In-situ real-time plasma chamber condition monitoring
In-situ monitoring of free radicals in semiconductor chambers addresses the challenges of chamber seasoning and drift by providing real-time feedback for efficient chamber maintenance and performance optimization.
Patent Information
- Application Number
- JP2024164697
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2019-03-15
- Filing Date
- 2024-09-24
- Publication Date
- 2024-12-24
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The accumulation of unstable etching by-products on chamber walls in semiconductor manufacturing leads to particle formation and processing drift, necessitating time-consuming chamber cleaning and seasoning procedures that introduce moisture and increase downtime.
In-situ monitoring of free radical frequencies and wavelengths within the chamber to provide real-time feedback on chamber conditions, allowing for adjustments and reducing the need for seasoning and qualification cycles.
Enables real-time monitoring and control of chamber conditions, minimizing downtime and improving etch performance by maintaining optimal chemical levels without production interruptions.
Smart Images

Figure 2024178337000001_ABST
Abstract
Description
background
[0001] FIELD OF THE DISCLOSURE FIELD OF THE DISCLOSURE Embodiments of the present disclosure relate generally to semiconductor processing techniques, and more particularly to in-situ process monitoring methods.
[0002] Description of Related Art Semiconductor manufacturing involves a series of processes in which electrical circuits are fabricated in semiconductors (e.g., silicon wafers) according to a circuit design. These processes are carried out in a series of chambers. Successful operation of a modern semiconductor manufacturing facility requires a steady stream of wafers moving from one chamber to another in the process of forming the electrical circuits in the wafers. Such processes inevitably generate various types of etch by-products. Some of the by-products deposit on the interior surfaces of the chamber in which the plasma etch process takes place.
[0003] The continued accumulation of by-products on internal surfaces such as chamber walls presents two challenges to semiconductor manufacturing. First, the structure of the accumulated by-products is not stable. Because of this, the by-products tend to flake off the chamber walls, creating particles and debris that can fall onto the wafer surface, causing product defects such as shorts between two conductive lines or discontinuities where an overlying layer cannot cover the debris. Second, by-products remaining on the chamber walls can react with the plasma and adversely affect etch performance, a phenomenon also known as "process drift."
[0004] To mitigate the effects of etch by-products, periodic chamber cleaning is required to remove deposits from the chamber walls. This is accomplished by removing the chamber from production and introducing a cleaning plasma into the chamber. This plasma reacts with the deposits and the products of this reaction are exhausted from the chamber. However, it has been observed that after such chamber cleaning, the clean chamber walls make the chamber unsuitable for immediate etching of production wafers.
[0005] When the chamber is open, moisture is introduced into the chamber from the environment. Moisture is a contaminant that can affect the performance of the chamber. Chamber seasoning is a procedure in which a series of blank silicon wafers are etched to restore the chamber walls to a condition suitable for etching production wafers. After chamber seasoning, a thin layer of silicon oxide covers the chamber walls. After chamber seasoning, the chamber goes through a qualification cycle in which qualification wafers are produced and inspected to determine whether the chamber is OK to return to the production cycle. If the qualification wafers meet specifications, the chamber is returned to production wafer etching. If the qualification wafers do not meet specifications, more wafers are processed, the chamber is further seasoned, and then more qualification wafers are processed and inspected.
[0006] The chamber seasoning time and wafer qualification time are very time consuming. If the qualified wafers do not meet the specifications the first time, additional time is spent re-seasoning the chamber and testing more wafers for qualification. This results in longer chamber downtime.
[0007] It would therefore be highly desirable to develop a process monitoring and chamber seasoning method that does not rely on etch rate measurements and therefore does not require interruptions in production or seasoning. It would also be preferable for such a method to monitor the condition of the chamber walls, providing real-time and accurate information about process drift and chamber seasoning.
[0008] FIELD OF THE DISCLOSURE Embodiments herein relate generally to semiconductor processing techniques, and more particularly to a method for in-situ process monitoring. Some embodiments herein generally relate to a method for in-situ and real-time chamber condition monitoring. For example, in one embodiment, the frequency and wavelength of free radicals in a chamber are monitored in-situ for each wafer in the chamber. The frequency and wavelength of the free radicals are associated with at least one selected chemical product. The associated free radicals are compared to an index. The index includes a target range for each chemical in the at least one selected chemical product.
[0009] In another embodiment, at least one wafer is inserted into the chamber. A gas is injected into the chamber. The gas includes a source of free radicals. The frequency and wavelength of the free radicals in the probe gas are monitored. The frequency and wavelength of the free radicals are associated with at least one selected chemical product. The associated free radicals are compared to an index. The index includes target ranges for each chemical in the at least one selected chemical product. The in situ chamber condition recipe is altered in response to the comparison if the associated free radicals are not within the target ranges.
[0010] In yet another embodiment, the frequency and wavelength of fluorine free radicals in the chamber are monitored in situ for each wafer in the chamber. The frequency and wavelength of the free radicals indicative of fluorine are compared to an index. The index includes a target range for fluorine. The fluorine level in the chamber can be increased when the free radicals indicative of fluorine approach the lower limit of the target range. [Brief description of the drawings]
[0011] In order that the above features of the present disclosure may be understood in detail, a more particular description of the present disclosure briefly summarized above will be made with reference to embodiments, some of which are illustrated in the accompanying drawings. However, since the present disclosure may include other equally effective embodiments, it should be noted that the accompanying drawings merely illustrate exemplary embodiments of the present disclosure and therefore should not be construed as limiting the scope thereof. [Figure 1] 1 is a schematic diagram of an inductively coupled plasma reactor according to an embodiment disclosed herein. [Diagram 2] 1 illustrates a method for in-situ process monitoring according to an embodiment disclosed herein. [Diagram 3] 1 illustrates a graph according to an embodiment disclosed herein. [Figure 4] 1 illustrates a graph according to an embodiment disclosed herein.
[0012] To facilitate understanding, wherever possible, identical reference numbers have been used to refer to identical elements common to the drawings.
[0013] In the following description, numerous specific details are set forth to allow the present disclosure to be more fully understood. However, as will be apparent to those skilled in the art, various modifications can be made using different configurations without departing from the scope of the present disclosure. In other embodiments, well-known configurations are not described to avoid obscuring the disclosure. Therefore, the present disclosure is not to be considered limited to the specific exemplary embodiments shown herein, and all alternative embodiments of such kind are intended to be included within the scope of the appended claims.
[0014] A schematic diagram of the reactor is shown in FIG. 1. The plasma reactor / chamber 100 comprises an upper chamber body 105 and a lower chamber body 108. Above the upper chamber body 105 is a dielectric dome 120. Chamber walls 118 surround the lower chamber body. The dome material is The material is alumina and the material of the chamber walls 118 is anodized aluminum. The dielectric dome 120 and the chamber walls 118 are maintained at constant temperatures (80° C. for the dome and 65° C. for the chamber walls 118).
[0015] A high density plasma is generated by applying a 12.56 MHz radio frequency (RF) power supply 110 to an inductive coil 115 on a dielectric dome 120, which is mounted on top of the upper chamber body 105. A cathode 125 with an electrostatic chuck (not shown) holds a semiconductor wafer 130, which is also thermally controlled by helium backside cooling during etching. A separate 13.56 MHz RF bias power supply 135 is applied to the cathode 125 to control the ion bombardment energy at the wafer 130. The chamber 100 is evacuated by a 2000 liters / second turbomolecular pump 140 assisted by a dry mechanical pump (not shown).
[0016] Process gases are introduced at controlled flow rates through four gas distribution rings 145 in the chamber wall. The chamber pressure is maintained at a specified value by a throttle valve 150. An optical spectrometer 160, which includes a prism 162, a monochromator 164, and a photodiode array (PDA) detector 166, is used to record the emission spectrum. The optical signal is collected from a sensor 170 in the chamber and sent through a multi-strand optical fiber 175. The sensor 170 transmits the data to the spectrometer 160. The spectrometer 160 classifies the free radicals into chemicals related to their frequencies and wavelengths. As used herein, "chemicals" is broadly defined to include chemical products and chemical compounds. The spectrometer 160 may classify the free radicals into chemicals (e.g., fluorine or oxygen) and / or chemical compounds (e.g., fluorine nitrate or silicon nitrate). To conserve processor resources, the customer / end user can select the chemicals to be identified.
[0017] A control system 180, comprising a CPU 182, a memory 184, and support circuitry 186 for the CPU 182, controls the various components of the plasma reactor 100. The CPU 182 executes software routines or sequences of program instructions stored in the memory 184, which cause the reactor 100 to perform the processing of materials as disclosed herein. In various embodiments, the memory 184 also contains programs (not shown) to implement the embodiments described herein.
[0018] In other embodiments, the memory 184 includes programs (not shown) to identify chemicals in the chamber 100 based on the wavelength and frequency of free radicals, issue alerts based on the relationship between target ranges and associated free radical frequencies and wavelengths, and increase the Green-2-Green Time by in-situ monitoring of the frequency and wavelength of free radicals in the chamber 100.
[0019] The processor 182 cooperates with conventional support circuits 186 (such as power supplies, clock circuits, cache memory, etc.), as well as circuitry that aids in the execution of software routines (not shown) stored in memory 184. Thus, it is contemplated that some of the process steps discussed herein as software processes may be loaded from a storage device (e.g., optical drive, floppy drive, disk drive, etc.) and implemented in memory 184 for operation by the processor 182. Thus, the various processes and methods of the present disclosure may be stored on a computer-readable medium.
[0020] The chamber 100 periodically requires preventive maintenance ("PM") maintenance, referred to herein as a "PM cycle." The chamber 100 may be taken out of production and subjected to a PM cycle based on the performance of the tool. A PM cycle may take the form of a "minor PM" cycle or a "major PM" cycle. A minor PM cycle is most commonly Most parts are not changed, whereas in major PM many parts are changed. To perform PM, the chamber 100 must be opened and exposed to the atmosphere outside the chamber 100. By changing more parts in the chamber 100, the chamber 100 is opened for a longer period of time, increasing the exposure to moisture (e.g., OH) in the chamber 100. Parts introduced into the chamber 100 also bring moisture (e.g., OH) into the chamber 100.
[0021] To return the chamber 100 to production, the chamber 100 undergoes a seasoning cycle and a qualification cycle. According to embodiments described herein, the wavelength and frequency of the chemicals may be monitored to shorten the seasoning and qualification cycles. For illustrative purposes, the change in OH levels during the seasoning and qualification cycles is shown in FIG.
[0022] 2 illustrates an exemplary method 200 for in-situ monitoring the frequency and wavelength of a selected gas in a process chamber 100. For example, during a seasoning cycle, the method 200 begins at block 202 and proceeds to block 204. At block 204, the sensor 170 monitors in-situ the frequency and wavelength of free radicals of the gas in the chamber 100.
[0023] At block 206, the free radicals are associated with at least one chemical product based on their frequency and wavelength. For example, free radicals having wavelengths and frequencies indicative of chlorine are associated with chlorine and are classified into a channel for chlorine. Examples of other gases that may be associated by their frequency and wavelength include carbon nitride, oxygen, fluorine, xenon, carbon, argon, silicon, silicon nitride, hydrogen, and hydrides.
[0024] Additionally, a first silicon of one wavelength and frequency may be monitored and classified differently from a second silicon of a different wavelength and frequency than the first silicon. In various embodiments, a selection is made (e.g., by the customer) to monitor one chemical (such as fluorine during a manufacturing cycle) or any combination of chemicals (such as all chemicals typically found in the chamber during a seasoning cycle).
[0025] At block 208, the free radicals in each channel (i.e., associated free radicals) are compared to an index (not shown). The index includes a target range for each chemical in at least one chemical product. For example, if one chemical (e.g., fluorine) is selected, a target range for fluorine is selected to compare to the free radicals associated with the fluorine identified to be in the chamber 100. A user may select which associated free radicals to monitor. A user may also graph the change in intensity of the associated free radicals. For example, during a seasoning cycle, a part is replaced in the chamber 100 and then the chamber 100 is closed. The wavelength and frequency of the free radicals indicative of OH may be monitored and compared to the index for OH. During a seasoning cycle, the OH level in the chamber 100 is reduced by processing a seasoning wafer.
[0026] Optionally, the associated free radicals may be graphed, for example, OH concentration, providing a graphical representation of the decrease in OH concentration in chamber 100. By continuously processing seasoned wafers while monitoring the OH associated free radicals in situ, a customer can process wafers to reduce OH, until OH reaches a minimum level (i.e., OH levels are stable and no longer decreasing rapidly).
[0027] The curve of the OH level decrease when a new part is placed in the chamber 100 is different than the curve of the OH level decrease when a cleaned part is used over the course of seasoning 25 wafers. Figure 3 shows a graph 300 having a curve 302 and a curve 304. Curve 302 shows the decrease in OH level when a new part is placed in the chamber 100, and the decrease in OH level when a cleaned part is used over the course of seasoning 25 wafers. Curve 304 shows the decrease in OH concentration after each seasoning wafer while the cleaned parts are in chamber 100. The steady state (low rate of change or no change) of OH in chamber 100 is different for new parts and used parts.
[0028] When OH is at an acceptable or steady state level, the wavelength and frequency of free radicals indicative of fluorine may be monitored to measure the fluorine level in the chamber 100 due to the parts in the chamber 100. New parts in the chamber 100 will produce higher fluorine levels than used parts that have been cleaned. The use of new parts may result in a faster etch rate as the wafers may be processed, changing the etch rate. The use of cleaned parts may result in lower fluorine levels (compared to the levels produced by new parts) and a slower etch rate.
[0029] Once the fluorine and OH levels are at acceptable levels, the seasoning cycle is terminated. After the seasoning cycle is completed, a dummy wafer is processed for a qualification cycle. The OH levels are monitored in situ, and wafers are processed until the rate of OH reduction is minimal (or no OH is present), greatly reducing the qualification time. For example, seasoning 25 wafers followed by a qualification cycle will result in a shorter qualification time than seasoning 10 wafers followed by a qualification cycle. In addition, the methods disclosed herein allow seasoning and subsequent qualification testing without the need to run a seasoning cycle, running a qualification cycle, then rerunning the seasoning cycle if the wafer fails the first qualification, and then running a second qualification cycle. Additionally, in situ monitoring of associated free radicals can indicate the efficiency when variables such as new or cleaned parts (both minor and major PM) are used, or when using one chamber compared to another.
[0030] Returning to Figure 2, method 200 also includes optional blocks 210 and 212. In optional block 210, an in-situ chamber condition ("ICC") recipe is altered in response to the comparison if the relevant free radicals are not within the target range. For example, the rate at which chemicals are added to the chamber may be altered and the etch trench altered if the relevant free radicals are not within the target range.
[0031] At optional block 212, if the associated free radicals fall below the target range, the chamber 100 may be removed from the production cycle for preventative maintenance for servicing, seasoning, and qualification testing, and if the associated free radicals are within the target range, the chamber 100 may be returned to the production cycle.
[0032] The average time between cleanings may also be increased using embodiments disclosed herein. Additionally, in situ monitoring of free radical wavelengths and frequencies allows for chemicals to be added as needed. For example, FIG. 4 shows a graphical comparison 400 of in situ monitoring and increasing fluorine levels using embodiments disclosed herein versus the prior art. The graphical comparison 400 includes a target range 402, which indicates an acceptable operating range for fluorine levels. An unacceptable level 404 indicates that the chamber 100 should be taken out of production and placed in preventive maintenance if the fluorine level falls below the unacceptable level 404. Waveform 406 shows that the fluorine has dropped to a level for preventive maintenance and then returned to production. Waveform 408 utilizes materials disclosed herein and includes in situ monitoring of free radical wavelengths and frequencies indicative of fluorine and increasing the fluorine level when the fluorine is approaching the lower limit of the target range. The results of adding fluorine are shown in waveform portion 410. Over time, the fluorine level begins to drop, as shown by waveform portion 412. When the fluorine level again approaches the lower end of the target range, fluorine can again be added to the chamber 100. By in situ monitoring and repeated addition of fluorine to the chamber, the average time between cleanings can be increased. For example, the average time between cleanings can be increased well beyond 600 RF hours without the need for minor or major PMs.
[0033] As used herein, terms such as "having," "containing," "including," "comprising," and the like are open-ended terms indicating the presence of stated elements or features, but not excluding the addition of additional elements or features. The articles "a," "an," and "the" are intended to include plural and singular, unless the context clearly dictates otherwise.
[0034] While the forgoing is directed to embodiments of the present disclosure, other and further embodiments of the present disclosure may be made without departing from the basic scope thereof, which is defined by the following claims.
Claims
[Claim 1] monitoring in situ the frequency and wavelength of the free radicals in the chamber for each wafer in the chamber; Associating the frequencies and wavelengths of the free radicals with at least one selected chemical product; and comparing the associated free radicals to an index, the index including a target range for each chemical in the at least one selected chemical product.
Citation Information
Patent Citations
Plasma treatment method
JP2002110642A
Plasma processing apparatus and method therefor
JP2007073751A
Plasma processing method and apparatus
JP2007324341A
Etching rate measuring apparatus and method
JP2008244107A
Plasma processing method, plasma processing apparatus, and moisture content detecting method of plasma processing apparatus
JP2010147052A
