by Chi-I Lang, VP of workflow products and applications, Intermolecular Inc.
June 28, 2011 - Greetings from the American Vacuum Society’s Atomic Level Deposition Conference in Cambridge, MA, where over 400 international attendees have gathered for an in-depth look at ALD and its emergence as an enabling technology not just in microelectronics, but also in nanostructures, energy, lighting, and other applications.
One interesting overall theme so far has been the attention paid to interface engineering. Several presenters have noted that achieving control of the interface between the substrate and the ALD layer is, arguably, more important than the ALD process itself in terms of obtaining the desired characteristics in the finished device. As a result, there has been significant discussion of supporting processes like cleaning and surface treatment.
|Atomic Level Deposition Conference 2011|
|Day 1: Interface engineering, rabbit ears and Roy Gordon|
|Day 2: Manufacturability takes center stage|
|Day 3: Precursor needs, spatial ALD, and butterfly wings|
A Sunday workshop session offered four long-form presentations, including one from A.C. Kummel of the University of California/San Diego on the novel use of in-situ scanning-tunneling electron microscopy for monitoring of ALD gate oxide deposition. He showed very good results, with evidence of the importance of interface engineering for III-V oxide processes. Even on simple precursor and water processes, interface quality has a big effect on electrical performance.
Also notable was a workshop session by Harvard’s Roy Gordon, who offered an excellent high-level survey of ALD chemistry, including precursor materials. He noted that there is a huge range of potential precursors and oxidizers beyond the commonly available choices, and a variety of processing parameters to play with, all of which play into film quality and electrical performance. Gordon is a star at the conference — many of his former students, including conference chair Jill Becker of Cambridge NanoTech, are in attendance — and Gordon was awarded the first ALD Innovation Award in recognition of his groundbreaking work and wide influence on the field.
Monday’s keynote was given by MIT professor Robert Langer, a brilliant biotechnologist and entertaining speaker, who provided an inspiring review of his work developing drug delivery systems and other treatment technologies that combine biology, chemistry, and electronics. His current work involves cartilage tissue engineering, including the remarkable accomplishment of growing an ear-shaped structure on a rabbit, and also the re-growth of spinal tissues in animals. He encouraged the audience to persevere, noting that in his early days of work, "no one believed in what I was doing," and predicted that ALD will become a mainstream technology with wide applicability.
Langer also gave an endorsement to high-throughput experimentation techniques (which my Intermolecular colleague, Russell Kempt, and I were gratified to hear). In response to an audience question, he said high-throughput techniques are ideal for addressing challenging problems when fundamental understanding is lacking, and new materials candidates need to be identified.
Another strong presentation came from Martyn Pemble of Tyndall National Institute in Ireland, highlighting the use of interface control layers, primarily nanometer-scale Al2O3, to boost performance of high-k gate oxides in InGaAs devices. III-V materials have great electrical properties, and high mobility, but the interface is very tricky. Pemble noted that cleaning and surface treatment are critical to ultimate success.
Also looking at better control was Shaista Babar from the U. of Illinois at Urbana Champaign, who discussed nucleation enhancement for the improvement of uniformity in ALD layers, which can experience roughness if the nucleation "islands" form in different sizes. Chemical or plasma surface activation are possible, but both have issues; Babar’s approach uses inhibitors to decrease the growth rate of the islands, helping to create continuous nucleation on the bare surface (with control of the interface again being called out as a critical factor).
K.H. Lee of South Korean equipment developer Wonik IPS Co. described an interesting mini-batch ALD system that targets higher throughput and lower cost-of-ownership. It utilizes rotating wafer stages that spin at 10-40rpm to provide better manufacturability using water, ozone, and metal precursors that are suitable for nanostructures.
Sunday night’s reception at the nearby Museum of Science was a fantastic opportunity for networking and conversation; we look forward to tonight’s poster session, and many more presentations tomorrow.