by Debra Vogler, senior technical editor, Solid State Technology
July 15, 2008 – ATMI’s exclusive partnership with Intermolecular, which has included a quiet equity investment and then a tool purchase, has now extended further with ATMI’s launch of four “High-Productivity Development Centers” in the US and Asia (see map below) to provide more rapid and efficient methods of integrating new development and manufacturing processes.
Utilizing Intermolecular’s integrated combinatorial R&D technology, ATMI has been able to accelerate the development of new materials and process flows for end users for such applications as: copper post-CMP cleaning, non-selective oxide etching, high-dose implant strip, copper post-etch residue removal, electroless cobalt deposition, and high-k/metal gate cleaning.
According to ATMI, the combinatorial method is used in conjunction with informatics software to design experiments and machine programming that controls and executes multiple simultaneous experiments. “We start with the electrical properties that are required, then we work our way back through the impact of various integration steps, and then we use combinatorial science to screen potential materials candidates,” ATMI’s EVP of process solutions, Tod Higinbotham, told SST.
At the new centers, ATMI conducts experiments designed in collaboration with end users that — by processing as little as a single wafer or coupon — can evaluate up to 192 different test chemistries at once. The company said that its combinatorial screening method allows independent control of formulation stir rate and exposure time at each independent test site on a wafer, enabling data collection in a fraction of the time and cost associated with conventional qualification methods.
For one post-CMP cleaning formulation, ATMI was able to use its high-productivity development tools and methods to conduct 1300 experiments using only 16 wafers in 3.5 weeks — looking at all the possible chemistries to come up with a new clean process with a stable copper surface and low roughness. According to Higinbotham, the same exercise would normally have taken at least 1300 wafers and span approximately six months. “We used ~$100 in chemicals, whereas normally we would have used >$100,000 worth,” he said.
Considering the savings in 300mm patterned wafer alone when comparing traditional materials screening and integration methods vs. the new one, he observed that the efficiency of the process is the attraction. “You’re doing this for many process steps and many chemistries, so you’re talking a major reduction in cost for the end user. But the real benefit is that the solution comes so much faster,” he said. When qualification and final testing begins, end users have a very high probability of meeting the electrical performance requirements because of the new methodology. Higinbotham added that ATMI thinks so highly of the process that it’s invested in multiple platforms (the locations in the US and Asia) — and is Intermolecular’s exclusive, primary materials development partner for the next 2-3 years. — D.V.