In the art there is a need for apparatus for accurately- displacing an object in two mutually substantially perpendicular directions, which are normally located in a horizontal plane, and which in the context of the present application will be designated as X- and Y-directions. More particularly, there is a need for such an XY displacement device which is able to displace an object with a positional accuracy better than l nm (nanometer) over distances in the range of less than 1 nm up to more than 100 um. An important field of application is the submicron technology and the subnanotechnology, that is, the technology for manufacturing structures with dimensions of even less than 1 μm. Another important field of application is the field of imaging surface structures with dimensions in the nanometer range and subnanometer range. Such imaging occurs, for instance, by means of a Scanning Probe Microscope (SPM) , such as for instance a Scanning Tunneling Microscope (STM) , and the present invention will be further explained, by way of non-limiting example, for this specific field of application. An STM is a device, well-known in the art, for forming an image of the structure of the surface of an object. A probe with a sharp tip is brought to a point at a short distance from the surface under examination, and between that probe and that surface a potential difference is applied. Although there is no electrical contact between that probe and that surface, still a current starts to flow, the so-called tunnel current, and the strength of this tunnel current is a measure for the distance between the probe and the surface, which can also be regarded as an indication of the Z-position of the surface at the location of the probe, to be designated as z(x,y) . For obtaining an image of the structure of the surface, it is desired to measure the Z-position of the surface over the entire surface. To that end, the probe is caused to make a scanning movement over the surface.

Since the operation of an STM as such does not constitute a subject of the present invention, and knowledge thereof is not required for a skilled person to properly understand the present invention, this will not be further descibed. Suffice it to note that the scanning movement mentioned is generally effected by displacing the object under examination in the X-direction and the Y-direction relative to the probe or, conversely, by displacing the probe in the X-direction and in the Y-direction relative to the object under examination. It will be clear that to obtain an accurate and faithful image of the surface structures to be examined, it is desired that such displacement is particularly accurate. This is especially the case if the SPM is used for making structures of dimensions in the nanometer range. Known XY displacement devices suffer from one or more disadvantages.

An important disadvantage of known XY displacement devices is that the X-displacement and the Y-displacement are not independent of each other. According to a known construction, an object mount is mounted on a subframe for displacement in the Y-direction, and that subframe is mounted on a main frame for displacement in the X-direction. A consequence thereof is that upon displacement of the subframe in the X-direction, it is virtually unavoidable that a parasitic displacement of the object mount in the Y-direction will occur.

It is therefore an object of the present invention to provide an XY displacement device where the X-displacement and the Y-displacement are uncoupled relative to each other. A further consequence of the above-mentioned known construction is that undesired displacements can occur as a result of temperature variations, while moreover the temperature characteristic in the Y-direction differs from the temperature characteristic in the X-direction. It is therefore a further object of the present invention to provide an XY displacement device which is symmetrical. More particularly, the present invention contemplates the provision of an XY

displacement device where temperature variations will substantially not cause any displacement of the object to be examined.

For displacing the object, actuators are needed. Since a high accuracy is desired (better than l nm) , while the scanning range can vary from a few nanometers to more than 100 micrometers, typically piezoelectric actuators are used, since an actuator of this type is particularly suitable for the purpose mentioned, inter alia because of the relatively simple control thereof and the relatively slight dimensions thereof. As is well known, a piezoelectric actuator has as a property that a length dimension varies depending on an electrical voltage applied to the actuator; such applied voltage is therefore typically used as a measure for the instantaneous XY position of the object.

Piezoelectric actuators, however, have a few properties which adversely affect the positional accuracy of the object to be examined. Examples of such adverse properties are, for instance, hysteresis, non-linearity, drift, creep, ageing, temperature dependence. A complicating factor here is that the hysteresis is dependent inter alia on the magnitude of the range to be scanned, so that it is particularly difficult, if not impossible, to calibrate a piezoelectric actuator. The properties mentioned give rise to geometric deformations of the images obtained, which deformations in turn are position- dependent, with the positional errors that occur possibly running up to 20% of the scanning range.

The problems mentioned are known as such, and different attempts have already been made to solve those problems. However, those attempts have not led to fully satisfactory results. In one particular attempt, the piezoelectric actuator is controlled with a non-linear driving voltage, the non¬ linear character of the driving voltage having been chosen with a view to compensating the non-linear behaviour of the piezoelectric actuator. This approach cannot lead to accurate results because the non-linear behaviour of the piezoelectric actuator is dependent on the scanning range and is changeable

as a result of, for instance, ageing, temperature influences and depolarisation effects.

In another approach the images obtained are corrected afterwards through application of image processing techniques. The application of such techniques is typically based on the presumed presence of particular periodic structures on the surface of the object, that is, a particular expectation of the investigator. The danger is then present that particular non-linear effects with a fairly large characteristic length are filtered out of the image. If the surface has no periodic features at all, image reconstruction is very difficult, if not impossible.

The invention contemplates the provision of an XY displacement device where the difficulties mentioned do not occur.

To that end, an XY displacement device according to the invention has the features as described in claim l.

The above-mentioned and other aspects, features and advantages of the present invention will be clarified through the following description of a preferred embodiment of an XY displacement device according to the invention, with reference to the drawings, in which: Fig. 1 is a top plan view of a preferred embodiment of an XY displacement device according to the invention; Fig. 2 diagrammatically illustrates the principle of the suspension of the object mount in the reference frame; Fig. 3 is a cross section of an STM which includes the XY displacement device represented in Fig. 1; and

Fig. 4 shows a block diagram illustrating the data processing with an XY displacement device according to the invention.

Fig. l shows a top plan view of an XY displacement device 1 according to the present invention. In this figure, the X-direction is represented horizontally and the Y-direction vertically. The embodiment of the XY displacement device 1 as

shown is intended for use in an SPM to have a probe make a scanning movement along a fixedly arranged specimen whose surface is to be examined. It will be clear, however, that it is also possible to use the XY displacement device 1, optionally after a few modifications, to have the specimen under examination make a scanning movement along a fixedly arranged probe. In the context of the present invention, the probe to be displaced, or the specimen to be displaced, will be designated by the term "object". Reference numeral 10 designates a reference frame, which is intended to be fixedly attached to, for instance, an apparatus frame or the fixed environment, for which purpose fixing holes 11 can be used.

Centrally in the reference frame 10, an object mount 20 is present, which is displaceable. The object mount 20 is intended for attaching thereto the object to be displaced (probe) and to that end is provided with fixing holes 21.

The reference frame 10 is further intended for attaching thereto a support means for a specimen or a specimen mount, as will be described in more detail hereinbelow, in such a manner that a specimen can be retained centrally with respect to the reference frame 10, opposite the probe fastened to the object mount 20. It will be clear that displacement of the object mount 20 with respect to the reference frame 10 is equivalent to displacement of a probe, attached to the object mount 20, relative to a specimen fixed with respect to the reference frame 10.

For the purpose of mounting such a support means, the fixing holes 11 can also be used. As shown in Fig. 1, and represented diagrammatically in Fig. 2, an X-manipulator 100 is coupled between the object mount 20 and the reference frame 10. The X-manipulator 100 has three main functions. Firstly, the X-manipulator 100 should provide a rigid coupling between the object mount 20 and an X-actuator, such as, for instance, a piezoelectric actuator. For the sake of simplicity, such an actuator is not represented in Figs, l and 2, but the force to be exerted by

the X-actuator is symbolically represented by the arrow F _x . By virtue of the rigid coupling contemplated, the object mount 20 accurately follows an X-displacement relative to the reference frame 10 as imposed by the X-actuator. Secondly, the X-manipulator 100 should allow a Y-displacement of the object mount 20 relative to the reference frame 10, on the one hand without this generating a (parasitic) X-displacement of the object mount 20 relative to the reference frame 10, and, on the other hand, without a transverse force being exerted on the X-actuator.

Thirdly, the X-manipulator 100 should should provide a rigid coupling between the object mount 20 and the reference frame 10 in the Z-direction (that is, perpendicular to the X- and Y-directions) in order to prevent a Z-displacement of the object mount 20 relative to the reference frame 10

(perpendicular to the plane of the paper) , again without a transverse force being exerted on the X-actuator. According to the invention, to that end, the X-manipulator 100 is constructed as a two-stage suspension for the object mount 20. The X-manipulator 100 comprises an actuator input part 110 which is intended to be coupled to an X-actuator; a first coupling part 120 for coupling the actuator input part 110 to the reference frame 10, and a second coupling part 130 for coupling the actuator input part 110 to the object mount 20.

In the embodiment shown in Fig. 1, the actuator input part 110 has a substantially rectangular shape, whose Y-dimension substantially corresponds with the Y-dimension of the object mount 20. During use, a piezoelectric actuator, not shown in this figure for clarity, will act on the actuator input part 110, as designated by the arrow F _x , which piezoelectric actuator on the other side thereof is coupled with the reference frame 10, as will be clear to a skilled person. By presenting a suitable control signal to such a piezoelectric actuator, an X-dimension of that actuator will change, and hence the X-distance between the actuator input part 110 and the reference frame 10, which amounts to the

imposition of an X- isplacement relative to the reference frame 10 on the actuator input part 110.

The first coupling part 120 of the X-manipulator 100 is designed to retain the actuator input part no with respect to the reference frame 10, in such a manner that the actuator input part 110 is displaceable in the X-direction and thus can follow a displacement imposed by the piezoelectric actuator, while further this actuator input part 110 is substantially fixed in the Y-direction, and preferably also in the Z-direction. For fixing the actuator input part 110 in the Y-direction, the first coupling part 120 comprises a first bar-shaped member 121 which, with its longitudinal direction parallel to the Y-direction, extends from the actuator input part 110 to a first attachment point 12 of the reference frame 10. In view of the desired symmetry, the first coupling part 120 further comprises a second bar-shaped member 122, which extends in line with the first bar-shaped member 121, from the actuator input part 110 to a second attachment point 13 of the reference frame 10, the second bar-shaped member 122 being preferably as long as the first bar-shaped member 121.

Owing to this orientation, the bar-shaped members 121 and 122 will counteract any Y-displacement of the actuator input part 110, because such a displacement will generate in those bar-shaped members 121 and 122 a tensile stress and a compression stress, respectively.

The two bar-shaped members 121 and 122 have an X-dimension (width) which is sufficiently smaller than their Y-dimension (length) , so that those bar-shaped members 121 and 122 are relatively slack in the X-direction and behave more or less as a leaf spring. Accordingly, they do not resist displacement of the actuator input part 110 in the X-direction. As a result of such an X-displacement of the actuator input part 110, those bar-shaped members 121 and 122 will bend slightly and a tensile stress will be generated in those bar-shaped members 121 and 122. The consequences of this for the Y-position of the object mount 20, however, are nil or

in any case negligible, since the X-displacement of the actuator input part 110 is particularly small in proportion to the length of the bar-shaped members 121 and 122, while moreover, because of the symmetrical arrangement of the bar-shaped members 121 and 122, the Y-displacements, if any, caused by the bending of the two bar-shaped members 121 and 122 substantially cancel each other out.

The second coupling part 130 of the X-manipulator 100 is designed to transmit an X-displacement, if any, of the actuator input part 110 to the object mount 20. To that end, the second coupling part 130 comprises a third bar-shaped member 131 which, with its longitudinal direction parallel to the X-direction, extends between the actuator input part 110 and the object mount 20, so that this third bar-shaped member 131 counteracts an X-displacement of the object mount 20 with respect to the actuator input part 110. This third bar-shaped member 131, like the first and second bar- shaped members 121 and 122, has a width (Y-dimension) much smaller than its length (X-dimension) , so that this third bar- shaped member 131 is relatively slack in the Y-direction and behaves as a leaf spring. As a result, the third bar-shaped member 131 allows a Y-displacement of the object mount 20 relative to the actuator input part 110, while the resultant reaction force in the Y-direction exerted on the actuator input part 110 is relatively small.

For fixing the object mount 20 in the Z-direction, separate means can be present. According to the invention, however, this effect can already be accomplished by choosing a suitable configuration for the bar-shaped members 121, 122, 131. If the Z-dimension of the third bar-shaped member 131 is selected to be sufficiently greater than the Y-dimension thereof, and preferably is substantially as great as the X-dimension thereof, the third bar-shaped member 131 is relatively stiff in the Z-direction and counteracts a Z-displacement of the object mount 20 relative to the actuator input part 110. If the Z-dimension of the first and second bar-shaped members 121 and 122 is selected to be sufficiently

greater than their Y-dimension, and is preferably substantially as great as their X-dimension, the first and second bar-shaped members 121 and 122 are relatively stiff in the Z-direction and counteract a Z-displacement of the actuator input part 110 relative to the reference frame 10.

This means that the object mount 20 is maintained fixed in the Z-direction without the X-actuator being loaded in the Z-direction.

For reasons to be discussed later, the third bar-shaped member 131 is preferably arranged next to an axis of symmetry of the device 1 extending in the X-direction, and on the other side of that axis of symmetry 2 a fourth bar-shaped member 132 is arranged, substantially identical to and parallel to the third bar-shaped member 131, as illustrated in Fig. 1. If in some manner or other a Y-displacement of the object mount 20 occurs, the third bar-shaped member 131 will bend. Since the Y- isplacement of the object mount 20 will always be much smaller than the length of the third bar-shaped member 131, the resultant bending of the third bar-shaped member 131 will be minimal, so that any X-displacement of the object mount 20 thereby caused, will be particularly small. For such an X-displacement to be cancelled as well, the object mount 20 is preferably coupled via a complementary X- manipulator 100' with the reference frame 10. The complementary X-manipulator 100' is arranged mirror- symmetrically with respect to the X-manipulator 100, is located on the side of the object mount 20 remote from the X- manipulator 100, and can be identical in construction to the X-manipulator 100. For simplicity, in Fig. 1 the parts of the complementary X-manipulator 100' are designated with the same reference numerals as the corresponding parts of the X- manipulator 100, but primed.

What is achieved through the symmetrical arrangement of the two X- anipulators 100 and 100' is not only that a Y-displacement of the object mount 20 causes no, or at best a negligible, X-displacement of the object mount 20, but also

that temperature changes have no net influence on the X-position of the object mount 20.

Since the operation of the complementary X-manipulator 100' corresponds with that of the X-manipulator 100, that operation will not be separately discussed. However, it is noted that between the actuator input part 110' and the reference frame 10 no piezoelectric actuator is coupled, but a resilient member, such as a compression spring to press the section consisting of the means 110', 20 and 110 against the piezoelectric actuator under a bias, which improves the stiffness in the X-direction. For simplicity, such a compression spring is not shown in the drawing; only the biassing force exerted by it is symbolically represented by the arrow Fχ- _b i _a s-

Further coupled between the object mount 20 and the reference frame 10 is a Y-manipulator 200, which also has three main functions, which mutatis mutandis are equal to those of the X-manipulator 100. Firstly, the Y-manipulator 200 should provide a rigid coupling between the object mount 20 and a Y-actuator; the force to be exerted by the Y-actuator is symbolically represented by the arrow F _y . Owing to the rigid coupling contemplated, the object mount 20 accurately follows a Y-displacement relative to the reference frame 10 as imposed by the Y-actuator.

Secondly, the Y-manipulator 200 should allow an X-displacement of the object mount 20 relative to the reference frame 10, on the one hand without this generating a (parasitic) Y-displacement of the object mount 20 relative to the reference frame 10 and, on the other hand, without a transverse force being exerted on the Y-actuator.

Thirdly, the Y-manipulator 200 should provide a rigid coupling between the object mount 20 and the reference frame 10 in the Z-direction in order to prevent a Z- displacement of the object mount 20 relative to the reference frame 10 (perpendicular to the plane of the paper) , again without a transverse force being exerted on the Y-actuator.

The Y-manipulator 200 represented in Fig. 1 is identical to the X-manipulator 100 already discussed, with the understanding that the orientation thereof is rotated 90° relative to the orientation of the X-manipulator 100. The parts of the Y-manipulator 200 are provided with a reference numeral higher by 100 than the reference numerals of the corresponding parts of the X-manipulator 100. A separate discussion of the operation of the Y-manipulator 200 is omitted here, because that operation will be clear to a skilled person after reading the above discussion of the operation of the X-manipulator 100.

Further provided is a complementary Y-manipulator 200', whose operation and parts correspond with those of the complementary X-manipulator 100' . It is noted that the first coupling part 220 of the Y-manipulator 200 can be connected to the same points of connection 13 and 13' of the reference frame 10, as illustrated, or with separate attachment points. A comparable remark applies to the complementary Y-manipulator 200'.

With the construction described in the foregoing, the invention already provides an improvement over existing XY displacement devices, in particular because of the symmetry. It is readily seen that a change in temperature of the whole can give rise to a change in the mechanical stress prevailing in the reference frame 10 and the manipulators, but that, owing to the symmetrical construction, this does not induce any displacement of the object mount 20 relative to the axis of symmetry in the Z-direction. Further, it will be clear that with such a construction as described, the X- and Y-displacements of the object mount 20 are substantially mutually uncoupled.

In a further preferred embodiment, the invention provides a further improvement over the state of the art, in that measuring means are provided for measuring at the object mount 20 the actual displacement thereof. With reference to

Fig. 3, a preferred embodiment of such measuring means will be discussed.

Fig. 3 shows a schematic cross section, taken along the line of symmetry 2, of the device 1 illustrated in Fig. l, in a ready-for-use condition, that is, provided with a piezoelectric actuator 140, a biassing spring 150, a probe 22, and a specimen mount 31.

The piezoelectric actuator 140 is arranged outside the reference frame 10. Via a throughbore 145 in the reference frame 10 a first end 141 of the piezoelectric actuator 140 is coupled with the actuator input part 110; the other end 142 of the piezoelectric actuator 140 is immovably connected to the reference frame 10 via an arcuate support member 143. Lead wires for the piezoelectric actuator 140 are designated by 146 and 147.

The biassing spring 150 is comparably mounted outside the reference frame 10 through an arcuate support member 143' .

The specimen mount 31 is fastened to a supporting plate 30, which in turn is immovably connected to the reference frame 10, for instance by means of screws, which is not shown in the figure for the sake of simplicity.

The probe 22, which can have a construction known per se, is mounted in the object mount 20, and reaches through an opening 32 in the supporting plate 30, up to a point in the vicinity of a specimen (not shown) to be mounted in the specimen mount 31. The probe 22 is provided with means 23 for displacing the probe 22 in the Z-direction, which means 23 can likewise have a construction which is known per se. According to the invention, there are provided X-measuring means 300 for providing a measuring signal which is indicative of the X-position of the object mount 20 with respect to the reference frame 10. In an embodiment which is preferred because of its simplicity, those X-measuring means 300 comprise capacitive measuring sensors 310, 310'. The measuring sensor 310 comprises a reference measuring electrode 311 fixedly mounted with respect to the reference frame 10, and an object measuring electrode 312 which is

fixedly mounted to the object mount 20. Preferably, and as illustrated, the reference measuring electrode 311 is fastened to the supporting plate 30 for the specimen mount 31 via an electrode support 313. The two measuring electrodes 311 and 312 - the size of the distance between them is exaggerated for the purpose of clarity in the representation of Fig. 3 - each have substantially the shape of an insulating flat plate on which conducting structures are arranged, and they are spaced apart a short distance, parallel to each other, so that between them a capacity C is defined. The two measuring electrodes 311 and 312 are provided with lead wires, not shown for the sake of simplicity, for connection to a measuring apparatus, likewise not shown for the sake of simplicity, so that the capacity C can be measured. As is well known, such a capacity C is inversely proportional to the distance between the plate electrodes 311 and 312, so that the measured capacity C can be regarded as a measuring signal that is representative of that distance and hence for the X-position of the object mount 20 with respect to the reference frame 10. Since the manner in which the capacity C is measured does not constitute a subject of the present invention, and knowledge thereof is not required for a skilled person to properly understand the present invention, and use can be made of measuring apparatus known and available for this purpose, the capacity measurement will not be further described. Suffice it to note that here a measuring signal (for instance a frequency or a time) can be obtained which, without requiring that the capacity be actually calculated therefrom, is already representative as such of the X-position of the object mount 20 with respect to the reference frame 10.

For obtaining a measuring signal which is representative of the X-position of the object mount 20 as well as possible, the measuring sensor 310 is preferably arranged in the space 133 between the bar-shaped members 131 and 132 of the second coupling part 130.

The capacity to be measured depends not only on the relative distance between the plate electrodes 311 and 312, but also on factors of the environment, such as the dielectric constant of the medium (for instance, air) , if any is present, between the plate electodes 311 and 312. Further, the relative distance between the plate electrodes 311 and 312, and hence the capacity to be measured, can change if parts of the device 1 change in shape, for instance if the object mount 20 expands or shrinks as a result of temperature changes, even if in the process the object 22 retained by the object mount 20 does not change position. The factors mentioned can therefore be regarded as a source of inaccuracies. To correct for them as well as possbible, preferably two measuring sensors 310, 310' of the type mentioned are arranged on opposite sides of the object mount 20. Upon a displacement of the object mount 20 in the X-direction (to the right in Fig. 3) the electrode distance of the first measuring sensor 310 will increase and so the measuring signal Cl provided by it will become smaller, while the electrode distance of the second measuring sensor 310' will become smaller and hence the measuring signal C2 provided by it will become larger. By contrast, upon a change of the dielectric constant of the medium or a thermal expansion of the object mount 20, both measuring signals will change to an equal extent. Thus, for instance, the difference signal C1-C2 is representative to an improved extent of the X-position of the object mount 20. An additional advantage of the use of the difference signal C1-C2 as the signal representing the X-position of the object mount 20 is related to the fact that the nominal capacity of the measuring sensors (order of magnitude pF) is much greater than the required resolution in the capacity changes (order of magnitude aF) . When the object mount 20 is properly centered with respect to the measuring sensors, the nominal value of the difference signal C1-C2 is zero, at least smaller than the expectable changes in the difference signal C1-C2.

In practice, it is virtually inevitable that the electrode plates 311 and 312 are not exactly parallel to each other. A consequence of this is that the electrode plates 311 and 312 are not directed exactly parallel to each other, in which case the formula which describes the relationship between the capacity C and the electrode distance acquires a more complex form, in which the extent of misalignment plays a role. Since in practice such misalignment is usually small and constant, it is possible that the measuring inaccuracy caused by this is negligible. It is also possible, however, to design the X-measuring means such that they also provide information about any misalignment of the electrode plates with respect to each other, so that this can be corrected for. To that end, one of the electrode plates 311, 312 can be divided into four segments, so that the measuring sensor 310 in fact comprises four measuring capacities yielding four partial capacity signals cl, c2, c3, c4, the sum signal cl+c2+c3+c4 corresponding with the above-mentioned measuring signal C. By suitably processing the four partial capacity signals cl, c2, c3, c4, it is possible to set the electrode plates parallel to each other. As a result, it is possible to determine the displacement of the object mount 20 relative to the reference frame 10.

The XY-displacement device 1 further comprises

Y-measuring means 400 for supplying a measuring signal that is representative of the Y-position of the object mount 20 with respect to the reference frame 10. These Y-measuring means 400 are preferably identical to the above-discussed X-measuring means 300, with the understanding that they have been displaced 90° with respect to the X-measuring means 300, as will be clear to a skilled person. For this reason, the Y-measuring means 400 will not be discussed separately, nor are they separately illustrated in Fig. 3.

Presently, with reference to Fig. 4, by way of example, an SPM 500 will be discussed, in which the XY-displacement

device l according to the present invention is used. The SPM 500 comprises an actuator driving unit 510 for supplying control signals a _x and a _y for the X- and Y-actuators 140 and 240. The SPM 500 further comprises a first data processing device 520 for receiving and processing the measuring signals C _x and C _y supplied by the X- and Y-measuring means 300 and 400. The SPM 500 further comprises a second data processing device 530 for receiving and processing the information received by means of the probe 22. The invention provides two different ways in which the measuring signals C _x and C _y provided by the X- and Y-measuring means 300 and 400 can be utilized during the performance of a displacement of the object 22 retained by the object mount 20. These two ways will be explained in the following, where the target X- and Y-positions (coordinates) will be designated as X _T and y _T , respectively, and the real X- and Y-positions will be designated as X _R and y _& , respectively. Those real X- and Y-positions are calculated (or at least approximated) with the aid of the measuring signals C _x and C _y , supplied by the X- and Y-measuring means 300 and 400, and not, as is conventionally customary, on the basis of control signals a _x and a _y .

According to a first method, the X- and Y-actuators 140 and 240 are driven in a conventional manner for the displacement of the object mount 20 to different target positions (x _τ , y _T ) • The actuator driving unit 510 has information about the characteristic of the X- and Y-actuators 140 and 240, and calculates, on the basis of that information, control signals a _x (xτ) and a _y (yτ) as a function of the intended X- and Y-coordinates X _T and y _τ . As has been explained in the foregoing, the above-mentioned information is probably incorrect, so that the real X- and Y-coordinates x _R (a _x (x _τ )) and y _R (a _y (y _τ )) resulting from the control signals a _x (xτ) and a _y (yτ) will be different from the intended coordinates x _τ and y _τ . The second data processing device 530 of the SPM 500 receives a measuring signal from the probe 22, which will be designated with the letter Ψ. That measuring signal Ψ can for

instance involve a Z-coordinate of the surface to be examined. According to this first method according to the present invention, in addition, the real X- and Y-coordinates x _R en y _R calculated on the basis of the measuring signals C _x and C _y supplied by the X- and Y-measuring means 300 and 400 are supplied to the second data processing device 530, as indicated with the signal path 540 ¹ , and the second data processing device 530 processes the measuring signal Ψ as belonging to the real position (x _R , y ) . This first method has the advantage that no changes are necessary for an existing control device. It is true that a measurement is not performed exactly at the intended position, but the real position (x _R , y ) where the measurement is performed, is known, and the measuring signals Ψ(x _R , y _R ) are correct. A contour measurement performed in this way provides a correct image of the contour of the part of the specimen surface actually examined.

A second method according to the present invention is suitable in particular for applications where it is desired for the real position to be equal to the intended position. An example of such an application is a contour measurement where it is desired to obtain the measuring signals Ψ at specific, priorly known X- and Y-coordinates X _T and y _T - Another example of such an application is a machining technique for fabricating a structure in or on the specimen surface.

According to this second method according to the present invention, the object mount 20 is displaced each time in such a manner that the real position (x _R , y _R ) is equal to the target position (X _T , yr) • To that end, the actuator driving unit 510 is provided, via a feedback loop, with information representative of the real position (x _R , y _R ) , as indicated with the signal path 540 ¹¹ , and the actuator driving unit 510 adjusts the control signals a _x and a _y to reduce the differences x _R -xτ and y -yτ to substantially zero. In a variant, the actuator driving unit 510 receives information that is representative of those differences x _R -xτ and y _R -y

which information can take three values, viz. "too large", "too small", and "within a pre-set tolerance".

In a variant, it is desired to displace the object mount 20 linearly, for instance in the X- direction, that is, while keeping the Y-coordinate constant. Then the displacement in the X-direction is performed in accordance with either of the two above-discussed methods, while the Y-coordinate is kept equal to an initial Y-coordinate by controlling the Y- actuator 240 such that the measuring signal C _y representing the Y-coordinate remains constant. It is then not necessary to calculate the real Y-coordinate each time. Such a control variant occurs, for instance, when scanning a surface with an SPM according to a back and forth movement. In the foregoing, it has been explained that by virtue of the design of the XY displacement device 1 as proposed by the invention, the X- and Y-displacements are uncoupled. If nevertheless, for whatever reason, an undesired displacement in, for instance, the Y-direction threatens to arise during an imposed movement in, for instance, the X-direction, such threatening undesired displacement can be counteracted in a simple manner in accordance with the invention by adjustment of the control signal (a _y ) for the (Y-) actuator in question. The same applies with regard to undesired displacements that might occur as a result of external disturbances, such as vibrations, drift, creep, etc.

In principle, it is possible to build up the XY-displacement device 1 from loose parts, which are then joined together, for instance by gluing or welding. This involves a disadvantage, however, in that non-symmetrical stresses can thereby be introduced into the device 1, whose magnitude, direction and distribution characteristic are unknown. In view of the desired accuracy, it is therefore preferred to manufacture the XY-displacement device 1 as an integrated whole, by making suitable recesses in a massive workpiece, for instance through spark machining or laser

cutting. An embodiment which has been found suitable is manufactured starting from a rectangular block of stainless steel of a thickness (Z-dimension) of 2 cm, a length (X-dimension) of 11 cm, and a width (Y-dimension) of 11 cm. The X- and Y- external dimensions of the object mount 20 were both 2 cm. The lengths of the bar-shaped members were all 15 mm, while the thicknesses thereof were all 2 mm, with mutual distances of 16 mm. The actuator input parts were rectangular, as shown in Fig. l, with dimensions of 20x10x20 mmxmmxmm.

It will be clear to a skilled person that it is possible to change or modify the embodiment of the device according to the invention as shown, without departing from the concept of the invention or the scope of protection. Thus, it is for instance possible that the bar-shaped members of the first coupling means 120, 120', 220, 220' have a different length than the bar-shaped members of the second coupling means 130, 130', 230, 230'. It is also possible that the number of bar-shaped members of the second coupling means 130, 130', 230, 230' is greater than two. It is also possible that the first coupling means 120, 120', 220, 220' comprise several, preferably parallel, bar-shaped members. It is also possible that the first coupling means 120, 120", 220, 220' comprise bar-shaped members directed in the Z-direction.

Further, the two data processing devices 520 and 530 can be formed by several cooperating modules or be integrated into a single device, or the first data processing device 520 and the actuator driving unit 510 can be integrated into a single device, or the two data processing devices 520 and 530 together with the actuator driving device 510 can be integrated into a single device.

Further, the resilient means 150 can be replaced by a piezoelectric actuator. In such a case, the two oppositely mounted actuators are driven in combination: a "basic" signal for both actuators defines the bias, while a displacement is

effected by presenting to one actuator a greater and to the other a smaller voltage. Such a configuration is referred to by the term "push/pull stage" .