Method and device for cleaning semiconductor devices and carriers for semiconductor devices
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- BROOKS AUTOMATION GERMANY
- Filing Date
- 2025-11-06
- Publication Date
- 2026-05-15
Smart Images

Figure EP2025082182_15052026_PF_FP_ABST
Abstract
Description
[0001] Method and device for cleaning semiconductor devices and carriers for semiconductor devices
[0002] Technical Field
[0003] The invention relates to cleaning methods and devices for cleaning items used in the field of semiconductor manufacturing, such as wafers or reticles, or carriers or pods for storage or transport of wafers or reticles. This invention especially relates to methods and systems for cleaning wafer carriers, specifically Front Opening Unified Pods (FOUPs), used in semiconductor manufacturing. The invention addresses the challenges associated with traditional FOUP cleaning methods, which rely on fixed cleaning recipes and struggle to effectively handle varying types and levels of contamination, particularly Airborne Molecular Contamination (AMC).
[0004] Background
[0005] Consistent contamination control and cleaning quality are critical in the field of semiconductor manufacturing plants, so called fabs. Traditional cleaning tools make use of fixed cleaning sequences, which can lead to inefficiencies due to widely varying contamination levels. Pods with high levels of contamination may not be adequately cleaned, while lightly contaminated pods may undergo unnecessarily long cleaning cycles, reducing overall system efficiency and throughput.
[0006] Cleaning and treatment of surfaces of all kinds is an important part of semiconductor fabrication processes and there are many different tools and cleaning methods in the semiconductor industry used to achieve this. As semiconductor device size, i.e. especially the size of circuits and transistors on semiconductor devices, decreases and the number of processing steps increases, there is an increasing necessity for the optimization and development of cleaning approaches. Cleaning of carrier pods (e.g. FOUP, FOSB, EUV, etc.), and their inner surfaces is especially important as wafers, reticles, masks and such devices spend long time periods inside these pods between processing steps, and carrier pods as well as wafers, mask and reticles can be a source of contaminant accumulation and carry a cross-contamination risk. Cleaning and modification of these devices, their constituent materials, and surfaces is required for removing particles and organic, metallic, molecular, and other contaminants, which can be achieved by a variety of methods. Traditional FOUP cleaning systems rely on a fixed, one-size-fits-all cleaning recipe. While effective for general cleaning, these systems struggle with varying contamination types and levels in different FOUPs, leading to suboptimal cleaning results and the potential for residual contamination. Especially Airborne Molecular Contamination (AMC) has become a growing concern due to its ability to persist after standard cleaning processes. Residual AMC can dramatically impact the performance and yield of semiconductor devices. As semiconductor devices evolve and processes become more advanced, the need for an adaptable and precise cleaning method that can be tailored to specific contamination levels is critical.
[0007] Traditional FOUR cleaning methods are limited by their reliance on a single, static cleaning recipe, which does not account for the varying contamination levels or types across different FOUPs. This results, for example, in issues such as suboptimal cleaning, as fixed cleaning recipes may not efficiently remove all forms of contamination, especially AMC, which can be particularly challenging to eliminate, and residual contamination, as inadequate cleaning can leave residual AMC, leading to contamination risks that negatively impact product yield and quality.
[0008] The need for specialized cleaning techniques depending on contaminant type is evident as process complexity grows and new materials and processes are introduced.
[0009] A variety of cleaning tools exist in the market in the semiconductor industry for carriers and pods. In these tools there are three primary cleaning steps, which are performed: Wet cleaning and drying, exposure to vacuum, and purging with an expedient purge gas such as clean dry air CDA, nitrogen or argon (these steps will be referred to in the following as the main cleaning process). Typically, these steps occur in different units, stations or chambers within the cleaning tool, namely in a cleaning and drying chamber, a vacuum chamber, and a purge station respectively.
[0010] The main cleaning process can be summarized as follows: In the wet cleaning step, the semiconductor carriers are sprayed with or immersed in liquid cleaning media, which can contain a mix of chemicals or ultrapure water, to remove contaminants like organic matter, particles, and metal ions from the surface. Subsequently, the semiconductor carriers are dried with air spray and throughout this step they may be heated to improve contaminant removal. After this wet cleaning and drying process, the semiconductor carriers are placed in a vacuum chamber. The vacuum helps to remove any remaining liquid and molecular contaminants by reducing pressure, which encourages the evaporation and outgassing of contaminants and other residual substances. Finally, an inert purge gas (such as CDA, nitrogen or argon) is introduced to further clean the semiconductor carrier. The purge gas helps to remove any remaining moisture and contaminants, ensuring the mini-environment provided by the semiconductor carrier is filled with pure and dry gas, making it ready for further processing.
[0011] With increasing cleanliness demands and sensitivity of fabrication processes to new contaminant types, the main cleaning process as described above is not always sufficient to achieve the desired or required cleaning results. Carriers must be cleaned sufficiently often in order to guarantee their continued cleanliness, and cleaned to stringent specifications to maintain high yields in manufacturing lines. Cleaning performance and throughput are two of the most important factors in the cleaning process. Additionally, the cost of cleaning each carrier can be of concern and is primarily affected by the cleaning tool’s consumption of power, cleaning media (e g. ultrapure water, chemicals), compressed air, and purge gas. Improvements to the efficiency of the main cleaning process can be beneficial to improve cleaning performance as well as reducing cost of ownership of cleaning tools, while also maintaining or improving throughput.
[0012] To address these issues, according to a first and second aspect, the present invention provides a method comprising the features of claim 1 and a device comprising the features of claim 9. To address these issues, according to a third and fourth aspect, the present invention further provides a method comprising the features of claim 12 and a device comprising the features of claim 16.
[0013] Embodiments of the invention are the subject matter of the dependent claims.
[0014] The following paragraphs especially apply to the first and second aspects of the invention, but can also be applied to the third and fourth aspects. In addition to the main treatment process, performing a pre-treatment process and / or a post-treatment process according to the invention in case of a contamination level as measured exceeding or not exceeding predetermined contamination levels, effectively increases cleaning efficiency, while maintaining or increasing throughput, thereby for example reducing the cost of ownership. Pre / post treatment processes can provide a specialized cleaning process based on contaminant types and improve the overall effectiveness of cleaning tools. Items such as semiconductor carriers to be cleaned, for example FOUPs, can benefit from adding various cleaning treatment processes that cannot be implemented in the main cleaning process without reducing throughput or increasing cost and complexity. If, for example, a contamination measurement is performed before the main cleaning process, it can, depending on the type and level of contamination measured, be determined that a further contamination measurement after the main cleaning process is necessary / expedient or not. It is also possible to perform a contamination measurement only after the main cleaning process or to perform a contamination measurement before and after the main cleaning process, and subsequently implement or adapt further cleaning steps or processes based on these measurements
[0015] According to an embodiment, the item configured and adapted for use in the field of semiconductor manufacturing is a wafer or a reticle, or a container, especially a carrier and / or storage pod for wafers or reticles, such as a FOIIP, a FOSB, an EUV double pod, or a surface of any such item. The description mainly refers to the cleaning of containers, especially FOUPs. Be it noted that the concept of the invention may also be applied to wafers or reticles, wherein, however, the specific cleaning and treatment steps would typically have to be adapted, for example taking into account that they are made from different materials than containers / camers such as FOUPs.
[0016] According to an embodiment, the contamination measurement is performed before the main cleaning process, and in case the contamination as measured exceeds a predetermined level, the pre-treatment process, also referred to as pre-treatment in the following, is performed. This measurement is usually referred to as pre-measurement in the following. According to an embodiment, a contamination measurement is performed after the main cleaning process, and in case the contamination as measured exceeds a predetermined level, the post-treatment process, also referred to as post-treatment in the following, is performed. This measurement is usually referred to as post-measurement in the following.
[0017] According to an embodiment, the contamination measurement comprises at least one of a measurement of organic contamination, a measurement of airborne molecular contamination AMC and a measurement of particle contamination. These are the predominant types of contamination observed in the semiconductor manufacturing environment.
[0018] According to an embodiment, the main cleaning process comprises a wet cleaning treatment, preferably including a subsequent drying treatment, a vacuum treatment, in which the item to be cleaned is subjected to a vacuum, and a purging treatment, in which the item to be cleaned is subjected to a purge gas.
[0019] According to an embodiment, the pre-treatment process comprises at least one of a heating process, a cooling process, a steam treatment process, a gas treatment process, a purging treatment process, a plasma and / or ultraviolet and ozone treatment process (UVO treatment process), a snow-jet treatment process and an electrostatic discharge treatment process.
[0020] According to a preferred embodiment, the post-treatment comprises at least one of a heating process, a cooling process, a steam treatment process, a gas treatment process, a purging treatment process, a plasma and / or UVO treatment process, a snow-jet treatment process and an electrostatic discharge treatment process.
[0021] Pre-treatment and / or post treatment can comprise one or more of a wide range of processes which can selectively target different types and levels of contamination. There are some technologies or methods that, for various reasons, can or should not be implemented in the main cleaning process as defined by some cleaning tools in the semiconductor industry. However, providing such additional processes in the context of pre-treatment and / or post-treatment, which are only performed in case certain contamination conditions are met, increases the cleaning efficiency of the main cleaning process which follows the pre-treatment and / or precedes the post-treatment. Pretreatment can especially be provided in a separate pre-treatment station or unit before the main cleaning process so as not to reduce the throughput of the entire process. The pre-treatment processes / steps mentioned above and further expanded on in the following are advantageous, especially in that they can be individually chosen and / or adapted to deal with specific contamination situations. Pre-measurement of contamination before cleaning can identify the contamination levels of particles, volatile organics, acids, amines, and other common types of contaminants found in the semiconductor manufacturing environment, for example in wafer or reticle carriers. Pre / Post treatment processes can be selected based on pre-cleaning measurement and / or post-cleaning measurement of contamination levels and targeted at specific types of contamination. One or more pre-treatment or post treatment units could be specially adapted to perform the following pre-treatment or post-treatment processes:
[0022] - a heating step, especially comprising heating the items to be cleaned. An example of such a heating step includes use of hot purge gas introduced into a FOIIP before the main cleaning process. Simulations have shown increased efficiency of molecular contamination removal from FOUPs when using a hot purge gas over use of a cold purge gas. Combined heating and creation of concentration gradients by introducing hot purge gas increases the rate of diffusion of contaminants to the surface of carrier materials, so that they can be more efficiently removed in a subsequent main cleaning process. Thus, a pretreatment step including heating carrier pods with hot gas spray, radiant heat sources, and / or hot purge gas can increase the efficiency of contaminant removal and can improve removal rate of the combined pre-treatment and main cleaning process. Another example of the benefits of heating during pretreatment before the wet cleaning of the main cleaning process is that the heat helps with weakening the bonding of particles to the surface of the FOIIP, resulting in more efficient particle removal during the main cleaning process, i.e. during the wet cleaning and drying steps. The heating process can occur by use of hot purge gas, by spraying hot air onto the surface of the item to be cleaned, or by a radiation heat source, or a combination of such heat sources. - an electrostatic neutralization step as part of the pre-treatment comprises for example exposing the surfaces of items to be cleaned to positively and / or negatively ionized air. Static forces between charged particles and charged surfaces can be dissipated by such a pre-treatment with ionized air, reducing the Coulomb forces acting between particles and the surface . Advantageously, the purge gas supply mechanisms of cleaning devices for FOUPs or reticle pods can be used to introduce ionized air to neutralize charge on the surface of items to be cleaned. With reduced bonding forces, the particle removal efficiency of the main cleaning process is significantly improved.
[0023] - an UVO (combination of UV light and oxygen / ozone) treatment process as part of the pre-treatment comprises for example exposing the items to be cleaned to a combination of UV light sources and oxygen, forming ozone and further oxygen radicals. The combination of ozone and UV light may be effectively used to remove organic contamination and contamination from residual photoresist originating from wafers or reticles from items to be cleaned. Cross-contamination of items to be cleaned, such as FOUPS and reticle pods, due to resist and organics from wafers can be difficult to remove in the main cleaning process and requires special attention. UVO treatment can reduce organic contaminants into by-products such as H20, CO2, and NOx, which can be more easily and efficiently removed during the main cleaning process. Additionally, inorganic contaminants can be converted into highly oxidized states which can then be easily removed by rinsing with ultrapure water in the main cleaning process. UV- ozone processing has been shown to remove organic contaminants from common thermoplastics which carriers and pods are typically made from. Including an UV-ozone pre-treatment step can thus be a tool to improve the effectiveness of cleaning carriers and pods of organics, resist, and other contamination in combination with the main cleaning process.
[0024] - a plasma treatment as part of the pre-treatment may also be advantageous. Plasma cleaning of contaminants involves both physical and chemical reactions with contaminants. The physical mechanisms involve plasma ions and atoms colliding with contaminants and transferring their energy, which can break bonds and decompose them into by-products or detach contaminants from the surfaces of the items to be cleaned. Chemically, plasma can react with contaminants and / or oxidize them, forming functional groups, which can further be transformed into CO2, H20, and other by-products which can be more easily removed in the main cleaning process. Choice of expedient gases (in plasma different gases can be used) can be adapted for the removal of different types of contamination. Plasma can effectively remove most types of organic contamination and can uniformly cover the item to be cleaned, as well as penetrate small pores, which are not easily accessible using traditional cleaning methods. Depending on the gases chosen as a basis for creating a plasma, different reactions during the plasma process will take place, and different by-products will be produced. Such by-products can be more efficiently removed during the subsequent main cleaning process.
[0025] - a snow jet treatment as part of the pre-treatment may also be advantageously implemented, wherein the items to be cleaned are exposed to a CO2 snow jet. A CO2 snow jet is a cleaning tool that uses solid carbon dioxide (CO2) in the form of snow or pellets to remove contaminants from surfaces. The process involves blasting a jet of CO2 snow or pellets onto the surface to be cleaned. When the solid CO2 comes into contact with the surface, it rapidly sublimates (turns from solid directly into gas), causing a cleaning effect through a combination of physical impact and thermal shock. This can reduce particles or metal contamination without damaging the device or item. Snow cleaning can be particularly effective at removing sub-micron particles, which can be difficult to remove during the main cleaning process. Additionally, it has been demonstrated that wet cleaning combined with CO2 snow pre-treatment results in higher efficiency of removal of residual photoresist.
[0026] - a steam treatment process comprising steam treating of the items to be cleaned can also be advantageously implemented as part of the pre-treatment. Steam can affect contamination removal in different ways. For example, high temperature and pressure of steam will help with the removal of particles. High- velocity, fine steam droplets can penetrate pores and bring better efficiency in contamination removal. Also, some organic residuals could break down and dissolve better under or following steam treatment. Additionally, steam from ultrapure water would condense on all surfaces and leach molecular contaminants from the bulk material.
[0027] The pre-treatment may also comprise a measurement or detection process. While this does not directly affect the contamination, such a detection step would measure the contamination and particle level by using a smart tool, such that subsequent cleaning procedures, either within the pre-treatment process or the subsequent main cleaning process or the post treatment process, could be adapted and optimised.
[0028] It is especially possible to incorporate a plurality of these pre-treatment processes or steps, especially performed in corresponding pre-treatment stations, in the cleaning process to enhance contamination removal.
[0029] After the main cleaning process, which especially comprises at least one of a wet cleaning treatment, preferably including a subsequent drying treatment, a vacuum treatment, in which the item to be cleaned is subjected to a vacuum, and a purging treatment, in which the item to be cleaned is subjected to a purging gas, an additional post treatment station can be provided, in which one or more post treatment steps are performed, which may further enhance the removal of contamination and / or prolong Q time. Post-treatment steps may include a wide range of processes. While all steps discussed above in connection with pre-treatment are also advantageously implemented in post treatment, the following steps are especially advantageous in connection with post-treatment:
[0030] The post-treatment processes / steps listed in the following are advantageous, especially in that they can be individually chosen and / or adapted to deal with specific contamination situations. Other post-treatment processes, although not explicitly mentioned, could also be used. One or more post-treatment stations could especially be adapted to perform at least one the following post-treatment processes:
[0031] - a cooling process with cold purge gas to be performed to cool the temperature of the item to be cleaned. The main cleaning process heats the item to an elevated temperature which can take an hour or more to cool down to an acceptable temperature for further processing. However, the device is in the cleanest state immediately after the cleaning process, as diffusion and outgassing of contaminants from the constituent material occurs over time after cleaning. A cooling process as a post-treatment would allow the device to be introduced into the manufacturing line faster and in a cleaner state and thus improve the effective cleaning efficiency of the process. The cooling effect can be achieved by using the purge function of the cleaning system and the items to be cleaned, for example FOUPs, and introducing a cold purge gas. This would also have the effect of slowing the diffusion and outgassing of contaminants from the material.
[0032] - an electrostatic discharge as discussed above in connection with pre-treatment may also be used for post-treatment, whereby electrostatic attraction of contaminations to the surface will be reduced.
[0033] - a plasma treatment process as discussed above in connection with pre-treatment may also be used for post-treatment. Advantageously, choice of an expedient gas, which serves as basis for the plasma, will not result in any by-product and will help with the removal of remaining.
[0034] - a gas treatment process, especially utilizing a noble gas such as Argon, may advantageously be used in post-treatment. By subjecting the items to be cleaned to Argon, cooling can be facilitated, and surface cleaning by sputtering (for example by ionization of Argon and accelerating the ions such that they impinge on the surface, which has a negative charge that knocks out the contamination atoms on the surface) may be affected, and contamination removal in general can be improved.
[0035] A measurement or detection process may also be implemented in connection with posttreatment. While such a post-treatment step does not directly affect the contamination, a detection step could measure the contamination and particle level by using a smart tool, such that it can then be determined whether or not the device / item should be resubjected to any of the previously mentioned cleaning processes. On the other hand, by using such a measurement process it can reliably be ascertained that an achieved level of contamination is acceptable or sufficient. It is possible to implement multiple pre-treatment and / or post-treatment steps in the cleaning process to enhance contamination removal. It is conceivable to use the same treatment unit as pre-treatment unit and post-treatment unit.
[0036] Providing at least one pre-treatment step and / or at least one post-treatment step in connection with current cleaning methods is advantageous to increase the removal efficiency of contamination (e.g. AMC, metal, particle, etc.) and throughput, and may also reduce cost and consumptions of the main cleaning process.
[0037] The following paragraphs especially apply to the third and fourth aspects of the invention, but can also be applied to the first and second aspects previously discussed.
[0038] The third and fourth aspects, as well as the first and second aspects, especially address the problem of inconsistent results in cleaning procedures, which are oftentimes due to a lack of adaptability in the cleaning process. According to the invention, it is possible to increase variability and reduce the overall efficiency of the cleaning system.
[0039] The invention, especially as embodied in the third and fourth aspects, provides a flexible, adaptive method that optimizes FOUR cleaning based on real-time contamination data.
[0040] The method may advantageously take into account historical or empirical contamination data, especially using an optimization module that employs historical contamination data for run-to-run cleaning performance improvement.
[0041] According to an embodiment, the adjusting of the at least one cleaning parameter is performed in real time.
[0042] According to an embodiment, the initial cleaning procedure comprises cleaning based on a default or initial cleaning recipe.
[0043] The plurality of cleaning parameters may advantageously include one or more of the following: cleaning duration, chemical concentration, cleaning temperature, and cleaning pressure. The cleaning system may comprise a cleaning module, a contamination analysis module and an adaptive feedback module.
[0044] By utilizing real-time contamination data and adaptive cleaning processes, the invention provides a solution for optimizing FOUR cleaning, reducing residual contamination, and improving overall cleaning efficiency and consistency in semiconductor production.
[0045] The invention provides a solution for optimizing FOUR cleaning by utilizing real-time contamination data and adaptive cleaning processes, which reduces residual contamination and improves overall cleaning efficiency.
[0046] 1. A typical operation flow for a FOIIP implementing the invention is as follows: Firstly, the FOIIP is cleaned using an initial or default cleaning recipe within the FOIIP cleaning system. Then, the residual contamination (AMC) after cleaning is measured using an AMC analyzer. In response to this measurement, the cleaning process is adapted, i.e. based on feedback from the AMC analysis. This measurement can detect various forms of AMC, allowing the cleaning process to be assessed with precision. As a result, the cleaning recipes may be optimized dynamically using the data thus obtained as “historical data” to improve future cleaning runs. In other words, contamination level data gathered from the AMC analyzer are fed back into the cleaning system. Based on this data, the cleaning system adjusts its parameters, such as cleaning duration, pressure, temperature, or chemical concentration, in real-time to address any residual contamination identified. The system can track cleaning performance over multiple batches (runs), using historical data to optimize future cleaning cycles. The system can predict the contamination levels of incoming FOUPs based on prior results, continuously improving the cleaning recipe and process efficiency. This and any other type of information / data may be continuously collected in order to complete a full feedback loop.
[0047] By adjusting the cleaning process dynamically based on real-time contamination data, the system can ensure more thorough cleaning, reducing the risk of residual AMC contamination. The use of run-to-run optimization helps minimize the variability in cleaning performance, leading to more consistent results. The improved cleaning process reduces the likelihood of contamination-related defects in semiconductor manufacturing, thereby improving product yield and quality. By avoiding unnecessary rework due to suboptimal cleaning and reducing contamination-related failures, this method can lower operational costs and enhance overall manufacturing efficiency. The dynamic nature of this solution allows it to be scaled to accommodate varying contamination types, process demands, and specific FOUR cleaning requirements, making it adaptable to both current and future semiconductor manufacturing needs.
[0048] The primary application of this technology is in semiconductor wafer production, where FOUR cleanliness is critical to maintaining high-quality yields. The method according to the invention can also be applied in advanced packaging processes, where contamination control is equally crucial. Any cleanroom or controlled environment that requires precise cleaning of carriers or pods for sensitive materials can benefit from the invention.
[0049] The invention will now further be explained in connection with the appended Figures. Herein
[0050] Figure 1 shows a flow diagram illustrating an advantageous embodiment of the present invention,
[0051] Figure 2 shows a more detailed flow diagram further illustrating the embodiment of Figure 1 ,
[0052] Figure 3 shows a schematical view of an advantageous embodiment of a device according to the invention,
[0053] Figure 4 shows a schematical view of an advantageous embodiment of a pre-treatment station or a post-treatment station as included in the device of Figure 3, and
[0054] Figure 5 shows an example of an advantageous combination of a specific pre-treatment process with a main cleaning process. Figure 6 shows a schematic diagram illustrating a preferred embodiment of the invention in connection with the third and fourth aspects of the invention.
[0055] Be it noted that the Figures will deal with a FOUR as a specific item to be cleaned. The invention, especially as illustrated in the Figures, is applicable to any item configured and adapted for use in the field of semiconductor manufacturing, such as a wafer or a reticle, or a container, especially a carrier and / or storage pod for wafers or reticles, such as a FOUR, a FOSB, an EUV double pod, or a surface of any such item.
[0056] An advantageous embodiment of the cleaning method according to the invention is shown in Figure 1. Herein, a FOIIP 100 is contaminated, for example coming from a fab, and is subjected to a pre-cleaning measurement to measure contamination levels of at least one contaminant, referred to as pre-measurement in the following, in step 110. In case the FOIIP 100 exhibits a predetermined level of contamination, it is subjected to a pre-treatment ( in step 110a), after which the method proceeds to step 120, in which a wet cleaning and a drying of the FOIIP are performed. In case the level of contamination as measured in step 110 is below the predetermined level of contamination, the method proceeds directly from step 110 to step 120, in which a wet cleaning and a drying of the FOIIP are performed. In a subsequent step 130, the FOIIP is then subjected to a vacuum treatment followed by a purge gas treatment in step 140. Steps 120,130 and 140 constitute the main cleaning process as discussed above.
[0057] In a subsequent step 150, the FOIIP is subjected to a post-cleaning measurement, similar to the pre-measurement as performed in step 110, and referred to as postmeasurement in the following. Alternatively, as indicated by the line and arrow 140a, it can be determined, in connection with step 140, i.e. without post-measurement, for a FOIIP 100 that has been subjected to a pre-treatment process and a main cleaning process, or has been subjected only to a main cleaning process, that no further treatment is necessary. In this case, the method proceeds from step 140 to step 160, which means that the FOIIP is deemed to be sufficiently clean and in a state for further use within the fab environment. If it is determined in step 140 that the contamination of the FOIIP needs to be measured, the method continues to step 150. In case it is determined in step 150 that the contamination of the FOIIP is above a predetermined level, the FOIIP is subjected to a post-treatment process in step 150a, after which it may be is deemed to be sufficiently clean and in a state for further use within the fab environment, so that the method proceeds to step 160. If the post measurement in step 150 shows an acceptable level of contamination, the method directly proceeds to step 160, so that the FOUR can enter the fab environment. Thus, FOUPs will only be subjected to pre-treatment and / or post-treatment processes (steps 110a, 150a), in case their level of contamination lies above certain predetermined values, whereby throughput of a cleaning device may be enhanced, as not all FOUPs are subjected to an enhanced cleaning process comprising pre- or post-treatment. At the same time, it can be ensured that even FOUPs which exhibit high levels of contamination are sufficiently cleaned. It is to be noted that pre-treatment and post-treatment can be performed in a combined pre- and post-treatment station. Also, separate stations for pre- and post-treatment may be provided.
[0058] Further details of this embodiment of the invention are shown in Figure 2. Figure 2 shows only FOUP 100 and steps 110, 110a, 120, 130, 140 and 160 as already described in connection with Figure 1 . Step 110 is shown in greater detail as comprising sub-steps 111 to 113.. Step 110a is shown in greater detail as comprising sub-steps 114 to 118.
[0059] FOUP 100 to be cleaned is, in step 110, subjected to a pre-measurement. This premeasurement comprises at least one of a measurement step 111 of organic contaminants, an AMC measurement step 112 and a particle contamination measurement step 113. Depending on the types and levels of contamination measured in step 110, pre-treatment step 110a is performed, comprising steps 114 to 118. If a level of organic contaminants as measured is above a predetermined level, an UVO treatment (step 114) and / or a plasma treatment (step 115) are performed. If the level of AMC as measured is above a predetermined level, a steam treatment (step 116) and / or a preheating treatment (step 117) are performed. If a particle level as measured is above a predetermined level, an electrostatic treatment (step 118) is performed. All premeasurement steps 111 ,112,113 may be performed simultaneously before the corresponding pre-treatment steps 114 to 118 are performed. It is also possible to perform one pre-measurement step at a time, and then perform a corresponding pretreatment step, before performing a second pre-measurement step etc. When all pretreatment steps as determined to be feasible or necessary have been performed, the method proceeds to the main cleaning process (steps 120,130 and 140). In case of all pre-measurements ascertaining contamination levels below the various predetermined values, the method immediately proceeds to the main cleaning process (steps 120, 130 and 140), as indicated by arrow 119. The main cleaning process comprises, as mentioned previously, a wet cleaning and a drying step (step 120), a vacuum step (step 130) and a purge step (140). Subsequently, in step 160, the FOUR is deemed to be sufficiently clean and in a state for further use within the fab environment. As described above in connection with Figure 1 , it is possible to perform a post-measurement before proceeding to step 160.
[0060] A combined cleaning and inspection system 400 for a container such as a FOUR 100 according to a preferred embodiment of the invention is shown in Figure 3 in a purely schematic manner. The system 400 comprises a housing 401 including a cleaning unit
[0061] 410 comprising a main cleaning unit 412, especially adapted to perform steps
[0062] 120, 130, 140 as discussed in connection with Figures 1 and 2, a pre-treatment unit 411 , especially adapted to perform steps 110, 110a, 111 , 112 and 113 as discussed in connection with Figures 1 and 2, and a post-treatment unit 413, especially adapted to perform steps 150, 150a as discussed in connection with Figures 1 and 2, as well as an inspection unit 430.
[0063] The system 400 comprises at least one load port 402, configured to introduce a container, such as FOIIP 100, which is to be cleaned, into the cleaning system, an Equipment Front End Module EFEM 404 adapted to cooperate with the at least one load port 402, a gate opening section 406 connected to the EFEM 404, and a gate closing section 409 connected to the EFEM 404. The system can comprise a buffer station 408 between the gate opening section 406 and the cleaning unit 410. The buffer station 408 is adapted to transport the FOIIP 100 into the cleaning unit 410 . The system is provided with at least one control unit 440.
[0064] Pre-treatment measurement in pre-treatment unit 411 is adapted to measure contamination of FOIIP 100, especially to analyze air or gas within the FOIIP 100 while it is entering in the main chambers of unit 412 412 . Pre-treatment measurement unit
[0065] 411 is, for example, adapted to measure at least one of organic contamination, AMC, particle contamination and humidity. Post-treatment measurement at post treatment unit 413 is adapted to measure contamination of FOUR 100, especially to analyze air or gas within the after it exits the main chambers of unit 412. Post-treatment measurement at unit 413 can, for example, be adapted to measure at least one of organic contamination, AMC, particle contamination and humidity.
[0066] The inspection unit 430 is adapted to detect at least one of a plurality of FOIIP characteristics, such that if a FOIIP is repeatedly placed within the inspection unit, for example at regular intervals during its lifetime, or after each or a specified number of usages within a fab environment, a development of these characteristics over time can be monitored. Such a monitoring of container or FOIIP characteristics over time provides valuable information regarding the state of the FOIIP, and whether it can be used further, requires maintenance, or must be replaced, for example. The inspection unit 430 can be provided, for example, with an imaging unit for imaging of at least part of the FOIIP, for example a gasket, a pressure detection unit for monitoring a pressure within the FOIIP following a pressurization or a de-pressurization, a detection unit for monitoring a heat reaction following an exposure of the container to a heat source, and / or a humidity detection unit for monitoring humidity within the FOIIP.
[0067] According to a preferred embodiment of the invention, a plurality of sensors 420 can be positioned throughout the system 400, especially in at least one of the load port 402, the EFEM 404, the gate opening section 406, the buffer station 408, the cleaning unit 410, the inspection unit 430, or the gate closing section 409. Only some of these sensors are schematically shown in Figure 3 . These sensors can be adapted to provide various data for monitoring FOIIP characteristics.
[0068] Typical paths of a FOIIP within such a cleaning system 400 are indicated by arrows in Figure 3. Especially referring to the embodiments of the method according to the invention as outlined above with reference to Figures 1 and 2, a FOIIP 100 to be cleaned, after entering the system via load port 402, EFEM 404, gate opening station 406 and buffer station 408 is first subjected to a pre-measurement in pre-measurement unit 411 . Depending on contamination values measured, the FOIIP is then subjected to cleaning process, possibly preceded by a pre-treatment in pre-treatment unit 411 and / or followed by a post-treatment unit in post-treatment unit 413. Before being subjected to a post-treatment, it is, as outlined above, possible to perform a post-treatment measurement in post-treatment measurement unit 413.
[0069] A schematical view of an advantageous embodiment of a pre-treatment station or unit 422 or a post-treatment station or unit 423 as included in the device according to the invention is shown in Figure 4. For ease of reference, it will be assumed in the following that Figure 4 shows a pre-treatment unit 422. A post-treatment unit may be provided with at least one of the features as shown in Figure 3. Pre-treatment unit 422 is adapted to pre-treat a FOUR 100, as described above.
[0070] Pre-treatment unit 422 comprises a sealable chamber 500. A UVO / Plasma generator unit 510 is provided to generate UVO and / or plasma and transport it into chamber 500. A radiative heating element 520 for performing a pre-heating treatment is provided within chamber 500. A purge interface, by means of which a purge gas can be supplied to chamber 500 and / or FOIIP 100 is schematically shown and designated 530. The purge interface is connected to a media supply unit 540. The media supply unit 540 for example is adapted to ionise a cleaning medium 580 in an ionising unit 542. The ionised medium may then by heated in a heating unit 544 to achieve electrostatic heating or cooled in a cooling unit 546, before being transported to the chamber 500 via a three- way-valve 548 and purge interface 530. After interacting with FOIIP 100, cleaning medium 580 leaves chamber 500 via exhaust interface 532. An additional measurement unit 550 for measuring contamination of medium 580 after being exhausted from chamber 500 may be provided in media supply unit 540. Furthermore, an articulated control arm 590 equipped with a nozzle 595 configured and adapted to spray a medium such as a snow jet, steam, air or other cleaning media onto FOIIP 100 within chamber 500 is provided.
[0071] Figure 5, comprising Figures 5a to 5d, shows a typical example to illustrate how pretreatment performed on a FOIIP within a pre-treatment unit 422 facilitates the main cleaning of the FOIIP within the main cleaning process: The situation shown in Figure 5a is that the inside surface of FOIIP 100 arranged within chamber 500 is contaminated with electrostatically charged contamination 610. Due to its charge, which in the example shown is positive, this contamination adheres relatively strongly to the inside surface of the FOUR, which exhibits a negative charge. By means of introducing ionised air 680 via purge interface or nozzle 530, which is subsequently exhausted from chamber 500 via exhaust interface 532, the electrostatic charges as mentioned can be effectively neutralised, as illustrated in Figure 5b.
[0072] Figures 5c and 5d serve to illustrate the difference in effectiveness achievable by a subsequent main cleaning process, for example performed by a wet cleaning step as symbolised by wet cleaning feature 620. In case of a pre-treatment with ionised air as shown in Figure 5a, removal of contamination of contamination 610 is relatively ineffective, as illustrated in Figure 5c. It is substantially more effective in the case of pretreatment with ionised air, as shown in Figure 5d.
[0073] To summarize, the invention provides a continuous-flow cleaning system designed for industrial use. Items, such as for example FOUPs, can continuously or intermittently be fed into the system and undergo an automated inspection process to determine contamination levels and types. Based on this assessment, each item may be dynamically routed. For example, items identified in the pre-measurement as excessively contaminated may be diverted into a dedicated pre-cleaning unit, where they undergo targeted pre-treatment, as discussed above. Items showing light or acceptable levels of contamination can be sent directly to the main cleaning unit, bypassing the precleaning unit.
[0074] After the main cleaning process, a secondary inspection or post-measurement may be performed. If items are still insufficiently clean, they are automatically diverted to a posttreatment unit for further processing before exiting the system.
[0075] The system according to the invention significantly increases operational throughput, ensures a consistent standard of cleanliness, and minimizes unnecessary resource consumption by tailoring the cleaning process to actual contamination conditions.
[0076] A centralized control system may be provided to manage the detection or measurement of contamination, routing, and cleaning operations, and can adapt parameters based on operational data or historical contamination patterns. The third and fourth aspects of the invention will now be further explained especially with reference to Figure 6. Figure 6 shows a preferred embodiment of a method for controlling a cleaning process according to the invention, performed by a cleaning system adapted to perform the method.
[0077] As can be seen from Figure 6, a used and thus potentially contaminated FOUR 100 is subjected to a first (default) cleaning procedure in a step 720 using a cleaner tool or cleaning module. Subsequently to this step 720, a measurement of residual contamination, especially airborne molecular contamination, in or on the FOUR, especially an airborne molecular contamination (AMC) measurement is performed on the FOIIP using an analyzer or contamination analysis module, especially an AMC analyzer, in step 740. If it is subsequently determined in a step 760 that the FOIIP 100 meets a predetermined cleanliness level, for example a cleanliness target set by a customer, the cleaning of the FOIIP 100 is deemed completed (step 780).
[0078] If this is not the case, the method branches back to step 720, such that the FOIIP 100 is re-subjected to a further cleaning procedure. This further cleaning procedure may correspond to the first cleaning procedure, as indicated by arrow 765. It may, alternatively, be a modified or adjusted cleaning procedure compared to the first cleaning procedure, the modification being determined in a step 770 based on data obtained in the AMC measurement in step 740, as indicated by arrows 767 and 769. Be it noted that the modification determined in step 770 may additionally or alternatively be based on other data, such as historical data, for example relating to previously cleaned FOUPs. This procedure can then be repeated until the FOIIP 100 meets the predetermined cleanliness level. Be it noted that step 770 may be performed such that at least one of a plurality of cleaning parameters defining the cleaning procedure of the FOIIP performed in step 720 is adapted, especially using an adaptive feedback module, based on contamination analysis data acquired during the contamination analysis.
[0079] Steps 720, 740, 760, 765 and 780 may collectively be referred to as the FOIIP cleaning process. Step 770 constitutes a so called Run-to-Run or R2R process, as will be outlined in the following. Figure 6 depicts an automated process to optimize the recipes using a “R2R” (Run-To- Run) approach. As to the terminology used herein, “Run” means a complete cleaning process, for example comprising steps 720,740 and 760. R2R optimization means that the recipes will be automatically adjusted (in step 770) based on data as discussed above towards the defined cleanliness, throughput and consumption targets.
[0080] For example, historical statistical data show that removing a 30 ppb NH3 (ammonia) to < 1 ppb level, can be achieved with a likelihood of essentially 100% by a default cleaning recipe, e.g. using cleaning substances of a certain composition and / or concentration at a certain pressure and / or temperature.
[0081] Postulating, for example, a cleanliness target of < 1 ppb level, while reducing the consumption of cleaning substances and increasing the throughput as much as possible, the software will automatically in the next process change some process parameters, for example at least one of composition, concentration, pressure and temperature as outlined above, using artificial intelligence in every run.
[0082] After a certain number of runs, statistical data shows that achieving the same cleanliness target 100% of the time can be achieved by using an optimized recipe which saves x% of energy. Then the R2R optimization is completed.
[0083] As to cleaning results, by using such an R2R approach, targeted cleanliness levels with optimized throughput and reduced consumption can be achieved in an effective way. For example, a maximum cleanliness level can be achieved, if the optimization target is set to focus on cleanliness only, herein disregarding or reducing the relevance of consumption and / or throughput.
[0084] The default recipe as referenced above can especially be equated with an initial cleaning recipe with default parameter settings, for example chosen and / or adapted for a for a specific type of cleaner. A type of FOUR type used is typically based on customer requirements, and can vary. Consumption of facilities such as power, water and gas is usually based on AMC measured data and R2R results. The default cleaning recipe will typically be tuned by adjusting recipe parameters, such as lower water flow rate or temperature, which will make the cleaning process more efficient and reduce facility usage requirements. By using the methods according to the invention a substantive amount of data can be acquired, especially from an R2R control, whereby relationships for example between AMC data (TVOC, NH3, HF...) and the parameters of the cleaning recipe can be effectively determined . The adaptive cleaning function can, for example, be implemented or enhanced by analyzing these relationships.
Claims
23Claims1. Method for cleaning an item configured and adapted for use in the field of semiconductor manufacturing, wherein a contamination measurement (110, 150) of the item is performed, and, depending on a contamination level as measured, either only a main cleaning process (120, 130, 140) is performed, or a main cleaning process (120, 130, 140) is performed, preceded by a pretreatment process (110) and / or followed by a post-treatment process (150).
2. Method according to claim 1 , wherein the item configured and adapted for use in the field of semiconductor manufacturing is a wafer or a reticle, or a container, especially a carrier and / or storage pod for wafers or reticles, such as a FOUR (212), a FOSB, an EUV double pod, or a surface of any such item.
3. Method according to claim 1 or 2, wherein the contamination measurement (110) is performed before the main cleaning process (120, 130, 140), and in case the contamination as measured exceeds a predetermined level, the pretreatment process (110) is performed.
4. Method according to any one of the preceding claims, wherein a contamination measurement (150) is performed after the main cleaning process (120,130, 140), and in case the contamination as measured exceeds a predetermined level, the post-treatment process (150) is performed.
5. Method according to any one of the preceding claims, wherein the contamination measurement comprises at least one of a measurement of organic contamination (111 ), a measurement of airborne molecular contamination AMC (112) and a measurement of particle contamination (113).
6. Method according to any one of the preceding claims, wherein the main cleaning process (120, 130, 140) comprises a wet cleaning treatment, preferably including a subsequent drying treatment, a vacuum treatment, in which the item to be cleaned is subjected to a vacuum, and a purging treatment, in which the item to be cleaned is subjected to a purging gas.
7. Method according to any one of the preceding claims, wherein the pre-treatment process (110) comprises at least one of a heating process, a cooling process, a steam treatment process, a gas treatment process, a purging treatment process, a plasma and / or UVO treatment process, a snow-jet treatment process and an electrostatic discharge treatment process.
8. Method according to any one of the preceding claims, wherein the posttreatment (150) comprises at least one of a heating process, a cooling process, a steam treatment process, a gas treatment process, a purging treatment process, a plasma and / or UVO treatment process, a snow-jet treatment process and an electrostatic discharge treatment process.
9. Device for cleaning an item configured and adapted for use in the field semiconductor manufacturing, comprising at least one pre-treatment unit (422) for performing a pre-treatment, a main cleaning unit for performing a main cleaning process, and at least one post-treatment unit (423) for performing a post-treatment process.
10. Device according to claim 9, wherein the item configured and adapted for use in the field of semiconductor manufacturing is a semiconductor device such as wafer or reticle, a container, especially a carrier and / or storage pod such as FOUP, FOSB, EUV double pod, or a surface of any such item.11 . Device according to claim 9 or 10, configured and adapted to perform the method of any one of claims 1 to 8.
12. A method for controlling the cleaning process of a Front Opening Unified Pod (FOUP) in semiconductor manufacturing, comprising:- a) subjecting (720) the FOUP to an initial cleaning procedure defined by a plurality of cleaning parameters, especially using a cleaning module configured to clean a FOUP,- b) subjecting (740) the FOUR to a contamination analysis, especially using a contamination analysis module configured to measure residual contamination, especially airborne molecular contamination, in or on the FOUR,- c) in case it is determined (760) in the contamination analysis that the FOIIP has not reached a predetermined level of cleanliness in the cleaning procedure, re-subjecting (720) the FOIIP to a second cleaning procedure, especially to a modified cleaning procedure, in which- at least one of the plurality of cleaning parameters defining the cleaning of the FOIIP is adapted, especially using an adaptive feedback module, based on contamination analysis data acquired during the contamination analysis, and, if necessary,-repeating the steps b) and c) until the predetermined level of cleanliness has been achieved.
13. Method according to claim 12, further taking into account historical or empirical contamination data, especially using an optimization module that employs historical contamination data for run-to-run cleaning performance improvement.
14. Method according to claim 12 or 13, wherein the adjusting of the at least one cleaning parameter is performed in real time.
15. Method according to any one of the preceding claims 12 to 14, wherein the initial cleaning procedure comprises cleaning based on a default or initial cleaning recipe.
16. Method according to any one of the preceding claims 12 to 15, wherein the plurality of cleaning parameters includes one or more of the following: cleaning duration, chemical concentration, cleaning temperature, and cleaning pressure.
17. Cleaning system configured and adapted to perform the method according to any one of the preceding claims 12 to 16.
18. Cleaning system according to claim 17, comprising a cleaning module, a contamination analysis module and an adaptive feedback module.