|BASF is already using ionic liquids on large-scale chemical processes as part of its Strategy 2015 sustainable manufacturing initiative.
Supercritical fluids are so named because they are in the “supercritical” state, above the critical temperature and pressure points at which vapor and liquid phases would be in equilibrium. They are part of a new class of “tunable” solvents, whose properties, including density, dielectric constant and solubility, can be easily varied with changes in pressure or other process conditions. The solvents can even “reverse” activity, and function as an acid and base, as needed, within the same synthesis.
Pharmaceutical manufacturers are focusing their evaluation of these new solvents in powder processes , which are used to make about 80% of pharmaceuticals on the market. Tunable solvents used in precipitation and collection have been shown to improve particle uniformity and shape. Eventually, however, the fluids could be used outside of powder processing, in virtually every avenue of pharmaceutical manufacturing as a medium for reactions, extractions and separations.
Tunable solvents require specialized and expensive equipment, so validation and other issues challenge their use in full-scale drug production today. However, the solvents offer a number of compelling long-term benefits; they are flexible, safe, inexpensive, inflammable and environmentally benign.
By selectively precipitating catalysts and products or reactants, they offer advantages in chiral drug synthesis and isolation, and chromatography. They also offer flexibility for synthesis and processing. For example, alcohols can be expanded with CO2 to form reversible alkylcarbonic acids, while CO2-expanded amines yield reversible alkylcarbamic acids, for clean, in-situ acid catalysis. These acid catalysts require no neutralization, eliminating disposal issues associated with salt waste generation.
Tunable solvents also offer environmental benefits. Consider phase transfer catalysis (PTC), which has long been used as a technique for the contact of hydrophobic and hydrophilic species and which accounts for more than $20 billion in annual production today . Applying gaseous CO2 to post-reaction PTC mixtures can alter PTC partitioning by up to two orders of magnitude, representing a savings in washwater recovery of over 95% , and significant economic savings.
As the pharmaceutical industry moves to new and radically different forms of molecules and modes of delivery, the need for new processing and manufacturing techniques is growing at an exponential rate. Tunable fluids applications in pharmaceutical manufacturing will provide solvent media for synthesis as well as processing that meets GMP requirements and offers distinct advantages over conventional methods.
This article will examine the promise of tunable solvents, focusing on new synthesis routes, solvent types and advantages in heterogeneous catalytic processes. It will also highlight development projects that are now under way at leading pharmaceutical companies (see Going Green: The Spirit's Willing, But… below).
The advantages of tunable fluid processing have been studied closely, yielding the following modified drug syntheses, each of which has its advantages and weaknesses:
- Rapid Expansion of Supercritical Solution (RESS),
- in which a drug is dissolved in an SCF such as CO
- and then sprayed into a collection chamber. The solvent is then rapidly removed, resulting in well-defined, uniform drug particles. RESS is a simple and effective technique, but is restricted by the limited solubility of drugs in SCFs.
- Gas Anti-Solvent (GAS)
- is another popular technique, shown in
- ). In this case, a gas such as CO
- is added to an organic solution of a desired drug, resulting in the expansion of the organic solvent and the precipitation of uniform drug particles. The precipitation is easily tuned by adjusting temperature and CO
- pressure, allowing close control of particle size and morphology. GAS takes advantage of the organic solvent’s strength coupled with the tunable solvent expandability. However, it requires semi-batch operation.
- Supercritical Anti-Solvent (SAS) and Solution Enhanced Dispersion by Supercritical fluids (SEDS)
- allow for continuous operation and control. In both of these processes, the drug solution is sprayed into the SCF or mixed with the SCF and sprayed into a collection vessel. Nozzle design and inlet stream flow rates can be adjusted to control the process while utilizing the solvent’s tunability.
Conventional tunable solvents, specifically, supercritical fluids, have been shown to improve particle shape and uniformity, and to facilitate new drug delivery systems. But new opportunities are presenting themselves for other types of tunable solvents , specifically:
- Gas eXpanded Liquids (GXLs),
- organic solvents mixed with carbon dioxide gas, which expands their volume by a factor of 10 or more. Nutraceuticals production has moved in the direction of GXLs, by using ethanol cosolvent as a polar modifier with scCO
- . This results in a solvent medium that can combine synthesis and product recovery in a simplified and efficient “one pot” system.
- Near-critical Water (NCW),
- which offers organic solubilities and the potential for reversible acid/base catalysis.
- Organic/Aqueous Tunable Solvents (OATS),
- or water-soluble catalysts (including both aqueous organometallic complexes and enzymatic biocatalysts) that can perform difficult transformations on highly hydrophobic substrates.
Near-critical Water (NCW), maintained at 200-300 °C, offers an alternative to liquid water, which is a poor solvent choice for many nonpolar chemicals. It also offers an alternative to organic solvents, since it is able to dissolve both organics and salts. NCW also acts as a self-neutralizing acid and base, so that no catalysts need be added for certain reactions, eliminating post-reaction neutralization and greatly reducing salt waste.
NCW offers substantial cost savings in separation reactions, which, in typical processes, account for 60-80% of the capital and operating costs. With organic reactions in NCW, the separation can be as simple as mere cooling and decanting. The utility of this medium has been demonstrated for a diverse group of organic syntheses, including acylations , alkylations , and condensation reactions .
Self-neutralizing alkylcarbonic acids have also been used for acid-catalyzed synthesis. Alkylcarbonic acids are formed by CO2 addition to primary alcohols — for example, methylcarbonic acid from methanol as shown in Figure 2 (below) (b), by the same mechanism as carbonic acid in carbonated water, as shown in 2(a).
This provides in-situ acid formation which is easily neutralized without the addition of base . Alkylcarbonic acids combine good organic solubility with simple neutralization via depressurization. The use of in-situ acid completely eliminates the solid wastes associated with many acid processes. We have compared the reaction rates of several alkylcarbonic acids, the effect of CO2, and demonstrated the use of alkylcarbonic acids for acetal formation  and the hydrolysis of β-pinene . In specific cases, the necessary methylcarbonic acid is formed by simply bubbling CO2 into the methanol solution. This is shown in Figure 3 (below), where the protonation of Reichardt’s dye results in a distinct color change.
As mentioned, tunable solvents are proving useful in novel homogeneous catalysts. Homogeneous catalysts typically offer higher activities and selectivities than heterogeneous catalysts, but it is very difficult to separate these complexes from reaction products. In addition, they are extremely expensive and toxic, particularly asymmetric types, so their recovery and re-use are imperative.
Separating these catalysts becomes even more important, and difficult, in pharmaceutical manufacturing processes, since even trace levels of metallic contamination are unacceptable in product. The use of recoverable homogeneous catalysts enables the development of highly active organometallic complexes that can be easily removed from the product stream, yielding pure products and recyclable catalytic complexes.
|Figure 4. Illustrative example of the OATS concept for phase separation. The picture on the left is a 50:50 mixture of tetrahydrofuran (THF) and water. After exposure to 15 bar CO2 pressure (right), the mixture splits into two phases. The top phase contains an organic dye (methyl yellow) while the bottom phase contains an aqueous dye (Blue #1).
Figure 4 (at right) demonstrates this “miscibility switch” with a mixture of water and a water-miscible organic solvent (THF). The high-pressure sapphire tube assembly contains an organic soluble pharmaceutical analog (methyl yellow) and a water soluble catalyst analog (Blue Dye #1). At ambient conditions, the two phases are completely miscible, as shown by the green color. With the addition of 15 bar of CO2, the THF organic phase is expanded, tuning the solvent properties, inducing a complete phase separation shown by the distinct yellow and blue phases.
This approach represents an interdisciplinary effort aimed at designing solvent and catalytic systems whereby a reversible stimulus induces a change in the system’s phase behavior, enabling easy recovery of the homogeneous catalysts through simple separation techniques, such as filtering and extraction, which are normally applied to heterogeneous or biphasic catalytic systems. Specific examples include the application of gaseous CO2as a benign agent in GXLs to induce catalytic recycle of:
- Miscible water/organic systems,
- where a catalyst is modified for aqueous solubility;
- Miscible poly(ethylene glycol)/organic systems,
- where a catalyst is modified for PEG solubility;
- Immiscible fluorous/organic liquid/liquid or solid/liquid systems,
- where the catalyst is modified for fluorous solubility;
- Immiscible water/organic systems
- involving phase transfer catalysts.
For pharmaceutical manufacturing, the Organic/Aqueous Tunable Solvents (OATS) are very promising. For example, many enzymatic transformations are inhibited by the limited solubility of most hydrocarbon substrates in aqueous enzymatic solutions. In order to overcome this limitation, we have incorporated a series of hydrophilic organic co-solvents (i.e.,THF, acetonitrile, dioxane) that enhance the solubility of very nonpolar organic compounds in aqueous enzymatic systems .
After the reaction has been performed in this mixed solvent, the application of moderate CO2 pressures (~10-20 bar) induces the separation of the organic co-solvent, now containing the hydrophobic product, from the aqueous enzymatic stream, which can then be reused. This technique represents a powerful opportunity to eliminate catalyst contamination of product streams and also recycle highly selective catalytic species. The OATS technique has been demonstrated in many other key reactions, such as the recovery of non-enzymatic aqueous catalysts .
In short, the door is now open for many novel methodologies that are especially applicable to the asymmetric synthesis common to much pharmaceutical production. Though they are a departure from current practice in pharmaceutical manufacturing, such processes have already been applied in a host of other applications, and in each case they give superior products and cost benefits.
|Figure 1. Illustration of Gas Anti-Solvent crystallization for pharmaceutical processing.|
|Figure 2. Comparison of structures of (a) carbonic acid; (b) methylcarbonic acid; (c) peroxycarbonic acid.|
|Figure 3. Alkylcarbonic acids are capable of protonating many molecules. Shown here is an example of the protonation of Reichardt’s betaine dye by addition of CO2 to methanol.|
We know the party line on green chemistry: renewable reagents and cleaner processes lead to reduced environmental burden. But what do risk-phobic pharmaceutical companies really hope to gain from alternative solvents and processing?
“It depends on whom you ask,” says David J.C. Constable, Ph.D., team leader at GlaxoSmithKline’s (King of Prussia, Pa.) Environment, Health and Safety department. “Going green has been shown over and over again to save money. I have stacks of literature about this.”
Early on, during GSK’s assessment of what constitutes green and what does not, the company evaluated “an enormous number” of chemistries and chemical process technologies and published its results widely. But even with senior management’s blessing, adopting green chemistry is anything but a walk in the park. There are the usual objections of loss of time, regulatory hurdles, revalidation, and the “if it ain’t broke don’t fix it” mentality.
It’s not that much easier during early R&D, either. Not only do preclinical and clinical projects leave little time for chemical or process development, but many synthetic organic chemists are reluctant to try new syntheses and processes. “There is a need for greater collegial collaboration between the chemist, chemical engineer and biotechnologist,” Constable says.
Constable also believes the regulatory trajectory of new chemical entities severely inhibits process innovation. “FDA's approach to marrying process to product, as opposed to just a rigorous product specification, locks bad processes in place,” he says. “It is enormously difficult and expensive to change.”
The more far-out the change, the less likely it will be taken seriously. For example, ionic liquids are a hot topic in green industrial chemistry, but have received scant attention from pharmaceutical chemists (no less a change agent than Constable described GSK’s forays into ionic liquids as “unproductive”).
Ionic liquids are organic salts formed from an almost infinite combination of imidazolium or pyridinium cations, and select anions, mixed in any quantity. Ionic liquids have no measurable vapor pressure and are nonflammable, so they are potential substitutes for volatile organic solvents. At the same time, they dissolve almost any organic compound and catalyze reactions, thereby serving as agents of process intensification.
What more could a pharmaceutical chemist ask for?
According to a 2004 Chemical Week article, Novartis (Basel, Switzerland) has replaced six chemical steps in an established process with a two-step Friedel-Crafts alkylation in an ionic liquid. Not only is the process run at pilot scale, but the use of ionic liquids was deemed patentable. Other major manufacturers are also investigating ionic liquids, according to chemical suppliers quoted in the article.
K.R. Seddon, Ph.D., chair of inorganic chemistry at Queen's University (Belfast, Northern Ireland), a leading authority on ionic liquids, sees “no great obstacles” to adopting liquids in drug-making.
He dismisses the notion that these solvents are toxic. “The knock on ionic liquids is more a result of image than of fact. Ionic liquids consist of mostly simple ions, many of which have already been designated as safe for ingestion.”
BASF (Florham Park, N.J.) is already using ionic liquids on large-scale chemical processes as part of its Strategy 2015 sustainable manufacturing initiative. The company’s BASIL (biphasic acid scavenging using ionic liquids) process is applicable to about one-third of all industrial chemical processes that require acid scavenging.
Normally, acidic by-products are scavenged with amines, which results in viscous, white slurries that are difficult to remove from the product. BASF’s work-around uses methyl imidazole, whose hydrochloride salt happens to be an easily-removable ionic liquid. According to Calvin J. Emanuel, Ph.D., manager of new business development, benefits include easier product isolation, higher yield, and an 80,000-fold improvement in operating efficiency (you read that right). “Without all those solvents around, the process requires less energy to overcome heat transfer issues,” he explains. After the reaction, the ionic liquid is converted back to imidazole and reused.
BASF sees ionic liquids as more than simple solvents. “Calling them solvents doesn’t do them justice,” Emanuel says. “It’s impossible to compare the utility of a simple solvent like THF with ionic liquids.” Prof. Seddon agrees. In his estimation, ionic liquids’ true benefit is their ability to provide greater selectivity and higher yield than traditional solvents.
Since BASIL was introduced in 2002, chemical companies have used it in production of alkoxyphenylphosphines on a multiton scale. An “important” pharmaceutical industry customer has also licensed the process, says William Pagano, a BASF spokesman. BASIL’s potential applications include chlorinations by nucleophilic HCL, azeotropic distillations, and extractions. In addition to using ionic liquids, BASF offers a broad portfolio of them in bulk quantities through its BASIONICS product line. Investigators can purchase development and laboratory quantities through Sigma-Aldrich.
About the Authors
Dr. Charles A. Eckert and Dr. Charles L. Liotta have been research partners for 16 years; they occupy a common laboratory space and codirect students in both Chemistry and Chemical Engineering. Their research focus is on the use of novel solution chemistry for sustainable technology, and for their contributions to industry they jointly received the 2004 Presidential Green Chemistry Challenge Award. Both have joint appointments at Georgia Tech in the Schools of Chemical and Biomolecular Engineering and in the School of Chemistry and Biochemistry. Their group’s website is http://www.che.gatech.edu/ssc/eckert/.
Dr. Christopher L. Kitchens is a Post Doctoral Researcher in the Eckert-Liotta Joint Research Group at Georgia Institute of Technology. He received a B.S. in Chemistry from Appalachian State University and a Ph.D. in Chemical Engineering from Auburn University, where he worked on nanoparticle synthesis and processing in tunable fluids.
Dr. Jason P. Hallett is a Research Engineer in the Eckert-Liotta Joint Research Group at Georgia Institute of Technology. He received a B.S. in Chemical Engineering from the University of Maine and a Ph.D. in Chemical Engineering from Georgia Institute of Technology, where he worked on novel methods for homogeneous catalyst recycle.
- McHugh, M; Krukonis, V., Supercritical Fluid Extraction: Principals and Practice, Butterworth Publishers, Stoneham, Mass., 1986.
- York, P.; Kompella, U.; Shekunov, B., Supercritical Fluid Technology for Drug Product Development. Drugs and the pharmaceutical sciences ; v. 138 M. Dekker, New York, 2004.
- Staks, C.; Liotta, C.; Halpern, M., Phase-Transfer Catalysis: Fundamentals, Applications, and Industrial Perspectives. Chapman & Hall, New York, 1994.
- Xie, X.; Brown, J.; Joseph, P.; Liotta, C.; Eckert, C.,“Phase-Transfer Catalyst Separation by CO2 Enhanced Aqueous Extraction”, Chemical Communications, 2002, 1156.
- Eckert, Charles A.; Liotta, Charles L.; Bush, David; Brown, James S.; Hallett, Jason P. “Sustainable Reactions in Tunable Solvents.” J. Phys. Chem. B 2004, 108, 18108.
- Brown, J.; Gläser, R.; Liotta, C.; Eckert, C.,“Acylation of Activated Aromatics without Added Acid Catalyst”, Chem. Commun., 2000, 1295-1296.
- Chandler, K.; Deng, F.; Dillow, A.; Liotta, C.; Eckert, C., “Alkylation Reactions in Near-Subcritical Water in the Absence of Acid Catalysts,” Ind. Eng. Chem. Res., 1997, 36, 5175.
- Nolen, S.; Liotta, C.; Eckert, C.,“The Catalytic Opportunities of Near-Critical Water: A Benign Medium for Conventionally Acid and Base Catalyzed Organic Synthesis,” Green Chem., 2003, 663.
- West, K.; Wheeler, C.; McCarney, J.; Griffith, K.; Bush, D.; Liotta, C.; Eckert, C.; “In Situ Formation of Alkylcarbonic Acid with CO2” J. Phys. Chem. A, 2001. 105, 3947.
- Xie, X.; Liotta, C.; Eckert, C., CO2-Catalyzed Acetal Formation in CO2-Expanded Methanol and Ethylene Glycol. Ind. Eng. Chem. Res. 2004, 43, 2605.
- Chamblee, T.S., R.R. Weikel, S.A. Nolen, C.L. Liotta, and C.A. Eckert, “Reversible in situ acid formation from beta-pinene hydrolysis using CO2 expanded liquid and hot water” Green Chem., 2004, 6.
- Nolen, S.; Lu, J.; Brown, J.; Pollet, P.; Eason, B.; Griffith, K.; Glaser, R.; Bush, D.; Lamb, D.; Eckert, C.; Liotta, C.; Thiele, G.; Bartels, K.,“Olefin Epoxidations Using Supercritical Carbon Dioxide and Hydrogen Peroxide Without Added Metallic Catalysts or Peroxy Acids,” Ind. Eng. Chem. Res., 2002. 41, 316.
- West, K.; Hallett, J.; Jones, R.; Bush, D.; Liotta, C.; Eckert, C. "CO2-Induced Miscibility of Fluorous and Organic Solvents at Ambient Temperatures." Ind. Eng. Chem. Res., 2004. 43, 4827.
- Lu, J.; Lazzaroni, M.; Hallett, J.; Bommarius, A.; Liotta, C.; Eckert, C. “Tunable Solutions for Homogeneous Catalyst Recycle.” Ind. Eng. Chem. Res. 2004, 43, 1586.