Greening Flash Chromatography - ACS Sustainable Chemistry


Greening Flash Chromatography - ACS Sustainable Chemistry...

1 downloads 87 Views 3MB Size

Research Article pubs.acs.org/journal/ascecg

Greening Flash Chromatography Ray McClain,*,† Vanessa Rada,† Ashley Nomland,‡ Matt Przybyciel,§ Dave Kohler,§ Rolf Schlake,∥ Philippe Nantermet,† and Christopher J. Welch⊥ †

Automation and Capabilities Enhancement, Merck Research Laboratories, West Point, Pennsylvania 19486, United States Discovery Chemistry, Merck Research Laboratories, West Point, Pennsylvania 19486, United States § ES Industries, Inc., West Berlin, New Jersey 08091, United States ∥ Applied Separations, Allentown, Pennsylvania 18101, United States ⊥ Process Research and Development, Merck Research Laboratories, Rahway, New Jersey 07065, United States ‡

ABSTRACT: Proof of principle is demonstrated for a novel approach for using pressurized liquid carbon dioxide (CO2) as a mobile phase for flash chromatography purification, rather than traditionally used organic solvents. Regulation of a pressurized CO2 supply line rather than specialized and expensive CO2 pumps and chillers typically used in supercritical fluid chromatography applications greatly simplifies operations while reducing costs. Evaluation of a prototype instrument reveals encouraging purification, recovery, and reduction in organic solvent usage.

KEYWORDS: Flash chromatography, Green chemistry, Liquid CO2



INTRODUCTION In just two decades, green chemistry and engineering has evolved from an unfamiliar and somewhat threatening concept to becoming a critical component of research, development, and manufacturing operations of many chemistry-related industries.1−4 While much of the effort to date has understandably focused on developing greener manufacturing and large-scale development processes, significant progress has also been made in improving the environmental footprint of research operations, where the amount of waste generated per researcher is relatively small, but the cumulative impact of a large number of researchers can be substantial. Advances in the miniaturization of organic reactions and in the replacement of toxic and harmful reagents with benign alternatives have made rapid progress in recent years.5−7 Even the analytical technologies that underlie modern organic chemical research have come under scrutiny, with many new alternatives that reduce solvent use and green alternatives being developed to replace harmful solvents.8−10 The recent focus on reducing the environmental impact of flash chromatography, the “go to” purification technology still used on a daily basis by almost all synthetic organic chemists, has been on changing the nature of the eluents used.11−15 Flash chromatography, a term coined by Clark Still in an influential 1978 publication, is a normal phase separation technique employing a silica gel stationary phase and various organic solvents such as ethyl acetate/hexane or methanol/dichloromethane as a mobile phase.16 Flow through the column is often facilitated with nitrogen overpressure or the © XXXX American Chemical Society

use of a low pressure liquid pump, with manual fraction collection into test tubes often facilitated by the use of an automated fraction collector. A typical automated flash chromatography system is presented in Figure 1. Flash column separations are run on a daily basis by many thousands of synthetic chemists, with disposal of all solvents and stationary phase after a single use being typical. The practice is a testimony to the adage that “in order to make a small amount of something pure we often make a large amount of something else dirty”, with a kilogram or more of waste solvent and silica often accompanying the purification of a few milligrams of a new investigational compound. Given the large environmental footprint of flash chromatography, much thought and planning has been devoted to the development of green alternatives. Currently, flash chromatography is carried out by individual researchers with relatively inexpensive equipment. Thus, any “one-to-one” replacement technology must be both inexpensive and user friendly. On the other hand, any envisioned “many-to-one” replacement strategy must be able to provide user-friendly software for sample queuing and “hands off” automated and reliable separations for a variety of low purity samples. In recent years, normal phase purification techniques in larger scale chemical development have to a large extent been supplanted by greener supercritical Received: June 1, 2016 Revised: July 19, 2016

A

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 1. Schematic diagram illustrating components of a typical automated flash chromatography system.

Figure 2. Schematic diagram illustrating the components of a typical preparative SFC system.

fluid chromatography (SFC) purification technologies employing pressurized carbon dioxide-based mobile phases, where complex and expensive purification equipment is typically operated by separation specialty teams, rather than individual synthetic chemists.17−19 Such tools are prohibitively expensive for a one-to-one replacement strategy and poorly suited to the large number of customers and impure samples required by the many-to-one replacement strategy. A typical preparative SFC system is presented in Figure 2. The idea of an inexpensive SFC instrument serving the needs of individual synthetic chemists has heretofore been impractical, largely due to the added cost and complexity associated with the pumping of carbon dioxide in the sub/supercritical state, where a chiller and a separate carbon dioxide pump are

required, in addition to all the other components of a conventional preparative liquid chromatography system. We now introduce a new concept for liquefied CO2 flash purification that bypasses the need for this chiller and carbon dioxide pump, greatly simplifying and reducing the cost of green flash chromatography to the point where a one-to-one replacement of current flash chromatography may be feasible.



EXPERIMENTAL SECTION

Materials. Sorbic acid, 4-hydroxybenzoic acid, niacinamide, antipyrine, ketoprofen, caffeine, and (S,R) noscapin were obtained from Sigma-Aldrich (St. Louis, MO). N-Benzylbenzamide and 4dimethylamino pyridine (DMAP) were obtained from Acros (New Jersey). Ibuprofen was obtained from Alfa Aesar (Heysham, England. Beverage grade CO2 was obtained from Praxair (Danbury, CT). B

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 3. Schematic diagram illustrating the components of the eCO2Chrom Liquid CO2 Flash Chromatography system. Optima grade methanol, ethyl acetate, and ammonium hydroxide were obtained from Fisher (Fair Lawn, NJ). Ethyl alcohol (200 proof) was obtained from Pharmco-Aaper (Shelbyville, KY). Prepacked silica cartridge columns as well as prepacked imidazole cartridge columns were obtained from ES Industries (West Berlin, NJ). Instrument and Column Housing. The prototype instrument used in these studies is the eCO2Chrom Liquid CO2 Flash Chromatography system from Applied Systems (Allentown, PA) which is shown in Figure 3.The column used in conjuction with the Ecoflash system is assembled by the user through introduction of a silica cartridge to a stainless steel tube with detachable end caps. More details about both the instrument and separation cartridge are discussed in the Results and Discussion section. It is important to note that all CO2 tubing is stainless steel to avoid any possible corrosion from residual moisture present. Typical Method. 60 mL/min isocratic CO2 flow at 65 bar outlet pressure 40 g, 50 μm silica column, 2 cm i.d. × 15 cm length ∼0−30 mL/min linear cosolvent gradient executed over 15 min 50 °C temperature control on CO2 inlet valve 30 °C temperature on back pressure control valve 10 mL/min make up flow 214 and 254 nm wavelength detection from four channel diode array detector Detector autozeroed at 2 min

collaborations between members of our team led to the development of the Ecoflash Liquid CO2 Flash Chromatography prototype described herein. The eCO2Chrom unit consists of a heated micrometering valve to accurately dispense the liquid CO2, a liquid CO2 flow meter, a cosolvent pump capable of 100 mL/min flow, a threeway valve to allow use of an external loading cartridge or direct loading on the head of the separation cartridge, a four channel diode array detector, a back pressure regulator to maintain the liquid state of the CO2, a gas liquid separation chamber, a makeup pump, and an open-bed fraction collector. The 40 g silica cartridge used as the separation column is loaded by the user into a permanent, reusable, heavy gauge stainless steel tube with threads that allow screw on, stainless steel end caps to be added by the end user prior to each separation. The end caps contain O-rings which seal the disposable, plastic cartridge and essentially make an HPLC column capable of withstanding the 60−70 bar of pressure frequently experienced during routine operation. The system currently is limited to a column format of 40 g as dictated by the existing stainless steel housing but conceptually could be scaled up through machining of a larger stainless steel cartridge housing and production of the properly sized plastic cartridges. The larger column housing and cartridge would allow the purification of multigram scale separations frequently performed with traditional flash chromatography systems. In order to have the potential for displacing conventional liquid-based flash chromatography approaches, the flash purification system utilizing liquid CO2 as the mobile phase must deliver a level of chromatographic performance that is at least equivalent to that of traditional flash chromatography. To assess the capabilities of the prototype system, several compounds of various structural classes including acidic, neutral, and basic functionalities were selected as test probes and multicomponent standard solutions were prepared. The selected compounds were prepared as shown in Table 1, and 1



RESULTS AND DISCUSSION Chromatographic Performance. The key innovation in the development of the eCO2Chrom Liquid CO2 Flash Chromatography system is the realization that in a laboratory with a pressurized CO2 supply line the chiller and CO2 pump of a conventional preparative SFC instrument are in some ways redundant and that elimination of these components can lead to considerable cost savings and reduced complexity. With access to a pressurized CO2 supply line, simple metering of the CO2 supply can be used to supply the fluid required for chromatographic separations. Simple proof of concept demonstrations showed the feasibility of this approach, and C

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

particles. The flow of CO2 was increased to 60 mL/min to mimic the higher flow rates traditionally used with sub/ supercritical CO2 separations, with comparable results. The methanol gradient was next decreased from 0 to 40 mL/min over 15 min to 0−23 mL/min to maximize resolving power, as all the test probes had eluted in the first half of the previous gradient. An example chromatogram showing separation of a six component mixture using this method is displayed in Figure 4. The chromatographic peak shape afforded by this method was quite good and is comparable to what is typically obtained in liquid flash chromatography experiments. The partial coelution of antiprine and niacinamide in the test mixture under these conditions was not a concern since our focus at this point was on overall instrument performance rather than complete resolution of all peaks. It is widely understood that basic compounds do not elute readily and sometimes display undesirable peak shape on silica columns without the use of basic additives. This was confirmed through analysis of a 0.5 mL aliquot of the basic standard mixture containing noscapine and DMAP on a 40 g silica cartridge. The DMAP did not elute from the silica cartridge during the 0−23 mL/min methanol gradient executed over the course of 15 min. A much more aggressive gradient of 0−40 mL/min of 1% NH4OH in methanol executed over 15 min was

Table 1. Standard Solutions Used in Evaluation conc (mg/mL) neutral mix benzylbenzamide niacinamide acid mix ibuprofen 4-hydroxybenzoic acid basic mix noscapine DMAP

diluent

76.4 93.7

methanol methanol

94.9 113.2

1:1 DCM:methanol 1:1 DCM:methanol

208.8 222.5

1:1 DCM:methanol 1:1 DCM:methano

mL aliquots of the various mixtures were then transferred to the head of the 2 cm i.d. × 15 cm length silica cartridges containing 40 g of 50 um silica particles, effectively preparing the cartridges for separation. The initial chromatographic conditions evaluated consisted of maintaining the liquid CO2 flow at 40 mL/min flow rate and 65 bar of pressure at the column outlet for the duration of the 15 min analysis, while increasing the methanol cosolvent flow from 0−40 mL/min in a linear profile. The initial results for the acid and neutral mixtures were favorable, with all four test probes being successfully resolubilized and eluted in Gaussian shaped peaks typically observed with flash cartridges containing 50 μm dp silica

Figure 4. Mixture containing approximately 15 mg each of sorbic acid, ibuprofen, 4-hydroxybenzoic acid, caffeine, antipyrine, and niacinamide chromatographed according to the following conditions: 60 mL/min liquid CO2 at 65 bar, 0−30 mL/min methanol gradient over 15 min, 50 °C temperature control on CO2 inlet valve, 30 °C temperature control on BPR, 10 mL/min ethanol makeup flow, and 214 nm UV detection. Antipyrine and niacinamide partially coelute under these conditions but is not of concern as selectivity is not being evaluated in this case. D

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 5. Comparison of noscapine and DMAP chromatographed on silica column versus GreenSep Basic bonded phase. Both components elute within the linear gradient range with acceptable peak shape on the imidazole-bonded phase, while the DMAP required increasing the gradient slope and addition of NH4OH on the silica cartridge to elute from the silica cartridge.

and solvent can be clearly seen in the collected fractions or in the waste line. However, in the initial design of the liquid CO2 flash system at cosolvent levels less than 10 mL/min, no solvent was observed in either the fraction collection vial or the waste line. Evaluation of the cosolvent pump showed it to dependably deliver flow down to a flow rate of 3−4 mL/min, suggesting that the gas−liquid separation mechanism of the unit was aerosolizing the column eluent at liquid flows of less than 10 mL/min. Eluent aerosolization at low percentages of liquid solvents has been observed in preparative SFC systems, where incorporation of a makeup pump to supply liquid flow just before the gas−liquid separation mechanism eliminates aerosolization and increases sample recovery. Application of this strategy afforded excellent performance, as summarized in Figure 6. The neutral binary compound mix was analyzed at various incremental masses on the column to evaluate not only recovery but recovery linearity. Figure 6 displays the chromatographic traces. Table 2 displays the volumes loaded on the head of the column and the calculated percent recovery. It is immediately apparent that the recovery of both early eluting peaks as well as late eluting peaks is very high, ranging from 84.4% to 95.5%. It is worth highlighting that the ability to load samples in diluents composed of high percentages of methanol is an improvement over conventional liquid flash chromatography, where even the slightest amount of methanol in the diluent can cause severe peak distortion.

required to begin elution of the DMAP from the silica cartridge. We have previously investigated the use of specific bonded phases for improved separation of amines.20,21 Accordingly, a GreenSep Basic column was prepared for evaluation with the imidazole-based ligand bonded to 50 μm silica, using an identical cartridge format and packing technique as the bare silica cartridges. Half a milliliter of the binary basic mixture was chromatographed on the GreenSep Basic cartridge using the same method as for the acidic and neutral mixtures (60 mL/ min CO2 flow, a linear flow gradient of 0−23 mL/min methanol (no additive) over 15 min) leading to elution of DMAP in the middle of the gradient with desirable peak shape (Figure 5). Preliminary data generated with standard test probes suggest that an acceptable general protocol for liquid CO2 flash would be the use of a silica column for acidic and neutral compounds and the use of the GreenSep Basic bonded phase for basic compounds. Recovery. Flash chromatography is a purification technique, and as such, high recovery of purified compounds is essential. It is therefore essential that any prototype CO2-based flash chromatography system exhibits recoveries that are comparable with those of conventional flash chromatography. The liquid CO2 flash system was designed with a fraction collector in which the collection of purified compounds can be triggered by UV threshold or specified time collection. In a typical organic flash system, solvent is continually flowing through the system, E

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 6. Recovery data of five incremental aliquots of the binary neutral mix found in Table 2. Data acquired according to the chromatographic method presented in the Experimental Section. Volumes were noted on the niacinamide peak as the ideal peak fronting allows for easy identification.

considerable solvent savings vs the conventional liquid flash chromatography approach. Clearly, the flow rate of the makeup pump could be adjusted such that it is only active when the flow rate of the cosolvent pump drops below 10 mL/min, ensuring that the combined flow of these two liquid pumps is always greater than 10 mL/min. Example of Actual Drug Discovery Compound. It is common to utilize silica-based, thin-layer chromatography (TLC) to develop chromatographic conditions for use in a flash separation. The use of TLC as a predictor for preparative SFC separation has received some attention from researchers, with mixed results.22,23 Ashraf-Khorassani and Taylor found some correlation between TLC and SFC with silica columns, but Miller warned that it is not possible to use TLC to predict retention for SFC using silica columns. In our hands, the chromatographic behavior of sample analytes on TLC plates allowed for successful scale-up to the liquid CO2 flash system. An in-house synthesized developmental compound was analyzed via TLC with the mobile phase composition of 1:1 hexane:(3:1 ethyl acetate:ethanol) delivering a response factor (Rf) of 0.7 for the compound of interest and Rfs of 0.5 and 0.3 for the two impurities present. Five hundred milligrams of the developmental compound was then purified on a 40 g silica cartridge on a traditional flash chromatograph using 3:1 ethyl acetate:ethanol in hexane as the mobile phase. The 15 min gradient ranging from 100% hexane to 50% 3:1 ethyl acetate:ethanol in hexane was performed at 40 mL/min, generating 600 mL of liquid waste. The desired compound eluted at 8 min, with the impurities eluting at 13 and 16 min, as predicted from the TLC scouting analysis. A similar 500 mg separation was performed on a 40 g silica cartridge on the liquid

Table 2 recovery

0.5 mL

1.0 mL

2.0 mL

2.5 mL

5 mL

benzylbenzamide niacinamide

89.2 93.4

84.4 92.4

90.0 91.4

92.4 93.8

92.9 95.5

The effectiveness of the makeup pump was further evaluated by testing with sorbic acid, a compound that elutes at low cosolvent levels. A chromatographic method was developed for the analysis of duplicate 530 mg aliquots of sorbic acid, one aliquot analyzed with the makeup pump off and with the other with the makeup pump set at 10 mL/min, the point where the gas−liquid separation mechanism begins delivery of constant flow to the fraction collector. The 15 min method consisted of a 60 mL/min constant flow of liquid CO2 along with a 0−8 mL/min linear flow ramp of methanol. The sorbic acid began elution when the methanol flow was at 3 mL/min and concluded elution when the methanol flow is at 4 mL/min. The chromatographic traces for the duplicate analysis are nearly identical, as they should be, but the fractions obtained for the duplicate runs could not be more different. The fraction volume of the analysis with the makeup was 27 mL, while the fraction volume for the analysis without the makeup was 0 mL. The 27 mL was rotovapped under low pressure and found to provide 95% recovery of the 530 mgs of sorbic acid loaded, clearly demonstrating the effectiveness of the makeup pump as well as the gas−liquid separator. In contrast, virtually no sorbic acid was found in the collection vial from the run without makeup flow. The need for this additional make up solvent pump is less than ideal from a cost, complexity, and green chemistry perspective but does provide excellent separations with F

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

Figure 7. (a) TLC analysis with developing solvent of 1:1 heptane:(3:1ethyl acetate:ethanol). (b) Liquid CO2 flash chromatogram using a 0−50 mL/min gradient 3:1ethyl acetate:ethanol in 65 bar liquid CO2 performed over 15 min. (c) Traditional flash chromatogram using 0−50% 3:1ethyl acetate:ethanol in heptane over 15 min.

CO2-based flash chromatograph with the CO2 flow rate set at 60 mL/min, the BPR set at 65 bar, and a 0−50 mL/min 3:1ethyl acetate:ethanol in CO2 gradient performed over 15 min, mimicking the conditions used in the traditional flash system. This liquid CO2 flash purification generated 375 mL of waste solvent from the chromatographic separation combined with 150 mL of waste solvent from the makeup solvent introduction prior to the back pressure valve. The peak of interest eluted at 4 min with the impurities eluting at 7.8 and 9 min, as predicted by the TLC analysis and as shown by the traditional flash analysis. The data for TLC plate, liquid CO2 flash chromatogram, and traditional flash chromatogram are shown in Figure 7a, b, and c, respectively. Our study demonstrates liquid CO2 delivers more similar chromatographic selectivity to heptane than does CO2 in the sub- or super-critical state, enabling scale-up to the liquid CO2 flash unit without the use of an expensive analytical SFC. The ability to gauge separation success via TLC was one of the key milestones achieved during the evaluation of this system. As highlighted in the above real-world sample, the liquid CO2 flash system utilized 37.5% less solvent for the chromatographic resolution of the sample when compared to a traditional flash chromatography system. However, the solvent consumed by the makeup pump reduced solvent savings to 12.5% for this separation. The gas−liquid separator on the flash system provides excellent recovery with an inlet flow of 10 mL/min, which allows the makeup pump to be stopped when the system flow achieves this critical flow rate to reduce solvent consumption and related waste disposal. For this example, the makeup pump could be stopped 3 min into the purification run as the cosolvent delivery would then be above the critical flow. This would translate to a total waste reduction of 32.5% when comparing the liquid CO2 flash system to the traditional flash system. Subsequent analyses on the liquid CO2 flash unit have employed this abbreviated makeup flow and confirmed its success. Safety. Safety is a major concern when using any compressed gas in the laboratory. In order to be useful for routine laboratory work, any CO2-based flash system must minimize risks arising from the use of a compressed gas, the exposure of lab occupants to aerosolized samples, or asphyxiation by CO2. The presence of a CO2-metering valve to deliver liquid CO2 instead of high pressure CO2 from a pump minimizes these concerns on this new style of flash

chromatography system. The new system has a post-gas−liquid separator CO2 vent line that is sent to a liquid scrubber to capture any residual aerosolized compound. The outlet line of the scrubber is sent to an exhaust hood which removes the gaseous CO2 from the laboratory area. The recovery data presented in Table 2 demonstrates that aerosolization is typically not a concern. Furthermore, the engineered mechanism for removing gaseous CO2 from the system provides secondary containment to eliminate possible exposure to lab occupants. Finally, the performance of the system for “unusual” samples is encouraging. A developmental compound with known solubility deficiencies was loaded on a 40 g silica cartridge and subjected to the standard gradient described in the experimental station. Purity assessment of isolated fractions revealed none of the desired compound. Upon opening the cartridge housing, bright white precipitate at the head of the column was analyzed via LC/MS and found to be the compound of interest, in greater than 95% purity. The ability of the prototype flash system to handle this insoluble compound and maintain the integrity of the plastic separation column demonstrates an ability to withstand “unusual” samples that precipitate out of solution, an event that can lead to pressure rises in SFC systems. Overall, we have found the instrument to have a record of safety, dependability, and ease of use that is comparable with that of established, much more expensive, preparative SFC instruments.



CONCLUSION A new concept in flash chromatography using liquid carbon dioxide has been developed and shown to deliver chromatographic performance comparable to that of traditional flash systems while reducing waste generation by about one-third. As optimization of gradient slope, gradient range, run times, flow rates, and other operating parameters is still ongoing, further improvements in waste reduction are likely as our work continues. The unit can accurately deliver CO2 from a micrometering valve to afford efficient chromatographic performance while eliminating the need for an expensive CO2 pump and associated chiller. The embedded gas liquid separation apparatus has been proven to provide excellent recovery to ensure successful recovery of the material being purified. The makeup pump present just prior to the gas liquid separator ensures recovery of solutes eluting at low cosolvent G

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX

Research Article

ACS Sustainable Chemistry & Engineering

enantiopurity in Pharmaceutical Process Research. LC-GC Europe 2005, 16−29. (19) Miller, L.; Potter, M. Preparative resolution of racemates using HPLC and SFC in a pharmaceutical discovery environment. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2008, 875, 230−236. (20) McClain, R.; Przybyciel, M. A systematic study of achiral stationary phases using analytes selected with a molecular diversity model. LC/GC North America 2011, 29, 894−906. (21) McClain, R.; Hyun, M. H.; Li, Y.; Welch, C. J. Design, synthesis and evaluation of stationary phases for improved achiral SFC separations. J. Chromatogr. A 2013, 1302, 163−73. (22) Miller, L.; Mahoney, M. Evaluation of flash supercritical chromatography and alternative sample loading techniques for pharmaceutical medicinal chemistry purifications. J. Chromatogr. A 2012, 1250, 264−273. (23) Ashraf-Khorassani, M.; Yan, Q.; Akin, A.; Riley, F.; Aurigemma, C.; Taylor, L. T. Feasibility of correlating separation of ternary mixture of neutral analytes via TLC with SFC in support of green flash separations. J. Chromatogr. A 2015, 1418, 210−217.

percentages. Laboratory evaluations have shown scouting investigations using TLC to be a good predictor of performance using liquid CO2 flash chromatography. The liquid CO2 flash unit has proven to be safe, rugged, reliable, and easy to use through the course of our evaluation using hundreds of flash cartridges and the study of real world samples possessing demanding physical properties. This new style of flash unit is a green alternative to the powerful tool delivered by Clark Still almost 40 years ago.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



REFERENCES

(1) Horvath, I. T.; Anastas, P. T. Innovations and green chemistry. Chem. Rev. (Washington, DC, U. S.) 2007, 107 (6), 2169−2173. (2) Trost, B. M. On inventing reactions for atom economy. Acc. Chem. Res. 2002, 35 (9), 695−705. (3) Dunn, P. J. The importance of Green Chemistry in Process Research and Development. Chem. Soc. Rev. 2012, 41 (4), 1452−1461. (4) Constable, D. J. C.; Dunn, P. J.; Hayler, J. D.; Humphrey, G. R.; Leazer, J. L., Jr.; Linderman, R. J.; Lorenz, K.; Manley, J.; Pearlman, B. A.; Wells, A.; et al. Key green chemistry research areas-a perspective from pharmaceutical manufacturers. Green Chem. 2007, 9 (5), 411− 420. (5) Geyer, K.; Codee, J. D. C.; Seeberger, P. H. Microreactors as tools for synthetic chemists - the chemists’ round-bottomed flask of the 21st century? Chem. - Eur. J. 2006, 12 (33), 8434−8442. (6) Santanilla, A. B.; Regalado, E. L.; Pereira, T.; Shevlin, M.; Bateman, K.; Campeau, L. C.; Schneeweis, J.; Berritt, S.; Shi, Z. C.; Nantermet, P.; Liu, Y.; Helmy, R.; Welch, C. J.; Vachal, P.; Davies, I. W.; Cernak, T.; Dreher, S. D. Nanomole Scale High-Throughput Chemistry for the Synthesis of Complex Molecules. Science 2015, 347, 49−53. (7) Sheldon, R. A. E factors, green chemistry and catalysis: an odyssey. Chem. Commun. (Cambridge, U. K.) 2008, 29, 3352−3365. (8) Welch, C. J.; Wu, N.; Biba, M.; Hartman, R.; Brkovic, T.; Gong, X.; Helmy, R.; Schafer, W.; Cuff, J.; Pirzada, Z. Greening Analytical Separation Technologies. TrAC, Trends Anal. Chem. 2010, 29, 667− 680. (9) Armenta, S.; Garrigues, S.; de la Guardia, M. Green Analytical Chemistry. TrAC, Trends Anal. Chem. 2008, 27 (6), 497−511. (10) Peterson, E.; Dillon, B.; Raheem, I.; Richardson, P.; Richter, D.; Schmidt, R.; Sneddon, H. Sustainable chromatography (an oxymoron?). Green Chem. 2014, 16, 4060−4075. (11) MacMillan, D. S.; Murray, J.; Sneddon, H. F.; Jamieson, C.; Watson, A. J. B. Green Chem. 2012, 14, 3016−3019. (12) Taygerly, J. P.; Miller, L. M.; Yee, A.; Peterson, E. A. Green Chem. 2012, 14, 3020−3025. (13) Drueckhammer, D. G.; Gao, S. Q.; Liang, X.; Liao, J. ACS Sustainable Chem. Eng. 2012, 87−90. (14) Hobbs, A. N.; Young, R. J. Drug Discovery Today 2013, 18 (3− 4), 148−154. (15) Chardon, F. M.; Blaquiere, N.; Castanedo, G. M.; Koenig, S. G. Green Chem. 2014, 16, 4102−4105. (16) Still, W. C.; Kahn, M.; Mitra, A. Rapid chromatographic technique for preparative separations with moderate resolution. J. Org. Chem. 1978, 43, 2923−2925. (17) Terfloth, G. J. Enantioseparations in super and subcritical chromatography. J. Chromatogr. A 2001, 906, 301−307. (18) Welch, C. J.; Leonard, W. R., Jr.; DaSilva, J. O.; Biba, M.; Albaneze-Walker, J.; Henderson, D. W.; Laing, B.; Mathre, D. J. Preparative chiral SFC as a ’green’ technology for rapid access to H

DOI: 10.1021/acssuschemeng.6b01219 ACS Sustainable Chem. Eng. XXXX, XXX, XXX−XXX