Various exemplary details are described with reference to the following figures, wherein:
The systems and methods described herein may be particularly applicable to microelectromechanical (MEMS) devices, wherein the vias may be required to carry a relatively large amount of current. MEMS devices are often fabricated on a composite silicon-on-insulator (SOI) wafer, consisting of a relatively thick (about 400-725 μm) “handle” layer of silicon overcoated with a thin (about 1-4 μm) layer of buried silicon dioxide, and covered with a silicon “device” layer about 5-100 μm thick. In the embodiment described below, the handle layer is about 500 μm thick, the oxide layer is about 2 μm thick, and the device layer is about 80 μm thick, and the through wafer vias may be formed in this device layer. However, it should be understood that this embodiment is exemplary only, and that the through wafer vias may be formed in simple a Si substrate, or in the device layer or handle layer of an SOI substrate, for example, before the formation of the MEMS device on top of the substrate and vias. The MEMS device may then be made, for example, by additive and subtractive thin-film processing. Movable features may be freed by, for example, wet or dry etching a selectively etchable, intermediate sacrificial layer from beneath the moveable feature. The moveable features may then be hermetically encapsulated in a cap or lid wafer, which is bonded or otherwise adhered to the top of the via/device wafer, to protect the moveable features from damage from handling and/or to seal a particular gas in the device as a preferred environment for operation of the MEMS device. After formation and encapsulation of the MEMS device, the back side of the via substrate may be ground to remove the dead end walls of the blind holes, and form the through wafer vias. Each of these processes is described in further detail below.
Through-hole vias are particularly convenient for encapsulated MEMS devices, because they may allow electrical access to the encapsulated devices. Without such through holes, electrical access to the MEMS device may have to be gained by electrical leads routed laterally under the lid wafer which is hermetically sealed over the MEMS device. It may be problematic, however, to achieve a hermetic seal over terrain that includes the electrical leads unless more complex and expensive processing steps are employed. This approach also makes radio-frequency applications of the device limited, as electromagnetic coupling will occur from the metallic bondline disposed over the laterally oriented leads. Alternatively, the electrical access may be achieved with vertically oriented through-wafer vias formed in a simple silicon substrate for example, or through the handle layer or device layer of an SOI wafer, using the systems and methods described here.
As mentioned above, the through hole vias may be constructed by first forming a blind trench in the substrate, and then forming a plating seed layer in the blind trench. It should be understood that although the word “trench” is used, the term should be construed as including any shape of opening, including a circular hole. A “blind trench” or “blind hole” may be an opening formed in one side of a substrate, which does not extend through the whole thickness of the substrate, such that the trench ends with a dead end wall in the substrate material. A “through hole” or “through wafer via” should be construed to mean an electrical conduit which extends completely through a material, for example, through a layer, wafer or substrate.
Gold is chosen for the conductive plating base layer despite its relatively high cost, because it appears to have outstanding throwing power, or ability to be deposited into deeply recessed features. Gold appears to surpass the commonly used metals, such as copper (Cu), tungsten (W) or nickel (Ni) in this regard, as a plating base layer. In addition, because the deposition is performed at relatively low temperature, other previously deposited or defined features may exist on the wafer prior to the deposition of the gold plating base. Another advantage of using Au is that it does not reverse plate or etch during the Cu reverse pulse plating process which follows. Relatively high reverse plating currents can be applied, while maintaining the electrical integrity of the plating base.
The plating of the conductive species may be accomplished using reverse pulse plating. Generally, the method comprises plating the conductive species on all the conductive surfaces of the blind holes, and then de-plating a portion of the conductive species on the upper portions of the blind holes, near the top corners. The de-plating process generally requires performing the electroplating process with the electrical bias polarities reversed between the cathode and the anode terminals of the electroplating system. By adjusting the ratio of the charge driven between the terminals in the reverse direction relative to the charge driven between the terminals in the forward direction, the ratio of the amount of conductive species deposited on the upper sidewalls relative to the amount at the base of the blind hole, can be adjusted.
In various exemplary embodiments, the ratio of the forward charge to the reverse charge is about two to one, and more generally less than about ten to one. For example, a forward current of about 0.4 A may be applied to the terminals for about 100 msec, for a total forward charge of about 40 mCoulombs. A reverse current of about 4.0 A is then applied to the terminals, with the bias polarity reversed, for about 5 msec, for a total reverse charge of about −20 mCoulombs. More generally, the forward current may be between about 0.1 A to about 1 A, for about 10 msec to about 200 msec. A reverse current of about −1 A to about −10 A may then be applied to the terminals for between about 1 msec to about 10 msec. The design considerations involved in choosing the amount of forward charge relative to the amount of reverse charge may be the plating time required to plate up the entire via blind hole, balanced against the tendency of the via to close at the top before the via is fully plated, thereby forming a void. Using the recommended values set forth above, a plating time of about eight hours may be required. Decreasing the amount of forward charge relative to the amount of reverse charge, while assuring that the voids are not formed within the via, may increase the required plating time.
A pause between the reverse current pulse and the next positive current pulse may be used to allow the plating bath to circulate and the plating species concentrations to equilibrate. A pause of between about 1 msec and about 200 msec may be inserted between the reverse current pulse and the forward current pulse. More preferably, a pause of about 10 msec between each reverse current pulse and each following forward current pulse may be sufficient. A suitable exemplary pulse train for the combination forward and reverse pulse plating process is shown in
Combination forward and reverse pulse plating processes, such as that depicted in
The plated species may be copper (Cu), for example, plated by immersing the substrate 100 in a plating solution containing copper sulfate and sulfuric acid and performing the reverse pulse plating process. However, it should be understood that this embodiment is exemplary only, and that any other suitably conductive material which can be plated on the substrate, including gold (Au) or nickel (Ni), may be used in place of copper.
The via substrate 100 may, at this point, be bonded to another substrate, which has a previously fabricated MEMS device on its surface. The substrate may be either Si, or an SOI substrate, for example. The bonding mechanism may be, for example, eutectic, glass flit, polymer, or another other low-temperature bonding method, typically less than 300 degrees centigrade. The MEMS device may also be made first on top of the via substrate 100, then bonded to a lid wafer, also with a relatively low temperature process, such as eutectic bonding or glass frit bonding. Because there are metals on the via substrate, high temperature bonding may not be used. Additional details describing the formation of the MEMS device are set forth below, in reference to
Finally, the through wafer vias may need to be formed from the blind trenches in the via substrate 100, by removing material from the exposed backside of the via substrate 100 up to the dead-end walls of the blind trenches. The through wafer vias may be formed by, for example, isotropic dry etching, single-sided wet etching or grinding, lapping, and polishing the back side 190 of the substrate 100, to remove material from the back side to a point 170 at which the dead end walls have been removed. In one embodiment, the means for removing material from the back side 190 of the substrate 100 may be a precision wafer grinder, such as a model VG-401 available from Okamoto of Japan. The grinder may use a metal wheel with diamond grit embedded in it as an abrasive. The rotation rate of the grinding wheel may be about 800 μm, and the rotation rate of the table holding the substrate 100 may be about 80 μm. Using these parameters, the grinding tool may be programmed to remove material at a rate of about 25 Mm per minute for about 15 minutes, to remove about 400 μm to about 450 μm of material, leaving the through wafer via substrate 100 having a thickness of about 50 μm to about 100 μm. At this point, the blind trenches 122 and 124 may become the through wafer vias 222 and 224.
Other techniques for removing material may be used, such as dry or single-sided wet etching, either alone or in combination with grinding, to remove about 400 μm of silicon from a 500 μm thick substrate, leaving about 100 μm of material as substrate 100. The etching can be done either before, but typically after the via substrate 100 is bonded to a device substrate. Accordingly, using the methods described here, through wafer vias of diameter less than about 50 μm and depths of at least about 100 μm may be made. More particularly, the aspect ratio of the via, that is, the ratio of the depth of the via to its width, may be at least one-to-one, and as great as about eight-to-one.
Alternatively, instead of removing material from the back side of the substrate, the through wafer vias 222 and 224 may be made using a silicon-on-insulator (SOD) composite substrate. The blind trenches may be etched as described above through a thick, 50 μm-100 μm device wafer, and coated with the seed layer structure and plated as before. However, using the silicon-on-insulator wafer, the handle wafer may be dry or wet etched, using the buried oxide as an etch stop. Vias may then be patterned in the now exposed, but previously buried oxide to ultimately allow a conductive path from subsequently defined metal pads and the through wafer vias. The buried oxide, being left on the majority of the substrate, also serves the purpose of electrical isolation between the top metallization layer and the substrate. In another alternative, the blind holes may be formed in the SOI handle layer, and the SOI device layer subsequently removed to form the through wafer vias.
Alternatively, the MEMS device 300 may be made by forming moveable features in the device layer of another SOI wafer by, for example, deep reactive ion etching (DRIE) with the oxide layer forming a convenient etch stop. The movable feature is then freed by, for example, wet etching the oxide layer from beneath the moveable feature. The device layer may then be bonded face to face with a via substrate. The inner surface of the via substrate may be an integral part of the MEMS device, for example, switch contacts may be placed directly over the vias. Additional details as to the method of manufacture of such a cantilevered MEMS switch may be found in U.S. patent application Ser. No. 11/211,623 (Attorney Docket No. IMT-Wallis), U.S. patent application Ser. No. 11/211,624 (Attorney Docket No. IMT-Blind Trench) and U.S. patent application Ser. No. 11/359,558 (Attorney Docket No. IMT-SOI Release). Additionally, the through wafer via may be part of the lid wafer, carrying signals to the MEMS device through a connection in the bond line, while not actually being an active part of the device.
However, it should be understood that the MEMS device 300 may be any of a number of devices other than the switches described in the incorporated '912 application, '623 application, '558 application, or '624 application, such as accelerometers, sensors, actuators, and the like. Since the details of the MEMS device 300 are not necessary to the understanding of the systems and methods described here, it is depicted only schematically in
The through wafer vias 222 and 224 may be made using the reverse pulse plating process described above, to form a low-cost, highly conductive via with excellent thermal conductivity. However, the ability of the plated via to form a hermetic seal from one side of the wafer to the other may be limited by such factors as grain boundaries, and the propensity of the plated metal to crack and delaminate from the surrounding substrate, especially at elevated temperatures. A more hermetic seal may be made by providing thin pads 312 and 314 over the through wafer vias 222 and 224, as shown in
The pads 312 and 314 may be formed before the MEMS device 300, by first depositing an adhesion layer such as chromium (Cr), followed by a layer of Au. A barrier layer, for example, molybdenum (Mo) may also be used to prevent the chromium of the adhesion layer from diffusing into the gold of the pad. The thickness of the pads 312 and 314 may be sufficient to provide a barrier to the transmission of gasses through the pad and therefore through the via, and therefore may increase the hermeticity of the encapsulated MEMS device 300. A thickness of between about 2500 Angstroms and about 1 μm of gold may be sufficient to provide this barrier. The Cr adhesion layer may be between about 50 Angstroms and about 1500 Angstroms thick, and the optional Mo layer may be about 100 Angstroms thick. Also, pads may be placed on the outer surface of the substrate, as probe pads 322 and 324, which in addition to providing a hermetic seal, allow the encapsulated device to be probed electrically from outside the encapsulation.
The gold pads 312 and 314 may be formed so that the edges extend slightly beyond the vias, about 5 μm beyond is typically sufficient to allow a misalignment tolerance and a good seal. The gold pad 312 and 314 can be formed using a lift-off process, or deposited, patterned and etched using dry or wet processes.
In addition to providing a barrier to the transmission of gasses, the pads 312 and 314 may also serve to keep the copper vias 222 and 224 from oxidizing during processing. Pad 314 may also be used as a switch contact as shown in
After formation of the gold pads 312 and 314, and formation of the MEMS device 300, the MEMS cantilevered device 300 may be encapsulated in a cap or lid wafer 500, which has been relieved in areas to provide clearance for the movement of MEMS device 300. A hermetic seal may be made using any suitable adhesive 400, which may be applied to the bonding surfaces of the lid wafer 500. For example, the hermetic seal may be an alloy seal as taught in greater detail in U.S. patent application Ser. No. 11/211,625 (Attorney Docket No. IMT-Interconnect) and U.S. patent application Ser. No. 11/211,622 (Attorney Docket No. IMT-Preform) incorporated by reference herein in their entireties. The alloy seal may be an alloy of gold (Au) layers and indium (In) layer, in the stoichiometry of AuIn2. Alternatively, the hermetic seal may be formed using a glass frit with embedded particles as a standoff, as taught in U.S. patent application Ser. No. 11/390,085 (Attorney Docket No. IMW-Standoff), incorporated by reference herein in its entirety.
Electrical contact to the encapsulated MEMS cantilevered device 1000 may be obtained with the through wafer vias 222 and 224. Contacts may be made by depositing a layer of a conductive material 322 and 324, onto the back side of substrate 100. The conductive material 322 and 324 may be, for example, gold pads about 0.5 μm or greater in thickness. As previously mentioned, these contacts 322 and 324 may serve as probe pads for testing the functionality of the encapsulated device 1000.
In step S800, material may be removed from the back side of the substrate to remove the dead end wall of the blind trench to form the through wafer vias. In various exemplary embodiments, the back side of the substrate may be ground or etched to remove the dead end wall of the blind trench.
The process may continue in step S900, wherein the contact probe pads may be deposited over the through wafer vias on the back side of the substrate. In step S1000, the devices may be singulated from the device wafer by, for example, saw cutting. The process ends in step S1100.
It should be understood that not all of the steps of the method illustrated in
While various details have been described in conjunction with the exemplary implementations outlined above, various alternatives, modifications, variations, improvements, and/or substantial equivalents, whether known or that are or may be presently unforeseen, may become apparent upon reviewing the foregoing disclosure. For example, while the disclosure describes an embodiment including a MEMS switch, it should be understood that this embodiment is exemplary only, and that the systems and methods disclosed here may be applied to any number of alternative MEMS or non-MEMS devices. Accordingly, the exemplary implementations set forth above, are intended to be illustrative, not limiting.