How To Reach C&ENACS Membership Number
Visit SGI


July 21, 2003
Volume 81, Number 29
CENEAR 81 29 pp. 14-17
ISSN 0009-2347

Combinatorial screening methods are catching on for improving catalysts used in the manufacture of bulk chemicals


Combinatorial chemistry--a term applied to high-throughput methods of creating and screening large numbers of compounds simultaneously for particular properties--is gaining traction in catalysis for large-volume industrial chemicals. It is becoming like a catalyst itself, making the process of catalyst discovery less labor intensive and more selective.


SETTING UP An array of hundreds of reagents is being prepared for a batch screening program of a catalyst library at Avantium's laboratory in Amsterdam. AVANTIUM PHOTO

Combinatorial methods have long been used to synthesize and screen pharmaceutical compounds. They have also been making strides in finding new polymer catalysts. For example, Symyx Technologies and Dow Chemical recently unveiled a new class of polyolefin catalysts that were developed through a four-year combinatorial chemistry collaboration (C&EN, April 7, page 10).

But in the more conservative industrial chemicals industry, where the priority is not discovering new products but improving the production of old ones, combinatorial chemistry is just beginning to yield successes, says John J. Murphy, a consultant with the Catalyst Group, Spring House, Pa. "There has been a lot of testing and evaluation over the past few years, but we are now beginning to see some of the fruits of the labor," he says.

Finding catalysts for bulk chemical processes can be a complicated task because most of the catalysts in use are heterogeneous--solids that work with liquid or gaseous reagents. Homogeneous catalysts, in contrast, work in solution. Some chemical processes, such as ExxonMobil Chemical's oxo process for making alcohols, do use homogeneous catalysts.

Murphy says combinatorial chemistry is more commercially advanced for homogeneous catalysts than it is for heterogeneous catalysts largely because high-throughput methods for screening homogeneous systems have been used in pharmaceuticals for some time. He now detects growing interest in combinatorial chemistry for heterogeneous catalysis.

Andreas Mader, chief executive officer of Zurich-based Thales Technologies, explains that there is a technical reason why the development of combinatorial chemistry for the two types of catalysts differs. "Homogeneous is better developed because it is easier to explore and to understand than heterogeneous," he says. However, this complexity makes heterogeneous catalysis a promising market for combinatorial chemistry companies because they have an opportunity to add real value for clients who are outsourcing catalysis R&D. "It makes it more attractive because there is a lot more to improve on," Mader notes.

SOLIDS ARE MORE difficult to work with than liquids, points out Jennifer Holmgren, director of exploratory and fundamentals research at UOP. "There is a lot of treatment and preparation of the solids to get them ready for testing," because preparation has a dramatic effect on a catalyst's properties, she says. "I can make a 0.5% platinum-on-alumina catalyst five different ways and get a different catalytic result each time. There is a tremendous amount of complexity in the creation of the heterogeneous catalyst itself."


LITTLE PLANTS Avantium's nanoflow reactors are loaded with catalysts and connected, 64 at a time, to a gas manifold. AVANTIUM PHOTO

In fact, Murphy says, because of the larger number of combinations usually possible in heterogeneous catalysis, combinatorial chemistry may be a necessity for catalyst discovery in bulk chemicals. "The experimental space is enormous. You would not get to look at all the combinations without high-throughput screening," he says.

Moreover, homogeneous catalysts go into fewer processes than heterogeneous catalysts, which can be used in fixed-bed, fluidized-bed, slurry, and many other reaction technologies, according to Troy J. Campione, vice president of collaborations and strategic alliances at Symyx. "There are fewer variables and fewer obstacles to overcome with homogeneous catalysts than in heterogeneous," he says. "It is harder to make a one-reactor-fits-all system in heterogeneous catalysis than it is in homogeneous catalysis." He notes that after initial catalyst screening steps that are universal for all catalyst types, Symyx uses fixed-bed reactors in a secondary screening.

Holmgren says that Torial, a combinatorial chemistry business focusing on heterogeneous catalysis that UOP launched last year, tries to simulate commercial reaction conditions as best it can. "We work very hard to create scalable and realistic systems so they can go from combi to large-scale very quickly and not have to redo everything on the lab scale," she says.

Other companies that focus on heterogeneous catalysis do the same. Heidelberg, Germany-based hte--the name stands for high-throughput experimentation--has heterogeneous catalysis as its primary focus, says Wolfgang Strehlau, the company's business development manager. "Our approach has been derived from classical chemical process engineering," he says. "We are looking at classical processes and miniaturizing and parallelizing them." This applies, he says, to both the first and second stages of experimentation.

Strehlau points to hte's work in automotive catalytic converters, a heterogeneous catalyst system in which a washcoat of a metal oxide is placed on a ceramic substrate. He says hte makes small amounts of the washcoat and applies them to tiny substrates for parallel tests.

Gert-Jan Gruter, vice president of chemical technology at Amsterdam-based Avantium Technologies, says his company has a similar approach. He is critical of methods used by companies like Symyx that apply a simple screening method across a range of processes. "Our philosophy is that we don't start at that level, because we think that once you find something interesting, it is extremely difficult to translate it to a real heterogeneous particle with a defined pore structure," he says. "In our approach, rather than run as many experiments as possible, we try to minimize the number of experiments by developing models for correlating structure with performance."

However, Symyx has secondary screening steps that are customized to particular processes.

Thales's Mader says his company sorts through large numbers of catalysts using a mathematical model that simulates reactions, shrinking the need to synthesize actual catalysts from, say, 1,000 down to 10 or 15. "When we do synthesize, it matches exactly with what we do with the model," he claims. For example, using this method, Thales found a polymerization catalyst with improved kinetic properties over an existing catalyst.

But no matter what the method is, combinatorial chemistry all boils down to saving time. And because combichem can decrease the time needed for catalyst discovery by orders of magnitude, it is leading some chemical companies to try injecting new life into catalytic processes they may not have otherwise tinkered with.

Campione says combinatorial chemistry is often 10 to 50 times more efficient than conventional research methods and is sometimes 100 to 1,000 times more efficient. For example, he says that, using combinatorial chemistry, a team of three to four Symyx scientists can conduct 100 to 200 polymerizations per day and analyze the results. The same 100 polymerizations using conventional methods would take about 20 people, and the analysis probably wouldn't come back the same day or even the same week.

"What combichem enables is breakthroughs--the really long term things that no one has the stomach to fund anymore."

SIMILARLY, UOP's Holmgren says Torial prepared and tested 512 catalysts in five weeks in its search for a new paraffin isomerization catalyst for a UOP process, coming up with multiple leads. On the same project, she says, it took a single researcher three years to test 271 catalysts without even finding a catalyst that was better than the existing state-of-the-art catalyst.


SEA OF GREEN Catalyst samples are displayed for inspection and storage after catalytic testing in a Torial reactor assay module. UOP/SINTEF PHOTO

This kind of savings is sparking interest in improving catalysis for chemical processes, Avantium's Gruter says. "The methods that are now available to screen catalysts in a systematic way have given new interest in trying to improve catalysts that have not been looked at for the past five to 10 years," he says.

Holmgren says combinatorial chemistry makes innovative research on chemical process catalysts accessible for companies that don't want to spend too much time on risky projects. "What combichem enables is breakthroughs--the really long term things that no one has the stomach to fund anymore," she says. "Twenty years ago, the chemical industry was investing heavily in very difficult problems. Now you find that the investment is not there anymore. Now, all of a sudden, the rate of breakthrough innovation is really going to increase in our industry."

Symyx's Campione agrees. "Management has limited patience with which they will wait for success. You can maybe come up with a breakthrough in 10 years, but no one will wait for you to do that."

Developing catalysts that allow the use of cheaper feedstocks--like alkanes for processes that traditionally use alkenes--is one such project. Campione says Symyx has discovered catalysts that can make acrylic acid from propane instead of from propylene and acetic acid from ethane instead of ethylene, both reactions that chemical companies have pursued unsuccessfully for years. "They are materials that show much greater activity and selectivity than what's been discovered before," he says.

Working with Celanese, and taking up where an old Union Carbide program left off, Symyx has found catalysts for the oxo dehydrogenation of ethane to ethylene that Campione says could be put into a new process that would be competitive with the steam cracking of ethane. Such a process would be subject to the typical risks associated with an entirely new approach, but he argues that the risk is offset by much lower capital costs.

Symyx isn't alone. Gruter says Avantium is working on the direct oxidation of propane. However, he says most projects have more modest aims. "The majority of our customer projects are improving existing processes using the same feedstocks they already use," he says. "It is usually increasing catalyst lifetime, increasing selectivity, and making the process simpler to operate."

Indeed, according to Murphy, even though combinatorial chemistry can reduce risks for ambitious projects, chemical companies are more receptive to modest ideas. "If the sales pitch is for 'blue sky' R&D, your audience is one out of 10; if it is making process or product improvements, it is eight out of 10," he says.

For example, Avantium was asked to improve on a customer's selective hydrogenation catalyst. The catalyst had a problem: As it lost activity over time, the customer had to increase temperatures to maintain 100% conversion at the expense of selectivity. "We had to screen catalysts not only for selectivity but also for lifetime and for a minimum selectivity loss when you increase temperature," Gruter says.


HIGH VOLUME Symyx uses microfluidic parallel reactors for primary catalyst screening. SYMYX PHOTO

BECAUSE OF THE high volumes involved, combinatorial chemistry companies view petrochemicals as a more promising market for their techniques than fine, or even specialty, chemicals. Campione says Symyx has all but abandoned its research collaboration business in fine chemicals because the applications are small. "There wasn't a sustaining research program that would allow us to run our business efficiently," he says.

Moreover, he says manufacturing is not as big a part of the cost for fine chemicals--which are typically going into pharmaceuticals--as it is for petrochemicals. In heterogeneous catalysis for petrochemicals, Symyx is collaborating with BP, ExxonMobil, and Celanese, among others.

"We tend to gravitate toward the big commodity players, because that is where the biggest opportunities are," Campione says. "In commodities, once you get one breakthrough catalyst, you have paid for all the research because it is worth so much money." Despite all its efforts, Symyx still hasn't come out with a commercial chemical catalyst, although some are in the pilot phase.

Avantium's Gruter says the bulk of his firm's efforts are in similar programs. "From an intellectual interest point of view, commodities and specialties are equal," he says. "But from a project-size perspective, there is much more work for us in the commodity area. The reason is that in specialties the volumes are relatively small and the R&D budgets are also relatively small. In larger volume projects, there is much more budget available to do a serious attempt to try to improve the catalyst technology."

However, not every competitor has the same priorities. Thales's Mader says his company started out doing only polymer chemistry but is now branching off into pharmaceuticals--in terms of finding both ingredients and the catalysts to make them--and agrochemicals.

Despite combinatorial chemistry's potential, UOP's Holmgren says she still encounters a lot of skepticism from the conservative chemical industry because of the uniqueness of heterogeneous catalyst processes. "Working with materials is not an easy thing, and a lot of people find it difficult to accept that you can still achieve the same results combinatorially that you would with traditional methods," she says.

Duncan Akporiaye, Torial's chief technology officer and supervisor of combinatorial chemistry for Norway's Sintef, says his firm has an antidote to this skepticism through its association with UOP, a successful licensor of many heterogeneous-catalyst-based processes. "UOP is very well known in the arena, and it gives us a comfort factor and security with customers," he says.

But hte's Strehlau expects that the skepticism will fade universally and that chemical makers will start turning to combinatorial chemistry for help in catalyst development. "We have developed an infrastructure for making catalysts and for evaluation," he says. "For us, a key driver is the chemical industry realizing that it must come up with innovations or new products because otherwise it will only become a commodity business."


Chemical & Engineering News
Copyright © 2003 American Chemical Society

Visit Eastman

Cover Story

Combinatorial screening methods are catching on for improving catalysts used in the manufacture of bulk chemicals

Gas-to-liquid fuels are becoming a reality and looming as a robust market for catalysts

Related Story
[C&EN, April 7, 2003]
Related Sites
E-mail this article to a friend
Print this article
E-mail the editor

Home | Table of Contents | Today's Headlines | Business | Government & Policy | Science & Technology |
About C&EN | How To Reach Us | How to Advertise | Editorial Calendar | Email Webmaster

Chemical & Engineering News
Copyright © 2003 American Chemical Society. All rights reserved.
• (202) 872-4600 • (800) 227-5558

CASChemPortChemCenterPubs Page