The Development and Application of Two-Chamber Reactors and

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article pubs.acs.org/accounts

The Development and Application of Two-Chamber Reactors and Carbon Monoxide Precursors for Safe Carbonylation Reactions Stig D. Friis,*,† Anders T. Lindhardt,*,‡ and Troels Skrydstrup*,† †

Carbon Dioxide Activation Center (CADIAC), Interdisciplinary Nanoscience Center (iNANO), and Department of Chemistry, Aarhus University, Gustav Wieds Vej 14, 8000 Aarhus C, Denmark ‡ Carbon Dioxide Activation Center (CADIAC), Interdisciplinary Nanoscience Center (iNANO), and Department of Engineering, Aarhus University, Finlandsgade 22, 8200 Aarhus N, Denmark CONSPECTUS: Low molecular weight gases (e.g., carbon monoxide, hydrogen, and ethylene) represent vital building blocks for the construction of a wide array of organic molecules. Whereas experimental organic chemists routinely handle solid and liquid reagents, the same is not the case for gaseous reagents. Synthetic transformations employing such reagents are commonly conducted under pressure in autoclaves or under atmospheric pressure with a balloon setup, which necessitates either specialized equipment or potentially hazardous and nonrecommended installations. Other safety concerns associated with gaseous reagents may include their toxicity and flammability and, with certain gases, their inability to be detected by human senses. Despite these significant drawbacks, industrial processes apply gaseous building blocks regularly due to their low cost and ready availability but nevertheless under a strictly controlled manner. Carbon monoxide (CO) fits with all the parameters for being a gas of immense industrial importance but with severe handling restrictions due to its inherent toxicity and flammability. In academia, as well as research and development laboratories, CO is often avoided because of these safety issues, which is a limitation for the development of new carbonylation reactions. With our desire to address the handling of CO in a laboratory setting, we designed and developed a two-chamber reactor (COware) for the controlled delivery and utilization of stoichiometric amounts of CO for Pd-catalyzed carbonylation reactions. In addition to COware, two stable and solid CO-releasing molecules (COgen and SilaCOgen) were developed, both of which release CO upon activation by either Pd catalysis or fluoride addition, respectively. The unique combination of COware with either COgen or SilaCOgen provides a simple reactor setup enabling synthetic chemists to easily perform safe carbonylation chemistry without the need for directly handling the gaseous reagent. With this technology, an array of low-pressure carbonylations were developed applying only near stoichiometric amounts of carbon monoxide. Importantly, carbon isotope variants of the CO precursors, such as 13COgen, Sila13COgen, or even 14COgen, provide a simple means for performing isotope-labeling syntheses. Finally, the COware applicability has been extended to reactions with other gases, such as hydrogen, CO2, and ethylene including their deuterium and 13C-isotopically labeled versions where relevant. The COware system has been repeatedly demonstrated to be a valuable reactor for carrying out a wide number of transition metal-catalyzed transformations, and we believe this technology will have a significant place in many organic research laboratories.

■

INTRODUCTION

laboratory setting necessitates CO detectors as a safety measure for the operator.1 To circumvent the issue of handling gaseous carbon monoxide, a variety of solid or liquid CO-releasing molecules have been developed and employed for a number of carbonylation reactions, for example, formates, aldehydes, and metal carbonyls to mention a few.2,3 Unfortunately, such compounds and the byproducts formed by CO release are not necessarily innocent, unreactive spectator compounds, which narrows the window of applicable carbonylation protocols or the functional group tolerance.4 Furthermore, in situ decarbonylation protocols lead to byproducts from the CO surrogate in

Among the gases used in organic synthesis, carbon monoxide is considered a key reagent along with other important gases such as hydrogen and carbon dioxide. The attractive feature of installing carbonyl groups into organic products while forming two new bonds renders this class of transformations highly valuable. However, even if carbonylation reactions exhibit high atom efficiency compared with other means for carbonyl group synthesis, the handling concerns associated with the high toxicity of carbon monoxide gas have restricted the academic exploitation of this reaction class. What is more, CO cannot be detected by any of the human senses, making accidental exposure to lethal concentrations a serious threat when working with larger amounts of the gas. Pressurized cylinders are the primary source for CO, and hence their use in a research © 2016 American Chemical Society

Received: October 19, 2015 Published: March 21, 2016 594

DOI: 10.1021/acs.accounts.5b00471 Acc. Chem. Res. 2016, 49, 594−605

Article

Accounts of Chemical Research

reactions discussed in this Account were generally performed at pressures not exceeding 5 bar. Two different types of compounds were developed as CO precursors, both releasing CO only upon activation. First, 9methylfluorene-9-carbonyl chloride (COgen), a bench stable and crystalline acid chloride, releases CO when subjected to a tri-tert-butylphosphine ligated Pd catalyst and an amine base in a number of aprotic solvents. Upon heating, the CO release is nearly quantitative as demonstrated by the isolation of amide 2 in 96% yield when a limiting amount of COgen in COware was applied (Scheme 1). Alternatively, methyldiphenylsilacarboxylic acid (SilaCOgen), also a highly crystalline solid, will effectively release CO when treated with a fluoride source in a variety of solvents. Depending on the reactions studied, one advantage of SilaCOgen, is its ability to effectively and rapidly release CO even at temperatures below 40 °C, while COgen is more applicable where controlled CO release is required.5,6 Both reagents have demonstrated their worth in numerous carbonylative transformations as will be highlighted in the following sections.

the reaction mixture with the desired product, complicating reaction workup and product purification.

■

DESIGN OF COware AND CO-RELEASING MOLECULES In 2010, we initiated a research program in Pd-catalyzed carbonylation chemistry. Like researchers before us, we were hesitant to employ gaseous carbon monoxide in combination with autoclaves for pressurized reactors or balloons due to the safety concerns discussed above. Nevertheless, the thought of adding large excesses of a carbon monoxide surrogate to the reaction mixture for in situ CO release was neither appealing because such additives would likely complicate the development of novel carbonylative transformations. Therefore, we constructed a setup whereby the direct handling of CO could be avoided. This would require first a stable, preferentially solid CO precursor from which release of CO can be controlled. Second, only the CO generated from the CO release should come in contact with the carbonylation reaction. The second part of this plan was easily executed by the design of a twochamber reactor: one chamber for the CO release and the secondary chamber for the CO-consuming reaction. The spatial separation of the two reactor chambers allows CO to diffuse across via the headspace (Figure 1). Although, COware is

■

ISOTOPIC LABELING The preparation of isotopically labeled compounds with either stabile or radioactive isotopes is central to modern drug discovery, to metabolic studies, as internal standards, and for other uses. Each synthetic step applied to reach the target structure after incorporation of the isotope is in general a costly manipulation. This is especially true within the field of 14Cisotope labeling, whereby the isotope is radioactive, requiring specially trained personal and specialized laboratory facilities to effect the synthesis. Because both COgen and SilaCOgen are synthesized from carbon dioxide, facile preparation of their 13Cand 14C-isotopically labeled counterparts is possible using stoichiometric amounts of the isotopically labeled CO2 (Scheme 1). Finally, because many Pd-catalyzed reactions are characterized by a high functional group tolerance, incorporation of isotopically labeled CO can be achieved with advanced intermediates or even in the final step of the synthesis, making carbonylation chemistry suitable for isotopic labeling. With 13COgen and Sila13COgen at hand, their applicability to the preparation of a range of biologically interesting compounds has been demonstrated. Thus, with COware, the desired carbonylative transformations are easily optimized using the unlabeled precursors, COgen or SilaCOgen. Changing the

Figure 1. COware two-chamber system.

constructed of pyrex glass, stress-testing revealed these reactors to endure pressures up to 15 bar without failure, although all

Scheme 1. Synthesis and Decarbonylation of COgen and SilaCOgen

595


Article


this transformation.13 During the development of both COgen and SilaCOgen, a simple aminocarbonylation was used to quantify the amount of generated CO (Scheme 1). Applying a limiting amount of COgen or 13COgen for the generation of unlabeled or labeled CO, we prepared the pharmacologically active compounds 6−9 and [13C]-6−[13C]-9 in comparable and good yields from their corresponding iodides (Scheme 3),

CO precursor for its 13C-labeled version under otherwise identical conditions provides the 13C-labeled target. In this way, 13 C2-labeled FSB, a selective β-amyloid fibril binder, could be prepared via a double alkoxycarbonylation (Scheme 2).7 Additional examples are presented in the following sections. Scheme 2. Carbonylation in COware Using 13COgen To Realize Carbon-13 Labeled FSB

Scheme 3. Aminocarbonylation in COware Using a Limiting Amount of COgen or 13COgen

In a similar manner, 14COgen was prepared from 14CO2 (Scheme 1) and then employed as the limiting reagent (0.1 mmol scale) for carbonylation reactions in COware affording olaparib (anticancer agent), thalidomide (immunomodulator), and fenofibrate (blood cholesterol lowering drug) in 37%, 70%, and 72% radiochemical yields, respectively, applying conditions developed using unlabeled COgen (Figure 2).8

while aryl bromides were also competent substrates.6 Similarly, applying Sila13COgen, [13C]-piclamilast (10) was furnished in a 74% isolated yield with a limiting amount of the 13CO precursor (Scheme 4).5 Scheme 4. Aminocarbonylation in COware Applying a Limiting Amount of Sila13COgen

Figure 2. Carbonyl radiolabeling of three approved drug molecules, using a limiting amount of 14COgen.

In a study by Larhed and co-workers, COware was combined with the cheap and readily available carbon monoxide precursor, Mo(CO)6, furnishing a range of amides from their corresponding aryl iodides or bromides and amine counterparts.14 Thus, with the two-chamber system setup, complications associated with the catalytic activity of molybdenum carbonyl complexes and their removal after reaction completion were avoided, thereby facilitating access to amides 11a−11f and others (Scheme 5). As demonstrated by the work of Larhed, aryl bromides generally require higher temperatures to be viable substrates in the aminocarbonylation. Nonetheless, these remain more appealing substrates due to their lower cost, wider commercial availability, and higher stability. To solve this problem, Buchwald et al. proposed the use of a more active Pd catalyst, generated from a palladacycle precatalyst, which rapidly and cleanly generates a ligated palladium(0) catalyst. Employing this strategy, the reaction temperature required to efficiently transform both electron rich and electron poor aryl bromides could be lowered to 45 °C from the usual temperatures of 80 °C and higher.15 At this temperature, SilaCOgen still provides near quantitative decarbonylation also allowing for product

■

CARBONYLATIONS USING COware Although Pd-catalyzed carbonylation reactions have been known for over four decades, these carbonyl-installing reactions are still highly popular because CO is an inexpensive and versatile C1-building block.9−12 By eliminating the handling of gaseous CO, the two-chamber reactor encourages the safe development of new carbonylations, as well as the improvement of existing protocols. Because the carbonylation reaction in the second chamber of COware is only in contact with CO in the headspace, in principle, transition to scale-up synthesis applying carbon monoxide from pressurized cylinders does not require substantial alterations of the reaction conditions. Aminocarbonylation

Amides represent a key functional group in both medicinal chemistry and material science, as well as in naturally occurring compounds. Whereas aminocarbonylations are well documented in the literature, challenges are still associated with 596


Article

Accounts of Chemical Research Scheme 5. Aminocarbonylation in COware Using Mo(CO)6

Scheme 7. Two-Step Synthesis of Clenbuterol, with a Double Carbonylation as the Key Step

selectivity between 12j and 12k by adjustment of the SilaCOgen amounts (Scheme 6). By modification of the aminocarbonylation conditions, formation of α-ketoamides via double carbonylation can be achieved. In 2006, Kondo and co-workers published a high yielding, low temperature, and low pressure version of the double carbonylation forming α-ketoamides.16 By application of these conditions in combination with COware and COgen, αketoamides and their 13C2-isotopic variants can be furnished under mild and safe conditions without the use of gaseous CO or 13CO.17 As exemplified in Scheme 7, subsequent and selective reduction of the double 13C-labeled α-ketoamide 13 provided [13C2,D3]-clenbutarol. Switching the carbon monoxide source to SilaCOgen also provides good yields of the double carbonylated products and sometimes in better yields as with electron poor aryl iodides, which could be explained by a more rapid CO release with SilaCOgen compared with COgen (Scheme 8).5 The selective formation of primary amides represents a particularly challenging aminocarbonylation, because of the handling of two toxic gases, namely, CO and ammonia. Ammonium carbamate is an attractive ammonia surrogate, which upon heating releases two equivalents of NH3 along with an equivalent of CO2. Thus, with this carbamate and CO from COgen, a selection of primary amides were prepared from various aryl bromides in the presence of a JosiPhos-type ligated palladium catalyst, thereby eliminating the direct handling of two gaseous reagents (Scheme 9).18 Amides not only are attractive products in their own right but with appropriately designed nucleophiles also represent viable substrates for heterocycle synthesis. In a carbonylative approach to oxadiazoles with aryl bromides, acylhydrazine and

Scheme 8. Comparison of COgen and SilaCOgen in a Double Carbonylation

Scheme 9. Carbonylative Synthesis of Primary Amides

amidoxime were found to be excellent nucleophiles generating 1,3,4-oxadiazoles and 1,2,4-oxadiazoles, respectively.19 As illustrated in Scheme 10, starting from Boc-hydrazine, a four-

Scheme 6. Low Temperature Aminocarbonylation Applying a Buchwald-type Palladacycle Precatalyst and SilaCOgen

597


Article

Accounts of Chemical Research Scheme 10. Two Sequential Aminocarbonylations Employing 13COgen To Afford a Double Carbon-13 Labeled Nonsymmetrical 1,3,4-Oxadiazole

Scheme 12. Alkoxycarbonylation with Sterically Hindered Tertiary Sodium Alkoxides

Scheme 11. Alkoxycarbonylation Using a near Stoichiometric Amount of SilaCOgen

Scheme 13. Thiocarbonylation versus Thiolation of Aryl Iodides Is Highly Dependent on Ligand and Base

step sequence involving N-carbonylation, deprotection, N′carbonylation, and dehydrative cyclization provided the 13C2labeled nonsymmetrical 1,3,4-oxadiazole 20 in a 64% overall yield with 13CO from 13COgen. The employment of poorer nucleophiles also produces viable aminocarbonylations. To this end, both N-substituted cyanamides, ureas, and isocyanates have proven to be competent reaction partners in the aminocarbonylation of aryl iodides and bromides, when using COware in combination with either COgen or SilaCOgen.20,21

Scheme 14. Thiocarbonylation of Organic Bromides

Alkoxy- and Thiocarbonylation

While similar to the aminocarbonylation, the use of alcohol nucleophiles does present unique challenges due to their lower reactivity under carbonylative conditions. In the COware twochamber system, heteroaryl tosylate 21 was converted to ester 22 in a 75% isolated yield under modified literature conditions using only mild base (Scheme 11).5 On the other hand, when more sterically hindered alcohols were applied in the preparation of esters from aryl bromides, use of the corresponding sodium salts was required. In combination with a palladium catalyst ligated by an electron rich ferrocene type ligand (DiPrPF), tertiary esters could be generated in good yields, even allowing for the formation of adamantyl and trityl esters (Scheme 12).22 With only 1.3 equiv of CO-loading required, this methodology provides convenient access to O-protected carboxylic acids from aryl and heteroaryl bromides. Recent work has demonstrated that less electron rich oxygen nucleophiles, such as N-hydroxysuccinimide, can also be competent reaction partners furnishing activated esters.23 Thiols can also be exploited as nucleophiles as was earlier shown by Alper and co-workers with aryl iodides in a phosphonium salt ionic liquid as solvent.24 We recently published another protocol whereby the product selectivity turned out to be highly ligand and base dependent with aryl 598


Article

Accounts of Chemical Research Scheme 15. Carbonylative α-Arylation of Ketones with (Hetero)Aryl Iodides and Bromides

Scheme 16. Magnesium(II)chloride Mediated β-Ketoester (A) and 1,3-Diketone (B) Synthesis

iodides.25 Under otherwise identical conditions, a DPEphos derived Pd-catalyst provided almost exclusively the thioester when combined with sodium acetate as base, while tBuJosiPhos and sodium tert-butoxide furnished the thioether (Scheme 13). Applying the former catalytic system enabled the conversion of both electron rich and electron poor aryl iodides into their corresponding thioesters. Conducting the thiocarbonylation in anisole at 60−120 °C allows for the use of aryl, vinyl, and benzyl bromides, as well as benzyl chloride, to generate the corresponding thioesters (Scheme 14).26 With these thioesters at hand, a range of different amides can be accessed through simple nucleophilic acyl substitution.

temperature increase in combination with a ligand exchange to DPPP (Scheme 15).30 In our hands, excellent selectivity toward the desired Cacylated product in the carbonylative α-arylation was generally observed, while Beller and co-workers demonstrated a high preference for O-acylation under slightly different catalytic conditions.31 While the method developed for the carbonylative αarylation of ketones provided a wide scope of 1,3-diketones, the use of a strong and hygroscopic base, such as NaHMDS, does limit the convenience of the reaction setup, as well as the functional group tolerance. Employing 1,3-dicarbonyls such as malonates as starting materials eliminates the need for a strong base, and postcoupling decarboxylation provides the equivalent 1,3-ketoesters. In order to enhance the α-proton acidity while simultaneously increasing the C- to O-acylation ratio, MgCl2 was introduced. In combination with triethylamine, this Lewis acid promotes efficient enolization of monoester potassium malonates enabling their participation in carbonylative cross coupling. This strategy transformed a diverse collection of both aryl and heteroaryl bromides into their corresponding 1,3ketoesters (Scheme 16A).32 Additionally, aryl triflates, electron poor aryl chlorides, and benzyl chloride were found to be competent electrophiles. Instead of a decarboxylative protocol to 1,3-dicarbonyl compounds, we exploited an acid mediated postdeacetylation step. In doing so, such compounds were obtained under the same mild conditions using the Et3N/MgCl2 base combination. Hence, subjecting acetylacetone to identical carbonylative

Carbonylative α-Arylation

With the concurrent development of intermolecular Pdcatalyzed α-arylation of ketones in 1997 by the groups of Buchwald and Hartwig, efficient access to benzylic and allylic carbonyl derivatives starting from a simple enolizable compound was realized.27,28 The carbonylative counterpart to this useful transformation provides access to 1,3-dicarbonyls, which are key components in the preparation of heterocycles. Employing COware, the first intermolecular carbonylative αarylation of non-malonate type reagents was realized.29 Subjecting an array of (hetero)aryl iodides, as well as different aryl,alkyl- and alkyl,alkyl-ketones to the optimized reaction conditions efficiently furnished a broad scope of 1,3-diketones (Scheme 15). Inclusion of the aryl bromides for this carbonylative α-arylation of ketones only required a slight 599


Article

Accounts of Chemical Research Scheme 17. Carbonylative α-Arylation To Form β-Ketoamides and β-Keto(thio)esters

conditions used for the β-ketoesters, followed by deacetylation with 2 M hydrochloric acid, afforded the desired 1,3-diketones with excellent C- to O-acylation and acyl hydrolysis selectivity (Scheme 16B).33 Conducting the reaction on gram scale only served to improved the yields, and carbonyl labeling was once again realized by exchanging COgen with 13COgen under identical conditions. Similarly, β-ketoamides, β-ketoesters, and β-ketothioesters proved reactive under the carbonylative α-arylation coupling conditions employing the mild basic system with MgCl2 and triethylamine, furnishing β-aryl-β-ketoamides, -esters, and -thioesters after acidic deacetylation (Scheme 17).34 Next, 2-pyridylacetone was examined as the nucleophile, speculating that chelation of MgCl 2 to 1,3-dicarbonyl compounds would be efficient for the carbonylative α-arylation of substrates displaying other heteroatoms than oxygen. Both aryl and heteroaryl bromides proved to be substrates when utilizing the XantPhos ligated palladium catalyst and a slight excess of COgen. Subsequent acidic treatment efficiently

Scheme 19. Carbonylative α-Arylation To Afford 3-Acyl-2Oxindoles

Scheme 18. Carbonylative α-Arylation of (2-Azaaryl)methyl Anion Equivalents deacetylated the initially formed diketones. Exchanging 2pyridylacetone for other heteroaryl acetone substrates with the ring nitrogen correctly positioned for magnesium chelation successfully provided the corresponding α-arylated products (Scheme 18).35 In the carbonylative α-arylation of 2-oxindoles, the MgCl2 and triethylamine base-system proved highly competent, although in this case, only a single carbonyl functionality was necessary for magnesium coordination. Omitting the Lewis acid halted product formation. The combination of Pd(OAc)2 as the metal source with XantPhos proved ideal for this reaction thus providing access to biologically important 3-acyl-2-oxindoles (Scheme 19).36 The electron withdrawing nature of nitro-groups and consequently the low pKa of α-positioned protons in nitroalkanes are central to a number of classical transformations. This same feature could make them ideal substrates for carbonylative α-arylations, and indeed the same combination of magnesium chloride and triethylamine set the stage for the carbonylative formation of α-nitroketones. Applying nitromethane as the solvent, we obtained the corresponding α-nitroketones in good yields, being versatile precursors for 600


Article


ligands added (Scheme 22). Key to the high selectivity toward the desired benzophenone product over biaryl formation was the employment of a low catalyst loading.40 The inclusion of aryl bromides as substrates for the carbonylative Suzuki−Miyaura cross coupling under low pressure conditions in COware was effected in collaboration with Prof. G. A. Molander.41 The use of cataCXium A·HI proved essential to obtaining good product to biaryl selectivities. Interestingly, applying either aryl trihydroxyborates or DABO boronates in combination with aryl and heteroaryl bromides allowed for base-free and high yielding carbonylative Suzuki−Miyaura cross couplings (Scheme 23). For both catalytic systems, the simple switch from unlabeled COgen to 13 COgen allowed for the synthesis of carbon-13 labeled ketones.

both heterocyclic synthesis and phenylethylamine type compounds (Scheme 20).37 Scheme 20. Three-Step Pyrrole Synthesis via Carbonylative α-Arylation of Nitromethane

Carbonylative Heck

In line with the use of substrates bearing a single electron withdrawing group, Beller and co-workers recently developed a carbonylative α-arylation of nitriles, using strong base and elevated CO pressures in an autoclave system. However, this setup is not suitable for isotopic acyl labeling, because a large excess of high pressure carbon monoxide is applied. In a collaboration, reaction conditions were adapted to the COware system, which allowed for the isolation of carbon-13 labeled βketo nitriles from 13COgen.38 Combining the carbon isotope labeling technology developed with COware and 13COgen with the classical construction of heterocycles starting from the generated 1,3-dicarbonyl compounds presented above allows for an array of both isotopically labeled and unlabeled heterocycles to be realized using literature protocols (Scheme 21).29,32,35

Another Nobel Prize winning C−C bond forming reaction is the Pd-catalyzed Heck reaction for the synthesis of functionalized olefins. Limited attention has been paid to the threecomponent carbonylative version of this transformation despite it being an attractive route to α,β-unsaturated ketones.42 With a special emphasis on chalcone derivatives, mild carbonylative Heck conditions were developed in COware (Scheme 23).43 Intriguingly, the optimized reaction conditions feature a 5:1 ratio of palladium to phosphine ligand, with reductive carbonylative Heck reaction representing the major byproduct. A selected chalcone was cyclized to its indanone derivative using neat TFA affording 53h in a 74% isolated yield (Scheme 23). Utilizing SilaCOgen as a CO surrogate retarded the reaction, possibly due to its faster decarbonylation. Reducing the rate of carbon monoxide release from SilaCOgen by using a catalytic amount of fluoride ensured yields comparable to those obtained with COgen.5 Changing the focus to butyl vinyl ether as the nucleophile paved the way to 1,3-ketoaldehyde synthons via a carbonylative Heck type reaction.44 Starting from a range of aryl bromides, these products were obtained in good yields by applying tri-tertbutylphosphine in the unusual palladium to ligand ratio of 2:1. Additionally, increasing the amount of COgen and thus CO applied in the transformation to 2.2 equiv proved beneficial

Carbonylative Suzuki−Miyaura

The Pd-catalyzed Suzuki−Miyaura reaction is one of the most widely used transition metal catalyzed cross couplings in organic chemistry.39 In its carbonylative version, this transformation enables the formation of nonsymmetrical ketones. With a special focus on the generation of benzophenones, aryl iodides were found to be good substrates even under an ambient oxygen-containing atmosphere using PdCl2 in combination with potassium carbonate and no additional

Scheme 21. Heterocycle Synthesis through Carbonylative α-Arylation

601


Article

Accounts of Chemical Research Scheme 22. Carbonylative Suzuki−Miyaura Reaction Starting from (Hetero)Aryl Iodides or Bromides

employing COgen in combination with COware, we could prepare the desired alkynones in a high yielding transformation starting from their corresponding aryl bromides with an excellent substrate scope (Scheme 25).46 These products not

Scheme 23. Carbonylative Heck Reaction with Styrenes

Scheme 25. Carbonylative Sonogashira Reaction with (Hetero)Aryl Bromides

only are interesting in their own right but also represent attractive starting materials for the preparation of a range of heterocycles, including pyrimidines.

(Scheme 24). With these 1,3-ketoaldehyde synthons and their labeled versions at hand, easy access to a number of nitrogen containing heterocycles, as well as acetophenones became a reality.

Carbonylative C−H Activation

While classical carbonylative and direct cross coupling generally ensure excellent regioselectivity by employing a nucleophilic coupling partner capable of undergoing transmetalation, the corresponding C−H functionalization is more attractive in terms of atom economy, cost efficiency, and waste generation. With the COware reactor, a comprehensive screening and optimization of reaction conditions allowed for the development of an intermolecular Pd-catalyzed carbonylative C−H functionalization with aryl bromides as the coupling partner.47 With polyfluoronated arenes, benzopolyfluorophenones could be synthesized at reaction temperatures of only 80 °C. Such reaction temperatures being typically required for promoting the oxidative addition step with aryl bromides suggests that the C−H functionalization step is relatively facile at this temperature (Scheme 26).

Scheme 24. Carbonylative Heck Reactions Applying Butyl Vinyl Ether

■

OTHER GASES IN COware Although the COware system was initially developed for COrelated chemistry, in principle many gaseous reagents generated from stable precursors could be utilized with this setup. The two-chamber reactor presents a number of advantages, including ease of setup, safety, and the ability to generate and utilize isotopically labeled gases in stoichiometric quantities.

Carbonylative Sonogashira

Today, the Sonogashira cross coupling is a key reaction for the preparation of functionalized alkynes, especially when aryl or vinyl substituents are desired. Introducing carbon monoxide into the reaction mixture leads to the formation of propargylic ketones from the generation of two new C−C bonds.45 By 602


Article


usefulness of such compounds for a variety of tasks. Using specifically functionalized styrene and the well-known metathesis reaction allowed for the facile preparation of unlabeled ethylene in COware, while the corresponding terminally isotope labeled styrene derivatives efficiently generate [D4]ethylene, [13C2]-ethylene, and for the first time fully isotopically labeled [13C2,D4]-ethylene.49 Combining this controlled production of otherwise unavailable or costly isotopically labeled gases in one chamber of COware with an ethylene utilizing transformation allows for the incorporation of isotopic labels into the core of, for example, benzene or pyridine rings (Scheme 28).

Scheme 26. C−H Activation Dependent Carbonylation of (Hetero)Aryl Bromides

Scheme 28. Isotopically Labeled Ethylene Generated and Utilized in COware Hydrogenation

Hydrogen is a key reagent most commonly known for its use in reductive transformations. Chemists around the world handle this diatomic gas routinely from small-scale applications to multiton industrial processes. Moreover, the installation of deuterium and tritium into biologically active compounds is of continuous interest to both academia and industry for a number of reasons. Using the COware two-chamber system, hydrogen gas is easily generated from the classical reaction of metallic zinc with hydrochloric acid in one chamber and subsequently utilized in the secondary chamber. While this method for the generation of gaseous hydrogen is certainly no new reaction, it does provides a convenient way of subjecting a transformation to pressurized hydrogen in COware (up to 5 bar). In addition, simply changing from aqueous hydrogen chloride to deuterium chloride enables the formation and utilization of pressurized deuterium gas from common, inexpensive, and storable reagents (Scheme 27).48 Ethylene Generation and Utilization

■

Ethylene is the most commercially produced organic compound with a global yearly production of over 140 million tons. Nonetheless, isotopically labeled ethylene with combinations of both deuterium and carbon-13 is neither easily prepared nor easily handled, which contrasts the potential

CONCLUSIONS AND FUTURE PERSPECTIVES The two-chamber system, COware, has established itself as a competent tool for conducting safe reactions with gaseous reagents on a laboratory scale. In particular, this reactor has

Scheme 27. Reduction and Deuterium Incorporation Using H2 or D2 in COware

603


Article

Accounts of Chemical Research been developed and utilized for a wide number of both wellknown and new transition metal catalyzed carbonylative transformations, because its combination with COgen, SilaCOgen, or similar carbon monoxide releasing molecules eliminates the requirement for handling large volumes of gaseous CO from a pressurized cylinder. Milligram to gram scale transformations can be routinely undertaken. When combined with 13COgen, Sila13COgen, or 14COgen, this carbonylative technology is ideal for the facile and late-stage introduction of isotopic labels into a wide array of biologically relevant compounds because many of the carbonylative transformations have been optimized with near stoichiometric amounts of generated CO. The two-chamber system has also proven to be valuable for other gases than CO. At present, hydrogen and deuterium gas are produced and employed for reductions, as well as hydrogen−deuterium exchange, while ethylene can be prepared in a near quantitative manner in a number of isotopic derivatives. Future work into other applicable gases is currently underway and includes carbon-13 and -14 acetylene, as well as hydrogen cyanide.

■

at the Institut de Chimie des Substances Naturelles, France (Prof. G. Ourisson and Dr. D. Grierson), and at the Carlsberg Laboratories (Prof. K. Bock), he was employed as Chargé de Recherche (CR1) first at the Université d’Orléans and then at Université Paris XI (1992− 1997). In 1997, he became Associate Professor at the Department of Chemistry, Aarhus University, and was promoted to full Professor of Organic Chemistry in 2002. Troels Skrydstrup was elected as a fellow of the Royal Danish Academy of Sciences in 2008, and in 2012, he was knighted by the Danish Queen.

■

ACKNOWLEDGMENTS The authors thank Dennis U. Nielsen for insightful discussions during the preparation of this manuscript.

■

REFERENCES

(1) While CO surrogates significantly improve safety; appropriate safety measures should still be taken, for example, proper ventilation. (2) Morimoto, T.; Kakiuchi, K. Evolution of carbonylation catalysis: No need for carbon monoxide. Angew. Chem., Int. Ed. 2004, 43, 5580− 5588. (3) Gautam, P.; Bhanage, B. M. Recent advances in the transition metal catalyzed carbonylation of alkynes, arenes and aryl halides using CO surrogates. Catal. Sci. Technol. 2015, 5, 4663−4702 and references therein.. (4) Roberts, B.; Liptrot, D.; Alcaraz, L.; Luker, T.; Stocks, M. Molybdenum-Mediated Carbonylation of Aryl Halides with Nucleophiles Using Microwave Irradiation. Org. Lett. 2010, 12, 4280−4283. (5) Friis, S. D.; Taaning, R. H.; Lindhardt, A. T.; Skrydstrup, T. Silacarboxylic Acids as Efficient Carbon Monoxide Releasing Molecules: Synthesis and Application in Palladium-Catalyzed Carbonylation Reactions. J. Am. Chem. Soc. 2011, 133, 18114−18117. (6) Hermange, P.; Lindhardt, A. T.; Taaning, R. H.; Bjerglund, K.; Lupp, D.; Skrydstrup, T. Ex Situ Generation of Stoichiometric and Substoichiometric 12CO and 13CO and Its Efficient Incorporation in Palladium Catalyzed Aminocarbonylations. J. Am. Chem. Soc. 2011, 133, 6061−6071. (7) Burhardt, M. N.; Taaning, R.; Nielsen, N. C.; Skrydstrup, T. Isotope-Labeling of the Fibril Binding Compound FSB via a PdCatalyzed Double Alkoxycarbonylation. J. Org. Chem. 2012, 77, 5357− 5363. (8) Lindhardt, A. T.; Simonssen, R.; Taaning, R. H.; Gøgsig, T. M.; Nilsson, G. N.; Stenhagen, G.; Elmore, C. S.; Skrydstrup, T. 14Carbon monoxide made simple − novel approach to the generation, utilization, and scrubbing of 14carbon monoxide. J. Labelled Compd. Radiopharm. 2012, 55, 411−418. (9) Wu, X.-F.; Neumann, H.; Beller, M. Synthesis of Heterocycles via Palladium-Catalyzed Carbonylations. Chem. Rev. 2013, 113, 1−35. (10) Liu, Q.; Zhang, H.; Lei, A. Oxidative Carbonylation Reactions: Organometallic Compounds (R-M) or Hydrocarbons (R-H) as Nucleophiles. Angew. Chem., Int. Ed. 2011, 50, 10788−10799. (11) Catalytic Carbonylation Reactions; Beller, M., Ed.; Topics in Organometallic Chemistry; Springer, 2006; Vol. 18. (12) Wu, L. P.; Liu, Q.; Jackstell, R.; Beller, M. Carbonylations of Alkenes with CO Surrogates. Angew. Chem., Int. Ed. 2014, 53, 6310− 6320. (13) Schoenberg, A.; Heck, R. F. Palladium-catalyzed amidation of aryl, heterocyclic, and vinylic halides. J. Org. Chem. 1974, 39, 3327− 3331. (14) Nordeman, P.; Odell, L. R.; Larhed, M. Aminocarbonylations Employing Mo(CO)6 and a Bridged Two-Vial System: Allowing the Use of Nitro Group Substituted Aryl Iodides and Aryl Bromides. J. Org. Chem. 2012, 77, 11393−11398. (15) Friis, S. D.; Skrydstrup, T.; Buchwald, S. L. Mild Pd-Catalyzed Aminocarbonylation of (Hetero)Aryl Bromides with a Palladacycle Precatalyst. Org. Lett. 2014, 16, 4296−4299.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. *E-mail: [email protected]. Funding

We are deeply appreciative of the generous financial support from the Danish National Research Foundation (Grant No. DNRF118), the Danish Innovation Foundation through the iDEA project, the Villum Foundation, the Danish Council for Independent Research: Technology and Production Sciences, and Aarhus University for generous financial support. The Authors also thank Dennis U. Nielsen for insightful discussions during the preparation of this manuscript. Notes

The authors declare the following competing financial interest(s): Anders T. Lindhardt and Troels Skrydstrup are co-owners of SyTracks a/s, which commercializes the twochamber technology. Biographies Stig D. Friis received his M.Sc. and Ph.D. degrees from Aarhus University in 2012 and 2014, respectively, for his research in the area of transition metal catalyzed carbonylations and carbon monoxide releasing molecules under the supervision of Prof. Troels Skrydstrup. During his Ph.D. studies, he spent time in the laboratories of Prof. Stephen L. Buchwald in the spring of 2013. Stig Friis has since then returned to MIT to carry out his postdoctoral studies with Prof. Buchwald. Anders T. Lindhardt obtained his M.Sc. and Ph.D. degrees from Aarhus University (2005 and 2007) for his research in the field of Pdand Ni-catalyzed coupling reactions. Following 4 years of postdoctorial stays at Aarhus University, Anders Lindhardt left academia for an industrial position within Research & Development. In 2013, he returned to Aarhus University to start his independent carrier, working at the Department of Engineering within the field of flow chemistry and continuous processes. Troels Skrydstrup received his M.Sc. and Ph.D. degrees from the Technical University of Denmark (1985 and 1988) under the supervision of Prof. Anders Kjær. After several postdoctoral periods 604


Article

Accounts of Chemical Research (16) Iizuka, M.; Kondo, Y. Remarkable ligand effect on the palladium-catalyzed double carbonylation of aryl iodides. Chem. Commun. 2006, 1739−1741. (17) Nielsen, D. U.; Neumann, K.; Taaning, R. H.; Lindhardt, A. T.; Modvig, A.; Skrydstrup, T. Palladium-Catalyzed Double Carbonylation Using Near Stoichiometric Carbon Monoxide: Expedient Access to Substituted 13C2-Labeled Phenethylamines. J. Org. Chem. 2012, 77, 6155−6165. (18) Nielsen, D. U.; Taaning, R. H.; Lindhardt, A. T.; Gøgsig, T. M.; Skrydstrup, T. Palladium-Catalyzed Approach to Primary Amides Using Nongaseous Precursors. Org. Lett. 2011, 13, 4454−4457. (19) Andersen, T. L.; Caneschi, W.; Ayoub, A.; Lindhardt, A. T.; Couri, M. R. C.; Skrydstrup, T. 1,2,4- and 1,3,4-Oxadiazole Synthesis by Palladium-Catalyzed Carbonylative Assembly of Aryl Bromides with Amidoximes or Hydrazides. Adv. Synth. Catal. 2014, 356, 3074− 3082. (20) Lian, Z.; Friis, S. D.; Lindhardt, A. T.; Skrydstrup, T. PalladiumCatalyzed Carbonylation of Aryl Bromides with N-Substituted Cyanamides. Synlett 2014, 25, 1241−1245. (21) Yin, H.; de Almeida, A. M.; de Almeida, M. V.; Lindhardt, A. T.; Skrydstrup, T. Synthesis of Acyl Carbamates via Four Component PdCatalyzed Carbonylative Coupling of Aryl Halides, Potassium Cyanate, and Alcohols. Org. Lett. 2015, 17, 1248−1251 and reference 13a therein.. (22) Xin, Z.; Gøgsig, T. M.; Lindhardt, A. T.; Skrydstrup, T. An Efficient Method for the Preparation of Tertiary Esters by PalladiumCatalyzed Alkoxycarbonylation of Aryl Bromides. Org. Lett. 2012, 14, 284−287. (23) de Almeida, A. M.; Andersen, T. L.; Lindhardt, A. T.; de Almeida, M. V.; Skrydstrup, T. General Method for the Preparation of Active Esters by Palladium-Catalyzed Alkoxycarbonylation of Aryl Bromides. J. Org. Chem. 2015, 80, 1920−1928. (24) Cao, H.; McNamee, L.; Alper, H. Palladium-Catalyzed Thiocarbonylation fo Iodoarenes with Thiols in Phosphonium Salt Ionic Liquids. J. Org. Chem. 2008, 73, 3530−3534. (25) Burhardt, M. N.; Taaning, R. H.; Skrydstrup, T. Pd-Catalyzed Thiocarbonylation with Stoichiometric Carbon Monoxide: Scope and Applications. Org. Lett. 2013, 15, 948−951. (26) Burhardt, M. N.; Ahlburg, A.; Skrydstrup, T. Palladiumcatalyzed thiocarbonylation of aryl, vinyl, and benzyl bromides. J. Org. Chem. 2014, 79, 11830−11840. (27) Palucki, M.; Buchwald, S. L. Palladium-catalyzed alpha-arylation of ketones. J. Am. Chem. Soc. 1997, 119, 11108−11109. (28) Hamann, B. C.; Hartwig, J. F. Palladium-catalyzed direct alphaarylation of ketones. Rate acceleration by sterically hindered chelating ligands and reductive elimination from a transition metal enolate complex. J. Am. Chem. Soc. 1997, 119, 12382−12383. (29) Gøgsig, T. M.; Taaning, R. H.; Lindhardt, A. T.; Skrydstrup, T. Palladium-Catalyzed Carbonylative α-Arylation for Accessing 1,3Diketones. Angew. Chem., Int. Ed. 2012, 51, 798−801. (30) Nielsen, D. U.; Lescot, C.; Gogsig, T. M.; Lindhardt, A. T.; Skrydstrup, T. Pd-Catalyzed Carbonylative alpha-Arylation of Aryl Bromides: Scope and Mechanistic Studies. Chem. - Eur. J. 2013, 19, 17926−17938. (31) Schranck, J.; Tlili, A.; Neumann, H.; Alsabeh, P. G.; Stradiotto, M.; Beller, M. A Selective Palladium-Catalyzed Carbonylative Arylation of Aryl Ketones to Give Vinylbenzoate Compounds. Chem. - Eur. J. 2012, 18, 15592−15597. (32) Korsager, S.; Nielsen, D. U.; Taaning, R. H.; Skrydstrup, T. Access to β-Keto Esters by Palladium-Catalyzed Carbonylative Coupling of Aryl Halides with Monoester Potassium Malonates. Angew. Chem., Int. Ed. 2013, 52, 9763−9766. (33) Korsager, S.; Nielsen, D. U.; Taaning, R. H.; Lindhardt, A. T.; Skrydstrup, T. Direct Route to 1,3-Diketones by Palladium-Catalyzed Carbonylative Coupling of Aryl Halides with Acetylacetone. Chem. Eur. J. 2013, 19, 17687−17691. (34) Nielsen, D. U.; Korsager, S.; Lindhardt, A. T.; Skrydstrup, T. A Palladium-Catalyzed Carbonylative-Deacetylative Sequence to 1,3Keto Amides. Adv. Synth. Catal. 2014, 356, 3519−3524.

(35) Jusseau, X.; Yin, H.; Lindhardt, A. T.; Skrydstrup, T. PalladiumCatalyzed Carbonylative Coupling of (2-Azaaryl)methyl Anion Equivalents with (Hetero)Aryl Bromides. Chem. - Eur. J. 2014, 20, 15785−15789. (36) Lian, Z.; Friis, S. D.; Skrydstrup, T. Palladium-Catalyzed Carbonylative alpha-Arylation of 2-Oxindoles with (Hetero) aryl Bromides: Efficient and Complementary Approach to 3-Acyl-2oxindoles. Angew. Chem., Int. Ed. 2014, 53, 9582−9586. (37) Lian, Z.; Friis, S. D.; Skrydstrup, T. Palladium-catalysed carbonylative alpha-arylation of nitromethane. Chem. Commun. 2015, 51, 3600−3603. (38) Schranck, J.; Burhardt, M.; Bornschein, C.; Neumann, H.; Skrydstrup, T.; Beller, M. Palladium-Catalyzed Carbonylative alphaArylation to beta-Ketonitriles. Chem. - Eur. J. 2014, 20, 9534−9538. (39) Colacot, T. J. The 2010 Nobel Prize in Chemistry: PalladiumCatalysed Cross-Coupling. Platinum Met. Rev. 2011, 55, 84−90. (40) Ahlburg, A.; Lindhardt, A. T.; Taaning, R. H.; Modvig, A. E.; Skrydstrup, T. An Air-Tolerant Approach to the Carbonylative SuzukiMiyaura Coupling: Applications in Isotope Labeling. J. Org. Chem. 2013, 78, 10310−10318. (41) Bjerglund, K. M.; Skrydstrup, T.; Molander, G. A. Carbonylative Suzuki Couplings of Aryl Bromides with Boronic Acid Derivatives under Base-Free Conditions. Org. Lett. 2014, 16, 1888−1891. (42) Satoh, T.; Itaya, T.; Okuro, K.; Miura, M.; Nomura, M. Palladium-Catalyzed Cross-Carbonylation of Aryl Iodides with 5Membered Cyclic Olefins. J. Org. Chem. 1995, 60, 7267−7271. (43) Hermange, P.; Gøgsig, T. M.; Lindhardt, A. T.; Taaning, R. H.; Skrydstrup, T. Carbonylative Heck Reactions Using CO Generated ex Situ in a Two-Chamber System. Org. Lett. 2011, 13, 2444−2447. (44) Gogsig, T. M.; Nielsen, D. U.; Lindhardt, A. T.; Skrydstrup, T. Palladium Catalyzed Carbonylative Heck Reaction Affording Monoprotected 1,3-Ketoaldehydes. Org. Lett. 2012, 14, 2536−2539. (45) Kobayashi, T.; Tanaka, M. Carbonylation of organic halides in the presence of terminal acetylenes; novel acetylenic ketone synthesis J. Chem. Soc. J. Chem. Soc., Chem. Commun. 1981, 333−334. (46) Neumann, K. T.; Laursen, S. R.; Lindhardt, A. T.; BangAndersen, B.; Skrydstrup, T. Palladium-Catalyzed Carbonylative Sonogashira Coupling of Aryl Bromides Using Near Stoichiometric Carbon Monoxide. Org. Lett. 2014, 16, 2216−2219. (47) Lian, Z.; Friis, S. D.; Skrydstrup, T. C-H activation dependent Pd-catalyzed carbonylative coupling of (hetero)aryl bromides and polyfluoroarenes. Chem. Commun. 2015, 51, 1870−1873. (48) Modvig, A.; Andersen, T. L.; Taaning, R. H.; Lindhardt, A. T.; Skrydstrup, T. Two-Chamber Hydrogen Generation and Application: Access to Pressurized Deuterium Gas. J. Org. Chem. 2014, 79, 5861− 5868. (49) Min, G. K.; Bjerglund, K.; Kramer, S.; Gogsig, T. M.; Lindhardt, A. T.; Skrydstrup, T. Generation of Stoichiometric Ethylene and Isotopic Derivatives and Application in Transition-Metal-Catalyzed Vinylation and Enyne Metathesis. Chem. - Eur. J. 2013, 19, 17603− 17607.

605


The Development and Application of Two-Chamber Reactors and

Recommend Documents