This Small Business Innovation Research Phase I project is to demonstrate the feasibility of batch fabricating high-aspect ratio atomic force microscopy (AFM) probes. These probes have excellent mechanical and electrical properties and are customizable to a wide range of applications and substrates. Currently each probe is individually fabricated by dipping a silver-coated probe into melted gallium at room temperature, resulting in the self-assembly of a long, constant-diameter metal nanoneedle on the probe tip. Current production throughput is only five probes per hour. Because of their unique form and function there is a growing demand for these probes which can only be met if they are fabricated in parallel. In Phase I, a batch process will be developed, with the goal of moderate yield (25%) over a 1 cm square area. One innovative aspect of the project is the use of a gallium coated substrate that has an elastomeric underlayer to provide a degree of self-alignment that ensures intimate contact of the thick gallium film layer with surfaces that are not perfectly flat. The extension of this concept - in future studies - to the patterning of arrays of freestanding nanoneedles over curved and multilevel substrates appears reasonable. Based on the attainment of adequate yields in Phase I, Phase II will focus on the development of a semi-automated tool for wafer-scale growth of probes.
The broader impact/commercial potential of this project is a dramatically increased commercial viability of a new kind of specialized AFM probe. The total market for AFM probes is $385 million, of which up to $100 million is addressable, if such probes can be fabricated in larger quantities. Since the launch of this technology in late 2008, customer feedback has been overwhelmingly positive. Current customers of these probes have made it clear that this new probe technology represents an enabling tool which will help advance and accelerate the pace of research and discovery in areas including nanomanipulation, biophysical probing, nanomechanics, nanoelectronics and metrology. The long range economic and societal impact will be a new manufactured product which will help to maintain U.S. leadership in nanotechnology and create high-paying technical jobs for scientists and engineers in Kentucky, a state where such opportunities have traditionally been extremely limited.
Project outcome report: The ultimate goal of this SBIR project is to design and build a tool that enables the batch fabrication of specialized AFM probes (NeedleProbesTM) over entire 4-inch wafers. In the fabrication of NeedleProbes, a single freestanding metal nanoneedle is grown onto the tip of an AFM probe. Currently, NaugaNeedles manufactures such probes by a low throughput, one-at-a-time, and man-in-the-loop process. In the proposed batch processing tool, the process will be performed in parallel, similar to a mask aligner, wafer bonder or nano-imprinter. Prior to the start of the Phase I project a few arrays of nanoneedles had been fabricated in parallel by a fairly uncontrolled, manual pulling process. The results demonstrated that the yield and uniformity need to be greatly improved before batch fabrication can become reasonably practical. In Phase I, efforts were made to fine tune the processing steps and to control process variables. An improved, semi-automated experimental pulling station (EPS) was constructed (Figure 3 from the Phase I report). The system includes high optical magnification (0.5 micron resolution) side view optics that provide live viewing of the pulling process — from the time that the gallium reactant on one substrate contacts the silver film on the AFM tips on the other substrate, and through substrate separation and growth of the nanoneedles (See Figure 4 from the Phase I report). The pulling system has enabled numerous experiments (over 60 experimental needle arrays have been grown as part of this study) and rapid feedback, on substrates that are 0.2 to 1 cm wide. The goal of Phase II is the development a full 4-inch wafer-scale prototype of a nanoneedle pulling system (NPS). The highest yield achieved to date is shown in Figure 1. A flat, smooth gallium film was brought into contact with an array of sharpened pillars that resemble AFM tips. The EPS provided good control of alignment and process parameters, so that each silver-coated tip was nearly identically exposed to the same amount of gallium during the entire process of contacting and retracting the gallium. At this point of the study, the process variables for the nanoneedles pulling process (pulling velocities, gallium film thickness, and silver film thickness) are being correlated with the parameters of the resulting arrays (yield, average diameter and length of the needles). (See Table 1 from the Phase I report.) Additional data logging and analysis will continue throughout the Phase II study. Also, during Phase I, we evaluated several different approaches to making gallium films that are constant in thickness and locally smooth. Originally, we proposed to address these goals through the evaporation of gallium. With this technique, we found that the average film thickness over the entire substrate was well-controlled and near uniform, but the gallium tended to coalesce in small droplets that increased in diameter with film thickness (Figure 2a), thereby increasing local roughness. We found that brushing over the film (while it was in a melted state) with a straight-edged blade produces films that have a local roughness of only a few nanometers (Figure 2b). This led to our Phase IB proposal of directly applying gallium by blade-brushing to planar substrates, pillar arrays, and even substrates with shallow recessed wells. Arrays of gallium droplets also were patterned onto pillars using a conformable elastomeric membrane. This technique led to a dramatic improvement in yield (Figure 2b-d). Given the critical importance of uniformity, we plan to continue to evaluate alternate approaches to making gallium substrates, beyond the current Phase IB and into Phase II. Overall, in Phase I, a process that previously had extremely low yields was improved through controlled alignments, pillar array preparation, and gallium preparation to reach yields that exceeding 90% in three trials. While improved uniformity in needle diameter, length, and orientation will be important in full production, the results to date show clearly that, with further refinement, full 4-inch wafers can be patterned with high yields. Also, there is substantial room for improvement in alignment, gallium preparation, and post-fabrication processing, all of which could contribute to making the needles more uniform. Once these improvements to the fabrication process are developed, the extension of the method to patterning the needles on actual AFM probes should be straightforward. Thus, we believe that the likelihood of demonstrating a proof-of-performance prototype tool that can produce complete, salable NeedleProbes on a full wafer is reasonably high—as early as the end of Phase II. Initially variability might still be high enough to require somewhat tedious hand selection at first. Beyond Phase II, additional manufacturing process R&D is planned in order to continue to improve uniformity and reduce the number of probes sampled and amount of time involved in inspection during quality control.
