Asset Management Automation Lean Manufacturing Management

Take Risk Out Of Full Scale Projects

Michelle Segrest | May 27, 2019

EPIC engineers run quality tests on a completed pilot plant system for an energy client. All photos courtesy of EPIC Systems

Pilot plants demonstrate the feasibility of proposed process technologies for real-world operations.

De-risking has always been the driving force behind pilot plants and that force is as strong, if not stronger, in today’s environment of ever-shorter product life cycles. For example, when a company’s research and development group has an idea for a new process technology, testing may be performed at a “glass scale.”

“This can mean making only one liter of a certain chemical, similar to the type of tests you do in a high school chemistry class,” according to John Schott P.E., president of EPIC Systems Inc., St. Louis (epicsysinc.com), a manufacturing and service company that designs, integrates, and builds pilot-plant solutions for manufacturers worldwide. “Let’s say all goes well. The tests show the technology works at that scale. But are they convinced enough to take the risk to build a plant for $150 million that produces 10,000 pounds of the material each day? In most cases, it’s just too big a jump to spend that kind of money and hope a plant works.”

A pilot plant is meant to show that a process technology can be made at a smaller industrial scale, or perhaps two evolutions of that scale, before a larger, full-scale production plant is built. 

“The money spent on a pilot plant always makes sense,” Schott said. “One of three things will happen. The company will find out what it thought would work really does. Or, second, it will discover something unexpected and can then find solutions and feel confident moving forward. Or, third, it can find things that cannot be remedied, which will halt the full-scale project.” In the end, he concluded, maybe the company spends $5 to $10 million on a pilot plant rather than $150 million on a full-scale operation.

“Pilot plants are the true workhorse of the process-development industry,” said David Edwards, P.E., vice president, sales and marketing for Zeton Inc., Burlington, Ontario, Canada (zeton.com), which completes laboratory, pilot-plant, demonstration-plant, and small modular-production-plant projects around the world. “As chemical or biological processes are scaled from the laboratory to production, pilot plants provide the first window into continuous, as opposed to batch, processing. It will often incorporate unreacted feed/product recycle so the mass balance can be closed.”

Companies are using this idea to experiment with everything from plasma technology to spinning-disc reactors, to reactive distillation units. “There are a lot of new technologies out there that companies are now really working to create scaled pilot plants for,” Schott said. “We’ve developed projects in just about every industry, including pharmaceutical, food and beverage, industrial chemicals, consumer products, petrochemicals, it’s all over the board. And [they’re] all for the fundamental purposes of de-risking the expense of building a full-scale facility that may not work. De-risking is the key word and the main reason companies build pilot plants.”

Many companies, however, will skip pilot plants and go straight to full-scale production. Schott and Edwards agree this approach is a mistake, especially with innovative, untested process technologies. 

“Often, the thinking is that the cost of a pilot plant is too high. But [investing] millions of dollars in a commercial-sized system that doesn’t work is ultimately more expensive,” Schott said. “Pilot plants are an excellent risk-mitigation strategy for process scale-up at a fraction of the investment.” 

There is a substantial amount of money on the line when start to scale a process from a laboratory setting to a commercial plant. “You might have reaction data and theoretical models describing every conceivable detail of the reaction taking place. Yet taking them from process to production scale is still a big unknown. Turning the unknowns into something manageable and scalable is the role of a pilot plant.”

Process equipment does not scale linearly, which, although it is an unnoticeable effect at the small scale, can have huge ramifications in a production plant, Schott explained. 

“Modeling the process can catch some of these changes, but there are still a lot of assumptions in even the most sophisticated of computer models,” he said. “Although it can be easy to let yourself think that a reactor is a reactor, whether it’s 50 milliliters or 5,000 gallons, there are some fundamental challenges that must be vetted out and overcome. An investment made in a pilot plant can demonstrate your process technology is ready for production scale. It saves money by catching design issues early on, and it increases the likelihood of securing funding for new technologies. A pilot plant will greatly reduce your risk and provide a feasible path to successful commercial production.”

This pilot plant system tests the process of carbon dioxide removal from flue gas at coal-based power plants.

Advantages 

A pilot plant allows companies to collect real-time data that can help to ensure a full-scale production plant runs properly. It allows experimentation with such items as inputs, outputs, and processing time for streamlining the process.

“Jumping from lab scale straight to full production can result in problematic mixing, product output, and slow or hard-to-control reactions,” Schott explained. “Since chemical processes don’t scale linearly, it can be difficult to predict how a full-scale commercial process will actually behave.” 

While some issues can be addressed in simulations, the physical version that runs in the real world will often still behave differently than the simulations predict. “Modeling every factor takes time and may be too complex,” Schott noted. “The realities of physical-system layout and equipment constraints also play a large role in pilot-plant design. Non-uniform concentration gradients can cause less-than-ideal behavior and can invalidate some assumptions.”

Many pilot plants are built to study catalysts and catalytic processes, in which one chemical compound or biological organism is transformed into another. “Catalyst performance tests are carried out in the pilot plant to determine yield and selectivity data, and the lifetime of the catalyst is measured under a variety of operating conditions,” Edwards added.

Argo Genesis partnered with EPIC Systems to take a newly formulated asphalt additive and scale it to a pilot plant level of production. EPIC designed and fabricated this two-phase batch-reaction-to-distillation pilot plant.

Trends

Pilot plants have been around for decades, but the ways they are built and how their process technologies are gauged for feasibility have changed significantly over time.

Among the latest trends are those associated with modular design, which make it easier and quicker to upscale equipment and ship it anywhere, regardless of location. “The offsite construction of modular pilot plants can shorten timelines by as much as 40%, especially on projects with other site work that needs to happen,” Schott explained. 

Modular design offers a number of  distinct advantages for most pilot-plant projects, including:

• Pilot modules are constructed in ideal shop conditions, indoors, which leads to better quality and faster completion times.

• Fabrication, assembly, and testing of the new system will not disrupt any current operations since it is built off-site. This also provides a layer of discretion.

• Off-site construction is also a great advantage when plant upgrades are simultaneously occurring. On-site improvements can be made in parallel with system construction, speeding up the entire project timeline.

• Components, layout, and system arrangement are simplified and organized by pilot-plant design experts, reducing overall construction costs, space requirements, and required plant upgrades.

When pilot plants first began to emerge, a stick-build process was used. “These projects were built piece by piece at a certain location,” Schott explained. “The single biggest reason customers now modularize is because of schedule compression. In a traditional construction, everything is sequential. Until a full-scale manufacturing facility is built, you can’t put in the first piece of process equipment. With modularization, you bring construction into parallel with the field work. The moment the building is constructed, trucks are ready to deliver fully assembled and tested process systems. It shaves literally weeks or months off the process schedule.” 

Another important pilot-plant trend has been made possible through the development of digital-twin technology. Companies such as the chemical giant DuPont have seen safety, environmental, and productivity improvements by creating digital twins of manufacturing processes and using them for pilot-plant fabrication. 

“Digital twins ensure fewer mistakes on implementation because engineers can visually see what’s going to happen,” said Tracy Clarke-Pringle, Ph.D, a process modeling and control consultant for DuPont. “For many processes, it is impossible to think through all the consequences of a change,” she explained. “A digital twin doesn’t need to be perfect to be useful, and it doesn’t take years to develop useful models, with the right tools.”

Clark-Pringle emphasized the incredible amount of learning that occurs simply by developing a digital twin. “This is often underestimated, and a hidden value of the twin,” she said. “As sites experience more turnover with operators, digital twins provide the gold standard for effective training. There is no other way for inexperienced operators to see and feel what emergency situations are like. By definition, these situations occur very rarely. You do not want their first experience with an emergency to be during the emergency itself. This is why pilots use flight simulators. The chemical industry needs to be better about training their operators in the same way.”

A chemical pilot plant scaled-up a proprietary ion exchange process for a prominent agricultural chemicals company.

Examples 

According to experts, pilot plants can be used to produce usable product, start building a market, and convince investors of long-term viability. While the process technologies are often innovative or new for their industry, the idea of a pilot plant is to test commercial equipment and figure out how to lower the production costs for whatever the pilot plant is producing. Here are several examples:

Gas Technology Institute, Des Plaines, IL (gti.energy) — a non-profit organization supported by the U.S. Department of Energy, Washington (energy.gov) — engineered and fabricated a pilot-plant system that tests the process of carbon dioxide removal from flue gas at coal-based power plants.

During the front-end engineering (FEE) process, it was discovered that the second stage of the process (recovery and regeneration of the CO2) could not be accomplished through the proposed membrane process. Instead, a more traditional thermodynamic separation process was implemented, which required changes to the process-flow diagram. 

The pilot project delivered a scalable process system for testing CO2 removal technology and avoided unnecessary delays through flexibility and engineering experience. It produced faster and safer skid fabrication and installation due to modular design and off-site fabrication.

Argo Genesis Chemical LLC is an agribusiness company that took a newly formulated asphalt additive and scaled it to a pilot-plant level of production. A two-phase batch-reaction-to-distillation processing plant was designed and fabricated. (Argo Genesis is a sister plant of Seneca Petroleum Co., Crestwood, IL. The two businesses are members of the Seneca Companies, headquartered in Des Moines (senacoco.com).

The pilot module for Argo Genesis represents an investment in commercializing biopolymer technology. The process scale-up was simulated with the help of Aspen Plus software. Proper demonstration-plant function was ensured by simulation before mechanical design and skid fabrication began. To reduce safety concerns and the cost of designing an indoor hazardous process, the modular plant was designed for an outdoor installation. The software is produced by Aspen Technology Inc., Bedford, MA (aspentech.com).

New biopolymer formulations were produced in 32 working hours, significantly speeding up the development process and creating a biopolymer demonstration plant that is the first of its kind. There is now a flexible process, allowing the testing of multiple formulas and proving functional for commercialization at a demonstration-plant scale.

EPIC Systems used the client-provided sequence of operations to build a modular chemical pilot plant. Based on a 3D model, the pilot plant was able to fit in the required small footprint. A highly corrosive acid was the main component of the system.

To meet the client’s industrial-grade requirements, a variety of atypical materials was used to fabricate the pilot plant. Hastelloy- and Teflon-lined lines and other piping specifications were used throughout, along with fiberglass tanks. PVC- and Teflon-lined piping required custom-built flanges due to tight tolerance specifications. The highly corrosive nature of the material, combined with the low-flow rate, demanded installation of unique flow meters.

The pilot system required measurement at many instrumentation points, including flow, temperature, pressure, and chemical level in the tanks. A bubbler system was installed to allow easier and constant measurement of the liquid level. Compressed air was released into the bottom of tanks and, based on the resulting level, the liquid level was calculated.

This pilot plant was the intermediate step before a large industrial, international process plant was built for this technology. The many sampling and instrumentation points allowed the company to collect the necessary data for a larger industrial-scale operation. 

An automotive-chemicals company developed a custom pilot plant that produces a petroleum additive. The company upgraded its system to an automated operation, featuring custom formulation capabilities for internal testing and end-customer formulation flexibility. Production was increased 60% through a customized automation-integration plan, new batch-management software, and an engineered approach to process scale-up.

Moving from manual to automated batching improved consistency of batches between runs, with better tracking of process parameters. Using the distributed-control system DeltaV, the automation software enables users to quickly gain access to recipe management and data collection running throughout the plant. DeltaV is a product of Emerson, St. Louis (emerson.com).

“This plant improves our client’s ability to provide custom formulations to their customers and increase internal testing of new formulations,” said project manager Matt Benz. “Higher production rates mean more product is run through the system in a day. A broader temperature range and key equipment updates drastically improved system efficiency. The range of products they are able to run through this one pilot skid increased greatly.” EP

Michelle Segrest is president of Navigate Content Inc., Gulf Shores, AL, and specializes in creating content for the processing industries. Contact her at michelle@navigatecontent.com

FEATURED VIDEO

CURRENT ISSUE

ABOUT THE AUTHOR

Michelle Segrest

Michelle Segrest

View Comments

Sign up for insights, trends, & developments in
  • Machinery Solutions
  • Maintenance & Reliability Solutions
  • Energy Efficiency
Return to top