Hundreds of chemists can’t possibly share a 450-sq-ft lab—unless, that is, they don’t have to be there. Eli Lilly & Co. designed its Automated Synthesis Lab around this idea. Logging in from their desks, Lilly drug discovery chemists and collaborators around the world can remotely set up and run their syntheses at the Indianapolis-based ASL.
The lab works by integrating multiple reactors, including flow hydrogenation and microwave units. The robotically operated reactors are linked to automated reaction workup and analysis bays. Users can ask the system to sample reaction vessels for testing and photograph them to look for changes. The ASL is designed to make 100-mg samples, which are large enough for early-stage discovery work. Products can be stored, sent to the owner, used in another reaction, or passed along for biological testing.
In any given month, up to 40 chemists access the ASL, says Scott Sheehan, the firm’s senior director for discovery chemistry research and technology. The lab employs two engineers to maintain the equipment and a small staff of chemists to load reagents and off-load products. The lab has run more than 24,000 reactions across a range of reaction types over the past three years. “Our longest linear synthesis in the ASL has been eight chemical steps, but on average the syntheses are between three and four steps,” Sheehan says.
Enabling chemists to run their routine chemical transformations remotely in the ASL frees up time for more challenging work, he explains. “And we continue to expand what we consider to be routine for our automation to run.” Additions might include photochemistry and small-scale optimization methods such as microfluidic reactors.
“We measure the value of automation not by the number of compounds it turns out, but by the diversity of the compounds,” Sheehan says. Some 70% of the molecules made in the ASL fit this bill.
Ten years ago, the ASL wouldn’t have been possible because the equipment didn’t exist. For many chemists, the Lilly lab is still a dream, but benchtop synthesis equipment with varying degrees of automation is making inroads into industrial and academic labs. Suppliers of microwave synthesis and flow chemistry instruments in particular hope to see new generations of chemists become more familiar with their lab tools.
Using any new technology takes some adjustment, Sheehan acknowledges. “When chemists started to realize that these instruments weren’t designed to compete with them, but to augment their capabilities and provide them with more time to focus on judgment-based, value-added activities, they really embraced the idea.”
Microwave synthesis began in the mid-1980s, sometimes in home microwaves, says Grace S. Vanier, a product manager in the life sciences division of the microwave equipment firm CEM Corp. The company’s first machines were for drying and materials testing uses, but by the early 2000s the firm had expanded into organic synthesis. Interest has continued to grow as users explore the capabilities of microwave-assisted chemistry (C&EN, Sept. 24, page 32).
Microwave chemistry makes sense for any reaction that will benefit from heating. Benchtop synthesis units focus microwave energy on milliliter-sized samples to heat them evenly and efficiently. High temperatures, often above a solvent’s boiling point, and high pressures generated inside sealed vials can accelerate reactions that once took hours so that they take only minutes, while helping them proceed cleanly with fewer side reactions.
“The payoff for the chemist is not only that it’s faster, but they can iterate much more quickly,” says John Urh, Americas market manager at Biotage, a lab equipment firm based in Sweden. “You can load vials in an autosampler and run different sets of conditions, quickly do a mass spec on those, and in an hour or so get an answer that would otherwise take weeks to get.”
Lab-scale microwave units can sell for $10,000 and up, according to manufacturers. But unlike cheaper home microwaves, these systems have the power, temperature, and pressure controls needed to run reactions reproducibly, Vanier explains. “A domestic unit also doesn’t have the safety features.”
Although pharmaceutical, fine chemicals, and contract research firms have been leading customers, microwave systems are starting to be found in specialty chemical, nanomaterial, and petrochemical labs. The Los Angeles-based market research firm Strategic Directions International estimates that the microwave-assisted chemistry market is worth $133 million annually and will grow about 4.5% this year.
Customers want devices that offer a high return on investment by providing answers quickly, Vanier says. “Our focus has been on how we can make microwave-assisted chemistry easier to use while expanding what they can do with it.”
Soon after the first benchtop units appeared, UpScale Microwave, formerly known as Accel Synthesis, decided to offer larger reactors. “Many of us realized that we needed a scale-up option,” says Richard Wagner, who had implemented several new synthetic technologies in R&D and manufacturing while a manager at GlaxoSmithKline. He is now UpScale’s vice president of chemistry and business development.
Initially, UpScale developed a floor unit that could handle reaction sizes up to 10 L. But feedback from process chemists led it to design a benchtop unit that handles 50-mL to 2-L reactions and a production-scale reactor, Wagner explains. Called the R2P-1, the benchtop reactor was launched this summer.
University of Connecticut chemistry professor Nicholas E. Leadbeater demonstrated that scale-up between UpScale’s reactors is possible using identical conditions (Org. Process Res. Dev., DOI: 10.1021/op900287j). The reactors use standard glassware and prepressurize the reaction chamber, rather than relying on heating inside closed vials.
Austrian equipment maker Anton Paar, which entered the microwave synthesis market in 2004, has developed benchtop units in collaboration with the Christian Doppler Laboratory for Microwave Chemistry at the University of Graz.
Indeed, academia, like industry, is interested in the rapid synthesis and optimization capabilities of microwaves in research labs. But many students are still taught decades-old synthesis methods and aren’t learning to use the equipment they’ll see in industrial settings, Biotage’s Urh says.
Some schools, such as Allegheny College in Meadville, Pa., are including the technology in teaching chemistry, Urh says. Rather than spend their three-hour organic chemistry lab completing one cookbook recipe, students use microwave-assisted methods to rapidly run multistep reactions under a variety of conditions. “It allows students to more closely mimic the reality of a chemist’s job,” he says.
Peptide synthesis is one chemistry drudge job that microwave synthesis can ease. Instrument suppliers are finding a growing pool of chemists interested in peptides and have increased their focus on this market area, which they say is about a quarter of the overall microwave synthesis market. Offerings range from simple manual devices to fully automated, parallel units that can make multiple peptides at once and cost upward of $100,000.
Although microwave reactors can reach 300 °C, heating to a mere 75 °C can bring tremendous benefits in multistep solid-phase peptide synthesis. Along with significantly faster reactions, microwaves improve the efficiency of deprotection and coupling steps by driving reactions toward completion, which results in higher purity and yields.
CEM introduced the first automated microwave peptide synthesizer in 2003, according to Vanier. Its Liberty instrument now comes in single- or 12-channel models. In mid-2011, the company launched the semiautomated Discover SPS Plus, which customers can upgrade to the automated version.
Biotage began making dedicated microwave peptide synthesizers after discovering that customers had been using its microwave units for manual peptide synthesis for about a decade, explains Amit Mehrotra, Biotage’s peptide product manager. About a year ago, the company launched the Initiator+ SP Wave semiautomated unit, followed in September by the fully automated Initiator+ Alstra.
Along with computer control, some microwave suppliers offer robotic sample handling and flow chemistry options for peptide and general organic synthesis. Sweden’s WaveCraft is working with the England-based flow chemistry firm Uniqsis to develop continuous-flow, microwave-assisted organic synthesis systems. Uppsala University and WaveCraft researchers recently evaluated a prototype (Org. Process Res. Dev., DOI: 10.1021/op300003b), and the firms introduced the ArrheniusOneFlow system in May.
Researchers use microwave heating widely at the lab scale, but it appears less in production because of the challenge in effectively irradiating large batch reaction volumes. Microwaves can, however, penetrate and uniformly heat the small volumes in the micro- and mesoscale channels that are at the core of lab-scale flow chemistry reactors.
Although chemists can combine microwave and flow technologies, microreactor-based flow chemistry systems can produce high temperatures and pressures without requiring microwave heating, points out chemist Andrew Mansfield. “By applying higher back pressure, you can superheat your reaction solvents,” he says.
A former pharma industry scientist, Mansfield founded Flow Chemistry Solutions in 2011 to provide applications and consulting expertise. Several instrument suppliers also play in the emerging flow chemistry market.
The precise regulation possible in miniaturized flow systems extends many benefits, flow chemistry experts say. “The big promise of flow chemistry is that you can attempt reactions that you’d never be able to do in a flask,” says David Griffin, application engineer at Vapourtec in England. The flow chemistry company has contracted with Mansfield’s firm to develop applications.
Operating in flow mode at tiny volumes allows for the control of many parameters, including temperature and stoichiometry, which translates into reaction speed, efficiency, and safety, Mansfield explains. Particularly amenable are transformations that are exothermic, produce gases, and use or produce hazardous compounds. Highly reactive intermediates or unstable catalytic species can be generated and immediately reacted in flow mode.
Flow reactions also are scalable and reproducible because all reactants see the same conditions. “If you increase the size of a round-bottom flask, you are not going to have efficient mixing, and you’ll get temperature and concentration hot spots,” Mansfield says. “In small capillaries you get precise mixing, and the high surface area dissipates any heat that builds up. Or, conversely, you can control the temperature without having a temperature gradient.”
Many energetic reactions are run in batch mode at cryogenic temperatures to control them, Griffin notes. In a flow system, with its implicit control, these same reactions can be safely run at higher temperatures to speed them up. Reactions involving volatile substances or gases that would have to be run under high pressures also are not challenging for flow systems, he adds.
Safety concerns about explosive or runaway reactions prevent some chemists from conducting them at large scale. “A rig capable of making a kilogram per day in batch might be big enough that if something happens it would be a terrible incident,” Griffin says. In a flow reactor, only a small amount of material is in transit at any time, and the consequences of a mishap are easily managed. “And if you want to make 5 kg, it doesn’t mean a bigger machine, it doesn’t mean more risk, it just means running the reactor longer,” he adds.
The ability to handle these difficult-to-scale or hazardous chemistries has made microreactor technology more attractive in recent years. “Flow chemistry has become more accepted,” Mansfield says. “It is somewhat analogous to microwave applications, which was a very niche market for a long time.” Microreactor technology has become the subject of recent special issues in scientific journals, and Switzerland-based Flow Chemistry Society emerged in 2010 with a dedicated journal.
Sales of microreactor modules that go into lab systems have more than doubled in the past five years, according to the French consulting firm Yole Développement. In 2012, sales of the modules, which represent about 20% of the cost of a microreactor system, are expected to reach about $18 million. Yole expects the R&D market to grow a bit more but then level off by 2016, while sales of production systems are forecast to more than triple.
“Ten years ago you couldn’t buy a flow chemistry system for your lab; they just didn’t exist,” says Mike Hawes, chief executive officer of England-based Syrris, an 11-year-old company that makes lab-scale batch and flow reactors. Today, Syrris and other flow reactor providers point to a customer base that has grown in academia and across industries including pharma, fine chemicals, petrochemicals, food, and nanomaterials.
Flow chemistry appeals to process chemists wanting to generate data on reactions and optimize them, Hawes explains. Rather than running multiple individual reactions, chemist can test, and even automate, a range of reaction conditions using a flow system. Integrating the reactor with a sample collector and analytical apparatus yields both reaction products and informative results. Although the same approach could be taken in microtiter plates using parallel synthesis, the range of reactions may be more limited.
But equipment makers want to reach more than just the process chemist. “Our products are modular, and customers can buy different levels of functionality,” Hawes says. “A basic kit might be feasible for an academic or someone with a modest budget, while adding modules can take you all the way to the most advanced automated system.” Syrris’ sophisticated Africa system, launched in 2004, and the more compact Asia model, which debuted in 2011, use similar reactor types.
Chemtrix sells a fully automated, glass microreactor system, called the Labtrix S1. The unit is popular with process chemists who are “more versed in using design-of-experiment methods and having a more hands-off approach,” Chief Technology Officer Charlotte Wiles says. Younger researchers also tend to like the automation aspects.
After feedback from other scientists who indicated they’d like a device that kept them more in touch with their chemistry, the Netherlands-based firm launched a manual system, the Labtrix Start, in 2010. It uses the same microreactor chips and has the same flow rate, temperature, and pressure ranges as the automated system, but it costs significantly less.
“It is targeted at academic and early-stage researchers who are just starting out in the field and want to see what flow chemistry can do for them,” Wiles says. Another way suppliers help researchers test the waters is by renting out equipment. Although flow chemistry has become more widespread, “a great number of chemists have never experienced it,” she adds, noting that adoption has been slower in the U.S. than in Europe.
Flow Chemistry Solutions’ Mansfield says that a number of flow chemistry firms are “stripping down” their flagship apparatus to lower prices and make the instruments more accessible to academic teaching labs. In addition to the Labtrix Start from Chemtrix, FutureChemistry in the Netherlands has the FlowStart Evo.
Vapourtec has an easy-Scholar version of its three-month-old E-Series system, Griffin says. It and other versions selling for less than $20,000 target polymer and medicinal chemists as well. “They are not process chemists and don’t want to spend a lot of time playing with a piece of equipment,” he says. “If they are going to try out a flow process, it has to be simple, cheap, and pretty bulletproof because they will throw things in that they don’t even know will react.” Vapourtec also sells an expandable, fully automated R-Series system.
Many firms offer training courses and teaching materials to instruct about flow chemistry. “This has really helped with acceptance of the technology, since many people are not trained in it and don’t have the time to learn it on their own,” Wiles says. “We are taking a bottom-up approach by training students so they’ll go out into industry having experienced another synthetic tool.”
Griffin agrees that two factors influence flow chemistry’s adoption in the lab: Chemists don’t know about it, and it isn’t taught in universities. “Until people come out of university knowing as much about flow chemistry as they do about round-bottom flasks, there will always be opportunities that are missed,” he says. And if medicinal chemists are aware that flow chemistry is an option in process development, they might consider chemistries they wouldn’t have before.
Similarly, Wiles believes that microscale flow chemistry “needs to be demonstrated more outside the lab to become more widely adopted within the lab.” Within the next 18 months, she expects, some of the big pharma and contract manufacturers using the technology in production will be able to talk about their work. “The first few cases that manage to go all the way through the development pipeline will really make people sit up and take notice,” she says.
According to Mansfield, chemists like to see real reaction examples to understand the capabilities of these lab tools. In response, suppliers continually demonstrate methods and extend the technology to new chemistries with special accessories. Some firms, however, are known for offering very specialized systems. For example, Hungary’s ThalesNano has developed dedicated flow hydrogenation, ozonolysis, and carbonylation reactors.
“Although most flow apparatus is designed to be user-friendly and quite simple to operate, it is a transition away from the tried and tested synthetic techniques, and it takes a bit of experience for people to get up to speed with that,” Mansfield says. In particular, chemists moving to flow systems must become accustomed to thinking in terms of solutions, concentrations, solubilities, alternatives for solid reagents, and reaction kinetics.
Not every reaction is suited for flow mode. The technology has drawbacks, including limited throughput, low production capacity, and difficulties coping with solids. To deal with solids, chemists and equipment providers are developing methods using supported reagents, such as catalysts. They also have found ways to mitigate clogging by controlling particle size and even by keeping products molten until they exit the reactor.
Mansfield believes that flow reactors are versatile pieces of apparatus that can handle a wide range of reactions. “But you could also say that about a round-bottom flask, and they have been around for hundreds of years, so why would you want to change?” he asks. “The idea is that you don’t replace existing techniques, but you complement them with new synthetic tools.”