Methods for Etching Fuse Openings in a Semiconductor Device

1. The Field of the Invention

The present invention relates to the manufacture of semiconductor devices. More particularly, the present invention is directed to improved methods for etching fuse openings in the manufacture of semiconductor devices.

2. The Relevant Technology

In order to improve yield in the manufacture of semiconductor devices, redundant circuit elements may be provided in a circuit layout. The redundant elements may be selectively connected to or disconnected from the circuit as needed to replace defective circuit elements by selectively blowing fuses in the circuit. In highly dense memory circuits, for example, spare rows and columns are formed during fabrication. If a defective bit is found during testing, a spare row or column is substituted for the defective bit by selectively blowing fuses included in the circuit for that purpose.

In state of the art memory circuit layouts, laser fuses take the form of sections of gate stacks including a polysilicon conductive layer. The stack is enclosed laterally by dielectric spacers and upwardly by a dielectric cap. A fuse is "blown" by irradiating the dielectric cap from above with laser radiation at a selected location along the gate stack. The polysilicon is heated by the laser radiation and expands, popping off the dielectric cap at the selected location, and the polysilicon is vaporized or burned off, creating a break in the conductive polysilicon layer.

Fabrication of reliable gate stack type fuses is complicated by the thickness of the overlying layers. As much as 30,000 to 40,000 Angstroms or more of overlying layers must be removed to expose the dielectric cap of the gate stack so that laser radiation may be used to blow a fuse. The overlying layers must be removed to within approximately 3000 Angstroms or less of the top of the cap in order for laser irradiation to reliably blow a selected fuse. But at etch depths as great as 30,000 or 40,000 Angstroms, variations in etch rate over the surface of a wafer, together with variations over the surface of the wafer in the layers to be etched resulting from previous process steps, can result in a difference as large as 6500 Angstroms or more between the deepest and shallowest effective depth of the etch. Etching too deep may destroy the gate stack or expose the substrate, allowing contamination and shorting. Etching not deep enough results in fuses that cannot be reliably blown.

Etching fuse openings thus typically requires painstaking control, such as a timed etch followed by an etch depth measurement for every wafer, followed by a second timed etch for a time calculated individually for each wafer. This type of control is cumbersome and time consuming. Even with control of this type, achieving an etch depth within process limits across the entire surface of a wafer is not always possible. Hence an improved method of etching fuse openings in a semiconductor device is needed.

An object of the present invention is to provide a reliable, easily performed method of etching fuse openings in a semiconductor device to etch depths within process limits.

Another object of the present invention is to provide a method for etching fuse openings in a semiconductor device providing increased uniformity of etch depth across the surface of a wafer.

Yet another object of the present invention is to provide a method for etching fuse openings in a semiconductor device providing increased uniformity of etch depth across the surface of a wafer, which method is easily integrated into standard process flows.

Still another object of the present invention is to provide a method for etching fuse openings in a semiconductor device providing increased uniformity of etch depth across the surface of a wafers which method also provides an increase in the process limits.

Still another object of the present invention is to increase yield in the manufacture of a circuit device by providing an improved method of etching fuse openings, which method increased the number of fuses per wafer which can successfully be blown.

In accordance with one embodiment of the present invention, a silicon nitride cap and silicon nitride spacers are employed on a gate stack serving as a fuse. The etch of the fuse opening is then performed by first etching with an etch process that etches both silicon nitride and silicon oxide, then later etching with an etch process that is selective to silicon nitride. The later etch process etches the silicon nitride of the cap and spacers on the gate stack little or not at all, allowing a wider variation in etch depths without destroying the gate stack. Thus the process limits are increased, and the etch of the fuse opening is more easily kept within those limits.

In accordance with another embodiment of the present invention, a silicon nitride cap and silicon nitride spacers are employed on the gate stack serving as a fuse, and a patch is provided in the overlying layers above the gate stack serving as a fuse. The steps for etching the fuse opening are then so arranged that, at the time the patch is reached, an etch process is used that is selective to a material of the patch, resulting in an etch stop effect at the level of the patch. The etch process is then changed to an etch process that is not selective to the patch. The etch of the fuse opening is then completed with an etch process that is selective to silicon nitride so as to stop at the cap and spacers on the gate stack. In addition to allowing wider variation in etch depths by protecting the gate stack, this embodiment and variations thereof decrease the variation in the etch depth over the surface of a wafer.

Both embodiments summarized above increase the reliability of the fuse opening etch, providing increased yield and tighter, more easily maintained process control.

These and other objects and features of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth hereinafter.

In order that the manner in which the above-recited and other advantages and objects of the invention are obtained may be more fully explained, a more particular description of the invention briefly described above will be rendered by reference to specific applications thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical applications of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:

FIG. 1 is a cross section of a partially completed semiconductor device prepared for etching of a fuse opening according to the methods of the present invention.

FIG. 2 is a cross section of the structure shown in FIG. 1 after etching of a fuse opening according to the methods of the present invention.

FIG. 3 is another cross section of a partially completed semiconductor device prepared for etching of a fuse opening according to the methods of the present invention.

FIG. 4 is a cross section of the structure of FIG. 3 after partial etching of a fuse opening according to the methods of the present invention.

FIG. 5 is a cross section of the structure of FIG. 4 after completion of etching of a fuse opening according to the methods of the present invention.

FIG. 6 is yet another cross section of a partially completed semiconductor device prepared for etching of a fuse opening according to the methods of the present invention.

As an overview, the present invention provides methods of etching fuse openings during the manufacture of a semiconductor device, which methods improve the reliability of the fuse opening etch. Each of the methods includes providing a gate stack serving as a fuse with spacers and a cap comprised of silicon nitride, then altering the etch process at some point prior to completion of the etch from an etch process that etches silicon nitride and silicon oxide to an etch process that is selective to silicon nitride. Some methods of the present invention further include providing a patch in the overlying layers above the gate stack serving as a fuse, and then etching, at the time during the etch at which the patch is reached, with an etch process selective to the material of the patch, and, after reaching the patch, switching to an etch process that is not selective to the material of the patch. Various currently preferred embodiments of such methods are described in more detail below.

FIG. 1 shows a cross section of a partially completed semiconductor device prepared for etching of a fuse opening according to one embodiment of the present invention. A gate stack 14 comprising a polysilicon gate 16, a metal silicide layer 18, a thin oxide layer 20 and a silicon nitride cap 22 has been formed on a substrate 12. Gate stack 14 is enclosed by silicon nitride spacers 24. Overlying layers have been deposited over and around gate stack 14. The overlying layers include various silicon oxide layers 26, 28', 30, and 32, which may include such materials as grown silicon oxide, TEOS, BPSG, etc., and silicon oxide and silicon nitride passivation layer 34. A layer of polysilicon which comprises polysilicon gate 16 is conductive and extends, along with the rest of gate stack 14, for a length along a direction perpendicular to the plane of FIG. 1.

Gate stack 14 is intended to act as a fuse. But a fuse opening must be etched in the overlying layers above gate stack 14 before laser irradiation can be used to "blow" the fuse. The total thickness of the layers above nitride cap 22 of gate stack 14 can exceed 30,000 or even 40,000 Angstroms, resulting in significant variations in total etch depth during etching of the fuse opening. Even with lower total thickness of the layers above nitride cap 22, process variations can result in significant variations in total etch depth. If the etch goes too deep, nitride cap 22 and thin oxide layer 20 may be etched away, leaving the conductive metal silicide and polysilicon layers exposed to potential contamination and electrical shorts, and making the fuse difficult to reliably blow. Nitride spacers 24 and silicon oxide layer 26 may even be completely removed at the bottom of an over-etched fuse opening, exposing substrate 12 to possible contamination and electrical shorts. If the etch is not deep enough, the layers overlying nitride cap 22 will prevent nitride cap 22 from popping off, and the polysilicon, even if melted by the laser energy, will remain in place and will remain conductive after cooling, so that the fuse cannot be successfully blown.

According to the embodiment of the present invention employed with the structure shown in FIG. 1, a fuse opening is etched in the overlying layers above gate stack 14 using at least a two-step etch process. Using standard masking technology, a mask is provided on silicon nitride passivation layer 34 to shield areas not to be etched. An opening in the mask is formed over gate stack 14. Openings in the mask will also typically be formed over bonding pad areas, and so that fuse openings in the overlying layers above gate stack 14 and bonding pad openings in silicon nitride and silicon oxide passivation layers 34, 32 may be etched simultaneously. If the bonding pad openings are indeed etched simultaneously with the fuse openings, the etch processes chosen for use must be selective to the material uppermost within the bonding pads, typically aluminum.

The first step in the at least two-step etch process is etching with an etch process that etches silicon nitride, and preferably etches both silicon nitride and silicon oxide. This step must at least etch through silicon nitride passivation layer 34, and is preferred to etch through silicon nitride passivation layer 34, oxide passivation layer 32, silicon oxide layer 30, and into, but not through, silicon oxide layer 28. A last step in the at least two-step etch process is etching with an etch process that is selective to silicon nitride, i.e., an etch process that etches silicon nitride relatively little or not at all. This last step is preferably accomplished by using a CH.sub.2 F.sub.2 /CF.sub.4 /CHF.sub.3 /argon chemistry as described in U.S. Pat. No. 5,286,344 entitled "Process of Selectively Etching a Layer of Silicon Dioxide on an Underlying Layer of Silicon Nitride". This last step is applied long enough to extend the opening begun by the first step downward to, at most, the level of a bottom 38 of a fuse opening 36 shown in FIG. 2. At the lowest desirable level as shown in FIG. 2 for bottom 38 of fuse opening 36, at least 2000 Angstroms of silicon oxide layer 26 remain above substrate 12. At the highest desirable level for the bottom of the fuse opening, not more than about 3000 Angstroms of silicon oxide layers 26 and 28 should remain above nitride cap 22. This corresponds to a fuse opening bottom located at dotted line A in FIG. 2.

Without the last step of etching with an etch process selective to silicon nitride, nitride cap 22, thin oxide layer 20, and nitride spacers 24 would begin to be etched away as downward progress of the fuse opening continued below nitride cap 22. Including the last step of etching with an etch process selective to silicon nitride allows the bottom of the fuse opening to extend below nitride cap 22 without damaging gate stack 14. The embodiment of the present invention employed with the structure of FIGS. 1 and 2 thus results in an increase in distance D, the vertical distance between the lowest and highest desirable levels for the bottom of the fuse opening. At some areas of a wafer, the completed fuse etch may stop as high as the level of dotted line A in FIG. 2, while at other areas of the same wafer, the completed fuse etch may reach the depth shown in FIG. 2, where only a relatively small thickness of silicon oxide layer 26 remains at the bottom 38 of fuse opening 36. Thus, the acceptable limits of variation of the fuse opening depth are increased, resulting in easier and more reliable process control.

The first and last steps of the embodiment of the present invention employed with the structure of FIGS. 1 and 2 may be performed in separate pieces of etch equipment, or may be performed in a single piece of etch equipment, or may preferably be performed in situ in a single process chamber by merely varying the etch chemistry appropriately.

The currently preferred method for performing the first step of etching with an etch process that etches silicon nitride includes etching in an Applied Materials 5000, a single wafer plasma etching apparatus manufactured by Applied Materials of Santa Clara, Calif. The currently preferred process parameters include etching at a pressure of 150 mTorr, a power level of 700W, a magnetic field strength of 75 Gauss, with gas flows of 25 sccm CF.sub.4, 35 sccm CHF.sub.3, and 60 sccm Argon, and with helium backside cooling at a pressure of 7000 mTorr. This etch recipe etches both silicon nitride and silicon oxide, and is preferably used to etch through silicon nitride passivation layer 34, silicon oxide passivation layer 32, silicon oxide layer 30, and into, but not through, silicon oxide layer 26. The etch is preferably a timed etch performed for a time in the range of about 180 to 220 seconds.

The currently preferred method for performing the last step of etching with an etch process that is selective to silicon nitride includes etching in situ the same process chamber as the first etch step, but with process parameters of pressure of 175 mTorr, power of 435W, magnetic field strength of 75 Gauss, gas flows of 35 sccm CF.sub.4, 90 sccm Argon, 16 sccm CH.sub.2 F.sub.2, and 20 sccm CHF.sub.3, and helium backside cooling at a pressure of 4500 mTorr.

FIG. 3 shows a cross section of structure identical to that of FIG. 1, except that a patch 40 is included in the overlying layers above gate stack 14. Patch 40 is preferably comprised of a material deposited during steps already performed during a standard fabrication process. Thus, to deposit patch 40, all that is required is that the mask at a certain step in the fabrication process be altered to allow patch 40 to be deposited, or to allow patch 40 to not be removed after deposition. Thus patch 40 may be incorporated without additional fabrication steps.

Patch 40 is preferably comprised of polysilicon or silicon nitride. Any suitable polysilicon layer or silicon nitride layer normally deposited after the formation of silicon oxide layer 26 may be employed. Thus patch 40 need not have exactly the location shown in FIG. 3, but may be located anywhere above gate stack 14 and below silicon nitride and silicon oxide passivation layers 34, 32. As currently most preferred, patch 40 is comprised of polysilicon that is deposited to form, on other areas of the wafer, the upper elements of stacked capacitors. This polysilicon, typically referred to as "cell poly," is not normally present above gate stack 14 in the fuse area, but may be used to form patch 40 over gate stack 14 by appropriate changes to the mask employed to pattern the cell poly. Thus the use of cell poly to form patch 40 is easily integrated into existing process flows. The use of cell poly to form patch 40 has the additional advantage of being deposited, in typical process flows, on a planarized surface, and hence it is flat, such that etch selectivity to the polysilicon surface is enhanced. It is also spaced away from the top the nitride cap over gate stack 14, such that the etch process used can easily be changed to an etch process selective to silicon nitride before nearing the nitride cap.

The embodiment of the present invention employed with the structure shown in FIG. 3 includes at least the following steps: (1) etching with an etch process selective to a material of which patch 40 is comprised, (2) etching with an etch process not selective to a material of which patch 40 is comprised, and (3) etching with an etch process selective to silicon nitride. If patch 40 is comprised of silicon nitride, an additional step of etching with an etch process that etches silicon nitride is required to be performed before step (1), above, in order to etch through silicon nitride passivation layer 34.

The embodiment of the present invention employed with the structure shown in FIG. 3 provides the advantages discussed above with respect to the embodiment described with reference to FIGS. 1 and 2. The embodiment of the present invention employed with the structure shown in FIG. 3 further provides the advantage of increasing the uniformity of the fuse opening etch depth across the surface of a wafer. The increased uniformity is achieved by including patch 40 above gate stack 14 and by etching, from sometime before the downward progression of the fuse opening reaches patch 40, with an etch process selective to a material of which patch 40 is comprised. The downward progress of the fuse opening then slows or stops at the level of patch 40.

The downward progress of the fuse opening need not actually stop on the upper surface of patch 40 for patch 40 to have the desired effect. All that is required is that patch 40 etch at a slower rate than the overlying layers above patch 40. At the areas of the wafer where patch 40 is reached earliest, the slower etch rate of patch 40 allows other areas of the wafer to at least partly catch up, decreasing the variation in the depth of the etch. The effects of etching with an etch process selective to the material of which patch 40 is comprised are shown in FIG. 4. The downward progress of the partially completed fuse opening 36 is stopped or at least significantly slowed at patch 40.

At some time after the downward progress of a fuse opening somewhere on a wafer first reaches patch 40, and preferably some time before the downward progress of any fuse opening on the entire wafer reaches completely through patch 40, the step is commenced of etching with an etch process not selective to the material of which patch 40 is comprised. This step removes the remaining portions of patch 40 over the entire surface of the wafer, and allows further etching to proceed with more uniform etch depth. Last, the step of etching with an etch process selective to silicon nitride is performed until bottom 38 of fuse opening 36 is located somewhere within the acceptable range described above with reference to FIG. 2, such as the position of bottom 38 shown in FIG. 5. The actual target for the depth of the fuse opening is preferably at the center of the acceptable range. Additional etch steps may optionally be employed before patch 40 is reached, or before the nitride covering gate stack 14 is reached, or both, in order to increase the etch rates of the various layers to be removed.

The various steps of the embodiment of the present invention employed with the structure of FIGS. 3 and 4 may be performed in separate pieces of etch equipment, or may be performed in a single piece of etch equipment, or may preferably be performed in situ in a single process chamber by merely varying the process recipe appropriately.

The currently most preferred etch processes for performing the steps of the embodiment of the present invention employed with the structure of FIGS. 3 and 4 are as follows. Recipe one: in an Applied Materials 5000, etching through silicon nitride passivation layer 34, silicon oxide passivation layer 32, silicon oxide layer 30, and into, but not through, silicon oxide layer 28 with a process recipe of 150 mTorr pressure, 700W power, 75 Gauss magnetic field, gas flows of 25 sccm CF.sub.4, 35 sccm CHF.sub.3, and 60 sccm argon, and with helium backside cooling at a pressure of 7000 mTorr. This corresponds to the additional step employed to etch through nitride passivation layer 34. Recipe two: etching in situ but with a recipe of pressure of 175 mTorr, power of 435W, magnetic filed strength of 75 Gauss, gas flows of 35 sccm CF.sub.4, 90 sccm argon, 16 sccm CH.sub.2 F.sub.2, and 20 sccm CHF.sub.3, and helium backside cooling at a pressure of 4500 mTorr. This etch step corresponds the step of etching with an etch process selective to a material of which patch 40 is comprised. This etch recipe is selective both to silicon nitride and to polysilicon, and so may be used with patches comprised of either material. Recipe three: etching in situ but with a recipe of 130 mTorr pressure, 225W power, 50 Gauss magnetic field strength, gas flows of 50 sccm Cl.sub.2 and 10 sccm HBr, and helium backside cooling at a pressure of 4500 mTorr. This recipe etches polysilicon and corresponds to the step of etching with a process not selective to a material of which patch 40 is comprised, particularly if patch 40 is comprised of polysilicon. If patch 40 is comprised of silicon nitride, recipe one described above for the additional step may be used here as well. Recipe four: etching in situ, again with a recipe of pressure of 175 mTorr, power of 435W, magnetic filed strength of 75 Gauss, gas flows of 35 sccm CF.sub.4, 90 sccm argon, 16 sccm CH.sub.2 F.sub.2, and 20 sccm CHF.sub.3, and helium backside cooling at a pressure of 4500 mTorr. Recipe four corresponds to the step of etching with an etch process selective to silicon nitride.

While the above etch processes are currently most preferred for performing the steps of the embodiment of the present invention employed with the structure of FIGS. 3 and 4, as an alternative preferred embodiment, if patch 40 is comprised of polysilicon, recipe one above may be used both to etch through silicon nitride passivation layer 34 and to stop or slow down at patch 40. Recipe one above would then correspond to the step of etching with an etch process selective to a material of which patch 40 is comprised. Then recipes three and four above would be used as described above, corresponding respectively to the steps of etching with an etch process not selective to a material of which patch 40 is comprised, and etching with an etch process selective to silicon nitride.

FIG. 6 shows a cross section of structure identical to that of FIG. 3, except that a patch 40 includes two layers, layer 42 and layer 44. In a currently preferred embodiment, layer 42 is cell poly and layer 44 is a silicon nitride insulating layer that is patterned in the same step as the cell poly and hence is present in patch 40 with layer 44. Patch 40 is not limited to one or two layers or to one or two materials. All that is required is that: (1) during the step of etching with an etch process selective to a material of which patch 40 is comprised, at least one of the materials of patch 40 act as an etch stop, and, (2) that during the subsequent step of etching with an etch process not selective to the material of which patch 40 is comprised, the etch process used be not selective to the materials of which any remaining layers of patch 40 are comprised.

The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.