Active solid-state devices (e.g. – transistors – solid-state diode – Field effect device – Having insulated electrode
Reexamination Certificate
2002-05-16
2004-03-02
Clark, Jasmine (Department: 2815)
Active solid-state devices (e.g., transistors, solid-state diode
Field effect device
Having insulated electrode
C257S209000, C257S363000
Reexamination Certificate
active
06700161
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a structure and method for forming a controllably variable resistor having wide application to a variety of unique digital and analog circuits, including both programmable digital integrated circuit devices, and programmable analog integrated circuit devices.
The present invention even more particularly relates to a structure and method for forming and setting a programmable resistive element such as a fuse or antifuse used in programming digital integrated circuit devices including redundant memory elements, for example. This invention also relates to a programmable or trimming resistor used in analog RF circuit tuning applications. More specifically, this disclosure relates to a structure and method for non-ablatively forming a fuse, antifuse, or trimming resistor element without ablation by selectively altering resistivity of the variable resistor to a finite value.
2. Description of Related Art
The continued progress in improving integrated circuit (IC) performance, either by device scaling or by more efficient utilization of chip area, is directed to allowing faster and smaller devices to be manufactured, as well as to allowing a reduction in manufacturing process time and expense, both during manufacture of the semiconductor device itself, as well as during the testing of the device.
Traditionally, programmable devices or devices incorporating redundant circuitry are manufactured to provide end-user flexibility in the ultimate application of the device, and/or to increase production yield. Applications, which often use such redundant circuitry or programmable elements, include, for example, programmable logic arrays (PLA) or dynamic random access memory (DRAM) devices.
Fuses are employed in integrated circuits to encode or “program” information on a circuit chip at the time of manufacture. A fusible link, or “fuse”, is one that provides a closed or low resistance connection when first formed, and which is modified to provide an open or high resistance circuit when programmed.
The encoded information is used to later identify the chip, to enable or tune circuits depending on test results, or to repair defective regions of the chip by enabling spare or redundant circuits. The redundant circuitry can be selectively removed from the final device configuration by the use of fusible conductive links, or fuses.
In the case of fusible conductive links, an ablative approach is often used to provide flexibility and improve production yield. However, there is an area penalty incurred on the chip by the inclusion of the redundant circuitry, and damage may result to the surrounding circuitry.
The typical method for providing fuses is to form small conductive paths that can be selectively ablated with a precisely positioned laser beam, or by providing a current that is high enough to melt the conductive material. Such, ablation of the conductive link encodes the necessary information as a series of bits, or selectively enables or disables one or more circuits in the integrated circuit. A drawback with this method is that the area required by the region damaged by ablation is relatively large, so that features and devices cannot be fabricated near the fuses. Another drawback with laser-ablated fuses is that they do not scale well with lithography.
The ablative damage associated with such a fuse blowing process typically extends at least a few microns around the fuse, and often extends into the top few layers of the so-called “back-end-of-the-line” (BEOL) structure. BEOL processing, also known in the industry as “back end”, generally is considered to include steps from contact to the semiconductor substrate through completion of the wafer, prior to electrical test. Structure resulting from back-end processing may include, for example, addition of insulating or conductive material, e.g., Copper (Cu) used in high performance processors, and dielectric insulators such as silicon dioxide.
The damage to such back-end structures poses reliability concerns, such as the electrical shorting of elements that the manufacturer does not intend to be shorted. Damage resulting from blowing fuses also imposes limits on the proximity of adjacent fuses, and hence fuse pitch reduction which directly affects circuit packing density, and which may be cumulative to the area penalty imposed by the inclusion of the redundant circuitry. Typical fuse pitches are limited to the range of 3 to 10 &mgr;m, are conventionally available, with a fuse pitch of greater than 3 &mgr;m being common in processes that open or “blow” the fusible links ablatively.
The damage and debris that occurs around the programmed link after ablation can limit the achievable pitch in manufacturing the device and, consequently, the level of integration and miniaturization. Therefore, both the ultimate scalability and reliability of the device are necessarily affected adversely by conventional approaches to programming a fuse. The technology used for integrated circuit manufacturing is migrating to low k and ultra low k materials which are mechanically very weak and very susceptible to damage from disruptive fuse programming like ablation. Non-ablative fuses can play an important role for advanced CMOS interconnect technology.
The use of fusible links for device personalization has further inherent limitations. Specifically, the fuse link can only be blown open or left closed; they cannot be used to close a previously opened link. Personalization where a previously open connection is made closed requires another approach.
Redundant circuitry can be selectively added to the final device configuration by the use of antifusible links, or “antifuses”, which are structures that, when first fabricated, are an open or high resistance circuit. When the antifuse is “programmed” the open circuit becomes closed, and conduction across the antifuse becomes possible. Thus, antifuses are used to perform the opposite and complementary function of a fuse.
In the case of either using a fuse or antifuse to program a device, making such discretionary connections alters the function or operating characteristics of integrated circuits. Typically, when a sufficient voltage called a “fusing voltage” is applied across an antifuse structure, the resulting current flow and energy imparted into the fuse element causes the structure to change into an electrically conductive state, or become permanently shorted, and an electrical connection is made. Antifuses are also used in a wide variety of applications, including Field Programmable Gate Arrays (FPGA).
Conventional antifuse technology has several disadvantages. For example, many conventional antifuses require specific metal types to be used as electrodes. These metals are not always compatible with common fabrication technologies. For example, some conventional approaches require a transparent electrode, and thus cannot use electrodes consisting of aluminum or polysilicon, which are opaque. Furthermore, these antifuse structures generally require 12-15 volts to fuse the antifuse. Applying such a voltage to the antifuse can also cause damage to other circuit elements, and thus these antifuses may be incompatible with modem low-voltage semiconductor devices that commonly operate at 3.3 volts or 2.5 volts. Additionally, these structures will be difficult to scale to the significantly smaller sizes that will be required as semiconductor device density increases, for similar reasons to those noted with respect to fuse pitch reduction limits imposed by ablative damage to adjacent BEOL structure.
In other applications, for example, radio frequency (RF) integrated circuit applications, impedance matching between devices and circuits is very important to ensure that a low VSWR is attained, so that proper operation of the circuits can be maintained. As various semiconductor devices are usually incorporated into such an RF integrated circuit, the interactions between the various signals present and the numerous devices can be complex, and corr
Hsu Louis Lu-Chen
Narayan Chandrasekhar
Radens Carl J.
Clark Jasmine
Connolly Bove Lodge & Hutz LLC
Hume Larry J.
Trepp Robert M.
LandOfFree
Variable resistor structure and method for forming and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Variable resistor structure and method for forming and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Variable resistor structure and method for forming and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3268201