MOLDING OF OPTICAL ELEMENTS USING A TOOL HAVING AN OVERFLOW VOLUME

FIELD OF THE INVENTION

The invention is in the field of manufacturing miniature optical or mechanical elements, in particular refractive optical elements or diffractive micro-optical elements, by means of a replication process that includes embossing or moulding steps. More concretely, it deals with a method of replicating an optical element and a replication tool therefor.

BACKGROUND OF THE INVENTION

Replicated optical elements include diffractive and/or refractive micro-optical elements for influencing an optical beam in any pre-defined manner, refractive elements such as lenses, potentially at least partially reflecting elements etc..

When optical elements are produced by replication, there is often a basic configuration involving a substrate and replication material on a surface thereof, which replication material is shaped and hardened in the course of a replication process. Often, the dimension perpendicular to the named substrate surface — the thickness or height of the replicated structures, also termed z-dimension — is important and must be well-defined and controlled. Since the other dimensions of the element are defined by the replication tool - this being the nature of the replication

process - also the volume of the replicated element is well defined. However, small volumes of dispensed liquid or viscous material are generally difficult and costly to control. Since elements that are only partially filled are defective and lost, it is therefore advantageous to dispense excess replication material. By this, one makes sure that also for replication material volumes that fluctuate between different elements, no or only few elements are lost.

Of special interest are the wafer-scale fabrication processes, where an array of optical elements is fabricated on a disk-like ("wafer-") structure, which subsequently to replication is separated ("diced") into the individual elements or stacked on other wafer-like elements and after stacking separated into the individual elements, as for example described in WO 2005/083 789. 'Wafer scale' refers to the size of disk like or plate like substrates of sizes comparable to semiconductor wafers, such as disks having diameters between 2 in and 12 in. In conventional wafer-scale replication processes, replication material for the entire, wafer-scale replica is disposed on the substrate in a single blob. However, there might be areas sideward of the element where replication material is not wanted in later replication steps. In certain applications, the fabricated elements must for example be used in combination with other elements, and the residual material will impair the function of the combined structure. In a co-pending application "Method and Tool for Manufacturing Optical Elements" by the same inventors and filed on the same day as the present application, an array replication method is disclosed according to which for every optical element or sub-group of optical elements to be created, a blob of replication material is dispensed in an array like manner, either on the substrate or on the tool.

In such an array replication process, excess material will ooze out sideward from the element volume. For example, miniature optical lenses may be replicated above the surface of a wafer carrying semiconductor chips each embodying a CCD or CMOS- camera sensor array. The residual material, if it covers critical areas, may interfere

with further processing steps of the stack comprising the semiconductor wafer and the lenses, e.g. bonding.

WO 2004/068198 by the same applicant, herewith incorporated by reference in its entirety, describes a replication process for creating micro-optical elements. A structured (or micro-structured) element is manufactured by replicating/shaping (moulding or embossing or the like) a 3D- structure in a preliminary product using a replication tool. The replication tool comprises a spacer portion protruding from a replication surface. A replicated micro- optical element is referred to as replica.

The spacer portions allow for an automated and accurate thickness control of the deformable material on the substrate. They may comprise "leg like" structures built into the tool. In addition the spacers prevent the deformation of the micro optical topography since the spacers protrude further than the highest structural features on a tool.

The spacer portion is preferably available in a manner that it is 'distributed' over at least an essential fraction of the replication tool, for example over the entire replication tool or at the edge. This means that features of the spacer portion are present in an essential fraction of the replication tool, for example, the spacer portion consists of a plurality of spacers distributed over the replication surface of the replication tool. The spacers allow for an automated and accurate thickness control of the replication material layer.

The replication process may be an embossing process, where the plastically deformable or viscous or liquid replication material for the product to be shaped is placed on a surface of a substrate, which can have any size. In the embossing step, the spacer portions abut against the top surface of the substrate. Said surface thus serves as a stop face for the embossing.

- A -

For these reasons, the replication process described in WO 2004/068198 is one particularly advantageous possibility of controlling the thickness (height, z- dimension) of the replicated elements. Other ways of controlling the z-dimension include measuring the distance between a tool plane and a substrate plane and actively adjusting this distance at different places by a robot.

For the reasons stated above, the embossing step causes residual material to remain in the areas between the elements, and for example also around the periphery of each of the elements. If the replication tool comprises a spacer portion, this may also be true for the spacer area surrounding an element.

DESCRIPTION OF THE INVENTION

It is therefore an object of the invention to create a method of replicating an element and a replication tool of the type mentioned initially, which overcomes the disadvantages mentioned above.

According to a first aspect of the invention, a method of manufacturing an element by means of a replication tool is provided, the method comprising the steps of

• providing a replication tool that defines the shape of the element;

• providing a substrate;

• pressing the tool and the substrate against each other, with a replication material in a liquid or viscous or plastically deformable state located between the tool and the substrate;

• confining the replication material to a predetermined area of the substrate, which predetermined area exceeds the desired area of the element on the substrate, in at least one direction along the surface of the substrate by less than a predetermined distance; • hardening the replication material to form the element.

The replication material is confined between the tool and the surface of the substrate. By confining the replication material to only part of the substrate surface, the resulting element will, after hardening by e.g. curing only cover part of the substrate. The element will not extend to cover the substrate in predetermined areas, leaving them free for e.g. bonding.

Preferably, the replication tool comprises a plurality of sections each defining the (negative of) the shape of an component (such as an optical element, for example a lens) later to be separated from the other elements by dicing the substrate or an assembly including the substrate into the individual elements. Then, the confining of replication material to a predetermined area of the substrate for example includes confining the replication material to a plurality of regions, each region around a replication section, the regions preferably not overlapping. For example, the replication sections may be present as an array of identical replication sections, where around each of the replication sections the replication material is confined to a region.

The replication tool may comprise a spacer portion. In such a tool, at least one cavity of the tool defines a replication surface with negative structural features, being a negative of at least some of the structural features of the element to be produced. The cavity contains the element volume and may additionally comprise at least one buffer and/or overflow volume. The spacer or spacer portions protrude from the replication surface. In the replication process, the spacer or spacer portions abut against the substrate and/or float on a thin basis layer of replication material.

The force by which the tool and the substrate are pressed against each other may be chosen based on specific requirements. For example, the force may be just the weight of the replication tool lying, by way of spacer portions abutting the substrate surface and/or floating on a thin basis layer of replication material, on the substrate.

Alternatively, the substrate may lie on the replication tool. The force may according to yet another alternative be higher or lower than the weight and may for example be applied by a mask aligner or similar device, which controls the distance of the substrate and the replication tool during the replication process.

Before the replication tool and the substrate are brought together for the replication process, replication material in a liquid or viscous or plastically deformable state is placed on the replication tool and/or the substrate. The replication tool may, as mentioned above, comprise a plurality of sections each defining an element to be replicated. Then, preferably the method comprises applying a (possibly pre-defined) volume of replication material locally and individually, at laterally displaced positions, each position corresponding to one section, to at least one of the tool and the substrate prior to pressing the tool against the substrate. This allows providing a plurality of cavities, each corresponding to an optical element, with an optimal amount of replication material. By this, the volume of surplus replication material that must be removed or diverted from the critical areas is reduced or eliminated, as compared to the case where a plurality of elements would be formed from a single blob of replication material.

While the replication tool and the substrate are in the replication position - in which the replication tool and the substrate are brought together, for example the replication tool is placed on the substrate - the replication material is hardened. Depending on the replication material chosen, it may be hardened by curing, for example UV curing. As an alternative it may be hardened by cooling. Depending on the replication material chosen, other hardening methods are possible. Subsequently, the replication tool and the replication material are separated from each other. For most applications, the replication material remains on the substrate. The optical element typically is a refractive or diffractive optical element, but also may e.g. also have a micromechanical function at least in regions.

The element volume covers a part of the substrate and constitutes the functional part of the element. The remainder of the cured replication material may fill a volume at the sides of the element, i.e. the region of space adjacent to both the substrate and the functional part of the element, and does not interfere with the function of the element. The invention allows controlling how far the replication material may move along the substrate at each side of the element volume.

In a preferred embodiment of the invention, the flow of the replication material is controlled and/or limited by capillary forces and/or surface tension. This exploits the property of geometric features to further or to hinder the flow of the replication material between the tool and the substrate.

As an example, the replication tool may be chosen to comprise a plurality of cavities each defining the shape of one element or a group of elements, each cavity being limited, at least in one lateral direction, by a flat section. An inner edge is formed between the cavity and the flat section. The replication tool further comprises a plurality of overflow volumes or one contiguous overflow volume between the cavities. And an outer edge is formed between the flat section and the overflow volume. The dispensed replication material (per cavity) is chosen to be larger than the volume of the cavity. The flat section then serves as floating (non-contact) spacer, which preferably surrounds the cavity. The outer edge constitutes a discontinuity stopping a flow the replication material. Without such discontinuities, capillary forces would cause the replication material to eventually drain the replication material from the element volume.

The cavity, in this example, may for example consist of the element volume only. It may be dome-shaped so that the element is a convex refractive lens adjacent to which a thin base layer is formed, the base layer being what replication material remains underneath the floating spacer.

Even in the case of a cylinder symmetric optical element, the shape of the flat section, when seen in the direction perpendicular to the substrate surface, e.g. along a central axis of the element, may be asymmetrical so that a bulge of replication material forming along the outer edge in the overflow volume is farther away from the replication element towards one side of the element than towards an other side.

Here and in the following, for the sake of convenience, the dimension perpendicular to the surface of the substrate, which comprises an essentially flat surface - is denoted as "height". In actual practice, the entire arrangement may also be used in an upside down configuration or also in a configuration where the substrate surface is vertical or at an angle to the horizontal. The according direction perpendicular to the surface is denoted z-direction. The terms "periphery", "lateral" and "sides" relate to a direction perpendicular to the z-direction.

In another example control of the flow is done by a cavity in the tool defining the shape of the element, and the cavity including a buffer volume along at least one side of the element, which buffer volume is separated from the element volume by an inner edge. Furthermore, the predetermined volume of replication material applied individually to the element volume of the cavity is smaller than the volume of the cavity. This causes the inner edge to limit the flow of the replication material into the buffer volume by capillary forces acting at the inner edge and by surface tension.

Especially, the predetermined volume of replication material may be about the volume of the element volume (or slightly smaller or slightly larger). The element volume is the volume of the functional element, extending from the outer shape of the element defined by the tool on one side to the substrate on the other side. The replication material will then be stopped by fluid forces acting at the inner edge from flowing into the buffer volume.

In yet another preferred embodiment of the invention, when pressing the tool against the substrate, an inclined spacer displaces the replication material towards the element volume, and in particular a buffer volume adjacent to the element volume. The inclined spacer has an inclined surface that is to be brought into contact with the surface of the substrate. The inclined surface, when no pressure is applied, touches the substrate at an outer periphery, and in regions closer to the element volume gradually moves away from the substrate. When, during embossing or moulding, pressure is applied to the tool, the tool, being slightly elastic, is deformed, and the inclined surface causes replication material to be displaced from under the inclined spacer.

In a preferred embodiment of the invention, the method comprises the further step of • confining the flow of the replication material towards at least one side of the tool by a contact spacer that touches the substrate; and • enabling the flow of the replication material towards another side of the tool by an overflow channel.

This allows diverting the replication material away from the critical areas and guiding it to an overflow volume located in a noncritical area.

Also according to the invention, a replication tool for replicating an element from a replication material is provided, the replication tool comprising a replication side, a plurality of cavities on the replication side, each defining the shape of one element or a group of elements, the replication tool further comprising at least one spacer portion, protruding, on the replication side, from the cavities, the replication tool further comprising means for confining the replication material to a predetermined area of the tool, when the tool is pressed against a substrate, which predetermined area exceeds the desired volume of the element in at least one direction along the surface of the substrate by less than a predetermined distance.

Such means for confining the replication material, or flow confining features are constituted by the inner edge, the buffer volume, the outer edge, the spacer and the inclined spacer; each of them alone, or several of them in combination. They may be combined to form a "multi-tiered" flow confinement, which, according to the amount of replication material actually present, stops the flow at an earlier or a later limit. This allows controlling the flow despite inaccuracies when dispensing the replication material to individual cavities or onto corresponding individual locations on the substrate.

In other words, the cavity comprises an element volume and a further volume, at a periphery of the element volume, the boundaries of the further volume comprising discontinuities for selectively inhibiting and/or enabling capillary flow of the replication material when pressing the tool against the substrate, with the replication material in between.

The discontinuities for example are, for circular optical elements, also circular in shape and concentric. For other shapes of optical elements, e.g. rectangles or rounded rectangles, the consecutive discontinuities may follow the shape of the optical element at increasing distances.

In a preferred embodiment of the invention, the discontinuities lie between ridges and recesses formed in the replication tool. Thus, the discontinuities are constituted e.g. by the edges between circular (or rectangular etc, see above) ridges and channels formed around the section of the replication tool that defines the optical element. A series of consecutive ridges and channels therefore defines a quantisation of the extension of the surplus replication material, since the outward flow of the replication material is inhibited or stopped at each edge or discontinuity, and continues only if the volume of the replication material in relation to the volume of the element volume exceeds a certain limit.

In order to minimise the area surrounding the optical element that remains covered by nonfunctional replication material, a floating spacer surrounding the element volume and defining its outer border is for example made as thin as possible while still providing the function of a spacer, i.e. providing sufficient support for the tool. Furthermore, the recess or several recesses outside the floating spacer, or outside the outer edge of the element volume (when there is no floating spacer), is preferably made as deep as possible, e.g. up to the depth of the element volume. In consequence, the volume defined by the recess is increased, and the volume of replication material it can absorb before the material spills over into the next recess is also increased.

The volume of each circumferential recess or channel is preferably correlated with the precision with which the volume of the replication material deposited dropwise can be controlled. For example, if the latter volume can be controlled to a high degree, then it is known that the size or volume of the surplus material will only vary within narrow bounds. In consequence, a recess is preferable located and sized such as to cover this variation in surplus size or volume. That is, with the minimum expected surplus size, according to the deposition accuracy, the recess will not be filled at all, and for the maximum surplus size the recess will be filled just up to its limit. In other words, the dimensions of means for limiting the flow of the replication material (i.e. the dimensions of discontinuities or edges and of the intervening recesses and their volumes) are designed in accordance with expected values of the volume of replication material being applied.

Depending on other constraints, the volume of the recess is adjusted by selecting the depth and the width of the recess. The depth is e.g. limited by the process for creating the replication tool, and the width of the recess is limited by design constraints limiting the overall size of the optical element plus surplus material. So the overall design of the recess and drop deposition constitutes an optimal selection of the

interrelated features of drop deposition precision and recess geometry, according to optimisation criteria that depend on the individual product.

In a further preferred embodiment, the replication tool comprises a spacer dimensioned to stop the flow of the replication material by touching the substrate at one side of the cavity; and an overflow channel enabling the flow of the replication material towards another side of the cavity.

In a further preferred embodiment, the replication tool comprises a buffer volume at at least one side of the element volume defined by the cavity, the buffer volume and the element volume defining, at their common boundary, an inner edge for inhibiting the flow of the replication material into the buffer volume.

In a further preferred embodiment, the replication tool comprises further edges in the surface of the buffer volume for inhibiting the flow of the replication material into the buffer volume. The further edges follow the shape of the inner edge at least roughly in parallel curves.

The tool comprises a plurality of cavities, thus preferably allowing for the simultaneous manufacturing of an array of elements on a common substrate. This common substrate preferably is part of an opto-electronic or micro-opto-electronic assembly comprising optical and electronic elements produced on a wafer scale and later diced into separate units.

Further preferred embodiments are evident from the dependent patent claims. Features of the method claims may be combined with features of the device claims and vice versa.

The replica (for example a micro-optical element or micro-optical element component or an optical micro-system) may be made of epoxy. The hardening step,

which is done while the replication tool is still in place - may then be an UV curing step. UV light curing is a fast process that allows for a good control of the hardening process. The skilled person will know other materials and other hardening processes.

"Optical" elements include elements that are capable of influencing electromagnetic radiation not only in the visible part of the spectrum. Especially, optical elements include elements for influencing visible light, Infrared radiation, and potentially also UV radiation. The word "wafer" in this text does not mean any restriction as to the shape of the substrate.

BRIEF DESCRIPTION OF THE DRAWINGS

The subject matter of the invention will be explained in more detail in the following text with reference to preferred exemplary embodiments, which are illustrated in the attached drawings, which schematically show:

Figures 1 and 2 cross sections through a tools placed on a substrate;

Figure 3 an elevated view of the arrangement of Figure 2;

Figure 4 an example of an alternative geometrical shape of a transition between a buffer volume and an overflow volume;

Figures 5 - 9 cross sections through further tools;

Figure 10 an elevated view of the arrangement of Figure 9;

Figures 11 - 13 cross sections through further tools; and

Figure 14 a flow diagram of the method according to the invention.

The reference symbols used in the drawings, and their meanings, are listed in summary form in the list of reference symbols. In principle, identical parts are provided with the same reference symbols in the figures.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Figure 1 schematically shows a cross section through a tool 10 placed on a substrate 12. The tool 10 forms a cavity 8 that defines the shape of the element to be formed by an element volume 1. In the shown case, the optical element is simply a refractive lens. The element volume 1 lies between the tool 10 and the substrate 12. It is surrounded by a protruding element of the tool 10 which here is denoted as floating spacer 14. A flat surface 17 of the spacer is runs approximately parallel to the surface of the substrate 12 and here is at a distance of about 5μm to 15μm therefrom. Underneath the floating spacer 14, between the flat surface 17 and the substrate 12, a small buffer volume 3 forms. Between the element volume 1 and the buffer volume 3, the tool 10 comprises an inner edge 2. Between the buffer volume 3 and an overflow volume 5, the tool 10 comprises an outer edge 4.

The main function of the floating spacer 14 is to pull out excess materially capillary forces. The flow stops at the outer edge 4 and forms a bulge 18 and therefore prevents that the element volume 1 is emptied by the capillary forces. In this way, the width of the floating spacer 14 and the shape and shape and size of the overflow volume 5 define where excess material is to go. Therefore, by keeping the replication material volume below a certain maximum volume, the replication material is confined.

The inner edge 2 constitutes a first discontinuity, stopping the flow an outer boundary of the replication material 13, as is also shown in following Figures. The outer edge 4 constitutes a second discontinuity, stopping the replication material 13 from flowing to the buffer volume 5 adjacent to the buffer volume 3. Without such discontinuities, capillary forces would cause the replication material 13 to continuously flow along the channel formed by the buffer volume 3, eventually draining the replication material 13 from the element volume 1.

Figure 2 shows a variation of the above principle. In this variation, the floating spacer 14 surrounding the element volume 1 is asymmetric. By this, the excess material can be transported to areas where it is not disturbing other processes. A top view of the configuration of Figure 2 is shown in Figure 3. The bulge of replication material (shown in Figures 1 and 2 but not in Figure 3, since the latter only shows the tool 10 without the replication material) extending around the outer edge 4 may for example be approximately constant in its cross section. By the asymmetric shape of the floating spacer, the length of the outer edge 4 is increased. For these reasons, the asymmetric solution allows to confine by the replication material especially well in one desired direction, corresponding to the lower left corner in the sketched configuration, as may be especially desired in configurations with an off-center optical element.

The tool preferably comprises, as it may in the embodiment of Figure 1 and in all of the hereafter-described embodiments, multiple sections each corresponding to an element to be replicated. The sections are arranged array-like, for instance in a grid with grid 11 lines corresponding to cutting or dicing lines for later separation of the substrate 12 carrying the manufactured optical elements or corresponding to bonding areas where other elements are later to be bonded to.

As shown in Figures 2 and 3, an asymmetry of material flow between different directions can be implemented way is based on different distances. However, it is also possible to influence the replication material flow by other means such as different surface properties at different locations or by geometrical shape. The outside portions of the spacers 14 can be formed in a way so that differing surface tensions can be used to control the excess material. An example is shown in Figure 4. The spacer 14 at one side comprises a geometrical feature 20 that causes the flow towards this side to be different from the flow towards the other side.

Figure 5 shows a cross section of a tool 10 with replication material 13 just filling the element volume 1 and being contained by the discontinuity of the inner edge 2 between the element volume 1 and the buffer volume 3. The length of the buffer volume 3 preferably lies in the range of 100 to 300 or 500 or 800 micrometers.

In Figure 5, the buffer volume 3 is within the cavity 8. Also, the z-dimension and thus the element height and ultimately the element volume are fixed by a contact spacer 9 surrounding the cavity 8. The contact spacer 9 may for example be of the kind described in WO 2004/068198. Figure 5 thus shows an example, where the replication material is confined by a combination of an exact dispensing of the replication material volume corresponding to the element volume 1 (or to a slightly smaller or larger volume) and the effect of surface tension in combination with the impact of an edge 2.

The embodiment relying on a more or less exact dispensing of the replication material and a geometrical element (such as an edge) limiting the replication material flow in at least one direction my means of surface tension and/or capillary forces thus not rely on there being a contact spacer surrounding the cavity. This is illustrated in Figure 6. Figure 6 shows part of a cross section of a tool 10 in which on one side, an (optional) elevated spacer section 14 is shown. In such an embodiment, the z-dimension is defined in an other way, for example by contact spacers on an other side (not shown) or at an other, for example peripheral lateral position, by active distance adjusters and/or controllers, or other means.

Figure 7 shows a cross section of a tool 10 with further edges 21 formed at the surface of the buffer volume 3. These further edges 21 confine the flow of the replication material 13, and come into action depending on the total volume of the replication material 13, which may vary when applying the replication material 13 individually with a doser, such as a dosing syringe, to the cavity 8, to the substrate 12 at locations opposite to the cavities 8, or generally, if no spacers and thus no cavities

are present, on the lateral positions of the elements to be replicated, either to the substrate or to the replication tool or to both.

Figure 8 shows part of a cross section of a tool 10 that has an inclined spacer 15 prior to being pressed against the substrate 12. The arrow shows the direction of flow of the replication material 13 under the inclined spacer 15, as it is being compressed.

Usually, the weight of the replication tool, with optional additional weights, is sufficient to generate the required pressure. The buffer volume 3 takes up the replication material 13 displaced from under the inclined spacer 15. In this embodiment, it is the inclined spacer that limits the flow.

Figure 9 schematically shows a cross section through a tool 10 placed on a substrate 12. Figure 10 shows a corresponding elevated view. The tool 10 comprises a cavity 8 that defines the shape of the element to be formed by an element volume 1. The element volume 1 lies between the tool 10 and the substrate 12, and is surrounded by a buffer volume 3. Between the element volume 1 and the buffer volume 3, the tool 10 comprises an inner edge 2. Between the buffer volume 3 and an overflow volume 5, and between the buffer volume 3 and a free volume 6, the tool 10 comprises an outer edge 4, 4'. The buffer volume 3 constitutes an outlet or overflow channel 16 for surplus material, in the case that the amount of replication material 13 exceeds the volume of the element volume 1.

For cases in which a large volume tolerance is required, the cavity 8 comprises an overflow volume 5 on one side of the element volume 1. On the other side, the outer edge 4, or the free volume 6 or the spacer 9 defines the limit of flow for the replication material 13, keeping the replication material 13 away from critical areas of the substrate. This outer edge 4, together with the outer limit of the overflow volume 5, defines a predetermined area 7 that gives the maximum area of substrate 12 that can be covered by the replication material 13.

The outer edge 4, 4' is shaped differently between the transition 4 from the buffer volume 3 to the free volume 6 on the one hand and the transition 4' from the buffer volume 3 to the overflow volume 5 on the other hand, so that surface tension and/or capillary forces cause excess replication material to flow into the overflow volume 5 but not to the free volume 6. For example, the outer edge 4, 4' may be sharper at the transition 4 to the free volume 6 and rounder at the transition 4' to the overflow volume 5.

The tool 10 here rests on (optional) contact spacers 9 placed against the substrate 12. The function of the free volume 6, which is not to be filled by replication material, is, in combination with the outer edge 4, to stop the flow of the replication material and also to thereby prevent it from flowing underneath the contact spacer 9. Depending on the viscosity of the replication material, surface tension and capillary forces, this may not be necessary, and the flow may be stopped by the contact spacer itself. In that case, the contact spacer may be immediately adjacent to the element volume 1, without there being a need for the buffer volume and the free volume 6.

Since the overflow volume 5 is higher than the buffer volume 3, following a discontinuity or step in height at the outer edge 4, capillary forces are no longer relevant (For the sake of convenience, the dimension perpendicular to the surface of the substrate 12 is denoted as "height". In actual practice, the entire arrangement may also be used upside down.). The overflow volume 5 will simply be filled in accordance with the surplus replication material 13 volume.

In an exemplary embodiment of the invention, a diameter of the element volume 1 is between 1 and 2 millimetres and has a height around 250 micrometers, the height of the buffer volume 3, i.e. the distance between the cavity 8 and the substrate 12 in the region of the buffer volume 3 is ca. 10 micrometers, the length of the buffer volume 3, i.e. the distance from the inner edge 2 to the outer edge 4 is ca. 50 to 200 micrometers.

Figures 11 through 13 show cross sections through further tools which comprise buffer volumes with recesses adapted to the expected size or volume of individual drops of replication material. Figure 11 shows a tool 10 similar to that of Figure 7, that is, without a floating spacer, in which the further edges 21 constitute the boundaries between ridges 23 and recesses 19', 19". The ridges 23 and recesses 19', 19", as in the other figures, run around the element volume 1, be it as concentric circles or following the contour of a noncircular optical element. In the latter case, the width and depth of each circumferential ridge or channel formed in this manner preferably remains constant around its the circumference. In Figure 11, a first, inner recess 19' has a larger volume, since its width and/or depth is larger than that of a second, outer recess 19". The inner recess 19' can accept a relatively large volume of surplus replication material and is preferably located and sized such that the volume of replication material required to reach a first, inner edge 21' of the inner recess 19' corresponds to an expected minimum volume deposited by a drop deposition device (with a given probability); and

- the volume of replication material required to reach a second, outer edge 21" of the inner recess 19' corresponds to an expected maximum volume deposited by a drop deposition device (with a given probability);. For cases in which the replication material exceeds the expected maximum volume (with low probability, but not to be ruled out completely), the second, outer recess 19" may be arranged to create a limit according to its edges. This arrangement of ridges may be combined with an overflow volume 5, indicated by dashed lines, or not.

Figure 12 shows a tool 10 with recesses 19', 19", 19"' and ridges 23 arranged and dimensioned in an analogue fashion as in Figure 11, but separated from the element volume 1 by a elevated (floating) spacer 14, as in Figure 1. As a variant of the recess/ridge arrangement of Figure 11, a further recess 19'" is arranged inside the larger recess 19', in order to account for the low probability cases in which the

suφlus material is less than the expected minimum volume of deposited material, and to provide for a defined contour of the replication material. Again, this arrangement of ridges may be combined with an overflow volume 5, indicated by dotted lines, or not.

Figure 13 shows a tool 10 with an inclined or sloped surface 22 extending outwards from an elevated spacer 14 with increasing height. As a result, the relation between the distance from the element volume 1 and the volume of replication material required to fill the cavity under the inclined surface 22 up to this distance is non- linear. This nonlinearity is not only caused by the area covered increasing with the square of the radius, but further also because of the height of the sloped surface increasing with the radius. Depending on the viscosity and other flow properties (in particular adhesion vs. cohesion) of the replication material, this geometry can be of advantage. This geometry may be combined with regular or irregularly sized recesses and ridges as in Figures 11 and 12, and with or without an overflow volume 5. A geometry with an inclined surface may also be used in set-ups without the contact spacers 9, for example it may be used in configurations like the one in Fig. 1 as a surface of the surplus volume 5.

Figure 14 shows a flow diagram of the method described.

While the invention has been described in present preferred embodiments of the invention, it is distinctly understood that the invention is not limited thereto, but may be otherwise variously embodied and practised within the scope of the claims.