Patent Application
20090166564
1. Field of the Invention
The present invention relates generally to semiconductor fabrication. More particularly, the present invention relates to monitoring ion implantation procedures on wafers.
2. Background of the Invention
Semiconductor device construction is highly sensitive to particulate contamination. High current ion implantation is a significant source of particles that are generated from a variety of sources such as mechanical friction and lead rubbing. Additionally, small variations in the wafer tilt angle can cause errors in the implant angle, which leads to defective implantations. Implanters are regularly tested for defects by running monitor wafers through the implanter. These monitor wafers are tested for particulate contaminants (PCs), ion beam implant angle errors, and so on. However, current testing methods do not provide reliable results. Testing conditions often do not simulate realistic manufacturing conditions. Thus, improved methods for testing the efficacy of an ion implanter are needed.
Ion implantation is a surface modification process in which ions are injected into the near-surface region of a substrate. High-energy ions are produced in an accelerator and directed as a beam onto the surface of the substrate where they form an alloy with the surface upon impact. A typical ion-implanter used in many semiconductor fabrication plants today is the Varian E500 implanter. This type of implanter processes semiconductor wafers in batches, or “lots.” A single wafer is placed on a wafer holder, and an ion beam is aimed at the wafer. The wafer holder has the ability to rotate the wafer, and “tilt” the wafer around a horizontal axis normal to the incident angle of the ion beam (beam entry angle, or BEA). See
The use of angled implants with wafer rotation, tilt, and twist has added new failure modes. One such failure mode is the generation of Particulate Contaminants (PC). Wafer rotation, tilt, and twist can introduce new PC generation mechanisms that must be quantified. These movements are a source of PC buildup on wafers. To rotate, the wafer holder sits on ball bearings. This mechanical movement causes friction and wearing/rubbing of parts. The ultra-clean vacuum environment within the implanter makes the wafers susceptible to particulate matter, dust, etc. generated by this friction. PCs lead to increased failure rate among components, and reduced overall yield. Additionally, at certain tilt angles, movement of mechanical parts around the wafer holder causes rubbing of parts against the walls of the chamber. In the E500 for instance, a phenomenon called “lead rubbing” involves rubber insulation around wires rubbing against the inner walls of the implanter. This releases more particles into the implanter environment, resulting in greater particulate contamination.
Another failure mode is caused by inaccurate dose placement. The tilt angle error is an important controlling variable affecting implant profile and dose placement. The reason is that in a crystallographic structure such as a silicone substrate or wafer, dopants and ions achieve depths of penetration that vary with the angle of the wafer. Small mechanical misalignments may cause dramatic variations in effective current/gate length, and may lower the yield, especially for advanced 65 and 45 nm technologies. Occasionally wafers are subjected to multiple operations on several machines. It is therefore very beneficial to ensure uniform calibration across these stations to increase overall yield.
There are several existing methods known in the art for testing production defects in an ion implanter. For instance, to detect PC contamination, a test wafer or monitor wafer is subjected to an ion implantation. Then, statistical process control (SPC) is used to detect abnormal or inconsistent PC counts on the monitor wafer. SPC involves using statistical techniques to measure and analyze the variation in processes. SPC comparison detects any unusual variation in the manufacturing process, which could indicate a problem with the process. However, current testing methods used for older implanter technologies are not effective in measuring PCs in present situations. It has been found that using one or two monitor wafers may show no signs of PC contamination, but processing large production lots of batches still leads to reduced yields due to contamination. Additionally, phenomena like lead rubbing are not adequately considered in present testing methods, which only test single wafers at small tilt angles between 0 and 7 degrees.
Furthermore, miscalibrations in the tilt angle of the wafer holder are not adequately compensated for in conventional testing methods. SPC is used to monitor dose accuracy after subjecting monitor wafers to tilt angles of only 0-7 degrees. Thus, what is needed is a method for testing implanters that simulates real-life conditions and provides measurable and quantifiable results.
The present invention discloses methods for testing an ion implanter. Batches of wafers in an ion implanter are subjected to a series of mechanical operations characteristic of the implanting process. According to one exemplary embodiment, the present invention is a method involving subjecting a plurality of wafers to rotation, twist, and extreme tilt. Multiple wafers are used because testing only 1-2 monitor wafers tested by themselves would not yield realistic and measurable PC contamination because often times the efficacy of the implanter machines decreases with increased number of wafers used. Furthermore, extreme tilt and twist guarantee that lead rubbing and other phenomena inherent in an implanter are not ignored. Thus, the last few wafers from a batch subjected to rotation, extreme tilt, and twist may be used to monitor particles using SPC.
In another exemplary embodiment, monitor wafers are implanted immediately after processing a production lot subjected to rotation, tilt, and twist, loading the monitor wafers in a cassette position such that they are implanted after the production lot.
In another exemplary embodiment, instead of full production conditions, only extreme tilt is used. This can be used to isolate PC contamination to that generated from phenomenon like lead rubbing alone.
In an alternative exemplary embodiment, the present invention is a method for testing an ion implanter and correcting implanter profile, dose placement and accuracy. This can be achieved by monitoring minimum sheet resistance of the dosage area as a function of the tilt angle around a known crystallographic channel. The implanter setup is adjusted to minimize the sheet resistance. By defining this angle as the “standard qual angle”, the implanter setup angle accuracy can be statistically tracked. Thereby, monitoring sheet resistance at this known angle of high sensitivity allows one to calibrate all implanters to the same physical angle.
The present invention discloses methods for testing an ion implanter by subjecting multiple wafers to rotation, twist, and tilt. A wafer holder within an ion implanter, such as the E500 implanter, processes a lot or batch of at least four wafers. Multiple wafers are used so that a measurable number of PCs can be generated. In one exemplary embodiment of the invention, any wafer beyond the first three wafers can be used as a monitor wafer. The wafer holder twists the wafer by rotating its surface in 90 degree increments. This ensures more accurate testing conditions because it produces PCs that will be generated by the friction of ball bearings within the rotation mechanism, thus simulating actual production conditions. The wafer holder also tilts the wafer to an extreme angle (at least 30 degrees) about its horizontal axis. It was found that large tilt angles on the E500 series cause “lead rubbing”, forcing a rubber insulated wire against the side wall of the implanter. This combination of physical movements generates sufficient PCs such that in a batch of 10 wafers, the 10th wafer would have the highest PC levels. It was also found that a minimum of 4 wafers are required to be tested to detect measurable PC levels.
The present invention also addresses the issues relating to dose placement and accuracy. Around a specific implant tilt angle, in this case found to be around 36 degrees, a crystallographic channel within the silicon wafer is aligned with the implant beam. At this angle, the silicon crystal lattice structure provides minimum resistance to implant ions and dopants within the ion beam. Thus, dopants and impurities within the ion beam traverse deeper within the substrate. This lowers the sheet resistance at the surface of the wafer. Specifically, at a tilt angle of 36 degrees, sheet resistance is found to be at a minimum. Furthermore, at this angle, there is maximum sensitivity in sheet resistance to small variations in implant angle. This allows the wafer holder in an ion implanter to be calibrated to a more precise angle. Thus, it is an objective of an exemplary embodiment of the present invention to improve dose accuracy and placement by performing high-angle implants at 36 degrees, measuring sheet resistance, and correlating the sheet resistance to known configurations to accurately calibrate the tilt angle and consequently the beam angle.
For the purposes of the present invention, a “substrate” or “wafer” includes any thin slice of semiconducting material, such as a silicon crystal, upon which microcircuits are constructed by doping, chemical etching, and deposition. Substrates may undergo Shallow Trench Isolation (STI), Chemical-Mechanical Planarization (CMP), lithography, ion implantation, and other processes. A “lot” or “batch” of wafers comprises at least four wafers that are processed by an ion implanter within a short time span. At least four wafers will be required for a successful PC contamination test.
A “Monitor Wafer” or simply “monitor” is a wafer from the batch of wafers that is subjected to testing after undergoing an ion implantation process. According to an exemplary embodiment of the present invention, the monitor wafer is the last wafer in a batch of wafers. In another exemplary embodiment, the monitor wafer is at least the 4th wafer in the batch, or any higher-numbered wafer. In yet another exemplary embodiment, the monitor wafer may be any wafer implanted immediately after a production batch of wafers such that it collects PCs from the production batch.
“Tilt” and “Twist” are physical movements applied to a wafer positioned in a wafer holder within an ion implanter. The respective orientations of these movements are shown in
Additionally, the twist angle is helpful to completely describe the direction of incidence of the ion beam. Wafer holder 110 can also twist a wafer by rotating along the normal axis of the wafer. This rotation is shown by circle 149. In a typical E500 implanter, wafer holder 110 rotates over a rollerboard while positioned on ball bearings. The friction between the ball bearings and the rollerboard is one cause of PC generation. Wafers are twisted to vary the angle of the ion beam 120. The twisting mechanism causes additional stress on the ball bearings, increasing the rate of PC generation, and may also cause lead rubbing problems.
The effect of using multiple wafers is shown in
The graph shows that a single pilot test is insufficient to cause PCs; rather, multiple wafer implants are required to cause measurable buildup. Additionally, since implant self-cleaning occurs fairly quickly (within less than two batches of wafers), it is difficult to diagnose the situation without using multiple wafers. Finally, it is apparent that tilt or twist is required for PC generation. Thus, some combination of rotation, tilt, twist, along with multiple wafers is required to achieve measurable PC buildup.
In an alternative exemplary embodiment, monitor wafers are implanted immediately after a production lot undergoing these processes. The production lot uses twist rotation and extreme tilt. The monitor wafer(s) should be implanted immediately after the production lot. This ensures PC buildup from the production lot affects the monitor wafer as it would affect wafers within the production lot.
As described herein, another problem addressed by the present invention relates to the implant dose accuracy. In one exemplary embodiment, the present invention improves implant profile and calibration of dose placement. This is achieved by monitoring sheet resistance as a function of tilt angle for a monitor wafer around a known crystallographic channel, and adjusting the implanter setup to minimize sheet resistance. Furthermore, by selecting one angle near or in the crystallographic channel as the “standard qual angle”, the implanter setup angle accuracy can be statistically tracked. This allows one to monitor changes in sheet resistance at an angle of maximum sensitivity, thus allowing all implanters to be calibrated to the same physical angle. The same monitor wafer may be used for both PC checking and dose accuracy. After undergoing SPC for PC contamination, the wafer may be annealed, thus activating the electric components. Sheet resistance is then tested. This serves to prevent waste of wafers while minimizing experimental runs.
In one exemplary embodiment, after implantation, the monitor wafer is annealed and then checked for resistivity. After implantation, the wafer is placed in an annealer, such that the dopant material interacts with the crystalline structure making it electrically active. Sheet resistance is then tested to determine a minimum value or to compare to a known minimum value. The tool can thus be recalibrated to match minimum resistance at the standard qual angle.
The advantages of this process are numerous. For one, the disclosed method assures accurate particle monitoring, allowing for higher standards. Complying with the standards saves costs of end-of-line fault analysis and diagnoses. Additionally, this method reduces the number of pilot runs, and can be applied to multiple existing and future technologies undergoing rotation and twist implants. Since multiple implanters including the entire E500 series are subject to the problems from lead rubbing, this method would work on these devices, reducing the risks of lot scrapping. Furthermore, since newer implanters are more accurate, older implanters are able to be reprogrammed to match the accuracy of the new implanters using the standard qual angle check.
The foregoing disclosure of the exemplary embodiments of the present invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many variations and modifications of the embodiments described herein will be apparent to one of ordinary skill in the art in light of the above disclosure. The scope of the invention is to be defined only by the claims appended hereto, and by their equivalents.
Further, in describing representative embodiments of the present invention, the specification may have presented the method and/or process of the present invention as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described. As one of ordinary skill in the art would appreciate, other sequences of steps may be possible. Therefore, the particular order of the steps set forth in the specification should not be construed as limitations on the claims. In addition, the claims directed to the method and/or process of the present invention should not be limited to the performance of their steps in the order written, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the present invention.
