Isolator/minienvironment progress report: One degree of separation that radically changed advanced manufacturing.
By Sheila Galatowitsch
Enclosing critical processes in steel boxes—whether to protect the product or the worker from contamination—has grown into a mainstay of high-tech manufacturing.
In the case of minienvironments used in chipmaking, the technology has “gone about as far as it will go in convincing people,” says Ken Goldstein, Ph.D., principal of Cleanroom Consultants Inc. (Scottsdale, Ariz.) and co-chair of the Institute of Environmental Sciences and Technology's (IEST) Working Group 28 (minienvironments).
“Nobody asks the question, 'Do minienvironments work?' anymore. Nobody questions whether they are financially or economically viable. Everybody knows they cost more money but make it up in terms of yield and speeding up production ramps,” says Goldstein.
Meanwhile, isolation technology in pharmaceutical manufacturing has “ridden itself of the noise and confusion” surrounding its use, says James Agalloco, president of Agalloco & Associates (Belle Mead, N.J.), a technical service firm. “There are enough success stories out there that people are confident about the technology.”
These maturing innovations, however, still have a way to go before widespread penetration in the user base. Here's a progress report on the one degree of separation that radically changed advanced manufacturing.
Enclosure enthusiasts-and new prospects
Most leading-edge wafer fab operators view minienvironments as a fact of life, especially on the front-end of tools where they are nearly universal in one configuration or another. They are also starting to be widely applied on back-end assembly and test areas as these become more sensitive to contamination. “I'm not sure it's possible to build a 300-mm fab without minienvironments, and it's increasingly difficult without them in 200-mm fabs,” says Goldstein, who also serves as a CleanRooms Editorial Advisory Board member.
But the vast majority of the world's fabs are 10 to 15 years old and still using 150-mm tool sets in ISO Class 5 cleanrooms. While they continue to operate for at least another five years, these prospective users are struggling to introduce minienvironments into existing cleanrooms that were never designed to accommodate them. “Most people now understand you can't just slap minienvironments in existing fabs and go about your business. They require extensive integration, including things often overlooked like sprinkler head placement and maintenance access,” says Goldstein.
Although retrofitting can be difficult, the reward is worth it in many cases: higher yields and a longer lifespan for the fab and its equipment. That's why minienvironment vendor Asyst Technologies Inc. (Fremont, Calif.) is targeting these users with what it calls “adaptive” solutions. “We see a strong demand for adapting existing fabs with minienvironments,” says Norma Riley, vice president and general manager of Asyst's interface division. Another growing minienvironment user group includes equipment and component manufacturers, adds Riley.
null
|
Yet some high-tech electronics manufacturers will never have a need for the technology. Many ISO Class 7 and 8 cleanroom applications simply don't require that level of contamination control; others don't have the production volume to warrant it. Apart from these application issues, the biggest hold on growth of minienvironments presently is the continuing high-tech recession and dearth of new fab starts.
In the pharmaceutical world, approximately 10 percent of global aseptic filling lines use isolators. In 1998, there were 84 such lines; in 2000, 172. Today, there are 191 lines, according to Jack Lysfjord, vice president of consulting, Valicare division of Bosch Packaging (Minneapolis, Minn.). “The technology has many systems in operation, it has passed the rigors of offshore and U.S. inspectors, and we are moving forward with gaining knowledge and confidence,” says Lysfjord, who conducts biennial surveys on the topic.
Drugmakers that have embraced the technology have multiple isolators. Baxter International Inc. (Deerfield, Ill.) has the most with 22 isolators; Aventis Pharmaceuticals (Strasbourg, France; Bridgewater, N.J.) and Pharmacia Corp. (Peapack, N.J.) have 12 and 11 respectively. While growth in Europe and North America has slowed somewhat over the past two years, use is growing rapidly in Japan, where the majority of new filling lines now incorporate isolators, says Lysfjord.
Isolators moved from sterility testing to aseptic filling lines in the past decade, but fueling today's growth are containment isolators for potent compound applications. An increasing number of drugs produced aseptically are now so potent that operators must also be protected from contamination. Vaccine makers, in particular, are turning to the technology to ensure asepsis and safe removal of thimerosal, a mercury-containing preservative.
Potent compound applications present a challenge to isolator makers who must design a unit that satisfies both conditions.
Carlisle Life Sciences (New Lisbon, Wis.) came up with one solution to this dilemma by engineering a positive pressure isolator that could be decontaminated with hydrogen peroxide vapors to protect the product. “Then we put in controls that would sense a leak and immediately create a negative pressure to protect the operator,” says Gary Partington, technical sales representative. Several of these units, which target aseptic/potent powder applications, include programmable logic controllers to maintain strict environmental requirements and are operational today.
One problem some isolator manufacturers have with delivering workable solutions to containment problems is that they do not understand the process taking place inside the isolator and are not able to deliver an ergonomically integrated product to the end user, says Hank Rahe of EnGuard Systems (Indianapolis, Ind.). As understanding of these processes increases, applications in this area will explode, says Rahe.
Still, as in the microelectronics world, there are pharmaceutical manufacturers who have yet to implement a single isolator—whether out of concern over tougher regulatory scrutiny or lack of time and resources. Latecomers to enclosures will find that industry working groups have been busy over the past few years developing guidelines that will help jumpstart future implementation.
Educating the user base
The most recent guidelines include ISO 14644-7, which covers all enclosures; Parenteral Drug Association (PDA; Bethesda, Md.) Technical Report (TR)-34, “Design and Validation of Isolator Systems for the Manufacturing and Testing of Health Care Products,” which focuses on health care issues; and the IEST Recommended Practice (RP) CC-028.1 on minienvironments for microelectronics and similar applications, including medical devices.
Because it addresses the wide range of enclosures available to the industry—clean air hoods, glove boxes, isolators and minienvironments—and targets the international user community, the 48-page ISO document is the most general of the guidelines. It defines what it terms “separative devices” and covers issues such as design and construction, material ingress and egress, personnel interface, installation and maintenance, support services, testing and certification. A final international vote on ISO 14644-7 was scheduled for year-end 2002.
The working group that crafted language for the document included representatives from both the microelectronics and pharmaceutical industries. The two groups at first thought they had no common ground, says Richard A. Matthews, chairman of the International Organization for Standardization (ISO, Geneva) Technical Committee (TC)-209, “Cleanrooms and Associated Clean Environments.”
null
“Each group was sure their technology was applicable only to their field, but a breakthrough came when they realized they had a common problem. They were working with a cleanroom that doesn't have any people in it,” says Matthews. “Once they got over that hurdle, they realized there were common denominators, such as the need for transfer devices and a pressurization scheme.”
New users may want to start with the generic ISO guideline, then turn to documents that address industry-specific needs. PDA's TR-34 gives pharmaceutical users direction on implementation of isolators for a broad range of manufacturing, development, testing and potent compound applications. The technical report also clarifies variables such as isolator airflow rates, leakage and decontamination versus sterilization.
“In my opinion, TR-34 remains the most useful and pragmatic guideline in terms of implementation of isolation technology. None of the regulatory authorities in the world have offered such a comprehensive practical document,” says James Akers of Akers, Kennedy and Associates (Kansas City, Mo.), who along with Agalloco co-chaired the TR-34 committee. “It's a shame that the U.S. Food and Drug Administration (FDA) and industry haven't standardized the recommendations in TR-34 because if they did so, it would result in a generalized common understanding of what is necessary in a practical sense to have a well-designed and well-validated isolation system in use,” he adds.
Six years in the making, IEST RP-028.1 describes the various uses, configurations and benefits of minienvironments and offers a checklist of parameters to consider when designing and implementing them. “It gives users things to worry about, but it doesn't tell them how to do it,” says Goldstein. Sections include applications and concepts, planning, design considerations, and evaluation and testing.
Lessons learned and challenges ahead
These guidelines help advance understanding of enclosure technology by tapping into more than a decade's worth of lessons learned. For isolators, a major insight involved leaks. “There used to be an idea that an isolator could never leak, but we learned that they have to leak. They can't be hermetically sealed,” says Agalloco. Also discarded was the assumption that airflow and air velocity paradigms for cleanrooms and isolators should be similar; they are not. “A million questions have been answered. We've gotten some sanity” on the topic, adds Agalloco.
Improvements in materials of construction have led to more efficient decontamination methods, but the question of whether to sterilize or decontaminate an isolator has yet to be resolved. “The regulators will accept decontamination, but companies are loath to give up the idea that they should sterilize the interior,” says Agalloco. Another challenge lies in how to decontaminate an isolator used for potent compounds so that the line can be used for other products down the road. “We have much more to learn about the potency issue,” says Lysfjord.
There's also still the fear on the part of some pharmaceutical manufacturers that the FDA hasn't clearly supported isolation technology—a fear borne out in part by inconsistent inspection outcomes. The agency “damns by faint praise saying they like it, but then they place so many qualifiers on it you have to wonder whether they really mean it,” says Agalloco.
The qualifiers, which include concerns over aseptic technique, glove tears and the like, apply equally to cleanrooms, but some users think isolators get tougher scrutiny. In fact, the FDA's recently released draft update to the 1987 aseptic processing guide includes a section on isolators with the warning that users should not adopt a “false sense of security” with the technology. (See sidebar for a response from the FDA's Office of Pharmaceutical Science within the Center for Drug Evaluation and Research.)
This atmosphere of uncertainty has led to “a failure on behalf of the industry and regulatory community to set reasonable standards-standards that are attainable,” says Akers. “There's been a tendency to strive for a level of perfection that goes well beyond what would be expected in a typical cleanroom. The objective with isolators should be to effect a clear benefit in terms of process control over existing cleanrooms. The target with any technology shouldn't be perfection.”
Standardization on minienvironments for 300-mm tool sets is helping the chipmaking industry achieve full automation. “Setting standards on minienvironment load port configurations was a real breakthrough,” says Tom Garrett, who manages Intel's (Portland, Ore.) D1D program within the company's Logic Technology Development Group. Before standardization, every minienvironment had a load port at a different position; now load ports are set at the same height and location so tools can work with a variety of automation schemes and tool interfaces.
The adoption of standards has allowed both equipment and device makers to introduce a level of simplicity into the manufacturing process that improves performance and lowers costs, says Asyst's Riley. The next test lies ahead as 300-mm fabs begin ramping up to high-volume production.
“It's a challenge for those of us in the automation and minienvironment space to make sure all the components work together—that the equipment communicates automatically with other equipment, that hand-offs between automated material transport and loading systems work, and that this entire system is reliable, robust and clean,” says Riley. “The key to achieving that is more communication and more standards.”
The future of enclosures
Any discussion of the future of enclosure technologies includes the question of whether they will displace cleanrooms altogether, and at this stage of the technologies' maturity, there are still camps on both sides of the issue.
“I don't see isolators as a direct replacement for cleanrooms. Each has its appropriate use,” says Matthews. “You can build a nice ISO Class 7 cleanroom for about $200 per square foot and have total freedom of movement, whereas an isolator might cost $1,000 per square foot for a comparable performance with restricted access.”
Others see a long-term trend away from large cleanrooms and toward these small, sophisticated enclosures. “Minienvironments in wafer fabs have been so successful that we no longer need an ISO Class 3 or 4 fab,” says Goldstein.
In fact, some experts predict that in the next 15 years, fab cleanrooms will drift to ISO Class 8 or 9, while minienvironments and robotic equipment maintain contamination control where it counts—in and around the process chamber and the product itself. “This would be a significant paradigm shift if this comes to pass,” says Goldstein.
As minienvironments control ever-smaller particles, what worked in the past might not work in the future. “The physics of particle transfer change as you get down a magnitude lower, so the mechanics of minienvironment construction and design will have to change with it. For example, minienvironments will need hard seals to minimize particle infiltration, whereas in the past, all that was needed was a simple pressure differential,” says Goldstein.
Meanwhile, some pharmaceutical manufacturers can envision the day when regulators will actually demand isolators for certain drugs. Although they are still evolving, enclosure technologies are firmly entrenched in these disparate, ultra-clean industries. III
Cooney: CDER's OPS supports isolators, aims to spread message
The Office of Pharmaceutical Science (OPS) within the U.S. Food and Drug Administration's (FDA) Center for Drug Evaluation and Research (CDER) is without question a supporter of isolation technology, according to Peter Cooney, Ph.D., the office's associate director for new drug microbiology.
Making the rounds of industry seminars and discussion groups, Cooney has experienced first-hand the industry perception that the agency has tougher standards for isolators. “The OPS doesn't condone that,” says Cooney. “We don't want to institute more rigorous requirements simply because somebody installs an isolator. Further, I don't think it is appropriate at this time that the acceptance criteria are different for isolators than for conventional cleanrooms.”
It may not be appropriate, but it is understandable given the learning curve required for new technologies, says David Hussong, Ph.D., an OPS review microbiologist and captain in the Commissioned Corps of the United States Public Health Service.
“This technology has come a long way in 10 years and there have been changes to our understanding of it,” says Hussong. “That's typical of any new experience because not everybody learns at the same rate, but we are getting there.”
The OPS is the arm within CDER responsible for the scientific review of new and generic drug product applications. “We're lucky because we get all the standards and testing put before us so we can see where the industry is going,” says Hussong. “The downside is that we don't get our hands on these processes; it's the investigators who go to these firms.”
The FDA current good manufacturing practices (cGMP) initiative announced last summer may help get regulators and industry on the same page (see CleanRooms' December 2002 Special Report). In addition to a plan to bolster cGMPs with science-based policies and standards, the initiative also includes measures that will encourage use of the latest manufacturing innovations and better coordinate the product submission review program with the field inspection organization. The industry is optimistic that this “holistic” approach will yield more informed reviewers and investigators and more consistency in reviewing applications.
One of the intentions of the agency-wide initiative is to devise a better process for the approval of supplements containing new technology. The agency's actions will hopefully “alleviate some of the fears companies have that there might be problems after approval,” says Cooney.