In Part One of this article, we discussed the increased materials consumption requirements and new purity challenges required to support the underlying integrated circuit technology of mega-trends like AI, self-driving cars and the Internet of Things.
In Part Two, we will look at some additional materials purity improvement approaches and how the microelectronics industry must commit to these purity best practices as a necessary strategy to help drive ongoing advanced technology innovation.
Understanding Contamination Mapping
The concept of contamination mapping examines the entire fluid system, breaks it down into individual components, and maps out where the particle contaminants come from in the materials used and/or in the processing of them.
If the end chemical has a particle spike, mapping the components can help detect which component is contributing the particles. Once identified, the component supplier conducts a contamination mapping evaluation of their process, including the equipment, the materials, and the processes used to manufacture the component.
If it’s determined the material is contributing the contamination, then the process goes back one step further in the supply chain to the material supplier, who then maps their manufacturing process to determine if the contaminants are in their raw feed materials or in their processing of the polymer. The goal is to take active steps throughout the supply chain to remove contaminant sources.
A key step toward ensuring overall system cleanliness is for end users to partner with component and equipment suppliers that have conducted contamination mapping up the supply chain to understand and lessen contamination sources. Leveraging the contamination analysis capabilities of component suppliers is also key to ensuring consistently clean component delivery to OEMs and fabs.
In addition to evaluating materials of construction that are less prone to particle shedding, there should be continued focus on component configurations and minimizing dead legs and entrapment areas, as well as reducing particles generated by moving parts from valves, pumps, and filters. Computational fluid dynamics analysis is another tool that component manufacturers can use to optimize product designs and improve component-flushing performance.
Reducing Leachable Metallic Contamination
Semiconductor processes, such as photolithography and wet etch and clean, have become more metal sensitive at advanced process nodes. In all fluoropolymer delivery systems, metals leaching from material in components such as PFA tubing, valves, and purifiers impact device yield.
No matter how much time or money is invested in flushing systems to rid them of particle contamination, a much longer-term problem exists that flushing will not resolve.
As chemicals sit in storage containers and long tubing runs, metallic ions are pulled off or leach out of the material and contaminate the fluid stream. So in addition to identifying causes for particle contamination, chemical manufacturers and device makers spend significant time interrogating their entire chemical delivery system to determine the source of those metals.
It can take months for metal contaminants to leach out of materials. Containment vessel manufacturers are evaluating ways to reduce metallic contamination by testing innovative polymers that have lower metal content. Component manufacturers are also working to reduce contamination in tubing and other components by researching ultra-clean PFA materials.
Chemical manufacturers and fabs are both very interested in clean tubing and are engaged in studies with component suppliers to evaluate tubing materials to determine what impact they have on their systems.
It’s important to note that not all suppliers perform metals extraction testing on their products, nor have they optimized their processes to reduce contamination. Regular component analysis is critical to detecting process and material excursions that lead to metallic contamination.
Different PFA grades and how those materials are processed result in dissimilar purity levels. A contaminant that starts in the raw feed can end up in the fluoropolymer, which can then be extruded into the tubing or valve wall, or into the filter membrane, where it then moves through the fluid delivery system—ending up on a wafer and causing a defect. Contamination mapping can be instrumental in finding the contaminant source.
Resin manufacturing starts with a monomer, which may already have its own contamination, and then is put into a polymerization chamber, which, if made of metal, can contribute contamination to the polymer powder. The polymer is then dried and extruded into a pellet, which can also introduce more contaminants into the polymer pellet.
When the component manufacturer converts the pellet into tubing, that extrusion process adds another potential contamination source to the final product. Here again, contamination mapping of the material supply chain can be critical in identifying the metallic contamination source, so immediate corrective action can be taken.
Cleaning process tools and components like photoresist dispense pumps can also reduce the impact of metallic contamination. Photoresist and other lithography solvents do not have a strong affinity to pull out metal contaminants, but they still leach out over time, and chemical interactions between photoresist and metal ions create gel defects and micro-bridged circuits, which cause electrical shorts on wafers.
At advanced logic nodes, these contaminants are detrimental to device yield. The use of solvent purifiers can be especially effective at removing both dissolved and colloidal metal contaminants from a variety of ultrapure, polar, and non-polar solvents used in photoresist applications. Purifiers today remove sub-ppt contaminants and successfully reduce metal-induced defects from raw material manufacturing to wafer dispense.
Making an Industry Commitment to Ensure Materials Purity
Not all material handling components are equal in terms of particles and metals content. Evaluating mechanical components and materials of construction will increase purity levels and help maintain a safe and contamination-controlled chemical delivery environment. It starts with chemical manufacturers around the world working to meet or exceed stringent purity specifications while controlling particle and metallic contamination in containment vessels and distribution systems.
Another key step toward ensuring overall system cleanliness throughout the fluid stream is for end users to commit to partner with component and equipment suppliers that have conducted contamination mapping up the supply chain to understand and lessen contamination sources. Evaluating component cleanliness and sourcing the cleanest chemical delivery, filtration, metrology, and fluid handling systems is an important part of the contamination reduction effort.
The entire semiconductor industry stands to benefit from chemical manufacturers, and equipment and materials suppliers providing contamination-controlled, clean products at each step throughout chemical manufacture to point of use. These new levels of materials purity will be essential to the ongoing adoption of—and public trust in—technologies such as AI and the Internet of Things in our personal and work lives.
The stakes are high, but through industry collaboration and focus on achieving the highest standards of chemical purity, the outcome will include better device performance, increased yield, and reduced costs.
Brett Reichow is Director, Fluid Handling with the Advanced Materials Handling Division at Entegris, a leading advanced sciences and technology company.
