed and the beam can condense on a solid surface. Many types of material sources are used to form gaseous beams, such as thermal evaporators, sputter sources, pulsed laser deposition sources, pulsed valve sources. ), Ion beam sources, spray sources, etc. can be used.
e) During the fabrication process, the substrate and shadow mask can be heated and cooled. The temperature of the substrate and shadow mask can be controlled independently.
f) The method of the present invention can be applied to various substrates. For the production of a smooth structure, a substrate having a smooth surface is preferred. Since there is no process for applying mechanical stress to the production, a mechanically fragile substrate can be used. Since the substrate is not exposed to a solvent or a resist during the manufacturing process, a chemically sensitive or porous substrate can be easily used.
g) The method of the invention requires only a few steps. This develops the device and reduces manufacturing costs.
h) The method of the present invention can be performed in a well-controlled environment. Undesirable effects on device performance due to contamination are avoided.
i) The method of the present invention can also be used to fabricate many structures in parallel.

The second invention method has the following advantages.
All of the advantages of the first invention method also apply to this second invention method.
・ The most important point is that the number of processes is small, so even if the number of materials increases, the number of additional processes is small.
If more than 2 materials are used, the fabrication method can be further optimized, for example, the interface properties of different layers can be improved, for example, appropriate lattice constants and coupling constants can be selected, or A diffusion barrier can be provided.

The third invention method has the following advantages.
The advantages of the first and second invention methods are all applicable to the third invention method. A characteristic application of the third invention method will be described. The advantages of the third invention method are as follows.
a) A small array pitch (p) in a one-dimensional structure array allows the creation of a very high density crossbar structure.
b) The surface of the one-dimensional structure array is smooth. Therefore, the functional layer and the top one-dimensional structural arrangement are less likely to be distorted.
c) The method of the present invention can be applied to the production of structures composed of various materials (metals, alloys, semiconductors and insulators). Considering functional layers, studies of suitable functional layers that are possible for various applications are ongoing, and the flexibility of material selection in the present invention will become increasingly important.
d) During the fabrication process, the substrate and shadow mask can be heated and cooled. The temperature of the substrate and shadow mask can be controlled independently.
e) The method of the present invention can be applied to various types of substrates.
f) The method of the invention requires only a relatively small number of steps.
g) The method of the invention can be performed in a well-controlled environment.
h) The method of the present invention can also be used to produce a number of structures in parallel.

BEST MODE FOR CARRYING OUT THE INVENTION

The method of the present invention (first, second and third methods) will be described in more detail with reference to the drawings (an example of a production apparatus is also described in Non-Patent Documents 6 and 7).
FIG. 1 is a schematic longitudinal sectional view of a typical reaction chamber (manufacturing apparatus) used in the method of the present invention. The evaporation source assembly (1) with at least two independent material sources (1a, 1b) generates at least two different gas beams.
The shadow mask (5) is placed parallel to the surface of the substrate and close to the substrate. The shadow mask and the substrate are exposed to the material beam (4). By providing a high-precision mechanical drive device, the mask can be moved with respect to the substrate (fixed) (the mask may be fixed and the substrate may be moved). The opening shape (6) in the shadow mask (5) determines the lateral geometric shape of the structure (8) produced on the substrate (7).
Typical values for these sizes are: For example, the size (s) of the raw material source is 2 mm, the distance (a) between the raw material source and the mask is 20 cm, and the distance (b) between the mask and the substrate is about 1 μm. These values determine the width of the boundary region (q = sb / a), but it can be changed as required.
An additional variable heat radiator (20) may be used to compensate for temperature changes when the source gas is thermally radiated. Thereby, the alignment error due to temperature change can be minimized.
This manufacturing method can be applied to various types of substrates. Examples of the substrate (7) include well-known substrates such as silicon, silicon oxide, silicon nitride, sapphire, and metal. In addition, since there is no process in which mechanical stress is applied to the production, a substrate such as a mechanically fragile thin film can be used. Since the substrate is not exposed to a solvent or a resist in the manufacturing process, a chemically sensitive substrate (for example, a halide) or a porous substrate can be easily used.

  Usable material sources or material source sources (1 / 1a / 1b) include thermal evaporators, sputter sources, ion sources, pulsed laser deposition sources. source, pulsed valve source, spray sources, or other types of material sources can be used. Furthermore, electrons and photons can be used for structural modification / modification (for example, polymerizing organic substances).

  The material source generally consists of several components (1a / 1b / ...) to deposit different substances. In a typical case, the material source is binary (1a / 1b), one of which is a gas substance for conductors and the other is a gas substance for insulators.

The reaction chamber (12) is preferably a vacuum chamber. In order to produce a clean one, an ultra-high vacuum chamber (UHV) having a pressure of 10 ⁻⁷ or less is most preferably used, but a slightly higher degree of vacuum is sufficient. What is important for evaporation assuming passing through a shadow mask is that the mean free path of atoms / molecules is not smaller than the source-substrate distance (a + b, ie, a). is there. At a pressure of 0.1 Pa, the average free movement distance of atoms / molecules is about 40 cm, which is a sufficient value. If the average free movement distance of atoms / molecules is smaller than the distance between the source and the substrate, the collision of the beam particles causes a diffusion beam, resulting in blurring of the final structure and loss of accuracy.

The shadow mask (5) is usually made of a thin film having holes (openings). The opening is a slit, and the slit may be linear or curved. The opening can be created by various techniques (eg, focused ion beam (FIB), electron beam lithography, photolithography, etc.). Two examples of shadow masks with slits are shown in FIG. In order to produce a dense one-dimensional structure array whose surface is kept smooth or planar, the slit width (d) is preferably 20 nm to 1 μm, and the slit pitch (f) is preferably 100 nm to 10 μm.
The predetermined moving distance of the shadow mask relative to the substrate (fixed) (and vice versa) (u ₁ , u ₂ ) determines the line width (w ₁ , w ₂ ) of the one-dimensional structure array on the substrate, This must be smaller than the slit width (d). Preferably, it is 5 nm-250 nm.
The distance b between the shadow mask and the substrate is preferably 100 nm to 10 μm.
As shown in FIG. 3, the line bundle is produced while being partially overlapped laterally. The technique of forming microstructures while overlapping allows the use of relatively large openings and thus contributes to reducing the problem of clogging of openings (or slits).
To avoid undesired deposition of structural fragments, a second shadow mask having a larger opening may be used in combination with the opening or closure of the first shadow mask.
The substrate and / or shadow mask can be heated or cooled. As is well known in thin film and surface science, depending on the material to be deposited (or a combination of the deposited material and the substrate), temperature control can improve the quality of the thin film or interface. Temperature control also helps prevent (or delay) clogging of the shadow mask.

Next, referring to FIG. 3 and FIG. 1, the manufacturing process of the one-dimensional structure array will be described according to the first invention method. In the example of FIG. 3, the sample (substrate) is fixed and the shadow mask is moved.
Set the substrate in the reaction chamber. The shadow mask has a slit opening (two examples of the shadow mask layout are shown in FIGS. 2A and 2B) (the direction of the slit is perpendicular to the cross section).
Two material sources are used for two different substances. One is a conductor (deposition by the supply source 1a), and the other is an insulator (deposition by the supply source 1b). The substrate is an insulator.
The shadow mask is disposed close to the substrate surface (typically a distance of 1 μm or less).
<Production>
(i) Activate the first material supply source (1a) and generate a gaseous beam of material therefrom. Part of the generated gaseous beam passes through the opening of the shadow mask (5) and is deposited on the substrate surface. As a result, a first layer of structural elements is formed (8a ′, 8a ″, FIG. 3A).
The horizontal shape of this structural element is determined by the opening of the shadow mask (5), and its thickness is determined by the processing time (when the release rate of the substance originating from the material source is constant).
(ii) The shadow mask is moved by a predetermined distance u _{1 in} a direction perpendicular to the length direction of the slit parallel to the substrate surface. Here, the movement of the shadow mask may be the movement v ₁ in the slit length direction together with the movement (u ₁ ).
(iii) The second material supply source (1b) is activated, and a second layer is formed on the top as shown in FIG. 3B (8b ′, 8b ″).
(iv) The shadow mask is moved in parallel with the substrate surface by a predetermined distance u _{2 in} a direction perpendicular to the length direction of the slit. Here, the movement of the shadow mask may include the movement v ₂ in the slit length direction together with the movement (u ₂ ).
The above steps (i) to (iv) are repeated several times until the total movement distance (of the shadow mask) in the direction perpendicular to the slit length reaches the slit pitch (f). (The eleventh structural layer is shown in FIGS. 3C and 4). The completed structure is shown in FIG.

<Some additional items regarding production>
In the case of manufacturing a long line, it is preferable to use a shadow mask having a slit opening as shown in FIG. If the slit is too long, the shadow mask becomes fragile, which is considered to avoid it. When used, each structural element is created in two steps. During each step, the shadow mask is moved along the direction of the slit (in FIG. 2B, the dotted line indicates the position of the slit in the second stage).
In order to avoid errors due to thermal expansion, it is preferable to keep the thermal radiation on the shadow mask and the sample (substrate) constant. For this reason, additional heat radiation that keeps the total power supply at a certain level is preferably used.
To increase the production speed, all material sources can always be kept “in operation”, but unused material beams need to be blocked using a shutter.
The source of material can be moved by a mechanical drive system. This is because the central axis is aligned so that the position of the structure formed on the substrate through the shadow mask is constant even if the source used is changed.

<Description of the fabricated structure>
The properties of the final structure shown in FIGS. 5 and 6 are as follows.
The structure consists of a conductor (dark gray in FIGS. 5 and 6) and an insulator (light gray in FIGS. 5 and 6). All conductive layers are isolated from each other by an insulator. The width D of this structural element (see FIG. 6A) is approximately equal to the slit width d of the shadow mask.
The surface of the structure, the insulating material of the electrically conductive material apparent linewidth w ₁ and the apparent line width w ₂ is one-dimensional structure array arranged alternately. The line widths w ₁ and w ₂ are determined by the position of the shadow mask during fabrication (w ₁ or u ₁ , w ₂ or u ₂ ), and therefore a value much smaller than the width D of the structural element can be selected. Which is smaller than q ₁ and q ₂ in the boundary region. (Total _{w 1} and _{w 2)} line pitch p can also 10nm or less.
The topography of the structural surface is smooth. The surface of the one-dimensional structure of the conductor and the one-dimensional structure of the insulator are in the same plane (except for the one-dimensional structure produced in the initial stage). A smooth topography can be reached on top of the structure created by the different shadow mask openings if positioning is chosen optimally. This is shown in FIG. 5 and is made using the first slit openings (8a ′, 8b ′, 8c ′, etc.) and the second slit openings (8a ′, 8b ′, 8c ′). , Etc.) are connected to each other.

  Instead of two different gas material sources (1a / 1b), three or more gas material sources (1a / 1b / 1c /...) Can be used, and the above operations can be performed in the same manner. Although details are omitted, this is the second invention method of the present application.

Next, a third invention method, that is, a method for producing a crossbar structure will be described (see FIGS. 9 to 11). The crossbar structure is composed of two mutually intersecting one-dimensional structural arrays and a functional layer between them. Since the production of the one-dimensional structure array is performed in the same manner as described above, a detailed description is omitted.
Set the substrate in the reaction chamber. The shadow mask (5) has three types of openings (possible layout is shown in FIG. 8 (I)), that is, (1) parallel in one direction as shown in FIG. 8 (I) A. A slit, (2) a square opening as shown in FIG. 8 (I) B, and (3) a second direction (preferably, a first direction) as shown in FIG. 8 (I) C. Has parallel slits aligned in the vertical direction.
The source of various materials (assembly) (1) consists of at least two material sources for one-dimensional structural arrangement and at least one material source for functional substances. A substrate which is an electrical insulator is set in the reaction chamber. A similar device (see FIG. 1) is used again. A shadow mask is placed close to the substrate surface (typically about 1 μm).

The crossbar structure is manufactured in the following three steps.
(1) A dense one-dimensional structure array having a smooth surface or a flat surface is produced on the substrate by any one of the above methods (the first or second invention method). At this time, an opening of the first type as shown in FIG. 8A is used. During the production of the one-dimensional structure array, the mask is moved stepwise as described above. Here, the average position of the shadow mask with respect to the sample (substrate) is P1 (see FIG. 9).
(2) Move the shadow mask to the position P2. A functional layer is deposited on the one-dimensional structure array (see FIG. 9).
(3) A second one-dimensional structure array is formed on the functional layer. Let P3 be the average position of the shadow mask with respect to the sample (substrate) (see FIG. 9).
The final structure obtained is shown in FIGS. This is the three parts described above, namely the first one-dimensional structure array (13), the functional layer (14) and the second one-dimensional structure array (15). The one-dimensional structure array is composed of a conductor (17) and an insulator (16). In the case of a crossbar structure, the functional layer is usually sandwiched between conductive wires. This crossbar device consists of two end sub-devices, many horizontal lines x many vertical lines, each connected to one horizontal wire and one vertical wire. One scheme of such a device is shown in FIG.

In order to facilitate electrical connection, the line spacing at the end R can be chosen to be larger than the pitch p (see FIG. 11). This is done by moving extra distances v ₁ and v ₂ along the line direction.
If necessary, it can be made longer than the length L ₁ of the line length L ₂ of the conductor lines of the insulator to avoid shorting.

Crossbar devices often require functional materials having at least two (meta) stable states. Several types of effects provide this property. The production methods described herein are not limited to one specific effect or one specific substance. Here is a list of potential systems (not all):
The functional material consists of a single layer or multiple layers of the following:
・ Organic substances (molecular switches)
・ Photogens ・ Ionic conductors (for atomic switches)
・ Ferroelectric materials ・ Magnetic materials ・ Additional layers of metals, semiconductors, and insulators ・ Combination of such functional materials The structure of the one-dimensional structure arrangement is replaced with other types of metal conductors (electronic conductivity). It can also be composed of a conductor (spin conductor, magnetoresistive conductor, ionic conductor, etc.).

In the above example of making a crossbar structure, all structural elements were made in the reaction chamber (FIG. 1). However, the crossbar structure manufacturing process is not limited to this, and the following process is also possible.
(1) The first one-dimensional structure array is created in the reaction chamber as described above.
(2) The functional layer is applied using different fabrication methods, such as standard lithographic techniques, contact imprinting or self assembly. To do this, the sample (substrate) is moved outside the reaction chamber 12 to suit the selected method. The smooth surface of our one-dimensional structure array is expected to be suitable for adding functional layers with various technologies.
(3) Finally, the sample is again placed in the reaction chamber 12 and a second one-dimensional structural array is created on the functional layer. This shadow mask fabrication method is so gentle that in most cases the functional material is not damaged.

The opening for producing the functional material may be different from the quadrangular shape shown in FIG. Another shape is shown in FIG. 8 (II), which is a circular hole that is two-dimensionally aligned and used for the production of a functional substance. This opening can be moved (sideways) to cover most of this area while making the functional material. Compared to the rectangular opening shown in FIG. 8I, this type of opening has the following advantages: (i) A large opening can be avoided (a large opening makes a shadow mask film weaker). (Ii) If the functional layer is constructed in the lateral direction, the properties of the functional layer will be improved (for example, the leakage current will be reduced).
Three different types of openings for different components of the crossbar structure (consisting of two mutually intersecting one-dimensional structural arrays and functional materials between them) can also be formed on separate shadow masks. In this case, the shadow mask will be simple, allowing more dense patterning and avoiding the creation of unnecessary structural elements.
In order to prevent the production of unnecessary structural elements, a shadow mask to which a larger opening pattern is added can also be used.

We can illustrate another application of a dense one-dimensional structure array (not a crossbar structure) as follows.
• Templates in electrical deposition (applying different voltages to different insulated lines) • Imprinting for nanoprinting (eg using hydrophobic and hydrophilic lines)
・ Grating nanomaterials for light, X-rays, neurons or electrons (for example, when using materials that act as catalysts for nanomaterial growth)
・ Sensors ・ Magnetic devices, Spintronics

Example 1
<Production of one-dimensional structure array>
We created a partially completed one-dimensional structural array (similar to the structure shown in FIGS. 3C and 4) using the apparatus shown in FIG.
Details of the mechanism used and the construction procedure:
The reaction chamber 12 is an ultra vacuum system having a base pressure of 10 ⁻⁷ (Pa) or less.
In this mechanism, the shadow mask was fixed and the sample (substrate) was moved during the structure fabrication.
A homemade inertial sliding member and inchworm (UHVL-025, Burleigh Instruments, USA) were used to coarsely align before starting fabrication. The work of accurately arranging the shadow mask and the substrate in parallel was performed using a homemade inertial sliding member tilting stage. The difference in angle was detected by the difference in the angle of the reflected laser beam.
A stage with a capacitive position sensor (P-733.2UD, PI, Germany) was used for precise alignment of the substrate during structure fabrication.
The material source 1 is a thermal evaporator (electron beam heating) that can handle four different substances (EGC04, Oxford Applied Research, UK; commercially available). The material supply (assembly) source was mounted on a homemade material source alignment device (driven by two stepper motors) for alignment of the active material source in the central axis. The substance supply rate (namely, the film thickness formed per unit time) was measured using a crystal resonator (XTM / 2, Inficon, USA).

The shadow mask 5 is a 200 nm thick silicon nitride film supported by a 100 μm thick silicon frame. A similar film is commercially available, for example, used as a TEM mesh. The film is coated with chromium (2 nm) and gold (30 nm). The opening was cut using a focused ion beam (FIB) (see FIG. 12). The slit width d was about 250 nm. The slit length L was 5 μm, and the slit pitch f was 1 μm.
The properties of the silicon nitride film were examined after the FIB treatment. FIG. 12 shows an image of a portion of the silicon nitride film after the opening has been cut using a focused ion beam (FIB). The image was recorded in the same device imaging mode immediately after FIB processing.
As the substrate 7, silicon nitride (70 nm) on silicon covered with PdAu (20 nm) was used. Samples were coated with conductive PdAu to avoid charging problems during characterization by non-contact atomic force microscope (nc-AFM) and electron microscope (SEM). An insulating substrate will be required as the structure of the functional layer applied to the device.

The moving distance (u ₁ and u ₂ ) of the shadow mask with respect to the substrate was 50 nm.
・ Distance a = 20 cm (see FIG. 1), distance b = about 3 μm
Deposited material 8: Cu and CaF ₂
-Crucible material: Mo for Cu, Ta for CaF ₂
・ Size of raw material opening (crucible opening) s = 4 mm
・ Evaporation rate: About 0.3 nm / min for both substances ・ Power during thermal evaporation of Cu: 34 W (0.8 kV, 42 mA)
Of · CuF ₂ heat蒸時power: 24W (0.8kV, 30mA)
-Pressure during evaporation: ^10-6 Pa

The electric power applied during evaporation causes the temperature of the shadow mask to rise through heat radiation.
The power change causes a thermal drift of the mask position due to a temperature change. In order to minimize the problem of thermal drift, it was necessary to set the heating power to the same level for different materials. (A possible improvement is to further compensate for temperature changes using infrared radiation—this is indicated by 20 in FIG. 1).

<Structural characterization>
The fabricated structure was evaluated using a scanning electron microscope (SEM) (shown in the main part of FIG. 13) and a Kelvin probe microscope (KPM). KPM enables the measurement of surface potential (as shown in the inset (A) in FIG. 13) and the topography (surface feature) simultaneously (inset (B) in FIG. ) As shown). These data provide evidence of the feasibility of the present invention. (I) The bundle of lines has a surface shape (curve B) as theoretically expected (see FIG. 3C). (Ii) The boundary between the insulator line and the conductor line is well defined as shown by the contrast (curve A) in the contact potential scan.

The typical longitudinal section showing the typical reaction room used with the method of the present invention. The figure (top view) which looked at the example of the shadow mask for preparation of the one-dimensional structure arrangement | sequence from the top. It is sectional drawing explaining preparation of a one-dimensional structure arrangement | sequence, (A) is a 1st layer, (B) is the 2nd layer, (C) demonstrates preparation of the 9th layer. The top view of the one-dimensional structure arrangement | sequence after producing the 9th layer shown in FIG.3 (C) (finished partially). (A) is sectional drawing and (B) is a top view of the one-dimensional structure arrangement | sequence which completed preparation. The elements on larger scale of FIG. Schematic plan view of a one-dimensional structure array produced with line slits with errors (errors are exaggerated).

The top view which shows two examples (I and II) of the shadow mask arrangement | positioning for preparation of a crossbar structure. The top view explaining alignment of the mask for producing the element from which a crossbar structure differs. The position of the crossbar structure is indicated by a dotted circle. P0: Stand-by state (warm-up of material source), P1: Mask position for production of bottom one-dimensional structure array, P2: Mask position for production of functional material, P3: Top one-dimensional structure arrangement Mask position for fabrication. The top view of an example of a crossbar structure. The enlarged view of the (A) and (B) part of FIG. An image of a part of a silicon nitride film after piercing with a focused ion beam (FIB). The image was recorded by the imaging mode of the FIB device. The electron microscope image which recorded the structure of the prototype of the one-dimensional structure arrangement | sequence produced by the method of this invention ex-situ. The inset is a Kelvin probe scan recorded in-situ after preparation, with (A) the contact potential and (B) the topography. FIG. 3 is a schematic cross-sectional view of a specific molecular device including a crossbar structure that can be produced using the method of the present invention.

Explanation of symbols

1: evaporation source assembly 1a of various materials: first material source 1b: second material source 4: material beam 5: shadow mask 6: opening 7: substrate 8: fabricated structure 8a ', 8a'' , 8a ″ ′,..., 8a ^(m) ′: Structures 8b ′, 8b ″, 8b ′ ″,..., 8b ^(m) ′: produced in the first deposition step fabricated structures 8c ', 8c'', 8c ''', ..., 8c (m) ': a third of the fabricated structure 8n in the deposition process', 8n'',8n''', ..., 8n ( ^m) ''': Structure produced in the nth deposition step 12: Reaction chamber 13: Bottom one-dimensional structure array 14: Functional material 15: Top one-dimensional structure array 16: Insulator 17: Conductor 20 : Radiant heater

s: Diameter of the material supply source a: Distance from the material supply source to the mask b: Distance from the mask to the substrate q, q ₁ , q ₂ : Boundary portion of the structure d: Width of the opening (slit) D: Structure Width of object (full width half maximum)
f: Distance (pitch) between openings (slits)
L: Length of the opening (slit) p: Pitch of the one-dimensional structure array w ₁ , w ₂ : Effective line width on the surface t ₁ , t ₂ : Layer thickness R: Distance between conductors at the line end
E: error in structural form u, u ₁ , u ₂ : displacement of shadow mask relative to sample (substrate) in the transverse direction perpendicular to the slit length direction v, v ₁ , v ₂ : in the slit length direction Displacement of shadow mask relative to sample (substrate) in parallel lateral direction

Claims (6)

A method for preparing a dense one-dimensional structure array whose surface is maintained in a smooth or flat state on a substrate, comprising the following steps.
(i) Set the substrate in the reaction chamber;
(ii) A shadow mask having a slit (the slit width is d and the slit pitch is f) is set close to the substrate;
(iii) A raw material is supplied from a raw material supply source for a predetermined time, and a first layer is deposited on the substrate so as to correspond to the slit of the shadow mask;
(iv) moving the shadow mask in a direction perpendicular to the length direction of the slit by a predetermined distance (u / u ₁ ) with respect to the substrate while maintaining parallelism with the substrate; Is less than d,
(v) supplying another raw material from the raw material supply source for a predetermined time, and depositing a second layer corresponding to the slit of the shadow mask on the substrate;
(vi) moving the shadow mask in a direction perpendicular to the length direction of the slit by a predetermined distance (u / u ₂ ) while maintaining parallelism with the substrate;
(vii) The above steps (iii) to (vi) are repeated until the total movement distance of the shadow mask or the substrate reaches the slit pitch (f). The preparation method according to claim 1, wherein the one raw material is a conductor raw material, and the other gas raw material is an insulator raw material. The preparation method according to claim 1, wherein a vacuum chamber is used as the reaction chamber. The preparation method according to claim 1, wherein the slit width (d) is 20 nm-1 μm, the slit pitch (f) is 100 nm-10 μm, and the predetermined distance (u) is 5 nm-250 nm. A method for preparing a dense one-dimensional structure array whose surface is maintained in a smooth or flat state on a substrate, comprising the following steps.
(i) Set the substrate in the reaction chamber;
(ii) A shadow mask having a slit (the slit width is d and the slit pitch is f) is set close to the substrate;
(iii) A raw material is supplied from a raw material supply source for a predetermined time, and a first layer is deposited on the substrate so as to correspond to the slit of the shadow mask;
(iv) moving the shadow mask in a direction perpendicular to the length direction of the slit by a predetermined distance (u / u ₁ ) with respect to the substrate while maintaining parallelism with the substrate; Is less than d,
(v) supplying another raw material from the raw material supply component for a predetermined time and depositing a second layer corresponding to the slit of the shadow mask on the substrate;
(vi) moving the shadow mask in a direction perpendicular to the length direction of the slit by a predetermined distance (u / u ₂ ) with respect to the substrate while maintaining parallelism with the substrate;
(vii) An n-th (n) -th source material is supplied from a source material supply source for a predetermined time, and an n-th layer corresponding to the shadow mask slit is deposited on the substrate;
(viii) the shadow mask, while maintaining the parallelism between the substrate, a predetermined distance (u / u _n) with respect to the substrate, moves in the direction perpendicular to the length direction of the slit;
(ix) Repeat steps (iii) to (viii) until the total movement distance of the shadow mask or substrate reaches the slit pitch (f). A method for preparing a crossbar structure, comprising the following steps.
(1) According to the preparation method of claim 1 or 5, a dense one-dimensional structure array whose surface is kept smooth or flat is prepared on a substrate;
(2) depositing a functional layer on the one-dimensional structure array using a shadow mask having predetermined holes;
(3) The functional layer is formed by the same method as in claim 1 or 5 except that a shadow mask having a slit whose length direction is perpendicular to the slit used in the step (1) is used. On top of it, a dense one-dimensional structural array is prepared with the surface kept smooth or planar.