CA1328131C

CA1328131C - Encoder

Info

Publication number: CA1328131C
Application number: CA000575623A
Authority: CA
Inventors: Yuko Morikawa; Yoshihiro Yanagisawa; Akihiko Yamano; Hiroyasu Nose; Hiroshi Matsuda; Hisaaki Kawade; Kunihiro Sakai; Haruki Kawada; Kenji Saitoh; Minoru Yoshii; Eigo Kawakami; Toshimitsu Kawase; Ken Eguchi
Original assignee: Canon Inc
Current assignee: Canon Inc
Priority date: 1987-08-25
Filing date: 1988-08-24
Publication date: 1994-03-29
Anticipated expiration: 2011-03-29

Abstract

ABSTRACT OF THE DISCLOSURE
An encoder includes an electrically conductive reference scale having surface steps formed at predetermined positions; an electrically conductive probe having a tip disposed opposed to the reference scale; wherein the reference scale and the probe are relatively movable in a direction different from the opposing direction of the tip of the probe and the reference scale; a portion for applying an electric voltage to between the reference scale and the probe, a portion for detecting a change in a tunnel current between the reference scale and the probe, to between which the electric voltage is applied by the voltage applying portion at the time of the relative movement between the scale and the probe, the detecting portion detecting the change in the tunnel current when the probe passes a position opposed to a surface step of the reference scale; and portion for detecting the amount of the relative movement between the scale and probe, on the basis of the detection by the change detecting portion.

Description

~32813~

This invention relates to an encoder for use in the detection of positional information during fine positioning, dimension measurement, distance measurement, velocity measurement or otherwise. More particularly, it is concerned with an encoder for use in measurements in which resolution of an order of atoms (a few angstroms) is required.
Usually, encoders of the type described comprises a reference scale, which bears information related to position or angle, and detecting means movable relatively to the reference scale, for detecting the information related to the position or angle. In respect to the reference scale and the detecting means used, encoders may be classified into different types, e.g., optical encQders, magnetic encoders, electrostatic capacitance type encoders, etc.
Of these encoders, those exhibiting the highest resolution are optical encoders which use, for example, the principle of grating interference.
In grating interference type optical encoders, the resolution performance is determined chiefly by the pitch of the grating. Thus, precise fabrication of the grating with a minute pitch and precise detection of fringes produced by the grating are important factors. Current fine machining techniques (for example, the electron beam pattern drawing technique or the ion beam technique) can at the best assure a precision of 0.01 micron (100 '~

~ t~

132~

angstroms). Current detection techniques (for example, the optical heterodyne method) also have a resolution limited to 0.01 micron. It is thus difficult to satisfy the requirement for a higher resolution encoder, which would be particularly useful in semiconductor microcircuit manufacturing apparatus.
It is accordingly a primary object of the present invention to provide an encoder which can provide resolution of an order comparable to interatomic distances.
According to the invention, there is provided an encoder, comprising: an electrically conductive reference scale having surface steps formed at predetermined positions; an electrically conductive probe having a tip disposed opposed to said reference scale;
wherein said reference scale and said probe are relatively movable in a direction different from the opposing direction of said tip of said probe and said reference scale; means for applying an electric voltage to between said reference scale and said probe; means for detecting a change in a tunnel current between said reference scale and said probe, to b~tween which the electric voltage is applied by said voltage applying means at the time of the relative movement between said scale and said probe, said detecting means detecting the change in the tunnel current when the probe passes a position opposed to a surface step of said reference .,.~,, '~ , .

~l328~31 scale; and means for detecting the amount of the relative movement between said scale and probe, on the basis of the detection by said change detecting means.
These and other objects, features and advantages of the present invention will ~ecome more apparent upon a consideration of the following description of the preferred embodiments of the present invention taken in conjunction with the accompanying drawings, in which:
Figure 1 is a schematic view showing the structure of a known type optical encoder which uses the principle of grating interference.
Figure 2 is a schematic and diagrammatic view showing the structure of an encoder according to one embodiment of the present invention, wherein detection of tunnel currPnt is used.
Figure 3 is a waveform view showing slgnals obtained at different constituent portions of the encoder shown in Figure 2.
Figure 4 is a schematic view showing an example wherein a periodic atomic or molecular arrangement on the surface of an electrically conductive material is used as a reference scale.
Figure 5A is a schematic section showing an example wherein graduations fabricated on the surface :

1328~31 of an electrically conductive material by means of converged ion beam treatment or electron beam treatment, are used as a reference scale.
Figure 5B is a schematic sectlon showing an example wherein the surface of a material having a shape like a diffraction grating and coated with an electrically conductive material is used as a reference scale.
Figure 6 is a schematic section showing an example of a compound scale wherein one reference origin is provided on a reference scale.
Figures 7A - 7C are schematic sections, respectively, showing examples of compound scale, in each of which plural reference origins are provided on a reference scale.
Figure 8 is a schematic section showing an example wherein use is made of a compound scale in which plural reference origins having predetected widths and intervals are provided on a reference scale.
Figures 9A and 9B are a schematic section and a schematic perspactive view, respectively, showing examples in each of which use is made of a compound scale wherein absolute positional information is recorded on or adjacent to a reference origin provided on a reference scale.
Figure 10 is a schematic section showing an example wherein non-periodic graduations are used as a .
' ~ 32813~

reference scale.
Figure 11 is a schematic and diagrammatic view showing the structure of an encoder wherein an asymmetric-shape reference scale is used as a reference scale.
Figure 12 is a waveform view showing signals which are obtainable in the embodiment of Figure 11.
Figures 13A and 13B are schematic views showing a plane (1,1,1) of a face~centered cubic lattice, which is a specific example of an asymmetric reference scale, wherein Figure 13A is a top plan view and Figure 13B is a sectional view.
Figure 14 is a schematic section showing an embodiment wherein a probe is oscillated laterally for the detection of direction.
Figure 15 is a waveform view showing signals which are obtainable in the embodiment of Figure 14.
Figure 16 is a schematic section showing an embodiment wherein a probe is oscillated laterally for the detection of direction~
Figure 17 is a waveform view showing signals which are obtainable in the embodiment of Figure 160 Figure 18 is a schematic and diagrammatic view showing the structure of an embodiment wherein plural probes are used to allow detection of direction.
Figure 19 is a waveform view showing signals which are obtainable in the embodiment of Figure 18.

:

.

-6- ~2~131 Figure 20 is a schematic and diagrammatic view showing the structure of an encoder in which the vertical position of a probe is detected.
Figure 21 is a black diagram showing the sequence of operations for the detection of lateral relative displacement based on the detection of probe vertical position, under a condition of constant tunnel current.
Figure 22 is a waveform view showing signals which are obtainable at different constituent portions of the embodiment of Figure 20.
Figure 23 is a schematic section showing an example of the Figure 20 embodiment but, in this example, an asymmetric-shape scale is used as a reference scale.
Figure 24 is a waveform view showing signals which are obtainable in the Figure 23 example.
Figure 25 is a schematic section showing an example of the Figure 20 embodiment butl in this example, a probe is laterally oscillated for the detection of direction.
Figure 26 is a waveform view showing signals which are obtainable in the Figure 25 example.
- Figure 27 is a schematic section showing an example of the Figure 20 embodiment, wherein a probe is laterally oscillated for the detection of direction.
Figure 28 is a waveform view showing signals . .. . .

;: 7 :~8~

which are obtainable in the Figure 27 example.
Figure 29 is a schematic and diagrammatic view showing the structure of an embodiment wherein plural probes are used to allow detection of direction.
Figure 30 is a waveform view showing signals which are obtainable in the Figure 29 embodiment.
Figure 31 is a schematic and diagrammatic view of an encoder according to one embodiment of the present invention, wherein a tunnel current is -detected.
Figures 32 and 33 are waveform views, respectively, showing signals which are obtainable at various constituent portions of the Figure 31 embodiment.
Figure 34 is a schematic and diagrammatic view of an encoder according to another embodiment of the present invention.
Figures 35 and 36 are waveform views, respectively, showing signals which are obtainable at 20 various constituent portions of the Figure 34 embodiment.
Figure 37 is a perspective view schematically showing an encoder according to a fuxther embodiment of the present invention.
Figure 38 is a diagrammatic view showing control maans of the encoder of the Figure 37 embodiment.

~'' . .

, ,:

~ -8- 1~2~3~

Figure 39 is a schematic view explicating the principle of a scanning tunneling microscope, which principle is used in the present invention.
Figure 40 is a timing chart showing the interrelationship among the signals obtainable in the device of Figure 37.
Figure 41 is a diagrammatic view of a pulse counter used in the device of Figure 37.
~, Figure 42 is a schematic section showing a 10 code plate used in the device of Figure 37.
Figure 43 is a side view showing a major portion of an encoder according to a still further embodiment of the present invention.
Figure 44 is a perspective view showing a 15 major portion of an encoder according to a yet another embodiment of the present invention.
Figures 45A and 45B are a sectional view and a top plan view, respectively, showing an encoder according to a still further embodiment of the present 20 invention~
Figure 46 is a diagrammatic view showing control means used in the encoder of the embodiment shown in Figures 45A and 45B.
Figure 47 is a timiny chart showing the 25 interrelationship among the signals obtainable in the ; device of the embodiment shown in Figures 45A and 45B~
Figure 48 is a flow chart showing the . .

, .
-: , 9 132~13~

sequence of measurement made in the device of the embodiment shown in Figures 45A and 45B.
" Figure 49 is a schematic section showing an encoder according to another embodiment of the present invention.
Figure 50 is a principle view showing the interrelationship between a coordinate axis and ; recording positions, in one embodiment of the present invention.
Figure 51 is a block diagram schematically showing a recording and reproducing (playback) apparatus, arrangad in accordance with the embodiment of Figure 50.
Figure 52A is a plan view showing an example 15 of a recording medium used in the Figure 50 embodiment.
Figure 52B is a section taken on a line A-A' in Figure 52~.
Figure 53 is a schematic view showing an example ~f the positional relationship between a 20 coordinate axis and recording positions, on the surface of a recording medium used in the embodiment of Figure 50.
Figure 54 is a block diagram schematically showing a recording and reproducing apparatus according 25 to another embodiment of the present invention.
Figure 55A is a plan view of a recording medium used in the Figure 54 apparatus.

:. :, - . : i .

~ ~ :

-10- 1~28~3~

Figure 55B is a section taken on a line A-A' in Figure 55B.
Figure 56 is a schematic view showing recording positions on the recording medium of the Figure 55 example.
Figure 57A is a plan view of another recording medium which is usable in the Figure 54 apparatus.
Figure 57B is a section taken on a line A-A' 10 in Figure 57A.
Figure 58 is a schematic view showing the positional relationship among recording areas on the surface of the recording medium of the Figure 57 example.
; 15 Figure 59 is a schematic and diagrammatic view of a recording and reproducing apparatus according to a further embodiment of the present invention.
Figure 60 is a block diagram of a control system used in the Figure 59 embodiment.
Figure 61 is a diagram showing the algorism in the central processing unit included in the Figure 59 embodiment.
Figure 62 is a principle view schematically showing a recording and reproducing apparatus according 25 to a still further embodiment of the present invention.
Figure 63 is a schematic and diagrammatic view of a recording and reproducing apparatus according :; , . :, : ::

' -13~-8~3:~

to a further another embodiment of the present invention.
Figure 64 is a schematic view showing positional reference patterns and recording/reproducing areas, defined on a recording medium used in the Figure 63 embodiment.
Figure 65 is a schematic view showing recorded signals, in the form of two-dimensional images, in the embodiment of Figure 63~
Figure 66 is a schemati~ view showing the manner of scan of recorded signals by a readout system, in accordance with the embodiment of Figure 63.
Figure 1 shows the structure of a known type optical encoder. In this example, light source 3001 emits monochromatic light 3002 which is projected on a diffraction grating 3003 which functions as a reference scale. Of the light diffracted by the grating, positive and negative first-order diffraction components 3004 and 3005 are directed by reflection mirrors 3006 into a part silvered mirror 3007, which combines the diffraction light components and thus engenders interference.
resulting interference fringe is detected and photoelectrically converted by photodetector 3008, the relative displacement between the optical system and the reference scale being detected on the basis of the interference fringe position.
Briefly, in accordance with one aspect of the present invention, an electric voltage is applied between - ~:
'`' ' ~ ~
:

.

1328~31 -llA ~
an electrically conductive reference scale, which functions as a reference with respect to length, and an electrically conductive probe having a tip disposed close to the surface of the reference scale. The electric current (tunnel current) which flows between the reference scale and the probe is detected to obtain a tunnel current signal which, in turn, is subjected to waveform processing. From the processed signal, the amount and direction of the relative movement between the reference scale and the probe in a lateral direction, namely, a........

,/.'~

, ,:
. . . .
~ ' ~

~, . ... .
. . . ..

-12- ~32813~

direction along the scale surface, is detected to obtain electric signals representing the lateral relative moving amount and the relative moving direction. Then, from these signals, the lateral 6 relative displacement (quantity of deviation) between the reference scale and the probe is measured and, finally, the quantity of lateral relative movement between the probe and the reference scale is dètected.
The invention will be described below in 10 graater detail.
The present invention utilizes the phenomenon that their occurs a flow of tunnel current when an electric voltage is applied to between an electrically conductive material and an electrically conductive probe and when they come close to each other to a clearance of about 1 nm ("Solid State Physics" Vol, 22, No. 3, 1987, pp. 176-186). The tunnel current depends on the distance between the electrically conductive material and the electrically conductive probe as well 20 as a work function upon the surface. Therefore, it is possible to read out the information related to various ; surface shape state.
Where an electric voltage V which is lower than the work function ~ (~ > eV wherein e is the 25 electric charge of electron) is applied to between an electrically conductive probe and an electrically : conductive reference atomic arrangement which are .
.... . .

.

~32~3~

spaced by a distance Z, electrons tunnel through a potential barrier between the electrically conductive probe and the electrically conductive reference electron arrangement. Tunnel current density JT may be determined by free electron approximation and can be expressed as follows:
JT = (~V/2~AZ)exp~-2Z/A) ... (1) on condition that:
~ ~ ~/ ~ and ~ = e2/~
wherein ~ is the decay distance of the wave function in a vacuum or an atmosphere outside the metal (~ is in the range of 1 - 5 eV and A is in the range of 1 - 2 angstroms);
is equal to h/2 ~ (h is a Planck's constant); and m is mass of electron.
In equation 11), the tunnel current density JT changes with the distance Z. Namely, if there 20 occurs relative positional deviation between the probe and the reference atomic arrangement on condition that the average interval between the probe and the reference atomic arrangement is maintained constant, the tunnel current changes periodically in accordance 25 with the reference atomic arrangement. Thus, from that change in the tunnel current, the quantity of positional deviation can be determined.

. : ~
~; ~

.: . , : . .

Also, in equation (1), if control is made with regard to the position of the probe in the direction of a normal to the reference atomic arrangement so as to maintain the tunnel current JT
constant, namely, to maintain the distance Z constant, then the probe position controlling signal (controlling the probe position in the direction of the normal) changes periodically in accordance with the reference atomic arrangement. Thus, from that change in the 10 position controlling signal, the quantity of positional deviation can be determined.
The resolution ~ in the scale surface can be higher with a finer tip end of the probe and, according to model calculation, it can be expressed as follows:
~ = [2~(R~Z)]1/2 wherein, R is the radius of curvature of the metal probe tip.
Where th~ probe tip is provided by one atom, R = 1 angstrom, and thus, the resolution ~ in the plane 20 is of an order of 3 angstroms. The above-described method, using the tunnel current, is operable not only in the atmosphere but also in a liguid, and provides various advantageous features such as, for example, high resolution.
The present invention utilizes the above-described principle and provides in one aspect thereof an encoder wherein a periodic surface structure '' ~ '' . :

' ' '' ., : ~

provided by a regular atomic arrangement on the surface of a material is used as a reference and wherein use is made of a tunnel current which flows between the reference surface and an electrically conductive probe.
Thus, the present invention can assure a highest resolution of an atom order (a few angstroms) which is far higher than the maximum resolution of the conventional encoders.
~eferring to the drawings, embodiments of the 10 present invention will be described.
Figure 2 shows the structure of an encoder according to one embodiment of the present invention.
Figure 3 shows signals which are obtainable at various constituent portions of the present embodiment.
In Figure 2, an object 101 and another object 102 are made relatively movable only in a lateral direction (horizontal direction in the sheet of the drawing). The object 101 is provided with an electrically conductive reference scale 103, and the 20 object 102 is provided with an electrically conductive probe 104. Between the probe 104 and the reference scale 103, a bias voltage is applied by means of a bias voltage source 106. The tip of the probe 104 and the reference scale 103 are disposed so close to each other 25 that a tunnel current schematically illustrated at 105 flows therebetween. The tunnel current 105 may be detected by a current detecting means 107 and may be ' - `''.;' '"' -` ','` ' , " ' ~ ':
;

-16- ~328~

amplified by a current amplifying means 108.
~ fter the amplification, the tunnel current signal 118 (its waveform being illustrated in Figure 3, (a)) is applied to a waveform processing means 110, wherein it is subjected to binary-coding by the slicing with a certaln threshold level, waveform processing by detecting edges of the resultant binary-coded signal for pulse formation, as well as the direction detection which will be described later with reference to a 10 specific means that is provided for the detection of the direction of the lateral relative displacement.
Thereafter, the signal is transformed into a moving amount signal 112 (Figure 3, (b)) and a moving direction signal 113 (Figure 3, (c)), this being made by a converting means 111. These signals are applied to a counter 114. In this counter 114, the moving amount signal 114 and the direction signal 113 applied thereto are transformed into a lateral relative displacement. Further, by using an initial position 20 data having been preparatorily stored in a memory 115, subtracting means 116 operates to calculate the lateral relative displacement, at signal 117 (Figure 3, (d)), from the initial position, between the objects 101 and 102.- The resolution of the encoder of the present 25 embodiment can be of an order of the pitch of the reference scale 103 (a few angstroms where a periodic arrangement of atoms on the surface of the electrically . ~ , .

-17- ~328131 conductive material is used as the reference scale).
However, a further improved resolution less than 1 angstrom is attainable where a detected position signal is divided (Figure 3, (e)~.
Next, description will be made of materials usable for the probe in the present invention. Those materials having electric conductivity may be used as the probe material. Examples are a metal such as Au, Pt, Ag, Pd, Al, In, Sn/ Pb, W or otherwise; and an 10 alloy of appropriate ones of these materials. Further, graphite, silicide or an electrically conductive oxide material such as, for example, ITO (In-Sn oxide), may be used.
In order to assure high resolution of the encoder, it is necessary that the tip of the probe is pointed as sharp as possible to an order of atom.
The present embodiment uses a probe 104 which has been formed by mechanically grinding a tungsten wire of a diameter ~ = 1 mm to a diameter of about 0.2 20 mm and thereafter by sharpening the tip by means of electropolishing. However, the manner of treatment is not limited thereto.
Next, the reference scale will be explained.
Figure 4 is a schematic view showing a probe 104 and an 25 electrically conductive reference scale 103 used in the present embodiment. Shown in Figure 4 is an example wherein a periodic atomic or molecular arrangement 301 ,,, ,, . , , ~ .
.. ,... " . - :, 1 3 2 ~

on the surface of an electrically conductive material (whose lattice spacing is predetected) is used as the reference scale. By using the periodic arrangement of a minute structure of atoms or molecules as a reference scalq 103, as described, it i.s possible to provide an encoder having a resolution of an order of interatomic distance, namely, of an order of angstrom. Examples of such an electrically conductive material usablè in the present invention are a metal such as Au, Pt, TaS2 or 1 n otherwise; an alloy such as CuAu, PtIr or otherwise; a submetal such as graphite; a semiconductor such as Si, MoS2, GaAs, SiC or otherwise; an electrically conductive LB film (Langmuir Blodgett film); and an electrically conductive organic material such as TTF-TCNQ (tetrathiafluvalene-tetracyanoquinodimethane).
In place of the atomic or molecular arrangement of the above-described materials, a scale such as shown in Figure 5A in which graduations 402 are formed on the surface 401 of an electrically conductive material by 20 means of a focused ion beam process or an electron beam process; or alternatively a scale such as shown in Figure SB in which the surface 403 of an article having a shape like a diffraction grating i.s coated with a film 404 of an electrically conductive material, may be 25 used as a reference scale.
Next, description will be made of examples wherein a periodic arrangement of atoms, molecules or ,.

-19- i328131 otherwise is used in combination with one or more fabricated graduations to provide a reerence scale (hereinafter such scale will be referred to as a "compound scale").
Figure 6 is a sectional view showing an embodiment of a compound scale. As shown in this Figure, a reference origin 501 is formed on a reference scale 103. The moment at which a probe 104 comes to the position of this origin 501 may be treated as 10 the initial setting point. By doing so, it is always possible to determine the amount of movement of the probe from this reference origin 501. Therefore, it is possible to detect the absolute moving amount of the object 102 with respect to the object 101. Except for thisl the present embodiment can be handled essentially in the same way as having been described with reference to the detection of the relative moving amount.
Figures 7A - 7C are sectional views, respectively, showing further embodiments of the 20 compound scale. Of these drawings, Figure 7A shows an example wherein portions of the surface of an atomic arrangement of a reference scale 103 are substituted by a different material to provide reference origins 601a1 - 601a3. Figure 7B shows an example wherein a 25 different material is adhered to portions of the surface of an atomic arrangement of a reference scale 103, to provide reference origins 601b1 - 601b3.

:.

-20- ~2~3~

Figure 7C shows an example wherein portions in the neighborhood of the surface of an atomic arrangement of a reference scale 103 are removed by etching to provide reference origins 601C1 - 601c3.
In these examples, as illustrated, a plurality of reference origins (three in the illustrated examples) are provided. Referring particularly to Figure 7A for explanation of the operation, each time a probe 104 passes a reference 10 origin 601ai (i = 1 - m~, a rough-pitch signal is outputted. During a time period during which the probe goes from the reference origin 601ai to the next reference origin 601ai+1, fine-pitch signals are outputted. Essentially the same operation is made in each of the compound scales of the Figure 7A and Figure 7B examples. Where the amount of relative movement between the objects 101 and 102 is large, the detecting process and the detecting time can be simplified and reduced by selectively extracting such rough-pitch 20 signals.
Figure 8 is a s~ction showing another embodiment of a compound scale. Reference origins 701d1 - 701d4 as illustrated in this Figure are different in respect to the width of origin and the 25 interval between the originsO However, such widths and intervals are predetected by preparatory measurement, the predetected values being memorized in a memory 115 ' ' .... 1j~ `
.
, ` 1~2813~

(Figure 2). In the Figure 8 example, the widths of the reference origins 701d1 - 701d3 are denoted, reSpectivelyl by Lm_1, Lm and Lm+1 while the intervals between the origins 701d1 and 701d2 and between the origins 701d2 and 702d3 are denoted, respectively, by Wm_1 and Wm. These values are preparatorily stored into the memory 115 (Figure 2). Each time a probe 104 passes a reference origin, a corresponding width and/or intexval, being memorized, are counted. In other 10 words, each time a reference origin is passed, a rough-pitch signal (representing the interval Wm_1 or Wm between corresponding reference origins) is outputted while, on the other hand, a signal representing the width Lm_1, Lm or Lm+1 of that reference origin is 15 outputted. By doing so, the detecting process and the detecting time can be simplified and reduced, similarly to the embodiments shown in Figures 7A - 7C.
Figures 9A and 9B are sections, respectively, schematically showing yet further embodiments of 20 compound scales. In the Figure 9A example, there is a reference origin 801. Absolute position information of bits n (i1 ~ in)~ concerning this reference origin, has been recorded on the reference origin 801. When a probe 104 passes this reference origin 801, the 25 absolute position information related to this reference origin is picked up, in a manner similar to the case of the fine-pitch signal. Therefore, it is possible to .
, ~ , . . . . . ............................... ..

.

-22- ~2813~

discriminate such reference origin as just having been passed by the probe. Accordingly, it is possible to obtain positional information in a reduced time.
Similarly, in the Figure 9B example, reference origins 802, 803 and 804 have recorded thereon respective absolute position information.
Grooved structures (surface steps) of shapes such as depicted at 802', 803' and 804' are formed as the reference origins 802, 803 and 804, respectively. Each 10 grooved structure itself bears the absolute position information. Thus, each reference origin can be identified. Namely, on these grooved structures 802', 803' and 804', there are recorded the addresses i~
` and i+1 of respective re~erence origins. The position 15 of each referenca origin may be predetermined. For example, the position of the leftmost side surface of each protrusion (grooved structure) may be treated as the position of a corresponding origin. Denoted at 805 and 806 are atom graduations~
In this embodiment, the bits of the absolute position information are arrayed in the advancing direction of the probe 104. However, they may be arrayed in the neighborhood of the reference origin.
As a further alternative, the objects 101 and 102 may ~5 be made relatively movable also in a direction perpendicular to the above-described relatively moving direction (horizontal in the sheet of the drawing) and -23- 1~28131 these bits may be arrayed ln that perpendicular direction. In this case, the objects 101 and 102 are relatively moved in that perpendicular direction to allow read-out of the absolute position information.
Figure 10 is a sectional view showing an embodiment wherein non-periodic graduations are used as a reference scale. In this example, the intervals of the non-periodic graduations 901 are preparatorily measured by use of a known-type scale, the obtained 10 interval information being memorized in a memory 115 (Figure 2). The amount of relative movement between two objects 101 and 102 is detected by use of (i) those signals which are obtalnable as a result of the scan of the non~periodic graduations by a probe 104 caused by 15 the relative movement between the objects 101 and 102 and (ii) the interval information having been stored in - the memory 115.
Referring now to Figures 11 - 19, description will be made of means for detecting the direction of 20 the lateral relative displacement between objects 101 and 102O
; Figure 11 shows an embodiment which uses a reference scale 1001 having an asymmetric shape (of an electron cloud distribution) in the direction of the 25 relative displacement.
It is now assumed that, with the relative displacement between the objects 101 and 102, a tunnel 1~2~131 current signal 118 changes as shown in Figure 12, (a).
In such case, within a lateral relative position information extracting means 109, there are produced a moving amount signal (edge detection pulse signal) 112 and a direction signal ~differentiated signal) 113 whose waveforms are such as shown in Figure 12, (b) and (c). By adding or subtracting the pulse number of the moving amount signal 112, in accordance with the sign (positiveness or negativeness) of the direction signal 10 113, there can be detected a lateral relative displacement such as shown in Figure 12, ~d) in terms of pulse count. Accordingly, in this example, it is detected that the object 102 has been displaced with respect to the object 101, rightwardly in the sheet of the drawing and by an amount as depicted in Figure 12, (d).
Similarly, where the tunnel current changes with the relative displacement, in a manner such as shown in Figure 12, ~e), there are produced after the 20 subsequent signal waveform processing a moving amount signal 112 and a direction signal 113 which are such as shown in Figure 12, (f~ and (g)~ From this, it is discriminated that the object 101 has been displaced with reference to the object 101, leftwardly in the 25 sheet of the drawing and by an amount as depicted in Figure 12, (h).
Also, where the tunnel current shows changes ~328~31 such as shown in Figure 12, (i), it is discriminated that the direction of the relative displacement of the object 102 with respect to the object 101 has been changed at polnt T, from right to left in the sheet of the drawing.
As for examples of reference scale having such an asymmetric shape (of an electron cloud distribution), there are a plane (111) of a face-centered cubic lattice as shown in Figures 13A and 13B
(wherein Figure 13A is a top plan view while Figure 13B
is a sectional view), an electrically conductive LB
film in which asymmetry occurs in the pull-up direction, and so on.
Figure 14 shows another embodiment wherein 15 the direction of lateral relative displacement between objects 101 and 102 is detected. In the Figure 14 example, laterally oscillating means 1301 is used to minutely oscillate a probe 104 in the direction of the relative displacement. By oscillating the probe 104 20 sufficiently quickly as compared with the speed of the relative movement between the objects 101 and 102, there can be obtained a position differentiation coefficient for the tunnel current signal, with respect to the lateral relative displacement and, from which, 25 the direction of lateral relative displacement is detectable.
This will be explained. Where the probe is . . ~ ;
. .. .
~ ~ '' ' , . .

oscillated in the direction of the relative displacement, the produced signal has such shape that a minute oscillation signal ~requency resulting from the oscillation of the probe is superposed upon a tunnel current signal which corresponds to the position on the scale, the minute oscillation signal frequency having a smaller amplitude and a higher frequency as compared with those of the tunnel current signal. The resultant signal is divided and the split portions are applied to 10 a low-pass filter and a high-pass filter, respectively.
The output of the low-pass filter is a substantially pure tunnel current signal (Figure 15, (a), (e), (i) and (m)) which corresponds to the scale position and in which the minute oscillation signal component is 15 excluded. Therefore, it is used for the measurement o~
the relative displacement. On the other hand, the output of the high-pass filter corresponds only to the minute oscillation signal (caused by the probe oscillation) in which the tunnel current signal 20 component is excluded. The minute oscillation signal is such that, where the motion of the probe in one direction is denoted by "positive" while the motion of the probe in the opposite direction is denoted by "negative" and when the probe moves in the positive 25 direction relatively to the scale: in an occasion where the tunnel current signal increases, there occurs a minute oscillation signal of the same phase as the .
- :.
:
:

-27- ~328131 oscillation of the probe and of such amplitude that corresponds to the ~uantity of the increase; whereas in an occasion where the tunnel signal decreases, there occurs a minute oscillation signal of the opposite phase as the oscillation of the probe and of such amplitude that corresponds to the quantity of the decrease. Therefore, by applying the thus obtained signals to a lock-in amplifier while using the probe oscillation signal as a reference signal, and on the 10 basis of the amplitude and phase obtained thereby, it is possible to obtain such a signal that corresponds to the position differentiation coefficient for the tunnel current with respect to the positive-direction relative displacement, namely such a signal as shown in Figure 15 15, (c). Where the probe moves in the negative direction relatively to the scale, there is produced a minute oscillation signal having the opposite phase as the probe oscillation ~where the tunnel current signal increases) or a minute oscillation signal having the 20 same phase as the probe oscillation (where the tunnel current signal decreases). Similarly, these signals are applied to the lock-in amplifier to obtain a signal which is one that corresponds to the position differentiation coefficient for the tunnel current with 25 respect to the negative-direction relative displacement, namely, such a signal as shown in Figure 15, (g). As described, the position differentiation ' ............... ' ., - . . ~
, -28- ~32~3~

coefficient signals related to the relative displacement and obtainable with respect to the same tunnel current signal may have surely the opposite phases in accordance with the difference in the advancing direction. Therefore, by detecting the phase, it is possible to detect the direction of advancement.
For example, where the tunnel current signal and the position differentiation coefficient therefor, 10 caused with the lateral relative displacement between the objects 101 and 102, show changes such as shown in Figure 15~ (a) and ~c), wherein the wave (c) goes ahead of the wave (a) by a quarter wave, it is discriminated that in the Figure 14 example the object 102 has been displaced with respect to the object 101, rightwardly in the sheet of the drawing and by an amount as depicted in Figure 15, (d).
Similarly, when the tunnel current changes with the relative displacement in the manner as shown 20 in Figure 15, (e), wherein the wave (g) has a delay by a quarter wave as compared with the wave ~e), after the execution of the signal waveform processing shown in Figure 15, (f) and (g~, it is discriminated that the ; object 101 has ~een displaced with respect to the 25 object 102, leftwardly in the sheet of the drawing and by an amount as depicted in Figure 15, (h).
Further, where the tunnel current shows changes such . : , ~ . ' ' - ' -29- 13~13~

as, for example, shown in Figure 15, (i) and ~m), it is discriminated that the direction of the relative displacement of the obje~t 102 with respect to the ob;ect 101 has changed at point T, from rightwardly to leftwardly.
Figure 16 shows a yet another embodiment wherein the direction of lateral relative displacement between objects 101 and 102 is detected. In the Figure 16 example, laterally oscillating means 1501 is 10 effective to oscillate a probe 104 in the direction of the relative displacement of the objects. At different two points A and B, tunnel current signals A and B are detectad. In the present embodiment, it is necessary that the probe 104 is oscillated sufficiently quickly 15 as compared with the speed of the relative movement between the objects 101 and 102. Where the range of oscillation of the probe 104 is denoted by d, the method of the present embodîment is equivalent to such method in which two probes spaced by an interval d are 20 used to detect the direction from two tunnel current signals having a phase difference with respect to the same reference scale 103.
Here, as an example, the pitch of the reference scale 103 is denoted by p and the range d of 25 the oscillation of the probe 104 is determined so as to satisfy the following relation:
d = p/4 ~ N-(p/2) _30_ 132~13~

wherein N = 0, 1, 2 ~..
In this case, and where the tunnel current changes with the lateral relative displacement between the ob;ects 101 and 102, in such manner as shown in Figure 17, (a), it is discriminated that in the Figure 16 embodiment the object 102 has been displaced with respect to the object 101, rightwardly in the sheet of the drawing. On the other hand, the tunnel current changes such as shown in Figure 17, (e), it is 10 discriminated that the object 102 has been relatively displaced leftwardly. Further, the change in the tunnel current such as shown in Figure 17, (i) or (m) represents that, at point T, the direction of relative displacement of the object 102 with respect to the 15 object 101 has changed from rightwardly to leftwardly.
Figure 18 shows a yet another embodiment wherein the direction of lateral relative displacement between objects 101 and 102 is detected. The Figure 18 embodiment is an example wherein two probes are 20 used~ Probes 1701A and 1701B are spaced by an interval d. By detecting tunnel current signals 1705A
and 1705B at the two probes, respectively, it is possible to detect the direction of the lateral relative displacement. For example, where the pitch 25 of a reference scale 103 is denoted by p, the interval d between the probes 1701A and 1701B may be so determined as to satisfy the following equation:

:

_31- 1328131 d = p/4 ~ N~(p/2) wherein N = 0, 1, 2 ...
~ n this case and where the tunnel currents 1705A and 1705B at the two probes 1701A and 1701B
changes with the lateral relative displacement between the objects 101 and 102, in such manner as shown in Figure 19, (a~, it is discriminated that in the Figure 18 embodiment the object 102 has been displaced with respect to the object 101, rightwardly in the sheet of 10 the drawing. On the other hand, where the tunnel current changes such as shown in Figure 19, ~e), it is discriminated that the object has been displaced rightwardly.
Further, where the tunnel current changes 15 such as shown in Figure 19, (i) or (m), it is discriminated that at point T, the direction of the relative displacement of the object 101 with respect to the object 102 has changed, from rightwardly to leftwardly.
Next, description will be made of . embodiments in which the vertical position of a probe : is controlled by use of a probe vertical position controlling means so as to maintain the tunnel current at a predetermined level and in which, while doing so, 25 the amount of lateral relative movement is calculated.
Figure 20 shows the structure of an encoder according to a further embodiment of the present ', , .

-32- 1328~3~

invention. Figure 21 is a block diagram for explicating the operation of the present embodiment.
In Figure 20, for the detection of the lateral relative displacement between objects 101 and 102, first a set tunnel current level Io is determined ~step B1 in Figure 21)u The vertical position of a probe 104 is controlled by use of a probe vertical position controlling means 1901. For th~s control, first a tunnel current level I is detected (step B2) 10 and then the detected current I is compared with the set tunnel current level Io (step B4). If I ~ Iot feedback control is made from a feedback control means 1902 to the probe vertical position controlling means 1901 (step B3). When the detected current I becomes 15 equal to the set current Io (step B4), an absolute vertical position signal representing the absolute vertical position of the probe 104 is outputted to a lateral relative position information extracting means 109 (step BS). After this and during the subsequent 20 lateral relative displacement between the objects 101 and 102, the probe vertical position controlling means 1901 and the feedback control means 1902 are used to execute the feedback contrcl of the vertical po~ition of the probe 104 so that the detected current I is 25 constantly maintained to be equal to the set current level Io~ the absolute vertical position signal concerning the probe 104 being successively outputted -33- .

to the lateral relative position information extracting means 109. In this case, the time required for the above-described feedback control is sufficiently short as compared with the time necessary for the objects 101 and 102 to laterally and relatively displace by an amount corresponding to one pitch of the reference scale 103.
Within the lateral relative position information extracting means 109, the amount and 10 direction of the lateral relative movement are detected (step B7), this being made on the basis of the supplied absolute vertical position signal 1903 of the probe 104 and the information which concerns the reference scale 103 and the pxobe 104 (step B6). More 15 specifically, the absolute vertical position signal 1903 (whose waveform is shown in Figure 22, (a)) is subjected to waveform processing by a waveform processing means 110 and, thereafter, it is transformed by a converting means 111 into a moving 20 amount signal 112 (Figure 22, (b)) and a direction signal 113 (Figure 22, (c)~, these signals being applied to a counter 114. In this counter 114, the moving amount signal 112 and the direction signal 113 applied thereto are transformed into a lateral 25 relative displacement. Further, by using an initial position data having been pr~paratorily stored in a memory 115, a subtracting means 116 operates to ` "' ' ' .~ . ..

_34_ ~328:~3~

determine the lateral relative displacement~ at signal 117 (Figure 22, (d)), between the objects 101 and 102 (step B8).
The resolution of the encoder of the present embodiment may be of an o~der of the pitch of the reference scale 103 (a few angstroms where a periodic arrangement of atoms on the surface of an electrically conductive material is used as the reference scale).
However, a plurality of thresholds (Figure 22~ (e)) 10 may be used upon formation of the moving amount pulse signal from the probe vertical displacement signal, so that a multi-coded pulse signal such as shown in Figure 22, (f~, is produced~ By using this in combination with the relative displacement direction signal shown in Figure 22, (g~, it is possible to obtain those signals corresponding to the divided portions of the position detection signal and, therefore, to ensure the resolution less than 1 angstrom (Figure 22, (h)). With regard to the 20 vertical position control, the reference scale may be moved vertically with the detection of the vertical position thereof. This is also the case w1th the other embodiments which will be described later.
Next, referring to Figures 23 - 30, 25 description will be made of means for detecting the direction of the lateral relative displacement between the objects 101 ~nd 102, which means is provided in an :;. ' . ' ~. : . ' ~3~31 embodiment similar to the foregoing embodiment wherein the probe vertical position is detected.
Figure 23 shows one example, of the Figure 20 embodiment, in which an asymmetric shape (of an electron cloud distribution~ in the direction of the relative displacement of the objects, as illustrated in Figure 13, is used as a reference scale. Where the vertical position of a probe changes with the relative displacement of the objects 101 and 102, in a manner 10 as shown in Figure 24, (a), there are produced a moving amount signal 112 and a direction signal 113 within a subsequent lateral relative position information extracting means 109, such as illustrated in Figure 24, (b) and (c), respectively. From these signals, the lateral relative displacement such as shown in Figure 24, (d), is detectable. Therefore, in this case, it is discriminated that in the Figure 23 the object 102 has been displaced with respect to the object 101 t rightwardly in the sheet of the drawing and 20 by an amount as depicted in Figure 24, ~d).
Similarly, where tha vertical displacement of tha probe 104 changes with the relative displacement of the objects, in a manner as shown in Figure 24, (e), after the signal waveform processing 25 such as shown in Figure 24, (f~ and ~g), it is discriminated that the object 102 has been displaced with respect to the object 101, leftwardly in the ,~. .

132813~

sheet of the drawing and by an amount as depicted in Figure 24, (h).
Further, where, for example, the vertical displacement of the probe 104 changes such as shown in Figure 24, (i), it is discriminated that, at point T, the direction of relative displacement of the object 102 with respect to the object 101 has changed, from rightwardly to leftwardly.
Figure 25 shows another example, of the 10 above-described embodiment wherein the probe vertical position is detected, which example is arranged to detect the direction of the lateral relative displacement between the objects 101 and 102. In the Figure 2S example, a probe 104 is oscillated minutely in the direction of the relative displacement of the objects, by use of a laterally oscillating means 1301.
By oscillating the probe 104 sufficiently quickly as compared with the relative moving speed of the objects 101 and 102, it is possible to obtain a position 20 differentiation coefficient for the absolute vertical position signal of the probe 104 with respect to the lateral relative displacement. From this, it is : possible to detect the direction of the lateral . relative displacement.
For example, where the vertical displacement of the probe 104 and the position differentiation coefficient therefor, caused as a result of the ~ - ~
; ' , ' ` ~

_37_ ~28~3~

lateral relative displacement between the objects 101 and 102, change in the manner as shown in Figure 26, (a) and (c), then it is discriminated that in Figure 25 the object 102 has been displaced with respect to the object 101, ri~htwardly in the sheet o~ the drawing . . .
and by an amount as depicted in Figure 26, (d~.
Similarly, where the vertical displacement of the probe 104 changes with the relative displacement of the objects in the manner shown in 10 Figure 26, (e~, after the execution of the processing shown in Figure 26, (f) and (g), it is discriminated that the object 101 has been displaced with respect to the object 102, leftwardly in the sheet of the drawing and by an amount as depicted in Figure 26, (h).
Further, where there occurs a change such as shown in Figure 26, (i) or (m), it is discriminated that, at point T, the direction of the relative displacement of the object 102 with respect to the object 101 has changed, from rightwardly to leftwardly.
Figure ~7 shows a yet another ~-xample, of the above-described embodiment wherein the probe vertical position is detected, which example is arranged to detect the direction of the lateral relative displacement of the objects 101 and 102O In ~5 this example, a probe laterally oscillating means 1501 is used to oscillate a probe 104 in the direction of the relative displacement of the objects, so that at : :

-38- ~32~13~

dif~erent two points A and B absolute vertical position signals A and B of the probe 104 axe detected. It is necessary that the probe 104 is oscillated sufficiently quickly as compared with the relative moving speed o~ the objects 101 and 102 and also that, within sufficiently short time as compared with the cycle of oscillation of the probe 104, the vertical position control for the probe 104 is made by use of a probe vertical position controlling means 10 1901 (i.e. the feedback control for maintaining a tunnel current 105 constant). Where the range of oscillation of the probe 104 is denoted by d, the method of the present embodiment is equivalent to such method in which two probes spaced by an interval d are used to detect the direction on the basis of two signals having a phase difference with respect to the same reference scale.
Here, as an example, the pitch of the referellce scale is denoted by p and the range d of the 20 oscillation of the probe 104 is selested so as to satisfy the following relation:

d = P/4 + N IP/2 wherein N = 0, 1, 2 ...
In this case and where the vertical 25 displacement of the pro~e 104 changes with the lateral relative displacement of the objects 101 and 102, in the manner as shown in Figure 28, (a), it is 132813~

discriminated that in Figure 27 the object 102 has been displaced with respect to the object 101, rightwardly in the sheet of the drawing. Where the vertical displacement of the probe 104 is such as shown in Figure 28, (e), it is discriminated that the object has been displaced leftwardly.
Further, where the vertical displacement of the probe 104 changes such as shown in Figure 28, (i) or (m), it is discriminated that, at point T, the 10 direction of the relative displacement of the object 102 with respect to the ob~ect 101 has changed, from rightwardly to leftwardly.
Figure 29 shows a further example, of the above-described embodiment wherein the probe vertical position is detected, which example is arranged to detect the direction of the lateral relative displacement of the objects 101 and 102. In this example, two probes are used~ In Figure 29, two probe vertical position controlling means 2801A and 2801B
20 are adapted to execute, independently of each other, the vertical position control for two probes 1701A and 1701B spaced by an interval d (iOe. feedback control for maintaining a tunnel current constant). By detecting two probe vertical displacement signals 25 2803A and 2803B, it is possible to detect the direction of the lateral relative displacement.
For example, the pitch of a reference scale . :
: :' :................................... ... . . .

103 is denoted by p and the interval d between the probes 1701A and 1701B is selected so as to satisfy the following relation:
d = p/4 ~ N (p/2 wherein N = 0, 1, 2, ...
In this case and where the vertical displacements of the two probes 1701A and 1701B change with the lateral xelative displacement of the objects 101 and 102 in the manner as shown in Figure 30, (a), 10 it is discriminated that in Figure 29 the object 102 has been displaced with respect to the object 101, rightwardly in the sheet of the drawing. Where the vertical displacements of the two probes show the changes such as shown in Figure 30, (e), it is 15 discriminated that the object has been displaced rightwardly.
Further, where the vertical displacements of the two probes 1701A and 1701B, with the lateral relative displacement of the objects 101 and 102, 20 change in the manner such as shown in Figure 30, (i~
or ~m), it is discriminated that, at point T, the direction of the relative displacement of the object 102 with respect to the object 101 has changed, from rightwardly to leftwardly.
In accordanca with those embodiments of the present invention, having been described herein~efore, an atomic arrangement or otherwise is used as a ~328131 reference scale and a tunnel current flowing between a probe and a reference scale is detected to determine any relative positional deviation between the probe and the reference scale. Accordingly, the invention in these aspects can provide an encoder of high resolutlon of an order of interatomic distance (angstrom).
Further, in one aspect of the invention, a tunnel current is detected while maintaining, constant, the interval between a probe and a reference scale. This assures higher speed position detection.
On the other hand, the vertical position of a probe may be detected while maintaining a constant tunnel current. This assures higher-precision position detection.
15Additionally, as a reference scale, a combined structure of an atomic arrangement and a fabricated graduation or graduations may be used.
This allows the detection of an absolute position and the position detection over a wider range as well as higher-speed position detection.
Figure 31 shows the structure of an encoder according to another embodiment of the present invention. Figures 32 and 33 show signals which are obtainable at various constituent portions of the present embodiment.
In Figure 31, an object 101 and another object 102 are made relatively movable only in a : . , .
.
:~ ' ` ' ' ` `
.

-42- i3~8~3~

lateral direction (horizontal direction in the sheet of the drawing~. The object 101 is provided with an electrically conductive reference scale 103, while the object 102 is provided with an electrically conductive probe 104. To between the probe 104 and the reference scale 103, a bias voltage is applied by means of a bias voltage source 106. The probe 104 has a tip which is disposed close to the reference scale 103 so that a tunnel current schematically illustrated at 105 flows therebetween. The tunnel current 105 is converted into an electric voltage by means of a current-to-voltage converting circuit 4107 and, after being amplified by an amplifying circuit 4108, the output signal is subjected to logarithmic transforma-; 15 tion by a logarithmic transformation circuit 4109.
This is made in order that the output signal becomes i, proportional to the interval between the probe and the scale.
There is provided a probe oscillating means , 20 4110 which is operable to oscillate the probe 104 in the direction of the relative movement between the objects 101 and 102, at an oscillation frequency f and an amplitude d. The oscillation speed is made sufficiently higher than the relative moving speed of 25 the objects 101 and 102.
Probe oscillating signal is obtained in the following manner:

. . , , ~ .

1 3 ~

Oscillator 4111 produces a rectangular wave 2a having an oscillation frequency nf which is applied to a frequency dividing circuit 4112 so that the rectangular wave is transformed Into an output signal 2b. The output signal 2b is applied to a waveform transformation circuit 4113 by which it is transformed into a triangular wave (signal 2c) having an oscillation frequency f. After being amplified by an amplifier 4114, the signal 2c is applied to the probe oscillating means 4110. An appropriate oscillating means may be provided on the object 102 side to oscillate the reference scale, in place of oscillating the probe.
In order that the average interval between the probe and the reference scale is maintained constant (i.e. the average tunnel current is maintained constant~ during the lateral relative movement between the objects 101 and 102, there are provided a feedback loop. The feedback loop comprises an average tunnel current level setting circuit 4115 which is adapted to detect the output signal of the logarithmic transforma-tion circuit 4109 and to produce, in an occasion where the detected tunnel current lavel is deviated from a set level, such a signal that corrects the detected error. The feedback loop further comprises a low-pass filter 4116 and an amplifying circuit 4117. The output signal of the amplifying circuit 4117 is applied to a ~328~3~
-~4-probe vertlcal position controlling means 4118 to adjust the interval between the probe and the reference scale. Here, the cut-off frequency of the low-pass filter 4115 is so selected that (i) a quick modulation component of the tunnel current which results from the variation in height of the portion of the reference scale as opposed to the probe, which variation occurs as a consequence of the ~can of the referenca scale by the probe being oscillated laterally relative to the reference scale, can be excluded and that (ii) a gradually changing component of the tunnel current which results from the inclination or otherwise of the reference scale during the lateral relative movement between the objects 101 and 102, can be transmitted.
As a consequence, the probe vertical position controlling means 4118 does not follow the change in the tunnel current caused by the oscillation of the probe, but it follows only the change in the tunnel current caused as a result of the relative movement bekween the objects 101 and 102. In this manner, the vertical position of the probe is controlled.
As a result of the oscillation of the probe by the probe oscillating means 4110, there is produced in the tunnel current 105 flowing between the probe and the reference scale a modulation component of a frequency (2p/d)f (wherein p is the pitch of the reference scale) which is in accordance with the scan _45_ 132~13~

of the reference scale by the probe. If, at this time, the objects 101 and 102 relatively move in a lateral direction, there occurs a phase shift of the afore-mentioned modulation component of a frequency (2p/d)f, appearing in the tunnel..current 105, with respect to a reference signal (which may be the probe oscillating signal, for example)O Since one cycle of the signal (phase shift of 2~) corresponds to the lateral relative shift of the probe and the reference scale by a distance corresponding to the unit scale (the interval between two adjacent graduations) of the reference scale, the amount of lateral relative movement between the objects 101 and 102 can be detected by detecting such phase shift. In this casel even if there is a defect or otherwise on the reference scale, only a portion of the waveform of the signal is disturbed and the phase shift is not substantially affected thereby.
Therefore, the precision is hardly deteriorated due to the external disturbance such as the defect or otherwise.
Rsferring now to Figures 32 and 33, details of the signal processing method will be described.
The modulation component of the frequency (2p/d)f which appears in the tunnel current is picked up (as signal 2d), after being processed by the current-to-voltage converting circuit 4107, the amplifying circuit 4108, the logarithmic transformation ,.
, ~28~31 circuit 410g and a bandpass filter 4118. The signal 2d is binary-coded by a binarizing circuit 4119, whereby a signal 2e is produced.
Here, the amplitude of the probe oscillating signal 2c to be applied to the probe oscillating means 4110 (i.e. the gain of the amplifying circuit 4114) is adjusted to satisfy the relation "d=2p/n", such that the frequency of the signal 2e is made equal to "nf".
Additionally, while using as a reference signal the frequency-di~ided signal 2b, having been produced by frequency-dividing the output signal 2a of the oscillator 4111 by means of the frequency dividing circuit 4112 so that the frequency is demultiplied to 1/n, the signal 2e is separated into two signals 2f and 2g by use of an analog switch 4120.
Also, by use of another analog switch 4121, the signal 2a i5 separated into two signals 2h and 2i, while using the signal 2b as a reference signal.
Here, the signals 2f and 2h are applied to a phase comparator 4122 to obtain a phase difference output signal 2j which i5 in turn averaged by an averaging circuit 4123, whereby an output signal 2k i5 produced. When the objects 101 and 102 moves relative~
ly and laterally, the signal 2k changes, such as a signal 3a, in accordance with the amount of the relative movement. Further, each tlme the phase difference becomes equal to "2n~" ~n i5 an integral .

_47_ 1328131 number), the zero-cross point, ~or example, of the phase difference output signal 3a is detected by use of a binariæing circuit 4124 to produce pulses (signal 3b). The number of these pulses is counted by an 5 up/down counter 4125. By doing so, it is possible to detect the amount of relative phase deviation between the signals 2f and 2h, as an encoder output signal 3c.
At this time, the phase deviation direction signal to be applied to the counter 4125, namely, the up/down 10 condition (sign~, can be determined in the following manner:
First, by using the output signal 2a of the oscillator 4111 and by using a phase shifter 4126 and an analog switch 4127, a signal 21 whose phase is t 15 shifted by 90 degrees as compared with the signal 2h is produced. The signal 2l as well as the signal 2f are applied to a phase comparator 4128 to produce a phase difference output signal 2m which, in turn, is averaged by an averaging circuit 4129, whereby a signal 2n is 20 produced.
Similarly to the signal 2k, the signal 2n changes such as a signal 3d with the lateral relative movement between the objects 101 and 102 and in accordance with the amount of such relative movement.
; 25 Further, the signal 3d is binary-coded by a binarizing circuit 4130, such that a phase deviation direction signal, namely, an up/down signal 3e to be -48- ~328~31 appli~d to the up/down counter 4125 is produced. If the sign of this up/down signal 3e is positive at the moment of the rise point (3b1, 3b2 and 3b3) of the moving amount pulse signal 3b, the up/down counter 4125 operates to count up the number of pulses. If, to the contrary, the sign of the up/down signal 3e at the moment of the rise point is negative, the counter 4125 operates to count down the same. In this manner, the amount of lateral relative movement between the objects 101 and 102 can be detectedO In the method used in the present embodiment, one cycle (2~) of phase shift corresponds to the relative displacement by a distance corresponding to the unit scale (pitch) of the reference scale. Even if there is a defect or otherwise of the reference scale, only a portion of the waveform of the signal is disturbed and the phase shift is not substantially affected thereby. Therefore, the correctness of the measured value can be retained irrespective of the presence of any external disturbance. In place of the above described signal processing made to the signals 2f and 2h, the signals 2g and 2i may be processed in a similar manner. The relative displacement can be detected also in this way.
Figure 34 shows the structure of an encoder according to a further embodiment of the present invention.
Figures 35 and 36 show signals which are obtainable at various constituent portions of the present embodiment.

.

-``` 1328~31 In Figure 34, an object 101 and anotherob;ect 102 are made relatively movable only in a lateral direction (a horiæontal direction in the sheet of the drawing). The object 101 is provided with an electrically conductive reference scale 103, while the object 102 is provided with an electrically conductive probe 104. ~o between the probe 104 and the reference scale 103, a bias voltage is applied by means of a bias voltage source 106. The prohe 104 has a tip which is disposed so close to the reference scale 103 that a tunnel current, schematically illustrated at 105, flows therebetween. The tunnel current 105 is converted into an electric voltage by means of a current to-voltage converting circuit 4107 and, after being amplified by an amplifying circuit 4108, it is subjected to logarithm transformation by means of a logarithm transformation circuit 4109.
There is provided a probe oscillating means 4110 which is operable to oscillate the probe 104 in ~he direction of the relative movement between the objects 101 and 102, at an oscillation frequency f and an amplitude d. The oscillation speed is made sufficiently higher than the relative moving speed of - the objects 101 and 102.
Probe oscillating signal is obtained in the following manner:
Oscillator 4111 produces a rectangular wave ~ , . - . : -.~

~328~ 31 2a having an oscillation frequency f which is applied to a waveform transformation circuit 4113 by which it is transformed into a triangular wave. After being amplified by an amplifier 41141 the signal ~signal 5b) is applied to the probe oscillating means 4110. An appropriate oscillating means may he provided on the object 102 side to oscillate the reference scale, in place of oscillating the probe.
In order that the average interval between -the probe and the reference scale is maintained constant (i.e. the average tunnel current is maintained constant) during the lateral relative movement between the objects 101 and 102, there are provided a feedback loop. The feedback loop comprises an average tunnel current level setting circuit 4115 which is adapted to detect the output signal of the logarithmic transfor-mation circuit 4109 and to produce, in an occasion where the detected tunnel current level is deviated from a set level, such a signal that corrects the 20 detected error. The feedback loop further comprises a low-pass filter 4116 and an amplifying circuit 4117.
The output signal of the amplifying circuit 4117 is applied to a probe vertical position controlling means 4118 to adjust the interval between the probe and the 25 reference scale. Here, the cut-off frequency of the low-pass filter 4115 is so selected that (i~ a quick modulation component of the tunnel current which .

results from the variation in height of the portion of the reference scale as opposed to the probe, which variation occurs as a consequence of the scan of the reference scale by the probe being oscillated laterally relative to the reference scale, can be excluded and that (ii) a gradually changing component of the tunnel current which results from the inclination or otherwise of the reference scale during the lateral relative movement between the objects 101 and 102, can be -transmitted. As a consequence, the probe vertical position controlling means 4118 does not follow the change in the tunnel current caused by the oscillation of the probe, but it follo~s only the change in the tunnel current caused as a result of the relative movement between the objects 101 and 102. In this manner, the vertical position of the probe is controlled.
As a result of the oscillation of the probe by the probe oscillating means 4110, there is produced in the tunnel current 105 flowing between the probe and the reference scale a modulation component of a frequency ~2p~d)f ~wherein p is the pitch of the reference scale~ which is in accordance with the scan of the reference scale by the probe. If, at this time, the objects 101 and 102 relatively move in a lateral direction, there occurs a phase shift of the afore-mentioned modulation component of a frequency (2p/d~f, appearing in the tunnel current 105, with respect to a reference signal (which may be the probe oscillating signal, for example). Since one cycle of the signal (phase shift of 2~) corresponds to the lateral relative shift of the probe and the reference scale by a distance corresponding to the unit scale (the interval between two adjacent graduations) of the reference scale, the amount of lateral relative movement between the objects 101 and 102 can be detected by detecting such phase shift. In this case, even if there is a defect or otherwise on the reference scale, only a portion of the waveform of the signal is disturbed and the phase shift is not substantially affected thereby.
Therefore, the precision is hardly deteriorated due to the external disturbance such as the defect or otherwise.
Referring now to Figures 35 and 36, details of tha signal processing method will ba described.
The modulation component of the freguency (2p/d)f which appears in the tunnel currènt is picked up (as signal 5c~, after being processed by the current-to-voltage converting circuit 4107, the amplifying circuit 4108, the logarithmic transformation circuit 4109 and a bandpass filter 4118. The signal 5c 25 is binary-coded by a binarizing circuit 4119, whereby a signal 5D is produced.
Additionally, while using as a reference .~

. .. .

53 ~32~31 signal the output signal 5e of the oscillator 4111, the signal Sd is separated into two signals 5e and 5f by use of an analog switch 4120.
Also, a shlft register 4431 operates to provide a delay of time 1/2f, whereby a signal 5g is produced.
Here, the signals 5f and 5g are applied to a phase comparator 4122 to obtain a phase difference output signal 5h which is in turn averaged by an averaging circuit 4123, whereby an output signal 5i is produced. When the objects 101 and 102 moves relatively and laterally, the signal 5i changes, such as a signal 6a, in accordance with the amount of the relative movement. Furthex, each time the phase difference becomes equal to "2n~' (n is an integral number), the zero-cross point, ~or example, of the phase difference output signal 6a is detected by use of a binarizing circuit 4124 to produce pulses ~signal 6b). The number of these pulses is counted by an up/down counter 4125. By doing so, it i5 possible to detect the amount of relative phase deviation between the signals 5f and 5g, as an encoder output signal 6c.
At this time, the phase deviation direction signal to be applied to the counter 4125, namely, the up/down 25 condition ~sign), can be determined in the following manner:
First, a signal 5~ whose phase is shifted by :, :

132~131 90 degrees as compared with the signal 5g is produced.
The signal 5f as well as the signal 5; are applied to a phase comparator 4128 to produce a phase difference output signal 5k which, in turn~ is averaged by an averaging circuit 4129, whereby a signal 21(6d) is produced.
Simllarly to the signal 5i, the signal 51 changes such as a signal 6d with the lateral relative movement between the objects 101 and 102 and in accordance with the amount of such relative movement.
Further, the signal 6d is binary-coded by a binarizing circuit 4130, such that a phase deviation direction signal, namely, an up/down signal 6e to be applied to the up/down counter 4125 is produced. In this embodiment, the up-down counter 4125 operates to count down the number of pulses if the sign of the signal 6e at the moment of the pulse rise point of the signal 6b is positive, whereas the counter 4125 counts up if the sign of the signal 6e at the same moment is negative.
In this manner, the lateral relative displacement between the objects 101 and 102 can be detected. In the method used in the present embodiment, one cycle ~2~) of phase shift corresponds to the relative displacement by a distance corresponding to a half of the unit scale (pitch) of the reference scale.

" : ' ;
':: '' ` ' ~28~31 , ID accordance with these embodiments of the present invention, as described hereinbefore, an atomic arrangement or otherwise is used as a reference scale and a tunnel current flowing between a probe and the referenae scale is detected to thereby detect the relative positional deviation between the probe and the reference scale. Accordingly, the present invention can provide an encoder having a high resolution of an order of interatomic distance (angstrom). Further, by relatively oscillating the probe and the reference scale in the directlon of the relative positional deviation, it is possible tv ensure stable and reliable encoder in which substantially no error occurs, irrespective of the presence of any local defect of the reference scale used or any external disturbance such as vibration.
Next, description will ~e made of an aspect of the present invention according to which an encoder capable of conducting hiqh precision and wider range 20 measurement is provided.
Figure 37 shows an embodiment of the encoder according to this aspect of the present invention. In Figure 37, denoted at 5001 is a light emitting element; at 5002, a collimator lens; at 5003, a light receiving portion which comprises two light receiving elements 5003a and 5003b; ancl at 5005, a perforated code plate. Denoted at 5040 is a metallic code plate ~.

.

~,,`' ' ,:: . . :
.:.~ . .

-56- ~3~813~

which is movable in the direction of an arrow C. The code plate 5040 is provided with a scale 5041 to be used with a first position detecting means and another scale 5042 to be used with a second position detecting means, these scales being formed on the same plane.
The code plate 5040 is used as a re~lection type and, for this reason, the scale 5041 comprises what is called "black-and-white pattern" formed by a combination of such portions as having a large 10 reflection factor and such portions as having substantially no reflectivity, those two types of portions being formed at a regular pitch. A portion of the light reflected by the pattern of the scale 5041 passes through the perforated code plate 5005, and the intensity of tha light passing therethrough changes periodically in accordance with the relative displacement between the two code plates 5005 and 5040. The scale 5042 has a pitch which is very fine as compared with that of the scale 5041. The scale 20 5042 ma~ be made by the following manner: A substrate coated with a resist material is used and a scale pattern is transferred thereonto by means of electron-beam pattern drawing or lithography. After resist develspment, etching treatment is made, whereby grooves 25 each having a predetermined depth from the surface of the code plate are formed successively in the direction of arrow C. In this embodiment, the pitch of the scale ' :, , ' ' '`' ,,' ,,. ' :

~57~ 8 1 3 ~

5042 is set to be equal to 1/N of the pitch of the scale 5041, wherein N is an integral number.
There are provided a probe 5007 made of an electrically conductive material, and a probe driving means 5008 operable to move the probe in a direction substantially perpendicular to the surface of the code plate 5040. The probe driving means 5008 comprises two piezoelectic actuators 5008a and 5008b. The actuator 5008a comprises a single piezoelectric material and functions as a first probe driver. The actuator 5008a has an end connected to the probe 5007, while the other end of it is fixed to an end of the piezoelectric actuator 5008b which comprises a multi-layered type piezoelectric device and which functions as a second 15 probe driver. The other end of the actuator 5008b are fixedly secured, together with the first position detecting means, to a casing of a position detecting head (both being not shown). The first position detecting means is provided by various elements such as 20 the light emitting element 5001, the collimator lens 5002, the light receiving element 5003 and the perforated code plate 5005. The probe 5007 and the probe driving means 5008 cooperate with a control means, which comprises various blocks shown in Figure 25 38, to provide the second position detecting means.
In Figure 38, denoted at 5009 is a detecting means for detecting an electric current Ip which flows ., . ~

--5~- 132~31 through the probe 5007 when a certain voltage Vp is applied to between the code plate 5040 and the probe 5007~ Controller 5010 produces a signal Vs whiah is applied, when a switch 5011 is closed to, a piezo-electric device driver 5012 for controlling thepiezoelectric actuator 5~08a, in order that the distance between the probe 5007 and the scale 5042 on the code plate 5040 is maintained constant and, as a result of which, the electric current Ip is maintained constant. Electric voltage Vb is produced by a driver (not shown) and is used to drive the second driver 5008b so as to move the probe 5007 so that it comes close to the scale 5042, to such a distance khat allows detectlon of the electric curxent Ie. Once the voltage 15 Vb is set, the magnitude thereof is retained fixed during the use of the encoder.
Next, the operation of the second position detecting means will be explained.
The second position detecting means utilizes 20 an operational principle similar to that of a scanning tunneling microscope (STM) which is discussed in "Scanning Tunneling Microscope" hy Kajimura et al, Solid State Physics, Vol. 22, No. 3~ pp. 176-185 (1987~, or in "Recent Progress in the Scanning 25 Tunneling Microscope" by H. Adachi, Proc. 6th Sensor Symposium, pp. 137-142 (1986) or otherwise. This is also with the case of the embodiments described in the ~ ~ ' ; ! .
'. ' ' '' . ~

_59_ 1328131 foregoing.
More specifically, it is known that, when as shown in Figure 39 an electric voltage V is applied to between a sample 5044 and a probe 50Q7' made of an electrically conductive material and the interval therebetween is gradually reduced, and when the interval is reduced to a distance Z, there occurs a flow of minute electric current JT which is called a "tunnel current". The scanning tunneling microscope (STM) uses this principle. The tunnel current changes exponentially with the distance Z, as represented by - ~
equation (1) and as described hereinbefore.
Here, if the probe 5007' is moved in the direction of an arrow C while moving the same upwardly and downwardly so that the distance Z is maintained constant, the vertical movement itself of the probe 5007' corresponds to the surface shape of the sample 5044 in the direction of the arrow C. Also, ik has been confirmed that the resolution in the direction along the surface is of an order of atom.
The second position detecting means of the present embodiment utilizes this principle, as in the cases of the embodiments described with reference to Figures 20 - 36. In operation, first in Figure 38 a 25 constant voltage Vp is applied to betwean the code plate 5040 and the probe 5007. Then, by using the piezoelectric actuator 5008b while adjusting an . ' ~

-60- 1~2~131 electric voltage Vb, the probe 5007 is displaced toward the code plate 5040 and to the position of a distance through which a tunnel current can flow.
Subsequently, the switch 5011 is closed and, in order that the tunnel current Ip to be detected by a current detecting means 5009 is made constant, an electric voltage Va to be applied to the piezoelectric actuator 5008a is adjusted by use of a signal Vs supplièd from the controller 5010 to the driver 5012. If, in this state, there occurs relative movement between the probe 5007 and the code plate 5040 in the direction of ~rrow C, the tunnel current Ip changes with the surface shape (change in the surface height) of the scale 5042. Accordingly, the controller 5010 supplies 15 a signal Vs to the driver 5012 for the piezoelectric actuator 5008a, so as to maintain the tunnel current Ip constant as described hereinbefore, namely, so as to assure that the interval between the probe 5007 and the scale ~042 is maintained continuously constant.
20 Since the amount of prolongation/contraction of the piezoelectric actuator 5008a is proportional to the applied voltage Va, the signal Vs is a signal that corresponds to the surface shape of the scale 5042.
By binary-coding and waveform rectification made to 25 this signal, a signal S as illustrated in Figure 40 is obtainable. As an alternative, the voltage Va may be maintained constant while keeping the switch 5011 open;

: ,............. ............, ` , , .:. ~ , .
.: ... ~ ., ~ . . . . .
~ . ' ' ' ;, ' `` ~32813~

the spatial position of the probe 5007 in the direction of the thickness of the code plate 5040 may be retained constant; and the electric current Ip as obtainable at the current detecting means 5009 may be quantized. A
similar signal S is obtainable also in this manner.
In Figure 40, denoted at A and B are those signals of the light receiving elements 5003a and 5003b shown in Figure 37, as having been binary-coded and waveform-rectified. By the function of the perforated code plate 5005l these signals have a ; mutual phase difference of 90 degrees (the axis of abscissa denotes the posltion). In the example of Figure 40, the pitches of the scales 5041 and 5042 have a ratio of 50 tno phase shift; duty = 50:50) and, 15 also, in Figure 37 the position of a spot defined on the scale 5041 by the light emitting element 5001 of the first position detecting means as well as the position, on the scale 5042, of the probe 5007 of the second position detecting means are both on a straight 20 line D-D' whish extends orthogonally to these two scales.
Next, while referring to Figures 40 and 41, description will be made of the manner of position detection by use of the encoder structured as 25 described above.
Figure 41 shows the structure of a pulse counter used in the device of Figure 37.

. ..

, .. . ..
. :.: ~

-62 ~8131 Where the position of a point X1 with respect to the positional origin (leftward: not shown) is to be detected, the signal S from the position detecting means (Figure 38) may be counted up or down with reference to the origin, because the positiveness (or negativeness) of the moving direction can be discriminated by a flip-flop 5013 (Figure 41) and from the signals A and B produced by the first positlon detecting means. Since, however, the pitch ratio of the two scales 5041 and 5042 ~P:p = 50:1) is predetermined; first the signal A (or B) is used to execute rough counting and, thereafter, the signal S
is counted up from the point of the last rise X0 (or fall) of the signal A (or B). By doing so, the structure of the counter can be simplified. For example, as shown in Figure 41, a counter 5016 may be used to count the signal A while another counter 5017 may be used to count the signal S. In this case, the counter 5017 may be cleared at each rise (or fall) of the signal A. By doing so, the position of the point X1 can be detected. This is because:
Where the count ~or the signal A is denoted by M and the count for the signal S is denoted by N, it follows that, since P=50 p:
X1 = M-P ~ N-p = (50~M ~ N~p Denoted in Figure 41 at 5014 and 5015 each is a multiplexer for changing the input to the up/down .
, . . .
'' - . ' ~ . , -63- ~32~131 terminals (U and D) of the counter 5016 or 5017, in accordance with the moving direction, namely, the output of the flip-flop 5013. Denoted at 5018 and 5019 are displays for displaying the counts of the S counters 5016 and 5017, respectively. The resolution of the encoder can be increased in this manner.
However, if further improvement in the resolution is desired, an atomic arrangement of the surface of the code plate may be used as a scale (the size of the atom can be predetected).
Such an example is shown in Figure 42.
Figure 42 is a vertical section of the surface portion of the scale 5042. Denoted at 5042a are metal atoms constituting the code plate 5040. Denoted at 5042b are grooves having been formed by an electron-beam pattern drawing technique or otherwise so that they extend in directions perpendicular to the direction of position detectionr Generally, it is difficult to provide these grooves with the intervals of the 20 precision of an order of atomsO However, by ; preparatorily measuring the position of each groove with respect to the positional origin (by counting the number of atoms by use of the second position detecting ` means) and by storing the measured positions into a 25 memory or otherwise, it is possible to detect the position of an m-th groove 5042b in the manner as having been described with reference to Figure 41.

' -64- ~32~31 Accordingly, by starting the counting of the number n of atoms, from the position of the rise of the m-th groove, the position detection of the order of atom is assured.
Figure 43 shows a major portion of an encoder according to another embodiment. This example corresponds to a case wherein the above-described second position detectinq means includes two probes~
Reference numerals 5071 and 5072 denote such two probes which are disposed opposed to the same scale.
Denoted at 5081a and 5082a are piezoelectric actuators each functions as a second probe driver for corresponding one of the probes. Denoted at 5008b is a piezoelectric actuator which functions as a first probe driver. Denoted at d is the interval between the two probes 5071 and 5072 in the direction of the position detection tarrow C). The interval d is set with respect to the pitch p of the above-described scale 5042 to satisfy the relation:
d = (2n~ p/4 wherein n is an integral number.
Further, with each of the probes 5071 and 5072 and the second probe drivers 5081a and 5082a, the control means (Figure 38) is operationally associated 25 such that, in a similar manner as described, the distance between the scale 5042 and each probe 5071 or 5072 can be controlled and maintained constant. With .

' ~ '' . :, .'',. .' ' . '. ' . :'-~3~13~

this arrangement, two signals having a phase difference of 90 degrees are obtainable in the same way as described and, therefore, it is possible to detect the moving direation even if the distance of movement is so minute that the direction thereof can not be detected by the first position detecting means.
Figure 44 is a perspective view showing a major portion of an encoder according to a still further embodiment of the present invention. In this embodiment, the scale 5042 of the foregoing embodiment to be used with the second position detecting means is provided by two rows of scales 5421 and 5422 having a mutual phase shift of an amount corresponding to a quarter of the pitch (i.e. p/4). In place thereof, any phase shift between two probes 5071 and 5072 (as in the case of the foregoing embodiment) is avoided and they are disposed above the scales 5021 and 5022, respectively. Alsol in the present embodiment, substantially the same advantageous effects as 20 attainable in the Figure 43 embodiment are attained.
Figures 45A and 45B show an encoder according to a yet another embodiment of the present invention, wherein Figure 45A is a sectional side view while Figure 45B shows the encoder as viewed from the 25 below in Figure 45A. In these Figures, denoted at 6007 is a probe made of an electrically conductive material;
and at 6008, a driving means which comprises two piezo-. .

-66- 132~131 electric actuators 6008a and 6008b and which is operable to move the probe 6007 in a direction substan-tially perpendicular to the surface of a second code plate 6042 which will be described later. The piezo-electric actuator 6008a comprises a singlepiezoelectric material and functions as a first probe driver. The actuator 6008a has an end connected to the probe 6007, and the other end thereof is fixed to an end of the piezoelectric actuator 6008b which is a multi-layered type piezoelectric device and which functions as a second probe driver. The other end of the piezoelectric actuator 6008b is fixedly secured to a holding member 6100 for holding, in a predetermined positional relationship, various elements such as a light emitting element 6001, a collimator lens 6002, a light receiving element 6003 and a perforated code plate (mask) 6005 which cooperate with each other to provide a first position detecting means. The probe 6007 and the probe driving means 6008 cooperate with a control means, comprising various blocks shown in Figure 46, to provide a second position detecting means. Denoted in Figure 45A at 6041 is a first code plate having a surface 6410 on which, for allowing that the first code plate 6041 is used as a reflection type, there is provided a scale, not shown, formed by what is called a "black-and-white pattern" which comprises such areas as having a large reflection factor and such ' . ' ~ ' ' '' ., ,';',''.' '' - '-."'. ~':' ' ' ,, , ' ~

-67- 13~13~

areas as havlng substantially no reflectivity, these two types of areas being disposed at regular pitch.
Denoted at 6101 is a casing having four parallel leaf-springs of metal plates 6102, made of phosphor bronze or otherwise, for supporting the holding member 6100 for relative movement in the direction of an arrow C.
The casing as a whole may be referred to as a "measuring head". The measuring head 6101 is made movable relatively to the first code plate 6041.
Denoted at 6042 is a second code plate which is made of an electrically conductive material and which is disposed on the inside surface of the measuring head ~101 to be opposed to the probe 6007. A constant electric voltage is applied to between the probe 6007 and the second code plate. On the surface 6420 of the second code plate 6042, there is formed a scale 6021 which comprises grooves or protrusions of a predetermined depth (height) having a pitch which is very fine as compared with that of the scale of the first code plate 6041. The grooves of the scale 6421 are formed successively in the direction of arrow C, by means of electron-beam pattern drawing process or lithography process.
For example, the pitch of the scale 6421 is set to be equal to 1 /N of the pitch of the scale of the first code plate 6041, wherein N is an integral number.
Denoted at 6108 is a drlving means which is provided -68- ~328~31 between the holding member 6100 and the measuring head 6101, for moving the holding member 6100 relatively to the measuring head 6101. This driving means comprises a multilayered type piezoelectric actuator, for example. Denoted at 6003 is a light recelving device which comprises two light receiving elements 6003a and 6003b, as in the case of Figure 37.
Figure 46 is a block diagram showing a control means of the encoder of the Figure 45 embodiment.
In Figuxe 46, denoted at 6009 is a current detecting means for detecting an electric current Ip which flows through the probe 6007 when a certain electric voltage Vp is applied to between the probe 6007 and the second code plate 6042. Controller 6010 produces a signal Vs to be applied, when a switch 6011 is closed, to a piezoalectric device driver 6012 for controlling the pie~oelectric actuator 6008a, in order that the distance between the probe 6007 and the scale 6021 on the second code plate 6042 is maintained constant and, as a result of which, the electric current Ip is maintained constant. Electric voltage Vb is produced by a driver (not shown) and is used to drive the piezoelectric actuator 6008b of multilayered type so as to move the probe 6007 toward the scale 6421 and to the position of such distance by which the current Ip can be detected. Once the voltage Vb is , :. -' :' , . .
' ' ' ,., ', ~ :

-69- ~2~3~

set, the magnitude thereof is retained fixed during the use of the encoder.
The operation of the second position detecting means is similar to that of the Figure 37 embodiment. And, for this reason, description thereof is omitted here for simplicity.
Figure 47 is a timing chart showing the relationship among the signals produced at various portions of the device of Figure 45. Denoted in Figure 47 at A and B are those signals from the light receiving device 6003 (the elements 6003a and 6003b~
as having been binary-coded and waveform-rectified. By the function of the mask 6005, these signals have a mutual phase shift of 90 degrees (the axis of abscissa denotes the position). Pulses PC are formed by generating a pulse of a predetermined width, starting from each rise or fall of these two signals and, thus, correspond to those in which the original pulse generation frequency is multiplied by four.
The manner of position detection by use of the encoder structured as described above, will now be explained taken in conjunction with Figures 45A, 45B
and 47 as well as the flow chart of Figure 48, showing the -sequence of measuring process.
In an occasion where, in Figure 47, the distance from a positional origin X0 to a position Xk is to be measured and if only the first position ~32~3~

detecting means is used, the pulse counting starts at the leftmost point in Figure 47 to count the pulses PC. Thus, the value "M x P" as determined by the number M of the pulses, from the origin X0 to the position X1 and by the pitch P of the pulses (which is equal to 1/4 of the pitch of the scale of the code plate 6041), is obtained as the distance from the origin to the position Xk. Wi~h this method, however, the length S from the position X1 to the position X2 is not measured. In other words, the measured value contains an error corresponding to one pulse spacing (~P) of the signal PC. According to the present embodiment, this length S can also be measured by use - of the second position detecting means, to thereby improve the measurement accuracy (resolution).
The measurement process starts at the execution of the preparation (such as described below) to be made prior to the measuring operation.
Referring back to Figures 45~ and 45B, first, the driving means 6108 moves the holding member 6100 to displace the probe 6007 to a reference position on the code plate 6042 (step 6110). This is done in order that the first position detecting means is placed at a fixed position with respect to the measuring head 25 casing 6101. Subsequently, a counter (not shown) for counting pulses PF is reset IsteP 6120). The aforesaid reference position can be discriminated by, .;
. ~
.

for example, preparatorily and locally changing the pitch of the scale 6421 on the second code plate 6042.
Subsequently, the measuring head 6101 is moved to the positional origin X0 (step 6130~ and a counter (not 5 shown) for counting the pulses PC is reset (step 61 40). For the positioning at this positional origin, a separate light switch such as a photo-interrupter may be used. Alternatively, a specific mark for the positional origin may be provided on the first code 10 plate 6G41 and this may be detecked by the first position detecting means.
If, at step 6051, the discrimination result shows that the measurement is to be efected, first the measuring head 6101 is moved toward the measuring 15 position lstep 6160). While counting pulses PC by use of the first position detecting means (step 6170), the measuring head is moved. When it reaches the measuring position (step 6180), the measuring head 6101 is stopped (step 61 90~. Further, for measuring 20 the minute distance S in Figuxe 47, the holding member 6100 is displaced by the driving means 6108 in a direction opposite to that in which the measuring head 6101 has moved, until the last-counted pulse PC
appears (step 6200). During this displacement, pulses 25 PF are counted by use of the probe 6007 and the PF
pulse counting counter (step 6210). When the pulse PC
appears ~although at this time the PC pulse counting - , .

-72- 13~131 counter is not decremented by one) (step 6220), the distance of the measuring position from the positional origin is calculated from the counts of the PC pulse counting counter and the PF pulse counting counter ~step 6240). For example, if the two counts are M and N, then the distance X to be measured can be expressed as follows:
X = M-P ~ N p Finally, the holding member 6100 is returned to the reference position and the PF pulse counting counter is reset, for preparation for the next measurement.
In this manner, the measurement with higher resolution as compared with those of the conventional encoders is attained. If, however, further improvement in the resolution is desirsd, an atomic arrangement on the surface of the second code plate may be used as a scale, as in the case of the Figure 42 embodiment (the size of atom may be predetected).
20 Since such a scale has been explained with reference to Figure 42, description thereof is omitted here for simplicity.
It i5 sufficient for the second position detecting means to play a role supplementing the first 25 position detecting means, and for this reason the range of measurement thereof may be small. At the minimum, a range not smaller than the pitch P of the pulses PC is . . . . .. .
.. . -- .
.
. . ! ' . .
`'` ' ' ~ ~, ' "" " '' ' ""'' ' ''' "

sufficient. For this reason, the size of the second code plate and/or the actuator (driving means~ may be small. As a result, the measuring head including these components may have substantially the same size as of that of the conventional type measuring head.
As a matter of course, the second position detecting means of the Figure 45 embodiment may be replaced by that as having been described with`
reference to Figure 43 or 44. Also, as shown in Figure 49, it is possible that a probe 6007 and a probe driving means 6008 are provided on the inside surface of a measuring head 6101 while a code plate 6042 is provided on a holding member 6100, such that the relationship between the probe 6007 and the code plate 6042 is reversed as compared with the foregoing embodiment. Similarly r in the other embodiments, the relationship between the scale and the first or second position detecting means may be reversed.
In the embodlments described with reference 20 to Figures 37 - 49, the first position detecting means is not limited to an optical type, but a magnetic or other type detecting means may be used. Further, ~hile the foregoing embodiments have been described with referenca to a linear type measuring system, the 25 device may be arranged as a rotary type (namely, as an angle measuring device).

. .

.

_74 1328~3~

Next, description will be made of another aspect of the present invention. In this aspect, it is intendeA to provide a recording/reproducing apparatus wherein a probe electrode is used to conduct 6 electric high-density recording/reproducing and wherein the functions of high-precision positlon detection and position control are introduced to assure the recording/reproducing with high density and good reproducibility.
Briefly, in accordance with this aspect of the present invention, there is provided a recording/reproducing apparatus which includes a recording medium having a position coordinate axis functioning as a reference, voltage applying means having a probe electrode, and detecting means for detecting the position on the position coordinate axis, such that the recording of information and/or the reproduction of recorded information can be made at such position on the recording medium that corresponds to the detected coordinate position.
The position detecting device and recording/reproducing of information, employed in some embodiments in this aspect of the invention, utilize the phenomenon that, when an electrically conductive 25 probe (probe electrode) and an electrically conductive material are brought close to each other to a clearance of about 1 nm while applyi.ng an electric ~' :``
;: . - , - .. . . .
, .
. ' : ' ~ :

132~131 voltage to therebetween, there occurs a flow of tunnel current therebetween. Since such a tunnel current depends on the work function on the surface of a conductor, information concerning various surface electron states can be read out. The present invention uses this. More specifically, a recording medium having a regular atomic arrangement or an arbitrary formed origin, operable as a reference, is used and, to such recording medium, a position coordinate system based on the regular atomic arrangement or the reference origin, is introduced.
The position detection is effected by detecting a characteristic change in the tunnel current corresponding to the position coordinate system.
Also, on the basis of the result of the position detection, the recording/reproducing position on the recording medium, which is in a determined positional relation with the position coordinate system, is specified, and the positioning of the probe electrode to such recording/reproducing position is controlled.
Thereafter, by use of the probe electrode, the recording/reproducing is effected.
The positional relation between the coordinate axis and the recording position at this 25 time is schematically shown in Figure 50. The positional information (A - I) in the form of graduations on the coordinate axis is always in a .

...,. ~
'' ~
~ ' ' -76- 1328~31 "relative" positional relation with the recording position (A' - I'). Namely, A:A', for example.
Therefore, by detecting the positlonal information A -I, the recording position A' - Il can be necessarily specified. In this case, it is not always necessary that the points (graduations) on the coordinate axis and the recording positions are placed in a univocal relative disposition. For example, there may be plural recording positions such as a point A'', a point Al'l, etc. as well as a point Al, all correspond-ing to the positional information A. However, from the viewpoint of precision, a univocal relation (one-to-one relation) is preferable. Further, use of a single coordinate axis is not a limitation, but plural axes may be used as required. Additionally, the coordinate system is not limited to one-dimensional, and a two-dimensional system ~grid-like system~ may be used. In such case, in accordance with the grid points of the two-dimensional coordinate system, the recording positions may be distributed two-dimensionally.

Coordinate Axis - The coordinate axis used in the present invention as a member of a position detecting system may be formed by use of a regular atomic arrangement or, alternatively, an arbitrary formed reference , -77- ~ 328~3~

point. As for such a regular atomic arrangement, an electrically conductive material whose lattice spacing is predetected (for example, one of various metals, graphite monocrystal or otherwise) may be used.
Additionally, since the tunnel current used in this aspect of the present invention has a magnitude of an order of nA, those materials as having an electric conductivity not less than 10-1 (ohm~cm)~1 may be used as the aforementioned conductive material.
10 Accordingly, a monocrystal of semiconductor such as silicons, may be used. As a representative of these materials, a metal sample is now considered.
In equation (1), the tunnel current density JT changes with the distance Z. Therefore, if a probe electrode is scanned in an arbitrary rectilinear direction relative to the surface of the metal sample, while maintalning constant an average distance between the probe and the reference atomic arrangement~ the tunnel current changes periodically in accordance with 20 the metal atomic arrangement. Where such a metal sample whose lattice constant is predetected is used, the state of the atomic arrangement in an arbitrary direction with reference to a certain lattice point upon an arbitrary crystal plane is self-evident and, 25 therefore, the periodic change in the tunnel current which occurs as a result of the scan of the probe electric in that direction can be satisfactorily :. ~

:

-78- ~ 328 ~ 31 predicted. Thus, by correcting the scan direction of the probe electrode so as to assure that the predicted value of the change in the tunnel current and the measured value of the change in the tunnel current, caused as a result of actual scan o~ the probe electrode, become equal to each othex, it is possible that the motion of the probe electrode accords with the atomic arrangement of the sample. Namely, if the atomic arrangement is considered as a coordinate axis, 10 the probe electrode moves just along this coordinate axis. Assuming now that the probe electrode on this coordinate axis is movable in a particular direction and to a position spaced by a particular distance and that the end of transfer (destination) is at a 15 recordable/reproduceable region, the recording/re-producing can be made at each position corresponding, in a one-to-one relation, to each point on the coordinate axis. In this case, it is not always necessary that the probe electrode moves between the 20 coordinate axis and the recording xegion. For example, to a probe electrode (position detecting probe electrode) moving along the coordinate axis, a separate recording/reproducing probe electrode may be provided at a predetermined position and they may be 25 operationally associated with each other.
In any case, the position of the probe electrode in the recording region, namely, the , .

-7g- ~3~31 recording position, can be determined univocally with respect to the coordinate axls which uses the crystal lattice of a metal sample.
In an occasion where a portion of or all of the surface of a recording medium has a regular atomic arrangement and where the state of such arrangement is predetected, it is possible to set such a recording region as having an X-Y coordinate system which shows a univocal relation with a coordinate axis using the 10 crystal lattice of such atomic arrangement.
As for the coordinate axis for the position detecting, as other alternatives, a recessed or protruded structure may be formed on the surface of a sample, or a plurality of reference points may be artificially formed by ion injection of atoms. The surface structure or these points can be used as a position coordinate, although the precision as a coordinate axis may be inferior, as compared with one in which the atomic arrangement is used.
In the described manner, a position coordinate can be set on a recording medium and, at each point corresponding to the position coordinate, the recording/reproducing can be made. Upon actual recording/reproducing, however, it is necessary to 25 specify the start point. Namely, it is necessary to define an origin on the coordinate axis, which functions as a reference. As for such reference - ~ :

~2813~

origin, a surface step (recess or protrusion) may be formed on the coordinate axis by an etching process or otherwise or, alternatively, ion injection may be used to change the state of the surface of the recording medium. However, as described hereinbefore, the precision thereof is not satisfactory for use as an origin of a coordinate axis using an atomic arrangement. When in Figure 50 the point A on the coordinate axis is to be selected as a reference origin, discriminating the point A is equivalent to discriminating a point A' on the recording region which is in a univocal positional relation with the point A. In other words, if it is possible to discriminate the point A', the coordinate axis and the positions of the points on the coordinate axis are determined univocally. As for the method of setting a reference origin at the point A', a method of inputting information concerning an orlgin to the point Al, like a method of record writing, is superior in respect to the precision as well as ease in the formation. The number of such reference origin is not limited to one, and plural reference origins may be formed in accordance with requirements such as enlargement of the recording region.

Recording Medium As for the recording medium usable in this ,. ' ~

.
' .

aspect of the present invention, those materials as providing, in current/voltage characteristics, a memory switching phenomenon (electric memory effect) may be used. Examples are as follows:
~1) Amorphous semiconductors such as oxide glass, borate glass or chalcogenide glass containing Se, Te or As combined with elements of groups III; IV, V and VI in the periodic table. These materials are intrinsic semiconductors having optical band gap Eg of 0.6 - 1~4 eV or electric activation energy ~E of about 0.7 -1.6 eV. Examples of chalcogenide glass are: As-Se-Te series, Ge-As-Se series, Si~Ge-As-Te series, for ample Si16Ge14AS5Te65 (suffix is atom ~), Ge-Te-x series, Si-Te-X series (X = a small amount of V or VI
group element~ for example Ge1sTe81Sb2S2. Further, Ge-Sb-Se series chalcogenide glass may be used.
In an amorphous semiconductor layer wherein the above-described compound is deposited on an electrode, the electric memory effect of the medium can be presented by applying an electric`voltage perpendicularly to the film surface by use of a probe electrode.
As for the method of depositing such material, any well-known thin film forming process may be used to sufficiently achieve the ob~ects of the present invention. Preferable examples of film forming method are: vacuum deposition method and - . : .
' , -82~ 132~31 cluster ion beam method. Generally, the electric memory effect of such a material is observed on the condition of thickness not greater than a few microns and in respect to the recording resolution as a recording medium, thinner thickness is preferable.
However, from the view points of uniformness and recording performance, a thickness not less than 100 angstroms and not greater than 1 micron, preferably not greater than 1000 angstroms, is desirable.
l2) An organic semiconductor formed by depositing on an electrode a salt composed of a metal such as Cu or Ag having a relatively low reduction potential and an electron accepting compound such as tetraquinodimethane (TCNQ), TCNQ derivative, for example tetrafluorotetracyanoquinodimethane (TCNQF4~, tetracyanoethylene (TCNE) and tetracyanonaphthoquino-dimethane (TNAP).
As for the method of forming such an oryanic semiconductor layer, use may be made of a method of 20 vacuum-depositing the above-described electron accepting compound on an electrode of Cu or Ag.
; The electric memory effect of such an organic semiconductor is observed on the condition of a thickness not greater than several tens microns.
25 However, from the view points of film formation property and uniformness, a thickness in a range of 100 angstroms to 1 micron is preferable.

' ~3~8~31 (3) Recording mediums made of amorphous silicon material~ Examples are those recording mediums having a layered structure of metal/A-Si (p+-layer/n-layer/i-layer) or metal/A-Si (n+-layer/p-layer/i-layer).
Deposition of each layer of A-Si can be satisfactorily made by use of any well-known method. Preferably, in the present invention, glow discharge method (GD
method) is used. As for the film thickness of the A-Si material, the n-layer preferably has a thicknsss of 2000 - 8000 angstroms while each of the i-layer and p+-layer preferably has a thickness of about 1000 angstromsO The total thickness is preferably about 0.5 - 1 micron.
(4~ Recording mediums having on an electrode a laminated structure of those molecules as having a group o~ ~-electron level and a group of only ~-electron level.
As the structure of the dye having ~-electron system suitable for the present invention, 20 there may be included, for example, dyes having porphyrine skeleton such as phthalocyanine, tetraphenylporphyrine, etc.; azulene type dyes having squarilium group and croconic methine group as the bonding chain and cyanine-like dyes having two or more 25 nitrogen-containing heterocyclic rings such as quinoline, benzothiazole, benzoxazole, etc. bonded through squarilium group and croconic methine group;

. '. . ~ . -..... .

-84- 132~13t or cyanine dyes, condensed polycyclic aromatic compounds such as anthracene and pyrene, and chain compounds obtained by polymerization of aromatic ring and heterocyclic compounds; and polymers of diacetylene group; further derivatives of tetraquinodimethane or tetrathiafluvalene and analogues thereof and charge transfer complexes thereof; and further metal complex compounds such as ferrocene, tris-bipyridine ruthenium complexes, etc.
As for the formation of an organic recording medium, specifically a deposition method, cluster ion beam method or otherwise are usable. However, from the points of con~rollability, easiness and reproducibility, the LB method is very suitable one among various conventional processes.
According to this LB method, a monomolecular film or its built-up film of an organic compound having, in one molecular, a hydrophobic site and a hydrophilic site, can be easily formad on a substrate~
20 Thus, with a thickness of an order of molecule and over a wide area, a uniform and homogeneous organic super-thin film can be produced stably.
The LB method is a method for forming a monomolecular film or a built-up film by utilizing the 25 phenomenon that, where, in the molecule structure having in the molecule a hydrophobic site and a hydrophilic site, a suitable balance (amphiphatic ' ~328~31 balance) is maintained therebetween, molecules form a monomolecular layer on the surface of water with the hydrophilic radical facing downwardly.
As for the radical constituting the hydrophobic site, there are various hydrophobic radicals such as, for example, prevalently known saturated and non-saturated hydrocarbon group, condensed polycyclic aromatic group and chain polycyclic phenyl group. They can provide singly or in combination a hydrophobic portion. On the other hand, most typical examples as the constituent elements of the hydrophiIic portion are hydrophilic radicals such as carboxyl group, ester group, acid amlde group, imide group, hydroxyl group and amino group (first, second and third classes and fourth class). Similarly, they can provide singly or in combination the hydrophilic portion of the molecule.
Those dye molecules, if they have well-balanced hydrophobic and hydrophilic sites and ~-electron system of a suitable magnitude, can form amonomolecular film on the surface of water and, therefore, they can be used as very suitable materials for the present invention.
As specific examples, there may be molecules such as described below.

, -86- 1328~:3~

[ I ] Croconic methine dyes ~N> ~CH--C~

~f~C=CH$~CH--C6 3) O

>C = C H~C H--C~

.
' .
,~ .

~ ., .

-87- ~ 8~ 3~

5) 0 ~>C C H h 6~

6) 0 1 0 &~5 >C = C H~ C H--C~

7 ) O
~3= C H~ =C H
R l ~ O ~ ' ,, .
CH ~ e , 8 ) 1 0 IX >C = C H~C H--C~

:`' ...
''', ' ~8~3~

9 ) C~CH 3 0 C~CH 3 ~N>C=CH~ ~N~

10) 0 RI > ~3~N<CRH 3 O O

~5 R ~ R

In the above examples, R1 denotes a long chain alkyl group corresponding to a group having the above-described ~ electron level and haviny been introduced to make easier the formation of a monomolecular film on the surface of water. The carbon number n thereof is preferably in the range of ' - ,, 89 1328~1 5 < n < 30. Those compounds listed above as specific examples are only of fundamental structure, and various substitution products of these compounds may of course be preferably used in the present invention.

~II] Squarilium dyes Compound in which croconic methine group of the compound as described in item [I] i5 substituted by a squarilium group having the following structure: -1a ~ .
O

: ... .. ~ .

, . ' .

go 132~131 [ III ] Porphyrine type dye compound R ~ R~, R2 . R3, R4 = H, ~ ~.

~N)= /CH 3 R 4 ~I¢N--M N~ --O--CH 2--C--CH 3 ~N~ I --O C S H ~1 .
~ --C(CH3)3, M=H 2, Cu. Ni, Al--Cl, SiC 1~ 2 and rare earth metal ion 2) \~ R=OCH (COO~I) ~N M N~
N ¦ M=H 2, CU. Ni, Zn, Al--Cl, R \c=/ ~ Sic 2 2 and rare earth metal ion . ~:

-91- 1 3~813~

3) Br R--N ~ R=CnH2n~l 5 < n < 25 N M N ~
~N ¦ Br- M =H 2. CU, Ni, Zn, Al-CI
_ ~ +

~ =J ~ N- R SiC ~ 2 and rare earth metal ion Br-In these examples, R has been introduced to make the formation of a monomolecular film easier and it is not limited to the substituent group ~entioned above. Also, R1 - R4 and R correspond to the above-described groups having ~-electron level.

-92- :L3~8~3~.

I IV] Condensed polycyclic aromatic compound ~0~'0~ `' R=C 4 H g ~C 12 H 25 (CH 2) 2 COOH

CnH 2 n COOH

o < n ~ 20 ~<
N-- (CH 2 ) n --COOH

~O~- C--O (CH 2 ) n --COOH

~93~ ~32~131 [V] Diacetylene compounds CH 3 ~ CH 2 3-n C--C--C--C~CH 2 3-m X
0 < n, m < 20 wherein nt-m ~ 10 X denotes a hydrophilic radical and usually -COOH is used. However, -OH, -CONH2 or otherwise may be used.

[VI] Others Quinquethienyl 2 ) CH 3 ~CH 2 ) 4 ~ CN

~ i _94_ ~ 3~8~ 3~

3 ) ~ R
R=CONHC 18 H 37, Fe ~- , 4 ) ~ OCOC 17 H 35 '1-Fe ~ OCOC 17 H 35 5 ) ¦ H4sCæ-- N~N ¦

20 ~c>=~ <C~

6 ) R
NC \ CN
~=,~=,~/ R=C 18 H 37 NC / ~ \ CN

:, :

.
- .
_., .

132~31 Other than the aforementioned materials, those dye materials suitable to the LB method may of course be conveniently used in the present invention.
For example, biological material (such as, for example, bacteriorhodopsin or cytochrome c), synthetic polypeptides (PBLG, etc.) or otherwise are usable.
The electric memory effect of these compounds having ~-electron level is observed on the condition of a thickness not greater than several tens microns. However, from the viewpoints of film formation property and uniformness, a thickness in the range of 15 - 2000 angstroms is preferable.
The substrate for supporting those materials described in items (1) - (4) and having the electric memory effect should have characteristics as an electrode. However, any material can be used provided that it is an electric conductor having a conductivity not less than 10-6 (ohm.cm~1). Namely, many varieties of materials may be used. Examples are: a metal plate such as Au, Pt, Pd, Ag, Al, In, Sn, Pb, W or otherwise; an alloy of suitable ones of these materials; a glass, ceramics or plastic material on which such metal or alloy is deposited; Si (crystal, amorphous) or graphite; electrically conductive oxide such as ITO or otherwise, etc.
The recording medium of the present invention may be provided by a combination of such an .: ' . : ' -96- 132~3~

electric memory effect exhibiting material and a supporting base (electrode). Where an atomic arrangement is used as the above-described coordinate axis, the atomic arrangement of such an electric 5 memory effect exhibiting material itself is, in many cases, inferior in respect to the regularity and, therefore, it is not desirable to use the same as the coordinate axis. Accordingly, it is desirable to use, as the supporting base, a material having a regular - t 10 atomic arrangement such as, for exampla, metal, crystal Si or graphite and, while defining on that material a portion in which the electric memory effect exhibiting material is not deposited, to use the atomic arrangement of that portion of the base as the coordinate axis.

Probe Electrode In order to assure increased resolution of recording, reproducing and/or erasing, it is necessary 20 to point, as sharp as possible, the tip of a probe electrode used in the present invention. In one embodiment, a probe is formed by mechanically grinding the tip of a tungsten wire of a diameter 1~ so that it is shaped .into a cone of 9O degrees and, thereafter, 25 by applying an electric field to it in a super high vacuum to cause vaporization of surface atoms.
However, the shape of the probe or the process of ~ , .~ ., :. .

~32~31 treatment is not limited thereto.
Further, it is not always necessary to use separate probe electrodes for the pos~tion detection and for the recording/reproducingO A common probe electrode may be used~ Also, different probe electrodes may be used for each of the position detection and the recording/reproducing.
Figure 51 is a schematic and diagrammatic view of a recording apparatus according to one embodiment of the present invention and, in this example, two probe electrodes are provided for the position detection and for the recording/reproducing, respectively. Denoted in Figure 51 at 7102 and 7103 are the probe electrodes used for the recording/reproducing and for the position detection, respectively. Although the distance between these two probe electrodes can be adjusted minutely by means of a probe electrode interval minuta adjusting mechanism 7112 using a piezoelectric device, usually they are maintained at a fixed interval. Denoted at 7106 is a probe current amplifier andl at 7109, is a servo circuit for controlling fine motion driving mechanisms 7107 and 7108 each using a piezoelectric device, so as to maintain the probe current constant. Denoted at 7113 is a voltage source for applying, to between the recording/reproducing probe electrode 7102 and a base electrode 7104, a pulse voltage for the ..
.. . .

-98- ~32~13~

recordinglerasing.
Since the probe current changes radically when the pulse voltage is applied, the servo circuit 7109 controls to turn a hold circuit on so that the output voltage is maintained constant during that period.
; Denoted at 7110 is an X-Y scan drive circuit for controllably moving the paired prove electrodes 7102 and 7103 in the X and Y directions. Rough-motion drive mechanisms 7114 and 7115 are used to pre-: paratorily and roughly control the distance between the recording medium 7001 and the probe electrode 7102 (7103) so that a probe current of an order of 10~9 A
; is produced, or to allow a large relative displacement between the probe electrode and the substrate in the Xand Y directions (beyond the range of the fine-motion control mechanism).
These components are controlled by centralized control by use of a microcomputer 7116.
20 Reference numeral 7117 d~notes a display device.
Mechanical performance in the movement control by use of piezoelectric devices is as follows:

Z-axis fine-motion controllable range:
0.1 nm - 1 micron Z-axis rough-motion controllable range:
., .~ 10 nm - 10 mm i, . . .

. . .
:

gg ~328~ 31 X-Y scan range: 0.1 nm - 1 micron X-Y rough-motion controllable range:
10 nm - 10 mm Tolerance for measurement and control:
< 0.1 nm (fine-motion control) Tolerance for measurement and control:
< 1 nm (rough-motion control~

Details of the manner of recording/reproduc-ing in the present embodiment will be explained below, particularly with reference to some examples thereof.

Example 1 As the recording/reproducing apparatus, that shown in Figure 51 was used and, as the probe electrode 7102 ~7103), a probe electrode made of tungsten was used. The probe electrode 7102 (7103) i~
proYided for controlling the distance Z to the surface of the recording medium 7001, and the distance Z of 20 each probe electrode can be controlled by fine motion, independently from the other, by means of a piezoelectric device such that the electric current is maintained constant. Further, the fine-motion controlling mechanism is arranged to effect fine-25 motion control also in the direction within a plane(X-Y), while retaining the distance Z constant.
Of the two pxobe electrcdes, the position . .

~ . . , . ,:

-1 00 - ~L 3 2 8 1 3 1 detecting probe electrode 7103 is used for the detection of the atomic arrange.ment of the base 7105, which functions as a position coordinate. On the other hand, the recording/reproducing probe electrode 7102 is held at a fixed position with respect to the position detecting probe electrode 7103 in the X and Y
directions (the interval therebetween being ad;us~able by use of the probe electrode interval minute adjusting mechanism 7112), and ls used for the 10 recording, reproducing and/or erasing on the recording layer 7101.
While these two probe electrodes are adapted to be controllably moved, as a unit, by fin~-motion control in the X and Y directions within the plane, with respect to the Z direction they can be controlled independently from each other by fine-motion. The recording medium 7001 is placed on a high-precision X-Y stage 7118 and can be moved to a desired position ~X-Y rough-motion mechanism). The directions X and Y
20 of the rough-motion mechanism and the directions X and Y of the fine-motion mechanism can be in accord with each other, within the range of errors due to the difference in precision between these movement controlling mechanism.
Details of the recording medium used in the present example will be descrlbed. Figure 52 shows the structure of the recording medium. P-type Si '' , '' ~ : .

~ , wafer (B-dope 0.3 mm thickness) ha~ing a diameter of 1/2 inch with a plane (111) being exposed, was used as a substrate 7105O The substrate is cut along a line B-B' for the purpose of attaining substantially constant orientation thereof when it is placed on the X-Y stage 7118 of the recording/reproducing apparatus.
The direction B-B' is substantially parallel to the direction [211] of the Si crystal. Subsequently, at the position at a distance of 1 mm from the middle point of the side B-B' to the center of the base, a portion of the base was etched by a range of 1-micron square and 0.2-micron depth, to form a reference origin (rough) 7201. The method of forming such a reference origin (rough~ will be explained below.

First, the surface of a Si base plate was ~ k coated with polymethyl methacrylate ~PMMA; trade-~a~
"OEBR-1000" manufactured by Tokyo Ohka Kogyo, Inc.
Japan) which is an electron-beam resist material.
Then, an electron beam of a diameter 0.1 micron was 20 projected with an acceleration voltage 20 keV to draw a pattern of a size of 1 micron square. Subsequently, by using a specific developing liguid, the portion irradiated by the electron beam was solved. As for the etching, a mixed gas of CF4 and H2 was used to effect 25 sputter etching with a pressure 3Pa and a discharging electric power 100 W, for 20 minutes. The etching depth was ~.2 micron. Finally, by using methyl ethyl ' `

` -102- 1328~31 ketone, the PMMA was solved.
Subsequently, after masking the substrate portion in the neighborhood of the reference origin (rough~ 7201, Cr was deposited by vacuum evaporation to a thickness 50 angstroms to provide a background layer.
Further, Au was deposited by the same method to a thickness of 400 angstroms to provide a base electrode 7104.
Subsequently, on such Au electrode, an LB
10 film (eight layers) of squarilium-bis-6-octylazulene (hereinafter simply "SOAZ"3 was laminated to provide a recording layer 7101. Details of the method of forming such a recording layer are as follows:
First, benzene solution in which SOAZ was solved at concentration 0.2 mg/ml, was spread on the surface of water having a temperature of 20 C to form on the water surface a monomolecular film. After evaporation of solvent, the surface pxessure of the monomolecular film was increased to 20 mN/m and, while 20 maintaining that constant, the base plate was slowly and repeatedly dipped and pulled up in the direction across the water surface and at a speed of 3 mm/min., whereby an eight-layer built-up film of SOAZ
monomolecular film was formed on the base electrode 25 7104.
By using the recording medium 7001 formed in the described manner, recordinglreproducing 1328~31 -~03-expariments were made.
Details are as follows:
The recording medium 7001 having a recording layer 7101 provided by accumulated eight SOAZ layers, was placed on the X-Y stage 7118 with the cutout (B-B') of the substrate of the recording medlum 1 aligned with respect to a predetermined direction.
Subsequently f the position detecting probe electrode 7102 was moved to a position spaced by about 1 mm 10 inwardly of the substrate from the side edge B-B' and, while applying a probe voltage of 0.6 V to between the position detecting probe electrode and the Si substrate 7105 and after temporally aligning the X
direction of the X-Y fine-motion mechanism (7110, 7111) so as to be substantially parallel to the direction B-B', X-direction scan was made through a length of 1 micron. Thereafter, Y-direction scan (perpendicular to the direction X) was made through a length of 1 micronO At that time, the measurement of 20 the surface state was repaated while variously changing the manner of determination of the X-Y
coordinate axis~ and adjustment was made so that the pitches of the obtained Si atomic arrangement became closest to 6.65 angstroms and 3.84 angstroms, 25 respectively. By such adjustment, the X axis of the X-Y fine-motion mechanism can be made coincident with the direction 1211] of the Si substrate, while the Y

.
..
-. ; :

axis thereof can be made coincident with the direction -1011]. At the same time, adjustment was made so that the X and Y directions of the rough-motion mechanism were brought ln accord with the X and Y directions of the adjusted fine-motion mechanism, within the range of control errors of the rough-mo~ion mechanism.
Subsequently, with respect to the X and Y directions, the rough-motion mechanism was used to scan the position detecting probe electrode and the position of 10 the reference origin ~rough) 7201 was detected. ~hen, at a position spaced in the Y-axis direction by 2 mm from the center of the reference origin (rough) 7201 to the center of the substrate, and by using the fine-motion mechanism, the Si lattice point was detected~
15 The position of that lattice point (point C in Figure 53) w~s taken as a coordinate axis origin 7301, and the position detecting probe electrode 7103 was scanned in the X direction (direction [211]3. By checking each lattice point with respect to the 20 direction 1211] of Si at that time, confirmation was made with regard to the direction control correction and the position coordinate (lattics pitch). In Figure 53, each point denoted at 7302 (i.e. each lattice point) is illustrated as a representative of 25 ten lattice points. It will be understood that, during the operation described above, the recording/reproducing pro~e electrode 7102 moves over - . -the recording layer 7101 surface, with the movement of the position detecting probe electrode 7103.
In the present example, the distance between the two probe electrodes was 3 mm in the Y-axis 5 direction. While the recording/reproducing probe electrode 7102 was used to effect desired information recording, prior to the actual recording, a reference origin (fine) 7303 was formed at a recording position (point C' in Figure 53) corresponding to the position 10 coordinate axis origin 7301.
Such a reference origin (fine) can be formed by utilizing the electric memory ~ffect of the recording layer 7101. Namely, a probe voltage of 1.0 V was applied to between the recording/reproducing 15 electrode 7102 and the Au electrode 7104 and, by using the Z-axis direction fine-motion control mechanism 7107, the distance Z between the recording/reproducing probe electrode 7102 and the recording layer 7101 surface was adjusted so that the probe curren~ Ip 20 became equal to 10-9 A. 5ubsequently, while placing the recording/reproducing probe electrode 7102 on a positive side and placing the Au electrode 7104 on a negative side, a rectangular pulse voltage (18 V, 0.1 microsecond) greater than a threshold voltage ~th ON, 25 from which the electric memory material (SOAZ, eight-layer LB film) changes into a low resistance state (ON
state), was applied to create the "ON" state. While : . ~

~328~3~

retaining the distance Z between the recording/reproducing probe electrode 7102 and the recording layer 7101, a probe voltage of 1.0 V was applied between the recording/reproducing probe 5 electrode 7102 and the Au electrode 7104. The probe current Ip was measured and it wa~ confirmed that there was an electric current of about Q.5 mA and that the "ON" state was establish~d. By the ~oregoing operation, the reference origin (fine) 7303 was 10 determined~ In that case, the region on the recording layer of a size of 10 nm square was rendered into the "ON" state, and that was done to prevent that the origin positional information concerning the reference origin (fine) 7303 and record information, to be 15 recorded later, were reproduced mixedly (Figure 53).
However, the shape of the reference origin (fine) 7303 . is not lim~ted to that of the present example.
Subsequently, the position detecting probe electrode 7103 was scanned in the direction [211]
20 while checking the lattice point and, at each 15-pitch point (9.98 nm), the recording was made by using the recording/reproducing probe electrode 7103 which was moving sirnultaneously therewith. Thus, the pitch of . the recording points 7304 was 9.98 nm (Figure 53).
25 The recording was effected in accordance with a method similar to the formation of the reference origin (fine) 7303 and by defining on the recording layer .
. .
' ~ ., ,~

-107- 1~28131 (SOAZ, eight-layer LB film) 7101 the "ON" state and the "OFF" state (high-resistance state before the recording).
The recording medium having a record formed thereon by the above-described process, was once demounted from the recording/reproducing apparatus and, thereafter, was placed again on the X-Y stage 7118, and reproduction experimants were made. First, like the recording, the X and Y directions of the 10 position control system were aligned, using the Si lattice with the direction [211] and the direction [011], respectively. Thereafter, with respect to the X and Y directions, the rough-motion mechanism was used to scan the position detecting probe electrode 7103, and the position of the reference origin (rough) 7201 was detected. On the basis of the detected reference origin (rough) 7201, the rough-motion and fine-motion mechanisms were used to scan the recording/reproducing probe electrode 7102, and the 20 position of the reference origin (fine) 7303 was de~ected.
It was confirmed that, at the same time, the position detecting probe alectrode 7103 was on the Si lattice point ~position coordinate axis origin 7301).
25 If there was deviation, the fine-motion mechanism could be used to correct the X-Y coordinate system to bring the position detecting probe electrode 7103 and ~. '.' ' ~ , . ~ :

the Si lattice point into agreement with each other.
Subsequently, a probe voltage of 0.6 V was applied to between the position detecting prove electrode 7103 and the Au electrode 7104 and, while detecting the positions of the S.t lattice points, the position detecting probe was scanned in the direction [211] ~X-axis direction). At that time, a probe voltage of 1.0 V was applied to between the Au electrode 7104 and the recording/reproducing probe ~.lectrode 7102 which was 10 being moved simultaneously therewith, and the reproduction of the recorded information was effected by (i) directly reading the changes in the probe current based on the "ON" state or the "OFF" state at each recording point orl alternatively, by (ii) 15 reading through the servo circuit 7109 the changes in the distance Z between the recording/reproducing probe electrode 7102 and the recording layer 7101 surface, at the time of scan of the recording/reproducing probe electrode 7102 made so as to maintain a constant probe 20 current Ip.
Aacording to the described reproduction experiments, the bit error rate was 2x10-6.
Additionally~ a probe voltage of 10 V, greater than a threshold voltage Vth OFF ~rom which 25 the electric memory material changes from the "ON"
state to the "OFF" state was set and the recording positions were traced again. As a result, it was :. , : , confirmed that all the records were erased and the transition to the "OFF" state had occurred.

Example 2 In place of the SOAZ four-layer LB film used in Example 1, CuTCNQF~ was used to provide a recording layer 7101. Recording and reproducing experiments were made in a similar manner as in Example 1.
As for the electric voltage to be applied for the recording, a rectangular pulse of 2 Vmax. and 10 ns was used, and the electric voltage to be applied for the reproduction was 0.1 V. Also, as for the electric voltage to be applied for the erasing, a 15 rectangular pulse of 5 Vmax. and 100 ns was used. The recording pitch was 99.8 nm (150 Si lattice point pitch), and the size of the reference origin (fine) 7303 was changed to 0.1-micron square. According to the results of reproduction experiments, the bit error 20 rate was 1x10-9.
The method of forming a CuTCNQF4 recording ; layer 7101 will be explained. On the Au electrode ~104, Cu and TCNQF4 were co-deposited by vacuum evaporation method, and a Cu~TCNQF4 layer was 25 deposited to a thickness of 2000 angstroms (substrate temperature: room temperature). At that time, an electric current which was preset to provide ` . ': ' ' - :

deposition speeds of about Cu: 5 angstrom/s and TCNQF4: 20 angstrom/s, was applied and heating was made. It was confirmed that, as a result, a blue film was deposiked due to the production of CuTCNQF4~

Example 3 In Example 1, the probe electrodes 710~ and 7103 were scanned not only in the X direction (direction [211]) but also in the Y direction (direction l011]), and the recording was made two-di.mensionally. At that time, the recording pitch in the Y direction was 11.5 nm (for each thirty Si lattice points). Similar reproduction experiments were made, and the bit error rate was 3x10-6.

Example 4 In Example 1, the substrate 7105 was replaced by a Ga As wafer and the recording layer 7101 was replaced by an eight-layer LB film of t-butyl 20 substitution product of silicon chloride-phthalocyanine (PcSiCl2). Except for those points, the present example is substantially the same as Example 1 and, recording/reproducing experiments were made. Description will be m~de to the differences in 25 comparison with Example 1~ The structure of the recording medium 1 was such as shown in Figure 52. As for a substrate, a P type Ga-As wafer (Zn dopet 0.3 mm thickness) of a diameter of 1/2 inch with its plane (110) exposed, was used. The direction B-B' of the cutout of the substrate is substantially parallel to the direction ~001] of the Ga-As crystal.
6 From the middle point of the side B-B' to the center of the substrate, etching was made to a size of 1-micron square and a depth of 0.2 micron, to form a reEerence origin (rough). The manner of forming such a reference oxigin (rough) will be 10 explained below.
First, a ultraviolet ray resist ltrade name:
"AZ 1350") was applied to the Ga-As substrate surface to a thickness of 1 micron and, after prebaking, a mask corresponding to the shape shown in Figure 52 was used, and the treatments of ultraviolet-ray exposure, development and post-baking were conducted. As a consequence, a mask pattern was formed on the Ga-As substrate. Subsequently, by using a BCl3 gas, sputter etching was made under the condition of gas pressure 20 of 1 Pa and discharging voltage of 100 W, for three minutes. The resultant etching depth was 0.2 micron.
~he AZ 1350 material of the mask was removed by acetone cleaning.
Similarly to Example 1, a Cr/Au base 2~ electrode 710~ was formed on the thus produced substrate and, thereafter, on that base electrode 7104 an eight-layer LB film of t-butyl substitution product .. : ~ , , . ' -. ; ' . . :

112 ~l328131 of silicon chloride-phthalocyanine (PcSiC12) was built to provide a recording layer 7101. The recording layer forming conditions were as follows:

Solvent: CH3CCl3 Conc~ of solution: 1 mg/ml Aqueous phase: pH 8.2 (pure water adjusted by NaOH) Water surfacP pressure: 25 mN/m Vertical moving speed of substrate: 5 mm/min.
(Z type accumulation) Using the recording medium 7001 formed in the described manner, recording/reproducing experiments were made. First, with regard to the setting o~ the X-Y coordinate system, adjustment was made so that the X-axis and the Y-axis became in accord with the direction [001~ and the direction [110] of the Ga-As crystal, respectively. Since the 20 pitch on the coordinate axis at that time was based on the Ga-Ga interatomic distance, it was 5.65 angstroms with respect to the direction ~001] and 400 angstroms with respect to the direction [110]. The recording was made at 10.17 nm pitch (each eighteen Ga lattice 25 points) in the X direction and at 12 nm pitch (each thirty Ga lattice points) in the Y direction.
Reproductlon experiments were made in the same manner as in Example 1, and the bit error rate was 3x10-S.

Example 5 In Example 2, the recording layar 7101 was replaced by a film of amorphous semiconductor, represented by an atomlc composition ratio of Si16Ge14As5Te65, made by well-known vacuurn evaporation method to a thickness of 2000 angstroms. In a similar manner as in Example 2, recording/reproducing 10 experiments were made. However, the recording voltage, the reproducing voltage and the erasing voltage were changed to a rectangular pulse of 20 Vmax., 0.1 microsecond; a voltage of 1.0 V; and a rectangular pulse of 50 Vmax., 10 microseconds, respectively).
The recording pitch was 96.05 nm (each one hundred and sixty Ga lattice points) in the X
direction and 120 nm ~each three hundreds Ga lattice points) in the Y direction. The size of the reference 20 origin ~fine) was changed to 0.1-micron square. As a result of the reproduction experiments, the bit error rate was 1x10-9.

Figure 54 is a schematic and diagrammatic 25 view of a recording apparatus according to a still further embodiment of the present invention. In Figure 54, denoted at 8105 is a probe current .
: ' " .' ~, , ''' '" ' ~''," ;

' ~, ' .

amplifier, and, at 8106, is a servo circuit for controlling a fine-motion mechanism 8107 using a piezoelectric device so as to maintain the probe current constant. Denoted at 8108 is an electric power source for applying a recording/erasing pulse voltage to between a probe electrode 8102 and an electrode 8103.
Since the probe current changes radically when the pulse voltage is applied, the servo circuit 10 8109 controls to turn a hold circuit on so that the output voltage is maintained constant during that period.
Denoted at 8109 is an X-Y scan drive circuit for controllably moving the prove electrode 8102 in the X and Y directions. Rough-motion drive mechanisms 8110 and 8111 are used to preparatorily and roughly control the distance between the recording medium 8001 and the probe electrode 8102 so that a probe current of an order of 10-9 A is produced, or to allow a large 20 relative displacement between the probe electrode and the substrate in the X and Y directions (beyond the range of the fine-motion control mechanism).
These components are controlled by centralized control by use of a microcomputer 8112.
25 Reference numeral 8113 denotes a display device.
Mechanical performance in the movement control by use of piezoelectric devices ls as follows:

Z-axis fine-motion controllable range.
0.1 nm 1 mlcron Z-axis rough-motion controllable range:
10 nm - 10 mm X-Y scan range: 0.1 nm - 1 micron X-Y rough-motion controllable range:
10 nm - 10 mm Tolerance for measurement and control: -< 0.1 nm (fine-motion control~
Tolerance for measurement and control:
~ 1 nm (rough-motion control) Details of the manner of recording/reproduc-15 ing in the present embodiment will be explained below, particularly with reference to some examples thereof.

Example 6 As the recording/reproducing apparatus, that 20 shown in Figure 54 was used and, as the probe electrode 8102, a probe electrode made of tungsten was used. The probe electrode 8102 is provided for controlling the distance Z to the surface of the recording medium 8001, and the distance Z of the probe 25 electrode can be controlled by fine motion, such that the electric current is maintained constantO Further, the fine-motion controlling mechanism is arranged to . .

-116- 132813~

effect fine-motion control also in the direction within a plane ~X-Y), while retaining the distance Z
constant.
The probe electrode 8102 is usable for detecting the position in the relative direction within the surface of the recording medium as well as for the recording, reproducing and erasing. Also, the recording medium 8001 is placed on a high-precision X-Y stage 8114 and can be moved to a desired position (X-Y rough-motion mechanism). The X and Y directions of the rough-motion mechanism and the X-Y directions of the fine-motion mechanism are in accord with each other within the range of error resulting from the difference in precision between these movement control mechanisms.
Details of the recording medium usable in the present embodiment will be described. The structure of such a recording medium is shown in Figures 55A and 55B. Figure 55A is a plan view of a 20 recording medium used in the present embodiment, and Figure 55B is a section taken on a line A-A'~
P-type Si wafer (B-dope 0.3 mm thickness) having a diameter of 1t2 inch with a plane ~111) being exposed, was used as a substrate 8104. The substrate 25 is cut along a line B-B' for the purpose of attaining substantially constant orlentation thereof when it is placed on the X-Y stage 8114 of the -~, : . .

-117- 1328~31 recording/reproducing apparatus. The direction B-B' is substantially parallel to the direction [211] of the Si crystal. Subsequently, at the position at a distance ~f 1 mm from the middle point of the side B-B' to the center of the base, a portion of the basewas etched by a range of 1-micron square and 0.2-micron depth, to form a reference origin (rough). The method of forming such a reference origin (rough) is substantially the same as has been described with 10 reference to Example 1.
The manner of forming the base electrode 8103 and the recording layer 8101 i5 essentially the same as having been described with reference to Figure 52.
By using the recording medium 8001 formed in the described manner, recording/reproducing experiments were made.
Details are as follows:
The recording medium 8001 having a recording 20 layer 8101 provided by accumulated eight SOAZ layers, was placed on the X-Y stage 8114 with the cutout (B-B'3 of the substrate of the recording medium 1 aligned with respect to a predetermined direction. Sub-sequantly, the probe electrode 8102 was moved to a 25 position spaced by about 1 mm inwardly of the substrate from the side edge B-B' and, while applying a probe voltage of 0.6 V to between the probe .;:

': ' ` ~ - ''' : .

-118- 1328~3t electrode and the Si substrate 8104 and after temporally aligning the X direction of the X-Y fine-motion mechanism (8107, 8109) so as to be substantial-ly parallel to the direction B-B', X-direction scan was made through a length of 1 micron. Thereafter, Y-direction scan (perpendicular to the direction X) was made through a length of 1 micron. At that time, the measurement of the surface state was repeated while variously changing the manner of determination of the 10 X-Y coordinate axis, and adjustment was made so that the pitches of the obtained Si atomic arrangement became closest to 6O65 angstroms and 3.84 angstroms, respectively. By such adjustment, the X axis of the X-Y fine-motion mechanism can be made coincident with 15 the direction [211] of the Si substrate, while the Y
axis thereof can be made coincident with the direction [011]. At the same time, adjustment was made so that the X and Y directions of the rough-motion mechanism were brought in accord with the X and Y directions of 20 the adjusted fine-motion mechanism, within the range of control errors of the rough-motion mechanism.
Subsequently, with respect to the X and Y directions, the rough-motion mechanism was used to scan the probe electrode and the position of the reference origin (rough) 8201 was detected.
A reference origin (fine) 8202 was formed at a position spaced by 2 mm in the Y-axis direction from .
` ' ' ;' ,. : ,' ;

-1 1 9 - ~ 3 2 8 1 3 1 the center of the reference origin (rough~ to the center of the substrate.
Such a reference origin ~fine) can be formed by utilizing the electric memory effect of the recording layer 8101. Namely, a probe voltage of 1.0 V was applied to between the electrode 8102 and the Au electrode 8103 and, by using the Z-axis direction fine-motion contro1 mechanism 8107, the distance Z
between the probe electrode 8102 and the recording 10 layer 8101 surface was adjusted so that the probe current Ip became equal to 10~9 A. Subsequently, while placing the probe electrode 8102 on a positive side and placing the Au electrode on a negative side, a rectangular pulse voltage (18 V, 0.1 microsecond) 15 greater than a threshold voltage Vth ON, from which the el`ectric memory material (SO~Z, eight-layer LB
film) changes into a low resistance state ~ON state~, was applied to create the "ON" state. While retaining the distance Z between the probe electrode 8102 and 20 the recording layer 8101, a probe voltage of 1.0 V was applied between the probe electrode 8102 and the Au electrode 8103. The probe current Ip was measured and it was confirmed that there was an electric current of about 0.5 mA and that the "ON" state was established.
~5 By the foregoing operation, the reference origin (fine~ 8202 was determinedO In that case, the region on the recording layer of a size of 10 nm square was :. - ~ . . . .

, .

-120- 132813~

rendered into the "ON" state, and that was done to prevent that the origin positional information concerning the reference origin (fine) 8202 and record information, to be recorded later, were reproduced mixedly (Figure 56). However, the shape of the reference origin (fine) 8202 is not limited to that of the present example.
Subsequently, while using such reference origin (fine) as an origin on ~he X-Y coordinate of the probe electrode position control system, the probe electrode 8102 was scanned by fine motion and the information recording was conducted at a pitch of 0.01 micron. The recording position per one bit on the recording surface 8101 is schematically shown in 15 Figure 56. The recording was made in a similar manner as the formation of the reference origin (fine) and by defining on the electric memory material (SOAZ four-layer LB film) an "ON" state and an "OFF" state thigh resistance state before recording).
The recording medium having a record formed thereon by the above-described process, was once demounted from the recording/reproducing apparatus and, thereafter, was placed again on the X-Y stage 8114, and reproduction experiments were made. First, 25 like the recording, the X and Y directions of the position control system were aligned, using the Si atom scale with the direction ~211] and the direction .: :

132813t [011], respectively. Thareafter, with respect to the X and Y directions, the rough-motion mechanlsm was used to scan the probe electrode and the position of the reference origin (rough) 8201 was detected. On the basis of the detected reference origin (rough), the rough-motion and fine-motion mechanisms were used to detect the reference origin (fine)~
By using such reference origin (fine) as an origin of the X-Y coordinate system, the reproduction of the recorded information was conducted. More specifically, a reproducing probe voltage of 1.0 V was applied to between the probe electrode 8102 and the Au electrode 8103, and the detection of the position of the reference origin (fine) 8202 as well as the reproduction of the recorded information were effected by (i~ directly reading the change in the probe current flowing in the "ON" state region and the "OFF"
statè region or, alternatively, by ~ii) reading through the servo circuit 8106 the changes in the 20 distance Z between the probe electrode 8102 and the recording layer 8101 surface, at the time of scan of the probe electrode 8102 made to maintain a constant probe current Ip. In the reproduction experiments described above, the bit error rate was 5x10-6.
Additionally, a probe voltage of 10 V, greater than a threshold voltage Vth OFF from which the electric memory material changes from the "ON"

"

state to the "OFF" state was set and the recording positions were traced again. As a result, it was confirmed that all the records were erased and the transition to the "OFF" state had occurred.

ComParing Example 1 In the reproduction experiments in Example 6, the setting of the X-Y coordinate system of the probe electrode scan mechanism using the atom scale as 10 well as the setting of the position coordinate origin based on the detection of reference origins (rough and fine) were omitted. In that case, it was very difficult to find out the record writing region and, further, the reproduction was substantially impossible.

Example 7 Description will be made of an Exampls wherein the X-Y coordinate system of a probe electrode scan system is set by use of a reference-scale, using a plurality of referance origîns, to execute the recording/reproducing.
The structure of a recording medium 8001 usable in this example is schematically shown in 25 Figures 57A and 57B. In this example, as a substrate 8104, an optically ground glass plate ~1 mm thickness) of a size 0~7 x 1.5 cm was used. Then, at a position ~ ' :

' :"

-123- 1 328~31 spaced by 1 mm from the middle point on the side B-B' toward the center of substrate, a reference origin trough) 8202 having a size of 1-micron square and a depth of 0.1 micron was formed.
The method of forming such a reference origin (rough~ will be explained below.
By using a known photoresist process, a resist material ~trade name: "AZ 1350") was applied to the substrate to a thickness of 1 micron. After the o prebaking, and by using a mask havlng a shape shown in Figure 56, ultraviolet ray exposure, development and post-baking were conducted and a mask pattern was ~ormed on the glass plate. Subsequently, based on a known CF4 gas plasma etching process, dry etching was made to the glass surface to a depth of 0.1 micron under the condition of etching power of S0 W, thus pressure of 1 Pa and gas flow rate of 15 SCCM. The AZ 1350 material of the mask was removed by acetone cleaning.
Then, the substrate was left in a saturated vapor of hexamethyldisilazane and hydrophobic treatment was made to the surface thereof. Then, onto the substrate, Cr was deposited to a thickness of 50 angstroms by vacuum evaporation method to provide a 25 hackground layer. Further, by the same method, Au was deposited to a thickness of 400 angstroms to provide a base electrode 8103. Then, on that Au electrode, a :. , : : :
. ' ~; ;::

-124- 132~131 ten-layer LB film of t-butyl substitution product of luthetium diphthalocyanine [LuH(Pc)2) is laminated, such that a recording layer 8101 was formed. That was done in the manner that the recording layer 8101 was not deposited in the neighborhood of the reference origin (rough) 8201.
The conditions for the formation of the t-butyl substitution LB film of LuH(Pc)2 were as follows: -Solvent: chloroform/trimethylbenzene/acetone = 1/1/2 (V/V) Concantration: 0.5 mg/ml Aqueous phase: pure water of temperature 20 C
Water surface pressure: 20 mN/m Vertical moving speed of substrate~ 3 mm/min.

By using the recording medium 8001 formed in the described manner and by using the 20 recording/reproducing apparatus described with reference to Example 6, recording/reproducing experiments were made. Details are as follows:
The recording medium 8001, having a recording layer 8101 formed by the accumulation of 25 ten-layer LB film of t-butyl substitution of LuH(Pc)2, was placed on the X-Y stage 8114 with the direction B-B' of the recording medium aligned with respect to the -:

`': ` ' ~ ' ' '~ 7 ~ ' , ' ~ .
. ': ' . ' ~

-125~ 813~

X-axis direction of the X-Y s~age 8114. Subsequently r as in Example 6, with respect to the X and Y
directions, the rough-motion mechanism 8110 was used to scan the probe electrode 8102 to thereby detect the position of the reference origin ~rough) 8201. The probe voltage was 0.1 V. At the position ~on the recording layer 8101) spaced by 2 mm in the Y-axis direction from the center of the reference origin (rough) 8201 to the center o the substrate, a first reference origin ~fine) 8401 was formed by a similar manner as the formation of the reference origin (fine) in Example 6. At that time, the X and Y directions of the rough-motion mechanism and the X and Y directions of the fine-motion mechanism had been in accord with each other within the range of control error of the rough-motion mechanism. Then, by using the fine-motion mechanism, a second reference origin (fine) 8402 was formed at a position spaced by 1 micron in the Y-axis direction from the first reference origin (fine) 8401. Such a second reference origin (fine3 8402 can be formed in a slmilar manner as of the first reference origin (fine). For distinguishability, each origin may have a different shape. However, this is not-always necessary. What is required is only that these origins are not confused with the other, ordinary record information~ By using one of the first and second reference origins (fine~ 8401 and : ' . ' ~ . ' ~,;, ... . ~ ..
, -126~ ~328~3~

8402 as an origin of the X-Y coordinate sys~em and by using the fine-motion mechanism, the recording of information was made at a pitch of 0.01 micron. The recording was made in the same manner as in Example 6.
The recording medium 8001 on which ~ecords had been formed in the described process was temporally demounted from the recording/reproducing apparatus, and thereafter, it was placed again on the X-Y stage 8114, and reproduction experiments were made. First, like the xecording, the probe electrode was scanned in ths X and Y directions by the rough-motion mechanism to detect the reference origin (rough) 8201 and, on the basis of the detected reference origin (rough) and by use of the rough-motion and fine-motion mechanisms, the first reference origin (fine) 8401 was found. Then, by using the fine-motion mechanism, the second reference origin (finè) ~402 was detected and, thereafter, the X-Y
coordinate system was adjusted (re-set) so that the direction of a line segment connecting the first and second reference origins (fine) and the Y-axis direction of the probe electrode scan system came into accord with each other. The setting at that time was done so that the first reference origin (fine) B401 provided an origin of such X-Y coordinate system. The reproduction of recorded information was made by using such position coordinate system as a reference, and : -, -- :

-127- 1328~31 the bit error rate was 3x10-6.
Reproduction was tried without the re-setting of the X-Y coordinate system using the described reference origins, but the reproduction was not possible.

Example 8 Another example wherein plural reference points are provided outside the range of fine-motion scan, ln order to increase the recording capacity, and wherein the recording/reproducing is effected by using each point as a reference, will now be described. As a recording/reproducing apparatus, that the same as used in Example 6 (Figure 54) was used.
15Recording medium 8001 was manufactured in the same manner as in Example 6, and a reference origln (fine) 8202 was formed in a similar manner.
Subsequently, by using the fine-motion mechanism, the probe electrode was moved to a position spaced by 5 20 mi rons along the Y axis from the reference origin (fine~ 8202 to the center of the substrate, and a reference point (fine) A 8501 was formed in a similar manner as the case of the reference origin (fine) 8202. Further, in a similar manner, another reference origin (fine) B 8502 was formed at a position spaced by 5 microns along the Y axis from the reference point (fine~ A to the center of the substrate. By using : - .
; ' . .

each of these points as a reference and by us~ of the fine-motion control mechanism, the recording was made in each of a recording reglon 8503 (O), a recording region 8504 (A) and a recording region 8505 (B) (see Figure 58).
Subsequently, the recording medium 8001 was temporally demounted from the recording/reproducing apparatus, and after placing it again, the reproduction of the recorded information was effected in a similar manner as in Example 6. In each of the recording regions 8503 (O), 8504 (A) and 8505 (B), the bit error rate was 5x10-6 Comparinq Example 2 Like Example 8, a recording medium 8001 was manufactured and the recording was made. In this example, the reference point (~ine) A 8501 and the reference point (fine3 B 8502 were not formed, and the ' recording was made at positions corresponding to the recording region A 8504 and the recording region B
8505. To such recording medium 8001, having records thereon, reproduction experiments were made in the same manner as in Example 8~ According to the results, the bit error rate in the recording region O
25 8503 was 5X10-6, and no change occurred as compared with Example 8. However, with regard to the recording region A 8504 and the r~cording region B 8505, it was ~`

.... .. .
.

:

very difficul-t to detect their positions. Also, the bit error rate was in the range of 2x10 3 to 5x10-4, and there was reduction in precision by not less than two figures, as compared with Example 8.

Example 9 In Example 6, the substrate 8104 was replaced by a Ga-As wafer and the recording layer 8101 was replaced by an eight-layer LB film of t-butyl 1C substitution product of silicon chloride-phthalocyanine (PcSniCl2). Exc~pt for those points, the present example is substantially the same as Example 6 and, recording/reproducing experiments were made. Description will be made to the differences in comparison with Example 6. The structure of the recording medium 1 was such as shown in Figure 55. As for a substrate, a P type Ga-As wa~er (Zn dope, 0.3 mm thickness~ of a diameter of 1/2 inch with its plane (110~ exposed, was used. The directlon B-B' of the cutout of the substrate is substantially parallel to the dlrection [001] of the Ga-As crystal.
From the middle point of the side B-B' to the center of the substrate, etchlng was made to a size of 1-micron square and a depth of 0.2 micron, to 25 form a reference origin (rough3. The manner of forming such a reference origin (rough) will be explained below.

' --130_ 1328131 First, a ultraviolet ray resist (trade name:
"AZ 1350") was applied to the Ga-As substrate surface to a thickness of 1 micron and, after prebaking, a mask corresponding to the shape shown in Figure 55 was used, and the treatments of ultraviolet-ray exposure r development and post-baking were conducted. As a consequence, a mask pattern was formed on the Ga-As ~ubstrate. Subseguently, by using a BC13 gas, sputter etching was made under the condition of gas pressure of 1 Pa and discharging voltage of 100 W, for three minutes. The resultant etching depth was 0.2 micron.
The AZ 1350 material of the mask was removed by acetone cleaning.
Similarly to Example 6, a Cr/Au base electrode 8103 was formed on the thus produced substrate and, thereafter, on that base electrode 8103 an eight-layer LB film of t-butyl substitution silicon chloride-phthalocyanine (PcSiC12) was built to provide a recording layer 8101O The recording layer forming 20 conditions were as follows:
Solvent: CH3CC13 Conc. of solution: 1 mg/ml Aqueous phase: pH 8.2 (pure water adjusted by NaOH) Water surface pressure: 25 mN/m Vertical moviny speed of substrate: 5 mm/min.
(Z type accumulation) ;' - ~ .

By using the recording medium 8001 manufactured in the described manner, the recording and reproducing experiments were made. The bit error rate at the time of reproduction was 5x10-6.
It is added that, with regard to the setting o~ the X-Y coordinate axis using atom scale, the adjustment was made so that the X axis and the Y axis were brought into accord with the direction [001] and the diraction [110l of the Ga-As crystal, 10 respectlvely. The Ga-Ga interatomic pitch at that time was 5.65 angstroms with respect to the direction [001] and 4.00 angstroms with respect to the directlon [ 1 1 0 ] .

15 Example 10 In Example 7, the recording layer 8101 was replaced by a film of amorphous semiconductor, represented by an atomic composition ratio of Si16Ge14As5Te65, made by well-known vacuum evaporation 20 method to a thickness of 2000 angstroms. In a similar manner as in Example 7 r recording/reproducing experiments were made. However, the recording voltage, the reproducing voltage and the erasing voltage were changed to a rectangular pulse of 20 25 Vmax., 0~1 microsecond; a voltage of 1.0 V; and a rectangular pulse of 50 Vmax., 10 microseconds respectively).

' ` , ' .

.~ . -The recording pitch was 0.1 micron, and the size of each of the first and second reference origins (fine) was changed to 0.1 micron square. According to the results of reproduction experiments, the bit error rate was 1x10-9.

Example 11 In place of the SOAZ four-layer LB film used in Example 6, CuTCNQF4 was used to provide a recording 10 layer 8101. Recording and reproducing experiments were made in a similar manner as in Example 6.
As for the electric voltage to be applied for the recording, a rectangular pulse of 2 Vmax. and 10 ns was used, and the electric voltage to be applied 15 for the reproduction was 0.1 VO Also, as for the electric voltage to be applied for the erasing, a rectangular pulse of 5 Vmax. and 100 ns was used. The recording pitch was 0.1 micron, and the size of thP
reference origin (fine3 8201 was changed to 0.1-micron 20 square. According to the results of reproduction experiments, the bit error rate was 1x10-9.
The method of forming a CuTCNQF4 recording layer 8101 will be explained. On the Au electrode 8103, Cu and TCNQF4 were co-deposited by vacuum 25 evaporation method, and a Cu~TCNQF~ layer was deposited to a thickness of 2000 angstroms (substrate temperature: room temperature~. At that time, an -133- 132813~, electric current which was preset to provide deposition speeds of about Cu: 5 angstrom/s and TCNQF4: 20 angstrom/s, was applied and heating was made. It was conflrmed that, as a result, a blue film was deposited due to the production of CuTCNQF4.

Example 12 In this example, the base electrode (Cr-undercoated Au electrode) 8103 used in Example 10 10 (wherein plural reference origins and an amorphous semiconductor recording layer 8101 are used) is replaced by a Cr material of a thickness 500 angstroms and the recording layer 8101 is replaced by an amorphous silicon layer of the structure P~-layer/n-15 layer/i-layer. Except for these points, the presant example is the same as Example 10. Thus, in the same manner, the recording and reproducing experiments were made. According to the results of reproduction experiments, the bit error rate was 1x10~9.
During the experiments, the following ,t voltages were applied for the recordingl reproducing and erasing.
( Recording: 20 V
Reproducing: 0.5 V

Erasing: -5 V

' The method of forming the recording layer 8101 will now be explained. First, Cr was vacuu~-deposited to a film thickness of 500 angstroms to form an electrode. Then, by using a glow discharge method, a P~-type amorphous silicon film of a thickness 1000 angstroms was formed. The forming conditions at that time were as follows:

Introduced gas: B2H6/SiH4 (NBH/NSiH = 10 (diluted to 0.025 mol % by H2 gas~
rf Power: 0.01 W/cm2 Pressure: 0.5 Torr Substrate temp~: 300 C
Deposition speed: 30 angstroms/min.

Subsequently, excessive source material gas was exhausted, and, thereafter r a new source material gas was supplied and n type amorphous silicon was 20 deposited to a thickness of 5000 angstroms. The formation conditions were as ollows:

Introduced ga5: PH3/SiH4 (NPH/NSiH = 5x10 ~diluted to 0.05 mol % by H2 gas) rf Power: 0.01 W/cm2 Pressure: 0.5 Torr Substrate temp.: 300 C
Deposition speed: 40 angstroms/min.

After exhausting the source material gas, SiH~ gas diluted to 0.05 mol % by H2 gas was introduced into a chamber and, while retaining the other conditions, i-phase amorphous silicon was deposited to a thickness of 1000 angstroms.
Although various recording medium forming 10 methods have been described with reference to the foregoing embodiments and examples, the formation of a recording medium is not limited to the methods of the disclosed embodiments and examples. Any film forming method may be used provided that a very uniform film can be ormeda Further, it is to be understood that the present invention does not limit the material, shape and surface structure of a substrate.
` Figure 59 is a schematic view of a recording/reproducing apparatus according to a further 20 embodiment of the present invention, in which two probe electrodes are separately provided or the position detection and for the recording/reproducing.
Denoted in Figure 59 at 8601 and 8602 are the recording/reproducing probe electrode and the 25 position detecting probe electrode, each being made by electropolishing a tungsten wire of a dlameter 1 mm.

The position detecting probe electrode 8602 is used -136- 132~3~

~or detecting the atomic arrangement of a substrate 8603 which functions as a position coordinate. On the other hand, the recording/reproducing probe electrode 8601 is held at a ~ixed position with respect to the position detecting probe electrode 8602 and is made movable as a unit therewith. The recording/reproduc-ing probe electrode is used for the recording, reproducing and erasing on a recording medium 8604.
Denoted at 8605 P and 8605 R are X-Y-Z axis fine- -10 motion mechanisms, each having a tube type fine-motion element (tube scanner), ~or controlling the distance Z
between the recording medium 8604 and the probe electrode 8601 or 8602. These ~ine-motion mechanisms are adapted to independently control the positions of 5 the probe electrodes 8601 and 8602 so that an average interval therebetween can be maintained constant.
Further, t'nese fine-motion mechanisms are arranged to effect fine position control in a plane (X and Y~
while maintaining the distance Z constant.
The recording medium 8604 is placed on an X-Y rough-motion mechanism 9606 which includes a high-precision resilient hinge mechanism and which uses parallel springs, so that the recording medium can be moved to a desired position. Denoted at 8607 is a 25 super rough-motion mechanism mainly comprising a multilayered-structure type piezoelectric device, for posing relative displacement in the Z-axis direction beyond the range of fine motion. These components and mechanisms are controlled under the influence of a control system 8608. Also, these components and . mechanisms are placed on an anti-vibration table 8609.
The mechanical performance of the present embodiment in the movement control is such as follows:
Z-axis fine-motion control range:
0.1 nm - 1 micron Z-axis rough-motion control range: -10 nm - 10 mm X-Y fine-motion scan range:
not less than 0.01 nm X-Y rough-motion control range:
10 nm - 10 mm Figure 60 is a block diagram of the control system 8608 used in the present embodiment.
The position of the probe electrode 8601 (8602) is controlled and maintained constant by detecting through a comparator 8620 an electric 20 current (probe current) flowing between the probe electrode and the recording medium (or the atoms of the substrate for the position detection); by comparing the detected current with a set level (10 9 A); and by controlling the Z-axis position of the 25 probe electrode through probe Z-axis fine-motion mechanisms 8605 Pz and 8605 Rz so that the probe current beco~es approximately equal to the level 10-9 -138- ~ ~2~13~

A. Upon recording and erasing, the probe current radically changes simultaneously with the application of a writing or erasing pulse voltage. Therefore, during such perlod, a hold circuit 8611 1s turned on, which circuit is operable to maintain constant the output voltage to the Z-axis fine-motion mechanism 865 PZ for the probe 8602. The position of the position detecting probe electrode 8602 is adjusted in the X
direction by means of an X-axis fine-motion mechanism 10 8605 PX, while it is oscillated minutely at a frequency of a few kHz in the Y direction by means of a Y-axis fine-motion mechanism 8605 Py. Probe current signal S83 which is obtainable at that time is subjected to phase-lock detection with a scan signal 15 S81 of the Y-axis fine-motion mechanism 9605 Py which is obtained from a central processing unit (hereinafter "CPU") 8616 and through a counter 8612 and a`dlgital-to-analog converter (hereinafter "D/A
converter"~ 8613. By this, a signal S82 corresponding : 20 to the deviation of the probe electrode in the Y
direction is produced as an output of a phase-lock detection circuit 8614. While adding the thus produced output to a Y-axis fine-motion mechanism scan signal S81, the rough-motion mechanism 8615 and the X-25 axis fine-motion mechanism 8605 Px cooperate to effect the scan in the X-axis direction, too. By this, the movement of the position detecting probe 8602 in the X

:`

: `

139 ~3281~

direction can be made along the atomic arrangement~
While such Y-axis oscillation and the scan of the probe electrode relative to the atomic arrangement cause an oscillation component of high frequency in the probe current, such component can be intercepted by a low pass filter 8618. Therefore, the Z-axis fine-motion mechanisms 8605 Pz and 8605 Rz do not move the probe, while following the oscillation component of high frequency. As a consequence, the 10 movement of the recording/reproducing probe electrode 8601, which i5 fixed in a predetermined relation with the position detecting probe electrode 8602 by means of the X-Y fine-motion mechanisms 8605 Rx and 8605 Ry and being movable together with the position detecting probe electrode, can be made accordingly.
The signal from the position detecting probe electrode 8602 is transmitted to the CPU 8610 by way of a low-pass filter 8618A (which is adapted to transmit such frequency component of the probe 20 current that causes as a result of the X-direction scan of the probe electrode relative to the atomic arrangement but which is adapted to intercept such component of the probe current as having higher frequency) and the A/D converter 8617. From this 25 signal and by counting the number of atoms scanned, the position of the probe electrode can be detected.
The bias voltage and the ' ' . ` :

-~40- ~ 328131 recording/reproducing pulse voltage which are to be applied to between the probe electrodes and the recording medium 8604 or the position detecting medium ~atoms of the substrate) 8604A, are supplied from two 5 mutually independent D/A converters 8613.
For the rough-motion drive in the X and ~
directions, parallel springs 8615B, functioning as a rough-motion mechanism 8615, are used. Each X Y-Z
fine-motion mechanism 8605P (8605R) includes a 10 cylindrical type piezoelectric device. Such a piezoelectric device has a high mechanical resonance frequency ~not greater than 8 K~z in the X and Y
directions; not greater than 40 KHz in the Z
direction) and, therefore, a not so large scan range 15 is attainable (not greater than 500 nm). However, it is suitable for high-speed scan.
Figure 61 shows the control algorism in the CPU used in the present embodiment. The CPU 8610 operates in response to a start signal to reset the 20 elements such as, for example, the A/D converter 8612 and the counter 8613, and to set respective initial conditions. Subsequently, the position detecting probe electrode 8602 and the recordinglreproducing probe electrode 8601 are stepwisely and separately 25 moved toward the substrate and, at each stepwise movement, the level of electric current (Jt) that flows to the probe electrode ls analog-to-digital .. . .. .

1328~31 converted. Until a set level is reachedl the probe electrode 8601 (8602) is brought close to the recording medium 8604 or the substrate 8603. After this, by applying scan signals S84x and S84y to the 5 parallel spring means 8615, the probe electrode 8601 is scanned along the plane (X and Y directions) over the whole region of the recording medium. Also, together with such scan, the probe electrode 8602 is scanned over the substrate. The probe current Jt 10 during the scan is converted at suitable timing by the A/D converter 8617, and the position of a reference origin on the recording medium is calculated on the basis of the X-Y scan signal, the level of the current Jt from the probe electrode 8601 and the positional 15 information as obtained from the position detecting medium 8604A. The probe electrode is moved to the thus determined reference oxigin~
For the recording, a small region i5 scanned from the position of the r~ference origin and along 20 the plane and, in response to a trigger signal corresponding to the frequency at the time of signal sampling, first a control signal with respect to the Z-axis direction is held by a hold circuit 8611 and, thereafter, a recording pulse is outputted to effect 25 the recording. For reproduction, a similar trigger signal is used, and an elec:tric current flowing through the probe electrode is analog-digital . . .

.` ~' ' -142- ~3~813~

converted and the information is reproduced. A~ter completion of the recording or reproducing, the probe electrode is moved away from the recording medium and to its home position and is held thereat.
In the present embodiment, the scanning mechanism is separated into a rough-motion scanning portion and a fine-motion scanning portion, and for the higher frequency scan, the fine-motion scanning portion i5 used, whereas for lower-frequency scanning, 10 the rough-motion scanning portion i5 used. This makes it possible to reduce the stroke of the fine-motion scanning portion and to increase the rigidity thereof.
Therefore, it is possible to increase the scan speed.
As a result, it is possible to increase the information transfer speed. For the fine-motion mechanism, use of an actuator having a piezoelectric device is preferable in respect to the precision. Any one of prevalently used fine-motion mechanisms or devices may be used. Examples are "piezoelectric 20 tripod" (IBM J.Res. & Dev. 30 ~1986~ 355 or Helv. Phys.
Acta. 55 (1982) 726, by G. Binnig and H. Rohrer); "3-d continuous fine-positioning system" (Rev. Sci.
Instrum. 56 11985) 1573, by G.F.A. Van De Walle, J.W.
Gerritsen, H. Van Ken-pen & P. Wyder; and "tube 25 scanner" (Rev. Sci. Instrum. 57 (1986) 168, by G.
Binnig & D.P.E. Smith).
The recording medium used in the present ,~ . .

:~ :

.: :. : . ' 132813~

embodimant is similar to that in the Figure 51 embodiment. As an alternative embodiment, such as shown in Figure 62, a medium 8604B ~or the positioning may be provided on the opposite side of the recording 5 medium 8604.
Figure 63 is a schematic and diagrammatic view of a recording/reproducing apparatus according to a still further embodiment of the present inventlon.
Danoted in Figure 63 at 9001 is a probe electrode 10 which is used for the recording and reproducing. For enhanced rasolution in the recording and reproducing, the probe electrode is made by mechanical-polishing and electro-polishing the tip of a tungsten needle.
However, any other material such as Pt-Ir, Pt or 15 otherwise may be used. Also, any other machining process may be used.
Denoted at 9002 is a recording medium which, in this embod1ment, is provided by an accumulation of eight layers of squarilium-bis-6-octylazulene, having 20 a memory effect with respect to the switching characteristics of voltage and current, the layered structure being formed on a graphite substrate by use of the LB method.
Denoted at 9003 is a current amplifier for 25 converting, into an electric voltage, and for amplifying an ~lectric current flowing through the probe electrode. Denoted at 9004 is a fine-motion ::

, .

~32813~

mechanlsm which comprises a cylindrical piezoelectric device and which operates to adjust the position of the probe electrode with respect to the direction in the plane of the recording medium (i.e. X and Y
6 directions) and in a direction from the probe electrode to the recording medium (i.e. in the Z
direction). Denoted at 9005 is a rough-motion mechanism for moving the recording medium in the X and Y directions. ~his mechanism comprises parallel 10 springs mainly using resilient hinge.
Denoted at 9006 is an X-Y controller for applying an operation voltage to the control mechanisms 9004 and 9005. Denoted at 9007 is a processing unit which includes an A/D converting circuit for digitizing the output of the probe current amplifier, a microprocessor for controlling various device systems and a processor for pattern-recognition and code transformation to be made to the position reference scale on the recording medium and 20 to the recording signals.
Denoted at 9008 is a voltage applying circuit for applying recording, reproducing and erasing voltages. The voltage applying circuit is arranged to apply a desired bias voltage in the range 25 of -10 V - +10 V. Denoted at 9009 is an anti-vibration table which is effective to reduce the noise due to the vibration of the apparatus during the -14~- ~ 3 2 gl 3 recording and reproduction.
Denoted at 9010 is a super rough-motion mechanism for controlling such a region as having a size larger than the control range of the fine-motion 5 and rough-motion mechanisms. The portion denoted by a broken line 9011 is a structure made of invar, for supporting various structural components.
Figure 64 is a schematic view showing positional reference patterns and signal recording 10 regions on the recording medium used in the present embodlment. In this embodiment, position reference patterns are preparatorily recorded on the recording medium, prior to the recording of signal information.
Also, before starting the recording or reproduction of signal information, the probe electrode is scanned to discriminate the po~itional reference pattern and, the recording/reproducing is effected on the basis of the position of at least one recording/reproducing region on the recording medium obtained as a result of that 20 and the information concerning the scan direction of the probe electrode during the recording/reproducing.
Before the recordingtreproducing, four patterns 9012a - 9012d shown in Figure 64 are preparatorily recorded on the recording medium. Owing 25 to the mechanical precision of the above-described structure, the position of the probe electrode upon the recording medium is assured, with a precision of . ':
: .

. .

10 microns, at such position as spaced by about ~00 microns in the Z-axis direction from the home posit.ion on the recording medium.
As a sequence, an electric voltage of 300 mV
is applied by a voltage applying device 9008 to between the poles and, in this state, the output of the current amplifier is processed by a processing unit 9007O Based on this, the super-rough motion mechanism is controlled to reduce the interval betw~en the electrodes to such distance so close as to allow the flow of an electric current of 1 pA.
Subsequently, in a similar manner, the fine-motion mechanism is controlled so that an electric current of 1 nA flows between the electrodes.
Thereafter, the entire region as encircled by a broken line in Figure 64 is scanned to write the patterns 9012a - 9012d shown in Figure 64.
In the initial state of the recording medium, the whole region is in an "OFF" stateO In order that the portion onto which a pattern should be recorded is xendered into an "ON" state, an electric voltage o~ +10 V which is higher than a threshold level concerning the "ON-OFF transition" of the recording medium is applied to between the electrodes when the probe electrode comes to the pattern recording position. By this, the illustrated pattern can be produced. In the drawing, the painted area ~ '. ' ..

-147_ 1328131 shows such portion which is in the "ON" state. Slnce, at the time of recording, the electric current flowing between the electrodes increases by about three figures. Accordingly, to avoid the effect thereof to 5 the interval control for the electrodes, the fine-motion mechanism is controlled so that an average current flowing between the electrodes is maintained at 0.9 nA and, by doing so, an average distance between the electrodes can be maintained constant~ -For the recording/reproducing upon the ordinary recording region, first, the probe electrode is positioned with respect to the positional reference pattern. For this purpose, a voltage of +5 V lower than the threshold concerning the "ON-OFF transition"
15 of the recording medium is applied to between the electrodes and, in this state, the fine-mQtion and rough-motion mechanisms are controlled to cause the probe electrode to scan over the whole recording medium. During the scan, the "ON/OFF" state is 20 detected on the basis of the changes in the tunnel current between the electrodes to thereby detect the positional reference pattern. Since the positional reference pattern occupies a wider area, than the record unit of the ordinary recording information, ~5 when the "ON/OFF" state thereof is detected by the scan of the probe electrode, the detected ON/OFF
frequency is remarkably lower than that of the `

ordinary recording information. Therefore, by using a simple band fllter group, the scanning detection signals of the positional reference pattern and the ordinary recording information can be easily separated 5 as a low frequency component and a high frequency component. On the basis of two-dimensional information concerning the plural positional reference patterns, the probe electrode is moved to a position in the neighborhood of one of the four patterns.
10 Subsequently, such a range whlch is slightly wider than the pattern region is scanned and the signals obtained therefrom are stored into a memory as two-dimensional information. As shown in Figure 64, each reference pattern can be discriminated on the basis of 15 the direction of the array of directional patterns 9013 which are arrayed in the X and Y directions specifically. Also, the position of the pattern being detected can be discriminated, on the basis of the relative position of a positional pattern 9014 with 20 respect to the positional reference pattern. Namely, the positional reference pattern contains directional information and positional information. Therefore, by using prevalently adopted image processing method, the information concerning the scan direction at the time 25 of actual recording/repxoducing as well as the information concerning the position of the pattern being detected~ can be obtained. Further, from the : ~;

stored positional relation between the patterns, the information concerning the approx~mate positlon of the positional reference pattern related to the next ordinary recording region.
For the recording of the ordinary recording information, the probe electrode is moved to a position in the neighborhood of the ordinary recording region on the basis of the positional information.
Then, such a range as being slightly wider than the 10 recording region is scanned. After a synchronization signal in the positional reference pattern is detected, the information is written at a predetermined position on the basis of the positional information and with the precision of the 15 piezoelectric device. For the reproduction of the ordinary recording information, two-dimensional information as obtained by the scan of the whole region after the synchronization signal is divided into a record unit (unit information region) as shown ; 20 in Figure 65, while taking into account the hysteresis of the piezoelectric device, thermal ; expansion/contraction of the recordlng medium and other information, having been inputted separately.
- Each unit region is image-processed to reproduce the ; 25 recorded information (1 or O). After the ~`~ recording/reproducing is completed, the probe electrode is moved to a position in the neighborhood `
., .

. ' .. , ' ~ ~: ' -150- 1328~31 of the next pattern, on the basis of the obtained positional information.
In the present embodiment, the scanning mechanism is separated into a rough-motlon scanning 5 portion for low-frequency scanning and a fine-motion scanning portion for high-frequency scanning. This is effective to make it possible to reduce the stroke o~
the fine-motion scanning portion and to increase the rigidity thereof. Therefore, the scan speed can be 10 increased and as a result of which, the information transfer speed can be increased. As for the fine-motion mechanism, known type fine-motion mechanisms or devices are usable, axamples of which have been described hereinbefore. In the present embodiment, a 15 tube scanner having high natural frequency is used.
On the other hand, as for the rough-motion mechanism, parallel spring means having high-performance resilient hinge mechanism, in which the freedom of motion is restricted to the movable direction, is 20 used.
By using parallel spring means comprising a resilient hlnge mechanism, as the rough-motion mechanism for the recording medium, it is possible to suppress any distortion of the recording medium at the 25 rough-motion in the direction perpendicular to the recording medium surface ti.e. in the Z-axis direction). Therefore, it is possible to impxove the signal-to-noise ratio at the time of recording/reproducing~ Also, it is po~sible to enlarge the rough-motion range while retaining high rigidity of the piezoelectric device.
Further, of various three-dimensional scanners using ordinary piezoelectric devices, a cylindrical type piezoelectric device having high resonance frequency is used as the fine-motlon mechanism for the probe electrode. By this, the response frequency of the fine-motion mechanism can be improved and, therefore, the scan speed can be increased. As a result, it is possible to improve the record information transfer speed.
By using a positional reference pattern and by providing unit for the recordinglreproducing of signals, the signal processing at the time of recording/reproducing of signals, the signal processing at the time of recording/reproducing can be made simple.
Since the positional reference-pattern contains two-dimensional information, ik is possible to cancel or compensate for during the scan, any positional deviat~on of the record information due to any-rotation or distortion of the recording medium 9 25 caused when it is mounted to the apparatus.
In the present embodiment, the information is not treated one-dimensionally but lt is treated :,~ .. . .
`,~' " '"~, :

two-dimensionally. This results in improved degree of freedom of position change at the time of reproduction. As a result, without positioning by use of faedback control or otherwise, high-density recorded information can be read out without error.
Since the racorded information is processed as a unit of area, it is not necessary to reproduce, for every scan, the position of the signal at the time of recording. Namely, as shown in Figure 66, even if during the reproduction the scanning actuator is not correctly on tha recorded signal for a short time period due to any external disturbance, it is possible to reproduce a two-dimensional image representing the record unit and the positional relation between the record units, provided that the two-dimensional image obtained by the whole-region scan is within the tolerance of image processing. Therefore, substantially no problem occurs for the reproduction of information. Also, even if there occurs expansion/contraction of the medium due to the difference in temperature or humidity between the time of raproduction and the time of recording, the change in the obtained two-dimensional image of the information is merely an analogous deformation of the original image and, therefore, there occurs substantially no readout error. For a similar reason, recording mediums can be interchangeably used between :~ . . . .
. .

-153- ~ 328131 different apparatuses~ -Further, the mounting of the recording medium to the apparatus does not require high-precision mechanisms. ~hereforet the recording/reproducing apparatus can be produced with a low cost and by mass-production.
While the invention has been described with reference to the structures disclosed herein, it is not confined to the details set forth and this application is intended to co~er such modifications or changes as may come within the purposes of the improvements or the scope of the following claims.

., '

Claims

1. An encoder, comprising:
an electrically conductive reference scale having surface steps formed at predetermined positions;
an electrically conductive probe having a tip disposed opposed to said reference scale;
wherein said reference scale and said probe are relatively movable in a direction different from the opposing direction of said tip of said probe and said reference scale;
means for applying an electric voltage to between said reference scale and said probe;
means for detecting a change in a tunnel current between said reference scale and said probe, to between which the electric voltage is applied by said voltage applying means at the time of the relative movement between said scale and said probe, said detecting means detecting the change in the tunnel current when the probe passes a position opposed to a surface step of said reference scale; and means for detecting the amount of the relative movement between said scale and probe, on the basis of the detection by said change detecting means.

2. An encoder according to Claim 1, wherein said amount detecting means detects the amount of the relative movement on the basis of the amount of change in the tunnel current as detected by said change detecting means.

3. An encoder according to Claim 1, wherein said amount detecting means includes means for moving at least one of said reference scale and said probe in said opposing direction on the basis of the amount of change in the tunnel current as detected by said change detecting means, and means for detecting the amount of the relative movement between said reference scale and said probe on the basis of the amount of movement made by said moving means.

4. An encoder according to Claim 1, further comprising means for relatively oscillating said reference scale and said probe in a direction different from said opposing direction, and means for detecting the direction of the relative movement between said reference scale and said probe on the basis of a change in phase of the change in the tunnel current between the reference scale and said probe being relatively oscillated by said oscillating means.

5. An encoder according to Claim 1, further comprising means bearing positional information and being provided on one of said reference scale and said probe at a pitch larger than the pitch of said surface step, and means for roughly detecting the amount of the relative movement between said reference scale and said probe on the basis of the positional information.

6. An encoder according to Claim 1, further comprising first position detecting means for detecting first and second scales, provided for rough detection of the amount of relative movement between relatively movable first and second objects, for detecting the amount of the relative movement between the first and second objects, wherein one of said reference scale and said probe is provided on said first object and the other is provided on one of said second scale and said first position detecting means and wherein, for fine detection of the amount of the relative movement between the first and second objects, the amount of the relative movement between said reference scale and said probe is detected.

7. A relative position detecting device for detecting the positionaI relation between first and second objects in a predetermined direction; said device comprising:
an electrically conductive reference scale having positional information for each predetermined position of said first object;

an electrically conductive probe disposed on said second object and having a tip opposed to said reference scale;
wherein said reference scale and said probe are opposed to each other in a direction different from said predetermined direction:
means for applying an electric voltage to between said reference scale and said probe;
means for detecting a change in a tunnel current between said reference scale and said probe, to between which the voltage is applied by said voltage applying means when said reference scale and said probe relatively move in said predetermined direction, said change detecting means detecting the change in the tunnel current when said probe passes a position opposed to said positional information; and means for detecting the positional relation in said predetermined direction between said first and second objects, on the basis of the detection by said, change detecting means.

8. A device according to Claim 7, further comprising a recording medium provided on said first object, and means for recording or reproducing information at a predetermined position on said recording medium, on the basis of the detection by said positional relation detecting means.

9. A device according to Claim 8, wherein said recording means has an electric memory effect and wherein said information recording or reproducing means includes an electrically conductive second probe having a tip disposed opposed to said recording medium and means for applying an electric voltage to between said probe and said information recording or reproducing means, said second probe being used for the information recording or reproducing.

10. A device according to Claim 8, wherein said recording medium has an electric memory effect, wherein said reference scale is provided on said recording medium and wherein said information recording or reproducing means effects the information recording or reproducing by use of said probe.

11. A device according to Claim 10, wherein said reference scale has a two-dimensional pattern which contains positional information.

12. An encoder according to claim 1, further comprising driving means for imparting relative movement between said scale and probe in a direction different from the opposing direction.

13. An encoder according to claim 12, wherein said driving means is a piezoelectric actuator.

14. An encoder according to claim 7, further comprising driving means for imparting relative movement between said first and second object.

15. An encoder according to claim 14, wherein said driving means is a piezoelectric actuator.

16. An encoder according to claim 8, wherein the recording medium is provided on a first side of the first object, and said scale is on a second side of the first object.