Micromachined machine parts and their manufacturing method

Abstract

The present invention relates to a method for fabricating one or more freestanding microfabricated components, which method involves reactive ion etching of a photoresist and a tungsten-based layer supported on a carrier substrate to define one or more microfabricated components, and separating the one or more microfabricated components from the carrier substrate to free-standing the components. The present invention also relates to a tungsten-based microprobe obtained by the method, the microprobe having a substantially square or rectangular cross-section perpendicular to the longitudinal axis of the microprobe. The present invention also relates to a probe card including a plurality of such microprobes.

Description

[Technical Field]

[0001] MICROMACHINED MECHANICAL COMPONENTS AND METHODS FOR MANUFACTURING SAME FIELD OF THE INVENTION The present invention relates to micromachined mechanical components and methods for their manufacture, including particularly, but not exclusively, freestanding micromachined probes suitable for use in applications such as integrated circuit (IC) testing. [Background technology]

[0002] Micromechanical components (generally defined as components with at least one linear dimension in the range of 1 μm to 100 μm) have a wide range of applications. However, one important use of such components is as microprobes suitable for use in applications such as integrated circuit (IC) testing. For IC testing, the probes are attached to a probe card, which acts as an interface between an electronic test system and a semiconductor wafer. The purpose of the probe card is to electrically connect the circuitry on the wafer to the test system, allowing for testing and verification of the circuitry at the wafer level.

[0003] There is a growing demand for probe cards with finer probe pitches and improved manufacturing precision. This allows for more devices to be tested per wafer while reducing costs, and also allows for more I / O test pads per device, enabling more detailed testing. Furthermore, new IC technologies require improved probe performance in the radio frequency (RF) band and higher current carrying capacity (CCC).

[0004] In an attempt to address some of these issues, some work has been done to fabricate small-scale probes. For example, International Publication No. 2005 / 043594 discloses two different methods for forming photo-defined micro-electrical contacts. The first method involves forming probes from flat stock of thin beryllium copper. Respective layers of photoresist are deposited directly on opposite surfaces of a planar CuBe alloy sheet. The photoresist layers are then exposed (using a mask) and developed, leaving positive images of multiple probes on each surface of the alloy. Each surface is then chemically (wet) etched to form multiple probes within the CuBe alloy. This technique takes advantage of the "undercut" resulting from the wet etching process to produce the cross-sectional profile shown in Figure 10. A second method is also disclosed in which negative images of multiple probes are formed on a stainless steel mandrel using photoresist and a mask. Next, Ni or NiCo alloy is plated onto the exposed areas of the steel mandrel using electroforming techniques, thereby forming multiple Ni or NiCo probes on the exposed surface of the mandrel. Neither of these methods is suitable for forming tungsten into microprobes. Furthermore, the first method requires careful placement of masks on opposite sides of a planar CuBe sheet. Therefore, the resulting microprobes must have a large tolerance for shape due to the difficulty of precisely aligning the two sets of masks.

[0005] However, the material of the microfabricated components is important for many applications. For microprobe applications, tungsten has many unique properties. For example, tungsten has high electrical conductivity, which reduces resistive losses when used as a microprobe, thereby reducing self-heating (Joule heating) of the probe. This allows for higher currents to be carried in a given probe design. Additionally, tungsten has a high melting point, allowing the probe to operate at higher temperatures. Finally, tungsten is hard, resulting in less wear on the probe needle, resulting in a longer lifespan than other materials. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] International Publication No. 2005 / 043594 [Non-patent literature]

[0007] [Non-Patent Document 1] Shenglin et al.(2016) Summary of the Invention [Problem to be solved by the invention]

[0008] However, a significant problem to overcome is that it has not been possible to reliably manufacture fine-pitch microprobes (or other similarly sized micromachined components) from tungsten using conventional prior art methods. Three main technologies exist for making such products: a. Wire drawing - tungsten parts can be drawn, but the reproducibility of the part shape is poor and the manufacturing cost is high; b. Stamping - the minimum size of parts that can be produced is limited, with part widths on the order of 100 μm, which is not suitable for many micromechanical applications (and in the case of microprobes, not for fine-pitch applications); c. Electroplating - While electroplating allows for good control of geometric size, tungsten cannot be electroplated. Therefore, this process is only suitable for more easily electroplatable materials such as nickel or palladium alloys (as in WO 2005 / 043594, supra). These materials have inferior properties compared to tungsten.

[0009] Although some research has been conducted on producing tungsten microneedles using deep reactive ion etching (Shenglin et al. (2016) - Non-Patent Document 1), this process has many problems. First, this process uses deep reactive ion etching to deeply etch bulk tungsten, forming an array of upstanding members (microneedles). The length of each microneedle is determined by the etching depth. Thus, the microneedles are necessarily attached as extensions of the bulk material. The resulting microneedles cannot be separated from the bulk tungsten base to provide standalone individual microneedles. Therefore, it is not possible to fabricate free-standing microfabricated components using this method. Furthermore, because this process uses a hard aluminum mask to define the shape of the microneedle array, there is a potential problem of Al redeposition during the DRIE process. Therefore, it is not suitable for mass production of components. Finally, the resulting microneedles are described as having a sidewall slope angle of approximately 14°. Therefore, it is not possible to fabricate components with vertical or near-vertical sidewalls, which are required (or advantageous) for some applications of microfabricated components. The present invention was devised in light of the above considerations. [Means for solving the problem]

[0010] The present inventors have discovered that the use of reactive ion etching of a tungsten layer in the fabrication of microfabricated components can reduce or eliminate some or all of the above problems. Accordingly, a first aspect of the present invention is a method for fabricating one or more freestanding microfabricated components, comprising the steps of providing a tungsten-based layer supported on a carrier substrate, applying a photoresist, exposing the photoresist through a mask having a predetermined shape, and removing portions of the photoresist to provide a photoresist mask on or above the tungsten-based layer, reactive ion etching the photoresist and the tungsten-based layer to define one or more microfabricated components, and separating the one or more microfabricated components from the carrier substrate.

[0011] As used herein, a "tungsten-based layer" is defined as a layer that comprises, consists essentially of, or consists of tungsten (W). Preferably, tungsten is present in the layer at an atomic percent of 50 at% or greater. In other words, tungsten is preferably the major component of the tungsten-based layer. Tungsten may be present in the layer at an atomic percent of 60 at% or greater, 70 at% or greater, 80 at% or greater, 90 at% or greater, 95 at% or greater, or 99 at% or greater. One or more other elements may be present as desired. For example, the tungsten-based layer may be a tungsten-based alloy, such as a tungsten-rhenium alloy (W—Re alloy). The use of a W—Re alloy layer may be particularly advantageous in certain applications of the method of the present invention, particularly in the fabrication of microprobes. While W—Re microprobes may have slightly reduced bulk electrical conductivity compared to pure W microprobes, they experience less tip oxidation, resulting in reduced contact resistance over time compared to pure W microprobes. When the tungsten-based layer is a W-Re alloy, Re may be present in the range of about 1 at% to 10 at%, more preferably about 3 at% to about 5 at%, where Re exists as a solid solution in W.

[0012] Reactive ion etching (RIE) is a dry etching technique that uses a chemically reactive plasma to remove material from the layer being etched. This chemically active plasma is generated under low pressure (vacuum) by an electromagnetic field, and high-energy ions from the plasma attack and react with the material surface. As a dry etching technique, RIE has characteristics that distinguish it from wet etching techniques. In particular, the nearly vertical delivery of reactive ions allows reactive ion etching to produce highly anisotropic etching profiles, as opposed to the isotropic profiles typically produced as a result of wet chemical etching techniques.

[0013] The inventors have discovered that etching tungsten with RIE allows for the production of freestanding micromachined components on a relatively large scale at a relatively low cost per component. Furthermore, the use of RIE can enable the production of micromachined components with lithographic precision and fine component details with sub-micrometer accuracy compared to existing techniques used for tungsten micromachining (e.g., when the micromachined components are microprobes, with fine probe pitch and near-vertical probe sidewalls). The present invention enables the production of tungsten-based components with a level of precision not possible using existing techniques (e.g., wire extrusion and sheet stamping processes described in the "Background" section above). As used herein, the term "freestanding" is generally defined as a part that, in practical use, does not require attachment or support from another structure. The part may be attached to or supported by another structure. This is in contrast to non-freestanding parts, which must always be connected to or supported by another structure due to their manufacturing method or inability to support their own weight. It will be appreciated that microfabricated components according to the present invention may initially be fabricated as freestanding components, but may be attached to or supported by one or more other components in subsequent use.

[0014] The tungsten-based layer may be provided as a foil layer. In this specification, the term "foil layer" refers to a thin, self-supporting metal plate, typically manufactured by hammering or rolling (although other manufacturing methods are also possible). The foil layer may have a thickness ranging, for example, from 1 μm to 100 μm, more preferably from 5 μm to 50 μm. The foil layer may be provided as an initially freestanding layer that is transferred to a carrier substrate so as to be supported on the carrier substrate. In other words, the foil layer is not deposited directly on the carrier substrate. Alternatively, the tungsten-based layer may be provided as a deposited layer, possibly deposited directly on the carrier substrate by a process such as chemical vapor deposition (CVD).

[0015] Providing the tungsten-based layer as a foil layer may be preferable to providing it as a deposited layer. Providing the tungsten-based layer as a foil layer has many advantages over using a deposited tungsten-based layer, including increased manufacturing efficiency and reduced processing costs due to reduced processing time and complexity. Furthermore, using a tungsten-based layer that is a foil layer allows for greater control over the layer thickness, resulting in greater repeatability of the part thickness. This allows for the fabrication of multiple micromachined parts where the thickness of each part is substantially consistent or varies only slightly (they do not vary by more than 1 μm relative to any other of the multiple micromachined parts).

[0016] The method according to the present invention can include a step of polishing the tungsten-based layer. Polishing can be performed on one or both surfaces of the tungsten-based layer. If the tungsten-based layer is initially provided as a free-standing layer (e.g., a foil layer), polishing one or both surfaces (defined herein as the major surfaces of the layer) of the tungsten-based layer can be performed before the tungsten-based layer is supported on a carrier substrate. Alternatively, polishing one surface (the free surface) of the tungsten-based layer can be performed "in situ" while the tungsten-based layer is supported on a carrier substrate.

[0017] A polishing step can be advantageous regardless of the morphology of the tungsten-based layer, but it can be particularly advantageous when the tungsten-based layer is provided as a foil layer, since foil layers tend to have significant roughness due to the manufacturing process. For example, rolled tungsten foil may have an initial arithmetic mean roughness Ra on the order of 100 nm and a maximum peak-to-valley roughness Rt of more than 1 μm. This roughness can cause problems in further process steps and, additionally or alternatively, can affect the mechanical and electrical performance of the resulting microfabricated component. Polishing the tungsten-based layer can reduce or avoid such problems.

[0018] The polishing step may include one or more different polishing substeps, including one or both of a grinding step and / or a chemical-mechanical polishing (CMP) step. In a preferred method, the tungsten-based layer may first be subjected to a grinding step to remove most of the roughness before being polished using a chemical-mechanical polishing (CMP) step. One or both major surfaces of the tungsten-based layer may be polished to a surface roughness of 5 nm or less Ra and / or 50 nm or less Rt. In some cases, the major surfaces of the tungsten-based layer may be polished to a surface roughness as low as 0.1 nm or less Ra and / or 1 nm or less Rt. Surface roughness can be measured using any conventional technique known in the art, for example, using atomic force microscopy (AFM), non-contact optical profilometry, or contact profilometry. Here, surface roughness is defined with respect to the polished surface of the layer. Typically, the polished surface is one or both of the major surfaces of the layer. Providing a layer with a surface roughness in the above range can reduce the risk of manufacturing defects during further processing.

[0019] The final thickness of the micromachined component produced according to this method can be determined in part by the thickness of the tungsten-based layer after polishing, and therefore the method can include polishing the tungsten-based layer to achieve a predetermined thickness.

[0020] The final thickness of the microfabricated component is determined by the thickness of the tungsten-based layer before reactive ion etching the photoresist and tungsten-based layer (or after polishing, if any). The thickness of this layer is not particularly limited, but the thickness before reactive ion etching the photoresist and tungsten-based layer may be 5 μm or greater to provide a freestanding component suitable for use in many microelectromechanical systems (MEMS).

[0021] The material of the carrier substrate is not particularly limited, but preferably the carrier substrate comprises a material having a coefficient of thermal expansion (CTE) within ±10% of the coefficient of thermal expansion (CTE) of the tungsten-based layer. One suitable material for the carrier substrate is glass. An alternative material is silicon. The carrier substrate is preferably transparent or translucent, which facilitates manufacturing by allowing the use of manufacturing steps that require light transmission through the carrier substrate.

[0022] The tungsten-based layer may be supported on or bonded to the carrier substrate, for example, by a suitable adhesive. Bonding the tungsten-based layer to the carrier substrate facilitates handling of the tungsten-based layer and allows for processing using a wider range of processing techniques. When using an adhesive to bond the tungsten-based layer to the carrier substrate, the adhesive preferably has a CTE within ±10% of the CTE of the tungsten-based layer. Examples of suitable adhesives include BrewerBOND® 220 and WaferBOND® HT-10.11, although any suitable bonding material can be used. If used, the adhesive should be temporary and allow for subsequent separation of the microfabricated component from the carrier substrate. If desired, the adhesive layer may consist of or include a release layer, such as a UV release layer.

[0023] By providing a carrier substrate and / or adhesive with a CTE that is within ±10% of the CTE of the tungsten-based layer, stress in the tungsten-based layer caused by heating during the reactive ion etching process can be reduced or avoided. The carrier substrate and / or adhesive preferably also has high thermal conductivity to allow heat transfer from the tungsten-based layer to the adhesive and / or carrier substrate during etching. For example, the combined substrate and adhesive layer has a thermal conductivity of 0.8-1.4 Wm at 90°C. -1 K -1 The carrier substrate and / or adhesive may have a thermal conductivity of at least 130°C to prevent degradation of these materials during high-temperature processing. This helps to reduce the maximum temperature reached in the tungsten-based layer during etching, as excessively high temperatures can have undesirable effects on the microstructure of the tungsten-based layer. The carrier substrate and / or adhesive are preferably thermally stable up to temperatures of at least 130°C to prevent degradation of these materials during high-temperature processing steps.

[0024] Conventional photolithography techniques can be used to form the photoresist mask. The photoresist can be applied on or above the surface of the tungsten-based layer. That is, the photoresist can be applied directly onto the tungsten-based layer. If the photoresist is formed directly on the tungsten-based layer, removing a portion of the photoresist can expose at least a portion of the tungsten-based layer. Alternatively, or additionally, one or more intermediate layers can be provided on the tungsten-based layer, and the photoresist can be applied directly onto such intermediate layers. In this context, "above" refers to the direction or order of layer deposition, not necessarily the layer orientation. The photoresist can be applied to cover only a portion of the tungsten-based layer. Alternatively, the photoresist can be applied to cover the entire tungsten-based layer. The photoresist can be applied at a predetermined thickness selected based on the initial thickness and etch selectivity of the tungsten-based layer. This allows the foil to be completely etched without etching through a resist mask. For example, if the tungsten base layer is 28 μm thick and has an etch selectivity of 7:1, the photoresist must be applied at a thickness of at least 4 μm to allow complete etching of the foil without etching through the resist mask.

[0025] The use of a photoresist mask (sometimes called a "soft" mask) rather than a "hard" mask (such as an aluminum mask) has the advantage of reducing the risk of redeposition of mask material during the reactive ion etching process, making it more suitable for mass-produced parts.

[0026] The photoresist may be a positive photoresist or a negative photoresist. If the photoresist is a positive photoresist, the predetermined shape of the mask substantially corresponds to the intended shape of the one or more freestanding microfabricated components. If the photoresist is a negative photoresist, the predetermined shape of the mask substantially corresponds to a negative image of the intended shape of the one or more freestanding microfabricated components. In this manner, components of appropriate shapes can be formed.

[0027] Reactive ion etching can be performed using any suitable plasma, such as SF, CF, Cl, HBr, O, H, and / or CH (References 1, 2, 3). Preferably, reactive ion etching is performed using a sulfur hexafluoride (SF) plasma. Tungsten can be etched with an SF plasma according to the following reaction (1): (1) W+SF6 → WF6+S

[0028] The by-products of the reaction are tungsten hexafluoride (WF6) and sulfur. WF6 has a low boiling point of 17.1°C. Preferably, reactive ion etching is performed at temperatures above 17.1°C. When RIE is performed at temperatures above 17.1°C, the WF6 by-product of the etch quickly becomes gaseous and can be vented from the etcher without causing problems with metal deposition on dissimilar materials or on the ceramic of the etch chamber, which can result when etching at lower temperatures.

[0029] Rhenium can be etched in an SF6 plasma according to the following reaction (2): (2) Re+SF6 → ReF6+S

[0030] Although ReF4 and ReF5 may also be formed, the primary by-product is ReF6, which has a boiling point of 33.7° C. If the tungsten-based layer is a W-Re layer, the reactive ion etching is preferably performed at a temperature above 33.7° C. This allows both the WF6 and ReF6 by-products to be quickly gasified by the etch and easily evacuated from the etcher.

[0031] Reactive ion etching can use the Bosch process (also known as pulse or time-division etching). That is, the reactive ion etching step can include substeps that alternate between etching and passivation steps. As mentioned above, sulfur hexafluoride (SF6) or any other suitable gas can be used for the etching step. The passivation step involves depositing a chemically inert passivation layer and can be performed using a gas such as octafluorocyclobutane (CF8). The etching and passivation steps are performed alternately for a predetermined number of cycles, each for a predetermined length of time. The relative duration of the etching and passivation steps affects the resulting etch profile. Generally, shorter cycle times result in smoother etched walls with reduced scalloping, while longer etching times result in higher etch rates. The inventors have found that alternating etching and passivation steps with etching durations ranging from 1 to 5 seconds and passivation durations ranging from 0.005 to 0.1 seconds can provide steep, smooth sidewalls suitable for microfabricated components. The inventors have found that alternating etching and passivation steps of about 3.5 seconds and about 0.01 seconds, respectively, provides a good balance between etching rate and reducing sidewall scalloping.

[0032] Once the RIE etching to define the micromachined components is complete, the residual photoresist can be removed from the tungsten-based layer. This can be done by any suitable process. In methods where the tungsten-based layer is bonded to a carrier substrate, the resist stripper must be selected so that it does not react with the adhesive and cause premature delamination of the micromechanical components. In one preferred method, the carrier substrate and tungsten-based layer are immersed in a resist stripper, such as 1-methyl-2-pyrrolidone (NMP), at a temperature of about 90°C for about 5 minutes before being rinsed in deionized water for about 5 minutes.

[0033] The step of separating one or more microfabricated components from the carrier substrate may include several substeps, including: adhering a free surface of one or more of said microfabricated components to a releasable tape; Separating the microfabricated component from the carrier substrate; and Peeling the microfabricated component from the releasable tape to obtain a freestanding microfabricated component.

[0034] If the tungsten-based layer is initially bonded to the carrier layer using an adhesive, separating the microfabricated component from the carrier substrate may include peeling the microfabricated component from the carrier substrate. Such peeling may be performed by any suitable process. Peeling may be laser-assisted. After peeling, residual adhesive may be removed by a cleaning process.

[0035] The inventors have discovered that attaching the microfabricated components to a releasable tape before separating them from the carrier substrate makes handling the microfabricated components easier. Furthermore, it is possible to maintain the relative alignment of each of the multiple microfabricated components. The releasable tape may be a UV-release dicing tape. Accordingly, the step of releasing the microfabricated components from the tape may include exposing the releasable tape to UV light. Examples of suitable dicing tapes include Nitto PF-02 or Lintec D-511T. The tape is preferably UV-curable so that it does not leave adhesive residue on the final product when peeled from the dicing tape.

[0036] The freestanding microfabricated component or components produced using the method of the first aspect are not particularly limited in size, shape, or purpose. The inventors have found that this method may be particularly suitable for producing microprobes suitable for use in integrated circuit (IC) testing. However, this method may also be applied to the production of microfabricated components including, but not limited to, gears used in watches and drones, microfabricated medical tools, and X-ray masks (collimators) for use in silicon processing or X-ray imaging techniques.

[0037] Thus, in a second aspect, the present invention provides a microfabricated component obtainable or obtainable by the method of the first aspect, wherein the microfabricated component has a substantially square or rectangular cross-section perpendicular to its longitudinal axis. Preferably, the microfabricated component is a microprobe.

[0038] The term "substantially square" is used herein to define that the sidewalls of the microfabricated component or microprobe are substantially vertical and the major surface of the microfabricated component or microprobe is substantially horizontal. In other words, the sidewalls of the microfabricated component or microprobe preferably extend at an angle of approximately 90° relative to the major surface of the microfabricated component or microprobe. While manufacturing tolerances may allow for some variation, the sidewalls of the microfabricated component or microprobe preferably extend at an angle between 88° and 92°, and more preferably between 89° and 91°, relative to the major surface of the microfabricated component or microprobe.

[0039] When the microfabricated component is a microprobe, the microprobe may have a width of less than 1 mm. The exact width is not particularly limited and can be appropriately selected depending on the desired commercial application. In some preferred configurations, the width of the microprobe is 25 μm or less, 10 μm or less, or 5 μm or less. The length of the microprobe is not limited other than the dimensions of the initial tungsten-based layer. However, the length of the microprobe may range from 10 μm to 10 mm. The length of the microprobe may be 50 μm or more, 100 μm or more, 500 μm or more, 1 mm or more, or 2 mm or more. For most applications, the length of the microprobe can be selected from the range of 2 mm to 10 mm.

[0040] At least one surface of the tungsten-based microprobe may have a low surface arithmetic mean roughness Ra of 5 nm or less, more preferably 1 nm or less. Preferably, at least two opposing surfaces of the tungsten-based microprobe have a surface arithmetic mean roughness Ra of 5 nm or less, more preferably 1 nm or less.

[0041] In a third aspect, the present invention provides a plurality of tungsten-based microprobes according to the second aspect, wherein the thickness of each of the plurality of microprobes varies within 1 μm of any other of the plurality of microprobes. This uniformity of thickness across the plurality of microprobes has the technical advantage that, in use (e.g., as part of a probe card), the overdrive / force required for electrical contact is approximately the same for all microprobes. This improves measurement reproducibility and extends the useful life of devices incorporating the plurality of microprobes, since no single microprobe is subjected to undue stress during use. The plurality of tungsten-based microprobes may be produced by the method according to the first aspect, in which the tungsten-based layer is provided as a foil layer. Because the foil thickness substantially determines the thickness of the resulting microfabricated part, it is possible to reduce or eliminate thickness variations across multiple parts fabricated from the same foil layer.

[0042] In a fourth aspect, the present invention provides a probe card including a plurality of tungsten-based microprobes according to the second or third aspect, wherein the pitch of the plurality of microprobes is 45 μm or less, where the pitch is defined as the distance between the centers of two adjacent microprobes in a direction perpendicular to the longitudinal axis of the microprobes. The pitch is typically equal to the total width of a single microprobe plus the width of the gap between adjacent microprobes. The width of the microprobes may be 25 μm or less, for example, 10 μm or less to about 5 μm. The width of the gap between adjacent microprobes is not particularly limited, but may be, for example, between 10 μm and 20 μm. Thus, the pitch of the plurality of tungsten-based microprobes is 45 μm or less, about 15 μm.

[0043] A probe card can be manufactured in a conventional manner using a plurality of tungsten-based microprobes according to the second or third embodiment.

[0044] The present invention includes combinations of the described aspects and preferred features except where such combinations are clearly impermissible or explicitly avoided.

[0013] Embodiments and experiments illustrating the principles of the invention will now be described with reference to the accompanying drawings. [Brief explanation of the drawings]

[0045] [Figure 1] 1 is a schematic illustration of a series of process steps in a method according to the invention; [Figure 2] 1A-1C are various scanning electron microscope images of a first example of a plurality of microprobes according to the present invention. [Figure 3] 10A-10C are various scanning electron microscope images of a second example of a plurality of microprobes according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0046] Aspects and embodiments of the present invention will now be described with reference to the accompanying drawings. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned herein are incorporated by reference.

[0047] 1 shows a schematic diagram of the sequence of process steps in one example of a method according to the present invention. It should be understood that the method is not limited to the exact steps shown, and variations can be made to the method as described in the "Summary of the Invention" above. The steps of the method are summarized as follows: a) Step 1: Providing a tungsten base layer, provided as a tungsten foil of desired dimensions; b) Step 2: Polish both sides of the tungsten foil; c) Step 3: Bonding the tungsten foil to a carrier substrate (here a glass carrier wafer) to provide a tungsten-based layer supported on the carrier substrate; d) Step 4: Apply photoresist to the tungsten base layer; e) Step 5: Exposing the photoresist through a mask having a predetermined shape, and removing a portion of the photoresist to provide a photoresist mask on or above the tungsten-based layer; f) Step 6: Perform RIE etching to define the microfabricated components; g) Step 7: Remove residual photoresist; h) Step 8: Temporarily adhering the free surface of the microfabricated component to a releasable tape; i) Step 9: Separating the microfabricated component from the carrier substrate; j) Step 10: Peel the microfabricated component from the releasable tape to obtain a free-standing microfabricated component.

[0048] An example of the method of the present invention is described in more detail below. In this exemplary method, multiple microprobes are fabricated from rolled tungsten foil. The electrical and mechanical properties of the resulting microprobes (also called probe needles) are substantially determined by the properties of the starting tungsten foil.

[0049] The inventors have found that tungsten foil produced using a rolling manufacturing process imparts high electrical conductivity and mechanical strength to the resulting microfabricated components. For manufacturing convenience, the foil is first cut into circles with diameters of, for example, 100 mm or 150 mm, that fit the subsequent processing steps. The foil thickness is selected to be close to, but greater than, the desired final thickness of the probe needle (Figure 1(a), step 1).

[0050] Due to the nature of the rolled tungsten manufacturing process, the foil used in this example is rough, with an arithmetic mean roughness (Ra) on the order of 100 nm and a maximum peak-to-valley roughness (Rt) of over 1 μm. This roughness can affect the mechanical and electrical performance of the final probe needle, so a polishing step is performed (Figure 1(b), step 2). The foil is first ground to remove most of the roughness and then polished using chemical-mechanical polishing (CMP). These process steps are applied to both sides of the tungsten foil. The required surface finish of the polished foil in this example is an Ra of 5 nm or less with an Rt of 50 nm or less. The surface of the final probe needle, with both sides finished, is given by the polished foil. The thickness of the foil after polishing is the final thickness of the probe needle.

[0051] After polishing, the tungsten foil is temporarily bonded to the glass wafer using a thermally compatible temporary adhesive (such as BrewerBOND® 220, WaferBOND® HT-10.11, or another suitable adhesive). The adhesive layer can also include a release layer for subsequent UVB laser delamination (Figure 1(c), step 3) [Reference 4].

[0052] As mentioned above, proper selection of the thermal properties of the adhesive helps ensure thermal equilibrium between the glass wafer and the tungsten foil during the etching process. Proper selection of the thermal properties of the carrier and adhesive can reduce the maximum temperature of the foil during etching (due to high thermal conductivity) and also reduce stress in the foil at high temperatures.

[0053] A UV-sensitive positive photoresist is applied to the temporarily adhered tungsten foil (FIG. 1(d), step 4). An example of a suitable photoresist is the Megaposit™ SPR220 series resist available from Dow Chemicals company, although any other suitable photoresist can be used.

[0054] The resist is coated to a thickness of 12 μm and then allowed to relax for at least an hour. This helps reduce or prevent stress-induced cracking of the resist. The resist is then soft-baked by gradually increasing the temperature from room temperature (approximately 20°C) to approximately 115°C over approximately 5 minutes to bake out the solvent. It is then held at the elevated temperature (approximately 115°C) for approximately 5 minutes. This gradual increase in temperature helps prevent stress from being introduced into the foil.

[0055] The resist is then exposed through a mask to UV radiation. In one example, the resist is then exposed to 122 mJ / cm 2 The photoresist is exposed through a mask with UV radiation at ±10% energy, which is the product of the lamp power and exposure time. The energy used determines the resolution, adhesion, and depth of exposure. The exact energy used to expose the photoresist can be appropriately selected based on the resist used, its thickness, and the desired resolution of the mask.

[0056] After exposure, the wafer is left for at least 45 minutes. This allows moisture to be reabsorbed from the environment, helping to complete the polymer cross-linking within the resist. Next, a post-exposure bake is performed to complete the resist cross-linking process. This bake step is performed by increasing the temperature from room temperature (approximately 20°C) to approximately 115°C over approximately 5 minutes, and then holding the temperature at approximately 115°C for 90 seconds. The resist is then developed in a chemically reactive solution (such as Megaposit™ MF-26A developer) to remove any exposed photoresist remaining behind the photopolymer resist mask (Figure 1(e), step 5).

[0057] The tungsten foil is then etched using a PlasmaTherm DSE multiplex inductively coupled plasma machine (although other etching equipment may be used). This equipment is selected for this example due to its fast gas switching, fast and stable pressure control, and solid-state RF tuning capabilities. A coil around the etching chamber is used to generate the plasma, and a platen coil is connected to the wafer electrode and controls the RF bias potential of the wafer relative to the plasma. Backside helium pressure is used to reach the electrode through the temporary bonding material and carrier wafer, providing heat transfer between the tungsten foil and maintaining a constant wafer temperature. Once the wafer is placed in the machine, it is secured to the electrode by a ceramic ring. Alternating etching and passivation steps of 3.5 seconds and 0.01 seconds are used. These fast switching steps avoid sidewall damage due to overetching, which can lead to scalloping, and allow for the formation of substantially vertical sidewalls during the etching process.

[0058] The gas used for the etching process is SF6, and the gas used for passivation is C4F8. The flow rates of SF6 and C4F8 are 350 standard cubic centimeters per minute (sccm) and 100 sccm, respectively. For the deposition (passivation stage), the pressure in the chamber is 35 mTorr, the platen power is 100 W, and the coil power is 2000 W. For the etching stage, the pressure in the chamber is 150 mTorr, the platen power is 150 W, and the coil power is 3000 W.

[0059] RIE has an etch rate of approximately 750 nm / min for the tungsten foil layer using these process parameters. Therefore, the required etch time is about 34 min for a 25 μm tungsten foil thickness. In this example, the choice of material and etching process results in an etch selectivity ratio of tungsten to resist of about 7:1 (Figure 1(f), step 6).

[0060] After etching the wafer, the microstructures on the wafer were inspected using an optical microscope, and the step height was measured using a profilometer. The photoresist mask of the tungsten microstructures was subsequently removed by immersing the wafer in a resist stripper such as NMP at a temperature of 90 °C and rinsing with deionized water for 5 min each. NMP was chosen as the stripper because it does not react with the temporary adhesive used to attach the tungsten foil to the glass carrier wafer (Figure 1(g), step 7).

[0061] The etched microstructure is subjected to a series of electrical tests to extract the foil's properties before peeling. In one example, a four-point probe resistance measurement was performed on the etched foil to calculate the electrical resistivity of the tungsten-based layer. The sheet resistance was found to be approximately 5.4 μΩ·cm.

[0062] Finally, the microprobe is separated from the carrier substrate. To achieve this, the free surface of the microprobe is adhered to UV-release dicing tape (Figure 1(h) Step 8) before separating the microprobe from the glass carrier wafer (Figure 1(i) Step 9) [Reference 5].

[0063] In an example where the adhesive layer bonding the tungsten-based layer to the carrier substrate includes a release layer for subsequent UV laser debonding, UV light is shone through the clear glass carrier substrate and absorbed by the adhesive release layer. The absorbed UV laser energy breaks the chemical bonds in the adhesive or release layer without generating substantial heat, thereby removing the microfabricated component from the carrier substrate with minimal or no thermal stress. The UV light does not penetrate with sufficient strength to affect the adhesion of the microfabricated component to the UV release dicing tape. After laser debonding of the microfabricated component and carrier substrate, the microfabricated component and carrier substrate can be separated with near-zero mechanical force applied to the microfabricated component.

[0064] The UV-release dicing tape should be UV-curable and have a sufficiently high adhesive strength to allow solvent cleaning of the microfabricated components attached to the dicing tape after separation from the carrier substrate. Examples of suitable dicing tapes are Nitto PF-02 or Lintec D-511T, which have adhesive strengths of 1840 mN / 22 mm and 11800 mN / 25 mm, respectively.

[0065] The microprobe is then cleaned while still attached to the dicing tape to remove any residual bonding adhesive with either a solvent or plasma, or a combination thereof. Suitable solvents include mesitylene with IPA or mesitylene with methyl ethyl ketone.

[0066] Finally, the probes are peeled from the UV-release dicing tape using ultraviolet (UV) light, removed from the tape, and packaged appropriately. This step can be performed with a pick-and-place tool if necessary (Figure 1(j), step 10).

[0067] Figures 2 and 3 show various scanning electron microscope images of microprobes made in accordance with the present invention. All microprobes shown have the same thickness of 22 μm.

[0068] Figure 2 shows 25 μm wide, 2 mm long probes spaced 50 μm apart. The microprobe has a wide area on its surface that is 40 μm wide and 200 μm long.

[0069] Figure 3 shows a 25 μm wide, 4 mm long probe with a 225 μm pitch between parallel sections. The probe has two bends, the first bend shown in Figure 3(a) and a zoom of the other bend shown in Figure 3(b). Figure 3(b) shows that the sidewalls of the probe are substantially vertical and substantially smooth (minimal or no scalloping is observed).

[0070] The features described above, or the following claims, the accompanying drawings, expressed in a particular form or in terms of means for performing a disclosed function, or methods or processes for obtaining the disclosed results, as appropriate, may be used individually or in any combination of such features to realize the invention in diverse forms thereof.

[0071] While the present invention has been described in conjunction with the above exemplary embodiments, many equivalent modifications and variations will become apparent to those skilled in the art given this disclosure. Accordingly, the above exemplary embodiments of the present invention are considered to be illustrative and not limiting, and various changes to the described embodiments can be made without departing from the spirit and scope of the present invention. For the avoidance of doubt, the theoretical explanations set forth herein are provided for the purpose of enhancing the reader's understanding. The inventors do not wish to be bound by any of these theoretical explanations. The section headings used herein are for organizational purposes only and should not be construed as limiting the subject matter described.

[0072] Throughout this specification, including the claims that follow, unless the context requires otherwise, the use of terms such as "comprises," "has," "including," "having," and variations thereof implies the inclusion of a stated "integer" or "step" or group of "integers" or "steps," but does not exclude other "integers" or "steps" or groups of "integers" or "steps."

[0073] As used in the specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" one particular value and / or to "about" another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values ​​are expressed as approximations, by use of the prefix "about," it will be understood that the particular value forms another embodiment. The term "about" in reference to numerical values ​​is arbitrary and means, for example, ±10%.

[0074] (References) A number of publications have been cited above in order to more fully describe and disclose the present invention and the state of the art to which it pertains. Full citations for these references are set forth below. Each of these references is incorporated herein in its entirety. 1.Ma,Shenglin&Xia,Yanming&Wang,Yaohua&en,Kuili&Luo,Rongfeng&Song,Lu&Chen,Xian&Chen,Jing&Jin,Yufeng.(2016).“Fabrication and characterization of a tungsten microneedle array based on deep reactive ion etching technology”.Journal of Vacuum Science&Technology B,Nanotechnology and Microelectronics:Materials,Processing,Measurement, and Phenomena.34.052002.10.1116 / 1.4960715. 2.“Study of the conditions for aniostropic plasma etching of tungsten and tungsten nitride using Sf6 / Ar gas mixtures”.C.Reyes-Betanzo et al.Journal of the electrochemical society,149(3)G179-G183(2002) 3.“Microfabrication of thick tungsten films for use as absorbers of deep x-ray lithography masks”.H.Okuyama et al.Microsystems technologies7(2001)80-84 4.“High Temperature-Resistant Spin-On Adhesive for Temporary Wafer Mounting Using an Automated High-Throughput Tooling Solution”A.Smith et al.,CS MANTECH Conference,May14-17,2007,Austin,Texas,USA 5.United States Patent US9,029,238B2“Advanced Handler wafer bonding and debonding”

Claims

1. providing a tungsten-based layer supported on a carrier substrate; applying a photoresist, exposing the photoresist through a mask having a predetermined shape, and removing portions of the photoresist to provide a photoresist mask on or above the tungsten-based layer; reactive ion etching the photoresist and the tungsten-based layer to define one or more micro-fabricated features; and Separating one or more of the microfabricated components from the carrier substrate.

1. A method for manufacturing one or more freestanding microfabricated components, comprising:

2. 10. The method of claim 1, wherein the tungsten-based layer comprises tungsten at an atomic percent of 50 at% or greater.

3. The method of claim 1 or 2, wherein the tungsten-based layer is a W—Re alloy.

4. The method according to any one of claims 1 to 3, wherein the tungsten-based layer is provided as a foil layer.

5. The method of any one of claims 1 to 4, further comprising polishing the tungsten-based layer.

6. 6. The method of claim 5, wherein one or both surfaces of the tungsten-based layer are polished to an average surface roughness Ra of 5 nm or less and / or Rt of 50 nm or less.

7. 7. The method of claim 1, wherein the tungsten-based layer has a thickness of 10 μm or more prior to the step of reactive ion etching the photoresist and the tungsten-based layer.

8. The method of any one of claims 1 to 7, wherein the carrier substrate comprises a material having a coefficient of thermal expansion CTE that is within ±10% of the coefficient of thermal expansion CTE of the tungsten-based layer.

9. 9. The method of claim 1, wherein the tungsten-based layer is bonded to a carrier substrate using an adhesive having a coefficient of thermal expansion (CTE) within ±10% of the coefficient of thermal expansion (CTE) of the tungsten-based layer.

10. (i) the photoresist is a positive photoresist and the predetermined shape of the mask substantially corresponds to the intended shape of one or more freestanding microfabricated components; or (ii) the photoresist is a negative photoresist, and the predetermined shape of the mask substantially corresponds to a negative image of the intended shape of one or more freestanding microfabricated components; The method according to any one of claims 1 to 9.

11. The reactive ion etching is performed using sulfur hexafluoride (SF) 6 The method according to any one of claims 1 to 10, which is carried out using plasma.

12. The method according to any one of claims 1 to 11, wherein the reactive ion etching uses the Bosch process with alternating etching and passivation steps of 3.5 seconds and 0.01 seconds, respectively.

13. The method of any one of claims 1 to 12, wherein the reactive ion etching is performed at a temperature above 17.1°C.

14. The method according to any one of claims 1 to 13, wherein the step of separating one or more microfabricated components from the carrier substrate comprises the following substeps: adhering a free surface of one or more of said microfabricated components to a releasable tape; Separating the microfabricated component from the carrier substrate; and Peeling the microfabricated component from the releasable tape to obtain a freestanding microfabricated component.

15. 15. The method of claim 14, wherein the releasable tape is a UV release dicing tape, and peeling the micro-fabricated components from the releasable tape includes exposing the tape to UV light.

16. The method of any one of claims 1 to 15, wherein the one or more freestanding microfabricated components are microprobes suitable for use in integrated circuit (IC) testing.

17. A tungsten-based microprobe obtainable by the method according to any one of claims 1 to 16, comprising: The microprobe is a tungsten-based microprobe having a substantially square or rectangular cross section perpendicular to the longitudinal axis of the microprobe.

18. 20. The tungsten-based microprobe of claim 17, wherein at least one surface of the microprobe has an arithmetic mean surface roughness Ra of 5 nm or less.

19. A tungsten-based microprobe according to claim 17 or 18, wherein the width of the microprobe is between 5 μm and 25 μm.

20. 20. A tungsten-based microprobe according to any one of claims 17 to 19, comprising: A tungsten-based microprobe, wherein the thickness of each of the plurality of microprobes varies within 1 μm compared to any other of the plurality of microprobes.

21. A probe card comprising a plurality of tungsten-based microprobes according to any one of claims 17 to 20, A probe card, wherein the pitch of the plurality of microprobes is 45 μm or less, the pitch being defined as the distance between the centers of two adjacent microprobes in a direction perpendicular to the longitudinal axes of the microprobes.

Images