A Modular Approach to Process Analysis

While elegant in concept, the change from manual, laboratory-based analysis to real-time, online instrumentation has long been problematic.

While it's hard to argue the potential gains in reduced variability and improved control promised by implementing what the FDA refers to as "Process Analytical Technologies" (PAT), wide adoption has been limited by the need for complex, expensive and custom-built systems for automatically extracting process samples and conditioning them for analysis.

Now, visionary process analytical chemists in the petrochemical and chemical process industries, spurred by the successful use of modular sampling systems and manifolds for the analysis of ultra-pure gases in the semiconductor industry, have launched a major effort to define and adopt a new generation of modular process analyzer sampling systems, sensors, connectivity (communication between the sensor and control computer) and the diagnostics and intelligence to optimize their performance.

Ultimately, the effort promises to package analytical technologies in modular, shoe-box-size black boxes that contain the sampling system as well as combinations of analytical instruments.

"It's a platform for implementing microanalytical instrumentation," explains Mel Koch, Ph.D., director of the Center for Process Analytical Chemistry (CPAC) at the University of Washington, Seattle. Microanalytical instrumentation implies the use of ever smaller instruments "and ever smaller sample sizes "to reduce waste and increase instrument response times. "As soon as you do that, you increase the capability of things that go on--in combinatorial chemistry, in process development, in optimization and control."

Formally created at the ISA Expo 2000 meeting of the Instrumentation, Systems, and Automation Society (ISA) of Research Triangle Park, N.C., the New Sampling/Sensor Initiative (NeSSI), is a vendor-neutral and non-affiliated effort that operates under the auspices of CPAC. Since that time, it has evolved and broadened into "an ad hoc industry initiative to drive permanent change in how we do process analytical chemistry," says Peter van Vuuren, ExxonMobil Chemical, Baytown, Tex.

Their plan: to facilitate three generations of development and evaluation of modular sampling systems, associated software and plug-and-play hardware, open standards for communication and hardware for miniaturized analytical instrumentation.

Exactly when the concept of a modular sampling system first originated is hard to pin down. Work that started as early as 1994 and resulted in a patent for a "Modular Sample Conditioning System" may represent the beginnings. (U.S. Patent 5,841,036, was filed in August 1996 and issued to Don Mayeaux, president, A+ Corporation, Prairieville, La., in November 1998.)

However, it appears that the industry-wide initiative on the need for new approaches to sampling started within the Oil and Petrochemical Industry Focus Group at CPAC.

"There was a consensus that the sampling problem was not unique to the petrochemical industry," recalls CPAC director Koch, who formerly had global responsibility for analytical sciences at Dow Chemical. The group realized that sampling was a concern in the food and consumer products, pharmaceutical, biotech and the chemical process industries, and that there was a need for projects in sampling and sampling systems, he says.

"The CPAC group wanted to develop a concept for a flexible and smart sampling system that would be low-maintenance, require lower volumes--smaller samples--and be amenable to microanalytical devices," elaborates consultant Jim Tatera of Tatera & Associates, Inc., Madison, Ind. "The modular approach really took off when people piggybacked the concept with the ISA's SP76 standard for a modular sampling interface, allowing it to move forward with some flexibility and interchangeability."

As a tool for process analytical technology, NeSSI clearly falls under FDA's process analytical technology (PAT) initiative. "I think it has a valuable place in the manufacture of drug substance," says Ajaz Hussain, deputy director of FDA's Office of Pharmaceutical Science and chair of the agency's PAT subcommittee.

And despite the FDA's professed support and the technology's demonstrated potential, Hussain recognizes that regulatory concerns will likely slow the implementation of NeSSI-based technologies. "The PAT initiative was designed to essentially remove perceptions that regulatory agencies won' understand this new technology and remove the [regulatory] hurdles for bringing it in [to new and existing processes]," he says.

In addition, the research exemption or safe harbor concept (for allowing the testing of new technologies in a manufacturing context) should facilitate implementation of PAT-related technologies, notes Paul David, associate fellow in chemical process R&D for Pfizer in Skokie, Ill. "Industry must be allowed to learn from the results of NeSSI-based measurements without the threat of FDA enforcement action--assuming there is no risk to the safety and efficacy of the product," David says.

Inspector training should help, too. "FDA people who will review or inspect PAT-based applications will have specialized training and certification," Hussain says. "PAT-based applications won't be handled in a routine way."

"Our philosophy is to minimize the variability of a chemical process during development and scale-up so that expensive continuous monitoring is not needed in commercial production," Pfizer's Paul David says. "We are interested in the NeSSI concept, especially the introduction of microanalytical instrumentation for these systems, and can envision using several of these devices as sensors."