Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 177
Midsize Facilities: The Infrastructure for Materials Research E Report of a Site Visit Team INTRODUCTION The Boston Area Site Visit was conducted by three members of the Committee on Smaller Facilities (COSF)—Donald Tennant (Lucent), Frances Ross (IBM), and John Soures (University of Rochester)—and facilitated by the Board on Physics and Astronomy’s director, Donald Shapero. As in most of the regional visits, the trip was arranged such that a number of facilities and centers could be accessed in a 2- to 3-day period, and therefore much of the time was devoted to interviews and tours rather than point-to-point travel. The Boston area trip therefore permitted visits to three institutions, which represented many more research centers, all within a 2+ day stay. The trip was conducted from July 30 to August 1, 2003. Trip reports were filed for internal committee use from each of the area groups, and summary reviews were conducted by the group leaders at the COSF meeting at the Stanford University campus in October 2003. The host institutions for the Boston area included the Massachusetts Institute of Technology (MIT) and Northeastern University on July 31 and Harvard University on August 1. All three institutions were generous with their time and made access to staff and facilities amply available to the committee. The specific individuals and research organizations visited included:
OCR for page 178
Midsize Facilities: The Infrastructure for Materials Research Massachusetts Institute of Technology Center for Materials Science and Engineering, Michael Rubner, Director Materials Processing Center, Lionel Kimerling, Director Microphotonics Center, George Kenney, Associate Director Northeastern University NASA Center for Advanced Microgravity Materials Processing, Albert Sacco, Jr., Director Electronic Materials Research Institute, Donald Heiman, Principal Investigator (PI) and Professor of Physics Harvard University Center for Imaging and Mesoscale Structures (CIMS), Bill Appleton, Director Venkatesh Narayanamurti, Dean of Engineering and Applied Sciences Rowland Institute of Science, Frans Spaepen, Director Materials Research Science and Engineering Center, David Weitz, Director Nanoscale Science and Engineering Center (NSEC), Robert Westervelt, Principal Investigator (PI) and Director OBSERVATIONS Massachusetts Institute of Technology Forum The site visit team met most of the time with the leadership of the Center for Materials Science and Engineering (CMSE) and the Materials Processing Center (MPC), had tours of the CMSE facilities, and conducted impromptu interviews with staff. Facility Mission/Purpose The mission of the MIT CMSE is to foster collaborative interdisciplinary research and education in the fundamental science of materials and in the engineering of materials for specific applications. CMSE promotes collaborations among MIT faculty as well as between MIT research faculty and researchers at other universities, industry, government, and nonprofit laboratories. Collaborations are encouraged through several mechanisms including: Interdisciplinary Research Groups (IRGs), Initiative Projects, Shared Experimental Facilities (SEFs), and Outreach Programs. The CMSE SEFs include electron microscopy, analytical, x-ray diffraction, and crystal growth facilities; these facilities are operated for both internal (MIT) and external users. Background/Origins/History The MIT CMSE is one of 29 existing Materials Research Science and Engineering Centers (MRSEC) sponsored by the National Science Foundation (NSF). MIT has been part of the MRSEC program since 1994.
OCR for page 179
Midsize Facilities: The Infrastructure for Materials Research The CMSE has benefited from the creation of a “virtual center”—the MIT MPC—in 1980. The MPC is an umbrella organization that focuses on creating shared structures and partnerships. Within MPC, MIT has created the Microphotonics Center, whose function is to provide for the creation of new materials, structures, and architectures to enable the evolution of photonics from single, discrete devices to integrated photonic systems. The MIT CMSE was recently renewed by NSF for a period of 6 years. The Center supports an interdisciplinary research program with emphasis on micro-and nanostructured materials in the areas of photonics, polymer assemblies, and semiconductor and magnetic structures. The CMSE has also developed a strong educational program, with graduate, undergraduate, and K-12 elements. Shared Experimental Facilities Operations The cost structure for the shared facilities was discussed. Issues such as maintenance, salaries, administration, equipment, and materials were explored. Funding discussions followed, including the budget portions that are supplied by NSF-sponsored MRSEC core funding, user fees, cost sharing, and so on. Specifics were discussed in order to help identify typical practices when compared with other institutions. The site visit team learned that there are eight full-time support staff to operate the four SEFs. The staff titles range from Project Technician to Principal Research Scientist. There are two staff assigned to each of the primary SEF elements: Materials Analysis, Crystal Growth and Preparation, Electron Microscopy, and X-ray Diffraction. The shared facilities are most often operated by the individual users after hands-on training by the facility staff. The group discussed planning for operations, maintenance, and upgrades. Users sign up to use the facilities on an as-needed basis. Facility maintenance, upgrades, and so on are coordinated with the major users. The management of the SEFs is apparently given a high priority with the CMSE. A high-level senior research scientist oversees the operation of the SEFs. Faculty user groups are established as required, to guide the selection of capital equipment and to assess the facilities’ performance. Postdoctoral associates are often incorporated in the facility organization to introduce and develop new instrumentation. Instrumentation and Services Provided The CMSE SEFs provide state-of-the-art capabilities in a timely manner in four primary areas: Materials Analysis: Atomic force microscope, X-ray photoemission spectrometer, scanning Auger microprobe, ultraviolet-visible-near infrared spectrophotometers, spectrofluorimeter, two differential scanning calorimeters,
OCR for page 180
Midsize Facilities: The Infrastructure for Materials Research dynamic mechanical rheology analysis, Fourier transform infrared spectrometer, micro-Raman spectrometer, stress measurement system, thermal co-evaporator, and profilometer. Crystal Growth and Preparation: Three floating-zone furnaces, six furnaces for top-seeded solution growth, cutting and polishing apparatus, differential thermal analyzer/thermogravimetric analyzer, and two Superconducting Quantum Interference Device (SQUID) magnetometers. Electron Microscopy: 250 kV field-emission scanning transmission electron microscopes (STEMs), three 200 kV TEMs, field emission gun SEM, environmental SEM with x-ray analysis, specimen preparation, and image analysis equipment. X-ray Diffraction: Four rotating anode x-ray generators, two sealed-tube x-ray generators, a small-angle x-ray scattering diffractometer, four circle diffractometer, automated crystal analysis system, precession and Laue cameras, pole figure, glancing angle and high-temperature attachments. Annual use of the SEFs averages 600 to 750 individuals. Approximately 72 percent of the users are MIT internal. The majority of the external users have an academic affiliation. Only about 1 percent are industrial users. Almost all the users are local/regional. Since September 2002, over 600 individuals have used the SEFs, including students and postdoctoral associates of 100 MIT faculty in 23 academic departments and centers, students and staff of 17 faculty from other academic/research institutions, and the staff of 6 senior-level industrial managers. Education is one of the most important functions of the SEFs. The educational function is exercised in many ways: hands-on operational training provided to users (faculty, staff, and students), minicourses open to the MIT and external academic communities, and the interchange and intermingling provided by the interaction of the multifaceted user community. Northeastern University NASA Center for Advanced Microgravity Materials Processing Forum Discussion and tour of laboratories by the director Albert Sacco, Jr., brief introduction to some research (shown by a postdoctoral associate). Facility Mission/Purpose The aim of the Center for Advanced Microgravity Materials Processing (CAMMP) is to collaborate with industry to develop materials or processes of use both to the company and to NASA (mostly by ground-based experiments, a few in space).
OCR for page 181
Midsize Facilities: The Infrastructure for Materials Research Background/Origins/History This is an unusual facility in that it is funded jointly by NASA (60 percent) and by the university (40 percent). It has been going since 1997. It develops its user base by aggressive solicitation (company visits) and by personal contacts of the director. However, only about 60 percent of proposals are accepted. Shared Experimental Facilities (SEF) Operations CAMMP has little formal management structure; it is mostly run by its founder/director. To become a user, the company pays a membership fee, but after this there are no user fees. So this is really a consortium of industries and the university. It appeared that the membership fee depends on the scope of the collaboration. The biggest user is a commercial chemical company; others included U.S. and European universities. These university collaborators must have industrial links to use CAMMP. The industrial partners fund one or two postdoctoral associates or graduate students, usually for 3 years. If the industrial partner drops the project, Northeastern University (NE) pays for the remainder of the student’s term, allowing the student to continue with related research. There are no per hour fees; the belief is that the faculty could not afford them, and instead faculty often supply other value or resources to the operation. Major equipment includes: SEM (used 25 hr/week), x-ray diffraction (XRD) (35 hr/week), GADDS (general area detector diffraction system, an unusual capability, 5 hr/week), atomic force microscope, focused-ion beam, mass spectrometer, Fourier transform infrared spectrometer, particle-size analysis. The only equipment desired is a TEM (currently they go to an industrial partner’s central laboratory); the director believes that the current size is about right. CAMMP uses service contracts and buys new to avoid maintenance costs; lower-use equipment is given to NE faculty. In contrast to MIT, CAMMP SEFs have no technical staff, and technical support is done by postdoctoral associates on extended terms (called research assistants), who stay 5 to 10 years and are often coauthors. There are typically 8 faculty involved, 8 students (3 postdoctoral) from NE, 8 from outside universities, 1 finance administrator, and the director. Instrumentation and Services Provided The CAMMP facility is smaller and less conventionally funded and therefore more than most reflects the philosophy of the director: The director has informal collaborative arrangements for equipment. He believes that user fees serve no purpose, because in small universities faculty are too poor to pay them. Instead, “ideas are currency, not money.” Apart from the funded graduate students, the center hosts 3 to 4 honors students per year, many school teachers, and does some evening classes on techniques (NE is strong on evening classes).
OCR for page 182
Midsize Facilities: The Infrastructure for Materials Research Electronic Materials Research Institute Forum Group discussion with about 15 people, including the Dean of the university, PI (Donald Heiman), other faculty members, a few users, and some of the technical staff, followed by a brief tour of the outside of clean rooms and the SEM/XRD laboratory. Facility Mission/Purpose This facility is under development and currently has no physical location. Its aim is to bring all shared equipment together into one location, run it, provide matching funds when groups of faculty apply for equipment, determine overall needs, and solicit donors. Background/Origins/History The Electronic Materials Research Institute (EMRI) started in 2002 with a modest seed grant from the university to hire and invest in equipment; 15 faculty are involved now, with planned increase to 20 to 25. It was hard to get numbers for the projected final size and budget. The dean has a vision for this center that he compared to the plans for biotechnology (his specialty). According to him, NE is very healthy with its student programs and plenty of government (National Institutes of Health) and industrial (biotechnology) support. It has already (in 1973) set up a biological and chemical effort, the Barnett Institute, with a $100 million building, two endowed chairs, 20 faculty, an endowment, ample research funds, and a notable patent income, and would like to make the same size of effort in materials. As was done with the biotechnology area, he wants industry to define the equipment and curriculum; the biology faculty have a track record in getting funding, and the dean expects EMRI to actually generate revenue. Shared Experimental Facilities (SEF) Operations An instrumentation committee, with a member from each department and one of the technical staff, considers user feedback and equipment on campus and at other area research institutions. Users are to include those in the areas of biology, pharmacology, materials, physics, and so on. User fees are nominal but are structured as hourly and yearly; revenue is generated from educational activities. The EMRI staff is currently 15, but the number is very fluid, and presumably very few of these (secretary, some students) are funded by EMRI itself. The technicians are still funded directly by the university. Teaching activities include student training, plus classes in techniques and pre-college courses. Research training is projected into broad areas, which include photonic materials, fuel cells, theory, metamaterials (left-handed materials), pharmacology/biotechnology; the special feature of EMRI seems to be the biological connection.
OCR for page 183
Midsize Facilities: The Infrastructure for Materials Research Instrumentation and Services Provided This facility seems still very much under development. But much was already present in NE: SQUID, near-field scanning optical microscope (NSOM), TEM, AFM, XRD, microelectromechanical systems (MEMS) fabrication, and a computing facility. Their first success as EMRI was in obtaining SEM and NSOM. Their next requests will be for TEM, nuclear magnetic resonance (NMR), and MEMS capabilities. They expect to add more equipment by applying for Nanoscale Interdisciplinary Research Team and NSEC funding. Harvard University Forum The site team visit to Harvard was primarily hosted by the CIMS and two major NSF centers, MRSEC and NSEC. Tours were conducted among several locations, since the new building to house most of the shared facilities is still in the planning stages. The director of the Rowland Institute at Harvard, Frans Spaepen, also briefly described the activities of the Rowland Institute, which was recently acquired by Harvard. Facility Mission/Purpose The CIMS mission is to manage the shared facilities, partially fund visiting scientists, provide small grants to seed research projects, and provide staff assistants. Current research missions include synthesis of nanoscale structures, fabrication of mesoscale devices, advanced imaging and characterization services, and development of new fabrication and imaging methods. The participants draw from five different departments. Background/Origins/History CIMS is a facility that supports the research activities of two major NSF centers at Harvard, MRSEC and NSEC, in addition to individual research grants and the undergraduate research mission. The center had a proposal pending as a node in the NSF-funded National Nanotechnology Infrastructure Network which has since been successful. While the current equipment is distributed among several locations, a new building and clean room being built by Harvard will eventually house most of the shared facilities. This is an investment on the order of $100 million. The merging of the shared facilities also allows cost share of operational expenses, including support staff. The site visit team also heard briefly about the Rowland Institute but did not visit those facilities. Shared Experimental Facilities (SEF) Operations About 30 faculty are participating in the CIMS activities. Thirteen technical staff (20 eventually) support the suite of instruments, with 3 postdoctoral associates and 4 visiting scientists. About half of the operation budget comes from the dean from the division budget, the rest from fees and grant cost shares. Since this is a new undertaking, this formula is
OCR for page 184
Midsize Facilities: The Infrastructure for Materials Research expected to change as more user fees and grants accrue. The construction costs and new faculty are funded by the Harvard endowment. Harvard has recently brought the physical sciences faculty and the division of engineering and applied science together under one dean. This seems to be a high-level effort to maximize the interaction of the various faculty. CIMS has a non-faculty director plus a faculty scientific director. Great attention has been paid to developing specialized space in an effort provide good tool performance. The university has jump-started the programs with new faculty hires and is refurbishing temporary space in advance of the completion of facility construction. Harvard offers both short courses and credit courses in some of the technical areas being fostered by CIMS. Undergraduate, not just graduate students, have access to these world-class research facilities. An applied physics course (credit) was also offered in which 19 professors gave tutorial lectures on their research fields. Attracting and retaining high-quality support staff is a high priority. While support staff are hired on soft money, they can be given rolling 3-year contracts. This adds stability, but requires the university to assume the cost if a grant is discontinued or a funding shortfall is experienced. To encourage and attract high-quality staff scientists, contributions are recognized as authorships in publications. Instrumentation and Services Provided The CIMS facilities were distributed and incomplete but growing in capability rapidly, and with plans to form a centralized user facility in a new building. The instruments merged analysis and fabrication capabilities in one organization. Materials research instruments include TEM, SEM, ultra-high-vacuum-STM, Rutherford backscattering, XRD, ion accelerators, and fabrication instruments such as electron-beam lithography, FIB, optical pattern generator, furnaces, deposition, and etching tools. FINDINGS, BEST PRACTICES, AND DISCUSSION POINTS Following is a cumulative list reflecting the observations of the site visit team based on all of the Boston area site visits. On the Importance of Shared Facilities The innovation is at the intersection of the sciences, and central instrumentation facilities attract individuals and foster cooperation, sharing, and collaboration. Shared facilities such as the MIT CMSE provide an essential service to MIT researchers, faculty, and students as well as to external users engaged in
OCR for page 185
Midsize Facilities: The Infrastructure for Materials Research materials science research and engineering activities. They provide educational as well as research value for the materials science community. SEFs provide opportunities for sharing experiences and knowledge among a large group of users with diverse backgrounds. Off-campus fabrication facilities are less desirable than off-campus characterization facilities. Much of the value of smaller facilities is nonmonetary—enabling faculty to mix and changing the mind set from independent, competing research programs to joint efforts—as well as monetary: providing a focus for large donations (e.g., buildings) and for institutional commitment. Regarding how to make facilities more visible: If smaller facilities were more visible, would they get funded and maintained more readily? The team discussed how to emphasize to NSF reviewers the value, monetary and nonmonetary, of such facilities in awarding other grants. On Management/Philosophy There is a new recognition that universities take responsibility for investing in research facilities (are these the new libraries of undergraduate education, as one member of faculty put it?). To build a significant facility, high-level support (president, provost, dean) for facilities is essential. For shared facilities to work, there needs to be recognition of a service component in everyone’s participation. There are efficiencies that industry can teach the universities with regard to interaction among staff, but there are important differences—for example, NSF must take a much longer view. While the sharing of experimental facilities is key to keeping down the overall cost of materials science research activities, it is evident that such sharing is most effective when the shared facilities are not too distant from the users (local or regional) and when the facility schedule allows for relatively quick turnaround and scheduling (<1-2 weeks). MRSEC is vitally important to convincing a university to invest. It is a long-standing example of shared central facilities in the materials area. Some facilities have only basic equipment (that seems to be the philosophy of the MIT MRSEC shared-equipment facility), and the complex pieces seen by the team (the ultra-high-vacuum STM at MIT) required hiring a postdoctoral associate or technician to run; it was not clear how the concept of users versus collaborators would apply if there are only a few experiments each year.
OCR for page 186
Midsize Facilities: The Infrastructure for Materials Research Successful facilities frequently passed on low-use equipment to faculty or took over the running of heavily used equipment; a bureaucracy that allows such flexibility appears to be a good practice. Another question was how much instrument development should be part of the mix. A component of instrument development (as at the Harvard CIMS) is probably appropriate, as there is little other funding avenue for it and it makes use of the skilled technical staff (and makes their jobs more interesting, too). Several people made the comment that multiple clean rooms are needed because of materials incompatibility, so similar equipment even in nearby facilities is necessary. And too much emphasis on a few large centers prevents students from getting hands-on experience. There seemed to be no uniform way to report income and expenditures of a facility. For example, depreciation is usually ignored (see the section below) and if the university provides secretarial or accounting help, how much is this worth? On Funding Sources and Gaps The shared experimental facilities face challenges in several areas, including equipment renewal. Due to NSF’s budget limitations and grant limitations, equipment costing in excess of $500,000 is nearly impossible to include in the MRSEC grant budget. The single-item cost of state-of-the-art instrumentation for these facilities ranges from $100,000 to over $1 million per major system. Some of the equipment on the center inventory is over 10 years old and will be in need of replacement or renewal in the immediate future. The discussion of user fees raised lively discussion. Most facilities only expect to raise 25 percent of their running costs through user fees. The typically $250,000 raised per year is useful in buying consumables, but did not appear to be the make-or-break factor in running a facility. An argument for fees was that they lead to extra efficiency on the part of users. Arguments against are that it is hard on a low-budget faculty, especially in less-well-funded schools, and it is philosophically wrong because it hinders the educational value of the institution. High user fees discourage wide use of facilities. Some facility users and managers think that a facility should be free, but most think that user fees demonstrate that instruments and services have value. It seemed important that users have input and form a consensus and buy in with respect to the usage, fees, priorities, and so on. Many small users feel that large users are receiving subsidized use of facilities.
OCR for page 187
Midsize Facilities: The Infrastructure for Materials Research Should a university-funded user facility be considered part of the cost share when applying for a grant? There were several ideas about how this should be calculated, but most facility users and managers agreed that it should be, or at least that the concept of cost sharing should be made more flexible. As an example of the high cost of the building facilities required to carry out such work, the MIT SEFs occupy a total of 9,200 ft2 of space. The facility managers estimated that the cost of building such space is higher than $2,500/ft2. Future upgrades and improvements may therefore require significant expenditures in building construction and renewal—an area not normally covered by grant funding. Several faculty noted that the MRSEC has been flat funded for many years and that the budget fraction earmarked for infrastructure has shrunk over the years and has become too small to foster renewal even with normal cost shares. This, some believed, reflects the review panel weight placed on the IRG proposals and the added budget required for K-12 outreach. There is some sentiment among faculty that many funding agency program managers are flexible in the management of centers, but that the review panel expectations are more rigid. Therefore any changes (e.g., in budget allowances) in guidelines need to be clearly announced to the reviewers. On this trip, the site visit team visited well-endowed universities, and the donation of a $100 million building seemed almost required to get a major new facility going! Most facilities have a business plan for accepting donations. The team noted that donors really seem to like facilities but are reluctant to donate equipment. It seemed clear that the operation of the SEFs is dependent on strong support from the university and on subsidies from the primary (NSF) grant. If the actual cost of operations had to be charged to all the users, either the shared experimental facilities would not be able to operate or the scope of work of the center would have to be significantly curtailed. On Professional Staffing Since the facilities are highly specialized, significant experience and a high level of specialized education are required to adequately operate the SEFs. On the other hand, due to the small size of the facilities, there is sometimes little opportunity for personal growth and advancement. Departure or retirement of key individuals remains a major concern for smooth continuing operations. Everyone the team visited agreed that technical support staff are key to the facility success. The team saw either successful facilities or those that had
OCR for page 188
Midsize Facilities: The Infrastructure for Materials Research only just started. MIT permits its technicians to move around, learn more skills, and be coauthors or presenters. Moving around only works in large facilities, but one possibly more universal improvement that the team discussed was establishing a fund to get staff to conferences or classes. There was wide support for the idea that there should be a modest budgetary component for professional growth and the training of support staff in MRSEC, NSEC, and other facilities. There were surprisingly many staff at the postdoctoral level rather than people with specific training as technicians. In one case (CAMMP), it was impossible to find a technician for one piece of equipment and a long-term postdoctoral associate was hired instead. Such people are motivated by the work environment, or maybe like research but prefer not to have the responsibility of directing their own research. If the supply of such postdoctoral associates is good, perhaps the lack of technician training schools and lack of career paths are not such a problem. Job security can be important to retaining highly skilled staff. Harvard guaranteed 3 years’ employment even if the NSF grant were to be terminated, giving technical staff some job security. This is a way that a university can facilitate the quality of shared facilities.