Background of the Invention
Anti-fuse devices permit connecting underlying semiconductor devices otherwise isolated by rupturing, or converting to a lower resistance layer, an overlying fusing element by a high voltage electrical pulse. This permits connection of the underlying semiconductor device(s) with a predetermined circuit within the integrated circuit chip through a wire line overlying the fusing element. Anti-fuse modules may be used, for example, to compensate for defective devices elsewhere on the chip discovered during testing of the chip (wafer sort) by permitting access to underlying back-up semiconductor devices.
For example, field-programmable gate arrays, (FPGA) have been designed to contain the needed row of arrays, as well as additional rows of spare arrays, accessed if needed to replace ineffective counterparts, or to modify a specific design. These additional or spare arrays, or one-time fusible link structures, are sometimes comprised of an antifuse based programmable interconnect structure. The structure consists of an antifuse layer, usually a thin dielectric layer, placed between electrodes or conductive materials. When needed this antifuse material can be ruptured, or converted to a lower resistance layer, via a high voltage electrical pulse, resulting in creation of the replacement array structure.
Anti-fuse modules are fabricated along with other devices on the semiconductor chip. A prior art method of fabricating an anti-fuse module along with a normal device, such as a copper (Cu) dual damascene interconnect, required etching the antifuse stack over the normal device area which could damage the via portion of the normal device.
U.S. Pat. No. 5,877,075 to Dai, Jr. et al. describes a method of forming a dual damascene pattern employing a single photoresist layer.
U.S. Pat. No. 5,741,26 to Jain et al. describes a method for forming a dual damascene structure using an anti-reflective coating (ARC) layer of dielectric tantalum nitride (Ta.sub.3 N.sub.5) to reduce the amount of reflectance from the underlying conductive region and thus reduce the distortion of the photoresist pattern.
U.S. Pat. Nos. 5,705,849 and 5,602,053, both to Zheng et al. describe an anti-fuse structure and the method of making it, respectively. The anti-fuse structure comprises two dual damascene connectors separated by a structure comprising a pair of alternating layers of silicon nitride and amorphous silicon.
Summary of the Invention
Accordingly, it is an object of the present invention to provide a method of forming an anti-fuse module compatible with a Cu dual damascene process.
Another object of the present invention is to provide a method of integrally forming an anti-fuse module with a dual damascene process without impacting on the baseline of that dual damascene process.
Yet another object of the present invention is to provide a method of forming an anti-fuse module on a smaller design rule.
A further object of the present invention is to provide a method of forming an anti-fuse module while protecting the via portion of the normal device.
Other objects will appear hereinafter.
It has now been discovered that the above and other objects of the present invention may be accomplished in the following manner. Specifically, a semiconductor structure having at least two exposed metal lines covered by a first dielectric layer is provided. A first metal line is within an anti-fuse area and a second metal line is within an interconnect area. A first metal via is formed within the first dielectric layer within the anti-fuse area with the first metal via contacting the first metal line. A SiN layer is deposited over the first dielectric layer and the first metal via. The SiN layer is patterned to form at least two openings. A first opening exposes the first metal via, and a second opening exposes a portion of the first dielectric layer above the second metal line. A fusing element layer is deposited and patterned over the patterned SiN layered structure to form a fusing element over the first metal via. Simultaneously, an anti-fuse metal line is formed over the fusing element to form an anti-fuse module within the anti-fuse area, and a dual damascene interconnect is formed over, and contacting with, the second metal line and within the interconnect area.
Brief Description of the Drawings
The features and advantages of the method of fabricating an anti-fuse module integrated with a dual damascene process according to the present invention will be more clearly understood from the following description taken in conjunction with the accompanying drawings in which like reference numerals designate similar or corresponding elements, regions and portions and in which:
FIGS. 1 through 4 schematically illustrate in cross-sectional representation a preferred embodiment of the present invention.
Detailed Description of the Preferred Embodiment
Accordingly as shown in FIG. 1, starting semiconductor structure 10 includes two exposed metal lines 12, 14 and is understood to possibly include a semiconductor wafer or substrate, active and passive devices formed within the wafer, conductive layers and dielectric layers (e.g., inter-poly oxide (IPO), intermetal dielectric (IMD), etc.) formed over the wafer surface.
The term "semiconductor structure" is meant to include devices formed within a semiconductor wafer and the layers overlying the wafer. Unless otherwise specified, all structures, layers, etc. may be formed or accomplished by conventional methods known in the prior art.
First dielectric layer 20 is deposited over semiconductor structure 10 and is patterned to form trench 22 from about 6000 to 8000 .ANG. deep. First dielectric layer 20 may comprise a low-k (low dielectric constant) material or other oxide material.
Barrier layer 24 may be formed, lining trench 22. A planarized, first metal via 26 is formed in trench 22, filling trench 22. over barrier layer 24. First metal via 26 defines the bottom contact of the anti-fuse module to be formed within anti-fuse area 16 and may comprise tungsten (W), aluminum (Al) or copper (Cu). Barrier layer 24 may comprise an appropriate material for the selected metal used to form first metal via 26. Second metal line 14 is within interconnect area 18.
A layer 28, preferably silicon nitride (SiN), is deposited over first dielectric layer 20 and first metal via 26. Layer 28 is from about 500 to 1000 .ANG. thick and may also comprise two layers of SiN and oxide, or three layers of oxide, SiN, then oxide.
As shown in FIG. 2, SiN layer 28 is patterned to form openings 30, 32 over first metal via 26 within anti-fuse area 16 and over first dielectric layer 20 within interconnect area 18 and over second metal line 14, respectively. Opening 30 exposes, and preferably extends beyond, first metal via 26. Opening 32 exposes a portion of first dielectric layer 20 and preferably does not extend beyond underlying second metal line 14.
Fusing element layer 34 is formed over the structure, covering patterned SiN layer 28 and filling openings 30, 32. Fusing element layer 34 is from about 1000 to 2000 .ANG. thick and preferably comprises alternating layers of tantalum nitride (TaN) 36/amorphous silicon (a-Si) 38/TaN 40 but may comprise other alternating layers of materials, such as TaN/oxide/TaN, TiN/amorphous-Si/TiN, or TiN/oxide/TiN.
Fuse mask 42 is formed over the portion of fusing element layer 34 within anti-fuse area 16. Fuse mask 42 is preferably comprised of photoresist.
As shown in FIG. 3, the unmasked fusing element layer 34 is removed leaving fusing element 44 under fuse mask 42. Fuse mask 42 is then removed.
Second dielectric layer 46 is deposited over the structure. Second dielectric layer 46 and part of first dielectric layer 20 are then patterned to form the following openings. Anti-fuse metal line opening 48 is from about 3000 to 7000 .ANG. deep and is formed over fusing element 44 within second dielectric layer 46. Upper dual damascene metal line opening 50 is from about 4000 to 8000 .ANG. deep and is formed over, and extends beyond, opening 32 in patterned SiN layer 28. Lower dual damascene via opening 52 is from about 6500 to 9000 .ANG. deep and is formed within first dielectric layer 20, and includes opening 32 within patterned SiN layer 28, over second metal line 14. Via opening 52 and metal line opening 50 form dual damascene opening 54.
FIG. 4 shows the results of a dual damascene process. A metal layer (not shown), preferably comprised of copper (Cu), is deposited over the structure shown in FIG. 3 at least as high as the upper surface of patterned second dielectric layer 46. The Cu layer fills anti-fuse metal line opening 48 over fusing element 44, and fills dual damascene opening 54 over second metal line 14.
The Cu layer is then planarized to remove the excess Cu and forms anti-fuse Cu metal line 56, completing the anti-fuse module within anti-fuse area 16, and forms dual Cu damascene interconnect 58 within interconnect area 18. Anti-fuse metal line 56 contacts fusing element 44 and dual damascene interconnect 58 contacts second metal line 14.
When/if desired, fusing element 44 can be ruptured, or converted to a lower resistance layer, via a high voltage electrical pulse through anti-fuse metal line 56 thereby permitting an electrical connection between anti-fuse metal line 56 and first metal line 12 through first metal via 26, thus activating or connecting the underlying device(s) connected through first metal line 12.
The process of forming the anti-fuse module of the present invention may thus be used with a dual damascene process without damaging the via portion of the dual damascene interconnect structure. In contrast, the normal oxide-CMP (chemical mechanical polishing) and tungsten (W)-CMP process open fuse via and normal via together. When an anti-fuse structure is built on the fuse via, many etching and cleaning steps are required for the fuse and will affect the normal fuse surface. This is especially true for cleaning solvents that attack the normal via barrier metal which causes higher via resistance and via missing problem.
The process of the present invention also permits formation of anti-fuse modules on a smaller design rule because a dual damascene process provides a very advanced design for local interconnections.
While particular embodiments of the present invention have been illustrated and described, it is not intended to limit the invention, except as defined by the following claims.