View source for Robert Kottwitz

{{multiple issues|
{{very long|date=December 2013}}
{{copy edit|for=style|date=January 2014}}
}}

'''Photoredox catalysis''' is a branch of [[catalysis]] that harnesses the energy of [[visible light]] to accelerate a [[chemical reaction]] via a [[electron transfer|single-electron transfer]].<ref>{{cite journal|last=Tucker|first=Joseph W.|coauthors=Stephenson, Corey R. J.|title=Shining Light on Photoredox Catalysis: Theory and Synthetic Applications|journal=The Journal of Organic Chemistry|date=17 February 2012|volume=77|issue=4|pages=1617–1622|doi=10.1021/jo202538x|pmid=22283525}}</ref><ref>{{cite journal|last=Prier|first=Christopher K.|coauthors=Rankic, Danica A.; MacMillan, David W. C.|title=Visible Light Photoredox Catalysis with Transition Metal Complexes: Applications in Organic Synthesis|journal=Chemical Reviews|date=10 July 2013|volume=113|issue=7|pages=5322–5363|doi=10.1021/cr300503r|pmid=23509883}}</ref><ref>{{cite journal|last=Yoon|first=Tehshik P.|coauthors=Ischay, Michael A.; Du, Juana|title=Visible light photocatalysis as a greener approach to photochemical synthesis|journal=Nature Chemistry|date=23 June 2010|volume=2|issue=7|pages=527–532|doi=10.1038/NCHEM.687|pmid=20571569}}</ref><ref>{{cite journal|last=Xuan|first=Jun|coauthors=Xiao, Wen-Jing|title=Visible-Light Photoredox Catalysis|journal=Angewandte Chemie International Edition|date=9 July 2012|volume=51|issue=28|pages=6828–6838|doi=10.1002/anie.201200223}}</ref><ref>{{cite journal|last=Fagnoni|first=Maurizio|coauthors=Dondi, Daniele; Ravelli, Davide; Albini, Angelo|title=Photocatalysis for the Formation of the C−C Bond|journal=Chemical Reviews|date=June 2007|volume=107|issue=6|pages=2725–2756|doi=10.1021/cr068352x|pmid=17530909}}</ref>   This area is named as a combination of “photo-” referring to light and [[redox]], a condensed expression for the chemical processes of reduction and oxidation.  In particular, photoredox catalysis employs small quantities of a light-sensitive compound that, when excited by light, can mediate the transfer of [[electrons]] between chemical compounds that otherwise would not react.  Photoredox catalysts are generally drawn from three classes of materials: transition-metal complexes, organic dyes, and semiconductors.  While each class of materials has advantages, transition-metal complexes are used most often because, unlike semiconductors, they are homogeneous catalysts and because they tend to have stronger redox capabilities than most organic dyes.

[[File:Ru(bipy) Schematic.png|center|frameless|Schematic diagram of Ru(bipy)<sub>3</sub><sup>2+</sup>, a typical photoredox catalyst]]

Study of this branch of catalysis has led to the development of many new methods to accomplish important, but well known, chemical transformations, and additionally has led to the development of new chemical transformations.  Some advantages that photoredox catalysts offer in well-known transformations compared to traditional reagents is that photoredox catalysts are often less toxic than other reagents often used to generate [[free radicals]], such as organotin reagents.  Furthermore, while photoredox catalysts are potent redox agents while directly exposed to light, they are also very stable to ambient conditions and do not change oxidation state.  This can make transition-metal complex photoredox catalysts much easier to work with than stoichiometric redox agents such as quinones.  Finally, the properties of photoredox catalysts can be easily modified by changing the type of metal used in the catalyst or by changing one or more of the ligands bound to the metal center.  The somewhat modular nature of the catalyst allows straightforward synthesis of a range of analogs and makes it easy to "tune" the catalysts to the needs of a particular chemical system.

While photoredox catalysis has most often been applied to generate known reactive intermediates in a novel way, the study of this mode of catalysis has additionally led to the discovery of new organic reactions, such as the first direct functionalization of the β-arylation of saturated aldehydes.  Although the C<sub>3</sub>-symmetric transition-metal complexes used in many photoredox-catalyzed reactions are chiral, the use of enantioenriched photoredox catalysts led to low levels of enantioselectivity in a photoredox-catalyzed aryl-aryl coupling reaction studied by Ohkubo et al., suggesting that the chiral nature of these catalysts is not yet a highly effective means of transmitting stereochemical information in photoredox reactions.<ref>{{cite journal|last=Hamada|first=Taisuke|coauthors=Ishida, Hitoshi; Usui, Satoshi; Watanabe, Yoshiro; Tsumura, Kazunori; Ohkubo, Katsutoshi|title=A novel photocatalytic asymmetric synthesis of (R)-(+)-1,1?-bi-2-naphthol derivatives by oxidative coupling of 3-substituted-2-naphthol with ?-[Ru(menbpy)3]2+[menbpy = 4,4?-di(1R,2S,5R)-(?)-menthoxycarbonyl-2,2?-bipyridine], which posseses molecular helicity|journal=Journal of the Chemical Society, Chemical Communications|year=1993|issue=11|pages=909|doi=10.1039/C39930000909}}</ref>   However, while synthetically useful levels of enantioselectivity have not been achieved using chiral photoredox catalysts alone, optically-active products have been obtained through the synergistic combination of photoredox catalysis with chiral organocatalysts such as secondary amines and Brønsted acids.<ref>{{cite journal|last=Rono|first=Lydia J.|coauthors=Yayla, Hatice G.; Wang, David Y.; Armstrong, Michael F.; Knowles, Robert R.|title=Enantioselective Photoredox Catalysis Enabled by Proton-Coupled Electron Transfer: Development of an Asymmetric Aza-Pinacol Cyclization|journal=Journal of the American Chemical Society|date=27 November 2013|volume=135|issue=47|pages=17735–17738|doi=10.1021/ja4100595|pmid=24215561}}</ref>

== Photochemistry of transition metal photocatalysts ==

The activity of a photoredox catalyst can be described in three steps.  First, a molecule of the catalyst in its [[ground state]], where the electrons are distributed among the lowest-energy combination of states, interacts with light and moves into a long-lived [[excited state]], where the electrons are not distributed among the lowest-energy combination of available states.  Second, the photoexcited catalyst interacts by an [[outer-sphere electron transfer]] process to “quench” the excited state and to activate one of the other components of the chemical reaction.  Finally, a second single electron transfer occurs to return the catalyst to its original oxidation state and electron arrangement.

The first step of this process, [[photoexcitation]], is initiated by absorption of a [[photon]], promoting an electron from the highest occupied molecular orbital (HOMO) of the photocatalyst to another spin-allowed state.  Thus, since the ground state of the photocatalyst is a singlet state (a state with no total [[electron spin]]), the photon absorption excites the catalyst to another singlet state of higher energy.  This excitation is realized as a [[metal-to-ligand charge transfer]], where the electron moves from an orbital centered on the metal (e.g. a d orbital) to an orbital localized on the ligands (e.g. the π* orbital of an aromatic ligand).  While absorption of a photon can occur for any energy corresponding to the difference between two singlet states, the excited electronic state quickly relaxes to the lowest energy singlet excited state through [[Internal conversion (chemistry)|internal conversion]], a process where energy is dissipated as vibrational energy rather than as electromagnetic radiation.  This singlet excited state can relax further by two distinct processes: the catalyst may [[fluorescence|fluoresce]], radiating a photon and returning to the singlet ground state, or it can move to the lowest energy triplet excited state (a state where two unpaired electrons have the same spin) by a second non-radiative process termed [[intersystem crossing]].   Direct relaxation of the excited triplet to the ground state, termed [[phosphorescence]], requires both emission of a photon and inversion of the spin of the excited electron.  The spin-forbidden nature of this pathway means that it is a slow process and therefore that the triplet excited state has a substantial average lifetime.  For the common photocatalyst [[tris(bipyridine)ruthenium(II) chloride|tris-(2,2’-bipyridyl)ruthenium chloride]] (also known as Ru(bipy)<sub>3</sub><sup>2+</sup>), the lifetime of the triplet excited state has been measured to be approximately 1100 ns.  This span of time is long enough that other relaxation pathways, specifically electron-transfer pathways, can occur more rapidly than decay of the catalyst to its ground state.

[[File:Jablonski Diagram.png|600px|center|Jablonski diagram illustrating the electronic states accessible during photoexcitation.  Note: ISC stands for Intersystem Crossing.  E<sub>0,0</sub> is a measurement of the energy gap between the ground state and the lowest energy triplet state.  This parameter is proportional to the phosphorescence wavelength and is used to compute the redox potentials of the triplet state.]]

The long-lived triplet excited state accessible by photoexcitation is chemically interesting because it is both a more potent [[reducing agent]] and a more potent [[oxidizing agent]] than the ground state of the catalyst molecule.  In other words, the catalyst can more readily give up one its electrons or accept an electron from an external source.  The catalyst excited state is a stronger reductant because its highest energy electron has been excited to an even higher energy state through the photoexcitation process.  Similarly, the excited catalyst is a stronger oxidant because one of the catalyst’s lowest energy orbitals, which is fully occupied in the ground state, is only singly occupied after photoexcitation and is therefore available for an external electron to occupy.  Since organometallic photocatalysts consist of a coordinatively saturated metal complex, i.e. a structure that cannot form any additional bonds, electron transfer cannot take place by an [[inner sphere electron transfer|inner sphere]] mechanism through a direct bond of the metal complex to another reagent.  Instead, electron transfer must take place via an [[outer sphere electron transfer|outer sphere]] process, where the electron [[quantum tunnelling|tunnels]] between the catalyst and another molecule.

The [[Marcus theory|theory of outer sphere electron transfer]], developed by [[Rudolph Marcus]], predicts that such a tunneling process will occur most quickly in systems where not only is the electron transfer thermodynamically favorable (i.e. between strong reductants and oxidants) but also where the electron transfer has a low intrinsic barrier.  The intrinsic barrier of electron transfer derives from the [[Franck-Condon principle]], stating that electronic transition takes place more quickly if there is greater overlap between the initial and final electronic states.  Interpreted loosely, this principle suggests that the barrier of an electronic transition is related to the degree to which the system seeks to reorganize.  For an electronic transition with a system, the barrier is therefore related to the "overlap" between the initial and final wavefunctions of the electron being excited - i.e. the degree to which the electron needs to "move" in the transition.  In an intermolecular electron transfer, a similar role is played by the degree to which the nuclei seek to move in response to the change in their new electronic environment.  Immediately after electron transfer, the nuclear arrangement of the molecule, previously an equilibrium, now represents a vibrationally excited state and must relax to its new equilibrium geometry.  Rigid systems, whose geometry is not greatly dependent on oxidation state, will therefore experience less vibrational excitation during electron transfer, and will have a lower intrinsic barrier.  Photocatalysts such as Ru(bipy)3, are held in a rigid arrangement by flat, bidentate ligands arranged in an octahedral geometry around the metal center.  Therefore, the complex will not undergo much reorganization during an electron transfer and the process is therefore likely to be fast.  Since electron transfer of these complexes is fast, it is very likely to take place within the duration of the catalyst’s active state, i.e. during the lifetime of the triplet excited state.

[[File:Photocatalyst Cycle.png|200px|center]]

The final step in the photocatalytic cycle is the regeneration of the photocatalyst in its ground state.  At this stage, the catalyst exists as the ground state of either its oxidized or reduced forms, depending on whether it acted received or gave up an electron from its photoexcited state.  These oxidation states of the catalyst have a strong driving force to return to their equilibrium oxidation state and will act as a potent single-electron reductant or oxidant in order to satisfy that driving force.  In order to regenerate the original ground state, the catalyst must participate in a second outer-sphere electron transfer.  In many cases, this electron transfer takes place with a stoichiometric two-electron reductant or oxidant, although in some recent cases this step has also been implemented to activate a second reagent.  The case where the excited state catalyst first is reduced a reaction component and then is oxidized to return to its resting state is known as a reductive quenching cycle.  Conversely the case where the excited state catalyst first is oxidized and then reduced to return to its resting state is known as an oxidative quenching cycle.  These two possible cycles can be distinguished by a [[Stern-Volmer relationship|Stern-Volner experiment]].  Since the electron transfer step of the catalytic cycle takes place from the triplet excited state, it competes with phosphorescence as a relaxation pathway.  The Stern-Volner experiment measures the intensity of phosphorescence while varying the concentration of each possible quenching agent.  When the concentration of the actual quenching agent is varied, the rate of electron transfer therefore the degree of phosphorescence is affected.  This relationship is modeled by the equation:

<math>\left ( \frac{I_0}{I} \right ) = {k_q} * {\tau_0} \times [ Q ]</math>

Here, I<sub>0</sub> and I denote the emission intensity with and without quenching agent present, k<sub>q</sub> the rate constant of the quenching process, τ<sub>0</sub> the excited-state lifetime in the absence of quenching agent, and [Q] the concentration of quenching agent.  Thus, if the excited-state lifetime of the photoredox catalyst is known from other experiments, the rate constant of quenching in the presence of a single reaction component can be determined by measuring the change in emission intensity as the concentration of quenching agent changes.

== Photophysical properties of common photoredox catalysts ==

In the study of photoredox catalysis, it is essential to choose a catalyst for which the reduction and oxidation potentials are matched to the other components of the reaction.  While ground state redox potentials are easily measured by cyclic voltammetry or other electrochemical methods, measuring the redox potential of an electronically excited state cannot be accomplished directly by these methods.<ref>{{cite journal|last=Jones|first=Wayne E.|coauthors=Fox, Marye Anne|title=Determination of Excited-State Redox Potentials by Phase-Modulated Voltammetry|journal=The Journal of Physical Chemistry|date=May 1994|volume=98|issue=19|pages=5095–5099|doi=10.1021/j100070a025}}</ref>  However, two methods exist that allow estimation of the excited-state redox potentials and one method exists for the direct measurement of these potentials.  To estimate the excited-state redox potentials, one method is to compare the rates of electron transfer from the excited state to a series of ground-state reactants whose redox potentials are known.  A more common method to estimate these potentials is to use an equation developed by Rehm and Weller that describes the excited-state potentials as a correction of the ground-state potentials:

<math>{{E^*}_{1/2}}^{red} = {E_{1/2}}^{red} + E_{0,0} + w_r </math>

<math>{{E^*}_{1/2}}^{ox} = {E_{1/2}}^{ox} - E_{0,0} + w_r </math>

In these formulas, E*<sub>1/2</sub> represents the reduction or oxidation potential of the excited state, E<sub>1/2</sub> represents the reduction or oxidation potential of the ground state, E<sub>0,0</sub> represents the difference in energy between the zeroth vibrational states of the ground and excited states, and w<sub>r</sub> represents the [[work function]], an electrostatic interaction that arises due to the separation of charges that occurs during electron-transfer between two chemical species.  The zero-zero excitation energy, E<sub>0,0</sub> is usually approximated by the corresponding transition in the fluorescence spectrum.  This method allows calculation of approximate excited-state redox potentials from more easily measured ground-state redox potentials and spectroscopic data.

Direct measurement of the excited-state redox potentials is possible by applying a method known as phase-modulated voltammetry.  This method works by shining light onto an electrochemical cell in order to generate the desired excited-state species, but to modulate the intensity of the light sinusoidally, so that the concentration of the excited-state species is not constant.  In fact, the concentration of excited-state species in the cell should change exactly in phase with the intensity of light incident on the electrochemical cell.  If the potential applied to the cell is strong enough for electron transfer to occur, the change in concentration of the redox-competent excited state can be measured as an alternating current (AC).  Furthermore, the phase shift of the AC current relative to the intensity of the incident light corresponds to the average lifetime of an excited-state species before it engages in electron transfer.

Note on redox potentials: In the following table, all redox potentials are quoted as reduction potentials measured in comparison to the [[saturated calomel electrode]] (SCE).  Thus, the oxidation potential of a chemical species is listed as the reduction potential of its oxidized form.  More positive numbers indicate that the catalyst is reduced more readily, i.e. that it is a more strongly oxidizing species.  Negative numbers indicate that the electrochemical reaction does not occur spontaneously in the direction written.  In such a case, a voltage of the written magnitude must be applied to the catalyst in order to induce reduction.  Thus, more negative reduction potentials indicate that the chemical species is more easily oxidized, and is a more strongly reducing species.

{| class="wikitable sortable"
|-
! Photocatalyst !! Structure !! E<sub>1/2</sub>(C<sup>+</sup>/C) (V vs SCE) !! E<sub>1/2</sub>(C/C<sup>-</sup>) (V vs SCE) !! E<sub>1/2</sub>(C<sup>+</sup>/C*) (V vs SCE) !! E<sub>1/2</sub>(C*/C<sup>-</sup>) (V vs SCE) !! Excited-State Lifetime (ns) !! Peak Excitation Wavelength (nm) !! Peak Emission Wavelength (nm) !! Reference
|-
| tris-(2,2'-bipyrimidine)ruthenium<sup>2+</sup>  (Ru(bpm)<sub>3</sub><sup>2+</sup>) || [[File:Ru(bpm) Schematic.png|thumb|Schematic of tris-(2,2'-bipyrimidyl)ruthenium(II)]] || 1.69 || -0.91 || -0.21 || 0.99 || 131 || 454 || 639 ||<ref>{{cite journal|last=Rillema|first=D. Paul|coauthors=Allen, G.; Meyer, T. J.; Conrad, D.|title=Redox properties of ruthenium(II) tris chelate complexes containing the ligands 2,2'-bipyrazine, 2,2'-bipyridine, and 2,2'-bipyrimidine|journal=Inorganic Chemistry|year=1983|volume=22|issue=11|pages=1617–1622|doi=10.1021/ic00153a012}}</ref>
|-
| tris-(2,2'-bipyrazine)ruthenium<sup>2+</sup>  (Ru(bpz)<sub>3</sub><sup>2+</sup>) || [[File:Ru(bpz) Schematic.png|thumb|Schematic of tris-(2,2-bipyrazyl)ruthenium(II)]] || 1.86 || -0.80 || -0.26 || 1.45 || 740 || 443 || 591 ||<ref>{{cite journal|last=Crutchley|first=R. J.|coauthors=Lever, A. B. P.|title=Ruthenium(II) tris(bipyrazyl) dication - a new photocatalyst|journal=Journal of the American Chemical Society|date=September 1980|volume=102|issue=23|pages=7128–7129|doi=10.1021/ja00543a053}}</ref>
|-
| tris-(2,2'-bipyridine)ruthenium<sup>2+</sup>  (Ru(bipy)<sub>3</sub><sup>2+</sup>) || [[File:Ru(bipy) Schematic.png|thumb|Schematic diagram of Ru(bipy)<sub>3</sub><sup>2+</sup>, a typical photoredox catalyst]] || 1.29 || -1.33 || -0.81 || 0.77 || 1100 || 452 || 615 ||<ref>{{cite journal|last=Juris|first=Alberto|coauthors=Balzani, Vincenzo; Belser, Peter; von Zelewsky, Alex|title=Characterization of the Excited State Properties of Some New Photosensitizers of the Ruthenium (Polypyridine) Family|journal=Helvetica Chimica Acta|date=4 November 1981|volume=64|issue=7|pages=2175–2182|doi=10.1002/hlca.19810640723}}</ref>
|-
| tris-(1,10-phenanthroline)ruthenium<sup>2+</sup>  (Ru(phen)<sub>3</sub><sup>2+</sup>) || [[File:Ru(phen) Schematic.png|thumb|Schematic of tris-(1,10-phenanthroline)ruthenium(II)]] || 1.26 || -1.36 || -0.87 || 0.82 || 500 || 422 || 610 ||<ref>{{cite journal|last=Young|first=Roger C.|coauthors=Meyer, Thomas J.; Whitten, David G.|title=Electron transfer quenching of excited states of metal complexes|journal=Journal of the American Chemical Society|date=January 1976|volume=98|issue=1|pages=286–287|doi=10.1021/ja00417a073}}</ref>
|-
| bis-(2-(2',4'-difluorophenyl)-5-trifluoromethylpyridine)(di-tert-butylbipyridine)iridium<sup>+</sup>  (Ir(dF(CF<sub>3</sub>)ppy)<sub>2</sub>(dtbbpy)<sup>+<sup>) || [[File:Ir(dfptfmpy)(dtbbpy) Schematic.png|thumb|Schematic of bis-(2-(2',4'-difluorophenyl)-5-trifluoromethylpyridine)(di-tert-butylbipyridine)iridium(III)]] || 1.69 || -1.37 || -0.89 || 1.21 || 2300 || 380 || 470 ||<ref name="Lowry2005">{{cite journal|last=Lowry|first=Michael S.|coauthors=Goldsmith, Jonas I.; Slinker, Jason D.; Rohl, Richard; Pascal, Robert A.; Malliaras, George G.; Bernhard, Stefan|title=Single-Layer Electroluminescent Devices and Photoinduced Hydrogen Production from an Ionic Iridium(III) Complex|journal=Chemistry of Materials|date=November 2005|volume=17|issue=23|pages=5712–5719|doi=10.1021/cm051312}}</ref>
|-
| bis-(2-phenylpyridine)(di-tert-butylbipyridine)iridium<sup>+</sup>  (Ir(ppy)<sub>2</sub>(dtbbpy)<sup>+</sup>) || [[File:Ir(ppy)(dtbbpy) Schematic.png|thumb|bis-(2-phenylpyridine)(di-tert-butylbipyridine)iridium(III)]] || 1.21 || -1.51 || -0.96 || 0.66 || 557 || || 581 ||<ref name="Lowry2005" />
|-
| fac-(tris-(2,2'-phenylpyridine))iridium  (Ir(ppy)<sub>3</sub><sup>+</sup>) || [[File:Ir(ppy) Schematic.png|thumb|Schematic of fac-(tris-(2,2'-phenylpyridine))iridium(III)]] || 0.77 || -2.19 || -1.73 || 0.31 || 1900 || 375 || 494 ||<ref>{{cite journal|last=Flamigni|first=Lucia|coauthors=Barbieri, Andrea; Sabatini, Cristiana; Ventura, Barbara; Barigelletti, Francesco|title=Photochemistry and Photophysics of Coordination Compounds: Iridium|journal=Topics in Current Chemistry|year=2007|pages=143–203|doi=10.1007/128_2007_131|series=Topics in Current Chemistry|isbn=978-3-540-73348-5|volume=281}}</ref>
|}

The relative reducing and oxidizing natures of these photocatalysts can be understood intuitively by considering the electronegativity of the ligands and the metal center of the catalyst complex. More electronegative metals and ligands will tend to stabilize electrons better than their less electronegative counterparts.  Therefore, complexes with more electronegative complexes will be more easily reduced (i.e. be more powerfully oxidizing) than more electropositive complexes.  For example, the ligands 2,2'-bipyridine and 2,2'-phenylpyridine are isoelectronic structures, containing the same number and arrangement of electrons.  However, phenylpyridine replaces one of the nitrogen atoms in bipyridine with a carbon atom.  Carbon is less electronegative than nitrogen is, so it holds electrons less tightly than nitrogen.  Since the remainder of the ligand molecule is identical, this effect is transferred to the structure as a whole: phenylpyridine holds is electrons less tightly than bipyridine, i.e. it is more strongly electron-donating and less electronegative as a ligand.  Complexes with phenylpyridine ligands will therefore be more strongly reducing and less strongly oxidizing than complexes with bipyridine ligands because the phenylpyridine complexes are more electron-rich and hold their electrons with slightly less strength.  In the same way, a fluorinated derivative of phenylpyridine will be more electronegative than the corresponding simple phenylpyridine and complexes with fluorine-containing ligands will be more strongly oxidizing and less strongly reducing than the unsubstituted phenylpyridine complex.
The electronic influence of the metal center on the complex is somewhat more complex than the effect of the ligands.  According to the Pauling scale of electronegativity, both ruthenium and iridium have an electronegativity of 2.2.  If this was the sole factor relevant to the redox potentials, then complexes of ruthenium and iridium with the same ligands should be equally powerful photoredox catalysts.  However, considering the Rehm-Weller equation, the spectroscopic properties of the metal also play a role in determining the redox character of the excited state.<ref name="Tucker2012">{{cite journal|last=Tucker|first=Joseph W.|coauthors=Stephenson, Corey R. J.|title=Shining Light on Photoredox Catalysis: Theory and Synthetic Applications|journal=The Journal of Organic Chemistry|year=2012|volume=77|issue=4|pages=1617–1622|doi=10.1021/jo202538x|pmid=22283525}}</ref>  In particular, the parameter E<sub>0,0</sub> is related to the emission wavelength of the complex, and therefore to the size of the Stokes shift - the difference in energy between the maximum absorption and emission of a molecule.  Ruthenium complexes typically have large Stokes shifts, and therefore have low energy emission wavelengths and small zero-zero excitation energies, when compared to iridium complexes.  In effect, this means that although ground-state ruthenium complexes can be potent reductants, that the excited-state complex will be a much less potent reductant or oxidant than a comparable iridium complex.  This makes iridium photocatalysts more favorable for the development of general organic transformations because the stronger redox potentials of the excited catalyst allows the use of weaker stoichiometric reductants and oxidants or the use of less reactive substrates.<ref name="Tucker2012" />

== Selected applications of photoredox catalysis in organic chemistry ==

=== Reductive dehalogenation ===

Reductive dehalogenation is a convenient method for removing [[halogen]] atoms that are introduced into a molecule by a method such as [[halolactonization]].  However, the standard method for removing halogen atoms requires the use of stoichiometric [[organotin chemistry|organotin]] reagents, such as [[tributyltin hydride]].  While this reaction is very powerful and orthogonal to other functional groups commonly present in organic molecules, organotin reagents are highly toxic and the development of alternative reagents is desirable.  The cleavage of activated and reductively labile functional groups including sulfoniums and halogens is the earliest application of photoredox catalysis to organic synthesis, but early attempts were hampered by the need for specific or uncommon substrates or by the preferential formation of dimeric coupling products.<ref>{{cite journal|last=Hedstrand|first=David M.|coauthors=Kruizinga, Wim H.; Kellogg, Richard M.|title=Light induced and dye accelerated reductions of phenacyl onium salts by 1,4-dihydropyridines|journal=Tetrahedron Letters|date=January 1978|volume=19|issue=14|pages=1255–1258|doi=10.1016/S0040-4039(01)94515-0}}</ref><ref>{{cite journal|last=Willner|first=Itamar|coauthors=Tsfania, Tamar; Eichen, Yoav|title=Photocatalyzed and electrocatalyzed reduction of vicinal dibromides and activated ketones using ruthenium(I) tris(bipyridine) as electron-transfer mediator|journal=The Journal of Organic Chemistry|date=April 1990|volume=55|issue=9|pages=2656–2662|doi=10.1021/jo00296a023}}</ref><ref>{{cite journal|last=Hironaka|first=Katsuhiko|coauthors=Fukuzumi, Shunichi; Tanaka, Toshio|title=Tris(bipyridyl)ruthenium(II)-photosensitized reaction of 1-benzyl-1,4-dihydronicotinamide with benzyl bromide|journal=Journal of the Chemical Society, Perkin Transactions 2|year=1984|issue=10|pages=1705|doi=10.1039/P29840001705}}</ref><ref>{{cite journal|last=Kern|first=Jean-Marc|coauthors=Sauvage, Jean-Pierre|title=Photoassisted C?C coupling via electron transfer to benzylic halides by a bis(di-imine) copper(I) complex|journal=Journal of the Chemical Society, Chemical Communications|year=1987|issue=8|pages=546|doi=10.1039/C39870000546}}</ref><ref>{{cite journal|last=Fukuzumi|first=Shunichi.|coauthors=Mochizuki, Seiji.; Tanaka, Toshio.|title=Photocatalytic reduction of phenacyl halides by 9,10-dihydro-10-methylacridine: control between the reductive and oxidative quenching pathways of tris(bipyridine)ruthenium complex utilizing an acid catalysis|journal=The Journal of Physical Chemistry|date=January 1990|volume=94|issue=2|pages=722–726|doi=10.1021/j100365a039}}</ref>   Recently, the Stephenson lab developed a more general method for photoredox-mediated reductive dehalogenation.<ref>{{cite journal|last=Narayanam|first=Jagan M. R.|coauthors=Joseph W. Tucker, and Corey R. J. Stephenson|title=Electron-Transfer Photoredox Catalysis: Development of a Tin-Free Reductive Dehalogenation Procedure|journal=JACS|date=June 5, 2009|volume=131|issue=25|pages=8756–8757|doi=10.1021/ja9033582}}</ref>   Stevenson's original method employs Ru(bipy)<sub>3</sub><sup>2+</sup> as the photocatalyst and a stoichiometric amine reductant to reduce "activated" carbon-halogen bonds, such as those with an adjacent carbonyl group or arene.  These bonds are considered to be activated because the radical they produce upon fragmentation is stabilized by conjugation with the carbonyl group or arene, respectively.  The stoichiometric reductant present in this reaction transfers an electron to reduce the excited-state catalyst to the Ru(I) oxidation state.  The reduced catalyst can then shuttle the transferred electron to the halogenated substrate, reducing the weak C-X bond and inducing fragmentation.

[[File:Reductive Dehalogenation Figure.png|400 px|frameless|center|Diagram of the catalytic cycle proposed by Stephenson et al. for the mechanism of their tin-free reduction of activated halogens]]

In subsequent work published in 2012, Stephenson accomplished the reduction of unactivated carbon-iodine bonds using the strongly reducing photocatalyst tris-(2,2’-phenylpyridine)iridium (Ir(ppy)<sub>3</sub><sup>2+</sup>).<ref>{{cite journal|last=Nguyen|first=John D.|coauthors=D'Amato, Erica M.; Narayanam, Jagan M. R.; Stephenson, Corey R. J.|title=Engaging unactivated alkyl, alkenyl and aryl iodides in visible-light-mediated free radical reactions|journal=Nature Chemistry|year=2012|volume=4|issue=10|pages=854–859|doi=10.1038/nchem.1452|pmid=23001000}}</ref>   This updated reaction is mechanistically distinct from the previous transformation of activated bromides and chlorides.  In this version of the reaction, fac-Ir(ppy)<sub>3</sub> is used as a photocatalyst.  The increased reduction potential of this catalyst compared to Ru(bipy)<sub>3</sub><sup>2+</sup> allows direct reduction of the carbon-iodine bond without first interacting with a stoichiometric reductant.  Thus, the iridium complex transfers an electron to the substrate, causing fragmentation of the substrate and oxidizing the catalyst to the Ir(IV) oxidation state.  The oxidized photocatalyst is then easily returned to its original oxidation state through interaction with one of the reaction additives.

[[File:Unactivated Reductive Dehalogenation Diagram.png|400px|frameless|center|Mechanistic diagram of the reductive dehalogenation of unactivated carbon-iodine bonds]]

Just as tin-mediated radical dehalogenation reactions can be used to initiate cascade cyclizations to rapidly generate molecular complexity, Stephenson's tin-free reductive dehalogenation allows access to the same types of complex products.<ref>{{cite journal|last=Tucker|first=Joseph W.|coauthors=Nguyen, John D.; Narayanam, Jagan M. R.; Krabbe, Scott W.; Stephenson, Corey R. J.|title=Tin-free radical cyclization reactions initiated by visible light photoredox catalysis|journal=Chemical Communications|date=28 May 2010|volume=46|issue=27|pages=4985–4987|doi=10.1039/c0cc00981d|pmid=20512181}}</ref> In this work, Stephenson et al. presented a radical cascade cyclization that closed two five-membered rings and formed two new stereocenters, proceeding in good yield.
Furthermore, the Stephenson group has made use of their reductive dehalogenation protocol in a key step of their total synthesis of the natural product (+)-Gliocladin C.<ref>{{cite journal|last=Furst|first=Laura|coauthors=Narayanam, Jagan M. R.; Stephenson, Corey R. J.|title=Total Synthesis of (+)-Gliocladin C Enabled by Visible-Light Photoredox Catalysis|journal=Angewandte Chemie International Edition|date=4 October 2011|volume=50|issue=41|pages=9655–9659|doi=10.1002/anie.201103145}}</ref>

[[File:Applications of Reductive Dehalogenation.png|400px|frameless|center|a) Diagram of a radical cascade cyclization initiated by a photoredox-catalyzed reductive dehalogenation reaction.  b) Diagram of a radical fragment coupling initiated by a photoredox-catalyzed reductive dehalogenation reaction.  This reaction was performed as part of the Stephenson synthesis of the natural product (+)-Gliocladin C]]

=== Oxidative generation of iminium ions ===

[[Iminium]] ions are potent [[electrophiles]] useful for generating C-N bonds in complex molecules.  However, the condensation of [[amines]] with [[carbonyl]] compounds to form iminium ions is often an unfavorable process, sometimes requiring harsh dehydration conditions.  For this reason, alternative methods for the generation of iminium ions, particularly by oxidation from the corresponding amine, is a valuable synthetic tool.  In the first report of this type of reactivity, Stephenson et al. achieved the generation of iminium ions from activated amines by the use of Ir(dtbbpy)(ppy)<sub>2</sub>PF<sub>6</sub> as a photoredox catalyst.<ref>{{cite journal|last=Condie|first=Allison G.|coauthors=González-Gómez, José C.; Stephenson, Corey R. J.|title=Visible-Light Photoredox Catalysis: Aza-Henry Reactions via C−H Functionalization|journal=Journal of the American Chemical Society|date=10 February 2010|volume=132|issue=5|pages=1464–1465|doi=10.1021/ja909145y|pmid=20070079}}</ref>  The substrates investigated for this reaction were aryltetrahydroisoquinolines, substrates which would stabilize generation of nitrogen-centered radicals through conjugation with one of the arenes and would promote rapid and regioselective H-atom abstraction to generate the conjugated benzylic imine. Stephenson et al. propose that this transformation occurs by oxidation of the amine to the aminium radical cation by the excited photocatalyst, which is strongly oxidizing due to the electrophilicity of iridium and due to the electron-poor nature of the fluorinated ligands, followed by H-atom transfer to a superstoichimetric oxidant, such as [[nitromethane]] (also functioning in the reaction as a nucleophile and as the solvent) or molecular oxygen.  Finally, the reactive iminium ion formed by the H-atom transfer is quenched by reaction with nitromethane, the nucleophile present in the reaction.  Many research groups, including the laboratories of Magnus Rueping and Wujiong Xia, have investigated related transformations of amines with a wide variety of other nucleophiles, such as [[cyanide]] ([[Strecker amino-acid synthesis|Strecker reaction]]), [[silyl enol ethers]] ([[Mannich reaction]]), dialkylphosphates, allyl silanes (aza-[[Sakurai reaction]]), [[indoles]] ([[Friedel-Crafts reaction]]), and copper acetylides.<ref>{{cite journal|last=Rueping|first=Magnus|coauthors=Zhu, Shaoqun; Koenigs, René M.|title=Visible-light photoredox catalyzed oxidative Strecker reaction|journal=Chemical Communications|year=2011|volume=47|issue=47|pages=12709–11|doi=10.1039/C1CC15643H|pmid=22041859}}</ref><ref>{{cite journal|last=Zhao|first=Guolei|coauthors=Yang, Chao; Guo, Lin; Sun, Hongnan; Chen, Chao; Xia, Wujiong|title=Visible light-induced oxidative coupling reaction: easy access to Mannich-type products|journal=Chemical Communications|year=2012|volume=48|issue=17|pages=2337–9|doi=10.1039/C2CC17130A|pmid=22252544}}</ref><ref>{{cite journal|last=Rueping|first=Magnus|coauthors=Zhu, Shaoqun; Koenigs, René M.|title=Photoredox catalyzed C–P bond forming reactions—visible light mediated oxidative phosphonylations of amines|journal=Chemical Communications|year=2011|volume=47|issue=30|pages=8679–81|doi=10.1039/C1CC12907D|pmid=21720622}}</ref><ref>{{cite journal|last=Freeman|first=David B.|coauthors=Furst, Laura; Condie, Allison G.; Stephenson, Corey R. J.|title=Functionally Diverse Nucleophilic Trapping of Iminium Intermediates Generated Utilizing Visible Light|journal=Organic Letters|date=6 January 2012|volume=14|issue=1|pages=94–97|doi=10.1021/ol202883v|pmid=22148974|pmc=3253246}}</ref><ref>{{cite journal|last=Rueping|first=Magnus|coauthors=Koenigs, René M.; Poscharny, Konstantin; Fabry, David C.; Leonori, Daniele; Vila, Carlos|title=Dual Catalysis: Combination of Photocatalytic Aerobic Oxidation and Metal Catalyzed Alkynylation Reactions-C≡C Bond Formation Using Visible Light|journal=Chemistry - A European Journal|date=23 April 2012|volume=18|issue=17|pages=5170–5174|doi=10.1002/chem.201200050}}</ref>

[[File:Iminium Functionalization.png|600px|frameless|center|Photocatalytic generation and functionalization of iminium ions]]

Similar photoredox generation of iminium ions has furthermore been achieved using purely organic photoredox catalysts, such as Rose Bengal and Eosin Y.<ref>{{cite journal|last=Pan|first=Yuanhang|coauthors=Wang, Shuai; Kee, Choon Wee; Dubuisson, Emilie; Yang, Yuanyong; Loh, Kian Ping; Tan, Choon-Hong|title=Graphene oxide and Rose Bengal: oxidative C–H functionalisation of tertiary amines using visible light|journal=Green Chemistry|year=2011|volume=13|issue=12|pages=3341|doi=10.1039/C1GC15865A}}</ref><ref>{{cite journal|last=Fu|first=Weijun|coauthors=Guo, Wenbo; Zou, Guanglong; Xu, Chen|title=Selective trifluoromethylation and alkynylation of tetrahydroisoquinolines using visible light irradiation by Rose Bengal|journal=Journal of Fluorine Chemistry|date=August 2012|volume=140|pages=88–94|doi=10.1016/j.jfluchem.2012.05.009}}</ref><ref>{{cite journal|last=Hari|first=Durga Prasad|coauthors=König, Burkhard|title=Eosin Y Catalyzed Visible Light Oxidative C–C and C–P bond Formation|journal=Organic Letters|date=5 August 2011|volume=13|issue=15|pages=3852–3855|doi=10.1021/ol201376v|pmid=21744842}}</ref>

[[File:Organic Photoredox Catalysts.png|400px|frameless|center|Two organic dyes used in photoredox catalysis]]

Furthermore, an asymmetric variant of this reaction has been developed by Rovis et al. utilizing acyl nucleophile equivalents generated by N-heterocyclic carbene catalysis.<ref>{{cite journal|last=DiRocco|first=Daniel A.|coauthors=Rovis, Tomislav|title=Catalytic Asymmetric α-Acylation of Tertiary Amines Mediated by a Dual Catalysis Mode: N-Heterocyclic Carbene and Photoredox Catalysis|journal=Journal of the American Chemical Society|date=16 May 2012|volume=134|issue=19|pages=8094–8097|doi=10.1021/ja3030164|pmid=22548244|pmc=3354013}}</ref>   This reaction method sidesteps the problem of poor enantioinduction from chiral photoredox catalysts by moving the source of enantioselectivity to the N-heterocyclic carbene.

[[File:Carbene Catalyst.png|200px|frameless|center|Carbene Catalyst used in alpha-acylation of amines]]

=== Oxidative generation of oxocarbenium ions ===

The development of orthogonal protecting group chemistry is a crucial problem in organic synthesis because it is the use of these protecting groups that allows each instance of a common functional group, such as the hydroxyl group, to be distinguished during the synthesis of a complex molecule.  One very common protecting group for the hydroxyl functional group is the ''para''-methoxy benzyl (PMB) ether.  This protecting group is chemically very similar to the less electron-rich benzyl ether.  The usual method for selective cleavage of a PMB ether in the presence of a benzyl ether is through the use of strong stoichiometric oxidants such as [[2,3-dichloro-5,6-dicyano-1,4-benzoquinone]] (DDQ) or [[ceric ammonium nitrate]] (CAN).  PMB ethers are far more susceptible to oxidation than benzyl ethers because they are more electron-rich.  Research by Stephenson, et al. has shown that the selective deprotection of PMB ethers can be achieved through the use of bis-(2-(2',4'-difluorophenyl)-5-trifluoromethylpyridine)-(4,4'-ditertbutylbipyridine)iridium(III) hexafluorophosphate (Ir[dF(CF<sub>3</sub>)ppy]<sub>2</sub>(dtbbpy)PF<sub>6</sub>) and a mild stoichiometric oxidant such as bromotrichloromethane, CBrCl<sub>3</sub>.<ref>{{cite journal|last=Tucker|first=Joseph W.|coauthors=Narayanam, Jagan M. R.; Shah, Pinkey S.; Stephenson, Corey R. J.|title=Oxidative photoredox catalysis: mild and selective deprotection of PMB ethers mediated by visible light|journal=Chemical Communications|year=2011|volume=47|issue=17|pages=5040–5042|doi=10.1039/c1cc10827a|pmid=21431223}}</ref>   The photoexcited iridium catalyst is reducing enough to fragment the polyhalomethane compound to form trichloromethyl radical, bromide anion and the Ir(IV) oxidation state of the catalyst.  The electron-poor nature of the fluorinated ligands means that this iridium complex can be readily reduced: in particular, by an electron-rich arene such as a para-methoxy benzyl ether.  After the arene is oxidized, it will readily participate in H-atom transfer with trichloromethyl radical to form chloroform and an [[oxocarbenium]] ion, which is readily hydrolyzed to reveal the free hydroxide.  This reaction was demonstrated to be orthogonal to many common protecting groups, especially with the addition of a base to counteract the buildup of HBr during the reaction.

[[File:PMB Deprotection.png|600px|frameless|center|Photoredox-catalyzed PMB deprotection]]

=== Cycloadditions ===

[[Cycloaddition]]s and other [[pericyclic reactions]] are powerful transforms in organic synthesis because of their potential to rapidly generate complex molecular architectures and particularly because of their capacity to set multiple adjacent [[stereocenter]]s in a highly controlled manner.  However, only certain cycloadditions are allowed under thermal conditions according to the [[Woodward-Hoffmann rules]] of orbital symmetry, or other equivalent models such as [[frontier molecular orbital theory]] (FMO) or the Dewar-Zimmermann model.  Cycloadditions which are not thermally allowed, such as the [2+2] cycloaddition, can be enabled by photochemical activation of the reaction.  Under uncatalyzed conditions, this activation requires the use of high energy ultraviolet light capable of altering the orbital populations of the reactive compounds.  Alternatively, metal catalysts such as cobalt and copper have been reported to catalyze thermally-forbidden [2+2] cycloadditions via single electron transfer.

[[File:Enone Cycloaddition Figure.png|400px|frameless|center|Diagram of Photocatalytic Crossed Enone 2+2 cycloaddition]]

Recent work by the Yoon lab has demonstrated that the required change in orbital populations can be achieved by electron transfer with a photocatalyst sensitive to lower energy visible light.<ref>{{cite journal|last=Ischay|first=Michael A.|coauthors=Anzovino, Mary E.; Du, Juana; Yoon, Tehshik P.|title=Efficient Visible Light Photocatalysis of [2+2] Enone Cycloadditions|journal=Journal of the American Chemical Society|date=October 2008|volume=130|issue=39|pages=12886–12887|doi=10.1021/ja805387f|pmid=18767798}}</ref><ref>{{cite journal|last=Du|first=Juana|coauthors=Yoon, Tehshik P.|title=Crossed Intermolecular [2+2] Cycloadditions of Acyclic Enones via Visible Light Photocatalysis|journal=Journal of the American Chemical Society|date=21 October 2009|volume=131|issue=41|pages=14604–14605|doi=10.1021/ja903732v|pmid=19473018|pmc=2761970}}</ref><ref>{{cite journal|last=Ischay|first=Michael A.|coauthors=Lu, Zhan; Yoon, Tehshik P.|title=[2+2] Cycloadditions by Oxidative Visible Light Photocatalysis|journal=Journal of the American Chemical Society|date=30 June 2010|volume=132|issue=25|pages=8572–8574|doi=10.1021/ja103934y|pmid=20527886|pmc=2892825}}</ref><ref>{{cite journal|last=Tyson|first=Elizabeth L.|coauthors=Farney, Elliot P.; Yoon, Tehshik P.|title=Photocatalytic [2 + 2] Cycloadditions of Enones with Cleavable Redox Auxiliaries|journal=Organic Letters|date=17 February 2012|volume=14|issue=4|pages=1110–1113|doi=10.1021/ol3000298|pmid=22320352|pmc=3288794}}</ref><ref>{{cite journal|last=Ischay|first=Michael A.|coauthors=Ament, Michael S.; Yoon, Tehshik P.|title=Crossed intermolecular [2 + 2] cycloaddition of styrenes by visible light photocatalysis|journal=Chemical Science|year=2012|volume=3|issue=9|pages=2807–2811|doi=10.1039/c2sc20658g|pmid=22984640|pmc=3439822}}</ref> Yoon's work has demonstrated the efficient intra- and intermolecular [2+2] cycloadditions of activated olefins: particularly enones and styrenes.  Enones, or electron-poor olefins, were discovered to react via a radical-anion pathway, utilizing diisopropylethylamine as a transient source of electrons.  For this electron-transfer, Ru(bipy)<sub>3</sub><sup>2+</sup> was discovered to be an efficient photocatalyst.  Interestingly, the anionic nature of the cyclization proved to be crucial: performing the reaction in acid rather than with a lithium counterion favored a non-cycloaddition pathway.<ref>{{cite journal|last=Du|first=Juana|coauthors=Espelt, Laura Ruiz; Guzei, Ilia A.; Yoon, Tehshik P.|title=Photocatalytic reductive cyclizations of enones: Divergent reactivity of photogenerated radical and radical anion intermediates|journal=Chemical Science|year=2011|volume=2|issue=11|pages=2115–2119|doi=10.1039/c1sc00357g|pmid=22121471|pmc=3222952}}</ref>  Zhao et al. have likewise discovered that a still different cyclization pathway is available to chalcones with a samarium counterion.<ref>{{cite journal|last=Zhao|first=Guolei|coauthors=Yang, Chao; Guo, Lin; Sun, Hongnan; Lin, Run; Xia, Wujiong|title=Reactivity Insight into Reductive Coupling and Aldol Cyclization of Chalcones by Visible Light Photocatalysis|journal=The Journal of Organic Chemistry|date=20 July 2012|volume=77|issue=14|pages=6302–6306|doi=10.1021/jo300796j|pmid=22731518}}</ref>   Conversely, electron-rich styrenes were found to react via a radical-cation mechanism, utilizing methyl viologen or molecular oxygen as a transient electron sink.  While Ru(bipy)<sub>3</sub><sup>2+</sup> proved to be a competent catalyst for intramolecular cyclizations using methyl viologen, it could not be used with molecular oxygen as an electron sink or for intermolecular cyclizations.  For intermolecular cyclizations, Yoon et al. discovered that the more strongly oxidizing photocatalyst Ru(bpm)<sub>3</sub><sup>2+</sup> and molecular oxygen provided a catalytic system better suited to access the radical cation necessary for the cycloaddition to occur.  Ru(bpz)<sub>3</sub><sup>2+</sup>, a still more strongly oxidizing photocatalyst, proved to be problematic because it was not only strong enough to oxidize the reagents and catalyze the desired [2+2] cycloaddition, but also was strong enough to oxidize the cycloadduct and catalyze the retro-[2+2] reaction.  This comparison of photocatalysts highlights the importance of tuning the redox properties of a photocatalyst to the reaction system as well as demonstrating the value of polypyridyl compounds as ligands due to the ease with which they can be modified to adjust the redox properties of their complexes.

[[File:Styrene Cycloaddition Figure.png|400px|frameless|center|Diagram of Photocatalytic Crossed Styrene 2+2 Cycloaddition]]

Research on photoredox-catalyzed [2+2] cycloadditions has also been performed with a triphenylpyrylium organic photoredox catalyst by Nicewicz et al. as part of their study of cyclobutane lignans.<ref>{{cite journal|last=Riener|first=Michelle|coauthors=Nicewicz, David A.|title=Synthesis of cyclobutane lignans via an organic single electron oxidant–electron relay system|journal=Chemical Science|year=2013|volume=4|issue=6|pages=2625|doi=10.1039/c3sc50643f}}</ref>

[[File:Triphenylpyrylium Catalyst.png|400px|frameless|center|Triphenylpyrylium Photoredox catalyst]]

In addition to research on the thermally-forbidden [2+2] cycloaddition, research in the Yoon group has also studied photoredox catalysis of the [4+2] cyclization, also known as the Diels-Alder reaction.  Yoon et al. discovered that bis-enones, similar to the substrates used for the photoredox [2+2] cyclization, but with a longer linker joining the two enone functional groups, underwent intramolecular radical-anion hetero-Diels-Alder reactions more rapidly than [2+2] cycloaddition.<ref>{{cite journal|last=Hurtley|first=Anna E.|coauthors=Cismesia, Megan A.; Ischay, Michael A.; Yoon, Tehshik P.|title=Visible light photocatalysis of radical anion hetero-Diels–Alder cycloadditions|journal=Tetrahedron|date=June 2011|volume=67|issue=24|pages=4442–4448|doi=10.1016/j.tet.2011.02.066|pmid=21666769|pmc=3110713}}</ref>

[[File:Enone Diels-Alder Figure.png|400px|frameless|center|Diagram of Photocatalytic bis-enone hetero-Diels-Alder reaction]]

Similarly, Yoon et al. discovered that electron-rich styrenes participated in intra- or intermolecular Diels-Alder cyclizations via a radical cation mechanism.<ref name="Lin2011">{{cite journal|last=Lin|first=Shishi|coauthors=Ischay, Michael A.; Fry, Charles G.; Yoon, Tehshik P.|title=Radical Cation Diels–Alder Cycloadditions by Visible Light Photocatalysis|journal=Journal of the American Chemical Society|date=7 December 2011|volume=133|issue=48|pages=19350–19353|doi=10.1021/ja2093579|pmid=22032252|pmc=3227774}}</ref><ref>{{cite journal|last=Lin|first=Shishi|coauthors=Padilla, Christian E.; Ischay, Michael A.; Yoon, Tehshik P.|title=Visible light photocatalysis of intramolecular radical cation Diels–Alder cycloadditions|journal=Tetrahedron Letters|date=June 2012|volume=53|issue=24|pages=3073–3076|doi=10.1016/j.tetlet.2012.04.021|pmid=22711942|pmc=3375996}}</ref> In contrast to the Yoon's lab discovery that Ru(bipy)<sub>3</sub><sup>2+</sup> was competent for intramolecular, but not intermolecular [2+2] cyclizations, Yoon et al. discovered that Ru(bipy)<sub>3</sub><sup>2+</sup> was a competent catalyst for intermolecular, but not intramolecular, Diels-Alder cyclizations.  One particular advantage of this photoredox-catalyzed Diels-Alder reaction is that it allows cycloaddition between two electronically mismatched substrates.  The normal electronic demand for the Diels-Alder reaction calls for an electron-rich diene to react with an electron-poor olefin (or "dienophile"), while the inverse electron-demand Diels-Alder reaction takes place between the opposite case of an electron-poor diene and a very electron-rich dienophile.  The photoredox case, since it takes place by a different mechanism than the thermal Diels-Alder reaction, allows cycloaddition between an electron-rich diene and an electron-rich dienophile, allowing access to new classes of Diels-Alder adducts.

[[File:Styrene Diels-Alder Figure.png|400px|frameless|center|Diagram of Photocatalytic styrene Diels-Alder reaction]]

The synthetic value of Yoon's photoredox-catalyzed styrene Diels-Alder reaction was demonstrated via the total synthesis of the natural product Heitziamide A.<ref name="Lin2011" />  In this synthesis, Yoon et al. demonstrated that the thermal Diels-Alder reaction favors the undesired regioisomer, but the photoredox-catalyzed reaction gives the desired regioisomer in improved yield.

[[File:Heitziamide A Synthesis.png|400px|frameless|center|Key Photoredox Cycloaddition in Total Synthesis of Heitziamide A]]

=== Photoredox organocatalysis ===

[[Organocatalysis]] is a separate subfield of catalysis that explores the potential of organic small molecules as catalysts, particularly for the enantioselective creation of chiral molecules.  One major strategy in this subfield is the use of chiral secondary amines to activate carbonyl compounds.  In this case, condensation of the amine with the carbonyl compound generates a nucleophilic enamine.  The chiral amine is designed so that one face of the enamine is sterically shielded and so that only the unshielded face is free to react.  Despite the power of this approach to catalyze the enantioselective functionalization of carbonyl compounds, certain valuable transformations, such as the catalytic enantioselective α-alkylation of aldehydes, remained elusive.  Recent work by the MacMillan group revealed that the synergistic combination of organocatalysis with photoredox methods provides a simple catalytic solution to this problem.<ref>{{cite journal|last=Nicewicz|first=D. A.|coauthors=MacMillan, D. W. C.|title=Merging Photoredox Catalysis with Organocatalysis: The Direct Asymmetric Alkylation of Aldehydes|journal=Science|date=3 October 2008|volume=322|issue=5898|pages=77–80|doi=10.1126/science.1161976|pmid=18772399|pmc=2723798}}</ref>
In the MacMillan group's approach for the α-alkylation of aldehydes, Ru(bipy)<sub>3</sub><sup>2+</sup> reductively fragments an activated alkyl halide, such as bromomalonate or phenacyl bromide, which can then add to catalytically-generated enamine in an enantioselective manner.  The oxidized photocatalyst then oxidatively quenches the resulting α-amino radical to form an iminium ion, which hydrolyzes to give the functionalized carbonyl compound.  This photoredox transformation was shown to be mechanistically distinct from another organocatalytic radical process developed in the MacMillan lab, termed singly-occupied molecular orbital (SOMO) catalysis.  The method of SOMO catalysis employs superstoichiometric ceric ammonium nitrate (CAN) to oxidize the catalytically-generated enamine to the corresponding radical cation, which can then add to a suitable coupling partner such as allyl silane.  Nicewicz and MacMillan were able to exclude this type of mechanism from the photocatalytic alkylation reaction because whereas enamine radical cation was observed to cyclize onto pendant olefins and open cyclopropane radical clocks in SOMO catalysis, these structures were unreactive in the photoredox reaction.

[[File:Synergistic Alkylation Figure.png|400px|frameless|center|Diagram of Enantioselective Alkylation of Aldehydes via the synergistic combination of organo- and photoredox catalysis]]

Continued research in the MacMillan has expanded the scope of this transformation to include alkylation with other classes of activated alkyl halides of synthetic interest.  In particular, the use of the photocatalyst Ir(dtbbpy)(ppy)<sub>2</sub><sup>+</sup> allowed the direct and enantioselective α-trifluoromethylation of aldehydes while the use of Ir(ppy)<sub>3</sub> allowed the enantioselective coupling of aldehydes with electron-poor benzylic bromides.<ref>{{cite journal|last=Nagib|first=David A.|coauthors=Scott, Mark E.; MacMillan, David W. C.|title=Enantioselective α-Trifluoromethylation of Aldehydes via Photoredox Organocatalysis|journal=Journal of the American Chemical Society|date=12 August 2009|volume=131|issue=31|pages=10875–10877|doi=10.1021/ja9053338|pmid=19722670|pmc=3310169}}</ref><ref>{{cite journal|last=Shih|first=Hui-Wen|coauthors=Vander Wal, Mark N.; Grange, Rebecca L.; MacMillan, David W. C.|title=Enantioselective α-Benzylation of Aldehydes via Photoredox Organocatalysis|journal=Journal of the American Chemical Society|date=6 October 2010|volume=132|issue=39|pages=13600–13603|doi=10.1021/ja106593m|pmid=20831195|pmc=3056320}}</ref>  The productive merger of photoredox and organocatalytic methods to achieve enantioselective alkylation of aldehydes has also been investigated by Zeitler et al.<ref>{{cite journal|last=Neumann|first=Matthias|coauthors=Füldner, Stefan; König, Burkhard; Zeitler, Kirsten|title=Metal-Free, Cooperative Asymmetric Organophotoredox Catalysis with Visible Light|journal=Angewandte Chemie International Edition|date=24 January 2011|volume=50|issue=4|pages=951–954|doi=10.1002/anie.201002992}}</ref>  In this research, the same chiral imidazolidinone organocatalyst was used to form enamine and introduce chirality as in MacMillan's research.  However, the organic photoredox catalyst Eosin Y was used rather than a ruthenium or iridium complex.

MacMillan and coworkers have also developed the first direct β-arylation of saturated aldehydes and ketones through the combination of photoredox and organocatalytic methods.<ref>{{cite journal|last=Pirnot|first=M. T.|coauthors=Rankic, D. A.; Martin, D. B. C.; MacMillan, D. W. C.|title=Photoredox Activation for the Direct  -Arylation of Ketones and Aldehydes|journal=Science|date=28 March 2013|volume=339|issue=6127|pages=1593–1596|doi=10.1126/science.1232993|pmid=23539600|pmc=3723331}}</ref>   The only previous method to accomplish any direct β-functionalization of a saturated carbonyl consisted of a one-pot consists of a two-step process, both catalyzed by a secondary amine organocatalyst: stoichiometric reduction of an aldehyde with IBX followed by addition of an activated alkyl nucleophile to the beta-position of the resulting enal.<ref>{{cite journal|last=Zhang|first=Shi-Lei|coauthors=Xie, He-Xin; Zhu, Jin; Li, Hao; Zhang, Xin-Shuai; Li, Jian; Wang, Wei|title=Organocatalytic enantioselective β-functionalization of aldehydes by oxidation of enamines and their application in cascade reactions|journal=Nature Communications|date=1 March 2011|volume=2|pages=211|doi=10.1038/ncomms1214|pmid=21364550}}</ref> MacMillan's transformation, which like other photoredox transformations takes place by a radical mechanism, is limited to the addition of highly electrophilic arenes to the beta position.  The severe limitations on the scope of the arene component in this reaction is due primarily to the need for an arene radical anion which is stable enough not to react directly with enamine or enamine radical cation.  In MacMillan's proposed mechanism, the activated photoredox catalyst is quenched oxidatively by an electron-deficient arene, such as 1,4-dicyanobenzene.  The photocatalyst then oxidizes an enamine species, transiently generated by the condensation of an aldehyde with a secondary amine cocatalyst, such as isopropyl benzylamine, the catalyst determined to be optimal by MacMillan et al.  The resulting enamine radical cation usually reacts as a 3 π-electron system, but due to the stability of the radical coupling partners, deprotonation of the β-methylene position gives rise to a 5 π-electron system with strong radical character at the newly accessed β-carbon.  Although this reaction relies on the use of a secondary amine organocatalyst to generate the enamine species which is oxidized in MacMillan's proposed mechanism, an enantioselective variant of this reaction has not yet been developed.

[[File:Beta Arylation Diagram.png|400px|frameless|center|Diagram of Photoredox beta-arylation of aldehydes]]

The development of this direct β-arylation of aldehydes has led to related reactions for the β-functionalization of cyclic ketones.  In particular, β-arylation of cyclic ketones has been achieved under similar reaction conditions, but using azepane as the secondary amine cocatalyst.  More recently, a photocatalytic "homo-aldol" reaction has been developed for cyclic ketones, allowing the coupling of the beta-position of the ketone to the ipso carbon of aryl ketones, such as benzophenone and acetophenone.<ref>{{cite journal|last=Petronijević|first=Filip R.|coauthors=Nappi, Manuel; MacMillan, David W. C.|title=Direct β-Functionalization of Cyclic Ketones with Aryl Ketones via the Merger of Photoredox and Organocatalysis|journal=Journal of the American Chemical Society|date=22 November 2013|pages=131122154626007|doi=10.1021/ja410478a|volume=135|issue=49}}</ref>   In addition to the azepane cocatalyst, this reaction required the use of the more strongly reducing photoredox catalyst Ir(ppy)<sub>3</sub> and the addition of lithium hexafluoroarsenide (LiAsF<sub>6</sub>) in order to promote single-electron reduction of the aryl ketone.

===Additions to olefins===

The use of photoredox catalysis to generate reactive heteroatom-centered radicals was first explored by Barton et al. in 1994.<ref>{{cite journal|last=Barton|first=Derek H.R.|coauthors=Csiba, Maria A.; Jaszberenyi, Joseph Cs.|title=Ru(bpy)32+-mediated addition of Se-phenyl p-tolueneselenosulfonate to electron rich olefins|journal=Tetrahedron Letters|date=May 1994|volume=35|issue=18|pages=2869–2872|doi=10.1016/S0040-4039(00)76646-9}}</ref>  In this work, Barton noted that Ru(bipy)<sub>3</sub><sup>2+</sup> catalyzes the initial fragmentation of tosylphenylselenide to phenylselenolate anion and tosyl radical and that a radical chain propagation mechanism allowed the addition of tosyl radical and phenylseleno- radical across the double bond of electron rich alkyl vinyl ethers.  Since phenylselenolate anion is readily oxidized to diphenyldiselenide, the low quantities of diphenyldiselenide observed was taken as an indication that photoredox-catalyzed fragmentation of tosylphenylselenide was only important as an initiation step, and that most of the reactivity was due to a radical chain process.

[[File:Olefin Additions.png|400px|frameless|center|Addition of Tosylphenylselenide across an olefin]]

More recent examples of heteroaromatic additions to olefins include multicomponent oxy- and aminotrifluoromethylation reactions developed by Akita et al.<ref>{{cite journal|last=Yasu|first=Yusuke|coauthors=Koike, Takashi; Akita, Munetaka|title=Three-component Oxytrifluoromethylation of Alkenes: Highly Efficient and Regioselective Difunctionalization of C=C Bonds Mediated by Photoredox Catalysts|journal=Angewandte Chemie International Edition|date=17 September 2012|volume=51|issue=38|pages=9567–9571|doi=10.1002/anie.201205071}}</ref><ref>{{cite journal|last=Yasu|first=Yusuke|coauthors=Koike, Takashi; Akita, Munetaka|title=Intermolecular Aminotrifluoromethylation of Alkenes by Visible-Light-Driven Photoredox Catalysis|journal=Organic Letters|date=3 May 2013|volume=15|issue=9|pages=2136–2139|doi=10.1021/ol4006272|pmid=23600821}}</ref>  These reactions use Umemoto's reagent, a sulfonium salt that serves as an electrophilic source of the trifluoromethyl group and which is precedented to react via a single-electron transfer pathway.  Thus, single-electron reduction of Umemoto's reagent releases trifluoromethyl radical which adds to the reactive olefin.  Subsequently, single-electron oxidation of the alkyl radical generated by this addition produces a cation which can be trapped by water, an alcohol, or a nitrile.  In order to achieve high levels of regioselectivity, this reactivity has been explored mainly for styrenes, which are biased towards formation of the benzylic radical intermediate.

[[File:Olefin Trifluoromethylation.png|400px|frameless|center|Photoredox-catalyzed oxy- and aminotrifluoromethylation]]

Nicewicz et al. have studied the related hydrotrifluoromethylation of styrenes and aliphatic alkenes with a mesityl acridinium organic photoredox catalyst and Langlois' reagent as the source of CF<sub>3</sub> radical.<ref>{{cite journal|last=Wilger|first=Dale J.|coauthors=Gesmundo, Nathan J.; Nicewicz, David A.|title=Catalytic hydrotrifluoromethylation of styrenes and unactivated aliphatic alkenes via an organic photoredox system|journal=Chemical Science|year=2013|volume=4|issue=8|pages=3160|doi=10.1039/c3sc51209f}}</ref>   In this reaction, it was found that trifluoroethanol and substoichiometric amounts of an aromatic thiol, such as methyl thiosalicylate, employed in tandem served as the best source of hydrogen radical to complete the catalytic cycle.

[[File:Hydrotrifluoromethylation Reagents.png|400px|frameless|center|Diagram of two reagents for olefin hydrotrifluoromethylation]]

Nicewicz has also studied intramolecular hydroetherifications and hydroaminations that proceed with anti-Markovnikov selectivity.<ref>{{cite journal|last=Hamilton|first=David S.|coauthors=Nicewicz, David A.|title=Direct Catalytic Anti-Markovnikov Hydroetherification of Alkenols|journal=Journal of the American Chemical Society|date=14 November 2012|volume=134|issue=45|pages=18577–18580|doi=10.1021/ja309635w|pmid=23113557|pmc=3513336}}</ref><ref>{{cite journal|last=Nguyen|first=Tien M.|coauthors=Nicewicz, David A.|title=Anti-Markovnikov Hydroamination of Alkenes Catalyzed by an Organic Photoredox System|journal=Journal of the American Chemical Society|date=3 July 2013|volume=135|issue=26|pages=9588–9591|doi=10.1021/ja4031616|pmid=23768239|pmc=3754854}}</ref>   Mechanistically, Nicewicz invokes the single-electron oxidation of the olefin, trapping of the radical cation by a pendant hydroxyl or amine functional group, and quenching of the resulting alkyl radical by H-atom transfer from a highly labile donor species.  Extensions of this reactivity to intermolecular systems have resulted in i) a new synthetic route to complex tetrahydrofurans by a "polar-radical-crossover cycloaddition" (PRCC reaction) of an allylic alcohol with an olefin, and ii) the anti-Markovnikov addition of carboxylic acids to olefins.<ref>{{cite journal|last=Grandjean|first=Jean-Marc M.|coauthors=Nicewicz, David A.|title=Synthesis of Highly Substituted Tetrahydrofurans by Catalytic Polar-Radical-Crossover Cycloadditions of Alkenes and Alkenols|journal=Angewandte Chemie International Edition|date=2 April 2013|volume=52|issue=14|pages=3967–3971|doi=10.1002/anie.201210111}}</ref><ref>{{cite journal|last=Perkowski|first=Andrew J.|coauthors=Nicewicz, David A.|title=Direct Catalytic Anti-Markovnikov Addition of Carboxylic Acids to Alkenes|journal=Journal of the American Chemical Society|date=17 July 2013|volume=135|issue=28|pages=10334–10337|doi=10.1021/ja4057294|pmid=23808532|pmc=3757928}}</ref>

== References ==

{{Reflist}}

[[Category:Catalysis]]