DROPLET EJECTOR

Abstract

A droplet ejector for a printhead comprises: a substrate having a mounting surface and an opposite nozzle surface; a nozzle-forming layer formed on at least a portion of the nozzle surface of the substrate; a fluid chamber defined at least in part by the substrate and at least in part by the nozzle-forming layer, the fluid chamber having a fluid chamber outlet defined at least in part by a nozzle portion of the said nozzle-forming layer, the said nozzle portion comprising an inner portion located closer to the fluid chamber outlet and an outer portion located closer to a periphery of the nozzle portion; and either or both of an inner actuator arrangement formed on the inner portion of the nozzle portion of the nozzle-forming layer and an outer actuator arrangement formed on the outer portion of the nozzle portion of the nozzle-forming layer.

Claims

1. A droplet ejector for a printhead, the droplet ejector comprising: a substrate having a mounting surface and an opposite nozzle surface; a nozzle-forming layer formed on at least a portion of the nozzle surface of the substrate; a fluid chamber defined at least in part by the substrate and at least in part by the nozzle-forming layer, the fluid chamber having a fluid chamber outlet defined at least in part by a nozzle portion of the said nozzle-forming layer, the said nozzle portion comprising an inner portion located closer to the fluid chamber outlet and an outer portion located closer to a periphery of the nozzle portion; and either or both of an inner actuator arrangement formed on the inner portion of the nozzle portion of the nozzle-forming layer and an outer actuator arrangement formed on the outer portion of the nozzle portion of the nozzle-forming layer.

2. The droplet ejector according to claim 1, wherein the outer portion of the nozzle portion of the nozzle-forming layer at least partially surrounds the inner portion of the nozzle portion of the nozzle-forming layer.

3. The droplet ejector according to claim 1, wherein the inner actuator arrangement at least partially surrounds the fluid chamber outlet.

4. The droplet ejector according to claim 1, wherein both the inner and/or outer actuator arrangements are substantially annular.

5. (canceled)

6. The droplet ejector according to claim 1 comprising an inner actuator arrangement which comprises one or more inner piezoelectric actuators, at least one of said one or more inner piezoelectric actuators comprising an inner piezoelectric body provided between an inner pair of drive electrodes.

7. The droplet ejector according to claim 6, wherein the inner actuator arrangement consists of a single inner piezoelectric actuator which is substantially annular.

8. The droplet ejector according to claim 1 comprising an outer actuator arrangement which comprises one or more outer piezoelectric actuators, at least one of said one or more outer piezoelectric actuators comprising an outer piezoelectric body provided between an outer pair of drive electrodes.

9. The droplet ejector according to claim 8, wherein the outer actuator arrangement consists of a single outer piezoelectric actuator which is substantially annular.

10. The droplet ejector according to claim 9, wherein the single outer piezoelectric actuator surrounds the single inner piezoelectric actuator.

11. (canceled)

12. The droplet ejector according to claim 6, wherein the inner piezoelectric body or bodies and/or the outer piezoelectric body or bodies comprise one or more piezoelectric materials processable at a temperature below 450° C.

13. The droplet ejector according to claim 6, wherein the inner piezoelectric body or bodies and/or the outer piezoelectric body or bodies comprise one or more piezoelectric materials depositable at a temperature below 450° C.

14. The droplet ejector according to claim 12, wherein the one or more piezoelectric materials are PVD-deposited piezoelectric materials.

15. The droplet ejector according to claim 12, wherein the one or more piezoelectric materials comprise aluminium nitride and/or zinc oxide.

16.-17. (canceled)

18. The droplet ejector according to claim 12, wherein the one or more piezoelectric materials are non-ferroelectric piezoelectric materials.

19. The droplet ejector according to claim 6, wherein the inner piezoelectric body or bodies and/or the outer piezoelectric body or bodies have d.sub.31 piezoelectric constants having magnitudes less than 20 pC/N.

20. (canceled)

21. The droplet ejector according to claim 1, wherein the mounting surface of the substrate comprises a fluid inlet aperture in fluid communication with the fluid chamber.

22. The droplet ejector according to claim 1, wherein the fluid chamber is substantially cylindrical and the nozzle portion of the nozzle-forming layer is substantially annular.

23. (canceled)

24. A printhead comprising a plurality of droplet ejectors according to claim 1.

25. The printhead according to claim 24, wherein the plurality of droplet ejectors share a common substrate.

26. A printer comprising one or more printheads according to claim 24.

27. A method of actuating a droplet ejector according to claim 1, the method comprising: actuating the inner actuator arrangement and/or actuating the outer actuator arrangement to thereby cause displacement of at least a portion of the nozzle portion of the nozzle-forming layer and consequently ejection of fluid from the fluid chamber through the fluid chamber outlet.

28. The method according to claim 27, wherein the droplet ejector comprises both an inner actuator arrangement and an outer actuator arrangement, the method comprising: actuating both the inner actuator arrangement and the outer actuator arrangement to thereby cause displacement of at least a portion of the nozzle portion of the nozzle-forming layer and consequently ejection of fluid from the fluid chamber through the fluid chamber outlet.

29. The method according to claim 28, wherein the steps of actuating the inner actuator arrangement and actuating the outer actuator arrangement take place concurrently.

30.-32. (canceled)

Description

DESCRIPTION OF THE DRAWINGS

[0163] An example embodiment of the present invention will now be illustrated with reference to the following Figures in which:

[0164] FIG. 1 is a view of a monolithic fluid droplet ejector device including integrated fluidics, electronic circuitry, nozzles and actuators according to a first embodiment;

[0165] FIG. 2 is a cross-sectional view of the monolithic droplet ejector device along the line F2 shown in FIG. 1;

[0166] FIG. 3 is a plan view of a nozzle showing features of the monolithic droplet ejector shown in FIG. 1 with a protective coating removed;

[0167] FIGS. 4(a) and 4(b) show a schematic of drive pulse implementations for the droplet ejector device of FIG. 1;

[0168] FIG. 5 is a schematic of the manufacturing process flow for manufacturing the droplet ejector device of FIG. 1;

[0169] FIG. 6 is a cross-sectional view showing an alternative implementation of the electrode structure according to a second example embodiment of the invention;

[0170] FIG. 7 is a schematic showing an alternative drive pulse implementation for the droplet ejector device of FIG. 6;

[0171] FIG. 8 is a schematic showing a cross section through an alternative implementation of the nozzle structure according to a third example embodiment of the invention;

[0172] FIG. 9 is a cross-sectional view showing an alternative implementation of bond pad structures according to a fourth example embodiment of the invention;

[0173] FIG. 10 is a cross-sectional view through the nozzle structure on actuation of any of the droplet ejector devices of FIG. 1, FIG. 6, FIG. 8 or FIG. 9;

[0174] FIG. 11 provides both a cross-sectional view and a plan view of showing an alternative monolithic droplet ejector having only an inner actuator arrangement according to a fifth example embodiment of the invention;

[0175] FIG. 12 is a cross-sectional view through the nozzle structure on actuation of the droplet ejector device of FIG. 11;

[0176] FIG. 13 provides both a cross-sectional view and a plan view of showing an alternative monolithic droplet ejector having only an outer actuator arrangement according to a sixth example embodiment of the invention;

[0177] FIG. 14 is a cross-sectional view through the nozzle structure on actuation of the droplet ejector device of FIG. 13;

[0178] FIG. 15 is a plot of showing the volume swept by a droplet ejector device diaphragm as a function of the location of the actuator arrangement;

[0179] FIG. 16 shows in 3D the shape assumed by a diaphragm of a droplet ejector device according to FIGS. 1, 6, 8, 9, 11 and 13 on actuation;

[0180] FIG. 17 is a plot showing the deflection of the droplet ejector diaphragm for four different actuation implementations; and

[0181] FIG. 18 is a plot showing the deflection of the droplet ejector diaphragm for two different actuator configurations as a function of location of the actuator arrangements on the diaphragm.

[0182] FIG. 19 shows a table listing some common piezoelectric materials and the manufacturing methods associated with them, along with typical d.sub.31 values.

DETAILED DESCRIPTION OF AN EXAMPLE EMBODIMENT

First Example Embodiment

[0183] The first example embodiment is described with reference to FIGS. 1 to 5 and FIGS. 10 and 11.

[0184] FIG. 1 shows a monolithic fluid droplet ejector device 1 including integrated fluidics, electronic circuitry, nozzles and actuators according to the first example embodiment of the invention. FIG. 2 is a cross sectional view of the monolithic droplet ejector device 1 along the line F2 shown in FIG. 1.

[0185] As shown in FIG. 1 and FIG. 2, the fluid droplet ejector device is a monolithic chip that includes a substrate 100, fluid inlet channel 101, electronic circuitry 200, interconnect layer 300 comprising wiring, inner piezoelectric actuator 400, outer piezoelectric actuator 450, nozzle plate 500, protective front surface 600, nozzle 601 and bond pad 700. FIG. 1 shows a bond pad region 102 and a nozzle region 103.

[0186] The substrate 100 is typically between 20 and 1000 micrometres in thickness. The interconnect layer 300, inner piezoelectric actuator 400, outer piezoelectric actuator 450, nozzle plate 500 and protective front surface 600 are typically between 0.5 and 5 micrometres in thickness. The nozzle 601 is typically between 3 and 50 micrometres in diameter. The fluid inlet channel 103 has a characteristic dimension of between 50 and 800 micrometres.

[0187] The monolithic chip shown in FIG. 1 comprises 4 rows of nozzles. Each row is offset relative to adjacent rows in an alternating pattern. Any number of nozzle rows in different configurations are possible. The arrangement of the nozzles on the chip is configured to achieve a target print density (i.e. number of dots per inch (dpi)), a target firing frequency and/or a target print speed. A range of different nozzle configurations are possible which satisfy the particular printing requirements. Different printhead nozzle configurations are effected by arranging individual nozzle and nozzle specific drive electronics 201 and 202.

[0188] The substrate 100 is formed from a silicon wafer and comprises a supporting body 102, fluid inlet channels 101 and electronic circuitry 200.

[0189] The fluid inlet channels 101 are formed through the thickness of the substrate 100 with an opening at one surface at a fluid inlet 103 and are terminated at the other end by the nozzle plate 500 and nozzles 601. The walls of the fluid inlet channels 101 have a similar cross section through the substrate 100 and interconnect layer 300. The fluid inlet channels 101 are substantially cylindrical (i.e. substantially circular in cross section in the plane of the substrate). The corners of the fluid inlet channels 101, at the interface with the nozzle plate and at the fluid inlet interface, are rounded to minimize stress concentrations.

[0190] The electronic circuitry 200 is formed on the opposite surface of the substrate 100 to the surface that includes the fluid inlets 103. The electronic circuitry 200 can include digital and/or analog circuitry. Portions of the electronic circuitry, 201 and 202, are connected directly to the inner and outer piezoelectric actuators 400 and 450 by way of wiring 301 and 302 through the interconnect layer 300 and are located close to the actuators 400 and 450 to optimize the application of a drive wave form. The electrode actuator wiring interconnects 301 and 302 may be a continuous single construction or they may be constructed from multiple layers of wiring. The drive electronics may be configured to apply a set voltage or shaped voltage to the piezoelectric actuators for a set period of time.

[0191] Portions of the electronic circuitry 203 are associated with the overall operation of the entire monolithic droplet ejector device and can be located separate to the actuator drive circuitry 201 and 202. The circuitry 203 associated with the general operation of the chip can perform a range of functionalities including data routing, authentication, chip monitoring (e.g. chip temperature monitoring), lifecycle management, yield information processing and/or dead nozzle monitoring. The circuitry 203 is connected to the bond pads 700 and the specific electrode drive circuitry 201 and 202 through the interconnect layer 300. The chip drive electronics 203 may include analog and/or digital circuits configured to perform different functions such as data caching, data routing, bus management, general logic, synchronization, security, authentication, power routing and/or input/output. The chip drive electronics 203 may comprise circuitry components such as timing circuitry, interface circuitry, sensors and/or clocks.

[0192] There may be a number of general drive electronics areas located in different sections of the chip—for example between nozzle rows or around the periphery of the chip.

[0193] The electronic drive circuitry includes 200 CMOS drive circuitry.

[0194] The interconnect layer 300 is formed directly on top of the electronics circuitry 200 and the substrate 100 and comprises electrical insulator and wiring. Wiring in the interconnect layer 300 connects chip electronic circuitry 203 to both the bond pads 700 and to the actuator electrode drive circuitry 201 and 202. The interconnect layer 300 includes power and data routing wiring which is routed between nozzles, around the periphery of the chip and/or over drive electronics. The interconnect layer 300 typically comprises multiple layers having different wiring paths.

[0195] A nozzle plate 500 is formed on top of the interconnect layer 300. The nozzle plate 500 is formed from either a single material or a laminate of multiple materials. The nozzle plate 500 is continuous across the front surface of the chip with electrical openings for wiring between the interconnect layer 300 below and actuator electrodes 401 above.

[0196] The nozzle plate 500 is formed from one or more materials which must be manufacturable with the CMOS electronic drive circuitry 200 in terms of deposition temperatures, compositions, and chemical processing steps. The nozzle plate materials must also be chemically stable and impervious to the jetted fluids. The nozzle plate materials must also be compatible with the functioning of the piezoelectric actuator. For example, the Young's modulus of suitable materials lies in the range of 70 GPa to 300 GPa. However, variations in Young's modulus can be accommodated by changing the thickness of the nozzle plate 500. Example nozzle plate materials include one or more of (e.g. including combinations and/or laminates of) silicon dioxide (SiO.sub.2), silicon nitride (Si.sub.3N.sub.4), silicon carbide (SiC) and silicon oxynitride (SiO.sub.xN.sub.y).

[0197] Each outer piezoelectric actuator 450 comprises a laminate of a first electrode 451, a piezoelectric layer 452 and a second electrode 453. The first electrode 451 is attached to the nozzle plate 500. The piezoelectric layer 452 is attached to the first electrode 451. The second electrode 403 is attached to the piezoelectric layer surface opposite the first electrode attachment surface. The first electrode 451 is electrically connected to a wiring connection 301 in the interconnect layer 300. The second electrode 453 is electrically connected to a wiring connection 302 in the interconnect layer 300. The first electrode 451 and second electrode 453 are electrically isolated from each other.

[0198] Each inner piezoelectric actuator 400 comprises a laminate of a first electrode 401, a piezoelectric layer 402 and a second electrode 403. The first electrode 401 is attached to the nozzle plate 500. The piezoelectric layer 402 is attached to the first electrode 401. The second electrode 403 is attached to the piezoelectric layer surface opposite the first electrode attachment surface. The first electrode 401 is electrically connected to the second electrode 453 of the outer piezoelectric actuator. The second electrode 403 is electrically connected to the first electrode 451 of the outer piezoelectric actuator. The first electrode 401 and second electrode 403 of the inner piezoelectric actuator are electrically isolated from each other.

[0199] The electrode materials are electrically conductive and are typically formed from metals or intermetallic compounds such as titanium (Ti), aluminium (Al), titanium-aluminide (TiAL), tungsten (W) or platinum (Pt), or alloys thereof. These materials are manufacturable (in terms of deposition temperature and chemical process compatibility) with CMOS drive circuitry and the piezoelectric layer.

[0200] The piezoelectric layers 402 and 452 are formed from materials chosen for compatibility with the manufacture of CMOS and interconnect circuitry. CMOS drive circuitry can typically survive a temperature of up to about 450° C. However, high yield manufacturing requires a much lower peak manufacturing temperature, typically 300° C. Deposition methods that subject the CMOS drive electronics to temperatures over a duration can degrade performance, typically affecting dopant mobility and the degradation of wiring within the interconnect layer. The temperature limit restricts deposition methods for the piezoelectric layers. Suitable piezoelectric materials include aluminium nitride (AlN), aluminium nitride compounds (in particular scandium aluminium nitride (ScAlN)) and zinc oxide (ZnO), which are compatible with CMOS electronics. The composition of the piezoelectric material is chosen to optimise the piezoelectric properties. For example, the concentrations of any additional elements in aluminium nitride compounds (such as the concentration of scandium in scandium aluminium nitride) are typically chosen to optimise the magnitude of the d.sub.31 piezoelectric constant. The higher the concentration of scandium in scandium aluminium nitride, the typically larger the value of d.sub.31. The mass percentage of scandium in scandium aluminium nitride may be as high as 50%.

[0201] The piezoelectric actuator material is not continuous over the surface of the nozzle plate 500. The piezoelectric material is located primarily over the nozzle plate and includes a number of openings including electrode openings 404 and a region around the nozzle 405.

[0202] The protective front surface 600 is formed on the outer surface of the droplet ejector device 100 and covers the piezoelectric layers 402 and 452, the electrodes 401, 403, 451 and 453 and the nozzle plate 500. The protective front surface has openings for the nozzles 601 and for the bond pads 700. The protective front surface material is chemically inert and impermeable. The protective front surface material may also be repellent to the fluid to be ejected. The mechanical properties of the protective front surface material are chosen carefully to minimize the effect on the forcing action of the piezoelectric actuators 400 and 450 and nozzle plate 500. The protective front surface material is chosen to be manufacturable with a CMOS compatible process flow, for example in terms of processing temperature and chemical process compatibility. The protective front surface 600 prevents contact of fluid with any of the electrodes or the piezoelectric layers. Suitable protective front surface materials include polyimides, polytetrafluoroethylene (PTFE), diamond-like carbon (DLC) or related materials.

[0203] FIG. 3 is a plan view of a nozzle showing features of the monolithic droplet ejector structure 1 with the protective coating 600 removed according to the first embodiment. The dashed line shows the underlying position of the fluid inlet 103 in relation to the piezoelectric inner actuator 400 and the outer piezoelectric actuator 450.

[0204] In use, the fluid droplet ejector device 1 is mounted on a substrate that can supply fluid to the fluid inlet 103. Fluid pressure is typically slightly negative at the fluid inlet 103 and the fluid inlet channels 101 typically “prime” or fill with fluid by surface tension driven capillary action. The nozzles 601 prime up to the outer surface of the protective front surface 600 due to capillary action once the fluid inlets 103 are primed. The fluid does not move onto the outer surface of the protective surface 600 past the nozzles 601 due to the combination of negative fluid pressure and the geometry of the nozzle 601.

[0205] The actuator drive circuitry 201 and 202 controls the application of a voltage pulse to the drive electrodes 401, 403, 451 and 453, according to a timing signal from the overall drive circuitry 203. The application of electrode voltage across the piezoelectric material layers 402 and 452 creates two electric fields. The electric fields cause deformation of the piezoelectric material layers 402 and 452. The deformation can either be a tensile or compressive strain depending on the orientation of the electric field with respect to the local direction of polarization in the material. The induced strain caused by the expansion or contraction of the piezoelectric materials 402 and 452 typically induces a strain gradient through the thickness of the nozzle plate 500, piezoelectric actuators 400 and 450 and the protective front layer 600, causing a movement or displacement of the nozzle plate relative to a neutral position.

[0206] The piezoelectric properties of the piezoelectric materials can be characterized in part by the transverse piezoelectric constant d.sub.31. d.sub.31 is the particular component of the piezoelectric coefficient tensor which relates the electric field applied across the piezoelectric material in a first direction to the strain induced in the piezoelectric material along a second direction perpendicular to said first direction. The piezoelectric actuators 400 and 450 shown are configured such that the applied electric fields induce strains in the material layers in directions perpendicular to the directions in which the fields are applied and are therefore characterized by the d.sub.31 constant.

[0207] Due to the uniform thickness and composition of both piezoelectric material layers 402 and 452, and due to the electrical cross-connections between electrodes 403 and 451 and electrodes 401 and 453, the application of a constant voltage or a voltage pulse results in a first potential difference being applied across the inner actuator layer and a second potential difference being applied across the outer actuator layer, wherein the first and second potential differences are equal in magnitude but opposite in polarity. Expressed in a different way, an electric field E.sub.1 is set up across the inner actuator piezoelectric layer and an electric field E.sub.2 is set up across the outer actuator piezoelectric layer, wherein E.sub.1 and E.sub.2 are equal in magnitude but act in opposite directions. Because E.sub.1 and E.sub.2 act in opposite directions, the inner and outer actuator layers deform in opposite senses. Dependent on the polarity of E.sub.1 and E.sub.2, displacement X of the nozzle plate 500 is either positive or negative relative to a neutral position (i.e. when there are no applied electric fields). A positive displacement of the nozzle plate is shown in the upper portion of FIG. 4(a) whereas a negative displacement of the nozzle plate is shown in the lower portion of the figure.

[0208] The application of pulsed electric fields can cause oscillations of the nozzle plate 500. Oscillation of the nozzle plate typically induces a pressure in the fluid inlet 103 under the nozzle plate 500 which causes droplet ejection out of the nozzle 601. The frequency and amplitude of the nozzle plate oscillation is primarily a function of the mass and stiffness characteristics of the nozzle plate 500, piezoelectric actuators 400 and 450, the protective layer 600, the fluid properties (for example, the fluid density, fluid viscosity (either Newtonian or non-Newtonian) and surface tension), nozzle and fluid inlet geometries and the configuration of both drive pulses.

[0209] FIGS. 4(a) and 4(b) show two drive pulse implementations. Voltage pulses across the inner actuator electrodes 401 and 403 are shown in the diagram. It will be understood that voltage pulses equal in magnitude but opposite in polarity are simultaneously applied across outer actuator electrodes 451 and 453.

[0210] In a first implementation, the application of a steady state or DC electric field across the electrode pairs causes a distortion of the piezoelectric layers 402 and 452 and a steady state deflection of the nozzle plate away from the fluid inlet as shown in the upper portion of FIG. 4 (a). The fluid pressure under the nozzle plate is the same as the fluid inlet supply pressure. Strain energy is stored in the nozzle plate 500, the piezoelectric actuators 400 and 450 and the protective layer 600.

[0211] The electric fields are then removed and a reverse electric field pulse is applied as shown in the lower portion of FIG. 4 (a). This causes both a release of the stored strain energy and further distortion of the piezoelectric materials in the opposite direction. The nozzle plate moves towards the fluid inlet, which causes a positive pressure in the fluid inlet and nozzle region and droplet ejection out of the nozzle 601. The reverse electric field pulse may come immediately after the removal of the DC field or at a slightly delayed duration.

[0212] The final removal of the electric fields across the piezoelectric materials causes the nozzle plate 500 to return to a neutral position with no induced strain.

[0213] The application of electric fields of opposing polarity across the inner and outer actuators causes the nozzle plate to deform into the shape shown in FIG. 10. The nozzle plate in the region of the inner actuator curves in an opposite sense relative to the curvature of the nozzle plate in the region of the outer actuator, resulting in a sigmoidal cross-section. This particular shape significantly increases the maximum displacement of the nozzle portion of the nozzle plate from the neutral position when compared to the displacement achievable when a nozzle plate is provided with only one actuator causing curvature in only one sense. By increasing the maximum displacement of the nozzle plate away from the neutral position, a much greater ejection force can be exerted when the applied field is removed or reversed in polarity. This enables the use of piezoelectric materials having low d.sub.31 constants, which are normally considered unsuitable for use in inkjet printers due to the low forces they are capable of generating. These low-d.sub.31 materials are typically processable at lower temperatures, enabling closer integration of the droplet ejector with CMOS components. The larger ejection forces achievable also permit the overall ejector size to be reduced so that increased printhead nozzle densities are possible.

[0214] In a second implementation, the DC electric field configuration described in FIG. 4(a) with a pulse field configuration as shown in FIG. 4(b). This has the advantage of minimizing any applied strain effects over longer durations. An additional advantage of the dual pulsed approach is enabled by the timing of the field pulse switching application. The application of the first pulse will induce an oscillation with an initial nozzle plate movement away from the fluid inlet as shown in the upper portion of FIG. 4(b). This oscillation will introduce a negative fluid pressure under the nozzle plate which introduces a net fluid flow towards the nozzle which can additionally augment the fluid ejection flows through the nozzle.

[0215] FIG. 5 is a schematic showing the manufacturing process flow for the droplet ejector device. The first manufacturing step, as shown in FIG. 5(a), is to create drive circuitry and the interconnect layer 300, for example CMOS drive circuitry and interconnects, on a surface of a silicon wafer substrate. CMOS drive circuitry is formed by standard processes—for example ion implantation on p-type or n-type substrates followed by the creation of a wiring interconnect layer by standard CMOS fabrication processes (e.g. ion implantation, chemical vapour deposition (CVD), physical vapour deposition (PVD), etching, chemical-mechanical planarization (CMP) and/or electroplating).

[0216] Subsequent manufacturing steps are implemented to define features and structures of the monolithic droplet ejector device. Subsequent steps are chosen not to damage structures formed in previous steps. A key manufacturing parameter is the peak processing temperature. Problems associated with processing CMOS at high temperatures include the degradation of dopant mobility and interconnect wiring schemes. CMOS electronics are known to survive temperatures of 450° C. However, a much lower temperature (i.e. below 300° C.) is desirable for high yield.

[0217] The nozzle plate 500, the piezoelectric actuators 400 and 450, the protective layer 600 and the bond pads 700 are formed on top of the interconnect layer as shown in FIG. 5(b).

[0218] The nozzle plate 500 is deposited using a CVD or PVD process.

[0219] The formation of a CMOS compatible piezoelectric material 402 and 452 is of particular interest as this is the key driving element of the actuator. FIG. 19 shows Table 1 which lists some common piezoelectric materials and the manufacturing methods associated with them, along with typical d.sub.31 values. It can be seen that materials with the highest d.sub.31 values are incompatible with manufacture of monolithic CMOS structures. Materials that are compatible with CMOS structures have low d.sub.31 values and hence a much lower forcing capability.

[0220] As can be seen from the table, although lead zirconate titanate (PZT) can be deposited by PVD (including sputtering) at low temperatures, it subsequently requires a post process anneal at a temperature above the allowable temperature for CMOS. PZT can also be deposited by sol gel methods, but this again requires a high temperature anneal above the CMOS limit. PZT also has a very slow rate of deposition that is not viable commercially. PZT additionally contains lead, which is undesirable environmentally.

[0221] ZnO, AlN and AlN compounds (such as ScAlN) materials can be deposited using low-temperature PVD (e.g. sputtering) processes that do not require post processing such as annealing. These materials also do not require poling. A poling step is required for PZT, wherein the material is subjected to a very high electric field which orients all the electric dipoles in the direction of the field.

[0222] ZnO, AlN and AlN compounds (e.g. ScAlN) materials are therefore commercially viable materials for the fabrication of a monolithic droplet ejector device. However, the value of d.sub.31 for these materials is significantly lower than that of PZT. The particular configuration of the nozzle (i.e. the actuatable nozzle plate), which improves ejection efficiency, and the use of two pairs of control electrodes, which improves actuation efficiency, counter the lower d.sub.31 value associated with these materials.

[0223] Actuator electrode materials are deposited using a CMOS compatible process such as PVD (including low-temperature sputtering). Typical electrode materials may include titanium (Ti), platinum (Pt), aluminium (Al), tungsten (W), molybdenum (Mo) or alloys thereof. The electrodes are defined by standard patterning and etch methods.

[0224] Protective materials can be deposited and patterned using a spin on and cure method (suitable for polyimides or other polymeric materials). Some materials, such as PTFE, may require more specific deposition and patterning approaches.

[0225] Bond pads are deposited using methods such as CVD or PVD (e.g. sputtering).

[0226] The fluid inlet channels are defined using high aspect ratio Deep Reactive Ion Etching (DRIE) methodologies as shown in FIG. 5(c). The fluid inlets are aligned to the nozzle structures using a wafer front-back side alignment tool. The wafer may be mounted on a handle wafer during the front-back alignment and etch steps.

[0227] The DRIE approach may also be used to singulate the die, however, other approaches may be used such as a wafer saw.

Second Example Embodiment

[0228] FIG. 6 is a cross sectional view showing an alternative implementation of the electrode structure. In this embodiment the electrodes 403 and 453 are connected by wiring, 302, to a ground line 204 rather than drive circuitry. The ground line 204 is located within the interconnect layer 300 and is connected to the drive circuitry region 203 or directly to grounded bond pads 700.

Third Example Embodiment

[0229] FIG. 7 is a schematic showing an alternative drive pulse implementation compatible with this droplet ejector device. A voltage pulse, as shown in FIG. 7, is applied to only one electrode of each electrode pair, for example 401 and 453. This creates an electric field through the piezoelectric actuators 400 and 450 that creates a downward overall displacement of the nozzle plate 500. It is also possible to configure the device with a drive pulse applied to electrodes 403 and 451 and a ground voltage applied to electrode 401 and 453.

Fourth Example Embodiment

[0230] FIG. 8 is a schematic showing a cross section of an alternative implementation of the nozzle structure and shows the extension of the interconnect layer 304 attached to the nozzle plate layer 500 in the vicinity of the fluid inlet 101. The interconnect layer extension 304 may comprise solely dielectric material without any wiring. In another variation, the device has no nozzle plate layer and only an interconnect layer attached to the piezoelectric actuator.

Fifth Example Embodiment

[0231] FIG. 9 is a cross-sectional view showing an alternative implantation of the bond pad structures. The protective front surface has been removed in the vicinity of the bond pads 701. This geometry improves accessibility of external wiring schemes and reduces the overall height of wire bonding above the height of the chip.

Sixth and Seventh Example Embodiments

[0232] FIG. 11 is a schematic showing a cross section and plan view of an alternative implementation of the nozzle structure which includes only an inner piezoelectric actuator 400 adjacent the fluid outlet 601. In this embodiment, the piezoelectric material only extends between the electrodes 401 and 402 and does not extend beyond the electrodes over the remainder of the nozzle plate layer 500 (i.e. it does not extend into the region 450 where an outer piezoelectric actuator might be expected to be located).

[0233] The application of an electric field across the inner actuator causes the nozzle plate to deform into the shape shown in FIG. 12. Actuation of the inner actuator causes the inner portion of the nozzle plate to curve in a first sense. The outer portion of the nozzle plate in response curves in an opposite sense, resulting in a sigmoidal cross-section. This particular shape significantly increases the maximum displacement of the nozzle portion of the nozzle plate from the neutral position when compared to the displacement achievable when a nozzle plate is provided with only one actuator extending over the majority of the nozzle plate, which typically causes curvature in only one sense.

[0234] In addition, FIG. 13 is a schematic showing a cross section and plan view of an alternative implementation of the nozzle structure which includes only an outer piezoelectric actuator 450 adjacent the fluid outlet 601. In this embodiment, the piezoelectric material only extends between the electrodes 401 and 402 and does not extend beyond the electrodes over the remainder of the nozzle plate layer 500 (i.e. it does not extend into the region 400 where an inner piezoelectric actuator might be expected to be located).

[0235] The application of an electric field across the outer actuator causes the nozzle plate to deform into the shape shown in FIG. 12. Actuation of the outer actuator causes the outer portion of the nozzle plate to curve in a first sense. The inner portion of the nozzle plate in response curves in an opposite sense, resulting in a sigmoidal cross-section. This particular shape significantly increases the maximum displacement of the nozzle portion of the nozzle plate from the neutral position when compared to the displacement achievable when a nozzle plate is provided with only one actuator extending over the majority of the nozzle plate, which typically causes curvature in only one sense.

[0236] FIG. 15 shows the volume swept by the nozzle plate on actuation as a function of the radial location of a single annular actuator positioned symmetrically about the fluid outlet. In this case the layer of piezoelectric material extends across the entire nozzle plate and the location of the actuator is defined by the location of first and second actuator electrodes. The nozzle plate has an outer radius of 125 microns. It can be seen from this Figure that the maximum swept volume (and therefore fluid ejection) is achievable for an actuator located close to the outer periphery (at a location 105 microns from the centre) of the nozzle plate. FIG. 16 shows the 3D shape taken up by the nozzle plate on actuation of a single annular actuator located closer to the outer periphery. The inner portion of the nozzle plate can be seen to curve in an opposite sense from the outer portion of the nozzle plate.

[0237] FIG. 17 shows how the deflection of the nozzle plate from a neutral position (i.e. before actuation of any actuators) varies as a function of radial location across the nozzle plate in embodiments comprising both inner and outer piezoelectric actuators. The Figure shows data sets for: “Reversed Polarity” (both inner and outer annular actuators are provided, each being actuated concurrently by electric fields having opposed polarities); “Similar Polarity” (both inner and outer annular actuators are provided, each being actuated concurrently by electric fields having the same polarity); “Inner only” (both inner and outer annular actuators are provided, but only the inner actuator is actuated); and “Outer only” (both inner and outer annular actuators are provided, but only the outer actuator is actuated). In such embodiments, maximum deflection is achieved when electric fields having opposing polarities are applied to the inner and outer actuators.

[0238] FIG. 18 also shows how the deflection of the nozzle plate from the neutral position varies as a function of radial location across the nozzle plate for embodiments comprising only a single piezoelectric actuator in which the piezoelectric material does not extend beyond said piezoelectric actuator. In such embodiments, maximum deflection is achieved when an inner actuator is provided. The lack of piezoelectric material in the region not containing an actuator leads to increased flexibility and therefore potentially greater deflections can be achieved by ejectors incorporating a single annular piezoelectric actuator (whether inner or outer) compared to ejectors incorporating both inner and outer piezoelectric actuators.

[0239] Further variations and modifications may be made within the scope of the invention herein disclosed.

[0240] The device may be formed on a silicon wafer substrate. Alternatively, the substrate may comprise a silicon-on-insulator wafer or III-V semiconductor wafer.

[0241] The fluid inlet channels may be substantially cylindrical and therefore have substantially circular cross-sections in the plane of the substrate. Alternatively, the fluid inlet channels may take a variety of other cross-sections including multiple-sided, regular or irregular shapes. The shape of the fluid inlet channels is typically dependent on other aspects of the monolithic chip design such as the layout of nozzles, the drive electronics placement and the wiring routing in the interconnect layer 300.

[0242] The cross-sectional shapes may also be selected to minimize the width of the printhead chip without introducing failure mechanisms. Failure mechanisms may be structural (for example, too many fluid inlets may reduce the robustness of the chip) or they may be operational (for example, interconnect wires may be insufficient to carry the appropriate current). A reduced printhead width is desirable because it increases the number of chips which can be manufactured on a single wafer.

[0243] Further variations and modifications may be made within the scope of the invention herein disclosed.