Protecting group-free, selective cross-coupling ofalkyltrifluoroborates with borylated aryl bromidesvia photoredox/nickel dual catalysisYohei Yamashitaa,b,1, John C. Tellisa,1, and Gary A. Molandera,2

aRoy and Diana Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, Philadelphia, PA 19104-6323; and bProcess ChemistryLaboratories, Astellas Pharma Inc., Ibaraki 318-0001, Japan

Edited by John F. Hartwig, University of California, Berkeley, CA, and approved August 19, 2015 (received for review May 18, 2015)

Orthogonal reactivity modes offer substantial opportunities forrapid construction of complex small molecules. However, moststrategies for imparting orthogonality to cross-coupling reactionsrely on differential protection of reactive sites, greatly reducingboth atom and step economies. Reported here is a strategy fororthogonal cross-coupling wherein a mechanistically distinct acti-vation mode for transmetalation of sp3-hybridized organoboron re-agents enables C-C bond formation in the presence of variousprotected and unprotected sp2-hybridized organoborons. This mani-fold has the potential for broad application, because orthogonalityis inherent to the activation mode itself. The diversification po-tential of this platform is shown in the rapid elaboration of atrifunctional lynchpin through various transition metal-catalyzed pro-cesses without nonproductive deprotection or functional groupmanipulation steps.

orthogonal cross-coupling | photoredox/nickel dual catalysis |single-electron transmetalation | organotrifluoroborate

The development of strategies for rapid generation of molec-ular complexity from simple building blocks is central to the

advancement of organic synthesis. A general platform for achieve-ment of this ideal would have a profound impact on those disci-plines whose progress is reliant on the efficient production ofcomplex small molecules. Transition metal-catalyzed cross-cou-pling reactions have been instrumental to advancement in thisrealm, because these mild and functional group-tolerant methodsare naturally amenable to the late-stage union of fully elaboratedand minimally protected molecular fragments (1). Within thiscontext, a variety of methods have been developed for iterativeassembly of small-molecule scaffolds through orthogonal cross-coupling enabled by judicious attenuation of reactivity at eitherthe organometallic or organic (pseudo-)halide site (2–5).Among these various approaches, the greatest advancements

have been realized in Suzuki cross-coupling, because organoboronreagents are chemically robust and can be rendered inactivethrough topological and/or electronic differentiation or appropri-ate selection of heteroatomic substituents (5, 6). In an example ofthe former, Morken and coworkers (7–10) have reported the se-lective cross-coupling of geminal and vicinal diboronates throughPd catalysis. However, the requirement for proximal boryl sub-stitution severely restricts the utility of these methods as a result ofreduced substrate availability and reaction scope.More general approaches to orthogonal cross-coupling have

been developed through the use of various boronic acid maskinggroups as a means to control site selectivity. These strategies relyon the protection of one organoboron reagent in a latent formunable to engage the Pd catalyst in transmetalation, whereasanother, existing in a reactive form, participates in the initial C-Cbond formation. Continuation of the iterative process is thenenabled by subsequent deprotection of the masked boronic acid.A variety of protecting groups have been explored for thesepurposes, most notably N-methyliminodiacetic acid (MIDA)(6, 11–13) and 1,8-diaminonaphthalene (BDAN) (14–18). In

addition to these reagents, isolated reports have used organo-trifluoroborates (RBF3K) (19) and catecholboronates (20) asprotected boronic acid equivalents in specific settings. Althougheffective, these protecting group strategies are inherently sub-optimal in terms of atom and step economy (21). Nevertheless,this manifold of orthogonal protection is the only possible recoursewhen the desired iterative cross-coupling makes use of the samefundamental activation mode [i.e., transmetalation of an organo-boron reagent to a Pd(II) intermediate].In view of this paradigmatic limitation, we sought to develop

an unprecedented class of orthogonal cross-coupling in whichorganoboron sites are mechanistically differentiated, thus allowingorthogonal reactivity without artificial attenuation of reactivity. Thisstrategy would be enabled by our recently developed single-electrontransmetalation manifold, wherein sp3-hybridized organoboron re-agents participate in Ni-catalyzed cross-coupling through oxidativefragmentation to an alkyl radical promoted by an Ir photoredoxcatalyst (22–24). Our previous observation (22) that an sp3-hybridizedorganotrifluoroborate participates in photoredox/nickel dual-catalytic cross-coupling selectively in the presence of an sp2-hybridized organotrifluoroborate confirmed that the single-electrontransmetalation activation mode is, indeed, orthogonal to the tra-ditional two-electron regime under relevant reaction conditions.Furthermore, we were confident that the trivalent sp2-hybridizedorganoborons [pinacolboronate (BPin), neopentylglycolboronate(BNeop), and B(OH)2] used in conventional cross-couplingreactions would be left intact during the course of the photo-redox cross-coupling, because decomposition pathways, such as

Significance

Efficient assembly of small-molecule scaffolds is among themost fundamental goals of organic synthesis. Iterative syn-thesis, wherein predefined building blocks are unified in an“assembly line” fashion using only a small number of reactiontypes, is an attractive means for achieving this ideal. Thesemethods are particularly well-suited for applications in drugdiscovery, agrochemistry, and materials science, where rapidgeneration of structural diversity is a central objective. Astrategy is described in which two reactive sites are differen-tiated by their preferred mode of reactivity (single vs. twoelectron). This unique platform allows discrimination betweenthe two sites without artificial blocking of reactivity, stream-lining the iterative process by removing the need for depro-tection or interconversion of functional groups.

Author contributions: J.C.T. and G.A.M. designed research; Y.Y. and J.C.T. performedresearch; Y.Y. and J.C.T. analyzed data; and J.C.T. and G.A.M. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1Y.Y. and J.C.T. contributed equally to this work.2To whom correspondence should be addressed. Email: gmolandr@sas.upenn.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1509715112/-/DCSupplemental.

12026–12029 | PNAS | September 29, 2015 | vol. 112 | no. 39 www.pnas.org/cgi/doi/10.1073/pnas.1509715112

Dow

nloa

ded

by g

uest

on

May

30,

202

0

protodeboronation and oxidation, are unlikely to occur underthe exceptionally mild conditions used therein (Fig. 1).Our studies were initiated with analysis of the reaction of

potassium benzyltrifluoroborate with 4-bromophenylboronic acidpinacol ester. Slight modification of the reaction conditions pre-viously reported for the photoredox/Ni dual-catalytic cross-couplingof primary benzyltrifluoroborates allowed the efficient C-C bondformation between these partners to form the borylated diaryl-methane product selectively. [No evidence of biaryl formationarising from competitive cross-coupling of the BPin moiety wasobserved by GC-MS analysis.] Specifically, use of ethereal solvents,such as THF and dioxane, and 2,2,6,6-tetramethylpiperidine inplace of 2,6-lutidine suppressed formation of undesired side prod-ucts and improved conversion. [The major side product observedby GC-MS was benzylBPin (PhCH2BPin). We suggest that thismay be formed by fluoride abstraction from RBF3K by BF3(generated during the course of the reaction by oxidativefragmentation of RBF3K) followed by transesterification withthe aryl boronate.] Increased catalyst loading [3 mol % Ir[dFCF3ppy]2(bpy)PF6 Ir (dFCF3ppy, 2-(2,4-difluorophenyl)-5-(trifluoromethyl)pyridine; bpy, 2,2′-bipyridine) and 5 mol % Ni(COD)2 (COD, 1,5-cyclooctadiene)] was necessary to inducecomplete conversion within 24 h, although reduced loadings wereeffective at prolonged reaction times. Under these conditions, theorthogonal cross-coupling could be achieved in 76% yield afteroxidation to the corresponding alcohol.We next sought to explore the scope of the reaction with

regard to the substituents about the sp2-hybridized organoboronreagent. Trivalent boronate reagents were oxidized before iso-lation, because these compounds are prone to decomposition viahydrolysis and/or protodeboronation during column chroma-tography, thereby artificially suppressing the observed yields. Thiscross-coupling/oxidation procedure proved more consistent andreliable for assessment of reaction efficiency compared with di-rect isolation of the boronate esters or boronic acids themselves.11B NMR analysis of crude reaction mixtures before oxidation

confirmed that the boronate was intact after cross-coupling, andsubsequent studies involving transition metal-catalyzed se-quential functionalization unequivocally confirm the survivalof the boronate functional group (vide infra). Nearly all com-monly used boronic acid derivatives smoothly underwent or-thogonal cross-coupling, including those of neopentylglycol,MIDA, and 1,8-diaminonaphthalene. Particularly remarkableis the successful cross-coupling of unprotected 4-bromophenylbo-ronic acid, albeit in moderate yield. This example is especiallynotable given the high reactivity and reduced stability of freeboronic acids in cross-coupling relative to the related boronateesters (25). To our knowledge, this transformation is the firstreported Csp3-Csp2 cross-coupling that is tolerant of boronic acids aslatent functional groups, effectively highlighting the mildness andfunctional group tolerance of this photoredox/nickel dual-catalysisplatform (Fig. 2).

A

B

Fig. 1. Strategies for orthogonal cross-coupling. (A) Conventional or-thogonal protection approach requiring three discrete steps. (B) Photo-redox cross-coupling approach relying on single-electron transmetalationof sp3-hybridized organoboron reagents to establish orthogonal reactivitypatterns, allowing formation of two bonds in a single two-stage process. 9-BBN,9-borabicyclo[3.3.1]nonane; BMIDA, N-methyliminodiacetic acid boronate.

Fig. 2. Substrate scope for orthogonal cross-coupling of benzylic trifluorobo-rates with borylated aryl bromides. BMIDA, N-methyliminodiacetic acid boro-nate; CFL, compact fluorescent lamp; dtbbpy, 4,4′-di-tert-butylbipyridyl; HTMP,2,2,6,6-tetramethylpiperidine.

Yamashita et al. PNAS | September 29, 2015 | vol. 112 | no. 39 | 12027

CHEM

ISTR

Y

Dow

nloa

ded

by g

uest

on

May

30,

202

0

Additional exploration of the reaction scope revealed toleranceof 3-boryl substitution of the aryl bromide as well as trifluoromethyland cyano substituents. Although 2-bromophenylpinacol boro-nate proved a reluctant partner, the cross-coupled product wasformed in 25% yield with the sterically demanding ortho-pina-colboronate substituent. A variety of substituted potassium benzyl-trifluoroborates also proved competent partners (4f–4i). Product 4fis particularly striking, because the reaction manifold was shown tobe selective for cross-coupling of aryl bromides in preference toaryl chlorides, offering yet another level of orthogonality and di-versification. In general, functional group tolerance was found to becomparable with that reported for the related coupling with non-borylated aryl bromides (22) (Fig. 3).In an effort to showcase the potential that this orthogonal

cross-coupling paradigm has in the rapid diversification of simplebuilding blocks, a protocol was developed for the modular func-tionalization of bromochloroborylarene 6. In one pathway, photo-redox/Ni sp3-sp2 cross-coupling was followed directly by Pd-catalyzedSuzuki coupling of the arylBPin with 3-bromopyridine withoutintermediate purification. The product thus generated was sub-jected to Buchwald–Hartwig amination with morpholine to afford 9in 53% overall yield in a sequence involving three bond-formingprocesses and only two purification steps without need for non-productive functional group manipulations (Fig. 4).In a second transformation, the arylBPin intermediate was

diverted into a Rh(I)-catalyzed conjugate addition with methylvinyl ketone to afford 10 in 70% yield. Subsequent Pd-catalyzedSuzuki coupling proceeded in good yield, generating 11 in 60%overall yield. These representative examples attest to the di-versification potential of this platform, because it is easy to en-vision library development by simple alteration of the modularbuilding blocks used in each step or using any of the myriadmethods for functionalization of arylboronates and/or aryl ha-lides through transition metal-catalyzed protocols. Furthermore,nontransition metal-catalyzed transformations of boronic acidsand esters could be used for scaffold elaboration, includingMatteson homologation, diazo insertion, amination, and halo-genation (25–28). Indeed, the rich chemistry of boronic acids andrelated derivatives allows them to serve effectively as a “universalfunctional group” from which nearly any structural element canbe readily accessed.Although the benzylic cross-coupling reported herein is meant

primarily as a proof of concept that the single-electron trans-metalation manifold can be exploited for orthogonal cross-coupling,it is clear that the impact of this concept is dependent on the di-versity of sp3-hybridized organotrifluoroborates that can participatein the photoredox/Ni dual-catalytic C-C bond-forming process.

Our laboratory is actively expanding the palette of competentpartners, the results of which will be disclosed in due course. It isreasonable to suggest that orthogonal cross-coupling should beachievable with each new class of substrates, because the successof the platform is not dependent on the nature of substrates butrather the unique mechanistic features of the single-electrontransmetalation manifold, features that will be shared by allnewly reported methods. In an effort to support this perceivedgenerality, secondary α-alkoxyalkyltrifluoroborate 12 was engagedin orthogonal cross-coupling with 2a under the conditions opti-mized for use with primary benzylic trifluoroborates. Thus, sp3-sp2

Fig. 3. Modular functionalization of lynchpin 6 through photoredox/Ni dual-catalytic orthogonal cross-coupling. Reaction conditions: (A) shown in Fig. 2;(B) 3-bromopyridine (1.2 eq), Pd(PPh3)4 (5 mol %), K3PO4 (3 eq), and THF:H2O (5:1) at 80 °C for 20 h; (C ) morpholine (1.5 eq), Pd2(dba)3 (10 mol %), RuPhos(10 mol %), NaOt-Bu (3 eq), and dioxane at 100 °C for 14 h; (D) methyl vinyl ketone (1.2 eq), [Rh(OH)(COD)]2 (5 mol %), and THF:H2O (5:1) at 60 °C for 20 h;and (E) potassium 3-(trifluoroborato)-2,6-dimethoxypyridine (1.2 eq), Pd(OAc)2 (10 mol %), XPhos (20 mol %), K3PO4 (3 eq), and dioxane:H2O (4:1) at 90 °C for 22 h.dba, dibenzylideneacetone; RuPhos, 2-Dicyclohexylphosphino-2′,6′-diisopropoxybiphenyl; XPhos, 2-Dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl.

B

A

Fig. 4. Photoredox orthogonal cross-coupling of secondary alkyltrifluorobo-rates. (A) Cross-coupling of 1-benzyloxy-3-phenylpropyltrifluoroborate with 13.(B) Cross-coupling of various unactivated secondary alkyltrifluoroborates.

12028 | www.pnas.org/cgi/doi/10.1073/pnas.1509715112 Yamashita et al.

Dow

nloa

ded

by g

uest

on

May

30,

202

0

cross-coupling product 14 was afforded in 54% yield under theseunoptimized conditions, with no detectable cross-coupling oc-curring at the trivalent sp2 organoboron site. Furthermore,various unactivated secondary alkyltrifluoroborates were shownto cross-couple with 2a under conditions previously developed inour laboratory (23). These examples effectively validate photoredoxorthogonal cross-coupling as a general manifold for achievingsp3-sp2 C-C bond formation in the presence of sp2-hybridizedorganoborons that can be directly used in subsequent conven-tional cross-coupling reactions.The results reported herein assume even greater significance

in light of Burke and coworkers’ (29) recent report of automatediterative cross-coupling of brominated MIDA boronates. Here, avast array of natural products, pharmaceutical agents, and drug-like compounds were synthesized through cross-coupling pro-cedures strictly limited to sp2- and primary sp3-hybridized orga-noborons. Despite the tremendous power shown in this report,this limitation handicaps the potential impact of the technology,because no general protocol exists for the Suzuki cross-couplingof secondary alkylboron compounds (30, 31). Thus, althoughsimple hydrocarbon substructures often provide satisfactory yields,organoboron compounds displaying α-branching [e.g., 17b; themethods of Biscoe (31) and Molander (30) for Suzuki cross-cou-pling of secondary alkylboron reagents generate isomerized sideproducts with branched substrates] and a variety of functionalgroups and/or heteroatoms (e.g., 17c and 17d) cannot be cross-coupled using conventional technologies. [Suzuki cross-couplingof the piperidine- and tetrahydropyran-based alkylboron com-pounds leading to 18c and 18d has never been reported.] To ourknowledge, photoredox/Ni dual-catalytic cross-coupling offers forthe first time a general family of protocols for use with this subclassof reagents. This strategy, combined with the ability for orthogonalcross-coupling in the presence of a variety of sp2-hybridized boronic

acid derivatives, including MIDA boronates, presents an opportu-nity for greatly expanding the capabilities of this potentially im-portant technology. The ability of the orthogonal cross-couplings tobe performed without intermediate purification bodes favorablyfor application in an automated fashion, and the compatibility ofphotoredox chemistry with continuous flow reactors has beenpreviously established (32).In conclusion, we have delineated a conceptually unique strategy

for orthogonal cross-coupling that makes use of a mechanisticallydistinct manifold for the activation of sp3-hybridized reagents fortransmetalation. The described mechanistic differentiation of orga-noboron sites represents a fundamental departure from conventionalprotocols for orthogonal cross-coupling, which most often rely oninefficient, differential protection strategies. This reaction plat-form provides unprecedented opportunities for rapid diversifica-tion of polyfunctional building blocks in a straightforward mannerand without intermediate purification, deprotection, or functionalgroup manipulation procedures. Evidence has been providedwith regard to the broader utility of this protocol for the cross-coupling of a variety of sp3-hybridized organotrifluoroborates, andpotential applications in automated, iterative small-molecule pro-duction have been discussed. We anticipate that the proof of con-cept provided herein will serve as a powerful tool to practitioners inthe field and that future developments and advances will greatlyexpand the power of the described methods (SI Appendix).

ACKNOWLEDGMENTS. Dr. Rakesh Kohli is acknowledged for collection ofmass spectrometric data used for characterization purposes. David Primer isacknowledged for helpful discussions. Frontier Scientific is acknowledged fordonation of the organoboron compounds used in this study. Sigma-Aldrichand Johnson-Matthey are thanked for donations of IrCl3. Funding forthis research was provided by the National Institute of General Medical Sci-ences Grant R0I-GM113878.

1. Nicolaou KC, Bulger PG, Sarlah D (2005) Palladium-catalyzed cross-coupling reactionsin total synthesis. Angew Chem Int Ed Engl 44(29):4442–4489.

2. Hooper JF, Young RD, Pernik I, Weller AS, Willis MC (2013) Carbon-carbon bondconstruction using boronic acids and aryl methyl sulfides: Orthogonal reactivity inSuzuki-type couplings. Chem Sci 4:1568–1572.

3. Maingot L, et al. (2012) Regioselective syntheses of 2,7-(het)arylpyrido[2,3-d]pyrimi-dines by an orthogonal cross-coupling strategy. Synlett 23:2449–2452.

4. Hadei N, Achonduh GT, Valente C, O’Brien CJ, Organ MG (2011) Differentiating C-Brand C-Cl bond activation by using solvent polarity: Applications to orthogonal alkyl-alkyl Negishi reactions. Angew Chem Int Ed Engl 50(17):3896–3899.

5. Tobisu M, Chatani N (2009) Devising boron reagents for orthogonal functionalizationthrough Suzuki-Miyaura cross-coupling. Angew Chem Int Ed Engl 48(20):3565–3568.

6. Gillis EP, Burke MD (2009) Iterative cross-coupling with MIDA boronates: Towards ageneral strategy for small-molecule synthesis. Aldrichimica Acta 1:17–27.

7. Sun C, Potter B, Morken JP (2014) A catalytic enantiotopic-group-selective Suzukireaction for the construction of chiral organoboronates. J Am Chem Soc 136(18):6534–6537.

8. Miller SP, Morgan JB, Nepveux FJ, 5th, Morken JP (2004) Catalytic asymmetric car-bohydroxylation of alkenes by a tandem diboration/Suzuki cross-coupling/oxidationreaction. Org Lett 6(1):131–133.

9. Mlynarski SN, Schuster CH, Morken JP (2014) Asymmetric synthesis from terminalalkenes by cascades of diboration and cross-coupling. Nature 505:386–390.

10. Le H, Kyne RE, Brozek LA, Morken JP (2013) Catalytic enantioselective allyl-allyl cross-coupling with a borylated allylboronate. Org Lett 15(7):1432–1435.

11. Gillis EP, Burke MD (2007) A simple and modular strategy for small molecule synthesis:Iterative Suzuki-Miyaura coupling of B-protected haloboronic acid building blocks.J Am Chem Soc 129(21):6716–6717.

12. Lee SJ, Gray KC, Paek JS, Burke MD (2008) Simple, efficient, and modular syntheses ofpolyene natural products via iterative cross-coupling. J Am Chem Soc 130(2):466–468.

13. Woerly EM, Roy J, Burke MD (2014) Synthesis of most polyene natural product motifsusing just 12 building blocks and one coupling reaction. Nat Chem 6(6):484–491.

14. Iwadate N, Suginome M (2010) Differentially protected diboron for regioselectivediboration of alkynes: Internal-selective cross-coupling of 1-alkene-1,2-diboronic acidderivatives. J Am Chem Soc 132(8):2548–2549.

15. Iwadate N, Suginome M (2010) Rhodium-catalyzed dehydroborylation of styreneswith naphthalene-1,8-diaminatoborane [(dan]BH]: New synthesis of masked beta-borylstyrenes as new phenylene-vinylene cross-coupling modules. Chem Lett 39:558–560.

16. Noguchi H, Shioda T, Chou CM, Suginome M (2008) Differentially protected benze-nediboronic acids: Divalent cross-coupling modules for the efficient synthesis of boron-substituted oligoarenes. Org Lett 10(3):377–380.

17. Noguchi H, Hojo K, Suginome M (2007) Boron-masking strategy for the selective

synthesis of oligoarenes via iterative Suzuki-Miyaura coupling. J Am Chem Soc 129(4):

758–759.18. Lee JCH, McDonald R, Hall DG (2011) Enantioselective preparation and chemo-

selective cross-coupling of 1,1-diboron compounds. Nat Chem 3(11):894–899.19. Molander GA, Sandrock DL (2008) Orthogonal reactivity in boryl-substituted orga-

notrifluoroborates. J Am Chem Soc 130(47):15792–15793.20. Pelz NF, Morken JP (2006) Modular asymmetric synthesis of 1,2-diols by single-pot

allene diboration/hydroboration/cross-coupling. Org Lett 8(20):4557–4559.21. Newhouse T, Baran PS, Hoffmann RW (2009) The economies of synthesis. Chem Soc

Rev 38(11):3010–3021.22. Tellis JC, Primer DN, Molander GA (2014) Dual catalysis. Single-electron transmetalation

in organoboron cross-coupling by photoredox/nickel dual catalysis. Science 345(6195):

433–436.23. Primer DN, Karakaya I, Tellis JC, Molander GA (2015) Single-electron transmetalation:

An enabling technology for secondary alkylboron cross-coupling. J Am Chem Soc

137(6):2195–2198.24. Gutierrez O, Tellis JC, Primer DN, Molander GA, Kozlowski MC (2015) Nickel-catalyzed

cross-coupling of photoredox-generated radicals: Uncovering a general manifold

for stereoconvergence in nickel-catalyzed cross-couplings. J Am Chem Soc 137(15):

4896–4899.25. Hall DG, ed (2005) Boronic Acids (Wiley-VCH, Weinheim, Germany).26. Coeffard V, Moreau X, Thomassigny C, Greck C (2013) Transition-metal-free amination

of aryl boronic acids and their derivatives. Angew Chem Int Ed Engl 52(22):5684–5686.27. Argintaru OA, Ryu D, Aron I, Molander GA (2013) Synthesis and applications of α-tri-

fluoromethylated alkylboron compounds. Angew Chem Int Ed Engl 52(51):13656–13660.28. Molander GA, Ryu D (2014) Diastereoselective synthesis of vicinally bis(trifluoromethylated)

alkylboron compounds through successive insertions of 2,2,2-trifluorodiazoethane.

Angew Chem Int Ed Engl 53(51):14181–14185.29. Li J, et al. (2015) Synthesis of many different types of organic small molecules using

one automated process. Science 347:1221–1226.30. Dreher SD, Dormer PG, Sandrock DL, Molander GA (2008) Efficient cross-coupling of

secondary alkyltrifluoroborates with aryl chlorides–reaction discovery using parallel

microscale experimentation. J Am Chem Soc 130(29):9257–9259.31. Li L, Zhao S, Joshi-Pangu A, Diane M, Biscoe MR (2014) Stereospecific Pd-catalyzed

cross-coupling reactions of secondary alkylboron nucleophiles and aryl chlorides. J Am

Chem Soc 136(40):14027–14030.32. Tucker JW, Zhang Y, Jamison TF, Stephenson CRJ (2012) Visible-light photoredox

catalysis in flow. Angew Chem Int Ed Engl 51(17):4144–4147.

Yamashita et al. PNAS | September 29, 2015 | vol. 112 | no. 39 | 12029

CHEM

ISTR

Y

Dow

nloa

ded

by g

uest

on

May

30,

202

0