2

The Nanofabrication Facility

SCOPE AND MISSION

The role of the Nanofabrication Facility (NanoFab) in the CNST mission is to provide “economical access to and training on a commercial state-of-the-art tool set.”¹ More specifically, the NanoFab is intended to perform the following functions:

1. Provides access to state-of-the-art, commercial nanoscale measurement and fabrication tools and methods, along with associated technical expertise, in a shared-access, shared-cost environment to industry, academia, NIST, and other government agencies;
2. Enables processing and characterization of a wide range of nanoscale materials, structures, and devices critical to the nation’s measurement and technology needs; and
3. Fosters internal collaboration in nanotechnology across NIST’s laboratories and external collaboration with NIST’s partners through its shared environment.²

These mission statements are consistent with the statements on NIST’s website that the NanoFab

STAFFING

The NanoFab Operations Group consists of 21 people—a manager, 2 assistant managers, 6 process engineers, 3 microscopists, 2 user coordinators, 6 technicians (which includes 4 equipment engineers), and an administrative specialist. Of the process engineers, 3 hold doctorates in science or engineering, and 3 hold masters degrees. This represents an increase of 4 positions since the 2011 National Research Council (NRC) assessment.⁴ The group has, collectively, slightly more than 200 person-years of experience. In addition, the NanoFab draws on CNST support staff, who provide operational infrastructure, such as administration, information technology, facilities, and engineering. This is a reasonable level of staffing given the mission and scope of the facility.

___________________

¹ NIST, “Center for Nanoscale Science and Technology” http://www.nist.gov/cnst/, accessed June 10, 2016.

² Vincent Luciani, NanoFab Operations Manager, CNST, NIST, “Overview of the NanoFab Operations Group,” presentation to the panel, Gaithersburg, Md., May 2, 2016.

³ NIST, “User Facilities” http://www.nist.gov/user-facilities.cfm, accessed June 10, 2016.

⁴ National Research Council (NRC), An Assessment of the National Institute of Standards and Technology Center for Nanoscale Science and Technology: Fiscal Year 2011, The National Academies Press, Washington, D.C., 2011.

Process engineers are responsible for approximately 11 tools each. On those tools, they provide training, develop processes, and provide process services for remote users. In addition, they consult with users to help them develop processes or integrate processes to fabricate a desired structure. One of the process technicians works in the evening to provide onsite user support and to perform remote user support. A second process technician performs tool maintenance in the evenings. The facility relies heavily on external maintenance support through contracts on 10 of the 100 tools. In addition, the facility uses outside consultants to support annual safety audits. The NanoFab manager justifiably puts considerable emphasis on laboratory safety.

Staffing and equipment have increased since the 2011 NRC assessment in a way that seems commensurate with the expanded scope of work. Based on the growing numbers of research participants over the past few years, it would be expected that the staffing and equipment set would continue to grow accordingly, to fill what appears to be available space in the cleanroom and adjacent laboratories.

ALIGNMENT WITH MISSION

Consistent with the observation in the 2011 NRC assessment, the NanoFab operation, with its excellent facility and staff, is well aligned with the mission to serve as a national user facility. The presentations provided during the review showcased how these facilities were leveraged to support NIST research projects and industry and, through collaborations, academia. Performance against mission objectives was difficult to assess because a detailed strategic plan with a roadmap and performance metrics was not presented to the panel. Only limited user information was provided, as discussed below.

QUALITY OF RESEARCH AND FACILITIES

The CNST facility is equipped to support a broad range of research activities, including many materials and substrates. This requires a highly flexible tool set that can handle anything from 1 cm samples to full 200 mm wafers. In that respect, it is reasonable to compare the facility to those found in leading research universities. One would see similar equipment at Cornell University or Stanford University. The CNST facility has somewhat more equipment and gives users access to a few higher-end capabilities, but the difference is not dramatic.

Overall, the NanoFab tool set is robust. The lithography capabilities are strong, with standard contact printing, an i-line ASML stepper, direct write optical, and two JEOL 6300 electron-beam lithography systems. An array of resist-processing equipment exists, including an unusual resist-dispense system that supports highly nonplanar substrates. The NanoFab has multiple dual-beam, focused-ion beam systems that are used for fabrication and transmission electron microscope (TEM) sample preparation. The NanoFab hosts multiple field emission scanning electron microscopes (FE-SEMs) and an FEI Titan S/TEM for extreme (<0.2 nm) resolution microscopy. Other tools include several thin film deposition systems, including an impressive physical vapor deposition (PVD) system capable of ultra-flat interfaces of metals and oxides. A soft lithography bay is currently being set up.

The current capability of the facility is somewhat above the leading university fabrication facilities, and tool acquisition is continuing at a strong pace. The NanoFab is on a path to become one of the best fabrication laboratories in the country that are open to outside researchers. A key distinguishing feature of the CNST facility is the pervasive use of statistical process monitoring. All of the major tools had charts showing the history of some standard process results. This helps to provide more consistent process results, although it requires routine running of standard processes, and so adds cost.

Examples of several user projects were given during the review. These included several that required considerable process development and/or unusual capabilities.

One issue facing university and other small fabrication facilities is the dearth of laboratory management platforms that can support a large, shared-use facility in a way that is compliant with respect

to cost and to network security and versatile enough to accommodate the multitude and variety of user transactions. The commercial options are vulnerable, one-person operations. Many fabrication facilities have resorted to in-house software development, which is challenging and beyond the financial wherewithal of most. With in-house software development resources, the NanoFab staff have developed the most advanced, elegantly designed in-house laboratory management system that the panel has seen. The system is compliant with even the stringent security constraints of NIST. Moreover, in contrast to the monolithic design of most available laboratory management systems, the NanoFab Equipment Management Operation (NEMO) was built using current software development methods. It is module-based, making it easy to modify, upgrade, and maintain. The CNST would do the country well by making NEMO available to the fabrication research community.

URGENT NEED FOR EVALUATION OF THE NANOFAB AS A USER FACILITY

As noted in the 2011 NRC assessment, the NanoFab and its highly skilled staff are responsible for a major advance in the research capabilities of the CNST. The NanoFab and its staff address the core mission of operating a shared-use facility. Facility usage was difficult to assess, because the information provided by the CNST director was insufficient. When pressed on this point, he provided some additional information, but detailed analysis was difficult because the information was partial and provided only at the last minute, and there was insufficient opportunity to ask follow-up questions.

Considering the limited information that was provided, the following represents an attempt to benchmark the CNST NanoFab against the National Nanotechnology Infrastructure Network (NNIN) facilities, which are reasonably comparable in mission, size, staffing, and number of tools. The former NNIN consortium of 14 national user facilities collected user statistics over its 12 years of operation. Over the course of the NNIN, seven of the sites built new facilities or joined with new facilities. They aggressively marketed themselves following the open, shared-use model that has been demonstrated to be an effective way to cultivate a critical mass of users. For each of the new NNIN facilities, the user base was largely established in 5 years, with growth slowing thereafter. The CNST NanoFab has been in operation for about 9 years, and the expectation is that this would have provided sufficient opportunity to establish the bulk of its user base.

Without the needed data, it was difficult to determine how much capacity is available; however, during a nearly hour-long in-laboratory tour in the middle of the day, only one non-staff person was observed. Furthermore, in comparison to similarly sized NNIN facilities, the user base numbers are much lower, by at least 25 percent and perhaps up to 300 percent. This leads to the conclusion that despite its enviable location and advanced capabilities, the CNST NanoFab is not aggressively recruiting new users, particularly external ones. This conclusion was confirmed by comments made by the CNST management during the review.

A more reliable source of data is user fee income, which captures both user number and use intensity for both local and remote users. With a current operating budget of about $7 million, about half is provided by user fees. This is a reasonable balance, judging by the experience of NNIN facilities. External NanoFab income has increased significantly since 2011 and is now approximately $1.2 million per year. Income from NIST sources outside of the CNST has declined somewhat over the same period. The fiscal year (FY) 2015 users fees came from the following sources: $1.5 million from users employed by CNST, $0.9 million from users employed by NIST (outside of the CNST), and approximately $1.2 million from users external to NIST. Overall, the user fee income is increasing by about 10 percent per year; however, this was heavily influenced by a large spike in CNST usage in FY2015. Ignoring that spike, total user fee income has been nearly flat since the NRC assessment in 2011.⁵

___________________

⁵ NRC, An Assessment of the National Institute of Standards and Technology Center for Nanoscale Science and Technology: Fiscal Year 2011, The National Academies Press, Washington, D.C., 2011.

The NanoFab is a key asset for the CNST and the Washington, D.C., area. Proximity to a large number of universities, major agencies such as the National Institutes of Health (NIH), and a significant number of companies, all of which have little or no capabilities in nanoscale fabrication and characterization, makes the CNST NanoFab a tremendous potential asset for economic development and training of nanoscale researchers in the region. Furthermore, given its level of support and access to the process research team, it could become a national resource for research fabrication facilities. Considerations pertaining to the NanoFab and its management are summarized below.

The detailed quantitative information needed for evaluation of the NanoFab was difficult to obtain. This is completely unacceptable for a user facility with this level of federal support. The lack of evidence of external users during the review tour, combined with the opacity of the data given, creates an impression that the picture being provided was overly optimistic. The NanoFab needs to maintain and freely provide accurate year-to-year data on the number of users, the sources of these users, and the amount of user income. The latter provides the most unambiguous measure of facility usage. Furthermore, to provide an accurate assessment of the NanoFab, this data cannot be conflated with that of the NanoLab. The former consists of commercially available equipment designed to fabricate micro- and nanostructures. The latter develops and hosts unique equipment, generally designed for exquisitely detailed measurements.

If it has not already done so, the NanoFab needs to start a process that provides a detailed tracking of all users to determine their origin and their fields of interest (microelectromechanical systems [MEMS], materials, electronics, etc.). Laboratory operating software can then be used to determine laboratory usage by group. This is routinely done by leading research fabrication facilities, and it is an essential tool for understanding usage trends to make rational decisions about staffing, equipment upgrades, and new equipment acquisition. Furthermore, polling of users is essential and needs to be done regularly. The difficulty experienced by the panel in getting this data is a glaring deficiency.

The NanoFab management needs to begin an outreach program to recruit new users. Since they have not yet begun significant recruiting efforts, additional growth is possible. More users decrease idle equipment time, spreading costs more broadly. They also increase the impact of the facility. Contrary to the evaluation provided to the panel, the panel assesses that the laboratory is nowhere near capacity. Approaching capacity is to be viewed as a positive goal and would yield an increased impact. Furthermore, the NanoFab can increase staffing levels and/or add equipment or start to attract more external users by the quality of the projects.

The demonstration of support for industry users that distinguishes the NanoFab from university facilities is encouraging. More rigorous process control provides more reproducible processes that enable commercial success for companies. This comes with an increase in user rates relative to university laboratories. For example, the NanoFab electron-beam lithography rates, even subsidized, are 2 to 3 times those of comparable machines at universities. Concerns were raised about the application of the NIST subsidy to private companies. In the Department of Energy (DOE) model, companies are given access free of charge, but applications go through a reasonably rigorous external review process. The current CNST review process appears to be cursory. This can be an issue, especially if the facility begins to approach capacity and starts to become more selective.

The leadership needs to define a strategic roadmap for the NanoFab. The CNST director and the staff expressed the desire to see more users from a broader range of disciplines and external organizations, particularly from industry, given the association with the Department of Commerce (DOC); yet the action plan and metrics for success were not articulated. Developing and incorporating metrics to manage operations would enable the benchmarking of performance against that of similar facilities and would also be a means of goal setting for strategic planning.

The laboratory needs to continue drawing down resources devoted to complementary metal-oxide semiconductor (CMOS) technology because research in this area has declined sharply. Support for the new nanobiomedical area will be required, but details of the nature of this support are difficult to predict at this time due to the embryonic nature of the current effort. The investment of CNST resources to recruit a leading researcher to break new ground in the nanobiomedical area for the NanoLab is acknowledged;

however, the NanoFab only needs to look to similar facilities in universities where nanobiomedical research is taking hold, or to engage with NIH, to learn which equipment and processes are most needed to support research efforts in this area.

Its high level of sustained funding and collaboration with the process research team gives the CNST NanoFab the opportunity to be a leading resource in the country. As such, they need to be more outward looking and more broadly engaged with the fabrication community beyond the traditional mechanisms of research collaborations and peer-reviewed publications, through service in professional organizations, sharing fabrication protocols, and proliferating best practices (such as NEMO). Playing a highly visible role in University/Government/Industry Micro/Nanotechnology (UGIM) Symposium and the National Nanotechnology Coordinated Infrastructure (NNCI) are two possible avenues. An effort to make the CNST process detail readily available for publication in the NIST research journal is an excellent idea, as is the lithography toolbox developed by a project leader in the NRG. Still, distribution mechanisms need to be developed if the work is to have any impact. NIST could do the nanofabrication world a great service by playing a leading role in developing and propagating fabrication research processes and practices.