1593ISSN 1756-8919Future Med. Chem. (20 09) 1(9), 1593–161210.4155/FMC.09.132 © 2009 Future Science LtdReviewIn 2001, Wess and coworkers stated that, within the pharmaceutical industry, organic chemistry contributes to three main areas: lead compound identification, optimization of leads into clinical candidates and the supply of these materials in sufficient quantity for fur-ther investigations [1]. The laborious and often time-consuming nature of these steps, however, still presents the industry with a bottleneck that has not been alleviated, as first hoped, with the advent of techniques such as high-throughput screening and combinatoria l chemistr y [2]. Consequently, there remains a need for fur-ther advances in the way that chemistry is performed, in order to reduce the time taken to synthesize libraries, identify potential leads, optimize the ability to afford drug candidates and put compounds into production. Continuous-flow reactor methodology, in particular microreaction technology (MRT), has been shown to afford the user numerous advantages over traditional stirred-batch reactors, namely increased reaction control, which arises from predictable mixing regimes and inherently high surface-to-volume ratios. In addition, the ability to define with increasing accuracy the reaction condition employed, enables the opera-tor to conduct synthetic transformations with increased efficiency, reproducibility and safety when compared with conventional stirred ves-sels [3–7]. As such, this review aims to highlight the potential of continuous-flow chemistry as a valuable synthetic tool for the modern medicinal chemist, by initially discussing general transfor-mations by reaction type and then focusing on the relevant technological advancements that have enabled the use of flow reactors for the syn-thesis of biologically active compounds. For those readers who are new to the field, details of reactor fabrication, along with how the equipment is used, can be found in elsewhere [8]. Examples of synthetic ow reactions C-C bond-forming reactionsOver the past decade, a significant a mount of research has been conducted into the for-mation of C-C bonds under continuous flow, with authors publishing their findings on Aldol reactions [9] , Michael additions [10,11], acyla-tions [12], alkylations [13], Knoevenagel conden-sations [14,15,16] and Baylis–Hillman reactions to afford allylic alcohols [17], to name but a few.The complexity of such transformations was recently increased by Odedra and Seeberger [18], who reported a series of 5-(pyrrolidin-2-yl)tetrazole (1)-catalyzed asymmetric aldol reac-tions, under continuous flow. Employing a glass microreactor (mixing zone = 161 µm [wide] × 1240 µm [deep] × 5.4 cm [long]; reaction channel = 391 µm [wide] × 1240 µm [deep] × 18.4 cm [long]), the authors evaluated the effect of reactant residence time and tem-perature on the 1 (5–10 mol%)-catalysed reac-tion bet ween 4-nitrobenzaldehyde (1.0 mmol) and acetone (2) in dimethyl sulfoxide (DMSO). Using a screening approach, the authors iden-tified that a reaction temperature of 60°C af forded an optimal yield of 79% and an enantioselectivit y of 76% . The se result s compared favourably to previously reported data where 81% yield and 79% enantiomeric excess (ee) had been obtained using 20 mol% of catalyst 1; the use of a flow reactor, how-ever, resu lted in a dramatic reduction in reaction time and the proportion of catalyst employed. Based on the findings of this initial Continuous-ow organic synthesis: a tool for the modern medicinal chemistMedicinal chemists are under increasing pressure, not only to identify lead compounds and optimize them into clinical candidates, but also to produce materials in sufcient quantities for subsequent investigation. With this in mind, continuous-ow methodology presents an opportunity to reduce the time taken to, rst, identify the compound and, second, scale the process for evaluation and, where necessary, production. It is therefore the aim of this review to provide the reader with an insight into the advantages associated with the use of continuous-ow chemistry through the use of strategically selected literature examples.Charlotte Wiles1† Paul Watts2†Author for correspondence1Chemtrix BV, Burgemeester Lemmensstraat 358, 6163JT Geleen, The Netherlands2Department of Chemistry, University of Hull, Cottingham Road, Hull, HU6 7RX, UKTel.: +44 148 246 6459Fax: +44 148 246 6410E-mail: c.wiles@chemtrix.comMicRoReactoRA device in which chemical reactions take place in a continuously owing stream with lateral dimensions greater than 1 mmMicRoReaction tech nologyThe application of microreactors to the task of intensifying chemical reactions and/or unit operations, such as separationslead coMpoundChemical compound in drug discovery that has pharmacological or biological activity and is used as a starting point for chemical modication in order to increase pharmacokinetic featuresFor reprint orders, please contact reprints@future-science.com 10 9 8OOOOOOHH60°C+1098OOOOOOHH60°C+Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1594 future science groupstudy, the authors eva luated the reaction of a series of aromatic aldehydes observing good yields and enantioselectivities for aldehydes 3, 4 and 5 (tabl e 1). However, poor yields were obtained for benzaldehyde (6) and 2-naphthyl-aldehyde (7) due to competing dehydration of the reaction product.CycloadditionsCycloadditions present a useful route to com-pounds containing six-membered rings and, while they offer synthetic diversity when gen-erating compound libraries, reaction conditions can be harsh in order to attain acceptable yields. Using a microcapillary flow disk, Ley and co-workers demonstrated the scalable synthesis of 3a,5,7a-trimethyl-3a,4,7,7a-tetrahydro-isobenzo-furan-1,3-dione (8) via the Diels–Alder reaction, depicted in Fi guRe 1 [19]. To perform a reaction, solutions of maleic anhydride (9) and isoprene (10) (2 eq.) were brought together, under pres-sure-driven flow, where they mixed prior to heating (60°C) in a microcapillary flow reactor. Employing a flow rate of 6 ml min-1 (residence time = 28 min), the authors obtained the target compound 8 in a 98% yield, which equates to an impressive throughput of 3.9 kg day-1.More recent ly, Organ and co-workers reported a series of Diels –Alder reactions conducted in Pd-coated capillaries, identify-ing the Pd film as having a dual role: that of a heat source and a catalyst when microwave irradiation was employed [20] . As table 2 illus-trates, when conducting the reaction in the absence of the Pd-coating, in an oil-bath at 205°C, 54% conversion to (1R,4S)-dimethyl-7-oxabicyclo[2.2.1]hepta-2,5-diene-2,3-carbox-ylate (11) was obtained; this was subsequently increased to 72% upon employing a Pd-coated capillary. Utilizing microwave irradiation as a means of heating the flow reactor, the authors obtained 10% conversion to 11 in the absence of Pd; 47% conversion to 11 when the reactor was coated on the outside (due to a thermal effect) and 90% conversion to 11 when the Pd coating was in contact the with reactant solution (cata-lytic effect). Based on these findings, a series of Diels–Alder cycloadditions were performed, whereby conversions were consistently higher than those obtained in an oil bath. Click chemistryAs can be seen throughout this review, click chemistry is a powerful synthetic tool used for the synthesis of an array of pharmaceutical inter-mediates and products. In an example demon-strating the use of multiple solid-supported cata-lysts and reagents, Ley and Baxendale reported the design and application of a modular flow reactor in which 14 1,4-disubstituted-1,2,3-triazoles were synthesized [21]. Coupling an immobilized copper(I) iodide reagent (12) with two immobilized scavengers, QuadraPure-TU (thiourea) (13) and a phosphane resin (14), the authors investigated the [3 +2] cycloaddition of an array of azides (0.15–0.20 M) and ter-minal alkynes (0.1 M; 30 µl min-1) to afford 1,4-disubstituted-1,2,3-triazoles in excellent yield ( table 3). In analytical mode, the reactor Table 1. Results obtained for the continuous-flow aldol reaction between acetone 2 and a series of aromatic acceptors.260°C, DMSOOOAr HNHHN NNN1OArOH+Ar 1 (mol%) Residence time (min)Yield (%)*ee (%)‡4-F3CC6H4 (3) 5 30 77 684-NCC6H4 (4) 5 30 77 702-BrC6H4 (5) 5 30 78 75C6H5 (6) 10 30 38 (18)§632-Naphthyl (7) 10 30 44 (19)§57*After purification.‡Determine d by HPLC.§Dehydration product.ee: Enatiomeric excess.Table 2. Microwave effect on Diels–Alder reactions conducted in flow. 205°C, 1-5 barDMSOOCO2MeMeO2COCO2MeCO2Me11Capillary treatment Temperature (°C) Heat source Conversion (%)*None 205 Oil bath 54Pd-coated, inside 205 Oil bath 72None 115 MW 10Pd-coated, outside 205 MW 47Pd-coated, inside 205 MW 90*Determined by 1H NMR spectroscopy.MW: Microwave irradiation. Figure 1. Diels–Alder cycloaddition performed under continuous flow.Meso-ReactoRDevice in which chemical reactions take place in a continuously owing stream with lateral dimensions of typically 1–5 mmpRocess intens iFicationThe reduction in the size of equipment used for production while the desired output volume is maint ainedlaMinaR FlowNonturbulent ow of a viscous uid, resulting in the formation of coowing uid layers or streams; whereby mixing is diffusion-limited and occurs at the interface between uid streams 17MeOH, 100°C , 30 minOR H18ON2POOMeOMe KOtBuHRUnreacted aldehyde andphosphoric residues123HR70°C19NH2NMe2SOHO OContinuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1595future science groupwas optimized to afford 20–200 mg of product; however, in the case of propargylic alcohol (R1 = CH2OH) and benzyl azide (R1 = CH2Ph), the reactor was operated continuously for 3 h, to afford 1.50 g of the desired product (15 16) in 85% yield and greater than 95% purity. Therefore, by employing an inline scavenger module, the authors were able to synthesize the target compounds in excellent purity, compared with more traditional methodology.Using the Bestmann–Ohira reagent (17) combined with a series of polymer-supported reagents, Baxendale and co-workers subsequently investigated the multistep synthesis of terminal alkynes under continuous flow [22]. As Fi guRe 2 illustrates, the protocol employed involved reacting the aldehyde under investigation (0.13–0.15 M) with Bestmann–Ohira reagent (17) (0.10 M), in the presence of KOtBu (18) (0.12 M) at 100°C, in a polytetrafluoroethylene (PTFE) reactor (residence time = 30 min). Upon exiting the reactor, the reaction products were pumped through a series of scavenger columns in order to remove unreacted starting materials and acidic residues. The resulting terminal alkyne was obtained in excellent purity and moderate yield (65–82% yield) upon evaporation of the reaction solvent. In the case of polar aldehydes, such as N-methyl-2-formylindole, the polar acetylene was found to undergo hydrolysis to afford the respective ketone; to prevent this, the authors replaced amberlyst-15 (19) with alumina, enabling isolation of polar acetylenes in 64–78% yield. This approach provided a facile means of generating diverse libraries of acetylenes suitable for subsequent click-chemistry applications. C-C cross-coupling reactionsAnother area of significant interest to those in the field of continuous-flow synthesis is that of C-C cross-coupling reactions, with an array of examples reported, highlighting reduced reac-tion times, increased product purity and ease of catalyst recycle, a broader overview of which can be found in a recent review by Parmar et al. [23]. Combining the emerg ing technologies of microwave and microreaction chemistry, Comer and Organ demonstrated an array of advantages when conducting reactions within glass capillar y reactors (200 µm [internal diameter]) [24] . Introducing reactants into the reactor through a stainless steel mixing chamber, the coupling of 4-iodooct- 4-ene (20) (0.20 M) and 4-methoxyboronic acid (21) (0.24 M), using palladium tetrakis(triphenylphosphine) (22) as the catalyst, was investigated (Fig uR e 3). Under 100 W, the authors obtained 100% 1-methoxy- 4 - (1-propylpent-1-enyl)benzene (23) with a residence time of 28 min, leading them to investigate an array of aryl halides and boronic acids, obtaining the tar-get compounds in 37–100% yield. Compared with conventional batch reactions, these results Table 3. Selection of the 1,4-disubstituted-1,2,3-triazoles synthesized under continuous flow.R N3R1HNNNRR1NMe2.CuINH1.2. NH2S3. PPh3121314Triazole Yield (%)NNNHONO293NNNHO 1585NNNPh 1685NNNSOOCF391OOOHNNNPhCF388Figure 2. Reaction sequence used for the continuous-flow synthesis of terminal alkynes. I20MeOB(OH)2MeO21 2322+Pd(PPh3)4 22THF, MWH2NOHOONOH2735 min+H2NOHOONOH2735 min+Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1596 future science groupillustrate dramat ic improvements in y ield and purity, largely due to the suppression of competing side reactions.More recently, Kirschning et al. evaluated the activity of an immobilized oxime-pallada-cycle (24), towards a series of cross-coupling reactions, including the Suzuki–Miyaura and Heck–Mizoroki reaction [25]. Employing a flow reactor, comprising poly(vinylpyridine)-coated Raschig rings, functionalized with a palladacycle (10 mmol Pd ring-1), the authors initially evalu-ated the Suzuki–Miyaura reaction, as illustrated in table 4. Reactions were conducted by circu-lating a solution of aryl bromide (1.0 mmol), boronic acid (25) (1.5 mmol) and CsF (26) (2.4 mmol), through the heated reactor (100°C) at 2.5 ml min-1. After a period of 24 h, the reac-tor was washed with DMF-H2O and the wash-ings diluted with H2O prior to extraction with EtOAc and purification by flash chromatogra-phy. It is important to note that some leaching of Pd was experienced during this investigation and if the technique is to be used for an industrial application, this would first have to be resolved to ensure high product purity and sufficient catalyst lifetimes. C-heteroatom bond-forming reactionsThe formation of C-X bonds is of fundamen-tal importance in the synthesis of biologically active compounds. As such, many examples have been evaluated within continuous-flow reactors, including the large-scale synthesis of carba-mates [26] , the esterification of benzodiazepine ligands [27], preparation of deuterium-labeled amides [28], a-aminonitriles [29,30] and chrome-nones [31] , multistep peptide couplings [32 ,33] and polymer-assisted syntheses of oligonucle-otides [34] ; those of significant importance are detailed in the following section. In addition to the many challenges associated with the manipulation of large volumes of haz-ardous materials, the performance of exothermic reactions can be problematic on a large scale due to the inefficient removal of heat from stirred tank reactors. With this in mind, Schwalbe et al. investigated the advantages associated with per-forming the Paal–Knorr synthesis under continu-ous flow, utilizing the CYTOS® college system (FiguRe 4) [35]. Unlike batch methodology, where dropwise addition of reactants was required in order to maintain thermal control, perform-ing the reaction in a stainless-steel flow reactor (residence time = 5.2 min) enabled the use of neat reactants, affording 2-(2,5-dimethyl-1H- pyrrol-1yl)ethanol (27) in 91% yield (260.0 g h-1).In a second example, Warrington and co-workers demonstrated the synthesis of fane-tizole (28) , a compound used for the treat-ment of rheumatoid arthritis, within a glass flow reactor [3 6] . As Fi gu Re 5 illustrates, the active ingredient was synthesized via reaction of 2-bromoaceto phenone (29) (1.4 × 10-2 M) and phenylethylthiourea (3 0) (2.1 × 10-2 M) at 70°C in NMP. Under the aforementioned condit ions, the t arget comp ound 28 was obtained in 99% conversion, as determined by offline LC –MS. ArylationsProviding an alternative to the Buchwa ld–Hartwig cross-coupling reaction, Stevens et al. investigated a series of copper-mediated N- (tabl e 5) and O-arylations with arylboronic acids, within a stainless steel continuous-flow reactor [37]. The copper-mediated heteroatom arylation has grown in use since its report in the late 1990s [38,39,4 0]; however, as with many synthetic reactions, the exothermic nature of this transformation makes performing it on a large scale problematic. With this in mind, the authors investigated the ability to prepare Table 4. Results obtained for Suzuki–Miyaura under continuous flow.R1Br B(OH)225R1PdClNOHN24CsF 26, DMF-H2O 100°CR1Residence Time (h) Yield (%)H 24 89CH324 56COCH39 91OCH324 50Figure 3. Suzuki–Miyaura reaction conducted using microwave heating.Figure 4. An exothermic reaction conducted in a CYTOS® flow reactor. NHNH2SOBrNHSN70°CNMP282930+NHNH2SOBrNHSN70°CNMP282930+B(OH)2NHNClOSNNClOS31 2535+Cu(OAc)2 32DCM, Et3N 33/Py34 (1:2) rt, 0.2–2 hB(OH)2NHNClOSNNClOS312535+Cu(OAc)232DCM, Et3N33/Py34 (1:2) rt, 0.2–2 hContinuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1597future science groupmoderate quantities of the target compounds by conducting the reactions in a commercially available microreactor (CYTOS).In order to develop their methodology, initial investigations were conducted using the model reaction illustrated in Fi guRe 5. To optimize the reaction conditions, a premixed solution of pyr-azinone (31) (1.0 mmol) and boronic acid (25) (2.0 mmol) was introduced into the microreac-tor, where it mixed with a solution of Cu(AcO)2 (32) (1.0 mmol)/ Et3N (33) (1.0 mmol)/pyri-dine (34) (2.0 mmol) and the effect of flow rate was evaluated (residence times = 0.25–2 h). Compared with conventional batch methodol-ogy, conducting the reaction under flow pro-vided a greater than 80% reduction in reaction time, affording 90% conversion within 1.5 h. By systematically increasing the concentration of reactants to 10.0 mmol, the authors increased the throughput to afford N-arylated pyrazinone (35) in 3.92 mmol h-1. The investigation was subsequently expanded to evaluate the tech-niques scope (table 5), obtaining moderate-to-high yields for all anilines with the exception of 2,4,6-trichloroaniline (36) (steric effect).AzidationsDue to the risk of hydrazoic acid (HN3) build-up within the headspace of batch vessels, stringent engineering controls are required when perform-ing the synthesis of organic azides. By compari-son, continuous-flow reactors can be completely filled, removing the space for HN3 build-up and, thus, reducing the risk of explosion asso-ciated with the large-scale preparation of such compounds. With this in mind, Kopach and co workers recently communicated a facile tech-nique for the continuous-flow synthesis of organic azides from benzyl halides [41]. As FiguRe 7 illus-trates, the model reaction under investigation was the azidation of 3,5-bis-(trifluoromethyl)benzyl chloride (37) to afford 1-(azidomethyl)-3,5-bis-(trifluoromethyl)benzene (38); a mole-cule of interest to the authors due to its use in the synthesis of NK1-antagonists 39–40 (FiguR e 8). Employing a 316 stainless steel tube reac-tor (dimensions = 2.16 mm [outside diam-eter] × 0.64 mm [internal diameter] × 63.1 m [long], reactor volume = 20 ml), housed within an oven, the authors investigated the effect of temperature on the formation of azide (38). Reactions were conducted by mixing a solu-tion of 3,5-bis-(trifluoromethyl)benzyl chloride (37) (0.66 M, 745 µl min-1) in DMSO with an aqueous solution of sodium azide (41) (2.32 M, 255 µl min-1, 1.2 eq.) at room temperature, prior to pumping the reaction mixture through the heated reactor (50–90°C) for 20 min. Using this approach, the authors identified 90°C as the opti-mal temperature, whereby 97.3% conversion of 37 to azide 38 was obtained. Once optimized, the reactor was operated for 2.8 h, after which time the reaction products were partitioned between n-heptane and water to isolate 1-(azidomethyl)-3,5-bis- (trif luoromethyl)benzene (38) (94% yield) at a throughput of 32.1 mmol h-1. In addi-tion to increasing process safety, the use of a continuous-flow reactor enabled access to higher operating temperatures, which led to reduced reaction times, affording an efficient method for the large-scale production of organic azides. In a second example, Ley and coworkers reported the fabrication of an azide monolith (2.00 mmol g-1) within a flow reactor (15 mm Table 5. N-arylations conducted under continuous flow.NH2R1B(OH)2R2Cu(OAc)232DCM, Et3N33/Py34(1:2) rt, 2hHNR1R2R1R2Isolated yield (g) Isolated yield (%)H H (25) 1.2 71H 3-OEt 1.5 734-Cl H (25) 1.3 674-Cl 3-OEt 1.7 692,4,6-Cl (36) 3-OEt No reaction 02,4-NO23-OEt 1.7 56Figure 5. Protocol used for the synthesis of fanetizole (28). Figure 6. N-arylation of a pyrazinone (31). F3C CF3Cl37F3C CF3N338NaN34190°C, 20 minF3CCF3Cl37F3CCF3N338NaN34190°C,20minReview | Wiles WattsFuture Med. C hem. (20 09) 1(9)1598 future science group[internal diameter] × 10 cm [long]), demonstrat-ing its use for the synthesis of organic azides [42]. Employing a solution of acyl chloride (1.0 M) and a residence time of 13 min, the respective acyl azide was obtained in excellent yield and purity upon removal of the reaction solvent in vacuo. Compared with the solution-phase exam-ple discussed previously, this technique is advan-tageous, as the reaction can be performed in the absence of water, making it suitable for readily hydrolyzable precursors. The synthetic utility of the technique was subsequently demonstrated for the in situ decomposition of acyl azides (120°C), to afford the respective isocyanate in high-to-excellent yield (64–90%). Oxidations reductionsOxidationsAlthough synthetically useful, oxidation reac-tions are problematic when conducted on large scales due to the formation of byproducts (via overoxidation of target compounds), the use of toxic reactants, the generation of large exo-therms and the accumulation of hazardous inter-mediates. To address these points, several authors have investigated the transfer of oxidations to flow processes; these include Haswell et al. [43] (1° alcohols), Wiles, Hammond and Watts [4 4] (epoxidation of alkenes), and Kemperman [45] (Swern–Moffat oxidation).In a recent example, Uozimi, Yamada and Torii reported a facile method for the contin-uous-flow synthesis of carbinols, utilizing the potentially explosive reagent Oxone® (42) [46] . The presence of such functionality within thera-peutic agents and biologically active compounds means there is currently considerable interest in the development of rapid and efficient methods suitable for large-scale applications. Within a heated microtubular reactor (1 mm [internal diameter] × 5.0 cm [long]), the authors evalu-ated the oxidative cyclization of (Z) -4-decen-1-ol (43) (0.05 M in iPrOH) in the presence of Oxone (0.10 M in H2O). Employing aqueous Table 6. Results obtained for the oxidative cyclization of alkenols conducted in flow.R2R1OHnnOR1HO R2Oxone 42iPrOH/H2O80°C, 5 minAlkenol Product Conversion (%)OH43OHO 4499OHOHO90OHOHO88HOOOH90*HOOHO70**Flow rate = 2 µl min-1 (per reactant); 80°C, residence time = 10 min.Table 7. Results obtained for the epoxidation of alkenes using in situ-generated HOF:MeCN 45.(i) MeCN/H2O 10 % F246 on N2(ii) 45HOF:MeCN 45RRRROAlkene Epoxide Yield (%)C10H21C10H21O98PhPhOPhHHPh99PhPhOPhPhHH81PhCO2EtOHPhCO2EtH39*63‡OO94*1 eq. 45; ‡2 eq. 45.Figure 7. Model reaction used to evaluate the synthesis of organic azides derived from benzyl halides. 39NNNF3CCF3NOCl40NNNF3CCF3NOOONHOCl39NNNF3CCF3NOCl40NNNF3CCF3NOOONHOClContinuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1599future science groupNa2S2O3 as a quench agent, the authors collected several fractions of the reaction products and analyzed them by GC and NMR spectroscopy in order to confirm the development of a stable reaction system. Utilizing a reaction tempera-ture of 80°C and a residence time of 5 min, the authors obtained the corresponding cyclic ether, treo-1-(2-tetrahydrofuranyl)hexan-1-ol (44), in 99% conversion. As tabl e 6 illustrates, the reac-tion protocol developed could also be applied to the synthesis of pyranyl derivatives, affording all cyclic ethers in high yield. Using the highly effective oxidizing agent HOF:MeCN (45), Murray and Sandford dem-onstrated the epoxidation of alkenes within a nickel/polychloro trifluoroethene reactor [47]. To increase the safety associated with this transfor-mation, the authors demonstrated the in situ prep-aration of HOF:MeCN from elemental fluorine (46), with the alkene under investigation subse-quently reacted (1:1) at room temperature. Under the aforementioned conditions, the authors dem-onstrated the epoxidation of numerous alkenes, a selection of which can be found in table 7.ReductionsThe dibal-H (47) -mediated partial reduction of esters to aldehydes is of considerable interest in the pharmaceutical industry for the tonne-scale production of aldehyde raw materials. As such, Ducry and Roberge evaluated the effect of oper-ating temperature on yield and selectivity for the model reaction illustrated in Figu Re 9 [48]. In order to obtain the desired selectivity, batch reac-tions are conducted at -65 to -55°C; however, using a multi-injection concept, the authors were able to perform the partial reduction of methyl butyrate (48) in a microreactor at temperatures as high as -20°C, obtaining butyraldehyde (49) in 89% [11% n-butanol (50)] compared with 63% in batch. The ability to operate reactors at ele-vated temperatures and obtain increased selec-tivities has the potential to significantly reduce operating costs associated with such reactions.Possibly the largest area of synthetic chem-istry to benefit from advances made in the field of flow-reaction methodology is that of hydrogenations, where the development of commercially available reactors (H-cube™) have enabled researchers to perform triphasic reactions with increased ease and reproduc-ibility as t he catalysts are contained within cartridges, removing the hazards associated with handling catalytic material. In addition to dramatically reducing the time taken to per-form common hydrogenations, typically from Table 8. Model reactions performed using continuous-flow hydrogenation apparatus*.Substrate Product Catalyst Temperature (°C)Yield (%)‡NHO2NNHH2N10% Pd/C 25 95HN PhNH210% Pd/C 80 89Ph PhPhPh10% Pd/C 25 94OO10% Pd/C 25 85HNOHNH2Raney-Ni 70§88NCH2N2Raney-Ni 70§99*Unless other wise stated, reactions were conduc ted at 1 bar.‡Isolated yields.§70 bar.Figure 8. NK1 antagonists 39 and 40, which employ azide 38 as a raw material. CO2Me48Dibal-H (47)H49ODibal-H (47) OH50R1N3R2NNNR1R2+bCAll 52 in PBSaq. buffer (pH 7.4) 37°C, 40 hReview | Wiles WattsFuture Med. C hem. (20 09) 1(9)1600 future science grouphours to minutes, the hazards associated with hydrogenations are reduced by generating H2 in situ via the electrolysis of H2O; removing the need for H2 cylinders within the research laboratory. Recent examples reported by Darvas and co-workers illustrate the reduction of nitro groups, alkenes, oximes and nitriles, along with debenzylations ( table 8) [49] .  FluorinationsAs a large number of active compounds con-tain fluorinated moieties, efficient methods for their introduction are often sought. With this in mind, numerous examples using elemental F2 (46) have been demonstrated by Chambers and Sandford [50,51], with Miyake and co-work-ers reporting the incorporation of fluorine into small organic molecules via Michael additions, Horner–Wadsworth–Emmons reactions and trifluoromethylations [52]. In a recent communication, Seeberger reported the use of diethylaminosulfur trifluoride (51) within a continuous-flow reactor, as a means of increasing the ease with which aldehydes, car-boxylic acids and alcohols could be converted to the respective fluorinated analogue [53]. Employing a heated PTFE tube reactor (70°C) and a residence time of 16 min, the authors demonstrated a series of deoxy fluorinations using dichloromethane as the reaction solvent. As table 9 illustrates, this approach afforded the target compounds in moderate-to-excellent yields, following an offline aqueous extraction and purification via column chromatography.  Niche applications In addition to the use of conventional reaction methodology under flow conditions, the flex-ible reactor configurations attainable enable the large-scale application of techniques that have usually been reserved for niche applications, such as electrochemistry [54,55], photochemistry [56,57], biocatalysis [58,59] and radiopharmaceutical syn-thesis [60,61]. Based on these initial examples, it is hoped that the use of these techniques will spread beyond the laboratory, affording elegant meth-odology for the production-scale synthesis of complex organic molecules (see the biologically active compounds prepared under continuous flow section and the examples therein).  High-throughput screening While target-guided synthesis has been used with moderate success for the preparation of various protein inhibitors [62], broader application of the Table 9. Deoxyfluorinations conducted within a heated polytetrafluoroethylene reactor.Et2N SF351(1.0 to 2.0 eq.)ROHOR HOR OHR FOR HR FFFor or or orSubstrate Product Yield (%)*OHF70‡H HHHOHHF61§OOHBnOBnO OBnOBnOFBnOBnO OBnOBn(α:β1:4)100MeOOOHMeOOF100OHMeOHMeOF F89*Isolated yield.‡5:1 mixture of diaste reomers.§6:1 mixture of diastereomers. Figure 9. Model reaction used to evaluate flow reactors for the large-scale production of aldehydes.Figure 10. In situ click chemistry reactions performed in an integrated microfluidic device. SH2NOOSH2NOONHOSH2NOONHOSH2NOONHOOHOOH2NOOHFmocHNClONHN3ONHN3N3H3CO SONH2OOO NHN3NOON3H3COOON3ONHN3SHNOH3COON3N3BrFNHON3ONHN3O OON3FNN3ONHON3ONHN3SOON353a 53b 53c 53d53e 53f 53g 53hSH2NOOSH2NOONHOSH2NOONHOSH2NOONHOOHOOH2NOOHFmocHNClONHN3ONHN3N3H3COSONH2OOONHN3NOON3H3COOON3ONHN3SHNOH3COON3N3BrFNHON3ONHN3OOON3FNN3ONHON3ONHN3SOON353a53b53c53d53e53f53g53hAcetylenesAzides54a 54b 54c 54d54e 54f 54g 54h54i54j 54k 54l54m54n 54o 54pContinuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1601future science grouptechnique has been hampered by the high con-sumption of frequently scarce/expensive reagents and the inflexibility of the synthetic platforms used, restricting the number of permutations per-formed. With this in mind, Tseng and coworkers investigated the use of an integrated microfluidic device as a means of rapidly producing poten-tial biligand inhibitors, by assembling azide and acetyl ene building blocks within the binding pockets of target proteins (FiguRe 10) [63]. Coupling a multiplexed approach with improvements in screening techniques, the authors aimed to reduce the time taken to pre-pare and screen such compounds. Using this approach, the authors focused their efforts on reducing the quantities of reagent employed for the preparation of such triazoles [64] , initially demonstrating a two to 12-fold reduction in reactant volume compared with a standard 96-well approach ( table 10) [65]. Table 10. Comparison of time and reactant savings made by performing screening reactions under flow.Parameter Type of reactor96-well plate First generation Second generationNumber of reactions 96 32 1024Enzyme (bCAII 52) (µg) 94.00 19.00 0.36Acetylene (nmol) 6.00 2.40 0.12Azide (nmol) 40.00 3.60 0.12Total reaction volume (µl) 100.00 4.00 0.40Sample preparation time Few min 58 s 15 sDetection method LC–MS LC–MS MS/MSHit identification time 40 min 40 min 15 sFigure 11. Reactants employed in the large-scale click chemistry screen. OMeOOOO2N55Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1602 future science groupIn addition to affording the same experimental fidelity and a significant reduction in sample vol-umes, the system was readily automated, which enabled each reactant mixture to be dosed in less than 1.0 min. Regardless of the speed of sample preparation, however, the limiting factor was the analytical evaluation, which took approximately 40 min per sample. Consequently, the authors developed a second microfluidic device, which comprised four functional fluidic components:A microfluidic multiplexer, to regulate the individually addressed reagent feeds; A 150-nl rotary mixer;A 250-nl serpentine channel in which reagent mixing was completed; A PTFE tube (20 cm long) in which the reac-tion slugs were collected prior to off-line screening by MS/multiple reaction monitoring. Using a reversed-phase clean-up step (to remove polar charged reagents) enabled the use of ESI–MS, which reduced the ana lysis time from 40 min to 15 s per sample. In addition, the sensitivity of the analytical technique was improved reducing the required sample volume and leading to a 20–50-fold decrease in reagent consumption ( table 10).Having ach ieved their g oal of reduc-ing reac tant consumption and increasi ng re action processi n g /screening ti me, the authors demonstrated the synt hetic utilit y of their reactor, investigating four different reaction conditions:128 Cu(I)-catalyzed reactions;128 duplicated reactions between acetylenes 53a–h and azides 54a–p (Fig uRe 11), using the en z yme bovine c a rbonic anydra s e I I (bCAII) (52) ; As above, in the absence of bCAII; As above, in the presence of an inhibitor to confirm active site specificity.As table 11 illustrates, using this approach, the authors were able to rapidly identify 35 compounds as hits and four as modest hits, with ESI–MS multiple reaction monitoring Table 11. Compounds identified as a hit using multiple reaction monitoring-based identificationAzides (54a–p) Acetylenes (53a–h) 53a b c d e f g h54a No hit Hit No hit No hit No hit Hit Hit No hit bNo hit No hit No hit No hit No hit No hit No hit No hit cNo hit No hit No hit No hit No hit No hit Hit No hit dNo hit No hit No hit No hit No hit No hit No hit No hit eNo hit No hit No hit No hit No hit No hit No hit No hit fNo hit No hit No hit No hit No hit No hit No hit No hit gNo hit No hit No hit No hit No hit Hit No hit No hit hHit Modest hit No hit No hit No hit Hit Hit No hit iNo hit No hit No hit No hit No hit No hit No hit No hit jHit No hit No hit No hit No hit Modest hit Hit No hit kNo hit Hit No hit No hit Hit Hit Hit No hit lHit Modest hit No hit No hit No hit Modest hit Hit No hit mHit Hit Hit No hit Hit Hit Hit No hit nHit Hit No hit No hit No hit Hit Hit No hit oHit Hit No hit No hit Hit Hit Hit No hit p Hit Hit No hit No hit No hit Hit Hit No hit Hit: 15%; Modest hit: 10–15%; No hit: 5% response.Figure 12. Naproxcinod (55), a pharmaceutical synthesized on a pilot-scale using flow reactor methodology. ROHHORONO2HORONO2O2NO56HNO357+ROHHORONO2HORONO2O2NO56HNO357+NOONHHHH58NOONHHHH58Crude aq. alkaloid(1% formic acid + 1%)NH3, 1:2 v/v; pH 9–10Organic phase1% formic acid (pH 3.5)Polar compounds(basic aq. phase)Nonpolar compounds(organic phase)Pure alkaloid 58(acidic aq. phase)Crude aq. alkaloid(1% formic acid + 1%)NH3, 1:2 v/v; pH 9–10Organic phase1% formic acid (pH 3.5)Polar compounds(basic aq. phase)Nonpolar compounds(organic phase)Pure alkaloid58(acidic aq. phase)Continuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1603future science grouptechnology again proving advantageous as it enabled computerized interpretation of the results. Using this approach, the authors predict that this technology has the potential to increase the throughput of enzyme-inhibitor discovery.  Process chemistry The role of process chemistry is to develop a synthetic route used for the preparation of milli-grams of compound into a robust, scalable and economically viable method, suitable for the preparation of kilograms and, potentially, tonnes of material, should it be identified as an API. It is this final bottleneck that can cause a discovery program to fail, irrespective of how promising the screening results have been thus far [66]. With compounds becoming bigger and more stereo-chemically demanding, chemists are faced with an increasingly difficult task of reproducing such synthetic methodology on a larger scale. When employing continuous-flow reactors, it is the rapid nature with which the reaction scale can be altered that is of particular interest to the process chemist as, unlike conventional upscaling, which involves increasing the reac-tor volume to achieve the desired production volumes, flow reactors use an approach known as scale out or numbering up. This principle is based on the use of multiple reactors in paral-lel, with each controlled using the same condi-tions, identified by the R D chemist, whereby the number of reactors employed is chosen to attain the required process throughput. As the ability to rapidly scale a reaction is viewed as a ‘strategic competitive advantage’, R D and process chemists are beginning to see the value in the use of continuous-flow reactor technology at all stages of the development process [67–69]. Pilot-scale synthesis using continuous-ow reactorsExploiting the reaction control and accompany-ing safety aspects associated with flow reactor methodology, DSM recently disclosed the results of an investigation into the large-scale prepara-tion of the pharmaceutical, Naproxcinod (55) (Fig uRe 12) [70]. Employing glass-based micro-reactors, designed in conjunction with Corning Inc. [71], the researchers developed a pilot-scale process capable of delivering hundred s of kilograms of material 55.As nitration reactions often prove difficult to perform on a commercial scale, one of the chal-lenges faced was designing a system capable of producing volumes in the range of tonnes per day. Although, in this case, the reaction enthalpy was moderate, it was the potential exothermic Table 12. Effect of AS/S ratio on the particle size and concentration of solubilized danzol (62) at 30°C. AS/S ratio(v/v)Solubility of 62(µg ml-1)Average particle size (nm)0 35000.00 NA*1 1336.03 NA*2 79.30 NA*5 4.40 125010 2.83 90020 2.16 51040 1.63 50520‡1.63 364*No precipitation observed.‡Reactor temperature 4° C.AS/S: Antisolvent /solvent; NA: Not applicable. Figure 13. Selective nitration performed on a pilot-scale by DSM.Figure 14. Alkaloid strychnine (58).Figure 15. Three-phase extraction principle employed for the isolation of the alkaloid strychnine (58). The dashed line represent pillars within the microchannel and the shaded area represents the hydrophobic microchannel. NHOHOONHOHOOImmobilizedacylaseH2NOHO60 59 60++NHOHOONHOHOOImmobilizedacylaseH2NOHO605960++Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1604 future science groupdecomposition of the product 55 that presented the safety risk; careful control of the reac-tion conditions is therefore paramount when conducted on a production scale.To perform t he organic nitration, sub-strate 56 and solvent were brought together in a glass microstructure to afford a fine emul-sion [72], this was followed by the addition of neat HNO3 (57); immediately initiating the reaction (FiguRe 13). After the desired period of time, water was added to quench the reaction, prior to neutralization, which was performed in stages to maintain control over the reactor temperature. Using this approach, the authors evaluated a reactor volume of 150 ml, which was capable of producing 13 kg h-1 of Naproxcinod (55). Having demonstrated the viability of the system, no scale-up step was required in order to reach the production targets set; the authors simply employed a production unit comprising eight micro reactors, all operated under the afore-mentioned conditions. Employing this strategy enabled the users to produce 100 kg h-1 of 55, equivalent to annual production volumes of 800 tons. This investigation therefore goes some way towards illustrating the feasibility of perform-ing hazardous reactions on a large scale through the use of continuous-flow methodology and the ability to increase production volume without the need to re-optimize the process. Postsynthetic application of ow technologyAlthough numerous advantages of flow tech-nology have been illustrated thus far, a remain-ing challenge is that of product purification, which raises the question: do you collect the reaction products and perform conventional batch-wise purificat ions, negating some of the advantages harnessed by the use of a flow process, or employ inline solutions to this problem? Until now, examples of both tech-niques have been reported, with users favoring batch-wise protocols due to familiarity with this approach. However, when constructing libraries of compounds, this step can prove time consuming a nd, as such, inline prod-uct purification is widely considered to be a solution to this problem. Liquid–liquid extractionsOne of the most common purification tech-niques used in synthetic chemistry is that of liquid–liquid extraction (LLE), a technique that has been shown to be suited to continu-ous processing due to the high interfacial areas obtained between phases within microfluidic systems [73]. An early example of this approach was reported by Kitamori and coworkers, who observed that by conducting extractions within a Y-shaped microchannel, extraction efficiencies were at least one order of magnitude greater than those obtained in a conventional separating funnel [74] . Based on these initial observations, the authors recently reported the development of a circulation microchannel as a means of further increasing reaction efficiency, whilst reducing the volume of solvent required to perform a LLE [75] . In more complex situations, it can be neces-sary to perform three-phase extractions, where an analyte is extracted from an aqueous phase (feed phase) into an organic phase (transport phase) and then into a second aqueous phase (acceptor phase) [76] . An example of a three-phase separation conducted under continuous flow was recently communicated by Beek and co-workers [77], whereby a system capable of per-forming the ‘simultaneous extraction and back extraction’ of strychnine (58) (Figu Res 14 15) was developed [78]. Using a glass reactor with a contact length of 3.56 cm (channel dimen-sions = 100 µm [wide] × 40 µm [deep]), semi-toroidal pillar structures between the channels and a hydrophobic coated middle channel, the authors investigated the purification of a crude alkaloid solution (containing some polar and non polar impurities) using offline HPLC ana lysis to monitor the extraction efficiency.Employ ing a ba sic aqueous feed phase (0.5 µl min-1), a transport phase of CDCl3 (1.0 µl min-1) and an acceptor phase of aqueous formic acid (0.5 µl min-1), the crude alkaloid (strych-nine nitrate) underwent a simultaneous extrac-tion and back-extraction with a residence time of 25 s. Using this approach, 91% strychnine (58) was extracted from the feed phase into the trans-port phase and 93% from the transport phase into the acceptor phase. Having demonstrated the extraction of an alkaloid from a simulated Figure 16. Reaction protocol employed for a continuous-flow resolution. OORON3OClIn tolueneNaN3410.40 M in aq. NaOHIn aq. toluene Phase separator aq. waste105°C -N2NCOR OHHNWhere R = CH3, C2H5 or CH2PhON3In tolueneOORhCl(PPh3)3 61H2 (1 atm)Toluene, rtContinuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1605future science groupreaction mixture, the authors subsequently investigated the purification of a strychnos seed extract, whereby alkaloids 58 and Brucine were separated from a series of polar impurities. This latter example illustrates the potential applica-tion of micro reactors as tools for the rapid iso-lation of active ingredients from natural prod-ucts, to provide the medicinal chemist with biologically active target compounds.Inline continuous separationsAs can be seen from the examples described thus far, numerous authors have demonstrated the fabrication of separation devices capable of purifying the reaction products generated under continuous flow; as such, several authors have incorporated these techniques into systems used for chemical synthesis. Combining continuous-f low biocatalysis with an inline LLE, Maeda and co-workers demonstrated the abil it y to resolve race-mic amino acids under continuous f low (Fi guR e 16) [79,8 0]. Using immobilized amino-acylase, the first step of the reaction was an enzyme-catalyzed enantioselective hydrolysis of (rac) -acetyl-phenylalanine to afford l-phenyl-alanine (59) (99.2–99.9% ee) and unreacted acetyl-d-phenylalanine 60. Acidif ication of the reaction products, prior to the introduc-tion of EtOAc, enabled efficient extraction of l-phenyl alanine 59 into the aqueous phase, whilst acetyl-d-phenylalanine remained in the organic phase (84–92% efficiency). Using 0.5 µl min-1 for the enzymatic reaction and 2.0 µl min-1 for the LLE, the authors were able to resolve 240 nmol h-1 of the racemate. In 2007, Jensen and coworkers developed a LLE device that was based on capillary forces via the exploitation of selective wetting surfaces [81]. Separations were achieved using a thin porous fluoropolymer membrane (0.1–1 µm pores) that wetted nonaqueous solvents, enabling the separa-tion of aqueous slugs (segmented flow) in a con-tinuous organic phase or fluorous-aqueous sys-tems and partially miscible compounds. Proof of principle was demonstrated using aqueous dyes, with visual inspection confirming efficient sepa-ration of the phases and the technique further exemplified with the efficient separation of DMF from dichloromethane (DCM)/water and ether/water. The LLE was subsequently integrated into a multistep synthetic process, whereby the authors reported the continuous-flow synthesis of carbamates (Fi guRe 17) [82]. A previous limitation of continuous-flow chemistry was perceived to be the need to per-form all steps in the same reaction solvent; this example clearly illustrates the feasibility of per-forming multistep reactions in which separations are required in order to remove byproducts, spent reagents or even change the reaction solvent.Solid-phase sequestration of trace metalsWith an increasing number of synthetic reactions performed within the pharmaceutical industry that employ homogeneous metal-based cataly-sis and the quantities of trace metals tolerated Figure 17. Multistep reaction used to demonstrate the continuous flow liquid–liquid separation of water.Aq.: Aqueous.Figure 18. Wilkinson’s catalyst (61)-mediated hydrogenation used to demonstrate the sequestration of Rh under flow conditions. HHHHOON62HHHHOON6224% aq. N2H4.H2O6693°C, 106 min, 96%OHOOEtHOHONHHNH265 64i) 85% H3PO4672°C, 2 minii) 10% aq. NaNO268, 2 min, 94%OHONHN69NDCE/EtOH (1:1)60oC , 20 h, 84%OHHNH63 OOEt2°CReview | Wiles WattsFuture Med. C hem. (20 09) 1(9)1606 future science groupwithin pharmaceuticals ever decreasing, con-tinuous-flow reactors are finding novel applica-tion in the reduction of trace metals within final products. Employing a series of model reactions, Pitts demonstrated the continuous removal of trace metals from reaction products through the incorporation of immobilized scavengers into simple polyethylene cartridges [83]. Using this approach, the authors identified QuadraPure-TU (13) to be the most effective scavenger for Pd (Suzuki reaction) and Cu (Sonogashira and Rosemund von-Braun reactions), QuadraPure-AMPA was used to sequester Fe from a Michael addition and Rh from a hydrogenation reaction (Figu Re 18). ta ble 3 also illustrates an example of Cu sequestration conducted while performing a continuous-flow click reaction [21].Continuous-ow distillationsIn addition to the techniques discussed thus far, distillation is a useful tool for the separation of reaction products from solvents, additives and byproducts based on volatility. Until recently, its use in conjunction with continuous-flow processes has been precluded by the need to prepare large quantities of material in order to use conventional apparatus.Recently, Jensen and coworkers disclosed details of miniaturized distillation equipment, realizing for the first time its potential as a separation technique within microreaction sys-tems [84] . By establishing a vapour–liquid equi-librium using segmented flow (N2), the authors demonstrated the separation of liquid mixtures based on differences in boiling point. Employing a silicon device with an ‘on-chip’ condenser and an integrated vapour–liquid membrane separator (Pall Zeflour membrane) [80], the authors investigated the separation of MeOH/toluene (50:50) and DCM/toluene (50 :50), at 70°C, whereby mole fractions of 0.22 ± 0.03 (liquid):0.79 ± 0.06 (vapor) and 0.16 ± 0.07 (liquid): 0.63 ± 0.05 (vapor) were obtained, respectively. In both cases, the data obtained were consistent with phase equilibrium pre-dictions, confirming the techniques’ potential for the separation of liquid mixtures generated under continuous flow. Precipitation of active ingredientsIn addition to providing methodology capable of affording rapid compound generation and inline purification, continuous-flow methodology can also assist with steps such as post-production processing of pharmaceuticals. With the solu-bility of active compounds having a significant effect on bioavailability, companies have sought solutions to the problem of compound insolu-bility in water, with one approach focusing on increasing solubility by reducing the compounds’ particle size. In 2007, Chen and coworkers demonstrated the ability to perform the controlled precipitation of a hydrophobic pharmaceutical, danazol (62) (FiguR e 19), within a microchannel reactor (Inlet channels = 300 µm [wide] × 300 µm [deep]; reaction channel = 600 µm [wide] × 300 µm [deep]) [85]. Whilst it may be thought that pre-cipitation within microchannels would be prob-lematic, the authors employ large reaction chan-nels relative to the particle size (0.15% of the channel depth) and, as such, observe no prob-lems with channel blocking or clogging. Based on the fact that the driving force for precipita-tion is the supersaturation of a solution induced by the mixing of a compound solution (S) with an antisolvent (AS), the authors investigated the effect of AS/S ratio on particle size and (62) solubility. To achieve this, the authors dissolved the active ingredient 62, in EtOH and employed various flow rates of deionized water (AS) to afford a range of AS/S ratios. Upon contacting Figure 19. Danazol (62) , the hydrophobic pharmaceutical precipitated within a microreactor.Figure 20. Steps used to generate (1R,2S,4S) - (7-oxa-bicyclo[2.2.1]hept-2-yl) - carbamic acid ethyl ester (63) under continuous flow.Aq.: Aqueous. NHBocOBnNClHHONHONMe2OHOOHOHO70 71OHNHBocOBnNClHHONHONMe2OHOOHOHO7071OHOHOAc76OHOH7372OHOO74ChirazymeL2–C2 75Continuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1607future science groupof the two liquid streams, the authors observed precipitation of 62 and collected the compound by filtering the effluent (0.45 µm). As table 12 illustrates, with decreasing S-flow rate, a marked effect on danazol particle size was observed: an average reduction from 55 µm (20 –120 µm) obtained in batch to 505 nm under continuous flow. Furthermore, reducing the reactor temperature from 30 to 4°C afforded a decrease in particle size to 364 nm and a dra-matic reduction in particle size distribution. In all cases, the morphology of the danazol remained the same; however, the regularity and particle size distribution was greatly improved under continuous flow. Comparison of the dissolution profiles between the nanoparticles and raw danazol illustrated 100% dissolution of the nanoparticles in less than 5 min, com-pared with 35% of the raw material 62 (com-plete dissolution = 40 min). Analysis of 62 by Fourier transform infared and x-ray diffraction confirmed that the physical characteristics of the compound were not affected by the anti-solvent-based nanoization process. The authors concluded that a liquid antisolvent precipitation technique within flow reactors is an easy method for controlling and reducing, the particle size of hydrophobic drug molecules, facilitating the dissolution of sparingly soluble compounds and increasing bioavailability.Examples of biologically active compounds prepared under continuous owHaving provided a brief insight into the many ways that continuous-flow chemistry can be used to assist the medicinal chemist on a daily basis, the next section describes a selection of pharmaceutically interesting or biologically active compounds that have been synthesized utilizing continuous-flow techniques. Pharmaceutical intermediatesBy conducting the three-step synthesis of (1R,2S,4S)-(7-oxa-bicyclo[2.2.1]hept-2-yl)-car-bamic acid ethyl ester (63), in a series of stainless steel microreactors, Bannwarth and coworkers were able to synthesise the pharmaceutical inter-mediate 63 in a safe and expedient manner [86] .As FiguRe 20 illustrates, the first step of the reac-tion involved the formation of a hydrazide (64), which was achieved by the treatment of ethyl ester (65) with hydrazine monohydrate 66 (1.1 equivalent) at 93°C. Employing a residence time of 106 min, the authors were able to obtain the intermediate 64 in 96% yield upon cooling to 0°C and filtration. In a second microreactor, main-tained at 2°C, the hydrazide 64 was acidified with 85% aqueous phosphoric acid (67) over a period of 2 min, prior to the addition of 10% aqueous sodium nitrite (68) to afford the highly reactive azide intermediate 69 (2 min) in 94% yield after extraction into DCE. In the final step, the Curtius rearrangement was performed in a 1:1 mixture of DCE/EtOH and the reactor maintained at 60°C for 20 h to afford the target compound 63 (84% yield) in an overall yield of 75%. Compared with batch, a remarkable rate acceleration was observed for the hydrazide 64 formation (1.8 vs 24 h) and the flow reactor provided a safe means of handling potentially unstable intermediates.Figure 21. Role of aminonaphthalene 70 as a building block in the synthesis of the prodrug 71.Figure 22. Biocatalytic synthesis of (E)-retinyl acetate 76. HO NH282NaHCO380NH278OCl OO679THF/H2ONH81OOO6MeONa83NH77OONHOH+HONH282NaHCO380NH278OClOO679THF/H2ONH81OOO6MeONa83NH77OONHOH+THF125oCAlMe384HOOMe87 O88BrOODMF75°C batch K2CO3OOMeOOO89OHNOOO85NH290HCO2H9190°C flow OHNOHOO86+ flow Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1608 future science groupIn a second example of industrial interest, Tietze and Liu reported the development of a continuous-flow process for the synthesis of an aminonaphthalene derivative (70) on a kilogram scale, for use as a starting material towards the preparation of novel anticancer agents, such as the duocarmycin prodrug 71 (Fig uRe 21) [87]. The authors were pleased to report that, in most cases, similar or better results were obtained when comparing the reaction steps performed under continuous flow and batch; the f low method ology did, however, pro-vide increased sa fety and reduced reaction times, with an empirical accelerating factor of F = 3–10 reported.In an example that highlights the potential associated with combining biocatalysis with continuous-flow reaction methodolog y orsat, Wirz and Bischof demonstrated the preparation of an intermediate in the synthesis of retinol (vita-min A) (72; FiguRe 22) [88]. To perform a reaction, the authors pumped a solution of alcohol 73, in vinyl acetate 74 (10 w/v) through a packed bed, containing the biocatalyst (Chirazyme L2-C2) (75) (5.0 g), collecting the target compound 76 in a throughput of 49 g day-1. To enable the strategy to be employed for the large-scale synthesis of this intermediate, the authors increased the size of their packed bed to accommodate 120 g of bio catalyst 75; affording 99% conversion to 76 (1.6 hg day-1). However, when conducting the enzyme-catalyzed acetylation over a long period of time, the authors noted a deterioration in the product quality, an observation that was attributed to the presence of impurities within the feedstock. To reduce this effect, the authors incorporated a precolumn into the system containing EDTA tetrasodium salt and, with this modification in place, the authors successfully operated the reactor for over 100 days before replacing the precolumn.  Continuous-ow synthesis of active ingredients Suberoylanilide hydroxamic acid synthesisAs part of a medicinal chemistry project, Martinelli and coworkers developed a flow proto-col for the rapid conversion of esters to hydroxamic acids (52 to 100% yield), with no carboxylic acid formation (as observed in batch) [89]. In an exten-sion to this, the authors investigated the synthesis of suberoylanilide hydroxamic acid (SAHA) 77, an early histone deacetylase inhibitor (FiguRe 23), in a PTFE reactor. As Fig uRe 23 illustrates, the first step of the reaction involved the alkylation of aniline (78) (1.0 M) with suberoyl chloride (79) (1.0 M), in the presence of NaHCO3 (80) (1.0 M), to afford ester 81. The crude ester 81 was subsequently treated with 50% aqueous hydroxylamine (82) and sodium methoxide (83) (0.5 M), at 90°C, to afford SAHA. The crude product 77 was then passed through a packed-bed reactor contain-ing a solid-supported quaternary amine scaven-ger (1.50 g) to trap any carboxylic acid formed as a byproduct. Upon collection, the solvent was Figure 23. Methodology employed for the continuous-flow synthesis of suberoylanilide hydroxamic acid.Figure 24. Combined approach for the synthesis of efaproxiral (86). Cl92NNNHOClClNCl92NNNHOClClNContinuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1609future science groupevaporated, water added and the pH adjusted to 7 with acetic acid, resulting in the precipi-tation of SAH A. Recrystallization of the pre-cipitate afforded SAHA in an 80% yield (over two steps).EfaproxiralCombining batch and flow regimes, Seeberger et al. demonstrated the trimethylaluminum 84 -medi-ated synthesis of amide 85 and its subsequent deprotection to afford efaproxiral (RSR13; 86), a pharmaceutical agent used for the enhancement of radiation therapy [90,91]. Due to the acknowl-edged problems associated with the manipulation of hetero geneous solutions within microfabricated reactors, the authors opted to combine flow and batch regimes. As depicted in FiguRe 24, phenol derivative 87 was alkylated using the tert-butyl ester of 2-bromo-2-methyl-propionic acid (88) to afford ester 89 in 75% yield. The second step of the reac-tion involved an aluminium-mediated amide bond formation, between the ester 89 and 3,5-dimeth-ylaniline (90), which was conducted under con-tinuous flow to afford amide 85 in quantitative selectivity towards the methyl ester (residence time = 2 min, 77% yield). In the final step, the tert-butyl ester-protecting group was hydrolyzed, using for-mic acid 91 (90°C), to afford efaproxiral (86; 89% yield) at a throughput of 24 mmol h-1.In a second example employing an alu-minium-mediated amide bond formation, the authors demonstrated the synthesis of rimonabant (SR141716; 92) (Fig uRe 25) , a CB-1 receptor blocker (anti-obesity drug). Performing the entire reaction sequence under continuous flow, the authors isolated the target compound in an overall yield of 49% (three steps).As we l l as those ex a mples disc u ssed herein, many ot her examples of pharm a-ceutically relevant compou nds have been synthesized utilizing continuous-flow tech-nology, includ i ng Ciprof loxa cin® [9 2 ,9 3] , Sildenafil® [94], natural products pristane [95] , (±) -oxomaritidine [96 ] and grossamide [97] , and even radiopharm aceuticals, suc h as 2-[18F]-fluoro-2-deoxy-2-d-glucose [59,6 0].Future perspectiveWhether it be lead compound identification, lead compound optimization or the upscaling of a synthetic route in preparation for produc-tion, it ca n be seen from the exa mples dis-cussed herein that MRT and continuous-flow chemistry are valuable tools for the medicinal chemist, affording time and cost savings while reducing the hazards associated with conven-tional batch methodology. With an increasing number of commercially available reaction plat-forms becoming available, it is only a matter of time before a step change is made within industr y, enabling both researchers and pro-duction chemists to use the same equipment. However, for this approach to be rea lized, further research is required into the develop-ment and commercial availability of contin-uous-flow reactors with throughputs suitable for use within a production environment. It is also imperative that the technolog y becomes competitively priced, enabling implementa-tion of MRT into process environments for the same cost per litre-1 as a new batch process. Furthermore, advances in the area of online ana lysis are required, as a means of reducing the analytical bottleneck that currently exists with the use of these information-rich systems at an R D level and to enable process control from a production perspective. Figure 25. Rimonabant (SR141716) (92) synthesized under continuous flow.Executive summary Microreaction technology is an emerging technique that has been applied to many forms of synthetic chemistry over the past decade.  Reactors have been fabricated from a range of substrates, including polymers, metals, ceramics and glass. Through the selection of reaction routes and products that are of interest to the medicinal chemist the advantages associated with continuous-flow techniques for drug discovery are presented. Examples include systems used to identify hits, optimise processes and attain production volumes all within the confines of continuous-flow reactors. As such it can be seen that continuous-flow chemistry can be used in both early stage research and development and production providing a tool that can span all stages necessary for identifying, evaluating and producing a pharmaceutical agent. Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1610 future science groupBibliography1 Wess G, Urmann M, Sickenberger B. Medicina l chemistry: challenges a nd opportunities. Angew. Chem. Int. Ed . 40(18), 3341–3350 (20 01). 2 Persidis A. Combinatoria l chemistry. Nat. Biotechnol. 15, 391–392 (1997). 3 Weigl BH, Bardell RL , Cabrera CR. Lab-on-a-chip for drug development. Adv. Drug Deliver y Rev. 55, 349–377 (2003). 4 Glasnov TN, Kappe CO. Microwave-assisted synthesis u nder continuous-flow conditions. Macromol. Rapid Commun. 28, 395– 410 (20 07). 5 Mason BP, Price KE , Steinbacher JL, Bogdan A R, McQu ade DT. Greener approaches to organ ic synthesis using microreac tor technology. Chem. Rev. 107( 6), 2300–2318 (2007). 6 W iles C, Watts P. Continuous flow reactors, a tool for the modern s ynthetic chemist. Eur. J. Org. Che m. 10, 1655–1671 (2008). 7 Guthrie D. Follow the pharma flow. Spec. Chem. Mag. 29, 28–29 (2009). 8 E hrfeld W, Hessel V, Lowe H. Micro Reactors New Technolog y for Modern Chemistry. W ile y VCH, Germa ny (2000). 9 W iles C, Watts P, Haswell SJ, Pombo-Vi lla r E. The aldol reaction of silyl enol ethers wit hin a micro reactor. Lab Chip 1, 100–101 (2001). 10 Wiles C, Watts P, Haswell SJ, Pombo-Vi lla r E. 1,4-addition of enolates to a,b-unsaturated ketones wit hin a micro reactor. Lab Chip 2, 62– 64 (2002). 11 Lowe H, Hessel V, Lob P, Hubbard S. Addition of secondary amines to a,b-unsaturated carbonyl compound s and nitriles by using microstr uctures reactors. Org. Proc. Res. Dev. 10, 1144–1152 (20 06). 12 Wiles C, Watts P, Haswell SJ, Pombo-Vi lla r E. The regioselective prepa ration of 1,3-d iketones in a microreac tor. Chem. Commun. 43(16), 1034–1035 (2002). 13 Wiles C, Watts P, Haswell SJ, Pombo-Villar E. Stereoselective alkylation of a n evans auxiliary derivative w ithin a pressure-driven micro reactor. Lab Chip 4, 171–173 (2004). 14 Wile s C, Watts P, Haswell SJ, Pombo-Villar E. A n investigation into the use of silica-supported bases w ithin EOF-based flow reactors. Tetrahedron 60, 8421–8427 (200 4). 15 Wile s C, Watts P, Haswell SJ. Acid-catalysed synthesis and deprotection of dimethyl acetals in a miniaturised electroosmotic flow reactor. Tetrahedron 61, 5209–5217 (2005).16 El Kadib A, Chimenton R, Sachse A, Fajula F, Gala rneau A, Coq B. Functiona lised inorganic monolit hic microreactors for high productivity in fine chemic al c ata lytic synthesis. Angew. Chem. Int. Ed. 48, 4969– 4972 (2009). 17 Acke DRJ, Stevens C V. Study of the Baylis-Hi llman react ion in a microreac tor environment : first continuous production of Baylis-Hi llman adducts. Org. Proc. Res. Dev. 10, 417–422 (2006). 18 Odedra A, Seeberger PH. 5- (py rrolidin-2-yl)tetrazole-catalysed aldol and mannich reactions : acceleration and lower catalyst loading in a continuous flow reactor. Angew. Chem . Int. Ed. 48(15), 2699–2702 (2009). 19 Hornung CH, Mackley MR, Baxendale IR, Ley SV. A microcapillary flow disk reactor for organic synthesis. Org. Proc. Res. Dev. 11, 399– 405 (2007). 20 Shore G, Orga n MG. Diels-Alder cycloadditions by microwave-assisted, continuous flow organic synt hesis (MACOS): the role of metal films in the flow tube. Chem. Commun. DOI: 10.1002/chin.200825029 838–840 (2008). 21 Smith CD, Baxendale IR, Lanners S, Hayward JJ, Smith SC , Le y SV. [3+2] cycloaddition of acet ylenes with azides to give 1,4-disubst ituted 1,2,3-triazoles in a modular flow reactor. Org. Biomol. Chem. 5, 1559–1561 (20 07). 22 Baxendale IR, Ley SV, Mansfield AC, Smith CD. Multistep synthesis using modula r flow reactors: Bestma nn-Ohira rea gent for the formation of alk ynes and triazoles. Angew. Chem . Int. Ed. 48, 4017–4021 (2009). 23 Singh BK, K aval N, Tomar S, Van der Eycken E, Parmar VS. Transition metal-cata lysed carbon-carbon bond formation Suzuki, Heck and Sonogashira reactions using microwave and microtechnology. Org. Proc. Res. Dev. 12, 468– 474 (2008). 24 Comer E, Organ MG. A microreactor for microwave-assisted capillar y (continuous flow) organic synthesis (M ACOS). J. Am. Chem. Soc. 127(22), 8160–8167 (2005). 25 Mennecke K, Soledenko W, Kirschning A. Carbon-carbon cross-coupling reaction under continuous flow conditions using poly(vinylpyridine) doped with palladium. Synthesis 10, 1589–1599 (2008). 26 Zhang X, Stefanick S, Villani FJ. Application of microreactor technology in process development. Org. Proc. Res. Dev. 8, 455– 460 (2004).27 Lu S, Watts P, Chin FT, Hong J et al. Syntheses of 11C and 18F-labeled carboxylic esters within a hydrodynamically-driven micro-reactor. Lab Chip 6, 523–525 (2004).28 Hooper J, Watts P. Expedient synthesis of deuterium-labelled a mides within micro reactors. J. Labelled Compd. Radiopharm. 50, 189–196 (2007). 29 Wiles C , Watts P. An integrated microreactor for the multi-component synthesis of a-aminonitriles. Org. Proc. Res. Dev. 12, 1001–1006 (2008).30 Wiles C , Watts P. Evaluation of the heterogeneously catalysed strecker reaction conducted under continuous flow. Eur. J. Org. Chem. 33, 5597–5613 (2008).31 Acke DRJ, Stevens CV. A HCN-based reaction under microreactor conditions: industrially feasible and continuous synthesis of 3,4-diamino-1H-isochromen-1-ones. Green Chem. 9, 386 –390 (2007). 32 Watts P, Wiles C, Haswell SJ, Pombo-Villar E. Solution phase synthesis of b-peptides using micro reactors. Tetrahedron 58, 5427 (20 02). 33 Flogel O, Codee JDC, Seebach D, Seeberger PH. Microreactor system for the synthesis of b-peptides. Angew. Chem. Int. Ed. 45, 7000–7003 (2006). 34 Huang Y, Castrataro P, Lee CC, Qua ke SR. Solvent resistant microfluidic DNA synthesizer. Lab Chip 7, 24–26 (2007). 35 Schwalbe T, Autze V, Hohmann M, Stirner W. Novel innovation systems for a cellular approach to continuous process chemistry from discovery to market. Org. Proc. Res. Dev. 8, 440– 454 (2004).36 Garcia-Egido E, Wong SYF, Warrington BH. A Hantzsch synthesis of 2-aminothiazoles Financial competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options , expert t estimony, grants or patents received or pending, or royalties.No writing assistance was utilized in the production of this manuscript. Continuous-ow organic synthesis: a tool for the modern medicinal chemist | Reviewwww.fut ure- scie nce.com 1611future science groupperformed in a heated microreactor system. Lab Chip 2, 31–33 (2002).37 Singh BK, Stevens CV, Acke DRJ, Parmar VS, Van der Eycken EV. Copper-mediated N- and O-ar ylations with arylboronic acids in a continuous flow microreactor: a new avenue for efficient scalability. Tetrahedron Lett. 50, 15–18 (2009). 38 Chan DMT, Monaco KL, Wang RP, Winters MP. New N- and O-arylations with phenylboronic acids and cupric acetate. Tetrahedron Lett. 39, 2933–2936 (1998). 39 Evans DA, Katz JL , West TR. Synthesis of diar yl ethers through the copper-promoted arylation of phenols with arylboronic acids. An expedient synthesis of thyroxine. Tetrahedron Lett. 39, 2937–2940 (1998). 40 Lam PYS, Clark CG, Saubern S et al. New aryl/heteroaryl CN bond cross-coupling reactions via arylboronic acid/cupric acetate arylation. Tetrahedron Lett. 39, 2941–2944 (1998). 41 Kopach ME, Murray MM, Braden TM, Kobierski ME, Williams OL. Improved synthesis of 1-(azidomethyl)-3,5-bis-(trifluoromethyl)benzene: development of batch and microflow azide processes. Org. Proc. Res. Dev. 13, 152–160 (2009).42 Baumann M, Baxendale IR, Ley SV, Nikbin N, Smith CD. Azide monoliths as convenient flow reactors for efficient curtius rearrangement reactions. Org. Biomol. Chem. 6, 1587–1593 (20 08). 43 Wiles C , Watts P, Haswell SJ. An efficient continuous flow technique for the chemoselective synthesis of thioacetals. Tetrahedron Lett. 48, 7362–7365 (2007). 44 Wiles C , Hammond MJ, Watts P. The development and evaluation of a continuous flow process for the lipase-mediated oxidation of alkenes. Beilstein J. Org. Chem. 5, 27 (2009). 45 Van der Linden JJM, Hilberink PW, Kronenburg CMP, Kemperman GJ. Investigation of the Moffatt-Swern oxidation in a continuous flow microreactor system. Org. Proc. Res. Dev. 12, 911–920 (2008). 46 Yamada YM, Torii K, Uozumi Y. Oxidative cyclization of alkenols with oxone using miniflow reactor. Beilstein J. Org. Chem. 5(18) (2009). 47 McPake CB, Murray CB, Sandford G. Epoxidation of alkenes using HOF:MeCN by a continuous flow process. Tetrahedron Lett. 50, 1674–1676 (2009). 48 Ducry L , Roberge DM. Dibal-H reduction of methyl butyrate into butyra ldehyde using microreactors. Org. Proc. Res. Dev. 12, 163–167 (2008). 49 Jones RV, Godorhazy L, Varga N, Szalay D, Urge L, Darvas F. Continuous flow high pressure hydrogenation reactor for optimization and high-throughput synthesis. J. Comb. Chem. 8, 110–116 (2006). 50 Chambers RD, Sandford G, Trmcic J, Okazoe T. Elemental fluorine. Part 21. Direct fluorination of benzaldehyde derivatives. Org. Proc. Res. Dev. 12, 339–344 (2008). 51 Chambers RD, Fox MA, Sandford G, Trmcic J, Goeta A. Elemental fluorine: Part 20. Direction fluorination of deactivated aromatic systems using microreactor techniques. J. Fluorine Chem. 128, 29–33 (2007). 52 Miyake N, Kita zume T. Microreactors for the synthesis of fluorinated materials. J. Fluorine Chem. 122, 243 –246 (2003). 53 Gustafsson T, Gilmour R, Seeberger PH. fluorination reactions in microreactors. Chem. Commun. 26, 3022–3024 (2008). 54 Suga S, Okajima M, Fujiwara K, Yoshida J. Electrochemical combinatorial organic syntheses using microflow systems. QSAR 24(6), 728–741 (2005). 55 Nagaki A, Takabayashi N, Tomida Y, Yoshida J. Synthesis of unsymmetrically substituted biaryls via sequential lithiation of dibromobiaryls using integrated microflow systems. Beilstein J. Org. Chem. 5, 16 (2009). 56 Fukuyama T, Hino Y, Kamata N, Ryu I. Quick excecution of [2+2] type photochemical cycloaddition reaction by continuous flow system using a glass-made microreactor. Chem. Lett. 33, 1430 (2004). 57 Sugimoto A, Sumino Y, Ta kagi M, Fukuyama T, Ryu I. The Barton reaction using a microreactor and black light. Continuous-flow synthesis of a key steroid intermediate for an endothelin receptor antagonist. Tetrahedron Lett. 47, 6197–6200 (2006).58 Csajagi C, Szatzker G, Toke RR, Urge L, Darvas F, Poppe L. Enantiomer selective acylation of racemic alcohols by lipases in continuous-flow bioreactors. Tetrahedron Asymm. 19, 237–246 (2008). 59 Miyazaki M, Maeda H. Microchannel enzyme reactors and their applications for processing. Trends Biotechnol. 24, 463–470 (2006). 60 Liow E, O’Brien A, Luthra S, Brady F, Steel C. Preliminary studies of conducting high level production radiosyntheses using microfluidic devices. J. Label. Comp. Radiopharm. 48, S28–S28 (2005). 61 Elizarov AM. Microreactors for radiopharmaceutical synthesis. Lab Chip 9, 1326–1333 (2009). 62 Ramstrom O, Lehn JM. Drug discovery by dynamic combinatorial libraries. Nat. Rev. Drug Discov. 1, 26–36 (2002). 63 Wang Y, Lin W Y, Liu K et al. An integrated microflu idic device for large-scale in situ click chemistry screening. Lab Chip DOI: 10.1039/b907430a (2009) (Epub ahead of print).64 Tron GC, Pirali T, Billington R A, Canonico PL, Sorba G, Genazzani AA. Click chemist ry reactions in medicinal chemistr y: applications of the 1,3-dipolar c ycloaddition between azides and alky nes. Med. Res. Rev. 28(2), 278 –308 (2009). 65 Wang J, Sui G, Mocharla VP et al. Integrated microflu idics for parallel screening of an in situ click chemistry library. Angew. Chem. Int. Ed . 45, 5276–5281 (2006). 66 Schwa lbe T, Kurawe A, Sommer J. Application report on operat ing cellular process chemistry pla nts in fine chemical and contract manufacturing industries. Chem. Eng. Technol. 28(4), 408–419 (2005). 67 Lavric ED, Woehl P. Advanced-flow™ gla ss reactors for seamless scale-up. Chimica Og gi 27(3), 45–48 (2009). 68 Anderson NG. Practical use of continuou s processing in developing and scaling-up laboratory processes. Org. Proc. Res. Dev. 5, 613–621 (2001). 69 Yoshida J. Flash Chemistry : Fast Organic Synthesis in Microsystems. Wiley, London, UK (2008).70 Thayer AM. Handle with care. Chem. Eng. News 87(11), 17–19 (2009). 71 Braune S, Pochlauer P, Reintjens R et al. Selective nitration in a microreactor for pharmaceutical produc tion under cGMP conditions. Chimica Oggi 26, 5, 1– 4 (2008). 72 Cheva lier B, Lavric ED, Cerato-Noyerie C, Horn CR, Woehl P. Microreactors for industrial multi-phase applic ations. Chimica Oggi 26(2), 38– 42 (2008). 73 Aota A, Mawatari K, Kitamori T. Parallel multi-phase m icroflows: fundamental physics, stabilization methods and applications. Lab Chip DOI: 10.1039/b904430m (2009) (Epub ahead of print). 74 Tokeshi M, Minagawa T, Kitamori T. Integration of a microextraction system on a glass chip: ion-pair solvent extraction of Fe(II) with 4,7-diphenyl-1,10-phenanthrolinedisulfonic acid and tri-N-octylmet hylammoniu m chloride. Anal. Chem. 73, 5551–5556 (2001). 75 Kikutani Y, Mawatari K, Hiba ra A, Kita mori T. Circulation microcha nnel for liquid–liquid microextraction. Microchim Acta 164, 241–247 (2009). 76 Tokeshi M, Minagawa T, Uchiyama K, Hibara A et al. Continuous-flow chemica l processing on a micro-chip by combining Review | Wiles WattsFuture Med. C hem. (20 09) 1(9)1612 future science groupmicrounit operations and a mu ltiphase flow network. Anal. Chem. 74, 1565–1571 (2002). 77 Tetala KKR, Swarts JW, Chen B, Janssen AEM, van Beek TA. A three-phase microfluidic chip for rapid sample clean-up of alkaloids from plant extracts. Lab Chip 9, 2085–2092 (2009). 78 Maruyama T, Matsushita H, Uchida J, Kubota F, Kamiya N, Goto M. Liquid membrane operations in a microfluidic device for selective separation of metal ions. Anal. Chem. 76, 4495–4500 (2004). 79 Honda T, Miyazaki M, Nakamura H, Maeda H. Facile preparation of an enzyme-immobilised microreactor using a cross-linking enzyme membrane on a microchannel surface. Adv. Synth. Catal. 348, 2163–2171 (2006). 80 Honda T, Miyazaki M, Yamaguchi Y, Nakamura H, Maeda H. Integrated microreaction system for optical resolution of racemic amino acids. Lab Chip 7, 366–372 (2007). 81 Kralj JG, Sahoo HR, Jensen KF. Integrated continuous microfluidic liquid-liquid extraction. Lab Chip 7, 256–263 (2007). 82 Sahoo HR, Kralj JG, Jensen KF. Multi-step continuous flow microchemica l synthesis involving multiple reactions and separations. Angew. Chem. Int. Ed. 46, 5704–5708 (2007). 83 Hinchcliffe A, Hughes C, Pears DA, Pitts MR. QuadraPure cartridges for removal of trace metal from reaction mixtures in flow. Org. Proc. Res. Dev. 11, 477–481 (2007). 84 Hartman RL, Sa hoo HR, Yen BC, Jensen KF. Distillation in microchemical systems using capillary forces and segmented flow. Lab Chip 9, 1843–1849 (2009). 85 Zhao H, Wang JX, Wang QA, Chen JF, Yun J. Controlled liquid antisolvent precipitation of hydrophobic pharmaceutical nanoparticles in a microchannel reactor. Ind . Eng. Chem. Res. 46, 8229–8235 (2007). 86 Rumi L, Pfleger C , Spurr P, Kilinkhammer U, Bannwarth W. Adaptation of an exothermic and acylazide-involving synthesis sequence to microreactor technolog y. Org. Proc. Res. Dev. 13, 747–750 (2009). 87 Tietze LF, Liu D. Continuous-flow microreactor multi-step synthesis of an aminonaphthalene derivative as starting material for the preparation of novel antica ncer agents. Arkivoc, vii 193–210 (2008). 88 Orsat B, Wirz B, Bischof S. A continuous lipase-cata lysed acylation process for the large-scale production of vitamin A precursors. Chimia 53, 579–584 (1999). 89 Riva E, Gagliardi S, Mazzoni C, Passerella D et al. Efficient continuous flow synthesis of hydroxamic acid prepa ration. J. Org. Chem. 9, 3540–3543 (2009). 90 Gustafsson T, Ponten F, Seeberger PH. Trimethylaluminium mediated amide bond formation in a continuous flow microreactor as a key to the synthesis of rimonabant and efaproxiral. Chem. Commun. 24, 1100–1102 (2008). 91 Gustafsson T, Gilmor R, Seeberger PH. Fluorination in microreactors. Chem. Commun. 21, 3022–3024 (2008). 92 Schwalbe T, Autze V, Wille G. Chemical synthesis in microreactors. Chimia 56, 636– 646 (2002). 93 CPC. Cellular Process Chemistry Systems GmbH, EP 0116024-A2 (2001). 94 Panke G, Schwalbe T, Stirner W, Taghavi-Moghadam S, Wille G. A practical approach of continuous processing to high energetic nitration reactions in microreactors. Synthesis 18, 2827–2830 (2003). 95 Tanaka K, Motomatsu S, Koyama K, Tanaka S, Fukase K. Large-scale synthesis of immunoactivating natural product, pristane, by continuous microfluidic dehydration as the key step. Org. Lett. 9, 299–302 (2007). 96 Baxendale IR, Deeley J, Griffiths-Jones CM, Ley SV, Saaby S, Tranmer GK. A flow process for the multi-step synthesis of the alkaloid natural product oxomaritidine: a new paradigm for molecular a ssembly. Chem . Commun. 18, 2566–2568 (2006). 97 Baxendale IR, Griffit hs-Jones CM, Ley SV, Tranmer GK. Preparation of the neolignan natural product grossamide by a continuous flow process. Synlett 3, 427–430 (2006).