NL2004052C2

NL2004052C2 - Capacitive sensing system.

Info

Publication number: NL2004052C2
Application number: NL2004052A
Authority: NL
Inventors: Guido Boer; Stijn Willem Herman Karel Steenbrink; Joannes Jacobus Baar; Kaustubh Prabodh Padhye; Robert Mossel; Niels Vergeer
Original assignee: Mapper Lithography Ip Bv
Priority date: 2010-01-05
Filing date: 2010-01-05
Publication date: 2011-07-06

Description

-1-

CAPACITIYE SENSING SYSTEM BACKGROUND OF THE INVENTION

1. Field of the Invention

[0001] The present invention relates to a capacitive sensing system comprising two or 5 more capacitive sensors for measuring distance, in particular capacitive sensors for measuring distance to a target in a lithography apparatus.

2. Description of the Related Art

[0002] Charged particle and optical lithography machines and inspection machines are used to expose patterns onto wafers and other targets, typically as part of a semiconductor 10 device manufacturing process. In a lithography system a wafer is usually exposed at multiple locations by optical or particle exposure beams generated by the lithography machine. The wafer is usually positioned on a wafer table and multiple exposures are typically achieved by controlled displacement of the wafer table with respect to a stationary electron/optics column. The exposures are typically performed continuously on the wafer 15 surface.

[0003] The wafer surface which is to be exposed is almost never completely flat. A typical wafer may have a bow in it of up to 50 pm without clamping to the wafer table. Apart from the wafer bow the wafer surface may have other non-uniformities over its surface. The wafer bow and other non-uniformities result in height variations in the wafer 20 surface. To achieve the extremely high precision required of modem lithography machines, it necessary to correct for this height variation to maintain the wafer surface that is exposed in the focal plane of the projection lens used to focus the optical or particle exposure beams onto the wafer.

[0004] The wafer table that holds the wafer may be adjusted to compensate for these 25 variations in height of the wafer surface. The height of the wafer table may be adjusted to bring the wafer surface to be exposed into the focal plane of the projection lens. Control of the wafer table height may be accomplished using signals transmitted from sensors which measure the height of the wafer surface, e.g. the distance between the projection lens and the wafer surface. Highly sensitive sensors are required to ensure correct control of wafer 30 position at the extreme precision required for modem lithography machines. Various types -2- of sensors have been used for this type of application, including capacitive probes.

However, the existing capacitive probes and associated measurement and control systems have suffered from several drawbacks.

[0005] Existing capacitive sensors are relatively large, both in height and sensor area.

5 Figs. 1A and IB show the structure of a prior art capacitive sensor. Fig. 1A shows a cross-sectional view and Fig. IB shows an end view of the sensor probe. A conductive sensing electrode 2 is surrounded by a conductive guard electrode 3. An insulating layer 4 separates the two electrodes and another insulating layer 5 may be used to separate the guard electrode 3 from the housing 6. An electrical cable 7 and connector 8 connects the sensor to 10 a signal processing system to derive the desired final measurement signal. The operating range of the sensor is dependent upon the sensing area under the sensing electrode 2. The guard electrode 3 is set at the same potential as the sensing electrode to confine the electric field within the sensing area to generate a relatively uniform electric field between the sensing electrode 2 and the target 9. This type of construction leads to a relatively tall 15 sensor, generally about 20 mm in height, and a relatively large sensing electrode.

[0006] The relatively large height and width of the sensors requires that the sensors need to be located relatively far from the projection lens, introducing errors due to variation in the relative positioning of the sensors and the projection lens due to manufacturing tolerances and thermal expansion. The relatively large size of existing capacitive probes also 20 requires that individual sensors in multi-sensor configurations are spaced relatively far apart, reducing the spatial resolution of the sensing system so that non-uniformities in the wafer surface occurring over a small area of the wafer surface may not be detected. The relatively wide spacing also results in a slower measurement process, reducing throughput of a lithography machine using these systems. The requirement to calibrate existing sensors in 25 combination with the sensor wiring requires recalibration whenever a sensor is replaced, making the replacement complex, time-consuming, and expensive. The cost of existing capacitive probes is also very high.

BRIEF SUMMARY OF THE INVENTION

[0007] The invention seeks to solve or reduce the above drawbacks to provide an 30 improved capacitive sensor.

-3-

[0008] In one aspect of the invention, the invention relates to a capacitive sensing system comprising two or more capacitive sensors, one or more AC power sources for energizing the capacitive sensors, and a signal processing circuit for processing a signal from the capacitive sensors. The sensors are arranged in pairs and the one or more AC power 5 sources are arranged to energize a first sensor of a pair of sensors with an alternating current or voltage 180 degrees out of phase to a current or voltage for a second sensor of the pair of sensors. The system may comprise four or more capacitive sensors arranged in pairs, wherein each sensor of a first pair of sensors is energized by a current or voltage which is out of phase with respect to a current energizing each sensor of an adjacent second 10 pair of sensors.

[0009] The system may also include an active guarding circuit for energizing a guard electrode of at least one of the capacitive sensors with the same potential as present on the sensing electrode of the sensor. The system may also comprise a cable for connecting at least one of the capacitive sensors to the signal processing circuit, and an active guarding 15 circuit for energizing a conductor of the cable with the same potential as present on the sensing electrode of the sensor. The cable may be a coaxial cable with a center conductor and an outer conductor, the center conductor for electrically connecting the current source to the capacitive sensor and the outer conductor for electrically connecting to the active guarding circuit. The cable may be a triaxial cable with a shield conductor grounded at an 20 end of the cable remote from the sensors. Furthermore, each of the capacitive sensors of the system may comprise a thin film structure having the features described herein, or an array of sensors as described herein.

[0010] The signal processing circuit for processing a signal from the capacitive sensors may comprise a synchronous detection circuit having an oscillator and phase shift circuit for 25 generating a signal to drive the one or more AC power sources. The system may further comprise an automated calibration algorithm for adjustment of the phase shift circuit, and the automated calibration algorithm may adjust a phase shift of the phase shift circuit to maximize an output of mixer circuit, and may adjust a phase shift of the phase shift circuit to be substantially equal to a phase shift occurring in the sensing system, and may adjust a 30 signal for driving the one or more AC power sources.

-4-

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Various aspects of the invention will be further explained with reference to embodiments shown in the drawings wherein:

[0012] Fig. 1A is a cross sectional view of a capacitive sensor; 5 [0013] Fig. IB is an end view of the capacitive sensor of Fig. 1A;

[0014] Fig. 2 is a simplified schematic diagram of a parallel plate electrode arrangement;

[0015] Fig. 3 is a diagram of a capacitive sensor probe and grounded conductive target;

[0016] Fig. 4 is a diagram of two capacitive sensor probes in a differential measurement arrangement with a grounded conductive target; 10 [0017] Fig. 5 is a cross sectional view of a capacitive sensor comprising a thin film structure;

[0018] Fig. 6 is a cross sectional view of a thin film sensor with protective layers;

[0019] Fig. 7 is a top view of the sensor of Fig. 6;

[0020] Fig. 8A is a top view of a thin film sensor with a square sensing electrode; 15 [0021] Fig. 8B is a cross sectional view of the sensor of Fig. 8A;

[0022] Fig. 9A is a top view of a thin film sensor with a circular sensing electrode;

[0023] Fig. 9B is a cross sectional view of the sensor of Fig. 8A;

[0024] Figs. 10A to 10D are cross sectional views of thin film capacitive sensors;

[0025] Fig. 11 is a top view of a sensor with connecting lines and contact pads; 20 [0026] Figs. 12A and 12B are cross sectional views of contact pad structures;

[0027] Figs. 13A to 13D are diagrams of sensors, connecting lines and contact pads formed on a common substrate;

[0028] Fig. 14 is a side view of sensors mounted on a lithography machine;

[0029] Figs. 15A and 15B are diagrams of a flex print connector; 25 [0030] Fig. 16A and 16B are cross sectional views of a projection lens stack of a charged particle lithography machine;

[0031] Figs. 17A to 17D are diagrams of a flexible printed circuit structure with multiple sensors and integrated flex print connectors;

[0032] Fig. 18 is another connection arrangement of sensors on a lithography machine; -5-

[0033] Figs. 19A and 19B are diagrams of an arrangement for mounting an integrated flexible printed circuit structure on a lithography machine;

[0034] Figs. 20A and 20B are diagrams of configurations of capacitive sensors on a mounting plate; 5 [0035] Fig. 21 is a schematic diagram of a sensor system and signal processing system;

[0036] Fig. 22A is a simplified circuit diagram of a high-impedance amplifier circuit with current source;

[0037] Fig. 22B is a simplified circuit diagram of a differential sensor arrangement with current source; 10 [0038] Fig. 23 is a simplified circuit diagram of a whetstone bridge arrangement with voltage source;

[0039] Fig. 24 is a simplified circuit diagram of a differential sensor arrangement with voltage source;

[0040] Fig. 25 is a simplified circuit diagram of a differential sensor circuit arrangement; 15 [0041] Fig. 26 is a simplified circuit diagram of a synchronous detector circuit;

[0042] Fig. 27 is a schematic diagram showing capacitances in a sensor system;

[0043] Fig. 28 is a simplified circuit diagram of an arrangement with a cable connecting a sensor to a signal processing circuit;

[0044] Fig. 29 is a simplified circuit diagram of another embodiment of a synchronous 20 circuit;

[0045] Fig. 30 is a simplified circuit diagram of an arrangement for processing signals from a differential pair of sensors;

[0046] Fig. 31 is a schematic diagram of a control system for positioning of a wafer for a lithography machine; and 25 [0047] Fig. 32 is a diagram of a sensor arrangement for use with the control system of

Fig. 31.

DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0048] The following is a description of various embodiments of the invention, given by way of example only and with reference to the drawings.

30 Theory of capacitive sensors -6-

[0049] A capacitive sensor uses a homogeneous electric field set up between two conductive surfaces. Over short distances, the applied voltage is proportional to the distance between the surfaces. Single-plate sensors measure the distance between a single sensor plate and an electrically conductive target surface.

5 [0050] Fig. 2 shows a parallel plate electrode arrangement. The capacitance between the two electrodes 11, 12 is given by the charge induced on one of the electrodes due to a potential difference between the two electrodes, divided by the potential difference, as represented in equation (1), C = -Q~. (1)

AV

10 [0051] The two parallel electrodes are separated by a distance d. Capacitance between the two electrodes is given by equation (2), neglecting the effects of field bending and nonhomogeneity of the dielectric, c = £oM (2) d where C is the capacitance between the two electrodes (F), A is the overlap area of the two 15 electrodes (m2), £„ is the permittivity of free space (8.85 x 10'12 F/m), £r is the relative permittivity of the medium between the electrodes, and d is the distance between the electrodes (m).

[0052] When an alternating electrical current source 13 is used to charge a parallel plate capacitor, a voltage potential develops between the electrodes depending on the impedance 20 of the electrodes. The impedance of a parallel plate capacitance is given by equation (3), Z = —ï— (3)

2nfC

where Z is impedance (£!),ƒ is frequency (Hz), and C is capacitance (F).

[0053] From equation (3) it can be seen that capacitive impedance is inversely proportional to the value of the capacitance and frequency of the signal applied to the 25 capacitor. In case of a capacitive sensor, the change in an electrical parameter (voltage or current) is measured which corresponds to the change in the impedance of the sensor. When the frequency of the signal applied to the sensor is kept constant, the impedance can be made inversely proportional to the change in capacitance. Equation (2) shows that the -7- capacitance is directly proportional to the overlap area of the sensor electrodes and inversely proportional to the change in distance between the electrodes. Combining equations (2) and (3) yields the equation: v =----i (4)

2?t/e0M

5 where i = current.

[0054] By keeping the electrode overlap area and the frequency of the electrical signal (current) applied to the sensor constant, the change in distance between the electrodes results in a change in the impedance of the capacitive sensor. The voltage across the sensor will be proportional to the impedance, and proportional to the distance (d) between the 10 sensor electrodes, enabling accurate measurement of the distance. Various measurement concepts may be used as described below.

Measurement principle for capacitive sensor

[0055] Fig. 3 shows a single capacitive sensor probe 1 measuring the separation distance to grounded conductive target 9. When supplied with an AC current, current will flow 15 along a path 15 from the sensor to the target through the sensor-target capacitance 16, and from the target to ground through the target-ground impedance 17. The accuracy of measurement of the distance from the sensor to the target depends on how accurately the sensor can measure the sensor-target capacitance 16. The capacitance of the target-ground impedance 17 will often greatly exceed the sensor-target capacitance 16, and may be over 20 100 times as much. This high capacitance results in a low impedance 17 so its effect on the sensor is small. However, variations the impedance 17 will affect the distance measurement and it is desirable to minimize this effect.

[0056] Fig. 4 shows an arrangement of two capacitive sensor probes la and lb for a differential measurement of the separation distance to target 9. The sensors are supplied 25 with an AC current offset by 180 degrees, so that current will flow along a path 18 from one sensor to the target through the sensor-target capacitance 16a, and from the target to the other sensor through the other sensor-target capacitance 16b. The differential measurement arrangement greatly reduces the influence of the target-ground capacitance and increases the sensitivity of the sensing system.

-8-

Structure of sensors

[0057] Fig. 5 shows a cross-sectional view of a capacitive sensor according to an embodiment of the invention comprising a thin film structure. The conductive sensing electrode 31 and conductive side guard electrode 32 are formed on or attached to an 5 insulating film 34. A conductive back guard electrode 35 is disposed on the back side of the insulating film 34. The gap 39 between the sensing electrode and guard electrode is narrow, typically a few tenths of micrometers, and may be an air gap or filled with an insulating material.

[0058] The electric field generated between the sensing electrode and the target bends 10 near the edges of the sensing electrode. The presence of a conductor near the edge of the sensing electrode can have a large and unpredictable effect on the electric field and thus on the sensor’s measurement. To avoid this situation (and make the sensor measurement more predictable and easier to model so the electric field can be calculated analytically) the sensing electrode is surrounded by a guard electrode which is energized by the same 15 potential as the sensing electrode. The guard electrode functions as a shield against external interference and also moves the electric field bending effects out of the sensing area under the sensing electrode, reducing parasitic capacitance. An electric field is generated between the guard electrode and the target on each side of the electric field between the sensing electrode and the target. There is no electric field generated between the sensing electrode 20 and guard electrode since they are at the same potential. This results in a substantially homogeneous electric field in the area under the sensing electrode while the field bending occurs at the outside edges of the guard electrodes.

[0059] The area of the sensing electrode 31 should be large in comparison to the distance separating the sensing electrode from the target. Also, the gap 39 between the sensing 25 electrode 31 and the side guard electrode 32 should be small in comparison to the distance between the sensing electrode and the target, and the width of the side guard electrode 32 should be large in comparison to the distance between the sensing electrode and the target. In one embodiment, the width of the sensing electrode is at least five times the distance between the sensor electrode and the target, the gap between the sensing electrode and the 30 guard electrode is less than or equal to one fifth of the distance between the sensor -9- electrode and the target, and the width of the guard electrode is at least five times the distance between the sensing electrode and the target. Following these comparative design rules provides an embodiment of the capacitive sensor design rules with a highly predictable capacitance, e.g. a predictability of the capacitance of 1 ppm. Predictability is here defined 5 as the relative error made if the ideal plate-distance capacitance formula of equation (2) above is used for calculating capacitance for finite electrode dimensions instead of infinite electrode dimensions.

[0060] In addition to the sensor capacitance Ci between the sensing electrode and the target which is being measured, the sensor has intrinsic parasitic capacitances C2 and C3 10 between each of the separate elements of the structure. The parasitic capacitances C2 and C3 are small in comparison to the capacitance Ci being measured. In the embodiment of Fig. 5, the parasitic capacitances include capacitance C2 between the sensing electrode and side guard electrodes and capacitance C3 between the sensing electrode and back guard electrode.

15 [0061] In one embodiment of the sensor, the capacitance Ci between the sensing electrode and target is 0.1 pF to 1 pF, while the parasitic capacitance C2 between the sensing and side guard electrodes is a factor of 100 to 1000 times smaller, typically of the order of 0.001 pF (i.e. 10"15 F). The parasitic capacitance C3 between the sensing electrode and back electrode is typically larger and dominates, typically about 1 to 1000 pF (i.e. 10"12 20 F to 10"9 F). The effect of these parasitic capacitances is reduced by energizing the guard electrode with the same potential as the sensing electrode. This can be accomplished by electrically connecting the side guard and sensing electrodes, or by the use of active guarding, discussed in more detail below. Active guarding may also be used for the back guard electrode.

25 [0062] For applications with lithography machines operating in a clean environment in a vacuum chamber, the sensors are preferably constructed to give off very low levels of contaminants when in the vacuum environment. A protective layer may be formed over the conductors for sensors used in this type of application, such as a Kapton polyimide film or similar protective film, particularly where materials are used which may contaminate the 30 vacuum environment. Fig. 6 shows a cross-sectional view of another embodiment of the -10- thin film sensor including protective layers 37 and 38, and Fig. 7 shows a top view of the sensor.

[0063] The sensing electrode 31 is formed in a circular shape, with a “C” shaped side guard electrode 32 almost completely surrounding the sensing electrode, leaving a narrow 5 gap between the two electrodes around the periphery of the sensing electrode 31. In this embodiment the side guard electrode 32 and back electrode 35 are optionally electrically connected, by means of an opening 37 in the insulating film 34 permitting the guard and back electrodes to come into contact. A single “C” shaped opening is used in this embodiment, although the other shapes could be used and/or multiple openings could be 10 used. Connecting the guard and back electrodes puts both electrodes at the same potential to eliminate the effect of any capacitance between them, and by using active guarding the effect of any capacitance between the guard and back electrodes and the sensing electrode can also be eliminated.

[0064] The sensor of Fig. 6 is shown attached to a plate 40. This is typically a part of the 15 structure of the equipment requiring the distance measurement, or may be attached to part of the structure, e.g. a spacer plate around the projection lens of a lithography machine where the sensor is measuring the distance between the projection lens and a wafer under the lens. The plate 40 may be conductive, and thus can also serve as a shield electrode for the sensor. Alternatively, a conductive shield electrode may be formed on the insulating film 20 38 as part of the structure of the sensor.

[0065] In Fig. 7, the internal sensing electrode 31 has one or more extensions forming connecting lines 41 for making an electrical connection from the sensing electrode to external signal processing circuits, and the side guard electrode 32 similarly has one or more extensions forming connecting lines 42 for making electrical connections. The sensing 25 electrode 31, side guard electrode 32, and connecting lines 41 and 42 are formed from thin films. In the embodiment shown, the electrodes 31 and 32 and connecting lines 41 and 42 are all in the same plane, and may be formed from the same thin film by depositing or forming the film removing portions using a laser, by etching, or other suitable removal techniques. The side guard electrode 32 substantially surrounds the sensing electrode 31, 30 leaving a small gap for the connecting lines 41 to extend outwards from the sensing -11- electrode to provide an electrical connection between the sensing electrode and the signal processing circuits. The connecting lines also add parasitic capacitances that should be taken into account in the design of the sensor.

[0066] In one embodiment, the electrodes 31, 32 and 35 are formed from conductive 5 layers about 18 micron thick, the insulating film 34 is about 25 micron thick, and the protective layers 37, 38 are about 50 micron thick. The thin film sensor may be constructed with a total thickness of about 100-200 microns, and a thickness between the back surface of the sensor structure and the front surface of the sensing electrode (i.e. the surface facing the direction in which the distance measurement is taken) of 50-150 microns, preferably 10 about 100 microns. The thin film structure, the small area, and the very small height (thickness) of the sensor makes it possible to apply the sensors in applications where there is very little room available (particularly where the available height is limited), and where close spacing between sensors or between sensors and other equipment is required.

[0067] The small size of the thin film sensor shown in Figs. 6 and 7 (and also shown in 15 other embodiments described below) provides many advantages. The thin film structure results in minimal height, and the width or area of the sensor may also be very small. This enables the sensor to be mounted in close proximity to the point at which the distance measurement is desired. When used with a lithography machine for measuring the distance between the projection lens and the target being exposed, the sensors can be mounted next 20 to the projection lens and on the same mounting structure so that both the sensors and the projection lens are fixed to the same reference point. This greatly reduces errors due to relative movement between the sensors and projection lens, eliminates the need for correction for sensor mounting variation, and reduces requirements for calibration. The small size of the sensor also reduces flatness requirements for the sensor itself.

25 [0068] Figs. 8 and 9 show additional embodiments of the thin film sensor with the insulating layer 34 formed only between the sensing electrode 31 and back electrode 35, so that the side guard electrode 32 and back electrode 35 can directly connect to each other.

[0069] Fig. 8A shows a top view and Fig. 8B shows a cross-sectional view of the sensor with square sensing electrode. In one embodiment the square sensor is designed with a 30 nominal sensor capacitance (capacitance Q between the sensing electrode and the target) of -12- 1 pF at a nominal distance of 100 microns between the sensor and target. The sensing electrode has a width of 3.5 mm (+/- 0.01 mm) with area of 12.25 mm2. The guard electrode has width 1.5 mm (+/- 0.01 mm), and the gap between the sensing and guard electrodes is 0.015 mm (+/- 0.001 mm). In another embodiment the sensor is designed with 5 a nominal sensor capacitance of 10 pF at a nominal distance of 100 microns between the sensor and target. The sensing electrode has a width of 11 mm (+/- 0.01 mm) with area of 121 mm2. The guard electrode width and gap are unchanged at 1.5 mm (+/- 0.01 mm) and 0.015 mm (+/- 0.001 mm) respectively.

[0070] Fig. 9A shows a top view and Fig. 9B shows a cross-sectional view of the sensor 10 with circular sensing electrode. In one embodiment the circular sensor is designed with a nominal sensor capacitance of 1 pF at a nominal distance of 100 microns between the sensor and target. The sensing electrode has a diameter of 4 mm (+/- 0.001 mm) with area of 12.25 mm2. The guard electrode has an inner diameter of 4.015 mm (+/- 0.001 mm) and outer diameter of 8 mm (+/- 0.001 mm). In another embodiment the sensor is designed with 15 a nominal sensor capacitance of 10 pF at a nominal distance of 100 microns between the sensor and target. The sensing electrode has a diameter of 6.2 mm (+/- 0.001 mm) with area of 121 mm2. The guard electrode has an inner diameter of 6.215 mm (+/- 0.001 mm) and outer diameter of 12.4 mm (+/- 0.001 mm).

[0071] The embodiments of Figs. 5-9 may be constructed to have a measurement range 20 (in the z axis perpendicular to the sensing electrode surface) of 80- 180 microns between the sensor and target. The dimensions of the sensors may be changed to accommodate a different measurement range, as will be appreciated by one of skill in the art.

[0072] The embodiments of Figs. 5-9 may be made of very thin layers of material, the sensing electrode 31 with thickness of 100 nm (+/- 10 nm), the side guard electrode 32 and 25 back electrode 35 of thickness 150 nm (+/- 10 nm), and the insulating layer 34 of thickness 50 nm (+/- 10 nm). The sensing electrode in these embodiments is square or circular, providing a large sensing area to maximize sensitivity of the sensor while minimizing the overall dimensions of the sensor. However, the sensor may deviate from these shapes, with a sensing electrode (and similarly guard electrodes) taking the form of a rectangle, oval or 30 other shape to maximize the sensing area.

-13-

[0073] The embodiments in Figs. 5-9 may be constructed with a conductive layer for the electrodes 31, 32 deposited onto the insulating layer 35 or affixed to the insulating layer with an adhesive or bonding layer. The gap 39 between the sensing and guard electrodes may be formed by forming a single conductive layer for both the sensing and guard 5 electrodes and removing material using a laser or etching to create the gap. A laser is preferred for making very small gap widths, and can be used to make a 25 micron wide gap with small deviation, while etching is generally less precise.

[0074] The sensors can be manufactured using various techniques, for example using lithographic techniques, MEMS (Micro Electro Mechanical Systems) technology, or 10 flexible printed circuit technology. Using flexible printed circuit technology, the insulating layer 34 may be provided as a flexible sheet or tape of suitable material, such as a Kapton polyimide film or similar flexible insulating film. The conductive electrodes 31, 32 and 35 may be formed of a thin layer of copper or other suitable conductive material, fixed to the insulating layer 34 using adhesive, formed as an adhesiveless laminate, e.g. those using a 15 direct metallization process, or printed onto the insulating layer using conductive inks or other suitable printing techniques. The protective insulating films 37 and 38 may be formed of the same types of materials as layer 34.

[0075] The flexible printed thin film sensor is easy to manufacture and can be made quickly resulting in a short lead time for manufacture. The sensor can be made with robust 20 connections from the sensor to the signal processing circuit. The small size provides more flexibility for placement at or very near the point where the distance is to be measured. The sensors can be glued in place as individual sensor elements to quickly and simply assemble the sensing system. The flatness and tilt of the individual sensors may be checked after they are glued in place and calibrated out in the measurement procedure. Where a flexible sheet 25 of material is used for the insulation layers, the entire sensor can be constructed to be flexible.

[0076] The gap width between the sensing and side guard electrodes in some of the above embodiments above do not satisfy the comparative design rules described above, e.g. the gap between the sensing and guard electrodes is more than one fifth of the distance between -14- the sensing electrode and the target. However, the advantages of the thin film structure will outweigh this for many applications of the sensors.

[0077] Figs. 10A to 10D show various configurations for a thin film capacitive sensor using different materials for the sensor substrate. These embodiments are suitable for 5 construction using lithographic techniques, which allow the manufacture of very precisely shaped electrodes with very small gap sizes. This enables a sensor to be constructed to satisfy the comparative design rules described above, and having very high resolution for measuring very small features and very small distances. Lithographic processes also enable the connecting lines and contact/bond pads to be made with very small track widths and 10 precise dimensions. Furthermore, lithographic processes are well known to those of skill in the art and once a process flow is developed can be readily applied in the manufacture of sensors having higher resolution. However, initial development of the process results in a longer manufacturing lead time, and will require short loop experiments to verify different process steps. Figs. 10A-10D only show the arrangement of layers, and do not show a side 15 guard electrode, but if included it would be formed on the same layer as the sensing electrode, and do not show the optional connection a side guard electrode and back guard electrode.

[0078] For applications where one or more sensors are mounted on a machine such as a lithography machine, the substrate in these embodiments may be common to more than one 20 sensor so that a set of sensors are constructed in one unit. An example of this type of arrangement if shown in Figs. 13A-13D and described below. The substrate may then be connected to a mounting plate or the substrate could be used as the mounting plate for mounting the sensors to the machine.

[0079] The embodiment in Fig. 10A has a silicon substrate 45 with insulating layers 47a, 25 47b formed on both sides. A sensing electrode 31 is formed on a surface of one of the insulating layers and a back guard electrode 35 is formed on a surface of the other insulating layer. This embodiment may require active biasing of the guard electrode to function effectively, which may require patterning of the plate on which the sensor is fixed. In addition, capacitive coupling between the projection lens and the sensors may be a 30 problem.

-15-

[0080] The embodiment in Fig. 10B has a silicon substrate 45 with multiple layers formed on one side, comprising a first insulating layer 47a, a sensing electrode 31 formed on the first insulating layer, a second insulating layer 47b, and a back guard electrode 35 formed on the second insulating layer. This embodiment avoids the need for patterning the 5 mounting plate to which the sensor is fixed, and also avoids capacitive coupling between the projection lens and the sensors. However, an additional insulating layer is required compared to the embodiment of Fig. 10A.

[0081] The silicon substrate is not a good insulator so that insulating layers are included in these embodiments. A further disadvantage of the silicon substrate is that parasitic 10 currents can be generated in the silicon due to impurities in the silicon, and these currents can disturb the capacitance measurement of the sensor.

[0082] The embodiment in Fig. 10C has a pyrex substrate 46 with a sensing electrode 31 is formed on one surface and a back electrode 35 formed on the other surface. This embodiment also requires patterning of the plate on which the sensor is fixed if active 15 biasing of the guard electrodes is implemented, although active guarding may be omitted with a decrease in sensitivity and an addition of a small amount of non-linearity in the sensor. An embodiment with this structure, with substrate thickness of 100 pm and gap between sensing electrode and side guard electrode of 16 pm, when energized with current of 50 pA at 150 kHz, may produce an effective output voltage of about 11.5V at a distance 20 of 0.8 pm between sensor and target, and an effective output voltage of about 13.5V at a distance of 1.8 pm between sensor and target.

[0083] The embodiment in Fig. 10D has a pyrex substrate 46 with multiple layers formed on one side, comprising a guard electrode 49, an insulating layer 47 formed on the guard electrode, and a sensing electrode 48 is formed on the insulating layer. Patterning of the 25 plate on which the sensor is fixed is not required for this arrangement, and capacitive coupling between the projection lens and the sensors is reduced due to the 100 pm Pyrex layer. Pyrex is a good insulator and the insulating layers used with the silicon embodiments can be omitted for the embodiments using a pyrex substrate.

[0084] Making the electrical connections between the sensor electrodes (sensing, side 30 guard and back guard electrodes) and the signal processing system requires making a robust -16- low impedance connection to the small sensor elements. The connection should be able to withstand the mechanical stresses expected, while avoiding the introduction of additional parasitic capacitances in the sensor arrangement. For sensor applications with lithography machines, the connections should also avoid use of materials that would give off 5 contaminants into a vacuum environment.

[0085] Fig. 11 shows a sensor having contact pads 50a, 50b formed at the ends of connecting lines 41 and 42, for making external connections from the sensor to signal processing circuits. Figs. 12A and 12B show cross sectional views of the structure of contact pads for making electrical contact to the sensor electrodes. These are particularly 10 suited for the embodiments using substrates of silicon, pyrex and similar materials. These embodiments provide a contact pad on the back side of a substrate for electrically connecting to sensor electrodes on a front side of the substrate. Fig. 12A shows an embodiment with a via hole through a substrate 55. A conductive contact pad 50 is formed on a back side of the silicon substrate and a conductive connection 51 is formed through the 15 via hole to connect with a conductive layer 52 on the front side of the substrate. Fig. 12B shows an embodiment with an electrical connection made over the edge of the substrate 55. A conductive contact pad 50 is formed on a back side of the substrate and a conductive connection 51 is formed at the edge of the substrate to connect with conductive layer 52 on the front side of the substrate.

20 [0086] For embodiments using a silicon or other non-dielectric substrate, an insulating layer 53 separates the conductive layer 52 from the substrate, and a small insulating layer 54 separates the contact pad 50 from the substrate. The via hole is also coated with an insulating layer in Fig. 12A, and the edge of the substrate under the conductive connection 51 is coated with an insulating layer in Fig. 12B. The additional insulating layer required for 25 the contact pad gives rise to an additional small parasitic capacitance. For embodiments using a dielectric substrate such as pyrex, the additional insulating layers are optional and additional parasitic capacitances are reduced.

[0087] Figs. 13A to 13D show an embodiment of a sensing system with multiple sensors constructed on a single substrate 102 surrounding a projection lens 104 of a lithography 30 machine. Fig. 13A shows a front side of the substrate, i.e. the side facing downwards and -17- towards the wafer to be exposed. Eight sensors (comprising four sensor pairs) are formed on the substrate spaced in pairs around the projection lens. In this embodiment, a conductive sensing electrode 31 is formed on the front side of the substrate for each sensor. A side guard electrode 32 surrounds each sensing electrode with a small gap formed 5 between them. Connecting lines 105 make electrical connections between each sensing and guard electrode and the edge of the substrate. In this embodiment the substrate is made from a dielectric material such as pyrex or kapton, and no additional insulating layer is used between the electrodes and the substrate. A thin protective insulating layer may also be formed over the sensor electrodes on the front side of the substrate.

10 [0088] Fig. 13B shows the back side of the substrate, i.e. the side facing upwards away from the wafer to be exposed. A conductive back guard electrode 35 is formed on the back side of the substrate for each sensor. For each sensor, the back electrode is aligned with the sensing and side guard electrodes on the front side of the substrate. In this embodiment with circular electrodes, the centers of all of the electrodes are aligned for each sensor. The back 15 electrode 35 has a larger diameter than the sensing electrode 31, and may be equal to or greater than the diameter of the side guard electrode 32 on the front side. Connecting lines 106 make electrical connections between the back guard electrodes and the edge of the substrate.

[0089] The connecting lines 105 on the front side and 106 on the back side of the 20 substrate may be arranged to form contact pads 50a and 50b on the back side of the substrate at the edge, e.g. using the construction shown in Figs. 11,12A or 12B, where contact pads 50a are electrically connected to the sensing electrodes 31 and the contact pads 50b are connected to the side guard electrodes 32 and back guard electrodes 35. In this embodiment, the contact pad areas alternate, with each contact pad 50a from a sensing 25 electrode having a contact pad 50b from a corresponding side guard and back guard electrode on either side. An additional contact pad 50c is also formed at the edge of the substrate to connect to a shield electrode, which may be connected to a shield for the cable connecting the sensors to the measurement system. The contact pad areas together form contact pads 50, arranged in separate areas corresponding to the sensor pair arrangement 30 on the substrate.

-18-

[0090] Fig. 13C shows the back side of the substrate with an insulating layer 110 formed over the substrate, leaving a gap around the edge of the substrate so that the contact pads 50 are exposed for making connections. Fig. 13D shows the substrate 102 mounted onto a spacer/mounting plate 112. The mounting plate 112 may be conductive and can function as 5 a shield electrode, and may be grounded, or alternatively a conductive shield plate functioning as a shield electrode may be provided as a separate component. The contact pad 50c functions as a connecting area for making electrical connections to the shield electrode, e.g. to connect to the sensor shield. The insulating layer 110 electrically separates the guard electrodes from the mounting plate/shield electrode. In this embodiment the mounting plate 10 has cut-outs around its edge to leave the contact pads 50 exposed for making electrical connections.

[0091] In one embodiment, the arrangement in Figs. 13A-13D may comprise a pyrex substrate of 50 mm diameter with the square hole, e.g. 19 x 19 mm or 26 x 26 mm, to accommodate a projection lens. The sensing electrodes have a diameter 3.8 mm and a gap 15 of 16 pm between the sensing and guard electrodes, the guard electrodes have a width of 1 mm, and the back electrodes a diameter of 6 mm. The connecting lines 105 have a width of 0.05 mm and a separation of 16 pm, and the connecting lines 106 have a width of 0.5 mm and a separation of 0.5 mm, and the contact pads may be 0.5 mm wide and 1.4 mm long, the pads separated from each other by a 0.5 mm gap. The sensor may be powered with a 20 current of 10 pA at 200 kHz.

[0092] The sensors in the embodiment shown in Figs. 12A-12D, or in any of the other sensor arrangement described herein, may be arranged in differential pairs where each sensor in a pair is driven by a voltage or current which is out-of-phase from the other sensor of the pair. For example, a first sensor of a pair may be driven by a current 180 degrees out- 25 of-phase from the other sensor of the pair. To reduce coupling between pairs of sensors so that multiple differential pairs of sensors may be used together, each sensor pair may be driven by a voltage or current which is phase offset from the adjacent sensor pair. For example, adjacent pairs of sensors may be driven by a current 90 degrees out-of-phase from each other. For example, the pair of sensors at the top of Fig. 13A may be energized by 30 currents at phase 0 and 180 degrees, while the pair of sensors on the right side and the pair -19- on the left side are each energized at phase 90 and 270 degrees, and the pair of sensors at the bottom are energized at phase 0 and 180 degrees. In this way, a phase division technique is used with orthogonal biasing of adjacent sensor pairs, to separate the pairs and reduce interference between them. Other techniques, such as frequency division or time 5 division may alternatively or additionally be used to reduce interference between sensor pairs.

[0093] An electrical connection from the sensor probes to a signal processing system is needed to transfer the electrical signals off the sensor and transmit them for converting the raw sensor signals into a usable format. Fig. 14 shows a side view of a sensor arrangement 10 with sensors 30 mounted on a front side of a substrate 102 around a lithography machine projection lens 104. Contact pads 50 are formed on the back side of the substrate, and connection wires 60 in the form of metal contact springs make electrical contact with the pads to connect to a signal processing system.

[0094] Figs. 15A and 15B show an alternative connection arrangement using a flexible 15 printed circuit connecting member 110, comprising a flexible membrane 111 on which conductive tracks 112-114 are printed or affixed. A protective insulating layer may be formed over the conductive tracks. One end of the flex print connector 110 is bonded to contact pads 50 or connecting areas of the sensor electrodes so that the conductive tracks make an electrical connection to the sensor electrodes. In the embodiment shown, 20 conductive track 113 connects to a contact pad for the sensing electrode, and conductive tracks 112 and 114 connect to contact pads for the guard electrode and/or back electrode.

A connector plug or socket 116 is affixed to the other end of the flex print connector 110 with contact terminals 117 for making electrical contact with wires or connecting pins 120 for transferring the sensor signals to the signal processing system. Fig. 15A shows the 25 underside of an embodiment of the flex print connector 110 showing the conductive tracks 112-114, alongside a topside view of a set of contact pads 50 to which the conductive tracks would connect. Fig. 15B shows a side view of the flex print connector 110 when connected to the contact pads. The flex print connector is flexible and may be used with any of the sensor embodiments described herein. The maximum bending radius of the flex print 30 connector should be considered, particularly for very small conductive track widths, and -20- alignment between the flex print connector and sensor contact pads during assembly is important.

[0095] Fig. 16A shows a cross section through a projection lens and deflector stack 132 of a charged particle lithography machine. The stack 132 typically comprises vertically 5 stacked projection lens elements and beamlet deflector elements for focusing charged particle beamlets generated by the lithography machine onto the surface of the wafer and deflecting them across the surface of the wafer in a scanning pattern. Each vertically stacked projection lens element may actually comprise an array of projection lenses for simultaneously focusing a large number of beamlets onto the wafer surface, each beamlet 10 for exposing a different portion of the wafer, and each deflector element may similarly comprise an array of deflectors.

[0096] The lens stack 132 is mounted in a housing frame member 130. A mounting plate 112, which may also function as a spacer between two electrostatic lens elements of the lens stack, is positioned below and affixed to the frame member 130, with a centrally located 15 hole through which the charged particle beam is projected. The plate/spacer 112 may be made from glass or other suitable insulating material to provide an insulating layer between the high voltages present in the lens stack 132 and the bottom lens 104, wafer, sensors, and other components nearby. The plate 112, together with generally cylindrically shaped frame member 130 and upper mounting plate 133, forms a housing structure for the projection 20 lens and deflector stack 132.

[0097] The plate 112 may alternatively be conducting or include a conducting layer which functions as a shield for the sensors. In the embodiment shown, the projection lens stack 132 comprises a series of projection lens elements arranged in a vertical stack, located mostly above the plate 112 but with the final lens element 104 of the stack located below 25 the plate on the bottom surface of the plate 112.

[0098] Fig. 16B shows an arrangement of sensors for measuring a distance related to the distance between the bottom projection lens 104 of the projection lens stack and the wafer 9 resting on moveable wafer table 134. Note that the wafer and table are shown schematically for convenience, the width thereof actually being much larger than that of the 30 lens stack housing. Typically the wafer is 200 or 300 mm in diameter versus 50-70 mm for -21- the lens stack housing. The sensors 30 are mounted on the same plate 112 as the bottom projection lens 104, and in close proximity to the lens 104. The sensors are preferably smaller than the size of the exposure field of the lithography machine, and some or all of the sensors may be located closer to the edge of the projection lens than a distance equal to the 5 width or length of the exposure field size.

[0099] In this arrangement the sensors are mounted in a fixed relationship to the projection lens, so that the distance between the bottom projection lens 104 and the wafer 9 can be determined from the measured distance between the sensors and the wafer. The very small size of the sensors described herein makes it possible to mount the sensors in close 10 proximity to the projection lens and permits them to be mounted on the same support element so they are both fixed to the same reference point. Because the sensors are integrated with the bottom projection lens on a single structure, this greatly reduces errors due to changes in the relative positions of the sensors and the projection lens, caused for example by thermal expansion and contraction and movement between the support element 15 on which the bottom projection lens is mounted and the support element on which the sensors are mounted, and due to mounting imprecision of different base structures for the sensors and bottom projection lens respectively. This results in eliminating the need for calibration of the sensor system for variation in the x and y axis (i.e. parallel to the surface of the wafer) and in the z axis (i.e. perpendicular to the surface of the wafer), or at least 20 reducing the need for such calibration. Conventional capacitive sensors are too tall and wide to mount on the plate 112, and would have to be mounted further away from the projection lens 104, for example on the frame member 131.

[00100] The sensors are disposed on plate 112 adjacent to the bottom of the projection lens 104. For sensors constructed separately, e.g. according to a thin film construction of 25 the type shown in Figs. 5-9, the individual sensors may be fixed directly to the mounting plate 112, e.g. with adhesive. For sensors formed on a common substrate, e.g. of the type shown in Figs. 13A-13D, the sensor substrate may be fixed to the mounting plate 112, also using adhesive or other attachment means. It is also possible for the common sensor substrate to also function as the mounting plate 112.

-22-

[00101] The bottom of the projection lens 104 may be substantially at the same height as the bottom surface of the sensing electrodes of the sensors 30, or may be slightly lower. By designing the system so that the distance that is desired to be measured is nearly equal to the distance actually measured (i.e. the distance between the bottom surface of the sensing 5 electrodes and the target being measured) the sensitivity of the system is increased. In one embodiment, when used for a lithography machine, the bottom of the projection lens 104 extends 50 pm below the bottom surface of the sensing electrodes of the sensors 30. The focal plane of the projection is 50 pm below the bottom of the projection lens and 100 pm below the sensing electrodes. The wafer table has a z-axis (vertical) movement range of 80-10 180 pm below the bottom surface of the sensing electrodes, with a positioning accuracy of 100 nm, the top of this range bringing the wafer within 80 pm of the sensing electrodes and the bottom of this range moving the wafer to 180 pm below the sensing electrodes.

[00102] Figs. 17A to 17D show a flexible printed circuit structure 120 with multiple sensors 30 and integrated flex print connector 110. The structure 120 comprises a flexible 15 base of insulating material such as a Kapton polyimide film or similar flexible insulating film.

The conductive electrodes for the sensors and conductive tracks for making the connecting lines are formed of a thin layer of copper or other suitable conductive material, fixed to the insulating base layer using adhesive, formed as an adhesiveless laminate, e.g. using a direct metallization process, or printed onto the insulating layer using conductive inks or other 20 suitable printing techniques. A protective insulating film may then be formed over the conductive layers.

[00103] In the embodiment shown, eight pairs of sensors are arranged in a square array around a square cutout 121 for accommodating the bottom projection lens 104. Single sensors may also be used in place of the sensor pairs, and different spatial arrangements of 25 the sensors or sensor pairs may also be used. The flexible base layer includes extended portions which function as flex print connectors 110, constructed as described above, for making electrical connections between the sensors and a signal processing system. The flex print connectors 110 connect to connectors 116, which provide stress relief and interface to triaxial cables 210 for connecting to a remote signal processing system.

-23-

[00104] The integration of a sensor and the connections and wiring necessary to transfer signals from the sensor to a location away from the sensor where a larger and more robust connection can be made solves several problems. The extremely small size of the sensor makes it difficult to make electrical connections due to size limitations of the wiring and 5 connector hardware. The capacitances introduced by wiring must be controlled so that they do not dominate the sensor system. Any small movement or vibration of wiring may result in damage or a need to recalibrate the sensor. The integration of both sensor and sensor wiring onto a single flex print flexible base layer enables a connection at the sensor with very small dimensions, the capacitances introduced by the wiring can be controlled during 10 the design of the system, and the integration onto a single base layer produces a mechanically robust design greatly reducing problems with movement of wiring.

[00105] The integration of multiple sensors with their associated wiring onto a single base layer provides additional advantages. By forming the array of sensors on a single base, the spatial arrangement of the sensors is fixed when manufactured, and a larger integrated 15 structure makes for easier handling and affixing to the equipment, e.g. a lithography machine.

[00106] Signal preprocessing circuits 200 may be integrated on the flex print connector by printing or otherwise forming the circuits onto the flexible base layer. The signal preprocessing circuits 200 may include a buffer/amplifier used for active biasing of the 20 sensor guard electrodes (described below), may include additional circuitry, or may be omitted so that only connection hardware local to the projection lens and all active components are located remotely. As the lithography machine operates in a vacuum environment, placing active components close to the sensors and in the vacuum may result in problems dissipating heat from the active components due to lack of heat transfer in a 25 vacuum. However, locating the components needed for active guarding close to the guard electrodes increases performance of the system. In the embodiment shown, the signal preprocessing circuits 200 are located next to the connectors 116 so that heat generated by the circuits may be more effectively conducted through the connectors 116 to the cables 210 and away from the sensor arrangement.

-24-

[00107] Fig. 18 shows an alternative connection arrangement. A flex print connector 110 connects at one end to a sensor 30 on mounting plate 112 and at the other end to a signal preprocessing circuit 200 via connecting wires or pins 201. The signal preprocessing unit 200 may be mounted on a frame member 131, preferably in a recess or compartment. The 5 output of the signal preprocessing circuits 200 is transmitted to a control system via wires or pins 202, connector 204, and triaxial cable 210.

[00108] Figs. 19A and 19B show an arrangement for mounting the integrated flexible printed circuit structure 120 (shown in Figs. 17A to 17D) in a lithography machine. Fig.

19A shows the arrangement from above, with the projection lens stack removed to improve 10 visibility, and the flexible printed circuit structure 120 located in a well in frame member 136. The connectors 116 connect via triaxial cables 210 to cabling bundle 212 which connects in turn to cables 214 for connecting to a remote signal processing system. Fig.

19B shows a bottom view showing the flexible printed circuit structure 120 and sensors 30 facing towards the wafer.

15 [00109] The flexible printed circuit structure 120 may be fixed to the bottom surface of a mounting plate, e.g. the mounting plate 112 shown in Figs. 16A and 16B, using an adhesive or other suitable attachment method. This results in the integration of the sensor array with associated wiring with the mounting plate 112 and bottom projection lens 104, all in a single structure. The sensors are thereby mounted in close proximity to the projection lens 20 and in a fixed relationship to the projection lens, resulting in the benefits as described above for the embodiment in Figs. 16A and 16B.

[00110] Figs. 20A and 20B show various configurations of capacitive sensors on a mounting plate 112 surrounding a lithography machine projection lens 104. In Fig. 20A, four sensors are distributed in the four quadrants of the mounting plate 112, the sensors 25 arranges in pairs for differential sensing. On the bottom side of the mounting plate, each sensor comprises a sensing electrode 31 and side guard electrode 32. This arrangement is particularly suited to measurement of height and tilt of a wafer. Fig. 20B shows an arrangement of sixteen sensors arranged in pairs in a square matrix, the middle of the square having no sensors where the projection lens 104 is located. For all of the configurations -25- described above, back guard electrodes may be included on the back side of a sensor substrate.

Electronic circuits

[00111] Fig. 21 shows a sensor system 300 comprising one or more sensor probes 30, a 5 signal processing system 301, and a connection system 302 for conveying signals from the sensor probes to the signal processing system. The signal processing system 301 may include current or voltage source circuitry for driving the sensor probes, amplifier/buffer circuitry for amplifying the raw sensor signal, circuitry for biasing the sensor guard electrodes and connecting cable conductors, signal processing circuitry for processing the 10 signals received from the probes and for output of the processed signals as measurement data, and circuitry for calibrating the system. The connection system 302 my include cables to connect the sensors to the signal processing system.

[00112] Each part of the system may be a source for various types of measurement errors and factors reducing sensitivity. Errors are introduced by the sensor probes due to the finite 15 geometry of the probes and the limitations of the manufacturing process resulting in irregularities and imprecision in the geometry of the sensor electrodes and other components. Intrinsic/parasitic capacitances, due to the structure of the sensor probe, and interaction with other components near to the probes, may reduce sensitivity of the sensors.

[00113] Errors may be introduced by the mounting of the probes, as a result of tilt or non-20 flatness of the mounting surface or probes, and tolerances in the position and other factors relating to the mounting. Errors may be introduced by the signal processing system, due to signal processing errors, component tolerances, external or internal interference, and other factors. Errors may also be introduced by the connection system, such as additional capacitances introduced by the connecting components such as connecting lines, contact 25 pads, connection wires and cabling.

[00114] To detect a change in the capacitance of the sensors, which represents distance between the sensor and the target, various amplifier configurations can be used. Fig. 22A is a simplified circuit diagram of a basic high-impedance amplifier circuit. The high input impedance amplifier uses a unity gain non-inverting configuration of an amplifier 305. An 30 AC current source 306 is connected as an input to the amplifier, in parallel with the sensor -26- probe 30. The circuit produces a linear output 309 proportional to the change in capacitance, which varies with the distance between the sensor probe 30 and target 9.

[00115] The sensor 30 is connected between the input of the amplifier and ground or virtual ground, i.e. one electrode of the capacitance being measured is connected to ground.

5 However, for distance measurements to a wafer, the sensing electrode of the sensor forms one electrode of the measured capacitance while a conducting layer in the wafer forms the other electrode. The conducting layer of wafer 9 is typically capacitively coupled to ground via the wafer table 134 and other lithography machine components. The capacitance between the wafer and ground typically varies between 6 pF to 70 nF, and the nominal 10 value of the sensor capacitance is typically about 0.1 pF to 1 pF. To measure the small changes in distance between the sensor 30 and wafer 9 accurately it is necessary to have the wafer to ground capacitance at least 1000 times larger than the nominal capacitance of the sensor. Since the range of variation of the wafer to ground capacitance is quite large, small changes in this capacitance can affect the distance measurement. If the wafer to ground 15 capacitance is not at least 1000 times larger than the nominal capacitance of the sensor, then small changes in wafer to ground capacitance will cause changes in the measured capacitance and unwanted changes in the distance measurement.

[00116] In the differential measurement principle the two sensors 30a, 30b of a differential pair are driven by current sources 306a, 306b which are 180 degrees out of phase, as shown 20 in Fig. 22B. The current through one probe finds a path through the conductive layer in the target. A virtual ground, i.e. the lowest potential or a constant potential in the current path is created at the centre of the current path. High impedance amplifiers 305a, 305b are used to measure the voltage signals corresponding to the change in the distance between the sensors and a conductive layer in the target. The differential measurement principle makes 25 the distance measurement independent of the variation in the wafer to ground impedance.

[00117] Fig. 23 shows an alternative circuit for biasing the sensor 30 with a voltage source. Two fixed impedances 71 and 72, a variable capacitance 73, and the sensor 30 (shown as a variable capacitance) are arranged in a whetstone bridge arrangement energized by voltage source 306. The bridge has two legs in arranged in a parallel circuit, the 30 impedance 71 and variable capacitance 73 connected together at a first node and forming -27- one leg, and the impedance 72 and the sensor 30 connected at a second node and forming the other leg. The fixed impedances 71 and 72 have identical impedance values, and the variable capacitance 73 is adjusted to match the nominal capacitance of the sensor 30. The two nodes at the mid point of each leg of the bridge are connected to the two inputs of a 5 differential amplifier 75, which measures the difference in the voltage across nodes, i.e.

across the variable capacitance 73 and sensor 30. The variable capacitance is tuned to adjust the null value of the differential amplifier, and may be adjusted by an automated calibration algorithm.

[00118] The bridge components and differential amplifier may be located at the sensor.

10 When used in conjunction a flex print construction as shown in Figs. 17A-17D, the fixed impedances 71, 72 may be formed on the same flexible base layer as the sensor. The variable capacitance 73 may be formed using a variable capacitance diode (varicap) or other suitable component. The variable capacitance 73 may also be integrated on the flexible base layer, as a component mounted or formed on the base layer, or integrated into the flexible 15 structure itself using copper and insulating layers. The differential amplifier may be formed on the base layer, but the considerations regarding active components in a vacuum environment discussed above also apply. When the differential amplifier is remotely located and the same cable length is used for connection to the sensor and variable capacitance, the effect cable capacitance can be removed and common mode disturbances may be cancelled 20 out.

[00119] Fig. 24 shows the circuit of Fig. 23 implemented for a differential sensor pair.

Each sensor 30a, 30b of the pair is connected to a fixed impedance 71, 72 and biased by a voltage source 306a, 306b. The whetstone bridge arrangement is now formed by the fixed capacitances 71 and 72 and the sensor pair 30a and 30b, which are connected via 25 conductive wafer resist on the target.

[00120] Fig. 25 shows an embodiment using differential measurement with high impedance amplifier circuits. Two sensors 30a, 30b are arranged in a differential pair. The sensing electrode 31a of sensor 30a is driven by AC current source 306a, and sensing electrode 31b of sensor 30b is driven by AC current source 306b. The two current sources 306a, 306b are 30 180 degrees out of phase to each other. During one half cycle, current 307 flows in one -28- direction through sensor 30a and sensor-to-target capacitance 16a, through a conductive layer of the target 9, and through sensor-to-target capacitance 16b and through sensor 30b. During the next half cycle the current flows in the reverse direction.

[00121] Amplifier/buffer 305a amplifies the raw output voltage of sensor 30a to generate 5 output signal 309a for further processing. The output 309a may also be fed back to the side guard electrode 32a and/or back guard electrode 35a of the sensor 30a. This implements active guarding by energizing the guard electrodes with the same voltage that is present in the sensing electrode, so that there is no electric field formed between the sensing electrode and the guard electrode, so that the electric field between the sensing electrode and the 10 target is as uniform as possible. Amplifier/buffer 305b similarly amplifies the raw output voltage of sensor 30b to generate output signal 309b, and provide an active biasing signal for the guard electrodes of sensor 30b. The output signals 309a, 309b may be input to synchronous detector circuits 330a, 330b respectively.

[00122] The amplifiers 305a, 305b are preferably located close to the sensors 30a, 30b, 15 particularly when active biasing of the guard electrodes is implemented, to prevent the introduction of errors caused by additional capacitances introduced by cabling from the sensors to a remote location where the signal processing takes place. For capacitive sensor applications for a lithography machine operating in a vacuum, placing active components close to the sensors usually requires putting these components in the vacuum chamber, 20 which may result in heat dissipation problems due to lack of heat transfer by radiation in a vacuum (although heat transfer by conduction still occurs in a vacuum). For this reason, the current sources 306a, 306b and further signal processing circuits such as the synchronous detectors 330a, 330b may be located remotely from the sensors outside the vacuum chamber. However, the amplifiers 305a, 305b are preferably located in the vacuum chamber 25 close to the sensors to achieve lower measurement error, in a configuration that permits conduction of heat away from the active components.

[00123] Fig. 26 shows one embodiment of a synchronous detector circuit 330. A reference oscillator 331 generates a reference frequency fl which is used by current source 306 to generate alternating current 332 for driving the sensor, and is used by phase shifter 333 to 30 generate reference signal 334 (also at frequency fl) which has a phase shift in relation to the -29- reference frequency. The phase shift of the reference signal 334 is tuned to be equal to the phase shift between the reference frequency and signal 309 from the sensor, to account for phase shift occurring in the cabling between the synchronous detector circuit 330 and sensor and within the sensing arrangement.

5 [00124] The output 309 from the sensor at frequency f2 is the input to input buffer 335.

Multiplier 337 receives the buffered or amplified input signal 336 at frequency J2 and reference signal 334 at frequency//. The output from multiplier 337 will include components of the sum of the two input frequencies (fl + J2) and the difference between the two input frequencies (fl -f2). The output from multiplier 337 is passed through low 10 pass filter 338 to filter out the higher frequencies to leave the low frequency component representing the difference between the two input frequencies (fl -f2). This signal is amplified by amplifier 339 to generate measurement signal 340. This measurement signal 340 represents the change in impedance measured by the sensor, which depends on the change in distance between the sensor and the target.

15 [00125] As noted above, the current source circuits for driving the sensors and signal processing circuits may be located remotely from the sensors. However, a cabling connection used for connecting the sensors to the remote circuits will introduce additional undesirable capacitances in the system. Fig. 27 is a diagram showing capacitances in the sensor system. Capacitance Cin represents the cable capacitance plus the output capacitance 20 of the current source 306, capacitance Cs represents the sensor-to-target capacitance being measured plus the intrinsic (parasitic) capacitance of the sensor and sensor connecting lines, and capacitance Cout represents the cable capacitance plus the input capacitance of the buffer 305. The stray capacitances Cin and Cout should be small in comparison to Cs and the sensor-to-target capacitance being measured, because large stray capacitances will 25 reduce the sensitivity of the sensor.

[00126] The capacitance of the cable connecting the sensors to the signal processing circuits may be large and have an adverse effect on the sensitivity of the sensor. Active guarding may be used for the cable conductors to reduce or eliminate this problem. Fig. 28 shows an arrangement with a cable 350 connecting sensor 30 to signal processing circuit 30 330. In the arrangement shown, amplifier/buffer 305 is located locally to the sensor 30 (to -30- the left of the vertical dashed line) and the current source 306 is located remotely from the sensor (to the right of the vertical dashed line). The current source 306 transmits current to the sensor 30 via the center conductor 351 of the cable. The buffer 305 energizes the coaxial conductor 352 with the same voltage as present on the sensor’s sensing electrodes.

5 Since the center conductor 351 and coaxial conductor 352 have the same voltage on them, the effect of any stray capacitance between the conductors is effectively nullified.

[00127] The cable in this embodiment is a triaxial cable with a third coaxial conductor 353 which functions as a shield and is grounded. The shield is preferably connected to a separate ground at the remote end of the cable, e.g. at the circuit 330. This ground is just a shielding 10 ground and is preferably not connected to any ground at the sensor.

[00128] Many alternatives to this arrangement are possible. For example, a non-coaxial cable can also be used, and the buffer 305 may be located remotely from the sensor to energize the conductor 352 from the far end of the cable 350 locally to the signal processing circuits 330.

15 [00129] The physical separation of sensor and signal processing circuitry has been avoided in conventional designs, in which the sensors and circuitry were supplied together by one manufacturer and calibrated together in the factory as a set to match the electronic circuits to the sensor and avoid non-linearities. In the system described herein, an automated script may be used in conjunction with the wafer table control system to calibrate the sensing 20 system quickly, e.g. by moving the wafer table to known positions and taking measurements from the sensors. This removes the necessity of treating the sensor and signal processing circuitry as a matched pair and allows replacement of a sensor without also replacing the signal processing circuitry, greatly simplifying maintenance and reducing non-productive time for the lithography machine and thus increasing throughput.

25 [00130] Fig. 29 shows another embodiment of a synchronous circuit 360. A digital reference oscillator 376 generates a reference frequency fl, which forms a reference signal input to mixer 372, and is also fed to digital phase shifter 377 to introduce a phase delay. The phase delayed reference frequency is converted to an analog signal by digital-to-analog converter 378, and the analog phase delayed signal drives the current source 306 which 30 energizes the sensor 30. The phase delay is tuned to equal the phase shift occurring in the -31- sensing system and cable 350, equal to the phase difference between the reference frequency fl and the sensor system output signal 361 at the input to the synchronous detector circuit. The phase delay may be set by adjusting the phase shifter 377 until a maximum output is obtained from the mixer 372. Note that the phase shift in the sensing 5 system and cable should remain constant even when the capacitance of the sensor changes.

[00131] Input buffer 362 receives the sensor signal 361, and the buffered signal is input to band pass filter 363 to filter out noise and interference from the signal. The filtered signal is converted to a digital signal by analog-to-digital converter 364. Digital processing is then used, the circuit 360 thus combining analog and digital processing of the signals to use the 10 best features of both. The digital filtered sensor signal is then input to a single-input differential-output, differential amplifier (or phase splitter) 371 with two differential outputs, one in phase with the input signal and the other output 180 degrees out of phase with the input signal. Mixer 372 receives the differential outputs (at frequency f2) and the reference signal (at frequency fl). The input frequencies fl (the reference frequency) and f2 15 (the frequency of the sensor signal 361) are eliminated in mixer 372 and the mixer output includes components with frequencies of the sum and difference of the input frequencies (i.e. fl + f2 and fl - fl). The low pass filter 373 filters out the sum of the frequencies leaving the low frequency component representing the difference between the reference frequency and the sensor signal frequency (fl - fl). When the frequencies fl and fl are 20 equal and in phase, the mixer outputs a DC value proportional to the amplitude of the sensor signal 361, which is proportional to the sensor capacitance and proportional to the distance between the sensor and the target. Amplifier 374 amplifies the output from the low pass filter and it may then be input to a control system.

[00132] A calibration algorithm may be used to automatically calibrate the synchronous 25 detector circuit to the sensor system. The algorithm adjusts the digital phase shifter 377 in steps to increment or decrement the phase delay, and monitors the output of the mixer 372. The phase-shifter is adjusted until a maximum value is achieved at the output of mixer 372, indicating that the reference signal and sensor signal are in phase.

-32-

[00133] The digital processing of the synchronous detection circuit may be implemented using a field programmable gate array (FPGA), and the calibration algorithm can also be implemented in the FPGA and/or in software working in conjunction with the FPGA.

[00134] This design with automated calibration allows for replacing a sensor or array of 5 sensors without the need for also replacing or recalibrating the signal processing circuitry which processes the measurement signals from the sensor or sensors. Conventional sensor system designs involved sensors and signal processing circuitry calibrated together in the factory as a matched set, so that replacing a sensor required also replacing the signal processing circuitry. The automated calibration algorithm enables the sensors to be replaced 10 independently of the signal processing circuits and swift calibration of the new sensors with the existing signal processing circuits, greatly simplifying maintenance and reducing nonproductive time for the lithography machine. This enables less downtime and consequently higher throughput of the lithography machine.

[00135] Fig. 30 is a simplified diagram of a processing circuit for a differential pair of 15 sensors 30a, 30b. The sensors are biased by current (or voltage) sources 306a, 306b operating 180 degrees out of phase. The output of each sensor 30a, 30b is fed to synchronous detector circuits 360a, 360b respectively, and the outputs of the synchronous detector circuits are input to circuit 380 implementing a subtraction function. The outputs of the synchronous detector circuit outputs are 180 degrees out-of-phase, and so the 20 subtraction operates to remove common mode interference from the two synchronous detector circuit outputs while retaining the sensor signals.

[00136] The capacitive sensors may be used for control of the positioning of a wafer for a lithography machine. Fig. 31 is a schematic diagram of such a control system. The sensors 30 measure a distance related to the distance between the projection lens 104 of the 25 lithography machine and wafer 9 resting on moveable wafer table 134. In this arrangement the sensors are mounted in a fixed relationship to the projection lens, so that the distance between the projection lens and the wafer can be determined from the measured distance between the sensors and the wafer. The sensor signals are transmitted to signal processing unit 301, and the measured data output by the signal processing unit is transmitted to 30 control unit 400. The control unit 400 may be used to control the z-axis (vertical) -33- movement of the wafer table to maintain the wafer in the focal plane of the projection lens of the lithography machine, so that the charged particle beamlets generated by the machine remain focused on the surface of the wafer as the wafer moves in the x and y axis.

[00137] The sensor system in conjunction with the control system may be used to 5 accurately determine the distance between the projection lens and the wafer at various points as the wafer is moved. This enables the system to determine the topology of the surface of the wafer, detecting the presence of a tilt or bow in the wafer and other irregularities in the surface of the wafer. The wafer surface is almost never completely flat.

A typical wafer may have a bow in it of up to 50 pm without clamping. Apart from the 10 wafer bow the wafer surface may have other non-uniformities over its surface. The wafer bow and other non-uniformities result in height variations in the wafer surface, which result in the wafer surface not being in the focal plane of the projection lens. The control system may be used to correct for this height variation to maintain the wafer surface in the focal plane of the projection lens used to focus the optical or particle exposure beams onto the 15 wafer. The vertical position of the wafer table may be adjusted to compensate for these variations in height of the wafer surface using signals transmitted from the sensors.

[00138] The system is designed to measure the topology of the wafer surface while processing (e.g. exposing) the wafer, rather than doing the measurements in advance. This reduces the overall wafer processing time and increases throughput.

20 [00139] In one embodiment, an arrangement is used of eight sensors in a square matrix around the projection lens, as shown in Fig. 32. This arrangement permits a measurement of wafer tilt, wafer bow and other irregularities and the exposure of the wafer to be done in a single scan of the wafer. In a typical arrangement, the wafer is moved in a mechanical scan direction while the optical or charged particle beams of the lithography machine are scanned 25 over the surface of the wafer to expose the wafer according to an exposure pattern.

[00140] As the wafer is moved in a mechanical scan direction 405, the first row of sensors A-C measure distance to the wafer surface at three points in a line on the surface of the wafer, corresponding to the sensors A, B and C. The presence and magnitude of a tilt in the wafer position, bow in the wafer, or other surface irregularities can be calculated by 30 comparing the measurement from sensor B with measurements from sensors A and C. The -34- calculated value of the wafer tilt, bow or irregularity along the line is stored in memory in the control system. As the wafer advances in direction 405, the line previously measured by sensors A-C falls under the projection lens 104 and is ready to be exposed. The sensors D and E on either side of the projection lens measure distance to the points on the line 5 previously measured by sensors A and C. The height of the wafer at the point under the projection lens can now be calculated on the basis of the stored value and the current measurements from sensors D and E. The control system can use this calculation of the height of the wafer at the point under the projection lens to adjust the height of the wafer table to ensure the surface of the wafer remains in the focal plane of the projection lens.

10 This enables the compensation for a tilt, bow or other irregularity in the surface of the wafer to be measured and corrected for in a single scan of the wafer.

[00141] The third row of sensors F-H are provided to enable the same process described above to be used in both mechanical scan directions, i.e. when the wafer is moving in direction 405 or in the opposite direction. Fig. 32 illustrates a matrix of eight sensors, 15 however the method described above may be implemented with a different number of sensors. Furthermore, each sensor shown in Fig. 32 may be a differential pair of sensors, making a differential measurement at each point A-H.

[00142] The invention has been described by reference to certain embodiments discussed above. It should be noted various constructions and alternatives have been described, 20 which may be used with any of the embodiments described herein, as would be know by those of skill in the art. Furthermore, it will be recognized that these embodiments are susceptible to various modifications and alternative forms well known to those of skill in the art without departing from the spirit and scope of the invention. Accordingly, although specific embodiments have been described, these are examples only and are not limiting 25 upon the scope of the invention, which is defined in the accompanying claims.

Claims

A capacitive measuring system, comprising two or more capacitive sensors (30), one or more alternating current sources (306) for supplying energy to the capacitive sensors and a signal processing circuit (301) for processing a signal from the capacitive sensors wherein the sensors are paired and the one or more AC power source is included to supply energy to the first sensor of a pair of sensors with an AC or 180 degree out of phase voltage with the current or voltage of the second sensor in the pair of 10 sensors.

The capacitive measuring system according to claim 1, wherein the system comprises four or more capacitive sensors (30) which are incorporated in pairs and wherein each sensor of a first pair of sensors is supplied with energy by an alternating current or voltage which has been phase-out relative to a current or voltage which supplies energy to each sensor in an adjacent pair of sensors.

The capacitive measurement system according to claim 1 or 2, wherein the current or voltage supplying power to adjacent pairs of sensors is 90 degrees out of phase using a synchronous detection feature.

The capacitive measuring system according to any one of the preceding claims, further comprising an active field monitoring circuit (305) for supplying energy to a field monitoring electrode (32, 35) of at least one of the capacitive sensors, with the same potential as on a recording electrode ( 31) of the sensor.

5. The capacitive measuring system according to any of the preceding claims, further comprising a cable (350) for connecting the capacitive sensor (30) to the signal processing circuit and an active monitoring circuit (305) for supplying energy to a conductor ( 352) of the cable with the same potential as present on the sensor's receiving electrode (31).

6. The capacitive measuring system according to claim 5, wherein the cable (350) is a coaxial cable with a middle conductor (351) and an outer conductor (352), the middle conductor for electrically connecting the power source to the capacitive 2004052 sensors and the outer conductor for electrically connecting to the active protection circuit (305).

The capacitive measuring system according to claim 6, wherein the cable (350) comprises a triaxial cable with a shielding conductor (353) grounded at one end of the cable, remote from the sensors.

The capacitive measuring system according to any of the preceding claims, wherein each of the capacitive sensors comprises a thin-film structure comprising an insulating layer (34) and a first and a second conductive film formed on the first surface of the insulating layer with the first the conductive film comprises a recording electrode (31) and the second conductive film comprises a lateral field monitoring electrodes (32) which is formed near the recording electrode and defines a gap (39) between the field monitoring electrode and the recording electrode.

9. The capacitive measuring system according to claim 8, wherein the thin film structure of each sensor further comprises a third conductive film, comprising a back-field field monitoring electrode (35) formed on a second surface of the insulating layer (34).

The capacitive measuring system according to claim 9, wherein the lateral field monitoring electrodes (32) and the backside field monitoring electrode (35) are electrically connected.

The capacitive measuring system according to any of claims 8-10, wherein the thin film structure is flexible.

12. The capacitive measuring system according to any of the preceding claims 8-11, wherein each sensor further comprises a conductive protection electrode (40), separated from the back-field field monitoring electrode (35) by a second insulating layer (38).

The capacitive measuring system according to any of the preceding claims 8-12, wherein each sensor further comprises a connecting track formed of the first conductive layer for electrically connecting the receiving electrode (31) to a contact block (50, 50a) by a recess in the lateral field monitoring electrodes (32).

The capacitive measuring system according to any of claims 8-13, wherein each sensor further comprises an elongated connecting element (110), comprising a flexible membrane on which the conductive tracks (112-114) are imprinted or attached, where the conductive tracks are electrically connected are with the receiving electrode (31) and the lateral field monitoring electrodes (32) of the sensor on one end and a connector (160) on the other end.

The capacitive measuring system of claim 14, wherein the connecting element (110) of each sensor comprises active monitoring circuits.

The capacitive measuring system according to claim 14 or 15, wherein the conductive 10 tracks (112-114) are formed on an insulating layer (34) and wherein the insulating layer comprises a first surface on which the measuring electrode (31) and the lateral field monitoring electrodes (32) are formed and a second stretched surface on which the conductive tracks are formed.

The capacitive measuring system according to any of the preceding claims 1-7, wherein the capacitive sensors comprise a series of sensors (30) formed on an insulating base layer (34, 102, 120) and comprising a thin film structure, each sensor comprising a first conductive film and a second conductive film formed on a first surface of the insulating base layer, the first conductive layer comprises for each sensor a receiving electrode (31) and the second conductive layer comprises a lateral field monitoring electrodes (32), formed adjacent to the receiving electrode and defining a gap (39) between the protective electrode and the receiving electrode.

The capacitive measuring system of claim 17, wherein the insulating base layer (34, 102, 120) is flexible.

The capacitive measuring system according to any of the preceding claims 17 or 18, wherein each sensor further comprises a third conductive film, comprising a back-field field monitoring electrode (35) formed on a second surface of the insulating base layer.

The capacitive measuring system of claim 19, wherein the lateral field monitoring electrodes (32) and the backside field monitoring electrode (35) are electrically connected.

The capacitive measuring system according to any of the preceding claims 17-20, wherein the insulating base layer comprises a first surface where the measuring electrodes (31) and the side protection electrodes (32) of the sensors are formed, and a second stretched surface on which the conductive paths are formed, the conductive paths are electrically connected to the electrode at one end and to a connector (116) at the other end.

The capacitive measuring system according to any of the preceding claims, wherein the signal processing circuit (301) for processing a signal from the capacitive sensors comprises a synchronous detection circuit, which comprises an oscillator (331, 376) and a phase shifting circuit ( 377, 333) for generating a signal to drive the at least one alternating current power source (306).

The capacitive measurement system of claim 22, further comprising an automated calibration algorithm for adjusting the phase shift circuit (377, 333).

The capacitive measurement system of claim 23, wherein the automated calibration algorithm adjusts a phase shift of the phase shift circuit 20 (377, 333) to maximize the output of the mixer circuit (372).

The capacitive measurement system according to any of the preceding claims 22-24, wherein the automated calibration algorithm adjusts a phase shift of the phase shift circuit (377, 333) to be substantially equal to the phase shift that takes place in the measuring system.

The capacitive measuring system according to any of the preceding claims 22-25, wherein the automated calibration algorithm adjusts a signal for driving the at least one alternating current source (306). 2004052