WO2009039552A1

WO2009039552A1 - Reactive ion etching process for etching metals

Info

Publication number: WO2009039552A1
Application number: PCT/AU2007/001425
Authority: WO
Inventors: Darrell Larue Mcreynolds; C. S. Lakshmi; Kia Silverbrook
Original assignee: Silverbrook Research Pty Ltd
Priority date: 2007-09-26
Filing date: 2007-09-26
Publication date: 2009-04-02
Also published as: EP2205775A1; TW200914638A

Abstract

A method of etching a metal by a reactive ion etching process is provided. The etchant gas chemistry for the reactive ion etching process consists essentially of NH3. The process is particularly suitable for etching superalloys, which etch only slowly using conventional metal etching techniques.

Description

REACTIVE ION ETCHING PROCESS FOR ETCHING METALS

Field of the Invention This invention relates to a method of etching metals. It has been developed primarily to provide a method for etching metals that cannot be etched using conventional metal etching methods.

Background of the Invention CMOS and MEMS processes typically require one or more metal etching steps.

Conventionally, reactive ion etching (RIE) of metals uses a plasma of BCI3 and/or Cl₂. Indeed, the present Applicant has employed a gas chemistry of BCI₃ and Cl₂ for etching inkjet heater materials, such as titanium nitride and titanium aluminium nitride. This gas chemistry provides excellent etch rates and selectivity in most RIE processes for metals (including metal nitrides and the like).

In the development of improved heater elements for thermal inkjet (bubblejet) nozzles, the present Applicant has conducted considerable research effort in evaluating suitable material candidates. The resistive heaters in inkjet printheads operate in an extremely harsh environment. They must heat and cool in rapid succession to form bubbles in the ejectable liquid - usually a water soluble ink with a superheat limit of approximately 300⁰C. Under these conditions of cyclic stress, in the presence of hot ink, water vapor, dissolved oxygen and possibly other corrosive species, the heaters will increase in resistance and ultimately fail via a combination of oxidation, cracking and fatigue.

To protect against the effects of oxidation, corrosion and cavitation on the heater material, some inkjet manufacturers use protective layers, typically made from Si₃N₄, SiC and/or Ta. However, protective layers affect heat transfer from the heater element to the ink, and therefore reduce the overall efficiency of the printhead.

To this end, the Applicant developed inkjet heater elements formed of superalloys (See US Application Nos 11/482953 (Docket No. MTDOOlUS) and 11/482953 (Docket No. MTD002US) both filed July 10, 2006. Superalloys, such as Inconel® offer high temperature strength, corrosion and oxidation resistance far exceeding that of conventional thin film

MEMS31US heaters used in known thermal inkjet printheads. The primary advantage of superalloys is that they can provide sufficient strength, oxidation- and corrosion-resistance to allow heater operation without protective coatings. Thus, the energy wasted in heating a protective coating is removed from the design. As a result, the input energy required to form a bubble with a particular impulse is reduced, lowering the level of residual heat in the printhead and improving overall efficiency.

Inconel™ is a hard, nickel-based superalloy that has been shown by the present Applicant to have excellent efficacy in inkjet nozzles. However, whilst superalloys and other metals may have excellent properties for use in inkjet nozzles, the inherent hardness of such materials presents its own manufacturing problems. In particular, conventional metal RIE processes using BCI3/CI2 fail to etch such materials, or etch at rates that are too slow to be practicable.

An alternative metal etch process has been described by Nakatani and co-workers. US Patent No. 6,391,216 describes an RIE process employing a gas chemistry of CO and NH3. This gas chemistry is reported to be highly effective for etching hard metal alloys, such as Fe- Ni and Co-Cr alloys, which cannot be etched by conventional methods. Nakatani provides a rationale for etching transition metal alloys using a CO/NH₃ gas chemistry. It is understood that the formation of volatile transition metal carbonyl compounds, such as Fe(CCOs, Ni(CO)4, Mn₂(CO)I₀, Cr(CO)₆, V(CO)₆, Mo(CO)₆ and W(CO)₆, drives the etch. Further, NH₃ is believed to play the role of inhibiting disproportionation of CO into C and CO₂ in the plasma.

Whilst Nakatani has developed one method of etching hard, metal alloys, considerable problems still remain. Carbon monoxide, in its pure form, is a highly toxic gas requiring special safety precautions. In general, MEMS fabs are not equipped to plumb carbon monoxide into their plasma etching tools. Accordingly, it would be desirable to provide a process for etching metals, which is suitable for etching hard, metal alloys, and which does not employ highly toxic gases, such as carbon monoxide, requiring special safety provisions.

It would be further desirable to explore other uses of any new etching conditions, which may replace or complement existing MEMS processes.

MEMS31US Summary of the Invention

In a first aspect, there is provided a method of etching a metal by a reactive ion etching process, wherein an etchant gas chemistry for said reactive ion etching process consists essentially of NH₃. This method provides superior etching rates whilst avoiding the use of highly toxic gases, such as CO.

Optionally, the metal is an alloy or a superalloy, which are particularly difficult to etch by conventional methods.

Optionally, the metal has a microstructure with a grain size of less than 100 nanometres. Optionally, the metal has a Fe content of up to 60% by weight.

Optionally, the metal has a Ni content of between 25% by weight and 70% by weight.

Optionally, the metal has a Co content of between 35% by weight and 65% by weight.

Optionally, the metal is a superalloy comprising at least one metal selected from the group comprising: Cr, Al, Mo, Nb, Ta, Y, La, Ti, Fe, Ni and Co Optionally, the superalloy is of formula MCrAlX, where M is one or more of Ni,Co,Fe with M contributing at least 50% by weight, Cr contributing 8% and 35% by weight, Al contributing up to 8% by weight, and X contributing from 0 to 25% by weight, with X being selected from at least one of: Mo, Re, Ru, Ti, Ta, V, W, Nb, Zr, B, C, Si, Y and Hf.

Optionally, the superalloy comprises Ni, Fe, Cr and Al. Optionally, the metal is selected from:

INCONEL™ Alloy 600, Alloy 601, Alloy 617, Alloy 625, Alloy 625LCF, Alloy 690, Alloy 693, Alloy 718, Alloy X-750, Alloy 783, Alloy 725, Alloy 751, Alloy MA754, Alloy MA758, Alloy 925, or Alloy HX;

INCOLOY™ Alloy 330, Alloy 800, Alloy 800H, Alloy 800HT, Alloy MA956, Alloy A-286, or Alloy DS;

NIMONIC™ Alloy 75, Alloy 80A, or Alloy 90;

BRIGHTRAY^® Alloy B, Alloy C, Alloy F, Alloy S, or Alloy 35; or,

FERRY^® Alloy or Thermo-Span^® Alloy

Optionally, the gas chemistry additionally consists of an inert gas, selected from helium, argon and xenon. Alternativey, the etchant gas chemistry consists OfNH₃ only.

Optionally, etch regions of the metal are defined by an etch mask.

MEMS31US Optionally, the etch mask is a hard mask comprised of at least one of: silicon dioxide, silicon nitride, tantalum, tungsten, zirconium and hafnium.

Optionally, the etch mask comprises a layer of silicon nitride and a layer silicon nitride. Optionally, the method of the invention provides an etch rate of at least 200 Angstroms per minute.

Optionally, the method is a step of a MEMS fabrication process.

Optionally, the metal is a heater element for an inkjet nozzle assembly, and the method is a step of a printhead fabrication process.

Brief Description of the Drawings

Optional embodiments of the present invention will now be described, by way of example only with reference to the accompanying drawings in which:

Figure 1 is a schematic cross-sectional view through an ink chamber of a unit cell of a printhead with a suspended heater element at a particular stage during its operative cycle;

Figure 2 is a schematic cross-sectional view through the ink chamber Figure 1, at another stage of operation;

Figure 3 is a schematic cross-sectional view through the ink chamber Figure 1 , at yet another stage of operation; Figure 4 is a schematic cross-sectional view through the ink chamber Figure 1 , at yet a further stage of operation;

Figure 5 is a diagrammatic cross-sectional view through a unit cell of a printhead in accordance with an embodiment of the invention showing the collapse of a vapor bubble;

Figure 6 is a schematic cross-sectional view through an ink chamber of a unit cell of a printhead with a floor bonded heater element, at a particular stage during its operative cycle;

Figure 7 is a schematic cross-sectional view through the ink chamber of Figure 6, at another stage of operation;

Figure 8 is a schematic cross-sectional view through an ink chamber of a unit cell of a printhead with a roof bonded heater element, at a particular stage during its operative cycle; Figure 9 is a schematic cross-sectional view through the ink chamber of Figure 8, at another stage of operation;

MEMS31US Figures 10, 12, 14, 15, 17, 18, 20, 23, 25, 27, 28, 30, 32, 34 and 36 are schematic perspective views of a unit cell of a printhead in accordance with a suspended heater embodiment of the invention, at various successive stages in the production process of the printhead;

Figures 11, 13, 16, 19, 21, 24, 26, 28, 31, 33 and 35 are each schematic plan views of a mask suitable for use in performing the production stage for the printhead, as represented in the respective immediately preceding figures;

Figures 37 and 38 are a schematic section view and perspective view respectively of a partially complete second embodiment of the invention, wherein the passivation layer has been deposited on the CMOS; Figures 39, 40 and 41 are a perspective, mask and section view respectively showing the etch through the passivation layer to the top layer of the CMOS of the second embodiment;

Figures 42 and 43 are a perspective and section views respectively showing the deposition of the heater material of the second embodiment;

Figures 44, 45 and 46 are a perspective, mask and section view respectively showing the etch patterning the heater material of the second embodiment;

Figures 47, 48 and 49 are a perspective, mask and section view respectively showing the deposition of a photoresist layer and subsequent etch for the dielectric etch of the front ink hole;

Figures 50 and 51 are a perspective and section view respectively showing the dielectric etch into the wafer for the front ink hole;

Figures 52 and 53 are a perspective and section view respectively showing the deposition of a new photoresist layer;

Figures 54, 55 and 56 are a perspective, mask and section view respectively showing the patterning of the photoresist layer; Figures 57 and 58 are a perspective and section view respectively showing the deposition of the roof layer;

Figures 59, 60 and 61 are a perspective, mask and section view respectively showing the etch of the nozzle rims into the roof layer;

Figures 62, 63 and 64 are a perspective, mask and section view respectively showing the etch of the nozzle openings;

MEMS31US Figures 65 and 66 are a perspective and section view respectively showing the deposition of the protective photoresist overcoat;

Figures 67 and 68 are a perspective and section view respectively showing the back etch of the wafer; Figure 69 is a section view showing the release etch removing the remaining photoresist; Figure 70 is plan view of the completed unit cell of the second embodiment; and Figure 71 is a Weibull chart showing the reliability of a Inconel™ 718 heater element with a nanocrystalline microstructure compared to a TiAlN heater.

Detailed Description of the Invention

The invention will now be described in the context of fabrication of thermal inkjet nozzles. Two types of thermal inkjet nozzles, and their corresponding fabrication processes, will now be described.

InkJet Nozzles Having Suspended Heater Elements

With reference to Figures 1 to 4, the unit cell 1 of a printhead according to an embodiment of the invention comprises a nozzle plate 2 with nozzles 3 therein, the nozzles having nozzle rims 4, and apertures 5 extending through the nozzle plate. The nozzle plate 2 is plasma etched from a silicon nitride structure which is deposited, by way of chemical vapor deposition (CVD), over a sacrificial material which is subsequently etched.

The printhead also includes, with respect to each nozzle 3, side walls 6 on which the nozzle plate is supported, a chamber 7 defined by the walls and the nozzle plate 2, a multi-layer substrate 8 and an inlet passage 9 extending through the multi-layer substrate to the far side (not shown) of the substrate. A looped, elongate heater element 10 is suspended within the chamber 7, so that the element is in the form of a suspended beam. The printhead as shown is a microelectromechanical system (MEMS) structure, which is formed by a lithographic process which is described in more detail below.

When the printhead is in use, ink 11 from a reservoir (not shown) enters the chamber 7 via the inlet passage 9, so that the chamber fills to the level as shown in Figure 1. Thereafter, the

MEMS31US heater element 10 is heated for somewhat less than 1 microsecond (μs), so that the heating is in the form of a thermal pulse. It will be appreciated that the heater element 10 is in thermal contact with the ink 11 in the chamber 7 so that when the element is heated, this causes the generation of vapor bubbles 12 in the ink. Accordingly, the ink 11 constitutes a bubble forming liquid. Figure 1 shows the formation of a bubble 12 approximately 1 μs after generation of the thermal pulse, that is, when the bubble has just nucleated on the heater elements 10. It will be appreciated that, as the heat is applied in the form of a pulse, all the energy necessary to generate the bubble 12 is to be supplied within that short time.

Turning briefly to Figure 35, there is shown a mask 13 for forming a heater 14 (as shown in Figure 34) of the printhead (which heater includes the element 10 referred to above), during a lithographic process, as described in more detail below. As the mask 13 is used to form the heater 14, the shapes of several of its parts correspond to the shape of the element 10. The mask 13 therefore provides a useful reference by which to identify various parts of the heater 14. The heater 14 has electrodes 15 corresponding to the parts designated 15.34 of the mask 13 and a heater element 10 corresponding to the parts designated 10.34 of the mask. In operation, voltage is applied across the electrodes 15 to cause current to flow through the element 10. The electrodes 15 are much thicker than the element 10 so that most of the electrical resistance is provided by the element. Thus, nearly all of the power consumed in operating the heater 14 is dissipated via the element 10, in creating the thermal pulse referred to above.

When the element 10 is heated as described above, the bubble 12 forms along the length of the element, this bubble appearing, in the cross-sectional view of Figure 1, as four bubble portions, one for each of the element portions shown in cross section.

The bubble 12, once generated, causes an increase in pressure within the chamber 7, which in turn causes the ejection of a drop 16 of the ink 11 through the nozzle 3. The rim 4 assists in directing the drop 16 as it is ejected, so as to minimize the chance of drop misdirection.

MEMS31US The reason that there is only one nozzle 3 and chamber 7 per inlet passage 9 is so that the pressure wave generated within the chamber, on heating of the element 10 and forming of a bubble 12, does not affect adjacent chambers and their corresponding nozzles. However, it is possible to feed ink to several chambers via a single inlet passage as long as pressure pulse diffusing structures are positioned between chambers. The embodiment shown in Figures 37 to 70 incorporates these structures for the purpose of reducing cross talk to an acceptable level.

The advantages of the heater element 10 being suspended rather than embedded in any solid material, are discussed below. However, there are also advantages to bonding the heater element to the internal surfaces of the chamber. These are discussed below with reference to Figures 6 to 9.

Figures 2 and 3 show the unit cell 1 at two successive later stages of operation of the printhead. It can be seen that the bubble 12 generates further, and hence grows, with the resultant advancement of ink 11 through the nozzle 3. The shape of the bubble 12 as it grows, as shown in Figure 3, is determined by a combination of the inertial dynamics and the surface tension of the ink 11. The surface tension tends to minimize the surface area of the bubble 12 so that, by the time a certain amount of liquid has evaporated, the bubble is essentially disk- shaped.

The increase in pressure within the chamber 7 not only pushes ink 11 out through the nozzle 3, but also pushes some ink back through the inlet passage 9. However, the inlet passage 9 is approximately 200 to 300 microns in length, and is only about 16 microns in diameter. Hence there is a substantial inertia and viscous drag limiting back flow. As a result, the predominant effect of the pressure rise in the chamber 7 is to force ink out through the nozzle 3 as an ejected drop 16, rather than back through the inlet passage 9.

Turning now to Figure 4, the printhead is shown at a still further successive stage of operation, in which the ink drop 16 that is being ejected is shown during its "necking phase" before the drop breaks off. At this stage, the bubble 12 has already reached its maximum size and has then begun to collapse towards the point of collapse 17, as reflected in more detail in Figure 5.

MEMS31US The collapsing of the bubble 12 towards the point of collapse 17 causes some ink 11 to be drawn from within the nozzle 3 (from the sides 18 of the drop), and some to be drawn from the inlet passage 9, towards the point of collapse. Most of the ink 11 drawn in this manner is drawn from the nozzle 3, forming an annular neck 19 at the base of the drop 16 prior to its breaking off.

The drop 16 requires a certain amount of momentum to overcome surface tension forces, in order to break off. As ink 11 is drawn from the nozzle 3 by the collapse of the bubble 12, the diameter of the neck 19 reduces thereby reducing the amount of total surface tension holding the drop, so that the momentum of the drop as it is ejected out of the nozzle is sufficient to allow the drop to break off.

When the drop 16 breaks off, cavitation forces are caused as reflected by the arrows 20, as the bubble 12 collapses to the point of collapse 17. It will be noted that there are no solid surfaces in the vicinity of the point of collapse 17 on which the cavitation can have an effect.

Manufacturing process for suspended heater element

Relevant parts of the manufacturing process of a printhead according to embodiments of the invention are now described with reference to Figures 10 to 33.

Referring to Figure 10, there is shown a cross-section through a silicon substrate portion 21, being a portion of a Memjet™ printhead, at an intermediate stage in the production process thereof. This figure relates to that portion of the printhead corresponding to a unit cell 1. The description of the manufacturing process that follows will be in relation to a unit cell 1, although it will be appreciated that the process will be applied to a multitude of adjacent unit cells of which the whole printhead is composed.

Figure 10 represents the next successive step, during the manufacturing process, after the completion of a standard CMOS fabrication process, including the fabrication of CMOS drive transistors (not shown) in the region 22 in the substrate portion 21, and the completion of

MEMS31US standard CMOS interconnect layers 23 and passivation layer 24. Wiring indicated by the dashed lines 25 electrically interconnects the transistors and other drive circuitry (also not shown) and the heater element corresponding to the nozzle.

Guard rings 26 are formed in the metallization of the interconnect layers 23 to prevent ink 11 from diffusing from the region, designated 27, where the nozzle of the unit cell 1 will be formed, through the substrate portion 21 to the region containing the wiring 25, and corroding the CMOS circuitry disposed in the region designated 22.

The first stage after the completion of the CMOS fabrication process consists of etching a portion of the passivation layer 24 to form the passivation recesses 29.

Figure 12 shows the stage of production after the etching of the interconnect layers 23, to form an opening 30. The opening 30 is to constitute the ink inlet passage to the chamber that will be formed later in the process.

Figure 14 shows the stage of production after the etching of a hole 31 in the substrate portion 21 at a position where the nozzle 3 is to be formed. Later in the production process, a further hole (indicated by the dashed line 32) will be etched from the other side (not shown) of the substrate portion 21 to join up with the hole 31, to complete the inlet passage to the chamber. Thus, the hole 32 will not have to be etched all the way from the other side of the substrate portion 21 to the level of the interconnect layers 23.

If, instead, the hole 32 were to be etched all the way to the interconnect layers 23, then to avoid the hole 32 being etched so as to destroy the transistors in the region 22, the hole 32 would have to be etched a greater distance away from that region so as to leave a suitable margin (indicated by the arrow 34) for etching inaccuracies. But the etching of the hole 31 from the top of the substrate portion 21, and the resultant shortened depth of the hole 32, means that a lesser margin 34 need be left, and that a substantially higher packing density of nozzles can thus be achieved.

MEMS31US Figure 15 shows the stage of production after a four micron thick layer 35 of a sacrificial resist has been deposited on the layer 24. This layer 35 fills the hole 31 and now forms part of the structure of the printhead. The resist layer 35 is then exposed with certain patterns (as represented by the mask shown in Figure 16) to form recesses 36 and a slot 37. This provides for the formation of contacts for the electrodes 15 of the heater element to be formed later in the production process. The slot 37 will provide, later in the process, for the formation of the nozzle walls 6 that will define part of the chamber 7.

Figure 17 shows the stage of production after the deposition, on the layer 35, of a 0.5 micron thick layer 38 of heater material, which may be titanium aluminium nitride or a superalloy, such as Inconel™ 718, as described below.

Figure 18 shows the stage of production after patterning and etching of the heater material 38 to form the heater 14, including the heater element 10 and electrodes 15. In accordance with the present invention, the heater material 38 is etched using an NH₃ etchant gas chemistry. The NH₃ gas chemistry is particularly suitable for etching heater materials formed from hard, superalloys, such as Inconel™ 718. The gas chemistry contains no carbon monoxide and can readily be incorporated into fabs. Usually, the gas chemistry contains only NH3, although inert gases, such as helium and argon, may also be present. Etch rates of at least 200 Angstoms/min, at least 300 Angstroms/min or at least 400 Angstroms/min are typically achievable. Typically etch rates are at least 50% greater than etch rates using standard metal etch gas chemistries (e.g. BCI3/CI2). Moreover, these high etch rates are achievable without the use of highly toxic CO gas. In addition, carbon deposits resulting from the use of CO are avoided. Carbon deposits are particularly problematic since they have a micromasking effect and result in an undesirably high surface roughness of the etched metal.

The skilled person will appreciate that optimal etch conditions may be achieved by varying standard etch parameters, such as ICP power, bias, chamber pressure, temperature etc. Specific examples of suitable etch parameters, which have been used to etch Inconel™ 718, are detailed in Table 1 below. These etch parameters are readily achievable in any standard etching tool.

MEMS31US

Table 1

A hard mask is typically used for the NH₃ etch. Suitable hard mask materials include silicon dioxide, silicon nitride, tantalum, tungsten, zirconium and hafnium. Optionally, the hard mask comprises a layer of silicon nitride and a layer of silicon oxide. The hard mask may be etched to define etch regions of the underlying metal using any standard hard mask etch conditions. For example, a silicon nitride/silicon oxide hard mask may be etched using a standard C4F8/O2 etchant gas chemistry.

Figure 20 shows the stage of production after another sacrificial resist layer 39, about 1 micron thick, has been added.

Figure 22 shows the stage of production after a second layer 40 of heater material has been deposited. In a preferred embodiment, this layer 40, like the first heater layer 38, is of 0.5 micron thick and comprised of the same material as the first heater layer 38.

Figure 23 then shows this second layer 40 of heater material after it has been etched to form the pattern as shown, indicated by reference numeral 41. In this illustration, this patterned layer does not include a heater layer element 10, and in this sense has no heater functionality. However, this layer of heater material does assist in reducing the resistance of the electrodes 15 of the heater 14 so that, in operation, less energy is consumed by the electrodes which allows greater energy consumption by, and therefore greater effectiveness of, the heater elements 10.

MEMS31US Figure 25 shows the stage of production after a third layer 42, of sacrificial resist, has been deposited. The uppermost level of this layer will constitute the inner surface of the nozzle plate 2 to be formed later. This is also the inner extent of the ejection aperture 5 of the nozzle. The height of this layer 42 must be sufficient to allow for the formation of a bubble 12 in the region designated 43 during operation of the printhead. However, the height of layer 42 determines the mass of ink that the bubble must move in order to eject a droplet. In light of this, the printhead structure of the present invention is designed such that the heater element is much closer to the ejection aperture than in prior art printheads. The mass of ink moved by the bubble is reduced. The generation of a bubble sufficient for the ejection of the desired droplet will require less energy, thereby improving efficiency.

Figure 27 shows the stage of production after the roof layer 44 has been deposited, that is, the layer which will constitute the nozzle plate 2. Instead of being formed from 100 micron thick polyimide film, the nozzle plate 2 is formed of silicon nitride, just 2 microns thick.

Figure 28 shows the stage of production after the chemical vapor deposition (CVD) of silicon nitride forming the layer 44, has been partly etched at the position designated 45, so as to form the outside part of the nozzle rim 4, this outside part being designated 4.1

Figure 30 shows the stage of production after the CVD of silicon nitride has been etched all the way through at 46, to complete the formation of the nozzle rim 4 and to form the ejection aperture 5, and after the CVD silicon nitride has been removed at the position designated 47 where it is not required.

Figure 32 shows the stage of production after a protective layer 48 of resist has been applied. After this stage, the substrate portion 21 is then ground from its other side (not shown) to reduce the substrate portion from its nominal thickness of about 800 microns to about 200 microns, and then, as foreshadowed above, to etch the hole 32. The hole 32 is etched to a depth such that it meets the hole 31.

MEMS31US Then, the sacrificial resist of each of the resist layers 35, 39, 42 and 48, is removed using an oxidative plasma, to form the structure shown in Figure 34, with walls 6 and nozzle plate 2 which together define the chamber 7 (part of the walls and nozzle plate being shown cutaway). It will be noted that this also serves to remove the resist filling the hole 31 so that this hole, together with the hole 32 (not shown in Figure 34), define a passage extending from the lower side of the substrate portion 21 to the nozzle 3, this passage serving as the ink inlet passage, generally designated 9, to the chamber 7.

Figure 36 shows the printhead with the nozzle guard and chamber walls removed to clearly illustrate the vertically stacked arrangement of the heater elements 10 and the electrodes 15.

Bonded Heater Element Embodiments

Figures 6 to 9 schematically show two bonded heater embodiments; in Figures 6 and 7 the heater 10 is bonded to the floor of the chamber 7, and Figures 8 and 9 bonded the heater to the roof of the chamber. These figures generally correspond with Figures 1 and 2 in that they show bubble 12 nucleation and the early stages of growth. In the interests of brevity, figures corresponding to Figures 3 to 5 showing continued growth and drop ejection have been omitted.

Referring firstly to Figures 6 and 7, the heater element 10 is bonded to the floor of the ink chamber 7. In this case the heater layer 38 is deposited on the passivation layer 24 after the etching the passivation recesses 29 (best shown in Figure 10), before etching of the ink inlet holes 30 and 31 and deposition of the sacrificial layer 35 (shown in Figures 14 and 15). This re-arrangement of the manufacturing sequence prevents the heater material 38 from being deposited in the holes 30 and 31. In this case the heater layer 38 lies underneath the sacrificial layer 35. This allows the roof layer 50 to be deposited on the sacrificial layer 35, instead of the heater layer 38 as is the case in the suspended heater embodiments. No other sacrificial layers are required if the heater element 10 is bonded to the chamber floor, whereas suspended heater embodiments need the deposition and subsequent etching of the second sacrificial layer 42 above described with reference to Figures 25 to 35. To maintain the efficiency of the printhead, a low thermal product layer 25 can be deposited on the passivation layer 24 so that

MEMS31US it lies between the heater element 10 and the rest of the substrate 8. The thermal product of a material and its ability to thermally isolate the heater element 10 is discussed above and in greater detail below with reference to equation 3. However, in essence it reduces thermal loss into the passivation layer 24 during the heating pulse.

Figures 8 and 9 show the heater element 10 is bonded to the roof of the ink chamber 7. In terms of the suspended heater fabrication process described with reference to Figures 10 to 36, the heater layer 38 is deposited on top of the sacrificial layer 35, so the manufacturing sequence is unchanged until after the heater layer 38 is patterned and etched. At that point the roof layer 44 is then deposited on top of the etched heater layer 38, without an intervening sacrificial layer. A low thermal product layer 25 can be included in the roof layer 44 so that the heater layer 38 is in contact with the low thermal product layer, thereby reducing thermal loss into the roof 50 during the heating pulse.

Bonded Heater Element Manufacturing Process

The unit cells shown in Figures 6 to 9 are largely schematic and purposely correspond to the unit cells shown in Figures 1 to 4 where possible so as to highlight the differences between bonded and suspended heater elements. Figures 37 to 70 show the fabrication steps of a more detailed and complex bonded heater embodiment. In this embodiment, the unit cell 21 has four nozzles, four heater elements and one ink inlet. This design increases the nozzle packing density by supplying a plurality of nozzle chambers from a single ink inlet, using elliptical nozzle openings, thinner heater elements and staggering the rows of nozzles. The greater nozzle density affords greater print resolution.

Figures 37 and 38 show the partially complete unit cell 1. In the interests of brevity, this description begins at the completion of the standard CMOS fabrication on the wafer 8. The CMOS interconnect layers 23 are four metal layers with interlayer dielectric in between. The topmost metal layer, M4 layer 50 (shown in dotted line) has been patterned to form heater electrode contacts covered by the passivation layer 24. M4 layer is in fact made up of three layers; a layer if TiN, a layer of Al/Cu (>98% Al) and another layer of TiN which acts as an anti-reflective coating (ARC). The ARC stops light from scattering during subsequent

MEMS31US exposure steps. A TiN ARC has a resistivity suitable for the heater materials (discussed below).

The passivation layer may be a single silicon dioxide layer is deposited over the interconnect layers 23. Optionally, the passivation layer 24 can be a silicon nitride layer between two silicon dioxide layers (referred to as an "ONO" stack). The passivation layer 24 is planarised such that its thickness on the M4 layers 50 is preferably 0.5 microns. The passivation layer separates the CMOS layers from the MEMS structures and is also used as a hard mask for the ink inlet etch described below.

Figures 39 and 41 show the windows 54 etched into the passivation layer 24 using the mask 52 shown in Figure 40. As usual, a photoresist layer (not shown) is spun onto passivation layer 24. The clear tone mask 52 - the dark areas indicate where UV light passes through the mask - is exposed and the resist developed in a positive developing solution to remove the exposed photoresist. The passivation layer 24 is then etched through using an oxide etcher (for example, a Centura DPS (Decoupled Plasma Source) Etcher by Applied Materials). The etch needs to stop on the top, or partly into the TiN ARC layer but not the underlying Al/Cu layer. Then the photoresist layer (not shown) is stripped with O₂ plasma in a standard CMOS asher.

Figures 42 and 43 show the deposition of a 0.2 micron layer of heater material 56. Suitable heater materials, such as TiAl, TiAlN and Inconel™ 718, are discussed elsewhere in the specification. As shown in Figures 44 and 46, the heater material layer 56 is etched using the clear tone mask 58 shown in Figure 45. As described above, an RIE process employing an NH₃ gas chemistry is suitable for this etching step, especially when the heater material 56 is comprised of a superalloy, such as Inconel™ 718. Suitable etch conditions were described in detail above. Typically, a hard mask is used to define the etch.

After this step, a layer photoresist 42 is again spun onto the wafer 8 as shown in Figure 47. The dark tone mask 60 (dark areas block the UV light) shown in Figure 48, exposes the resist which is then developed and removed to define the position of the ink inlet 31 on the

MEMS31US passivation layer 24. As shown in Figure 49, the removal of the resist 42 at the site of the ink inlet 31 exposes the passivation layer 24 in preparation for the dielectric etch.

Figures 50 and 51 shows the dielectric etch through the passivation layer 24, the CMOS interconnect layers 23 and into the underlying wafer 8. This is a deep reactive ion etch

(DRIE) using any standard CMOS etcher (e.g. Applied Materials Centura DPS (Decoupled

Plasma Source) Etcher), and extends about 20 microns to 30 microns into the wafer 8. In the embodiment shown, the front side ink inlet etch is about 25 microns deep. The accuracy of the front side etch is important as the backside etch (described below) must be deep enough to reach it in order to establish an ink flow path to the nozzle chamber. After the front side etch of the ink inlet 31, the photoresist 42 is ashed away with O₂ plasma (not shown).

Once the photoresist layer 42 is removed, another layer of photoresist 35 is spun onto the wafer as shown in Figures 52 and 53. The thickness of this layer is carefully controlled as it forms a scaffold for the subsequent deposition of the chamber roof material (described below). In the present embodiment, the photoresist layer 35 is 8 microns thick (except where it plugs the ink inlet 31 as best shown in Figure 53). Next the photoresist layer 35 is patterned according to the mask 62 shown in Figure 55. The mask is a clear tone mask in that the dark areas indicate the areas of exposure to UV light. The exposed photoresist is developed and removed so that the layer 35 is patterned in accordance with Figure 54. Figure 56 is a section view of the patterned photoresist layer 35.

With the photoresist 35 defining the chamber roof and support walls, a layer of roof material, such as silicon nitride, is deposited onto the sacrificial scaffolding. In the embodiment shown in Figures 57 and 58, the layer of roof material 44 is 3 microns thick (except at the walls or column features).

Figure 59, 60 and 61 show the etching of the nozzle rims 4. A layer of photoresist (not shown) spun onto the roof layer 44 and expose under the clear tone mask 64 (the dark areas are exposed to UV). The roof layer 44 is then etched to a depth of 2 microns leaving the

MEMS31US raised nozzle rims 4 and the bubble vent feature 66. The remaining photoresist is then ashed away.

Figures 62, 63 and 64 show the nozzle aperture etch through the roof layer 44. Again, a layer of photoresist (not shown) is spun onto the roof layer 44. It the then patterned with the dark tone mask 68 (clear areas exposed) and then developed to remove the exposed resist. The underlying SiN layer is then etched with a standard CMOS etcher down to the underlying layer of photoresist 35. This forms the nozzle apertures 3. The bubble vent hole 66 is also etched during this step. Again the remaining photoresist is removed with O₂ plasma.

Figures 65 and 66 show the application of a protective photoresist overcoat 74. This prevents the delicate MEMS structures from being damaged during further handling. Likewise, the scaffold photoresist 35 is still in place to provide the roof layer 44 with support.

The wafer 8 is then turned over so that the 'backside' 70 (see Figure 67) can be etched. Then the front side of the wafer 8 (or more specifically, the photoresist overcoat 74) is stuck to a glass handle wafer with thermal tape or similar. It will be appreciated that wafers are initially about 750 microns thick. To reduce the thickness, and therefore the depth of etch needed to establish fluid communication between the front and the back of the wafer, the reverse side 70 of the wafer is ground down until the wafer is about 160 microns thick and then DRIE etched to remove any pitting in the ground surface. The backside is then coated with a photoresist layer (not shown) in preparation for the channel 32 etching. The clear tone mask 72 (shown in Figure 68) is positioned on the back side 70 for exposure and development. The resist then defines the width of the channel 32 (about 80 microns in the embodiment shown). The channels 32 are then etched with a DRIE (Deep Reactive Ion Etch) down to and marginally beyond the plugged front side ink inlets 31. The photoresist on the backside 72 is then ashed away with O₂ plasma, and the wafer 8 is again turned over for the front side ashing of the protective overcoat 74 and the scaffold photoresist 35.

Figures 69 and 70 show the completed unit cell 1. While Figure 70 is a plan view, the features obscured by the roof have been shown in full line for the purposes of illustration.

MEMS31US In use, ink is fed from the backside 70 into the channel 32 and into the front side inlet 31. Gas bubbles are prone to form in the ink supply lines to the printhead. This is due to outgassing where dissolved gasses come out of solution and collect as bubbles. If the bubbles are fed into the chambers 7 with the ink, they can prevent ink ejection from the nozzles. The compressible bubbles absorb the pressure generated by the nucleating bubbles on the heater elements 10 and so the pressure pulse is insufficient to eject ink from the aperture 3. As the ink primes the chambers 7, any entrained bubbles will tend to follow the columnar features on either side of the ink inlet 31 and be pushed toward the bubble vent 66. Bubble vent 66 is sized such that the surface tension of the ink will prevent ink leakage, but trapped gas bubbles can vent. Each heater element 10 is enclosed on three sides by chamber walls and by additional columnar features on the fourth side. These columnar features diffuse the radiating pressure pulse to lower cross-talk between chambers 7.

Superalloy Heaters

Superalloys are a class of materials developed for use at elevated temperatures. They are usually based on elements from Group VIIA of the Periodic Table and predominantly used in applications requiring high temperature material stability such as jet engines, power station turbines and the like. Their suitability in the thermal inkjet realm has until now gone unrecognized. Superalloys can offer high temperature strength, corrosion and oxidation resistance far exceeding that of conventional thin film heaters (such as tantalum aluminium, tantalum nitride or hafnium diboride) used in known thermal inkjet printheads. The primary advantage of superalloys is that they can have sufficient strength, oxidation and corrosion resistance to allow heater operation without protective coatings, so that the energy wasted in heating the coatings is removed from the design.

Testing has indicated that superalloys can in some cases have far superior lifetimes compared to conventional thin film materials when tested without protective layers. Figure 71 is a Weibull Plot of heater reliability for two different heater materials tested in open pool boiling (the heaters are simply actuated in an open pool of water i.e. not within a nozzle). Skilled artisans will appreciate that Weibull charts are a well recognized measure of heater reliability.

MEMS31US The chart plots the probability of failure, or unreliability, against a log scale of the number of actuations. It should be noted that the Key shown in Figure 71 also indicates the number of failed and suspended data points for each alloy. For example, F = 8 below Inconel 718 in the key indicates that eight of the heaters used in the test were tested to the point of open circuit failure, while S = I indicates that one of the test heaters was suspended or in other words, still operating when the test was suspended. The known heater material, TiAlN is compared with the superalloy Inconel 718. The registered trademark Inconel is owned by Huntington Alloys Canada Ltd 2060 Flavelle Boulevard, Mississauga, Ontario L5K 1Z9 Canada.

The applicant's prior work indicates that oxidation resistance is strongly correlated with heater lifetime. Adding Al to TiN to produce TiAlN greatly increased the heater's oxidation resistance (measured by Auger depth profiling of oxygen content after furnace treatment) and also greatly increased heater lifetime. The Al diffused to the surface of the heater and formed a thin oxide scale with a very low diffusivity for further penetration of oxygen. It is this oxide scale which passivates the heater, protecting it from further attack by an oxidative or corrosive environment, permitting operation without protective layers. Sputtered Inconel 718 also exhibits this form of protection and also contains Al, but has two other advantageous properties that further enhance oxidation resistance; the presence of Cr, and a nanocrystalline structure.

Chromium behaves in a similar fashion to aluminium as an additive, in that it provides self passivating properties by forming a protective scale of chromium oxide. The combination of Cr and Al in a material is thought to be better than either in isolation because the alumina scale grows more slowly than the chromia scale, but ultimately provides better protection The Cr addition is beneficial because the chromia scale provides short term protection while the alumina scale is growing, allowing the concentration of Al in the material required for short term protection to be reduced. Reducing the Al concentration is beneficial because high Al concentrations intended for enhanced oxidation protection can jeopardize the phase stability of the material.

MEMS31US X-ray diffraction and electron microscope studies of the sputtered Inconel 718 showed a crystalline micro structure, with a grain size less than lOOnm (a "nano crystalline" micro structure). The nano crystalline microstructure of Inconel 718 is beneficial in that it provides good material strength yet has a high density of grain boundaries. Compared to a material with much larger crystals and a lower density of grain boundaries, the nanocrystalline structure provides higher diffusivity for the protective scale forming elements Cr and Al (more rapid formation of the scale) and a more even growth of the scale over the heater surface, so the protection is provided more rapidly and more effectively. The protective scales adhere better to the nanocrystalline structure, which results in reduced spalling. Further improvement in the mechanical stability and adherence of the scale is possible using additives, such as rare earth metals e.g. yttrium, lanthanum etc.

It should be noted that superalloys are typically cast or wrought and this does not yield a nanocrystalline microstructure: the benefits provided by the nanocrystalline structure are specific to the sputtering technique used in the MEMS heater fabrication of this application. It should also be noted that the benefits of superalloys as heater materials are not solely related to oxidation resistance: their microstructure is carefully engineered with additives to encourage the formation of phases that impart high temperature strength and fatigue resistance. Potential additions comprise the addition of aluminium, titanium, niobium, tantalum, hafnium or vandium to form the gamma prime phase of Ni based superalloys; the addition of iron, cobalt, chrome, tungsten, molybdenum, rhenium or ruthenium to form the gamma phase or the addition of C, Cr, Mo, W, Nb, Ta, Ti to form carbides at the grain boundaries. Zr and B may also be added to strengthen grain boundaries. Controlling these additives, and the material fabrication process, can also act to suppress undesirable age- induced Topologically Close Packed (TCP) phases, such as sigma, eta, mu phases which can cause embrittlement, reducing the mechanical stability and ductility of the material. Such phases are avoided as they may also act to consume elements that would otherwise be available for the favoured gamma and gamma prime phase formation. Thus, while the presence of Cr and Al to provide oxidation protection is preferred for the heater materials, superalloys in general can be considered a superior class of materials from which selection of heater material candidates may be made, since considerably more effort has been put into

MEMS31US designing them for high temperature strength, oxidation and corrosion resistance than has been put into improving the conventional thin film heater materials used in MEMS.

The Applicant's results indicate that superalloys : a Cr content between 2% by weight and 35% by weight; a Al content of between 0.1% by weight and 8% by weight; a Mo content of between 1% by weight and 17% by weight; a Nb+Ta content of between 0.25% by weight and 8.0% by weight; a Ti content of between 0.1% by weight and 5.0% by weight; a Fe content of up to 60% by weight; a Ni content of between 26% by weight and 70% by weight; and or, a Co content of between 35% by weight and 65% by weight; are suitable for use as a thin film heater element within a MEMS bubble generator (e.g. suspended heater element, bonded heater element and so on).

Superalloy's having the generic formula MCrAlX where:

M is one or more of Ni,Co,Fe with M contributing at least 50% by weight; Cr contributing between 8% and 35% by weight; Al contributing more than zero but less than 8% by weight; and, X contributing less than 25% by weight, with X consisting of zero or more of Mo, Re,

Ru, Ti, Ta, V, W, Nb, Zr, B, C, Si, Y, Hf; provide good results in open pool testing (described above).

In particular, superalloys with Ni, Fe, Cr and Al together with additives comprising zero or more of Mo, Re, Ru, Ti, Ta, V, W, Nb, Zr, B, C, Si, Y, or Hf, show superior results.

Using these criteria, suitable superalloy materials for thermal inkjet printhead heaters may be selected from:

INCONEL™ Alloy 600, Alloy 601, Alloy 617, Alloy 625, Alloy 625LCF, Alloy 690, Alloy 693, Alloy 718, Alloy X-750, Alloy 725, Alloy 751, Alloy MA754, Alloy MA758, Alloy 783, Alloy 925, or Alloy HX;

MEMS31US INCOLOY™ Alloy 330, Alloy 800, Alloy 800H, Alloy 800HT, Alloy MA956, Alloy A-286, or Alloy DS;

NIMONIC™ Alloy 75, Alloy 80A, or Alloy 90; BRIGHTRAY^® Alloy B, Alloy C, Alloy F, Alloy S, or Alloy 35; or, FERRY^® Alloy or Thermo- Span^® Alloy

Brightray, Ferry and Nimonic are the registered trademarks of Special Metals Wiggin Ltd Holmer Road HEREFORD HR4 9FL UNITED KINGDOM.

Thermo-Span is a registered trademark of CRS holdings Inc., a subsidiary of Carpenter Technology Corporation

MEMS31US

Claims

1. A method of etching a metal by a reactive ion etching process, wherein an etchant gas chemistry for said reactive ion etching process consists essentially OfNH₃.

2. The method of claim 1 , wherein said metal is an alloy.

3. The method of claim 1 , wherein said metal is a superalloy .

4. The method of claim 1 , wherein said metal has a microstructure with a grain size of less than 100 nanometres.

5. The method of claim 1, wherein said metal has a Fe content of up to 60% by weight.

6. The method of claim 5, wherein said metal has a Ni content of between 25% by weight and 70% by weight.

7. The method of claim 5, wherein said metal has a Co content of between 35% by weight and 65% by weight.

8. The method of claim 3, wherein said superalloy comprises at least one metal selected from the group comprising: Cr, Al, Mo, Nb, Ta, Y, La, Ti, Fe, Ni and Co

9. The method of claim 3, wherein the superalloy is of formula MCrAlX, where M is one or more of Ni,Co,Fe with M contributing at least 50% by weight, Cr contributing 8% and 35% by weight, Al contributing up to 8% by weight, and X contributing from 0 to 25% by weight, with X being selected from at least one of: Mo, Re, Ru, Ti, Ta, V, W, Nb, Zr, B, C, Si, Y and Hf.

10. The method of claim 3, wherein said superalloy comprises Ni, Fe, Cr and Al.

MEMS31US

11. The method of claim 1 , wherein said metal is selected from:

INCONEL™ Alloy 600, Alloy 601, Alloy 617, Alloy 625, Alloy 625LCF, Alloy 690, Alloy 693, Alloy 718, Alloy X-750, Alloy 783, Alloy 725, Alloy 751, Alloy MA754, Alloy MA758, Alloy 925, or Alloy HX; INCOLOY™ Alloy 330, Alloy 800, Alloy 800H, Alloy 800HT, Alloy MA956, Alloy A-286, or Alloy DS;

NIMONIC™ Alloy 75, Alloy 80A, or Alloy 90; BRIGHTRAY^® Alloy B, Alloy C, Alloy F, Alloy S, or Alloy 35; or, FERRY^® Alloy or Thermo-Span^® Alloy

12. The method of claim 1, wherein said gas chemistry additionally consists of an inert gas, selected from helium, argon and xenon.

13. The method of claim 1, wherein said etchant gas chemistry consists OfNH₃ only.

14. The method of claim 1 , wherein etch regions of said metal are defined by an etch mask.

15. The method of claim 14, wherein said etch mask is a hard mask.

16. The method of claim 15, wherein said hard mask is comprised of at least one of: silicon dioxide, silicon nitride, tantalum, tungsten, zirconium and hafnium.

17. The method of claim 15, wherein said etch mask comprises a layer of silicon nitride and a layer silicon nitride.

18. The method of claim 1, wherein an etch rate is at least 200 Angstroms per minute.

19. The method of claim 1, wherein said method is a step of a MEMS fabrication process.

MEMS31US

20. The method of claim 1 , wherein said metal is a heater element for an inkjet nozzle assembly, said method being a step of a printhead fabrication process.

MEMS31US